You are on page 1of 335

Fundamentals of Compressible Fluid Mechanics

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

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

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

from The Metalogicon by John in 1159

CONTENTS

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

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

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

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

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

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

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

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

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

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

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

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

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

v 53 62 63 63 70 70 73 73 74

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

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

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

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

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

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

. .4. . 14. . 14. . . . . . . . . . . . . . 14. . . .1 Alternative Approach to Governing Equations . θ . . . . . . . . . . . . . . . . .5 d’Alembert’s Paradox . . . . . . .1. . . . . 14.2 Geometrical Explanation . . . . . . .2 Introduction to Prandtl–Meyer Function . . . . . . . . . . . . . . . 14. .4. . . . . . . .6 Small δ “Weak Oblique shock” . . . . . . . . . . . . . . . .6 Flat Body with an Angle of Attack .4. . . . . . . . . . . 14. . . . . . . . . . . .8 Combination of the Oblique Shock and Isentropic Expansion . . . . . 15. . . . .2. . . . . . . . . . . .4 Given Two Angles. . . . . . . . . . . 12. . .4. . . . . . . .3 Oblique Shock . . . . . .4. . . . . . . . . .7 Close and Far Views of the Oblique Shock .4. .4. . . . . .11Oblique Shock Examples . . . . 14. . . . .1. . . . .2 Summary . .CONTENTS 12 Evacuating/Filing Chambers under External Volume Control 12. . . . . . . . . . . . . . . . .3 Introduction to Zero Inclination . . . . . . . . . . . . . M1 . . . . . . . . and Shock Angle. . . . . . . . . . . . . . . . . . . . . . . . . . 14. . . . . . . .1 Introduction . . . δ and θ . . . . . . . 14. . . .2 Introduction . . . . . . . 14. . . .3 Upstream Mach Number.4. 15. .12Application of Oblique Shock . . . . .4 The Working Equations for the Prandtl-Meyer Function . . . 14. . . . . . . . . .1 Upstream Mach Number. . . . . . . . . . . . 15. . .3 Direct Connection . . . . . . . 15. . . . . . . . . . .2 Comparison And Limitations between the Two Approaches 15. . 14. . . . . . .5 Summary . . . . . . . . . . .13Optimization of Suction Section Design . . 14. . . . . . . .1. 15. . .4. . . . . . . . . . . .1 General Model . . . . . . . . . 14. . . . . .7 Examples For Prandtl–Meyer Function . .10Issues Related to the Maximum Deflection Angle . . . . . . . . . . . . . . . . . . . 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. . . . . . . . . . 14. . . . . . . . .1 Introduction to Oblique Shock . . . . . . . . . . 14. . . . δ 14. . . . .2 When No Oblique Shock Exist or When D > 0 . . . . . .1 Rapid Process . . . . . . . . . . . . . . . . . . . . . .2 Examples . 13 Topics in Unsteady one Dimensional gas dynamics 14 Oblique-Shock 14. . 14. . . . . . . . . . . .4 Solution of Mach Angle . 14. . 15. . . . . . . . . . .5 Flow in a Semi–2D Shape . . . . .8 Maximum Value of Oblique shock . . . . . . . . . . . .1 Preface to Oblique Shock . . . 12. . 14.4. . . . . . . . . 14. . . . . . .3 The Maximum Turning Angle . . .4. . . . . . . . . . . . .4. . . . . . . . . . . . 15. . . . . . . . . . . . . . . . . . . . . . . . . 12. . . . . . . . . . . .2.2.2. . . . . . and Deflection Angle. . . . . . . . . . . . . . . . . . . . . .2. . . . . . . . . .4. . . . . . M1 . .9 Detached shock . . . . . . . . . .6 Appendix: Oblique Shock Stability Analysis . . . . . . . . . . 15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

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

. . . . 1. 7 9 11 12 16 17 17 18 19 20 21 22 28 28 32 4. . 46 ix . . . . . . . . . . . . . 1. . . . . . . .6 Photo of Ernest Mach . . . . . . . . . . . . k = 1. . . . . 3. . . . .4 Flow rate as a function of the back pressure taken from Stodola 1927 Steam and Gas Turbines .4 Control volume inside a converging-diverging nozzle. . . .2 The schematic of deLavel’s turbine after Stodola. .7 The photo of thebullet in a supersonic flow not taken in a wind tunnel 1. . . . . . . . . . . . . .3 The Compressibility Chart . . . . . . . . . 41 4. . . . Steam and Gas Turbine . . . . . . 1. . . .5 Portrait of Galileo Galilei . . . . . . 1. . . . . . . . . . . . . .1 Flow of a compressible substance (gas) through a converging–diverging nozzle. . . . . . . 1. . . . . . .4 44 4. . . . .LIST OF FIGURES 1. . . . . . . . . . . . .10 The photo of Gino Fanno approximately in 1950 . . . . . . . . . 1. . . . . . .3 The stagnation properties as a function of the Mach number. . . . . . . . . . . . . . . . . . . . . 1. . . . . . . . . . . .11 Photo of Prandtl . . . . . . . . . . . . .3 The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines . . . 3. . . . . . . . . . . . . . . . . . .8 Photo of Lord Rayleigh . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. . . . . . . .2 Stationary sound wave and gas moves relative to the pulse . . . . . . . . . . . . . . . . . . .9 Portrait of Rankine . 43 4. . . . . . . . . . . . . . . . . .1 The shock as connection of Fanno and Rayleigh lines after Stodola. . . . . . . . . . . . . . . . . . . . . . . . .2 Perfect gas flows through a tube .12 The photo of Ernst Rudolf George Eckert with the author’s family . . . . . . . . . . 3. . . . . . . . . . . . . . . 1.1 A very slow moving piston in a still gas . 1. Steam and Gas Turbine . . . . . . . . . . . . . . . . . . .

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

. . . . 182 The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow . . . . . . . .15 9. . . . . 178 The pressure distribution as a function of D for a long D The effects of pressure variations on Mach number profile as a funcfL fL tion of 4D when the total resistance 4D = 0. . .5 9. . . . 166 The maximum length as a function of specific heat. . . . .10 9. . 171 The development properties in of converging nozzle . . .2 The Temperature Entropy Diagram For Rayleigh Line . . . . . . 11. . . . . . . .16 9. . . .12 9. . . . . . . 181 The maximum entrance Mach number. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 The control volume of the “Cylinder” . . . . k . . . . . .17 9. . . . . . .2 9. . . . . . . . . .13 9. . . . . . . . . . . . . . . . .4) . . .1 The control volume of Rayleigh Flow . . . . . . . . . . . . . . . 155 Various parameters in Fanno flow as a function of Mach number . . . 11. . . . 195 11. . . .3 The basic functions of Rayleigh Flow (k=1. . . . . . . . . . . .1) . . . . . 185 10. . . . . . .4 The pressure assumptions in the chamber and tube entrance .3 A schematic of the control volumes used in this model . . . .1 The two different classifications of models that explain the filling or evacuating of a single chamber . . . . . . .14 9. . . .3 The Mach number at the entrance to a tube under isothermal flow fL . . .11 9. . . 143 8. . . . . . . . . 170 fL The effects of increase of 4D on the Fanno line . . . . . . . . . 174 fL M1 as a function M2 for different 4D for supersonic entrance velocity. . . . .8 9. . . . . . .3 . . . .6 9. . 181 fL The extra tube length as a function of the shock location. . .3 for Fanno Flow . . . . . . .3 9. . . 218 . .4 9. . . 201 202 202 203 210 212 12.2 Description of the pressure. . . .175 fL fL The pressure distribution as a function of 4D for a short 4D . . . . . .LIST OF FIGURES xi 8. . .18 Control volume of the gas flow in a constant cross section . . . . 153 model as a function 4D 9. . . . . . 11. 189 10. M1 to the tube as a function fL of 4D supersonic branch . . . . . . . . . 179 fL fL Fanno Flow Mach number as a function of 4D when the total 4D = 0. . . . . . 11. . 163 Schematic of Example (9. . .6 The reduced time as a function of the modified reduced pressure . . . . . . . . . . . . .5 The reduced time as a function of the modified reduced pressure . . . . . . . . . .9 9. . . . . . . . . 191 10.2) . . . . . . . . .7 9. . . . temperature relationships as a function of the Mach number for isothermal flow . . . . . . . . . . . . 4D supersonic branch . . . . . . . . . . . 172 The Mach numbers at entrance and exit of tube and mass flow rate fL for Fanno Flow as a function of the 4D . . . . . . . . . 164 The schematic of Example (9. . . 180 Schematic of a “long” tube in supersonic branch .2 A schematic of two possible connections of the tube to a single chamber . . . . . . . . . . . . . 173 4f L M1 as a function M2 for various D . . . . . . . . . . . . . . . 177 4f L 4f L . . . . . . . . . . . . . . .1 9. 11. .

. . . . . . . . . . . . . .xii LIST OF FIGURES 12. . .17Two variations of inlet suction for supersonic flow . . . . . . . . . . 241 14. . . . .19Schematic for Example (14. . . .5 The view of a large inclination angle from different points in the fluid field. . Notice that both have a maximum point and a “no solution” zone. . . . . . . . . . .15Oblique shock occurs around a cone. . . . . .21Typical examples of unstable and stable situations . .18Schematic for Example (14. . . . . . . . . . . . .4 Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. 243 14. the cone half angle is 15◦ and the Mach number is 2. .16Maximum values of the properties in an oblique shock . . . 264 15. . 254 14. . . . . . . . . . . . . . . . . . . . .3 The pressure ratio as a function of the dimensionless time for choked condition . . . .8 The calculation of D (possible error). . . . . . . 229 14. . . . . . . . . . . . 252 14. . . . . . . Prandtl-Meyer function approaches closer to a zero deflection angle. . It can be noticed that the normal shock. . . . . . . . . . . . . . . . . . . . . . 244 14. . . . . . .0. . . shock angle. . 250 14. . . . . .4 The pressure ratio as a function of the dimensionless time . 228 14. . .13The schematic for a symmetrical suction section with Mach reflection 248 14. . . 224 12. . . . . . . . . 235 14. . . . . . . . . . . . . . . 223 12. . . . . . . . 253 14. . 251 14. . . . . . . . . . . 265 15. . . . . . . . . . . . . . . . . .9 The possible range of solutions for different parameters for given upstream Mach numbers . . . a Research Professor at Cuernavaco University of Mexico. .20Schematic of two angles turn with two weak shocks . . . . . .2 The pressure ratio as a function of the dimensionless time for chokeless condition . .3 A typical oblique shock schematic . . . . . . . .1 A view of a straight normal shock as a limited case for oblique shock 227 14. . . . 236 14. . . . . . . . 247 14. . . . .2 The angles of the Mach line triangle . . . . . . . . . . . which is around zero. . . . . . According to his measurement. . . . . . . . . . . .2. . 266 . . . . . . . . . . . .6 The various coefficients of three different Mach numbers to demonstrate that D is zero . . .11A local and a far view of the oblique shock. . . .12The schematic for a round–tip bullet in a supersonic flow . . . . . . . . . .14The “detached” shock in a complicated configuration sometimes referred to as Mach reflection . However. . . . . . . . . . . . . . . . 265 15. . and exit Mach number for M1 = 3 . . . 239 14. . . . . . . . . . . . . . . . . . . . the strong shock. . . . . . . . . .240 14.7 The Mach waves that are supposed to be generated at zero inclination. . This photo is courtesy of Dr.2 The regions where oblique shock or Prandtl–Meyer function exist. . . . . .10Schematic of finite wedge with zero angle of attack. 251 14. . . . . .1 The definition of the angle for the Prandtl–Meyer function. . . . . .22The schematic of stability analysis for oblique shock.5) . . . . . . 263 14. . . . . . . . and the weak shock coexist. . . . . . . . . . . . . . . . . . . . . . Grigory Toker. . . . . . . . . . . . . . . . . . . . . . . .4) . . . . . . . . . . . 224 14. . .3 The schematic of the turning flow . . . . 245 14. 248 14. . . . . . . . . . . . .

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

xiv LIST OF FIGURES .

. . .4 . . 3. . . . John. . . . . . . 1972 . . . . .3) 120 5. . . . . . . . . .1 The shock wave table for k = 1. 35 36 37 58 59 60 62 63 73 74 5. . . . . . . . . .4) . . . . . . John. . . . . . . . . . . 4. . . . .2 Liquids speed of sound. . . . . . . . . . . . . . Manual of Sound Recording. . . .4 Table for shock propagating from a suddenly opened valve (k=1. . . .1 The shock wave table for k = 1. . . 4. . . . . 116 5.LIST OF TABLES 3.3 Table for shock propagating from suddenly opened valve (k=1. . . . 115 5. . . . .3 Table for shock propagating from suddenly opened valve (k=1.4) . . . . London: Fountain Press. .4 (continue) . . . .4) . . 3. . .2 Table for a Reflective Shock from a suddenly closed end (k=1. . . . .4 . .3) 119 5. . 119 5. . . . . . . 4. . . . . . . .1 The shock wave table for k = 1. . . . . .3 Isothermal Table . . . . .1 Water speed of sound from different sources . . .2 Isentropic Table k = 1. 1972 . .1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . . .4 Table for shock propagating from a suddenly opened valve (k=1. . . . . .3 Solids speed of sound. . . . . .1 Fliegner’s number and other paramters as function of Mach number (continue) . . . . . . . . . 117 5. . . . after Aldred. .4 (continue) . . . . . . . . . . . . . . . . . . .4 (continue) . . . 4. . . . . . . . . . . . . . .3 Isothermal Table (continue) .4 Table for shock propagating from a suddenly opened valve (k=1. . . . . . . . . . . . . . . . . London:Fountain Press. . . . 118 5.4)(continue)118 5. . . . . . . .2 Table for Reflective Shock from suddenly closed valve (end) (k=1. 117 5. . . . Manual of Sound Recording. 4. . . after Aldred. . . . . . . . . . . . . . . . .1 Fliegner’s number and other paramters as function of Mach number 4. 4. . .3) 121 xv . . . .2 Isentropic Table k=1. . . . . . . . . . . . .

. . . 187 Fanno Flow Standard basic Table (continue) .1 Rayleigh Flow k=1. . . . . . . . . 246 . . . .1 Rayleigh Flow k=1. .4 . . . . . . . . . . .1 Maximum values of oblique shock (continue) k=1. . . . . . . 194 10. . . . . . 152 Fanno Flow Standard basic Table . . . . . . . . . . . . . . . . . . . . . .4 (continue) . . . 147 The flow parameters for unchoked flow . . . . . .4 . 193 10. . . . . . . . . . . . . . . . . . . . . 245 14. . . . . . . . . . . .1 LIST OF TABLES The Isothermal Flow basic parameters .1 Table of maximum values of the oblique Shock k=1.4 9. . . . . . . . . . . .4 . . . . . . .4 (continue) . . . . .1 Rayleigh Flow k=1. . . . . . . . . 188 10. .1 8.1 9. . . . . . . . . 195 14.xvi 8. . . . . . . .

Fifth Floor. with or without modifying it. which is a copyleft license designed for free software. Boston. because free software needs free documentation: a free program should xvii . or other functional and useful document ”free” in the sense of freedom: to assure everyone the effective freedom to copy and redistribute it.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. which means that derivative works of the document must themselves be free in the same sense. but changing it is not allowed.2001. while not being considered responsible for modifications made by others. 51 Franklin St. Secondarily. November 2002 Copyright ©2000. either commercially or noncommercially.2.2002 Free Software Foundation. textbook. Inc. This License is a kind of ”copyleft”. GNU Free Documentation License Version 1. this License preserves for the author and publisher a way to get credit for their work. Preamble The purpose of this License is to make a manual. We have designed this License in order to use it for manuals for free software. MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document. It complements the GNU General Public License.

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

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

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

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

forming one section Entitled ”History”. and that you preserve all their Warranty Disclaimers. The combined work need only contain one copy of this License. but you may replace the old one. and list them all as Invariant Sections of your combined work in its license notice. In the combination. provided that you include in the combination all of the Invariant Sections of all of the original documents. 5.xxii LIST OF TABLES You may add a section Entitled ”Endorsements”. or else a unique number. 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. you may not add another. to the end of the list of Cover Texts in the Modified Version. and multiple identical Invariant Sections may be replaced with a single copy. and any sections Entitled ”Dedications”. provided you insert a copy of this License into . likewise combine any sections Entitled ”Acknowledgements”. under the terms defined in section 4 above for modified versions. and a passage of up to 25 words as a Back-Cover Text. you must combine any sections Entitled ”History” in the various original documents. previously added by you or by arrangement made by the same entity you are acting on behalf of. unmodified. COLLECTIONS OF DOCUMENTS You may make a collection consisting of the Document and other documents released under this License. statements of peer review or that the text has been approved by an organization as the authoritative definition of a standard. You may extract a single document from such a collection. provided that you follow the rules of this License for verbatim copying of each of the documents in all other respects. If the Document already includes a cover text for the same cover. 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 title of each such section unique by adding at the end of it. and distribute it individually under this License. on explicit permission from the previous publisher that added the old one. If there are multiple Invariant Sections with the same name but different contents. 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. COMBINING DOCUMENTS You may combine the Document with other documents released under this License. You must delete all sections Entitled ”Endorsements”. in parentheses. 6. You may add a passage of up to five words as a Front-Cover Text. and replace the individual copies of this License in the various documents with a single copy that is included in the collection. the name of the original author or publisher of that section if known.

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

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

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

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

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

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

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

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

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

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

chapters or sections can be added after the skeleton has been written.1 0.0 0.0. Someone has to write the skeleton and hopefully others will add “flesh and skin.0. The undersigned of this document intends to be the organizer/author/coordinator of the projects in the following areas: project name Die Casting Mechanics Statics Dynamics Strength of Material Compressible Flow Fluid Mechanics Thermodynamics Heat Transfer Open Channel Flow Two/Multi phases flow progress alpha not started yet not started yet not started yet not started yet early beta alpha early alpha not started yet not started yet not started yet remarks version 0. while in other cases This may merely be the person who decides what will go into the book and what will not (gate keeper).0.0. But more than that. These books should be considered more as a project than to fit the traditional definition of “plain” books.0.) can be redone by anyone who has the time to do it. better discussions or better explanations are all welcome to these books. These books are written in a similar manner to the open source software process.0. Additionally. the contributions to books can be done by many who are not experts. These data (tables.0 0. It is also hoped that others will contribute to the question and answer sections in the book. .3 0. The organizer of the book in some instances will be the main author of the work.0 0.4 0.0.0 0.0. some errors are possible and expected. Thus. 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). 5 Data are not copyrighted.CREDITS xxxiii adding a question and perhaps the solution. Even if not complete. contributions can be made from any part of the world by those who wish to translate the book. this method is expected to accelerate the creation of these high quality books. Nevertheless. It is hoped that the book will be error-free. other books contain data5 which can be A typeset in L TEX.0. Unlike a regular book. the traditional role of author will be replaced by an organizer who will be the one to compile the book.” In this process. Thus.01 0. Thus.0 0. graphs and etc.0 0. these works will have a version number because they are alive and continuously evolving.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.

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

Menikoff pointed to this inconsistency. Several sections were add to this book with many new ideas for example on the moving shock tables. While it moves earlier but the name was not changed. Nevertheless. 2006 The title of this section is change to reflect that it moved to beginning of the book.A. 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.000 downloaded and is encouraging to receive over 200 thank you eMail (only one from U. While all these requests are important.Prologue For This Book Version 0.4. the ideas of moving shock will reduced the work for many student of compressible flow.3 Sep.000. I am sorry will not be entertaining work for non Linux/Unix systems. Version 0. For example. If one want to use the software engine it is okay and permitted by the license of this work.4./Arizona) and some other reactions.2 It was surprising to find that over 14. Dr. this author cannot add all the things that he was asked and want to the book in instant fashion. the time is limited and they will be inserted as time permitted. The moving shock issues are not completed and more work is needed also in the shock tube. However. and the author is apologizing for this omission. 15. 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). xxxv . one of the reader ask why not one of the example of oblique shock was not turn into the explanation of von Neumann paradox. I also got request from a India to write the interface for Microsoft.S. especially for Microsoft. The download to this mount is over 25.

The only exception is the addition to Taylor–Maccoll flow (axisymmetricale conical flow) in Prandtl -Meyer function (currently in a note form). This assumption drives students from other fields away from this knowledge. Thus. POTTO project books are open content textbooks. 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 time. The chapter about oblique shock was added and it contains the analytical solution. This premise should not be assumed. the connection between Prandtl–Meyer flow and oblique is an note form. This “rejection” is especially true when students feel that they have to go through a “shock wave” in their understanding. However. an individual asked whether this author intended to write about the unsteady section.potto. This knowledge should be spread to other fields because it needed there as well. At this stage. and hopefully it will be near future. The a brief chapter on Prandtl–Meyer flow was added. This led to comparison between maximum to ideal gas model to more general model.xxxvi LIST OF TABLES This textbook has sections which are cutting edge research8 . The reason the topic of Com8 A reader asked this author to examine a paper on Triple Shock Entropy Theorem and Its Consequences by Le Roy F. The other questions that appeared related two phase and connecting several chambers to each other. most readers reached to www. In the introductory chapter. the two phase aspects were added. it reaches to about 30% (2011). Version 0.3) several individuals sent me remarks and suggestions. Furthermore. the questions that appear on the net will guide this author on what is really need to be in a compressible flow book.4 Since the last version (0. The isothermal nozzle was combined with the isentropic chapter. most additions of the next version will be concentrated on Fanno flow and Rayleigh flow. Also. Version 0. The additions of this version focus mainly on the oblique shock and related issues as results of questions and reactions on this topic. This book is the second book in the series of POTTO project books. If the total combined variation search of terms “Fanno” and “Rayleigh” (mostly through google) is accounted. The fifth chapter deals now with normal shock in variable area ducts.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.org by searching for either terms “Rayleigh flow” (107) and “Fanno flow” ((93). In the chapter on speed of sound. several questions were about compressibility factor and two phase flow in Fanno flow and other kind of flow models. The sixth chapter deals with external forces fields. Henderson and Ralph Menikoff. extensive description of the compressible flow history was written.

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

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

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

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

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

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

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

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

These omissions. Further topics related to non-ideal gas such as steam and various freons are in the process of being added to this book especially in relationship to Fanno flow. Specific missing parts from every chapters are discussed below. It is hoped that the style file will be converged to the final form rapidly. One of the virtue of this book lay in the fact that it contains a software that is extensible. This part will be incorporated in the future hopefully with the help of others. For example. Also issues that associated with the book like the software has to be improved. the Fanno module can be extended to include effects of real gases. approach problems are sometime appears in the book under the Meta simple like this sample this part. there are specific issues which are on the “table” and they are described herein.To Do List and Road Map Meta This book is not complete and probably never will be completed. illustration or photo of experiment. several chapters are missing. At this stage. It A is hoped the changes in TEX and L TEX related to this book in future will be minimal and minor. On occasions a footnote was used to point out for a need of improvement. There will always new problems to add or to polish the explanations or include more new materials. You are always welcome to add a new material: problem. mistakes. Material can xlv . Meta End Questions/problems appear as a marginal note. question. Nevertheless. 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. The dimensional analysis of the flow when the flow can be considered as isothermal. 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. Normal Shock Extend the partialy (open/close) moving shock theory. Nozzle The effect of external forces (add problems). Mixed gases and liquids. More problems in relationship to two phase.0 version). The combined effects of isentropic nozzle with heat transfer (especially with relationship to the program.xlvi LIST OF TABLES be further illuminate. . 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. Speed of Sound Discussion about the movement in medium with variation in speed of sound. Speed of sound in wet steam. Additional material can be provided to give a different angle on the issue at hand.

VERSION 4. 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++. The FORTRAN version will not be included.1. 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. Provide discussion on variations of the effecting parameters. Application Marcell–Taylor (from the notes) Examples Transient problem .

xlviii LIST OF TABLES .

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

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

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

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

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

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

The fact that the conditions and direction were known did not bring the solution to the equations. view the work of G. and Mach number was described as impossible to obtain. In fact. Thus. 20 Some add discussion about the general relationships. I. Today this equation/condition is known as Prandtl’s equation or condition (1908). There were until recently (version 0. deflection angle. the supersonic branch is missing from his understanding (see Figure (1. Berlin. Steam and Gas was found by Landau & Lifshitz Turbine and expanded by Kolosnitsyn & Stanyukovich (1984).1)). In fact Prandtl is the one who introduced the name of Rankine-Hugoniot’s conditions not aware of the earlier developments of this condition. 7 Here. the relationship between upstream Mach number and the angles. A solution Rayleigh lines after Stodola. HISTORICAL BACKGROUND conditions to a more general form19 .1.1: The shock as connection of Fanno and munist regimes. No. It was probably later that Stodola (Fanno’s adviser) realized that the shock is the intersection of the Fanno line with the Rayleigh line. ub.3. Taylor from England as the proof (of course utilizing the second law) 21 Theodor Meyer in Mitteil. the second law has been around for over 40 years and yet the significance of it was not was well established. Theodor Meyer (Prandtl’s student) derived the conditions for oblique shock in 190821 as a byproduct of the expansion work. to be add to oblique shock Since early in the 1950s the analytical relationships between the oblique shock. ¨ Forsch-Arb. The probable reason that analytical solution was not published because the claim chapter. 1. shock angle. The “last nail” of understanding was put by Landau. page 62. Yet. Stodola suggested the graphical solution utilizing the Fanno line. a Jewish scientist who worked in Moscow University in the 1960’s during the ComFig. 19 To .3 of this book) several equations that tied various properties/quantities for example. The first full analytical solution connecting the angles with upstream Mach number was published in this book version 0.3. 62. 1908. it took over 50 years for Prandtl to arrive at and to demonstrate that the shock has only one direction20 .

No 5 p. This difference probably results in not recognizing that The boundary conditions play an important role in the stability of the shock24 . His analysis was based on Hamilton’s principle when he ignore the boundary condition.” Proceedings of the National Academy of Science. S. p. part of analysis of this book is also redoing old work. While the actual solution all the roots are real and positive for the attached shock.. It is interesting that transformation result in one of root being negative. New York. The effect of real gases was investigated very early since steam was used move turbines.F.90). Carrier26 was first to recognize that strong and weak shocks stable. similar analysis was published by Briggs. Vol. even this author sinned in this erroneous conclusion.” Quarterly of Applied Mathematics. pp. Emanuel’s partial solution just redone 36 years work (how many times works have to be redone in this field).. In fact analysis based on Hamilton’s principle isn’t suitable for stability because entropy creation was recognized 1955 by Herivel25 . “Comment on Calculation of Oblique shock waves. F. 22 Since writing this book. 2. Vol. “On the Stability of the supersonic Flows Past as a Wedge. Yet. George. The researchers understood that flow after a strong shock was governed by elliptic equation while the flow after a weak shock was governed by hyperbolic equations. 179–210. pp.” published by Interscience Publishers. Pt. pp. but no comparison between real gases and ideal gas is offered there. 6. Henderson and Menikoff27 dealt with only the procedure to find the maximum of oblique shock.. . 23 Epstein. The presentation was missing the condition for the detachment or point where the model collapse. Vol. 17. The first that attempt this question where in 1931 by Epstein23 . G. 1948.8 CHAPTER 1. recently.” Proceedings of the Cambridge philosophical society. Even all books that were published recently claimed that no strong shock was ever observed in flow around cone (Taylor–Maccoll flow). “On the air resistance of Projectiles. That solution is based on a transformation of sin θ to tan β . Inc. 51. ”Triple Shock Entropy Theorem. 1949. 26 Carrier. Hence.” Journal of Fluid Mechanics 366 (1998) pp. 1963. 344-349. 367–378. But more surprisingly. P. 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. INTRODUCTION in the famous report of NACA 1135 that explicit analytical solution isn’t possible22 . 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. 1931. 25 Herivel. 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..4 that it was clear that the Normal shock along with strong shock and weak shock “live” together peacefully and in stable conditions. The results of that analysis was that strong shock is unstable. December 2000 (US$ 124. 317. 1955. J.” AIAA Journal Vol 2. several individuals point out that a solution was found in book “Analytical Fluid Dynamics” by Emanuel. 27 Henderson and Menikoff. It is amazing that early research concluded that only the weak solution is possible or stable as opposed to the reality. For example. In fact. 24 In study this issue this author realized only after examining a colleague experimental Picture 14. J. If fact the confusion on this issue was persistent until now. second edition. “The Derivation of The Equations of Motion On an Ideal Fluid by Hamilton’s Principle. In a way. 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. 974. 532-547.

4. The theoretical first work on this issue was done by Zeuner. 29 Rayleigh was the first to develop the model that bears his name. 1.” Leipzig 1899. . Choking wasn’t clearly to be observed. p. HISTORICAL BACKGROUND 9 The moving shock and shock tube were study even before World War Two. No one was looking for or expecting the choking to occur. there was no realization that the flow is choked just that the flow moves faster than speed of sound.3 of this book (as far it is known. Steam and Gas Turbine operation by another inventor (Curtis) 1896 used in steam turbine.. 1.1. and when it was found the significance of the choking phenomenon was not clear. As a result deLavel’s nozzle was invented by Carl Gustaf Patrik Fig. The introduction of the steam engine and other thermodynamics cycles led to the choking problem. 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. 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. even when researcher stumbled over it. The first to discover (try to explain) the choking phenomenon was Rayleigh29 . The first experimental choking phenomenon was discovered by Fliegner’s experiments which were conducted some time in the middle of 186x28 on air flow through a converging nozzle. 28 Fliegner Schweizer Bauztg. Later. According the thermodynamics theory (various cycles) the larger heat supply for a given temperature difference (larger higher temperature) the larger the output. 68–72. but did not produce any model or conduct successful experimental work.2: The schematic of deLavel’s turbine afde Laval in 1882 and first successful ter Stodola. “Theorie die Turbinen. but after a certain point it did matter (because the steam was choked). It is likely that others had noticed that flow is choked. an explicit analytical solution for shock a head of piston movement (special case of open valve) was originally published in this book for the first time. Vol 31 1898. leading to the study and development of Rayleigh flow.3 Choking Flow The choking problem is almost unique to gas dynamics and has many different forms. this book is first to publish this tables). trial and error method was the only method to solve this problem. Up to this version 0.3. page 268 f. Only after the dimensionless presentation of the problem and the construction of the moving shock table the problem became trivial. Yet.3.

R.. In the literature appeared reference to article by Lorenz in the Physik Zeitshr. The actual results will be determined by the modified Eckert number to which model they are closer. heft 62. of the publications were not named after Prandtl but rather by his students like Meyer. 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. Ing. “Theorie der Turbinen. 119. Vol 12. erschienen in den Mitteilungen uber ¨ Forsch.G. Later Romer at el extended it to isothermal variable area flow (1955). In this book. Vol. Comptes R. Uber ¨ zweidimensionals Bewegungsvordange eines Gases. In 1908 Meyer has extend this work to make two dimensional calculations32 . 33 Parenty. However. Leipzig 1899 page 268 f. this author adapts E. Berlin 1908. Theodor. 8. Recently this concept was proposed (not explicitly) by Dutton and Converdill (1997)35 . 31 Some 30 Zeuner. as if in 1904. 113.10 CHAPTER 1.. Perhaps. Dissertation 1907. there are also other works that this author did not come crossed. Ecert’s idea of dimensionless parameters control which determines where the reality lay between the two extremes. 8 1896. The √ first one to analyzed that choking occurs at 1/ k for isothermal flow was Shapiro (195x). 116. Vol.-Arb. . in many cases the reality is somewhere between the adiabatic and the isothermal flow. INTRODUCTION After the introduction of the deLavel’s converging–diverging nozzle theoretical work was started by Zeuner30 .-Wes. 35 These researchers demonstrate results between two extremes and actually proposed this idea. Namely. It is so strange that a giant like Shapiro did not realize his model on isothermal contradict his conclusion from his own famous paper. Later continue by Prandtl’s group31 starting 1904. Phys. 32 Meyer. Th. It was commonly believed34 that the choking occurs only at M = 1. Paris. Chim. Ann. 1897. that the presentation here suggests that topic should be presented case between two extremes. Experimental work by Parenty33 and others measured the pressure along the converging-diverging nozzle.

A different reason to study the converging-diverging nozzle was the Venturi meter which was used in measuring the flow rate of gases.3: The measured pressure in a nozzle taken from Stodola 1927 Steam and Gas Turbines and 1947-48. Between the turn of the century Fig. At the end of the arm was the object that was under observation and study. the spinning arm was no longer used in testing. several models that explained choking at the throat have been built.545) that creates the maximum flow rate. 37. Yet. 1907. As a result. HISTORICAL BACKGROUND Nozzle flow 11 The first “wind tunnel” was not a tunnel but a rotating arm attached at the center. The arm’s circular motion could reach a velocity above the speed of sound at its end. Bendemann 36 carried experiments to study the accuracy of these flow meters and he measured and refound that the flow reaches a critical value (pressure ratio of 0. due to the circular motion. . 36 Bendemann Mitteil uber ¨ Forschungsarbeiten. Berlin.1.3. No. when the first supersonic wind tunnel was built. There are two main models or extremes that describe the flow in the nozzle: isothermal and adiabatic. 1. in 1904 the Wright brothers demonstrated that results from the wind tunnel and spinning arm are different.

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

a shock wave occurs which increases the resistance. The friction factor is the main component in the analysis as Darcy f 41 had already proposed in 1845. the Fanno model wasn’t able to produce a prediction useful for the industry. Fanno attributes the main pressure reduction to friction. the question cannot be answered yet. 1. The author invites others to help in this information.3. Thus. was ever produced to verify this finding. 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. Without the existence of the friction factor data. but can be explained by choking at a lower Mach number. In reality. which built on Hunter Rouse’s (194x) work made Darcy– Weisbach’s equation universally useful. However at a Mach number above 0. If it turned out that no one had done it before Shapiro. flow that is dominantly adiabatic could be simplified and analyzed.8 (relative of velocity of the body to upstream velocity) a local Mach number (local velocity) can reach M = 1. But because the actual Fanno’s thesis is not available.4 External flow When the flow over an external body is about . At that stage. No experimental evidence. 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 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.). The arrival of the Moody diagram. known by the undersigned. 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. f based radius is only one quarter of the Darcy f which is based on diameter . HISTORICAL BACKGROUND 13 Mumenthaler from UTH University.1. this flow model should be called Shapiro’s flow. no copy of the thesis can be found in the original University and perhaps only in the personal custody of the Fanno family40 . This fact was evident in industrial applications where the expectation of the choking is at Mach one. The Navier-Stokes equations which describe the flow (or even 41 Fanning 40 This material is very important and someone should find it and make it available to researchers.8 Mach or more the flow must be considered to be a compressible flow. Shapiro organized all the material in a coherent way and made this model useful.3. Additionally an understating of the supersonic branch of the flow was unknown (The idea of shock in tube was not raised at that time.

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

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

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

Later. but he declined.3. In 1863. he named it Mach’s Principle.R. Mach’s revolutionary experiment demonstrated the existence of the shock wave as shown in Figure 1. and in 1918. Mach and E. There Mach wrote his thesis ”On Electrical Discharge and Induction. His experiments required exact timing.1. was a high school teacher who taught Ernst at home until he was 14. Johann. 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. physics and philosophy. 1. before he entered the university of Vienna were he studies mathematics. . He graduated from Vienna in 1860. This was one of the primary sources of inspiration for Einstein’s theory of General Relativity. Einstein was greatly influenced by it.” Mach was Fig. Ernest Mach (1836 . It is amazing that Mach was able to photograph the phenomenon using the spinning arm technique (no wind tunnel was available at that time and most definitely nothing that Fig. Mach’s father. Note it was not taken in a wind tunnel sonic speeds.G. He then turned to physics. HISTORICAL BACKGROUND Ernest Mach (1838-1916) 17 Ernst Mach was born in 1838 in Chrlice (now part of Brno). He was not able to attach the camera to the arm and utilize the remote control (not existent at that time). 1.6: Photo of Ernest Mach interested also in physiology of sensory perception. and in 1867 he received a position in the Technical University in Prague47 where he taught experimental physics for the next 28 years. Mach was also a great thinker/philosopher and influenced the theory of relativity dealing with frame of reference. when he studied in Kromeriz Gymnasium.7: The Photo of the bullet in a supersonic flow that could take a photo at superMach made. Eckert.7. when Czechia was still a part of the Austro–Hungary empire. Mach’s shadowgraph 47 It is interesting to point out that Prague provided us two of the top influential researchers[:] E.1916) published Die Machanik in which he formalized this argument.

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

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

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

with his student Theodor Meyer. turbulence mixing theories etc. During this time Prandtl developed his boundary layer theory and studied supersonic fluid flows through nozzles. In 1904. As a result. he was known worldwide as the leader in the science of fluid dynamics. His work and achievements in fluid dynamics resulted in equations that simplified . Germany (1901).3. His interest changed when. Bavaria. developed the first theory for calculating the properties of shock and expansion waves in supersonic flow in 1908 (two chapters in this book). As a byproduct they produced the theory for oblique shock. He became the director of the ¨ Institute for Technical Physics at the University of Gottingen. in 1874. ¨ Ludwig Prandtl worked at Gottingen until his death on August 15.1. In 1925 Prandtl became the director ¨ of the Kaiser Wilhelm Institute for Flow Investigation at Gottingen. Prandtl started his studies at the age of 20 in Munich. the young Ludwig spent more time with his father which made him interested in his father’s physics and maFig. Germany and he graduated at the age of 26 with a Ph.D. 1. HISTORICAL BACKGROUND Ludwig Prandtl 21 Perhaps Prandtl’s greatest achievement was his ability to produce so many great scientists. This upbringing fostered the young Prandtl’s interest in science and experimentation. Ludwig Prandtl was born in Freising. the paper which describes his boundary layer theory. There is no one who educated as many great scientists as Prandtl.D. he was required to design factory equipment that involved problems related to the field of fluid mechanics (a suction device).11: Photo of Prandtl chinery books. in his first job. His father was a professor of engineering and his mother suffered from a lengthy illness. 1953. his Ph. was focused on solid mechanics. His 1904 paper raised Prandtl’s prestige. Prandtl also contributed to research in many areas. Interestingly. He developed the Prandtl-Glauert rule for subsonic airflow. Prandtl changed the field of fluid mechanics and is called the modern father of fluid mechanics because of his introduction of boundary layer. By the 1930s. Prandtl. he presented the revolutionary paper “Flussigkeitsbewegung Bei Sehr Kleiner Reibung” (Fluid Flow in Very Little Friction). Later he sought and found a job as a professor of mechanics at a technical school in Hannover. such as meteorology and structural mechanics. It is mind boggling to look at the long list of those who were his students and colleagues.

high speed flow (correction formula for subsonic compressible flows). During World War II. 1. Therefore many referred to him as ¨ the father of modern fluid mechanics. and served as a consultant to the U. wing-fuselage. Air Force and the National Advisory Committee for Aeronautics before coming to Minnesota. Hence. wing-propeller. He emigrated to the United States after the war. Prandtl’s other contributions include: the introduction of the Prandtl number in fluid mechanics. etc).S. airfoils and wing theory (including theories of aerodynamic interference.22 CHAPTER 1.Fig. and many are still used today. 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. His book ”Introduction to the Transfer of Heat and Mass.R. Eckert developed the under. biplane. he developed methods for jet engine turbine blade cooling at a research laboratory in Prague. which is associated with the Mach number. Ludwig Prandtl died in Gottingen. Germany on August 15th 1953.12: The photo of Ernst Rudolf George Eckert with the author’s family standing of heat dissipation in relation to kinetic energy. especially in compressible flow. fundamental studies in the wind tunnel. INTRODUCTION understanding.G. In addition to being named to the National Academy of Engineering in 1970. theory of turbulence. where he studied at the German Institute of Technology. He authored more than 500 articles and received several medals for his contributions to science. . Eckert Eckert was born in 1904 in Prague. Schlichting suggested this dimensionless group in honor of Eckert.” published in 1937. the dimensionless group has been designated as the Eckert number. is still considered a fundamental text in the field.

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

INTRODUCTION .24 CHAPTER 1.

The mass after a very short time of dt is dm. Several application of the fluid mechanics will demonstrated. At the initial time the mass that was in the tube was m0 .3 2. In addition the basic Bernoulli’s equation will be derived for incompressible flow and later for compressible flow. 25 .CHAPTER 2 Fundamentals of Basic Fluid Mechanics 2.1 Introduction This chapter is a review of the fundamentals that the student is expected to know. For simplicity. 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 terms will be reviewed such as stream lines. This material is not covered in the history chapter. The basic principles are related to the basic conservation principle.2 2. The flow through a stream tube is assumed to be one-dimensional so that there isn’t any flow except at the tube opening.4 Fluid Properties Control Volume Reynold’s Transport Theorem For simplification the discussion will be focused on one dimensional control volume and it will be generalzed later. 2. it is assumed the control volume is a fixed boundary.

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

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

To answer this question consider a piston moving from the left to the right at a relatively small velocity (see Figure 3. the pulse will be infinitesimally small.28 CHAPTER 3. The ears c sense the variations by frequency P+dP P ρ ρ+dρ and magnitude which are transferred to the brain which translates to voice. P is a function of two independent variables.4) yields c2 dρ ρ = dP dP =⇒ c2 = ρ dρ (3. it raises the question: what is the speed of the Fig. Thus. Here.3) Substituting the expression for dU from equation (3.5). Applying the mass balance P+dP yields P ρ+dρ ρ ρc = (ρ + dρ)(c − dU ) (3. 3.2) From the energy equation (Bernoulli’s equation).1) Fig.1: A very slow moving piston in a still gas small disturbance travel in a “quiet” medium. For an ideal gas. the pressure and density can be assumed to be continuous.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. SPEED OF SOUND The people had recognized for several hundred years that sound is sound wave dU velocity=dU a variation of pressure.2) into equation (3. it is considered . assuming isentropic flow and neglecting the gravity results (c − dU )2 − c2 dP + =0 2 ρ neglecting second term (dU 2 ) yield −cdU + dP =0 ρ (3.” It is assumed that if the velocity of the piston is infinitesimally small.5) An expression is needed to represent the right hand side of equation (3.1). 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 information that the piston is moving passes thorough a single “pressure pulse. 3. This velocity is referred to as the speed of sound.4) (3. Thus.

Example 3.7) s Note that the equation (3. therefore it can be written ∂P dP = dρ ∂ρ (3.8) (3.2) is P F R cs U (ρU dA) (P + dP ) − P = (ρ + dρ)(c − dU )2 − ρc2 Neglecting all the relative small terms results in : ∼ 0 ∼0   : dU 2 dP = (ρ + dρ) c2 −  2cdU +  − ρc2 (3.1: Demonstrate that equation (3.10) 3. The pressure for an ideal gas can be expressed as a simple function of density.3.5) can be obtained by utilizing the momentum equation instead of the energy equation. SPEED OF SOUND IN IDEAL AND PERFECT GASES 29 that P = P (ρ.11) . k namely P = constant × ρk (3. (3. s) where s is the entropy. S OLUTION The momentum equation written for the control volume shown in Figure (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.9) dP = c2 dρ This yields the same equation as (3.6) In the derivations for the speed of sound it was assumed that the flow is isentropic.5). ρ.5) can be derived from the momentum equation.3. The full differential of the pressure can be expressed as follows: dP = ∂P ∂ρ dρ + s ∂P ∂s ds ρ (3. and a function “molecular structure” or ratio of specific heats.

