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

. . . . . . . .10Issues Related to the Maximum Deflection Angle . . . . . . .7 Close and Far Views of the Oblique Shock . . . . . . . . . . . . 12. . . . . . . . . . . . . 14. . . . . . . . . . .2 Introduction to Prandtl–Meyer Function . . . . .4. . . . . . . . . . . . . . . . . . . . . 14. . . . . .4 Solution of Mach Angle . . .4. . . . . . . . . . . . . . . . . . . . . . .4. . .4. . . . . . . . . . . 14. . . . . . . . . . . . . . . . . .4. . .1 Introduction . . . . . . . . . . 14. 15. . . . . . 14. . . . . . . 14. . . .8 Combination of the Oblique Shock and Isentropic Expansion . . . . .3 Introduction to Zero Inclination . . . . and Deflection Angle. 15. . . . . . . . 14.2 Comparison And Limitations between the Two Approaches 15. . . . . . .6 Flat Body with an Angle of Attack . . . . .4 The Working Equations for the Prandtl-Meyer Function . . . . . . .2. . . . . . . . . . . . . δ 14. . .2 Geometrical Explanation . .2. . . . .2 Examples . . . . . . . 15. . 12. . . . . . . . 14. . .1 General Model . . . . . . . . . . . . . . . . 14. .3 The Maximum Turning Angle . . .3 Oblique Shock . . .4. . . . . .CONTENTS 12 Evacuating/Filing Chambers under External Volume Control 12. . . . . .1. . . . . . 15. . . . . . . . . .6 Appendix: Oblique Shock Stability Analysis . . . . . . 15. . . . . . . . . . . . . . . . . 14. . . . 12. . . . . . M1 . . . . . . . 15. .1 Upstream Mach Number. .1. . . . . 15. . . .11Oblique Shock Examples . . . .4. . . . .7 Examples For Prandtl–Meyer Function . . . . .2 Summary . . . . . . . . . . . . . . 13 Topics in Unsteady one Dimensional gas dynamics 14 Oblique-Shock 14. . . . . . . . . .2 When No Oblique Shock Exist or When D > 0 . . . . . . . . . . . . . . . . . . . .3 Upstream Mach Number. . . . . .2. . . . 12. . . . . . . 14. . . . . θ . .1 Preface to Oblique Shock . 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 Summary . . . . . . .4.13Optimization of Suction Section Design . . . . . . 14. . 14. . . . . .4. . δ and θ . . . . . . . . . . . . .3 Direct Connection .8 Maximum Value of Oblique shock . . . .4. . . . . . . . .9 Detached shock .1 Alternative Approach to Governing Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14. . . and Shock Angle. .5 Flow in a Semi–2D Shape . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . .1 Introduction to Oblique Shock . . . . . . . . . . . . . . . . . .4. . . . . . . . . . . . . . . .4. . .1 Rapid Process . . 14. . . .2. . . . 14. . . . . . . . 15. . . . . . M1 . . . . . . .1. . . . . . . .2 Introduction . . . . . . . . . . . .2. . .5 d’Alembert’s Paradox .4. . .6 Small δ “Weak Oblique shock” .12Application of Oblique Shock . . . . . .4 Given Two Angles. . . . . . . . . . . . . . . . . . . . . . . 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. . . . . 14. . . . . . . . .

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

. . . . . . . 3. . . 1. . 41 4. . . . . . . . . . . . . . . . .4 44 4. . . . . . . . . . . . . . . . . . 1. . . . . .LIST OF FIGURES 1. . . . . . . 1. . . . . . . . . . . . . . . . . . 1. . . . . 3. . . . . . . . . . . . 1. . . . . . . . . . . . . . . . . 43 4. . .6 Photo of Ernest Mach . . . . . . . . . . . . . . . .2 The schematic of deLavel’s turbine after Stodola. . . . .2 Stationary sound wave and gas moves relative to the pulse . . . . . . . .4 Control volume inside a converging-diverging nozzle. . . . . . . . . . . . . . . 1. . . . . . . . . . .1 The shock as connection of Fanno and Rayleigh lines after Stodola. 46 ix . . . . .4 Flow rate as a function of the back pressure taken from Stodola 1927 Steam and Gas Turbines . . . . . . . . . Steam and Gas Turbine . .1 Flow of a compressible substance (gas) through a converging–diverging nozzle. . . . . . . . . . . . . . . . . . . . . . . . . . . .3 The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines . . . . . . . . .11 Photo of Prandtl . . . . . . . . . . . . 1. . .8 Photo of Lord Rayleigh . . . . .7 The photo of thebullet in a supersonic flow not taken in a wind tunnel 1. . . . . . . . . . . .9 Portrait of Rankine . . . . 1. . . . . . k = 1. . . . . . . . . . . . .1 A very slow moving piston in a still gas . . . . . . . . . . . . . . . . . . . Steam and Gas Turbine . . . . . . . . . . 1. . 1. . . . .10 The photo of Gino Fanno approximately in 1950 . . . . . .5 Portrait of Galileo Galilei . . . . . . . . . . . . . . . . . .12 The photo of Ernst Rudolf George Eckert with the author’s family . . . . . . . . . . . .3 The Compressibility Chart . . . . . . . . . . . . . 7 9 11 12 16 17 17 18 19 20 21 22 28 28 32 4.2 Perfect gas flows through a tube . . . . . . . .3 The stagnation properties as a function of the Mach number. . . . 3. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . .10) . . . . . . . . . . 5. . . . . . . . . . . . . . .16 The shock tube schematic with a pressure ”diagram. . . . . . . . . . . .8) . . . . . . . . .7 The moving shock Mach numbers as a result of a sudden and complete stop. . . . . 5. . . .3 6. . . . . . . . . . . . . . . . . . . .3 The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. . . . 5. . . . . . . . . . . . . . . .” . . 5. . . . . . . . . . . . . . Schematic of a supersonic tunnel in a continuous region (and also for example (6. . . . . . . . . . . . .2 The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. . . . . . . . . . . Description to clarify the definition of diffuser efficiency . . . . . . . . . . . . . . . . 4. . . . . . . . . .1 Control volume for isothermal flow . . . . . .17 Figure for Example (5.2 6. 5. .8 A shock moves into a still medium as a result of a sudden and complete opening of a valve . 5. . 5. . . . . . . . .9 Schematic to explain the significances of the Impulse function . . . . . . . . . 5. . . . . 4. . . . . . 5. . . . .4 The flow in the nozzle with different back pressures . . . . . . . . . . . 5. . . . . . . . . . . . . . . . . . . . . . .1 A shock wave inside a tube. . . . . . . . . . . . . . . . . . . . . . . . . 5. . . . . . 137 . . . . . . . 4.7 The comparison of nozzle flow . . . . . . . . . . . . . . . . . . 5. . . . 5. . . . . . . . . . . . 4. . . . . k . . . . . . . . . . . . . . . .3) . . . . . . . . 4. . .6 Various ratios as a function of Mach number for isothermal Nozzle . . . . . . . 5. . but it can also be viewed as a one– dimensional shock wave. . . . 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. . . . . . . .10 The maximum of “downstream” Mach number as a function of the specific heat. .4 The ratios of the static properties of the two sides of the shock. . . .18 The results for Example (5. . . . . . . . . . . . . . . . . . . . . .5 Comparison between stationary shock and moving shock in ducts . . . . . . . . . 5. . 5. . . . . . .15 Figure for Example (5. . . . . . . . . . . . .13 A shock as a result of a sudden and partially a valve closing or a narrowing the passage to the flow . . . . . . . . . . . . . . . . . . .6 Comparison between a stationary shock and a moving shock in a stationary medium in ducts. . .12 The results of the partial opening of the valve. . . . .14 Schematic of a piston pushing air in a tube. . . . .9 The number of iterations to achieve convergence. . . . . . . . . 6. . . 5. . .1 6. . . . . . . . . .x LIST OF FIGURES 4. . . . . . . . . . . . . . A nozzle with normal shock . . .8 Comparison of the pressure and temperature drop as a function of the normalized length (two scales) . . .5 The relationship between the cross section and the Mach number on the subsonic branch . . .11 A shock moves into a moving medium as a result of a sudden and complete open valve. . . .10 Schematic of a flow of a compressible substance (gas) thorough a converging nozzle for example (4. . . .7) . . . . . . . . . .10) . . . . . . . . .

. .12 9. . 11.4) . . . . . . . . . . . .2 The Temperature Entropy Diagram For Rayleigh Line . . . 155 Various parameters in Fanno flow as a function of Mach number . . . . . . . . . . .3 The Mach number at the entrance to a tube under isothermal flow fL . . . . 172 The Mach numbers at entrance and exit of tube and mass flow rate fL for Fanno Flow as a function of the 4D . . . 185 10. . .9 9.1 9. . . . . 170 fL The effects of increase of 4D on the Fanno line . . . . . . . . . . . . temperature relationships as a function of the Mach number for isothermal flow . . . . . .6 9. . . 166 The maximum length as a function of specific heat. .13 9. . .2) . . . . . . . . . . . . . . . . . . . . . . . . .175 fL fL The pressure distribution as a function of 4D for a short 4D . . . . M1 to the tube as a function fL of 4D supersonic branch . . . . . . .3 for Fanno Flow . . . . . . . . 171 The development properties in of converging nozzle . . . 153 model as a function 4D 9. . 11. . 143 8. . . . . . 177 4f L 4f L . .4 The pressure assumptions in the chamber and tube entrance .15 9. . . . . . .2 9. . . .17 9. . . . . . . . 4D supersonic branch . . . . . . . . . . . . . . . .5 The reduced time as a function of the modified reduced pressure . . 164 The schematic of Example (9. . . . . . . . . . . .3 9. . . . . . . . . . . .1) . . . . . . . . . . . . . . . . . . . . . . . . . 191 10. . . . .18 Control volume of the gas flow in a constant cross section . . . . . 179 fL fL Fanno Flow Mach number as a function of 4D when the total 4D = 0. . . . .2 A schematic of two possible connections of the tube to a single chamber . 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. . . . . . . . . . . . . . . . .1 The control volume of the “Cylinder” . . . . . . . . . . 189 10. . . . .2 Description of the pressure. . . . . . . . .8 9. . . . .5 9. 182 The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow . . 174 fL M1 as a function M2 for different 4D for supersonic entrance velocity.10 9. . . . . . . . . . .3 . . . . . . . . . . . . . .4 9. . . . . . . .1 The control volume of Rayleigh Flow . . . . . . . . . . . 11. . . . . . . . . . . . . . . . .3 A schematic of the control volumes used in this model . . . . . . . . .3 The basic functions of Rayleigh Flow (k=1. . 195 11. .16 9. . .7 9. . . . 218 . . . . . . . . . k . . . . . . . . . .14 9. . . . . . 181 The maximum entrance Mach number. . . . . . . . . . . . .6 The reduced time as a function of the modified reduced pressure . . . . . 11. . . . 201 202 202 203 210 212 12.11 9. . . . . .LIST OF FIGURES xi 8. . . . . 181 fL The extra tube length as a function of the shock location. . . . . . .1 The two different classifications of models that explain the filling or evacuating of a single chamber . . 173 4f L M1 as a function M2 for various D . . . . . . 163 Schematic of Example (9. . 11. . . . . . . . . 180 Schematic of a “long” tube in supersonic branch . .

. . . .8 The calculation of D (possible error). . . . . . .1 The definition of the angle for the Prandtl–Meyer function. . . . . . . . . . . It can be noticed that the normal shock. . . .18Schematic for Example (14. . . . . .1 A view of a straight normal shock as a limited case for oblique shock 227 14. . . . . . . . . . . . However. . .7 The Mach waves that are supposed to be generated at zero inclination. . . .10Schematic of finite wedge with zero angle of attack. . . . . .4 The pressure ratio as a function of the dimensionless time . . . . . . . 245 14. .5 The view of a large inclination angle from different points in the fluid field. . . 265 15. . . . . .19Schematic for Example (14. . . 253 14. . . . . . . . . .2.2 The angles of the Mach line triangle . This photo is courtesy of Dr. .21Typical examples of unstable and stable situations . . . 241 14. . shock angle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 14. . . . . . . . . . . . . . . . 266 . . . . .22The schematic of stability analysis for oblique shock. .3 The schematic of the turning flow . . . . . . . . . .13The schematic for a symmetrical suction section with Mach reflection 248 14. . . 264 15. . 251 14. . . . . . . . . . . . . . . . 248 14. . . . . . . . . . . the strong shock. . . . . . .6 The various coefficients of three different Mach numbers to demonstrate that D is zero . . . . . . . According to his measurement. . . . . . . . . . 252 14. which is around zero. . . . . . . . . . . . . . . . . . .xii LIST OF FIGURES 12. . . . and the weak shock coexist. . .12The schematic for a round–tip bullet in a supersonic flow . . . . . . . . Grigory Toker. . . . . . . . . . . . . . . . . . . 235 14. . . .15Oblique shock occurs around a cone. . . . . . . . . . . . . . . . 239 14. Prandtl-Meyer function approaches closer to a zero deflection angle. . . . .4) . . . and exit Mach number for M1 = 3 . . . . . . . . .17Two variations of inlet suction for supersonic flow . . . . . . .4 Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. a Research Professor at Cuernavaco University of Mexico. . . .20Schematic of two angles turn with two weak shocks . . . . 236 14. . . . 223 12. . . . . . . . . . . . 229 14. . . . . . . . . 250 14. . . . . . . . . . . . . . . . . . . .11A local and a far view of the oblique shock. 224 12. . . . . .3 A typical oblique shock schematic . . . . 254 14. . . . . . . 247 14. . . . . . 265 15. . . . . . . . . . . Notice that both have a maximum point and a “no solution” zone. . . . .2 The pressure ratio as a function of the dimensionless time for chokeless condition . . . . 228 14. . .2 The regions where oblique shock or Prandtl–Meyer function exist. . .0. . . . . . . . .240 14. . . . . . . the cone half angle is 15◦ and the Mach number is 2. . . . . . . . . . . . . . . . . . . . . . . . .14The “detached” shock in a complicated configuration sometimes referred to as Mach reflection . . . . . 224 14. . . . . . . . .5) . . . . . . . .9 The possible range of solutions for different parameters for given upstream Mach numbers . . . . . . . .16Maximum values of the properties in an oblique shock . . . 244 14. . . 251 14. . . . . 243 14.3 The pressure ratio as a function of the dimensionless time for choked condition . . .

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

xiv LIST OF FIGURES .

. . . . . . . . . . . . .4 . .3 Solids speed of sound. . . . . . . . 1972 . . . 116 5. . . . 4. . Manual of Sound Recording. . . . . .1 The shock wave table for k = 1. . . . .3) 121 xv . . . . . . London: Fountain Press. .3 Table for shock propagating from suddenly opened valve (k=1. . . . .3 Isothermal Table (continue) . . . . . . . . 4. . . . . . . . . . . . . . . . .1 The shock wave table for k = 1. 35 36 37 58 59 60 62 63 73 74 5. . . . . . . . . . . . .4 . . . . . . . . . . . . . . . 4. . 117 5. . . 4. . . after Aldred. . . . . 4. . . . .3) 120 5. . . .4 Table for shock propagating from a suddenly opened valve (k=1.3 Isothermal Table . . 3. . . 1972 . .1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . . . . after Aldred. . . . . . . . . . . . . . .LIST OF TABLES 3. . . . 3. . . . . . . . . . . . . . . . . .3) 119 5.4) . . . .4 (continue) . . . . . . . . . . . . . . . . . . . . . . . . . . .4)(continue)118 5. . . . .4 Table for shock propagating from a suddenly opened valve (k=1. . . London:Fountain Press. . .2 Table for Reflective Shock from suddenly closed valve (end) (k=1. . . . . . . . . . 118 5. . . .2 Isentropic Table k=1. . .4 Table for shock propagating from a suddenly opened valve (k=1. . . .2 Table for a Reflective Shock from a suddenly closed end (k=1. .1 Fliegner’s number and other paramters as function of Mach number (continue) . . .4) . .2 Isentropic Table k = 1. . .4 (continue) . . . . . Manual of Sound Recording. 119 5. . . . . . . 115 5. . John. . . . . .4) .4 (continue) .3 Table for shock propagating from suddenly opened valve (k=1.1 Fliegner’s number and other paramters as function of Mach number 4. . . . . . .1 The shock wave table for k = 1. . . . John. . .2 Liquids speed of sound. . . . . . . . . . 117 5.1 Water speed of sound from different sources . 4. .

. . . . .4 . . . . . . . .1 Rayleigh Flow k=1. . . . . . . 188 10. . . . . . . . . . . . . . . . . .1 Rayleigh Flow k=1. . 147 The flow parameters for unchoked flow . . .1 9. . . . 187 Fanno Flow Standard basic Table (continue) . . . . . . . . . .1 Table of maximum values of the oblique Shock k=1. . . 246 . 194 10. . . . . .4 (continue) . . . . . 245 14. . . . .xvi 8. . . . 195 14. . . . . . . . . 152 Fanno Flow Standard basic Table . . . . . . . .4 . . . . . . . . . . . . . . . . .1 Rayleigh Flow k=1. . .4 . . . . . . . . . . . . . . . . . . .4 9. . . . . .1 8.1 Maximum values of oblique shock (continue) k=1. . . .1 LIST OF TABLES The Isothermal Flow basic parameters . . . 193 10. . . . . .4 (continue) . . . . . . .

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

and a Back-Cover Text may be at most 25 words. either copied verbatim. in any medium. Any member of the public is a licensee. royalty-free license. as being those of Invariant Sections. or of legal. APPLICABILITY AND DEFINITIONS This License applies to any manual or other work. A Front-Cover Text may be at most 5 words. has been arranged to thwart or discourage subsequent modification . in the notice that says that the Document is released under this License. and that is suitable for input to text formatters or for automatic translation to a variety of formats suitable for input to text formatters. refers to any such manual or work. or with modifications and/or translated into another language. If the Document does not identify any Invariant Sections then there are none. But this License is not limited to software manuals. regardless of subject matter or whether it is published as a printed book. it can be used for any textual work. as Front-Cover Texts or Back-Cover Texts. if the Document is in part a textbook of mathematics. and is addressed as ”you”. or absence of markup. 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. modify or distribute the work in a way requiring permission under copyright law. represented in a format whose specification is available to the general public. If a section does not fit the above definition of Secondary then it is not allowed to be designated as Invariant. Such a notice grants a world-wide. 1. The ”Invariant Sections” are certain Secondary Sections whose titles are designated. You accept the license if you copy. in the notice that says that the Document is released under this License. A ”Transparent” copy of the Document means a machine-readable copy. a Secondary Section may not explain any mathematics. below.) The relationship could be a matter of historical connection with the subject or with related matters. ethical or political position regarding them. We recommend this License principally for works whose purpose is instruction or reference. to use that work under the conditions stated herein. The ”Cover Texts” are certain short passages of text that are listed. philosophical. (Thus.xviii LIST OF TABLES come with manuals providing the same freedoms that the software does. commercial. A copy made in an otherwise Transparent file format whose markup. The Document may contain zero Invariant Sections. A ”Modified Version” of the Document means any work containing the Document or a portion of it. that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. unlimited in duration. 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. The ”Document”.

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

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

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

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

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

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

3. Disclaimer of warranty of the original author You may copy and distribute exact replicas of this document as you receive it. The original version of this document may be found at http://www. you must document the nature of those modifications in the ”Credits” section along with your name. without prior written permission of the Dr. 2. 4. .shtml LICENSE Terms and Conditions for Copying. You may distribute such modifications.” or alleged association with this author nor may “POTTO” or “POTTO Project” appear in their name.org/copyright. 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. 5. and information concerning how you may be contacted. You may at your option offer warranty in exchange for a fee. and Modifying 1. You must have a reasonable way to contact you. 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. keep intact all the copyright notices that refer to this document. You may offer instructional support for this document and software exchange for a fee.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. all the material based on this original content or work. Genick BarMeir. org/opl. Distributing.html This document is derived from open content license http://opencontent. under the terms of Section 1 above. potto. You may at your discretion charge a fee for the media and/or handling involved in creating a unique copy of this document. Derived Name(s) Products derived from this software may not be called “POTTO Project. in any medium. provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty of the original author. Your Name and Communication With You If you wish to modify this text or software in any way.

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

) are listed by name only for reasons of brevity. etc. The only ”catch” is that credit must be given where credit is due. contact info. • Contact at: barmeir@gmail. Credits All entries arranged in alphabetical order of surname.CONTRIBUTOR LIST How to contribute to this book As a copylefted work. just smaller in the sense of less text changed.” it is in no way inferior to the effort or value of a ”major” contribution. Please understand that when I classify a contribution as ”minor. you must follow the same guidelines as for any other GDL copyrighted work. etc. time. and resources to make this a better book! • Date(s) of contribution(s): 2004 to present • Nature of contribution: Original author.com John Martones • Date(s) of contribution(s): June 2005 xxvii . date. Minor contributions (typo corrections. this book is open to revision and expansion by any interested parties. Any and all contributions are gratefully accepted. 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. 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).

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

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

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

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

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

0.” In this process. These data (tables.3 0. These books should be considered more as a project than to fit the traditional definition of “plain” books.0. Thus. the traditional role of author will be replaced by an organizer who will be the one to compile the book.CREDITS xxxiii adding a question and perhaps the solution. 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.) can be redone by anyone who has the time to do it. It is hoped that the book will be error-free.0 0.0.0.1 0. while in other cases This may merely be the person who decides what will go into the book and what will not (gate keeper). other books contain data5 which can be A typeset in L TEX. Nevertheless. these works will have a version number because they are alive and continuously evolving.01 0. 5 Data are not copyrighted.0 0.0 0. The organizer of the book in some instances will be the main author of the work. .0 0. this method is expected to accelerate the creation of these high quality books. These books are written in a similar manner to the open source software process. better discussions or better explanations are all welcome to these books. graphs and etc. But more than that. Someone has to write the skeleton and hopefully others will add “flesh and skin.0. It is also hoped that others will contribute to the question and answer sections in the book. chapters or sections can be added after the skeleton has been written. Thus. the contributions to books can be done by many who are not experts.0.0. some errors are possible and expected.4 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). contributions can be made from any part of the world by those who wish to translate the book. Thus.0 0.0. Unlike a regular book. Additionally.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.0.0 0. Even if not complete.

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

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). While all these requests are important.A. this author cannot add all the things that he was asked and want to the book in instant fashion. Version 0. The download to this mount is over 25.Prologue For This Book Version 0./Arizona) and some other reactions. 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. I also got request from a India to write the interface for Microsoft.4. Several sections were add to this book with many new ideas for example on the moving shock tables. and the author is apologizing for this omission. Dr. xxxv .4.3 Sep. For example. the ideas of moving shock will reduced the work for many student of compressible flow. Menikoff pointed to this inconsistency.000 downloaded and is encouraging to receive over 200 thank you eMail (only one from U. However. I am sorry will not be entertaining work for non Linux/Unix systems.000. While it moves earlier but the name was not changed. the time is limited and they will be inserted as time permitted. especially for Microsoft. The moving shock issues are not completed and more work is needed also in the shock tube. one of the reader ask why not one of the example of oblique shock was not turn into the explanation of von Neumann paradox.S. If one want to use the software engine it is okay and permitted by the license of this work. Nevertheless. 2006 The title of this section is change to reflect that it moved to beginning of the book.2 It was surprising to find that over 14. 15.

POTTO project books are open content textbooks. The additions of this version focus mainly on the oblique shock and related issues as results of questions and reactions on this topic. most readers reached to www. . Version 0.potto. several questions were about compressibility factor and two phase flow in Fanno flow and other kind of flow models. the two phase aspects were added. Some examples were added to the normal shock chapter. The other questions that appeared related two phase and connecting several chambers to each other. Thus. The isothermal nozzle was combined with the isentropic chapter.org by searching for either terms “Rayleigh flow” (107) and “Fanno flow” ((93). This knowledge should be spread to other fields because it needed there as well. At this stage. The a brief chapter on Prandtl–Meyer flow was added. The sixth chapter deals with external forces fields. the connection between Prandtl–Meyer flow and oblique is an note form. Henderson and Ralph Menikoff. and hopefully it will be near future. This “rejection” is especially true when students feel that they have to go through a “shock wave” in their understanding. Furthermore. it reaches to about 30% (2011). The only exception is the addition to Taylor–Maccoll flow (axisymmetricale conical flow) in Prandtl -Meyer function (currently in a note form). This premise should not be assumed. an individual asked whether this author intended to write about the unsteady section. This assumption drives students from other fields away from this knowledge. This book is the second book in the series of POTTO project books. the questions that appear on the net will guide this author on what is really need to be in a compressible flow book. This led to comparison between maximum to ideal gas model to more general model. However. At this time. Also. The chapter about oblique shock was added and it contains the analytical solution.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. The fifth chapter deals now with normal shock in variable area ducts. In the chapter on speed of sound. extensive description of the compressible flow history was written. If the total combined variation search of terms “Fanno” and “Rayleigh” (mostly through google) is accounted. Version 0. most additions of the next version will be concentrated on Fanno flow and Rayleigh flow. This indicates that these topics are highly is demanded and not many concerned with the shock phenomena as this author believed and expected. 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. In the introductory chapter.4 Since the last version (0.3) several individuals sent me remarks and suggestions.xxxvi LIST OF TABLES This textbook has sections which are cutting edge research8 .

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

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

If a community will pull together. He made me realized that the common approach to supersonic branch of isothermal as non–existent is the wrong approach.How This Book Was Written This book started because I needed an explanation for manufacturing engineers. Anyone can be part of it. 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 . there are some chapters in this book which are original materials never published before. The presentation of some of the chapters is different from other books. 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. The book 13 Still in untyped note form. Apparently many manufacturing engineers and even some researchers in manufacturing engineering were lack of understanding about fluid mechanics in particularly about compressible flow. Therefore. I wrote notes and asked myself what materials should be included in such a book so when I provide consultation to a company. xxxix . I realized that books in compressible flow are written in a form that is hard for non fluid mechanic engineer to understand. even my 10 years old son. Eliezer made me change the chapter on isothermal flow. I do not need to explain the fundamentals. 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. this book is designed to be in such form that is easy to understand.” Later. Therefore. I wrote to myself some notes and I converted one of the note to a chapter in my first book. “Fundamentals Of Die Casting Design. the best books can be created. There is no reason why a text book which cost leas than 10$ to publish/produce will cost about 150 dollars. For example. I realized that people need down to earth book about compressible flow and this book was born. Therefore.

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

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

1. . Change the logtable to tabular environment for short tables.7 had several bug fixes and add two angle calculations to the oblique shock.xlii LIST OF TABLES Version 4.7 Version 4.1.

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

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

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

). Speed of sound in wet steam. 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. This concept in relation of the wind tunnel and atmosphere with varied density and temperature. . Nozzle The effect of external forces (add problems).xlvi LIST OF TABLES be further illuminate. Real gases effects (only temperature effects) Flow with “tabulated gases” calculations Phase change and two phase flow (multi choking points) effects (after 1. The combined effects of isentropic nozzle with heat transfer (especially with relationship to the program. Normal Shock Extend the partialy (open/close) moving shock theory. Mixed gases and liquids. Additional material can be provided to give a different angle on the issue at hand. Speed of Sound Discussion about the movement in medium with variation in speed of sound. 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. The dimensional analysis of the flow when the flow can be considered as isothermal. More problems in relationship to two phase.0 version).

Provide discussion on variations of the effecting parameters. The FORTRAN version will not be included. Application Marcell–Taylor (from the notes) Examples Transient problem .1.VERSION 4. Oblique Shock Add application to design problems Real Gas effects Prandtl–Meyer The limitations (Prandtl-Meyer). 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++.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.

xlviii LIST OF TABLES .

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

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

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

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

Stokes in 1848 retreated from his work and wrote an apology on his “mistake. The major breakthrough in the understanding of compressible flow came when Ernest Mach “stumbled” over the discontinuity. in his 1860 thesis. conservation of energy was a concept that was applied only to mechanical energy. 14 (1808).” Philos. He had then correctly derived the jump conditions that discontinuous solutions must satisfy. France. XXXII (1848). Mathematical and Physical Papers. From Classic Papers in Shock Compression Science. Neither of these groups would agree to “merge” or “relinquish” control of their “territory” to the other.3. 1803-1882. Challis had noticed contradictions concerning some solutions of the equations of compressible gas dynamics13 . 3-65. these problems are more common than one might be expected. Ernst Mach studied several fields in physics and also studied philosophy. There are even conflicts from not giving enough credit. while many of the mathematical techniques are currently practiced. Matter. French mathematician. New York. Reprinted from the original journals and transactions. ”M’emoire sur la th’eorie du son. 1880-1905. worked at Cambridge.3. Ec. Poisson found a “solution” to the Euler’s equations with certain boundary conditions which required discontinuity12 which had obtained an implicit form in 1808. Riemann. perhaps the most famous is that of Newton’s netscaping (stealing and embracing) Leibniz[’s] invention of calculus. Major material like Fanno flow isn’t taught in many classes. University Press. Again the “jumping” conditions were redeveloped by two different researchers independently: Stokes and Riemann. with additional notes by the author. HISTORICAL BACKGROUND 5 1. Even on this subject. Yet for many it was still debatable up to the middle of the nineteen century. was not sure whether or not discontinuity is only a mathematical creature or a real creature. 319-392. Cambridge. At first.1. 1781-1840 worked in Paris. So. In the early 18xx. His research centered on optic aspects which lead him to study acoustic and therefore supersonic flow (high speed. 12 Simeon ´ Denis Poisson. Aside from the conflicts of scientists with the Catholic Church and Muslim religion. Later. It is widely believed that Mach had done his research as purely intellectual research. Shock Compression Condens. England UK. ”On the velocity of sound. 494-499 14 Stokes George Gabriel Sir. English Astronomer.”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). Even the undersigned encountered individuals who have tried to ride on his work. .2 The shock wave puzzle Here is where the politics of science was a major obstacle to achieving an advancement10 . He was mostly interested in experimental physics. Springer. 13 James Challis. 1998.” J. Polytech. like Moody not giving the due credit to Rouse. a different group of scientists dealt with calorimetry (internal energy). 11 This recognition of the first law is today the most “obvious” for engineering students. The other kind of problem is “hijacking” by a sector. High-press. It was easier to publish articles about the second law of thermodynamics than to convince anyone of the first law of thermodynamics. since no Mach number was known at that time). It took about a century to establish the first law11 . At this stage some experimental evidence was needed. the Aeronautic sector “took over” gas dynamics as did the emphasis on mathematics like perturbations methods or asymptotic expansions instead on the physical phenomena. Poisson showed that solutions could approach a discontinuity by using conservation of mass and momentum. it is logical to believe that his interest had risen due to the need to achieve 10 Amazingly. However. On the other side. science is full of many stories of conflicts and disputes. Mag.

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

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

The first that attempt this question where in 1931 by Epstein23 . part of analysis of this book is also redoing old work. In fact analysis based on Hamilton’s principle isn’t suitable for stability because entropy creation was recognized 1955 by Herivel25 . J. 1931. For example.. 1963. similar analysis was published by Briggs. 1948. 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. 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 results of that analysis was that strong shock is unstable.. Emanuel’s partial solution just redone 36 years work (how many times works have to be redone in this field).” Journal of Fluid Mechanics 366 (1998) pp. Henderson and Menikoff27 dealt with only the procedure to find the maximum of oblique shock. December 2000 (US$ 124. 25 Herivel.F. New York. 51. Carrier26 was first to recognize that strong and weak shocks stable. Inc.. 532-547.” AIAA Journal Vol 2. . That solution is based on a transformation of sin θ to tan β . even this author sinned in this erroneous conclusion. but no comparison between real gases and ideal gas is offered there. The effect of real gases was investigated very early since steam was used move turbines. 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. Vol. Even all books that were published recently claimed that no strong shock was ever observed in flow around cone (Taylor–Maccoll flow). 2. pp. 179–210. 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. G. F. J.. S. INTRODUCTION in the famous report of NACA 1135 that explicit analytical solution isn’t possible22 . 27 Henderson and Menikoff.8 CHAPTER 1. 974. Yet. “The Derivation of The Equations of Motion On an Ideal Fluid by Hamilton’s Principle.” Proceedings of the National Academy of Science. His analysis was based on Hamilton’s principle when he ignore the boundary condition. No 5 p.” Quarterly of Applied Mathematics. recently. p. 1949. “On the Stability of the supersonic Flows Past as a Wedge. 317.” published by Interscience Publishers. pp.4 that it was clear that the Normal shock along with strong shock and weak shock “live” together peacefully and in stable conditions. 23 Epstein. “Comment on Calculation of Oblique shock waves.” Proceedings of the Cambridge philosophical society. In a way. 1955. “On the air resistance of Projectiles. George. 22 Since writing this book. ”Triple Shock Entropy Theorem. While the actual solution all the roots are real and positive for the attached shock. 6. The presentation was missing the condition for the detachment or point where the model collapse. P. Vol. 17. In fact. 344-349. It is amazing that early research concluded that only the weak solution is possible or stable as opposed to the reality. several individuals point out that a solution was found in book “Analytical Fluid Dynamics” by Emanuel. If fact the confusion on this issue was persistent until now. second edition. 367–378. Pt. pp.90). Vol. 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. 24 In study this issue this author realized only after examining a colleague experimental Picture 14. This difference probably results in not recognizing that The boundary conditions play an important role in the stability of the shock24 . Hence. But more surprisingly. It is interesting that transformation result in one of root being negative. 26 Carrier.

and when it was found the significance of the choking phenomenon was not clear.4. It is likely that others had noticed that flow is choked. Choking wasn’t clearly to be observed. HISTORICAL BACKGROUND 9 The moving shock and shock tube were study even before World War Two. 68–72. 28 Fliegner Schweizer Bauztg. there was no realization that the flow is choked just that the flow moves faster than speed of sound. p. As a result deLavel’s nozzle was invented by Carl Gustaf Patrik Fig.1. trial and error method was the only method to solve this problem. leading to the study and development of Rayleigh flow.. . even when researcher stumbled over it. The theoretical first work on this issue was done by Zeuner. 1.” Leipzig 1899.3 of this book (as far it is known. 29 Rayleigh was the first to develop the model that bears his name. 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. 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. No one was looking for or expecting the choking to occur. page 268 f. The first to discover (try to explain) the choking phenomenon was Rayleigh29 . “Theorie die Turbinen. Later. 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. this book is first to publish this tables). 1. but did not produce any model or conduct successful experimental work. 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. but after a certain point it did matter (because the steam was choked). The introduction of the steam engine and other thermodynamics cycles led to the choking problem. Only after the dimensionless presentation of the problem and the construction of the moving shock table the problem became trivial. Yet. Vol 31 1898.3. Steam and Gas Turbine operation by another inventor (Curtis) 1896 used in steam turbine.3 Choking Flow The choking problem is almost unique to gas dynamics and has many different forms.2: The schematic of deLavel’s turbine afde Laval in 1882 and first successful ter Stodola.3. According the thermodynamics theory (various cycles) the larger heat supply for a given temperature difference (larger higher temperature) the larger the output. Up to this version 0.

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

1. when the first supersonic wind tunnel was built. 36 Bendemann Mitteil uber ¨ Forschungsarbeiten. A different reason to study the converging-diverging nozzle was the Venturi meter which was used in measuring the flow rate of gases. As a result. 1907.3: The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines and 1947-48. There are two main models or extremes that describe the flow in the nozzle: isothermal and adiabatic. Berlin. At the end of the arm was the object that was under observation and study. the spinning arm was no longer used in testing. No. 37. in 1904 the Wright brothers demonstrated that results from the wind tunnel and spinning arm are different. . due to the circular motion. Between the turn of the century Fig. Yet.3. HISTORICAL BACKGROUND Nozzle flow 11 The first “wind tunnel” was not a tunnel but a rotating arm attached at the center. The arm’s circular motion could reach a velocity above the speed of sound at its end. 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. several models that explained choking at the throat have been built.545) that creates the maximum flow rate.1.

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

). Thus. Fanno attributes the main pressure reduction to friction. HISTORICAL BACKGROUND 13 Mumenthaler from UTH University. No experimental evidence. flow that is dominantly adiabatic could be simplified and analyzed. 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. but can be explained by choking at a lower Mach number.8 Mach or more the flow must be considered to be a compressible flow. known by the undersigned. was ever produced to verify this finding. The author invites others to help in this information. At that stage. 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. 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. 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. a shock wave occurs which increases the resistance.4 External flow When the flow over an external body is about . The friction factor is the main component in the analysis as Darcy f 41 had already proposed in 1845. However at a Mach number above 0. If it turned out that no one had done it before Shapiro. In reality. the flow is √ choked somewhere between 1/ k to one for cases that are between Fanno (adiabatic) and isothermal flow. the question cannot be answered yet.1. Shapiro organized all the material in a coherent way and made this model useful. no copy of the thesis can be found in the original University and perhaps only in the personal custody of the Fanno family40 . The choking was assumed only to be in the subsonic flow. This fact was evident in industrial applications where the expectation of the choking is at Mach one. 1. But because the actual Fanno’s thesis is not available. Without the existence of the friction factor data. The arrival of the Moody diagram.8 (relative of velocity of the body to upstream velocity) a local Mach number (local velocity) can reach M = 1. f based radius is only one quarter of the Darcy f which is based on diameter .3. Additionally an understating of the supersonic branch of the flow was unknown (The idea of shock in tube was not raised at that time. the Fanno model wasn’t able to produce a prediction useful for the industry. this flow model should be called Shapiro’s flow. which built on Hunter Rouse’s (194x) work made Darcy– Weisbach’s equation universally useful.3.

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

In this section there is no originality and none should be expected. Later. not realizing that garbage in is garbage out. The analytical solution for forced volume and the “balloon” problem (airbag’s problem) model were published first in this book (version 0. There are many figures that should be included and a biased selection was required. such as the Internet.1. the model and solution to the nozzle attached to chamber issue in his book “Fundamentals of Gas Dynamics.35) in 2005. in die casting conference there was a discussion and presentation on which turbulence model is suitable for a complete still liquid. Other “strange” models can be found in the undersigned’s book “Fundamentals of Die Casting Design. the emergence of the CFD gave the illusion that there are solutions at hand. 46 In 45 International . 1. The classification of filling or evacuating the chamber as external control and internal control (mostly by pressure) was described in version 0. Much information can be obtained from other resources. Later several researchers mostly from the University in Illinois extended this work to isothermal nozzle (choked and unchoked). Pennsylvania. fact. the model has to be based on scientific principles and not detached from reality.3.3 of this book by this author. 1964. The simplest model of nozzle. Textbook Co. HISTORICAL BACKGROUND 15 1. The earlier work dealing with this issue was by Giffen.3.6 Biographies of Major Figures In this section a short summary of major figures that influenced the field of gas dynamics is present. As anecdotal story explaining the lack of progress. A. Scranton.. and was republished by Owczarek. He also extended the model to include the unchoked case. In 1993 the first reasonable models for forced volume were published by the undersigned. J..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. 1940.”45 . Since World War II considerable works have been carried out in this area but with very little progress46 . that model was extended by several research groups. is not sufficient in many cases and a connection by a tube (rather just nozzle or orifice) is more appropriated..e. i.3.

and he was put under house arrest for the remainder of his life. Fig. (1602). During the time after his study. Galileo started his studying at the University of Pisa in 1581. this relationship did not last and Marina married Giovanni Bartoluzzi and Galileo’s son. Galileo’s observations got him into trouble with the Catholic Church. Galileo also proved that objects fell with the same velocity regardless of their size. Vincenzio. where she bore him three children. joined him in Florence (1613). Galileo died in 1642 in his home outside of Florence. Galileo invented many mechanical devices such as the pump and the telescope (1609). the church was not harsh with him. he made numerous discoveries such as that of the pendulum clock. The oldest of six children. however. . His telescopes helped him make many astronomic observations which proved the Copernican system. 1. because of his noble ancestry. Galileo was convicted after publishing his book Dialogue. 1564 to musician Vincenzo Galilei and Giulia degli Ammannati. He then became a professor of mathematics at the University of Padua in 1592. Galileo moved with his family in early 1570 to Florence.5: Portrait of Galileo Galilei Galileo had a relationship with Marina Gamba (they never married) who lived and worked in his house in Padua. INTRODUCTION Galileo was born in Pisa. Italy on February 15.16 Galileo Galilei CHAPTER 1. However.

before he entered the university of Vienna were he studies mathematics. Einstein was greatly influenced by it. when Czechia was still a part of the Austro–Hungary empire. 1.1916) published Die Machanik in which he formalized this argument. This was one of the primary sources of inspiration for Einstein’s theory of General Relativity. Eckert. and in 1918. physics and philosophy.1. Mach’s revolutionary experiment demonstrated the existence of the shock wave as shown in Figure 1. He graduated from Vienna in 1860. he named it Mach’s Principle. Mach’s shadowgraph 47 It is interesting to point out that Prague provided us two of the top influential researchers[:] E. and in 1867 he received a position in the Technical University in Prague47 where he taught experimental physics for the next 28 years. His experiments required exact timing. Mach and E.7: The Photo of the bullet in a supersonic flow that could take a photo at superMach made. There Mach wrote his thesis ”On Electrical Discharge and Induction.7. 1. was a high school teacher who taught Ernst at home until he was 14.” Mach was Fig. 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.R.6: Photo of Ernest Mach interested also in physiology of sensory perception. Note it was not taken in a wind tunnel sonic speeds. when he studied in Kromeriz Gymnasium. Johann. Mach’s father. He then turned to physics. He was not able to attach the camera to the arm and utilize the remote control (not existent at that time).G. Later. 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. In 1863. Mach was also a great thinker/philosopher and influenced the theory of relativity dealing with frame of reference.3. Ernest Mach (1836 . . but he declined. HISTORICAL BACKGROUND Ernest Mach (1838-1916) 17 Ernst Mach was born in 1838 in Chrlice (now part of Brno).

