This action might not be possible to undo. Are you sure you want to continue?

Genick Bar–Meir, Ph. D. 2729 West Jarvis Ave Chicago, IL 60645-1335 email:barmeir at gmail.com

Copyright © 2010, 2009, 2008, 2007, and 2006 by Genick Bar-Meir See the ﬁle copying.fdl or copyright.tex for copying conditions. Version (0.3.0.0 October 25, 2010)

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

from The Metalogicon by John in 1159

CONTENTS

Nomenclature 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 How to contribute to this book . . . . . . . . . . . . . . . . . Credits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Steven from artofproblemsolving.com . . . . . . . . . . . Dan H. Olson . . . . . . . . . . . . . . . . . . . . . . . . Richard Hackbarth . . . . . . . . . . . . . . . . . . . . . . John Herbolenes . . . . . . . . . . . . . . . . . . . . . . . Eliezer Bar-Meir . . . . . . . . . . . . . . . . . . . . . . Henry Schoumertate . . . . . . . . . . . . . . . . . . . . . Your name here . . . . . . . . . . . . . . . . . . . . . . . Typo corrections and other ”minor” contributions . . . . . Version 0.1.8 August 6, 2008 . . . . . . . . . . . . . . . . . . . pages 189 size 2.6M . . . . . . . . . . . . . . . . . . . . . Version 0.1 April 22, 2008 . . . . . . . . . . . . . . . . . . . . .

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

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

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

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

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

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

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

xv xxiii xxiv xxv xxv xxvi xxviii xxviii xxix xxix xxix xxix xxx xxxi xxxi xxxi xxxii xxxii xxxii xxxii xxxii xxxii xxxiii xliii xliii xliii

iii

iv

CONTENTS pages 151 size 1.3M . . . . . . . . . . . . . . . . . . . . . . . . . . . . xliii Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xlix Open Channel Flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . xlix

1 Introduction 1.1 What is Fluid Mechanics? . . . . . 1.2 Brief History . . . . . . . . . . . . 1.3 Kinds of Fluids . . . . . . . . . . . 1.4 Shear Stress . . . . . . . . . . . . 1.5 Viscosity . . . . . . . . . . . . . . 1.5.1 General . . . . . . . . . . 1.5.2 Non–Newtonian Fluids . . 1.5.3 Kinematic Viscosity . . . . 1.5.4 Estimation of The Viscosity 1.5.5 Bulk Modulus . . . . . . . 1.6 Surface Tension . . . . . . . . . . 1.6.1 Wetting of Surfaces . . . .

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

1 1 3 5 6 9 9 10 11 12 20 22 24 33 33 41 41 43 43 44 44 44 45 47 51 52 53 54 55 59 59 59 61 61 62 66 70 74

2 Review of Thermodynamics 2.1 Basic Deﬁnitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Review of Mechanics 3.1 Kinematics of of Point Body . . . . . 3.2 Center of Mass . . . . . . . . . . . . 3.2.1 Actual Center of Mass . . . . 3.2.2 Aproximate Center of Area . . 3.3 Moment of Inertia . . . . . . . . . . . 3.3.1 Moment of Inertia for Mass . . 3.3.2 Moment of Inertia for Area . . 3.3.3 Examples of Moment of Inertia 3.3.4 Product of Inertia . . . . . . . 3.3.5 Principal Axes of Inertia . . . . 3.4 Newton’s Laws of Motion . . . . . . . 3.5 Angular Momentum and Torque . . . 3.5.1 Tables of geometries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4 Fluids Statics 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 The Hydrostatic Equation . . . . . . . . . . . . . . . . . . . 4.3 Pressure and Density in a Gravitational Field . . . . . . . . . 4.3.1 Constant Density in Gravitational Field . . . . . . . . 4.3.2 Pressure Measurement . . . . . . . . . . . . . . . . 4.3.3 Varying Density in a Gravity Field . . . . . . . . . . 4.3.4 The Pressure Eﬀects Due To Temperature Variations 4.3.5 Gravity Variations Eﬀects on Pressure and Density .

CONTENTS 4.4 4.3.6 Liquid Phase . . . . . . . . . . . . . . . Fluid in a Accelerated System . . . . . . . . . . 4.4.1 Fluid in a Linearly Accelerated System . 4.4.2 Angular Acceleration Systems: Constant Fluid Forces on Surfaces . . . . . . . . . . . . . 4.5.1 Fluid Forces on Straight Surfaces . . . . 4.5.2 Forces on Curved Surfaces . . . . . . . Buoyancy and Stability . . . . . . . . . . . . . 4.6.1 Stability . . . . . . . . . . . . . . . . . 4.6.2 Surface Tension . . . . . . . . . . . . . Rayleigh–Taylor Instability . . . . . . . . . . . . . . . . . . . . . . . . . . . Density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v . 76 . 77 . 77 . 79 . 81 . 81 . 90 . 97 . 104 . 114 . 115

4.5

4.6

4.7

I

Integral Analysis

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

121

123 123 124 125 127 127 134 136 142 145 147 147 147 148 149 149 150 154 157 158 165 166 169 171 171 183 184 185

5 Mass Conservation 5.1 Introduction . . . . . . . . . . . . . . . . . . . . 5.2 Control Volume . . . . . . . . . . . . . . . . . . 5.3 Continuity Equation . . . . . . . . . . . . . . . . 5.3.1 Non Deformable Control Volume . . . . . 5.3.2 Constant Density Fluids . . . . . . . . . . 5.4 Reynolds Transport Theorem . . . . . . . . . . . 5.5 Examples For Mass Conservation . . . . . . . . . 5.6 The Details Picture – Velocity Area Relationship 5.7 More Examples for Mass Conservation . . . . . .

6 Momentum Conservation 6.1 Momentum Governing Equation . . . . . . . . . . . . . 6.1.1 Introduction to Continuous . . . . . . . . . . . . 6.1.2 External Forces . . . . . . . . . . . . . . . . . . 6.1.3 Momentum Governing Equation . . . . . . . . . 6.1.4 Momentum Equation in Acceleration System . . 6.1.5 Momentum For Steady State and Uniform Flow . 6.2 Momentum Equation Application . . . . . . . . . . . . . 6.2.1 Momentum for Unsteady State and Uniform Flow 6.2.2 Momentum Application to Unsteady State . . . . 6.3 Conservation Moment Of Momentum . . . . . . . . . . 6.4 More Examples on Momentum Conservation . . . . . . . 6.4.1 Qualitative Questions . . . . . . . . . . . . . . . 7 Energy Conservation 7.1 The First Law of Thermodynamics . . . 7.2 Limitation of Integral Approach . . . . . 7.3 Approximation of Energy Equation . . . 7.3.1 Energy Equation in Steady State . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vi 7.4 7.3.2 Energy 7.4.1 7.4.2 7.4.3 7.4.4

CONTENTS Energy Equation in Frictionless Flow and Steady State . Equation in Accelerated System . . . . . . . . . . . . . Energy in Linear Acceleration Coordinate . . . . . . . . Linear Accelerated System . . . . . . . . . . . . . . . . Energy Equation in Rotating Coordinate System . . . . . Energy Equation in Accelerated Coordinate with Uniform . . . . . . . . . . . . . . . Flow . . . . . 185 187 187 188 188 190

II

Diﬀerential Analysis

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

191

193 193 194 197 199 204 204 205 206 210 220 220 224 234 239 239 239 240 241 242 243 247 248 251 252 254 255 256 258 259 261 262 269

8 Diﬀerential Analysis 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.2 Mass Conservation . . . . . . . . . . . . . . . . . . . . . . . . . 8.2.1 Mass Conservation Examples . . . . . . . . . . . . . . . 8.2.2 Simpliﬁed Continuity Equation . . . . . . . . . . . . . . 8.3 Conservation of General Quantity . . . . . . . . . . . . . . . . . 8.3.1 Generalization of Mathematical Approach for Derivations 8.3.2 Examples of Generalized of Quantities . . . . . . . . . . 8.4 Momentum Conservation . . . . . . . . . . . . . . . . . . . . . 8.5 Derivations of the Momentum Equation . . . . . . . . . . . . . 8.6 Boundary Conditions and Driving Forces . . . . . . . . . . . . . 8.6.1 Boundary Conditions Categories . . . . . . . . . . . . . 8.7 Examples for Diﬀerential Equation (Navier-Stokes) . . . . . . . 8.7.1 Interfacial Instability . . . . . . . . . . . . . . . . . . . . 9 Multi–Phase Flow 9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2 History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3 What to Expect From This Chapter . . . . . . . . . . . . . 9.4 Kind of Multi-Phase Flow . . . . . . . . . . . . . . . . . . . 9.5 Classiﬁcation of Liquid-Liquid Flow Regimes . . . . . . . . . 9.5.1 Co–Current Flow . . . . . . . . . . . . . . . . . . . 9.6 Multi–Phase Flow Variables Deﬁnitions . . . . . . . . . . . . 9.6.1 Multi–Phase Averaged Variables Deﬁnitions . . . . . 9.7 Homogeneous Models . . . . . . . . . . . . . . . . . . . . . 9.7.1 Pressure Loss Components . . . . . . . . . . . . . . 9.7.2 Lockhart Martinelli Model . . . . . . . . . . . . . . 9.8 Solid–Liquid Flow . . . . . . . . . . . . . . . . . . . . . . . 9.8.1 Solid Particles with Heavier Density ρS > ρL . . . . 9.8.2 Solid With Lighter Density ρS < ρ and With Gravity 9.9 Counter–Current Flow . . . . . . . . . . . . . . . . . . . . . 9.9.1 Horizontal Counter–Current Flow . . . . . . . . . . . 9.9.2 Flooding and Reversal Flow . . . . . . . . . . . . . . 9.10 Multi–Phase Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CONTENTS A Mathematics For Fluid Mechanics A.1 Vectors . . . . . . . . . . . . . . . . . . . . . . . . . . A.1.1 Vector Algebra . . . . . . . . . . . . . . . . . . A.1.2 Diﬀerential Operators of Vectors . . . . . . . . A.1.3 Diﬀerentiation of the Vector Operations . . . . A.2 Ordinary Diﬀerential Equations (ODE) . . . . . . . . . A.2.1 First Order Diﬀerential Equations . . . . . . . . A.2.2 Variables Separation or Segregation . . . . . . A.2.3 Non–Linear Equations . . . . . . . . . . . . . . A.2.4 Second Order Diﬀerential Equations . . . . . . A.2.5 Non–Linear Second Order Equations . . . . . . A.2.6 Third Order Diﬀerential Equation . . . . . . . A.2.7 Forth and Higher Order ODE . . . . . . . . . . A.2.8 A general Form of the Homogeneous Equation A.3 Partial Diﬀerential Equations . . . . . . . . . . . . . . A.3.1 First-order equations . . . . . . . . . . . . . . A.4 Trigonometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii 271 271 272 274 276 282 282 283 285 288 290 293 295 297 297 298 299

Index 301 Subjects Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Authors Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304

viii CONTENTS .

. . . . . . . . . .18 3. The schematic to explain the summation of moment of inertia. .11 1. . . . . .13 1. . . . . . . . . . . . .10 1. . The raising height as a function of the radii. . . . . . . . . . . . .12 1. . . . . . . . . . . . . . . Cylinder with an element for calculation moment of inertia .5 3. .9 1. . Description of rectangular in x–y plane. Reduced viscosity as function of the reduced temperature. . . .2 3. Reduced viscosity as function of the reduced temperature.5 1. . . . . The shear stress as a function of the shear rate. . . . . . . . . . . . . . . . . . . . Nitrogen and Argon viscosity. . . . . . . . . . . . . . . The diﬀerence of power ﬂuids. . . . . . . .2 1. . . . . . Surface tension control volume analysis. . . . . . . . . . . . .1 1.6 1. . . . . . . A square element for the calculations of inertia. . . . . . . . . . . . . . . .16 1. . . . . . 2 6 6 7 9 10 10 11 12 15 17 18 22 24 25 26 29 29 42 43 44 45 46 47 47 48 ix . . .8 Diagram to explain part of relationships of ﬂuid mechanics branches. . . . . . . .LIST OF FIGURES 1. . The raising height as a function of the radius. . . Density as a function of the size of sample.7 1. Description of liquid surface. . . . . . . . . . . . . . . . . . . . . .3 3. . . . . . The schematic that explains the summation of moment of inertia. . . . . . Air viscosity as a function of the temperature. . . . . . . . . . . . . . . The deformation of ﬂuid due to shear stress. . . . . Water viscosity as a function temperature. . . . . . . . . . . . . . Description of the extinguish nozzle . . . . . . . . . . . Schematics to describe the shear stress in ﬂuid mechanics. . . .7 3. . . . Thin body center of mass/area schematic. . . .6 3. . . . . . . . . . . . . Description of how the center of mass is calculated . . .17 1. . . . . . . . .3 1. . .4 1. . . . . Description of wetting and non–wetting ﬂuids. . . . . . . . . . . .8 1.4 3. . . . . . . . . .14 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 3. . . . . . Forces in Contact angle. . . . . . . . . . .15 1. . . . Liquid metals viscosity as a function of the temperature. . . . .

