PDF An Introduction To Compressible Flows With Applications Quasi One Dimensional Approximation and General Formulation For Subsonic Transonic and Supersonic Flows Jose Pontes Ebook Full Chapter

An Introduction to Compressible Flows
with Applications Quasi One

Dimensional Approximation and
General Formulation for Subsonic
Transonic and Supersonic Flows José
Pontes
Visit to download the full and correct content document:
https://textbookfull.com/product/an-introduction-to-compressible-flows-with-application
s-quasi-one-dimensional-approximation-and-general-formulation-for-subsonic-transon
ic-and-supersonic-flows-jose-pontes/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Calculus of Variations An Introduction to the One

Dimensional Theory with Examples and Exercises 1st
Edition Hansjörg Kielhöfer (Auth.)
https://textbookfull.com/product/calculus-of-variations-an-
introduction-to-the-one-dimensional-theory-with-examples-and-
exercises-1st-edition-hansjorg-kielhofer-auth/
Flows of Non Smooth Vector Fields and Degenerate

Elliptic Equations With Applications to the Vlasov
Poisson and Semigeostrophic Systems Maria Colombo
https://textbookfull.com/product/flows-of-non-smooth-vector-
fields-and-degenerate-elliptic-equations-with-applications-to-
the-vlasov-poisson-and-semigeostrophic-systems-maria-colombo/
Computational Techniques for Multiphase Flows 2nd

Edition Guan Heng
https://textbookfull.com/product/computational-techniques-for-
multiphase-flows-2nd-edition-guan-heng/
Analytic Solutions for Flows Through Cascades Peter

Jonathan Baddoo
https://textbookfull.com/product/analytic-solutions-for-flows-
through-cascades-peter-jonathan-baddoo/
Non Newtonian Fluid Mechanics and Complex Flows Angiolo
Farina
https://textbookfull.com/product/non-newtonian-fluid-mechanics-
and-complex-flows-angiolo-farina/
Introduction to the Variational Formulation in

Mechanics Fundamentals and Applications 1st Edition
Edgardo O. Taroco
https://textbookfull.com/product/introduction-to-the-variational-
formulation-in-mechanics-fundamentals-and-applications-1st-
edition-edgardo-o-taroco/
Mathematical Modeling of Unsteady Inviscid Flows Jeff

D. Eldredge
https://textbookfull.com/product/mathematical-modeling-of-
unsteady-inviscid-flows-jeff-d-eldredge/
Design Synthesis and Applications of One Dimensional

Chalcogenide Hetero Nanostructures Tao-Tao Zhuang
https://textbookfull.com/product/design-synthesis-and-
applications-of-one-dimensional-chalcogenide-hetero-
nanostructures-tao-tao-zhuang/
New Trends and Results in Mathematical Description of

Fluid Flows Miroslav Bulí■ek
https://textbookfull.com/product/new-trends-and-results-in-
mathematical-description-of-fluid-flows-miroslav-bulicek/
SPRINGER BRIEFS IN MATHEMATICS
José Pontes
Norberto Mangiavacchi
Gustavo R. Anjos
An Introduction
to Compressible Flows
with Applications
Quasi-One-Dimensional
Approximation and
General Formulation
for Subsonic, Transonic
and Supersonic Flows
123
SpringerBriefs in Mathematics
Series Editors
Nicola Bellomo, Torino, Italy
Michele Benzi, Pisa, Italy
Palle Jorgensen, Iowa, USA
Tatsien Li, Shanghai, China
Roderick Melnik,Waterloo, Canada
Otmar Scherzer, Linz, Austria
Benjamin Steinberg, New York, USA
Lothar Reichel, Kent, USA
Yuri Tschinkel, New York, USA
George Yin, Detroit, USA
Ping Zhang, Kalamazoo, USA
SpringerBriefs in Mathematics showcases expositions in all areas of mathematics

and applied mathematics. Manuscripts presenting new results or a single new result
in a classical field, new field, or an emerging topic, applications, or bridges between
new results and already published works, are encouraged. The series is intended for
mathematicians and applied mathematicians.
More information about this series at http://www.springer.com/series/10030

SBMAC SpringerBriefs
Editorial Board
Luis Gustavo Nonato

University of São Paulo (USP)
Institute of Mathematical and Computer Sciences (ICMC)
Department of Applied Mathematics and Statistics
São Carlos, Brazil
Paulo J. S. Silva
University of Campinas (UNICAMP)
Institute of Mathematics, Statistics and Scientific Computing (IMECC)
Department of Applied Mathematics
Campinas, Brazil
The SBMAC SpringerBriefs series publishes relevant contributions in the fields

of applied and computational mathematics, mathematics, scientific computing, and
related areas. Featuring compact volumes of 50 to 125 pages, the series covers a
range of content from professional to academic.
The Sociedade Brasileira de Matemática Aplicada e Computacional (Brazilian

Society of Computational and Applied Mathematics, SBMAC) is a professional
association focused on computational and industrial applied mathematics. The
society is active in furthering the development of mathematics and its applications
in scientific, technological, and industrial fields. The SBMAC has helped to develop
the applications of mathematics in science, technology, and industry, to encourage
the development and implementation of effective methods and mathematical tech-
niques for the benefit of science and technology, and to promote the exchange of
ideas and information between the diverse areas of application.
http://www.sbmac.org.br/
José Pontes • Norberto Mangiavacchi
Gustavo R. Anjos
An Introduction to
Compressible Flows
with Applications
Quasi-One-Dimensional Approximation and
General Formulation for Subsonic, Transonic
and Supersonic Flows
José Pontes Norberto Mangiavacchi
Faculty of Engineering Faculty of Engineering
State University of Rio de Janeiro State University of Rio de Janeiro
Rio de Janeiro Rio de Janeiro
Rio de Janeiro, Brazil Rio de Janeiro, Brazil
Gustavo R. Anjos
COPPE/Mechanical Engineering Program
Federal University of Rio de Janeiro
Rio de Janeiro
Rio de Janeiro, Brazil
ISSN 2191-8198 ISSN 2191-8201 (electronic)

SpringerBriefs in Mathematics
ISBN 978-3-030-33252-5 ISBN 978-3-030-33253-2 (eBook)
https://doi.org/10.1007/978-3-030-33253-2
Mathematics Subject Classification: 76B07, 76G25, 76J20
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019

This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This book is designed as an introduction to the problem of fluid flows where

compressibility effects play a major role and appear not only in consequence
of the primary high velocity imposed to the flow, but also in other engineering
configurations, like in failures and in the opening of safety valves of pipelines
employed for the transport and distribution of pressurized gases. Being devoted to
applications, the authors opted to develop the material not in the standard form found
in mathematical books, with results presented as proved theorems, but by deriving
and simplifying the governing equations.
The first chapter presents the full three-dimensional evolution equation of
potential flows, commonly found around slender aerospace structures as wings,
control surface, and rockets. The chapter discusses the propagation of weak sound
waves and the numerical solution of the linear wave equation in the frequency
domain.
The second chapter is devoted to quasi-one-dimensional isentropic gas dynamics,
and also the problem of shock compression, entropy production across shock waves,
and the thickness of such waves. Still in this chapter, the book addresses the cases of
flows in ducts with constant transversal section with heat addition and, in separate,
with friction. The cases of flow in ducts with variable transversal section with
friction or heat transfer, and with both occurring simultaneously are addressed in
sequence. In these cases, the critical condition of Mach number equal to one does not
occur at the minimum transversal section of the duct. The case of practical interest
of isothermal flow in ducts with constant transversal section, found in pipelines is
discussed next. The chapter ends pointing to the analogy between open channel
hydraulics and one-dimensional gas dynamics.
Chapter 3 deals with oblique shock waves, supersonic flows over corners and
wedges, and Riemman problems. Chapter 4 is devoted to the study of high velocity
flows subject to small perturbations, as found in the case of slender bodies, aligned
or almost aligned to the flow. The case of subsonic and supersonic flows past
a periodic-shaped wall is addressed in particular. The book ends with a chapter
devoted to the derivation of the basic equations of compressible fluid flows, with
v
vi Preface
attention given to the one-dimensional case, where the velocity is defined as the
ratio between volumetric flow and the local area of the transversal section.
At the end of each chapter, problems are proposed to further develop theoretical
and computational aspects of compressible flows.
The text is suitable for advanced undergraduate students or beginning graduate
students in Mathematics, Physics, and Engineering, and it may be useful for
researchers in academia and in the industry.
Contents
1 Compressible Potential Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 The Equation of Compressible Potential Flows . . . . . . . . . . . . . . . . . . . . . . . . 1
3 A Classification of Partial Differential Equations . . . . . . . . . . . . . . . . . . . . . . 6
4 Weak Waves: Sound Velocity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5 Solution of the Wave Equation in the Frequency Domain . . . . . . . . . . . . . 11
6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
2 One-Dimensional Compressible Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2 One-Dimensional Isentropic Gasdynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1 Effect of the Mach Number on Flow Parameters . . . . . . . . . . . . . . . 34
2.2 The Pressure Coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3 Strong Waves: Shock Compression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.1 Basic Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.2 The Rankine-Hugoniot Relation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.3 Entropy Production and Irreversibility of Shock Waves I . . . . . . 39
3.4 Entropy Production and Irreversibility of Shock Waves II . . . . . 43
3.5 The Thickness of Shock Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4 Rayleigh Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5 Fanno Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
6 Friction in Ducts with Variable Cross Section. . . . . . . . . . . . . . . . . . . . . . . . . . 61
7 Heat Transfer in Ducts with Variable Cross Section . . . . . . . . . . . . . . . . . . . 64
8 Friction and Heat Transfer in Ducts with Constant Cross Section . . . . 66
9 Friction in Isothermal Ducts with Constant Cross Section . . . . . . . . . . . . 67
10 Analogy with Open Channel Hydraulics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
10.1 The Area × Velocity Relation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
10.2 Shallow Water Theory: Hydraulic Jumps . . . . . . . . . . . . . . . . . . . . . . . 74
11 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
vii
viii Contents
3 Oblique Shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
2 Oblique Shock Waves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3 The Shock Polar Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
4 Supersonic Flows Over Corners and Wedges . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5 Riemann Problems for the Euler Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.1 Shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
5.2 Rarefactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
6 Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4 Uniform Flows with Small Perturbations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
2 The Equations of the Steady Perturbed Flows. . . . . . . . . . . . . . . . . . . . . . . . . . 103
3 The Pressure Coefficient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4 Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5 Two-Dimensional Flow Past a Periodic-Shaped Wall. . . . . . . . . . . . . . . . . . 108
5.1 Subsonic Flow Past a Periodic-Shaped Wall . . . . . . . . . . . . . . . . . . . . 108
5.2 Supersonic Flow Past a Periodic-Shaped Wall. . . . . . . . . . . . . . . . . . 112
5 The Basic Equations of Compressible Fluid Flow . . . . . . . . . . . . . . . . . . . . . . . . 115
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
2 Mass Conservation: Continuity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
3 Transport and Accumulation of a Scalar Variable. . . . . . . . . . . . . . . . . . . . . . 116
4 Momentum Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5 Bernoulli’s Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
6 Equation of the Circulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
7 Energy Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
8 The Equations of Steady One-Dimensional Flows . . . . . . . . . . . . . . . . . . . . . 123
8.1 Continuity Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
8.2 Momentum Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
8.3 Energy Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
8.4 Entropy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
List of Figures
Fig. 1.1 Characteristics of potential compressible flows . . . . . . . . . . . . . . . . . . . . . . 9

