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 . . . . . . . . . . . . . . . . . . . . 9.9.1 Variations of The Tube Length ( 4f L ) Effects . . . D 9.9.2 The Pressure Ratio, P2 , effects . . . . . . . . . . P1 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 . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . .2 Examples . .11Oblique Shock Examples . .2 Comparison And Limitations between the Two Approaches 15. .1 Alternative Approach to Governing Equations . . . . . . . . . . .1. . . . . . . . . . . . .2 Geometrical Explanation . .4. . . .1. . . .4. . . . 14. . . . . . . . . . . . . . θ . . . . . . . . . . . . 14. . . . . . . . . 14. . . . . 15. . . . . .2 When No Oblique Shock Exist or When D > 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . .4. . . . . . . . 14. . . . . . . . . . . . . . . M1 . . . . . . . . . . . . . . . . . . . . .2. . .3 Upstream Mach Number. . . . . . . . . . . . . . . . . .4. . . . . . . . . . . . . . . .5 Flow in a Semi–2D Shape . . . . . 15.4. . . . . . 14. . . . . . . . . 15.CONTENTS 12 Evacuating/Filing Chambers under External Volume Control 12. . . . . . . . . . . M1 . . . .6 Small δ “Weak Oblique shock” . . 14. . . . . . . . . . . . . . . . .2. . . . . . . 15. . . . . . . . . . . . . . . . . . . . . .3 The Maximum Turning Angle . δ and θ . . . . . 14. . . . . . . . . . . . . . . . . . . . . and Shock Angle. . . . .3 Direct Connection . . 15. . . .2 Summary . . . . . δ 14. . . . 14. . . . . . . . . . . . . . . . . . . . . .4. . . . .8 Combination of the Oblique Shock and Isentropic Expansion . . . . .6 Flat Body with an Angle of Attack . .1 General Model . . . . . . . . . . . . . . . . . . . . 13 Topics in Unsteady one Dimensional gas dynamics 14 Oblique-Shock 14. . and Deflection Angle.4. . . .2. . . . . .3 Introduction to Zero Inclination . . .10Issues Related to the Maximum Deflection Angle . . . .2 Introduction .7 Close and Far Views of the Oblique Shock . . 14. . . . . . . . . . . 12. . . . .8 Maximum Value of Oblique shock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12. . . . . . . . . . . 14. . . . . . . . . . . . . . . . . . . . . . . .7 Examples For Prandtl–Meyer Function . . .1 Introduction .2. . . . . . . . . . . . .4. .2 Introduction to Prandtl–Meyer Function . . . . .9 Detached shock . . . 15. . . . . . . . . . . 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. . . . . . . . 12. . . . .6 Appendix: Oblique Shock Stability Analysis . 14. . .12Application of Oblique Shock . . . . 14. . . . . .1 Preface to Oblique Shock . 14. 14. . . .4 The Working Equations for the Prandtl-Meyer Function . . . . . . . . . . . . . . .1 Rapid Process . . . . .4 Given Two Angles. . . . . . . . . . . . . . .5 d’Alembert’s Paradox . . 14. . . . .1. . . . . .3 Oblique Shock . . . . . 14.4. .1 Upstream Mach Number. . . . . . . .2. . . . . . .4. . . . . . . . . . .4 Solution of Mach Angle . . . . . . . . . . 14. . . . . . .13Optimization of Suction Section Design . . . . . . . . . . . . . . . . . . . . . . . 14. 12. . . . . . . . . . . . . . . 15. . .4. . . . . . . . . . . . . . . . . . .4. . . . 15. . . 14. . . . . .1 Introduction to Oblique Shock . . 14. .4.5 Summary . .

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

. . . . . . . . . . . k = 1. .4 Flow rate as a function of the back pressure taken from Stodola 1927 Steam and Gas Turbines . .9 Portrait of Rankine . . 3. . . . . . . . . . . .1 Flow of a compressible substance (gas) through a converging–diverging nozzle. . . . . . . . . . . . 1. . 43 4. . . . .3 The stagnation properties as a function of the Mach number. .3 The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines . . . . . . . . . . . . . . . . . . . . .3 The Compressibility Chart . . . . . . . . . . . 1. . . . . . . . . . . . . . . . . . . . . . . .1 A very slow moving piston in a still gas . . . . . . . . .LIST OF FIGURES 1. . . . . Steam and Gas Turbine . . . . . . . . . . . . . . . . . . . . . 1. . . . . . . . .6 Photo of Ernest Mach . . . . . . . . . . . . . . . . . . 1. .1 The shock as connection of Fanno and Rayleigh lines after Stodola. . . . . . . 1. .4 44 4. . . . . . . . . . . 3. . . .4 Control volume inside a converging-diverging nozzle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. . . . . 41 4.12 The photo of Ernst Rudolf George Eckert with the author’s family . . . . . . 1. . . . . . . . . . . .2 Perfect gas flows through a tube . . . . . .11 Photo of Prandtl . 1. . . . . . . . . . . .7 The photo of thebullet in a supersonic flow not taken in a wind tunnel 1. . . . . . . 7 9 11 12 16 17 17 18 19 20 21 22 28 28 32 4. . 46 ix . . . . . . . . . . . . . . . . . . .10 The photo of Gino Fanno approximately in 1950 . . . .5 Portrait of Galileo Galilei .8 Photo of Lord Rayleigh . . . . . . . . . . . . . . . . . . . . . . . Steam and Gas Turbine . . . . . . .2 Stationary sound wave and gas moves relative to the pulse . . . .2 The schematic of deLavel’s turbine after Stodola. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. . . . . . . . . 1. . . 3. . . . . . . . . . . . . . . . .

.7) . . . . . . . . . . . . . . . 5. . . but it can also be viewed as a one– dimensional shock wave. . .x LIST OF FIGURES 4. . 5. . . . . . . . . . . . . . . . . . . . . . 5.5 The relationship between the cross section and the Mach number on the subsonic branch .10 The maximum of “downstream” Mach number as a function of the specific heat. . . . . 4. . . . . . . . . . . . . . . . . . . 4. . . . . . . . . . .8) . . . . 5.1 Control volume for isothermal flow .4 The flow in the nozzle with different back pressures . . . . . . .1 6. . . . . 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. . . . .7 The comparison of nozzle flow . . .16 The shock tube schematic with a pressure ”diagram. . . . . .8 Comparison of the pressure and temperature drop as a function of the normalized length (two scales) . . . . 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 8.7 The moving shock Mach numbers as a result of a sudden and complete stop. . . . . . . . 5. . . . . . .” . . . . . . . . . . . . 5. . . . . . .14 Schematic of a piston pushing air in a tube. .12 The results of the partial opening of the valve. . . . . . . . .3 6. . . . . . . . 5. . . . . . . . . . . . . . . . . .10 Schematic of a flow of a compressible substance (gas) thorough a converging nozzle for example (4. . . . . . . . . . . . . . . . . . 5. . . . .6 Various ratios as a function of Mach number for isothermal Nozzle . . . . . . . . . . . . . . . Description to clarify the definition of diffuser efficiency . . . . .17 Figure for Example (5. . . 4. . . . .3) . . .9 Schematic to explain the significances of the Impulse function . . . . . . .5 Comparison between stationary shock and moving shock in ducts . . . . . . . . .8 A shock moves into a still medium as a result of a sudden and complete opening of a valve . . . . . . . 137 . . . . . . 5. . .9 The number of iterations to achieve convergence. . .15 Figure for Example (5. . . . . . . . . . . . 5. . . . . . . . . . . . .13 A shock as a result of a sudden and partially a valve closing or a narrowing the passage to the flow . . . . . . . . . . . . . . . . . . . . . . . k. . . . . . . . . . . . . . . . . . . . . . . .10) . 4. . . .4 The ratios of the static properties of the two sides of the shock. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6. . 5. . 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10) . . . . . . 5. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 6. . . . . . . . . . .11 A shock moves into a moving medium as a result of a sudden and complete open valve. . .1 A shock wave inside a tube. . . . . . . 4. . . . .6 Comparison between a stationary shock and a moving shock in a stationary medium in ducts. . . . . . . . . A nozzle with normal shock . . . . . . . . Schematic of a supersonic tunnel in a continuous region (and also for example (6. . . . .3 The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. 5. . . . 5. 5. . .18 The results for Example (5. . . . .2 The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. . . .

. . . . . . 172 The Mach numbers at entrance and exit of tube and mass flow rate for Fanno Flow as a function of the 4f L . 185 10.3 A schematic of the control volumes used in this model .16 9. . . . 164 The schematic of Example (9. . . .4 The pressure assumptions in the chamber and tube entrance . . . . .3 The Mach number at the entrance to a tube under isothermal flow model as a function 4f L . .3 9.3 . . . . . 11. . . .1 9. . . . . . . . . . . . . . . . .7 9. .6 9. . . .175 D The pressure distribution as a function of 4f L for a short 4f L . . . . . . . . . . . . . . 195 11. . . . . . . . 177 D D The pressure distribution as a function of 4f L for a long 4f L . . . . 166 The maximum length as a function of specific heat. . . . . . . . . . . .3 for Fanno Flow . .2 Description of the pressure. . . . . . . . . . . . . . . . . . . .8 9. . . . . . . . 4f L superD sonic branch . . . . . . .3 The basic functions of Rayleigh Flow (k=1. . . . . . . .1 The control volume of the “Cylinder” . . . . . .2 The Temperature Entropy Diagram For Rayleigh Line . 11. . M1 to the tube as a function of 4f L supersonic branch . 170 The effects of increase of 4f L on the Fanno line . 218 . . . . .1 The control volume of Rayleigh Flow . . . 179 D D Fanno Flow Mach number as a function of 4f L when the total 4f L = D D 0. . . . . . . . . . . . . 182 D The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow . . . 11. . .5 9. . . . . . . . .9 9. . . . . . . k . . . . . .4) . . . . . . . . . . . . . . . .2 A schematic of two possible connections of the tube to a single chamber . . . . . .13 9. . . . . . . . . . . . . . . . . . 174 D M1 as a function M2 for different 4f L for supersonic entrance velocity. . . . . . . . . . . . . . . . . . . . . . . 155 Various parameters in Fanno flow as a function of Mach number . 178 D D The effects of pressure variations on Mach number profile as a function of 4f L when the total resistance 4f L = 0. . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 D 9. . . . 11. . . . . . . . . . . . . 143 8. . 201 202 202 203 210 212 12. . temperature relationships as a function of the Mach number for isothermal flow . . . . .17 9. . . . .5 The reduced time as a function of the modified reduced pressure . . . . . . .10 9. . 191 10. . . . .1 The two different classifications of models that explain the filling or evacuating of a single chamber . . . .6 The reduced time as a function of the modified reduced pressure . . . . . .11 9. 173 D M1 as a function M2 for various 4f L . .18 Control volume of the gas flow in a constant cross section . . 181 The maximum entrance Mach number. . . . . 11. . . . . . . . . . . . .15 9.LIST OF FIGURES xi 8. . . . . . . . .4 9. . . . . .12 9. . . . . . . . . . . . . . 180 Schematic of a “long” tube in supersonic branch . 163 Schematic of Example (9. . 189 10. . . . . . .14 9. . . . . 181 The extra tube length as a function of the shock location.1) . . . . . . . . . . . . . . . . .2 9. . . 171 D The development properties in of converging nozzle . . . .2) . . . . . . . . . . . . .

. . . . . . . . . . . .22The schematic of stability analysis for oblique shock. .3 The pressure ratio as a function of the dimensionless time for choked condition .10Schematic of finite wedge with zero angle of attack. . . . . . .4 Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. 263 14. . . . . . . . . . . . . . Prandtl-Meyer function approaches closer to a zero deflection angle. . .11A local and a far view of the oblique shock. . . However. and the weak shock coexist. . . . . . . . . . . . . . . . . . . . . . . . . Notice that both have a maximum point and a “no solution” zone. . . . . . . . . . . . . . . . . . . . . . 228 14. . . . . . . . . 224 14. . . . .2 The regions where oblique shock or Prandtl–Meyer function exist. . . . . 266 . . and exit Mach number for M1 = 3 . . . . . . . . . . . . .9 The possible range of solutions for different parameters for given upstream Mach numbers . . . . . . . . a Research Professor at Cuernavaco University of Mexico. . . . . . . . 265 15.7 The Mach waves that are supposed to be generated at zero inclination. 223 12. . . . . . . . . . . 250 14. . . . . . . .5 The view of a large inclination angle from different points in the fluid field. . 229 14. . . . . . . . . . .14The “detached” shock in a complicated configuration sometimes referred to as Mach reflection . . . . . . 244 14. . . . . .18Schematic for Example (14. . 264 15. . . 253 14. . . .3 The schematic of the turning flow . . . . . . . . 254 14. 241 14. . . . . . . 252 14. . . . . . . . . . . . . . . . . .xii LIST OF FIGURES 12. . . . . . . . . shock angle. . .0. . . . . . 236 14. . . .6 The various coefficients of three different Mach numbers to demonstrate that D is zero . . . 243 14. . . . . . 247 14. . .2 The angles of the Mach line triangle . .240 14. . . . . . the cone half angle is 15◦ and the Mach number is 2.15Oblique shock occurs around a cone. . . . According to his measurement. . . . . . . .16Maximum values of the properties in an oblique shock . . . .8 The calculation of D (possible error). . . . . . . . . . . . . 265 15. . . . . . . . . . . . . . .1 The definition of the angle for the Prandtl–Meyer function. . . . .4) . . . . 235 14. . . . . . . . . . . . . 248 14. . . .3 A typical oblique shock schematic . . . 224 12. . . . . . . . . . . It can be noticed that the normal shock. . which is around zero. . . . . .1 A view of a straight normal shock as a limited case for oblique shock 227 14. . . . . . .4 The pressure ratio as a function of the dimensionless time . . . . . . . . . .12The schematic for a round–tip bullet in a supersonic flow . This photo is courtesy of Dr. .13The schematic for a symmetrical suction section with Mach reflection 248 14.21Typical examples of unstable and stable situations . . . . . . . . . . . . 239 14. .2 The pressure ratio as a function of the dimensionless time for chokeless condition . . . . . .5) . . . . . Grigory Toker. . . . . . . . . . the strong shock. . . . . 251 14. . . . . . . . .19Schematic for Example (14. . . .17Two variations of inlet suction for supersonic flow . . . .2.20Schematic of two angles turn with two weak shocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 14. . . . . . 251 14. . . . . . . . . . . . . . . . . . . . . . . . .

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

xiv LIST OF FIGURES .

. . . . . . . . . . . . . . . . . . . . . . . . .4) . . . . . . . . . . . . . 4. . . . . . . . .2 Table for Reflective Shock from suddenly closed valve (end) (k=1.4 Table for shock propagating from a suddenly opened valve (k=1.2 Isentropic Table k = 1. .3 Isothermal Table (continue) . . . . .3) 121 xv . . . . . . . 1972 . . . . . . . . . .1 Fliegner’s number and other paramters as function of Mach number 4. . .4 (continue) . . . . . . . after Aldred. . . . . . . . . 117 5. . . . . after Aldred. . . .3 Table for shock propagating from suddenly opened valve (k=1. . . . . . . . . . . .2 Liquids speed of sound. . . . . . . . Manual of Sound Recording.4 Table for shock propagating from a suddenly opened valve (k=1.3 Isothermal Table . 116 5. . . . .3) 119 5. . .LIST OF TABLES 3. . . . . . . . . . . . .4 .3 Solids speed of sound. .4 (continue) . . . .4)(continue)118 5. 1972 . . . . . . . . . .1 The shock wave table for k = 1. . . . . . . . . . . . . . . . . .4) . 118 5. . . 4. London: Fountain Press. . . . . .4 (continue) . . . . . .4 . .2 Table for a Reflective Shock from a suddenly closed end (k=1. . . . . . . . London:Fountain Press. . . 3. . 4. . . .3) 120 5. 119 5.1 The shock wave table for k = 1. . . .2 Isentropic Table k=1. . . . . . . .4) . . . . . . . . . . . . . . . . 35 36 37 58 59 60 62 63 73 74 5. . . . 3. . 115 5. . 4. . . . . . . . Manual of Sound Recording. . . . . . . . .4 Table for shock propagating from a suddenly opened valve (k=1.1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . 117 5. . 4. . . .1 The shock wave table for k = 1. . . . .1 Water speed of sound from different sources . . . John. . . . 4.3 Table for shock propagating from suddenly opened valve (k=1. John. . . . . . . .1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . .

1 Maximum values of oblique shock (continue) k=1. . . . . .4 . . . 245 14. . . . . . . . . .4 9. . . . . . . . .1 Rayleigh Flow k=1. . . . .4 (continue) . . . . . . .1 LIST OF TABLES The Isothermal Flow basic parameters . . . . . . 152 Fanno Flow Standard basic Table . .4 . . . . . 187 Fanno Flow Standard basic Table (continue) .1 Rayleigh Flow k=1. . . . 188 10. 195 14. . . . . . . 194 10. . . . . . . . . . . . . . . . . 246 . . .1 Table of maximum values of the oblique Shock k=1. . . . . . . . .xvi 8.4 . . . . . . . . . . . . . .1 Rayleigh Flow k=1. . . . . . . . . .1 8. . . . 193 10. . . . . . . . . . . . .4 (continue) . . .1 9. . . . . . . . . . . . . . . 147 The flow parameters for unchoked flow . . . . . . . . .

We have designed this License in order to use it for manuals for free software. either commercially or noncommercially. which means that derivative works of the document must themselves be free in the same sense. This License is a kind of ”copyleft”. It complements the GNU General Public License. while not being considered responsible for modifications made by others. Fifth Floor. Secondarily. November 2002 Copyright ©2000. which is a copyleft license designed for free software. or other functional and useful document ”free” in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it. this License preserves for the author and publisher a way to get credit for their work. textbook.2. Inc. GNU Free Documentation License Version 1.2001. but changing it is not allowed.2002 Free Software Foundation. Preamble The purpose of this License is to make a manual. 51 Franklin St. because free software needs free documentation: a free program should xvii . with or without modifying it. MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document. Boston.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.

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

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

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

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

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

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

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

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

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

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

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

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

While he known to look like he know about many things. While he writes his programs and does other computer chores. The author lives with his wife and three children. practically from scratch. 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. More than that. he often feels clueless about computers and programing. 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.

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

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

this method is expected to accelerate the creation of these high quality books. The organizer of the book in some instances will be the main author of the work.0 0. better discussions or better explanations are all welcome to these books.0.0 0.0. These books are intended to be “continuous” in the sense that there will be someone who will maintain and improve the book with time (the organizer).4 0. while in other cases This may merely be the person who decides what will go into the book and what will not (gate keeper). .0. the traditional role of author will be replaced by an organizer who will be the one to compile the book.1 0.0 0. other books contain data5 which can be A typeset in LTEX.3 0. But more than that. 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. these works will have a version number because they are alive and continuously evolving.0. Thus. 5 Data are not copyrighted.) can be redone by anyone who has the time to do it.0. These data (tables. graphs and etc.0. Someone has to write the skeleton and hopefully others will add “flesh and skin.” In this process.0 0.01 0. These books should be considered more as a project than to fit the traditional definition of “plain” books.0.CREDITS xxxiii adding a question and perhaps the solution. It is hoped that the book will be error-free.0 0.0 Based on Eckert Tel-Aviv’notes The meaning of the progress is as: • The Alpha Stage is when some of the chapters are already in rough draft. Even if not complete. Unlike a regular book. some errors are possible and expected.0. chapters or sections can be added after the skeleton has been written. Thus. Thus. contributions can be made from any part of the world by those who wish to translate the book.0. Additionally. the contributions to books can be done by many who are not experts. Nevertheless.0 0. These books are written in a similar manner to the open source software process. It is also hoped that others will contribute to the question and answer sections in the book.

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

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

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

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

is suggested to referred to this model as Shapiro flow . The second chapter deals with speed of sound. Gas Dynamics Calculator (CDC-POTTO). It is hoped that this book will serve the purposes that was envisioned for the book. is currently under construction. 11 If 12 It you have better and different examples or presentations you are welcome to submit them. In the oblique shock. The next three chapters are dealing with three models of choked flow: Isothermal flow12 . Fanno flow and Rayleigh flow. The chapter on one-dimensional unsteady state. weak and unrealistic shocks. etc). again under construction. and properties change effects (ideal gases and real gases) (under construction). The fourth chapter deals with a simple case of discontinuity (a simple shock wave in a nozzle). deals with a basic introduction of thermodynamics concepts (under construction). which is excellent tool to explain the strong. The third chapter provides the first example of choked flow (isentropic flow in a variable area). Isothermal flow provides useful tools for the pipe systems design. and Prandtl–Meyer expansion will be included only after the gamma version unless someone will provide discussion(s) (a skeleton) on these topics. It is further hoped that others will contribute to this book and find additional use for this book and enclosed software. Discussions on the flow around bodies (wing. The program design and how to use the program are described (briefly). for the first time analytical solution is presented. The last chapter deals with the computer program. The book is organized into several chapters which.xxxviii LIST OF TABLES demonstrated11 . as a traditional textbook. the Isothermal flow is introduced because of the relative ease of the analytical treatment. 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. First. The topics of filling and evacuating of gaseous chambers are presented. normally missed from traditional textbooks. There are two advanced topics which included here: oblique shock wave. Every chapter can be “ripped” out and printed independently.

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

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

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

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

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

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

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

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

Application Marcell–Taylor (from the notes) Examples Transient problem .VERSION 4. Provide discussion on variations of the effecting parameters. The FORTRAN version will not be included. Oblique Shock Add application to design problems Real Gas effects Prandtl–Meyer The limitations (Prandtl-Meyer).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.1. Examples: provide categorization Evacuation and filling semi rigid Chambers To construct the Rayleigh flow in the tube (thermal chocking) Energy equation (non isentropic process) Examples classifications Software (converting the FORTRAN program to c++) Evacuating and filling chambers under external forces Comparison with chemical reaction case Energy equation (non isentropic process) Examples Software transformation from FORTRAN to c++.

xlviii LIST OF TABLES .

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

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

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

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

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

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

20 Some add discussion about the general relationships. page 62. The fact that the conditions and direction were known did not bring the solution to the equations. to be add to oblique shock Since early in the 1950s the analytical relationships between the oblique shock. Berlin. HISTORICAL BACKGROUND conditions to a more general form19 .1. 1.1: The shock as connection of Fanno and munist regimes. In fact Prandtl is the one who introduced the name of Rankine-Hugoniot’s conditions not aware of the earlier developments of this condition.3. 62. it took over 50 years for Prandtl to arrive at and to demonstrate that the shock has only one direction20 . Taylor from England as the proof (of course utilizing the second law) 21 Theodor Meyer in Mitteil. Today this equation/condition is known as Prandtl’s equation or condition (1908). 1908.1)). the supersonic branch is missing from his understanding (see Figure (1. the second law has been around for over 40 years and yet the significance of it was not was well established. ub. The first full analytical solution connecting the angles with upstream Mach number was published in this book version 0. 7 Here. There were until recently (version 0. Steam and Gas was found by Landau & Lifshitz Turbine and expanded by Kolosnitsyn & Stanyukovich (1984). Thus. 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. Stodola suggested the graphical solution utilizing the Fanno line. In fact. Theodor Meyer (Prandtl’s student) derived the conditions for oblique shock in 190821 as a byproduct of the expansion work. shock angle.3. It was probably later that Stodola (Fanno’s adviser) realized that the shock is the intersection of the Fanno line with the Rayleigh line. the relationship between upstream Mach number and the angles. Yet. The “last nail” of understanding was put by Landau. view the work of G. No. deflection angle. I. A solution Rayleigh lines after Stodola. and Mach number was described as impossible to obtain. ¨ 19 To . Forsch-Arb. The probable reason that analytical solution was not published because the claim chapter.

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

1. Choking wasn’t clearly to be observed. . an explicit analytical solution for shock a head of piston movement (special case of open valve) was originally published in this book for the first time. this book is first to publish this tables). but after a certain point it did matter (because the steam was choked). 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. Vol 31 1898.3 Choking Flow The choking problem is almost unique to gas dynamics and has many different forms. The introduction of the steam engine and other thermodynamics cycles led to the choking problem.3 of this book (as far it is known. No one was looking for or expecting the choking to occur. page 268 f.. Steam and Gas Turbine operation by another inventor (Curtis) 1896 used in steam turbine. even when researcher stumbled over it. Up to this version 0. HISTORICAL BACKGROUND 9 The moving shock and shock tube were study even before World War Two. Only after the dimensionless presentation of the problem and the construction of the moving shock table the problem became trivial.3. As a result deLavel’s nozzle was invented by Carl Gustaf Patrik Fig. The first to discover (try to explain) the choking phenomenon was Rayleigh29 . leading to the study and development of Rayleigh flow. 1. The first experimental choking phenomenon was discovered by Fliegner’s experiments which were conducted some time in the middle of 186x28 on air flow through a converging nozzle. The 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.1. there was no realization that the flow is choked just that the flow moves faster than speed of sound.3. Yet. 68–72. 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. “Theorie die Turbinen. and when it was found the significance of the choking phenomenon was not clear. trial and error method was the only method to solve this problem. According the thermodynamics theory (various cycles) the larger heat supply for a given temperature difference (larger higher temperature) the larger the output. 28 Fliegner Schweizer Bauztg.” Leipzig 1899. 29 Rayleigh was the first to develop the model that bears his name.4. It is likely that others had noticed that flow is choked. Later. p. but did not produce any model or conduct successful experimental work.

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

in 1904 the Wright brothers demonstrated that results from the wind tunnel and spinning arm are different.3. At the end of the arm was the object that was under observation and study. severalGas Turbines explained choking at the throat have been built.1. There are two main models or extremes that describe the flow in the nozzle: isothermal and adiabatic. 36 Bendemann Mitteil uber Forschungsarbeiten.3: The measured pressure in a nozzle taken from Stodola 1927 Steam and models that and 1947-48. Yet. A different reason to study the converging-diverging nozzle was the Venturi meter which was used in measuring the flow rate of gases. 37. when the first supersonic wind tunnel was built. As a result. Between the turn of the century Fig. 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. ¨ . HISTORICAL BACKGROUND Nozzle flow 11 The first “wind tunnel” was not a tunnel but a rotating arm attached at the center. 1907. No.545) that creates the maximum flow rate. the spinning arm was no longer used in testing. The arm’s circular motion could reach a velocity above the speed of sound at its end. due to the circular motion. 1. Berlin.

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

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

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

The earlier work dealing with this issue was by Giffen. The simplest model of nozzle.e. As anecdotal story explaining the lack of progress. J.. A. 1964. HISTORICAL BACKGROUND 15 1.. Much information can be obtained from other resources. Later. and was republished by Owczarek. Since World War II considerable works have been carried out in this area but with very little progress46 . fact. Textbook Co.”45 . i. not realizing that garbage in is garbage out. in die casting conference there was a discussion and presentation on which turbulence model is suitable for a complete still liquid. In 1993 the first reasonable models for forced volume were published by the undersigned.1.3. Scranton. the model has to be based on scientific principles and not detached from reality. Other “strange” models can be found in the undersigned’s book “Fundamentals of Die Casting Design.35) in 2005. such as the Internet. In this section there is no originality and none should be expected. the emergence of the CFD gave the illusion that there are solutions at hand. 1. the model and solution to the nozzle attached to chamber issue in his book “Fundamentals of Gas Dynamics. Pennsylvania.3 of this book by this author.3. is not sufficient in many cases and a connection by a tube (rather just nozzle or orifice) is more appropriated. 46 In 45 International . The analytical solution for forced volume and the “balloon” problem (airbag’s problem) model were published first in this book (version 0.3.. There are many figures that should be included and a biased selection was required. The classification of filling or evacuating the chamber as external control and internal control (mostly by pressure) was described in version 0.5 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. Later several researchers mostly from the University in Illinois extended this work to isothermal nozzle (choked and unchoked). that model was extended by several research groups. 1940. 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.

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

Mach and E. Mach was also a great thinker/philosopher and influenced the theory of relativity dealing with frame of reference. In 1863. physics and philosophy.” Mach was Fig.G. 1. he named it Mach’s Principle. . Mach’s father.1.R. HISTORICAL BACKGROUND Ernest Mach (1838-1916) 17 Ernst Mach was born in 1838 in Chrlice (now part of Brno). when he studied in Kromeriz Gymnasium. and in 1867 he received a position in the Technical University in Prague47 where he taught experimental physics for the next 28 years. He was not able to attach the camera to the arm and utilize the remote control (not existent at that time). His experiments required exact timing.3.7: The Photo of the bullet in a supersonic flow that could take a photo at superMach made. Einstein was greatly influenced by it. Eckert. was a high school teacher who taught Ernst at home until he was 14. but he declined. before he entered the university of Vienna were he studies mathematics. 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. and in 1918.7. when Czechia was still a part of the Austro–Hungary empire. Note it was not taken in a wind tunnel sonic speeds.1916) published Die Machanik in which he formalized this argument. He graduated from Vienna in 1860. 1. Later. He then turned to physics. Mach’s shadowgraph 47 It is interesting to point out that Prague provided us two of the top influential researchers[:] E.6: Photo of Ernest Mach interested also in physiology of sensory perception. There Mach wrote his thesis ”On Electrical Discharge and Induction. Mach’s revolutionary experiment demonstrated the existence of the shock wave as shown in Figure 1. Ernest Mach (1836 . This was one of the primary sources of inspiration for Einstein’s theory of General Relativity. 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. Johann.

started studies in compressible flow mostly from a mathematical approach. Rayleigh was also a contributer to the Encyclopedia Britannica. concerning optics and vibrating systems. Lord Rayleigh was a British physicist born near Maldon. It seems that Rayleigh was the first who realized that flow with chemical reactions (heat transfer) can be choked. on November 12. clearly 48 Mach dealt with only air. and photography. six volumes issued during 1889-1920. hydrodynamics. . He published 446 papers which. the year of his marriage. covering sound. Yet. elasticity. wave theory. light scattering. In 1861 he entered Trinity College at Cambridge. Rayleigh’s later work was concentrated on electric and magnetic problems. but his later work ranged over almost the whole field of physics. He served for six years as the Fig. Lord Rayleigh’s first research was mainly mathematical. The name Mach Number (M) was coined by J. viscosity. electrodynamics. In 1866 he obtained a fellowship at Trinity which he held until 1871. Mach was the first to note the transition that occurs when the ratio U/c goes from being less than 1 to greater than 1.8: Photo of Lord Rayleigh president of the government committee on explosives. Essex.18 CHAPTER 1. electromagnetism. He was Lord Lieutenant of Essex from 1892 to 1901. density of gases. 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). 1842. flow of liquids. At that time there wasn’t the realization that the flow could be choked. reprinted in his collected works. color vision. Ackeret (Prandtl’s student) in 1932 in honor of Mach. INTRODUCTION technique and a related method called Schlieren Photography are still used today. Rayleigh was considered to be an excellent instructor. but it is reasonable to assume that he understood that this ratio was applied to other gases. capillarity. His Theory of Sound was published in two volumes during 1877-1878. John William Strutt (Lord Rayleigh) A researcher with a wide interest. He graduated in the Mathematical Tripos in 1865 as Senior Wrangler and Smith’s Prizeman. and from 1896 to 1919 he acted as Scientific Adviser to Trinity House. 1. and his other extensive studies are reported in his Scientific Papers. Mach’s contributions to supersonic flow were not limited to experimental methods alone. where he commenced reading mathematics. His exceptional abilities soon enabled him to overtake his colleagues.

and his only brother and parents died before him. London. was a Justice of the Peace and the recipient of honorary science and law degrees. They had three sons. 1. 1919 at Witham. he was head of the Cavendish Laboratory at Cambridge from 1879-1884. Essex. strength of materials. Rankine. Rankine was born in Edinburgh to British Fig.3. William John Macquorn Rankine William John Macquorn Rankine (July 2. He received the Nobel Prize in 1904. and as President from 1905 to 1908. never allowing politics to interfere with science. He intervened in debates of the House of Lords only on rare occasions.9: Portrait of Rankine Army lieutenant David Rankine and Barbara Grahame. As a successor to James Clerk Maxwell. the eldest of whom was to become a professor of physics at the Imperial College of Science and Technology. at Witham. His steam engine manuals were used for many decades. Rayleigh died on June 30. Lord Rayleigh. and naval engineering in which he was involved in applying scientific principles to building ships. Rankine developed a theory of the steam engine. a Chancellor of Cambridge University. 1872) was a Scottish engineer and physicist. . sister of the future prime minister. HISTORICAL BACKGROUND 19 show his capacity for understanding everything just a little more deeply than anyone else. 1820 . He was a Fellow of the Royal Society (1873) and served as Secretary from 1885 to 1896. Rankine never married. and in 1887 became Professor of Natural Philosophy at the Royal Institute of Great Britain. In 1871 he married Evelyn. Rankine was well rounded interested beside the energy field he was also interested in civil engineering. Lord Rayleigh died on June 30.1. Essex. 1919.December 24. He was a founding contributor to the science of thermodynamics (Rankine Cycle). the Earl of Balfour (of the famous Balfour declaration of the Jewish state).

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

During this time Prandtl developed his boundary layer theory and studied supersonic fluid flows through nozzles. Interestingly. His interest changed when.D. By the 1930s. Bavaria. he was known worldwide as the leader in the science of fluid dynamics. developed the first theory for calculating the properties of shock and expansion waves in supersonic flow in 1908 (two chapters in this book). He developed the Prandtl-Glauert rule for subsonic airflow. he was required to design factory equipment that involved problems related to the field of fluid mechanics (a suction device). such as meteorology and structural mechanics. his Ph. Ludwig Prandtl was born in Freising. It is mind boggling to look at the long list of those who were his students and colleagues. Prandtl changed the field of fluid mechanics and is called the modern father of fluid mechanics because of his introduction of boundary layer. in his first job. There is no one who educated as many great scientists as Prandtl. His work and achievements in fluid dynamics resulted in equations that simplified . As a result. His father was a professor of engineering and his mother suffered from a lengthy illness. His 1904 paper raised Prandtl’s prestige. Prandtl also contributed to research in many areas. 1. was focused on solid mechanics. in 1874. Prandtl started his studies at the age of 20 in Munich. Germany and he graduated at the age of 26 with a Ph. Prandtl. 1953.11: Photo of Prandtl chinery books. he presented the revolutionary paper “Flussigkeitsbewegung Bei Sehr Kleiner Reibung” (Fluid Flow in Very Little Friction). Germany (1901). In 1925 Prandtl became the director ¨ of the Kaiser Wilhelm Institute for Flow Investigation at Gottingen. He became the director of the ¨ Institute for Technical Physics at the University of Gottingen. the paper which describes his boundary layer theory. In 1904.1. HISTORICAL BACKGROUND Ludwig Prandtl 21 Perhaps Prandtl’s greatest achievement was his ability to produce so many great scientists. with his student Theodor Meyer.3.D. Later he sought and found a job as a professor of mechanics at a technical school in Hannover. turbulence mixing theories etc. This upbringing fostered the young Prandtl’s interest in science and experimentation. As a byproduct they produced the theory for oblique shock. the young Ludwig spent more time with his father which made him interested in his father’s physics and maFig. ¨ Ludwig Prandtl worked at Gottingen until his death on August 15.

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

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

24 CHAPTER 1. INTRODUCTION .

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

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 .

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

SPEED OF SOUND The people had recognized for several hundred years that sound is sound wave dU velocity=dU a variation of pressure.4) (3.1) Fig.5) An expression is needed to represent the right hand side of equation (3. 3.28 CHAPTER 3.1: A very slow moving piston in a still gas small disturbance travel in a “quiet” medium. it raises the question: what is the speed of the Fig. 3.2) into equation (3. 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. The information that the piston is moving passes thorough a single “pressure pulse. Here. For an ideal gas.” It is assumed that if the velocity of the piston is infinitesimally small. 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 (3.2) From the energy equation (Bernoulli’s equation). it is considered . Thus.4) yields c2 dρ ρ = dP dP =⇒ c2 = ρ dρ (3.1).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. Thus. P is a function of two independent variables. To answer this question consider a piston moving from the left to the right at a relatively small velocity (see Figure 3. The ears c sense the variations by frequency P+dP P ρ ρ+dρ and magnitude which are transferred to the brain which translates to voice. This velocity is referred to as the speed of sound.5). 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. the pulse will be infinitesimally small.3) Substituting the expression for dU from equation (3.

(3.1: Demonstrate that equation (3. The pressure for an ideal gas can be expressed as a simple function of density. ρ.11) . k namely P = constant × ρk (3.3. and a function “molecular structure” or ratio of specific heats.8) (3.7) s Note that the equation (3. therefore it can be written ∂P dP = dρ ∂ρ (3.6) In the derivations for the speed of sound it was assumed that the flow is isentropic.3.9) dP = c2 dρ This yields the same equation as (3. s) where s is the entropy.5).5) can be derived from the momentum equation.5) can be obtained by utilizing the momentum equation instead of the energy equation.10) 3.3 Speed of sound in ideal and perfect gases The speed of sound can be obtained easily for the equation of state for an ideal gas (also perfect gas as a sub set) because of a simple mathematical expression. Example 3. SPEED OF SOUND IN IDEAL AND PERFECT GASES 29 that P = P (ρ. The full differential of the pressure can be expressed as follows: dP = ∂P ∂ρ dρ + s ∂P ∂s ds ρ (3.2) 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 dUX ∼ 0 dP = (ρ + dρ) c2 −$$X $$$ 2 2cdU + − ρc2 (3. S OLUTION The momentum equation written for the control volume shown in Figure (3.

and equation (3.12) can be written as √ (3.12) Remember that P/ρ is defined for an ideal gas as RT .9563 At 18[bar] and 350◦ C: s = 7.13216 kg m3 kg m3 kg m3 After interpretation of the temperature: kJ At 18[bar] and 335.2: Calculate the speed of sound in water vapor at 20[bar] and 350◦ C.5 0. table A 8.46956 ρ = 7.7◦C: s ∼ 6.13) c = kRT Example 3.5 sec Note that a better approximation can be done with a steam table. 3 This data is taken form Van Wylen and Sontag “Fundamentals of Classical Thermodynamics” 2nd edition . S OLUTION The solution can be estimated by using the data from steam table3 c= At 20[bar] and 350◦ C: s = 6.30 and hence CHAPTER 3.14) s=constant ρ = 6. and it will be part of the future program (potto–GDC).8226 ∆P ∆ρ kJ K kg kJ K kg kJ K kg (3. Classical Thermodynamics.32823 sec kg m3 (3.) √ m c = kRT ∼ 1.0100 At 18[bar] and 300◦ C: s = 6. SPEED OF SOUND P c= dP constant × ρk = k × constant × ρk−1 = k × dρ ρ =k× P ρ (3. (a) utilizes the steam table (b) assuming ideal gas.9563 K kg ρ ∼ 6.61376 ρ = 6.327 × 461 × (350 + 273) ∼ 771.15) for ideal gas assumption (data taken from Van Wylen and Sontag.94199 and substituting into the equation yields c= m 200000 = 780.

SPEED OF SOUND IN REAL GAS 31 Example 3.7).16) For assumption of constant temperature the time is t= √ Hence the correction factor tcorrected = t TA TA 2 ¯ T 3 (TB − TA ) TB TA −1 (3. x T = (TB − TA ) + TA h Where the distance x is the variable distance.18) (3.? S OLUTION The temperature is denoted at “A” as TA and temperature in “B” is TB . The distance between “A” and “B” is denoted as h. 3. The compressibility factor represents the deviation from the ideal gas.19) The speed of sound of any gas is provided by equation (3. 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. 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 (3.19) some mathematical expressions are needed.4 Speed of Sound in Real Gas The ideal gas model can be improved by introducing the compressibility factor.3. Recalling from thermodynamics. To obtain the expression for a gas that obeys the law expressed by (3. the Gibbs function (3. Thus. What is the time it take for sound to travel from point “A” to point “B” under this assumption.20) .17) This correction factor approaches one when TB −→ TA . a real gas equation can be expressed in many cases as P = zρRT (3.4.

23) P Van Wylen p. 372 SI version. it can be shown 4 dh = Cp dT + v − T 4 See ∂v ∂T (3.21) P The definition of volumetric specific heat for a pure substance is Cv = ∂u ∂T =T ρ ∂s ∂T (3.22) ρ From thermodynamics.20) The definition of pressure specific heat for a pure substance is Cp = ∂h ∂T =T P ∂s ∂T (3. .32 CHAPTER 3. perhaps to insert the discussion here. SPEED OF SOUND Fig.3: The Compressibility Chart is used to obtain T ds = dh − dP ρ (3. 3.

28) and (3.30) yields dρ dP Cv = ρ P Cp z+T z+T ∂z ∂T P ∂z ∂T ρ (3.26) P Utilizing Gibbs equation (3. However.3.24) into equation (3.24) Substituting the equation (3.20) dh T ds = Cp dT − T z ∂z ∂T zRT P dP dP dP − = Cp dT − ρ ρ ρ ∂z ∂T T z ∂z ∂T +1 P =Cp dT − dP P P ρ T z +1 P (3.28) P Equation (3.4.27) Letting ds = 0 for isentropic process results in dP R dT = z+T T P Cp ∂z ∂T (3.31) .25) to became dh = Cp dT − Tv z ∂z ∂T   dh = Cp dT + v − T   RT    P ∂z ∂T + P zR    dP P  ∂z ∂T dP ρ (3. 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 b zR ∂T&& + & P ∂T P & 1 (3.29) results in dρ R z+T ρ Cv ∂z ∂T = ρ dP R z+T P Cp ∂z ∂T (3.30) P Rearranging equation (3.29) ρ Equating the right hand side of equations (3.23) results  v  v  z T Simplifying equation (3.28) can be integrated by parts. 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 (3.25) P dP = Cp dT − T z (3.

Now. SPEED OF SOUND If the terms in the braces are constant in the range under interest in this study.995times287 × 300 = 346. However.34 CHAPTER 3. This relationship (3.11). But the definition of n in equation (3. the speed of sound for a real gas can be obtained in the same manner as for an ideal gas.407. n = 1. a change in temperature can have a dramatical change in the speed of sound. For example.34) Example 3.4 factor (0. and in–fact any thermodynamics book shows this relationship. at relative moderate pressure but low temperature common in atmosphere.33) Equation (3.3 and n ∼ 1 which means that speed of sound is only 0.33) isn’t new.31) yields ρ1 ρ2 n = P1 P2 (3. What is different in these derivations is that a relationship between coefficient n and k was established. equation (3.4: Calculate the speed of sound of air at 30◦ C and atmospheric pressure ∼ 1[bar]. n is defined as k n= Cp Cv z+T z+T ∂z ∂T ρ ∂z ∂T P (3.403.3 1.7[m/sec] The correction factor for air under normal conditions (atmospheric conditions or even increased pressure) is minimal on the speed of sound.403 × 0.32) Note that n approaches k when z → 1 and when z is constant. S OLUTION According to the ideal gas model the speed of sound should be √ √ c = kRT = 1.5) to calculated by ideal gas model. dP = nzRT dρ (3. . Assume that R = 287[j/kg/K].31) can be integrated.407 × 287 × 300 ∼ 348. The integration of equation (3. Make the calculation based on the ideal gas model and compare these calculations to real gas model (compressibility factor).1[m/sec] For the real gas first coefficient n = 1.32) provides a tool to estimate n.403 has √ √ c = znRT = 1. the compressibility factor. z = 0.995. and z = 0.33) is similar to equation (3. The specific heat for air is k = 1. For short hand writing convenience.