. . The forces on curved area. . . . . . . . . . Schematic of ﬂoating bodies . . . . . . . . . . . Inverted manometer . . . . . . . . Polynomial shape dam description . . . .37 4. .17 4. . . . . . . . . . . . . .5 . . . . . . .13 4. . . . . . . . . .35 4. . . . . . . . . The ﬂoating forces on Immersed Cylinder . . . . . . . . . . . . . . .7 . . . . . . . . . . . . . . . . . Schematic of gas measurement utilizing the “U” tube. .15 4. . .x 3. .12 3. . . . . . . . . . . . . . . . Two adjoin layers for stability analysis . . . . . . . . . . . .14 4. . Forces diagram of cart sliding on inclined plane .12 4. . . . . . . . . . . . . . . Schematic of sensitive measurement device. . . . . . . . . . . . . . . . . . . . . . . . . . . . .1 4. .26 4. . . . . . . . . . . . . . . . . . . . . . . Pressure lines in a static constant density ﬂuid. . . . . . . . . . .36 4. . The general forces acting on non symmetrical straight area. center of .23 4.8 4. . . . . . . . . . . . . . . . . . . . . . . . Rectangular area under pressure. . . . .33 4. . . . . . . . . .38 The ratio of the moment of inertia 2D to 3D. . . . . . . . . . . . A cart slide on inclined plane . . . . Cubic body dimensions for stability analysis. . . . . . . . . . . . . Schematic of Net Force on ﬂoating body. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 4. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The varying gravity eﬀects on density and pressure . . . . . . . . . . . . . The general forces acting on non symmetrical straight area . . . area . . . . . Schematic of ﬂoating cubic . . . . . . . . . . . . . . . . . . . . . . . . . . . A schematic to explain the atmospheric pressure measurement . . . . . . . . . .30 4. . . . .3 4. . . . . . . .19 4. . Schematic to explain the angular angle . . . . . . . . . . . .moment of inertia and Triangle for example 3. . . . .34 4. . . . . . . . . . . . . . . . . . . Schematic angular angle to explain example 4. Area above the dam arc subtract triangle. . Inclined manometer . . . . . . . . . . . Description of parabola . . . . . . . . . . . . . . . .21 4. . . . The eﬀects of multi layers density on static forces .22 4.11 3. Schematic of a thin wall ﬂoating body . The diﬀerence between the slop and the direction angle. . Moment on arc element around Point “O” . . . Area above the dam arc calculation for the center. . . . . . . The general forces acting on submerged area . . . . . . . . . . . . . . Product of inertia for triangle . . Schematic of submerged area. . . . . 48 49 49 51 52 59 62 62 63 64 66 66 69 72 74 77 78 79 79 80 81 82 83 85 86 89 90 91 92 92 93 94 95 96 97 98 99 104 105 105 107 107 108 Description of a ﬂuid element in accelerated system. . . .6 4. . . . . . . . . .29 4. . . . . . . . . . . . . . . .20 4. .5 4. . . .2 4. .32 4. . . .27 4. . . .25 4. . . . . . . . . . . . . . .28 4. . . . . . . .10 4. .9 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10 3. Moment of inertia for rectangular . .31 4. The eﬀective gravity is for accelerated cart . . Hydrostatic pressure when there is compressibility in the liquid phase.24 4. . . . . . Schematic of Immersed Cylinder . . . .18 4. The maximum height reverse as a function of density ratio. . . . . . . . . . . . . Circular shape Dam . . .13 4. . . . . .16 4. . . . . Stability analysis of ﬂoating body. LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . .11 4. . . . . . . . . . . . . . . . . . . . . . . . . . Stability of cubic body inﬁnity long. . . . . .4 4.7 4. .

. The impeller of the centrifugal pump and the velocities diagram Nozzle schematics water rocket . . . . . . . .39 4. . . . . . .46 4. . . . . . . . . . . . . . . . . . . .LIST OF FIGURES 4. . . . . . . . .4 8. . Calculations of GM for abrupt shape body. . . . . . .14 . . . . . . . . . . . . . . . . . . . . xi 109 110 112 113 115 116 117 118 119 123 124 125 126 129 132 137 142 143 144 145 146 148 151 153 155 156 157 160 161 165 167 170 172 174 175 184 194 196 198 199 207 The explaination for the direction relative to surface . . . . . .8 6. .12 6. . . . . . Schematic of a tank seating on wheels . . . . . . . . . . Discharge from a Large Container . . . . . . . . .9 6. . . . . . . . . . . . . . . . . Piston control volume . . . .7 6. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .41 4. . . . . . . . . . . . . . . . .5 Stability of two triangles put tougher . . . . .45 4. . . . .44 4. . . . . . . . . . . Schematics of area impinged by a jet . . . . . . . . .43 4. .4 5. . Boundary Layer control mass . . . . . . . .10 5. . A heavy needle is ﬂoating on a liquid. . . . . . . . Flow out of un symmetrical tank . .2 7. . . .1 6. . . . . A rocket with a moving control volume . . . . . . . . . . . . . . . . . . . . . . Control volume usage to calculate local averaged velocity Control volume and system in the motion . . . .5 6. . . . . . . . .1 5. . . . . . . Schematics of velocities at the interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . Toy Sled pushed by the liquid jet . . . . . . . . Mass ﬂow due to temperature diﬀerence . . . . . . . . . . The eﬀects of liquid movement on the GM . . . .42 4. . Propeller schematic to explain the change of momentum .6 5. Mass ﬂow in coating process . . . . . . . . Circular cross section for ﬁnding Ux . . . . . . . . . .40 4. . . . . . . . Height of the liquid for example 5. . . . . . . . . . . . . . . . .2 5. . . . . . . . .11 5.3 6. . . . . .10 6. . . Velocity for a circular shape . . . . . . . . . . . Description of depression to explain the Rayleigh–Taylor instability Description of depression to explain the instability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 8. . . . . . . . . . Nozzle schematic for forces calculations . The work on the control volume . . . . Control volume and system in motion . . . .8 5. . . . . . . . . . . .6 6. . . .1 8. . . . . . . . . . . . . . . . . .2 6. . Stress diagram on a tetrahedron shape . . . . .9 5. . . . . . . . . . . . .4 . . . . . . . . .5 5.47 5. . . . . . .4 8. . . . . . . . . Filling of the bucket and choices of the control volumes . . .1 7. Three liquids layers under rotation . . . The cross section of the interface for max liquid. . . . . . . . . . . . . . . Boat for example 5. . . . . . . . . . . . . . . .3 8. . . .7 5. . . . . . . . . . . . .4 6. . . . . . . . . . . . . . . . . .3 7. . . . . . . . . . . . . .11 7. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The mass balance on the inﬁnitesimal control volume The mass conservation in cylindrical coordinates . . . . . . . . . A new control volume to ﬁnd the velocity in discharge tank . . . . . . . . Schematics of ﬂow in a pipe with varying density . .3 5. . . . . . . . . . . . . . . Kinetic Energy and Averaged Velocity Flow in an oscillating manometer . . . . . . . . . . Measurement of GM of ﬂoating body. . . . . . . . . .

. . . . . . . . . . .9 8. . .4 9. Counter–current ﬂow in a can. . . . . . . . . . . . . . . . Vector in Cartesian coordinates system . . . . . . . . . . . . . . . . The ﬂow patterns in solid-liquid ﬂow. . . A ﬂow map to explain the horizontal counter–current ﬂow.3 9. . . . . . . . . .5 A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . (a) Deformations of the isosceles triangular . . . . . . . .8 9. . . . .1 9. . . .13 8. . . . . Deformations of diﬀerent rectangles . . . . . . . . . . . . . . . . . . . . .7 Diﬀerent ﬁelds of multi phase ﬂow. .13 9. . . . . . . . . . . . . . .5 9. . . . . . . . . . . . . . . . . . . .15 9. . .8 8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 8. . One dimensional ﬂow with shear between plates . . . . The general Orthogonal with unit vectors . . . . . .20 9. . . . . . . . . . . . . Spherical Coordinate System . . .12 8. . . . . . . . .6 9. . . . . . . . . . . .3 A. .14 9. . . . . . Counter–ﬂow in vertical tubes map. . . . . . . . . . .4 A. The terminal velocity that left the solid particles. . . . . . (b) Deformations of the straight angle triangle . . . . . . . . . . . . . . .1 A. . . . . . . . . . . . Liquid ﬂow due to gravity . . . . . . . . . .9 9. . . . . . . . . . . . . Parabolic coordinates by user WillowW using Blender The tringle angles sides . . . . .17 8. . . . . .11 8. .10 9. Flow of Liquid between concentric cylinders . Modiﬁed Mandhane map for ﬂow regime in horizontal tubes. . . . . . . . . . . . . .19 8. . . . . . . 1–Dimensional free surface . A diagram to explain the ﬂood in a two dimension geometry. Mass ﬂow due to temperature diﬀerence . . . . . . . . . . . . . The control volume in pipe ﬂow . . . . . . . . Plug ﬂow in horizontal tubes with the liquids ﬂow is faster. . . . 208 210 212 213 215 215 215 216 221 223 224 224 226 227 230 232 234 241 243 244 244 245 246 247 257 258 259 260 260 261 262 262 268 271 272 278 279 280 281 299 8. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . General forces diagram to calculated the in a two dimension geometry. . . . . . . . . . . .16 A. . . The right hand rule . . Linear strain of the element . . . . . . . . . . . . . .7 9. . . . . . Flow in kerosene lamp . . . . . . . . . Gas and liquid in Flow in verstical tube against the gravity. . . . . .2 9. . . . . . . . . . . . . .12 9.18 8. . . . . . . . . . . Flow between two plates top moving . . .7 8. . . . . . . . . . . . . . . . . . . . . . . . . . . . Cylindrical Coordinate System . . . . . . . .10 LIST OF FIGURES Diagram to analysis the shear stress tensor . . .14 8. . . .16 8. . . . . . . . . . . . . . . . . . . . .6 A. . . . . . The shear stress at diﬀerent surfaces . . . . . Image of counter-current ﬂow in liquid–gas/solid–gas conﬁgurations. . . . . . . . . . . . . . . . . . . Flow driven by surface tension . . . . . . . . . . . . . . A dimensional vertical ﬂow map low gravity against gravity. . . . . . Control volume at t and t + dt under continuous angle deformation Shear stress at two coordinates in 45◦ orientations . . . . . . . . . Kind of Stratiﬁed ﬂow in horizontal tubes. . . .11 9. . . . . . Stratiﬁed ﬂow in horizontal tubes when the liquids ﬂow is very slow. . .2 A. . . . .6 8. . . . .xii 8. . . . . . . . . . Flood in vertical pipe. .

. .7 2. . . . . .1 3. . . Viscosity of selected gases . . . Moment of inertia for various plane surfaces . . . . Moments of Inertia full shape. . .LIST OF TABLES 1 1. . . . . . . . . . . . . . . Properties at the critical stage . . . . . .1 Orthogonal coordinates systems (under construction please ignore) . . . . . . . . . . . . . . . . . . . . . . . . . . . Viscosity of selected liquids . .2 1. . . . The contact angle for air/water with selected materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . The surface tension for selected materials. . . . . . . . . . .1 1. . . . . . . . . . . . . . . . . . . . . . .4 1. . . Sutherland’s equation coeﬃcients . . . . . . . . . . . . . .2 Books Under Potto Project . . . . . . 282 xiii . . . . . . .7 1. . . . .3 1. . . . . Bulk modulus for selected materials . . . . . . . . . . . . continue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5 1. . . . . . . . . . . . . . . . . . . .1 3. . . . . . . . . A. . xl 13 13 14 15 20 25 31 32 38 56 57 Properties of Various Ideal Gases [300K] . . . . . . . . . . . .6 1. .

xiv LIST OF TABLES .

26).0). page 34 Internal Energy per unit mass. see equation (6. page 12 reference viscosity at reference temperature. see equation (9. see equation (2. see equation (5.9). page 59 Body force. see equation (2. see equation (4. page 35 subscribe for control volume.23). page 148 Units length. see equation (2. page 33 M µ µ0 F ext U Ξ A a Bf c. see equation (6. page 90 The acceleration of object or system.22). page 37 Internal energy. page 149 The velocity taken with the direction. see equation (2.7). see equation (2. Cp Cv EU Eu Angular Momentum. see equation (1.0).17). see equation (2.3). see equation (6.1). page 124 Speciﬁc pressure heat. page 34 xv .121). page 37 The shear stress Tenser. Ti0 . see equation (4. page 147 Martinelli parameter. see equation (2. see equation (1. page 37 Speciﬁc volume heat.v.. page 255 The area of surface.38).11). page 12 External forces by non–ﬂuids means.17).NOMENCLATURE ¯ R τ Universal gas constant.6).1). page 165 viscosity at input temperature T. see equation (6.43).

24). page 12 velocity . page 172 Angular momentum.17). page 34 Speciﬁc gas constant. page 36 the ratio of the speciﬁc heats.13). see equation (5. page 34 The gravitation constant. page 54 LIST OF TABLES Patmos Atmospheric Pressure. page 59 Enthalpy. see equation (2.1.40). page 38 Entropy of the system.14). page 34 The work done by the system between state 1 and state 2. page 54 reference temperature in degrees Kelvin. page 12 input temperature in degrees Kelvin. page 36 Speciﬁc enthalpy.4). see equation (2. page 124 . page 12 Torque.89). page 75 general Body force.6). see equation (1.18). see equation (2.27). see equation (4.17). see equation (3. see equation (2.xvi Ei G gG H h k kT L System energy at state i. see equation (4.0). page 34 The energy transfered to the system between state 1 and state 2. see equation (4.6).2).3).2). see equation (4. page 61 Subscribe says. see equation (3. see equation (2.42). page 37 Fluid thermal conductivity. see equation (2. page 36 Suth is Sutherland’s constant and it is presented in the Table 1. see equation (1. page 34 Work per unit mass.2). see equation (2.18). see equation (2. see equation (2. page 83 q Q12 R S Suth Tτ Ti0 Tin U w W12 z says Energy per unit mass. see equation (1. page 34 the coordinate in z direction.17).0). see equation (7.66). see equation (2. see equation (2.

9.0. 2010 (3. Add example to dif chap on concentric cylinders poiseuille ﬂow. Correct the ideal gas relationship typo thanks to Michal Zadrozny. Add example. Add discussion about inclined manometer Improve many ﬁgures and equations in Static chapter. xvii .3 M 354 pages) Change the emphasis equations to new style in Static chapter.2.0 Oct 24.3 M 344 pages) Change the emphasis equations to new style in Thermo chapter.1 Oct 11. change to the new empheq format and improve cylinder ﬁgure. Improve english in statice and moslty in diﬀerential analysis chapter. Minor correction to to the wording in page 11 viscosity density issue (thanks to Prashant Balan). Add example of falling liquid gravity as driving force in pressence of shear streas. 2010 (3. Add to the appendix the diﬀerentiation of vector operations.3.The Book Change Log Version 0. Version 0.

Add the macro protect to insert ﬁgure in lower right corner thanks to Steven from www.2. Additions to the mathematical appendix on variables separation in second order ode equations.xviii LIST OF TABLES Version 0.6 March 10.3 Jan 01.2. Thanks to Eliezer Bar-Meir Version 0. 2010 (2. Add the macro ekes to equations in examples thanks to Steven from www. English corrections.9 M 280 pages) add example to Mechanical Chapter and some spelling corrected. Version 0. Corrections to Static Chapter. Additions to the mathematical appendix on vector algebra.8 M 241 pages) The momentum conservation chapter was released. Some additions to mass conservation chapter on averaged velocity. 2010 (2. Version 0. Some additions to momentum conservation chapter.artofproblemsolving.4 March 01.3 M 338 pages) Initial release of the diﬀerential equations chapter. 2010 (2. Improve the emphasis macro for the important equation and useful equation. 2010 (3. Add the macro to improve emphases equation thanks to Steven from www.9 M 280 pages) The energy conservation chapter was released.com.9 Sep 20. Thanks to Eliezer Bar-Meir .artofproblemsolving.artofproblemsolving Add example about the the third component of the velocity.2.2. English corrections.

