Fundamentals of Compressible Fluid Mechanics

Genick Bar–Meir, Ph. D. 1107 16th Ave S. E. Minneapolis, MN 55414-2411 email:barmeir@gmail.com

Copyright © 2006, 2005, and 2004 by Genick Bar-Meir See the file copying.fdl or copyright.tex for copying conditions. Version (0.4.4.2 aka 0.4.4.1j May 21, 2007)

‘We are like dwarfs sitting on the shoulders of giants”

from The Metalogicon by John in 1159

CONTENTS

GNU Free Documentation License . . . . . . . . . . . . . . . . 1. APPLICABILITY AND DEFINITIONS . . . . . . . . . . 2. VERBATIM COPYING . . . . . . . . . . . . . . . . . . 3. COPYING IN QUANTITY . . . . . . . . . . . . . . . . . 4. MODIFICATIONS . . . . . . . . . . . . . . . . . . . . . 5. COMBINING DOCUMENTS . . . . . . . . . . . . . . . 6. COLLECTIONS OF DOCUMENTS . . . . . . . . . . . 7. AGGREGATION WITH INDEPENDENT WORKS . . . 8. TRANSLATION . . . . . . . . . . . . . . . . . . . . . . 9. TERMINATION . . . . . . . . . . . . . . . . . . . . . . 10. FUTURE REVISIONS OF THIS LICENSE . . . . . . . ADDENDUM: How to use this License for your documents Potto Project License . . . . . . . . . . . . . . . . . . . . . . . How to contribute to this book . . . . . . . . . . . . . . . . . . Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . John Martones . . . . . . . . . . . . . . . . . . . . . . . . Grigory Toker . . . . . . . . . . . . . . . . . . . . . . . . . Ralph Menikoff . . . . . . . . . . . . . . . . . . . . . . . . Your name here . . . . . . . . . . . . . . . . . . . . . . . Typo corrections and other ”minor” contributions . . . . . Version 0.4.3 Sep. 15, 2006 . . . . . . . . . . . . . . . . . . . . Version 0.4.2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Version 0.4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Version 0.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Version 4.3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Version 4.1.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Speed of Sound . . . . . . . . . . . . . . . . . . . . . . . iii

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

xvii xviii xix xix xx xxii xxii xxiii xxiii xxiii xxiii xxiv xxv xxvii xxvii xxvii xxviii xxviii xxviii xxviii xxxv xxxv xxxvi xxxvi xli xlii xlvi

iv Stagnation effects . . . . . . . . . . . . . . . . . . . . Nozzle . . . . . . . . . . . . . . . . . . . . . . . . . . Normal Shock . . . . . . . . . . . . . . . . . . . . . . . Isothermal Flow . . . . . . . . . . . . . . . . . . . . . . Fanno Flow . . . . . . . . . . . . . . . . . . . . . . . . Rayleigh Flow . . . . . . . . . . . . . . . . . . . . . . . Evacuation and filling semi rigid Chambers . . . . . . Evacuating and filling chambers under external forces Oblique Shock . . . . . . . . . . . . . . . . . . . . . . Prandtl–Meyer . . . . . . . . . . . . . . . . . . . . . . Transient problem . . . . . . . . . . . . . . . . . . . . 1 Introduction 1.1 What is Compressible Flow ? . . . . . . . . . . . . 1.2 Why Compressible Flow is Important? . . . . . . . 1.3 Historical Background . . . . . . . . . . . . . . . . 1.3.1 Early Developments . . . . . . . . . . . . . 1.3.2 The shock wave puzzle . . . . . . . . . . . 1.3.3 Choking Flow . . . . . . . . . . . . . . . . . 1.3.4 External flow . . . . . . . . . . . . . . . . . 1.3.5 Filling and Evacuating Gaseous Chambers 1.3.6 Biographies of Major Figures . . . . . . . . 2 Fundamentals of Basic Fluid Mechanics 2.1 Introduction . . . . . . . . . . . . . . . 2.2 Fluid Properties . . . . . . . . . . . . . 2.3 Control Volume . . . . . . . . . . . . . 2.4 Reynold’s Transport Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xlvi xlvi xlvi xlvi xlvii xlvii xlvii xlvii xlvii xlvii xlvii 1 1 2 2 4 5 9 13 15 15 25 25 25 25 25 27 27 27 29 31 35 36 37 41 41 41 44 45 46 50

3 Speed of Sound 3.1 Motivation . . . . . . . . . . . . . . . . . . . . . . 3.2 Introduction . . . . . . . . . . . . . . . . . . . . . 3.3 Speed of sound in ideal and perfect gases . . . . 3.4 Speed of Sound in Real Gas . . . . . . . . . . . 3.5 Speed of Sound in Almost Incompressible Liquid 3.6 Speed of Sound in Solids . . . . . . . . . . . . . 3.7 Sound Speed in Two Phase Medium . . . . . . .

4 Isentropic Flow 4.1 Stagnation State for Ideal Gas Model . . . . . . . . . . 4.1.1 General Relationship . . . . . . . . . . . . . . . 4.1.2 Relationships for Small Mach Number . . . . . 4.2 Isentropic Converging-Diverging Flow in Cross Section 4.2.1 The Properties in the Adiabatic Nozzle . . . . . 4.2.2 Isentropic Flow Examples . . . . . . . . . . . .

CONTENTS 4.2.3 Mass Flow Rate (Number) . . . . . . . . . 4.3 Isentropic Tables . . . . . . . . . . . . . . . . . . . 4.3.1 Isentropic Isothermal Flow Nozzle . . . . . 4.3.2 General Relationship . . . . . . . . . . . . . 4.4 The Impulse Function . . . . . . . . . . . . . . . . 4.4.1 Impulse in Isentropic Adiabatic Nozzle . . 4.4.2 The Impulse Function in Isothermal Nozzle 4.5 Isothermal Table . . . . . . . . . . . . . . . . . . . 4.6 The effects of Real Gases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v 53 62 63 63 70 70 73 73 74

5 Normal Shock 81 5.1 Solution of the Governing Equations . . . . . . . . . . . . . . . . . . 84 5.1.1 Informal Model . . . . . . . . . . . . . . . . . . . . . . . . . . 84 5.1.2 Formal Model . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 5.1.3 Prandtl’s Condition . . . . . . . . . . . . . . . . . . . . . . . . 88 5.2 Operating Equations and Analysis . . . . . . . . . . . . . . . . . . . 89 5.2.1 The Limitations of the Shock Wave . . . . . . . . . . . . . . . 90 5.2.2 Small Perturbation Solution . . . . . . . . . . . . . . . . . . . 90 5.2.3 Shock Thickness . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.3 The Moving Shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 5.3.1 Shock Result from a Sudden and Complete Stop . . . . . . . 94 5.3.2 Moving Shock into Stationary Medium (Suddenly Open Valve) 96 5.3.3 Partially Open Valve . . . . . . . . . . . . . . . . . . . . . . . 101 5.3.4 Partially Closed Valve . . . . . . . . . . . . . . . . . . . . . . 103 5.3.5 Worked–out Examples for Shock Dynamics . . . . . . . . . . 104 5.4 Shock Tube . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 5.5 Shock with Real Gases . . . . . . . . . . . . . . . . . . . . . . . . . 113 5.6 Shock in Wet Steam . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 5.7 Normal Shock in Ducts . . . . . . . . . . . . . . . . . . . . . . . . . . 113 5.8 More Examples for Moving Shocks . . . . . . . . . . . . . . . . . . . 114 5.9 Tables of Normal Shocks, k = 1.4 Ideal Gas . . . . . . . . . . . . . . 115 6 Normal Shock in Variable Duct Areas 123 6.1 Nozzle efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 6.2 Diffuser Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 7 Nozzle Flow With External Forces 135 7.1 Isentropic Nozzle (Q = 0) . . . . . . . . . . . . . . . . . . . . . . . . 136 7.2 Isothermal Nozzle (T = constant) . . . . . . . . . . . . . . . . . . . 136 8 Isothermal Flow 8.1 The Control Volume Analysis/Governing equations 8.2 Dimensionless Representation . . . . . . . . . . . 8.3 The Entrance Limitation of Supersonic Branch . . 8.4 Comparison with Incompressible Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 138 138 142 143

vi 8.5 8.6 8.7 8.8 9 Supersonic Branch . . . . . Figures and Tables . . . . . Isothermal Flow Examples . Unchoked situation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CONTENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 146 147 152 155 155 156 157 160 161 164 169 169 170 171 176 178 185 186 187 189 189 190 193 196 201 202 204 205 206 206 207 207 209 209 209 211 212 213 213 215

Fanno Flow 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 Non–dimensionalization of the equations . . . . . . . . 9.4 The Mechanics and Why the Flow is Choked? . . . . . . 9.5 The working equations . . . . . . . . . . . . . . . . . . . 9.6 Examples of Fanno Flow . . . . . . . . . . . . . . . . . . 9.7 Supersonic Branch . . . . . . . . . . . . . . . . . . . . . 9.8 Maximum length for the supersonic flow . . . . . . . . . 9.9 Working Conditions . . . . . . . . . . . . . . . . . . . . fL 9.9.1 Variations of The Tube Length ( 4D ) Effects . . . P2 9.9.2 The Pressure Ratio, P1 , effects . . . . . . . . . . 9.9.3 Entrance Mach number, M1 , effects . . . . . . . 9.10 The Approximation of the Fanno flow by Isothermal Flow 9.11 More Examples of Fanno Flow . . . . . . . . . . . . . . 9.12 The Table for Fanno Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10 RAYLEIGH FLOW 10.1 Introduction . . . . . . . . . . 10.2 Governing Equation . . . . . 10.3 Rayleigh Flow Tables . . . . . 10.4 Examples For Rayleigh Flow

11 Evacuating and Filling a Semi Rigid Chambers 11.1 Governing Equations and Assumptions . . . 11.2 General Model and Non-dimensioned . . . . 11.2.1 Isentropic Process . . . . . . . . . . . 11.2.2 Isothermal Process in The Chamber . 11.2.3 A Note on the Entrance Mach number 11.3 Rigid Tank with Nozzle . . . . . . . . . . . . . 11.3.1 Adiabatic Isentropic Nozzle Attached . 11.3.2 Isothermal Nozzle Attached . . . . . . 11.4 Rapid evacuating of a rigid tank . . . . . . . 11.4.1 With Fanno Flow . . . . . . . . . . . . 11.4.2 Filling Process . . . . . . . . . . . . . 11.4.3 The Isothermal Process . . . . . . . . 11.4.4 Simple Semi Rigid Chamber . . . . . 11.4.5 The “Simple” General Case . . . . . . 11.5 Advance Topics . . . . . . . . . . . . . . . . .

.5 Flow in a Semi–2D Shape .4 The Working Equations for the Prandtl-Meyer Function . . . . . . . 14. . .1. . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14. . . . . . . . . . . . δ and θ . . . .1 Introduction . . . . . . . 15. . . .4. . .4. . . . . . . vii 217 217 218 221 221 222 225 227 227 228 228 228 229 229 232 232 235 241 242 244 244 245 245 246 247 249 251 262 262 263 265 265 266 267 270 271 271 272 273 274 276 15 Prandtl-Meyer Function 15. . . . . . . 14. . .2 When No Oblique Shock Exist or When D > 0 . . . . . . . . . .4. . . . . . . . . . . . . . . . . . . . . 12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . and Shock Angle. . .2 Summary . . θ . . . . . . . . .2 Introduction . . . . . . . .1 Upstream Mach Number.2. . . . . . . . . .5 Summary . . . . . . . . . . . . . . . . . . . .4. . . . . . . . 14. . . 14. . . . 15. .10Issues Related to the Maximum Deflection Angle . .1 Introduction to Oblique Shock . . . . .8 Combination of the Oblique Shock and Isentropic Expansion . . . . . . 14. . . . . . . . . . . . . .2. . . . .2 Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.2. . . . . . . . . .3 Oblique Shock . . . . . . . . . . . . . . . . . .1 Alternative Approach to Governing Equations . . . δ 14. . .5 d’Alembert’s Paradox . . . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . . . 14. . . . . . .4. . . .3 Upstream Mach Number. . . . . . . . . . . . . . . . . . . . . . . . . . . 14. . . . . . . . . . . . . . . . . . 14. .CONTENTS 12 Evacuating/Filing Chambers under External Volume Control 12. . .4. . . . . . . . .4 Given Two Angles. .7 Examples For Prandtl–Meyer Function . . . . . . .1.4. .2 Comparison And Limitations between the Two Approaches 15. . . . 15. .2. 14. 14. . . 14. . . . .8 Maximum Value of Oblique shock . . . .3 The Maximum Turning Angle . .2. . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . . . 15. . . . . . . . . 14. . . .2 Geometrical Explanation . . . . . . . . . . . . . . . . . 14. . . . . . . . . . . . . 15. . . . . . . . . . .1 General Model . . .6 Appendix: Oblique Shock Stability Analysis . . . . . . . . . . . . 15.1 Rapid Process . . . . .9 Detached shock . . .1. . . . . . . . . . 14. .2 Introduction to Prandtl–Meyer Function . .4. . . . . . . . . . . 13 Topics in Unsteady one Dimensional gas dynamics 14 Oblique-Shock 14. . . . . . . . . . . . . .4. . . . . . . . . . . . . . . . . . M1 . .1 Preface to Oblique Shock . . .4. . . . . . . .7 Close and Far Views of the Oblique Shock . . . .11Oblique Shock Examples .4. . . . . . . . .13Optimization of Suction Section Design . . . . . . 14. . . . . . . . . . . . . . .6 Flat Body with an Angle of Attack . . . . . . .4. 12. . .6 Small δ “Weak Oblique shock” . . . . . . . . . . . . .3 Direct Connection . . . . .3 Introduction to Zero Inclination . .4. . . . . . and Deflection Angle. . . 15. . . 14. . . . . . 12. . . 15.4 Solution of Mach Angle . . . . . M1 . . . . . . . . .12Application of Oblique Shock . . . . . . . . . . . . . . . .

. .viii 16 Topics in Steady state Two Dimensional flow CONTENTS 279 A Computer Program 281 A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 Program listings . . . . . . . 287 . . . . . . .2 Usage . . . . . . 285 Authors index . . . .1 About the Program . . . . . . . . . . . 284 Index 285 Subjects index . . . . 281 A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . 1. . . . . . . . . . . . . . . . . . . . . . . . . 1. . 1. 43 4. . . . 3. . 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . .LIST OF FIGURES 1. . 46 ix .3 The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines .2 Perfect gas flows through a tube . . . . . 1. 1. . . . . . . k = 1. . . . .6 Photo of Ernest Mach . . . . . . . .1 The shock as connection of Fanno and Rayleigh lines after Stodola. . . . . . . . . . . . .9 Portrait of Rankine . . . . . . . . .12 The photo of Ernst Rudolf George Eckert with the author’s family .1 A very slow moving piston in a still gas . . . . . . . . . . 1. . . 3.1 Flow of a compressible substance (gas) through a converging–diverging nozzle. . . . . . . . . . . . . . . . . 41 4. . . . . . . . . . . . . . . . . . .5 Portrait of Galileo Galilei . . . . 7 9 11 12 16 17 17 18 19 20 21 22 28 28 32 4. .3 The stagnation properties as a function of the Mach number. . . . . . . . 1. . . . . . . . . . . . . . . . . .2 Stationary sound wave and gas moves relative to the pulse . . .4 Control volume inside a converging-diverging nozzle. . . . . . . . . . . . . . . . . . . . Steam and Gas Turbine . . . Steam and Gas Turbine . . . . . . . . . .8 Photo of Lord Rayleigh . . . . . .7 The photo of thebullet in a supersonic flow not taken in a wind tunnel 1. . . . . . . . 1. . . . .2 The schematic of deLavel’s turbine after Stodola. . .10 The photo of Gino Fanno approximately in 1950 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 Photo of Prandtl . . . 3. . . . . .3 The Compressibility Chart . . . . . . . . . . . . . . . . .4 Flow rate as a function of the back pressure taken from Stodola 1927 Steam and Gas Turbines . . . . . . . . . . . . . . . . . . . . . . 1. . . . . . . .4 44 4. . . . . . . . .

. . . . . . . . 5. . . . . .9 The number of iterations to achieve convergence. . 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 A shock moves into a moving medium as a result of a sudden and complete open valve.5 The relationship between the cross section and the Mach number on the subsonic branch . . . . . .1 6. . . . . . . .2 6. . . . . . . . .” . . 5. . . . . . . . . .15 Figure for Example (5. . . . . . . . . . . .4 The ratios of the static properties of the two sides of the shock. . . . . . . 5. .3 The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number.18 The results for Example (5. . . . . . . . . . . . . . . . . 4.8 Comparison of the pressure and temperature drop as a function of the normalized length (two scales) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3) . . . 4. 6. . .10) . 5. . . . . .1 Control volume for isothermal flow . . . . . . . . . . . . . . . 5. . . . . . . . . . A nozzle with normal shock . .7 The moving shock Mach numbers as a result of a sudden and complete stop. . . . . but it can also be viewed as a one– dimensional shock wave. . . . . . . . . . . . . . . . .1 A shock wave inside a tube. . . . . .16 The shock tube schematic with a pressure ”diagram. . . .3 6. . 5. . . .13 A shock as a result of a sudden and partially a valve closing or a narrowing the passage to the flow . . . . . . . . . . . . . . .9 Schematic to explain the significances of the Impulse function . . . . . . . . . . . . 5. . . . . . . . . . . .17 Figure for Example (5. . .4 The flow in the nozzle with different back pressures . . . . . . . . . . . . 5. 4. . . .8) . . . . . . . .7) . . . . . . . 5. . . . . 5. . 5. . . . . . . . . . . . . . .7 The comparison of nozzle flow .8 A shock moves into a still medium as a result of a sudden and complete opening of a valve . . . . . . . 5. 5. . . Schematic of a supersonic tunnel in a continuous region (and also for example (6. . . .6 Various ratios as a function of Mach number for isothermal Nozzle . . . . . . 4. . . . . . . .10) . . . . . . . . . . . . . . . . 5. . . . . . .6 Comparison between a stationary shock and a moving shock in a stationary medium in ducts. . . . . . . . . . . . . . . . . . . . . . .x LIST OF FIGURES 4. . . . . . . . . . . . . . . . . . . . . . . .10 Schematic of a flow of a compressible substance (gas) thorough a converging nozzle for example (4.14 Schematic of a piston pushing air in a tube. 5. . . 4. . .12 The results of the partial opening of the valve. . . . . Description to clarify the definition of diffuser efficiency . . . . . . . . . . . . . . 137 . . . . . . .5 Comparison between stationary shock and moving shock in ducts . . . . . . . . . . . . . . . . . . . 5. . . . . . . . . . . . . . . . .2 The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. . . . . .10 The maximum of “downstream” Mach number as a function of the specific heat. . . . . . . . . . 50 66 67 68 71 72 81 83 87 89 91 94 95 96 97 99 102 103 103 107 109 110 114 115 123 124 130 130 8. . 5. . . . . . . . . . . . . k . . . . . . . . . . . . . . . . . . .

. . . . 166 The maximum length as a function of specific heat. . . . . . . .3 A schematic of the control volumes used in this model . .175 fL fL The pressure distribution as a function of 4D for a short 4D . . . . . . . . . . . .1) .6 9. . 185 10. . . . . . . . . . . . . . 11. . . . . . . . . .4 9. . . . . . . . . . k . . . . . . . . 4D supersonic branch . . . . .2 9. . . . . 172 The Mach numbers at entrance and exit of tube and mass flow rate fL for Fanno Flow as a function of the 4D . . . . . . . . . .1 9. . .3 The Mach number at the entrance to a tube under isothermal flow fL . . . . . .17 9. . . . . . 11. . . . .8 9. . . 171 The development properties in of converging nozzle . . .7 9. . . . . . . . . . . . . . . . . .2 A schematic of two possible connections of the tube to a single chamber . . . . . . . . . . . . . . . 11. . . . 218 . . . . . .2) . . . . . .2 The Temperature Entropy Diagram For Rayleigh Line . . . . . .1 The control volume of Rayleigh Flow . . . . . . . . . 180 Schematic of a “long” tube in supersonic branch . . . . . . . . .6 The reduced time as a function of the modified reduced pressure . . . . . . . . . . . . . . . .3 for Fanno Flow . . . . . . .5 The reduced time as a function of the modified reduced pressure . . . . . 11. . . .9 9. . . . . .4 The pressure assumptions in the chamber and tube entrance . . .10 9.3 9. . . . . . . . . . . 191 10. . . 163 Schematic of Example (9. . . 177 4f L 4f L . . . . . . . .12 9. . . . . 155 Various parameters in Fanno flow as a function of Mach number . . . . . .1 The two different classifications of models that explain the filling or evacuating of a single chamber . . . . 181 fL The extra tube length as a function of the shock location. . . . . . .3 . . temperature relationships as a function of the Mach number for isothermal flow . . . . . . . . . . .5 9. . . . 11. . . . . . . . . . . . . . . . . . . 179 fL fL Fanno Flow Mach number as a function of 4D when the total 4D = 0. .14 9. . . . . . . . . . . . . . .13 9. . . 174 fL M1 as a function M2 for different 4D for supersonic entrance velocity. . .3 The basic functions of Rayleigh Flow (k=1. . . . . . . 201 202 202 203 210 212 12. . . 164 The schematic of Example (9. . . . . . M1 to the tube as a function fL of 4D supersonic branch . . . . . . . . . . . . . . . . . .11 9. . . . . . . . . . . . . 178 The pressure distribution as a function of D for a long D The effects of pressure variations on Mach number profile as a funcfL fL tion of 4D when the total resistance 4D = 0. . . .LIST OF FIGURES xi 8. .2 Description of the pressure. . .4) . . 195 11. . . 189 10. . . . . . . . . . .15 9. . . . . . . . . . . .18 Control volume of the gas flow in a constant cross section . . . . . 143 8. . . 170 fL The effects of increase of 4D on the Fanno line . .16 9. . 181 The maximum entrance Mach number. 182 The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow . . . . . 153 model as a function 4D 9.1 The control volume of the “Cylinder” . . . . 173 4f L M1 as a function M2 for various D . . . . . . . .

. . . .4 Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. . . . . However. . . . . 248 14. . . . . . . . . . . .10Schematic of finite wedge with zero angle of attack. . .12The schematic for a round–tip bullet in a supersonic flow . . . . . . . . .15Oblique shock occurs around a cone. . . . . . . . . . . . . . . . 263 14. . . . 229 14. .5 The view of a large inclination angle from different points in the fluid field. . . 251 14. 224 12. Notice that both have a maximum point and a “no solution” zone. . . . . . . . . . shock angle. . . . . . . . . 250 14. . . .1 The definition of the angle for the Prandtl–Meyer function.9 The possible range of solutions for different parameters for given upstream Mach numbers . . . . .16Maximum values of the properties in an oblique shock . . . 236 14. . . .0. . . . . . .2 The angles of the Mach line triangle . . . . . . .7 The Mach waves that are supposed to be generated at zero inclination. . . . . .3 The pressure ratio as a function of the dimensionless time for choked condition . . .240 14. . . . . . . and exit Mach number for M1 = 3 . . . . 254 14. and the weak shock coexist. .19Schematic for Example (14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5) . . . . . . . . . .2 The pressure ratio as a function of the dimensionless time for chokeless condition . . . . . . . . . . . . . . .21Typical examples of unstable and stable situations . . . . . . Prandtl-Meyer function approaches closer to a zero deflection angle. . . . . . . . a Research Professor at Cuernavaco University of Mexico. . . . . . . . . 244 14. . . . . the strong shock. . . .11A local and a far view of the oblique shock. . . . . . . . . . 245 14. . . . . . . . . 228 14. . . .6 The various coefficients of three different Mach numbers to demonstrate that D is zero . . . . 251 14. . . .17Two variations of inlet suction for supersonic flow . . According to his measurement. It can be noticed that the normal shock. . 243 14. . . . . . . . 241 14. . . . . . 239 14. . .3 A typical oblique shock schematic . . . . 265 15. . . . . . . . . . . . . 247 14. . . . . . . . . . .4) . . . . . 264 15. the cone half angle is 15◦ and the Mach number is 2. . . . . . . . . . . . . 266 . . . . . 235 14. . . 252 14. .22The schematic of stability analysis for oblique shock. . . . . . . . . . . . . . . . . . . . . Grigory Toker. . . . . . . . . . . . . 223 12. . 253 14. .2 The regions where oblique shock or Prandtl–Meyer function exist. 224 14. . . . . . . . . . .4 The pressure ratio as a function of the dimensionless time . . . which is around zero. . . . . . . . . . . . . . . . . . . .xii LIST OF FIGURES 12. . . . . . . . . . . . . . . . . . . . . . . This photo is courtesy of Dr. . .2.14The “detached” shock in a complicated configuration sometimes referred to as Mach reflection . . . . 265 15. . . . . . . . . . . . . . . . . . . .1 A view of a straight normal shock as a limited case for oblique shock 227 14. . .8 The calculation of D (possible error). . . . . . . . . .18Schematic for Example (14. . . . . . .3 The schematic of the turning flow . . . . . . . . . . . . . .13The schematic for a symmetrical suction section with Mach reflection 248 14. . . . . . . . . .20Schematic of two angles turn with two weak shocks .

. . . . . . .LIST OF FIGURES 15.1 Schematic diagram that explains the structure of the program . . . . . . . . . .2) . xiii 267 272 272 273 273 274 275 A. .1 . . . . . . . . . . .8 The definition of the angle for the Prandtl–Meyer function. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15. 15. . . . . . 15. . . . .5 Expansion of Prandtl-Meyer function when it exceeds the maximum angle . . . . . . . . . . . . 282 . . . . . . . . . .9 The schematic of Example 15. .4 The schematic of the coordinate based on the mathematical description . .6 The angle as a function of the Mach number . .7 A simplified diamond shape to illustrate the supersonic d’Alembert’s Paradox . . . . . . . . . . . . . . . . . . . . . . . . . . . 15. . . . . . . . . . . . . 15. . . . . 15. . . . . .10The schematic for the reversed question of example (15. . . . .

xiv LIST OF FIGURES .

2 Table for a Reflective Shock from a suddenly closed end (k=1. .3) 119 5. . . . .1 The shock wave table for k = 1. . . . . .4 (continue) . Manual of Sound Recording. . 119 5. . . . . . . 4. . . . 3. . . . . . . . . . . 115 5.4) . . . . .2 Isentropic Table k=1. . . 35 36 37 58 59 60 62 63 73 74 5. . 117 5. . 116 5. . after Aldred. .4)(continue)118 5. London: Fountain Press. . . . . . . .4 Table for shock propagating from a suddenly opened valve (k=1. . . . . .4 Table for shock propagating from a suddenly opened valve (k=1.4 (continue) . .1 The shock wave table for k = 1. . . . . . . . . . . . . . . .3) 120 5. . .3 Solids speed of sound. .4 (continue) . . .1 Fliegner’s number and other paramters as function of Mach number (continue) . . .1 The shock wave table for k = 1. . . . . . .LIST OF TABLES 3. . . . John. . . 117 5. . 118 5. .2 Liquids speed of sound. London:Fountain Press. . .3 Table for shock propagating from suddenly opened valve (k=1. . . . . . .3) 121 xv . . . 4. after Aldred. . 1972 . . . .3 Table for shock propagating from suddenly opened valve (k=1. . . . . . . . . . . . . . . 1972 . . . . . . John. . . . .4) . 4. . . . . . . . . . . . . . . . .1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . 4. . . .3 Isothermal Table . . .1 Fliegner’s number and other paramters as function of Mach number 4. . . . . . . . Manual of Sound Recording. . . . . . . . . . . . . . . . . . . . . . . .4) . . . . .3 Isothermal Table (continue) . . . . . . 3. . . . . . . . . . . . . .2 Isentropic Table k = 1. . . .2 Table for Reflective Shock from suddenly closed valve (end) (k=1. .4 . . . . . . . . . . .4 Table for shock propagating from a suddenly opened valve (k=1. . 4. . . . . . . . . . . . . . .1 Water speed of sound from different sources . . . . . . . .4 . . . . . . . . . . . . . 4. . . .

. . . . 246 . . . .4 (continue) . . . . . . .4 . . . . . . . . . 152 Fanno Flow Standard basic Table . . . .1 LIST OF TABLES The Isothermal Flow basic parameters . . 245 14. . . . . . . . . 195 14. . . .1 Maximum values of oblique shock (continue) k=1. . .1 Rayleigh Flow k=1. 188 10. . . .4 . . . . . . . . . . . . . . . . . . . 194 10. . . . . .4 (continue) . . 193 10. . . . . . .4 9. . . . . . . . . .1 Rayleigh Flow k=1. . . . . . . .1 8. .4 . . . . . . . . . . . . 147 The flow parameters for unchoked flow . . . . . . . . . . . . . . . . . . . . . . . . . . . 187 Fanno Flow Standard basic Table (continue) . . .1 9. . . . .1 Table of maximum values of the oblique Shock k=1. . . . .xvi 8. . . . . .1 Rayleigh Flow k=1.

2001. because free software needs free documentation: a free program should xvii . textbook. Inc.Notice of Copyright For This Document: This document is published under dual licenses: You can choose the license under which you use the document and associate files and software. November 2002 Copyright ©2000. Preamble The purpose of this License is to make a manual. which means that derivative works of the document must themselves be free in the same sense. which is a copyleft license designed for free software. or other functional and useful document ”free” in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it. 51 Franklin St. Boston. this License preserves for the author and publisher a way to get credit for their work. It complements the GNU General Public License. We have designed this License in order to use it for manuals for free software. while not being considered responsible for modifications made by others. either commercially or noncommercially. but changing it is not allowed.2. Secondarily. MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document. with or without modifying it. This License is a kind of ”copyleft”. Fifth Floor.2002 Free Software Foundation. GNU Free Documentation License Version 1.

refers to any such manual or work.xviii LIST OF TABLES come with manuals providing the same freedoms that the software does. philosophical. A Front-Cover Text may be at most 5 words. unlimited in duration. as being those of Invariant Sections. to use that work under the conditions stated herein. A ”Modified Version” of the Document means any work containing the Document or a portion of it. that is suitable for revising the document straightforwardly with generic text editors or (for images composed of pixels) generic paint programs or (for drawings) some widely available drawing editor. either copied verbatim. You accept the license if you copy. Such a notice grants a world-wide. or with modifications and/or translated into another language. or of legal. But this License is not limited to software manuals. The ”Document”. (Thus. If a section does not fit the above definition of Secondary then it is not allowed to be designated as Invariant.) The relationship could be a matter of historical connection with the subject or with related matters. as Front-Cover Texts or Back-Cover Texts. APPLICABILITY AND DEFINITIONS This License applies to any manual or other work. and a Back-Cover Text may be at most 25 words. A ”Transparent” copy of the Document means a machine-readable copy. If the Document does not identify any Invariant Sections then there are none. A ”Secondary Section” is a named appendix or a front-matter section of the Document that deals exclusively with the relationship of the publishers or authors of the Document to the Document’s overall subject (or to related matters) and contains nothing that could fall directly within that overall subject. and is addressed as ”you”. in any medium. that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. royalty-free license. regardless of subject matter or whether it is published as a printed book. The Document may contain zero Invariant Sections. 1. if the Document is in part a textbook of mathematics. in the notice that says that the Document is released under this License. commercial. represented in a format whose specification is available to the general public. modify or distribute the work in a way requiring permission under copyright law. ethical or political position regarding them. or absence of markup. a Secondary Section may not explain any mathematics. in the notice that says that the Document is released under this License. A copy made in an otherwise Transparent file format whose markup. The ”Cover Texts” are certain short passages of text that are listed. below. The ”Invariant Sections” are certain Secondary Sections whose titles are designated. it can be used for any textual work. Any member of the public is a licensee. and that is suitable for input to text formatters or for automatic translation to a variety of formats suitable for input to text formatters. We recommend this License principally for works whose purpose is instruction or reference. has been arranged to thwart or discourage subsequent modification .

The Document may include Warranty Disclaimers next to the notice which states that this License applies to the Document. 2. The ”Title Page” means. and the license notice saying this License applies to the Document are reproduced in all copies. You may also lend copies. under the same conditions stated above. A section ”Entitled XYZ” means a named subunit of the Document whose title either is precisely XYZ or contains XYZ in parentheses following text that translates XYZ in another language. legibly. the title page itself.GNU FREE DOCUMENTATION LICENSE xix by readers is not Transparent.) To ”Preserve the Title” of such a section when you modify the Document means that it remains a section ”Entitled XYZ” according to this definition. for a printed book. Examples of transparent image formats include PNG. you may accept compensation in exchange for copies. ”Dedications”. For works in formats which do not have any title page as such. ”Endorsements”. plus such following pages as are needed to hold. PostScript or PDF produced by some word processors for output purposes only. These Warranty Disclaimers are considered to be included by reference in this License. the copyright notices. XCF and JPG. SGML or XML using a publicly available DTD. However. You may not use technical measures to obstruct or control the reading or further copying of the copies you make or distribute. VERBATIM COPYING You may copy and distribute the Document in any medium. preceding the beginning of the body of the text. ”Title Page” means the text near the most prominent appearance of the work’s title. SGML or XML for which the DTD and/or processing tools are not generally available. and standard-conforming simple HTML. but only as regards disclaiming warranties: any other implication that these Warranty Disclaimers may have is void and has no effect on the meaning of this License. and the machine-generated HTML. either commercially or noncommercially. and the Document’s . the material this License requires to appear in the title page. and you may publicly display copies. LaTeX input format. and that you add no other conditions whatsoever to those of this License. numbering more than 100. Opaque formats include proprietary formats that can be read and edited only by proprietary word processors. An image format is not Transparent if used for any substantial amount of text. provided that this License. Texinfo input format. such as ”Acknowledgements”. or ”History”. COPYING IN QUANTITY If you publish printed copies (or copies in media that commonly have printed covers) of the Document. A copy that is not ”Transparent” is called ”Opaque”. If you distribute a large enough number of copies you must also follow the conditions in section 3. Examples of suitable formats for Transparent copies include plain ASCII without markup. (Here XYZ stands for a specific section name mentioned below. 3. PostScript or PDF designed for human modification.

. one or more persons or entities responsible for authorship of the modifications in the Modified Version. and from those of previous versions (which should. thus licensing distribution and modification of the Modified Version to whoever possesses a copy of it. provided that you release the Modified Version under precisely this License. 4. Use in the Title Page (and on the covers. You may add other material on the covers in addition. as the publisher. but not required. you should put the first ones listed (as many as fit reasonably) on the actual cover. B. be listed in the History section of the Document). You may use the same title as a previous version if the original publisher of that version gives permission. you must take reasonably prudent steps. If the required texts for either cover are too voluminous to fit legibly. Both covers must also clearly and legibly identify you as the publisher of these copies. as long as they preserve the title of the Document and satisfy these conditions. can be treated as verbatim copying in other respects. if any) a title distinct from that of the Document. if it has fewer than five). or state in or with each Opaque copy a computernetwork location from which the general network-using public has access to download using public-standard network protocols a complete Transparent copy of the Document. The front cover must present the full title with all words of the title equally prominent and visible. If you use the latter option. together with at least five of the principal authors of the Document (all of its principal authors. It is requested. with the Modified Version filling the role of the Document. you must enclose the copies in covers that carry. State on the Title page the name of the publisher of the Modified Version. all these Cover Texts: Front-Cover Texts on the front cover. MODIFICATIONS You may copy and distribute a Modified Version of the Document under the conditions of sections 2 and 3 above. and continue the rest onto adjacent pages. to ensure that this Transparent copy will remain thus accessible at the stated location until at least one year after the last time you distribute an Opaque copy (directly or through your agents or retailers) of that edition to the public. unless they release you from this requirement. you must do these things in the Modified Version: A. to give them a chance to provide you with an updated version of the Document. you must either include a machine-readable Transparent copy along with each Opaque copy. free of added material. if there were any. and Back-Cover Texts on the back cover. In addition. C.xx LIST OF TABLES license notice requires Cover Texts. If you publish or distribute Opaque copies of the Document numbering more than 100. when you begin distribution of Opaque copies in quantity. Copying with changes limited to the covers. clearly and legibly. List on the Title Page. as authors. that you contact the authors of the Document well before redistributing any large number of copies.

. authors. and publisher of the Document as given on its Title Page. F. Delete any section Entitled ”Endorsements”. H. L. year. Add an appropriate copyright notice for your modifications adjacent to the other copyright notices. These titles must be distinct from any other section titles. Preserve all the Invariant Sections of the Document. For any section Entitled ”Acknowledgements” or ”Dedications”. a license notice giving the public permission to use the Modified Version under the terms of this License. These may be placed in the ”History” section. N. Preserve the section Entitled ”History”. unaltered in their text and in their titles. and publisher of the Modified Version as given on the Title Page. xxi E. M. if any. Preserve all the copyright notices of the Document. given in the Document for public access to a Transparent copy of the Document. and preserve in the section all the substance and tone of each of the contributor acknowledgements and/or dedications given therein. If the Modified Version includes new front-matter sections or appendices that qualify as Secondary Sections and contain no material copied from the Document. Include. then add an item describing the Modified Version as stated in the previous sentence. Preserve any Warranty Disclaimers. Preserve its Title. Such a section may not be included in the Modified Version. J. Section numbers or the equivalent are not considered part of the section titles. I. If there is no section Entitled ”History” in the Document. year. Preserve the Title of the section. new authors.GNU FREE DOCUMENTATION LICENSE D. Preserve in that license notice the full lists of Invariant Sections and required Cover Texts given in the Document’s license notice. add their titles to the list of Invariant Sections in the Modified Version’s license notice. Do not retitle any existing section to be Entitled ”Endorsements” or to conflict in title with any Invariant Section. and add to it an item stating at least the title. Preserve the network location. K. Include an unaltered copy of this License. and likewise the network locations given in the Document for previous versions it was based on. in the form shown in the Addendum below. To do this. You may omit a network location for a work that was published at least four years before the Document itself. or if the original publisher of the version it refers to gives permission. you may at your option designate some or all of these sections as invariant. G. create one stating the title. immediately after the copyright notices. O.

you must combine any sections Entitled ”History” in the various original documents. under the terms defined in section 4 above for modified versions. unmodified. on explicit permission from the previous publisher that added the old one. and any sections Entitled ”Dedications”. to the end of the list of Cover Texts in the Modified Version. previously added by you or by arrangement made by the same entity you are acting on behalf of. statements of peer review or that the text has been approved by an organization as the authoritative definition of a standard. provided you insert a copy of this License into . and a passage of up to 25 words as a Back-Cover Text. COLLECTIONS OF DOCUMENTS You may make a collection consisting of the Document and other documents released under this License. and distribute it individually under this License. and multiple identical Invariant Sections may be replaced with a single copy. and that you preserve all their Warranty Disclaimers. 5. 6. Make the same adjustment to the section titles in the list of Invariant Sections in the license notice of the combined work. and list them all as Invariant Sections of your combined work in its license notice. You must delete all sections Entitled ”Endorsements”. provided it contains nothing but endorsements of your Modified Version by various parties–for example. but you may replace the old one. in parentheses. and replace the individual copies of this License in the various documents with a single copy that is included in the collection. likewise combine any sections Entitled ”Acknowledgements”. the name of the original author or publisher of that section if known. Only one passage of Front-Cover Text and one of BackCover Text may be added by (or through arrangements made by) any one entity. The author(s) and publisher(s) of the Document do not by this License give permission to use their names for publicity for or to assert or imply endorsement of any Modified Version. forming one section Entitled ”History”. provided that you follow the rules of this License for verbatim copying of each of the documents in all other respects. COMBINING DOCUMENTS You may combine the Document with other documents released under this License. You may extract a single document from such a collection. you may not add another. provided that you include in the combination all of the Invariant Sections of all of the original documents. If there are multiple Invariant Sections with the same name but different contents. make the title of each such section unique by adding at the end of it. If the Document already includes a cover text for the same cover.xxii LIST OF TABLES You may add a section Entitled ”Endorsements”. The combined work need only contain one copy of this License. You may add a passage of up to five words as a Front-Cover Text. In the combination. or else a unique number.

TRANSLATION Translation is considered a kind of modification. and any Warranty Disclaimers. 8. You may include a translation of this License. However. and follow this License in all other respects regarding verbatim copying of that document. and will automatically terminate your rights under this License. ”Dedications”. Otherwise they must appear on printed covers that bracket the whole aggregate. provided that you also include the original English version of this License and the original versions of those notices and disclaimers. so you may distribute translations of the Document under the terms of section 4. the Document’s Cover Texts may be placed on covers that bracket the Document within the aggregate. If the Cover Text requirement of section 3 is applicable to these copies of the Document. or rights. FUTURE REVISIONS OF THIS LICENSE . sublicense. the requirement (section 4) to Preserve its Title (section 1) will typically require changing the actual title. sublicense or distribute the Document is void. AGGREGATION WITH INDEPENDENT WORKS A compilation of the Document or its derivatives with other separate and independent documents or works. modify. In case of a disagreement between the translation and the original version of this License or a notice or disclaimer. 10. from you under this License will not have their licenses terminated so long as such parties remain in full compliance. Replacing Invariant Sections with translations requires special permission from their copyright holders. then if the Document is less than one half of the entire aggregate. the original version will prevail. or the electronic equivalent of covers if the Document is in electronic form. in or on a volume of a storage or distribution medium. If a section in the Document is Entitled ”Acknowledgements”. TERMINATION You may not copy.GNU FREE DOCUMENTATION LICENSE xxiii the extracted document. parties who have received copies. and all the license notices in the Document. 7. this License does not apply to the other works in the aggregate which are not themselves derivative works of the Document. or ”History”. is called an ”aggregate” if the copyright resulting from the compilation is not used to limit the legal rights of the compilation’s users beyond what the individual works permit. but you may include translations of some or all Invariant Sections in addition to the original versions of these Invariant Sections. Any other attempt to copy. modify. When the Document is included in an aggregate. or distribute the Document except as expressly provided for under this License. 9.

See http://www. Front-Cover Texts and Back-Cover Texts. ADDENDUM: How to use this License for your documents To use this License in a document you have written. If your document contains nontrivial examples of program code.” line with this: with the Invariant Sections being LIST THEIR TITLES.Texts. Each version of the License is given a distinguishing version number.xxiv LIST OF TABLES The Free Software Foundation may publish new. If the Document specifies that a particular numbered version of this License ”or any later version” applies to it.. you may choose any version ever published (not as a draft) by the Free Software Foundation. include a copy of the License in the document and put the following copyright and license notices just after the title page: Copyright ©YEAR YOUR NAME. If you have Invariant Sections without Cover Texts.. such as the GNU General Public License. no FrontCover Texts. If the Document does not specify a version number of this License. A copy of the license is included in the section entitled ”GNU Free Documentation License”. with the FrontCover Texts being LIST.2 or any later version published by the Free Software Foundation. and no Back-Cover Texts. to permit their use in free software. Version 1. If you have Invariant Sections. and with the Back-Cover Texts being LIST. Such new versions will be similar in spirit to the present version.gnu. Permission is granted to copy. distribute and/or modify this document under the terms of the GNU Free Documentation License. we recommend releasing these examples in parallel under your choice of free software license. replace the ”with. . but may differ in detail to address new problems or concerns. merge those two alternatives to suit the situation.org/copyleft/. with no Invariant Sections. you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. or some other combination of the three. revised versions of the GNU Free Documentation License from time to time.

You may at your option offer warranty in exchange for a fee. you must document the nature of those modifications in the ”Credits” section along with your name. No Endorsement The names ”POTTO Project” and ”Fundamentals of Compressible Fluid Mechanics” or the author of this document must not be used to endorse or promote products derived from this text (book or software) without prior written permission. keep intact all the copyright notices that refer to this document. . all the material based on this original content or work. Modification and distribution of modified material You may modify your copy or copies of this document and the attached software or any portion of it. The original version of this document may be found at http://www.org/copyright.html This document is derived from open content license http://opencontent. and information concerning how you may be contacted. Derived Name(s) Products derived from this software may not be called “POTTO Project. in any medium. 2.POTTO PROJECT LICENSE xxv Potto Project License This document may be redistributed provided a pointer appears in a prominent place showing clearly where the original version was published and/or was obtained. Distributing. and Modifying 1. Disclaimer of warranty of the original author You may copy and distribute exact replicas of this document as you receive it. provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty of the original author. You may distribute such modifications. 4. potto.” or alleged association with this author nor may “POTTO” or “POTTO Project” appear in their name. You must have a reasonable way to contact you. org/opl. Your Name and Communication With You If you wish to modify this text or software in any way. under the terms of Section 1 above. without prior written permission of the Dr. You may offer instructional support for this document and software exchange for a fee. You may at your discretion charge a fee for the media and/or handling involved in creating a unique copy of this document. 3. 5. Genick BarMeir.shtml LICENSE Terms and Conditions for Copying.

the POTTO Project residence is the state of Minnesota. incidental or consequential damages arising out of the use or inability to use this manuscript. or by deriving works here from. Except when otherwise stated in writing the copyright holders and/or other parties provide these manuscripts “AS IS” without warranty of any kind. These actions are prohibited by law if you do not accept this License. No Liability In no event unless required by applicable law or agreed to in writing will any copyright holder. but not limited to. . including. since you have not signed it. Applicability of this license You are not required to accept this License. distributing or translating these materials. or otherwise unacceptable you assume the cost of all necessary repair or correction. to the extent permitted by applicable law. Should this manuscript prove faulty. even if such holder or other party has been advised of the possibility of such damages. Therefore. you indicate your acceptance of this License to do so. Currently. by distributing or translating these materials. there is no warranty for the manuscript. distribute or modify these materials. 8. or any other party who may mirror and/or redistribute these materials as permitted above. inaccurate. special. 7. nothing else grants you permission to copy. The various provisions of these terms and conditions are severable and if any provision is held to be invalid or unenforceable by any court of competent jurisdiction then such invalidity or unenforceability shall not affect the remaining provisions. you use the book and or the software must be terminated immediately. If these terms and conditions are not accepted in full. including any general. the implied warranties of merchantability and fitness for a particular purpose. The entire risk of use of this manuscript is with you. either expressed or implied.xxvi LIST OF TABLES 6. 9. However. Jurisdiction These terms and conditions are governed by and will be interpreted in accordance with the state of POTTO Project residence law and any disputes relating to these terms and conditions will be exclusively subject to the jurisdiction of the courts of POTTO Project residence. be liable to you for damages. No Warranty Because these materials are licensed free of charge. and all its terms and conditions for copying.

This is a copyrighted work: it is not in the public domain! If you wish to cite portions of this book in a work of your own. Minor contributions (typo corrections. I am indebted to all those who have given freely of their own knowledge. Major contributions are listed by individual name with some detail on the nature of the contribution(s).” it is in no way inferior to the effort or value of a ”major” contribution. date. just smaller in the sense of less text changed. Any and all contributions are gratefully accepted.) are listed by name only for reasons of brevity. you must follow the same guidelines as for any other GDL copyrighted work. etc.CONTRIBUTOR LIST How to contribute to this book As a copylefted work. The only ”catch” is that credit must be given where credit is due. time. Credits All entries arranged in alphabetical order of surname. • Contact at: barmeir@gmail. Please understand that when I classify a contribution as ”minor. and resources to make this a better book! • Date(s) of contribution(s): 2004 to present • Nature of contribution: Original author. this book is open to revision and expansion by any interested parties. contact info.com John Martones • Date(s) of contribution(s): June 2005 xxvii . etc.

. • Roy Tate November 2006. some error corrections. Ralph Menikoff • Date(s) of contribution(s): July 2005 • Nature of contribution: Some discussion about the solution to oblique shock and about the Maximum Deflection of the oblique shock. D. Your name here • Date(s) of contribution(s): Month and year of contribution • Nature of contribution: Insert text here. describing how you contributed to the book. Grigory Toker • Date(s) of contribution(s): August 2005 • Nature of contribution: Provided pictures of the oblique shock for oblique shcok chapter.net Typo corrections and other ”minor” contributions • H. Ph. • Contact at: my email@provider.xxviii LIST OF TABLES • Nature of contribution: HTML formatting. September 2005. Suggestions on improving english and gramer. . Gohrah. some LaTeX issues.

R. In his early part of his professional life. Much of his time has been spend doing research in the field of heat and mass transfer (this includes fluid mechanics) related to manufacturing processes and design. will promote new re1 Where the mathematicians were able only to prove that the solution exists.About This Author Genick Bar-Meir holds a Ph. the critical piston velocity in a partially filled chamber (related to hydraulic jump). Bar-Meir was the last student of the late Dr. Now. All the models have practical applicability. For example. Spain. supply and demand to rapid change power system and etc. Bar-Meir developed models that explained several manufacturing processes such the rapid evacuation of gas from containers. Moreover. Currently.E. in GM. He developed models for Mass Transfer in high concentration that became a building blocks for many other models. As the change in the view occurred. the author’s models were used in numerical works. These models are based on analytical solution to a family of equations1 .G.D. These models have been extended by several research groups (needless to say with large research grants). as in the past. the Spanish Comision Interministerial provides grants TAP97-0489 and PB98-0007. Bar-Meir was mainly interested in elegant models whether they have or not a practical applicability. books and softwares. and even Iran. The author enjoys to encourages his students to understand the material beyond the basic requirements of exams. in Mechanical Engineering from University of Minnesota and a Master in Fluid Mechanics from Tel Aviv University. xxix . he spends time writing books and software for the POTTO project (see Potto Prologue). this author’s views had changed and the virtue of the practical part of any model becomes the essential part of his ideas. The author believes that this book. British industry. Eckert. and the CICYT and the European Commission provides 1FD97-2333 grants for minor aspects of that models. Dr.

practically from scratch. The author spent years working on the sea (ships) as a engine sea officer but now the author prefers to remain on solid ground. A past project of his was building a four stories house. The author lives with his wife and three children. he often feels clueless about computers and programing. While he writes his programs and does other computer chores. this author believes that the book will blaze a trail of new understanding. More than that.xxx LIST OF TABLES search. While he known to look like he know about many things. the author just know to learn quickly. .

Ashcroff (see http://cyber. punishes the small “entrepreneur” who tries to defend his or her work. On one hand. The study of technical 2 After the last decision of the Supreme Court in the case of Eldred v.edu/openlaw/eldredvashcroft for more information) copyrights practically remain indefinitely with the holder (not the creator). the creation of the POTTO Project. It is unacceptable that the price of the college books will be over $150 per book (over 10 hours of work for an average student in The United States). like an open source. The second issue that prompted the writing of this book is the fact that we as the public have to deal with a corrupted judicial system. 3 In some sense one can view the encyclopedia Wikipedia as an open content project (see http: //en. The writing of a technical book is really a collection of information and practice. Writing a book in the technical field is not the same as writing a novel. when applied to “small” individuals who are not able to hire a large legal firm. law. Hence. The wikipedia is an excellent collection of articles which are written by various individuals. It has become very clear to the author and founder of the POTTO Project that this situation must be stopped. The first issue is the enormous price of college textbooks. one of this author’s sages. judges simply manufacture facts to make the little guy lose and pay for the defense of his work. the corrupted court system defends the “big” guys and on the other hand. There is always someone who can add to the book. said instead of whining about arrogance and incorrectness. xxxi . one should increase wisdom.Prologue For The POTTO Project This series of books was born out of frustrations in two respects.harvard. Kook.wikipedia. is a new idea3 . As individuals we have to obey the law. particularly the copyright law with the “infinite2 ” time with the copyright holders. It is apparent that writing textbooks especially for college students as a cooperation. The POTTO Project has far greater goals than simply correcting an abusive Judicial system or simply exposing abusive judges. As R. However. This project is to increase wisdom and humility.org/wiki/Main Page).

The author has not found any technique that is more useful for this purpose than practicing the solving of problems and exercises.” American Scientist.html)4. In these cases.xxxii LIST OF TABLES material isn’t only done by having to memorize the material.ac. In a way. Thus. they will become the most popular books and the most read books in their respected field.uk/ bugclub/raiders. but also by coming to understand and be able to solve related problems. 77:139. they now have an opportunity to put their notes to use for others. the popularity of the books should be one of the incentives for potential contributors. The answer varies from individual to individual.. Why would someone volunteer to be an author or organizer of such a book? This is the first question the undersigned was asked. For others. there are always others who can see new aspects of or add to the material. so can we in creating textbooks by this powerful tool. The collective material is much richer than any single person can create by himself. ”Army Ants: A Collective Intelligence. Whatever the reasons. One component is to come to know and socialize with many in the profession. While a single ant is not intelligent enough to attack and hunt large prey. To reach this possibility the collective book idea was created/adapted. For some authors. then that contributor will be the copyright holder of that specific section (even within question/answer sections). Experience has shown that in explaining to others any given subject.ex. For others the social part is as simple as a desire to reduce the price of college textbooks. The desire to be an author of a well-known book (at least in his/her profession) will convince some to put forth the effort. contributing to this book will help one to understand the material better. The secret of the ants’ power lies in their collective intelligence. The following example explains this point: The army ant is a kind of carnivorous ant that lives and hunts in the tropics. While one can be as creative as possible. the collective power of their networking creates an extremely powerful intelligence to carry out this attack ( see for information http://www. 1989 . the writing of or contributing to this kind of book will serve as a social function. The book’s contributor’s names could be written by their sections. It is not just for experts to contribute. hunting animals that are even up to a hundred kilograms in weight. One can be successful when one solves as many problems as possible. especially for family members or relatives and those students lacking funds. but also students who happened to be doing their homework. For some contributors/authors. the undersigned believes that personal intentions are appropriate and are the author’s/organizer’s private affair. Nigel R. It is hoped that because of the open nature of these books. So when an insect which is blind can be so powerful by networking. The student’s contributions can be done by 4 see also in Franks. The social function can have at least two components. If a contributor of a section in such a book can be easily identified. the reason is the pure fun of writing and organizing educational material. one also begins to better understand the material. in the course of their teaching they have found that the textbook they were using contains sections that can be improved or that are not as good as their own notes.

0 0.” In this process. .0. The organizer of the book in some instances will be the main author of the work.0.) can be redone by anyone who has the time to do it. Thus. these works will have a version number because they are alive and continuously evolving.0. graphs and etc. Someone has to write the skeleton and hopefully others will add “flesh and skin. this method is expected to accelerate the creation of these high quality books. Thus. better discussions or better explanations are all welcome to these books.0.0. It is hoped that the book will be error-free.4 0.0 0.0. Unlike a regular book. These books should be considered more as a project than to fit the traditional definition of “plain” books.0. It is also hoped that others will contribute to the question and answer sections in the book.0 0. These books are written in a similar manner to the open source software process. Additionally.3 0. contributions can be made from any part of the world by those who wish to translate the book. The undersigned of this document intends to be the organizer/author/coordinator of the projects in the following areas: project name Die Casting Mechanics Statics Dynamics Strength of Material Compressible Flow Fluid Mechanics Thermodynamics Heat Transfer Open Channel Flow Two/Multi phases flow progress alpha not started yet not started yet not started yet not started yet early beta alpha early alpha not started yet not started yet not started yet remarks version 0.0. Even if not complete.0 0.1 0.0 0. the traditional role of author will be replaced by an organizer who will be the one to compile the book. These data (tables. some errors are possible and expected.0.0 Based on Eckert Tel-Aviv’notes The meaning of the progress is as: • The Alpha Stage is when some of the chapters are already in rough draft. chapters or sections can be added after the skeleton has been written. other books contain data5 which can be A typeset in L TEX. But more than that.CREDITS xxxiii adding a question and perhaps the solution. while in other cases This may merely be the person who decides what will go into the book and what will not (gate keeper).0 0. the contributions to books can be done by many who are not experts. 5 Data are not copyrighted. Thus. Nevertheless.01 0. These books are intended to be “continuous” in the sense that there will be someone who will maintain and improve the book with time (the organizer).

Writing a textbook is comprised of many aspects. A new version is created every several years. a process in which books have a new version every a few years. many of whom volunteered to help. There are book(s) that have continued after their author passed away. traditionally books have been improved by the same author(s). The text processes.. and • In Gamma Stage is when all the chapters are written and some of the chapters are in a mature form. such as OpenOffice. the Boundary Layer Theory originated6 by Hermann Schlichting but continues to this day.xxxiv LIST OF TABLES • In Beta Stage is when all or almost all of the chapters have been written and are at least in a draft stage. etc.e.) are already presented. However. The idea that a book can be created as a project has mushroomed from the open source software concept. While some terms are defined in a relatively clear fashion. The mature stage of a section is when all of the topics in the section are written. other definitions give merely a hint on the status. are not appropriate for these projects. • The Advanced Stage is when all of the basic material is written and all that is left are aspects that are active. Schlichting. especially L TEX. advanced topics. Further. figures. Again. However. These chores can be done independently from each other and by more than one individual. The mature stage of a chapter is when all or nearly all of the sections are in a mature stage and have a mature bibliography as well as mature and numerous examples for every section. only L TEX. and Microsoft Word software. have the ability to produce the quality that one expects for these A writings. i. any text that is produced by Microsoft and kept in “Microsoft” format are against the spirit of this project In that they force spending money on Microsoft software. But more A than that. because of the open nature of this project. which include the actual writing of the text. and special cases. who passed way some years ago. creating diagrams and figures. and writing A the L TEX macros7 which will put the text into an attractive format. pieces of material and data can be used by different books. . and all of the examples and data (tables. are the only ones which have a cross platform ability to produce macros and a uniform feel and quality. Abiword. and perhaps troff. Word processors. projects such as the Linux Documentation project demonstrated that books can be written as the cooperative effort of many individuals. 7 One can only expect that open source and readable format will be used for this project. writing examples. 6 Originally authored by Dr. But such a thing is hard to define and should be enough for this stage. but it has roots in the way science progresses.

4.000.3 Sep. If one want to use the software engine it is okay and permitted by the license of this work. and the author is apologizing for this omission. this author cannot add all the things that he was asked and want to the book in instant fashion. I am sorry will not be entertaining work for non Linux/Unix systems.S.A.Prologue For This Book Version 0. The author was asked by a former client why he didn’t insert his improved tank filling and evacuating models (the addtion of the energy equation instead of isentropic model). Dr. the ideas of moving shock will reduced the work for many student of compressible flow. Version 0. especially for Microsoft. Menikoff pointed to this inconsistency. the time is limited and they will be inserted as time permitted. The download to this mount is over 25. However. For example. While all these requests are important.2 It was surprising to find that over 14. Several sections were add to this book with many new ideas for example on the moving shock tables. The moving shock issues are not completed and more work is needed also in the shock tube. 2006 The title of this section is change to reflect that it moved to beginning of the book. one of the reader ask why not one of the example of oblique shock was not turn into the explanation of von Neumann paradox. Nevertheless. I also got request from a India to write the interface for Microsoft. For example solving homework problem from other text books became either just two mouse clicks away or just looking at that the tables in this book. xxxv . 15./Arizona) and some other reactions.4. While it moves earlier but the name was not changed.000 downloaded and is encouraging to receive over 200 thank you eMail (only one from U.

Version 0. If the total combined variation search of terms “Fanno” and “Rayleigh” (mostly through google) is accounted.4 Since the last version (0. The a brief chapter on Prandtl–Meyer flow was added. most additions of the next version will be concentrated on Fanno flow and Rayleigh flow. most readers reached to www. In the chapter on speed of sound.org by searching for either terms “Rayleigh flow” (107) and “Fanno flow” ((93). The reason the topic of Com8 A reader asked this author to examine a paper on Triple Shock Entropy Theorem and Its Consequences by Le Roy F. the two phase aspects were added. However.3 In the traditional class of compressible flow it is assumed that the students will be aerospace engineers or dealing mostly with construction of airplanes and turbomachinery. and hopefully it will be near future.potto. The isothermal nozzle was combined with the isentropic chapter. This book is the second book in the series of POTTO project books. The fifth chapter deals now with normal shock in variable area ducts. The additions of this version focus mainly on the oblique shock and related issues as results of questions and reactions on this topic. This assumption drives students from other fields away from this knowledge. Thus. POTTO project books are open content textbooks. This premise should not be assumed. This led to comparison between maximum to ideal gas model to more general model. In the introductory chapter. Version 0. At this stage. This indicates that these topics are highly is demanded and not many concerned with the shock phenomena as this author believed and expected. The chapter about oblique shock was added and it contains the analytical solution. extensive description of the compressible flow history was written. it reaches to about 30% (2011). . Furthermore. several questions were about compressibility factor and two phase flow in Fanno flow and other kind of flow models. the connection between Prandtl–Meyer flow and oblique is an note form.3) several individuals sent me remarks and suggestions. This knowledge should be spread to other fields because it needed there as well. the questions that appear on the net will guide this author on what is really need to be in a compressible flow book. The only exception is the addition to Taylor–Maccoll flow (axisymmetricale conical flow) in Prandtl -Meyer function (currently in a note form). Henderson and Ralph Menikoff. At this time. The other questions that appeared related two phase and connecting several chambers to each other. Also. Some examples were added to the normal shock chapter.xxxvi LIST OF TABLES This textbook has sections which are cutting edge research8 . This “rejection” is especially true when students feel that they have to go through a “shock wave” in their understanding. The sixth chapter deals with external forces fields. an individual asked whether this author intended to write about the unsteady section.

These unique issues of compressible flow are to be emphasized and shown. the study of heat transfer was without any dimensional analysis. This book is written in the spirit of my adviser and mentor E. Unfortunately. there is no known place in world that does not teach according to Eckert’s doctrine. 10 The fundamental misunderstanding of choking results in poor models (research) in the area of die casting. Ernst Schmidt.). With all the above. He wrote his book because he realized that the dimensional analysis utilized by him and his adviser (for the post doc). Instead. is because of the realization that manufacture engineering simply lacks fundamental knowledge in this area and thus produces faulty designs and understanding of major processes.. which in turn results in many bankrupt companies and the movement of the die casting industry to offshore. His book met strong criticism in which some called to burn his book. Their applicability to real world processes is to be 9 Please read the undersigned’s book “Fundamentals of Die Casting Design.” which demonstrates how ridiculous design and research can be. especially for external compressible flow (e.R. almost suddenly things became clear. while important. aside from his research activity. show me what is wrong”. while relatively simple topics like fundamentals of strength of material were delayed. there were classes during which most students had a period of confusion. Who. As many instructors can recall from their time as undergraduates.VERSION 0.g. wrote the book that brought a revolution in the heat transfer field of education. The other issues. Up to Eckert’s book. This book offers a more balanced emphasis which focuses more on internal compressible flow than the traditional classes. Eckert. As a wise person says “don’t tell me that it is wrong. Today. Compressible flow is essentially different from incompressible flow in mainly two respects: discontinuity (shock wave) and choked flow. . flow around a wing. it must be emphasized that this book will not revolutionize the field even though considerable new materials that have never been published are included. manufacturing engineering. must be taught in engineering classes. etc. the undersigned observed that many researchers who are dealing with manufacturing processes are lack of understanding about fluid mechanics in general but particularly in relationship to compressible flow. This criticism will not change the future or the success of the ideas in this work. and then later. it will provide a new emphasis and new angle to Gas Dynamics.G. It is clear that an understanding of Compressible Flow is very important for areas that traditionally have ignored the knowledge of this topic10 . the lack of competitive advantage moves many of the die casting operations to off shore9 .g. when the dust settled. For example. The internal flow topics seem to be common for the “traditional” students and students from other fields. This situation is typical also for Compressible Flow classes. however. In fact one of the reasons that many manufacturing jobs are moving to other countries is because of the lack of understanding of fluid mechanics in general and compressible in particular. It is assumed that the same kind of individuals who criticized Eckert’s work will criticize this work. are not that crucial to the understanding of the unique phenomena of compressible flow. and their colleagues.3 xxxvii pressible Flow was chosen. e. this is the only reply.

is currently under construction. First. is suggested to referred to this model as Shapiro flow . which is excellent tool to explain the strong. weak and unrealistic shocks. and properties change effects (ideal gases and real gases) (under construction). The book is organized into several chapters which. 11 If 12 It you have better and different examples or presentations you are welcome to submit them. The program design and how to use the program are described (briefly). It is further hoped that others will contribute to this book and find additional use for this book and enclosed software. It is hoped that this book will serve the purposes that was envisioned for the book. for the first time analytical solution is presented. There are two advanced topics which included here: oblique shock wave. In the oblique shock. Gas Dynamics Calculator (CDC-POTTO). The fourth chapter deals with a simple case of discontinuity (a simple shock wave in a nozzle).xxxviii LIST OF TABLES demonstrated11 . The topics of filling and evacuating of gaseous chambers are presented. These chapters are presented almost independently. normally missed from traditional textbooks. again under construction. and Prandtl–Meyer expansion will be included only after the gamma version unless someone will provide discussion(s) (a skeleton) on these topics. etc). The last chapter deals with the computer program. the Isothermal flow is introduced because of the relative ease of the analytical treatment. The third chapter provides the first example of choked flow (isentropic flow in a variable area). The chapter on one-dimensional unsteady state. Fanno flow and Rayleigh flow. The second chapter deals with speed of sound. deals with a basic introduction of thermodynamics concepts (under construction). Discussions on the flow around bodies (wing. The next three chapters are dealing with three models of choked flow: Isothermal flow12 . as a traditional textbook. Isothermal flow provides useful tools for the pipe systems design. The next chapter is dealing with isothermal flow with and without external forces (the moving of the choking point). Every chapter can be “ripped” out and printed independently.