5.35) This agrees well with the measured speed of sound in water.. The amount of compression of almost all liquids is seen to be very small as given in Table (3. 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].1: Water speed of sound from different sources The effect of impurity and temperature is relatively large.5) Remark Fresh Water (20 ◦ C) Distilled Water at (25 ◦ C) Water distilled reference Cutnell.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.36) dP dρ (3.2×109 N/m2 . Ohio: Chemical Rubber Co.2 × 109 N/m2 = 1493m/s 1000kg/m3 elastic property = inertial property B ρ (3. the pressure is about 4 × 107 N/m2 . 1482 m/s at 20◦ C. 1997: 468. Physics.000 meters. For example. 1967-1968: E37 Value [m/sec] 1492 1496 1494 Table 3.8% even under this pressure nevertheless it is a change. 601 Handbook of Chemistry and Physics. The ratio of the change in the fractional volume to pressure or compression is referred to as the bulk modulus of the material. At a depth of about 4. Many researchers have looked at this velocity. the average bulk modulus for water is 2.5). The fractional volume change is only about 1. The mathematical definition of bulk modulus as following B=ρ In physical terms can be written as c= For example for water c= 2. New York: Wiley. John D. For example. According .3. The World Book Encyclopedia. The compressibility of the substance is the reciprocal of the bulk modulus. Johnson. Chicago: World Book. & Kenneth W. SPEED OF SOUND IN ALMOST INCOMPRESSIBLE LIQUID 35 3. as can be observed from the equation (3. 1999. and for purposes of comparison it is given in Table (3.37).

using a general tabulated value for the bulk modulus. Soc. salinity.36 CHAPTER 3.37) where c0 = 1449. and hydrostatic pressure. Acoust. John. 1960.14 is about clean/pure water. Nevertheless. Speed of sound in solid of steel. SPEED OF SOUND to Wilson5 .2: Liquids speed of sound. 1357. (3. with different speeds in different directions. 5 J. 1972 In summary. p. cT is a function temperature.32. the speed of sound in liquids is about 3 to 5 relative to the speed of sound in gases. and cST P is a correction factor between coupling of the different parameters. T. Wilson’s empirical formula appears as follows: c(S. in different kinds of geometries.38) Compared to one tabulated value the example values for stainless steel lays between the speed for longitudinal and transverse waves. N 10. and cS is a function salinity. material Glycerol Sea water Mercury Kerosene Methyl alcohol Carbon tetrachloride reference 25 ◦ C Value [m/sec] 1904 1533 1450 1324 1143 926 Table 3. gives a sound speed for structural steel of E = ρ 160 × 109 N/m2 = 4512m/s 7860Kg/m3 c= (3. after Aldred. Manual of Sound Recording. vol. the speed of sound in solids is larger than in liquids and definitely larger than in gases.. and differences between transverse and longitudinal waves. Wilson’s formula is accepted by the National Oceanographic Data Center (NODC) USA for computer processing of hydrological information. cP is a function pressure. London: Fountain Press. Young’s Modulus for a representative value for the bulk modulus for steel is 160 109 N /m2 . 3.6 Speed of Sound in Solids The situation with solids is considerably more complicated. P ) = c0 + cT + cS + cP + cST P . . Amer. the speed of sound in sea water depends on temperature.

and liquid with some bubbles. In that case. there could be more than one speed of sound for two phase flow.3: Solids speed of sound. 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 ˙ where ξ = mb is the mass ratio of the materials. Manual of Sound Recording. Topic for none homogeneous mixing are beyond the scope of this book. three extreme cases suggest themselves: the flow is mostly gas with drops of the other phase (liquid or solid).40) can be approximated as ρ =1+m ρa (3. There can be several models that approached this problem. for homogeneous and under certain condition a single velocity can be considered.41) . The first case is analyzed. It further assumed that no heat and mass transfer occurs between the particles. John. In actuality. m ˙ For small value of ξ equation (3. 1972 3. For simplicity. about equal parts of gas and the liquid phase. However. London:Fountain Press.3.39) (3. Indeed there is double chocking phenomenon in two phase flow.40) (3. after Aldred. it assumed that two materials are homogeneously mixed. 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 3.7.7 Sound Speed in Two Phase Medium The gas flow in many industrial situations contains other particles.

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 (3.42) 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. Generally. For example. It must noticed that m = constant. 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. The gas density can be replaced by equation (3. It can be noticed that Rmix and γ are smaller than similar variables in a pure gas.6).41) R P = T ρ 1+m (3.47) (3.39) and substituted into equation (3.44) Therefore. the velocity of sound in slightly wed steam can be about one third of the pure steam speed of sound.48) At this stage the other models for two phase are left for next version (0.46) into γ= Cp + mC Cv + mC (3. If the droplets (or the solid particles) can be assumed to have the same velocity as the gas with no heat transfer or fiction between the particles isentropic relation can be assumed as P = constant ρa k (3. the velocity of mixtures with large gas component is smaller of the pure gas.43) Assuming that partial pressure of the particles is constant and applying the second law for the mixture yields droplets gas 0 = mC dP (Cp + mC)dT dP dT dT + Cp −R = −R T T P T P γ−1 γ (3.45) In a way the definition of γ was so chosen that effective specific pressure heat C +mC and effective specific volumetric heat are p and Cv +mC respectively. Hence. .38 mb ˙ ma ˙ CHAPTER 3. SPEED OF SOUND where m = is mass flow rate per gas flow rate.

3. speed of sound can be expressed as c2 = where X is defined as X= s − sf (PB ) sf g (PB ) (3.50) ∂P ∂P [f (X)] = ∂ρ ∂ρ (3.7. SOUND SPEED IN TWO PHASE MEDIUM Meta 39 For a mixture of two phases.49) Meta End .

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

˙ Q = Cp (T0 B − T0 A )m ˙

(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 ture, T0 is provided in power series. Expanding of the equations according to the 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

Examination of the relation between properties can then be carried out.28) Differentiation of the equation state (perfect gas). the energy equation.25) with (4. the stagnation density is constant through the flow. Therefore.26) The thermodynamic relationship between the properties can be expressed as T ds = dh − dP ρ (4. is constant through the adiabatic flow because there isn’t heat transfer. T ρ P U T+dT ρ+dρ P+dP U+dU 4.4). Conversely. Second is isentropic and isothermal Fig.11) are the same. equation (4. The control volume is shown in Figure (4.29) .4: Control volume inside a convergingmodel. reads dh + U dU = 0 Differentiation of continuity equation.46 CHAPTER 4. Thus. T0 . ISENTROPIC FLOW The important sub case in this chapter is the flow in a converging–diverging nozzle. it is constant for the other. perature.27) For isentropic process ds ≡ 0 and combining equations (4.25) hen there is no external work and heat transfer. If the right hand side is constant for one variable.2. 4. and dividing the results by the equation of state (ρRT ) yields dP dρ dT = + P ρ T (4. Clearly. the stagnation temdiverging nozzle. ρAU = m = constant.27) yields dP + U dU = 0 ρ (4.9) and equation (4. In the same argument. in mathematical terms.1 The Properties in the Adiabatic Nozzle (4. P = ρRT . and dividing by the ˙ continuity equation reads dρ dA dU + + =0 ρ A U (4. There are two models that assume variable area flow: First is isentropic and adiabatic model. knowing the Mach number or the temperature will provide all that is needed to find the other properties. the stagnation pressure is also constant through the flow because the flow isentropic. The only properties that need to be connected are the cross section area and the Mach number.

33) is a differential equation for the pressure as a function of the cross section area. Now. dρ. If Mach number is smaller than .32) Equation (4.34) The pressure Mach number relationship Before going further in the mathematical derivation it is worth looking at the physical meaning of equation (4. It can be observed that the critical Mach number is one. It is convenient to rearrange equation (4.34).31) = U2 dA A (4.31) to obtain dP ρ Or in a dimensionless form dP dA 1 − M2 = U2 ρ A (4.26) and using it in equation (4.30) so that the density.28) reads dU U dP dA dρ − U2 + =0 ρ A ρ (4. If the Mach number is larger than one than dP has opposite sign of dA.33) 1− U c 2    dρ dP   =U  + A dP ρ    2  dA  (4. The meaning of the sign change for the pressure differential is that the pressure can increase or decrease.2. can be replaced by the static pressure. 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. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 47 Obtaining an expression for dU/U from the mass balance equation (4.4.33) to obtain a variables separation form of dP = dA ρU 2 A 1 − M2 (4. The term ρU 2 /A is always positive (because all the three terms can be only positive). it can be observed that dP can be positive or negative depending on the dA and Mach number.30) Rearranging equation (4.

In that case. dA > 0 ⇒ dP < 0 dA < 0 ⇒ dP > 0 This behavior is opposite to incompressible flow behavior. For the special case of M = 1 (sonic flow) the value of the term 1 − M 2 = 0 thus mathematically dP → ∞ or dA = 0. ρ. 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. From the speed of sound. ISENTROPIC FLOW one dP and dA have the same sign. The relationship between the velocity and the pressure can be observed from equation (4. not necessarily means that when dA = 0 that M = 1. the equations are applicable for any gas (perfect or imperfect gas). For the supersonic branch M > 1. Conversely. dρ = 1 dP c2 (4.35) it is obvious that dU has an opposite sign to dP (since the term P U is positive). Note that the pressure decrease is larger in compressible flow compared to incompressible flow. Thus. if the area decreases (as a function of x) the pressure decreases. increases with pressure and vice versa (see equation 4. An increase in area results in an increase of the static pressure (converting the dynamic pressure to a static pressure). the equation of state was not used.It must also be noted that when M = 1 occurs only when dA = 0. The second law (isentropic relationship) dictates that ds = 0 and from thermodynamics dP dT −R ds = 0 = Cp T P . one can observe that the density. dU = − dP PU (4.36). Hence the pressure increases when the velocity decreases and vice versa.36). However.35) From equation (4. the phenomenon is different. For M > 1 the term 1/1 − M 2 is negative and change the character of the equation.4.35 . Since physically dP can increase only in a finite amount it must that dA = 0.36) It can be noted that in the derivations of the above equations (4. the opposite. it is possible that dM = 0 thus the diverging side is in the subsonic branch and the flow isn’t choked.28) by solving it for dU .48 CHAPTER 4.

29). mass (4. Note.43) momentum equation is not used normally in isentropic process.42) changes equation (4.4.33).38) yields dT (k − 1)M dM =− T 1 + k−1 M 2 2 Relationship Between the Mach Number and Cross Section Area The equations used in the solution are energy (4. the temperature varies according to the same way that pressure does. 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. why? .40) P =− M dM 1 + k−1 M 2 2 (4. First stage equation (4.2.33) isn’t the solution but demonstration of certain properties on the pressure. 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 (4.37) T k P Thus.41) into 2 P P (4.39) is combined with equation (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 49 and for perfect gas dT k − 1 dP = (4. Differentiation of equation (4.37) and the relationship of pressure and cross section area (4.39). The relationship between temperature and the cross section area can be obtained by utilizing the relationship between the pressure and temperature (4. The relationship between the Mach number and the temperature can be obtained by utilizing the fact that the process is assumed to be adiabatic dT0 = 0. second law (4.38) and simplifying equation (4.37).39) (4.42) M2 − 1 dA = dM A M 1 + k−1 M 2 2 1 The (4. state (4. equation (4.9).41) The following identify.37) and becomes (k − 1)M dM (k − 1) dP =− k P 1 + k−1 M 2 2 Combining equation (4.26)1 .33) yields 1 k ρU 2 dA A 1−M 2 (4.40) with equation (4.

Assume that the ratio of specific heat is k = Cp /Cv = 1. an increase of the cross section decreases the velocity and Mach number (see Figure (4. In subsonic flow branch. Much nubmer the cross section area.1: Air is allowed to flow from a reservoir with temperature of 21◦ C and with pressure of 5[MPa] through a tube. The special case is when M = 1 which Fig. an object can be moved in arbitrary speed. Therefore. the factors cro A.4.43). on the other hand. In the case of M > 1 the Mach number increases with x the cross section area and vice versa. calculate the Mach number.” Again. indexsonic trasition This minimum area is referred to as “throat. the opposite conclusion that when dA = 0 implies that M = 1 is not correct because possibility of dM = 0. a decrease of the cross section increases the velocity and the Mach number.50 CHAPTER 4. 2 This condition does not impose any restrictions for external flow. from the mathematical point of view: on one hand.5)).43) is very important because it relates the geometry M. Assume that process is isentropic and neglect the velocity at the reservoir. 4. ction M 1 + k−1 M 2 and A are positive re2 se gardless of the values of M or A. ISENTROPIC FLOW Equation (4. At some point on the tube static pressure was measured to be 3[MPa]. In equation (4. the only factor that affects relationship between the cross area and the Mach number is M 2 − 1. 4. Hence the stagnation temperature can be written T0 = constant and P0 = constant and both of them are known (the condition at the reservoir). and the cross section area at that point where the static pressure was measured. the Mach number can be calculated by utilizing the pressure ratio. velocity. In external flow.2 Isentropic Flow Examples Example 4. For the point where the static pressure is known. S OLUTION The stagnation conditions at the reservoir will be maintained throughout the tube because the process is isentropic. With the known Mach number. A (area) with the relative velocity (Mach ss number).2.5: The relationship between the cross requires that dA = 0. For M < 1 the Mach number is varied opposite to M. 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.    ©   ¨ ¦ ¤ ©§¥¡ ¢ £¡ . It was measured that air mass flow rate is 1[kg/sec].

Also. T = 0.2. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 51 the temperature.69428839 × J ρ0 RT0 287.88639 0.53105 With these values the static temperature and the density can be calculated.26 × 10−5 [m3 ] ρU kRT = 0.0115 0.86420 0.4. Calculate the Mach number at point B under the isentropic flow assumption. and velocity can be calculated.4 × 287 × 294 = 304[m/sec] (4.5[Bar]. estimate the temperature at point B. Finally. In the point where the static pressure known 3[M P a] P ¯ = = 0. Assume that the specific heat ratio k = 1. 4 Well.69428 1. the cross section can be calculated with all these information.88638317 × √ 1.6 P = P0 5[M P a] From Table (4.4 and assume a perfect gas model. Downstream at point B the pressure was measured to be 1. or simply using the equations shows that M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.44) For a circular tube the diameter is about 1[cm].3) or utilizing the enclosed program from PottoGDC. Here it is subsonic and this technique is not suitable.60693 0. there is no known way for the author to directly measure the Mach number. 4 This .60000 0.076K ρ0 ρ= ρ P0 5 × 106 [P a] = 0. pressure is about two atmospheres with temperature of 250[K] this question is for academic purposes.0 kgK × 294[K] kg m3 = 41.1416 The velocity at that point is c U =M √ The tube area can be obtained from the mass conservation as A= m ˙ = 8.2: The Mach number at point A on tube is measured to be M = 23 and the static pressure is 2[Bar]4 . Example 4.2) or from Figure (4.86420338 × (273 + 21) = 254. The best approximation is by using inserted cone for supersonic flow and measure the oblique shock.

0000 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.53884934. Both solution are possible and acceptable. The supersonic branch solution is possible only if there where a transition at throat where M=1.59309 With this information the pressure at point B can be expressed as from the table 4.2 1.17040879.3: Gas flows through a converging–diverging duct. The two possible solutions: the first supersonic M = 1. At M = 2 (supersonic flow) the ratios are M 2.17040604 PB 1.272112 With the value of AB from the Table (4. Therefore.8137788 and TB = 0.55555556 271. The stagnation temperature can be “bypassed” to P0 calculate the temperature at point B M =2 M =1.4. S OLUTION To obtain the Mach number at point B by finding the ratio of the area to the critical area.. TB = T A × T0 × TA TB T0 = 250[K] × 1 × 0.60315132 PB = 0.6265306 and second subsonic M = 0.12780 0. 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.23005 1. Find the Mach number at point B.55556 0.60315132 0.6875 0.2 @ M = 2 PA = P0 PB P0 × PA 2.59014 = 1. This relationship can be obtained by AB AB AA 40 = × ∗ = × A∗ AA A 50 from the Table 4. At point “A” the cross section area is 50 [cm2 ] and the Mach number was measured to be 0.5 The corresponding Mach number for this pressure ratio is 1.0 = 0. the stagnation pressure at point A is known and stagnation temperature can be calculated. .81.42[K] Example 4.4.21567 0.52 CHAPTER 4. Assume that the flow is isentropic and the gas specific heat ratio is 1.2) or from Potto-GDC two solutions can A∗ be obtained.12780453 × = 0. At point B in the duct the cross section area is 40 [cm2 ].

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 (4.2721 0.49) Equation (4. 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. The area ratio is defined as the ratio of the cross section at any point to the throat area (the narrow area). It is convenient to rearrange the equation (4.48) by equation (4. Hence equation (4. to examine the maximum flow rate and to see what is the effect of the compressibility on the flow rate.e.53887 0.45) to be expressed in terms of the stagnation properties as f (M.46) Expressing the temperature in terms of Mach number in equation (4.22617 0.6266 0.34585 1.47) It can be noted that equation (4.28772 0. The throat area can be denoted as by A∗ . specifically and explicitly the relationship for the chocked flow.86838 1.k) m ˙ P P0 U √ = A P0 kRT k R T0 1 P √ = √0 M T T0 T0 k P R P0 T0 T (4.65396 0. dividing equations (4.94511 0. temperature.47) yields 1 A = ∗ A M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (4. pressure) do not change.2721 0.0440 4. .47) holds everywhere in the converging-diverging duct and this statement also true for the throat.46) results in m ˙ = A kM P0 √ kRT0 1+ k−1 2 M 2 k+1 − 2(k−1) (4.49) relates the Mach number at any point to the cross section area ratio.82071 1. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 53 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 1.4.2. The area ratio as a function of the Mach number needed to be established.45) This parameter is studied here.47) obtained the form √ k+1 − 2(k−1) m ˙ kP0 k−1 (4.48) = √ 1+ A∗ 2 RT0 Since the mass flow rate is constant in the duct.2.

52) is known as Fliegner’s Formula on the name of one of the first engineers who observed experimentally the choking phenomenon.51) max The maximum flow rate for air (R = 287j/kgK) becomes.53) (4.040418 A ∗ P0 (4. √ m T0 ˙ = 0.5)). Carrying this calculation results at M = 1.68473 P √0 ∼ √ T0 R (4. It can be noticed that Fliengner’s equation can lead to definition of the Fliengner’s Number. k = 1.54 CHAPTER 4.48) in with respect to M and equating to zero.55) and the maximum point for F n at M = 1 is Fn = k k+1 2 k+1 − 2(k−1) (4. (see for example. ISENTROPIC FLOW The maximum flow rate can be expressed either by taking the derivative of equation (4.48) results in F n = kM k−1 2 M 1+ 2 k+1 − 2(k−1) (4.52) Equation (4. m ˙ A∗ P √0 = T0 k R k+1 2 k+1 − 2(k−1) (4. While these .4 m ˙ A∗ 0.50) max For specific heat ratio.54) Utilizing Fliengner’s number definition and substituting it into equation (4. Shapiro (problem 4.56) “Naughty Professor” Problems in Isentropic Flow To explain the material better some instructors invented problems. which have mostly academic proposes. c0 √ m kRT0 ˙ mc0 ˙ m T0 ˙ 1 √ =√ =√ ∗P ∗P ∗P A 0 kRA 0 RA 0 k The definition of Fliengner’s number (Fn) is Fn ≡ √ mc0 ˙ RA∗ P0 Fn (4.

59) Equation (4. 5 Since    (4.9) and substituting for Mach number M = m/Aρc results in ˙ k−1 T0 =1+ T 2 Rearranging equation (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 55 problems have a limit applicability in reality. In other words.2.60) The only physical solution is when the density is positive and thus the only solution is    1 P ρ=  + 2  RT0  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 = RT0 ). ρ when all other variables are known. the terms for the group over the under–brace approaches zero when the flow rate (Mach number) is very small.61) version 0. e. The situation where the mass flow rate per area given with one of the stagnation properties and one of the static properties.59) is quadratic equation for density. The use of the regular isentropic Table is not possible because there isn’t variable represent this kind problems. Using energy equation (4.44 of this book. It is convenient to change it into ρ2 − k−1 Pρ − T0 R 2kRT0 m ˙ A 2 =0 (4. For this kind of problems a new Table was constructed and present here5 .57) 1/kR T0 ρ 2 = T ρ ρ + T c2 k−1 2 m ˙ A 2 (4.57) result in p R m ˙ Aρc 2 (4.58) transformed it into ρ2 = k−1 Pρ + T0 R 2kRT0 m ˙ A 2 (4. . P0 and T or T0 and P present difficulty for the calculations.g. The case of T0 and P This case considered to be simplest case and will first presented here.58) And further Rearranging equation (4. they have substantial academic value and therefore presented here.4.

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 = 1 + 1 + 2 ˆ 2 k2 A∗ P AP 0 2   (4.66) m ˙ A 2 = F n2 k A ∗ P0 AP 2 T T0 P0 P 2 (4. k also). And the values of A∗ P0 were tabulated in Table (4.1).2) and Fn is tabulated in the next Table (4.64) RT0 P2 m ˙ A 2 = F n2 k A ∗ P0 AP 2 (4.65) Again notice that the right hand side of equation (4.63) =f (M ) RT0 P2 Thus.56 CHAPTER 4. Thus. this dimensionless group is function of Mach number only. First. m ˙ A 2 = 1 c0 2 k P0 2 m ˙ A∗ 2 A∗ A 2 P0 P 2 (4. again the specific heat ratio. also the specific heat.61) is transformed into   2 ˙ 1 (k − 1)RT0 m  1+ 1+2 ρ= ˆ 2 kP 2 A (4. RT P0 2 It can be noticed that F n2 = k T T0 P0 P 2 Hence.62) The dimensionless density now is related to a dimensionless group that is a function of Fn number and Mach number only! Thus. T .66) is only function of Mach AP number (well. k). ISENTROPIC FLOW It is convenient to denote a new dimensionless density as ρ= ˆ ρ p RT0 = 1 ρRT0 = ¯ P T (4. and static temperature.67) (4. P0 . the problems is reduced to finding tabulated values. F n2 A ∗ P0 AP With this new definition equation (4.68) . The case of P0 and T A similar problem can be described for the case of stagnation pressure.

This example shows how a dimensional analysis is used to solve a problems without actually solving any equations.2).67) became RT P0 2 m ˙ A 2 = A ∗ P0 AP 2 (4. if it was shown that a group of parameters depends only Mach number than the Mach is determined by this group. (4.2. Based on the same arguments. 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).71) It was hidden in the derivations/explanations of the above analysis didn’t explicitly state under what conditions these analysis is correct. . ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 57 Thus equation (4. mass flow rate per area. c0 2 2 2 1 Rρ0 P m ˙ A = kRT0 P kRP0 P0 P0 m ˙ A = c0 2 P kRP0 2 P0 m ˙ A 2 = F n2 k P0 P (4. not all the analysis valid for the same conditions and is as the regular “isentropic” Table. For example the first “naughty professor” question is sufficient that process is adiabatic only (T0 .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. First. P .). the dimensionless approach is used later analytical method is discussed (under construction). 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. The heat/temperature part is valid for enough adiabatic condition while the pressure condition requires also isentropic process. Unfortunately. The actual solution of the equation is left as exercise (this example under construction).4.69) The right hand side is tabulated in the “regular” isentropic Table such (4. All the above conditions/situations require to have the perfect gas model as the equation of state.

51000 0.23137 0.25000 0.41073 1.39000 0.28000 0.22634 0.78382 0.35000 0.40000 0.050001 0. ISENTROPIC FLOW Table 4.002 0.017813 0.18428 0.91838 1.41997 0.36550 0.013 0.42000 0.17397 0.30214 0.43919 0.89613 1.48360 0.055 0.20458 0.23155 0.028 0.079722 0.14927 0.51932 0.47000 0.27000 0.057647 0.003 0.002 0.46000 0.62E−05 0.44363 0.095449 0.33226 0.74912 1.41338 0.54733 0.10752 0.60047 0.35316 0.85261 1.37474 0.050 0.68875 0.021 0.28677 1.18992 0.059212 0.42683 1.023 0.49249 0.071 0.001 0.042 0.49285 1.007 0.57253 0.14889 0.012 0.57656 0.37896 1.54000 0.044110 0.060404 0.47609 1.074254 0.31203 0.36329 1.0 0.010476 0.34000 0.71967 0.66098 0.00865 0.031 0.14276 0.32220 0.17728 0.014268 0.21000 0.58 CHAPTER 4.400E−06 1.024585 0.14592 0.003 0.31703 1.11710 0.24000 0.62915 0.39701 0.21703 0.45951 1.072487 0.24674 0.13284 0.00707 0.0 2.59736 1.081847 0.61550 1.035 0.16522 0.065654 0.27608 0.00352 0.38000 0.20442 0.60706 0.49485 0.000424 0.0 0.033229 0.060 0.55637 0.30000 0.51882 0.67129 1.017 0.81034 1.63889 .36000 0.46677 0.41000 0.001 0.46633 0.25535 0.006 0.65246 1.26495 0.12039 0.005 0.015 0.005 0.077 0.34775 1.30418 0.11928 0.92149 0.87421 1.094654 0.52858 0.14084 1.050518 0.000 0.39478 1.53000 0.20109 0.52690 1.33000 0.88588 0.75136 0.012593 0.32000 0.50000 0.83132 1.31000 0.083989 0.44192 0.73995 0.29247 0.0 0.41855 0.00E+00 1.011 0.48000 0.15963 0.44309 1.028651 0.49000 0.76924 1.57709 0.014197 0.29663 0.63386 1.20000 0.000 0.34330 0.78965 1.065 0.25018 0.070106 1.37432 0.33378 0.046 0.001 0.65857 0.69036 1.27358 0.10397 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 Rρ0 2 T m 2 ˙ A 0.81706 0.004 0.089910 0.065557 0.40333 0.13232 0.019 0.074314 0.44215 0.015027 0.45000 0.17381 0.13796 0.18896 0.38884 0.13342 0.23000 0.00351 0.54422 1.36764 0.56172 1.20316 0.12239 0.038365 0.72927 1.16581 0.038 0.12724 0.16121 0.21584 0.39596 0.95791 0.58952 0.54531 0.37000 0.43000 0.00747 0.23777 0.087372 0.098460 0.003 0.000 0.31480 0.70969 1.29000 0.067111 0.69948 0.44000 0.62436 0.15372 0.020986 0.49305 0.33465 0.85107 0.27926 0.26264 0.10639 0.008 0.46798 0.55000 0.22085 0.28307 0.18709 0.50978 1.57944 1.24773 0.11294 0.30185 1.26000 0.52000 0.009 0.10000 0.33233 1.029920 0.026 0.22000 0.52485 0.35361 0.

278 1.971 3.699 3.018 2.84000 0.383 1.305 1.508 1.044 1.104 3.924 1.109 1.740 2.217 1.513 5.657 1.805 0.74624 0.287 3.535 2.59000 0.92000 0.92366 0.74290 0.124 1.72000 0.258 4.423 1.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.610 1.94848 0.271 1.288 1.99507 1.720 2.355 1.65000 0.660 1.389 3.646 3.385 1.381 2.037 1.819 1.869 7.667 1.033 1.324 1.937 1.62936 0.67000 0.240 1.368 1.642 1.78250 0.82000 0.897 5.557 1.105 1.326 1.57000 0.096 2.738 1.86000 0.152 1.963 3.70675 0.671 2.536 2.083 1.279 1.998 3.88142 0.006 6.82722 0.343 1.225 1.113 1.587 1.541 3.211 1.977 2.135 1.671 1.955 3.141 1.090 1.747 1. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 59 Table 4.56000 0.477 1.980 2.516 1.424 6.148 1.74000 0.670 2.68565 0.840 4.94000 Fn 0.297 1.094 1.607 1.405 1.414 1.173 1.340 2.70000 0.581 1.284 2.73000 0.61000 0.422 1.761 1.072 1.736 1.703 1.084 1.99514 1.450 1.709 3.509 1.20 0.86101 0.94096 0.613 6.151 1.790 2.131 1.931 2.500 1.244 3.842 2.415 1.011 1.90382 0.87424 0.953 2.156 2.78062 0.853 1.059 1.212 1.869 1.194 1.979 4.462 3.412 2.059 2.83000 0.882 2.260 1.214 1.448 1.216 2.469 2.457 3.043 4.69000 0.554 10.721 4.500 1.661 1.865 .810 2.381 8.236 1.317 1.161 1.758 1.93000 0.712 1.187 1.177 2.407 2.198 1.89000 0.859 1.98717 1.571 4.244 5.640 1.62000 0.461 1.105 0.235 5.90000 0.64000 0.122 1.058 2.806 P0 A∗ AP 2 RT0 P2 m 2 ˙ A 1 Rρ0 P m 2 ˙ A 1 Rρ0 2 T m 2 ˙ A 0.320 1.602 2.676 1.95665 1.68000 0.949 9.972 5.427 1.81000 0.202 1.791 1.595 2.255 1.813 2.66000 0.79000 0.574 1.241 1.121 3.814 1.446 1.035 1.582 1.269 1.87000 0.80000 0.71000 0.278 2.553 1.784 1.60000 0.4.81996 0.526 1.2.81139 0.168 2.602 2.349 1.898 1.77000 0.489 2.112 1.241 2.91000 0.85000 0.607 1.474 1.342 7.538 1.029 3.67210 0.484 4.708 1.277 2.185 1.97562 1.097 1.991 2.030 1.75000 0.58000 0.362 1.081 2.323 1.78000 0.101 2.382 1.147 1.266 4.821 1.082 2.76000 0.541 1.96389 0.404 2.771 1.846 8.220 3.162 1.144 ρ ˆ 1.88000 0.63000 0.038 2.088 1.903 1.

S OLUTION The second academic condition is when the static temperature is given with the stagnation pressure.843 1.920 1.804 5.046 P0 A∗ AP 2 RT0 P2 m 2 ˙ A 1 Rρ0 P m 2 ˙ A 1 Rρ0 2 T m 2 ˙ A 3. The second model which there is significant heat transfer but insignificant pressure loss (Rayleigh flow like). Assume that the process is isentropic and k=1.500 3.037 4. it is clear that the function f (P0 .88 11. A constant flow rate requires that mA = mB . Calculate for that point the Mach number.016 Example 4.72).95000 0. in a case where the flow isn’t isentropic or adiabatic the total pressure and the total temperature will change (due to friction.72) From equation (4. There are two possible models that can be used to simplify the calculations.436 6.98 6. .324 2.4: A gas flows in the tube with mass flow rate of 1 [kg/sec] and tube cross section is ◦ 0.99000 1.259 3.3.96000 0.06 14. The first model for neglected heat transfer (adiabatic) flow and in which the total temperature remained constant (Fanno flow like).001[m2 ].961 2.37 13.19 14.98000 0.47) is appropriate regardless the flow is isentropic or adiabatic.371 2.583 10.003 2. T0 .112 6.233 2.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0. The third academic condition is of static temperature and the static pressure. At some point the static pressure was measured to be 1.000 Fn 2.181 3.60 CHAPTER 4.136 7.48) can be equated as kP0 A∗ RT0 k−1 2 1+ M 2 k−1 − 2(k−1) = constant (4. A∗ ) = constant.60 12.913 4. the velocity. Flow with pressure losses The expression for the mass flow rate (4.217 4. and heat transfer). Denoting sub˙ ˙ script A for one point and subscript B for another point mass equation (4.188 2.600 4.777 7. ISENTROPIC FLOW Table 4.881 1.278 2.404 4.5[Bar]. and the stagnation pressure.338 3.515 7. 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.419 3. Clearly.97000 0.419 ρ ˆ 1. The temperature at Chamber supplying the pressure to tube is 27 C . 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.

calculated the total pressure lost.05853 0.39498 1. M 1. thus the area ratios can be calculated.5.55401 0.2. comparison of mass flow rate at point A and point B leads to P 0 A ∗ |A = P 0 A ∗ |B .01[m2 ]. 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]. and utilizing equation (4. S OLUTION Both Mach numbers are known.32039 0. and the duct section area is 0.15432 0.75) 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. the cross section area is 0. The total pressure can be calculated because the Mach number and static pressure are known.1762 0.13169 2.5000 T T0 ρ ρ0 A A P P0 0.62693 A×P A∗ ×P0 F F∗ .5000 2. Downstream at exit of tube.4. With these information.68966 0.76) Example 4.015[m2 ] and Mach number is 1.6367 0.5.44444 0.27240 0. And utilizing the equality of A∗ = P 0 |A A∗ | = ∗A P 0 |B A |B A∗ A A (4. Assume no mass lost and adiabatic steady state flow. point B.73) For adiabatic flow.75) the stagnation pressure at point B can be obtained. Mach number is 2.74) leads to A A ∗ MA A A ∗ MB P 0 |A = P 0 |B A|A A|B (4.

62 CHAPTER 4.0000 0.912 1.96899 0. 0.8E + 5 11.500 0.268 0.01 1.4E+5 4.72632 0.030 1.015 4.5 A = 3 = 51.38484 0.800 0.55425 0.200 1.700 0.35036 0.094 1.100 1.008 1.77640 0.600 0.050 0.48290 0.340 1.590 1.74738 0.36091 0.00 1.99303 0.99206 0.71839 0.882 1.000 0.781 2.4 M 0.27240307 × 15.25781291 × Hence P0 |A − P0 |B = 51.59 5.53039 0.115 1.00000 0.013[Bar] Note that the large total pressure loss is much larger than the static pressure loss (Pressure point B the pressure is 0.41238 0.88652 0.68110 0.99800 0.78400 0.89699 0.93155 0.32039 A×P A∗ ×P0 F F∗ 2.8E+5 11.443 1.27240 5.188 1.53807 0.65602 0.200 0.424 1.59126 0.88517 0.58377 0.75) provides P0 |B = 51.130 0.83333 0.97250 0.39498 5.6367187 0.99875 0.009 1.52989 0.035 1.95238 0.42493 0.500 T T0 ρ ρ0 A A P P0 1.78896 0.257 − 15.99950 0.52828 0.84045 0. the stagnation at point A is obtained from Table (4.58170 0.146[Bar]).86059 0.900 1.100 0.038 1.52828 0.31424 0.98028 0.1761671 × ≈ 15.058527663 by utilizing equation (4.84302 0.89561 0.72093 0.300 0.80515 0.25781291[Bar] 0.99502 0.066 1.2) as P 0 |A = P P P0 M =2.92427 0.243 = 4.822 2.93947 0.2: Isentropic Table k = 1.3 Isentropic Tables Table 4.79158 0.91075 0.55401 .243[Bar] 2.300 1.93284 0.43742 0.57 5.63394 0.243 = 36.400 1. ISENTROPIC FLOW First.98232 0.59650 0.53974 0.838 2.400 0.47207 0.54655 0.000 1.95638 0.73999 0.00000 0.53399 0.176 1.63535 0.99825 0.52828 0.964 2.68966 1.68704 0.53114 0.46835 0.

62693 0.900 2.66138 0.73257 0.092593 0.700 1.72 16.90E−5 4.20259 0.23810 0.017321 0.2: Isentropic Table k=1.000495 1.44444 0.56 25.15432 0.000385 0.600 1.3.69983 0.0E+2 1.14184 0.000633 0.28682 0.63371 0.052493 0.00107 0.28986 0.000242 0.027662 0.057227 0.022046 0.013957 0. .21567 0.12780 0.3.29414 0.235 6.4.028962 0.500 7.011340 0.74058 0.25044 0.081633 0.58549 0.19802 0.000815 0.00346 0.000 5.00107 0.637 4.1 4.3E+2 4.73723 0.31969 0.058140 0.2E+2 5.9E+2 2.00189 0.11528 0.800 1. Again in reality the heat transfer is somewhere in between the two extremes.4E+2 1. knowing the two limits provides a tool to examine where the reality should be expected.74E−5 3.77) into the momentum equation6 yields U dU + RT dP =0 P (4.000102 6.039628 0.g.35573 0.25699 0.033682 0.058528 0.00 36.076226 0.73903 0.23005 0.500 6.23527 0.16667 0.78) (4.67320 0.59309 0.027224 0.60680 0.65326 0.00758 0. dP = dρRT Substituting equation (4.015504 0. (e.68830 0.71578 0.00364 0.17404 0.790 10.4 (continue) 63 M 1.555 1.338 1.58072 0.047251 0.025156 0.500 3.3.5E+2 3. ISENTROPIC TABLES Table 4.87 53.013111 0.77) 6 The one dimensional momentum equation for steady state is U dU/dx = −dP/dx+0(other effects) which are neglected here.000 6. Eckert number is very small) is presented.56182 0.500 4.000 4.000 3.00659 0.55556 0.00141 0.72136 0.31E−5 2.019473 0. The perfect gas model is again assumed (later more complex models can be assumed and constructed in a future versions).72586 0.56976 0.250 1.439 1. the other extreme case model where the heat transfer to the gas is perfect.070595 0.36E−5 0.00519 0.72953 0.000631 0.35714 0.500 5.00 T T0 ρ ρ0 A A P P0 0.000155 0.072464 0.688 2.18 75.045233 0. So.00261 0.047619 0.017449 0.00190 0.70876 0.089018 0.2 Isentropic Isothermal Flow Nozzle General Relationship In this section.000 9.000 8.500 10.000 7.13 1.012628 A×P A∗ ×P0 F F∗ 0.27099 0.4E+2 0.74192 4.14924 0.10582 0.500 9.064725 0.73510 0.500 8.13169 0.23211 0.57768 0.12195 0.000 2. In isothermal process the perfect gas model reads P = ρRT .

P2 k(M2 2 − M1 2 ) = ln 2 P1 (4. U2 2 − U 1 2 P2 + RT ln =0 2 P1 (4. ISENTROPIC FLOW Integration of equation (4.78) yields the Bernoulli’s equation for ideal gas in isothermal process which reads .84) reads k(M1 2 −M2 2 ) P2 2 = =e P1 e M1 2 e M2 2 k 2 (4.79) Thus.64 CHAPTER 4.87) .85) As oppose to the adiabatic case (T0 = constant) in the isothermal flow the stagnation temperature ratio can be expressed 1 ! ¡ T0 1 T1 1 + = ¡ T0 2 T ¡2 1+ 2 k−1 2 M1 2 k−1 2 M2 = 1+ 1+ 2 k−1 2 M1 2 k−1 2 M2 (4.81) Or in explicit terms of the stagnation properties the velocity is U= 2RT ln P P0 (4.80) The velocity at point 2 for stagnation point.82) Transform from equation (4.83) yields . U1 ≈ 0 reads U2 = 2RT ln P2 P1 (4.83) Simplifying equation (4. the velocity at point 2 becomes U2 = 2RT ln P2 − U1 2 P1 (4. T 2 P1 (4.79) to a dimensionless form becomes constant constant    T (M kR  2 2 − M1 2 ) P2    = R ln .84) Or in terms of the pressure ratio equation (4.86) Utilizing conservation of the mass AρM = constant to yield M 2 P2 A1 = A2 M 1 P1 (4.

93) Of course in isothermal process T = T ∗ .88) The change in the stagnation pressure can be expressed as P0 2 P2 = P0 1 P1 1+ 1+ 2 k−1 2 M2 2 k−1 2 M1 e M1 = 2 e M1 2 k 2 (4.85) yields M1 A2 = A1 M2 65 e M2 2 e M1 2 k k−1 k 2 (4.4.85) to read (1−M 2 )k ρ P = ∗ =e 2 ∗ P ρ (4.3. at this stage. the critical stagnation pressure reads 2 k−1 (1−M )k 2 1 + P0 2 M1 =e 2 P0 ∗ k+1 k k−1 k k−1 (4. All these equations are plotted in Figure (4. From the Figure 4.91) (4.87) and equation (4. The minimum of the curve is when area is minimum and at the point where the flow is choked.89) The critical point. 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.6). The critical pressure ratio can be obtained from (4. Here the critical point defined as the point where M = 1 so results can be compared to the adiabatic case and denoted by star. Again it has to emphasis that this critical point is not really related to physical critical point but it is arbitrary definition. is unknown (at what Mach number the nozzle is choked is unknown) so there are two possibilities: the choking point or M = 1 to normalize the equation. ISENTROPIC TABLES Combing equation (4.94) .90) Equation (4.3 it can be observed that minimum of the curve A/A∗ isn’t on M = 1.92) (4. The true critical point is when flow is choked and the relationship between two will be presented. The mathematical procedure to find the minimum is simply taking the derivative and equating to zero as following d A A∗ dM = kM 2 e k(M 2 −1) 2 M2 −e k(M 2 −1) 2 =0 (4.88) 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 + k−1 M1 2 T0 2 = T0 ∗ k+1 Finally.

dividing the √ critical adiabatic velocity by k results in Uthroatmax = √ RT (4.5 0 0 0. Thus. M = √ k (4.5 3 2.5 4 k=14 P/P * A/A * P0 / P0 T 0 / T0 T/T * * * Tue Apr 5 10:20:36 2005 Fig.6: Various ratios as a function of Mach number for isothermal Nozzle Equation (4. the pressure loss in adiabatic flow is milder as can be seen in Figure (4.5 3 3.5 2 M 2.96) On the other hand.5 2 1.5 1 1.7(a)).66 CHAPTER 4. Utilizing equation .5 1 0.94) simplified to 1 kM 2 − 1 = 0 .95) It can be noticed that a similar results are obtained for adiabatic flow. It should be emphisized that the stagnation pressure decrees. ISENTROPIC FLOW Isothermal Nozzle 4 3. The velocity √ at the throat of isothermal model is smaller by a factor of k. 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. 4.

Again. as opposed to the adiabatic case.5 Distance (normalized distance two scales) 2 M isoT M isentropic Uisntropic/UisoT k=14 k=14 Tue Apr 5 10:39:06 2005 Thu Apr 7 14:53:49 2005 (a) Comparison between the isothermal nozzle and adiabatic nozzle in various variables (b) The comparison of the adiabatic model and isothermal model Fig.5 3 3. the heat transfer is expressed as (4.5 4 A / A iso * A / A adiabatic * P / P iso * P / P adiabatic * 67 Comparison between the two models 5 4.4.5 3 2. It also has to be emphasized that the stagnation values of the isothermal model are not constant.5 0 0 0.97) e− 2 = 0.5 1 0.5 2 M 2.5 1 1.90) the following relationship can be obtained P ∗ Pthroat Pthroat = = P0initial P0initial P ∗ 1 e (1−02 )k 2 e „ 1− “ 1 √ k ”2 « k 2 = (4.5 1.5 4 3.7: The comparison of nozzle flow (4.5 0 0 1 0.5 2 1.5 1 0.60653 1 Notice that the critical pressure is independent of the specific heat ratio. ISENTROPIC TABLES Isothermal Nozzle 4 3. 4.5 3 2.98) Q = Cp (T02 − T02 ) .5 2 1. k.3.

4. The ratio of the velocities can be expressed as √ Ms kRTs Us √ = (4. As can be observed from Figure (4.4 bounded.4).68 CHAPTER 4.2 the distance from the entrance to the throat is normalized (to 0 0 1 2 0. The Mach number was computed for the two models and plotted in Figure (4.7(b)). In this Fig.100) the velocity ratio was obtained and is plotted in Figure (4. the isentropic for the subsonic branch will be over prediction. This P / P0 isentropic profile isn’t an ideal profile 0. using the isentropic model results in under prediction of the actual results for the velocity in the supersonic branch. The initial stagnation temperatures are almost the same and can be canceled out to obtain Us ∼ UT MT Ms 1+ 2 k−1 2 Ms (4. . The prediction of the Mach number are similarly shown in Figure (4.6 between the two models so in an actual situation it can be 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. 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.100) By utilizing equation (4.5 one (1)). The Mach number for the isentropic is larger for the supersonic branch but the velocity is lower. Thus. To make sense and eliminate unnecessary details 0.7(b)).7(b)). While. 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).99) UT MT kRTs It can be noticed that temperature in the isothermal model is constant while temperature in the adiabatic model can be expressed as a function of the stagnation temperature. In this comparison it has to be remembered that critical area for the two models are different by about 3% (for k = 1.8 T / T0 isentropic but rather a simple sample P / P0 isothermal T/T0 isothermal just to examine the difference 0.5 1.7(b)).

8) shows that the range between the predicted temperatures of the two models is very large.102) and for the isothermal process the stagnation pressure varies and has to be taken into account as the following: isentropic P0 P0z Pz = P0int P0int P0 ∗ ∗ Pz P0z (4. ISENTROPIC TABLES 69 Two other ratios need to be examined: temperature and pressure. The stagnation pressure is 5[Bar] and the stagnation temperature is 27◦ C.103) where z is an arbitrary point on the nozzle. To demonstrate the relativity of the approach advocated in this book consider the following example.4. Calculate the velocity at the exit using the adiabatic model. The pressure ratio for the isentropic model is P = P0 int 1+ 1 k−1 2 2 M k−1 k (4. The initial stagnation temperature is denoted as T0 int . with known area ratio and known stagnation Potto–GDC provides the following table: . 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. Thus. Figure (4. the sought ratio is provided.6: Consider a diverging–converging nozzle made out of wood (low conductive material) with exit area equal entrance area.4. The throat area ratio to entrance area is 1:4 respectively. Example 4.101) While the temperature ratio of the isothermal model is constant and equal to one (1). Using equations (4. The temperature ratio of T /T0 int can be obtained via the isentropic model as T 1 = k−1 T0 int 1 + 2 M2 (4.89) and the isentropic relationship. If the nozzle was made from copper (a good heat conductor) a larger heat transfer occurs. the conservation of the stagnation properties. Assume that the back pressure is low enough to have supersonic flow without shock and k = 1. while the range between the predicted pressure by the two models is relatively small.e. 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.3.