John William Strutt (Lord Rayleigh) A researcher with a wide interest. He graduated in the Mathematical Tripos in 1865 as Senior Wrangler and Smith’s Prizeman. electrodynamics. Lord Rayleigh’s first research was mainly mathematical. clearly 48 Mach dealt with only air. In 1866 he obtained a fellowship at Trinity which he held until 1871. Lord Rayleigh was a British physicist born near Maldon. elasticity. Mach’s contributions to supersonic flow were not limited to experimental methods alone. light scattering. density of gases. Rayleigh was considered to be an excellent instructor. His Theory of Sound was published in two volumes during 1877-1878. In 1861 he entered Trinity College at Cambridge. Rayleigh was also a contributer to the Encyclopedia Britannica. He published 446 papers which. and from 1896 to 1919 he acted as Scientific Adviser to Trinity House. At that time there wasn’t the realization that the flow could be choked. . viscosity. wave theory. flow of liquids. but it is reasonable to assume that he understood that this ratio was applied to other gases. reprinted in his collected works. He was Lord Lieutenant of Essex from 1892 to 1901. Ackeret (Prandtl’s student) in 1932 in honor of Mach. 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). started studies in compressible flow mostly from a mathematical approach. His exceptional abilities soon enabled him to overtake his colleagues. six volumes issued during 1889-1920. on November 12. 1842. and photography. Rayleigh’s later work was concentrated on electric and magnetic problems. and his other extensive studies are reported in his Scientific Papers. where he commenced reading mathematics. The name Mach Number (M) was coined by J. color vision. concerning optics and vibrating systems.8: Photo of Lord Rayleigh president of the government committee on explosives. but his later work ranged over almost the whole field of physics. hydrodynamics. It seems that Rayleigh was the first who realized that flow with chemical reactions (heat transfer) can be choked. covering sound. INTRODUCTION technique and a related method called Schlieren Photography are still used today. the year of his marriage. Yet. Essex.18 CHAPTER 1. electromagnetism. capillarity. 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 served for six years as the Fig. 1.

Rankine. was a Justice of the Peace and the recipient of honorary science and law degrees. HISTORICAL BACKGROUND 19 show his capacity for understanding everything just a little more deeply than anyone else. Rankine was born in Edinburgh to British Fig. As a successor to James Clerk Maxwell.December 24. the Earl of Balfour (of the famous Balfour declaration of the Jewish state). 1820 . he was head of the Cavendish Laboratory at Cambridge from 1879-1884. Lord Rayleigh died on June 30.3. Rankine developed a theory of the steam engine. and his only brother and parents died before him. Lord Rayleigh. sister of the future prime minister.9: Portrait of Rankine Army lieutenant David Rankine and Barbara Grahame. 1872) was a Scottish engineer and physicist. Rayleigh died on June 30. William John Macquorn Rankine William John Macquorn Rankine (July 2. never allowing politics to interfere with science. He received the Nobel Prize in 1904. London. He was a Fellow of the Royal Society (1873) and served as Secretary from 1885 to 1896. and as President from 1905 to 1908. They had three sons. Rankine never married. 1. In 1871 he married Evelyn. He intervened in debates of the House of Lords only on rare occasions.1. a Chancellor of Cambridge University. and naval engineering in which he was involved in applying scientific principles to building ships. He was a founding contributor to the science of thermodynamics (Rankine Cycle). Essex. Essex. and in 1887 became Professor of Natural Philosophy at the Royal Institute of Great Britain. 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. at Witham. . strength of materials. Rankine was well rounded interested beside the energy field he was also interested in civil engineering. 1919 at Witham. 1919.

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

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

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

3.T. degree) in 1946 in mechanical engineering from MIT.” to transform the gas dynamics field to a coherent text material for engineers. turbomachinery and propulsion by jet engines and rockets. (It is M. HISTORICAL BACKGROUND 23 Eckert was an excellent mentor to many researchers (including this author). In 1965 he become the head of the Department of Mechanical Engineering until 1974.B. Shapiro moved to the field of biomedical engineering where he was able to pioneer new work. 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. In his first 25 years Shapiro focused primarily on power production. emphysema and glaucoma.D. He was assistant professor in 1943. While Shapiro viewed gas dynamics mostly through aeronautic eyes. Shapiro was instrumental in the treatment of blood clots. 51 Parts taken from Sasha Brown. The undersigned believes that Shapiro was the first one to propose an isothermal flow model that is not part of the aeronautic field.’s equivalent of a Ph.1. Therefore it is proposed to call this model Shapiro’s Flow.I. Ascher Shapiro passed way in November 2004. Shapiro grew up in New York City and received his S.D.D. MIT . and he had a reputation for being warm and kindly. Shapiro spent most of his active years at MIT. the Eckert equivalent for the compressible flow. Unfortunately for the field of Gas Dynamics. asthma. Furthermore. was instrumental in using his two volume book “The Dynamics of Thermodynamics of the Compressible Fluid Flow. Ascher Shapiro MIT Professor Ascher Shapiro51 . in 1938 and the Sc. three years before receiving his Sc. highspeed flight. He was also a leading Figure in bringing together engineering in the East and West during the Cold War years.

24 CHAPTER 1. INTRODUCTION .

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

26 CHAPTER 2.1) ∆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.

The mouse go out to investigate. The mouse ask the cat I thought I hear a dog. yes you right. In this chapter an introductory discussion is given about different situations which can appear the industry in regards to speed of sound. Yet. 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 cat reply. With the “new” knowledge from the consultation the calculations were within the range of acceptable results. the engineer was not ware of the effect of particles on the speed of sound. 1 Aerospace 27 . and cat is catching the mouse. one language is not enough today.CHAPTER 3 Speed of Sound 3.1 Motivation In traditional compressible flow classes there is very little discussion about the speed of sound outside the ideal gas. The above situation is not unique in the industry. The author thinks that this approach has many shortcomings. but a joke is must in this situation. Thus. As it will shown in this chapter. My teacher was right. In a recent consultation an engineer1 design a industrial system that contains converting diverging nozzle with filter to remove small particles from air. Hence. A cat is pursuing a mouse and the mouse escape and hide in the hole. The engineer was well aware of the calculation of the nozzle. Aerospace Department.2 Introduction engineer that alumni of University of Minnesota. 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. 2 Pardon. 3. the actual flow rate was only half of his prediction. the particles can. reduces the speed of sound by almost as half. Suddenly. the engineer was able to predict that was a chocking point.

5). Here. it is considered . Thus.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. it raises the question: what is the speed of the Fig.3) Substituting the expression for dU from equation (3. SPEED OF SOUND The people had recognized for several hundred years that sound is sound wave dU velocity=dU a variation of pressure.5) An expression is needed to represent the right hand side of equation (3.28 CHAPTER 3.4) yields c2 dρ ρ = dP dP =⇒ c2 = ρ dρ (3. To answer this question consider a piston moving from the left to the right at a relatively small velocity (see Figure 3. Applying the mass balance P+dP yields P ρ+dρ ρ ρc = (ρ + dρ)(c − dU ) (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. P is a function of two independent variables. 3. The information that the piston is moving passes thorough a single “pressure pulse.1) Fig. 3. Thus. 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. This velocity is referred to as the speed of sound. 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.” It is assumed that if the velocity of the piston is infinitesimally small.4) (3.2) From the energy equation (Bernoulli’s equation). the pulse will be infinitesimally small.2) into equation (3. the pressure and density can be assumed to be continuous.1).1: A very slow moving piston in a still gas small disturbance travel in a “quiet” medium. For an ideal gas.

3.5) can be derived from the momentum equation. S OLUTION The momentum equation written for the control volume shown in Figure (3.9) dP = c2 dρ This yields the same equation as (3. k namely P = constant × ρk (3. s) where s is the entropy.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. The pressure for an ideal gas can be expressed as a simple function of density.11) .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. and a function “molecular structure” or ratio of specific heats. Example 3. (3.6) In the derivations for the speed of sound it was assumed that the flow is isentropic.1: Demonstrate that equation (3. The full differential of the pressure can be expressed as follows: dP = ∂P ∂ρ dρ + s ∂P ∂s ds ρ (3. ρ.7) s Note that the equation (3.10) 3. therefore it can be written ∂P dP = dρ ∂ρ (3.5). SPEED OF SOUND IN IDEAL AND PERFECT GASES 29 that P = P (ρ.5) can be obtained by utilizing the momentum equation instead of the energy equation.3.

8226 ∆P ∆ρ kJ K kg kJ K kg kJ K kg (3. 3 This data is taken form Van Wylen and Sontag “Fundamentals of Classical Thermodynamics” 2nd edition .12) Remember that P/ρ is defined for an ideal gas as RT .327 × 461 × (350 + 273) ∼ 771.94199 and substituting into the equation yields c= m 200000 = 780.5 sec Note that a better approximation can be done with a steam table.61376 ρ = 6.30 and hence CHAPTER 3.7◦C: s ∼ 6. Classical Thermodynamics. (a) utilizes the steam table (b) assuming ideal gas.) √ m c = kRT ∼ 1.5 0.15) for ideal gas assumption (data taken from Van Wylen and Sontag.12) can be written as √ (3.32823 sec kg m3 (3. table A 8. and equation (3.0100 At 18[bar] and 300◦ C: s = 6.2: Calculate the speed of sound in water vapor at 20[bar] and 350◦ C. S OLUTION The solution can be estimated by using the data from steam table3 c= At 20[bar] and 350◦ C: s = 6.14) s=constant ρ = 6.13216 kg m3 kg m3 kg m3 After interpretation of the temperature: kJ At 18[bar] and 335. and it will be part of the future program (potto–GDC).13) c = kRT Example 3.9563 K kg ρ ∼ 6.46956 ρ = 7.9563 At 18[bar] and 350◦ C: s = 7. SPEED OF SOUND P c= dP constant × ρk = k × constant × ρk−1 = k × dρ ρ =k× P ρ (3.

the Gibbs function (3. Thus. The compressibility factor represents the deviation from the ideal gas. Recalling from thermodynamics. 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. To obtain the expression for a gas that obeys the law expressed by (3.17) This correction factor approaches one when TB −→ TA . What is the time it take for sound to travel from point “A” to point “B” under this assumption. It should be noted that velocity is provided as a function of the distance and not the time (another reverse problem).3.19) The speed of sound of any gas is provided by equation (3. x T = (TB − TA ) + TA h Where the distance x is the variable distance. 3. SPEED OF SOUND IN REAL GAS 31 Example 3.4.16) 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 (3.4 Speed of Sound in Real Gas The ideal gas model can be improved by introducing the compressibility factor. a real gas equation can be expressed in many cases as P = zρRT (3. The distance between “A” and “B” is denoted as h.? S OLUTION The temperature is denoted at “A” as TA and temperature in “B” is TB .3: The temperature in the atmosphere can be assumed to be a linear function of the height for some distances.19) some mathematical expressions are needed.7).18) (3.20) .

3: The Compressibility Chart is used to obtain T ds = dh − dP ρ (3. . 372 SI version. SPEED OF SOUND Fig.22) ρ From thermodynamics. perhaps to insert the discussion here.32 CHAPTER 3. 3.21) P The definition of volumetric specific heat for a pure substance is Cv = ∂u ∂T =T ρ ∂s ∂T (3.20) The definition of pressure specific heat for a pure substance is Cp = ∂h ∂T =T P ∂s ∂T (3.23) P Van Wylen p. it can be shown 4 dh = Cp dT + v − T 4 See ∂v ∂T (3.

30) yields dρ dP Cv = ρ P Cp z+T z+T ∂z ∂T P ∂z ∂T ρ (3. However.28) P Equation (3.29) ρ Equating the right hand side of equations (3.25) 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 (3.24) into 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 (3.28) and (3.25) P dP = Cp dT − T z (3.3. 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 (3.23) results  v  v  z T Simplifying equation (3.29) results in dρ R z+T ρ Cv ∂z ∂T = ρ dP R z+T P Cp ∂z ∂T (3.4.31) .24) Substituting the equation (3.28) can be integrated by parts.30) P Rearranging equation (3.27) Letting ds = 0 for isentropic process results in dP R dT = z+T T P Cp ∂z ∂T (3.26) P Utilizing Gibbs equation (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.

995times287 × 300 = 346. dP = nzRT dρ (3. S OLUTION According to the ideal gas model the speed of sound should be √ √ c = kRT = 1. The integration of equation (3.407. SPEED OF SOUND If the terms in the braces are constant in the range under interest in this study.403. What is different in these derivations is that a relationship between coefficient n and k was established. z = 0.31) yields ρ1 ρ2 n = P1 P2 (3. Assume that R = 287[j/kg/K ].11). and in–fact any thermodynamics book shows this relationship. equation (3.3 1.32) Note that n approaches k when z → 1 and when z is constant. .4 factor (0. Make the calculation based on the ideal gas model and compare these calculations to real gas model (compressibility factor).32) provides a tool to estimate n.7[m/sec] The correction factor for air under normal conditions (atmospheric conditions or even increased pressure) is minimal on the speed of sound.5) to calculated by ideal gas model. n is defined as k n= Cp Cv z+T z+T ∂z ∂T ρ ∂z ∂T P (3.995.407 × 287 × 300 ∼ 348.33) Equation (3.31) can be integrated.33) is similar to equation (3. the compressibility factor.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. However. Now. and z = 0. For example. a change in temperature can have a dramatical change in the speed of sound.34 CHAPTER 3.34) Example 3. n = 1. the speed of sound for a real gas can be obtained in the same manner as for an ideal gas.403 has √ √ c = znRT = 1.33) isn’t new. For short hand writing convenience. The specific heat for air is k = 1.403 × 0.3 and n ∼ 1 which means that speed of sound is only 0. But the definition of n in equation (3.1[m/sec] For the real gas first coefficient n = 1. This relationship (3.

Ohio: Chemical Rubber Co. The compressibility of the substance is the reciprocal of the bulk modulus.5). the average bulk modulus for water is 2. For example.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. Johnson. Physics.37). New York: Wiley.000 meters. SPEED OF SOUND IN ALMOST INCOMPRESSIBLE LIQUID 35 3. Many researchers have looked at this velocity. and for purposes of comparison it is given in Table (3. The ratio of the change in the fractional volume to pressure or compression is referred to as the bulk modulus of the material.5. At a depth of about 4. John D.36) dP dρ (3. The mathematical definition of bulk modulus as following B=ρ In physical terms can be written as c= For example for water c= 2. 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. The fractional volume change is only about 1.35) This agrees well with the measured speed of sound in water. The World Book Encyclopedia. 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]. For example. 601 Handbook of Chemistry and Physics.3. 1997: 468. According . the pressure is about 4 × 107 N/m2 . 1999.5) Remark Fresh Water (20 ◦ C) Distilled Water at (25 ◦ C) Water distilled reference Cutnell. 1482 m/s at 20◦ C. as can be observed from the equation (3. Chicago: World Book..2 × 109 N/m2 .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.8% even under this pressure nevertheless it is a change. & Kenneth W.

37) where c0 = 1449. using a general tabulated value for the bulk modulus. 3. the speed of sound in sea water depends on temperature. salinity. 5 J. and cST P is a correction factor between coupling of the different parameters. with different speeds in different directions. the speed of sound in solids is larger than in liquids and definitely larger than in gases. Wilson’s formula is accepted by the National Oceanographic Data Center (NODC) USA for computer processing of hydrological information. 1960.2: Liquids speed of sound. John. and hydrostatic pressure. Soc. cP is a function pressure. 1357.38) Compared to one tabulated value the example values for stainless steel lays between the speed for longitudinal and transverse waves. Amer.14 is about clean/pure water. after Aldred. Manual of Sound Recording. T.32. P ) = c0 + cT + cS + cP + cST P . p. Nevertheless. Wilson’s empirical formula appears as follows: c(S..36 CHAPTER 3. material Glycerol Sea water Mercury Kerosene Methyl alcohol Carbon tetrachloride reference 25 ◦ C Value [m/sec] 1904 1533 1450 1324 1143 926 Table 3. and differences between transverse and longitudinal waves. Acoust. cT is a function temperature. Speed of sound in solid of steel. and cS is a function salinity. London: Fountain Press. (3. vol.6 Speed of Sound in Solids The situation with solids is considerably more complicated. the speed of sound in liquids is about 3 to 5 relative to the speed of sound in gases. 1972 In summary. SPEED OF SOUND to Wilson5 . in different kinds of geometries. N 10. . gives a sound speed for structural steel of E = ρ 160 × 109 N/m2 = 4512m/s 7860Kg/m3 c= (3. Young’s Modulus for a representative value for the bulk modulus for steel is 160 109 N /m2 .

it assumed that two materials are homogeneously mixed. three extreme cases suggest themselves: the flow is mostly gas with drops of the other phase (liquid or solid).7.39) (3. 1972 3. 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) can be approximated as ρ =1+m ρa (3. about equal parts of gas and the liquid phase. London:Fountain Press.40) (3.3. In actuality. Topic for none homogeneous mixing are beyond the scope of this book. Manual of Sound Recording. However. For simplicity. and liquid with some bubbles. Indeed there is double chocking phenomenon in two phase flow. It further assumed that no heat and mass transfer occurs between the particles. 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. For small value of ξ equation (3. In that case.3: Solids speed of sound.7 Sound Speed in Two Phase Medium The gas flow in many industrial situations contains other particles. after Aldred. There can be several models that approached this problem. for homogeneous and under certain condition a single velocity can be considered. John.41) . there could be more than one speed of sound for two phase flow. The first case is analyzed.

47) (3. this analysis results in lower speed of sound compared to pure gas. It must noticed that m = constant. . Generally.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.39) and substituted into equation (3.6). the mixture isentropic relationship can be expressed as P where γ−1 R = γ Cp + mC Recalling that R = Cp − Cv reduces equation (3. Hence.46) T = constant (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 (3.43) 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 (3.44) Therefore.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. The gas density can be replaced by equation (3. It can be noticed that Rmix and γ are smaller than similar variables in a pure gas. the velocity of sound in slightly wed steam can be about one third of the pure steam speed of sound. For example. SPEED OF SOUND where m = is mass flow rate per gas flow rate. the velocity of mixtures with large gas component is smaller of the pure gas.41) R P = T ρ 1+m (3.38 m ˙b m ˙a CHAPTER 3.46) into γ= Cp + mC Cv + mC (3. The 1+m 1+m correction factors for the specific heat is not linear. 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 (3.48) At this stage the other models for two phase are left for next version (0.

3.49) Meta End . SOUND SPEED IN TWO PHASE MEDIUM Meta 39 For a mixture of two phases. speed of sound can be expressed as c2 = where X is defined as X= s − sf (PB ) sf g (PB ) (3.7.50) ∂P ∂P [f (X )] = ∂ρ ∂ρ (3.

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

In the same argument.27) For isentropic process ds ≡ 0 and combining equations (4. Conversely.11) are the same. P = ρRT .28) Differentiation of the equation state (perfect gas).25) with (4. Clearly. T ρ P U T+dT ρ+dρ P+dP U+dU 4. and dividing the results by the equation of state (ρRT ) yields dP dρ dT = + P ρ T (4.25) hen there is no external work and heat transfer. The control volume is shown in Figure (4. in mathematical terms. equation (4.4).46 CHAPTER 4. reads dh + U dU = 0 Differentiation of continuity equation. Therefore. it is constant for the other. the stagnation density is constant through the flow. the energy equation. 4.27) yields dP + U dU = 0 ρ (4.4: Control volume inside a convergingmodel.26) The thermodynamic relationship between the properties can be expressed as T ds = dh − dP ρ (4.1 The Properties in the Adiabatic Nozzle (4. the stagnation temdiverging nozzle. and dividing by the continuity equation reads dρ dA dU + + =0 ρ A U (4. There are two models that assume variable area flow: First is isentropic and adiabatic model. ρAU = m ˙ = constant. knowing the Mach number or the temperature will provide all that is needed to find the other properties. Second is isentropic and isothermal Fig. Examination of the relation between properties can then be carried out.29) . Thus. ISENTROPIC FLOW The important sub case in this chapter is the flow in a converging–diverging nozzle. is constant through the adiabatic flow because there isn’t heat transfer. T0 . If the right hand side is constant for one variable. The only properties that need to be connected are the cross section area and the Mach number.2. the stagnation pressure is also constant through the flow because the flow isentropic. perature.9) and equation (4.

2. dρ. it can be observed that dP can be positive or negative depending on the dA and Mach number.26) and using it in equation (4.33) is a differential equation for the pressure as a function of the cross section area. The meaning of the sign change for the pressure differential is that the pressure can increase or decrease.32) Equation (4.34) The pressure Mach number relationship Before going further in the mathematical derivation it is worth looking at the physical meaning of equation (4.33) 1− U c 2    dρ dP   =U  +  A dP ρ   2  dA  (4. The term ρU 2 /A is always positive (because all the three terms can be only positive). ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 47 Obtaining an expression for dU/U from the mass balance equation (4. It is convenient to rearrange equation (4.4. can be replaced by the static pressure.31) to obtain dP ρ Or in a dimensionless form dP dA 1 − M2 = U2 ρ A (4. Now. If the Mach number is larger than one than dP has opposite sign of dA.30) Rearranging equation (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 (4.34). If Mach number is smaller than .31) = U2 dA A (4.33) to obtain a variables separation form of dP = dA ρU 2 A 1 − M2 (4. It can be observed that the critical Mach number is one.28) reads dU U dP dA dρ − U2 + =0 ρ A ρ (4.30) so that the density.

dρ = 1 dP c2 (4.35) From equation (4. one can observe that the density. For the supersonic branch M > 1. the opposite.48 CHAPTER 4. The relationship between the velocity and the pressure can be observed from equation (4. Conversely. the equation of state was not used. the phenomenon is different. Thus. if the area decreases (as a function of x) the pressure decreases. the equations are applicable for any gas (perfect or imperfect gas). 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. An increase in area results in an increase of the static pressure (converting the dynamic pressure to a static pressure).36). Since physically dP can increase only in a finite amount it must that dA = 0. ISENTROPIC FLOW one dP and dA have the same sign. Hence the pressure increases when the velocity decreases and vice versa. For the special case of M = 1 (sonic flow) the value of the term 1 − M 2 = 0 thus mathematically dP → ∞ or dA = 0. dU = − dP PU (4.35 . In that case.28) by solving it for dU . dA > 0 ⇒ dP < 0 dA < 0 ⇒ dP > 0 This behavior is opposite to incompressible flow behavior.4.36). ρ. not necessarily means that when dA = 0 that M = 1.36) It can be noted that in the derivations of the above equations (4. increases with pressure and vice versa (see equation 4. The second law (isentropic relationship) dictates that ds = 0 and from thermodynamics dP dT −R ds = 0 = Cp T P . 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. it is possible that dM = 0 thus the diverging side is in the subsonic branch and the flow isn’t choked. For M > 1 the term 1/1 − M 2 is negative and change the character of the equation. From the speed of sound. However.35) it is obvious that dU has an opposite sign to dP (since the term P U is positive).

33). ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 49 and for perfect gas dT k − 1 dP = (4.37) and becomes (k − 1)M dM (k − 1) dP =− 1 2 k P 1 + k− 2 M Combining equation (4. second law (4. mass (4.38) and simplifying equation (4.41) The following identify.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. 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. The relationship between temperature and the cross section area can be obtained by utilizing the relationship between the pressure and temperature (4.42) M2 − 1 dA = dM 1 2 A M 1 + k− 2 M 1 The (4.39). why? . state (4.39) is combined with equation (4.40) with equation (4. Note. equation (4.37). First stage equation (4.42) changes equation (4.37) T k P Thus.33) isn’t the solution but demonstration of certain properties on the pressure.26)1 .40) P =− M dM 1 2 1 + k− 2 M (4. the temperature varies according to the same way that pressure does.41) into 2 P P (4.33) yields 1 k ρU 2 dA A 1−M 2 (4.2.39) (4.4. 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. Differentiation of equation (4.29).37) and the relationship of pressure and cross section area (4.43) momentum equation is not used normally in isentropic process.9). 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.

the factors ros 1 2 . ISENTROPIC FLOW Equation (4. There¡£¢ fore. A (area) with the relative velocity (Mach s number).50 CHAPTER 4. the only factor that affects relation¡¥¤§¦©¨ . In equation (4.43) is very important because it relates the geometry M.43). ction A M and A are positive reM 1 + k− 2 sec gardless of the values of M or A.

Hence the stagnation temperature can be written T0 = constant and P0 = constant and both of them are known (the condition at the reservoir). velocity. calculate the Mach number. on the other hand. S OLUTION The stagnation conditions at the reservoir will be maintained throughout the tube because the process is isentropic. 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. For M < 1 the Mach number is varied opposite to M. It was measured that air mass flow rate is 1[kg/sec]. 4. ship between the cross area and the  © Mach number is M 2 − 1. In the case of M > 1 the Mach number increases with x the cross section area and vice versa.5)). 4.” Again. Assume that process is isentropic and neglect the velocity at the reservoir. an increase of the cross section decreases the velocity and Mach number (see Figure (4. Much nubmer the cross section area.4.2. the opposite conclusion that when dA = 0 implies that M = 1 is not correct because possibility of dM = 0. With the known Mach number. the Mach number can be calculated by utilizing the pressure ratio.2 Isentropic Flow Examples Example 4. an object can be moved in arbitrary speed. . For the point where the static pressure is known.1: Air is allowed to flow from a reservoir with temperature of 21◦ C and with pressure of 5[MPa] through a tube.5: The relationship between the cross requires that dA = 0. At some point on the tube static pressure was measured to be 3[MPa]. 2 This condition does not impose any restrictions for external flow. indexsonic trasition This minimum area is referred to as “throat. The special case is when M = 1 which Fig. and the cross section area at that point where the static pressure was measured. a decrease of the cross section increases the velocity and the Mach number. Assume that the ratio of specific heat is k = Cp /Cv = 1. from the mathematical point of view: on one hand. In external flow. In subsonic flow branch.

1416 The velocity at that point is c U =M √ The tube area can be obtained from the mass conservation as A= m ˙ = 8. there is no known way for the author to directly measure the Mach number. Assume that the specific heat ratio k = 1.88638317 × √ 1.2) or from Figure (4.2: The Mach number at point A on tube is measured to be M = 23 and the static pressure is 2[Bar]4 .69428 1.5[Bar].3) or utilizing the enclosed program from PottoGDC. 4 This . Also. Here it is subsonic and this technique is not suitable. pressure is about two atmospheres with temperature of 250[K ] this question is for academic purposes. or simply using the equations shows that M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0. T = 0. estimate the temperature at point B. The best approximation is by using inserted cone for supersonic flow and measure the oblique shock. and velocity can be calculated. Finally.86420338 × (273 + 21) = 254.0115 0.69428839 × J ρ0 RT0 × 294[K ] 287. Example 4.4 and assume a perfect gas model.44) For a circular tube the diameter is about 1[cm].4.26 × 10−5 [m3 ] ρU kRT = 0.076K ρ0 ρ= ρ P0 5 × 106 [P a] = 0.2.60000 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 51 the temperature. In the point where the static pressure known ¯ = P = 3[M P a] = 0.6 P P0 5[M P a] From Table (4.0 kgK kg m3 = 41. 4 Well.53105 With these values the static temperature and the density can be calculated.86420 0. Calculate the Mach number at point B under the isentropic flow assumption. the cross section can be calculated with all these information.60693 0.88639 0. Downstream at point B the pressure was measured to be 1.4 × 287 × 294 = 304[m/sec] (4.

6265306 and second subsonic M = 0.5 The corresponding Mach number for this pressure ratio is 1.12780453 × = 0. Therefore. Assume that the flow is isentropic and the gas specific heat ratio is 1. 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.272112 B With the value of A A∗ from the Table (4.23005 1.53884934. At point “A” the cross section area is 50 [cm2 ] and the Mach number was measured to be 0.0000 T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.12780 0.55556 0. Both solution are possible and acceptable. Find the Mach number at point B.21567 0. At M = 2 (supersonic flow) the ratios are M 2.59309 With this information the pressure at point B can be expressed as from the table 4.. The supersonic branch solution is possible only if there where a transition at throat where M=1.2) or from Potto-GDC two solutions can be obtained.0 = 0.17040879.2 1.4.60315132 P P0 = 0.6875 0. The stagnation temperature can be “bypassed” to calculate the temperature at point B M =2 M =1. the stagnation pressure at point A is known and stagnation temperature can be calculated.42[K ] Example 4.8137788 and TB = B 0. .2 @ M = 2 PA = P0 PB P0 × PA 2. At point B in the duct the cross section area is 40 [cm2 ].55555556 271.17040604 PB 1.81.59014 = 1. This relationship can be obtained by AB AB AA 40 = × ∗ = × A∗ AA A 50 from the Table 4. S OLUTION To obtain the Mach number at point B by finding the ratio of the area to the critical area. The two possible solutions: the first supersonic M = 1.3: Gas flows through a converging–diverging duct.4.52 CHAPTER 4. TB = T A × T0 × TA TB T0 = 250[K ] × 1 × 0.60315132 0.

e.2721 0.49) Equation (4.2.4.45) This parameter is studied here.49) relates the Mach number at any point to the cross section area ratio.47) It can be noted that equation (4.82071 1. It is convenient to rearrange the equation (4.47) obtained the form √ k+1 − 2( k−1) m ˙ kP0 k−1 √ (4.3 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 0.65396 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 53 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 1.94511 0.2721 0.47) holds everywhere in the converging-diverging duct and this statement also true for the throat.46) results in m ˙ = A kM P0 √ kRT0 1+ k−1 2 M 2 k+1 − 2( k−1) (4.47) yields 1 A = ∗ A M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (4.22617 0. The area ratio is defined as the ratio of the cross section at any point to the throat area (the narrow area). . The area ratio as a function of the Mach number needed to be established.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. specifically and explicitly the relationship for the chocked flow. Hence equation (4. to examine the maximum flow rate and to see what is the effect of the compressibility on the flow rate. pressure) do not change.34585 1. temperature.28772 0.2. 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.48) by equation (4. dividing equations (4.48) = 1 + ∗ A 2 RT0 Since the mass flow rate is constant in the duct.0440 4.53887 0. The throat area can be denoted as by A∗ .45) to be expressed in terms of the stagnation properties as f (M.46) Expressing the temperature in terms of Mach number in equation (4.86838 1.

68473 P √0 ∼ √ T0 R (4.4 m ˙ A∗ 0. It can be noticed that Fliengner’s equation can lead to definition of the Fliengner’s Number.53) (4.5)).55) and the maximum point for F n at M = 1 is Fn = k k+1 2 k+1 − 2( k−1) (4.54) Utilizing Fliengner’s number definition and substituting it into equation (4.52) Equation (4.51) max The maximum flow rate for air (R = 287j/kgK ) becomes. k = 1.52) is known as Fliegner’s Formula on the name of one of the first engineers who observed experimentally the choking phenomenon. √ m ˙ T0 = 0.54 CHAPTER 4. While these .040418 A ∗ P0 (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.48) in with respect to M and equating to zero. Shapiro (problem 4.48) results in F n = kM k−1 2 M 1+ 2 k+1 − 2( k−1) (4. (see for example. 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 (4.50) max For specific heat ratio. Carrying this calculation results at M = 1. which have mostly academic proposes. m ˙ A∗ P √0 = T0 k R k+1 2 k+1 − 2( k−1) (4.

ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 55 problems have a limit applicability in reality.58) transformed it into ρ2 = k−1 Pρ + T0 R 2kRT0 m ˙ A 2 (4. e.9) and substituting for Mach number M = m/Aρc ˙ results in k−1 T0 =1+ T 2 Rearranging equation (4. .59) Equation (4. the terms 0 for the group over the under–brace approaches zero when the flow rate (Mach number) is very small.61) version 0.57) 1/kR T0 ρ 2 = T ρ ρ + T c2 k−1 2 m ˙ A 2 (4. ρ when all other variables are known. The situation where the mass flow rate per area given with one of the stagnation properties and one of the static properties.59) is quadratic equation for density. For this kind of problems a new Table was constructed and present here5 .g. In other words. P0 and T or T0 and P present difficulty for the calculations.2.60) 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 ). It is convenient to change it into ρ2 − k−1 Pρ − T0 R 2kRT0 m ˙ A 2 =0 (4.4. The case of T0 and P This case considered to be simplest case and will first presented here.58) And further Rearranging equation (4.44 of this book. they have substantial academic value and therefore presented here. Using energy equation (4.57) result in p R m ˙ Aρc 2 (4. The use of the regular isentropic Table is not possible because there isn’t variable represent this kind problems. 5 Since    (4.

First. k also).62) The dimensionless density now is related to a dimensionless group that is a function of Fn number and Mach number only! Thus.68) . this dimensionless group is function of Mach number only. also the specific heat. Thus.56 CHAPTER 4. it is shown that the dimensionless group is a function of Mach number only (well. RT P0 2 It can be noticed that F n2 = k T T0 P0 P 2 Hence. And the values of A ∗P 0 in Table (4.1).61) is transformed into   2 ˙ 1 (k − 1)RT0 m  1+ 1+2 ρ ˆ= 2 kP 2 A (4.66) is only function of Mach AP were tabulated number (well. again the specific heat ratio. k ).66) m ˙ A 2 = F n2 k A ∗ P0 AP 2 T T0 P0 P 2 (4. the dimensionless density is  1 (k − 1)F n2 ρ ˆ = 1 + 1 + 2 2 k2 A∗ P AP 0 2   (4. m ˙ A 2 = 1 c0 2 k P0 2 m ˙ A∗ 2 A∗ A 2 P0 P 2 (4. ISENTROPIC FLOW It is convenient to denote a new dimensionless density as ρ ˆ= ρ p RT0 = 1 ρRT0 = ¯ P T (4.63) =f (M ) RT0 P2 Thus. the problems is reduced to finding tabulated values.67) (4.65) Again notice that the right hand side of equation (4. The case of P0 and T A similar problem can be described for the case of stagnation pressure. T .64) RT0 P2 m ˙ A 2 = F n2 k A ∗ P0 AP 2 (4.2) and Fn is tabulated in the next Table (4. and static temperature. P0 . F n2 A ∗ P0 AP With this new definition equation (4.

For example the first “naughty professor” question is sufficient that process is adiabatic only (T0 . mass flow rate per area. The heat/temperature part is valid for enough adiabatic condition while the pressure condition requires also isentropic process. if it was shown that a group of parameters depends only Mach number than the Mach is determined by this group.2). The actual solution of the equation is left as exercise (this example under construction).67) became RT P0 2 m ˙ A 2 = A ∗ P0 AP 2 (4.70) 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. Based on the same arguments. the dimensionless approach is used later analytical method is discussed (under construction). Unfortunately. .71) It was hidden in the derivations/explanations of the above analysis didn’t explicitly state under what conditions these analysis is correct.4.69) The right hand side is tabulated in the “regular” isentropic Table such (4. not all the analysis valid for the same conditions and is as the regular “isentropic” Table.).2). First. (4. 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. All the above conditions/situations require to have the perfect gas model as the equation of state. P . 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.2. 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. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 57 Thus equation (4. This example shows how a dimensional analysis is used to solve a problems without actually solving any equations.

27926 0.58 CHAPTER 4.52690 1.30214 0.000 0.46677 0.78965 1.010476 0.060404 0.33465 0.15963 0.044110 0.10752 0.74912 1.002 0.001 0.81034 1.18896 0.36550 0.26264 0.83132 1.10397 0.91838 1.24674 0.39701 0.88588 0.56172 1.17397 0.031 0.17381 0.021 0.63386 1.65857 0.11294 0.13284 0.71967 0.18428 0.47000 0.000 0.13796 0.003 0.37896 1.49249 0.25000 0.20316 0.31480 0.42000 0.11710 0.23777 0.0 2.62436 0.43919 0.22085 0.62915 0.035 0.009 0.53000 0.31203 0.13232 0.18709 0.33226 0.050001 0.25018 0.69036 1.002 0.49485 0.35000 0.41855 0.65246 1.42683 1.30185 1.012593 0.005 0.16522 0.00747 0.30000 0.074314 0.28000 0.079722 0.51932 0.094654 0.55637 0.12724 0.074254 0.37474 0.087372 0.29247 0.14592 0.065 0.57656 0.54000 0.400E −06 1.14889 0.055 0.33378 0.49305 0.78382 0.46000 0.067111 0.16121 0.20442 0.54422 1.070106 1.66098 0.52858 0.57253 0.92149 0.54733 0.38884 0.023 0.26495 0.077 0.12039 0.49000 0.22634 0.046 0.34000 0.46798 0.081847 0.73995 0.61550 1.51882 0.50000 0.62E −05 0.006 0.47609 1.014197 0.029920 0.70969 1.24000 0.27358 0.36764 0.57709 0.45951 1.003 0.69948 0.67129 1.37432 0.34775 1.012 0.24773 0.008 0.44215 0.065654 0.52000 0.065557 0.76924 1.35361 0.026 0.00707 0.52485 0.00351 0.68875 0.44309 1.003 0.028 0.028651 0.85107 0.51000 0.20109 0.23000 0.017813 0.14276 0.10639 0.81706 0.38000 0.37000 0.020986 0.87421 1.098460 0.038365 0.00865 0.024585 0.083989 0.11928 0.057647 0.48000 0.33233 1.060 0.36329 1.10000 0.050518 0.58952 0.14084 1.29000 0.48360 0.45000 0.14927 0.28677 1.072487 0.059212 0.004 0.15372 0.042 0.34330 0.44192 0.31000 0.21000 0.95791 0.21703 0.26000 0.46633 0.28307 0.55000 0.23155 0.33000 0.44363 0.050 0.001 0.0 0.22000 0.40000 0.30418 0.41997 0. ISENTROPIC FLOW Table 4.017 0.49285 1.007 0.1: 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 0.033229 0.16581 0.00E +00 1.54531 0.40333 0.60706 0.57944 1.41338 0.50978 1.63889 .59736 1.21584 0.12239 0.095449 0.89613 1.089910 0.43000 0.000 0.005 0.001 0.75136 0.019 0.00352 0.85261 1.39000 0.013 0.014268 0.011 0.36000 0.17728 0.20458 0.18992 0.015 0.35316 0.41073 1.071 0.72927 1.23137 0.0 0.60047 0.32000 0.31703 1.39596 0.32220 0.20000 0.29663 0.27000 0.13342 0.39478 1.0 0.27608 0.000424 0.038 0.015027 0.25535 0.41000 0.44000 0.