13216 kg m3 kg m3 kg m3 After interpretation of the temperature: kJ At 18[bar] and 335.7◦C: s ∼ 6. SPEED OF SOUND P c= dP constant × ρk = k × constant × ρk−1 = k × dρ ρ =k× P ρ (3.61376 ρ = 6.9563 K kg ρ ∼ 6.5 sec Note that a better approximation can be done with a steam table.12) Remember that P/ρ is defined for an ideal gas as RT .94199 and substituting into the equation yields c= m 200000 = 780.14) s=constant ρ = 6. (a) utilizes the steam table (b) assuming ideal gas.32823 sec kg m3 (3. S OLUTION The solution can be estimated by using the data from steam table3 c= At 20[bar] and 350◦ C: s = 6.5 0.0100 At 18[bar] and 300◦ C: s = 6. and equation (3.327 × 461 × (350 + 273) ∼ 771.30 and hence CHAPTER 3.46956 ρ = 7. Classical Thermodynamics. table A 8.) √ m c = kRT ∼ 1. and it will be part of the future program (potto–GDC). 3 This data is taken form Van Wylen and Sontag “Fundamentals of Classical Thermodynamics” 2nd edition .9563 At 18[bar] and 350◦ C: s = 7.8226 ∆P ∆ρ kJ K kg kJ K kg kJ K kg (3.12) can be written as √ (3.15) for ideal gas assumption (data taken from Van Wylen and Sontag.2: Calculate the speed of sound in water vapor at 20[bar] and 350◦ C.13) c = kRT Example 3.

x T = (TB − TA ) + TA h Where the distance x is the variable distance. Recalling from thermodynamics.18) (3. 3. What is the time it take for sound to travel from point “A” to point “B” under this assumption.19) The speed of sound of any gas is provided by equation (3.4.3.7). The distance between “A” and “B” is denoted as h.4 Speed of Sound in Real Gas The ideal gas model can be improved by introducing the compressibility factor.17) This correction factor approaches one when TB −→ TA . The compressibility factor represents the deviation from the ideal gas. It should be noted that velocity is provided as a function of the distance and not the time (another reverse problem).? S OLUTION The temperature is denoted at “A” as TA and temperature in “B” is TB .16) For assumption of constant temperature the time is t= √ Hence the correction factor tcorrected = t TA TA 2 ¯ 3 (TB − TA ) T TB TA −1 (3. 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. a real gas equation can be expressed in many cases as P = zρRT (3.19) some mathematical expressions are needed. To obtain the expression for a gas that obeys the law expressed by (3. the Gibbs function (3. Thus.20) . SPEED OF SOUND IN REAL GAS 31 Example 3.3: The temperature in the atmosphere can be assumed to be a linear function of the height for some distances.

.22) ρ From thermodynamics.20) The definition of pressure specific heat for a pure substance is Cp = ∂h ∂T =T P ∂s ∂T (3.3: The Compressibility Chart is used to obtain T ds = dh − dP ρ (3. perhaps to insert the discussion here. 3.23) P Van Wylen p. SPEED OF SOUND Fig. 372 SI version.21) P The definition of volumetric specific heat for a pure substance is Cv = ∂u ∂T =T ρ ∂s ∂T (3.32 CHAPTER 3. it can be shown 4 dh = Cp dT + v − T 4 See ∂v ∂T (3.

it is more convenient to express dT /T in terms of Cv and dρ/ρ as follows dρ R dT z+T = T ρ Cv ∂z ∂T (3. However.29) results in dρ R z+T ρ Cv ∂z ∂T = ρ dP R z+T P Cp ∂z ∂T (3.24) into equation (3.23) results  v  v  z T Simplifying equation (3.4.31) .29) ρ Equating the right hand side of equations (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) can be integrated by parts.28) P Equation (3.27) Letting ds = 0 for isentropic process results in dP R dT = z+T T P Cp ∂z ∂T (3. SPEED OF SOUND IN REAL GAS The specific volumetric is the inverse of the density as v = zRT /P and thus ∂v ∂T ∂ zRT P 33 = P ∂T P RT = P ∂z ∂T P > zR ∂T +  P ∂T P  1 (3.25) P dP = Cp dT − T z (3.26) P Utilizing Gibbs equation (3.24) Substituting the equation (3.30) P Rearranging equation (3.30) yields dρ dP Cv = ρ P Cp z+T z+T ∂z ∂T P ∂z ∂T ρ (3.25) to became dh = Cp dT − Tv z ∂z ∂T   dh = Cp dT + v − T   RT    P ∂z ∂T + P  zR    dP P  ∂z ∂T dP ρ (3.28) and (3.3.

SPEED OF SOUND If the terms in the braces are constant in the range under interest in this study. For example. The integration of equation (3. n is defined as k n= Cp Cv z+T z+T ∂z ∂T ρ ∂z ∂T P (3. a change in temperature can have a dramatical change in the speed of sound.7[m/sec] The correction factor for air under normal conditions (atmospheric conditions or even increased pressure) is minimal on the speed of sound. . This relationship (3. the speed of sound for a real gas can be obtained in the same manner as for an ideal gas.995times287 × 300 = 346. n = 1. Now. The specific heat for air is k = 1.403 has √ √ c = znRT = 1. For short hand writing convenience. z = 0.5) to calculated by ideal gas model.32) Note that n approaches k when z → 1 and when z is constant.31) can be integrated. S OLUTION According to the ideal gas model the speed of sound should be √ √ c = kRT = 1. But the definition of n in equation (3.3 1.995.3 and n ∼ 1 which means that speed of sound is only 0. Make the calculation based on the ideal gas model and compare these calculations to real gas model (compressibility factor).403 × 0.33) is similar to equation (3.407 × 287 × 300 ∼ 348.1[m/sec] For the real gas first coefficient n = 1. and in–fact any thermodynamics book shows this relationship. equation (3.4 factor (0. the compressibility factor. at relative moderate pressure but low temperature common in atmosphere.34) Example 3.33) Equation (3. dP = nzRT dρ (3. However.31) yields ρ1 ρ2 n = P1 P2 (3.4: Calculate the speed of sound of air at 30◦ C and atmospheric pressure ∼ 1[bar].407.34 CHAPTER 3. What is different in these derivations is that a relationship between coefficient n and k was established. and z = 0.11). Assume that R = 287[j/kg/K ].403.33) isn’t new.32) provides a tool to estimate n.

the pressure is about 4 × 107 N/m2 . 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]. New York: Wiley. 1997: 468. Johnson.2 × 109 N/m2 = 1493m/s 1000kg/m3 elastic property = inertial property B ρ (3. According . the average bulk modulus for water is 2. Physics. 1999. The World Book Encyclopedia.. The compressibility of the substance is the reciprocal of the bulk modulus.000 meters.3. Chicago: World Book. as can be observed from the equation (3. Ohio: Chemical Rubber Co. 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 4.5). For example. 601 Handbook of Chemistry and Physics. The mathematical definition of bulk modulus as following B=ρ In physical terms can be written as c= For example for water c= 2.2 × 109 N/m2 .37).5) Remark Fresh Water (20 ◦ C) Distilled Water at (25 ◦ C) Water distilled reference Cutnell.35) This agrees well with the measured speed of sound in water. The fractional volume change is only about 1.1: Water speed of sound from different sources The effect of impurity and temperature is relatively large. The amount of compression of almost all liquids is seen to be very small as given in Table (3. 1967-1968: E37 Value [m/sec] 1492 1496 1494 Table 3.36) dP dρ (3. and for purposes of comparison it is given in Table (3. & Kenneth W. SPEED OF SOUND IN ALMOST INCOMPRESSIBLE LIQUID 35 3. John D.5.5 Speed of Sound in Almost Incompressible Liquid Even liquid normally is assumed to be incompressible in reality has a small and important compressible aspect. Many researchers have looked at this velocity.8% even under this pressure nevertheless it is a change. For example. 1482 m/s at 20◦ C.

after Aldred. gives a sound speed for structural steel of E = ρ 160 × 109 N/m2 = 4512m/s 7860Kg/m3 c= (3.2: Liquids speed of sound. the speed of sound in solids is larger than in liquids and definitely larger than in gases. with different speeds in different directions. SPEED OF SOUND to Wilson5 . salinity. Soc.14 is about clean/pure water. . P ) = c0 + cT + cS + cP + cST P . 1972 In summary.6 Speed of Sound in Solids The situation with solids is considerably more complicated. 1960. London: Fountain Press. the speed of sound in liquids is about 3 to 5 relative to the speed of sound in gases. (3.37) where c0 = 1449. 5 J. Amer. using a general tabulated value for the bulk modulus. cT is a function temperature. Speed of sound in solid of steel. Wilson’s formula is accepted by the National Oceanographic Data Center (NODC) USA for computer processing of hydrological information. material Glycerol Sea water Mercury Kerosene Methyl alcohol Carbon tetrachloride reference 25 ◦ C Value [m/sec] 1904 1533 1450 1324 1143 926 Table 3.38) Compared to one tabulated value the example values for stainless steel lays between the speed for longitudinal and transverse waves.32.36 CHAPTER 3. 3. and differences between transverse and longitudinal waves. Wilson’s empirical formula appears as follows: c(S. and cS is a function salinity. T. 1357. vol. the speed of sound in sea water depends on temperature. Nevertheless. John. Acoust. Manual of Sound Recording. N 10. and hydrostatic pressure. and cST P is a correction factor between coupling of the different parameters. in different kinds of geometries. Young’s Modulus for a representative value for the bulk modulus for steel is 160 109 N /m2 . p. cP is a function pressure..

7. three extreme cases suggest themselves: the flow is mostly gas with drops of the other phase (liquid or solid). for homogeneous and under certain condition a single velocity can be considered.7 Sound Speed in Two Phase Medium The gas flow in many industrial situations contains other particles. For small value of ξ equation (3. However.40) (3. In actuality. after Aldred. For simplicity. 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. There can be several models that approached this problem.3.39) (3. London:Fountain Press. about equal parts of gas and the liquid phase. Manual of Sound Recording. Indeed there is double chocking phenomenon in two phase flow.40) can be approximated as ρ =1+m ρa (3. In that case. The equation of state for the gas can be written as Pa = ρa RTa The average density can be expressed as ξ 1−ξ 1 = + ρm ρa ρb ˙b where ξ = m m ˙ is the mass ratio of the materials.3: Solids speed of sound. and liquid with some bubbles. It further assumed that no heat and mass transfer occurs between the particles. The first case is analyzed. John. there could be more than one speed of sound for two phase flow. 1972 3. Topic for none homogeneous mixing are beyond the scope of this book.41) .

Since the equations are the same as before hence the familiar equation for speed of sound can be applied as c= Insert example with small steel particles with air up to 20% γRmix T (3.48) At this stage the other models for two phase are left for next version (0.44) Therefore. The gas density can be replaced by equation (3.46) T = constant (3.41) R P = T ρ 1+m (3.6). Hence. this analysis results in lower speed of sound compared to pure gas.46) into γ= Cp + mC Cv + mC (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. It must noticed that m = constant.43) Assuming that partial pressure of the particles is constant and applying the second law for the mixture yields droplets gas 0 = mC dP (Cp + mC )dT dP dT dT + Cp −R = −R T T P T P γ −1 γ (3. It can be noticed that Rmix and γ are smaller than similar variables in a pure gas.45) In a way the definition of γ was so chosen that effective specific pressure heat C +mC +mC and effective specific volumetric heat are p and Cv respectively.38 m ˙b m ˙a CHAPTER 3. Generally.39) and substituted into equation (3. the mixture isentropic relationship can be expressed as P where γ−1 R = γ Cp + mC Recalling that R = Cp − Cv reduces equation (3. SPEED OF SOUND where m = is mass flow rate per gas flow rate. The 1+m 1+m correction factors for the specific heat is not linear. . 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 (3.47) (3. For example. the velocity of sound in slightly wed steam can be about one third of the pure steam speed of sound. the velocity of mixtures with large gas component is smaller of the pure gas.

SOUND SPEED IN TWO PHASE MEDIUM Meta 39 For a mixture of two phases. speed of sound can be expressed as c2 = where X is defined as X= s − sf (PB ) sf g (PB ) (3.49) Meta End .7.3.50) ∂P ∂P [f (X )] = ∂ρ ∂ρ (3.

40

CHAPTER 3. SPEED OF SOUND

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

PB = P 0 P P0
Subsonic M <1

Supersonic
M >1

4.1 Stagnation State for Ideal Gas Model
4.1.1 General Relationship

distance, x

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

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

42

CHAPTER 4. ISENTROPIC FLOW

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

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

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

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

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

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

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

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

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

4.1. STAGNATION STATE FOR IDEAL GAS MODEL
 

43

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

˙ = Cp (T0 B − T0 A )m Q ˙

(4.10)

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

=

1+

k−1 2 M 2

k k−1

(4.11)

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

=

1+

k−1 2 M 2

1 k−1

(4.12)

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

P∗ = P0

2 k+1

k k−1

(4.14)

ρ∗ = ρ0

2 k+1

1 k−1

(4.15)

44

CHAPTER 4. ISENTROPIC FLOW

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

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

4.1.2

Relationships for Small Mach Number

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

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

4.2. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 45

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

(4.18)

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

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

(4.19)

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

M∗ =

1−

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

1+

M2 +··· 4

(4.21)

1−

kM 2 +··· 4

(4.22)

1+

k−1 2 M +··· 4

(4.23)

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

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

(4.24)

4.2

Isentropic Converging-Diverging Flow in Cross Section

Therefore. Thus. knowing the Mach number or the temperature will provide all that is needed to find the other properties.26) The thermodynamic relationship between the properties can be expressed as T ds = dh − dP ρ (4. T0 . the stagnation density is constant through the flow. The only properties that need to be connected are the cross section area and the Mach number. T ρ P U T+dT ρ+dρ P+dP U+dU 4. it is constant for the other. equation (4. ISENTROPIC FLOW The important sub case in this chapter is the flow in a converging–diverging nozzle. in mathematical terms. is constant through the adiabatic flow because there isn’t heat transfer. and dividing by the continuity equation reads dρ dA dU + + =0 ρ A U (4. the energy equation. Second is isentropic and isothermal Fig.4: Control volume inside a convergingmodel. the stagnation temdiverging nozzle. There are two models that assume variable area flow: First is isentropic and adiabatic model.25) with (4.4). The control volume is shown in Figure (4. reads dh + U dU = 0 Differentiation of continuity equation.46 CHAPTER 4. the stagnation pressure is also constant through the flow because the flow isentropic. 4.27) For isentropic process ds ≡ 0 and combining equations (4.11) are the same. In the same argument.27) yields dP + U dU = 0 ρ (4. Examination of the relation between properties can then be carried out. perature.28) Differentiation of the equation state (perfect gas).29) . P = ρRT .25) hen there is no external work and heat transfer. If the right hand side is constant for one variable.9) and equation (4. and dividing the results by the equation of state (ρRT ) yields dP dρ dT = + P ρ T (4.1 The Properties in the Adiabatic Nozzle (4. Clearly.2. ρAU = m ˙ = constant. Conversely.

33) to obtain a variables separation form of dP = dA ρU 2 A 1 − M2 (4.31) = U2 dA A (4. If Mach number is smaller than .28) reads dU U dP dA dρ − U2 + =0 ρ A ρ (4.4.34) The pressure Mach number relationship Before going further in the mathematical derivation it is worth looking at the physical meaning of equation (4. can be replaced by the static pressure.26) and using it in equation (4.33) 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.32) Equation (4. It can be observed that the critical Mach number is one.31) to obtain dP ρ Or in a dimensionless form dP dA 1 − M2 = U2 ρ A (4.2. dP/ρ yields  1 c2 dP = U2 ρ dA dρ dP + A ρ dP Recalling that dP/dρ = c2 and substitute the speed of sound into equation (4.33) is a differential equation for the pressure as a function of the cross section area. It is convenient to rearrange equation (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 47 Obtaining an expression for dU/U from the mass balance equation (4. it can be observed that dP can be positive or negative depending on the dA and Mach number.30) Rearranging equation (4.34). If the Mach number is larger than one than dP has opposite sign of dA. Now. The term ρU 2 /A is always positive (because all the three terms can be only positive).30) so that the density. dρ.

it is possible that dM = 0 thus the diverging side is in the subsonic branch and the flow isn’t choked. ISENTROPIC FLOW one dP and dA have the same sign. Thus.It must also be noted that when M = 1 occurs only when dA = 0. From the speed of sound.48 CHAPTER 4. The second law (isentropic relationship) dictates that ds = 0 and from thermodynamics dP dT −R ds = 0 = Cp T P . 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. one can observe that the density. dU = − dP PU (4. dA > 0 ⇒ dP < 0 dA < 0 ⇒ dP > 0 This behavior is opposite to incompressible flow behavior.35 . Hence the pressure increases when the velocity decreases and vice versa. Since physically dP can increase only in a finite amount it must that dA = 0. the phenomenon is different.36). ρ. the opposite. Conversely. the equation of state was not used.36). An increase in area results in an increase of the static pressure (converting the dynamic pressure to a static pressure). For the special case of M = 1 (sonic flow) the value of the term 1 − M 2 = 0 thus mathematically dP → ∞ or dA = 0. if the area decreases (as a function of x) the pressure decreases. For the supersonic branch M > 1. the equations are applicable for any gas (perfect or imperfect gas).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. However. dρ = 1 dP c2 (4. In that case. The relationship between the velocity and the pressure can be observed from equation (4. For M > 1 the term 1/1 − M 2 is negative and change the character of the equation. increases with pressure and vice versa (see equation 4.36) It can be noted that in the derivations of the above equations (4.35) From equation (4. not necessarily means that when dA = 0 that M = 1.28) by solving it for dU .4.

second law (4.38) and simplifying equation (4.37) and becomes (k − 1)M dM (k − 1) dP =− 1 2 k P 1 + k− 2 M Combining equation (4.33) yields 1 k ρU 2 dA A 1−M 2 (4.43) momentum equation is not used normally in isentropic process. First stage equation (4.4.26)1 . Note. why? .41) into 2 P P (4.39) is combined with equation (4. ρU 2 = kM P can be proved as M2 U2 U2 kM P = k 2 ρRT = k ρRT = ρU 2 c kRT Using the identity in equation (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 49 and for perfect gas dT k − 1 dP = (4.33) isn’t the solution but demonstration of certain properties on the pressure.37) and the relationship of pressure and cross section area (4.9).40) with equation (4.39). the temperature varies according to the same way that pressure does. 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 = 0.33).42) M2 − 1 dA = dM 1 2 A M 1 + k− 2 M 1 The (4.42) changes equation (4.2.38) yields dT (k − 1)M dM =− 1 2 T 1 + k− 2 M Relationship Between the Mach Number and Cross Section Area The equations used in the solution are energy (4. Differentiation of equation (4. mass (4. the relationship between the temperature and the stagnation temperature becomes dT0 = 0 = dT 1+ k−1 2 M 2 + T (k − 1)M dM (4.39) (4. equation (4.37).37) T k P Thus.41) The following identify.40) P =− M dM 1 2 1 + k− 2 M (4. state (4. The relationship between temperature and the cross section area can be obtained by utilizing the relationship between the pressure and temperature (4.29).

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

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

6 P P0 5[M P a] From Table (4.88639 0. Also.2) or from Figure (4. Assume that the specific heat ratio k = 1.076K ρ0 ρ= ρ P0 5 × 106 [P a] = 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 51 the temperature.0115 0.60000 0.5[Bar].4 × 287 × 294 = 304[m/sec] (4. Example 4.2. 4 Well. there is no known way for the author to directly measure the Mach number. In the point where the static pressure known ¯ = P = 3[M P a] = 0. Here it is subsonic and this technique is not suitable.3) or utilizing the enclosed program from PottoGDC.4 and assume a perfect gas model.69428 1.26 × 10−5 [m3 ] ρU kRT = 0.88638317 × √ 1.86420338 × (273 + 21) = 254. and velocity can be calculated. Calculate the Mach number at point B under the isentropic flow assumption.69428839 × J ρ0 RT0 × 294[K ] 287.1416 The velocity at that point is c U =M √ The tube area can be obtained from the mass conservation as A= m ˙ = 8. T = 0.60693 0. pressure is about two atmospheres with temperature of 250[K ] this question is for academic purposes.2: The Mach number at point A on tube is measured to be M = 23 and the static pressure is 2[Bar]4 .4. Finally.53105 With these values the static temperature and the density can be calculated. The best approximation is by using inserted cone for supersonic flow and measure the oblique shock. 4 This . the cross section can be calculated with all these information.44) For a circular tube the diameter is about 1[cm]. Downstream at point B the pressure was measured to be 1. estimate the temperature at point B.86420 0.0 kgK kg m3 = 41. or simply using the equations shows that M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.

The stagnation temperature can be “bypassed” to calculate the temperature at point B M =2 M =1.53884934.2) or from Potto-GDC two solutions can be obtained. .17040879. At M = 2 (supersonic flow) the ratios are M 2. S OLUTION To obtain the Mach number at point B by finding the ratio of the area to the critical area.8137788 and TB = B 0. At point “A” the cross section area is 50 [cm2 ] and the Mach number was measured to be 0. At point B in the duct the cross section area is 40 [cm2 ].17040604 PB 1. Assume that the flow is isentropic and the gas specific heat ratio is 1. This relationship can be obtained by AB AB AA 40 = × ∗ = × A∗ AA A 50 from the Table 4.272112 B With the value of A A∗ from the Table (4.2 @ M = 2 PA = P0 PB P0 × PA 2. The supersonic branch solution is possible only if there where a transition at throat where M=1. TB = T A × T0 × TA TB T0 = 250[K ] × 1 × 0. The two possible solutions: the first supersonic M = 1.3: Gas flows through a converging–diverging duct.2 1.23005 1.12780453 × = 0.42[K ] Example 4. Find the Mach number at point B..6875 0.5 The corresponding Mach number for this pressure ratio is 1.55556 0.59014 = 1. Both solution are possible and acceptable.6265306 and second subsonic M = 0. the stagnation pressure at point A is known and stagnation temperature can be calculated.4.21567 0.55555556 271.81. Therefore. 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.60315132 P P0 = 0.0 = 0.52 CHAPTER 4.4.12780 0.60315132 0.0000 T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 0.59309 With this information the pressure at point B can be expressed as from the table 4.

65396 0. Hence equation (4.e.49) Equation (4.46) Expressing the temperature in terms of Mach number in equation (4.k) m ˙ P P0 U √ = A P0 kRT k R T0 1 P √ = √0 M T T0 T0 k P R P0 T0 T (4.47) holds everywhere in the converging-diverging duct and this statement also true for the throat. dividing equations (4.28772 0.2.22617 0.6266 0.47) It can be noted that 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. pressure) do not change.46) results in m ˙ = A kM P0 √ kRT0 1+ k−1 2 M 2 k+1 − 2( k−1) (4.45) This parameter is studied here.0440 4. The area ratio as a function of the Mach number needed to be established.4.34585 1.86838 1.2.48) = 1 + ∗ A 2 RT0 Since the mass flow rate is constant in the duct.82071 1. The area ratio is defined as the ratio of the cross section at any point to the throat area (the narrow area).47) yields 1 A = ∗ A M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (4. to examine the maximum flow rate and to see what is the effect of the compressibility on the flow rate. specifically and explicitly the relationship for the chocked flow. It is convenient to rearrange the equation (4.47) obtained the form √ k+1 − 2( k−1) m ˙ kP0 k−1 √ (4.2721 0.45) to be expressed in terms of the stagnation properties as f (M.48) by equation (4.53887 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 53 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 1.94511 0.2721 0. The throat area can be denoted as by A∗ .49) relates the Mach number at any point to the cross section area ratio.3 Mass Flow Rate (Number) One of the important engineering parameters is the mass flow rate which for ideal gas is m ˙ = ρU A = P UA RT (4. temperature. .

While these . c0 √ m ˙ kRT0 mc ˙ 0 m ˙ T0 1 √ =√ =√ ∗ ∗ ∗ A P0 kRA P0 RA P0 k The definition of Fliengner’s number (Fn) is Fn ≡ √ mc ˙ 0 RA∗ P0 Fn (4. It can be noticed that Fliengner’s equation can lead to definition of the Fliengner’s Number.56) “Naughty Professor” Problems in Isentropic Flow To explain the material better some instructors invented problems. ISENTROPIC FLOW The maximum flow rate can be expressed either by taking the derivative of equation (4. k = 1.48) in with respect to M and equating to zero.52) is known as Fliegner’s Formula on the name of one of the first engineers who observed experimentally the choking phenomenon.48) results in F n = kM k−1 2 M 1+ 2 k+1 − 2( k−1) (4. which have mostly academic proposes.54) Utilizing Fliengner’s number definition and substituting it into equation (4. (see for example.54 CHAPTER 4. √ m ˙ T0 = 0.51) max The maximum flow rate for air (R = 287j/kgK ) becomes. Shapiro (problem 4.5)).52) Equation (4.55) and the maximum point for F n at M = 1 is Fn = k k+1 2 k+1 − 2( k−1) (4.50) max For specific heat ratio.53) (4.4 m ˙ A∗ 0. Carrying this calculation results at M = 1.040418 A ∗ P0 (4.68473 P √0 ∼ √ T0 R (4. m ˙ A∗ P √0 = T0 k R k+1 2 k+1 − 2( k−1) (4.

61) version 0.g. ρ when all other variables are known.58) transformed it into ρ2 = k−1 Pρ + T0 R 2kRT0 m ˙ A 2 (4.4.57) result in p R m ˙ Aρc 2 (4. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 55 problems have a limit applicability in reality.59) Equation (4. Using energy equation (4. the terms 0 for the group over the under–brace approaches zero when the flow rate (Mach number) is very small. It is convenient to change it into ρ2 − k−1 Pρ − T0 R 2kRT0 m ˙ A 2 =0 (4. e.59) is quadratic equation for density.60) The only physical solution is when the density is positive and thus the only solution is    P 1 ρ=  + 2 RT  0 P RT0 2 k−1 +2 kRT0 m ˙ A →(M →0)→0 2  For almost incompressible flow the density is reduced and the familiar form of perfect gas model is seen since stagnation temperature is approaching the static P temperature for very small Mach number (ρ = RT ).44 of this book. The situation where the mass flow rate per area given with one of the stagnation properties and one of the static properties.9) and substituting for Mach number M = m/Aρc ˙ results in k−1 T0 =1+ T 2 Rearranging equation (4.2. For this kind of problems a new Table was constructed and present here5 . they have substantial academic value and therefore presented here.58) And further Rearranging equation (4. The case of T0 and P This case considered to be simplest case and will first presented here. 5 Since    (4. In other words. P0 and T or T0 and P present difficulty for the calculations. . The use of the regular isentropic Table is not possible because there isn’t variable represent this kind problems.57) 1/kR T0 ρ 2 = T ρ ρ + T c2 k−1 2 m ˙ A 2 (4.

F n2 A ∗ P0 AP With this new definition equation (4.67) (4.1). m ˙ A 2 = 1 c0 2 k P0 2 m ˙ A∗ 2 A∗ A 2 P0 P 2 (4.66) m ˙ A 2 = F n2 k A ∗ P0 AP 2 T T0 P0 P 2 (4.65) Again notice that the right hand side of equation (4.64) RT0 P2 m ˙ A 2 = F n2 k A ∗ P0 AP 2 (4. And the values of A ∗P 0 in Table (4.61) is transformed into   2 ˙ 1 (k − 1)RT0 m  1+ 1+2 ρ ˆ= 2 kP 2 A (4. k ). First. T . again the specific heat ratio. this dimensionless group is function of Mach number only.56 CHAPTER 4. and static temperature.66) is only function of Mach AP were tabulated number (well. the problems is reduced to finding tabulated values.68) . Thus. P0 . The case of P0 and T A similar problem can be described for the case of stagnation pressure. 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.63) =f (M ) RT0 P2 Thus. k also).2) and Fn is tabulated in the next Table (4. ISENTROPIC FLOW It is convenient to denote a new dimensionless density as ρ ˆ= ρ p RT0 = 1 ρRT0 = ¯ P T (4. it is shown that the dimensionless group is a function of Mach number only (well. also the specific heat. the dimensionless density is  1 (k − 1)F n2 ρ ˆ = 1 + 1 + 2 2 k2 A∗ P AP 0 2   (4.

P . First. The method of solution for given The case of ρ0 and T or P The last case sometimes referred to as the “naughty professor’s question” case dealt here is when the stagnation density given with the static temperature/pressure.70) The last case dealt here is of the stagnation density with static pressure and the following is dimensionless group c0 2 2 2 1 Rρ0 2 T m ˙ A = kRT0 T0 kRP0 2 T m ˙ A = c 0 2 T0 kRP0 2 T m ˙ A 2 = F n2 k T0 T (4.2). (4.71) It was hidden in the derivations/explanations of the above analysis didn’t explicitly state under what conditions these analysis is correct. the dimensionless approach is used later analytical method is discussed (under construction). For example the first “naughty professor” question is sufficient that process is adiabatic only (T0 . The heat/temperature part is valid for enough adiabatic condition while the pressure condition requires also isentropic process.). not all the analysis valid for the same conditions and is as the regular “isentropic” Table. 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. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 57 Thus equation (4.2). The actual solution of the equation is left as exercise (this example under construction). c0 2 2 2 1 Rρ0 P m ˙ A = kRT0 P kRP0 P0 P 0 m ˙ A = c0 2 P kRP0 2 P 0 m ˙ A 2 = F n2 k P0 P (4. This example shows how a dimensional analysis is used to solve a problems without actually solving any equations.69) The right hand side is tabulated in the “regular” isentropic Table such (4. Unfortunately. mass flow rate per area. . Based on the same arguments.67) became RT P0 2 m ˙ A 2 = A ∗ P0 AP 2 (4.2. All the above conditions/situations require to have the perfect gas model as the equation of state. if it was shown that a group of parameters depends only Mach number than the Mach is determined by this group.4.

36550 0.10639 0.008 0.18428 0.000424 0.68875 0.44192 0.31203 0.020986 0.30000 0.33465 0.44215 0.30185 1.42683 1.13284 0.75136 0.39478 1.038 0.31000 0.85261 1.52858 0.065 0.42000 0.00E +00 1.16522 0.62915 0.10752 0.54422 1.00351 0.077 0.14889 0.20458 0.24000 0.20442 0.012593 0.003 0.079722 0.17397 0.25000 0.35361 0.074314 0.098460 0.25018 0.024585 0.44309 1.13342 0.26264 0.010476 0.00747 0.038365 0.51000 0.30214 0.38884 0.005 0.000 0.66098 0.10397 0.11710 0.71967 0.28000 0.46677 0.88588 0.34775 1.017 0.39701 0.050 0.095449 0.69036 1.41000 0.37000 0.48360 0.0 0.001 0.20000 0.56172 1.45000 0.009 0. ISENTROPIC FLOW Table 4.72927 1.001 0.58 CHAPTER 4.13796 0.31703 1.55637 0.36764 0.046 0.071 0.47609 1.34000 0.25535 0.15963 0.81706 0.013 0.10000 0.52485 0.005 0.35000 0.000 0.40333 0.24773 0.14084 1.37896 1.23777 0.62E −05 0.95791 0.083989 0.74912 1.65246 1.38000 0.026 0.028 0.055 0.41338 0.63386 1.057647 0.50978 1.006 0.031 0.21703 0.094654 0.29000 0.36329 1.59736 1.072487 0.54531 0.007 0.33378 0.22000 0.044110 0.40000 0.067111 0.16581 0.0 2.21584 0.26495 0.17381 0.41073 1.85107 0.1: Fliegner’s number and other paramters as function of Mach number M Fn ρ ˆ P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ 0 2 T m ˙ 2 A 0.35316 0.015 0.18896 0.18709 0.003 0.57656 0.24674 0.43000 0.023 0.63889 .60706 0.035 0.27358 0.081847 0.49305 0.012 0.29663 0.41997 0.16121 0.83132 1.0 0.51932 0.17728 0.089910 0.76924 1.22085 0.57709 0.34330 0.39596 0.20316 0.78965 1.015027 0.065557 0.27608 0.014268 0.33233 1.49285 1.44000 0.57944 1.46000 0.36000 0.11928 0.12039 0.20109 0.14592 0.46633 0.050001 0.27000 0.18992 0.45951 1.26000 0.31480 0.042 0.060404 0.14276 0.00352 0.48000 0.017813 0.074254 0.0 0.53000 0.23000 0.50000 0.400E −06 1.73995 0.49485 0.11294 0.050518 0.21000 0.002 0.060 0.004 0.033229 0.81034 1.27926 0.69948 0.23155 0.028651 0.52690 1.91838 1.87421 1.003 0.37474 0.65857 0.15372 0.22634 0.49000 0.44363 0.014197 0.37432 0.52000 0.23137 0.021 0.12724 0.60047 0.059212 0.46798 0.32000 0.67129 1.43919 0.92149 0.61550 1.13232 0.33000 0.001 0.070106 1.12239 0.011 0.029920 0.002 0.89613 1.28307 0.14927 0.019 0.55000 0.54000 0.41855 0.62436 0.70969 1.30418 0.000 0.54733 0.78382 0.29247 0.28677 1.39000 0.33226 0.00707 0.51882 0.47000 0.32220 0.00865 0.065654 0.58952 0.57253 0.087372 0.49249 0.

865 .278 2.582 1.971 3.79000 0.63000 0.814 1.77000 0.083 1.70000 0.035 1.94096 0.508 1.657 1.018 2.67000 0.20 0.500 1.187 1.949 9.385 1.806 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ0 2 T m ˙ 2 A 0.081 2.082 2.671 2.258 4.469 2.405 1.699 3.595 2. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 59 Table 4.489 2.484 4.006 6.87424 0.113 1.96389 0.57000 0.74624 0.144 ρ ˆ 1.708 1.500 1.211 1.554 10.423 1.62000 0.853 1.81996 0.212 1.474 1.67210 0.368 1.953 2.349 1.541 1.76000 0.58000 0.99514 1.277 2.244 3.324 1.972 5.202 1.553 1.676 1.736 1.95665 1.607 1.937 1.78000 0.602 2.56000 0.670 2.461 1.61000 0.897 5.185 1.85000 0.92366 0.011 1.059 2.86101 0.88142 0.094 1.840 4.084 1.581 1.105 0.842 2.541 3.148 1.97562 1.90000 0.088 1.738 1.122 1.033 1.151 1.297 1.216 2.771 1.2.038 2.220 3.790 2.1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.859 1.607 1.097 1.991 2.526 1.88000 0.516 1.59000 0.81139 0.90382 0.382 1.043 4.846 8.71000 0.381 8.326 1.424 6.587 1.161 1.60000 0.791 1.661 1.241 2.090 1.75000 0.320 1.030 1.241 1.389 3.244 5.124 1.269 1.029 3.404 2.68565 0.761 1.924 1.882 2.305 1.709 3.104 3.255 1.571 4.62936 0.92000 0.70675 0.147 1.044 1.869 1.72000 0.819 1.784 1.342 7.096 2.538 1.963 3.477 1.362 1.82722 0.340 2.059 1.703 1.287 3.660 1.955 3.979 4.78062 0.94000 Fn 0.225 1.152 1.131 1.747 1.73000 0.667 1.412 2.869 7.509 1.109 1.072 1.156 2.105 1.415 1.68000 0.448 1.89000 0.450 1.64000 0.66000 0.535 2.101 2.266 4.93000 0.168 2.323 1.162 1.198 1.236 1.414 1.135 1.177 2.931 2.427 1.355 1.980 2.94848 0.721 4.977 2.260 1.65000 0.557 1.271 1.903 1.646 3.407 2.642 1.173 1.86000 0.740 2.810 2.574 1.610 1.4.80000 0.278 1.87000 0.240 1.78250 0.381 2.613 6.214 1.217 1.74000 0.91000 0.671 1.758 1.513 5.640 1.462 3.288 1.457 3.383 1.235 5.112 1.317 1.446 1.121 3.898 1.81000 0.99507 1.284 2.805 0.141 1.058 2.602 2.037 1.712 1.813 2.74290 0.279 1.69000 0.720 2.998 3.821 1.343 1.98717 1.536 2.82000 0.84000 0.83000 0.194 1.422 1.

217 4.136 7.920 1.233 2.003 2.4: A gas flows in the tube with mass flow rate of 1 [kg/sec] and tube cross section is ◦ 0. At some point the static pressure was measured to be 1.404 4.72) From equation (4. Flow with pressure losses The expression for the mass flow rate (4. Calculate for that point the Mach number. it is clear that the function f (P0 .37 13.804 5. Assume that the process is isentropic and k=1.515 7.06 14.60 12.5[Bar].500 3.338 3. Denoting subscript A for one point and subscript B for another point mass equation (4. There are two possible models that can be used to simplify the calculations.600 4. S OLUTION The second academic condition is when the static temperature is given with the stagnation pressure.419 3.046 P0 A∗ AP 2 RT0 P2 m ˙ 2 A 1 Rρ0 P m ˙ 2 A 1 Rρ 0 2 T m ˙ 2 A 3.60 CHAPTER 4.48) can be equated as kP0 A∗ RT0 k−1 2 1+ M 2 k−1 − 2( k−1) = constant (4.72).961 2. 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.98000 0. A constant flow rate requires that m ˙A = m ˙ B .777 7.95000 0. Clearly.181 3.97000 0.3.037 4.188 2. T0 . in a case where the flow isn’t isentropic or adiabatic the total pressure and the total temperature will change (due to friction.278 2.259 3. .88 11. The temperature at Chamber supplying the pressure to tube is 27 C . ISENTROPIC FLOW Table 4.436 6.881 1.371 2. and heat transfer).1: Fliegner’s number and other paramters as function of Mach number (continue) M 0.96000 0.016 Example 4.583 10. A∗ ) = constant.419 ρ ˆ 1. the velocity.98 6.324 2.000 Fn 2.843 1. The first model for neglected heat transfer (adiabatic) flow and in which the total temperature remained constant (Fanno flow like).112 6.47) is appropriate regardless the flow is isentropic or adiabatic. The third academic condition is of static temperature and the static pressure.913 4.001[m2 ].19 14.99000 1. and the stagnation pressure. The second model which there is significant heat transfer but insignificant pressure loss (Rayleigh flow like). 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.

With these information. calculated the total pressure lost.5. and the duct section area is 0.73) For adiabatic flow.5000 T T0 ρ ρ0 A A P P0 0.13169 2.01[m2 ].68966 0. And utilizing the equality of A∗ = P 0 |A A∗ | = ∗A P 0 |B A |B A∗ A A (4.015[m2 ] and Mach number is 1.55401 0.5000 2. Mach number is 2.4. M 1.75) the stagnation pressure at point B can be obtained. Assume no mass lost and adiabatic steady state flow. thus the area ratios can be calculated. S OLUTION Both Mach numbers are known.5: At point A of the tube the pressure is 3[Bar]. Downstream at exit of tube.74) leads to A A ∗ MA A A ∗ MB P 0 |A = P 0 |B A| A A| B (4.27240 0. ISENTROPIC CONVERGING-DIVERGING FLOW IN CROSS SECTION 61 If the mass flow rate is constant at any point on the tube (no mass loss occur) then k RT0 2 k+1 k+1 k−1 m ˙ =A ∗ P0 (4. the cross section area is 0.76) Example 4. and utilizing equation (4.05853 0.15432 0. comparison of mass flow rate at point A and point B leads to P 0 A ∗ |A = P 0 A ∗ |B .62693 A×P A∗ ×P0 F F∗ .32039 0.5.39498 1.2.1762 0.44444 0. The total pressure can be calculated because the Mach number and static pressure are known.6367 0. point B.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.

99303 0.99825 0.93155 0.243 = 36.99875 0.57 5.058527663 by utilizing equation (4.243 = 4.53114 0.300 0.75) provides P0 |B = 51.65602 0.25781291 × Hence P0 |A − P0 |B = 51.32039 A×P A∗ ×P0 F F∗ 2.800 0.92427 0.00000 0.008 1.25781291[Bar] 0.53399 0.39498 5. the stagnation at point A is obtained from Table (4.340 1.424 1.99950 0.600 0.97250 0.79158 0.99206 0. ISENTROPIC FLOW First.54655 0.58377 0.882 1.91075 0.38484 0. 0.31424 0.000 1.48290 0.700 0.000 0.590 1.86059 0.00000 0.59126 0.55401 .3 Isentropic Tables Table 4.68110 0.176 1.74738 0.066 1.27240 5.35036 0.72632 0.63535 0.0000 0.95638 0.400 0.98028 0.84302 0.52989 0.53974 0.8E +5 11.200 0.009 1.27240307 × 15.912 1.4E +5 4.500 T T0 ρ ρ0 A A P P0 1.013[Bar] Note that the large total pressure loss is much larger than the static pressure loss (Pressure point B the pressure is 0.78400 0.8E + 5 11.96899 0.300 1.094 1.964 2.035 1.89699 0.77640 0.52828 0.80515 0.130 0.52828 0.5 A = 3 = 51.41238 0.838 2.038 1.99502 0.2) as P 0 |A = P P P0 M =2.015 4.030 1.62 CHAPTER 4.443 1.78896 0.2: Isentropic Table k = 1.42493 0.63394 0.53807 0.99800 0.88652 0.050 0.88517 0.59 5.84045 0.781 2.89561 0.43742 0.73999 0.146[Bar]).47207 0.4 M 0.01 1.93947 0.1761671 × ≈ 15.46835 0.822 2.68704 0.100 1.95238 0.55425 0.257 − 15.6367187 0.200 1.72093 0.68966 1.93284 0.58170 0.243[Bar] 2.98232 0.53039 0.500 0.188 1.83333 0.59650 0.115 1.900 1.100 0.400 1.00 1.71839 0.52828 0.268 0.36091 0.