0000 4.104) As was discussed before.9402 1. the maximum velocity cannot exceed the 691.4940 0.98511 0.9405 0.70 M 0.4 4. The exit velocity. ISENTROPIC FLOW A A P P0 A×P A∗ ×P0 0. Using the area ratio in Figure (4.9).99572 0.0000 0. the initial stagnation temperature is given as 300K.9910 1.08129 4.9402 T T0 ρ ρ0 CHAPTER 4. the velocity can be calculated.02979 3. but in the literature some denote this function as I.14655 2.777777778 1 T0 ∗ 1 + k−1 k 1 + k−1 2 k Thus the stagnation temperature at the exit is T0ini = 1.50225 The exit Mach number is known and the initial temperature to the throat temperature ratio can be calculated as the following: 1 T0ini 1 = = = 0.4.9 ∼ 617.4 × 287 × 109.36644 × 300 = 109.98934 0.12556 0. 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 ˙ (4.93[m/sec] 1. is √ √ U = M kRT = 2.2K.921 T0exit The exit stagnation temperature is 1.36644 0. The Impulse function is denoted here as F . the velocity in the copper nozzle will be larger than the velocity in the wood nozzle.0 = 691. To explain the motivation for using this definition consider the calculation of the net forces that acting on section shown in Figure (4.51183 4. The exit velocity can be determined by utilizing the following equation √ √ Uexit = M kRT = 1.9910 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 With the known Mach number and temperature at the exit.92 × 300 = 576.1 The Impulse Function Impulse in Isentropic Adiabatic Nozzle One of the functions that is used in calculating the forces is the Impulse function.0000 0.105) .9K. The exit temperature is 0.11915 Even for the isothermal model.777777778 = 1.6) or using the Potto–GDC obtains the following table M 1. then.4940/0. However.4 × 287 × 300.253[m/sec] 4.253[m/sec] (4.

4.106) and by utilizing the identity in equation (4. The mass conservation also can be applied to our control volume m = ρ 1 A 1 U1 = ρ 2 A 2 U2 ˙ (4. if the right hand side is only a function of the Mach number and k than the left hand side must be function of only the same parameters.108) which makes the ratio a function of k and the Mach number. 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.107) P2 A 2 P1 A 1 Fnet = 1 + kM2 2 − 1 + kM1 2 ∗ ∗ P0 A P0 A P0 A ∗ (4.108) P1 A1 1 + kM1 F = ∗ ∗ F∗ P A (1 + k) 2 (4.108) Examining equation (4. Defining a function that depends only Fig. M and k.110) = 1 P∗ P0 2 ( k+1 ) k−1 k P1 A 1 1 + kM1 2 P0 A ∗ 1 (1 + k) (4. Thus. Hence. the net force is Fnet k+1 = P0 A (1 + k) 2 ∗ k k−1 F2 F1 − ∗ ∗ F F (4. Hence. THE IMPULSE FUNCTION 71 The net force is denoted here as Fnet .105) with equation (4.42) results in Rearranging 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 (4.109) (4.4. k. 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 (4.112) .108) shows that the right hand side is only a function of x-direction Mach number and specific heat ratio.106) Combining equation (4.111) This ratio is different only in a coefficient from the ratio defined in equation (4.

72 CHAPTER 4.1 A ∗ P0 70061.27353 0.009m2 2 A2 = 0.1949 P0 P1 = 5.058 0.52).0 × 400 × 287 A ∗ P0 = = ∼ 70061.2227 The pressure at point 1 is P1 = P 2 0. A2 A1 0. To obtained the Mach number. provides the following √ √ m RT ˙ 1.2121 With the area ratio of A A 0.003[m ].Fig.99132 5.2121 the area ratio of at point 1 can be calculated.10: Schematic of a flow of a compressible subzle and pressure at point 1.96355 2. To obtain this ratio the denominator is needed to be obtained. 4.0times0.003m2 entrance area is 0.94934/0.7: 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 = 0.99380 5. the ratio of P1 A1 /A∗ P0 is needed to be calculated. The stagnation temperature is 400K and the pressure at point 2 was measured as 5[Bar] Calculate the net force acting on the noz.99380 ∼ 4.776[Bar] P2 P0 .98526 0.1000 0. stance (gas) thorough a converging nozzle S OLUTION The solution is obtained by getting the data for the Mach number.003 And utilizing again Potto-GDC provides M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.96666 = 2.2121 × = 5.94934 2.009[m ] and T0 = 400K 2 P2 = 50[Bar] the exit area is 0.2227 A A A2 0.7) 0.058 and 500000 × 0.11164 0. ISENTROPIC FLOW To demonstrate the usefulness of the this function consider a simple situation of the flow through a converging nozzle Example 4.76 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 1 for example (4.009 A1 = = 2.76[N ] 0.003 A 2 P2 = ∼ 2. Utilizing Fliegner’s equation (4.99751 0.1774 2.

949 5.225 A×P A∗ ×P0 F F∗ .5 × (2.00 4.2E+5 20.05 0.0E+6 4.23.4.4 × 1.112) Fnet = P2 A2 P0 A ∗ k + 1 k−1 F2 F1 (1 + k) − ∗ P2 A 2 2 F∗ F 1 × 2.00 0.96666) ∼ 614[kN ] = 500000 × 2.010 2.1 T0 T0 P0 P0 A A P P 0.014 2.064 5. the definition of the Impulse function is reused.0E + 5 9.114) Since U 2 /RT = kM 2 and the ratio of equation (4.362 10.53199 1.52921 0. The ratio of the impulse function for two points on the nozzle is P2 A 2 + ρ 2 U 2 2 A 2 F2 = F1 P1 A 1 + ρ 1 U 1 2 A 1 (4.001 2. The same is done here for the isothermal nozzle flow model.113) Utilizing the ideal gas model for density and some rearrangement results in P2 A 2 1 + F2 = F1 P1 A 1 1 + U2 2 RT U1 2 RT (4.1 k 73 4.4.5 Isothermal Table Table 4. ISOTHERMAL TABLE The net force is obtained by utilizing equation (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. As previously.000 1.064 1.2 The Impulse Function in Isothermal Nozzle Previously Impulse function was developed in the isentropic adiabatic flow.00 8.115) 4.116) (4.064 1.87) transformed equation into (4.5.3: Isothermal Table M 0.52828 0.1949 − 0.

0 1.287 1.250 1.000 5.020 1.000 4.0 0.142 1.719 5.89348 1.608 3.500 5.431 4.247 0.000 9.00370 0.599 0.0 8.111 1.13333 0.41686 0.540 4.62500 0.763 1.500 10.000 0.99967 1.000 1.007 5.690 1.015317 7.0 1.4E + 3 2.000 3.125 1.500 2.1E + 3 0.21 1.97274 1.328 1.281 1.2E + 3 7.1E + 4 1.80 1.62665 0.586 5.26634 0.14286 0.0 5.281 1.800 1.8E + 2 4.194 1.50 3.00 T0 T0 P0 P0 A A P P 0.389 1.000 0.000 1.52632 0.0 0.047 1.079 1.0 7.375 1.665 0.004 0.063 1.026 19.333 2.6E + 4 2.75344 1.0 5.6E + 5 1.41E−5 4.61693 0.33333 0.41 40.055 1.0 0.70 1.94147 1.3E + 2 1.0 5.97376 1.9 1.271 0.00221 9.2E + 3 3.3 0.667 1.20 1.143 4.73278 0.7E + 3 1.0 0.98750 0.33554 0.7 0.021 0.16090 0.75E−5 1.4E+26 0.500 4.00 1.044 1.100000 A×P A∗ ×P0 F F∗ 2.000 3.12500 0.2 0.958 1.56232 0.000 8.237 1.0 0.20000 0.6 0.6E+13 0.80844 1.50000 0.245 2.50 1.464 1.5E + 2 0.4E+20 0.86329 0.50618 4. ISENTROPIC FLOW Table 4.78 0.80528 0.8E+18 0.4E + 8 0.429 1.8 0.2E+29 1.58985 0.2E + 4 1.90 2.6 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- .22222 0.55556 0.3E + 9 0.553 1.429 1.114 1.500 7.565 1.438 2.18182 0.10526 0.083 0.000 7.284 3.047 0.11765 0.28571 0.97156 1.0 0.053 0.000 3.74 CHAPTER 4.60 1.000 6.681 1.0 0.0 0.891 1.3: Isothermal Table (continue) M 0.10 1.500 6.059 1.15385 0.98796 1.035 3.12246 0.161 2.29 80.83333 0.025349 0.90909 0.14 0.16667 0.062 1.134 9.134 0.569 3.40000 0.0 5.889 2.5 0.30 1.000 1.85853 1.036 0.0 1.3E+23 0.275 1.718 3.51069 0.90302 1.54322 0.11111 0.296 5.064 2.0E + 6 0.5E + 2 2.071758 90.863 0.939 2.399 0.209 0.128 1.183 1.3E+15 0.0 0.73492 0.625 1.875 4.41E−6 0.66667 0.25000 0.000215 1.4 0.000 0.3E+11 0.76923 0.200 1.67383 0.58824 0.856 4.9E + 2 8.56954 3.065 1.000 1.69449 2.00 2.500 8.22881 15.71429 0.0 0.0 2.564 1.63276 2.500 9.153 1.4E + 3 5.40 1.000380 2.0 1.20846 0.

Differentiating equation (3.119) The definition of pressure specific heat is Cp ≡ Maxwell relation hence. ∂h ∂P =v−T ∂s ∂T T P First.120) P Equations (4. that for perfect gas the n is substituted by k. the differential of enthalpy is calculated for real gas equation of state as dh = Cp dT − T Z ∂z ∂T dP ρ (4.4. a real gas equation can be expressed in equation (3.27) is reused to related the entropy change to the change in thermodynamics properties and applied on non-ideal gas.124) . The following power relationship is assumed ρ = ρ0 P P0 1 n (4.19) and dividing by equation (3.121) P The mechanical energy equation can be expressed as d U2 2 =− dP ρ (4. To carry the integration of the right hand side the relationship between the pressure and the density has to be defined. Since ds = 0 and utilizing the equation of the state dh = dP/ρ. With integration of equation (4.19) are combined to form Cp dT ds = −z 1+ R R T T Z ∂z ∂T dP P (4. Gibb’s equation (4. P ) and full differential is dh = ∂h ∂T dT + P ∂h ∂P dP T ∂h ∂T (4.123) Notice. Thus. The enthalpy is a function of the temperature and pressure thus.122) At the stagnation the definition requires that the velocity is zero.19).117) Again.118) and second derivative is (4.6.122) when using relationship which is defined in equation (4. h = h(T. THE EFFECTS OF REAL GASES 75 erties like the critical pressure and critical temperature.123) results U2 = 2 P1 P0 dP = ρ P P0 1 ρ0 P0 P 1 n dP (4.27) and (3.19) yields dz dρ dT dP = + + P z ρ T (4.

125) For n > 1 the integration results in 2n 1− n−1 n−1 P ( n ) P0 U= z0 RT0 (4.132) .128) Where ρ∗ is the density at the throat (assuming the chocking condition) and A∗ is the cross area of the throat.19) results U2 = 2 P P0 z0 RT0 P0 P0 P 1 n dP (4. Thus.127) It must be noted that n is a function of the critical temperature and critical pressure.34) to defined the Mach number as M=√ U znRT (4.129) For the case of n = 1 ρ∗ U ∗∗ 1 n m = A∗ ˙ P0 z0 RT0 P P0 2z0 RT0 ln P0 P (4.76 CHAPTER 4. The mass flow rate is regardless to equation of state as following m = ρ ∗ A∗ U ∗ ˙ (4. the mass flow rate in our properties ρ∗ U∗ P0 m=A ˙ z0 RT0 ∗ P P0 1 n 2n z0 RT0 1− n−1 n−1 P ( n ) P0 (4.126) For n = 1 the integration becomes U= 2z0 RT0 ln P0 P (4.130) The Mach number can be obtained by utilizing equation (3.121) when ds = 0 results T2 T1 Cp dT = R T P2 z 1+ P1 T Z ∂z ∂T P dP P (4.131) Integrating equation (4. ISENTROPIC FLOW Substituting relation for stagnation density (3.

134) It was shown in Chapter (3) that (3.132) looks at Bernnolli’s equation which is dU 2 =− 2 After integration of the velocity dU 2 =− 2 P/P0 1 dP ρ (4.139) Fliegner’s number for n = 1 is Fn = mc0 ˙ =2 A ∗ P0 P∗ P0 2 − ln P∗ P0 (4. THE EFFECTS OF REAL GASES 77 To carryout the integration of equation (4. 2n n−1 P P0 n−1 n U= When n = 1 or when n → 1 z0 RT0 1− (4.33) is applicable for some ranges of relative temperature and pressure (relative to critical temperature and pressure and not the stagnation conditions).6.136) The mass flow rate for the real gas m = ρ∗ U ∗ A∗ ˙ A ∗ P0 m= √ ˙ z0 RT0 And for n = 1 A ∗ P0 m= √ ˙ z0 RT0 Fliegner’s number in this case is Fn = mc0 ˙ A ∗ P0 2n n−1 P∗ P0 1 n 2n n−1 P∗ P0 1 n 1− P∗ P0 (4.135) U= 2z0 RT0 ln P0 P (4.4.137) 2n n−1 2z0 RT0 ln P0 P (4.138) 1− P∗ P0 (4.140) .133) ρ0 d ρ P P0 (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. equation (4.142) To obtain the relationship between the temperature and pressure. ISENTROPIC FLOW n n−1 (4.78 The critical ratio of the pressure is P∗ = P0 2 n+1 CHAPTER 4.145) For n = 1 the Mach number is M= 2 z 0 T0 P 0 ln z T P (4.141) When n = 1 or more generally when n → 1 this is a ratio approach P∗ √ = e P0 (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 (4.147) for n = 1 T0 = T The critical temperature is given by T∗ = T0 eM [z+T ( 2 ∂z ∂T )P ] (4.143) The power of the pressure ratio is approaching that T0 z0 = T z P0 P k−1 k 1−n n when z approaches 1.148) ∂z n z+T ( ∂T ) ] P 1 + n ( 1−n )[ 2 (4. Note (4.149) .132) can be integrated T0 = T P0 P R Cp ∂z [z+T ( ∂T )P ] (4.

82K T 0. .9066 0. ii.8: A design is required that at a specific point the Mach number should be M = 2.42027 The stagnation temperature is T0 = 3.150) The mass flow rate as a function of the Mach number is P0 n m= ˙ M c0 n−1 2 1+ M 2 n+1 n−1 (4.151) For the case of n = 1 the mass flow rate is P0 A ∗ n m= ˙ c0 e M2 n−1 2 M 1+ 2 n+1 n−1 (4.6.42027 0.61 the results are M 2. The stagnation pressure is obtained from P0 = P0 2. and temperature 300K . the pressure 2[Bar]. iii.61 P = ∼ 52.802[Bar] P 0.04943 0.152) Example 4.4. S OLUTION 1.14366 2. Are the stagnation pressure and temperature at the entrance different from the point? You can assume that k = 1.61. Calculate the stagnation pressure and the stagnation temperature.6100 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.405. the stagnation pressure is constant for isentropic flow. Calculate the area ratio between the point and the throat. THE EFFECTS OF REAL GASES and for n = 1 T∗ = T0 79 e−[z+T ( ∂z ∂T )P ] (4. i. Of course. The solution is simplified by using Potto-GDC for M = 2.04943 T0 300 T = ∼ 713.11761 2.

ISENTROPIC FLOW .80 CHAPTER 4.

In the speed of sound analysis. and the gas flows from left to right. 81 4 53 ( )' % &$ 120 " #!   . are 1 Currently under construction. without any abrupt changes. In this discussion. The second is when many sound waves “run into” each other and accumulate (some refer to it as “coalescing”) into a large difference. the sound wave can be viewed as an extremely weak shock. low pressure side. to the left and to the right of the shock. Here. and when the membrane bursts (see the discussion about the shock tube). A discussion on the shock thickness will be presented in a forthcoming section1 . A shock can occur in at least two different mechanisms. The first is when a large differflow direction ence (above a small minimum value) between the two sides of a membrane. In Figure (5. the relationship for a perfect gas is constructed. and the thickness of the shock is assumed to be very small.CHAPTER 5 Normal Shock In this chapter the relationships between the two sides of normal shock are presented. The conditions. but it can also be viewed as a one–dimensional shock wave. the flow is assumed to be in a steady state. the shock travels from the high pressure to the Fig. In fact. Of course. it was assumed the medium is continuous. c. 5.1: A shock wave inside a tube.1) a control volume for this analysis is shown. This assumption is no longer valid in the case of a shock. which is the shock wave.v.

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

5) is ln Ty Py − (k − 1) >0 Tx Px (5.. the degree of freedom is now energy. 5. continuity. This flow is dealt with in greater detail in Chapter (10). one can solve only four (4) equations (including the second law). one can have a whole range of solutions. If the energy. producing various resistance to the flow. which will produce a range of possible solutions. the intersection is shown and two solutions are obtained. U. the intersection of these two curves is what really Rayleigh Fanno line line happened in the shock. can be expressed as a function of the other parameters. which are needed to be solved. The mathematical explanation is given Chapter (9) in greater detail. and this flow is known as Rayleigh flow. equation (5.6) It can also be noticed that entropy. i. the parameters are P. The entropy increases from point x to Fig. and state equations are solved for the arbitrary value of the Ty . and ρ. It is also worth noting that the temperature at M = 1 on Rayleigh flow is larger than that on the Fanno line. The first set is the energy.83 Note that the greater–equal signs were not used. and state equations. the energy amount “added” to the shock.2: The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. the energy equation is used.2). and state equations. In the first case. which will require additional parameters. On the other hand. d b eca C B @8A u sqi rt p h g ecf V` WV Y WV TSXQSX¥U 9 7 @86 ER FE IF FE TSGQPHG¥D . The reason is that the process is irreversible. Clearly. Instead of solving all the equations that were presented. a parabola in the T –s diagram will be obtained. and Chapter (9) deals extensively with this topic. This case is called Fanno flow. T. and the second set is the momentum. This situation is similar to a frictionless flow with the addition of heat. Since the shock has no heat transfer (a special subsonic flow case of Rayleigh flow) and supersonic flow there isn’t essentially any T momentum transfer (a speshock jump cial case of Fanno flow). continuity. In Figure (5. Thus. Mathematically. point y. continuity.e. The solution of every set of these equations produces one additional degree of freedom. when the momentum equation is solved instead of the energy equation. the increase of the entropy determines the direction of s the flow. For ideal gas. and therefore no equality can exist. Now one can view these equations as two different subsets of equations. s.

Note that the Mach number is a function of temperature.8) results in Py 2 Px 2 Ux 2 = 2 2 Uy 2 R 2 Tx 2 R Ty (5.1 5. (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 .1). as follows: Px 2 Py 2 Ux 2 = Uy 2 Tx kRTx Ty kRTy cx 2 cy 2 (5.10) Note that the speed of sound on the different sides of the shock is different. and (5.11) .2 Formal Model Equations (5.3) can be converted into a dimensionless form. c2 = kRT to be used for the Mach number definition. provides a way to obtain the speed of sound definition/equation for perfect gas. The analytical solution is discussed in the next section. NORMAL SHOCK 5. The reason that dimensionless forms are heavily used in this book is because by doing so it simplifies and clarifies the solution. It can also be noted that in many cases the dimensionless equations set is more easily solved.2). 5.9) (5. at least. the equation of state yields Px Py Ux = Uy RTx RTy Squaring equation (5. numerically.1. ρ. T0 x = T0 y . Utilizing the definition of Mach number results in Py 2 Px 2 Mx 2 = My 2 Tx Ty (5.8) Multiplying the two sides by the ratio of the specific heat. and thus for known Mx all the other quantities can be determined. Ty = Tx Ty T0y Tx T0 x = 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5.1) substituting for density.1. 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.84 CHAPTER 5. k.7) All the other relationships are essentially derived from this equation.

Under the perfect gas model.13) It can also be observed that equation (5.5. This relationship can be obtained from the combination of mass.1.15) yields Py 1 + kMx 2 = Px 1 + kMy 2 (5. SOLUTION OF THE GOVERNING EQUATIONS Rearranging equation (5. From equation (5.11)) as Py 1 + P0 y = P0x Px 1 + 2 k−1 2 My 2 k−1 2 Mx k k−1 k k−1 (5.18) . ρU 2 is identical to kP M 2 because M2 ρ ρU 2 = Using the identity (5. Py M y Px M x 2 = 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5.15) P  U2   kRT = kP M 2  RT  kRT  c2   (5.16) can be interpreted as the loss of the static pressure.2) into Px + kPx Mx 2 = Py + kPy My 2 Rearranging equation (5.17) The relationship between Mx and My is needed to be solved from the above set of equations.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 (5.12) Energy equation (5. and energy equations. The loss of the total pressure ratio can be expressed by utilizing the relationship between the pressure and total pressure (see equation (4.14) The pressure ratio in equation (5. T0 y = T0 x .12) (mass) the temperature ratio can be eliminated.11) results in Ty = Tx Py Px 2 85 My Mx 2 (5.13) means that the stagnation temperature is the same.14) transforms the momentum equation (5.16) (5.13) (energy) and equation (5. momentum.

23) The density and upstream Mach number relationship can be obtained in the same fashion to became ρy Ux (k + 1)Mx 2 = = ρx Uy 2 + (k − 1)Mx 2 (5. and Rosen.e.19) Equation (5.16) and equation (5.20) is the trivial solution in which the two sides are identical and no shock wave occurs. NORMAL SHOCK Combining the results of (5. the pressure and the temperature from both sides of the nonexistent shock are the same. Mx .18) with equation (5. 1990. K.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 a symmetrical equation in the sense that if My is substituted with Mx and Mx substituted with My the equation remains the same.” Ch. for example.19) is biquadratic. i. . Thus. 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 (5.24) 3 Ireland. one solution is My = M x (5.19) as a polynomial (fourth order).86 CHAPTER 5. 9 in A Classical Introduction to Modern Number Theory. Clearly. The second solution is where the shock wave occurs.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. ”Cubic and Biquadratic Reciprocity.22) The first solution (5. M. The pressure ratio between the two sides can now be as a function of only a single Mach number. 108-137. Tx = Ty . It is also possible to cross–multiply equation (5. New York: Springer-Verlag.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. in this case. 2nd ed.20) It can be observed that equation (5.21) becomes My 2 = Mx 2 + 2k k−1 Mx 2 k−1 2 −1 (5. pp. The solution can be obtained by rewriting equation (5. Utilizing equation (5. Px = Py .

And the temperature ratio. SOLUTION OF THE GOVERNING EQUATIONS 87 The fact that the pressure ratio is a function of the upstream Mach number.5 0.1.26) which is associated with the shock wave.25) In the same way.6 0. The Maximum Conditions The maximum speed of sound is when the highest temperature is achieved.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 (5.30) .9 0.29) Mx Fig.4 0.3 0. Mx . The Star Conditions The speed of sound at the critical condition can also be a good reference velocity.7 0.2 0.3: The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. as a function of pressure ratio.5. the relationship between the density ratio and pressure ratio is k+1 1 + k−1 ρx = k+1 ρy + k−1 Py Px Py Px (5. The maximum temperature that can be achieved is the stagnation temperature Umax = 2k RT0 k−1 (5. 5.8 0.28) Based on this definition a new Mach number can be defined M0 = U c0 (5. provides additional way of obtaining an additional useful relationship. The speed of sound at that velocity is √ c∗ = kRT ∗ (5.27) My 1 0. is transformed into Ty = Tx Py Px k+1 k−1 Py Px k+1 Py k−1 Px + 1+ (5.

88 CHAPTER 5. It is therefore convenient to define the star Mach number that will be independent of the specific Mach number (independent of the temperature). c U c U = ∗ = ∗M c∗ c c c M∗ = (5. Therefore.1. NORMAL SHOCK In the same manner.3 Prandtl’s Condition It can be easily observed that the temperature from both sides of the shock wave is discontinuous. c2 U2 c∗ 2 c∗ 2 k + 1 ∗2 + = + = c k−1 2 k−1 2 2(k − 1) (5.33) Dividing the mass equation by the momentum equation and combining it with the perfect gas model yields c1 2 c2 2 + U1 = + U2 kU1 kU2 Combining equation (5.34) 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. an additional Mach number can be defined as M∗ = U c∗ (5.34) .33) and (5.36) (5.35) the following can be obtained: U1 U2 = c ∗ 2 or in a dimensionless form M ∗ 1 M ∗ 2 = c∗ 2 (5.35) (5.37) (5.32) The jump condition across the shock must satisfy the constant energy.31) 5. the speed of sound is different in these adjoining mediums.

Finding the Mach number for a pressure ratio of 8. OPERATING EQUATIONS AND ANALYSIS 89 5.47642 2. pressure.0 60.30879 and k = 1. This kind of equations requires examining Table (5. First.0272237 = 55.0 90.0 30.0 0. The working equations were presented earlier.9360 8. the velocity can readily be calculated. Using these data.7245 My 0.4 or utilizing Potto-GDC for for value of the specific heat ratio.1: Air flows with a Mach number of Mx = 3. and velocity downstream of the shock.0 70. total pressure. Also with the temperature.0 80. The typical situations in which these equations can be used also include the moving shocks.0 Py/Py.2 Operating Equations and Analysis In Figure (5. Example 5. The equations should be used with the Mach number (upstream or downstream) for a given pressure ratio or density ratio (velocity ratio). an example is provided. the Mach number after the shock.1110 3..1) for k = 1.0272237 =⇒ P0x = 1.32 is only a few mouse clicks away from the following table. and the ratio of the total pressure.38109 To illustrate the use of the above equations. My . ρy/ρx and Ty/Tx as a function of Mx Shock Wave relationship Py/Px ρy/ρx Ty/Tx S OLUTION Fig.4 × 287 × 273 = 331.5[bar] and Tx = 273K.5 [bar] and a temperature of 0◦ C goes through a normal shock. P0y /P0x .4: The ratios of the static properties of the two sides Analysis: of the shock. 5. the total pressure can be obtained (through an isentropic relationship in Table (4.0 20. P0x is known). 120.2. at a pressure of 0.5/0. the known information are Mx = 3. Ty Tx ρy ρx Py Px P0y P0 x Mx 2.0 100. The relationship that was calculated will be utilized to obtain the ratios for the downstream of the normal Px shock. i.0 50. It can be noticed that the density ratio (velocity ratio) also has a finite value regardless of the upstream Mach number.1[bar] √ √ cx = kRTx = 1.e.5. px = 1.2). P0x = 0. are plotted as a function of the entrance Mach number. Note that the My has a minimum value which depends on the specific heat ratio.0 10.3).3088 0.0 110.0 40. Tx . Calculate the temperature.2m/sec 1 2 3 Fri Jun 18 15:48:25 2004 4 5 Mx 6 7 8 9 10 .

38) This result is shown in Figure (5. This technique also has an academic value and therefore will be described in the next version (0.40) (5.47519 Ty Tx ρy ρx CHAPTER 5. The strength of the shock wave is defined as Py Py − P x ˆ = −1 P= Px Px By using equation (5.23) transforms equation (5.6[m/sec] P0y = P0y P0x × P0x = 0.0000 My 0.5 series).85714 = 257.6[m/sec] Now the velocity downstream is determined by the inverse ratio of ρy /ρx = Ux /Uy = 3.32834 × 55.3). the downstream Mach number (see equation (5. 5. it refers to a case where only a “small shock” occurs.6790 3.2.2 = 993.2 Small Perturbation Solution The small perturbation solution refers to an analytical solution where only a small change (or several small changes) occurs. which is up to Mx = 1.90 Mx 3.85714.1 The Limitations of the Shock Wave When the upstream Mach number becomes very large. In this case.38). This approach had a major significance and usefulness at a time when personal computers were not available.6/3.22)) is limited by My = 2 $ ∼0 $ X 1 + $$2 $ (k−1)Mx 2 2k k−1 b & − M1& &x 2 ∼0 = k−1 2k (5.16) and by utilizing the limit that was obtained in equation (5.8571 10.3333 0. Now. NORMAL SHOCK Py Px P0 y P0 x 2.39) into ˆ P= 2k Mx 2 − 1 k+1 (5.2.39) .09[bar] 5. The limits of the pressure ratio can be obtained by looking at equation (5.3.1[bar] = 18.32834 Ux = Mx × cx = 3 × 331. this technique is used mostly in obtaining analytical expressions for simplified models. Uy = 993. during the writing of this version of the book.

v. which this section will be dealing with. when a valve is suddenly 4 closed and a shock propagates upstream. Moving Coordinates There are more genera cases where the moving shock is created which include a change in the physical prop. The last situation is the most general case. which creates two shock stages. 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. In some industrial applications. the issue should be presented. but this book will not deal with 4 It will be explained using dimensional analysis what is suddenly open ƒ ˜‚ o m l pn5k  ¥}p{Xx € ~ y | z y g e fd  Ž ˆ i jh Œ ‹sˆp{X„ Š ‰ … ‡ † … w q t r pvusq ™ — ˜– . the shock can move downstream or upstream.Fig. On the other extreme. 5. however.24) the following is obtained: ˆ P= 2k k−1 2 k−1 ρy ρx 91 −1 −1 − ρy ρx (5.3. flow direction For example.v.3 Shock Thickness The issue of shock thickness (which will be presented in a later version) is presented here for completeness.5: Comparison between stationary shock and moving shock in ducts erties.5. This issue has a very limited practical application for most students. to convince the students that indeed the assumption of very thin shock is validated by analytical and experimental studies.41) 5.3 The Moving Shocks † ‡… y €x w Sv ’ “‘ ‰ ˆ „ #ƒ ‚ S • “” In some situations. when a valve is suddenly opened or a membrane is ruptured. the shock wave is not stationary. propagates downstream (the opposite Stationary Coordinates direction of the previous case). a shock occurs and c.2. As a general rule. 5. The shock thickness can be defined in several ways. This liquid (metal) is pushing gas (mostly) air. a liquid (metal) is pushed in two rapid stages to a cavity through a pipe system. THE MOVING SHOCKS or by utilizing equation (5. c. This kind of situation arises in many industrial applications.

The general case is where two shocks collide into one shock and moves upstream or downstream is the general case. In these changes (of opening a valve and closing a valve on the other side) create situations in which different shocks are moving in the tube. the steady state is obtained in the moving control value.92 CHAPTER 5. In cases where the shock velocity can be approximated as a constant (in the majority of cases) or as near constant. The downstream prime Mach number can be expressed c as Us Us − U y = − My = Msy − My (5.43) The temperature measured by the observer is Tx = T x Ty = T y Assuming that the shock is moving to the right. equations.44) .42) (5. Here. 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 reason for the deviation is that this choice reduces the programing work (especially for object–oriented programing like C++). the previous analysis.e. For this analysis. the prime ’ denote the values of the static coordinates. In such a case.5)) the velocity measured by the observer is Where Us is the shock velocity which is moving to the right.45) Ux = U s − U x (5. the coordinates move with the shock.47) My = cy cy Uy = U s − U y (5. 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. (refer to Figure (5. The upstream prime Mach number can be defined as Us − U x Us Mx = = − Mx = Msx − Mx (5. The problem can be reduced to the previously studied shock. Note that this notation is contrary to the conventional notation found in the literature. An observer moving with the shock will notice that the pressure in the shock is Px = P x Py = P y (5.46) cx cx It can be noted that the additional definition was introduced for the shock upstream s Mach number. Msx = Ux . to the stationary case when the coordinates are attached to the shock front. i. The reluctance to deal with the most general case is due to fact it is highly specialized and complicated even beyond early graduate students level. NORMAL SHOCK them at this stage.. and the tools developed in this chapter can be employed.

51) leads to T0y − T0x = Us Ty k − 1 Tx k − 1 (Msx − 2Mx ) − (Msy − 2My ) cx 2 cy 2 (5.53) Again. 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 (5.48) The “upstream” stagnation temperature of the fluid is T0x = Tx 1 + k−1 Mx 2 2 (5. this difference in the moving shock is expected because moving material velocity (kinetic energy) is converted into internal energy. and is “converted” or added into the total temperature.49) and the “upstream” prime stagnation pressure is P0x = Px k−1 Mx 2 1+ 2 k k−1 (5. at a very high velocity.v. The rise in the total temperature is due to the fact that a new material has entered the c. This difference can also be viewed as a result of the unsteady state of the shock.51) It should be noted that the stagnation temperature (in the stationary coordinates) rises as opposed to the stationary normal shock.50) The same can be said for the “downstream” side of the shock. Msy . an additional definition was introduced for the shock downstream Mach number. THE MOVING SHOCKS 93 Similar to the previous case.5. The relationship between the two new shock Mach numbers is cy Us Us = cx cx cy Msx = Ty Msy Tx (5. .3. The difference between the stagnation temperature is in the moving coordinates T0y − T0x = 0 (5.52) − Tx 1 + k−1 Mx 2 2 −Tx Msx and according to equation (5.

Fig. This shock is viewed by some as a reflective shock. NORMAL SHOCK 5. the shock velocity “downstream” is Us .53) can be transformed into simpler equations when Mx is zero and Us is a positive value.55) Again. in a sudden and complete closing of a valve.v. In many such cases the steady state is established quite rapidly. » ¸ º ¹ 5¥p2¸ ½ ‡¼ ¦ ¡ ¤ ¢  ¥j£¡ ¬§«G2§ ª ©¨  Sœ ˜£ 5•— – c.42) to (5. 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. for example.56) Stationary Coordinates Moving Coordinates c. ¿ ›¾ · µ ´ ¶Q³ Ç X‡Ä›‡PÀ ÆÅ Á à  Á ¯® °‡­ The prominent question in this situation is what will be the shock wave velocity for a given fluid velocity. In the moving coordinates. The relationship between the Mach number on the two sides of the shock are tied . and for a given specific heat ratio. In such a case. Ux . or Msy instead of the velocity because it combines the temperature and the velocity into one parameter. It turns out that it is very useful to use the dimensionless parameter Msx . The “upstream” Mach number reads Mx = Us + U x = Msx + Mx cx (5. The information propagates upstream in which the gas velocity is converted into temperature. The stagnation temperature increases by T0y − T0x = Us Tx k − 1 Ty k − 1 (Msx + 2Mx ) − (Msy ) cx 2 cy 2 š ²‡± ›™ ” 2“ ’ Q‘ Ÿ  ž (5. This situation arises in many cases in the industry.6: Comparison between a stationary shock and a moving shock in a stationary medium in ducts. the observer (with the shock) sees the flow moving from the left to the right. the shock is moving to the left. 5.54) The “downstream” Mach number reads My = |Us | = Msy cy (5.94 CHAPTER 5. Equations (5.v. The sudden closing of the valve must result in a zero velocity of the gas.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.3. The flow is moving to the right. The upstream is on the left of the shock.

55) by 2 95 (My ) = Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. Two numerical methods and the algorithm employed to solve this problem for given.54) and (5. Mx .58) The temperature ratio in equation Shock in A Suddenly Close Valve (5. Here no such device is needed! Luckily and hopefully. For very small values of the upstream Mach Thu Aug 3 18:54:21 2006 number. which can easily be obtained in a Fanno flow.7).7: The moving shock Mach numbers as a 1 result of a sudden and complete stop. Note that to achieve supersonic velocity (in stationary coordinates) a diverging– converging nozzle is required. 1 − 2 (the coefficient is only approximated as 0.5) as shown in Figure (5. From the same figure it can also be observed that a high velocity can result in a much larger velocity for the reflective shock.1 1 Mx the “smart” guessing of M sx.57) And substituting equation (5.57) into (5.58) but it seems utilizing numerical methods is much more simple. The solution to equation (5. Mx ∼ equation (5. Only 2 one real solution is possible. 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. .5.58) pro1 vides that Msx ∼ 1 + 2 and Msy = Fig. 1 Note. a Mach number close to one (1). Sometimes this phenomenon can have a tremendous significance in industrial applications.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). For example. THE MOVING SHOCKS through equations (5. The typical method is 0 0.48) results in f (Msx ) Mx = Tx Ty Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. the result is about double the sonic velocity of the reflective shock. is provided herein: (a) Guess Mx > 1. an analytical solution can be obtained for equation (5. 5.58) can be obtained by several numerical methods.3.

The “upstream” will be on the right (see Figure (5. (a) Guess Mx = 1 + Mx .3. and adjust the new guess Mx > 1 accordingly.v. the shock moves in a constant velocity (5. c.v. 5. (a) Stationary coordinates (b) Moving coordinates Fig. (c) calculate the Mx = Mx − Tx Ty M y (d) compare the new Mx approach the old Mx . (c) Calculate the Mx = Mx − Tx Ty M y (d) Compare to the calculated Mx to the given Mx . (b) using the shock table or use Potto–GDC to calculate the temperature ratio and My .8(b)). c.2 Moving Shock into Stationary Medium (Suddenly Open Valve) General Velocities Issues When a valve or membrane is suddenly opened. The second method is “successive substitutions. if not satisfactory use the new Mx to calculate Mx = 1 + Mx then return to part (b).96 CHAPTER 5. 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.” which has better convergence to the solution initially in most ranges but less effective for higher accuracies. 5. With the exception of close proximity to the valve.8: A shock moves into a still medium as a result of a sudden and complete opening of a valve æ çå à ß Þ Í8ÊÝ ä á ã â 2©#Há íé è ì ë è ê é TH ›2ćHè Ö Ü 5Û Ì Ë É Í8ÊÈ Ô Ò Ñ ÐÏ ÕÓ¥XHÎ Upstream ÙØ ÚH× .8(a)).8(b))). NORMAL SHOCK (b) Using shock table or use Potto–GDC to calculate temperature ratio and My . a shock is created and propagates downstream.

25 2 1.3 Mx My Ty/Tx Ty/Tx 0. The same question that was prominent in the previous case appears now. 5.3 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.9: The number of iterations to achieve convergence.3 (b) My = 1.5. My’ = 1.75 5 10 Number of Iteration 15 20 Wed Aug 23 17:20:59 2006 Wed Aug 23 17:46:15 2006 (a) My = 0.5 0 Number of Iteration 10 0 Shock in A Suddenly Open Valve k = 1 4.3 Fig.60) Us = Msx cx (5.5 4 3. The “upstream” Mach number is Mx = The “downstream” Mach number is My = Us − U y = Msy − My cy (5.5 3 2. THE MOVING SHOCKS 97 Similar definitions of the right side and the left side of the shock Mach numbers can be utilized.3. what will be the shock velocity for a given upstream Mach number? Again.62) Shock in A Suddenly Open Valve 1. It has to be noted that the “upstream” and “downstream” are the reverse from the previous case.75 k = 1 4. My’ = 0. the relationship .61) A similar rearrangement to the previous case results in T0y − T0x = Ty 1 + k−1 −2Msy My + My 2 2 2 (5. The stagnation temperature (of moving coordinates) is T0y − T0x = Ty 1 + k−1 2 (Msy − My ) 2 − Tx 1 + k−1 2 (Mx ) 2 =0 (5.5 1 1 0.5 1.

(b) Calculate the Mach number My by utilizing the tables or Potto–GDC.9)). stop or return to stage (b).63) Since Msx can be represented by Msy theoretically equation (5. the larger number of the iterations required to achieve the same accuracy. sufficient results can be achieved after 3-4 iterations.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. The other two solutions are Msx = (k + 1)Uy ± Uy (1 + k) 4 cx 2 + 16cx 2 (5. The unknown data are the piston velocity. the temperature.0. The results show that the convergence occurs very rapidly (see Figure (5. Piston Velocity When a piston is moving. To illustrate the convergence of the procedure. and.64) is a quadratic equation for Msx .3.65) . (c) Utilizing Mx = calculate the new “improved” Mx . the calculations for Us can be obtained from equation (5. NORMAL SHOCK (Msx ) + 2k k−1 2 (Msx ) − 1 2 k−1 2 (5. consider the case of My = 0. It is common practice to solve this equation by numerical methods.64) Ty My + M y Tx Equation (5. no Mach number is given but pieces of information on both sides of the shock.” This method is applied by the following algorithm: (a) Assume that Mx = 1.63) can be solved. for most practical purposes.98 between the two sides is Msy = My + CHAPTER 5. In this case. it creates a shock that moves at a speed greater than that of the piston itself. If it is satisfactory. Yet. other conditions ahead of the shock.3 and My = 1. Therefore. The larger the value of My . One such methods is “successive substitutions. (d) Check the new and improved Mx against the old one. There are three solutions of which the first one is Msx = 0 and this is immediately disregarded.

67) Where the “stange” Mach number is Myx = Uy /cx . and the downstream Mach y y(max) Maximum My’ . The limit of the equation when cx → ∞ leads to Msx = (k + 1)Myx 4 (5.25 has no limit. To explain this phenomenon.75 suddenly and completely clos0. Some use equation (5. k ing of valve results in no limit (at least from the model point Thu Aug 24 17:46:07 2006 of view). This phenomenon occurs in many industrial applications. k.25 discussed earlier in a nozzle 2 flow and in other pipe flow mod1. THE MOVING SHOCKS 99 The negative sign provides a negative value which is disregarded.5. 5.68) As one additional “strange” it can be seen that the shock is close to the piston when the gas ahead of the piston is very hot.66) or in a dimensionless form Msx = (k + 1)Myx + Myx (1 + k) 4 2 + 16 (5. shock.68) to explain the next Shock-Choke phenomenon. such as the internal combustion engines and die casting.5 The spesific heat ratio. This shock–choking phenomenon Shock in A Suddenly Open Valve Maximum M ’ possible is somewhat similar to the 2. Mx = Msx → ∞.5 M choking phenomenon that was 2. what is the maximum velocity that can be reached before this model fails? In other words. The differ1. the shock Mach number cannot be increased. Shock–Choke Phenomenon Assuming that the gas velocity is supersonic (in stationary coordinates) before the shock moves. 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.10: The maximum of “downstream” Mach number as a function of the specific heat. Consider when the “upstream” Mach approaches infinity.75 els (later chapters). look at the normal Fig. and the only solution left is Msx = (k + 1)Uy + Uy (1 + k) 4 cx 2 + 16cx 2 (5.3. It must be noted 1 that in the previous case of 0.5 ence is that the actual velocity 1.

69) Mx = Msx Thus.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. From the upstream side.100 CHAPTER 5. the results is My = 2 k(k − 1) (5. the Mach number is approaching infinity because of the temperature ratio but the velocity is finite.38). the Mach number is ∞    k−1 Ty =   Tx 2k   (5.73) What happens when a gas with a Mach number larger than the maximum Mach number possible is flowing in the tube? Obviously. the semi steady state described by the moving shock cannot be sustained. The Mach number is reduced to the maximum value very rapidly. One can view this as the source of the shock–choking phenomenon. A similar phenomenon to the choking in the nozzle and later in an internal pipe flow is obtained. To understand this limit.23) to this situation the following is obtained: Msx = k+1 2k Px −1 +1 Py (5. is approaching to (k − 1)/2k. These limits determine the maximum velocity after the shock since Umax = cy My . consider that the maximum Mach number is obtained Py when the pressure ratio is approaching infinity Px → ∞.71) 1− Substituting equations (5.72) + When the pressure ratio is approaching infinity (extremely strong pressure ratio). By applying equation (5. NORMAL SHOCK number.26) and (5. 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.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 (5.25) into equation (5. .

86 216507. .8898 1.5. The general case is when one gas flows into another gas with a given velocity.56 668.02 773.3 Partially Open Valve The previous case is a special case of the moving shock. THE MOVING SHOCKS k 1.54 750.75) Since the right hand side of equation (5.43301 0.93 243332.40 1.1785 1.54006 0.75) is constant.20 2.2964 1.05 226871.70 1.62 695.93048 0.3.40 2.2645 1. is cy ρy m ˙ = U y ρ y = M y cy ρ y = M y A √ My kPy = RTy kRTy Py RTy (5.25) can be transferred for large pressure ratios into m ˙ ∼ A Ty Px k − 1 Tx k + 1 (5.00 2. 5.90 2.81 My 0.30 1.81786 0.54772 My 2.80 1.23 873.37797 0. with the exception of Ty the mass flow rate is approaching infinity when the pressure ratio is approaching infinity.51177 0.74) Equation (5.50000 0.40825 0. the shock–choke phenomenon means that the Mach number is only limited in stationary coordinates but the actual flow rate isn’t.50 1.73029 Ty Tx 101 169842.52223 0. ∞.44 273861.09 265805.13). The mass flow rate when the pressure ratio is approaching infinity.50 Mx 1073.64 255883.99 235702.00 729.60 1.87039 0.96 204124.85 922.53161 0.78 261117.29 188982.74 681.4434 1.79 250000.25 985.36 270031.48667 0.74) and equation (5.85 Table of maximum values of the shock-choking phenomenon. The only limitation is that the “downstream’ gas velocity is higher than the “upstream” gas velocity as shown in Figure (5.0815 1.6330 1.47141 0.00000 0.61 801.33968 0.10 2.09 833.45374 0.30 2.3.56 711. Thus.77151 0.

