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

.1. . . . . . . . . . . . . .CONTENTS 12 Evacuating/Filing Chambers under External Volume Control 12. . . . . . . . . .2 Summary . . . . .4. . 14. . . . . . . . . . 15. . . . . . . . . . . . .3 Upstream Mach Number. .1 Preface to Oblique Shock . . . .5 Flow in a Semi–2D Shape . . .3 Introduction to Zero Inclination . . . .8 Maximum Value of Oblique shock .11Oblique Shock Examples . . . . . . . .6 Flat Body with an Angle of Attack . . . . . . . . . . .4. . . . 12. . . . . . . 15. . . . . . . . . . .2 Geometrical Explanation . . . .7 Examples For Prandtl–Meyer Function . . . . . . . .1. . . . . . . .4. . . . . . . . . . . . . . . δ and θ . . . . M1 . . . .1 Alternative Approach to Governing Equations . . . .2 When No Oblique Shock Exist or When D > 0 . .3 Direct Connection . . . . .6 Small δ “Weak Oblique shock” . 15. . . 15. . . . . . . . . . . . . . . . . . .1 Rapid Process . . . . . . . . . . 15. . . M1 . .4. . . . . . . . . . . .13Optimization of Suction Section Design . . 12. . . . . . δ 14. . . . .5 Summary . . . . 14. . . . .4. . . . . . θ . . . . . . . . . . .2. . . . . . . . . . . . . . . . . . . . . .2. .8 Combination of the Oblique Shock and Isentropic Expansion . 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. . . . . . .9 Detached shock . . .4. . . . . . 14. . . . . . . . 12. . .2. . . . . .4. and Deflection Angle. . . . . . . .3 Oblique Shock . . .4. . 14. . . . . . .1 Introduction to Oblique Shock . . . . . 14. . . . . . . . . . . 15. .7 Close and Far Views of the Oblique Shock . . . . . .4 Solution of Mach Angle . . . . . . . . . . . . . . . . .4. .2 Examples . . . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.4. . . 14. . . . . . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . and Shock Angle. . . . . 13 Topics in Unsteady one Dimensional gas dynamics 14 Oblique-Shock 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Upstream Mach Number. . . . . .4 Given Two Angles. 14. 14. . . . . . . .1 Introduction . . . .4 The Working Equations for the Prandtl-Meyer Function . . .10Issues Related to the Maximum Deflection Angle . . . 14. 14. . 14. .5 d’Alembert’s Paradox . . . . . . . . . . . . . . . . . . 15. . .2 Introduction to Prandtl–Meyer Function . . . . . . . . . . . . . . . . .2 Introduction . . . . . . . . . . . .4.3 The Maximum Turning Angle . . . . . . . . . . 14. 14. . . . . . . . . . .6 Appendix: Oblique Shock Stability Analysis . .2. . . . . .1 General Model . .2 Comparison And Limitations between the Two Approaches 15. . . . . . 14. . . . .4. 12. . . . . . . . . . . . . . . . . . . . . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. . . . . . . . . . . . . . 14. . . . . . . . . . . 15. . . . . . . . . . 14.1. . . . . . .4. . . . . . . . . . . . .12Application of Oblique Shock . . . . . . . . .

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

. . . . . . . . . . . . 3.8 Photo of Lord Rayleigh . . . . . . . . . 1. . . . . . . . . . . . . . 1. . . . . . . . . . . k = 1. . . . . . . .10 The photo of Gino Fanno approximately in 1950 . . . . . . . . . .11 Photo of Prandtl . . . . . . . 1. . . . . . . .9 Portrait of Rankine . . . . . . . . . . . .LIST OF FIGURES 1. . 1. . . . . . . . 43 4. .5 Portrait of Galileo Galilei . . . . . 46 ix . . . . .6 Photo of Ernest Mach . . . . . . . Steam and Gas Turbine . . . . .4 Flow rate as a function of the back pressure taken from Stodola 1927 Steam and Gas Turbines . . . . . . . .1 Flow of a compressible substance (gas) through a converging–diverging nozzle. . . . 1. . . . . .2 Stationary sound wave and gas moves relative to the pulse . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 9 11 12 16 17 17 18 19 20 21 22 28 28 32 4. . . . . . . . . . . . . . . . . . . . .3 The stagnation properties as a function of the Mach number. . 1. . . . . . . . . . .1 The shock as connection of Fanno and Rayleigh lines after Stodola. . . . . .4 Control volume inside a converging-diverging nozzle. . . . . . . . . . . . . . . . . . . . . . . . . 3. . . . . . . . . . . . . . . . . 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines . 41 4. . . . . . . . . . . . . . . . . . . . . . . .1 A very slow moving piston in a still gas . . . . . . . . . . Steam and Gas Turbine . . . . . . . . 1. . 1.12 The photo of Ernst Rudolf George Eckert with the author’s family . . . . . . . .4 44 4. . . . . . . . . . . . . . . . . . . . . . . . . . .2 The schematic of deLavel’s turbine after Stodola. . . . . . . . .2 Perfect gas flows through a tube . . .7 The photo of thebullet in a supersonic flow not taken in a wind tunnel 1.3 The Compressibility Chart . . . 1. 3. . . . . . . . . . . .

. . .3 6. . .2 The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. . . 5. . . . . 5. . . . . . 4.4 The flow in the nozzle with different back pressures . .10) . . . . . . . . . . . . .17 Figure for Example (5. . . . . .14 Schematic of a piston pushing air in a tube. . . . . . . . . . . . . 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . k . 5. .10 The maximum of “downstream” Mach number as a function of the specific heat. . . . . . . . . . . . . . . 5. .6 Comparison between a stationary shock and a moving shock in a stationary medium in ducts. . . . . . . . . . . . . . . . . . 5. . . . . 5. . . . . . . . . . . . . . . . . . . .8 A shock moves into a still medium as a result of a sudden and complete opening of a valve . . . . . . . . . . . . . . . . . . . . . . . . . . .9 The number of iterations to achieve convergence.11 A shock moves into a moving medium as a result of a sudden and complete open valve. . . . . . . . . 5. . . . . . . .5 Comparison between stationary shock and moving shock in ducts . 5. . . . . . . .10) . . . . . . . . . . . . . . . . . . . . . .7) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Schematic to explain the significances of the Impulse function . 4.12 The results of the partial opening of the valve.15 Figure for Example (5. . 5. . .7 The moving shock Mach numbers as a result of a sudden and complete stop. . . . .x LIST OF FIGURES 4. . . . . . Description to clarify the definition of diffuser efficiency . . . . . . . . . .18 The results for Example (5.” . . . . . . . . . 5. . . .3) . . . . . . . . . . . . . 6. .3 The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. . . . . . . . . . . . but it can also be viewed as a one– dimensional shock wave. . . 5. . 5. . . . . . . . . . . . . . . . .8) . . . . . . . . . . . 5. . . . . . . 4. . .13 A shock as a result of a sudden and partially a valve closing or a narrowing the passage to the flow . 5. . . . . . . . . . . . . . . . .6 Various ratios as a function of Mach number for isothermal Nozzle . . . . . . . . . . . . . . . . 137 . 4. . . . .1 6. . . . 5. 5. . . . . . . . . . . .7 The comparison of nozzle flow . Schematic of a supersonic tunnel in a continuous region (and also for example (6. . . . . . . . . . . . . . . . . .10 Schematic of a flow of a compressible substance (gas) thorough a converging nozzle for example (4.2 6. .5 The relationship between the cross section and the Mach number on the subsonic branch . .16 The shock tube schematic with a pressure ”diagram. . . . . . . . . .1 A shock wave inside a tube. . . . . . . . . . . . . . . . . . .8 Comparison of the pressure and temperature drop as a function of the normalized length (two scales) . . . . . . . . . . . A nozzle with normal shock . . . . . . . . . . . . . . 5.4 The ratios of the static properties of the two sides of the shock. . . . . . . . . . . . . . . . . . . . 4. . . . . . 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. . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 Control volume for isothermal flow . . . . . . . . . . . .

. 181 fL The extra tube length as a function of the shock location. . . . . . . . . . .14 9. . . . . . . . . . . M1 to the tube as a function fL of 4D supersonic branch . .1 9. . . . .3 . . . . . . . . . .4 The pressure assumptions in the chamber and tube entrance . . . . . . . . . . . . . . temperature relationships as a function of the Mach number for isothermal flow .3 The Mach number at the entrance to a tube under isothermal flow fL . . . . 185 10. . . . . . . . . . . . . . . . 189 10. . . .9 9. . . 218 . .8 9. . . . .6 The reduced time as a function of the modified reduced pressure . . . . . . . . . . .1 The control volume of the “Cylinder” .5 The reduced time as a function of the modified reduced pressure . . . . . . . . 163 Schematic of Example (9. . . . .1) . . 11. . . . . . . . . . . . . k . . . . .16 9. . . . . . . 166 The maximum length as a function of specific heat. . . . . . . . .5 9. . . . 179 fL fL Fanno Flow Mach number as a function of 4D when the total 4D = 0. . . . . . . . . . 201 202 202 203 210 212 12. 171 The development properties in of converging nozzle . . . . . . . .18 Control volume of the gas flow in a constant cross section . . . . . .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. . . . . . . . . . .13 9. .3 The basic functions of Rayleigh Flow (k=1. . . 153 model as a function 4D 9.6 9. . 170 fL The effects of increase of 4D on the Fanno line . .1 The control volume of Rayleigh Flow . .175 fL fL The pressure distribution as a function of 4D for a short 4D . . . . . 4D supersonic branch . . 11. . . 143 8. . . . 11. .10 9. .15 9. . . . .11 9. . . . . . . . .3 for Fanno Flow . . . . . . . . . . . . . . . . . . .4 9. .3 9. . . . . . .2 Description of the pressure. . . . . .2) . . . . . . . . . . . . . 172 The Mach numbers at entrance and exit of tube and mass flow rate fL for Fanno Flow as a function of the 4D . . . . 11. . . . . . . . . . . . . . . . . . . . . 191 10. . . . . . . . . . . . . 155 Various parameters in Fanno flow as a function of Mach number . .2 The Temperature Entropy Diagram For Rayleigh Line . . . . . . . . 177 4f L 4f L . . . . . . . . 180 Schematic of a “long” tube in supersonic branch . . . . . . . . . . . . . . . . . . .LIST OF FIGURES xi 8. . . . . 195 11. . . . . . . . . . . . . . . . . . 174 fL M1 as a function M2 for different 4D for supersonic entrance velocity. . . . . . . 11. . .12 9. 182 The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow .17 9. . .4) . . . . . . . . . 181 The maximum entrance Mach number. . . . . . . . . . . . . . . .2 9.3 A schematic of the control volumes used in this model . . . . . . . . . 164 The schematic of Example (9. . . . . . .1 The two different classifications of models that explain the filling or evacuating of a single chamber . . . . .7 9. . . . . . . . . . 173 4f L M1 as a function M2 for various D .

. .21Typical examples of unstable and stable situations . . This photo is courtesy of Dr. . . . . 229 14.4) . . . . 235 14.12The schematic for a round–tip bullet in a supersonic flow . 264 15. . . .0. . .6 The various coefficients of three different Mach numbers to demonstrate that D is zero . . . . . . . . . . . .240 14.11A local and a far view of the oblique shock. . a Research Professor at Cuernavaco University of Mexico. . 243 14. . . . . 250 14. . . .10Schematic of finite wedge with zero angle of attack. 248 14. . . 236 14. . . . .3 The schematic of the turning flow . . .1 A view of a straight normal shock as a limited case for oblique shock 227 14. . . the strong shock. . . .17Two variations of inlet suction for supersonic flow . . . . . . . . .2 The pressure ratio as a function of the dimensionless time for chokeless condition . . According to his measurement. . shock angle. . 239 14. . 247 14. . . . . . . . . . . . . . . . . .2 The angles of the Mach line triangle .3 The pressure ratio as a function of the dimensionless time for choked condition . . . 251 14. . . 241 14. . . .4 The pressure ratio as a function of the dimensionless time . . . . . . . . . 244 14. . . . . . . . . . Notice that both have a maximum point and a “no solution” zone. . . . . .19Schematic for Example (14. . . . . . . 252 14. . . . . . . . . . . . . . . . . . . . . . . . 253 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . and exit Mach number for M1 = 3 . .15Oblique shock occurs around a cone. . .18Schematic for Example (14. . . . . . .xii LIST OF FIGURES 12. . . . . . . . . and the weak shock coexist. However. . . . . Prandtl-Meyer function approaches closer to a zero deflection angle.2. . . . . . 254 14. . . . . . . . . . . . . . . 263 14. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 12. . . . . . . . . . . . . .20Schematic of two angles turn with two weak shocks . . It can be noticed that the normal shock. . . . . . . . . . . . . . . . .5) . 251 14. . . . .8 The calculation of D (possible error). . . . . .4 Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. 224 14. . . . . . . . . . . . . . Grigory Toker. . . . . . .5 The view of a large inclination angle from different points in the fluid field.3 A typical oblique shock schematic . . . . . . .2 The regions where oblique shock or Prandtl–Meyer function exist. . 223 12. . . .22The schematic of stability analysis for oblique shock. . . . . . . . . which is around zero. . . 265 15. . . . . . . . . . . . . . . . 266 . . . . . . . . . . . . . .9 The possible range of solutions for different parameters for given upstream Mach numbers . . .13The schematic for a symmetrical suction section with Mach reflection 248 14. . . 245 14. . . . . . .14The “detached” shock in a complicated configuration sometimes referred to as Mach reflection . . . . . . . . . . . . 265 15. . . .16Maximum values of the properties in an oblique shock . . . . . . . . . . . . . . . . . . . . the cone half angle is 15◦ and the Mach number is 2. . . . . . . . . . . . . . . 228 14. . . . .1 The definition of the angle for the Prandtl–Meyer function. . . . . . . . .7 The Mach waves that are supposed to be generated at zero inclination. . . . . . . . . . . . . .

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

xiv LIST OF FIGURES .

. . . . . . . . .3 Solids speed of sound. . . . . . . . . . . . . . . . . . . . . . 4. . . . . 4. . . . . . . . . . . . . . . 115 5. .3 Table for shock propagating from suddenly opened valve (k=1. . .1 The shock wave table for k = 1. . . . . 119 5.2 Isentropic Table k=1. . .3 Table for shock propagating from suddenly opened valve (k=1. . . . . . . . . 1972 . . . . . . after Aldred. . . . . . London: Fountain Press. . .4 .3) 119 5. . . . .1 Water speed of sound from different sources . . . . . . 4. . . . . . . 116 5. . . London:Fountain Press. .4 Table for shock propagating from a suddenly opened valve (k=1.2 Table for Reflective Shock from suddenly closed valve (end) (k=1.4 Table for shock propagating from a suddenly opened valve (k=1. . . . . . . 3.1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . . . . . . . . . . . . . . . . . 4.4) . . . 35 36 37 58 59 60 62 63 73 74 5. .1 Fliegner’s number and other paramters as function of Mach number 4. . 117 5.4 (continue) . . . . . .2 Liquids speed of sound. . . . . .3 Isothermal Table . . . .3 Isothermal Table (continue) . . . . . Manual of Sound Recording.1 Fliegner’s number and other paramters as function of Mach number (continue) .1 The shock wave table for k = 1. . . . . . .4 Table for shock propagating from a suddenly opened valve (k=1. . . . 3. . . . .2 Isentropic Table k = 1. . . .3) 120 5. .4)(continue)118 5. . . . John. . . . .4 . Manual of Sound Recording. after Aldred.LIST OF TABLES 3. . . . . . . . . . . . . . . . . . . . . . . . . John. . . .4) . . . . . . 117 5. . . . . 1972 . 4. . . . . .2 Table for a Reflective Shock from a suddenly closed end (k=1. 4.1 The shock wave table for k = 1. . . 118 5. . . . . . . . . . . .4 (continue) . . .3) 121 xv . . . .4 (continue) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4) .

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

Boston. 51 Franklin St. Fifth Floor. Preamble The purpose of this License is to make a manual. but changing it is not allowed. this License preserves for the author and publisher a way to get credit for their work. because free software needs free documentation: a free program should xvii . or other functional and useful document ”free” in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it. either commercially or noncommercially.2001.2002 Free Software Foundation. which means that derivative works of the document must themselves be free in the same sense. textbook. We have designed this License in order to use it for manuals for free software. MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document. GNU Free Documentation License Version 1. Inc. while not being considered responsible for modifications made by others.2. November 2002 Copyright ©2000. Secondarily.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. This License is a kind of ”copyleft”. It complements the GNU General Public License. which is a copyleft license designed for free software. with or without modifying it.

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

3 0.0 0. But more than that. better discussions or better explanations are all welcome to these books. this method is expected to accelerate the creation of these high quality books. 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).0 0. Unlike a regular book. the contributions to books can be done by many who are not experts. contributions can be made from any part of the world by those who wish to translate the book.0. Thus. . 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. Thus. chapters or sections can be added after the skeleton has been written.CREDITS xxxiii adding a question and perhaps the solution.” In this process.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. the traditional role of author will be replaced by an organizer who will be the one to compile the book. Someone has to write the skeleton and hopefully others will add “flesh and skin. these works will have a version number because they are alive and continuously evolving.0.) can be redone by anyone who has the time to do it.0. Thus. Additionally. 5 Data are not copyrighted. Even if not complete.0 0.1 0.0. Nevertheless.0.01 0.0.0 0. The undersigned of this document intends to be the organizer/author/coordinator of the projects in the following areas: project name Die Casting Mechanics Statics Dynamics Strength of Material Compressible Flow Fluid Mechanics Thermodynamics Heat Transfer Open Channel Flow Two/Multi phases flow progress alpha not started yet not started yet not started yet not started yet early beta alpha early alpha not started yet not started yet not started yet remarks version 0.0.0.0 0. It is also hoped that others will contribute to the question and answer sections in the book. These books should be considered more as a project than to fit the traditional definition of “plain” books. graphs and etc. some errors are possible and expected.4 0. These data (tables. The organizer of the book in some instances will be the main author of the work.0. It is hoped that the book will be error-free. These books are written in a similar manner to the open source software process.

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

2006 The title of this section is change to reflect that it moved to beginning of the book. While all these requests are important.4. However. If one want to use the software engine it is okay and permitted by the license of this work.2 It was surprising to find that over 14. For example. Menikoff pointed to this inconsistency. The download to this mount is over 25. this author cannot add all the things that he was asked and want to the book in instant fashion. Dr. 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).000.S. 15. especially for Microsoft. The moving shock issues are not completed and more work is needed also in the shock tube. xxxv . and the author is apologizing for this omission. Nevertheless. the ideas of moving shock will reduced the work for many student of compressible flow. Several sections were add to this book with many new ideas for example on the moving shock tables. I am sorry will not be entertaining work for non Linux/Unix systems. 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. Version 0. I also got request from a India to write the interface for Microsoft.3 Sep.4. one of the reader ask why not one of the example of oblique shock was not turn into the explanation of von Neumann paradox.A./Arizona) and some other reactions. While it moves earlier but the name was not changed. the time is limited and they will be inserted as time permitted.000 downloaded and is encouraging to receive over 200 thank you eMail (only one from U.Prologue For This Book Version 0.

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

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

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

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

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

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

7 had several bug fixes and add two angle calculations to the oblique shock.1.7 Version 4.xlii LIST OF TABLES Version 4.1. Change the logtable to tabular environment for short tables. .

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

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

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

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

Application Marcell–Taylor (from the notes) Examples Transient problem .7 xlvii Fanno Flow More examples: various categories Some improvement on the software (clean up) Real gas effects (compressible factor) Tablated gas Rayleigh Flow To mature the chapter: discussion on the “dark” corners of this model. Provide discussion on variations of the effecting parameters. 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++.VERSION 4. The FORTRAN version will not be included.1. Oblique Shock Add application to design problems Real Gas effects Prandtl–Meyer The limitations (Prandtl-Meyer).

xlviii LIST OF TABLES .

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

While there are excellent books about the history of fluid mechanics (hydraulic) see for example book by Rouse6 . including steam. For instance. the flow of natural gas in a pipe system. should be considered a compressible flow. The die casting process is a process in which liquid metal. Compressible flow deals with more than air. 1957) . Some aspects of these systems require consideration of the unique phenomena of compressible flow. a problem that led to the Fanno flow model. These processes include the flow of gas in the exhaust system of an internal combustion engine. many books on compressible flow ignore the isothermal flow5 .s. Traditionally. Aero Engineers are not the only ones who have to deal with some aspect of compressible flow.2 CHAPTER 1. For manufacturing engineers there are many situations where the compressibility or compressible flow understating is essential for adequate design. History of Hydraulics (Iowa City: Institute of Hydraulic Research. a common method of heating in the u. 6 Hunter Rouse and Simon Inc. 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. nitrogen and helium. Clearly.. in a matter of milliseconds. There are very few courses that deal with isothermal flow. 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). 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. natural gas. the control engineers who are using pneumatic systems use compressed substances. For instance. Compressible flow also includes flow around bodies such as the wings of an airplane. injection molding. These processes include situations not expected to have a compressible flow. In fact. mostly aluminum. The air is displaced by the liquid metal in a very rapid manner. and also gas turbine. etc. is injected into a mold to obtain a near final shape. and is considered an external flow. a greater emphasis is on the internal flow. This doesn’t in any way meant that the important topics such as shock wave and oblique shock wave should be neglected. The cooling of some manufacturing systems and design of refrigeration systems also utilizes compressed air flow knowledge. INTRODUCTION 1.2 Why Compressible Flow is Important? Compressible flow appears in many natural and many technological processes. such as manufacturing process such as the die casting. therefore the compressibility has to be taken into account. 1. This book contains several chapters which deal with external flow as well. In this book.

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

In the twentieth century the flight industry became the pushing force. Newton’s equation is missing the heat ratio.1 Early Developments The compressible flow is a subset of fluid mechanics/hydraulics and therefore the knowledge development followed the understanding of incompressible flow. an Italian scientist. Therefore some sound for gas as c = √ referred to coefficient k as Maxwell’s coefficient. aerospace engineering played a significant role in the development of this knowledge. capillary flow and phase change. Robert Boyle discovered that sound waves must travel in a medium (1660) and this lead to the concept that sound is a pressure change. the modern basic classes became “solidified. Galileo. people found that sound travels in different speeds through different mediums. measured the choking but theoretical science did not provide explanation for it (or was award that there is an explanation for it. Understandably. for a long time the question of the speed of sound was bounced around. INTRODUCTION To add history from the work. while most later contributions were driven by necessity. . so almost all text books in this field are written from an aerodynamic prospective. There is no clear evidence as to who came up with this concept. The early approach to the speed of sound was by the measuring of the speed of sound.” Contributions by researchers and educators from other fields were not as dominant and significant. Speed of Sound The idea that there is a speed of sound and that it can be measured is a major achievement. dominated the field. 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). as well as others like Shapiro. was one of the earliest contributors to our understanding of sound. Early contributors were motivated from a purely intellectual curiosity. Giants like Prandtl and his students like Van Karman. During that time. √k (late 1660’s). As a result. once there was a realization that sound can be measured. Marin Mersenne was the first to measure the speed of sound in air (1640). Maxwell was the first to derive the speed of kRT from particles (statistical) mechanics. Topics that should be included in this history review but that are not yet added to this section are as follows: Multi Phase flow. 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. 1.4 CHAPTER 1. Other milestones in the speed of sound understanding development were by Leonardo Da Vinci. sound) was a major part of Galileo’s work. but some attribute it to Galileo Galilei: 166x.). Dealing with the difference between the two speeds (light. However. who discovered that sound travels in waves (1500).3.

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

Shock was assumed to be one dimensional and mass. Thus. Nevertheless. he was not able to analyze it (neither was he aware of Poisson’s work or the works of others. momentum. New York. Scottish engineer. ”On the thermodynamic theory of waves of finite longitudinal disturbance. 277-288. What they could not prove or find was that shock occurs only when upstream is supersonic. He utilized the variations of the air density to clearly show shock line at the front of the bullet. 161-243.7). and energy equations18 lead to a solution which ties the upstream and downstream properties. Polytech. Classic papers in shock compression science. Shock Compression Condens. I. While Mach proved shock wave and oblique shock wave existence. 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.” Philos. 3-97.e. one anecdotal story suggests a way to make money by inventing a better killing machine for the Europeans. 1998 17 Pierre Henri Hugoniot. Classic papers in shock compression science. He worked in Glasgow. He further showed that the discontinuity occurs at M = 1 and realized that the velocity ratio (Mach number). At the time. Springer. Shock Compression Condens. He also observed the existence of a conical shock wave (oblique shock wave). Matter. 16 William John Macquorn Rankine. Mach had good understanding of the fundamentals of supersonic flow and the effects on bullet movement (supersonic flow). 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. 1998 18 Today it is well established that shock has three dimensions but small sections can be treated as one dimensional. and not the velocity. Trans. High-press. Later. . II” J. Scotland UK. 1-125. 160 (1870). French engineer. he brought confidence to the theoreticians to publish their studies. At that time many inventions dealt with machine guns which were able to shoot more bullets per minute. A need for bullets that could travel faster to overcome these obstacles was created. Mach used his knowledge in Optics to study the flow around bullets. 245-358.6 CHAPTER 1. Therefore. the jump conditions were redeveloped and now named after Rankine16 and Hugoniot17 .. 57 (1887). Rankine and Hugoniot. New York. High-press. 1851-1887. direction of the flow. no known15 equations or explanations resulted from these experiments. others expanded Rankine-Hugoniot’s 15 The words “no known” refer to the undersigned. Springer. 133-147. While the machine gun turned out to be a good killing machine. It is possible that some insight was developed but none of the documents that were reviewed revealed it to the undersigned. Back to the pencil and paper. INTRODUCTION powerful/long–distance shooting rifles/guns. Mach’s paper from 1876 deals with the flow around bullets. He was the first to photograph the shock wave. 1820-1872. ”Sur la propagation du mouvement dans les corps et sp’ecialement dans les gaz parfaits. Ec.). redeveloped independently the equation that governs the relationship of the shock wave. Matter. defense techniques started to appear such as sand bags. 58 (1889). Mach’s paper from 1876 demonstrated shock wave (discontinuity) and suggested the importance of the ratio of the velocity to the speed of sound. part II. i. is the important parameter in the study of the compressible flow. Mach’s contributions can be summarized as providing an experimental proof to discontinuity.

19 To . A solution Rayleigh lines after Stodola. Theodor Meyer (Prandtl’s student) derived the conditions for oblique shock in 190821 as a byproduct of the expansion work. Berlin. There were until recently (version 0. Yet. I.1)). 1. to be add to oblique shock Since early in the 1950s the analytical relationships between the oblique shock. ub. Steam and Gas was found by Landau & Lifshitz Turbine and expanded by Kolosnitsyn & Stanyukovich (1984). The first full analytical solution connecting the angles with upstream Mach number was published in this book version 0. the relationship between upstream Mach number and the angles. it took over 50 years for Prandtl to arrive at and to demonstrate that the shock has only one direction20 . The probable reason that analytical solution was not published because the claim chapter. In fact Prandtl is the one who introduced the name of Rankine-Hugoniot’s conditions not aware of the earlier developments of this condition. 1908. Taylor from England as the proof (of course utilizing the second law) 21 Theodor Meyer in Mitteil. 20 Some add discussion about the general relationships. Stodola suggested the graphical solution utilizing the Fanno line. Thus. the supersonic branch is missing from his understanding (see Figure (1. page 62.1. Today this equation/condition is known as Prandtl’s equation or condition (1908). ¨ Forsch-Arb.3. The fact that the conditions and direction were known did not bring the solution to the equations. In fact. HISTORICAL BACKGROUND conditions to a more general form19 . 62. view the work of G. It was probably later that Stodola (Fanno’s adviser) realized that the shock is the intersection of the Fanno line with the Rayleigh line. No.3. The “last nail” of understanding was put by Landau. deflection angle. 7 Here.1: The shock as connection of Fanno and munist regimes. the second law has been around for over 40 years and yet the significance of it was not was well established. shock angle. and Mach number was described as impossible to obtain. a Jewish scientist who worked in Moscow University in the 1960’s during the ComFig.3 of this book) several equations that tied various properties/quantities for example.

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

3. 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. 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. there was no realization that the flow is choked just that the flow moves faster than speed of sound. It is likely that others had noticed that flow is choked. Up to this version 0. Later. 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.4. but did not produce any model or conduct successful experimental work.” Leipzig 1899.1. 1. 68–72. p. The first to discover (try to explain) the choking phenomenon was Rayleigh29 . As a result deLavel’s nozzle was invented by Carl Gustaf Patrik Fig. this book is first to publish this tables). “Theorie die Turbinen. . leading to the study and development of Rayleigh flow. 28 Fliegner Schweizer Bauztg. page 268 f. 1. No one was looking for or expecting the choking to occur. but after a certain point it did matter (because the steam was choked). 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.3. Only after the dimensionless presentation of the problem and the construction of the moving shock table the problem became trivial. The introduction of the steam engine and other thermodynamics cycles led to the choking problem. HISTORICAL BACKGROUND 9 The moving shock and shock tube were study even before World War Two. Choking wasn’t clearly to be observed. even when researcher stumbled over it. The theoretical first work on this issue was done by Zeuner..2: The schematic of deLavel’s turbine afde Laval in 1882 and first successful ter Stodola. trial and error method was the only method to solve this problem. and when it was found the significance of the choking phenomenon was not clear. According the thermodynamics theory (various cycles) the larger heat supply for a given temperature difference (larger higher temperature) the larger the output. Vol 31 1898.3 of this book (as far it is known. Yet.3 Choking Flow The choking problem is almost unique to gas dynamics and has many different forms. Steam and Gas Turbine operation by another inventor (Curtis) 1896 used in steam turbine.

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

the spinning arm was no longer used in testing.3. 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. due to the circular motion. . 1907. There are two main models or extremes that describe the flow in the nozzle: isothermal and adiabatic. The arm’s circular motion could reach a velocity above the speed of sound at its end. Between the turn of the century Fig. HISTORICAL BACKGROUND Nozzle flow 11 The first “wind tunnel” was not a tunnel but a rotating arm attached at the center. 1. 36 Bendemann Mitteil uber ¨ Forschungsarbeiten. several models that explained choking at the throat have been built. Berlin.1. At the end of the arm was the object that was under observation and study.3: The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines and 1947-48. when the first supersonic wind tunnel was built. As a result. No. 37.545) that creates the maximum flow rate. in 1904 the Wright brothers demonstrated that results from the wind tunnel and spinning arm are different. Yet. A different reason to study the converging-diverging nozzle was the Venturi meter which was used in measuring the flow rate of gases.

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

8 (relative of velocity of the body to upstream velocity) a local Mach number (local velocity) can reach M = 1.1. At that stage. no copy of the thesis can be found in the original University and perhaps only in the personal custody of the Fanno family40 . However at a Mach number above 0. the question cannot be answered yet. HISTORICAL BACKGROUND 13 Mumenthaler from UTH University. Thus. But because the actual Fanno’s thesis is not available. This fact was evident in industrial applications where the expectation of the choking is at Mach one. If it turned out that no one had done it before Shapiro. was ever produced to verify this finding. but can be explained by choking at a lower Mach number. flow that is dominantly adiabatic could be simplified and analyzed. 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. No experimental evidence. In reality. the Fanno model wasn’t able to produce a prediction useful for the industry. f based radius is only one quarter of the Darcy f which is based on diameter . 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.3. The author invites others to help in this information. a shock wave occurs which increases the resistance. which built on Hunter Rouse’s (194x) work made Darcy– Weisbach’s equation universally useful. known by the undersigned. Additionally an understating of the supersonic branch of the flow was unknown (The idea of shock in tube was not raised at that time.3. 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.8 Mach or more the flow must be considered to be a compressible flow.4 External flow When the flow over an external body is about . Without the existence of the friction factor data. this flow model should be called Shapiro’s flow.). The choking was assumed only to be in the subsonic flow. 1. Fanno attributes the main pressure reduction to friction. 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. The friction factor is the main component in the analysis as Darcy f 41 had already proposed in 1845. The arrival of the Moody diagram. Shapiro organized all the material in a coherent way and made this model useful. the flow is √ choked somewhere between 1/ k to one for cases that are between Fanno (adiabatic) and isothermal flow.

He introduced the lifting line theory. slide rule is sold for about 7. but the propellers (which are also airfoils) did exceed the speed of sound.14 CHAPTER 1.5$ on the net. one of Prandtl’s assistants. thus exhibiting inefficiency. simplified the shock equations so that they became easy to use. Lanchester’s concept of circulatory flow (1894). there is no reason to teach it in a regular school. and the KuttaJoukowski circulation theory of lift (1906). Prandtl claimed that he was not aware of Lanchester’s model when he had begun his work in 1911. Yet.” Cambridge University Press. . After the First World War aviation became important and in the 1920s a push of research focused on what was called the compressibility problem. Anderson’s Book “History of Aerodynamics and Its Impact on Flying Machines. Briggs and Dryden showed that the problem was related to the shock wave. This is because the English Astronomer Frederick Lanchester published the foundation for Prandtl’s theory in his 1907 book Aerodynamics. After World War Two. Theoreticians tried to simplify the equations and arrive at approximate solutions representing specific cases. Ackeret. which was published in 1918-1919 and enabled accurate calculations of induced drag and its effect on lift42 . John D. Airplanes could not yet fly fast. However. Practitioners like the Wright brothers relied upon experimentation to figure out what theory could not yet tell them. 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. That is the “dinosaur44 ” reason that even today some instructors are teaching mostly the perturbations and asymptotic methods in Gas Dynamics classes. This problem led to two consequences. Ludwig Prandtl in 1904 explained the two most important causes of drag by introducing the boundary layer theory. Prandtl’s boundary layer theory allowed various simplifications of the Navier-Stokes equations. Prandtl created his thin–airfoil theory that enabled the calculation of lift for thin. By the way. Von Karman reduced the equations for supersonic flow into a single equation. especially at the propeller tips. J. More information on external flow can be found in . He later contributed to the PrandtlGlauert rule for subsonic airflow that describes the compressibility effects of air at high speeds. INTRODUCTION Euler equations) were considered unsolvable during the mid 18xx because of the high complexity. During World War I. Later. Prandtl worked on calculating the effect of induced drag on lift. 43 This undersigned is aware of only one case that these methods were really used to calculations of wing. the research had continued and some technical solutions were found. 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. See for the full story in the shock section. 1997 42 The English call this theory the Lanchester-Prandtl theory. Today these methods of perturbations and asymptotic are hardly used in wing calculations43 . cambered airfoils. Meanwhile in Germany. Some of the solutions lead to tedious calculations which lead to the creation of Computational Fluid Dynamics (CFD). Examples of such work are Hermann von Helmholtz’s concept of vortex filaments (1858). Prandtl’s student. 44 It is like teaching using slide ruler in today school.

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

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

. He graduated from Vienna in 1860. Einstein was greatly influenced by it. This was one of the primary sources of inspiration for Einstein’s theory of General Relativity. Johann. He was not able to attach the camera to the arm and utilize the remote control (not existent at that time). 1.G. Note it was not taken in a wind tunnel sonic speeds.3. Ernest Mach (1836 . and in 1918. Mach’s shadowgraph 47 It is interesting to point out that Prague provided us two of the top influential researchers[:] E. when Czechia was still a part of the Austro–Hungary empire. Mach was also a great thinker/philosopher and influenced the theory of relativity dealing with frame of reference. Mach’s revolutionary experiment demonstrated the existence of the shock wave as shown in Figure 1. At first he received a professorship position at Graz in mathematics (1864) and was then offered a position as a professor of surgery at the university of Salzburg. HISTORICAL BACKGROUND Ernest Mach (1838-1916) 17 Ernst Mach was born in 1838 in Chrlice (now part of Brno). was a high school teacher who taught Ernst at home until he was 14.7: The Photo of the bullet in a supersonic flow that could take a photo at superMach made. Mach and E.R. but he declined. He then turned to physics.7.1.” 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. Mach’s father. before he entered the university of Vienna were he studies mathematics. when he studied in Kromeriz Gymnasium. and in 1867 he received a position in the Technical University in Prague47 where he taught experimental physics for the next 28 years. In 1863.6: Photo of Ernest Mach interested also in physiology of sensory perception. Eckert. 1. There Mach wrote his thesis ”On Electrical Discharge and Induction. physics and philosophy. His experiments required exact timing. he named it Mach’s Principle.1916) published Die Machanik in which he formalized this argument. Later.

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

1. He received the Nobel Prize in 1904. Rankine. Rayleigh died on June 30. the Earl of Balfour (of the famous Balfour declaration of the Jewish state). Lord Rayleigh. HISTORICAL BACKGROUND 19 show his capacity for understanding everything just a little more deeply than anyone else. Rankine was well rounded interested beside the energy field he was also interested in civil engineering. . Rankine never married. and in 1887 became Professor of Natural Philosophy at the Royal Institute of Great Britain. Lord Rayleigh died on June 30. he was head of the Cavendish Laboratory at Cambridge from 1879-1884. 1919 at Witham. Essex. William John Macquorn Rankine William John Macquorn Rankine (July 2. and naval engineering in which he was involved in applying scientific principles to building ships. a Chancellor of Cambridge University. at Witham. They had three sons. As a successor to James Clerk Maxwell. Essex. His steam engine manuals were used for many decades. was a Justice of the Peace and the recipient of honorary science and law degrees.3.December 24. 1820 . strength of materials. Rankine was born in Edinburgh to British Fig. the eldest of whom was to become a professor of physics at the Imperial College of Science and Technology. sister of the future prime minister. He was a founding contributor to the science of thermodynamics (Rankine Cycle). and his only brother and parents died before him. Rankine developed a theory of the steam engine.1. 1919. and as President from 1905 to 1908. never allowing politics to interfere with science.9: Portrait of Rankine Army lieutenant David Rankine and Barbara Grahame. 1872) was a Scottish engineer and physicist. He was a Fellow of the Royal Society (1873) and served as Secretary from 1885 to 1896. He intervened in debates of the House of Lords only on rare occasions. London. In 1871 he married Evelyn.

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

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

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

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

24 CHAPTER 1. INTRODUCTION .

The flow through a stream tube is assumed to be one-dimensional so that there isn’t any flow except at the tube opening. 25 .3 2.4 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.2 2. The mass after a very short time of dt is dm. Several terms will be reviewed such as stream lines. For simplicity.CHAPTER 2 Fundamentals of Basic Fluid Mechanics 2. This material is not covered in the history chapter.1 Introduction This chapter is a review of the fundamentals that the student is expected to know. At the initial time the mass that was in the tube was m0 . Several application of the fluid mechanics will demonstrated. The basic principles are related to the basic conservation principle. it is assumed the control volume is a fixed boundary. In addition the basic Bernoulli’s equation will be derived for incompressible flow and later for compressible flow. 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. 2.

FUNDAMENTALS OF BASIC FLUID MECHANICS Supposed that the fluid has a property η dNs dt = lim Ns (t0 + ∆t) − Ns (t0 ) ∆t (2.26 CHAPTER 2.1) ∆t→0 .

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

SPEED OF SOUND The people had recognized for several hundred years that sound is sound wave dU velocity=dU a variation of pressure. 3. This velocity is referred to as the speed of sound. Thus. To answer this question consider a piston moving from the left to the right at a relatively small velocity (see Figure 3. Thus.1) Fig. it is considered . 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 (3.28 CHAPTER 3. P is a function of two independent variables. 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.2) into equation (3.3) Substituting the expression for dU from equation (3.4) yields c2 dρ ρ = dP dP =⇒ c2 = ρ dρ (3.2) From the energy equation (Bernoulli’s equation). the pulse will be infinitesimally small.5) An expression is needed to represent the right hand side of equation (3.1). it raises the question: what is the speed of the Fig.” It is assumed that if the velocity of the piston is infinitesimally small. the pressure and density can be assumed to be continuous. Applying the mass balance P+dP yields P ρ+dρ ρ ρc = (ρ + dρ)(c − dU ) (3. Here.1: A very slow moving piston in a still gas small disturbance travel in a “quiet” medium. 3.4) (3. The information that the piston is moving passes thorough a single “pressure pulse.5). 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.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. For an ideal gas.