2: Isentropic Table k=1.00364 0.047619 0.77) 6 The one dimensional momentum equation for steady state is U dU/dx = −dP/dx +0(other effects ) which are neglected here. Again in reality the heat transfer is somewhere in between the two extremes.31E −5 2.000385 0.14924 0. ISENTROPIC TABLES Table 4.4.2E +2 5.56182 0.000 5.500 4.35714 0.g.14184 0.00 T T0 ρ ρ0 A A P P0 0.35573 0.74E −5 3.089018 0.688 2.77) into the momentum equation6 yields U dU + RT dP =0 P (4.058140 0. Eckert number is very small) is presented.00190 0. The perfect gas model is again assumed (later more complex models can be assumed and constructed in a future versions).15432 0.57768 0.000 3.057227 0.55556 0.5E +2 3.000495 1. .72953 0.69983 0.3E +2 4.017321 0.033682 0.500 10.076226 0.31969 0.013111 0.73510 0.11528 0.028962 0.63371 0.0E +2 1.13 1.73903 0.052493 0.000242 0.3.000 9.039628 0.4 (continue) 63 M 1.16667 0.017449 0. In isothermal process the perfect gas model reads P = ρRT .87 53.28986 0.9E +2 2.235 6.66138 0.000 2. (e.56 25.17404 0.13169 0.500 6.000631 0.4E +2 0.027224 0. the other extreme case model where the heat transfer to the gas is perfect.000 8.00346 0.21567 0.900 2.3.00758 0.12780 0.250 1.000633 0.59309 0.047251 0.338 1.23810 0.072464 0.73257 0.00519 0.10582 0.72586 0.012628 A×P A∗ ×P0 F F∗ 0.000155 0.23211 0.00189 0.27099 0.500 5.23005 0.71578 0.72136 0. knowing the two limits provides a tool to examine where the reality should be expected.070595 0.081633 0.015504 0.600 1.58549 0.20259 0.439 1.064725 0.72 16.000 4.00659 0.500 9.28682 0. So.18 75.00261 0.000 7.25044 0.2 Isentropic Isothermal Flow Nozzle General Relationship In this section.00 36.12195 0.500 7.800 1.013957 0.000 6.045233 0.555 1.44444 0.000102 6.000815 0.790 10.025156 0.78) (4.00141 0.027662 0.74192 4. dP = dρRT Substituting equation (4.60680 0.4E +2 1.00107 0.500 3.68830 0.00107 0.637 4.56976 0.3.700 1.65326 0.022046 0.1 4.62693 0.73723 0.23527 0.500 8.019473 0.29414 0.67320 0.58072 0.70876 0.092593 0.058528 0.74058 0.36E −5 0.90E −5 4.19802 0.25699 0.011340 0.

79) to a dimensionless form becomes constant constant  (M2 2 − M1 2 ) kRT P2   ln = R  .83) yields . T 2 P1 (4.79) Thus.87) .78) yields the Bernoulli’s equation for ideal gas in isothermal process which reads . P2 k (M2 2 − M1 2 ) = ln 2 P1 (4.64 CHAPTER 4. the velocity at point 2 becomes U2 = 2RT ln P2 − U1 2 P1 (4.80) The velocity at point 2 for stagnation point. ISENTROPIC FLOW Integration of equation (4.82) Transform from equation (4.86) Utilizing conservation of the mass AρM = constant to yield M 2 P2 A1 = A2 M 1 P1 (4.84) Or in terms of the pressure ratio equation (4.81) Or in explicit terms of the stagnation properties the velocity is U= 2RT ln P P0 (4.85) As oppose to the adiabatic case (T0 = constant) in the isothermal flow the stagnation temperature ratio can be expressed 1  ¡ 1+ T0 1 T1 = ¡ T0 2 T2 1 + ¡ 2 k−1 2 M1 2 k−1 2 M2 = 1+ 1+ 2 k −1 2 M1 2 k −1 2 M2 (4. U2 2 − U 1 2 P2 + RT ln =0 2 P1 (4. U1 ≈ 0 reads U2 = 2RT ln P2 P1 (4.84) reads k(M1 2 −M2 2 ) P2 2 = =e P1 e M1 2 e M2 2 k 2 (4.83) Simplifying equation (4.

the critical stagnation pressure reads 2 k−1 (1−M )k 2 1 + P0 2 M1 2 = e P0 ∗ k+1 k k−1 (4. at this stage. The critical pressure ratio can be obtained from (4. It should be noted that the stagnation temperature is not constant as in the adiabatic case and the critical point is the only one constant. All these equations are plotted in Figure (4. The minimum of the curve is when area is minimum and at the point where the flow is choked.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. From the Figure 4.85) yields M1 A2 = A1 M2 65 e M2 2 e M1 2 k k−1 k 2 (4.4.6).93) Of course in isothermal process T = T ∗ .88) is reduced to obtained the critical area ratio writes A 1 (1−M 2 )k = e 2 ∗ A M Similarly the stagnation temperature reads 2 1 2 1 + k− T0 2 M1 = T0 ∗ k+1 k k−1 (4. is unknown (at what Mach number the nozzle is choked is unknown) so there are two possibilities: the choking point or M = 1 to normalize the equation. 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 (4.3. The true critical point is when flow is choked and the relationship between two will be presented.94) .89) The critical point.87) and equation (4.90) Equation (4.3 it can be observed that minimum of the curve A/A∗ isn’t on M = 1.91) (4.92) Finally. ISENTROPIC TABLES Combing equation (4.85) to read (1−M 2 )k ρ P = ∗ =e 2 ∗ P ρ (4. Again it has to emphasis that this critical point is not really related to physical critical point but it is arbitrary definition. Here the critical point defined as the point where M = 1 so results can be compared to the adiabatic case and denoted by star.

5 3 2.66 CHAPTER 4. It should be emphisized that the stagnation pressure decrees.7(a)).6: Various ratios as a function of Mach number for isothermal Nozzle Equation (4. ISENTROPIC FLOW Isothermal Nozzle 4 3. M = √ k (4. the pressure loss in adiabatic flow is milder as can be seen in Figure (4.5 2 1. The velocity √ k .5 1 0.5 0 0 0.5 3 3.95) It can be noticed that a similar results are obtained for adiabatic flow. 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. dividing the at the throat of isothermal model is smaller by a factor of √ critical adiabatic velocity by k results in Uthroatmax = √ RT (4.96) On the other hand. 4.5 1 1.94) simplified to 1 kM 2 − 1 = 0 . Thus. Utilizing equation .5 2 M 2.5 4 k=14 P/P * A/A * P0 / P0 T 0 / T0 T/T * * * Tue Apr 5 10:20:36 2005 Fig.

97) e− 2 = 0.5 4 3.5 0 0 1 0.4.5 1 0.5 2 1. ISENTROPIC TABLES Isothermal Nozzle 4 3.5 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. 4.60653 1 Notice that the critical pressure is independent of the specific heat ratio. k .5 2 M 2.5 3 3.5 1 0.5 1 1. It also has to be emphasized that the stagnation values of the isothermal model are not constant. as opposed to the adiabatic case.3.90) the following relationship can be obtained P ∗ Pthroat Pthroat = = P0initial P0initial P ∗ 1 e (1−02 )k 2 e „ 1− “ 1 √ k ”2 « k 2 = (4.5 2 1.5 0 0 0. Again.5 3 2.98) Q = Cp (T02 − T02 ) . the heat transfer is expressed as (4.7: The comparison of nozzle flow (4.5 1.5 4 A / A iso * A / A adiabatic * P / P iso * P / P adiabatic * 67 Comparison between the two models 5 4.5 3 2.

4. In this comparison it has to be remembered that critical area for the two models are different by about 3% (for k = 1. The ratio of the velocities can be expressed as √ Ms kRTs Us √ = (4. using the isentropic model results in under prediction of the actual results for the velocity in the supersonic branch. To make sense and eliminate unnecessary details 0.5 one (1)). The prediction of the Mach number are similarly shown in Figure (4.68 CHAPTER 4.8: 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. 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 (4.8 T / T0 isentropic but rather a simple sample P / P0 isothermal T/T0 isothermal just to examine the difference 0.2 the distance from the entrance to the throat is normalized (to 0 0 1 2 0. the isentropic for the subsonic branch will be over prediction.7(b)). . The Mach number was computed for the two models and plotted in Figure (4.7(b)). As can be observed from Figure (4. While.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.7(b)). 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. In this Fig.4).5 1. This P / P0 isentropic profile isn’t an ideal profile 0.100) the velocity ratio was obtained and is plotted in Figure (4. Thus.100) By utilizing equation (4.7(b)). The Mach number for the isentropic is larger for the supersonic branch but the velocity is lower.4 bounded.6 between the two models so in an actual situation it can be 0. 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).

Thus. The pressure ratio for the isentropic model is P = P0 int 1+ 1 k−1 2 2 M k−1 k (4.3.4.101) While the temperature ratio of the isothermal model is constant and equal to one (1). The stagnation pressure is 5[Bar] and the stagnation temperature is 27◦ C. with known area ratio and known stagnation Potto–GDC provides the following table: . while the range between the predicted pressure by the two models is relatively small. Calculate the velocity at the exit using the adiabatic model.6: Consider a diverging–converging nozzle made out of wood (low conductive material) with exit area equal entrance area.89) and the isentropic relationship. The temperature ratio of T /T0 int can be obtained via the isentropic model as T 1 = k−1 T0 int 1 + 2 M2 (4. Using equations (4. If the nozzle was made from copper (a good heat conductor) a larger heat transfer occurs.e. Example 4. 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. ISENTROPIC TABLES 69 Two other ratios need to be examined: temperature and pressure.4. The initial stagnation temperature is denoted as T0 int . Assume that the back pressure is low enough to have supersonic flow without shock and k = 1.102) and for the isothermal process the stagnation pressure varies and has to be taken into account as the following: isentropic P0 P0 z Pz = P0 int P0 int P0 ∗ ∗ Pz P0 z (4.8) shows that the range between the predicted temperatures of the two models is very large. To demonstrate the relativity of the approach advocated in this book consider the following example. should the velocity increase or decrease? What is the maximum possible increase? S OLUTION The first part of the question deals with the adiabatic model i. the conservation of the stagnation properties. the sought ratio is provided. Figure (4. The throat area ratio to entrance area is 1:4 respectively.103) where z is an arbitrary point on the nozzle.

the velocity can be calculated.9402 1.9405 0.777777778 11 k−1 T0 ∗ 1 + k− 1 + 2 k k Thus the stagnation temperature at the exit is T0ini = 1.9910 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 With the known Mach number and temperature at the exit. The exit velocity can be determined by utilizing the following equation √ √ Uexit = M kRT = 1.253[m/sec] (4. The exit velocity.98511 0.0000 0.99572 0. To explain the motivation for using this definition consider the calculation of the net forces that acting on section shown in Figure (4. 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 (4.4940 0.98934 0. then.36644 0.4 4.14655 2. ISENTROPIC FLOW A A P P0 A×P A∗ ×P0 0. the velocity in the copper nozzle will be larger than the velocity in the wood nozzle.92 × 300 = 576.0000 4.777777778 = 1.4940/0.921 T0exit The exit stagnation temperature is 1.12556 0.08129 4.11915 Even for the isothermal model. the maximum velocity cannot exceed the 691.0 = 691. The exit temperature is 0.50225 The exit Mach number is known and the initial temperature to the throat temperature ratio can be calculated as the following: 1 T0ini 1 = = = 0.104) As was discussed before.253[m/sec] 4.4 × 287 × 300.4.51183 4.105) .6) or using the Potto–GDC obtains the following table M 1.0000 0. is √ √ U = M kRT = 2. However.9402 T T0 ρ ρ0 CHAPTER 4.9K .4 × 287 × 109.02979 3. The Impulse function is denoted here as F .1 The Impulse Function Impulse in Isentropic Adiabatic Nozzle One of the functions that is used in calculating the forces is the Impulse function.70 M 0.2K . but in the literature some denote this function as I . the initial stagnation temperature is given as 300K .9 ∼ 617.93[m/sec] 1. Using the area ratio in Figure (4.9910 1.36644 × 300 = 109.9).

Defining a function that depends only Fig.106) Combining equation (4.108) shows that the right hand side is only a function of x-direction Mach number and specific heat ratio. Hence. defining the Impulse function as F = P A 1 + kM2 2 In the Impulse function when F (M = 1) is denoted as F ∗ F ∗ = P ∗ A∗ (1 + k ) The ratio of the Impulse function is defined as see function (4.108) which makes the ratio a function of k and the Mach number.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. Hence.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 (4.111) This ratio is different only in a coefficient from the ratio defined in equation (4.42) results in Rearranging equation (4.108) P1 A1 1 + kM1 F = ∗ ∗ F∗ P A (1 + k ) 2 (4.4. k .109) (4. 4. M and k . 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.112) .108) Examining 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. Thus.106) and by utilizing the identity in equation (4.105) with equation (4.107) P2 A 2 P1 A 1 Fnet = 1 + kM2 2 − 1 + kM1 2 ∗ ∗ P0 A P0 A P0 A ∗ (4.4. the net force is Fnet k+1 = P0 A (1 + k ) 2 ∗ k k−1 F2 F1 − ∗ ∗ F F (4. THE IMPULSE FUNCTION 71 The net force is denoted here as Fnet .110) = 1 P∗ P0 2 ( k+1 ) k−1 k P1 A 1 1 + kM1 2 P0 A ∗ 1 (1 + k ) (4.

10: Schematic of a flow of a compressible subzle and pressure at point 1.98526 0.94934/0.1000 0.94934 2.2227 The pressure at point 1 is P1 = P 2 0.058 0.776[Bar] P2 P0 .003 A 2 P2 = ∼ 2.99751 0.1949 P0 P1 = 5. Utilizing Fliegner’s equation (4.0times0.7) 0. the ratio of P1 A1 /A∗ P0 is needed to be calculated.1 A ∗ P0 70061.7: 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 = 0.2227 A A A2 0. 4.96355 2.0 × 400 × 287 A ∗ P0 = = ∼ 70061.2121 With the area ratio of A A 0.99380 5. A2 A1 0. To obtained the Mach number.058 and 500000 × 0. To obtain this ratio the denominator is needed to be obtained.76 M T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ 1 for example (4.11164 0. ISENTROPIC FLOW To demonstrate the usefulness of the this function consider a simple situation of the flow through a converging nozzle Example 4. provides the following √ √ m ˙ RT 1.99380 ∼ 4.96666 = 2.003m2 entrance area is 0.009 A1 = = 2.99132 5.2121 × = 5.27353 0.1774 2.2121 the area ratio of at point 1 can be calculated.76[N ] 0.009m2 2 A2 = 0.52). 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 ∗ ×P 0 F F∗ 0.003[m ].009[m ] and T0 = 400K 2 P2 = 50[Bar] the exit area is 0.72 CHAPTER 4. The stagnation temperature is 400K and the pressure at point 2 was measured as 5[Bar] Calculate the net force acting on the noz.Fig.

2 The Impulse Function in Isothermal Nozzle Previously Impulse function was developed in the isentropic adiabatic flow.96666) ∼ 614[kN ] = 500000 × 2.113) Utilizing the ideal gas model for density and some rearrangement results in P2 A 2 1 + F2 = F1 P1 A 1 1 + U2 2 RT U1 2 RT (4.4.010 2.064 1.5.23.87) transformed equation into (4.014 2.114) Since U 2 /RT = kM 2 and the ratio of equation (4.53199 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 (4.4.000 1.064 5. The same is done here for the isothermal nozzle flow model.112) Fnet = P2 A2 P0 A ∗ k + 1 k−1 F2 F1 (1 + k ) − ∗ P2 A 2 2 F∗ F 1 × 2.4 × 1.2E +5 20.115) 4.1 T0 T0 P0 P0 A A P P 0.05 0. As previously.949 5. the definition of the Impulse function is reused.064 1.1949 − 0.00 8.00 0.52828 0.0E + 5 9.1 k 73 4.001 2.3: Isothermal Table M 0.5 × (2.00 4.5 Isothermal Table Table 4.0E +6 4. ISOTHERMAL TABLE The net force is obtained by utilizing equation (4.116) (4.225 A×P A∗ ×P0 F F∗ .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 (4.362 10.

5E + 2 0.438 2.875 4.94147 1.000 6.3: Isothermal Table (continue) M 0.891 1.9 1.14286 0.608 3.025349 0.5 0.500 9.431 4.681 1.247 0.062 1.0 1.85853 1.800 1.20000 0.74 CHAPTER 4.78 0.0 0.000 8.015317 7.565 1.111 1.58824 0.000 1.61693 0.059 1.958 1.889 2.0 0.564 1.275 1.86329 0.200 1.296 5.000 3.26634 0.540 4.40000 0.30 1.245 2.599 0.25000 0.000 1.11111 0.8E + 2 4.6E + 4 2.8E +18 0.6 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- .75E −5 1.73492 0.0 0.271 0.20 1.7E + 3 1.6 0.020 1.047 1.66667 0.000 0.10526 0.50000 0.90 2.50 3.004 0.41 40.4E +26 0.51069 0.194 1.000215 1.7 0.500 8.667 1.99967 1.143 4.763 1.047 0.690 1.718 3.0 1.0 1.333 2.1E + 3 0.4 0.079 1.83333 0.0 2.11765 0.0 7.500 4.0 0.287 1.4E + 8 0.2E + 3 7.75344 1.98796 1.90302 1.000 3.12246 0.399 0.142 1.0 5.0 0.569 3.000 4.18182 0.80 1.5E + 2 2.0 1.000 5.134 0.6E + 5 1.10 1.281 1.41E −6 0.50 1.00221 9.250 1.58985 0.76923 0.56232 0.21 1.4E + 3 5.60 1.500 10.000 7.80844 1.500 2.56954 3.3 0.62665 0.73278 0.55556 0.000 0.70 1.63276 2.00 T0 T0 P0 P0 A A P P 0.0 0.3E +15 0.053 0.000380 2.67383 0.16667 0.071758 90.500 7.939 2.036 0.6E +13 0.284 3. ISENTROPIC FLOW Table 4.000 0.1E + 4 1.3E + 2 1.114 1.50618 4.0 5.625 1.0 5.083 0.281 1.62500 0.69449 2.97274 1.153 1.000 1.065 1.89348 1.52632 0.9E + 2 8.14 0.0 0.97376 1.8 0.100000 A×P A∗ ×P0 F F∗ 2.035 3.54322 0.026 19.0E + 6 0.41E −5 4.375 1.71429 0.0 0.134 9.90909 0.80528 0.20846 0.2E + 4 1.719 5.553 1.856 4.064 2.0 5.0 0.3E +11 0.429 1.161 2.41686 0.429 1.2 0.000 3.464 1.863 0.16090 0.000 1.13333 0.22222 0.063 1.00370 0.3E +23 0.98750 0.33554 0.29 80.500 5.586 5.0 8.0 0.500 6.055 1.328 1.389 1.97156 1.000 9.00 2.021 0.12500 0.2E + 3 3.28571 0.00 1.007 5.40 1.3E + 9 0.128 1.209 0.044 1.22881 15.33333 0.665 0.237 1.183 1.125 1.4E + 3 2.4E +20 0.2E +29 1.15385 0.

To carry the integration of the right hand side the relationship between the pressure and the density has to be defined. The following power relationship is assumed ρ = ρ0 P P0 1 n (4. Since ds = 0 and utilizing the equation of the state dh = dP/ρ.120) P Equations (4.4. a real gas equation can be expressed in equation (3. h = h(T.119) The definition of pressure specific heat is Cp ≡ Maxwell relation hence.122) when using relationship which is defined in equation (4.19).19) and dividing by equation (3. The enthalpy is a function of the temperature and pressure thus.121) P The mechanical energy equation can be expressed as d U2 2 =− dP ρ (4. Differentiating equation (3. the differential of enthalpy is calculated for real gas equation of state as dh = Cp dT − T Z ∂z ∂T dP ρ (4. Thus. With integration of equation (4.124) . P ) and full differential is dh = ∂h ∂T dT + P ∂h ∂P dP T ∂h ∂T (4.19) yields dz dρ dT dP = + + P z ρ T (4. that for perfect gas the n is substituted by k .19) are combined to form Cp dT ds = −z 1+ R R T T Z ∂z ∂T dP P (4.117) Again.118) and second derivative is (4.6.27) and (3.122) At the stagnation the definition requires that the velocity is zero. THE EFFECTS OF REAL GASES 75 erties like the critical pressure and critical temperature.27) is reused to related the entropy change to the change in thermodynamics properties and applied on non-ideal gas. ∂h ∂P =v−T ∂s ∂T T P First.123) Notice. Gibb’s equation (4.123) results U2 = 2 P1 P0 dP = ρ P P0 1 ρ0 P0 P 1 n dP (4.

121) when ds = 0 results T2 T1 Cp dT = R T P2 z 1+ P1 T Z ∂z ∂T P dP P (4.76 CHAPTER 4.130) The Mach number can be obtained by utilizing equation (3.19) results U2 = 2 P P0 z0 RT0 P0 P0 P 1 n dP (4.126) For n = 1 the integration becomes U= 2z0 RT0 ln P0 P (4.132) . Thus. The mass flow rate is regardless to equation of state as following m ˙ = ρ ∗ A∗ U ∗ (4.131) Integrating equation (4.34) to defined the Mach number as M=√ U znRT (4.125) For n > 1 the integration results in 2n 1− n−1 n−1 P ( n ) P0 U= z0 RT0 (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. 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. ISENTROPIC FLOW Substituting relation for stagnation density (3.128) Where ρ∗ is the density at the throat (assuming the chocking condition) and A∗ is the cross area of the throat.127) It must be noted that n is a function of the critical temperature and critical pressure.

136) The mass flow rate for the real gas m ˙ = ρ ∗ U ∗ A∗ A ∗ P0 m ˙ =√ z0 RT0 And for n = 1 A ∗ P0 m ˙ =√ z0 RT0 Fliegner’s number in this case is Fn = mc ˙ 0 A ∗ P0 2n n−1 P∗ P0 1 n 2n n−1 P∗ P0 1 n 1− P∗ P0 (4.139) Fliegner’s number for n = 1 is Fn = mc ˙ 0 =2 A ∗ P0 P∗ P0 2 − ln P∗ P0 (4.140) .6.137) 2n n−1 2z0 RT0 ln P0 P (4.132) looks at Bernnolli’s equation which is dU 2 =− 2 After integration of the velocity dU 2 =− 2 P/P0 1 dP ρ (4.135) U= 2z0 RT0 ln P0 P (4.4. 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− (4.134) It was shown in Chapter (3) that (3.138) 1− P∗ P0 (4.33) is applicable for some ranges of relative temperature and pressure (relative to critical temperature and pressure and not the stagnation conditions).133) ρ0 d ρ P P0 (4.

ISENTROPIC FLOW n n−1 (4. Note (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 (4.142) To obtain the relationship between the temperature and pressure.147) for n = 1 T0 = T The critical temperature is given by T∗ = T0 eM [z+T ( 2 ∂z ∂T )P ] (4.132) can be integrated T0 = T P0 P R Cp ∂z )P ] [z+T ( ∂T (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 1.146) 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 (4.148) ∂z n z +T ( ∂T )P ] 1 + n ( 1−n )[ 2 (4.141) When n = 1 or more generally when n → 1 this is a ratio approach P∗ √ = e P0 (4.78 The critical ratio of the pressure is P∗ = P0 2 n+1 CHAPTER 4.149) . equation (4.145) For n = 1 the Mach number is M= 2 z 0 T0 P 0 ln z T P (4.

152) Example 4. Are the stagnation pressure and temperature at the entrance different from the point? You can assume that k = 1. the stagnation pressure is constant for isentropic flow. THE EFFECTS OF REAL GASES and for n = 1 T∗ = T0 79 e−[z+T ( ∂z ∂T )P ] (4.42027 0. Calculate the area ratio between the point and the throat. iii.6100 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.150) 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 (4.802[Bar] P 0.04943 T0 300 T = ∼ 713.8: A design is required that at a specific point the Mach number should be M = 2. Of course.6.11761 2. i.14366 2.151) For the case of n = 1 the mass flow rate is P0 A ∗ n m ˙ = c0 e M2 n−1 2 M 1+ 2 n+1 n−1 (4.405.4. The stagnation pressure is obtained from P0 = P0 2. The solution is simplified by using Potto-GDC for M = 2. and temperature 300K .61 the results are M 2.9066 0. the pressure 2[Bar].61 P = ∼ 52. Calculate the stagnation pressure and the stagnation temperature. ii.82K T 0.42027 The stagnation temperature is T0 = 3. S OLUTION 1.04943 0.61. .

80 CHAPTER 4. ISENTROPIC FLOW .

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

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

one can have a whole range of solutions. U. The first set is the energy.. s. This situation is similar to a frictionless flow with the addition of heat. which will produce a range of possible solutions. when the momentum equation is solved instead of the energy equation. and state equations. The solution of every set of these equations produces one additional degree of freedom. Now one can view these equations as two different subsets of equations. Clearly. the degree of freedom is now energy.2: The intersection of Fanno flow and Rayleigh flow produces two solutions for the shock wave. continuity. On the other hand. which are needed to be solved.6) It can also be noticed that entropy. In Figure (5. the energy equation is used.546flow there isn’t essentially any T =?>A@ momentum transfer (a speshock jump cial case of Fanno flow).83 Note that the greater–equal signs were not used. one can solve only four (4) equations (including the second law). 5.2). In the first case.5) is ln Ty Py − (k − 1) >0 Tx Px (5. the increase of the entropy determines the direction of s the flow. Instead of solving all the equations that were presented. and Chapter (9) deals extensively with this topic. i. The reason is that the process is irreversible. continuity. producing various resistance to the flow. For ideal gas. Mathematically. . the energy amount “added” to the shock. Since the shock has $&%(' no heat transfer (a special EGFI J H K subsonic flow case of Rayleigh flow) and supersonic . and ρ. B?CAD the intersection of these two curves is what really Rayleigh Fanno line line happened in the shock. continuity. equation (5.38:95<68 tions are obtained.e.213-/. This case is called Fanno flow. It is also worth noting that the temperature at M = 1 on Rayleigh flow is larger than that on the Fanno line. and the second set is the momentum. the parameters are P. can be expressed as a function of the other parameters. If the energy. point y .0. and state equations. and this flow is known as Rayleigh flow. T. The mathematical explanation is given Chapter (9) in greater detail.¥-/. which will require additional parameters. and state equations are solved for the arbitrary value of the Ty . and therefore no equality can exist. The entropy increases from point x to Fig. This flow is dealt with in greater detail in Chapter (10). Thus. the intersec)&*(+ tion is shown and two solu7¥8:95. a parabola in the T –s diagram will be obtained.

1. NORMAL SHOCK 5.84 CHAPTER 5. and thus for known Mx all the other quantities can be determined.1) substituting for density. ρ. c2 = kRT to be used for the Mach number definition. The relationship between the temperature and the stagnation temperature provides the relationship of the temperature for both sides of the shock.2). The analytical solution is discussed in the next section. k.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. at least. 5. (5.11) . The only issue left to derive is the relationship between Mx and My .7) All the other relationships are essentially derived from this equation. Ty = Tx Ty T0 y Tx T0 x = 1+ 1+ 2 k −1 2 Mx 2 k−1 2 My (5. Utilizing the definition of Mach number results in Py 2 Px 2 Mx 2 = My 2 Tx Ty (5.10) Note that the speed of sound on the different sides of the shock is different.9) (5. the equation of state yields Px Py Ux = Uy RTx RTy Squaring equation (5.8) Multiplying the two sides by the ratio of the specific heat.1). numerically. From the continuity equation (5. It can also be noted that in many cases the dimensionless equations set is more easily solved. provides a way to obtain the speed of sound definition/equation for perfect gas. The reason that dimensionless forms are heavily used in this book is because by doing so it simplifies and clarifies the solution. Note that the Mach number is a function of temperature. T0 x = T0 y . and (5.2 Formal Model Equations (5. as follows: Px 2 Py 2 Ux 2 = Uy 2 Tx kRTx Ty kRTy cx 2 cy 2 (5.1 5.3) can be converted into a dimensionless form.1.8) results in Py 2 Px 2 2 U = Uy 2 x R 2 Tx 2 R 2 Ty 2 (5.

14) The pressure ratio in equation (5. T0 y = T0 x . This relationship can be obtained from the combination of mass.16) can be interpreted as the loss of the static pressure.13) It can also be observed that equation (5.16) (5.13) means that the stagnation temperature is the same. ρU 2 is identical to kP M 2 because M2 ρ ρU 2 = Using the identity (5. From equation (5. and energy equations.11) results in Ty = Tx Py Px 2 85 My Mx 2 (5.11)) as Py 1 + P0 y = P0 x Px 1 + 2 k−1 2 My 2 k−1 2 Mx k k−1 k k−1 (5.5.13) (energy) and equation (5.14) transforms the momentum equation (5. SOLUTION OF THE GOVERNING EQUATIONS Rearranging equation (5.12) (mass) the temperature ratio can be eliminated. momentum.15) yields Py 1 + kMx 2 = Px 1 + kMy 2 (5.15) 2  P   U  kRT = kP M 2  RT kRT  c2   (5. The loss of the total pressure ratio can be expressed by utilizing the relationship between the pressure and total pressure (see equation (4. Py M y Px M x 2 = 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5.18) .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 (5.1. Under the perfect gas model.2) into Px + kPx Mx 2 = Py + kPy My 2 Rearranging equation (5.12) Energy equation (5.

1990. pp. 108-137.19) is a symmetrical equation in the sense that if My is substituted with Mx and Mx substituted with My the equation remains the same. M. K. ”Cubic and Biquadratic Reciprocity.20) is the trivial solution in which the two sides are identical and no shock wave occurs. and Rosen. the pressure and the temperature from both sides of the nonexistent shock are the same. for example. According to the Gauss Biquadratic Reciprocity Theorem this kind of equation has a real solution in a certain range3 which will be discussed later.16) and equation (5. The solution can be obtained by rewriting equation (5. one solution is My = M x (5.16) results in 1 + kMx 2 1 + kMy 2 2 = Mx My 2 1+ 1+ 2 k−1 2 Mx 2 k−1 2 My (5.23) 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. New York: Springer-Verlag. 2nd ed.22) The first solution (5. . The pressure ratio between the two sides can now be as a function of only a single Mach number. Thus. Utilizing equation (5. The second solution is where the shock wave occurs.21) Equation (5. in this case. i.20) It can be observed that equation (5.86 CHAPTER 5. Mx .19) and divide it by My 2 − My 2 1+ k−1 My 2 + My 2 − kMy 2 My 2 = 0 2 (5.24) 3 Ireland.19) Equation (5. Px = Py .e. Clearly.21) becomes My 2 = Mx 2 + 2k k−1 Mx 2 k−1 2 −1 (5. Tx = Ty .19) as a polynomial (fourth order). NORMAL SHOCK Combining the results of (5.” Ch.19) is biquadratic. 9 in A Classical Introduction to Modern Number Theory.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. It is also possible to cross–multiply equation (5.18) with equation (5.

27) My 1 0.28) Based on this definition a new Mach number can be defined M0 = U c0 (5.3 0. And the temperature ratio. 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 (5.7 0. Mx .5.1.3: The exit Mach number and the stagnation pressure ratio as a function of upstream Mach number. The speed of sound at that velocity is √ c∗ = kRT ∗ (5.30) .5 0. The Maximum Conditions The maximum speed of sound is when the highest temperature is achieved.8 0.1 1 2 3 Fri Jun 18 15:47:34 2004 0 4 5 6 7 8 9 10 My P0y/P0x Shock Wave relationship My and P0y/P0x as a function of Mx The stagnation speed of sound is c0 = kRT0 (5.9 0. The Star Conditions The speed of sound at the critical condition can also be a good reference velocity. provides additional way of obtaining an additional useful relationship.29) Mx Fig.6 0.25) In the same way. as a function of pressure ratio.2 0.4 0. SOLUTION OF THE GOVERNING EQUATIONS 87 The fact that the pressure ratio is a function of the upstream Mach number. is transformed into Ty = Tx Py Px k+1 k−1 Py Px k+1 Py k−1 Px + 1+ (5.26) which is associated with the shock wave. 5. The maximum temperature that can be achieved is the stagnation temperature Umax = 2k RT0 k−1 (5.

35) (5. an additional Mach number can be defined as M∗ = U c∗ (5.35) the following can be obtained: U1 U2 = c ∗ 2 or in a dimensionless form M ∗ 1 M ∗ 2 = c∗ 2 (5.31) 5.3 Prandtl’s Condition It can be easily observed that the temperature from both sides of the shock wave is discontinuous.1.34) .32) The jump condition across the shock must satisfy the constant energy. It is therefore convenient to define the star Mach number that will be independent of the specific Mach number (independent of the temperature). NORMAL SHOCK In the same manner.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. the speed of sound is different in these adjoining mediums.37) (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. c2 U2 c∗ 2 c∗ 2 k + 1 ∗2 + = + = c k−1 2 k−1 2 2(k − 1) (5. c U c U = ∗ = ∗M c∗ c c c M∗ = (5.36) (5.33) and (5. Therefore.88 CHAPTER 5.

the total pressure can be obtained (through an isentropic relationship in Table (4.0 110. The relationship that was calculated will be utilized to obtain the ratios for the downstream of the normal Px shock.0272237 = 55. It can be noticed that the density ratio (velocity ratio) also has a finite value regardless of the upstream Mach number. an example is provided.7245 My 0.2m/sec 1 2 3 Fri Jun 18 15:48:25 2004 4 5 Mx 6 7 8 9 10 . at a pressure of 0.47642 2. the known information are Mx = 3. OPERATING EQUATIONS AND ANALYSIS 89 5. Finding the Mach number for a pressure ratio of 8. Ty Tx ρy ρx Py Px P0y P0 x Mx 2. the Mach number after the shock. 5. Tx .4 or utilizing Potto-GDC for for value of the specific heat ratio.1) for k = 1.3088 0.0272237 =⇒ P0x = 1.3).0 60. My .30879 and k = 1. pressure.32 is only a few mouse clicks away from the following table. 120. and velocity downstream of the shock.9360 8.5/0. ρy/ρx and Ty/Tx as a function of Mx Shock Wave relationship Py/Px ρy/ρx Ty/Tx S OLUTION Fig.0 70. P0x is known).0 100.0 30. The typical situations in which these equations can be used also include the moving shocks.2.0 0.2). Example 5.0 Py/Py. are plotted as a function of the entrance Mach number.2 Operating Equations and Analysis In Figure (5. First. P0y /P0x .0 90. the velocity can readily be calculated. P = 0.0 20. px = 1. The working equations were presented earlier. and the ratio of the total pressure. Also with the temperature.0 40.5[bar] and Tx = 273K .0 50.4 × 287 × 273 = 331.e.1: Air flows with a Mach number of Mx = 3.0 80.38109 To illustrate the use of the above equations. This kind of equations requires examining Table (5. total pressure.1[bar] √ √ 0x cx = kRTx = 1.. The equations should be used with the Mach number (upstream or downstream) for a given pressure ratio or density ratio (velocity ratio). Note that the My has a minimum value which depends on the specific heat ratio.5 [bar] and a temperature of 0◦ C goes through a normal shock.5.4: The ratios of the static properties of the two sides Analysis: of the shock. Using these data. Calculate the temperature.0 10. i.1110 3.

it refers to a case where only a “small shock” occurs. The limits of the pressure ratio can be obtained by looking at equation (5. during the writing of this version of the book.0000 My 0.5 series).38).3.39) . This technique also has an academic value and therefore will be described in the next version (0. Now.2 Small Perturbation Solution The small perturbation solution refers to an analytical solution where only a small change (or several small changes) occurs.3333 0. the downstream Mach number (see equation (5. This approach had a major significance and usefulness at a time when personal computers were not available.40) (5.90 Mx 3. Uy = 993.2.09[bar] 5.6/3. In this case. which is up to Mx = 1.1[bar] = 18.32834 Ux = Mx × cx = 3 × 331.22)) is limited by My = 2 :  ∼0 2  1 + (k −1)Mx 2 2k k−1 >   − M1 x 2 ∼0 = k−1 2k (5. NORMAL SHOCK Py Px P0 y P0 x 2.47519 Ty Tx ρy ρx CHAPTER 5.85714 = 257.6790 3.8571 10.6[m/sec] P0y = P0y P0x × P0x = 0. 5.6[m/sec] Now the velocity downstream is determined by the inverse ratio of ρy /ρx = Ux /Uy = 3.16) and by utilizing the limit that was obtained in equation (5.3).32834 × 55.38) This result is shown in Figure (5. The strength of the shock wave is defined as ˆ = Py − P x = Py − 1 P Px Px By using equation (5. this technique is used mostly in obtaining analytical expressions for simplified models.2.85714.1 The Limitations of the Shock Wave When the upstream Mach number becomes very large.39) into ˆ= P 2k Mx 2 − 1 k+1 (5.23) transforms equation (5.2 = 993.

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. to convince the students that indeed the assumption of very thin shock is validated by analytical and experimental studies.24) the following is obtained: ˆ= P 2k k−1 2 k−1 ρy ρx 91 −1 −1 − ρy ρx (5. This kind of situation []\ ^]_ arises in many industrial applications. the shock wave is not stationary. THE MOVING SHOCKS or by utilizing equation (5.41) 5. when a valve is suddenly hji NPO  S T f d e b ` .2.3 Shock Thickness The issue of shock thickness (which will be presented in a later version) is presented here for completeness. The shock thickness can be defined in several ways.3 The Moving Shocks In some situations. however. the issue should be presented.3. flow L5M g direction Q5R For example.5. 5. This issue has a very limited practical application for most students.

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

e.47) My = cy cy Uy = U s − U y (5. In cases where the shock velocity can be approximated as a constant (in the majority of cases) or as near constant. the previous analysis. The general case is where two shocks collide into one shock and moves upstream or downstream is the general case. An observer moving with the shock will notice that the pressure in the shock is Px = P x Py = P y (5. The downstream prime Mach number can be expressed as Us Us − U y = − My = Msy − My (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 steady state is obtained in the moving control value. equations. and the tools developed in this chapter can be employed. Here. The problem can be reduced to the previously studied shock. Note that this notation is contrary to the conventional notation found in the literature. The upstream prime Mach number can be defined as Us − U x Us Mx = = − Mx = Msx − Mx (5. the prime ’ denote the values of the static coordinates. 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.5)) the velocity measured by the observer is Where Us is the shock velocity which is moving to the right. NORMAL SHOCK them at this stage. In such a case.43) The temperature measured by the observer is Tx = T x Ty = T y Assuming that the shock is moving to the right. In these changes (of opening a valve and closing a valve on the other side) create situations in which different shocks are moving in the tube. Msx = U cx .46) cx cx It can be noted that the additional definition was introduced for the shock upstream s Mach number. The reason for the deviation is that this choice reduces the programing work (especially for object–oriented programing like C++). to the stationary case when the coordinates are attached to the shock front.92 CHAPTER 5.44) .42) (5. For this analysis.45) Ux = U s − U x (5. the coordinates move with the shock. 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.. i.

5. and is “converted” or added into the total temperature.3.49) and the “upstream” prime stagnation pressure is P0x = Px k−1 Mx 2 1+ 2 k k−1 (5. The difference between the stagnation temperature is in the moving coordinates T0y − T0x = 0 (5.48) The “upstream” stagnation temperature of the fluid is T0x = Tx 1 + k−1 Mx 2 2 (5.50) The same can be said for the “downstream” side of the shock. T0y − T0x =Ty 1 + k−1 Msy − My 2 T0y 2 − Tx 1 + k−1 Msx − Mx 2 2 0 = Ty 1 + k−1 My 2 T0x 2 +Ty Msy k−1 (Msy − 2My ) 2 k−1 (Msx − 2Mx ) 2 (5. at a very high velocity. The rise in the total temperature is due to the fact that a new material has entered the c. .v.52) − Tx 1 + k−1 Mx 2 2 −Tx Msx and according to equation (5.51) leads to T0y − T0x = Us Ty k − 1 Tx k − 1 (Msx − 2Mx ) − (Msy − 2My ) cx 2 cy 2 (5. THE MOVING SHOCKS 93 Similar to the previous case. The relationship between the two new shock Mach numbers is cy Us Us = cx cx cy Msx = Ty Msy Tx (5. this difference in the moving shock is expected because moving material velocity (kinetic energy) is converted into internal energy. This difference can also be viewed as a result of the unsteady state of the shock.53) Again.51) It should be noted that the stagnation temperature (in the stationary coordinates) rises as opposed to the stationary normal shock. an additional definition was introduced for the shock downstream Mach number. Msy .

54) The “downstream” Mach number reads My = |Us | = Msy cy (5.42) to (5.v. The “upstream” or the “downstream” Mach number is not known even if the pressure and the temperature downstream are given. Fig. the shock is moving to the left. The relationship between the Mach number on the two sides of the shock are tied ¡£¢j¤¥¡ ¦ •£– ˜ —# œ5 § ¨/©«ª§¬ c. The information propagates upstream in which the gas velocity is converted into temperature. In many such cases the steady state is established quite rapidly.3.1 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.v. 5. The flow is moving to the right. The difficulty lies in the jump from the stationary coordinates to the moving coordinates. the shock velocity “downstream” is Us . for example.53) can be transformed into simpler equations when Mx is zero and Us is a positive value. Equations (5. NORMAL SHOCK 5. In the moving coordinates. The stagnation temperature increases by T0y − T0x = Us Tx k − 1 Ty k − 1 (Msx + 2Mx ) − (Msy ) cx 2 cy 2 ž Ÿ ‘3’ “ ” ™›š ±W² (5. The upstream is on the left of the shock. or Msy instead of the velocity because it combines the temperature and the velocity into one parameter. and for a given specific heat ratio.55) Again.56) The prominent question in this situation is what will be the shock wave velocity for a given fluid velocity. This situation arises in many cases in the industry. in a sudden and complete closing of a valve. the observer (with the shock) sees the flow moving from the left to the right. It turns out that it is very useful to use the dimensionless parameter Msx . . The sudden closing of the valve must result in a zero velocity of the gas. The “upstream” Mach number reads Mx = Us + U x = Msx + Mx cx (5.6: Comparison between a stationary shock and a moving shock in a stationary medium in ducts. Ux . This shock is viewed by some as a reflective shock. ­W®°¯ Stationary Coordinates À2ÁW›ÃÄÁWÅ:Æ Ç ³3´¶µZ· ¾›¿ ¼W½ ¸ p ¹ º¥¸#» Moving Coordinates c.94 CHAPTER 5. In such a case.

Mx ∼ equation (5. a Mach number close to one (1).57) And substituting equation (5.5.7).58) pro1 vides that Msx ∼ 1 + 2 and Msy = Fig.1 1 Mx the “smart” guessing of M sx.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). 1 Note.48) results in f (Msx ) Mx = Tx Ty Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. Mx . From the same figure it can also be observed that a high velocity can result in a much larger velocity for the reflective shock. the result is about double the sonic velocity of the reflective shock.3. The typical method is 0 0.7: The moving shock Mach numbers as a 1 result of a sudden and complete stop. which can easily be obtained in a Fanno flow. Note that to achieve supersonic velocity (in stationary coordinates) a diverging– converging nozzle is required.55) by 2 95 ( My ) = Mx + Msx 2k k−1 2 + Mx + Msx 2 k−1 2 −1 (5. . is provided herein: (a) Guess Mx > 1. 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.58) The temperature ratio in equation Shock in A Suddenly Close Valve (5. For example. Only 2 one real solution is possible.58) can be obtained by several numerical methods. Two numerical methods and the algorithm employed to solve this problem for given.57) into (5. an analytical solution can be obtained for equation (5. For very small values of the upstream Mach Thu Aug 3 18:54:21 2006 number. The solution to equation (5. Here no such device is needed! Luckily and hopefully.54) and (5. THE MOVING SHOCKS through equations (5. 1 − 2 (the coefficient is only approximated as 0. Sometimes this phenomenon can have a tremendous significance in industrial applications.5) as shown in Figure (5.58) but it seems utilizing numerical methods is much more simple. 5.