135 1.427 1.87000 0.326 1.791 1.92000 0.383 1.194 1.535 2.56000 0.287 3.571 4.156 2.477 1.241 2.037 1.74000 0.771 1.244 3.187 1.977 2.4.60000 0.088 1.225 1.67000 0.20 0.407 2.869 7.859 1.083 1.62936 0.82000 0.235 5.424 6. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 59 Table 4.642 1.963 3.148 1.70675 0.758 1.018 2.343 1.105 1.998 3.65000 0.385 1.805 0.94000 Fn 0.553 1.70000 0.216 2.84000 0.77000 0.96389 0.846 8.953 2.168 2.006 6.340 2.297 1.660 1.869 1.949 9.646 3.81000 0.671 1.096 2.595 2.038 2.538 1.214 1.389 3.94848 0.819 1.422 1.368 1.78062 0.68000 0.260 1.937 1.131 1.362 1.75000 0.258 4.072 1.708 1.212 1.279 1.95665 1.462 3.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.414 1.541 1.461 1.574 1.78250 0.508 1.657 1.185 1.211 1.244 5.670 2.882 2.81139 0.122 1.382 1.82722 0.043 4.99514 1.278 1.404 2.69000 0.94096 0.90000 0.640 1.72000 0.68565 0.320 1.278 2.897 5.255 1.87424 0.810 2.61000 0.415 1.202 1.972 5.790 2.821 1.500 1.035 1.931 2.736 1.541 3.661 1.173 1.105 0.161 1.323 1.412 2.610 1.67210 0.720 2.587 1.162 1.814 1.059 2.88142 0.79000 0.011 1.738 1.2.971 3.044 1.469 2.607 1.699 3.59000 0.489 2.903 1.266 4.62000 0.526 1.121 3.840 4.73000 0.92366 0.667 1.084 1.124 1.83000 0.147 1.500 1.090 1.979 4.198 1.241 1.030 1.058 2.109 1.761 1.516 1.709 3.554 10.484 4.78000 0.240 1.721 4.269 1.991 2.740 2.349 1.094 1.033 1.029 3.853 1.236 1.536 2.865 .784 1.88000 0.86101 0.217 1.924 1.448 1.141 1.712 1.64000 0.284 2.104 3.66000 0.58000 0.607 1.317 1.059 1.93000 0.277 2.602 2.101 2.324 1.842 2.602 2.81996 0.513 5.081 2.97562 1.671 2.097 1.63000 0.80000 0.57000 0.474 1.813 2.955 3.457 3.703 1.90382 0.152 1.91000 0.423 1.85000 0.381 8.082 2.676 1.86000 0.288 1.89000 0.405 1.144 ρ ˆ 1.980 2.446 1.98717 1.220 3.381 2.74624 0.305 1.613 6.151 1.271 1.71000 0.74290 0.99507 1.747 1.581 1.342 7.112 1.76000 0.450 1.806 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ0 2 T m ˙ 2 A 0.557 1.177 2.582 1.898 1.113 1.355 1.509 1.

48) can be equated as kP0 A∗ RT0 k−1 2 1+ M 2 k−1 − 2( k−1) = constant (4. Flow with pressure losses The expression for the mass flow rate (4.324 2. At some point the static pressure was measured to be 1. and the stagnation pressure.920 1.112 6.804 5.99000 1. The third academic condition is of static temperature and the static pressure.777 7. A constant flow rate requires that m ˙A = m ˙ B .881 1.97000 0. Assume that the process is isentropic and k=1.016 Example 4.5[Bar].404 4.47) is appropriate regardless the flow is isentropic or adiabatic. and heat transfer).419 3.046 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ 0 2 T m ˙ 2 A 3.95000 0.259 3.37 13. .500 3.233 2.583 10.188 2.419 ρ ˆ 1.961 2.72).4: A gas flows in the tube with mass flow rate of 1 [kg/sec] and tube cross section is ◦ 0.913 4.136 7.843 1.98000 0.371 2. Clearly. Denoting subscript A for one point and subscript B for another point mass equation (4.515 7.72) From equation (4.278 2.60 CHAPTER 4.037 4.98 6. ISENTROPIC FLOW Table 4. in a case where the flow isn’t isentropic or adiabatic the total pressure and the total temperature will change (due to friction.003 2.000 Fn 2.06 14.96000 0. The second model which there is significant heat transfer but insignificant pressure loss (Rayleigh flow like).3.181 3. 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. the velocity.338 3. There are two possible models that can be used to simplify the calculations. 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.88 11.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0. it is clear that the function f (P0 .600 4.19 14. The first model for neglected heat transfer (adiabatic) flow and in which the total temperature remained constant (Fanno flow like).001[m2 ].436 6. The temperature at Chamber supplying the pressure to tube is 27 C . Calculate for that point the Mach number.60 12.217 4. T0 . A∗ ) = constant. S OLUTION The second academic condition is when the static temperature is given with the stagnation pressure.

Assume no mass lost and adiabatic steady state flow. the cross section area is 0.27240 0. And utilizing the equality of A∗ = P 0 |A A∗ | = ∗A P 0 |B A |B A∗ A A (4. comparison of mass flow rate at point A and point B leads to P 0 A ∗ |A = P 0 A ∗ |B .5000 2.39498 1.4.2.015[m2 ] and Mach number is 1.55401 0.5. Mach number is 2. point B.5. thus the area ratios can be calculated.74) leads to A A ∗ MA A A ∗ MB P 0 |A = P 0 |B A| A A| B (4. 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 (4.32039 0.68966 0. With these information. The total pressure can be calculated because the Mach number and static pressure are known.75) the stagnation pressure at point B can be obtained. calculated the total pressure lost.01[m2 ].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 (4.6367 0.44444 0.5000 T T0 ρ ρ0 A A P P0 0.62693 A×P A∗ ×P0 F F∗ .13169 2. Downstream at exit of tube.15432 0.5: At point A of the tube the pressure is 3[Bar]. and utilizing equation (4. M 1.76) Example 4.1762 0. and the duct section area is 0. S OLUTION Both Mach numbers are known.05853 0.73) For adiabatic flow.

000 0.243 = 4.1761671 × ≈ 15.52828 0.95638 0.2: Isentropic Table k = 1.27240307 × 15.99303 0.176 1.146[Bar]).83333 0.050 0.900 1.86059 0.008 1.41238 0.97250 0.55425 0.013[Bar] Note that the large total pressure loss is much larger than the static pressure loss (Pressure point B the pressure is 0.912 1.8E + 5 11.99800 0.77640 0.257 − 15.8E +5 11.27240 5.59126 0.300 0.188 1.72093 0.340 1.93155 0.500 T T0 ρ ρ0 A A P P0 1.58170 0.590 1.63394 0.130 0.094 1.800 0.4 M 0.781 2.63535 0.01 1.38484 0.65602 0.98028 0.53974 0. the stagnation at point A is obtained from Table (4. ISENTROPIC FLOW First.52828 0.68966 1.4E +5 4.75) provides P0 |B = 51.300 1.5 A = 3 = 51.73999 0.838 2.58377 0.93284 0.47207 0.6367187 0.066 1.32039 A×P A∗ ×P0 F F∗ 2.72632 0. 0.99825 0.84045 0.015 4.68704 0.99875 0.443 1.038 1.92427 0.2) as P 0 |A = P P P0 M =2.500 0.59 5.424 1.100 0.822 2.243 = 36.53399 0.268 0.25781291 × Hence P0 |A − P0 |B = 51.88517 0.99206 0.400 0.53807 0.78896 0.91075 0.00000 0.43742 0.95238 0.57 5.53114 0.243[Bar] 2.39498 5.74738 0.52989 0.400 1.00 1.100 1.96899 0.59650 0.93947 0.89699 0.42493 0.36091 0.700 0.99502 0.009 1.54655 0.25781291[Bar] 0.0000 0.80515 0.200 1.115 1.78400 0.058527663 by utilizing equation (4.53039 0.88652 0.600 0.00000 0.46835 0.31424 0.030 1.55401 .89561 0.79158 0.882 1.68110 0.84302 0.000 1.48290 0.3 Isentropic Tables Table 4.200 0.964 2.035 1.62 CHAPTER 4.71839 0.98232 0.35036 0.52828 0.99950 0.

000631 0.78) (4.56 25.00364 0.000 6. knowing the two limits provides a tool to examine where the reality should be expected.000 7.000 5.69983 0.092593 0.500 8.039628 0.29414 0.3.439 1.70876 0.g.67320 0.20259 0.73903 0.74192 4.14924 0.31E −5 2.600 1.4 (continue) 63 M 1.25044 0.00189 0.500 4.028962 0.35714 0.072464 0.56182 0.025156 0.12195 0. So.000495 1.62693 0. the other extreme case model where the heat transfer to the gas is perfect.2 Isentropic Isothermal Flow Nozzle General Relationship In this section.12780 0.235 6.000 2.052493 0.2: Isentropic Table k=1.027224 0.000 9.033682 0.65326 0.16667 0.25699 0.14184 0.00141 0.012628 A×P A∗ ×P0 F F∗ 0.27099 0.500 3.0E +2 1.72 16. dP = dρRT Substituting equation (4.36E −5 0. The perfect gas model is again assumed (later more complex models can be assumed and constructed in a future versions).23527 0.00758 0.070595 0.000242 0.23005 0.13169 0.900 2.058528 0.4.047251 0.015504 0.637 4.057227 0.022046 0.90E −5 4. In isothermal process the perfect gas model reads P = ρRT .19802 0.555 1.800 1.60680 0.076226 0.5E +2 3.9E +2 2.000 8.77) 6 The one dimensional momentum equation for steady state is U dU/dx = −dP/dx +0(other effects ) which are neglected here.00 T T0 ρ ρ0 A A P P0 0.28986 0.56976 0.250 1.28682 0.000155 0.2E +2 5.3.87 53.74E −5 3.55556 0.71578 0.500 9.500 10.72586 0. .17404 0.013957 0.000 4.1 4.31969 0.11528 0.73723 0.21567 0.72136 0.00346 0.58072 0.15432 0.00519 0.73257 0.4E +2 1.000815 0.13 1.68830 0.081633 0.047619 0.44444 0.013111 0.00190 0.000385 0.63371 0. (e.57768 0.500 6.66138 0.73510 0.790 10.688 2.59309 0.045233 0.58549 0.77) into the momentum equation6 yields U dU + RT dP =0 P (4.011340 0.000 3.00261 0.23810 0.00659 0.000633 0.027662 0.00107 0.017321 0.500 5.4E +2 0.00 36.089018 0.000102 6. Again in reality the heat transfer is somewhere in between the two extremes.064725 0.18 75.338 1.00107 0. Eckert number is very small) is presented.017449 0.10582 0.3.019473 0. ISENTROPIC TABLES Table 4.700 1.058140 0.500 7.3E +2 4.23211 0.72953 0.35573 0.74058 0.

the velocity at point 2 becomes U2 = 2RT ln P2 − U1 2 P1 (4. U2 2 − U 1 2 P2 + RT ln =0 2 P1 (4. T 2 P1 (4.84) Or in terms of the pressure ratio equation (4.81) Or in explicit terms of the stagnation properties the velocity is U= 2RT ln P P0 (4. P2 k (M2 2 − M1 2 ) = ln 2 P1 (4.83) yields .79) to a dimensionless form becomes constant constant  (M2 2 − M1 2 ) kRT P2   ln = R  .64 CHAPTER 4.87) . ISENTROPIC FLOW Integration of equation (4.78) yields the Bernoulli’s equation for ideal gas in isothermal process which reads .83) Simplifying equation (4.84) reads k(M1 2 −M2 2 ) P2 2 = =e P1 e M1 2 e M2 2 k 2 (4.79) Thus.86) Utilizing conservation of the mass AρM = constant to yield M 2 P2 A1 = A2 M 1 P1 (4.82) Transform from equation (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. U1 ≈ 0 reads U2 = 2RT ln P2 P1 (4.80) The velocity at point 2 for stagnation point.

All these equations are plotted in Figure (4. 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.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. 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 4.4. 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. Here the critical point defined as the point where M = 1 so results can be compared to the adiabatic case and denoted by star. ISENTROPIC TABLES Combing equation (4.3 it can be observed that minimum of the curve A/A∗ isn’t on M = 1. The minimum of the curve is when area is minimum and at the point where the flow is choked.3.91) (4. The critical pressure ratio can be obtained from (4. The true critical point is when flow is choked and the relationship between two will be presented.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 (4.85) yields M1 A2 = A1 M2 65 e M2 2 e M1 2 k k−1 k 2 (4.94) . 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. at this stage.92) Finally.87) and equation (4.90) Equation (4.85) to read (1−M 2 )k ρ P = ∗ =e 2 ∗ P ρ (4. Again it has to emphasis that this critical point is not really related to physical critical point but it is arbitrary definition.93) Of course in isothermal process T = T ∗ .6).89) The critical point.

M = √ k (4.5 2 M 2. ISENTROPIC FLOW Isothermal Nozzle 4 3.6: Various ratios as a function of Mach number for isothermal Nozzle Equation (4. Utilizing equation .5 1 0. the pressure loss in adiabatic flow is milder as can be seen in Figure (4. 4.5 1 1.5 3 3.5 3 2.94) simplified to 1 kM 2 − 1 = 0 .95) It can be noticed that a similar results are obtained for adiabatic flow.96) On the other hand. 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 0 0 0.5 2 1.5 4 k=14 P/P * A/A * P0 / P0 T 0 / T0 T/T * * * Tue Apr 5 10:20:36 2005 Fig. Thus. The velocity √ k .7(a)). 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.66 CHAPTER 4.

5 1 1.5 1 0.98) Q = Cp (T02 − T02 ) .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.5 2 1.5 3 2.90) 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 = (4.97) e− 2 = 0.5 1 0. Again. the heat transfer is expressed as (4. It also has to be emphasized that the stagnation values of the isothermal model are not constant.4.5 2 M 2. as opposed to the adiabatic case.7: The comparison of nozzle flow (4.5 3 3.5 4 3.5 1.3. ISENTROPIC TABLES Isothermal Nozzle 4 3.5 2 1.60653 1 Notice that the critical pressure is independent of the specific heat ratio.5 3 2.5 0 0 1 0.5 4 A / A iso * A / A adiabatic * P / P iso * P / P adiabatic * 67 Comparison between the two models 5 4.5 0 0 0. k . 4.

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). 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.5 1.8: 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 20.7(b)). In this Fig.2 the distance from the entrance to the throat is normalized (to 0 0 1 2 0.100) By utilizing equation (4. 4.68 CHAPTER 4.100) the velocity ratio was obtained and is 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 0.99) 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.6 between the two models so in an actual situation it can be 0.4). Thus. As can be observed from Figure (4. The Mach number for the isentropic is larger for the supersonic branch but the velocity is lower. The ratio of the velocities can be expressed as √ Ms kRTs Us √ = (4. the isentropic for the subsonic branch will be over prediction. using the isentropic model results in under prediction of the actual results for the velocity in the supersonic branch. The prediction of the Mach number are similarly shown in Figure (4. .7(b)). This P / P0 isentropic profile isn’t an ideal profile 0.7(b)).5 one (1)).4 bounded. 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. In this comparison it has to be remembered that critical area for the two models are different by about 3% (for k = 1. To make sense and eliminate unnecessary details 0. While. The Mach number was computed for the two models and plotted in Figure (4.

101) While the temperature ratio of the isothermal model is constant and equal to one (1). the sought ratio is provided. Example 4.89) and the isentropic relationship. 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. To demonstrate the relativity of the approach advocated in this book consider the following example. Calculate the velocity at the exit using the adiabatic model. The throat area ratio to entrance area is 1:4 respectively. the conservation of the stagnation properties. with known area ratio and known stagnation Potto–GDC provides the following table: .4.4. Using equations (4. Assume that the back pressure is low enough to have supersonic flow without shock and k = 1.e.3. If the nozzle was made from copper (a good heat conductor) a larger heat transfer occurs.8) shows that the range between the predicted temperatures of the two models is very large. Figure (4. The pressure ratio for the isentropic model is P = P0 int 1+ 1 k−1 2 2 M k−1 k (4. Thus. The initial stagnation temperature is denoted as T0 int . 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. while the range between the predicted pressure by the two models is relatively small. 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. ISENTROPIC TABLES 69 Two other ratios need to be examined: temperature and pressure. The stagnation pressure is 5[Bar] and the stagnation temperature is 27◦ C.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 (4.103) where z is an arbitrary point on the nozzle.6: Consider a diverging–converging nozzle made out of wood (low conductive material) with exit area equal entrance area.

9402 T T0 ρ ρ0 CHAPTER 4.9). To explain the motivation for using this definition consider the calculation of the net forces that acting on section shown in Figure (4.92 × 300 = 576.99572 0. The exit velocity.36644 × 300 = 109.4 × 287 × 300.104) As was discussed before.9K .253[m/sec] (4.11915 Even for the isothermal model.777777778 11 k−1 T0 ∗ 1 + k− 1 + 2 k k Thus the stagnation temperature at the exit is T0ini = 1.36644 0.0000 0.0000 0.4940 0. The exit temperature is 0.12556 0. The Impulse function is denoted here as F .02979 3. the velocity can be calculated. The exit velocity can be determined by utilizing the following equation √ √ Uexit = M kRT = 1.50225 The exit Mach number is known and the initial temperature to the throat temperature ratio can be calculated as the following: 1 T0ini 1 = = = 0.4. the maximum velocity cannot exceed the 691.4 × 287 × 109.2K .70 M 0. but in the literature some denote this function as I .98511 0.9 ∼ 617.9402 1.6) or using the Potto–GDC obtains the following table M 1. the velocity in the copper nozzle will be larger than the velocity in the wood nozzle. 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 (4.9910 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 With the known Mach number and temperature at the exit.777777778 = 1.93[m/sec] 1. is √ √ U = M kRT = 2.4940/0. then.4 4.9405 0.08129 4. However.0 = 691.51183 4.1 The Impulse Function Impulse in Isentropic Adiabatic Nozzle One of the functions that is used in calculating the forces is the Impulse function.14655 2.921 T0exit The exit stagnation temperature is 1. ISENTROPIC FLOW A A P P0 A×P A∗ ×P0 0.9910 1.98934 0.253[m/sec] 4.0000 4. the initial stagnation temperature is given as 300K .105) . Using the area ratio in Figure (4.

107) P2 A 2 P1 A 1 Fnet = 1 + kM2 2 − 1 + kM1 2 ∗ ∗ P0 A P0 A P0 A ∗ (4.109) (4.108) shows that the right hand side is only a function of x-direction Mach number and specific heat ratio.42) results in Rearranging equation (4. M and k . THE IMPULSE FUNCTION 71 The net force is denoted here as Fnet . 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 (4.105) with equation (4.108) which makes the ratio a function of k and the Mach number. The mass conservation also can be applied to our control volume m ˙ = ρ 1 A 1 U1 = ρ 2 A 2 U2 (4.110) = 1 P∗ P0 2 ( k+1 ) k−1 k P1 A 1 1 + kM1 2 P0 A ∗ 1 (1 + k ) (4.111) This ratio is different only in a coefficient from the ratio defined in equation (4. Thus.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 (4.9: 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.112) . 4. Hence.106) and by utilizing the identity in equation (4. Defining a function that depends only Fig. the net force is Fnet k+1 = P0 A (1 + k ) 2 ∗ k k−1 F2 F1 − ∗ ∗ F F (4.4. Hence.4.108) Examining equation (4. 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.106) Combining equation (4. k .108) P1 A1 1 + kM1 F = ∗ ∗ F∗ P A (1 + k ) 2 (4.

96355 2.2227 A A A2 0. To obtain this ratio the denominator is needed to be obtained.99751 0. Utilizing Fliegner’s equation (4. stance (gas) thorough a converging nozzle S OLUTION The solution is obtained by getting the data for the Mach number.99380 5.11164 0.94934/0.003m2 entrance area is 0.003[m ].27353 0.2121 the area ratio of at point 1 can be calculated.76 M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 1 for example (4. the ratio of P1 A1 /A∗ P0 is needed to be calculated.009m2 2 A2 = 0.009 A1 = = 2.1000 0.1949 P0 P1 = 5.7: 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 = 0.1 A ∗ P0 70061.776[Bar] P2 P0 .0 × 400 × 287 A ∗ P0 = = ∼ 70061. A2 A1 0.058 0.94934 2.7) 0. ISENTROPIC FLOW To demonstrate the usefulness of the this function consider a simple situation of the flow through a converging nozzle Example 4.2121 With the area ratio of A A 0.003 A 2 P2 = ∼ 2.52).98526 0.003 And utilizing again Potto-GDC provides M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.009[m ] and T0 = 400K 2 P2 = 50[Bar] the exit area is 0. provides the following √ √ m ˙ RT 1.76[N ] 0.Fig.99380 ∼ 4.10: Schematic of a flow of a compressible subzle and pressure at point 1.0times0.058 and 500000 × 0. To obtained the Mach number.1774 2.72 CHAPTER 4.99132 5.2227 The pressure at point 1 is P1 = P 2 0.2121 × = 5.96666 = 2. 4. The stagnation temperature is 400K and the pressure at point 2 was measured as 5[Bar] Calculate the net force acting on the noz.

0E + 5 9.010 2.4.4.2 The Impulse Function in Isothermal Nozzle Previously Impulse function was developed in the isentropic adiabatic flow.52828 0. the definition of the Impulse function is reused.4 × 1.00 4.114) Since U 2 /RT = kM 2 and the ratio of equation (4.5.2E +5 20.000 1. 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 (4.23. The same is done here for the isothermal nozzle flow model.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 (4.116) (4.064 5. ISOTHERMAL TABLE The net force is obtained by utilizing equation (4.014 2.949 5.5 Isothermal Table Table 4.1 T0 T0 P0 P0 A A P P 0.1 k 73 4.115) 4.0E +6 4.1949 − 0.52921 0.112) Fnet = P2 A2 P0 A ∗ k + 1 k−1 F2 F1 (1 + k ) − ∗ P2 A 2 2 F∗ F 1 × 2.3: Isothermal Table M 0.00 8.00 0.362 10.064 1.064 1.96666) ∼ 614[kN ] = 500000 × 2.53199 1. As previously.001 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 (4.05 0.5 × (2.225 A×P A∗ ×P0 F F∗ .87) transformed equation into (4.

3 0.079 1.464 1.75E −5 1.209 0.55556 0.50618 4.62500 0.75344 1.863 0.111 1.62665 0.40000 0.939 2.26634 0.565 1.1E + 3 0.10 1.500 5.0 5.667 1.50000 0.90 2.80528 0.719 5.134 9.0 0.0 0.287 1.281 1.73492 0.12246 0.0 0.000 1.000 3.28571 0.194 1.41E −6 0.6 0.055 1.97376 1.50 1.97274 1.86329 0.000 0.4E + 3 2.80 1.0 5.22222 0.20846 0.0 0.5 0.065 1.20 1.800 1.6E + 5 1.00 1.161 2.41 40.3: Isothermal Table (continue) M 0.100000 A×P A∗ ×P0 F F∗ 2.500 4.58985 0.14286 0.3E + 9 0.071758 90.608 3.237 1.90909 0.1E + 4 1.000 1.67383 0.52632 0.247 0.2 0.63276 2.66667 0.004 0.4E +26 0.58824 0.690 1.15385 0.76923 0.12500 0.891 1.564 1.94147 1.000 4.3E +11 0.4E + 3 5.429 1.4E + 8 0.33333 0.70 1.114 1.0 1.0 0.5E + 2 2.61693 0.183 1.71429 0.569 3.020 1.83333 0.29 80.333 2.500 7.13333 0.015317 7.85853 1.0 1.021 0.90302 1.74 CHAPTER 4.128 1.11111 0.142 1.134 0.000 9.059 1.281 1.000 3.73278 0.500 6.540 4.0 5.000380 2.97156 1.00 2.60 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- .56954 3.98750 0.8E + 2 4.047 1.3E + 2 1.245 2.763 1.026 19.500 2.78 0.21 1.625 1.586 5.375 1.500 10.064 2.5E + 2 0.18182 0.33554 0.062 1.875 4.047 0.7E + 3 1.000 0.40 1.718 3.889 2.41686 0.6E +13 0.399 0.22881 15.30 1.438 2.044 1.000 6.856 4.681 1.429 1.99967 1.00221 9.8 0.599 0.0E + 6 0.56232 0.0 7.11765 0.000215 1.9E + 2 8.284 3.0 1.00 T0 T0 P0 P0 A A P P 0.125 1.8E +18 0.69449 2.036 0.000 1.14 0.54322 0.4E +20 0.000 3.98796 1.200 1.4 0.296 5.0 0.053 0.0 5.3E +23 0.500 8.000 7.7 0.0 0.10526 0.143 4.035 3.083 0.025349 0.25000 0.431 4.000 0.16090 0.000 1.007 5.16667 0.0 0.9 1.2E +29 1.000 8.0 1.2E + 4 1.2E + 3 3. ISENTROPIC FLOW Table 4.958 1.00370 0.500 9.50 3.250 1.89348 1.20000 0.271 0.0 0.328 1.553 1.41E −5 4.665 0.2E + 3 7.0 8.6E + 4 2.153 1.389 1.0 0.000 5.0 2.3E +15 0.063 1.51069 0.80844 1.275 1.

Thus.120) P Equations (4. To carry the integration of the right hand side the relationship between the pressure and the density has to be defined. P ) and full differential is dh = ∂h ∂T dT + P ∂h ∂P dP T ∂h ∂T (4. a real gas equation can be expressed in equation (3.19) are combined to form Cp dT ds = −z 1+ R R T T Z ∂z ∂T dP P (4. THE EFFECTS OF REAL GASES 75 erties like the critical pressure and critical temperature.123) results U2 = 2 P1 P0 dP = ρ P P0 1 ρ0 P0 P 1 n dP (4.27) and (3. ∂h ∂P =v−T ∂s ∂T T P First.19) and dividing by equation (3.27) is reused to related the entropy change to the change in thermodynamics properties and applied on non-ideal gas.19) yields dz dρ dT dP = + + P z ρ T (4. h = h(T.122) At the stagnation the definition requires that the velocity is zero.4. Differentiating equation (3.121) P The mechanical energy equation can be expressed as d U2 2 =− dP ρ (4.118) and second derivative is (4. The enthalpy is a function of the temperature and pressure thus. The following power relationship is assumed ρ = ρ0 P P0 1 n (4.119) The definition of pressure specific heat is Cp ≡ Maxwell relation hence.19).122) when using relationship which is defined in equation (4.124) . Gibb’s equation (4. Since ds = 0 and utilizing the equation of the state dh = dP/ρ.123) Notice. that for perfect gas the n is substituted by k . the differential of enthalpy is calculated for real gas equation of state as dh = Cp dT − T Z ∂z ∂T dP ρ (4. With integration of equation (4.117) Again.6.

Thus.121) when ds = 0 results T2 T1 Cp dT = R T P2 z 1+ P1 T Z ∂z ∂T P dP P (4.132) .127) It must be noted that n is a function of the critical temperature and critical pressure.129) For the case of n = 1 ρ∗ U ∗∗ 1 n m ˙ = A∗ P0 z0 RT0 P P0 2z0 RT0 ln P0 P (4.125) For n > 1 the integration results in 2n 1− n−1 n−1 P ( n ) P0 U= z0 RT0 (4.128) Where ρ∗ is the density at the throat (assuming the chocking condition) and A∗ is the cross area of the throat.76 CHAPTER 4.34) to defined the Mach number as M=√ U znRT (4.131) Integrating equation (4. 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 (4.19) results U2 = 2 P P0 z0 RT0 P0 P0 P 1 n dP (4.130) The Mach number can be obtained by utilizing equation (3. 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. ISENTROPIC FLOW Substituting relation for stagnation density (3.

133) ρ0 d ρ P P0 (4.137) 2n n−1 2z0 RT0 ln P0 P (4.138) 1− P∗ P0 (4.139) Fliegner’s number for n = 1 is Fn = mc ˙ 0 =2 A ∗ P0 P∗ P0 2 − ln P∗ P0 (4. 2n n−1 P P0 n−1 n U= When n = 1 or when n → 1 z0 RT0 1− (4.140) .135) U= 2z0 RT0 ln P0 P (4.33) is applicable for some ranges of relative temperature and pressure (relative to critical temperature and pressure and not the stagnation conditions).132) looks at Bernnolli’s equation which is dU 2 =− 2 After integration of the velocity dU 2 =− 2 P/P0 1 dP ρ (4.134) It was shown in Chapter (3) that (3.6.136) 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 (4.4. THE EFFECTS OF REAL GASES 77 To carryout the integration of equation (4.

78 The critical ratio of the pressure is P∗ = P0 2 n+1 CHAPTER 4.145) For n = 1 the Mach number is M= 2 z 0 T0 P 0 ln z T P (4.132) can be integrated T0 = T P0 P R Cp ∂z )P ] [z+T ( ∂T (4. equation (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 1.144) 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 (4.148) ∂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 ] (4.142) To obtain the relationship between the temperature and pressure. Note (4.141) When n = 1 or more generally when n → 1 this is a ratio approach P∗ √ = e P0 (4. ISENTROPIC FLOW n n−1 (4.149) .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.

. Calculate the area ratio between the point and the throat.61 P = ∼ 52.82K T 0.04943 T0 300 T = ∼ 713.405.61 the results are M 2.152) Example 4. S OLUTION 1. ii. Calculate the stagnation pressure and the stagnation temperature. Are the stagnation pressure and temperature at the entrance different from the point? You can assume that k = 1. Of course.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 (4.42027 0.14366 2.151) 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 (4.04943 0. the stagnation pressure is constant for isentropic flow. The solution is simplified by using Potto-GDC for M = 2.11761 2.6. The stagnation pressure is obtained from P0 = P0 2.61. iii.4. and temperature 300K .42027 The stagnation temperature is T0 = 3.802[Bar] P 0. the pressure 2[Bar].8: A design is required that at a specific point the Mach number should be M = 2. i.6100 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.9066 0. THE EFFECTS OF REAL GASES and for n = 1 T∗ = T0 79 e−[z+T ( ∂z ∂T )P ] (4.

80 CHAPTER 4. ISENTROPIC FLOW .

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

The only tool that brings us to the direction of the flow is the second law of thermodynamics. Thus.3). it can be written that The process is adiabatic. the gas flows from a supersonic flow to a subsonic flow. the entropy must increase. In mathematical terms. The transition in the shock is abrupt and in a very narrow width. then there are four unknown conditions downstream. Therefore. 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. Physics dictates that there is only one possible solution. and therefore the energy equation can be written as C p Tx + Ux 2 Uy 2 = C p Ty + 2 2 (5. . the momentum is the quantity that remains constant because there are no external forces. and the shock occurs at a very narrow section. The conditions to the right of the shock wave are uniform.1) In a shock wave. The chemical reactions (even condensation) are neglected.82 CHAPTER 5. 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). This law dictates the direction of the flow. It is customary in this field to denote x as the upstream condition and y as the downstream condition. Mathematically. or nearly adiabatic. A system of four unknowns and four equations is solvable. and as it will be shown. the second law is expressed by the entropy.2) The equation of state for perfect gas reads P = ρRT (5.3) Px − P y = ρ x U y 2 − ρ y U x 2 (5. Clearly. the increase of the entropy is fundamental to the phenomenon and the understanding of it. the isentropic transition assumption is not appropriate in this case because the shock wave is a discontinued area. One cannot deduce the direction of the flow from the pressure on both sides of the shock wave. it can be written as follows: sy − s x > 0 (5. For the adiabatic process. Nevertheless.4) If the conditions upstream are known. one can note that there are two solutions because of the quadratic of equation (5. NORMAL SHOCK assumed to be uniform2 . These two possible solutions refer to the direction of the flow.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. but different from the left side.

i. producing various resistance to the flow. the degree of freedom is now energy. continuity. It is also worth noting that the temperature at M = 1 on Rayleigh flow is larger than that on the Fanno line. point y . which are needed to be solved. Since the shock has $&%(' no heat transfer (a special EGFI J H K subsonic flow case of Rayleigh flow) and supersonic .38:95<68 tions are obtained. This situation is similar to a frictionless flow with the addition of heat. and state equations. The solution of every set of these equations produces one additional degree of freedom.2). U.¥-/. continuity. one can solve only four (4) equations (including the second law). s. and state equations are solved for the arbitrary value of the Ty . and ρ. T. the increase of the entropy determines the direction of s the flow. On the other hand.546flow there isn’t essentially any T =?>A@ momentum transfer (a speshock jump cial case of Fanno flow). and state equations. one can have a whole range of solutions. The reason is that the process is irreversible. and therefore no equality can exist. For ideal gas.83 Note that the greater–equal signs were not used. Now one can view these equations as two different subsets of equations. This case is called Fanno flow.0. can be expressed as a function of the other parameters. and the second set is the momentum. Thus..213-/.5) is ln Ty Py − (k − 1) >0 Tx Px (5.6) It can also be noticed that entropy. In the first case. This flow is dealt with in greater detail in Chapter (10). The mathematical explanation is given Chapter (9) in greater detail. If the energy. continuity. The entropy increases from point x to Fig.e. B?CAD the intersection of these two curves is what really Rayleigh Fanno line line happened in the shock. a parabola in the T –s diagram will be obtained. and this flow is known as Rayleigh flow. Instead of solving all the equations that were presented. when the momentum equation is solved instead of the energy equation. In Figure (5. . which will produce a range of possible solutions. the parameters are P. equation (5. Mathematically. the energy amount “added” to the shock. 5. the intersec)&*(+ tion is shown and two solu7¥8:95. which will require additional parameters. the energy equation is used. and Chapter (9) deals extensively with this topic. The first set is the energy. Clearly.2: The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave.

11) . The only issue left to derive is the relationship between Mx and My . From the continuity equation (5. c2 = kRT to be used for the Mach number definition. (5.2 Formal Model Equations (5. provides a way to obtain the speed of sound definition/equation for perfect gas. T0 x = T0 y . at least.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.1 5.8) Multiplying the two sides by the ratio of the specific heat. ρ.7) All the other relationships are essentially derived from this equation.1).2). 5. and thus for known Mx all the other quantities can be determined. Note that the Mach number is a function of temperature.1) substituting for density. It can also be noted that in many cases the dimensionless equations set is more easily solved. the equation of state yields Px Py Ux = Uy RTx RTy Squaring equation (5. NORMAL SHOCK 5.1 Solution of the Governing Equations Informal Model Accepting the fact that the shock is adiabatic or nearly adiabatic requires that total energy is conserved. Utilizing the definition of Mach number results in Py 2 Px 2 Mx 2 = My 2 Tx Ty (5.84 CHAPTER 5. The analytical solution is discussed in the next section.1. k. The relationship between the temperature and the stagnation temperature provides the relationship of the temperature for both sides of the shock.9) (5. as follows: Px 2 Py 2 Ux 2 = Uy 2 Tx kRTx Ty kRTy cx 2 cy 2 (5.1. and (5.8) results in Py 2 Px 2 2 U = Uy 2 x R 2 Tx 2 R 2 Ty 2 (5. numerically. Ty = Tx Ty T0 y Tx T0 x = 1+ 1+ 2 k −1 2 Mx 2 k−1 2 My (5.3) can be converted into a dimensionless form.

This relationship can be obtained from the combination of mass.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. T0 y = T0 x .13) (energy) and equation (5.2) into Px + kPx Mx 2 = Py + kPy My 2 Rearranging equation (5.18) .14) The pressure ratio in equation (5.5. ρU 2 is identical to kP M 2 because M2 ρ ρU 2 = Using the identity (5.15) 2  P   U  kRT = kP M 2  RT kRT  c2   (5.12) Energy equation (5. Under the perfect gas model. and energy equations.16) (5.13) means that the stagnation temperature is the same.16) can be interpreted as the loss of the static pressure.12) (mass) the temperature ratio can be eliminated.11) results in Ty = Tx Py Px 2 85 My Mx 2 (5. momentum.1.14) transforms the momentum equation (5. The loss of the total pressure ratio can be expressed by utilizing the relationship between the pressure and total pressure (see equation (4.13) It can also be observed that equation (5. Py M y Px M x 2 = 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5.15) yields Py 1 + kMx 2 = Px 1 + kMy 2 (5. SOLUTION OF THE GOVERNING EQUATIONS Rearranging equation (5. From equation (5.17) The relationship between Mx and My is needed to be solved from the above set of equations.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.

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 (5.21) Equation (5. for example. It is also possible to cross–multiply equation (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 (5.19) and divide it by My 2 − My 2 1+ k−1 My 2 + My 2 − kMy 2 My 2 = 0 2 (5. 2nd ed.22) The first solution (5.24) 3 Ireland. Utilizing equation (5. M. 1990. Mx .21) becomes My 2 = Mx 2 + 2k k−1 Mx 2 k−1 2 −1 (5. Thus.19) Equation (5.19) as a polynomial (fourth order). the pressure and the temperature from both sides of the nonexistent shock are the same. The solution can be obtained by rewriting equation (5. K. i. in this case.” Ch. According to the Gauss Biquadratic Reciprocity Theorem this kind of equation has a real solution in a certain range3 which will be discussed later.16) and equation (5. The second solution is where the shock wave occurs. 9 in A Classical Introduction to Modern Number Theory.19) is biquadratic.e.18) with equation (5. 108-137. Px = Py . New York: Springer-Verlag.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. Tx = Ty . pp.20) It can be observed that equation (5. one solution is My = M x (5. ”Cubic and Biquadratic Reciprocity. The pressure ratio between the two sides can now be as a function of only a single Mach number. Clearly. NORMAL SHOCK Combining the results of (5.86 CHAPTER 5.20) is the trivial solution in which the two sides are identical and no shock wave occurs.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. and Rosen. .

5. SOLUTION OF THE GOVERNING EQUATIONS 87 The fact that the pressure ratio is a function of the upstream Mach number. Mx .3 0.1.5 0.7 0. as a function of pressure ratio.30) .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 (5.26) which is associated with the shock wave.28) Based on this definition a new Mach number can be defined M0 = U c0 (5. The maximum temperature that can be achieved is the stagnation temperature Umax = 2k RT0 k−1 (5. 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 (5.4 0.6 0. The Maximum Conditions The maximum speed of sound is when the highest temperature is achieved.27) My 1 0. The speed of sound at that velocity is √ c∗ = kRT ∗ (5. The Star Conditions The speed of sound at the critical condition can also be a good reference velocity.2 0.29) Mx Fig.8 0.9 0.5. And the temperature ratio.3: The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. is transformed into Ty = Tx Py Px k+1 k−1 Py Px k+1 Py k−1 Px + 1+ (5. provides additional way of obtaining an additional useful relationship.25) In the same way.

c U c U = ∗ = ∗M c∗ c c c M∗ = (5.1.37) (5.88 CHAPTER 5.3 Prandtl’s Condition It can be easily observed that the temperature from both sides of the shock wave is discontinuous. Therefore.35) (5.31) 5.33) and (5.33) 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 (5.34) . c2 U2 c∗ 2 c∗ 2 k + 1 ∗2 + = + = c k−1 2 k−1 2 2(k − 1) (5.34) 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. It is therefore convenient to define the star Mach number that will be independent of the specific Mach number (independent of the temperature).36) (5. NORMAL SHOCK In the same manner.35) the following can be obtained: U1 U2 = c ∗ 2 or in a dimensionless form M ∗ 1 M ∗ 2 = c∗ 2 (5. the speed of sound is different in these adjoining mediums. an additional Mach number can be defined as M∗ = U c∗ (5.32) The jump condition across the shock must satisfy the constant energy.

0 100. an example is provided.0272237 =⇒ P0x = 1. P0x is known).0 Py/Py.2 Operating Equations and Analysis In Figure (5.9360 8.0 20. the total pressure can be obtained (through an isentropic relationship in Table (4. Using these data.2m/sec 1 2 3 Fri Jun 18 15:48:25 2004 4 5 Mx 6 7 8 9 10 .0272237 = 55.32 is only a few mouse clicks away from the following table. It can be noticed that the density ratio (velocity ratio) also has a finite value regardless of the upstream Mach number.1: Air flows with a Mach number of Mx = 3.. My .4 or utilizing Potto-GDC for for value of the specific heat ratio. and the ratio of the total pressure.5 [bar] and a temperature of 0◦ C goes through a normal shock. are plotted as a function of the entrance Mach number. the velocity can readily be calculated.47642 2.0 50.0 40. P0y /P0x . ρy/ρx and Ty/Tx as a function of Mx Shock Wave relationship Py/Px ρy/ρx Ty/Tx S OLUTION Fig.0 90.7245 My 0.2).2.0 10.1) for k = 1.0 0. Tx .3088 0. Note that the My has a minimum value which depends on the specific heat ratio.30879 and k = 1.0 30.38109 To illustrate the use of the above equations. The equations should be used with the Mach number (upstream or downstream) for a given pressure ratio or density ratio (velocity ratio).1[bar] √ √ 0x cx = kRTx = 1.1110 3. at a pressure of 0. The typical situations in which these equations can be used also include the moving shocks. 5.4: The ratios of the static properties of the two sides Analysis: of the shock.4 × 287 × 273 = 331. P = 0.0 60.0 80. Finding the Mach number for a pressure ratio of 8. i. Calculate the temperature.e. The working equations were presented earlier. and velocity downstream of the shock.0 70. OPERATING EQUATIONS AND ANALYSIS 89 5. This kind of equations requires examining Table (5. the known information are Mx = 3.0 110. 120.5/0. Also with the temperature. Example 5. px = 1.5. First. pressure. total pressure.5[bar] and Tx = 273K . Ty Tx ρy ρx Py Px P0y P0 x Mx 2. The relationship that was calculated will be utilized to obtain the ratios for the downstream of the normal Px shock.3). the Mach number after the shock.