The presentation of some of the chapters is different from other books. even my 10 years old son. I realized that people need down to earth book about compressible flow and this book was born.How This Book Was Written This book started because I needed an explanation for manufacturing engineers. There is no reason why a text book which cost leas than 10$ to publish/produce will cost about 150 dollars. this book is designed to be in such form that is easy to understand. I realized that books in compressible flow are written in a form that is hard for non fluid mechanic engineer to understand. I wrote notes and asked myself what materials should be included in such a book so when I provide consultation to a company. the best books can be created. there are some chapters in this book which are original materials never published before. If a community will pull together.” Later. For example. Anyone can be part of it. Therefore. It should be included because this section provides the explanation and direction on what Fanno flow model will approach if heat transfer is taken into account13 . He made me realized that the common approach to supersonic branch of isothermal as non–existent is the wrong approach. Apparently many manufacturing engineers and even some researchers in manufacturing engineering were lack of understanding about fluid mechanics in particularly about compressible flow. The book 13 Still in untyped note form. It was farther extended by the allegation of free market and yet the academic education cost is sky rocketing without a real reason and real competition. Therefore. The free/open content of the book was created because the realization that open content accelerated the creation of books and reaction to the corruption of the court implementing the copyright law by manufacturing facts and laws. xxxix . Therefore. “Fundamentals Of Die Casting Design. Eliezer made me change the chapter on isothermal flow. I do not need to explain the fundamentals. I wrote to myself some notes and I converted one of the note to a chapter in my first book.

this book was written on Linux (MicrosoftLess book). . graphic designer. a program that cannot be used on new system and I had to keep my old Linux to make it work14 . The old figures where done by grap (part the old Troff). Grace is a problematic program but is the best I have found. This book was written using the vim editor for editing (sorry never was able to be comfortable with emacs).xl LIST OF TABLES does not provide the old style graphical solution methods yet provide the graphical explanation of things. The graphics were done by TGIF. The figure in cover page was created by Michael Petschauer. Of course. please contact me. 14 If you would like to to help me to write a new spell check user interface. Unfortunately.com). The spell checking was done by gaspell. I did not have any access to grap and switched to Grace. and is open/free content copyright by him ( happy circle@yahoo. the best graphic program that this author experienced so far. I hope someone will write a new spell check so I can switch to a new system.

The graphical interface was created only after the engine was written. the program changed to contain finding parameters for a given 4D a primitive interface to provide parameters to carry out the proper calculations. So. the code became a new program to find the root of something between the values of the tables e. The graphical interface was written to provide a filter for the unfamiliar user. the stagnation table was one such program. At that stage.3 This version add several feature among them is the shock dynamics calculation with the iteration. Yet. (Potto–GDC) was created to generate various tables for the book either at end the chapters or for the exercises. The last freature is good for homework either for the students or the instroctors. I decided to include Potto–GDC to the book. Version 4. then.g. xli . Later. every flow model was a different program. It also remove the need to recompile the code everytime. For example. When it become cumbersome to handle several programs. a need to intermediate stage of tube flow model (the PipeFlow class) was created and new classes were created. fL . the Potto-GDC was many small programs for specific tasks. the author utilized the object oriented feature of C++ and assigned functions to the common tasks to a base class and the specific applications to the derived classes. Initially. This calculator was given to several individuals and they found Potto–GDC to be very useful.About Gas Dynamics Calculator Gas Dynamic Calculator. Later.

7 had several bug fixes and add two angle calculations to the oblique shock. Change the logtable to tabular environment for short tables.1.xlii LIST OF TABLES Version 4.1.7 Version 4. .

16 At 15 To xliii . without form. However. This book is designed to replace the book(s) or instructor’s notes for the compressible flow in (mostly) undergraduate classes for engineering/science students. Nevertheless. You have to remember that this book is a work in progress. This means that a lot of information is presented which is not necessary for everyone. and secondly a reference manual only as a lucky coincidence. Fundamentals of Compressible Flow. Let This book. and basic science such as calculus. Fanno flow. etc. This book is written and maintained on a volunteer basis. Like all volunteer work. These explanations have been marked as such and can be skipped. And the Fingers of the Author the face of the keyboard. The structure of this book is such that many of the chapters could be usable independently. you can read just chapter 9. And the Author there be words. of the bits moved upon said. I hope this makes the book easier to use as a reference manual. the POTTO project was and void. describes the fundamentals of compressible flow phenomena for engineers and others. and there were words. the book is not well organized. I believe professionals working in many engineering the power and glory of the mighty God. Moreover. the present. I have tried to describe why the theories are the way they are. say. and emptiness was upon the face and files. physics. the explanations are not as good as if I had a few years to perfect them. It is hoped that the computer program enclosed in the book will take on a life of its own and develop into an open content or source project.Preface "In the beginning. naturally. there is a limit on how much effort I was able to put into the book and its organization. For example. increase your understanding of the fundamentals of compressible fluid flow. if you need information about." 15 . this manuscript is first and foremost a textbook. rather than just listing “seven easy steps” for each task. This book is only to explain his power. It is hoped that the book could be used as a reference book for people who have at least some knowledge of the basics of fundamental fluid mechanics. due to the fact that English is my third language and time limitations.16 Reading everything will.

but those are solely for the author’s purposes. and assistance. ignore them please. I am interested in it all. I have tried to make this text of the highest quality possible and am interested in your comments and ideas on how to make it better. Even reaction/comments from individuals like David Marshall17 Several people have helped me with this book. You may contact me via Email at “barmeir@gmail. . This book contains many original models. this process leaves a large room to blockage of novel ideas and plagiarism. this book contains material that never was published before. If you have Internet e-mail access. other email that imply that someone will take care of this author aren’t appreciated. While his comment looks like unpleasant reaction. G. This is mostly for the author’s purposes and also for your amusement. I hope to improve or to add to these areas in the near future. I have left some issues which have unsatisfactory explanations in the book. more mathematics (or less mathematics). or proofreading. I also would like to thank Amy Ross for her advice ideas. and material knowledge and a desire to provide open content textbooks and to improve them to join me in this project. you can contact me at “barmeir@gmail. directly or indirectly. While peer review and publication in a professional publication is excellent idea in theory. 17 Dr. Incorrect language. The symbol META was added to provide typographical conventions to blurb as needed. rewritten sections. I hope that many others will participate of this project and will contribute to this book (even small contributions such as providing examples or editing mistakes are needed). Furthermore. In practice. I would like to especially thank to my adviser. marked with a Mata mark. literature review is always good?). Over ten individuals wrote me about this letter. errors. E. Naturally. If you want to be involved in the editing. A L TEX knowledge. editing. it brought or cause the expansion the oblique shock chapter.com”. I encourage anyone with a penchant for writing. They will be removed gradually as the version number advances. I am asking from everyone to assume that his reaction was innocent one.com”. This material never went through a peer review. please drop me a line. more fundamental material. If you would like be “peer reviews” or critic to my new ideas please send me your idea(s).xliv LIST OF TABLES fields will benefit from this information. R. There are also notes in the margin. whose work was the inspiration for this book. and explanations never published before. Dr. ideas for new areas to cover. graphic ability. Eckert. graphic design. However. Marshall wrote to this author that the author should review other people work before he write any thing new (well.

It A is hoped the changes in TEX and L TEX related to this book in future will be minimal and minor. Specific missing parts from every chapters are discussed below. question. There will always new problems to add or to polish the explanations or include more new materials. illustration or photo of experiment. One of the virtue of this book lay in the fact that it contains a software that is extensible. For example. These omissions. Further topics related to non-ideal gas such as steam and various freons are in the process of being added to this book especially in relationship to Fanno flow. there are specific issues which are on the “table” and they are described herein. Nevertheless. On occasions a footnote was used to point out for a need of improvement. You are always welcome to add a new material: problem. mistakes. approach problems are sometime appears in the book under the Meta simple like this sample this part. Meta End Questions/problems appear as a marginal note. This part will be incorporated in the future hopefully with the help of others. Also issues that associated with the book like the software has to be improved. several chapters are missing. The effects of the deviations from the ideal gas model on the properties should be included. Material can xlv . the Fanno module can be extended to include effects of real gases.To Do List and Road Map Meta This book is not complete and probably never will be completed. At this stage. It is hoped that the style file will be converged to the final form rapidly.

Normal Shock Extend the partialy (open/close) moving shock theory. The combined effects of isentropic nozzle with heat transfer (especially with relationship to the program. Speed of Sound Discussion about the movement in medium with variation in speed of sound.). More problems in relationship to two phase. The dimensional analysis of the flow when the flow can be considered as isothermal. Stagnation effects Extend the applicability with examples Cp as a function of temperature (deviation of ideal gas model) “real gas”’ like water vapor History – on the teaching (for example when the concept of stagnation was first taught. Real gases effects (only temperature effects) Flow with “tabulated gases” calculations Phase change and two phase flow (multi choking points) effects (after 1. Additional material can be provided to give a different angle on the issue at hand. Mixed gases and liquids. This concept in relation of the wind tunnel and atmosphere with varied density and temperature. Nozzle The effect of external forces (add problems). Speed of sound in wet steam.0 version). Provide more examples on the preveious topic Shock in real gases like water vapor Shock in (partially) two phase gases like air with dust particals Isothermal Flow Classification of Problems Comparison of results with Fanno flow Pipes Network calculations. .xlvi LIST OF TABLES be further illuminate.

The FORTRAN version will not be included.7 xlvii Fanno Flow More examples: various categories Some improvement on the software (clean up) Real gas effects (compressible factor) Tablated gas Rayleigh Flow To mature the chapter: discussion on the “dark” corners of this model. Provide discussion on variations of the effecting parameters. Application Marcell–Taylor (from the notes) Examples Transient problem . Oblique Shock Add application to design problems Real Gas effects Prandtl–Meyer The limitations (Prandtl-Meyer).VERSION 4.1. Examples: provide categorization Evacuation and filling semi rigid Chambers To construct the Rayleigh flow in the tube (thermal chocking) Energy equation (non isentropic process) Examples classifications Software (converting the FORTRAN program to c++) Evacuating and filling chambers under external forces Comparison with chemical reaction case Energy equation (non isentropic process) Examples Software transformation from FORTRAN to c++.

xlviii LIST OF TABLES .

Here one has to learn to use his intuition as a tool for future use. Thus. one has to realize that intuition is really a condition where one uses his past experiences to predict other situations. not only aeronautic engineers. Further. The shock wave and choking are not intuitive for most people. Thus. 3 The thermal choking is somewhat different but similarity exists. But attempt is made to present the material in introductory level.1 What is Compressible Flow ? This book deals with an introduction1 to the flow of compressible substances (gases). Rather. which are beyond a critical value(s). 4 This book is intended for engineers and therefore a discussion about astronomical conditions isn’t presented.CHAPTER 1 Introduction 1. However. the difference is in two phenomena that do not exist in incompressible flow2 . but other engineers. Choking is referred to as the situation where downstream conditions. etc. The main difference between compressible flow and almost incompressible flow is not the fact that compressibility has to be considered. this richness is unique to gas dynamics. 1 . 2 It can be argued that in open channel flow there is a hydraulic jump (discontinuity) and in some ranges no effect of downstream conditions on the flow. 1 This book gradually sliding to include more material that isn’t so introductory. there also are situations of choking in general (external) flow4 . Though choking occurs in certain pipe flows in astronomy. and even manufacturing engineers will be able use this “intuition” in design and even research. The first phenomenon is the very sharp discontinuity (jump) in the flow in properties. The second phenomenon is the choking of the flow. the uniqueness of the phenomena in the gas dynamics provides spectacular situations of a limited length (see Fanno model) and thermal choking. doesn’t affect the flow. However. there is no equivalent to oblique shock wave. Choking is when downstream variations don’t effect the flow3 .

.2 CHAPTER 1. Clearly. therefore the compressibility has to be taken into account. the control engineers who are using pneumatic systems use compressed substances. For instance. The above flows that were mentioned are called internal flows. This doesn’t in any way meant that the important topics such as shock wave and oblique shock wave should be neglected. 6 Hunter Rouse and Simon Inc. For manufacturing engineers there are many situations where the compressibility or compressible flow understating is essential for adequate design. Some aspects of these systems require consideration of the unique phenomena of compressible flow. should be considered a compressible flow. The cooling of some manufacturing systems and design of refrigeration systems also utilizes compressed air flow knowledge. and also gas turbine. mostly aluminum. History of Hydraulics (Iowa City: Institute of Hydraulic Research. and is considered an external flow. including steam. 1957) . etc. injection molding. Aero Engineers are not the only ones who have to deal with some aspect of compressible flow. While there are excellent books about the history of fluid mechanics (hydraulic) see for example book by Rouse6 . There are very few courses that deal with isothermal flow. natural gas. is injected into a mold to obtain a near final shape. Traditionally. INTRODUCTION 1. a common method of heating in the u. most gas dynamics (compressible flow) classes deal mostly with shock waves and external flow and briefly teach Fanno flows and Rayleigh flows (two kind of choking flows).s. There are numerous sources dealing with the history of flight and airplanes (aeronau5 Any search on the web on classes of compressible flow will show this fact and the undersigned can testify that this was true in his first class as a student of compressible flow. many books on compressible flow ignore the isothermal flow5 . In this book. a greater emphasis is on the internal flow.2 Why Compressible Flow is Important? Compressible flow appears in many natural and many technological processes. Compressible flow also includes flow around bodies such as the wings of an airplane. The die casting process is a process in which liquid metal. a problem that led to the Fanno flow model. The air is displaced by the liquid metal in a very rapid manner. These processes include the flow of gas in the exhaust system of an internal combustion engine.3 Historical Background In writing this book it became clear that there is more unknown and unwritten about the history of compressible fluid than known. in a matter of milliseconds. 1. This book contains several chapters which deal with external flow as well. such as manufacturing process such as the die casting. In fact. For instance. the flow of natural gas in a pipe system. These processes include situations not expected to have a compressible flow. nitrogen and helium. Compressible flow deals with more than air.

There is no book8 that describes the history of these models.. researchers were aware that the speed of sound is some kind of limit. which are the core of gas dynamics. Sometimes. Mach observed shock and Fliegner J. for many early contributors this topic was a side issue. Furthermore. . A History of Aerodynamics: And Its Impact on Flying Machines. This was Lord Rayleigh. 8 The only remark found about Fanno flow that it was taken from the Fanno Master thesis by his adviser. Aeronautics is an overlapping part of compressible flow. Even before Prandtl’s idea of boundary layer appeared. D. however these two fields are different. Is there any earlier reference to this model? 7 Anderson. It must be noted that researchers were interested in the speed of sound even long before applications and knowledge could demand any utilization. Thus. to the undersigned. Cambridge University Press. the Fanno flow and isothermal flow.1. Possible reasons for the lack of written documentation are one. and two. are not part of aerodynamics. only one contributor of the three main models of internal compressible flow (Isothermal. Here is a challenge: find any book describing the history of the Fanno model. unknown. At this stage there are more questions about the history of compressible flow needing to be answered. the significant and importance of compressibility emerged. In fact. for whom the Rayleigh flow was named. respectively. Additional example. HISTORICAL BACKGROUND 3 tic)7 . For instance. The early application in which compressibility has a major effect was with fire arms. For example. In the first half of nineteen century there was little realization that the compressibility is important because there were very little applications (if any) that required the understanding of this phenomenon. 1997. the question. The technological improvements in fire arms led to a gun capable of shooting bullets at speeds approaching to the speed of sound. could not be answered by any of the undersigned’s colleagues in University of Minnesota or elsewhere. they will appear in a footnote like this9 . Rayleigh) was described by any text book. a large part of this knowledge is relatively new. these questions will appear in a section with a title but without text or with only a little text. taking compressible flow as if it were incompressible flow. this author did not find any reference to isothermal flow model earlier to Shapiro’s book. For example. Mach and Fliegner “stumbled” over the shock wave and choking. it is obvious that Shapiro published the erroneous conclusion that all the chocking occurred at M = 1 in his article which contradicts his isothermal model. Perhaps this flow should be named after the Shapiro. and when did he live. The other two models were. who was the first to “conclude” the “all” the chocking occurs at M = 1? Is it Shapiro? Originally. Cambridge. The research and interest in the speed of sound was a purely academic interest. Sometimes. In the second half of the nineteen century. As there were no motivations to investigate the shock wave or choked flow both were treated as the same.3. who was Fanno. 9 Who developed the isothermal model? The research so far leads to Shapiro. Jr. Some researchers even have suggested that compressibility can be “swallowed” into the ideal flow (Euler’s equation’s flow is sometimes referred to as ideal flow). Fanno. England. there was no idea that there are special effects and phenomena of compressible flow.

aerospace engineering played a significant role in the development of this knowledge. Early contributors were motivated from a purely intellectual curiosity. 1. Newton’s equation is missing the heat ratio. who discovered that sound travels in waves (1500). INTRODUCTION To add history from the work. In the twentieth century the flight industry became the pushing force. people found that sound travels in different speeds through different mediums. However. During that time. Maxwell was the first to derive the speed of kRT from particles (statistical) mechanics. dominated the field. .” Contributions by researchers and educators from other fields were not as dominant and significant. √k (late 1660’s). for a long time the question of the speed of sound was bounced around.). sound) was a major part of Galileo’s work. Therefore some sound for gas as c = √ referred to coefficient k as Maxwell’s coefficient. capillary flow and phase change. Topics that should be included in this history review but that are not yet added to this section are as follows: Multi Phase flow. so almost all text books in this field are written from an aerodynamic prospective. Giants like Prandtl and his students like Van Karman. Understandably.1 Early Developments The compressible flow is a subset of fluid mechanics/hydraulics and therefore the knowledge development followed the understanding of incompressible flow. As a result.4 CHAPTER 1. Dealing with the difference between the two speeds (light. Galileo. Newton was the first to formulate a relationship between the speed of sound in gases by relating the density and compressibility in a medium (by assuming isothermal process). an Italian scientist. Other milestones in the speed of sound understanding development were by Leonardo Da Vinci. the modern basic classes became “solidified. The early approach to the speed of sound was by the measuring of the speed of sound. was one of the earliest contributors to our understanding of sound. There is no clear evidence as to who came up with this concept. measured the choking but theoretical science did not provide explanation for it (or was award that there is an explanation for it.3. Speed of Sound The idea that there is a speed of sound and that it can be measured is a major achievement. A possible explanation to this discovery lies in the fact that mother nature exhibits in every thunder storm the difference between the speed of light and the speed of sound. as well as others like Shapiro. but some attribute it to Galileo Galilei: 166x. while most later contributions were driven by necessity. Robert Boyle discovered that sound waves must travel in a medium (1660) and this lead to the concept that sound is a pressure change. once there was a realization that sound can be measured. Marin Mersenne was the first to measure the speed of sound in air (1640).

The other kind of problem is “hijacking” by a sector. Ec. From Classic Papers in Shock Compression Science. 13 James Challis. science is full of many stories of conflicts and disputes. 1803-1882. So. New York. It is widely believed that Mach had done his research as purely intellectual research. Even on this subject. Yet for many it was still debatable up to the middle of the nineteen century. Poisson found a “solution” to the Euler’s equations with certain boundary conditions which required discontinuity12 which had obtained an implicit form in 1808. 14 (1808). ”M’emoire sur la th’eorie du son. Springer. In the early 18xx. Major material like Fanno flow isn’t taught in many classes. It was easier to publish articles about the second law of thermodynamics than to convince anyone of the first law of thermodynamics. Shock Compression Condens. Matter. 12 Simeon ´ Denis Poisson. University Press. Polytech. French mathematician. with additional notes by the author. Neither of these groups would agree to “merge” or “relinquish” control of their “territory” to the other. 3-65.” Philos. However. in his 1860 thesis.1. conservation of energy was a concept that was applied only to mechanical energy. the Aeronautic sector “took over” gas dynamics as did the emphasis on mathematics like perturbations methods or asymptotic expansions instead on the physical phenomena. 1998. He was mostly interested in experimental physics. Cambridge. . Later. 494-499 14 Stokes George Gabriel Sir. 11 This recognition of the first law is today the most “obvious” for engineering students. Mathematical and Physical Papers. was not sure whether or not discontinuity is only a mathematical creature or a real creature. Mag.” J. 1781-1840 worked in Paris. Riemann. At this stage some experimental evidence was needed. France. a different group of scientists dealt with calorimetry (internal energy). it is logical to believe that his interest had risen due to the need to achieve 10 Amazingly.”14 Stokes was convinced by Lord Rayleigh and Lord Kelvin that he was mistaken on the grounds that energy is conserved (not realizing the concept of internal energy). He had then correctly derived the jump conditions that discontinuous solutions must satisfy. 319-392.3. Again the “jumping” conditions were redeveloped by two different researchers independently: Stokes and Riemann. At first. XXXII (1848). Reprinted from the original journals and transactions. while many of the mathematical techniques are currently practiced. It took about a century to establish the first law11 . On the other side. these problems are more common than one might be expected. Poisson showed that solutions could approach a discontinuity by using conservation of mass and momentum. perhaps the most famous is that of Newton’s netscaping (stealing and embracing) Leibniz[’s] invention of calculus. There are even conflicts from not giving enough credit. ”On the velocity of sound. 1880-1905. Stokes in 1848 retreated from his work and wrote an apology on his “mistake. England UK. like Moody not giving the due credit to Rouse.2 The shock wave puzzle Here is where the politics of science was a major obstacle to achieving an advancement10 . High-press.3. HISTORICAL BACKGROUND 5 1. Ernst Mach studied several fields in physics and also studied philosophy. Aside from the conflicts of scientists with the Catholic Church and Muslim religion. since no Mach number was known at that time). Challis had noticed contradictions concerning some solutions of the equations of compressible gas dynamics13 . English Astronomer. worked at Cambridge. His research centered on optic aspects which lead him to study acoustic and therefore supersonic flow (high speed. Even the undersigned encountered individuals who have tried to ride on his work. The major breakthrough in the understanding of compressible flow came when Ernest Mach “stumbled” over the discontinuity.

He also observed the existence of a conical shock wave (oblique shock wave). one anecdotal story suggests a way to make money by inventing a better killing machine for the Europeans. and energy equations18 lead to a solution which ties the upstream and downstream properties. 1820-1872. 58 (1889). II” J. Mach’s paper from 1876 deals with the flow around bullets. 133-147. he was not able to analyze it (neither was he aware of Poisson’s work or the works of others. Springer. What makes Mach’s achievement all the more remarkable was the technique he used to take the historic photograph: He employed an innovative approach called the shadowgraph. Classic papers in shock compression science. 1998 17 Pierre Henri Hugoniot. defense techniques started to appear such as sand bags. Polytech. ”On the thermodynamic theory of waves of finite longitudinal disturbance. Back to the pencil and paper. 16 William John Macquorn Rankine. Scottish engineer. At that time many inventions dealt with machine guns which were able to shoot more bullets per minute. Trans. Mach’s paper from 1876 demonstrated shock wave (discontinuity) and suggested the importance of the ratio of the velocity to the speed of sound. Nevertheless. part II. Therefore. Later. Matter. momentum. Shock was assumed to be one dimensional and mass. French engineer. Matter. 1998 18 Today it is well established that shock has three dimensions but small sections can be treated as one dimensional. the jump conditions were redeveloped and now named after Rankine16 and Hugoniot17 . . he brought confidence to the theoreticians to publish their studies. High-press. It is possible that some insight was developed but none of the documents that were reviewed revealed it to the undersigned. While Mach proved shock wave and oblique shock wave existence.7). He further showed that the discontinuity occurs at M = 1 and realized that the velocity ratio (Mach number).). and not the velocity. New York. i. 160 (1870). INTRODUCTION powerful/long–distance shooting rifles/guns. Mach’s contributions can be summarized as providing an experimental proof to discontinuity. Shock Compression Condens. Rankine and Hugoniot. While the machine gun turned out to be a good killing machine. At the time. is the important parameter in the study of the compressible flow. I. direction of the flow. Classic papers in shock compression science. ”Sur la propagation du mouvement dans les corps et sp’ecialement dans les gaz parfaits.e. Ec. Scotland UK. 161-243. Thus. Mach used his knowledge in Optics to study the flow around bullets. A need for bullets that could travel faster to overcome these obstacles was created. 1-125. He utilized the variations of the air density to clearly show shock line at the front of the bullet. Mach had good understanding of the fundamentals of supersonic flow and the effects on bullet movement (supersonic flow). no known15 equations or explanations resulted from these experiments. He was the first to photograph the shock wave. 245-358. He worked in Glasgow. High-press. 277-288. 3-97. What they could not prove or find was that shock occurs only when upstream is supersonic. others expanded Rankine-Hugoniot’s 15 The words “no known” refer to the undersigned. Springer.” Philos. redeveloped independently the equation that governs the relationship of the shock wave. Shock Compression Condens. New York. 57 (1887). In his paper discussing ”Photographische Fixierung der durch Projektile in der Luft eingeleiten Vorgange” he showed a picture of a shock wave (see Figure 1..6 CHAPTER 1. 1851-1887.

Berlin. Thus. HISTORICAL BACKGROUND conditions to a more general form19 . I. ¨ Forsch-Arb. The “last nail” of understanding was put by Landau. It was probably later that Stodola (Fanno’s adviser) realized that the shock is the intersection of the Fanno line with the Rayleigh line. the supersonic branch is missing from his understanding (see Figure (1. The first full analytical solution connecting the angles with upstream Mach number was published in this book version 0. 19 To . Today this equation/condition is known as Prandtl’s equation or condition (1908). 20 Some add discussion about the general relationships. No. The probable reason that analytical solution was not published because the claim chapter. view the work of G. 62. 1908. to be add to oblique shock Since early in the 1950s the analytical relationships between the oblique shock.3. and Mach number was described as impossible to obtain. The fact that the conditions and direction were known did not bring the solution to the equations. the second law has been around for over 40 years and yet the significance of it was not was well established. Theodor Meyer (Prandtl’s student) derived the conditions for oblique shock in 190821 as a byproduct of the expansion work. Steam and Gas was found by Landau & Lifshitz Turbine and expanded by Kolosnitsyn & Stanyukovich (1984). 1.1. a Jewish scientist who worked in Moscow University in the 1960’s during the ComFig. deflection angle.1)).1: The shock as connection of Fanno and munist regimes. Yet. Taylor from England as the proof (of course utilizing the second law) 21 Theodor Meyer in Mitteil. it took over 50 years for Prandtl to arrive at and to demonstrate that the shock has only one direction20 . A solution Rayleigh lines after Stodola. ub. the relationship between upstream Mach number and the angles. Stodola suggested the graphical solution utilizing the Fanno line. In fact Prandtl is the one who introduced the name of Rankine-Hugoniot’s conditions not aware of the earlier developments of this condition. 7 Here.3. In fact. page 62.3 of this book) several equations that tied various properties/quantities for example. There were until recently (version 0. shock angle.

recently.” Proceedings of the Cambridge philosophical society.. That solution is based on a transformation of sin θ to tan β . The question whether the oblique shock is stable or which root is stable was daunting since the early discovery that there are more than one possible solution. In a way. even this author sinned in this erroneous conclusion. 25 Herivel. Vol. In general the mathematical treatment was left to numerical investigation and there is relatively very little known on the difference between ideal gas model and real gas. The presentation was missing the condition for the detachment or point where the model collapse. p. It is amazing that early research concluded that only the weak solution is possible or stable as opposed to the reality. 23 Epstein.. 179–210.90). While the actual solution all the roots are real and positive for the attached shock. 24 In study this issue this author realized only after examining a colleague experimental Picture 14. “On the Stability of the supersonic Flows Past as a Wedge. 1931. 1955. In fact. Vol. Carrier26 was first to recognize that strong and weak shocks stable. similar analysis was published by Briggs. . His analysis was based on Hamilton’s principle when he ignore the boundary condition. Yet. pp.8 CHAPTER 1.” Quarterly of Applied Mathematics. G. The effect of real gases was investigated very early since steam was used move turbines. If fact the confusion on this issue was persistent until now. pp. Inc. No 5 p. 1963. George. 1948. Pt.” Proceedings of the National Academy of Science. part of analysis of this book is also redoing old work. 17. 1949. For example. pp. J. The real question isn’t if they exist rather under what conditions these shocks exist which was suggested by Courant and Friedrichs in their book “Supersonic Flow and Shock Waves. In fact analysis based on Hamilton’s principle isn’t suitable for stability because entropy creation was recognized 1955 by Herivel25 . The first that attempt this question where in 1931 by Epstein23 .4 that it was clear that the Normal shock along with strong shock and weak shock “live” together peacefully and in stable conditions. Hence. 532-547. This difference probably results in not recognizing that The boundary conditions play an important role in the stability of the shock24 . 2. 6. 344-349. The researchers understood that flow after a strong shock was governed by elliptic equation while the flow after a weak shock was governed by hyperbolic equations. F. 317. 26 Carrier. “Comment on Calculation of Oblique shock waves.F.” AIAA Journal Vol 2. 367–378. The results of that analysis was that strong shock is unstable. It is interesting that transformation result in one of root being negative. 51. Even all books that were published recently claimed that no strong shock was ever observed in flow around cone (Taylor–Maccoll flow). But more surprisingly. New York. J. INTRODUCTION in the famous report of NACA 1135 that explicit analytical solution isn’t possible22 .” Journal of Fluid Mechanics 366 (1998) pp. Vol. S. Emanuel’s partial solution just redone 36 years work (how many times works have to be redone in this field). P. 974.” published by Interscience Publishers. “The Derivation of The Equations of Motion On an Ideal Fluid by Hamilton’s Principle. but no comparison between real gases and ideal gas is offered there. 22 Since writing this book. second edition. December 2000 (US$ 124. what is new in this work is completeness of all the three roots and the analytical condition for detached shock and breaking of the model. Henderson and Menikoff27 dealt with only the procedure to find the maximum of oblique shock. ”Triple Shock Entropy Theorem. “On the air resistance of Projectiles. several individuals point out that a solution was found in book “Analytical Fluid Dynamics” by Emanuel.. 27 Henderson and Menikoff..

It is likely that others had noticed that flow is choked. Up to this version 0. Choking wasn’t clearly to be observed. 1. Steam and Gas Turbine operation by another inventor (Curtis) 1896 used in steam turbine. but did not produce any model or conduct successful experimental work.1. an explicit analytical solution for shock a head of piston movement (special case of open valve) was originally published in this book for the first time. . As a result deLavel’s nozzle was invented by Carl Gustaf Patrik Fig. 28 Fliegner Schweizer Bauztg.” Leipzig 1899.4. The realization that in most cases the moving shock can be analyzed as steady state since it approaches semi steady state can be traced early of 1940’s. page 268 f.3. Later. “Theorie die Turbinen. there was no realization that the flow is choked just that the flow moves faster than speed of sound. According the thermodynamics theory (various cycles) the larger heat supply for a given temperature difference (larger higher temperature) the larger the output.3 Choking Flow The choking problem is almost unique to gas dynamics and has many different forms. Yet. p.3 of this book (as far it is known. No one was looking for or expecting the choking to occur. Vol 31 1898. The introduction of the steam engine and other thermodynamics cycles led to the choking problem. 1. even when researcher stumbled over it. The first to discover (try to explain) the choking phenomenon was Rayleigh29 . The first experimental choking phenomenon was discovered by Fliegner’s experiments which were conducted some time in the middle of 186x28 on air flow through a converging nozzle. 29 Rayleigh was the first to develop the model that bears his name. this book is first to publish this tables). trial and error method was the only method to solve this problem. HISTORICAL BACKGROUND 9 The moving shock and shock tube were study even before World War Two. 68–72.3. Only after the dimensionless presentation of the problem and the construction of the moving shock table the problem became trivial. The theoretical first work on this issue was done by Zeuner. leading to the study and development of Rayleigh flow.2: The schematic of deLavel’s turbine afde Laval in 1882 and first successful ter Stodola. The problem was introduced because people wanted to increase the output of the Engine by increasing the flames (larger heat transfer or larger energy) which failed.. and when it was found the significance of the choking phenomenon was not clear. but after a certain point it did matter (because the steam was choked).

33 Parenty. Berlin 1908. Namely. 35 These researchers demonstrate results between two extremes and actually proposed this idea. of the publications were not named after Prandtl but rather by his students like Meyer. Uber ¨ zweidimensionals Bewegungsvordange eines Gases. this author adapts E. Th.10 CHAPTER 1. Vol 12. Recently this concept was proposed (not explicitly) by Dutton and Converdill (1997)35 .. . The actual results will be determined by the modified Eckert number to which model they are closer. Later Romer at el extended it to isothermal variable area flow (1955). Later continue by Prandtl’s group31 starting 1904. 8.G. In the literature appeared reference to article by Lorenz in the Physik Zeitshr. Perhaps. erschienen in den Mitteilungen uber ¨ Forsch. as if in 1904. there are also other works that this author did not come crossed. However. that the presentation here suggests that topic should be presented case between two extremes. Ing. 8 1896. Chim. Leipzig 1899 page 268 f. It was commonly believed34 that the choking occurs only at M = 1. 116. Experimental work by Parenty33 and others measured the pressure along the converging-diverging nozzle. 1897. In this book. 31 Some 30 Zeuner. Vol.-Arb. Ann. 113. The √ first one to analyzed that choking occurs at 1/ k for isothermal flow was Shapiro (195x). “Theorie der Turbinen.-Wes. INTRODUCTION After the introduction of the deLavel’s converging–diverging nozzle theoretical work was started by Zeuner30 . Ecert’s idea of dimensionless parameters control which determines where the reality lay between the two extremes. in many cases the reality is somewhere between the adiabatic and the isothermal flow. 119. Paris. It is so strange that a giant like Shapiro did not realize his model on isothermal contradict his conclusion from his own famous paper. heft 62. In 1908 Meyer has extend this work to make two dimensional calculations32 . Theodor. Phys.. 32 Meyer.R. 34 The personal experience of this undersigned shows that even instructors of Gas Dynamics are not aware that the chocking occurs at different Mach number and depends on the model. Vol. Dissertation 1907. Comptes R.

Between the turn of the century Fig. The arm’s circular motion could reach a velocity above the speed of sound at its end. when the first supersonic wind tunnel was built. 36 Bendemann Mitteil uber ¨ Forschungsarbeiten. in 1904 the Wright brothers demonstrated that results from the wind tunnel and spinning arm are different. Bendemann 36 carried experiments to study the accuracy of these flow meters and he measured and refound that the flow reaches a critical value (pressure ratio of 0. . At the end of the arm was the object that was under observation and study.3: The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines and 1947-48.1.545) that creates the maximum flow rate. 37.3. 1907. As a result. There are two main models or extremes that describe the flow in the nozzle: isothermal and adiabatic. No. Yet. Berlin. HISTORICAL BACKGROUND Nozzle flow 11 The first “wind tunnel” was not a tunnel but a rotating arm attached at the center. 1. due to the circular motion. A different reason to study the converging-diverging nozzle was the Venturi meter which was used in measuring the flow rate of gases. several models that explained choking at the throat have been built. the spinning arm was no longer used in testing.

the material reaches choking point due to heat transfer. In general when the model is assumed to be isothermal√ the choking occurs at 1/ k. 398 December 1955. 27 no 322. In dimensionless form. Piosson was the first one to realize the shock wave possibility. However.4: Flow rate as a function of the back pressure taken from Stodola 1927 Steam and Gas Turbines were the first ones to real√ ized that choking can occurs at different Mach number (1/ k other then the isothermal pipe. there is also a possibility that someone found it earlier. The concept that the choking point can move from the throat introduced by38 a researcher unknown to this author. In Rayleigh flow. Rayleigh’s work was during the time when it was debatable as to whether there are two forms of energies (mechanical. the first to suggest a model for frictionless flow with a constant heat transfer. Fanno Flow The most important model in compressible flow was suggested by Gino Fanno in his Master’s thesis (1904). It is very interesting that the isothermal nozzle was proposed by Romer at el 1955 (who was behind the adviser or the student?). The 1/ k question. Rayleigh looked at flow without mechanical energy transfer (friction) but only thermal energy transfer. Romer et al37 analyzed the isothermal flow in a nozzle. It is remarkable that √ choking was found as 1/ k as opposed to one (1). thermal). some astronomy books showing this effect in a dimensional form without mentioning the original researcher. this phenomenon produces a dimensionless number similar to Ozer number and therefor the name Ozer number adapted in this book. For example. Rudolf 38 Romer. to insert information about the detonation wave and relationship to Rayleigh line. according to Dr. even though Watt and others found and proved that they are the same.” Aircraft Eng. Yet. p. The model bears his name. INTRODUCTION to insert the isothermal nozzle with external forces like gravity and to show that choking location can move depending on the direction of the force. 39 As most of the history research has shown. Rayleigh Flow Rayleigh was probably39 . . and Ali Bulent Cambel.. 38 This undersign didn’t find the actual trace to the source of proposing this effect. Therefore. These researchers Fig. no additional flow can occur. hence term “thermally choked” is used. vol. “Analysis of Isothermal Variable Area Flow. 1. To find where Rayleigh did understand that √ his model leads to 1/ k point flow and graphical representation √ of the flow. I Carl Jr.12 Nozzle flow CHAPTER 1.

known by the undersigned. At that stage.8 (relative of velocity of the body to upstream velocity) a local Mach number (local velocity) can reach M = 1.3.). The arrival of the Moody diagram.1. When was Gas Dynamics (compressible flow) as a separate class started? Did the explanation for the combination of diverging-converging nuzzle with tube for Fanno flow first appeared in Shapiro’s book? expanding model by others The earliest reference to isothermal flow √ was found in Shapiro’s Book. Shapiro organized all the material in a coherent way and made this model useful. the question cannot be answered yet. Without the existence of the friction factor data. but can be explained by choking at a lower Mach number. the flow is √ choked somewhere between 1/ k to one for cases that are between Fanno (adiabatic) and isothermal flow.4 External flow When the flow over an external body is about .3. was ever produced to verify this finding. The model suggests that the choking occurs at 1/ k and it appears that Shapiro was the first one to realize this difference compared to the other models. flow that is dominantly adiabatic could be simplified and analyzed. Fanno attributes the main pressure reduction to friction. The friction factor is the main component in the analysis as Darcy f 41 had already proposed in 1845. a shock wave occurs which increases the resistance. this flow model should be called Shapiro’s flow. HISTORICAL BACKGROUND 13 Mumenthaler from UTH University. The author invites others to help in this information. No experimental evidence. The choking was assumed only to be in the subsonic flow. If it turned out that no one had done it before Shapiro. which built on Hunter Rouse’s (194x) work made Darcy– Weisbach’s equation universally useful. f based radius is only one quarter of the Darcy f which is based on diameter . But because the actual Fanno’s thesis is not available. However at a Mach number above 0.8 Mach or more the flow must be considered to be a compressible flow. no copy of the thesis can be found in the original University and perhaps only in the personal custody of the Fanno family40 . Additionally an understating of the supersonic branch of the flow was unknown (The idea of shock in tube was not raised at that time. Meta Meta End Isothermal Flow Did Fanno realize that the flow is choked? It appears at least in Stodola’s book that choking was understood in 1927 and even earlier. This fact was evident in industrial applications where the expectation of the choking is at Mach one. In reality. Thus. the Fanno model wasn’t able to produce a prediction useful for the industry. 1. The Navier-Stokes equations which describe the flow (or even 41 Fanning 40 This material is very important and someone should find it and make it available to researchers.

and the KuttaJoukowski circulation theory of lift (1906). INTRODUCTION Euler equations) were considered unsolvable during the mid 18xx because of the high complexity. Prandtl’s student. Prandtl worked on calculating the effect of induced drag on lift. J. Frank Caldwell and Elisha Fales demonstrated in 1918 that at a critical speed (later renamed the critical Mach number) airfoils suffered dramatic increases in drag and decreases in lift. After World War Two. Examples of such work are Hermann von Helmholtz’s concept of vortex filaments (1858). 44 It is like teaching using slide ruler in today school. Theoreticians tried to simplify the equations and arrive at approximate solutions representing specific cases. By the way. Yet. Prandtl created his thin–airfoil theory that enabled the calculation of lift for thin. During World War I. one of Prandtl’s assistants. That is the “dinosaur44 ” reason that even today some instructors are teaching mostly the perturbations and asymptotic methods in Gas Dynamics classes. More information on external flow can be found in . . Practitioners like the Wright brothers relied upon experimentation to figure out what theory could not yet tell them. Meanwhile in Germany. Ackeret. cambered airfoils. Some of the solutions lead to tedious calculations which lead to the creation of Computational Fluid Dynamics (CFD). 43 This undersigned is aware of only one case that these methods were really used to calculations of wing. Prandtl’s boundary layer theory allowed various simplifications of the Navier-Stokes equations. Today these methods of perturbations and asymptotic are hardly used in wing calculations43 . slide rule is sold for about 7. Airplanes could not yet fly fast. especially at the propeller tips. However.” Cambridge University Press. This is because the English Astronomer Frederick Lanchester published the foundation for Prandtl’s theory in his 1907 book Aerodynamics. Prandtl claimed that he was not aware of Lanchester’s model when he had begun his work in 1911. Anderson’s Book “History of Aerodynamics and Its Impact on Flying Machines. but the propellers (which are also airfoils) did exceed the speed of sound.5$ on the net. which was published in 1918-1919 and enabled accurate calculations of induced drag and its effect on lift42 . Later. Von Karman reduced the equations for supersonic flow into a single equation. John D. He later contributed to the PrandtlGlauert rule for subsonic airflow that describes the compressibility effects of air at high speeds. 1997 42 The English call this theory the Lanchester-Prandtl theory.14 CHAPTER 1. simplified the shock equations so that they became easy to use. This claim seems reasonable in the light that Prandtl was not ware of earlier works when he named erroneously the conditions for the shock wave. See for the full story in the shock section. there is no reason to teach it in a regular school. thus exhibiting inefficiency. the research had continued and some technical solutions were found. Lanchester’s concept of circulatory flow (1894). After the First World War aviation became important and in the 1920s a push of research focused on what was called the compressibility problem. This problem led to two consequences. Ludwig Prandtl in 1904 explained the two most important causes of drag by introducing the boundary layer theory. He introduced the lifting line theory. Briggs and Dryden showed that the problem was related to the shock wave.

There are many figures that should be included and a biased selection was required. 1. The simplest model of nozzle. Much information can be obtained from other resources.e. i.3. In this section there is no originality and none should be expected. In 1993 the first reasonable models for forced volume were published by the undersigned. 46 In 45 International . in die casting conference there was a discussion and presentation on which turbulence model is suitable for a complete still liquid. and was republished by Owczarek. Textbook Co. J. As anecdotal story explaining the lack of progress. the model has to be based on scientific principles and not detached from reality. A. 1940. fact.3..3. is not sufficient in many cases and a connection by a tube (rather just nozzle or orifice) is more appropriated. that model was extended by several research groups.6 Biographies of Major Figures In this section a short summary of major figures that influenced the field of gas dynamics is present. such as the Internet. Since World War II considerable works have been carried out in this area but with very little progress46 .35) in 2005. Scranton.. The classification of filling or evacuating the chamber as external control and internal control (mostly by pressure) was described in version 0. the emergence of the CFD gave the illusion that there are solutions at hand. Later several researchers mostly from the University in Illinois extended this work to isothermal nozzle (choked and unchoked). not realizing that garbage in is garbage out..5 Filling and Evacuating Gaseous Chambers It is remarkable that there were so few contributions made in the area of a filling or evacuation gaseous chamber. HISTORICAL BACKGROUND 15 1.3 of this book by this author. 1964. He also extended the model to include the unchoked case. The analytical solution for forced volume and the “balloon” problem (airbag’s problem) model were published first in this book (version 0.”45 . Pennsylvania. The earlier work dealing with this issue was by Giffen.1. Later. Other “strange” models can be found in the undersigned’s book “Fundamentals of Die Casting Design. the model and solution to the nozzle attached to chamber issue in his book “Fundamentals of Gas Dynamics.

5: Portrait of Galileo Galilei Galileo had a relationship with Marina Gamba (they never married) who lived and worked in his house in Padua. During the time after his study. INTRODUCTION Galileo was born in Pisa. However. 1. The oldest of six children. Galileo invented many mechanical devices such as the pump and the telescope (1609). 1564 to musician Vincenzo Galilei and Giulia degli Ammannati. where she bore him three children. Fig. Galileo started his studying at the University of Pisa in 1581. because of his noble ancestry. Vincenzio. . Italy on February 15. however. Galileo died in 1642 in his home outside of Florence. (1602). His telescopes helped him make many astronomic observations which proved the Copernican system.16 Galileo Galilei CHAPTER 1. and he was put under house arrest for the remainder of his life. Galileo also proved that objects fell with the same velocity regardless of their size. Galileo was convicted after publishing his book Dialogue. He then became a professor of mathematics at the University of Padua in 1592. this relationship did not last and Marina married Giovanni Bartoluzzi and Galileo’s son. joined him in Florence (1613). Galileo’s observations got him into trouble with the Catholic Church. Galileo moved with his family in early 1570 to Florence. he made numerous discoveries such as that of the pendulum clock. the church was not harsh with him.

6: Photo of Ernest Mach interested also in physiology of sensory perception. Mach was also a great thinker/philosopher and influenced the theory of relativity dealing with frame of reference.R. Mach’s shadowgraph 47 It is interesting to point out that Prague provided us two of the top influential researchers[:] E. His experiments required exact timing. He graduated from Vienna in 1860.G. Note it was not taken in a wind tunnel sonic speeds. and in 1867 he received a position in the Technical University in Prague47 where he taught experimental physics for the next 28 years. In 1863. He then turned to physics. There Mach wrote his thesis ”On Electrical Discharge and Induction. . Eckert. before he entered the university of Vienna were he studies mathematics. 1. he named it Mach’s Principle. when Czechia was still a part of the Austro–Hungary empire. and in 1918. Einstein was greatly influenced by it. Later.1. At first he received a professorship position at Graz in mathematics (1864) and was then offered a position as a professor of surgery at the university of Salzburg. was a high school teacher who taught Ernst at home until he was 14. It is amazing that Mach was able to photograph the phenomenon using the spinning arm technique (no wind tunnel was available at that time and most definitely nothing that Fig. HISTORICAL BACKGROUND Ernest Mach (1838-1916) 17 Ernst Mach was born in 1838 in Chrlice (now part of Brno). when he studied in Kromeriz Gymnasium. Ernest Mach (1836 . physics and philosophy.” Mach was Fig. Mach’s father.7: The Photo of the bullet in a supersonic flow that could take a photo at superMach made. Mach and E. This was one of the primary sources of inspiration for Einstein’s theory of General Relativity.3.1916) published Die Machanik in which he formalized this argument.7. Johann. Mach’s revolutionary experiment demonstrated the existence of the shock wave as shown in Figure 1. He was not able to attach the camera to the arm and utilize the remote control (not existent at that time). 1. but he declined.

Mach understood the basic characteristics of external supersonic flow where the most important variable affecting the flow is the ratio of the speed of the flow48 (U) relative to the speed of sound (c). Mach’s contributions to supersonic flow were not limited to experimental methods alone. hydrodynamics. The name Mach Number (M) was coined by J. reprinted in his collected works. Mach was the first to note the transition that occurs when the ratio U/c goes from being less than 1 to greater than 1. He was Lord Lieutenant of Essex from 1892 to 1901. Ackeret (Prandtl’s student) in 1932 in honor of Mach. It seems that Rayleigh was the first who realized that flow with chemical reactions (heat transfer) can be choked. Lord Rayleigh was a British physicist born near Maldon. In 1861 he entered Trinity College at Cambridge. color vision. and his other extensive studies are reported in his Scientific Papers. clearly 48 Mach dealt with only air. concerning optics and vibrating systems. INTRODUCTION technique and a related method called Schlieren Photography are still used today. Essex. flow of liquids. At that time there wasn’t the realization that the flow could be choked. He served for six years as the Fig. His Theory of Sound was published in two volumes during 1877-1878. density of gases. and from 1896 to 1919 he acted as Scientific Adviser to Trinity House. started studies in compressible flow mostly from a mathematical approach. electrodynamics. 1. and photography. the year of his marriage. Lord Rayleigh’s first research was mainly mathematical. light scattering. John William Strutt (Lord Rayleigh) A researcher with a wide interest. but it is reasonable to assume that he understood that this ratio was applied to other gases. electromagnetism. Rayleigh was considered to be an excellent instructor. elasticity. 1842. . but his later work ranged over almost the whole field of physics. He published 446 papers which. six volumes issued during 1889-1920. viscosity. His exceptional abilities soon enabled him to overtake his colleagues. wave theory. capillarity. where he commenced reading mathematics. covering sound. Rayleigh was also a contributer to the Encyclopedia Britannica.18 CHAPTER 1. Rayleigh’s later work was concentrated on electric and magnetic problems. He graduated in the Mathematical Tripos in 1865 as Senior Wrangler and Smith’s Prizeman. on November 12.8: Photo of Lord Rayleigh president of the government committee on explosives. In 1866 he obtained a fellowship at Trinity which he held until 1871. Yet.

at Witham. They had three sons. and naval engineering in which he was involved in applying scientific principles to building ships. He was a Fellow of the Royal Society (1873) and served as Secretary from 1885 to 1896. the Earl of Balfour (of the famous Balfour declaration of the Jewish state). 1872) was a Scottish engineer and physicist.December 24. 1. 1919 at Witham. He was a founding contributor to the science of thermodynamics (Rankine Cycle). never allowing politics to interfere with science. he was head of the Cavendish Laboratory at Cambridge from 1879-1884. Rankine was born in Edinburgh to British Fig. and in 1887 became Professor of Natural Philosophy at the Royal Institute of Great Britain. Lord Rayleigh died on June 30. the eldest of whom was to become a professor of physics at the Imperial College of Science and Technology. His steam engine manuals were used for many decades.1. Rankine was well rounded interested beside the energy field he was also interested in civil engineering. 1820 . and as President from 1905 to 1908. He received the Nobel Prize in 1904. In 1871 he married Evelyn. strength of materials. Lord Rayleigh. a Chancellor of Cambridge University. and his only brother and parents died before him. Essex. Rankine. 1919. William John Macquorn Rankine William John Macquorn Rankine (July 2. Rankine never married. HISTORICAL BACKGROUND 19 show his capacity for understanding everything just a little more deeply than anyone else. Essex. sister of the future prime minister. He intervened in debates of the House of Lords only on rare occasions. was a Justice of the Peace and the recipient of honorary science and law degrees.3.9: Portrait of Rankine Army lieutenant David Rankine and Barbara Grahame. Rayleigh died on June 30. As a successor to James Clerk Maxwell. Rankine developed a theory of the steam engine. . London.

Faced with anti–semitism. In July 1904 he received his diploma (master). on February 1939 Fanno was degraded (denounced) and he lost his Ph. Fanno had a cache of old Italian currency (which was apparently still highly acceptable) which helped him and his wife survive the war. Fanno passed way in 1960 without world recognition for his model. Marco. Fig. Fanno converted to Catholicism. was childless. As were many Jews at that time. However. Clearly. 1888. After the war. He likely had a good knowledge of Yiddish and possibly some Hebrew. 50 In 49 Missing . Additionally. Isaak Baruch Weil’s family. Fanno was not as lucky as his brother. Apparently. Fanno was only able to work in agriculture and agricultural engineering. When one of Professor Stodola’s assistants attended military service this temporary position was offered to Fanno. from Regian Istituto Superiore d’Ingegneria di Genova. Later Fanno had to go back to Italy to find a job in industry. “Why didn’t a talented guy like Fanno keep or obtain a position in academia after he published his model?” The answer is tied to the fact that somehow rumors about his roots began to surface.D. even though for short time he went to live in a Jewish home. data about friction factor some places. He studied in a technical institute in Venice and graduated with very high grades as a mechanical engineer. the fact that his model was not a “smashing49 success” did not help. He married. He obtained a Ph. mentioned earlier Marco Fanno is a famous economist who later developed fundamentals of the supply and demand theory.” To further camouflage himself. In this new place he was able to pose as a Roman Catholic. he had to be under house arrest to avoid being sent to the “vacation camps. Fanno turned out to be a good engineer and he later obtained a management position. Consequently. (is this the first case in history) because his of his Jewish nationality50 . the ridicules claims that Jews persecuted only because their religion.20 Gino Girolamo Fanno CHAPTER 1. Switzerland in 1900 to attend graduate school for his master’s degree. Fanno’s older brother.10: The photo of Gino Fanno approximately in 1950 and French. 1. During the War (WWII). Fanno left Italy for Zurich. and like his brother. he did not have a problem studying in a different language. English. Fanno was fluent in several languages including Italian. German. Fanno was not part of the Jewish religion (see his picture) only his nationality was Jewish.D. who was able to get into academia. INTRODUCTION Fanno a Jewish Engineer was born on November 18.

his Ph.D. the paper which describes his boundary layer theory. It is mind boggling to look at the long list of those who were his students and colleagues. 1953. Germany (1901).11: Photo of Prandtl chinery books. Germany and he graduated at the age of 26 with a Ph. This upbringing fostered the young Prandtl’s interest in science and experimentation.3. such as meteorology and structural mechanics. the young Ludwig spent more time with his father which made him interested in his father’s physics and maFig. he was known worldwide as the leader in the science of fluid dynamics. His interest changed when. His father was a professor of engineering and his mother suffered from a lengthy illness. with his student Theodor Meyer. During this time Prandtl developed his boundary layer theory and studied supersonic fluid flows through nozzles. turbulence mixing theories etc. His work and achievements in fluid dynamics resulted in equations that simplified . Prandtl also contributed to research in many areas. HISTORICAL BACKGROUND Ludwig Prandtl 21 Perhaps Prandtl’s greatest achievement was his ability to produce so many great scientists. developed the first theory for calculating the properties of shock and expansion waves in supersonic flow in 1908 (two chapters in this book). By the 1930s.1. 1. Prandtl started his studies at the age of 20 in Munich. Prandtl changed the field of fluid mechanics and is called the modern father of fluid mechanics because of his introduction of boundary layer. His 1904 paper raised Prandtl’s prestige. Ludwig Prandtl was born in Freising. He became the director of the ¨ Institute for Technical Physics at the University of Gottingen. he presented the revolutionary paper “Flussigkeitsbewegung Bei Sehr Kleiner Reibung” (Fluid Flow in Very Little Friction). He developed the Prandtl-Glauert rule for subsonic airflow. Prandtl. Interestingly. In 1904. he was required to design factory equipment that involved problems related to the field of fluid mechanics (a suction device). As a byproduct they produced the theory for oblique shock. in 1874. As a result. Later he sought and found a job as a professor of mechanics at a technical school in Hannover. in his first job. was focused on solid mechanics. ¨ Ludwig Prandtl worked at Gottingen until his death on August 15. There is no one who educated as many great scientists as Prandtl. In 1925 Prandtl became the director ¨ of the Kaiser Wilhelm Institute for Flow Investigation at Gottingen. Bavaria.D.

Eckert Eckert was born in 1904 in Prague. where he studied at the German Institute of Technology. Ludwig Prandtl died in Gottingen. Germany on August 15th 1953. especially in compressible flow. and many are still used today. INTRODUCTION understanding. Prandtl’s other contributions include: the introduction of the Prandtl number in fluid mechanics. Schlichting suggested this dimensionless group in honor of Eckert. airfoils and wing theory (including theories of aerodynamic interference. .12: The photo of Ernst Rudolf George Eckert with the author’s family standing of heat dissipation in relation to kinetic energy. wing-fuselage. high speed flow (correction formula for subsonic compressible flows). His book ”Introduction to the Transfer of Heat and Mass.R. theory of turbulence. biplane.G.22 CHAPTER 1. the dimensionless group has been designated as the Eckert number. In addition to being named to the National Academy of Engineering in 1970. Hence. Air Force and the National Advisory Committee for Aeronautics before coming to Minnesota. During World War II. Therefore many referred to him as ¨ the father of modern fluid mechanics. he developed methods for jet engine turbine blade cooling at a research laboratory in Prague. He authored more than 500 articles and received several medals for his contributions to science.Fig. fundamental studies in the wind tunnel. which is associated with the Mach number.” published in 1937.S. etc). He emigrated to the United States after the war. His name is linked to the following: • Prandtl number (heat transfer problems) • Prandtl-Glauert compressibility correction • Prandtl’s boundary layer equation • Prandtl’s lifting line theory • Prandtl’s law of friction for smooth pipes • Prandtl-Meyer expansion fans (supersonic flow) • Prandtl’s Mixing Length Concept (theory of turbulence) E. wing-propeller. Eckert developed the under. 1. is still considered a fundamental text in the field. and served as a consultant to the U.

T. asthma. in 1938 and the Sc. MIT . HISTORICAL BACKGROUND 23 Eckert was an excellent mentor to many researchers (including this author).3. While Shapiro viewed gas dynamics mostly through aeronautic eyes. In his first 25 years Shapiro focused primarily on power production. Ascher Shapiro MIT Professor Ascher Shapiro51 . (It is M.D. turbomachinery and propulsion by jet engines and rockets.’s equivalent of a Ph. and he had a reputation for being warm and kindly. degree) in 1946 in mechanical engineering from MIT. emphysema and glaucoma. 51 Parts taken from Sasha Brown.D. Shapiro spent most of his active years at MIT. Shapiro grew up in New York City and received his S. the Eckert equivalent for the compressible flow.B. He was also a leading Figure in bringing together engineering in the East and West during the Cold War years. highspeed flight. Shapiro moved to the field of biomedical engineering where he was able to pioneer new work. was instrumental in using his two volume book “The Dynamics of Thermodynamics of the Compressible Fluid Flow. three years before receiving his Sc.D. Furthermore. In 1965 he become the head of the Department of Mechanical Engineering until 1974.1. Therefore it is proposed to call this model Shapiro’s Flow.” to transform the gas dynamics field to a coherent text material for engineers. Ascher Shapiro passed way in November 2004. Shapiro’s knowledge of fluid mechanics enabled him to “sew” the missing parts of the Fanno line with Moody’s diagram to create the most useful model in compressible flow. Unfortunately for the field of Gas Dynamics. He was assistant professor in 1943. The undersigned believes that Shapiro was the first one to propose an isothermal flow model that is not part of the aeronautic field.I. Shapiro was instrumental in the treatment of blood clots.

INTRODUCTION .24 CHAPTER 1.

2. 25 .CHAPTER 2 Fundamentals of Basic Fluid Mechanics 2. At the initial time the mass that was in the tube was m0 .3 2. it is assumed the control volume is a fixed boundary.4 Fluid Properties Control Volume Reynold’s Transport Theorem For simplification the discussion will be focused on one dimensional control volume and it will be generalzed later. Several application of the fluid mechanics will demonstrated. In addition the basic Bernoulli’s equation will be derived for incompressible flow and later for compressible flow.2 2. The basic principles are related to the basic conservation principle. Several terms will be reviewed such as stream lines. For simplicity. The mass after a very short time of dt is dm.1 Introduction This chapter is a review of the fundamentals that the student is expected to know. The flow through a stream tube is assumed to be one-dimensional so that there isn’t any flow except at the tube opening. The flow on the right through the opening and on the left is assumed to enter the stream tube while the flow is assumed to leave the stream tube. This material is not covered in the history chapter.

1) ∆t→0 . FUNDAMENTALS OF BASIC FLUID MECHANICS Supposed that the fluid has a property η dNs dt = lim Ns (t0 + ∆t) − Ns (t0 ) ∆t (2.26 CHAPTER 2.

With the “new” knowledge from the consultation the calculations were within the range of acceptable results. In this chapter an introductory discussion is given about different situations which can appear the industry in regards to speed of sound. The above situation is not unique in the industry. the engineer was not ware of the effect of particles on the speed of sound. The mouse ask the cat I thought I hear a dog. The author thinks that this approach has many shortcomings. Thus. Suddenly.2 Introduction engineer that alumni of University of Minnesota. In a recent consultation an engineer1 design a industrial system that contains converting diverging nozzle with filter to remove small particles from air. 1 Aerospace 27 . and cat is catching the mouse. yes you right. Hence. My teacher was right. The cat reply. It should be expected that engineers know how to manage this situation of non pure substances (like clean air). the mouse hear a barking dog and a cat yelling. Yet.CHAPTER 3 Speed of Sound 3. As it will shown in this chapter. in some situations. The fact that the engineer knows about the chocking is great but it is not enough for today’s sophisticated industry2 . the engineer was able to predict that was a chocking point. reduces the speed of sound by almost as half. one language is not enough today. A cat is pursuing a mouse and the mouse escape and hide in the hole. The mouse go out to investigate.1 Motivation In traditional compressible flow classes there is very little discussion about the speed of sound outside the ideal gas. but a joke is must in this situation. 3. Aerospace Department. The engineer was well aware of the calculation of the nozzle. 2 Pardon. the particles can. the actual flow rate was only half of his prediction.

it raises the question: what is the speed of the Fig. assuming isentropic flow and neglecting the gravity results (c − dU )2 − c2 dP + =0 2 ρ neglecting second term (dU 2 ) yield −cdU + dP =0 ρ (3. In the control volume it is convenient to look at a control volControl volume around ume which is attached to a pressure the sound wave c-dU c pulse. Thus. For an ideal gas.2) From the energy equation (Bernoulli’s equation).1). This velocity is referred to as the speed of sound. it is considered .2: Stationary sound wave and gas moves relative to the pulse or when the higher term dU dρ is neglected yields ρdU = cdρ =⇒ dU = cdρ ρ (3. Here. To answer this question consider a piston moving from the left to the right at a relatively small velocity (see Figure 3. 3.1: A very slow moving piston in a still gas small disturbance travel in a “quiet” medium. 3. SPEED OF SOUND The people had recognized for several hundred years that sound is sound wave dU velocity=dU a variation of pressure. Applying the mass balance P+dP yields P ρ+dρ ρ ρc = (ρ + dρ)(c − dU ) (3. the pulse will be infinitesimally small.28 CHAPTER 3. The information that the piston is moving passes thorough a single “pressure pulse.5) An expression is needed to represent the right hand side of equation (3. Thus.3) Substituting the expression for dU from equation (3. the pressure and density can be assumed to be continuous.4) yields c2 dρ ρ = dP dP =⇒ c2 = ρ dρ (3. The ears c sense the variations by frequency P+dP P ρ ρ+dρ and magnitude which are transferred to the brain which translates to voice. P is a function of two independent variables.2) into equation (3.5).4) (3.1) Fig.” It is assumed that if the velocity of the piston is infinitesimally small.

k namely P = constant × ρk (3. and a function “molecular structure” or ratio of specific heats.11) . Example 3. S OLUTION The momentum equation written for the control volume shown in Figure (3.3.5).10) 3.5) can be derived from the momentum equation. s) where s is the entropy.3.5) can be obtained by utilizing the momentum equation instead of the energy equation. therefore it can be written ∂P dP = dρ ∂ρ (3. (3.6) In the derivations for the speed of sound it was assumed that the flow is isentropic. SPEED OF SOUND IN IDEAL AND PERFECT GASES 29 that P = P (ρ. The full differential of the pressure can be expressed as follows: dP = ∂P ∂ρ dρ + s ∂P ∂s ds ρ (3. ρ.1: Demonstrate that equation (3.9) dP = c2 dρ This yields the same equation as (3.2) is P F R cs U (ρU dA) (P + dP ) − P = (ρ + dρ)(c − dU )2 − ρc2 Neglecting all the relative small terms results in : ∼ 0 ∼0   : dU 2 dP = (ρ + dρ) c2 −  2cdU +  − ρc2 (3. The pressure for an ideal gas can be expressed as a simple function of density.7) s Note that the equation (3.3 Speed of sound in ideal and perfect gases The speed of sound can be obtained easily for the equation of state for an ideal gas (also perfect gas as a sub set) because of a simple mathematical expression.8) (3.

0100 At 18[bar] and 300◦ C: s = 6.94199 and substituting into the equation yields c= m 200000 = 780. 3 This data is taken form Van Wylen and Sontag “Fundamentals of Classical Thermodynamics” 2nd edition . and equation (3.30 and hence CHAPTER 3.2: Calculate the speed of sound in water vapor at 20[bar] and 350◦ C.13) c = kRT Example 3.8226 ∆P ∆ρ kJ K kg kJ K kg kJ K kg (3. table A 8.7◦C: s ∼ 6. and it will be part of the future program (potto–GDC). SPEED OF SOUND P c= dP constant × ρk = k × constant × ρk−1 = k × dρ ρ =k× P ρ (3.5 sec Note that a better approximation can be done with a steam table.13216 kg m3 kg m3 kg m3 After interpretation of the temperature: kJ At 18[bar] and 335.15) for ideal gas assumption (data taken from Van Wylen and Sontag. (a) utilizes the steam table (b) assuming ideal gas. Classical Thermodynamics.32823 sec kg m3 (3. S OLUTION The solution can be estimated by using the data from steam table3 c= At 20[bar] and 350◦ C: s = 6.46956 ρ = 7.327 × 461 × (350 + 273) ∼ 771.) √ m c = kRT ∼ 1.12) Remember that P/ρ is defined for an ideal gas as RT .61376 ρ = 6.9563 K kg ρ ∼ 6.14) s=constant ρ = 6.5 0.12) can be written as √ (3.9563 At 18[bar] and 350◦ C: s = 7.

What is the time it take for sound to travel from point “A” to point “B” under this assumption.18) (3.20) .19) The speed of sound of any gas is provided by equation (3.17) This correction factor approaches one when TB −→ TA . The distance between “A” and “B” is denoted as h.4 Speed of Sound in Real Gas The ideal gas model can be improved by introducing the compressibility factor.3: The temperature in the atmosphere can be assumed to be a linear function of the height for some distances. x T = (TB − TA ) + TA h Where the distance x is the variable distance. To obtain the expression for a gas that obeys the law expressed by (3.19) some mathematical expressions are needed. Recalling from thermodynamics. The compressibility factor represents the deviation from the ideal gas. For an infinitesimal time dt is equal to dt = dx = kRT (x) dx kRTA (TB −TA )x TA h +1 integration of the above equation yields 2hTA t= √ 3 kRTA (TB − TA ) h ¯ kRT 3 2 TB TA 3 2 −1 (3.4. Thus. the Gibbs function (3. 3. It should be noted that velocity is provided as a function of the distance and not the time (another reverse problem).3.? S OLUTION The temperature is denoted at “A” as TA and temperature in “B” is TB . a real gas equation can be expressed in many cases as P = zρRT (3. SPEED OF SOUND IN REAL GAS 31 Example 3.7).16) For assumption of constant temperature the time is t= √ Hence the correction factor tcorrected = t TA TA 2 ¯ 3 (TB − TA ) T TB TA −1 (3.