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

62) Shock in A Suddenly Open Valve 1.61) A similar rearrangement to the previous case results in T0 y − T 0 x = T y 1 + k−1 −2Msy My + My 2 2 2 (5. The same question that was prominent in the previous case appears now.5 4 3. the relationship .5 0 Number of Iteration 10 0 Shock in A Suddenly Open Valve k = 1 4.3 Fig.3 (b) My = 1.5 1 1 0.9: The number of iterations to achieve convergence.5 3 2. The stagnation temperature (of moving coordinates) is T0 y − T 0 x = T y 1 + k−1 2 (Msy − My ) 2 − Tx 1 + k−1 2 ( Mx ) 2 =0 (5.3.3 Mx My 1.25 2 1.60) Us = Msx cx (5.5. My’ = 1. The “upstream” Mach number is Mx = The “downstream” Mach number is My = Us − U y = Msy − My cy (5.5 1.75 k = 1 4. THE MOVING SHOCKS 97 Similar definitions of the right side and the left side of the shock Mach numbers can be utilized. 5.3 Mx My Ty/Tx Ty/Tx 0.75 5 10 Number of Iteration 15 20 Wed Aug 23 17:20:59 2006 Wed Aug 23 17:46:15 2006 (a) My = 0. what will be the shock velocity for a given upstream Mach number? Again. My’ = 0.59) 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. It has to be noted that the “upstream” and “downstream” are the reverse from the previous case.

64) is a quadratic equation for Msx . If it is satisfactory.0.65) . the larger number of the iterations required to achieve the same accuracy.3 and My = 1.63) can be solved. There are three solutions of which the first one is Msx = 0 and this is immediately disregarded. (b) Calculate the Mach number My by utilizing the tables or Potto–GDC. the temperature. no Mach number is given but pieces of information on both sides of the shock. The larger the value of My . The other two solutions are Msx = (k + 1)Uy ± Uy (1 + k ) 4 cx 2 + 16cx 2 (5. To illustrate the convergence of the procedure. and. The results show that the convergence occurs very rapidly (see Figure (5. Therefore. other conditions ahead of the shock.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. NORMAL SHOCK (Msx ) + 2k k−1 2 (Msx ) − 1 2 k−1 2 (5.3. sufficient results can be achieved after 3-4 iterations. consider the case of My = 0. Yet.9)). for most practical purposes.63) Since Msx can be represented by Msy theoretically equation (5. It is common practice to solve this equation by numerical methods. One such methods is “successive substitutions.” This method is applied by the following algorithm: (a) Assume that Mx = 1. (d) Check the new and improved Mx against the old one.98 between the two sides is Msy = My + CHAPTER 5. Piston Velocity When a piston is moving.64) Ty My + M y Tx Equation (5. In this case. The unknown data are the piston velocity. it creates a shock that moves at a speed greater than that of the piston itself. the calculations for Us can be obtained from equation (5. (c) Utilizing Mx = calculate the new “improved” Mx . stop or return to stage (b).

THE MOVING SHOCKS 99 The negative sign provides a negative value which is disregarded.10: The maximum of “downstream” Mach number as a function of the specific heat. This shock–choking phenomenon Shock in A Suddenly Open Valve Maximum M ’ possible is somewhat similar to the 2. and the only solution left is Msx = (k + 1)Uy + Uy (1 + k ) 4 cx 2 + 16cx 2 (5.66) or in a dimensionless form Msx = (k + 1)Myx + Myx (1 + k ) 4 2 + 16 (5. Consider when the “upstream” Mach approaches infinity. look at the normal Fig.67) Where the “stange” Mach number is Myx = Uy /cx . such as the internal combustion engines and die casting. To explain this phenomenon.25 discussed earlier in a nozzle 2 flow and in other pipe flow mod1.75 suddenly and completely clos0.68) 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.25 has no limit. and the downstream Mach y y(max) Maximum My’ .5 ence is that the actual velocity 1. The limit of the equation when cx → ∞ leads to Msx = (k + 1)Myx 4 (5.5 M choking phenomenon that was 2.68) to explain the next Shock-Choke phenomenon. The differ1. is there a point where the moving shock is fast enough to reduce the “upstream” relative Mach number below the speed of sound? This is the point where regardless of the pressure difference is. Mx = Msx → ∞. 5.5. the shock Mach number cannot be increased. k. k ing of valve results in no limit (at least from the model point Thu Aug 24 17:46:07 2006 of view).5 The spesific heat ratio. shock. Shock–Choke Phenomenon Assuming that the gas velocity is supersonic (in stationary coordinates) before the shock moves. It must be noted 1 that in the previous case of 0. Some use equation (5. This phenomenon occurs in many industrial applications.3.75 els (later chapters). what is the maximum velocity that can be reached before this model fails? In other words.

One can view this as the source of the shock–choking phenomenon. To understand this limit.71) 1− Substituting equations (5. By applying equation when the pressure ratio is approaching infinity Px (5.38).23) to this situation the following is obtained: Msx = k+1 2k Px −1 +1 Py (5. consider that the maximum Mach number is obtained Py → ∞.69) Mx = Msx Thus. NORMAL SHOCK number. the Mach number is ∞    Ty k − 1 =   T 2k   x (5. . A similar phenomenon to the choking in the nozzle and later in an internal pipe flow is obtained.100 CHAPTER 5. the Mach number is approaching infinity because of the temperature ratio but the velocity is finite. 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. is approaching to (k − 1)/2k .72) + When the pressure ratio is approaching infinity (extremely strong pressure ratio). The Mach number is reduced to the maximum value very rapidly.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. These limits determine the maximum velocity after the shock since Umax = cy My . the results is My = 2 k (k − 1) (5. according to equation (5.25) into equation (5. the semi steady state described by the moving shock cannot be sustained.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 (5.26) and (5.73) What happens when a gas with a Mach number larger than the maximum Mach number possible is flowing in the tube? Obviously. From the upstream side.

40 1.61 801.64 255883.74) and equation (5.50000 0.2964 1.53161 0.10 2. The mass flow rate when the pressure ratio is approaching infinity.81 My 0.44 273861.33968 0.43301 0.40825 0.54772 My 2.48667 0.23 873. Thus. the shock–choke phenomenon means that the Mach number is only limited in stationary coordinates but the actual flow rate isn’t.29 188982.8898 1.4434 1.45374 0.90 2. is cy ρy m ˙ = U y ρ y = M y cy ρ y = M y A √ My kPy = RTy kRTy Py RTy (5.13).85 922.93 243332.56 668.1785 1.74 681.25) can be transferred for large pressure ratios into m ˙ ∼ A Ty Px k − 1 Tx k + 1 (5.75) Since the right hand side of equation (5.3 Partially Open Valve The previous case is a special case of the moving shock.70 1.80 1. 5. with the exception of Ty the mass flow rate is approaching infinity when the pressure ratio is approaching infinity.51177 0.3.75) is constant.85 Table of maximum values of the shock-choking phenomenon.37797 0.79 250000.36 270031.05 226871.00 2.78 261117.02 773.54006 0.93048 0. The general case is when one gas flows into another gas with a given velocity. The only limitation is that the “downstream’ gas velocity is higher than the “upstream” gas velocity as shown in Figure (5. .2645 1.50 1.50 Mx 1073.20 2.81786 0.25 985.62 695.54 750.5.87039 0.96 204124.77151 0. ∞. THE MOVING SHOCKS k 1.00 729.0815 1.09 833.6330 1.40 2.00000 0.73029 Ty Tx 101 169842.30 1.99 235702.3.56 711.52223 0.30 2.09 265805.74) Equation (5.86 216507.47141 0.60 1.

(b) Calculate the Mach number My by utilizing the tables or Potto–GDC. 5. 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 (5.102 CHAPTER 5. stop or return to stage (b).v. NORMAL SHOCK ÿ¡ £¢Äÿ¥¤§¦ ÿ¡ ©¨ Uy > U x ô0õÚö Ux ÷#ø c.11: A shock moves into a moving medium as a result of a sudden and complete open valve. (a) Stationary coordinates (b) Moving coordinates Fig. If it is satisfactory.77) (5.v. The procedure for the calculations can be done by the following algorithm: (a) Assume that Mx = Mx + 1. A common problem is to find the moving shock velocity when the velocity “downstream” or the pressure is suddenly increased. It has to be mentioned that the temperature “downstream” is unknown (the flow of the gas with the higher velocity).76) An additional parameter has be supplied to solve the problem. ò îÊï&ðÍñ ó Ux = Us − Ux Upstream ùÊú&ûÍü ýçþ c. Ty (Msy ) − Mx Tx . (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.

4 0.8 1.2 2 2.9 nomenon also occurs in the case where M ’ = 0.5.8 1.7 The mathematics is cumbersome but re0. not the actual flow).8 M ’ = 0.6 sults show that the shock choking phe0. 5. My .2 a faster flow is running into a slower fluid. x x x My y 5.3.4 2. Figure (5. Mx .4 limited.0 0. The figure demonstrates Fig. M ’ = 0.6 M’ bers for various static Mach numbers.12) 0.3. it was shown that the shock chokk=14 ing phenomenon occurs when the flow is 1 running into a still medium. Thu Oct 19 10:34:19 2006 and for various static “upstream” Mach numbers.9 0.4 Partially Closed Valve .3 exhibits some “downstream” Mach num0. This phe0. vicinity of the previous value (see following question/example).12: The results of the partial opening that the maximum can also occurs in the of the valve. THE MOVING SHOCKS 103 Shock in A Suddenly Open Valve Earlier.5 nomenon is still there (the Mach number is 0.

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

296 1. My = √ U kRT =√ 450 ∼ 1. the Mach velocity ahead of the shock has to calculated.5 Worked–out Examples for Shock Dynamics Example 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. Thus. If it is satisfactory.80) My = M s + M y (5. S OLUTION It can be observed that the gas behind the shock is moving while the gas ahead of the shock is still.3.104 CHAPTER 5.1206 My Mx My 1. Assume the specific heat ratio is 1. stop or return to stage (b). it is the case of a shock moving into still medium (suddenly opened valve case). First.0 1.3 × 287 × 300 By utilizing Potto–GDC or Table (5.81) For given static Mach numbers the procedure for the calculation is as follows: (a) Assume that Mx = Mx + 1.4) one can obtain the following table: Mx 2. (b) .953 0.54220 0.604 4.63955 .3.2: 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.132 Ty Tx Py Px P0 y P0 x 0. Ty (Msy ) + Mx Tx 5. NORMAL SHOCK Mx = M s + M x (5. Compute the pressure and the temperature behind the shock.

045 2.0445 My Mx My 0.56994 0.2961.72 sec Example 5.0 0.0 0.0 Ty Tx Py Px P0 y P0 x 0.2) you can find the proper Mx .042 2.724 1.56995 Ty Tx My 0.0 0. THE MOVING SHOCKS Using the above table. The gas is brought into a complete stop by a sudden closing a valve.0 1.3.3: Gas flows in a tube with a velocity of 450[m/sec].044 My 0. The static pressure at the tube is 2Bar and the (static) temperature of 300K .953[Bar] Px The velocity behind the shock wave is obtained √ m Uy = Mx cx = 1.722 1.724 The table was obtained by utilizing Potto–GDC with the iteration request.0 ∼ 4.56995 1.9432 1. the temperature behind the shock is Ty = T y = Ty Tx = 1.2K Tx 105 In same manner.0 0. If this detail is not sufficient then simply utilize the iterations procedure described earlier and obtain the following: i 0 1 2 3 4 Mx 2.2961. The specific heat ratio can be assumed to be k = 1.70009 If you insist on doing the steps yourself.5. Then using Table (5.56995 0.57040 0.044 2. the prime properties can be found. Then. Mx to be 1.724 1.132 × 1.2961 2.4. Simply using the Potto–GDC provides for the temperature and velocity the following table: Mx 2.710 0. .2961 1.604 × 300 ∼ 481. it can be done for the pressure ratio as following Py = P y = Py Px = 4.724 4. At this stage the reflecting shock velocity is unknown. find the upstream prime Mach. S OLUTION The first thing that needs to be done is to find the prime Mach number Mx = 1.953 × 1. Calculate the velocity and the pressure behind the reflecting shock.3 × 287 × 300 ∼ 378.53487 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 (5.5: A gas is flowing in a pipe with a Mach number of 0. S OLUTION Refer to section (5.52778 Ty Tx ρy ρx Py Px P0 y P0 x 2. Calculate the speed of the shock when a valve is closed in such a way that the Mach number is reduced by half.3). Hint.99813 If the information about the iterations is needed please refer to the following table.0789 1. Potto-GDC provides the solution of the above data Mx 1.3.20000 1.0000 3.3020 0.3166 0.4) for the calculation procedure. It can also be obtained from the stationary normal shock wave table.000 6. Potto-GDC provides for this temperature ratio the following table: Mx 2.78928 0.3574 My Mx My Ty Tx Py Px P0 y P0 x 0. .52778 0. this is the case of a partially closed valve case in which the ratio of the prime Mach number is half (the new parameter that is added in the general case).317 0.3574 in the moving shock table provides Mx 2.106 CHAPTER 5.4: What should be the prime Mach number (or the combination of the velocity with the temperature.1583 6.1220 My Mx My Ty Tx Py Px P0 y P0 x 0.55832 using the required Mx = 2.89509 0.0 2.55830 Example 5.3574 My 0.4. NORMAL SHOCK Example 5.40000 0.

0789 1.20000 0.20000 0.89509 0.20000 0.0788 1.89517 0.3037 1.3.1443 1. The piston is Mx = 0.0030 1.0789 1.20000 0.0789 1.1220 My 0.89508 0.20000 0.0836 1.1208 1.20000 0.1220 1.3020 1.89789 0.73971 0.3199 1.3020 1.0789 2.0930 1.20000 0. Calculate the time Fig.92479 0.20000 0.5.89509 0. Assume that there is no friction and the Fanno flow model is not applicable.0802 1.1241 1. 5.0789 1.1222 1.89461 0.20000 0.20000 0.2989 1.20000 0.0545 1.4000 1.0765 1.3020 Example 5.3020 1.0712 1.1288 1.3022 1. S OLUTION Using the procedure described in this section.1967 1.89512 0.0782 1.1220 1.3020 1.2922 1.1200 1.1220 1.20000 1.4 and 300◦C.89494 0.14: Schematic of a piston pushing air in a it takes for the shock to reach the tube.89504 0.89595 0.0045 1.3019 1.3025 1.2705 1.89510 0.20000 0.20000 0.20000 0.1099 1.20000 0.0m length.1216 1.3020 1.2032 1.0106 1.89509 Ty Tx Py Px 107 My 0.1259 1.0832 1.1221 1.89509 0.0789 1.20000 0.1220 1.8.20000 0.3075 1.87903 0.1219 1.3609 1.1226 1.6: A piston is pushing air that flows in a tube with a Mach number of M = 0. Calculate the velocity of the shock created by the piston in the air.89536 0.0789 1.20000 0.0789 1.2547 1.1220 1.0787 1.3017 1.20000 0.20000 0.84424 0.20000 0.1220 1.89354 0.3020 1.1220 1.90416 0.4 My = 0.0790 1.99548 0.0789 1.8 accelerated very rapidly and the air adjoined the piston obtains Mach number M = 0. end of the tube of 1.89009 0.0793 1. the solution is .5041 1.20000 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 1.1182 1.89509 0.3011 1.

81829 0.2380 1.50000 0.1519 2.1517 1.80000 0.1519 1.80000 0. i 0 1 2 3 4 5 6 Mx 1.2380 My Mx My CHAPTER 5.98860 The complete iteration is provided below.6273 1.6215 1.5000 1.3202 1.80000 0.5834 1.7: From the previous example (5.108 Mx 1.82716 0.2248 1.81940 0.2380 − 0.1519 −0.10) calculate the velocity difference between initial piston velocity and final piston velocity.80000 0.80000 1.4) Example 5.4 × 287 × 300(1.1531 1. S OLUTION The stationary difference between the two sides of the shock is: ∆U =Uy − Ux = cy Uy − cx Ux   √   1.81943 0.1519 1.1435 1.1519 1.80000 1. NORMAL SHOCK Ty Tx Py Px P0 y P0 x 0.2378 1.2381 1.6216 The time it takes for the shock to reach the end of the cylinder is t= length Us cx (Mx −Mx ) =√ 1 = 0.2400 1.5    q Ty Tx =  ∼ 124.81958 0.81942 0.6215 0.2380 My 0.4583 1.0034[sec] 1.6207 1.6217 1.70109 0.81942 Ty Tx Py Px My 0.8 × 1.4[m/sec] .80000 0.4 × 287 × 300 0.80000 0.

0 My 0. 1.0 The time for the shocks to collide is t= 1[m] length = ∼ 0.95890 Ty Tx Py Px P0y P0 x Fig.93451 Ty Tx Py Px P0y P0 x Uy cx 347. 5.66).173 0. Calculate the time it will take for the two shocks to collide. On one side the pressure is high.0013[sec] Usx 1 + Usx 2 (1. 109 1 [Bar] 300 K 40 m/sec shock waves 70 m/sec S OLUTION This situation is an open valve case where the prime information is given.4 Shock Tube The shock tube is a study tool with very little practical purposes. The distance between the two pistons is 1[m]. The solution is given by equation (5. SHOCK TUBE Example 5. It is used in many cases to understand certain phenomena.0715 My 0. The air between the pistons is at 1[Bar] and 300K . For this case the following table can easily be obtain from Potto–GDC for the left piston Mx 1. expansion waves occur with a reduction of pressure.15: Figure for Example (5.047 1.4.99959 40.0 My 0. At the back of the shock.0715 + 1.5. Other situations can be examined and extended from these phenomena.15)). This is the same case as in the suddenly opened valve case described previously. it is the explicit analytical solution.8) Uy cx 347. 1.083 1.93471 Mx 0. The temperature is known to reach several thousands degrees in a very brief period of time. The high pressure . and.99785 70. A cylinder with two chambers connected by a diaphragm. When the diaphragm is ruptured the gas from the high pressure section flows into the low pressure section.318 0. a shock is created that it travels to the low pressure chamber. while the pressure on the other side is low.1283 My 0.8: An engine is designed so that two pistons are moving toward each other (see Figure (5.1283)347. 5.89048 Mx 0.0 while the velocity of the right piston is Mx 1. When the pressure is high enough.

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

88) After rearranging equation (5.86) (5.111 √ range of the speed of sound.4. A typical design problem of a shock tube is finding the pressure to achieve the desired temperature or Mach number.89) . c ∼ 340 thus the whole process takes only a few milliseconds or less.85) The fluid velocity in zone 2 is the same U2 = U s − U 2 = U s 1 − From the mass conservation. a review of the material is provided thus far. Thus.88) the result is U2 = c1 k P2 −1 P1 2k k+1 P2 k−1 P1 1+k (5.). The pressure ratio between the two sides of the shock is 5.83) And the velocity ratio between the two sides of the shock is k+1 P2 1+ k ρ2 U1 −1 P1 = = k+1 P2 U2 ρ2 k−1 P 1 (5. it follows that U2 ρ1 = Us ρ2 U2 Us (5. The relationship between the different properties was discussed earlier and because it is a common problem.82) k − 1 k + 1 P2 + 2k 2k P1 (5.87) U2 = c 1 k − 1 k + 1 P2 + 2k 2k P1 1− 1+ k+1 k−1 P2 P1 k+1 P2 k−1 P1 + (5. these kinds of experiments require fast recording devices (a relatively fast camera and fast data acquisition devices.84) 2k Ms1 2 − 1 k−1 (5. SHOCK TUBE k−1 P2 = P1 k+1 Rearranging equation (5.82) becomes Ms1 = Or expressing the velocity as Us = Ms1 c1 = c1 k − 1 k + 1 P2 + 2k 2k P1 (5. The following equations were developed earlier and are repeated here for clarification.

112 CHAPTER 5. in zone 4.94) (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. d(ρU ) = 0.89) 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 (5.95) and equation (5.91) dρ (5. and dividing by the continuity equation the following is obtained: dU dρ =− ρ c (5. hence equation (5.92) Equation (5.96) is transformed into P5 P2  1 − = √ P1 P1   (k − 2k c1 1) c 5 P5 P3 −1 P2 P1 k − k2 −1 2k + (k + 1) −1    (5. applying the isentropic relationship (T ∝ ρk−1 ) yields c = c5 From equation (5.96) After some rearrangement.97) . taking the derivative of the continuity equation.95) As it was mentioned earlier the velocity at points 2 and 3 are identical. NORMAL SHOCK On the isentropic side. equation (5.92) can be integrated as follows: U3 ρ3 dU = U5 =0 ρ5 c5 ρ ρ5 k −1 2 dρ (5.93) The results of the integration are 2c5 U3 = k−1 Or in terms of the pressure ratio as 2c5 U3 = k−1 1− P3 P5 k −1 2k 1− ρ3 ρ5 k −1 2 (5.

5. Find the shock velocity and temperature behind the shock if the pressure P5 ratio is P = 40? 1 S OLUTION 5. the assumption of an uniform flow in the duct is closer to reality.99) By utilizing the isentropic relationship for zone 3 to 5 results in k5 −1 k5 k5 −1 k5 T3 = T5 P3 P5 = P2 P1 P5 P1 (5.5. Ms1 k−1 c1 k+1 c5 113 k1 − 1 P5 = P1 k+1+1 2k Ms1 2 − 1 k1 − 1 1− Ms1 Ms1 2 − 1 k − k2 −1 (5. It is normal to have a large Mach number with a large Re number. For a high Reynolds number. In that case.5 5. the assumptions in construction of these models are acceptable and reasonable. .7 Shock with Real Gases Shock in Wet Steam Normal Shock in Ducts The flow in ducts is related to boundary layer issues.100) Example 5.6 5. the following is obtained: 1+ T2 = T1 1+ k1 −1 k1 +1 k1 −1 k1 +1 P2 P1 P1 P2 (5.98) Using the Rankine–Hugoniot relationship and the perfect gas model. SHOCK WITH REAL GASES Or in terms of the Mach number.9: 5 A shock tube with an initial pressure ratio of P P1 = 20 and an initial temperature of 300K .

Therefore.17)).0 5. NORMAL SHOCK 5.668 Ty Tx Py Px P0 y P0 x 0.057811 The direct calculation will be by using the “upstream” Mach number. 668 sqrt 1 .41404 0. The temperature should not increase above a certain value.0850.8 More Examples for Moving Shocks Example 5. the time is t= distance 3 √ = ∼ 0. air is Fig. The time it takes for the material from the valve to reach the exit is distance 3 t= = ∼ 0. Where any of these advises make any sense in the light of the above assumptions? What will be your recommendations to the manufacturing company? Plot the exit temperature and the mass flow rate as a function of the time. Comment on this proposal.10) supposed to reach the pipe exit as quickly as possible when the valve is opened (see Figure (5.0017[sec] 5. The time it takes for the shock to travel from the Py valve depends on the pressure ratio Px = 30 Mx 5. Assume that there isn’t any resistance whatsoever in the pipe.114 CHAPTER 5.967 My kRTy . In this system.0021[sec] 1 . S OLUTION This problem is known as the suddenly open valve problem in which the shock choking phenomenon occurs.0002 [sec] to be opened). 4 × 287 × 300 × 5. An engineer is required to design a cooling system for a critical electronic deexit valve vice.00 0. Mx = Msx = 5. Assume that the time scale for opening the valve is significantly smaller than the typical time of the pipe (totally unrealistic even though the valve manufacture claims of 0.0850 My Mx My 1.0850sqrt1.10: This problem was taken from the real industrial manufacturdistance ing world.967 30.17: Figure for Example (5. 5. The ambient temperature is 27◦ C and 1[Bar]. opening valve probelm The distance between between the valve and the pipe exit is 3[m].4 × 287 × 300 Msx kRTx The mass flow rate after reaching the exit under these assumptions remains constant until the uncooled material reaches the exit. the 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. After building the system. The conditions upstream of the valve are 30[Bar] and 27◦ C .

15 1. K = 1. There is also the possibility of steps increase in which every step heat released will not be enough to over heat the device.30 My 1.4 Mx 1.9. The plot of the mass flow rate and the velocity are given in Figure (5.00000 1. It can only increase the rate after the initial heating spike A possible solution is to have the valve very close to the pipe exit.25 1. Try to estimate the temperature increase of the product. The last possible requirement a programmable valve and very fast which its valve probably exceed the moving shock the valve downstream. 5.08398 1.12799 1.11958 1.11: Example (5.05 1.51570 1.65625 1.97937 . k = 1.00 1.09658 1.42857 1.34161 1.99893 0. the heating time is reduced significantly. Time[Msec] To increase the pipe diameter will not change the temperature and therefore Fig.00000 0.1: The shock wave table for k = 1.20 1.25504 1.4 IDEAL GAS 115 During that difference of time the material is get heated instead of cooling down because of the high temperature.16908 1.37625 1.99669 0.51333 1.87502 0.91177 0.84217 0.78596 Ty Tx ρy ρx Py Px P0y P0x 1.10 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.03284 1.80500 1. TABLES OF NORMAL SHOCKS. Thus. Example 5.00000 1.98706 0.95313 0.99280 0.4 Ideal Gas Table 5.9 Tables of Normal Shocks.00000 0.10) will not change the effects of heating.99985 0.10) deals with a damaging of electronic product by the temperature increase.15938 1. Plot the pipe exit temperature as a function of the time.81264 0.5. S OLUTION 5.06494 1.19087 1. this suggestion contradicts the purpose of the required manufacturing needs.00000 1. Thus.18: The results for Example (5.18).24500 1.

79023 0.43496 0.85714 4.40625 3.02498 .07869 5.59562 0.48000 5.41908 0.36906 7.56935 1.45 2.81322 1.12500 8.21295 0.97833 5.41189 0.73971 0.95819 0.56148 0.03175 2.32022 1.17166 0.90138 1.08846 2.47519 0.75 1.00 6.45833 1.50 3.09170 0.51568 2.83457 0.96974 0.49458 1.72855 1.33333 3.60 1.32834 0.90625 23.01863 3.29878 6.95122 3.80000 6.55395 0.91319 0.04500 4.26829 5.45 1.05100 0.70109 0.70 1.07505 0.22625 5.67420 0.56906 0.00500 6.63625 2.62814 0.98511 3.13876 0.25 2.35922 2.68410 0.50000 20.25469 1.15625 14.11525 2.54014 0.60553 0.35 2.40897 0.80 1.61333 3.40625 41.52312 0.74420 0.04033 2.40642 0.89520 0.40623 0.88231 2.86207 1.94680 1.91156 5.00 3.62809 0.00 2.52861 0.57735 0.82180 7.00000 31.54055 0.87599 0.85686 1.67901 2.59188 2.45115 0.05 2.55333 6.69919 4.10 2.50 1.85572 0.13750 2.25 4.26958 4.50 4.98958 35.58329 0.81190 2.32544 5.07229 4.27907 2.45833 26.94484 0.65625 10.53637 1.49181 0.45833 2.54706 0.68966 1.82625 4.66844 0.76736 0.00 4.4 (continue) Mx 1.68750 1.74002 2.42355 0.61194 3.57143 4.21182 5.40625 0.61650 0.40 1.33333 12. NORMAL SHOCK Table 5.92979 0.02965 0.75 3.11256 0.28720 1.00000 5.51299 0.65105 0.77045 1.95 2.20500 3.40 2.66667 2.40416 0.64729 1.83625 7.1: The shock wave table for k = 1.43811 2.08455 3.56128 0.25 5.73625 4.72087 0.76175 0.12500 38.15 2.15625 29.31915 1.85 1.82000 3.64054 0.50 2.60570 0.28625 2.116 CHAPTER 5.60278 1.95958 2.94732 2.50000 4.66894 4.99311 2.35379 1.42280 1.31505 3.26451 0.19772 2.00 5.50 5.65396 0.27625 6.53158 1.04236 0.25 3.12000 2.71956 0.94059 8.39657 2.25 My 0.12500 16.60792 1.87509 5.06172 0.44231 0.26087 4.22261 1.83333 45.81268 0.65 1.90 1.14894 5.75 5.41523 0.20 2.04688 4.42878 0.58618 0.53441 0.51931 0.55 1.00958 3.46192 0.40216 Ty Tx ρy ρx Py Px P0y P0x 1.51792 0.49901 0.73958 6.44891 4.38797 1.03536 0.42623 4.69751 0.75 4.77614 1.35 1.75 6.23958 18.14897 3.27335 3.21190 3.30 2.81188 4.

014 1.016 1.62095 13.018 1.38750 5.0 0.766 My 0.008 1.92832 0.676 1.99997 0.028 1.0 0.055 1.118 1.00849 0.127 1.9.0 0.98227 0.56522 5.036 1.97099 0.61165 5.60 0.149 1.38807 0.38980 0.45833 69.65625 105.0 0.38672 14.12500 110.00000 0.98446 0.0 0.70019 5.08 0.39385 0.00 7.80 0.00739 0.09 0.323 1.50 9.15625 94.780 3.423 1.62425 Mx 0.38758 Ty Tx ρy ρx Py Px P0y P0x 9.475 2.17678 14.25 9.50000 0.39289 0.0 0.15643 9.76452 0.99973 0.268 1.037 1.40038 0.024 1.00343 0.39491 0.99980 0.506 1.67171 0.00387 0.75 9.082 1.82978 16.00304 Table 5.05 0.84463 0.1: The shock wave table for k = 1.99113 15.062 1.89918 0.70 0.99317 0.99991 0.95944 0.028 1.0 0.38860 0.38918 0.57997 18.73958 116.39121 0.42736 20.112 3.04 0.195 2.99790 0.4) Mx 1.36508 5.00 9.39879 0.040 1.44444 5.133 1.96506 0.0 0.020 1.40 0.58939 5.01316 0.15625 65.344 1.00979 0.00 8.80251 0.00565 0.5.4 IDEAL GAS Table 5.23958 84.75 7.25 7.0 0.33333 99.012 1.043 1.39607 0.0 0.024 1.0 0.00496 0.73029 0. TABLES OF NORMAL SHOCKS.087 1.95231 0.589 1.99994 0.442 1.99999 0.01 0.473 1.95387 0.66874 5.99403 0.0 Ty Tx Py Px P0y P0 x 1.39736 0.0 0.032 1.98812 0.502 1.02115 0.90 1.25 8.02 0.49152 19.68504 5.505 1.126 1.50 8.86537 0.65116 5.00000 61.12500 52.00645 0.50000 79.10 0.269 1.004 1.030 1.0 0.12500 89.71429 49.51020 5.30 0.4 (continue) 117 Mx 6.171 1.0 0.00437 0.710 1.64673 0.97647 0.00 My 0.2: Table for a Reflective Shock from a suddenly closed end (k=1.50 0.00000 1.47883 5.97074 0.63218 5.82755 .941 2.39201 0.0 0.20 0.16229 11.049 1.69273 17.00 My 0.006 1.50 6.316 1.219 1.0 0.50 7.69946 0.90625 74.01798 0.99998 0.196 1.46939 11.87948 12.40667 5.043 1.39048 0.102 1.01535 0. K = 1.0 0.03 0.012 1.89128 0.0 0.94291 0.99986 0.80077 10.98958 57.53890 5.057 1.01133 0.381 1.07 0.75 8.06 0.75 10.94836 0.072 1.

990 2.99763 0.41 15.94746 0.831 5.09 0.47855 0.03 0.51996 0.206 5.0 0.0 0.4) Mx 1.210 1.0 0.08 My 0.048 2.4E +2 1.0 0.20 0.02 0.024 1.256 6.613 My 0.004 1.118 CHAPTER 5.104 1.56312 0.94180 0.99999 0.60761 0.0 My 0.99985 0.50 0.018 1.146 2.0 0.087 2.0 0.99994 0.0 0.006 1.952 2.00000 0.297 2.08 0.0 0.05 0.60401 0.0 0.62 84.96465 0.70 1.058 1.728 5.50 1.0 0.98216 0.0 0.3: Table for shock propagating from suddenly opened valve (k=1.0 0.033 1.026517 0.51808 0.331 1.0 0.00 5.0 0.827 7.00 7.83607 0.0 0.073 1.552 2.65290 0.188 1.868 1.0 Ty Tx Py Px P0 y P0 x 1.087 2.4)(continue) Mx 1.38608 0.66462 Mx 0.40 0.88717 0.734 17.89 12.60 0.99998 0.99402 0.162 7.397 1.10 1.55453 0.280 4.98019 0.245 2.80 1.0 0.95315 0.0 0.50814 0.49912 0.248 1.90 2.00125 Table 5. NORMAL SHOCK Table 5.00 8.495 1.715 6.703 10.99181 0.0 0.057 1.39116 0.98807 0.40 1.656 2.317 1.727 1.031 1.0 0.014 1.14 62.295 1.136 1.0 0.05 19.30 1.38457 Mx 1.49092 0.39566 0.050 1.04 0.00350 0.0 0.431 8.10 0.00000 1.410 3.43921 0.810 1.38817 0.20 1.020 1.33 3.58578 0.70 Ty Tx Py Px P0 y P0 x 1.028 1.70283 0.061716 0.858 1.037 1.029 1.01 29.086 1.0 0.41523 0.0 0.576 1.0 My 0.60 1.40284 0.0 0.024 1.133 1.00639 0.0 0.00 10.69834 0.762 3.93133 0.74316 0.0 0.00 3.120 1.189 2.52904 0.2: Table for Reflective Shock from suddenly closed valve (end) (k=1.791 2.00204 0.325 11.044 1.016 1.43894 0.0 0.00 6.063 1.541 1.96069 0.441 2.066 8.859 5.0 0.97629 0.390 1.0 0.012492 0.088 1.0 0.89039 .346 2.862 4.7E +2 0.15637 0.000 6.30 0.00 44.00 9.897 1.649 1.008 1.517 9.99990 0.56935 0.78652 0.74403 0.06 0.134 1.448 2.47 1.012 1.0 0.336 8.012 1.25 24.00 4.037 1.0 0.041 1.1E +2 1.95888 0.78840 0.97045 0.043 1.152 1.99996 0.99971 0.40213 0.54114 0.800 8.07 0.99979 0.21 29.01 0.

0 0.00272 0.387 4.83661 0.24 11.040 0.029 My 0.74 66.050 Ty Tx Py Px P0y P0 x 1.61 1.830 1.0 0.1E +3 1.040 1.400 1. K = 1.62 17.9E + 3 5.18E −5 0.896 2.3E +2 7.889 1.300 1.744 1.067 1.79 My 0.810 1.62923 0.33 21.7E +2 9.3E +2 7.0 0.2E + 2 1.167 3. TABLES OF NORMAL SHOCKS.003 1.84 1.0 0.00000 1.38713 0.889 1.840 1.0 My 0.889 1.80 76.37944 0.59699 0.98290 0.76940 0.0 0.0 0.6E + 3 4.013 1.00000 0.18575 0.0 0.508 2.42622 0.54 15.536 4.0 0.37810 Mx 0.37814 0.83 71.40843 0.68907 0.97166 Mx 0.25 12.040812 0.840 7.0 0.39187 0.0 0.888 1.888 1.975 11.800 1.2E +2 9.90 1.07 90.200 1.46599 0.97726 0.38870 0.012 1.145 2.0 0.604 1.38974 0.02 83.00395 0.0 0.37813 0.09 35.69 68.49586 0.56619 0.3: Table for shock propagating from suddenly opened valve (k=1.92 42.00544 0.889 Ty Tx Py Px P0 y P0 x 1.3) Mx 1.00101 0.0 0.2E + 3 5.023 1.5E + 3 8.0 0.5E + 2 3.72 80.171 8.80 0.95 64.98857 0.53 63.823 5.78 80.020 0.38096 0.2E +2 8.012 1.37820 0.4 IDEAL GAS Table 5.0 0.9.100 1.99999 0.4E + 3 6.821 9.919 2.000497 0.931 14.9E + 3 6.0 0.889 1.0 0.500 1.054 1.2E +3 1.00721 0.368 5.9E + 2 2.0 0.0 0.600 1.0 0.0E + 2 1.0058 1.99 25.4E +3 3.7E +2 8.551 8.0 Table 5.068 2.799 3.450 2.0 0.62 14.35 25.37817 0.48823 0.881 3.51223 0.0 0.37812 0.9E + 3 7.39028 0.025 4.0 0.0 0.0 0.0 0.745 1.31 62.658 4.888 1.4: Table for shock propagating from a suddenly opened valve (k=1.62 61.8E + 2 7.491 1.860 1.37816 0.0 0.00175 0.99997 .32 16.269 2.0 0.07 18.457 10.1E +3 1.2E + 3 0.646 6.017 1.0 0.0 0.880 1.59649 0.5.9E +2 1.53817 0.0 0.57 31.010 0.293 8.0 0.009 1.0 0.5E + 3 5.00 1.0 0.10216 0.45 22.4E + 3 4.006 1.0 0.030 0.99998 0.700 1.38402 0.850 1.18 73.0 0.026 1.870 1.4) 119 Mx 1.28412 0.5E + 2 1.44536 0.0 0.888 1.7E + 2 4.820 1.0 0.38248 0.0 0.0 0.38557 0.888 1.0 0.37818 0.37822 0.000181 3.015 1.99427 0.37821 0.0 My 0.

611 4.58 23.96 0.0 0.900 1.681 1.92 23.30 23.58223 0.614 4.84227 0.200 1.45807 0.110 1.252 2.39624 0.202 3.0 0.0 0.840 1.612 4.96610 0.86274 0.052949 0.39027 0.997 3.800 1.060462 0.016 4.42089 0.3) Mx 1.103 1.088718 0.035 1.0 0.880 1.482 4.197 1.099 4.553 1.96631 0.467 2.644 2.80734 0.471 4.053088 0.060 0.592 3.0 0.66575 0.75301 0.707 5.706 1.95 23.39026 Mx 0.772 9.46 14.888 1.600 0.0 0.0 0.889 1.888 1.43882 0.0 0.093988 0.923 4.349 4.99995 0.889 Ty Tx Py Px P0 y P0 x 1.00 1.860 1.052879 0.426 2.749 3.042 1.452 1.052984 0.048 1.381 3.39031 0.99792 0.073863 0.083607 0.613 4.125 1.080 0.0 0.0 0.99987 0.870 1.140 1.39160 0.700 1.79611 0.0 0.49 20.286 4.100 4.448 3.31281 0.100 0.034 4.40257 0.90734 0.523 7. NORMAL SHOCK Table 5.0 0.078654 0.95 19.400 0.959 2.888 1.400 1.95506 0.519 3.052809 .052914 0.889 1.96056 0.842 3.367 11.99991 0.25 21.0 0.4: Table for shock propagating from a suddenly opened valve (k=1.0 0.053053 0.136 1.39033 0.99975 0.0 0.55074 0.109 4.820 1.126 1.47875 0.177 1.0 0.98290 0.95 23.054 1.098 1.0 0.64 21.91 23.94415 0.615 4.49333 0.081 1.275 1.0 0.069233 0.500 0.274 1.64073 0.800 0.850 1.669 3.236 2.191 2.0 0.0 0.097 4.89159 0.031 1.94156 0.0 0.810 1.553 4.024 1.060 1.663 1.39037 0.0 0.39035 0.200 0.064766 0.0 0.39314 0.830 1.0 0.615 4.91 23.021 1.94959 0.0 0.063 1.0 0.0 0.616 4.39780 0.0 0.67546 0.32 18.74136 0.053018 0.070 0.815 1.595 2.99288 0.0 0.120 CHAPTER 5.39938 0.22904 0.0 0.225 4.785 1.522 6.898 3.576 4.415 4.57853 0.0 0.100 1.0 0.300 0.44 18.028 1.359 1.52495 0.0 0.090 0.40226 0.489 1.93 23.166 4.71284 0.302 1.889 1.94 23.500 1.90 22.223 1.39468 0.39029 0.0 0.056322 0.888 1.400 1.15495 0.0 0.834 3.05 20.0 0.919 2.613 4.99981 0.300 1.60847 0.052844 0.478 1.069 2.053 4.0 0.93 23.980 2.333 1.098 4.888 1.50100 0.0 My 0.0 0.39034 0.600 1.40097 0.099 4.40418 0.018 1.570 1.102 4.700 0.101 4.39030 0.616 My 0.096 4.096 1.

97 0.4: Table for shock propagating from a suddenly opened valve (k=1.4 IDEAL GAS Table 5.9.3) 121 Mx 4.5.617 My 0.0 My 1.39025 Mx 0. TABLES OF NORMAL SHOCKS.052775 .104 23. K = 1.889 Ty Tx Py Px P0y P0 x 4.

NORMAL SHOCK .122 CHAPTER 5.

CHAPTER 6 Normal Shock in Variable Duct Areas In the previous two chapters. the flow in a convergent– divergent nuzzle was presented when the pressure ratio was above or below the special range. 6. P B is between sures w ic flo subson shock a after 123 . but also in steady state cases when there is no supersonic flow (in stationary coordinates).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. This Chapter will present the flow in this special range of pressure ratios. When the Fig. As it was shown in Chapter 5. As it is was presented before. the gas has to pass through a converging–diverging nozzle to obtain a supersonic flow. d In Figure (6.1) the reSupersonic duced pressure distribution in !¡" b the converging–diverging nozdistance. 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). the shock can occur only in steady state when there is a supersonic flow. x zle is shown in its whole range of pressure ratios. In the previous chapter.

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

It has to be realized that for a large tank.54743 Ty Tx ρy ρx Py Px P0y P0 x 1.1).1972 My 0. From equation (5. as Ae A∗ = 1. Ay /A∗ is known and the exit area can be calculated as Ae Ay 9 Ae = × ∗ = 1. My can be obtained.0268 Again utilizing the isentropic relationship the exit conditions can be evaluated.1972 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.50877 0. My .54746. the Mach number at the exit can be calculated by utilizing the isentropic relationship. For example.8882. in the case of ratio of 100 the Mach number is 0. one can obtain using the isentropic relationship .49) or the GDC– Potto.9474 5.54743 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.125 First.09393 0. Mx 2.0000 0. Thus.197 as shown table below: M 2. but can be shown that the correction is negligible for a typical dimension ratio that is over 100. A∗ . Ax 6 = =2 A∗ 3 With this ratio (A/A∗ = 2) utilizing the Table (5. the Mach number.94345 0. With known Mach number the new star area ratio. Mx is about 2.8544 2. the inside conditions are essentially the stagnation conditions (this statement is said without a proof. The star area (the throat area).1972 the Mach number.2) or from equation (4.18463 2.81568 1. This Mach number can be calculated by utilizing the isentropic relationship from the large tank to the shock (point “x”). With these values.4656 0. the stagnation temperature and pressure are known T0 = 308K and P0 = 4[Bar].00587 and the error is less than %0. Then the relationship developed for the shock can be utilized to calculate the Mach number after the shock.1) or equation (4.2) My ∼ = 0.22) or from Table (4. the exit Mach number has to be determined.2588 0. From the Mach number after the shock. Mx = 2. (point “y”).2588 × = 1.86457 1. before the shock is known and given as well.62941 From Table (4.18787 With this Mach number. the subsonic branch can be evaluated for the pressure and temperature ratios.8882 ∗ A Ay A 6 with this area ratio.11) the following Table for the isentropic relationship is obtained M 0.