3). The limits of the pressure ratio can be obtained by looking at equation (5. Uy = 993. this technique is used mostly in obtaining analytical expressions for simplified models.47519 Ty Tx ρy ρx CHAPTER 5.23) transforms equation (5. it refers to a case where only a “small shock” occurs. This approach had a major significance and usefulness at a time when personal computers were not available.85714 = 257.5 series).16) and by utilizing the limit that was obtained in equation (5. which is up to Mx = 1.2 = 993. This technique also has an academic value and therefore will be described in the next version (0.6[m/sec] Now the velocity downstream is determined by the inverse ratio of ρy /ρx = Ux /Uy = 3.1[bar] = 18.3333 0. the downstream Mach number (see equation (5.8571 10.32834 × 55. during the writing of this version of the book. The strength of the shock wave is defined as ˆ = Py − P x = Py − 1 P Px Px By using equation (5.09[bar] 5.0000 My 0.3.40) (5.6[m/sec] P0y = P0y P0x × P0x = 0.38) This result is shown in Figure (5. 5. In this case.39) into ˆ= P 2k Mx 2 − 1 k+1 (5.6/3.38).32834 Ux = Mx × cx = 3 × 331. NORMAL SHOCK Py Px P0 y P0 x 2.85714.90 Mx 3.2 Small Perturbation Solution The small perturbation solution refers to an analytical solution where only a small change (or several small changes) occurs.1 The Limitations of the Shock Wave When the upstream Mach number becomes very large. Now.2.22)) is limited by My = 2 :  ∼0 2  1 + (k −1)Mx 2 2k k−1 >   − M1 x 2 ∼0 = k−1 2k (5.2.39) .6790 3.

24) the following is obtained: ˆ= P 2k k−1 2 k−1 ρy ρx 91 −1 −1 − ρy ρx (5.5. THE MOVING SHOCKS or by utilizing equation (5. however.3 The Moving Shocks In some situations. to convince the students that indeed the assumption of very thin shock is validated by analytical and experimental studies.3 Shock Thickness The issue of shock thickness (which will be presented in a later version) is presented here for completeness. 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.41) 5. when a valve is suddenly hji NPO  S T f d e b ` .2.3. flow L5M g direction Q5R For example. This issue has a very limited practical application for most students. the issue should be presented. the shock wave is not stationary. The shock thickness can be defined in several ways. This kind of situation []\ ^]_ arises in many industrial applications.

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

For this analysis. (refer to Figure (5.45) Ux = U s − U x (5.44) . Msx = U cx . 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. The upstream prime Mach number can be defined as Us − U x Us Mx = = − Mx = Msx − Mx (5.46) cx cx It can be noted that the additional definition was introduced for the shock upstream s Mach number.5)) the velocity measured by the observer is Where Us is the shock velocity which is moving to the right. the prime ’ denote the values of the static coordinates.e. i. 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. NORMAL SHOCK them at this stage..92 CHAPTER 5. The general case is where two shocks collide into one shock and moves upstream or downstream is the general case.42) (5. The downstream prime Mach number can be expressed as Us Us − U y = − My = Msy − My (5. equations. In cases where the shock velocity can be approximated as a constant (in the majority of cases) or as near constant. the steady state is obtained in the moving control value. Here. Note that this notation is contrary to the conventional notation found in the literature. the coordinates move with the shock. In such a case. 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. the previous analysis. to the stationary case when the coordinates are attached to the shock front. The problem can be reduced to the previously studied shock. The reason for the deviation is that this choice reduces the programing work (especially for object–oriented programing like C++). An observer moving with the shock will notice that the pressure in the shock is Px = P x Py = P y (5. and the tools developed in this chapter can be employed.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.

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 (5.50) The same can be said for the “downstream” side of the shock. The rise in the total temperature is due to the fact that a new material has entered the c.3. The relationship between the two new shock Mach numbers is cy Us Us = cx cx cy Msx = Ty Msy Tx (5. . Msy .52) − Tx 1 + k−1 Mx 2 2 −Tx Msx and according to equation (5.48) The “upstream” stagnation temperature of the fluid is T0x = Tx 1 + k−1 Mx 2 2 (5.51) leads to T0y − T0x = Us Ty k − 1 Tx k − 1 (Msx − 2Mx ) − (Msy − 2My ) cx 2 cy 2 (5.5.49) and the “upstream” prime stagnation pressure is P0x = Px k−1 Mx 2 1+ 2 k k−1 (5.53) Again. this difference in the moving shock is expected because moving material velocity (kinetic energy) is converted into internal energy. 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.v. an additional definition was introduced for the shock downstream Mach number. and is “converted” or added into the total temperature. THE MOVING SHOCKS 93 Similar to the previous case. This difference can also be viewed as a result of the unsteady state of the shock. at a very high velocity.

94 CHAPTER 5.6: Comparison between a stationary shock and a moving shock in a stationary medium in ducts.54) The “downstream” Mach number reads My = |Us | = Msy cy (5. The flow is moving to the right. the shock is moving to the left.55) Again. Equations (5. ­W®°¯ Stationary Coordinates À2ÁW›ÃÄÁWÅ:Æ Ç ³3´¶µZ· ¾›¿ ¼W½ ¸ p ¹ º¥¸#» Moving Coordinates c. The difficulty lies in the jump from the stationary coordinates to the moving coordinates.56) The prominent question in this situation is what will be the shock wave velocity for a given fluid velocity. This shock is viewed by some as a reflective shock. in a sudden and complete closing of a valve. In many such cases the steady state is established quite rapidly. The information propagates upstream in which the gas velocity is converted into temperature.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. NORMAL SHOCK 5. for example. . It turns out that it is very useful to use the dimensionless parameter Msx . Fig.53) can be transformed into simpler equations when Mx is zero and Us is a positive value. the observer (with the shock) sees the flow moving from the left to the right. The relationship between the Mach number on the two sides of the shock are tied ¡£¢j¤¥¡ ¦ •£– ˜ —# œ5 § ¨/©«ª§¬ c. and for a given specific heat ratio. 5. In the moving coordinates.42) to (5. 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. or Msy instead of the velocity because it combines the temperature and the velocity into one parameter. Ux .v. The “upstream” Mach number reads Mx = Us + U x = Msx + Mx cx (5.v. The “upstream” or the “downstream” Mach number is not known even if the pressure and the temperature downstream are given.3. The upstream is on the left of the shock. This situation arises in many cases in the industry. In such a case. the shock velocity “downstream” is Us . The sudden closing of the valve must result in a zero velocity of the gas.

The typical method is 0 0.5.54) and (5. an analytical solution can be obtained for equation (5. 5. which can easily be obtained in a Fanno flow. Note that to achieve supersonic velocity (in stationary coordinates) a diverging– converging nozzle is required. 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. From the same figure it can also be observed that a high velocity can result in a much larger velocity for the reflective shock. a Mach number close to one (1). For example. 1 Note. For very small values of the upstream Mach Thu Aug 3 18:54:21 2006 number.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).5) as shown in Figure (5. 1 − 2 (the coefficient is only approximated as 0.58) pro1 vides that Msx ∼ 1 + 2 and Msy = Fig.58) can be obtained by several numerical methods.58) but it seems utilizing numerical methods is much more simple.57) into (5. THE MOVING SHOCKS through equations (5. Only 2 one real solution is possible.57) And substituting equation (5. .48) results in f (Msx ) Mx = Tx Ty Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. Mx . Mx ∼ equation (5.1 1 Mx the “smart” guessing of M sx. is provided herein: (a) Guess Mx > 1.7).7: The moving shock Mach numbers as a 1 result of a sudden and complete stop. 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. the result is about double the sonic velocity of the reflective shock.55) by 2 95 ( My ) = Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. Here no such device is needed! Luckily and hopefully. The solution to equation (5. Sometimes this phenomenon can have a tremendous significance in industrial applications.3.

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

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 (5.3.61) 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. The “upstream” Mach number is Mx = The “downstream” Mach number is My = Us − U y = Msy − My cy (5.3 Fig. what will be the shock velocity for a given upstream Mach number? Again.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.3 (b) My = 1. the relationship .5 3 2. It has to be noted that the “upstream” and “downstream” are the reverse from the previous case.5 1.5 0 Number of Iteration 10 0 Shock in A Suddenly Open Valve k = 1 4.75 k = 1 4.5.60) Us = Msx cx (5. The same question that was prominent in the previous case appears now.3 Mx My Ty/Tx Ty/Tx 0.25 2 1. 5.5 1 1 0. My’ = 1.3 Mx My 1. My’ = 0.9: The number of iterations to achieve convergence. THE MOVING SHOCKS 97 Similar definitions of the right side and the left side of the shock Mach numbers can be utilized.75 5 10 Number of Iteration 15 20 Wed Aug 23 17:20:59 2006 Wed Aug 23 17:46:15 2006 (a) My = 0.62) Shock in A Suddenly Open Valve 1.5 4 3.

63) Since Msx can be represented by Msy theoretically equation (5.3 and My = 1. (c) Utilizing Mx = calculate the new “improved” Mx . consider the case of My = 0. There are three solutions of which the first one is Msx = 0 and this is immediately disregarded. To illustrate the convergence of the procedure.98 between the two sides is Msy = My + CHAPTER 5. The other two solutions are Msx = (k + 1)Uy ± Uy (1 + k ) 4 cx 2 + 16cx 2 (5.” This method is applied by the following algorithm: (a) Assume that Mx = 1.64) is a quadratic equation for Msx . the larger number of the iterations required to achieve the same accuracy.65) . other conditions ahead of the shock.3. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. The unknown data are the piston velocity. Therefore. sufficient results can be achieved after 3-4 iterations. (d) Check the new and improved Mx against the old one. the temperature. and.64) Ty My + M y Tx Equation (5. In this case. Yet. If it is satisfactory. it creates a shock that moves at a speed greater than that of the piston itself.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. no Mach number is given but pieces of information on both sides of the shock.9)). Piston Velocity When a piston is moving.0. The results show that the convergence occurs very rapidly (see Figure (5. for most practical purposes. NORMAL SHOCK (Msx ) + 2k k−1 2 (Msx ) − 1 2 k−1 2 (5. It is common practice to solve this equation by numerical methods. The larger the value of My . stop or return to stage (b). the calculations for Us can be obtained from equation (5.63) can be solved. One such methods is “successive substitutions.

To explain this phenomenon. 5. look at the normal Fig. This shock–choking phenomenon Shock in A Suddenly Open Valve Maximum M ’ possible is somewhat similar to the 2. Some use equation (5. The differ1. what is the maximum velocity that can be reached before this model fails? In other words. and the downstream Mach y y(max) Maximum My’ .10: The maximum of “downstream” Mach number as a function of the specific heat. the shock Mach number cannot be increased. 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. Shock–Choke Phenomenon Assuming that the gas velocity is supersonic (in stationary coordinates) before the shock moves. shock.67) Where the “stange” Mach number is Myx = Uy /cx .68) to explain the next Shock-Choke phenomenon. The limit of the equation when cx → ∞ leads to Msx = (k + 1)Myx 4 (5. and the only solution left is Msx = (k + 1)Uy + Uy (1 + k ) 4 cx 2 + 16cx 2 (5.25 has no limit. 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. Consider when the “upstream” Mach approaches infinity. This phenomenon occurs in many industrial applications.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.75 suddenly and completely clos0.5.5 The spesific heat ratio.25 discussed earlier in a nozzle 2 flow and in other pipe flow mod1.5 M choking phenomenon that was 2.3. such as the internal combustion engines and die casting. k.75 els (later chapters).66) or in a dimensionless form Msx = (k + 1)Myx + Myx (1 + k ) 4 2 + 16 (5. Mx = Msx → ∞.5 ence is that the actual velocity 1. k ing of valve results in no limit (at least from the model point Thu Aug 24 17:46:07 2006 of view).

These limits determine the maximum velocity after the shock since Umax = cy My . . 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.26) and (5. One can view this as the source of the shock–choking phenomenon. is approaching to (k − 1)/2k .69) Mx = Msx Thus. the Mach number is approaching infinity because of the temperature ratio but the velocity is finite. The Mach number is reduced to the maximum value very rapidly.100 CHAPTER 5. NORMAL SHOCK number.73) What happens when a gas with a Mach number larger than the maximum Mach number possible is flowing in the tube? Obviously. consider that the maximum Mach number is obtained Py → ∞. the Mach number is ∞    Ty k − 1 =   T 2k   x (5.23) to this situation the following is obtained: Msx = k+1 2k Px −1 +1 Py (5.72) + When the pressure ratio is approaching infinity (extremely strong pressure ratio). according to equation (5.71) 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. From the upstream side.25) into equation (5. the semi steady state described by the moving shock cannot be sustained. A similar phenomenon to the choking in the nozzle and later in an internal pipe flow is obtained. To understand this limit.71) 1− Substituting equations (5. the results is My = 2 k (k − 1) (5.70) 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 (5.38). By applying equation when the pressure ratio is approaching infinity Px (5.

3.54006 0.20 2.96 204124.13).85 922.75) is constant. The general case is when one gas flows into another gas with a given velocity. 5.81 My 0.0815 1.25 985.50 Mx 1073.40 2.56 711. The mass flow rate when the pressure ratio is approaching infinity.60 1.54 750.50000 0. ∞.74) Equation (5.73029 Ty Tx 101 169842.2645 1. The only limitation is that the “downstream’ gas velocity is higher than the “upstream” gas velocity as shown in Figure (5.56 668.70 1.90 2. is cy ρy m ˙ = U y ρ y = M y cy ρ y = M y A √ My kPy = RTy kRTy Py RTy (5.80 1.2964 1. THE MOVING SHOCKS k 1. with the exception of Ty the mass flow rate is approaching infinity when the pressure ratio is approaching infinity.3. Thus.3 Partially Open Valve The previous case is a special case of the moving shock.09 833. the shock–choke phenomenon means that the Mach number is only limited in stationary coordinates but the actual flow rate isn’t.23 873.54772 My 2. .09 265805.62 695.99 235702.02 773.40 1.8898 1.6330 1.1785 1.40825 0.37797 0.78 261117.43301 0.53161 0.74 681.93 243332.10 2.87039 0.00 2.00000 0.77151 0.75) Since the right hand side of equation (5.86 216507.36 270031.93048 0.4434 1.30 2.64 255883.45374 0.05 226871.52223 0.29 188982.81786 0.30 1.25) can be transferred for large pressure ratios into m ˙ ∼ A Ty Px k − 1 Tx k + 1 (5.00 729.33968 0.50 1.47141 0.51177 0.61 801.79 250000.85 Table of maximum values of the shock-choking phenomenon.48667 0.5.44 273861.74) and equation (5.

Ty (Msy ) − Mx Tx . The procedure for the calculations can be done by the following algorithm: (a) Assume that Mx = Mx + 1.v. If it is satisfactory.11: A shock moves into a moving medium as a result of a sudden and complete open valve. stop or return to stage (b). (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.102 CHAPTER 5. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. It has to be mentioned that the temperature “downstream” is unknown (the flow of the gas with the higher velocity).v. 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. A common problem is to find the moving shock velocity when the velocity “downstream” or the pressure is suddenly increased.76) An additional parameter has be supplied to solve the problem. NORMAL SHOCK ÿ¡ £¢Äÿ¥¤§¦ ÿ¡ ©¨ Uy > U x ô0õÚö Ux ÷#ø c. 5.77) (5. (a) Stationary coordinates (b) Moving coordinates Fig. ò îÊï&ðÍñ ó Ux = Us − Ux Upstream ùÊú&ûÍü ýçþ c.

8 1. Thu Oct 19 10:34:19 2006 and for various static “upstream” Mach numbers.8 M ’ = 0.0 0.5 nomenon is still there (the Mach number is 0. x x x My y 5.2 a faster flow is running into a slower fluid.2 2 2.4 limited.3 exhibits some “downstream” Mach num0.3.12: The results of the partial opening that the maximum can also occurs in the of the valve. THE MOVING SHOCKS 103 Shock in A Suddenly Open Valve Earlier. 5.6 M’ bers for various static Mach numbers.4 2. vicinity of the previous value (see following question/example). Mx .7 The mathematics is cumbersome but re0. Figure (5.9 nomenon also occurs in the case where M ’ = 0.3.4 Partially Closed Valve . This phe0. M ’ = 0.4 0. My .9 0.12) 0. The figure demonstrates Fig. it was shown that the shock chokk=14 ing phenomenon occurs when the flow is 1 running into a still medium.5.8 1.6 sults show that the shock choking phe0. not the actual flow).

c. (a) Stationary coordinates (b) Moving coordinates Fig.v.v. 5.13: 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.78) Ux = U s + U x Uy = U s + U y (5.     Ux Uy Ux = Us + Ux Upstream Uy = Us + Uy ρ y Py Ty c. Similar equations can be written: (5.79) . The information propagates upstream in the same way as before.

the Mach velocity ahead of the shock has to calculated.953 0. 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.1206 My Mx My 1.2: 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.4) one can obtain the following table: Mx 2. Assume the specific heat ratio is 1.3 × 287 × 300 By utilizing Potto–GDC or Table (5.3. NORMAL SHOCK Mx = M s + M x (5.5 Worked–out Examples for Shock Dynamics Example 5. S OLUTION It can be observed that the gas behind the shock is moving while the gas ahead of the shock is still.63955 . First. Thus. Ty (Msy ) + Mx Tx 5. My = √ U kRT =√ 450 ∼ 1.132 Ty Tx Py Px P0 y P0 x 0.296 1. (b) .80) My = M s + M y (5.604 4. If it is satisfactory.54220 0. stop or return to stage (b). Compute the pressure and the temperature behind the shock. it is the case of a shock moving into still medium (suddenly opened valve case).104 CHAPTER 5.81) For given static Mach numbers the procedure for the calculation is as follows: (a) Assume that Mx = Mx + 1.3.0 1.

0 0.953 × 1.724 The table was obtained by utilizing Potto–GDC with the iteration request.4.710 0. Calculate the velocity and the pressure behind the reflecting shock. the prime properties can be found.953[Bar] Px The velocity behind the shock wave is obtained √ m Uy = Mx cx = 1. Simply using the Potto–GDC provides for the temperature and velocity the following table: Mx 2.0 ∼ 4.2961 1.3. S OLUTION The first thing that needs to be done is to find the prime Mach number Mx = 1.53487 0.604 × 300 ∼ 481.044 My 0. The specific heat ratio can be assumed to be k = 1.3 × 287 × 300 ∼ 378.044 2. Then. find the upstream prime Mach.722 1.0 0.56995 Ty Tx My 0.2961.0 Ty Tx Py Px P0 y P0 x 0.57040 0. Mx to be 1.5.042 2.56994 0. Then using Table (5. The static pressure at the tube is 2Bar and the (static) temperature of 300K .70009 If you insist on doing the steps yourself.56995 0.2961. The gas is brought into a complete stop by a sudden closing a valve.132 × 1.2) you can find the proper Mx . THE MOVING SHOCKS Using the above table.0445 My Mx My 0.3: Gas flows in a tube with a velocity of 450[m/sec].0 1. .56995 1.9432 1.724 4.2K Tx 105 In same manner.724 1.0 0. At this stage the reflecting shock velocity is unknown.045 2. the temperature behind the shock is Ty = T y = Ty Tx = 1. it can be done for the pressure ratio as following Py = P y = Py Px = 4.0 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 2.72 sec Example 5.2961 2.724 1.

3).55830 Example 5.3574 My 0. 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 (5.52778 Ty Tx ρy ρx Py Px P0 y P0 x 2.20000 1.0789 1.0000 3.78928 0.1583 6.3574 My Mx My Ty Tx Py Px P0 y P0 x 0.3574 in the moving shock table provides Mx 2. Hint. It can also be obtained from the stationary normal shock wave table.89509 0.3166 0.4: What should be the prime Mach number (or the combination of the velocity with the temperature.3020 0. S OLUTION Refer to section (5.0 2.52778 0. .1220 My Mx My Ty Tx Py Px P0 y P0 x 0.106 CHAPTER 5. Potto-GDC provides the solution of the above data Mx 1. 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).99813 If the information about the iterations is needed please refer to the following table.3. Potto-GDC provides for this temperature ratio the following table: Mx 2. NORMAL SHOCK Example 5.317 0.000 6.5: A gas is flowing in a pipe with a Mach number of 0.4.40000 0.4) for the calculation procedure.55832 using the required Mx = 2. Calculate the speed of the shock when a valve is closed in such a way that the Mach number is reduced by half.

The piston is Mx = 0.89509 0.89595 0.0930 1.89494 0.89509 0.20000 0.0106 1.0045 1.2705 1. Calculate the time Fig.2547 1.92479 0.3019 1.3020 1.1220 My 0.0789 1.20000 0.3609 1.20000 0.1220 1.8 accelerated very rapidly and the air adjoined the piston obtains Mach number M = 0.1226 1.0802 1.20000 0.89461 0.3017 1.3020 1.20000 0.1220 1.89504 0.1220 1.0789 1.0545 1.73971 0.20000 0.1288 1.3020 1.20000 1.20000 0.1182 1.4 and 300◦C.4 My = 0.14: Schematic of a piston pushing air in a it takes for the shock to reach the tube.20000 0.89508 0.1221 1.1241 1.20000 0.0m length.89009 0.1099 1. the solution is .89354 0.3075 1.2989 1.0790 1.89512 0.0789 1.20000 0.6: A piston is pushing air that flows in a tube with a Mach number of M = 0.3199 1.89536 0.5041 1.0789 1.89517 0.20000 0.20000 0. end of the tube of 1.0030 1.0787 1.89509 Ty Tx Py Px 107 My 0.3.0836 1.20000 0.1220 1.99548 0.1443 1.0789 1.1216 1.90416 0.20000 0.1200 1.8.0789 2.0832 1.3020 1.89510 0.3020 1.0712 1.20000 0.20000 0.3020 1.84424 0.0789 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 1.3011 1.1219 1.1220 1. Calculate the velocity of the shock created by the piston in the air.0793 1.1259 1.1967 1.89509 0. S OLUTION Using the procedure described in this section.20000 0.5. 5.1208 1.3025 1.89509 0.0789 1.20000 0.0789 1.3037 1.1222 1.0788 1.1220 1.2032 1.89789 0.0765 1.2922 1.87903 0.20000 0.1220 1.20000 0.20000 0. Assume that there is no friction and the Fanno flow model is not applicable.20000 0.0782 1.3020 Example 5.3022 1.4000 1.

80000 0.2380 − 0. S OLUTION The stationary difference between the two sides of the shock is: ∆U =Uy − Ux = cy Uy − cx Ux   √   1.6216 The time it takes for the shock to reach the end of the cylinder is t= length Us cx (Mx −Mx ) =√ 1 = 0.108 Mx 1.7: From the previous example (5.80000 1.2248 1.1519 1.1517 1.81942 Ty Tx Py Px My 0.82716 0.5000 1.1531 1.1519 1.98860 The complete iteration is provided below.4583 1.2378 1.4 × 287 × 300(1.1435 1.80000 0.4[m/sec] .1519 2.80000 0. NORMAL SHOCK Ty Tx Py Px P0 y P0 x 0.81958 0.2380 My 0.1519 −0.81940 0.80000 1.2381 1.5834 1.70109 0.0034[sec] 1.2380 My Mx My CHAPTER 5.50000 0.6215 1.8 × 1.10) calculate the velocity difference between initial piston velocity and final piston velocity.5    q Ty Tx =  ∼ 124.6207 1.3202 1.2380 1.81829 0.2400 1.6217 1.80000 0.4 × 287 × 300 0. i 0 1 2 3 4 5 6 Mx 1.6215 0.6273 1.80000 0.81942 0.80000 0.1519 1.81943 0.4) Example 5.

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.5. expansion waves occur with a reduction of pressure.15: Figure for Example (5.318 0. The high pressure . Calculate the time it will take for the two shocks to collide.89048 Mx 0. A cylinder with two chambers connected by a diaphragm. On one side the pressure is high.0 My 0.4 Shock Tube The shock tube is a study tool with very little practical purposes.173 0. It is used in many cases to understand certain phenomena.047 1.1283 My 0. SHOCK TUBE Example 5.95890 Ty Tx Py Px P0y P0 x Fig. 1. The distance between the two pistons is 1[m].93471 Mx 0. The solution is given by equation (5.0 The time for the shocks to collide is t= 1[m] length = ∼ 0.0 My 0.4. while the pressure on the other side is low.0013[sec] Usx 1 + Usx 2 (1.99785 70.66). When the diaphragm is ruptured the gas from the high pressure section flows into the low pressure section.93451 Ty Tx Py Px P0y P0 x Uy cx 347.1283)347. At the back of the shock.8: An engine is designed so that two pistons are moving toward each other (see Figure (5. 1. For this case the following table can easily be obtain from Potto–GDC for the left piston Mx 1.8) Uy cx 347.99959 40. When the pressure is high enough.0715 + 1. and.083 1.15)). Other situations can be examined and extended from these phenomena. a shock is created that it travels to the low pressure chamber.0715 My 0. 5. The air between the pistons is at 1[Bar] and 300K . The temperature is known to reach several thousands degrees in a very brief period of time. This is the same case as in the suddenly opened valve case described previously. 5.0 while the velocity of the right piston is Mx 1. it is the explicit analytical solution.

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.. 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. Initially. Zone 4 is where the gradual transition occurs between original high pressure to low pressure. On the other side. 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 . In zone 5. the expansion waves are moving into the high pressure chamber i. the gas from the driver section is coalescing from small shock waves into a large shock wave. The properties in the 5 1 4 3 2 different zones have different relationships.16)). 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 shock is moving at a supersonic speed (it depends on the definition.e. what reference temperature is being used) and the medium behind the shock is also moving but at a velocity. this is a case of sudden opened valve). U2 . The velocities in the expansion chamber vary between three zones. The expansion front is moving at the local speed of sound in the high pressure section.e. In fact. the driver section. In this analysis. The assumption is that the shock is very sharp with zero width. the velocity is zero and the pressure is in its original value. The expansion front wave velocity is larger than the velocity at the back front expansion wave velocity. In zone 3 is the original material that was in the high pressure chamber but is now the same pressure as zone 2. while the distance pressure in the two zones are the same. 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. 5. Zone 4 is expanding during the initial stage (until the expansion reaches the wall). The boundaries of zone 4 are defined by initial conditions.16: The shock tube schematic with a pressure ”diagram. which can be supersonic or subsonic in stationary coordinates. Fig. 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. it is assumed that this time is essentially zero. i. Zone 1 is an undisturbed gas and zone 2 is an area where the shock already passed.” The pressure. NORMAL SHOCK chamber is referred to in the literature is the driver section and the low section is referred to as the expansion section.110 CHAPTER 5.

82) becomes Ms1 = Or expressing the velocity as Us = Ms1 c1 = c1 k − 1 k + 1 P2 + 2k 2k P1 (5. a review of the material is provided thus far.).4. c ∼ 340 thus the whole process takes only a few milliseconds or less.85) The fluid velocity in zone 2 is the same U2 = U s − U 2 = U s 1 − From the mass conservation.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 (5.89) . Thus.111 √ range of the speed of sound.86) (5. A typical design problem of a shock tube is finding the pressure to achieve the desired temperature or Mach number.84) 2k Ms1 2 − 1 k−1 (5. The pressure ratio between the two sides of the shock is 5.82) k − 1 k + 1 P2 + 2k 2k P1 (5. The relationship between the different properties was discussed earlier and because it is a common problem. The following equations were developed earlier and are repeated here for clarification.88) After rearranging equation (5.88) the result is U2 = c1 k P2 −1 P1 2k k+1 P2 k−1 P1 1+k (5.87) 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. it follows that U2 ρ1 = Us ρ2 U2 Us (5. these kinds of experiments require fast recording devices (a relatively fast camera and fast data acquisition devices. SHOCK TUBE k−1 P2 = P1 k+1 Rearranging equation (5.

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    (5.91) dρ (5.90) it follows that dρ dU = −c = c5 ρ ρ ρ5 k−1 2 T = T5 ρ ρ5 k −1 2 (5.93) 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. hence equation (5.96) After some rearrangement. 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.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. NORMAL SHOCK On the isentropic side. equation (5.94) (5.112 CHAPTER 5.92) can be integrated as follows: U3 ρ3 dU = U5 =0 ρ5 c5 ρ ρ5 k −1 2 dρ (5.90) Since the process in zone 4 is isentropic.92) Equation (5.97) . taking the derivative of the continuity equation. and dividing by the continuity equation the following is obtained: dU dρ =− ρ c (5. d(ρU ) = 0. in zone 4.

In that case.5.100) Example 5.7 Shock with Real Gases Shock in Wet Steam Normal Shock in Ducts The flow in ducts is related to boundary layer issues. the following is obtained: 1+ T2 = T1 1+ k1 −1 k1 +1 k1 −1 k1 +1 P2 P1 P1 P2 (5. It is normal to have a large Mach number with a large Re number. SHOCK WITH REAL GASES Or in terms of the Mach number.98) Using the Rankine–Hugoniot relationship and the perfect gas model. the assumptions in construction of these models are acceptable and reasonable.6 5. . 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. For a high Reynolds number.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 (5.5 5.9: 5 A shock tube with an initial pressure ratio of P P1 = 20 and an initial temperature of 300K .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.

The time it takes for the material from the valve to reach the exit is distance 3 t= = ∼ 0.967 30. 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.0002 [sec] to be opened). In this system.0021[sec] 1 . 4 × 287 × 300 × 5.114 CHAPTER 5.00 0. Mx = Msx = 5. opening valve probelm The distance between between the valve and the pipe exit is 3[m].0 5. The temperature should not increase above a certain value. 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. The ambient temperature is 27◦ C and 1[Bar].41404 0. Assume that there isn’t any resistance whatsoever in the pipe. NORMAL SHOCK 5. the time is t= distance 3 √ = ∼ 0.17: Figure for Example (5.0850sqrt1. An engineer is required to design a cooling system for a critical electronic deexit valve vice.668 Ty Tx Py Px P0 y P0 x 0. Comment on this proposal. 5.10: This problem was taken from the real industrial manufacturdistance ing world.8 More Examples for Moving Shocks Example 5.17)). 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. Therefore. 668 sqrt 1 . After building the system. 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. The conditions upstream of the valve are 30[Bar] and 27◦ C .0017[sec] 5.0850.0850 My Mx My 1. air is Fig.057811 The direct calculation will be by using the “upstream” Mach number.967 My kRTy .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. S OLUTION This problem is known as the suddenly open valve problem in which the shock choking phenomenon occurs.

97937 .37625 1. TABLES OF NORMAL SHOCKS.03284 1.9.12799 1.78596 Ty Tx ρy ρx Py Px P0y P0x 1.34161 1.10) will not change the effects of heating. 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.91177 0.25 1.4 IDEAL GAS 115 During that difference of time the material is get heated instead of cooling down because of the high temperature.00000 0.24500 1.99985 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.80500 1.98706 0.20 1.00000 1.51570 1. K = 1. Example 5.99669 0.00000 1.11: Example (5. Time[Msec] To increase the pipe diameter will not change the temperature and therefore Fig.19087 1. The plot of the mass flow rate and the velocity are given in Figure (5.18).30 My 1.09658 1.4 Ideal Gas Table 5.99280 0. There is also the possibility of steps increase in which every step heat released will not be enough to over heat the device.00000 1.4 Mx 1.65625 1. Thus. Plot the pipe exit temperature as a function of the time.06494 1.18: The results for Example (5.99893 0. k = 1.42857 1. Thus.08398 1.5.1: The shock wave table for k = 1. S OLUTION 5.25504 1.16908 1. 5.00 1.10 1.84217 0. The last possible requirement a programmable valve and very fast which its valve probably exceed the moving shock the valve downstream.15938 1.10) deals with a damaging of electronic product by the temperature increase.81264 0. the heating time is reduced significantly.05 1.51333 1. Try to estimate the temperature increase of the product.00000 0.87502 0.9 Tables of Normal Shocks.95313 0.11958 1.15 1. this suggestion contradicts the purpose of the required manufacturing needs.

31915 1.35 2.67901 2.54014 0.53441 0.00 5.60 1.41189 0.90625 23.83333 45.68966 1.69919 4.40625 41.32544 5.75 6.04033 2.95958 2.79023 0.68750 1.38797 1.08846 2.96974 0.66844 0.40625 3.82180 7.83457 0.62814 0.77614 1.28720 1.05100 0.72855 1.44231 0.50 2.74420 0.80 1.41908 0.65625 10.68410 0.59188 2.40897 0.21182 5.45833 2.85 1.35379 1.42878 0.60278 1.95819 0.73958 6.13750 2.49901 0.00 3.62809 0.42623 4.15625 29.45833 1.75 5.88231 2.00 6.54055 0.51792 0.94732 2.41523 0.60570 0.50 5.51568 2.86207 1.07869 5.05 2.04236 0.63625 2.55 1.76736 0.81190 2.60792 1. NORMAL SHOCK Table 5.42280 1.00958 3.56906 0.95122 3.57143 4.89520 0.21295 0.45833 26.00500 6.35 1.46192 0.27335 3.81268 0.69751 0.35922 2.08455 3.25 My 0.57735 0.53158 1.12500 38.51299 0.42355 0.22261 1.47519 0.19772 2.58329 0.40 2.49458 1.52861 0.07505 0.92979 0.87509 5.99311 2.73625 4.43811 2.94680 1.81188 4.70109 0.02965 0.45 2.83625 7.85572 0.51931 0.25 5.40416 0.65396 0.12000 2.40623 0.32022 1.26087 4.40216 Ty Tx ρy ρx Py Px P0y P0x 1.74002 2.59562 0.65105 0.50 4.50000 4.85686 1.70 1.17166 0.27625 6.90 1.36906 7.4 (continue) Mx 1.04500 4.26958 4.00 2.33333 3.65 1.95 2.53637 1.21190 3.54706 0.09170 0.97833 5.22625 5.67420 0.50000 20.30 2.76175 0.98958 35.20500 3.28625 2.45115 0.61333 3.61650 0.43496 0.48000 5.56128 0.13876 0.01863 3.50 3.49181 0.25 4.75 3.72087 0.15 2.03536 0.03175 2.90138 1.64054 0.11525 2.45 1.50 1.82625 4.73971 0.94059 8.14894 5.44891 4.32834 0.31505 3.25 3.33333 12.56148 0.11256 0.60553 0.00000 5.00 4.85714 4.80000 6.20 2.15625 14.26451 0.75 1.12500 16.40625 0.23958 18.27907 2.02498 .64729 1.55395 0.26829 5.55333 6.10 2.40642 0.77045 1.98511 3.1: The shock wave table for k = 1.29878 6.116 CHAPTER 5.66667 2.66894 4.52312 0.94484 0.07229 4.75 4.06172 0.71956 0.25 2.87599 0.91156 5.25469 1.00000 31.58618 0.39657 2.61194 3.81322 1.12500 8.82000 3.04688 4.56935 1.40 1.14897 3.91319 0.

00 8.97099 0.219 1.195 2.39385 0.01 0.66874 5.89128 0.0 0.99790 0.00000 61.01798 0.71429 49.39048 0.423 1.028 1.99973 0.06 0.10 0.0 0.58939 5.710 1.90625 74.006 1.0 0.037 1.00000 1.043 1.00979 0.0 0.00645 0.94291 0.505 1.171 1.4 IDEAL GAS Table 5.024 1.92832 0.02115 0. K = 1.016 1.56522 5.00 7.941 2.47883 5.014 1.61165 5.98227 0.39607 0.126 1.0 0.062 1.149 1.381 1.98446 0.96506 0.99994 0.75 8.25 9.98958 57.01535 0.95944 0.99998 0.97074 0.99403 0.780 3.502 1.60 0.39289 0.02 0.004 1.269 1.08 0.12500 110.0 0.36508 5.506 1.80077 10.0 Ty Tx Py Px P0y P0 x 1.25 8.38758 Ty Tx ρy ρx Py Px P0y P0x 9.80 0.4) Mx 1.00849 0.5.01316 0.50000 79.38672 14.20 0.00 My 0.112 3.00739 0.69946 0.69273 17.38750 5.030 1.49152 19.028 1.97647 0.102 1.38860 0.68504 5.049 1.057 1.99999 0.2: Table for a Reflective Shock from a suddenly closed end (k=1.323 1.39121 0.676 1.0 0.25 7.50 7.04 0.055 1.86537 0.23958 84.0 0.012 1.12500 89.50000 0.040 1.39879 0.133 1.087 1.87948 12.0 0.50 0.0 0.30 0.012 1.95231 0.03 0.05 0.94836 0.57997 18.0 0.00437 0.043 1.01133 0.4 (continue) 117 Mx 6.9.95387 0.473 1.40038 0.44444 5.63218 5.80251 0.53890 5.00 My 0.07 0.020 1.38918 0.99986 0.0 0.442 1.40667 5.99997 0.072 1.00 9.73029 0.00496 0.50 6.766 My 0. TABLES OF NORMAL SHOCKS.65116 5.73958 116.50 8.84463 0.00343 0.42736 20.00000 0.99113 15.82755 .082 1.589 1.38807 0.0 0.15625 65.018 1.00387 0.75 9.032 1.45833 69.15643 9.82978 16.024 1.17678 14.0 0.00304 Table 5.0 0.70019 5.33333 99.09 0.475 2.316 1.99980 0.0 0.70 0.90 1.75 7.67171 0.1: The shock wave table for k = 1.196 1.75 10.50 9.0 0.15625 94.118 1.39491 0.64673 0.127 1.62095 13.00565 0.39736 0.40 0.036 1.99317 0.89918 0.46939 11.98812 0.65625 105.008 1.99991 0.344 1.12500 52.268 1.76452 0.0 0.16229 11.39201 0.62425 Mx 0.38980 0.51020 5.

78652 0.810 1.397 1.0 0.09 0.715 6.280 4.541 1.004 1.210 1.552 2.031 1.0 0.297 2.037 1.4)(continue) Mx 1.162 7.990 2.146 2.576 1.74316 0.0 0.99996 0.20 0.41523 0.38817 0.15637 0.56935 0.868 1.063 1.08 My 0.0 0.98019 0.43894 0.0 0.044 1.70283 0.800 8.317 1.016 1.20 1.448 2.0 0.50814 0.0 Ty Tx Py Px P0 y P0 x 1.762 3.014 1.90 2.0 0.336 8.21 29.020 1.043 1.858 1.862 4.00000 1.94746 0.00000 0.0 My 0.087 2.41 15.073 1.048 2.0 0.99402 0.256 6.859 5.0 0.98807 0.54114 0.152 1.62 84.050 1.3: Table for shock propagating from suddenly opened valve (k=1.4) Mx 1.99971 0.134 1.410 3.83607 0.00350 0.65290 0.0 0.188 1.295 1.0 0.50 0.827 7.0 0.1E +2 1.517 9.66462 Mx 0.96465 0.0 0.10 0.649 1.206 5.033 1.74403 0.0 0.120 1.189 2.00 8.086 1.2: Table for Reflective Shock from suddenly closed valve (end) (k=1.00 6.0 0.99763 0.0 0.0 0.40213 0.008 1.136 1.01 29.99998 0.93133 0.03 0.0 My 0.97629 0.831 5.897 1.99985 0.952 2.52904 0.4E +2 1.99999 0.08 0.0 0.60 1.88717 0.38457 Mx 1.38608 0.00 5.00125 Table 5.89 12.029 1.018 1.60 0.00 4.024 1.495 1.58578 0.104 1.248 1.0 0.0 0.49912 0.000 6.39566 0.703 10.40 0.346 2.01 0.30 1.39116 0.04 0.50 1.245 2.80 1.89039 .70 Ty Tx Py Px P0 y P0 x 1.60761 0.05 19.78840 0.0 0.47 1.25 24.728 5.0 0.734 17.0 0.0 0.97045 0.431 8.30 0.94180 0.07 0.55453 0.00204 0.087 2.02 0.441 2.061716 0.058 1.51996 0.0 0.0 0.0 0.0 0.012 1.00 9.00 3.0 0.024 1.33 3.066 8.026517 0.69834 0.00 44.037 1.057 1.43921 0.0 0.390 1.656 2.0 0.791 2.40284 0.012492 0.041 1.99994 0.7E +2 0.613 My 0.133 1.012 1.49092 0.10 1.40 1.96069 0.14 62.028 1.99990 0.06 0. NORMAL SHOCK Table 5.331 1.05 0.088 1.006 1.99979 0.56312 0.51808 0.00 7.70 1.325 11.98216 0.95315 0.118 CHAPTER 5.00639 0.727 1.47855 0.95888 0.00 10.0 0.60401 0.99181 0.