SPEED OF SOUND Fig. 3. .3: The Compressibility Chart is used to obtain T ds = dh − dP ρ (3.22) ρ From thermodynamics.23) P Van Wylen p. perhaps to insert the discussion here.32 CHAPTER 3.20) The definition of pressure specific heat for a pure substance is Cp = ∂h ∂T =T P ∂s ∂T (3.21) P The definition of volumetric specific heat for a pure substance is Cv = ∂u ∂T =T ρ ∂s ∂T (3. 372 SI version. it can be shown 4 dh = Cp dT + v − T 4 See ∂v ∂T (3.

26) P Utilizing Gibbs equation (3.23) results  v  v  z T Simplifying equation (3. However.24) into equation (3.28) and (3.4.30) yields dρ dP Cv = ρ P Cp z+T z+T ∂z ∂T P ∂z ∂T ρ (3.25) P dP = Cp dT − T z (3.29) results in dρ R z+T ρ Cv ∂z ∂T = ρ dP R z+T P Cp ∂z ∂T (3.3.28) P Equation (3.30) P Rearranging equation (3.29) ρ Equating the right hand side of equations (3.27) Letting ds = 0 for isentropic process results in dP R dT = z+T T P Cp ∂z ∂T (3. SPEED OF SOUND IN REAL GAS The specific volumetric is the inverse of the density as v = zRT /P and thus ∂v ∂T ∂ zRT P 33 = P ∂T P RT = P ∂z ∂T P > zR ∂T +  P ∂T P  1 (3.24) Substituting the equation (3.20) dh T ds = Cp dT − T z ∂z ∂T zRT P dP dP dP − = Cp dT − ρ ρ ρ ∂z ∂T T z ∂z ∂T +1 P =Cp dT − dP P P ρ T z +1 P (3. it is more convenient to express dT /T in terms of Cv and dρ/ρ as follows dρ R dT z+T = T ρ Cv ∂z ∂T (3.28) can be integrated by parts.31) .25) to became dh = Cp dT − Tv z ∂z ∂T   dh = Cp dT + v − T   RT    P ∂z ∂T + P  zR    dP P  ∂z ∂T dP ρ (3.

The integration of equation (3.34 CHAPTER 3. the speed of sound for a real gas can be obtained in the same manner as for an ideal gas.1[m/sec] For the real gas first coefficient n = 1.32) provides a tool to estimate n.33) isn’t new.3 1. Assume that R = 287[j/kg/K ]. What is different in these derivations is that a relationship between coefficient n and k was established. z = 0.31) can be integrated. .4: Calculate the speed of sound of air at 30◦ C and atmospheric pressure ∼ 1[bar]. at relative moderate pressure but low temperature common in atmosphere. n is defined as k n= Cp Cv z+T z+T ∂z ∂T ρ ∂z ∂T P (3. Make the calculation based on the ideal gas model and compare these calculations to real gas model (compressibility factor). the compressibility factor.407 × 287 × 300 ∼ 348. S OLUTION According to the ideal gas model the speed of sound should be √ √ c = kRT = 1.995. a change in temperature can have a dramatical change in the speed of sound.7[m/sec] The correction factor for air under normal conditions (atmospheric conditions or even increased pressure) is minimal on the speed of sound. dP = nzRT dρ (3.33) Equation (3. and in–fact any thermodynamics book shows this relationship.403 has √ √ c = znRT = 1. For short hand writing convenience. SPEED OF SOUND If the terms in the braces are constant in the range under interest in this study. n = 1. equation (3.3 and n ∼ 1 which means that speed of sound is only 0. But the definition of n in equation (3.403. However.995times287 × 300 = 346.33) is similar to equation (3.32) Note that n approaches k when z → 1 and when z is constant. Now. and z = 0.34) Example 3.403 × 0.31) yields ρ1 ρ2 n = P1 P2 (3. The specific heat for air is k = 1. This relationship (3.11).4 factor (0.5) to calculated by ideal gas model. For example.407.

5). 1967-1968: E37 Value [m/sec] 1492 1496 1494 Table 3. The amount of compression of almost all liquids is seen to be very small as given in Table (3. SPEED OF SOUND IN ALMOST INCOMPRESSIBLE LIQUID 35 3.2 × 109 N/m2 . as can be observed from the equation (3.5 Speed of Sound in Almost Incompressible Liquid Even liquid normally is assumed to be incompressible in reality has a small and important compressible aspect. According . Physics. 1999. Chicago: World Book.3.5. 1997: 468. Johnson. 601 Handbook of Chemistry and Physics.2 × 109 N/m2 = 1493m/s 1000kg/m3 elastic property = inertial property B ρ (3.1: Water speed of sound from different sources The effect of impurity and temperature is relatively large. the average bulk modulus for water is 2.000 meters. 1482 m/s at 20◦ C.37). John D.36) dP dρ (3.5) Remark Fresh Water (20 ◦ C) Distilled Water at (25 ◦ C) Water distilled reference Cutnell.8% even under this pressure nevertheless it is a change. New York: Wiley. The ratio of the change in the fractional volume to pressure or compression is referred to as the bulk modulus of the material. and for purposes of comparison it is given in Table (3. The compressibility of the substance is the reciprocal of the bulk modulus. At a depth of about 4. Ohio: Chemical Rubber Co. The fractional volume change is only about 1. & Kenneth W.. with an increase of 34 degrees from 0◦ C there is an increase in the velocity from about 1430 m/sec to about 1546 [m/sec]. Many researchers have looked at this velocity. The mathematical definition of bulk modulus as following B=ρ In physical terms can be written as c= For example for water c= 2.35) This agrees well with the measured speed of sound in water. For example. The World Book Encyclopedia. the pressure is about 4 × 107 N/m2 . For example.

P ) = c0 + cT + cS + cP + cST P . Acoust.6 Speed of Sound in Solids The situation with solids is considerably more complicated. material Glycerol Sea water Mercury Kerosene Methyl alcohol Carbon tetrachloride reference 25 ◦ C Value [m/sec] 1904 1533 1450 1324 1143 926 Table 3. SPEED OF SOUND to Wilson5 . and cST P is a correction factor between coupling of the different parameters. using a general tabulated value for the bulk modulus. Wilson’s empirical formula appears as follows: c(S. 1960. and hydrostatic pressure.38) Compared to one tabulated value the example values for stainless steel lays between the speed for longitudinal and transverse waves.32. the speed of sound in solids is larger than in liquids and definitely larger than in gases. T. vol. Amer.. and cS is a function salinity. Wilson’s formula is accepted by the National Oceanographic Data Center (NODC) USA for computer processing of hydrological information. gives a sound speed for structural steel of E = ρ 160 × 109 N/m2 = 4512m/s 7860Kg/m3 c= (3. the speed of sound in liquids is about 3 to 5 relative to the speed of sound in gases. Nevertheless. John. 5 J.2: Liquids speed of sound. (3. 1972 In summary. the speed of sound in sea water depends on temperature. Young’s Modulus for a representative value for the bulk modulus for steel is 160 109 N /m2 . after Aldred. London: Fountain Press. cT is a function temperature.36 CHAPTER 3. and differences between transverse and longitudinal waves. Manual of Sound Recording. salinity. . cP is a function pressure. Soc. Speed of sound in solid of steel. 3. p. 1357.14 is about clean/pure water. N 10. with different speeds in different directions. in different kinds of geometries.37) where c0 = 1449.

it assumed that two materials are homogeneously mixed. there could be more than one speed of sound for two phase flow. In actuality.7.3: Solids speed of sound. There can be several models that approached this problem. In that case. about equal parts of gas and the liquid phase. The first case is analyzed. Topic for none homogeneous mixing are beyond the scope of this book. 1972 3. However. SOUND SPEED IN TWO PHASE MEDIUM material Diamond Pyrex glass Steel Steel Steel Iron Aluminum Brass Copper Gold Lucite Lead Rubber reference longitudinal wave transverse shear longitudinal wave (extensional wave) Value [m/sec] 12000 5640 5790 3100 5000 5130 5100 4700 3560 3240 2680 1322 1600 37 Table 3.40) (3. Manual of Sound Recording.7 Sound Speed in Two Phase Medium The gas flow in many industrial situations contains other particles. for homogeneous and under certain condition a single velocity can be considered.3. For simplicity. Indeed there is double chocking phenomenon in two phase flow.39) (3. It further assumed that no heat and mass transfer occurs between the particles. and liquid with some bubbles.40) can be approximated as ρ =1+m ρa (3.41) . The equation of state for the gas can be written as Pa = ρa RTa The average density can be expressed as ξ 1−ξ 1 = + ρm ρa ρb ˙b where ξ = m m ˙ is the mass ratio of the materials. John. London:Fountain Press. three extreme cases suggest themselves: the flow is mostly gas with drops of the other phase (liquid or solid). after Aldred. For small value of ξ equation (3.

the velocity of sound in slightly wed steam can be about one third of the pure steam speed of sound.41) R P = T ρ 1+m (3.48) At this stage the other models for two phase are left for next version (0. the mixture isentropic relationship can be expressed as P where γ−1 R = γ Cp + mC Recalling that R = Cp − Cv reduces equation (3.45) In a way the definition of γ was so chosen that effective specific pressure heat C +mC +mC and effective specific volumetric heat are p and Cv respectively. Since the equations are the same as before hence the familiar equation for speed of sound can be applied as c= Insert example with small steel particles with air up to 20% γRmix T (3.44) Therefore.38 m ˙b m ˙a CHAPTER 3. . It must noticed that m = constant. Hence. It can be noticed that Rmix and γ are smaller than similar variables in a pure gas. For example. The gas density can be replaced by equation (3. the velocity of mixtures with large gas component is smaller of the pure gas. Generally. SPEED OF SOUND where m = is mass flow rate per gas flow rate. this analysis results in lower speed of sound compared to pure gas.43) Assuming that partial pressure of the particles is constant and applying the second law for the mixture yields droplets gas 0 = mC dP (Cp + mC )dT dP dT dT + Cp −R = −R T T P T P γ −1 γ (3.6). The 1+m 1+m correction factors for the specific heat is not linear.39) and substituted into equation (3.47) (3.46) T = constant (3.46) into γ= Cp + mC Cv + mC (3. If the droplets (or the solid particles) can be assumed to have the same velocity as the gas with no heat transfer or fiction between the particles isentropic relation can be assumed as P = constant ρa k (3.42) A approximation of addition droplets of liquid or dust (solid) results in reduction of R and yet approximate equation similar to ideal gas was obtained.

speed of sound can be expressed as c2 = where X is defined as X= s − sf (PB ) sf g (PB ) (3.49) Meta End .50) ∂P ∂P [f (X )] = ∂ρ ∂ρ (3.3. SOUND SPEED IN TWO PHASE MEDIUM Meta 39 For a mixture of two phases.7.

40

CHAPTER 3. SPEED OF SOUND

CHAPTER 4 Isentropic Flow
In this chapter a discussion on a steady state flow through a smooth and continuous area flow rate is presented. A discussion about the flow through a converging–diverging nozzle is also part of this chapter. The isentropic flow models are important because of two main reasons: One, it provides the information about the trends and important parameters. Two, the correction factors can be introduced later to account for deviations from the ideal state.

PB = P 0 P P0
Subsonic M <1

Supersonic
M >1

4.1 Stagnation State for Ideal Gas Model
4.1.1 General Relationship

distance, x

Fig. 4.1: Flow of a compressible substance (gas) through a converging– diverging nozzle.

It is assumed that the flow is one– dimensional. Figure (4.1) describes a gas flow through a converging–diverging nozzle.t has been found that a theoretical state known as the stagnation state is very useful in simplifying the solution and treatment of the flow. he stagnation state is a theoretical state in which the flow is brought into a complete motionless condition in isentropic process without other forces (e.g. gravity force). Several properties that can be represented by this theoretical process which include temperature, pressure, and density et cetera and denoted by the subscript “0.” 41

42

CHAPTER 4. ISENTROPIC FLOW

First, the stagnation temperature is calculated. The energy conservation can be written as h+ U2 = h0 2 (4.1)

Perfect gas is an ideal gas with a constant heat capacity, Cp . For perfect gas equation (4.1) is simplified into Cp T + U2 = C p T0 2 (4.2)

Again it is common to denote T0 as the stagnation temperature. Recalling from thermodynamic the relationship for perfect gas R = C p − Cv kR k−1 (4.3)

and denoting k ≡ Cp ÷ Cv then the thermodynamics relationship obtains the form Cp = (4.4)

and where R is a specific constant. Dividing equation (4.2) by (Cp T ) yields 1+ U2 T0 = 2Cp T T (4.5)

Now, substituting c2 = kRT or T = c2 /kR equation (4.5) changes into 1+ T0 kRU 2 = 2 2Cp c T (4.6)

By utilizing the definition of k by equation (4.4) and inserting it into equation (4.6) yields 1+ k − 1 U2 T0 = 2 c2 T (4.7)

It very useful to convert equation (4.6) into a dimensionless form and denote Mach number as the ratio of velocity to speed of sound as M≡ U c (4.8)

nserting the definition of Mach number (4.8) into equation (4.7) reads T0 k−1 2 =1+ M T 2 (4.9)

4.1. STAGNATION STATE FOR IDEAL GAS MODEL
 

43

B A The usefulness of Mach number and equation (4.9) can be demonT0 T0 P0 P0 strated by this following simple example. velocity ρ0 ρ0 In this example a gas flows through a tube (see Figure 4.2) of any shape can be expressed as a function of only the Fig. 4.2: Perfect gas flows through a tube stagnation temperature as opposed to the function of the temperatures and velocities. The definition of the stagnation state provides the advantage of compact writing. For example, writing the energy equation for the tube shown in Figure (4.2) can be reduced to

˙ = Cp (T0 B − T0 A )m Q ˙

(4.10)

The ratio of stagnation pressure to the static pressure can be expressed as the function of the temperature ratio because of the isentropic relationship as P0 = P T0 T
k k−1

=

1+

k−1 2 M 2

k k−1

(4.11)

In the same manner the relationship for the density ratio is ρ0 = ρ T0 T
1 k−1

=

1+

k−1 2 M 2

1 k−1

(4.12)

A new useful definition is introduced for the case when M = 1 and denoted by superscript “∗.” The special case of ratio of the star values to stagnation values are dependent only on the heat ratio as the following: 2 c∗ 2 T∗ = 2 = T0 c0 k+1 (4.13)

P∗ = P0

2 k+1

k k−1

(4.14)

ρ∗ = ρ0

2 k+1

1 k−1

(4.15)

44

CHAPTER 4. ISENTROPIC FLOW

Static Properties As A Function of Mach Number
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 1 2 3 Mon Jun 5 17:39:34 2006 0 4 5 Mach number 6 7 8 9 P/P0 ρ/ρ0 T/T0

Fig. 4.3: The stagnation properties as a function of the Mach number, k = 1.4

4.1.2

Relationships for Small Mach Number

Even with today’s computers a simplified method can reduce the tedious work involved in computational work. In particular, the trends can be examined with analytical methods. It further will be used in the book to examine trends in derived models. It can be noticed that the Mach number involved in the above equations is in a square power. Hence, if an acceptable error is of about %1 then M < 0.1 provides the desired range. Further, if a higher power is used, much smaller error results. First it can be noticed that the ratio of temperature to stagnation temperaT is provided in power series. Expanding of the equations according to the ture, T 0 binomial expansion of (1 + x)n = 1 + nx + n(n − 1)x2 n(n − 1)(n − 2)x3 + +··· 2! 3! (4.16)

will resutls in the same fashion P0 (k − 1)M 2 kM 4 2(2 − k )M 6 =1+ + + ··· P 4 8 48 (4.17)

4.2. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 45

(k − 1)M 2 kM 4 2(2 − k )M 6 ρ0 =1+ + + ··· ρ 4 8 48

(4.18)

The pressure difference normalized by the velocity (kinetic energy) as correction factor is
compressibility correction

M2 P0 − P (2 − k )M 4 = 1 + + +··· 1 2 4 24 2 ρU

(4.19)

From the above equation, it can be observed that the correction factor approaches zero when M −→ 0 and then equation (4.19) approaches the standard equation for incompressible flow. The definition of the star Mach is ratio of the velocity and star speed of soundat M = 1. U = c∗ k+1 M 2 k−1 2 M +··· 4 (4.20)

M∗ =

1−

kM 2 P0 − P = P 2 ρ0 − ρ M2 = ρ 2 The normalized mass rate becomes m ˙ = A kP0 2 M 2 RT0

1+

M2 +··· 4

(4.21)

1−

kM 2 +··· 4

(4.22)

1+

k−1 2 M +··· 4

(4.23)

The ratio of the area to star area is A = A∗ 2 k+1
k+1 2(k−1)

1 k+1 (3 − k )(k + 1) 3 + M+ M +··· M 4 32

(4.24)

4.2

Isentropic Converging-Diverging Flow in Cross Section

1 The Properties in the Adiabatic Nozzle (4. knowing the Mach number or the temperature will provide all that is needed to find the other properties. and dividing the results by the equation of state (ρRT ) yields dP dρ dT = + P ρ T (4. Therefore. the stagnation density is constant through the flow. and dividing by the continuity equation reads dρ dA dU + + =0 ρ A U (4. The control volume is shown in Figure (4. ρAU = m ˙ = constant. perature.29) . Second is isentropic and isothermal Fig.27) yields dP + U dU = 0 ρ (4. There are two models that assume variable area flow: First is isentropic and adiabatic model. 4. The only properties that need to be connected are the cross section area and the Mach number.4). Clearly. ISENTROPIC FLOW The important sub case in this chapter is the flow in a converging–diverging nozzle. the stagnation pressure is also constant through the flow because the flow isentropic.11) are the same.26) The thermodynamic relationship between the properties can be expressed as T ds = dh − dP ρ (4.2.46 CHAPTER 4. T ρ P U T+dT ρ+dρ P+dP U+dU 4. reads dh + U dU = 0 Differentiation of continuity equation. In the same argument. the stagnation temdiverging nozzle. equation (4. Conversely. Thus.25) hen there is no external work and heat transfer. the energy equation.28) Differentiation of the equation state (perfect gas). Examination of the relation between properties can then be carried out. If the right hand side is constant for one variable.4: Control volume inside a convergingmodel. T0 .27) For isentropic process ds ≡ 0 and combining equations (4. is constant through the adiabatic flow because there isn’t heat transfer. in mathematical terms.9) and equation (4. it is constant for the other. P = ρRT .25) with (4.

it can be observed that dP can be positive or negative depending on the dA and Mach number.33) to obtain a variables separation form of dP = dA ρU 2 A 1 − M2 (4.31) to obtain dP ρ Or in a dimensionless form dP dA 1 − M2 = U2 ρ A (4. Now.30) so that the density.33) 1− U c 2    dρ dP   =U  +  A dP ρ   2  dA  (4.34).4. dP/ρ yields  1 c2 dP = U2 ρ dA dρ dP + A ρ dP Recalling that dP/dρ = c2 and substitute the speed of sound into equation (4.33) is a differential equation for the pressure as a function of the cross section area. The term ρU 2 /A is always positive (because all the three terms can be only positive).31) = U2 dA A (4.32) Equation (4.30) Rearranging equation (4. dρ.34) The pressure Mach number relationship Before going further in the mathematical derivation it is worth looking at the physical meaning of equation (4. It can be observed that the critical Mach number is one. It is convenient to rearrange equation (4. The meaning of the sign change for the pressure differential is that the pressure can increase or decrease.26) and using it in equation (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 47 Obtaining an expression for dU/U from the mass balance equation (4.28) reads dU U dP dA dρ − U2 + =0 ρ A ρ (4. If the Mach number is larger than one than dP has opposite sign of dA. If Mach number is smaller than . can be replaced by the static pressure.2.

From the speed of sound. However. Note that the pressure decrease is larger in compressible flow compared to incompressible flow.It must also be noted that when M = 1 occurs only when dA = 0.4.36) It can be noted that in the derivations of the above equations (4. Hence the pressure increases when the velocity decreases and vice versa. dρ = 1 dP c2 (4. dA > 0 ⇒ dP < 0 dA < 0 ⇒ dP > 0 This behavior is opposite to incompressible flow behavior. For M > 1 the term 1/1 − M 2 is negative and change the character of the equation.48 CHAPTER 4.35) From equation (4. the equations are applicable for any gas (perfect or imperfect gas). if the area decreases (as a function of x) the pressure decreases. In that case. one can observe that the density. ρ.28) by solving it for dU .35 . Thus. An increase in area results in an increase of the static pressure (converting the dynamic pressure to a static pressure). Since physically dP can increase only in a finite amount it must that dA = 0. the opposite.36).35) it is obvious that dU has an opposite sign to dP (since the term P U is positive). it is possible that dM = 0 thus the diverging side is in the subsonic branch and the flow isn’t choked. For the supersonic branch M > 1. Conversely. For the special case of M = 1 (sonic flow) the value of the term 1 − M 2 = 0 thus mathematically dP → ∞ or dA = 0. For the subsonic branch M < 1 the term 1/(1 − M 2 ) is positive hence dA > 0 =⇒ dP > 0 dA < 0 =⇒ dP < 0 From these observations the trends are similar to those in incompressible fluid. dU = − dP PU (4. the equation of state was not used. The second law (isentropic relationship) dictates that ds = 0 and from thermodynamics dP dT −R ds = 0 = Cp T P . the phenomenon is different. ISENTROPIC FLOW one dP and dA have the same sign. increases with pressure and vice versa (see equation 4. not necessarily means that when dA = 0 that M = 1.36). The relationship between the velocity and the pressure can be observed from equation (4.

The relationship between temperature and the cross section area can be obtained by utilizing the relationship between the pressure and temperature (4. The relationship between the Mach number and the temperature can be obtained by utilizing the fact that the process is assumed to be adiabatic dT0 = 0.42) M2 − 1 dA = dM 1 2 A M 1 + k− 2 M 1 The (4.2.4. equation (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 49 and for perfect gas dT k − 1 dP = (4.39) is combined with equation (4.26)1 .33) yields 1 k ρU 2 dA A 1−M 2 (4. the relationship between the temperature and the stagnation temperature becomes dT0 = 0 = dT 1+ k−1 2 M 2 + T (k − 1)M dM (4.39) (4.33).41) The following identify.41) into 2 P P (4. second law (4.38) yields dT (k − 1)M dM =− 1 2 T 1 + k− 2 M Relationship Between the Mach Number and Cross Section Area The equations used in the solution are energy (4.9).42) changes equation (4.29).33) isn’t the solution but demonstration of certain properties on the pressure.39).43) momentum equation is not used normally in isentropic process.38) and simplifying equation (4. state (4.40) P =− M dM 1 2 1 + k− 2 M (4. First stage equation (4. mass (4. Note.37) T k P Thus.37). Differentiation of equation (4. ρU 2 = kM P can be proved as M2 U2 U2 kM P = k 2 ρRT = k ρRT = ρU 2 c kRT Using the identity in equation (4.40) with equation (4. the temperature varies according to the same way that pressure does. why? .37) and the relationship of pressure and cross section area (4.37) and becomes (k − 1)M dM (k − 1) dP =− 1 2 k P 1 + k− 2 M Combining equation (4.

ISENTROPIC FLOW Equation (4. There¡£¢ fore. A (area) with the relative velocity (Mach s number).50 CHAPTER 4. ction A M and A are positive reM 1 + k− 2 sec gardless of the values of M or A. the factors ros 1 2 . the only factor that affects relation¡¥¤§¦©¨ .43) is very important because it relates the geometry M.43). In equation (4.

indexsonic trasition This minimum area is referred to as “throat.4. The special case is when M = 1 which Fig. This condition section and the Mach number on the 2 imposes that internal flow has to pass subsonic branch a converting–diverging device to obtain supersonic velocity. . 4. In subsonic flow branch. In the case of M > 1 the Mach number increases with x the cross section area and vice versa. In external flow.5)). It was measured that air mass flow rate is 1[kg/sec].” Again. Assume that process is isentropic and neglect the velocity at the reservoir.1: Air is allowed to flow from a reservoir with temperature of 21◦ C and with pressure of 5[MPa] through a tube. the opposite conclusion that when dA = 0 implies that M = 1 is not correct because possibility of dM = 0. from the mathematical point of view: on one hand. a decrease of the cross section increases the velocity and the Mach number. and the cross section area at that point where the static pressure was measured. With the known Mach number. velocity. 4. For the point where the static pressure is known. S OLUTION The stagnation conditions at the reservoir will be maintained throughout the tube because the process is isentropic. calculate the Mach number. For M < 1 the Mach number is varied opposite to M. the Mach number can be calculated by utilizing the pressure ratio. on the other hand. an object can be moved in arbitrary speed. At some point on the tube static pressure was measured to be 3[MPa]. ship between the cross area and the  © Mach number is M 2 − 1.2. Hence the stagnation temperature can be written T0 = constant and P0 = constant and both of them are known (the condition at the reservoir). 2 This condition does not impose any restrictions for external flow. an increase of the cross section decreases the velocity and Mach number (see Figure (4. Assume that the ratio of specific heat is k = Cp /Cv = 1.5: The relationship between the cross requires that dA = 0.2 Isentropic Flow Examples Example 4. Much nubmer the cross section area.

88639 0. 4 This . pressure is about two atmospheres with temperature of 250[K ] this question is for academic purposes.1416 The velocity at that point is c U =M √ The tube area can be obtained from the mass conservation as A= m ˙ = 8.88638317 × √ 1. there is no known way for the author to directly measure the Mach number.86420 0.3) or utilizing the enclosed program from PottoGDC.26 × 10−5 [m3 ] ρU kRT = 0.5[Bar]. Also. In the point where the static pressure known ¯ = P = 3[M P a] = 0. The best approximation is by using inserted cone for supersonic flow and measure the oblique shock.2: The Mach number at point A on tube is measured to be M = 23 and the static pressure is 2[Bar]4 .4 × 287 × 294 = 304[m/sec] (4. Calculate the Mach number at point B under the isentropic flow assumption.69428839 × J ρ0 RT0 × 294[K ] 287.60000 0. and velocity can be calculated.076K ρ0 ρ= ρ P0 5 × 106 [P a] = 0.0115 0. estimate the temperature at point B. Downstream at point B the pressure was measured to be 1. T = 0.2) or from Figure (4.4 and assume a perfect gas model. Assume that the specific heat ratio k = 1.60693 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 51 the temperature.6 P P0 5[M P a] From Table (4.44) For a circular tube the diameter is about 1[cm].0 kgK kg m3 = 41. the cross section can be calculated with all these information. Here it is subsonic and this technique is not suitable.53105 With these values the static temperature and the density can be calculated.2.86420338 × (273 + 21) = 254. 4 Well.69428 1. Finally.4. Example 4. or simply using the equations shows that M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.

55556 0. Assume that the flow is isentropic and the gas specific heat ratio is 1.42[K ] Example 4. At point B in the duct the cross section area is 40 [cm2 ]..6265306 and second subsonic M = 0.3: Gas flows through a converging–diverging duct.8137788 and TB = B 0.2 1.81. At point “A” the cross section area is 50 [cm2 ] and the Mach number was measured to be 0.2) or from Potto-GDC two solutions can be obtained.4.0000 T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.55555556 271. S OLUTION To obtain the Mach number at point B by finding the ratio of the area to the critical area.12780453 × = 0. The supersonic branch solution is possible only if there where a transition at throat where M=1.5 The corresponding Mach number for this pressure ratio is 1.17040879. At M = 2 (supersonic flow) the ratios are M 2. ISENTROPIC FLOW S OLUTION With the known Mach number at point A all the ratios of the static properties to total (stagnation) properties can be calculated. Therefore. The stagnation temperature can be “bypassed” to calculate the temperature at point B M =2 M =1.2 @ M = 2 PA = P0 PB P0 × PA 2. TB = T A × T0 × TA TB T0 = 250[K ] × 1 × 0. .60315132 P P0 = 0.53884934.17040604 PB 1. the stagnation pressure at point A is known and stagnation temperature can be calculated.60315132 0.12780 0.6875 0. Both solution are possible and acceptable. Find the Mach number at point B. This relationship can be obtained by AB AB AA 40 = × ∗ = × A∗ AA A 50 from the Table 4.59014 = 1.4.21567 0.23005 1.0 = 0.272112 B With the value of A A∗ from the Table (4.52 CHAPTER 4.59309 With this information the pressure at point B can be expressed as from the table 4. The two possible solutions: the first supersonic M = 1.

e. It can be noticed that at the throat when the flow is chocked or in other words M = 1 and that the stagnation conditions (i.94511 0. dividing equations (4.2721 0.46) Expressing the temperature in terms of Mach number in equation (4. It is convenient to rearrange the equation (4.28772 0. temperature.86838 1.22617 0.49) Equation (4.53887 0.2.4.34585 1.2721 0.82071 1.45) to be expressed in terms of the stagnation properties as f (M. Hence equation (4.65396 0.47) obtained the form √ k+1 − 2( k−1) m ˙ kP0 k−1 √ (4.47) It can be noted that equation (4.46) results in m ˙ = A kM P0 √ kRT0 1+ k−1 2 M 2 k+1 − 2( k−1) (4. The area ratio is defined as the ratio of the cross section at any point to the throat area (the narrow area). The throat area can be denoted as by A∗ . to examine the maximum flow rate and to see what is the effect of the compressibility on the flow rate.k) m ˙ P P0 U √ = A P0 kRT k R T0 1 P √ = √0 M T T0 T0 k P R P0 T0 T (4.45) This parameter is studied here. pressure) do not change.48) by equation (4.49) relates the Mach number at any point to the cross section area ratio.3 Mass Flow Rate (Number) One of the important engineering parameters is the mass flow rate which for ideal gas is m ˙ = ρU A = P UA RT (4.6266 0. specifically and explicitly the relationship for the chocked flow.47) yields 1 A = ∗ A M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (4.0440 4.2.47) holds everywhere in the converging-diverging duct and this statement also true for the throat. The area ratio as a function of the Mach number needed to be established. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 53 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 1.48) = 1 + ∗ A 2 RT0 Since the mass flow rate is constant in the duct. .

53) (4. c0 √ m ˙ kRT0 mc ˙ 0 m ˙ T0 1 √ =√ =√ ∗ ∗ ∗ A P0 kRA P0 RA P0 k The definition of Fliengner’s number (Fn) is Fn ≡ √ mc ˙ 0 RA∗ P0 Fn (4.52) Equation (4. ISENTROPIC FLOW The maximum flow rate can be expressed either by taking the derivative of equation (4.56) “Naughty Professor” Problems in Isentropic Flow To explain the material better some instructors invented problems.54) Utilizing Fliengner’s number definition and substituting it into equation (4.51) max The maximum flow rate for air (R = 287j/kgK ) becomes.55) and the maximum point for F n at M = 1 is Fn = k k+1 2 k+1 − 2( k−1) (4. (see for example. It can be noticed that Fliengner’s equation can lead to definition of the Fliengner’s Number.52) is known as Fliegner’s Formula on the name of one of the first engineers who observed experimentally the choking phenomenon. k = 1.68473 P √0 ∼ √ T0 R (4.5)).040418 A ∗ P0 (4. Carrying this calculation results at M = 1. m ˙ A∗ P √0 = T0 k R k+1 2 k+1 − 2( k−1) (4.4 m ˙ A∗ 0. which have mostly academic proposes. While these .48) results in F n = kM k−1 2 M 1+ 2 k+1 − 2( k−1) (4.54 CHAPTER 4.48) in with respect to M and equating to zero.50) max For specific heat ratio. √ m ˙ T0 = 0. Shapiro (problem 4.

the terms 0 for the group over the under–brace approaches zero when the flow rate (Mach number) is very small. The situation where the mass flow rate per area given with one of the stagnation properties and one of the static properties. Using energy equation (4.61) version 0.g.59) Equation (4.58) And further Rearranging equation (4. .4.59) is quadratic equation for density. e.57) result in p R m ˙ Aρc 2 (4.57) 1/kR T0 ρ 2 = T ρ ρ + T c2 k−1 2 m ˙ A 2 (4. ρ when all other variables are known. P0 and T or T0 and P present difficulty for the calculations. The case of T0 and P This case considered to be simplest case and will first presented here. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 55 problems have a limit applicability in reality. In other words. It is convenient to change it into ρ2 − k−1 Pρ − T0 R 2kRT0 m ˙ A 2 =0 (4. For this kind of problems a new Table was constructed and present here5 . 5 Since    (4. The use of the regular isentropic Table is not possible because there isn’t variable represent this kind problems.9) and substituting for Mach number M = m/Aρc ˙ results in k−1 T0 =1+ T 2 Rearranging equation (4.44 of this book.60) The only physical solution is when the density is positive and thus the only solution is    P 1 ρ=  + 2 RT  0 P RT0 2 k−1 +2 kRT0 m ˙ A →(M →0)→0 2  For almost incompressible flow the density is reduced and the familiar form of perfect gas model is seen since stagnation temperature is approaching the static P temperature for very small Mach number (ρ = RT ).58) transformed it into ρ2 = k−1 Pρ + T0 R 2kRT0 m ˙ A 2 (4. they have substantial academic value and therefore presented here.2.

1).65) Again notice that the right hand side of equation (4. the problems is reduced to finding tabulated values. The case of P0 and T A similar problem can be described for the case of stagnation pressure.63) =f (M ) RT0 P2 Thus.66) is only function of Mach AP were tabulated number (well. this dimensionless group is function of Mach number only. ISENTROPIC FLOW It is convenient to denote a new dimensionless density as ρ ˆ= ρ p RT0 = 1 ρRT0 = ¯ P T (4. again the specific heat ratio. And the values of A ∗P 0 in Table (4. T . F n2 A ∗ P0 AP With this new definition equation (4. m ˙ A 2 = 1 c0 2 k P0 2 m ˙ A∗ 2 A∗ A 2 P0 P 2 (4.68) .61) is transformed into   2 ˙ 1 (k − 1)RT0 m  1+ 1+2 ρ ˆ= 2 kP 2 A (4. P0 . First. and static temperature.67) (4. RT P0 2 It can be noticed that F n2 = k T T0 P0 P 2 Hence. also the specific heat. k also). it is shown that the dimensionless group is a function of Mach number only (well.64) RT0 P2 m ˙ A 2 = F n2 k A ∗ P0 AP 2 (4.2) and Fn is tabulated in the next Table (4.62) The dimensionless density now is related to a dimensionless group that is a function of Fn number and Mach number only! Thus. the dimensionless density is  1 (k − 1)F n2 ρ ˆ = 1 + 1 + 2 2 k2 A∗ P AP 0 2   (4.66) m ˙ A 2 = F n2 k A ∗ P0 AP 2 T T0 P0 P 2 (4. Thus.56 CHAPTER 4. k ).

Based on the same arguments. Unfortunately.2. All the above conditions/situations require to have the perfect gas model as the equation of state. if it was shown that a group of parameters depends only Mach number than the Mach is determined by this group. First.69) The right hand side is tabulated in the “regular” isentropic Table such (4. c0 2 2 2 1 Rρ0 P m ˙ A = kRT0 P kRP0 P0 P 0 m ˙ A = c0 2 P kRP0 2 P 0 m ˙ A 2 = F n2 k P0 P (4. For example the first “naughty professor” question is sufficient that process is adiabatic only (T0 . The method of solution for given The case of ρ0 and T or P The last case sometimes referred to as the “naughty professor’s question” case dealt here is when the stagnation density given with the static temperature/pressure.70) The last case dealt here is of the stagnation density with static pressure and the following is dimensionless group c0 2 2 2 1 Rρ0 2 T m ˙ A = kRT0 T0 kRP0 2 T m ˙ A = c 0 2 T0 kRP0 2 T m ˙ A 2 = F n2 k T0 T (4.67) became RT P0 2 m ˙ A 2 = A ∗ P0 AP 2 (4.).2).2). What is the legitimacy of this method? The explanation simply based the previous experience in which for a given ratio of area or pressure ratio (etcetera) determines the Mach number. (4. mass flow rate per area.4. This example shows how a dimensional analysis is used to solve a problems without actually solving any equations.71) It was hidden in the derivations/explanations of the above analysis didn’t explicitly state under what conditions these analysis is correct. P . not all the analysis valid for the same conditions and is as the regular “isentropic” Table. The heat/temperature part is valid for enough adiabatic condition while the pressure condition requires also isentropic process. . ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 57 Thus equation (4. the dimensionless approach is used later analytical method is discussed (under construction). The actual solution of the equation is left as exercise (this example under construction).

000 0.0 0.10397 0.079722 0.65246 1.035 0.012 0.019 0.12724 0.074254 0.028 0.060404 0.23155 0.37432 0.55000 0.009 0.003 0.57656 0.25535 0.20316 0.57709 0.014268 0.067111 0.16522 0.12239 0.54422 1.62436 0.44192 0.00865 0.69948 0.34330 0.67129 1.44000 0.065 0.050 0.27000 0.54733 0.85261 1.35316 0.33378 0.49305 0.31480 0.20000 0.41073 1.36000 0.006 0.51932 0.27608 0.20442 0.57253 0. ISENTROPIC FLOW Table 4.003 0.26264 0.29247 0.41855 0.13796 0.023 0.22634 0.21703 0.42000 0.52000 0.54000 0.15372 0.46798 0.45000 0.060 0.71967 0.32220 0.004 0.56172 1.39478 1.40333 0.49285 1.51000 0.18896 0.29000 0.055 0.33233 1.26000 0.11928 0.78965 1.95791 0.39596 0.49000 0.52690 1.48360 0.028651 0.31000 0.74912 1.39000 0.85107 0.91838 1.23777 0.41997 0.021 0.41338 0.87421 1.0 0.031 0.00707 0.001 0.29663 0.077 0.42683 1.11294 0.32000 0.36550 0.37000 0.089910 0.36329 1.029920 0.003 0.059212 0.10752 0.012593 0.057647 0.60706 0.14276 0.72927 1.34000 0.57944 1.00E +00 1.14084 1.081847 0.81706 0.000 0.24773 0.65857 0.78382 0.75136 0.14592 0.050518 0.38884 0.18709 0.000424 0.0 2.095449 0.13342 0.58 CHAPTER 4.48000 0.30214 0.37896 1.071 0.044110 0.51882 0.30000 0.033229 0.010476 0.35361 0.038365 0.54531 0.087372 0.026 0.89613 1.24000 0.88588 0.28000 0.46000 0.12039 0.28677 1.36764 0.61550 1.015 0.038 0.17728 0.16581 0.34775 1.005 0.44215 0.76924 1.070106 1.17397 0.015027 0.30418 0.13232 0.62915 0.33465 0.005 0.30185 1.55637 0.60047 0.083989 0.065654 0.400E −06 1.18992 0.23137 0.074314 0.24674 0.33000 0.41000 0.66098 0.49249 0.27926 0.27358 0.098460 0.14927 0.43000 0.17381 0.50978 1.70969 1.44363 0.47000 0.065557 0.52485 0.10639 0.072487 0.23000 0.011 0.013 0.094654 0.58952 0.001 0.73995 0.13284 0.31703 1.1: Fliegner’s number and other paramters as function of Mach number M Fn ρ ˆ P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ 0 2 T m ˙ 2 A 0.001 0.042 0.69036 1.49485 0.007 0.20458 0.16121 0.000 0.20109 0.25000 0.28307 0.18428 0.52858 0.11710 0.017 0.63386 1.26495 0.017813 0.046 0.83132 1.22000 0.21000 0.002 0.59736 1.22085 0.47609 1.008 0.46633 0.92149 0.15963 0.81034 1.00747 0.14889 0.014197 0.40000 0.024585 0.10000 0.00352 0.0 0.62E −05 0.35000 0.45951 1.050001 0.37474 0.25018 0.002 0.31203 0.53000 0.39701 0.33226 0.43919 0.00351 0.46677 0.63889 .68875 0.020986 0.44309 1.50000 0.21584 0.38000 0.

814 1.541 3.058 2.211 1.269 1.161 1.541 1.407 2.68000 0.63000 0.037 1.072 1.538 1.109 1.342 7.535 2.81000 0.76000 0.94096 0.898 1.99514 1.72000 0.355 1.278 1.500 1.806 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ0 2 T m ˙ 2 A 0.949 9.368 1.422 1.93000 0.709 3.082 2.87424 0.536 2.853 1.96389 0.448 1.362 1.574 1.220 3.461 1.278 2.104 3.740 2.65000 0.288 1.20 0.738 1.937 1.661 1.771 1.096 2.043 4.869 1.66000 0.846 8.99507 1.124 1.144 ρ ˆ 1.135 1.924 1.699 3.657 1.59000 0.187 1.509 1.90000 0.500 1.86101 0.971 3.225 1.90382 0.67000 0.790 2.526 1.477 1.168 2.081 2.343 1.607 1.78062 0.87000 0.761 1.73000 0.791 1.2.57000 0.805 0.610 1.317 1.712 1.813 2.414 1.88142 0.006 6.462 3.89000 0.972 5.105 0.82000 0.81996 0.320 1.78000 0.819 1.74000 0.75000 0.78250 0.977 2.389 3.74624 0.131 1.083 1.244 5.81139 0.177 2.202 1.088 1.271 1.86000 0.044 1.84000 0.64000 0.121 3.457 3.667 1.489 2.95665 1.241 2.70675 0.326 1.156 2.94000 Fn 0.105 1.71000 0.557 1.821 1.810 2.758 1.284 2.642 1.581 1.602 2.92000 0.998 3.607 1.571 4.582 1.587 1.113 1.383 1.011 1.553 1.80000 0.033 1.216 2.101 2.70000 0.670 2.897 5.446 1.660 1.646 3.381 8.98717 1.469 2.382 1.97562 1.324 1.903 1.427 1.671 2.059 2.185 1.79000 0.059 1.58000 0.214 1.979 4.67210 0.62000 0.4.097 1.88000 0.244 3.094 1.991 2.112 1.090 1.018 2.676 1.721 4.56000 0.385 1.931 2.865 .405 1.82722 0.148 1.038 2.68565 0.840 4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 59 Table 4.381 2.173 1.784 1.423 1.554 10.340 2.703 1.260 1.029 3.736 1.671 1.953 2.084 1.91000 0.147 1.882 2.035 1.240 1.508 1.720 2.62936 0.869 7.236 1.595 2.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.77000 0.266 4.842 2.141 1.217 1.513 5.412 2.708 1.484 4.415 1.194 1.212 1.859 1.516 1.613 6.747 1.235 5.151 1.122 1.94848 0.450 1.162 1.74290 0.83000 0.349 1.404 2.474 1.980 2.258 4.255 1.963 3.61000 0.602 2.241 1.85000 0.69000 0.323 1.287 3.92366 0.152 1.60000 0.198 1.030 1.277 2.279 1.305 1.640 1.955 3.424 6.297 1.

Calculate for that point the Mach number.913 4.37 13.88 11. The second model which there is significant heat transfer but insignificant pressure loss (Rayleigh flow like).4: A gas flows in the tube with mass flow rate of 1 [kg/sec] and tube cross section is ◦ 0.003 2.419 ρ ˆ 1.5[Bar]. The third academic condition is of static temperature and the static pressure. A constant flow rate requires that m ˙A = m ˙ B .3.47) is appropriate regardless the flow is isentropic or adiabatic. There are two possible models that can be used to simplify the calculations.72) From equation (4.324 2.961 2.98000 0.804 5.97000 0.843 1.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.583 10. A∗ ) = constant.404 4.96000 0. in a case where the flow isn’t isentropic or adiabatic the total pressure and the total temperature will change (due to friction. Assume that the process is isentropic and k=1.60 CHAPTER 4. ISENTROPIC FLOW Table 4. and heat transfer).000 Fn 2.500 3.016 Example 4. At some point the static pressure was measured to be 1.136 7.181 3.188 2.95000 0. and the stagnation pressure. the velocity.278 2.371 2.436 6. it is clear that the function f (P0 . Denoting subscript A for one point and subscript B for another point mass equation (4.19 14. In the same manner the definition of A∗ referred to the theoretical minimum area (”throat area”) if the flow continues to flow in an isentropic manner. S OLUTION The second academic condition is when the static temperature is given with the stagnation pressure.600 4.06 14. That expression was derived based on the theoretical total pressure and temperature (Mach number) which does not based on the considerations whether the flow is isentropic or adiabatic. . Flow with pressure losses The expression for the mass flow rate (4.419 3. The first model for neglected heat transfer (adiabatic) flow and in which the total temperature remained constant (Fanno flow like).98 6. The temperature at Chamber supplying the pressure to tube is 27 C .72).99000 1.259 3.60 12.515 7.777 7.881 1.046 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ 0 2 T m ˙ 2 A 3.217 4. Clearly.037 4.112 6.001[m2 ]. T0 .338 3.233 2.920 1.48) can be equated as kP0 A∗ RT0 k−1 2 1+ M 2 k−1 − 2( k−1) = constant (4.

55401 0.6367 0.75) the stagnation pressure at point B can be obtained.75) For a flow with a constant stagnation pressure (frictionless flow) and non adiabatic flow reads T 0 |A = T 0 |B B A ∗ MB A A ∗ MA A| B A| A 2 (4.68966 0.13169 2. The total pressure can be calculated because the Mach number and static pressure are known.01[m2 ]. calculated the total pressure lost.05853 0. thus the area ratios can be calculated. S OLUTION Both Mach numbers are known. Mach number is 2. comparison of mass flow rate at point A and point B leads to P 0 A ∗ |A = P 0 A ∗ |B .015[m2 ] and Mach number is 1.5. and utilizing equation (4.1762 0.5: At point A of the tube the pressure is 3[Bar].62693 A×P A∗ ×P0 F F∗ .4. and the duct section area is 0. the cross section area is 0.5. And utilizing the equality of A∗ = P 0 |A A∗ | = ∗A P 0 |B A |B A∗ A A (4.74) leads to A A ∗ MA A A ∗ MB P 0 |A = P 0 |B A| A A| B (4.73) For adiabatic flow. M 1. Assume no mass lost and adiabatic steady state flow. point B.32039 0.15432 0.5000 T T0 ρ ρ0 A A P P0 0.44444 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 61 If the mass flow rate is constant at any point on the tube (no mass loss occur) then k RT0 2 k+1 k+1 k−1 m ˙ =A ∗ P0 (4. Downstream at exit of tube.76) Example 4.39498 1. With these information.5000 2.2.27240 0.

0000 0.340 1.79158 0.99303 0.39498 5.48290 0.882 1.050 0.98028 0.83333 0.96899 0.99800 0.58377 0.91075 0.600 0.066 1.964 2.88652 0.99502 0.53114 0.35036 0.008 1.000 1.2) as P 0 |A = P P P0 M =2.55425 0.57 5.86059 0.42493 0.99825 0.243 = 4.97250 0.4 M 0.92427 0.74738 0.59126 0. ISENTROPIC FLOW First.900 1.63535 0.400 1.8E + 5 11.52828 0.300 1.95238 0.78896 0.27240307 × 15.68704 0.800 0.41238 0.4E +5 4.52989 0.700 0. 0.53807 0.000 0.188 1.243 = 36.32039 A×P A∗ ×P0 F F∗ 2.300 0.25781291[Bar] 0.53399 0.500 0.038 1.2: Isentropic Table k = 1.01 1.015 4.93284 0.25781291 × Hence P0 |A − P0 |B = 51.781 2.00000 0.93155 0.89561 0.013[Bar] Note that the large total pressure loss is much larger than the static pressure loss (Pressure point B the pressure is 0.68966 1.838 2.78400 0.62 CHAPTER 4.058527663 by utilizing equation (4.100 1.146[Bar]).77640 0.52828 0.590 1.5 A = 3 = 51.63394 0.424 1.243[Bar] 2.268 0.176 1.75) provides P0 |B = 51.84045 0.88517 0.53974 0.31424 0.53039 0.912 1.52828 0.00000 0.54655 0.93947 0.6367187 0.72093 0.200 1.035 1.400 0.094 1. the stagnation at point A is obtained from Table (4.030 1.98232 0.72632 0.68110 0.200 0.80515 0.47207 0.59650 0.8E +5 11.500 T T0 ρ ρ0 A A P P0 1.65602 0.100 0.130 0.89699 0.36091 0.00 1.3 Isentropic Tables Table 4.58170 0.38484 0.822 2.27240 5.99206 0.59 5.95638 0.43742 0.99950 0.46835 0.73999 0.257 − 15.115 1.99875 0.443 1.55401 .1761671 × ≈ 15.009 1.71839 0.84302 0.

4 (continue) 63 M 1.1 4.700 1.022046 0.045233 0.2: Isentropic Table k=1.500 3.18 75.23005 0.4E +2 1.74E −5 3.039628 0.19802 0.790 10.500 4.00261 0.74058 0.27099 0.3.000242 0.90E −5 4.000631 0.000 7.69983 0.56976 0.015504 0.00346 0.58549 0.000155 0.3.637 4.00189 0.057227 0. ISENTROPIC TABLES Table 4.00190 0.017449 0.3E +2 4.28682 0.31E −5 2.500 7.013111 0.4. dP = dρRT Substituting equation (4.027662 0.59309 0.500 9.72136 0.3.00107 0.000102 6.20259 0.58072 0.00141 0.70876 0.500 5.73723 0.000 2.058140 0. The perfect gas model is again assumed (later more complex models can be assumed and constructed in a future versions). Eckert number is very small) is presented.17404 0.2 Isentropic Isothermal Flow Nozzle General Relationship In this section.5E +2 3.000 6.23810 0. knowing the two limits provides a tool to examine where the reality should be expected.500 10.11528 0.500 8.900 2.56 25.66138 0.013957 0.092593 0.25044 0.14924 0.072464 0. (e.21567 0.36E −5 0. the other extreme case model where the heat transfer to the gas is perfect.338 1.058528 0.73257 0.67320 0.77) 6 The one dimensional momentum equation for steady state is U dU/dx = −dP/dx +0(other effects ) which are neglected here. .73510 0.10582 0.57768 0.00364 0.00107 0.15432 0.65326 0.00 36.00519 0.011340 0.29414 0.439 1. Again in reality the heat transfer is somewhere in between the two extremes.000 8.000633 0.4E +2 0.55556 0.13169 0.052493 0.500 6.16667 0.72586 0.28986 0.62693 0.23211 0. So.0E +2 1.60680 0.235 6.72953 0.68830 0.012628 A×P A∗ ×P0 F F∗ 0.00758 0.047619 0.77) into the momentum equation6 yields U dU + RT dP =0 P (4.000 9.87 53.63371 0.00 T T0 ρ ρ0 A A P P0 0.74192 4.089018 0.00659 0.600 1.9E +2 2.800 1.71578 0.72 16.73903 0.081633 0.033682 0.35573 0.000815 0.g.025156 0.14184 0.12780 0.688 2.25699 0.2E +2 5.13 1.076226 0.064725 0.028962 0.070595 0.000 4.31969 0.000 3.027224 0.555 1.56182 0.000 5.019473 0.44444 0.78) (4.047251 0.250 1.000385 0.017321 0.000495 1.12195 0.23527 0. In isothermal process the perfect gas model reads P = ρRT .35714 0.

U2 2 − U 1 2 P2 + RT ln =0 2 P1 (4. P2 k (M2 2 − M1 2 ) = ln 2 P1 (4.87) .82) Transform from equation (4.84) Or in terms of the pressure ratio equation (4.78) yields the Bernoulli’s equation for ideal gas in isothermal process which reads .84) reads k(M1 2 −M2 2 ) P2 2 = =e P1 e M1 2 e M2 2 k 2 (4.79) to a dimensionless form becomes constant constant  (M2 2 − M1 2 ) kRT P2   ln = R  . U1 ≈ 0 reads U2 = 2RT ln P2 P1 (4. ISENTROPIC FLOW Integration of equation (4. T 2 P1 (4.81) Or in explicit terms of the stagnation properties the velocity is U= 2RT ln P P0 (4.64 CHAPTER 4.83) yields . the velocity at point 2 becomes U2 = 2RT ln P2 − U1 2 P1 (4.85) As oppose to the adiabatic case (T0 = constant) in the isothermal flow the stagnation temperature ratio can be expressed 1  ¡ 1+ T0 1 T1 = ¡ T0 2 T2 1 + ¡ 2 k−1 2 M1 2 k−1 2 M2 = 1+ 1+ 2 k −1 2 M1 2 k −1 2 M2 (4.79) Thus.80) The velocity at point 2 for stagnation point.83) Simplifying equation (4.86) Utilizing conservation of the mass AρM = constant to yield M 2 P2 A1 = A2 M 1 P1 (4.

the critical stagnation pressure reads 2 k−1 (1−M )k 2 1 + P0 2 M1 2 = e P0 ∗ k+1 k k−1 (4.85) to read (1−M 2 )k ρ P = ∗ =e 2 ∗ P ρ (4.88) The change in the stagnation pressure can be expressed as P0 2 P2 = P0 1 P1 1+ 1+ 2 k−1 2 M2 2 k−1 2 M1 e M1 = 2 e M1 2 k 2 (4.93) Of course in isothermal process T = T ∗ . is unknown (at what Mach number the nozzle is choked is unknown) so there are two possibilities: the choking point or M = 1 to normalize the equation. ISENTROPIC TABLES Combing equation (4.92) Finally.89) The critical point.4. Again it has to emphasis that this critical point is not really related to physical critical point but it is arbitrary definition.87) and equation (4. Here the critical point defined as the point where M = 1 so results can be compared to the adiabatic case and denoted by star. The critical pressure ratio can be obtained from (4.91) (4.3. The minimum of the curve is when area is minimum and at the point where the flow is choked.90) Equation (4.94) . All these equations are plotted in Figure (4.6).3 it can be observed that minimum of the curve A/A∗ isn’t on M = 1. The true critical point is when flow is choked and the relationship between two will be presented. The mathematical procedure to find the minimum is simply taking the derivative and equating to zero as following d A A∗ dM = kM 2 e k(M 2 −1) 2 M2 −e k(M 2 −1) 2 =0 (4. From the Figure 4. It should be noted that the stagnation temperature is not constant as in the adiabatic case and the critical point is the only one constant. at this stage.88) is reduced to obtained the critical area ratio writes A 1 (1−M 2 )k = e 2 ∗ A M Similarly the stagnation temperature reads 2 1 2 1 + k− T0 2 M1 = T0 ∗ k+1 k k−1 (4.85) yields M1 A2 = A1 M2 65 e M2 2 e M1 2 k k−1 k 2 (4.

7(a)).96) On the other hand. The velocity √ k .5 3 2. Thus.5 2 1. M = √ k (4. It is convenient to find expression for the ratio of the initial stagnation pressure (the stagnation pressure before entering the nozzle) to the pressure at the throat. the pressure loss in adiabatic flow is milder as can be seen in Figure (4. ISENTROPIC FLOW Isothermal Nozzle 4 3.5 0 0 0. It should be emphisized that the stagnation pressure decrees. dividing the at the throat of isothermal model is smaller by a factor of √ critical adiabatic velocity by k results in Uthroatmax = √ RT (4.5 4 k=14 P/P * A/A * P0 / P0 T 0 / T0 T/T * * * Tue Apr 5 10:20:36 2005 Fig.5 3 3.5 1 1. 4.95) It can be noticed that a similar results are obtained for adiabatic flow.6: Various ratios as a function of Mach number for isothermal Nozzle Equation (4.5 1 0.66 CHAPTER 4.94) simplified to 1 kM 2 − 1 = 0 . Utilizing equation .5 2 M 2.

7: The comparison of nozzle flow (4.5 Distance (normalized distance two scales) 2 M isoT M isentropic Uisntropic/UisoT k=14 k=14 Tue Apr 5 10:39:06 2005 Thu Apr 7 14:53:49 2005 (a) Comparison between the isothermal nozzle and adiabatic nozzle in various variables (b) The comparison of the adiabatic model and isothermal model Fig. the heat transfer is expressed as (4.5 3 2.3.60653 1 Notice that the critical pressure is independent of the specific heat ratio.5 4 3. k .90) the following relationship can be obtained P ∗ Pthroat Pthroat = = P0initial P0initial P ∗ 1 e (1−02 )k 2 e „ 1− “ 1 √ k ”2 « k 2 = (4.5 2 1.5 1 0. Again.4.5 1 1. ISENTROPIC TABLES Isothermal Nozzle 4 3. 4.5 3 2.5 0 0 1 0.5 2 M 2.5 4 A / A iso * A / A adiabatic * P / P iso * P / P adiabatic * 67 Comparison between the two models 5 4.98) Q = Cp (T02 − T02 ) .5 2 1.5 0 0 0. It also has to be emphasized that the stagnation values of the isothermal model are not constant.5 3 3.5 1 0. as opposed to the adiabatic case.5 1.97) e− 2 = 0.

100) the velocity ratio was obtained and is plotted in Figure (4. The Mach number was computed for the two models and plotted in Figure (4. .7(b)).8 T / T0 isentropic but rather a simple sample P / P0 isothermal T/T0 isothermal just to examine the difference 0. using the isentropic model results in under prediction of the actual results for the velocity in the supersonic branch. The initial stagnation temperatures are almost the same and can be canceled out to obtain Us ∼ UT MT Ms 1+ 2 k−1 2 Ms (4. The Mach number for the isentropic is larger for the supersonic branch but the velocity is lower.6 between the two models so in an actual situation it can be 0. ISENTROPIC FLOW For comparison between Comparison between the two models the adiabatic model and the k=14 isothermal a simple profile of nozzle area as a function of 1 the distance is assumed.7(b)). The ratio of the velocities can be expressed as √ Ms kRTs Us √ = (4. Thus. As can be observed from Figure (4.5 1. In this Fig.8: Comparison of the pressure and temperature drop as a function of the normalized length comparison the entrance area (two scales) ratio and the exit area ratio are the same and equal to 20. 4. While. This P / P0 isentropic profile isn’t an ideal profile 0.68 CHAPTER 4.4).4 bounded. To make sense and eliminate unnecessary details 0.7(b)). the isentropic for the subsonic branch will be over prediction. The prediction of the Mach number are similarly shown in Figure (4.2 the distance from the entrance to the throat is normalized (to 0 0 1 2 0.7(b)).99) UT MT kRTs It can be noticed that temperature in the isothermal model is constant while temperature in the adiabatic model can be expressed as a function of the stagnation temperature. In the same fashion Distance (normalized distance two scales) the distance from the throat to Fri Apr 8 15:11:44 2005 the exit is normalized (to one (1)) (it doesn’t mean that these distances are the same). In this comparison it has to be remembered that critical area for the two models are different by about 3% (for k = 1.100) By utilizing equation (4.5 one (1)).

The temperature ratio of T /T0 int can be obtained via the isentropic model as T 1 = k−1 T0 int 1 + 2 M2 (4. If the nozzle was made from copper (a good heat conductor) a larger heat transfer occurs.103) where z is an arbitrary point on the nozzle. To demonstrate the relativity of the approach advocated in this book consider the following example. Example 4.4.101) While the temperature ratio of the isothermal model is constant and equal to one (1).3.4. Figure (4. the sought ratio is provided. Using equations (4. The initial stagnation temperature is denoted as T0 int .89) and the isentropic relationship. The stagnation pressure is 5[Bar] and the stagnation temperature is 27◦ C. The meaning of this analysis is that transferred heat affects the temperature to a larger degree but the effect on the pressure is much less significant. The pressure ratio for the isentropic model is P = P0 int 1+ 1 k−1 2 2 M k−1 k (4.e.102) and for the isothermal process the stagnation pressure varies and has to be taken into account as the following: isentropic P0 P0 z Pz = P0 int P0 int P0 ∗ ∗ Pz P0 z (4. Calculate the velocity at the exit using the adiabatic model. Assume that the back pressure is low enough to have supersonic flow without shock and k = 1.8) shows that the range between the predicted temperatures of the two models is very large. with known area ratio and known stagnation Potto–GDC provides the following table: . should the velocity increase or decrease? What is the maximum possible increase? S OLUTION The first part of the question deals with the adiabatic model i. the conservation of the stagnation properties. The throat area ratio to entrance area is 1:4 respectively.6: Consider a diverging–converging nozzle made out of wood (low conductive material) with exit area equal entrance area. Thus. while the range between the predicted pressure by the two models is relatively small. ISENTROPIC TABLES 69 Two other ratios need to be examined: temperature and pressure.

6) or using the Potto–GDC obtains the following table M 1.92 × 300 = 576.02979 3.98934 0.4940 0. the velocity can be calculated.36644 × 300 = 109. The exit velocity.777777778 = 1.777777778 11 k−1 T0 ∗ 1 + k− 1 + 2 k k Thus the stagnation temperature at the exit is T0ini = 1.9910 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 With the known Mach number and temperature at the exit. ISENTROPIC FLOW A A P P0 A×P A∗ ×P0 0. is √ √ U = M kRT = 2.9 ∼ 617.104) As was discussed before.9K .0000 4.4940/0.12556 0.4 × 287 × 300.11915 Even for the isothermal model.70 M 0.93[m/sec] 1.0000 0.50225 The exit Mach number is known and the initial temperature to the throat temperature ratio can be calculated as the following: 1 T0ini 1 = = = 0.99572 0. then.14655 2.2K .36644 0.9402 1. However.08129 4. Using the area ratio in Figure (4.253[m/sec] 4.4 4. but in the literature some denote this function as I . To explain the motivation for using this definition consider the calculation of the net forces that acting on section shown in Figure (4.105) .1 The Impulse Function Impulse in Isentropic Adiabatic Nozzle One of the functions that is used in calculating the forces is the Impulse function. To calculate the net forces acting in the x–direction the momentum equation has to be applied Fnet = m ˙ (U2 − U1 ) + P2 A2 − P1 A1 (4.0000 0.9). the maximum velocity cannot exceed the 691.4 × 287 × 109.921 T0exit The exit stagnation temperature is 1.98511 0.9402 T T0 ρ ρ0 CHAPTER 4. the initial stagnation temperature is given as 300K .51183 4.253[m/sec] (4.4. The Impulse function is denoted here as F .9405 0. The exit temperature is 0. The exit velocity can be determined by utilizing the following equation √ √ Uexit = M kRT = 1. the velocity in the copper nozzle will be larger than the velocity in the wood nozzle.0 = 691.9910 1.

106) and by utilizing the identity in equation (4. Thus. Hence.106) Combining equation (4.109) (4.42) results in Rearranging equation (4.108) which makes the ratio a function of k and the Mach number.4. 4. defining the Impulse function as F = P A 1 + kM2 2 In the Impulse function when F (M = 1) is denoted as F ∗ F ∗ = P ∗ A∗ (1 + k ) The ratio of the Impulse function is defined as see function (4.111) This ratio is different only in a coefficient from the ratio defined in equation (4.4. The mass conservation also can be applied to our control volume m ˙ = ρ 1 A 1 U1 = ρ 2 A 2 U2 (4.108) P1 A1 1 + kM1 F = ∗ ∗ F∗ P A (1 + k ) 2 (4.108) shows that the right hand side is only a function of x-direction Mach number and specific heat ratio. Defining a function that depends only Fig.107) P2 A 2 P1 A 1 Fnet = 1 + kM2 2 − 1 + kM1 2 ∗ ∗ P0 A P0 A P0 A ∗ (4.9: Schematic to explain the signifion the Mach number creates the concances of the Impulse function venience for calculating the net forces acting on any device.105) with equation (4. M and k . k .107) and dividing it by P0 A∗ results in f (M2 ) f (M2 ) f (M1 ) f (M1 ) Fnet = kP2 A2 M2 2 − kP1 A1 M1 2 − P2 A2 − P1 A1 (4.108) Examining equation (4. THE IMPULSE FUNCTION 71 The net force is denoted here as Fnet . if the right hand side is only a function of the Mach number and k than the left hand side must be function of only the same parameters.110) = 1 P∗ P0 2 ( k+1 ) k−1 k P1 A 1 1 + kM1 2 P0 A ∗ 1 (1 + k ) (4. the net force is Fnet k+1 = P0 A (1 + k ) 2 ∗ k k−1 F2 F1 − ∗ ∗ F F (4.112) . Hence.

009 A1 = = 2.009[m ] and T0 = 400K 2 P2 = 50[Bar] the exit area is 0.2121 With the area ratio of A A 0.003 And utilizing again Potto-GDC provides M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.10: Schematic of a flow of a compressible subzle and pressure at point 1. The stagnation temperature is 400K and the pressure at point 2 was measured as 5[Bar] Calculate the net force acting on the noz.96666 = 2. provides the following √ √ m ˙ RT 1.99132 5.058 and 500000 × 0.99380 5.776[Bar] P2 P0 .2121 × = 5.11164 0.003m2 entrance area is 0.99751 0.94934 2.1774 2.2121 the area ratio of at point 1 can be calculated. the ratio of P1 A1 /A∗ P0 is needed to be calculated.72 CHAPTER 4.1000 0. stance (gas) thorough a converging nozzle S OLUTION The solution is obtained by getting the data for the Mach number.99380 ∼ 4.7) 0.003[m ]. To obtain this ratio the denominator is needed to be obtained.27353 0.76[N ] 0.009m2 2 A2 = 0. 4.98526 0.76 M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 1 for example (4. A2 A1 0.52).94934/0.Fig.7: Consider a flow of gas into a 2 converging nozzle with a mass m ˙ = 1[kg/sec] flow rate of 1[kg/sec] and the A1 = 0.058 0. Utilizing Fliegner’s equation (4.0 × 400 × 287 A ∗ P0 = = ∼ 70061.2227 The pressure at point 1 is P1 = P 2 0.1949 P0 P1 = 5.0times0.96355 2. To obtained the Mach number.2227 A A A2 0.003 A 2 P2 = ∼ 2.1 A ∗ P0 70061. ISENTROPIC FLOW To demonstrate the usefulness of the this function consider a simple situation of the flow through a converging nozzle Example 4.