126 M 0.854 × 0.97318 0. In that case we don’t have to go through that shock transition.92882 × ∼ =2.951 Ty Tx Tx T0 T0 =0.0000 0.6728[Bar] .466 × 0.97912 0.11310 3. Pexit = Pexit P0 P0 Py Py Px Px P0 P0 =0. the exit conditions can be calculated.98077 0.14190 Pexit = Pexit P0 P0 = 0.97[Bar] For the supersonic sonic branch Pexit = Pexit P0 P0 = 0.81568 × 5.04730 2. The area ratio for both cases.8882 0.41820 × 4 ∼ =1.509 × 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.19745 2.6374 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.41820 0.99226 0.94862 1.9195 0.094 × 4 T0 Ty 1 0.9K × 1.92882 1. Yet we have to pay attention that there two possible back pressures that can “achieve” it or target.99226 × 4 ∼ =3. is A/A∗ = 3 In the subsonic branch (either using equation or the isentropic Table or GDC-Potto as M 0.34[Bar] The exit temperature is Texit = Texit T0 1 0.0000 3. NORMAL SHOCK IN VARIABLE DUCT AREAS T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.98133 × ∼ =299.32651 CHAPTER 6.7538 Since the stagnation pressure is constant as well the stagnation temperature.

7K T∗   R  T0 T0  T∗ A √ kRT ∗ = P∗ P0 P0 R T∗ T0 A kR T0 T∗ T0 T0 The temperature at the throat reads P∗ = The speed of sound is c= √ 1. While the pressure to achieve full supersonic flow through the nozzle the pressure has to be below the 42% the original value. Thus. .7 = 321. In these applications a small pressure difference can produce a shock wave and a chock flow.12[m/sec] P∗ P0 P0 = 0.1) where would be shock’s location when the back pressure is 2[Bar]? 1 The meaning of the word practical is that in reality the engineer does not given the opportunity to determine the location of the shock but rather information such as pressures and temperature. these kind situations exist.13[kg/sec] 287 × 256.113[Bar] And the mass flow rate reads m ˙ = 4105 3 × 10−4 × 321.12 = 0.833 × 308 = 256. Example 6.5283 × 4 = 2.7 It is interesting to note that in this case the choking condition is obtained (M = 1) when the back pressure just reduced to less than 5% than original pressure (the pressure in the tank).4 × 287 × 256.127 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. 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. In fact in many industrial applications. over 50% of the range of pressure a shock occores some where in the nozzle.2: In the data from the above example (6. For more practical example1 from industrial application point of view.

.905 4 Looking up in the Table (4. see also equation (4.113[Bar]. NORMAL SHOCK IN VARIABLE DUCT AREAS S OLUTION The solution procedure is similar to what was shown in previous Example (6.52628 Ty Tx ρy ρx Py Px P0 y P0 x 2. the computer can be use to carry this calculations in a sophisticate way. there are two main possible ways to obtain the solution.38034 0. Utilizing the Table (4.5[unitless!] Px 0 × A x ∗ Py 0 × A y ∗ 4×3 A With the knowledge of the ratio PP ∗ which was calculated and determines the exit 0A Mach number. The ratio of the ratio of stagnation pressure obtained by f or Mexit P0 y = P0 x P0 y Pexit Pexit P0 x = 2 1 × = 0. it requires some iterations by “smart” guessing the different shock locations. The exit total pressure can be obtained (if needed). utilizing the isentropic Table (4.0128 3. In fact.1). the exit Mach number. The second (recommended) method is noticing that the flow is adiabatic and the mass flow rate is constant which means that the ratio of the P0 × A∗ = Py0 × A∗ |@y (upstream conditions are known. 2 Of course. Here.5000 0.5525 0.93118 1.3709 My 0. More importantly the pressure ratio exit is known.2). Thus.2) or utilizing the GDC-Potto provides Mx 2.g.75158 With these values the relationship between the stagnation pressures of the shock are obtainable e. the needed pressure is only 2[Bar] which means that the next guess for the shock location should be with a larger area2 . the exit pressure is between point “a” and point “b”.3914 0.2) or the GDC-Potto provides the following table is obtained T T0 ρ ρ0 A A P P0 A×P A ∗ ×P 0 F F∗ M 0. Pexit Aexit 2×9 Pexit Aexit = = = 1.97188 0.72)). The area (location) that the previous example did not “produce” the “right” solution (the exit pressure was 2.6575 0. It follows that there must exist a shock in the nozzle.1755 6. is known. In the first method. The conditions in the tank are again the stagnation conditions. the previous example information used and expanded. Mathematically.55250 With the information of Mach number (either Mx or My ) the area where the shock (location) occurs can be found.90500 1. The solution process starts at the nozzle’s exit and progress to the entrance.128 CHAPTER 6. First. My .

3396 × 3 ∼ = 7. Vc Vc = √ η= (Uactual )2 (Uideal ) 2 h0 − hexit (Uactual )2 = 2 h0 − hexit s (Uideal ) (6. In the literature some define also velocity coefficient as the ratio of the actual velocity to the ideal velocity.1 Nozzle efficiency Obviously nozzles are not perfectly efficient and there are several ways to define the nozzleefficiency.47076 0.3) (6.5) 6.1) where hexit s is the enthalpy if the flow was isentropic.6.1.2 Diffuser Efficiency . The actual energy that was used is E = h0 − hexit (6.15205 2. 6.07158 0.16747 Approaching the shock location from the upstream (entrance) yields A= A ∗ A = 2.3709 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 129 0. 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.99. NOZZLE EFFICIENCY M 2. The total energy that can be converted is during isentropic process is E = h0 − hexit s (6.2) The efficiency can be defined as η= The typical efficiency of nozzle is ranged between 0.4) There is another less used definition which referred as the coefficient of discharge as the ratio of the actual mass rate to the ideal mass flow rate. as “simple” check this value is larger than the value in the previous example. Cd = m ˙ actual m ˙ ideal (6.0188[cm2] A∗ Note.3396 0.9 to 0.

7[Bar] and temperature of 250K . The cross section in area between the nuzzle Compressor and diffuser is 0.7) s. S OLUTION The condition at M = 3 is summarized in following table M 3. 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.3: A wind tunnel combined from Diffuser nozzle a nozzle and a diffuser (actually two nozzles connected by a 1 f£gMh 2 3 i£jMk 4 constant area see Figure (6.8) Example 6.02722 0.4: Schematic of a supersonic tunnel in a continoccurs in the test section.0 and prescapacitor sure of 0.130 CHAPTER 6.35714 0.3: Description to clarify the definition of diffuser efficiency And further expanding equation (6.02[m2 ].entropy Fig.11528 0.4 can be assumed. Assume that a shock Fig.07623 4.4)) the required condition at point 3 are: M = 3.6) P02 P2 01 02 2 P1 1 For perfect gas equation (6. η= 2(h3 − h1 ) h3 − h 1 = h01 − h1 U1 2 (6.3) 0.0000 T T0 ρ ρ0 A A P P0 uous region (and also for example (6. 6. 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.65326 A×P A ∗ ×P 0 F F∗ .6) can be converted to η= 2Cp (T3 − T1 ) U1 2 (6. 6.2346 0. k = 1.7) results in η= kR 2k −1 T1 T3 T1 − c1 2 M1 2 1 = 2 k−1 T3 T1 M1 2 −1 2 = 2 M1 (k − 1) T3 T1 k−1 k −1 (6.

0047 ∼ 0.2963 34.3333 0.6790 3.32834 A∗ d = P0 n ∗ 1 A n∼ 0. The “downstream” shock number is Msy = Us k ∗ 287.37554 0.2346 = 0.2.479 34.0 5.0000 My 0. Calculate the conditions after the shock.5346 My Mx My 1.02/4.8571 10.3.50 0. For example Potto-GDC (this code was produce by the program) Mx 5.989 Ty Tx Py Px P0 y P0 x 0.02172 The calculation of the temperature and pressure ratio also can be obtain by the same manner.0143[m3] P0 d 0. DIFFUSER EFFICIENCY The nozzle area can be calculated by A∗ n = A A = 0.37554 Ty Tx ρy ρx Py Px P0y P0 x 5. P0 A∗ is constant (constant mass flow).021717 The calculations were carried as following: First calculate the Mx as M x = Us / (k ∗ 287.4: A shock is moving at 200 [m/sec] in pipe with gas with k = 1.09668 .32834 Example 6.4968 0. pressure of 2[Bar] and temperature of 350K .5346 My 0. temperature and pressure “upstream” known. ∗ Tx ) Then calculate the My by using Potto-GDC or utilize the Tables. S OLUTION This is a case of completely and suddenly open valve with the shock velocity.0047[m2] A 131 In this case.6.47519 Ty Tx ρy ρx Py Px P0y P0 x 2. In this case Potto–GDC provides the following table Mx 5. ∗ Tx ∗ Ty Tx ∼ 2.4789 6. First the stagnation behind the shock will be Mx 3.

the “upstream” Mach is given.75593 Ty Tx ρy ρx Py Px P0 y P0 x 1.16[m/sec] Example 6. If the steady state is achieved. S OLUTION This is the case of a closed valve in which mass flow rate with the area given. For this pressure ratio of Py /Px = 2 the following table can be obtained or by using Potto–GDC Mx 1.6250 2.989 Example 6.5: An inventor interested in a design of tube and piston so that the pressure is doubled in the cylinder when the piston is moving suddenly.3628 My 0. Ux = m ˙ mRT ˙ 2 × 287 × 350 = = ∼ 502.25[m/sec] ρA PA 200000 × 0.15 cx 1. Thus.41087 ∼ 1.24[◦C] Ty = T x Tx The velocity of the piston is then √ Uy = My ∗ cy = 0.2308 = 369.0000 0. what will be the piston velocity? S OLUTION This is an open valve case in which the pressure ratio is given.96697 The temperature ratio and the Mach numbers for the velocity of the air (and the piston) can be calculated.25 Ux =√ ∼ 2.2308 1. The propagating piston is assumed to move into media with temperature of 300K and atmospheric pressure of 1[Bar].4 ∗ 287 ∗ 369. NORMAL SHOCK IN VARIABLE DUCT AREAS Finally utilizing the equation to calculate the following My = Msy − My = 2.002 502.6: A flow of gas is brought into a sudden stop.002[m3 ].091 × 143 × 350 Thus the static Mach number. Mx is Mx = .091 (Butane?). The imaginary gas conditions are temperature is 350K and pressure is 2[Bar] and R = 143[j/kg K ] and k = 1.24 ∼ 291.75593 ∗ 1.132 CHAPTER 6. Calculate the conditions behind the shock wave. The mass flow rate of the gas is 2 [kg/sec] and cross section A = 0.09668 − 0. The temperature at “downstream” (close to the piston) is Ty = 300 × 1.

923 2.47996 Ty Tx Py Px My 0.1500 2.6.589 9.0 0.1500 2.804 9.796 9.0 0.590 2.35101 This table was obtained by using the procedure described in this book.9222 My Mx My 0. The iteration of the procedure are i 0 1 2 3 4 5 Mx 3.922 2.796 0.922 My 0.796 .4096 9.589 2.796 9.47996 2.0 0.609 2.2.46689 0.914 9.47996 0.922 2.0 0.47988 0.589 11.0 0.0 2.940 2.0 Ty Tx Py Px P0 y P0 x 133 0.8598 2.589 2.47886 0.47995 0. DIFFUSER EFFICIENCY With this value for the Mach number Potto-GDC provides Mx 2.

NORMAL SHOCK IN VARIABLE DUCT AREAS .134 CHAPTER 6.

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

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

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

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 (8.6) . ISOTHERMAL FLOW 8. (8. Hence.2) where A is the cross section area (it doesn’t have to be a perfect circle.1 The Control Volume Analysis/Governing equations Figure (8. The Awetted area is the area that shear stress acts on. The heat transfer up stream (or down stream) is assumed to be negligible.).138 CHAPTER 8. equation of state is expressed as the following: P = ρRT (8.1) (8.3) Again it is assumed that the gas is a perfect gas and therefore. equation of state reads dP dρ = P ρ It is convenient to define a hydraulic diameter DH = 4 × Cross Section Area wetted perimeter (8. First it must be recalled that the temperature is constant and therefore. The shear stress is the force per area that acts on the fluid by the tube wall. Dimensionless Representation In this section the equations are transformed into the dimensionless form and presented as such.1) describes the flow of gas from the left to the right.5) 8.4) (8. the energy equation can be written as the following: U2 dQ = cp dT + d = cp dT0 m ˙ 2 The momentum equation is written as the following −AdP − τw dAwetted area = mdU ˙ Perhaps more quantitative discussions about how “circular” the shape should be. a close enough shape is sufficient.7) (8.2 it seems obvious to write this equation perhaps to consult with others.

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. DIMENSIONLESS REPRESENTATION 139 Now.8) into momentum equation (8. the Fanning friction factor2 is introduced. thus “Fanning f” values are only 1/4th of “Darcy f” values.11) yields should be noted that Fanning factor based on hydraulic radius.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. 2 It where are the stagnation equations? put them in a table put explanation how to derive this expression.8) Substituting equation (8. U 2 = kRT M 2 Differentiation of equation (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 (8.2) yields 4dx −dP − f DH 1 2 ρU 2 m ˙ A = ρU dU (8.11) d(M 2 ) d(U 2 ) dT = − M2 U2 T (8.8.15) (8.12) (8. U .11) yields d(U 2 ) = kR M 2 dT + T d(M 2 ) (8.9) Rearranging equation (8. instead of diameter friction equation.2. . P as a function of the Mach number has to substitute along with velocity.10) Now the pressure.

20) to become dU dP =− P U 3 Assuming (8. T. state are described above. P. There are 4 unknowns (M.22) the upstream variables are known. energy.18) dT0 (k − 1) M 2 dM 2 = 1 T0 M2 2 1 + k− 2 (8. These different solutions are supersonic and subsonic solution.20) The four equations momentum.140 dP0 dP = + P0 P 1 CHAPTER 8. fL The distance friction. 4D . continuity (mass). The variable. the equations need to be obtained as a function of 4D density is eliminated from equation (8.17) Notice that dT0 = 0 in an isothermal flow. 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). Thus.17) yields: T0 1 k−1 2 2 d M 1 2 + k− 2 M dT0 = Rearranging equation (8. ISOTHERMAL FLOW 1 2 2 kM k−1 + 2 M2 dM 2 M2 (8. ρ)3 and with these four equations the solution is attainable.16) Differentiation of equation (4.21) (8.19) By utilizing the momentum equation it is possible to obtain a relation between the pressure and density.9) yields: dT0 = dT 1+ k−1 2 M 2 +T k−1 dM 2 2 (8. Substituting T for equation (8. is selected as the choice for the independent fL .18) yields M2 M2 (8.15) when combined with equation (8. Recalling that an isothermal flow (T = 0) and combining it with perfect gas model yields dρ dP = P ρ From the continuity equation (see equation (8. One can notice that there are two possible solutions (because of the square power). .14)) leads dM 2 2dU = 2 M U (8.

24) Similarly or by other path the stagnation pressure can be expressed as a function fL of 4D 2 kM 2 1 − k+1 dx dP0 2 M = 4f k − 1 P0 D 2 (kM 2 − 1) 1 + 2 M 2 (8. Thus. By denoting the superscript symbol ∗ for the choking condition.27) It can be noticed that at the entrance (x = 0) for which M = Mx=0 (the initial fL is positive for any x. one can obtain − dP 4f dx − P DH kP M 2 2 = kP M 2 dP P 141 (8.23) Equation (8. The term 4D the other side has to be positive as well.24) can be separated to obtain integrable form as follows L 0 dT0 dx k (1 − k ) M 2 4f = k−1 2 2 T0 D 2 (1 − kM ) 1 + 2 M (8. Integration of equation (8. thus.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 . one can obtain that 1/k M2 = ∗2 U2 U (8.22) into equation (8. Heat transfer has a limited value therefore the model of the flow must be changed. the term on velocity in the tube isn’t zero).25) The variables in equation (8. A more appropriate model is an adiabatic flow model yet it can serve as a bounding boundary (or limit). When the value M = √ k 1 Mach number larger than M > √ it makes the right hand side of the integrate k negative.10).2.27) yields 4f Lmax D = 1 − kM 2 + ln kM 2 kM 2 (8.29) . DIMENSIONLESS REPRESENTATION After substituting the velocity (8. To obtain this restriction 1 = kM 2 . The physical meaning of this value is similar to M = 1 choked flow which was discussed in a variable area flow in Chapter (4). Since the stagnation temperature (T0 ) has a finite value which means that dT0 → ∞.26) that when M → √ the value k 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. 1 Further it can be noticed from equation (8. 1 is the limiting case from a mathematical point of view.26) 4f dx = D 1/k M2 1 − kM 2 dM 2 kM 2 (8.

It is very useful to obtain the relationship between the entrance and the exit condition for this case.2) 8. Denote 1 and 2 as the conditions at the inlet and exit respectably.33) equation (8. ISOTHERMAL FLOW Rearranging equation (8.142 CHAPTER 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∗ (8.24) 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 (8.32) 1+ 1 k−1 2 2 M k −1 + 2k k k−1 (8. From equation (8.3 The Entrance Limitation of Supersonic Branch Situations where the conditions at the tube exit have not arrived at the critical conditions are discussed here.35) These equations (8.30)-(8.29) is transfered into √ U kM = U∗ Utilizing the continuity equation provides ρU = ρ∗ U ∗ .30) ρ 1 =√ ρ∗ kM (8.36) For the case that M1 >> M2 and M1 → 1 equation (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 (8.37) .31) (8.36) is reduced into the following approximation ∼0 4f L D = 2 ln M1 − 1 − 1 − kM2 2 kM2 2 (8.34) And the stagnation temperature at the critical point can be expressed as 1 2 T 1 + k− T0 2k 2 M = ∗ = ∗ k − 1 T0 T 3k − 1 1 + 2k 1+ k−1 2 M2 (8.35) are presented on in Figure (8.

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. 8.4 Comparison with Incompressible Flow The Mach number of the flow in some instances is relatively small.5. even small 4D > 2 results in a Mach number which is larger than 4. This velocity requires a large entrance length to achieve good heat transfer.01 1 Mach number 10 Fig.1 0.38) This relationship shows the maximum limit that Mach number can approach when fL the heat transfer is extraordinarily fast. ρ/ρ and T0/T0 as a function of M 1e+02 4fL  D P or  ρ  * ∗ P ρ T0/T0 P0/P0 1 * * * * * 1e+01 0.1 Fri Feb 18 17:23:43 2005 0. In these cases.2: Description of the pressure.8. COMPARISON WITH INCOMPRESSIBLE FLOW 143 Isothermal Flow P/P . 8. temperature relationships as a function of the Mach number for isothermal flow Solving for M1 results in M1 ∼ e 1 2 „ 4f L D +1 « (8.4. In reality. Yet this model provides the directions of the heat transfer effects on the flow.

39) one can obtained that M2 = M 1 P1 P2 (8. ISOTHERMAL FLOW incompressible flow. a relationship between M2 and M1 and pressures has to be derived. Expanding the solution for small pressure ratio drop. denote χ= P1 − P 2 P1 (8.42) Now 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 (8.46) . From equation (8. To obtain a similar expression for isothermal flow.43) 4f L D = 1 2 1 − (1 − χ) − ln kM1 2 1 1−χ 2 2 (8. by some mathematics.144 CHAPTER 8. the pressure loss is expressed as follows P1 − P 2 = 4f L D U2 2 (8.40) Substituting this expression into (8. P1 − P2 /P1 .40) yields 4f L D = 1 kM1 2 1− P2 P1 2 − ln P2 P1 2 (8.39) fL represent the ratio of Now note that for incompressible flow U1 = U2 = U and 4D the traditional h12 .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. For 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.

48) yields 4f L D = χ (2 − χ) − kM1 2 (2 − χ) + f χ3 kM1 2 (8.5 Supersonic Branch Apparently.5. 8.47) similarly it can be shown that f (χ = 0) = 1 equation (8. Equation (8. Currently and ashamedly the author is looking for a more simplified explanation. and is based on estimates.52) A few observations can be made about equation (8.52).51) value of The pressure difference can be plotted as a function of the M1 for given 4f L D .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 (8.48) rearranging equation (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 (8.49) and further rearrangement yields 4f L D = χ 2(1 − kM1 2 ) − 1 + kM1 2 χ + f χ3 kM1 2 (8.8.50) in cases that χ is small 4f L D ≈ χ 2(1 − kM1 2 ) − 1 + kM1 2 χ kM1 2 (8.45) now can be approximated as 4f L D = 1 (2χ − χ2 ) − 2χ − χ2 + f χ3 kM1 2 (8. this analysis/model is over simplified for the supersonic branch and does not produce reasonable results since it neglects to take into account the heat transfer effects. The current explanation is correct but based on hands waving and definitely does not satisfy the author. 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. .

7 D On the other hand a typical value of friction coefficient f = 0. even fL for relatively small tubes with 4D = 10 the inlet speed is over 56 [km/sec].37 which means that the maximum amount of energy is insufficient. the feeding diverging nozzle somewhat reduces the required entry length (as opposed to converging feeding). even with low temperature like 200K the speed of sound of air is 283[m/sec]. this model can provide a better understanding to the trends and deviations of the Fanno flow model. Nevertheless. if Mach number is changing from 10 to 1 the kinetic energy change is T0 about T ∗ = 18. are 0. 8. ν . The flow speed at the entrance is very large. The thermal entry length is in the order of the hydrodynamic entry length (look at the Prandtl number. Almost all the perfect gas model substances dealt with in this book. value for the common gases. For example. a gas flows in a tube with 4D number is over 200.005 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).6 5 see Figures and Tables Kays and Crawford “Convective Heat Transfer” (equation 12-12). .0000130034 kg/m-sec at 200K. Now.146 CHAPTER 8. For this illustration. the hydraulic entry length is very large as will be shown below. for most gas cases the speed of sound is about 300[m/sec]. This requires that the entrance length to be larger than the actual length of the tub for air. For example.0000185 kg/m-sec at 300K and 0. (0. the speed of sound is a function of temperature. ISOTHERMAL FLOW assumption of the model and the model breaks.06 UD ν (8.0). Combine this information with our case of 4f L D = 10 Lentrance = 250746268. So.).53) The typical values of the the kinetic viscosity. this topic will be covered in the next version because it provide some insight and boundary to the Fanno Flow model. 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 .71. However. Remember from Fluid Dynamic book Lentrance = 0. 0 Now with limitation. over hundred of meters per fL = 10 the required entry Mach second. In the supersonic flow.005 results in Lmax 10 = = 500 D 4 × 0.

89644 0.4147 2. etc. resistance (diameter.00626 0.6903 1.056 1.55000 0.93800 0.98700 0.83637 0.87528 0.33 279.1289 16.70000 0.82758 0.99741 1. and mass flow rate questions. 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).5644 9.89075 0.6109 8.1: The Isothermal Flow basic parameters 147 M 0.).4515 4.2553 10.96075 0.92794 0.98982 0.1269 1.3002 1.8.8172 2.87675 0.3524 1.87516 0.88594 0.00205 0.25000 0. In this model no questions about shock (should) exist7 .60000 0.99485 0.45000 0.5920 6.1599 13. 7 Those who are mathematically inclined can include these kinds of questions but there are no real world applications to isothermal model with shock.021 1.1289 16. These questions or examples will appear in the later versions.81879 0.009 1.0495 1.3906 8.97344 0.8781 1. Here is an example.1827 1.000896 0.7.09000 0.5366 1.06 192.40000 0.65000 0.90300 0.0859 12.75000 0.7040 66.4147 2.08000 0.81000 0.000 28.8791 1.2565 1.2074 1.08085 0.16552 0.2258 3.3334 2.00371 0.007 1.005 1.8650 3.9181 5.03000 0.6500 5.2258 3.000 17.20000 0.05000 0.8172 2.1718 21.87612 0.2074 1.8493 7.021 1.87544 0.87642 0.1269 1.12 139.011 1.9925 4.5644 9.35000 0.50000 0.6651 13. ISOTHERMAL FLOW EXAMPLES Table 8.032 1.10000 0.3906 8.0736 10.0859 12.7230 2.1259 1.0 28.1129 1.0736 10.7 Isothermal Flow Examples There can be several kinds of questions aside from the proof questions6 Generally.07000 0.89 82.4784 1.1718 21.2668 0.03095 0.000 8.9031 14.2126 1.0823 1.79 105.003 1.1129 1.6903 1.87586 0.50207 0.04000 0.3806 2. The driving force questions deal with what should be the pressure difference to obtain certain flow rate.80732 0.29895 0. friction factor. the “engineering” or practical questions can be divided into driving force (pressure difference).99232 0.5366 1.0677 1.3002 1.97 439.3806 2.4086 1.043 1.043 1.6470 1.0255 1.87563 0.88200 0. friction coefficient.9682 1.91875 0.06000 0.000220 0.4515 4.032 1.001 1.011 1.84515 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 785.9747 7.94894 0.91044 0. .30000 0.8781 1.80000 0.056 1.4086 1.000 0.9031 14.

0 m ˙ = kg sec P1 P1 P1 kU kU A =√ = AkM1 A√ RT k c kRT kRT .04331 × √ m .19 RT 290 × 300 m3 π × (0.31 × 290 × 300 ∼ = 14.005 and the exit pressure is 1[bar]. 450 ∼ kg = = 23.25)2 kg × 14. the flow is not choked.5921 0.25 Utilizing Table (8.00 20. the density is a √ the entrance Mach number. The specific J heat for the gas. fL Calculating the resistance. ρ.25 [m] diameter and 5000 [m] in length is attached to a pump.0 0. surroundings temperature 27◦ C.31. k = 1. 017. In incompressible flow.1743 12.9 4 sec The maximum flow rate then reads m ˙ = ρAU = 23.04331 400.1) or the program provides M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 0. What should be the pump pressure so that a flow rate of 2 [kg/sec] will be achieved? Assume that friction factor f = 0.89446 The maximum flow rate (the limiting case) can be calculated by utilizing the above table.e. Hint: calculate the maximum flow rate and then check if this request is reasonable. The density reads 1. check whether flow is choked (or even possible). S OLUTION If the flow was incompressible then for known density. The velocity of the gas at the entrance U = cM = 0.62 sec ρ= P 2.1: A tube of 0. It is note worthy to mention that since the isothermal model breaks around the choking point. the velocity can be calcufL U2 function of lated by utilizing ∆P = 4D 2g . R = 290 Kkg . 4D 4f L D = 4 × 0. It is more appropriate to assume an isothermal model hence our model is appropriate. The exit Mach number is not necessarily 1/ k i.148 CHAPTER 8.0055000 = 400 0.62 ∼ = 16. the flow rate is really some what different. First. To solve this problem the flow rate has to be calculated as m ˙ = ρAU = 2. ISOTHERMAL FLOW Example 8.19 × The maximum flow rate is larger then the requested mass rate hence the flow is not choked.

6779 8.89442 Note that tables in this example are for k = 1.5 [km] (500 [m]). the maximum M1 that can be obtained when the M2 is at its maximum and back pressure is at the atmospheric pressure.103 .4. 31 100000 × π×(0 4 P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 66.566[bar] Example 8.8.0 0. The friction coefficient for the tube can be assumed as 0.89567 The entrance Mach number is obtained by 4f L D 1 = 66.04014 466. A flow rate of 0. . calculate the required diameter of tube.68 The pressure should be 21. ISOTHERMAL FLOW EXAMPLES Now combining with equation (8. with the value of M1 either by utilizing Table (8.1) or using the provided program yields 8 It is unfortunate.59 mc ˙ = = 0.10300 4fL D P P∗ 2 × 337.5 [in]8 . Thus.4826 = 2.). Due to safety concerns.76780 × 8.1) or by utilizing the program M 0. but it seems that this standard will be around in USA for some time.5844 0. S OLUTION At first. Mmax 1 1 P2 = √ = 0.6779 + 400 ∼ = 466.0845 M1 = M 2 P1 k 10 Now.31 P = 21.4826 5.40) yields m ˙ = M2 = M2 P2 Ak c 149 From Table (8. Note that tubes are provided in increments of 0. You can assume that the soundings temperature to be 27◦ C.2 [kg/sec] is required.0 0.2: A flow of gas was considered for a distance of 0.68 P P∗ P0 P0 ∗ ρ ρ∗ T0 T0 ∗ M 4fL D 0.7.3249 0. the minimum diameter will be obtained when the flow is choked.7678 13.02 (A relative smooth tube of cast iron.25)2 P2 Ak × 1 . Assume that the flow is isothermal and k=1. the maximum pressure allowed for the gas is only 10[bar].

0853 × √ 287 × 300 √ 1.4318[m].0018 6.150 M 0.2991 0.68[in] However.08528 × 347. 000 kg P = = 11.0 0. ISOTHERMAL FLOW P0 P0 ∗ ρ ρ∗ T0 T0 ∗ 94.3[kg/sec] Since 50.87627 To check whether the flow rate satisfies the requirement m ˙ = 106 × π ×0.6 sec .4318 P P∗ P0 P0 ∗ 92.43182 4 × 0.61 RT 287 × 300 m3 The velocity at the entrance should be m U = M ∗ c = 0.9110 6. The new 4D 4f L D = 4 × 0. 000.5 increments and the next size is 17[in] or 0.08527 4fL D 92.3 ≥ 0.2 the mass flow rate requirement is satisfied.2 ∼ = 29.6400 9. The only point that must be emphasized is that all properties (like Mach number.4310 10.43 0.87625 = 94.4 ≈ 50.08450 With 4f Lmax D 4fL D P P∗ CHAPTER 8.02 × 500 0.4 × 287 × 300 ∼ = 347. D= 4f L 4f Lmax D 4 × 0. It should be noted that P should be replaced by P0 in the calculations.431 the value of minimum diameter. the pipes are provided only in 0.2 sec and the density is ρ= 1.2424 0. pressure and fL is etc) have to be taken at the same point. The speed of sound at the entrance is √ √ m c = kRT = 1.0 0.42359[m] = 16. 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.64 ρ ρ∗ T0 T0 ∗ M 0.02 × 500 94.

21[bar] 20.7.027 π × 29. Therefore. P2 = P0 ∗T P2 P0 ∗T = 2. The gas and the sounding temperature can be assumed to be 300 K.01.52828 = 1.19235 12. Example 8.66915 From the table M1 ≈ 0. the information whether the flow is choked needs to be found.4. The solution is an iterative process. k=1. Calculate the Mach number at the entrance to pipe and the flow rate.01 × 4000 = 400 0. for the sake of the exercise the other parameters will be calculated. The flow rate is given with the diameter of the pipe. ISOTHERMAL FLOW EXAMPLES The diameter should be D= 4m ˙ = πU ρ 4 × 0.and P0 ∗T ∼ = 28 12. at first it will be assumed that the whole length is the maximum length.87531 P0 P0 ∗T 20.67 2.21 × 0.67 The pressure at point (b) by utilizing the isentropic relationship (M = 1) pressure ratio is 0.17[bar] As the pressure at point (b) is smaller than the actual pressure P ∗ < P2 than the actual pressure one must conclude that the flow is not choked. It should be noted that the flow isn’t choked. and the average friction f=0. S OLUTION First.72021 0. 1. This situation is reversed question. guess reasonable value of M1 and calculate 4f L D .4 with 4f Lmax D = 400 the following can be written 4f L D T0 T0 ∗T ρ ρ∗T P P∗T P0 P0 ∗T M 0.8.19235 ≈ 12. 4f Lmax D = 4 × 0.0419 400.2 ∼ = 0.61 151 Nevertheless.4[m] diameter and 4000 [m] length to a different station (b). Assume that the flow is isothermal.3: A gas flows of from a station (a) with pressure of 20[bar] through a pipe with 0.52828. The pressure at the exit (station (b)) is 2[bar].6 × 11.0419 .

32807 0.32795 0.36780 0.8 Unchoked situation Table 8.08971 0.83889 0.10000 0. Calculate the pressure.05000 0.5708 M2 0.84095 0.83740 4f Lmax D 1 4f L D P2 P1 0.32131 400.152 2.0419 4 300 × 287 RT 42.10000 0. there are no examples on isothermal with supersonic flow.5914 0.00000 0.10000 0.08978 0. ISOTHERMAL FLOW by subtracting 4f L D 1 − 4f L D 3.7272 0. 4. P2 bear in mind that this isn’t the real pressure but based on assumption 5. Obtain M2 from the Table ? or by using the Potto–GDC.16922 0.6483 0.6934 0.12949 0. M1 0.4 P k π × D2 M= √ m ˙ = ρAM c = √ π × 0.05005 0.46[kg/sec] In this chapter.10000 0.5807 0.84018 0.83997 0. Compare the results of guessed pressure P2 with the actual pressure and choose new Ma number M1 accordingly.10000 .0419 The flow rate is √ √ 2000000 1. Calculate the value of 4f L D 2 CHAPTER 8. Now the process has been done for you and is provided in Figure (??) or in the table obtained from the provided program.40754 0.10000 0.16912 0.10000 8.40737 0.6684 0.59338 4f Lmax D 1 4f L D P2 P1 400.4: The flow parameters for unchoked flow M1 0.12942 0.36766 0.10000 0.83827 0. M2 0.83920 0.22 × 0.

4 0.10 Fri Feb 25 17:20:14 2005 Fig.8. 8. UNCHOKED SITUATION 153 M1 isothermal flow 1 0.8.9 0.7 0.3: The Mach number at the entrance to a tube under isothermal flow model as a fL function 4D .2 = 0.2 0.8 = 0.3 0.5 = 0.1 0 0 10 20 30 40 50 4fL  D 60 70 80 90 100 M1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.5 0.8 0.6 0.

ISOTHERMAL FLOW .154 CHAPTER 8.

etc.± 9. This model is the second ¥¦¨§Ÿ¥ pipe flow model described here. As this model raised from need to explain the steam flow in turbines.CHAPTER 9 Fanno Flow An adiabatic flow with friction is ¤ ›8œžŸ› named after Ginno Fanno a Jewish “ flow  ¢¡ž£u  direction engineer.1: Control volume of the gas flow in a constant cross section number. This model explains many industrial flow processes which includes emptying of pressured container through a relatively short tube. Later. 155 . exhaust system of an internal combustion engine. compressed air systems.1)). ”•!–˜—D™„š © ¦¨§ ©«ª~¬­ ¦¨§ ¯ ­ ® The main restriction for this model is that heat transfer is negligible and ². It is advantages to examine the simplest situation and yet without losing the core properties of the process. plicable to flow processes which are No heat transer very fast compared to heat transfer mechanisms with small Eckert Fig. °. 9. This model is apc.v. discussed on the ideal gas model and the entry length issues. more general cases will be examined2 .1 Introduction Consider a gas flowing through a conduit with a friction (see Figure (9.³ can be ignored 1 . 1 Even 2 Not the friction does not convert into heat ready yet.

2 Model m ˙ = ρAU = constant → ρ 1 U1 = ρ 2 U2 (9.6) 3 The equation of state is written again here so that all the relevant equations can be found when this chapter is printed separately.2) (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. .156 CHAPTER 9.8) 4 × Cross Section Area wetted perimeter (9.3) Again for simplicity.4) U1 = 2cp 2 T0 2 T2 + U2 2cp 2 (9. P = ρRT P2 P1 = → ρ 1 T1 ρ 2 T2 (9.5) It is assumed that the flow can be approximated as one–dimensional.1) The mass (continuity equation) balance can be written as The energy conservation (under the assumption that this model is adiabatic flow and the friction is not transformed into thermal energy) reads T0 1 = → T1 + Or in a derivative form Cp dT + d U2 2 = 0 (9. the perfect gas model is assumed3 . FANNO FLOW 9.

10) into momentum equation (9. The infinitesimal area that shear stress is acting on is dAw = πDdx (9.9. By utilizing the definition of the sound speed to produce the following identities for perfect gas M2 = U c 2 = U2 k RT P ρ (9. 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.3.14) Utilizing the definition of the perfect gas results in M2 = ρU 2 kP (9.12) The second law is the last equation to be utilized to determine the flow direction.13) 9.11) by the cross section area.11) yields Dividing equation (9. s2 ≥ s 1 (9.2) and substituting equation (9.10) By utilizing equation (9.6) yields A τw 2 m ˙ A − πD dP − πDdx f 4 1 2 ρU 2 = A ρU dU (9.9) Introducing the Fanning friction factor as a dimensionless friction factor which is some times referred to as the friction coefficient and reads as the following: f= τw 1 2 ρU 2 (9.3 Non–dimensionalization of the equations Before solving the above equation a dimensionless process is applied.15) . A and rearranging 4f dx D 1 2 ρU 2 −dP + = ρU dU (9.

14) yields d(M 2 ) d(U 2 ) dT = − 2 M U2 T (9.17).5) results dρ dT dP = + P ρ dT (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 (9.158 CHAPTER 9.20) Dividing the energy equation (9.21) Equations (9. These equa- .18).16) By further rearranging equation (9.14) and substituting it into equation (9.20).17) It is convenient to relate expressions of (dP/P ) and dU/U in terms of the Mach number and substituting it into equation (9.17). (9. (9. and (9.18) The derivation of the equation of state (9. Derivative of mass conservation ((9.19) Derivation of the Mach identity equation (9.16) results in − dP 4f dx − P D kM 2 2 = kM 2 dU U (9. (9.19).14) and dividing by equation (9. Thus.5) and dividing the results by equation of state (9.2)) results in dU U dρ 1 dU 2 + =0 ρ 2 U2 (9. FANNO FLOW Using the identity in equation (9. the original limitations must be applied to the resulting equation. (9.21) need to be solved.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.

24) and results in 1 − M 2 dM 2 4f dx = 1 2 D kM 4 (1 + k− 2 M ) Rearranging equation (9.21) by eliminating dT /T .21) when eliminating dT /T results dP dρ (k − 1)M 2 dU 2 = − P ρ 2 U2 (9.9. 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).24) (9.27) and (9. Explicit explanation is provided for only two variables. sistance. M variable is obtained by combining equation (9. 4D Combining equations (9.29) 1 dc k (k − 1)M 4 4f dx dT = =− T 2 c 2(1 − M 2 ) D (9.18) and equation (9.19) and (9.25) results in 1 2 kM 2 1 + k− 4f dx dM 2 2 M = 2 2 M 1−M D (9.22). Then dρ/ρ and U are eliminated by utilizing equation (9. The dimensionless friction.3.28) . is chosen as the independent variable since the change in the dimensionless refL .23) kM 2 1 + (k − 1)M 2 4f dx dP =− P 2(1 − M 2 ) D (9.22) and rearrangement yields 1 + (k − 1)M 2 dU 2 dP =− P 2 U2 The term dU 2 /U 2 can be eliminated by using (9. fL the rest variables can be done in a similar fashion.27) and the relationship for the temperature is density is obtained by utilizing equations (9. The only variable that is left is P (or dP/P ) which can be eliminated by utilizing equation (9.18) to obtain dρ 4f dx kM 2 =− 2 ρ 2 (1 − M ) D (9.18) and substituting it into equation (9.26) After similar mathematical manipulation one can get the relationship for the velocity to read dU kM 2 4f dx = U 2 (1 − M 2 ) D (9. 4D .23) The second equation for Mach number.25) (9. causes the change in the other variables.22) The term dρ ρ can be eliminated by utilizing equation (9.20) and (9.

flow with speed below the speed of sound.34) 9. When M < 1 the pressure decreases downstream as can be seen from equation (9.33) The first law requires that the stagnation temperature remains constant.30) dP dT − R ln T P (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. FANNO FLOW (9. This pressure increase is what makes compressible flow so different from “conventional” flow.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. For example. in the supersonic branch. . the pressure increases downstream. 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. M > 1. from equation (9. flow above speed of sound. (dT0 = 0).35) (9.24) because f dx and M are positive. Thus the discussion will be divided into two cases: One. Two.24) through (9.31) The stagnation temperature expresses as T0 = T (1 + (1 − k )/2M 2 ).24) it can be observed that the critical point is when M = 1.4 The Mechanics and Why the Flow is Choked? The trends of the properties can be examined by looking in equations (9. For the same reasons.34).

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

39) can be integrated to yield: 1 P = ∗ P M k+1 2 k −1 2 2 M 1+ (9.27).39) The symbol “*” denotes the state when the flow is choked and Mach number is equal to 1. (9.24) can eliminate term D and describe the pressure on the Mach number.23) is represented by 4D 4f L and Mach number. (9.41) ρ 1 = ρ∗ M 1+ k−1 2 2 M k+1 2 (9. (9.42) U = U∗ ρ ρ∗ −1 =M 1+ k+1 2 k−1 2 2 M (9.24). Dividing equation (9.24) in equation (9.44) .28). FANNO FLOW By utilizing the mean average theorem equation (9. Now equation (9.36) yields k+1 2 ¯ max 1 1 − M2 k+1 4fL 2 M = + ln 1 2 D k M2 2k 1 + k− 2 M (9.43) The stagnation pressure decreases and can be expressed by k 2 k−1 (1+ 1− 2 M ) k P0 = P0 ∗ P0 P P0 ∗ P∗ 2 ( k+1 ) k−1 k P P∗ (9. M = 1 when P = P ∗ Equation (9. It is common to replace the f Equations (9.40) In the same fashion the variables ratio can be obtained k+1 c2 T 2 = = 1 2 T∗ c∗ 2 1 + k− 2 M (9.26) yields dP P dM 2 M2 =− 1 + (k − 1M 2 dM 2 1 2 M 2M 2 1 + k− 2 (9.29).30) can be solved.162 CHAPTER 9. and (9. For example. (9. fL . the pressure as written in equation (9. Thus.38) ¯ with f which is adopted in this book.29).

2: Various parameters in Fanno flow as a function of Mach number many cases shockless and therefore a relationship between two points should be .40) and substituting it into equation (9.9. THE WORKING EQUATIONS 163 Using the pressure ratio in equation (9.44) yields P0 = P0 ∗ 1+ k−1 2 2 M k+1 2 k k−1 1 M 1+ k−1 2 2 M k+1 2 (9. 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 0.1 Fri Sep 24 13:42:37 2004 0.5.34) yields s − s∗ = ln M 2 cp k+1 1 2 1 + k− 2 M k+1 k 2M 2 (9.01 1 Mach number 10 Fig. D The results of these equations are plotted in Figure (9.46) The integration of equation (9.45) And further rearranging equation (9.1 0.47) discussion about Reynolds number and dimensionless 4f L friction parameter.45) provides 1 P0 ∗ = P0 M 1+ k−1 2 2 M k+1 2 k+1 2(k−1) (9.2) The Fanno flow is in P/P . 9.