5) can be derived from the momentum equation.10) 3.9) dP = c2 dρ This yields the same equation as (3.6) In the derivations for the speed of sound it was assumed that the flow is isentropic. The full differential of the pressure can be expressed as follows: dP = ∂P ∂ρ dρ + s ∂P ∂s ds ρ (3. SPEED OF SOUND IN IDEAL AND PERFECT GASES 29 that P = P (ρ.3. and a function “molecular structure” or ratio of specific heats.1: Demonstrate that equation (3.8) (3. s) where s is the entropy. ρ.5).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.3. k namely P = constant × ρk (3.5) can be obtained by utilizing the momentum equation instead of the energy equation. S OLUTION The momentum equation written for the control volume shown in Figure (3. Example 3. (3.7) s Note that the equation (3. The pressure for an ideal gas can be expressed as a simple function of density.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. therefore it can be written ∂P dP = dρ ∂ρ (3.11) .

94199 and substituting into the equation yields c= m 200000 = 780.12) Remember that P/ρ is defined for an ideal gas as RT . and equation (3.61376 ρ = 6.32823 sec kg m3 (3.13216 kg m3 kg m3 kg m3 After interpretation of the temperature: kJ At 18[bar] and 335.5 0.) √ m c = kRT ∼ 1.5 sec Note that a better approximation can be done with a steam table.8226 ∆P ∆ρ kJ K kg kJ K kg kJ K kg (3.15) for ideal gas assumption (data taken from Van Wylen and Sontag.2: Calculate the speed of sound in water vapor at 20[bar] and 350◦ C. 3 This data is taken form Van Wylen and Sontag “Fundamentals of Classical Thermodynamics” 2nd edition .9563 At 18[bar] and 350◦ C: s = 7. S OLUTION The solution can be estimated by using the data from steam table3 c= At 20[bar] and 350◦ C: s = 6.7◦C: s ∼ 6.30 and hence CHAPTER 3. and it will be part of the future program (potto–GDC).46956 ρ = 7.9563 K kg ρ ∼ 6. (a) utilizes the steam table (b) assuming ideal gas.13) c = kRT Example 3. table A 8.327 × 461 × (350 + 273) ∼ 771.14) s=constant ρ = 6.12) can be written as √ (3. SPEED OF SOUND P c= dP constant × ρk = k × constant × ρk−1 = k × dρ ρ =k× P ρ (3.0100 At 18[bar] and 300◦ C: s = 6. Classical Thermodynamics.

x T = (TB − TA ) + TA h Where the distance x is the variable distance.3. the Gibbs function (3.20) . The distance between “A” and “B” is denoted as h. What is the time it take for sound to travel from point “A” to point “B” under this assumption.4 Speed of Sound in Real Gas The ideal gas model can be improved by introducing the compressibility factor. SPEED OF SOUND IN REAL GAS 31 Example 3. Recalling from thermodynamics. It should be noted that velocity is provided as a function of the distance and not the time (another reverse problem).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.19) The speed of sound of any gas is provided by equation (3. Thus.17) This correction factor approaches one when TB −→ TA .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. The compressibility factor represents the deviation from the ideal gas. For an infinitesimal time dt is equal to dt = dx = kRT (x) dx kRTA (TB −TA )x TA h +1 integration of the above equation yields 2hTA t= √ 3 kRTA (TB − TA ) h ¯ kRT 3 2 TB TA 3 2 −1 (3.4. a real gas equation can be expressed in many cases as P = zρRT (3. To obtain the expression for a gas that obeys the law expressed by (3. 3.7).18) (3.? S OLUTION The temperature is denoted at “A” as TA and temperature in “B” is TB .

20) The definition of pressure specific heat for a pure substance is Cp = ∂h ∂T =T P ∂s ∂T (3.22) ρ From thermodynamics.21) P The definition of volumetric specific heat for a pure substance is Cv = ∂u ∂T =T ρ ∂s ∂T (3.3: The Compressibility Chart is used to obtain T ds = dh − dP ρ (3.32 CHAPTER 3. SPEED OF SOUND Fig.23) P Van Wylen p. . perhaps to insert the discussion here. 372 SI version. it can be shown 4 dh = Cp dT + v − T 4 See ∂v ∂T (3. 3.

31) . 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 (3.25) P dP = Cp dT − T z (3.3.24) into equation (3.26) P Utilizing Gibbs equation (3.27) Letting ds = 0 for isentropic process results in dP R dT = z+T T P Cp ∂z ∂T (3.4.28) P Equation (3.28) can be integrated by parts.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 (3.24) Substituting the equation (3.28) and (3.23) results  v  v  z T Simplifying equation (3.30) yields dρ dP Cv = ρ P Cp z+T z+T ∂z ∂T P ∂z ∂T ρ (3.29) results in dρ R z+T ρ Cv ∂z ∂T = ρ dP R z+T P Cp ∂z ∂T (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. However.30) P Rearranging 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.

1[m/sec] For the real gas first coefficient n = 1.407. Now. the compressibility factor. For short hand writing convenience.32) provides a tool to estimate n. the speed of sound for a real gas can be obtained in the same manner as for an ideal gas. But the definition of n in equation (3.4: Calculate the speed of sound of air at 30◦ C and atmospheric pressure ∼ 1[bar]. What is different in these derivations is that a relationship between coefficient n and k was established. equation (3.4 factor (0.3 and n ∼ 1 which means that speed of sound is only 0. and in–fact any thermodynamics book shows this relationship. z = 0. SPEED OF SOUND If the terms in the braces are constant in the range under interest in this study.995times287 × 300 = 346. Assume that R = 287[j/kg/K ]. .403 × 0. at relative moderate pressure but low temperature common in atmosphere.5) to calculated by ideal gas model. dP = nzRT dρ (3.3 1.33) is similar to equation (3. However. n = 1. For example.407 × 287 × 300 ∼ 348.34) Example 3. S OLUTION According to the ideal gas model the speed of sound should be √ √ c = kRT = 1. and z = 0.34 CHAPTER 3.33) Equation (3. The integration of equation (3. This relationship (3.31) yields ρ1 ρ2 n = P1 P2 (3. a change in temperature can have a dramatical change in the speed of sound.11).7[m/sec] The correction factor for air under normal conditions (atmospheric conditions or even increased pressure) is minimal on the speed of sound.403.31) can be integrated.32) Note that n approaches k when z → 1 and when z is constant.995.403 has √ √ c = znRT = 1. n is defined as k n= Cp Cv z+T z+T ∂z ∂T ρ ∂z ∂T P (3. The specific heat for air is k = 1.33) isn’t new. Make the calculation based on the ideal gas model and compare these calculations to real gas model (compressibility factor).

1: Water speed of sound from different sources The effect of impurity and temperature is relatively large.000 meters.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. At a depth of about 4. SPEED OF SOUND IN ALMOST INCOMPRESSIBLE LIQUID 35 3. as can be observed from the equation (3.5) Remark Fresh Water (20 ◦ C) Distilled Water at (25 ◦ C) Water distilled reference Cutnell.2 × 109 N/m2 = 1493m/s 1000kg/m3 elastic property = inertial property B ρ (3.36) dP dρ (3. Chicago: World Book. The amount of compression of almost all liquids is seen to be very small as given in Table (3.8% even under this pressure nevertheless it is a change. 1967-1968: E37 Value [m/sec] 1492 1496 1494 Table 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. the pressure is about 4 × 107 N/m2 .2 × 109 N/m2 . Ohio: Chemical Rubber Co. According . 601 Handbook of Chemistry and Physics.. the average bulk modulus for water is 2. The World Book Encyclopedia. The ratio of the change in the fractional volume to pressure or compression is referred to as the bulk modulus of the material.37). The fractional volume change is only about 1. The compressibility of the substance is the reciprocal of the bulk modulus.3. Johnson. Many researchers have looked at this velocity.5). Physics. New York: Wiley.5. and for purposes of comparison it is given in Table (3. & Kenneth W. For example.35) This agrees well with the measured speed of sound in water. For example. 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]. 1482 m/s at 20◦ C. 1997: 468. John D. 1999.

Soc. Manual of Sound Recording. John. and cST P is a correction factor between coupling of the different parameters. Young’s Modulus for a representative value for the bulk modulus for steel is 160 109 N /m2 . Acoust. the speed of sound in liquids is about 3 to 5 relative to the speed of sound in gases. Nevertheless. the speed of sound in solids is larger than in liquids and definitely larger than in gases. after Aldred. material Glycerol Sea water Mercury Kerosene Methyl alcohol Carbon tetrachloride reference 25 ◦ C Value [m/sec] 1904 1533 1450 1324 1143 926 Table 3.36 CHAPTER 3. Wilson’s formula is accepted by the National Oceanographic Data Center (NODC) USA for computer processing of hydrological information. 5 J. gives a sound speed for structural steel of E = ρ 160 × 109 N/m2 = 4512m/s 7860Kg/m3 c= (3. . N 10. cP is a function pressure.37) where c0 = 1449. London: Fountain Press.38) Compared to one tabulated value the example values for stainless steel lays between the speed for longitudinal and transverse waves. Speed of sound in solid of steel. 1960.32.2: Liquids speed of sound. Wilson’s empirical formula appears as follows: c(S.. and cS is a function salinity. p. T. and hydrostatic pressure. with different speeds in different directions. (3. salinity. 1972 In summary. Amer. SPEED OF SOUND to Wilson5 . P ) = c0 + cT + cS + cP + cST P . cT is a function temperature. using a general tabulated value for the bulk modulus. the speed of sound in sea water depends on temperature. 1357. vol.14 is about clean/pure water. in different kinds of geometries. and differences between transverse and longitudinal waves. 3.6 Speed of Sound in Solids The situation with solids is considerably more complicated.

For simplicity.7 Sound Speed in Two Phase Medium The gas flow in many industrial situations contains other particles.39) (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. John. and liquid with some bubbles. after Aldred.7. In that case.40) (3. three extreme cases suggest themselves: the flow is mostly gas with drops of the other phase (liquid or solid). 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. London:Fountain Press. However. for homogeneous and under certain condition a single velocity can be considered. there could be more than one speed of sound for two phase flow. it assumed that two materials are homogeneously mixed.3: Solids speed of sound. Topic for none homogeneous mixing are beyond the scope of this book. There can be several models that approached this problem. It further assumed that no heat and mass transfer occurs between the particles.40) can be approximated as ρ =1+m ρa (3.41) . 1972 3. Manual of Sound Recording. Indeed there is double chocking phenomenon in two phase flow. For small value of ξ equation (3.3. In actuality. about equal parts of gas and the liquid phase. The first case is analyzed.

For example. 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. It must noticed that m = constant.44) Therefore. this analysis results in lower speed of sound compared to pure gas. The 1+m 1+m correction factors for the specific heat is not linear.47) (3.45) In a way the definition of γ was so chosen that effective specific pressure heat C +mC +mC and effective specific volumetric heat are p and Cv respectively. The gas density can be replaced by equation (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. It can be noticed that Rmix and γ are smaller than similar variables in a pure gas. Generally.38 m ˙b m ˙a CHAPTER 3. .6). the velocity of sound in slightly wed steam can be about one third of the pure steam speed of sound. the velocity of mixtures with large gas component is smaller of the pure gas. Hence.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.41) R P = T ρ 1+m (3. the mixture isentropic relationship can be expressed as P where γ−1 R = γ Cp + mC Recalling that R = Cp − Cv reduces equation (3.46) T = constant (3. 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 (3.39) and substituted into equation (3.48) At this stage the other models for two phase are left for next version (0.46) into γ= Cp + mC Cv + mC (3. SPEED OF SOUND where m = is mass flow rate per gas flow rate.

3. 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 ) (3.7.49) Meta End .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

26) The thermodynamic relationship between the properties can be expressed as T ds = dh − dP ρ (4. Conversely.2. in mathematical terms. In the same argument. the stagnation density is constant through the flow.25) with (4. ρAU = m ˙ = constant. Therefore. the stagnation pressure is also constant through the flow because the flow isentropic. and dividing by the continuity equation reads dρ dA dU + + =0 ρ A U (4.27) For isentropic process ds ≡ 0 and combining equations (4. The only properties that need to be connected are the cross section area and the Mach number. Second is isentropic and isothermal Fig.27) yields dP + U dU = 0 ρ (4. P = ρRT . Examination of the relation between properties can then be carried out. ISENTROPIC FLOW The important sub case in this chapter is the flow in a converging–diverging nozzle.11) are the same.9) and equation (4.4: Control volume inside a convergingmodel.25) hen there is no external work and heat transfer.4). T0 . is constant through the adiabatic flow because there isn’t heat transfer. There are two models that assume variable area flow: First is isentropic and adiabatic model.29) . it is constant for the other. reads dh + U dU = 0 Differentiation of continuity equation. knowing the Mach number or the temperature will provide all that is needed to find the other properties. Clearly. Thus.28) Differentiation of the equation state (perfect gas).46 CHAPTER 4. and dividing the results by the equation of state (ρRT ) yields dP dρ dT = + P ρ T (4. The control volume is shown in Figure (4. the energy equation. the stagnation temdiverging nozzle. equation (4. T ρ P U T+dT ρ+dρ P+dP U+dU 4. 4. perature.1 The Properties in the Adiabatic Nozzle (4. If the right hand side is constant for one variable.

34) The pressure Mach number relationship Before going further in the mathematical derivation it is worth looking at the physical meaning of equation (4.31) = U2 dA A (4.31) to obtain dP ρ Or in a dimensionless form dP dA 1 − M2 = U2 ρ A (4. It can be observed that the critical Mach number is one. The meaning of the sign change for the pressure differential is that the pressure can increase or decrease. it can be observed that dP can be positive or negative depending on the dA and Mach number. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 47 Obtaining an expression for dU/U from the mass balance equation (4.4. can be replaced by the static pressure.30) Rearranging equation (4.28) reads dU U dP dA dρ − U2 + =0 ρ A ρ (4. Now. 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 (4.30) so that the density.2.33) to obtain a variables separation form of dP = dA ρU 2 A 1 − M2 (4.26) and using it in equation (4.32) Equation (4. If the Mach number is larger than one than dP has opposite sign of dA. dρ. If Mach number is smaller than . The term ρU 2 /A is always positive (because all the three terms can be only positive). It is convenient to rearrange equation (4.34).33) 1− U c 2    dρ dP   =U  +  A dP ρ   2  dA  (4.33) is a differential equation for the pressure as a function of the cross section area.

one can observe that the density.36). ρ. dU = − dP PU (4. From the speed of sound. 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.48 CHAPTER 4.It must also be noted that when M = 1 occurs only when dA = 0. For the special case of M = 1 (sonic flow) the value of the term 1 − M 2 = 0 thus mathematically dP → ∞ or 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. increases with pressure and vice versa (see equation 4. the equations are applicable for any gas (perfect or imperfect gas). For the supersonic branch M > 1.35 .36). Thus. the equation of state was not used. An increase in area results in an increase of the static pressure (converting the dynamic pressure to a static pressure).35) From equation (4. Hence the pressure increases when the velocity decreases and vice versa. dρ = 1 dP c2 (4.4. However. Note that the pressure decrease is larger in compressible flow compared to incompressible flow.35) it is obvious that dU has an opposite sign to dP (since the term P U is positive). if the area decreases (as a function of x) the pressure decreases. dA > 0 ⇒ dP < 0 dA < 0 ⇒ dP > 0 This behavior is opposite to incompressible flow behavior. In that case. Since physically dP can increase only in a finite amount it must that dA = 0. The second law (isentropic relationship) dictates that ds = 0 and from thermodynamics dP dT −R ds = 0 = Cp T P . Conversely. not necessarily means that when dA = 0 that M = 1. The relationship between the velocity and the pressure can be observed from equation (4. ISENTROPIC FLOW one dP and dA have the same sign.36) It can be noted that in the derivations of the above equations (4. the opposite. the phenomenon is different.28) by solving it for dU .

37).40) with equation (4.41) into 2 P P (4.39) (4.33) isn’t the solution but demonstration of certain properties on the pressure.42) changes equation (4. state (4.2. the temperature varies according to the same way that pressure does. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 49 and for perfect gas dT k − 1 dP = (4. why? .26)1 .33) yields 1 k ρU 2 dA A 1−M 2 (4. the relationship between the temperature and the stagnation temperature becomes dT0 = 0 = dT 1+ k−1 2 M 2 + T (k − 1)M dM (4. 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. equation (4. mass (4. second law (4.29).40) P =− M dM 1 2 1 + k− 2 M (4.38) yields dT (k − 1)M dM =− 1 2 T 1 + k− 2 M Relationship Between the Mach Number and Cross Section Area The equations used in the solution are energy (4.37) and the relationship of pressure and cross section area (4.41) The following identify. 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. Differentiation of equation (4.33).38) and simplifying equation (4. The relationship between temperature and the cross section area can be obtained by utilizing the relationship between the pressure and temperature (4.37) and becomes (k − 1)M dM (k − 1) dP =− 1 2 k P 1 + k− 2 M Combining equation (4.39). First stage equation (4.9).4.39) is combined with equation (4.43) momentum equation is not used normally in isentropic process.42) M2 − 1 dA = dM 1 2 A M 1 + k− 2 M 1 The (4. Note.37) T k P Thus.

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

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

1416 The velocity at that point is c U =M √ The tube area can be obtained from the mass conservation as A= m ˙ = 8. Finally.0115 0. T = 0. 4 Well.26 × 10−5 [m3 ] ρU kRT = 0.2) or from Figure (4. Calculate the Mach number at point B under the isentropic flow assumption.60693 0. pressure is about two atmospheres with temperature of 250[K ] this question is for academic purposes.4 × 287 × 294 = 304[m/sec] (4. Also.4 and assume a perfect gas model.60000 0. Downstream at point B the pressure was measured to be 1. and velocity can be calculated. The best approximation is by using inserted cone for supersonic flow and measure the oblique shock. In the point where the static pressure known ¯ = P = 3[M P a] = 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 51 the temperature.88639 0.3) or utilizing the enclosed program from PottoGDC.2.86420338 × (273 + 21) = 254. 4 This . there is no known way for the author to directly measure the Mach number.53105 With these values the static temperature and the density can be calculated. Example 4.69428839 × J ρ0 RT0 × 294[K ] 287.44) For a circular tube the diameter is about 1[cm]. Assume that the specific heat ratio k = 1.5[Bar]. estimate the temperature at point B.076K ρ0 ρ= ρ P0 5 × 106 [P a] = 0. the cross section can be calculated with all these information.88638317 × √ 1. or simply using the equations shows that M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.69428 1. Here it is subsonic and this technique is not suitable.86420 0.6 P P0 5[M P a] From Table (4.0 kgK kg m3 = 41.2: The Mach number at point A on tube is measured to be M = 23 and the static pressure is 2[Bar]4 .4.

8137788 and TB = B 0.55556 0.12780453 × = 0. At point “A” the cross section area is 50 [cm2 ] and the Mach number was measured to be 0. The two possible solutions: the first supersonic M = 1.17040879.3: Gas flows through a converging–diverging duct.52 CHAPTER 4.53884934. Find the Mach number at point B.0 = 0.59309 With this information the pressure at point B can be expressed as from the table 4.4.5 The corresponding Mach number for this pressure ratio is 1. Both solution are possible and acceptable.60315132 0.0000 T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.60315132 P P0 = 0. Therefore.2 1. At point B in the duct the cross section area is 40 [cm2 ].81..272112 B With the value of A A∗ from the Table (4.12780 0.4. .42[K ] Example 4. At M = 2 (supersonic flow) the ratios are M 2. S OLUTION To obtain the Mach number at point B by finding the ratio of the area to the critical area. the stagnation pressure at point A is known and stagnation temperature can be calculated. 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.6875 0.2 @ M = 2 PA = P0 PB P0 × PA 2. This relationship can be obtained by AB AB AA 40 = × ∗ = × A∗ AA A 50 from the Table 4.55555556 271. TB = T A × T0 × TA TB T0 = 250[K ] × 1 × 0.17040604 PB 1.6265306 and second subsonic M = 0. Assume that the flow is isentropic and the gas specific heat ratio is 1.59014 = 1.21567 0. 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. The stagnation temperature can be “bypassed” to calculate the temperature at point B M =2 M =1.23005 1.

2.45) to be expressed in terms of the stagnation properties as f (M.e. Hence equation (4.2. 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 (4.46) Expressing the temperature in terms of Mach number in equation (4.49) Equation (4. The area ratio is defined as the ratio of the cross section at any point to the throat area (the narrow area).6266 0.47) It can be noted that equation (4.34585 1.47) holds everywhere in the converging-diverging duct and this statement also true for the throat.45) This parameter is studied here.53887 0. specifically and explicitly the relationship for the chocked flow.47) yields 1 A = ∗ A M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (4.49) relates the Mach number at any point to the cross section area ratio. pressure) do not change.28772 0. .k) m ˙ P P0 U √ = A P0 kRT k R T0 1 P √ = √0 M T T0 T0 k P R P0 T0 T (4.22617 0.82071 1. The area ratio as a function of the Mach number needed to be established.2721 0.47) obtained the form √ k+1 − 2( k−1) m ˙ kP0 k−1 √ (4. to examine the maximum flow rate and to see what is the effect of the compressibility on the flow rate.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 (4.46) results in m ˙ = A kM P0 √ kRT0 1+ k−1 2 M 2 k+1 − 2( k−1) (4. dividing equations (4.48) = 1 + ∗ A 2 RT0 Since the mass flow rate is constant in the duct.2721 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 53 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 1.65396 0. temperature. The throat area can be denoted as by A∗ .0440 4.86838 1.4. It is convenient to rearrange the equation (4.94511 0.

Carrying this calculation results at M = 1.68473 P √0 ∼ √ T0 R (4.4 m ˙ A∗ 0.50) max For specific heat ratio.54) Utilizing Fliengner’s number definition and substituting it into equation (4.51) max The maximum flow rate for air (R = 287j/kgK ) becomes.48) in with respect to M and equating to zero. k = 1.52) is known as Fliegner’s Formula on the name of one of the first engineers who observed experimentally the choking phenomenon. (see for example. √ m ˙ T0 = 0. It can be noticed that Fliengner’s equation can lead to definition of the Fliengner’s Number. which have mostly academic proposes.54 CHAPTER 4. m ˙ A∗ P √0 = T0 k R k+1 2 k+1 − 2( k−1) (4.52) Equation (4.040418 A ∗ P0 (4. c0 √ m ˙ kRT0 mc ˙ 0 m ˙ T0 1 √ =√ =√ ∗ ∗ ∗ A P0 kRA P0 RA P0 k The definition of Fliengner’s number (Fn) is Fn ≡ √ mc ˙ 0 RA∗ P0 Fn (4.48) results in F n = kM k−1 2 M 1+ 2 k+1 − 2( k−1) (4.56) “Naughty Professor” Problems in Isentropic Flow To explain the material better some instructors invented problems. Shapiro (problem 4.53) (4. ISENTROPIC FLOW The maximum flow rate can be expressed either by taking the derivative of equation (4. While these .5)).55) and the maximum point for F n at M = 1 is Fn = k k+1 2 k+1 − 2( k−1) (4.

57) result in p R m ˙ Aρc 2 (4. ρ when all other variables are known. For this kind of problems a new Table was constructed and present here5 . In other words. .4. the terms 0 for the group over the under–brace approaches zero when the flow rate (Mach number) is very small.58) And further Rearranging equation (4.58) transformed it into ρ2 = k−1 Pρ + T0 R 2kRT0 m ˙ A 2 (4. The case of T0 and P This case considered to be simplest case and will first presented here. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 55 problems have a limit applicability in reality. they have substantial academic value and therefore presented here.9) and substituting for Mach number M = m/Aρc ˙ results in k−1 T0 =1+ T 2 Rearranging equation (4.2.57) 1/kR T0 ρ 2 = T ρ ρ + T c2 k−1 2 m ˙ A 2 (4. 5 Since    (4. It is convenient to change it into ρ2 − k−1 Pρ − T0 R 2kRT0 m ˙ A 2 =0 (4. e.59) is quadratic equation for density. The use of the regular isentropic Table is not possible because there isn’t variable represent this kind problems. Using energy equation (4. The situation where the mass flow rate per area given with one of the stagnation properties and one of the static properties.61) version 0.59) Equation (4. P0 and T or T0 and P present difficulty for the calculations.g.44 of this book.60) The only physical solution is when the density is positive and thus the only solution is    P 1 ρ=  + 2 RT  0 P RT0 2 k−1 +2 kRT0 m ˙ A →(M →0)→0 2  For almost incompressible flow the density is reduced and the familiar form of perfect gas model is seen since stagnation temperature is approaching the static P temperature for very small Mach number (ρ = RT ).

1). m ˙ A 2 = 1 c0 2 k P0 2 m ˙ A∗ 2 A∗ A 2 P0 P 2 (4. F n2 A ∗ P0 AP With this new definition equation (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.64) RT0 P2 m ˙ A 2 = F n2 k A ∗ P0 AP 2 (4. and static temperature. k ). P0 . T . k also). it is shown that the dimensionless group is a function of Mach number only (well. the dimensionless density is  1 (k − 1)F n2 ρ ˆ = 1 + 1 + 2 2 k2 A∗ P AP 0 2   (4.67) (4. And the values of A ∗P 0 in Table (4.61) is transformed into   2 ˙ 1 (k − 1)RT0 m  1+ 1+2 ρ ˆ= 2 kP 2 A (4.68) . also the specific heat. again the specific heat ratio. the problems is reduced to finding tabulated values.65) Again notice that the right hand side of equation (4.66) is only function of Mach AP were tabulated number (well.2) and Fn is tabulated in the next Table (4.62) The dimensionless density now is related to a dimensionless group that is a function of Fn number and Mach number only! Thus.56 CHAPTER 4. this dimensionless group is function of Mach number only. First. The case of P0 and T A similar problem can be described for the case of stagnation pressure.66) m ˙ A 2 = F n2 k A ∗ P0 AP 2 T T0 P0 P 2 (4. RT P0 2 It can be noticed that F n2 = k T T0 P0 P 2 Hence. Thus.

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. Unfortunately. the dimensionless approach is used later analytical method is discussed (under construction).2). if it was shown that a group of parameters depends only Mach number than the Mach is determined by this group. First. not all the analysis valid for the same conditions and is as the regular “isentropic” Table.). 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. (4.67) became RT P0 2 m ˙ A 2 = A ∗ P0 AP 2 (4.69) The right hand side is tabulated in the “regular” isentropic Table such (4. 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). For example the first “naughty professor” question is sufficient that process is adiabatic only (T0 . This example shows how a dimensional analysis is used to solve a problems without actually solving any equations. mass flow rate per area.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. The heat/temperature part is valid for enough adiabatic condition while the pressure condition requires also isentropic process.2.4. . Based on the same arguments.71) It was hidden in the derivations/explanations of the above analysis didn’t explicitly state under what conditions these analysis is correct. The actual solution of the equation is left as exercise (this example under construction). ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 57 Thus equation (4.

007 0.71967 0.020986 0.47000 0.24000 0.41073 1.094654 0.10397 0.031 0.081847 0.41997 0.62915 0.003 0.001 0.019 0.53000 0.36000 0.35316 0.22000 0.27926 0.52485 0.24674 0.15372 0.16121 0.35000 0.44000 0.31000 0.49285 1.36550 0.52690 1.30000 0.18709 0.065 0.33378 0.026 0.33226 0.20316 0.89613 1.45951 1.13284 0.59736 1.044110 0.38000 0.001 0.51932 0.41855 0.079722 0.52000 0.43000 0.10000 0.57253 0.67129 1.16522 0.63386 1.18896 0.49249 0.85107 0.00747 0.95791 0.22085 0.005 0.40333 0.56172 1.33233 1.17381 0.46000 0.29247 0.65857 0.13232 0.28000 0.000 0.00E +00 1.12039 0.17397 0.81034 1.92149 0.26264 0.060 0.76924 1.28677 1.050518 0.042 0.001 0.087372 0.00707 0.29000 0.074314 0.005 0.70969 1.10639 0.050001 0.68875 0.57944 1.050 0.37000 0.74912 1.0 2.87421 1.002 0.077 0.11294 0.029920 0.51000 0.72927 1.010476 0.35361 0.18992 0.059212 0.37474 0.18428 0.015 0.54000 0.038365 0.62436 0.41000 0.000424 0.11928 0.27608 0.0 0.78965 1.44215 0.014197 0.43919 0.002 0.46798 0.49305 0.000 0.017813 0.39478 1.31703 1.024585 0.55000 0.83132 1.13796 0.014268 0.012 0.12239 0.26000 0.38884 0.39000 0.028 0.36764 0.58 CHAPTER 4.55637 0.008 0.27000 0.33465 0.25000 0.49000 0.067111 0.25018 0.52858 0.41338 0.48360 0.85261 1.28307 0.20458 0.34330 0.23137 0.14889 0.39596 0.36329 1.021 0.20000 0.30214 0.34000 0.23155 0.44192 0.003 0.30185 1.57656 0.54422 1.37896 1.75136 0.21703 0.20109 0.14084 1.69036 1. ISENTROPIC FLOW Table 4.69948 0.60047 0.50000 0.072487 0.26495 0.011 0.21584 0.47609 1.0 0.33000 0.070106 1.46677 0.30418 0.20442 0.31203 0.66098 0.24773 0.34775 1.11710 0.91838 1.013 0.095449 0.54733 0.035 0.65246 1.32220 0.098460 0.22634 0.88588 0.028651 0.14276 0.0 0.015027 0.29663 0.089910 0.17728 0.57709 0.12724 0.023 0.033229 0.400E −06 1.13342 0.78382 0.73995 0.31480 0.62E −05 0.54531 0.25535 0.60706 0.012593 0.16581 0.074254 0.23000 0.046 0.42000 0.58952 0.038 0.057647 0.81706 0.15963 0.00351 0.21000 0.006 0.23777 0.065654 0.10752 0.083989 0.51882 0.63889 .000 0.00352 0.42683 1.40000 0.004 0.017 0.49485 0.44309 1.060404 0.009 0.14592 0.14927 0.61550 1.071 0.44363 0.39701 0.46633 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 0.32000 0.27358 0.50978 1.45000 0.48000 0.003 0.055 0.00865 0.37432 0.065557 0.

57000 0.385 1.269 1.177 2.771 1.156 2.225 1.105 0.62936 0.78062 0.161 1.94096 0.98717 1.082 2.383 1.574 1.991 2.86101 0.613 6.135 1.642 1.819 1.865 .853 1.667 1.949 9.088 1.646 3.78250 0.185 1.305 1.88000 0.500 1.141 1.508 1.018 2.602 2.198 1.56000 0.323 1.381 8.610 1.842 2.381 2.761 1.77000 0.806 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ0 2 T m ˙ 2 A 0.121 3.477 1.882 2.602 2.513 5.214 1.91000 0.216 2.87000 0.278 1.461 1.277 2.74624 0.68000 0.68565 0.92366 0.937 1.526 1.557 1.105 1.62000 0.006 6.266 4.840 4.813 2.20 0.58000 0.814 1.287 3.66000 0.70675 0.340 2.810 2.846 8.081 2.288 1.535 2.404 2.670 2.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.152 1.271 1.414 1.89000 0.699 3.74000 0.88142 0.94848 0.362 1.122 1.109 1.805 0.61000 0.79000 0.931 2.65000 0.220 3.011 1.784 1.104 3.977 2.422 1.541 3.446 1.740 2.187 1.95665 1.148 1.553 1.113 1.671 2.241 2.73000 0.84000 0.581 1.82722 0.607 1.924 1.096 2.712 1.971 3.571 4.94000 Fn 0.059 1.211 1.71000 0.083 1.255 1.131 1.217 1.703 1.097 1.448 1.99507 1.101 2.63000 0.2.389 3.60000 0.072 1.4.96389 0.173 1.83000 0.821 1.660 1.897 5.747 1.084 1.038 2.324 1.979 4.368 1.320 1.657 1.423 1.297 1.998 3.236 1.554 10.090 1.963 3.859 1.382 1.342 7.260 1.450 1.955 3.92000 0.70000 0.124 1.97562 1.427 1.469 2.037 1.587 1.241 1.035 1.790 2.094 1.326 1.661 1.235 5.791 1.044 1.99514 1.489 2.85000 0.758 1.194 1.903 1.69000 0.86000 0.412 2.980 2.538 1.671 1.721 4.78000 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 59 Table 4.144 ρ ˆ 1.258 4.030 1.343 1.595 2.240 1.349 1.244 3.484 4.90382 0.64000 0.279 1.355 1.708 1.953 2.76000 0.640 1.81996 0.317 1.738 1.509 1.898 1.93000 0.168 2.147 1.058 2.81139 0.869 7.82000 0.869 1.536 2.75000 0.607 1.059 2.474 1.72000 0.112 1.462 3.244 5.516 1.582 1.74290 0.67210 0.709 3.151 1.59000 0.500 1.87424 0.415 1.90000 0.043 4.81000 0.033 1.541 1.720 2.405 1.162 1.029 3.212 1.424 6.736 1.457 3.676 1.972 5.80000 0.284 2.407 2.67000 0.202 1.278 2.

48) can be equated as kP0 A∗ RT0 k−1 2 1+ M 2 k−1 − 2( k−1) = constant (4.000 Fn 2.217 4.804 5.436 6. 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.3.338 3.112 6. Clearly.913 4. 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.97000 0.324 2. The temperature at Chamber supplying the pressure to tube is 27 C .60 12.5[Bar]. Flow with pressure losses The expression for the mass flow rate (4. The second model which there is significant heat transfer but insignificant pressure loss (Rayleigh flow like). Denoting subscript A for one point and subscript B for another point mass equation (4.777 7. A∗ ) = constant. the velocity.98 6.98000 0.500 3.233 2. The third academic condition is of static temperature and the static pressure.003 2.600 4.06 14.37 13. T0 . Calculate for that point the Mach number. S OLUTION The second academic condition is when the static temperature is given with the stagnation pressure. A constant flow rate requires that m ˙A = m ˙ B . .72) From equation (4.181 3.88 11. Assume that the process is isentropic and k=1. in a case where the flow isn’t isentropic or adiabatic the total pressure and the total temperature will change (due to friction.583 10.016 Example 4.920 1.259 3.4: A gas flows in the tube with mass flow rate of 1 [kg/sec] and tube cross section is ◦ 0. and the stagnation pressure. it is clear that the function f (P0 .95000 0.60 CHAPTER 4.843 1.037 4.419 ρ ˆ 1.371 2.278 2.47) is appropriate regardless the flow is isentropic or adiabatic. and heat transfer).001[m2 ].961 2.136 7. At some point the static pressure was measured to be 1.72).188 2.515 7.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0. There are two possible models that can be used to simplify the calculations.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. ISENTROPIC FLOW Table 4.99000 1.96000 0.881 1.19 14.419 3. The first model for neglected heat transfer (adiabatic) flow and in which the total temperature remained constant (Fanno flow like).404 4.

point B.68966 0. Assume no mass lost and adiabatic steady state flow. With these information.27240 0. The total pressure can be calculated because the Mach number and static pressure are known.01[m2 ]. calculated the total pressure lost.74) leads to A A ∗ MA A A ∗ MB P 0 |A = P 0 |B A| A A| B (4.015[m2 ] and Mach number is 1. Mach number is 2.76) Example 4. And utilizing the equality of A∗ = P 0 |A A∗ | = ∗A P 0 |B A |B A∗ A A (4.75) the stagnation pressure at point B can be obtained.39498 1.05853 0. and the duct section area is 0.13169 2. the cross section area is 0.6367 0.5. comparison of mass flow rate at point A and point B leads to P 0 A ∗ |A = P 0 A ∗ |B .73) For adiabatic flow.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 (4.1762 0.44444 0.15432 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 61 If the mass flow rate is constant at any point on the tube (no mass loss occur) then k RT0 2 k+1 k+1 k−1 m ˙ =A ∗ P0 (4.5: At point A of the tube the pressure is 3[Bar].4. and utilizing equation (4. thus the area ratios can be calculated. S OLUTION Both Mach numbers are known.2.5000 T T0 ρ ρ0 A A P P0 0. M 1.55401 0.5.5000 2.32039 0. Downstream at exit of tube.62693 A×P A∗ ×P0 F F∗ .

00 1.36091 0.57 5.32039 A×P A∗ ×P0 F F∗ 2.838 2.52989 0.009 1.964 2.75) provides P0 |B = 51.100 1.99206 0.88652 0.27240307 × 15.99950 0.243 = 4.268 0.83333 0.200 0.72093 0.97250 0.013[Bar] Note that the large total pressure loss is much larger than the static pressure loss (Pressure point B the pressure is 0.65602 0.99800 0.68704 0.77640 0.8E +5 11.55401 .95238 0.590 1.52828 0.59 5.900 1.00000 0.31424 0.74738 0.42493 0.80515 0.115 1.050 0.822 2.188 1.52828 0.99825 0.300 0.030 1.93947 0.99502 0.100 0.058527663 by utilizing equation (4.01 1.4E +5 4.89699 0.52828 0.84045 0.99303 0.25781291 × Hence P0 |A − P0 |B = 51.094 1.35036 0. 0.400 1.000 1.500 T T0 ρ ρ0 A A P P0 1.96899 0.038 1.95638 0.88517 0.53039 0.43742 0.63394 0.93284 0.146[Bar]).882 1.58377 0.000 0.2) as P 0 |A = P P P0 M =2.58170 0.46835 0.2: Isentropic Table k = 1.41238 0.78896 0. ISENTROPIC FLOW First.71839 0.8E + 5 11.62 CHAPTER 4.1761671 × ≈ 15.008 1.72632 0.243[Bar] 2.98028 0.015 4.68966 1.68110 0.424 1.79158 0.200 1.130 0.59126 0.63535 0.6367187 0.3 Isentropic Tables Table 4.781 2.47207 0.600 0.89561 0.443 1.99875 0.73999 0.27240 5.243 = 36.00000 0.98232 0.53399 0.0000 0.700 0.035 1.400 0.78400 0.25781291[Bar] 0.39498 5.38484 0.53807 0.176 1.92427 0.53114 0.4 M 0.54655 0.91075 0.5 A = 3 = 51.59650 0.55425 0.800 0.340 1.257 − 15.48290 0.912 1.066 1.500 0. the stagnation at point A is obtained from Table (4.84302 0.300 1.53974 0.86059 0.93155 0.

68830 0.00190 0.092593 0.15432 0.439 1.73510 0.047251 0.72953 0.67320 0.000815 0.500 8.052493 0.77) 6 The one dimensional momentum equation for steady state is U dU/dx = −dP/dx +0(other effects ) which are neglected here.019473 0.00 T T0 ρ ρ0 A A P P0 0.790 10.16667 0.4.2 Isentropic Isothermal Flow Nozzle General Relationship In this section.28682 0.2E +2 5.00189 0.69983 0.000 6.000631 0.4E +2 0.35714 0.500 3.000633 0. knowing the two limits provides a tool to examine where the reality should be expected.12780 0.027224 0.012628 A×P A∗ ×P0 F F∗ 0.022046 0.500 5.77) into the momentum equation6 yields U dU + RT dP =0 P (4.0E +2 1.064725 0.500 9.338 1.3.00364 0.000385 0.033682 0.4E +2 1.000 4.23810 0.000 5.045233 0.66138 0.58072 0.000 3.14924 0.058140 0.00659 0.013111 0. Again in reality the heat transfer is somewhere in between the two extremes.9E +2 2. the other extreme case model where the heat transfer to the gas is perfect.800 1.27099 0.00 36.57768 0.35573 0.500 6.500 10. So.015504 0.000242 0.72586 0.00346 0. dP = dρRT Substituting equation (4.3.60680 0.74058 0.74192 4. In isothermal process the perfect gas model reads P = ρRT .36E −5 0.23005 0.56 25.039628 0.3E +2 4.00107 0.71578 0.00519 0.20259 0.058528 0.44444 0.00758 0.56182 0.g.73723 0.1 4.87 53.72136 0.072464 0.23211 0.12195 0.000102 6.25699 0.000155 0.73903 0.21567 0.78) (4.19802 0.31969 0.3.028962 0. ISENTROPIC TABLES Table 4.017321 0. Eckert number is very small) is presented.58549 0.10582 0.089018 0.13 1.000 2.025156 0.31E −5 2.250 1.637 4.4 (continue) 63 M 1.076226 0.29414 0.23527 0.17404 0.14184 0.00261 0.688 2. (e.55556 0.56976 0.011340 0.00107 0.13169 0.500 4.000 8.5E +2 3.047619 0.65326 0.600 1.700 1.057227 0.017449 0.74E −5 3.2: Isentropic Table k=1.63371 0.235 6.72 16.500 7.90E −5 4.62693 0.070595 0.000 7.013957 0.081633 0.900 2.25044 0. The perfect gas model is again assumed (later more complex models can be assumed and constructed in a future versions).000 9.28986 0.00141 0.73257 0.000495 1.11528 0.027662 0.555 1.59309 0. .18 75.70876 0.