4 × 1.362 10.00 4.064 1.52828 0.225 A×P A∗ ×P0 F F∗ .115) 4.96666) ∼ 614[kN ] = 500000 × 2.5 Isothermal Table Table 4. The same is done here for the isothermal nozzle flow model. As previously.116) (4.010 2.52921 0.114) Since U 2 /RT = kM 2 and the ratio of equation (4.112) Fnet = P2 A2 P0 A ∗ k + 1 k−1 F2 F1 (1 + k ) − ∗ P2 A 2 2 F∗ F 1 × 2.1949 − 0.5 × (2.113) Utilizing the ideal gas model for density and some rearrangement results in P2 A 2 1 + F2 = F1 P1 A 1 1 + U2 2 RT U1 2 RT (4.1 k 73 4.4.014 2.2 The Impulse Function in Isothermal Nozzle Previously Impulse function was developed in the isentropic adiabatic flow.949 5.114) M1 1 + kM2 2 F2 = F1 M2 1 + kM1 2 At the star condition (M = 1) (not the minimum point) results in F2 1 1 + kM2 2 = F∗ M2 1 + k (4.5.001 2.53199 1.3: Isothermal Table M 0.4.1 T0 T0 P0 P0 A A P P 0.064 5. ISOTHERMAL TABLE The net force is obtained by utilizing equation (4.00 8.000 1.23.064 1.87) transformed equation into (4.0E +6 4.05 0. The ratio of the impulse function for two points on the nozzle is P2 A 2 + ρ 2 U 2 2 A 2 F2 = F1 P1 A 1 + ρ 1 U 1 2 A 1 (4.00 0.0E + 5 9.2E +5 20. the definition of the Impulse function is reused.

67383 0.062 1.0 7.763 1.000 6.000 3.540 4.800 1.608 3.80528 0.287 1.053 0.125 1.083 0.56232 0.0 1.000 3.00 T0 T0 P0 P0 A A P P 0.025349 0.209 0.000 8.161 2.2 0.000 7.063 1.5E + 2 0.100000 A×P A∗ ×P0 F F∗ 2.41E −6 0.55556 0.718 3.5E + 2 2.284 3.3E +23 0.21 1.26634 0.183 1.69449 2.000 1.296 5.271 0.0 5.035 3.281 1.11111 0.0 0.000 4.059 1.16090 0.065 1.153 1.7 0.4E + 3 2.429 1.237 1.61693 0.15385 0.6E + 4 2.14286 0.142 1.939 2.18182 0.41686 0.3 0.70 1.3E +11 0.0 0.863 0.0 1.4E +26 0.429 1.464 1.74 CHAPTER 4.8 0.500 7.000 1.0 1.51069 0.0E + 6 0.399 0.111 1.665 0.12500 0.586 5.0 8.29 80.85853 1.90909 0.004 0.33333 0.97376 1.3: Isothermal Table (continue) M 0.8E +18 0.40 1.90302 1.0 0.500 5.0 2.021 0.75344 1.2E + 4 1.0 0.71429 0.0 1.73278 0.56954 3.625 1.2E +29 1.000 0.000 1.889 2.4E + 3 5.0 0.98796 1.9E + 2 8.500 10.50 3.54322 0.000 3.134 9.000215 1.047 0.500 9.50000 0.99967 1.00370 0.89348 1.52632 0.22222 0.28571 0.044 1.50 1.194 1.000 9.33554 0.80844 1.73492 0.047 1.75E −5 1.000380 2.328 1.000 5.064 2.333 2.7E + 3 1.16667 0.58985 0.500 2.079 1.036 0.6E +13 0.3E +15 0.553 1.015317 7.500 8.200 1.2E + 3 3.1E + 4 1.6 The effects of Real Gases To obtained expressions for non–ideal gas it is communally done by reusing the ideal gas model and introducing a new variable which is a function of the gas prop- .134 0.0 0.3E + 9 0.86329 0.00 1.20846 0.80 1.007 5.247 0.00 2.000 0.0 0.58824 0.667 1.281 1. ISENTROPIC FLOW Table 4.690 1.4E +20 0.0 0.0 5.4 0.62665 0.958 1.76923 0.719 5.25000 0.10526 0.1E + 3 0.114 1.071758 90.599 0.97156 1.6 0.250 1.8E + 2 4.143 4.30 1.14 0.245 2.0 5.564 1.875 4.90 2.565 1.20000 0.66667 0.026 19.40000 0.000 0.00221 9.0 0.11765 0.020 1.50618 4.13333 0.97274 1.6E + 5 1.94147 1.500 6.0 5.60 1.891 1.438 2.000 1.569 3.055 1.375 1.275 1.4E + 8 0.3E + 2 1.41 40.78 0.128 1.681 1.41E −5 4.63276 2.856 4.9 1.5 0.0 0.2E + 3 7.389 1.20 1.22881 15.83333 0.12246 0.10 1.431 4.98750 0.62500 0.500 4.

121) P The mechanical energy equation can be expressed as d U2 2 =− dP ρ (4. The following power relationship is assumed ρ = ρ0 P P0 1 n (4.4.6.123) Notice.118) and second derivative is (4.19). Thus.27) is reused to related the entropy change to the change in thermodynamics properties and applied on non-ideal gas. Differentiating equation (3. h = h(T. Since ds = 0 and utilizing the equation of the state dh = dP/ρ.19) are combined to form Cp dT ds = −z 1+ R R T T Z ∂z ∂T dP P (4.122) At the stagnation the definition requires that the velocity is zero.117) Again.122) when using relationship which is defined in equation (4.27) and (3. To carry the integration of the right hand side the relationship between the pressure and the density has to be defined. that for perfect gas the n is substituted by k . The enthalpy is a function of the temperature and pressure thus. the differential of enthalpy is calculated for real gas equation of state as dh = Cp dT − T Z ∂z ∂T dP ρ (4. a real gas equation can be expressed in equation (3. With integration of equation (4.119) The definition of pressure specific heat is Cp ≡ Maxwell relation hence. ∂h ∂P =v−T ∂s ∂T T P First.123) results U2 = 2 P1 P0 dP = ρ P P0 1 ρ0 P0 P 1 n dP (4. THE EFFECTS OF REAL GASES 75 erties like the critical pressure and critical temperature.19) and dividing by equation (3.120) P Equations (4.124) .19) yields dz dρ dT dP = + + P z ρ T (4. P ) and full differential is dh = ∂h ∂T dT + P ∂h ∂P dP T ∂h ∂T (4. Gibb’s equation (4.

125) For n > 1 the integration results in 2n 1− n−1 n−1 P ( n ) P0 U= z0 RT0 (4.19) results U2 = 2 P P0 z0 RT0 P0 P0 P 1 n dP (4.132) .34) to defined the Mach number as M=√ U znRT (4.130) The Mach number can be obtained by utilizing equation (3. the mass flow rate in our properties ρ∗ U∗ P0 m ˙ =A z0 RT0 ∗ P P0 1 n 2n z0 RT0 1− n−1 n−1 P ( n ) P0 (4. The mass flow rate is regardless to equation of state as following m ˙ = ρ ∗ A∗ U ∗ (4.126) For n = 1 the integration becomes U= 2z0 RT0 ln P0 P (4.131) Integrating equation (4.76 CHAPTER 4.129) For the case of n = 1 ρ∗ U ∗∗ 1 n m ˙ = A∗ P0 z0 RT0 P P0 2z0 RT0 ln P0 P (4. ISENTROPIC FLOW Substituting relation for stagnation density (3.127) It must be noted that n is a function of the critical temperature and critical pressure.128) Where ρ∗ is the density at the throat (assuming the chocking condition) and A∗ is the cross area of the throat. Thus.121) when ds = 0 results T2 T1 Cp dT = R T P2 z 1+ P1 T Z ∂z ∂T P dP P (4.

134) It was shown in Chapter (3) that (3. 2n n−1 P P0 n−1 n U= When n = 1 or when n → 1 z0 RT0 1− (4.137) 2n n−1 2z0 RT0 ln P0 P (4.132) looks at Bernnolli’s equation which is dU 2 =− 2 After integration of the velocity dU 2 =− 2 P/P0 1 dP ρ (4.138) 1− P∗ P0 (4.140) . THE EFFECTS OF REAL GASES 77 To carryout the integration of equation (4.139) Fliegner’s number for n = 1 is Fn = mc ˙ 0 =2 A ∗ P0 P∗ P0 2 − ln P∗ P0 (4.6.135) U= 2z0 RT0 ln P0 P (4.136) The mass flow rate for the real gas m ˙ = ρ ∗ U ∗ A∗ A ∗ P0 m ˙ =√ z0 RT0 And for n = 1 A ∗ P0 m ˙ =√ z0 RT0 Fliegner’s number in this case is Fn = mc ˙ 0 A ∗ P0 2n n−1 P∗ P0 1 n 2n n−1 P∗ P0 1 n 1− P∗ P0 (4.33) is applicable for some ranges of relative temperature and pressure (relative to critical temperature and pressure and not the stagnation conditions).133) ρ0 d ρ P P0 (4.4.

132) can be integrated T0 = T P0 P R Cp ∂z )P ] [z+T ( ∂T (4.145) For n = 1 the Mach number is M= 2 z 0 T0 P 0 ln z T P (4.141) When n = 1 or more generally when n → 1 this is a ratio approach P∗ √ = e P0 (4.149) .144) The Mach number at every point at the nozzle can be expressed as M= 2 n−1 z 0 T0 1− z T P −0 P 1−n n (4.78 The critical ratio of the pressure is P∗ = P0 2 n+1 CHAPTER 4. equation (4.142) To obtain the relationship between the temperature and pressure. ISENTROPIC FLOW n n−1 (4.146) The pressure ratio at any point can be expressed as a function of the Mach number as n−1 ∂z z +T ( ∂T )P ] T0 n − 1 2 ( n )[ = 1+ M T 2 (4. Note (4.148) ∂z n z +T ( ∂T )P ] 1 + n ( 1−n )[ 2 (4.147) for n = 1 T0 = T The critical temperature is given by T∗ = T0 eM [z+T ( 2 ∂z ∂T )P ] (4.143) The power of the pressure ratio is approaching that T0 z0 = T z P0 P k−1 k 1−n n when z approaches 1.

Are the stagnation pressure and temperature at the entrance different from the point? You can assume that k = 1.42027 0.11761 2. S OLUTION 1. the stagnation pressure is constant for isentropic flow.04943 T0 300 T = ∼ 713.14366 2. Calculate the stagnation pressure and the stagnation temperature. THE EFFECTS OF REAL GASES and for n = 1 T∗ = T0 79 e−[z+T ( ∂z ∂T )P ] (4. i. .82K T 0.42027 The stagnation temperature is T0 = 3. Calculate the area ratio between the point and the throat.405. The stagnation pressure is obtained from P0 = P0 2.152) Example 4. and temperature 300K . The solution is simplified by using Potto-GDC for M = 2.61 the results are M 2. Of course. the pressure 2[Bar].61.6.9066 0.6100 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. ii. iii.150) The mass flow rate as a function of the Mach number is P0 n m ˙ = M c0 n−1 2 1+ M 2 n+1 n−1 (4.151) For the case of n = 1 the mass flow rate is P0 A ∗ n m ˙ = c0 e M2 n−1 2 M 1+ 2 n+1 n−1 (4.8: A design is required that at a specific point the Mach number should be M = 2.61 P = ∼ 52.802[Bar] P 0.4.04943 0.

ISENTROPIC FLOW .80 CHAPTER 4.

A discussion on the shock thickness will be presented in a forthcoming section1 . c.v. and the gas flows from left to right. Of course. but it can also be viewed as a one–dimensional shock wave.CHAPTER 5 Normal Shock In this chapter the relationships between the two sides of normal shock are presented. low pressure side. The conditions. the shock travels from the high pressure to the Fig. The first is when a large differ  flow direction ence (above a small minimum   value) between the two sides   !#" of a membrane.1) a control volume for this analysis is shown. and when the membrane bursts (see the discussion about the shock tube). to the left and to the right of the shock. Here. the flow is assumed to be in a steady state. without any abrupt changes. it was assumed the medium is continuous. In the speed of sound analysis. are 1 Currently under construction. In this discussion. In fact.1: A shock wave inside a tube. In Figure (5. the relationship for a perfect gas is constructed. and the thickness of the shock is assumed to be very small. 5. 81 . The second is when many sound waves “run into” each other and accumulate (some refer to it as “coalescing”) into a large difference. the sound wave can be viewed as an extremely weak shock. which is the shock wave. A shock can occur in at least two different mechanisms. This assumption is no longer valid in the case of a shock.

3).4) If the conditions upstream are known. the entropy must increase. it can be written as follows: sy − s x > 0 (5. the momentum is the quantity that remains constant because there are no external forces. This law dictates the direction of the flow.1) In a shock wave. The chemical reactions (even condensation) are neglected. Physics dictates that there is only one possible solution. NORMAL SHOCK assumed to be uniform2 . then there are four unknown conditions downstream. and therefore the energy equation can be written as C p Tx + Ux 2 Uy 2 = C p Ty + 2 2 (5. but different from the left side. The transition in the shock is abrupt and in a very narrow width. It is further assumed that there is no friction or heat loss at the shock (because the heat transfer is negligible due to the fact that it occurs on a relatively small surface). the gas flows from a supersonic flow to a subsonic flow. These two possible solutions refer to the direction of the flow. One cannot deduce the direction of the flow from the pressure on both sides of the shock wave. Nevertheless. one can note that there are two solutions because of the quadratic of equation (5.3) Px − P y = ρ x U y 2 − ρ y U x 2 (5. the increase of the entropy is fundamental to the phenomenon and the understanding of it. the second law is expressed by the entropy. The conditions to the right of the shock wave are uniform. A system of four unknowns and four equations is solvable. In mathematical terms. Clearly. . The only tool that brings us to the direction of the flow is the second law of thermodynamics. It is customary in this field to denote x as the upstream condition and y as the downstream condition. Therefore. and the shock occurs at a very narrow section. Mathematically. For the adiabatic process. the isentropic transition assumption is not appropriate in this case because the shock wave is a discontinued area. The mass flow rate is constant from the two sides of the shock and therefore the mass balance is reduced to ρ x Ux = ρ y Uy (5.2) The equation of state for perfect gas reads P = ρRT (5.82 CHAPTER 5. it can be written that The process is adiabatic. and as it will be shown. or nearly adiabatic.5) 2 Clearly the change in the shock is so significant compared to the changes in medium before and after the shock that the changes in the mediums (flow) can be considered uniform. Thus.

the intersec)&*(+ tion is shown and two solu7¥8:95. In Figure (5. continuity. B?CAD the intersection of these two curves is what really Rayleigh Fanno line line happened in the shock. and state equations. when the momentum equation is solved instead of the energy equation. and the second set is the momentum. T. U. This flow is dealt with in greater detail in Chapter (10).2). and state equations. s.38:95<68 tions are obtained.5) is ln Ty Py − (k − 1) >0 Tx Px (5.0. On the other hand. Now one can view these equations as two different subsets of equations.¥-/. continuity. This situation is similar to a frictionless flow with the addition of heat. the energy equation is used. the degree of freedom is now energy. Since the shock has $&%(' no heat transfer (a special EGFI J H K subsonic flow case of Rayleigh flow) and supersonic . the parameters are P. one can solve only four (4) equations (including the second law). and ρ. the increase of the entropy determines the direction of s the flow. Mathematically. and therefore no equality can exist.6) It can also be noticed that entropy. one can have a whole range of solutions. Instead of solving all the equations that were presented. continuity.213-/. . For ideal gas.2: The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. 5.e. which will produce a range of possible solutions. The first set is the energy. The entropy increases from point x to Fig. and state equations are solved for the arbitrary value of the Ty . The reason is that the process is irreversible. and Chapter (9) deals extensively with this topic. The solution of every set of these equations produces one additional degree of freedom. can be expressed as a function of the other parameters.. This case is called Fanno flow. It is also worth noting that the temperature at M = 1 on Rayleigh flow is larger than that on the Fanno line. If the energy. i. and this flow is known as Rayleigh flow. point y . Thus.546flow there isn’t essentially any T =?>A@ momentum transfer (a speshock jump cial case of Fanno flow). which are needed to be solved. which will require additional parameters. The mathematical explanation is given Chapter (9) in greater detail. In the first case. the energy amount “added” to the shock. producing various resistance to the flow. equation (5. Clearly.83 Note that the greater–equal signs were not used. a parabola in the T –s diagram will be obtained.

11) . The relationship between the temperature and the stagnation temperature provides the relationship of the temperature for both sides of the shock. (5. and thus for known Mx all the other quantities can be determined.1 Solution of the Governing Equations Informal Model Accepting the fact that the shock is adiabatic or nearly adiabatic requires that total energy is conserved.1 5.1. It can also be noted that in many cases the dimensionless equations set is more easily solved.8) Multiplying the two sides by the ratio of the specific heat. provides a way to obtain the speed of sound definition/equation for perfect gas. Note that the Mach number is a function of temperature.9) (5. The analytical solution is discussed in the next section.3) can be converted into a dimensionless form. as follows: Px 2 Py 2 Ux 2 = Uy 2 Tx kRTx Ty kRTy cx 2 cy 2 (5. ρ.1).10) Note that the speed of sound on the different sides of the shock is different. The reason that dimensionless forms are heavily used in this book is because by doing so it simplifies and clarifies the solution. T0 x = T0 y .1. k.1) substituting for density. at least.7) All the other relationships are essentially derived from this equation. c2 = kRT to be used for the Mach number definition. Utilizing the definition of Mach number results in Py 2 Px 2 Mx 2 = My 2 Tx Ty (5. The only issue left to derive is the relationship between Mx and My .84 CHAPTER 5. numerically.8) results in Py 2 Px 2 2 U = Uy 2 x R 2 Tx 2 R 2 Ty 2 (5. 5. From the continuity equation (5.2 Formal Model Equations (5.2). Ty = Tx Ty T0 y Tx T0 x = 1+ 1+ 2 k −1 2 Mx 2 k−1 2 My (5. and (5. NORMAL SHOCK 5. the equation of state yields Px Py Ux = Uy RTx RTy Squaring equation (5.

12) Energy equation (5.2) into Px + kPx Mx 2 = Py + kPy My 2 Rearranging equation (5.17) The relationship between Mx and My is needed to be solved from the above set of equations. This relationship can be obtained from the combination of mass. SOLUTION OF THE GOVERNING EQUATIONS Rearranging equation (5.14) transforms the momentum equation (5.5. momentum.16) (5.11)) as Py 1 + P0 y = P0 x Px 1 + 2 k−1 2 My 2 k−1 2 Mx k k−1 k k−1 (5.18) .15) yields Py 1 + kMx 2 = Px 1 + kMy 2 (5. From equation (5.1. Py M y Px M x 2 = 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5. and energy equations.15) 2  P   U  kRT = kP M 2  RT kRT  c2   (5.16) can be interpreted as the loss of the static pressure.11) results in Ty = Tx Py Px 2 85 My Mx 2 (5.13) It can also be observed that equation (5.13) (energy) and equation (5. T0 y = T0 x .3) can be converted to a dimensionless form which can be expressed as Ty 1 + k−1 My 2 2 = Tx 1 + k−1 Mx 2 2 (5. Under the perfect gas model. The loss of the total pressure ratio can be expressed by utilizing the relationship between the pressure and total pressure (see equation (4.12) (mass) the temperature ratio can be eliminated.13) means that the stagnation temperature is the same.14) The pressure ratio in equation (5. ρU 2 is identical to kP M 2 because M2 ρ ρU 2 = Using the identity (5.

Clearly. New York: Springer-Verlag.e.86 CHAPTER 5.19) Equation (5.19) is biquadratic. The second solution is where the shock wave occurs. The pressure ratio between the two sides can now be as a function of only a single Mach number.18) with equation (5. Px = Py . K.19) and divide it by My 2 − My 2 1+ k−1 My 2 + My 2 − kMy 2 My 2 = 0 2 (5.22) provides the pressure ratio as only a function of the upstream Mach number as 2k k−1 Py = Mx 2 − Px k+1 k+1 2k Py =1+ Mx 2 − 1 Px k+1 (5. .20) It can be observed that equation (5. in this case. NORMAL SHOCK Combining the results of (5. M.21) Equation (5. The solution can be obtained by rewriting equation (5. i. the pressure and the temperature from both sides of the nonexistent shock are the same. According to the Gauss Biquadratic Reciprocity Theorem this kind of equation has a real solution in a certain range3 which will be discussed later.23) The density and upstream Mach number relationship can be obtained in the same fashion to became ρy Ux (k + 1)Mx 2 = = ρx Uy 2 + (k − 1)Mx 2 (5. one solution is My = M x (5. and Rosen. Utilizing equation (5.21) becomes My 2 = Mx 2 + 2k k−1 Mx 2 k−1 2 −1 (5. It is also possible to cross–multiply equation (5. 108-137. ”Cubic and Biquadratic Reciprocity.16) and equation (5. Tx = Ty .20) is the trivial solution in which the two sides are identical and no shock wave occurs. 2nd ed.16) results in 1 + kMx 2 1 + kMy 2 2 = Mx My 2 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5. Thus.22) The first solution (5. 9 in A Classical Introduction to Modern Number Theory. Mx .” Ch. 1990.24) 3 Ireland.19) is a symmetrical equation in the sense that if My is substituted with Mx and Mx substituted with My the equation remains the same.19) as a polynomial (fourth order). pp. for example.

8 0. The Star Conditions The speed of sound at the critical condition can also be a good reference velocity. The speed of sound at that velocity is √ c∗ = kRT ∗ (5.28) Based on this definition a new Mach number can be defined M0 = U c0 (5.9 0.1 1 2 3 Fri Jun 18 15:47:34 2004 0 4 5 6 7 8 9 10 My P0y/P0x Shock Wave relationship My and P0y/P0x as a function of Mx The stagnation speed of sound is c0 = kRT0 (5. SOLUTION OF THE GOVERNING EQUATIONS 87 The fact that the pressure ratio is a function of the upstream Mach number.30) .2 0.25) In the same way.7 0. And the temperature ratio.3 0.5. as a function of pressure ratio.5 0.27) My 1 0. The Maximum Conditions The maximum speed of sound is when the highest temperature is achieved. the relationship between the density ratio and pressure ratio is k+1 1+ k −1 ρx = k+1 ρy + k−1 Py Px Py Px (5. 5. provides additional way of obtaining an additional useful relationship.1. The maximum temperature that can be achieved is the stagnation temperature Umax = 2k RT0 k−1 (5.29) Mx Fig.3: The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. Mx .6 0. is transformed into Ty = Tx Py Px k+1 k−1 Py Px k+1 Py k−1 Px + 1+ (5.4 0.26) which is associated with the shock wave.

35) (5. c2 U2 c∗ 2 c∗ 2 k + 1 ∗2 + = + = c k−1 2 k−1 2 2(k − 1) (5.3 Prandtl’s Condition It can be easily observed that the temperature from both sides of the shock wave is discontinuous.36) (5. the speed of sound is different in these adjoining mediums.33) Dividing the mass equation by the momentum equation and combining it with the perfect gas model yields c1 2 c2 2 + U1 = + U2 kU1 kU2 Combining equation (5.35) the following can be obtained: U1 U2 = c ∗ 2 or in a dimensionless form M ∗ 1 M ∗ 2 = c∗ 2 (5.1. c U c U = ∗ = ∗M c∗ c c c M∗ = (5.34) .34) results in 1 1 k + 1 ∗2 k − 1 k + 1 ∗2 k − 1 c − U1 + U 1 = c − U2 + U 2 kU1 2 2 kU2 2 2 After rearranging and diving equation (5.88 CHAPTER 5.32) The jump condition across the shock must satisfy the constant energy. an additional Mach number can be defined as M∗ = U c∗ (5.37) (5. NORMAL SHOCK In the same manner.31) 5. It is therefore convenient to define the star Mach number that will be independent of the specific Mach number (independent of the temperature).33) and (5. Therefore.

7245 My 0.0 30.5[bar] and Tx = 273K . the known information are Mx = 3. the Mach number after the shock. at a pressure of 0.0 20. px = 1.3088 0. This kind of equations requires examining Table (5. are plotted as a function of the entrance Mach number. 120.0 60. and velocity downstream of the shock. P0x is known). Finding the Mach number for a pressure ratio of 8. pressure. Ty Tx ρy ρx Py Px P0y P0 x Mx 2.47642 2. the total pressure can be obtained (through an isentropic relationship in Table (4.0 70.3).5 [bar] and a temperature of 0◦ C goes through a normal shock.0 90. the velocity can readily be calculated. P = 0. total pressure.e.2).38109 To illustrate the use of the above equations.0272237 = 55. Tx .5/0.32 is only a few mouse clicks away from the following table. The relationship that was calculated will be utilized to obtain the ratios for the downstream of the normal Px shock. The equations should be used with the Mach number (upstream or downstream) for a given pressure ratio or density ratio (velocity ratio).0272237 =⇒ P0x = 1.9360 8.5. P0y /P0x .0 100.1110 3.0 0. ρy/ρx and Ty/Tx as a function of Mx Shock Wave relationship Py/Px ρy/ρx Ty/Tx S OLUTION Fig.4 × 287 × 273 = 331.30879 and k = 1.0 50. Using these data. First.0 80. an example is provided. 5.1[bar] √ √ 0x cx = kRTx = 1. Also with the temperature. My .0 10.4 or utilizing Potto-GDC for for value of the specific heat ratio.4: The ratios of the static properties of the two sides Analysis: of the shock. Calculate the temperature.0 110. The typical situations in which these equations can be used also include the moving shocks.0 Py/Py.0 40.2m/sec 1 2 3 Fri Jun 18 15:48:25 2004 4 5 Mx 6 7 8 9 10 . i.2. Example 5. Note that the My has a minimum value which depends on the specific heat ratio.1: Air flows with a Mach number of Mx = 3. The working equations were presented earlier. It can be noticed that the density ratio (velocity ratio) also has a finite value regardless of the upstream Mach number.. and the ratio of the total pressure. OPERATING EQUATIONS AND ANALYSIS 89 5.2 Operating Equations and Analysis In Figure (5.1) for k = 1.

47519 Ty Tx ρy ρx CHAPTER 5.6790 3. The limits of the pressure ratio can be obtained by looking at equation (5.85714 = 257.22)) is limited by My = 2 :  ∼0 2  1 + (k −1)Mx 2 2k k−1 >   − M1 x 2 ∼0 = k−1 2k (5.38).3333 0. This approach had a major significance and usefulness at a time when personal computers were not available. NORMAL SHOCK Py Px P0 y P0 x 2.39) into ˆ= P 2k Mx 2 − 1 k+1 (5.23) transforms equation (5.2 = 993. Uy = 993. Now.85714.6/3.32834 Ux = Mx × cx = 3 × 331.32834 × 55.5 series).38) This result is shown in Figure (5.09[bar] 5. during the writing of this version of the book.6[m/sec] P0y = P0y P0x × P0x = 0.3). this technique is used mostly in obtaining analytical expressions for simplified models.6[m/sec] Now the velocity downstream is determined by the inverse ratio of ρy /ρx = Ux /Uy = 3. the downstream Mach number (see equation (5.39) .0000 My 0.40) (5.1 The Limitations of the Shock Wave When the upstream Mach number becomes very large. it refers to a case where only a “small shock” occurs.3. which is up to Mx = 1.2 Small Perturbation Solution The small perturbation solution refers to an analytical solution where only a small change (or several small changes) occurs.90 Mx 3. The strength of the shock wave is defined as ˆ = Py − P x = Py − 1 P Px Px By using equation (5.2.16) and by utilizing the limit that was obtained in equation (5.8571 10. In this case.2.1[bar] = 18. 5. This technique also has an academic value and therefore will be described in the next version (0.

when a valve is suddenly hji NPO  S T f d e b ` . THE MOVING SHOCKS or by utilizing equation (5. flow L5M g direction Q5R For example.3.3 Shock Thickness The issue of shock thickness (which will be presented in a later version) is presented here for completeness.24) the following is obtained: ˆ= P 2k k−1 2 k−1 ρy ρx 91 −1 −1 − ρy ρx (5. The most common definition is by passing a tangent to the velocity at the center and finding out where the theoretical upstream and downstream conditions are meet. The shock thickness can be defined in several ways. to convince the students that indeed the assumption of very thin shock is validated by analytical and experimental studies.3 The Moving Shocks In some situations. the issue should be presented. however. This issue has a very limited practical application for most students.2. 5. the shock wave is not stationary.41) 5. This kind of situation []\ ^]_ arises in many industrial applications.5.

which this section will be dealing with.5: Comparison between stationary shock and moving shock in ducts erties. but this book will not deal with 4 It will be explained using dimensional analysis what is suddenly open . a c 4 closed and a shock propagates upUWV XZY stream. the shock can move downstream or upstream. In some industrial applications. a liquid (metal) is pushed in two rapid stages „:…{†p‡ˆ…s‰‹Š Œ to a cavity through a pipe system. qsrutvqpw ‚bƒ As a general rule.v. The last situation is the most general case.Fig. propagates downstream (the opposite Stationary Coordinates direction of the previous case).v. a shock occurs and c. Moving Coordinates There are more genera cases where the moving shock is created which include a change in the physical prop. On the other extreme. This x:y{zp|}y~¥€  liquid (metal) is pushing gas (mostly) Žˆ k#lnmpo air. c. 5. which creates two shock stages. when a valve is suddenly opened or a membrane is ruptured.

the prime ’ denote the values of the static coordinates. to the stationary case when the coordinates are attached to the shock front. i. In such a case.5)) the velocity measured by the observer is Where Us is the shock velocity which is moving to the right. For this analysis. the coordinates move with the shock. and the tools developed in this chapter can be employed. An observer moving with the shock will notice that the pressure in the shock is Px = P x Py = P y (5.45) Ux = U s − U x (5.46) cx cx It can be noted that the additional definition was introduced for the shock upstream s Mach number. The reluctance to deal with the most general case is due to fact it is highly specialized and complicated even beyond early graduate students level. the steady state is obtained in the moving control value. Msx = U cx . Note that this notation is contrary to the conventional notation found in the literature. equations. The downstream prime Mach number can be expressed as Us Us − U y = − My = Msy − My (5. In cases where the shock velocity can be approximated as a constant (in the majority of cases) or as near constant.42) (5. The reason for the deviation is that this choice reduces the programing work (especially for object–oriented programing like C++).47) My = cy cy Uy = U s − U y (5.43) The temperature measured by the observer is Tx = T x Ty = T y Assuming that the shock is moving to the right. In these changes (of opening a valve and closing a valve on the other side) create situations in which different shocks are moving in the tube.e. The general case is where two shocks collide into one shock and moves upstream or downstream is the general case. the previous analysis. (refer to Figure (5. The upstream prime Mach number can be defined as Us − U x Us Mx = = − Mx = Msx − Mx (5.92 CHAPTER 5. Here. The “downstream” velocity is The speed of sound on both sides of the shock depends only on the temperature and it is assumed to be constant. A specific example is common in die–casting: after the first shock moves a second shock is created in which its velocity is dictated by the upstream and downstream velocities.44) .. NORMAL SHOCK them at this stage. The problem can be reduced to the previously studied shock.

T0y − T0x =Ty 1 + k−1 Msy − My 2 T0y 2 − Tx 1 + k−1 Msx − Mx 2 2 0 = Ty 1 + k−1 My 2 T0x 2 +Ty Msy k−1 (Msy − 2My ) 2 k−1 (Msx − 2Mx ) 2 (5. . This difference can also be viewed as a result of the unsteady state of the shock. this difference in the moving shock is expected because moving material velocity (kinetic energy) is converted into internal energy.53) Again. an additional definition was introduced for the shock downstream Mach number.52) − Tx 1 + k−1 Mx 2 2 −Tx Msx and according to equation (5. THE MOVING SHOCKS 93 Similar to the previous case. at a very high velocity.3. The relationship between the two new shock Mach numbers is cy Us Us = cx cx cy Msx = Ty Msy Tx (5.v. and is “converted” or added into the total temperature. The difference between the stagnation temperature is in the moving coordinates T0y − T0x = 0 (5.51) It should be noted that the stagnation temperature (in the stationary coordinates) rises as opposed to the stationary normal shock. The rise in the total temperature is due to the fact that a new material has entered the c.50) The same can be said for the “downstream” side of the shock.48) The “upstream” stagnation temperature of the fluid is T0x = Tx 1 + k−1 Mx 2 2 (5.49) and the “upstream” prime stagnation pressure is P0x = Px k−1 Mx 2 1+ 2 k k−1 (5. Msy .5.51) leads to T0y − T0x = Us Ty k − 1 Tx k − 1 (Msx − 2Mx ) − (Msy − 2My ) cx 2 cy 2 (5.

This situation arises in many cases in the industry. In the moving coordinates.v. Fig. the shock velocity “downstream” is Us . for example. In many such cases the steady state is established quite rapidly. and for a given specific heat ratio.3. The upstream is on the left of the shock. Ux . 5. NORMAL SHOCK 5.94 CHAPTER 5. The flow is moving to the right. It turns out that it is very useful to use the dimensionless parameter Msx .56) The prominent question in this situation is what will be the shock wave velocity for a given fluid velocity. Equations (5. the observer (with the shock) sees the flow moving from the left to the right. .v. In such a case. ­W®°¯ Stationary Coordinates À2ÁW›ÃÄÁWÅ:Æ Ç ³3´¶µZ· ¾›¿ ¼W½ ¸ p ¹ º¥¸#» Moving Coordinates c.42) to (5.53) can be transformed into simpler equations when Mx is zero and Us is a positive value. in a sudden and complete closing of a valve. The relationship between the Mach number on the two sides of the shock are tied ¡£¢j¤¥¡ ¦ •£– ˜ —# œ5 § ¨/©«ª§¬ c.6: Comparison between a stationary shock and a moving shock in a stationary medium in ducts.1 Shock Result from a Sudden and Complete Stop The general discussion can be simplified in the extreme case when the shock is moving from a still medium. The difficulty lies in the jump from the stationary coordinates to the moving coordinates.55) Again. The “upstream” or the “downstream” Mach number is not known even if the pressure and the temperature downstream are given. The “upstream” Mach number reads Mx = Us + U x = Msx + Mx cx (5. The sudden closing of the valve must result in a zero velocity of the gas.54) The “downstream” Mach number reads My = |Us | = Msy cy (5. the shock is moving to the left. This shock is viewed by some as a reflective shock. The information propagates upstream in which the gas velocity is converted into temperature. or Msy instead of the velocity because it combines the temperature and the velocity into one parameter. The stagnation temperature increases by T0y − T0x = Us Tx k − 1 Ty k − 1 (Msx + 2Mx ) − (Msy ) cx 2 cy 2 ž Ÿ ‘3’ “ ” ™›š ±W² (5.

which can easily be obtained in a Fanno flow. Two numerical methods and the algorithm employed to solve this problem for given.58) The temperature ratio in equation Shock in A Suddenly Close Valve (5.7).1 1 Mx the “smart” guessing of M sx.58) but it seems utilizing numerical methods is much more simple.5. an analytical solution can be obtained for equation (5.3.5) as shown in Figure (5.58) pro1 vides that Msx ∼ 1 + 2 and Msy = Fig. engineers who are dealing with a supersonic flow when installing the nozzle and pipe systems for gaseous mediums understand the importance of the reflective shock wave.7: The moving shock Mach numbers as a 1 result of a sudden and complete stop.54) and (5.58) can be obtained by several numerical methods.57) into (5. Only 2 one real solution is possible. Note that to achieve supersonic velocity (in stationary coordinates) a diverging– converging nozzle is required. 1 Note.57) And substituting equation (5. a Mach number close to one (1). is provided herein: (a) Guess Mx > 1. For very small values of the upstream Mach Thu Aug 3 18:54:21 2006 number. Here no such device is needed! Luckily and hopefully. For example. the result is about double the sonic velocity of the reflective shock. 5. Mx ∼ equation (5. The solution to equation (5. 1 − 2 (the coefficient is only approximated as 0. . From the same figure it can also be observed that a high velocity can result in a much larger velocity for the reflective shock.58) and the rest of the right–hand k=14 3 side show clearly that Msx has four Msx possible solutions (fourth–order polyMsy nomial Msx has four solutions). THE MOVING SHOCKS through equations (5.55) by 2 95 ( My ) = Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. Mx . Sometimes this phenomenon can have a tremendous significance in industrial applications. The typical method is 0 0.48) results in f (Msx ) Mx = Tx Ty Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5.

(a) Stationary coordinates (b) Moving coordinates Fig. Using a coordinates system which moves with the shock results in a stationary shock and the flow is moving to the left see Figure (5.8(a)). è0éWêÄè ë›ì è0é6í á0âã©á ä Upstream ÝÊÞ&ßÍà åçæ c.v.” which has better convergence to the solution initially in most ranges but less effective for higher accuracies.8: A shock moves into a still medium as a result of a sudden and complete opening of a valve . the shock moves in a constant velocity (5. 5. (b) using the shock table or use Potto–GDC to calculate the temperature ratio and My . With the exception of close proximity to the valve. (c) Calculate the Mx = Mx − Tx Ty M y (d) Compare to the calculated Mx to the given Mx . and adjust the new guess Mx > 1 accordingly. NORMAL SHOCK (b) Using shock table or use Potto–GDC to calculate temperature ratio and My .2 Moving Shock into Stationary Medium (Suddenly Open Valve) General Velocities Issues When a valve or membrane is suddenly opened.v. (a) Guess Mx = 1 + Mx .8(b))).3. 5. (c) calculate the Mx = Mx − Tx Ty M y (d) compare the new Mx approach the old Mx . a shock is created and propagates downstream.96 CHAPTER 5. The second method is “successive substitutions. ×0ØÚÙ Î0Ï:Ð¥ÑÓÒÕÔ ÈÊÉ&ËÍÌ Û#Ü Ö c. if not satisfactory use the new Mx to calculate Mx = 1 + Mx then return to part (b).8(b)). The “upstream” will be on the right (see Figure (5.

5. It has to be noted that the “upstream” and “downstream” are the reverse from the previous case.5 0 Number of Iteration 10 0 Shock in A Suddenly Open Valve k = 1 4.3 (b) My = 1.5. The same question that was prominent in the previous case appears now.61) A similar rearrangement to the previous case results in T0 y − T 0 x = T y 1 + k−1 −2Msy My + My 2 2 2 (5.59) Note that in this case the stagnation temperature in stationary coordinates changes (as in the previous case) whereas the thermal energy (due to pressure difference) is converted into velocity.60) Us = Msx cx (5.3 Mx My 1. My’ = 0.5 1 1 0. The stagnation temperature (of moving coordinates) is T0 y − T 0 x = T y 1 + k−1 2 (Msy − My ) 2 − Tx 1 + k−1 2 ( Mx ) 2 =0 (5. THE MOVING SHOCKS 97 Similar definitions of the right side and the left side of the shock Mach numbers can be utilized. The “upstream” Mach number is Mx = The “downstream” Mach number is My = Us − U y = Msy − My cy (5.3 Fig. My’ = 1.9: The number of iterations to achieve convergence.5 3 2. what will be the shock velocity for a given upstream Mach number? Again.5 1. the relationship .3.3 Mx My Ty/Tx Ty/Tx 0.25 2 1.75 k = 1 4.5 4 3.62) Shock in A Suddenly Open Valve 1.75 5 10 Number of Iteration 15 20 Wed Aug 23 17:20:59 2006 Wed Aug 23 17:46:15 2006 (a) My = 0.

Piston Velocity When a piston is moving.65) . and.3 and My = 1.64) is a quadratic equation for Msx .24) that relate the shock velocities and Shock Mach number as Msx (k + 1)Msx 2 Ux = = U Uy 2 + (k − 1)Msx 2 Msx − cy x (5. other conditions ahead of the shock. The results show that the convergence occurs very rapidly (see Figure (5. In this case. If it is satisfactory. The unknown data are the piston velocity. the temperature. To illustrate the convergence of the procedure. The other two solutions are Msx = (k + 1)Uy ± Uy (1 + k ) 4 cx 2 + 16cx 2 (5. consider the case of My = 0. It is common practice to solve this equation by numerical methods. for most practical purposes.63) can be solved. no Mach number is given but pieces of information on both sides of the shock.9)). One such methods is “successive substitutions. the larger number of the iterations required to achieve the same accuracy.63) Since Msx can be represented by Msy theoretically equation (5. stop or return to stage (b). There are three solutions of which the first one is Msx = 0 and this is immediately disregarded. the calculations for Us can be obtained from equation (5. sufficient results can be achieved after 3-4 iterations. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. The larger the value of My . (d) Check the new and improved Mx against the old one.98 between the two sides is Msy = My + CHAPTER 5. it creates a shock that moves at a speed greater than that of the piston itself. (c) Utilizing Mx = calculate the new “improved” Mx . NORMAL SHOCK (Msx ) + 2k k−1 2 (Msx ) − 1 2 k−1 2 (5.0.64) Ty My + M y Tx Equation (5.” This method is applied by the following algorithm: (a) Assume that Mx = 1.3. Therefore. Yet.

25 discussed earlier in a nozzle 2 flow and in other pipe flow mod1. such as the internal combustion engines and die casting. look at the normal Fig. shock. is there a point where the moving shock is fast enough to reduce the “upstream” relative Mach number below the speed of sound? This is the point where regardless of the pressure difference is. Mx = Msx → ∞.25 has no limit. Some use equation (5. This phenomenon occurs in many industrial applications.68) to explain the next Shock-Choke phenomenon. k ing of valve results in no limit (at least from the model point Thu Aug 24 17:46:07 2006 of view). It must be noted 1 that in the previous case of 0. THE MOVING SHOCKS 99 The negative sign provides a negative value which is disregarded.10: The maximum of “downstream” Mach number as a function of the specific heat.75 suddenly and completely clos0. The limit of the equation when cx → ∞ leads to Msx = (k + 1)Myx 4 (5.67) Where the “stange” Mach number is Myx = Uy /cx . This shock–choking phenomenon Shock in A Suddenly Open Valve Maximum M ’ possible is somewhat similar to the 2. the shock Mach number cannot be increased. and the downstream Mach y y(max) Maximum My’ .5 ence is that the actual velocity 1.5 The spesific heat ratio. and the only solution left is Msx = (k + 1)Uy + Uy (1 + k ) 4 cx 2 + 16cx 2 (5.75 els (later chapters). k. To explain this phenomenon.66) or in a dimensionless form Msx = (k + 1)Myx + Myx (1 + k ) 4 2 + 16 (5.68) As one additional “strange” it can be seen that the shock is close to the piston when the gas ahead of the piston is very hot. Consider when the “upstream” Mach approaches infinity. The differ1. what is the maximum velocity that can be reached before this model fails? In other words.5 M choking phenomenon that was 2.5. 5.3. Shock–Choke Phenomenon Assuming that the gas velocity is supersonic (in stationary coordinates) before the shock moves.

72) + When the pressure ratio is approaching infinity (extremely strong pressure ratio).100 CHAPTER 5. These limits determine the maximum velocity after the shock since Umax = cy My .23) to this situation the following is obtained: Msx = k+1 2k Px −1 +1 Py (5. the Mach number is ∞    Ty k − 1 =   T 2k   x (5. NORMAL SHOCK number. is approaching to (k − 1)/2k . The Mach number is reduced to the maximum value very rapidly. the results is My = 2 k (k − 1) (5.38). By applying equation when the pressure ratio is approaching infinity Px (5. A similar phenomenon to the choking in the nozzle and later in an internal pipe flow is obtained. according to equation (5. The reduction occurs by an increase of temperature after the shock or a stationary shock occurs as it will be shown in chapters on internal flow.73) What happens when a gas with a Mach number larger than the maximum Mach number possible is flowing in the tube? Obviously.25) into equation (5. the Mach number is approaching infinity because of the temperature ratio but the velocity is finite.70) and the mass conservation leads to Uy ρ y = U s ρ x Us − U y My = Ty Tx ρy = U s ρx ρx ρy Msx (5.69) Mx = Msx Thus.26) and (5. From the upstream side. To understand this limit.71) 1− Substituting equations (5. the semi steady state described by the moving shock cannot be sustained. .71) results in 1 = k Py 1− Px 2k k+1 Py k−1 Px + k+1 1+ × My k+1 k−1 Py Px Py Px k+1 k−1 (5. consider that the maximum Mach number is obtained Py → ∞. One can view this as the source of the shock–choking phenomenon.

25 985.80 1.85 Table of maximum values of the shock-choking phenomenon.64 255883.10 2.6330 1.87039 0.54 750.1785 1.30 2.09 833.13).4434 1.53161 0.00000 0.43301 0.60 1. with the exception of Ty the mass flow rate is approaching infinity when the pressure ratio is approaching infinity.40825 0.05 226871.00 729. the shock–choke phenomenon means that the Mach number is only limited in stationary coordinates but the actual flow rate isn’t.90 2.79 250000.37797 0.75) is constant.81786 0.81 My 0.52223 0.78 261117.85 922.73029 Ty Tx 101 169842.70 1.40 2.3 Partially Open Valve The previous case is a special case of the moving shock.20 2.40 1.74) and equation (5. is cy ρy m ˙ = U y ρ y = M y cy ρ y = M y A √ My kPy = RTy kRTy Py RTy (5.50 Mx 1073.45374 0. Thus.61 801. The general case is when one gas flows into another gas with a given velocity.02 773.3. ∞.50000 0. .50 1.56 711.36 270031.93 243332.62 695. The only limitation is that the “downstream’ gas velocity is higher than the “upstream” gas velocity as shown in Figure (5.74 681.48667 0.56 668.09 265805.75) Since the right hand side of equation (5.29 188982.23 873.51177 0. The mass flow rate when the pressure ratio is approaching infinity.25) can be transferred for large pressure ratios into m ˙ ∼ A Ty Px k − 1 Tx k + 1 (5.44 273861.30 1. THE MOVING SHOCKS k 1. 5.77151 0.5.33968 0.86 216507.96 204124.54772 My 2.47141 0.2645 1.54006 0.2964 1.0815 1.74) Equation (5.93048 0.00 2.99 235702.3.8898 1.

Ty (Msy ) − Mx Tx . If it is satisfactory.77) (5. (c) Calculate the “downstream” shock Mach number Msy = My + My (d) Utilizing Mx = calculate the new “improved” Mx (e) Check the new and improved Mx against the old one.v. 5. NORMAL SHOCK ÿ¡ £¢Äÿ¥¤§¦ ÿ¡ ©¨ Uy > U x ô0õÚö Ux ÷#ø c. The relationship between the different Mach numbers on the “upstream” side is Mx = Msx − Mx The relationship between the different Mach on the “downstream” side is My = Msy − My (5.76) An additional parameter has be supplied to solve the problem. The procedure for the calculations can be done by the following algorithm: (a) Assume that Mx = Mx + 1. It has to be mentioned that the temperature “downstream” is unknown (the flow of the gas with the higher velocity). (a) Stationary coordinates (b) Moving coordinates Fig. A common problem is to find the moving shock velocity when the velocity “downstream” or the pressure is suddenly increased. stop or return to stage (b).102 CHAPTER 5.v. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. ò îÊï&ðÍñ ó Ux = Us − Ux Upstream ùÊú&ûÍü ýçþ c.11: A shock moves into a moving medium as a result of a sudden and complete open valve.

6 sults show that the shock choking phe0.5.2 2 2. This phe0.8 1. Thu Oct 19 10:34:19 2006 and for various static “upstream” Mach numbers. 5.3 exhibits some “downstream” Mach num0.8 1.4 Partially Closed Valve .0 0.6 M’ bers for various static Mach numbers. it was shown that the shock chokk=14 ing phenomenon occurs when the flow is 1 running into a still medium.7 The mathematics is cumbersome but re0.4 limited.12) 0.12: The results of the partial opening that the maximum can also occurs in the of the valve.2 a faster flow is running into a slower fluid. My .5 nomenon is still there (the Mach number is 0. vicinity of the previous value (see following question/example). M ’ = 0.3.8 M ’ = 0. Mx . x x x My y 5.4 2.3.9 nomenon also occurs in the case where M ’ = 0. Figure (5.9 0. not the actual flow).4 0. THE MOVING SHOCKS 103 Shock in A Suddenly Open Valve Earlier. The figure demonstrates Fig.

(a) Stationary coordinates (b) Moving coordinates Fig.13: A shock as a result of a sudden and partially a valve closing or a narrowing the passage to the flow The totally closed valve is a special case of a partially closed valve in which there is a sudden change and the resistance increases in the pipe.     Ux Uy Ux = Us + Ux Upstream Uy = Us + Uy ρ y Py Ty c.78) Ux = U s + U x Uy = U s + U y (5.79) .v.v. Similar equations can be written: (5. 5. The information propagates upstream in the same way as before. c.

3. Calculate the Mach number My by utilizing the tables or Potto–GDC (c) Calculate the “downstream” shock Mach number Msy = My − My (d) Utilizing Mx = calculate the new “improved” Mx (e) Check the new and improved Mx against the old one. Compute the pressure and the temperature behind the shock.3 × 287 × 300 By utilizing Potto–GDC or Table (5.3.63955 .104 CHAPTER 5. it is the case of a shock moving into still medium (suddenly opened valve case).54220 0.604 4.2: A shock is moving at a speed of 450 [m/sec] in a stagnated gas at pressure of 1 [Bar] and temperature of 27◦ C. (b) .4) one can obtain the following table: Mx 2.296 1. Ty (Msy ) + Mx Tx 5.132 Ty Tx Py Px P0 y P0 x 0.80) My = M s + M y (5. If it is satisfactory.1206 My Mx My 1.0 1. Assume the specific heat ratio is 1. First. S OLUTION It can be observed that the gas behind the shock is moving while the gas ahead of the shock is still.5 Worked–out Examples for Shock Dynamics Example 5. My = √ U kRT =√ 450 ∼ 1. stop or return to stage (b).953 0. NORMAL SHOCK Mx = M s + M x (5.81) For given static Mach numbers the procedure for the calculation is as follows: (a) Assume that Mx = Mx + 1. the Mach velocity ahead of the shock has to calculated. Thus.

4.53487 0.045 2. The specific heat ratio can be assumed to be k = 1. S OLUTION The first thing that needs to be done is to find the prime Mach number Mx = 1.0 Ty Tx Py Px P0 y P0 x 0.710 0.3 × 287 × 300 ∼ 378.72 sec Example 5. find the upstream prime Mach. The gas is brought into a complete stop by a sudden closing a valve.0 0. At this stage the reflecting shock velocity is unknown.56995 1.953 × 1.70009 If you insist on doing the steps yourself.2) you can find the proper Mx .722 1. it can be done for the pressure ratio as following Py = P y = Py Px = 4.56995 0. the prime properties can be found.9432 1.042 2. Calculate the velocity and the pressure behind the reflecting shock.56995 Ty Tx My 0.2K Tx 105 In same manner.57040 0.0 0. If this detail is not sufficient then simply utilize the iterations procedure described earlier and obtain the following: i 0 1 2 3 4 Mx 2. the temperature behind the shock is Ty = T y = Ty Tx = 1.2961. Mx to be 1.0 1. .5.0 0.044 My 0.3. Simply using the Potto–GDC provides for the temperature and velocity the following table: Mx 2.604 × 300 ∼ 481. THE MOVING SHOCKS Using the above table. The static pressure at the tube is 2Bar and the (static) temperature of 300K .0 0.953[Bar] Px The velocity behind the shock wave is obtained √ m Uy = Mx cx = 1.2961 2.724 The table was obtained by utilizing Potto–GDC with the iteration request. Then using Table (5.724 1.044 2.724 4.0 ∼ 4.3: Gas flows in a tube with a velocity of 450[m/sec].0445 My Mx My 0.2961 1. Then.56994 0.2961.132 × 1.724 1.

Potto-GDC provides for this temperature ratio the following table: Mx 2.3574 My Mx My Ty Tx Py Px P0 y P0 x 0.40000 0.000 6. Hint.0 2.52778 Ty Tx ρy ρx Py Px P0 y P0 x 2. for those who like an additional step) in order to double the temperature when the valve is suddenly and totally closed? S OLUTION The ratio can be obtained from Table (5.1220 My Mx My Ty Tx Py Px P0 y P0 x 0.0789 1.78928 0.55830 Example 5.317 0.3166 0.0000 3.3020 0. S OLUTION Refer to section (5. NORMAL SHOCK Example 5.55832 using the required Mx = 2.52778 0.4.3). this is the case of a partially closed valve case in which the ratio of the prime Mach number is half (the new parameter that is added in the general case).3574 in the moving shock table provides Mx 2. It can also be obtained from the stationary normal shock wave table.3574 My 0.5: A gas is flowing in a pipe with a Mach number of 0. Calculate the speed of the shock when a valve is closed in such a way that the Mach number is reduced by half.3.20000 1.89509 0. Potto-GDC provides the solution of the above data Mx 1.99813 If the information about the iterations is needed please refer to the following table. .106 CHAPTER 5.4) for the calculation procedure.4: What should be the prime Mach number (or the combination of the velocity with the temperature.1583 6.

3022 1.1216 1.20000 1.0789 1.1220 My 0.20000 0.20000 0.0545 1.89354 0.3020 1.3020 1.1219 1.89510 0.1967 1. 5.1220 1. S OLUTION Using the procedure described in this section.2922 1. Assume that there is no friction and the Fanno flow model is not applicable.20000 0.1099 1.20000 0.0930 1.89509 0.90416 0.20000 0.20000 0.0802 1.20000 0.3025 1.20000 0.3609 1.3199 1.20000 0.5. end of the tube of 1.89494 0.3019 1.1226 1.20000 0.4 My = 0.1220 1.89536 0.20000 0.89509 0.0106 1.8 accelerated very rapidly and the air adjoined the piston obtains Mach number M = 0.1221 1.0793 1.0787 1.3020 1.87903 0.8.20000 0.1220 1.1288 1.3020 Example 5.89009 0.0m length.73971 0.3017 1.1443 1.1200 1.2989 1.14: Schematic of a piston pushing air in a it takes for the shock to reach the tube.1182 1.3020 1.0030 1.1220 1.0789 1.0712 1.1220 1.89789 0.89504 0. Calculate the velocity of the shock created by the piston in the air.0789 1.3011 1.4 and 300◦C.1220 1.0765 1.20000 0.89512 0.3020 1.0789 2.0789 1. THE MOVING SHOCKS i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 Mx 1.2705 1. The piston is Mx = 0.20000 0.89509 Ty Tx Py Px 107 My 0.1259 1. Calculate the time Fig.3037 1.99548 0.20000 0.89509 0.0789 1.0832 1.1208 1.0789 1.89517 0.20000 0.0836 1.92479 0.1220 1.0790 1.89595 0.20000 0.84424 0.4000 1.0045 1.20000 0.0789 1.0789 1.6: A piston is pushing air that flows in a tube with a Mach number of M = 0.3020 1. the solution is .20000 0.89508 0.89461 0.2032 1.2547 1.3075 1.1222 1.20000 0.89509 0.20000 0.1241 1.5041 1.3.20000 0.0788 1.0782 1.

82716 0.80000 0.0034[sec] 1.4 × 287 × 300 0. S OLUTION The stationary difference between the two sides of the shock is: ∆U =Uy − Ux = cy Uy − cx Ux   √   1.80000 0.2400 1.81829 0.2380 My Mx My CHAPTER 5.6215 0.1519 1.8 × 1.2378 1.50000 0.1519 2.80000 0.70109 0.1517 1.2381 1.1435 1.80000 1.81942 Ty Tx Py Px My 0.5834 1.1519 −0.80000 0.81943 0.6215 1. i 0 1 2 3 4 5 6 Mx 1.6216 The time it takes for the shock to reach the end of the cylinder is t= length Us cx (Mx −Mx ) =√ 1 = 0.4 × 287 × 300(1.4) Example 5. NORMAL SHOCK Ty Tx Py Px P0 y P0 x 0.81958 0.7: From the previous example (5.108 Mx 1.81942 0.2380 − 0.5    q Ty Tx =  ∼ 124.2380 1.1519 1.80000 0.4583 1.2380 My 0.81940 0.6217 1.1531 1.80000 0.6273 1.98860 The complete iteration is provided below.80000 1.6207 1.10) calculate the velocity difference between initial piston velocity and final piston velocity.3202 1.2248 1.4[m/sec] .1519 1.5000 1.

When the diaphragm is ruptured the gas from the high pressure section flows into the low pressure section.99785 70. For this case the following table can easily be obtain from Potto–GDC for the left piston Mx 1. The air between the pistons is at 1[Bar] and 300K .93451 Ty Tx Py Px P0y P0 x Uy cx 347.5. Other situations can be examined and extended from these phenomena.99959 40.4 Shock Tube The shock tube is a study tool with very little practical purposes. The temperature is known to reach several thousands degrees in a very brief period of time. A cylinder with two chambers connected by a diaphragm. It is used in many cases to understand certain phenomena.0 My 0. while the pressure on the other side is low.66). The solution is given by equation (5.318 0. and. expansion waves occur with a reduction of pressure. When the pressure is high enough.0715 My 0.95890 Ty Tx Py Px P0y P0 x Fig. 1. The high pressure .0 while the velocity of the right piston is Mx 1.0 The time for the shocks to collide is t= 1[m] length = ∼ 0.4.1283)347. it is the explicit analytical solution. The distance between the two pistons is 1[m]. a shock is created that it travels to the low pressure chamber. This is the same case as in the suddenly opened valve case described previously.89048 Mx 0.1283 My 0.0013[sec] Usx 1 + Usx 2 (1. SHOCK TUBE Example 5. At the back of the shock. Calculate the time it will take for the two shocks to collide.047 1. 1.8) Uy cx 347.93471 Mx 0.8: An engine is designed so that two pistons are moving toward each other (see Figure (5.0715 + 1. 109 1 [Bar] 300 K 40 m/sec shock waves 70 m/sec S OLUTION This situation is an open valve case where the prime information is given.083 1.0 My 0. 5.15: Figure for Example (5.15)). 5. On one side the pressure is high.173 0.

Zone 4 is expanding during the initial stage (until the expansion reaches the wall).e. the driver section. In this analysis.16: The shock tube schematic with a pressure ”diagram. In fact. Zone 4 is where the gradual transition occurs between original high pressure to low pressure. Diaphragm The relationship bet tween zone 1 and zone 2 is that of a moving reflective shock into still medium some where shock reflective wave wave (again.16)). The shock tube is a relatively small length 1 − 2[m] and the typical velocity is in the expansion front Co nt ac t Su rf ac e back fr on t . material in the expansion chamber and the front are moving to the left while the actual flow of the gas is moving to the right (refer to Figure (5. The assumption is that the shock is very sharp with zero width.. the velocity is zero and the pressure is in its original value. In zone 3 is the original material that was in the high pressure chamber but is now the same pressure as zone 2. In zone 5. 5. Initially. this is a case of sudden opened valve). i. The properties in the 5 1 4 3 2 different zones have different relationships. the expansion waves are moving into the high pressure chamber i. The boundaries of zone 4 are defined by initial conditions. The expansion front is moving at the local speed of sound in the high pressure section. The shock is moving at a supersonic speed (it depends on the definition. the temperature and their properties in zone 4 aren’t constant and continuous between the conditions in zone 3 to the conditions in zone 5. The expansion back front is moving at the local speed of sound velocity but the actual gas is moving in the opposite direction in U2 . The velocities in the expansion chamber vary between three zones. On the other side. while the distance pressure in the two zones are the same. The material in zone t1 2 and 3 is moving e wav ck at the same velocity sho (speed) but the temperature and the entropy are different.” The pressure.e. it is assumed that this time is essentially zero.110 CHAPTER 5. the gas from the driver section is coalescing from small shock waves into a large shock wave. Fig. which can be supersonic or subsonic in stationary coordinates. NORMAL SHOCK chamber is referred to in the literature is the driver section and the low section is referred to as the expansion section. what reference temperature is being used) and the medium behind the shock is also moving but at a velocity. U2 . Zone 1 is an undisturbed gas and zone 2 is an area where the shock already passed. The expansion front wave velocity is larger than the velocity at the back front expansion wave velocity.

83) And the velocity ratio between the two sides of the shock is k+1 P2 1+ k ρ2 U1 −1 P1 = = k+1 P2 U2 ρ2 k−1 P 1 (5. it follows that U2 ρ1 = Us ρ2 U2 Us (5. c ∼ 340 thus the whole process takes only a few milliseconds or less.87) U2 = c 1 k − 1 k + 1 P2 + 2k 2k P1 1− 1+ k+1 k−1 P2 P1 k+1 P2 k−1 P1 + (5.82) k − 1 k + 1 P2 + 2k 2k P1 (5. SHOCK TUBE k−1 P2 = P1 k+1 Rearranging equation (5. The pressure ratio between the two sides of the shock is 5.).89) .4. a review of the material is provided thus far. The following equations were developed earlier and are repeated here for clarification.88) the result is U2 = c1 k P2 −1 P1 2k k+1 P2 k−1 P1 1+k (5. A typical design problem of a shock tube is finding the pressure to achieve the desired temperature or Mach number. The relationship between the different properties was discussed earlier and because it is a common problem.82) becomes Ms1 = Or expressing the velocity as Us = Ms1 c1 = c1 k − 1 k + 1 P2 + 2k 2k P1 (5.86) (5.84) 2k Ms1 2 − 1 k−1 (5.111 √ range of the speed of sound.85) The fluid velocity in zone 2 is the same U2 = U s − U 2 = U s 1 − From the mass conservation.88) After rearranging equation (5. these kinds of experiments require fast recording devices (a relatively fast camera and fast data acquisition devices. Thus.

97) .95) and equation (5.96) is transformed into P5 P2  1 − = √ P1 P1   (k − 2k c1 1) c 5 P5 P3 −1 P2 P1 k − k2 −1 2k + (k + 1) −1    (5. d(ρU ) = 0.94) (5. applying the isentropic relationship (T ∝ ρk−1 ) yields c = c5 From equation (5.95) As it was mentioned earlier the velocity at points 2 and 3 are identical.92) Equation (5. hence equation (5. equation (5.90) Since the process in zone 4 is isentropic.93) The results of the integration are 2c5 U3 = k−1 Or in terms of the pressure ratio as 2c5 U3 = k−1 1− P3 P5 k −1 2k 1− ρ3 ρ5 k −1 2 (5. NORMAL SHOCK On the isentropic side.90) it follows that dρ dU = −c = c5 ρ ρ ρ5 k−1 2 T = T5 ρ ρ5 k −1 2 (5.96) After some rearrangement.89) can be combined to yield 2c5 k−1 1− P3 P5 k−1 2k = c1 k P2 −1 P1 2k k+1 P2 k−1 P1 1+k (5. in zone 4.91) dρ (5.92) can be integrated as follows: U3 ρ3 dU = U5 =0 ρ5 c5 ρ ρ5 k −1 2 dρ (5.112 CHAPTER 5. and dividing by the continuity equation the following is obtained: dU dρ =− ρ c (5. taking the derivative of the continuity equation.

In that case.99) By utilizing the isentropic relationship for zone 3 to 5 results in k5 −1 k5 k5 −1 k5 T3 = T5 P3 P5 = P2 P1 P5 P1 (5.5.5. Ms1 k−1 c1 k+1 c5 113 k1 − 1 P5 = P1 k+1+1 2k Ms1 2 − 1 k1 − 1 1− Ms1 Ms1 2 − 1 k − k2 −1 (5.100) Example 5. SHOCK WITH REAL GASES Or in terms of the Mach number. For a high Reynolds number.7 Shock with Real Gases Shock in Wet Steam Normal Shock in Ducts The flow in ducts is related to boundary layer issues.9: 5 A shock tube with an initial pressure ratio of P P1 = 20 and an initial temperature of 300K . the following is obtained: 1+ T2 = T1 1+ k1 −1 k1 +1 k1 −1 k1 +1 P2 P1 P1 P2 (5.98) Using the Rankine–Hugoniot relationship and the perfect gas model.6 5.5 5. the assumptions in construction of these models are acceptable and reasonable. Find the shock velocity and temperature behind the shock if the pressure P5 ratio is P = 40? 1 S OLUTION 5. the assumption of an uniform flow in the duct is closer to reality. It is normal to have a large Mach number with a large Re number. .