A common problem is to find the moving shock velocity when the velocity “downstream” or the pressure is suddenly increased. c. The procedure for the calculations can be done by the following algorithm: (a) Assume that Mx = Mx + 1. 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.v.76) An additional parameter has be supplied to solve the problem.102 CHAPTER 5. (a) Stationary coordinates (b) Moving coordinates Fig. (c) Calculate the “downstream” shock Mach number Msy = My + My (d) Utilizing Mx = calculate the new “improved” Mx (e) Check the new and improved Mx against the old one.v. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. Ty (Msy ) − Mx Tx þ çý ü û ú Í8Êù ¨  ÿ ¦ ¤ ÿ ¢   ©¡ §¥Ä£¡ÿ Ux = Us − Ux Upstream ó ñ ð ï Í8Êî ò ø 5÷ öõ ÚHô . stop or return to stage (b).11: A shock moves into a moving medium as a result of a sudden and complete open valve. 5.77) (5. If it is satisfactory. NORMAL SHOCK Ux Uy > U x c. It has to be mentioned that the temperature “downstream” is unknown (the flow of the gas with the higher velocity).

7 The mathematics is cumbersome but re0. Figure (5.5.12) 0.13: 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.3.12: The results of the partial opening that the maximum can also occurs in the of the valve. vicinity of the previous value (see following question/example). My .4 limited. THE MOVING SHOCKS 103 Shock in A Suddenly Open Valve Earlier.3 exhibits some “downstream” Mach num0.4 0. Similar equations can be written: (5.79) . c. Thu Oct 19 10:34:19 2006 and for various static “upstream” Mach numbers.v. x x x My y 5. The figure demonstrates Fig.6 sults show that the shock choking phe0. (a) Stationary coordinates (b) Moving coordinates Ux = U s + U x Uy = U s + U y (5.8 M ’ = 0.2 a faster flow is running into a slower fluid.9 nomenon also occurs in the case where M ’ = 0.v.2 2 2. it was shown that the shock chokk=14 ing phenomenon occurs when the flow is 1 running into a still medium.4 Partially Closed Valve Ux Uy Ux = Us + Ux Upstream Uy = Us + Uy ρ y Py Ty Fig.3. not the actual flow). This phe0.0 0.8 1.4 2. Mx . The information propagates upstream in the same way as before. 5.6 M’ bers for various static Mach numbers. M ’ = 0.78) !  "      c.9 0.8 1. 5.5 nomenon is still there (the Mach number is 0.

132 Ty Tx Py Px P0 y P0 x 0.3. it is the case of a shock moving into still medium (suddenly opened valve case). the Mach velocity ahead of the shock has to calculated.1206 My Mx My 1.604 4.54220 0.3 × 287 × 300 By utilizing Potto–GDC or Table (5.3.80) My = M s + M y (5. First.5 Worked–out Examples for Shock Dynamics Example 5.2: 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.104 CHAPTER 5. 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. NORMAL SHOCK Mx = M s + M x (5. stop or return to stage (b). Assume the specific heat ratio is 1.4) one can obtain the following table: Mx 2.81) For given static Mach numbers the procedure for the calculation is as follows: (a) Assume that Mx = Mx + 1.0 1.953 0.296 1. If it is satisfactory. (b) . Ty (Msy ) + Mx Tx 5. Compute the pressure and the temperature behind the shock. Thus.63955 . 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.

724 4.56995 1.0 0.56994 0.2) you can find the proper Mx .0 1. find the upstream prime Mach.2961.2961 2.724 The table was obtained by utilizing Potto–GDC with the iteration request.53487 0.5. Calculate the velocity and the pressure behind the reflecting shock.953 × 1.604 × 300 ∼ 481.56995 Ty Tx My 0.0 0.724 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 2.724 1. Simply using the Potto–GDC provides for the temperature and velocity the following table: Mx 2.9432 1.710 0. the prime properties can be found.0 Ty Tx Py Px P0 y P0 x 0. THE MOVING SHOCKS Using the above table.0 0.56995 0.0 ∼ 4.3: Gas flows in a tube with a velocity of 450[m/sec].0445 My Mx My 0.57040 0.044 2. The gas is brought into a complete stop by a sudden closing a valve.953[Bar] Px The velocity behind the shock wave is obtained √ m Uy = Mx cx = 1.2961. Mx to be 1.3. The specific heat ratio can be assumed to be k = 1. Then. At this stage the reflecting shock velocity is unknown.4.045 2. the temperature behind the shock is Ty = T y = Ty Tx = 1. The static pressure at the tube is 2Bar and the (static) temperature of 300K.3 × 287 × 300 ∼ 378.70009 If you insist on doing the steps yourself.72 sec Example 5.722 1.044 My 0. Then using Table (5. S OLUTION The first thing that needs to be done is to find the prime Mach number Mx = 1.132 × 1. .0 0.2K Tx 105 In same manner.042 2. it can be done for the pressure ratio as following Py = P y = Py Px = 4.2961 1.

3020 0.3574 My 0. for those who like an additional step) in order to double the temperature when the valve is suddenly and totally closed? S OLUTION The ratio can be obtained from Table (5.4) for the calculation procedure.3.89509 0.3574 My Mx My Ty Tx Py Px P0 y P0 x 0.55830 Example 5.55832 using the required Mx = 2.1583 6. S OLUTION Refer to section (5.52778 0. Calculate the speed of the shock when a valve is closed in such a way that the Mach number is reduced by half.3).78928 0.5: A gas is flowing in a pipe with a Mach number of 0.52778 Ty Tx ρy ρx Py Px P0 y P0 x 2.20000 1.4: What should be the prime Mach number (or the combination of the velocity with the temperature.106 CHAPTER 5.3166 0.0789 1.1220 My Mx My Ty Tx Py Px P0 y P0 x 0. NORMAL SHOCK Example 5. Potto-GDC provides the solution of the above data Mx 1. It can also be obtained from the stationary normal shock wave table.0 2.0000 3.3574 in the moving shock table provides Mx 2.000 6. 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). Potto-GDC provides for this temperature ratio the following table: Mx 2.99813 If the information about the iterations is needed please refer to the following table.40000 0. .317 0.4. Hint.

3025 1.5.0836 1.84424 0.3199 1. Calculate the velocity of the shock created by the piston in the air.0793 1.1200 1.20000 0.20000 0.3020 1.99548 0.20000 0.3017 1.1220 My 0.3020 Example 5.89512 0.89509 0.3020 1.0106 1. 5.89536 0.14: Schematic of a piston pushing air in a it takes for the shock to reach the tube.89009 0.1226 1.0789 1.20000 0.0789 1.0045 1.2032 1.20000 0.8 accelerated very rapidly and the air adjoined the piston obtains Mach number M = 0.20000 0.1220 1.3020 1.2989 1.90416 0. 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 1.0787 1.0930 1.3609 1.89508 0.20000 0.20000 0.0789 1.1220 1. Assume that there is no friction and the Fanno flow model is not applicable.4 My = 0.1220 1.3020 1.8.1222 1.1967 1. the solution is .0789 1.89789 0.1220 1.2922 1.0832 1.3019 1. Calculate the time Fig.92479 0.0782 1.89354 0.89509 0.20000 0.20000 0.89517 0.0789 1.20000 1.3020 1.3011 1.0788 1.1220 1.1220 1.89595 0.1288 1.20000 0.4 and 300◦C.0545 1.89509 0. end of the tube of 1.1443 1.0789 1.0030 1.20000 0.0790 1.89461 0.20000 0.20000 0.20000 0.5041 1.20000 0.20000 0.1259 1.2705 1.3020 1.20000 0.6: A piston is pushing air that flows in a tube with a Mach number of M = 0.4000 1.0712 1. S OLUTION Using the procedure described in this section.1216 1.1099 1.2547 1.1220 1.3075 1.73971 0.20000 0.20000 0.89494 0.3022 1.1219 1.0m length.0765 1.89510 0.1221 1.20000 0.3037 1.1208 1.0789 2.0789 1.1241 1.20000 0.89504 0.1182 1.87903 0. The piston is Mx = 0.89509 0.0789 1.3.0802 1.89509 Ty Tx Py Px 107 My 0.

5000 1.80000 0.1517 1.2380 1.7: From the previous example (5.80000 0.4 × 287 × 300 0.1519 2.3202 1.80000 0.70109 0.82716 0.2378 1.81942 0.1519 1.80000 0.2380 My 0.4 × 287 × 300(1.5    q Ty Tx =  ∼ 124.2248 1.1519 −0.108 Mx 1.4) Example 5.80000 1.4583 1.10) calculate the velocity difference between initial piston velocity and final piston velocity.80000 1.81958 0.50000 0. S OLUTION The stationary difference between the two sides of the shock is: ∆U =Uy − Ux = cy Uy − cx Ux   √   1.2380 My Mx My CHAPTER 5.80000 0.81829 0.81943 0.6273 1.6207 1.4[m/sec] .2381 1.8 × 1.5834 1.6215 0.6217 1.80000 0.2380 − 0.2400 1. NORMAL SHOCK Ty Tx Py Px P0 y P0 x 0.0034[sec] 1.6216 The time it takes for the shock to reach the end of the cylinder is t= length Us cx (Mx −Mx ) =√ 1 = 0.81942 Ty Tx Py Px My 0.1519 1.1531 1.1435 1.1519 1. i 0 1 2 3 4 5 6 Mx 1.81940 0.98860 The complete iteration is provided below.6215 1.

1. When the pressure is high enough.1283)347. At the back of the shock.0 My 0. a shock is created that it travels to the low pressure chamber. A cylinder with two chambers connected by a diaphragm.0715 + 1. The distance between the two pistons is 1[m].0 while the velocity of the right piston is Mx 1.99959 40. 5. On one side the pressure is high. It is used in many cases to understand certain phenomena.0715 My 0.99785 70.89048 Mx 0. Calculate the time it will take for the two shocks to collide.1283 My 0. The solution is given by equation (5. 5.5.93471 Mx 0.083 1.318 0. For this case the following table can easily be obtain from Potto–GDC for the left piston Mx 1.8: An engine is designed so that two pistons are moving toward each other (see Figure (5. and.4 Shock Tube The shock tube is a study tool with very little practical purposes. This is the same case as in the suddenly opened valve case described previously.15)).4. Other situations can be examined and extended from these phenomena.95890 Ty Tx Py Px P0y P0 x Fig.66). expansion waves occur with a reduction of pressure. while the pressure on the other side is low.93451 Ty Tx Py Px P0y P0 x Uy cx 347.0 The time for the shocks to collide is t= 1[m] length = ∼ 0. it is the explicit analytical solution. The high pressure .173 0. The temperature is known to reach several thousands degrees in a very brief period of time.0 My 0.8) Uy cx 347.15: Figure for Example (5.047 1. SHOCK TUBE Example 5. 1.0013[sec] Usx 1 + Usx 2 (1. The air between the pistons is at 1[Bar] and 300K. 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. When the diaphragm is ruptured the gas from the high pressure section flows into the low pressure section.

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

111 √ range of the speed of sound.88) After rearranging equation (5.88) the result is U2 = c1 k P2 −1 P1 2k k+1 P2 k−1 P1 1+k (5.85) The fluid velocity in zone 2 is the same U2 = U s − U 2 = U s 1 − From the mass conservation.89) . The relationship between the different properties was discussed earlier and because it is a common problem. A typical design problem of a shock tube is finding the pressure to achieve the desired temperature or Mach number. The following equations were developed earlier and are repeated here for clarification.82) k − 1 k + 1 P2 + 2k 2k P1 (5.83) And the velocity ratio between the two sides of the shock is k+1 1 + k−1 P2 ρ2 U1 = = k+1 P2P1 U2 ρ2 k−1 P 1 (5. these kinds of experiments require fast recording devices (a relatively fast camera and fast data acquisition devices. The pressure ratio between the two sides of the shock is 5. c ∼ 340 thus the whole process takes only a few milliseconds or less.87) U2 = c 1 k − 1 k + 1 P2 + 2k 2k P1 1− + 1+ (5.82) becomes Ms1 = Or expressing the velocity as Us = Ms1 c1 = c1 k − 1 k + 1 P2 + 2k 2k P1 (5. a review of the material is provided thus far. it follows that U2 ρ1 = Us ρ2 k+1 k−1 P2 P1 k+1 P2 k−1 P1 U2 Us (5. SHOCK TUBE k−1 P2 = P1 k+1 Rearranging equation (5. Thus.86) (5.84) 2k Ms1 2 − 1 k−1 (5.).4.

96) After some rearrangement.112 CHAPTER 5.96) is transformed into P5 P2  1 − = √ P1 P1   (k − 2k c1 1) c5 P5 P3 −1 P2 P1 2k − k−1 2k + (k + 1) −1    (5.91) dρ (5. taking the derivative of the continuity equation.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 (5.89) 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. d(ρU ) = 0. hence equation (5. NORMAL SHOCK On the isentropic side. and dividing by the continuity equation the following is obtained: dU dρ =− ρ c (5.94) (5. applying the isentropic relationship (T ∝ ρk−1 ) yields c = c5 From equation (5.97) .92) can be integrated as follows: U3 ρ3 dU = U5 =0 ρ5 c5 ρ ρ5 k−1 2 dρ (5.92) Equation (5.90) Since the process in zone 4 is isentropic.90) it follows that dρ dU = −c = c5 ρ ρ ρ5 k−1 2 T = T5 ρ ρ5 k−1 2 (5.95) As it was mentioned earlier the velocity at points 2 and 3 are identical. equation (5. in zone 4.95) and equation (5.

6 5. the assumption of an uniform flow in the duct is closer to reality. In that case. Find the shock velocity and temperature behind the shock if the pressure P5 ratio is P1 = 40? S OLUTION 5.5. SHOCK WITH REAL GASES Or in terms of the Mach number.98) Using the Rankine–Hugoniot relationship and the perfect gas model.100) Example 5.7 Shock with Real Gases Shock in Wet Steam Normal Shock in Ducts The flow in ducts is related to boundary layer issues. 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 2k − k−1 (5. For a high Reynolds number. It is normal to have a large Mach number with a large Re number. the following is obtained: 1+ T2 = T1 1+ k1 −1 k1 +1 k1 −1 k1 +1 P2 P1 P1 P2 (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 (5.9: A shock tube with an initial pressure ratio of P5 = 20 and an initial temperature of P1 300K.5. . the assumptions in construction of these models are acceptable and reasonable.5 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.967 My kRTy . In this system.0021[sec] 1.0017[sec] 5. Assume that there isn’t any resistance whatsoever in the pipe. An engineer is required to design a cooling system for a critical electronic deexit valve vice. 5.057811 The direct calculation will be by using the “upstream” Mach number.668sqrt1.0 5.00 0.4 × 287 × 300 × 5. Therefore.17)).668 Ty Tx Py Px P0 y P0 x 0. opening valve probelm The distance between between the valve and the pipe exit is 3[m].0850sqrt1.0850 My Mx My 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. the time is t= distance 3 √ = ∼ 0.8 More Examples for Moving Shocks Example 5. 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. S OLUTION This problem is known as the suddenly open valve problem in which the shock choking phenomenon occurs.17: Figure for Example (5. Comment on this proposal.0850.114 CHAPTER 5. The time it takes for the shock to travel from the Py valve depends on the pressure ratio Px = 30 Mx 5. Mx = Msx = 5.41404 0. The time it takes for the material from the valve to reach the exit is distance 3 t= = ∼ 0. The ambient temperature is 27◦ C and 1[Bar].10) supposed to reach the pipe exit as quickly as possible when the valve is opened (see Figure (5.967 30.10: This problem was taken from the real industrial manufacturdistance ing world.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. air is Fig. The conditions upstream of the valve are 30[Bar] and 27◦ C .0002 [sec] to be opened). After building the system. NORMAL SHOCK 5. The temperature should not increase above a certain value.

18). Thus. 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. TABLES OF NORMAL SHOCKS.99893 0.00000 1.25504 1.06494 1. 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.25 1. The last possible requirement a programmable valve and very fast which its valve probably exceed the moving shock the valve downstream.51570 1.98706 0.5.15938 1.78596 Ty Tx ρy ρx Py Px P0y P0x 1.81264 0.4 Mx 1.20 1.84217 0.4 Ideal Gas Table 5.03284 1. The plot of the mass flow rate and the velocity are given in Figure (5.42857 1. Example 5.95313 0. Plot the pipe exit temperature as a function of the time.00 1.65625 1.30 My 1.80500 1.05 1. k = 1.00000 1.34161 1.9.99280 0.00000 0.18: The results for Example (5. the heating time is reduced significantly.51333 1.11958 1.00000 1. Thus.9 Tables of Normal Shocks. this suggestion contradicts the purpose of the required manufacturing needs.10 1.10) deals with a damaging of electronic product by the temperature increase.11: Example (5.12799 1.97937 .08398 1.4 IDEAL GAS 115 During that difference of time the material is get heated instead of cooling down because of the high temperature.10) will not change the effects of heating.99669 0.37625 1. Time[Msec] To increase the pipe diameter will not change the temperature and therefore Fig.15 1.1: The shock wave table for k = 1.91177 0. Try to estimate the temperature increase of the product. S OLUTION 5.24500 1.16908 1.99985 0. K = 1.87502 0.19087 1. There is also the possibility of steps increase in which every step heat released will not be enough to over heat the device. 5.00000 0.09658 1.

56906 0.27335 3.12500 16.51299 0.00 5.68966 1.92979 0.94059 8.49458 1.90 1.85714 4.03175 2.44891 4.70 1.40625 0.51792 0.57735 0.42623 4.43811 2.29878 6.45 1.32022 1.50 5.40897 0.51931 0.53441 0.75 1.40623 0.12000 2.68410 0.50 3.26087 4.90138 1.02498 .40625 3.25469 1.00958 3.61650 0.15625 29.81268 0.14894 5.40216 Ty Tx ρy ρx Py Px P0y P0x 1.11525 2.36906 7.89520 0.47519 0.59188 2.96974 0.31915 1.83625 7.09170 0.80 1.4 (continue) Mx 1.42355 0.53158 1.00 6.22261 1.75 5.83457 0.73971 0.85572 0.40 1.82625 4.11256 0.50 2.26958 4.72087 0.54706 0.98511 3.51568 2.35 1.97833 5.41189 0.99311 2.25 2.35 2.95122 3.46192 0.68750 1.12500 8.85 1.42878 0.54055 0.64054 0.25 4.33333 12.48000 5.39657 2.40625 41.74002 2.38797 1.85686 1.50000 4.75 3.08846 2.95958 2.04500 4.41523 0.94484 0.60 1.54014 0.28720 1.28625 2.41908 0.25 5.25 My 0.32834 0.55 1.40416 0.04033 2.61194 3.95819 0.77614 1.43496 0.116 CHAPTER 5.26451 0.44231 0.60792 1.60278 1.82180 7.45833 1.75 6.31505 3.71956 0.65625 10.03536 0.58618 0.64729 1.81188 4.76736 0.21182 5.81322 1.27625 6.87599 0.60570 0.98958 35.79023 0.69919 4.60553 0.57143 4.06172 0.88231 2.55395 0.27907 2.00 2.61333 3.07869 5.94732 2.17166 0.76175 0.91319 0.32544 5.74420 0.08455 3.73958 6.12500 38.67901 2.15 2. NORMAL SHOCK Table 5.33333 3.45115 0.66844 0.72855 1.10 2.35922 2.53637 1.07505 0.21295 0.05100 0.66667 2.94680 1.56148 0.69751 0.75 4.56128 0.65396 0.01863 3.49181 0.00000 5.04236 0.83333 45.23958 18.05 2.67420 0.91156 5.25 3.95 2.40 2.1: The shock wave table for k = 1.13876 0.63625 2.58329 0.49901 0.86207 1.40642 0.07229 4.22625 5.65105 0.65 1.59562 0.00 4.62814 0.20500 3.21190 3.80000 6.62809 0.70109 0.66894 4.00500 6.04688 4.90625 23.45 2.19772 2.50 4.00000 31.00 3.45833 2.55333 6.15625 14.82000 3.87509 5.13750 2.81190 2.20 2.35379 1.50 1.42280 1.52861 0.56935 1.77045 1.14897 3.26829 5.52312 0.73625 4.50000 20.02965 0.30 2.45833 26.

TABLES OF NORMAL SHOCKS.66874 5.040 1.423 1.16229 11.25 8.00739 0.99403 0.50 7.39121 0.87948 12.0 0.0 0.40038 0.00645 0.0 0.47883 5.014 1.02115 0.08 0.118 1.04 0.99986 0.56522 5.004 1.4) Mx 1.016 1.94836 0.219 1.00387 0.25 9.055 1.99999 0.38758 Ty Tx ρy ρx Py Px P0y P0x 9.95231 0.00000 61.12500 52. K = 1.51020 5.33333 99.506 1.710 1.01798 0.057 1.42736 20.4 IDEAL GAS Table 5.00496 0.126 1.39385 0.76452 0.133 1.00000 0.0 0.82755 .57997 18.008 1.0 0.00979 0.69273 17.99998 0.38860 0.0 0.475 2.67171 0.75 10.96506 0.65625 105.07 0.15625 65.0 0.75 8.05 0.780 3.80 0.38750 5.61165 5.5.53890 5.0 0.0 0.171 1.020 1.0 0.01133 0.39048 0.98446 0.39736 0.1: The shock wave table for k = 1.01316 0.30 0.344 1.676 1.006 1.63218 5.012 1.94291 0.73029 0.10 0.00 My 0.65116 5.0 0.50 6.12500 110.80251 0.00 8.99980 0.102 1.89128 0.043 1.40 0.087 1.037 1.0 0.028 1.38980 0.028 1.97647 0.03 0.00 9.4 (continue) 117 Mx 6.95944 0.39201 0.50 9.39491 0.589 1.00 My 0.80077 10.38672 14.46939 11.99113 15.0 0.323 1.90 1.043 1.049 1.018 1.49152 19.9.70019 5.196 1.64673 0.95387 0.268 1.149 1.44444 5.98812 0.024 1.030 1.12500 89.06 0.75 9.39289 0.062 1.50 0.98227 0.89918 0.82978 16.45833 69.00849 0.127 1.71429 49.316 1.0 0.082 1.01 0.036 1.17678 14.473 1.502 1.20 0.0 0.39879 0.39607 0.75 7.442 1.15625 94.90625 74.00 7.00343 0.99994 0.98958 57.60 0.02 0.99991 0.50 8.38807 0.92832 0.0 Ty Tx Py Px P0y P0 x 1.00437 0.0 0.072 1.766 My 0.99997 0.0 0.00000 1.97074 0.69946 0.38918 0.58939 5.99317 0.112 3.505 1.86537 0.195 2.09 0.68504 5.97099 0.50000 79.024 1.00304 Table 5.84463 0.99973 0.73958 116.62425 Mx 0.269 1.25 7.0 0.23958 84.50000 0.01535 0.032 1.00565 0.99790 0.40667 5.70 0.15643 9.381 1.62095 13.941 2.2: Table for a Reflective Shock from a suddenly closed end (k=1.36508 5.012 1.

3: Table for shock propagating from suddenly opened valve (k=1.25 24.02 0.99996 0.89 12.576 1.012 1.21 29.762 3.09 0.15637 0.063 1.050 1.649 1.016 1.95888 0.014 1.703 10.0 0.41 15.99763 0.0 0.97629 0.07 0.000 6.517 9.0 0.990 2.0 0.00 5.93133 0.70283 0.0 0.03 0.061716 0.0 0.00 7.810 1.858 1.390 1.026517 0.54114 0.0 0.028 1.99990 0.325 11.0 Ty Tx Py Px P0 y P0 x 1.008 1.317 1.0 0.00 6.94746 0.70 Ty Tx Py Px P0 y P0 x 1.133 1.897 1.734 17.20 1.05 19.00125 Table 5.0 0.47 1.00000 1.40 1.0 0.06 0.50 1.96069 0.14 62.0 0.031 1.0 0.033 1.256 6.99971 0.057 1.715 6.043 1.52904 0.791 2.01 29.99181 0.0 0.51996 0.162 7.0 0.441 2.60 1.10 0.037 1.39116 0.104 1.431 8.146 2.43921 0.012492 0.00 4.346 2.56935 0.0 0.245 2.56312 0.728 5.297 2.0 0.50814 0.4E+2 1.99979 0.0 0.029 1.727 1.97045 0.30 0.0 0.40213 0.78652 0.2: Table for Reflective Shock from suddenly closed valve (end) (k=1.552 2.74316 0.656 2.058 1.33 3.448 2.7E+2 0.4) Mx 1.280 4.088 1.118 CHAPTER 5.88717 0.98216 0.83607 0.49912 0.120 1.012 1.30 1.00 8.0 0.0 My 0.004 1.99985 0.47855 0.55453 0.206 5.0 0.495 1.49092 0.336 8.90 2.00 10.020 1.952 2.087 2.331 1.99402 0.70 1.08 0.69834 0.0 0.74403 0.0 0.38457 Mx 1.65290 0.94180 0.018 1.10 1.95315 0.210 1.96465 0.00 44.60 0.859 5.38817 0.00 3.0 0.00639 0.152 1.39566 0.831 5.397 1.086 1.0 0.0 0.024 1.51808 0.04 0.541 1.0 0.087 2.78840 0.00000 0.134 1.4)(continue) Mx 1.0 0.0 0.38608 0.136 1.048 2.58578 0.01 0.295 1.89039 .006 1.50 0.0 0.43894 0.073 1.98019 0.024 1.99994 0.827 7.20 0.188 1.1E+2 1.60761 0.66462 Mx 0.98807 0.40284 0. NORMAL SHOCK Table 5.62 84.189 2.037 1.0 0.0 My 0.99998 0.41523 0.00204 0.041 1.066 8.868 1.60401 0.248 1.800 8.00 9.044 1.40 0.613 My 0.862 4.00350 0.05 0.99999 0.410 3.0 0.80 1.0 0.08 My 0.

62 17.09 35.067 1.5.0 0.799 3.889 1.8E + 2 7.300 1.604 1.293 8.0 0.0 Table 5.3: Table for shock propagating from suddenly opened valve (k=1.18 73.017 1.00721 0.0 0.0 0.0 0.98857 0.889 1.810 1.42622 0.39187 0.32 16.54 15.90 1.026 1.0 My 0.821 9.0 0.508 2.3E+2 7.7E + 2 4.9.023 1.99999 0.010 0.820 1.37820 0.1E+3 1.10216 0.840 7.0 0.7E+2 9.95 64.84 1.3) Mx 1.0 0.0 0.00272 0.0 0.62 61.931 14.4 IDEAL GAS Table 5.79 My 0.44536 0.35 25.53817 0.000181 3.0 0.450 2.171 8.0 0.68907 0.0 0.37944 0.551 8.76940 0.009 1.49586 0.0 0.59699 0.4E + 3 4.0 0.888 1.0 My 0.025 4.0 0.78 80.536 4.98290 0.31 62. K = 1.0 0.0 0.2E+3 1.38402 0.2E+2 8.400 1. TABLES OF NORMAL SHOCKS.46599 0.269 2.840 1.4) 119 Mx 1.9E + 2 2.888 1.881 3.0 0.100 1.37814 0.38248 0.2E + 3 0.9E + 3 5.888 1.7E+2 8.050 Ty Tx Py Px P0y P0 x 1.015 1.0058 1.029 My 0.37813 0.99 25.500 1.744 1.37816 0.00101 0.25 12.83661 0.006 1.4E + 3 6.99998 0.48823 0.9E + 3 7.28412 0.59649 0.0 0.45 22.880 1.800 1.030 0.92 42.9E+2 1.888 1.80 0.99997 .0 0.37810 Mx 0.02 83.5E + 2 3.1E+3 1.0 0.975 11.823 5.040812 0.37822 0.000497 0.9E + 3 6.57 31.0 0.889 1.3E+2 7.56619 0.18E−5 0.83 71.850 1.200 1.658 4.600 1.5E + 2 1.0 0.00 1.2E + 3 5.40843 0.00000 1.012 1.2E+2 9.5E + 3 5.97726 0.00544 0.61 1.4: Table for shock propagating from a suddenly opened valve (k=1.37817 0.145 2.0 0.0 0.012 1.888 1.38096 0.0 0.24 11.0 0.97166 Mx 0.4E+3 3.491 1.53 63.919 2.387 4.38870 0.0 0.889 1.07 90.0 0.39028 0.07 18.2E + 2 1.0 0.6E + 3 4.889 Ty Tx Py Px P0 y P0 x 1.646 6.72 80.0 0.0 0.040 0.054 1.896 2.99427 0.0 0.870 1.167 3.38713 0.368 5.74 66.00000 0.860 1.5E + 3 8.745 1.068 2.38974 0.00175 0.62 14.0 0.38557 0.69 68.37818 0.37812 0.0 0.700 1.003 1.0 0.040 1.020 0.0 0.18575 0.00395 0.0 0.0E + 2 1.0 0.830 1.013 1.51223 0.37821 0.457 10.80 76.62923 0.0 0.33 21.

600 0.820 1.090 0.888 1.359 1.40097 0.3) Mx 1.166 4.99975 0.39026 Mx 0.099 4.080 0.093988 0.202 3.0 0.100 0.39034 0.0 0.94959 0.109 4.50100 0.0 0.070 0.40257 0.706 1.39030 0.997 3.898 3.522 6.980 2.523 7.052984 0.100 1.99987 0.919 2.482 4.275 1.042 1.40226 0.95 23.99995 0.0 0.889 1.0 0.500 1.96 0.707 5.098 1.42089 0.0 0.22904 0.0 0.830 1.840 1.110 1.67546 0.700 0.060462 0.0 0.096 1.800 1.71284 0.083607 0.39029 0.103 1.053018 0.31281 0.553 1.39033 0.069233 0.223 1.98290 0.252 2.052914 0.55074 0.0 0.888 1.054 1.810 1.815 1.098 4.99991 0.613 4.034 4.614 4.0 0.39160 0.592 3.94156 0.0 0.681 1.0 0.060 0.99981 0.448 3.90 22.096 4.0 0.0 My 0.101 4.200 0.126 1.048 1.663 1.46 14.612 4.0 0.39035 0.52495 0.016 4.021 1.39468 0.426 2.0 0.081 1.05 20.381 3.615 4.30 23.0 0.700 1.0 0.90734 0.0 0.031 1.102 4.0 0.39031 0.96610 0.078654 0.286 4.300 0.136 1.120 CHAPTER 5.39314 0.616 4.088718 0. NORMAL SHOCK Table 5.834 3.028 1.053053 0.45807 0.052809 .060 1.467 2.236 2.471 4.097 4.880 1.93 23.0 0.58223 0.86274 0.0 0.860 1.669 3.888 1.0 0.177 1.400 1.073863 0.99792 0.125 1.889 1.0 0.44 18.452 1.785 1.64 21.035 1.60847 0.613 4.99288 0.32 18.615 4.0 0.91 23.93 23.225 4.89159 0.64073 0.800 0.80734 0.75301 0.0 0.489 1.74136 0.40418 0.0 0.0 0.500 0.600 1.0 0.302 1.069 2.49 20.66575 0.519 3.96631 0.053 4.95 23.100 4.063 1.056322 0.064766 0.570 1.95 19.25 21.0 0.84227 0.0 0.553 4.95506 0.644 2.39027 0.57853 0.400 0.850 1.300 1.0 0.39938 0.274 1.0 0.94415 0.0 0.888 1.052949 0.94 23.749 3.58 23.842 3.15495 0.889 Ty Tx Py Px P0 y P0 x 1.92 23.923 4.49333 0.4: Table for shock propagating from a suddenly opened valve (k=1.0 0.053088 0.140 1.00 1.200 1.39037 0.0 0.0 0.367 11.024 1.018 1.333 1.900 1.478 1.052844 0.870 1.349 4.611 4.39780 0.96056 0.959 2.39624 0.595 2.79611 0.43882 0.400 1.91 23.772 9.0 0.47875 0.888 1.415 4.197 1.052879 0.576 4.889 1.099 4.616 My 0.191 2.0 0.

97 0.052775 .5.104 23.39025 Mx 0.617 My 0.9.3) 121 Mx 4.0 My 1.4 IDEAL GAS Table 5.889 Ty Tx Py Px P0y P0 x 4.4: Table for shock propagating from a suddenly opened valve (k=1. K = 1. TABLES OF NORMAL SHOCKS.

NORMAL SHOCK .122 CHAPTER 5.

CHAPTER 6 Normal Shock in Variable Duct Areas In the previous two chapters. the gas has to pass through a converging–diverging nozzle to obtain a supersonic flow. P B is between sures w ic flo subson shock a after 123 0& ( ' ©§)%& 7 6 ¡35 4 2 ¡31 $ %# # . but also in steady state cases when there is no supersonic flow (in stationary coordinates). d In Figure (6. This Chapter will present the flow in this special range of pressure ratios.1: The flow in the nozzle with different back prespressure ratio. the flow in a variable area duct and a normal shock (discontinuity) were discussed. A discussion of the occurrences of shock in flow in a variable is presented. It is c interesting to note that a normal a Subsonic shock must occur in these situations (pressure ratios). 6. the shock can occur only in steady state when there is a supersonic flow.1) the reSupersonic duced pressure distribution in b the converging–diverging nozdistance. In the previous chapter. x zle is shown in its whole range of pressure ratios. the flow in a convergent– divergent nuzzle was presented when the pressure ratio was above or below the special range. As it is was presented before. As it was shown in Chapter 5. When the Fig.

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. the over–expanded nozzle is worse than the under–expanded nozzle because the nozzle’s large exit area results in extra drag. this example provides a demonstration of x y the calculations required for the location even if it isn’t realistic. 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. once the flow becomes supersonic.) Also determine the critical points for the back pressure (point “a” and point “b”). ed ™ — – ” ‘ ‰ yp%˜ eV•“ ’ "ˆ ‡† … ƒ ‚  xvut rph„ eB§€ ywr"s qi g d c b a rphf eV¥` YXWU T R PPPVSQ IGF CA @ 9 PH£E DB)8 . this example will provide the fundamentals to Fig. the nozzle doesn’t provide the maximum thrust possible. Obviously if the back pressure. Calculate the back pressure and the temperature of the flow. It troat has to be recognized that the shock wave isn’t easily visible (see Mach’s photography techexit point "e" niques). 6. Example 6. only exact geometry can achieve continuous pressure flow. In the under–expanded case. (It should be noted that the temperature of the surrounding is irrelevant in this case. Only in one point where P B = Pb continuous pressure exist. Nevertheless. Nozzle throat area is 3[cm2 ] and the exit area is 9[cm2 ]. no continuous pressure possibly can exists. In this case. is lower than the critical value (the only value that can achieve continuous pressure) a shock occurs outside of the nozzle. P B .2: A nozzle with normal shock explain the usage of the tools (equations and tables) that were developed so far. The shock occurs in a location where the cross section area is 6[cm2 ]. The location of the shock is determined by geometry to achieve the right back pressure.124 CHAPTER 6. Therefore. In comparison of nozzle performance for rocket and aviation. In conclusion. NORMAL SHOCK IN VARIABLE DUCT AREAS point “a” and point “b” the flow is different from what was discussed before. If the back pressure. 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. The first example is for academic reasons. P B is smaller than Pb a discontinuous point (a shock) will occur. In the literature. This nozzle is called an over–expanded nozzle.1: A large tank with compressed air is attached into a converging–diverging nozzle at pressure 4[Bar] and temperature of 35[◦ C]. 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. On the other hand.

1972 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.86457 1. Then the relationship developed for the shock can be utilized to calculate the Mach number after the shock. Thus.54743 Ty Tx ρy ρx Py Px P0y P0 x 1.9474 5. Mx is about 2.125 First.8882.0268 Again utilizing the isentropic relationship the exit conditions can be evaluated. the stagnation temperature and pressure are known T0 = 308K and P0 = 4[Bar]. It has to be realized that for a large tank. Mx = 2.8544 2. one can obtain using the isentropic relationship . (point “y”). in the case of ratio of 100 the Mach number is 0. the inside conditions are essentially the stagnation conditions (this statement is said without a proof.1972 My 0.0000 0. The star area (the throat area).2588 0.81568 1. the Mach number at the exit can be calculated by utilizing the isentropic relationship.54743 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.4656 0. From equation (5.18787 With this Mach number.50877 0. but can be shown that the correction is negligible for a typical dimension ratio that is over 100.197 as shown table below: M 2.2) or from equation (4.11) the following Table for the isentropic relationship is obtained M 0. Ax 6 = =2 A∗ 3 With this ratio (A/A∗ = 2) utilizing the Table (5.2) My ∼ 0.18463 2. For example.62941 From Table (4. This Mach number can be calculated by utilizing the isentropic relationship from the large tank to the shock (point “x”).8882 ∗ A Ay A 6 with this area ratio.49) or the GDC– Potto.22) or from Table (4. My can be obtained. the exit Mach number has to be determined. the subsonic = branch can be evaluated for the pressure and temperature ratios. From the Mach number after the shock. With known Mach number the new star area ratio. Ay /A∗ is known and the exit area can be calculated as Ae Ay 9 Ae = × ∗ = 1.1). My .54746.94345 0.1972 the Mach number.00587 and the error is less than %0. Mx 2.1) or equation (4.09393 0. With these values. before the shock is known and given as well. A∗ .2588 × = 1. the Mach number. as Ae A∗ = 1.

Yet we have to pay attention that there two possible back pressures that can “achieve” it or target.9K × 1.6374 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.41820 × 4 ∼ =1.98077 0.97912 0. The area ratio for both cases.94862 1.7538 Since the stagnation pressure is constant as well the stagnation temperature.19745 2. is A/A∗ = 3 In the subsonic branch (either using equation or the isentropic Table or GDC-Potto as M 0.97[Bar] For the supersonic sonic branch Pexit = Pexit P0 P0 = 0.951 Ty Tx Tx T0 T0 =0. In that case we don’t have to go through that shock transition.32651 CHAPTER 6.97318 0. Pexit = Pexit P0 P0 Py Py Px Px P0 P0 =0.8882 0.81568 × 5.14190 Pexit = Pexit P0 P0 = 0.41820 0.92882 1.04730 2.99226 0.6728[Bar] . NORMAL SHOCK IN VARIABLE DUCT AREAS T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.094 × 4 T0 Ty 1 0.0000 3.34[Bar] The exit temperature is Texit = Texit T0 1 0.9195 0.11310 3.466 × 0.0000 0.126 M 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.92882 × ∼ =2. the exit conditions can be calculated.854 × 0.98133 × ∼ =299.99226 × 4 ∼ =3.

Thus. In fact in many industrial applications. these kind situations exist.833 × 308 = 256.1) where would be shock’s location when the back pressure is 2[Bar]? 1 The meaning of the word practical is that in reality the engineer does not given the opportunity to determine the location of the shock but rather information such as pressures and temperature.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. In these applications a small pressure difference can produce a shock wave and a chock flow. While the pressure to achieve full supersonic flow through the nozzle the pressure has to be below the 42% the original value.5283 × 4 = 2.12[m/sec] P∗ P0 P0 = 0. For more practical example1 from industrial application point of view. .12 = 0.7 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). over 50% of the range of pressure a shock occores some where in the nozzle.4 × 287 × 256. Example 6.7 = 321.13[kg/sec] 287 × 256.113[Bar] And the mass flow rate reads m= ˙ 4105 3 × 10−4 × 321.7K T∗  R  T0   T0  T∗ A √ kRT ∗ = P∗ P0 P0 R T∗ T0 A kR T0 T∗ T0 T0 The temperature at the throat reads P∗ = The speed of sound is c= √ 1. The flow rate is expressed as following  ∗  P ρ∗ P∗ m=ρ A U = ˙ A cM = RT ∗ ∗ ∗ M =1 P ∗     P0   P0   c The temperature and pressure at the throat are: T∗ = T∗ T0 T0 = 0.2: In the data from the above example (6.

see also equation (4. the previous example information used and expanded.90500 1.38034 0. First. the exit pressure is between point “a” and point “b”.1755 6.3709 My 0.93118 1.0128 3. . The area (location) that the previous example did not “produce” the “right” solution (the exit pressure was 2. there are two main possible ways to obtain the solution.1).g. Utilizing the Table (4.2) or the GDC-Potto provides the following table is obtained T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ M 0. Thus.5[unitless!] Px 0 × A x ∗ Py 0 × A y ∗ 4×3 A With the knowledge of the ratio PP A∗ which was calculated and determines the exit 0 Mach number.55250 With the information of Mach number (either Mx or My ) the area where the shock (location) occurs can be found.5525 0. utilizing the isentropic Table (4. It follows that there must exist a shock in the nozzle.6575 0.905 4 Looking up in the Table (4. Pexit Aexit 2×9 Pexit Aexit = = = 1.97188 0.52628 Ty Tx ρy ρx Py Px P0 y P0 x 2. the exit Mach number.2). The exit total pressure can be obtained (if needed).72)). The solution process starts at the nozzle’s exit and progress to the entrance.75158 With these values the relationship between the stagnation pressures of the shock are obtainable e.2) or utilizing the GDC-Potto provides Mx 2. NORMAL SHOCK IN VARIABLE DUCT AREAS S OLUTION The solution procedure is similar to what was shown in previous Example (6. Here. More importantly the pressure ratio exit is known. is known. In fact.3914 0.128 CHAPTER 6. The ratio of the ratio of stagnation pressure obtained by f or Mexit P0 y = P0x P0 y Pexit Pexit P0x = 2 1 × = 0. it requires some iterations by “smart” guessing the different shock locations. 2 Of course.113[Bar]. Mathematically. The conditions in the tank are again the stagnation conditions. My . the computer can be use to carry this calculations in a sophisticate way. 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.5000 0. the needed pressure is only 2[Bar] which means that the next guess for the shock location should be with a larger area2 . In the first method.

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. as “simple” check this value is larger than the value in the previous example.2 Diffuser Efficiency . Cd = mactual ˙ mideal ˙ (6.1. 6.99.3) (6.1 Nozzle efficiency Obviously nozzles are not perfectly efficient and there are several ways to define the nozzleefficiency. 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.15205 2. The actual energy that was used is E = h0 − hexit (6.5) 6.0188[cm2] = A∗ Note.1) where hexit s is the enthalpy if the flow was isentropic.9 to 0. The total energy that can be converted is during isentropic process is E = h0 − hexit s (6.16747 Approaching the shock location from the upstream (entrance) yields A= A ∗ A = 2.6.07158 0.2) The efficiency can be defined as η= The typical efficiency of nozzle is ranged between 0.47076 0.3396 0.3396 × 3 ∼ 7. NOZZLE EFFICIENCY M 2. In the literature some define also velocity coefficient as the ratio of the actual velocity to the ideal velocity.3709 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 129 0. Vc Vc = √ η= (Uactual )2 (Uideal ) 2 h0 − hexit (Uactual )2 = 2 h0 − hexit s (Uideal ) (6.