Fig. 1.2 Eigenfunctions associated with a steady acoustic wave . . . . . . . . . . . . . . 14
Fig. 1.3 Eigenfunctions associated with a steady acoustic wave . . . . . . . . . . . . . . 17
Fig. 1.4 Numerical results: velocity and pressure fields . . . . . . . . . . . . . . . . . . . . . . . 21
Fig. 1.5 Geometries and numerical meshes used in tests with forcing. . . . . . . . 24
Fig. 1.6 Frequency response of forced cavities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Fig. 1.7 Solution of pressure, velocity, and entropy for the wave equation . . . 25
Fig. 2.1 The scheme of a convergent-divergent nozzle . . . . . . . . . . . . . . . . . . . . . . . . 28
Fig. 2.2 Mach number, pressure, and temperatures along a nozzle . . . . . . . . . . . 34
Fig. 2.3 The flow through a shock wave. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Fig. 2.4 Increase of the specific mass through shock wave . . . . . . . . . . . . . . . . . . . . 39
Fig. 2.5 Rayleigh line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Fig. 2.6 Equilibrium of forces—Fanno Line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Fig. 2.7 Rayleigh and Fanno lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Fig. 2.8 Isothermal gas flow in ducts with friction and heat transfer . . . . . . . . . 70
Fig. 2.9 Flow in a variable cross section channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Fig. 3.1 Flow through oblique shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Fig. 3.2 Deflection angle β of oblique shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Fig. 3.3 Shock polar plot for M1 = 2, 5, and ∞ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Fig. 3.4 Oblique shock over dihedrals and wedges . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
Fig. 3.5 Riemann’s problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Fig. 3.6 Exact solution of a Riemann’s problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Fig. 3.7 Solution of pressure and velocity in a linear shock tube . . . . . . . . . . . . . 97
Fig. 3.8 Solution of pressure and velocity in a linear shock tube . . . . . . . . . . . . . 101
Fig. 5.1 A control volume and the mass conservation principle . . . . . . . . . . . . . . 116
Fig. 5.2 Bernoulli’s equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Fig. 5.3 Circulation around a moving mass of fluid . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
ix
Chapter 1
Compressible Potential Flows
1 Introduction
This chapter deals with compressible irrotational isentropic flows to which a

velocity potential can be associated [5, 6, 10]. Viscous effects are neglected. The
chapter starts with derivation of the three-dimensional, compressible, and time-
dependent equation governing the evolution and the steady state potential. Particular
cases of this equation are the ones governing the dynamics of weak waves, the
tridimensional compressible flow close to slender bodies aligned or quasi aligned
to an incident flow, Laplace’s equation, which governs the incompressible potential
flow under M 1. Section 3 discusses the classification of PDEs as elliptic,
parabolic, and hyperbolic.
In sequence, the chapter addresses sound propagation questions, including the
numerical solution of the wave equation in one and two dimensions, in the frequency
domains, using Finite Differences and Finite Elements methods.
2 The Equation of Compressible Potential Flows
Irrotational flows admit a velocity potential φ, from which the velocity field v
derives:
v = grad φ. (1.1)
In addition, irrotational flows satisfy Bernoulli’s equation, given by Eq. 5.11.

Conditions for an irrotational flow to stay in this condition are given by Kelvin’s
theorem Eq. 5.14 which states that vorticity is produced by thermodynamic and
viscous effects. We assume that these conditions are satisfied in the derivation of
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 1

J. Pontes et al., An Introduction to Compressible Flows with Applications,
SpringerBriefs in Mathematics, https://doi.org/10.1007/978-3-030-33253-2_1
2 1 Compressible Potential Flows
the evolution equation of the time-dependent potential associated with compressible

irrotational flows. We derive now an equation for the time evolution of φ = φ(x, t).
Euler’s equation can be written as:
∂v v2 1
+ grad − v × rot v = − grad p,
∂t 2 ρ
where v 2 = v · v. In the case of irrotational flows, this equation becomes:
∂vi ∂ v2 1 ∂p
+ = − .
∂t ∂xi 2 ρ ∂xi
Expressing the velocity components as functions of the velocity potential, vi =

∂φ/∂xi :
∂ ∂φ 1 ∂p ∂ v2
+ + = 0.
∂t ∂xi ρ ∂xi ∂xi 2
Upon integrating this last equation, we have:

∂ ∂φ 1 ∂p ∂ v2 ∂φ dp v 2
dxi + dxi + dxi = + + = F (t).
∂t ∂xi ρ ∂xi ∂xi 2 ∂t ρ 2
The time dependency expressed by the function F (t) can be incorporated to the
potential φ, leading to:

∂φ dp v 2
+ + = 0.
∂t ρ 2
We derive this last equation with respect to time, to obtain:

∂ 2φ ∂ dp 1 ∂v 2
2
+ + = 0. (1.2)
∂t ∂t ρ 2 ∂t
At this point, we introduce the hypothesis of isentropic flow, by further assuming

that no heat is transferred between the fluid particles, and that no heat is generated
in the medium. In consequence of these assumptions, along with the hypothesis of
inviscid flow, the equation of the entropy, given by:
Ds 1 κ Q̇
T = τ : grad v + ∇ 2 T + ,
Dt ρ ρ ρ
simplifies and takes the form:
Ds
T = 0,
Dt
2 The Equation of Compressible Potential Flows 3
i.e., the flow is isentropic. At each point of the field, we have, thus:
p = p(ρ, s)

∂p ∂p ∂p
dp = dρ + ds = dρ = a 2 dρ, (1.3)
∂ρ s ∂s ρ ∂ρ s
where a 2 = (∂p/∂ρ)s and ds = 0. The term containing the integral in Eq. 1.2 can
be rewritten as:

∂ dp ∂ a2 ∂ dρ ∂ a 2 ∂ρ
= dρ = a 2 = a 2 ln ρ = .
∂t ρ ∂t ρ ∂t ρ ∂t ρ ∂t
The term a 2 = (∂p/∂ρ)s has units of square of a velocity. In fact,

∂p F M ×L 1 L3 L2
= = = .
∂ρ s L2 × ρ t2 L2 M t2
In the case of perfect gases, we have:

∂p
a2 = = γ RT .
∂ρ s
As we shall see below, a 2 is the square of the sound velocity. By introducing this
definition in Eq. 1.2 we have:
∂ 2φ a 2 ∂ρ 1 ∂v 2
+ + = 0. (1.4)
∂t 2 ρ ∂t 2 ∂t
Upon multiplying Euler’s equation by vi and mentioning that vi ∂vi /∂t =

1/2 ∂v 2 /∂t, we obtain:
1 ∂v 2 ∂vi vi ∂p
+ vi vj = −
2 ∂t ∂xj ρ ∂xi
or
1 ∂v 2 ∂vi a 2 ∂ρ
+ vi vj = − vi ,
2 ∂t ∂xj ρ ∂xi
since ∂p/∂xi = a 2 ∂ρ/∂xi . By using the continuity equation, we replace the term
vi ∂ρ/∂xi in the above equation by:

∂ρ ∂ρ ∂vi
vi = − +ρ
∂xi ∂t ∂xi
to obtain:

1 ∂v 2 ∂vi a2 ∂ρ ∂vi
+ vi vj = +ρ .
2 ∂t ∂xj ρ ∂t ∂xi
Taking Eq. 1.4 into account, we replace the term (a 2 /ρ)(∂ρ/∂t) in the above
equation:
1 ∂v 2 ∂vi ∂vi ∂ 2φ 1 ∂v 2
+ vi vj = a2 − 2 − .
2 ∂t ∂xj ∂xi ∂t 2 ∂t
By replacing 1/2 ∂v 2 /∂t = vi ∂vi /∂t we have:
∂vi ∂vi ∂vi ∂ 2φ

2vi + vi vj = a2 − 2.
∂t ∂xj ∂xi ∂t
We replace now vi = ∂φ/∂xi to obtain:
∂φ ∂ 2 φ ∂φ ∂φ ∂ 2 φ ∂ 2φ ∂ 2φ
2 + = a2 2 − 2 ,
∂xi ∂t∂xi ∂xi ∂xj ∂xi ∂xj ∂xi ∂t
and finally:

∂ 2φ 1 ∂φ ∂φ ∂ 2 φ ∂φ ∂ 2 φ ∂ 2φ
2
= 2 +2 + 2 . (1.5)
∂xi a ∂xi ∂xj ∂xi ∂xj ∂xi ∂xi ∂t ∂t
In vector notation:

1 ∂∇φ ∂ 2φ
∇ φ = 2
2
∇φ ⊗ ∇φ : ∇∇φ + 2∇φ · + 2 . (1.6)
a ∂t ∂t
The potential equation of the compressible flows is written, in indexes and in vector
notations, by Eqs. 1.5 and 1.6 respectively. The equation holds in domains without
entropy production and where the inflow is irrotational.
Boundary conditions for the potential φ:
1. Stationary solids:
∂φ
= 0,
∂n
where n is the coordinate along the direction n, perpendicular to the solid surface.
The condition is equivalent to grad φ · n = 0.
2 The Equation of Compressible Potential Flows 5
2. Solids moving with steady velocity U:
∂φ
= n · U.
∂n
This condition may also be written as: n · (v − U) = 0, where v is local velocity.