0850. 4 × 287 × 300 × 5. NORMAL SHOCK 5.668 Ty Tx Py Px P0 y P0 x 0. the time is t= distance 3 √ = ∼ 0.0850sqrt1.17: Figure for Example (5.41404 0.0021[sec] 1 . The ambient temperature is 27◦ C and 1[Bar]. The conditions upstream of the valve are 30[Bar] and 27◦ C .967 My kRTy .8 More Examples for Moving Shocks Example 5. Assume that there isn’t any resistance whatsoever in the pipe.0 5.00 0. After building the system. In this system. Mx = Msx = 5.4 × 287 × 300 Msx kRTx The mass flow rate after reaching the exit under these assumptions remains constant until the uncooled material reaches the exit.17)).057811 The direct calculation will be by using the “upstream” Mach number. Assume that the time scale for opening the valve is significantly smaller than the typical time of the pipe (totally unrealistic even though the valve manufacture claims of 0. The temperature should not increase above a certain value. the engineer notices that the system does not cool the device fast enough and proposes to increase the pressure and increase the diameter of the pipe. opening valve probelm The distance between between the valve and the pipe exit is 3[m].114 CHAPTER 5.967 30. S OLUTION This problem is known as the suddenly open valve problem in which the shock choking phenomenon occurs. air is Fig. The time it takes for the material from the valve to reach the exit is distance 3 t= = ∼ 0.0017[sec] 5. An engineer is required to design a cooling system for a critical electronic deexit valve vice.10: This problem was taken from the real industrial manufacturdistance ing world. Where any of these advises make any sense in the light of the above assumptions? What will be your recommendations to the manufacturing company? Plot the exit temperature and the mass flow rate as a function of the time. The time it takes for the shock to travel from the Py valve depends on the pressure ratio Px = 30 Mx 5.10) supposed to reach the pipe exit as quickly as possible when the valve is opened (see Figure (5. 668 sqrt 1 . Comment on this proposal. Therefore.0002 [sec] to be opened). 5.0850 My Mx My 1.

81264 0. k = 1.91177 0.99985 0.10 1. Thus.99893 0. Example 5.25504 1.65625 1. Try to estimate the temperature increase of the product.00000 0.9.20 1.98706 0.00000 1.97937 .4 Mx 1.16908 1.78596 Ty Tx ρy ρx Py Px P0y P0x 1.42857 1.00 1.11: Example (5.84217 0.05 1.5.19087 1.08398 1.25 1. There is also the possibility of steps increase in which every step heat released will not be enough to over heat the device. this suggestion contradicts the purpose of the required manufacturing needs.4 IDEAL GAS 115 During that difference of time the material is get heated instead of cooling down because of the high temperature.87502 0.80500 1.00000 1.4 Ideal Gas Table 5.10) deals with a damaging of electronic product by the temperature increase.99669 0.9 Tables of Normal Shocks.06494 1.30 My 1.03284 1. Time[Msec] To increase the pipe diameter will not change the temperature and therefore Fig.51333 1. It can only increase the rate after the initial heating spike A possible solution is to have the valve very close to the pipe exit.15 1. The last possible requirement a programmable valve and very fast which its valve probably exceed the moving shock the valve downstream.09658 1.51570 1. 5. S OLUTION 5.15938 1. TABLES OF NORMAL SHOCKS.95313 0.18).18: The results for Example (5.11958 1.1: The shock wave table for k = 1.99280 0. Thus. the heating time is reduced significantly.10) will not change the effects of heating.12799 1. K = 1.00000 1.37625 1. The plot of the mass flow rate and the velocity are given in Figure (5.34161 1.00000 0. The suggestion of the engineer to inMass Flow Rate crease the pressure will decrease the time but will increase the temperature at the exit during this critical time peVelocity riod.24500 1. Plot the pipe exit temperature as a function of the time.

42623 4.66667 2.72087 0.52861 0.94732 2.02498 .45833 26.07505 0.26087 4.75 6.25 3.04688 4.60278 1.09170 0.55333 6.40623 0.40 1.23958 18.75 5.87599 0.33333 12.49458 1.57735 0.80 1.1: The shock wave table for k = 1.75 3.64729 1.45 2.13876 0.05 2.97833 5.06172 0. NORMAL SHOCK Table 5.44231 0.11525 2.32022 1.00 6.42355 0.96974 0.40625 0.27625 6.62809 0.43496 0.55395 0.07229 4.50 5.47519 0.00 5.50 4.36906 7.69751 0.56148 0.81322 1.44891 4.41189 0.40216 Ty Tx ρy ρx Py Px P0y P0x 1.51568 2.46192 0.64054 0.91156 5.67420 0.04236 0.40897 0.98958 35.12500 38.54706 0.80000 6.11256 0.00500 6.95122 3.56906 0.38797 1.94059 8.21190 3.22261 1.85572 0.14897 3.73971 0.42280 1.76175 0.66894 4.69919 4.83333 45.68966 1.50 3.32834 0.65105 0.12500 8.40625 41.81190 2.21295 0.60 1.27907 2.74420 0.26829 5.53441 0.32544 5.60570 0.20500 3.57143 4.73958 6.59188 2.61333 3.15625 14.25 4.65396 0.98511 3.15625 29.95958 2.51792 0.28625 2.03536 0.49181 0.41523 0.25 My 0.35922 2.50 2.81268 0.26958 4.72855 1.88231 2.30 2.25469 1.77045 1.71956 0.90625 23.14894 5.12500 16.21182 5.53158 1.40642 0.58618 0.56935 1.85714 4.08455 3.94484 0.50 1.67901 2.90 1.03175 2.75 1.25 5.70 1.27335 3.74002 2.77614 1.39657 2.19772 2.76736 0.45 1.85 1.90138 1.68750 1.70109 0.45833 2.13750 2.65625 10.58329 0.07869 5.04033 2.50000 4.49901 0.00000 31.87509 5.05100 0.17166 0.68410 0.00 2.48000 5.52312 0.41908 0.86207 1.56128 0.63625 2.79023 0.12000 2.65 1.35 2.28720 1.82180 7.15 2.60553 0.35 1.116 CHAPTER 5.25 2.81188 4.43811 2.40416 0.00000 5.54014 0.31505 3.94680 1.02965 0.22625 5.53637 1.51931 0.92979 0.33333 3.20 2.10 2.00 4.82625 4.91319 0.51299 0.95819 0.62814 0.29878 6.00958 3.45115 0.61650 0.85686 1.45833 1.60792 1.83457 0.40 2.54055 0.66844 0.08846 2.55 1.59562 0.35379 1.42878 0.26451 0.04500 4.99311 2.01863 3.4 (continue) Mx 1.31915 1.73625 4.82000 3.89520 0.61194 3.83625 7.50000 20.00 3.75 4.40625 3.95 2.

67171 0.45833 69.23958 84.127 1.38980 0.442 1.42736 20.02115 0.004 1.06 0.024 1.39048 0.149 1.40038 0.73958 116.82755 .00387 0.043 1.99403 0.36508 5.344 1.99973 0.80 0.0 0.062 1.028 1.01798 0.082 1.00739 0.38807 0.0 0.020 1.133 1.95387 0.016 1.80077 10.20 0.126 1.38758 Ty Tx ρy ρx Py Px P0y P0x 9.64673 0.16229 11.219 1.99997 0.4 IDEAL GAS Table 5.00979 0.86537 0.00849 0.057 1.75 8.00000 61.95944 0.032 1.69946 0.03 0.00437 0.51020 5.15643 9.70 0.0 0.38918 0.269 1.04 0.47883 5.0 0.941 2.036 1.87948 12.01133 0.00343 0.780 3.76452 0.049 1.00304 Table 5.66874 5.028 1.38860 0.102 1.0 0.381 1.037 1.00000 0.92832 0.00 9.0 0.96506 0.055 1.75 9.676 1.38672 14.99317 0.09 0.50000 79.12500 110.00496 0.99998 0.50 6.50 7.50 8.475 2.196 1.97099 0.0 0.0 0.00 My 0.56522 5.50000 0.087 1.710 1.10 0.072 1.99113 15.61165 5.50 0.05 0.07 0.39385 0.99999 0.9.171 1.99986 0. TABLES OF NORMAL SHOCKS.00 7.65116 5.15625 65.99790 0.018 1.84463 0.0 0.53890 5.0 0.1: The shock wave table for k = 1.98812 0.80251 0. K = 1.70019 5.12500 52.00 My 0.118 1.195 2.12500 89.39491 0.99994 0.4) Mx 1.00565 0.50 9.71429 49.505 1.57997 18.89918 0.473 1.316 1.46939 11.423 1.17678 14.97074 0.39607 0.589 1.39736 0.0 0.60 0.90625 74.008 1.33333 99.73029 0.043 1.0 0.5.65625 105.00 8.40667 5.75 10.25 7.39289 0.39201 0.40 0.323 1.62425 Mx 0.0 0.01316 0.766 My 0.506 1.44444 5.95231 0.97647 0.0 0.014 1.25 8.94291 0.89128 0.49152 19.39879 0.01535 0.012 1.00645 0.0 Ty Tx Py Px P0y P0 x 1.08 0.98446 0.024 1.94836 0.0 0.99991 0.38750 5.68504 5.75 7.2: Table for a Reflective Shock from a suddenly closed end (k=1.502 1.98227 0.98958 57.0 0.012 1.0 0.62095 13.82978 16.99980 0.30 0.63218 5.90 1.02 0.268 1.39121 0.112 3.25 9.00000 1.01 0.15625 94.4 (continue) 117 Mx 6.040 1.0 0.030 1.58939 5.006 1.69273 17.

15637 0.50 0.20 1.95315 0.49092 0.10 0.397 1.50814 0.066 8.134 1.30 1.30 0.028 1.0 0.0 0.0 0.727 1.70 Ty Tx Py Px P0 y P0 x 1.088 1.014 1.65290 0.40284 0.858 1.952 2.99985 0.061716 0.576 1.791 2.38457 Mx 1.057 1.00000 1.331 1.000 6.256 6.0 0.56935 0.94746 0.70283 0.541 1.280 4.38817 0.026517 0.00 6.25 24.728 5.3: Table for shock propagating from suddenly opened valve (k=1.00 3.62 84.38608 0.033 1.07 0.448 2.99996 0.00 10.50 1.162 7.40 0.78652 0.037 1.0 0.98807 0.39566 0.99763 0.55453 0.317 1.495 1.024 1.56312 0.104 1.133 1.43921 0.0 0.00350 0.087 2.00639 0.0 0.006 1.0 Ty Tx Py Px P0 y P0 x 1.0 0.0 0.20 0.859 5.00 44.99971 0.0 0.00 5.990 2.33 3.1E +2 1.02 0.78840 0.88717 0.60 0.21 29.715 6.024 1.93133 0.004 1.0 0.336 8.99979 0.058 1.018 1.66462 Mx 0.51996 0.00204 0.0 0.245 2.012492 0.49912 0.10 1.40213 0.47 1.831 5.95888 0.0 0.03 0.136 1.74316 0.031 1.01 29.0 0.97045 0.0 0.0 0.2: Table for Reflective Shock from suddenly closed valve (end) (k=1.410 3.69834 0.99402 0.00 4.09 0.06 0.99990 0.39116 0.0 0.47855 0.99994 0.044 1.734 17.0 0.0 0.016 1.94180 0.029 1.90 2.60401 0.210 1.0 My 0.297 2.0 0.0 0.012 1.00 7.862 4.810 1. NORMAL SHOCK Table 5.05 0.58578 0.517 9.99998 0.041 1.762 3.41 15.54114 0.97629 0.087 2.146 2.41523 0.4)(continue) Mx 1.346 2.43894 0.98216 0.52904 0.152 1.189 2.14 62.613 My 0.0 0.51808 0.441 2.552 2.00 8.01 0.96465 0.0 0.0 0.012 1.05 19.020 1.00 9.120 1.0 0.7E +2 0.60 1.70 1.40 1.868 1.83607 0.0 0.89 12.248 1.037 1.74403 0.118 CHAPTER 5.08 My 0.89039 .325 11.897 1.008 1.295 1.048 2.0 0.96069 0.800 8.4E +2 1.0 0.00125 Table 5.649 1.99181 0.08 0.390 1.656 2.827 7.00000 0.0 My 0.050 1.086 1.4) Mx 1.206 5.188 1.073 1.0 0.60761 0.043 1.0 0.98019 0.0 0.0 0.431 8.99999 0.063 1.80 1.04 0.703 10.

0 0.7E +2 8.33 21.80 76.0 0.0 0.5E + 3 5.99427 0.830 1. TABLES OF NORMAL SHOCKS.39028 0.00721 0.07 18.00175 0.83661 0.293 8.457 10.49586 0.24 11.700 1.38248 0.00000 0.4E +3 3.5.37817 0.48823 0.860 1.74 66.6E + 3 4.84 1.810 1.00544 0.387 4.003 1.889 Ty Tx Py Px P0 y P0 x 1.2E +3 1.69 68.59699 0.4E + 3 4.9E + 2 2.56619 0.0 0.3E +2 7.37821 0.97166 Mx 0.72 80.0 0.040812 0.840 7.888 1.54 15.35 25. K = 1.167 3.07 90.0 0.2E + 3 0.919 2.145 2.0 0.98857 0.97726 0.39187 0.02 83.0 0.59649 0.3) Mx 1.0 0.0058 1.030 0.0 My 0.0 0.79 My 0.0 0.9E + 3 7.45 22.1E +3 1.18E −5 0.012 1.799 3.5E + 2 1.025 4.62923 0.745 1.000181 3.000497 0.99 25.0 0.0 Table 5.2E +2 8.4E + 3 6.37816 0.0E + 2 1.4) 119 Mx 1.888 1.880 1.889 1.010 0.006 1.99998 0.9E +2 1.46599 0.821 9.604 1.38557 0.0 0.823 5.023 1.4: Table for shock propagating from a suddenly opened valve (k=1.38974 0.040 0.800 1.500 1.889 1.8E + 2 7.28412 0.067 1.5E + 2 3.38713 0.029 My 0.0 0.62 61.300 1.508 2.7E + 2 4.37818 0.92 42.3: Table for shock propagating from suddenly opened valve (k=1.0 0.0 0.9E + 3 5.551 8.53817 0.38096 0.37814 0.38402 0.0 0.269 2.44536 0.536 4.0 0.10216 0.00000 1.889 1.0 0.054 1.0 0.850 1.0 0.61 1.00 1.888 1.0 0.2E + 2 1.62 17.37812 0.0 0.2E + 3 5.0 0.171 8.51223 0.37944 0.90 1.881 3.888 1.040 1.37822 0.0 0.015 1.0 0.98290 0.100 1.870 1.0 0.00101 0.32 16.0 0.017 1.4 IDEAL GAS Table 5.9.840 1.975 11.0 0.450 2.020 0.18 73.57 31.25 12.931 14.820 1.0 0.0 0.83 71.009 1.40843 0.2E +2 9.0 0.026 1.76940 0.37820 0.600 1.0 0.013 1.00395 0.18575 0.400 1.9E + 3 6.0 0.5E + 3 8.3E +2 7.012 1.7E +2 9.1E +3 1.491 1.888 1.0 0.62 14.37813 0.78 80.38870 0.889 1.31 62.99997 .0 My 0.744 1.0 0.99999 0.0 0.200 1.00272 0.050 Ty Tx Py Px P0y P0 x 1.068 2.0 0.896 2.53 63.0 0.37810 Mx 0.368 5.42622 0.09 35.80 0.68907 0.95 64.658 4.0 0.646 6.0 0.

99288 0.202 3.0 0.069 2.553 1.95 23.39780 0.40257 0.064766 0.048 1.236 2.471 4.31281 0.30 23.0 0.39160 0.0 0.0 0.39037 0.0 0.053053 0.888 1.94 23. NORMAL SHOCK Table 5.39314 0.100 4.0 0.0 0.592 3.0 0.0 My 0.052914 0.166 4.707 5.200 1.0 0.22904 0.300 1.136 1.096 4.96 0.519 3.980 2.522 6.0 0.49 20.0 0.900 1.39938 0.500 1.889 Ty Tx Py Px P0 y P0 x 1.830 1.58223 0.523 7.95506 0.053 4.4: Table for shock propagating from a suddenly opened valve (k=1.0 0.95 19.45807 0.700 0.80734 0.66575 0.100 0.94415 0.098 1.500 0.889 1.99975 0.55074 0.32 18.0 0.600 1.39035 0.0 0.99991 0.959 2.39030 0.302 1.39034 0.088718 0.0 0.0 0.997 3.96056 0.333 1.86274 0.400 1.90 22.359 1.67546 0.452 1.842 3.0 0.681 1.274 1.140 1.39029 0.40097 0.99995 0.099 4.0 0.800 1.070 0.054 1.39033 0.060462 0.0 0.98290 0.64 21.103 1.400 1.64073 0.870 1.0 0.888 1.0 0.400 0.785 1.100 1.0 0.89159 0.367 11.749 3.052984 0.95 23.018 1.042 1.615 4.39026 Mx 0.0 0.083607 0.772 9.177 1.91 23.75301 0.0 0.489 1.90734 0.031 1.05 20.58 23.614 4.034 4.615 4.40418 0.0 0.46 14.39031 0.93 23.120 CHAPTER 5.286 4.810 1.00 1.093988 0.94959 0.47875 0.0 0.613 4.0 0.467 2.84227 0.053088 0.820 1.42089 0.923 4.71284 0.40226 0.57853 0.300 0.200 0.060 0.052809 .706 1.021 1.478 1.43882 0.252 2.60847 0.613 4.098 4.125 1.073863 0.663 1.553 4.860 1.700 1.069233 0.191 2.99987 0.223 1.50100 0.576 4.0 0.99792 0.888 1.15495 0.889 1.0 0.016 4.0 0.0 0.800 0.024 1.25 21.0 0.0 0.840 1.381 3.669 3.063 1.035 1.080 0.081 1.93 23.49333 0.349 4.197 1.74136 0.79611 0.92 23.39624 0.99981 0.0 0.94156 0.126 1.275 1.44 18.595 2.91 23.078654 0.600 0.052879 0.919 2.850 1.101 4.888 1.612 4.616 My 0.570 1.0 0.52495 0.097 4.889 1.3) Mx 1.644 2.888 1.090 0.611 4.028 1.052844 0.482 4.056322 0.39027 0.426 2.448 3.880 1.053018 0.102 4.096 1.96631 0.109 4.898 3.0 0.815 1.060 1.0 0.0 0.225 4.834 3.415 4.96610 0.099 4.052949 0.616 4.110 1.39468 0.

5.97 0.0 My 1.104 23.4 IDEAL GAS Table 5.3) 121 Mx 4.9.889 Ty Tx Py Px P0y P0 x 4. TABLES OF NORMAL SHOCKS.052775 .4: Table for shock propagating from a suddenly opened valve (k=1.39025 Mx 0.617 My 0. K = 1.

NORMAL SHOCK .122 CHAPTER 5.

It is §© c interesting to note that a normal  a  #!$¡% Subsonic shock must occur in these situations (pressure ratios). When the Fig. the shock can occur only in steady state when there is a supersonic flow. x zle is shown in its whole range of pressure ratios. 6.1) the reSupersonic duced pressure distribution in !¡" b the converging–diverging nozdistance.CHAPTER 6 Normal Shock in Variable Duct Areas In the previous two chapters. the flow in a convergent– divergent nuzzle was presented when the pressure ratio was above or below the special range. A discussion of the occurrences of shock in flow in a variable is presented. the flow in a variable area duct and a normal shock (discontinuity) were discussed. In the previous chapter. P B is between sures w ic flo subson shock a after 123 .1: The flow in the nozzle with different back prespressure ratio. As it was shown in Chapter 5. This Chapter will present the flow in this special range of pressure ratios. d In Figure (6. but also in steady state cases when there is no supersonic flow (in stationary coordinates). the gas has to pass through a converging–diverging nozzle to obtain a supersonic flow. As it is was presented before.

S OLUTION Since the key word “large tank” was used that means that the stagnation temperature and pressure are known and equal to the conditions in the tank. Calculate the back pressure and the temperature of the flow. Obviously if the back pressure. If the back pressure. Nevertheless. The location of the shock is determined by geometry to achieve the right back pressure. . this example will provide the fundamentals to Fig.124 CHAPTER 6. point "e" niques). Nozzle throat area is 3[cm2 ] and the exit area is 9[cm2 ]. In the under–expanded case.+ -£. This nozzle is called an over–expanded nozzle. no continuous pressure possibly can exists.0/21 (see Mach’s photography techexit 354687292:2. In this case. In the literature. It XY Z\[ ]_^8`Aa bcFdOe troat has to be recognized that the shock wave isn’t easily visible &'(*). P B is smaller than Pb a discontinuous point (a shock) will occur. Therefore. In conclusion. (It should be noted that the temperature of the surrounding is irrelevant in this case.2: A nozzle with normal shock explain the usage of the tools (equations and tables) that were developed so far. The first example is for academic reasons. In comparison of nozzle performance for rocket and aviation. On the other hand. If the back pressure is within the range of Pa to Pb than the exact location determines that after the shock the subsonic branch will match the back pressure. some refer to a nozzle with an area ratio such point b as above the back pressure and it is referred to as an under–expanded nozzle. when the nozzle exit area is too large a shock will occur and other phenomenon such as plume will separate from the wall inside the nozzle.1: A large tank with compressed air is attached into a converging–diverging nozzle at pressure 4[Bar] and temperature of 35[◦ C]. Only in one point where P B = Pb continuous pressure exist. once the flow becomes supersonic. The shock occurs in a location where the cross section area is 6[cm2 ]. the nozzle doesn’t provide the maximum thrust possible. this example provides a demonstration of x y the calculations required for the ¥ <  = 8 > A ? @ D B F C H E G  I H J M K O L N § P Q*RAS TDUFVHW location even if it isn’t realistic. Example 6. 6. is lower than the critical value (the only value that can achieve continuous pressure) a shock occurs outside of the nozzle. NORMAL SHOCK IN VARIABLE DUCT AREAS point “a” and point “b” the flow is different from what was discussed before. only exact geometry can achieve continuous pressure flow. P B .) Also determine the critical points for the back pressure (point “a” and point “b”). the over–expanded nozzle is worse than the under–expanded nozzle because the nozzle’s large exit area results in extra drag.

11) the following Table for the isentropic relationship is obtained M 0. Mx 2.0000 0.18463 2.125 First. My .50877 0. (point “y”).54746.1972 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. as Ae A∗ = 1. Thus. A∗ . the Mach number at the exit can be calculated by utilizing the isentropic relationship. From equation (5. For example. the subsonic branch can be evaluated for the pressure and temperature ratios. the exit Mach number has to be determined. before the shock is known and given as well.22) or from Table (4.54743 Ty Tx ρy ρx Py Px P0y P0 x 1. It has to be realized that for a large tank.09393 0.197 as shown table below: M 2. one can obtain using the isentropic relationship . in the case of ratio of 100 the Mach number is 0.8882. With these values. Then the relationship developed for the shock can be utilized to calculate the Mach number after the shock.8882 ∗ A Ay A 6 with this area ratio.8544 2.9474 5. The star area (the throat area).2) My ∼ = 0.1).1972 the Mach number.18787 With this Mach number. but can be shown that the correction is negligible for a typical dimension ratio that is over 100. the Mach number. With known Mach number the new star area ratio. From the Mach number after the shock. Ax 6 = =2 A∗ 3 With this ratio (A/A∗ = 2) utilizing the Table (5.1972 My 0.2588 × = 1.2588 0.94345 0. Ay /A∗ is known and the exit area can be calculated as Ae Ay 9 Ae = × ∗ = 1.81568 1.4656 0.0268 Again utilizing the isentropic relationship the exit conditions can be evaluated. Mx = 2. the inside conditions are essentially the stagnation conditions (this statement is said without a proof.62941 From Table (4. My can be obtained.1) or equation (4.2) or from equation (4. This Mach number can be calculated by utilizing the isentropic relationship from the large tank to the shock (point “x”).86457 1. the stagnation temperature and pressure are known T0 = 308K and P0 = 4[Bar]. Mx is about 2.54743 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.00587 and the error is less than %0.49) or the GDC– Potto.

99226 0.99226 × 4 ∼ =3.97[Bar] For the supersonic sonic branch Pexit = Pexit P0 P0 = 0.11310 3.32651 CHAPTER 6.97318 0. In that case we don’t have to go through that shock transition.14190 Pexit = Pexit P0 P0 = 0.466 × 0. the exit conditions can be calculated.509 × 308 For the “critical” points ”a” and ”b” are the points that the shock doesn’t occur and yet the flow achieve Mach equal 1 at the throat.92882 1.0000 3.41820 0.951 Ty Tx Tx T0 T0 =0.6374 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.19745 2. Pexit = Pexit P0 P0 Py Py Px Px P0 P0 =0.7538 Since the stagnation pressure is constant as well the stagnation temperature.41820 × 4 ∼ =1.9K × 1.98133 × ∼ =299.04730 2.9195 0.92882 × ∼ =2. NORMAL SHOCK IN VARIABLE DUCT AREAS T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. is A/A∗ = 3 In the subsonic branch (either using equation or the isentropic Table or GDC-Potto as M 0. Yet we have to pay attention that there two possible back pressures that can “achieve” it or target.6728[Bar] .0000 0.97912 0.094 × 4 T0 Ty 1 0. The area ratio for both cases.98077 0.94862 1.854 × 0.8882 0.34[Bar] The exit temperature is Texit = Texit T0 1 0.81568 × 5.126 M 0.

5283 × 4 = 2.7 = 321. Example 6.7K T∗   R  T0 T0  T∗ A √ kRT ∗ = P∗ P0 P0 R T∗ T0 A kR T0 T∗ T0 T0 The temperature at the throat reads P∗ = The speed of sound is c= √ 1. While the pressure to achieve full supersonic flow through the nozzle the pressure has to be below the 42% the original value.12[m/sec] P∗ P0 P0 = 0. In these applications a small pressure difference can produce a shock wave and a chock flow.4 × 287 × 256.1) where would be shock’s location when the back pressure is 2[Bar]? 1 The meaning of the word practical is that in reality the engineer does not given the opportunity to determine the location of the shock but rather information such as pressures and temperature. these kind situations exist.833 × 308 = 256. For more practical example1 from industrial application point of view.13[kg/sec] 287 × 256. . Thus.7 It is interesting to note that in this case the choking condition is obtained (M = 1) when the back pressure just reduced to less than 5% than original pressure (the pressure in the tank). In fact in many industrial applications. The flow rate is expressed as following  ∗  P ρ∗ P∗ m ˙ =ρ A U = A cM = RT ∗ ∗ ∗ M =1 P ∗     P0   P0   c The temperature and pressure at the throat are: T∗ = T∗ T0 T0 = 0.127 It should be noted that the flow rate is constant and maximum for any point beyond the point ”a” even if the shock is exist.2: In the data from the above example (6. over 50% of the range of pressure a shock occores some where in the nozzle.113[Bar] And the mass flow rate reads m ˙ = 4105 3 × 10−4 × 321.12 = 0.

the previous example information used and expanded.5000 0. More importantly the pressure ratio exit is known.2) or the GDC-Potto provides the following table is obtained T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ M 0.0128 3. The ratio of the ratio of stagnation pressure obtained by f or Mexit P0 y = P0 x P0 y Pexit Pexit P0 x = 2 1 × = 0. utilizing the isentropic Table (4.90500 1. is known. 2 Of course.5525 0. The second (recommended) method is noticing that the flow is adiabatic and the mass flow rate is constant which means that the ratio of the P0 × A∗ = Py0 × A∗ |@y (upstream conditions are known. NORMAL SHOCK IN VARIABLE DUCT AREAS S OLUTION The solution procedure is similar to what was shown in previous Example (6. Mathematically. Pexit Aexit 2×9 Pexit Aexit = = = 1. The solution process starts at the nozzle’s exit and progress to the entrance. the needed pressure is only 2[Bar] which means that the next guess for the shock location should be with a larger area2 .5[unitless!] Px 0 × A x ∗ Py 0 × A y ∗ 4×3 A With the knowledge of the ratio PP ∗ which was calculated and determines the exit 0A Mach number. there are two main possible ways to obtain the solution. the exit pressure is between point “a” and point “b”.3914 0. First. In the first method. The exit total pressure can be obtained (if needed).52628 Ty Tx ρy ρx Py Px P0 y P0 x 2. Thus. . the exit Mach number. It follows that there must exist a shock in the nozzle.97188 0. The area (location) that the previous example did not “produce” the “right” solution (the exit pressure was 2. the computer can be use to carry this calculations in a sophisticate way.2) or utilizing the GDC-Potto provides Mx 2.1755 6.2).75158 With these values the relationship between the stagnation pressures of the shock are obtainable e.g. Here.3709 My 0.72)).55250 With the information of Mach number (either Mx or My ) the area where the shock (location) occurs can be found.905 4 Looking up in the Table (4. see also equation (4. Utilizing the Table (4. In fact. it requires some iterations by “smart” guessing the different shock locations. My .6575 0.38034 0. The conditions in the tank are again the stagnation conditions.113[Bar].93118 1.1).128 CHAPTER 6.

6. as “simple” check this value is larger than the value in the previous example.16747 Approaching the shock location from the upstream (entrance) yields A= A ∗ A = 2.0188[cm2] A∗ Note.1.4) There is another less used definition which referred as the coefficient of discharge as the ratio of the actual mass rate to the ideal mass flow rate. One of the effective way is to define the efficiency as the ratio of the energy converted to kinetic energy and the total potential energy could be converted to kinetic energy.3396 0. In the literature some define also velocity coefficient as the ratio of the actual velocity to the ideal velocity.3) (6.1) where hexit s is the enthalpy if the flow was isentropic.3709 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 129 0.9 to 0. Vc Vc = √ η= (Uactual )2 (Uideal ) 2 h0 − hexit (Uactual )2 = 2 h0 − hexit s (Uideal ) (6.15205 2.5) 6.47076 0.1 Nozzle efficiency Obviously nozzles are not perfectly efficient and there are several ways to define the nozzleefficiency.2) The efficiency can be defined as η= The typical efficiency of nozzle is ranged between 0.3396 × 3 ∼ = 7. The total energy that can be converted is during isentropic process is E = h0 − hexit s (6. Cd = m ˙ actual m ˙ ideal (6.2 Diffuser Efficiency .99. The actual energy that was used is E = h0 − hexit (6.07158 0. NOZZLE EFFICIENCY M 2. 6.

Assume that a shock Fig.3) 0. 6.130 CHAPTER 6.07623 4.11528 0.0000 T T0 ρ ρ0 A A P P0 uous region (and also for example (6. NORMAL SHOCK IN VARIABLE DUCT AREAS P01 h The efficiency of the diffuser is defined as the ratio of the enthalpy change that occurred between the entrance to exit stagnation pressure to the kinetic energy. k = 1.02[m2 ].35714 0. η= 2(h3 − h1 ) h3 − h 1 = h01 − h1 U1 2 (6.02722 0.65326 A×P A ∗ ×P 0 F F∗ .7) s. S OLUTION The condition at M = 3 is summarized in following table M 3. What is cooler area of nozzle’s throat and what is area of the diffuser’s throat to maintain chocked diffuser with heat subsonic flow in the expansion out section. 6.4: Schematic of a supersonic tunnel in a continoccurs in the test section.4 can be assumed.0 and prescapacitor sure of 0.3: A wind tunnel combined from Diffuser nozzle a nozzle and a diffuser (actually two nozzles connected by a 1 f£gMh 2 3 i£jMk 4 constant area see Figure (6.6) can be converted to η= 2Cp (T3 − T1 ) U1 2 (6.entropy Fig.6) P02 P2 01 02 2 P1 1 For perfect gas equation (6.7[Bar] and temperature of 250K .7) results in η= kR 2k −1 T1 T3 T1 − c1 2 M1 2 1 = 2 k−1 T3 T1 M1 2 −1 2 = 2 M1 (k − 1) T3 T1 k−1 k −1 (6.2346 0.3: Description to clarify the definition of diffuser efficiency And further expanding equation (6.4)) the required condition at point 3 are: M = 3.8) Example 6. The cross section in area between the nuzzle Compressor and diffuser is 0.

32834 A∗ d = P0 n ∗ 1 A n∼ 0.5346 My 0.021717 The calculations were carried as following: First calculate the Mx as M x = Us / (k ∗ 287.02172 The calculation of the temperature and pressure ratio also can be obtain by the same manner.47519 Ty Tx ρy ρx Py Px P0y P0 x 2.3333 0.2963 34.4789 6.09668 . P0 A∗ is constant (constant mass flow). DIFFUSER EFFICIENCY The nozzle area can be calculated by A∗ n = A A = 0.02/4.32834 Example 6.3.989 Ty Tx Py Px P0 y P0 x 0. First the stagnation behind the shock will be Mx 3. In this case Potto–GDC provides the following table Mx 5. pressure of 2[Bar] and temperature of 350K . Calculate the conditions after the shock. S OLUTION This is a case of completely and suddenly open valve with the shock velocity.37554 0.0143[m3] P0 d 0.6790 3.50 0.6.479 34.0047[m2] A 131 In this case.2346 = 0.4: A shock is moving at 200 [m/sec] in pipe with gas with k = 1.8571 10. For example Potto-GDC (this code was produce by the program) Mx 5. temperature and pressure “upstream” known.0047 ∼ 0.37554 Ty Tx ρy ρx Py Px P0y P0 x 5.5346 My Mx My 1.0000 My 0.0 5. The “downstream” shock number is Msy = Us k ∗ 287. ∗ Tx ∗ Ty Tx ∼ 2.4968 0. ∗ Tx ) Then calculate the My by using Potto-GDC or utilize the Tables.2.

6250 2.15 cx 1. Thus.25 Ux =√ ∼ 2.09668 − 0. The imaginary gas conditions are temperature is 350K and pressure is 2[Bar] and R = 143[j/kg K ] and k = 1.25[m/sec] ρA PA 200000 × 0.96697 The temperature ratio and the Mach numbers for the velocity of the air (and the piston) can be calculated.4 ∗ 287 ∗ 369.989 Example 6.41087 ∼ 1.132 CHAPTER 6. S OLUTION This is the case of a closed valve in which mass flow rate with the area given.091 (Butane?).2308 1.091 × 143 × 350 Thus the static Mach number. NORMAL SHOCK IN VARIABLE DUCT AREAS Finally utilizing the equation to calculate the following My = Msy − My = 2.75593 ∗ 1. Ux = m ˙ mRT ˙ 2 × 287 × 350 = = ∼ 502. the “upstream” Mach is given.75593 Ty Tx ρy ρx Py Px P0 y P0 x 1.3628 My 0.0000 0. what will be the piston velocity? S OLUTION This is an open valve case in which the pressure ratio is given. Calculate the conditions behind the shock wave.5: An inventor interested in a design of tube and piston so that the pressure is doubled in the cylinder when the piston is moving suddenly.24 ∼ 291. The propagating piston is assumed to move into media with temperature of 300K and atmospheric pressure of 1[Bar]. If the steady state is achieved. The temperature at “downstream” (close to the piston) is Ty = 300 × 1. The mass flow rate of the gas is 2 [kg/sec] and cross section A = 0. Mx is Mx = .16[m/sec] Example 6. For this pressure ratio of Py /Px = 2 the following table can be obtained or by using Potto–GDC Mx 1.6: A flow of gas is brought into a sudden stop.002 502.2308 = 369.002[m3 ].24[◦C] Ty = T x Tx The velocity of the piston is then √ Uy = My ∗ cy = 0.

589 2.0 2.47996 0.4096 9.923 2.922 My 0.590 2.796 .47996 Ty Tx Py Px My 0. DIFFUSER EFFICIENCY With this value for the Mach number Potto-GDC provides Mx 2.2.914 9.47988 0.796 0.0 0.0 0.47996 2.6.8598 2.1500 2.589 11.35101 This table was obtained by using the procedure described in this book. The iteration of the procedure are i 0 1 2 3 4 5 Mx 3.804 9.0 Ty Tx Py Px P0 y P0 x 133 0.940 2.922 2.796 9.796 9.0 0.47886 0.1500 2.0 0.589 2.922 2.46689 0.609 2.9222 My Mx My 0.47995 0.589 9.0 0.

NORMAL SHOCK IN VARIABLE DUCT AREAS .134 CHAPTER 6.

the working equations are also different and this author isn’t aware of material in the literature which provides any working table for the gravity effect. The isothermal nozzle model is suitable in cases where the flow is relatively slow (small Eckert numbers) while as the isentropic model is more suitable for large Eckert numbers. The equations results in slightly different conditions for the chocking and different chocking speed. These models can served as limiting cases for more realistic flow. isothermal. 135 . The effects of the gravity of the nozzle flow in two models isentropic and isothermal is analyzed here. You can help especially if you have photos showing these effects. Flow in a vertical or horizontal nozzle are different because the gravity. In cases where more refined calculations have to carried the gravity or other forces have to be taken into account. In the previous chapters a simple model describing the flow in nozzle was explained. The simplified models that suggests them–self are: friction and adiabatic. seem the most applicable. please do so.CHAPTER 7 Nozzle Flow With External Forces This chapter is under heavy construction. The two models produces slightly different equations. to this chapter. Please ignore. Moreover. If you want to contribute and add any results of experiments.

136 CHAPTER 7. NOZZLE FLOW WITH EXTERNAL FORCES 7.e.1) 7.1 Isentropic Nozzle (Q = 0) external work or potential difference. i. z × g dh + U dU = f (x)dx The energy equation for isentropic nozzle provides (7.2 Isothermal Nozzle (T = constant) .

in a perfect gas. to be kept in mind that the gas undergoes an isothermal process. as a constant . for engineering purposes. 1/ k . is relatively high and the isothermal flow model is not valid anymore. For instance. To main.S. Such situations are common in large cities in U. Close to the critical point (about.1: Control volume for isothermal flow tain conservation of mass.v. Ž The high speed of the gas is obtained or explained by the combination  vxwzy{v l flow of heat transfer and the friction to the |~}z€| direction flow. 8.CHAPTER 8 Isothermal Flow In this chapter a model dealing with gas that flows through a long tube is described. √ explanation is not correct as it will be shown later on.). For a long pipe. This model has a applicability to situations which occur in a relatively long distance and where heat transfer is relatively rapid so that the temperature can be treated.A. It is more predominant (more applicable) in situations where the gas is pumped over a length of kilometers. Therefore. this model is applicable when a natural gas flows over several hundreds of meters. the pressure dif‚„ƒz…{‚ mnFoqpsrut †‡ƒz…!†!ˆq‰sŠ ƒ‹… ŠŒ ference reduces the density of the gas. 1 This To put discussion for what the “relatively rapid” means. where natural gas is used for heating. the velocity increases inversely to the pressure. At critical point the velocity reaches the speed of sound at the exit and hence the flow will be choked1 . the den‘’ sity is inverse of the pressure (it has c.Fig. the heat transfer. 137 . the study of the isothermal flow above this point is only an academic discussion but also provides the upper limit for Fanno Flow. For example.

4) (8. First it must be recalled that the temperature is constant and therefore. Hence.5) 8. The Awetted area is the area that shear stress acts on.3) Again it is assumed that the gas is a perfect gas and therefore.). Dimensionless Representation In this section the equations are transformed into the dimensionless form and presented as such. The second law of thermodynamics reads T2 k − 1 P2 s2 − s 1 = ln − ln Cp T1 k P1 The mass conservation is reduced to m ˙ = constant = ρU A (8. a close enough shape is sufficient.6) . (8.1) (8. equation of state is expressed as the following: P = ρRT (8. The shear stress is the force per area that acts on the fluid by the tube wall.1 The Control Volume Analysis/Governing equations Figure (8.2 it seems obvious to write this equation perhaps to consult with others. equation of state reads dP dρ = P ρ It is convenient to define a hydraulic diameter DH = 4 × Cross Section Area wetted perimeter (8. ISOTHERMAL FLOW 8.138 CHAPTER 8.1) describes the flow of gas from the left to the right. the energy equation can be written as the following: U2 dQ = cp dT + d = cp dT0 m ˙ 2 The momentum equation is written as the following −AdP − τw dAwetted area = mdU ˙ Perhaps more quantitative discussions about how “circular” the shape should be.2) where A is the cross section area (it doesn’t have to be a perfect circle.7) (8. The heat transfer up stream (or down stream) is assumed to be negligible.

U .11) d(M 2 ) d(U 2 ) dT = − M2 U2 T (8.9) Rearranging equation (8.11) yields d(U 2 ) = kR M 2 dT + T d(M 2 ) (8. P as a function of the Mach number has to substitute along with velocity. 2 It where are the stagnation equations? put them in a table put explanation how to derive this expression.8.13) Now it can be noticed that dT = 0 for isothermal process and therefore d(U 2 ) 2U dU 2dU d(M 2 ) = = = M2 U2 U2 U The dimensionalization of the mass conservation equation yields dρ dU dρ 2U dU dρ d(U 2 ) + = + = + =0 ρ U ρ 2U 2 ρ 2 U2 (8. .10) Now the pressure.8) Substituting equation (8.8) into momentum equation (8. U 2 = kRT M 2 Differentiation of equation (8. instead of diameter friction equation.2) yields 4dx −dP − f DH 1 2 ρU 2 m ˙ A = ρU dU (8.9) and using the identify for perfect gas M 2 = ρU 2 /kP yields: − 4f dx dP − P DH kP M 2 2 = kP M 2 dU U (8. this factor is a dimensionless friction factor sometimes referred to as the friction coefficient as f= τw 1 2 2 ρU (8.15) (8. the Fanning friction factor2 is introduced.14) Differentiation of the isotropic (stagnation) relationship of the pressure (4.11) yields should be noted that Fanning factor based on hydraulic radius.12) (8.2. thus “Fanning f” values are only 1/4th of “Darcy f” values. DIMENSIONLESS REPRESENTATION 139 Now.

state are described above.22) the upstream variables are known.9) yields: dT0 = dT 1+ k−1 2 M 2 +T k−1 dM 2 2 (8. T.19) By utilizing the momentum equation it is possible to obtain a relation between the pressure and density. energy. 4D .140 dP0 dP = + P0 P 1 CHAPTER 8.20) The four equations momentum. P.20) to become dU dP =− P U 3 Assuming (8.17) Notice that dT0 = 0 in an isothermal flow.15) when combined with equation (8. There is no change in the actual temperature of the flow but the stagnation temperature increases or decreases depending on the Mach number (supersonic flow of subsonic flow). There are 4 unknowns (M. One can notice that there are two possible solutions (because of the square power). Thus. fL The distance friction.14)) leads dM 2 2dU = 2 M U (8. Recalling that an isothermal flow (T = 0) and combining it with perfect gas model yields dρ dP = P ρ From the continuity equation (see equation (8. is selected as the choice for the independent fL . . The variable. Substituting T for equation (8.18) dT0 (k − 1) M 2 dM 2 = 1 T0 M2 2 1 + k− 2 (8. ISOTHERMAL FLOW 1 2 2 kM k−1 + 2 M2 dM 2 M2 (8.16) Differentiation of equation (4. the equations need to be obtained as a function of 4D density is eliminated from equation (8. continuity (mass).18) yields M2 M2 (8. ρ)3 and with these four equations the solution is attainable.21) (8.17) yields: T0 1 k−1 2 2 d M 1 2 + k− 2 M dT0 = Rearranging equation (8. These different solutions are supersonic and subsonic solution.

Since the stagnation temperature (T0 ) has a finite value which means that dT0 → ∞.2. To obtain this restriction 1 = kM 2 . The term 4D the other side has to be positive as well. Integration of equation (8.22) into equation (8.29) . one can obtain that 1/k M2 = ∗2 U2 U (8. Thus.24) can be separated to obtain integrable form as follows L 0 dT0 dx k (1 − k ) M 2 4f = k−1 2 2 T0 D 2 (1 − kM ) 1 + 2 M (8.8. When the value M = √ k 1 Mach number larger than M > √ it makes the right hand side of the integrate k negative. 1 is the limiting case from a mathematical point of view.27) yields 4f Lmax D = 1 − kM 2 + ln kM 2 kM 2 (8. By denoting the superscript symbol ∗ for the choking condition. The physical meaning of this value is similar to M = 1 choked flow which was discussed in a variable area flow in Chapter (4).28) The definition for perfect gas yields M 2 = U 2 /kRT and noticing that √ T = constant is used to describe the relation of the properties at M = 1/ k .23) can be rearranged into dρ dU 1 dM 2 kM 2 dx dP = =− =− = − 4f P ρ U 2 M2 2 (1 − kM 2 ) D (8.26) 4f dx = D 1/k M2 1 − kM 2 dM 2 kM 2 (8.26) that when M → √ the value k of right hand side approaches infinity (∞).24) Similarly or by other path the stagnation pressure can be expressed as a function fL of 4D 2 kM 2 1 − k+1 dx dP0 2 M = 4f k − 1 P0 D 2 (kM 2 − 1) 1 + 2 M 2 (8.25) The variables in equation (8. Heat transfer has a limited value therefore the model of the flow must be changed. the term on velocity in the tube isn’t zero). 1 Further it can be noticed from equation (8. A more appropriate model is an adiabatic flow model yet it can serve as a bounding boundary (or limit).10).23) Equation (8.27) It can be noticed that at the entrance (x = 0) for which M = Mx=0 (the initial fL is positive for any x. DIMENSIONLESS REPRESENTATION After substituting the velocity (8. one can obtain − dP 4f dx − P DH kP M 2 2 = kP M 2 dP P 141 (8. thus.

142 CHAPTER 8.24) 4f L D = 4f Lmax D 1 − 4f Lmax D 2 = 1 − kM1 2 1 − kM2 2 − + ln kM1 2 kM2 2 M1 M2 2 (8. =⇒ Reusing the perfect–gas relationship ρ 1 P = ∗ =√ ∗ P ρ kM Now utilizing the relation for stagnated isotropic pressure one can obtain P0 P ∗ = P∗ P0 Substituting for P P∗ (8.29) is transfered into √ U kM = U∗ Utilizing the continuity equation provides ρU = ρ∗ U ∗ .35) are presented on in Figure (8. ISOTHERMAL FLOW Rearranging equation (8. From equation (8.37) .32) and rearranging yields 2k 3k − 1 k k−1 P0 1 √ ∗ = P0 k 1+ k−1 2 M 2 k k −1 1 M (8.36) For the case that M1 >> M2 and M1 → 1 equation (8.36) is reduced into the following approximation ∼0 4f L D = 2 ln M1 − 1 − 1 − kM2 2 kM2 2 (8. Denote 1 and 2 as the conditions at the inlet and exit respectably.35) These equations (8.30) ρ 1 =√ ρ∗ kM (8.34) And the stagnation temperature at the critical point can be expressed as 1 2 T 1 + k− T0 2k 2 M = ∗ = ∗ k − 1 T0 T 3k − 1 1 + 2k 1+ k−1 2 M2 (8.31) (8.33) equation (8.3 The Entrance Limitation of Supersonic Branch Situations where the conditions at the tube exit have not arrived at the critical conditions are discussed here.32) 1+ 1 k−1 2 2 M k −1 + 2k k k−1 (8. It is very useful to obtain the relationship between the entrance and the exit condition for this case.30)-(8.2) 8.

even small 4D > 2 results in a Mach number which is larger than 4. one should expect that the isothermal flow should have similar characteristics as .4 Comparison with Incompressible Flow The Mach number of the flow in some instances is relatively small. This velocity requires a large entrance length to achieve good heat transfer.2: Description of the pressure. 8. ρ/ρ and T0/T0 as a function of M 1e+02 4fL  D P or  ρ  * ∗ P ρ T0/T0 P0/P0 1 * * * * * 1e+01 0.4.01 1 Mach number 10 Fig. Yet this model provides the directions of the heat transfer effects on the flow. COMPARISON WITH INCOMPRESSIBLE FLOW 143 Isothermal Flow P/P . In reality.5. With this conflicting mechanism obviously the flow is closer to the Fanno flow model.1 Fri Feb 18 17:23:43 2005 0.8. temperature relationships as a function of the Mach number for isothermal flow Solving for M1 results in M1 ∼ e 1 2 „ 4f L D +1 « (8. 8. In these cases.38) This relationship shows the maximum limit that Mach number can approach when fL the heat transfer is extraordinarily fast.1 0.

P1 − P2 /P1 .43) 4f L D = 1 2 1 − (1 − χ) − ln kM1 2 1 1−χ 2 2 (8. For incompressible flow.40) Substituting this expression into (8. From equation (8. To obtain a similar expression for isothermal flow.46) .42) Now equation (8.41) Because f is always positive there is only one solution to the above equation even though M2. a relationship between M2 and M1 and pressures has to be derived.44) 4f L D = 1 2χ − χ2 − ln kM1 2 1 1−χ (8.39) one can obtained that M2 = M 1 P1 P2 (8. denote χ= P1 − P 2 P1 (8. Expanding the solution for small pressure ratio drop.45) now we have to expand into a series around χ = 0 and remember that f (x) = f (0) + f (0)x + f (0) x2 + 0 x3 2 (8.39) fL represent the ratio of Now note that for incompressible flow U1 = U2 = U and 4D the traditional h12 .144 CHAPTER 8. by some mathematics.40) yields 4f L D = 1 kM1 2 1− P2 P1 2 − ln P2 P1 2 (8. the pressure loss is expressed as follows P1 − P 2 = 4f L D U2 2 (8.41) can be transformed into 4f L D 1 = kM1 2 1− P2 − P 1 + P 1 P1 2 − ln 1 P2 P1 2 (8. ISOTHERMAL FLOW incompressible flow.

51) value of The pressure difference can be plotted as a function of the M1 for given 4f L D .47) similarly it can be shown that f (χ = 0) = 1 equation (8.8. The current explanation is correct but based on hands waving and definitely does not satisfy the author. Equation (8. and is based on estimates.51) can be solved explicitly to produce a solution for χ= 1 − kM1 2 − 1 + kM1 2 kM1 2 4f L 1 − kM1 2 − 1 + kM1 2 1 + kM1 2 D (8.5 Supersonic Branch Apparently. A dimensionless analysis4 demonstrates that all the common materials that the author is familiar which creates a large error in the fundamental 4 This dimensional analysis is a bit tricky.52). .52) A few observations can be made about equation (8.48) rearranging equation (8. Currently and ashamedly the author is looking for a more simplified explanation.5.50) in cases that χ is small 4f L D ≈ χ 2(1 − kM1 2 ) − 1 + kM1 2 χ kM1 2 (8.48) yields 4f L D = χ (2 − χ) − kM1 2 (2 − χ) + f χ3 kM1 2 (8.45) now can be approximated as 4f L D = 1 (2χ − χ2 ) − 2χ − χ2 + f χ3 kM1 2 (8.49) and further rearrangement yields 4f L D = χ 2(1 − kM1 2 ) − 1 + kM1 2 χ + f χ3 kM1 2 (8. 8. this analysis/model is over simplified for the supersonic branch and does not produce reasonable results since it neglects to take into account the heat transfer effects. SUPERSONIC BRANCH and for example the first derivative of d ln dχ 2 145 1 1−χ 2 = χ=0 χ=0 (1 − χ) × (−2)(1 − χ)−3 (−1) = 2 (8.

Combine this information with our case of 4f L D = 10 Lentrance = 250746268. . In the supersonic flow. This requires that the entrance length to be larger than the actual length of the tub for air.).005 results in Lmax 10 = = 500 D 4 × 0.7 D On the other hand a typical value of friction coefficient f = 0. (0. So. However.0). Most of the heat transfer is hampered in the sublayer thus the core assumption of isothermal flow (not enough heat transfer so the temperature isn’t constant) breaks down5 . Nevertheless. ν . a gas flows in a tube with 4D number is over 200. over hundred of meters per fL = 10 the required entry Mach second. even fL for relatively small tubes with 4D = 10 the inlet speed is over 56 [km/sec].6 5 see Figures and Tables Kays and Crawford “Convective Heat Transfer” (equation 12-12). Almost all the perfect gas model substances dealt with in this book. Remember from Fluid Dynamic book Lentrance = 0. 0 Now with limitation.0000185 kg/m-sec at 300K and 0.005 The fact that the actual tube length is only less than 1% of the entry length means that the assumption is that the isothermal flow also breaks (as in a large response time).06 UD ν (8. For this illustration. the hydraulic entry length is very large as will be shown below. even with low temperature like 200K the speed of sound of air is 283[m/sec]. The flow speed at the entrance is very large. are 0.71.53) The typical values of the the kinetic viscosity. For example. for most gas cases the speed of sound is about 300[m/sec]. The thermal entry length is in the order of the hydrodynamic entry length (look at the Prandtl number. value for the common gases. ISOTHERMAL FLOW assumption of the model and the model breaks. 8. Now. if Mach number is changing from 10 to 1 the kinetic energy change is T0 about T ∗ = 18.37 which means that the maximum amount of energy is insufficient. this model can provide a better understanding to the trends and deviations of the Fanno flow model.146 CHAPTER 8. this topic will be covered in the next version because it provide some insight and boundary to the Fanno Flow model.0000130034 kg/m-sec at 200K. the feeding diverging nozzle somewhat reduces the required entry length (as opposed to converging feeding). the speed of sound is a function of temperature. For example.

80732 0.3906 8.2258 3.021 1.007 1.04000 0.8781 1.1718 21.98982 0.8493 7.35000 0. friction coefficient.06 192.1289 16.032 1.2074 1.33 279.00626 0.80000 0.4086 1.2553 10.9031 14.83637 0.87612 0.93800 0.).8172 2.3806 2.4515 4.09000 0.88200 0.005 1.0 28.9682 1.000 17.0255 1. ISOTHERMAL FLOW EXAMPLES Table 8.87528 0.40000 0.5920 6.1827 1.87563 0.056 1.6109 8.021 1.5644 9.8781 1.1599 13. Here is an example.87642 0.011 1.000896 0.70000 0.06000 0.3524 1. resistance (diameter.99741 1.00371 0.3906 8.96075 0.003 1.6470 1.91044 0.94894 0.82758 0.4086 1.7230 2.4147 2.1269 1.87586 0.4147 2.99232 0. the “engineering” or practical questions can be divided into driving force (pressure difference).25000 0.7040 66.8650 3.9181 5.03095 0.7 Isothermal Flow Examples There can be several kinds of questions aside from the proof questions6 Generally.6903 1.009 1. etc.90300 0.30000 0.50207 0.1259 1. .0736 10.000220 0. friction factor.9747 7.05000 0.000 28.1129 1.3334 2.89075 0.97344 0.65000 0.043 1.89 82.98700 0.6500 5.81000 0.87516 0.2258 3.89644 0.3002 1.000 8.00205 0.12 139.2074 1.92794 0.4784 1.5366 1.87675 0.97 439.8791 1.2565 1.20000 0.08000 0.5366 1.07000 0.55000 0.08085 0.0823 1.03000 0.043 1.29895 0.88594 0.1129 1.8172 2.75000 0.9031 14. 7 Those who are mathematically inclined can include these kinds of questions but there are no real world applications to isothermal model with shock.3806 2.7.60000 0.79 105.4515 4.056 1.1289 16.81879 0.5644 9.1: The Isothermal Flow basic parameters 147 M 0.10000 0.0677 1. In this model no questions about shock (should) exist7 .0859 12.1269 1.0495 1. The driving force questions deal with what should be the pressure difference to obtain certain flow rate.0859 12. and mass flow rate questions.84515 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 785. 6 The proof questions are questions that ask for proof or for finding a mathematical identity (normally good for mathematicians and study of perturbation methods). These questions or examples will appear in the later versions.3002 1.6651 13.50000 0.1718 21.8.0736 10.9925 4.001 1.011 1.91875 0.87544 0.2126 1.032 1.000 0.45000 0.2668 0.99485 0.16552 0.6903 1.

04331 400. R = 290 Kkg .0 m ˙ = kg sec P1 P1 P1 kU kU A =√ = AkM1 A√ RT k c kRT kRT . To solve this problem the flow rate has to be calculated as m ˙ = ρAU = 2.5921 0.25 Utilizing Table (8. S OLUTION If the flow was incompressible then for known density. In incompressible flow. The velocity of the gas at the entrance U = cM = 0. fL Calculating the resistance. What should be the pump pressure so that a flow rate of 2 [kg/sec] will be achieved? Assume that friction factor f = 0. ISOTHERMAL FLOW Example 8.31.e.04331 × √ m .0055000 = 400 0. ρ. surroundings temperature 27◦ C.005 and the exit pressure is 1[bar]. The exit Mach number is not necessarily 1/ k i.00 20. Hint: calculate the maximum flow rate and then check if this request is reasonable.25)2 kg × 14.62 ∼ = 16. First.31 × 290 × 300 ∼ = 14. the velocity can be calcufL U2 function of lated by utilizing ∆P = 4D 2g . 017. It is more appropriate to assume an isothermal model hence our model is appropriate. The density reads 1.25 [m] diameter and 5000 [m] in length is attached to a pump. check whether flow is choked (or even possible).148 CHAPTER 8.19 RT 290 × 300 m3 π × (0.1: A tube of 0. 450 ∼ kg = = 23. It is note worthy to mention that since the isothermal model breaks around the choking point.62 sec ρ= P 2.1) or the program provides M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 0. the flow rate is really some what different. the density is a √ the entrance Mach number. the flow is not choked.1743 12.0 0.9 4 sec The maximum flow rate then reads m ˙ = ρAU = 23. 4D 4f L D = 4 × 0. k = 1. The specific J heat for the gas.89446 The maximum flow rate (the limiting case) can be calculated by utilizing the above table.19 × The maximum flow rate is larger then the requested mass rate hence the flow is not choked.

68 The pressure should be 21.10300 4fL D P P∗ 2 × 337. ISOTHERMAL FLOW EXAMPLES Now combining with equation (8. Note that tubes are provided in increments of 0.5 [in]8 .7.5844 0. Assume that the flow is isothermal and k=1. You can assume that the soundings temperature to be 27◦ C.1) or using the provided program yields 8 It is unfortunate.25)2 P2 Ak × 1 . the maximum M1 that can be obtained when the M2 is at its maximum and back pressure is at the atmospheric pressure.4826 5.).5 [km] (500 [m]). the maximum pressure allowed for the gas is only 10[bar].2 [kg/sec] is required. 31 100000 × π×(0 4 P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 66. S OLUTION At first.2: A flow of gas was considered for a distance of 0.76780 × 8.8.0 0. Due to safety concerns.103 .7678 13.566[bar] Example 8. but it seems that this standard will be around in USA for some time. Mmax 1 1 P2 = √ = 0. with the value of M1 either by utilizing Table (8.89567 The entrance Mach number is obtained by 4f L D 1 = 66.68 P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ M 4fL D 0.31 P = 21.0 0.6779 + 400 ∼ = 466. The friction coefficient for the tube can be assumed as 0.04014 466. calculate the required diameter of tube.3249 0.0845 M1 = M 2 P1 k 10 Now.4826 = 2.02 (A relative smooth tube of cast iron.1) or by utilizing the program M 0. the minimum diameter will be obtained when the flow is choked. A flow rate of 0. .40) yields m ˙ = M2 = M2 P2 Ak c 149 From Table (8. Thus.59 mc ˙ = = 0.89442 Note that tables in this example are for k = 1.6779 8.4.

9110 6.4 × 287 × 300 ∼ = 347.0018 6.4318 P P∗ P0 P0 ∗ 92. 000 kg P = = 11. the pipes are provided only in 0.2 the mass flow rate requirement is satisfied.08528 × 347.4318[m].02 × 500 0. pressure and fL is etc) have to be taken at the same point. The new 4D 4f L D = 4 × 0.02 × 500 94.2 sec and the density is ρ= 1. The speed of sound at the entrance is √ √ m c = kRT = 1.08527 4fL D 92.61 RT 287 × 300 m3 The velocity at the entrance should be m U = M ∗ c = 0.4310 10.431 the value of minimum diameter.42359[m] = 16.6 sec . The only point that must be emphasized is that all properties (like Mach number.2424 0. It should be noted that P should be replaced by P0 in the calculations.150 M 0.5 increments and the next size is 17[in] or 0. D= 4f L 4f Lmax D 4 × 0.2 ∼ = 29. 000.6400 9.3[kg/sec] Since 50.0 0. ISOTHERMAL FLOW P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 94.87627 To check whether the flow rate satisfies the requirement m ˙ = 106 × π ×0.0 0.3 ≥ 0. With this pipe size the calculations are to be repeated in reverse and produces: (Clearly the maximum mass is determined with) √ √ P P AM k m ˙ = ρAU = ρAM c = AM kRT = √ RT RT The usage of the above equation clearly applied to the whole pipe.2991 0.43182 4 × 0.08450 With 4f Lmax D 4fL D P P∗ CHAPTER 8.64 ρ ρ∗ T0 T0 ∗ M 0.0853 × √ 287 × 300 √ 1.43 0.68[in] However.4 ≈ 50.87625 = 94.

1.01 × 4000 = 400 0.21[bar] 20.67 2.6 × 11.66915 From the table M1 ≈ 0.52828.17[bar] As the pressure at point (b) is smaller than the actual pressure P ∗ < P2 than the actual pressure one must conclude that the flow is not choked.3: A gas flows of from a station (a) with pressure of 20[bar] through a pipe with 0. Calculate the Mach number at the entrance to pipe and the flow rate. ISOTHERMAL FLOW EXAMPLES The diameter should be D= 4m ˙ = πU ρ 4 × 0. It should be noted that the flow isn’t choked. at first it will be assumed that the whole length is the maximum length.7.027 π × 29.2 ∼ = 0. Example 8.87531 P0 P0 ∗T 20.0419 .4[m] diameter and 4000 [m] length to a different station (b). for the sake of the exercise the other parameters will be calculated.4.61 151 Nevertheless. Therefore.19235 12.8.and P0 ∗T ∼ = 28 12. and the average friction f=0.21 × 0. guess reasonable value of M1 and calculate 4f L D .4 with 4f Lmax D = 400 the following can be written 4f L D T0 T0 ∗T ρ ρ∗T P P∗T P0 P0 ∗T M 0. S OLUTION First.0419 400. P2 = P0 ∗T P2 P0 ∗T = 2. The flow rate is given with the diameter of the pipe. Assume that the flow is isothermal.52828 = 1.19235 ≈ 12. 4f Lmax D = 4 × 0. the information whether the flow is choked needs to be found. The gas and the sounding temperature can be assumed to be 300 K. k=1. This situation is reversed question.01.72021 0. The pressure at the exit (station (b)) is 2[bar].67 The pressure at point (b) by utilizing the isentropic relationship (M = 1) pressure ratio is 0. The solution is an iterative process.

59338 4f Lmax D 1 4f L D P2 P1 400. Calculate the value of 4f L D 2 CHAPTER 8.10000 0.16922 0.22 × 0.0419 The flow rate is √ √ 2000000 1.46[kg/sec] In this chapter.10000 8.7272 0.08978 0.32807 0.10000 0.36766 0. 4.05000 0. Compare the results of guessed pressure P2 with the actual pressure and choose new Ma number M1 accordingly.32131 400.83920 0.6934 0.5807 0.36780 0. M2 0.40737 0.152 2.83889 0.8 Unchoked situation Table 8.05005 0.08971 0.10000 0.4 P k π × D2 M= √ m ˙ = ρAM c = √ π × 0.84018 0.16912 0.32795 0.12942 0. P2 bear in mind that this isn’t the real pressure but based on assumption 5. there are no examples on isothermal with supersonic flow. Calculate the pressure.10000 .0419 4 300 × 287 RT 42.83827 0.00000 0. Now the process has been done for you and is provided in Figure (??) or in the table obtained from the provided program.84095 0.10000 0.10000 0.12949 0.5708 M2 0.83740 4f Lmax D 1 4f L D P2 P1 0.4: The flow parameters for unchoked flow M1 0.10000 0.6684 0. ISOTHERMAL FLOW by subtracting 4f L D 1 − 4f L D 3.40754 0.6483 0.5914 0. Obtain M2 from the Table ? or by using the Potto–GDC. M1 0.83997 0.

10 Fri Feb 25 17:20:14 2005 Fig.3 0.6 0.7 0.3: The Mach number at the entrance to a tube under isothermal flow model as a fL function 4D .1 0 0 10 20 30 40 50 4fL  D 60 70 80 90 100 M1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.2 = 0.8.5 = 0. UNCHOKED SITUATION 153 M1 isothermal flow 1 0. 8.5 0.8 = 0.8.9 0.8 0.2 0.4 0.

ISOTHERMAL FLOW .154 CHAPTER 8.

± 9. etc. Later. This model explains many industrial flow processes which includes emptying of pressured container through a relatively short tube. compressed air systems. plicable to flow processes which are No heat transer very fast compared to heat transfer mechanisms with small Eckert Fig. °.1 Introduction Consider a gas flowing through a conduit with a friction (see Figure (9. 9. It is advantages to examine the simplest situation and yet without losing the core properties of the process. exhaust system of an internal combustion engine.³ can be ignored 1 . 1 Even 2 Not the friction does not convert into heat ready yet.1)). 155 . As this model raised from need to explain the steam flow in turbines.CHAPTER 9 Fanno Flow An adiabatic flow with friction is ¤ ›8œžŸ› named after Ginno Fanno a Jewish “ flow  ¢¡ž£u  direction engineer. ”•!–˜—D™„š © ¦¨§ ©«ª~¬­ ¦¨§ ¯ ­ ® The main restriction for this model is that heat transfer is negligible and ². This model is the second ¥¦¨§Ÿ¥ pipe flow model described here.1: Control volume of the gas flow in a constant cross section number. discussed on the ideal gas model and the entry length issues. more general cases will be examined2 . This model is apc.v.

7) (9. FANNO FLOW 9.6) 3 The equation of state is written again here so that all the relevant equations can be found when this chapter is printed separately.5) It is assumed that the flow can be approximated as one–dimensional. .1) The mass (continuity equation) balance can be written as The energy conservation (under the assumption that this model is adiabatic flow and the friction is not transformed into thermal energy) reads T0 1 = → T1 + Or in a derivative form Cp dT + d U2 2 = 0 (9.3) Again for simplicity.2) (9. P = ρRT P2 P1 = → ρ 1 T1 ρ 2 T2 (9.4) U1 = 2cp 2 T0 2 T2 + U2 2cp 2 (9.156 CHAPTER 9.2 Model m ˙ = ρAU = constant → ρ 1 U1 = ρ 2 U2 (9. The force acting on the gas is the friction at the wall and the momentum conservation reads −AdP − τw dAw = mdU ˙ It is convenient to define a hydraulic diameter as DH = Or in other words A= πDH 2 4 (9. the perfect gas model is assumed3 .8) 4 × Cross Section Area wetted perimeter (9.