6 Examples of Fanno Flow Example 9.49) Hence. FANNO FLOW derived. . the “star” values are imaginary values that represent the value at choking. finding the total pressure and temperature at the pipe inlet is the solution. ¾ 5 ¿ À form pipe with a diameter of 0.164 CHAPTER 9. The air exits to the atÁÃÂ. For given Mach number (M = 0. There is no Mach meter. The following conditions preԐÕAÖ·×sØsÙ Ú vail at the exit: P2 = 1[bar] temperature Ê È ÉAËÍÌDÎ ÏÑÐDÒsÓ T2 = 27◦ C M2 = 0. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.1: Û£ÜAÝ·Þß à Air flows from a reservoir and enters a uni´€µ·¶¸ ¶s¹©º »§¼ 0 ½ . Thus.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. the temperature and the total pressure at the pipe inlet are the same as those in the reservoir. the flow to the pipe inlet.3: Schematic of Example (9.0089 1.1291 4 This 1. Estimate the total temperature and total pressure in the reservoir under the Fanno flow model.90000 0. With the Mach number and temperature known at the exit. 4f Lmax D = 2 4f Lmax D 1 − 4f L D (9.004 and that the flow from the reservoir up to the Fig.1) pipe inlet is essentially isentropic.01451 1.50) 9. Assume that the average friction factor to be f = 0. the total temperature at the entrance can be fL obtained by knowing the 4D .05 [m] and áãâåäÑæ©ç è§é length of 10 [m]. In most times.94 .Ä_ÅMÆ Ç mosphere. S OLUTION For isentropic.9146 1. The real ratio can be obtained by two star ratios as an example T2 = T1 T T ∗ M2 T T ∗ M1 (9.0934 0.0327 property is given only for academic purposes.9) the following is obtained. 9.

1699 Note that the subsonic branch is chosen.17 × 1.01451 0.0327 0.975 T∗ T 348K = 75◦ C Another academic question: .915 T1 T ∗ T01 = T2 T T0 ∗ 2 T 1 T 1 1 1 =300 × × 1. EXAMPLES OF FANNO FLOW So.014 × = 2.35886 0.35886 3. M 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.38814 1.05 4f L D The rest of the parameters can be obtained with the new (9.93840 1.9.21 either from Table 1 = 4f L D + 4f L D 2 = 4 × 0.91[Bar] 1.97489 0.5[K ] 1.7405 2.91484 1. The stagnation ratios has to be added for M = 0.5922 0. the total temperature at the exit is T ∗ |2 = T∗ T T2 = 2 165 300 = 290.12913 0.2100 3.004 × 10 + 0.1) by interpolations or by utilizing the attached program.35886 M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.6.5764 0.0327 4f L D To ”move” to the other side of the tube the 4f L D is added as 3.7405 0.0140 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.

166 Example 9. (b) the temperature at the pipe exit. FANNO FLOW ó©ô2õö ÷Mø ùHúMû ü ýMþ ÿ  ¢¡¤£¦¥¨§©§¨ ¦. A normal shock wave occurs in the tube and the flow discharges to the atmosphere. and CHAPTER 9.5 [cm] in diameter and 1.65 [Bar].2: A system is composed of a convergentdivergent nozzle followed by a tube with length of 2. The system is supplied by a vessel. The vessel conditions are at 29.0. 400 K. determine: (a) the mass flow rate through the system.0 [m] long. With these conditions a pipe inlet Mach number is 3.

005. With the isentro relationship the Mach number at point one (1) is known. m ˙ = ρAM c The density and speed of sound are unknowns and need to be computed.07623 4.11528 0. S OLUTION (a) Assuming that the pressure vessel is very much larger than the pipe.357 × 400 = 142.02722 0.65[P ar] The mass flow rate through the system is constant and for simplicity point 1 is chosen which. It is further assumed t the flow through the nozzle can be approximated as isentropic. therefore the velocity the vessel can be assumed to be small enough so it can be neglected. Hence. T01 = 400K a P01 = 29. 9. Thus.35714 0.4.65326 T1 = T1 T01 = 0. R = 287 [J/kgK ] and f = 0. Take k = 1. the stagnat conditions can be approximated for the condition in the tank.2346 0.0000 The temperature is 0.4: The schematic of Example (9. then the following can be found eith from Table (9.2) (c) determine the Mach number when a normal shock wave occurs [Mx ].1) or the Potto–GDC T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ M 3. ¦  "!"#  shock d-c nozzle êãë§ìMí ìHîOïMð ñ_ò    atmosphere conditions Fig.8K T01 .

3333 0.42857 and the conditions of the tube are 4f L D = 4 × 0.1) or by using the Potto–GDC.69 4 sec 8.97 × kg π × 0. to obtain the following M 3. the shock at the entrance of the tube.1)) . if a shock exists.8 1.81[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. First.005 × 1.8 0.47519.52216 0.2346 0.8571 10.9640 0. and two.0000 My 0.50918 1.47519 Ty Tx ρy ρx Py Px P0 y P0 x 2.0 = 0.2) m ˙ = 1.9.0252 × 3 × 239. by comparing “possible” Pexit to PB .54[m/sec] The pressure at point 1 can be calculated as P1 = P1 P01 = 0.97 kg m3 (b) First. From the Table (9.8 > 0.6790 3.027 × 30 P01 0. The exit pressure determines the location of the shock.6. check whether the flow is shockless by comparing the flow resistance and the maximum possible resistance. (Fanno flow Table (9.54 = 0. the possibility that the shock wave occurs immediately at the entrance for which the ratio for Mx are (shock wave Table (5.0000 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0. Two possibilities are needed to be checked.025 Since 0. one.1 × 104 287 × 142.32834 After shock wave the flow is subsonic with “M1 ”= 0.8 239.52216 the flow is choked and with a shock wave. shock at the exit and comparing the pressure ratios. the speed of sound can be calculated as √ √ c1 = kRT = 1. EXAMPLES OF FANNO FLOW Using the temperature.21822 4.4 × 287 × 142.1)) Mx 3.

One possible way to find the e temperature. 2) Calculate the entrance Mach assuming shock at the entrance.89545 1.50917 1.168 M 4fL D P P∗ P0 P0 ∗ CHAPTER 9. With the location of shock.12476 The actual pressure ratio 1/29.2549 1. calculate the parameters for shock locat fL with known 4D in the “y” side.85676 1.0338 is smaller than the case in which shock occurs the entrance.65 = 0. T2 is by finding the location of the shock. To check Secant Method. the shock is somewhere downstream. FANNO FLOW ρ ρ∗ U U∗ T T∗ 0.1481 The stagnation values for M = 0.47519 0. Thus. Then either by utilizing shock table or the program.1912 0.2919 2. a) set M2 = 1 fL and calculated M1 ’ for subsonic branch b) add 4D c) calculated Mx for M1 ’ Note this Mach number is the low Value. but be-careful second order methods like Newton-Rapson method can be unstable.47519 are M T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 F F∗ 0.65326 The ratio of exit pressure to the chamber total pressure is 1 1 ∗ P2 = P0 = = P2 P∗ P0 y P P0 x P1 P1 P0 y P0 x P0 1 1× × 0. “claiming” upstream from exit through shock to the entrance. To find the location of the shock ra 2 of the pressure ratio. For example.9640 0. 5 You can use any method you which.3904 0.8568 × 0. 3) According your root finding algorithm5 calculate or guess the shock location and th compute as above the new M1 . P P1 is needed. to obt the upstream Mach number.3904 1. .2549 0. The procedure for the calculations: 1) Calculate the entrance Mach number assuming the shock occurs at the exit: a) set M2 = 1 assume the flow in the entire tube is supersonic: b) calculated M1 Note this Mach number is the high Value.32834 × 1 2.95679 0.47519 1.

The author seeks for a nice explanation of this concept for non–fluid mechanics engineers.57910 4f L D 0. It has to be noted and recognized that as opposed to subsonic branch the supersonic branch has a limited length.).gov/digidoc/report/tm/44/NACA-TM-844.0000 M2 1.9899 My 0. the comparison with reality shows that heat transfer cause changes to the flow and they need to be expected. 7 See on the web http://naca.22019 0. M1 3. experiments from many starting with 1938 work by Frossel7 has shown that the error is not significant. Insert also example on percentage of heat transfer.0000 4fL D up 4fL D down Mx 1. . furthermore. In the Fanno model there is no heat transfer. because the very limited amount of heat transformed it is closer to an adiabatic flow. Many of the pictures in the literature carry copyright statements.57981 (c) The way of the numerical procedure for solving this problem is by finding produce M1 = 3.larc. Nevertheless.7. on the comparison of the maximum length of isothermal model and Fanno Model. up that will 9. repeat previous stage until the solution is obtained.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. SUPERSONIC BRANCH 169 a) set M2 = 1 fL b) for the new 4D and compute the new My ’ for the subsonic branch c) calculated Mx ’ for the My ’ fL and calculated the M1 d) Add the leftover of 4D 4) guess new location for the shock according to your finding root procedure and according to the result. please provide to the undersign so they can be added to this book.8 Maximum length for the supersonic flow To insert example on the change in the flow rate between isothermal flow to Fanno Flow. 9 If you have experiments demonstrating this point. However. 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. The information from the wall to the tube center6 is slower in reality.nasa. These changes include the choking point at lower Mach number.).PDF 8 Many in the industry have difficulties in understanding this concept. The only limitation of the model is its uniform velocity (assuming parabolic flow for laminar and different profile for turbulent flow. These results were obtained from the mathematical derivations but were verified by numerous experiments9 . and. 9. This solicitation is about how to explain this issue to non-engineers or engineer without a proper background. It also must be recognized that there is a maximum length for which only supersonic flow can exist8 . In the process Mx and My must be calculated (see the chapter on the program with its algorithms.9.

1 0 1. From the above analysis.5 1.  D .55 1. k = 1. The maximum length in supersonic flow In Fanno Flow 1. k as shown in Figure (9.2 1.25 1.4 1.5 1.5).2 0.8 0.6 1.3 1.5 0.4 0.45 1.4) = 0.170 as follows: CHAPTER 9.3 0. it can be observed that no matter how high the entrance Mach number will be the tube length is limited and depends only on specific heat ratio.5: The maximum length as a function of specific heat. k Thu Mar 3 16:24:00 2005 Fig.8215 The maximum length of the supersonic flow is limited by the above number. 9.1 1 0.2 1. FANNO FLOW k+1 2 1 − M2 k+1 4f Lmax 2 M = = + ln k − 1 2 D kM 2k 2 1 + 2 M2 4f L D (M → ∞) ∼ −∞ k + 1 (k + 1)∞ + ln k×∞ 2k (k − 1)∞ (k + 1) −1 k + 1 + ln = k 2k 2(k − 1) = 4f L D (M → ∞. Even the flow in the tube starts as a supersonic in parts of the tube can be transformed into the subsonic 4fLmax maximum length.65 spesific heat.9 0.7 0.4 1.3 1.9 Working Conditions It has to be recognized that there are two regimes that can occur in Fanno flow model one of subsonic flow and the other supersonic flow.35 1.6 0. k 9.

the treatment of the two branches are separated.6: The effects of increase of 4f L D on the Fanno line In the analysis of this effect. the mass flow rate decreases. A shock wave can occur and some portions of the tube will be in a subsonic flow pattern. First. Three parameters. In this process. However. increasing the tube length results in increasing the exit Mach number (normally denoted herein as M2 ).43 5&687 9 ' ( Fig.9. Only a combination of these two parameters is truly independent. it should be assumed that back pressure is constant and/or low as possible as needed to maintain a choked flow. the entrance Mach number. The entrance temperature increases as can be seen from Figure (9. It is worth noting that entrance Mach number is reduced (as some might explain it to reduce the flow rate). all the three parameters can be varied and they are discussed separately here. The .>=? ? I I X K<L N ?@BA G J MN U¨V V V W NOBP&Q8R4S T $&% )+*-. P2 /P1 are controlling the flow. WORKING CONDITIONS 171 branch./. Once the Mach number reaches maximum (M = 1). The discussion has to differentiate between two ways of feeding the tube: converging nozzle or a converging-diverging nozzle. and the pressure ratio.9. the dimensionless fL friction. 9. 4D .102.9. Subsonic branch For converging nozzle feeding. no further increase of the exit Mach number can be achieved. M1 . 9.1 fL Variations of The Tube Length ( 4D ) Effects C8DFE H¨I :<.7).

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

the entrance Mach number. The entrance Mach number.9. fL M1 . The entrance Mach continues to be constant and exit Mach number is constant. 9. The exit Mach number. M2 decreases (remember this flow is fL = 0) as M2 = M1 . WORKING CONDITIONS 0 4f L D 4f L D 4f L D choking shockless chokeless 173 4f L D 4f L D 4f L D 4f L D < < < < < < < < 4f L D 4f L D 4f L D choking shockless chokeless 0→a a→b b→c c→∞ ∞ wyx a | z+{ all supersonic flow ^`_ba b mixed supersonic with subsonic flow with a shock between c d c j lonqpsrutqv km the nozzle is still choked }y~ efhg i Fig. However. the mass flow rate is constant because the flow is choked at the nozzle. In the range b − c.8: The Mach numbers at entrance and exit of tube and mass flow rate for Fanno Flow fL as a function of the 4D The 0-a range. In the range of a − b the flow is all supersonic. is only a theoretical length in which the supersonic flow can occur if nozzle is provided with a larger Mach number (a change to the nozzle area ratio which also reduces the mass flow rate). It is worth noticing that in the a − −c the mass flow rate nozzle entrance velocity max . M2 is still one. 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. 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. Notice that exit Mach number. it is a more practical point.9. reduces with the increase of 4D . fL The total maximum available for supersonic flow b − −b . M1 is constant because it is a function of the nozzle design only. M2 = 1. The mass flow rate continues to be constant. 4D . At the end of the on the supersonic branch) and starts ( 4D range a.

This situation is because of the “double” choked condition in the nozzle.1 D = 1.9 1 Exit Mach number Tue Oct 19 09:56:15 2004 Fig.3 0. M1 as a function of M2 4fL  = 0. the exit Mach number remains constant and equal to one. In this range.2 0. Somewhat different then the subsonic branch the mass flow rate is constant even if the flow in the tube is completely subsonic.8) exhibits the development of M1 .3 0. this situation is rather strange to explain.7 Entrace Mach number 0. the flow rate decreases since (m ˙ ∝ M1 )11 .9: M1 as a function M2 for various 4f L D 10 On a personal note. 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.5 0.1 0.8 0. fL M2 mass flow rate as a function of 4D . 1 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.0 Fanno Flow 1 0.5 0.9 0. the resistance increases and on the other hand.7 0. The exit Mach M2 is a continuous monotonic function that decreases with 4D The entrance Mach M1 is a non continuous function with a jump at the point when shock occurs at the entrance “moves” into the nozzle.6 0. On one hand. Figures (9.2 0.1 0 0 0.4 0.6 0.0 = 100.174 CHAPTER 9. 9.8 0.4 0. To summarize the above discussion. fL .0 = 10.

5 1 0. and as can be observed. below the maximum critical length. fL For a given 4D . 9.1 shock = 0.5 2 1.10) exhibits the entrance Mach number as a function of the M2 . The Figure was calculated by max utilizing the data from Figure (9.4 1.5 3 2.2 0. Obviously. Obviously there can be two extreme possibilities for the subsonic exit branch.4 0.1  D = 0. shock at the entrance. Fanno Flow 5 4.2 1. The Figure (9.4 two extremes are shown.5 4 3. The fL larger 4D larger M1 must occurs even for shock at the entrance. Below.10: M1 as a function M2 for different 4f L D for supersonic entrance velocity.8 2 M1 M1 as a function of M2 for the subsonic brench 4fL = 0. shock at the exit.6 0.8 1 M2 1. One.10) only for 4f L 4f L 4f L D = 0.9. the maximum . tow. shockless flow. when the shock wave occurs at the tube exit and two.2) by obtaining the 4f L for M2 and subtracting D 4f L the given D and finding the corresponding M1 .5 0 0 0.6 1. Subsonic velocity occurs for supersonic entrance velocity. WORKING CONDITIONS 175 Figure (9.4 = 0. In Figure (9.2 shown with only fL creates shock at the exit only. and three. the larger 4D larger differences between exit Mach number for the different shock locations.2 = 0. at the tube entrance.9.9) exhibits the M1 as a function of M2 . one.4 Tue Jan 4 11:26:19 2005 Fig. the supersonic entrance flow has three different regimes which depends on the back pressure. For D = 0.1 and D = 0.

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

The back pressure below point c has some adjustment as it occurs with exceptions of point d. In this range and further reduction of the pressure the mass flow rate is constant no matter how low the back pressure is reduced. Up to this stage the nozzle feeding the tube increases the mass flow rate (with decreasing back pressure). Once the back pressure is less than point b the supersonic reaches to the tube. 9. M2 < 1 and is not 1.9.11: The pressure distribution as a function of 4f L D for a short 4f L D Short 4f L D Figure (9. .12) shows different pressure profiles for different back pressures.9. WORKING CONDITIONS ƒ‚ &€ „†… 177 a shock in the nozzle fully subsoinic flow Œ† ŒƒŽ critical Point a criticalPoint b critical Point c ‡qˆŠ‰ ‹ critical Point d Fig. 13 It is common misconception that the back pressure has to be at point d. Before the flow reaches critical point a (in the Figure) the flow is subsonic. A back pressure that is at the critical point c results in a shock wave that is at the exit. Note however that exit Mach number. the tube is “clean” of any shock13 . Between point a and point b the shock is in the nozzle. When the back pressure is below point c.

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

4 Mach Number 1. in this situation.8 1. fL As was shown before.6 0.25 Tue Jan 4 12:11:20 2005 Fig. 9.8 0.13: The effects of pressure variations on Mach number profile as a function of fL = 0.9. It is P2 assumed that in this discussion that the pressure ratio P is large enough to create 1 4f L a choked flow and D is small enough to allow it to happen.2 1 0. first is the total 4f L maximum D of the supersonic which depends only on the specific heat.9. Practically.05 0. WORKING CONDITIONS 179 Mach number in Fanno Flow 4fL  D 2 1.2 0.3 for Fanno Flow when the total resistance 4D 4f L D The entrance Mach number.15 4fL  D 0. This effect is the third parameter discussed here. 4f L D is equal to 4f Lmax D as a The process of decreasing the converging–diverging nozzle’s throat increases the . there are two different maximums for 4D .4 0. the nozzle area ratio is changed by changing the throat area. supersonic branch reached otherwise the flow is subsonic with regular effects. M1 is a function of the ratio of the nozzle’s throat area to the nozzle exit area and its efficiency.2 0 0 0.1 shock at 75% 50% 5% 0.6 1. and second the maximum depends on the entrance Mach number. M1 . D Obviously. k . This analysis fL max deals with the case where 4D is shorter than total 4f L . the critical point is where result in the entrance Mach number.

By doing so.8 0. decrease of the throat area results in flushing the shock out of the tube.15 4fL  D 0.6 3. The reference to the tube is because it is the focus of the study.6 1.14: Fanno Flow Mach number as a function of 4f L D when the total 4f L D = 0. The process of decreasing the throat area also results in increasing the pressure drop of the nozzle (larger resistance in the nozzle15 )16 . 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. area) decreases the flow rate while in a different way (increasing the 4D .25 5% 50 % 75 % Fri Nov 12 04:07:34 2004 Fig. 9.180 CHAPTER 9. The mass flow rate is proportionally linear to the throat area and therefore the mass flow rate reduces.2 0. Decreasing the nozzle throat area moves the shock location downstream. The “payment” for increase in the supersonic length is by reducing the mass flow.4 4 3. the throat area decreases. fL max In the case of large tube 4D > 4f L the exit Mach number increases with the D 14 The word “entrance” referred to the tube and not to the nozzle.4 2 1. Further.05 0.1 0.8 4. FANNO FLOW P2/P1 Fanno Flow 4fL  D 4.2 P2/P1 2.4 0 0 0.8 2. If the tube contains no supersonic flow then reducing the nozzle throat area wouldn’t increase the entrance Mach number.2 0. 15 Strange? Frictionless nozzle has a larger resistance when the throat area decreases 16 It is one of the strange phenomenon that in one way increasing the resistance (changing the throat fL ) does not affect the flow rate.3 entrance14 Mach number.

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

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

9.51) is smaller than 4f L D retreat . Let denote ∆ Note that 4f L D sup 4f L D ¯L f − = 4D actual 4f L D max∞ 4f L D sup (9. 4f L D and subtract the actual extra 4f L D assuming shock at the left side 4f L D and subtract the actual extra 4f L D assuming shock at the right s iii) According to the positive or negative utilizes your root finding procedure. the Mach number equal infinity when left side assumes result in infinity length of possible extra (the whole flow in the tube is subsonic).16). cRd3ePf g 1 4f L D max∞ ^`_Ia b 4f L D Fig. To overcome this numerical problem it is suggested to start the calculation from distance from the right hand side. FANNO FLOW shock location is at the maximum length.182 CHAPTER 9. The proposed procedure is based on Figure (9. The requirement that has to be sat- isfied is that denote as difference between the maximum possible of length in which the supersonic flow is achieved and the actual length in which the . ii) Calculate the extra (at the entrance). M1 to the tube as a function of sonic branch super- From numerical point of view. i) Calculate the extra the max length). then shock at Mx = 1 results in My = 1.17: The maximum entrance Mach number.

D exceed the maximum length D for this entrance fL Mach number. Hence.3102 Now the solution is somewhere exceeds the maximum 4f L D 4f L D max for the supersonic flow.4 (Mexit = 1).00849 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 2.9.4 The solution is obtained by an iterative process.1318 = 2. the value of left side is −0.5000 0.52) Figure (9.9 − 0.9.42390 1.5652 74.1318. In a summary of the actions is done by the following algorithm: (a) check if the (b) Guess 4f L D is 2. (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 .17) shows the entrance Mach number.39289 The extra ∆ 4f L D My 0. 4f L D = 0.1641 between the negative of left side to the positive of the right side17 . Accordingly continue. The retreating length is expressed as subsonic but 4f L D retreat = 4f L D max∞ − 4f L D sup (9. Or in a very long tube the whole flow will be subsonic.76820 (flow is choked and no additional 4D ). WORKING CONDITIONS 183 flow is supersonic see Figure (9. The right side is when the shock is at the entrance at which the extra 4f L D is calculated for Mx and My is 4f L D = 0.3591 0. Hence.3: Calculate the shock location for entrance Mach number M1 = 8 and for assume that k = 1.3867 5.39289 Ty Tx ρy ρx Py Px P0y P0 x 13.821508116. Example 9.0000 With (M1 ) M 0.7461 1.9 S OLUTION max for k = 1.76820.442 − 0.4417 2. M1 reduces after the maximum length is exceeded. The left side is when the shock occurs at Mx 8.1318.6136 2.15).76820 = 0. . thus the extra tube is ∆ 4f L D fL = 0. The maximum for M1 = 8 is 4D = 0. The maximum 4f L D 4f L 4f L is 0.

My calculated previous (e) Calculate 4f L D for supersonic branch for the Mx 4f L D up (f) Calculate the “new and improved” (g) Compute the “new 4f L D down = 4f L D − 4f L D up (h) Check the new and improved to stage (b).32932 0.6554 1.32932 0.90000 0.6382 1.6706 1. .57068 0.90000 0.32878 0.31783 0.6706 1.64830 0.64872 0. If it is satisfactory stop or retu 4fL D up 4fL D down Mx 1.57068 0.64834 0.57079 0.32930 0.32932 0.90000 0.90000 This procedure rapidly converted to the solution.57068 0.32907 0.64830 0. FANNO FLOW (d) Calculate the associate Mach number.57068 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 1.90000 0.64830 0.90000 0.57605 0.58217 0.66779 0.0000 4f L D down against the old one.6706 1.90000 0.57122 0.27830 0.6706 1.22574 0.64839 0.3838 1.64850 0.90000 0.57069 0.32395 0.57184 0.6691 1.30494 0.64920 0.67426 0.90000 0.57070 0.6635 1.5286 1.65728 0.0000 M2 1.69119 0.32931 0.90000 0. Shock location are: M1 8.64830 0.90000 0.64831 0.59506 0.90000 0.6706 1.64830 4fL D 0.32927 0.57093 0.90000 0.57318 0.65246 0.6699 1.74664 0.57068 0.64831 0.90000 0.32932 0.6673 1.32921 0.6705 1.32682 0.184 CHAPTER 9.57068 0.6703 1.64832 0.90000 0.32816 0.6706 My 0.65023 0.62170 0. Mx with the Mach number.6706 1.90000 0.90000 0.32932 0.6021 1.90000 0.6706 My 0.57073 0.

4D .3 M1 0.2) or the results from computer program attached to this book shows that reduction of the mass flow is very rapid.1 = 0. 9. THE APPROXIMATION OF THE FANNO FLOW BY ISOTHERMAL FLOW185 9.8 iso = 0. Hence. One must notice that the maximum temperature at the entrance is T0 1 .10.10 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.1 0 0 10 20 30 40 Wed Mar 9 11:38:27 2005 50 4fL  D 60 70 80 90 100 Fig.2 = 0.1 iso = 0.5 = 0.18) the dominating parameter is 4D . friction. The results are very similar for isothermal flow. When the Mach number decreases the temperature approaches the stagnation temperature (T → T0 ). Hence the mass flow rate is a function of 4D because M1 changes.15 even for large k = 1.8 M1 Fanno flow 0.2 0.9. The only difference is in small dimensionless fL . to insert a question or example about this issue in end with comperison to Isothermal Flow 0. say about 1% that flow can be assumed to be isothermal. Looking at the table or Figure (9.67.99 which requires that enough for M1 < 0.4 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 P 2 / P1 = 0.18: The entrance Mach number as a function of dimensionless resistance and comparison with Isothermal Flow fL As it can be seen for the Figure (9. This tolerance requires that (T0 − T )/T0 = 0. This fL requirement provides that somewhere (depend) in the vicinity of 4D = 25 the fL flow can be assumed isothermal. if one allows certain deviation of temperature.

99233 0.11 More Examples of Fanno Flow Example 9.98874 0.0000 40.1 the flow is choked.11637 0.48 287300 kg sec .0 2.3.7569 67. FANNO FLOW 9. = 4 × 0.99556 4. 0.99873 0.02 From Figure (9.50000 0.0000 40.99195 0.99683 4. Hence.0000 40.1 M1 ≈ 0.4206 0. ρ ρ0 M 0.4 ≈ 0.99097 0.127 × RT 4 1. or accurately by utilizing the program as in the following table. the mass rate is a function of the Mach number.8.0733 7.4: To demonstrate the utility in Figure (9. 0.0000 40.22697 0.1.6523 5.5.07975 M2 1. T ≈ T0 and is the same for the pressure. L = 4[m].022 k = 300000 × × 0.18) for P2 /P1 = 0.30000 0.02[m] and pressure ratio P2 /P1 = 0.12728 0.0000 0.1697 50.05.1196 7.42 0.12728 0. 4f L D 4f L D .1697 10.99354 0. M1 0.99677 0. 0.09965 4fL D 4fL D 1 4fL D 2 P2 P1 40.99692 0. D = 0.11392 0.05 × 4 = 40 0.5393 4. Find the mass flow rate for f = 0.07975 T T0 A A P P0 0.99741 0.13 etc. The mass flow rate is m ˙ = P AM π × 0.12420 0.7569 107.2842 0.186 CHAPTER 9. S OLUTION First calculate the dimensionless resistance.2519 A×P A∗ ×P0 Therefore.0000 42.98928 0. The stagnation conditions at the entrance are 300K and 3[bar] air.11392 0.80000 Only for the pressure ratio of 0.5910 4.40790 0.7027 5. 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.18) consider the following example.12420 0.

3085 1.1315 1.72805 0.03633 0.0691 0.61237 0.061 1.4626 1.00000 1.12 The Table for Fanno Flow Table 9.1381 1.017 1.48 × = 0.00 2.0787 1.0929 1.0207 1.0922 1.40825 1.2993 3.6191 2.1994 1.5333 8.87037 0.6400 2.2915 13.964 1.48 × = 0.66 106.1533 1.4027 3.53452 0.2200 13.633 1.30500 0.5901 2.9216 14.08 0.4961 66.1194 1.3184 2.1356 1.3817 14.08758 0.9.40 0.07664 0.35 0.1981 1.4487 2.3546 2.0382 1.1618 6.02 193.0351 3.5 ∼ 0.27217 0.5753 1.2549 1.21822 4.2893 1.1976 1.1273 kg sec kg sec kg sec 187 9.77894 0.7092 1.6183 1.06 0.1991 1.1489 1.9635 4.0944 1.4815 21.0638 1.35 280.48326 0.3665 1.1882 1.95781 1.1429 1.45 0.2616 12.90 0.4834 5.3005 27.3865 1.4935 1.06570 0. THE TABLE FOR FANNO FLOW and for the rest m ˙ 0.01451 0.9341 1.2119 1.91460 0.1996 1.04381 0.43 P1 0.43133 0.0485 1.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 787.002 1.49082 0.95 1.1242 P2 = 0.2508 9.0702 2.37879 0.3 ∼ 0.65 0.09 0.1273 0.12.0934 1.20814 0.00328 0.1628 1.07975 P2 = 0.1512 9.4182 10.044 1.1065 1.8218 5.70 0.08 440.32459 0.50918 0.5116 19.000 0.6958 1.6843 7.0693 1.52216 36.000 0.05476 0.73179 0.1713 1.12728 0.3398 1.03286 0.9435 5.4525 2.1852 1.1905 1.42857 .6416 8.07229 0.30 0.2620 15.66667 0.0624 1.4318 22.688 0.235 30.6659 15.000 0.5826 3.1378 4.1788 1.2047 1.09851 0.21822 0.72 83.468 P1 0.8254 18.8 ∼ 0.0474 11.55 0.5914 18.0089 1.1: Fanno Flow Standard basic Table M 0.03 0.8708 1.04 0.6742 3.2838 1.1985 1.48 × = 0.1139 P2 m ˙ = 0.7780 2.20 0.80 0.1988 1.50 0.3848 1.00 3.75 0.7634 1.82514 0.05 0.1998 1.03 140.58506 0.85 0.1273 0.25 0.10 0.63481 0.4613 10.1291 1.00 1.4554 2.5664 1.0327 1.0 0.30 m ˙ P1 0.32572 0.68374 0.10944 0.9034 11.60 0.07 0.

81755 0.446 2.40859 0.00200 2.82033 0.1E +5 0.00 70.236 2.4E +2 0.434 2.81582 0.00271 1.11111 0.188 CHAPTER 9.5E +5 0.00 9.3E +2 0. FANNO FLOW Table 9.00296 0.46771 0.448 2.00153 4.40875 0.42066 0.138 2.00 60.6E +4 0.390 2.43568 0.00 5.063758 53.00663 0.42857 0.3E +6 0.00142 0.295 2.440 2.448 0.42390 0.72988 0.9E +2 0.00122 .00 40.00166 0.029348 3.447 2.00 4fL D P P∗ P0 P0 ∗ ρ ρ∗ U U∗ T T∗ 0.089443 25.00 30.77899 0.333 2.00 8.1: Fanno Flow Standard basic Table (continue) M 4.20000 0.0E +2 0.5E +4 0.82008 0.057143 0.44721 0.00 55.448 2.4E +6 0.000809 2.446 2.00 7.40908 0.00952 0.00609 1.00488 0.00 50.00 35.8E +6 0.00 25.14634 0.63306 0.81860 0.014815 0.000979 1.69380 0.41833 0.81928 0.28571 0.40889 0.00240 0.40938 0.82052 0.069767 0.000579 5.40849 0.00198 0.377 2.00121 8.000680 3.6E +5 0.13363 10.00 0.00 20.036860 1.82066 0.445 2.00374 0.00 6.40846 2.76819 0.000500 7.023905 5.81975 0.443 2.40988 0.6E +6 0.447 2.359 2.8E +5 0.72 0.5E +6 0.41079 0.047619 1.18 0.00 10.40866 0.82078 0.78683 0.00 45.81265 0.40853 0.086957 0.00390 4.00 65.75280 0.

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

2).3) and (10.2 Governing Equation Q = Cp (T0 2 − T0 1 ) (10. Thus. One can notice that equations (10.4) can further assist in obtaining the temperature ratio as T2 P2 ρ 1 = (10. In analysis of the flow. for smaller range of temperature and length the calculations are more accurate. P2 1 + kM1 2 = P1 1 + kM2 2 (10. Yet. As appose to Fanno flow in which the resistance always oppose the the flow direction. RAYLEIGH FLOW situations involve chemical reaction. The density change though the heat transfer (temperature change).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 (10.6) T1 P1 ρ 2 The density ratio can be expressed in term of mass conservation as U2 r r √ kRT2 = kRT1 M2 M1 T2 T1 (10.4) are similar to the equations that were solved for the shock wave.7) U2 ρ1 = = ρ2 U1 kRT2 U1 kRT1 . in Rayleigh flow also cooling can be applied. the main characteristic of the flow such as chocking condition etc. one has to be aware that properties do change significantly for a large range of temperature.190 CHAPTER 10. (10. if the upstream conditions are known (or downstream condition are known).5) The equation of state (10.1) (10. are encapsulated in this model.2) (10. The basic physics of the flow revolves around the fact that the gas is highly compressible. a solution can be obtained. Nevertheless. 10. The flow velocity acceleration change the direction when the cooling is applied.4) A(P1 − P2 ) = m ˙ (V2 − V1 ) ρ 1 U1 A = ρ 2 U2 A = m ˙ There are four equations with four unknown.

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 = 0.6) yields T2 1 + kM1 2 M2 = T1 1 + kM2 2 M1 T2 T1 191 (10.8) Transferring the temperature ratio to left hand side and squaring results in T2 1 + kM1 2 = T1 1 + kM2 2 2 M2 M1 2 (10.10.12) . GOVERNING EQUATION Substituting equations (10. and s2 . 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 (10.2.11) results in 2(1 − M 2 ) 1 ds = Cp dM M (1 + kM 2 ) (10.9) {N|~ }€ y ‰XŠˆ‹ ne ŒŽˆŒ N‚„ƒ Pr es su re li co ns ta …X†ˆ‡ nt z Fig. The second maximum can be expressed as dT /ds = ∞ The second law is used to find the expression for derivative. A derivative of equation (10. 10.10) (10.5) and (10. and s1 are constant then the variable parameters are M2 .11) Let the initial condition M1 .

The derivative is equal to infinity. when cooling applied to a tube the velocity decreases and heating applied the velocity √ increases. Yet. increase of heating results in increase of entropy. RAYLEIGH FLOW Take the derivative of the equation (10. The peculiars point of M = 1/ k when additional heat is applied the temperature is decreasing. Yet for every curve.16) U∗ ρ1 = = ∗ ρ U1 1 The √U kRT ∗ kRT ∗ √ √ U1 kRT1 kRT1 ∗ √ = 1 M1 T∗ T1 (10. But the entropy of supersonic flow is lower (see the Figure (10.192 CHAPTER 10. The chocking is occurred only when M = 1 because it violate the second law.17) star is an asterisk.). yet note this point is not the choking point.15) (10.14) (10. Hence. 1 + kM1 2 P∗ = P1 1+k The temperature ratio is 1 T∗ = 2 T1 M 1 + kM1 2 1+k 2 (10. and M2 results in dT 1 − kM 2 = constant × 3 dM (1 + kM 2 ) Combining equations (10.2)) and therefore it is not possible (the maximum entropy at M = 1.5) can be written between chocking point and any point on the curve.12) and (10. dT /ds < 0. chocking can be explained by the fact increase of energy must accompanied by increase of entropy. The derivative is equal to zero when 1 − kM 2 = 0 or M = 1/ k or when M → 0.13) by eliminating dM results in M (1 − kM 2 ) dT = constant × ds (1 − M 2 )(1 + kM 2 )2 (10. It is convent to referrers to the value of M = 1. And cooling results in reduction of entropy. The equation (10. The derivative is negative. dT /ds = ∞ when M = 1. . From thermodynamics.13) On T-s diagram a family of curves can be drawn for a given constant. These value referred as the “star”1 values.12) when letting the variable parameters be T2 . some what similarly to Fanno flow. several observations can √ be generalized. The transition to supper sonic flow occurs when the area changes.

961 1.00431 0.259 1.00383 0.07 0.25096 0.395 2.029 1.178 1.95693 .019 1.57447 0.014300 0.34686 0.40 0.90850 0.020529 0.85 0.31373 0.61393 0.011 1.508 1.014 1.379 2.265 1.35 0.005 1.69751 0.80 0.384 2.00765 0.023669 0.75524 0.20661 0.030215 0.235 1.207 2. Several observations can be made in regards to the stagnation temperature.392 2.094 1.91670 0.52903 0.70 0.423 1.20 0.30440 0.25 0.017119 0.00216 0.001 0.61515 0.79012 0.266 1.343 1.99814 2.10 0. Table 10.43894 0.90 0.25684 0.3 Rayleigh Flow Tables The “star” values are tabulated in Table (10.135 1.056020 0.55 0.81013 0.08 0.388 2.075 1.058 1.3.03 0.40887 0.686 1.05 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 (10.011922 0.63713 0.00917 0.13793 0.94009 0.30 0.045616 0.870 1.090909 0.1: Rayleigh Flow k=1.114 1.98097 0.17355 0.86204 0.011680 0.267 1.60 0.00598 0.025 1.99207 0.04 0.261 1.70804 0.81892 0.036212 0.015 0.262 1.266 1.019222 0.264 1.4 M 0.373 2.96395 0.10.37865 0.09 0.96081 0.65 0.69136 0.75 0.00860 0.397 2.273 2.049 1.19183 0.060 1.1).99290 1.91097 0.50 0.44444 0.367 2.95 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 0.86833 0.45 0.218 1.266 1.596 1.157 1.75991 0.06 0.025 1.85987 0.193 1.125 1.131 2.51413 0.027841 0.030 1.778 1.199 1.038075 0.00517 0.023223 0.015224 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 (10.043 1.51001 0.19) 10.046777 0.

51349 0.146 1.699 1.11111 0.00469 0.0 9.240 1.4 2.18788 0.1E +4 3.28028 0.0E +4 1.256 1.4E +2 1.316 1.641 1.067263 0.52893 0.83628 0.49122 1.046693 0.3E +2 3.4 (continue) M 1.9 2.475 1.078 1.71006 0.61580 0.73954 0.33454 0.8E +2 1.0 25.79576 0.31486 0.5 10.026490 0.65377 0.044910 0.044 1.194 CHAPTER 10.21417 0.0 3.020979 0.503 1.675 1.403 1.4 1.58909 0.017021 0.30864 0.018846 0.00 0.20040 0.52438 0.176 1.22936 0.5 6.455 1.667 1.580 1.0 30.43353 0.3 1.00274 0.019 1.0 35.13223 0.56982 0.690 1.79339 0.26478 0.53633 0.227 12.95 54.49174 0.57831 0.055363 0.49259 0.33439 0.50702 0.93425 0.424 5.5 5.43122 0.028972 0.431 1.0 5.136 3.70174 0.023495 0.6 1.51647 0.81414 0.066667 0.0 4.686 1.000 1.058264 0.68 75.681 1.375 1.17647 0.49415 0.21 38.5 7.00140 1.52970 0.21419 0.00190 0.56734 0.698 1.36364 0.538 1.5 2.1 1.405 1.5 9.85917 0.494 1.743 1.52356 0.00732 0.2E +4 8.51098 0.0 2.713 1.50 18.7 2.7 1. RAYLEIGH FLOW Table 10.54473 0.16831 0.00428 0.588 1.80539 0.039900 0.64103 0.078 1.0 T T∗ T0 T0 ∗ P P∗ P0 P0 ∗ ρ∗ ρ 1.00 0.205 1.75613 0.092719 0.0 8.0 7.8 2.24615 0.5 4.620 1.0E +2 1.40384 0.95798 0.0 6.77406 0.045 2.10256 0.6 2.3 2.71301 0.00240 1.97872 0.35561 0.00 0.46106 0.13540 0.701 1.0 1.550 1.90928 0.050943 0.00326 0.68494 0.41 1.078487 0.1: Rayleigh Flow k=1.032053 0.96031 0.418 2.66350 0.5 1.873 3.99392 0.52004 0.60894 0.50885 0.702 1.711 1.37870 0.328 8.67380 0.7E +5 1.47562 0.85971 0.693 1.0E +2 3.034483 0.030094 0.301 1.712 1.00 1.72421 0.1 2.571 1.28551 0.55556 0.886 2.634 2.29687 0.49356 0.39643 0.525 1.122 1.69700 0.8E +2 2.005 1.89087 0.75250 0.035650 0.656 1.340 1.2 1.713 .695 1.91185 0.0 20.9 3.081772 0.2 2.8 1.63 27.65398 0.616 1.222 2.510 1.561 1.039883 0.5 8.88419 0.

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

195

M 40.0 45.0 50.0 55.0 60.0 65.0 70.0

T T∗

T0 T0 ∗

P P∗

P0 P0 ∗

ρ∗ ρ

0.00184 0.00145 0.00117 0.000971 0.000816 0.000695 0.000600

0.49089 0.49066 0.49050 0.49037 0.49028 0.49021 0.49015

0.00107 0.000846 0.000686 0.000567 0.000476 0.000406 0.000350

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

1.714 1.714 1.714 1.714 1.714 1.714 1.714

The Data is presented in Figure (10.3).