We consider now some limit cases for Eq. 1.5.
Notation We denote the derivation by an index containing the variable relative to
which the potential φ is derived: ∂φ∂t = φt , ∂ 2 φ/∂x 2 = φxx , etc.
1. Two-dimensional steady flows:
(φx2 − a 2 )φxx + (φy2 − a 2 )φyy + 2φx φy φxy = 0 (1.7)
or
(u2 − a 2 )φxx + (v 2 − a 2 )φyy + 2vx vy φxy = 0,
where u and v stand for the velocity components along the x and y directions,
respectively.
2. Tridimensional steady flows, with vx vy , vx vz (slender bodies with low
angle of attack and out of the transonic region):
(1 − M 2 )φxx + φyy + φzz = 0.
In this case, the slender body aligned or quasi aligned to the inflow gives raise to
small velocity components in the orthogonal directions.
3. Steady tridimensional flow, with all velocity components small, when compared
to the sound velocity. In this case we consider 1/a 2 −→ 0 to obtain:
φxx + φyy + φzz = 0.
The last one is the well-known Laplace’s equation, which can also be derived
from the continuity equation for incompressible flows, also assuming that
irrotational flows admit a potential from which the velocity field derives (Eq. 1.1).
By replacing this equation in the continuity equation we have:
∂vi ∂ ∂φ
= = ∇ 2 φ = 0. (1.8)
∂xi ∂xi ∂xi
We stress that the hypothesis of incompressible flows is equivalent to assume that

the sound velocity diverges. The right-hand side expression of Eq. 1.5 vanishes.
Equation 1.8 applies to any incompressible potential flow and not only to cases
where vx vy and vx vz . Problems to which this equation applies belong to a
particular class where the hydrodynamic field is obtained by solving the potential
equation, followed by a post-processing where the velocity components are

obtained as the gradient of the potential. When necessary, the pressure gradient
is obtained with the Euler equation.
Equation 1.8 has the important feature of being linear. Due to that, the sum of
two solutions of the potential equation is also a solution of Laplace’s equation.
The superposition principle applies.
4. Unsteady one-dimensional flows
1 2
φxx = φ φ xx + φ tt + 2φ x φ tx
a2 x
or
1 2
φxx = vx φ xx + φ tt + 2vx φ tx .
a2
5. Unsteady three-dimensional flows, vi a 2 :
1
∇ 2φ = φtt . (1.9)
a2
We have in this case the weak waves equation, which governs the evolution of
small perturbations propagating in the medium with velocity a, namely, a is the
sound velocity (see also Sect. 4).
3 A Classification of Partial Differential Equations
We consider a steady two-dimensional potential flow and assume that φ and grad φ
are known on a given curve Σ. We rewrite Eq. 1.7 as:
Aφxx + Bφxy + Cφyy + D = 0,
with independent variables x and y, dependent variable φ, and coefficients A, B,

C, and D, as functions of a, x, y, φ, φx , and φy . Let us assume that x = x(σ ) and
y = y(σ ). We denote:
p = φx r = φxx
q = φy s = φxy t = φyy .
We have, then:
Ar + Bs + Ct = −D
dp dφx dx dy
= = r+ s
dσ dσ dσ dσ
3 A Classification of Partial Differential Equations 7
dq dφy dx dy
= = s+ t.
dσ dσ dσ dσ
By rewriting the above system of equations in matrix form:
⎛ ⎞ ⎛ ⎞
A B C ⎛ ⎞ −D
⎜ dx dy ⎟ r ⎜ dp ⎟
⎜ ⎟ ⎜ ⎟
⎜ dσ dσ 0 ⎟ ⎝ s ⎠ = ⎜ dσ ⎟ .
⎝ dx dy ⎠ ⎝ dq ⎠
0 t
dσ dσ dσ
Let us find conditions for the discontinuity of the velocity derivatives and, in
consequence the conditions for the divergence of higher order derivatives, φxx , φxy ,
and φyy . Let σ be the parameter of a characteristic curve, along which the highest
order derivatives namely the velocity derivatives are discontinuous. We prescribe
this condition by imposing that the determinant of the matrix of coefficients of the
above equation vanishes. This condition writes as:
2 2
dy dx dy dx
A −B +C = 0
dσ dσ dσ dσ
or, dividing the last equation by (dx/dσ )2 :

2
dy dy
A −B + C = 0,
dx dx
from where we have:
√
dy B± B 2 − 4AC
= .
dx 2A
The derivatives of the velocity field are continuous when the roots of the equation
are imaginaries, and discontinuous if real. Equation 1.7 may be classified in three
groups, depending on the value of the associated parameters:
1. Parabolic, if B 2 − 4AC = 0;
2. Hyperbolic, if B 2 − 4AC > 0;
3. Elliptic, if B 2 − 4AC < 0 (no characteristic curves exist in the field, along which
the derivatives of the velocity are discontinuous);
Some examples:
1. Laplace’s equation:
∂ 2φ ∂ 2φ
+ = 0.
∂x 2 ∂y 2
No real characteristics exist in the fields obeying Laplace’s equation. All the
derivatives are continuous.
2. One-dimensional, time-dependent temperature equation
∂T ∂ 2T
= .
∂t ∂x 2
This equation presents one family of characteristics. The equation is parabolic.
3. Wave equation:
∂ 2φ 2
2∂ φ
= a .
∂t 2 ∂x 2
This equation presents two families of characteristics, along the straight lines
dx/dt = ±a. The field always presents a a discontinuous region. The equation
is, in this case, hyperbolic.
4. In the case of Eq. 1.7, written in the form:

u2 − a 2 φxx + 2uvφxy + v 2 − a 2 φyy = 0,

where u = φx , v = φy , A = u2 − a 2 , B = 2uv, and C = v 2 − a 2 ,
we have:

B 2 − 4AC = a 2 u2 + v 2 − a 2 .

If |v|2 = u2 + v 2 < a 2 , i.e., if the flow regime is subsonic, the equation
and 2there are no discontinuities at the velocity derivatives. If |v| =
is 2
2elliptic
u + v > a , namely, if the flow is supersonic, B2 − 4AC > 0, the equation
2
is hyperbolic. In this case:
v a2 √
dy uv ± a 2 u2 + v 2 − a 2 ± 2 M2 − 1
= = u u = tan (α ± θ ) ,
dx u2 − a 2 a2
1− 2
u
(1.10)
where:
√
u2 + v 2 v a 1
M = tan α = sin θ = √ = .
a u u +v
2 2 M
If v = 0, Eq. 1.10 takes the following simplified form:
dy 1
= tan (±θ ) = ± √ ,
dx M2 − 1
4 Weak Waves: Sound Velocity 9
(a) (b) (c)
Fig. 1.1 Characteristic lines of a steady two-dimensional compressible flow. (a): A point
moving with subsonic velocity emits small perturbations that propagate faster than the point; no
discontinuities exist in the velocity derivatives. The equation of the potential is elliptic. (b): The
point moves with the sound velocity. A family of characteristics exists, along which the velocity
derivatives are discontinuous. The region ahead (at left) of the characteristics is the silent zone and
does not receive signals emitted by the point. The equation of the potential is parabolic. (c): The
point moves with supersonic velocity. The field admits two families of characteristics, along which
the velocity derivatives are discontinuous. Points located ahead of the characteristics are in the
zone of silence. The equation of the potential is hyperbolic. A small perturbation initially at the
point O stays in the characteristic line defining the Mach’s Cone and moves perpendicularly to it,
whereas the particle emitting perturbations move attached to the cone’s vertex
where θ is the angle between the characteristics and the velocity vector of a point
moving in the field. We also have:
√
a 1 M2 − 1
sin θ = = , and cos θ = .
u M M
Figure 1.1c shows that, at the end of an elapsed time equal to one, a perturbation
emitted by a point moving with velocity u covers a distance numerically equal to a,
while the point covers a distance numerically equal to u. From the same figure we
infer that sin θ = a/u = 1/M.
For a point moving with v = u, the characteristic lines along which the velocity
derivatives are discontinuous and the regions of the field receiving signals of small
perturbations emitted by the point are shown schematically in Fig. 1.1. The inner
region delimited by the characteristics is named as the Mach’s Cone.
4 Weak Waves: Sound Velocity
In the previous section we showed that the potential equation of compressible flows
(Eq. 1.5) reduces, in the case of the evolution of weak perturbations of rest states, to
the wave equation for the flow potential φ (Eq. 1.9). Weak perturbations of the rest
state in a gas are, by definition, sound waves. In addition, for a perfect gas, we have:
a 2 = γ RT .
We show now that not only perturbations of the rest potential, but also, of velocity,
specific mass and pressure perturbations do obey the equation of weak waves and
propagate with velocity a, namely, a is effectively the sound velocity.
The potential φ can be written in the form of a sum of the uniform, time-
independent one, φ̄, plus a perturbation, φ̃:
φ = φ̄ + φ̃,
with the uniform and time-independent φ̄, trivially obeying Eq. 1.9. The perturbation
term φ̃ also satisfies the same equation:
1
∇ 2 φ̃ = φ̃tt .
a2
Upon applying the gradient operator to this last equation we find:
∂ φ̃ 1 ∂ 2 ∂ φ̃
∇2 = 2 2 ,
∂xi a ∂t ∂xi
or
1 ∂ 2 ṽi
∇ 2 ṽi = , (1.11)
a 2 ∂t 2
where ṽi is the velocity perturbation, associated with φ̃. In order to prove that spe-
cific mass perturbations around a steady value ρ̄, we consider the three-dimensional
Euler’s equation applicable to a rest fluid submitted to a small perturbation. Specific
mass and pressure are written as:
ρ = ρ̄ + ρ̃
p = p̄ + p̃,
where p̄ and ρ̄ are the steady unperturbed values of pressure and specific mass,
whereas p̃ and ρ̃ stand for the time-dependent perturbations of the steady state
variables. Assuming isentropic flow and neglecting gravitational effects, we write
Euler’s equation in the form:

∂ ∂ ∂ρ
(ρ̄ + ρ̃) v̄i + ṽj + v̄j + ṽj (v̄i + ṽi ) = −a 2 .
∂t ∂xj ∂xi
Euler’s equations are now linearized. Reminding that the base state variables
are uniform in space and time independent, that perturbations are small, and,
additionally, that v̄i ≡ 0, we obtain the linearized isentropic Euler’s equations:
∂ ṽi ∂ ρ̃
ρ̄ + a2 = 0. (1.12)
∂t ∂xi
5 Solution of the Wave Equation in the Frequency Domain 11
By deriving Eq. 1.12 with respect to time and rearranging terms, and using Eq. 1.11
we obtain:
1 ∂ 2 ṽi 1 ∂ 2 ρ̃
∇ 2 ṽi = = − .
a 2 ∂t 2 ρ̄ ∂t ∂xi
Derivation of the above equation with respect to time leads to:
∂ ṽi 1 ∂ 3 ρ̃
∇2 = − . (1.13)
∂t ρ̄ ∂t 2 ∂xi
From Eq. 1.12 we obtain:
∂ ṽi a 2 ∂ ρ̃
= − . (1.14)
∂t ρ̄ ∂xi
Upon replacing the expression of ∂vi /∂t, given by Eq. 1.14 into Eq. 1.13 we find:
∂ρ 1 ∂ 3 ρ̃
∇2 = 2 2 . (1.15)
∂xi a ∂t ∂xi
Integration with respect to xi leads to the sought wave equation for the specific
mass perturbation:
1 ∂ 2 ρ̃
∇ 2ρ = . (1.16)
a 2 ∂t 2
The pressure perturbation p̃ also obeys a wave equation. In fact, for an isentropic
process we can write:
1
d ρ̃ = d p̃.
a2
Replacing the above expression of d ρ̃ in Eq. 1.16 leads to the wave equation for the
small pressure perturbations:
1 ∂ 2 p̃
∇ 2 p̃ = . (1.17)
a 2 ∂t 2
5 Solution of the Wave Equation in the Frequency Domain
In this section we describe the method for finding time periodic solutions of the
pressure perturbation equation. As a first step we assume that the solution of Eq. 1.17
may be written as a superposition of harmonic modes in the form:
p = p̂ ej ωt ,
where p̂ stems for the pressure scalar field in the frequency domain, √ p, for the
pressure field in the time domain. At the frequency domain j = −1. In order
to analyze the system response to periodic forcing, we consider the solutions
associated with a given frequency ω. Upon replacing the above solution in Eq. 1.16
and dividing the result by ej ωt we obtain a Helmholtz equation identical to the above
referred with time derivatives replaced by j ω:
κ 2 p + ∇ 2 p = 0, (1.18)
where κ 2 = 1 2
ω is the square of the wave number, and a0 the sound velocity. In the
a02
case of acoustic waves propagating in the air at 0 ◦ C, at the sea level, typical values
of the parameters are γ = 1.4, ρ0 = 1.29 kg/m3 , and the P0 = 1.01 × 105 n/m2 for
the atmospheric pressure, leading to a0 = 330 m/s.
Formulation and solution of the eigenvalue problem in the frequency domain in
one and two dimensions using the Finite Differences and Finite Elements Methods,
and using both the strong and the weak formulations are presented in the exercises
at the end of this chapter.
6 Problems
1. Formulation of the eigenvalue problem in the frequency domain, in one

dimension, using the strong form:
Show that the one-dimensional pressure wave equation writes in the form:
κ 2 p̂ + D 2 p̂ = 0,
where D = d/dx. Consider now the problem of finding if the natural frequencies
of a one-dimensional rectangular cavity, open at one end and close at the other,
is close to a reference frequency ω0 = 2πf0 . The problem also involves the
identification of the vibration modes associated with the natural frequencies.
Define the nondimensional wave numbers and natural frequencies, by κ ∗ =
2πf0∗ and f0∗ = ff0 , drop the asterisks, and rewrite the eigenvalue-eigenfunction
problem as:
D 2 p̂ = −ω2 p̂. (1.19)
The eigenvalues ω = κ/a02 represent the resonant angular frequencies of the

cavity, associated with the steady waves, of the transient solution. Boundary
conditions at a rigid surface with a local normal vector n require
n · ∇p = 0. (1.20)
6 Problems 13
Show that in the one-dimensional case, occurring, for instance, when we consider
the sound propagation in a tube closed at one end, boundary conditions reduce to:

∂p
=0 (for the closed end at x = 0),
∂x x=0
p(L) = 0 (for the open end at x = L).
2. Discrete formulation and resolution of the one-dimensional eigenvalue-

eigenfunction problem in Finite Differences:
Finite Differences discretization of the one-dimensional real eigenvalue problem
replaces the space derivatives by a discrete representation. If a centered second
order differencing scheme is adopted and node points are uniformly distributed,
xi = x0 + Δx i , i = 0, 1, 2, . . . , nx , the discrete formula reads:
d 2p pi−1 − 2pi + pi+i

= + O(Δx 2 ). (1.21)
dx 2 Δx 2
Verify that upon replacing the space derivative by Eq. 1.21, one obtains a linear
system of equations, written in matrix form, as:
Aph = −κ 2 Mph ,
where nonvanishing elements of the matrices A and M read:
1 2 1
A(i, i − 1) = , A(i, i) = − , A(i, i + 1) = , M(i, i) = − 1.
Δx 2 Δx 2 Δx 2
A and M are the rigidity and M mass matrices, respectively, and ph stems for
the pressure at the computational nodes.
Define a computational mesh, including information about the boundary
conditions and the medium properties, and solve the eigenvalue-eigenfunction
problem associated with Eq. 1.19, using the following algorithm:
clear;clc;
% Pre-processing: Domain Definition
nnodes=100; % number of nodes
Xmin=0; Xmax=1;
dx=(Xmax-Xmin)/(nnodes-1);
x=Xmin:dx:Xmax;
% Assembling matrices (operator d^2/dx^2)

for i=2:nnodes-1
A(i,i-1)=1/(dx*dx);
A(i,i)=-2/(dx*dx);
A(i,i+1)=1/(dx*dx);
M(i,i)=-1;
end
% Application of Dirichlet BCs

A(1,1)=1;
M(1,1)=0;
A(nnodes,nnodes)=1;
M(nnodes,nnodes)=0;
% Application of Neumann BCs

A(nnodes,nnodes)=1;
A(nnodes,nnodes-1)=-1;
% Solving eigenvalues problem

[V,D] = eigs(A,M,6,’sm’);
% Plotting first 4 eigenfunctions

plot(x,V(:,1)’,x,V(:,2)’,x,V(:,3)’,x,V(:,4)’);
xlabel(’domain lenght x’,’FontSize’, 20)
xtickformat(’%.1f’)
ylabel(’amplitude’,’FontSize’, 20)
ytickformat(’%.2f’)
set(gca,’FontSize’,20)
Validation Tests: Resonance in a Tube

Validate the methodology by evaluating the eigenvalues and eigenfunctions
associated with a steady wave in a tube, in order to confirm the exactitude of
the method. A possible algorithm written in Octave language, for defining the
computational mesh, assembling the matrices, with the boundary conditions,
and solving the problem is given below. Validate the above methodology by
evaluating the eigenvalues and eigenfunctions associated with a steady wave in a
tube, and show that modes present the form shown in Fig. 1.2.
Fig. 1.2 Eigenfunctions

associated with a steady
one-dimensional acoustic
wave in a tube open at the left
end and closed at x = L. The
results show that the
eigenfunctions are Fourier
modes satisfying the Dirichlet
BC at left and the Neumann
ones at right
6 Problems 15
3. Discrete formulation of the eigenvalue-eigenfunction problem in one dimen-

sion with the Finite Elements Method:
Galerkin formulation of eigenvalue-eigenfunction problems consists in choosing
base and weighing functions in finite dimensional spaces, P h and W h . The
discrete problem turns in finding ph ∈ P h and ω ∈ R such that for all w h ∈ W h :

Dph , Dw h = ωph , ωw h .
By using Finite Element canonical basis in Cartesian coordinates, show that the
following linear system, written in matrix form, is obtained:
AP h = −ω2 MP h ,
where:

Aij = − Dwih , Dwjh and M = wih , wjh
are matrices associated with the stiffness and the mass, respectively, and P h
represents the pressure at the computational nodes. Define the computational
mesh consisting of nodes and elements containing information on the boundary
conditions and medium properties according to the algorithm below:
clear;clc;
% Pre-processing: Domain Definition
kappa=1; Xmin=0; Xmax=1;
% Generation of nodes (coordinates)
nnodes=100; % Assigning nodes numbers
dx=(Xmax-Xmin)/(nnodes-1);
x=Xmin:dx:Xmax;
x1=Xmin;
x2=Xmax;
% Total number of elements
nele=nnodes-1;
% Conectivity matrix
ien=zeros(nele,2);
for e=1:nele
ien(e,1)=e;
ien(e,2)=e+1;
end
% BC, Dirichlet nodes
iccd=[1 nnodes];
uccd=[0 0];
% BC Newmann nodes
iccn=[];
uccn=[];
% Degrees of freedom
nccd=size(iccd,2);
nccn=size(iccn,2);
The vector iccd contains the nodes indices at the Dirichlet boundary.
Show that transforming the original eigenvalue-eigenfunction problem in a

generalized eigenvalue-eigenvector problem with eigenvalues ω2 and eigenvec-
tors P h further requires the application of discrete BCs as described below:
Dirichlet boundary conditions are straightly imposed in Finite Elements
formulations, being then denoted as essential. As a first step, boundary nodes
are identified. In one-dimensional problems, boundary nodes are the ones at the
domain extremes, i = 1 and i = n nodes. For a node i, at the system boundary
we have:
(a) Prescribe the condition P (i) = 0, by assign zeroes to all elements A and M
in line i;
(b) Impose that nodal values also vanish, by doing A(i, i) = 1.
Vanishing derivatives boundary conditions are implicitly imposed in Finite
Element formulations, being thus denoted as natural boundary conditions.
Verify that Dirichlet boundary conditions are properly assigned with the
following algorithm, written in Octave:
for e=1:nele
lenght=(x(ien(e,2))-x(ien(e,1)));
k=[1 -1;-1 1]/lenght; % stiffness matrix
m=[2 1;1 2]*lenght/6; % mass matrix
for ilocal=1:2
iglobal=ien(e,ilocal);
for jlocal=1:2
jglobal=ien(e,jlocal);
K(iglobal,jglobal)=K(iglobal,jglobal)+k(ilocal,jlocal);
M(iglobal,jglobal)=M(iglobal,jglobal)+kappa*m(ilocal,jlocal);
end
end
end
The indices of the Dirichlet boundary are stored in the vector idbcd.
The following procedure written in Octave language illustrates the implemen-
tation of BCs to matrices K and M:
% setting BCs (zeroing line and setting 1 to main diagonal)
for i=1:nccd
% zeroing line iccd[i]
K(iccd(i),:)=0;
M(iccd(i),:)=0;
% setting 1 to main diagonal
K(iccd(i),iccd(i))=1;
end;
% Solving eigenvalues problem

neigs=5;
[V,d] = eigs(K,M,neigs,1);
sqrt(d/4/pi/pi);
hold off

for i=1:neigs
6 Problems 17
plot(x,real(V(:,i)),x,imag(V(:,i)));
hold on
end

xlabel(’domain lenght x’,’FontSize’, 20)
ylabel(’amplitude’,’FontSize’, 20)
Validation Tests: Resonance in a Tube with Open Ends

Validate the above methodology by evaluating the eigenvalues and eigenfunc-
tions associated with a steady wave in a tube with open ends (p = 0) to
demonstrate the correctness of the method. Show that the eigenfunctions take
the form shown in Fig. 1.3.
4. Formulation of the real eigenvalue-eigenfunction problem in 2D:
Consider now a closed two-dimensional cavity and the problem of finding the
resonance frequency, near a frequency of interest, ω0 = 2πf0 , and the associated
vibration modes, known as standing or stationary waves. Write, in dimensional
form, κ ∗ = 2πf0∗ , where f0∗ = f/f0 is the nondimensional resonant frequency.
Show that dropping the asterisks leads to an eigenvalue-eigenfunction problem
given by:
∇ 2 p = −ω2 p.
The eigenvalues ω = κ represent the nondimensional resonant angular frequen-

cies of the cavity, and the associated standing modes are the eigenfunctions of the
Laplacian operator. Verify that, at an impermeable boundary, with a local normal
unitary vector n, boundary conditions can be written in the form:
n · ∇p = 0. (1.22)
Fig. 1.3 Eigenfunctions

associated with a steady wave
in a tube with open ends
(p = 0) to demonstrate the
correctness of the method.
The result confirms that the
eigenfunctions are Fourier
modes satisfying the Dirichlet
homogeneous boundary
conditions
Formulation of the Real Eigenvalue-Eigenfunction Problem in 2D, Through

the Finite Elements Method
Discrete formulation of the real eigenvalue-eigenfunction problem in 2D
through the Finite Elements Method
The Galerkin formulation of the two-dimensional eigenvalue-eigenfunction
problem consists in choosing base and weighing functions in finite dimensional
spaces P h and W h . The discrete problem turns out to finding ph ∈ P h and
ω ∈ R is such that, for all Wh ∈ W h

∇ph , ∇W h = − κph , κW h .
Upon using canonical Finite Element basis in Cartesian coordinates, one obtains
the following linear system in matrix form:
AP = −κ 2 MP ,
where A and M are the associated rigidity and mass matrices, respectively,
whereas P stems for p values at the computational nodes. In order to approx-
imate the original problem to a generalized eigenvalue-eigenvector problem with
eigenvalue κ 2 and eigenvector P , complete the problem statement by applying
the appropriate boundary conditions. In the case of a closed cavity, BCs consist
in prescribing impermeable walls, which are naturally satisfied in the boundaries
by the variational formulation.
Computer Implementation of the Finite Elements Scheme
Implementation of the main program and of the auxiliary functions in Octave to
run the above scheme is suggested below:
Main Program
% %% --------------- Main ------------------------------- %%%
% Mesh parameters
Ly=1; %%% vertical size
Lx=2; %%% horizontal size
Nx=21; Ny=11; # number of mesh points
X=zeros(Nx*Ny,1); Y=zeros(Nx*Ny,1);
dx=1/(Nx-1)*Lx; dy=1/(Ny-1)*Ly;
% Mesh generation
[TRI,X,Y] = meshing(Nx,Ny,dx,dy,X,Y);
% Assembling of matrices
[Ak,Mk,Dx,Dy] = assemblingmatrices(TRI,X,Y);
% Solution of the linear system
nmodes=15;
[V,D]= eigs(Ak,Mk,nmodes,’sm’);
% Plotting eigenvalue and eigenvector

for imode=2:11 % plotting 10 first modes
colormap jet
% eigen plot
subplot(1,2,1); trisurf(TRI,X,Y,V(:,imode)); shading interp
6 Problems 19
title([’Pressure field - f = ’ num2str(D(imode,imode))]);

set(get(gca,’title’),’Position’,[2.6 1.9])
pbaspect([1 1 1])
xlabel(’x’,’FontSize’, 16)
ylabel(’y’,’FontSize’, 16)
zlabel(’amplitude’,’FontSize’, 16)
ztickformat(’%.2f’)
% quiver plot
u=-Mk\(Dx*V(:,imode));v=-Mk\(Dy*V(:,imode));
subplot(1,2,2);quiver(X,Y,u,v);
title([’Velocity field - f = ’ num2str(D(imode,imode))]);
set(get(gca,’title’),’Position’,[0.7 1.19])
pbaspect([1 1 1])
xlabel(’x’,’FontSize’, 16)
xlim([-0.1 2.1])
ylabel(’y’,’FontSize’, 16)
ylim([-0.1 1.1])
drawnow
end
% %% --------------- End Main --------------------------- %%%
For the Mesh Generation