11) by the cross section area.6) yields A τw 2 m ˙ A − πD dP − πDdx f 4 1 2 ρU 2 = A ρU dU (9.11) yields Dividing equation (9. A and rearranging 4f dx D 1 2 ρU 2 −dP + = ρU dU (9. By utilizing the definition of the sound speed to produce the following identities for perfect gas M2 = U c 2 = U2 k RT P ρ (9.10) By utilizing equation (9.12) The second law is the last equation to be utilized to determine the flow direction.2) and substituting equation (9.15) . NON–DIMENSIONALIZATION OF THE EQUATIONS 157 It is convenient to substitute D for DH and yet it still will be referred to the same name as the hydraulic diameter. The infinitesimal area that shear stress is acting on is dAw = πDdx (9.9) Introducing the Fanning friction factor as a dimensionless friction factor which is some times referred to as the friction coefficient and reads as the following: f= τw 1 2 ρU 2 (9.13) 9.10) into momentum equation (9.3.9. s2 ≥ s 1 (9.14) Utilizing the definition of the perfect gas results in M2 = ρU 2 kP (9.3 Non–dimensionalization of the equations Before solving the above equation a dimensionless process is applied.

14) yields d(M 2 ) d(U 2 ) dT = − 2 M U2 T (9.4) by Cp and by utilizing the definition Mach number yields dT + T 1 kR (k − 1) Cp 1 U2 d T U2 U2 2 = → (k − 1) U 2 dT + d T kRT U 2 c2 U2 2 = k − 1 2 dU 2 dT + M =0 → T 2 U2 This equation is obtained by combining the definition of Mach number with equation of state and mass conservation.18).19). (9.14) and dividing by equation (9.19) Derivation of the Mach identity equation (9.20) Dividing the energy equation (9. and (9.5) results dρ dT dP = + P ρ dT (9.17).14) and substituting it into equation (9. Derivative of mass conservation ((9.11) and after some rearrangement yields ρU 2 4f dx −dP + DH 1 kP M 2 2 ρU 2 dU = dU = kP M 2 U U (9. FANNO FLOW Using the identity in equation (9.2)) results in dU U dρ 1 dU 2 + =0 ρ 2 U2 (9.158 CHAPTER 9. (9.16) results in − dP 4f dx − P D kM 2 2 = kM 2 dU U (9. (9. the original limitations must be applied to the resulting equation.5) and dividing the results by equation of state (9.18) The derivation of the equation of state (9.16) By further rearranging equation (9.17).17) It is convenient to relate expressions of (dP/P ) and dU/U in terms of the Mach number and substituting it into equation (9.21) Equations (9. These equa- . (9. Thus.20).21) need to be solved.

20) and (9.22).22) and rearrangement yields 1 + (k − 1)M 2 dU 2 dP =− P 2 U2 The term dU 2 /U 2 can be eliminated by using (9.18) and equation (9.18) to obtain dρ 4f dx kM 2 =− 2 ρ 2 (1 − M ) D (9.28) .3.21) when eliminating dT /T results dP dρ (k − 1)M 2 dU 2 = − P ρ 2 U2 (9. NON–DIMENSIONALIZATION OF THE EQUATIONS 159 tions are separable so one variable is a function of only single variable (the chosen as the independent variable). causes the change in the other variables.24) (9.25) results in 1 2 kM 2 1 + k− 4f dx dM 2 2 M = 2 2 M 1−M D (9.22) The term dρ ρ can be eliminated by utilizing equation (9. Explicit explanation is provided for only two variables. sistance. 4D Combining equations (9.18) and substituting it into equation (9. The only variable that is left is P (or dP/P ) which can be eliminated by utilizing equation (9. M variable is obtained by combining equation (9. The dimensionless friction.23) The second equation for Mach number. fL the rest variables can be done in a similar fashion. 4D .23) kM 2 1 + (k − 1)M 2 4f dx dP =− P 2(1 − M 2 ) D (9.25) (9.27) and (9.26) After similar mathematical manipulation one can get the relationship for the velocity to read dU kM 2 4f dx = U 2 (1 − M 2 ) D (9.27) and the relationship for the temperature is density is obtained by utilizing equations (9.9.19) and (9.29) 1 dc k (k − 1)M 4 4f dx dT = =− T 2 c 2(1 − M 2 ) D (9. is chosen as the independent variable since the change in the dimensionless refL .24) and results in 1 − M 2 dM 2 4f dx = 1 2 D kM 4 (1 + k− 2 M ) Rearranging equation (9. Then dρ/ρ and U are eliminated by utilizing equation (9.21) by eliminating dT /T .

(dT0 = 0). When M < 1 the pressure decreases downstream as can be seen from equation (9. in the supersonic branch. M > 1.24) through (9. Thus the discussion will be divided into two cases: One.34) 9.24) it can be observed that the critical point is when M = 1. flow with speed below the speed of sound.31) The stagnation temperature expresses as T0 = T (1 + (1 − k )/2M 2 ).34).160 The stagnation pressure is similarly obtained as dP0 kM 2 4f dx =− P0 2 D The second law reads ds = Cp ln CHAPTER 9. Two.32) In similar fashion the relationship between the stagnation pressure and the pressure can be substituted into the entropy equation and result in ds = Cp ln dT0 dP0 − R ln T0 P0 (9.4 The Mechanics and Why the Flow is Choked? The trends of the properties can be examined by looking in equations (9. FANNO FLOW (9.30) dP dT − R ln T P (9. Taking derivative of this expression when M remains constant yields dT0 = dT (1 + (1 − k )/2M 2 ) and thus when these equations are divided they yield dT /T = dT0 /T0 (9.35) (9. For the same reasons. . flow above speed of sound. This pressure increase is what makes compressible flow so different from “conventional” flow. from equation (9.24) because f dx and M are positive.33) The first law requires that the stagnation temperature remains constant. the pressure increases downstream. For example. Therefore the entropy change is ds (k − 1) dP0 =− Cp k P0 Using the equation for stagnation pressure the entropy equation yields ds (k − 1)M 2 4f dx = Cp 2 D (9.

It demonstrates that the Mach number increases downstream when the flow is subsonic. T Density. when the flow is supersonic. First.26).37) . M = 1 at the last point. the explanation is based on the equations developed earlier and there is no known explanation that is based on the physics. For example. This situation is called choked flow. On the other hand. the pressure has to be a monotonic function which means that flow cannot crosses over the point of M = 1.5 The working equations 4 D Lmax k+1 2 1 1 − M2 k+1 2 M + ln k − 1 k M2 2k 1 + 2 M2 Integration of equation (9. For example.24) that increase or decrease from subsonic just below one M = (1 − ) to above just above one M = (1 + ) requires a change in a sign pressure direction.5. M Velocity. the pressure decreases. P Mach number.24)).36) A representative friction factor is defined as ¯= f 1 Lmax 0 Lmax f dx (9. The Trends The trends or whether the variables are increasing or decreasing can be observed from looking at the equation developed.25) yields f dx = L (9. dP (@M = 1) = ∞ and mathematically it is a singular point (see equation (9. U Temperature. the pressure can be examined by looking at equation (9. Observing from equation (9. T0 9. However. This constrain means that because the flow cannot “crossover” M = 1 the gas has to reach to this speed. it has to be recognized that the critical point is when M = 1 it will show a change in the trend and it is singular point by itself. THE WORKING EQUATIONS Why the flow is choked? 161 Here. ρ Stagnation Temperature. The summary of the properties changes on the sides of the branch Subsonic decrease increase increase decrease decrease decrease Supersonic increase decrease decrease increase increase increase Pressure.9.

(9.36) yields k+1 2 ¯ max 1 1 − M2 k+1 4fL 2 M = + ln 1 2 D k M2 2k 1 + k− 2 M (9. Now equation (9.24) in equation (9.23) is represented by 4D 4f L and Mach number. (9. and (9.162 CHAPTER 9. For example.41) ρ 1 = ρ∗ M 1+ k−1 2 2 M k+1 2 (9.39) can be integrated to yield: 1 P = ∗ P M k+1 2 k −1 2 2 M 1+ (9.39) The symbol “*” denotes the state when the flow is choked and Mach number is equal to 1. Thus.24). (9.29).42) U = U∗ ρ ρ∗ −1 =M 1+ k+1 2 k−1 2 2 M (9.30) can be solved. fL .43) The stagnation pressure decreases and can be expressed by k 2 k−1 (1+ 1− 2 M ) k P0 = P0 ∗ P0 P P0 ∗ P∗ 2 ( k+1 ) k−1 k P P∗ (9. FANNO FLOW By utilizing the mean average theorem equation (9.40) In the same fashion the variables ratio can be obtained k+1 c2 T 2 = = 1 2 T∗ c∗ 2 1 + k− 2 M (9. the pressure as written in equation (9.26) yields dP P dM 2 M2 =− 1 + (k − 1M 2 dM 2 1 2 M 2M 2 1 + k− 2 (9.29). (9.28).27).24) can eliminate term D and describe the pressure on the Mach number.38) ¯ with f which is adopted in this book. Dividing equation (9. M = 1 when P = P ∗ Equation (9.44) . It is common to replace the f Equations (9.

46) The integration of equation (9. THE WORKING EQUATIONS 163 Using the pressure ratio in equation (9.1 0. 9.44) yields P0 = P0 ∗ 1+ k−1 2 2 M k+1 2 k k−1 1 M 1+ k−1 2 2 M k+1 2 (9.34) yields s − s∗ = ln M 2 cp k+1 1 2 1 + k− 2 M k+1 k 2M 2 (9.01 1 Mach number 10 Fig.1 Fri Sep 24 13:42:37 2004 0.47) discussion about Reynolds number and dimensionless 4f L friction parameter.45) provides 1 P0 ∗ = P0 M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (9. ρ/ρ and T/T as a function of M 1e+02 4fL  D P  * P * T0/T0 P0/P0 U/U* * * * Fanno Flow * 1e+01 1 0.2: Various parameters in Fanno flow as a function of Mach number many cases shockless and therefore a relationship between two points should be .45) And further rearranging equation (9.5. D The results of these equations are plotted in Figure (9.40) and substituting it into equation (9.2) The Fanno flow is in P/P .9.

The real ratio can be obtained by two star ratios as an example T2 = T1 T T ∗ M2 T T ∗ M1 (9. FANNO FLOW derived. S OLUTION For isentropic.05 [m] and áãâåäÑæ©ç è§é length of 10 [m]. the flow to the pipe inlet. Assume that the average friction factor to be f = 0.94 . The following conditions preԐÕAÖ·×sØsÙ Ú vail at the exit: P2 = 1[bar] temperature Ê È ÉAËÍÌDÎ ÏÑÐDÒsÓ T2 = 27◦ C M2 = 0.1291 4 This 1. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.164 CHAPTER 9.9) the following is obtained.004 and that the flow from the reservoir up to the Fig. ¾ 5 ¿ À form pipe with a diameter of 0. Estimate the total temperature and total pressure in the reservoir under the Fanno flow model. . For given Mach number (M = 0.49) Hence. The air exits to the atÁÃÂ.0934 0.3: Schematic of Example (9. the temperature and the total pressure at the pipe inlet are the same as those in the reservoir.1) pipe inlet is essentially isentropic. Thus. 4f Lmax D = 2 4f Lmax D 1 − 4f L D (9.6 Examples of Fanno Flow Example 9.01451 1. There is no Mach meter. the total temperature at the entrance can be fL obtained by knowing the 4D .1: Û£ÜAÝ·Þß à Air flows from a reservoir and enters a uni´€µ·¶¸ ¶s¹©º »§¼ 0 ½ . the “star” values are imaginary values that represent the value at choking.90000 0.50) 9.0089 1. finding the total pressure and temperature at the pipe inlet is the solution.48) A special interest is the equation for the dimensionless friction as following L2 L1 4f L dx = D Lmax L1 4f L dx − D Lmax L2 4f L dx D (9. 9. With the Mach number and temperature known at the exit.Ä_ÅMÆ Ç mosphere. In most times.9146 1.0327 property is given only for academic purposes.

05 4f L D The rest of the parameters can be obtained with the new (9.2100 3.9.6.01451 0.35886 3.915 T1 T ∗ T01 = T2 T T0 ∗ 2 T 1 T 1 1 1 =300 × × 1.1699 Note that the subsonic branch is chosen.975 T∗ T 348K = 75◦ C Another academic question: . M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.7405 2.0327 0. EXAMPLES OF FANNO FLOW So.5[K ] 1.38814 1.17 × 1.7405 0.004 × 10 + 0. The stagnation ratios has to be added for M = 0.5764 0.93840 1.0327 4f L D To ”move” to the other side of the tube the 4f L D is added as 3.21 either from Table 1 = 4f L D + 4f L D 2 = 4 × 0. the total temperature at the exit is T ∗ |2 = T∗ T T2 = 2 165 300 = 290.12913 0.35886 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.97489 0.91484 1.35886 0.91[Bar] 1.5922 0.1) by interpolations or by utilizing the attached program.0140 1.014 × = 2.78305 The total pressure P01 can be found from the combination of the ratios as follows: P1 P ∗ P01 = P2 P∗ P 2 P P∗ 1 P0 P 1 1 1 =1 × × 3.

400 K. The vessel conditions are at 29. With these conditions a pipe inlet Mach number is 3. and CHAPTER 9. The system is supplied by a vessel.166 Example 9.0 [m] long.5 [cm] in diameter and 1. determine: (a) the mass flow rate through the system.0. (b) the temperature at the pipe exit. FANNO FLOW ó©ô2õö ÷Mø ùHúMû ü ýMþ ÿ  ¢¡¤£¦¥¨§©§¨ ¦.65 [Bar].2: A system is composed of a convergentdivergent nozzle followed by a tube with length of 2. A normal shock wave occurs in the tube and the flow discharges to the atmosphere.

It is further assumed t the flow through the nozzle can be approximated as isentropic. Take k = 1.1) or the Potto–GDC T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ M 3.2346 0.02722 0.11528 0. ¦  "!"#  shock d-c nozzle êãë§ìMí ìHîOïMð ñ_ò    atmosphere conditions Fig.2) (c) determine the Mach number when a normal shock wave occurs [Mx ].4.0000 The temperature is 0. the stagnat conditions can be approximated for the condition in the tank.005.07623 4. Thus. 9. then the following can be found eith from Table (9. m ˙ = ρAM c The density and speed of sound are unknowns and need to be computed.35714 0.4: The schematic of Example (9. therefore the velocity the vessel can be assumed to be small enough so it can be neglected. With the isentro relationship the Mach number at point one (1) is known. R = 287 [J/kgK ] and f = 0.8K T01 .357 × 400 = 142. S OLUTION (a) Assuming that the pressure vessel is very much larger than the pipe. T01 = 400K a P01 = 29. Hence.65[P ar] The mass flow rate through the system is constant and for simplicity point 1 is chosen which.65326 T1 = T1 T01 = 0.

(Fanno flow Table (9.8 > 0.1 × 104 287 × 142.3333 0.52216 the flow is choked and with a shock wave.9.1)) Mx 3.81[Bar] 167 The density as a function of other properties at point 1 is ρ1 = P RT = 1 The mass flow rate can be evaluated from equation (9.0252 × 3 × 239. EXAMPLES OF FANNO FLOW Using the temperature. by comparing “possible” Pexit to PB .8 239.0 = 0.8571 10.52216 0.50918 1. the possibility that the shock wave occurs immediately at the entrance for which the ratio for Mx are (shock wave Table (5. The exit pressure determines the location of the shock.9640 0.54 = 0.1)) .2346 0.8 0.97 × kg π × 0. check whether the flow is shockless by comparing the flow resistance and the maximum possible resistance. one.025 Since 0.32834 After shock wave the flow is subsonic with “M1 ”= 0.4 × 287 × 142.6. and two.005 × 1. the shock at the entrance of the tube. Two possibilities are needed to be checked. the speed of sound can be calculated as √ √ c1 = kRT = 1. if a shock exists.2) m ˙ = 1.0000 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.1) or by using the Potto–GDC.47519 Ty Tx ρy ρx Py Px P0 y P0 x 2.47519.8 1. First. shock at the exit and comparing the pressure ratios.42857 and the conditions of the tube are 4f L D = 4 × 0.54[m/sec] The pressure at point 1 can be calculated as P1 = P1 P01 = 0.21822 4.0000 My 0.97 kg m3 (b) First.027 × 30 P01 0. From the Table (9.69 4 sec 8.6790 3. to obtain the following M 3.

0338 is smaller than the case in which shock occurs the entrance. To check Secant Method. One possible way to find the e temperature.9640 0. . The procedure for the calculations: 1) Calculate the entrance Mach number assuming the shock occurs at the exit: a) set M2 = 1 assume the flow in the entire tube is supersonic: b) calculated M1 Note this Mach number is the high Value. 2) Calculate the entrance Mach assuming shock at the entrance.32834 × 1 2.65 = 0.3904 1. 5 You can use any method you which.2549 1.2919 2.65326 The ratio of exit pressure to the chamber total pressure is 1 1 ∗ P2 = P0 = = P2 P∗ P0 y P P0 x P1 P1 P0 y P0 x P0 1 1× × 0.50917 1. P P1 is needed. the shock is somewhere downstream. calculate the parameters for shock locat fL with known 4D in the “y” side.168 M 4fL D P P∗ P0 P0 ∗ CHAPTER 9.47519 1. a) set M2 = 1 fL and calculated M1 ’ for subsonic branch b) add 4D c) calculated Mx for M1 ’ Note this Mach number is the low Value. For example. To find the location of the shock ra 2 of the pressure ratio. Then either by utilizing shock table or the program. With the location of shock. but be-careful second order methods like Newton-Rapson method can be unstable. 3) According your root finding algorithm5 calculate or guess the shock location and th compute as above the new M1 . T2 is by finding the location of the shock.3904 0. to obt the upstream Mach number. “claiming” upstream from exit through shock to the entrance.2549 0. FANNO FLOW ρ ρ∗ U U∗ T T∗ 0.95679 0.12476 The actual pressure ratio 1/29.47519 0.1912 0.1481 The stagnation values for M = 0. Thus.47519 are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.85676 1.89545 1.8568 × 0.

It also must be recognized that there is a maximum length for which only supersonic flow can exist8 .nasa. Many of the pictures in the literature carry copyright statements.8 Maximum length for the supersonic flow To insert example on the change in the flow rate between isothermal flow to Fanno Flow. furthermore.9. on the comparison of the maximum length of isothermal model and Fanno Model. because the very limited amount of heat transformed it is closer to an adiabatic flow. Insert also example on percentage of heat transfer. The only limitation of the model is its uniform velocity (assuming parabolic flow for laminar and different profile for turbulent flow.0000 4fL D up 4fL D down Mx 1. and. SUPERSONIC BRANCH 169 a) set M2 = 1 fL b) for the new 4D and compute the new My ’ for the subsonic branch c) calculated Mx ’ for the My ’ fL and calculated the M1 d) Add the leftover of 4D 4) guess new location for the shock according to your finding root procedure and according to the result. repeat previous stage until the solution is obtained. please provide to the undersign so they can be added to this book.).57910 4f L D 0. In the Fanno model there is no heat transfer. The author seeks for a nice explanation of this concept for non–fluid mechanics engineers. up that will 9.gov/digidoc/report/tm/44/NACA-TM-844. This solicitation is about how to explain this issue to non-engineers or engineer without a proper background.57981 (c) The way of the numerical procedure for solving this problem is by finding produce M1 = 3. These results were obtained from the mathematical derivations but were verified by numerous experiments9 . 7 See on the web http://naca. 9.PDF 8 Many in the industry have difficulties in understanding this concept.9899 My 0. Nevertheless. In the process Mx and My must be calculated (see the chapter on the program with its algorithms.7. 9 If you have experiments demonstrating this point. experiments from many starting with 1938 work by Frossel7 has shown that the error is not significant. The information from the wall to the tube center6 is slower in reality.larc.). These changes include the choking point at lower Mach number.22019 0. M1 3. the comparison with reality shows that heat transfer cause changes to the flow and they need to be expected.7 Supersonic Branch In Chapter (8) it was shown that the isothermal model cannot describe adequately the situation because the thermal entry length is relatively large compared to the pipe length and the heat transfer is not sufficient to maintain constant temperature. . It has to be noted and recognized that as opposed to subsonic branch the supersonic branch has a limited length.0000 M2 1. The maximum length of the supersonic can be evaluated when M = ∞ 6 The word information referred to is the shear stress transformed from the wall to the center of the tube. However.

170 as follows: CHAPTER 9. k 9.5 1.  D . The maximum length in supersonic flow In Fanno Flow 1.3 0.7 0. FANNO FLOW k+1 2 1 − M2 k+1 4f Lmax 2 M = = + ln k − 1 2 D kM 2k 2 1 + 2 M2 4f L D (M → ∞) ∼ −∞ k + 1 (k + 1)∞ + ln k×∞ 2k (k − 1)∞ (k + 1) −1 k + 1 + ln = k 2k 2(k − 1) = 4f L D (M → ∞.3 1. 9.45 1.1 1 0.55 1. k as shown in Figure (9. it can be observed that no matter how high the entrance Mach number will be the tube length is limited and depends only on specific heat ratio.6 1.5 0. k = 1.2 1.9 0.6 0.4) = 0.25 1.8 0. k Thu Mar 3 16:24:00 2005 Fig.5).4 0.65 spesific heat.35 1.5 1. From the above analysis.4 1.8215 The maximum length of the supersonic flow is limited by the above number.4 1.5: The maximum length as a function of specific heat.9 Working Conditions It has to be recognized that there are two regimes that can occur in Fanno flow model one of subsonic flow and the other supersonic flow.1 0 1.2 0.2 1. Even the flow in the tube starts as a supersonic in parts of the tube can be transformed into the subsonic 4fLmax maximum length.3 1.

Only a combination of these two parameters is truly independent. the mass flow rate decreases. all the three parameters can be varied and they are discussed separately here.6: The effects of increase of 4f L D on the Fanno line In the analysis of this effect. Once the Mach number reaches maximum (M = 1). and the pressure ratio. no further increase of the exit Mach number can be achieved. P2 /P1 are controlling the flow. Three parameters.7). The discussion has to differentiate between two ways of feeding the tube: converging nozzle or a converging-diverging nozzle. WORKING CONDITIONS 171 branch. the entrance Mach number. the dimensionless fL friction.1 fL Variations of The Tube Length ( 4D ) Effects C8DFE H¨I :<. The . 4D . 9.9. It is worth noting that entrance Mach number is reduced (as some might explain it to reduce the flow rate). M1 .9.43 5&687 9 ' ( Fig. the treatment of the two branches are separated. it should be assumed that back pressure is constant and/or low as possible as needed to maintain a choked flow.9.102./. increasing the tube length results in increasing the exit Mach number (normally denoted herein as M2 ). In this process. However. The entrance temperature increases as can be seen from Figure (9. 9.>=? ? I I X K<L N ?@BA G J MN U¨V V V W NOBP&Q8R4S T $&% )+*-. A shock wave can occur and some portions of the tube will be in a subsonic flow pattern. First. Subsonic branch For converging nozzle feeding.

entrance and exit Mach number are discussed. is the point in which no supersonic flow is possible in the tube i. There is another point d.e. Point b is the maximum possible flow for supersonic flow and is not dependent on the nozzle.7: The development properties in of converging nozzle velocity therefore must decrease because the loss of the enthalpy (stagnation temP perature) is “used. The next point. 9. Point a is the choking point (for the supersonic branch) in which the exit Mach number reaches to one. The following ranges that has to be discussed includes (see Figure (9. the mass flow rate must decrease. At the starting point the flow is choked in the nozzle. Thus. to achieve supersonic flow. FANNO FLOW Y[Z \ 1’ 1 2 Fanno lines 2’ 1’’ 2’’ ] Fig. These results are applicable to the converging nozzle. results in a similar flow pattern as in the converging nozzle. in which no supersonic flow is possible in the entire nozzle–tube system. Between these transitional points the effect parameters such as mass flow rate.” The density decrease because ρ = RT and when pressure is remains almost constant the density decreases. the shock reaches to the nozzle. 4D the flow becomes choked a different flow pattern emerges. referred here as the critical point c. In the case of the converging–diverging feeding nozzle. Supersonic Branch There are several transitional points that change the pattern of the flow. Once less friction.172 constant pressure lines CHAPTER 9.8)): . increase of the dimensionfL .

In the range of a − b the flow is all supersonic.8: The Mach numbers at entrance and exit of tube and mass flow rate for Fanno Flow fL as a function of the 4D The 0-a range. reduces with the increase of 4D . It is worth noticing that in the a − −c the mass flow rate nozzle entrance velocity max . The mass flow rate continues to be constant. fL M1 . WORKING CONDITIONS 0 4f L D 4f L D 4f L D choking shockless chokeless 173 4f L D 4f L D 4f L D 4f L D < < < < < < < < 4f L D 4f L D 4f L D choking shockless chokeless 0→a a→b b→c c→∞ ∞ wyx a | z+{ all supersonic flow ^`_ba b mixed supersonic with subsonic flow with a shock between c d c j lonqpsrutqv km the nozzle is still choked }y~ efhg i Fig. In the next range a − −b The flow is double choked and make the adjustment for the flow rate at different choking points by changing the shock location. the entrance Mach number. M1 is constant because it is a function of the nozzle design only. M2 = 1. At the end of the on the supersonic branch) and starts ( 4D range a. 9. M2 decreases (remember this flow is fL = 0) as M2 = M1 . is only a theoretical length in which the supersonic flow can occur if nozzle is provided with a larger Mach number (a change to the nozzle area ratio which also reduces the mass flow rate).9.9. The entrance Mach continues to be constant and exit Mach number is constant. Notice that exit Mach number. fL The total maximum available for supersonic flow b − −b . The exit Mach number. However. the mass flow rate is constant because the flow is choked at the nozzle. In the range b − c. M2 is still one. The entrance Mach number. 4D . In semi supersonic flow b − −c (in which no supersonic is available in the tube but only in the nozzle) the flow is still double choked and the mass flow rate is constant. it is a more practical point.

6 0. Somewhat different then the subsonic branch the mass flow rate is constant even if the flow in the tube is completely subsonic. 9. the exit Mach number remains constant and equal to one.1 0 0 0.8 0.174 CHAPTER 9. fL M2 mass flow rate as a function of 4D .2 0. The exit Mach M2 is a continuous monotonic function that decreases with 4D The entrance Mach M1 is a non continuous function with a jump at the point when shock occurs at the entrance “moves” into the nozzle.1 0. M1 as a function of M2 4fL  = 0.5 0. 1 1 .8 0. To summarize the above discussion.3 0.9 1 Exit Mach number Tue Oct 19 09:56:15 2004 Fig.2 0.0 = 100.4 0. the resistance increases and on the other hand.7 Entrace Mach number 0.9: M1 as a function M2 for various 4f L D 10 On a personal note. fL .9 0.8) exhibits the development of M1 . this situation is rather strange to explain.1 D = 1. On one hand. In this range. the flow rate decreases since (m ˙ ∝ M1 )11 .7 0. FANNO FLOW and the exit velocity remains constant!10 In the last range c − −∞ the end is really the pressure limit or the break of the model and the isothermal model is more appropriate to describe the flow.0 Fanno Flow 1 0.6 0.3 0. Figures (9. This situation is because of the “double” choked condition in the nozzle.0 = 10.5 0.4 0. Does anyone have an explanation for this strange behavior suitable for non–engineers or engineers without background in fluid mechanics? 11 Note that ρ increases with decreases of M but this effect is less significant.

when the shock wave occurs at the tube exit and two.4 1.9. shock at the exit. For D = 0.4 two extremes are shown.5 4 3.8 2 M1 M1 as a function of M2 for the subsonic brench 4fL = 0. the maximum .6 1. Obviously.9.5 0 0 0. One. fL For a given 4D . one.2 1.5 2 1. shock at the entrance.2 shown with only fL creates shock at the exit only. tow.1 shock = 0. Obviously there can be two extreme possibilities for the subsonic exit branch. The Figure (9. at the tube entrance. The Figure was calculated by max utilizing the data from Figure (9.10) exhibits the entrance Mach number as a function of the M2 .8 1 M2 1.10: M1 as a function M2 for different 4f L D for supersonic entrance velocity. 9.4 = 0.10) only for 4f L 4f L 4f L D = 0. the supersonic entrance flow has three different regimes which depends on the back pressure.6 0.4 Tue Jan 4 11:26:19 2005 Fig.2 0. Below.4 0. In Figure (9. and three. below the maximum critical length. Fanno Flow 5 4. Subsonic velocity occurs for supersonic entrance velocity.9) exhibits the M1 as a function of M2 . shockless flow.2 = 0. The fL larger 4D larger M1 must occurs even for shock at the entrance.5 3 2. and as can be observed.5 1 0.2) by obtaining the 4f L for M2 and subtracting D 4f L the given D and finding the corresponding M1 .1  D = 0.1 and D = 0. the larger 4D larger differences between exit Mach number for the different shock locations. WORKING CONDITIONS 175 Figure (9.

However. f . The critical length is the maximum that associate with entrance D Mach number. the exit and entrance Mach fL numbers increase. can increases only if the area increase. This jump is referred to as a choked flow. Choking explanation for pressure variation/reduction Decreasing the pressure ratio or in actuality the back pressure. . 12 See more on the discussion about changing the length of the tube. One case is where the 4D 4f Lmax length. as it was shown in Chapter (4). Mach number. results in increase of the entrance and the exit velocity until a maximum is reached for the exit velocity. 9. the entrance pressure decreases as well because of the increase in the entrance Mach number (velocity). In our model the tube area is postulated as a constant therefore the velocity cannot increase any further. P2 ). P2 . For cases where the supply come from a reservoir with a constant pressure. If the feeding nozzle is converging than the flow will be only subsonic.3. for the flow to be continuous the pressure must decrease and for that the velocity must increase. Any additional reduction in the back pressure will not change the situation in the tube. At first the converging nozzle is presented and later the converging-diverging nozzle is explained. Something must break since there are conflicting demands and it result in a “jump” in the flow.2 The Pressure Ratio.176 critical length is mathematically 4f L D CHAPTER 9. As the pressure ratio increases (smaller back pressure. If the feeding nozzle is a “converging–diverging” then it has to be differentiated fL is short or equal to the critical between two cases.9. The maximum velocity is when exit Mach number equals one. the pressure ratio P P1 variations. Again a differentiation of the feeding is important to point out. is independent of the parameters such as. FANNO FLOW >− 1 1+k k+1 + ln k 2k k−1 fL For cases of 4D above the maximum critical length no supersonic flow can be over the whole tube and at some point a shock will occur and the flow becomes subsonic flow12 . If the nozzle is “converging– diverging” than in some part supersonic flow is possible. The Mach number. Reynolds number et cetera) thus the flow remains on the same Fanno line. P1 effects In this section the studied parameter is the variation of the back pressure and 2 thus. The only change will be at tube surroundings which are irrelevant to this discussion. According to Fanno model the value of 4D is constant (friction factor. For very low pressure ratio the flow can be assumed as incompressible with exit Mach number smaller than < 0.

When the back pressure is below point c. Note however that exit Mach number. In this range and further reduction of the pressure the mass flow rate is constant no matter how low the back pressure is reduced. Once the back pressure is less than point b the supersonic reaches to the tube. the tube is “clean” of any shock13 .11: The pressure distribution as a function of 4f L D for a short 4f L D Short 4f L D Figure (9.12) shows different pressure profiles for different back pressures.9. 9. Before the flow reaches critical point a (in the Figure) the flow is subsonic. Up to this stage the nozzle feeding the tube increases the mass flow rate (with decreasing back pressure).9. 13 It is common misconception that the back pressure has to be at point d. The back pressure below point c has some adjustment as it occurs with exceptions of point d. M2 < 1 and is not 1. Between point a and point b the shock is in the nozzle. WORKING CONDITIONS ƒ‚ &€ „†… 177 a shock in the nozzle fully subsoinic flow Œ† ŒƒŽ critical Point a criticalPoint b critical Point c ‡qˆŠ‰ ‹ critical Point d Fig. A back pressure that is at the critical point c results in a shock wave that is at the exit. .

FANNO FLOW “†” Ð&ÑÓÒsÔ ÐÖÕsÐØ×"Ù·Ô Ú¤Ô ×"ÑÜÛÞÝßà·áâ¦ã ç ämå æ a shock in the nozzle fully subsoinic œžsŸ ¢¡£ ¤BŸ¥¤¦œ§©¨¢ª¬«s­¯®±°³²´µ·¶q¸Š¹ flow ºm½» ¼ ™†š ™ƒ› ¾À¿ÂÁÄà critical Point a criticalPoint b critical Point c •q–Š— ˜ Fig. In this point the exit different from point c at the case of short tube 4D D Mach number is equal to 1 and the flow is double shock. point c in this case is process as explained in the short 4D fL max < 4f L . M1 fL . effects In this discussion. This dissection deals only with the flow when it reaches the { ųÆÇÈ·ÉÊ¦Ë ÌmÏÍ Î for a long 4f L D .12: The pressure distribution as a function of 4f L D Long 4f L D fL max In the case of 4D > 4f L reduction of the back pressure results in the same D fL up to point c. In reality these effects have significance and needs to be accounted for some instances. the effect of changing the throat area on the nozzle efficiency is neglected. However. The is no analytical solution for the location of this point c. At point c or location of the shock wave. is a function entrance Mach number.178 ‘ƒ’ & CHAPTER 9. Further reduction of the back pressure at this stage will not “move” the shock wave downstream the nozzle. 9. M1 .3 Entrance Mach number.9. The and the “extra” 4D procedure is (will be) presented in later stage. 9.

This effect is the third parameter discussed here. the nozzle area ratio is changed by changing the throat area. the critical point is where result in the entrance Mach number.15 4fL  D 0.4 0.1 shock at 75% 50% 5% 0. first is the total 4f L maximum D of the supersonic which depends only on the specific heat. It is P2 assumed that in this discussion that the pressure ratio P is large enough to create 1 4f L a choked flow and D is small enough to allow it to happen.6 0.4 Mach Number 1. 4f L D is equal to 4f Lmax D as a The process of decreasing the converging–diverging nozzle’s throat increases the .6 1. M1 .2 0 0 0.2 1 0.9.8 1. fL As was shown before. supersonic branch reached otherwise the flow is subsonic with regular effects. there are two different maximums for 4D . M1 is a function of the ratio of the nozzle’s throat area to the nozzle exit area and its efficiency. and second the maximum depends on the entrance Mach number. 9.05 0.13: The effects of pressure variations on Mach number profile as a function of fL = 0. in this situation. D Obviously. WORKING CONDITIONS 179 Mach number in Fanno Flow 4fL  D 2 1.9. Practically. k .25 Tue Jan 4 12:11:20 2005 Fig. This analysis fL max deals with the case where 4D is shorter than total 4f L .8 0.3 for Fanno Flow when the total resistance 4D 4f L D The entrance Mach number.2 0.

2 0. The mass flow rate is proportionally linear to the throat area and therefore the mass flow rate reduces.2 0. This part is for the case where some part of the tube is under supersonic regime and there is shock as a transition to subsonic branch. FANNO FLOW P2/P1 Fanno Flow 4fL  D 4. The reference to the tube is because it is the focus of the study.25 5% 50 % 75 % Fri Nov 12 04:07:34 2004 Fig.8 0. The process of decreasing the throat area also results in increasing the pressure drop of the nozzle (larger resistance in the nozzle15 )16 . The “payment” for increase in the supersonic length is by reducing the mass flow.6 3.3 entrance14 Mach number.4 2 1. By doing so.4 4 3.15 4fL  D 0. 15 Strange? Frictionless nozzle has a larger resistance when the throat area decreases 16 It is one of the strange phenomenon that in one way increasing the resistance (changing the throat fL ) does not affect the flow rate. decrease of the throat area results in flushing the shock out of the tube.14: Fanno Flow Mach number as a function of 4f L D when the total 4f L D = 0. Decreasing the nozzle throat area moves the shock location downstream. 9. area) decreases the flow rate while in a different way (increasing the 4D . the throat area decreases.05 0.180 CHAPTER 9.2 P2/P1 2. fL max In the case of large tube 4D > 4f L the exit Mach number increases with the D 14 The word “entrance” referred to the tube and not to the nozzle.1 0.8 2. Further.4 0 0 0.6 1.8 4. If the tube contains no supersonic flow then reducing the nozzle throat area wouldn’t increase the entrance Mach number.

WORKING CONDITIONS üþý ÿ ¡ ¢¢¢ ¢¢¢ ¢ £¥¤§¦©¨     .9.9.

is to find the furtherest shock location downstream.%4/0 B(C*D E FFF FFF F GIHKJ 4f L D Fig. When the entrance Mach number is infinity. 9. M1 = ∞.16: The extra tube length as a function of the shock location. M1 = ∞.16) shows the fL possible ∆ 4D as function of retreat of the location of the shock wave from the maximum location... 9. . -/. Figure (9. However.. §    181   "!# %$ ! èêé<ë ìêíïîñðò8óô¦õõ öø÷ ùûú shock Fig.102-3. the location of the shock wave approaches to the theoretical location if entrance Mach.. 57689. if the . supersonic branch The maximum location of the shock The main point in this discussion however.:(<*= ?A @ > SUTVXW YUZ[]\ LNMPORQ 0 &('*) + .15: Schematic of a “long” tube in supersonic branch decrease of the throat area. . Once the exit Mach number reaches one no further increases is possible.

9. The proposed procedure is based on Figure (9. FANNO FLOW shock location is at the maximum length.182 CHAPTER 9. the Mach number equal infinity when left side assumes result in infinity length of possible extra (the whole flow in the tube is subsonic). To overcome this numerical problem it is suggested to start the calculation from distance from the right hand side.17: The maximum entrance Mach number.16). The requirement that has to be sat- isfied is that denote as difference between the maximum possible of length in which the supersonic flow is achieved and the actual length in which the . ii) Calculate the extra (at the entrance). M1 to the tube as a function of sonic branch super- From numerical point of view. 4f L D and subtract the actual extra 4f L D assuming shock at the left side 4f L D and subtract the actual extra 4f L D assuming shock at the right s iii) According to the positive or negative utilizes your root finding procedure.51) is smaller than 4f L D retreat . Let denote ∆ Note that 4f L D sup 4f L D ¯L f − = 4D actual 4f L D max∞ 4f L D sup (9. i) Calculate the extra the max length). then shock at Mx = 1 results in My = 1. cRd3ePf g 1 4f L D max∞ ^`_Ia b 4f L D Fig.

WORKING CONDITIONS 183 flow is supersonic see Figure (9.442 − 0. 4f L D up = − 4f L D max 4f L D up .0000 With (M1 ) M 0.1318. Or in a very long tube the whole flow will be subsonic. the value of left side is −0.1318.5652 74.7461 1.3: Calculate the shock location for entrance Mach number M1 = 8 and for assume that k = 1.9 S OLUTION max for k = 1. D exceed the maximum length D for this entrance fL Mach number. 4f L D = 0. Hence.39289 The extra ∆ 4f L D My 0.42390 1. . thus the extra tube is ∆ 4f L D fL = 0.17) shows the entrance Mach number.39289 Ty Tx ρy ρx Py Px P0y P0 x 13.4 The solution is obtained by an iterative process. (c) Calculate the Mach number corresponding to the current guess of 17 What if the right side is also negative? The flow is chocked and shock must occur in the nozzle before entering the tube.9.52) Figure (9. The right side is when the shock is at the entrance at which the extra 4f L D is calculated for Mx and My is 4f L D = 0. In a summary of the actions is done by the following algorithm: (a) check if the (b) Guess 4f L D is 2.4 (Mexit = 1). Hence.1318 = 2.76820 (flow is choked and no additional 4D ). Accordingly continue.76820.9 − 0. The retreating length is expressed as subsonic but 4f L D retreat = 4f L D max∞ − 4f L D sup (9.3867 5.9.6136 2.4417 2.15).5000 0. M1 reduces after the maximum length is exceeded. Example 9. The maximum for M1 = 8 is 4D = 0.76820 = 0.3591 0. The maximum 4f L D 4f L 4f L is 0.821508116.1641 between the negative of left side to the positive of the right side17 .3102 Now the solution is somewhere exceeds the maximum 4f L D 4f L D max for the supersonic flow.00849 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 2. The left side is when the shock occurs at Mx 8.

32932 0.90000 0.0000 4f L D down against the old one.90000 0. FANNO FLOW (d) Calculate the associate Mach number.64830 4fL D 0.57184 0.64834 0.57069 0.6021 1.64830 0.66779 0.6705 1.22574 0.27830 0.57068 0. Mx with the Mach number.32932 0.65023 0.3838 1.57122 0.57318 0.6706 1.90000 0.64920 0.32682 0.90000 0.58217 0. My calculated previous (e) Calculate 4f L D for supersonic branch for the Mx 4f L D up (f) Calculate the “new and improved” (g) Compute the “new 4f L D down = 4f L D − 4f L D up (h) Check the new and improved to stage (b).6382 1.32931 0.90000 0.32921 0.67426 0.90000 0.90000 0.90000 0.6635 1.6699 1.57068 0.57070 0.64831 0.64830 0.64831 0.90000 0.64830 0.90000 0.0000 M2 1.90000 This procedure rapidly converted to the solution.90000 0.32930 0.6706 1.64832 0.6703 1. Shock location are: M1 8.64830 0.6706 1.90000 0.6673 1.57068 0.64839 0.64872 0.90000 0.62170 0.32907 0.90000 0.32927 0.6706 My 0.6706 1.32395 0.57068 0.6706 1.30494 0.31783 0.6706 1.57079 0.57068 0.69119 0. .90000 0.74664 0.90000 0.32932 The iteration summary is also shown below i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 4fL D up 4fL D down Mx 1.32932 0.57073 0.57068 0.32878 0.184 CHAPTER 9.32816 0.6691 1.59506 0.32932 0.57093 0.6706 My 0.57605 0.6554 1.64850 0.32932 0.90000 0.65246 0. If it is satisfactory stop or retu 4fL D up 4fL D down Mx 1.5286 1.65728 0.

9. This tolerance requires that (T0 − T )/T0 = 0. say about 1% that flow can be assumed to be isothermal. 9. When the Mach number decreases the temperature approaches the stagnation temperature (T → T0 ).2) or the results from computer program attached to this book shows that reduction of the mass flow is very rapid.18: The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow fL As it can be seen for the Figure (9. Hence.1 iso = 0.8 iso = 0. Hence the mass flow rate is a function of 4D because M1 changes. The only difference is in small dimensionless fL . friction. 4D .4 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.8 M1 Fanno flow 0.10. One must notice that the maximum temperature at the entrance is T0 1 . THE APPROXIMATION OF THE FANNO FLOW BY ISOTHERMAL FLOW185 9.67.2 = 0.99 which requires that enough for M1 < 0. The results are very similar for isothermal flow.1 0 0 10 20 30 40 Wed Mar 9 11:38:27 2005 50 4fL  D 60 70 80 90 100 Fig.15 even for large k = 1. Looking at the table or Figure (9.2 0.5 = 0. if one allows certain deviation of temperature. This fL requirement provides that somewhere (depend) in the vicinity of 4D = 25 the fL flow can be assumed isothermal.18) the dominating parameter is 4D .1 = 0.10 The Approximation of the Fanno flow by Isothermal Flow The isothermal flow model has an equation that theoreticians find it easier to use compared to the Fanno flow model.3 M1 0. to insert a question or example about this issue in end with comperison to Isothermal Flow 0.

99873 0. 4f L D 4f L D .98928 0. D = 0.7569 107.07975 M2 1.40790 0. L = 4[m]. 0.7569 67.4: To demonstrate the utility in Figure (9.4 ≈ 0.12728 0.18) consider the following example. Find the mass flow rate for f = 0.11 More Examples of Fanno Flow Example 9.127 × RT 4 1. or accurately by utilizing the program as in the following table.5.4206 0.5393 4. The mass flow rate is m ˙ = P AM π × 0.0000 0.1.12420 0.99097 0.1 M1 ≈ 0. ρ ρ0 M 0.6523 5.1697 10.0000 40.2519 A×P A∗ ×P0 Therefore.2842 0.99677 0.22697 0.05.1697 50. 0.0 2.3.0000 40.80000 Only for the pressure ratio of 0. FANNO FLOW 9.12420 0.0000 40. T ≈ T0 and is the same for the pressure.022 k = 300000 × × 0. M1 0.13 etc.48 287300 kg sec .186 CHAPTER 9.1 the flow is choked.11637 0.09965 4fL D 4fL D 1 4fL D 2 P2 P1 40.99692 0.99741 0.1196 7.11392 0.98874 0.11392 0.05 × 4 = 40 0.0000 40.99556 4. the mass rate is a function of the Mach number.50000 0.42 0.99233 0. The stagnation conditions at the entrance are 300K and 3[bar] air. 0.30000 0.02 From Figure (9.8. S OLUTION First calculate the dimensionless resistance.99354 0.7027 5.12728 0.0733 7.99683 4. Hence.0000 42.99195 0.5910 4.02[m] and pressure ratio P2 /P1 = 0.18) for P2 /P1 = 0. = 4 × 0.07975 T T0 A A P P0 0. The Mach number is indeed a function of the pressure ratio but mass flow rate is a function of pressure ratio only through Mach number.

1985 1.1976 1.2047 1.7092 1.2508 9.50 0.9034 11.6958 1.65 0.08 440.43 P1 0.1994 1.9216 14.3546 2.1139 P2 m ˙ = 0.468 P1 0.5826 3.5 ∼ 0.2119 1.1512 9.0691 0.3865 1.04381 0.49082 0.58506 0.35 0.1291 1.8708 1.87037 0.1713 1.00 3.09851 0.1381 1.061 1.1356 1.2620 15.85 0.7780 2.1882 1.8 ∼ 0.03286 0.000 0.2838 1.07664 0.25 0.95 1.4613 10.66 106.50918 0.48 × = 0.30500 0.6183 1.0702 2.52216 36.4961 66.1788 1.3085 1.4554 2.05476 0.1315 1.12 The Table for Fanno Flow Table 9.72 83.1998 1.1273 0.3848 1.0934 1.000 0.75 0.5664 1.21822 4.40 0.32572 0.2200 13.0089 1.21822 0.04 0.0 0.07 0.4626 1.45 0.60 0.42857 .235 30.4525 2.4834 5.4182 10.91460 0.4027 3.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 787.1065 1.2616 12.8254 18.1242 P2 = 0.1905 1.07975 P2 = 0.30 0.1194 1.0474 11.73179 0.20 0.9.5901 2.3817 14.27217 0.77894 0.633 1.12.9341 1.07229 0.5753 1.03 0.6191 2.55 0.00328 0.06 0.1981 1.6400 2.35 280.82514 0.2915 13.10944 0.1429 1.5116 19.00 1.48 × = 0.1996 1.1628 1.63481 0.3 ∼ 0.5914 18.9435 5.5333 8.32459 0.1533 1.002 1.2549 1.1273 kg sec kg sec kg sec 187 9.3005 27.1489 1.06570 0.01451 0.08758 0.70 0.6416 8.00 2.2893 1.0327 1.3665 1.6659 15.4487 2.43133 0.0929 1.08 0.3184 2.4815 21.1618 6.03633 0.3398 1.0944 1.0693 1.30 m ˙ P1 0.1273 0.03 140.1: Fanno Flow Standard basic Table M 0.90 0.02 193.688 0.964 1.09 0.9635 4.10 0.40825 1.05 0.20814 0.000 0.68374 0.00000 1.0351 3.8218 5.12728 0.0638 1.2993 3.017 1.7634 1.66667 0.1991 1.37879 0.0922 1. THE TABLE FOR FANNO FLOW and for the rest m ˙ 0.6843 7.0382 1.0624 1.61237 0.48326 0.4935 1.95781 1.4318 22.1378 4.0207 1.1852 1.0787 1.0485 1.6742 3.53452 0.72805 0.1988 1.80 0.48 × = 0.044 1.

00 60.00153 4.00663 0.82033 0.00 45.000809 2.00 30.00 8.44721 0.000579 5.8E +6 0.40866 0.4E +6 0.00609 1.4E +2 0.81860 0.76819 0.00 5.00 10.036860 1.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.8E +5 0.40875 0.40859 0.00 9.447 2.00271 1.00200 2.00 7.5E +4 0.014815 0.448 2.000500 7.81928 0.00 70.20000 0.00121 8.13363 10.42390 0.063758 53.00240 0.000979 1.000680 3.81582 0.82052 0.75280 0.443 2.40938 0.6E +6 0.377 2.69380 0.81975 0.3E +2 0.138 2.295 2.047619 1.40889 0.00488 0.82008 0.28571 0.81265 0.18 0.40846 2.3E +6 0.00952 0.40908 0.00 65.333 2.069767 0.188 CHAPTER 9.9E +2 0.40849 0.00122 .434 2.78683 0.1: Fanno Flow Standard basic Table (continue) M 4.77899 0.023905 5.40853 0.14634 0.445 2.448 0.00 55.00 25.1E +5 0.390 2.440 2.41079 0.00 6.0E +2 0.00198 0.00 20.00142 0.6E +5 0.057143 0.448 2.41833 0.5E +5 0.359 2.5E +6 0.447 2.00 0.46771 0.00296 0.446 2.089443 25.11111 0.086957 0.43568 0.00390 4. FANNO FLOW Table 9.82078 0.00 35.6E +4 0.00 40.446 2.63306 0.236 2.029348 3.72988 0.00374 0.40988 0.82066 0.00 50.42066 0.81755 0.42857 0.00166 0.72 0.

In practice Rayleigh flow is really provide good model for the real situation.CHAPTER 10 RAYLEIGH FLOW Rayleigh flow is (frictionless) flow with heat transfer through a pipe of constant cross sectional area.Fig. This fact create situation different compare to the previous two models. Yet. 10.1: The control volume of Rayleigh Flow fer and friction the main assumption was that relative length is so the heat transfer occurs between the surrounding and tube. This model usage is to have a rough predict the conditions occur mostly in 189 . As opposed to the two previous models. The density and pressure change due to external cooling or heating. Rayleigh flow is practical and useful concept in a obtaining trends and limits. the heat transfer can be in two directions not like the friction (there is no negative friction).1 Introduction The third simple model for an one dimenhji flow prn q k`l direction mo sional flow is for constant heat transfer for fric( s t uwv tionless flow. As before. 10. This model applied to case where the heat transfer is significant and the friction can be ignored. In contrast. While the isothermal flow model has heat trans. This flow referred in the literature as Rayleigh Flow (see historical notes). a simple model is built around assumption of constant properties (poorer prediction to case were chemical reaction take palace). the heat transfer in Rayleigh flow occurs either between unknown temperature to tube and the heat flux is maintained constant. This x flow is another extreme case in which the fricheat transfer (in and out) tion effect are neglected because their relative effect is much smaller the heat transfer effect.

Nevertheless. in Rayleigh flow also cooling can be applied. if the upstream conditions are known (or downstream condition are known). one has to be aware that properties do change significantly for a large range of temperature. RAYLEIGH FLOW situations involve chemical reaction. Thus.190 CHAPTER 10.2) (10. The flow velocity acceleration change the direction when the cooling is applied. a solution can be obtained. Yet. (10. The density change though the heat transfer (temperature change). 10.7) U2 ρ1 = = ρ2 U1 kRT2 U1 kRT1 . As appose to Fanno flow in which the resistance always oppose the the flow direction. P2 1 + kM1 2 = P1 1 + kM2 2 (10.4) are similar to the equations that were solved for the shock wave.4) can further assist in obtaining the temperature ratio as T2 P2 ρ 1 = (10. In analysis of the flow. for smaller range of temperature and length the calculations are more accurate.3) The energy balance on the control volume reads the momentum balance reads The mass conservation reads Equation of state P1 P2 = ρ 1 T1 ρ 2 T2 (10.1) (10.3) and (10.5) The equation of state (10. One can notice that equations (10. are encapsulated in this model. The basic physics of the flow revolves around the fact that the gas is highly compressible. the main characteristic of the flow such as chocking condition etc.2).2 Governing Equation Q = Cp (T0 2 − T0 1 ) (10.6) T1 P1 ρ 2 The density ratio can be expressed in term of mass conservation as U2 r r √ kRT2 = kRT1 M2 M1 T2 T1 (10.4) A(P1 − P2 ) = m ˙ (V2 − V1 ) ρ 1 U1 A = ρ 2 U2 A = m ˙ There are four equations with four unknown.

10. GOVERNING EQUATION Substituting equations (10.6) yields T2 1 + kM1 2 M2 = T1 1 + kM2 2 M1 T2 T1 191 (10. and s2 . and s1 are constant then the variable parameters are M2 .2. A derivative of equation (10.8) Transferring the temperature ratio to left hand side and squaring results in T2 1 + kM1 2 = T1 1 + kM2 2 2 M2 M1 2 (10. T2 k − 1 P2 s1 − s 2 = ln − ln Cp T1 k P1 s1 − s 2 1 + kM1 2 ) M2 k−1 1 + kM 212 = 2 ln ( + ln Cp k (1 + kM2 2 ) M1 1 + kM1 2 (10.10. The second maximum can be expressed as dT /ds = ∞ The second law is used to find the expression for derivative.7) into equation (10.12) .2: The Temperature Entropy Diagram For Rayleigh Line The Rayleigh line exhibits two possible maximums one for dT /ds = 0 and for ds/dT = 0.10) (10.11) Let the initial condition M1 .9) {N|~ }€ y ‰XŠˆ‹ ne ŒŽˆŒ N‚„ƒ Pr es su re li co ns ta …X†ˆ‡ nt z Fig.5) and (10.11) results in 2(1 − M 2 ) 1 ds = Cp dM M (1 + kM 2 ) (10.

chocking can be explained by the fact increase of energy must accompanied by increase of entropy. and M2 results in dT 1 − kM 2 = constant × 3 dM (1 + kM 2 ) Combining equations (10. It is convent to referrers to the value of M = 1.13) by eliminating dM results in M (1 − kM 2 ) dT = constant × ds (1 − M 2 )(1 + kM 2 )2 (10. several observations can √ be generalized.). The chocking is occurred only when M = 1 because it violate the second law. But the entropy of supersonic flow is lower (see the Figure (10. dT /ds < 0. These value referred as the “star”1 values.12) and (10. And cooling results in reduction of entropy. Yet for every curve. some what similarly to Fanno flow. Yet.17) star is an asterisk. The derivative is equal to infinity.14) (10. The derivative is equal to zero when 1 − kM 2 = 0 or M = 1/ k or when M → 0. dT /ds = ∞ when M = 1. increase of heating results in increase of entropy. The equation (10. Hence. The derivative is negative. 1 + kM1 2 P∗ = P1 1+k The temperature ratio is 1 T∗ = 2 T1 M 1 + kM1 2 1+k 2 (10. From thermodynamics.15) (10. The transition to supper sonic flow occurs when the area changes.192 CHAPTER 10.16) U∗ ρ1 = = ∗ ρ U1 1 The √U kRT ∗ kRT ∗ √ √ U1 kRT1 kRT1 ∗ √ = 1 M1 T∗ T1 (10.2)) and therefore it is not possible (the maximum entropy at M = 1.12) when letting the variable parameters be T2 . The peculiars point of M = 1/ k when additional heat is applied the temperature is decreasing. .5) can be written between chocking point and any point on the curve. yet note this point is not the choking point. when cooling applied to a tube the velocity decreases and heating applied the velocity √ increases. RAYLEIGH FLOW Take the derivative of the equation (10.13) On T-s diagram a family of curves can be drawn for a given constant.

80 0.00917 0.94009 0.31373 0.00765 0.025 1.014 1.51413 0.30440 0.3 Rayleigh Flow Tables The “star” values are tabulated in Table (10.379 2.056020 0. Table 10.015 0.61393 0.157 1.266 1.4 M 0.135 1.261 1.75524 0.25 0.373 2.50 0.75 0.00598 0.81013 0.69751 0.35 0.193 1.001 0.90 0.961 1.20661 0.207 2.265 1.65 0.397 2.95 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.04 0.95693 .005 1.036212 0.017119 0.273 2.19183 0.508 1.96395 0.00431 0.015224 0.63713 0.023669 0.423 1.090909 0.99290 1.30 0.96081 0.05 0.00517 0.367 2.85 0.384 2.03 0.199 1.99814 2.20 0.011922 0.131 2.06 0.86204 0.596 1.014300 0. RAYLEIGH FLOW TABLES 193 T1 1 + T0 1 ∗ = T0 T∗ 2 k−1 2 M1 1+k 2 = 2(1 + k )M1 2 (1 + kM 2 )2 1+ k−1 M1 2 2 (10.37865 0.00383 0.25096 0.395 2.09 0.86833 0.81892 0.267 1.69136 0.178 1.40 0.262 1.125 1.392 2.094 1.023223 0.60 0.99207 0.52903 0.025 1.79012 0.218 1.027841 0.91670 0.043 1.058 1.686 1.44444 0.61515 0.030 1.40887 0.343 1.046777 0.10 0.90850 0.85987 0.55 0.1: Rayleigh Flow k=1.08 0.045616 0.778 1.075 1.259 1.264 1.13793 0.18) The stagnation pressure ratio reads P1 1 + P0 1 ∗ = P0 P∗ 2 k−1 2 M1 1+k 2 = 1+k 1 + kM1 2 1 + kM1 2 (1+k) 2 k k−1 (10.10.266 1.19) 10.57447 0.75991 0.019222 0.70 0.030215 0.45 0.1). Several observations can be made in regards to the stagnation temperature.51001 0.25684 0.019 1.43894 0.038075 0.011 1.91097 0.870 1.00860 0.060 1.34686 0.388 2.266 1.011680 0.00216 0.3.049 1.020529 0.17355 0.114 1.98097 0.07 0.029 1.235 1.70804 0.

5 9.698 1.49259 0.405 1.256 1.95798 0.0 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 1.1: Rayleigh Flow k=1.403 1.424 5.00326 0.561 1.588 1.713 .5 1.0 25.0 20.693 1.6 1.5 5.675 1.873 3.75250 0.078 1.510 1.00428 0.71301 0.39643 0.030094 0.240 1.13223 0.81414 0.7E +5 1.72421 0.67380 0.375 1.79576 0.316 1.21419 0.616 1.702 1.078 1.032053 0.000 1.222 2.70174 0.571 1.5 7.0 3.5 6.5 10.49356 0.0 1.33454 0.418 2.3 2.634 2.7 2.50702 0.066667 0.47562 0.0E +2 3.2 2.9 3.301 1.65377 0.00 1.26478 0.0 8.4 2.51647 0.88419 0.0 5.743 1.039900 0.00274 0.00140 1.49122 1.690 1.8 1.2 1.328 8.93425 0.36364 0.64103 0.0 35.045 2.176 1.56982 0.017021 0.66350 0.8E +2 1.89087 0.28551 0.1 2.0 7.067263 0.80539 0.0E +4 1.85917 0.019 1.058264 0.50885 0.503 1.205 1.0 6.22936 0.092719 0.00 0.83628 0.13540 0.96031 0.52893 0.95 54.61580 0.00240 1.73954 0.046693 0.1E +4 3.044910 0.16831 0.57831 0.3 1.686 1.136 3.75613 0.538 1.56734 0.10256 0.8E +2 2.525 1.77406 0.227 12.194 CHAPTER 10.2E +4 8.52356 0.68 75.43353 0. RAYLEIGH FLOW Table 10.53633 0.6 2.455 1.340 1.431 1.90928 0.50 18.146 1.5 4.5 2.49415 0.034483 0.54473 0.21 38.00 0.17647 0.40384 0.79339 0.0 2.30864 0.52004 0.475 1.33439 0.122 1.039883 0.005 1.018846 0.4 (continue) M 1.0 9.712 1.044 1.99392 0.580 1.050943 0.9 2.695 1.4E +2 1.711 1.51098 0.11111 0.0E +2 1.31486 0.85971 0.18788 0.43122 0.28028 0.49174 0.00190 0.29687 0.1 1.41 1.494 1.20040 0.035650 0.00 0.081772 0.028972 0.620 1.641 1.69700 0.24615 0.63 27.701 1.35561 0.699 1.886 2.713 1.51349 0.656 1.4 1.023495 0.60894 0.026490 0.8 2.055363 0.21417 0.0 30.55556 0.667 1.0 4.078487 0.3E +2 3.71006 0.681 1.37870 0.550 1.00732 0.46106 0.7 1.020979 0.58909 0.91185 0.5 8.97872 0.68494 0.65398 0.52970 0.52438 0.00469 0.

10.3. RAYLEIGH FLOW TABLES
Table 10.1: Rayleigh Flow k=1.4 (continue)

195

M 40.0 45.0 50.0 55.0 60.0 65.0 70.0

T T∗

T0 T0 ∗

P P∗

P0 P0 ∗

ρ∗ ρ

0.00184 0.00145 0.00117 0.000971 0.000816 0.000695 0.000600

0.49089 0.49066 0.49050 0.49037 0.49028 0.49021 0.49015

0.00107 0.000846 0.000686 0.000567 0.000476 0.000406 0.000350

3.4E +5 6.0E +5 1.0E +6 1.6E +6 2.5E +6 3.8E +6 5.5E +6

1.714 1.714 1.714 1.714 1.714 1.714 1.714

The Data is presented in Figure (10.3).

Rayleigh Flow
4 k=14 T/T * T0 / T0 P/P * P0 / P0 ρ/ρ 2
∗ * *

3

@( M = ∞ ) 1.7143

1 @( M = ∞ ) 0.4898 0

0.1

1

M

10

Fri May 6 11:11:44 2005
Fig. 10.3: The basic functions of Rayleigh Flow (k=1.4)

196

CHAPTER 10. RAYLEIGH FLOW

10.4

Examples For Rayleigh Flow

Illustrative example The typical questions that raised in Rayleigh Flow are related to the maximum heat that can be transfered to gas (reaction heat) and to flow rate. Example 10.1: Air enters a pipe with pressure of 3[bar] and temperature of 27◦ C at Mach number of M = 0.25. Due internal combustion heat was released and the exit temperature was found to be 127◦C. Calculated the exit Mach number, the exit pressure, the total exit pressure, and heat released (transfered) to the air. After what amount of kJ energy the exit temperature will start to decrease? Assume CP = 1.004 kg ◦C S OLUTION The entrance Mach number and the exit temperature are given and from the Table 10.1 or from the program the initial ratio can be calculated. From the initial values the ratio at the exit can be computed as following. M 0.25000 and
T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ

0.30440

0.25684

2.2069

1.2177

0.13793

T1 T2 400 T2 = ∗ = 0.304 × = 0.4053 T∗ T T1 300 M 0.29831
T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ

0.40530

0.34376

2.1341

1.1992

0.18991

The exit Mach number is known, the exit pressure can be calculated as P2 = P 1 1 P ∗ P2 =3× × 2.1341 = 2.901[Bar] P1 P ∗ 2.2069

For the entrance the stagnation values are M
T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗

0.25000 0.98765 0.96942 2.4027

0.95745 2.3005

1.0424

The total exit pressure, P02 can be calculated as the following:
isentropic

P0 2 = P 1

P0 1 P1

P0 ∗ P0 2 1 1 × × 1.1992 = 3.08572[Bar] ∗ =3× P0 1 P0 0.95745 1.2177

10.4. EXAMPLES FOR RAYLEIGH FLOW

197

The heat release (heat transfer) can be calculated from obtaining the stagnation temperature form the both sides. The stagnation temperature at the entrance, T01
isentropic

T0 1 = T 1

T0 1 T1

= 300/0.98765 = 303.75[K ]

The isentropic conditions at the exit are M
T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗

0.29831 0.98251 0.95686 2.0454 The exit stagnation temperature is
isentropic

0.94012 1.9229

0.90103

T0 2 = T 2 The heat release becomes

T0 2 T2

= 400/0.98765 = 407.12[K ]

Q kJ = Cp (T02 − T01 ) 1 × 1.004 × (407.12 − 303.75) = 103.78 m ˙ seckg ◦ C The √ maximum temperature occurs at the point where the Mach number reaches 1/ k and at this point the Rayleigh relationship are: M 0.84515
T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ

1.0286

0.97959

1.2000

1.0116

0.85714

The maximum heat before the temperature can be calculated as following: Tmax = T1 T ∗ Tmax 300 × 1.0286 = 1013.7[K ] T1 T ∗ 0.3044

The isentropic relationship at the maximum are M
T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗

0.84515 0.87500 0.71618 1.0221 The stagnation temperature for this point is T0max = Tmax ∗

0.62666 0.64051 0.53376

T0max 1013.7 = = 1158.51[K ] Tmax 0.875

The maximum heat can be calculated as kJ Q = Cp (T0max − T01 ) = 1 × 1.004 × (1158.51 − 303.75) = 858.18 m ˙ kgsecK Note that this point isn’t the choking point.