015 1.83661 0.9E + 3 6.7E +2 8.95 64.00272 0.24 11.025 4.3) Mx 1.012 1.0 0.00 1.2E +3 1.44536 0.9E + 3 5.3: Table for shock propagating from suddenly opened valve (k=1.5E + 3 8.37814 0.07 18.536 4.067 1.98290 0.8E + 2 7.53817 0.42622 0.009 1.387 4.269 2.0 My 0.28412 0.023 1.56619 0.0 My 0.881 3.92 42.0 0.889 1.9E +2 1.0 0.48823 0.013 1.0 0.0 0.029 My 0.98857 0.003 1.975 11.59699 0.99 25.38713 0.4E + 3 6.889 Ty Tx Py Px P0 y P0 x 1.896 2.500 1.0 0.38248 0.99998 0.62 17.0 0.200 1.2E +2 8.0 0.2E + 2 1.72 80.400 1.830 1.79 My 0.62 61.026 1.054 1.508 2.1E +3 1.33 21.49586 0.62923 0.40843 0.0 0.4) 119 Mx 1.810 1.0 Table 5.800 1.61 1.919 2.37813 0.37818 0.7E + 2 4.00000 1.745 1.0 0.171 8.32 16.4E + 3 4.457 10.889 1.658 4.870 1.4: Table for shock propagating from a suddenly opened valve (k=1.040 0.37817 0.9E + 2 2.860 1.744 1.0 0.38974 0.37812 0.0058 1.050 Ty Tx Py Px P0y P0 x 1.888 1.35 25.0 0.90 1.600 1.18 73.2E +2 9.0 0.0 0.0 0.3E +2 7.38870 0.0E + 2 1.646 6.37822 0.0 0.38096 0.80 76.37816 0.99997 .888 1.293 8.37820 0.37944 0.00000 0.0 0.00395 0.0 0.0 0. K = 1.823 5.53 63.821 9.54 15.145 2.888 1.74 66.0 0.4 IDEAL GAS Table 5.551 8.020 0.00544 0.0 0.0 0.09 35.820 1.39028 0. TABLES OF NORMAL SHOCKS.9E + 3 7.99999 0.000497 0.57 31.0 0.5E + 3 5.37821 0.0 0.010 0.000181 3.840 1.0 0.931 14.84 1.5E + 2 1.0 0.0 0.62 14.69 68.68907 0.38402 0.0 0.97726 0.31 62.00721 0.9.46599 0.7E +2 9.300 1.99427 0.59649 0.51223 0.0 0.700 1.18575 0.040 1.0 0.2E + 3 5.0 0.368 5.030 0.10216 0.0 0.0 0.0 0.0 0.012 1.78 80.5.00101 0.07 90.37810 Mx 0.38557 0.00175 0.2E + 3 0.97166 Mx 0.0 0.25 12.840 7.02 83.39187 0.0 0.6E + 3 4.4E +3 3.491 1.45 22.18E −5 0.0 0.450 2.006 1.889 1.5E + 2 3.83 71.017 1.850 1.167 3.100 1.880 1.1E +3 1.040812 0.3E +2 7.889 1.76940 0.604 1.888 1.80 0.888 1.0 0.799 3.068 2.0 0.0 0.

89159 0.500 1.22904 0.0 0.048 1.426 2.191 2.98290 0.110 1.225 4.95 23.57853 0.109 4.035 1.64073 0.467 2.99987 0.0 0.39037 0.0 0.615 4.0 0.39031 0.74136 0.870 1.0 0.707 5.595 2.600 1.79611 0.100 0.15495 0.083607 0.31281 0.500 0.3) Mx 1.576 4.136 1.300 0.45807 0.980 2.889 Ty Tx Py Px P0 y P0 x 1.286 4.50100 0.92 23.44 18.223 1.052914 0.614 4.100 4.349 4.275 1.94415 0.43882 0.90734 0.553 1.064766 0.197 1.66575 0.55074 0.052879 0.898 3.706 1.95 23.0 0.94156 0.40097 0.0 0.888 1.063 1.616 My 0.081 1.80734 0.39624 0.096 4.40226 0.95 19.99991 0.016 4.031 1.053018 0.097 4.0 0.0 0.523 7.099 4.830 1.098 4.0 0.749 3.090 0.471 4.93 23.0 0.888 1.39160 0.96631 0.080 0.05 20.519 3.810 1.052984 0.052949 0.088718 0.842 3.0 0.300 1.800 1.663 1.400 0.75301 0.166 4.021 1.0 0.053 4.0 0.093988 0.0 0.0 0.060462 0.096 1.58 23.888 1.39027 0.052844 0.67546 0.274 1.177 1.482 4.39029 0.069 2.32 18.0 0.40257 0.700 1.600 0.060 1.200 1.96 0.923 4.0 My 0.99792 0.0 0.60847 0. NORMAL SHOCK Table 5.034 4.772 9.820 1.200 0.333 1.100 1.58223 0.073863 0.522 6.86274 0.054 1.616 4.056322 0.00 1.700 0.96610 0.018 1.101 4.613 4.0 0.042 1.919 2.99975 0.448 3.39034 0.098 1.91 23.024 1.91 23.140 1.834 3.800 0.95506 0.0 0.40418 0.90 22.053053 0.0 0.052809 .359 1.028 1.900 1.96056 0.0 0.0 0.0 0.30 23.644 2.0 0.236 2.39468 0.25 21.94959 0.102 4.0 0.070 0.400 1.84227 0.840 1.0 0.64 21.46 14.0 0.681 1.997 3.99288 0.099 4.71284 0.612 4.889 1.49333 0.49 20.367 11.0 0.553 4.4: Table for shock propagating from a suddenly opened valve (k=1.860 1.125 1.889 1.850 1.815 1.302 1.39780 0.94 23.52495 0.611 4.120 CHAPTER 5.785 1.39033 0.0 0.39030 0.39314 0.615 4.252 2.0 0.400 1.0 0.669 3.39938 0.93 23.0 0.888 1.42089 0.592 3.47875 0.39026 Mx 0.880 1.0 0.0 0.489 1.202 3.889 1.0 0.415 4.103 1.0 0.126 1.078654 0.888 1.613 4.452 1.381 3.069233 0.0 0.570 1.053088 0.959 2.99981 0.99995 0.478 1.39035 0.060 0.

K = 1.9.5.617 My 0.39025 Mx 0.0 My 1. TABLES OF NORMAL SHOCKS.052775 .3) 121 Mx 4.4 IDEAL GAS Table 5.97 0.4: Table for shock propagating from a suddenly opened valve (k=1.889 Ty Tx Py Px P0y P0 x 4.104 23.

122 CHAPTER 5. NORMAL SHOCK .

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

In the under–expanded case. point "e" niques). 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. 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. no continuous pressure possibly can exists.2: A nozzle with normal shock explain the usage of the tools (equations and tables) that were developed so far. the over–expanded nozzle is worse than the under–expanded nozzle because the nozzle’s large exit area results in extra drag. Nevertheless. Therefore. P B . .0/21 (see Mach’s photography techexit 354687292:2. The location of the shock is determined by geometry to achieve the right 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. NORMAL SHOCK IN VARIABLE DUCT AREAS point “a” and point “b” the flow is different from what was discussed before. On the other hand. 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. The shock occurs in a location where the cross section area is 6[cm2 ]. In the literature. It XY Z\[ ]_^8`Aa bcFdOe troat has to be recognized that the shock wave isn’t easily visible &'(*). In this case.+ -£. only exact geometry can achieve continuous pressure flow. Only in one point where P B = Pb continuous pressure exist. (It should be noted that the temperature of the surrounding is irrelevant in this case. the nozzle doesn’t provide the maximum thrust possible.124 CHAPTER 6. If the back pressure. The first example is for academic reasons.) Also determine the critical points for the back pressure (point “a” and point “b”). This nozzle is called an over–expanded nozzle. is lower than the critical value (the only value that can achieve continuous pressure) a shock occurs outside of 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]. Obviously if the back pressure. Nozzle throat area is 3[cm2 ] and the exit area is 9[cm2 ]. P B is smaller than Pb a discontinuous point (a shock) will occur. this example will provide the fundamentals to Fig. In conclusion. Example 6. In comparison of nozzle performance for rocket and aviation. 6. Calculate the back pressure and the temperature of the flow. once the flow becomes supersonic. 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.

0268 Again utilizing the isentropic relationship the exit conditions can be evaluated. as Ae A∗ = 1. For example. (point “y”).54743 Ty Tx ρy ρx Py Px P0y P0 x 1. Mx 2. My can be obtained.00587 and the error is less than %0.2588 × = 1. one can obtain using the isentropic relationship . Thus. Mx is about 2. My .8882.1972 My 0.18463 2.1972 the Mach number.11) the following Table for the isentropic relationship is obtained M 0.94345 0. With known Mach number the new star area ratio.50877 0.09393 0. but can be shown that the correction is negligible for a typical dimension ratio that is over 100.1972 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.49) or the GDC– Potto.1) or equation (4. the inside conditions are essentially the stagnation conditions (this statement is said without a proof.54746.22) or from Table (4. Mx = 2. the stagnation temperature and pressure are known T0 = 308K and P0 = 4[Bar]. A∗ .4656 0. This Mach number can be calculated by utilizing the isentropic relationship from the large tank to the shock (point “x”).18787 With this Mach number.2588 0.1). It has to be realized that for a large tank. With these values. before the shock is known and given as well. the Mach number at the exit can be calculated by utilizing the isentropic relationship. Then the relationship developed for the shock can be utilized to calculate the Mach number after the shock.2) My ∼ = 0.125 First.81568 1.8882 ∗ A Ay A 6 with this area ratio. Ay /A∗ is known and the exit area can be calculated as Ae Ay 9 Ae = × ∗ = 1.54743 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.197 as shown table below: M 2. in the case of ratio of 100 the Mach number is 0.8544 2. the subsonic branch can be evaluated for the pressure and temperature ratios.62941 From Table (4. the Mach number. Ax 6 = =2 A∗ 3 With this ratio (A/A∗ = 2) utilizing the Table (5.86457 1.0000 0. From the Mach number after the shock.2) or from equation (4. the exit Mach number has to be determined. The star area (the throat area).9474 5. From equation (5.

7538 Since the stagnation pressure is constant as well the stagnation temperature. Yet we have to pay attention that there two possible back pressures that can “achieve” it or target.81568 × 5. The area ratio for both cases.6728[Bar] .126 M 0.99226 × 4 ∼ =3.41820 0.6374 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.951 Ty Tx Tx T0 T0 =0.04730 2.97912 0.92882 1.41820 × 4 ∼ =1.97[Bar] For the supersonic sonic branch Pexit = Pexit P0 P0 = 0.97318 0.98133 × ∼ =299.19745 2. is A/A∗ = 3 In the subsonic branch (either using equation or the isentropic Table or GDC-Potto as M 0.0000 0.92882 × ∼ =2.34[Bar] The exit temperature is Texit = Texit T0 1 0.509 × 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.14190 Pexit = Pexit P0 P0 = 0.11310 3. the exit conditions can be calculated.32651 CHAPTER 6. In that case we don’t have to go through that shock transition.8882 0. Pexit = Pexit P0 P0 Py Py Px Px P0 P0 =0.094 × 4 T0 Ty 1 0.0000 3.854 × 0.466 × 0.99226 0.98077 0.94862 1.9K × 1.9195 0. NORMAL SHOCK IN VARIABLE DUCT AREAS T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.

2: In the data from the above example (6.113[Bar] And the mass flow rate reads m ˙ = 4105 3 × 10−4 × 321.5283 × 4 = 2.13[kg/sec] 287 × 256. .12[m/sec] P∗ P0 P0 = 0. Thus. In fact in many industrial applications. these kind situations exist. Example 6. For more practical example1 from industrial application point of view.7 = 321.127 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.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. 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 = 0.12 = 0. In these applications a small pressure difference can produce a shock wave and a chock flow.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.4 × 287 × 256.833 × 308 = 256. over 50% of the range of pressure a shock occores some where in the nozzle. While the pressure to achieve full supersonic flow through the nozzle the pressure has to be below the 42% the original value.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).

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.g.97188 0.2). My . Here. utilizing the isentropic Table (4. The conditions in the tank are again the stagnation conditions. 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. it requires some iterations by “smart” guessing the different shock locations.3914 0.113[Bar].0128 3. The area (location) that the previous example did not “produce” the “right” solution (the exit pressure was 2. 2 Of course.72)).128 CHAPTER 6. NORMAL SHOCK IN VARIABLE DUCT AREAS S OLUTION The solution procedure is similar to what was shown in previous Example (6.5000 0.5[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.75158 With these values the relationship between the stagnation pressures of the shock are obtainable e. there are two main possible ways to obtain the solution. the exit pressure is between point “a” and point “b”. 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.55250 With the information of Mach number (either Mx or My ) the area where the shock (location) occurs can be found.90500 1. the computer can be use to carry this calculations in a sophisticate way.2) or utilizing the GDC-Potto provides Mx 2. Utilizing the Table (4.3709 My 0.6575 0.905 4 Looking up in the Table (4. Thus.38034 0. First. . In fact.1755 6. Mathematically.93118 1. is known. It follows that there must exist a shock in the nozzle. the needed pressure is only 2[Bar] which means that the next guess for the shock location should be with a larger area2 . In the first method. the previous example information used and expanded.52628 Ty Tx ρy ρx Py Px P0 y P0 x 2. The exit total pressure can be obtained (if needed). The solution process starts at the nozzle’s exit and progress to the entrance. the exit Mach number. see also equation (4. More importantly the pressure ratio exit is known. Pexit Aexit 2×9 Pexit Aexit = = = 1.1).5525 0.

07158 0.1 Nozzle efficiency Obviously nozzles are not perfectly efficient and there are several ways to define the nozzleefficiency. as “simple” check this value is larger than the value in the previous example.3396 0.16747 Approaching the shock location from the upstream (entrance) yields A= A ∗ A = 2. The total energy that can be converted is during isentropic process is E = h0 − hexit s (6.4) 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. Vc Vc = √ η= (Uactual )2 (Uideal ) 2 h0 − hexit (Uactual )2 = 2 h0 − hexit s (Uideal ) (6.2 Diffuser Efficiency . NOZZLE EFFICIENCY M 2.47076 0.0188[cm2] A∗ Note.3709 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 129 0.3) (6. Cd = m ˙ actual m ˙ ideal (6.2) The efficiency can be defined as η= The typical efficiency of nozzle is ranged between 0. In the literature some define also velocity coefficient as the ratio of the actual velocity to the ideal velocity.15205 2. 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. 6.3396 × 3 ∼ = 7.5) 6.9 to 0.6.1) where hexit s is the enthalpy if the flow was isentropic.1. The actual energy that was used is E = h0 − hexit (6.99.

η= 2(h3 − h1 ) h3 − h 1 = h01 − h1 U1 2 (6.4: Schematic of a supersonic tunnel in a continoccurs in the test section.entropy Fig.65326 A×P A ∗ ×P 0 F F∗ .7[Bar] and temperature of 250K .02[m2 ].35714 0.11528 0.3: Description to clarify the definition of diffuser efficiency And further expanding equation (6.130 CHAPTER 6.4 can be assumed.0000 T T0 ρ ρ0 A A P P0 uous region (and also for example (6. k = 1.8) Example 6.02722 0. 6.6) can be converted to η= 2Cp (T3 − T1 ) U1 2 (6.07623 4.3) 0.0 and prescapacitor sure of 0. 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. Assume that a shock Fig. 6.4)) the required condition at point 3 are: M = 3.7) s. 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.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. The cross section in area between the nuzzle Compressor and diffuser is 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) P02 P2 01 02 2 P1 1 For perfect gas equation (6.2346 0. S OLUTION The condition at M = 3 is summarized in following table M 3.

37554 0.0047 ∼ 0.4968 0.50 0.09668 .6790 3.4: A shock is moving at 200 [m/sec] in pipe with gas with k = 1. ∗ Tx ) Then calculate the My by using Potto-GDC or utilize the Tables.5346 My 0.3333 0. In this case Potto–GDC provides the following table Mx 5.2346 = 0. The “downstream” shock number is Msy = Us k ∗ 287.37554 Ty Tx ρy ρx Py Px P0y P0 x 5.3.0047[m2] A 131 In this case.32834 Example 6.6. First the stagnation behind the shock will be Mx 3.8571 10. For example Potto-GDC (this code was produce by the program) Mx 5.0 5. ∗ Tx ∗ Ty Tx ∼ 2.02172 The calculation of the temperature and pressure ratio also can be obtain by the same manner.02/4.2.0143[m3] P0 d 0.5346 My Mx My 1. temperature and pressure “upstream” known.989 Ty Tx Py Px P0 y P0 x 0.0000 My 0.2963 34.47519 Ty Tx ρy ρx Py Px P0y P0 x 2. pressure of 2[Bar] and temperature of 350K .4789 6. P0 A∗ is constant (constant mass flow). S OLUTION This is a case of completely and suddenly open valve with the shock velocity.32834 A∗ d = P0 n ∗ 1 A n∼ 0.021717 The calculations were carried as following: First calculate the Mx as M x = Us / (k ∗ 287. DIFFUSER EFFICIENCY The nozzle area can be calculated by A∗ n = A A = 0.479 34. Calculate the conditions after the shock.

132 CHAPTER 6. Ux = m ˙ mRT ˙ 2 × 287 × 350 = = ∼ 502.24 ∼ 291.25 Ux =√ ∼ 2.75593 ∗ 1. what will be the piston velocity? S OLUTION This is an open valve case in which the pressure ratio is given.25[m/sec] ρA PA 200000 × 0.002 502.091 × 143 × 350 Thus the static Mach number. For this pressure ratio of Py /Px = 2 the following table can be obtained or by using Potto–GDC Mx 1. If the steady state is achieved.4 ∗ 287 ∗ 369.15 cx 1. The propagating piston is assumed to move into media with temperature of 300K and atmospheric pressure of 1[Bar]. Mx is Mx = . S OLUTION This is the case of a closed valve in which mass flow rate with the area given.24[◦C] Ty = T x Tx The velocity of the piston is then √ Uy = My ∗ cy = 0.2308 = 369.09668 − 0.6: A flow of gas is brought into a sudden stop. Thus. The imaginary gas conditions are temperature is 350K and pressure is 2[Bar] and R = 143[j/kg K ] and k = 1.989 Example 6.41087 ∼ 1.3628 My 0.16[m/sec] Example 6. NORMAL SHOCK IN VARIABLE DUCT AREAS Finally utilizing the equation to calculate the following My = Msy − My = 2.75593 Ty Tx ρy ρx Py Px P0 y P0 x 1.96697 The temperature ratio and the Mach numbers for the velocity of the air (and the piston) can be calculated.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.0000 0.6250 2. the “upstream” Mach is given.002[m3 ]. 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 = 0.2308 1.091 (Butane?). Calculate the conditions behind the shock wave.

796 0.609 2.0 2.0 0.796 9.46689 0.0 0.47886 0.6.0 0.923 2.922 2. The iteration of the procedure are i 0 1 2 3 4 5 Mx 3.914 9.47996 Ty Tx Py Px My 0.35101 This table was obtained by using the procedure described in this book.1500 2.8598 2.589 9.1500 2.9222 My Mx My 0.0 Ty Tx Py Px P0 y P0 x 133 0.796 9.0 0.922 2.47988 0.922 My 0.2.804 9.590 2.796 .589 11. DIFFUSER EFFICIENCY With this value for the Mach number Potto-GDC provides Mx 2.47996 2.47995 0.47996 0.589 2.589 2.0 0.940 2.4096 9.

NORMAL SHOCK IN VARIABLE DUCT AREAS .134 CHAPTER 6.

Moreover. 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. Flow in a vertical or horizontal nozzle are different because the gravity. In the previous chapters a simple model describing the flow in nozzle was explained. Please ignore. These models can served as limiting cases for more realistic flow. isothermal. In cases where more refined calculations have to carried the gravity or other forces have to be taken into account. The two models produces slightly different equations. 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.CHAPTER 7 Nozzle Flow With External Forces This chapter is under heavy construction. If you want to contribute and add any results of experiments. The equations results in slightly different conditions for the chocking and different chocking speed. 135 . The effects of the gravity of the nozzle flow in two models isentropic and isothermal is analyzed here. please do so. seem the most applicable. to this chapter. The simplified models that suggests them–self are: friction and adiabatic. You can help especially if you have photos showing these effects.

1 Isentropic Nozzle (Q = 0) external work or potential difference.2 Isothermal Nozzle (T = constant) .1) 7.e. z × g dh + U dU = f (x)dx The energy equation for isentropic nozzle provides (7.136 CHAPTER 7. i. NOZZLE FLOW WITH EXTERNAL FORCES 7.

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

1) describes the flow of gas from the left to the right.1) (8. 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 (8.5) 8.138 CHAPTER 8. equation of state is expressed as the following: P = ρRT (8. 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. First it must be recalled that the temperature is constant and therefore.1 The Control Volume Analysis/Governing equations Figure (8.2 it seems obvious to write this equation perhaps to consult with others. The Awetted area is the area that shear stress acts on.).7) (8. The heat transfer up stream (or down stream) is assumed to be negligible.2) where A is the cross section area (it doesn’t have to be a perfect circle. a close enough shape is sufficient. Hence.4) (8. 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. The shear stress is the force per area that acts on the fluid by the tube wall.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. (8.6) .

14) Differentiation of the isotropic (stagnation) relationship of the pressure (4. U . 2 It where are the stagnation equations? put them in a table put explanation how to derive this expression.11) yields should be noted that Fanning factor based on hydraulic radius. DIMENSIONLESS REPRESENTATION 139 Now. thus “Fanning f” values are only 1/4th of “Darcy f” values.8) into momentum equation (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 (8.8) Substituting equation (8.2.2) yields 4dx −dP − f DH 1 2 ρU 2 m ˙ A = ρU dU (8.12) (8.11) yields d(U 2 ) = kR M 2 dT + T d(M 2 ) (8.11) d(M 2 ) d(U 2 ) dT = − M2 U2 T (8.15) (8.9) Rearranging equation (8. the Fanning friction factor2 is introduced.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 (8.8.10) Now the pressure. this factor is a dimensionless friction factor sometimes referred to as the friction coefficient as f= τw 1 2 2 ρU (8. P as a function of the Mach number has to substitute along with velocity. instead of diameter friction equation. U 2 = kRT M 2 Differentiation of equation (8. .

Thus.15) when combined with equation (8. is selected as the choice for the independent fL . These different solutions are supersonic and subsonic solution.21) (8. continuity (mass).17) Notice that dT0 = 0 in an isothermal flow. 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 (8.18) yields M2 M2 (8. The variable.20) The four equations momentum.9) yields: dT0 = dT 1+ k−1 2 M 2 +T k−1 dM 2 2 (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). energy. T. ISOTHERMAL FLOW 1 2 2 kM k−1 + 2 M2 dM 2 M2 (8. One can notice that there are two possible solutions (because of the square power).22) the upstream variables are known.19) By utilizing the momentum equation it is possible to obtain a relation between the pressure and density.16) Differentiation of equation (4. fL The distance friction. P. . Substituting T for equation (8.140 dP0 dP = + P0 P 1 CHAPTER 8.18) dT0 (k − 1) M 2 dM 2 = 1 T0 M2 2 1 + k− 2 (8. state are described above.20) to become dU dP =− P U 3 Assuming (8.17) yields: T0 1 k−1 2 2 d M 1 2 + k− 2 M dT0 = Rearranging equation (8. There are 4 unknowns (M. 4D .14)) leads dM 2 2dU = 2 M U (8. the equations need to be obtained as a function of 4D density is eliminated from equation (8. ρ)3 and with these four equations the solution is attainable.

the term on velocity in the tube isn’t zero). By denoting the superscript symbol ∗ for the choking condition.25) The variables in equation (8. Heat transfer has a limited value therefore the model of the flow must be changed.10). 1 is the limiting case from a mathematical point of view. one can obtain − dP 4f dx − P DH kP M 2 2 = kP M 2 dP P 141 (8. The physical meaning of this value is similar to M = 1 choked flow which was discussed in a variable area flow in Chapter (4).23) Equation (8. Since the stagnation temperature (T0 ) has a finite value which means that dT0 → ∞. Integration of equation (8.24) 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. thus.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 (8.2.24) 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 (8.26) 4f dx = D 1/k M2 1 − kM 2 dM 2 kM 2 (8. A more appropriate model is an adiabatic flow model yet it can serve as a bounding boundary (or limit). To obtain this restriction 1 = kM 2 . Thus.8.22) into equation (8. When the value M = √ k 1 Mach number larger than M > √ it makes the right hand side of the integrate k negative.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 .27) yields 4f Lmax D = 1 − kM 2 + ln kM 2 kM 2 (8.26) that when M → √ the value k of right hand side approaches infinity (∞). 1 Further it can be noticed from equation (8. The term 4D the other side has to be positive as well. one can obtain that 1/k M2 = ∗2 U2 U (8.29) .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.

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 (8.36) is reduced into the following approximation ∼0 4f L D = 2 ln M1 − 1 − 1 − kM2 2 kM2 2 (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) and rearranging yields 2k 3k − 1 k k−1 P0 1 √ ∗ = P0 k 1+ k−1 2 M 2 k k −1 1 M (8.37) . It is very useful to obtain the relationship between the entrance and the exit condition for this case.2) 8. ISOTHERMAL FLOW Rearranging equation (8.36) For the case that M1 >> M2 and M1 → 1 equation (8. From equation (8. Denote 1 and 2 as the conditions at the inlet and exit respectably.31) (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 (8.30)-(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∗ (8.35) These equations (8.35) are presented on in Figure (8.33) equation (8.142 CHAPTER 8.30) ρ 1 =√ ρ∗ kM (8.32) 1+ 1 k−1 2 2 M k −1 + 2k k k−1 (8.29) is transfered into √ U kM = U∗ Utilizing the continuity equation provides ρU = ρ∗ U ∗ .

38) This relationship shows the maximum limit that Mach number can approach when fL the heat transfer is extraordinarily fast. 8.1 0.2: Description of the pressure. In reality.4.1 Fri Feb 18 17:23:43 2005 0.01 1 Mach number 10 Fig. In these cases. 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. Yet this model provides the directions of the heat transfer effects on the flow. This velocity requires a large entrance length to achieve good heat transfer. 8.4 Comparison with Incompressible Flow The Mach number of the flow in some instances is relatively small. 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. even small 4D > 2 results in a Mach number which is larger than 4. With this conflicting mechanism obviously the flow is closer to the Fanno flow model.5.8. COMPARISON WITH INCOMPRESSIBLE FLOW 143 Isothermal Flow P/P . one should expect that the isothermal flow should have similar characteristics as .

44) 4f L D = 1 2χ − χ2 − ln kM1 2 1 1−χ (8.42) Now equation (8.45) 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. a relationship between M2 and M1 and pressures has to be derived.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. by some mathematics. denote χ= P1 − P 2 P1 (8. the pressure loss is expressed as follows P1 − P 2 = 4f L D U2 2 (8. To obtain a similar expression for isothermal flow.39) one can obtained that M2 = M 1 P1 P2 (8.40) Substituting this expression into (8.39) fL represent the ratio of Now note that for incompressible flow U1 = U2 = U and 4D the traditional h12 .46) .144 CHAPTER 8. Expanding the solution for small pressure ratio drop. From equation (8. ISOTHERMAL FLOW incompressible flow.43) 4f L D = 1 2 1 − (1 − χ) − ln kM1 2 1 1−χ 2 2 (8. For incompressible flow.40) yields 4f L D = 1 kM1 2 1− P2 P1 2 − ln P2 P1 2 (8.41) Because f is always positive there is only one solution to the above equation even though M2. P1 − P2 /P1 .

8.5 Supersonic Branch Apparently.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 (8.45) now can be approximated as 4f L D = 1 (2χ − χ2 ) − 2χ − χ2 + f χ3 kM1 2 (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.47) similarly it can be shown that f (χ = 0) = 1 equation (8.5. Equation (8.48) yields 4f L D = χ (2 − χ) − kM1 2 (2 − χ) + f χ3 kM1 2 (8. Currently and ashamedly the author is looking for a more simplified explanation. .51) value of The pressure difference can be plotted as a function of the M1 for given 4f L D . and is based on estimates. 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 (8.52) A few observations can be made about equation (8.52). The current explanation is correct but based on hands waving and definitely does not satisfy the author. 8. 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.48) rearranging equation (8.49) and further rearrangement yields 4f L D = χ 2(1 − kM1 2 ) − 1 + kM1 2 χ + f χ3 kM1 2 (8.50) in cases that χ is small 4f L D ≈ χ 2(1 − kM1 2 ) − 1 + kM1 2 χ kM1 2 (8.

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). this model can provide a better understanding to the trends and deviations of the Fanno flow model. are 0. Combine this information with our case of 4f L D = 10 Lentrance = 250746268. ν . for most gas cases the speed of sound is about 300[m/sec]. The flow speed at the entrance is very large. a gas flows in a tube with 4D number is over 200.71. Nevertheless. For example. So.53) The typical values of the the kinetic viscosity. even with low temperature like 200K the speed of sound of air is 283[m/sec]. over hundred of meters per fL = 10 the required entry Mach second. The thermal entry length is in the order of the hydrodynamic entry length (look at the Prandtl number.6 5 see Figures and Tables Kays and Crawford “Convective Heat Transfer” (equation 12-12).06 UD ν (8. However. For example. 8. For this illustration. this topic will be covered in the next version because it provide some insight and boundary to the Fanno Flow model.0). the feeding diverging nozzle somewhat reduces the required entry length (as opposed to converging feeding). Remember from Fluid Dynamic book Lentrance = 0.146 CHAPTER 8. even fL for relatively small tubes with 4D = 10 the inlet speed is over 56 [km/sec]. .). Almost all the perfect gas model substances dealt with in this book. 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 . the speed of sound is a function of temperature. ISOTHERMAL FLOW assumption of the model and the model breaks.7 D On the other hand a typical value of friction coefficient f = 0. (0.37 which means that the maximum amount of energy is insufficient. 0 Now with limitation.005 results in Lmax 10 = = 500 D 4 × 0. Now. This requires that the entrance length to be larger than the actual length of the tub for air.0000185 kg/m-sec at 300K and 0. the hydraulic entry length is very large as will be shown below. value for the common gases. if Mach number is changing from 10 to 1 the kinetic energy change is T0 about T ∗ = 18.0000130034 kg/m-sec at 200K. In the supersonic flow.

00205 0.001 1.032 1.97344 0.91875 0.000896 0.011 1.1599 13.032 1. 7 Those who are mathematically inclined can include these kinds of questions but there are no real world applications to isothermal model with shock. In this model no questions about shock (should) exist7 .99485 0.10000 0.3002 1.08085 0.94894 0.87642 0.0736 10.20000 0.5366 1.7.2126 1.16552 0.30000 0.89644 0.88200 0.000 28.6109 8.043 1.99741 1.5644 9.98700 0.05000 0.8781 1.4086 1.9925 4.7230 2.9682 1.6651 13.03095 0.056 1.6903 1. and mass flow rate questions.07000 0.7040 66.043 1.1: The Isothermal Flow basic parameters 147 M 0.000 8.81000 0.8172 2.81879 0.3806 2. resistance (diameter.87586 0.009 1.2074 1.09000 0. etc.87544 0.011 1.9747 7.0 28.80000 0.2565 1.1129 1.87612 0.2668 0.6470 1.2258 3.70000 0.99232 0. friction coefficient.75000 0.25000 0.0736 10. 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).03000 0.91044 0.2074 1.6903 1.4515 4.1289 16.1718 21.4515 4.000 17.9031 14. ISOTHERMAL FLOW EXAMPLES Table 8. the “engineering” or practical questions can be divided into driving force (pressure difference).60000 0. .79 105.056 1.0495 1. friction factor.89 82.06000 0.50000 0.84515 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 785.8781 1.35000 0.0859 12.2258 3.4784 1.80732 0.92794 0.40000 0.000 0.45000 0.08000 0.87563 0.8.88594 0.007 1.89075 0.0255 1.0677 1.3906 8.021 1.9181 5. Here is an example.7 Isothermal Flow Examples There can be several kinds of questions aside from the proof questions6 Generally.9031 14.04000 0.8493 7.90300 0.87516 0.06 192.29895 0.4086 1.5644 9.).87675 0.1269 1.12 139.5920 6.1129 1.6500 5.0859 12.96075 0.5366 1.93800 0.4147 2.8650 3.4147 2.65000 0.1718 21.00626 0.8791 1.000220 0.1259 1.82758 0.1269 1.00371 0. The driving force questions deal with what should be the pressure difference to obtain certain flow rate.003 1. These questions or examples will appear in the later versions.33 279.3334 2.3906 8.98982 0.1289 16.005 1.87528 0.8172 2.97 439.3524 1.021 1.1827 1.55000 0.2553 10.0823 1.3806 2.83637 0.3002 1.50207 0.

148 CHAPTER 8. Hint: calculate the maximum flow rate and then check if this request is reasonable. 450 ∼ kg = = 23. the flow rate is really some what different. The exit Mach number is not necessarily 1/ k i.19 RT 290 × 300 m3 π × (0.9 4 sec The maximum flow rate then reads m ˙ = ρAU = 23.25 Utilizing Table (8.25 [m] diameter and 5000 [m] in length is attached to a pump. the velocity can be calcufL U2 function of lated by utilizing ∆P = 4D 2g .0 m ˙ = kg sec P1 P1 P1 kU kU A =√ = AkM1 A√ RT k c kRT kRT .00 20. 017. 4D 4f L D = 4 × 0. S OLUTION If the flow was incompressible then for known density.1: A tube of 0. In incompressible flow. It is more appropriate to assume an isothermal model hence our model is appropriate.31. ρ.1) or the program provides M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 0. the flow is not choked. R = 290 Kkg .04331 × √ m . What should be the pump pressure so that a flow rate of 2 [kg/sec] will be achieved? Assume that friction factor f = 0.62 ∼ = 16.5921 0. The density reads 1.31 × 290 × 300 ∼ = 14. check whether flow is choked (or even possible).0055000 = 400 0.25)2 kg × 14. surroundings temperature 27◦ C. The specific J heat for the gas. ISOTHERMAL FLOW Example 8. fL Calculating the resistance.1743 12.89446 The maximum flow rate (the limiting case) can be calculated by utilizing the above table. The velocity of the gas at the entrance U = cM = 0.62 sec ρ= P 2.0 0. k = 1. To solve this problem the flow rate has to be calculated as m ˙ = ρAU = 2.005 and the exit pressure is 1[bar].04331 400. First. It is note worthy to mention that since the isothermal model breaks around the choking point.e.19 × The maximum flow rate is larger then the requested mass rate hence the flow is not choked. the density is a √ the entrance Mach number.

4826 5.1) or using the provided program yields 8 It is unfortunate.68 The pressure should be 21. Thus. ISOTHERMAL FLOW EXAMPLES Now combining with equation (8.5844 0. 31 100000 × π×(0 4 P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 66.25)2 P2 Ak × 1 .31 P = 21.04014 466. S OLUTION At first. the minimum diameter will be obtained when the flow is choked. The friction coefficient for the tube can be assumed as 0.0845 M1 = M 2 P1 k 10 Now.103 . calculate the required diameter of tube.2: A flow of gas was considered for a distance of 0.7678 13.0 0. Mmax 1 1 P2 = √ = 0.6779 8. the maximum M1 that can be obtained when the M2 is at its maximum and back pressure is at the atmospheric pressure.3249 0.10300 4fL D P P∗ 2 × 337. with the value of M1 either by utilizing Table (8.68 P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ M 4fL D 0.5 [km] (500 [m]).0 0. A flow rate of 0.89567 The entrance Mach number is obtained by 4f L D 1 = 66. the maximum pressure allowed for the gas is only 10[bar]. Assume that the flow is isothermal and k=1.6779 + 400 ∼ = 466.89442 Note that tables in this example are for k = 1.40) yields m ˙ = M2 = M2 P2 Ak c 149 From Table (8.5 [in]8 .7.566[bar] Example 8.4826 = 2. but it seems that this standard will be around in USA for some time.).8.02 (A relative smooth tube of cast iron. Note that tubes are provided in increments of 0.1) or by utilizing the program M 0.76780 × 8. .4. You can assume that the soundings temperature to be 27◦ C.59 mc ˙ = = 0.2 [kg/sec] is required. Due to safety concerns.

4318 P P∗ P0 P0 ∗ 92.2 sec and the density is ρ= 1.150 M 0.2991 0.5 increments and the next size is 17[in] or 0.0 0. 000 kg P = = 11. 000.9110 6.4310 10.3[kg/sec] Since 50.02 × 500 94.43182 4 × 0.43 0.68[in] However.61 RT 287 × 300 m3 The velocity at the entrance should be m U = M ∗ c = 0.2424 0. The new 4D 4f L D = 4 × 0.87627 To check whether the flow rate satisfies the requirement m ˙ = 106 × π ×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.0018 6.0853 × √ 287 × 300 √ 1.2 the mass flow rate requirement is satisfied.431 the value of minimum diameter.6400 9.08450 With 4f Lmax D 4fL D P P∗ CHAPTER 8. It should be noted that P should be replaced by P0 in the calculations.08527 4fL D 92.02 × 500 0.4 × 287 × 300 ∼ = 347. the pipes are provided only in 0.08528 × 347. The only point that must be emphasized is that all properties (like Mach number. D= 4f L 4f Lmax D 4 × 0.42359[m] = 16.6 sec . pressure and fL is etc) have to be taken at the same point.87625 = 94.3 ≥ 0. ISOTHERMAL FLOW P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 94.64 ρ ρ∗ T0 T0 ∗ M 0.4318[m].0 0.2 ∼ = 29.4 ≈ 50. The speed of sound at the entrance is √ √ m c = kRT = 1.

52828. the information whether the flow is choked needs to be found.0419 .and P0 ∗T ∼ = 28 12. Assume that the flow is isothermal.8. and the average friction f=0. 4f Lmax D = 4 × 0. ISOTHERMAL FLOW EXAMPLES The diameter should be D= 4m ˙ = πU ρ 4 × 0. The gas and the sounding temperature can be assumed to be 300 K. It should be noted that the flow isn’t choked.21 × 0. k=1.2 ∼ = 0.21[bar] 20.72021 0.01 × 4000 = 400 0. guess reasonable value of M1 and calculate 4f L D . The flow rate is given with the diameter of the pipe.6 × 11.67 2. This situation is reversed question. for the sake of the exercise the other parameters will be calculated.52828 = 1. 1.17[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.01.027 π × 29.4.87531 P0 P0 ∗T 20.61 151 Nevertheless. Therefore. Calculate the Mach number at the entrance to pipe and the flow rate.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 0.19235 12. P2 = P0 ∗T P2 P0 ∗T = 2. S OLUTION First. The pressure at the exit (station (b)) is 2[bar].7.19235 ≈ 12.0419 400.67 The pressure at point (b) by utilizing the isentropic relationship (M = 1) pressure ratio is 0.4[m] diameter and 4000 [m] length to a different station (b). The solution is an iterative process.66915 From the table M1 ≈ 0. at first it will be assumed that the whole length is the maximum length. Example 8.3: A gas flows of from a station (a) with pressure of 20[bar] through a pipe with 0.

05000 0.83740 4f Lmax D 1 4f L D P2 P1 0.16922 0.83920 0.83889 0.10000 0.10000 0.10000 0.08978 0.12942 0.10000 0.0419 4 300 × 287 RT 42.40754 0. Compare the results of guessed pressure P2 with the actual pressure and choose new Ma number M1 accordingly.5914 0. 4.6483 0. Obtain M2 from the Table ? or by using the Potto–GDC.5807 0. P2 bear in mind that this isn’t the real pressure but based on assumption 5.6684 0.22 × 0. Calculate the pressure.152 2. there are no examples on isothermal with supersonic flow.84095 0.00000 0.36766 0.32131 400.7272 0.10000 0. Now the process has been done for you and is provided in Figure (??) or in the table obtained from the provided program.4 P k π × D2 M= √ m ˙ = ρAM c = √ π × 0.0419 The flow rate is √ √ 2000000 1.59338 4f Lmax D 1 4f L D P2 P1 400. ISOTHERMAL FLOW by subtracting 4f L D 1 − 4f L D 3.83997 0.05005 0. M1 0.32795 0.46[kg/sec] In this chapter. Calculate the value of 4f L D 2 CHAPTER 8.08971 0.5708 M2 0.10000 0.6934 0.84018 0.16912 0. M2 0.40737 0.12949 0.4: The flow parameters for unchoked flow M1 0.10000 .83827 0.32807 0.8 Unchoked situation Table 8.36780 0.10000 8.