Add example on angular rotation to statics chapter. Add an example to mechanics chapter.5 M 197 pages) Continue ﬁxing the long titles issues.1 Sep 17. 2009 (2. Add Reynold’s Transform explanation. Version 0. 2009 (2.5 Nov 01. English corrections. 2008 (2.9 Dec 01. Correcting the gas properties tables (thanks to Heru and Micheal) Move the gas tables to common area to all the books.8 Aug 6.8a July 5.1. 2009 (2.1.4 M 189 pages) Add the chapter on introduction to muli–phase ﬂow . Version 0.8. Thanks to Eliezer Bar-Meir Version 0.1. Two open questions were released.5 M 203 pages) First true draft for the mass conservation. Add the ﬁrst draft of the temperature-velocity diagram to the Therm’s chapter. Improve the dwarﬁng macro to allow ﬂexibility with sub title. Add some examples to static chapter.6 M 183 pages) Fixing some long titles issues. 2009 (2.1.LIST OF TABLES xix Version 0. Add the open question concept. Version 0.1.6 M 219 pages) The mass conservation chapter was released.8.

1.6 Jun 30. 2008 . 2008 (1. Fix very minor issues (English) in the static chapter. 2008 (1. Improve the Index (thanks to Irene). LIST OF TABLES Version 0. Add the product of inertia to mechanics chapter.4 M 155 pages) Add the constant table list for the introduction chapter. Fix minor issues (English) in the introduction chapter.1.5 Jun 5. Version 0. 2008 (1.1a April 23. Version 0.4 M 149 pages) Add the introduction.3 M 151 pages) Fix the English in the introduction chapter.1 M 111 pages) Major English corrections for the three chapters. Add the Rayleigh–Taylor instability. 2008 (1. Improve the doChap scrip to break up the book to chapters.xx Again additional improvement to the index (thanks to Irene). Minor corrections for all three chapters.1. viscosity and other properties of ﬂuid.1. Version 0. Version 0.5a Jun 11.1 May 8. (thanks to Tousher). Remove the multiphase chapter (it is not for public consumption yet).

LIST OF TABLES xxi Version 0. The mechanics chapter was released. . The static chapter was released (the most extensive and detailed chapter). 2008 The Thermodynamics chapter was released.1a April 23.

xxii LIST OF TABLES .

either commercially or noncommercially. ”If you print more than 200 copies. MA 02110-1301 USA Everyone is permitted to copy and distribute verbatim copies of this license document.” Version 1. textbook. you are required to furnish the author with two (2) copies of the printed book. or other functional and useful document ”free” in the sense of freedom: to assure everyone the eﬀective freedom to copy and redistribute it. The Potto Project License doesn’t long apply to this document and associated docoments. 51 Franklin St.Notice of Copyright For This Document: This document is published under modiﬁed FDL. Fifth Floor. this License preserves for the author and xxiii . The change of the license is to prevent from situations that the author has to buy his own book. Boston. GNU Free Documentation License The modiﬁcation is that under section 3 “copying in quantity” should be add in the end. Inc.2. November 2002 Copyright ©2000. with or without modifying it. Preamble The purpose of this License is to make a manual. but changing it is not allowed.2002 Free Software Foundation. Secondarily.2001.

The Document may contain zero Invariant Sections. as Front-Cover Texts or Back-Cover Texts. below.) The relationship could be a matter of historical connection with the subject or with related matters. A Front-Cover Text may be at most 5 words. royalty-free license. and is addressed as ”you”. (Thus. that contains a notice placed by the copyright holder saying it can be distributed under the terms of this License. modify or distribute the work in a way requiring permission under copyright law. to use that work under the conditions stated herein. or with modiﬁcations and/or translated into another language. a Secondary Section may not explain any mathematics. The ”Cover Texts” are certain short passages of text that are listed. It complements the GNU General Public License. If the Document does not identify any Invariant Sections then there are none. and a Back-Cover Text may be at most 25 words. unlimited in duration. regardless of subject matter or whether it is published as a printed book. that is suitable for revising the document straightforwardly with generic text editors or (for . You accept the license if you copy. while not being considered responsible for modiﬁcations made by others. commercial. represented in a format whose speciﬁcation is available to the general public.xxiv LIST OF TABLES publisher a way to get credit for their work. ethical or political position regarding them. Any member of the public is a licensee. This License is a kind of ”copyleft”. The ”Invariant Sections” are certain Secondary Sections whose titles are designated. APPLICABILITY AND DEFINITIONS This License applies to any manual or other work. either copied verbatim. in the notice that says that the Document is released under this License. The ”Document”. A ”Transparent” copy of the Document means a machine-readable copy. 1. If a section does not ﬁt the above deﬁnition of Secondary then it is not allowed to be designated as Invariant. or of legal. if the Document is in part a textbook of mathematics. in any medium. in the notice that says that the Document is released under this License. But this License is not limited to software manuals. We recommend this License principally for works whose purpose is instruction or reference. it can be used for any textual work. We have designed this License in order to use it for manuals for free software. as being those of Invariant Sections. because free software needs free documentation: a free program should come with manuals providing the same freedoms that the software does. Such a notice grants a world-wide. philosophical. which is a copyleft license designed for free software. which means that derivative works of the document must themselves be free in the same sense. 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. A ”Modiﬁed Version” of the Document means any work containing the Document or a portion of it.

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

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

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

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

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

or some other combination of the three. and with the Back-Cover Texts being LIST. Front-Cover Texts and Back-Cover 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. 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. If your document contains nontrivial examples of program code. with no Invariant Sections.” line with this: with the Invariant Sections being LIST THEIR TITLES. you may choose any version ever published (not as a draft) by the Free Software Foundation. to permit their use in free software. replace the ”with. we recommend releasing these examples in parallel under your choice of free software license. no Front-Cover Texts.2 or any later version published by the Free Software Foundation. Permission is granted to copy. If you have Invariant Sections without Cover Texts. ADDENDUM: How to use this License for your documents To use this License in a document you have written. and no BackCover Texts. with the FrontCover Texts being LIST. If the Document does not specify a version number of this License. distribute and/or modify this document under the terms of the GNU Free Documentation License. merge those two alternatives to suit the situation. A copy of the license is included in the section entitled ”GNU Free Documentation License”. If you have Invariant Sections.Texts. . Version 1. such as the GNU General Public License.xxx LIST OF TABLES version” applies to it.

Minor contributions (typo corrections. Major contributions are listed by individual name with some detail on the nature of the contribution(s). Contact at: barmeir at gmail. Credits All entries have been arranged in alphabetical order of surname (hopefully. Dec. etc. time. just smaller in the sense of less text changed. date. Please understand that when I classify a contribution as ”minor.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.” it is in no way inferior to the eﬀort or value of a ”major” contribution. and resources to make this a better book! Date(s) of contribution(s): 1999 to present Nature of contribution: Original author. Any and all contributions are gratefully accepted. contact info. The only ”catch” is that credit must be given where credit is due. this book is open to revisions and expansions by any interested parties. I am indebted to all those who have given freely of their own knowledge. etc. 2009 xxxi .) are listed by name only for reasons of brevity.com Steven from artofproblemsolving.com Date(s) of contribution(s): June 2005. 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.

Henry Schoumertate Date(s) of contribution(s): Nov 2009 Nature of contribution: Discussion on the mathematics of Reynolds Transforms. John Herbolenes Date(s) of contribution(s): August 2009 Nature of contribution: Provide some example for the static chapter. Nature of contribution: In 2009 creating the exEq macro to have diﬀerent counter for example. describing how you contributed to the book.net . help on building the useful equation and important equation macros. Contact at: my email@provider. Dec 2009 Nature of contribution: Correct many English mistakes Mass. Olson Date(s) of contribution(s): April 2008 Nature of contribution: Some discussions about chapter on mechanics and correction of English. Dan H.xxxii LIST OF TABLES Nature of contribution: LaTeX formatting. Your name here Date(s) of contribution(s): Month and year of contribution Nature of contribution: Insert text here. Eliezer Bar-Meir Date(s) of contribution(s): Nov 2009. Richard Hackbarth Date(s) of contribution(s): April 2008 Nature of contribution: Some discussions about chapter on mechanics and correction of English. Nature of contribution: Correct many English mistakes Momentum.

Gupta. Tousher Yang April 2008.CREDITS xxxiii Typo corrections and other ”minor” contributions R.38) by Michal Zadrozny. review of statics and thermo chapters. Corretion to equation (2. (Nov 2010) Corretion to wording in viscosity density Prashant Balan. January 2008. help with the original img macro and other ( LaTeX issues). (Nov 2010) .

xxxiv LIST OF TABLES .

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

The author’s explanations on missing diameter and other issues in fanno ﬂow and ““naughty professor’s question”” are used in the industry. Engineers have constructed design that based on this conclusion. In his book “Basics of Fluid Mechanics”. He described and categorized the ﬁlling and evacuating of chamber by compressible ﬂuid in which he also found analytical solutions to cases where the working ﬂuid was ideal gas. He built a model to explain the ﬂooding problem (two phase ﬂow) based on the physics. practically from scratch. He also build analytical solution to several moving shock cases. The author spent years working on the sea (ships) as a engine sea oﬃcer but now the author prefers to remain on solid ground. Bar-Meir demonstrated that ﬂuids must have wavy surface when the materials ﬂow together. The common explanation to Prandtl–Meyer function shows that ﬂow can turn in a sharp corner. . he often feels clueless about computers and programing. the author just know to learn quickly. A past project of his was building a four stories house. All the previous models for the ﬂooding phenomenon did not have a physical explanation to the dryness. While he is known to look like he knows about many things. While he writes his programs and does other computer chores. He also constructed and explained many new categories for two ﬂow regimes. Bar-Meir demonstrated that common Prandtl–Meyer explanation violates the conservation of mass and therefor the turn must be around a ﬁnite radius.xxxvi LIST OF TABLES Meir discovered the analytical solution for oblique shock and showed that there is a quiet buﬀer between the oblique shock and Prandtl–Meyer. The author lives with his wife and three children.

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

For others. For some authors. While one can be as creative as possible.ex. In these cases. The following example explains this point: The army ant is a kind of carnivorous ant that lives and hunts in the tropics. the books on compressible ﬂow and die casting became the most popular books in their respective area. The student’s contributions can be done by adding a question and perhaps the solution. The collective material is much richer than any single person can create by himself. The answer varies from individual to individual. Experience has shown that in explaining to others any given subject.uk/bugclub/raiders. hunting animals that are even up to a hundred kilograms in weight. For others the social part is as simple as a desire to reduce the price of college textbooks. The book’s contributor’s names could be written by their sections. one also begins to better understand the material. there are always others who can see new aspects of or add to the material. 77:139. ”Army Ants: A Collective Intelligence. the reason is the pure fun of writing and organizing educational material. but also students who happened to be doing their homework. The social function can have at least two components.html) . One component is to come to know and socialize with many in the profession. especially for family members or relatives and those students lacking funds. It is not just for experts to contribute. It is hoped that because of the open nature of these books. the undersigned believes that personal intentions are appropriate and are the author’s/organizer’s private aﬀair. When an insect which is blind can be so powerful by networking. One can be successful when one solves as many problems as possible. Thus. To reach this possibility the collective book idea was created/adapted.xxxviii LIST OF TABLES related problems. Why would someone volunteer to be an author or organizer of such a book? This is the ﬁrst question the undersigned was asked. this method is expected to accelerate the creation of these high quality books. so can we in creating textbooks by this powerful tool. then that contributor will be the copyright holder of that speciﬁc section (even within question/answer sections).” American Scientist. Nigel R. Thus. they now have an opportunity to put their notes to use for others. The secret of the ants’ power lies in their collective intelligence. In a way. 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. 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. If a contributor of a section in such a book can be easily identiﬁed. the collective power of their networking creates an extremely powerful intelligence to carry out this attack4 . The author has not found any technique that is more useful for this purpose than practicing the solving of problems and exercises. For example. While a single ant is not intelligent enough to attack and hunt large prey. the writing of or contributing to this kind of books will serve as a social function. they will become the most popular books and the most read books in their respected ﬁeld.ac.. 1989 (see for information http://www. Whatever the reasons. These books are written in a similar manner to the open source software 4 see also in Franks. For some contributors/authors. the popularity of the books should be one of the incentives for potential contributors. contributing to these books will help one to understand the material better.