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. U2 2 − U 1 2 P2 + RT ln =0 2 P1 (4. T 2 P1 (4.82) Transform from equation (4.64 CHAPTER 4.86) Utilizing conservation of the mass AρM = constant to yield M 2 P2 A1 = A2 M 1 P1 (4.87) .84) Or in terms of the pressure ratio equation (4. U1 ≈ 0 reads U2 = 2RT ln P2 P1 (4. the velocity at point 2 becomes U2 = 2RT ln P2 − U1 2 P1 (4.79) Thus.80) The velocity at point 2 for stagnation point.79) to a dimensionless form becomes constant constant  (M2 2 − M1 2 ) kRT P2   ln = R  .78) yields the Bernoulli’s equation for ideal gas in isothermal process which reads . P2 k (M2 2 − M1 2 ) = ln 2 P1 (4.84) reads k(M1 2 −M2 2 ) P2 2 = =e P1 e M1 2 e M2 2 k 2 (4.83) Simplifying equation (4.83) yields .81) Or in explicit terms of the stagnation properties the velocity is U= 2RT ln P P0 (4. ISENTROPIC FLOW Integration of equation (4.

93) Of course in isothermal process T = T ∗ . From the Figure 4.87) and equation (4. The critical pressure ratio can be obtained from (4.88) The change in the stagnation pressure can be expressed as P0 2 P2 = P0 1 P1 1+ 1+ 2 k−1 2 M2 2 k−1 2 M1 e M1 = 2 e M1 2 k 2 (4. at this stage.6).89) The critical point. The true critical point is when flow is choked and the relationship between two will be presented. Here the critical point defined as the point where M = 1 so results can be compared to the adiabatic case and denoted by star. the critical stagnation pressure reads 2 k−1 (1−M )k 2 1 + P0 2 M1 2 = e P0 ∗ k+1 k k−1 (4.92) Finally.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. 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. Again it has to emphasis that this critical point is not really related to physical critical point but it is arbitrary definition.94) . ISENTROPIC TABLES Combing equation (4. All these equations are plotted in Figure (4.85) yields M1 A2 = A1 M2 65 e M2 2 e M1 2 k k−1 k 2 (4. The minimum of the curve is when area is minimum and at the point where the flow is choked.3 it can be observed that minimum of the curve A/A∗ isn’t on M = 1.85) to read (1−M 2 )k ρ P = ∗ =e 2 ∗ P ρ (4.91) (4.3.4. 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 (4.90) Equation (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.

Utilizing equation . M = √ k (4. the pressure loss in adiabatic flow is milder as can be seen in Figure (4. The velocity √ k .7(a)).5 1 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. 4.94) simplified to 1 kM 2 − 1 = 0 .96) On the other hand.5 3 2. 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.5 0 0 0. ISENTROPIC FLOW Isothermal Nozzle 4 3.5 2 M 2.5 3 3.6: Various ratios as a function of Mach number for isothermal Nozzle Equation (4. dividing the at the throat of isothermal model is smaller by a factor of √ critical adiabatic velocity by k results in Uthroatmax = √ RT (4.95) It can be noticed that a similar results are obtained for adiabatic flow.66 CHAPTER 4.5 1 1. Thus. It should be emphisized that the stagnation pressure decrees.

5 1 1.5 2 M 2.5 1 0.97) e− 2 = 0.5 3 2.5 0 0 0.5 2 1.5 0 0 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. as opposed to the adiabatic case.60653 1 Notice that the critical pressure is independent of the specific heat ratio.7: The comparison of nozzle flow (4.3.5 3 3.4.5 4 A / A iso * A / A adiabatic * P / P iso * P / P adiabatic * 67 Comparison between the two models 5 4. k .5 3 2. Again.5 4 3. ISENTROPIC TABLES Isothermal Nozzle 4 3. 4. the heat transfer is expressed as (4.5 1 0.90) 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. It also has to be emphasized that the stagnation values of the isothermal model are not constant.5 1.5 2 1.

100) By utilizing equation (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).6 between the two models so in an actual situation it can be 0. 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. Thus.2 the distance from the entrance to the throat is normalized (to 0 0 1 2 0.8: 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 20.100) the velocity ratio was obtained and is plotted in Figure (4. 4. The Mach number for the isentropic is larger for the supersonic branch but the velocity is lower. As can be observed from Figure (4. To make sense and eliminate unnecessary details 0. In this Fig. In this comparison it has to be remembered that critical area for the two models are different by about 3% (for k = 1. The prediction of the Mach number are similarly shown in Figure (4. .4).7(b)).5 one (1)).7(b)).99) UT MT kRTs It can be noticed that temperature in the isothermal model is constant while temperature in the adiabatic model can be expressed as a function of the stagnation temperature. The Mach number was computed for the two models and plotted in Figure (4.5 1.8 T / T0 isentropic but rather a simple sample P / P0 isothermal T/T0 isothermal just to examine the difference 0. This P / P0 isentropic profile isn’t an ideal profile 0.7(b)).4 bounded.7(b)). While. The ratio of the velocities can be expressed as √ Ms kRTs Us √ = (4. the isentropic for the subsonic branch will be over prediction. 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.68 CHAPTER 4. using the isentropic model results in under prediction of the actual results for the velocity in the supersonic branch.

101) While the temperature ratio of the isothermal model is constant and equal to one (1). The temperature ratio of T /T0 int can be obtained via the isentropic model as T 1 = k−1 T0 int 1 + 2 M2 (4.6: Consider a diverging–converging nozzle made out of wood (low conductive material) with exit area equal entrance area. 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. If the nozzle was made from copper (a good heat conductor) a larger heat transfer occurs. Calculate the velocity at the exit using the adiabatic model.8) shows that the range between the predicted temperatures of the two models is very large.3. the conservation of the stagnation properties. while the range between the predicted pressure by the two models is relatively small. Example 4.4. 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. To demonstrate the relativity of the approach advocated in this book consider the following example.e. Using equations (4. The stagnation pressure is 5[Bar] and the stagnation temperature is 27◦ C.103) where z is an arbitrary point on the nozzle. ISENTROPIC TABLES 69 Two other ratios need to be examined: temperature and pressure. with known area ratio and known stagnation Potto–GDC provides the following table: . Assume that the back pressure is low enough to have supersonic flow without shock and k = 1.102) 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. Thus. The pressure ratio for the isentropic model is P = P0 int 1+ 1 k−1 2 2 M k−1 k (4.89) and the isentropic relationship. The initial stagnation temperature is denoted as T0 int .4. The throat area ratio to entrance area is 1:4 respectively. Figure (4. the sought ratio is provided.

777777778 11 k−1 T0 ∗ 1 + k− 1 + 2 k k Thus the stagnation temperature at the exit is T0ini = 1.0000 0.12556 0.253[m/sec] (4.70 M 0.9402 T T0 ρ ρ0 CHAPTER 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 = 1.253[m/sec] 4.36644 × 300 = 109. The exit velocity can be determined by utilizing the following equation √ √ Uexit = M kRT = 1. The exit velocity. Using the area ratio in Figure (4.4940 0.9).98511 0.921 T0exit The exit stagnation temperature is 1.105) .9910 1. the velocity can be calculated.08129 4.99572 0. the initial stagnation temperature is given as 300K . is √ √ U = M kRT = 2. the velocity in the copper nozzle will be larger than the velocity in the wood nozzle. ISENTROPIC FLOW A A P P0 A×P A∗ ×P0 0.9 ∼ 617.51183 4.4940/0.92 × 300 = 576. the maximum velocity cannot exceed the 691.2K . but in the literature some denote this function as I .9405 0. However.0000 4.50225 The exit Mach number is known and the initial temperature to the throat temperature ratio can be calculated as the following: 1 T0ini 1 = = = 0.14655 2.9402 1.4 × 287 × 109. 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 (4.4.11915 Even for the isothermal model. then.1 The Impulse Function Impulse in Isentropic Adiabatic Nozzle One of the functions that is used in calculating the forces is the Impulse function.02979 3.4 4. The Impulse function is denoted here as F .98934 0.6) or using the Potto–GDC obtains the following table M 1.9K . To explain the motivation for using this definition consider the calculation of the net forces that acting on section shown in Figure (4.36644 0.0 = 691.4 × 287 × 300.93[m/sec] 1. The exit temperature is 0.0000 0.104) As was discussed before.

Hence.105) with equation (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. 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. THE IMPULSE FUNCTION 71 The net force is denoted here as Fnet .106) Combining equation (4.107) 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 (4.4.106) and by utilizing the identity in equation (4. M and k .42) results in Rearranging equation (4.107) P2 A 2 P1 A 1 Fnet = 1 + kM2 2 − 1 + kM1 2 ∗ ∗ P0 A P0 A P0 A ∗ (4. the net force is Fnet k+1 = P0 A (1 + k ) 2 ∗ k k−1 F2 F1 − ∗ ∗ F F (4.109) (4. Hence.108) shows that the right hand side is only a function of x-direction Mach number and specific heat ratio. The mass conservation also can be applied to our control volume m ˙ = ρ 1 A 1 U1 = ρ 2 A 2 U2 (4. k .110) = 1 P∗ P0 2 ( k+1 ) k−1 k P1 A 1 1 + kM1 2 P0 A ∗ 1 (1 + k ) (4. 4.4. Defining a function that depends only Fig.112) .111) This ratio is different only in a coefficient from the ratio defined in equation (4. Thus. 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 (4.108) Examining equation (4.108) P1 A1 1 + kM1 F = ∗ ∗ F∗ P A (1 + k ) 2 (4.108) which makes the ratio a function of k and the Mach number.

stance (gas) thorough a converging nozzle S OLUTION The solution is obtained by getting the data for the Mach number. A2 A1 0.99380 ∼ 4.11164 0.009 A1 = = 2.94934 2. To obtained the Mach number.1 A ∗ P0 70061. provides the following √ √ m ˙ RT 1.058 and 500000 × 0.Fig.10: Schematic of a flow of a compressible subzle and pressure at point 1. 4.99751 0.52).003m2 entrance area is 0.003 A 2 P2 = ∼ 2.72 CHAPTER 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.1949 P0 P1 = 5.003[m ].009m2 2 A2 = 0.27353 0.1000 0.003 And utilizing again Potto-GDC provides M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.0times0.2121 With the area ratio of A A 0.0 × 400 × 287 A ∗ P0 = = ∼ 70061. the ratio of P1 A1 /A∗ P0 is needed to be calculated. To obtain this ratio the denominator is needed to be obtained.99380 5.7) 0. Utilizing Fliegner’s equation (4.7: 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. ISENTROPIC FLOW To demonstrate the usefulness of the this function consider a simple situation of the flow through a converging nozzle Example 4.2227 The pressure at point 1 is P1 = P 2 0.98526 0.776[Bar] P2 P0 .96355 2.76[N ] 0.2121 × = 5.058 0.2227 A A A2 0.94934/0.009[m ] and T0 = 400K 2 P2 = 50[Bar] the exit area is 0.96666 = 2.2121 the area ratio of at point 1 can be calculated.1774 2.99132 5.76 M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 1 for example (4.

1 T0 T0 P0 P0 A A P P 0.00 8.4.114) 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 (4.00 4.00 0.23.064 1. As previously.52828 0.114) Since U 2 /RT = kM 2 and the ratio of equation (4. ISOTHERMAL TABLE The net force is obtained by utilizing equation (4.4 × 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 (4.362 10.2 The Impulse Function in Isothermal Nozzle Previously Impulse function was developed in the isentropic adiabatic flow.5.064 5.53199 1.4.1 k 73 4.001 2.05 0.010 2.0E +6 4.113) 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. The same is done here for the isothermal nozzle flow model.52921 0.014 2.3: Isothermal Table M 0.115) 4.5 Isothermal Table Table 4. the definition of the Impulse function is reused.96666) ∼ 614[kN ] = 500000 × 2.2E +5 20.5 × (2.0E + 5 9.1949 − 0.116) (4.225 A×P A∗ ×P0 F F∗ .112) Fnet = P2 A2 P0 A ∗ k + 1 k−1 F2 F1 (1 + k ) − ∗ P2 A 2 2 F∗ F 1 × 2.064 1.000 1.949 5.87) transformed equation into (4.

718 3.035 3.569 3.0 7.062 1.690 1.000 5.0 5.143 4.8 0.3E +15 0.065 1.4E +20 0.064 2.2 0.100000 A×P A∗ ×P0 F F∗ 2.134 0.56232 0.3 0.0 0.73492 0.97156 1.000 1.875 4.055 1.429 1.763 1.10526 0.389 1.429 1.015317 7.681 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- .0 0.0 5.0 2.99967 1.375 1.20 1.026 19.111 1.6E +13 0.4E + 3 5.000 7.586 5.58824 0.0 0.237 1.40000 0.33333 0.6E + 4 2.245 2.000 3.500 5.0 0.4E + 8 0.11111 0.12246 0.60 1.7 0.500 10.000 6.0E + 6 0.0 5.75E −5 1.000 0.7E + 3 1.20000 0.14 0.431 4.399 0.83333 0.73278 0.41E −6 0.0 8.047 1.00370 0.128 1.134 9.90302 1.66667 0.281 1.800 1.665 0.500 7.540 4.14286 0.891 1.52632 0.287 1.3E +11 0.000 0.142 1.153 1.89348 1.939 2.78 0.00221 9.67383 0.4E + 3 2.41 40.856 4.8E + 2 4.5 0.2E + 3 3.51069 0.97274 1.025349 0.247 0.94147 1.9E + 2 8.000 3.6 0.1E + 3 0.2E +29 1.2E + 4 1.044 1.083 0.250 1.63276 2.719 5.90909 0.125 1.69449 2.020 1.62500 0.0 1.0 1.97376 1.564 1.0 0.000 1.50000 0.284 3.036 0.0 5.18182 0.12500 0.58985 0.333 2.11765 0.80844 1.667 1.209 0.194 1.079 1.98796 1.0 0.625 1.80 1.50618 4.3: Isothermal Table (continue) M 0.8E +18 0.889 2.41E −5 4.26634 0.000 4.0 0.5E + 2 0.500 8.15385 0.863 0.86329 0.000 0.21 1.00 2.10 1.000 1.500 2.000215 1.275 1.500 4.80528 0.281 1.74 CHAPTER 4.75344 1.000380 2.16090 0.33554 0.50 3.41686 0.114 1.000 1.22222 0.0 1.053 0.071758 90.3E + 9 0.271 0.958 1.9 1.00 T0 T0 P0 P0 A A P P 0.3E + 2 1.5E + 2 2.500 9.0 0.54322 0.007 5.047 0.000 9.464 1.20846 0.1E + 4 1.0 1.0 0.553 1.22881 15.599 0.200 1.004 0.70 1.000 3.296 5.021 0.61693 0.565 1.059 1.16667 0.3E +23 0.4E +26 0.500 6.30 1.90 2.183 1.6E + 5 1.50 1.0 0.13333 0.28571 0.56954 3.85853 1. ISENTROPIC FLOW Table 4.438 2.608 3.55556 0.161 2.000 8.2E + 3 7.71429 0.40 1.063 1.98750 0.00 1.29 80.62665 0.25000 0.4 0.328 1.76923 0.

19) are combined to form Cp dT ds = −z 1+ R R T T Z ∂z ∂T dP P (4.124) . Gibb’s equation (4. P ) and full differential is dh = ∂h ∂T dT + P ∂h ∂P dP T ∂h ∂T (4. With integration of equation (4.123) results U2 = 2 P1 P0 dP = ρ P P0 1 ρ0 P0 P 1 n dP (4. a real gas equation can be expressed in equation (3. Thus.122) when using relationship which is defined in equation (4.123) Notice. To carry the integration of the right hand side the relationship between the pressure and the density has to be defined.27) is reused to related the entropy change to the change in thermodynamics properties and applied on non-ideal gas. the differential of enthalpy is calculated for real gas equation of state as dh = Cp dT − T Z ∂z ∂T dP ρ (4.118) and second derivative is (4.120) P Equations (4.19).19) and dividing by equation (3.119) The definition of pressure specific heat is Cp ≡ Maxwell relation hence.117) Again. Since ds = 0 and utilizing the equation of the state dh = dP/ρ.6.122) At the stagnation the definition requires that the velocity is zero. h = h(T.121) P The mechanical energy equation can be expressed as d U2 2 =− dP ρ (4. that for perfect gas the n is substituted by k . THE EFFECTS OF REAL GASES 75 erties like the critical pressure and critical temperature.27) and (3. The enthalpy is a function of the temperature and pressure thus. The following power relationship is assumed ρ = ρ0 P P0 1 n (4.4. Differentiating equation (3.19) yields dz dρ dT dP = + + P z ρ T (4. ∂h ∂P =v−T ∂s ∂T T P First.

121) when ds = 0 results T2 T1 Cp dT = R T P2 z 1+ P1 T Z ∂z ∂T P dP P (4.129) For the case of n = 1 ρ∗ U ∗∗ 1 n m ˙ = A∗ P0 z0 RT0 P P0 2z0 RT0 ln P0 P (4.131) Integrating equation (4.19) results U2 = 2 P P0 z0 RT0 P0 P0 P 1 n dP (4.126) For n = 1 the integration becomes U= 2z0 RT0 ln P0 P (4.76 CHAPTER 4.127) It must be noted that n is a function of the critical temperature and critical pressure.132) . 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 (4.130) The Mach number can be obtained by utilizing equation (3. ISENTROPIC FLOW Substituting relation for stagnation density (3. Thus.128) Where ρ∗ is the density at the throat (assuming the chocking condition) and A∗ is the cross area of the throat.125) For n > 1 the integration results in 2n 1− n−1 n−1 P ( n ) P0 U= z0 RT0 (4.34) to defined the Mach number as M=√ U znRT (4. The mass flow rate is regardless to equation of state as following m ˙ = ρ ∗ A∗ U ∗ (4.

4.133) ρ0 d ρ P P0 (4.33) is applicable for some ranges of relative temperature and pressure (relative to critical temperature and pressure and not the stagnation conditions). 2n n−1 P P0 n−1 n U= When n = 1 or when n → 1 z0 RT0 1− (4.134) It was shown in Chapter (3) that (3.139) Fliegner’s number for n = 1 is Fn = mc ˙ 0 =2 A ∗ P0 P∗ P0 2 − ln P∗ P0 (4.136) The mass flow rate for the real gas m ˙ = ρ ∗ U ∗ A∗ A ∗ P0 m ˙ =√ z0 RT0 And for n = 1 A ∗ P0 m ˙ =√ z0 RT0 Fliegner’s number in this case is Fn = mc ˙ 0 A ∗ P0 2n n−1 P∗ P0 1 n 2n n−1 P∗ P0 1 n 1− P∗ P0 (4.6.140) .137) 2n n−1 2z0 RT0 ln P0 P (4. THE EFFECTS OF REAL GASES 77 To carryout the integration of equation (4.138) 1− P∗ P0 (4.132) looks at Bernnolli’s equation which is dU 2 =− 2 After integration of the velocity dU 2 =− 2 P/P0 1 dP ρ (4.135) U= 2z0 RT0 ln P0 P (4.

132) can be integrated T0 = T P0 P R Cp ∂z )P ] [z+T ( ∂T (4.141) When n = 1 or more generally when n → 1 this is a ratio approach P∗ √ = e P0 (4.145) For n = 1 the Mach number is M= 2 z 0 T0 P 0 ln z T P (4.149) .148) ∂z n z +T ( ∂T )P ] 1 + n ( 1−n )[ 2 (4.144) 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 (4.147) for n = 1 T0 = T The critical temperature is given by T∗ = T0 eM [z+T ( 2 ∂z ∂T )P ] (4.78 The critical ratio of the pressure is P∗ = P0 2 n+1 CHAPTER 4.146) The pressure ratio at any point can be expressed as a function of the Mach number as n−1 ∂z z +T ( ∂T )P ] T0 n − 1 2 ( n )[ = 1+ M T 2 (4. equation (4.142) To obtain the relationship between the temperature and pressure. ISENTROPIC FLOW n n−1 (4. Note (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.

The stagnation pressure is obtained from P0 = P0 2. the pressure 2[Bar].04943 T0 300 T = ∼ 713.82K T 0.42027 0. The solution is simplified by using Potto-GDC for M = 2.802[Bar] P 0.04943 0. .8: A design is required that at a specific point the Mach number should be M = 2.42027 The stagnation temperature is T0 = 3. Calculate the stagnation pressure and the stagnation temperature.61 the results are M 2.405. S OLUTION 1.11761 2.6100 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. THE EFFECTS OF REAL GASES and for n = 1 T∗ = T0 79 e−[z+T ( ∂z ∂T )P ] (4. ii. Are the stagnation pressure and temperature at the entrance different from the point? You can assume that k = 1. and temperature 300K .152) Example 4.151) For the case of n = 1 the mass flow rate is P0 A ∗ n m ˙ = c0 e M2 n−1 2 M 1+ 2 n+1 n−1 (4.4.14366 2.150) 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 (4.6. Of course. i. Calculate the area ratio between the point and the throat.9066 0. the stagnation pressure is constant for isentropic flow. iii.61.61 P = ∼ 52.

80 CHAPTER 4. ISENTROPIC FLOW .

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

Clearly. . The chemical reactions (even condensation) are neglected.3) Px − P y = ρ x U y 2 − ρ y U x 2 (5.2) The equation of state for perfect gas reads P = ρRT (5. one can note that there are two solutions because of the quadratic of equation (5. then there are four unknown conditions downstream. it can be written that The process is adiabatic. The conditions to the right of the shock wave are uniform.3). For the adiabatic process. Mathematically. the second law is expressed by the entropy. One cannot deduce the direction of the flow from the pressure on both sides of the shock wave. It is customary in this field to denote x as the upstream condition and y as the downstream condition. and therefore the energy equation can be written as C p Tx + Ux 2 Uy 2 = C p Ty + 2 2 (5. 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. but different from the left side. the isentropic transition assumption is not appropriate in this case because the shock wave is a discontinued area. and as it will be shown. Physics dictates that there is only one possible solution. This law dictates the direction of the flow. A system of four unknowns and four equations is solvable. or nearly adiabatic. The only tool that brings us to the direction of the flow is the second law of thermodynamics. Therefore. These two possible solutions refer to the direction of the flow. NORMAL SHOCK assumed to be uniform2 .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. In mathematical terms. the momentum is the quantity that remains constant because there are no external forces. 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). Nevertheless. The transition in the shock is abrupt and in a very narrow width. the increase of the entropy is fundamental to the phenomenon and the understanding of it.82 CHAPTER 5. Thus. and the shock occurs at a very narrow section. the entropy must increase. it can be written as follows: sy − s x > 0 (5.4) If the conditions upstream are known.1) In a shock wave. the gas flows from a supersonic flow to a subsonic flow.

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

11) . The reason that dimensionless forms are heavily used in this book is because by doing so it simplifies and clarifies the solution. T0 x = T0 y . c2 = kRT to be used for the Mach number definition. Note that the Mach number is a function of temperature. 5. provides a way to obtain the speed of sound definition/equation for perfect gas. and (5.9) (5. It can also be noted that in many cases the dimensionless equations set is more easily solved. as follows: Px 2 Py 2 Ux 2 = Uy 2 Tx kRTx Ty kRTy cx 2 cy 2 (5.1) substituting for density. NORMAL SHOCK 5. Ty = Tx Ty T0 y Tx T0 x = 1+ 1+ 2 k −1 2 Mx 2 k−1 2 My (5.1.8) results in Py 2 Px 2 2 U = Uy 2 x R 2 Tx 2 R 2 Ty 2 (5. The relationship between the temperature and the stagnation temperature provides the relationship of the temperature for both sides of the shock. The only issue left to derive is the relationship between Mx and My .3) can be converted into a dimensionless form. at least.1. k.8) Multiplying the two sides by the ratio of the specific heat.84 CHAPTER 5. (5.1 5.2 Formal Model Equations (5.10) Note that the speed of sound on the different sides of the shock is different.2). and thus for known Mx all the other quantities can be determined. ρ.7) All the other relationships are essentially derived from this equation. numerically. From the continuity equation (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.1). Utilizing the definition of Mach number results in Py 2 Px 2 Mx 2 = My 2 Tx Ty (5. the equation of state yields Px Py Ux = Uy RTx RTy Squaring equation (5. The analytical solution is discussed in the next section.

T0 y = T0 x .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 (5.12) (mass) the temperature ratio can be eliminated. momentum.13) (energy) and equation (5.3) 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 (5.16) (5.14) The pressure ratio in equation (5.14) transforms the momentum equation (5.15) 2  P   U  kRT = kP M 2  RT kRT  c2   (5.17) The relationship between Mx and My is needed to be solved from the above set of equations.1. The loss of the total pressure ratio can be expressed by utilizing the relationship between the pressure and total pressure (see equation (4.11) results in Ty = Tx Py Px 2 85 My Mx 2 (5.16) can be interpreted as the loss of the static pressure.18) .13) means that the stagnation temperature is the same. ρU 2 is identical to kP M 2 because M2 ρ ρU 2 = Using the identity (5. Under the perfect gas model.5.2) into Px + kPx Mx 2 = Py + kPy My 2 Rearranging equation (5. Py M y Px M x 2 = 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5. From equation (5. SOLUTION OF THE GOVERNING EQUATIONS Rearranging equation (5.13) It can also be observed that equation (5.12) Energy equation (5. This relationship can be obtained from the combination of mass. and energy equations.15) yields Py 1 + kMx 2 = Px 1 + kMy 2 (5.

2nd ed. M. 1990.20) It can be observed that equation (5. The solution can be obtained by rewriting equation (5. for example.16) and equation (5. Mx .19) as a polynomial (fourth order). . The pressure ratio between the two sides can now be as a function of only a single Mach number. Thus.e.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. According to the Gauss Biquadratic Reciprocity Theorem this kind of equation has a real solution in a certain range3 which will be discussed later.21) Equation (5. pp. i. The second solution is where the shock wave occurs.24) 3 Ireland.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. Utilizing equation (5. one solution is My = M x (5. 108-137. Clearly. NORMAL SHOCK Combining the results of (5.18) with equation (5. Tx = Ty . 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) Equation (5. Px = Py .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. and Rosen. the pressure and the temperature from both sides of the nonexistent shock are the same.86 CHAPTER 5. New York: Springer-Verlag.22) The first solution (5.19) and divide it by My 2 − My 2 1+ k−1 My 2 + My 2 − kMy 2 My 2 = 0 2 (5.19) is biquadratic.20) is the trivial solution in which the two sides are identical and no shock wave occurs. 9 in A Classical Introduction to Modern Number Theory. ”Cubic and Biquadratic Reciprocity. in this case.” Ch. K.21) becomes My 2 = Mx 2 + 2k k−1 Mx 2 k−1 2 −1 (5.

is transformed into Ty = Tx Py Px k+1 k−1 Py Px k+1 Py k−1 Px + 1+ (5. The Maximum Conditions The maximum speed of sound is when the highest temperature is achieved.9 0.2 0. The speed of sound at that velocity is √ c∗ = kRT ∗ (5.29) Mx Fig.4 0. The maximum temperature that can be achieved is the stagnation temperature Umax = 2k RT0 k−1 (5. And the temperature ratio.25) In the same way. 5.5 0. as a function of pressure ratio.5.1 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 (5.6 0.28) Based on this definition a new Mach number can be defined M0 = U c0 (5. Mx .7 0.3 0.26) which is associated with the shock wave.3: The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number.30) . The Star Conditions The speed of sound at the critical condition can also be a good reference velocity. SOLUTION OF THE GOVERNING EQUATIONS 87 The fact that the pressure ratio is a function of the upstream Mach number.8 0. provides additional way of obtaining an additional useful relationship.1.27) My 1 0. 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.

88 CHAPTER 5.34) .33) Dividing the mass equation by the momentum equation and combining it with the perfect gas model yields c1 2 c2 2 + U1 = + U2 kU1 kU2 Combining equation (5.35) the following can be obtained: U1 U2 = c ∗ 2 or in a dimensionless form M ∗ 1 M ∗ 2 = c∗ 2 (5. the speed of sound is different in these adjoining mediums.37) (5.3 Prandtl’s Condition It can be easily observed that the temperature from both sides of the shock wave is discontinuous. Therefore. It is therefore convenient to define the star Mach number that will be independent of the specific Mach number (independent of the temperature).35) (5. NORMAL SHOCK In the same manner. c2 U2 c∗ 2 c∗ 2 k + 1 ∗2 + = + = c k−1 2 k−1 2 2(k − 1) (5.36) (5. an additional Mach number can be defined as M∗ = U c∗ (5.32) The jump condition across the shock must satisfy the constant energy. c U c U = ∗ = ∗M c∗ c c c M∗ = (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 (5.1.31) 5.33) and (5.

4 × 287 × 273 = 331. the Mach number after the shock.0 80. Finding the Mach number for a pressure ratio of 8.0 60.5[bar] and Tx = 273K .2m/sec 1 2 3 Fri Jun 18 15:48:25 2004 4 5 Mx 6 7 8 9 10 .4: The ratios of the static properties of the two sides Analysis: of the shock.2 Operating Equations and Analysis In Figure (5.1110 3.5/0.3088 0.0 10. It can be noticed that the density ratio (velocity ratio) also has a finite value regardless of the upstream Mach number. First.0 70.0 90.0 100.0 20.0 110. at a pressure of 0. The working equations were presented earlier.47642 2. Ty Tx ρy ρx Py Px P0y P0 x Mx 2.7245 My 0. P0y /P0x . OPERATING EQUATIONS AND ANALYSIS 89 5.38109 To illustrate the use of the above equations.0272237 =⇒ P0x = 1.9360 8.2. Note that the My has a minimum value which depends on the specific heat ratio. Calculate the temperature. the velocity can readily be calculated. P0x is known). The typical situations in which these equations can be used also include the moving shocks. 120. pressure. i. The relationship that was calculated will be utilized to obtain the ratios for the downstream of the normal Px shock.0 Py/Py.1) for k = 1. an example is provided.32 is only a few mouse clicks away from the following table. 5. and velocity downstream of the shock.5. the known information are Mx = 3. Example 5.0272237 = 55.30879 and k = 1.1: Air flows with a Mach number of Mx = 3.4 or utilizing Potto-GDC for for value of the specific heat ratio.5 [bar] and a temperature of 0◦ C goes through a normal shock.0 30. the total pressure can be obtained (through an isentropic relationship in Table (4. P = 0.3).0 50. are plotted as a function of the entrance Mach number. The equations should be used with the Mach number (upstream or downstream) for a given pressure ratio or density ratio (velocity ratio). Tx . and the ratio of the total pressure.0 0.0 40. Also with the temperature. px = 1. total pressure.e. My . Using these data. ρy/ρx and Ty/Tx as a function of Mx Shock Wave relationship Py/Px ρy/ρx Ty/Tx S OLUTION Fig..2). This kind of equations requires examining Table (5.1[bar] √ √ 0x cx = kRTx = 1.

39) .16) and by utilizing the limit that was obtained in equation (5. this technique is used mostly in obtaining analytical expressions for simplified models.85714.40) (5.47519 Ty Tx ρy ρx CHAPTER 5.8571 10. Uy = 993.3).1 The Limitations of the Shock Wave When the upstream Mach number becomes very large. it refers to a case where only a “small shock” occurs. This approach had a major significance and usefulness at a time when personal computers were not available. This technique also has an academic value and therefore will be described in the next version (0.32834 Ux = Mx × cx = 3 × 331. The limits of the pressure ratio can be obtained by looking at equation (5. NORMAL SHOCK Py Px P0 y P0 x 2.09[bar] 5.2.2 = 993.32834 × 55.2 Small Perturbation Solution The small perturbation solution refers to an analytical solution where only a small change (or several small changes) occurs.0000 My 0. 5.1[bar] = 18.6[m/sec] Now the velocity downstream is determined by the inverse ratio of ρy /ρx = Ux /Uy = 3. Now.23) transforms equation (5.5 series). In this case. during the writing of this version of the book.6[m/sec] P0y = P0y P0x × P0x = 0.38).39) into ˆ= P 2k Mx 2 − 1 k+1 (5.38) This result is shown in Figure (5. which is up to Mx = 1.90 Mx 3. the downstream Mach number (see equation (5.3.22)) is limited by My = 2 :  ∼0 2  1 + (k −1)Mx 2 2k k−1 >   − M1 x 2 ∼0 = k−1 2k (5.6/3.85714 = 257.3333 0.6790 3.2. The strength of the shock wave is defined as ˆ = Py − P x = Py − 1 P Px Px By using equation (5.

flow L5M g direction Q5R For example.5.2.3 Shock Thickness The issue of shock thickness (which will be presented in a later version) is presented here for completeness. however. 5. the shock wave is not stationary. to convince the students that indeed the assumption of very thin shock is validated by analytical and experimental studies. THE MOVING SHOCKS or by utilizing equation (5. This issue has a very limited practical application for most students. The shock thickness can be defined in several ways.24) the following is obtained: ˆ= P 2k k−1 2 k−1 ρy ρx 91 −1 −1 − ρy ρx (5.3 The Moving Shocks In some situations. 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. when a valve is suddenly hji NPO  S T f d e b ` . This kind of situation []\ ^]_ arises in many industrial applications. the issue should be presented.41) 5.3.

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

The problem can be reduced to the previously studied shock. An observer moving with the shock will notice that the pressure in the shock is Px = P x Py = P y (5. The reluctance to deal with the most general case is due to fact it is highly specialized and complicated even beyond early graduate students level. The upstream prime Mach number can be defined as Us − U x Us Mx = = − Mx = Msx − Mx (5. (refer to Figure (5. Here. 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.45) Ux = U s − U x (5.47) My = cy cy Uy = U s − U y (5. In such a case.44) . and the tools developed in this chapter can be employed. Note that this notation is contrary to the conventional notation found in the literature. to the stationary case when the coordinates are attached to the shock front.43) The temperature measured by the observer is Tx = T x Ty = T y Assuming that the shock is moving to the right. In cases where the shock velocity can be approximated as a constant (in the majority of cases) or as near constant. equations. The general case is where two shocks collide into one shock and moves upstream or downstream is the general case. For this analysis.92 CHAPTER 5.42) (5.e..5)) the velocity measured by the observer is Where Us is the shock velocity which is moving to the right. The reason for the deviation is that this choice reduces the programing work (especially for object–oriented programing like C++). the steady state is obtained in the moving control value. NORMAL SHOCK them at this stage. the previous analysis. 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. i. 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 coordinates move with the shock. Msx = U cx .46) cx cx It can be noted that the additional definition was introduced for the shock upstream s Mach number. The downstream prime Mach number can be expressed as Us Us − U y = − My = Msy − My (5. the prime ’ denote the values of the static coordinates.

and is “converted” or added into the total temperature. The relationship between the two new shock Mach numbers is cy Us Us = cx cx cy Msx = Ty Msy Tx (5. at a very high velocity.53) Again.5.50) The same can be said for the “downstream” side of the shock. . This difference can also be viewed as a result of the unsteady state of the shock. Msy .51) leads to T0y − T0x = Us Ty k − 1 Tx k − 1 (Msx − 2Mx ) − (Msy − 2My ) cx 2 cy 2 (5.v.51) It should be noted that the stagnation temperature (in the stationary coordinates) rises as opposed to the stationary normal shock. this difference in the moving shock is expected because moving material velocity (kinetic energy) is converted into internal energy. The rise in the total temperature is due to the fact that a new material has entered the c.48) The “upstream” stagnation temperature of the fluid is T0x = Tx 1 + k−1 Mx 2 2 (5.3. THE MOVING SHOCKS 93 Similar to the previous case.49) and the “upstream” prime stagnation pressure is P0x = Px k−1 Mx 2 1+ 2 k k−1 (5. 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 (5.52) − Tx 1 + k−1 Mx 2 2 −Tx Msx and according to equation (5. The difference between the stagnation temperature is in the moving coordinates T0y − T0x = 0 (5. an additional definition was introduced for the shock downstream Mach number.

the shock velocity “downstream” is Us . . The sudden closing of the valve must result in a zero velocity of the gas. In the moving coordinates. NORMAL SHOCK 5. Ux . Equations (5. or Msy instead of the velocity because it combines the temperature and the velocity into one parameter. Fig. The upstream is on the left of the shock. ­W®°¯ Stationary Coordinates À2ÁW›ÃÄÁWÅ:Æ Ç ³3´¶µZ· ¾›¿ ¼W½ ¸ p ¹ º¥¸#» Moving Coordinates c.v. the shock is moving to the left.v. The information propagates upstream in which the gas velocity is converted into temperature.54) The “downstream” Mach number reads My = |Us | = Msy cy (5. The flow is moving to the right. The “upstream” Mach number reads Mx = Us + U x = Msx + Mx cx (5. and for a given specific heat ratio. In many such cases the steady state is established quite rapidly. 5.56) The prominent question in this situation is what will be the shock wave velocity for a given fluid velocity. in a sudden and complete closing of a valve.42) to (5. It turns out that it is very useful to use the dimensionless parameter Msx .55) Again. The “upstream” or the “downstream” Mach number is not known even if the pressure and the temperature downstream are given. The difficulty lies in the jump from the stationary coordinates to the moving coordinates. In such a case.53) can be transformed into simpler equations when Mx is zero and Us is a positive value.3.94 CHAPTER 5. for example.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. This situation arises in many cases in the industry. This shock is viewed by some as a reflective shock. 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. 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 (5.6: Comparison between a stationary shock and a moving shock in a stationary medium in ducts.

From the same figure it can also be observed that a high velocity can result in a much larger velocity for the reflective shock.58) but it seems utilizing numerical methods is much more simple. Only 2 one real solution is possible. 1 − 2 (the coefficient is only approximated as 0.7).58) pro1 vides that Msx ∼ 1 + 2 and Msy = Fig. For example.7: The moving shock Mach numbers as a 1 result of a sudden and complete stop.55) by 2 95 ( My ) = Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. Mx ∼ equation (5. Note that to achieve supersonic velocity (in stationary coordinates) a diverging– converging nozzle is required.57) And substituting equation (5. 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. Mx .58) can be obtained by several numerical methods.57) into (5.58) and the rest of the right–hand k=14 3 side show clearly that Msx has four Msx possible solutions (fourth–order polyMsy nomial Msx has four solutions). the result is about double the sonic velocity of the reflective shock.5) as shown in Figure (5. a Mach number close to one (1). which can easily be obtained in a Fanno flow. For very small values of the upstream Mach Thu Aug 3 18:54:21 2006 number. . Sometimes this phenomenon can have a tremendous significance in industrial applications. 5. Two numerical methods and the algorithm employed to solve this problem for given. is provided herein: (a) Guess Mx > 1.3.58) The temperature ratio in equation Shock in A Suddenly Close Valve (5.54) and (5. an analytical solution can be obtained for equation (5.48) results in f (Msx ) Mx = Tx Ty Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5.5. THE MOVING SHOCKS through equations (5. Here no such device is needed! Luckily and hopefully. The typical method is 0 0. 1 Note. The solution to equation (5.1 1 Mx the “smart” guessing of M sx.

and adjust the new guess Mx > 1 accordingly. ×0ØÚÙ Î0Ï:Ð¥ÑÓÒÕÔ ÈÊÉ&ËÍÌ Û#Ü Ö c. 5.8(b))).” 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. the shock moves in a constant velocity (5. if not satisfactory use the new Mx to calculate Mx = 1 + Mx then return to part (b). With the exception of close proximity to the valve.v. (a) Guess Mx = 1 + Mx . Using a coordinates system which moves with the shock results in a stationary shock and the flow is moving to the left see Figure (5.3.8: A shock moves into a still medium as a result of a sudden and complete opening of a valve . (c) calculate the Mx = Mx − Tx Ty M y (d) compare the new Mx approach the old Mx . NORMAL SHOCK (b) Using shock table or use Potto–GDC to calculate temperature ratio and My . (a) Stationary coordinates (b) Moving coordinates Fig. a shock is created and propagates downstream. 5.8(a)). (b) using the shock table or use Potto–GDC to calculate the temperature ratio and My . The “upstream” will be on the right (see Figure (5. è0éWêÄè ë›ì è0é6í á0âã©á ä Upstream ÝÊÞ&ßÍà åçæ c. The second method is “successive substitutions.v.8(b)).96 CHAPTER 5. (c) Calculate the Mx = Mx − Tx Ty M y (d) Compare to the calculated Mx to the given Mx .