8.8 = 0.10 Fri Feb 25 17:20:14 2005 Fig. UNCHOKED SITUATION 153 M1 isothermal flow 1 0.8.8 0.9 0.5 = 0.4 0.3: The Mach number at the entrance to a tube under isothermal flow model as a fL function 4D .2 0.7 0.5 0.1 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 = 0.3 0.8.6 0.2 = 0.

ISOTHERMAL FLOW .154 CHAPTER 8.

155 . Later.1)). etc.v. This model explains many industrial flow processes which includes emptying of pressured container through a relatively short tube. 9. exhaust system of an internal combustion engine.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. plicable to flow processes which are No heat transer very fast compared to heat transfer mechanisms with small Eckert Fig. This model is the second ¥¦¨§Ÿ¥ pipe flow model described here.± 9.³ can be ignored 1 . This model is apc. discussed on the ideal gas model and the entry length issues. It is advantages to examine the simplest situation and yet without losing the core properties of the process. 1 Even 2 Not the friction does not convert into heat ready yet. As this model raised from need to explain the steam flow in turbines.1: Control volume of the gas flow in a constant cross section number.1 Introduction Consider a gas flowing through a conduit with a friction (see Figure (9. compressed air systems. more general cases will be examined2 .

3) Again for simplicity. P = ρRT P2 P1 = → ρ 1 T1 ρ 2 T2 (9.8) 4 × Cross Section Area wetted perimeter (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 (9.6) 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. 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 . FANNO FLOW 9.7) (9.2) (9.1) 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.

6) yields A τw 2 m ˙ A − πD dP − πDdx f 4 1 2 ρU 2 = A ρU dU (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 (9.15) . 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 (9.9. 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 (9.3.3 Non–dimensionalization of the equations Before solving the above equation a dimensionless process is applied.13) 9. A and rearranging 4f dx D 1 2 ρU 2 −dP + = ρU dU (9.10) into momentum equation (9. s2 ≥ s 1 (9.11) by the cross section area.14) Utilizing the definition of the perfect gas results in M2 = ρU 2 kP (9.11) yields Dividing equation (9.10) By utilizing equation (9.2) and substituting equation (9.12) The second law is the last equation to be utilized to determine the flow direction.

5) and dividing the results by equation of state (9. (9.14) and substituting it into equation (9. Derivative of mass conservation ((9.14) and dividing by equation (9.19).158 CHAPTER 9.17). Thus.2)) results in dU U dρ 1 dU 2 + =0 ρ 2 U2 (9. the original limitations must be applied to the resulting equation. FANNO FLOW Using the identity in equation (9.14) yields d(M 2 ) d(U 2 ) dT = − 2 M U2 T (9. and (9.16) By further rearranging equation (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.20) Dividing the energy equation (9.18) The derivation of the equation of state (9.21) need to be solved.20).18).17) It is convenient to relate expressions of (dP/P ) and dU/U in terms of the Mach number and substituting it into equation (9.11) 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. These equa- .5) results dρ dT dP = + P ρ dT (9. (9.19) Derivation of the Mach identity equation (9.17). (9.21) Equations (9. (9.16) results in − dP 4f dx − P D kM 2 2 = kM 2 dU U (9.

9. causes the change in the other variables.25) results in 1 2 kM 2 1 + k− 4f dx dM 2 2 M = 2 2 M 1−M D (9.27) and (9.22). The dimensionless friction.22) The term dρ ρ can be eliminated by utilizing equation (9. 4D Combining equations (9.21) by eliminating dT /T . Then dρ/ρ and U are eliminated by utilizing equation (9. The only variable that is left is P (or dP/P ) which can be eliminated by utilizing equation (9.3.29) 1 dc k (k − 1)M 4 4f dx dT = =− T 2 c 2(1 − M 2 ) D (9.27) and the relationship for the temperature is density is obtained by utilizing equations (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).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 (9.22) 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 (9.20) and (9.25) (9.18) and substituting it into equation (9.24) (9.19) and (9.23) The second equation for Mach number.21) when eliminating dT /T results dP dρ (k − 1)M 2 dU 2 = − P ρ 2 U2 (9. fL the rest variables can be done in a similar fashion. 4D .18) to obtain dρ 4f dx kM 2 =− 2 ρ 2 (1 − M ) D (9. Explicit explanation is provided for only two variables.28) . sistance.23) kM 2 1 + (k − 1)M 2 4f dx dP =− P 2(1 − M 2 ) D (9. is chosen as the independent variable since the change in the dimensionless refL .18) and equation (9. M variable is obtained by combining equation (9.24) and results in 1 − M 2 dM 2 4f dx = 1 2 D kM 4 (1 + k− 2 M ) Rearranging equation (9.

FANNO FLOW (9. This pressure increase is what makes compressible flow so different from “conventional” flow. flow above speed of sound. (dT0 = 0).33) The first law requires that the stagnation temperature remains constant. flow with speed below the speed of sound. M > 1.32) 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 (9.24) it can be observed that the critical point is when M = 1.34). When M < 1 the pressure decreases downstream as can be seen from equation (9. For the same reasons.24) because f dx and M are positive.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. 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. 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 (9.24) through (9. the pressure increases downstream. from equation (9.35) (9. For example. Two.4 The Mechanics and Why the Flow is Choked? The trends of the properties can be examined by looking in equations (9.30) dP dT − R ln T P (9.31) The stagnation temperature expresses as T0 = T (1 + (1 − k )/2M 2 ).34) 9. Thus the discussion will be divided into two cases: One. in the supersonic branch. .

First.25) yields f dx = L (9. M Velocity.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. T Density.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.24)). This situation is called choked flow. the pressure has to be a monotonic function which means that flow cannot crosses over the point of M = 1. when the flow is supersonic.36) A representative friction factor is defined as ¯= f 1 Lmax 0 Lmax f dx (9. the pressure can be examined by looking at equation (9.26). 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. This constrain means that because the flow cannot “crossover” M = 1 the gas has to reach to this speed. dP (@M = 1) = ∞ and mathematically it is a singular point (see equation (9. On the other hand. THE WORKING EQUATIONS Why the flow is choked? 161 Here.9. M = 1 at the last point. the pressure decreases. ρ Stagnation Temperature. the explanation is based on the equations developed earlier and there is no known explanation that is based on the physics. For example. T0 9. 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 Trends The trends or whether the variables are increasing or decreasing can be observed from looking at the equation developed. However.37) . P Mach number.5. U Temperature. It demonstrates that the Mach number increases downstream when the flow is subsonic. For example. Observing from equation (9.

26) yields dP P dM 2 M2 =− 1 + (k − 1M 2 dM 2 1 2 M 2M 2 1 + k− 2 (9. For example.38) ¯ with f which is adopted in this book. (9. Now equation (9.29).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 (9.24).41) ρ 1 = ρ∗ M 1+ k−1 2 2 M k+1 2 (9.44) . M = 1 when P = P ∗ Equation (9.30) can be solved.39) can be integrated to yield: 1 P = ∗ P M k+1 2 k −1 2 2 M 1+ (9.43) 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∗ (9.42) U = U∗ ρ ρ∗ −1 =M 1+ k+1 2 k−1 2 2 M (9. Dividing equation (9. FANNO FLOW By utilizing the mean average theorem equation (9.23) is represented by 4D 4f L and Mach number.27). It is common to replace the f Equations (9. and (9.24) in equation (9. (9.29). (9.162 CHAPTER 9.28). fL . (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 (9.39) The symbol “*” denotes the state when the flow is choked and Mach number is equal to 1. Thus.24) can eliminate term D and describe the pressure on the Mach number.

45) And further rearranging equation (9. D The results of these equations are plotted in Figure (9.2: Various parameters in Fanno flow as a function of Mach number many cases shockless and therefore a relationship between two points should be .1 0.45) provides 1 P0 ∗ = P0 M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (9.47) discussion about Reynolds number and dimensionless 4f L friction parameter.40) and substituting it into equation (9.9.1 Fri Sep 24 13:42:37 2004 0. 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. 9.01 1 Mach number 10 Fig.46) The integration of equation (9.5.2) The Fanno flow is in P/P .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. THE WORKING EQUATIONS 163 Using the pressure ratio in equation (9.34) yields s − s∗ = ln M 2 cp k+1 1 2 1 + k− 2 M k+1 k 2M 2 (9.

In most times. ¾ 5 ¿ À form pipe with a diameter of 0.1: Û£ÜAÝ·Þß à Air flows from a reservoir and enters a uni´€µ·¶¸ ¶s¹©º »§¼ 0 ½ .164 CHAPTER 9. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0. Thus. the “star” values are imaginary values that represent the value at choking. 4f Lmax D = 2 4f Lmax D 1 − 4f L D (9.01451 1. the flow to the pipe inlet.1) pipe inlet is essentially isentropic. .004 and that the flow from the reservoir up to the Fig.Ä_ÅMÆ Ç mosphere. FANNO FLOW derived. Estimate the total temperature and total pressure in the reservoir under the Fanno flow model.90000 0. The air exits to the atÁÃÂ.6 Examples of Fanno Flow Example 9.3: Schematic of Example (9. There is no Mach meter.0934 0.49) Hence.0327 property is given only for academic purposes. the temperature and the total pressure at the pipe inlet are the same as those in the reservoir.9146 1.0089 1. S OLUTION For isentropic.05 [m] and áãâåäÑæ©ç è§é length of 10 [m]. finding the total pressure and temperature at the pipe inlet is the solution.94 .9) the following is obtained. 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 1.50) 9. The real ratio can be obtained by two star ratios as an example T2 = T1 T T ∗ M2 T T ∗ M1 (9. 9. For given Mach number (M = 0. the total temperature at the entrance can be fL obtained by knowing the 4D . With the Mach number and temperature known at the exit. Assume that the average friction factor to be f = 0.48) 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.

91484 1.7405 0.35886 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.35886 0. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.1699 Note that the subsonic branch is chosen.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 × × 3.9.5764 0.2100 3.5[K ] 1.21 either from Table 1 = 4f L D + 4f L D 2 = 4 × 0.35886 3.38814 1.915 T1 T ∗ T01 = T2 T T0 ∗ 2 T 1 T 1 1 1 =300 × × 1.05 4f L D The rest of the parameters can be obtained with the new (9.17 × 1.97489 0. the total temperature at the exit is T ∗ |2 = T∗ T T2 = 2 165 300 = 290.1) by interpolations or by utilizing the attached program.0327 0.12913 0.5922 0.7405 2. The stagnation ratios has to be added for M = 0.014 × = 2.91[Bar] 1.0327 4f L D To ”move” to the other side of the tube the 4f L D is added as 3.975 T∗ T 348K = 75◦ C Another academic question: . EXAMPLES OF FANNO FLOW So.93840 1.01451 0.6.004 × 10 + 0.0140 1.

0. The vessel conditions are at 29. 400 K. and CHAPTER 9. determine: (a) the mass flow rate through the system.5 [cm] in diameter and 1.2: A system is composed of a convergentdivergent nozzle followed by a tube with length of 2.0 [m] long. The system is supplied by a vessel.166 Example 9. (b) the temperature at the pipe exit.65 [Bar]. With these conditions a pipe inlet Mach number is 3. FANNO FLOW ó©ô2õö ÷Mø ùHúMû ü ýMþ ÿ  ¢¡¤£¦¥¨§©§¨ ¦. A normal shock wave occurs in the tube and the flow discharges to the atmosphere.

9. Take k = 1. S OLUTION (a) Assuming that the pressure vessel is very much larger than the pipe.2) (c) determine the Mach number when a normal shock wave occurs [Mx ]. therefore the velocity the vessel can be assumed to be small enough so it can be neglected. then the following can be found eith from Table (9.65[P ar] The mass flow rate through the system is constant and for simplicity point 1 is chosen which.005.1) or the Potto–GDC T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ M 3.357 × 400 = 142. m ˙ = ρAM c The density and speed of sound are unknowns and need to be computed.8K T01 .4: The schematic of Example (9.35714 0. T01 = 400K a P01 = 29.0000 The temperature is 0.02722 0.4. ¦  "!"#  shock d-c nozzle êãë§ìMí ìHîOïMð ñ_ò    atmosphere conditions Fig. Thus. the stagnat conditions can be approximated for the condition in the tank. Hence.11528 0. It is further assumed t the flow through the nozzle can be approximated as isentropic.65326 T1 = T1 T01 = 0. R = 287 [J/kgK ] and f = 0.07623 4. With the isentro relationship the Mach number at point one (1) is known.2346 0.

47519 Ty Tx ρy ρx Py Px P0 y P0 x 2.027 × 30 P01 0.3333 0.21822 4. and two. From the Table (9. (Fanno flow Table (9.42857 and the conditions of the tube are 4f L D = 4 × 0.47519. shock at the exit and comparing the pressure ratios. one.1)) Mx 3.1) or by using the Potto–GDC.52216 the flow is choked and with a shock wave.54[m/sec] The pressure at point 1 can be calculated as P1 = P1 P01 = 0. check whether the flow is shockless by comparing the flow resistance and the maximum possible resistance. if a shock exists.4 × 287 × 142.0000 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0. by comparing “possible” Pexit to PB .8 1.9640 0.005 × 1.8571 10.97 kg m3 (b) First.0000 My 0. EXAMPLES OF FANNO FLOW Using the temperature.6. the speed of sound can be calculated as √ √ c1 = kRT = 1.1)) . Two possibilities are needed to be checked. the shock at the entrance of the tube.50918 1.6790 3.2) m ˙ = 1.1 × 104 287 × 142.0252 × 3 × 239.025 Since 0.8 > 0.2346 0.0 = 0.54 = 0.9.81[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 (9.97 × kg π × 0.69 4 sec 8. First. The exit pressure determines the location of the shock. to obtain the following M 3. the possibility that the shock wave occurs immediately at the entrance for which the ratio for Mx are (shock wave Table (5.8 239.8 0.52216 0.32834 After shock wave the flow is subsonic with “M1 ”= 0.

2549 1. Thus.1912 0.2919 2. to obt the upstream Mach number. 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.47519 0.3904 0.0338 is smaller than the case in which shock occurs the entrance. 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.3904 1. 2) Calculate the entrance Mach assuming shock at the entrance.168 M 4fL D P P∗ P0 P0 ∗ CHAPTER 9. P P1 is needed. but be-careful second order methods like Newton-Rapson method can be unstable. . One possible way to find the e temperature. “claiming” upstream from exit through shock to the entrance.85676 1.2549 0. To check Secant Method. calculate the parameters for shock locat fL with known 4D in the “y” side.1481 The stagnation values for M = 0. With the location of shock.9640 0.65 = 0. FANNO FLOW ρ ρ∗ U U∗ T T∗ 0. 5 You can use any method you which.32834 × 1 2. 3) According your root finding algorithm5 calculate or guess the shock location and th compute as above the new M1 .89545 1. Then either by utilizing shock table or the program.8568 × 0.95679 0. T2 is by finding the location of the shock.47519 1. the shock is somewhere downstream.50917 1.12476 The actual pressure ratio 1/29. To find the location of the shock ra 2 of the pressure ratio.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. For example.47519 are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.

and. In the process Mx and My must be calculated (see the chapter on the program with its algorithms. M1 3.larc. experiments from many starting with 1938 work by Frossel7 has shown that the error is not significant. These results were obtained from the mathematical derivations but were verified by numerous experiments9 . In the Fanno model there is no heat transfer. Nevertheless.22019 0.7.0000 M2 1. The author seeks for a nice explanation of this concept for non–fluid mechanics engineers.8 Maximum length for the supersonic flow To insert example on the change in the flow rate between isothermal flow to Fanno Flow. Many of the pictures in the literature carry copyright statements. However.9. These changes include the choking point at lower Mach number.57981 (c) The way of the numerical procedure for solving this problem is by finding produce M1 = 3. It also must be recognized that there is a maximum length for which only supersonic flow can exist8 . 9 If you have experiments demonstrating this point. please provide to the undersign so they can be added to this book. 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.). It has to be noted and recognized that as opposed to subsonic branch the supersonic branch has a limited length. furthermore.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.).PDF 8 Many in the industry have difficulties in understanding this concept. Insert also example on percentage of heat transfer.9899 My 0. 7 See on the web http://naca.57910 4f L D 0. The information from the wall to the tube center6 is slower in reality.nasa. 9. up that will 9. repeat previous stage until the solution is obtained. . The only limitation of the model is its uniform velocity (assuming parabolic flow for laminar and different profile for turbulent flow. 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.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. on the comparison of the maximum length of isothermal model and Fanno Model.0000 4fL D up 4fL D down Mx 1. because the very limited amount of heat transformed it is closer to an adiabatic flow. the comparison with reality shows that heat transfer cause changes to the flow and they need to be expected.

 D .45 1.35 1.5 0.8 0.2 1.65 spesific heat.4 0. 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.5 1.3 1.7 0.5: The maximum length as a function of specific heat. 9. k Thu Mar 3 16:24:00 2005 Fig.5 1.5).6 0. k = 1. From the above analysis.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.3 0.170 as follows: CHAPTER 9.6 1. 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 → ∞.55 1.1 0 1.4 1. k 9.8215 The maximum length of the supersonic flow is limited by the above number.25 1.1 1 0.4) = 0. The maximum length in supersonic flow In Fanno Flow 1.4 1.3 1. k as shown in Figure (9. Even the flow in the tube starts as a supersonic in parts of the tube can be transformed into the subsonic 4fLmax maximum length.2 0.9 0.2 1.

6: The effects of increase of 4f L D on the Fanno line In the analysis of this effect. 4D . WORKING CONDITIONS 171 branch. Subsonic branch For converging nozzle feeding. no further increase of the exit Mach number can be achieved. Only a combination of these two parameters is truly independent.9. Once the Mach number reaches maximum (M = 1). it should be assumed that back pressure is constant and/or low as possible as needed to maintain a choked flow.43 5&687 9 ' ( Fig. The . the dimensionless fL friction. The entrance temperature increases as can be seen from Figure (9.102. increasing the tube length results in increasing the exit Mach number (normally denoted herein as M2 ). It is worth noting that entrance Mach number is reduced (as some might explain it to reduce the flow rate).7). In this process. The discussion has to differentiate between two ways of feeding the tube: converging nozzle or a converging-diverging nozzle. the treatment of the two branches are separated.1 fL Variations of The Tube Length ( 4D ) Effects C8DFE H¨I :<. However. and the pressure ratio. Three parameters. M1 . P2 /P1 are controlling the flow. the entrance Mach number.9.>=? ? I I X K<L N ?@BA G J MN U¨V V V W NOBP&Q8R4S T $&% )+*-.9. A shock wave can occur and some portions of the tube will be in a subsonic flow pattern. 9. 9. First. the mass flow rate decreases. all the three parameters can be varied and they are discussed separately here./.

Point b is the maximum possible flow for supersonic flow and is not dependent on the nozzle. 4D the flow becomes choked a different flow pattern emerges.e. referred here as the critical point c. Thus. Once less friction. Point a is the choking point (for the supersonic branch) in which the exit Mach number reaches to one. 9. results in a similar flow pattern as in the converging nozzle. to achieve supersonic flow.172 constant pressure lines CHAPTER 9. In the case of the converging–diverging feeding nozzle. is the point in which no supersonic flow is possible in the tube i.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. increase of the dimensionfL . Between these transitional points the effect parameters such as mass flow rate. The following ranges that has to be discussed includes (see Figure (9. At the starting point the flow is choked in the nozzle. in which no supersonic flow is possible in the entire nozzle–tube system. Supersonic Branch There are several transitional points that change the pattern of the flow. FANNO FLOW Y[Z \ 1’ 1 2 Fanno lines 2’ 1’’ 2’’ ] Fig. the shock reaches to the nozzle.8)): . These results are applicable to the converging nozzle. the mass flow rate must decrease. There is another point d. entrance and exit Mach number are discussed.” The density decrease because ρ = RT and when pressure is remains almost constant the density decreases.

The exit Mach number. reduces with the increase of 4D . 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 mass flow rate is constant because the flow is choked at the nozzle. 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.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.9. 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 number. it is a more practical point. fL M1 . fL The total maximum available for supersonic flow b − −b . M2 decreases (remember this flow is fL = 0) as M2 = M1 . At the end of the on the supersonic branch) and starts ( 4D range a. It is worth noticing that in the a − −c the mass flow rate nozzle entrance velocity max . However. Notice that exit Mach number. The entrance Mach continues to be constant and exit Mach number is constant. M2 is still one. In the range of a − b the flow is all supersonic. In the range b − c. M2 = 1. M1 is constant because it is a function of the nozzle design only. 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. The mass flow rate continues to be constant.

2 0. fL . On one hand.3 0. the flow rate decreases since (m ˙ ∝ M1 )11 .8 0. this situation is rather strange to explain. 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 0 0.7 0. the resistance increases and on the other hand.9 1 Exit Mach number Tue Oct 19 09:56:15 2004 Fig. To summarize the above discussion. 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.4 0.1 D = 1.174 CHAPTER 9.8 0.5 0.5 0. Figures (9.0 = 10.8) exhibits the development of M1 . 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. the exit Mach number remains constant and equal to one.1 0.6 0.7 Entrace Mach number 0.2 0.3 0. 1 1 .9: M1 as a function M2 for various 4f L D 10 On a personal note. This situation is because of the “double” choked condition in the nozzle. Somewhat different then the subsonic branch the mass flow rate is constant even if the flow in the tube is completely subsonic.0 = 100.9 0. 9. M1 as a function of M2 4fL  = 0. In this range. fL M2 mass flow rate as a function of 4D .6 0.0 Fanno Flow 1 0.4 0.

4 0. In Figure (9. at the tube entrance.10) exhibits the entrance Mach number as a function of the M2 .4 1. The Figure (9. one.5 2 1.9) exhibits the M1 as a function of M2 .4 Tue Jan 4 11:26:19 2005 Fig. shockless flow.5 0 0 0.10: M1 as a function M2 for different 4f L D for supersonic entrance velocity.9. and as can be observed. shock at the entrance.2 = 0. the larger 4D larger differences between exit Mach number for the different shock locations.2) by obtaining the 4f L for M2 and subtracting D 4f L the given D and finding the corresponding M1 .5 1 0.8 1 M2 1. The fL larger 4D larger M1 must occurs even for shock at the entrance.4 two extremes are shown. Obviously there can be two extreme possibilities for the subsonic exit branch. fL For a given 4D .1 shock = 0. shock at the exit. WORKING CONDITIONS 175 Figure (9. Fanno Flow 5 4.1 and D = 0. below the maximum critical length. the maximum . For D = 0.1  D = 0.4 = 0. The Figure was calculated by max utilizing the data from Figure (9.5 4 3. and three. the supersonic entrance flow has three different regimes which depends on the back pressure. One.6 1.8 2 M1 M1 as a function of M2 for the subsonic brench 4fL = 0.10) only for 4f L 4f L 4f L D = 0.2 shown with only fL creates shock at the exit only.9.2 0. tow.2 1. Obviously.6 0. when the shock wave occurs at the tube exit and two. Below.5 3 2. Subsonic velocity occurs for supersonic entrance velocity. 9.

the pressure ratio P P1 variations. 12 See more on the discussion about changing the length of the tube. . For very low pressure ratio the flow can be assumed as incompressible with exit Mach number smaller than < 0. Choking explanation for pressure variation/reduction Decreasing the pressure ratio or in actuality the back pressure. The Mach number.2 The Pressure Ratio. the entrance pressure decreases as well because of the increase in the entrance Mach number (velocity). If the feeding nozzle is converging than the flow will be only subsonic. f . If the nozzle is “converging– diverging” than in some part supersonic flow is possible.3. One case is where the 4D 4f Lmax length. Reynolds number et cetera) thus the flow remains on the same Fanno line. Mach number. for the flow to be continuous the pressure must decrease and for that the velocity must increase. the exit and entrance Mach fL numbers increase. The critical length is the maximum that associate with entrance D Mach number. as it was shown in Chapter (4). The only change will be at tube surroundings which are irrelevant to this discussion. 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. However. 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. results in increase of the entrance and the exit velocity until a maximum is reached for the exit velocity. The maximum velocity is when exit Mach number equals one. This jump is referred to as a choked flow. For cases where the supply come from a reservoir with a constant pressure. 9. 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 .9. P2 .176 critical length is mathematically 4f L D CHAPTER 9. Again a differentiation of the feeding is important to point out. Something must break since there are conflicting demands and it result in a “jump” in the flow. As the pressure ratio increases (smaller back pressure. can increases only if the area increase. In our model the tube area is postulated as a constant therefore the velocity cannot increase any further. P2 ). P1 effects In this section the studied parameter is the variation of the back pressure and 2 thus. is independent of the parameters such as. According to Fanno model the value of 4D is constant (friction factor.

Up to this stage the nozzle feeding the tube increases the mass flow rate (with decreasing back pressure). A back pressure that is at the critical point c results in a shock wave that is at the exit. 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.12) shows different pressure profiles for different back pressures. the tube is “clean” of any shock13 .9. Note however that exit Mach number.9. Between point a and point b the shock is in the nozzle.11: The pressure distribution as a function of 4f L D for a short 4f L D Short 4f L D Figure (9. 9. In this range and further reduction of the pressure the mass flow rate is constant no matter how low the back pressure is reduced. When the back pressure is below point c. Before the flow reaches critical point a (in the Figure) the flow is subsonic. Once the back pressure is less than point b the supersonic reaches to the tube. . The back pressure below point c has some adjustment as it occurs with exceptions of point d. 13 It is common misconception that the back pressure has to be at point d. M2 < 1 and is not 1.

This dissection deals only with the flow when it reaches the { ųÆÇÈ·ÉÊ¦Ë ÌmÏÍ Î for a long 4f L D .9. 9. M1 fL . However. 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. At point c or location of the shock wave. 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 reality these effects have significance and needs to be accounted for some instances.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.178 ‘ƒ’ & CHAPTER 9. is a function entrance Mach number. 9.3 Entrance Mach number. point c in this case is process as explained in the short 4D fL max < 4f L . Further reduction of the back pressure at this stage will not “move” the shock wave downstream the nozzle. effects In this discussion. the effect of changing the throat area on the nozzle efficiency is neglected. The and the “extra” 4D procedure is (will be) presented in later stage. M1 . The is no analytical solution for the location of this point c.

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. M1 is a function of the ratio of the nozzle’s throat area to the nozzle exit area and its efficiency. supersonic branch reached otherwise the flow is subsonic with regular effects. in this situation. the critical point is where result in the entrance Mach number.13: The effects of pressure variations on Mach number profile as a function of fL = 0.6 1.9. WORKING CONDITIONS 179 Mach number in Fanno Flow 4fL  D 2 1. 9.1 shock at 75% 50% 5% 0. 4f L D is equal to 4f Lmax D as a The process of decreasing the converging–diverging nozzle’s throat increases the .6 0. there are two different maximums for 4D . This analysis fL max deals with the case where 4D is shorter than total 4f L .2 0 0 0.4 0.3 for Fanno Flow when the total resistance 4D 4f L D The entrance Mach number. Practically. first is the total 4f L maximum D of the supersonic which depends only on the specific heat.15 4fL  D 0.05 0.8 1. k . and second the maximum depends on the entrance Mach number.25 Tue Jan 4 12:11:20 2005 Fig.4 Mach Number 1. This effect is the third parameter discussed here. fL As was shown before. M1 . the nozzle area ratio is changed by changing the throat area.2 0. D Obviously.2 1 0.8 0.9.

4 4 3. 9. decrease of the throat area results in flushing the shock out of the tube. The reference to the tube is because it is the focus of the study.2 0. By doing so. The process of decreasing the throat area also results in increasing the pressure drop of the nozzle (larger resistance in the nozzle15 )16 .4 0 0 0.6 1. The mass flow rate is proportionally linear to the throat area and therefore the mass flow rate reduces. The “payment” for increase in the supersonic length is by reducing the mass flow. 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. 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. Decreasing the nozzle throat area moves the shock location downstream.2 0. If the tube contains no supersonic flow then reducing the nozzle throat area wouldn’t increase the entrance Mach number.8 0. area) decreases the flow rate while in a different way (increasing the 4D .8 4.25 5% 50 % 75 % Fri Nov 12 04:07:34 2004 Fig.8 2.6 3. the throat area decreases. FANNO FLOW P2/P1 Fanno Flow 4fL  D 4.2 P2/P1 2.1 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.15 4fL  D 0.05 0. Further.4 2 1.3 entrance14 Mach number.180 CHAPTER 9.14: Fanno Flow Mach number as a function of 4f L D when the total 4f L D = 0.

9. WORKING CONDITIONS üþý ÿ ¡ ¢¢¢ ¢¢¢ ¢ £¥¤§¦©¨     .9.

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

17: The maximum entrance Mach number. FANNO FLOW shock location is at the maximum length. To overcome this numerical problem it is suggested to start the calculation from distance from the right hand side. the Mach number equal infinity when left side assumes result in infinity length of possible extra (the whole flow in the tube is subsonic). ii) Calculate the extra (at the entrance). then shock at Mx = 1 results in My = 1. Let denote ∆ Note that 4f L D sup 4f L D ¯L f − = 4D actual 4f L D max∞ 4f L D sup (9. M1 to the tube as a function of sonic branch super- From numerical point of view.182 CHAPTER 9. i) Calculate the extra the max length). 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 . 9. cRd3ePf g 1 4f L D max∞ ^`_Ia b 4f L D Fig. 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. The proposed procedure is based on Figure (9.51) is smaller than 4f L D retreat .16).

The left side is when the shock occurs at Mx 8.9.3591 0.52) Figure (9. Example 9.9 S OLUTION max for k = 1.821508116. The maximum 4f L D 4f L 4f L is 0.3867 5. Accordingly continue.00849 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 2.442 − 0. 4f L D = 0.9 − 0.5000 0. In a summary of the actions is done by the following algorithm: (a) check if the (b) Guess 4f L D is 2.39289 Ty Tx ρy ρx Py Px P0y P0 x 13. M1 reduces after the maximum length is exceeded.4 The solution is obtained by an iterative process. thus the extra tube is ∆ 4f L D fL = 0.3: Calculate the shock location for entrance Mach number M1 = 8 and for assume that k = 1. Or in a very long tube the whole flow will be subsonic.17) shows the entrance Mach number.9.1318.76820 (flow is choked and no additional 4D ). The retreating length is expressed as subsonic but 4f L D retreat = 4f L D max∞ − 4f L D sup (9.15). 4f L D up = − 4f L D max 4f L D up . .39289 The extra ∆ 4f L D My 0.1641 between the negative of left side to the positive of the right side17 . D exceed the maximum length D for this entrance fL Mach number.0000 With (M1 ) M 0.5652 74. (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.6136 2. Hence.4417 2.4 (Mexit = 1).76820 = 0.1318.76820.3102 Now the solution is somewhere exceeds the maximum 4f L D 4f L D max for the supersonic flow.42390 1.1318 = 2. WORKING CONDITIONS 183 flow is supersonic see 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. Hence. The maximum for M1 = 8 is 4D = 0. the value of left side is −0.7461 1.

69119 0.6706 My 0.31783 0.64830 0.90000 0.90000 0.22574 0.32816 0.90000 0. .90000 0.90000 0.3838 1.57070 0.32878 0.6706 1.58217 0.90000 0.57184 0.65728 0.6706 1.90000 0.5286 1.32932 0.64830 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 1.6673 1.6699 1.90000 0.64872 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). Shock location are: M1 8.64832 0.90000 This procedure rapidly converted to the solution.32682 0.90000 0.62170 0.90000 0.6635 1.90000 0.6382 1.6705 1.64830 0.0000 M2 1.64850 0.66779 0.57068 0.64830 0.64839 0.64830 4fL D 0. If it is satisfactory stop or retu 4fL D up 4fL D down Mx 1.57318 0.90000 0.184 CHAPTER 9.32921 0. FANNO FLOW (d) Calculate the associate Mach number.57068 0.32932 0.6021 1.32932 0.64920 0.57068 0.6706 1.6706 1.27830 0.57069 0.32907 0.90000 0.32927 0.64834 0.64831 0.57122 0.57093 0.32932 0.32931 0.90000 0.6554 1.6703 1. Mx with the Mach number.90000 0.6691 1.65023 0.90000 0.57605 0.0000 4f L D down against the old one.90000 0.57073 0.57068 0.57079 0.57068 0.30494 0.6706 1.64831 0.6706 1.32932 0.32930 0.57068 0.65246 0.74664 0.6706 My 0.32395 0.59506 0.67426 0.

18) the dominating parameter is 4D .8 iso = 0.1 = 0.10. 4D .1 iso = 0. THE APPROXIMATION OF THE FANNO FLOW BY ISOTHERMAL FLOW185 9.99 which requires that enough for M1 < 0. friction.2 0. The results are very similar for isothermal flow.3 M1 0. When the Mach number decreases the temperature approaches the stagnation temperature (T → T0 ).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.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. This fL requirement provides that somewhere (depend) in the vicinity of 4D = 25 the fL flow can be assumed isothermal. The only difference is in small dimensionless fL .5 = 0. 9. Hence the mass flow rate is a function of 4D because M1 changes.2) or the results from computer program attached to this book shows that reduction of the mass flow is very rapid.4 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0. to insert a question or example about this issue in end with comperison to Isothermal Flow 0. Hence. This tolerance requires that (T0 − T )/T0 = 0.9.1 0 0 10 20 30 40 Wed Mar 9 11:38:27 2005 50 4fL  D 60 70 80 90 100 Fig. say about 1% that flow can be assumed to be isothermal.2 = 0.67. One must notice that the maximum temperature at the entrance is T0 1 . Looking at the table or Figure (9. if one allows certain deviation of temperature.15 even for large k = 1.8 M1 Fanno flow 0.

99683 4.5393 4. Hence.99097 0.5910 4.5.98874 0. T ≈ T0 and is the same for the pressure.7569 67.05.0000 40. or accurately by utilizing the program as in the following table.07975 T T0 A A P P0 0.0 2.18) consider the following example. = 4 × 0.0000 0.4: To demonstrate the utility in Figure (9. ρ ρ0 M 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.22697 0.98928 0.2842 0. S OLUTION First calculate the dimensionless resistance. 0.99195 0. The mass flow rate is m ˙ = P AM π × 0. the mass rate is a function of the Mach number.02 From Figure (9.1697 50.1196 7.99873 0.127 × RT 4 1.80000 Only for the pressure ratio of 0. 0.30000 0.6523 5.48 287300 kg sec .0000 40.11392 0.11 More Examples of Fanno Flow Example 9.8.02[m] and pressure ratio P2 /P1 = 0.4 ≈ 0.7027 5.40790 0. 4f L D 4f L D .3.1 the flow is choked.11637 0.022 k = 300000 × × 0.12728 0.99692 0.186 CHAPTER 9. FANNO FLOW 9. Find the mass flow rate for f = 0.07975 M2 1.13 etc.12728 0.12420 0.99556 4.05 × 4 = 40 0.0733 7.4206 0.42 0.1697 10.99677 0. D = 0.1. M1 0.0000 40. 0.99233 0.0000 42. The stagnation conditions at the entrance are 300K and 3[bar] air.50000 0.7569 107.2519 A×P A∗ ×P0 Therefore.99354 0.1 M1 ≈ 0.99741 0. L = 4[m].0000 40.12420 0.11392 0.18) for P2 /P1 = 0.09965 4fL D 4fL D 1 4fL D 2 P2 P1 40.

1994 1.77894 0.2508 9.1991 1.1985 1.00328 0.08758 0.9.50918 0.52216 36.1618 6.3817 14.0485 1.66667 0.60 0.1988 1.0693 1.27217 0.9341 1.30500 0.07664 0.0351 3.08 0.32459 0.01451 0.55 0. THE TABLE FOR FANNO FLOW and for the rest m ˙ 0.1378 4.95781 1.4182 10.000 0.07229 0.37879 0.1628 1.05476 0.0474 11.1852 1.95 1.2838 1.30 m ˙ P1 0.43 P1 0.4815 21.43133 0.0 0.49082 0.48 × = 0.61237 0.1315 1.66 106.91460 0.05 0.72 83.9635 4.48326 0.0327 1.5664 1.000 0.07975 P2 = 0.03633 0.4487 2.8 ∼ 0.73179 0.4935 1.633 1.2915 13.1273 0.1512 9.1788 1.1533 1.1998 1.4626 1.35 0.1065 1.50 0.6183 1.04381 0.3184 2.964 1.0787 1.32572 0.1976 1.1273 kg sec kg sec kg sec 187 9.70 0.85 0.017 1.4525 2.0944 1.4554 2.2620 15.1: Fanno Flow Standard basic Table M 0.10944 0.4027 3.5914 18.08 440.40825 1.03286 0.68374 0.12728 0.0207 1.65 0.3865 1.7780 2.06570 0.3398 1.2047 1.2616 12.48 × = 0.3665 1.1882 1.20814 0.40 0.2549 1.06 0.4834 5.4961 66.5753 1.00 1.45 0.6843 7.21822 0.688 0.0691 0.1242 P2 = 0.9034 11.02 193.1996 1.1291 1.58506 0.5826 3.3 ∼ 0.90 0.5901 2.6416 8.04 0.3005 27.0929 1.42857 .002 1.7092 1.00 2.6191 2.12.044 1.21822 4.87037 0.4613 10.2119 1.0702 2.9216 14.00000 1.2993 3.03 0.80 0.48 × = 0.82514 0.72805 0.07 0.1713 1.53452 0.03 140.63481 0.6958 1.8218 5.4318 22.1429 1.5 ∼ 0.1981 1.1194 1.3085 1.1356 1.5116 19.1489 1.0089 1.6400 2.12 The Table for Fanno Flow Table 9.00 3.3546 2.6742 3.3848 1.5333 8.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 787.0624 1.2893 1.0934 1.061 1.1381 1.1273 0.75 0.7634 1.10 0.8254 18.468 P1 0.35 280.09 0.20 0.235 30.09851 0.8708 1.1905 1.1139 P2 m ˙ = 0.6659 15.25 0.9435 5.0922 1.0382 1.2200 13.000 0.30 0.0638 1.

40988 0.00 65.6E +6 0.00374 0.440 2.40849 0.00198 0.82008 0.81265 0.00 40.43568 0.443 2.0E +2 0.42857 0.72988 0.023905 5.4E +2 0.81755 0.20000 0.3E +2 0.000500 7.14634 0.46771 0.42390 0.446 2.448 2.5E +4 0.81582 0.00 7.00663 0.390 2.00952 0.000579 5.00271 1.82066 0.00609 1. FANNO FLOW Table 9.5E +5 0.9E +2 0.447 2.00 6.00 25.00488 0.00 20.18 0.3E +6 0.063758 53.82078 0.000979 1.00 10.069767 0.40846 2.377 2.446 2.00 50.00153 4.13363 10.11111 0.1E +5 0.00240 0.8E +6 0.82033 0.6E +5 0.44721 0.333 2.40875 0.6E +4 0.00122 .295 2.000680 3.63306 0.00 70.40853 0.089443 25.40866 0.28571 0.00 35.40938 0.448 2.40889 0.448 0.00 5.014815 0.42066 0.434 2.000809 2.78683 0.057143 0.77899 0.138 2.00 60.00200 2.445 2.036860 1.00 30.69380 0.00 55.00 8.81860 0.00 45.00390 4.00 0.40859 0.81928 0.4E +6 0.00166 0.81975 0.1: Fanno Flow Standard basic Table (continue) M 4.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.8E +5 0.188 CHAPTER 9.72 0.5E +6 0.82052 0.029348 3.00296 0.447 2.00 9.00121 8.047619 1.00142 0.236 2.359 2.75280 0.41833 0.086957 0.76819 0.41079 0.40908 0.

10.CHAPTER 10 RAYLEIGH FLOW Rayleigh flow is (frictionless) flow with heat transfer through a pipe of constant cross sectional area. This model usage is to have a rough predict the conditions occur mostly in 189 . This model applied to case where the heat transfer is significant and the friction can be ignored. the heat transfer in Rayleigh flow occurs either between unknown temperature to tube and the heat flux is maintained constant. In practice Rayleigh flow is really provide good model for the real situation. a simple model is built around assumption of constant properties (poorer prediction to case were chemical reaction take palace). Rayleigh flow is practical and useful concept in a obtaining trends and limits. This fact create situation different compare to the previous two models. 10.1 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 opposed to the two previous models. Yet.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. The density and pressure change due to external cooling or heating. the heat transfer can be in two directions not like the friction (there is no negative friction). As before. 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. While the isothermal flow model has heat trans.Fig. This flow referred in the literature as Rayleigh Flow (see historical notes). In contrast.