For example. The book on compressible ﬂow is also used by “young engineers and scientists” in NASA according to Dr. These data (tables. This may merely be the person who decides what will go into the book and what will not (gate keeper). Every month people from about 110 diﬀerent countries download these books.CREDITS xxxix process.000 copies a month. These books are intended to be “continuous” in the sense that there will be someone who will maintain and improve the books with time (the organizer(s)). Unlike a regular book. Thus. graphs and etc. chapters or sections can be added after the skeleton has been written. better discussions or better explanations are all welcome to these books. Nevertheless. these works will have a version number because they are alive and continuously evolving. In addition. The combined number downloads of these books is over half a million (December 2009) or in a rate of 20. some errors are possible and expected. It is also hoped that others will contribute to the question and answer sections in the book. Additionally.” In this process. . Thus. It is hoped that the books will be error-free. Potto books on compressible ﬂow and ﬂuid mechanics are used as the main textbook or as a reference book in several universities around the world. contributions can be made from any part of the world by those who wish to translate the book. other books contain data5 which can be typeset in A LTEX. Someone has to write the skeleton and hopefully others will add “ﬂesh and skin. the traditional role of author will be replaced by an organizer who will be the one to compile the book.) can be redone by anyone who has the time to do it. NASA Langley Research Center. These books contain innovative ideas which make some chapters the best in the world. These books should be considered more as a project than to ﬁt the traditional deﬁnition of “plain” books. In the last 5 years three textbooks have been constructed which are available for download. many algorithms for calculating Fanno ﬂow parameters which are not found in any other book. The undersigned of this document intends to be the organizer/author/coordinator of the projects in the following areas: 5 Data are not copyrighted. the chapters on Fanno ﬂow and Oblique shock contain many original ideas such as the full analytical solution to the oblique shock. Potto has auxiliary materials such as the gas dynamics tables (the largest compressible ﬂow tables collection in the world). while in other cases only the gate keeper. The books are used in more than 165 diﬀerent countries around the world. etc. Even if not complete. Farassat. But more than that. Gas Dynamics Calculator (Potto-GDC). the contributions to books can be done by many who are not experts. The organizer of the book in some instances will be the main author of the work.

advanced topics.0. The mature stage of a section is when all of the topics in the section are written.0.0 Based on Eckert Availability for Public Download Mechanics Open Channel Flow Statics Strength of Material Thermodynamics Two/Multi ﬂow phases NSY NSY early alpha NSY early alpha NSY ﬁrst chapter 0. in Gamma Stage is when all the chapters are written and some of the chapters are in a mature form.0 0.0.) are already presented.1. other deﬁnitions give merely a hint on the status. and special cases. The mature stage of a chapter is when all or nearly all the sections are in a mature stage and have a mature bibliography as well as numerous examples for every section.0. a process in which books .01 TelAviv’notes 0. But such a thing is hard to deﬁne and should be enough for this stage.0 NSY = Not Started Yet The meaning of the progress is as: The Alpha Stage is when some of the chapters are already in a rough draft. and the Advanced Stage is when all of the basic material is written and all that is left are aspects that are active. However.xl Table -1. LIST OF TABLES Project Name Compressible Flow Die Casting Dynamics Fluid Mechanics Heat Transfer Progress beta alpha NSY alpha NSY Remarks Version 0. ﬁgures. While some terms are deﬁned in a relatively clear fashion.0 0. The idea that a book can be created as a project has mushroomed from the open source software concept.1 0.0.0 0. but it has roots in the way science progresses.8. Books under development in Potto project.0. in Beta Stage is when all or almost all of the chapters have been written and are at least in a draft stage. traditionally books have been improved by the same author(s). and all of the examples and data (tables.0 0.0.1 0.3 0.0. etc.0.2 0.4.

and perhaps troﬀ. projects such as the Linux Documentation project demonstrated that books can be written as the cooperative eﬀort of many individuals. only LTEX. the Boundary Layer Theory originated6 by Hermann Schlichting but continues to this day. pieces of material and data can be used by diﬀerent books. many of whom volunteered to help. which include the actual writing of the text.CREDITS xli have a new version every a few years. have the ability to produce the quality that one expects for A these writings. These chores can be done independently from each other and by more than one individual. Writing a textbook is comprised of many aspects. writing examples. and writing the A LTEX macros7 which will put the text into an attractive format. Schlichting. 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. . especially LTEX. Again. 6 Originally authored by Dr. Word processors. and Microsoft Word software. creating diagrams and ﬁgures. Further. are not appropriate for these projects. because of the open nature of this project. Abiword. who passed way some years ago..e. i. such as OpenOﬃce. There are book(s) that have continued after their author passed away. But more A than that. The text processes. 7 One can only expect that open source and readable format will be used for this project. are the only ones which have a cross platform ability to produce macros and a uniform feel and quality. A new version is created every several years. However.

xlii LIST OF TABLES .

3M The topic of ﬂuid mechanics is common to several disciplines: mechanical engineering. In fact.1 April 22. This author was bothered by this explanation. the common material is presented and hopefully can be used by all. it was proven that this wavy interface is created due to the need to satisfy the continuous velocity and shear stress at the interface and not a disturbance.1. in this version. 2008 pages 151 size 1.Prologue For This Book Version 0. This explanation (on the wavy interface) demonstrates this characteristic of Potto project books. The number of downloads of this book results from this quality. The introduction to multi–phase is another example to this quality. it is also related to disciplines like industrial engineering. it is very satisfying for the author. This issue is related to renewal energy of extracting energy from brine solution (think about the Dead Sea. The common explanation to the wave existence was that there is always a disturbance which causes instability. and civil engineering. and electrical engineering. One can only admire the wonderful advances done by the xliii . While it is a hard work to discover and develop and bring this information to the students. Potto project books are characterized by high quality which marked by presentation of the new developments and clear explanations. Even in this early development stage. 2008 pages 189 size 2. While the emphasis is somewhat diﬀerent in this book. he spend time to study the wave phenomenon at the interface of open channel ﬂow.8 August 6. so much energy). Version 0.6M When this author was an undergraduate student. aerospace engineering. number of downloads per month is about 5000 copies. Now. chemical engineering.

. ideal ﬂow will be presented with the issues of added mass and added forces (under construction). The topic of ﬂuid mechanics was chosen just to ﬁll the introduction chapter to compressible ﬂow. When the Static Chapter was written. there is no known place in world that does not teach according to Eckert’s doctrine. anonymous Wikipedia user name EMBaero claimed that the material in the book is plagiarizing. During the writing it became apparent that it should be a book in its own right. These criticisms will not change the future or the success of the ideas in this work. aside from his research activity. wrote the book that brought a revolution in the education of the heat transfer. will be presented next (again under construction). The third book chapter is a review of mechanics. His book met strong criticism in which some called to “burn” his book. this author did not realize that so many new ideas will be inserted into this topic. Clearly. POTTO project books are open content textbooks so everyone are welcome to joint in. This book is the third book in the series of POTTO project books. what and when a certain model is suitable than other models. dimensional analysis will be present (again under construction). As traditional texts in this ﬁeld. Indeed. The book is organized into several chapters which. deals with a basic introduction to the ﬂuid properties and concepts (under construction). as a traditional textbook. For example.R. In writing the chapter on ﬂuid statics.xliv LIST OF TABLES previous geniuses who work in this ﬁeld. He wrote his book because he realized that the dimensional analysis utilized by him and his adviser (for the post doc). It is hoped that the other chapters will be as good this one. The next two chapters will deals with open channel ﬂow and gas dynamics. the study of heat transfer was without any dimensional analysis. not a traditional topic. the previous book. An introduction to multi–phase ﬂow. The classic issue of turbulence (and stability) will be presented. The second chapter deals with Thermodynamics. Eckert. multiphase is an advance topic. In this book it is hoped to insert. however. This book is written in the spirit of my adviser and mentor E. some minimal familiarity can be helpful for many engineers who have to deal with non pure single phase ﬂuid. It is assumed that the same kind of individual(s) who criticized Eckert’s work will criticize this work. Maybe that was the reason that he felt that is okay to plagiarize the book on Wikipedia. Today. it must be emphasized that this book is not expected to revolutionize the ﬁeld but change some of the way things are taught. this is the only reply. With all the above. must be taught in engineering classes. At this stage. Ernst Schmidt. and their colleagues. The next topic is statics.G. met its opposition. show me what is wrong”. One of the diﬀerence in this book is the insertion of the introduction to multiphase ﬂow. Eckert. As a wise person says “don’t tell me that it is wrong. there was a realization that it is the best chapter written on this topic. he just doesn’t know from where and what. on compressible ﬂow. However. Up to Egret’s book.

So. The ﬁgure in cover page was created by Genick Bar-Meir. The graphics were done by TGIF. the best graphic program that this author experienced so far. This author was inﬂuenced by Streeter and Wylie book which was his undergrad textbooks. and hope to ﬁnd a way to use gaspell. it was assumed that introductory book on ﬂuid mechanics should not contained many new ideas but should be modern in the material presentation. The structure of Hansen. This book was written using the vim editor for editing (sorry never was able to be comfortable with emacs). Of course. This statement really meant that the book is intent to be used by students to solve their exams and also used by practitioners when they search for solutions for practical problems. this book was written on Linux (Micro$oftLess book). The spell checking was done by ispell. Otherwise. The approach adapted in this book is practical. and is copyleft by him. Streeter and Wylie.” Later. In writing this book. After a while it seems that is easier to write a whole book than the two original planned chapters. multi–phase ﬂow chapter was written.How This Book Was Written This book started because I needed an introduction to the compressible ﬂow book. The chapters are not written in order. The book does not provide the old style graphical solution methods yet provides the graphical explanation of things. xlv . and more hands–on approach. and Shames books were adapted and used as a scaﬀolding for this book. There are numerous books on ﬂuid mechanics but none of which is open content. a program that currently cannot be used on new Linux systems. The presentation of some of the chapters is slightly diﬀerent from other books because the usability of the computers. this book avoids this kind of issues. The ﬁgures were done by gle. The ﬁrst 4 chapters were written ﬁrst because they were supposed to be modiﬁed and used as ﬂuid mechanics introduction in “Fundamentals of Compressible Flow. issue of proofs so and so are here only either to explain a point or have a solution of exams.

xlvi LIST OF TABLES .

9 At 8 To xlvii . without form. It is hoped that the book could be used as a reference book for people who have at least some basics knowledge of science areas such as calculus. You have to remember that this book is a work in progress. I hope to improve or to add to these areas in the near future. say. there is a limit on how much eﬀort I was able to put into the book and its organization. marked with a Mata mark. This book is written and maintained on a volunteer basis. increase your understanding of the many aspects of ﬂuid mechanics. which can be very useful for many. Basics of Fluid Mechanics. and there were words. The structure of this book is such that many of the chapters could be usable independently. This means that a lot of information is presented which is not necessary for everyone. And the Author there be words. etc.Preface "In the beginning. if you need information about. you can read just chapter (4). and secondly a reference manual only as a lucky coincidence. Nevertheless. And the Fingers of the Author the face of the keyboard. and emptiness was upon the face and files. the present. the explanations are not as good as if I had a few years to perfect them. These explanations have been marked as such and can be skipped. the power and glory of the mighty God. I believe professionals working in many engineering ﬁelds will beneﬁt from this information. statics’ equations.9 Reading everything will. I have tried to describe why the theories are the way they are. Like all volunteer work. the POTTO project was and void. rather than just listing “seven easy steps” for each task. of the bits moved upon said. physics. describes the fundamentals of ﬂuid mechanics phenomena for engineers and others. However. This book contains many worked examples. For example. I have left some issues which have unsatisfactory explanations in the book. this manuscript is ﬁrst and foremost a textbook. This book is only to explain his power." 8 . I hope this makes the book easier to use as a reference manual. Moreover. Let This book. naturally. This book is designed to replace all introductory textbook(s) or instructor’s notes for the ﬂuid mechanics in undergraduate classes for engineering/science students but also for technical peoples. the book is not well organized. due to the fact that English is my third language and time limitations.

graphic design.com”. more mathematics (or less mathematics). However. rewritten sections. Dr. G. You may contact me via Email at “barmeir@gmail. If you would like be “peer reviews” or critic to my new ideas please send me your comment(s). E. This material never went through a close content review. There are also notes in the margin. editing. Several people have helped me with this book. The symbol META was added to provide typographical conventions to blurb as needed. If you want to be involved in the editing. or proofreading. you can contact me at “barmeir@gmail. This is mostly for the author’s purposes and also for your amusement. In practice. R. literature review is always good. directly or indirectly. this process leaves a large room to blockage of novel ideas and plagiarism. While close content peer review and publication in a professional publication is excellent idea in theory. 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. . please drop me a line. ideas. Eckert. but those are solely for the author’s purposes. isn’t it?). this book contains material that never was published before (sorry cannot avoid it). more fundamental material. and assistance.xlviii LIST OF TABLES Furthermore. I would like to especially thank to my adviser. Incorrect language. While his comment looks like unpleasant reaction. If you have Internet e-mail access. other email that imply that someone will take care of this author aren’t appreciated. errors. it brought or cause the expansion of the explanation for the oblique shock. 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). graphic ability. I also would like to thank to Jannie McRotien (Open Channel Flow chapter) and Tousher Yang for their advices. whose work was the inspiration for this book. and material knowledge and a desire to provide open content textbooks and to improve them to join me in this project.com”. I am particularly interested in the best arrangement of the book. Over ten individuals wrote me about this letter. ideas for new areas to cover. ignore them please. Marshall wrote to this author that the author should review other people work before he write any thing new (well. LTEX knowledge. Naturally. 10 Dr. Even reaction/comments from individuals like David Marshall10 . I am asking from everyone to assume that his reaction was innocent one. A I encourage anyone with a penchant for writing. I am interested in it all. They will be removed gradually as the version number advances.

Open Channel Flow The chapter isn’t in the development stage yet. Additional material can be provided to give a diﬀerent angle on the issue at hand. Material can be further illuminate. illustration or photo of experiment. many chapters are missing. mistakes. Properties The chapter isn’t in development stage yet. Some parts were taken from Fundamentals of Die Casting Design book and are in a process of improvement. Nevertheless. It is hoped the A changes in TEX and LTEX related to this book in future will be minimal and minor. At this stage. there are speciﬁc issues which are on the “table” and they are described herein. xlix . There will always new problems to add or to polish the explanations or include more new materials. Speciﬁc missing parts from every chapters are discussed below. It is hoped that the style ﬁle will be converged to the ﬁnal form rapidly. Meta End You are always welcome to add a new material: problem. approach problems are sometime appears in the book under the Meta simple like this Meta sample this part. Also issues that associated with the book like the software has to be improved. question. These omissions.To Do List and Road Map This book isn’t complete and probably never will be completed.

l LIST OF TABLES .