60) Us = Msx cx (5.62) Shock in A Suddenly Open Valve 1. what will be the shock velocity for a given upstream Mach number? Again.5 0 Number of Iteration 10 0 Shock in A Suddenly Open Valve k = 1 4. The “upstream” Mach number is Mx = The “downstream” Mach number is My = Us − U y = Msy − My cy (5. The same question that was prominent in the previous case appears now. It has to be noted that the “upstream” and “downstream” are the reverse from the previous case.3 Mx My Ty/Tx Ty/Tx 0.5 3 2.3 (b) My = 1. the relationship . THE MOVING SHOCKS 97 Similar definitions of the right side and the left side of the shock Mach numbers can be utilized. 5.5 4 3.3.3 Mx My 1.59) 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.5 1 1 0. My’ = 1. 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. My’ = 0.5 1.9: The number of iterations to achieve convergence.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 (5.3 Fig.25 2 1.75 k = 1 4.75 5 10 Number of Iteration 15 20 Wed Aug 23 17:20:59 2006 Wed Aug 23 17:46:15 2006 (a) My = 0.5.

3 and My = 1. the calculations for Us can be obtained from equation (5. the temperature.” This method is applied by the following algorithm: (a) Assume that Mx = 1.65) . There are three solutions of which the first one is Msx = 0 and this is immediately disregarded.64) Ty My + M y Tx Equation (5. The results show that the convergence occurs very rapidly (see Figure (5.0.63) can be solved. and. it creates a shock that moves at a speed greater than that of the piston itself. no Mach number is given but pieces of information on both sides of the shock.98 between the two sides is Msy = My + CHAPTER 5. Therefore. for most practical purposes. sufficient results can be achieved after 3-4 iterations.3. One such methods is “successive substitutions. It is common practice to solve this equation by numerical methods.63) Since Msx can be represented by Msy theoretically equation (5. NORMAL SHOCK (Msx ) + 2k k−1 2 (Msx ) − 1 2 k−1 2 (5. To illustrate the convergence of the procedure. other conditions ahead of the shock. the larger number of the iterations required to achieve the same accuracy. (d) Check the new and improved Mx against the old one. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. The other two solutions are Msx = (k + 1)Uy ± Uy (1 + k ) 4 cx 2 + 16cx 2 (5. In this case.64) is a quadratic equation for Msx . If it is satisfactory. The unknown data are the piston velocity.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. stop or return to stage (b). Yet. consider the case of My = 0. Piston Velocity When a piston is moving. (c) Utilizing Mx = calculate the new “improved” Mx .9)). The larger the value of My .

and the only solution left is Msx = (k + 1)Uy + Uy (1 + k ) 4 cx 2 + 16cx 2 (5. Shock–Choke Phenomenon Assuming that the gas velocity is supersonic (in stationary coordinates) before the shock moves.5 ence is that the actual velocity 1. shock.5. Mx = Msx → ∞. The differ1.3. and the downstream Mach y y(max) Maximum My’ . 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. the shock Mach number cannot be increased. It must be noted 1 that in the previous case of 0.66) or in a dimensionless form Msx = (k + 1)Myx + Myx (1 + k ) 4 2 + 16 (5.75 suddenly and completely clos0. This shock–choking phenomenon Shock in A Suddenly Open Valve Maximum M ’ possible is somewhat similar to the 2.68) to explain the next Shock-Choke phenomenon. k ing of valve results in no limit (at least from the model point Thu Aug 24 17:46:07 2006 of view).10: The maximum of “downstream” Mach number as a function of the specific heat. 5.67) Where the “stange” Mach number is Myx = Uy /cx . what is the maximum velocity that can be reached before this model fails? In other words. k. This phenomenon occurs in many industrial applications.25 has no limit.75 els (later chapters). Consider when the “upstream” Mach approaches infinity. To explain this phenomenon. look at the normal Fig. Some use equation (5.5 M choking phenomenon that was 2.68) 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. The limit of the equation when cx → ∞ leads to Msx = (k + 1)Myx 4 (5. such as the internal combustion engines and die casting.5 The spesific heat ratio. THE MOVING SHOCKS 99 The negative sign provides a negative value which is disregarded.25 discussed earlier in a nozzle 2 flow and in other pipe flow mod1.

NORMAL SHOCK number.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. the results is My = 2 k (k − 1) (5. The Mach number is reduced to the maximum value very rapidly.26) and (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. To understand this limit. From the upstream side.25) into equation (5. is approaching to (k − 1)/2k .72) + When the pressure ratio is approaching infinity (extremely strong pressure ratio). . according to equation (5. The reduction occurs by an increase of temperature after the shock or a stationary shock occurs as it will be shown in chapters on internal flow. A similar phenomenon to the choking in the nozzle and later in an internal pipe flow is obtained.69) Mx = Msx Thus. These limits determine the maximum velocity after the shock since Umax = cy My .73) What happens when a gas with a Mach number larger than the maximum Mach number possible is flowing in the tube? Obviously. One can view this as the source of the shock–choking phenomenon.23) to this situation the following is obtained: Msx = k+1 2k Px −1 +1 Py (5. the Mach number is approaching infinity because of the temperature ratio but the velocity is finite. the Mach number is ∞    Ty k − 1 =   T 2k   x (5. the semi steady state described by the moving shock cannot be sustained. By applying equation when the pressure ratio is approaching infinity Px (5.38). consider that the maximum Mach number is obtained Py → ∞.71) 1− Substituting equations (5.100 CHAPTER 5.

the shock–choke phenomenon means that the Mach number is only limited in stationary coordinates but the actual flow rate isn’t.54 750. THE MOVING SHOCKS k 1.40 2.33968 0.43301 0.80 1.00 2.45374 0.56 668.85 Table of maximum values of the shock-choking phenomenon.86 216507.56 711.81 My 0.74 681.87039 0.54772 My 2.51177 0.29 188982.36 270031.1785 1.00 729.52223 0.20 2.48667 0.10 2.50 Mx 1073.25 985.70 1.50000 0.0815 1.53161 0.13).78 261117.93 243332.64 255883.62 695.25) can be transferred for large pressure ratios into m ˙ ∼ A Ty Px k − 1 Tx k + 1 (5. The only limitation is that the “downstream’ gas velocity is higher than the “upstream” gas velocity as shown in Figure (5.96 204124.37797 0.09 265805.09 833. .2964 1.30 1.3.47141 0.02 773.99 235702.3 Partially Open Valve The previous case is a special case of the moving shock. ∞.44 273861.90 2.50 1.4434 1.05 226871. 5.75) Since the right hand side of equation (5.60 1.2645 1.23 873.5. with the exception of Ty the mass flow rate is approaching infinity when the pressure ratio is approaching infinity.61 801.54006 0.79 250000.81786 0.85 922.30 2.40825 0. The general case is when one gas flows into another gas with a given velocity.75) is constant.8898 1. The mass flow rate when the pressure ratio is approaching infinity.3.93048 0.74) Equation (5.6330 1.74) and equation (5.73029 Ty Tx 101 169842.00000 0.77151 0. is cy ρy m ˙ = U y ρ y = M y cy ρ y = M y A √ My kPy = RTy kRTy Py RTy (5. Thus.40 1.

102 CHAPTER 5. The procedure for the calculations can be done by the following algorithm: (a) Assume that Mx = Mx + 1. 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.77) (5. stop or return to stage (b). If it is satisfactory.v. It has to be mentioned that the temperature “downstream” is unknown (the flow of the gas with the higher velocity). (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. Ty (Msy ) − Mx Tx . (a) Stationary coordinates (b) Moving coordinates Fig. 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. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. ò îÊï&ðÍñ ó Ux = Us − Ux Upstream ùÊú&ûÍü ýçþ c. NORMAL SHOCK ÿ¡ £¢Äÿ¥¤§¦ ÿ¡ ©¨ Uy > U x ô0õÚö Ux ÷#ø c.11: A shock moves into a moving medium as a result of a sudden and complete open valve.v.

vicinity of the previous value (see following question/example).4 Partially Closed Valve .8 1. THE MOVING SHOCKS 103 Shock in A Suddenly Open Valve Earlier.6 sults show that the shock choking phe0. it was shown that the shock chokk=14 ing phenomenon occurs when the flow is 1 running into a still medium.4 2. M ’ = 0.4 0. My .9 0.3. x x x My y 5.3. This phe0. Figure (5.6 M’ bers for various static Mach numbers.0 0.9 nomenon also occurs in the case where M ’ = 0.8 M ’ = 0. 5.5 nomenon is still there (the Mach number is 0.12: The results of the partial opening that the maximum can also occurs in the of the valve.3 exhibits some “downstream” Mach num0.12) 0. not the actual flow). Mx .5.2 a faster flow is running into a slower fluid. Thu Oct 19 10:34:19 2006 and for various static “upstream” Mach numbers.4 limited.8 1.2 2 2.7 The mathematics is cumbersome but re0. The figure demonstrates Fig.

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. c.79) . The information propagates upstream in the same way as before.78) Ux = U s + U x Uy = U s + U y (5. (a) Stationary coordinates (b) Moving coordinates Fig. Similar equations can be written: (5.v.     Ux Uy Ux = Us + Ux Upstream Uy = Us + Uy ρ y Py Ty c.v.

0 1.104 CHAPTER 5.132 Ty Tx Py Px P0 y P0 x 0.4) one can obtain the following table: Mx 2. Assume the specific heat ratio is 1. Thus. the Mach velocity ahead of the shock has to calculated.5 Worked–out Examples for Shock Dynamics Example 5. First.63955 . Compute the pressure and the temperature behind the shock.604 4.1206 My Mx My 1.80) My = M s + M y (5.3 × 287 × 300 By utilizing Potto–GDC or Table (5.296 1.3.81) For given static Mach numbers the procedure for the calculation is as follows: (a) Assume that Mx = Mx + 1. 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. Ty (Msy ) + Mx Tx 5.953 0.2: 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.54220 0. stop or return to stage (b).3. (b) . NORMAL SHOCK Mx = M s + M x (5. If it is satisfactory. S OLUTION It can be observed that the gas behind the shock is moving while the gas ahead of the shock is still. My = √ U kRT =√ 450 ∼ 1. it is the case of a shock moving into still medium (suddenly opened valve case).

0 Ty Tx Py Px P0 y P0 x 0. Calculate the velocity and the pressure behind the reflecting shock.724 1. If this detail is not sufficient then simply utilize the iterations procedure described earlier and obtain the following: i 0 1 2 3 4 Mx 2. THE MOVING SHOCKS Using the above table.2) you can find the proper Mx .0445 My Mx My 0. Simply using the Potto–GDC provides for the temperature and velocity the following table: Mx 2.3.724 1.53487 0.56995 Ty Tx My 0.042 2.2K Tx 105 In same manner.56995 0.0 ∼ 4. .2961 1.2961 2.3 × 287 × 300 ∼ 378.044 2.045 2.604 × 300 ∼ 481. The specific heat ratio can be assumed to be k = 1.4. Then using Table (5.953[Bar] Px The velocity behind the shock wave is obtained √ m Uy = Mx cx = 1.710 0.70009 If you insist on doing the steps yourself.72 sec Example 5.724 4. The static pressure at the tube is 2Bar and the (static) temperature of 300K .9432 1. find the upstream prime Mach.0 0. The gas is brought into a complete stop by a sudden closing a valve. it can be done for the pressure ratio as following Py = P y = Py Px = 4. S OLUTION The first thing that needs to be done is to find the prime Mach number Mx = 1.2961.722 1.0 1. the prime properties can be found.0 0.724 The table was obtained by utilizing Potto–GDC with the iteration request.56994 0. At this stage the reflecting shock velocity is unknown.132 × 1.56995 1.044 My 0.0 0. the temperature behind the shock is Ty = T y = Ty Tx = 1. Then.2961.0 0.953 × 1.5. Mx to be 1.3: Gas flows in a tube with a velocity of 450[m/sec].57040 0.

1220 My Mx My Ty Tx Py Px P0 y P0 x 0.0000 3.55830 Example 5.3574 in the moving shock table provides Mx 2.52778 Ty Tx ρy ρx Py Px P0 y P0 x 2.3.89509 0.1583 6.4) for the calculation procedure.4: What should be the prime Mach number (or the combination of the velocity with the temperature.99813 If the information about the iterations is needed please refer to the following table.52778 0. Hint. Potto-GDC provides the solution of the above data Mx 1.0 2.3).55832 using the required Mx = 2.106 CHAPTER 5.3574 My Mx My Ty Tx Py Px P0 y P0 x 0. S OLUTION Refer to section (5.317 0.3574 My 0. Potto-GDC provides for this temperature ratio the following table: Mx 2. It can also be obtained from the stationary normal shock wave table.20000 1.0789 1. Calculate the speed of the shock when a valve is closed in such a way that the Mach number is reduced by half.3166 0. 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).40000 0. .3020 0. NORMAL SHOCK Example 5. 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 (5.4.000 6.5: A gas is flowing in a pipe with a Mach number of 0.78928 0.

S OLUTION Using the procedure described in this section.4000 1.20000 0.20000 0.0106 1. Calculate the time Fig.89508 0.1288 1.1220 My 0. end of the tube of 1.1216 1.0789 2.1220 1.89504 0.0789 1.89509 0.1221 1.0789 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 1.0789 1.20000 0.20000 0.3025 1.3019 1.4 and 300◦C.89494 0.1222 1.0030 1.0793 1.92479 0.20000 0.2922 1.0789 1.3609 1.1220 1.20000 1.20000 0.8 accelerated very rapidly and the air adjoined the piston obtains Mach number M = 0.20000 0.1226 1.1241 1.0789 1.1219 1.90416 0.14: Schematic of a piston pushing air in a it takes for the shock to reach the tube.5041 1.0765 1.89509 0.3075 1.0787 1.2032 1.89354 0.89509 Ty Tx Py Px 107 My 0.89461 0. 5.6: A piston is pushing air that flows in a tube with a Mach number of M = 0.0832 1.1220 1.20000 0.3020 1.3020 1.20000 0.0930 1.20000 0.0782 1.0045 1.20000 0.73971 0.0788 1.0545 1.0802 1.20000 0.20000 0.3037 1.3020 Example 5.0789 1.1443 1.0712 1. the solution is .5.3020 1.3020 1.84424 0.0790 1. The piston is Mx = 0.1967 1.0789 1.3011 1.2547 1.20000 0.1208 1.89536 0.3017 1.20000 0.1200 1.8.1182 1.89512 0.3. Assume that there is no friction and the Fanno flow model is not applicable.89510 0.2705 1.1099 1.0836 1.0789 1.1220 1.1220 1.99548 0.89595 0.0m length.3022 1.89509 0. Calculate the velocity of the shock created by the piston in the air.87903 0.89517 0.20000 0.1259 1.20000 0.3020 1.1220 1.20000 0.89509 0.3199 1.4 My = 0.20000 0.2989 1.3020 1.1220 1.89789 0.89009 0.20000 0.20000 0.20000 0.

6215 0.80000 0.2380 − 0.2248 1.81942 0.80000 0.2380 My Mx My CHAPTER 5.50000 0.6215 1.1519 2.1517 1.5000 1.80000 1.1519 1.81958 0.80000 0.4[m/sec] .80000 1.1519 1.108 Mx 1.6217 1.82716 0.2380 My 0.0034[sec] 1.4) Example 5.1519 −0.81943 0.2400 1.81829 0.4583 1.81942 Ty Tx Py Px My 0.4 × 287 × 300(1.6273 1.2378 1.80000 0.3202 1.1531 1.98860 The complete iteration is provided below.7: From the previous example (5.6207 1. i 0 1 2 3 4 5 6 Mx 1.6216 The time it takes for the shock to reach the end of the cylinder is t= length Us cx (Mx −Mx ) =√ 1 = 0.8 × 1. NORMAL SHOCK Ty Tx Py Px P0 y P0 x 0. S OLUTION The stationary difference between the two sides of the shock is: ∆U =Uy − Ux = cy Uy − cx Ux   √   1.4 × 287 × 300 0.70109 0.1519 1.5834 1.2380 1.80000 0.2381 1.5    q Ty Tx =  ∼ 124.1435 1.80000 0.10) calculate the velocity difference between initial piston velocity and final piston velocity.81940 0.

expansion waves occur with a reduction of pressure. A cylinder with two chambers connected by a diaphragm. This is the same case as in the suddenly opened valve case described previously.99785 70. it is the explicit analytical solution.1283)347.0715 + 1.318 0.66). The distance between the two pistons is 1[m]. 5. 1. 5.083 1.1283 My 0.5. Calculate the time it will take for the two shocks to collide.0 My 0.93451 Ty Tx Py Px P0y P0 x Uy cx 347. For this case the following table can easily be obtain from Potto–GDC for the left piston Mx 1.99959 40.173 0.0 My 0.047 1. The solution is given by equation (5. It is used in many cases to understand certain phenomena. while the pressure on the other side is low. Other situations can be examined and extended from these phenomena. 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.0013[sec] Usx 1 + Usx 2 (1.0715 My 0. and.0 while the velocity of the right piston is Mx 1.8) Uy cx 347. a shock is created that it travels to the low pressure chamber. At the back of the shock. On one side the pressure is high.89048 Mx 0.0 The time for the shocks to collide is t= 1[m] length = ∼ 0.15: Figure for Example (5. 1.95890 Ty Tx Py Px P0y P0 x Fig.93471 Mx 0. The high pressure . When the pressure is high enough. The temperature is known to reach several thousands degrees in a very brief period of time.15)). SHOCK TUBE Example 5.8: An engine is designed so that two pistons are moving toward each other (see Figure (5. The air between the pistons is at 1[Bar] and 300K . When the diaphragm is ruptured the gas from the high pressure section flows into the low pressure section.4 Shock Tube The shock tube is a study tool with very little practical purposes.4.

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

88) After rearranging equation (5.82) becomes Ms1 = Or expressing the velocity as Us = Ms1 c1 = c1 k − 1 k + 1 P2 + 2k 2k P1 (5.111 √ range of the speed of sound. The following equations were developed earlier and are repeated here for clarification.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 + (5.89) . The pressure ratio between the two sides of the shock is 5. Thus. The relationship between the different properties was discussed earlier and because it is a common problem.82) k − 1 k + 1 P2 + 2k 2k P1 (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. a review of the material is provided thus far. c ∼ 340 thus the whole process takes only a few milliseconds or less.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 (5.86) (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 (5.4.85) The fluid velocity in zone 2 is the same U2 = U s − U 2 = U s 1 − From the mass conservation. it follows that U2 ρ1 = Us ρ2 U2 Us (5.88) the result is U2 = c1 k P2 −1 P1 2k k+1 P2 k−1 P1 1+k (5.

91) dρ (5.92) can be integrated as follows: U3 ρ3 dU = U5 =0 ρ5 c5 ρ ρ5 k −1 2 dρ (5.96) After some rearrangement.90) it follows that dρ dU = −c = c5 ρ ρ ρ5 k−1 2 T = T5 ρ ρ5 k −1 2 (5.90) Since the process in zone 4 is isentropic. and dividing by the continuity equation the following is obtained: dU dρ =− ρ c (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 (5.95) and equation (5.92) Equation (5. d(ρU ) = 0. taking the derivative of the continuity equation. in zone 4.96) 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    (5.89) 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 (5.94) (5.112 CHAPTER 5. NORMAL SHOCK On the isentropic side. equation (5. hence equation (5.97) . 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.

5 5. SHOCK WITH REAL GASES Or in terms of the Mach number. . the following is obtained: 1+ T2 = T1 1+ k1 −1 k1 +1 k1 −1 k1 +1 P2 P1 P1 P2 (5.100) Example 5.9: 5 A shock tube with an initial pressure ratio of P P1 = 20 and an initial temperature of 300K .6 5. Find the shock velocity and temperature behind the shock if the pressure P5 ratio is P = 40? 1 S OLUTION 5.99) 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 assumptions in construction of these models are acceptable and reasonable.5.5. It is normal to have a large Mach number with a large Re number.7 Shock with Real Gases Shock in Wet Steam Normal Shock in Ducts The flow in ducts is related to boundary layer issues. the assumption of an uniform flow in the duct is closer to reality.98) Using the Rankine–Hugoniot relationship and the perfect gas model. 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. In that case. For a high Reynolds number.

The temperature should not increase above a certain value. In this system.0021[sec] 1 . The ambient temperature is 27◦ C and 1[Bar].17: Figure for Example (5. 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.0850 My Mx My 1.967 My kRTy . The time it takes for the shock to travel from the Py valve depends on the pressure ratio Px = 30 Mx 5. An engineer is required to design a cooling system for a critical electronic deexit valve vice.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. the time is t= distance 3 √ = ∼ 0. NORMAL SHOCK 5. opening valve probelm The distance between between the valve and the pipe exit is 3[m].057811 The direct calculation will be by using the “upstream” Mach number. 668 sqrt 1 . 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 0.0850. S OLUTION This problem is known as the suddenly open valve problem in which the shock choking phenomenon occurs. 4 × 287 × 300 × 5. Assume that there isn’t any resistance whatsoever in the pipe.8 More Examples for Moving Shocks Example 5.0 5. Mx = Msx = 5. After building the system.0017[sec] 5.41404 0. 5.17)). Comment on this proposal. Therefore. The conditions upstream of the valve are 30[Bar] and 27◦ C .114 CHAPTER 5.668 Ty Tx Py Px P0 y P0 x 0. 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.00 0.967 30. The time it takes for the material from the valve to reach the exit is distance 3 t= = ∼ 0.0850sqrt1.0002 [sec] to be opened). air is Fig.10: This problem was taken from the real industrial manufacturdistance ing world.10) supposed to reach the pipe exit as quickly as possible when the valve is opened (see Figure (5.

k = 1.20 1.18: The results for Example (5. S OLUTION 5. TABLES OF NORMAL SHOCKS.18).00000 1. There is also the possibility of steps increase in which every step heat released will not be enough to over heat the device.99280 0.00000 1.10) deals with a damaging of electronic product by the temperature increase.09658 1.30 My 1.00000 0. Time[Msec] To increase the pipe diameter will not change the temperature and therefore Fig.42857 1.00000 0. The plot of the mass flow rate and the velocity are given in Figure (5.12799 1.84217 0.37625 1.11: Example (5.19087 1.05 1. Try to estimate the temperature increase of the product.34161 1.87502 0. Plot the pipe exit temperature as a function of the time.25 1.5.51570 1. the heating time is reduced significantly.03284 1.4 IDEAL GAS 115 During that difference of time the material is get heated instead of cooling down because of the high temperature.78596 Ty Tx ρy ρx Py Px P0y P0x 1.81264 0.51333 1.65625 1.10 1.08398 1.10) will not change the effects of heating.99985 0. Thus. Example 5.4 Mx 1.15938 1.11958 1. 5.4 Ideal Gas Table 5.15 1.9.95313 0.99669 0.80500 1.91177 0.16908 1.00 1.00000 1. It can only increase the rate after the initial heating spike A possible solution is to have the valve very close to the pipe exit.06494 1.25504 1.97937 .99893 0. Thus.9 Tables of Normal Shocks. K = 1. The last possible requirement a programmable valve and very fast which its valve probably exceed the moving shock the valve downstream. The suggestion of the engineer to inMass Flow Rate crease the pressure will decrease the time but will increase the temperature at the exit during this critical time peVelocity riod.24500 1. this suggestion contradicts the purpose of the required manufacturing needs.98706 0.1: The shock wave table for k = 1.

87599 0.65105 0.53637 1.27907 2.65396 0.02498 .75 1.20500 3.74420 0.57143 4.83457 0.12000 2.53441 0.01863 3.00000 31.82000 3.06172 0.45833 26.40623 0.07869 5.87509 5.43496 0.42355 0.40625 0.12500 16.40216 Ty Tx ρy ρx Py Px P0y P0x 1.77614 1.12500 8.08455 3.40416 0.51792 0.03175 2.76175 0.79023 0.68410 0.91156 5.25 My 0.35379 1.59562 0.10 2.76736 0.33333 12.99311 2.49181 0.60 1.54055 0.38797 1.88231 2.80 1.64729 1.13750 2.85714 4.15625 14.74002 2.70 1.85 1.52861 0.40897 0.94732 2.40 1.28720 1.35922 2.92979 0.43811 2.1: The shock wave table for k = 1.94680 1.96974 0.15625 29.31505 3.32022 1.00000 5.42623 4.00 4.53158 1.08846 2.11525 2.14897 3.66667 2.40625 41.49901 0.59188 2.44231 0.00 2.95958 2.21190 3.50 1.26958 4.61194 3.46192 0.00958 3.00 5.4 (continue) Mx 1.95 2.15 2.23958 18.68966 1.33333 3.54014 0.35 2.66844 0.27625 6.86207 1.26451 0.75 4.80000 6.22261 1.64054 0.00500 6.44891 4.75 5.60278 1.22625 5.58618 0.67420 0.63625 2.25469 1.94484 0.85572 0.00 3.26087 4.60570 0.19772 2.75 6.45833 2.17166 0.42280 1.12500 38.45 2.13876 0.48000 5.56935 1.57735 0.25 5.32544 5.39657 2.05100 0.56148 0.25 2.65625 10.71956 0.28625 2.50 2.90138 1.50000 20.55395 0.20 2.81190 2.25 3.04236 0.04033 2.26829 5.09170 0.58329 0.45 1.50 3.02965 0.70109 0.82625 4.30 2.50 5.54706 0.81268 0.50 4.73958 6.89520 0.91319 0.40 2.51931 0.69919 4.45833 1.40625 3.72087 0.45115 0.94059 8.66894 4.21295 0.21182 5.07505 0.47519 0.67901 2.41523 0.68750 1.29878 6.56906 0.95819 0.04688 4.81322 1.90625 23.85686 1.72855 1.00 6.51568 2.81188 4.98958 35.65 1.83625 7.60792 1.69751 0.14894 5.77045 1.95122 3.11256 0.49458 1.73971 0.35 1.116 CHAPTER 5.32834 0.36906 7.60553 0.41908 0.82180 7.05 2.55 1.51299 0.62809 0.27335 3.50000 4.73625 4.75 3.90 1.03536 0.55333 6.31915 1.41189 0.52312 0.07229 4.25 4.62814 0.40642 0.56128 0.04500 4.97833 5. NORMAL SHOCK Table 5.98511 3.61333 3.61650 0.42878 0.83333 45.

86537 0.082 1.004 1.00 My 0.043 1.98812 0.0 0.710 1.00 My 0.0 0.0 0.00000 0.024 1.95387 0.95944 0.01133 0.39879 0.97647 0.012 1.00 9.0 0.99113 15.505 1.56522 5.316 1.92832 0.99997 0.0 0.20 0.50 8.012 1.49152 19.99998 0.4) Mx 1.00979 0.45833 69.196 1.04 0.99994 0.75 9.71429 49.269 1.00 8.0 0.006 1.99986 0.057 1.75 7.90 1.80077 10.33333 99.01316 0.42736 20.40038 0.037 1.0 0.99790 0.75 8.344 1.16229 11.5.040 1.95231 0.780 3.99973 0.96506 0.69946 0.75 10.049 1.323 1.50000 0.46939 11.60 0.39289 0.53890 5.38860 0.00387 0.73958 116.442 1.05 0.89128 0.118 1.17678 14.008 1.09 0.0 0.89918 0.2: Table for a Reflective Shock from a suddenly closed end (k=1.84463 0.62425 Mx 0.10 0.62095 13.82978 16.80251 0.68504 5.08 0.25 7.15625 94.38758 Ty Tx ρy ρx Py Px P0y P0x 9.97074 0.087 1.024 1.38672 14.47883 5.57997 18.00304 Table 5.99317 0.44444 5.766 My 0.69273 17.016 1.40 0.38750 5.12500 52.82755 .018 1.99991 0.64673 0.66874 5.00739 0.70 0.70019 5.00000 1.102 1.036 1.39201 0.65116 5.80 0.12500 110.25 9.0 0.423 1.0 0.502 1.4 (continue) 117 Mx 6.473 1.50 6.014 1.00437 0.00849 0.39385 0.112 3.00645 0.99999 0.58939 5.12500 89.87948 12.072 1.195 2.39121 0.589 1. K = 1.00 7.61165 5.0 0.98227 0.0 0.15643 9.506 1.02115 0.50000 79.39491 0.30 0.39607 0.0 0.0 Ty Tx Py Px P0y P0 x 1.73029 0.65625 105.00343 0.020 1.50 7.133 1.062 1.07 0.94836 0.03 0.032 1.51020 5.67171 0.01798 0.15625 65.38918 0.38807 0.39048 0.02 0.00496 0.0 0.0 0.676 1.043 1.06 0.00000 61.171 1.36508 5.0 0.99403 0.0 0.40667 5.268 1.0 0.63218 5.4 IDEAL GAS Table 5.1: The shock wave table for k = 1.028 1.98446 0.9.50 0.01 0.01535 0.90625 74.028 1.941 2.98958 57.38980 0.127 1.030 1.99980 0.055 1. TABLES OF NORMAL SHOCKS.219 1.381 1.25 8.475 2.50 9.149 1.126 1.94291 0.00565 0.23958 84.97099 0.76452 0.39736 0.

0 0.00 6.99985 0.727 1.78840 0.0 0.21 29.0 0.50814 0.048 2.99181 0.07 0.0 0.810 1.441 2.495 1.49912 0.00350 0.033 1.868 1.057 1.410 3.390 1.38608 0.058 1.02 0.552 2.41523 0.066 8.54114 0.0 0.087 2.50 0.90 2.30 1.4E +2 1.990 2.98807 0.576 1.99994 0.037 1.791 2.83607 0.01 0.004 1.715 6.000 6.43894 0.69834 0.0 0.99990 0.50 1.99996 0.295 1.859 5.118 CHAPTER 5.050 1.162 7.152 1.0 Ty Tx Py Px P0 y P0 x 1.95315 0.97629 0.120 1.05 0.0 0.14 62.94180 0.47855 0.008 1.00 8.51808 0.800 8.248 1.331 1.89 12.74403 0.397 1.031 1.00 4.97045 0.56935 0.0 0.00 9.029 1.20 1.00 5.952 2.043 1.0 0.25 24.897 1.006 1.70283 0.762 3.38457 Mx 1.018 1.0 0.58578 0.99998 0.613 My 0.40 1.189 2.0 0.012 1.0 0. NORMAL SHOCK Table 5.734 17.431 8.862 4.96465 0.016 1.89039 .0 0.66462 Mx 0.012 1.60 1.70 1.280 4.09 0.40213 0.55453 0.00 44.3: Table for shock propagating from suddenly opened valve (k=1.74316 0.649 1.827 7.78652 0.00 3.858 1.08 My 0.20 0.39116 0.98216 0.133 1.104 1.00 7.7E +2 0.47 1.703 10.026517 0.0 0.60 0.087 2.346 2.0 0.05 19.30 0.024 1.188 1.56312 0.0 0.0 My 0.088 1.0 0.99402 0.831 5.62 84.4) Mx 1.012492 0.80 1.063 1.33 3.0 0.04 0.70 Ty Tx Py Px P0 y P0 x 1.041 1.01 29.00000 1.99999 0.98019 0.0 My 0.94746 0.88717 0.256 6.99763 0.086 1.95888 0.037 1.40 0.136 1.0 0.43921 0.39566 0.0 0.00000 0.00639 0.297 2.0 0.014 1.245 2.146 2.024 1.0 0.0 0.206 5.0 0.93133 0.2: Table for Reflective Shock from suddenly closed valve (end) (k=1.99979 0.4)(continue) Mx 1.020 1.0 0.10 0.0 0.00 10.0 0.0 0.317 1.134 1.51996 0.028 1.10 1.52904 0.728 5.60401 0.06 0.08 0.325 11.99971 0.0 0.448 2.65290 0.38817 0.0 0.517 9.40284 0.336 8.96069 0.0 0.15637 0.073 1.03 0.60761 0.656 2.00204 0.41 15.210 1.00125 Table 5.49092 0.541 1.044 1.061716 0.0 0.1E +2 1.

0 0.72 80.00101 0.00000 0.0 0.98290 0.0 0.840 7.7E +2 8.040812 0.604 1.0 0.00175 0.18E −5 0.850 1.9E + 3 6.600 1.76940 0.00000 1.97726 0.74 66.003 1.888 1.536 4.46599 0.37822 0.9E + 2 2.889 1.59699 0.0 0.99998 0.009 1.0 0.491 1.38974 0.919 2.889 1.4: Table for shock propagating from a suddenly opened valve (k=1.020 0.5E + 3 8.889 1.18 73.2E +2 9.37817 0.931 14.015 1.9.8E + 2 7.0 0.0 0.450 2.2E +2 8.49586 0.38248 0.0 0.9E + 3 7.658 4.0 0.551 8.38096 0.84 1.0 0.9E +2 1.99 25.37821 0.57 31.050 Ty Tx Py Px P0y P0 x 1.013 1.368 5.0 0.067 1.0 0.293 8.37814 0.800 1.006 1.0 0.18575 0.80 76.39187 0.0 0.31 62.0 0.53817 0.5.0 0.95 64.821 9.10216 0.2E + 3 0.51223 0.3E +2 7.28412 0.38402 0.09 35.62 61.7E + 2 4.054 1.40843 0.171 8.646 6.99999 0.59649 0.48823 0.00395 0.79 My 0.025 4.00544 0.98857 0.39028 0.42622 0.4E +3 3.0 0.38713 0.896 2.32 16.80 0.017 1.0 0.5E + 2 3.012 1.6E + 3 4.888 1.00721 0.38870 0.012 1.400 1.870 1.62 17.37818 0.0 0.0 0.0 0.744 1.3: Table for shock propagating from suddenly opened valve (k=1.0 0.2E +3 1.54 15.0 0.029 My 0.010 0.200 1.799 3.62 14.0 0.1E +3 1.99997 .07 90.83661 0.99427 0.0 0.387 4.823 5.37816 0.3E +2 7.0058 1.7E +2 9.889 Ty Tx Py Px P0 y P0 x 1.92 42.24 11.745 1.889 1.820 1.068 2.0 0.167 3.888 1.1E +3 1.0 0.0 0.4E + 3 4.25 12.145 2.030 0.07 18.38557 0.5E + 2 1.61 1.33 21.023 1.4E + 3 6.69 68.040 1.888 1.00 1.0 0. K = 1.44536 0.0 0.37820 0.0 My 0.9E + 3 5.68907 0.4) 119 Mx 1.2E + 2 1.880 1.000181 3.830 1.860 1.0 0.90 1.508 2.0 0.00272 0.37813 0.881 3.62923 0.840 1.100 1.0 0.700 1.53 63.37812 0.02 83.026 1.2E + 3 5.0 0.0 0.97166 Mx 0.269 2.0E + 2 1.040 0.0 0.0 0.78 80.500 1.45 22.37944 0.0 Table 5.810 1.3) Mx 1.300 1.56619 0.0 0.000497 0.975 11.37810 Mx 0.888 1.0 0.0 0.5E + 3 5.35 25.457 10. TABLES OF NORMAL SHOCKS.0 My 0.83 71.4 IDEAL GAS Table 5.

080 0.89159 0.166 4.0 0.0 0.0 0.0 0.98290 0.054 1.40226 0.519 3.888 1.15495 0.0 0.615 4.612 4.880 1.39026 Mx 0.500 1.4: Table for shock propagating from a suddenly opened valve (k=1.300 1.90 22.052809 .67546 0.523 7.100 1.64073 0.096 1.448 3.94 23.381 3.553 4.053053 0.810 1.0 0.39938 0.426 2.028 1.58 23.101 4.274 1.39160 0.093988 0.31281 0.553 1.39030 0.613 4.0 0.0 0.91 23. NORMAL SHOCK Table 5.900 1.860 1.39624 0.223 1.40418 0.482 4.083607 0.031 1.888 1.39034 0.052914 0.99981 0.275 1.707 5.39027 0.471 4.39468 0.576 4.0 0.0 My 0.616 4.800 1.073863 0.126 1.103 1.99991 0.200 0.700 1.681 1.333 1.785 1.0 0.053 4.0 0.467 2.99792 0.0 0.021 1.39033 0.75301 0.500 0.55074 0.39029 0.888 1.200 1.060 0.0 0.94415 0.097 4.489 1.39031 0.225 4.25 21.0 0.888 1.0 0.96610 0.0 0.100 4.700 0.522 6.980 2.0 0.898 3.830 1.834 3.078654 0.0 0.064766 0.93 23.0 0.367 11.099 4.706 1.749 3.0 0.40257 0.197 1.024 1.30 23.39780 0.099 4.889 1.842 3.84227 0.136 1.50100 0.800 0.959 2.0 0.0 0.400 1.349 4.052879 0.125 1.99288 0.39314 0.95 23.94959 0.600 1.0 0.46 14.018 1.096 4.40097 0.056322 0.42089 0.302 1.048 1.034 4.611 4.22904 0.570 1.663 1.0 0.99987 0.870 1.400 1.79611 0.0 0.919 2.063 1.0 0.850 1.39037 0.060 1.0 0.052949 0.66575 0.772 9.815 1.102 4.053018 0.00 1.86274 0.43882 0.888 1.614 4.0 0.478 1.0 0.191 2.592 3.400 0.0 0.47875 0.613 4.93 23.0 0.49 20.252 2.64 21.99975 0.60847 0.0 0.052984 0.595 2.44 18.177 1.0 0.80734 0.100 0.069 2.0 0.49333 0.090 0.92 23.016 4.96056 0.052844 0.088718 0.94156 0.05 20.042 1.035 1.96631 0.840 1.95 19.644 2.52495 0.0 0.820 1.669 3.053088 0.889 1.889 1.923 4.110 1.58223 0.615 4.99995 0.300 0.415 4.90734 0.95 23.060462 0.0 0.95506 0.098 4.109 4.069233 0.39035 0.452 1.359 1.202 3.96 0.070 0.0 0.286 4.081 1.3) Mx 1.32 18.889 Ty Tx Py Px P0 y P0 x 1.74136 0.236 2.098 1.600 0.45807 0.57853 0.0 0.616 My 0.120 CHAPTER 5.71284 0.140 1.91 23.997 3.

39025 Mx 0.104 23.052775 .617 My 0.0 My 1.97 0.5.3) 121 Mx 4. K = 1.9.889 Ty Tx Py Px P0y P0 x 4.4: Table for shock propagating from a suddenly opened valve (k=1. TABLES OF NORMAL SHOCKS.4 IDEAL GAS Table 5.