3) 0. S OLUTION The condition at M = 3 is summarized in following table M 3.0000 T T0 ρ ρ0 A A P P0 uous region (and also for example (6.130 CHAPTER 6.7) s.07623 4.3: A wind tunnel combined from Diffuser nozzle a nozzle and a diffuser (actually two nozzles connected by a 1 2 4 3 constant area see Figure (6.2346 0.0 and prescapacitor sure of 0.6) P02 P2 01 02 2 P1 1 For perfect gas equation (6.7) results in η= kR 2 k−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 (6.8) Example 6.11528 0. 6. Assume that a shock Fig. k = 1.4)) the required condition at point 3 are: M = 3. What is cooler area of nozzle’s throat and what is area of the diffuser’s throat to maintain chocked diffuser with heat subsonic flow in the expansion out section.6) can be converted to η= 2Cp (T3 − T1 ) U1 2 (6.3: Description to clarify the definition of diffuser efficiency And further expanding equation (6.02[m2 ]. The cross section in area between the nuzzle Compressor and diffuser is 0.7[Bar] and temperature of 250K.65326 A×P A∗ ×P0 kj w£i hg w£f F F∗ . 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.02722 0. 6.4 can be assumed.4: Schematic of a supersonic tunnel in a continoccurs in the test section.35714 0. η= 2(h3 − h1 ) h3 − h 1 = h01 − h1 U1 2 (6.entropy Fig.

pressure of 2[Bar] and temperature of 350K.479 34.0000 My 0.3333 0. temperature and pressure “upstream” known.5346 My Mx My 1.0047 ∼ 0. S OLUTION This is a case of completely and suddenly open valve with the shock velocity. ∗ Tx ∗ Ty Tx ∼ 2. In this case Potto–GDC provides the following table Mx 5.0047[m2] A 131 In this case.0143[m3] P0 d 0.02172 The calculation of the temperature and pressure ratio also can be obtain by the same manner. P0 A∗ is constant (constant mass flow).32834 A∗ d = P0 n ∗ 1 A n∼ 0.47519 Ty Tx ρy ρx Py Px P0y P0 x 2. The “downstream” shock number is Msy = Us k ∗ 287.37554 0.4789 6.2.989 Ty Tx Py Px P0 y P0 x 0.09668 .8571 10.4968 0.3.6. Calculate the conditions after the shock.32834 Example 6.6790 3.2963 34.2346 = 0.4: A shock is moving at 200 [m/sec] in pipe with gas with k = 1. For example Potto-GDC (this code was produce by the program) Mx 5.5346 My 0.50 0. First the stagnation behind the shock will be Mx 3.021717 The calculations were carried as following: First calculate the Mx as M x = Us / (k ∗ 287.37554 Ty Tx ρy ρx Py Px P0y P0 x 5. DIFFUSER EFFICIENCY The nozzle area can be calculated by A∗ n = A A = 0.0 5.02/4. ∗ Tx ) Then calculate the My by using Potto-GDC or utilize the Tables.

41087 ∼ 1.15 cx 1.25 Ux =√ ∼ 2.4 ∗ 287 ∗ 369.5: 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. what will be the piston velocity? S OLUTION This is an open valve case in which the pressure ratio is given. If the steady state is achieved.2308 = 369.2308 1.25[m/sec] ρA PA 200000 × 0.75593 ∗ 1.091 × 143 × 350 Thus the static Mach number.09668 − 0.96697 The temperature ratio and the Mach numbers for the velocity of the air (and the piston) can be calculated.3628 My 0. Calculate the conditions behind the shock wave.002[m3 ]. NORMAL SHOCK IN VARIABLE DUCT AREAS Finally utilizing the equation to calculate the following My = Msy − My = 2.75593 Ty Tx ρy ρx Py Px P0 y P0 x 1.132 CHAPTER 6. Mx is Mx = . Ux = m ˙ mRT ˙ 2 × 287 × 350 = = ∼ 502. the “upstream” Mach is given. Thus. The propagating piston is assumed to move into media with temperature of 300K and atmospheric pressure of 1[Bar]. The temperature at “downstream” (close to the piston) is Ty = 300 × 1.24 ∼ 291.0000 0.002 502.24[◦C] Ty = T x Tx The velocity of the piston is then √ Uy = My ∗ cy = 0.6250 2. S OLUTION This is the case of a closed valve in which mass flow rate with the area given.091 (Butane?).16[m/sec] Example 6.6: A flow of gas is brought into a sudden stop.989 Example 6. The mass flow rate of the gas is 2 [kg/sec] and cross section A = 0. The imaginary gas conditions are temperature is 350K and pressure is 2[Bar] and R = 143[j/kg K] and k = 1. For this pressure ratio of Py /Px = 2 the following table can be obtained or by using Potto–GDC Mx 1.

0 Ty Tx Py Px P0 y P0 x 133 0.0 0.47995 0.922 2.590 2.2.0 0.0 2.0 0.9222 My Mx My 0.796 0. DIFFUSER EFFICIENCY With this value for the Mach number Potto-GDC provides Mx 2.589 9.47996 2.804 9.923 2.4096 9. The iteration of the procedure are i 0 1 2 3 4 5 Mx 3.46689 0.6.796 9.922 2.1500 2.922 My 0.47996 0.589 2.8598 2.914 9.35101 This table was obtained by using the procedure described in this book.47996 Ty Tx Py Px My 0.47988 0.796 9.0 0.0 0.609 2.589 11.796 .1500 2.47886 0.589 2.940 2.

134 CHAPTER 6. NORMAL SHOCK IN VARIABLE DUCT AREAS .

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

NOZZLE FLOW WITH EXTERNAL FORCES 7. i.1) 7.1 Isentropic Nozzle (Q = 0) external work or potential difference.136 CHAPTER 7.e.2 Isothermal Nozzle (T = constant) . z × g dh + U dU = f (x)dx The energy equation for isentropic nozzle provides (7.

the study of the isothermal flow above this point is only an academic discussion but also provides the upper limit for Fanno Flow. in a perfect gas.v. the pressure difference reduces the density of the gas. The high speed of the gas is obtained or explained by the combination flow of heat transfer and the friction to the direction flow.1: Control volume for isothermal flow tain conservation of mass.Fig. To main. For instance.).A. is relatively high and the isothermal flow model is not valid anymore. It is more predominant (more applicable) in situations where the gas is pumped over a length of kilometers. 1 This To put discussion for what the “relatively rapid” means. for engineering purposes. the density is inverse of the pressure (it has c. the heat transfer. For example.S. this model is applicable when a natural gas flows over several hundreds of meters. At critical point the velocity reaches the speed of sound at the exit and hence the flow will be choked1 .CHAPTER 8 Isothermal Flow In this chapter a model dealing with gas that flows through a long tube is described. Such situations are common in large cities in U. as a constant . 8. the velocity increases inversely to the pressure. where natural gas is used for heating. Therefore. to be kept in mind that the gas undergoes an isothermal process. √ explanation is not correct as it will be shown later on. For a long pipe. Close to the critical point (about. 137 Œ Š ‹ƒ … Š s‰qˆ33z‡† † … ƒ ‚ … ƒ {z„‚ |€z~|  } v y w {zxv  Ž ’ ‘ t rp o usqpnm  l . 1/ k. 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.

).2 it seems obvious to write this equation perhaps to consult with others.1) describes the flow of gas from the left to the right.138 CHAPTER 8. equation of state reads dP dρ = P ρ It is convenient to define a hydraulic diameter DH = 4 × Cross Section Area wetted perimeter (8.7) (8. (8.1 The Control Volume Analysis/Governing equations Figure (8. The shear stress is the force per area that acts on the fluid by the tube wall. First it must be recalled that the temperature is constant and therefore.4) (8. ISOTHERMAL FLOW 8. a close enough shape is sufficient.1) (8.5) 8. Dimensionless Representation In this section the equations are transformed into the dimensionless form and presented as such. The heat transfer up stream (or down stream) is assumed to be negligible. 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. Hence.2) where A is the cross section area (it doesn’t have to be a perfect circle. equation of state is expressed as the following: P = ρRT (8.6) .3) Again it is assumed that the gas is a perfect gas and therefore. The Awetted area is the area that shear stress acts on. 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.

the Fanning friction factor2 is introduced. U .11) d(M 2 ) d(U 2 ) dT = − M2 U2 T (8.14) Differentiation of the isotropic (stagnation) relationship of the pressure (4. this factor is a dimensionless friction factor sometimes referred to as the friction coefficient as f= τw 1 2 2 ρU (8.2) yields 4dx −dP − f DH 1 2 ρU 2 m ˙ A = ρU dU (8.9) and using the identify for perfect gas M 2 = ρU 2 /kP yields: − 4f dx dP − P DH kP M 2 2 = kP M 2 dU U (8.12) (8.11) yields should be noted that Fanning factor based on hydraulic radius.2.8) Substituting equation (8. instead of diameter friction equation. thus “Fanning f” values are only 1/4th of “Darcy f” values.10) Now the pressure. 2 It where are the stagnation equations? put them in a table put explanation how to derive this expression. U 2 = kRT M 2 Differentiation of equation (8.8. DIMENSIONLESS REPRESENTATION 139 Now.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. P as a function of the Mach number has to substitute along with velocity.8) into momentum equation (8.11) yields d(U 2 ) = kR M 2 dT + T d(M 2 ) (8. .9) Rearranging equation (8.15) (8.

ρ)3 and with these four equations the solution is attainable. ISOTHERMAL FLOW 1 2 2 kM k−1 + 2 M2 dM 2 M2 (8. There are 4 unknowns (M.17) yields: T0 1 k−1 2 2 d M + k−1 M 2 2 dT0 = Rearranging equation (8.17) Notice that dT0 = 0 in an isothermal flow.20) to become dU dP =− P U 3 Assuming (8. T. state are described above.15) when combined with equation (8.16) Differentiation of equation (4.21) (8. The distance friction. Thus. is selected as the choice for the independent D variable. 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 (8.20) The four equations momentum. the equations need to be obtained as a function of 4f L . These different solutions are supersonic and subsonic solution.14)) leads dM 2 2dU = 2 M U (8.140 dP0 dP = + P0 P 1 CHAPTER 8.19) By utilizing the momentum equation it is possible to obtain a relation between the pressure and density. 4f L .18) yields M2 M2 (8.22) the upstream variables are known. energy. .18) dT0 (k − 1) M 2 dM 2 = T0 2 1 + k−1 M 2 2 (8. P.9) yields: dT0 = dT 1+ k−1 2 M 2 +T k−1 dM 2 2 (8. Substituting T for equation (8. There is no change in the actual temperature of the flow but the stagnation temperature increases or decreases depending on the Mach number (supersonic flow of subsonic flow). continuity (mass). One can notice that there are two possible solutions (because of the square power). The D density is eliminated from equation (8.

thus. By denoting the superscript symbol ∗ for the choking condition. Thus. Since the stagnation temperature (T0 ) has a finite value which means that dT0 → ∞. Integration of equation (8.26) that when M → √k the value of right hand side approaches infinity (∞).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. the term on D the other side has to be positive as well.10). The term 4f L is positive for any x.24) can be separated to obtain integrable form as follows L 0 4f dx = D 1/k M2 1 − kM 2 dM 2 kM 2 (8.27) yields 4f Lmax D = 1 − kM 2 + ln kM 2 kM 2 (8.26) The variables in equation (8. A more appropriate model is an adiabatic flow model yet it can serve as a bounding boundary (or limit).22) into equation (8. one can obtain that 1/k M2 = ∗2 U2 U (8. To obtain this restriction 1 = kM 2 . Heat transfer has a limited value therefore the model of the flow must be changed.24) Similarly or by other path the stagnation pressure can be expressed as a function of 4f L D kM 2 1 − k+1 M 2 dx dP0 2 = 4f P0 D 2 (kM 2 − 1) 1 + k−1 M 2 2 dT0 dx k (1 − k) M 2 4f = 2 ) 1 + k−1 M 2 T0 D 2 (1 − kM 2 (8. one can obtain − dP 4f dx − P DH kP M 2 2 = kP M 2 dP P 141 (8.8.25) (8. 1 the value M = √k is the limiting case from a mathematical point of view. The physical meaning of this value is similar to M = 1 choked flow which was discussed in a variable area flow in Chapter (4).23) Equation (8.27) It can be noticed that at the entrance (x = 0) for which M = Mx=0 (the initial velocity in the tube isn’t zero). 1 Further it can be noticed from equation (8.28) 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. DIMENSIONLESS REPRESENTATION After substituting the velocity (8.2.29) . When 1 Mach number larger than M > √k it makes the right hand side of the integrate negative.

24) 4f L D = 4f Lmax D 1 − 4f Lmax D 2 = 1 − kM1 2 1 − kM2 2 − + ln kM1 2 kM2 2 M1 M2 2 (8. =⇒ Reusing the perfect–gas relationship ρ 1 P = ∗ =√ ∗ P ρ kM Now utilizing the relation for stagnated isotropic pressure one can obtain P0 P ∗ = P∗ P0 Substituting for P P∗ (8.35) are presented on in Figure (8.30)-(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.31) (8.34) And the stagnation temperature at the critical point can be expressed as T 1 + k−1 M 2 T0 2k 2 = ∗ = T∗ T0 3k − 1 1 + k−1 2k 1+ k−1 2 M2 (8.32) 1+ 1 k−1 2 2 M k−1 + 2k k k−1 (8. Denote 1 and 2 as the conditions at the inlet and exit respectably. ISOTHERMAL FLOW Rearranging equation (8.36) is reduced into the following approximation ∼0 4f L D = 2 ln M1 − 1 − 1 − kM2 2 kM2 2 (8.142 CHAPTER 8.37) .35) These equations (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 (8.29) is transfered into √ U = kM U∗ Utilizing the continuity equation provides ρU = ρ∗ U ∗ .2) 8.36) For the case that M1 >> M2 and M1 → 1 equation (8. It is very useful to obtain the relationship between the entrance and the exit condition for this case.30) ρ 1 =√ ρ∗ kM (8.33) equation (8. From equation (8.

1 Fri Feb 18 17:23:43 2005 0.8.2: Description of the pressure.38) This relationship shows the maximum limit that Mach number can approach when the heat transfer is extraordinarily fast. 8.5. COMPARISON WITH INCOMPRESSIBLE FLOW 143 Isothermal Flow P/P . Yet this model provides the directions of the heat transfer effects on the flow. ρ/ρ and T0/T0 as a function of M 1e+02 4fL  D P ρ  or  * ∗ P ρ T0/T0 P0/P0 1 * * * * * 1e+01 0.4. 8. This velocity requires a large entrance length to achieve good heat transfer. one should expect that the isothermal flow should have similar characteristics as . With this conflicting mechanism obviously the flow is closer to the Fanno flow model.01 1 Mach number 10 Fig. 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. In these cases. even small 4f L > 2 results in D a Mach number which is larger than 4.4 Comparison with Incompressible Flow The Mach number of the flow in some instances is relatively small. In reality.1 0.

ISOTHERMAL FLOW incompressible flow.45) now we have to expand into a series around χ = 0 and remember that f (x) = f (0) + f (0)x + f (0) x2 + 0 x3 2 (8.39) one can obtained that M2 = M 1 P1 P2 (8.42) Now equation (8.144 CHAPTER 8. P1 − P2 /P1 .40) yields 4f L D = 1 kM1 2 1− P2 P1 2 − ln P2 P1 2 (8. the pressure loss is expressed as follows P1 − P 2 = 4f L D U2 2 (8.46) . Expanding the solution for small pressure ratio drop.40) Substituting this expression into (8. a relationship between M2 and M1 and pressures has to be derived. To obtain a similar expression for isothermal flow. From equation (8.41) Because f is always positive there is only one solution to the above equation even though M2.44) 4f L D = 1 2χ − χ2 − ln kM1 2 1 1−χ (8.39) Now note that for incompressible flow U1 = U2 = U and 4f L represent the ratio of D the traditional h12 .43) 4f L D = 1 2 1 − (1 − χ) − ln kM1 2 1 1−χ 2 2 (8. For incompressible flow. denote χ= P1 − P 2 P1 (8. by some mathematics.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.

47) similarly it can be shown that f (χ = 0) = 1 equation (8. and is based on estimates. 8.52) A few observations can be made about equation (8.8.49) and further rearrangement yields 4f L D = χ 2(1 − kM1 2 ) − 1 + kM1 2 χ + f χ3 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. Currently and ashamedly the author is looking for a more simplified explanation.5 Supersonic Branch Apparently. 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.48) yields 4f L D = χ (2 − χ) − kM1 2 (2 − χ) + f χ3 kM1 2 (8. 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 (8.45) now can be approximated as 4f L D = 1 (2χ − χ2 ) − 2χ − χ2 + f χ3 kM1 2 (8. The current explanation is correct but based on hands waving and definitely does not satisfy the author.50) in cases that χ is small 4f L D ≈ χ 2(1 − kM1 2 ) − 1 + kM1 2 χ kM1 2 (8. .5.48) rearranging equation (8. Equation (8. A dimensionless analysis4 demonstrates that all the common materials that the author is familiar which creates a large error in the fundamental 4 This dimensional analysis is a bit tricky.52).51) value of The pressure difference can be plotted as a function of the M1 for given 4f L D .

Almost all the perfect gas model substances dealt with in this book. if Mach number is changing from 10 to 1 the kinetic energy change is T about T00∗ = 18.005 results in Lmax 10 = = 500 D 4 × 0. the hydraulic entry length is very large as will be shown below. (0. a gas flows in a tube with 4f L = 10 the required entry Mach D number is over 200. are 0. value for the common gases. Remember from Fluid Dynamic book Lentrance = 0. In the supersonic flow.7 D On the other hand a typical value of friction coefficient f = 0.0000185 kg/m-sec at 300K and 0. So.53) The typical values of the the kinetic viscosity. For example. this model can provide a better understanding to the trends and deviations of the Fanno flow model. Now. For this illustration. over hundred of meters per second.0000130034 kg/m-sec at 200K. . this topic will be covered in the next version because it provide some insight and boundary to the Fanno Flow model. Nevertheless. ν. even with low temperature like 200K the speed of sound of air is 283[m/sec]. The flow speed at the entrance is very large.71. However. the feeding diverging nozzle somewhat reduces the required entry length (as opposed to converging feeding).6 5 see Figures and Tables Kays and Crawford “Convective Heat Transfer” (equation 12-12). For example. 8. even for relatively small tubes with 4f L = 10 the inlet speed is over 56 [km/sec].06 UD ν (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). the speed of sound is a function of temperature. 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 . Now with limitation. The thermal entry length is in the order of the hydrodynamic entry length (look at the Prandtl number. Combine this information with our case of 4f L D = 10 Lentrance = 250746268. for most gas cases the speed of sound is about 300[m/sec].146 CHAPTER 8.).0). ISOTHERMAL FLOW assumption of the model and the model breaks.37 which means that the maximum amount of energy is insufficient. This D requires that the entrance length to be larger than the actual length of the tub for air.

8172 2.03000 0. ISOTHERMAL FLOW EXAMPLES Table 8.6903 1.50000 0.09000 0.4147 2.4515 4.00626 0.011 1.9682 1.0255 1.88594 0.9747 7.3906 8.3806 2.40000 0.87544 0.000 17.6651 13.1718 21.056 1. These questions or examples will appear in the later versions.1289 16.2668 0.2126 1.4784 1.89 82. .6109 8.06 192.00371 0.).04000 0.043 1. 7 Those who are mathematically inclined can include these kinds of questions but there are no real world applications to isothermal model with shock.99741 1.000 8.6500 5.1269 1.97 439.89644 0.2074 1.8172 2.1289 16. and mass flow rate questions.87563 0.5366 1.056 1.06000 0.1129 1. etc.9031 14.003 1.8.97344 0.7.0859 12.3524 1.10000 0.60000 0.29895 0.0495 1.000896 0.7 Isothermal Flow Examples There can be several kinds of questions aside from the proof questions6 Generally.0677 1.80732 0.1718 21.2565 1.3806 2.1827 1.9925 4.0736 10.8781 1.6470 1.45000 0.5920 6.009 1.87675 0.98982 0.75000 0.00205 0. In this model no questions about shock (should) exist7 .87612 0.84515 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 785.6903 1.2258 3.08085 0.93800 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).9181 5.032 1. friction coefficient.87516 0.98700 0.03095 0.8781 1.82758 0.8650 3.7040 66.87642 0.0736 10.33 279.50207 0.4086 1.043 1.4147 2.20000 0.5366 1.0859 12.000220 0.96075 0.79 105.87586 0.91044 0.9031 14.1: The Isothermal Flow basic parameters 147 M 0.55000 0.021 1.88200 0.2258 3.8791 1.005 1.05000 0.2074 1.1129 1.1259 1.007 1.12 139.83637 0. friction factor.4515 4.0823 1.92794 0.7230 2.8493 7.3906 8. the “engineering” or practical questions can be divided into driving force (pressure difference).99232 0.001 1.87528 0.3002 1.70000 0.81879 0.2553 10.021 1.000 0.35000 0.3334 2.3002 1.30000 0.99485 0. The driving force questions deal with what should be the pressure difference to obtain certain flow rate.94894 0. resistance (diameter.011 1.4086 1.89075 0.81000 0.1269 1.032 1.08000 0.0 28.80000 0.16552 0. Here is an example.1599 13.07000 0.91875 0.90300 0.5644 9.25000 0.5644 9.65000 0.000 28.

9 = 4 sec The maximum flow rate then reads m = ρAU = 23. ρ. check whether flow is choked (or even possible).e. surroundings temperature 27◦ C.19 RT 290 × 300 m3 π × (0.25 [m] diameter and 5000 [m] in length is attached to a pump.89446 The maximum flow rate (the limiting case) can be calculated by utilizing the above table. the flow is not choked. What should be the pump pressure so that a flow rate of 2 [kg/sec] will be achieved? Assume that friction factor f = 0.31. the flow rate is really some what different. In incompressible flow.0 0.00 20. 4f L D 4f L D = 4 × 0.148 CHAPTER 8. R = 290 Kkg . It is more appropriate to assume an isothermal model hence our model is appropriate.005 and the exit pressure is 1[bar].0055000 = 400 0.1) or the program provides M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 0. the density is a function of D 2g √ the entrance Mach number. ISOTHERMAL FLOW Example 8. To solve this problem the flow rate has to be calculated as m = ρAU = 2.1743 12.0 ˙ m= ˙ kg sec P1 P1 P1 kU kU A =√ = AkM1 A√ RT k c kRT kRT .19 × ˙ The maximum flow rate is larger then the requested mass rate hence the flow is not choked. Hint: calculate the maximum flow rate and then check if this request is reasonable.04331 × √ m 1. The velocity of the gas at the entrance U = cM = 0.31 × 290 × 300 ∼ 14. The density reads = ρ= P 2.25 Utilizing Table (8. k = 1.5921 0. The exit Mach number is not necessarily 1/ k i.62 sec .25)2 kg × 14.62 ∼ 16.1: A tube of 0. The specific J heat for the gas. First. the velocity can be calcu2 lated by utilizing ∆P = 4f L U . 017. S OLUTION If the flow was incompressible then for known density. It is note worthy to mention that since the isothermal model breaks around the choking point. Calculating the resistance. 450 ∼ kg = = 23.04331 400.

A flow rate of 0.103 2 P2 Ak 100000 × π×(0. The friction coefficient for the tube can be assumed as 0.1) or using the provided program yields 8 It is unfortunate. the maximum pressure allowed for the gas is only 10[bar].25) × 1. .6779 + 400 ∼ 466.5844 0.5 [in]8 .31 P = 21.7.8.89442 Note that tables in this example are for k = 1.89567 The entrance Mach number is obtained by 4f L D 1 = 66. Assume that the flow is isothermal and k=1. Due to safety concerns.0 0.04014 466. You can assume that the soundings temperature to be 27◦ C.02 (A relative smooth tube of cast iron.7678 13.).566[bar] Example 8. ISOTHERMAL FLOW EXAMPLES Now combining with equation (8. the minimum diameter will be obtained when the flow is choked. Note that tubes are provided in increments of 0. calculate the required diameter of tube.2: A flow of gas was considered for a distance of 0.2 [kg/sec] is required. with the value of M1 either by utilizing Table (8.31 4 P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 66.4.0845 M1 = M 2 P1 k 10 Now.40) yields m= ˙ M2 = M2 P2 Ak c 149 From Table (8.76780 × 8.0 0.68 = P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ M 4fL D 0.10300 4fL D P P∗ 2 × 337.1) or by utilizing the program M 0.59 mc ˙ = = 0.68 The pressure should be 21.6779 8. the maximum M1 that can be obtained when the M2 is at its maximum and back pressure is at the atmospheric pressure.4826 = 2. but it seems that this standard will be around in USA for some time. Mmax 1 1 P2 = √ = 0. S OLUTION At first. Thus.3249 0.5 [km] (500 [m]).4826 5.

0 0. With this pipe size the calculations are to be repeated in reverse and produces: (Clearly the maximum mass is determined with) √ √ P P AM k m = ρAU = ρAM c = ˙ AM kRT = √ RT RT The usage of the above equation clearly applied to the whole pipe.87625 = 94. pressure and etc) have to be taken at the same point. 000. the pipes are provided only in 0.42359[m] = 16.2424 0.61 RT 287 × 300 m3 The velocity at the entrance should be m U = M ∗ c = 0.02 × 500 0.87627 To check whether the flow rate satisfies the requirement m= ˙ 106 × π×0.4 ≈ 50.6 = sec .431 the value of minimum diameter.4310 10.68[in] However.0853 × √ 287 × 300 √ 1.4318[m].3[kg/sec] Since 50.2991 0. 000 kg P = = 11.4 × 287 × 300 ∼ 347.64 ρ ρ∗ T0 T0 ∗ M 0.0 0.2 = sec and the density is ρ= 1. The only point that must be emphasized is that all properties (like Mach number.9110 6. The speed of sound at the entrance is √ √ m c = kRT = 1.4318 P P∗ P0 P0 ∗ 92.08528 × 347.2 ∼ 29.43 0.2 the mass flow rate requirement is satisfied. The new 4f L is D 4f L D = 4 × 0. It should be noted that P should be replaced by P0 in the calculations.08527 4fL D 92.43182 4 × 0.08450 With 4f Lmax D 4fL D P P∗ CHAPTER 8.6400 9.0018 6.02 × 500 94. D= 4f L 4f Lmax D 4 × 0.150 M 0. ISOTHERMAL FLOW P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 94.3 ≥ 0.5 increments and the next size is 17[in] or 0.

Calculate the Mach number at the entrance to pipe and the flow rate. The flow rate is given with the diameter of the pipe.19235 ≈ 12.6 × 11. The gas and the sounding temperature can be assumed to be 300 K. ISOTHERMAL FLOW EXAMPLES The diameter should be D= 4m ˙ = πU ρ 4 × 0. k=1.and P0 ∗T ∼ = 28 12.7. Example 8.67 2.72021 0.21[bar] 20. 1. Assume that the flow is isothermal.4[m] diameter and 4000 [m] length to a different station (b).8.4. and the average friction f=0. for the sake of the exercise the other parameters will be calculated.3: A gas flows of from a station (a) with pressure of 20[bar] through a pipe with 0. Therefore.0419 . The pressure at the exit (station (b)) is 2[bar].61 151 Nevertheless. This situation is reversed question. at first it will be assumed that the whole length is the maximum length.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. 4f Lmax D = 4 × 0.01. S OLUTION First. It should be noted that the flow isn’t choked.0419 400.19235 12.4 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 0.027 = π × 29.52828.66915 From the table M1 ≈ 0.01 × 4000 = 400 0.87531 P0 P0 ∗T 20. the information whether the flow is choked needs to be found.2 ∼ 0. guess reasonable value of M1 and calculate 4f L D . P2 = P0 ∗T P2 P0 ∗T = 2. The solution is an iterative process.21 × 0.52828 = 1.67 The pressure at point (b) by utilizing the isentropic relationship (M = 1) pressure ratio is 0.

ISOTHERMAL FLOW by subtracting 4f L D 1 − 4f L D 3.10000 0.16922 0. Now the process has been done for you and is provided in Figure (??) or in the table obtained from the provided program. there are no examples on isothermal with supersonic flow.83740 4f Lmax D 1 4f L D P2 P1 0. 4. Calculate the pressure.00000 0.40737 0.4 P k π × D2 M= √ m = ρAM c = √ ˙ π × 0.5914 0.7272 0.6684 0.10000 0.83997 0. M1 0.8 Unchoked situation Table 8.0419 4 300 × 287 RT 42.32131 400.84095 0.10000 8.10000 0.152 2.0419 The flow rate is √ √ 2000000 1.83889 0.32795 0.10000 0.5807 0.12942 0.12949 0.16912 0.83827 0. P2 bear in mind that this isn’t the real pressure but based on assumption 5.08971 0. Calculate the value of 4f L D 2 CHAPTER 8.32807 0.40754 0.36780 0.08978 0.46[kg/sec] In this chapter.05000 0.10000 0.36766 0.05005 0.10000 . Compare the results of guessed pressure P2 with the actual pressure and choose new Ma number M1 accordingly.83920 0. Obtain M2 from the Table ? or by using the Potto–GDC.6934 0.4: The flow parameters for unchoked flow M1 0. M2 0.84018 0.59338 4f Lmax D 1 4f L D P2 P1 400.6483 0.10000 0.5708 M2 0.22 × 0.

1 0 0 10 20 30 40 50 4fL  D 60 70 80 90 100 M1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.5 = 0.5 0.6 0.10 Fri Feb 25 17:20:14 2005 Fig.3: The Mach number at the entrance to a tube under isothermal flow model as a function 4f L D .2 = 0. 8.8.2 0.3 0.8.9 0.8 = 0.8 0.4 0. UNCHOKED SITUATION 153 M1 isothermal flow 1 0.7 0.

154 CHAPTER 8. ISOTHERMAL FLOW .

As this model raised from need to explain the steam flow in turbines.1 Introduction Consider a gas flowing through a conduit with a friction (see Figure (9. more general cases will be examined2 . discussed on the ideal gas model and the entry length issues. compressed air systems.v. etc. This model explains many industrial flow processes which includes emptying of pressured container through a relatively short tube.CHAPTER 9 Fanno Flow An adiabatic flow with friction is named after Ginno Fanno a Jewish flow direction engineer. This model is apc. The main restriction for this model is that heat transfer is negligible and can be ignored 1 . This model is the second pipe flow model described here. 1 Even 2 Not the friction does not convert into heat ready yet. 9. exhaust system of an internal combustion engine.1: Control volume of the gas flow in a constant cross section number. plicable to flow processes which are No heat transer very fast compared to heat transfer mechanisms with small Eckert Fig.1)). Later. It is advantages to examine the simplest situation and yet without losing the core properties of the process. 155 ® ¯­ ¨¦ § § ­ %¬~ª«© ¨¦ © ¥ § ¦ Ÿ¨¥  už¢  £ ¡ ›  œ ŸžV› ± D° ³ D² š ™— – „h˜3•” ¤ “ . 9.

6) 3 The equation of state is written again here so that all the relevant equations can be found when this chapter is printed separately.2) (9.2 Model m = ρAU = constant ˙ → ρ 1 U1 = ρ 2 U2 (9.3) Again for simplicity.5) It is assumed that the flow can be approximated as one–dimensional. . the perfect gas model is assumed3 .8) 4 × Cross Section Area wetted perimeter (9.156 CHAPTER 9.4) U1 = 2cp 2 T0 2 T2 + U2 2cp 2 (9. P = ρRT P2 P1 = → ρ 1 T1 ρ 2 T2 (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. FANNO FLOW 9.7) (9. The force acting on the gas is the friction at the wall and the momentum conservation reads −AdP − τw dAw = mdU ˙ It is convenient to define a hydraulic diameter as DH = Or in other words A= πDH 2 4 (9.

11) by the cross section area. By utilizing the definition of the sound speed to produce the following identities for perfect gas M2 = U c 2 = U2 k RT P ρ (9.10) By utilizing equation (9.3 Non–dimensionalization of the equations Before solving the above equation a dimensionless process is applied.12) The second law is the last equation to be utilized to determine the flow direction.15) .14) Utilizing the definition of the perfect gas results in M2 = ρU 2 kP (9.13) 9. s2 ≥ s 1 (9.6) yields A τw 2 m ˙ A − πD dP − πDdx f 4 1 2 ρU 2 = A ρU dU (9.11) yields Dividing equation (9.3.2) and substituting equation (9.9.9) 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 ρU 2 2 (9. A and rearranging 4f dx D 1 2 ρU 2 −dP + = ρU dU (9. The infinitesimal area that shear stress is acting on is dAw = πDdx (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.10) into momentum equation (9.

Thus.16) results in − dP 4f dx − P D kM 2 2 = kM 2 dU U (9. (9. (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 (9.20) Dividing the energy equation (9.16) By further rearranging equation (9.19) Derivation of the Mach identity equation (9.20).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.14) and dividing by equation (9.17).21) need to be solved.14) yields d(M 2 ) d(U 2 ) dT = − 2 M U2 T (9. the original limitations must be applied to the resulting equation.158 CHAPTER 9.17). (9.5) results dρ dT dP = + P ρ dT (9.14) and substituting it into equation (9. Derivative of mass conservation ((9.19).5) and dividing the results by equation of state (9.2)) results in dU U dρ 1 dU 2 + =0 ρ 2 U2 (9.21) Equations (9. (9.18) The derivation of the equation of state (9. FANNO FLOW Using the identity in equation (9.17) It is convenient to relate expressions of (dP/P ) and dU/U in terms of the Mach number and substituting it into equation (9. and (9.18). These equa- .

28) .21) when eliminating dT /T results dP dρ (k − 1)M 2 dU 2 = − P ρ 2 U2 (9.25) results in kM 2 1 + k−1 M 2 4f dx dM 2 2 = M2 1 − M2 D (9.18) and equation (9.22) The term dρ can be eliminated by utilizing equation (9.3. M variable is obtained by combining equation (9. Explicit explanation is provided for only two variables.27) and the relationship for the temperature is density is obtained by utilizing equations (9. NON–DIMENSIONALIZATION OF THE EQUATIONS 159 tions are separable so one variable is a function of only single variable (the chosen as the independent variable).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. causes the change in the other variables.23) kM 2 1 + (k − 1)M 2 4f dx dP =− P 2(1 − M 2 ) D (9.18) to obtain dρ 4f dx kM 2 =− 2) D ρ 2 (1 − M (9.26) (9.25) 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 (9.23) The second equation for Mach number. Then dρ/ρ and U are eliminated by utilizing equation (9.18) and substituting it into ρ equation (9. 4f L . The dimensionless friction.20) and (9.24) (9.9.19) and (9.27) and (9. the rest variables can be done in a similar fashion. D Combining equations (9. The only variable that is left is P (or dP/P ) which can be eliminated by utilizing equation (9.29) 1 dc k(k − 1)M 4 4f dx dT = =− T 2 c 2(1 − M 2 ) D (9.22). D is chosen as the independent variable since the change in the dimensionless resistance.24) and results in 1 − M 2 dM 2 4f dx = D kM 4 (1 + k−1 M 2 ) 2 Rearranging equation (9.21) by eliminating dT /T . 4f L .

Two. When M < 1 the pressure decreases downstream as can be seen from equation (9.24) because f dx and M are positive. the pressure increases downstream.34). For example. This pressure increase is what makes compressible flow so different from “conventional” flow.24) through (9.34) 9. in the supersonic branch.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. 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 (9.32) 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. flow above speed of sound. flow with speed below the speed of sound.24) it can be observed that the critical point is when M = 1. FANNO FLOW (9. For the same reasons. M > 1. from equation (9.33) The first law requires that the stagnation temperature remains constant.4 The Mechanics and Why the Flow is Choked? The trends of the properties can be examined by looking in equations (9.30) dP dT − R ln T P (9.35) (9. (dT0 = 0). Thus the discussion will be divided into two cases: One. .31) The stagnation temperature expresses as T0 = T (1 + (1 − k)/2M 2 ). 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.

THE WORKING EQUATIONS Why the flow is choked? 161 Here. when the flow is supersonic.26). T0 9. M = 1 at the last point. M Velocity. For example.24)).36) A representative friction factor is defined as ¯ f= 1 Lmax 0 Lmax f dx (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. 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. ρ Stagnation Temperature. For example. the pressure can be examined by looking at equation (9.25) yields f dx = L (9. P Mach number. Observing from equation (9.5. On the other hand.9. T Density. It demonstrates that the Mach number increases downstream when the flow is subsonic. However. dP (@M = 1) = ∞ and mathematically it is a singular point (see equation (9.37) . This constrain means that because the flow cannot “crossover” M = 1 the gas has to reach to this speed. the pressure decreases. it has to be recognized that the critical point is when M = 1 it will show a change in the trend and it is singular point by itself. the pressure has to be a monotonic function which means that flow cannot crosses over the point of 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. U Temperature. This situation is called choked flow. the explanation is based on the equations developed earlier and there is no known explanation that is based on the physics. First. The Trends The trends or whether the variables are increasing or decreasing can be observed from looking at the equation developed.

Equations (9.26) yields dP P dM 2 M2 =− 1 + (k − 1M 2 dM 2 2M 2 1 + k−1 M 2 2 (9. and (9. Thus.24) in equation (9.44) .39) can be integrated to yield: 1 P = ∗ P M k+1 2 k−1 2 2 M 1+ (9.23) is represented by 4f L .43) The stagnation pressure decreases and can be expressed by (1+ 1−k M 2 ) k−1 2 P0 = P0 ∗ P0 P P0 ∗ P∗ 2 ( k+1 ) k−1 k k P P∗ (9.27).24) can eliminate term 4f L and describe the D pressure on the Mach number. For example.38) ¯ It is common to replace the f with f which is adopted in this book. (9. (9.40) In the same fashion the variables ratio can be obtained k+1 c2 T 2 = ∗2 = T∗ c 1 + k−1 M 2 2 (9.162 CHAPTER 9. (9. FANNO FLOW By utilizing the mean average theorem equation (9. Dividing equation (9.41) ρ 1 = ρ∗ M 1+ k−1 2 2 M k+1 2 (9.29).36) yields k+1 2 ¯ 1 1 − M2 k+1 4fLmax 2 M = + ln D k M2 2k 1 + k−1 M 2 2 (9.28).24).29). the pressure as written in equation (9. D and Mach number. Now equation (9.42) U = U∗ ρ ρ∗ −1 =M 1+ k+1 2 k−1 2 2 M (9. M = 1 when P = P ∗ Equation (9. (9.39) The symbol “*” denotes the state when the flow is choked and Mach number is equal to 1.30) can be solved.

D The results of these equations are plotted in Figure (9.46) The integration of equation (9.1 0.34) yields s − s∗ = ln M 2 cp k+1 1 + k−1 M 2 2 k+1 k 2M 2 (9. THE WORKING EQUATIONS 163 Using the pressure ratio in equation (9.45) And further rearranging equation (9.2: Various parameters in Fanno flow as a function of Mach number many cases shockless and therefore a relationship between two points should be .5.1 Fri Sep 24 13:42:37 2004 0.45) provides 1 P0 ∗ = P0 M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (9.01 1 Mach number 10 Fig.9.47) discussion about Reynolds number and dimensionless 4f L friction parameter. 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 0.40) and substituting it into equation (9.2) The Fanno flow is in P/P .44) yields P0 = P0 ∗ 1+ k−1 2 2 M k+1 2 k k−1 1 M 1+ k−1 2 2 M k+1 2 (9. 9.

There is no Mach meter. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.01451 1. The air exits to the atmosphere. 9.50) 9. In most times. FANNO FLOW derived. 4f Lmax D = 2 4f Lmax D 1 − 4f L D (9.164 CHAPTER 9.94 .9146 1. the total temperature at the entrance can be obtained by knowing the 4f L . Ç Æw•DÃÁ Å Ä À ¿¾ SDH½ 0.90000 0.1291 4 This 1. finding the total pressure and temperature at the pipe inlet is the solution. the temperature and the total pressure at the pipe inlet are the same as those in the reservoir.1: Air flows from a reservoir and enters a uniform pipe with a diameter of 0.1) pipe inlet is essentially isentropic.9) the following is D obtained. the flow to the pipe inlet.49) Hence.0089 1. With the Mach number and temperature known at the exit. Thus.3: Schematic of Example (9.004 and that the flow from the reservoir up to the Fig. The real ratio can be obtained by two star ratios as an example T2 = T1 T T ∗ M2 T T ∗ M1 (9. S OLUTION For isentropic. the “star” values are imaginary values that represent the value at choking. Assume that the average friction factor to be f = 0. The following conditions prevail at the exit: P2 = 1[bar] temperature T2 = 27◦ C M2 = 0.0327 .0934 property is given only for academic purposes. Estimate the total temperature and total pressure in the reservoir under the Fanno flow model.6 Examples of Fanno Flow é çæä â §è ©Ñåãá ¼ º¹ ¸¶ µ §» ©s¶ %·€´ à %·e£Û ßÞ ÝÜ ÓÒÐ ÎÌ Ë É shÑÏ hÍeÊÈ Ú ss·eÔ ÙØ× Ö Õ Example 9.05 [m] and length of 10 [m].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. For given Mach number (M = 0.

2100 3.93840 1.97489 0.1) by interpolations or by utilizing the attached program.5764 0.0327 4f L D To ”move” to the other side of the tube the 4f L D is added as 3.014 × = 2. the total temperature at the exit is T ∗ |2 = T∗ T T2 = 2 165 300 = 290.35886 3.004 × 10 + 0.0327 0.91484 1.9.01451 0.12913 0.21 either from Table 1 = 4f L D + 4f L D 2 = 4 × 0.5[K] 1.0140 1.35886 0.7405 2.1699 Note that the subsonic branch is chosen.35886 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.5922 0. The stagnation ratios has to be added for M = 0. EXAMPLES OF FANNO FLOW So.05 4f L D The rest of the parameters can be obtained with the new (9.915 T1 T ∗ T01 = T2 T T0 ∗ 2 T 1 T 1 1 1 =300 × × 1.78305 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 × × 3.7405 0.6.975 T∗ T 348K = 75◦ C Another academic question: .91[Bar] 1.38814 1.17 × 1. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.

T01 = 400K a P01 = 29. R = 287 [J/kgK] and f = 0. then the following can be found eith from Table (9.0 [m] long.65326 T1 = T1 T01 = 0. FANNO FLOW shock d-c nozzle Fig.11528 0. Hence.4. determine: (a) the mass flow rate through the system. The system is supplied by a vessel. The vessel conditions are at 29. Thus. A normal shock wave occurs in the tube and the flow discharges to the atmosphere. With the isentro relationship the Mach number at point one (1) is known. therefore the velocity the vessel can be assumed to be small enough so it can be neglected. With these conditions a pipe inlet Mach number is 3.8K T01 " ò ð ï î í ì #ë ! •ñ wyrì w§ãê  ) & 0( ' $% atmosphere conditions     5 3 2 44!1  ¦¦ ÿ wý ü wrù w÷ P©ó þ ûú ø ö õô  §§ ¥ £ ¡ ¨©¨¦¤¢  . It is further assumed t the flow through the nozzle can be approximated as isentropic.4: The schematic of Example (9.0. and CHAPTER 9. 400 K. m = ρAM c ˙ The density and speed of sound are unknowns and need to be computed.166 Example 9.2) (c) determine the Mach number when a normal shock wave occurs [Mx ]. Take k = 1.35714 0.2346 0.357 × 400 = 142.0000 The temperature is 0. the stagnat conditions can be approximated for the condition in the tank.1) or the Potto–GDC T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ M 3.02722 0. S OLUTION (a) Assuming that the pressure vessel is very much larger than the pipe. 9.2: A system is composed of a convergentdivergent nozzle followed by a tube with length of 2.65[P ar] The mass flow rate through the system is constant and for simplicity point 1 is chosen which.07623 4.5 [cm] in diameter and 1.005.65 [Bar]. (b) the temperature at the pipe exit.

8 1.9.21822 4.3333 0. the shock at the entrance of the tube. (Fanno flow Table (9.8 0. Two possibilities are needed to be checked.2346 0.1 × 104 287 × 142.8571 10.9640 0. if a shock exists.005 × 1.52216 0.32834 After shock wave the flow is subsonic with “M1 ”= 0. the speed of sound can be calculated as √ √ c1 = kRT = 1. The exit pressure determines the location of the shock. one.6.54 = 0. and two.52216 the flow is choked and with a shock wave.8 > 0.47519 Ty Tx ρy ρx Py Px P0 y P0 x 2.97 × ˙ kg π × 0.8 239.025 Since 0. to obtain the following M 3.1)) .50918 1.027 × 30 P01 0. by comparing “possible” Pexit to PB .1)) Mx 3. From the Table (9.0 = 0.6790 3. check whether the flow is shockless by comparing the flow resistance and the maximum possible resistance.69 4 sec 8.0000 My 0.54[m/sec] The pressure at point 1 can be calculated as P1 = P1 P01 = 0.0252 × 3 × 239.81[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. shock at the exit and comparing the pressure ratios.97 kg m3 (b) First.0000 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.2) m = 1.42857 and the conditions of the tube are 4f L D = 4 × 0.1) or by using the Potto–GDC.47519. the possibility that the shock wave occurs immediately at the entrance for which the ratio for Mx are (shock wave Table (5. EXAMPLES OF FANNO FLOW Using the temperature. First.4 × 287 × 142.