Fluid mechanics is a branch of continuous mechanics which deals with a relationship between forces. It is known that these materials have the ability to drown people. motions. ﬂuid statics.CHAPTER 1 Introduction 1. or ﬂow round bodies (solid or otherwise). but a closer look reveals that the glass is a liquid with a large viscosity. The ﬂuid mechanics can also be distinguish between a single phase ﬂow and multiphase ﬂow (ﬂow made more than one phase or single distinguishable material). and statical conditions in continuous material. The last boundary (as all the boundaries in ﬂuid mechanics) 1 . Even material such as aluminum just below the mushy zone also behaves as a liquid similarly to butter.). the boundary between the solid mechanics and ﬂuid mechanics is some kind of gray shed and not a sharp distinction (see Figure 1. Furthermore. glass appears as a solid material. This study area deals with many and diversiﬁed problems such as surface tension. the discussion in this book is limited to simple and (mostly) Newtonian (sometimes power ﬂuids) ﬂuids which will be deﬁned later. After it was established that the boundaries of ﬂuid mechanics aren’t sharp. ﬂow in enclose bodies. almost any action a person is doing involves some kind of a ﬂuid mechanics problem. Materials like sand (some call it quick sand) and grains should be treated as liquids. The ﬂuid mechanics study involve many ﬁelds that have no clear boundary between them. In fact. etc. ﬂow stability. Researchers distinguish between orderly ﬂow and chaotic ﬂow as the laminar ﬂow and the turbulent ﬂow.1 for the complex relationships between the diﬀerent branches which only part of it should be drawn in the same time.1 What is Fluid Mechanics? Fluid mechanics deals with the study of all ﬂuids under static and dynamic situations. A proof of the glass “liquidity” is the change of the glass thickness in high windows in European Churches after hundred years. The bottom part of the glass is thicker than the top part. For example.

com/HTML ) analyzed a ﬂow of a complete still liquid assuming a . http://ekkinc. Moreover. For example.2 CHAPTER 1. -1. Diagram to explain part of relationships of ﬂuid mechanics branches. air with dust particle). Or. engineers in software company (EKK Inc. when a general model is need because more parameters are eﬀecting the situation. After it was made clear that the boundaries of ﬂuid mechanics aren’t sharp.1. ﬂow with two phases (or materials) can be treated as a single phase (for example. It is this author’s personal experience that the knowledge and ability to know in what area the situation lay is one of the main problems. Then the dimensional analysis will be used explain why in certain cases one distinguish area/principle is more relevant than the other and some eﬀects can be neglected. the study must make arbitrary boundaries between ﬁelds. INTRODUCTION Continuous Mechanics Solid Mechanics something between Fluid Mechanics Fluid Statics Fluid Dynamics Boundaries problems Multi phase flow Internal Flow Laminar Flow Stability problems Turbulent Flow Fig. isn’t sharp because ﬂuid can go through a phase change (condensation or evaporation) in the middle or during the ﬂow and switch from a single phase ﬂow to a multi phase ﬂow.

almost all knowledge of the ancients can be summarized as application of instincts. The ideal ﬂow (frictionless ﬂow) should be expanded compared to the regular treatment. These aqueducts reached their greatest size and grandeur in those of the City of Rome and China. BRIEF HISTORY 3 complex turbulent ﬂow model. turbulence. people realized that wells have to be dug and crude pumping devices need to be constructed. with the exception Archimedes (250 B.2 Brief History The need to have some understanding of ﬂuid mechanics started with the need to obtain water supply. Bernoulli family. There are two main approaches of presenting an introduction of ﬂuid mechanics materials. 1. After his initial work. These two approaches pose a dilemma to anyone who writes an introductory book for the ﬂuid mechanics. boundary layer and internal and external ﬂow . This book is unique in providing chapter on multiphase ﬂow. The ﬁrst progress in ﬂuid mechanics was made by Leonardo Da Vinci (1452-1519) who built the ﬁrst chambered canal lock near Milan. to be followed by stability. and D’Alembert.) on the principles of buoyancy. Reviewing many books on ﬂuid mechanics made it clear. one of the main goals of this book is to explain what model should be applied. Euler. chapters on open channel ﬂow (as a sub class of the multiphase ﬂow) and compressible ﬂow (with the latest developments) are provided. As in thermodynamics. At some point.1. This book attempts to ﬁnd a hybrid approach in which the kinematic is presented ﬁrst (aside to standard initial four chapters) follow by Integral analysis and continued by Diﬀerential analysis. For example. larger tunnels built for a larger water supply. This fact was acknowledged by D’Alembert who stated that. and continue with Empirical Analysis. Naturally. aqueducts were constructed. The ﬁrst approach introduces the ﬂuid kinematic and then the basic governing equations. Such absurd analysis are common among engineers who do not know which model can be applied. etc.C. there isn’t a clear winner. a large population created a need to solve waste (sewage) and some basic understanding was created. The second approach deals with the Integral Analysis to be followed with Diﬀerential Analysis. For example. Yet. Later. Thus. These two approaches have justiﬁcations and positive points.” For example the concept of ideal liquid that leads to motion with no resistance. Newton. people realized that water can be used to move things and provide power. two diﬀerent of school of thoughts were created: the ﬁrst be- . At that stage theory and experiments had some discrepancy. Before dealing with the boundaries. “The theory of ﬂuids must necessarily be based upon experiment. the knowledge of ﬂuid mechanics (hydraulic) increasingly gained speed by the contributions of Galileo. the simpliﬁed private cases must be explained. There were no calculations even with the great need for water supply and transportation. When cities increased to a larger size. Torricelli. conﬂicts with the reality. This discrepancy between theory and practice is called the “D’Alembert paradox” and serves to demonstrate the limitations of theory alone in solving ﬂuid problems.2. He also made several attempts to study the ﬂight (birds) and developed some concepts on the origin of the forces.

Kirchhoﬀ. Therefore. This concept leads to mathematical basis for many approximations. Thus. and the Kutta-Joukowski circulation theory of lift (1906). Ganguillet. At the end of the twenty century. and Manning. the way how it was calculated changed. the empirical formulas generated by ﬁtting curves to experimental data (even sometime merely presenting the results in tabular form) resulting in formulas that the relationship between the physics and properties made very little sense. were Brahms. for example. Taylor. Examples of such work are Hermann von Helmholtz’s concept of vortexes (1858). as in thermodynamics. Fanning. On the theoretical side. The introduction of the computers during the 60s and much more powerful personal computer has changed the ﬁeld. While the understanding of the fundamentals did not change much. Meyer. transformed the ﬂuid mechanics to modern science that we have known today. Chezy. Fabre. This problem led to two consequences. Rankine. and many others. However. and Froude’s idea of the use of models change the science of the ﬂuid mechanics.4 CHAPTER 1. considerable contribution were made by Euler. and the second believed that solution is the pure practical (experimental) aspect of ﬂuid mechanics. La Grange. In the middle of the nineteen century. On the “experimental” side. at the same time proposed many correlations to many ﬂuid mechanics problems. people cannot relinquish control. creating a matching between the two school of thoughts: experimental and theoretical. Dubuat. mainly in pipes and open channels area. resistance by Darcy. Prandtl and his students Blasius. Rayleigh. Bhuckingham. d’Aubisson. Hagen. These programs in many cases can capture all the appropriate parameters and adequately provide a reasonable description of the physics. The experimentalists. the demand for vigorous scientiﬁc knowledge that can be applied to various liquids as opposed to formula for every ﬂuid was created by the expansion of many industries. But. ﬁrst Navier in the molecular level and later Stokes from continuous point of view succeeded in creating governing equations for real ﬂuid motion. The Navier-Stokes equations. the development of dimensional analysis by Rayleigh. Dupuit. This demand coupled with new several novel concepts like the theoretical and experimental researches of Reynolds. INTRODUCTION lieved that the solution will come from theoretical aspect alone. Bossut. The obvious happened without theoretical guidance. Lanchester’s concept of circulatory ﬂow (1894). Helmhoitz. and Kelvin. Gas Dynamics. after World War Two. As results it created today “strange” names: Hydrodynamics. Stanton. were considered unsolvable during the mid nineteen century because of the high complexity. von Karman. and Blasius and several other individuals as Nikuradse. and Aeronautics. Thus. and Poisseuille. Hydraulics. Theoreticians tried to simplify the equations and arrive at approximated solutions representing speciﬁc cases. Weisbach. Coulomb. which describes the ﬂow (or even Euler equations). Perhaps the most radical concept that eﬀects the ﬂuid mechanics is of Prandtl’s idea of boundary layer which is a combination of the modeling and dimensional analysis that leads to modern ﬂuid mechanics. there are many . There are many open source programs that can analyze many ﬂuid mechanics situations. Today many problems can be analyzed by using the numerical tools and provide reasonable results. Rose. many call Prandtl as the father of modern ﬂuid mechanics.

In this book. Gas has no free interface/surface (since it does ﬁll the entire volume). The second quantity discussed here is the area. Hence. these kinds of areas should be addressed inﬁnitesimally and locally. For example. e. This is a result of division of a vector by a vector and it is referred to as tensor. It is also said that liquid cannot return to their original state after the deformation. the change of water pressure by 1000% only change the volume by less than 1 percent. The diﬀerence between a gas phase to a liquid phase above the critical point are practically minor.1). The traditional quantity. In the best scenario. assuming turbulent ﬂow for still ﬂow simply provides erroneous results (see for example. It must be remember that force is a vector. The direction of area is perpendicular to the area. For example.1. Inc). The gas ﬁlls the volume and liquid cannot. and it is referred to the direction of the area. KINDS OF FLUIDS 5 other cases that numerical analysis cannot provide any meaningful result (trends). EKK. so at this stage the tensor will have to be broken into its components. for most practical purposes considered. There are several quantities that have to be addressed in this discussion. It is a known fact said that the ﬂuid continuously and permanently deformed under shear stress while solid exhibits a ﬁnite deformation which does not change with time. The ﬁrst is force which was reviewed in physics. it behaves like solid and under others it behaves like liquid (see Figure 1. Later. This diﬀerentiation leads to three groups of materials: solids and liquids. The main diﬀerence between the liquids and gases state is that gas will occupy the whole volume while liquids has an almost ﬁx volume. This test creates a new material group that shows dual behaviors.3. For the discussion here. This diﬀerence can be. Area of three–dimensional object has no single direction. these programs are as good as the input provided. But below the critical point. So. sharp even though in reality this diﬀerence isn’t sharp. The unit used to measure is [N]. no weather prediction program can produce good engineering quality results (where the snow will fall within 50 kilometers accuracy. It is evident from this discussion that when a liquid is at rest. any change in pressure directly aﬀects the volume. Building a car with this accuracy is a disaster). no shear stress is applied.3 Kinds of Fluids Some diﬀerentiate ﬂuid from solid by the reaction to shear stress. Thus. then the change of volume is at best 5%. the discussion on the mathematic meaning will be presented (later version). if the change of pressure is signiﬁcantly less than that. The pressure component in the area . 1. The area is measured in [m2 ]. a change in the volume by more 5% will required tens of thousands percent change of the pressure. This quantity was discussed in physics class but here it has an addtional meaning. Thus. the pressure will not aﬀect the volume. the emphasis is on the physics. one in the area direction and two perpendicular to the area. The ﬂuid is mainly divided into two categories: liquids and gases. The study of this kind of material called rheology and it will (almost) not be discussed in this book. under certain limits.g it has a direction. the pressure has three components. In gaseous phase. which is force per area has a new meaning.

Thus. Diﬀerent from solid. When this assumption is broken.2. the F denotes the force. thus the small distance analysis is applicable.2) Where A is the area. F. The density is a property which requires that ρ liquid to be continuous. h is the distance between the plates. it did not reach/reduced to the size where the atoms or molecular statistical calculations are signiﬁcant (see Figure 1. In solid mechanics.2 shows the density as log ℓ a function of the sample size. the density remains constant. It doesn’t mean ǫ that a sharp and abrupt change in the density cannot occur. Schematics to describe the shear short distance of two plates as shown in Figure stress in ﬂuid mechanics. In this discussion.2 for point where the green lines converge to constant density). (1. the aim is to develop diﬀerential equation.6 CHAPTER 1. Consider moving the plate with a zero lubricant (h ∼ 0) (results in large force) or a large amount of lubricant (smaller force). the principles of statistical mechanics must be utilized. 1. . -1. the velocity of the plate increases also. that is.1) It must be noted that ε is chosen so that the continuous assumption is not broken. Figure 1. ﬂuid cannot pull directly but through a solid surface. (1. It referred to density that is independent of the sampling size. here it will be treated as a separate issue.4 Shear Stress ∆ℓ The shear stress is part of the pressure tensor. After certain sample size. U0x F However. -1.3). From solid mechanics study. The units used for the pressure components is [N/m2 ]. The other two components are referred as the shear stresses. INTRODUCTION direction is called pressure (great way to confuse. Density as a function of ρ= ∆m ∆V −→ε ∆V lim the size of sample. isn’t it?). h) (1. Experiments show that the increase of height will increase the velocity up to a certain range.3. The density can be changed and it is a function of time and space (location) but must have a continues property. then. the density is deﬁned as Fig. The upper plate velocity generally will be U = f (A. the shear stress is h β considered as the ratio of the force acting on y area in the direction of the forces perpendicular x to area. Consider liquid that undergoes a shear stress between a Fig. it was shown that when the force per area increases.

In steady state.4 it can be noticed that for a small angle.1.6) yields τxy = µ δβ δt (1.11) . SHEAR STRESS For cases where the dependency is linear. the distance the t0 < t1 < t2 < t3 upper plate moves after small amount of time.8) it follows that U =h δβ δt (1.3) (1. Hence.5) it follows that ratio of the velocity to height is proportional to shear stress. then it can be written for small a angel that dU δβ = δt dy (1.3) can be rearranged to be U F ∝ h A Shear stress was deﬁned as τxy = F A hF A 7 (1.6) Where µ is called the absolute viscosity or dynamic viscosity which will be discussed later in this chapter in great length. the following can be written U∝ Equations (1. the regular approximation provides Fig.9) Combining equation (1.8) From equation (1. δt is d = U δt (1. The deformation of ﬂuid due to shear geometry stress as progression of time.5) From equations (1.4.7) From Figure 1.4) (1. d = U δt = h δβ (1.9) with equation (1. -1. applying the coeﬃcient to obtain a new equality as τxy = µ U h (1.4.4) and (1.10) If the velocity proﬁle is linear between the plate (it will be shown later that it is consistent with derivations of velocity).