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

3

@( M = ∞ ) 1.7143

1 @( M = ∞ ) 0.4898 0

0.1

1

M

10

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

196

CHAPTER 10. RAYLEIGH FLOW

10.4

Examples For Rayleigh Flow

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

0.30440

0.25684

2.2069

1.2177

0.13793

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

0.40530

0.34376

2.1341

1.1992

0.18991

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

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

0.25000 0.98765 0.96942 2.4027

0.95745 2.3005

1.0424

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

P0 2 = P 1

P0 1 P1

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

10.4. EXAMPLES FOR RAYLEIGH FLOW

197

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

T0 1 = T 1

T0 1 T1

= 300/0.98765 = 303.75[K ]

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

0.29831 0.98251 0.95686 2.0454 The exit stagnation temperature is
isentropic

0.94012 1.9229

0.90103

T0 2 = T 2 The heat release becomes

T0 2 T2

= 400/0.98765 = 407.12[K ]

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

1.0286

0.97959

1.2000

1.0116

0.85714

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

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

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

0.62666 0.64051 0.53376

T0max 1013.7 = = 1158.51[K ] Tmax 0.875

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

198

CHAPTER 10. RAYLEIGH FLOW

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

0.83333 0.63394 1.0000

0.52828 0.52828 0.52828

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

T0 2 1000 = = 1200.0[K ] T2 0.83333

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

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

P P∗

0.50160

0.42789

2.0589

1.1805

0.24362

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

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

199

0.28028

0.65398

0.17647

3.4245

1.5882

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

2.6790

3.8571

10.3333

0.32834

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

0.47519

0.40469

2.0802

1.1857

0.22844

M 0.47519

T T∗

T0 T0 ∗

P P∗

P0 P0 ∗

ρ∗ ρ

0.75086

0.65398

1.8235

1.1244

0.41176

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

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

200

CHAPTER 10. RAYLEIGH FLOW

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. These books are OCR. 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 . that Volume forced models in traditional classes for with compressExternal forces that control the tank volume ible flow (sometimes referred to as gas Fanno model dynamics) don’t provide a demonstrafor relativly short tube tion to applicability of the class mateš“›•œj ž Ÿ#¡1¢#£"¤ rial aside to aeronautical spectrum even Isothermal model such as turbomachinery. However. 1964. in life. There is a significant impor.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 . the situations. this example can further strength and enhance this conviction. by M. and “Compressible Fluid Flow. International Textbook Co. A. J.Fig. the undersigned discover two text books which to include some material related to this topic. 2 Even if the instructor feels that their students are convinced about the importance of the compressible. However.” 2nd Edition. A. are far Fanno model 1 After completion of these Chapters. Fundamentals of Gas Dynamics. It Isothermal model for relatively long tube was the undersigned experience.. 1985. Scranton. Pennsylvania. in many instances. Prentice Hall. 201 .. these books contained only limit discussions on the evacuation of chamber with attached nozzle. Saad.

The Isothermal model is more appropriate for cases where the tube is relatively long in–which a significant heat transfer occurs keeping the temperature almost constant. are presented (see Figure (11. at this stage. 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. Yet Isothermal flow model is used as the limiting case. Addition- . models. were the chamber’s volume is a function of external forces. were the chamber volume is controlled or a function of the pressure. Fanno flow model is the most appropriate in the case where the filling and Control volume for the filling case evacuating is relatively fast. 11. In the second Chapter. no discussion is ofÉÊIËKÌ ÍÂÎ Ï Ð Ñ(Ò Ó fered here. should be larger than 400. As it was shown in ChapfL ter (9) the resistance. here Isothermal flow and Fanno flow are used as limiting or bounding cases for the real flow. are discussed. 4D . most car today equipped with an airbag.202 CHAPTER 11. This author isn’t familiar with situations in which Control volume for the evacuating case Rayleigh flow model is applicable. And therefore.3: A schematic of the control volIsothermal flow is appropriate model. EVACUATING AND FILLING A SEMI RIGID CHAMBERS more complicate. Yet umes used in this model as it was stated before. 11.2: A schematic of two possible connections of the tube to a single chamber The Rayleigh flow model requires ¼½I¾K¿ ÀÂÁ Ã Ä that a constant heat transfer supplied either Å(Æ Ç È by chemical reactions or otherwise. 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. 11. For instance.1)). As it was shown in this book the most appropriate model for the flow in the tube for a relatively fast situation is Fanno Flow. In case the fL filling is relatively slow (long 4D than the Fig. For example. In the first Chapter models. The analysis is extended to include a semi rigid tank. the models in this Chapter are suitable for study of the filling the airbag or filling bicycle with air.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. fanno model reduced connection ²§³©´oµ ¶w·#¸¯¹2º¯» for relatively short tube fanno model 1 2 ¥§¦©¨oª «w¬#­¯®2°¯± A schematic of a direct connection 1 for relatively short tube 2 Isothermal model for a relatively long tube Isothermal model for a relatively long tube The connection is through a narrow passage Fig.

11.1. GOVERNING EQUATIONS AND ASSUMPTIONS

203

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

ÔXÕÖÔ*× ÜÞÝ©ß

Ø

ÙÛÚ

Ø
1 2

And for perfect gas the mass at any given time is

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

m=

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

(11.2)

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

204

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

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

d dt

PV RT

P2 )=0 P1

(11.3)

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

11.2

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

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

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

(11.5)

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

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

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

(11.6)

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

11.2. GENERAL MODEL AND NON-DIMENSIONED

205

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

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

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

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

−(k+1) 2(k−1)

(11.11)

−(k+1) 2(k−1)

(11.12)

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

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

11.2.1

Isentropic Process

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

¯ =P

k−1 k

(11.14)

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

206

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

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

(11.15)

(11.16)

(11.17a)

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

(11.17b)

(11.17c)

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

11.2.2 11.2.3

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

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

11.3. RIGID TANK WITH NOZZLE

207

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

=

4f Lmax D 1

4f Lmax D 2

(11.18)

and k−1 P2 = 1+ M2 2 P0 (t) 2
k 1−k

M1 M2

1+ 1+

2 k−1 2 M2 2 k−1 2 M1

k+1 k−1

(11.19)

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

11.3

Rigid Tank with Nozzle

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

11.3.1

Adiabatic Isentropic Nozzle Attached

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

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

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

=

k+1 2

−(k+1) 2(k−1)

(11.22)

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

208

CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS

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

(11.23)

Equation (11.23) can be integrated as
¯ P

P
1

1−k 2k

t

dP ±

dt = 0
0

(11.24)

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

(11.25)

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

(11.26)

And for smaller area

k+1 2

−()

(11.27)

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

−1

2.4 2

−2.4 0.8

= 0.013[sec]

(11.28)

4.22). ¯ = 1 and for isothermal nozzle T ¯ = 1 Thus.31) 11. 11. the general equation (11. For rigid volume.1 Rapid evacuating of a rigid tank With Fanno Flow ¯ (t) = 1. 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. equation (11. f [M ] is expressed by equation (11. Owczarek [1964] found an analytical solution which described here.30) Here.3.17) “shrinks” and doesn’t contain the relative volume term.2 Isothermal Nozzle Attached In this case the process in nozzle is assumed to isothermal but the process in the chamber is isentropic. Yet. V tank.32) .30) transformed into ¯= ln P k+1 2 " −(k+1) 2(k−1) ¯ t # ¯=e P 2(k−1) t ¯ ( k+1 2 ) −(k+1) (11. After the integration. Thus.13) is reduced into V ¯ dP ¯=0 = ±f [M ]P ¯ dt Separating the variables and rearranging equation (11. the differential temperature change in the chamber is slower than the temperature change in nozzle.29) ¯ dP ± f [M ] ¯ P ¯ t ¯= 0 dt 0 (11. is constant and equal one for a completely rigid The relative Volume. A reasonable model for the tank is isentropic (can be replaced polytropic relationship) and Fanno flow are assumed for the flow in the tube.4.11. the specific governing equation is ¯ −1 dP ¯ [M ]P ¯ 3k 2k =0 − k Mf ¯ dt (11. The temperature in the nozzle is changing because the temperature in the chamber is changing.29) converted into ¯ P 1 (11.4 11. In such case. equation (11.

The solution of equation (11.6 0. 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). Mmax ¯ = 1. But for the chocked case it appears as ¯ ¯ = 1 + (k − 1)M f [M ] t ¯ P 2 2k 1−k (11.2 0 0 0. The Figure (11.32) is obtained by noticing that and therefore M ¯ is not a function of time and by variables separation results in M ¯ t ¯ P ¯= dt 0 1 ¯ dP ¯ f [M ]P ¯ kM 3k−1 2k = 1 ¯ k M f [M ] ¯ P 1 3k ¯ ¯ 1− 2k dP P (11. Thus.0 Fig. The transition point is denoted here as chT .35) The function is drawn as shown here in Figure (11.e.34) ¯ is a function of P and can It has to be realized that this is “reversed” function i.36) .0 P(t) 0. 11.4 0. t be reversed for case.8 V(t) = P (t) V(t) = P (0) 0.33) results in ¯= t −k 2 ¯ 12 k − 1 P ¯ (k − 1)M f [M ] (11.33) direct integration of equation (11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS For a choked flow the entrance Mach number to the tube is at its maximum.2 0.8 1.6 0.34) has to include the entrance Mach under the integration sign as ¯ P ¯− t ¯chT = t PchT 3k 1 ¯ ¯ 1− 2k dP ¯ [M ] P k Mf (11. that when the modified reduced pressure equal to one the reduced time is zero.5: The reduced time as a function of the modified reduced pressure The big struggle look for suggestion for better notation.5).210 CHAPTER 11. equation (11.5) shows 1.4 ¯ t 0.

The transition point is denoted here as chT . equation .4. And in the choke region the calculations can use the choking formula and numerical calculations for the rest.2: A chamber with volume of 0. 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. when the flow became chokeless the entrance Mach number does not change much from the choking condition. Mmax ¯ = 1. Again.4. unless extra long time or extra low pressure is calculated/needed. At some point the flow becomes chokeless flow (unless the back pressure is a complete vacuum). The solution of equation (11. The Figure shows that when the modified reduced pressure equal to one the reduced time is zero.11. Further.2 Filling Process ¯ −1 dP ¯ ¯ 3k 2k =0 ¯ − k Mf [M ]P dt The governing equation is (11.1[m3 ] is filled with air at pressure of 10[Bar]. Example 11.0[m] S OLUTION The first parameter that calculated is 4f L 4f L D D =5 11.38) results in ¯= t −k 2 ¯ 12 k − 1 ¯ [M ] P (k − 1)Mf (11.5).01[m] and length of L = 5.37) For a choked flow the entrance Mach number to the tube is at its maximum. with today computer this should not be a problem and easily can be drawn as shown here in Figure (11. Nevertheless. The chamber is connected with a rubber tube with f = 0. Thus. for the special cases where the choked equation is not applicable the integration has to be separated into zones: choked and chokeless flow regions.025.37) is obtained by noticing that and therefore M ¯ is not a function of time and by variable separation results in M ¯ t ¯ P ¯= dt 0 1 ¯ dP ¯ [M ]P ¯ k Mf 3k−1 2k = 1 ¯ k Mf [M ] ¯ P 1 3k ¯ 1− ¯ 2k dP P (11. d = 0.39) It has to be realized that this is a reversed function.38) direct integration of equation (11. The reduced time increases with decrease of the pressure in the tank.

4. T isentropic tank and Isothermal flow in √the tube is different from Fanno flow in that the chocking condition occurs at 1/ k .8 0.6: The reduced time as a function of the modified reduced pressure (11.6 0. For the double isothermal (chamber and tube) the equation (11.8 1 or ñ™òIóõô/ö"ñø÷ùòP ô/ö¯ü(ý þ ÿ úû ò¡  Fig.6) reduced into ρ c(0) ¯V ¯ ¯1 P (0) P P (0)V (0) d P ¯ (t ¯) = 0 ± A kRT (0)Mmax M ¯ tc RT (0) dt R T (0) (11. the relative temperature. It has to be remembered that the chamber can undergo isothermal process.6 0.2 0 à™áPâäã3å¯æ(ç èˆéwêìë3í ð î ï 0.212 CHAPTER 11. The combination of the For Isothermal process.4 0. EVACUATING AND FILLING A SEMI RIGID CHAMBERS 1 V(t) = P(t) V(t) = V(0) 0.39) has to include the entrance Mach under the integration sign as ¯ P ¯− t ¯chT = t PchT 3k 1 ¯ 1− ¯ 2k dP P ¯ k Mf [M ] (11.41) . 11.3 The Isothermal Process ¯ = 1.40) 11.2 0.4 0. This model is reasonably appropriated when the chamber is insulated and not flat while the tube is relatively long and the process is relatively long.

42) into yields d ¯ 1+k P k ¯ dt Carrying differentiation result in ¯ +1 1 dP 1+k ¯k ¯ f [M ] = 0 ¯ k2 k M P −P ¯ k dt Similarly as before. Further. The analysis is based only on the mass conservation and if unsteady effects are required more terms (physical quantities) have taken into account. After all. .5 The “Simple” General Case The relationship between the pressure and the volume from the physical point of view must be monotonous. the relation must be also positive. Substituting equation (11. a general solution is easily can be obtained similarly to rigid tank. in the known situations to this author pressure increase results in volume decrease (at least for ideal gas.42) where a is a constant that represent the physics.46) The physical meaning that the pressure remains larger thorough evacuating process.45) integrated to obtain the form −1 2k 2 ¯ 3k ¯= 2k t 1−P ¯ Mf [M ](3k − 1)(1 + k ) (11. RAPID EVACUATING OF A RIGID TANK 213 11. the variables are separated as ¯ t ¯ −P k+1 2k ¯ f [M ] = 0 M (11. The physical explanation when it occurs beyond the scope of this book. it is assumed the ideal gas applied to the gas and this assumption isn’t relaxed here. This situation occurs at least in small ranges for airbag balloon etc.). In this analysis and previous analysis the initial effect of the chamber container inertia is neglected.45) The equation (11. increase of the pressure results in increase of the volume (as results of Hook’s law.43) (11.4. 11.44) dt = 0 k 1+k ¯ P 1 −1 ¯ k2 ¯ k dP P ¯ M f [M ] (11.4 Simple Semi Rigid Chamber A simple relation of semi rigid chamber when the volume of the chamber is linearly related to the pressure as V (t) = aP (t) (11.4.4. Further. Nevertheless.11. as results in faster reduction of the gas from the chamber.

4. it is waited to find such conditions before calling it as critical condition.47) When n can be any positive value including zero. After carrying–out differentiation results ¯ +1 1 + nk − k ¯ 1+nk−2k dP ¯ f [M ] = 0 ¯ k2 k k M P −P ¯ k dt (11. it is convenient for engineers to further study the point where the relationship between the reduced time and the reduced pressure are linear6 Value of n above it will Convex and below it concave. However. This undersigned is not aware situation where this indeed play important role. In reality the value of n lays between zero to one. as first approximation and simplified approach can be done by a single term with a different power as V (t) = aP n (11.49) reduced to equation (11. Yet.43). variables are separated and integrated as follows ¯ t 0 1 + nk − k dt = k ¯ P 1 −5k ¯ 1+2nk ¯ 2k P dP ¯ [M ] Mf (11.50) Carrying–out the integration for the initial part if exit results in nk−1 2k 2 ¯ 3k −2 ¯= 2k t ¯ [M ](3k − 2nk − 1)(1 + k ) 1 − P Mf (11. 0. similarly as before. When n is approaching to zero the chamber is approaches to a rigid tank and vis versa when the n → 1 the chamber is flexible like a balloon.49) Again.785714286) for k = 1.52) . Therefore. 3k − 2 2k (11. The physical meaning of n = 0 is that the tank is rigid. There isn’t a real critical value to n. d ¯ 1+nk−k k ¯ P dt +1 ¯ f [M ] = 0 ¯ k2 k M −P (11.51) The linear condition are obtain when 3k − 2nk − 1 = 1 −→ n = That is just bellow 1 (n = 0.214 CHAPTER 11. EVACUATING AND FILLING A SEMI RIGID CHAMBERS Any continuous positive monotonic function can be expressed into a polynomial function.48) Notice that when n = 1 equation (11. 6 Some suggested this border point as infinite evocation to infinite time for evacuation etc.

5 Advance Topics fL is very large for small values of the entrance Mach number which The term 4D requires keeping many digits in the calculation.55) This solution should used only for Min < 0. The solution of equation (11.19) as Min Pexit = . Goldstein.18) and (11.53) and equation (11.00286. otherwise equations (11. E R G.53) and (11. For small values of the Mach numbers. pp.11. P0 (t) Mexit The solution of two equations (11. Vol. 118. R. Journal of Manufacturing Science and Engineering (USA). May 1996.54) Min = . equation (11.18) and (11.5. 2. . J. G. (11. Eckert. 259-265.19) is described in “Pressure die casting: a model of vacuum pumping” Bar-Meir. ADVANCE TOPICS 215 11.18) can be approximated as 4f L D = 1 Mexit 2 − Min 2 k Mexit 2 Min 2 (11.19) must be solved numerically.54) yields 1− Pexit P0 (t) fL k 4D 2 (11. no.

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

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

1)). EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 12.1.1 Rapid Process Clearly under the assumption of rapid process the heat transfer can be neglected and Fanno flow can be assumed for the tube. isontropic process ¢¤£¦¥¨§ ©. The first approximation isotropic process describe the process inside the cylinder (see Figure (12.218CHAPTER 12.

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

3) is an nonlinear first order differential equation and can be rearranged as follows ¯ ¯ dt dP = k−1 ¯ tmax ¯ 1 − t ¯ ¯ 2k P k 1 − tc M f [M ]P .1.6) (k − 1) (1 − Bξ ) [ξ ] ¯ dt ¯ 1−t (12. The second characteristic time.5)).4)) can be obtained by transforming and by 2k −1 ¯ k2 ¯ = [ξ ] k−1 .2) The governing equation (11. tc and the “maximum” time.8) . The first characteristic time.3) ¯ = t/tmax .1) can be non–dimensionlassed as ¯ (t ¯) = 1 − t ¯ V 219 (12. Notice that in this case that there are two different characwhere t teristic times: the “characteristic” time. tmax is associated with the imposed time on the system (in this case the elapsed time of the piston stroke).12.4) is can be solved only when the flow is chocked In which case f [m] isn’t function of the time.5) Equation (12. Equation (12.7) ¯ =t (12.4) 2 [ξ ]( k−1 ) 2k −1 dξ 2k k−1 = where B = tmax ¯ tc M f [M ] And equation (12.6) can be integrated to obtain or in a different form 1 − Bξ 2 ¯ ln = − ln t (k − 1)B ξ 1 − Bξ ξ 2 (1−k)B (12.5) can be further simplified as ¯ dt 2dξ = ¯ (k − 1) (1 − Bξ ) ξ 1−t (12.10) that was developed in the previous Chapter (11) obtained the form as ¯ P 1 k ¯ dP ¯ ¯ 1V dV + ¯ ¯ k P dt dt + ¯ f (M ) tmax M ¯ P tc k+1 2k =0 (12. tc is associated with the ratio of the volume and the tube characteristics (see equation (11. The solution of equation (12. P (12. ¯ (0) = 1. The reduced Presk and therefore P introducing a new variable ξ = P 2k ¯ = 2k [ξ ]( k−1 )−1 dξ Utilizing this definition and there implication sure derivative.4) Equation (12. dP k−1 reduce equation (12. GENERAL MODEL Equation (12. tmax .

11) The analytical solution also can be approximated by a simpler equation as ¯ = [1 − t] P tmax tc (12. When tc > 1 the pressure increases significantly and verse versa. A can be expressed as A= The actual times ratio tmax tc @A A Ac Ac (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. Thus the exit area.2) and 12.10) P (0) = 1 (12. These Figures (12.2). to create a direct function equation (12. The solution is applicable to indirect connection.3) in which outside tube condition is in vacuum. In the literature.14) @Ac .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. Thus.3 demonstrate the imtmax portance of the ratio of tmax tc . = 1 as the critical area.220CHAPTER 12.9) The analytical solution is applicable only in the case which the flow is choked thorough all the process. the question remains how the time ratio can be transfered to parameters that can the engineer can design in the system. EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT Now substituting to the “preferred” variable 1− −1 tmax ¯ ¯ k2 k tc Mf [M ]P −1 ¯ k2 k P 2 t ¯ f [M ] (1−k) max M tc 1 ¯ =t ¯ P (12. In this case only some part of the flow is choked (the later part). The results of a choked case are presented in Figure (12. In case when the flow chokeless a numerical integration needed to be performed.4) is transformed into k 1− ¯ dP = ¯ dt with the initial condition of −1 tmax ¯ ¯ k2 k tc Mf [M ]P ¯ 1−t (12.). Ac Denoting the area that creates the ratio tmax tc provides the needed tool.13) can be expressed as 1 tmax tc = @A tmax tc @A tmax tc (12.

1. The effects of the area ratio. tmax = 0. For small values of A there pressure in this quasi steady state is a function of A c c A is no steady state stage. A Mexit .1. tc ∝ 1/A. GENERAL MODEL 221 According to equation (11. according to values. This parameter represents the ratio between the filling time and the evacuating time. The required solidification time. The pressure and the mass flow rate are maintained constant after this state is reached.2 Examples Example 12.4). Thus. It is important to point out the significance of the tmax tc .14) the tmax is canceled and reduced into tmax tc = @A A Ac (12. it was found that the influence of the parameter 4f L D on the pressure development in the cylinder is quite small. This quasi steady state is achieved when the volumetric air flow rate out is equal to the volume pushed by the piston. . A . and the friction Parameters influencing the process are the area ratio. the time which would be required to evacuate the cylinder for constant mass flow rate at the maximum Mach number when the gas temperature and pressure remain in their initial A . When Ac is greater than one the pressure is concave upA is less than one the pressure is concave downward as shown ward and when A c in Figures (12. D .3 Direct Connection In the above analysis is applicable to indirect connection. From other detailed calculations the author thesis (later to be published on this www. The influence is small on the residual air mass in the cylinder but larger on the Mach number. A c 4f L parameter.5) tc is inversely proportional to area. It should be noted that critical area. S OLUTION 12.9).15) A . The direct connection posts more mathematical difficulty because the critical area is not constant and time dependent. 12.org).12.03[sec]. This parameter also represents the dimensionless area. equation (12. A c the following equation Figure (12. Ac .1: Calculate the minimum required vent area for die casting process when the die volfL ume is 0. is not function of the time. The line that represents A = 1 is almost straight. c c A For large values of Ac the pressure increases the volume flow rate of the air until a quasi steady state is reached.4) describes the pressure as a function of the dimensionless A A time for various values of A . which was obtained by an integration of equation (12. The A A .001[m3] and 4D = 20. are studied here since it is the dominant c parameter.potto.1.

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

1 0.0 Fig.6 2.6 2.0 1.0 5.2 2.4 0.2 0. t.4 3. A/Ac 0. 12.0 2.0 0. Cylinder Volume Fraction Figure b 0.5 0.8 Dimensionless Area.0 1.2 0.0 0.2 1.0  ∗ ∗ ∗ o o o o o o ∗ ∇ o ∇ o ∗ ∇ ∇ ∗ ∇ ∇ ∗ ∗ ∇ ∇ ∇ ∗ ∇ ∗  ∇ ∗ ∇                ∗  ∇      ∇  ∇  ∇  ∗   ∇  ∗  ∗       ∗ ∇ ∇    ∗∗∇ ∇ ∇ ∗  ∗  ∇  ∇  ∇  ∇                               ∗    ∇ ∗ ∇ o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o = 5.6 4.2 1.4 Figure a 0.0 o  = 100.4 3.3 0.0 4.0 0.1 0.0 4.0 4fL D ∗ ∇ ∗ ∗ ∗ ∗ ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.4 1.0 0.8 1.2 0.6 0.2: The pressure ratio as a function of the dimensionless time for chokeless condition .8 Dimensionless Area.7 0.4 1.2 2.7  Dimensionless Time.5 0. SUMMARY 223 5.5 1.9 1.8 0.5 1.8 0.2 0.6 4.6 0.0 2.8 1.2 3.0 5.0 5.9 1.12. A/Ac 0.2 3.2. or.0 ∗ ∗ o o o o o ∗ ∇ o ∇ o ∗ ∇ ∇ ∗ ∇ ∇ ∗ ∇ ∇ ∇ ∗ ∇           ∗ ∗           ∇  ∇  ∇  ∗  ∇  ∇      ∗    ∗ ∗ ∇ ∇ ∇      ∗ ∗ ∇ ∇ ∇ ∗ ∗ ∇  ∇  ∇  ∇  ∇                               ∗    ∇ ∗ o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o ∇ ∇ 0.0 0.3 0.0 4fL D ∗ ∇ ∗ o ∗ ∗ ∗ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∇ ∗  ∗ P (t ) P (0)  3.

4: The pressure ratio as a function of the dimensionless time ..8 0. ∗ ∗ ∗ × × × × × × × × × × × × × × × × × 2 1...... . EVACUATING/FILING CHAMBERS UNDER EXTERNAL VOLUME CONT 2. . .2 0 0 × ×   × × ×   ×                          × × × 0. ∗ . . ∗ .....4 0.5 0...8 1.224CHAPTER 12. Ac .7 0. .1 .3: The pressure ratio as a function of the dimensionless time for choked condition Fig.∗..4  0..0 .5 ∗∗ ..6    × × × × × ×   × × × × × 0.. ∗ 0...5 ∗ . .2 ∗ . t. .. 4. 1.. 0. .. .. 1. CYLINDER VOLUME FRACTION 1 Fig.. ∗ 1.1 0. .. 12.3 0.9 DIMENSIONLESS TIME.6 0.   ..0 ∗ . ∗ .6 ...0 . ∗ . ∗ . ∗ . ∗ .2 P(t) P(0)  × 1  0.. .2 0... . A . .. ×. ∗ . ∗ . ∗ .8   0. ∗ . ∗ = 0... ∗ . 12.. ∗ ∗ ∗ .4 1. ∗ .. or... . . . ∗ . ∗ ...

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

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

The   other type of shock wave is the oblique shock. waves and asymptotics. The author believes that these issues are not relevant to most engineering students and practices. Soc.” Bull.” Phys. “Focusing of weak shock waves and the Von Neuman paradox of oblique shock reflection. Am. Furthermore. Rosales. Fluids 6.R. variation of the deflection angle from a zero (δ = 0) to a positive value results in oblique shock. In the literature oblique shock. “Rays. 14. even the whole book. Further changing the deflection angle to a negative value results in expansion waves. 227 .G.B. discussion on a normal shock was presented. 84. Those who would like to obtain more information. this section attempts to show the boundaries and the limits or connections of these models1 . and Prandtl–Meyer func tion are presented as three separate and different issues.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. Clearly. Von Neumann paradox. = 0 1 In this chapter. Math. 727 (1978). and E. However. Keller. 1874 (1994). should refer to J. one can view all these cases as three different Fig.  normal shock. these issues should not be introduced in introductory textbook of compressible flow.CHAPTER 14 Oblique-Shock 14.1 Preface to Oblique Shock In Chapter (5). However. a very limited discussion about reflection shocks and collisions of weak shock. etc are presented. A normal shock is a special type of shock wave. triple shock intersection. Tabak and R. The common representation is done by not showing the boundaries of these models.

2: The regions where oblique shock or locity must increase to achieve this rePrandtl–Meyer function exist. 14.1) which occurs when somewhere downstream a disturbance2 appears. sis. for a negative deflection angle (in this section’s notaNo Shock zone tion).2 Introduction to Prandtl–Meyer Function Decreasing the deflection angle re0◦ Prandtl sults in the same effects as before. This shock is commonly referred to as the oblique shock. the flow field is governed by a hyperbolic equation which deals with the case when information (like boundary conditions) reaches from downstream only if they are within the range of influence. For information such as the disturbance (boundary condition) to reach deep into the flow from the side requires time. Prandtl-Meyer function approaches closer to a zero deas opposed to oblique shock analyflection angle.Fig. Yet. Alternatively. This matching can occur only when there is a discontinuity in the flow field. 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.228 CHAPTER 14. the flow moves downstream and creates an angle. the gas flow must match the boundary conditions. During this time. The analysis shows that the flow ve. zero. . Notice quirement. which is around it will be shown in the next chapter. 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. normal shock occurs when a wall is straight/flat (δ = 0) as shown in Figure (14. 14. Thus. are examples of forces that create shock. 2 Zero velocity.2.1 Introduction Introduction to Oblique Shock A normal shock occurs when there is a disturbance downstream which imposes a boundary condition on the flow in which the fluid/gas can react only by a sharp change in the flow direction. The zero velocity can be found in a jet flowing into a still medium of gas. pressure boundary conditions.2 14. the increase in the upstream Mach number determines the downstream Mach number and the “negative” deflection angle. as discussed in Chapter (1)3 the flow behaves as it does in a hyperbolic field. It has to be pointed out that both the oblique shock and the Prandtl–Meyer function have a maximum point for M1 → ∞. ν∞(k ) θmax(k ) Meyer Oblique Function Shock The boundary conditions must match the geometry. In such a case. As it may be recalled. As a “no solution” zone. However. 3 This section is under construction and does not appear in the book yet.2. and different inclination angle.2)). OBLIQUE-SHOCK 14. When the deflection angle is increased. However. the maximum point for the Prandtl–Meyer function is much larger than the oblique shock by a factor of more than 2. the flow must be continuous.

If you do not like this description.-/. co-exist? Or perhaps a different model better describes the physics.” i. for a chosen arbi trary accuracy it requires a specific small area. which depends on the boundary layer. δ .” The tangent component does not change because it does not “move” across the shock line. You can use the traditional explanation. the change of the orientation makes the shock considerations two–dimensional. a one– dimensional shock can be Fig. The deflection angle. it was assumed that normal shock occurs at the same time. In this chapter. To match the boundary conditions. 14.1) 4 The author begs for forgiveness from those who view this description as offensive (There was an unpleasant email to the author accusing him of revolt against the holy of the holies.3) exhibits the schematic of the oblique shock.14.  However.3 Oblique Shock The shock occurs in realθ−δ $ %'& ity in situations where the ( )* + + shock has three–dimensional θ   !#" . the oblique shock and the Prandtl–Meyer function. OBLIQUE SHOCK 229 14. Alternately.). Figure (14.2. . 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 normal shock analysis dictates that after the shock. 0effects. Only the oblique shock’s normal component undergoes the “shock. using an infinite (or a two–dimensional) object produces a two–dimensional shock.. the mass balance reads Comp arsi on L ine ρ 1 U1 n = ρ 2 U2 n (14. the flow is always subsonic.3: A typical oblique shock schematic considered.3.e. you do not need the author’s permission. In such a case. The total flow after the oblique shock can also be supersonic. is the direction of the flow after the shock (parallel to the wall). The two– dimensional effects occur when the flow is affected from the “side. the stability issue will be examined in greater detail. the flow turns after the shock to be parallel to the inclination angle. please just ignore it.3 Introduction to Zero Inclination What happens when the inclination angle is zero? Which model is correct to use? Can these two conflicting models. Hence. 14. The three–dimensional 1#243 effects of the shock make it appear as a curved plane. Previously (see Chapter 5). a change in the flow direction4 .

1). Experiments and experience have shown that the common solution is the weak shock. . (14.7) The above velocity–geometry equations can also be expressed in term of Mach number. Yet the new relationship between the upstream Mach number. θ has to be solved. But because the literature referred to only two roots. Menikoff. and (14.6) U1 n U1 t (14.2). From the geometry it can be observed that tan θ = and tan(θ − δ ) = U2 n U2 t (14. and the Mach angle.2) (14. tan θ U1 n = tan(θ − δ ) U2 n (14. who raised the solution completeness issue. δ . and the third is an impossible solution (thermodynamically)6 . the term lesser extent is used.3) (14.4) are the same as the equations for normal shock with the exception that the total velocity is replaced by the perpendicular components.8) issue is due to R. The lesser extent angle is the unstable angle and the weak angle is the middle solution. OBLIQUE-SHOCK P1 + ρ 1 U 1 n 2 = P 2 + ρ 2 U 2 n 2 The momentum equation in the tangential direction yields U1 t = U 2 t The energy balance reads C p T1 + U1 n 2 U2 2 = C p T2 + n 2 2 (14. The first is referred to as the weak shock. here there are three possible pairs5 of solutions to these equations.230 The momentum equation reads CHAPTER 14.5) Unlike in the normal shock. the deflection angle. in which the shock turns to a lesser extent7 . 7 Actually this term is used from historical reasons.4) Equations (14. the second is the strong shock. The author is not aware of “simple” proof and a call to find a simple proof is needed. requires solving the entropy conservation equation. as sin θ = 6 The solution 5 This M1 n M1 (14.

As opposed to the normal shock. here angles (the second dimension) have to be determined.11).11) M1 t M1 M2 t M2 cos(θ − δ ) = The total energy across an oblique shock wave is constant.10) (14. and it follows that the total speed of sound is constant across the (oblique) shock. U1 t = U2 t the Mach number is M1 t = M2 t because the temperatures on both sides of the shock are different.14.12) The relationship between the properties can be determined by substituting M1 sin θ for of M1 into the normal shock relationship. (14.16) The Rankine–Hugoniot relations are the same as the relationship for the normal shock P2 − P 1 P2 − P 1 =k (14.13) The density and normal velocity ratio can be determined by the following equation (14.14) 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. and δ . It should be noted that although.8) through (14.15) (14. is a function of four unknowns of M1 . OBLIQUE SHOCK M2 n M2 231 sin(θ − δ ) = cos θ = (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. θ.17) ρ2 − ρ 1 ρ2 − ρ 1 . T1 = T2 . M2 .9) (14. Rearranging this set utilizing geometrical identities such as sin α = 2 sin α cos α results in tan δ = 2 cot θ M1 2 sin2 θ − 1 M1 2 (k + cos 2θ) + 2 (14.3. The solution from this set of four equations.

M1 and the deflection angle. M1 . 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. and the negative sign refers to the mirror image of the solution.4. 14. perhaps. 8 This point was pointed out by R. Other combinations of properties provide a relatively simple mathematical treatment. Clearly. and the solutions of selected pairs and selected relationships will be presented. Again. the most useful properties are upstream Mach number. thus. and Deflection Angle. OBLIQUE-SHOCK 14. the normal shock is a one–dimensional problem. Thus.18) requires that x has to be a real and positive number to obtain a real deflection angle8 .20) (14. sin θ must be positive.4 Solution of Mach Angle Oblique shock. and this will be discussed later. One has to be careful to use the normal components of the Mach numbers. The stagnation temperature contains the total velocity. the relationship between the properties can be determined by using the normal components or by utilizing the normal shock table developed earlier. only one parameter is required (to solve the problem). Probably. δ Again.1 Upstream Mach Number. Oblique shock is a two–dimensional problem and two properties must be provided so a solution can be found. if orientated to a coordinate perpendicular and parallel shock plane is like a normal shock.18) Equation (14. this set of parameters is. Thus. the most common and natural to examine. which create a somewhat complicated mathematical procedure.232 CHAPTER 14. He also suggested that θ is bounded by sin−1 1/M1 and .22) (14. the negative root of sin θ must be disregarded 1.19) (14.21) (14. Menikoff.

24) (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 . In the last case where D < 0. 9 The highest power of the equation (only with integer numbers) is the number of the roots. one root is real and two roots are complex. For example. θ. Maybe insert an example where imaginary solution results in no physical solution. so that the shock normal component is reduced to subsonic and yet parallel to the inclination angle.29) (14. all the roots are real and unequal.28) (14. to explain about complex numbers and imaginary numbers should be included. (14. in a quadratic equation there are two roots.4.27) 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. 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 (14.23) (14. SOLUTION OF MACH ANGLE 233 The solution of a cubic equation such as (14. From a mathematical point of view.25) R+ √ D.14. For the case D = 0.26) T = and where the definition of the D is 3 R− √ D (14. . 10 A call for suggestions. all the roots are real and at least two are identical. if D > 0. D > 0 occurs when no shock angle can be found.30) Only three roots can exist for the Mach angle.18) provides three roots9 .

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

in Figure (14. Only when the fluid is away from the object (smaller angle) liquid “sees” the object in a different inclination angle.0.4) and (14. M1 The first range is when the deflection angle reaches above the maximum point. a2 . The simple procedure For example. For example. k ). M1 −→ ∞ the maximum deflection angle is calculated when D = Q3 + R2 = 0. The flow is far away from the object and does not “see’ the object.4. This root. It can be noticed that the normal shock. in the case k = 1. For a given upstream Mach number. Once.4. 14.2 When No Oblique Shock Exist or When D > 0 Large deflection angle for given. the strong shock.14.” a change in the flow direction is no longer possible. In the alternative view. . M1 .4: Flow around spherically blunted 30◦ cone-cylinder with Mach number 2. Fig.4. and a3 for M1 = ∞. the imaginary angle is shown. the inclination angle reaches the “maximum potential energy. This can be done by evaluating the terms a1 . a1 = −1 − k sin2 δ 2 a2 = (k + 1) sin2 δ 4 a3 = 0 With these values the coefficients R and Q are 9(−)(1 + k sin2 δ ) (k+1)2 sin2 δ 4 R= and − (2)(−)(1 + k sin2 δ )2 54 Q= (1 + k sin2 δ )2 9 Solving equation (14. and the weak shock coexist. the wedge) in front of it and hence the normal shock occurs.28) after substituting these values of Q and R provides series of roots from which only one root is possible. a change in the inclination angle requires a larger energy to change the flow direction. the fluid “sees” the disturbance (in this case. is just above δmax ∼ π 4 (note that the maximum is also a function of the heat ratio. SOLUTION OF MACH ANGLE 235 14.5). for. This different inclination angle is sometimes referred to as an imaginary angle.

At this stage it is not clear who was the first to propose it. θmax is a function of only these parameters.12) and equating the results to zero creates relationship between the Mach number. 15 At first. Thus. θ. (14. When the right terms defined in (14.12). M1 . and (14. This can be shown mathematically by the argument that differentiating equation (14. J. M1 . one can conclude that the maximum Mach angle is only a function of the upstream Much number.5: The view of a large inclination angle from different points in the fluid field.20)-(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. Since in that equation there appears only the heat ratio k . The Procedure for Calculating The Maximum Deflection Point The maximum is obtained when D = 0.21).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.29).236 CHAPTER 14. it was seen as C. and Mach number. . M1 and the maximum Mach angle. English mathematician to be the creator but later an earlier version by several months was proposed by Bernard Grossman. It can be noticed that in equation (14.Chapman.30) are substituted into this equation and utilizing the trigonometrical sin2 δ + cos2 δ = 1 and other trigonometrical identities results in Maximum Deflection Mach Number’s equation in which is M1 2 (k + 1) (M1n 2 + 1) = 2(kM1n 4 + 2M1n 2 − 1) (14. the deflection angle is a function of the Mach angle and the upstream Mach number. 14. M1 .

Once M1n is found. The numerator of equation (14.79 . It should be noted that both the Maximum Mach Deflection equation and the maximum’s equation lead to the same conclusion that the maximum M1n is only a function of upstream the Mach number and the heat ratio k .32) is zero at different values of the denominator. d tan δ =0 dM1 (14. the maximum appears when θ has its maximum. the maximum deflection angle can be computed.36) . 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 must be pointed out that similar procedures can also be proposed (even though it does not appear in the literature).18 or θ = 67.12).34) provides a solution for Mmax .12) yields 2 2 kM1 4 sin4 θ + 2 − (k+1) M1 2 sin2 θ − 1 + (k+1) 2 M1 2 M1 d tan δ = 2M 2 1 dθ kM1 4 sin4 θ − (k − 1) + (k+1) M1 2 sin2 θ − 1 4 237 (14.14. Thus. Note that this procedure does not require an approximation of M1n to be made. It can be noticed that the Maximum Deflection Mach Number’s equation is also a quadratic equation for M1n 2 .4. a derivative can be taken with respect to M1 . it is sufficient to equate the numerator to zero to obtain the maximum.35) k+1 Hence. taking the derivative with respect to θ.34) and then solving equation (14. the Mach angle is sin θ = 2k .33) should be referred to as the maximum’s equation. Instead.32) Because tan is a monotonous function. To compare these two equations the simple case of Maximum for an infinite Mach number is examined. Thus.8). then the Mach angle can be easily calculated by equation (14. which makes ◦ θ = 1. The general solution of equation (14. SOLUTION OF MACH ANGLE The differentiation of the equation (14. With the value of θ utilizing equation (14.31) is (k + 1)M1 2 + 1 + M1n = (M1 2 M1 2 (k + 1)2 + 8(k 2 − 1) + 16(1 + k ) √ 2 k (14. for large Mach numbers. Thus. The nominator produces a quadratic equation for sin2 θ and only the positive value for sin2 θ is applied here.33) Equation (14. 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.

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 (14.1: 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 + .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. This can be partially demonstrated by evaluating the terms a1 .31) and neglect all the terms that are relatively small. This typical example is for those who like mathematics. Example 14. The analysis shows regardless of the value of the upstream Mach number D = 0 when δ = 0.238 CHAPTER 14. a2 . Start with equation (14. first the transition line in which D = 0 has to be determined. S OLUTION under construction The case of D ≥ 0 or 0 ≥ δ The second range in which D > 0 is when δ < 0. Thus. OBLIQUE-SHOCK Note that Maximum Deflection Mach Number’s equation can be extended to deal with more complicated equations of state (aside from the perfect gas model). Hint.39) . 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. This can be achieved by the standard mathematical procedure of equating D = 0.

6) is provided for the following demonstration. there are several papers suggesting zero strength Mach wave.39) into equation (14. The question of singular point or zero Mach wave strength are only of mathematical interest.0 3 0 16 can be generated from the wall 1 with zero deflection. mathematical challenge for those who like to work it out. In other ∞ -1 0 . Fluids 10 (1) . others suggest a singular point18 . M1 = 1.  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. which con. there was a transition from one kind of discontinuity to another. are several papers that attempt to prove this point in the past.0 -3 -1 . paper by Rosles.16 words. Nevertheless. 2. in the literature. Hence. SOLUTION OF MACH ANGLE 239 Substituting the values of Q and R equations (14. Note that. Once this analytical solution was published. “Caustics of weak shock waves. Substitution of all the above values into (14.14. 17 There 16 A . 18 See for example. the wall does not emit any signal to the flow (assuming zero viscosity).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 . ∞. 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 . Utilizing the symmetry and antisymmetry of the qualities of the cos and sin for δ < 0 demonstrates that D > 0 regardless of Mach number. Mach numbers to demonstrate that D is zero Nevertheless.2 ysis is no. And according to 9 this analysis. this proof became trivial.0.28) results in D = 0..6: The various coefficients of three different tradicts the common approach.28) provides the equation to be solved for δ . January 1998. In the range where δ ≤ 0.38) (14. But for non ideal gas (real gas) this solution is only an indication. 14. Tabak. Table (14.4. no Mach wave 2. the XX question is whether it is possiXXcoefficients XXX a1 a2 a3 ble for an oblique shock to exXX M1 X ist? The answer according to 3 this analysis and stability anal1. with a positive large deflection angle.Fig. in the previous case. this identity can be demonstrated by checking several points for example.” 206 Phys.

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

are far much simpler and a unique solution exists.4. M1 .3 Upstream Mach Number. and shock angle.5 2 1. θ δ 20 30 The solution for upstream Mach number. The deflection angle can be expressed as a function of these variables as cot δ = tan θ (k + 1)M1 2 −1 2(M1 2 sin2 θ − 1) (14. M1 .0 30.0 -0.5 0 90 80 70 60 50 40 30 20 10 0 -0.5 1 0. SOLUTION OF MACH ANGLE 241 mation of weak weak wave with nonzero strength has engineering applicability in a very limited cases. and Shock Angle. Oblique Shock 3 2.001 0.8: The calculation of D (possible error).4.001 k = 1 4 Mx=3 0 10 20 30 Wed Jun 22 15:03:35 2005 Fig. 14.0005 Myw Mys θs θw 0. especially in acoustic engineering.0005 0 0.0 δ 20.0 10. but for most cases it should be ignored. shock angle.14. and exit Mach number for M1 = 3 0 10 14.41) . θ.