50917 1.65 = 0.2549 0. To find the location of the shock ra of the pressure ratio. 3) According your root finding algorithm5 calculate or guess the shock location and th compute as above the new M1 . but be-careful second order methods like Newton-Rapson method can be unstable. 5 You can use any method you which. “claiming” upstream from P1 exit through shock to the entrance.47519 0.8568 × 0.168 M 4fL D P P∗ P0 P0 ∗ CHAPTER 9.3904 0. Then either by utilizing shock table or the program. With the location of shock.9640 0.2919 2.89545 1.0338 is smaller than the case in which shock occurs the entrance. 2) Calculate the entrance Mach assuming shock at the entrance.2549 1. a) set M2 = 1 b) add 4f L and calculated M1 ’ for subsonic branch D c) calculated Mx for M1 ’ Note this Mach number is the low Value. For example.95679 0. FANNO FLOW ρ ρ∗ U U∗ T T∗ 0. to obt D the upstream Mach number. . calculate the parameters for shock locat with known 4f L in the “y” side.32834 × 1 2. One possible way to find the e temperature.12476 The actual pressure ratio 1/29.1481 The stagnation values for M = 0.65326 The ratio of exit pressure to the chamber total pressure is 1 1 ∗ P2 = P0 = = P2 P∗ P0 y P P0x P1 P1 P0y P0x P0 1 1× × 0. the shock is somewhere downstream.3904 1.47519 1. P2 is needed. T2 is by finding the location of the shock. The procedure for the calculations: 1) Calculate the entrance Mach number assuming the shock occurs at the exit: a) set M2 = 1 assume the flow in the entire tube is supersonic: b) calculated M1 Note this Mach number is the high Value.47519 are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0. Thus.1912 0.85676 1. To check Secant Method.

57910 4f L D 0.0000 4fL D up 4fL D down Mx 1.7 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. These changes include the choking point at lower Mach number. because the very limited amount of heat transformed it is closer to an adiabatic flow. 9 If you have experiments demonstrating this point. . repeat previous stage until the solution is obtained. SUPERSONIC BRANCH 169 a) set M2 = 1 b) for the new 4f L and compute the new My ’ for the subsonic branch D c) calculated Mx ’ for the My ’ d) Add the leftover of 4f L and calculated the M1 D 4) guess new location for the shock according to your finding root procedure and according to the result.larc. The information from the wall to the tube center6 is slower in reality.8 Maximum length for the supersonic flow To insert example on the change in the flow rate between isothermal flow to Fanno Flow. 7 See on the web http://naca.22019 0.0000 M2 1. the comparison with reality shows that heat transfer cause changes to the flow and they need to be expected.9.). experiments from many starting with 1938 work by Frossel7 has shown that the error is not significant. up that will 9. However. The only limitation of the model is its uniform velocity (assuming parabolic flow for laminar and different profile for turbulent flow. In the process Mx and My must be calculated (see the chapter on the program with its algorithms. furthermore. on the comparison of the maximum length of isothermal model and Fanno Model. 9. and. M1 3. It has to be noted and recognized that as opposed to subsonic branch the supersonic branch has a limited length.). This solicitation is about how to explain this issue to non-engineers or engineer without a proper background.57981 (c) The way of the numerical procedure for solving this problem is by finding produce M1 = 3. The author seeks for a nice explanation of this concept for non–fluid mechanics engineers.gov/digidoc/report/tm/44/NACA-TM-844. Insert also example on percentage of heat transfer. 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.7.PDF 8 Many in the industry have difficulties in understanding this concept. In the Fanno model there is no heat transfer.nasa. please provide to the undersign so they can be added to this book.9899 My 0. Nevertheless. Many of the pictures in the literature carry copyright statements. 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 .

1 1 0.5).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. Even the flow in the tube starts as a supersonic in parts of the tube can be transformed into the subsonic 4fL maximum length. The maximum length in supersonic flow In Fanno Flow 1.1 0 1.5 1.170 as follows: CHAPTER 9.5 0.2 1.3 1.8215 The maximum length of the supersonic flow is limited by the above number.5 1.8 0.5: The maximum length as a function of specific heat.4) = 0.35 1. From the above analysis.4 0. 9. k = 1.55 1.3 0.9 0.65 spesific heat. k 9.4 1.3 1. k Thu Mar 3 16:24:00 2005 Fig. max D . 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 → ∞. k as shown in Figure (9. it can be observed that no matter how high the entrance Mach number will be the tube length is limited and depends only on specific heat ratio.2 0.7 0.6 0.6 1.2 1.45 1.25 1.4 1.

WORKING CONDITIONS 171 branch.7).9. Only a combination of these two parameters is truly independent. M1 .9. the mass flow rate decreases. 9. First.9. In this process. However.6: The effects of increase of on the Fanno line In the analysis of this effect. increasing the tube length results in increasing the exit Mach number (normally denoted herein as M2 ).1 Variations of The Tube Length ( 4f L ) Effects D † ¨… @ † † ‡ U T V8S W ƒ RV8fxy ‚  € 4f L D Fig. The discussion has to differentiate between two ways of feeding the tube: converging nozzle or a converging-diverging nozzle. The „ Q RPIGECA D H F D B wx x v `u ˆ s s s i ¨r pVg h t q e fcd ac b`X Y c 9 7 86 . Subsonic branch For converging nozzle feeding. it should be assumed that back pressure is constant and/or low as possible as needed to maintain a choked flow. no further increase of the exit Mach number can be achieved. 9. A shock wave can occur and some portions of the tube will be in a subsonic flow pattern. It is worth noting that entrance Mach number is reduced (as some might explain it to reduce the flow rate). Three parameters. the dimensionless friction. all the three parameters can be varied and they are discussed separately here. P2 /P1 are D controlling the flow. 4f L . the entrance Mach number. Once the Mach number reaches maximum (M = 1). and the pressure ratio. the treatment of the two branches are separated. The entrance temperature increases as can be seen from Figure (9.

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

At the end of the D range a.9. M1 . the entrance Mach number. it is a more practical point. However. 4f L . 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. The total maximum available for supersonic flow b − −b . The mass flow rate continues to be constant.8: The Mach numbers at entrance and exit of tube and mass flow rate for Fanno Flow as a function of the 4f L D The 0-a range. M1 is constant because it is a function of the nozzle design only. In the range b − c. is only a D 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). In the range of a − b the flow is all supersonic. the mass flow rate is constant because the flow is choked at the nozzle. 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→∞ ∞ a all supersonic flow mixed supersonic with subsonic flow with a shock between Fig. Notice that exit Mach number. 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. M2 decreases (remember this flow is on the supersonic branch) and starts ( 4f L = 0) as M2 = M1 . D It is worth noticing that in the a − −c the mass flow rate nozzle entrance velocity max ~ y} g f h!e i — • ˜–” x yw v t r p n qusqomk l j { Cz | ™ d b c the nozzle is still choked . The entrance Mach continues to be constant and exit Mach number is constant. reduces with the increase of 4f L . The exit Mach number.9. 9. The entrance Mach number. M2 is still one. M2 = 1.

M2 mass flow rate as a function of 4f L . the resistance increases and on the other hand.4 0. 1 1 . 9.4 0. This situation is because of the “double” choked condition in the nozzle.7 0.1 D = 1. 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. Figures (9.0 = 10.8 0.1 0 0 0. this situation is rather strange to explain.6 0.9: M1 as a function M2 for various 4f L D 10 On a personal note.9 0.7 Entrace Mach number 0.1 0.174 CHAPTER 9.0 = 100.6 0.5 0. ˙ To summarize the above discussion. the flow rate decreases since (m ∝ M1 )11 .5 0.2 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. Somewhat different then the subsonic D branch the mass flow rate is constant even if the flow in the tube is completely subsonic. D 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.8 0. On one hand.3 0.3 0. M1 as a function of M2 4fL  = 0. the exit Mach number remains constant and equal to one.0 Fanno Flow 1 0. The exit Mach M2 is a continuous monotonic function that decreases with 4f L .8) exhibits the development of M1 .9 1 Exit Mach number Tue Oct 19 09:56:15 2004 Fig.2 0. In this range.

the larger 4f L creates D larger differences between exit Mach number for the different shock locations. 9.5 4 3.1 and D = 0. and as can be observed.4 Tue Jan 4 11:26:19 2005 Fig.8 2 M1 M1 as a function of M2 for the subsonic brench 4fL  = 0.4 = 0. Obviously. shockless flow. shock at the entrance. For D = 0.2 = 0.1 D = 0.4 0. D For a given 4f L .4 1.2) by obtaining the 4f Lmax for M2 and subtracting D the given 4f L and finding the corresponding M1 . Subsonic velocity occurs for supersonic entrance velocity. Obviously there can be two extreme possibilities for the subsonic exit branch. below the maximum critical length.1 shock = 0. The larger 4f L larger M1 must occurs even for shock at the entrance.10) only for 4f L 4f L 4f L D = 0. Below.2 0.9. and three.2 shown with only shock at the exit only.5 1 0.4 two extremes are shown.5 2 1. In Figure (9.6 1. The Figure (9.10: M1 as a function M2 for different 4f L D for supersonic entrance velocity.5 0 0 0.8 1 M2 1.5 3 2. One. tow.9) exhibits the M1 as a function of M2 . at the tube entrance. The Figure was calculated by utilizing the data from Figure (9.10) exhibits the entrance Mach number as a function of the M2 . the supersonic entrance flow D has three different regimes which depends on the back pressure. one.6 0. shock at the exit.9.2 1. the maximum . when the shock wave occurs at the tube exit and two. D Fanno Flow 5 4. WORKING CONDITIONS 175 Figure (9.

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

When the back pressure is below point c. Up to this stage the nozzle feeding the tube increases the mass flow rate (with decreasing back pressure). M2 < 1 and is not 1. A back pressure that is at the critical point c results in a shock wave that is at the exit. 13 It is common misconception that the back pressure has to be at point d. Between point a and point b the shock is in the nozzle. The back pressure below point c has some adjustment as it occurs with exceptions of point d.12) shows different pressure profiles for different back pressures. ‰ ˆ Šq‡ ‹ ‚ ƒ Ž ƒŒ  †Œ 4f L D for a short 4f L D .11: The pressure distribution as a function of Short 4f L D Figure (9. the tube is “clean” of any shock13 . 9. 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. Note however that exit Mach number. In this range and further reduction of the pressure the mass flow rate is constant no matter how low the back pressure is reduced. WORKING CONDITIONS … †„ critical Point c critical Point d 177 € 8 fully subsoinic flow criticalPoint b a shock in the nozzle critical Point a Fig.9.9.

M1 and the “extra” 4f L . point c in this case is D different from point c at the case of short tube 4f L < 4f Lmax . The is no analytical solution for the location of this point c. the effect of changing the throat area on the nozzle efficiency is neglected.178  8 criticalPoint b CHAPTER 9. However. is a function entrance Mach number. 9. Further reduction of the back pressure at this stage will not “move” the shock wave downstream the nozzle. This dissection deals only with the flow when it reaches the { for a long Î ÏmÌ Í Ë Ê É ÆÇÈ ¦0·³Å 4f L D ¼ mº »½ ¹ ¸ ¶ ² Šq·´µ³±¯s¬¢©#¦¥f¤ ¢0sžœ ° ® ­ « ª ¨ § œ ¤ Ÿ £ ¡   Ÿ  à Á ¿ ÄÂÀ¾ æ çmä å ã â á Ýßà Û Ñ Ô Ô Ù × ¦0·ÞÜ4× ¤Ú ·4ØsÖÐ sÓ8Ð Ð Õ Ô Ò Ñ — – Šq• ˜ ’ ƒ‘ › ƒ™ š †™ fully subsoinic flow . 9. effects In this discussion. M1 . The D procedure is (will be) presented in later stage.12: The pressure distribution as a function of 4f L D Long 4f L D In the case of 4f L > 4f Lmax reduction of the back pressure results in the same D D process as explained in the short 4f L up to point c.3 Entrance Mach number. FANNO FLOW ” †“ critical Point c a shock in the nozzle critical Point a Fig. In reality these effects have significance and needs to be accounted for some instances. In this point the exit D D Mach number is equal to 1 and the flow is double shock.9. At point c or location of the shock wave.

D As was shown before.25 Tue Jan 4 12:11:20 2005 Fig. M1 is a function of the ratio of the nozzle’s throat area to the nozzle exit area and its efficiency.2 0. and D second the maximum depends on the entrance Mach number.4 0.4 Mach Number 1. M1 . D D Obviously.6 0. This analysis deals with the case where 4f L is shorter than total 4f Lmax .9.1 shock at 75% 50% 5% 0. Practically.6 1. the nozzle area ratio is changed by changing the throat area.15 4fL  D 0. first is the total D maximum 4f L of the supersonic which depends only on the specific heat. there are two different maximums for 4f L .13: The effects of pressure variations on Mach number profile as a function of when the total resistance 4f L = 0.3 for Fanno Flow D 4f L D The entrance Mach number. supersonic branch reached otherwise the flow is subsonic with regular effects.9. WORKING CONDITIONS 179 Mach number in Fanno Flow 4fL  D 2 1. 4f L D is equal to 4f Lmax D as a The process of decreasing the converging–diverging nozzle’s throat increases the . It is P2 assumed that in this discussion that the pressure ratio P1 is large enough to create a choked flow and 4f L is small enough to allow it to happen. the critical point is where result in the entrance Mach number.2 0 0 0.8 1. 9.8 0. k. This effect is the third parameter discussed here. in this situation.2 1 0.05 0.

6 3.4 0 0 0.4 2 1. FANNO FLOW P2/P1 Fanno Flow 4fL  D 4. 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.8 4. The reference to the tube is because it is the focus of the study. the throat area decreases. If the tube contains no supersonic flow then reducing the nozzle throat area wouldn’t increase the entrance Mach number.4 4 3. 9.6 1. The “payment” for increase in the supersonic length is by reducing the mass flow.3 entrance14 Mach number.2 0.2 P2/P1 2.8 2. By doing so.25 5% 50 % 75 % Fri Nov 12 04:07:34 2004 Fig.180 CHAPTER 9.8 0.14: Fanno Flow Mach number as a function of 4f L D when the total 4f L D = 0. In the case of large tube 4f L > 4f Lmax the exit Mach number increases with the D D 14 The word “entrance” referred to the tube and not to the nozzle.05 0. Further.15 4fL  D 0. D . 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 area) decreases the flow rate while in a different way (increasing the 4f L ) does not affect the flow rate. Decreasing the nozzle throat area moves the shock location downstream.2 0. The mass flow rate is proportionally linear to the throat area and therefore the mass flow rate reduces. decrease of the throat area results in flushing the shock out of the tube.1 0. The process of decreasing the throat area also results in increasing the pressure drop of the nozzle (larger resistance in the nozzle15 )16 .

is to find the furtherest shock location downstream. 9. 0 Fig.16) shows the possible ∆ 4f L as function of retreat of the location of the shock wave from the D maximum location.16: The extra tube length as a function of the shock location. WORKING CONDITIONS 181 shock Fig. 4f L D The maximum location of the shock The main point in this discussion however.15: Schematic of a “long” tube in supersonic branch decrease of the throat area. 9. the location of the shock wave approaches to the theoretical location if entrance Mach. Figure (9. if the ë `êè é ’ ‘  “…‰ ‡ † „ ˆ…ƒ #     !" § t ursq p ppp i ppp B@f h g 3 1 3 2672054120 ) ))) ( ))) '$ & % ú ûù ÷ øö H F H GR7QEPIFGE D DDD C DDD B@8 A 9 ¨©§¤¥£ ¢¢¢ ¦ ¢ ¢¢¢ ÿ ý þü ¡ õ õ ô ó ò ð !¦V!ñïêì î í e a `b X TVW cd B@YUS  y w ‚€xv supersonic branch . When the entrance Mach number is infinity.9. Once the exit Mach number reaches one no further increases is possible. M1 = ∞. However.9. M1 = ∞.

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 ˜ — • s–” gf ed 2€Q‚™ 1 4f L D super- (9. 4f L D max∞ Fig.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).17: The maximum entrance Mach number.51) . 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. then shock at Mx = 1 results in My = 1. ii) Calculate the extra (at the entrance). i) Calculate the extra the max length). M1 to the tube as a function of sonic branch From numerical point of view. The proposed procedure is based on Figure (9. FANNO FLOW shock location is at the maximum length. To overcome this numerical problem it is suggested to start the calculation from distance from the right hand side.182 CHAPTER 9. 9. Let denote ∆ Note that 4f L D sup 4f L D ¯ = 4f L actual − D 4f L D max∞ 4f L D sup is smaller than 4f L D retreat .

821508116. Example 9.7461 1.15). The maximum 4f Lmax for k = 1. (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. 4f L D up = − 4f L D max 4f L D up . The maximum for M1 = 8 is 4f L = 0. .0000 With (M1 ) M 0.6136 2.17) shows the entrance Mach number. 4f L D = 0. The retreating length is expressed as subsonic but 4f L D retreat = 4f L D max∞ − 4f L D sup (9.4 (Mexit = 1).76820 (flow is choked and no additional 4f L ).42390 1. The left side is when the shock occurs at My 0.3: Calculate the shock location for entrance Mach number M1 = 8 and for assume that k = 1.00849 2.76820.442 − 0.3591 0.4 D is 0.9. Or in a very long tube the whole flow will be subsonic. 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 Mx 8. 4f L exceed the maximum length 4f L for this entrance D D Mach number.3102 Now the solution is somewhere exceeds the maximum 4f L D 4f L D max for the supersonic flow.1318. Hence. M1 reduces after the maximum length is exceeded. thus the extra tube D is ∆ 4f L D = 0.9. Hence.4417 2.76820 = 0.5652 74.1641 between the negative of left side to the positive of the right side17 .5000 0. Accordingly continue. the value of left side D is −0.9 S OLUTION The solution is obtained by an iterative process. WORKING CONDITIONS 183 flow is supersonic see Figure (9.1318 = 2. In a summary of the actions is done by the following algorithm: (a) check if the (b) Guess 4f L D is 2.3867 5.52) Figure (9.1318.39289 The extra ∆ 4f L D 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 4f L D = 0.9 − 0.39289 Ty Tx ρy ρx Py Px P0y P0 x 13.

32932 0.90000 0.90000 0.62170 0.90000 This procedure rapidly converted to the solution.57069 0.32921 0.6706 1.57073 0.64834 0.32927 0.90000 0.57093 0.6673 1.27830 0.32932 0. FANNO FLOW (d) Calculate the associate Mach number.64830 0.31783 0.32907 0.6635 1.22574 0.6021 1.57184 0.57605 0.6699 1.65023 0.64832 0.32816 0.5286 1.64831 0.90000 0.90000 0.64830 0.64831 0.67426 0.57068 0.90000 0.90000 0.32878 0.64830 0.90000 0.57318 0.32932 0.0000 M2 1.65728 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 1.6706 1.0000 4f L D down against the old one.90000 0. Shock location are: M1 8.6691 1.90000 0.32932 0.57068 0.32395 0.64839 0.64920 0.32932 0.66779 0.32931 0.6706 1. If it is satisfactory stop or retu 4fL D up 4fL D down Mx 1.90000 0. Mx with the Mach number. 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).6706 1.57068 0.6706 1.90000 0.90000 0.64872 0.58217 0.184 CHAPTER 9.32930 0.57068 0.59506 0.6705 1.64830 0.64850 0.90000 0.6382 1.69119 0.6706 My 0.6554 1.6703 1.90000 0.3838 1.57068 0.64830 4fL D 0.6706 My 0.90000 0.32682 0.57079 0.65246 0.57070 0.30494 0.6706 1. .74664 0.90000 0.57122 0.57068 0.

1 iso = 0. The only difference is in small dimensionless friction.10.99 which requires that enough for M1 < 0.18: The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow As it can be seen for the Figure (9. One must notice that the maximum temperature at the entrance is T0 1 .2 0. When the Mach number decreases the temperature approaches the stagnation temperature (T → T0 ).8 M1 Fanno flow 0. This tolerance requires that (T0 − T )/T0 = 0. The results D are very similar for isothermal flow. 4f L . Hence. if one allows certain deviation of temperature.18) the dominating parameter is 4f L . This requirement provides that somewhere (depend) in the vicinity of 4f L = 25 the D flow can be assumed isothermal. D .8 iso = 0. Looking at the table or Figure (9.4 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.67.15 even for large k = 1.1 0 0 10 20 30 40 Wed Mar 9 11:38:27 2005 50 4fL  D 60 70 80 90 100 Fig.10 The Approximation of the Fanno flow by Isothermal Flow The isothermal flow model has an equation that theoreticians find it easier to use compared to the Fanno flow model.3 M1 0. THE APPROXIMATION OF THE FANNO FLOW BY ISOTHERMAL FLOW185 9.2 = 0. say about 1% that flow can be assumed to be isothermal.5 = 0.2) or the results from computer program attached to this book shows that reduction of the mass flow is very rapid.9. 9.1 = 0. Hence the mass flow rate is a function of 4f L D because M1 changes. to insert a question or example about this issue in end with comperison to Isothermal Flow 0.

18) consider the following example.99556 4.12420 0.11637 0.99741 0.12728 0.0000 40. T ≈ T0 and is the same for the pressure.2519 A×P A∗ ×P0 Therefore. FANNO FLOW 9.22697 0. Hence.0000 40.0000 42.18) for P2 /P1 = 0. the mass rate is a function of the Mach number.0000 40.09965 4fL D 4fL D 1 4fL D 2 P2 P1 40. The stagnation conditions at the entrance are 300K and 3[bar] air. = 4 × 0.5393 4.99097 0.11392 0.99692 0.42 0.5.7027 5. 4f L D 4f L D .40790 0. 0.7569 67.98928 0.12420 0.1697 10. M1 0.0000 40.11 More Examples of Fanno Flow Example 9.4: To demonstrate the utility in Figure (9.99233 0.0733 7. L = 4[m].1 the flow is choked.07975 M2 1.30000 0.0 2.1697 50.3.5910 4.99873 0. Find the mass flow rate for f = 0.98874 0. ρ ρ0 M 0. D = 0.2842 0.186 CHAPTER 9.99683 4.99354 0.99677 0.50000 0.7569 107.05 × 4 = 40 0.0000 0.1.99195 0.6523 5.022 k = 300000 × × 0.48 287300 kg sec .02[m] and pressure ratio P2 /P1 = 0.127 × RT 4 1.12728 0.1196 7.07975 T T0 A A P P0 0. or accurately by utilizing the program as in the following table. 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.02 From Figure (9.4206 0.11392 0.8.80000 Only for the pressure ratio of 0.05. 0.4 ≈ 0.1 M1 ≈ 0.13 etc. 0. S OLUTION First calculate the dimensionless resistance. The mass flow rate is m = P AM ˙ π × 0.

06570 0.2620 15.6958 1.0934 1.70 0.87037 0.1291 1.09851 0.4487 2.5333 8.30 m ˙ P1 0.1628 1.3546 2.05 0.95781 1.1996 1.2200 13.1273 0.0351 3.2915 13.7092 1.0929 1.0382 1.1429 1.3848 1.9034 11.53452 0.10 0.4182 10.4935 1.3085 1.4525 2.1998 1.1273 kg sec kg sec kg sec 187 9.1985 1.72805 0.1533 1.1315 1. THE TABLE FOR FANNO FLOW and for the rest m ˙ 0.00 1.00000 1.468 P1 0.5664 1.52216 36.1713 1.90 0.5826 3.05476 0.12.00 2.48 × = 0.3 ∼ 0.40825 1.72 83.2549 1.27217 0.6742 3.000 0.00 3.0638 1.4961 66.8708 1.0624 1.2993 3.3005 27.20814 0.12 The Table for Fanno Flow Table 9.32459 0.2047 1.964 1.85 0.8 ∼ 0.43 P1 0.6843 7.08758 0.03633 0.3665 1.1988 1.3184 2.03 0.66 106.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 787.061 1.0702 2.21822 0.044 1.1242 P2 = 0.1139 P2 m ˙ = 0.61237 0.1065 1.1852 1.3398 1.633 1.0207 1.0787 1.07975 P2 = 0.65 0.5116 19.68374 0.07664 0.1976 1.48326 0.0 0.58506 0.48 × = 0.002 1.0327 1.4613 10.6191 2.75 0.1273 0.8218 5.000 0.60 0.1: Fanno Flow Standard basic Table M 0.3817 14.01451 0.50918 0.2508 9.95 1.04381 0.4554 2.0693 1.00328 0.07 0.4815 21.5901 2.1378 4.5753 1.82514 0.8254 18.4318 22.03286 0.6400 2.6183 1.09 0.0474 11.2119 1.235 30.77894 0.4626 1.9635 4.2838 1.37879 0.30500 0.63481 0.1991 1.0944 1.1882 1.66667 0.6659 15.06 0.1618 6.12728 0.73179 0.35 280.07229 0.5 ∼ 0.42857 .25 0.43133 0.4027 3.0922 1.20 0.3865 1.1381 1.49082 0.4834 5.1788 1.9216 14.5914 18.02 193.10944 0.1905 1.2616 12.08 440.03 140.017 1.32572 0.1194 1.7634 1.45 0.1994 1.2893 1.55 0.9435 5.48 × = 0.0089 1.08 0.6416 8.50 0.80 0.40 0.35 0.1356 1.30 0.1489 1.9.04 0.91460 0.688 0.21822 4.7780 2.1512 9.9341 1.000 0.0485 1.1981 1.0691 0.

41079 0.00 30.089443 25.63306 0.00200 2.443 2.00 55.20000 0.40889 0.82008 0.000579 5.000809 2.00 70.4E+6 0.43568 0.81860 0.047619 1.40866 0.000979 1.00663 0.3E+6 0.40846 2.00 20.72988 0.42066 0.086957 0.440 2.5E+5 0.9E+2 0.72 0.014815 0.6E+6 0.00166 0.78683 0.5E+6 0.00271 1. FANNO FLOW Table 9.00488 0.40853 0.40875 0.11111 0.377 2.81582 0.036860 1.138 2.4E+2 0.00390 4.057143 0.6E+5 0.00374 0.1: Fanno Flow Standard basic Table (continue) M 4.390 2.00 65.82078 0.00 60.00 25.063758 53.00142 0.00121 8.00122 .5E+4 0.00198 0.446 2.81265 0.8E+6 0.81928 0.00153 4.82066 0.42390 0.69380 0.447 2.41833 0.000680 3.13363 10.448 2.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.023905 5.434 2.00240 0.00 8.359 2.00 10.00 45.00 6.3E+2 0.82052 0.445 2.40938 0.029348 3.81755 0.00 7.40908 0.77899 0.6E+4 0.448 0.44721 0.76819 0.446 2.40849 0.447 2.14634 0.00 9.8E+5 0.81975 0.40859 0.000500 7.75280 0.0E+2 0.00609 1.00952 0.18 0.188 CHAPTER 9.00 5.333 2.00296 0.448 2.46771 0.28571 0.00 40.069767 0.82033 0.236 2.00 50.40988 0.00 35.295 2.00 0.42857 0.1E+5 0.

This fact create situation different compare to the previous two models. 10.1: The control volume of Rayleigh Flow fer and friction the main assumption was that relative length is so the heat transfer occurs between the surrounding and tube. Rayleigh flow is practical and useful concept in a obtaining trends and limits. This flow referred in the literature as Rayleigh Flow (see historical notes). a simple model is built around assumption of constant properties (poorer prediction to case were chemical reaction take palace). This model usage is to have a rough predict the conditions occur mostly in 189 x .1 Introduction v wu n qo rp m t @s l –k i jh The third simple model for an one dimenflow direction sional flow is for constant heat transfer for frictionless flow. Yet.Fig. The density and pressure change due to external cooling or heating. As opposed to the two previous models. In contrast. This 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. the heat transfer can be in two directions not like the friction (there is no negative friction). As before. 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 applied to case where the heat transfer is significant and the friction can be ignored.CHAPTER 10 RAYLEIGH FLOW Rayleigh flow is (frictionless) flow with heat transfer through a pipe of constant cross sectional area. 10. In practice Rayleigh flow is really provide good model for the real situation.

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

10.12)  Œ  ˆŽŒ ƒ ‚ „x z ‡ † ˆˆ… (10.10) Let the initial condition M1 .2. GOVERNING EQUATION Substituting equations (10.7) 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 s2 .11) results in 2(1 − M 2 ) 1 ds = Cp dM M (1 + kM 2 ) (10. 10.5) and (10. T2 k − 1 P2 s1 − s 2 = ln − ln Cp T1 k P1 s1 − s 2 1 + kM1 2 ) M2 k−1 1 + kM 212 = 2 ln ( + ln Cp k (1 + kM2 2 ) M1 1 + kM1 2 (10.6) yields T2 1 + kM1 2 M2 = T1 1 + kM2 2 M1 T2 T1 191 (10. A derivative of equation (10. and s1 are constant then the variable parameters are M2 .11) € ~x{ } | ‹ Š ˆˆ‰ y . The second maximum can be expressed as dT /ds = ∞ The second law is used to find the expression for derivative.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 (10.9) nt Pr es su re li ne co ns ta Fig.

and M2 results in dT 1 − kM 2 = constant × 3 dM (1 + kM 2 ) Combining equations (10. 1 + kM1 2 P∗ = P1 1+k The temperature ratio is 1 T∗ = 2 T1 M 1 + kM1 2 1+k 2 (10. But the entropy of supersonic flow is lower (see the Figure (10. increase of heating results in increase of entropy. dT /ds = ∞ when M = 1. . And cooling results in reduction of entropy. several observations can be generalized.16) U∗ ρ1 = = ∗ ρ U1 1 The √U kRT ∗ kRT ∗ √ √ U1 kRT1 kRT1 ∗ √ = 1 M1 T∗ T1 (10. chocking can be explained by the fact increase of energy must accompanied by increase of entropy. It is convent to referrers to the value of M = 1. yet note this point is not the choking point. These value referred as the “star”1 values. RAYLEIGH FLOW Take the derivative of the equation (10. The equation (10. dT /ds < 0.192 CHAPTER 10. The derivative is negative. The derivative is equal to infinity. The chocking is occurred only when M = 1 because it violate the second law.15) (10.). some what similarly to Fanno flow. The peculiars point of M = 1/ k when additional heat is applied the temperature is decreasing. From thermodynamics. when cooling applied to a tube the velocity decreases and heating applied the velocity √ increases.2)) and therefore it is not possible (the maximum entropy at M = 1. Hence.14) (10.12) and (10.13) by eliminating dM results in M (1 − kM 2 ) dT = constant × ds (1 − M 2 )(1 + kM 2 )2 (10. The derivative is equal to zero √ when 1−kM 2 = 0 or M = 1/ k or when M → 0. Yet. Yet for every curve.17) star is an asterisk.13) On T-s diagram a family of curves can be drawn for a given constant. The transition to supper sonic flow occurs when the area changes.12) when letting the variable parameters be T2 .5) can be written between chocking point and any point on the curve.

090909 0.014300 0.69751 0.70804 0.262 1.001 0.273 2.011 1.075 1.00765 0.266 1.264 1. Several observations can be made in regards to the stagnation temperature.07 0.10 0.157 1.266 1.4 M 0.029 1.596 1.259 1.85987 0.193 1.94009 0.65 0.40887 0.265 1.373 2.049 1.43894 0.023223 0.86204 0.961 1.00431 0.027841 0.30 0.870 1.75 0.778 1.043 1.75524 0.011680 0.45 0.20661 0.3.85 0.015 0.207 2.056020 0.63713 0.86833 0.25096 0.55 0.114 1.058 1.267 1.03 0.99290 1.508 1.686 1.235 1.019 1.40 0.34686 0.00383 0.261 1.30440 0.3 Rayleigh Flow Tables The “star” values are tabulated in Table (10.90850 0.25 0.69136 0.019222 0.397 2.99814 2.61515 0.094 1.81892 0.218 1.95693 .125 1.37865 0.020529 0.50 0.030215 0.20 0.00216 0.025 1.09 0.51413 0.343 1.00917 0.10.91097 0.44444 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 (10.75991 0.04 0.045616 0.00517 0.423 1.036212 0.35 0.19183 0.96395 0.95 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0. Table 10.00598 0.038075 0.98097 0.395 2.79012 0.388 2.131 2.57447 0.367 2.96081 0.60 0.80 0.014 1.81013 0.00860 0.17355 0.52903 0.060 1.392 2.13793 0.023669 0. RAYLEIGH FLOW TABLES 193 T1 1 + T0 1 ∗ = T0 T∗ 2 k−1 2 M1 1+k 2 = 2(1 + k)M1 2 (1 + kM 2 )2 1+ k−1 M1 2 2 (10.90 0.61393 0.178 1.08 0.017119 0.99207 0.1).1: Rayleigh Flow k=1.91670 0.384 2.51001 0.379 2.25684 0.025 1.06 0.135 1.70 0.266 1.011922 0.05 0.015224 0.199 1.19) 10.030 1.046777 0.31373 0.005 1.

20040 0.205 1.046693 0.4E+2 1.36364 0.31486 0.11111 0.69700 0.73954 0.50 18.5 2.005 1.711 1. RAYLEIGH FLOW Table 10.713 .52356 0.081772 0.561 1.91185 0.712 1.045 2.699 1.5 8.65398 0.99392 0.8 1.3 1.686 1.26478 0.1 2.050943 0.403 1.620 1.41 1.52438 0.46106 0.1 1.75250 0.222 2.61580 0.078 1.667 1.256 1.0E+2 1.95798 0.72421 0.066667 0.53633 0.58909 0.00 0.00 1.194 CHAPTER 10.13540 0.494 1.35561 0.52004 0.538 1.675 1.030094 0.21419 0.525 1.21417 0.039900 0.90928 0.7 2.5 1.034483 0.028972 0.43353 0.80539 0.588 1.85971 0.693 1.97872 0.00326 0.56982 0.4 (continue) M 1.49415 0.71301 0.020979 0.571 1.475 1.67380 0.00190 0.65377 0.33454 0.018846 0.60894 0.79339 0.122 1.0E+4 1.044910 0.47562 0.68 75.00140 1.37870 0.00 0.63 27.56734 0.5 4.89087 0.2 1.328 8.873 3.71006 0.550 1.1E+4 3.16831 0.634 2.4 2.33439 0.713 1.136 3.510 1.83628 0.0E+2 3.503 1.405 1.50702 0.039883 0.0 25.49356 0.00274 0.035650 0.0 20.22936 0.00 0.00732 0.9 2.0 3.77406 0.66350 0.743 1.240 1.698 1.418 2.30864 0.1: Rayleigh Flow k=1.0 35.6 2.000 1.695 1.580 1.24615 0.13223 0.5 6.81414 0.017021 0.49259 0.340 1.078 1.0 7.79576 0.3E+2 3.5 10.95 54.28028 0.57831 0.5 5.00469 0.51647 0.43122 0.701 1.0 9.0 5.55556 0.00428 0.702 1.8E+2 2.2 2.7 1.52970 0.431 1.032053 0.4 1.023495 0.9 3.28551 0.6 1.092719 0.00240 1.176 1.3 2.40384 0.50885 0.51098 0.681 1.641 1.375 1.21 38.51349 0.424 5.690 1.0 4.49174 0.64103 0.026490 0.078487 0.0 1.616 1.68494 0.227 12.886 2.0 30.49122 1.146 1.75613 0.5 7.70174 0.96031 0.055363 0.301 1.5 9.8E+2 1.058264 0.8 2.18788 0.316 1.0 8.10256 0.0 2.044 1.88419 0.0 6.656 1.52893 0.85917 0.93425 0.7E+5 1.17647 0.019 1.455 1.54473 0.067263 0.39643 0.2E+4 8.0 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 1.29687 0.

RAYLEIGH FLOW TABLES Table 10.000695 0.000406 0.49021 0.49050 0.000476 0.0 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0. 10.0E+5 1.4898 0 0.3).49066 0.714 The Data is presented in Figure (10.5E+6 3.714 1.1: Rayleigh Flow k=1.49028 0.0 65.714 1.714 1.8E+6 5.4E+5 6.714 1.10.3.49037 0.000567 0.0 70.00107 0.49015 0.00117 0.0E+6 1.49089 0.714 1.0 55.1 1 M 10 Fri May 6 11:11:44 2005 Fig.000846 0.6E+6 2.000816 0.000600 0.4) .0 60.0 50.000350 3.000971 0.5E+6 1.3: The basic functions of Rayleigh Flow (k=1.00145 0.714 1. Rayleigh Flow 4 k=14 T/T * T0 / T0 P/P * P0 / P0 ρ/ρ 2 ∗ * * 3 @( M = ∞ ) 1.00184 0.0 45.4 (continue) 195 M 40.000686 0.7143 1 @( M = ∞ ) 0.

RAYLEIGH FLOW 10. Calculated the exit Mach number. Due internal combustion heat was released and the exit temperature was found to be 127◦C.25000 0.2177 .0424 The total exit pressure.13793 T1 T2 400 T2 = ∗ = 0.3005 1.1: Air enters a pipe with pressure of 3[bar] and temperature of 27◦ C at Mach number of M = 0. From the initial values the ratio at the exit can be computed as following. P02 can be calculated as the following: isentropic P0 2 = P 1 P0 1 P1 P0 ∗ P0 2 1 1 × × 1.196 CHAPTER 10.18991 The exit Mach number is known. Example 10.901[Bar] P1 P ∗ 2. the total exit pressure.1992 = 3.98765 0.2177 0.95745 1. and heat released (transfered) to the air. M 0.30440 0.1341 = 2.40530 0.2069 For the entrance the stagnation values are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.4053 T∗ T T1 300 M 0. After what amount of kJ energy the exit temperature will start to decrease? Assume CP = 1.95745 2.4 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.08572[Bar] ∗ =3× P0 1 P0 0.1 or from the program the initial ratio can be calculated.25000 and T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.96942 2.304 × = 0.34376 2.2069 1.25.004 kg◦ C S OLUTION The entrance Mach number and the exit temperature are given and from the Table 10.4027 0. the exit pressure.1992 0. the exit pressure can be calculated as P2 = P 1 1 P ∗ P2 =3× × 2.1341 1.29831 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.25684 2.

62666 0.0286 0.78 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 0.0286 = 1013.75) = 103.95686 2.98765 = 303.4.97959 1.875 The maximum heat can be calculated as kJ Q = Cp (T0max − T01 ) = 1 × 1.94012 1. T01 isentropic T0 1 = T 1 T0 1 T1 = 300/0.0221 The stagnation temperature for this point is T0max = Tmax ∗ 0.98765 = 407.9229 0.75) = 858.10.75[K] The isentropic conditions at the exit are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0. EXAMPLES FOR RAYLEIGH FLOW 197 The heat release (heat transfer) can be calculated from obtaining the stagnation temperature form the both sides.71618 1. The stagnation temperature at the entrance.004 × (1158.51[K] Tmax 0.84515 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 1.0454 The exit stagnation temperature is isentropic 0.12[K] Q kJ = Cp (T02 − T01 ) 1 × 1.90103 T0 2 = T 2 The heat release becomes T0 2 T2 = 400/0.98251 0.87500 0.53376 T0max 1013.0116 0.3044 The isentropic relationship at the maximum are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.12 − 303.7 = = 1158.85714 The maximum heat before the temperature can be calculated as following: Tmax = T1 T ∗ Tmax 300 × 1.004 × (407. .18 m ˙ kgsecK Note that this point isn’t the choking point.64051 0.29831 0.7[K] T1 T ∗ 0.2000 1.84515 0.51 − 303.

RAYLEIGH FLOW Example 10.1) or from the Potto-GDC the following ratios can be obtained.0589 1. M 1.42789 2. Calculated the entrance temperature and the entrance Mach number.34398 T T∗ T0 T0 ∗ = 0.2: Heat is added to the air until the flow is choked in amount of 600 [kJ/kg]. From the Table (10.0000 0.83333 0.5016 either by Table (10.83333 It must be noted that T02 = T0 ∗ . 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.3: Air with Mach 3 enters a frictionless duct with heating.58463 = 584.50160 0.38454 and the entrance temperature can be calculated as following T1 = 1000 × 0. 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. Therefore with or by Potto-GDC the following is obtained M 0.52828 0.63394 1.1805 0. T0 Q = Cp (T02 − T01 ) = Cp T0 ∗ 1 − 1 m ˙ T0 ∗ The table for M = 3 as following .5016 =1− = 1200 − = T0 2 T0 2 C P 1200 × 1.0000 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.004 T0 1 T0 ∗ T0 2 1000 = = 1200. entrance Mach number is 0. M2 = 1.52828 0.198 CHAPTER 10.24362 Thus.0[K] T2 0.6[K] T∗ The difference between the supersonic branch to subsonic branch T1 = T ∗ Example 10. What is the maximum heat that can be add so there is no subsonic flow.52828 The stagnation temperature T0 2 = T 2 The entrance temperature is Q/m ˙ 600 T0 1 ∼ 0.1) P0 P0 ∗ ρ∗ ρ P P∗ 0. The exit temperature is 1000 [K].

3333 0.8235 1.1244 0. .75086 0.0000 My 0.65398 1.4245 1.32834 With Mach number of M = 0.47519 0.22844 M 0.41176 It also must be noticed that stagnation temperature remains constant across shock wave.0802 1.47519 the maximum heat transfer requires information from Rayleigh flow as following M 0.8571 10. EXAMPLES FOR RAYLEIGH FLOW M 3.5882 The higher the entrance stagnation temperature the larger the heat amount that can be absorbed by the flow.1857 0.65398 0.10.0000 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 199 0.47519 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.17647 3.6790 3. T0 Q 1 − T0 1 ∗ m ˙ 1 − 0.47519 Ty Tx ρy ρx Py Px P0y P0 x 2.65398 1 − T0 ∗ m ˙ supersonic supersonic It is not surprising since the the shock wave are found on the Rayleigh flow.65398 subsonic subsonic = = =1 T0 1 Q 1 − 0. In subsonic branch the Mach number is after the shock is Mx 3.40469 2.28028 0.33138 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.4.

200

CHAPTER 10. RAYLEIGH FLOW

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 . It Isothermal model for relatively long tube was the undersigned experience, 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 material aside to aeronautical spectrum even Isothermal model such as turbomachinery. 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 . There is a significant impor- Fig. 11.1: The two different classifications of tance to the “pure” models such Isothermodels that explain the filling or mal flow and Fanno flow which have imevacuating of a single chamber mediate applicability. However, in many instances, the situations, in life, are far Fanno model

1 After completion of these Chapters, the undersigned discover two text books which to include some material related to this topic. These books are OCR, J. A., Fundamentals of Gas Dynamics, International Textbook Co., Scranton, Pennsylvania, 1964. and “Compressible Fluid Flow,” 2nd Edition, by M. A. Saad, Prentice Hall, 1985. However, these books contained only limit discussions on the evacuation of chamber with attached nozzle. 2 Even if the instructor feels that their students are convinced about the importance of the compressible, this example can further strength and enhance this conviction.

201

¤£¢¡Ÿ  œ › 45I5 ž j•“š

˜–” ’ ™— ©•“‘

202

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

more complicate. 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. In the first Chapter models, were the chamber volume is controlled or a function of the pressure, are discussed. In the second Chapter, models, were the chamber’s volume is a function of external forces, are presented (see Figure (11.1)).

11.1

Governing Equations and Assumptions

The process of filing or evacuating a semi flexible (semi rigid) chamber through a tube is very common in engineering. For example, most car today equipped with an airbag. For instance, the models in this Chapter are suitable for study of the filling the airbag or filling bicycle with air. The analysis is extended to include a semi rigid tank. 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. 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. 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. As it was shown in Chapter (9) the resistance, 4f L , should be larger than 400. Yet Isothermal flow model is D used as the limiting case.
fanno model

Isothermal model for a relatively long tube

The connection is through a narrow passage

A schematic of a direct connection

Fig. 11.2: A schematic of two possible connections of the tube to a single chamber

The Rayleigh flow model requires that a constant heat transfer supplied either by chemical reactions or otherwise. This author isn’t familiar with situations in which Control volume for the evacuating case Rayleigh flow model is applicable. And therefore, at this stage, no discussion is offered here. Fanno flow model is the most appropriate in the case where the filling and Control volume for the filling case evacuating is relatively fast. In case the filling is relatively slow (long 4f L than the D Fig. 11.3: A schematic of the control volIsothermal flow is appropriate model. Yet umes used in this model as it was stated before, here Isothermal flow and Fanno flow are used as limiting or bounding cases for the real flow. Addition-

Ó @Ñ Ò

Æ ÇÈ @Å

Î ÌË Ê ÏÐ ÂÍ us'É

Á ¿¾ ½ ÃÄ ÂÀ us'¼

± ®­¬ ª ¨ ¦ ¯P°¯5w« o©§¥

» ¹¸· µ ´ ³ ¯Pº¯5w¶ o©§²

reduced connection

for relatively short tube

fanno model

1

2

1

for relatively short tube

2

Isothermal model for a relatively long tube

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 resistance. The first kind connection tied the resistance, 4f L , with the tube area. D 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 − mout = 0 ˙ dt (11.1)

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)

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

Ø
1 2

Ú ÛÙ

Ø

× Ô Õ BÖˆÔ ß Ý ©ÞÜ

(11.2)

204

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

neglected (only for filling process) the most general equation ideal gas (11.1) reads
mout ˙ m U

d dt

PV RT

± ρ1 A c1 M1 ( 4f L , D

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)

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

¯ P1 P (0) ¯ ¯ ¯ ± ¯ A kRT1 T (0)Mmax M(t) = 0 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 ¯¯ d PV ¯ ¯ dt T ± ¯ ¯ tc AMmax kRT (0) P1 M1 ¯ M =0 ¯ V (0) T1 (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 ¯¯ ¯ ¯ d PV P1 M =0 (11.9) ± ¯ ¯ ¯ dt T 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 ¯ ¯ ¯¯ tmax P1 M d PV ± =0 (11.10) ˜ ¯ ¯ tc dt T T
1

¯ ¯ ¯ ¯ ˜ ˜ when P , V , T , and M are all are function of t in this case. And where t = t/tmax . 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 denoted with subscript 1. The ratio of √ 1¯ is substituted by
T1

¯ ¯ k−1 2 P P1 1+ =√ M ¯ ¯ 2 T T1 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 ¯ ¯ ¯ ¯¯ d PV P M(t)f [M ] √ ± =0 (11.13) ¯ ¯ ¯ dt T 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.