6)) F = 1 × 1. For this kind of substance τxy = µ dU dy (1. Many ﬂuids fall into this category such as air. This approximation is appropriate for many other ﬂuids but only within some ranges.5 A µU ∼ = 53. Thus. when the outer cylinder remains stationary. water etc. which is exhibited by all ﬂuids. The property of viscosity.5 m/sec. Assume steady state conditions. It can be assumed that the plates remains in equiledistance from each other and steady state is achived instanly. this interpretation is more suitable to explain the molecular mechanism of the viscosity. The units of absolute viscosity are [N sec/m2 ].9) can be interpreted as momentum in the x direction transfered into the y direction. The units of shear stress are the same as ﬂux per time as following F kg m 1 mU ˙ = 2 m2 A sec A kg m 1 sec sec m2 Thus.1 [m]. Solution Assuming Newtonian ﬂow.12) Newtonian ﬂuids are ﬂuids which the ratio is constant. In fact. the notation of τxy is easier to understand and visualize.1[m] diameters with height of 0.069 × 0.2[m] and 0.1: A space of 1 [cm] width between two large plane surfaces is ﬁlled with glycerine.2: Castor oil at 25◦ C ﬁlls the space between two concentric cylinders of 0. Some referred to shear stress as viscous ﬂux of x–momentum in the y–direction.01 End Solution Example 1. is due to the existence of cohesion and interaction between ﬂuid molecules. the following can be written (see equation (1. Solution .45[N ] h 0. Equation (1. Example 1. the viscosity is the resistance to the ﬂow (ﬂux) or the movement. INTRODUCTION Materials which obey equation (1.8 CHAPTER 1.10) referred to as Newtonian ﬂuid. Calculate the force that is required to drag a very thin plate of 1 [m2 ] at a speed of 0. These cohesion and interactions hamper the ﬂux in y–direction. Calculate the torque required to rotate the inner cylinder at 12 rpm.

liquid viscosities decrease as temperature increases.4 ¡ M= ∼ . The diﬀerent of power ﬂuids familys. The diﬀerence is due to their fundamentally different mechanism creating visτ cosity characteristics.986 0. This molecular activity is known to increase with temperature. Since these forces decrease rapidly with increases of temperature. in gases.5 Viscosity 1. This reasoning is a result of the considerations of the kinetic theory. the viscosity of gases will increase with temperature. exchange of momentum between layers brought as a result of molecular movement normal to the general direction of ﬂow. the momentum exchange due to molecular movement is small compared to the cohesive forces between the molecules. In gases.13 h 0. while in the dU liquids. Thus.5. ic op molecules are sparse and cohetr o ix th sion is negligible. However. ri = A µU ro − ri µ 2 π 2 0.0078[N m] h ¡ End Solution 1.5. the Fig. VISCOSITY The velocity is rps 9 ˙ U = r θ = 2 π ri rps = 2 × π × 0.1. In liquids. The same way as in example (1.1 × 12/60 = 0. This theory indicates that gas viscosities vary directly with the square root of temperature.5.1 General S Bi imp ng le ha m Viscosity varies widely with temperature. and it resists the ﬂow. Thus. the viscosity is primarily dependent on the magnitude of these cohesive forces. temperature variation has an opposite eﬀect on the viscosities of liqτ0 uids and gases.4 π ri Where rps is revolution per second.1). tic pl do ne Ne ic eu ct ei pe ps R re ho wt on ia n r-P hi as lip po ff di la ta nt . -1. thus. the molecules are more dx compact and cohesion is more dominate. the moment can be calculated as the force times the distance as ri 2 π ri h M =F In this case ro − ri = h thus.

Nitrogen (left) and Argon (right) viscosity as a function of the temperature and pressure after Lemmon and Jacobsen. When n = 1 equation represent Newtonian ﬂuid and K becomes the familiar µ. The lines in the above diagrams are only to show constant pressure lines.6 demonstrates that viscosity increases slightly with pressure. There is no such a thing of viscosity at 30% liquid. This relationship is referred to as power relationship Fig. Not all the materials obey this relationship. but this variation is negligible for most engineering problems. There is a large class of materials which shows a non-linear relationship with velocity for any shear stress.5. This class of materials can be approximated by a single polynomial term that is a = bxn . both materials are only a function of the temperature.5). K) in equation (1. 1.7. the relationship between the velocity and the shear stress was assumed to be linear. INTRODUCTION Fig. From the physical point of view. the pressure has minor eﬀect on the viscosity. -1. Figure 1.13) The new coeﬃcients (n. It must be stress that the viscosity in the dome is meaningless.2 Non–Newtonian Fluids In equation (1.10 CHAPTER 1. Well above the critical point. The shear stress as a function and it can be written as of the shear rate. Oils have the greatest increase of viscosity with pressure which is a good thing for many engineering purposes.13) are constant. On the liquid side below the critical point. It simply depends on the structure of the ﬂow as will be discussed in the chapter on multi phase ﬂow. viscosity τ =K dU dx n−1 dU dx (1. -1. The viscosity coeﬃcient is . the coeﬃcient depends on the velocity gradient.6.

for most practical purposes.14) dUx =0 dy if |τyx | < τ0 (1.000025 0.. The kinematic vis.0024 0. -1.002 0.16) The gas density decreases with the temperature.0026 0. VISCOSITY 11 always positive. according to Ferraris at el1 concrete behaves as shown in Figure 1.0008 0. F. This fact explains Fig. the liquid is dilettante.e-05 0. The reason for this new deﬁnition is that some experimental data are given in this form.000015 0.0028 0.5). The general relationship for simple Bingham ﬂow is τxy = −µ ± τ0 if |τyx | > τ0 (1.13) are referred to as purely viscous ﬂuids.0018 0. When n.003 0. The increase of the absolute viscosity with the temperature is enough to overcome the increase of density and thus. The liquids which satisfy equation (1. 1. de Larrard and N. The kinematic viscosity embraces both the viscosity and density properties of a ﬂuid.0004 0.0014 0.001 sec µ[ Nm2 ] 1. which are acceleration units (a combination of kinematic terms).0002 5.e-06 0 10 20 30 40 50 60 70 80 90 100 Temperature [◦ C ] May 1. 215-241 (2001) m ν[ sec ] 2 . 1 C.3 Kinematic Viscosity Air absolute & kinematic viscosity Atmospheric Pressure 0.0006 0. the kinematic viscosity also increase with the temperature for many materials. For example. In the simple case. However. this kind of ﬁgures isn’t used in regular engineering practice. Ferraris. 2008 ν= µ ρ (1. Materials Science of Concrete VI. When n is below one. the “liquid side” is like Newtonian ﬂuid for large shear stress.15) There are materials that simple Bingham model does not provide dequate explanation and a more sophisticate model is required.0016 0.5. cosity is deﬁned as 0. However. is above one. These results also explained better using the new deﬁnition.of the temperature. The above equation shows that the dimensions of ν to be square meter per second.0012 0.8. Materials which behave up to a certain shear stress as a solid and above it as a liquid are referred as Bingham liquids. Air viscosity as a function the name “kinematic” viscosity. eds. Martys. The Newtonian part of the model has to be replaced by power liquid. Mindess and J.5. Skalny. S. Many ﬂuids satisfy the above equation.1.00002 The kinematic viscosity is another way to look at the viscosity. [m2 /sec]. the ﬂuid is pseudoplastic.7.0022 0. Fluids that show increase in the viscosity (with increase of the shear) referred to as thixotropic and those that show decrease are called reopectic ﬂuids (see Figure 1.

0005 0 10 20 30 40 50 60 70 80 90 100 Temperature [◦ C ] May 1.555 Ti0 + Suth µ = µ0 0.555 × 800 + 120 800 524.9.17) viscosity at input temperature T reference viscosity at reference temperature. provides reasonable results2 for the range of −40◦ C to 1600◦ C m ν[ sec ] 2 µ[ N sec ] m2 0.555 Tin + Suth Where µ µ0 Tin Ti0 Suth .07 + 120 µ = 0. For isothermal ﬂow.1 Example 1.4 Estimation of The Viscosity Water absolute and kinematic viscosity Atmospheric Pressure 0. End Solution Liquid Metals 2 This author is ambivalent about statement.3: Calculate the viscosity of air at 800K based on Sutherland’s equation.0015 The absolute viscosity of many ﬂuids relatively doesn’t change with the pressure but very sensitive to temperature.001 0. INTRODUCTION 1.1. Ti0 input temperature in degrees Kelvin reference temperature in degrees Kelvin Suth is Sutherland’s constant and it is presented in the Table 1.002 0.555 × 524. -1.5. Use the data provide in Table 1.51 10−5 N sec m2 The viscosity increases almost by 40%.07 3 2 ∼ 2. The observed viscosity is about ∼ 3. The variations of air and water as a function of the temperature at atmospheric pressure are plotted in Figures 1.12 CHAPTER 1.00001827 × × 0. the viscosity can be considered constant in many cases. . Sutherland’s equation is used and according to the literature.9. Water viscosity as a func1.710−5 . have expressions that provide an estimate. Some common materials (pure and mixture) tion temperature. 2008 0.8 and Fig. T T0 3 2 (1. For many gases. Solution Appallying the constants from Suthelnd’s table provides 0.

0002018 0.67 518.00001827 0.5.67 528.1. The list for Sutherland’s equation coeﬃcients for selected materials. Substance formula i − C4 H10 CH4 CO2 O2 Hg Temperature T [◦ C] 23 20 20 20 380 Viscosity [ N sec ] m2 0.0000146 0.0000203 0.00000982 0.0000654 oxygen mercury vapor Table -1.93 540.00001720 0.2.1.0000876 0. VISCOSITY 13 coeﬃcients chemical formula Material ammonia standard air carbon dioxide carbon monoxide hydrogen nitrogen oxygen sulfur dioxide CO2 CO H2 N2 O2 SO2 N H3 Sutherland 370 120 240 118 72 111 127 416 TiO [K] 527.00001480 0.99 526.0001781 0.05 528.07 527. Viscosity of selected gases.0000076 0.0000109 0.67 524.0001254 Table -1.57 µ0 (N sec/m2 ) 0. .

INTRODUCTION Substance formula (C2 H5 )O C6 H6 Br2 C2 H5 OH Hg H2 SO4 Temperature T [◦ C] 20 20 26 20 25 25 25 25 25 20 ∼ 25◦ C ∼ 25◦ C ∼ 20◦ C ∼ 25◦ C ∼ 20◦ C 20 Viscosity [ N sec ] m2 0. Viscosity of selected liquids.000647 0.084 0.15-0. .6 0.01915 0.54 1.14 CHAPTER 1.05 0.001194 0.069 Olive Oil Castor Oil Clucuse Corn Oil SAE 30 SAE 50 SAE 70 Ketchup Ketchup Benzene Firm glass Glycerol Table -1.986 5-20 0.000652 ∼ 1 × 107 1.098 0.200 0.000946 0.001547 0.072 0.3.000245 0.

In Figure 1.256425 48. many aluminum alloys are behaving as a Newtonian liquid until the ﬁrst solidiﬁcation appears (assuming steady state thermodynamics properties). -1.Fig.00 30. The lower pressure is. Pr = P/Pc are drawn.636 58.26 44.5 1. Government Printing Oﬃce.3 28.07 16.4.5 151 289.865925 50.0 21. Even when there is a solidiﬁ. The lines of constant relative pressure.4 or similar information.4 305.S.54 15.183 39.9696 2.C. VISCOSITY chemical component H2 He Ne Ar Xe Air “mix” CO2 O2 C 2 H6 CH4 Water Molecular Weight 2.003 20. this graph also shows the trends. Washington D.6 26.096 K Pc [Bar] 12.016 4. In this graph. The simplest way is by obtaining the data from Table 1.97 44.0 0. if the information is available and is close enough to the critical point.10 exhibits several liquid metals (from The Reactor Handbook. other points can be estimated.2 154.1.3 19.0 18.11 the relative viscosity µr = µ/µc is plotted as a function of relative temperature.8 132 304.4 190.5 0 100 200 300 400 500 600 700 800 900 Temperature [◦ C ] May 1.01 32. ∼ 1[bar]. Atomic Energy Commission AECD-3646 U. if one point is well documented.5. 258.289945 27.) sec µ[ Nm2 ] 1.944 131. Furthermore.3 5. The critical pressure can be evaluated in the following three ways. 2008 The General Viscosity Graphs In case “ordinary” ﬂuids where information is limit. Figure 1.0 15. Vol.40685 22.7 647.9 15 Table -1. The second way.83865 46. The properties at the critical stage and their values of selected materials.4 49. Furthermore. Liquid Metal viscosity 2.7685 36. May 1995 p. then the critical .5 2.10.064 [MPa] µc [ N sec ] m2 3. the metal behavior can be estimated as a Newtonian material (further reading can be done in this author’s book “Fundamentals of Die Casting Design”).04 Tc [K] 33. cation (mushy zone). Liquid metals viscosity as a function of the temperature. Tr .47 2. µc is the viscosity at critical condition and µ is the viscosity at any given condition. for practical purpose. Hougen et al suggested to use graph similar to compressibility chart.358525 48. 19.8823 73.0 Li Na K Hg Pb Liquid metal can be considered as a Newtonian ﬂuid for many applications.

Physical Chemistry Japan Vol.6[N sec/m2 ] The observed value is 24[N sec/m2 ]3 . 26 No.19) Where vc is the critical molecular volume and M is molecular weight.41 154. Solution The critical condition of oxygen are Pc = 50. is by utilizing the following approximation µc = M Tc vc 2/3 ˜ (1.35[Bar] Tc = 154. only the mixture of low density gases is discussed for analytical expression. O2 at 100◦ C and 20[Bar].20) Calculate the reduced pressure and the reduced temperature and from the Figure 1.4 50.4 µc = 18 value of the reduced temperature is Tr ∼ The value of the reduced pressure is Pr ∼ 20 ∼ 0. Rev.4 N sec m2 The From Figure 1. INTRODUCTION µc = µ µr ﬁgure 1.11 it can be obtained µr ∼ 1.11 (1.16 viscosity is obtained as given CHAPTER 1.18) The third way. 2 1956. For most 3 Kyama.2 and the predicted viscosity is T able µ = µc µ µc = 18 × 1. In this book.15 ∼ 2. Example 1.4: Estimate the viscosity of oxygen. End Solution Viscosity of Mixtures In general the viscosity of liquid mixture has to be evaluated experimentally. Even for homogeneous mixture. . when none is available. Or ˜ √ µc = M Pc 2/3 Tc −1/6 (1. there isn’t silver bullet to estimate the viscosity.2 = 21.35 373. Makita.11 obtain the reduced viscosity.

2008 Fig. The mixture viscosity is highly nonlinear function of the fractions of the components.5. n µmix = i=1 xi µi n j=1 xi Φij 2 (1. The subscript i should be used for the j index. .22) Here. -1. xi is the mole fraction of component i.11.2 Pr=0.21) where Φi j is deﬁned as 1 Φij = √ 8 Mi 1+ Mj 1+ µi µj 4 Mj Mi (1. and µi is the viscosity of component i. the following Wilke’s correlation for gas at low density provides a result in a reasonable range. The dimensionless parameter Φij is equal to one when i = j. n is the number of the chemical components in the mixture. Reduced viscosity as function of the reduced temperature.5 Pr=1 Pr=2 Pr=3 Pr=5 Pr=25 4 5 6 7 8 9 10 0 5 2 2 3 T Tc 4 5 6 7 8 9 10 1 Reduced Temperature May 27.1. cases. VISCOSITY 17 Reduced Viscosity 2 10 liquid 5 dense gas Reduced Viscosity µ µc 2 two-phase region 1 critical point Pr=LD Pr=0.