% %% ----------- Creating mesh: X,Y and TRI ------------- %%%
function [TRI,X,Y] = meshing(Nx,Ny,dx,dy,X,Y)
for i=1:Nx
for j=1:Ny
kk=i+(j-1)*Nx;
X(kk)=(i-1)*dx;
Y(kk)=(j-1)*dy;
end
end
% setting connectivity matrix TRI
TRI=zeros((Nx-1)*(Ny-1)*2,3);
nel=0;
for i=1:Nx-1;
for j=1:Ny-1;
kk=i+(j-1)*Nx;
nel=nel+1;
TRI(nel,:)=[kk kk+Nx+1 kk+Nx];
nel=nel+1;
TRI(nel,:)=[kk kk+1 kk+Nx+1];
end
end
end
% %% --------- End Creating mesh: X,Y and TRI ----------- %%%
Another random document with
no related content on Scribd:
“Yes; in the best way.” Snell touched his fingers lightly to his
discolored eye. “In the best way,” he repeated.
CHAPTER XV
MARCH WINDS BLOW
It was clear that Snell’s idea of the “best way” to punish Don and his
aunt was a way that would also punish Hawkins, with whom Snell
was now on the bitterest of hostile terms; the two soldiers neither
spoke nor so much as glanced at each other. But whatever Snell’s
plan was, he and Tom were slow in carrying it out.
No doubt they were busy with other things, for the month of March
began in a way that promised to keep the Redcoats and the Tories
occupied for some little time. On the night of the second the
Continental batteries opened fire on the town.
Don and his Aunt Martha were in bed when the firing began. For a
long while they lay listening to the crash, crash of the shells, which
seemed to be landing somewhere on the Common. They heard Snell
and Hawkins descend the stairs and pass out into the street; then
Aunt Martha went to her nephew’s room. “Donald, my boy,” she said,
“what can it mean?”
“It means that General Washington is preparing to drive out old
Howe and his men,” Don replied confidently.
Don was nearer right than his aunt supposed. The two following
nights the bombardment was repeated; it seemed that every gun in
all the forts, both friendly and hostile, was crashing forth and
illuminating the sky every few seconds.
And on the next morning, the fifth of March and the anniversary of
the Boston Massacre, the whole town—and especially the British
high-command—opened their eyes wide with amazement. Strong
fortifications had sprung up, as if by magic, on Dorchester Heights.
Grim black guns were pointing at the town; grim black guns
threatened the British fleet, which lay at anchor out in the harbor.
Later in the morning Jud came hurrying into Pudding Lane and
entered the house; he was trembling with excitement. “The time’s
come!” he cried. “Have you seen Dorchester Heights? The Redcoats
have either got to attack the Heights the way they did Bunker’s Hill,
or they’ve got to clear out. I hope they attack!”
“Goodness!” exclaimed Aunt Martha.
“It’s true,” declared Jud, “everything I’ve told you. The Tories are
scared silly!”
“Have you seen Tom Bullard?” asked Don.
“Yes; I passed him on the way. He didn’t seem scared, though—
maybe he’s just too stupid. He shook his fist at me, and he said,
‘You’d better keep away from Pudding Lane if you know what’s good
for you.’”
“What did he mean?” asked Aunt Martha.
“Nothing, I guess,” said Don. “He likes to talk, that’s all.”
Throughout the rest of the day the town was a place of keen
excitement. Howe, it seems, had decided to accept Washington’s
challenge and attack the Heights. He was a brave man, and his own
honor as well as the honor of England was at stake; he did not mean
to yield weakly to a band of “rebels.” He ordered twenty-four hundred
men to embark at once in transports; under the command of Lord
Percy, they were to land at Castle William, from which place they
were to attack Dorchester Heights when night fell.
Don and Jud saw the Redcoats assemble for the attack and then
march off. They saw Hawkins, tall, erect, well-composed and
confident; they saw Snell marching in another rank—and he seemed
anything but confident.
By noon virtually everyone in Boston and the vicinity expected to see
a terrible battle. It seemed inevitable, for both sides were preparing
for it. While the British were mustering for the attack on Dorchester
Heights, Washington was preparing to attack the British lines in
Boston. A fine detachment of four thousand troops were on parade in
Cambridge; under the command of Generals Putnam, Greene and
Sullivan they were all ready to embark in boats at the mouth of the
river. And, as at Bunker Hill, people had taken up points of vantage
on the tops of houses and on some of the near-by hills.
But somewhat to the disappointment of Don and Jud the two armies
were not to meet. In the afternoon the wind blew furiously, and a wild
destructive surf crashed and pounded on the shores; no boat could
possibly land with safety in such a storm. Great limbs cracked and
crashed on the Common, and boards and shingles were torn from
many of the houses. The two boys, hurrying along Long Acre,
narrowly missed being struck with a pile of stones that came
tumbling from a chimney on a house near Rawson’s Lane.
“Say, that was close, wasn’t it?” exclaimed Jud. “A little more, Don,
and you and I might have been killed.”
Don laughed. “Come on, Jud, and let’s get home. Just look how dark
it’s getting! It’s going to rain too.”
That evening the rain came down in torrents, and the wind continued
to blow with unabated fury. And the next day, the sixth, found the
waves in the harbor high and confused. Both armies waited; and
Washington’s men strengthened their fortifications.
The next day Howe found himself in a critical and perplexing
situation. His army was at the mercy of the Continental batteries, and
the fleet was unable to ride in safety in the harbor. To remain in
Boston would be to expose his men to the greatest danger; to
withdraw would be to lose much valuable property. But Howe was
first of all a soldier, and after a hurried council he determined to
withdraw to save his army. Preparations began at once.
“They’re going, Aunt Martha!” cried Don, bursting in upon his aunt.
“The Redcoats are going to leave the town!”
“And what will they do to the town before they go? O Donald, what
will they do?”
“I don’t know,” replied Don thoughtfully. “They could do a lot of bad
things, I suppose, but, Aunt Martha, I don’t think they’ll do anything
very bad. I tell you it won’t be well for them if they set fire to any
buildings.”
“That’s what I’m afraid of most of all,” said his aunt.
Many other persons besides Aunt Martha were worried about the
fate of Boston. In the Continental army itself there was many a
soldier who wondered what would become of his home and of his
relatives who had refused to leave the town.
After a little group of prominent citizens had sent a petition to
Washington, begging him to take no measures that would injure the
town, the two armies reached a tacit understanding that the British
might embark without the Continentals molesting them.
Nevertheless, the American army held itself in readiness to act in
case the enemy did any damage.
Meanwhile, Washington was strengthening his defenses round the
town. On the evening of the ninth he sent a strong detachment to
plant a battery on Nook’s Hill to threaten the fleet; but the British
learned of his purpose, and almost all night Don and his aunt lay
awake and listened to the roar of cannon.
The next day Howe issued a printed proclamation that almost
caused Aunt Martha to lose heart. Don and Jud brought her word of
it.
“All linen and woolen goods have to be turned over to Crean Brush,
the Tory,” whispered Don—for Snell was up-stairs. “Old Howe knows
there’s stuff hidden in the town that our men can use. That’s why he
wants it.”
Aunt Martha glanced involuntarily at the door to the cellar. “We’d
best give it up, Donald,” she said. “I’d hoped we could keep it, but I
see now that we can’t. Oh, what a foolish woman I was!”
“No, Aunt Martha—no!” Don’s voice trembled in spite of himself.
“Nobody knows we have the stuff, and the Redcoats can’t possibly
search every cellar.”
“Don is right, Aunt Martha,” whispered Jud. “Don’t you turn it over to
them!”
“But if they come and search——” Aunt Martha checked herself
suddenly, for Snell was coming down the stairs.
Without looking to right or left, the Redcoat crossed the room and
went out on the street.
“Did he hear us?” asked Don’s aunt.
“Not likely,” replied Jud. “Now don’t you say one word about that stuff
in the cellar.”
Aunt Martha shook her head in uncertainty, but she finally decided to
do as the boys had advised.
The next day Crean Brush began his work of searching for hidden
supplies. Stores were broken into, and goods of all sorts were
carried off in violation of strict orders that Howe had issued. Lawless
bands of soldiers, sailors, marines and Tories went from house to
house and took what pleased them. And while they were doing that,
the army was transporting its equipment to the water-front to be
shipped aboard the vessels.
All day Don and his aunt remained in the house, anxiously expecting
every minute to hear the sound of Crean Brush’s men outside. Jud
did not put in an appearance until after dark, and then he remained
only for a few moments to say that a searching party had come to his
house but had found nothing. “If they had,” he added, “Ma and I
would have been as surprised as they, I guess.”
Don and his aunt laughed. Before Jud went away he got Don to one
side. “Say, Don,” he whispered, “you’ve got powder in the cellar
along with that other stuff, haven’t you?”
“Yes, a little,” Don replied.
“Well,” said Jud, “if I were you I’d move it somewhere else.”
“Why would you do that?”
“Well, at the rate Crean Brush and his gangs are going they’re pretty
sure to reach your house sooner or later; and if they search it and
find that powder—well, I don’t know what they won’t do.”
“I guess they’ll do enough even if they just find the cloth and stuff,”
said Don. “So it seems to me I might as well leave it all together.”
“No, Don; do as I tell you!” Jud’s voice was low and excited.
Don laughed. “I never knew you to be so cautious before, Jud.”
But Jud seized his friend’s arm. “Take my advice for once and do it,”
he urged. “I’ll help you move it now if you like.”
“No, not now,” replied Don. “Maybe later.”
“All right, then.” And Jud hurried out into the night.
CHAPTER XVI
CREAN BRUSH’S MEN
Don said nothing to his aunt about the powder in the cellar. Indeed
after Jud had gone he thought little of it and of the advice his
companion had given. Don and his aunt waited until Snell and
Hawkins had gone up-stairs, and then Aunt Martha said:
“Well, Donald, I think we’re almost at the end of the story.”
“What story?” asked Don.
Aunt Martha smiled. “I merely meant,” she replied, “that in a few
more days we’ll be all through with our suffering—or else there will
be more suffering, far more terrible than some of us can bear
perhaps.”
“You’re still afraid they’ll burn the town?”
“I can’t get it off my mind. Just look at Charlestown across the water.
What a snug little place to live in it used to be—and just see it now!”
Don was silent for a few moments. “Everything has gone pretty well
so far,” he said at last.
“And maybe before long we’ll see Uncle David and Glen.”
“O Donald, I’ve prayed for it!”
“I certainly wish that one or the other were here now.” Don was
thinking of Crean Brush and of his lawless men.
“Ah, yes. Well, we’d best go to bed now. Another night—another
night.”
“Yes, and before you know it General Washington will be here, and
the Redcoats will be on the water.”
Up-stairs in his room, Don lay for some time listening to the sound of
firing that seemed to come from the direction of Noddles Island. The
night was dark, and a strong wind was blowing against the little
windows. From across the hall came the sounds of snoring and of
heavy breathing; apparently both Snell and Hawkins were asleep.
Don closed his eyes and lay back on the pillow; but the position was
uncomfortable, and he turned on his side. That position also
uncomfortable, and he turned on his other side. Then his foot began
to itch, then his back, then his neck. He could not sleep.
At last he sat up in bed. Now he could hear the regular breathing of
his aunt; no doubt she was exhausted with the day’s worry. Once
more he tried to get to sleep, but it was of no use. He raised himself
on his elbow. “Now what in thunder ails me?” he thought.
There was something—something that somebody had said. What
was it? The next instant he thought of Jud and of what he had said
about the powder. “That’s it!” he said to himself. “What if Crean
Brush and his men should find it in the cellar and, drunk as some of
them were likely to be, touch a light to it!”
The thought made him spring part way out of bed. Aunt Martha was
still breathing regularly. That was enough for Don to make up his
mind.
He began softly to dress. The house was cold, and he shivered as
he put on his shirt and his trousers. In a few minutes he was all
dressed except for his shoes. Then he made his way cautiously to
the head of the stairs. Once he stepped on a loose, squeaky board
and heard his aunt turn and sigh; but she did not waken. Neither did
either of the soldiers.
Down the steep stairs Don went on all fours. In the kitchen he found
the candlestick, but he did not light it until he had opened the door to
the cellar. Half-way down the old steps he paused, undecided
whether to go the rest of the way. Then he took another step, but it
required courage. The flickering light of the candle sent grotesque,
ghostlike shadows dancing along the walls, like great unearthly black
vultures.
He wondered whether he were doing right and then wished that Jud
were with him. But, taking a fresh grip on himself, he went the rest of
the way.
Trembling with nervousness, he set the candle on a box and looked
about him. All around lay the goods that David Hollis had bought in a
hasty moment—large bales and small bales piled side by side and
on top of one another. With shaking fingers Don examined them,
going quickly from one to another. Then suddenly he came upon the
powder; there were one small keg and seven canvas bags of it lying
close to the foot of the steps.
He lifted the keg and then lifted one of the bags; the keg was much
the heavier. “Now what shall I do with the stuff?” he wondered.
For a few moments he stood in deep thought. The old cellar was
cold and damp, and a draft from somewhere was stirring the flame of
the candle. “I know,” he said at last and bent over the keg again.
With an effort he lifted it and started up the stairs. In a moment or
two he no longer felt cold. It was no easy task to get that heavy keg
up the stairs. From step to step he half rolled, half lifted it, and in a
few minutes he was sweating with the exertion. Another thing that
made the work hard was that he did not dare make any noise.
At last he got the keg to the top, and then after a brief rest he carried
it through the room to the back shed, the door to which had only a
latch. There he found another candle, and lighting it, set it on the
floor. Five minutes later he had the keg hidden well at the back of the
woodpile.
Then he returned for the bags. One at a time he carried them—all
seven of them—up the steps and stowed them close to the keg.
Having covered them well with the wood and having snuffed both
candles, he returned to his room and began hastily to undress. He
was congratulating himself on not having disturbed anyone when he
heard the voice of his aunt:
“Donald, are you awake?”
Don paused in the act of removing his shirt. He did not reply at once.
“Donald!”
“Uh-hm,” said Don.
“There—you are awake!”
“Didn’t you hear a noise down-stairs a few minutes ago?”
“Noise? H’m—what noise?” Don was in bed by this time and had the
covers well round his head.
He heard his aunt sigh heavily. How could her nephew sleep so
soundly? The good woman was really sorry that she had wakened
him!
It was not long before Don was asleep indeed. Nor did he waken