These D statements are correct for both Fanno flow and the Isothermal flow models. once the flow became choked the Mach number. 11.17a) (11. PB /P1 .206 CHAPTER 11.17) is a general equation for evacuating or filling for isentropic process in the chamber.2 11. M1 is a function of the resistance.2.3 Isothermal Process in The Chamber A Note on the Entrance Mach number The value of Mach number. The exit pressure. The situations where the chamber undergoes isentropic process but the flow in the tube is Isothermal are limited. EVACUATING AND FILLING A SEMI RIGID CHAMBERS The ratios can be expressed in term of the reduced pressure as followed: ¯ ¯ P P ¯1 = k−1 = P k ¯ ¯ k T P and ¯ P ¯ k+1 √ = P 2k ¯ T So equation (11.17c) Equation (11. 4f L and the ratio D of pressure in the tank to the back pressure. the model in the tube could be either Fanno flow or Isothermal flow.2.17b) (11. The method outlined in Chapters 8 and 9 is appropriate for solving for entrance Mach number. P2 is different from PB in some situations. These equations are combinations of . M1 is only a function of the resistance. the application of this model provide some kind of a limit where to expect when some heat transfer occurs. As it was shown before. Nevertheless. 4f L . in this stage. 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.15) ¯ ¯ 1 ¯ 1−k dP ¯ ¯ 1 dV ± P k+1 M(t)f [M ] = 0 ¯ 2k ¯ ¯ P k V +Pk ¯ ¯ k dt dt ¯ ¯ ¯ dP ¯ dV ¯ 3k−1 ¯ ¯ V ¯ + k P ¯ ± k P 2k M(t)f [M ] = 0 dt dt (11. It should be point out that.16) (11. Two equations must be solved for the Mach numbers at the duct entrance and exit when the flow is in a chokeless condition.13) is simplified into three different forms: d ¯ ¯1 VPk ¯ dt ¯ k+1 ¯ ¯ ± P 2k M (t)f [M ] = 0 (11. M1 .

It can be noticed that the characteristic time defined in equation (11. See advance topic about approximate solution for large resistance. Chapter 4. RIGID TANK WITH NOZZLE 207 the momentum and energy equations in terms of the Mach numbers. . The gas in the chamber assumed to go an isentropic processes and flow is bounded in nozzle between isentropic flow and isothermal flow5 .19) where 4f L is defined by equation (9. 4f L or small entrance Mach number.20) ¯ ¯ ¯ It was utilized that V = 1 and M definition is simplified as M = 1.1 Adiabatic Isentropic Nozzle Attached The mass flow out is given by either by Fliegner’s equation (4. D 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. it also will be assumed that the flow in the nozzle is either adiabatic or isothermal.49).12) simplified into k−1 2 f [M ] = 1 + 1 2 −(k+1) 2(k−1) = k+1 2 −(k+1) 2(k−1) (11.19) for given 4f L and Pexit yields the D P0 (t) entrance and exit Mach numbers.5) reduced into: tc = V (0) A kRT (0)) (11. D The solution of equations (11. are 4f L D = 4f Lmax D 1 − 4f Lmax D 2 (11.22) 5 This work is suggested by Donald Katze the point out that this issue appeared in Shapiro’s Book Vol 1.50).17) becomes ¯ 1 ¯ 1−k dP ¯ k+1 ¯ P k ¯ ± P 2k (t)f [M ] = 0 k dt (11.31.3. The characteristic equations for Fanno flow (9. M1 . 111 as a question 4.18) and (11.47) or simply use cM ρA∗ and equation (11. 11.11. Here. p.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 (11.21) Also it can be noticed that equation (11.3.

208

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

Equation (11.20) can be simplified as
1−k 1 ¯ P 2k dP ± f [m]dt = 0 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¯ ¯ ¯ ¯ sure, at P (t = 0) = 1 and P (t = t) = P . After the integration, equation (11.24) and rearrangement becomes ¯ P = 1± 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 ¯ P = = = 3.0 P (0) 1.5 The time is ¯ 1−k t = tmax P k − 1 k+1 2
−()

(11.26)

And for smaller area

(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)

11.4. 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. Owczarek [1964] found an analytical solution which described here.

11.3.2

Isothermal Nozzle Attached

In this case the process in nozzle is assumed to isothermal but the process in the chamber is isentropic. The temperature in the nozzle is changing because the temperature in the chamber is changing. Yet, the differential temperature change in the chamber is slower than the temperature change in nozzle. For rigid volume, ¯ ¯ V = 1 and for isothermal nozzle T = 1 Thus, equation (11.13) is reduced into ¯ dP ¯ = ±f [M ]P = 0 ¯ dt Separating the variables and rearranging equation (11.29) converted into
¯ P 1

(11.29)

¯ dP ¯ ± f [M ] P

¯ t

¯ dt = 0
0

(11.30)

Here, f [M ] is expressed by equation (11.22). After the integration, equation (11.30) transformed into ¯ ln P = k+1 2
"
−(k+1) 2(k−1)

¯ t
#

¯ P =e

¯ ( k+1 ) 2(k−1) t 2

−(k+1)

(11.31)

11.4
11.4.1

Rapid evacuating of a rigid tank
With Fanno Flow

¯ The relative Volume, V (t) = 1, is constant and equal one for a completely rigid tank. In such case, the general equation (11.17) “shrinks” and doesn’t contain the relative volume term. A reasonable model for the tank is isentropic (can be replaced polytropic relationship) and Fanno flow are assumed for the flow in the tube. Thus, the specific governing equation is ¯ dP ¯ ¯ 3k−1 − k Mf [M ]P 2k = 0 ¯ dt (11.32)

210

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

For a choked flow the entrance Mach number to the tube is at its maximum, Mmax ¯ and therefore M = 1. The solution of equation (11.32) is obtained by noticing that ¯ M is not a function of time and by variables separation results in
¯ t ¯ P

¯ dt =
0 1

¯ dP ¯ ¯ k M f [M ]P
3k−1 2k

=

1 ¯ f [M ] kM

¯ P 1

¯ 1−3k ¯ P 2k dP

(11.33)

direct integration of equation (11.33) results in ¯ t= 2 ¯ 1−k − 1 2k ¯ f [M ] P (k − 1)M (11.34)

¯ It has to be realized that this is “reversed” function i.e. t is a function of P and can be reversed for case. But for the chocked case it appears as ¯ (k − 1)M f [M ] ¯ ¯ P = 1+ t 2
2k 1−k

(11.35)

The function is drawn as shown here in Figure (11.5). The Figure (11.5) shows
1.0

P(t) 0.8

V(t) = P (t) V(t) = P (0)

0.6

0.4

0.2

0 0 0.2 0.4 ¯ t 0.6 0.8 1.0

Fig. 11.5: The reduced time as a function of the modified reduced pressure

The big struggle look for suggestion for better notation.

that when the modified reduced pressure equal to one the reduced time is zero. The reduced time increases with decrease of the pressure in the tank. At certain point the flow becomes chokeless flow (unless the back pressure is complete vacuum). The transition point is denoted here as chT . Thus, equation (11.34) has to include the entrance Mach under the integration sign as
¯ P

¯ ¯ t − tchT =

PchT

1 ¯ 1−3k ¯ P 2k dP ¯ k Mf [M ]

(11.36)

11.4. 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. Further, when the flow became chokeless the entrance Mach number does not change much from the choking condition. Again, for the special cases where the choked equation is not applicable the integration has to be separated into zones: choked and chokeless flow regions. And in the choke region the calculations can use the choking formula and numerical calculations for the rest. Example 11.2: A chamber with volume of 0.1[m3 ] is filled with air at pressure of 10[Bar]. The chamber is connected with a rubber tube with f = 0.025, d = 0.01[m] and length of L = 5.0[m] S OLUTION The first parameter that calculated is
4f L 4f L D D

=5

11.4.2

Filling Process
¯ dP ¯ ¯ 3k−1 = 0 2k ¯ − k Mf [M ]P dt

The governing equation is (11.37)

For a choked flow the entrance Mach number to the tube is at its maximum, Mmax ¯ and therefore M = 1. The solution of equation (11.37) is obtained by noticing that ¯ M is not a function of time and by variable separation results in
¯ t ¯ P

¯ dt =
0 1

¯ dP ¯ ¯ k Mf [M ]P
3k−1 2k

=

1 ¯ [M ] k Mf

¯ P 1

¯ 1−3k ¯ P 2k dP

(11.38)

direct integration of equation (11.38) results in ¯ t= 2 ¯ 1−k P 2k − 1 ¯ (k − 1)Mf [M ] (11.39)

It has to be realized that this is a reversed function. Nevertheless, with today computer this should not be a problem and easily can be drawn as shown here in Figure (11.5). The Figure shows that when the modified reduced pressure equal to one the reduced time is zero. The reduced time increases with decrease of the pressure in the tank. At some point the flow becomes chokeless flow (unless the back pressure is a complete vacuum). The transition point is denoted here as chT . Thus, equation

212

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS
1 V(t) = P(t) V(t) = V(0)

0.8

0.6

0.4

0.2

0

0.2

0.4

0.6

0.8

1

Fig. 11.6: The reduced time as a function of the modified reduced pressure

(11.39) has to include the entrance Mach under the integration sign as
¯ P

¯ ¯ t − tchT =

PchT

1 ¯ 1−3k dP ¯ 2k ¯ [M ] P k Mf

11.4.3

The Isothermal Process

¯ For Isothermal process, the relative temperature, T = 1. The combination of the isentropic tank and Isothermal flow in the tube is different from Fanno flow in that √ the chocking condition occurs at 1/ k. 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. It has to be remembered that the chamber can undergo isothermal process. For the double isothermal (chamber and tube) the equation (11.6) reduced into
ρ c(0)

¯¯ ¯ P1 P (0) P (0)V (0) d P V ¯ ¯ ± A kRT (0)Mmax M (t) = 0 ¯ tc RT (0) dt R T (0)

  ¡ò þ @¯Gû€ùø÷4Gõs™ñ ý üöô úÿò ñöô ó ò

ïð

î íëê é ç æåã â á Qìwˆè @¯Q䀙à

or

(11.40)

(11.41)

11.4. RAPID EVACUATING OF A RIGID TANK

213

11.4.4

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. This situation occurs at least in small ranges for airbag balloon etc. The physical explanation when it occurs beyond the scope of this book. Nevertheless, a general solution is easily can be obtained similarly to rigid tank. Substituting equation (11.42) into yields d ¯ 1+k P k ¯ dt Carrying differentiation result in ¯ 1 1 + k ¯ k dP ¯ ¯ k+1 M f [M ] = 0 2k P ¯ −P k dt Similarly as before, the variables are separated as
¯ t

¯ −P

k+1 2k

¯ M f [M ] = 0

(11.43)

(11.44)

dt =
0

k 1+k

¯ P 1

¯ k−1 ¯ P 2k dP ¯ M f [M ]

(11.45)

The equation (11.45) integrated to obtain the form 2k 2 ¯ 3k−1 ¯ t= ¯ 1 − P 2k Mf [M ](3k − 1)(1 + k) (11.46)

The physical meaning that the pressure remains larger thorough evacuating process, as results in faster reduction of the gas from the chamber.

11.4.5

The “Simple” General Case

The relationship between the pressure and the volume from the physical point of view must be monotonous. Further, the relation must be also positive, increase of the pressure results in increase of the volume (as results of Hook’s law. After all, in the known situations to this author pressure increase results in volume decrease (at least for ideal gas.). In this analysis and previous analysis the initial effect of the chamber container inertia is neglected. The analysis is based only on the mass conservation and if unsteady effects are required more terms (physical quantities) have taken into account. Further, it is assumed the ideal gas applied to the gas and this assumption isn’t relaxed here.

785714286) for k = 1. The physical meaning of n = 0 is that the tank is rigid.52) . 0. In reality the value of n lays between zero to one.49) Again.47) When n can be any positive value including zero. EVACUATING AND FILLING A SEMI RIGID CHAMBERS Any continuous positive monotonic function can be expressed into a polynomial function. similarly as before.43). Yet.48) Notice that when n = 1 equation (11. as first approximation and simplified approach can be done by a single term with a different power as V (t) = aP n (11. (11.51) 3k − 2 2k (11. This undersigned is not aware situation where this indeed play important role. 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. 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. There isn’t a real critical value to n. it is waited to find such conditions before calling it as critical condition. After carrying–out differentiation results ¯ 1 + nk − k ¯ 1+nk−2k dP ¯ ¯ k+1 M f [M ] = 0 k 2k P ¯ −P k dt (11.50) Carrying–out the integration for the initial part if exit results in 2k 2 ¯ 3k−2nk−1 ¯ 2k 1−P t= ¯ Mf [M ](3k − 2nk − 1)(1 + k) The linear condition are obtain when 3k − 2nk − 1 = 1 −→ n = That is just bellow 1 (n = 0. Therefore.214 CHAPTER 11.4. variables are separated and integrated as follows ¯ t 0 1 + nk − k dt = k ¯ P 1 ¯ 1+2nk−5k dP ¯ 2k P ¯ Mf [M ] (11. However. 6 Some suggested this border point as infinite evocation to infinite time for evacuation etc.49) reduced to equation (11. d ¯ 1+nk−k k ¯ P dt ¯ k+1 ¯ − P 2k M f [M ] = 0 (11.

11.19) as Min Pexit = . equation (11. E R G.5 Advance Topics The term 4f L is very large for small values of the entrance Mach number which D requires keeping many digits in the calculation. otherwise equations (11.53) and equation (11. For small values of the Mach numbers. Eckert. G.18) can be approximated as 4f L D = 1 Mexit 2 − Min 2 k Mexit 2 Min 2 (11. 118.19) is described in “Pressure die casting: a model of vacuum pumping” Bar-Meir. 2. R. Journal of Manufacturing Science and Engineering (USA).53) and (11. P0 (t) Mexit The solution of two equations (11. May 1996.00286.19) must be solved numerically. no. Vol. . 259-265. The solution of equation (11. J. ADVANCE TOPICS 215 11.18) and (11. Goldstein.55) This solution should used only for Min < 0.5.54) yields 1− Pexit P0 (t) 2 (11.54) Min = k 4f L D . (11. pp.18) and (11.

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

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

Nevertheless. These critical conditions are very significant that they provide guidelines for the design of processes. in some situations like die casting this description is appropriate. 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. For these reasons the optimum size is desired.1)). In the exhaust system cost of large exhaust valve increase with the size and in addition reduces the strength with the size of valve2 . 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. Obtaining conditions where the volume of displacing liquid is equal to the displaced liquid are called the critical conditions. eliminating the need for calculations of the gas flow in the tube to minimize the numerical calculations. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 12. V (t) = V (0) 1 − 2 After t tmax (12. The second objective is to provide a tool to “combine” the actual tube with the resistance in the tube. which is also indicated by experimental studies and utilized by practiced engineers. The first approximation isotropic process describe the process inside the cylinder (see Figure (12. this model can be extended numerical in cases where more complex function is applied. isontropic process  © § ¥ ¨¦¤¢ £ 1 2 Fanno model Fig.1.218CHAPTER 12. the possibility of crack increases. it is noteworthy to think about the process in qualitative terms. Obviously. . In reality.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. The conflicting requirements suggest an optimum area.1: The control volume of the “Cylinder” Before introducing the steps of the analysis.1) certain sizes. 12. The replacing incompressible liquid enter in the same amount as replaced incompressible liquid. A linear function is the simplest model that decibels changes the volume. 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. the best ventilation is achieved with a large tube or area. But in a compressible substance the situation can be totally different. thus.

The second characteristic time. (12. Notice that in this case that there are two different characteristic times: the “characteristic” time.3) is an nonlinear first order differential equation and can be rearranged as follows ¯ ¯ dt dP = k−1 ¯ tmax ¯ 1−t ¯ ¯ k 1 − tc M f [M ]P 2k P . tc and the “maximum” time.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)).10) that was developed in the previous Chapter (11) obtained the form as ¯ P 1 k ¯ ¯ ¯ 1 V dP dV ¯ dt + dt ¯ kP + ¯ tmax M f (M ) ¯ P tc k+1 2k =0 (12.1. ¯ P (0) = 1.4) Equation (12. The solution of equation (12.5) Equation (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. dP = k−1 [ξ]( k−1 ) dξ Utilizing this definition and there implication reduce equation (12.3) ¯ where t = t/tmax .12. GENERAL MODEL Equation (12.6) (k − 1) (1 − Bξ) [ξ] ¯ dt ¯ 1−t (12. tmax . Equation (12. The first characteristic time.7) ¯ =t (12. 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 (12.8) .4)) can be obtained by transforming and by 2k ¯ k−1 ¯ introducing a new variable ξ = P 2k and therefore P = [ξ] k−1 . The reduced Pres2k −1 2k ¯ sure derivative.1) can be non–dimensionlassed as ¯ ¯ ¯ V (t ) = 1 − t 219 (12. tmax is associated with the imposed time on the system (in this case the elapsed time of the piston stroke).4) 2 [ξ]( k−1 ) 2k −1 dξ 2k k−1 = where B = tmax ¯ tc M f [M ] And equation (12.

).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. This happen when vacuum is applied outside the tube (a technique used in die casting and injection molding to improve quality by reducing porosity.2) and 12. When tmax > 1 the pressure increases significantly tc tc and verse versa. These Figures (12. the question remains how the time ratio can be transfered to parameters that can the engineer can design in the system.220CHAPTER 12.14) @Ac .13) can be expressed as 1 tmax tc = @A tmax tc @A tmax tc (12. Thus.4) is transformed into k 1− ¯ dP = ¯ dt with the initial condition of tmax ¯ ¯ k−1 2k tc Mf [M ]P ¯ 1−t (12. The results of a choked case are presented in Figure (12.10) P (0) = 1 (12.2). In the literature. A can be expressed as A= The actual times ratio tmax tc @A A Ac Ac (12. Ac tc provides the needed tool. to create a direct function equation (12. The solution is applicable to indirect connection. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT Now substituting to the “preferred” variable 1− tmax ¯ ¯ k−1 2k tc Mf [M ]P 2 t ¯ (1−k) max M f [M ] tc 1 ¯ k−1 P 2k ¯ =t ¯ P (12. In this case only some part of the flow is choked (the later part). Denoting the area that creates the ratio tmax = 1 as the critical area. Thus the exit area.3 demonstrate the importance of the ratio of tmax .3) in which outside tube condition is in vacuum.11) The analytical solution also can be approximated by a simpler equation as ¯ P = [1 − t] tmax tc (12.9) The analytical solution is applicable only in the case which the flow is choked thorough all the process. In case when the flow chokeless a numerical integration needed to be performed.

Ac . The influence is small on the residual air mass in the cylinder but larger on the Mach number. Ac . . The effects of the area ratio. This parameter also represents the dimensionless area. The A A pressure in this quasi steady state is a function of Ac .5) tc is inversely proportional to area.org). tmax = 0. The required solidification time. D S OLUTION 12. The direct connection posts more mathematical difficulty because the critical area is not constant and time dependent. which was obtained by an integration of equation (12.2 Examples Example 12. is not function of the time.3 Direct Connection In the above analysis is applicable to indirect connection. A Mexit . It is important to point out the significance of the tmax . equation (12. according to the following equation Figure (12. are studied here since it is the dominant parameter.1.15) A Parameters influencing the process are the area ratio. Thus.9). Ac . This parameter reptc resents the ratio between the filling time and the evacuating time. and the friction parameter.4). When Ac is greater than one the pressure is concave upA ward and when Ac is less than one the pressure is concave downward as shown in Figures (12.03[sec].12. This quasi steady state is achieved when the volumetric air flow rate out is equal to the volume pushed by the piston. GENERAL MODEL 221 According to equation (11.4) describes the pressure as a function of the dimensionless A A time for various values of Ac .14) the tmax is canceled and reduced into tmax tc = @A A Ac (12.1. The pressure and the mass flow rate are maintained constant after this state is reached. 12. it was found that the influence of the parameter 4f L D on the pressure development in the cylinder is quite small. 4f L .potto.1. 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 values. For small values of Ac there A is no steady state stage. A For large values of Ac the pressure increases the volume flow rate of the air until a quasi steady state is reached.001[m3] and 4f L = 20. The line that represents Ac = 1 is almost straight. Ac . From other detailed calculations the author thesis (later to be D published on this www.1: Calculate the minimum required vent area for die casting process when the die volume is 0. tc ∝ 1/A. It should be noted that critical area.

.222CHAPTER 12. taking into account the compressibility of the gas flow. This critical area depends on the geometry and the filling time. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT To continue 12.2 Summary The analysis indicates there is a critical vent area below which the ventilation is poor and above which the resistance to air flow is minimal. The critical area also provides a mean to “combine” the actual vent area with the vent resistance for numerical simulations of the cavity filling.

2: The pressure ratio as a function of the dimensionless time for chokeless condition  0.0                                                                                         2.12.8 0.0 o ∗ ∇  3.0 0.0 0.6 ∇  4fL D = 100.0 5.5 0.0 o ∇  3.0 ∗ ∇ ∇ 0.9 1.2 2. SUMMARY 223 ∗ ∗ ∗ 1.8 1.6 0. 12.6 Dimensionless Area.7    4.8 0.0 1.5 1.8 ∇ ∗ ∇    4.5  ∗  2.2 0. or.3 0.1 0.2 0.4 1.5 1.0                                                                                         ∗ ∗ 0.2 ∗ ∗ 0.8 ∇ ∗ ∇ 0.0 ∗ ∇   P (t ) P (0) 3.8 1.0   0.2 0.0 ∗ ∗ ∇  4. A/Ac ∗ ∇  5.0 5.1 0.6   P (t ) P (0) 3.0 ∇ ∇ ∇ ∇ ∇ 0.0 2.2 2.2 0.2 0.4 0.4 ∗ ∇  ∗ ∇    3.6 Dimensionless Area.0 1.0 ∗ ∗ ∇  ∇ ∇ ∇ ∇ ∇ 4.0 . t.9 1.2 1.7 Dimensionless Time.4 Figure a 0.6 ∗  4fL D = 5.3 0. Cylinder Volume Fraction  ∗  2.4 1.2.2 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 ∗ ∗ 0.4 ∗ ∇  ∗ ∇   3. A/Ac ∗  ∇ 5.0 ∇ ∇ ∇ ∇ ∇ ∇ ∗ ∇ ∇ ∗ ∇ ∇ ∗ ∗ ∇ ∇ ∇ ∇ ∗ ∗ ∗ ∇ ∇ ∇ ∗ ∗ ∇ ∇ ∗ ∗ ∇ ∇ ∇ 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 ∗ ∗ ∗ ∗ ∗ ∗ Figure b Fig.

.. 1.8 0. ∗ × × × × × × × × × × × × × × × × × × 2 1... ∗∗ . EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 2. 12. ∗ . 0. ∗ 0..2 P(t) P(0) 0...7 0..4 0. ∗ ...2 ∗ . ∗ . CYLINDER VOLUME FRACTION Fig. ..8 0.5 ....0 ∗ ..6 0...224CHAPTER 12..2 0. ∗ .. .. ∗ ∗ . ∗ .. . ∗ . 4... ∗ .. Ac ∗ . A .. 1. ∗ . ∗ .3 0.5 0..1 ....4: The pressure ratio as a function of the dimensionless time                              0..3: The pressure ratio as a function of the dimensionless time for choked condition Fig.2 0 0 0..9 DIMENSIONLESS TIME. 12.4      0. ∗ × × × 1 . ∗ ..8 1..1 0. ∗ . . . ... ∗∗ . t.. ∗ ... ∗ = 0...5 ∗ .. .. ∗ .. or.4 1. . ∗ ...0 .6      1 × × × × × × × × × × × × × × × × × ×     1. .6 .0 ... ∗ ∗.

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. A tube with a membrane as shown in Figure 225 . First the physical case which is analyzed is described.

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

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

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

the stability issue will be examined in greater detail. In such a case. Alternately. a change in the flow direction4 .3) exhibits the schematic of the oblique shock. However. please just ignore it.2. the oblique shock and the Prandtl–Meyer function. it was assumed that normal shock occurs at the same time. Hence. Previously (see Chapter 5). is the direction of the flow after the shock (parallel to the wall). Only the oblique shock’s normal component undergoes the “shock.3 Introduction to Zero Inclination What happens when the inclination angle is zero? Which model is correct to use? Can these two conflicting models.3: A typical oblique shock schematic considered.3. using an infinite (or a two–dimensional) object produces a two–dimensional shock. which depends on the boundary layer. 14. the mass balance reads arsi on L ine Comp ρ 1 U1 n = ρ 2 U2 n 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. The two– dimensional effects occur when the flow is affected from the “side. The deflection angle.” i. To match the boundary conditions. the change of the orientation makes the shock considerations two–dimensional.). a one– dimensional shock can be Fig. δ. for a chosen arbitrary accuracy it requires a specific small area. 14. Figure (14. EH F E GD ( QR5I P 4 53 ) 1 20 (14. the flow turns after the shock to be parallel to the inclination angle.. The three–dimensional effects of the shock make it appear as a curved plane. The total flow after the oblique shock can also be supersonic.3 Oblique Shock C C 87 926 B 2@ A The shock occurs in realθ−δ ity in situations where the shock has three–dimensional θ effects. If you do not like this description.14. the flow is always subsonic. you do not need the author’s permission.e. You can use the traditional explanation. co-exist? Or perhaps a different model better describes the physics. OBLIQUE SHOCK 229 14. In some books and in the famous NACA report 1135 it was assumed that Mach wave and oblique shock co–occur in the same zone.” The tangent component does not change because it does not “move” across the shock line. The normal shock analysis dictates that after the shock. In this chapter.1) .

the deflection angle.3) (14. Experiments and experience have shown that the common solution is the weak shock. the second is the strong shock. and the third is an impossible solution (thermodynamically)6 .8) issue is due to R. θ has to be solved.230 The momentum equation reads CHAPTER 14. From the geometry it can be observed that tan θ = and tan(θ − δ) = U2 n U2 t (14.2). The first is referred to as the weak shock. Yet the new relationship between the upstream Mach number. who raised the solution completeness issue. the term lesser extent is used. tan θ U1 n = tan(θ − δ) U2 n (14.4) Equations (14.1). But because the literature referred to only two roots.4) are the same as the equations for normal shock with the exception that the total velocity is replaced by the perpendicular components.2) (14. here there are three possible pairs5 of solutions to these equations. 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 U1t = U2t The energy balance reads C p T1 + U1n 2 U2n 2 = C p T2 + 2 2 (14. as sin θ = 6 The solution 5 This M1n M1 (14. .6) U1 n U1 t (14. and the Mach angle. requires solving the entropy conservation equation. and (14. δ.5) Unlike in the normal shock. 7 Actually this term is used from historical reasons. The author is not aware of “simple” proof and a call to find a simple proof is needed. The lesser extent angle is the unstable angle and the weak angle is the middle solution.7) The above velocity–geometry equations can also be expressed in term of Mach number. (14. Menikoff. in which the shock turns to a lesser extent7 .

θ. U1t = U2t the Mach number is M1t = M2t because the temperatures on both sides of the shock are different.14.16) The Rankine–Hugoniot relations are the same as the relationship for the normal shock P2 − P 1 P2 − P 1 =k (14.15) (14. The solution from this set of four equations. and δ.12) The relationship between the properties can be determined by substituting M1 sin θ for of M1 into the normal shock relationship.9) (14.8) through (14. OBLIQUE SHOCK M2 n M2 231 sin(θ − δ) = cos θ = (14.17) ρ2 − ρ 1 ρ2 − ρ 1 .10) (14. It should be noted that although. is a function of four unknowns of M1 .11).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. T1 = T2 .11) M1 t M1 M2 t M2 cos(θ − δ) = The total energy across an oblique shock wave is constant. M2 . (14. and it follows that the total speed of sound is constant across the (oblique) shock.13) The density and normal velocity ratio can be determined by the following equation (14. As opposed to the normal shock. 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. here angles (the second dimension) have to be determined.3.

14. He also suggested that θ is bounded by sin−1 1/M1 and .20) (14. the normal shock is a one–dimensional problem. if orientated to a coordinate perpendicular and parallel shock plane is like a normal shock. and the negative sign refers to the mirror image of the solution.4 Solution of Mach Angle Oblique shock. Menikoff. Probably. and the solutions of selected pairs and selected relationships will be presented. and this will be discussed later. M1 .22) (14.1 Upstream Mach Number. the most common and natural to examine. this set of parameters is. Other combinations of properties provide a relatively simple mathematical treatment. thus.18) requires that x has to be a real and positive number to obtain a real deflection angle8 . the negative root of sin θ must be disregarded 1.21) (14.18) Equation (14. OBLIQUE-SHOCK 14. which create a somewhat complicated mathematical procedure. M1 and the deflection angle. perhaps. and Deflection Angle. 8 This point was pointed out by R. Oblique shock is a two–dimensional problem and two properties must be provided so a solution can be found.232 CHAPTER 14. Clearly.19) (14. the most useful properties are upstream Mach number. δ Again. One has to be careful to use the normal components of the Mach numbers. Again. Thus.4. Thus. only one parameter is required (to solve the problem). The stagnation temperature contains the total velocity. sin θ must be positive. Thompson (1950) has shown that the relationship of the shock angle is obtained from the following cubic equation: x3 + a 1 x2 + a 2 x + a 3 = 0 where x = sin2 θ and M1 2 + 2 − k sin2 δ M1 2 2M1 2 + 1 k−1 (k + 1)2 a2 = − + sin2 δ + 4 4 M1 M1 2 cos2 δ a3 = − M1 4 a1 = − (14. the relationship between the properties can be determined by using the normal components or by utilizing the normal shock table developed earlier.

one root is real and two roots are complex. 10 A call for suggestions. For example. 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. 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 .24) (14.14. 9 The highest power of the equation (only with integer numbers) is the number of the roots.29) (14. In the last case where D < 0.4. From a mathematical point of view. (14. all the roots are real and unequal. Maybe insert an example where imaginary solution results in no physical solution.28) (14.30) Only three roots can exist for the Mach angle.25) R+ √ D. . SOLUTION OF MACH ANGLE 233 The solution of a cubic equation such as (14.18) provides three roots9 . if D > 0.23) (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. D > 0 occurs when no shock angle can be found.26) T = and where the definition of the D is 3 R− √ D (14. For the case D = 0. so that the shock normal component is reduced to subsonic and yet parallel to the inclination angle. to explain about complex numbers and imaginary numbers should be included. all the roots are real and at least two are identical. in a quadratic equation there are two roots. θ.

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

Fig. and the weak shock coexist. in Figure (14.2 When No Oblique Shock Exist or When D > 0 Large deflection angle for given. It can be noticed that the normal shock. for. SOLUTION OF MACH ANGLE 235 14. the strong shock.4) and (14. In the alternative view. is just above δmax ∼ π (note that the maximum is also a function of the heat ratio. M1 The first range is when the deflection angle reaches above the maximum point. M1 . a change in the inclination angle requires a larger energy to change the flow direction. This root. and a3 for M1 = ∞. 14.” a change in the flow direction is no longer possible. the fluid “sees” the disturbance (in this case. M1 −→ ∞ the maximum deflection angle is calculated when D = Q3 + R2 = 0. the imaginary angle is shown. The flow is far away from the object and does not “see’ the object. 4 . 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) R= and Q= (1 + k sin2 δ)2 9 (k+1)2 sin2 δ 4 − (2)(−)(1 + k sin2 δ)2 54 Solving equation (14. This can be done by evaluating the terms a1 .4. For example.4. the wedge) in front of it and hence the normal shock occurs. Once. Only when the fluid is away from the object (smaller angle) liquid “sees” the object in a different inclination angle.0. a2 . For a given upstream Mach number. k).14.5). in the case k = 1.4: Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. This different inclination angle is sometimes referred to as an imaginary angle.28) after substituting these values of Q and R provides series of roots from which only one root is possible. The simple procedure For example. the inclination angle reaches the “maximum potential energy.4.

14.20)-(14. The Procedure for Calculating The Maximum Deflection Point The maximum is obtained when D = 0. one can conclude that the maximum Mach angle is only a function of the upstream Much number.12) and equating the results to zero creates relationship between the Mach number. and Mach number.21). M1 . it was seen as C.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. English mathematician to be the creator but later an earlier version by several months was proposed by Bernard Grossman. and (14. M1 . This can be shown mathematically by the argument that differentiating equation (14. (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.5: The view of a large inclination angle from different points in the fluid field. θmax is a function of only these parameters.12). J. When the right terms defined in (14.29).236 CHAPTER 14.Chapman. Thus. .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. It can be noticed that in equation (14. 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. θ. M1 . M1 and the maximum Mach angle. 15 At first.

the maximum appears when θ has its maximum.33) should be referred to as the maximum’s equation.33) Equation (14. Thus. for large Mach numbers. Thus. To compare these two equations the simple case of Maximum for an infinite Mach number is examined. The numerator of equation (14.14. SOLUTION OF MACH ANGLE The differentiation of the equation (14. d tan δ =0 dM1 (14. Once M1n is found.12).8). It must be pointed out that similar procedures can also be proposed (even though it does not appear in the literature).35) k+1 Hence.34) provides a solution for Mmax .18 or θ = 67. Note that this procedure does not require an approximation of M1n to be made. a derivative can be taken with respect to M1 . the maximum deflection angle can be computed. 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. The nominator produces a quadratic equation for sin2 θ and only the positive value for sin2 θ is applied here.12) yields kM1 4 sin4 θ + 2 − (k+1) M1 2 M1 2 sin2 θ − 1 + (k+1) M1 2 2 2 d tan δ = (k+1)2 M1 2 4 2 4 2 dθ kM1 sin θ − (k − 1) + M1 sin θ − 1 4 237 (14.4. Thus. The general solution of equation (14.36) . Instead. which makes ◦ θ = 1. With the value of θ utilizing equation (14. then the Mach angle can be easily calculated by equation (14. the Mach angle is sin θ = 2k . taking the derivative with respect to θ. 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. It can be noticed that the Maximum Deflection Mach Number’s equation is also a quadratic equation for M1n 2 . it is sufficient to equate the numerator to zero to obtain the maximum.32) is zero at different values of the denominator.32) Because tan is a monotonous function.31) is (k + 1)M1 2 + 1 + M1n = (M1 2 M1 2 (k + 1)2 + 8(k 2 − 1) + 16(1 + k) √ 2 k (14.34) and then solving equation (14. 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.79 .

Example 14.31) and neglect all the terms that are relatively small. S OLUTION under construction The case of D ≥ 0 or 0 ≥ δ The second range in which D > 0 is when δ < 0.37) − 27 −1 M1 4 −2 M1 2 +2 M1 2 2 = and 9 M1 + 2 2M1 54 + 1 + 27M1 2 − 2M1 2 M1 2 + 2 54M16 2 (14. Thus. Hint. and a3 for the specific value of M1 as following M1 2 + 2 M1 2 2M1 2 + 1 a2 = − M1 4 1 a3 = − M1 4 a1 = With values presented in equations (14.38) 3 Q= 2M1 2 +1 M1 4 − 9 M1 2 +2 M1 2 3 (14.37) for R and Q becoming 9 R= 2 M1 2 +2 M1 2 2M1 2 +1 M1 4 2 (14. The analysis shows regardless of the value of the upstream Mach number D = 0 when δ = 0. This can be partially demonstrated by evaluating the terms a1 .238 CHAPTER 14. This typical example is for those who like mathematics. a2 . Start with equation (14. OBLIQUE-SHOCK Note that Maximum Deflection Mach Number’s equation can be extended to deal with more complicated equations of state (aside from the perfect gas model). first the transition line in which D = 0 has to be determined.39) .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 + . This can be achieved by the standard mathematical procedure of equating D = 0.

Tabak. others suggest a singular point18 . the ˆˆ question is whether it is possiˆˆcoefficients ˆˆˆ a1 a2 a3 ble for an oblique shock to exˆˆ M1 ˆ ist? The answer according to 3 this analysis and stability anal1.2 ysis is no. And according to 9 this analysis. “Caustics of weak shock waves. the wall does not emit any signal to the flow (assuming zero viscosity).6: The various coefficients of three different tradicts the common approach. Fluids 10 (1) .28) provides the equation to be solved for δ. there was a transition from one kind of discontinuity to another. ∞. in the previous case. 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 . this proof became trivial.  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. there are several papers suggesting zero strength Mach wave. 2.0 3 0 16 can be generated from the wall 1 with zero deflection. But for non ideal gas (real gas) this solution is only an indication. no Mach wave 2. 14.” 206 Phys.16 words.38) (14. In other ∞ -1 0 . SOLUTION OF MACH ANGLE 239 Substituting the values of Q and R equations (14. The question of singular point or zero Mach wave strength are only of mathematical interest. which con.4. 18 See for example. Utilizing the symmetry and antisymmetry of the qualities of the cos and sin for δ < 0 demonstrates that D > 0 regardless of Mach number. in the literature.0. mathematical challenge for those who like to work it out. Substitution of all the above values into (14.Fig.39) into equation (14.6) is provided for the following demonstration. Mach numbers to demonstrate that D is zero Nevertheless.28) results in D = 0. Hence. In the range where δ ≤ 0.. Once this analytical solution was published. M1 = 1.14. with a positive large deflection angle. Table (14. paper by Rosles. are several papers that attempt to prove this point in the past. 17 There 16 A . Note that. January 1998.40) The author is not aware of any analytical demonstration in the literature which shows that the solution is identical to zero for δ = 016 .0 -3 -1 . Nevertheless. this identity can be demonstrated by checking several points for example.

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

Oblique Shock 3 2. SOLUTION OF MACH ANGLE 241 mation of weak weak wave with nonzero strength has engineering applicability in a very limited cases. M1 . and shock angle. M1 .0005 0 0. and Shock Angle. θ. are far much simpler and a unique solution exists.8: The calculation of D (possible error).0005 Myw Mys θs θw 0. θ δ 20 30 The solution for upstream Mach number. and exit Mach number for M1 = 3 0 10 14.001 k = 1 4 Mx=3 0 10 20 30 Wed Jun 22 15:03:35 2005 Fig. shock angle.0 10.0 δ 20.0 -0.5 2 1.41) .4. especially in acoustic engineering. 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) (14.4.5 0 90 80 70 60 50 40 30 20 10 0 -0.14.3 Upstream Mach Number.001 0.0 30.5 1 0. 14. but for most cases it should be ignored.

44) (14. The range of given θ.47) (14. OBLIQUE-SHOCK 2 cot θ(M1 2 sin2 θ − 1) 2 + M1 2 (k + 1 − 2 sin2 θ) (14.4 Given Two Angles.242 or tan δ = CHAPTER 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) (14.48) Even though the solution for these variables.9).45) (14. θ . is limited. The first upstream Mach number. δ and θ It is sometimes useful to obtain a relationship where the two angles are known.41) shows that the shock angle. has to be in the range of sin−1 (1/M1 ) ≥ θ ≥ (π/2) (see Figure 14. Examining equation (14. 14. is limited between ∞ and 1/ sin2 θ.49) . δ. M1 is M1 2 = 2(cot θ + tan δ) sin 2θ − (tan δ)(k + cos 2θ) (14.4. M1 and θ. the possible range deflection angle. is unique.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. upstream Mach number M1 .46) (14.

θ.50) For a large upstream Mach number M1 and a small shock angle (yet not approaching zero). . 14. SOLUTION OF MACH ANGLE subsonic weak solution 243 1.51) (14.14.9: The possible range of solutions for different parameters for given upstream Mach numbers The reduced pressure difference is 2 sin θ sin δ 2(P2 − P1 ) = ρU 2 cos(θ − δ) The reduced density is sin δ ρ2 − ρ 1 = ρ2 sin θ cos(θ − δ) (14.4. Equation (14.0 < M1 < ∞ Defection angle strong solution θmin = sin −1 1 M1 supersonic weak soution possible solution no solution zone θ. Shock angle π θ= 2 θmax ∼ π 2 θ=0 Fig. δ must also be small as well.41) can be simplified into k+1 θ∼ δ = 2 (14. the deflection angle.52) The results are consistent with the initial assumption which shows that it was an appropriate assumption.

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

the close proximity view provides a continuous turning of the deflection angle. δ.97131 δmax 1.L.53) tan θ = . 14. In fact. the A localbecomes possible. it should be covered in this book.11: imity to the wall. the oblique shock is an accumulation of many small (or again weak) “weak shocks. However. the shock and a far view of oblique shock..5152 θmax 76. 14..1: Table of maximum values of the oblique Shock k=1.14. In this theory. under construction. these small “shocks” are built or accumulate into a large and abrupt change (shock). In larger distance from the wall.1000 My 0. In reality. the boundary θ layer (B. Yet. studying this topic seems to come at the expense of other more important topics. For small deflection angles.) does not enter into the calculation. after the explicit analytical solution has been provided. the author admits that as long as there are instructors who examine their students on this issue.8 Maximum Value of Oblique shock Table 14. δ. At the local view close to the wall. and small normal upstream Mach numbers.” However.4.11). the boundary layer increases the zone where a continuous δ flow exists. M1 ∼ 1+ .4. The boundary layer reduces the upstream flow velocity and therefore the shock does not exist at close proxFig. The traditional approach to reconcile these two views is by suggesting that the far view shock is a collection of many small weak shocks (see Figure 14.2762 .4 The maximum values are summarized in the following Table . Mx 1.7 Close and Far Views of the Oblique Shock In many cases. 1 M1 2 − 1 (14. SOLUTION OF MACH ANGLE 245 author is not aware of any single case in which this topic is used in real–world calculations. 14. the far view shows a sharp transition.4. the oblique shock is a weak “weak oblique” shock. From the far view.