5: Calculate the viscosity of a mixutre (air) made of 20% oxygen. 2008 Fig. INTRODUCTION 5 Tr=0. M 32. Reduced viscosity as function of the reduced temperature. O2 and 80% netrogen N2 for the temperature of 20◦ C.0000203 0.1 Tr=1. 28. Mole Fraction. µ 0. x 0.12. Example 1.6 Tr=2 Tr=3 µ µ0 4 Reduced viscosity 3 2 1 -1 10 2 5 1 2 5 10 2 P Reduced Pressure [ Pc ] June 2.00001754 .18 6 CHAPTER 1.2 0.2 Tr=1. Solution The following table summarize the known details i 1 2 Component O2 N2 Molecular Weight.8 Viscosity.4 Tr=1. -1.8 Tr=1 Tr=1.

0024 0.86 1. This equation (1. An example for values for this formula.25) .23) provides reasonable value only up to 2 m τ = 0.0215 N sec . For gases with very long molecular structure or complexity structure these formulas cannot be applied.1.143 0. To estimate the viscosity of the mixture with n component Hougen and Watson’s method for pseudocritial properties is adapted.8 × 1.0000203 0.2 × 0.0 Φij 1.0 m2 N sec m2 The observed value is ∼ 0.2 × 0. The term τs is the characteristic shear stress of the mixture. The higher viscosity is more dominate at low shear stress.23) τxy dy 1 + τs Where the term µ∞ is the experimental value at high shear stress.875 1. It this method the following is deﬁned as n Pc mix = i=1 xi Pc i (1.0024 0. the viscosity is dominated by a liquid with high viscosity and at high shear stress to be dominated by a liquid with the low viscosity liquid.0000181 0. m2 m2 and τs = 0.00105 N sec .0 µi /µj 1. Reiner and Phillippoﬀ suggested the following formula 1 µ0 − µ∞ τ xy dUx µ + 2 = ∞ (1.0 19 µmix ∼ 0.2 × 1.996 1. m2 Figure 1.0 + 0. for Molten Sulfur at temperature 120◦ C are µ∞ = 0.5. the viscosity is only a function of the temperature with a “simple” molecular structure.996 + 0.24) n Tc mix = i=1 xi Tc i (1. For some mixtures of two liquids it was observed that at a low shear stress.0 1.001 kN .8 × 0.0000073 kN . µ0 = 0.157 .00001754 N sec + ∼ 0.0000182 . VISCOSITY i 1 2 j 1 2 1 2 Mi /Mj 1.8 × 1.0 1.0 1. End Solution In very low pressure.12 can be used for a crude estimate of dense gases mixture. The term µ0 is the experimental viscosity at shear stress approaching zero. in theory.

09 1. The bulk modulus is deﬁned as BT = −v ∂P ∂v (1.80 1.03-4.79 K na 647.5 [Bar] nf nf nf na na 4.5 Bulk Modulus Similar to solids (hook’s law).10 1.3 4.27) T Using the identity of v = 1/ρ transfers equation (1.5.109 [MPa] na 22.15-2.4 K 514 K nf 850 K 1750 K Est 513 nf nf nf na na 591.00 [MPa] Est 78.06 1.28 2.2-28.62 1.27) into BT = ρ ∂P ∂ρ (1.60 1. It can be noted that this property is not the result of the equation of state but related to it.32 1.5.5 [Bar] 172.74 [MPa] 4.97 2.5 2.096 K Pc 57.5.064 [MPa] . Table -1.8 [Bar] 48 [Bar] 4.49 0. INTRODUCTION µc mix = i=1 xi µc i (1.20 and n CHAPTER 1. chemical component Acetic Acid Acetone Benzene Carbon Tetrachloride Ethyl Alcohol Gasoline Glycerol Mercury Methyl Alcohol Nitrobenzene Olive Oil Paraﬃn Oil SAE 30 Oil Seawater Toluene Turpentine Water Bulk Modulus 109 N m 2.174 Tc 593K 508 K 562 K 556. The bulk modulus for selected material with the critical temperature and pressure na −→ not available and nf −→ not found (exist but was not found in the literature).28) T The bulk modulus for several liquids is presented in Table 1.49 [MPa] 6.3 [Mpa] nf 7.52 26.5 0.20 1. liquids have a property that describes the volume change as results of pressure change for constant temperature.34 1.26) 1.

The increase of the pressure increases the bulk modulus due to the molecules increase of the rejecting forces between each other when they are closer. v) The full pressure derivative is dP = ∂P ∂T dT + v (1. VISCOSITY 21 In the literature.33) From equation (1.32) On constant pressure lines.35) sometimes is used in measurement of the bulk modulus. .29) P This parameter indicates the change of volume due to temperature change when the pressure is constant. This relationship is obtained by the observation that the pressure as a function of the temperature and speciﬁc volume as P = f (T. additional expansions for similar parameters are deﬁned.33) follows that dv dT ∂P ∂T ∂P ∂v =− P =const v (1.34) indicates that relationship for these three coeﬃcients is βT = − βv βP (1.5. the temperature increase results in reduction of the bulk of modulus because the molecular are further away. The thermal expansion is deﬁned as βP = 1 v ∂v ∂T (1. and therefore equation (1. Another deﬁnition is referred as coeﬃcient of tension and it is deﬁned as βv = 1 P ∂P ∂T (1.1. In contrast.30) v This parameter indicates the change of the pressure due to the change of temperature (where v = constant).34) T Equation (1. These deﬁnitions are related to each other.31) ∂P ∂v dv T (1. dP = 0.32) is 0= ∂P ∂T dT + v ∂P ∂v dv T (1.35) The last equation (1.

01 = 2. INTRODUCTION Example 1. surface tension. In many (physics.7: Calculate the pressure needed to apply on water to reduce its volume by 1 per cent.22 CHAPTER 1. Solution Using the deﬁnition for the bulk modulus ∆P ∼ βT ∆v ∼ 2. There is a common mis. It also responsible for the breakage of a liquid jet into other medium/phase to R1 many drops (atomization).035 per cent of its volume by applying a pressure of 5[Bar].15 107 [N/m2 ] = 215[Bar] v End Solution 1. -1.6 Surface Tension The surface tension manifested itself by a rise or depression of the 2dβ1 liquid at the free surface edge.714[Bar] ∂v ∆v 0.Fig.6: Calculate the modulus of liquid elasticity that reduced 0. The R2 dℓ1 surface tension is force per length x and is measured by [N/m] and is acting to stretch the surface.15 109 .00035 End Solution Example 1.cribing preincipls radii. and ﬂuid mechanics) books explained that the surface tension is a result from unbalance molecular cohesive forces. y Surface tension is also responsi2dβ2 ble for the creation of the drops dℓ2 and bubbles. conception for the source of the surface tension.13. Assume the temperature to be 20◦ C. Solution Using the deﬁnition for the bulk modulus βT = −v v 5 ∂P ∼ ∆P = ∼ 14285. This explanation is wrong since it is in conﬂict with Newton’s . Surface tension results from a sharp change in the density between two adjoint phases or ma. Surface tension control volume analysis disterials.

8: A Tank ﬁlled with liquid.38) Other extreme is for a sphere for which the main radii are the same and equation (1. Furthermore. Calculate the minimum work required to increase the pressure in tank by ∆P .37) is reduced to 2σ ∆P = (1. This erroneous explanation can be traced to Adam’s book but earlier source may be found. the pressure diﬀerence has to balance the surface tension. Solution The work is due to the change of the bubbles volume.1. The pressure on one side is Pi and the pressure on the other side is Po . the angles are very small and thus (sin β ∼ β).37) predicts that pressure diﬀerence increase with inverse of the radius.40) Example 1. the surface tension forces are puling the surface upward.37) is reduced to ∆P = σ 1 R (1. SURFACE TENSION 23 second law (see example ?). The second with two equal radii.36) For a very small area. When the surface tension is constant. There are two extreme cases: one) radius of inﬁnite and radius of ﬁnite size.41) . the horizontal forces cancel each other because symmetry. The forces in the vertical direction reads (Pi − Po ) d 1 d 2 = ∆Pd 1 d 2 = 2 σd 1 sin β1 + 2 σd 2 sin β2 (1. thus the pressure inside the bubble is ∆P = 4σ R (1. it can be noticed that d i ∼ 2 Ri dβi . The work is rf w= r0 ∆P (v)dv (1.39) R Where R is the radius of the sphere. inner and outer. The relationship between the surface tension and the pressure on the two sides of the surface is based on geometry. which contains n bubbles with equal radii. The ﬁrst case is for an inﬁnite long cylinder for which the equation (1. r. Thus. Consider a small element of surface.6. A soap bubble is made of two layers. In the vertical direction. the equation (1.36) can be simpliﬁed as ∆P = σ 1 1 + R1 R2 (1.37) Equation (1. Assume that the liquid bulk modulus is inﬁnity. Thus.

24

CHAPTER 1. INTRODUCTION

The minimum work will be for a reversible process. The reversible process requires very slow compression. It is worth noting that for very slow process, the temperature must remain constant due to heat transfer. The relationship between pressure diﬀerence and the radius is described by equation (1.39) for reversible process. Hence the work is

∆P rf dv rf r0

w=

r0

2σ 4 π r2 dr = 8 π σ r

rdr = 4 π σ rf 2 − r0 2

(1.42)

Where, r0 is the radius at the initial stage and rf is the radius at the ﬁnal stage. The work for n bubbles is then 4 π σ n rf 2 − r0 2 . It can be noticed that the work is negative, that is the work is done on the system.

End Solution

1.6.1

Wetting of Surfaces

To explain the source of the contact angle, consider the point where three phases became in contact. This contact point occurs due to free surface G reaching a solid boundary. The surface tension occurs between gas phase (G) to liquid phase (L) S L and also occurs between the solid (S) and the liquid phases as well as between the gas phase and the solid phase. In Figure 1.14, forces diagram is shown when control volume is chosen so that Fig. -1.14. Forces in Contact angle. the masses of the solid, liquid, and gas can be ignored. Regardless to the magnitude of the surface tensions (except to zero) the forces cannot be balanced for the description of straight lines. For example, forces balanced along the line of solid boundary is σgs − σls − σlg cos β = 0 and in the tangent direction to the solid line the forces balance is Fsolid = σlg sin β substituting equation (1.44) into equation (1.43) yields σgs − σls = Fsolid tan β (1.45) (1.44) (1.43)

For β = π/2 =⇒ tan β = ∞. Thus, the solid reaction force must be zero. The gas solid surface tension is diﬀerent from the liquid solid surface tension and hence violating equation (1.43).

1.6. SURFACE TENSION

25

The surface tension forces must be balanced, thus, a contact angle is created to balNonWetting ance it. The contact angle is determined by Wetting fluid whether the surface tension between the gas fluid solid (gs) is larger or smaller then the surface tension of liquid solid (ls) and the local geometry. It must be noted that the solid boundary Fig. -1.15. Description of wetting and isn’t straight. The surface tension is a molec- non–wetting ﬂuids. ular phenomenon, thus depend on the locale structure of the surface and it provides the balance for these local structures. The connection of the three phases–materials–mediums creates two situations which are categorized as wetting or non–wetting. There is a common deﬁnition of wetting the surface. If the angle of the contact between three materials is larger than 90◦ then it is non-wetting. On the other hand, if the angle is below than 90◦ the material is wetting the surface (see Figure 1.15). The angle is determined by properties of the liquid, gas medium and the solid surface. And a small change on the solid surface can change the wetting condition to non–wetting. In fact there are commercial sprays that are intent to change the surface from wetting to non wetting. This fact is the reason that no reliable data can be provided with the exception to pure substances and perfect geometries. For example, water is described in many books as a wetting ﬂuid. This statement is correct in most cases, however, when solid surface is made or cotted with certain materials, the water is changed to be wetting (for example 3M selling product to “change” water to non–wetting). So, the wetness of ﬂuids is a function of the solid as well.