when Snell and Hawkins descended the stairs in the morning. Aunt
Martha had to call him four times before he roused and crawled
sleepily from his bed.
“My goodness,” said his aunt as she was putting the breakfast on the
table, “you’re surely a sleepyhead this morning, Donald Alden. Ah,
well, you’re a growing boy, and you need your rest.”
Don grinned up at her. “You have a speck on your specs, Aunt
Martha.”
“Donald!”
“A speck of dust on your spectacles, Aunt Martha.”
His aunt hastily removed the speck with the corner of her apron.
“Now just see that candle,” she said. “I thought it was just yesterday
that I put a fresh one in the stick—but see how short it is now!”
Don examined the candle with great care, as if to find out what had
become of the rest of it. “Why, it seems that——” he began and then
sprang to his feet.
From the street came the sound of shouting and of heavy footsteps
on the cobblestones.
“O Donald, they’ve come. It’s—it’s——”
“Now, you be easy, Aunt Martha,” Don interrupted her.
Though he spoke calmly he was anything but calm in his mind. He
went to the door, and just as he reached it someone pounded heavily
on the outside.
“Open the door, Donald,” said Aunt Martha, “or they’ll beat it down.”
Don flung the door open and to his great astonishment looked full
into the leering face of Tom Bullard. Beside him were three of Crean
Brush’s men, and behind them, grinning insolently, was the Redcoat
Snell. In a moment all were inside, and Snell was striding toward the
door to the cellar. “We’ll find something this time, boys!” he said
exultantly.
“Gentlemen, what is it you wish?” It was the voice of Aunt Martha,
and Don, glancing at her as she stood slight but well poised beside
the fireplace, thought she looked fully ten years younger.
There was something in her voice that made everyone turn and look
at her. “A-hem,” began one of the Tories—a big fellow who obviously
was the leader. “A-hem, we’ve come to search your house.”
“Yes,” said Snell, “we’ve come to get that powder which you’ve got in
the cellar.” With his bayonet he began to pry at the lock on the cellar
door.
Aunt Martha looked helplessly at her nephew. Tom Bullard, standing
near the door, made a sneering remark to the Tory beside him, and
Don clenched his fists and started for him. But he had taken only two
steps when he checked himself and turned to the leader. “You’ve no
right in that cellar!” he cried. “You’ve no right in this house!”
“Hold your young tongue,” said the Tory sharply. “There’s powder in
this cellar, and we know it. That’s what we want, and that’s what
we’re a-goin’ to get.”
“There’s not a grain of powder in the cellar,” Don replied.
Aunt Martha’s eyebrows lifted in astonishment; never in her life had
she known her nephew to tell an untruth, even in fun.
“No powder?” repeated the Tory. “Well, now that’s curious—very
curious—because both these fellows say there is.” He indicated
Snell and Tom.
“I’ll stake my life on it,” said Tom, stepping forward and throwing out
his chest.
“And I’ll stake mine,” said Snell.
“Well, hurry up and get that lock off, and we’ll soon see,” said the
leader.
Snell inserted the bayonet and gave a wrench. Don was thinking, not
of the powder, but of the bales of cloth at the foot of the stairs. In a
few minutes they would find them, and then things would go hard
with him and his aunt. Well, he had done his best, but what wouldn’t
he have done to keep them out of the cellar altogether!
“Blasted lock!” muttered Snell and gave another fierce wrench; there
was a sharp crack, and his bayonet was in two pieces.
Infuriated, the Redcoat hacked away with the short end that was in
his hand, and in a few moments the lock clattered to the floor. He
had opened the door and was about to go down when a sharp
command behind him made him turn as if he had seen a ghost.
“Snell, you hound, what does this mean!” Harry Hawkins, gun in
hand, crossed the threshold; he had just returned from the drill
grounds.
Snell’s face had gone suddenly white, and he only stood and looked.
“It means,” said the leader, “that we’re about to get some ammunition
that these rebels have hidden in the cellar.”
“It’s not true, sir!” cried Don, turning to Hawkins. “It’s not true. There
is no ammunition in the cellar—not a speck!”
Hawkins looked steadily at Aunt Martha. “That is true, I suppose?”
he inquired.
“My nephew has never told a lie in his life, and, sir, he—he is telling
the truth now. There is no ammunition in the cellar.”
“They’re both lying——” Tom Bullard stopped as abruptly as he had
begun as Hawkins whirled and faced him.
For a long moment no one spoke; then Aunt Martha addressed
Hawkins: “These men have taken it upon themselves to enter my
house unbidden. Five men against one boy and a woman! They
have no right here——”
“Oh, enough of that!” cried the leader and strode toward the cellar
door.
“Halt where you are!” exclaimed Hawkins, and as the Tory hesitated
the soldier raised his gun a few inches. “Let me see your orders.”
“Orders! Orders to search a rebel’s house?”
“Now, see here,”—Hawkins’s voice was hard and cold, and his eyes
were like points of fire—“this thing has gone about as far as I want to
see it go. I’ll stand sponsor for the boy and the woman—and I’ve got
a good reason for doing it. Now, my friends, you’ll oblige me by
leaving the house——”
“Why—why, you don’t mean to say——” began the leader.
“At once,” finished Hawkins and tapped the stock of his musket.
Tom Bullard was already outside the door, but Snell and the three
Tories did not move. Whereupon Hawkins stepped swiftly to the
cellar door and, slamming it shut, quickly drew his bayonet and
affixed it to the end of his piece.
“By heaven, you’ll hear of this!” cried the leader and backed slowly
across the room. “I promise you I’ll see you in the guard-house
before nightfall!”
“And,” added another, “we’ll be here again, and we’ll bring Brush,
himself, along.”
Hawkins made no answer but followed the three across the room
and, when they had gone out, held the door open for Snell, who lost
no time in joining them. The sudden turn of affairs had left the fellow
speechless, for he had expected to find the powder and then to
accuse Hawkins of knowing that it was hidden in the cellar.
“Oh, sir,” exclaimed Don a few moments later, “it’s true, what I told
you, every word of it, but, oh——”
“Say no more,” interrupted Hawkins, smiling. “Say no more. I don’t
doubt your word; and if I had I should have stopped them, Tories as
they are. But had they been the King’s men, I should not have
interfered in any circumstances.”
“But you’ve rendered us a great service——” Aunt Martha began.
“It is nothing compared with the service your nephew once rendered
me. I owe him my life, and I trust that sometime we may meet again
—in better days.” Hawkins turned and walked to the stairs.
Later in the afternoon Don explained to his aunt what he had done
with the powder the night before, and a look of relief came into her
tired eyes. “I knew there must be some explanation,” she said
simply. “And,” she added, smiling slightly, “that accounts for the noise
I heard last night and for the shortness of the candle.”
“Do you suppose they’ll return?” asked Don.
“I’m afraid so,” his aunt replied.
CHAPTER XVII
DON MEETS GENERAL WASHINGTON
Crean Brush’s men did not return to the house in Pudding Lane;
they had more than enough to do in the excitement of the
withdrawal. Nor did Snell give any further trouble; no doubt the fellow
feared the strong arm of Hawkins.
On the sixteenth of the month both Redcoats carried all their
equipment from the house and did not return. All that Hawkins said
as he left the room was “Better days, young sire; better days, my
good woman.”
“Ah, yes, let us hope for them,” replied Aunt Martha.
Don only smiled, and Hawkins, as he closed the door, smiled in
return. That was the last that Don ever saw of him during the war.
The following day, which was Sunday, the Redcoats began to
embark; and not only the soldiers left the town, but the Tories also.
Don and Jud caught a glimpse of Tom Bullard and his father carrying
some of their household effects down King Street. The faces of both
Tories showed anger and mortification.
“Come on,” whispered Jud, “let’s get ahead of ’em and then turn and
give ’em a yell. We’ll never see them again.”
“No, Jud,” Don replied, “I’m just a little sorry for them. Oh, yes, I
know Tom’s acted mean, but just think what’s happening to him and
his father; they’re going to Halifax, so I’ve heard, and all they can
take along is just that little bit of stuff they’re carrying. Their fine
house up on Hanover Street is lost, and they’ll never get it again,
because they daren’t ever return.”
Jud did not reply but glanced at his companion sidewise. And so the
two boys stood and watched their enemy until he and his father had
disappeared among the throngs of Redcoats and Tories at the foot of
the street.
The last boatload of soldiers and refugees had not been long away
from the shore when the Continental soldiers entered the town by
way of the Neck and by boats across the river. The boys spied one
of the first patrols on the southern end of the Common and hastened
toward them.
“I want to find out first of all about Uncle Dave and Glen,” Don said to
Jud.
But none of the men in the patrol knew either of the two men. Kindly
fellows they were, all of them, and they laughed and joked with the
boys and with one another as they marched along toward the Mall.
“Say!” exclaimed Jud when they had gone past. “I’m so glad to see
those buff and blue uniforms I can hardly say how I feel. I feel as if
I’d burst!”
“Me, too,” said Don, “except that I almost feel like—well, like when
you’re so happy it makes the tears come into your eyes. Look, here
come some more of our men!”
Probably most of the good people of Boston felt as Don and Jud felt;
certainly there were many who shed tears of joy as they stood in
their doorways and watched the various detachments of
Continentals arriving. There was good reason for the tears, for the
people who shed them had suffered like martyrs during long months
of privation, insult and oppression—to say nothing of disease, for
smallpox had broken out in the poorer parts of the town.
The first words that greeted Don as he entered his aunt’s house
were, “Donald, my boy, did—did you see your uncle?”
“No, Aunt Martha. I asked at least a score of our men about him, but
none of them seemed to know him. But, O Aunt Martha, ain’t it fine!
The Redcoats are gone!”
“When I’ve seen your uncle I shall rejoice,” his aunt replied and
turned quickly away.
One thing that annoyed Don the following day was that he failed to
see General Washington, who had entered the town and had dined
with Mr. James Bowdoin at the home of Mr. Erving, both of whom
were friends of Don’s uncle. Nor did Don see Washington the next
day, for the general had returned to Cambridge.
On Wednesday, the twentieth, the main body of the Continental
troops entered the town, with flags flying and drums beating.
“Watch out for my uncle and for Glen Drake,” Don said to Jud as the
two boys stood on a crowded street corner waiting for the head of
the column to appear.
“Yes, and you keep your eyes open for my father and for my
brothers.”
From far off came the sound of drums and fifes. The crowd at the
corner, mostly boys and women, moved uneasily. “It’s Yankee
Doodle they’re playing,” whispered Jud. “Say, doesn’t that sound
good!”
“It surely does!” agreed Don.
In a few minutes the regular tramp, tramp of marching feet reached
the ears of the eager little group.
“Here they are!”
A cavalcade of horses, white, black and chestnut, had turned a
corner. Behind them came the foot soldiers, resplendent in buff and
blue, ruddy of face, keen of eye.
“Hurrah! Hurrah!” Don and Jud swung their caps high into the air.
“Hurrah!” cried the rest of the little group.
But for the most part the main body of Continentals were greeted
with few cheers. The people of the besieged town had suffered too
much under Howe and the British; most of the inhabitants remained
at doors and windows and were content to wave their hands.
“There he is!” cried Jud.
“Who? Who?”
“General Washington! See, there on the big horse! Don, just look
how——”
But Don was not listening. All his attention was given to the man who
sat with such ease and dignity on the big horse. Never had he seen
anyone who looked so thoroughly like his ideal of a soldier. Tall and
well-proportioned, the general looked truly noble and majestic. His
coat was blue with buff facings, and on each shoulder he wore a rich
epaulette. His under dress was buff, and he wore a black cockade in
his hat. At his side hung an elegant small-sword.
The cavalcade swung past, and the two boys turned to each other at
the same time. “Jinks!” exclaimed Jud. “Wasn’t he fine!”
“Fine!” echoed Don and with a deep sigh turned again to view the
troops.
Several ranks of foot soldiers had already passed, but the boys
sharply scrutinized those who were approaching. Company after
company swung past. Then Jud suddenly spied his father and the
next instant one of his brothers. Both recognized him and smiled as
they passed. A few minutes later he spied the other brother.
Don was worried; not a man had he seen who looked in the least like
his uncle or Glen. Company after company, regiment after regiment,
marched by, and somehow Don felt his lips beginning to quiver.
“It’s too bad, Don,” said Jud. “I’d hoped we’d see Glen and your
uncle right off. Here’s the end of the column. Maybe they’ll come
later.”
Don made no answer; he was wondering how he could tell his aunt
that Uncle David had not entered with the troops. He bit his under lip.
Maybe his uncle’s wound had not healed. Maybe——
“I’ll see you later, Don,” Jud was saying. “Won’t Ma be glad when I
tell her!”
Don made his way dejectedly to the little house in Pudding Lane. He
could just picture his aunt’s face when he told her the news. He
opened the door and with head down stepped inside; the next
instant, when he lifted his eyes, he could hardly believe what he saw.
There, standing beside his aunt near the fireplace, resplendent in a
captain’s uniform was—David Hollis! His arm was round Aunt
Martha’s waist, and she was laughing and crying both at the same
time. And there in one corner of the room, looking almost as he had
looked when Don had first seen him, was Glen Drake!
“Donald, my lad!”
Don felt the breath almost squeezed from his body, for his uncle was
a big man. And then he felt the bones in his hand crunch as the old
trapper greeted him.
“Oh, this seems almost too good to be true!” Aunt Martha was
saying.
For the next fifteen minutes questions and answers followed one
another in quick succession. Then at last Don asked gravely: “Uncle
David, where did you come from? I never saw you in the column.”
David Hollis laughed heartily. “I saw you, though; my company led
the column. But you were so interested in our general that you didn’t
have eyes for anybody else.”
“Never mind, Don,” said Glen; “he’s a fine man to look at, the general
is.”
“What a hard time you’ve had here!” said Uncle David. “Your aunt
has been telling me. My boy, I’m proud of you for the way you’ve
acted and especially for the way you kept the Redcoats from getting
that stuff in the cellar.”
“Yes, Don, you sure played the fox that time,” said Glen. “And now
that the Redcoats have gone, I’m thinking you and I and that other
boy will be able to do a little trapping in the woods together.”
“Now, David,” said Aunt Martha, “what are we going to do with the
stuff in the cellar? I don’t want it to remain there; truly I don’t.”
David Hollis laughed and turned to his nephew. “It’s yours, Donald,”
he said; “you’ve earned the right to it, I think; do with it as you wish.
Perhaps you can sell it.”
“Me—sell it!” exclaimed Don.