122 CHAPTER 5. NORMAL SHOCK .

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

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. Nevertheless. 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. 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. If the back pressure. once the flow becomes supersonic. Only in one point where P B = Pb continuous pressure exist. In comparison of nozzle performance for rocket and aviation. Calculate the back pressure and the temperature of the flow. Therefore.1: A large tank with compressed air is attached into a converging–diverging nozzle at pressure 4[Bar] and temperature of 35[◦ C]. The first example is for academic reasons. In conclusion. This nozzle is called an over–expanded nozzle. P B is smaller than Pb a discontinuous point (a shock) will occur. 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. In the literature. . is lower than the critical value (the only value that can achieve continuous pressure) a shock occurs outside of the nozzle. this example will provide the fundamentals to Fig. the over–expanded nozzle is worse than the under–expanded nozzle because the nozzle’s large exit area results in extra drag. Nozzle throat area is 3[cm2 ] and the exit area is 9[cm2 ]. the nozzle doesn’t provide the maximum thrust possible. only exact geometry can achieve continuous pressure flow.) Also determine the critical points for the back pressure (point “a” and point “b”). In the under–expanded case. Obviously if the back pressure. 6.+ -£. It XY Z\[ ]_^8`Aa bcFdOe troat has to be recognized that the shock wave isn’t easily visible &'(*). P B .0/21 (see Mach’s photography techexit 354687292:2.2: A nozzle with normal shock explain the usage of the tools (equations and tables) that were developed so far. (It should be noted that the temperature of the surrounding is irrelevant in this case. The shock occurs in a location where the cross section area is 6[cm2 ]. In this case.124 CHAPTER 6. The location of the shock is determined by geometry to achieve the right back pressure. Example 6. 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. NORMAL SHOCK IN VARIABLE DUCT AREAS point “a” and point “b” the flow is different from what was discussed before. On the other hand. no continuous pressure possibly can exists.

22) or from Table (4.1972 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. Mx is about 2. My .09393 0.54743 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. before the shock is known and given as well.18463 2. From the Mach number after the shock.1). Mx = 2. but can be shown that the correction is negligible for a typical dimension ratio that is over 100.9474 5. With known Mach number the new star area ratio.197 as shown table below: M 2.0000 0.8882 ∗ A Ay A 6 with this area ratio.2) or from equation (4.125 First.50877 0.54743 Ty Tx ρy ρx Py Px P0y P0 x 1. in the case of ratio of 100 the Mach number is 0. Mx 2. For example. My can be obtained.4656 0. Ay /A∗ is known and the exit area can be calculated as Ae Ay 9 Ae = × ∗ = 1.8544 2.00587 and the error is less than %0.86457 1.8882. the inside conditions are essentially the stagnation conditions (this statement is said without a proof. The star area (the throat area).18787 With this Mach number. With these values.1972 My 0.1) or equation (4. one can obtain using the isentropic relationship . A∗ . the stagnation temperature and pressure are known T0 = 308K and P0 = 4[Bar].54746.2588 × = 1. (point “y”).2) My ∼ = 0. the subsonic branch can be evaluated for the pressure and temperature ratios.11) the following Table for the isentropic relationship is obtained M 0. the Mach number.0268 Again utilizing the isentropic relationship the exit conditions can be evaluated. 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.49) or the GDC– Potto. the exit Mach number has to be determined. Thus. This Mach number can be calculated by utilizing the isentropic relationship from the large tank to the shock (point “x”).81568 1.62941 From Table (4.94345 0. Ax 6 = =2 A∗ 3 With this ratio (A/A∗ = 2) utilizing the Table (5. as Ae A∗ = 1.1972 the Mach number. It has to be realized that for a large tank. From equation (5.2588 0.

6728[Bar] .99226 0.8882 0.97912 0. NORMAL SHOCK IN VARIABLE DUCT AREAS T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.9195 0.19745 2.34[Bar] The exit temperature is Texit = Texit T0 1 0. Pexit = Pexit P0 P0 Py Py Px Px P0 P0 =0.854 × 0. is A/A∗ = 3 In the subsonic branch (either using equation or the isentropic Table or GDC-Potto as M 0. the exit conditions can be calculated.41820 0.11310 3.9K × 1.92882 1.92882 × ∼ =2. Yet we have to pay attention that there two possible back pressures that can “achieve” it or target. The area ratio for both cases.41820 × 4 ∼ =1.94862 1.98077 0.81568 × 5. In that case we don’t have to go through that shock transition.97[Bar] For the supersonic sonic branch Pexit = Pexit P0 P0 = 0.126 M 0.97318 0.951 Ty Tx Tx T0 T0 =0.466 × 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.04730 2.6374 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.7538 Since the stagnation pressure is constant as well the stagnation temperature.32651 CHAPTER 6.99226 × 4 ∼ =3.0000 0.0000 3.094 × 4 T0 Ty 1 0.98133 × ∼ =299.14190 Pexit = Pexit P0 P0 = 0.

For more practical example1 from industrial application point of view. In fact in many industrial applications.13[kg/sec] 287 × 256.7 = 321.1) 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. 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. these kind situations exist.833 × 308 = 256. While the pressure to achieve full supersonic flow through the nozzle the pressure has to be below the 42% the original value. . Example 6. Thus. over 50% of the range of pressure a shock occores some where in the nozzle.113[Bar] And the mass flow rate reads m ˙ = 4105 3 × 10−4 × 321.2: In the data from the above example (6.127 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.5283 × 4 = 2.7 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).12 = 0.7K 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= √ 1.12[m/sec] P∗ P0 P0 = 0. In these applications a small pressure difference can produce a shock wave and a chock flow.4 × 287 × 256.

Mathematically.3709 My 0. the exit pressure is between point “a” and point “b”. More importantly the pressure ratio exit is known. In the first method.90500 1.3914 0. utilizing the isentropic Table (4. The solution process starts at the nozzle’s exit and progress to the entrance.128 CHAPTER 6. it requires some iterations by “smart” guessing the different shock locations.55250 With the information of Mach number (either Mx or My ) the area where the shock (location) occurs can be found.113[Bar]. Pexit Aexit 2×9 Pexit Aexit = = = 1. NORMAL SHOCK IN VARIABLE DUCT AREAS S OLUTION The solution procedure is similar to what was shown in previous Example (6.38034 0.1755 6.93118 1.72)). First.5000 0. 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 × = 0.0128 3.97188 0. there are two main possible ways to obtain the solution.905 4 Looking up in the Table (4.1).5525 0. The area (location) that the previous example did not “produce” the “right” solution (the exit pressure was 2. is known.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. The conditions in the tank are again the stagnation conditions. the computer can be use to carry this calculations in a sophisticate way.2) or utilizing the GDC-Potto provides Mx 2. In fact.75158 With these values the relationship between the stagnation pressures of the shock are obtainable e.52628 Ty Tx ρy ρx Py Px P0 y P0 x 2. . the needed pressure is only 2[Bar] which means that the next guess for the shock location should be with a larger area2 . see also equation (4.g. 2 Of course. Here. My .2). the exit Mach number. the previous example information used and expanded. It follows that there must exist a shock in the nozzle. Utilizing the Table (4. 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.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. Thus.6575 0. The exit total pressure can be obtained (if needed).

9 to 0.15205 2.99. NOZZLE EFFICIENCY M 2.2 Diffuser Efficiency .3) (6. 6.47076 0.07158 0.0188[cm2] A∗ Note.3396 × 3 ∼ = 7.5) 6.3396 0. as “simple” check this value is larger than the value in the previous example.3709 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 129 0.1 Nozzle efficiency Obviously nozzles are not perfectly efficient and there are several ways to define the nozzleefficiency.2) The efficiency can be defined as η= The typical efficiency of nozzle is ranged between 0.4) 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.16747 Approaching the shock location from the upstream (entrance) yields A= A ∗ A = 2.1) where hexit s is the enthalpy if the flow was isentropic. In the literature some define also velocity coefficient as the ratio of the actual velocity to the ideal velocity. Cd = m ˙ actual m ˙ ideal (6. The total energy that can be converted is during isentropic process is E = h0 − hexit s (6. Vc Vc = √ η= (Uactual )2 (Uideal ) 2 h0 − hexit (Uactual )2 = 2 h0 − hexit s (Uideal ) (6. The actual energy that was used is E = h0 − hexit (6.1. 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.

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. k = 1. Assume that a shock Fig.11528 0.02[m2 ].6) can be converted to η= 2Cp (T3 − T1 ) U1 2 (6.7) 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. 6. η= 2(h3 − h1 ) h3 − h 1 = h01 − h1 U1 2 (6. 6.6) P02 P2 01 02 2 P1 1 For perfect gas equation (6.8) Example 6. S OLUTION The condition at M = 3 is summarized in following table M 3.3: Description to clarify the definition of diffuser efficiency And further expanding equation (6.65326 A×P A ∗ ×P 0 F F∗ .3) 0.entropy Fig. 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.0000 T T0 ρ ρ0 A A P P0 uous region (and also for example (6. NORMAL SHOCK IN VARIABLE DUCT AREAS P01 h The efficiency of the diffuser is defined as the ratio of the enthalpy change that occurred between the entrance to exit stagnation pressure to the kinetic energy.130 CHAPTER 6.7[Bar] and temperature of 250K .7) s.0 and prescapacitor sure of 0. The cross section in area between the nuzzle Compressor and diffuser is 0.4)) the required condition at point 3 are: M = 3.4: Schematic of a supersonic tunnel in a continoccurs in the test section.07623 4.35714 0.02722 0.4 can be assumed.2346 0.

37554 Ty Tx ρy ρx Py Px P0y P0 x 5.4789 6.02172 The calculation of the temperature and pressure ratio also can be obtain by the same manner. P0 A∗ is constant (constant mass flow). In this case Potto–GDC provides the following table Mx 5.5346 My Mx My 1.0000 My 0.37554 0.021717 The calculations were carried as following: First calculate the Mx as M x = Us / (k ∗ 287.0143[m3] P0 d 0.3333 0.6790 3.8571 10.479 34. First the stagnation behind the shock will be Mx 3. Calculate the conditions after the shock.50 0.09668 .6.4968 0.989 Ty Tx Py Px P0 y P0 x 0.0 5.5346 My 0. temperature and pressure “upstream” known.32834 Example 6. pressure of 2[Bar] and temperature of 350K .2963 34.4: A shock is moving at 200 [m/sec] in pipe with gas with k = 1.0047[m2] A 131 In this case. DIFFUSER EFFICIENCY The nozzle area can be calculated by A∗ n = A A = 0.47519 Ty Tx ρy ρx Py Px P0y P0 x 2.3.02/4.32834 A∗ d = P0 n ∗ 1 A n∼ 0. The “downstream” shock number is Msy = Us k ∗ 287. ∗ Tx ∗ Ty Tx ∼ 2.0047 ∼ 0. S OLUTION This is a case of completely and suddenly open valve with the shock velocity. For example Potto-GDC (this code was produce by the program) Mx 5.2. ∗ Tx ) Then calculate the My by using Potto-GDC or utilize the Tables.2346 = 0.

6250 2.132 CHAPTER 6. the “upstream” Mach is given.75593 Ty Tx ρy ρx Py Px P0 y P0 x 1. what will be the piston velocity? S OLUTION This is an open valve case in which the pressure ratio is given. NORMAL SHOCK IN VARIABLE DUCT AREAS Finally utilizing the equation to calculate the following My = Msy − My = 2. Mx is Mx = .2308 1.41087 ∼ 1.002[m3 ].989 Example 6.25[m/sec] ρA PA 200000 × 0.0000 0.15 cx 1.96697 The temperature ratio and the Mach numbers for the velocity of the air (and the piston) can be calculated.16[m/sec] Example 6.5: An inventor interested in a design of tube and piston so that the pressure is doubled in the cylinder when the piston is moving suddenly.24[◦C] Ty = T x Tx The velocity of the piston is then √ Uy = My ∗ cy = 0.6: A flow of gas is brought into a sudden stop. The temperature at “downstream” (close to the piston) is Ty = 300 × 1. Calculate the conditions behind the shock wave.24 ∼ 291. Thus.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 1.3628 My 0.09668 − 0.75593 ∗ 1. Ux = m ˙ mRT ˙ 2 × 287 × 350 = = ∼ 502. The mass flow rate of the gas is 2 [kg/sec] and cross section A = 0.091 (Butane?). If the steady state is achieved.002 502. S OLUTION This is the case of a closed valve in which mass flow rate with the area given.4 ∗ 287 ∗ 369.2308 = 369.25 Ux =√ ∼ 2. The imaginary gas conditions are temperature is 350K and pressure is 2[Bar] and R = 143[j/kg K ] and k = 1. The propagating piston is assumed to move into media with temperature of 300K and atmospheric pressure of 1[Bar].

589 2.0 0.914 9.47886 0.589 9.796 9. The iteration of the procedure are i 0 1 2 3 4 5 Mx 3.0 0.0 2.609 2.589 11.1500 2.796 .47996 0.922 My 0.922 2.4096 9.9222 My Mx My 0.0 0.47996 Ty Tx Py Px My 0.6.796 9.589 2.47996 2.0 Ty Tx Py Px P0 y P0 x 133 0. DIFFUSER EFFICIENCY With this value for the Mach number Potto-GDC provides Mx 2.0 0.47988 0.0 0.47995 0.923 2.940 2.46689 0.796 0.35101 This table was obtained by using the procedure described in this book.590 2.8598 2.922 2.1500 2.2.804 9.

134 CHAPTER 6. NORMAL SHOCK IN VARIABLE DUCT AREAS .

You can help especially if you have photos showing these effects. The simplified models that suggests them–self are: friction and adiabatic. seem the most applicable.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 . In cases where more refined calculations have to carried the gravity or other forces have to be taken into account. The effects of the gravity of the nozzle flow in two models isentropic and isothermal is analyzed here. to this chapter. These models can served as limiting cases for more realistic flow. isothermal. Moreover. Flow in a vertical or horizontal nozzle are different because the gravity. please do so. In the previous chapters a simple model describing the flow in nozzle was explained. 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. Please ignore. The isothermal nozzle model is suitable in cases where the flow is relatively slow (small Eckert numbers) while as the isentropic model is more suitable for large Eckert numbers. The two models produces slightly different equations.

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

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

138 CHAPTER 8. Dimensionless Representation In this section the equations are transformed into the dimensionless form and presented as such.7) (8.).3) Again it is assumed that the gas is a perfect gas and therefore. equation of state reads dP dρ = P ρ It is convenient to define a hydraulic diameter DH = 4 × Cross Section Area wetted perimeter (8.2) where A is the cross section area (it doesn’t have to be a perfect circle.5) 8. The shear stress is the force per area that acts on the fluid by the tube wall.2 it seems obvious to write this equation perhaps to consult with others. The Awetted area is the area that shear stress acts on.4) (8.1 The Control Volume Analysis/Governing equations Figure (8. (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 (8.1) describes the flow of gas from the left to the right. a close enough shape is sufficient. First it must be recalled that the temperature is constant and therefore.6) . ISOTHERMAL FLOW 8.1) (8. Hence. 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 heat transfer up stream (or down stream) is assumed to be negligible. equation of state is expressed as the following: P = ρRT (8.

.2. 2 It where are the stagnation equations? put them in a table put explanation how to derive this expression.15) (8. thus “Fanning f” values are only 1/4th of “Darcy f” values.2) yields 4dx −dP − f DH 1 2 ρU 2 m ˙ A = ρU dU (8.12) (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) into momentum equation (8. this factor is a dimensionless friction factor sometimes referred to as the friction coefficient as f= τw 1 2 2 ρU (8.11) yields d(U 2 ) = kR M 2 dT + T d(M 2 ) (8. instead of diameter friction equation.8) Substituting equation (8.8.9) 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 (8.10) Now the pressure. U 2 = kRT M 2 Differentiation of equation (8. DIMENSIONLESS REPRESENTATION 139 Now.14) Differentiation of the isotropic (stagnation) relationship of the pressure (4. P as a function of the Mach number has to substitute along with velocity. U .11) d(M 2 ) d(U 2 ) dT = − M2 U2 T (8.11) yields should be noted that Fanning factor based on hydraulic radius.

state are described above. fL The distance friction. These different solutions are supersonic and subsonic solution.20) to become dU dP =− P U 3 Assuming (8.15) when combined with equation (8. ISOTHERMAL FLOW 1 2 2 kM k−1 + 2 M2 dM 2 M2 (8.14)) leads dM 2 2dU = 2 M U (8. is selected as the choice for the independent fL . energy. P.22) the upstream variables are known. Thus. ρ)3 and with these four equations the solution is attainable. There are 4 unknowns (M. Substituting T for equation (8.18) yields M2 M2 (8.19) By utilizing the momentum equation it is possible to obtain a relation between the pressure and density.9) yields: dT0 = dT 1+ k−1 2 M 2 +T k−1 dM 2 2 (8.17) yields: T0 1 k−1 2 2 d M 1 2 + k− 2 M dT0 = Rearranging equation (8.18) dT0 (k − 1) M 2 dM 2 = 1 T0 M2 2 1 + k− 2 (8. T. . 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). 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. The variable.140 dP0 dP = + P0 P 1 CHAPTER 8. 4D .16) Differentiation of equation (4.17) Notice that dT0 = 0 in an isothermal flow.21) (8. One can notice that there are two possible solutions (because of the square power). continuity (mass).20) The four equations momentum. the equations need to be obtained as a function of 4D density is eliminated from equation (8.

29) .2.26) 4f dx = D 1/k M2 1 − kM 2 dM 2 kM 2 (8. DIMENSIONLESS REPRESENTATION After substituting the velocity (8.23) 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. Integration of equation (8. When the value M = √ k 1 Mach number larger than M > √ it makes the right hand side of the integrate k negative.27) It can be noticed that at the entrance (x = 0) for which M = Mx=0 (the initial fL is positive for any x. thus. Thus.26) that when M → √ the value k of right hand side approaches infinity (∞). To obtain this restriction 1 = kM 2 .27) yields 4f Lmax D = 1 − kM 2 + ln kM 2 kM 2 (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).10).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 .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. The term 4D the other side has to be positive as well.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. Heat transfer has a limited value therefore the model of the flow must be changed.25) The variables in equation (8. Since the stagnation temperature (T0 ) has a finite value which means that dT0 → ∞. one can obtain that 1/k M2 = ∗2 U2 U (8.22) into equation (8. By denoting the superscript symbol ∗ for the choking condition. A more appropriate model is an adiabatic flow model yet it can serve as a bounding boundary (or limit). one can obtain − dP 4f dx − P DH kP M 2 2 = kP M 2 dP P 141 (8. the term on velocity in the tube isn’t zero). 1 Further it can be noticed from equation (8. 1 is the limiting case from a mathematical point of view.23) Equation (8.

It is very useful to obtain the relationship between the entrance and the exit condition for this case.37) .142 CHAPTER 8. Denote 1 and 2 as the conditions at the inlet and exit respectably.29) is transfered into √ U kM = U∗ Utilizing the continuity equation provides ρU = ρ∗ U ∗ .36) is reduced into the following approximation ∼0 4f L D = 2 ln M1 − 1 − 1 − kM2 2 kM2 2 (8.32) 1+ 1 k−1 2 2 M k −1 + 2k k k−1 (8.35) are presented on in Figure (8.33) equation (8.30)-(8.36) For the case that M1 >> M2 and M1 → 1 equation (8.2) 8.32) and rearranging yields 2k 3k − 1 k k−1 P0 1 √ ∗ = P0 k 1+ k−1 2 M 2 k k −1 1 M (8.34) 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) ρ 1 =√ ρ∗ kM (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. ISOTHERMAL FLOW Rearranging equation (8.31) (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. From equation (8.3 The Entrance Limitation of Supersonic Branch Situations where the conditions at the tube exit have not arrived at the critical conditions are discussed here.35) These equations (8.

1 0. In these cases.8. With this conflicting mechanism obviously the flow is closer to the Fanno flow model.5. 8. Yet this model provides the directions of the heat transfer effects on the flow. In reality. even small 4D > 2 results in a Mach number which is larger than 4. 8.38) This relationship shows the maximum limit that Mach number can approach when fL the heat transfer is extraordinarily fast.01 1 Mach number 10 Fig. COMPARISON WITH INCOMPRESSIBLE FLOW 143 Isothermal Flow P/P .4. 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. 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 0.2: Description of the pressure. This velocity requires a large entrance length to achieve good heat transfer.1 Fri Feb 18 17:23:43 2005 0. one should expect that the isothermal flow should have similar characteristics as .4 Comparison with Incompressible Flow The Mach number of the flow in some instances is relatively small.

P1 − P2 /P1 .44) 4f L D = 1 2χ − χ2 − ln kM1 2 1 1−χ (8.39) one can obtained that M2 = M 1 P1 P2 (8.144 CHAPTER 8.39) fL represent the ratio of Now note that for incompressible flow U1 = U2 = U and 4D the traditional h12 .40) yields 4f L D = 1 kM1 2 1− P2 P1 2 − ln P2 P1 2 (8.40) Substituting this expression into (8. the pressure loss is expressed as follows P1 − P 2 = 4f L D U2 2 (8. a relationship between M2 and M1 and pressures has to be derived. For incompressible flow. ISOTHERMAL FLOW incompressible flow. From equation (8.43) 4f L D = 1 2 1 − (1 − χ) − ln kM1 2 1 1−χ 2 2 (8. denote χ= P1 − P 2 P1 (8. Expanding the solution for small pressure ratio drop. by some mathematics.46) .41) can be transformed into 4f L D 1 = kM1 2 1− P2 − P 1 + P 1 P1 2 − ln 1 P2 P1 2 (8.41) Because f is always positive there is only one solution to the above equation even though M2.45) 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. To obtain a similar expression for isothermal flow.42) Now equation (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 .45) now can be approximated as 4f L D = 1 (2χ − χ2 ) − 2χ − χ2 + f χ3 kM1 2 (8.52). 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 (8.5. 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.8.49) and further rearrangement yields 4f L D = χ 2(1 − kM1 2 ) − 1 + kM1 2 χ + f χ3 kM1 2 (8. The current explanation is correct but based on hands waving and definitely does not satisfy the author.5 Supersonic Branch Apparently. A dimensionless analysis4 demonstrates that all the common materials that the author is familiar which creates a large error in the fundamental 4 This dimensional analysis is a bit tricky.50) in cases that χ is small 4f L D ≈ χ 2(1 − kM1 2 ) − 1 + kM1 2 χ kM1 2 (8.51) 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 (8.47) similarly it can be shown that f (χ = 0) = 1 equation (8. and is based on estimates. 8. .48) rearranging equation (8.48) yields 4f L D = χ (2 − χ) − kM1 2 (2 − χ) + f χ3 kM1 2 (8. Equation (8.

Now.71.0000185 kg/m-sec at 300K and 0. For this illustration. For example.005 results in Lmax 10 = = 500 D 4 × 0. the feeding diverging nozzle somewhat reduces the required entry length (as opposed to converging feeding). over hundred of meters per fL = 10 the required entry Mach second. So. . even with low temperature like 200K the speed of sound of air is 283[m/sec]. a gas flows in a tube with 4D number is over 200. the speed of sound is a function of temperature. The flow speed at the entrance is very large. However. 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 . this model can provide a better understanding to the trends and deviations of the Fanno flow model.53) The typical values of the the kinetic viscosity. The thermal entry length is in the order of the hydrodynamic entry length (look at the Prandtl number. 8. ISOTHERMAL FLOW assumption of the model and the model breaks. Almost all the perfect gas model substances dealt with in this book. this topic will be covered in the next version because it provide some insight and boundary to the Fanno Flow model. the hydraulic entry length is very large as will be shown below.06 UD ν (8.0). 0 Now with limitation.6 5 see Figures and Tables Kays and Crawford “Convective Heat Transfer” (equation 12-12). are 0.7 D On the other hand a typical value of friction coefficient f = 0. for most gas cases the speed of sound is about 300[m/sec]. even fL for relatively small tubes with 4D = 10 the inlet speed is over 56 [km/sec]. Nevertheless. For example. Remember from Fluid Dynamic book Lentrance = 0. if Mach number is changing from 10 to 1 the kinetic energy change is T0 about T ∗ = 18. In the supersonic flow.).0000130034 kg/m-sec at 200K.37 which means that the maximum amount of energy is insufficient.146 CHAPTER 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 requires that the entrance length to be larger than the actual length of the tub for air. (0. value for the common gases. Combine this information with our case of 4f L D = 10 Lentrance = 250746268. ν .

12 139. etc.3806 2.87612 0.29895 0.00205 0.043 1.1129 1.83637 0.87563 0. 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). friction coefficient.09000 0.84515 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 785.001 1.000 17.056 1.16552 0.5366 1.50000 0.1129 1.4784 1.99741 1.011 1.005 1.35000 0.1718 21.2074 1.2074 1.8650 3.9682 1.50207 0.3524 1.87675 0.03095 0.08085 0.82758 0.0859 12.7040 66. The driving force questions deal with what should be the pressure difference to obtain certain flow rate.91875 0.10000 0. the “engineering” or practical questions can be divided into driving force (pressure difference).3002 1.55000 0.0495 1.2258 3. In this model no questions about shock (should) exist7 .4515 4.06000 0.8791 1.011 1.00371 0.4147 2.2565 1.8172 2.7 Isothermal Flow Examples There can be several kinds of questions aside from the proof questions6 Generally. .4086 1.87516 0.6500 5.91044 0.1718 21.8781 1.90300 0. 7 Those who are mathematically inclined can include these kinds of questions but there are no real world applications to isothermal model with shock.99485 0.03000 0.99232 0.45000 0.65000 0. These questions or examples will appear in the later versions. Here is an example.8172 2.87586 0.5644 9.96075 0.032 1.05000 0.2553 10.81879 0.6470 1.000220 0.07000 0.1: The Isothermal Flow basic parameters 147 M 0.056 1.2258 3.0255 1.6651 13. and mass flow rate questions.043 1.6903 1.0 28.003 1.93800 0.00626 0.88200 0.7.1289 16.6109 8.4515 4.009 1.9031 14.0677 1.97 439.89075 0.).6903 1.70000 0.89 82.33 279.032 1.1289 16.0859 12.20000 0.7230 2.3334 2.000 8.75000 0.79 105.3002 1.81000 0.8493 7. friction factor.94894 0.89644 0.000896 0.06 192.88594 0.4147 2.9181 5.25000 0.5644 9.80000 0.80732 0. ISOTHERMAL FLOW EXAMPLES Table 8.1269 1.5920 6.1269 1.92794 0.8.0823 1.9925 4.0736 10.000 28.5366 1.1599 13.2668 0.1259 1.98982 0.9747 7.000 0.9031 14.4086 1.3906 8.87544 0.97344 0.87642 0.2126 1.021 1.0736 10.3906 8.30000 0.04000 0.98700 0.1827 1.60000 0.8781 1.3806 2. resistance (diameter.08000 0.007 1.87528 0.40000 0.021 1.

19 RT 290 × 300 m3 π × (0.04331 400. the density is a √ the entrance Mach number. The exit Mach number is not necessarily 1/ k i. the flow is not choked.62 ∼ = 16.31 × 290 × 300 ∼ = 14. 017.62 sec ρ= P 2.00 20. It is more appropriate to assume an isothermal model hence our model is appropriate.005 and the exit pressure is 1[bar]. Hint: calculate the maximum flow rate and then check if this request is reasonable.25)2 kg × 14.0055000 = 400 0.0 0.148 CHAPTER 8.89446 The maximum flow rate (the limiting case) can be calculated by utilizing the above table. the flow rate is really some what different.0 m ˙ = kg sec P1 P1 P1 kU kU A =√ = AkM1 A√ RT k c kRT kRT . What should be the pump pressure so that a flow rate of 2 [kg/sec] will be achieved? Assume that friction factor f = 0. The specific J heat for the gas. ISOTHERMAL FLOW Example 8. ρ.04331 × √ m . R = 290 Kkg . In incompressible flow. The velocity of the gas at the entrance U = cM = 0. check whether flow is choked (or even possible).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 . 4D 4f L D = 4 × 0.9 4 sec The maximum flow rate then reads m ˙ = ρAU = 23. It is note worthy to mention that since the isothermal model breaks around the choking point. 450 ∼ kg = = 23. To solve this problem the flow rate has to be calculated as m ˙ = ρAU = 2. surroundings temperature 27◦ C.5921 0. The density reads 1.25 Utilizing Table (8. fL Calculating the resistance.1: A tube of 0.19 × The maximum flow rate is larger then the requested mass rate hence the flow is not choked.e. k = 1. S OLUTION If the flow was incompressible then for known density. First.1743 12.1) or the program provides M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 0.31.

calculate the required diameter of tube.04014 466.5 [km] (500 [m]).89567 The entrance Mach number is obtained by 4f L D 1 = 66.7. 31 100000 × π×(0 4 P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 66.4826 = 2. the maximum pressure allowed for the gas is only 10[bar].31 P = 21.76780 × 8.68 P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ M 4fL D 0. the maximum M1 that can be obtained when the M2 is at its maximum and back pressure is at the atmospheric pressure.0 0. Thus.4826 5. A flow rate of 0. Assume that the flow is isothermal and k=1.59 mc ˙ = = 0. the minimum diameter will be obtained when the flow is choked.5 [in]8 .1) or using the provided program yields 8 It is unfortunate.8. ISOTHERMAL FLOW EXAMPLES Now combining with equation (8. .6779 8. with the value of M1 either by utilizing Table (8.2 [kg/sec] is required.7678 13.40) yields m ˙ = M2 = M2 P2 Ak c 149 From Table (8.89442 Note that tables in this example are for k = 1.0845 M1 = M 2 P1 k 10 Now. but it seems that this standard will be around in USA for some time.).02 (A relative smooth tube of cast iron.5844 0. You can assume that the soundings temperature to be 27◦ C.1) or by utilizing the program M 0.566[bar] Example 8.6779 + 400 ∼ = 466.4. Note that tubes are provided in increments of 0.0 0.68 The pressure should be 21. S OLUTION At first. The friction coefficient for the tube can be assumed as 0.103 .3249 0. Mmax 1 1 P2 = √ = 0. Due to safety concerns.10300 4fL D P P∗ 2 × 337.25)2 P2 Ak × 1 .2: A flow of gas was considered for a distance of 0.

9110 6.4 ≈ 50. pressure and fL is etc) have to be taken at the same point.08528 × 347.4 × 287 × 300 ∼ = 347.0 0.2 sec and the density is ρ= 1.4318[m].431 the value of minimum diameter.2 the mass flow rate requirement is satisfied.0 0. The only point that must be emphasized is that all properties (like Mach number.02 × 500 94. The speed of sound at the entrance is √ √ m c = kRT = 1. 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.08527 4fL D 92.43182 4 × 0.2 ∼ = 29.6400 9. ISOTHERMAL FLOW P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 94. The new 4D 4f L D = 4 × 0.6 sec .2424 0.43 0.64 ρ ρ∗ T0 T0 ∗ M 0.150 M 0.4310 10. 000.61 RT 287 × 300 m3 The velocity at the entrance should be m U = M ∗ c = 0.4318 P P∗ P0 P0 ∗ 92.2991 0.02 × 500 0. D= 4f L 4f Lmax D 4 × 0.0018 6.0853 × √ 287 × 300 √ 1.3[kg/sec] Since 50.87625 = 94. It should be noted that P should be replaced by P0 in the calculations. 000 kg P = = 11.3 ≥ 0.5 increments and the next size is 17[in] or 0.68[in] However.87627 To check whether the flow rate satisfies the requirement m ˙ = 106 × π ×0. the pipes are provided only in 0.42359[m] = 16.08450 With 4f Lmax D 4fL D P P∗ CHAPTER 8.

21[bar] 20. 1. at first it will be assumed that the whole length is the maximum length.52828 = 1.4.6 × 11. Assume that the flow is isothermal. S OLUTION First.66915 From the table M1 ≈ 0. The pressure at the exit (station (b)) is 2[bar].67 The pressure at point (b) by utilizing the isentropic relationship (M = 1) pressure ratio is 0.0419 .21 × 0. The gas and the sounding temperature can be assumed to be 300 K.027 π × 29.7.3: A gas flows of from a station (a) with pressure of 20[bar] through a pipe with 0. ISOTHERMAL FLOW EXAMPLES The diameter should be D= 4m ˙ = πU ρ 4 × 0.19235 ≈ 12. The flow rate is given with the diameter of the pipe. This situation is reversed question. k=1. P2 = P0 ∗T P2 P0 ∗T = 2.19235 12. Example 8. The solution is an iterative process.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 0.01 × 4000 = 400 0.17[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.4[m] diameter and 4000 [m] length to a different station (b).52828.0419 400.and P0 ∗T ∼ = 28 12. 4f Lmax D = 4 × 0. It should be noted that the flow isn’t choked. and the average friction f=0.72021 0. the information whether the flow is choked needs to be found.61 151 Nevertheless. for the sake of the exercise the other parameters will be calculated.8.67 2. guess reasonable value of M1 and calculate 4f L D .2 ∼ = 0.01.87531 P0 P0 ∗T 20. Calculate the Mach number at the entrance to pipe and the flow rate. Therefore.

0419 4 300 × 287 RT 42.4: The flow parameters for unchoked flow M1 0.6684 0.40737 0.10000 0.32795 0. Obtain M2 from the Table ? or by using the Potto–GDC.0419 The flow rate is √ √ 2000000 1.8 Unchoked situation Table 8. Calculate the pressure.05000 0.4 P k π × D2 M= √ m ˙ = ρAM c = √ π × 0.10000 0.5708 M2 0.40754 0.22 × 0. Calculate the value of 4f L D 2 CHAPTER 8.84018 0.08971 0.10000 .32807 0.6483 0.12942 0.5914 0.83997 0. ISOTHERMAL FLOW by subtracting 4f L D 1 − 4f L D 3.83740 4f Lmax D 1 4f L D P2 P1 0. M1 0. M2 0.00000 0.152 2. there are no examples on isothermal with supersonic flow.36780 0. Compare the results of guessed pressure P2 with the actual pressure and choose new Ma number M1 accordingly.12949 0.83920 0.83889 0.36766 0.46[kg/sec] In this chapter. Now the process has been done for you and is provided in Figure (??) or in the table obtained from the provided program.10000 0.83827 0.10000 8.08978 0. P2 bear in mind that this isn’t the real pressure but based on assumption 5.5807 0.10000 0.6934 0.59338 4f Lmax D 1 4f L D P2 P1 400.10000 0.32131 400.84095 0.10000 0.05005 0. 4.16922 0.16912 0.7272 0.

8.9 0. UNCHOKED SITUATION 153 M1 isothermal flow 1 0.3 0.8 0.7 0.3: The Mach number at the entrance to a tube under isothermal flow model as a fL function 4D .5 0.2 = 0.8.2 0.10 Fri Feb 25 17:20:14 2005 Fig.8 = 0.5 = 0.6 0.4 0.8.1 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.

ISOTHERMAL FLOW .154 CHAPTER 8.

1 Introduction Consider a gas flowing through a conduit with a friction (see Figure (9.CHAPTER 9 Fanno Flow An adiabatic flow with friction is ¤ ›8œžŸ› named after Ginno Fanno a Jewish “ flow  ¢¡ž£u  direction engineer.± 9. 155 .1)). As this model raised from need to explain the steam flow in turbines. 9. °.1: Control volume of the gas flow in a constant cross section number. more general cases will be examined2 . Later. 1 Even 2 Not the friction does not convert into heat ready yet. exhaust system of an internal combustion engine. This model is the second ¥¦¨§Ÿ¥ pipe flow model described here. This model explains many industrial flow processes which includes emptying of pressured container through a relatively short tube.v. plicable to flow processes which are No heat transer very fast compared to heat transfer mechanisms with small Eckert Fig. discussed on the ideal gas model and the entry length issues. This model is apc. It is advantages to examine the simplest situation and yet without losing the core properties of the process. compressed air systems. etc. ”•!–˜—D™„š © ¦¨§ ©«ª~¬­ ¦¨§ ¯ ­ ® The main restriction for this model is that heat transfer is negligible and ².³ can be ignored 1 .

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.156 CHAPTER 9.2 Model m ˙ = ρAU = constant → ρ 1 U1 = ρ 2 U2 (9.3) Again for simplicity. 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. .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. P = ρRT P2 P1 = → ρ 1 T1 ρ 2 T2 (9. FANNO FLOW 9.4) U1 = 2cp 2 T0 2 T2 + U2 2cp 2 (9.2) (9.7) (9.8) 4 × Cross Section Area wetted perimeter (9. the perfect gas model is assumed3 .5) It is assumed that the flow can be approximated as one–dimensional.

2) and substituting equation (9.3 Non–dimensionalization of the equations Before solving the above equation a dimensionless process is applied.10) By utilizing equation (9.3.14) Utilizing the definition of the perfect gas results in M2 = ρU 2 kP (9. The infinitesimal area that shear stress is acting on is dAw = πDdx (9.11) yields Dividing equation (9. By utilizing the definition of the sound speed to produce the following identities for perfect gas M2 = U c 2 = U2 k RT P ρ (9.9) 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) .9.13) 9. s2 ≥ s 1 (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.6) yields A τw 2 m ˙ A − πD dP − πDdx f 4 1 2 ρU 2 = A ρU dU (9.12) The second law is the last equation to be utilized to determine the flow direction.11) by the cross section area.10) into momentum equation (9. A and rearranging 4f dx D 1 2 ρU 2 −dP + = ρU dU (9.

21) need to be solved.21) Equations (9.5) and dividing the results by equation of state (9.19) Derivation of the Mach identity equation (9. Thus.20) Dividing the energy equation (9.17).158 CHAPTER 9. FANNO FLOW Using the identity in equation (9.19).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 (9.4) 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. These equa- .18) The derivation of the equation of state (9. (9.17). and (9. (9.14) and substituting it into equation (9. Derivative of mass conservation ((9. (9.14) and dividing by equation (9.14) yields d(M 2 ) d(U 2 ) dT = − 2 M U2 T (9. (9.16) By further rearranging equation (9. the original limitations must be applied to the resulting equation.5) results dρ dT dP = + P ρ dT (9.2)) results in dU U dρ 1 dU 2 + =0 ρ 2 U2 (9.18).16) results in − dP 4f dx − P D kM 2 2 = kM 2 dU U (9.20).

22).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. is chosen as the independent variable since the change in the dimensionless refL .21) when eliminating dT /T results dP dρ (k − 1)M 2 dU 2 = − P ρ 2 U2 (9.27) and (9.23) The second equation for Mach number.19) and (9.28) .29) 1 dc k (k − 1)M 4 4f dx dT = =− T 2 c 2(1 − M 2 ) D (9. The only variable that is left is P (or dP/P ) which can be eliminated by utilizing equation (9.24) and results in 1 − M 2 dM 2 4f dx = 1 2 D kM 4 (1 + k− 2 M ) Rearranging equation (9.22) The term dρ ρ can be eliminated by utilizing equation (9. 4D Combining equations (9.9.20) and (9. M variable is obtained by combining equation (9.18) and substituting it into equation (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).3. The dimensionless friction.24) (9. sistance.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. 4D . fL the rest variables can be done in a similar fashion.25) results in 1 2 kM 2 1 + k− 4f dx dM 2 2 M = 2 2 M 1−M D (9.21) by eliminating dT /T . causes the change in the other variables.18) and equation (9.18) to obtain dρ 4f dx kM 2 =− 2 ρ 2 (1 − M ) D (9. Then dρ/ρ and U are eliminated by utilizing equation (9.27) and the relationship for the temperature is density is obtained by utilizing equations (9.25) (9. Explicit explanation is provided for only two variables.23) kM 2 1 + (k − 1)M 2 4f dx dP =− P 2(1 − M 2 ) D (9.

This pressure increase is what makes compressible flow so different from “conventional” flow. (dT0 = 0). FANNO FLOW (9.4 The Mechanics and Why the Flow is Choked? The trends of the properties can be examined by looking in equations (9.24) because f dx and M are positive. from equation (9.34).30) dP dT − R ln T P (9.35) (9. the pressure increases downstream. When M < 1 the pressure decreases downstream as can be seen from equation (9.160 The stagnation pressure is similarly obtained as dP0 kM 2 4f dx =− P0 2 D The second law reads ds = Cp ln CHAPTER 9. Two.32) In similar fashion the relationship between the stagnation pressure and the pressure can be substituted into the entropy equation and result in ds = Cp ln dT0 dP0 − R ln T0 P0 (9. 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. . in the supersonic branch.33) The first law requires that the stagnation temperature remains constant. For example.34) 9.24) through (9. flow with speed below the speed of sound.24) it can be observed that the critical point is when M = 1. 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. flow above speed of sound. For the same reasons. Thus the discussion will be divided into two cases: One.31) The stagnation temperature expresses as T0 = T (1 + (1 − k )/2M 2 ). M > 1.