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) are similar to the equations that were solved for the shock wave. a solution can be obtained.190 CHAPTER 10. One can notice that equations (10. As appose to Fanno flow in which the resistance always oppose the the flow direction. Nevertheless.4) can further assist in obtaining the temperature ratio as T2 P2 ρ 1 = (10. (10. one has to be aware that properties do change significantly for a large range of temperature.2 Governing Equation Q = Cp (T0 2 − T0 1 ) (10. 10. The flow velocity acceleration change the direction when the cooling is applied. Yet. are encapsulated in this model. for smaller range of temperature and length the calculations are more accurate. if the upstream conditions are known (or downstream condition are known).5) The equation of state (10. in Rayleigh flow also cooling can be applied. the main characteristic of the flow such as chocking condition etc.1) (10. The density change though the heat transfer (temperature change). P2 1 + kM1 2 = P1 1 + kM2 2 (10.7) U2 ρ1 = = ρ2 U1 kRT2 U1 kRT1 . In analysis of the flow. Thus.2) (10. The basic physics of the flow revolves around the fact that the gas is highly compressible.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 (10.3) and (10.2). RAYLEIGH FLOW situations involve chemical reaction.4) A(P1 − P2 ) = m ˙ (V2 − V1 ) ρ 1 U1 A = ρ 2 U2 A = m ˙ There are four equations with four unknown.

10.10) (10.10.5) and (10. GOVERNING EQUATION Substituting equations (10. and s2 .9) {N|~ }€ y ‰XŠˆ‹ ne ŒŽˆŒ N‚„ƒ Pr es su re li co ns ta …X†ˆ‡ nt z Fig.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 (10.11) results in 2(1 − M 2 ) 1 ds = Cp dM M (1 + kM 2 ) (10.2. 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 (10.11) Let the initial condition M1 .7) into equation (10.6) yields T2 1 + kM1 2 M2 = T1 1 + kM2 2 M1 T2 T1 191 (10.2: The Temperature Entropy Diagram For Rayleigh Line The Rayleigh line exhibits two possible maximums one for dT /ds = 0 and for ds/dT = 0. The second maximum can be expressed as dT /ds = ∞ The second law is used to find the expression for derivative. and s1 are constant then the variable parameters are M2 .12) . A derivative of equation (10.

RAYLEIGH FLOW Take the derivative of the equation (10. From thermodynamics.14) (10. It is convent to referrers to the value of M = 1.13) On T-s diagram a family of curves can be drawn for a given constant. The derivative is equal to infinity.16) U∗ ρ1 = = ∗ ρ U1 1 The √U kRT ∗ kRT ∗ √ √ U1 kRT1 kRT1 ∗ √ = 1 M1 T∗ T1 (10. some what similarly to Fanno flow. The equation (10. dT /ds = ∞ when M = 1.192 CHAPTER 10. The transition to supper sonic flow occurs when the area changes. The derivative is equal to zero when 1 − kM 2 = 0 or M = 1/ k or when M → 0. several observations can √ be generalized.15) (10. and M2 results in dT 1 − kM 2 = constant × 3 dM (1 + kM 2 ) Combining equations (10.17) star is an asterisk. chocking can be explained by the fact increase of energy must accompanied by increase of entropy. Hence.12) when letting the variable parameters be T2 . increase of heating results in increase of entropy. Yet. And cooling results in reduction of entropy. 1 + kM1 2 P∗ = P1 1+k The temperature ratio is 1 T∗ = 2 T1 M 1 + kM1 2 1+k 2 (10. yet note this point is not the choking point. . But the entropy of supersonic flow is lower (see the Figure (10.5) can be written between chocking point and any point on the curve. The chocking is occurred only when M = 1 because it violate the second law.13) by eliminating dM results in M (1 − kM 2 ) dT = constant × ds (1 − M 2 )(1 + kM 2 )2 (10. The peculiars point of M = 1/ k when additional heat is applied the temperature is decreasing. The derivative is negative.). when cooling applied to a tube the velocity decreases and heating applied the velocity √ increases. These value referred as the “star”1 values. dT /ds < 0. Yet for every curve.2)) and therefore it is not possible (the maximum entropy at M = 1.12) and (10.

367 2.07 0.13793 0.10 0.10.114 1.596 1.423 1.094 1.235 1.86204 0.69751 0.61393 0.388 2.94009 0.65 0.262 1.178 1.3 Rayleigh Flow Tables The “star” values are tabulated in Table (10.51001 0.011922 0.264 1.43894 0.20661 0.020529 0.00431 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 (10.03 0.08 0.199 1.1).70804 0.45 0.1: Rayleigh Flow k=1.99814 2.90850 0.011680 0.80 0.015 0.397 2.81892 0.157 1.44444 0.261 1.75 0.81013 0.025 1.00917 0.011 1.075 1.001 0.00860 0.99290 1.20 0.00216 0.019 1.014300 0.379 2.005 1.17355 0.4 M 0.384 2.870 1.135 1.00765 0.090909 0.508 1.04 0.30 0.31373 0.25096 0.045616 0.265 1.273 2.25 0.85 0.027841 0.058 1.015224 0.060 1.218 1. Several observations can be made in regards to the stagnation temperature.96081 0.95 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.95693 .86833 0.030215 0.3.61515 0.267 1.131 2.52903 0.79012 0.259 1.207 2.343 1.40887 0.056020 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 (10.030 1.51413 0.023669 0.09 0.019222 0.75991 0.34686 0.043 1.35 0.046777 0.029 1.55 0.00598 0.266 1.60 0.06 0.85987 0.961 1.038075 0.266 1.50 0.266 1.00517 0.30440 0.025 1.91670 0.70 0.99207 0.373 2.05 0.57447 0.25684 0.90 0.778 1.193 1.91097 0.19183 0.014 1.395 2.96395 0.69136 0. Table 10.686 1.98097 0.049 1.40 0.125 1.017119 0.63713 0.19) 10.00383 0.75524 0.392 2.37865 0.036212 0.023223 0.

340 1.00 0.8E +2 2.65377 0.0 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 1.00274 0.0 9.713 .0 25.455 1.58909 0.227 12.580 1.71006 0.873 3.240 1.00240 1.3E +2 3.1 2.47562 0.034483 0.0 6.017021 0.675 1.43122 0.695 1.88419 0.713 1.078487 0.65398 0.081772 0.146 1.328 8.89087 0.5 8.743 1.16831 0.045 2.092719 0.194 CHAPTER 10.0 20.026490 0.52356 0.00326 0.63 27.00 1.5 2.023495 0.4 2.0E +2 3.13540 0.96031 0.019 1.97872 0.028972 0.3 2.7 2.0E +4 1.588 1.510 1.50885 0.2E +4 8.57831 0.525 1.56982 0.0 5.2 2.95798 0.68494 0.49122 1.00140 1.418 2.30864 0.49259 0.28551 0.475 1.52893 0.9 2.10256 0.22936 0.0 8.35561 0.503 1.039883 0.424 5.686 1.3 1.51647 0.00469 0.667 1.7E +5 1.005 1.50 18.1 1.616 1.701 1.066667 0.538 1.72421 0.85971 0.28028 0.656 1.0E +2 1.039900 0.41 1.403 1.0 7.256 1.6 1.26478 0.68 75.067263 0.222 2.561 1.044910 0.0 2.693 1.301 1.055363 0.99392 0.29687 0.55556 0.52970 0.21 38.044 1.51349 0.641 1.00428 0.75250 0.21417 0.39643 0.11111 0.00 0.032053 0.018846 0.69700 0.4E +2 1.00732 0.0 35.0 3.54473 0.43353 0.5 6.56734 0.634 2.6 2.20040 0.31486 0.5 9.17647 0.690 1.80539 0.21419 0.1E +4 3.058264 0.90928 0.136 3.702 1.030094 0.050943 0.49356 0.316 1.431 1.7 1.205 1.93425 0.550 1.24615 0.95 54.5 7.1: Rayleigh Flow k=1.40384 0.60894 0.886 2.67380 0.18788 0.000 1.712 1.00190 0.620 1.49415 0.0 30.494 1.50702 0.8 2.2 1.711 1.36364 0.4 1.37870 0.91185 0.66350 0.73954 0.122 1.70174 0.176 1.52004 0.75613 0.49174 0.8E +2 1.9 3.33454 0.71301 0.571 1.33439 0.405 1.85917 0.046693 0.81414 0.4 (continue) M 1.83628 0.00 0.698 1.5 5.61580 0.5 1.078 1. RAYLEIGH FLOW Table 10.0 1.13223 0.64103 0.020979 0.5 10.79339 0.0 4.8 1.53633 0.681 1.078 1.46106 0.52438 0.5 4.79576 0.51098 0.035650 0.699 1.77406 0.375 1.

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

International Textbook Co.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. However. Scranton. and “Compressible Fluid Flow. these books contained only limit discussions on the evacuation of chamber with attached nozzle. 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 . 1964. 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. this example can further strength and enhance this conviction. A. 11. Prentice Hall. 1985. the situations. It Isothermal model for relatively long tube was the undersigned experience..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 . by M. Fundamentals of Gas Dynamics. A. J. in life. Pennsylvania. in many instances. However.” 2nd Edition. 2 Even if the instructor feels that their students are convinced about the importance of the compressible. These books are OCR.Fig.. There is a significant impor. are far Fanno model 1 After completion of these Chapters. Saad. 201 . the undersigned discover two text books which to include some material related to this topic.

were the chamber’s volume is a function of external forces. Addition- . at this stage. 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. are presented (see Figure (11. were the chamber volume is controlled or a function of the pressure. Fanno flow model is the most appropriate in the case where the filling and Control volume for the filling case evacuating is relatively fast. most car today equipped with an airbag. In case the fL filling is relatively slow (long 4D than the Fig. 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. Yet umes used in this model as it was stated before. should be larger than 400. And therefore. For example. 4D . models.3: A schematic of the control volIsothermal flow is appropriate model. EVACUATING AND FILLING A SEMI RIGID CHAMBERS more complicate. 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. here Isothermal flow and Fanno flow are used as limiting or bounding cases for the real flow. In the first Chapter models.1)).202 CHAPTER 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. no discussion is ofÉÊIËKÌ ÍÂÎ Ï Ð Ñ(Ò Ó fered here.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.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. The analysis is extended to include a semi rigid tank. are discussed. the models in this Chapter are suitable for study of the filling the airbag or filling bicycle with air. 11. 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. 11. Yet Isothermal flow model is used as the limiting case. As it was shown in ChapfL ter (9) the resistance. This author isn’t familiar with situations in which Control volume for the evacuating case Rayleigh flow model is applicable. For instance. 11. In the second Chapter.

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)

29) converted into ¯ P 1 (11.17) “shrinks” and doesn’t contain the relative volume term. After the integration.22). the differential temperature change in the chamber is slower than the temperature change in nozzle. is constant and equal one for a completely rigid The relative Volume.4. 11.1 Rapid evacuating of a rigid tank With Fanno Flow ¯ (t) = 1. f [M ] is expressed by equation (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) (11.31) 11. equation (11.4 11.30) Here.4. V tank.3. the general equation (11. Yet. In such case. For rigid volume. The temperature in the nozzle is changing because the temperature in the chamber is changing.29) ¯ dP ± f [M ] ¯ P ¯ t ¯= 0 dt 0 (11. Owczarek [1964] found an analytical solution which described here.2 Isothermal Nozzle Attached In this case the process in nozzle is assumed to isothermal but the process in the chamber is isentropic. A reasonable model for the tank is isentropic (can be replaced polytropic relationship) and Fanno flow are assumed for the flow in the tube.11. the specific governing equation is ¯ −1 dP ¯ [M ]P ¯ 3k 2k =0 − k Mf ¯ dt (11.32) . 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.13) is reduced into V ¯ dP ¯=0 = ±f [M ]P ¯ dt Separating the variables and rearranging equation (11. equation (11. ¯ = 1 and for isothermal nozzle T ¯ = 1 Thus. Thus.

34) ¯ is a function of P and can It has to be realized that this is “reversed” function i. The Figure (11.4 0.6 0.6 0.2 0.5: The reduced time as a function of the modified reduced pressure The big struggle look for suggestion for better notation.36) .33) results in ¯= t −k 2 ¯ 12 k − 1 P ¯ (k − 1)M f [M ] (11. that when the modified reduced pressure equal to one the reduced time is zero.33) direct integration of equation (11.5). Mmax ¯ = 1.e.8 V(t) = P (t) V(t) = P (0) 0.2 0 0 0. 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS For a choked flow the entrance Mach number to the tube is at its maximum.35) The function is drawn as shown here in Figure (11.210 CHAPTER 11. But for the chocked case it appears as ¯ ¯ = 1 + (k − 1)M f [M ] t ¯ P 2 2k 1−k (11.0 P(t) 0.5) shows 1.32) 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 (11.34) 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 (11.4 ¯ t 0. t be reversed for case. The solution of equation (11.8 1. The transition point is denoted here as chT . equation (11. The reduced time increases with decrease of the pressure in the tank. At certain point the flow becomes chokeless flow (unless the back pressure is complete vacuum).0 Fig. Thus.

Example 11. with today computer this should not be a problem and easily can be drawn as shown here in Figure (11.4. equation .0[m] S OLUTION The first parameter that calculated is 4f L 4f L D D =5 11. The transition point is denoted here as chT . At some point the flow becomes chokeless flow (unless the back pressure is a complete vacuum). The reduced time increases with decrease of the pressure in the tank. when the flow became chokeless the entrance Mach number does not change much from the choking condition.2 Filling Process ¯ −1 dP ¯ ¯ 3k 2k =0 ¯ − k Mf [M ]P dt The governing equation is (11. The solution of equation (11. Nevertheless.01[m] and length of L = 5.5). The chamber is connected with a rubber tube with f = 0. 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. The Figure shows that when the modified reduced pressure equal to one the reduced time is zero.39) It has to be realized that this is a reversed function. Again.025.37) For a choked flow the entrance Mach number to the tube is at its maximum. for the special cases where the choked equation is not applicable the integration has to be separated into zones: choked and chokeless flow regions.38) direct integration of equation (11.38) results in ¯= t −k 2 ¯ 12 k − 1 ¯ [M ] P (k − 1)Mf (11.4. unless extra long time or extra low pressure is calculated/needed.11.1[m3 ] is filled with air at pressure of 10[Bar]. Further.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. Mmax ¯ = 1. And in the choke region the calculations can use the choking formula and numerical calculations for the rest.2: A chamber with volume of 0. d = 0. Thus.

40) 11.6: The reduced time as a function of the modified reduced pressure (11. The combination of the For Isothermal process.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 ] (11.4 0. the relative temperature.6 0. EVACUATING AND FILLING A SEMI RIGID CHAMBERS 1 V(t) = P(t) V(t) = V(0) 0. 11.2 0.3 The Isothermal Process ¯ = 1.8 0. T isentropic tank and Isothermal flow in √the tube is different from Fanno flow in that the chocking condition occurs at 1/ k . It has to be remembered that the chamber can undergo isothermal process.2 0 à™áPâäã3å¯æ(ç èˆéwêìë3í ð î ï 0.4.4 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.212 CHAPTER 11.41) . For the double isothermal (chamber and tube) the equation (11.8 1 or ñ™òIóõô/ö"ñø÷ùòP ô/ö¯ü(ý þ ÿ úû ò¡  Fig.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) (11.6 0.

5 The “Simple” General Case The relationship between the pressure and the volume from the physical point of view must be monotonous. RAPID EVACUATING OF A RIGID TANK 213 11. Substituting equation (11. in the known situations to this author pressure increase results in volume decrease (at least for ideal gas. In this analysis and previous analysis the initial effect of the chamber container inertia is neglected.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. The analysis is based only on the mass conservation and if unsteady effects are required more terms (physical quantities) have taken into account. This situation occurs at least in small ranges for airbag balloon etc.46) The physical meaning that the pressure remains larger thorough evacuating process.4. it is assumed the ideal gas applied to the gas and this assumption isn’t relaxed here. Nevertheless.43) (11.44) dt = 0 k 1+k ¯ P 1 −1 ¯ k2 ¯ k dP P ¯ M f [M ] (11.11.45) The equation (11. The physical explanation when it occurs beyond the scope of this book. Further. 11. the variables are separated as ¯ t ¯ −P k+1 2k ¯ f [M ] = 0 M (11.45) integrated to obtain the form −1 2k 2 ¯ 3k ¯= 2k t 1−P ¯ Mf [M ](3k − 1)(1 + k ) (11. a general solution is easily can be obtained similarly to rigid tank.4.). as results in faster reduction of the gas from the chamber.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. Further.42) where a is a constant that represent the physics. the relation must be also positive.4. increase of the pressure results in increase of the volume (as results of Hook’s law. After all. .

0. 6 Some suggested this border point as infinite evocation to infinite time for evacuation etc. In reality the value of n lays between zero to one.49) reduced to equation (11.214 CHAPTER 11.49) Again. similarly as before. Therefore. as first approximation and simplified approach can be done by a single term with a different power as V (t) = aP n (11.51) The linear condition are obtain when 3k − 2nk − 1 = 1 −→ n = That is just bellow 1 (n = 0. There isn’t a real critical value to n.48) Notice that when n = 1 equation (11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS Any continuous positive monotonic function can be expressed into a polynomial function. 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. This undersigned is not aware situation where this indeed play important role. However. 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. The physical meaning of n = 0 is that the tank is rigid. it is waited to find such conditions before calling it as critical condition.785714286) for k = 1.50) 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 (11.4.52) .43). 3k − 2 2k (11. 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. 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. d ¯ 1+nk−k k ¯ P dt +1 ¯ f [M ] = 0 ¯ k2 k M −P (11.47) When n can be any positive value including zero.

pp. equation (11. Vol.54) yields 1− Pexit P0 (t) fL k 4D 2 (11. P0 (t) Mexit The solution of two equations (11.19) is described in “Pressure die casting: a model of vacuum pumping” Bar-Meir.5.55) This solution should used only for Min < 0.18) and (11. . May 1996. otherwise equations (11. no. 259-265.19) must be solved numerically. 2. Eckert. J. E R G. 118. Journal of Manufacturing Science and Engineering (USA).19) as Min Pexit = .54) Min = . R.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 (11.00286.53) and (11.18) can be approximated as 4f L D = 1 Mexit 2 − Min 2 k Mexit 2 Min 2 (11. For small values of the Mach numbers. G. The solution of equation (11.53) and equation (11.11. Goldstein. ADVANCE TOPICS 215 11. (11.

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

direct and through reduced area are combined in this analysis. One can also view the exhaust systems of internal combustion engine in the same manner. Here the model deals with the case where the volume is controlled by external forces. It simplest model of such process is when a piston is displacing the gas though a long tube. The two different connections. It is further assumed that the process is relatively fast. 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. again to add the dimensional analysis what is rapid and what is slow. It assumed that no chemical reaction (or condensation/evaporation) occur in the piston or the tube 1 . This kind of model is applicable to many manufacturing processes such as die casting. sometime is vital to obtain a proper evacuation of the gas (air) from the cavity. 12. 217 . 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.1 General Model In this analysis. The last assumption is a appropriate assumption in process such as die casting. In these processes. extraction etc. For example. In general the process of the displacing the gas (in many cases air) with a liquid is a very common process. some simplified assumptions are made. Two extreme possibilities again suggest themselves: rapid and slow processes.CHAPTER 12 Evacuating/Filing Chambers under External Volume Control This chapter is the second on the section dealing with filling and evacuating chambers. in order to obtain the essence of the 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. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 12. The first approximation isotropic process describe the process inside the cylinder (see Figure (12.1. isontropic process ¢¤£¦¥¨§ ©.218CHAPTER 12.1)).

In reality. The conflicting requirements suggest an optimum area. 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. V (t) = V (0) 1 − 2 After t tmax (12. the possibility of crack increases. which is also indicated by experimental studies and utilized by practiced engineers. thus. the best ventilation is achieved with a large tube or area. For these reasons the optimum size is desired. 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 some situations like die casting this description is appropriate. The second objective is to provide a tool to “combine” the actual tube with the resistance in the tube. Obviously. These critical conditions are very significant that they provide guidelines for the design of processes. The replacing incompressible liquid enter in the same amount as replaced incompressible liquid. 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. it is noteworthy to think about the process in qualitative terms. 1 2 Fanno model Fig. 12. In the exhaust system cost of large exhaust valve increase with the size and in addition reduces the strength with the size of valve2 . Nevertheless. But in a compressible substance the situation can be totally different. this model can be extended numerical in cases where more complex function is applied.1) certain sizes. 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. A linear function is the simplest model that decibels changes the volume. eliminating the need for calculations of the gas flow in the tube to minimize the numerical calculations.

Equation (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. The first characteristic time.4)) can be obtained by transforming and by 2k −1 ¯ k2 ¯ = [ξ ] k−1 .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 . tmax .3) ¯ = t/tmax . P (12. GENERAL MODEL Equation (12.4) 2 [ξ ]( k−1 ) 2k −1 dξ 2k k−1 = where B = tmax ¯ tc M f [M ] And equation (12.1.8) . 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. ¯ (0) = 1. dP k−1 reduce equation (12.6) (k − 1) (1 − Bξ ) [ξ ] ¯ dt ¯ 1−t (12. The solution of equation (12. tc is associated with the ratio of the volume and the tube characteristics (see equation (11.4) Equation (12. The second characteristic time.7) ¯ =t (12.12.5) Equation (12.1) can be non–dimensionlassed as ¯ (t ¯) = 1 − t ¯ V 219 (12.4) is can be solved only when the flow is chocked In which case f [m] isn’t function of the time. Notice that in this case that there are two different characwhere t teristic times: the “characteristic” time.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 (12.2) The governing equation (11.5) can be further simplified as ¯ dt 2dξ = ¯ (k − 1) (1 − Bξ ) ξ 1−t (12.5)). tc and the “maximum” time. tmax is associated with the imposed time on the system (in this case the elapsed time of the piston stroke).

These Figures (12. the question remains how the time ratio can be transfered to parameters that can the engineer can design in the system.13) can be expressed as 1 tmax tc = @A tmax tc @A tmax tc (12. Thus.220CHAPTER 12. Thus the exit area. In the literature.11) The analytical solution also can be approximated by a simpler equation as ¯ = [1 − t] P tmax tc (12.2) and 12.9) The analytical solution is applicable only in the case which the flow is choked thorough all the process. The results of a choked case are presented in Figure (12. The solution is applicable to indirect connection. When tc > 1 the pressure increases significantly and verse versa.14) @Ac . In case when the flow chokeless a numerical integration needed to be performed. In this case only some part of the flow is choked (the later part).2). to create a direct function equation (12. This happen when vacuum is applied outside the tube (a technique used in die casting and injection molding to improve quality by reducing porosity.10) P (0) = 1 (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. = 1 as the critical area.3) in which outside tube condition is in vacuum.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. 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. Ac Denoting the area that creates the ratio tmax tc provides the needed tool.3 demonstrate the imtmax portance of the ratio of tmax tc . A can be expressed as A= The actual times ratio tmax tc @A A Ac Ac (12.

A c 4f L parameter.03[sec].4). The line that represents A = 1 is almost straight. S OLUTION 12. A Mexit .3 Direct Connection In the above analysis is applicable to indirect connection.1: Calculate the minimum required vent area for die casting process when the die volfL ume is 0. Ac . The effects of the area ratio. 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 .org). A .12. and the friction Parameters influencing the process are the area ratio. The direct connection posts more mathematical difficulty because the critical area is not constant and time dependent.2 Examples Example 12.001[m3] and 4D = 20. tmax = 0. A c the following equation Figure (12. This parameter represents the ratio between the filling time and the evacuating time. It should be noted that critical area. This parameter also represents the dimensionless area. The A A . The pressure and the mass flow rate are maintained constant after this state is reached.15) A . 12. 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. 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 (12.1.4) describes the pressure as a function of the dimensionless A A time for various values of A . The required solidification time. which was obtained by an integration of equation (12. it was found that the influence of the parameter 4f L D on the pressure development in the cylinder is quite small.5) tc is inversely proportional to area.14) the tmax is canceled and reduced into tmax tc = @A A Ac (12. . tc ∝ 1/A.potto.1. GENERAL MODEL 221 According to equation (11. This quasi steady state is achieved when the volumetric air flow rate out is equal to the volume pushed by the piston. are studied here since it is the dominant c parameter. Thus. It is important to point out the significance of the tmax tc . 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. according to values. is not function of the time. From other detailed calculations the author thesis (later to be published on this www.1. equation (12. The influence is small on the residual air mass in the cylinder but larger on the Mach number. D .9).

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. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT To continue 12.222CHAPTER 12. The critical area also provides a mean to “combine” the actual vent area with the vent resistance for numerical simulations of the cavity filling. taking into account the compressibility of the gas flow. . This critical area depends on the geometry and the filling time.

5 1.2: The pressure ratio as a function of the dimensionless time for chokeless condition .9 1.5 0.7  Dimensionless Time.0 2.0 0. or.2 3.7 0.3 0.0 o  = 100.0 0.0 4.4 Figure a 0.6 2.6 4.0 4fL D ∗ ∇ ∗ o ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.12.2 2.0 5.0 0.0 2.8 1.0 0. 12.0 0. Cylinder Volume Fraction Figure b 0.9 1.4 0.1 0.8 0.0 5.0 4fL D ∗ ∇ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.4 3.8 0.2 1.0 4.2 3.2 1.2 2. A/Ac 0. SUMMARY 223 5.4 1.2.0 Fig.1 0.8 1.0 5.5 0.6 2.0 1.2 0.4 1. t.2 0.5 1.8 Dimensionless Area.2 0.6 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 = 5.2 0.8 Dimensionless Area.0 1.6 4. A/Ac 0.4 3.0 ∗ ∗ 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 ∇ ∇ 0.3 0.6 0.

...8 1.. ∗ 0. ×. 1. . ..0 ∗ . ∗ .2 0 0 × ×   × × ×   ×                          × × × 0... Ac . ∗ = 0. ∗ . . ∗ ∗ ∗ × × × × × × × × × × × × × × × × × 2 1.224CHAPTER 12. 12... 4...5 ∗∗ .. .. .8   0...7 0.. . ∗ . ∗ .1 ..2 0. .. 12... ∗ ..3: The pressure ratio as a function of the dimensionless time for choked condition Fig.6 . ∗ . 0.. .. ... ∗ ..5 ∗ . ∗ ∗ ∗ . ∗ 1.. . CYLINDER VOLUME FRACTION 1 Fig.8 0.3 0. .   .4: The pressure ratio as a function of the dimensionless time .1 0. ∗ ..∗. 1.2 ∗ . ∗ . . ∗ ..4 1. .5 0. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 2...4  0.4 0..6    × × × × × ×   × × × × × 0.0 .. ∗ ..6 0.. ∗ . . ∗ .0 . ∗ .. . ∗ ..2 P(t) P(0)  × 1  0. or. A .9 DIMENSIONLESS TIME. .. t..

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

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

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

the flow moves downstream and creates an angle.2)). However. OBLIQUE-SHOCK 14. the direction of the flow is changed by a shock wave with an angle to the flow. As it may be recalled. This velocity increase is that both have a maximum point and referred to as the expansion wave. This shock is commonly referred to as the oblique shock. 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. ν∞(k ) θmax(k ) Meyer Oblique Function Shock The boundary conditions must match the geometry.2. Prandtl-Meyer function approaches closer to a zero deas opposed to oblique shock analyflection angle. For information such as the disturbance (boundary condition) to reach deep into the flow from the side requires time. 3 This section is under construction and does not appear in the book yet. 2 Zero velocity. the gas flow must match the boundary conditions. . normal shock occurs when a wall is straight/flat (δ = 0) as shown in Figure (14. the maximum point for the Prandtl–Meyer function is much larger than the oblique shock by a factor of more than 2.1 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.1) which occurs when somewhere downstream a disturbance2 appears. are examples of forces that create shock. the increase in the upstream Mach number determines the downstream Mach number and the “negative” deflection angle.2 Introduction to Prandtl–Meyer Function Decreasing the deflection angle re0◦ Prandtl sults in the same effects as before.2: The regions where oblique shock or locity must increase to achieve this rePrandtl–Meyer function exist. sis. and different inclination angle. 14.2 14. However. as discussed in Chapter (1)3 the flow behaves as it does in a hyperbolic field. Thus. Notice quirement. zero. The zero velocity can be found in a jet flowing into a still medium of gas. As a “no solution” zone. During this time.228 CHAPTER 14. The analysis shows that the flow ve. In such a case. Yet. which is around it will be shown in the next chapter. the flow must be continuous.Fig. When the deflection angle is increased.2. Alternatively. 14. This matching can occur only when there is a discontinuity in the flow field. 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. 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. for a negative deflection angle (in this section’s notaNo Shock zone tion).

. To match the boundary conditions.3 Introduction to Zero Inclination What happens when the inclination angle is zero? Which model is correct to use? Can these two conflicting models. The deflection angle. you do not need the author’s permission.” i. 14.2. Hence. which depends on the boundary layer.. In such a case. it was assumed that normal shock occurs at the same time. The normal shock analysis dictates that after the shock. Figure (14. co-exist? Or perhaps a different model better describes the physics.14. The two– dimensional effects occur when the flow is affected from the “side. a one– dimensional shock can be Fig. the change of the orientation makes the shock considerations two–dimensional. Only the oblique shock’s normal component undergoes the “shock. Previously (see Chapter 5). the flow turns after the shock to be parallel to the inclination angle. The three–dimensional 1#243 effects of the shock make it appear as a curved plane. You can use the traditional explanation. OBLIQUE SHOCK 229 14. Alternately.3: A typical oblique shock schematic considered. 0effects. 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.3.” The tangent component does not change because it does not “move” across the shock line.e. is the direction of the flow after the shock (parallel to the wall). δ . the flow is always subsonic. a change in the flow direction4 . the stability issue will be examined in greater detail. The total flow after the oblique shock can also be supersonic. In this chapter.-/. for a chosen arbi trary accuracy it requires a specific small area.).3 Oblique Shock The shock occurs in realθ−δ $ %'& ity in situations where the ( )* + + shock has three–dimensional θ   !#" . please just ignore it.3) exhibits the schematic of the oblique shock.  However. the oblique shock and the Prandtl–Meyer function.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. 14. using an infinite (or a two–dimensional) object produces a two–dimensional shock. If you do not like this description. the mass balance reads Comp arsi on L ine ρ 1 U1 n = ρ 2 U2 n (14.

230 The momentum equation reads CHAPTER 14. tan θ U1 n = tan(θ − δ ) U2 n (14. But because the literature referred to only two roots. 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. in which the shock turns to a lesser extent7 . The lesser extent angle is the unstable angle and the weak angle is the middle solution. δ . The author is not aware of “simple” proof and a call to find a simple proof is needed. requires solving the entropy conservation equation.1). here there are three possible pairs5 of solutions to these equations. The first is referred to as the weak shock. as sin θ = 6 The solution 5 This M1 n M1 (14. and the third is an impossible solution (thermodynamically)6 .6) U1 n U1 t (14. Yet the new relationship between the upstream Mach number.8) issue is due to R.2) (14.3) (14. and (14. . who raised the solution completeness issue. 7 Actually this term is used from historical reasons. From the geometry it can be observed that tan θ = and tan(θ − δ ) = U2 n U2 t (14.5) Unlike in the normal shock. (14.4) are the same as the equations for normal shock with the exception that the total velocity is replaced by the perpendicular components. the deflection angle. Menikoff. the second is the strong shock.7) The above velocity–geometry equations can also be expressed in term of Mach number.4) Equations (14. θ has to be solved. and the Mach angle. Experiments and experience have shown that the common solution is the weak shock. the term lesser extent is used.2).

and δ . here angles (the second dimension) have to be determined.3.12) The relationship between the properties can be determined by substituting M1 sin θ for of M1 into the normal shock relationship.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. M2 . OBLIQUE SHOCK M2 n M2 231 sin(θ − δ ) = cos θ = (14. It should be noted that although.17) ρ2 − ρ 1 ρ2 − ρ 1 . and it follows that the total speed of sound is constant across the (oblique) shock.13) The density and normal velocity ratio can be determined by the following equation (14.9) (14.11) M1 t M1 M2 t M2 cos(θ − δ ) = The total energy across an oblique shock wave is constant.14.15) (14. As opposed to the normal shock.16) The Rankine–Hugoniot relations are the same as the relationship for the normal shock P2 − P 1 P2 − P 1 =k (14. T1 = T2 . θ.8) through (14. is a function of four unknowns of M1 . The solution from this set of four equations.11). 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. 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. U1 t = U2 t the Mach number is M1 t = M2 t because the temperatures on both sides of the shock are different.10) (14. (14.

Again.19) (14. Thus. 8 This point was pointed out by R. the normal shock is a one–dimensional problem.18) Equation (14. and this will be discussed later. Clearly. One has to be careful to use the normal components of the Mach numbers. which create a somewhat complicated mathematical procedure. the negative root of sin θ must be disregarded 1. perhaps. M1 . this set of parameters is. Probably. sin θ must be positive. The stagnation temperature contains the total velocity.1 Upstream Mach Number. Oblique shock is a two–dimensional problem and two properties must be provided so a solution can be found. the most useful properties are upstream Mach number.20) (14. thus. if orientated to a coordinate perpendicular and parallel shock plane is like a normal shock.232 CHAPTER 14. the most common and natural to examine. M1 and the deflection angle.21) (14. He also suggested that θ is bounded by sin−1 1/M1 and . and the solutions of selected pairs and selected relationships will be presented. Thus. OBLIQUE-SHOCK 14.22) (14.4 Solution of Mach Angle Oblique shock. Other combinations of properties provide a relatively simple mathematical treatment.4. the relationship between the properties can be determined by using the normal components or by utilizing the normal shock table developed earlier. 14. δ Again.18) requires that x has to be a real and positive number to obtain a real deflection angle8 . and Deflection Angle. 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. only one parameter is required (to solve the problem). Menikoff. and the negative sign refers to the mirror image of the solution.

.4.18) provides three roots9 . all the roots are real and at least two are identical. 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 (14.30) Only three roots can exist for the Mach angle. In the last case where D < 0. (14. in a quadratic equation there are two roots. if D > 0. SOLUTION OF MACH ANGLE 233 The solution of a cubic equation such as (14.14. all the roots are real and unequal. one root is real and two roots are complex.29) (14. 9 The highest power of the equation (only with integer numbers) is the number of the roots. For example. to explain about complex numbers and imaginary numbers should be included. From a mathematical point of view.25) R+ √ D. 10 A call for suggestions. so that the shock normal component is reduced to subsonic and yet parallel to the inclination angle. For the case D = 0. θ. D > 0 occurs when no shock angle can be found.28) (14.26) T = and where the definition of the D is 3 R− √ D (14.27) 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 (14.23) (14.24) (14. Maybe insert an example where imaginary solution results in no physical solution. 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 .

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

and the weak shock coexist. The simple procedure For example. M1 . Fig. for. For a given upstream Mach number. and a3 for M1 = ∞. . M1 The first range is when the deflection angle reaches above the maximum point.4. a2 .14. k ). the wedge) in front of it and hence the normal shock occurs.4) and (14. SOLUTION OF MACH ANGLE 235 14. the strong shock. For example.” a change in the flow direction is no longer possible. This root.28) after substituting these values of Q and R provides series of roots from which only one root is possible.5). In the alternative view.0. This can be done by evaluating the terms a1 .4: Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. Only when the fluid is away from the object (smaller angle) liquid “sees” the object in a different inclination angle.4. The flow is far away from the object and does not “see’ the object. the imaginary angle is shown. the inclination angle reaches the “maximum potential energy. This different inclination angle is sometimes referred to as an imaginary angle. 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. Once.2 When No Oblique Shock Exist or When D > 0 Large deflection angle for given. the fluid “sees” the disturbance (in this case. M1 −→ ∞ the maximum deflection angle is calculated when D = Q3 + R2 = 0.4. 14. It can be noticed that the normal shock. in the case k = 1. in Figure (14. is just above δmax ∼ π 4 (note that the maximum is also a function of the heat ratio. a change in the inclination angle requires a larger energy to change the flow direction.

and (14.236 CHAPTER 14. This can be shown mathematically by the argument that differentiating equation (14. the deflection angle is a function of the Mach angle and the upstream Mach number.Chapman. θ.12). (14. M1 . and Mach number. . J. It can be noticed that in equation (14.21). 15 At first. Thus. 14. it was seen as C. Since in that equation there appears only the heat ratio k . When the right terms defined in (14.30) 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. one can conclude that the maximum Mach angle is only a function of the upstream Much number. English mathematician to be the creator but later an earlier version by several months was proposed by Bernard Grossman.5: The view of a large inclination angle from different points in the fluid field.29). M1 .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. At this stage it is not clear who was the first to propose it. M1 . The Procedure for Calculating The Maximum Deflection Point The maximum is obtained when D = 0.20)-(14. θmax is a function of only these parameters. 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. M1 and the maximum Mach angle.12) and equating the results to zero creates relationship between the Mach number.

taking the derivative with respect to θ.34) provides a solution for Mmax .33) should be referred to as the maximum’s equation. Thus.79 . 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 . The general solution of equation (14. then the Mach angle can be easily calculated by equation (14. Note that this procedure does not require an approximation of M1n to be made.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. To compare these two equations the simple case of Maximum for an infinite Mach number is examined. the Mach angle is sin θ = 2k .32) is zero at different values of the denominator. the maximum deflection angle can be computed. The nominator produces a quadratic equation for sin2 θ and only the positive value for sin2 θ is applied here. Once M1n is found.32) Because tan is a monotonous function.12).8).34) and then solving equation (14. the maximum appears when θ has its maximum. 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. Instead.4. d tan δ =0 dM1 (14. a derivative can be taken with respect to M1 . 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. With the value of θ utilizing equation (14. The numerator of equation (14. SOLUTION OF MACH ANGLE The differentiation of the equation (14.35) k+1 Hence. It must be pointed out that similar procedures can also be proposed (even though it does not appear in the literature).18 or θ = 67. which makes ◦ θ = 1.33) Equation (14. it is sufficient to equate the numerator to zero to obtain the maximum. Thus. It can be noticed that the Maximum Deflection Mach Number’s equation is also a quadratic equation for M1n 2 . for large Mach numbers.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. Thus.36) .

39) . This can be achieved by the standard mathematical procedure of equating D = 0. S OLUTION under construction The case of D ≥ 0 or 0 ≥ δ The second range in which D > 0 is when δ < 0.37) for R and Q becoming 9 R= 2 M1 2 +2 M1 2 2M1 2 +1 M1 4 2 (14. a2 . Example 14. 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. Start with equation (14.37) − 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. Thus. This can be partially demonstrated by evaluating the terms a1 . The analysis shows regardless of the value of the upstream Mach number D = 0 when δ = 0. first the transition line in which D = 0 has to be determined. This typical example is for those who like mathematics.31) and neglect all the terms that are relatively small.238 CHAPTER 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). Hint.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 + .

mathematical challenge for those who like to work it out. this identity can be demonstrated by checking several points for example. Once this analytical solution was published.0.16 words.6: The various coefficients of three different tradicts the common approach. Mach numbers to demonstrate that D is zero Nevertheless. others suggest a singular point18 .2 ysis is no. in the previous case. 17 There 16 A .0 3 0 16 can be generated from the wall 1 with zero deflection. 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 . Nevertheless.38) (14. 18 See for example. Note that. in the literature.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 . no Mach wave 2. 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 anal1.4.6) is provided for the following demonstration. this proof became trivial.. are several papers that attempt to prove this point in the past. there are several papers suggesting zero strength Mach wave. Fluids 10 (1) . 2. M1 = 1. SOLUTION OF MACH ANGLE 239 Substituting the values of Q and R equations (14. “Caustics of weak shock waves.28) results in D = 0. with a positive large deflection angle. the wall does not emit any signal to the flow (assuming zero viscosity). 14. Substitution of all the above values into (14. which con. Table (14. paper by Rosles. In the range where δ ≤ 0.28) provides the equation to be solved for δ .” 206 Phys.39) into equation (14. Utilizing the symmetry and antisymmetry of the qualities of the cos and sin for δ < 0 demonstrates that D > 0 regardless of Mach number. ∞.  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 (14.14. But for non ideal gas (real gas) this solution is only an indication. In other ∞ -1 0 . there was a transition from one kind of discontinuity to another. The question of singular point or zero Mach wave strength are only of mathematical interest. Hence. Tabak.0 -3 -1 . January 1998. And according to 9 this analysis.Fig.

detach normal shock). The approxi19 It is not a mistake. In reality. The first “weak” means more of compression “line” while the other means the weak shock. If the Mach number is large enough and the wall is rough enough. if it was assumed that no boundary layer exists and the wall is perfect. In the light of this discussion. It is well known. in the engineering world. OBLIQUE-SHOCK Suppose that there is a Mach wave at the wall at zero inclination (see Figure (14. any deviations from the zero inclination angle creates a jump from a positive angle (Mach wave) to a negative angle (expansion wave). If the sitFig. for simplicity’s sake. The imperfections of the wall can be. i. There must be another boundary condition(s) that causes the normal shock. the reality is that no continuous Mach wave can occur because of the viscosity (boundary layer). there are two “weaks. On the other hand.240 CHAPTER 14. no Mach wave at zero inclination should be assumed. For such a wall the zero inclination changes from small positive value to a negative value. After analysis of the oblique shock. there will be points where a weak19 weak will be created. .e.7)). Furthermore. 14. 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. The analysis of the oblique shock suggests that the inclination angle is not the source (boundary condition) that creates the shock. according the common explanation. This theoretical jump occurs because in a Mach wave the velocity decreases while in the expansion wave the velocity increases. But because the velocity after a Mach wave (even for an extremely weak shock µ1 µ2 µ3 µ∞ wave) is reduced. that there is no such thing as a perfect wall. At the very extreme case. both will not allow a situation of zero inclination with emission of Mach wave. distance. the same conclusion must be reached. However. another Mach wave occurs after a small distance. there are imperfections in the wall and in the flow and there is the question of boundary layer. emission of Mach wave can occur in other situations. only in several points (depending on the bumps) at the leading edge can a very weak shock occur.7: The Mach waves that are supposed uation keeps on occurring over a finite to be generated at zero inclination. This jump has to be in reality either smoothened out or has a physical meaning of jump (for example. the increase and the decrease depend on the upstream Mach number but in different directions. With these conflicting mechanisms. at least for a simple engineering analysis. Therefore. there will be a point where the Mach number will be 1 and a normal shock will occur. the boundary layer covers or smoothens out the bumps.” These words mean two different things. Nevertheless. for the purpose of an introductory class. Furthermore. The analysis started by looking at a normal shock which occurs when there is a zero inclination. the Mach angle will be larger (µ2 > µ1 ). that the normal shock can occur at zero inclination. thus. Obviously. assumed to be as a sinusoidal shape.