4272 12. 14.6465 64.8000 3.7000 1.94925 0. OBLIQUE-SHOCK Table 14.7023 66.92224 0.92165 0.0000 9.93083 0.96630 0.0464 66.0399 65.0000 3.6000 2.3934 37.0000 8.3645 67.0000 My 0.5676 65. In some cases.9442 6.95049 0.0119 19.8443 65.3000 1.1675 22.0922 38. this assumption might not be sufficient and different analysis is needed. Henderson and Menikoff20 suggested a procedure to calculate the maximum deflection angle for arbitrary equation of state21 .99337 0.4. 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.93629 0.9 Detached shock When the mathematical quantity D becomes positive.4419 It must be noted that the calculations are for the perfect gas model.2503 67.1619 44.6515 17.9087 66.9668 64. for large deflection angle.2546 43.1: Maximum values of oblique shock (continue) k=1.9555 69.92478 0.8000 4.1177 42.94387 0. 179–210.7532 64.1028 28.9020 67.7908 44. .98183 0.95435 0.7593 65. there isn’t a physical solution to an oblique shock.1833 21. 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.4000 1.4000 2.99047 0.3275 36.5787 65.5875 34.6074 64.2000 2.97214 0.5000 1.96942 0.3673 67.4 Mx 1.2000 1.4000 3.91997 0.93747 0.2309 65.91941 0.3059 38.6814 30.7739 41.91871 0.95897 0.8000 1.0000 7.8137 32.99559 δmax 3.9735 26.0000 6.0000 2.246 CHAPTER 14.4398 43.96335 0.1127 14.92683 0.0000 10.0734 35.6000 3.98714 0.3066 64.99440 0.7972 65.2000 3.5671 66. Since the flow “sees” the obstacle.4144 65.6621 9.0000 5.6000 1.4290 θmax 71.6934 64.9000 2.1196 67.

a graph and a general explanation to engineers is provided.54) .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). Yet. Upstream U∞ Figure 14. In such a case. a wedge or 22 See example 14. While at zone Fig. In the typical example. q gY d b p i Y hg ce cd `ab Y X W V `f ‚ †  „ ‰ ˆ t…vu ‡yx ‚† … „ ƒ sx vs € tu ‚ R9t € yx tuvw Rr r s s U T '2S (14. As a general rule. the increase in the upstream Mach results in a decrease of the detachment distance. it can be explained as the subsonic is “sucked” into gas in zone C. 14. Analysis of the detached shock can be carried out by looking at a body with a round section moving in a suM >1 personic flow (the absolute velocity Strong Shock isn’t important for this discussion). or.5. these calculations can be summarized by the flowing equation w ea k sh oc k detachment distance = constant × (θ − f (M∞ )) body thickness where f (M∞ ) is a function of the upstream Mach number which tabulated in the literature. The constant and the function are different for different geometries. Even though this topic has few applications. The flow in zone A has to go through some acceleration to became supersonic flow. Thus. The in a supersonic flow weak oblique shock is predicted to flow around the cone. The diszone A Normal Shock tance of the detachment is determined to a large degree by the upstream Mach number. the flow becomes “blinder” to obstacles. The detached shock’s distance from the body is a complex analysis and should be left to graduate class and researchers in this area.12: The schematic for a round–tip bullet C the flow must be supersonic.14. The explanation to such a phenomenon is above the level of this book (where is the “throat” area question22 . some might be used in certain situations which the author isn’t aware of. Regardless of the explanation. this phenomenon has a larger impact for a relatively smaller supersonic flow.4. This shock is referred to as the detach shock. Larger shock results in a smaller detachment distance.10 Issues Related to the Maximum Deflection Angle The issue of maximum deflection has a practical application aside from the obvious configuration used as a typical simple example. alternatively. 14.12 exhibits a round–tip bulzone B Subsonic Area θ let with a detached shock. Nevertheless. SOLUTION OF MACH ANGLE 247 from the body. the gas must go through a shock.

It can be observed that the maximum of the oblique shock for the δ1 θ1 perfect gas model depends only on the upstream Mach number i. The Mach number decreases after every shock. Consider. a maximum oblique shock that a contion section with Mach reflection stant deflection angle decrease of the Mach number results in an increase of δ1 Mach angle (weak shock only) M1 > θ1 M2 =⇒ θ1 < θ2 . due to the symmetry flow a slip plane angle can be guessed to be parallel to original flow. However. Such configurations seem sometimes at first glance not related to the detached shock issue. this situation causes the detached shock to appear in the second Fig. This situation is known in the scientific literature as the Mach reflection.13: The schematic for a symmetrical sucFig. Thus. 14.55) ˜ gf ‘ d˜ — – “ ’ hg'eh™¡'• ” ¡‘ j ki Additionally. the maximum deflection angle decreases with a decrease the Mach numA sub sonic ber. t z x l v t s q n m yyrwurrp o 5Gl . If the deflection angle exceeds the maximum possible. No detached shock issues are raised when only the first oblique shock is considered. yet the mathematics is somewhat complicated. a symmetrical suction section in which the deflection angle is just between the maximum deflection angle and above half of the maximum deflection angle. The analysis of this situation is logically very simple. It seems that the most straightforward method is the following: (a) Calculate M1B . for example. The calculations for such a case can be carried out by several approaches. Additionally.. for U B every upstream Mach number there C θ2 Slip Plane is only one maximum deflection anA δ2 gle.14: The “detached” shock in a complicated configuration sometimes reoblique shock. This detached shock ferred to as Mach reflection manifested itself in a form of curved shock (see Figure 14. a detached shock occurs.13). as before. the second oblique shock complicates the situation and the second oblique shock can cause a detached shock.248 CHAPTER 14.e. hence δ1 = δ2 . The maximum deflection angle in this case is.14). only a function of the upstream Mach number. δmax = f (M1 ) (14. at least two oblique shocks occur and after their interaction is shown in Figure (14. it can be observed for 14. However. there are configurations in which a detached shock occurs in design and engineers need to take it into consideration. OBLIQUE-SHOCK a cone moves into a still medium or gas flows into it. In this situation. ThereU B C fore.

calculate the weak.15) is exposed to supersonic flow and create an oblique shock.0000 My 0. δ2 utilizing equation (14.3: A cone shown in Figure (14. and the respective shock angles. In discussing these issues.4. the strong Mach numbers. For example. This procedure can be extended to calculate the maximum incoming Mach number. . 14. this zone increases as Mach number increases. at Mach 5 this zone is 8.4635 θw 0.31) equation (c) Calculate the deflection angle. Using the geometry provided in the photo.0000 My s 0.14. The wedge angle that satisfies this requirement is by equation (14.12) 249 (d) Use the deflection angle.7738 θmax 66. δ2 = δ1 and the Mach number M1 B to calculate M1 B . it can be ignored.5◦ . The maximum values are: Mx 4. one must be aware that there are zones of dual solutions in which sharp shock line coexists with a curved line. M1 by checking the relationship between the intermediate Mach number to M1 . What is the maximum wedge angle at which the oblique shock can occur? If the wedge angle is 20◦ .11 Oblique Shock Examples Example 14.56660 δ 0. utilizing (14. Is the shock shown in the photo weak or strong shock? Explain. θ2 .2: Air flows at Mach number (M1 ) or Mx = 4 is approaching a wedge.97234 δmax 38.34907 Example 14.5686 θs 1. predict at which Mach number was the photo taken based on the assumption that the cone is a wedge. POTTO–GDC can be used to calculate this ratio. In general.48523 My w 2. For engineering purposes when the Mach number reaches this value. S OLUTION The maximum wedge angle for (Mx = 4) D has to be equal to zero.28) (a side to the case proximity of δ = 0).0407 To obtain the results of the weak and the strong solutions either utilize the equation (14. Note that no maximum angle is achieved in this shock. SOLUTION OF MACH ANGLE (b) Calculate the maximum deflection angle.28) or the GDC which yields the following results Mx 4.4.

According to his measurement. the cone half angle is 15◦ and the Mach number is 2.4300 0.43◦ and the shock angle is 30. 14.88737 Because the flow is around the cone it must be a weak shock.49) or the Potto-GDC.0990 14. M1 3. . the shock would be weak because the maximum (transition to a strong shock) occurs at about 60◦ . OBLIQUE-SHOCK θ δ Fig. This photo is courtesy of Dr. S OLUTION The measurement shows that cone angle is 14. With given two angles the solution can be obtained by utilizing equation (14.250 CHAPTER 14. Even if the cone was a wedge.0143 30.15: Oblique shock occurs around a cone.2318 My s My w θs θw δ P0 y P0 x 0.099◦. Grigory Toker.56543 2.4522 71. a Research Professor at Cuernavaco University of Mexico.2. Note that the Mach number is larger than the one predicted by the wedge.

the following example is provided.17: Two variations of inlet suction for flow just below M = 1.0 3.0 5. SOLUTION OF MACH ANGLE 251 k=14 Oblique Shock 3 2.0 7.0[bar].0 4.4.12 Application of Oblique Shock Mx 2 3 4 5 6 7 8 9 10 One of the practical applications of the oblique shock is the design of an inlet suction for a supersonic flow.0 Mx Thu Jun 30 15:14:53 2005 My θ δ Fig.5 2 1.0 6. It is suggested that a series of weak shocks should replace one normal shock to increase the efficiency (see Figure (14. For details see Henderson and Menikoff “Triple Shock Entropy Theorem. Clearly.17))a . there is general proof that regardless to the equation of state (any kind of gas). 14.5 0 90 80 70 60 50 40 30 20 10 0 2.16: Maximum values of the properties in an oblique shock 14.14. Example 14. P = 1. 14. (1998) pp.0 10. a In fact. Compare the different conditions in the two ˜– •—•“ ”  …  Œ‹ ‡  ˆŽ hR ‰Š ˆ† … ’ w‘ „ R‚ y€ ¡~ } R{ ƒ|   | .4: The Section described in Figure 14. In such a case.0 8. and T = 17◦ C. To illustrate the design significance of the oblique shock.0. the entropy is to be minimized through a series of oblique shocks rather than through a single normal shock.18 air is flowing into a suction section at M = 2. the flow can be brought to a subsonic Fig. with a proper design.0 9.” Journal of Fluid Mechanics 366. supersonic flow there is less entropy production (less pressure loss).5 1 0.4. 179–210.

7498 85.6667 In the oblique shock.0000 My 0.0000 My s My w θs θw δ 7. the first angle shown is Mx 2.9831 51.99445 0.0000 And the additional information is 23 The results in this example are obtained using the graphical interface of POTTO–GDC thus. In the past the input file was given but the graphical interface it is no longer needed.2346 76.0000 P0 y P0 x 0. 14.252 CHAPTER 14. Assume that only a weak shock occurs.58974 1. no input explanation is given.72087 0.4) S OLUTION The first configuration is of a normal shock for which the results23 are Mx 2.7498 resulting in Mx 1.7498 My s My w θs θw δ P0 y P0 x 0. the new angle is 7◦ + 7◦ with new upstream Mach number of Mx = 1.18: Schematic for Example (14. OBLIQUE-SHOCK different configurations. neglect the detached distance 1 Fig.5000 0.7498 36.6875 2.1931 0.0000 My w θw δ 7.2098 and the additional information by the minimal info in the Potto-GDC is Mx 2.2098 In the new region.7021 36. « ¯® ¤ ­« ª © ¦ ¥ $•¡Gg¬Ge¨ § G$¤ 2 3 4 1.57735 Ty Tx ρy ρx Py Px P0 y P0 x 1.99445 1.5549 14.96524 .71761 1.0000 Py Px Ty Tx P0 y P0 x 1.2485 ´ µ³ ± ²° 7◦ £¢ š  Ÿ ž  œ› š •¡¡•y'uh'G™ Normal shock 7◦ 4.

19).6147 P1 P3 P2 P1 The loss in this case is much less than in a direct normal shock.285 = 2.4458.5 (see Figure 14. SOLUTION OF MACH ANGLE Mx 1.2574 Thus the area ratio has to be 1. 14. What is the requited “throat” area ratio to achieve acceleration from the subsonic region Fig. 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.8770 1.5000 My 0. the results are: Mx 1.45115 Ty Tx ρy ρx Py Px P0y P0 x 3.99549 An oblique shock is not possible and normal shock occurs.9496 P0 1 P0 3 P0 2 P0 1 The static pressure ratio for the second case is P4 P4 P3 P2 = = 1.2626 × 1.2609 14.6116 0.3151 4.4.4018 1.2346 My 0. .1853 0.96524 × 0. In fact.96089 0. the loss in the normal shock is above than 31% of the total pressure.82141 Ty Tx ρy ρx Py Px P0y P0 x 1.98903 With two weak shock waves and a normal shock the total pressure loss is P04 P03 P02 P04 = = 0.4458 0.0000 Py Px Ty Tx P0 y P0 x 253 1.5088 41. In such a case.7498 My w θw δ 7.1250 0.19: Schematic for Example (14.2626 1.14.4 yields M 0.90506 1.5: 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 3.1497 1.6116 × 1.5) 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 3.98903 × 0.21295 Now utilizing the isentropic relationship for k = 1.45115 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.86966 1. The half wedge angle is 10◦ .99445 = 0. Example 14.

8807 23.8482 88.0000 P0 y P0 x 0.0000 My s My w θs θw δ 6.0000 P0 y P0 x 0. 14.6: CHAPTER 14.7049 88.99894 And the isentropic relationships for M = 2.9476 21.254 Example 14.997731283 P00 P01 P00 .04263 0.48013 2.0000 P0 y P0 x 0.7008 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0. 1 S OLUTION The shock BD is an oblique shock with a response to a total turn of 6◦ .7080 0. Assume that there are no boundary layers and Fig.7049 2.40596 0.04290 0.0.99879 = 0. The total pressure ratio at zone 2 is P02 P02 P01 = = 0.5990 0.47641 2.48610 2.99879 For the shock BC the results are Mx 2. The conditions for this shock are: Mx 3.8482 My s My w θs θw δ 3.99894 × 0. 2.20).1978 3.0000 My s My w θs θw δ 3. 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 14.99105 The transition for shock AB is Mx 3.7008 87. Find the required angle of shock BE.10500 0. Then.20).20: Schematic of two angles turn with two weak shocks all the shocks are weak and straight.13632 0.10548 3.1854 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. Perform the calculation for M1 = 3.8912 22.13665 The combined shocks AB and BC provide the base of calculating the total pressure ratio at zone 3. explain why this description has internal conflict.9356 0.40669 0.7049.7008 are M 2.

00000 From the last calculation it is clear that the government proposed schematic of the double wedge is in conflict with the boundary condition. T1 = 27◦ C.04290 × 0. check whether the pressure at 2 is above or below or above the pressure (ratio) in zone 4.7◦ . Only when the flow is far away from the double wedge.49525 2. In reality the flow of double wedge will produce a curved shock surface with several zones.00000 The shock angle.0019 1.4). the flow behaves as only one theoretical angle of 6◦ exist. To do that.99920 Ty Tx ρy ρx Py Px P0y P0 x 1. the angle for BE shock which cause this pressure ratio needs to be found. SOLUTION OF MACH ANGLE On the other hand. The flow in zone 3 will flow into the wall in about 2.0 θw 21. the pressure at 4 has to be P4 P4 P0 4 = = 0.4 and R = 287J/KgK . The increase has to be P3 /P2 = 0.99105 = 0. θ can be calculated from θ = sin−1 1.7049 My s My w θs 0.042436789 = 1.001867743 To achieve this kind of pressure ratio the perpendicular component has to be Mx 1.0013 1.0008/2.7049 = 21.4.042436789 P0 2 P0 0 P0 2 P2 P Since P02 < P041 a weak shock must occur to increase the static pressure (see Figure 5.0005 1. Assume k = 1.04263 = 0.14.72 δ P0 y P0 x 0.0008 My 0. Example 14.026233 1.997731283 × 0. P0 P2 P2 = 2 = 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.042516045/0.7: Calculate the flow deflection angle and other parameters downstream when the Mach angle is 34◦ and P1 = 3[bar]. and U1 = 1000m/sec.7037 0.715320879◦ The deflection angle for such shock angle with Mach number is Mx 2.

0000 P0 y P0 x 0.4 × 287 × 418.48269 2.4 × 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 2.6[m/sec] Example 14. Yet.61 From the Table (5.66545 Ty Tx ρy ρx Py Px P0 y P0 x 1.5000 My s My w θs 85.88 sin(34. Mx 1.0443 0.128 1.8575 0.5K and the same for the pressure P2 = 2.256 CHAPTER 14.8: For Mach number 2. M1 .89145 The temperature ratio combined upstream temperature yield T2 = 1.5 and wedge with a total angle of 22◦ . the Upstream Mach number.8575 × 3 = 8.8124 δ 11.1280 0.0 θw 34. OBLIQUE-SHOCK S OLUTION The Mach angle of 34◦ is below maximum deflection which means that it is a weak shock.1) or GDC the following can be obtained.0) = 1.53431 2. calculate the ratio of the stagnation pressure. Utilizing GDC for Mach number 2.5 and the angle of 11◦ results in Mx 2.88 = 1. has to be determined M1 = √ U1 1000 = 2.78 P0 y P0 x 0. M1n = M1 ∗ sin θ = 2.15) and (14.6100 My 0.3949 2.3949 × 300 ∼ 418.96873 .0485 2.5 = 872.00 δ 15.57[bar] And the velocity √ √ Un2 = My w kRT = 2.0995 θw 32.13) or simply converting the M1 to perpendicular component.89127 The relationship for the temperature and pressure can be obtained by using equation (14.8800 My s My w θs 0.

then the Mach angle was calculated using the geometrical relationship of θ = sin−1 M1n /M1 . Example 14. With these two quantities.94021 4.9: What is the maximum pressure ratio that can be obtained on wedge when the gas is flowing in 2.5 Mach without any close boundaries? Would it make any difference if the wedge was flowing into the air? If so. Maximum Deflection Mach’s equation was used to calculate the normal component of the upstream. 3 stream line 2 1 θ M1 = 4 δ Assume that the upstream Mach number is 4 and the deflection angle is δ = 15◦ . S OLUTION This kind of problem is essentially two wedges placed in a certain geometry. utilizing equation (14. the following can be obtained: .6854 0. either use the Maximum Deflection Mach number’s equation or the Potto–GDC Mx 2. is obtained.12) the deflection angle. To obtain the maximum point for 2.3573 2. Utilizing the table or the Potto–GDC. SOLUTION OF MACH ANGLE 257 Example 14. For a weak shock the maximum pressure ratio is obtained at the deflection point because it is closest to a normal shock.4. what is the difference? S OLUTION It has to be recognized that without any other boundary condition. the shock is weak shock. It is clear that the flow must be parallel to the wall.7822 δ 29.5 Mach number. 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).10: Consider the schematic shown in the following figure. the upstream Mach number is known together with deflection angle.14. δ. For the first shock.5000 My max θmax 64.7974 Py Px Ty Tx P0 y P0 x 0.60027 In these calculations.

the second deflection angle is also 15◦ .9290 85.0000 1.0629 15.6695 1.7985 1.2028 84.9290 27.0629 15.8500 My s My w θs 0.0000 My w θw δ Py Px Ty Tx P0 y P0 x 2.0000 1.5851 27.7344 0.9290 My s My w θs θw δ P0 y P0 x 0. With these pieces of information by utilizing the Potto-GDC the following is obtained: Mx 2.7822 15.0000 0. OBLIQUE-SHOCK θw δ P0 y P0 x 0.51 P0 y P0 x 0.258 Mx 4.85. Note that hand calculations requires endless time looking up graphical representation of the solution.7822 15.80382 And the additional information by using minimal information ratio button in Potto– GDC is Mx 4.7985 × 1.6695 = 3.0026 P3 P2 P3 T1 T1 T2 = = 1.0 θw 29.0000 0.5764 0.7344 × 1. Utilizing the POTTO–GDC which provides a solution in just a few clicks.90041 With the combined tables the ratios can be easily calculated.2028 32. With these values the following can be obtained: Mx 2.3575 0.2808 32.0000 My s My w θs CHAPTER 14. S OLUTION Here the Mach number and the Mach angle are given.46152 2.51367 2. P1 P1 P2 = = 1.90041 and the additional information is Mx 2.00 δ 10. Again calculate the downstream ratios after the second shock and the deflection angle.5764 = 2.48469 2.11: A similar example as before but here Mach angle is 29◦ and Mach number is 2.96263 .632 T3 T2 T3 Example 14.80382 With a Mach number of M = 2.9290 My w θw δ Py Px Ty Tx P0 y P0 x 2.929.

9419 84.0000 0.8021 T3 T2 T3 Example 14.0058 0.5100 1.0 θw 30.0000 29.3268 0.0000 10.0590 10.41523 3.00 δ 20.3984 ∼ 1.3575 29.06172 While the results for the oblique shock are Mx 5.4089 1.49901 And the additional information is .51 the so called reflective shock gives the following information Mx 2.97569 P1 P1 P2 = = 1.0000 My 0.41523 Ty Tx ρy ρx Py Px P0y P0 x 5.9419 34. What is the deflection angle in this case? S OLUTION For the normal shock the results are Mx 5.0000 My s My w θs 0.0590 10.54894 1.3575 My w θw δ Py Px Ty Tx P0 y P0 x 1.97 P3 P2 P3 T1 T1 T2 = = 1.8500 My w θw δ Py Px Ty Tx P0 y P0 x 2.3575 My s My w θs θw δ P0 y P0 x 0.3268 ∼ 1.9398 34.3984 1. SOLUTION OF MACH ANGLE 259 and the additional information by utilizing the minimal info button in GDC provides Mx 2.4.8000 5.5100 0.14. 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◦ .5131 1.17 P0 y P0 x 0.3582 × 1.97569 and the additional information of Mx 2.4089 × 1.96263 With the deflection angle of δ = 10.3582 0.12: Compare a direct normal shock to oblique shock with a normal shock.

Based on these calculations.49901 The normal shock that follows this oblique is Mx 3. Example 14.6375 2.6375 × 10.47485 Ty Tx ρy ρx Py Px P0 y P0 x 2.0058 30.13: A flow in a tunnel ends up with two deflection angles from both sides (see the following Figure (14.260 Mx 5.0000 My w θw δ CHAPTER 14.5141 0. For the first direction of 15◦ and Mach number =5. S OLUTION The first two zones immediately after are computed using the same techniques that were developed and discussed earlier. 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 My 0. explain whether the slip angle is larger or smaller than the difference of the deflection angle.0000 20.6858 ∼ 6. P1 P1 P2 = = 2.374 ∼ 27.8625 10.36 P3 P2 P3 T1 T2 T1 = = 2.32671 The pressure ratios of the oblique shock with normal shock is the total shock in the second case.1736 2. . OBLIQUE-SHOCK Py Px Ty Tx P0 y P0 x 3.6858 3.13)).5141 × 2. C stream line δ2 θ2 1 D 4 slip plane B 0 2 stream line A 3 φ F θ1 δ1 Illustration for example (14.3740 0.75 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.13) For upstream Mach number of 5 and deflection angle of 12◦ and 15◦ .

5040 My w θw δ Py Px Ty Tx P0 y P0 x 2. it is assumed that the slip angle moves half of the angle to satisfy both of the deflection angles (first approximation).14.0000 0. Mx 5.6986 85.6819 27.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3. This non–continuous velocity profile can occur in our model because it is assumed that fluid is non–viscous.0000 My s My w θs θw δ P0 y P0 x 261 0. that only one deflection angle exist.6986 27.80600 The conditions in zone 4 and zone 3 have two things that are equal.0000 My s My w θs θw δ P0 y P0 x 0.3217 15.5040 24.5◦ which results in Mx 3.6668 13.0000 1. the slip angle has two extreme cases.4. For the two different deflection angles.5000 1. The first case is where match lower deflection angle and second is to match the higher deflection angle.5◦ and Mach 3.0000 1.6668 13. SOLUTION OF MACH ANGLE Mx 5.69317 For the second direction of 12◦ and Mach number =5. Under this assumption the conditions in zone 3 are solved by looking at the deflection angle of 12◦ + 1. In this case. It has to be noticed that the velocity magnitudes in zone 3 and 4 do not have to be equal.80600 And the additional conditions are Mx 5.6963 1.8006 21. If the two sides were equal because of symmetry the slip angle is also zero.5656 0.2845 12.69317 And the additional conditions are Mx 5. They are the pressure and the velocity direction.0000 0.5040 86.6247 1.9238 0. for the analysis. It is to say.43016 3.9791 1.88496 And in zone 4 the conditions are due to deflection angle of 13.2845 12.43914 3.47413 2.9122 21.5040 My s My w θs θw δ P0 y P0 x 0.8006 86.3217 15.6625 0.5◦ = 13.8006 .88496 with the additional information Mx 3.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3.0739 24.5000 0.

Consider the same values as previous analysis.5000 1. 14. Example 14.86179 From these tables the pressure ratio at zone 3 and 4 can be calculated P3 P2 P0 P1 1 P3 1 = = 1. the oblique shock with upstream Mach number. in oblique. There are researchers which suggest that the numerical work is the solution.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. M1 is always greater than 1. Thus.9035 85.86179 with the additional information Mx 3. However.4. 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).9791 ∼ 1.6577 1.6038 0.5 Summary As with normal shock.8006 My w θw δ Py Px Ty Tx P0 y P0 x 2.5000 0.3226 13.18192 P4 P2 P0 P1 P4 1. the . The optimum deflection angle is a function of the Mach number range in which the suction section is operated in.262 Mx 3.46259 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.9035 26. OBLIQUE-SHOCK θw δ P0 y P0 x 0.14: The previous example gave rise to another question on the order of the deflection angles.8006 My s My w θs CHAPTER 14.9316 26.3226 13.6038 To reduce the pressure ratio the deflection angle has to be reduced (remember that at weak weak shock almost no pressure change). To reduce the pressure the angle of slip plane has to increase from 1.6963 1.6247 × 1. as oppose to the normal shock. S OLUTION Waiting for the solution 14. the pressure at zone 3 has to be reduced.5◦ to a larger number.

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

has to be in the opposite direction of the change of the shock angle. Yet. There isn’t any known experimental evidence to show that flow is unstable for δ = 0.52). The change is determined from the solution provided before or from the approximation (14. ∆δ + to maintain the same direction stream lines. 14. the shock wave is unstable according to this analysis to one direction but stable to the other direction. 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. . the weak shock is stable. On the other hand. the virtual change in the deflection angle Fig.56) can be applied for either positive. The opposite happens when the deflection increment becomes negative. In fact.22: The schematic of stability analysis for oblique shock. ∆θ + or negative ∆θ− values. ∆θ = k+1 ∆δ 2 (14. Please note that this analysis doesn’t apply to the case of the close proximity of the δ = 0. OBLIQUE-SHOCK The change in the in∆θ + clination angle results ∆θ − in a different upstream Mach number and ∆δ − a different pressure.56) Equation (14. it must be pointed out that it doesn’t mean that the flow is unstable but rather that the model is incorrect. Thus.264 CHAPTER 14.

out any jump in properties.CHAPTER 15 Prandtl-Meyer Function 15.1) Fig. there is no limitation for the Prandtl-Meyer function to approach zero. when this model breaks down. and no known simple model can describe the situation.2)) µ = sin−1 1 M (15.2: The angles of the Mach line triangle ma xi mu m an gl e 265 À ¸ ¾ ½ » ¿G¼º ¹ · ¶ . The transition. the flow accelerates to match the boundary condition. As opposed to the oblique shock. 15. 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. flow separation occurs. which is defined as (as shown in Figure (15. for very small angles. with. Yet. Supersonic expansion and isentropic comU pression (Prandtl-Meyer function). as opposed to the oblique shock. the deflection angle is denoted as a positive when it is away from the flow (see Figure (15.1 Introduction positive angle As discussed in Chapter (14) when the deflection turns to the opposite direction of the flow.1: The definition of the angle for the Prandtl–Meyer function. Yet. it has to be assumed to be insignificant. the flow becomes complicated. are an extenc sion of the Mach line concept. The Mach line shows that a disturbance in a field of supersonic flow moves in an angle of µ. because of imperfections of the wall and the boundary layer. is smooth.Fig.1)). Here because of the tradition.

The change of the velocity in the flow direction. PRANDTL-MEYER FUNCTION 1 M1 − 1 µ = tan−1 √ (15. Once the contour is changed.6) (15. Later. a discussion on the relationship between the flow properties and the flow direction is presented. dy. and it is assumed here to be isotropic for a positive angle. This assumption. However. Hence. is dy = (U + dU ) sin(dν) = U dν The tan µ is the ratio of dy/dx (see Figure (15.2 Geometrical Explanation Ì ËÎ Í Ì ÊgµÊË x The change in the flow direction is assume to be result of the change in the tangential dx = dU cos(90 − µ) component. The velocity component 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 reason that a “negative” angle is not applicable is that the coalescing of the small Mach wave which results in a shock wave.4) dν ∼ sin(dν). is close to reality.3) . This Mach line is assumed to be a result of the positive angle. the velocity perpendicular to the flow. this assumption will be examflow ined. The Mach line is the chief line in the analysis because of the wall contour shape information propagates along this line.3)) tan µ = dx dU = dy U dν (15. cos(dν) ∼ 1 Ð Ú Õ Ù Ø ÖÕ ÐÓ Ñ ™Êw‡Ó Ï R× y¡gÔÒÐ Ï ch á ÃÅ Ä gµ²Ã à ß Þ Ý ÊÔ¡RÜ Û Á É ÊÈ Â Ç Æ (15. 15. dx is In the same manner. This direction change results in a change of the flow properties.2) A Mach line results because of a small disturbance in the wall contour. The typical simplifications for geometrical functions are used: y li ne Ma These simplifications are the core reasons why the change occurs only in the perpendicular direction (dν << 1). no shock is created from many small positive angles. In this chapter. the total Mach number increases. as it turns out. Therefore.3: The schematic of the turning only.266 or CHAPTER 15. the flow direction will change to fit the wall.5) dx = (U + dU ) cos ν − U = dU (15. 15. the Mach angle dν dy increase and result in a change in the direction of the flow.

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. a more rigorous explanation is provided. 15.11) (15. GEOMETRICAL EXPLANATION The ratio dU/U was shown to be dM 2 dU = U 2M 2 1 + k−1 M 2 2 Combining equations (15.7) (15.6) and (15.10) .2.4: The schematic of the coordinate based on the matheflow turns around a sinmatical description gle point.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. It must be recognized that here the cylindrical coordinates are advantageous because the Fig.1 Alternative Approach to Governing Equations In the previous secback Mach tion.2.7) transforms it into √ M 2 − 1dM 2 dν = − 2M 2 1 + k−1 M 2 2 After integration of equation (15.9) The constant can be chosen in a such a way that ν = 0 at M = 1.8) (15. a simplified verr Ur line sion was derived based Front Mach on geometrical arguUθ line θ ments. In this section.12) ∂Ur Uθ ∂Ur Uθ 2 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r ρ ∂r ρ ∂r (15. 15. For this coordinate system.15.

16) from equation (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. like the first term in the mass equation. the mass equation is reduced to ρUr + ∂ (ρUθ ) =0 ∂θ (15.14) The momentum equations now obtain the form of Uθ ∂Ur Uθ 2 − =0 r ∂θ r ∂Ur Uθ − Uθ = 0 ∂θ Uθ ∂Uθ Uθ Ur c2 ∂ρ − =− r ∂θ r rρ ∂θ ∂Uθ c2 ∂ρ Uθ − Ur = − ∂θ ρ ∂θ Substituting the term 1 ∂ρ ρ ∂θ (15. but only .15) (15.20) Ur + ∂Uθ ∂θ =0 (15. r. then all the derivatives with respect to the radius will vanish. Hence.17) Uθ or Uθ 2 Ur + ∂Uθ ∂θ = c 2 Ur + ∂Uθ ∂θ (15. One has to remember that when r enters to the function.13) Equation (15.19) it follows that Uθ = c (15. PRANDTL-MEYER FUNCTION If the assumption is that the flow isn’t a function of the radius. the derivative isn’t zero.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.16) results in ∂Uθ − Ur ∂θ = c2 Uθ Ur + ∂Uθ ∂θ (15.268 CHAPTER 15.14) into equation (15.

M = 1.26) + Ur 2 = 2h0 (15.15.27) incorporating the constant becomes Ur = 2h0 sin θ k−1 k+1 (15. This analysis can be also applied to the same equation when they are normalized by Mach number.24) Utilizing equation (15.21) If r isn’t approaching infinity. In fact. these results are associated with the characteristic line.2. based on the definition of the Mach angle.23) results in c2 Uθ 2 + U r 2 + = h0 k−1 2 (15. k. the component shown in Figure (15. The energy equation for any point on a stream line is h(θ) + Uθ 2 + U r 2 = h0 2 (15.20) for the speed of sound and substituting equation (15.25) 1 Cp c2 R T = RT = R (k − 1) Cv k−1 (15.24). GEOMETRICAL EXPLANATION 269 the tangential component. equation (15.26) becomes k+1 k−1 ∂Ur ∂θ 2 k−1 + ∂Ur 2 ∂θ + Ur 2 2 = h0 (15. equation (15.22) In the literature.15) becomes Uθ r ∂Ur − Uθ ∂θ =0 (15.3) under Uy is equal to the speed of sound. However.22) which is the radial velocity transforms equation (15. into equation (15.28) . the non–dimensionalization can be applied at this stage as well.23) Enthalpy in perfect gas with a constant specific heat.27) Note that Ur must be positive. The solution of the differential equation (15. ∞ and since Uθ = 0 leads to ∂Ur = Uθ ∂θ (15. equation (15. is c(θ)2 k h(θ) = Cp T = Cp and substituting this equality.25) into ∂Ur 2 ∂θ After some rearrangement. After some additional rearrangement.

27) because sin2 θ+cos2 θ = 1.30) Now utilizing the expression that was obtained for Ur and Uθ equations (15. The upstream Mach number is denoted in this segment as Mstarting .33) The deflection angle ν. the assumption .28) is chosen such that Ur (θ = 0) = 0.29) and (15.270 CHAPTER 15.2. The tangential velocity obtains the form Uθ = c = ∂Ur = ∂θ k−1 k+1 2 h0 cos θ k−1 k+1 (15.32) k+1 tan2 k−1 θ k−1 k+1 (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 (15.2 Comparison And Limitations between the Two Approaches The two models produce exactly the same results. PRANDTL-MEYER FUNCTION which satisfies equation (15. 15. In the geometrical model.6).34) (15. has to match to the definition of the angle that is chosen here (θ = 0 when M = 1).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.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. so ν(M ) = θ(M ) − θ(Mstarting ) = k+1 tan−1 k−1 k−1 k+1 M2 − 1 − tan−1 M2 − 1 (15.35) These relationships are plotted in Figure (15. but the assumptions for the construction of these models are different. The arbitrary constant in equation (15. For this upstream Mach number (see Figure (15.2)) tan ν = Mstarting 2 − 1 (15.

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. (1 + ) the radial velocity is small . However. Prandtl–Meyer function becomes ν∞ = π 2 k+1 −1 k−1 (15. At this point (r = 0) these models fail to satisfy the boundary conditions and something else happens there.5) 15. an increase in the Mach number can result in a very significant radial velocity. the question of boundary layer arises. 15. Normally. In the rigorous model.3.3 The Maximum Turning Angle The maximum turning angle is obtained when the starting Mach number is 1 and the end Mach number approaches infinity. it was assumed that radial velocity is only a function of θ. the fact remains that there is a radial velocity at Ur (r = 0) = constant. The radial velocity is “fed” through the reduction of the density. the maximum turning angle is much larger than the maximum deflection point because the process is isentropic. 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. In fact. some researchers recommend that.15.37) The change in the deflection angle is calculated by . 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.4 The Working Equations for the Prandtl-Meyer Function ν2 − ν1 = ν(M2 ) − ν(M1 ) (15. In this case. 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. For example. However. On top of the complication of the turning point. At a small Mach number. the sharp point should be replaced by a smoother transition.36) The maximum of the deflection point and the maximum turning point are only a function of the specific heat ratios. Thus. Regardless of the assumptions that were used in the construction of these models. 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). What happens when the deflection angel exceeds the maximum angle? The flow in this case behaves as if there is almost a maximum angle and in that region beyond the flow will became vortex street see Figure (15. in many instances.

it is Fig.38) It can be observed that only the area that “seems” to be by the flow was used in expressing equation (15.7). 4 Supposed that a two– dimensional diamond–shape 2 1 body is stationed in a su3 personic flow as shown in Figure (15. 15.5 d’Alembert’s Paradox In ideal inviscid incompressible flows. Again. and θ1 w θ2 this paradox is examined θ2 θ1 here. 4 This result is known as d’Alembert’s Paradox. M1 . . The relation between P2 and P4 is such that the flow depends on the upstream Mach number.7: A simplified diamond shape to illustrate the suassumed that the fluid is inpersonic d’Alembert’s Paradox viscid.38). PRANDTL-MEYER FUNCTION sl ip li ne Maximum turning Fig. the move3 ment of body does not 1 2 encounter any resistance. the drag. 15. k. The net force in flow direction. is D=2 w (P2 − P4 ) = w(P2 − P4 ) 2 (15. and the specific heat.272 CHAPTER 15.5: Expansion of Prandtl-Meyer function when it exceeds the maximum angle 15.

the slip condition is such that the 6 pressure in region 5 and 7 are the same. the Fig. M1 .6. Now. P4 .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. As before. As opposed to 5 the thickness of the body. 7 Again. the thickness of a body was 1 2 shown to have a drag. and additionally the direction of the velocity must be the same. 15. the body also obtains lift. P2 . specific heat. a body with 3 zero thickness but with an angle of attack will be examined. k. is larger than the pressure at zone 4. in addition 4 α to the drag. Thus.6 Flat Body with an Angle of Attack Previously. and the “visible” area of the object. w Slip plane . FLAT BODY WITH AN ANGLE OF ATTACK 273 Prandtl-Meyer Angle 100 80 k=1. there is always drag when the flow is supersonic which depends on the upstream Mach number. the pressure at zone 2. ent between the two regions. This drag is known in the literature as (shock) wave drag. 15. 15.6: The angle as a function of the Mach number Regardless in the equation of the state of the gas.8: The definition of the angle for the magnitude of the velocity will be differPrandtl–Meyer function.

22944 £¡ ¤¢  µ 52.1 The Mach number is then S OLUTION First.4449◦ and results in M 2. Assume the specific heat ratio is k = 1.4449 P P0 T T0 ρ ρ0 0.12734 0.4.1[m] what will the width of this imaginary tunnel after the bend? Calculate the “fan” angle. and the Mach number after the bending point.74448 0.3100 ν 6. A flow of air with a temperature of 20◦ C and a speed of U = 450m/sec flows (see Figure 15.0◦ an inclination.9).0024 ν 26.4620 ÿ þ ý û ù éyôüúø (& )'% $ 1 ¦ 0 §¥ "  © #!  ¨ ï î í ë ê è ç å ã y¡5ìGéræäâ ÷ ö õ ó ñ hyéôòð . Calculate the pressure reduction ratio. 15.35603 0.7 Examples For Prandtl–Meyer Function Example 15.4449 P P0 0.4449 + 20 = 26. the initial Mach number has to be calculated (the initial speed of sound). If the air flows in an imaginary two–dimensional tunnel with width of 0.31 343.6434 µ 63. Fig.47822 The “new” angle should be ν2 = 6. √ √ a = kRT = 1.55497 0.4 ∗ 287 ∗ 293 = 343.274 CHAPTER 15.1 T T0 ρ ρ0 this Mach number is associated with M 1. PRANDTL-MEYER FUNCTION 15.1: A wall is included with 20.9: The schematic of Example 15.1m/sec M= 450 = 1.

4 + 20.55497 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. 1 It isn’t really different from this explanation but shown in a more mathematical form. and the bend angle.0 − 52.67 flows over bend (see Figure 15. There is very little with practicality in this kind of problem.8◦ Reverse the example.15.31 0.74448 0. It isn’t present now because of the low priority to this issue.7.2). It will be presented in the future version. ρ1 x1 M1 c1 = ρ2 x2 M2 c2 =⇒ x2 = x1 ρ 1 M1 ρ 2 M2 T1 T2 275 1.35603 The “new” width can be calculated from the mass conservation equation. Example 15.47822 × = 0.2: Gas with k = 1. 15.1579[m] 0. and this time the pressure on both sides are given and the angle has to be obtained2 .2) after the bend.12734 = 1 = = 0.1 × fan angle = 63. S ¤R¢Q PIHGFED C B@8 A9 W U V §T bpiHhgf dec baB@X `Y 7 6 5¢2 43 .6 = 30.0024 0. Compute the Mach number Fig. EXAMPLES FOR PRANDTL–MEYER FUNCTION Note that P01 = P02 P2 P0 P2 0. The other alternative is using the exact solution1 . The expansion “fan” angle changes in the Mach angle between the two sides of the bend x2 = 0.22944 2. 2 This example is for academic understanding.10: The schematic for the reversed question of example (15. due to Landau and friends.35766 P1 P1 P02 0.

4 but k = 1.60365 0.0◦ 15.4576 ν 9. the following table can be obtained: M 3.1719 P P0 T T0 ρ ρ0 µ 55. Assume that the specific heat ratio at stage is k = 1. It should be noted that specific heat isn’t k = 1.4000 ν 7.31 + 4 for which the following table can be obtained (Potto-GDC) . calculate the drag coefficient and lift coefficient. The bend angle is ∆ν = 9.3.3.03965 With the angle of attack the region 3 will be at ν ∼ 62.47077 ¯ With this pressure ratio P = 0. The relative pressure downstream can be calculated by the relationship P2 P1 1 P2 = = × 0.58419 0. S OLUTION For M = 3.3: Consider two–dimensional flat thin plate at an angle of attack of 4◦ and a Mach number of 3.67.44831 For the rest of the calculation the initial condition is used.3.276 CHAPTER 15. M 1.2 M 1.4623 = 1.28418 0.2619 P0 2 P1 P0 1 1.4◦ ∆µ = 55.37972 0.7720 ∼ 1.5479 0.4623 0.3113 P P0 T T0 ρ ρ0 µ 73.1416 0.8 Combination of the Oblique Shock and Isentropic Expansion Example 15. PRANDTL-MEYER FUNCTION S OLUTION The Mach number is determined by satisfying the condition that the pressure downstream are and Mach given.26190 0.4576.5479 − 54.31424 = 0.1719 − 7.3000 ν 62.01506 0. The Mach number after the bend is M = 1.7720 P P0 T T0 ρ ρ0 µ 54.2619 require either locking in the table or using the enclosed program.

0000 Py Px Ty Tx P0 y P0 x 3.03093 On the other side.43534 3.35248 0.3467 0.7238) sin 4◦ ∼ .9313 20.3000 My w θw δ 4.054 1.15.3467 1.1115 88. a large lift can be obtained.0528 0.33.1066 0.1157 − 0. COMBINATION OF THE OBLIQUE SHOCK AND ISENTROPIC EXPANSION277 M 3.0039 1.7238 P1 P03 P01 P1 0. by clicking on the minimal button.4996 ν 66.1157 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. the oblique shock (assuming weak shock) results in Mx 3. .1157 − 0.8.3100 P P0 T T0 ρ ρ0 µ 74.99676 and the additional information.1115 20.0109 × 1 × ∼ 0.33.3000 My s My w θs θw δ 4.01506 The pressure ratio at point 4 is P3 = 1.0000 P0 y P0 x 0.32 2 (1.7238) cos 4◦ ∼ . Discussion on the optimum design is left for the next versions. provides Mx 3.99676 The pressure ratio at point 3 is P3 P3 P03 P01 1 = = 0.32 This shows that on the expense of a small drag.01090 0.1157 1.

278 CHAPTER 15. PRANDTL-MEYER FUNCTION .

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

279

280

CHAPTER 16. TOPICS IN STEADY STATE TWO DIMENSIONAL FLOW

APPENDIX A Computer Program
A.1 About the Program

The program is written in a C++ language. This program was used to generate all the data in this book. Some parts of the code are in FORTRAN (old code especially for chapters 11 and 12 and not included here.1 . The program has the base class of basic fluid mechanics and utilities functions to calculate certain properties given data. The derived class are Fanno, isothermal, shock and others. At this stage only the source code of the program is available no binary available. This program is complied under gnu g++ in /Gnu/Linux system. As much support as possible will be provided if it is in Linux systems. NO Support whatsoever will be provided for any Microsoft system. In fact even PLEASE do not even try to use this program under any Microsoft window system.

A.2

Usage

To use the program some information has to be provided. The necessary input A parameter(s), the kind of the information needed, where it has to be in a LTEX format or not, and in many case where it is a range of parameter(s). 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.

281

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. A.1: Schematic diagram that explains the structure of the program

APPENDIX A. COMPUTER PROGRAM

A.2. USAGE

283

M1ShockV Entrance Mach M1 when expected shock to the tube Fanno and isothermal classes FLDShockV FLD with shock in the in Fanno class M1fldV both M1 and
4f L D

are given
P1 P2 ,

M1fldP2P1V three part info MxV Mx or My

M1 and

4f L D

are given

infoStagnation print standard (stagnation) info infoStandard standard info for (Fanno, shock etc) infoTube print tube side info for (Fanno, 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 LTEX of Mx The following lines have to be inserted in the end of the main function.

int isTex = yes; int isRange = no; whatInfo = infoStandard ; variableName = MxV; Mx = 2.0 ; s.makeTable(whatInfo, isRange, isTex,

variableName, variableValue);

******************************************* 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. You can just can cut and paste it in your latex file. You must use longtable style file and dcolumn style files.

******************************************* \setlongtables \begin{longtable} {|D..{1.4}|D..{1.4}|D..{1.4}|D..{1.4}|D..{1.4}|D..{1.4}|D..{1.4}|} \caption{ ?? \label{?:tab:?}}\\

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. COMPUTER PROGRAM {$\rule[-0.1in]{0.pt}{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[-0.1in]{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 2.176& 2.152& 0.3608& \hline\end{longtable}

&

1.000&

0.5854&

3.773&

0.6164 \\

A.3

Program listings

Can be download from www.potto.org.

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

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

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