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. T0 9.24)).25) yields f dx = L (9. However. 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. For example. dP (@M = 1) = ∞ and mathematically it is a singular point (see equation (9.5 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. Observing from equation (9. U Temperature. M Velocity.9.5. ρ Stagnation Temperature. THE WORKING EQUATIONS Why the flow is choked? 161 Here. First. the explanation is based on the equations developed earlier and there is no known explanation that is based on the physics. when the flow is supersonic. the pressure can be examined by looking at equation (9. This constrain means that because the flow cannot “crossover” M = 1 the gas has to reach to this speed. This situation is called choked flow. For example.26). the pressure decreases. The Trends The trends or whether the variables are increasing or decreasing can be observed from looking at the equation developed.37) . M = 1 at the last point. P Mach number. 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. T Density.36) A representative friction factor is defined as ¯= f 1 Lmax 0 Lmax f dx (9. On the other hand. the pressure has to be a monotonic function which means that flow cannot crosses over the point of M = 1. It demonstrates that the Mach number increases downstream when the flow is subsonic.

26) yields dP P dM 2 M2 =− 1 + (k − 1M 2 dM 2 1 2 M 2M 2 1 + k− 2 (9.36) yields k+1 2 ¯ max 1 1 − M2 k+1 4fL 2 M = + ln 1 2 D k M2 2k 1 + k− 2 M (9.39) The symbol “*” denotes the state when the flow is choked and Mach number is equal to 1.24) can eliminate term D and describe the pressure on the Mach number. Thus.30) can be solved. It is common to replace the f Equations (9. (9.39) can be integrated to yield: 1 P = ∗ P M k+1 2 k −1 2 2 M 1+ (9.29).42) U = U∗ ρ ρ∗ −1 =M 1+ k+1 2 k−1 2 2 M (9.27).28). Dividing equation (9. fL . M = 1 when P = P ∗ Equation (9.29).38) ¯ with f which is adopted in this book. (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. the pressure as written in equation (9. (9. and (9. FANNO FLOW By utilizing the mean average theorem equation (9.44) .162 CHAPTER 9.23) is represented by 4D 4f L and Mach number.24). (9.24) in equation (9.41) ρ 1 = ρ∗ M 1+ k−1 2 2 M k+1 2 (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. Now equation (9. For example.

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. D The results of these equations are plotted in Figure (9.5.9.45) And further rearranging equation (9. 9.45) provides 1 P0 ∗ = P0 M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (9.1 Fri Sep 24 13:42:37 2004 0.2: Various parameters in Fanno flow as a function of Mach number many cases shockless and therefore a relationship between two points should be .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 (9.40) and substituting it into equation (9.01 1 Mach number 10 Fig.1 0.34) yields s − s∗ = ln M 2 cp k+1 1 2 1 + k− 2 M k+1 k 2M 2 (9. THE WORKING EQUATIONS 163 Using the pressure ratio in equation (9.47) discussion about Reynolds number and dimensionless 4f L friction parameter.46) The integration of equation (9.2) The Fanno flow is in P/P .

In most times.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. 4f Lmax D = 2 4f Lmax D 1 − 4f L D (9. the temperature and the total pressure at the pipe inlet are the same as those in the reservoir. Estimate the total temperature and total pressure in the reservoir under the Fanno flow model.50) 9. S OLUTION For isentropic.49) Hence.0327 property is given only for academic purposes.05 [m] and áãâåäÑæ©ç è§é length of 10 [m].164 CHAPTER 9.1291 4 This 1. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0. The real ratio can be obtained by two star ratios as an example T2 = T1 T T ∗ M2 T T ∗ M1 (9. finding the total pressure and temperature at the pipe inlet is the solution. the “star” values are imaginary values that represent the value at choking.90000 0.Ä_ÅMÆ Ç mosphere. the total temperature at the entrance can be fL obtained by knowing the 4D .1: Û£ÜAÝ·Þß à Air flows from a reservoir and enters a uni´€µ·¶¸ ¶s¹©º »§¼ 0 ½ .1) pipe inlet is essentially isentropic.9146 1.3: Schematic of Example (9. With the Mach number and temperature known at the exit. There is no Mach meter.004 and that the flow from the reservoir up to the Fig.94 . ¾ 5 ¿ À form pipe with a diameter of 0. the flow to the pipe inlet.01451 1. The following conditions preԐÕAÖ·×sØsÙ Ú vail at the exit: P2 = 1[bar] temperature Ê È ÉAËÍÌDÎ ÏÑÐDÒsÓ T2 = 27◦ C M2 = 0. The air exits to the atÁÃÂ.6 Examples of Fanno Flow Example 9.9) the following is obtained.0934 0.0089 1. Assume that the average friction factor to be f = 0. Thus. . For given Mach number (M = 0. FANNO FLOW derived. 9.

2100 3.975 T∗ T 348K = 75◦ C Another academic question: .35886 0.0140 1. the total temperature at the exit is T ∗ |2 = T∗ T T2 = 2 165 300 = 290.014 × = 2.6.5764 0.91[Bar] 1.5922 0.12913 0.17 × 1.1699 Note that the subsonic branch is chosen.1) by interpolations or by utilizing the attached program.35886 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0. EXAMPLES OF FANNO FLOW So.004 × 10 + 0.7405 2.91484 1. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.38814 1.0327 4f L D To ”move” to the other side of the tube the 4f L D is added as 3.5[K ] 1.9.7405 0.0327 0.01451 0.21 either from Table 1 = 4f L D + 4f L D 2 = 4 × 0.05 4f L D The rest of the parameters can be obtained with the new (9.78305 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 × × 3.93840 1.915 T1 T ∗ T01 = T2 T T0 ∗ 2 T 1 T 1 1 1 =300 × × 1. The stagnation ratios has to be added for M = 0.35886 3.97489 0.

determine: (a) the mass flow rate through the system.65 [Bar]. FANNO FLOW ó©ô2õö ÷Mø ùHúMû ü ýMþ ÿ  ¢¡¤£¦¥¨§©§¨ ¦. 400 K. A normal shock wave occurs in the tube and the flow discharges to the atmosphere. and CHAPTER 9.0. The system is supplied by a vessel.0 [m] long.5 [cm] in diameter and 1. The vessel conditions are at 29.2: A system is composed of a convergentdivergent nozzle followed by a tube with length of 2. With these conditions a pipe inlet Mach number is 3. (b) the temperature at the pipe exit.166 Example 9.

Thus.005. R = 287 [J/kgK ] and f = 0.0000 The temperature is 0.35714 0.65326 T1 = T1 T01 = 0.02722 0. S OLUTION (a) Assuming that the pressure vessel is very much larger than the pipe.11528 0. T01 = 400K a P01 = 29.1) or the Potto–GDC T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ M 3.65[P ar] The mass flow rate through the system is constant and for simplicity point 1 is chosen which. It is further assumed t the flow through the nozzle can be approximated as isentropic. Take k = 1. m ˙ = ρAM c The density and speed of sound are unknowns and need to be computed. With the isentro relationship the Mach number at point one (1) is known.4: The schematic of Example (9. Hence. therefore the velocity the vessel can be assumed to be small enough so it can be neglected.2) (c) determine the Mach number when a normal shock wave occurs [Mx ]. ¦  "!"#  shock d-c nozzle êãë§ìMí ìHîOïMð ñ_ò    atmosphere conditions Fig. the stagnat conditions can be approximated for the condition in the tank.357 × 400 = 142.4.2346 0. then the following can be found eith from Table (9. 9.07623 4.8K T01 .

8 0.42857 and the conditions of the tube are 4f L D = 4 × 0.50918 1. and two.8 239. the speed of sound can be calculated as √ √ c1 = kRT = 1.005 × 1.1)) .54 = 0. The exit pressure determines the location of the shock.8 1.0000 My 0.47519 Ty Tx ρy ρx Py Px P0 y P0 x 2.69 4 sec 8.1 × 104 287 × 142.97 × kg π × 0.6790 3. to obtain the following M 3. (Fanno flow Table (9. check whether the flow is shockless by comparing the flow resistance and the maximum possible resistance. EXAMPLES OF FANNO FLOW Using the temperature.4 × 287 × 142.9640 0. From the Table (9.47519. one.0252 × 3 × 239.2346 0.025 Since 0. the possibility that the shock wave occurs immediately at the entrance for which the ratio for Mx are (shock wave Table (5.3333 0.8571 10. if a shock exists.8 > 0.21822 4.54[m/sec] The pressure at point 1 can be calculated as P1 = P1 P01 = 0.027 × 30 P01 0. Two possibilities are needed to be checked.52216 0.1) or by using the Potto–GDC. by comparing “possible” Pexit to PB .52216 the flow is choked and with a shock wave. First.9.32834 After shock wave the flow is subsonic with “M1 ”= 0.0 = 0.1)) Mx 3. the shock at the entrance of the tube.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 (9.6.97 kg m3 (b) First. shock at the exit and comparing the pressure ratios.2) m ˙ = 1.0000 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.

One possible way to find the e temperature.168 M 4fL D P P∗ P0 P0 ∗ CHAPTER 9.3904 1.8568 × 0. Thus.95679 0. With the location of shock.2919 2. 5 You can use any method you which. T2 is by finding the location of the shock. For example.2549 0.2549 1. 3) According your root finding algorithm5 calculate or guess the shock location and th compute as above the new M1 .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× × 0.47519 0. calculate the parameters for shock locat fL with known 4D in the “y” side.1912 0.47519 are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.12476 The actual pressure ratio 1/29. “claiming” upstream from exit through shock to the entrance. . 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. FANNO FLOW ρ ρ∗ U U∗ T T∗ 0. Then either by utilizing shock table or the program.32834 × 1 2. the shock is somewhere downstream. but be-careful second order methods like Newton-Rapson method can be unstable.65 = 0.3904 0.1481 The stagnation values for M = 0.89545 1. To check Secant Method. 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.0338 is smaller than the case in which shock occurs the entrance. To find the location of the shock ra 2 of the pressure ratio.50917 1.9640 0. 2) Calculate the entrance Mach assuming shock at the entrance. to obt the upstream Mach number. P P1 is needed.47519 1.85676 1.

In the process Mx and My must be calculated (see the chapter on the program with its algorithms. However. The only limitation of the model is its uniform velocity (assuming parabolic flow for laminar and different profile for turbulent flow. 9 If you have experiments demonstrating this point. Nevertheless.57981 (c) The way of the numerical procedure for solving this problem is by finding produce M1 = 3. The information from the wall to the tube center6 is slower in reality. In the Fanno model there is no heat transfer.gov/digidoc/report/tm/44/NACA-TM-844.7 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.57910 4f L D 0. These changes include the choking point at lower Mach number. It has to be noted and recognized that as opposed to subsonic branch the supersonic branch has a limited length. the comparison with reality shows that heat transfer cause changes to the flow and they need to be expected.9899 My 0. please provide to the undersign so they can be added to this book. The author seeks for a nice explanation of this concept for non–fluid mechanics engineers.7. This solicitation is about how to explain this issue to non-engineers or engineer without a proper background.8 Maximum length for the supersonic flow To insert example on the change in the flow rate between isothermal flow to Fanno Flow. repeat previous stage until the solution is obtained.0000 M2 1.nasa.PDF 8 Many in the industry have difficulties in understanding this concept. Many of the pictures in the literature carry copyright statements. 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. 9. These results were obtained from the mathematical derivations but were verified by numerous experiments9 . It also must be recognized that there is a maximum length for which only supersonic flow can exist8 . 7 See on the web http://naca. experiments from many starting with 1938 work by Frossel7 has shown that the error is not significant. Insert also example on percentage of heat transfer.22019 0. M1 3.0000 4fL D up 4fL D down Mx 1. and.9. up that will 9. because the very limited amount of heat transformed it is closer to an adiabatic flow.). on the comparison of the maximum length of isothermal model and Fanno Model.larc.). . furthermore. 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.

5 1. k Thu Mar 3 16:24:00 2005 Fig.6 1.5: The maximum length as a function of specific heat.8215 The maximum length of the supersonic flow is limited by the above number. 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 → ∞.45 1.3 1.35 1.  D .25 1.9 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.4 0. k as shown in Figure (9. The maximum length in supersonic flow In Fanno Flow 1.6 0.9 0.5 0.5).4 1.3 1. k 9. k = 1.4) = 0.170 as follows: CHAPTER 9.2 1.55 1.1 0 1. Even the flow in the tube starts as a supersonic in parts of the tube can be transformed into the subsonic 4fLmax maximum length. 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.65 spesific heat.2 1.8 0.5 1.1 1 0. From the above analysis.3 0.2 0. 9.7 0.4 1.

The entrance temperature increases as can be seen from Figure (9.43 5&687 9 ' ( Fig. Three parameters. 9. It is worth noting that entrance Mach number is reduced (as some might explain it to reduce the flow rate). First. 4D . increasing the tube length results in increasing the exit Mach number (normally denoted herein as M2 ).9. the entrance Mach number. Once the Mach number reaches maximum (M = 1). A shock wave can occur and some portions of the tube will be in a subsonic flow pattern. and the pressure ratio. the mass flow rate decreases. all the three parameters can be varied and they are discussed separately here. Only a combination of these two parameters is truly independent. Subsonic branch For converging nozzle feeding./.9.6: The effects of increase of 4f L D on the Fanno line In the analysis of this effect. P2 /P1 are controlling the flow. no further increase of the exit Mach number can be achieved.102. The discussion has to differentiate between two ways of feeding the tube: converging nozzle or a converging-diverging nozzle. WORKING CONDITIONS 171 branch. 9.>=? ? I I X K<L N ?@BA G J MN U¨V V V W NOBP&Q8R4S T $&% )+*-.9. In this process. M1 .1 fL Variations of The Tube Length ( 4D ) Effects C8DFE H¨I :<. the dimensionless fL friction. The . it should be assumed that back pressure is constant and/or low as possible as needed to maintain a choked flow. the treatment of the two branches are separated.7). However.

Point b is the maximum possible flow for supersonic flow and is not dependent on the nozzle. the shock reaches to the nozzle. in which no supersonic flow is possible in the entire nozzle–tube system. Between these transitional points the effect parameters such as mass flow rate.7: The development properties in of converging nozzle velocity therefore must decrease because the loss of the enthalpy (stagnation temP perature) is “used.” The density decrease because ρ = RT and when pressure is remains almost constant the density decreases.e.172 constant pressure lines CHAPTER 9. The next point. 4D the flow becomes choked a different flow pattern emerges. Thus. increase of the dimensionfL . the mass flow rate must decrease. FANNO FLOW Y[Z \ 1’ 1 2 Fanno lines 2’ 1’’ 2’’ ] Fig. These results are applicable to the converging nozzle. entrance and exit Mach number are discussed. referred here as the critical point c.8)): . Point a is the choking point (for the supersonic branch) in which the exit Mach number reaches to one. In the case of the converging–diverging feeding nozzle. Once less friction. At the starting point the flow is choked in the nozzle. is the point in which no supersonic flow is possible in the tube i. The following ranges that has to be discussed includes (see Figure (9. Supersonic Branch There are several transitional points that change the pattern of the flow. There is another point d. to achieve supersonic flow. results in a similar flow pattern as in the converging nozzle. 9.

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

2 0.6 0.4 0.2 0.9 0.4 0. On one hand. Somewhat different then the subsonic branch the mass flow rate is constant even if the flow in the tube is completely subsonic.6 0.8 0. This situation is because of the “double” choked condition in the nozzle.8 0.9 1 Exit Mach number Tue Oct 19 09:56:15 2004 Fig.3 0.5 0. Does anyone have an explanation for this strange behavior suitable for non–engineers or engineers without background in fluid mechanics? 11 Note that ρ increases with decreases of M but this effect is less significant.0 Fanno Flow 1 0. In this range. Figures (9.9: M1 as a function M2 for various 4f L D 10 On a personal note.1 0 0 0. the resistance increases and on the other hand.8) exhibits the development of M1 .7 Entrace Mach number 0. the flow rate decreases since (m ˙ ∝ M1 )11 .5 0. 9.174 CHAPTER 9. this situation is rather strange to explain.0 = 10. To summarize the above discussion. 1 1 .3 0. FANNO FLOW and the exit velocity remains constant!10 In the last range c − −∞ the end is really the pressure limit or the break of the model and the isothermal model is more appropriate to describe the flow. fL .0 = 100. 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. fL M2 mass flow rate as a function of 4D . M1 as a function of M2 4fL  = 0.7 0.1 0.1 D = 1. the exit Mach number remains constant and equal to one.

tow.4 two extremes are shown.2 1.2 0. The fL larger 4D larger M1 must occurs even for shock at the entrance. Subsonic velocity occurs for supersonic entrance velocity.2 = 0.5 2 1. at the tube entrance. The Figure was calculated by max utilizing the data from Figure (9. 9. and as can be observed.6 1.4 1. when the shock wave occurs at the tube exit and two. shock at the entrance.10: M1 as a function M2 for different 4f L D for supersonic entrance velocity. the maximum .8 2 M1 M1 as a function of M2 for the subsonic brench 4fL = 0.9. below the maximum critical length. Below. the larger 4D larger differences between exit Mach number for the different shock locations.4 0. Obviously.5 0 0 0.1  D = 0.10) only for 4f L 4f L 4f L D = 0. Fanno Flow 5 4.10) exhibits the entrance Mach number as a function of the M2 .6 0.5 1 0.4 Tue Jan 4 11:26:19 2005 Fig. one.2 shown with only fL creates shock at the exit only.9) exhibits the M1 as a function of M2 .1 and D = 0. shock at the exit. In Figure (9.5 3 2. WORKING CONDITIONS 175 Figure (9. One. shockless flow. the supersonic entrance flow has three different regimes which depends on the back pressure.9. The Figure (9.8 1 M2 1. For D = 0.4 = 0.2) by obtaining the 4f L for M2 and subtracting D 4f L the given D and finding the corresponding M1 . fL For a given 4D . Obviously there can be two extreme possibilities for the subsonic exit branch.5 4 3. and three.1 shock = 0.

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

Note however that exit Mach number. Between point a and point b the shock is in the nozzle. In this range and further reduction of the pressure the mass flow rate is constant no matter how low the back pressure is reduced. Up to this stage the nozzle feeding the tube increases the mass flow rate (with decreasing back pressure). .12) shows different pressure profiles for different back pressures. A back pressure that is at the critical point c results in a shock wave that is at the exit.9.11: The pressure distribution as a function of 4f L D for a short 4f L D Short 4f L D Figure (9. When the back pressure is below point c. M2 < 1 and is not 1. 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.9. 9. 13 It is common misconception that the back pressure has to be at point d. The back pressure below point c has some adjustment as it occurs with exceptions of point d. 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. the tube is “clean” of any shock13 .

9.9. The is no analytical solution for the location of this point c. In reality these effects have significance and needs to be accounted for some instances. is a function entrance Mach number.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. 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. point c in this case is process as explained in the short 4D fL max < 4f L .3 Entrance Mach number. M1 fL .178 ‘ƒ’ & CHAPTER 9. The and the “extra” 4D procedure is (will be) presented in later stage. effects In this discussion. 9. 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. However. At point c or location of the shock wave. This dissection deals only with the flow when it reaches the { ųÆÇÈ·ÉÊ¦Ë ÌmÏÍ Î for a long 4f L D . the effect of changing the throat area on the nozzle efficiency is neglected. Further reduction of the back pressure at this stage will not “move” the shock wave downstream the nozzle. M1 .

Practically.6 1. the nozzle area ratio is changed by changing the throat area. k . there are two different maximums for 4D .9. the critical point is where result in the entrance Mach number.8 1.25 Tue Jan 4 12:11:20 2005 Fig.15 4fL  D 0. This analysis fL max deals with the case where 4D is shorter than total 4f L . M1 is a function of the ratio of the nozzle’s throat area to the nozzle exit area and its efficiency. WORKING CONDITIONS 179 Mach number in Fanno Flow 4fL  D 2 1. M1 .05 0. This effect is the third parameter discussed here.4 0. and second the maximum depends on the entrance Mach number.2 1 0. supersonic branch reached otherwise the flow is subsonic with regular effects.8 0. 4f L D is equal to 4f Lmax D as a The process of decreasing the converging–diverging nozzle’s throat increases the . first is the total 4f L maximum D of the supersonic which depends only on the specific heat. D Obviously.4 Mach Number 1. fL As was shown before.6 0.9.3 for Fanno Flow when the total resistance 4D 4f L D The entrance Mach number. 9.2 0 0 0.1 shock at 75% 50% 5% 0.13: The effects of pressure variations on Mach number profile as a function of fL = 0. 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.2 0. in this situation.

area) decreases the flow rate while in a different way (increasing the 4D . Decreasing the nozzle throat area moves the shock location downstream.4 4 3.15 4fL  D 0.1 0.14: Fanno Flow Mach number as a function of 4f L D when the total 4f L D = 0.2 P2/P1 2.05 0.8 2. By doing so.4 0 0 0.8 0.6 3.8 4. The process of decreasing the throat area also results in increasing the pressure drop of the nozzle (larger resistance in the nozzle15 )16 .4 2 1. decrease of the throat area results in flushing the shock out of the tube.25 5% 50 % 75 % Fri Nov 12 04:07:34 2004 Fig.3 entrance14 Mach number.2 0.2 0.6 1.180 CHAPTER 9. The “payment” for increase in the supersonic length is by reducing the mass flow. The mass flow rate is proportionally linear to the throat area and therefore the mass flow rate reduces. 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. the throat area decreases. Further. FANNO FLOW P2/P1 Fanno Flow 4fL  D 4. 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. The reference to the tube is because it is the focus of the study. 9. 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. If the tube contains no supersonic flow then reducing the nozzle throat area wouldn’t increase the entrance Mach number.

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

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

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

42390 1.1318. 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 13.52) Figure (9. Hence. M1 reduces after the maximum length is exceeded. Example 9. the value of left side is −0. .1641 between the negative of left side to the positive of the right side17 .1318. (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.0000 With (M1 ) M 0.5652 74.76820 = 0.3102 Now the solution is somewhere exceeds the maximum 4f L D 4f L D max for the supersonic flow.17) shows the entrance Mach number.821508116. Accordingly continue.9 − 0.76820 (flow is choked and no additional 4D ). The maximum 4f L D 4f L 4f L is 0. WORKING CONDITIONS 183 flow is supersonic see Figure (9.4 The solution is obtained by an iterative process. 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 = 0.1318 = 2. The maximum for M1 = 8 is 4D = 0.3: Calculate the shock location for entrance Mach number M1 = 8 and for assume that k = 1.39289 The extra ∆ 4f L D My 0.15). 4f L D up = − 4f L D max 4f L D up . The retreating length is expressed as subsonic but 4f L D retreat = 4f L D max∞ − 4f L D sup (9.9. The left side is when the shock occurs at Mx 8.5000 0. D exceed the maximum length D for this entrance fL Mach number.7461 1. 4f L D = 0.76820. thus the extra tube is ∆ 4f L D fL = 0.9.4417 2. Hence.6136 2.3867 5.3591 0.442 − 0.00849 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 2.4 (Mexit = 1).9 S OLUTION max for k = 1. Or in a very long tube the whole flow will be subsonic.

64850 0. FANNO FLOW (d) Calculate the associate Mach number.31783 0. Mx with the Mach number.90000 0.32932 The iteration summary is also shown below i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 4fL D up 4fL D down Mx 1.27830 0.32907 0.90000 0.57068 0.90000 0.32927 0.6706 1.90000 0.57069 0.57122 0.66779 0.64830 0. .32932 0.6699 1.6706 1.57068 0.57318 0.0000 4f L D down against the old one.32932 0.22574 0.32932 0.32932 0.57070 0.32932 0.6706 1.6703 1.90000 0.6554 1.32921 0.90000 0.90000 0.90000 0.57184 0.6635 1.32395 0.6673 1.57068 0.57605 0.90000 0.65246 0.30494 0.65728 0.57068 0.90000 0.90000 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 8.64872 0.6706 My 0.57068 0.90000 This procedure rapidly converted to the solution.6706 1.64830 0.74664 0.6021 1.64834 0.32878 0.6382 1.6705 1.58217 0.90000 0.59506 0.6706 1. If it is satisfactory stop or retu 4fL D up 4fL D down Mx 1.64830 4fL D 0.3838 1.57068 0.64831 0.57093 0.32682 0.6706 1.90000 0.184 CHAPTER 9.69119 0.64830 0.64839 0.64830 0.6691 1.0000 M2 1.90000 0.90000 0.6706 My 0.90000 0.5286 1.62170 0.32816 0.57073 0.64920 0.64831 0.64832 0.90000 0.67426 0.32930 0.65023 0.32931 0.57079 0.

99 which requires that enough for M1 < 0. Hence the mass flow rate is a function of 4D because M1 changes. The only difference is in small dimensionless fL .3 M1 0.2 0. Hence. 9. if one allows certain deviation of temperature. THE APPROXIMATION OF THE FANNO FLOW BY ISOTHERMAL FLOW185 9.10 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. The results are very similar for isothermal flow.4 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.8 iso = 0.1 = 0.5 = 0. This tolerance requires that (T0 − T )/T0 = 0. One must notice that the maximum temperature at the entrance is T0 1 .10. 4D .67. to insert a question or example about this issue in end with comperison to Isothermal Flow 0. Looking at the table or Figure (9.1 0 0 10 20 30 40 Wed Mar 9 11:38:27 2005 50 4fL  D 60 70 80 90 100 Fig.1 iso = 0. friction.18) the dominating parameter is 4D .2 = 0.18: The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow fL As it can be seen for the Figure (9.15 even for large k = 1. say about 1% that flow can be assumed to be isothermal. When the Mach number decreases the temperature approaches the stagnation temperature (T → T0 ).8 M1 Fanno flow 0. This fL requirement provides that somewhere (depend) in the vicinity of 4D = 25 the fL flow can be assumed isothermal.2) or the results from computer program attached to this book shows that reduction of the mass flow is very rapid.9.

0000 40.1196 7. 0.5393 4.5910 4. T ≈ T0 and is the same for the pressure.99556 4.5.7027 5.30000 0.2842 0.0000 40. FANNO FLOW 9. Hence.1.1697 50.7569 67.0000 42.8.05. The stagnation conditions at the entrance are 300K and 3[bar] air.022 k = 300000 × × 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.99677 0.127 × RT 4 1.80000 Only for the pressure ratio of 0. 0. ρ ρ0 M 0.0000 40. The mass flow rate is m ˙ = P AM π × 0. or accurately by utilizing the program as in the following table.12728 0. = 4 × 0.07975 M2 1.99097 0. L = 4[m].98928 0.99873 0.12420 0.02 From Figure (9. the mass rate is a function of the Mach number.0000 0.09965 4fL D 4fL D 1 4fL D 2 P2 P1 40.99354 0.4 ≈ 0.48 287300 kg sec .0000 40.1 M1 ≈ 0.99233 0. M1 0.4: To demonstrate the utility in Figure (9.11637 0.6523 5.99741 0.0733 7. D = 0.99683 4.12728 0.22697 0.99195 0.13 etc. Find the mass flow rate for f = 0.18) consider the following example.18) for P2 /P1 = 0. 4f L D 4f L D . S OLUTION First calculate the dimensionless resistance.1 the flow is choked.98874 0.2519 A×P A∗ ×P0 Therefore.11 More Examples of Fanno Flow Example 9.0 2.99692 0.50000 0.7569 107.07975 T T0 A A P P0 0.05 × 4 = 40 0.02[m] and pressure ratio P2 /P1 = 0.40790 0.42 0.11392 0.186 CHAPTER 9.1697 10.3.12420 0. 0.11392 0.4206 0.

4027 3.1512 9.3184 2.6191 2.68374 0.0382 1.30500 0.2838 1.35 280.12728 0.4626 1.2119 1.1533 1.0351 3.1291 1.4613 10.40 0.3817 14.1882 1.8218 5.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 787.1065 1.6958 1.1991 1.00 1.0934 1.5901 2.8 ∼ 0.1273 kg sec kg sec kg sec 187 9.48 × = 0.05476 0.633 1.63481 0.2620 15.77894 0.0702 2.0944 1.06570 0.09 0.2893 1.2047 1.21822 0.0624 1.4834 5.9034 11.07229 0.4525 2.5914 18.12 The Table for Fanno Flow Table 9.1489 1.3398 1.55 0.0922 1.70 0.32459 0.4182 10.1996 1.0929 1.40825 1.1: Fanno Flow Standard basic Table M 0.66 106.8708 1.49082 0.0207 1.5664 1.09851 0.08 440.5116 19.000 0.1981 1.0638 1.25 0.66667 0.9341 1.48 × = 0.1194 1.2508 9.20814 0.20 0.07975 P2 = 0.91460 0.1905 1.1242 P2 = 0.03 140.02 193.87037 0.017 1.30 m ˙ P1 0.10 0.9216 14.00 3.32572 0.48 × = 0.01451 0.42857 .95781 1.1788 1.7092 1. THE TABLE FOR FANNO FLOW and for the rest m ˙ 0.000 0.1429 1.4935 1.2200 13.04 0.1139 P2 m ˙ = 0.6659 15.48326 0.73179 0.08758 0.3848 1.3085 1.4487 2.90 0.2993 3.3546 2.12.7634 1.35 0.1713 1.00 2.72805 0.4961 66.0474 11.1356 1.45 0.53452 0.1378 4.3665 1.3 ∼ 0.6843 7.80 0.82514 0.37879 0.2915 13.06 0.04381 0.6742 3.75 0.2549 1.21822 4.468 P1 0.3005 27.72 83.60 0.00000 1.08 0.002 1.6400 2.1273 0.7780 2.5753 1.61237 0.0485 1.1315 1.5333 8.235 30.03286 0.061 1.4815 21.6416 8.044 1.43 P1 0.0693 1.1988 1.95 1.688 0.07 0.50918 0.0691 0.65 0.52216 36.1998 1.1976 1.03 0.0327 1.10944 0.8254 18.4318 22.0787 1.3865 1.43133 0.1628 1.03633 0.00328 0.1994 1.9435 5.4554 2.85 0.05 0.1381 1.0 0.1985 1.6183 1.1618 6.07664 0.50 0.5826 3.30 0.964 1.0089 1.9.1273 0.2616 12.5 ∼ 0.000 0.9635 4.27217 0.1852 1.58506 0.

00 45.000680 3.00200 2.00 70.00 65.00 50.72988 0.5E +4 0.4E +2 0.00 5.82033 0.8E +6 0.434 2.00 10.4E +6 0.6E +4 0.00166 0.00240 0.3E +2 0.00 9.13363 10.447 2.81265 0.377 2.41079 0.295 2.42857 0.82078 0.00952 0.236 2.46771 0.40859 0.82008 0.00 7.047619 1.81582 0.036860 1.089443 25.390 2.063758 53.20000 0.63306 0.00 8.138 2.18 0.40889 0.40875 0.000579 5.447 2.81755 0.445 2.00296 0.00 40.069767 0.40866 0.69380 0.029348 3.78683 0.28571 0.41833 0.00153 4.000809 2.000979 1.443 2.6E +6 0.446 2.448 0.42066 0.00122 .00 30.3E +6 0.00663 0.8E +5 0. FANNO FLOW Table 9.00142 0.00198 0.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.00390 4.1: Fanno Flow Standard basic Table (continue) M 4.82066 0.1E +5 0.446 2.5E +5 0.00 6.00 0.6E +5 0.42390 0.40849 0.014815 0.00 25.9E +2 0.00488 0.448 2.76819 0.000500 7.086957 0.00121 8.440 2.00374 0.057143 0.40988 0.40938 0.333 2.00 60.75280 0.40853 0.81975 0.77899 0.11111 0.00 55.72 0.00609 1.00271 1.359 2.00 35.023905 5.5E +6 0.40908 0.448 2.81928 0.81860 0.188 CHAPTER 9.00 20.43568 0.44721 0.14634 0.40846 2.0E +2 0.82052 0.

1 Introduction The third simple model for an one dimenhji flow prn q k`l direction mo sional flow is for constant heat transfer for fric( s t uwv tionless flow. As before. 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).CHAPTER 10 RAYLEIGH FLOW Rayleigh flow is (frictionless) flow with heat transfer through a pipe of constant cross sectional area. 10. This flow referred in the literature as Rayleigh Flow (see historical notes).Fig. As opposed to the two previous models. This fact create situation different compare to the previous two models. 10. In contrast. Yet. 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. the heat transfer in Rayleigh flow occurs either between unknown temperature to tube and the heat flux is maintained constant. This model usage is to have a rough predict the conditions occur mostly in 189 . a simple model is built around assumption of constant properties (poorer prediction to case were chemical reaction take palace). In practice Rayleigh flow is really provide good model for the real situation.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. Rayleigh flow is practical and useful concept in a obtaining trends and limits. This model applied to case where the heat transfer is significant and the friction can be ignored.

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 (10.4) A(P1 − P2 ) = m ˙ (V2 − V1 ) ρ 1 U1 A = ρ 2 U2 A = m ˙ There are four equations with four unknown. for smaller range of temperature and length the calculations are more accurate.4) are similar to the equations that were solved for the shock wave. are encapsulated in this model. the main characteristic of the flow such as chocking condition etc.7) U2 ρ1 = = ρ2 U1 kRT2 U1 kRT1 . a solution can be obtained. As appose to Fanno flow in which the resistance always oppose the the flow direction. P2 1 + kM1 2 = P1 1 + kM2 2 (10.5) The equation of state (10. if the upstream conditions are known (or downstream condition are known).2 Governing Equation Q = Cp (T0 2 − T0 1 ) (10. One can notice that equations (10. Thus. RAYLEIGH FLOW situations involve chemical reaction. in Rayleigh flow also cooling can be applied. In analysis of the flow.190 CHAPTER 10. one has to be aware that properties do change significantly for a large range of temperature.1) (10. Nevertheless.2). The basic physics of the flow revolves around the fact that the gas is highly compressible. Yet. The density change though the heat transfer (temperature change).4) can further assist in obtaining the temperature ratio as T2 P2 ρ 1 = (10. The flow velocity acceleration change the direction when the cooling is applied. (10.2) (10.3) and (10. 10.6) T1 P1 ρ 2 The density ratio can be expressed in term of mass conservation as U2 r r √ kRT2 = kRT1 M2 M1 T2 T1 (10.

12) .7) into equation (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. and s1 are constant then the variable parameters are M2 . 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 (10.8) Transferring the temperature ratio to left hand side and squaring results in T2 1 + kM1 2 = T1 1 + kM2 2 2 M2 M1 2 (10.2.10) (10.11) results in 2(1 − M 2 ) 1 ds = Cp dM M (1 + kM 2 ) (10. A derivative of equation (10.5) and (10. The second maximum can be expressed as dT /ds = ∞ The second law is used to find the expression for derivative. GOVERNING EQUATION Substituting equations (10. 10.6) yields T2 1 + kM1 2 M2 = T1 1 + kM2 2 M1 T2 T1 191 (10.11) Let the initial condition M1 .9) {N|~ }€ y ‰XŠˆ‹ ne ŒŽˆŒ N‚„ƒ Pr es su re li co ns ta …X†ˆ‡ nt z Fig. and s2 .10.

These value referred as the “star”1 values.12) when letting the variable parameters be T2 . The derivative is negative. when cooling applied to a tube the velocity decreases and heating applied the velocity √ increases. The peculiars point of M = 1/ k when additional heat is applied the temperature is decreasing.12) and (10. some what similarly to Fanno flow. dT /ds = ∞ when M = 1. And cooling results in reduction of entropy. increase of heating results in increase of entropy.17) star is an asterisk.14) (10.2)) and therefore it is not possible (the maximum entropy at M = 1. Yet for every curve. It is convent to referrers to the value of M = 1. The derivative is equal to zero when 1 − kM 2 = 0 or M = 1/ k or when M → 0. yet note this point is not the choking point. Yet. The chocking is occurred only when M = 1 because it violate the second law. The equation (10. dT /ds < 0.). and M2 results in dT 1 − kM 2 = constant × 3 dM (1 + kM 2 ) Combining equations (10. several observations can √ be generalized. 1 + kM1 2 P∗ = P1 1+k The temperature ratio is 1 T∗ = 2 T1 M 1 + kM1 2 1+k 2 (10.13) On T-s diagram a family of curves can be drawn for a given constant. RAYLEIGH FLOW Take the derivative of the equation (10. From thermodynamics. The transition to supper sonic flow occurs when the area changes.16) U∗ ρ1 = = ∗ ρ U1 1 The √U kRT ∗ kRT ∗ √ √ U1 kRT1 kRT1 ∗ √ = 1 M1 T∗ T1 (10. . Hence.5) can be written between chocking point and any point on the curve. chocking can be explained by the fact increase of energy must accompanied by increase of entropy. The derivative is equal to infinity. But the entropy of supersonic flow is lower (see the Figure (10.192 CHAPTER 10.13) by eliminating dM results in M (1 − kM 2 ) dT = constant × ds (1 − M 2 )(1 + kM 2 )2 (10.15) (10.

090909 0.392 2.10.367 2.023223 0.55 0.017119 0.80 0.57447 0.86833 0.13793 0.267 1.99207 0.99290 1.036212 0.81013 0. Several observations can be made in regards to the stagnation temperature.86204 0.00431 0.266 1.1).379 2.34686 0.778 1.030 1.99814 2.10 0.060 1.08 0.596 1.00765 0.388 2.70804 0.75991 0.40887 0.51001 0.075 1.75524 0.90 0.015224 0.51413 0.50 0.31373 0.273 2.69751 0.025 1.45 0.056020 0.343 1.023669 0.98097 0.3.25684 0.235 1.046777 0.65 0.178 1.44444 0.961 1.30440 0.37865 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 (10.193 1.265 1.95 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.423 1.94009 0.04 0.397 2.75 0.001 0.014300 0.79012 0.266 1.218 1.96395 0.266 1.049 1.114 1.40 0.00860 0.07 0.25 0.058 1.17355 0.1: Rayleigh Flow k=1.20 0.125 1.019222 0.09 0.029 1.094 1.027841 0.91097 0.014 1.261 1.4 M 0.384 2.81892 0.3 Rayleigh Flow Tables The “star” values are tabulated in Table (10.030215 0.020529 0.61515 0.25096 0.199 1.70 0.011922 0.373 2.011680 0.43894 0.157 1.06 0.96081 0.00917 0.508 1.259 1.19) 10.264 1. 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 (10.05 0.60 0.20661 0.85 0.043 1.35 0.91670 0.019 1.69136 0.00598 0.870 1.00216 0. Table 10.61393 0.262 1.03 0.005 1.045616 0.025 1.395 2.52903 0.207 2.038075 0.015 0.90850 0.85987 0.63713 0.19183 0.011 1.135 1.00383 0.00517 0.95693 .30 0.686 1.131 2.

85917 0.503 1.0E +2 3.0 9.017021 0.43122 0.00190 0.886 2.31486 0.525 1.39643 0.52438 0.33454 0.43353 0.675 1.1 2.68 75.701 1.571 1.5 9.18788 0.713 1.8 1.50885 0.078 1.79576 0.65377 0.375 1.222 2.89087 0.081772 0.00469 0.713 .68494 0.66350 0.405 1.75613 0.7 2. RAYLEIGH FLOW Table 10.030094 0.0 5.3 2.510 1.146 1.13223 0.79339 0.95 54.8E +2 2.28551 0.88419 0.699 1.69700 0.0 2.122 1.00240 1.0 3.418 2.21417 0.70174 0.77406 0.698 1.29687 0.9 2.5 5.51098 0.1 1.0 25.078 1.20040 0.205 1.4 1.00326 0.055363 0.494 1.431 1.2E +4 8.620 1.328 8.00732 0.0 1.56982 0.36364 0.51647 0.743 1.039900 0.078487 0.5 8.67380 0.24615 0.16831 0.0 7.0 8.0E +4 1.136 3.73954 0.2 2.33439 0.475 1.49122 1.0 30.52970 0.667 1.37870 0.57831 0.7 1.11111 0.588 1.538 1.00140 1.695 1.5 1.4E +2 1.656 1.6 1.52893 0.580 1.023495 0.8 2.634 2.21 38.00428 0.49415 0.60894 0.0 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 1.75250 0.044 1.034483 0.1: Rayleigh Flow k=1.227 12.194 CHAPTER 10.49259 0.80539 0.30864 0.95798 0.35561 0.046693 0.21419 0.712 1.3 1.058264 0.5 7.41 1.035650 0.52004 0.90928 0.17647 0.256 1.00 0.690 1.00 1.693 1.5 2.56734 0.97872 0.81414 0.93425 0.58909 0.51349 0.00274 0.72421 0.61580 0.702 1.240 1.711 1.0 35.22936 0.550 1.52356 0.9 3.018846 0.050943 0.5 10.301 1.019 1.686 1.8E +2 1.53633 0.71301 0.340 1.873 3.681 1.2 1.028972 0.3E +2 3.026490 0.5 6.044910 0.91185 0.49174 0.50702 0.641 1.039883 0.46106 0.26478 0.96031 0.0 6.020979 0.6 2.005 1.65398 0.455 1.40384 0.316 1.7E +5 1.85971 0.4 (continue) M 1.71006 0.0E +2 1.64103 0.1E +4 3.616 1.5 4.403 1.0 4.067263 0.066667 0.99392 0.561 1.4 2.00 0.28028 0.47562 0.10256 0.000 1.49356 0.00 0.176 1.045 2.424 5.032053 0.54473 0.83628 0.092719 0.50 18.13540 0.0 20.63 27.55556 0.