198

CHAPTER 10. RAYLEIGH FLOW

Example 10.2: Heat is added to the air until the flow is choked in amount of 600 [kJ/kg]. The exit temperature is 1000 [K]. Calculated the entrance temperature and the entrance Mach number. S OLUTION The solution involve finding the stagnation temperature at the exit and subtraction of the heat (heat equation) to obtain the entrance stagnation temperature. From the Table (10.1) or from the Potto-GDC the following ratios can be obtained. M 1.0000
T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗

0.83333 0.63394 1.0000

0.52828 0.52828 0.52828

The stagnation temperature T0 2 = T 2 The entrance temperature is Q/m ˙ 600 T0 1 ∼ =1− = 1200 − = 0.5016 T0 2 T0 2 C P 1200 × 1.004
T0 1 T0 ∗

T0 2 1000 = = 1200.0[K ] T2 0.83333

It must be noted that T02 = T0 ∗ . Therefore with or by Potto-GDC the following is obtained M 0.34398
T T∗ T0 T0 ∗

= 0.5016 either by Table (10.1)
P0 P0 ∗ ρ∗ ρ

P P∗

0.50160

0.42789

2.0589

1.1805

0.24362

Thus, entrance Mach number is 0.38454 and the entrance temperature can be calculated as following T1 = 1000 × 0.58463 = 584.6[K ] T∗ The difference between the supersonic branch to subsonic branch T1 = T ∗ Example 10.3: Air with Mach 3 enters a frictionless duct with heating. What is the maximum heat that can be add so there is no subsonic flow. If a shock is occurs immediately at the entrance what is the maximum heat that can be added? S OLUTION To achieve maximum heat transfer the exit Mach number has to be one, M2 = 1. T0 Q = Cp (T02 − T01 ) = Cp T0 ∗ 1 − 1 m ˙ T0 ∗ The table for M = 3 as following

10.4. EXAMPLES FOR RAYLEIGH FLOW M 3.0000
T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ

199

0.28028

0.65398

0.17647

3.4245

1.5882

The higher the entrance stagnation temperature the larger the heat amount that can be absorbed by the flow. In subsonic branch the Mach number is after the shock is Mx 3.0000 My 0.47519
Ty Tx ρy ρx Py Px P0y P0 x

2.6790

3.8571

10.3333

0.32834

With Mach number of M = 0.47519 the maximum heat transfer requires information from Rayleigh flow as following M 0.33138
T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ

0.47519

0.40469

2.0802

1.1857

0.22844

M 0.47519

T T∗

T0 T0 ∗

P P∗

P0 P0 ∗

ρ∗ ρ

0.75086

0.65398

1.8235

1.1244

0.41176

It also must be noticed that stagnation temperature remains constant across shock wave. T0 1 Q 1 − T0 ∗ m ˙ 1 − 0.65398 subsonic subsonic = = =1 T0 1 Q 1 − 0.65398 1 − T0 ∗ m ˙
supersonic supersonic

It is not surprising since the the shock wave are found on the Rayleigh flow.

200

CHAPTER 10. RAYLEIGH FLOW

In this Chapfor relativly long tube ter a discussion on application of comVolume is a function of pressure or rigid pressible flow to other fields like manu(the volume can be also a function of inertia and etc) Semi rigid tank facturing is presented2 . in life.” 2nd Edition. Scranton. 201 . There is a significant impor. 11. 1985. 2 Even if the instructor feels that their students are convinced about the importance of the compressible.Fig. are far Fanno model 1 After completion of these Chapters. A. this example can further strength and enhance this conviction. by M. However. Fundamentals of Gas Dynamics. that Volume forced models in traditional classes for with compressExternal forces that control the tank volume ible flow (sometimes referred to as gas Fanno model dynamics) don’t provide a demonstrafor relativly short tube tion to applicability of the class mateš“›•œj ž Ÿ#¡1¢#£"¤ rial aside to aeronautical spectrum even Isothermal model such as turbomachinery. A. These books are OCR. and “Compressible Fluid Flow. It Isothermal model for relatively long tube was the undersigned experience. Prentice Hall. International Textbook Co. in many instances. the undersigned discover two text books which to include some material related to this topic. Pennsylvania.CHAPTER 11 Evacuating and Filling a Semi Rigid Chambers for relatively short tube In some ways the next two Chapters “ ‘ • ’ © ” – ™ — ˜ contain materials is new to the traditional compressible flow text books1 . 1964. J..1: The two different classifications of tance to the “pure” models such Isothermodels that explain the filling or mal flow and Fanno flow which have imevacuating of a single chamber mediate applicability. these books contained only limit discussions on the evacuation of chamber with attached nozzle.. However. the situations. Saad.

11. For example.3: A schematic of the control volIsothermal flow is appropriate model. Yet Isothermal flow model is used as the limiting case. the models in this Chapter are suitable for study of the filling the airbag or filling bicycle with air. The Isothermal model is more appropriate for cases where the tube is relatively long in–which a significant heat transfer occurs keeping the temperature almost constant.202 CHAPTER 11. are presented (see Figure (11. Combination of gas compressibility in the chamber and flow out or through a tube post a special interest and these next two Chapters are dealing with these topics. The analysis is extended to include a semi rigid tank. were the chamber’s volume is a function of external forces. 4D . at this stage. here Isothermal flow and Fanno flow are used as limiting or bounding cases for the real flow. Yet umes used in this model as it was stated before. The term semi rigid tank referred to a tank that the volume is either completely rigid or is a function of the chamber’s pressure.1)).2: A schematic of two possible connections of the tube to a single chamber The Rayleigh flow model requires ¼½I¾K¿ ÀÂÁ Ã Ä that a constant heat transfer supplied either Å(Æ Ç È by chemical reactions or otherwise. As it was shown in this book the most appropriate model for the flow in the tube for a relatively fast situation is Fanno Flow. In the first Chapter models. In the second Chapter. Fanno flow model is the most appropriate in the case where the filling and Control volume for the filling case evacuating is relatively fast. should be larger than 400. As it was shown in ChapfL ter (9) the resistance. are discussed. most car today equipped with an airbag. In case the fL filling is relatively slow (long 4D than the Fig. And therefore. 11. models. This author isn’t familiar with situations in which Control volume for the evacuating case Rayleigh flow model is applicable. fanno model reduced connection ²§³©´oµ ¶w·#¸¯¹2º¯» for relatively short tube fanno model 1 2 ¥§¦©¨oª «w¬#­¯®2°¯± A schematic of a direct connection 1 for relatively short tube 2 Isothermal model for a relatively long tube Isothermal model for a relatively long tube The connection is through a narrow passage Fig. EVACUATING AND FILLING A SEMI RIGID CHAMBERS more complicate. Addition- . were the chamber volume is controlled or a function of the pressure. 11.1 Governing Equations and Assumptions The process of filing or evacuating a semi flexible (semi rigid) chamber through a tube is very common in engineering. For instance. no discussion is ofÉÊIËKÌ ÍÂÎ Ï Ð Ñ(Ò Ó fered here.

11.1. GOVERNING EQUATIONS AND ASSUMPTIONS

203

ally, the process in the chamber can be limited or bounded between two limits of Isentropic process or Isothermal process. In this analysis, in order to obtain the essence of the process, some simplified assumptions are made. The assumptions can be relaxed or removed and the model will be more general. Of course, the payment is by far more complex model that sometime clutter the physics. First, a model based on Fanno flow model is constructed. Second, model is studied in which the flow in the tube is isothermal. The flow in the tube in many cases is somewhere between the Fanno flow model to Isothermal flow model. This reality is an additional reason for the construction of two models in which they can be compared. Effects such as chemical reactions (or condensation/evaporation) are neglected. There are two suggested itself possibilities to the connection between the tube to the tank (see the Figure 11.2): one) direct two) through a reduction. The direct connection is when the tube is connect straight to tank like in a case where pipe is welded into the tank. The reduction is typical when a ball is filled trough an one–way valve (filling a baseball ball, also in manufacturing processes). The second possibility leads itself to an additional parameter that is independent of the fL resistance. The first kind connection tied the resistance, 4D , with the tube area. The simplest model for gas inside the chamber as a first approximation is the isotropic model. It is assumed that kinetic change in the chamber is negligible. Therefore, the pressure in the chamber is equal to the stagnation pressure, P ≈ P0 (see Figure (11.4)). Thus, the stagnation pressure at the tube’s entrance is the same as the pressure in the chamber. The mass in the chamber and mass flow out are expressed in terms of the chamber variables (see Figure 11.3. The mass in the tank for perfect gas reads dm −m ˙ out = 0 dt (11.1)

ÔXÕÖÔ*× ÜÞÝ©ß

Ø

ÙÛÚ

Ø
1 2

And for perfect gas the mass at any given time is

Fig. 11.4: The pressure assumptions in the chamber and tube entrance

m=

P (t)V (t) RT (t)

(11.2)

fL The mass flow out is a function of the resistance in tube, 4D and the pressure 4f L difference between the two sides of the tube m ˙ out ( D , P1 /P2 ). The initial conditions in the chamber are T (0), P (0) and etc. If the mass occupied in the tube is

204

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

neglected (only for filling process) the most general equation ideal gas (11.1) reads
m ˙ out m U fL ± ρ1 A c1 M1 ( 4D ,

d dt

PV RT

P2 )=0 P1

(11.3)

When the plus sign is for filling process and the negative sign is for evacuating process.

11.2

General Model and Non-dimensioned
¯ ¯ = T (t = t) T T (t = 0)

It is convenient to non-dimensioned the properties in chamber by dividing them by their initial conditions. The dimensionless properties of chamber as (11.4a) (11.4b) (11.4c) (11.4d)

¯ ¯ = V (t = t) V V (t = 0) ¯ ¯ = P (t = t) P P (t = 0) ¯= t t tc where tc is the characteristic time of the system defined as followed tc = V (0) AMmax kRT (0))

(11.5)

The physical meaning of characteristic time, tc is the time that will take to evacuate the chamber if the gas in the chamber was in its initial state, the flow rate was at its maximum (choking flow), and the gas was incompressible in the chamber. Utilizing these definitions (11.4) and substituting into equation (11.3) yields
ρ c(t)

¯V ¯ P (0)V (0) d P ¯ ¯ T tc RT (0) dt

¯1 P (0) P ¯1 T (0)Mmax M ¯ (t ¯) = 0 ± ¯ A kRT RT1 T (0)

(11.6)

where the following definition for the reduced Mach number is added as ¯ = M1 (t) M Mmax After some rearranging equation (11.6) obtains the form ¯V ¯ d P ¯ ¯ T dt ± ¯1 M ¯1 tc AMmax kRT (0) P ¯ =0 M ¯1 V (0) T (11.8) (11.7)

11.2. GENERAL MODEL AND NON-DIMENSIONED

205

and utilizing the definition of characteristic time, equation (11.5), and substituting into equation (11.8) yields ¯V ¯ ¯1 M ¯ d P P =0 (11.9) ± ¯ ¯ T ¯ dt T
1

Note that equation (11.9) can be modified by introducing additional parameter which referred to as external time, tmax 3 . For cases, where the process time is important parameter equation (11.9) transformed to ¯1 M ¯ ¯V ¯ tmax P d P ± =0 (11.10) ¯ ˜ T ¯ tc dt T
1

¯, V ¯,T ¯, and M ¯ are all are function of t ˜ in this case. And where t ˜ = t/tmax . when P It is more convenient to deal with the stagnation pressure then the actual pressure at the entrance to the tube. Utilizing the equations developed in Chapter 4 between the stagnation condition, denoted without subscript, and condition in a tube ¯ P 1 denoted with subscript 1. The ratio of √ is substituted by ¯
T1

¯ ¯1 k−1 2 P P 1+ =√ M ¯ ¯1 2 T T It is convenient to denote k−1 2 M f [M ] = 1 + 2

−(k+1) 2(k−1)

(11.11)

−(k+1) 2(k−1)

(11.12)

Equation (11.13) is a first order nonlinear differential equation that can be solved for different initial conditions. At this stage, the author isn’t aware that there is a general solution for this equation4 . Nevertheless, many numerical methods are available to solve this equation.

Note that f [M ] is a function of the time. Utilizing the definitions (11.11) and substituting equation (11.12) into equation (11.9) to be transformed into ¯M ¯ (t ¯V ¯ ¯)f [M ] d P P √ ± =0 (11.13) ¯ ¯ T ¯ dt T

11.2.1

Isentropic Process

The relationship between the pressure and the temperature in the chamber can be approximated as isotropic and therefore ¯ = T (t) = P (t) T T (0) P (0)
k−1 k

¯ =P

k−1 k

(11.14)

3 This notation is used in many industrial processes where time of process referred to sometime as the maximum time. 4 To those mathematically included, find the general solution for this equation.

206

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

The ratios can be expressed in term of the reduced pressure as followed: ¯ ¯ P P 1 ¯k = k −1 = P ¯ ¯ T P k and ¯ +1 P ¯ k2 k √ =P ¯ T So equation (11.13) is simplified into three different forms: d ¯ ¯1 VPk ¯ dt
+1 ¯ k2 ¯ (t ¯)f [M ] = 0 k M ±P

(11.15)

(11.16)

(11.17a)

¯ ¯ +1 1 dV 1 ¯ 1−k dP ¯ +P ¯k ¯ k2 ¯ (t ¯)f [M ] = 0 k M P k V ±P ¯ ¯ k dt dt ¯ ¯ −1 ¯ dP + k P ¯ (t ¯ dV ± k P ¯ 3k ¯)f [M ] = 0 2k M V ¯ ¯ dt dt

(11.17b)

(11.17c)

Equation (11.17) is a general equation for evacuating or filling for isentropic process in the chamber. It should be point out that, in this stage, the model in the tube could be either Fanno flow or Isothermal flow. The situations where the chamber undergoes isentropic process but the flow in the tube is Isothermal are limited. Nevertheless, the application of this model provide some kind of a limit where to expect when some heat transfer occurs. Note the temperature in the tube entrance can be above or below the surrounding temperature. Simplified calculations of the entrance Mach number are described in the advance topics section.

11.2.2 11.2.3

Isothermal Process in The Chamber A Note on the Entrance Mach number

fL and the ratio The value of Mach number, M1 is a function of the resistance, 4D of pressure in the tank to the back pressure, PB /P1 . The exit pressure, P2 is different from PB in some situations. As it was shown before, once the flow became fL choked the Mach number, M1 is only a function of the resistance, 4D . These statements are correct for both Fanno flow and the Isothermal flow models. The method outlined in Chapters 8 and 9 is appropriate for solving for entrance Mach number, M1 . Two equations must be solved for the Mach numbers at the duct entrance and exit when the flow is in a chokeless condition. These equations are combinations of

11.3. RIGID TANK WITH NOZZLE

207

the momentum and energy equations in terms of the Mach numbers. The characteristic equations for Fanno flow (9.50), are
4f L D

=

4f Lmax D 1

4f Lmax D 2

(11.18)

and k−1 P2 = 1+ M2 2 P0 (t) 2
k 1−k

M1 M2

1+ 1+

2 k−1 2 M2 2 k−1 2 M1

k+1 k−1

(11.19)

fL where 4D is defined by equation (9.49). fL exit and P The solution of equations (11.18) and (11.19) for given 4D P0 (t) yields the entrance and exit Mach numbers. See advance topic about approximate solution fL or small entrance Mach number, M1 . for large resistance, 4D

11.3

Rigid Tank with Nozzle

he most simplest possible combination is discussed here before going trough the more complex cases A chamber is filled or evacuated by a nozzle. The gas in the chamber assumed to go an isentropic processes and flow is bounded in nozzle between isentropic flow and isothermal flow5 . Here, it also will be assumed that the flow in the nozzle is either adiabatic or isothermal.

11.3.1

Adiabatic Isentropic Nozzle Attached

The mass flow out is given by either by Fliegner’s equation (4.47) or simply use cM ρA∗ and equation (11.17) becomes ¯ +1 1 ¯ 1−k dP ¯ k2 ¯)f [M ] = 0 k (t P k ¯ ±P k dt (11.20)

¯ = 1 and M ¯ definition is simplified as M ¯ = 1. It can be noticed It was utilized that V that the characteristic time defined in equation (11.5) reduced into: tc = V (0) A kRT (0)) (11.21)

Also it can be noticed that equation (11.12) simplified into k−1 2 f [M ] = 1 + 1 2
−(k+1) 2(k−1)

=

k+1 2

−(k+1) 2(k−1)

(11.22)

5 This work is suggested by Donald Katze the point out that this issue appeared in Shapiro’s Book Vol 1, Chapter 4, p. 111 as a question 4.31.

208

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

Equation (11.20) can be simplified as
1−k 1 ¯= 0 P 2k dP ± f [m]dt k

(11.23)

Equation (11.23) can be integrated as
¯ P

P
1

1−k 2k

t

dP ±

dt = 0
0

(11.24)

The integration limits are obtained by simply using the definitions of reduced pres¯ . After the integration, equation (11.24) and ¯ = 0) = 1 and P (t ¯= t ¯) = P sure, at P (t rearrangement becomes ¯ = 1± P k−1 2 ¯ f [M ] t
2k 1−k

(11.25)

Example 11.1: A chamber is connected to a main line with pressure line with a diaphragm and nozzle. The initial pressure at the chamber is 1.5[Bar] and the volume is 1.0[m3 ]. Calculate time it requires that the pressure to reach 5[Bar] for two different nozzles throat area of 0.001, and 0.1 [m2 ] when diaphragm is erupted. Assumed the stagnation temperature at the main line is the ambient of 27[◦ C]. S OLUTION The characteristic time is tmax = V 1.0 V √ = ∗ = = 0.028[sec] A∗ c A c 0.1 1.4 × 287 × 300 tmax = 1.0 √ = 2.8[sec] 0.001 1.4 × 287 × 300 ¯ = P (t) = 4.5 = 3.0 P P (0) 1.5 The time is
k ¯ 1− k −1 t = tmax P

(11.26)

And for smaller area

k+1 2

−()

(11.27)

Substituting values into equation (11.27) results t = 0.028 3
1−1.4 2.8

−1

2.4 2

−2.4 0.8

= 0.013[sec]

(11.28)

Owczarek [1964] found an analytical solution which described here.29) converted into ¯ P 1 (11. f [M ] is expressed by equation (11.2 Isothermal Nozzle Attached In this case the process in nozzle is assumed to isothermal but the process in the chamber is isentropic.29) ¯ dP ± f [M ] ¯ P ¯ t ¯= 0 dt 0 (11. For rigid volume. The temperature in the nozzle is changing because the temperature in the chamber is changing.17) “shrinks” and doesn’t contain the relative volume term. the general equation (11. RAPID EVACUATING OF A RIGID TANK Filling/Evacuating The Chamber Under Upchucked Condition 209 The flow in the nozzle can became upchucked and it can be analytically solved.30) Here.4.32) .3. Thus.4 11.30) transformed into ¯= ln P k+1 2 " −(k+1) 2(k−1) ¯ t # ¯=e P 2(k−1) t ¯ ( k+1 2 ) −(k+1) (11. In such case. equation (11.13) is reduced into V ¯ dP ¯=0 = ±f [M ]P ¯ dt Separating the variables and rearranging equation (11.11. the differential temperature change in the chamber is slower than the temperature change in nozzle.31) 11. V tank. After the integration.1 Rapid evacuating of a rigid tank With Fanno Flow ¯ (t) = 1. Yet. is constant and equal one for a completely rigid The relative Volume.4. equation (11.22). 11. A reasonable model for the tank is isentropic (can be replaced polytropic relationship) and Fanno flow are assumed for the flow in the tube. the specific governing equation is ¯ −1 dP ¯ [M ]P ¯ 3k 2k =0 − k Mf ¯ dt (11. ¯ = 1 and for isothermal nozzle T ¯ = 1 Thus.

11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS For a choked flow the entrance Mach number to the tube is at its maximum. The solution of equation (11.5: The reduced time as a function of the modified reduced pressure The big struggle look for suggestion for better notation.35) The function is drawn as shown here in Figure (11.34) ¯ is a function of P and can It has to be realized that this is “reversed” function i.6 0.0 P(t) 0. At certain point the flow becomes chokeless flow (unless the back pressure is complete vacuum).2 0 0 0. But for the chocked case it appears as ¯ ¯ = 1 + (k − 1)M f [M ] t ¯ P 2 2k 1−k (11.8 V(t) = P (t) V(t) = P (0) 0. The reduced time increases with decrease of the pressure in the tank. The transition point is denoted here as chT .33) direct integration of equation (11.36) . that when the modified reduced pressure equal to one the reduced time is zero. equation (11.2 0.4 0.4 ¯ t 0. t be reversed for case.34) has to include the entrance Mach under the integration sign as ¯ P ¯− t ¯chT = t PchT 3k 1 ¯ ¯ 1− 2k dP ¯ [M ] P k Mf (11.5).6 0.e.8 1.210 CHAPTER 11. Thus.33) results in ¯= t −k 2 ¯ 12 k − 1 P ¯ (k − 1)M f [M ] (11.32) is obtained by noticing that and therefore M ¯ is not a function of time and by variables separation results in M ¯ t ¯ P ¯= dt 0 1 ¯ dP ¯ f [M ]P ¯ kM 3k−1 2k = 1 ¯ k M f [M ] ¯ P 1 3k ¯ ¯ 1− 2k dP P (11.5) shows 1. Mmax ¯ = 1. The Figure (11.0 Fig.

1[m3 ] is filled with air at pressure of 10[Bar]. At some point the flow becomes chokeless flow (unless the back pressure is a complete vacuum).37) is obtained by noticing that and therefore M ¯ is not a function of time and by variable separation results in M ¯ t ¯ P ¯= dt 0 1 ¯ dP ¯ [M ]P ¯ k Mf 3k−1 2k = 1 ¯ k Mf [M ] ¯ P 1 3k ¯ 1− ¯ 2k dP P (11.0[m] S OLUTION The first parameter that calculated is 4f L 4f L D D =5 11. equation . Nevertheless.39) It has to be realized that this is a reversed function.4. The solution of equation (11. The transition point is denoted here as chT .2 Filling Process ¯ −1 dP ¯ ¯ 3k 2k =0 ¯ − k Mf [M ]P dt The governing equation is (11. when the flow became chokeless the entrance Mach number does not change much from the choking condition.37) For a choked flow the entrance Mach number to the tube is at its maximum. Mmax ¯ = 1.11. And in the choke region the calculations can use the choking formula and numerical calculations for the rest. Thus. unless extra long time or extra low pressure is calculated/needed.5). for the special cases where the choked equation is not applicable the integration has to be separated into zones: choked and chokeless flow regions. The chamber is connected with a rubber tube with f = 0.01[m] and length of L = 5.025. The reduced time increases with decrease of the pressure in the tank.2: A chamber with volume of 0. Example 11.38) results in ¯= t −k 2 ¯ 12 k − 1 ¯ [M ] P (k − 1)Mf (11. with today computer this should not be a problem and easily can be drawn as shown here in Figure (11.38) direct integration of equation (11. Further. Again. RAPID EVACUATING OF A RIGID TANK 211 For practical purposes if the flow is choked for more than 30% of the charecteristic time the choking equation can be used for the whole range. d = 0. The Figure shows that when the modified reduced pressure equal to one the reduced time is zero.4.

4.6: The reduced time as a function of the modified reduced pressure (11. For the double isothermal (chamber and tube) the equation (11.6 0. 11.212 CHAPTER 11. the relative temperature.8 1 or ñ™òIóõô/ö"ñø÷ùòP ô/ö¯ü(ý þ ÿ úû ò¡  Fig.8 0.6 0. T isentropic tank and Isothermal flow in √the tube is different from Fanno flow in that the chocking condition occurs at 1/ k . EVACUATING AND FILLING A SEMI RIGID CHAMBERS 1 V(t) = P(t) V(t) = V(0) 0. The combination of the For Isothermal process.4 0. This model is reasonably appropriated when the chamber is insulated and not flat while the tube is relatively long and the process is relatively long.2 0.2 0 à™áPâäã3å¯æ(ç èˆéwêìë3í ð î ï 0.4 0.6) reduced into ρ c(0) ¯V ¯ ¯1 P (0) P P (0)V (0) d P ¯ (t ¯) = 0 ± A kRT (0)Mmax M ¯ tc RT (0) dt R T (0) (11.40) 11.39) has to include the entrance Mach under the integration sign as ¯ P ¯− t ¯chT = t PchT 3k 1 ¯ 1− ¯ 2k dP P ¯ k Mf [M ] (11.41) . It has to be remembered that the chamber can undergo isothermal process.3 The Isothermal Process ¯ = 1.

Further. This situation occurs at least in small ranges for airbag balloon etc. as results in faster reduction of the gas from the chamber.43) (11.4. the relation must be also positive. The analysis is based only on the mass conservation and if unsteady effects are required more terms (physical quantities) have taken into account. in the known situations to this author pressure increase results in volume decrease (at least for ideal gas. the variables are separated as ¯ t ¯ −P k+1 2k ¯ f [M ] = 0 M (11. increase of the pressure results in increase of the volume (as results of Hook’s law. After all.4.5 The “Simple” General Case The relationship between the pressure and the volume from the physical point of view must be monotonous. In this analysis and previous analysis the initial effect of the chamber container inertia is neglected. Further.46) The physical meaning that the pressure remains larger thorough evacuating process. The physical explanation when it occurs beyond the scope of this book.). . 11.42) where a is a constant that represent the physics.45) The equation (11.4.11.4 Simple Semi Rigid Chamber A simple relation of semi rigid chamber when the volume of the chamber is linearly related to the pressure as V (t) = aP (t) (11.42) into yields d ¯ 1+k P k ¯ dt Carrying differentiation result in ¯ +1 1 dP 1+k ¯k ¯ f [M ] = 0 ¯ k2 k M P −P ¯ k dt Similarly as before. RAPID EVACUATING OF A RIGID TANK 213 11.44) dt = 0 k 1+k ¯ P 1 −1 ¯ k2 ¯ k dP P ¯ M f [M ] (11. it is assumed the ideal gas applied to the gas and this assumption isn’t relaxed here. a general solution is easily can be obtained similarly to rigid tank. Nevertheless. Substituting equation (11.45) integrated to obtain the form −1 2k 2 ¯ 3k ¯= 2k t 1−P ¯ Mf [M ](3k − 1)(1 + k ) (11.

214 CHAPTER 11. This undersigned is not aware situation where this indeed play important role. it is waited to find such conditions before calling it as critical condition. EVACUATING AND FILLING A SEMI RIGID CHAMBERS Any continuous positive monotonic function can be expressed into a polynomial function.49) reduced to equation (11.47) When n can be any positive value including zero. 6 Some suggested this border point as infinite evocation to infinite time for evacuation etc. After carrying–out differentiation results ¯ +1 1 + nk − k ¯ 1+nk−2k dP ¯ f [M ] = 0 ¯ k2 k k M P −P ¯ k dt (11. In reality the value of n lays between zero to one. 0.51) The linear condition are obtain when 3k − 2nk − 1 = 1 −→ n = That is just bellow 1 (n = 0.49) Again.43). There isn’t a real critical value to n. Therefore. 3k − 2 2k (11.52) . variables are separated and integrated as follows ¯ t 0 1 + nk − k dt = k ¯ P 1 −5k ¯ 1+2nk ¯ 2k P dP ¯ [M ] Mf (11. it is convenient for engineers to further study the point where the relationship between the reduced time and the reduced pressure are linear6 Value of n above it will Convex and below it concave. The physical meaning of n = 0 is that the tank is rigid. When n is approaching to zero the chamber is approaches to a rigid tank and vis versa when the n → 1 the chamber is flexible like a balloon. Yet.4.785714286) for k = 1. d ¯ 1+nk−k k ¯ P dt +1 ¯ f [M ] = 0 ¯ k2 k M −P (11.50) Carrying–out the integration for the initial part if exit results in nk−1 2k 2 ¯ 3k −2 ¯= 2k t ¯ [M ](3k − 2nk − 1)(1 + k ) 1 − P Mf (11. similarly as before.48) Notice that when n = 1 equation (11. However. as first approximation and simplified approach can be done by a single term with a different power as V (t) = aP n (11.

equation (11. 259-265.54) Min = .5.00286. J. P0 (t) Mexit The solution of two equations (11.18) and (11. otherwise equations (11. Vol. The solution of equation (11.19) is described in “Pressure die casting: a model of vacuum pumping” Bar-Meir. For small values of the Mach numbers. no. May 1996. Goldstein. 2.18) can be approximated as 4f L D = 1 Mexit 2 − Min 2 k Mexit 2 Min 2 (11. E R G.5 Advance Topics fL is very large for small values of the entrance Mach number which The term 4D requires keeping many digits in the calculation.18) and (11. 118. G. R.54) yields 1− Pexit P0 (t) fL k 4D 2 (11. ADVANCE TOPICS 215 11.53) and equation (11. Eckert. Journal of Manufacturing Science and Engineering (USA). pp.19) must be solved numerically.55) This solution should used only for Min < 0. .53) and (11.11. (11.19) as Min Pexit = .

216 CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS .

sometime is vital to obtain a proper evacuation of the gas (air) from the cavity. This kind of model is applicable to many manufacturing processes such as die casting. One can also view the exhaust systems of internal combustion engine in the same manner. direct and through reduced area are combined in this analysis. 217 .CHAPTER 12 Evacuating/Filing Chambers under External Volume Control This chapter is the second on the section dealing with filling and evacuating chambers. The two different connections. In general the process of the displacing the gas (in many cases air) with a liquid is a very common process. The last assumption is a appropriate assumption in process such as die casting. 1 such reaction are possible and expected to be part of process but the complicates the analysis and do not contribute to understand to the compressibility effects. For example. It simplest model of such process is when a piston is displacing the gas though a long tube. again to add the dimensional analysis what is rapid and what is slow. extraction etc. 12. Two extreme possibilities again suggest themselves: rapid and slow processes. in order to obtain the essence of the process.1 General Model In this analysis. in die casting process liquid metal is injected to a cavity and after the cooling/solidification period a part is obtained in near the final shape. some simplified assumptions are made. It is further assumed that the process is relatively fast. It assumed that no chemical reaction (or condensation/evaporation) occur in the piston or the tube 1 . In these processes. Here the model deals with the case where the volume is controlled by external forces.

The first approximation isotropic process describe the process inside the cylinder (see Figure (12.218CHAPTER 12.1)). EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 12.1. isontropic process ¢¤£¦¥¨§ ©.1 Rapid Process Clearly under the assumption of rapid process the heat transfer can be neglected and Fanno flow can be assumed for the tube.

1 2 Fanno model Fig. the possibility of crack increases. Nevertheless. Obviously. thus. which is also indicated by experimental studies and utilized by practiced engineers.1) certain sizes. The replacing incompressible liquid enter in the same amount as replaced incompressible liquid. A linear function is the simplest model that decibels changes the volume. 12. In reality. The second objective is to provide a tool to “combine” the actual tube with the resistance in the tube. in some situations like die casting this description is appropriate. The purpose of this analysis to yields a formula for critical/optimum vent area in a simple form is one of the objectives of this section. . it is noteworthy to think about the process in qualitative terms. V (t) = V (0) 1 − 2 After t tmax (12. But in a compressible substance the situation can be totally different. Obtaining conditions where the volume of displacing liquid is equal to the displaced liquid are called the critical conditions.1: The control volume of the “Cylinder” Before introducing the steps of the analysis. These critical conditions are very significant that they provide guidelines for the design of processes. In the exhaust system cost of large exhaust valve increase with the size and in addition reduces the strength with the size of valve2 . For these reasons the optimum size is desired. The conflicting requirements suggest an optimum area. eliminating the need for calculations of the gas flow in the tube to minimize the numerical calculations. the best ventilation is achieved with a large tube or area. it is possible to obtain a situation where that most of the liquid entered the chamber and yet most of the replaced gas can be still be in the chamber. In manufacture processes to minimize cost and the secondary machining such as trimming and other issues the exit area or tube has to be narrow as possible. this model can be extended numerical in cases where more complex function is applied.

8) . GENERAL MODEL Equation (12. The solution of equation (12. tc is associated with the ratio of the volume and the tube characteristics (see equation (11.5)).4) Equation (12.4) is can be solved only when the flow is chocked In which case f [m] isn’t function of the time. dP k−1 reduce equation (12.2) The governing equation (11.5) can be further simplified as ¯ dt 2dξ = ¯ (k − 1) (1 − Bξ ) ξ 1−t (12.6) (k − 1) (1 − Bξ ) [ξ ] ¯ dt ¯ 1−t (12. ¯ (0) = 1.10) that was developed in the previous Chapter (11) obtained the form as ¯ P 1 k ¯ dP ¯ ¯ 1V dV + ¯ ¯ k P dt dt + ¯ f (M ) tmax M ¯ P tc k+1 2k =0 (12. tmax .4)) can be obtained by transforming and by 2k −1 ¯ k2 ¯ = [ξ ] k−1 . The reduced Presk and therefore P introducing a new variable ξ = P 2k ¯ = 2k [ξ ]( k−1 )−1 dξ Utilizing this definition and there implication sure derivative. Notice that in this case that there are two different characwhere t teristic times: the “characteristic” time.3) is an nonlinear first order differential equation and can be rearranged as follows ¯ ¯ dt dP = k−1 ¯ tmax ¯ 1 − t ¯ ¯ 2k P k 1 − tc M f [M ]P . Equation (12.3) ¯ = t/tmax .12.6) can be integrated to obtain or in a different form 1 − Bξ 2 ¯ ln = − ln t (k − 1)B ξ 1 − Bξ ξ 2 (1−k)B (12.4) 2 [ξ ]( k−1 ) 2k −1 dξ 2k k−1 = where B = tmax ¯ tc M f [M ] And equation (12.1) can be non–dimensionlassed as ¯ (t ¯) = 1 − t ¯ V 219 (12. The first characteristic time. P (12.1. The second characteristic time.7) ¯ =t (12. tmax is associated with the imposed time on the system (in this case the elapsed time of the piston stroke). tc and the “maximum” time.5) Equation (12.

A can be expressed as A= The actual times ratio tmax tc @A A Ac Ac (12. When tc > 1 the pressure increases significantly and verse versa.9) The analytical solution is applicable only in the case which the flow is choked thorough all the process.10) P (0) = 1 (12. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT Now substituting to the “preferred” variable 1− −1 tmax ¯ ¯ k2 k tc Mf [M ]P −1 ¯ k2 k P 2 t ¯ f [M ] (1−k) max M tc 1 ¯ =t ¯ P (12. In the literature. Thus.2).11) The analytical solution also can be approximated by a simpler equation as ¯ = [1 − t] P tmax tc (12.4) is transformed into k 1− ¯ dP = ¯ dt with the initial condition of −1 tmax ¯ ¯ k2 k tc Mf [M ]P ¯ 1−t (12.14) @Ac . Thus the exit area.12) The results for numerical evaluation in the case when cylinder is initially at an atmospheric pressure and outside tube is also at atmospheric pressure are presented in Figure (12.3 demonstrate the imtmax portance of the ratio of tmax tc . the question remains how the time ratio can be transfered to parameters that can the engineer can design in the system. These Figures (12. to create a direct function equation (12.). In this case only some part of the flow is choked (the later part). Ac Denoting the area that creates the ratio tmax tc provides the needed tool. The solution is applicable to indirect connection. = 1 as the critical area. This happen when vacuum is applied outside the tube (a technique used in die casting and injection molding to improve quality by reducing porosity.220CHAPTER 12.2) and 12. The results of a choked case are presented in Figure (12. In case when the flow chokeless a numerical integration needed to be performed.3) in which outside tube condition is in vacuum.13) can be expressed as 1 tmax tc = @A tmax tc @A tmax tc (12.

This quasi steady state is achieved when the volumetric air flow rate out is equal to the volume pushed by the piston. When Ac is greater than one the pressure is concave upA is less than one the pressure is concave downward as shown ward and when A c in Figures (12. equation (12. the time which would be required to evacuate the cylinder for constant mass flow rate at the maximum Mach number when the gas temperature and pressure remain in their initial A . The direct connection posts more mathematical difficulty because the critical area is not constant and time dependent. Thus. The A A .5) tc is inversely proportional to area.15) A . For small values of A there pressure in this quasi steady state is a function of A c c A is no steady state stage.3 Direct Connection In the above analysis is applicable to indirect connection. D .001[m3] and 4D = 20. The effects of the area ratio.1. . which was obtained by an integration of equation (12. and the friction Parameters influencing the process are the area ratio.4) describes the pressure as a function of the dimensionless A A time for various values of A . Ac .12. The pressure and the mass flow rate are maintained constant after this state is reached. This parameter also represents the dimensionless area. The required solidification time.1. From other detailed calculations the author thesis (later to be published on this www. The influence is small on the residual air mass in the cylinder but larger on the Mach number. according to values.1. It is important to point out the significance of the tmax tc . is not function of the time. A . tc ∝ 1/A. are studied here since it is the dominant c parameter.4). This parameter represents the ratio between the filling time and the evacuating time.2 Examples Example 12. A c the following equation Figure (12. S OLUTION 12.1: Calculate the minimum required vent area for die casting process when the die volfL ume is 0. GENERAL MODEL 221 According to equation (11. It should be noted that critical area. The line that represents A = 1 is almost straight.potto. it was found that the influence of the parameter 4f L D on the pressure development in the cylinder is quite small.14) the tmax is canceled and reduced into tmax tc = @A A Ac (12.9).03[sec].org). c c A For large values of Ac the pressure increases the volume flow rate of the air until a quasi steady state is reached. tmax = 0. A c 4f L parameter. A Mexit . 12.

This critical area depends on the geometry and the filling time. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT To continue 12.222CHAPTER 12. taking into account the compressibility of the gas flow.2 Summary The analysis indicates there is a critical vent area below which the ventilation is poor and above which the resistance to air flow is minimal. . The critical area also provides a mean to “combine” the actual vent area with the vent resistance for numerical simulations of the cavity filling.

6 0.6 2.3 0.0 o  = 100.9 1. SUMMARY 223 5.0 0. or.0 Fig.12. A/Ac 0.9 1.2 1.6 4.2: The pressure ratio as a function of the dimensionless time for chokeless condition .4 3.0 5.6 0.2 2.4 0.5 0.0 4.8 0.2 3.5 0.0 4.0 4fL D ∗ ∇ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.8 Dimensionless Area.2 0.6 4. A/Ac 0.8 0.5 1.0 2.8 1.6 2.2 1.5 1.0 ∗ ∗ o o o o o ∗ ∇ o ∇ o ∗ ∇ ∇ ∗ ∇ ∇ ∗ ∇ ∇ ∇ ∗ ∇           ∗ ∗           ∇  ∇  ∇  ∗  ∇  ∇      ∗    ∗ ∗ ∇ ∇ ∇      ∗ ∗ ∇ ∇ ∇ ∗ ∗ ∇  ∇  ∇  ∇  ∇                               ∗    ∇ ∗ o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o ∇ ∇ 0.0 0.0 2.2 0.0 0.3 0.8 1. t.4 1.2 0.0 5.0 5.2 0.2.0 1.1 0.7  Dimensionless Time.4 1.2 3.2 2.1 0.0  ∗ ∗ ∗ o o o o o o ∗ ∇ o ∇ o ∗ ∇ ∇ ∗ ∇ ∇ ∗ ∗ ∇ ∇ ∇ ∗ ∇ ∗  ∇ ∗ ∇                ∗  ∇      ∇  ∇  ∇  ∗   ∇  ∗  ∗       ∗ ∇ ∇    ∗∗∇ ∇ ∇ ∗  ∗  ∇  ∇  ∇  ∇                               ∗    ∇ ∗ ∇ o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o = 5.7 0. Cylinder Volume Fraction Figure b 0.8 Dimensionless Area.0 0. 12.0 4fL D ∗ ∇ ∗ o ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.4 3.0 0.0 1.4 Figure a 0.

. ∗ = 0.0 .0 . ∗ 1. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 2.. . ..∗. ∗ ∗ ∗ × × × × × × × × × × × × × × × × × 2 1.. A . ×.1 .. ∗ ..4: The pressure ratio as a function of the dimensionless time ... ∗ ..   ..9 DIMENSIONLESS TIME. ∗ 0. .5 0.. ∗ .0 ∗ .6 0. ∗ . or.7 0..8 1. ∗ .. . ∗ . ∗ .. Ac .. . . ..... ∗ . 0..6 .4 1. ∗ ..224CHAPTER 12..8   0.. . .2 0..2 P(t) P(0)  × 1  0. 4.1 0. ∗ . ..3: The pressure ratio as a function of the dimensionless time for choked condition Fig. . . 12.8 0.. ..3 0..2 ∗ .... ∗ ....6    × × × × × ×   × × × × × 0. ∗ . ∗ . 12.5 ∗∗ .5 ∗ .. 1. . CYLINDER VOLUME FRACTION 1 Fig..4 0. ∗ ∗ ∗ . ∗ ... ∗ .2 0 0 × ×   × × ×   ×                          × × × 0. 1..4  0. . t. .

A tube with a membrane as shown in Figure 225 . First the physical case which is analyzed is described. The unsteady effect of the compressible flow are more complicated than in incompressible flow and a special techniques are need. The technique of characteristics is introduced.CHAPTER 13 Topics in Unsteady one Dimensional gas dynamics characteristic equations and the method of characteristics.

226CHAPTER 13. TOPICS IN UNSTEADY ONE DIMENSIONAL GAS DYNAMICS .

227 . 14.B. The   other type of shock wave is the oblique shock. The common representation is done by not showing the boundaries of these models.” Bull.  normal shock. waves and asymptotics. Keller.1 Preface to Oblique Shock In Chapter (5). 1874 (1994). 727 (1978). Tabak and R. Fluids 6.G. “Focusing of weak shock waves and the Von Neuman paradox of oblique shock reflection. Further changing the deflection angle to a negative value results in expansion waves. “Rays. and E. even the whole book.CHAPTER 14 Oblique-Shock 14. A normal shock is a special type of shock wave. this section attempts to show the boundaries and the limits or connections of these models1 . Soc. Von Neumann paradox. Math. triple shock intersection. these issues should not be introduced in introductory textbook of compressible flow. In the literature oblique shock. and Prandtl–Meyer func tion are presented as three separate and different issues. However. etc are presented.” Phys. a very limited discussion about reflection shocks and collisions of weak shock. 84. discussion on a normal shock was presented. Rosales. Furthermore.R. should refer to J. Am. variation of the deflection angle from a zero (δ = 0) to a positive value results in oblique shock. Those who would like to obtain more information. Clearly. = 0 1 In this chapter. However. The author believes that these issues are not relevant to most engineering students and practices. one can view all these cases as three different Fig.1: A view of a straight normal shock as regions of a flow over a plate with a dea limited case for oblique shock flection section.

are examples of forces that create shock. However. the maximum point for the Prandtl–Meyer function is much larger than the oblique shock by a factor of more than 2. Prandtl-Meyer function approaches closer to a zero deas opposed to oblique shock analyflection angle. and different inclination angle. This matching can occur only when there is a discontinuity in the flow field. zero. This shock is commonly referred to as the oblique shock. OBLIQUE-SHOCK 14. Yet. Thus. for a negative deflection angle (in this section’s notaNo Shock zone tion). Alternatively.228 CHAPTER 14. However. the increase in the upstream Mach number determines the downstream Mach number and the “negative” deflection angle. 14.1 Introduction Introduction to Oblique Shock A normal shock occurs when there is a disturbance downstream which imposes a boundary condition on the flow in which the fluid/gas can react only by a sharp change in the flow direction.2. In such a case. Notice quirement. What accounts for the larger maximum point is the effective turning (less entropy production) which will be explained in the next chapter (see Figure (14. as discussed in Chapter (1)3 the flow behaves as it does in a hyperbolic field. For information such as the disturbance (boundary condition) to reach deep into the flow from the side requires time. It has to be pointed out that both the oblique shock and the Prandtl–Meyer function have a maximum point for M1 → ∞. pressure boundary conditions. 3 This section is under construction and does not appear in the book yet. ν∞(k ) θmax(k ) Meyer Oblique Function Shock The boundary conditions must match the geometry. As it may be recalled. normal shock occurs when a wall is straight/flat (δ = 0) as shown in Figure (14. This velocity increase is that both have a maximum point and referred to as the expansion wave.2.Fig. the gas flow must match the boundary conditions.1) which occurs when somewhere downstream a disturbance2 appears. The analysis shows that the flow ve. the flow moves downstream and creates an angle.2: The regions where oblique shock or locity must increase to achieve this rePrandtl–Meyer function exist. which is around it will be shown in the next chapter. the flow field is governed by a hyperbolic equation which deals with the case when information (like boundary conditions) reaches from downstream only if they are within the range of influence.2 14. When the deflection angle is increased. 14. As a “no solution” zone. 2 Zero velocity. . During this time. the flow must be continuous. The zero velocity can be found in a jet flowing into a still medium of gas.2 Introduction to Prandtl–Meyer Function Decreasing the deflection angle re0◦ Prandtl sults in the same effects as before.2)). sis. the direction of the flow is changed by a shock wave with an angle to the flow.

.1) 4 The author begs for forgiveness from those who view this description as offensive (There was an unpleasant email to the author accusing him of revolt against the holy of the holies. Figure (14. is the direction of the flow after the shock (parallel to the wall). If you do not like this description. the change of the orientation makes the shock considerations two–dimensional. you do not need the author’s permission.” The tangent component does not change because it does not “move” across the shock line.3 Introduction to Zero Inclination What happens when the inclination angle is zero? Which model is correct to use? Can these two conflicting models. In this chapter. the oblique shock and the Prandtl–Meyer function. which depends on the boundary layer.  However. the stability issue will be examined in greater detail. 0effects. The normal shock analysis dictates that after the shock. The two– dimensional effects occur when the flow is affected from the “side. using an infinite (or a two–dimensional) object produces a two–dimensional shock. please just ignore it. co-exist? Or perhaps a different model better describes the physics. for a chosen arbi trary accuracy it requires a specific small area. . Alternately. Hence.3: A typical oblique shock schematic considered. the flow turns after the shock to be parallel to the inclination angle. the mass balance reads Comp arsi on L ine ρ 1 U1 n = ρ 2 U2 n (14. In such a case. To match the boundary conditions.14. Only the oblique shock’s normal component undergoes the “shock. Previously (see Chapter 5). it was assumed that normal shock occurs at the same time. You can use the traditional explanation.). the flow is always subsonic.-/. OBLIQUE SHOCK 229 14. In some books and in the famous NACA report 1135 it was assumed that Mach wave and oblique shock co–occur in the same zone. The three–dimensional 1#243 effects of the shock make it appear as a curved plane.e. The total flow after the oblique shock can also be supersonic. a one– dimensional shock can be Fig.” i. δ . a change in the flow direction4 . 14. 14.3) exhibits the schematic of the oblique shock.3.2.3 Oblique Shock The shock occurs in realθ−δ $ %'& ity in situations where the ( )* + + shock has three–dimensional θ   !#" . The deflection angle.

2) (14. who raised the solution completeness issue.2). OBLIQUE-SHOCK P1 + ρ 1 U 1 n 2 = P 2 + ρ 2 U 2 n 2 The momentum equation in the tangential direction yields U1 t = U 2 t The energy balance reads C p T1 + U1 n 2 U2 2 = C p T2 + n 2 2 (14. tan θ U1 n = tan(θ − δ ) U2 n (14. and the third is an impossible solution (thermodynamically)6 . the second is the strong shock.4) Equations (14. and the Mach angle.230 The momentum equation reads CHAPTER 14. θ has to be solved. The lesser extent angle is the unstable angle and the weak angle is the middle solution. Yet the new relationship between the upstream Mach number. the deflection angle. the term lesser extent is used.7) The above velocity–geometry equations can also be expressed in term of Mach number.3) (14. (14. .4) are the same as the equations for normal shock with the exception that the total velocity is replaced by the perpendicular components.6) U1 n U1 t (14.8) issue is due to R. here there are three possible pairs5 of solutions to these equations. From the geometry it can be observed that tan θ = and tan(θ − δ ) = U2 n U2 t (14. as sin θ = 6 The solution 5 This M1 n M1 (14. Menikoff.5) Unlike in the normal shock. The author is not aware of “simple” proof and a call to find a simple proof is needed.1). 7 Actually this term is used from historical reasons. δ . and (14. The first is referred to as the weak shock. Experiments and experience have shown that the common solution is the weak shock. in which the shock turns to a lesser extent7 . But because the literature referred to only two roots. requires solving the entropy conservation equation.

15) (14.9) (14. θ. It should be noted that although. and δ . here angles (the second dimension) have to be determined. which results in P2 2kM1 2 sin2 θ − (k − 1) = P1 k+1 ρ2 U1 n (k + 1)M1 2 sin2 θ = = ρ1 U2 n (k − 1)M1 2 sin2 θ + 2 (14.3.17) ρ2 − ρ 1 ρ2 − ρ 1 . Rearranging this set utilizing geometrical identities such as sin α = 2 sin α cos α results in tan δ = 2 cot θ M1 2 sin2 θ − 1 M1 2 (k + cos 2θ) + 2 (14. is a function of four unknowns of M1 .11). (14.11) M1 t M1 M2 t M2 cos(θ − δ ) = The total energy across an oblique shock wave is constant. The solution from this set of four equations. U1 t = U2 t the Mach number is M1 t = M2 t because the temperatures on both sides of the shock are different. and it follows that the total speed of sound is constant across the (oblique) shock.14) The temperature ratio is expressed as 2kM1 2 sin2 θ − (k − 1) (k − 1)M1 2 + 2 T2 = T1 (k + 1)2 M1 Prandtl’s relation for oblique shock is U n1 U n2 = c 2 − k−1 2 Ut k+1 (14.13) The density and normal velocity ratio can be determined by the following equation (14.14.8) through (14.16) The Rankine–Hugoniot relations are the same as the relationship for the normal shock P2 − P 1 P2 − P 1 =k (14. M2 .12) The relationship between the properties can be determined by substituting M1 sin θ for of M1 into the normal shock relationship.10) (14. As opposed to the normal shock. T1 = T2 . OBLIQUE SHOCK M2 n M2 231 sin(θ − δ ) = cos θ = (14.

Oblique shock is a two–dimensional problem and two properties must be provided so a solution can be found. perhaps. sin θ must be positive. Thus. Other combinations of properties provide a relatively simple mathematical treatment. One has to be careful to use the normal components of the Mach numbers. thus. 14. He also suggested that θ is bounded by sin−1 1/M1 and . only one parameter is required (to solve the problem).4. Again.1 Upstream Mach Number.18) requires that x has to be a real and positive number to obtain a real deflection angle8 . M1 .22) (14. Thompson (1950) has shown that the relationship of the shock angle is obtained from the following cubic equation: x3 + a 1 x2 + a 2 x + a 3 = 0 where x = sin2 θ and M1 2 + 2 − k sin2 δ M1 2 2M1 2 + 1 k−1 (k + 1)2 a2 = − + sin2 δ + 4 4 M1 M1 2 cos2 δ a3 = − M1 4 a1 = − (14. The stagnation temperature contains the total velocity.20) (14. Clearly.4 Solution of Mach Angle Oblique shock. M1 and the deflection angle.19) (14. and the negative sign refers to the mirror image of the solution. Probably. Menikoff. this set of parameters is. OBLIQUE-SHOCK 14. the normal shock is a one–dimensional problem. δ Again.232 CHAPTER 14. if orientated to a coordinate perpendicular and parallel shock plane is like a normal shock.21) (14. and the solutions of selected pairs and selected relationships will be presented. 8 This point was pointed out by R. Thus. and Deflection Angle. the negative root of sin θ must be disregarded 1. the most useful properties are upstream Mach number. which create a somewhat complicated mathematical procedure. and this will be discussed later. the relationship between the properties can be determined by using the normal components or by utilizing the normal shock table developed earlier. the most common and natural to examine.18) Equation (14.

26) T = and where the definition of the D is 3 R− √ D (14. to explain about complex numbers and imaginary numbers should be included.29) (14.24) (14. (14.28) (14. .25) R+ √ D. one root is real and two roots are complex. all the roots are real and unequal.4. SOLUTION OF MACH ANGLE 233 The solution of a cubic equation such as (14. if D > 0.18) provides three roots9 . Maybe insert an example where imaginary solution results in no physical solution.30) Only three roots can exist for the Mach angle.27) D = Q3 + R 2 and where the definitions of Q and R are Q= and R= 9a1 a2 − 27a3 − 2a1 3 54 3a2 − a1 2 9 (14. For the case D = 0. D > 0 occurs when no shock angle can be found. The physical meaning of the above analysis demonstrates that in the range where D > 0 no solution can exist because no imaginary solution can exist10 .14. From a mathematical point of view. θ. These roots can be expressed as 1 x1 = − a1 + (S + T ) 3 1 1 1 √ x2 = − a1 − (S + T ) + i 3(S − T ) 3 2 2 and 1 1 √ 1 x3 = − a1 − (S + T ) − i 3(S − T ) 3 2 2 Where S= 3 (14. 10 A call for suggestions. all the roots are real and at least two are identical. In the last case where D < 0. For example.23) (14. 9 The highest power of the equation (only with integer numbers) is the number of the roots. in a quadratic equation there are two roots. so that the shock normal component is reduced to subsonic and yet parallel to the inclination angle.

the body will freely rotate only around two axes with small and large moments of inertia. in steady–state situations. referred sometimes as a thermodynamically unstable root. These two angles of the strong and weak shock are stable for a two–dimensional wedge (see the appendix of this chapter for a limited discussion on the stability14 ). The cubical body has three symmetrical axes which the body can rotate around. is “unrealistic. The reader can simply try it. 13 See the hist/rical discussion on the stability. OBLIQUE-SHOCK Furthermore. the third root always turns the flow into subsonic and it is referred to as the strong solution. only in some cases when D = 0 does the solution have a physical meaning. at least. The shock will initially for a very brief time transient in it and will jump from this angle to the thermodynamically stable angles.23). the shock angle will jump to the third root13 . Hence. It should be noted that this case is where entropy increases in the largest amount. if a hand moves the shock angle starting from the deflection angle and reaching the first angle that satisfies the boundary condition. There is no theoretical evidence that showing that in strong unsteady state this angle is unstable. the shock wave keeps the flow almost all the time as a supersonic flow and it is referred to as the weak solution (there is a small section that the flow is subsonic). . the first solution does not occur in reality. Second. Though.234 CHAPTER 14. However. The “unstable” terms can be thermodynamcily stable in unsteady case. the solution in the case of D = 0 has to be examined in the light of other issues to determine the validity of the solution. it can be shown that the first solution(14. First. that the author has found. In summary. These two roots represent two different situations. There are those who view this question not as a stability equation but rather as under what conditions a strong or a weak shock will prevail. If an additional “push” is given. 14 This material is extra and not recommended for standard undergraduate students. by additional boundary conditions. When D < 0. The body rotation is unstable around the middle axes. 11 This situation is somewhat similar to a cubical body rotation. 12 There is no experimental or analytical evidence. for the second root. showing that it is totally impossible. Physically. this situation is unstable and the shock angle will jump to the second angle (root). This root has only a mathematical meaning for steady–state analysis12 . for example. the three unique roots are reduced to two roots at least for the steady state because thermodynamics dictates11 that. those who are dealing with rapid transient situations should be aware that this angle of oblique shock can exist. which is also related to a decrease in entropy.” Therefore.

a change in the inclination angle requires a larger energy to change the flow direction. M1 −→ ∞ the maximum deflection angle is calculated when D = Q3 + R2 = 0. It can be noticed that the normal shock.5).4. in the case k = 1. This can be done by evaluating the terms a1 . k ). This different inclination angle is sometimes referred to as an imaginary angle.4.28) after substituting these values of Q and R provides series of roots from which only one root is possible. and the weak shock coexist. For example. The flow is far away from the object and does not “see’ the object. 14. the wedge) in front of it and hence the normal shock occurs. for. Only when the fluid is away from the object (smaller angle) liquid “sees” the object in a different inclination angle. Fig.4: Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. the strong shock. the inclination angle reaches the “maximum potential energy. Once. . the fluid “sees” the disturbance (in this case. In the alternative view.4.14. a1 = −1 − k sin2 δ 2 a2 = (k + 1) sin2 δ 4 a3 = 0 With these values the coefficients R and Q are 9(−)(1 + k sin2 δ ) (k+1)2 sin2 δ 4 R= and − (2)(−)(1 + k sin2 δ )2 54 Q= (1 + k sin2 δ )2 9 Solving equation (14.0. The simple procedure For example. a2 . is just above δmax ∼ π 4 (note that the maximum is also a function of the heat ratio. M1 . This root. SOLUTION OF MACH ANGLE 235 14.2 When No Oblique Shock Exist or When D > 0 Large deflection angle for given. the imaginary angle is shown. For a given upstream Mach number. M1 The first range is when the deflection angle reaches above the maximum point. in Figure (14. and a3 for M1 = ∞.” a change in the flow direction is no longer possible.4) and (14.

This can be shown mathematically by the argument that differentiating equation (14. Since in that equation there appears only the heat ratio k . It can be noticed that in equation (14.29). M1 and the maximum Mach angle. M1 .20)-(14. When the right terms defined in (14. OBLIQUE-SHOCK The fluid doesn’t ’’see’ the object M∞ } } } The fluid ‘‘sees’’ the object with "imaginary" inclanation angle Intermediate zone The fluid "sees" the object infront Fig. the deflection angle is a function of the Mach angle and the upstream Mach number. J.21). θ. . 14. Thus.12). it was seen as C. M1 . At this stage it is not clear who was the first to propose it. one can conclude that the maximum Mach angle is only a function of the upstream Much number. The Procedure for Calculating The Maximum Deflection Point The maximum is obtained when D = 0. and Mach number.12) and equating the results to zero creates relationship between the Mach number. θmax is a function of only these parameters.30) are substituted into this equation and utilizing the trigonometrical sin2 δ + cos2 δ = 1 and other trigonometrical identities results in Maximum Deflection Mach Number’s equation in which is M1 2 (k + 1) (M1n 2 + 1) = 2(kM1n 4 + 2M1n 2 − 1) (14.5: The view of a large inclination angle from different points in the fluid field. M1 .236 CHAPTER 14. (14. 15 At first.31) This equation and its twin equation can be obtained by an alternative procedure proposed by someone15 who suggested another way to approach this issue. and (14. English mathematician to be the creator but later an earlier version by several months was proposed by Bernard Grossman.Chapman.

It can be noticed that the Maximum Deflection Mach Number’s equation is also a quadratic equation for M1n 2 .32) is zero at different values of the denominator.36) . SOLUTION OF MACH ANGLE The differentiation of the equation (14. The nominator produces a quadratic equation for sin2 θ and only the positive value for sin2 θ is applied here. Thus. Thus. The numerator of equation (14.14.12).33) Equation (14. The general solution of equation (14.18 or θ = 67. A simplified case of the Maximum Deflection Mach Number’s equation for large Mach number becomes M1n = k+1 M1 2k for M1 >> 1 (14. Thus. then the Mach angle can be easily calculated by equation (14. the Mach angle is sin θ = 2k .4. With the value of θ utilizing equation (14.12) yields 2 2 kM1 4 sin4 θ + 2 − (k+1) M1 2 sin2 θ − 1 + (k+1) 2 M1 2 M1 d tan δ = 2M 2 1 dθ kM1 4 sin4 θ − (k − 1) + (k+1) M1 2 sin2 θ − 1 4 237 (14. a derivative can be taken with respect to M1 .31) is (k + 1)M1 2 + 1 + M1n = (M1 2 M1 2 (k + 1)2 + 8(k 2 − 1) + 16(1 + k ) √ 2 k (14. the maximum appears when θ has its maximum.34) provides a solution for Mmax .79 . the sin2 θ is −1 + 2 k+1 4 M1 + (k + 1) 1 + kM1 2 sin θmax = 2 2 k−1 2 M1 + 4 k+1 2 M1 (14. taking the derivative with respect to θ. for large Mach numbers. To compare these two equations the simple case of Maximum for an infinite Mach number is examined. Note that this procedure does not require an approximation of M1n to be made. Once M1n is found. It must be pointed out that similar procedures can also be proposed (even though it does not appear in the literature).33) should be referred to as the maximum’s equation. d tan δ =0 dM1 (14.8). It should be noted that both the Maximum Mach Deflection equation and the maximum’s equation lead to the same conclusion that the maximum M1n is only a function of upstream the Mach number and the heat ratio k .34) and then solving equation (14.35) k+1 Hence. it is sufficient to equate the numerator to zero to obtain the maximum. the maximum deflection angle can be computed. which makes ◦ θ = 1.32) Because tan is a monotonous function. Instead.

a2 . This typical example is for those who like mathematics. The analysis shows regardless of the value of the upstream Mach number D = 0 when δ = 0.37) for R and Q becoming 9 R= 2 M1 2 +2 M1 2 2M1 2 +1 M1 4 2 (14. first the transition line in which D = 0 has to be determined. Start with equation (14. OBLIQUE-SHOCK Note that Maximum Deflection Mach Number’s equation can be extended to deal with more complicated equations of state (aside from the perfect gas model).31) and neglect all the terms that are relatively small. Hint. and a3 for the specific value of M1 as following M1 2 + 2 M1 2 2M1 2 + 1 a2 = − M1 4 1 a3 = − M1 4 a1 = With values presented in equations (14.1: Derive the perturbation of Maximum Deflection Mach Number’s equation for the case of a very small upstream Mach number number of the form M1 = 1 + . Thus. Example 14. This can be partially demonstrated by evaluating the terms a1 . This can be achieved by the standard mathematical procedure of equating D = 0.37) − 27 −1 M1 4 −2 M1 2 +2 M1 2 2 = and 9 M1 + 2 2M1 54 + 1 + 27M1 2 − 2M1 2 M1 2 + 2 54M16 2 (14.38) 3 Q= 2M1 2 +1 M1 4 − 9 M1 2 +2 M1 2 3 (14.39) .238 CHAPTER 14. S OLUTION under construction The case of D ≥ 0 or 0 ≥ δ The second range in which D > 0 is when δ < 0.

4. in the literature.0 -3 -1 . Fluids 10 (1) .. Table (14. In the range where δ ≤ 0.Fig. Tabak. no Mach wave 2.” 206 Phys. the physical interpretation of this fact is that either no shock exists and the flow is without any discontinuity or that a normal shock exists17 .6: The various coefficients of three different tradicts the common approach. Hence. others suggest a singular point18 . 17 There 16 A . SOLUTION OF MACH ANGLE 239 Substituting the values of Q and R equations (14. 14. in the previous case. mathematical challenge for those who like to work it out. which con. there are several papers suggesting zero strength Mach wave. Substitution of all the above values into (14. Nevertheless. M1 = 1. ∞. the XX question is whether it is possiXXcoefficients XXX a1 a2 a3 ble for an oblique shock to exXX M1 X ist? The answer according to 3 this analysis and stability anal1.  3  2M1 2 +1 M1 4 − 9 M1 2 +2 M1 2 3 3 9 M1 2 + 2 2M1 2 + 1 + 27M1 2 − 2M1 2 M1 2 + 2 54M16   + 2 2 = 0 (14.39) into equation (14. “Caustics of weak shock waves. Mach numbers to demonstrate that D is zero Nevertheless.14. 18 See for example.38) (14.6) is provided for the following demonstration. this proof became trivial.0. the wall does not emit any signal to the flow (assuming zero viscosity). And according to 9 this analysis. 2. January 1998. with a positive large deflection angle.40) The author is not aware of any analytical demonstration in the literature which shows that the solution is identical to zero for δ = 016 . Once this analytical solution was published. In other ∞ -1 0 .16 words. are several papers that attempt to prove this point in the past. there was a transition from one kind of discontinuity to another. paper by Rosles. The question of singular point or zero Mach wave strength are only of mathematical interest. Note that.2 ysis is no. Utilizing the symmetry and antisymmetry of the qualities of the cos and sin for δ < 0 demonstrates that D > 0 regardless of Mach number. this identity can be demonstrated by checking several points for example.0 3 0 16 can be generated from the wall 1 with zero deflection.28) results in D = 0. But for non ideal gas (real gas) this solution is only an indication.28) provides the equation to be solved for δ .

distance.7: The Mach waves that are supposed uation keeps on occurring over a finite to be generated at zero inclination. There must be another boundary condition(s) that causes the normal shock. i. the same conclusion must be reached. another Mach wave occurs after a small distance. If the Mach number is large enough and the wall is rough enough. for simplicity’s sake. if it was assumed that no boundary layer exists and the wall is perfect. both will not allow a situation of zero inclination with emission of Mach wave. the increase and the decrease depend on the upstream Mach number but in different directions. The first “weak” means more of compression “line” while the other means the weak shock. In the light of this discussion. there are two “weaks. no Mach wave at zero inclination should be assumed. OBLIQUE-SHOCK Suppose that there is a Mach wave at the wall at zero inclination (see Figure (14. in the engineering world. the zone in the proximity of zero inclination (small positive and negative inclination angle) should be viewed as a zone without any change unless the boundary conditions cause a normal shock. In reality. the reality is that no continuous Mach wave can occur because of the viscosity (boundary layer). at least for a simple engineering analysis. emission of Mach wave can occur in other situations. Furthermore. the Mach angle will be larger (µ2 > µ1 ). according the common explanation. Therefore. that the normal shock can occur at zero inclination. that there is no such thing as a perfect wall. assumed to be as a sinusoidal shape. for the purpose of an introductory class. On the other hand. For such a wall the zero inclination changes from small positive value to a negative value.” These words mean two different things. Furthermore. Obviously. 14.e. At the very extreme case. there will be points where a weak19 weak will be created. The analysis started by looking at a normal shock which occurs when there is a zero inclination.240 CHAPTER 14. If the sitFig. there are imperfections in the wall and in the flow and there is the question of boundary layer. there will be a point where the Mach number will be 1 and a normal shock will occur. The imperfections of the wall can be. This theoretical jump occurs because in a Mach wave the velocity decreases while in the expansion wave the velocity increases. detach normal shock). This jump has to be in reality either smoothened out or has a physical meaning of jump (for example. thus. Nevertheless. any deviations from the zero inclination angle creates a jump from a positive angle (Mach wave) to a negative angle (expansion wave). It is well known. With these conflicting mechanisms.7)). After analysis of the oblique shock. The analysis of the oblique shock suggests that the inclination angle is not the source (boundary condition) that creates the shock. But because the velocity after a Mach wave (even for an extremely weak shock µ1 µ2 µ3 µ∞ wave) is reduced. The approxi19 It is not a mistake. the boundary layer covers or smoothens out the bumps. However. only in several points (depending on the bumps) at the leading edge can a very weak shock occur. .