θ . is limited. is limited between ∞ and 1/ sin2 θ. δ and θ It is sometimes useful to obtain a relationship where the two angles are known.4 Given Two Angles. Examining equation (14.9).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 (14.49) .47) (14. The range of given θ. M1 and θ. has to be in the range of sin−1 (1/M1 ) ≥ θ ≥ (π/2) (see Figure 14.46) (14. the possible range deflection angle.44) (14. upstream Mach number M1 . The first upstream Mach number. is unique.41) shows that the shock angle.242 or tan δ = CHAPTER 14.4. δ .45) (14.48) Even though the solution for these variables. 14. OBLIQUE-SHOCK 2 cot θ(M1 2 sin2 θ − 1) 2 + M1 2 (k + 1 − 2 sin2 θ) (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. M1 is M1 2 = 2(cot θ + tan δ ) sin 2θ − (tan δ )(k + cos 2θ) (14.

.51) (14. 14. δ must also be small as well. θ.4.0 < 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. SOLUTION OF MACH ANGLE subsonic weak solution 243 1.50) For a large upstream Mach number M1 and a small shock angle (yet not approaching zero).41) can be simplified into k+1 θ∼ δ = 2 (14.52) The results are consistent with the initial assumption which shows that it was an appropriate assumption.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 ) (14.14. the deflection angle. Equation (14.

{ { { { { 14. In spite of the small area were the current solution can be assumed. a simple tool is sufficient for limit accuracy. dimensional analysis is not completed. It is recommended that this issue be skipped and the time be devoted to other issues.4.10: Schematic of finite wedge with zero be area where current solution can be angle of attack. a Even finite wedge with limiting wall can be considered as an example for this discussion if the B.10)). 14. for most engineering applications. b At this stage. The dimensional analysis shows that only the doted area to Fig. known to the author. The section where the current analysis is close to reality can be esti2-D oblique shock on both sides mated from a dimensional analysis for s i s aly n the required accuracy or by a numera edge range ical method. In spite of recent trends. reports that only a weak shock is possible. The author is not aware of any such analysis in literature. the first solution is thermodynamical unstable. All the literature.244 CHAPTER 14. The . the analysis for the middle section produces a close result to reality (because of symmetry). Clearly. As before. the strong shock is unstable. for steady state and no known experiments showing that it exist can be found in the literature.” Another geometry that can be considered as two–dimensional is the cone (some referred to it as Taylor–Maccoll flow).L. at least. ate i d i rm range inte for any finite length of the wedge. Even though. b assumed as correct . the numerical works require many times a “reality check. Yet. this solution is also act as a “reality check” to any numerical analysis. OBLIQUE-SHOCK 14.5 Flow in a Semi–2D Shape The discussion so far was about the straight infinite long wedgea which is no shock flow direction ysis anal a “pure” 2–D configuration. Experimental and analytical work shows that the weak solution is the stable solution and a discussion is provided in the appendix of this chapter. Additionally. The common approach is to carry out numerical analysis. The analysis also provides additional value of the expected range. the cone is a three–dimensional problem. The end of the wedge must have a different configuration (see Figure (14. The mathematics is much more complicated but there are three solutions. the symmetrical nature of the cone creates a semi–2D problem. the sis naly al a norm range analysis needs to account for edge efno shock fects.6 Small δ “Weak Oblique shock” This interest in this topic is mostly from an academic point of view. In this case there are no edge effects and the geometry dictates slightly different results. is neglected. As opposed to the weak shock.4.

14.4. 14. the close proximity view provides a continuous turning of the deflection angle. the shock becomes possible. From the far view. In reality. the boundary layer increases the zone where a continuous δ flow exists.. and small normal upstream Mach numbers. studying this topic seems to come at the expense of other more important topics. In larger distance from the wall.7 Close and Far Views of the Oblique Shock In many cases.4. it should be covered in this book.2762 . the oblique shock is a weak “weak oblique” shock. Mx 1. oblique shock.4.1: Table of maximum values of the oblique Shock k=1. the oblique shock is an accumulation of many small (or again weak) “weak shocks. M1 ∼ 1+ . under construction. these small “shocks” are built or accumulate into a large and abrupt change (shock).4 The maximum values are summarized in the following Table .14. the far view shows a sharp transition.53) tan θ = . Yet.97131 δmax 1. the boundary θ layer (B. 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 14.1000 My 0.L. At the local view close to the wall.” However. SOLUTION OF MACH ANGLE 245 author is not aware of any single case in which this topic is used in real–world calculations. δ . the author admits that as long as there are instructors who examine their students on this issue. For small deflection angles. In this theory.11: A local and a far view of the imity to the wall. In fact.5152 θmax 76. 1 M1 2 − 1 (14. 14.) does not enter into the calculation. after the explicit analytical solution has been provided..8 Maximum Value of Oblique shock Table 14. However. δ . The boundary layer reduces the upstream flow velocity and therefore the shock does not exist at close proxFig.11).

6465 64.9668 64.7532 64.99047 0.7739 41. .6515 17.94387 0.95049 0.2309 65.8000 1.5000 1.91997 0.99559 δmax 3.0119 19. for large deflection angle.8443 65.5787 65.92478 0.7972 65.6000 3.9735 26.2000 2. Henderson and Menikoff20 suggested a procedure to calculate the maximum deflection angle for arbitrary equation of state21 .96335 0.0000 6.7908 44.2503 67.7000 1.0000 9.91871 0.92683 0. 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.8000 4.95435 0.3059 38.0000 5.2000 3.4000 1.0464 66. OBLIQUE-SHOCK Table 14.1: Maximum values of oblique shock (continue) k=1.0000 7.99337 0.4000 3.99440 0.9442 6.0000 My 0.92224 0. there isn’t a physical solution to an oblique shock.8000 3.3673 67. Since the flow “sees” the obstacle.4419 It must be noted that the calculations are for the perfect gas model.96942 0.6074 64. 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.1619 44.3000 1.91941 0.93083 0.9555 69.4.4 Mx 1.4000 2.1675 22.0000 3.1127 14.0734 35.3934 37.5875 34.9020 67. In some cases.1833 21.246 CHAPTER 14.9087 66.1196 67. 14.0399 65.92165 0.5676 65.4290 θmax 71.0000 10.6621 9.94925 0.7593 65.9 Detached shock When the mathematical quantity D becomes positive.3645 67. 179–210. this assumption might not be sufficient and different analysis is needed.97214 0.93629 0.4272 12.4398 43.0000 8.6000 1.9000 2.2000 1.4144 65.96630 0.0000 2.3275 36.2546 43.1028 28.7023 66.1177 42.8137 32.6814 30.98183 0.6934 64.6000 2.0922 38.93747 0.5671 66.95897 0.98714 0.3066 64.

12: The schematic for a round–tip bullet C the flow must be supersonic. Thus. The in a supersonic flow weak oblique shock is predicted to flow around the cone.14.5. a graph and a general explanation to engineers is provided. The zone A is zone where the flow must be subsonic because at the body the velocity must be zero (the no–slip condition). The detached shock’s distance from the body is a complex analysis and should be left to graduate class and researchers in this area. Larger shock results in a smaller detachment distance. Yet. The flow in zone A has to go through some acceleration to became supersonic flow. While at zone Fig. The constant and the function are different for different geometries. the increase in the upstream Mach results in a decrease of the detachment distance. The explanation to such a phenomenon is above the level of this book (where is the “throat” area question22 . Upstream U∞ Figure 14. SOLUTION OF MACH ANGLE 247 from the body. this phenomenon has a larger impact for a relatively smaller supersonic flow. 14. Analysis of the detached shock can be carried out by looking at a body with a round section moving in a supersonic flow (the absolute velocity 8 9M : . alternatively. 14. In the typical example. or.4. the gas must go through a shock. Regardless of the explanation.CG isn’t important for this discussion). it can be explained as the subsonic is “sucked” into gas in zone C.4. In such a case. This shock is referred to as the detach shock. As a general rule. E Strong Shock F > @ .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. these calculations can be summarized by the flowing equation detachment distance = constant × (θ − f (M∞ )) body thickness w ea k sh oc k (14.54) where f (M∞ ) is a function of the upstream Mach number which tabulated in the literature. <> = > ?1@ ? ACBD< . The diszone A Normal Shock tance of the detachment is deter5 67 mined to a large degree by the upH4I J K L M NOI P J'H4Q R J K P L I S T U R V NONWI J U Q L K X Y T Z V R stream Mach number. a wedge or 22 See example 14. some might be used in certain situations which the author isn’t aware of. Even though this topic has few applications. . the flow becomes “blinder” to obstacles.12 exhibits a round–tip bulzone B Subsonic Area θ let with a detached shock. Nevertheless.

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

12) 249 (d) Use the deflection angle.4.5686 θs 1. δ2 utilizing equation (14.3: A cone shown in Figure (14. the strong Mach numbers. For engineering purposes when the Mach number reaches this value. Note that no maximum angle is achieved in this shock.48523 My w 2.0000 My s 0.97234 δmax 38.7738 θmax 66. this zone increases as Mach number increases. predict at which Mach number was the photo taken based on the assumption that the cone is a wedge. What is the maximum wedge angle at which the oblique shock can occur? If the wedge angle is 20◦ .2: Air flows at Mach number (M1 ) or Mx = 4 is approaching a wedge.31) equation (c) Calculate the deflection angle.0407 To obtain the results of the weak and the strong solutions either utilize the equation (14. This procedure can be extended to calculate the maximum incoming Mach number.11 Oblique Shock Examples Example 14.28) (a side to the case proximity of δ = 0). θ2 . POTTO–GDC can be used to calculate this ratio.4. In general. SOLUTION OF MACH ANGLE (b) Calculate the maximum deflection angle. For example. In discussing these issues. at Mach 5 this zone is 8. utilizing (14.5◦ . 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 . calculate the weak.34907 Example 14.56660 δ 0. δ2 = δ1 and the Mach number M1 B to calculate M1 B .14.0000 My 0.28) or the GDC which yields the following results Mx 4.4635 θw 0.15) is exposed to supersonic flow and create an oblique shock. it can be ignored. 14. Using the geometry provided in the photo. Is the shock shown in the photo weak or strong shock? Explain. S OLUTION The maximum wedge angle for (Mx = 4) D has to be equal to zero. The maximum values are: Mx 4. The wedge angle that satisfies this requirement is by equation (14. and the respective shock angles.

OBLIQUE-SHOCK θ δ Fig.099◦. According to his measurement.0143 30. With given two angles the solution can be obtained by utilizing equation (14.88737 Because the flow is around the cone it must be a weak shock. . Grigory Toker. the shock would be weak because the maximum (transition to a strong shock) occurs at about 60◦ .56543 2.2318 My s My w θs θw δ P0 y P0 x 0. a Research Professor at Cuernavaco University of Mexico.49) or the Potto-GDC.2. Note that the Mach number is larger than the one predicted by the wedge.4522 71.250 CHAPTER 14. M1 3.15: Oblique shock occurs around a cone. the cone half angle is 15◦ and the Mach number is 2.0990 14.4300 0.43◦ and the shock angle is 30. S OLUTION The measurement shows that cone angle is 14. 14. This photo is courtesy of Dr. Even if the cone was a wedge.

5 1 0.0 9.0. 179–210. 14. a In fact.12 Application of Oblique Shock Mx 2 3 4 5 6 7 8 9 10 One of the practical applications of the ‘w’ oblique shock is the design of an in{4| } ~¡ €O ‚4|ƒ „ … †ˆ‡ ‰ Š‹4Œh Žˆ…  let suction for a supersonic flow.5 2 1. SOLUTION OF MACH ANGLE 251 k=14 Oblique Shock 3 2. P = 1.4.17: Two variations of inlet suction for flow just below M = 1. In such a case. the following example is provided.0 10. there is general proof that regardless to the equation of state (any kind of gas). and T = 17◦ C.4: The Section described in Figure 14. Clearly. the entropy is to be minimized through a series of oblique shocks rather than through a single normal shock. It is suggested that a series of weak shocks “•”—–•˜ should replace one normal shock to increase the efficiency (see Figure (14.0 Mx Thu Jun 30 15:14:53 2005 My θ δ Fig. Example 14.0[bar].0 7.0 8.18 air is flowing into a suction section at M = 2.4. the flow can be brought to a subsonic Fig.0 3.17))a .0 6. To illustrate the design significance of the oblique shock.14. supersonic flow there is less entropy production (less pressure loss). Compare the different conditions in the two .5 0 90 80 70 60 50 40 30 20 10 0 2.16: Maximum values of the properties in an oblique shock 14.0 5. For details see Henderson and Menikoff “Triple Shock Entropy Theorem.0 4. 14. with a proper design. (1998) pp.” Journal of Fluid Mechanics 366.

7498 resulting in Mx 1.5549 14. . the new angle is 7◦ + 7◦ with new upstream Mach number of Mx = 1.18: Schematic for Example (14.5000 0. 14.1931 0.7498 36.0000 0.7498 85. the first angle shown is Mx 2. OBLIQUE-SHOCK different configurations.0000 Py Px Ty Tx P0 y P0 x 1. In the past the input file was given but the graphical interface it is no longer needed.7498 My s My w θs θw δ P0 y P0 x 0.7021 36.0000 My 0.0000 My s My w θs θw δ 7.2485 1. no input explanation is given.99445 In the new region.9831 51.0000 My w θw δ 7. Assume that only a weak shock occurs.2098 0.252 CHAPTER 14.0000 P0 y P0 x 0. °²± ™/š›DœužyŸ• ¡š¡¢•£ 1 neglect the detached distance 7◦ ¤¥/¦ § ¨e©/ª¬«C­/¤¡®•¯« 2 ³µ´ 3 4 Normal shock 7◦ Fig.6875 2.96524 And the additional information is 23 The results in this example are obtained using the graphical interface of POTTO–GDC thus.72087 In the oblique shock.71761 1.2098 1.58974 1.6667 4.57735 Ty Tx ρy ρx Py Px P0 y P0 x 1.4) S OLUTION The first configuration is of a normal shock for which the results23 are Mx 2.2346 76.99445 and the additional information by the minimal info in the Potto-GDC is Mx 2.

6116 0.98903 × 0.1250 0. SOLUTION OF MACH ANGLE Mx 1.7498 My w θw δ 7.4458 0.5000 My 0. What is the requited “throat” area ratio to achieve acceleration from the subsonic region Fig.6147 P1 P3 P2 P1 The loss in this case is much less than in a direct normal shock.9496 P0 1 P0 3 P0 2 P0 1 The static pressure ratio for the second case is P4 P4 P3 P2 = = 1.45115 Ty Tx ρy ρx Py Px P0y P0 x 3. 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.2626 1.4458.285 = 2.45115 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.2346 My 0.4.5: 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 3.2626 × 1.5 (see Figure 14.1497 1.99445 = 0. In such a case.82141 Ty Tx ρy ρx Py Px P0y P0 x 1. The half wedge angle is 10◦ .1853 0.2574 Thus the area ratio has to be 1.6116 × 1.3151 4.98903 With two weak shock waves and a normal shock the total pressure loss is P04 P03 P02 P04 = = 0. .4 yields M 0.21295 Now utilizing the isentropic relationship for k = 1. 14. In fact.90506 1.19).4018 1. Example 14.8770 1.5088 41.5) 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 3.96089 0.86966 1. the results are: Mx 1. the loss in the normal shock is above than 31% of the total pressure.19: Schematic for Example (14.96524 × 0.99549 An oblique shock is not possible and normal shock occurs.2609 14.14.0000 Py Px Ty Tx P0 y P0 x 253 1.

8807 23. 2.9476 21.0000 My s My w θs θw δ 6.8482 My s My w θs θw δ 3.20).0000 My s My w θs θw δ 3.99879 = 0.13665 The combined shocks AB and BC provide the base of calculating the total pressure ratio at zone 3.99894 And the isentropic relationships for M = 2.99105 The transition for shock AB is Mx 3.7008 87.47641 2. Then. The total pressure ratio at zone 2 is P02 P02 P01 = = 0.7080 0.40596 0. 1 S OLUTION The shock BD is an oblique shock with a response to a total turn of 6◦ .0000 P0 y P0 x 0.9356 0.10500 0.0000 P0 y P0 x 0.5990 0.13632 0.7049 88.7008 are M 2.0.99879 For the shock BC the results are Mx 2.10548 3.8482 88. 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 (14.1854 0. shock E or expension Adopt this description and wave M 2 assume that the turn of 1 0 ◦ 6 is made of two equal C A angles of 3◦ (see Figure 14.04290 0.7008 T T0 ρ ρ0 A A P P0 A×P A∗ ×P0 0.0000 P0 y P0 x 0. Perform the calculation for M1 = 3.254 Example 14.20: Schematic of two angles turn with two weak shocks all the shocks are weak and straight.7049 2.20).48013 2. Find the required angle of shock BE.04263 0. explain why this description has internal conflict.997731283 P00 P01 P00 .40669 0.99894 × 0.1978 3. 14.7049.6: CHAPTER 14. The conditions for this shock are: Mx 3.8912 22. Assume that there are no boundary layers and Fig.48610 2.

042436789 = 1.72 δ P0 y P0 x 0.0019 1.4 and R = 287J/KgK .99920 Ty Tx ρy ρx Py Px P0y P0 x 1.7: Calculate the flow deflection angle and other parameters downstream when the Mach angle is 34◦ and P1 = 3[bar]. SOLUTION OF MACH ANGLE On the other hand.997731283 × 0. check whether the pressure at 2 is above or below or above the pressure (ratio) in zone 4. θ can be calculated from θ = sin−1 1.7037 0. The increase has to be P3 /P2 = 0.0008 My 0.001867743 To achieve this kind of pressure ratio the perpendicular component has to be Mx 1.04290 × 0. Assume k = 1.0013 1.042516045/0.026233 1. T1 = 27◦ C.4.99105 = 0. the pressure at 4 has to be P4 P4 P0 4 = = 0.49525 2. the angle for BE shock which cause this pressure ratio needs to be found. In reality the flow of double wedge will produce a curved shock surface with several zones.14.4).0005 1. To do that.7049 My s My w θs 0.00000 From the last calculation it is clear that the government proposed schematic of the double wedge is in conflict with the boundary condition. P0 P2 P2 = 2 = 0. Only when the flow is far away from the double wedge. Example 14.7049 = 21. the flow behaves as only one theoretical angle of 6◦ exist. and U1 = 1000m/sec. The flow in zone 3 will flow into the wall in about 2.7◦ .00000 The shock angle.042436789 P0 2 P0 0 P0 2 P2 P4 Since P < P a weak shock must occur to increase the static pressure (see 02 01 Figure 5.042516045 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.0008/2.715320879◦ The deflection angle for such shock angle with Mach number is Mx 2.04263 = 0.0 θw 21.

3949 × 300 ∼ 418.8: For Mach number 2.88 = 1.48269 2.3949 2.5 and wedge with a total angle of 22◦ .88 sin(34.8800 My s My w θs 0.96873 . the Upstream Mach number.0485 2.1) or GDC the following can be obtained.13) or simply converting the M1 to perpendicular component. M1n = M1 ∗ sin θ = 2. M1 .78 P0 y P0 x 0.89145 The temperature ratio combined upstream temperature yield T2 = 1.4 × 287 × 418.8124 δ 11.1280 0.4 × 287 × 300 kRT Using this Mach number and the Mach deflection in either using the Table or the figure or POTTO-GDC results in Mx 2.128 1.5000 My s My w θs 85.6[m/sec] Example 14.8575 × 3 = 8.5 and the angle of 11◦ results in Mx 2. Yet.0) = 1.66545 Ty Tx ρy ρx Py Px P0 y P0 x 1.0995 θw 32.0443 0.00 δ 15. Utilizing GDC for Mach number 2.89127 The relationship for the temperature and pressure can be obtained by using equation (14.8575 0. has to be determined M1 = √ U1 1000 = 2.0000 P0 y P0 x 0.6100 My 0.0 θw 34.61 From the Table (5.53431 2.5K and the same for the pressure P2 = 2. Mx 1.15) and (14.5 = 872. OBLIQUE-SHOCK S OLUTION The Mach angle of 34◦ is below maximum deflection which means that it is a weak shock. calculate the ratio of the stagnation pressure.256 CHAPTER 14.57[bar] And the velocity √ √ Un2 = My w kRT = 2.

is obtained. For the first shock.9: What is the maximum pressure ratio that can be obtained on wedge when the gas is flowing in 2.7822 δ 29. To obtain the maximum point for 2. what is the difference? S OLUTION It has to be recognized that without any other boundary condition.60027 In these calculations. Example 14. δ . S OLUTION This kind of problem is essentially two wedges placed in a certain geometry.10: Consider the schematic shown in the following figure.7974 Py Px Ty Tx P0 y P0 x 0.6854 0. either use the Maximum Deflection Mach number’s equation or the Potto–GDC Mx 2. For a weak shock the maximum pressure ratio is obtained at the deflection point because it is closest to a normal shock.5000 My max θmax 64. the shock is weak shock. then the Mach angle was calculated using the geometrical relationship of θ = sin−1 M1n /M1 . Utilizing the table or the Potto–GDC.5 Mach without any close boundaries? Would it make any difference if the wedge was flowing into the air? If so. the following can be obtained: . Maximum Deflection Mach’s equation was used to calculate the normal component of the upstream.14.3573 2.5 Mach number. the upstream Mach number is known together with deflection angle. SOLUTION OF MACH ANGLE 257 Example 14.12) the deflection angle. With these two quantities. 3 stream line 2 1 θ M1 = 4 δ Assume that the upstream Mach number is 4 and the deflection angle is δ = 15◦ . utilizing equation (14.4.94021 4. It is clear that the flow must be parallel to the wall. 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).

2028 32.9290 27.2808 32.96263 . With these values the following can be obtained: Mx 2.51367 2. Utilizing the POTTO–GDC which provides a solution in just a few clicks.51 P0 y P0 x 0.00 δ 10.5764 = 2.9290 My w θw δ Py Px Ty Tx P0 y P0 x 2.90041 and the additional information is Mx 2.6695 1.80382 And the additional information by using minimal information ratio button in Potto– GDC is Mx 4.929.632 T3 T2 T3 Example 14.46152 2. S OLUTION Here the Mach number and the Mach angle are given. Again calculate the downstream ratios after the second shock and the deflection angle.7985 × 1.3575 0. OBLIQUE-SHOCK θw δ P0 y P0 x 0.80382 With a Mach number of M = 2.11: A similar example as before but here Mach angle is 29◦ and Mach number is 2.0000 My w θw δ Py Px Ty Tx P0 y P0 x 2.0000 1.2028 84. the second deflection angle is also 15◦ .0000 0.7822 15.0026 P3 P2 P3 T1 T1 T2 = = 1.6695 = 3. With these pieces of information by utilizing the Potto-GDC the following is obtained: Mx 2. Note that hand calculations requires endless time looking up graphical representation of the solution.7822 15.5851 27.0629 15.0000 My s My w θs CHAPTER 14.258 Mx 4.0000 0.0629 15.90041 With the combined tables the ratios can be easily calculated.85.0 θw 29.5764 0.9290 My s My w θs θw δ P0 y P0 x 0. P1 P1 P2 = = 1.7344 0.7344 × 1.7985 1.9290 85.0000 1.48469 2.8500 My s My w θs 0.

5100 0.49901 And the additional information is .8021 T3 T2 T3 Example 14.3575 29.4089 1.0000 10.9419 34.0590 10.54894 1.0058 0.0000 My s My w θs 0. SOLUTION OF MACH ANGLE 259 and the additional information by utilizing the minimal info button in GDC provides Mx 2.51 the so called reflective shock gives the following information Mx 2.97569 P1 P1 P2 = = 1.12: Compare a direct normal shock to oblique shock with a normal shock.8000 5.9419 84.00 δ 20.0000 0. What is the deflection angle in this case? S OLUTION For the normal shock the results are Mx 5.06172 While the results for the oblique shock are Mx 5.3984 1.97 P3 P2 P3 T1 T1 T2 = = 1.3268 0.3575 My w θw δ Py Px Ty Tx P0 y P0 x 1.96263 With the deflection angle of δ = 10.41523 Ty Tx ρy ρx Py Px P0y P0 x 5.17 P0 y P0 x 0.3582 × 1.3984 ∼ 1.5131 1.0000 My 0.0000 29.0590 10.4089 × 1.9398 34.97569 and the additional information of Mx 2.0 θw 30.5100 1.3268 ∼ 1.3575 My s My w θs θw δ P0 y P0 x 0.4. 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◦ .8500 My w θw δ Py Px Ty Tx P0 y P0 x 2.14.41523 3.3582 0.

13)). Example 14. Based on these calculations. explain whether the slip angle is larger or smaller than the difference of the deflection angle. P1 P1 P2 = = 2.32671 The pressure ratios of the oblique shock with normal shock is the total shock in the second case.6858 3.6375 × 10.13: A flow in a tunnel ends up with two deflection angles from both sides (see the following Figure (14.49901 The normal shock that follows this oblique is Mx 3.0058 My 0.3740 0.0000 20.0000 My w θw δ CHAPTER 14.260 Mx 5.75 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.5141 × 2. calculate the pressure at zones 3 and 4 based on the assumption that the slip plane is half of the difference between the two deflection angles. OBLIQUE-SHOCK Py Px Ty Tx P0 y P0 x 3.0058 30. For the first direction of 15◦ and Mach number =5.36 P3 P2 P3 T1 T2 T1 = = 2.8625 10.1736 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. S OLUTION The first two zones immediately after are computed using the same techniques that were developed and discussed earlier.6858 ∼ 6.47485 Ty Tx ρy ρx Py Px P0 y P0 x 2.13) For upstream Mach number of 5 and deflection angle of 12◦ and 15◦ .5141 0.6375 2. .374 ∼ 27.

88496 And in zone 4 the conditions are due to deflection angle of 13.3217 15. the slip angle has two extreme cases.5◦ = 13.6986 27.6668 13.5000 0.0000 0. In this case.3217 15.5040 86. The first case is where match lower deflection angle and second is to match the higher deflection angle.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3. that only one deflection angle exist.69317 For the second direction of 12◦ and Mach number =5.9791 1. for the analysis.8006 21.2845 12.5000 1.88496 with the additional information Mx 3.43016 3. Under this assumption the conditions in zone 3 are solved by looking at the deflection angle of 12◦ + 1. SOLUTION OF MACH ANGLE Mx 5.9122 21.0000 0.0000 My w θw δ Py Px Ty Tx P0 y P0 x 3.6247 1.43914 3.5◦ which results in Mx 3. If the two sides were equal because of symmetry the slip angle is also zero.69317 And the additional conditions are Mx 5.80600 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 261 0. They are the pressure and the velocity direction.80600 And the additional conditions are Mx 5.6986 85. It has to be noticed that the velocity magnitudes in zone 3 and 4 do not have to be equal.6963 1.5◦ and Mach 3.5040 24.0000 1. For the two different deflection angles.8006 86.2845 12.4.9238 0. It is to say. This non–continuous velocity profile can occur in our model because it is assumed that fluid is non–viscous. Mx 5.47413 2.5040 My s My w θs θw δ P0 y P0 x 0.0000 1.14.0000 My s My w θs θw δ P0 y P0 x 0.6668 13.8006 .5656 0. it is assumed that the slip angle moves half of the angle to satisfy both of the deflection angles (first approximation).5040 My w θw δ Py Px Ty Tx P0 y P0 x 2.6625 0.0739 24.6819 27.

in oblique.3226 13. as oppose to the normal shock. the .5◦ to a larger number.6963 1.5 Summary As with normal shock.262 Mx 3.5000 1. 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).86179 with the additional information Mx 3. The optimum deflection angle is a function of the Mach number range in which the suction section is operated in. 14.6247 × 1. M1 is always greater than 1.9791 ∼ 1. To reduce the pressure the angle of slip plane has to increase from 1. Example 14.86179 From these tables the pressure ratio at zone 3 and 4 can be calculated P3 P2 P0 P1 1 P3 1 = = 1.46259 2.13 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.5000 0.9035 26.14: The previous example gave rise to another question on the order of the deflection angles. S OLUTION Waiting for the solution 14. the pressure at zone 3 has to be reduced.6038 0. OBLIQUE-SHOCK θw δ P0 y P0 x 0.8006 My w θw δ Py Px Ty Tx P0 y P0 x 2.8006 My s My w θs CHAPTER 14. However.6577 1.4. 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. There are researchers which suggest that the numerical work is the solution. Consider the same values as previous analysis. Thus.3226 13.18192 P4 P2 P0 P1 P4 1.9035 85. the oblique shock with upstream Mach number.9316 26.6038 To reduce the pressure ratio the deflection angle has to be reduced (remember that at weak weak shock almost no pressure change).

if the answer is negative. Of course. for example. deflection angle. in that analysis it has to be remembered that when the flow becomes subsonic the equation changes from hyperbolic to an elliptic equation. the stable and unstable can be recognized. The similar analysis can be applied to strong shock.21: Typical examples of unstable and stable situations the situation will return to its original state then it is referred to as the stable situation. 14. Yet.6. In this example only the weak solution is explained. Given M1 and the deflection angle. The perpendicular component of the downstream Mach number. On the other hand. When D is equal to zero. As it results. then the situation is referred to as unstable. If the answer turns out to be that Fig. the situation moves away from the expected solution. The only difference is that here. for large deflection angle. the shock angle. M2 could be larger or smaller then 1. 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. is a ball shown in the Figure (14. a special case is created because the weak and strong solutions are equal (for large deflection angle). In the same manner. there are more than one parameter that can be changed. 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. 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.21). Unstable Stable .14. and three is strong shock. When D is positive there is no physical solution and only normal shock exist. two is weak shock. This change complicates the explanation and is omitted in this section. and upstream Mach number. M1 n is always smaller than 1.6 Appendix: Oblique Shock Stability Analysis The stability analysis is an analysis which answers the question of what happens if for some reason. the discussion is more complicated but similar analysis can be applied to the strong shock. When D > 0. APPENDIX: OBLIQUE SHOCK STABILITY ANALYSIS 263 downstream Mach number. the analysis for the oblique shock wave is carried out. Instinctively. 14.

The change is determined from the solution provided before or from the approximation (14. There isn’t any known experimental evidence to show that flow is unstable for δ = 0. Thus. the deflection angle becomes positive which increases the pressure at the wall. Yet. The pressure difference at the wall becomes a negative increment which tends to pull the shock angle to the opposite direction. OBLIQUE-SHOCK The change in the in∆θ + clination angle results ∆θ − in a different upstream Mach number and ∆δ − a different pressure. ∆θ = k+1 ∆δ 2 (14. On the other hand.56) can be applied for either positive. ∆δ + to maintain the same direction stream lines. ∆θ + or negative ∆θ− values. has to be in the opposite direction of the change of the shock angle. The opposite happens when the deflection increment becomes negative. the shock wave is unstable according to this analysis to one direction but stable to the other direction. the weak shock is stable. it must be pointed out that it doesn’t mean that the flow is unstable but rather that the model is incorrect.52). 14.56) Equation (14. Please note that this analysis doesn’t apply to the case of the close proximity of the δ = 0.22: The schematic of stability analysis for oblique shock. . In fact.264 CHAPTER 14. the virtual change in the deflection angle Fig.

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

dy .2) A Mach line results because of a small disturbance in the wall contour. Therefore. dx is In the same manner.6) (15. In this chapter. as it turns out.3: The schematic of the turning only. the Mach angle dν dy òĵÅCà Á increase and result in a change in the direction of the flow. 15. This direction change results in a change of the flow properties. 15. PRANDTL-MEYER FUNCTION 1 M1 − 1 µ = tan−1 √ (15. The Mach line is the chief line in the analysis because of the wall contour shape information propagates along this line.266 or CHAPTER 15.4) dν ∼ sin(dν ). this assumption will be examflow ined. The typical simplifications for geometrical functions are used: y Ma ch li ne These simplifications are the core reasons why the change occurs only in the perpendicular direction (dν << 1). is dy = (U + dU ) sin(dν ) = U dν The tan µ is the ratio of dy/dx (see Figure (15. no shock is created from many small positive angles.2 Geometrical Explanation x The change in the flow direction is assume ËÊÌ ÍµÎCËÊÌ to be result of the change in the tangential dx = dU cos(90 − µ) Æ Ç ÈÊÉ component.3)) tan µ = dx dU = dy U dν (15. The velocity compoÏ ÐÒÑÔÓCСÕOÖ ×4Ø Ï ÓWÙwÕÊÚcÐ nent in the direction of the Mach line is assumed to be constant to satisfy the assumption that the change is a result of the contour Fig. the total Mach numÛ Ü4Ý¡ÞÔßÊà á  ber increases.5) dx = (U + dU ) cos ν − U = dU (15. cos(dν ) ∼ 1 (15.3) . Later. This assumption. Hence. The reason that a “negative” angle is not applicable is that the coalescing of the small Mach wave which results in a shock wave. the flow direction will change to fit the wall. the velocity perpendicular to the flow. This Mach line is assumed to be a result of the positive angle. and it is assumed here to be isotropic for a positive angle. Once the contour is changed. a discussion on the relationship between the flow properties and the flow direction is presented. However. The change of the velocity in the flow direction. is close to reality.

4: The schematic of the coordinate based on the matheflow turns around a sinmatical description gle point. It must be recognized that here the cylindrical coordinates are advantageous because the Fig.7) (15. a simplified verr Ur line sion was derived based Front Mach on geometrical arguUθ line θ ments. a more rigorous explanation is provided.7) transforms it into √ M 2 − 1dM 2 dν = − 1 2 2M 2 1 + k− 2 M After integration of equation (15.2. In this section.6) and (15.11) (15. GEOMETRICAL EXPLANATION The ratio dU/U was shown to be dM 2 dU = 1 2 U 2M 2 1 + k− 2 M Combining equations (15.9) The constant can be chosen in a such a way that ν = 0 at M = 1.1 Alternative Approach to Governing Equations In the previous secback Mach tion.8) becomes ν (M ) = − k+1 tan−1 k−1 k−1 (M 2 − 1) + tan−1 k+1 (M 2 − 1) + constant 267 (15.2.8) (15.10) .12) ∂Ur Uθ ∂Ur Uθ 2 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r ρ ∂r ρ ∂r (15. 15. 15. For this coordinate system.15. the mass conservation can be written as ∂ (ρrUr ) ∂ (ρUθ ) + =0 ∂r ∂θ The momentum equations are expressed as Ur and Ur ∂Uθ Uθ ∂Uθ Uθ Ur 1 ∂P c2 ∂ρ + − =− =− ∂r r ∂θ r rρ ∂θ rρ ∂θ (15.

19) It is remarkable that the tangential velocity at every turn is at the speed of sound! It must be pointed out that the total velocity isn’t at the speed of sound. r.268 CHAPTER 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 (15.14) into equation (15. the derivative isn’t zero.16) results in ∂Uθ − Ur ∂θ = c2 Uθ Ur + ∂Uθ ∂θ (15.13) can be rearranged as transformed into − 1 Uθ Ur + ∂Uθ ∂θ = 1 ∂ρ ρ ∂θ (15. the mass equation is reduced to ρUr + ∂ (ρUθ ) =0 ∂θ (15.13) Equation (15. One has to remember that when r enters to the function. then all the derivatives with respect to the radius will vanish.16) from equation (15. like the first term in the mass equation.20) Ur + ∂Uθ ∂θ =0 (15. Hence. PRANDTL-MEYER FUNCTION If the assumption is that the flow isn’t a function of the radius. but only .15) (15.17) Uθ or Uθ 2 Ur + ∂Uθ ∂θ = c 2 Ur + ∂Uθ ∂θ (15.18) And an additional rearrangement results in c2 − U θ 2 From equation (15.19) it follows that Uθ = c (15.

22) which is the radial velocity transforms equation (15.27) incorporating the constant becomes Ur = 2h0 sin θ k−1 k+1 (15. these results are associated with the characteristic line. based on the definition of the Mach angle.22) In the literature. In fact.25) into ∂Ur 2 ∂θ After some rearrangement.24) Utilizing equation (15. ∞ and since Uθ = 0 leads to ∂Ur = Uθ ∂θ (15. equation (15.28) . the non–dimensionalization can be applied at this stage as well.27) Note that Ur must be positive.21) If r isn’t approaching infinity.3) under Uy is equal to the speed of sound.15) becomes Uθ r ∂Ur − Uθ ∂θ =0 (15.20) for the speed of sound and substituting equation (15. After some additional rearrangement. The energy equation for any point on a stream line is h(θ) + Uθ 2 + U r 2 = h0 2 (15.15.25) 1 Cp c2 R T = RT = R (k − 1) Cv k−1 (15.26) + Ur 2 = 2h0 (15.26) becomes k+1 k−1 ∂Ur ∂θ 2 k−1 + ∂Ur 2 ∂θ + Ur 2 2 = h0 (15. the component shown in Figure (15. GEOMETRICAL EXPLANATION 269 the tangential component. k . However. is c(θ )2 k h(θ) = Cp T = Cp and substituting this equality.23) results in c2 Uθ 2 + U r 2 + = h0 k−1 2 (15. The solution of the differential equation (15.2. equation (15. This analysis can be also applied to the same equation when they are normalized by Mach number.24).23) Enthalpy in perfect gas with a constant specific heat. into equation (15. M = 1. equation (15.

but the assumptions for the construction of these models are different. 15.31) What happens when the upstream Mach number is not 1? That is when the initial condition for the turning angle doesn’t start with M = 1 but is already at a different angle. has to match to the definition of the angle that is chosen here (θ = 0 when M = 1).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.34) (15.2.32) k+1 tan2 k−1 θ k−1 k+1 (15.30) Now utilizing the expression that was obtained for Ur and Uθ equations (15. The tangential velocity obtains the form Uθ = c = ∂Ur = ∂θ k−1 k+1 2 h0 cos θ k−1 k+1 (15.27) because sin2 θ +cos2 θ = 1. In the geometrical model.28) is chosen such that Ur (θ = 0) = 0. PRANDTL-MEYER FUNCTION which satisfies equation (15. The arbitrary constant in equation (15.270 CHAPTER 15. For this upstream Mach number (see Figure (15. The upstream Mach number is denoted in this segment as Mstarting .2)) tan ν = Mstarting 2 − 1 (15.35) These relationships are plotted in Figure (15.29) and (15. the assumption . so ν (M ) = θ(M ) − θ(Mstarting ) = k+1 tan−1 k−1 k−1 k+1 M2 − 1 − tan−1 M2 − 1 (15.6).33) The deflection angle ν .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.

The radial velocity is “fed” through the reduction of the density. Aside from its close proximity to turning point. 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 (15.5) 15.36) The maximum of the deflection point and the maximum turning point are only a function of the specific heat ratios. THE MAXIMUM TURNING ANGLE 271 is that the velocity change in the radial direction is zero. At a small Mach number. 15. In fact.37) The change in the deflection angle is calculated by . 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). the question of boundary layer arises. the maximum turning angle is much larger than the maximum deflection point because the process is isentropic. 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. some researchers recommend that. Normally. However. in many instances. At this point (r = 0) these models fail to satisfy the boundary conditions and something else happens there. the author recommends that this function be used everywhere beyond 2-4 the thickness of the boundary layer based on the upstream length. However. Prandtl–Meyer function becomes ν∞ = π 2 k+1 −1 k−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.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. the fact remains that there is a radial velocity at Ur (r = 0) = constant. it was assumed that radial velocity is only a function of θ. Thus. an increase in the Mach number can result in a very significant radial velocity. On top of the complication of the turning point. Regardless of the assumptions that were used in the construction of these models. (1 + ) the radial velocity is small . the sharp point should be replaced by a smoother transition.3.15. In this case. In the rigorous model.4 The Working Equations for the Prandtl-Meyer Function ν2 − ν1 = ν (M2 ) − ν (M1 ) (15. For example. mass balance is maintained by the reduction of the density.

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

15.6. FLAT BODY WITH AN ANGLE OF ATTACK

273

Prandtl-Meyer Angle
100

80

k=1.4

60 θ 40 20 0

1

2

3

4

5 6 Mach Number

7

8

9

10

Fri Jul 8 15:39:06 2005

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

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

15.6

Flat Body with an Angle of Attack

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

Slip

plane

274

CHAPTER 15. PRANDTL-MEYER FUNCTION

15.7

Examples For Prandtl–Meyer Function

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

âäãæårçéè/êìë#í¡îyï ðòñôóéõyöh÷

¨
©

   ¥§¦   


øúùüûôýyþéÿ

 ¢¡¤£

Fig. 15.9: The schematic of Example 15.1

The Mach number is then

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

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

µ 52.6434

0.35603

0.74448

0.47822

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

µ 63.4620

0.12734

0.55497

0.22944

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

275

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

¢!#"$ % :(;*<>= ?A@ BDC0E>F

6§7 8
9

&('*)+ ,.-0/21 3¢4¤5

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

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

276

CHAPTER 15. PRANDTL-MEYER FUNCTION

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

µ 54.4623

0.28418

0.60365

0.47077

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

µ 55.5479

0.26190

0.58419

0.44831

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

15.8

Combination of the Oblique Shock and Isentropic Expansion

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

µ 73.1416

0.01506

0.37972

0.03965

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

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

µ 74.0528

0.01090

0.35248

0.03093

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

0.43534 3.1115 88.9313 20.3467

0.99676

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

3.1115 20.3467

1.1157

1.1066

0.99676

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

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

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

PRANDTL-MEYER FUNCTION .278 CHAPTER 15.

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

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

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

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

{1.. 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. ******************************************* \setlongtables \begin{longtable} {|D.4}|D.{1. You must use longtable style file and dcolumn style files.{1.{1. variableName.4}|D.4}|D. int isTex = yes.. isTex. variableName = MxV.0 .4}|} \caption{ ?? \label{?:tab:?}}\\ . isRange.4}|D. whatInfo = infoStandard .. variableValue). int isRange = no.A.2. Mx = 2. You can just can cut and paste it in your latex file. ******************************************* 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.. 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 . shock etc) infoTube print tube side info for (Fanno.{1..4}|D.{1..4}|D.makeTable(whatInfo.{1.. etc) including infoShock print shock sides info infoTubeShock print tube info shock main info infoTubeProfile the Mach number and pressure ratio profiles infoTubeShockLimits print tube limits with shock A To get the shock results in L TEX of Mx The following lines have to be inserted in the end of the main function. s.

6164 \\ A.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[-0.152& 0.000& 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^{*}} $} \\\hline \endhead 2.284 \hline \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \multicolumn{1}{|c|} \\\hline APPENDIX A.176& 2.pt}{0.5854& 3.3 Program listings Can be download from www.potto.org. COMPUTER PROGRAM {$\rule[-0.3608& \hline\end{longtable} & 1.773& 0.1in]{0. .

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

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

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