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

195

M 40.0 45.0 50.0 55.0 60.0 65.0 70.0

T T∗

T0 T0 ∗

P P∗

P0 P0 ∗

ρ∗ ρ

0.00184 0.00145 0.00117 0.000971 0.000816 0.000695 0.000600

0.49089 0.49066 0.49050 0.49037 0.49028 0.49021 0.49015

0.00107 0.000846 0.000686 0.000567 0.000476 0.000406 0.000350

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

1.714 1.714 1.714 1.714 1.714 1.714 1.714

The Data is presented in Figure (10.3).

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

3

@( M = ∞ ) 1.7143

1 @( M = ∞ ) 0.4898 0

0.1

1

M

10

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

196

CHAPTER 10. RAYLEIGH FLOW

10.4

Examples For Rayleigh Flow

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

0.30440

0.25684

2.2069

1.2177

0.13793

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

0.40530

0.34376

2.1341

1.1992

0.18991

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

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

0.25000 0.98765 0.96942 2.4027

0.95745 2.3005

1.0424

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

P0 2 = P 1

P0 1 P1

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

10.4. EXAMPLES FOR RAYLEIGH FLOW

197

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

T0 1 = T 1

T0 1 T1

= 300/0.98765 = 303.75[K ]

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

0.29831 0.98251 0.95686 2.0454 The exit stagnation temperature is
isentropic

0.94012 1.9229

0.90103

T0 2 = T 2 The heat release becomes

T0 2 T2

= 400/0.98765 = 407.12[K ]

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

1.0286

0.97959

1.2000

1.0116

0.85714

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

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

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

0.62666 0.64051 0.53376

T0max 1013.7 = = 1158.51[K ] Tmax 0.875

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

198

CHAPTER 10. RAYLEIGH FLOW

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

0.83333 0.63394 1.0000

0.52828 0.52828 0.52828

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

T0 2 1000 = = 1200.0[K ] T2 0.83333

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

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

P P∗

0.50160

0.42789

2.0589

1.1805

0.24362

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

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

199

0.28028

0.65398

0.17647

3.4245

1.5882

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

2.6790

3.8571

10.3333

0.32834

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

0.47519

0.40469

2.0802

1.1857

0.22844

M 0.47519

T T∗

T0 T0 ∗

P P∗

P0 P0 ∗

ρ∗ ρ

0.75086

0.65398

1.8235

1.1244

0.41176

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

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

200

CHAPTER 10. RAYLEIGH FLOW

Pennsylvania. 2 Even if the instructor feels that their students are convinced about the importance of the compressible. Scranton. 1964.” 2nd Edition.Fig. International Textbook Co. These books are OCR. the undersigned discover two text books which to include some material related to this topic. Prentice Hall. and “Compressible Fluid Flow. In this Chapfor relativly long tube ter a discussion on application of comVolume is a function of pressure or rigid pressible flow to other fields like manu(the volume can be also a function of inertia and etc) Semi rigid tank facturing is presented2 . are far Fanno model 1 After completion of these Chapters. It Isothermal model for relatively long tube was the undersigned experience. 1985. 201 . these books contained only limit discussions on the evacuation of chamber with attached nozzle. However. However.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 . Fundamentals of Gas Dynamics. There is a significant impor. A. by M. 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... Saad.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. the situations. in life. this example can further strength and enhance this conviction. J. 11. in many instances. A.

here Isothermal flow and Fanno flow are used as limiting or bounding cases for the real flow. Yet umes used in this model as it was stated before. 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. 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. 4D . For example. are discussed.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.3: A schematic of the control volIsothermal flow is appropriate model. 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. The analysis is extended to include a semi rigid tank. Yet Isothermal flow model is used as the limiting case. As it was shown in ChapfL ter (9) the resistance. And therefore. This author isn’t familiar with situations in which Control volume for the evacuating case Rayleigh flow model is applicable. no discussion is ofÉÊIËKÌ ÍÂÎ Ï Ð Ñ(Ò Ó fered here. at this stage.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. models. 11. 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. were the chamber volume is controlled or a function of the pressure. are presented (see Figure (11. the models in this Chapter are suitable for study of the filling the airbag or filling bicycle with air. Fanno flow model is the most appropriate in the case where the filling and Control volume for the filling case evacuating is relatively fast. 11. In the second Chapter. Addition- . 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. most car today equipped with an airbag. For instance. should be larger than 400. In the first Chapter models. 11.202 CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS more complicate. were the chamber’s volume is a function of external forces.1)). In case the fL filling is relatively slow (long 4D than the Fig.

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)

For rigid volume. A reasonable model for the tank is isentropic (can be replaced polytropic relationship) and Fanno flow are assumed for the flow in the tube.1 Rapid evacuating of a rigid tank With Fanno Flow ¯ (t) = 1. The temperature in the nozzle is changing because the temperature in the chamber is changing. 11. In such case. 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. Yet. V tank.17) “shrinks” and doesn’t contain the relative volume term.3. Owczarek [1964] found an analytical solution which described here. the general equation (11. the specific governing equation is ¯ −1 dP ¯ [M ]P ¯ 3k 2k =0 − k Mf ¯ dt (11.4.30) transformed into ¯= ln P k+1 2 " −(k+1) 2(k−1) ¯ t # ¯=e P 2(k−1) t ¯ ( k+1 2 ) −(k+1) (11.30) Here. equation (11. equation (11.32) .2 Isothermal Nozzle Attached In this case the process in nozzle is assumed to isothermal but the process in the chamber is isentropic.29) converted into ¯ P 1 (11.13) is reduced into V ¯ dP ¯=0 = ±f [M ]P ¯ dt Separating the variables and rearranging equation (11.11. the differential temperature change in the chamber is slower than the temperature change in nozzle. After the integration.22). Thus. f [M ] is expressed by equation (11.4 11. ¯ = 1 and for isothermal nozzle T ¯ = 1 Thus.29) ¯ dP ± f [M ] ¯ P ¯ t ¯= 0 dt 0 (11.31) 11. is constant and equal one for a completely rigid The relative Volume.4.

34) ¯ is a function of P and can It has to be realized that this is “reversed” function i. EVACUATING AND FILLING A SEMI RIGID CHAMBERS For a choked flow the entrance Mach number to the tube is at its maximum.8 1.e.4 0. The Figure (11.32) is obtained by noticing that and therefore M ¯ is not a function of time and by variables separation results in M ¯ t ¯ P ¯= dt 0 1 ¯ dP ¯ f [M ]P ¯ kM 3k−1 2k = 1 ¯ k M f [M ] ¯ P 1 3k ¯ ¯ 1− 2k dP P (11. The reduced time increases with decrease of the pressure in the tank.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 (11.2 0.8 V(t) = P (t) V(t) = P (0) 0. t be reversed for case.0 P(t) 0.210 CHAPTER 11.5) shows 1. The transition point is denoted here as chT .4 ¯ t 0.5: The reduced time as a function of the modified reduced pressure The big struggle look for suggestion for better notation.2 0 0 0. 11. equation (11.36) .33) results in ¯= t −k 2 ¯ 12 k − 1 P ¯ (k − 1)M f [M ] (11.6 0. Thus. Mmax ¯ = 1.33) direct integration of equation (11. At certain point the flow becomes chokeless flow (unless the back pressure is complete vacuum).35) The function is drawn as shown here in Figure (11.0 Fig.6 0.5). The solution of equation (11. But for the chocked case it appears as ¯ ¯ = 1 + (k − 1)M f [M ] t ¯ P 2 2k 1−k (11. that when the modified reduced pressure equal to one the reduced time is zero.

38) results in ¯= t −k 2 ¯ 12 k − 1 ¯ [M ] P (k − 1)Mf (11. Again.01[m] and length of L = 5. The solution of equation (11.4. The Figure shows that when the modified reduced pressure equal to one the reduced time is zero. Mmax ¯ = 1. 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. unless extra long time or extra low pressure is calculated/needed.0[m] S OLUTION The first parameter that calculated is 4f L 4f L D D =5 11. for the special cases where the choked equation is not applicable the integration has to be separated into zones: choked and chokeless flow regions.4. equation .025.37) For a choked flow the entrance Mach number to the tube is at its maximum. Example 11. The reduced time increases with decrease of the pressure in the tank. with today computer this should not be a problem and easily can be drawn as shown here in Figure (11.2 Filling Process ¯ −1 dP ¯ ¯ 3k 2k =0 ¯ − k Mf [M ]P dt The governing equation is (11. d = 0. And in the choke region the calculations can use the choking formula and numerical calculations for the rest. The chamber is connected with a rubber tube with f = 0.5). Thus. At some point the flow becomes chokeless flow (unless the back pressure is a complete vacuum). Further. when the flow became chokeless the entrance Mach number does not change much from the choking condition.38) direct integration of equation (11.2: A chamber with volume of 0. Nevertheless.1[m3 ] is filled with air at pressure of 10[Bar].11.39) It has to be realized that this is a reversed function. The transition point is denoted here as chT .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.

6: The reduced time as a function of the modified reduced pressure (11. It has to be remembered that the chamber can undergo isothermal process. 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.3 The Isothermal Process ¯ = 1.41) .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) (11.212 CHAPTER 11. For the double isothermal (chamber and tube) the equation (11. The combination of the For Isothermal process.8 0. 11.4 0.6 0. EVACUATING AND FILLING A SEMI RIGID CHAMBERS 1 V(t) = P(t) V(t) = V(0) 0.39) 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 ] (11.8 1 or ñ™òIóõô/ö"ñø÷ùòP ô/ö¯ü(ý þ ÿ úû ò¡  Fig.4.40) 11.2 0 à™áPâäã3å¯æ(ç èˆéwêìë3í ð î ï 0.4 0. T isentropic tank and Isothermal flow in √the tube is different from Fanno flow in that the chocking condition occurs at 1/ k . the relative temperature.2 0.6 0.

RAPID EVACUATING OF A RIGID TANK 213 11. a general solution is easily can be obtained similarly to rigid tank.46) The physical meaning that the pressure remains larger thorough evacuating process.45) The equation (11.42) into yields d ¯ 1+k P k ¯ dt Carrying differentiation result in ¯ +1 1 dP 1+k ¯k ¯ f [M ] = 0 ¯ k2 k M P −P ¯ k dt Similarly as before.4. increase of the pressure results in increase of the volume (as results of Hook’s law. it is assumed the ideal gas applied to the gas and this assumption isn’t relaxed here. the variables are separated as ¯ t ¯ −P k+1 2k ¯ f [M ] = 0 M (11. .4. After all.44) dt = 0 k 1+k ¯ P 1 −1 ¯ k2 ¯ k dP P ¯ M f [M ] (11.43) (11. The physical explanation when it occurs beyond the scope of this book.4 Simple Semi Rigid Chamber A simple relation of semi rigid chamber when the volume of the chamber is linearly related to the pressure as V (t) = aP (t) (11.42) where a is a constant that represent the physics. In this analysis and previous analysis the initial effect of the chamber container inertia is neglected. This situation occurs at least in small ranges for airbag balloon etc.4.5 The “Simple” General Case The relationship between the pressure and the volume from the physical point of view must be monotonous. as results in faster reduction of the gas from the chamber. Nevertheless.). in the known situations to this author pressure increase results in volume decrease (at least for ideal gas. the relation must be also positive. 11. Further.11. Substituting equation (11. The analysis is based only on the mass conservation and if unsteady effects are required more terms (physical quantities) have taken into account. Further.45) integrated to obtain the form −1 2k 2 ¯ 3k ¯= 2k t 1−P ¯ Mf [M ](3k − 1)(1 + k ) (11.

214 CHAPTER 11. it is waited to find such conditions before calling it as critical condition. In reality the value of n lays between zero to one.48) Notice that when n = 1 equation (11. Therefore. 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. as first approximation and simplified approach can be done by a single term with a different power as V (t) = aP n (11.785714286) for k = 1. 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 (11.51) The linear condition are obtain when 3k − 2nk − 1 = 1 −→ n = That is just bellow 1 (n = 0. This undersigned is not aware situation where this indeed play important role. d ¯ 1+nk−k k ¯ P dt +1 ¯ f [M ] = 0 ¯ k2 k M −P (11. However. There isn’t a real critical value to n.52) . 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.49) Again.49) reduced to equation (11.47) When n can be any positive value including zero.43). The physical meaning of n = 0 is that the tank is rigid. 6 Some suggested this border point as infinite evocation to infinite time for evacuation etc. Yet. similarly as before.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. 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. EVACUATING AND FILLING A SEMI RIGID CHAMBERS Any continuous positive monotonic function can be expressed into a polynomial function. 0.4.

equation (11.53) and (11. May 1996. Eckert. Journal of Manufacturing Science and Engineering (USA).18) and (11. no. Vol. pp.5. G.54) yields 1− Pexit P0 (t) fL k 4D 2 (11.55) This solution should used only for Min < 0.18) 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.54) Min = . 118.19) is described in “Pressure die casting: a model of vacuum pumping” Bar-Meir.18) and (11. E R G. J. P0 (t) Mexit The solution of two equations (11.19) must be solved numerically. 259-265. R.00286.11. (11. Goldstein. 2. .53) and equation (11. ADVANCE TOPICS 215 11.5 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. The solution of equation (11. otherwise equations (11.19) as Min Pexit = .

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

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

1)). The first approximation isotropic process describe the process inside the cylinder (see Figure (12.218CHAPTER 12. isontropic process ¢¤£¦¥¨§ ©.1.1 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 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. the best ventilation is achieved with a large tube or area. The conflicting requirements suggest an optimum area. The second objective is to provide a tool to “combine” the actual tube with the resistance in the tube. Nevertheless. 12. Obviously.1: The control volume of the “Cylinder” Before introducing the steps of the analysis. eliminating the need for calculations of the gas flow in the tube to minimize the numerical calculations. Obtaining conditions where the volume of displacing liquid is equal to the displaced liquid are called the critical conditions. the possibility of crack increases.1) certain sizes. For these reasons the optimum size is desired. The replacing incompressible liquid enter in the same amount as replaced incompressible liquid. These critical conditions are very significant that they provide guidelines for the design of processes. In the exhaust system cost of large exhaust valve increase with the size and in addition reduces the strength with the size of valve2 . 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. thus. In reality. 1 2 Fanno model Fig. V (t) = V (0) 1 − 2 After t tmax (12. But in a compressible substance the situation can be totally different. it is noteworthy to think about the process in qualitative terms. . 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. this model can be extended numerical in cases where more complex function is applied. which is also indicated by experimental studies and utilized by practiced engineers. in some situations like die casting this description is appropriate. A linear function is the simplest model that decibels changes the volume.

4) Equation (12. Notice that in this case that there are two different characwhere t teristic times: the “characteristic” time. tc is associated with the ratio of the volume and the tube characteristics (see equation (11.5) can be further simplified as ¯ dt 2dξ = ¯ (k − 1) (1 − Bξ ) ξ 1−t (12.3) 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 .1) can be non–dimensionlassed as ¯ (t ¯) = 1 − t ¯ V 219 (12.12. Equation (12. tmax is associated with the imposed time on the system (in this case the elapsed time of the piston stroke).7) ¯ =t (12.4) is can be solved only when the flow is chocked In which case f [m] isn’t function of the time.2) The governing equation (11. The first characteristic time.3) ¯ = t/tmax . The solution of equation (12. P (12.5)). dP k−1 reduce equation (12.1.6) (k − 1) (1 − Bξ ) [ξ ] ¯ dt ¯ 1−t (12.10) 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 (12.4) 2 [ξ ]( k−1 ) 2k −1 dξ 2k k−1 = where B = tmax ¯ tc M f [M ] And equation (12. tc and the “maximum” time. GENERAL MODEL Equation (12. tmax .8) . The second characteristic time. ¯ (0) = 1.6) 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 (12.5) Equation (12. 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.4)) can be obtained by transforming and by 2k −1 ¯ k2 ¯ = [ξ ] k−1 .

10) P (0) = 1 (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. Ac Denoting the area that creates the ratio tmax tc provides the needed tool.11) The analytical solution also can be approximated by a simpler equation as ¯ = [1 − t] P tmax tc (12.14) @Ac .2). Thus.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. The solution is applicable to indirect connection.3) in which outside tube condition is in vacuum. the question remains how the time ratio can be transfered to parameters that can the engineer can design in the system. = 1 as the critical area. In this case only some part of the flow is choked (the later part).9) The analytical solution is applicable only in the case which the flow is choked thorough all the process.2) and 12. The results of a choked case are presented in Figure (12. to create a direct function equation (12. Thus the exit area. 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. A can be expressed as A= The actual times ratio tmax tc @A A Ac Ac (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. These Figures (12. In the literature.220CHAPTER 12.3 demonstrate the imtmax portance of the ratio of tmax tc . In case when the flow chokeless a numerical integration needed to be performed.13) can be expressed as 1 tmax tc = @A tmax tc @A tmax tc (12. When tc > 1 the pressure increases significantly and verse versa.

1: Calculate the minimum required vent area for die casting process when the die volfL ume is 0. 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 . it was found that the influence of the parameter 4f L D on the pressure development in the cylinder is quite small. . A c the following equation Figure (12.15) A . Ac .4). are studied here since it is the dominant c parameter.2 Examples Example 12.9). 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.03[sec]. This parameter represents the ratio between the filling time and the evacuating time. S OLUTION 12. D . This quasi steady state is achieved when the volumetric air flow rate out is equal to the volume pushed by the piston.5) tc is inversely proportional to area. tc ∝ 1/A. The pressure and the mass flow rate are maintained constant after this state is reached. 12. A c 4f L parameter.001[m3] and 4D = 20. From other detailed calculations the author thesis (later to be published on this www. and the friction Parameters influencing the process are the area ratio. The influence is small on the residual air mass in the cylinder but larger on the Mach number.1. It should be noted that critical area. GENERAL MODEL 221 According to equation (11.12. tmax = 0. 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. The effects of the area ratio. The line that represents A = 1 is almost straight. is not function of the time. which was obtained by an integration of equation (12. 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. The direct connection posts more mathematical difficulty because the critical area is not constant and time dependent. The A A . equation (12. A Mexit .4) describes the pressure as a function of the dimensionless A A time for various values of A . A . according to values.1.3 Direct Connection In the above analysis is applicable to indirect connection.potto.1. The required solidification time. Thus.org).14) the tmax is canceled and reduced into tmax tc = @A A Ac (12. This parameter also represents the dimensionless area. It is important to point out the significance of the tmax tc .

This critical area depends on the geometry and the filling time.222CHAPTER 12. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT To continue 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.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.

0 0.0 0.5 1.2 0.2 2.8 Dimensionless Area.4 Figure a 0.8 1. A/Ac 0.2 3.2 0.2 0.6 0.2 3.0 4.5 0. t.2.8 Dimensionless Area.2 1.8 1.0 1.6 4.0 5.0 0.1 0.6 4. A/Ac 0.0 4.1 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 ∇ ∇ 0.0 2.0 5.3 0.4 3.0 0.4 1.4 3.8 0. SUMMARY 223 5.3 0.4 1.0 0.5 1.2 2.6 2.5 0.0 Fig.4 0.0 4fL D ∗ ∇ ∗ o ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.0 4fL D ∗ ∇ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  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 = 5.7 0.8 0.12.7  Dimensionless Time. or.2 1.0 o  = 100. Cylinder Volume Fraction Figure b 0.0 5.9 1.9 1.6 2.2: The pressure ratio as a function of the dimensionless time for chokeless condition .0 1.0 2. 12.2 0.

. ∗ . ∗ 0... A .0 .. . . ... ...7 0.. 1.2 P(t) P(0)  × 1  0. ∗ .1 . . .5 0. ∗ . ∗ ..2 0 0 × ×   × × ×   ×                          × × × 0. .3: The pressure ratio as a function of the dimensionless time for choked condition Fig..3 0..6 . ×... .. Ac . ∗ .6    × × × × × ×   × × × × × 0.224CHAPTER 12. ∗ 1. ..   ..8   0.. 0.. ∗ = 0. . .4  0. . EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 2.. or. ∗ . ∗ .. 12..... ∗ ∗ ∗ .4 0. ∗ .. ∗ . ...2 0.8 1.6 0. . ...5 ∗ . ∗ . 1.0 . ∗ ..0 ∗ .1 0.4 1... 12. t.8 0. . CYLINDER VOLUME FRACTION 1 Fig... ∗ .4: The pressure ratio as a function of the dimensionless time . ∗ . 4..9 DIMENSIONLESS TIME..5 ∗∗ . ∗ ∗ ∗ × × × × × × × × × × × × × × × × × 2 1..∗. ∗ . ∗ .2 ∗ ..

The technique of characteristics is introduced.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. First the physical case which is analyzed is described. A tube with a membrane as shown in Figure 225 .

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

one can view all these cases as three different Fig. a very limited discussion about reflection shocks and collisions of weak shock. The   other type of shock wave is the oblique shock. even the whole book. The author believes that these issues are not relevant to most engineering students and practices. these issues should not be introduced in introductory textbook of compressible flow. Those who would like to obtain more information.R. 727 (1978). variation of the deflection angle from a zero (δ = 0) to a positive value results in oblique shock. Soc. However. 14. etc are presented. Further changing the deflection angle to a negative value results in expansion waves. In the literature oblique shock. and E. waves and asymptotics. 1874 (1994). and Prandtl–Meyer func tion are presented as three separate and different issues. discussion on a normal shock was presented. Keller. Rosales. “Rays. 227 .CHAPTER 14 Oblique-Shock 14. Fluids 6. should refer to J. Von Neumann paradox. Furthermore. However.B.  normal shock. 84.1 Preface to Oblique Shock In Chapter (5). this section attempts to show the boundaries and the limits or connections of these models1 .G. triple shock intersection. Math. Tabak and R. = 0 1 In this chapter.” Phys.” Bull. Clearly. “Focusing of weak shock waves and the Von Neuman paradox of oblique shock reflection. Am. A normal shock is a special type of shock wave. The common representation is done by not showing the boundaries of these models.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.

14. The zero velocity can be found in a jet flowing into a still medium of gas. This shock is commonly referred to as the oblique shock. As it may be recalled. and different inclination angle. The analysis shows that the flow ve. This velocity increase is that both have a maximum point and referred to as the expansion wave. However. ν∞(k ) θmax(k ) Meyer Oblique Function Shock The boundary conditions must match the geometry. When the deflection angle is increased. Thus. As a “no solution” zone. Yet. are examples of forces that create shock. In such a case.2)). This matching can occur only when there is a discontinuity in the flow field. . It has to be pointed out that both the oblique shock and the Prandtl–Meyer function have a maximum point for M1 → ∞.1 Introduction Introduction to Oblique Shock A normal shock occurs when there is a disturbance downstream which imposes a boundary condition on the flow in which the fluid/gas can react only by a sharp change in the flow direction.2: The regions where oblique shock or locity must increase to achieve this rePrandtl–Meyer function exist. zero.228 CHAPTER 14. the flow moves downstream and creates an angle. the gas flow must match the boundary conditions. the maximum point for the Prandtl–Meyer function is much larger than the oblique shock by a factor of more than 2. for a negative deflection angle (in this section’s notaNo Shock zone tion). During this time.2 Introduction to Prandtl–Meyer Function Decreasing the deflection angle re0◦ Prandtl sults in the same effects as before. 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. Prandtl-Meyer function approaches closer to a zero deas opposed to oblique shock analyflection angle. sis. 2 Zero velocity.2 14. However.2.1) which occurs when somewhere downstream a disturbance2 appears. which is around it will be shown in the next chapter. pressure boundary conditions. as discussed in Chapter (1)3 the flow behaves as it does in a hyperbolic field. 14. Notice quirement.Fig. OBLIQUE-SHOCK 14.2. 3 This section is under construction and does not appear in the book yet. Alternatively. the increase in the upstream Mach number determines the downstream Mach number and the “negative” deflection angle. the flow must be continuous. 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 direction of the flow is changed by a shock wave with an angle to the flow. normal shock occurs when a wall is straight/flat (δ = 0) as shown in Figure (14. For information such as the disturbance (boundary condition) to reach deep into the flow from the side requires time.

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

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. Experiments and experience have shown that the common solution is the weak shock. here there are three possible pairs5 of solutions to these equations. θ has to be solved. and the Mach angle.4) Equations (14. the second is the strong shock. (14.3) (14.2) (14. From the geometry it can be observed that tan θ = and tan(θ − δ ) = U2 n U2 t (14.4) are the same as the equations for normal shock with the exception that the total velocity is replaced by the perpendicular components.5) Unlike in the normal shock. 7 Actually this term is used from historical reasons. and the third is an impossible solution (thermodynamically)6 . as sin θ = 6 The solution 5 This M1 n M1 (14.1). The author is not aware of “simple” proof and a call to find a simple proof is needed.7) The above velocity–geometry equations can also be expressed in term of Mach number. requires solving the entropy conservation equation. tan θ U1 n = tan(θ − δ ) U2 n (14.230 The momentum equation reads CHAPTER 14.8) issue is due to R. in which the shock turns to a lesser extent7 . Yet the new relationship between the upstream Mach number. who raised the solution completeness issue. and (14.2). But because the literature referred to only two roots. Menikoff. The lesser extent angle is the unstable angle and the weak angle is the middle solution. the term lesser extent is used. The first is referred to as the weak shock.6) U1 n U1 t (14. . δ . the deflection angle.

8) through (14.17) ρ2 − ρ 1 ρ2 − ρ 1 . OBLIQUE SHOCK M2 n M2 231 sin(θ − δ ) = cos θ = (14. As opposed to the normal shock. It should be noted that although. M2 .9) (14.14. and it follows that the total speed of sound is constant across the (oblique) shock.15) (14. here angles (the second dimension) have to be determined.11).13) The density and normal velocity ratio can be determined by the following equation (14. θ.10) (14. (14.3.14) 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. 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 (14.16) The Rankine–Hugoniot relations are the same as the relationship for the normal shock P2 − P 1 P2 − P 1 =k (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 (14.12) The relationship between the properties can be determined by substituting M1 sin θ for of M1 into the normal shock relationship. The solution from this set of four equations. T1 = T2 .11) M1 t M1 M2 t M2 cos(θ − δ ) = The total energy across an oblique shock wave is constant. and δ . U1 t = U2 t the Mach number is M1 t = M2 t because the temperatures on both sides of the shock are different. is a function of four unknowns of M1 .

One has to be careful to use the normal components of the Mach numbers. Probably. 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. Other combinations of properties provide a relatively simple mathematical treatment. M1 and the deflection angle. 8 This point was pointed out by R. M1 . Clearly. this set of parameters is. the normal shock is a one–dimensional problem. and the negative sign refers to the mirror image of the solution. the most useful properties are upstream Mach number. which create a somewhat complicated mathematical procedure. and the solutions of selected pairs and selected relationships will be presented.18) Equation (14. sin θ must be positive.18) requires that x has to be a real and positive number to obtain a real deflection angle8 . Thus.22) (14. perhaps. the negative root of sin θ must be disregarded 1. only one parameter is required (to solve the problem). 14. and Deflection Angle. Oblique shock is a two–dimensional problem and two properties must be provided so a solution can be found. the relationship between the properties can be determined by using the normal components or by utilizing the normal shock table developed earlier.20) (14. and this will be discussed later.4 Solution of Mach Angle Oblique shock. OBLIQUE-SHOCK 14.4. Menikoff. Thus. δ Again.19) (14. The stagnation temperature contains the total velocity.232 CHAPTER 14.21) (14.1 Upstream Mach Number. thus. the most common and natural to examine. Again. if orientated to a coordinate perpendicular and parallel shock plane is like a normal shock. He also suggested that θ is bounded by sin−1 1/M1 and .

4.24) (14. one root is real and two roots are complex. all the roots are real and unequal. . 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 . 9 The highest power of the equation (only with integer numbers) is the number of the roots. 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 (14.27) 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. θ.29) (14. to explain about complex numbers and imaginary numbers should be included.28) (14.14. From a mathematical point of view. (14. in a quadratic equation there are two roots.30) Only three roots can exist for the Mach angle. Maybe insert an example where imaginary solution results in no physical solution. For example. For the case D = 0. SOLUTION OF MACH ANGLE 233 The solution of a cubic equation such as (14. In the last case where D < 0. 10 A call for suggestions. if D > 0. so that the shock normal component is reduced to subsonic and yet parallel to the inclination angle.26) T = and where the definition of the D is 3 R− √ D (14. D > 0 occurs when no shock angle can be found. all the roots are real and at least two are identical.23) (14.18) provides three roots9 .25) R+ √ D.

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

It can be noticed that the normal shock. and a3 for M1 = ∞. Fig. Once. in the case k = 1. k ).28) after substituting these values of Q and R provides series of roots from which only one root is possible. For a given upstream Mach number.4. Only when the fluid is away from the object (smaller angle) liquid “sees” the object in a different inclination angle. for. the strong shock. in Figure (14. the wedge) in front of it and hence the normal shock occurs. 14. M1 −→ ∞ the maximum deflection angle is calculated when D = Q3 + R2 = 0.4. is just above δmax ∼ π 4 (note that the maximum is also a function of the heat ratio.4) and (14. the fluid “sees” the disturbance (in this case. This root. SOLUTION OF MACH ANGLE 235 14. In the alternative view.4. M1 . and the weak shock coexist. For example. the inclination angle reaches the “maximum potential energy. . The flow is far away from the object and does not “see’ the object. a change in the inclination angle requires a larger energy to change the flow direction. The simple procedure For example.2 When No Oblique Shock Exist or When D > 0 Large deflection angle for given.” a change in the flow direction is no longer possible.14.4: Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. This can be done by evaluating the terms a1 . a2 . M1 The first range is when the deflection angle reaches above the maximum point. 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. This different inclination angle is sometimes referred to as an imaginary angle.0.5). the imaginary angle is shown.

Chapman. one can conclude that the maximum Mach angle is only a function of the upstream Much number. 15 At first. θ.12). M1 . .5: The view of a large inclination angle from different points in the fluid field. it was seen as C. At this stage it is not clear who was the first to propose it. the deflection angle is a function of the Mach angle and the upstream Mach number. Since in that equation there appears only the heat ratio k . When the right terms defined in (14.21). and (14.29). and Mach number.20)-(14.31) This equation and its twin equation can be obtained by an alternative procedure proposed by someone15 who suggested another way to approach this issue.12) and equating the results to zero creates relationship between the Mach number. English mathematician to be the creator but later an earlier version by several months was proposed by Bernard Grossman.236 CHAPTER 14. M1 and the maximum Mach angle. θmax is a function of only these parameters. (14. It can be noticed that in equation (14. M1 . M1 .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. The Procedure for Calculating The Maximum Deflection Point The maximum is obtained when D = 0. J. 14. OBLIQUE-SHOCK The fluid doesn’t ’’see’ the object M∞ } } } The fluid ‘‘sees’’ the object with "imaginary" inclanation angle Intermediate zone The fluid "sees" the object infront Fig. This can be shown mathematically by the argument that differentiating equation (14. Thus.

The general solution of equation (14. the maximum deflection angle can be computed.31) is (k + 1)M1 2 + 1 + M1n = (M1 2 M1 2 (k + 1)2 + 8(k 2 − 1) + 16(1 + k ) √ 2 k (14. the 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. which makes ◦ θ = 1. It can be noticed that the Maximum Deflection Mach Number’s equation is also a quadratic equation for M1n 2 .34) and then solving equation (14. for large Mach numbers.18 or θ = 67. Thus.79 . then the Mach angle can be easily calculated by equation (14.33) Equation (14.4.32) Because tan is a monotonous function.36) .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. The numerator of equation (14. To compare these two equations the simple case of Maximum for an infinite Mach number is examined. It must be pointed out that similar procedures can also be proposed (even though it does not appear in the literature). SOLUTION OF MACH ANGLE The differentiation of the equation (14. the maximum appears when θ has its maximum. it is sufficient to equate the numerator to zero to obtain the maximum. a derivative can be taken with respect to M1 .35) k+1 Hence. A simplified case of the Maximum Deflection Mach Number’s equation for large Mach number becomes M1n = k+1 M1 2k for M1 >> 1 (14. Thus.12). The nominator produces a quadratic equation for sin2 θ and only the positive value for sin2 θ is applied here. Thus.32) is zero at different values of the denominator. taking the derivative with respect to θ. 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 .8).34) provides a solution for Mmax . Once M1n is found.33) should be referred to as the maximum’s equation.14. Instead. With the value of θ utilizing equation (14. Note that this procedure does not require an approximation of M1n to be made. d tan δ =0 dM1 (14. the Mach angle is sin θ = 2k .

Start with equation (14. a2 . This typical example is for those who like mathematics.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 + .38) 3 Q= 2M1 2 +1 M1 4 − 9 M1 2 +2 M1 2 3 (14. Hint. Example 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 (14.31) and neglect all the terms that are relatively small. This can be partially demonstrated by evaluating the terms a1 . first the transition line in which D = 0 has to be determined. 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).39) .37) for R and Q becoming 9 R= 2 M1 2 +2 M1 2 2M1 2 +1 M1 4 2 (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 (14.238 CHAPTER 14. This can be achieved by the standard mathematical procedure of equating D = 0. Thus. S OLUTION under construction The case of D ≥ 0 or 0 ≥ δ The second range in which D > 0 is when δ < 0. The analysis shows regardless of the value of the upstream Mach number D = 0 when δ = 0.

17 There 16 A .28) results in D = 0.Fig. Utilizing the symmetry and antisymmetry of the qualities of the cos and sin for δ < 0 demonstrates that D > 0 regardless of Mach number. this identity can be demonstrated by checking several points for example.39) into equation (14.2 ysis is no. the wall does not emit any signal to the flow (assuming zero viscosity). Fluids 10 (1) .” 206 Phys. And according to 9 this analysis. paper by Rosles.0.  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. are several papers that attempt to prove this point in the past. M1 = 1.6) is provided for the following demonstration.6: The various coefficients of three different tradicts the common approach. But for non ideal gas (real gas) this solution is only an indication. mathematical challenge for those who like to work it out. 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 .4. “Caustics of weak shock waves. Table (14.40) The author is not aware of any analytical demonstration in the literature which shows that the solution is identical to zero for δ = 016 . in the literature. no Mach wave 2. Mach numbers to demonstrate that D is zero Nevertheless.28) provides the equation to be solved for δ . Tabak.16 words. Nevertheless. 14. this proof became trivial. Hence. In other ∞ -1 0 . there was a transition from one kind of discontinuity to another. Substitution of all the above values into (14. there are several papers suggesting zero strength Mach wave.14. which con.0 3 0 16 can be generated from the wall 1 with zero deflection. 2. In the range where δ ≤ 0. 18 See for example.0 -3 -1 . The question of singular point or zero Mach wave strength are only of mathematical interest. with a positive large deflection angle. ∞. Note that. January 1998.. Once this analytical solution was published. 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. SOLUTION OF MACH ANGLE 239 Substituting the values of Q and R equations (14. others suggest a singular point18 . in the previous case.38) (14.

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

θ.14. M1 .4.0005 0 0. are far much simpler and a unique solution exists. Oblique Shock 3 2.0 10. M1 . and Shock Angle. θ δ 20 30 The solution for upstream Mach number. especially in acoustic engineering.41) .0 -0. and shock angle.001 0.0 δ 20.5 1 0.3 Upstream Mach Number.4. 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. and exit Mach number for M1 = 3 0 10 14. but for most cases it should be ignored.5 2 1.001 k = 1 4 Mx=3 0 10 20 30 Wed Jun 22 15:03:35 2005 Fig. 14. shock angle.8: The calculation of D (possible error).0 30.5 0 90 80 70 60 50 40 30 20 10 0 -0.0005 Myw Mys θs θw 0. SOLUTION OF MACH ANGLE 241 mation of weak weak wave with nonzero strength has engineering applicability in a very limited cases.

is limited.4 Given Two Angles. is limited between ∞ and 1/ sin2 θ.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 (14.47) (14. Examining equation (14.42) 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. OBLIQUE-SHOCK 2 cot θ(M1 2 sin2 θ − 1) 2 + M1 2 (k + 1 − 2 sin2 θ) (14. M1 and θ.9). δ and θ It is sometimes useful to obtain a relationship where the two angles are known.49) .4.44) (14. The range of given θ. δ . M1 is M1 2 = 2(cot θ + tan δ ) sin 2θ − (tan δ )(k + cos 2θ) (14.46) (14. 14. is unique. has to be in the range of sin−1 (1/M1 ) ≥ θ ≥ (π/2) (see Figure 14.45) (14. the possible range deflection angle. upstream Mach number M1 .41) shows that the shock angle. θ . The first upstream Mach number.242 or tan δ = CHAPTER 14.48) Even though the solution for these variables.

θ.14. δ must also be small as well.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 ) (14.50) For a large upstream Mach number M1 and a small shock angle (yet not approaching zero). 14. Shock angle π θ= 2 θmax ∼ π 2 θ=0 Fig.0 < M1 < ∞ Defection angle strong solution θmin = sin −1 1 M1 supersonic weak soution possible solution no solution zone θ.4.52) The results are consistent with the initial assumption which shows that it was an appropriate assumption.51) (14. the deflection angle. SOLUTION OF MACH ANGLE subsonic weak solution 243 1. .41) can be simplified into k+1 θ∼ δ = 2 (14. Equation (14.

” Another geometry that can be considered as two–dimensional is the cone (some referred to it as Taylor–Maccoll flow). The analysis also provides additional value of the expected range. the first solution is thermodynamical unstable. for steady state and no known experiments showing that it exist can be found in the literature. is neglected. 14.10: Schematic of finite wedge with zero be area where current solution can be angle of attack. a Even finite wedge with limiting wall can be considered as an example for this discussion if the B. Clearly.4. 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. this solution is also act as a “reality check” to any numerical analysis. The end of the wedge must have a different configuration (see Figure (14. Even though.10)). All the literature. the strong shock is unstable. reports that only a weak shock is possible. OBLIQUE-SHOCK 14. the symmetrical nature of the cone creates a semi–2D problem. The author is not aware of any such analysis in literature. The . the analysis for the middle section produces a close result to reality (because of symmetry). ate i d i rm range inte for any finite length of the wedge. Additionally. dimensional analysis is not completed. for most engineering applications. In this case there are no edge effects and the geometry dictates slightly different results. { { { { { 14. The mathematics is much more complicated but there are three solutions. The common approach is to carry out numerical analysis. the cone is a three–dimensional problem. a simple tool is sufficient for limit accuracy. As before. In spite of the small area were the current solution can be assumed. known to the author. the numerical works require many times a “reality check. at least.5 Flow in a Semi–2D Shape The discussion so far was about the straight infinite long wedgea which is no shock flow direction ysis anal a “pure” 2–D configuration. b assumed as correct . In spite of recent trends.4.L. It is recommended that this issue be skipped and the time be devoted to other issues. As opposed to the weak shock. the sis naly al a norm range analysis needs to account for edge efno shock fects.244 CHAPTER 14. b At this stage.6 Small δ “Weak Oblique shock” This interest in this topic is mostly from an academic point of view. The dimensional analysis shows that only the doted area to Fig. Yet. Experimental and analytical work shows that the weak solution is the stable solution and a discussion is provided in the appendix of this chapter.

and small normal upstream Mach numbers. However. In this theory.L. it should be covered in this book. 1 M1 2 − 1 (14.14. after the explicit analytical solution has been provided. the oblique shock is an accumulation of many small (or again weak) “weak shocks.5152 θmax 76. 14.4. Mx 1.97131 δmax 1..11: A local and a far view of the imity to the wall. Yet. 14. 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 θ layer (B. M1 ∼ 1+ . under construction. oblique shock. 14.1000 My 0. the author admits that as long as there are instructors who examine their students on this issue.1: Table of maximum values of the oblique Shock k=1. the boundary layer increases the zone where a continuous δ flow exists. In reality. In fact. δ .8 Maximum Value of Oblique shock Table 14.” However.53) tan θ = .4.7 Close and Far Views of the Oblique Shock In many cases. the far view shows a sharp transition. these small “shocks” are built or accumulate into a large and abrupt change (shock).11). studying this topic seems to come at the expense of other more important topics. the oblique shock is a weak “weak oblique” shock. From the far view. For small deflection angles.4. the close proximity view provides a continuous turning of the deflection angle..4 The maximum values are summarized in the following Table . In larger distance from the wall.) does not enter into the calculation. 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. The boundary layer reduces the upstream flow velocity and therefore the shock does not exist at close proxFig.2762 . At the local view close to the wall. the shock becomes possible.