M1 . and Shock Angle.14.0005 0 0.0 30.3 Upstream Mach Number. θ δ 20 30 The solution for upstream Mach number.5 1 0. especially in acoustic engineering.0 δ 20. SOLUTION OF MACH ANGLE 241 mation of weak weak wave with nonzero strength has engineering applicability in a very limited cases.4.001 k = 1 4 Mx=3 0 10 20 30 Wed Jun 22 15:03:35 2005 Fig. The deflection angle can be expressed as a function of these variables as cot δ = tan θ (k + 1)M1 2 −1 2(M1 2 sin2 θ − 1) (14.001 0.0 -0. are far much simpler and a unique solution exists.4. and exit Mach number for M1 = 3 0 10 14.5 2 1.41) . θ.5 0 90 80 70 60 50 40 30 20 10 0 -0. M1 .0 10.8: The calculation of D (possible error). Oblique Shock 3 2. 14.0005 Myw Mys θs θw 0. and shock angle. shock angle. but for most cases it should be ignored.

the possible range deflection angle. M1 is M1 2 = 2(cot θ + tan δ ) sin 2θ − (tan δ )(k + cos 2θ) (14. is unique.42) The pressure ratio can be expressed as P2 2kM1 2 sin2 θ − (k − 1) = P1 k+1 The density ratio can be expressed as U1 n (k + 1)M1 2 sin2 θ ρ2 = = ρ1 U2 n (k − 1)M1 2 sin2 θ + 2 The temperature ratio expressed as 2kM1 2 sin2 θ − (k − 1) (k − 1)M1 2 sin2 θ + 2 c2 2 T2 = 2 = T1 c1 (k + 1)M1 2 sin2 θ The Mach number after the shock is M2 2 sin(θ − δ ) = or explicitly M2 2 = (k − 1)M1 2 sin2 θ + 2 2kM1 2 sin2 θ − (k − 1) (14. θ .9).45) (14. M1 and θ.49) . is limited. The first upstream Mach number. has to be in the range of sin−1 (1/M1 ) ≥ θ ≥ (π/2) (see Figure 14. The range of given θ.48) Even though the solution for these variables.242 or tan δ = CHAPTER 14. upstream Mach number M1 . OBLIQUE-SHOCK 2 cot θ(M1 2 sin2 θ − 1) 2 + M1 2 (k + 1 − 2 sin2 θ) (14.4. δ and θ It is sometimes useful to obtain a relationship where the two angles are known.46) (14.4 Given Two Angles.41) shows that the shock angle. Examining equation (14.44) (14.47) (14.43) The ratio of the total pressure can be expressed as P0 2 (k + 1)M1 2 sin2 θ = P0 1 (k − 1)M1 2 sin2 θ + 2 k k−1 (k + 1)2 M1 4 sin2 θ − 4(M1 2 sin2 θ − 1)(kM1 2 sin2 θ + 1) 2kM1 2 sin2 θ − (k − 1) (k − 1)M1 2 sin2 θ + 2 k+1 2kM1 2 sin2 θ − (k − 1) 1 k −1 (14. 14. is limited between ∞ and 1/ sin2 θ. δ .

.52) The results are consistent with the initial assumption which shows that it was an appropriate assumption.51) (14. Equation (14.14.41) can be simplified into k+1 θ∼ δ = 2 (14. SOLUTION OF MACH ANGLE subsonic weak solution 243 1. Shock angle π θ= 2 θmax ∼ π 2 θ=0 Fig.50) For a large upstream Mach number M1 and a small shock angle (yet not approaching zero). δ must also be small as well.4. the deflection angle.0 < M1 < ∞ Defection angle strong solution θmin = sin −1 1 M1 supersonic weak soution possible solution no solution zone θ. θ. 14.9: The possible range of solutions for different parameters for given upstream Mach numbers The reduced pressure difference is 2 sin θ sin δ 2(P2 − P1 ) = ρU 2 cos(θ − δ ) The reduced density is sin δ ρ2 − ρ 1 = ρ2 sin θ cos(θ − δ ) (14.

” Another geometry that can be considered as two–dimensional is the cone (some referred to it as Taylor–Maccoll flow). the symmetrical nature of the cone creates a semi–2D problem.6 Small δ “Weak Oblique shock” This interest in this topic is mostly from an academic point of view. As opposed to the weak shock. is neglected. this solution is also act as a “reality check” to any numerical analysis. Experimental and analytical work shows that the weak solution is the stable solution and a discussion is provided in the appendix of this chapter. for steady state and no known experiments showing that it exist can be found in the literature.4. The dimensional analysis shows that only the doted area to Fig. The . Even though.10)).10: Schematic of finite wedge with zero be area where current solution can be angle of attack. The section where the current analysis is close to reality can be esti2-D oblique shock on both sides mated from a dimensional analysis for s i s aly n the required accuracy or by a numera edge range ical method. The end of the wedge must have a different configuration (see Figure (14. at least. the sis naly al a norm range analysis needs to account for edge efno shock fects. The common approach is to carry out numerical analysis.L. known to the author. Additionally.5 Flow in a Semi–2D Shape The discussion so far was about the straight infinite long wedgea which is no shock flow direction ysis anal a “pure” 2–D configuration. b assumed as correct . the cone is a three–dimensional problem. All the literature. The analysis also provides additional value of the expected range. Clearly. the analysis for the middle section produces a close result to reality (because of symmetry).4. for most engineering applications. In spite of recent trends. the first solution is thermodynamical unstable. dimensional analysis is not completed. ate i d i rm range inte for any finite length of the wedge. Yet. reports that only a weak shock is possible. b At this stage.244 CHAPTER 14. OBLIQUE-SHOCK 14. { { { { { 14. As before. the strong shock is unstable. 14. a Even finite wedge with limiting wall can be considered as an example for this discussion if the B. In spite of the small area were the current solution can be assumed. In this case there are no edge effects and the geometry dictates slightly different results. the numerical works require many times a “reality check. The author is not aware of any such analysis in literature. The mathematics is much more complicated but there are three solutions. It is recommended that this issue be skipped and the time be devoted to other issues. a simple tool is sufficient for limit accuracy.

4 The maximum values are summarized in the following Table .) does not enter into the calculation. In this theory. In fact.53) tan θ = . and small normal upstream Mach numbers.11). From the far view. Mx 1. 14.4.97131 δmax 1.11: A local and a far view of the imity to the wall. However. In larger distance from the wall. the boundary layer increases the zone where a continuous δ flow exists. under construction. the oblique shock is an accumulation of many small (or again weak) “weak shocks.7 Close and Far Views of the Oblique Shock In many cases.. For small deflection angles.1000 My 0. SOLUTION OF MACH ANGLE 245 author is not aware of any single case in which this topic is used in real–world calculations.5152 θmax 76. The traditional approach to reconcile these two views is by suggesting that the far view shock is a collection of many small weak shocks (see Figure 14.2762 .” However. 14. 1 M1 2 − 1 (14. Yet. In reality.. M1 ∼ 1+ . 14. δ .8 Maximum Value of Oblique shock Table 14. The boundary layer reduces the upstream flow velocity and therefore the shock does not exist at close proxFig.4. these small “shocks” are built or accumulate into a large and abrupt change (shock). the shock becomes possible. it should be covered in this book. after the explicit analytical solution has been provided. oblique shock. δ .1: Table of maximum values of the oblique Shock k=1. the boundary θ layer (B. the author admits that as long as there are instructors who examine their students on this issue. At the local view close to the wall. studying this topic seems to come at the expense of other more important topics. the close proximity view provides a continuous turning of the deflection angle.14. the oblique shock is a weak “weak oblique” shock.4.L. the far view shows a sharp transition.

95897 0.9087 66.99337 0.0000 3. there isn’t a physical solution to an oblique shock.7908 44.3066 64.5875 34. In some cases.0399 65.8443 65.6000 2. 21 The effect of the equation of state on the maximum and other parameters at this state is unknown at this moment and there are more works underway.6515 17.6621 9.4000 1.93083 0.0734 35.9442 6.2000 3. for large deflection angle.5787 65.7972 65.94387 0.246 CHAPTER 14.4000 3.3000 1.1196 67.1028 28.4398 43.99440 0.9 Detached shock When the mathematical quantity D becomes positive.0000 8.94925 0.91941 0. 179–210.8000 4.9555 69.2000 2.96335 0.95435 0.6000 3.92478 0. OBLIQUE-SHOCK Table 14.0000 2.8000 3. this assumption might not be sufficient and different analysis is needed.0922 38.2309 65.7532 64.8000 1.1675 22.7593 65.7739 41.5000 1.6934 64.8137 32. the only possible reaction is by a normal shock which occurs at some distance 20 Henderson and Menikoff ”Triple Shock Entropy Theorem” Journal of Fluid Mechanics 366 (1998) pp.0000 My 0.1177 42.92683 0.1127 14.3645 67.6814 30.91997 0.4419 It must be noted that the calculations are for the perfect gas model.3059 38.4272 12.99047 0.1: Maximum values of oblique shock (continue) k=1. Henderson and Menikoff20 suggested a procedure to calculate the maximum deflection angle for arbitrary equation of state21 .4 Mx 1.92165 0.0000 10.7023 66.98714 0. 14.6465 64. Since the flow “sees” the obstacle.0119 19.93747 0.3934 37.0000 5.2503 67.6000 1.2000 1.2546 43.4.92224 0.0000 9.7000 1. .4290 θmax 71.95049 0.3275 36.91871 0.5676 65.3673 67.0000 6.9000 2.99559 δmax 3.0000 7.97214 0.4000 2.6074 64.1833 21.4144 65.96942 0.0464 66.96630 0.98183 0.93629 0.9020 67.1619 44.5671 66.9735 26.9668 64.

Regardless of the explanation.4. it can be explained as the subsonic is “sucked” into gas in zone C. Analysis of the detached shock can be carried out by looking at a body with a round section moving in a supersonic flow (the absolute velocity 8 9M : .CG isn’t important for this discussion).12 exhibits a round–tip bulzone B Subsonic Area θ let with a detached shock. 14. In the typical example.54) where f (M∞ ) is a function of the upstream Mach number which tabulated in the literature.12: The schematic for a round–tip bullet C the flow must be supersonic. 14. Thus. alternatively.4.5. The detached shock’s distance from the body is a complex analysis and should be left to graduate class and researchers in this area. Larger shock results in a smaller detachment distance. The explanation to such a phenomenon is above the level of this book (where is the “throat” area question22 . As a general rule. this phenomenon has a larger impact for a relatively smaller supersonic flow. The zone A is zone where the flow must be subsonic because at the body the velocity must be zero (the no–slip condition). This shock is referred to as the detach shock. The flow in zone A has to go through some acceleration to became supersonic flow. SOLUTION OF MACH ANGLE 247 from the body. the increase in the upstream Mach results in a decrease of the detachment distance. a wedge or 22 See example 14.14. The constant and the function are different for different geometries. The diszone A Normal Shock tance of the detachment is deter5 67 mined to a large degree by the upH4I J K L M NOI P J'H4Q R J K P L I S T U R V NONWI J U Q L K X Y T Z V R stream Mach number. Nevertheless.10 Issues Related to the Maximum Deflection Angle The issue of maximum deflection has a practical application aside from the obvious configuration used as a typical simple example. or. some might be used in certain situations which the author isn’t aware of. While at zone Fig. the flow becomes “blinder” to obstacles. Yet. . Even though this topic has few applications. In such a case. these calculations can be summarized by the flowing equation detachment distance = constant × (θ − f (M∞ )) body thickness w ea k sh oc k (14. Upstream U∞ Figure 14. E Strong Shock F > @ . the gas must go through a shock. a graph and a general explanation to engineers is provided. The in a supersonic flow weak oblique shock is predicted to flow around the cone. <> = > ?1@ ? ACBD< .

. 14. This detached shock ferred to as Mach reflection manifested itself in a form of curved shock (see Figure 14.55) Additionally. for U B ikj every upstream Mach number there C θ2 Slip Plane is only one maximum deflection anA δ2 gle. only a function of the upstream Mach number. The Mach number decreases after every shock. OBLIQUE-SHOCK a cone moves into a still medium or gas flows into it. Consider. This situation is known in the scientific literature as the Mach reflection.e. for example. 14. there are configurations in which a detached shock occurs in design and engineers need to take it into consideration. The calculations for such a case can be carried out by several approaches.14). Additionally. It can be observed that the maxi[ \¡] ^ _`¡acbDde[fCghb mum of the oblique shock for the δ1 θ1 perfect gas model depends only on the upstream Mach number i. The analysis of this situation is logically very simple. hence δ1 = δ2 . the maximum deflection angle decreases with a decrease the Mach numA sub sonic ber. If the deflection angle exceeds the maximum possible. ThereU B C fore. The maximum deflection angle in this case is. due to the symmetry flow a slip plane angle can be guessed to be parallel to original flow. No detached shock issues are raised when only the first oblique shock is considered. In this situation. it can be observed for Fig.13: The schematic for a symmetrical suca maximum oblique shock that a contion section with Mach reflection stant deflection angle decrease of the l/m#n o prqrsutwvrlyx zyt Mach number results in an increase of δ1 Mach angle (weak shock only) M1 > θ1 M2 =⇒ θ1 < θ2 . as before. However. It seems that the most straightforward method is the following: (a) Calculate M1 B . a detached shock occurs.14: The “detached” shock in a complicated configuration sometimes reoblique shock.248 CHAPTER 14. However. at least two oblique shocks occur and after their interaction is shown in Figure (14. Thus. a symmetrical suction section in which the deflection angle is just between the maximum deflection angle and above half of the maximum deflection angle. Such configurations seem sometimes at first glance not related to the detached shock issue. the second oblique shock complicates the situation and the second oblique shock can cause a detached shock. this situation causes the detached shock to appear in the second Fig.. δmax = f (M1 ) (14.13). yet the mathematics is somewhat complicated.

δ2 = δ1 and the Mach number M1 B to calculate M1 B .4. δ2 utilizing equation (14. M1 by checking the relationship between the intermediate Mach number to M1 .4. calculate the weak. SOLUTION OF MACH ANGLE (b) Calculate the maximum deflection angle.11 Oblique Shock Examples Example 14.48523 My w 2. utilizing (14. 14. In general.31) equation (c) Calculate the deflection angle.5686 θs 1. Using the geometry provided in the photo. predict at which Mach number was the photo taken based on the assumption that the cone is a wedge.14.97234 δmax 38.12) 249 (d) Use the deflection angle.3: A cone shown in Figure (14. one must be aware that there are zones of dual solutions in which sharp shock line coexists with a curved line.15) is exposed to supersonic flow and create an oblique shock.7738 θmax 66. .0000 My s 0. POTTO–GDC can be used to calculate this ratio. Note that no maximum angle is achieved in this shock.2: Air flows at Mach number (M1 ) or Mx = 4 is approaching a wedge.34907 Example 14.4635 θw 0. What is the maximum wedge angle at which the oblique shock can occur? If the wedge angle is 20◦ . and the respective shock angles. For engineering purposes when the Mach number reaches this value.28) or the GDC which yields the following results Mx 4. This procedure can be extended to calculate the maximum incoming Mach number. For example. In discussing these issues.0407 To obtain the results of the weak and the strong solutions either utilize the equation (14. at Mach 5 this zone is 8. the strong Mach numbers.56660 δ 0.0000 My 0. S OLUTION The maximum wedge angle for (Mx = 4) D has to be equal to zero. it can be ignored.28) (a side to the case proximity of δ = 0). The wedge angle that satisfies this requirement is by equation (14. The maximum values are: Mx 4. Is the shock shown in the photo weak or strong shock? Explain. θ2 .5◦ . this zone increases as Mach number increases.

According to his measurement.15: Oblique shock occurs around a cone. 14.2318 My s My w θs θw δ P0 y P0 x 0. S OLUTION The measurement shows that cone angle is 14. the shock would be weak because the maximum (transition to a strong shock) occurs at about 60◦ .88737 Because the flow is around the cone it must be a weak shock.56543 2. a Research Professor at Cuernavaco University of Mexico.49) or the Potto-GDC. Grigory Toker. OBLIQUE-SHOCK θ δ Fig. This photo is courtesy of Dr. . Note that the Mach number is larger than the one predicted by the wedge.2.099◦. Even if the cone was a wedge. the cone half angle is 15◦ and the Mach number is 2. With given two angles the solution can be obtained by utilizing equation (14.0990 14.43◦ and the shock angle is 30.4522 71.4300 0.250 CHAPTER 14. M1 3.0143 30.

0 8. Example 14.14. 14. 14. the entropy is to be minimized through a series of oblique shocks rather than through a single normal shock.” Journal of Fluid Mechanics 366.4.0 6. a In fact. SOLUTION OF MACH ANGLE 251 k=14 Oblique Shock 3 2.0[bar]. 179–210. In such a case. with a proper design. Compare the different conditions in the two .0 3. and T = 17◦ C.16: Maximum values of the properties in an oblique shock 14.17: Two variations of inlet suction for flow just below M = 1.0 9.5 2 1. the following example is provided. It is suggested that a series of weak shocks “•”—–•˜ should replace one normal shock to increase the efficiency (see Figure (14.12 Application of Oblique Shock Mx 2 3 4 5 6 7 8 9 10 One of the practical applications of the ‘w’ oblique shock is the design of an in{4| } ~¡ €O ‚4|ƒ „ … †ˆ‡ ‰ Š‹4Œh Žˆ…  let suction for a supersonic flow.0 5.17))a .0 Mx Thu Jun 30 15:14:53 2005 My θ δ Fig. (1998) pp.4: The Section described in Figure 14.0 10.0 4. supersonic flow there is less entropy production (less pressure loss). P = 1. For details see Henderson and Menikoff “Triple Shock Entropy Theorem.5 0 90 80 70 60 50 40 30 20 10 0 2. To illustrate the design significance of the oblique shock. Clearly.18 air is flowing into a suction section at M = 2.4.0.0 7. there is general proof that regardless to the equation of state (any kind of gas). the flow can be brought to a subsonic Fig.5 1 0.

0000 0. °²± ™/š›DœužyŸ• ¡š¡¢•£ 1 neglect the detached distance 7◦ ¤¥/¦ § ¨e©/ª¬«C­/¤¡®•¯« 2 ³µ´ 3 4 Normal shock 7◦ Fig.99445 In the new region. .0000 My w θw δ 7.5000 0. OBLIQUE-SHOCK different configurations.2346 76.96524 And the additional information is 23 The results in this example are obtained using the graphical interface of POTTO–GDC thus.72087 In the oblique shock.6875 2.0000 My s My w θs θw δ 7. 14.0000 Py Px Ty Tx P0 y P0 x 1.7498 My s My w θs θw δ P0 y P0 x 0.0000 P0 y P0 x 0.7498 85.5549 14.2098 0.9831 51.57735 Ty Tx ρy ρx Py Px P0 y P0 x 1.99445 and the additional information by the minimal info in the Potto-GDC is Mx 2.2485 1.4) S OLUTION The first configuration is of a normal shock for which the results23 are Mx 2.0000 My 0.7498 resulting in Mx 1. Assume that only a weak shock occurs.1931 0.71761 1. the new angle is 7◦ + 7◦ with new upstream Mach number of Mx = 1. In the past the input file was given but the graphical interface it is no longer needed.2098 1.18: Schematic for Example (14.58974 1.7498 36.6667 4.7021 36. the first angle shown is Mx 2. no input explanation is given.252 CHAPTER 14.

99445 = 0.5: A∗ My w A supersonic flow is approaching a very long two– 10 M dimensional bland wedge body and creates a detached shock at Mach 3.4458.19: Schematic for Example (14.4.96524 × 0. .2346 My 0.90506 1.2626 1. Example 14.2626 × 1. the loss in the normal shock is above than 31% of the total pressure. The half wedge angle is 10◦ .96089 0.19).6116 0.1250 0.2574 Thus the area ratio has to be 1.4 yields M 0.21295 Now utilizing the isentropic relationship for k = 1.9496 P0 1 P0 3 P0 2 P0 1 The static pressure ratio for the second case is P4 P4 P3 P2 = = 1.3151 4.86966 1.1853 0.82141 Ty Tx ρy ρx Py Px P0y P0 x 1.2609 14.98903 × 0.1497 1.4018 1. SOLUTION OF MACH ANGLE Mx 1. the results are: Mx 1. In fact.14.45115 Ty Tx ρy ρx Py Px P0y P0 x 3.5) to the supersonic region assuming the flow is one– dimensional? ◦ ys S OLUTION The detached shock is a normal shock and the results are Mx 3.5088 41.4458 0.8770 1.285 = 2. In such a case. 14.7498 My w θw δ 7. What is the requited “throat” area ratio to achieve acceleration from the subsonic region Fig.6116 × 1.6147 P1 P3 P2 P1 The loss in this case is much less than in a direct normal shock.0000 Py Px Ty Tx P0 y P0 x 253 1.99549 An oblique shock is not possible and normal shock occurs.5000 My 0. Note that the pressure after the weak shock is irrelevant to the area ratio between the normal shock and the “throat” according to the standard nozzle analysis.45115 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.5 (see Figure 14.98903 With two weak shock waves and a normal shock the total pressure loss is P04 P03 P02 P04 = = 0.

8807 23.7008 87. Perform the calculation for M1 = 3.8482 88.1978 3. The conditions for this shock are: Mx 3. Find the required angle of shock BE.8912 22.9356 0.04263 0.20).20).13632 0.40596 0.0000 My s My w θs θw δ 6.254 Example 14.7008 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.0.8482 My s My w θs θw δ 3.99105 The transition for shock AB is Mx 3.99879 For the shock BC the results are Mx 2. Then. 14. shock E or expension Adopt this description and wave M 2 assume that the turn of 1 0 ◦ 6 is made of two equal C A angles of 3◦ (see Figure 14. The total pressure ratio at zone 2 is P02 P02 P01 = = 0.7080 0. OBLIQUE-SHOCK D 4 Slip Plane The effects of a double P3 = P 4 B wedge are explained in the 3 government web site as weak weak oblique oblique shock shown in Figure (14. 2.47641 2.9476 21. explain why this description has internal conflict.6: CHAPTER 14.0000 P0 y P0 x 0.04290 0. Assume that there are no boundary layers and Fig.10548 3.99894 × 0.7008 are M 2.1854 0.5990 0.7049 88.7049 2.997731283 P00 P01 P00 .99894 And the isentropic relationships for M = 2.40669 0.48013 2.0000 P0 y P0 x 0.10500 0.99879 = 0.48610 2.7049.20: Schematic of two angles turn with two weak shocks all the shocks are weak and straight.0000 My s My w θs θw δ 3.0000 P0 y P0 x 0.13665 The combined shocks AB and BC provide the base of calculating the total pressure ratio at zone 3. 1 S OLUTION The shock BD is an oblique shock with a response to a total turn of 6◦ .

0005 1.0019 1. SOLUTION OF MACH ANGLE On the other hand.04290 × 0. check whether the pressure at 2 is above or below or above the pressure (ratio) in zone 4.042516045 P0 1 P0 4 P0 1 255 The static pressure at zone 4 and zone 3 have to match according to the government suggestion hence.99105 = 0. the pressure at 4 has to be P4 P4 P0 4 = = 0. Only when the flow is far away from the double wedge.4).042436789 = 1.99920 Ty Tx ρy ρx Py Px P0y P0 x 1.0008 My 0.7037 0.4. P0 P2 P2 = 2 = 0. the flow behaves as only one theoretical angle of 6◦ exist. In reality the flow of double wedge will produce a curved shock surface with several zones.0008/2.49525 2.042436789 P0 2 P0 0 P0 2 P2 P4 Since P < P a weak shock must occur to increase the static pressure (see 02 01 Figure 5.14.026233 1. The increase has to be P3 /P2 = 0.04263 = 0. Assume k = 1.997731283 × 0.715320879◦ The deflection angle for such shock angle with Mach number is Mx 2.7◦ . The flow in zone 3 will flow into the wall in about 2. θ can be calculated from θ = sin−1 1.7049 My s My w θs 0. Example 14. the angle for BE shock which cause this pressure ratio needs to be found.042516045/0.0 θw 21.0013 1.7: Calculate the flow deflection angle and other parameters downstream when the Mach angle is 34◦ and P1 = 3[bar].001867743 To achieve this kind of pressure ratio the perpendicular component has to be Mx 1. and U1 = 1000m/sec. To do that.72 δ P0 y P0 x 0.4 and R = 287J/KgK .00000 From the last calculation it is clear that the government proposed schematic of the double wedge is in conflict with the boundary condition. T1 = 27◦ C.00000 The shock angle.7049 = 21.

88 = 1.0000 P0 y P0 x 0.128 1.8124 δ 11.0) = 1. M1n = M1 ∗ sin θ = 2.1) or GDC the following can be obtained. has to be determined M1 = √ U1 1000 = 2. OBLIQUE-SHOCK S OLUTION The Mach angle of 34◦ is below maximum deflection which means that it is a weak shock.8: For Mach number 2.89145 The temperature ratio combined upstream temperature yield T2 = 1.0485 2.00 δ 15.48269 2. Yet.0 θw 34.5 and wedge with a total angle of 22◦ . the Upstream Mach number.78 P0 y P0 x 0.61 From the Table (5.6[m/sec] Example 14. Utilizing GDC for Mach number 2.8800 My s My w θs 0.5K and the same for the pressure P2 = 2.3949 × 300 ∼ 418.0443 0.5 = 872.57[bar] And the velocity √ √ Un2 = My w kRT = 2.5000 My s My w θs 85.5 and the angle of 11◦ results in Mx 2.15) and (14.3949 2.1280 0.8575 × 3 = 8. calculate the ratio of the stagnation pressure.53431 2.6100 My 0.4 × 287 × 300 kRT Using this Mach number and the Mach deflection in either using the Table or the figure or POTTO-GDC results in Mx 2.88 sin(34. M1 .96873 . Mx 1.66545 Ty Tx ρy ρx Py Px P0 y P0 x 1.8575 0.0995 θw 32.256 CHAPTER 14.13) or simply converting the M1 to perpendicular component.4 × 287 × 418.89127 The relationship for the temperature and pressure can be obtained by using equation (14.

what is the difference? S OLUTION It has to be recognized that without any other boundary condition. the following can be obtained: . Compute the pressure ratio and the temperature ratio after the second shock (sometimes referred to as the reflective shock while the first shock is called the incidental shock).5 Mach number. then the Mach angle was calculated using the geometrical relationship of θ = sin−1 M1n /M1 . For a weak shock the maximum pressure ratio is obtained at the deflection point because it is closest to a normal shock. S OLUTION This kind of problem is essentially two wedges placed in a certain geometry. utilizing equation (14.6854 0. the shock is weak shock.9: What is the maximum pressure ratio that can be obtained on wedge when the gas is flowing in 2.14. Maximum Deflection Mach’s equation was used to calculate the normal component of the upstream.5 Mach without any close boundaries? Would it make any difference if the wedge was flowing into the air? If so.7974 Py Px Ty Tx P0 y P0 x 0. 3 stream line 2 1 θ M1 = 4 δ Assume that the upstream Mach number is 4 and the deflection angle is δ = 15◦ .60027 In these calculations. Example 14.4.10: Consider the schematic shown in the following figure. is obtained.7822 δ 29. It is clear that the flow must be parallel to the wall. δ . For the first shock.5000 My max θmax 64. SOLUTION OF MACH ANGLE 257 Example 14. To obtain the maximum point for 2. the upstream Mach number is known together with deflection angle. With these two quantities.12) the deflection angle. Utilizing the table or the Potto–GDC.94021 4.3573 2. either use the Maximum Deflection Mach number’s equation or the Potto–GDC Mx 2.

Again calculate the downstream ratios after the second shock and the deflection angle. OBLIQUE-SHOCK θw δ P0 y P0 x 0.6695 1.9290 My w θw δ Py Px Ty Tx P0 y P0 x 2. P1 P1 P2 = = 1.7822 15.0000 0.3575 0.2028 32.0000 1.48469 2.80382 With a Mach number of M = 2. Utilizing the POTTO–GDC which provides a solution in just a few clicks.0629 15.0000 0.2028 84.90041 and the additional information is Mx 2.51367 2.2808 32.9290 27.7344 × 1.9290 My s My w θs θw δ P0 y P0 x 0. the second deflection angle is also 15◦ .11: A similar example as before but here Mach angle is 29◦ and Mach number is 2.632 T3 T2 T3 Example 14.7985 1.6695 = 3.929.0026 P3 P2 P3 T1 T1 T2 = = 1.5851 27.00 δ 10.0000 1.5764 = 2.0629 15.0000 My w θw δ Py Px Ty Tx P0 y P0 x 2. Note that hand calculations requires endless time looking up graphical representation of the solution. S OLUTION Here the Mach number and the Mach angle are given.96263 .7344 0.5764 0.51 P0 y P0 x 0.46152 2.7985 × 1.90041 With the combined tables the ratios can be easily calculated. With these pieces of information by utilizing the Potto-GDC the following is obtained: Mx 2.0000 My s My w θs CHAPTER 14.9290 85.8500 My s My w θs 0.7822 15.80382 And the additional information by using minimal information ratio button in Potto– GDC is Mx 4.85. With these values the following can be obtained: Mx 2.258 Mx 4.0 θw 29.

0590 10.0590 10.5100 0.12: Compare a direct normal shock to oblique shock with a normal shock.51 the so called reflective shock gives the following information Mx 2.0000 0.0000 10.3582 × 1. SOLUTION OF MACH ANGLE 259 and the additional information by utilizing the minimal info button in GDC provides Mx 2.3575 29.3582 0.41523 3.3268 0.0000 My s My w θs 0.41523 Ty Tx ρy ρx Py Px P0y P0 x 5. What is the deflection angle in this case? S OLUTION For the normal shock the results are Mx 5.54894 1.0000 My 0.3268 ∼ 1. Where will the total pressure loss (entropy) be larger? Assume that upstream Mach number is 5 and the first oblique shock has Mach angle of 30◦ .0058 0.06172 While the results for the oblique shock are Mx 5.0000 29.4.3984 1.9419 84.3575 My w θw δ Py Px Ty Tx P0 y P0 x 1.49901 And the additional information is .5131 1.0 θw 30.97 P3 P2 P3 T1 T1 T2 = = 1.8000 5.4089 × 1.5100 1.9419 34.3984 ∼ 1.00 δ 20.14.97569 P1 P1 P2 = = 1.8021 T3 T2 T3 Example 14.96263 With the deflection angle of δ = 10.3575 My s My w θs θw δ P0 y P0 x 0.8500 My w θw δ Py Px Ty Tx P0 y P0 x 2.17 P0 y P0 x 0.97569 and the additional information of Mx 2.4089 1.9398 34.

1736 2.5141 0.6375 × 10.260 Mx 5. C stream line δ2 θ2 1 D 4 slip plane B 0 2 stream line A 3 φ F θ1 δ1 Illustration for example (14.0000 20.13)).0000 My w θw δ CHAPTER 14.0058 30. S OLUTION The first two zones immediately after are computed using the same techniques that were developed and discussed earlier.13: A flow in a tunnel ends up with two deflection angles from both sides (see the following Figure (14. Based on these calculations.6858 3. OBLIQUE-SHOCK Py Px Ty Tx P0 y P0 x 3.49901 The normal shock that follows this oblique is Mx 3. P1 P1 P2 = = 2.6375 2. calculate the pressure at zones 3 and 4 based on the assumption that the slip plane is half of the difference between the two deflection angles.36 P3 P2 P3 T1 T2 T1 = = 2.374 ∼ 27.13) For upstream Mach number of 5 and deflection angle of 12◦ and 15◦ .47485 Ty Tx ρy ρx Py Px P0 y P0 x 2. .8625 10.6858 ∼ 6.3740 0.0058 My 0. Example 14. explain whether the slip angle is larger or smaller than the difference of the deflection angle.75 T3 T2 T3 Note the static pressure raised is less than the combination shocks as compared to the normal shock but the total pressure has the opposite result. For the first direction of 15◦ and Mach number =5.32671 The pressure ratios of the oblique shock with normal shock is the total shock in the second case.5141 × 2.

5040 My s My w θs θw δ P0 y P0 x 0. If the two sides were equal because of symmetry the slip angle is also zero. For the two different deflection angles.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3. the slip angle has two extreme cases.5040 My w θw δ Py Px Ty Tx P0 y P0 x 2.80600 The conditions in zone 4 and zone 3 have two things that are equal. it is assumed that the slip angle moves half of the angle to satisfy both of the deflection angles (first approximation).5000 0.88496 with the additional information Mx 3.80600 And the additional conditions are Mx 5.6986 27.47413 2.0000 0.3217 15.0000 My s My w θs θw δ P0 y P0 x 0.6625 0.4.8006 .6986 85. The first case is where match lower deflection angle and second is to match the higher deflection angle.9791 1.9238 0.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3.0000 0.5◦ which results in Mx 3. This non–continuous velocity profile can occur in our model because it is assumed that fluid is non–viscous.2845 12.0739 24.6247 1. Under this assumption the conditions in zone 3 are solved by looking at the deflection angle of 12◦ + 1.5656 0. that only one deflection angle exist.69317 And the additional conditions are Mx 5.3217 15. They are the pressure and the velocity direction.5040 24.88496 And in zone 4 the conditions are due to deflection angle of 13.5000 1.0000 My s My w θs θw δ P0 y P0 x 261 0.6963 1.0000 1.8006 21. In this case. for the analysis.2845 12. It is to say.43914 3.9122 21. SOLUTION OF MACH ANGLE Mx 5.43016 3.6668 13.69317 For the second direction of 12◦ and Mach number =5. Mx 5.6668 13. It has to be noticed that the velocity magnitudes in zone 3 and 4 do not have to be equal.8006 86.6819 27.14.5040 86.0000 1.5◦ and Mach 3.5◦ = 13.

Example 14. S OLUTION Waiting for the solution 14.4. 14.5 Summary As with normal shock.6963 1.86179 From these tables the pressure ratio at zone 3 and 4 can be calculated P3 P2 P0 P1 1 P3 1 = = 1.8006 My s My w θs CHAPTER 14.6038 0. will the oblique shock with first angle of 15◦ and then 12◦ or opposite order make a difference (M = 5)? If not what order will make a bigger entropy production or pressure loss? (No general proof is needed). the pressure at zone 3 has to be reduced.5◦ to a larger number. the . To reduce the pressure the angle of slip plane has to increase from 1. in oblique. the oblique shock with upstream Mach number.9035 26. There are researchers which suggest that the numerical work is the solution.9791 ∼ 1.86179 with the additional information Mx 3.5000 0. However. Thus.46259 2.8006 My w θw δ Py Px Ty Tx P0 y P0 x 2. Consider the same values as previous analysis.13 Optimization of Suction Section Design is presentation of the experimental works is useful here? or present the numerical works? Perhaps to present the simplified model. as oppose to the normal shock.5000 1.9316 26.14: The previous example gave rise to another question on the order of the deflection angles. M1 is always greater than 1.6038 To reduce the pressure ratio the deflection angle has to be reduced (remember that at weak weak shock almost no pressure change). OBLIQUE-SHOCK θw δ P0 y P0 x 0.9035 85. The optimum deflection angle is a function of the Mach number range in which the suction section is operated in.18192 P4 P2 P0 P1 P4 1.6247 × 1. Under heavy construction please ignore The question raised is what is the optimum design for inlet suction unit? There are several considerations that have to be taken into account besides supersonic flow which includes for example the material strength consideration and the operation factors.262 Mx 3.3226 13.3226 13.6577 1.

and three is strong shock.21: Typical examples of unstable and stable situations the situation will return to its original state then it is referred to as the stable situation. The similar analysis can be applied to strong shock.21). 14. the analysis for the oblique shock wave is carried out. The only difference is that here. there is a possibility of no two–dimensional solution resulting in a detached shock case. When D is equal to zero. a special case is created because the weak and strong solutions are equal (for large deflection angle).6. if the answer is negative. On the other hand. in that analysis it has to be remembered that when the flow becomes subsonic the equation changes from hyperbolic to an elliptic equation. the situation moves away from the expected solution. An example to this situation. There is also the situation where the ball is between the stable and unstable situations when the ball is on a plane field which is referred to as the neutrally stable. δ there could be three solutions: the first one is the “impossible” solution in the case where D is negative. APPENDIX: OBLIQUE SHOCK STABILITY ANALYSIS 263 downstream Mach number. the discussion is more complicated but similar analysis can be applied to the strong shock. This change complicates the explanation and is omitted in this section. and upstream Mach number.6 Appendix: Oblique Shock Stability Analysis The stability analysis is an analysis which answers the question of what happens if for some reason. Instinctively. the stable and unstable can be recognized. M2 could be larger or smaller then 1. there are more than one parameter that can be changed. deflection angle. for large deflection angle. is a ball shown in the Figure (14.14. If the answer turns out to be that Fig. Yet. M1 n is always smaller than 1. in the analysis the strong shock results in an elliptic solution (or region) as opposed to a hyperbolic in weak shock. Unstable Stable . When D > 0. then the situation is referred to as unstable. In this example only the weak solution is explained. In the same manner. two is weak shock. the shock angle. When D is positive there is no physical solution and only normal shock exist. As it results. Given M1 and the deflection angle. 14. Of course. The perpendicular component of the downstream Mach number. for example.

264 CHAPTER 14. The opposite happens when the deflection increment becomes negative. it must be pointed out that it doesn’t mean that the flow is unstable but rather that the model is incorrect. Thus.52). The pressure difference at the wall becomes a negative increment which tends to pull the shock angle to the opposite direction. On the other hand. .22: The schematic of stability analysis for oblique shock. ∆θ = k+1 ∆δ 2 (14. the weak shock is stable. has to be in the opposite direction of the change of the shock angle. Please note that this analysis doesn’t apply to the case of the close proximity of the δ = 0. Yet. the shock wave is unstable according to this analysis to one direction but stable to the other direction. the deflection angle becomes positive which increases the pressure at the wall. The change is determined from the solution provided before or from the approximation (14. the virtual change in the deflection angle Fig. 14.56) can be applied for either positive. In fact. There isn’t any known experimental evidence to show that flow is unstable for δ = 0. ∆θ + or negative ∆θ− values. OBLIQUE-SHOCK The change in the in∆θ + clination angle results ∆θ − in a different upstream Mach number and ∆δ − a different pressure. ∆δ + to maintain the same direction stream lines.56) Equation (14.

Fig. because of imperfections of the wall and the boundary layer. 15.1 Introduction positive angle As discussed in Chapter (14) when the deflection turns to the opposite direction of the flow. Here because of the tradition. the flow accelerates to match the boundary condition. when this model breaks down. is smooth.1)).2)) ¹ º¼»/½¿¾ µ = sin−1 1 M (15.CHAPTER 15 Prandtl-Meyer Function 15. flow separation occurs. Supersonic expansion and isentropic comÀ U pression (Prandtl-Meyer function). there is no limitation for the Prandtl-Meyer function to approach zero. Yet.1) Fig. it has to be assumed to be insignificant. Yet. As opposed to the oblique shock. The Mach line shows that a disturbance in a field of supersonic flow moves in an angle of µ. for very small angles. as opposed to the oblique shock. out any jump in properties. 15. the flow becomes complicated. The transition.1: The definition of the angle for the Prandtl–Meyer function. In a somewhat a similar concept to oblique shock there exists a “detachment” point above which this model breaks and another model has to be implemented. are an extenc · ¸ sion of the Mach line concept. which is defined as ¶ (as shown in Figure (15.2: The angles of the Mach line triangle 265 ma xi mu m an gl e . and no known simple model can describe the situation. with. the deflection angle is denoted as a positive when it is away from the flow (see Figure (15.

cos(dν ) ∼ 1 (15. and it is assumed here to be isotropic for a positive angle. is dy = (U + dU ) sin(dν ) = U dν The tan µ is the ratio of dy/dx (see Figure (15. as it turns out.2 Geometrical Explanation x The change in the flow direction is assume ËÊÌ ÍµÎCËÊÌ to be result of the change in the tangential dx = dU cos(90 − µ) Æ Ç ÈÊÉ component. 15. the Mach angle dν dy òĵÅCà Á increase and result in a change in the direction of the flow. The Mach line is the chief line in the analysis because of the wall contour shape information propagates along this line.2) A Mach line results because of a small disturbance in the wall contour. a discussion on the relationship between the flow properties and the flow direction is presented. Therefore. However.5) dx = (U + dU ) cos ν − U = dU (15. This direction change results in a change of the flow properties. 15. Later. PRANDTL-MEYER FUNCTION 1 M1 − 1 µ = tan−1 √ (15. The typical simplifications for geometrical functions are used: y Ma ch li ne These simplifications are the core reasons why the change occurs only in the perpendicular direction (dν << 1). This assumption. The change of the velocity in the flow direction. Hence. The velocity compoÏ ÐÒÑÔÓCСÕOÖ ×4Ø Ï ÓWÙwÕÊÚcÐ nent in the direction of the Mach line is assumed to be constant to satisfy the assumption that the change is a result of the contour Fig. is close to reality.3) .266 or CHAPTER 15. dx is In the same manner. the total Mach numÛ Ü4Ý¡ÞÔßÊà á  ber increases.4) dν ∼ sin(dν ). this assumption will be examflow ined. In this chapter. Once the contour is changed. no shock is created from many small positive angles. the velocity perpendicular to the flow. The reason that a “negative” angle is not applicable is that the coalescing of the small Mach wave which results in a shock wave.3)) tan µ = dx dU = dy U dν (15. the flow direction will change to fit the wall. dy .6) (15. This Mach line is assumed to be a result of the positive angle.3: The schematic of the turning only.

7) (15. It must be recognized that here the cylindrical coordinates are advantageous because the Fig.12) ∂Ur Uθ ∂Ur Uθ 2 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r ρ ∂r ρ ∂r (15.9) The constant can be chosen in a such a way that ν = 0 at M = 1.4: The schematic of the coordinate based on the matheflow turns around a sinmatical description gle point.11) (15. the mass conservation can be written as ∂ (ρrUr ) ∂ (ρUθ ) + =0 ∂r ∂θ The momentum equations are expressed as Ur and Ur ∂Uθ Uθ ∂Uθ Uθ Ur 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r rρ ∂θ rρ ∂θ (15.8) becomes ν (M ) = − k+1 tan−1 k−1 k−1 (M 2 − 1) + tan−1 k+1 (M 2 − 1) + constant 267 (15. GEOMETRICAL EXPLANATION The ratio dU/U was shown to be dM 2 dU = 1 2 U 2M 2 1 + k− 2 M Combining equations (15.7) transforms it into √ M 2 − 1dM 2 dν = − 1 2 2M 2 1 + k− 2 M After integration of equation (15. For this coordinate system.1 Alternative Approach to Governing Equations In the previous secback Mach tion.15.6) and (15. 15.2. a simplified verr Ur line sion was derived based Front Mach on geometrical arguUθ line θ ments. a more rigorous explanation is provided.2. 15.10) . In this section.8) (15.

17) Uθ or Uθ 2 Ur + ∂Uθ ∂θ = c 2 Ur + ∂Uθ ∂θ (15.13) can be rearranged as transformed into − 1 Uθ Ur + ∂Uθ ∂θ = 1 ∂ρ ρ ∂θ (15.14) The momentum equations now obtain the form of Uθ ∂Ur Uθ 2 − =0 r ∂θ r ∂Ur Uθ − Uθ = 0 ∂θ Uθ ∂Uθ Uθ Ur c2 ∂ρ − =− r ∂θ r rρ ∂θ ∂Uθ c2 ∂ρ Uθ − Ur = − ∂θ ρ ∂θ Substituting the term 1 ∂ρ ρ ∂θ (15. PRANDTL-MEYER FUNCTION If the assumption is that the flow isn’t a function of the radius.20) Ur + ∂Uθ ∂θ =0 (15.15) (15.19) It is remarkable that the tangential velocity at every turn is at the speed of sound! It must be pointed out that the total velocity isn’t at the speed of sound. r.14) into equation (15. then all the derivatives with respect to the radius will vanish. the derivative isn’t zero.13) Equation (15.19) it follows that Uθ = c (15.18) And an additional rearrangement results in c2 − U θ 2 From equation (15.16) from equation (15. but only . One has to remember that when r enters to the function.268 CHAPTER 15.16) results in ∂Uθ − Ur ∂θ = c2 Uθ Ur + ∂Uθ ∂θ (15. like the first term in the mass equation. the mass equation is reduced to ρUr + ∂ (ρUθ ) =0 ∂θ (15. Hence.

equation (15.27) Note that Ur must be positive. In fact. ∞ and since Uθ = 0 leads to ∂Ur = Uθ ∂θ (15.20) for the speed of sound and substituting equation (15.23) results in c2 Uθ 2 + U r 2 + = h0 k−1 2 (15.26) + Ur 2 = 2h0 (15. k . M = 1.23) Enthalpy in perfect gas with a constant specific heat.25) 1 Cp c2 R T = RT = R (k − 1) Cv k−1 (15. This analysis can be also applied to the same equation when they are normalized by Mach number. The solution of the differential equation (15.15. these results are associated with the characteristic line. equation (15.3) under Uy is equal to the speed of sound. the component shown in Figure (15.26) becomes k+1 k−1 ∂Ur ∂θ 2 k−1 + ∂Ur 2 ∂θ + Ur 2 2 = h0 (15. equation (15.21) If r isn’t approaching infinity. The energy equation for any point on a stream line is h(θ) + Uθ 2 + U r 2 = h0 2 (15.25) into ∂Ur 2 ∂θ After some rearrangement.15) becomes Uθ r ∂Ur − Uθ ∂θ =0 (15. based on the definition of the Mach angle.27) incorporating the constant becomes Ur = 2h0 sin θ k−1 k+1 (15. the non–dimensionalization can be applied at this stage as well.22) In the literature.24). GEOMETRICAL EXPLANATION 269 the tangential component.2. into equation (15.28) .22) which is the radial velocity transforms equation (15. is c(θ )2 k h(θ) = Cp T = Cp and substituting this equality.24) Utilizing equation (15. However. After some additional rearrangement.

31) What happens when the upstream Mach number is not 1? That is when the initial condition for the turning angle doesn’t start with M = 1 but is already at a different angle. the assumption . 15.30) Now utilizing the expression that was obtained for Ur and Uθ equations (15.28) is chosen such that Ur (θ = 0) = 0. so ν (M ) = θ(M ) − θ(Mstarting ) = k+1 tan−1 k−1 k−1 k+1 M2 − 1 − tan−1 M2 − 1 (15. The tangential velocity obtains the form Uθ = c = ∂Ur = ∂θ k−1 k+1 2 h0 cos θ k−1 k+1 (15.34) (15. PRANDTL-MEYER FUNCTION which satisfies equation (15.2.28) results for the Mach number is M2 = 1 + or the reverse function for θ is θ= k+1 tan−1 k−1 k−1 M2 − 1 k+1 (15.2)) tan ν = Mstarting 2 − 1 (15.29) The Mach number in the turning area is M2 = Uθ 2 + U r 2 Uθ 2 + U r 2 =1+ = c2 Uθ 2 Ur Uθ 2 (15. For this upstream Mach number (see Figure (15.33) The deflection angle ν . The upstream Mach number is denoted in this segment as Mstarting .270 CHAPTER 15. has to match to the definition of the angle that is chosen here (θ = 0 when M = 1).29) and (15.2 Comparison And Limitations between the Two Approaches The two models produce exactly the same results. The arbitrary constant in equation (15.32) k+1 tan2 k−1 θ k−1 k+1 (15.35) These relationships are plotted in Figure (15.6). In the geometrical model.27) because sin2 θ +cos2 θ = 1. but the assumptions for the construction of these models are different.

mass balance is maintained by the reduction of the density.5) 15. 15.3. the fact remains that there is a radial velocity at Ur (r = 0) = constant. the maximum turning angle is much larger than the maximum deflection point because the process is isentropic.15.37) The change in the deflection angle is calculated by . Prandtl–Meyer function becomes ν∞ = π 2 k+1 −1 k−1 (15. At this point (r = 0) these models fail to satisfy the boundary conditions and something else happens there. Normally. THE MAXIMUM TURNING ANGLE 271 is that the velocity change in the radial direction is zero. The radial velocity is “fed” through the reduction of the density. Thus. Aside from its close proximity to turning point. The statement for the construction of the geometrical model can be improved by assuming that the frame of reference is moving radially in a constant velocity. analysis of design commonly used in the industry and even questions posted to students show that many assume that the turning point can be sharp. it was assumed that radial velocity is only a function of θ. On top of the complication of the turning point.36) The maximum of the deflection point and the maximum turning point are only a function of the specific heat ratios. However. how did the gas accelerate to above the speed of sound when there is no nozzle (where is the nozzle?)? These questions +are of interest in engineering but are beyond the scope of this book (at least at this stage). However. Regardless of the assumptions that were used in the construction of these models. the sharp point should be replaced by a smoother transition. In the rigorous model. the author recommends that this function be used everywhere beyond 2-4 the thickness of the boundary layer based on the upstream length. (1 + ) the radial velocity is small . At a small Mach number. What happens when the deflection angel exceeds the maximum angle? The flow in this case behaves as if there is almost a maximum angle and in that region beyond the flow will became vortex street see Figure (15. For example. In fact.4 The Working Equations for the Prandtl-Meyer Function ν2 − ν1 = ν (M2 ) − ν (M1 ) (15. the question of boundary layer arises.3 The Maximum Turning Angle The maximum turning angle is obtained when the starting Mach number is 1 and the end Mach number approaches infinity. an increase in the Mach number can result in a very significant radial velocity. in many instances. In this case. some researchers recommend that.

38) It can be observed that only the area that “seems” to be by the flow was used in expressing equation (15. Again. 15.272 CHAPTER 15. The relation between P2 and P4 is such that the flow depends on the upstream Mach number.7). PRANDTL-MEYER FUNCTION sl ip li ne Maximum turning Fig. k .5 d’Alembert’s Paradox In ideal inviscid incompressible flows. . it is Fig. The net force in flow direction. the move3 ment of body does not 1 2 encounter any resistance. and the specific heat. and θ1 w θ2 this paradox is examined θ2 θ1 here. 4 This result is known as d’Alembert’s Paradox. M1 . 4 Supposed that a two– dimensional diamond–shape 2 1 body is stationed in a su3 personic flow as shown in Figure (15.7: A simplified diamond shape to illustrate the suassumed that the fluid is inpersonic d’Alembert’s Paradox viscid.38). the drag. 15. is D=2 w (P2 − P4 ) = w(P2 − P4 ) 2 (15.5: Expansion of Prandtl-Meyer function when it exceeds the maximum angle 15.

15.6. FLAT BODY WITH AN ANGLE OF ATTACK

273

Prandtl-Meyer Angle
100

80

k=1.4

60 θ 40 20 0

1

2

3

4

5 6 Mach Number

7

8

9

10

Fri Jul 8 15:39:06 2005

Fig. 15.6: The angle as a function of the Mach number

Regardless in the equation of the state of the gas, the pressure at zone 2, P2 , is larger than the pressure at zone 4, P4 . Thus, there is always drag when the flow is supersonic which depends on the upstream Mach number, M1 , specific heat, k , and the “visible” area of the object. This drag is known in the literature as (shock) wave drag.

15.6

Flat Body with an Angle of Attack

Previously, the thickness of a body was 1 2 shown to have a drag. Now, a body with 3 zero thickness but with an angle of attack will be examined. As opposed to 5 the thickness of the body, in addition 4 α to the drag, the body also obtains lift. 7 Again, the slip condition is such that the 6 pressure in region 5 and 7 are the same, and additionally the direction of the velocity must be the same. As before, the Fig. 15.8: The definition of the angle for the magnitude of the velocity will be differPrandtl–Meyer function. ent between the two regions.
w

Slip

plane

274

CHAPTER 15. PRANDTL-MEYER FUNCTION

15.7

Examples For Prandtl–Meyer Function

Example 15.1: A wall is included with 20.0◦ an inclination. A flow of air with a temperature of 20◦ C and a speed of U = 450m/sec flows (see Figure 15.9). Calculate the pressure reduction ratio, and the Mach number after the bending point. If the air flows in an imaginary two–dimensional tunnel with width of 0.1[m] what will the width of this imaginary tunnel after the bend? Calculate the “fan” angle. Assume the specific heat ratio is k = 1.4.

âäãæårçéè/êìë#í¡îyï ðòñôóéõyöh÷

¨
©

   ¥§¦   


øúùüûôýyþéÿ

 ¢¡¤£

Fig. 15.9: The schematic of Example 15.1

The Mach number is then

S OLUTION First, the initial Mach number has to be calculated (the initial speed of sound). √ √ a = kRT = 1.4 ∗ 287 ∗ 293 = 343.1m/sec M= 450 = 1.31 343.1
T T0 ρ ρ0

this Mach number is associated with M 1.3100 ν 6.4449
P P0

µ 52.6434

0.35603

0.74448

0.47822

The “new” angle should be ν2 = 6.4449 + 20 = 26.4449◦ and results in M 2.0024 ν 26.4449
P P0 T T0 ρ ρ0

µ 63.4620

0.12734

0.55497

0.22944

15.7. EXAMPLES FOR PRANDTL–MEYER FUNCTION Note that P01 = P02 P2 P0 P2 0.12734 = 1 = = 0.35766 P1 P1 P02 0.35603 The “new” width can be calculated from the mass conservation equation. ρ1 x1 M1 c1 = ρ2 x2 M2 c2 =⇒ x2 = x1 x2 = 0.1 × ρ 1 M1 ρ 2 M2 T1 T2

275

1.31 0.74448 0.47822 × = 0.1579[m] 0.22944 2.0024 0.55497 Note that the compression “fan” stream lines are note and their function can be obtain either by numerical method of going over small angle increments. The other alternative is using the exact solution1 . The expansion “fan” angle changes in the Mach angle between the two sides of the bend fan angle = 63.4 + 20.0 − 52.6 = 30.8◦ Reverse the example, and this time the pressure on both sides are given and the angle has to be obtained2 . Example 15.2: Gas with k = 1.67 flows over bend (see Figure 15.2). Compute the Mach number

¢!#"$ % :(;*<>= ?A@ BDC0E>F

6§7 8
9

&('*)+ ,.-0/21 3¢4¤5

Fig. 15.10: The schematic for the reversed question of example (15.2)

after the bend, and the bend angle.
1 It isn’t really different from this explanation but shown in a more mathematical form, due to Landau and friends. It will be presented in the future version. It isn’t present now because of the low priority to this issue. 2 This example is for academic understanding. There is very little with practicality in this kind of problem.

276

CHAPTER 15. PRANDTL-MEYER FUNCTION

S OLUTION The Mach number is determined by satisfying the condition that the pressure downstream are and Mach given. The relative pressure downstream can be calculated by the relationship P2 P1 1 P2 = = × 0.31424 = 0.2619 P0 2 P1 P0 1 1.2 M 1.4000 ν 7.7720
P P0 T T0 ρ ρ0

µ 54.4623

0.28418

0.60365

0.47077

¯ = 0.2619 require either locking in the table or using the With this pressure ratio P enclosed program. M 1.4576 ν 9.1719
P P0 T T0 ρ ρ0

µ 55.5479

0.26190

0.58419

0.44831

For the rest of the calculation the initial condition is used. The Mach number after the bend is M = 1.4576. It should be noted that specific heat isn’t k = 1.4 but k = 1.67. The bend angle is ∆ν = 9.1719 − 7.7720 ∼ 1.4◦ ∆µ = 55.5479 − 54.4623 = 1.0◦

15.8

Combination of the Oblique Shock and Isentropic Expansion

Example 15.3: Consider two–dimensional flat thin plate at an angle of attack of 4◦ and a Mach number of 3.3. Assume that the specific heat ratio at stage is k = 1.3, calculate the drag coefficient and lift coefficient. S OLUTION For M = 3.3, the following table can be obtained: M 3.3000 ν 62.3113
P P0 T T0 ρ ρ0

µ 73.1416

0.01506

0.37972

0.03965

With the angle of attack the region 3 will be at ν ∼ 62.31 + 4 for which the following table can be obtained (Potto-GDC)

15.8. COMBINATION OF THE OBLIQUE SHOCK AND ISENTROPIC EXPANSION277 M 3.4996 ν 66.3100
P P0 T T0 ρ ρ0

µ 74.0528

0.01090

0.35248

0.03093

On the other side, the oblique shock (assuming weak shock) results in Mx 3.3000 My s My w θs θw δ 4.0000
P0 y P0 x

0.43534 3.1115 88.9313 20.3467

0.99676

and the additional information, by clicking on the minimal button, provides Mx 3.3000 My w θw δ 4.0000
Py Px Ty Tx P0 y P0 x

3.1115 20.3467

1.1157

1.1066

0.99676

The pressure ratio at point 3 is P3 P3 P03 P01 1 = = 0.0109 × 1 × ∼ 0.7238 P1 P03 P01 P1 0.01506 The pressure ratio at point 4 is P3 = 1.1157 P1 dL = 2 2 (P4 −P3 ) cos α = kP1 M1 2 kM1 2 dd = 2 kM1 2 P3 P4 − P1 P1 sin α = P3 P4 − P1 P1 cos α = 2 (1.1157 − 0.7238) cos 4◦ ∼ .054 1.33.32

2 (1.1157 − 0.7238) sin 4◦ ∼ .0039 1.33.32

This shows that on the expense of a small drag, a large lift can be obtained. Discussion on the optimum design is left for the next versions.

278 CHAPTER 15. PRANDTL-MEYER FUNCTION .

method of characteristics 279 . 2D.CHAPTER 16 Topics in Steady state Two Dimensional flow shock–expansion theory. linearized potential flow: thin airfoil theory.

280 CHAPTER 16. TOPICS IN STEADY STATE TWO DIMENSIONAL FLOW .

This program is complied under gnu g++ in /Gnu/Linux system. Some parts of the code are in FORTRAN (old code especially for chapters 11 and 12 and not included here.2 Usage To use the program some information has to be provided. the kind of the information needed. machV The Mach number and it is used in stagnation class fldV The 4f L D and it is used in Fanno class isothermal class p2p1V The pressure ratio of the two sides of the tubes M1V Entrance Mach M1 to the tube Fanno and isothermal classes 1 when will be written in C++ will be add to this program. isothermal. At this stage only the source code of the program is available no binary available.1 . In fact even PLEASE do not even try to use this program under any Microsoft window system. This program was used to generate all the data in this book. A. shock and others. and in many case where it is a range of parameter(s). The necessary input A parameter(s). The derived class are Fanno. where it has to be in a L TEX format or not. 281 .1 About the Program The program is written in a C++ language. NO Support whatsoever will be provided for any Microsoft system. The program has the base class of basic fluid mechanics and utilities functions to calculate certain properties given data.APPENDIX A Computer Program A. As much support as possible will be provided if it is in Linux systems.

282 CompressibleFlow basic functions virtual functions Interpolation (root finding) LaTeX functions Representation functions pipe flow stagnation common functions discontinuity common functions real fluids common functions only contain P-M flow specific functions Fanno the actual functions the actual functions Isothermal Rayleigh the actual functions normal shock specific functions oblique shock specific functions Fig.1: Schematic diagram that explains the structure of the program APPENDIX A. COMPUTER PROGRAM . A.

makeTable(whatInfo...4}|} \caption{ ?? \label{?:tab:?}}\\ .{1..{1. isRange.A.4}|D. int isRange = no.{1.2. M1fldP2P1V three part info MxV Mx or My M1 and 4f L D are given infoStagnation print standard (stagnation) info infoStandard standard info for (Fanno.4}|D.. You can just can cut and paste it in your latex file. variableName = MxV.4}|D. USAGE 283 M1ShockV Entrance Mach M1 when expected shock to the tube Fanno and isothermal classes FLDShockV FLD with shock in the in Fanno class M1fldV both M1 and 4f L D are given P1 P2 .. ******************************************* \setlongtables \begin{longtable} {|D.{1.{1.. ******************************************* The following stuff is the same as above/below if you use showResults with showHeads but the information is setup for the latex text processing. s.4}|D. etc) including infoShock print shock sides info infoTubeShock print tube info shock main info infoTubeProfile the Mach number and pressure ratio profiles infoTubeShockLimits print tube limits with shock A To get the shock results in L TEX of Mx The following lines have to be inserted in the end of the main function. shock etc) infoTube print tube side info for (Fanno.{1. whatInfo = infoStandard . You must use longtable style file and dcolumn style files.4}|D.. Mx = 2.4}|D. variableValue). int isTex = yes.{1. variableName.0 . isTex.

5854& 3.3 in}\mathbf{M} $} \multicolumn{1}{|c|} {$\mathbf{4fL \over D} $} & \multicolumn{1}{|c|} {$\mathbf{P \over P^{*}} $} & \multicolumn{1}{|c|} {$\mathbf{P_0 \over {P_0}^{*}} $} & \multicolumn{1}{|c|} {$\mathbf{\rho \over \rho^{*}} $} & \multicolumn{1}{|c|} {$\mathbf{U \over {U}^{*}} $} & \multicolumn{1}{|c|} {$\mathbf{T \over T^{*}} $} \\\hline \endhead 2.3608& \hline\end{longtable} & 1.pt}{0.pt}{0.1in]{0.284 \hline \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \\\hline APPENDIX A.152& 0.773& 0.1in]{0.6164 \\ A.176& 2. COMPUTER PROGRAM {$\rule[-0.org.000& 0.3 Program listings Can be download from www.potto.3 in}\mathbf{M} $} {$\mathbf{4fL \over D} $} & {$\mathbf{P \over P^{*}} $} & {$\mathbf{P_0 \over {P_0}^{*}} $} & {$\mathbf{\rho \over \rho^{*}} $} & {$\mathbf{U \over {U}^{*}} $} & {$\mathbf{T \over T^{*}} $} & \endfirsthead \caption{ ?? (continue)} \\\hline \multicolumn{1}{|c|} {$\rule[-0. .

function. 2. 10 Emanuel’s partial solution to oblique shock. 9 piston velocity. Garl Gustaf Patrik diffuser efficiency. 4D choking. 93 M F Fanning Friction factor. see de Laval. 135 entrance issues. 12 fL 157 fanno flow. 234 Moody diagram. 133 airbag. 8 External flow. 7 isohtermal flow entrace length limitation. 3 maximum deflection angle. 215 clasifications of chambers. 28 chamber controled volume. 135 Mach. 155 Fanno flow. 137 de Laval. 137 fanno second law. 159 entrace Mach number calculations. 13 moving shock. 200 Bar-Meir’s solution to Oblique shock. 95 stagnation temperature. 226 deflection angle range. 228 nozzle efficiency. 3 N NACA 1135. 158 average friction factor. 9 friction factor. 240 deLavel’s nozzle. 140 fL 139 maximum . 5 intersection of Fanno and Rayleigh. 3. 158 shockless. see discontinuity internal energy. 164 star condition. 9 defection angle. Carl Gustaf Patrik. 13 Gibbs. 226 normal components. 8. 1 Eckert number. 227 negative deflection angle. 140 Isothermal Flow. 127 discontinuity. 127 O oblique shock . 233 long pipe flow. 8 Bernoulli’s equation. 31 gravity. 4D table. 13 G C H I Hydraulic Jump.SUBJECTS INDEX 285 Subjects index A B Fliegner experiment. see Shapiro flow isothermal flow. 153. 200 Darcy friction factor. 98 solution for closed valve. 160 Fliegner. 166 entropy. 145 D E L large defelection angle.

196 second law. 201 Shapiro Flow. 35 real gas. 86 small deflection angles. 86 star velocity. 98 Rayleigh Flow. 29 linear temperature. 87 table basic. 36 steam table. 109 shock wave. COMPUTER PROGRAM sub. 101 shock wave partial close valve. 28 strong solution. 12 rayleigh flow. 242 speed of sound. 8 APPENDIX A. 242 V von Neumann paradox. 31 solid. what. see isothermal nozzle science disputes. 225 oblique shock governing equations. 81 perturbation. 225 W Y Z weak solution. 36 zero diflection angle. 232 Young’s Modulus. 107 Taylor–Maccoll flow. 91 trivail solution. 128 T P piston velocity. 3 Shapiro flow. 238 S . 4 ideal gas. 115 thickness. 90 solution. 231 normal shock. 30 two phase. 190 tables. 37 speed of sound. 114 supersonic tunnel. 232 R table shock choking. 189 Romer. 229 Oblique shock stability. 31 liquid.286 condtions for solution. 187 entrance Mach number. 200 semirigid tank limits. 191 two maximums. 5 semi rigid chamber. 13 shock tube.

Lev.G. 5 Taylor. 14 . Theodor. 10 Fanno. Ernest. 248 Hugoniot. 7 T G H K L Galileo Galilei. 4. 14 Landau. 5 Converdill. 4 Henderson. 5 Rouse. E. 4 Mach. 10 Rankine. 230 V Van Karman. 7 Stokes. 4 Challis. Ludwig. 5 Menikoff. 5 Riemann. 6 Kutta-Joukowski. Robert. Pierre Henri.. 14 M N Newton. 6 Rayleigh. G. 4 O P Owczarek. 5. 4 Meyer. I. Gino Girolamo. 5 Shapiro. 4 W Wright brothers. Marin. 7 Moody. 7 Thompson. 4 Stodola.R.AUTHORS INDEX 287 Authors index B C R Boyle. 207 Poisson. 5 E F S Eckert. 248 Mersenne. 6 Prandtl. 7 Leonardo Da Vinci. John Macquorn.