PDF An Introduction To Compressible Flows With Applications Quasi One Dimensional Approximation and General Formulation For Subsonic Transonic and Supersonic Flows Jose Pontes Ebook Full Chapter

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PDF An Introduction To Compressible Flows With Applications Quasi One Dimensional Approximation and General Formulation For Subsonic Transonic and Supersonic Flows Jose Pontes Ebook Full Chapter

Uploaded by

Copyright:

Available Formats

An Introduction to Compressible Flows

with Applications Quasi One

Calculus of Variations An Introduction to the One

Flows of Non Smooth Vector Fields and Degenerate

Computational Techniques for Multiphase Flows 2nd

Analytic Solutions for Flows Through Cascades Peter

Introduction to the Variational Formulation in

Mathematical Modeling of Unsteady Inviscid Flows Jeff

Design Synthesis and Applications of One Dimensional

New Trends and Results in Mathematical Description of

SpringerBriefs in Mathematics showcases expositions in all areas of mathematics

More information about this series at http://www.springer.com/series/10030

Luis Gustavo Nonato

The SBMAC SpringerBriefs series publishes relevant contributions in the fields

The Sociedade Brasileira de Matemática Aplicada e Computacional (Brazilian

ISSN 2191-8198 ISSN 2191-8201 (electronic)

Mathematics Subject Classification: 76B07, 76G25, 76J20

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019

This book is designed as an introduction to the problem of fluid flows where

1 Compressible Potential Flows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Fig. 1.1 Characteristics of potential compressible flows . . . . . . . . . . . . . . . . . . . . . . 9

This chapter deals with compressible irrotational isentropic flows to which a

2 The Equation of Compressible Potential Flows

In addition, irrotational flows satisfy Bernoulli’s equation, given by Eq. 5.11.

© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2019 1

the evolution equation of the time-dependent potential associated with compressible

where v 2 = v · v. In the case of irrotational flows, this equation becomes:

Expressing the velocity components as functions of the velocity potential, vi =

Upon integrating this last equation, we have:

We derive this last equation with respect to time, to obtain:

At this point, we introduce the hypothesis of isentropic flow, by further assuming

simplifies and takes the form:

The term a 2 = (∂p/∂ρ)s has units of square of a velocity. In fact,

In the case of perfect gases, we have:

Upon multiplying Euler’s equation by vi and mentioning that vi ∂vi /∂t =

By replacing 1/2 ∂v 2 /∂t = vi ∂vi /∂t we have:

∂vi ∂vi ∂vi ∂ 2φ

We replace now vi = ∂φ/∂xi to obtain:

2. Solids moving with steady velocity U:

This condition may also be written as: n · (v − U) = 0, where v is local velocity.

(φx2 − a 2 )φxx + (φy2 − a 2 )φyy + 2φx φy φxy = 0 (1.7)

(u2 − a 2 )φxx + (v 2 − a 2 )φyy + 2vx vy φxy = 0,

(1 − M 2 )φxx + φyy + φzz = 0.

φxx + φyy + φzz = 0.

We stress that the hypothesis of incompressible flows is equivalent to assume that

equation, followed by a post-processing where the velocity components are

5. Unsteady three-dimensional flows, vi a 2 :

3 A Classification of Partial Differential Equations

Aφxx + Bφxy + Cφyy + D = 0,

with independent variables x and y, dependent variable φ, and coefficients A, B,

or, dividing the last equation by (dx/dσ )2 :

2. One-dimensional, time-dependent temperature equation

is hyperbolic. In this case:

If v = 0, Eq. 1.10 takes the following simplified form:

(a) (b) (c)

4 Weak Waves: Sound Velocity

Derivation of the above equation with respect to time leads to:

5 Solution of the Wave Equation in the Frequency Domain

1. Formulation of the eigenvalue problem in the frequency domain, in one

D 2 p̂ = −ω2 p̂. (1.19)

The eigenvalues ω = κ/a02 represent the resonant angular frequencies of the

2. Discrete formulation and resolution of the one-dimensional eigenvalue-

d 2p pi−1 − 2pi + pi+i