94925 0.5676 65.5000 1. 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.6074 64. .92224 0.3000 1.1619 44.4000 3.8443 65.3275 36.92165 0. this assumption might not be sufficient and different analysis is needed.6515 17.92478 0.6621 9.1833 21.6814 30.7532 64.0399 65.93629 0.98183 0.4290 θmax 71.0000 6.1177 42.5875 34.0000 5.95049 0.4398 43.7972 65.9020 67.95897 0.99440 0.95435 0.2503 67.9735 26.7739 41.1028 28.0000 8.1675 22.99337 0.4000 1.3673 67.99047 0.4419 It must be noted that the calculations are for the perfect gas model.0000 7.3059 38.3934 37.4.4272 12.9000 2.9442 6.6000 1.96942 0.2000 2.91941 0.99559 δmax 3.6465 64.8137 32.9 Detached shock When the mathematical quantity D becomes positive.1127 14.8000 4.2546 43.0000 My 0.7908 44.6000 2.9555 69.91871 0.96630 0.2000 1. In some cases.4 Mx 1.0000 9.93747 0.5671 66. OBLIQUE-SHOCK Table 14.9087 66.7593 65. Since the flow “sees” the obstacle. Henderson and Menikoff20 suggested a procedure to calculate the maximum deflection angle for arbitrary equation of state21 . 179–210.9668 64.98714 0.0464 66.7023 66.94387 0.2000 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.96335 0.0000 2.246 CHAPTER 14.0000 3.0000 10.3645 67.0734 35.0119 19.1: Maximum values of oblique shock (continue) k=1.4144 65.8000 3.7000 1.93083 0.4000 2. for large deflection angle. 14.92683 0.1196 67.6000 3.5787 65. there isn’t a physical solution to an oblique shock.0922 38.91997 0.6934 64.8000 1.2309 65.3066 64.97214 0.

Upstream U∞ Figure 14. The constant and the function are different for different geometries. E Strong Shock F > @ . The flow in zone A has to go through some acceleration to became supersonic flow. 14. The explanation to such a phenomenon is above the level of this book (where is the “throat” area question22 . In such a case.CG isn’t important for this discussion).12: The schematic for a round–tip bullet C the flow must be supersonic. the increase in the upstream Mach results in a decrease of the detachment distance. a wedge or 22 See example 14. <> = > ?1@ ? ACBD< . Even though this topic has few applications. This shock is referred to as the detach shock. While at zone Fig. it can be explained as the subsonic is “sucked” into gas in zone C. Yet. In the typical example.4. Thus.14. Regardless of the explanation. Larger shock results in a smaller detachment distance. 14. alternatively. a graph and a general explanation to engineers is provided. Nevertheless.4. The zone A is zone where the flow must be subsonic because at the body the velocity must be zero (the no–slip condition). As a general rule.5.10 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. this phenomenon has a larger impact for a relatively smaller supersonic flow.54) where f (M∞ ) is a function of the upstream Mach number which tabulated in the literature. . The detached shock’s distance from the body is a complex analysis and should be left to graduate class and researchers in this area. SOLUTION OF MACH ANGLE 247 from the body. or. The in a supersonic flow weak oblique shock is predicted to flow around the cone. these calculations can be summarized by the flowing equation detachment distance = constant × (θ − f (M∞ )) body thickness w ea k sh oc k (14. 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 : . some might be used in certain situations which the author isn’t aware of. the gas must go through a shock. 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. the flow becomes “blinder” to obstacles.12 exhibits a round–tip bulzone B Subsonic Area θ let with a detached shock.

for example. Such configurations seem sometimes at first glance not related to the detached shock issue. for U B ikj every upstream Mach number there C θ2 Slip Plane is only one maximum deflection anA δ2 gle. In this situation. there are configurations in which a detached shock occurs in design and engineers need to take it into consideration. the maximum deflection angle decreases with a decrease the Mach numA sub sonic ber. the second oblique shock complicates the situation and the second oblique shock can cause a detached shock.14: The “detached” shock in a complicated configuration sometimes reoblique shock.. δmax = f (M1 ) (14. hence δ1 = δ2 . Additionally. This detached shock ferred to as Mach reflection manifested itself in a form of curved shock (see Figure 14. The Mach number decreases after every shock. However. OBLIQUE-SHOCK a cone moves into a still medium or gas flows into it. yet the mathematics is somewhat complicated. . However. Thus. it can be observed for Fig. 14. It seems that the most straightforward method is the following: (a) Calculate M1 B . ThereU B C fore. this situation causes the detached shock to appear in the second Fig. Consider. due to the symmetry flow a slip plane angle can be guessed to be parallel to original flow.55) Additionally. 14.13). The maximum deflection angle in this case is.13: The schematic for a symmetrical suca maximum oblique shock that a contion section with Mach reflection stant deflection angle decrease of the l/m#n o prqrsutwvrlyx zyt Mach number results in an increase of δ1 Mach angle (weak shock only) M1 > θ1 M2 =⇒ θ1 < θ2 . as before.14). 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. This situation is known in the scientific literature as the Mach reflection. a symmetrical suction section in which the deflection angle is just between the maximum deflection angle and above half of the maximum deflection angle. If the deflection angle exceeds the maximum possible.e. at least two oblique shocks occur and after their interaction is shown in Figure (14. The analysis of this situation is logically very simple. No detached shock issues are raised when only the first oblique shock is considered. The calculations for such a case can be carried out by several approaches.248 CHAPTER 14. only a function of the upstream Mach number. a detached shock occurs.

5686 θs 1. This procedure can be extended to calculate the maximum incoming Mach number. one must be aware that there are zones of dual solutions in which sharp shock line coexists with a curved line. POTTO–GDC can be used to calculate this ratio.2: Air flows at Mach number (M1 ) or Mx = 4 is approaching a wedge. θ2 . For engineering purposes when the Mach number reaches this value.14.31) equation (c) Calculate the deflection angle.0000 My 0. this zone increases as Mach number increases. Using the geometry provided in the photo.15) is exposed to supersonic flow and create an oblique shock.7738 θmax 66. The wedge angle that satisfies this requirement is by equation (14. calculate the weak.5◦ . and the respective shock angles.34907 Example 14. For example.0000 My s 0.97234 δmax 38.28) or the GDC which yields the following results Mx 4. it can be ignored. M1 by checking the relationship between the intermediate Mach number to M1 . SOLUTION OF MACH ANGLE (b) Calculate the maximum deflection angle. predict at which Mach number was the photo taken based on the assumption that the cone is a wedge. The maximum values are: Mx 4. .0407 To obtain the results of the weak and the strong solutions either utilize the equation (14. at Mach 5 this zone is 8. utilizing (14.56660 δ 0. S OLUTION The maximum wedge angle for (Mx = 4) D has to be equal to zero. the strong Mach numbers.3: A cone shown in Figure (14. In general.11 Oblique Shock Examples Example 14. What is the maximum wedge angle at which the oblique shock can occur? If the wedge angle is 20◦ .4.4635 θw 0.12) 249 (d) Use the deflection angle.4.48523 My w 2. 14. Is the shock shown in the photo weak or strong shock? Explain. δ2 = δ1 and the Mach number M1 B to calculate M1 B . In discussing these issues. δ2 utilizing equation (14. Note that no maximum angle is achieved in this shock.28) (a side to the case proximity of δ = 0).

88737 Because the flow is around the cone it must be a weak shock.4522 71. . This photo is courtesy of Dr. the cone half angle is 15◦ and the Mach number is 2.43◦ and the shock angle is 30. Grigory Toker.2. 14. According to his measurement. a Research Professor at Cuernavaco University of Mexico. OBLIQUE-SHOCK θ δ Fig.56543 2. Even if the cone was a wedge.0990 14. M1 3. the shock would be weak because the maximum (transition to a strong shock) occurs at about 60◦ . With given two angles the solution can be obtained by utilizing equation (14.49) or the Potto-GDC.0143 30. S OLUTION The measurement shows that cone angle is 14.250 CHAPTER 14.099◦. Note that the Mach number is larger than the one predicted by the wedge.15: Oblique shock occurs around a cone.2318 My s My w θs θw δ P0 y P0 x 0.4300 0.

(1998) pp.0 Mx Thu Jun 30 15:14:53 2005 My θ δ Fig.5 0 90 80 70 60 50 40 30 20 10 0 2.4.” Journal of Fluid Mechanics 366.16: Maximum values of the properties in an oblique shock 14.0 6.0. there is general proof that regardless to the equation of state (any kind of gas).0[bar]. To illustrate the design significance of the oblique 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.0 10. the following example is provided.0 3. In such a case.5 1 0.0 9. P = 1. 14.5 2 1. the flow can be brought to a subsonic Fig.0 4. and T = 17◦ C. a In fact. 14. SOLUTION OF MACH ANGLE 251 k=14 Oblique Shock 3 2.17: Two variations of inlet suction for flow just below M = 1. the entropy is to be minimized through a series of oblique shocks rather than through a single normal shock. Example 14.0 5. Clearly.0 7.18 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 (14.14. 179–210. supersonic flow there is less entropy production (less pressure loss). with a proper design.4: The Section described in Figure 14. Compare the different conditions in the two . For details see Henderson and Menikoff “Triple Shock Entropy Theorem.17))a .0 8.4.

Assume that only a weak shock occurs.2098 0.9831 51.2346 76. no input explanation is given.0000 P0 y P0 x 0.252 CHAPTER 14.99445 and the additional information by the minimal info in the Potto-GDC is Mx 2.5000 0.0000 Py Px Ty Tx P0 y P0 x 1.96524 And the additional information is 23 The results in this example are obtained using the graphical interface of POTTO–GDC thus.72087 In the oblique shock. the new angle is 7◦ + 7◦ with new upstream Mach number of Mx = 1.4) S OLUTION The first configuration is of a normal shock for which the results23 are Mx 2. 14.7498 resulting in Mx 1.1931 0.0000 My 0.99445 In the new region.7498 85.0000 My s My w θs θw δ 7.7021 36.58974 1. OBLIQUE-SHOCK different configurations.7498 My s My w θs θw δ P0 y P0 x 0. °²± ™/š›DœužyŸ• ¡š¡¢•£ 1 neglect the detached distance 7◦ ¤¥/¦ § ¨e©/ª¬«C­/¤¡®•¯« 2 ³µ´ 3 4 Normal shock 7◦ Fig.2098 1.18: Schematic for Example (14. In the past the input file was given but the graphical interface it is no longer needed.57735 Ty Tx ρy ρx Py Px P0 y P0 x 1. .0000 0.6875 2.71761 1.5549 14.2485 1.0000 My w θw δ 7. the first angle shown is Mx 2.6667 4.7498 36.

2346 My 0.0000 Py Px Ty Tx P0 y P0 x 253 1.3151 4.99549 An oblique shock is not possible and normal shock occurs. 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. The half wedge angle is 10◦ .82141 Ty Tx ρy ρx Py Px P0y P0 x 1.4.2626 × 1.90506 1. Example 14.19: Schematic for Example (14.6116 × 1.7498 My w θw δ 7.45115 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. What is the requited “throat” area ratio to achieve acceleration from the subsonic region Fig.5088 41.98903 × 0.1250 0.5000 My 0.6116 0. the loss in the normal shock is above than 31% of the total pressure.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 3.45115 Ty Tx ρy ρx Py Px P0y P0 x 3.98903 With two weak shock waves and a normal shock the total pressure loss is P04 P03 P02 P04 = = 0.4 yields M 0.2626 1.2609 14.1853 0.6147 P1 P3 P2 P1 The loss in this case is much less than in a direct normal shock. In such a case. 14.9496 P0 1 P0 3 P0 2 P0 1 The static pressure ratio for the second case is P4 P4 P3 P2 = = 1.4458 0.99445 = 0. the results are: Mx 1.86966 1. .4018 1.21295 Now utilizing the isentropic relationship for k = 1.2574 Thus the area ratio has to be 1. In fact.14.4458.5: 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 3.5 (see Figure 14.1497 1.19).285 = 2.8770 1. SOLUTION OF MACH ANGLE Mx 1.96524 × 0.96089 0.

OBLIQUE-SHOCK D 4 Slip Plane The effects of a double P3 = P 4 B wedge are explained in the 3 government web site as weak weak oblique oblique shock shown in Figure (14.40669 0.47641 2.0.48013 2.7049 88.13665 The combined shocks AB and BC provide the base of calculating the total pressure ratio at zone 3.8912 22.7008 87.13632 0.7049. The total pressure ratio at zone 2 is P02 P02 P01 = = 0.6: CHAPTER 14. Find the required angle of shock BE.9356 0. 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 14.48610 2.997731283 P00 P01 P00 .99105 The transition for shock AB is Mx 3. Perform the calculation for M1 = 3.99894 × 0.20).04290 0.5990 0.0000 P0 y P0 x 0. 14.04263 0. Then.40596 0.99879 = 0.1978 3. Assume that there are no boundary layers and Fig. The conditions for this shock are: Mx 3.8482 My s My w θs θw δ 3.0000 My s My w θs θw δ 3.9476 21.254 Example 14.8482 88.1854 0. explain why this description has internal conflict.7049 2.10500 0.20: Schematic of two angles turn with two weak shocks all the shocks are weak and straight.99894 And the isentropic relationships for M = 2.8807 23.0000 My s My w θs θw δ 6.10548 3.7080 0.20). 1 S OLUTION The shock BD is an oblique shock with a response to a total turn of 6◦ .0000 P0 y P0 x 0. 2.99879 For the shock BC the results are Mx 2.7008 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.7008 are M 2.0000 P0 y P0 x 0.

04290 × 0.4.715320879◦ The deflection angle for such shock angle with Mach number is Mx 2. The increase has to be P3 /P2 = 0. the pressure at 4 has to be P4 P4 P0 4 = = 0.72 δ P0 y P0 x 0. θ can be calculated from θ = sin−1 1.0019 1.99920 Ty Tx ρy ρx Py Px P0y P0 x 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. the angle for BE shock which cause this pressure ratio needs to be found. In reality the flow of double wedge will produce a curved shock surface with several zones.99105 = 0. To do that.7037 0.042516045 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. P0 P2 P2 = 2 = 0.4).49525 2.04263 = 0. T1 = 27◦ C. SOLUTION OF MACH ANGLE On the other hand.7: Calculate the flow deflection angle and other parameters downstream when the Mach angle is 34◦ and P1 = 3[bar].997731283 × 0.0005 1.0008/2. The flow in zone 3 will flow into the wall in about 2.14.0013 1. the flow behaves as only one theoretical angle of 6◦ exist. check whether the pressure at 2 is above or below or above the pressure (ratio) in zone 4. Assume k = 1. Only when the flow is far away from the double wedge.7049 My s My w θs 0. Example 14.042516045/0.00000 The shock angle.4 and R = 287J/KgK .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 5.001867743 To achieve this kind of pressure ratio the perpendicular component has to be Mx 1.0 θw 21. and U1 = 1000m/sec.7049 = 21.0008 My 0.7◦ .042436789 = 1.026233 1.

8575 × 3 = 8.15) and (14.8: For Mach number 2.8124 δ 11.5 and wedge with a total angle of 22◦ .57[bar] And the velocity √ √ Un2 = My w kRT = 2. M1 .61 From the Table (5.5K and the same for the pressure P2 = 2. calculate the ratio of the stagnation pressure.0000 P0 y P0 x 0.8575 0.8800 My s My w θs 0.88 sin(34.88 = 1.3949 2.0995 θw 32.3949 × 300 ∼ 418.128 1.66545 Ty Tx ρy ρx Py Px P0 y P0 x 1.0) = 1.6100 My 0. Mx 1. has to be determined M1 = √ U1 1000 = 2.256 CHAPTER 14.53431 2.5000 My s My w θs 85. Utilizing GDC for Mach number 2.89145 The temperature ratio combined upstream temperature yield T2 = 1. Yet.89127 The relationship for the temperature and pressure can be obtained by using equation (14.48269 2.1) or GDC the following can be obtained. M1n = M1 ∗ sin θ = 2.0443 0.6[m/sec] Example 14.5 and the angle of 11◦ results in Mx 2. the Upstream Mach number. OBLIQUE-SHOCK S OLUTION The Mach angle of 34◦ is below maximum deflection which means that it is a weak shock.0 θw 34.13) or simply converting the M1 to perpendicular component.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 2.1280 0.96873 .4 × 287 × 418.5 = 872.00 δ 15.0485 2.78 P0 y P0 x 0.

3573 2. It is clear that the flow must be parallel to the wall. Utilizing the table or the Potto–GDC. utilizing equation (14. is obtained. S OLUTION This kind of problem is essentially two wedges placed in a certain geometry. With these two quantities.14.6854 0.9: What is the maximum pressure ratio that can be obtained on wedge when the gas is flowing in 2. SOLUTION OF MACH ANGLE 257 Example 14. the shock is weak shock. Maximum Deflection Mach’s equation was used to calculate the normal component of the upstream. 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). For a weak shock the maximum pressure ratio is obtained at the deflection point because it is closest to a normal shock. To obtain the maximum point for 2.94021 4. Example 14. then the Mach angle was calculated using the geometrical relationship of θ = sin−1 M1n /M1 .10: Consider the schematic shown in the following figure.7822 δ 29.7974 Py Px Ty Tx P0 y P0 x 0. δ . what is the difference? S OLUTION It has to be recognized that without any other boundary condition.12) the deflection angle. either use the Maximum Deflection Mach number’s equation or the Potto–GDC Mx 2. the following can be obtained: . For the first shock. 3 stream line 2 1 θ M1 = 4 δ Assume that the upstream Mach number is 4 and the deflection angle is δ = 15◦ .5 Mach number.60027 In these calculations.5 Mach without any close boundaries? Would it make any difference if the wedge was flowing into the air? If so. the upstream Mach number is known together with deflection angle.4.5000 My max θmax 64.

11: A similar example as before but here Mach angle is 29◦ and Mach number is 2. S OLUTION Here the Mach number and the Mach angle are given.0026 P3 P2 P3 T1 T1 T2 = = 1.7344 × 1.9290 My w θw δ Py Px Ty Tx P0 y P0 x 2. Again calculate the downstream ratios after the second shock and the deflection angle.90041 With the combined tables the ratios can be easily calculated.0629 15.3575 0.7985 1.7344 0.0000 My s My w θs CHAPTER 14.8500 My s My w θs 0.90041 and the additional information is Mx 2.96263 .0 θw 29.5764 0.2028 32.9290 85. With these values the following can be obtained: Mx 2.258 Mx 4.9290 27.7822 15.2808 32. the second deflection angle is also 15◦ . Note that hand calculations requires endless time looking up graphical representation of the solution.0000 1.6695 = 3. P1 P1 P2 = = 1. Utilizing the POTTO–GDC which provides a solution in just a few clicks.0000 1.51367 2.85.0629 15.46152 2.80382 With a Mach number of M = 2.51 P0 y P0 x 0.632 T3 T2 T3 Example 14.929. OBLIQUE-SHOCK θw δ P0 y P0 x 0.5851 27.80382 And the additional information by using minimal information ratio button in Potto– GDC is Mx 4.0000 0.0000 My w θw δ Py Px Ty Tx P0 y P0 x 2.7822 15.9290 My s My w θs θw δ P0 y P0 x 0.2028 84.00 δ 10.0000 0.5764 = 2. With these pieces of information by utilizing the Potto-GDC the following is obtained: Mx 2.48469 2.7985 × 1.6695 1.

96263 With the deflection angle of δ = 10.5131 1.3582 × 1.49901 And the additional information is .0000 0.8500 My w θw δ Py Px Ty Tx P0 y P0 x 2.54894 1.97 P3 P2 P3 T1 T1 T2 = = 1.5100 0.3268 0.4.97569 P1 P1 P2 = = 1.3984 1.3575 29.3268 ∼ 1.0058 0.41523 3.8000 5. What is the deflection angle in this case? S OLUTION For the normal shock the results are Mx 5.3984 ∼ 1.12: Compare a direct normal shock to oblique shock with a normal shock.0590 10.0000 My 0.0000 10.0000 My s My w θs 0. 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◦ .51 the so called reflective shock gives the following information Mx 2.8021 T3 T2 T3 Example 14. SOLUTION OF MACH ANGLE 259 and the additional information by utilizing the minimal info button in GDC provides Mx 2.17 P0 y P0 x 0.0590 10.9419 34.9419 84.3582 0.41523 Ty Tx ρy ρx Py Px P0y P0 x 5.0 θw 30.14.3575 My s My w θs θw δ P0 y P0 x 0.5100 1.4089 1.06172 While the results for the oblique shock are Mx 5.9398 34.97569 and the additional information of Mx 2.3575 My w θw δ Py Px Ty Tx P0 y P0 x 1.00 δ 20.0000 29.4089 × 1.

3740 0.13)).8625 10.6375 × 10. .374 ∼ 27.13) For upstream Mach number of 5 and deflection angle of 12◦ and 15◦ .1736 2. S OLUTION The first two zones immediately after are computed using the same techniques that were developed and discussed earlier. OBLIQUE-SHOCK Py Px Ty Tx P0 y P0 x 3.32671 The pressure ratios of the oblique shock with normal shock is the total shock in the second case.260 Mx 5. For the first direction of 15◦ and Mach number =5. explain whether the slip angle is larger or smaller than the difference of the deflection angle. Example 14.47485 Ty Tx ρy ρx Py Px P0 y P0 x 2.5141 0.49901 The normal shock that follows this oblique is Mx 3.6375 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. Based on these calculations. 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.0058 30.6858 ∼ 6.5141 × 2.75 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.6858 3.0000 My w θw δ CHAPTER 14.0000 20.0058 My 0.13: A flow in a tunnel ends up with two deflection angles from both sides (see the following Figure (14.36 P3 P2 P3 T1 T2 T1 = = 2. P1 P1 P2 = = 2.

5040 My s My w θs θw δ P0 y P0 x 0. it is assumed that the slip angle moves half of the angle to satisfy both of the deflection angles (first approximation).5◦ = 13.0000 My s My w θs θw δ P0 y P0 x 261 0.88496 And in zone 4 the conditions are due to deflection angle of 13.8006 21.88496 with the additional information Mx 3.9791 1.0000 My s My w θs θw δ P0 y P0 x 0. It is to say.3217 15.2845 12.43016 3. that only one deflection angle exist.0739 24.5◦ which results in Mx 3. for the analysis.3217 15.43914 3.6668 13.0000 1.5656 0. For the two different deflection angles.5000 1. the slip angle has two extreme cases.0000 1.69317 And the additional conditions are Mx 5.6986 27.8006 86.6963 1.9122 21. The first case is where match lower deflection angle and second is to match the higher deflection angle.69317 For the second direction of 12◦ and Mach number =5. This non–continuous velocity profile can occur in our model because it is assumed that fluid is non–viscous.4.5◦ and Mach 3.6986 85. Under this assumption the conditions in zone 3 are solved by looking at the deflection angle of 12◦ + 1.80600 And the additional conditions are Mx 5.2845 12. If the two sides were equal because of symmetry the slip angle is also zero.0000 0.8006 .0000 My w θw δ Py Px Ty Tx P0 y P0 x 3.6668 13.80600 The conditions in zone 4 and zone 3 have two things that are equal.5000 0.5040 86.6625 0. It has to be noticed that the velocity magnitudes in zone 3 and 4 do not have to be equal.5040 My w θw δ Py Px Ty Tx P0 y P0 x 2. SOLUTION OF MACH ANGLE Mx 5.9238 0.6247 1.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3.47413 2.14. In this case. Mx 5.5040 24.6819 27. They are the pressure and the velocity direction.0000 0.

Example 14. in oblique. the pressure at zone 3 has to be reduced.9035 26. OBLIQUE-SHOCK θw δ P0 y P0 x 0.9316 26.6038 To reduce the pressure ratio the deflection angle has to be reduced (remember that at weak weak shock almost no pressure change).3226 13.5 Summary As with normal shock.6963 1. 14.5000 0. will the oblique shock with first angle of 15◦ and then 12◦ or opposite order make a difference (M = 5)? If not what order will make a bigger entropy production or pressure loss? (No general proof is needed).5◦ to a larger number. S OLUTION Waiting for the solution 14. Thus.6247 × 1. There are researchers which suggest that the numerical work is the solution.46259 2. 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. However. M1 is always greater than 1.18192 P4 P2 P0 P1 P4 1.3226 13.5000 1.8006 My w θw δ Py Px Ty Tx P0 y P0 x 2.6577 1. Consider the same values as previous analysis.14: The previous example gave rise to another question on the order of the deflection angles. the oblique shock with upstream Mach number.9035 85. the . To reduce the pressure the angle of slip plane has to increase from 1.6038 0.86179 From these tables the pressure ratio at zone 3 and 4 can be calculated P3 P2 P0 P1 1 P3 1 = = 1.8006 My s My w θs CHAPTER 14.86179 with the additional information Mx 3.262 Mx 3.13 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.4. The optimum deflection angle is a function of the Mach number range in which the suction section is operated in. as oppose to the normal shock.9791 ∼ 1.

When D > 0. When D is positive there is no physical solution and only normal shock exist. Instinctively. the analysis for the oblique shock wave is carried out. the stable and unstable can be recognized.6. Yet.6 Appendix: Oblique Shock Stability Analysis The stability analysis is an analysis which answers the question of what happens if for some reason. the discussion is more complicated but similar analysis can be applied to the strong shock. if the answer is negative. two is weak shock. As it results. The only difference is that here. This change complicates the explanation and is omitted in this section. 14. then the situation is referred to as unstable. there is a possibility of no two–dimensional solution resulting in a detached shock case. When D is equal to zero. An example to this situation. In this example only the weak solution is explained.14. δ there could be three solutions: the first one is the “impossible” solution in the case where D is negative. and three is strong shock. The perpendicular component of the downstream Mach number. The similar analysis can be applied to 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. and upstream Mach number. for large deflection angle. On the other hand. Given M1 and the deflection angle. 14.21). M2 could be larger or smaller then 1. M1 n is always smaller than 1. Unstable Stable . APPENDIX: OBLIQUE SHOCK STABILITY ANALYSIS 263 downstream Mach number. for example. is a ball shown in the Figure (14. If the answer turns out to be that Fig. Of course. 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. in the analysis the strong shock results in an elliptic solution (or region) as opposed to a hyperbolic in weak shock. deflection angle. the shock angle. there are more than one parameter that can be changed. a special case is created because the weak and strong solutions are equal (for large deflection angle). the situation moves away from the expected solution. in that analysis it has to be remembered that when the flow becomes subsonic the equation changes from hyperbolic to an elliptic equation. In the same manner.

In fact. There isn’t any known experimental evidence to show that flow is unstable for δ = 0.264 CHAPTER 14. ∆δ + to maintain the same direction stream lines. OBLIQUE-SHOCK The change in the in∆θ + clination angle results ∆θ − in a different upstream Mach number and ∆δ − a different pressure. The opposite happens when the deflection increment becomes negative.22: The schematic of stability analysis for oblique shock. ∆θ = k+1 ∆δ 2 (14.52). the weak shock is stable. On the other hand. has to be in the opposite direction of the change of the shock angle. Thus. The change is determined from the solution provided before or from the approximation (14.56) Equation (14. Please note that this analysis doesn’t apply to the case of the close proximity of the δ = 0. 14. The pressure difference at the wall becomes a negative increment which tends to pull the shock angle to the opposite direction. . the deflection angle becomes positive which increases the pressure at the wall. ∆θ + or negative ∆θ− values. it must be pointed out that it doesn’t mean that the flow is unstable but rather that the model is incorrect. Yet. the virtual change in the deflection angle Fig. the shock wave is unstable according to this analysis to one direction but stable to the other direction.56) can be applied for either positive.

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

This assumption. the Mach angle dν dy òĵÅCà Á increase and result in a change in the direction of the flow. 15. This Mach line is assumed to be a result of the positive angle. a discussion on the relationship between the flow properties and the flow direction is presented.6) (15. cos(dν ) ∼ 1 (15. 15.2) A Mach line results because of a small disturbance in the wall contour. Hence. 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).5) dx = (U + dU ) cos ν − U = dU (15. is dy = (U + dU ) sin(dν ) = U dν The tan µ is the ratio of dy/dx (see Figure (15. the flow direction will change to fit the wall. The change of the velocity in the flow direction. However. Therefore. is close to reality. Once the contour is changed.3: The schematic of the turning only. The Mach line is the chief line in the analysis because of the wall contour shape information propagates along this line. The reason that a “negative” angle is not applicable is that the coalescing of the small Mach wave which results in a shock wave. no shock is created from many small positive angles. this assumption will be examflow ined.4) dν ∼ sin(dν ). and it is assumed here to be isotropic for a positive angle. as it turns out. dx is In the same manner. Later. the total Mach numÛ Ü4Ý¡ÞÔßÊà á  ber increases. In this chapter.3)) tan µ = dx dU = dy U dν (15. This direction change results in a change of the flow properties. PRANDTL-MEYER FUNCTION 1 M1 − 1 µ = tan−1 √ (15.3) .266 or CHAPTER 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. 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. the velocity perpendicular to the flow. dy .

2.2. 15. 15.8) becomes ν (M ) = − k+1 tan−1 k−1 k−1 (M 2 − 1) + tan−1 k+1 (M 2 − 1) + constant 267 (15.7) (15.6) and (15.1 Alternative Approach to Governing Equations In the previous secback Mach tion.11) (15.7) transforms it into √ M 2 − 1dM 2 dν = − 1 2 2M 2 1 + k− 2 M After integration of equation (15. In this section. a simplified verr Ur line sion was derived based Front Mach on geometrical arguUθ line θ ments.4: The schematic of the coordinate based on the matheflow turns around a sinmatical description gle point.12) ∂Ur Uθ ∂Ur Uθ 2 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r ρ ∂r ρ ∂r (15.15.10) . 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. For this coordinate system.9) The constant can be chosen in a such a way that ν = 0 at M = 1. It must be recognized that here the cylindrical coordinates are advantageous because the Fig.8) (15. GEOMETRICAL EXPLANATION The ratio dU/U was shown to be dM 2 dU = 1 2 U 2M 2 1 + k− 2 M Combining equations (15. a more rigorous explanation is provided.

19) it follows that Uθ = c (15.13) can be rearranged as transformed into − 1 Uθ Ur + ∂Uθ ∂θ = 1 ∂ρ ρ ∂θ (15.18) And an additional rearrangement results in c2 − U θ 2 From equation (15.16) from equation (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.17) Uθ or Uθ 2 Ur + ∂Uθ ∂θ = c 2 Ur + ∂Uθ ∂θ (15. PRANDTL-MEYER FUNCTION If the assumption is that the flow isn’t a function of the radius. the derivative isn’t zero. One has to remember that when r enters to the function.16) results in ∂Uθ − Ur ∂θ = c2 Uθ Ur + ∂Uθ ∂θ (15. Hence.14) into equation (15.15) (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 (15.13) Equation (15. but only .268 CHAPTER 15. like the first term in the mass equation. the mass equation is reduced to ρUr + ∂ (ρUθ ) =0 ∂θ (15. r.20) Ur + ∂Uθ ∂θ =0 (15. then all the derivatives with respect to the radius will vanish.

23) Enthalpy in perfect gas with a constant specific heat.27) incorporating the constant becomes Ur = 2h0 sin θ k−1 k+1 (15. The solution of the differential equation (15.22) In the literature. In fact.15. However. After some additional rearrangement.24).26) becomes k+1 k−1 ∂Ur ∂θ 2 k−1 + ∂Ur 2 ∂θ + Ur 2 2 = h0 (15. is c(θ )2 k h(θ) = Cp T = Cp and substituting this equality.26) + Ur 2 = 2h0 (15. This analysis can be also applied to the same equation when they are normalized by Mach number. equation (15.25) 1 Cp c2 R T = RT = R (k − 1) Cv k−1 (15. k .20) for the speed of sound and substituting equation (15. ∞ and since Uθ = 0 leads to ∂Ur = Uθ ∂θ (15.25) into ∂Ur 2 ∂θ After some rearrangement.23) results in c2 Uθ 2 + U r 2 + = h0 k−1 2 (15. these results are associated with the characteristic line. equation (15. The energy equation for any point on a stream line is h(θ) + Uθ 2 + U r 2 = h0 2 (15.27) Note that Ur must be positive. GEOMETRICAL EXPLANATION 269 the tangential component. the component shown in Figure (15.2.21) If r isn’t approaching infinity. M = 1. the non–dimensionalization can be applied at this stage as well.3) under Uy is equal to the speed of sound.24) Utilizing equation (15. based on the definition of the Mach angle.28) .15) becomes Uθ r ∂Ur − Uθ ∂θ =0 (15. equation (15.22) which is the radial velocity transforms equation (15. into equation (15.

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 (15.29) 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 (15.6). The arbitrary constant in equation (15. so ν (M ) = θ(M ) − θ(Mstarting ) = k+1 tan−1 k−1 k−1 k+1 M2 − 1 − tan−1 M2 − 1 (15.34) (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.32) k+1 tan2 k−1 θ k−1 k+1 (15.2)) tan ν = Mstarting 2 − 1 (15. In the geometrical model.2 Comparison And Limitations between the Two Approaches The two models produce exactly the same results. 15.35) These relationships are plotted in Figure (15. PRANDTL-MEYER FUNCTION which satisfies equation (15. The upstream Mach number is denoted in this segment as Mstarting .33) The deflection angle ν . has to match to the definition of the angle that is chosen here (θ = 0 when M = 1).270 CHAPTER 15.28) is chosen such that Ur (θ = 0) = 0.27) because sin2 θ +cos2 θ = 1.30) Now utilizing the expression that was obtained for Ur and Uθ equations (15. For this upstream Mach number (see Figure (15. The tangential velocity obtains the form Uθ = c = ∂Ur = ∂θ k−1 k+1 2 h0 cos θ k−1 k+1 (15. the assumption .29) and (15. but the assumptions for the construction of these models are different.2.

in many instances.5) 15. On top of the complication of the turning point. In the rigorous model. Thus. 15. In fact. The radial velocity is “fed” through the reduction of the density.15.4 The Working Equations for the Prandtl-Meyer Function ν2 − ν1 = ν (M2 ) − ν (M1 ) (15. However. In this case. Prandtl–Meyer function becomes ν∞ = π 2 k+1 −1 k−1 (15. Normally. an increase in the Mach number can result in a very significant radial velocity. the question of boundary layer arises. 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. However.3. 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). At this point (r = 0) these models fail to satisfy the boundary conditions and something else happens there. 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. (1 + ) the radial velocity is small . At a small Mach number. Regardless of the assumptions that were used in the construction of these models. the author recommends that this function be used everywhere beyond 2-4 the thickness of the boundary layer based on the upstream length. Aside from its close proximity to turning point.37) The change in the deflection angle is calculated by . the sharp point should be replaced by a smoother transition. some researchers recommend that. 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 (15.3 The Maximum Turning Angle The maximum turning angle is obtained when the starting Mach number is 1 and the end Mach number approaches infinity. For example. THE MAXIMUM TURNING ANGLE 271 is that the velocity change in the radial direction is zero. mass balance is maintained by the reduction of the density.36) The maximum of the deflection point and the maximum turning point are only a function of the specific heat ratios. the fact remains that there is a radial velocity at Ur (r = 0) = constant. the maximum turning angle is much larger than the maximum deflection point because the process is isentropic. it was assumed that radial velocity is only a function of θ.

5 d’Alembert’s Paradox In ideal inviscid incompressible flows. is D=2 w (P2 − P4 ) = w(P2 − P4 ) 2 (15. the drag. PRANDTL-MEYER FUNCTION sl ip li ne Maximum turning Fig. The relation between P2 and P4 is such that the flow depends on the upstream Mach number.7). and the specific heat. 15. k . and θ1 w θ2 this paradox is examined θ2 θ1 here.272 CHAPTER 15. M1 . the move3 ment of body does not 1 2 encounter any resistance. The net force in flow direction. 4 This result is known as d’Alembert’s Paradox. Again.5: Expansion of Prandtl-Meyer function when it exceeds the maximum angle 15.38). 15. 4 Supposed that a two– dimensional diamond–shape 2 1 body is stationed in a su3 personic flow as shown in Figure (15.38) It can be observed that only the area that “seems” to be by the flow was used in expressing equation (15. it is Fig.7: A simplified diamond shape to illustrate the suassumed that the fluid is inpersonic d’Alembert’s Paradox viscid. .

15.6. FLAT BODY WITH AN ANGLE OF ATTACK

273

Prandtl-Meyer Angle
100

80

k=1.4

60 θ 40 20 0

1

2

3

4

5 6 Mach Number

7

8

9

10

Fri Jul 8 15:39:06 2005

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

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

15.6

Flat Body with an Angle of Attack

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

Slip

plane

274

CHAPTER 15. PRANDTL-MEYER FUNCTION

15.7

Examples For Prandtl–Meyer Function

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

âäãæårçéè/êìë#í¡îyï ðòñôóéõyöh÷

¨
©

   ¥§¦   


øúùüûôýyþéÿ

 ¢¡¤£

Fig. 15.9: The schematic of Example 15.1

The Mach number is then

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

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

µ 52.6434

0.35603

0.74448

0.47822

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

µ 63.4620

0.12734

0.55497

0.22944

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

275

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

¢!#"$ % :(;*<>= ?A@ BDC0E>F

6§7 8
9

&('*)+ ,.-0/21 3¢4¤5

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

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

276

CHAPTER 15. PRANDTL-MEYER FUNCTION

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

µ 54.4623

0.28418

0.60365

0.47077

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

µ 55.5479

0.26190

0.58419

0.44831

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

15.8

Combination of the Oblique Shock and Isentropic Expansion

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

µ 73.1416

0.01506

0.37972

0.03965

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

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

µ 74.0528

0.01090

0.35248

0.03093

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

0.43534 3.1115 88.9313 20.3467

0.99676

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

3.1115 20.3467

1.1157

1.1066

0.99676

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

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

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

278 CHAPTER 15. PRANDTL-MEYER FUNCTION .

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

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

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

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

4}|D.4}|D..{1.4}|} \caption{ ?? \label{?:tab:?}}\\ ..{1.{1. Mx = 2.{1.. You can just can cut and paste it in your latex file.. whatInfo = infoStandard . You must use longtable style file and dcolumn style files.{1. int isTex = yes. ******************************************* 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.4}|D.. variableValue)..{1. ******************************************* \setlongtables \begin{longtable} {|D. variableName.A..2.4}|D. variableName = MxV. isRange. 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. 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.0 .makeTable(whatInfo.4}|D.4}|D. isTex.{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 . s. int isRange = no. shock etc) infoTube print tube side info for (Fanno.

pt}{0.1in]{0.3 Program listings Can be download from www.176& 2.potto.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.000& 0.6164 \\ A.152& 0.org.773& 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[-0.3608& \hline\end{longtable} & 1. . COMPUTER PROGRAM {$\rule[-0.5854& 3.1in]{0.pt}{0.3 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.

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

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

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

Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.