M1 .0005 0 0. and shock angle.0005 Myw Mys θs θw 0.41) . especially in acoustic engineering. θ.0 10.0 -0. shock angle.5 1 0. Oblique Shock 3 2.0 30.3 Upstream Mach Number.5 0 90 80 70 60 50 40 30 20 10 0 -0.14.001 0. are far much simpler and a unique solution exists. but for most cases it should be ignored. SOLUTION OF MACH ANGLE 241 mation of weak weak wave with nonzero strength has engineering applicability in a very limited cases. 14. and exit Mach number for M1 = 3 0 10 14.001 k = 1 4 Mx=3 0 10 20 30 Wed Jun 22 15:03:35 2005 Fig. θ δ 20 30 The solution for upstream Mach number. 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) (14.8: The calculation of D (possible error). M1 . and Shock Angle.5 2 1.4.0 δ 20.4.

14. M1 is M1 2 = 2(cot θ + tan δ ) sin 2θ − (tan δ )(k + cos 2θ) (14.41) shows that the shock angle. δ .49) .242 or tan δ = CHAPTER 14. δ and θ It is sometimes useful to obtain a relationship where the two angles are known. is limited.45) (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 (14.44) (14. is unique.4 Given Two Angles. is limited between ∞ and 1/ sin2 θ. has to be in the range of sin−1 (1/M1 ) ≥ θ ≥ (π/2) (see Figure 14. The first upstream Mach number.48) Even though the solution for these variables. OBLIQUE-SHOCK 2 cot θ(M1 2 sin2 θ − 1) 2 + M1 2 (k + 1 − 2 sin2 θ) (14. θ . upstream Mach number M1 .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. The range of given θ. the possible range deflection angle. Examining equation (14.46) (14.4.47) (14. M1 and θ.9).

52) The results are consistent with the initial assumption which shows that it was an appropriate assumption.14. 14.50) For a large upstream Mach number M1 and a small shock angle (yet not approaching zero). θ. Shock angle π θ= 2 θmax ∼ π 2 θ=0 Fig.4.0 < M1 < ∞ Defection angle strong solution θmin = sin −1 1 M1 supersonic weak soution possible solution no solution zone θ.41) can be simplified into k+1 θ∼ δ = 2 (14.51) (14. the deflection angle. SOLUTION OF MACH ANGLE subsonic weak solution 243 1. Equation (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. δ must also be small as well. .

for most engineering applications. ate i d i rm range inte for any finite length of the wedge. The common approach is to carry out numerical analysis. In this case there are no edge effects and the geometry dictates slightly different results. In spite of recent trends.6 Small δ “Weak Oblique shock” This interest in this topic is mostly from an academic point of view. for steady state and no known experiments showing that it exist can be found in the literature. a Even finite wedge with limiting wall can be considered as an example for this discussion if the B. The analysis also provides additional value of the expected range.10: Schematic of finite wedge with zero be area where current solution can be angle of attack. Experimental and analytical work shows that the weak solution is the stable solution and a discussion is provided in the appendix of this chapter. a simple tool is sufficient for limit accuracy.4. b assumed as correct . The end of the wedge must have a different configuration (see Figure (14.5 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. Clearly. 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.” Another geometry that can be considered as two–dimensional is the cone (some referred to it as Taylor–Maccoll flow). As opposed to the weak shock. As before. this solution is also act as a “reality check” to any numerical analysis. the sis naly al a norm range analysis needs to account for edge efno shock fects. Even though. OBLIQUE-SHOCK 14. The dimensional analysis shows that only the doted area to Fig. known to the author.L. { { { { { 14. Yet. In spite of the small area were the current solution can be assumed. the cone is a three–dimensional problem. b At this stage. The mathematics is much more complicated but there are three solutions. The . The author is not aware of any such analysis in literature. is neglected.10)). dimensional analysis is not completed.4. the first solution is thermodynamical unstable. the strong shock is unstable. reports that only a weak shock is possible. at least. It is recommended that this issue be skipped and the time be devoted to other issues. the numerical works require many times a “reality check. the symmetrical nature of the cone creates a semi–2D problem.244 CHAPTER 14. All the literature. the analysis for the middle section produces a close result to reality (because of symmetry). Additionally. 14.

δ . In this theory. the close proximity view provides a continuous turning of the deflection angle. In fact. 14. At the local view close to the wall. SOLUTION OF MACH ANGLE 245 author is not aware of any single case in which this topic is used in real–world calculations. the boundary layer increases the zone where a continuous δ flow exists. However. the far view shows a sharp transition. The boundary layer reduces the upstream flow velocity and therefore the shock does not exist at close proxFig.5152 θmax 76.7 Close and Far Views of the Oblique Shock In many cases. the oblique shock is an accumulation of many small (or again weak) “weak shocks.14.4 The maximum values are summarized in the following Table .4. 14. it should be covered in this book. studying this topic seems to come at the expense of other more important topics. the oblique shock is a weak “weak oblique” shock. Mx 1. 14. the author admits that as long as there are instructors who examine their students on this issue.8 Maximum Value of Oblique shock Table 14.1: Table of maximum values of the oblique Shock k=1. Yet.97131 δmax 1. and small normal upstream Mach numbers.2762 . From the far view.11: A local and a far view of the imity to the wall.. δ . 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. In reality. under construction. the boundary θ layer (B.1000 My 0.53) tan θ = . In larger distance from the wall. M1 ∼ 1+ .” However. after the explicit analytical solution has been provided. 1 M1 2 − 1 (14. these small “shocks” are built or accumulate into a large and abrupt change (shock). oblique shock. the shock becomes possible.L.. For small deflection angles.) does not enter into the calculation.4.11).4.

7000 1.91997 0.1675 22.97214 0.4.0922 38.5787 65.4419 It must be noted that the calculations are for the perfect gas model.2000 1.93083 0.0734 35. OBLIQUE-SHOCK Table 14.6000 3. 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.6814 30.9087 66.6515 17.6621 9. for large deflection angle.96335 0.7532 64.94387 0.3673 67.96942 0.3059 38. Henderson and Menikoff20 suggested a procedure to calculate the maximum deflection angle for arbitrary equation of state21 .0000 8.2000 3.4000 3.6934 64.1028 28.0119 19.7739 41.2000 2.0000 7.0000 My 0.4398 43. Since the flow “sees” the obstacle.92478 0.1: Maximum values of oblique shock (continue) k=1.94925 0.99559 δmax 3.2503 67.6000 1.99047 0.3066 64.4000 1.6000 2.4 Mx 1.91871 0.3000 1.95049 0. 179–210.98714 0. this assumption might not be sufficient and different analysis is needed.8000 4.6465 64.4290 θmax 71.99440 0.2309 65.5875 34.99337 0. In some cases.6074 64.0000 3. there isn’t a physical solution to an oblique shock.95435 0.98183 0.4272 12.9555 69.9 Detached shock When the mathematical quantity D becomes positive.0000 6.0000 9.96630 0.93629 0.0000 2.93747 0.1127 14.92683 0.0000 10.9000 2.7908 44.4000 2.9735 26.7593 65.7023 66. .4144 65. 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.9668 64.5671 66.7972 65.8000 1.0464 66.92165 0.2546 43.1833 21.3934 37.8443 65.91941 0.5000 1.9442 6.8000 3.3275 36.1177 42.9020 67.3645 67.0000 5.92224 0. 14.1196 67.1619 44.246 CHAPTER 14.8137 32.5676 65.0399 65.95897 0.

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.14. 14. As a general rule. it can be explained as the subsonic is “sucked” into gas in zone C. The flow in zone A has to go through some acceleration to became supersonic flow. these calculations can be summarized by the flowing equation detachment distance = constant × (θ − f (M∞ )) body thickness w ea k sh oc k (14.12 exhibits a round–tip bulzone B Subsonic Area θ let with a detached shock. some might be used in certain situations which the author isn’t aware of. E Strong Shock F > @ . the flow becomes “blinder” to obstacles.4. 14. Upstream U∞ Figure 14. Thus. 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.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. 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 : . While at zone Fig.4. alternatively.CG isn’t important for this discussion). or. a wedge or 22 See example 14.5. The explanation to such a phenomenon is above the level of this book (where is the “throat” area question22 .12: The schematic for a round–tip bullet C the flow must be supersonic. Even though this topic has few applications. . Regardless of the explanation. this phenomenon has a larger impact for a relatively smaller supersonic flow. <> = > ?1@ ? ACBD< . 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. Nevertheless. The constant and the function are different for different geometries. the increase in the upstream Mach results in a decrease of the detachment distance. a graph and a general explanation to engineers is provided.54) where f (M∞ ) is a function of the upstream Mach number which tabulated in the literature. SOLUTION OF MACH ANGLE 247 from the body. In the typical example. Yet. In such a case. the gas must go through a shock. The in a supersonic flow weak oblique shock is predicted to flow around the cone.

14: The “detached” shock in a complicated configuration sometimes reoblique shock.. The Mach number decreases after every shock. OBLIQUE-SHOCK a cone moves into a still medium or gas flows into it. hence δ1 = δ2 .e. a symmetrical suction section in which the deflection angle is just between the maximum deflection angle and above half of the maximum deflection angle. as before. the maximum deflection angle decreases with a decrease the Mach numA sub sonic ber. Thus. there are configurations in which a detached shock occurs in design and engineers need to take it into consideration. yet the mathematics is somewhat complicated. Additionally. 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.55) Additionally. It seems that the most straightforward method is the following: (a) Calculate M1 B . 14. The maximum deflection angle in this case is. This detached shock ferred to as Mach reflection manifested itself in a form of curved shock (see Figure 14. However. δmax = f (M1 ) (14. only a function of the upstream Mach number. a detached shock occurs. Consider.248 CHAPTER 14. due to the symmetry flow a slip plane angle can be guessed to be parallel to original flow.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 . If the deflection angle exceeds the maximum possible. 14.13). at least two oblique shocks occur and after their interaction is shown in Figure (14. No detached shock issues are raised when only the first oblique shock is considered. ThereU B C fore. it can be observed for Fig. In this situation. for example. The analysis of this situation is logically very simple. for U B ikj every upstream Mach number there C θ2 Slip Plane is only one maximum deflection anA δ2 gle. This situation is known in the scientific literature as the Mach reflection. However. Such configurations seem sometimes at first glance not related to the detached shock issue. .14). The calculations for such a case can be carried out by several approaches. 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.

11 Oblique Shock Examples Example 14.28) or the GDC which yields the following results Mx 4. Note that no maximum angle is achieved in this shock.4. δ2 = δ1 and the Mach number M1 B to calculate M1 B . What is the maximum wedge angle at which the oblique shock can occur? If the wedge angle is 20◦ .56660 δ 0.0000 My 0.34907 Example 14. For example. 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.7738 θmax 66. In general. the strong Mach numbers.48523 My w 2. one must be aware that there are zones of dual solutions in which sharp shock line coexists with a curved line. and the respective shock angles. θ2 . For engineering purposes when the Mach number reaches this value. Is the shock shown in the photo weak or strong shock? Explain.2: Air flows at Mach number (M1 ) or Mx = 4 is approaching a wedge. it can be ignored. In discussing these issues.4. 14. calculate the weak.15) is exposed to supersonic flow and create an oblique shock. SOLUTION OF MACH ANGLE (b) Calculate the maximum deflection angle. M1 by checking the relationship between the intermediate Mach number to M1 . . δ2 utilizing equation (14.28) (a side to the case proximity of δ = 0).5686 θs 1. The wedge angle that satisfies this requirement is by equation (14.14. This procedure can be extended to calculate the maximum incoming Mach number.5◦ .31) equation (c) Calculate the deflection angle.97234 δmax 38. utilizing (14.4635 θw 0.3: A cone shown in Figure (14. The maximum values are: Mx 4.0000 My s 0.0407 To obtain the results of the weak and the strong solutions either utilize the equation (14.12) 249 (d) Use the deflection angle. S OLUTION The maximum wedge angle for (Mx = 4) D has to be equal to zero. at Mach 5 this zone is 8. POTTO–GDC can be used to calculate this ratio. this zone increases as Mach number increases.

Even if the cone was a wedge.2.88737 Because the flow is around the cone it must be a weak shock.250 CHAPTER 14. a Research Professor at Cuernavaco University of Mexico.099◦.49) or the Potto-GDC. 14. S OLUTION The measurement shows that cone angle is 14. .0143 30. M1 3. Grigory Toker.56543 2. the cone half angle is 15◦ and the Mach number is 2.15: Oblique shock occurs around a cone. Note that the Mach number is larger than the one predicted by the wedge. This photo is courtesy of Dr. OBLIQUE-SHOCK θ δ Fig.43◦ and the shock angle is 30. the shock would be weak because the maximum (transition to a strong shock) occurs at about 60◦ . According to his measurement.0990 14.4300 0. With given two angles the solution can be obtained by utilizing equation (14.2318 My s My w θs θw δ P0 y P0 x 0.4522 71.

17))a . For details see Henderson and Menikoff “Triple Shock Entropy Theorem. a In fact.0 4.0 3. Compare the different conditions in the two .16: Maximum values of the properties in an oblique shock 14.0 6. supersonic flow there is less entropy production (less pressure loss).” Journal of Fluid Mechanics 366. and T = 17◦ C. with a proper design.0 10. To illustrate the design significance of the oblique shock. there is general proof that regardless to the equation of state (any kind of gas). the entropy is to be minimized through a series of oblique shocks rather than through a single normal shock.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.4: The Section described in Figure 14. 179–210.0 8.5 0 90 80 70 60 50 40 30 20 10 0 2.0.18 air is flowing into a suction section at M = 2. It is suggested that a series of weak shocks “•”—–•˜ should replace one normal shock to increase the efficiency (see Figure (14. the flow can be brought to a subsonic Fig.0[bar].17: Two variations of inlet suction for flow just below M = 1. (1998) pp. In such a case. Clearly.5 2 1. Example 14.4.14. 14. P = 1. the following example is provided.0 9.0 Mx Thu Jun 30 15:14:53 2005 My θ δ Fig.0 7.5 1 0.4. 14.0 5. SOLUTION OF MACH ANGLE 251 k=14 Oblique Shock 3 2.

99445 In the new region.0000 0. In the past the input file was given but the graphical interface it is no longer needed. no input explanation is given.7498 My s My w θs θw δ P0 y P0 x 0.0000 My s My w θs θw δ 7.6667 4.72087 In the oblique shock.4) S OLUTION The first configuration is of a normal shock for which the results23 are Mx 2.96524 And the additional information is 23 The results in this example are obtained using the graphical interface of POTTO–GDC thus.2098 0.58974 1.5000 0.6875 2.71761 1.18: Schematic for Example (14.252 CHAPTER 14.7498 36.2346 76.9831 51.1931 0.0000 P0 y P0 x 0.2485 1. the new angle is 7◦ + 7◦ with new upstream Mach number of Mx = 1. 14. °²± ™/š›DœužyŸ• ¡š¡¢•£ 1 neglect the detached distance 7◦ ¤¥/¦ § ¨e©/ª¬«C­/¤¡®•¯« 2 ³µ´ 3 4 Normal shock 7◦ Fig.7021 36. .7498 85.99445 and the additional information by the minimal info in the Potto-GDC is Mx 2.2098 1.0000 My w θw δ 7.0000 My 0.7498 resulting in Mx 1. OBLIQUE-SHOCK different configurations.0000 Py Px Ty Tx P0 y P0 x 1.5549 14.57735 Ty Tx ρy ρx Py Px P0 y P0 x 1. Assume that only a weak shock occurs. the first angle shown is Mx 2.

45115 Ty Tx ρy ρx Py Px P0y P0 x 3. the results are: Mx 1.19). .4 yields M 0.2346 My 0. What is the requited “throat” area ratio to achieve acceleration from the subsonic region Fig.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 3.2609 14.82141 Ty Tx ρy ρx Py Px P0y P0 x 1.285 = 2.98903 With two weak shock waves and a normal shock the total pressure loss is P04 P03 P02 P04 = = 0.21295 Now utilizing the isentropic relationship for k = 1.5000 My 0. In fact.2626 × 1.99549 An oblique shock is not possible and normal shock occurs.1250 0.4018 1.9496 P0 1 P0 3 P0 2 P0 1 The static pressure ratio for the second case is P4 P4 P3 P2 = = 1. the loss in the normal shock is above than 31% of the total pressure.5 (see Figure 14. SOLUTION OF MACH ANGLE Mx 1.99445 = 0.2626 1.3151 4.5088 41. Example 14.7498 My w θw δ 7.6116 0. The half wedge angle is 10◦ . 14.4.19: Schematic for Example (14.96524 × 0. In such a case.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 1.14.6116 × 1.86966 1.8770 1.98903 × 0.4458.1497 1.90506 1.1853 0.4458 0.45115 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.5) 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 3.2574 Thus the area ratio has to be 1. 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.96089 0.

0000 My s My w θs θw δ 3.13665 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◦ .8482 My s My w θs θw δ 3.04290 0.6: CHAPTER 14.8807 23.20). explain why this description has internal conflict.0000 P0 y P0 x 0. The total pressure ratio at zone 2 is P02 P02 P01 = = 0.10500 0.04263 0. 2.7049. The conditions for this shock are: Mx 3.7008 are M 2.99894 × 0.1854 0.8912 22.99879 For the shock BC the results are Mx 2.20).40596 0.99894 And the isentropic relationships for M = 2.7008 87.254 Example 14.40669 0.20: Schematic of two angles turn with two weak shocks all the shocks are weak and straight.0.7080 0.13632 0.997731283 P00 P01 P00 .0000 P0 y P0 x 0.9356 0.7049 2.0000 My s My w θs θw δ 6.10548 3.0000 P0 y P0 x 0.48013 2.7049 88. Assume that there are no boundary layers and Fig.8482 88.1978 3.7008 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.5990 0. Perform the calculation for M1 = 3.99879 = 0. Then. Find the required angle of shock BE.47641 2.99105 The transition for shock AB is Mx 3. 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 (14.9476 21. 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. 14.48610 2.

0 θw 21. To do that. Example 14.0013 1.00000 From the last calculation it is clear that the government proposed schematic of the double wedge is in conflict with the boundary condition.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.997731283 × 0. and U1 = 1000m/sec.715320879◦ The deflection angle for such shock angle with Mach number is Mx 2.7037 0.14.04263 = 0.0008 My 0. SOLUTION OF MACH ANGLE On the other hand.042516045/0. the flow behaves as only one theoretical angle of 6◦ exist. The flow in zone 3 will flow into the wall in about 2.042436789 = 1. The increase has to be P3 /P2 = 0.99920 Ty Tx ρy ρx Py Px P0y P0 x 1. check whether the pressure at 2 is above or below or above the pressure (ratio) in zone 4.7049 My s My w θs 0.72 δ P0 y P0 x 0.0008/2.49525 2.0019 1. T1 = 27◦ C.001867743 To achieve this kind of pressure ratio the perpendicular component has to be Mx 1.04290 × 0.4).026233 1.7◦ .0005 1.4 and R = 287J/KgK . Assume k = 1. P0 P2 P2 = 2 = 0.4.7049 = 21.042436789 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. θ can be calculated from θ = sin−1 1.99105 = 0.00000 The shock angle. Only when the flow is far away from the double wedge. In reality the flow of double wedge will produce a curved shock surface with several zones. the angle for BE shock which cause this pressure ratio needs to be found. the pressure at 4 has to be P4 P4 P0 4 = = 0.7: Calculate the flow deflection angle and other parameters downstream when the Mach angle is 34◦ and P1 = 3[bar].

5000 My s My w θs 85.8575 0. Utilizing GDC for Mach number 2.5 and the angle of 11◦ results in Mx 2.256 CHAPTER 14.8575 × 3 = 8.0995 θw 32. OBLIQUE-SHOCK S OLUTION The Mach angle of 34◦ is below maximum deflection which means that it is a weak shock.0) = 1.8800 My s My w θs 0.3949 2.0 θw 34.13) or simply converting the M1 to perpendicular component.96873 .78 P0 y P0 x 0. has to be determined M1 = √ U1 1000 = 2.128 1.4 × 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. M1 .00 δ 15.0485 2.89145 The temperature ratio combined upstream temperature yield T2 = 1.5 = 872.0000 P0 y P0 x 0.89127 The relationship for the temperature and pressure can be obtained by using equation (14.6[m/sec] Example 14.66545 Ty Tx ρy ρx Py Px P0 y P0 x 1.61 From the Table (5.88 sin(34.4 × 287 × 418. the Upstream Mach number.1280 0.8124 δ 11.88 = 1.53431 2.5K and the same for the pressure P2 = 2. Mx 1.5 and wedge with a total angle of 22◦ . calculate the ratio of the stagnation pressure.6100 My 0.15) and (14.3949 × 300 ∼ 418.1) or GDC the following can be obtained.8: For Mach number 2.57[bar] And the velocity √ √ Un2 = My w kRT = 2.48269 2. M1n = M1 ∗ sin θ = 2.0443 0. Yet.

14. the following can be obtained: . then the Mach angle was calculated using the geometrical relationship of θ = sin−1 M1n /M1 . 3 stream line 2 1 θ M1 = 4 δ Assume that the upstream Mach number is 4 and the deflection angle is δ = 15◦ .10: Consider the schematic shown in the following figure.5000 My max θmax 64. For a weak shock the maximum pressure ratio is obtained at the deflection point because it is closest to a normal shock. 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). either use the Maximum Deflection Mach number’s equation or the Potto–GDC Mx 2.94021 4.60027 In these calculations. To obtain the maximum point for 2.3573 2. is obtained. Example 14.6854 0. utilizing equation (14. SOLUTION OF MACH ANGLE 257 Example 14.5 Mach number.12) the deflection angle. Utilizing the table or the Potto–GDC. With these two quantities.7822 δ 29.4.7974 Py Px Ty Tx P0 y P0 x 0. For the first shock. δ .9: What is the maximum pressure ratio that can be obtained on wedge when the gas is flowing in 2.5 Mach without any close boundaries? Would it make any difference if the wedge was flowing into the air? If so. what is the difference? S OLUTION It has to be recognized that without any other boundary condition. the shock is weak shock. It is clear that the flow must be parallel to the wall. Maximum Deflection Mach’s equation was used to calculate the normal component of the upstream. S OLUTION This kind of problem is essentially two wedges placed in a certain geometry. the upstream Mach number is known together with deflection angle.

9290 85. With these pieces of information by utilizing the Potto-GDC the following is obtained: Mx 2.85.5851 27.2028 32.48469 2.80382 And the additional information by using minimal information ratio button in Potto– GDC is Mx 4.51 P0 y P0 x 0.0000 My w θw δ Py Px Ty Tx P0 y P0 x 2.51367 2.0000 1.7985 1.7344 0.5764 = 2.80382 With a Mach number of M = 2.7822 15.0000 My s My w θs CHAPTER 14.7985 × 1.90041 and the additional information is Mx 2.9290 My w θw δ Py Px Ty Tx P0 y P0 x 2.258 Mx 4.6695 1.0 θw 29.3575 0. S OLUTION Here the Mach number and the Mach angle are given. P1 P1 P2 = = 1. the second deflection angle is also 15◦ .7822 15.5764 0.0629 15.2028 84.0000 0.8500 My s My w θs 0.9290 27. Utilizing the POTTO–GDC which provides a solution in just a few clicks. Note that hand calculations requires endless time looking up graphical representation of the solution.9290 My s My w θs θw δ P0 y P0 x 0.632 T3 T2 T3 Example 14.7344 × 1.11: A similar example as before but here Mach angle is 29◦ and Mach number is 2.6695 = 3.00 δ 10.0000 0.929.2808 32.0629 15.90041 With the combined tables the ratios can be easily calculated.96263 .0026 P3 P2 P3 T1 T1 T2 = = 1. With these values the following can be obtained: Mx 2.0000 1.46152 2. OBLIQUE-SHOCK θw δ P0 y P0 x 0. Again calculate the downstream ratios after the second shock and the deflection angle.

41523 3.8021 T3 T2 T3 Example 14.4089 1.8000 5.8500 My w θw δ Py Px Ty Tx P0 y P0 x 2.3575 29.5100 0.9398 34.0000 29.49901 And the additional information is . SOLUTION OF MACH ANGLE 259 and the additional information by utilizing the minimal info button in GDC provides Mx 2.51 the so called reflective shock gives the following information Mx 2.3984 1.97569 P1 P1 P2 = = 1.0 θw 30.5100 1.00 δ 20.0000 10.96263 With the deflection angle of δ = 10.3575 My s My w θs θw δ P0 y P0 x 0.3582 × 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 0.0590 10.5131 1.97569 and the additional information of Mx 2.17 P0 y P0 x 0.3582 0.0000 My 0.3575 My w θw δ Py Px Ty Tx P0 y P0 x 1.0000 0.0590 10.9419 34.41523 Ty Tx ρy ρx Py Px P0y P0 x 5.3268 0.9419 84.06172 While the results for the oblique shock are Mx 5.3268 ∼ 1.3984 ∼ 1.4.4089 × 1.54894 1.0000 My s My w θs 0.14.97 P3 P2 P3 T1 T1 T2 = = 1.12: Compare a direct normal shock to oblique shock with a normal shock. What is the deflection angle in this case? S OLUTION For the normal shock the results are Mx 5.

6858 ∼ 6. 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.0000 20.3740 0.32671 The pressure ratios of the oblique shock with normal shock is the total shock in the second case. For the first direction of 15◦ and Mach number =5.36 P3 P2 P3 T1 T2 T1 = = 2.374 ∼ 27.8625 10.13) For upstream Mach number of 5 and deflection angle of 12◦ and 15◦ .47485 Ty Tx ρy ρx Py Px P0 y P0 x 2.49901 The normal shock that follows this oblique is Mx 3.260 Mx 5.6375 × 10.0000 My w θw δ CHAPTER 14.13: A flow in a tunnel ends up with two deflection angles from both sides (see the following Figure (14.0058 My 0. Example 14.75 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.1736 2.0058 30. OBLIQUE-SHOCK Py Px Ty Tx P0 y P0 x 3. S OLUTION The first two zones immediately after are computed using the same techniques that were developed and discussed earlier.6375 2. Based on these calculations.5141 0.5141 × 2. 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. . explain whether the slip angle is larger or smaller than the difference of the deflection angle. P1 P1 P2 = = 2.13)).6858 3.

0000 1.5040 24.6986 85. it is assumed that the slip angle moves half of the angle to satisfy both of the deflection angles (first approximation). In this case.9122 21.6625 0.0000 0.2845 12.6819 27. The first case is where match lower deflection angle and second is to match the higher deflection angle.4.5656 0. that only one deflection angle exist.3217 15.69317 For the second direction of 12◦ and Mach number =5. If the two sides were equal because of symmetry the slip angle is also zero.47413 2.9791 1.5◦ which results in Mx 3.88496 And in zone 4 the conditions are due to deflection angle of 13. Mx 5. It is to say.80600 And the additional conditions are Mx 5.6986 27.14.8006 86. Under this assumption the conditions in zone 3 are solved by looking at the deflection angle of 12◦ + 1. SOLUTION OF MACH ANGLE Mx 5. They are the pressure and the velocity direction.80600 The conditions in zone 4 and zone 3 have two things that are equal.5040 My w θw δ Py Px Ty Tx P0 y P0 x 2.0000 0.3217 15.88496 with the additional information Mx 3.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3. It has to be noticed that the velocity magnitudes in zone 3 and 4 do not have to be equal. for the analysis.5◦ = 13.0000 1.6668 13.5040 My s My w θs θw δ P0 y P0 x 0.5000 1.69317 And the additional conditions are Mx 5.6668 13.2845 12.9238 0.6963 1. For the two different deflection angles.43914 3.0000 My s My w θs θw δ P0 y P0 x 261 0.6247 1.0000 My s My w θs θw δ P0 y P0 x 0.8006 21.43016 3.5040 86.8006 . the slip angle has two extreme cases.5000 0.5◦ and Mach 3.0739 24.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3. This non–continuous velocity profile can occur in our model because it is assumed that fluid is non–viscous.

M1 is always greater than 1.5◦ to a larger number.3226 13.6038 To reduce the pressure ratio the deflection angle has to be reduced (remember that at weak weak shock almost no pressure change).4. as oppose to the normal shock. To reduce the pressure the angle of slip plane has to increase from 1. S OLUTION Waiting for the solution 14. the oblique shock with upstream Mach number.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.6038 0. the pressure at zone 3 has to be reduced.6963 1.9035 85.6247 × 1.8006 My w θw δ Py Px Ty Tx P0 y P0 x 2. Consider the same values as previous analysis.5 Summary As with normal shock.5000 1. 14. 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.86179 with the additional information Mx 3. the .6577 1.9035 26.3226 13. 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). OBLIQUE-SHOCK θw δ P0 y P0 x 0. Thus.14: The previous example gave rise to another question on the order of the deflection angles.262 Mx 3.9791 ∼ 1. in oblique. However. Example 14. There are researchers which suggest that the numerical work is the solution.18192 P4 P2 P0 P1 P4 1.5000 0.9316 26.46259 2.8006 My s My w θs CHAPTER 14. The optimum deflection angle is a function of the Mach number range in which the suction section is operated in.86179 From these tables the pressure ratio at zone 3 and 4 can be calculated P3 P2 P0 P1 1 P3 1 = = 1.

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

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

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

The Mach line is the chief line in the analysis because of the wall contour shape information propagates along this line. and it is assumed here to be isotropic for a positive angle.3: The schematic of the turning only. the total Mach numÛ Ü4Ý¡ÞÔßÊà á  ber increases.4) dν ∼ sin(dν ). Once the contour is changed. as it turns out. the flow direction will change to fit the wall. 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). PRANDTL-MEYER FUNCTION 1 M1 − 1 µ = tan−1 √ (15. This assumption. The change of the velocity in the flow direction.3) . Therefore.2) A Mach line results because of a small disturbance in the wall contour. no shock is created from many small positive angles. is dy = (U + dU ) sin(dν ) = U dν The tan µ is the ratio of dy/dx (see Figure (15. a discussion on the relationship between the flow properties and the flow direction is presented. is close to reality.3)) tan µ = dx dU = dy U dν (15. This Mach line is assumed to be a result of the positive angle.266 or CHAPTER 15. 15. 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. cos(dν ) ∼ 1 (15. The reason that a “negative” angle is not applicable is that the coalescing of the small Mach wave which results in a shock wave. the Mach angle dν dy òĵÅCà Á increase and result in a change in the direction of the flow.6) (15. Hence. dx is In the same manner. Later. the velocity perpendicular to the flow. However. This direction change results in a change of the flow properties. this assumption will be examflow ined.5) dx = (U + dU ) cos ν − U = dU (15. dy . 15.2 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. In this chapter.

12) ∂Ur Uθ ∂Ur Uθ 2 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r ρ ∂r ρ ∂r (15. 15.11) (15. 15.15.9) The constant can be chosen in a such a way that ν = 0 at M = 1.7) (15. It must be recognized that here the cylindrical coordinates are advantageous because the Fig.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. a more rigorous explanation is provided.8) (15.7) transforms it into √ M 2 − 1dM 2 dν = − 1 2 2M 2 1 + k− 2 M After integration of equation (15.1 Alternative Approach to Governing Equations In the previous secback Mach tion. In this section.4: The schematic of the coordinate based on the matheflow turns around a sinmatical description gle point.10) .2. a simplified verr Ur line sion was derived based Front Mach on geometrical arguUθ line θ ments. GEOMETRICAL EXPLANATION The ratio dU/U was shown to be dM 2 dU = 1 2 U 2M 2 1 + k− 2 M Combining equations (15.6) and (15.2. For this coordinate system. 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.

16) from equation (15.15) (15. then all the derivatives with respect to the radius will vanish. r.17) Uθ or Uθ 2 Ur + ∂Uθ ∂θ = c 2 Ur + ∂Uθ ∂θ (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.19) it follows that Uθ = c (15. like the first term in the mass equation.13) can be rearranged as transformed into − 1 Uθ Ur + ∂Uθ ∂θ = 1 ∂ρ ρ ∂θ (15.20) Ur + ∂Uθ ∂θ =0 (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. Hence. but only .14) into equation (15.268 CHAPTER 15.18) And an additional rearrangement results in c2 − U θ 2 From equation (15.13) Equation (15. One has to remember that when r enters to the function.16) results in ∂Uθ − Ur ∂θ = c2 Uθ Ur + ∂Uθ ∂θ (15. PRANDTL-MEYER FUNCTION If the assumption is that the flow isn’t a function of the radius. the mass equation is reduced to ρUr + ∂ (ρUθ ) =0 ∂θ (15. the derivative isn’t zero.

28) .24) Utilizing equation (15. into equation (15.22) In the literature.24). equation (15.15. The energy equation for any point on a stream line is h(θ) + Uθ 2 + U r 2 = h0 2 (15.23) results in c2 Uθ 2 + U r 2 + = h0 k−1 2 (15. the non–dimensionalization can be applied at this stage as well. is c(θ )2 k h(θ) = Cp T = Cp and substituting this equality.20) for the speed of sound and substituting equation (15.27) incorporating the constant becomes Ur = 2h0 sin θ k−1 k+1 (15. the component shown in Figure (15. This analysis can be also applied to the same equation when they are normalized by Mach number.3) under Uy is equal to the speed of sound.26) becomes k+1 k−1 ∂Ur ∂θ 2 k−1 + ∂Ur 2 ∂θ + Ur 2 2 = h0 (15.25) 1 Cp c2 R T = RT = R (k − 1) Cv k−1 (15.22) which is the radial velocity transforms equation (15. based on the definition of the Mach angle.15) becomes Uθ r ∂Ur − Uθ ∂θ =0 (15. equation (15. After some additional rearrangement. ∞ and since Uθ = 0 leads to ∂Ur = Uθ ∂θ (15. M = 1. In fact.2.27) Note that Ur must be positive. However. equation (15.23) Enthalpy in perfect gas with a constant specific heat.21) If r isn’t approaching infinity. these results are associated with the characteristic line.25) into ∂Ur 2 ∂θ After some rearrangement.26) + Ur 2 = 2h0 (15. GEOMETRICAL EXPLANATION 269 the tangential component. The solution of the differential equation (15. k .

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. has to match to the definition of the angle that is chosen here (θ = 0 when M = 1). In the geometrical model. For this upstream Mach number (see Figure (15.30) Now utilizing the expression that was obtained for Ur and Uθ equations (15.2.34) (15.6).27) because sin2 θ +cos2 θ = 1.28) is chosen such that Ur (θ = 0) = 0. PRANDTL-MEYER FUNCTION which satisfies equation (15.2)) tan ν = Mstarting 2 − 1 (15.31) 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. so ν (M ) = θ(M ) − θ(Mstarting ) = k+1 tan−1 k−1 k−1 k+1 M2 − 1 − tan−1 M2 − 1 (15.32) k+1 tan2 k−1 θ k−1 k+1 (15. but the assumptions for the construction of these models are different.270 CHAPTER 15. the assumption .33) The deflection angle ν .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. The tangential velocity obtains the form Uθ = c = ∂Ur = ∂θ k−1 k+1 2 h0 cos θ k−1 k+1 (15. 15. The upstream Mach number is denoted in this segment as Mstarting .29) and (15. The arbitrary constant in equation (15.35) These relationships are plotted in Figure (15.2 Comparison And Limitations between the Two Approaches The two models produce exactly the same results.

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. Normally. Aside from its close proximity to turning point. Prandtl–Meyer function becomes ν∞ = π 2 k+1 −1 k−1 (15. However. Thus. the question of boundary layer arises. 15. THE MAXIMUM TURNING ANGLE 271 is that the velocity change in the radial direction is zero.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. 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. in many instances. The radial velocity is “fed” through the reduction of the density. In the rigorous model. some researchers recommend that. the sharp point should be replaced by a smoother transition. the author recommends that this function be used everywhere beyond 2-4 the thickness of the boundary layer based on the upstream length. 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). 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. mass balance is maintained by the reduction of the density. the maximum turning angle is much larger than the maximum deflection point because the process is isentropic. an increase in the Mach number can result in a very significant radial velocity. (1 + ) the radial velocity is small .15. In fact. Regardless of the assumptions that were used in the construction of these models. On top of the complication of the turning point.37) The change in the deflection angle is calculated by . At this point (r = 0) these models fail to satisfy the boundary conditions and something else happens there.5) 15. the fact remains that there is a radial velocity at Ur (r = 0) = constant.4 The Working Equations for the Prandtl-Meyer Function ν2 − ν1 = ν (M2 ) − ν (M1 ) (15. At a small Mach number. it was assumed that radial velocity is only a function of θ.36) The maximum of the deflection point and the maximum turning point are only a function of the specific heat ratios. In this case.3. However. For example.

5: Expansion of Prandtl-Meyer function when it exceeds the maximum angle 15. 15. and the specific heat. it is Fig. M1 . the drag.38). 15.5 d’Alembert’s Paradox In ideal inviscid incompressible flows. The relation between P2 and P4 is such that the flow depends on the upstream Mach number.272 CHAPTER 15. is D=2 w (P2 − P4 ) = w(P2 − P4 ) 2 (15. Again. The net force in flow direction. the move3 ment of body does not 1 2 encounter any resistance.7: A simplified diamond shape to illustrate the suassumed that the fluid is inpersonic d’Alembert’s Paradox viscid.38) It can be observed that only the area that “seems” to be by the flow was used in expressing equation (15. and θ1 w θ2 this paradox is examined θ2 θ1 here.7). 4 This result is known as d’Alembert’s Paradox. . PRANDTL-MEYER FUNCTION sl ip li ne Maximum turning Fig. 4 Supposed that a two– dimensional diamond–shape 2 1 body is stationed in a su3 personic flow as shown in Figure (15. k .

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 .

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

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

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

COMPUTER PROGRAM .1: Schematic diagram that explains the structure of the program APPENDIX A. A.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. 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 . 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.{1. int isTex = yes.{1.4}|} \caption{ ?? \label{?:tab:?}}\ . variableName. ******************************************* 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. whatInfo = infoStandard .. Mx = 2. shock etc) infoTube print tube side info for (Fanno. variableName = MxV.4}|D. You must use longtable style file and dcolumn style files.A.4}|D.{1. isTex. s.2.0 . isRange.makeTable(whatInfo.4}|D. You can just can cut and paste it in your latex file..{1.4}|D.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. int isRange = no.4}|D.{1..{1. variableValue)..... ******************************************* \setlongtables \begin{longtable} {|D.

176& 2.3 Program listings Can be download from www. COMPUTER PROGRAM {$\rule[-0.1in]{0.000& 0.3 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.potto.6164 \ A.pt}{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.5854& 3. .org.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} & 1.773& 0.pt}{0.1in]{0.152& 0.

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

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

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