Table -1.6. The contact angle for air, distiled water with selected materials to demonstrat the inconconsitency.

chemical component Steel Steel,Nickel Nickel Nickel Chrome-Nickel Steel Silver Zink Bronze Copper Copper Copper

Contact Angle π/3.7 π/4.74 π/4.74 to π/3.83 π/4.76 to π/3.83 π/3.7 π/6 to π/4.5 π/3.4 π/3.2 π/4 π/3 π/2

Source [1] [2] [1] [3] [4] [5] [4] [4] [4] [7] [8]

1 R. Siegel, E. G. Keshock (1975) “Eﬀects of reduced gravity on nucleate boiling bubble

26

CHAPTER 1. INTRODUCTION dynamics in saturated water,” AIChE Journal Volume 10 Issue 4, Pages 509 - 517. 1975

2 Bergles A. E. and Rohsenow W. M. ”The determination of forced convection surface– boiling heat transfer, ASME, J. Heat Transfer, vol 1 pp 365 - 372. 3 Tolubinsky, V.I. and Ostrovsky, Y.N. (1966) “On the mechanism of boiling heat transfer”,. International Journal of Heat and Mass Transfer, Vol. 9, No 12, pages 1465-1470. 4 Arefeva E.I., Aladev O, I.T., (1958) “wlijanii smatchivaemosti na teploobmen pri kipenii,” Injenerno Fizitcheskij Jurnal, 11-17 1(7) In Russian. 5 Labuntsov D. A. (1939) “Approximate theory of heat transfer by developed nucleate boiling” In Sussian Izvestiya An SSSR , Energetika I transport, No 1. 6 Basu, N., Warrier, G. R., and Dhir, V. K., (2002) “Onset of Nucleate Boiling and Active Nucleation Site Density during Subcooled Flow Boiling,” ASME Journal of Heat Transfer, Vol. 124, papes 717 -728. 7 Gaetner, R. F., and Westwater, J. W., (1960) “Population of Active Sites in Nucleate Boiling Heat Transfer,” Chem. Eng. Prog. Symp., Ser. 56. 8 Wang, C. H., and Dhir, V. K., (1993), “Eﬀect of Surface Wettability on Active Nucleation Site Density During Pool Boiling of Water on a Vertical Surface,” J. Heat Transfer 115, pp. 659-669 To explain the contour of the surface, and the contact angle consider simple “wetting” liquid contacting a solid material in two– dimensional shape as depicted in Figure 1.16. To solve the shape of the liquid surface, the P h pressure diﬀerence between the two sides of P P free surface has to be balanced by the surface tension. In Figure 1.16 describes the raising of the liquid as results of the surface tension. The surface tension reduces the pressure in the liquid above the liquid line (the dotted line in Fig. -1.16. Description of liquid surface. the Figure 1.16). The pressure just below the surface is −g h(x) ρ (this pressure diﬀerence will be explained in more details in Chapter 4). The pressure, on the gas side, is the atmospheric pressure. This problem is a two dimensional problem and equation (1.38) is applicable to it. Appalling equation (1.38) and using the pressure diﬀerence yields

0 0 0

g h(x), ρ =

σ R(x)

(1.46)

**1.6. SURFACE TENSION The radius of any continuous function, h = h(x), is ˙ 1 + h(x) R(x) = ¨ h(x)
**

2 3/2

27

(1.47)

˙ Where h is the derivative of h with respect to x. Equation (1.47) can be derived either by forcing a circle at three points at (x, x+dx, and x+2dx) and thus ﬁnding the the diameter or by geometrical analysis of triangles build on points x and x+dx (perpendicular to the tangent at these points). Substituting equation (1.47) into equation (1.46) yields σ g h(x) ρ = (1.48) 2 3/2 ˙ 1 + h(x) ¨ h(x) Equation (1.48) is non–linear diﬀerential equation for height and can be written as

ghρ dh d2 h (1.49) 1+ − 2 =0 σ dx dx ˙ With the boundary conditions that specify either the derivative h(x = r) = 0 (sym˙ = β or heights in two points or other combinations. metry) and the derivative at hx An alternative presentation of equation (1.48) is ghρ = Integrating equation (1.50) transforms into gρ h dh = σ ¨ h ˙ 1 + h2

3/2

2

3/2

¨ σh ˙ 1 + h2

3/2

(1.50)

dh

(1.51)

The constant Lp σ/ρ g is refered to as Laplace’s capillarity constant. The units of this ˙ constant are meter squared. The diﬀerential dh is h. Using dummy variable and the ˙ ¨ ˙ identys h = ξ and hence, h = ξ = dξ transforms equation (1.51) into 1 h dh = Lp ξdξ (1 + ξ 2 )

3/2

(1.52)

**After the integration equation (1.52) becomes h2 + constant = − 2 Lp 1 ˙ 1 + h2
**

1/2

(1.53)

28

CHAPTER 1. INTRODUCTION

At inﬁnity, the height and the derivative of the height must by zero so constant + 0 = −1/1 and hence, constant = −1 . 1− h2 = 2 Lp 1 ˙ 1 + h2

1/2

(1.54)

**Equation (1.54) is a ﬁrst order diﬀerential equation that can be solved by variables separation4 . Equation (1.54) can be rearranged to be ˙ 1 + h2
**

1/2

=

1 2 1 − 2h Lp

2

(1.55)

Squaring both sides and moving the one to the right side yields ˙ h2 = 1 2 1 − 2h Lp −1 (1.56)

**The last stage of the separation is taking the square root of both sides to be dh ˙ h= = dx or dh 1 2 1 − 2h Lp
**

2

1 2 1 − 2h Lp

2

−1

(1.57)

= dx −1

(1.58)

Equation (1.58) can be integrated to yield dh = x + constant 2 1 −1 2 1 − 2h Lp -

(1.59)

The constant is determined by the boundary condition at x = 0. For example if h(x − 0) = h0 then constant = h0 . This equation is studied extensively in classes on surface tension. Furthermore, this equation describes the dimensionless parameter that aﬀects this phenomenon and this parameter will be studied in Chapter ?. This book is introductory, therefore this discussion on surface tension equation will be limited. 1.6.1.1

4 This

Capillarity

p equation has an analytical solution which is x = Lp 4 − (h/Lp)2 − Lp acosh(2 Lp/h) + constant where Lp is the Laplace constant. Shamefully, this author doesn’t know how to show it in a two lines derivations.

1.6. SURFACE TENSION The capillary forces referred to the fact that surface tension causes liquid to rise or penetrate into area (volume), otherwise it will not be there. It can be shown that the height that the liquid raised in a tube due to the surface tension is 2 σ cos β h= g ∆ρ r (1.60)

h

29

Theory

actual 0

working range

R

Fig. -1.17. The raising height as a

Where ∆ρ is the diﬀerence of liquid density to the function of the radii. gas density and r is the radius of tube. But this simplistic equation is unusable and useless unless the contact angle (assuming that the Capilary Height contact angel is constant or a repressive average can be found or provided or can be measured) is given. However, in reality there is no readily information for contact angle5 and therefore this equation is useful to show the treads. The maximum that the contact angle can be obtained in equation (1.60) when β = 0 and thus cos β = 1. This angle is obtained when a perfect half a sphere shape exist of the liquid surface. In that case equation (1.60) becomes Fig. -1.18. The raising height as a

1.2 1.0

{

Distilled water [23 C] Mercury [25 C] Equation

0.8

Height [cm]

0.6

0.4

0.2

0.0

0.0

0.3

0.6

0.9

1.2

1.5

1.8

2.1

2.4

2.7

Radii [cm]

May 29, 2008

hmax

2σ = g ∆ρ r

function of the radius.

(1.61)

Figure 1.18 exhibits the height as a function of the radius of the tube. The height based on equation (1.61) is shown in Figure 1.17 as blue line. The actual height is shown in the red line. Equation (1.61) provides reasonable results only in a certain range. For a small tube radius, equation (1.49) proved better results because the curve approaches hemispherical shaper (small gravity eﬀect). For large radii equation (1.49) approaches the strait line (the liquid line) strong gravity eﬀect. On the other hand, for extremely small radii equation (1.61) indicates that the high height which indicates a negative pressure. The liquid at a certain pressure will be vaporized and will breakdown the model upon this equation was constructed. Furthermore, the small scale indicates that the simplistic and continuous approach is not appropriate and a diﬀerent model is needed. The conclusion of this discussion are shown in Figure 1.17. The acutal dimension for many liquids (even water) is about 1-5 [mm]. The discussion above was referred to “wetting” contact angle. The depression of the liquid occurs in a “negative” contact angle similarly to “wetting.” The depression height, h is similar to equation (1.61) with a minus sign. However, the gravity is working against the surface tension and reducing the range and quality of the predictions of equation (1.61). The measurements of the height of distilled water and mercury are

5 Actially, there are information about the contact angle. However, that information conﬂict each other and no real information is avaible see Table 1.6.

30

CHAPTER 1. INTRODUCTION

presented in Figure 1.18. The experimental results of these materials are with agreement with the discussion above. The surface tension of a selected material is given in Table 1.7. In conclusion, the surface tension issue is important only in case where the radius is very small and gravity is negligible. The surface tension depends on the two materials or mediums that it separates. Example 1.9: Calculate the diameter of a water droplet to attain pressure diﬀerence of 1000[N/m2 ]. You can assume that temperature is 20◦ C. Solution The pressure inside the droplet is given by equation (1.39). D = 2R = 22σ 4 × 0.0728 = ∼ 2.912 10−4 [m] ∆P 1000

End Solution

Example 1.10: Calculate the pressure diﬀerence between a droplet of water at 20◦ C when the droplet has a diameter of 0.02 cm. Solution using equation ∆P = 2σ 2 × 0.0728 ∼ ∼ 728.0[N/m2 ] r 0.0002

End Solution

Example 1.11: Calculate the maximum force necessary to lift a thin wire ring of 0.04[m] diameter from a water surface at 20◦ C. Neglect the weight of the ring. Solution F = 2(2 π r σ) cos β The actual force is unknown since the contact angle is unknown. However, the maximum Force is obtained when β = 0 and thus cos β = 1. Therefore, F = 4 π r σ = 4 × π × 0.04 × 0.0728 ∼ .0366[N ] In this value the gravity is not accounted for.

End Solution

The surface tension for selected materials at temperature 20 C when not mentioned.70 26.6 5. chemical component Acetic Acid Acetone Aniline Benzene Benzylalcohol Benzylbenzoate Bromobenzene Bromobenzene Bromoform Butyronitrile Carbon disulﬁd Quinoline Chloro benzene Chloroform Cyclohexane Cyclohexanol Cyclopentanol Carbon Tetrachloride Carbon disulﬁd Chlorobutane Ethyl Alcohol Ethanol Ethylbenzene Ethylbromide Ethylene glycol Formamide Gasoline Glycerol Helium Mercury Methanol Methyl naphthalene Methyl Alcohol Neon Nitrobenzene Olive Oil Perﬂuoroheptane Perﬂuorohexane Surface Tension 27.90 43.50 41.0920 -0.1118 n/a n/a -0.2049 -0.95 36.10 22.6 25.15 43.6.4 28.1484 -0.10 32.1308 -0.0890 -0.00 45.88 39.1159 -0.1295 -0.30 23.067 -0.50 24.40 32.91 mN m T 20◦ C 22◦ C 25◦ C −269◦ C −247◦ C - correction mN mK n/a -0.1484 -0.1037 -0.30 43.1117 n/a -0.12 425-465.12 33.10 29.0773 -0.70 58.95 34.0842 n/a -0.3 22.1160 -0. SURFACE TENSION ◦ 31 Table -1.20 ∼ 21 64.1291 -0.0 12.85 11.20 24.1191 -0.1.60 27.0972 -0.1011 n/a -0.1120 -0.0935 .0-48.7.8 32.20 43.1160 -0.0 22.20 47.50 36.50 28.0 0.1066 -0.1063 -0.0832 -0.0966 -0.1177 -0.60 22.1211 -0.0598 n/a -0.70 38.1085 -0.1094 -0.

00 41.1172 -0.10 28.60 n/a 54-69 28.1101 -0.1100 n/a n/a -0.1372 -0.0902 -0.0777 -0.70 38.00 36.32 CHAPTER 1.1104 .50 23.7.1189 n/a -0.1514 -0.4 27 72.80 30.90 -0. The surface tension for selected materials (continue) chemical component Perﬂuorooctane Phenylisothiocyanate Propanol Pyridine Pyrrol SAE 30 Oil Seawater Toluene Turpentine Water o-Xylene m-Xylene Surface Tension mN m T 25◦ C - correction mN mK 14. INTRODUCTION Table -1.

System This term will be used in this book and it is deﬁned as a continuous (at least partially) ﬁxed quantity of matter. Work In mechanics. The dimensions of this material can be changed. Two. So.CHAPTER 2 Review of Thermodynamics In this chapter. it is assumed that the system speed is signiﬁcantly lower than that of the speed of light.1) This deﬁnition can be expanded to include two issues. there is a transfer of energy so that its eﬀect can cause work. This introduction is provided to bring the student back to current place with the material. 2. The ﬁrst issue that must be addressed. the work was deﬁned as mechanical work = F•d = P dV (2. a review of several deﬁnitions of common thermodynamics terms is presented.1 Basic Deﬁnitions The following basic deﬁnitions are common to thermodynamics and will be used in this book. In fact for almost all engineering purpose this law is reduced to two separate laws of mass conservation and energy conservation. 33 . It must be noted that electrical current is a work while heat transfer isn’t. In this deﬁnition. that work done on the surroundings by the system boundaries similarly is positive. the mass can be assumed constant even though the true conservation law applied to the combination of mass energy (see Einstein’s law).

3) is that the way the work is done and/or intermediate states are irrelevant to ﬁnal results. For such body force. = mU 2 2 (2. The internal energy is the energy that depends on the other properties of the system. From the ﬁrst law it directly implies that for process without heat transfer (adiabatic process) the following is true W12 = E1 − E2 (2. Q12 − W12 = E2 − E1 (2. and electrical energy.E. is the internal energy per unit mass. The statement describing the law is the following. A common body force is the gravity. work.4) Thus the energy equation can be written as Total Energy Equation mU1 mU2 2 + mgz1 + EU 1 + Q = + mgz2 + EU 2 + W 2 2 For the unit mass of the system equation (2.34 CHAPTER 2. the potential energy is mgz where g is the gravity force (acceleration). m is the mass and the z is the vertical height from a datum. Thermodynamics First Law This law refers to conservation of energy in a non accelerating system. There are several deﬁnitions/separations of the kind of works and they include kinetic energy. the conservation is applied to all systems. etc as long the mass remain constant the deﬁnition is not broken. The “new” internal energy. chemical potential.3) Interesting results of equation (2. The internal energy is denoted in this book as EU and it will be treated as a state property.6) 2 (2. For example for pure/homogeneous and simple gases it depends on two properties like temperature and pressure.5) is transformed into Spesiﬁc Energy Equation U1 2 U2 2 + gz1 + Eu 1 + q = + gz2 + Eu 2 + w 2 2 (2. REVIEW OF THERMODYNAMICS Our system can receive energy.5) where q is the energy per unit mass and w is the work per unit mass.2) The system energy is a state property. potential energy (gravity). Eu . The kinetic energy is K. Since all the systems can be calculated in a non accelerating systems. . The potential energy of the system is depended on the body force. etc.

1.7) In the same manner. BASIC DEFINITIONS 35 Since the above equations are true between arbitrary points.8) For the case were the body force.9) reduced to Time Dependent Energy Equation D EU DU Dz ˙ ˙ Q−W = + mU + mg Dt Dt Dt (2. The rate of change of the energy transfer is DQ ˙ =Q Dt (2.2. the integral is independent of the path. The most common mathematical form is Clausius inequality which state that δQ ≥0 T (2.10) The time derivative operator.11) The integration symbol with the circle represent integral of cycle (therefor circle) in with system return to the same condition. the work change rate transfered through the boundaries of the system is DW ˙ =W Dt Since the system is with a ﬁxed mass. the rate energy equation is DU D Bf z D EU ˙ ˙ + mU +m Q−W = Dt Dt Dt (2. D/Dt is used instead of the common notation because it referred to system property derivative.13) . Thus diﬀerentiating the energy equation with respect to time yields the rate of change energy equation. it is referred as a reversible process and the inequality change to equality. Thermodynamics Second Law There are several deﬁnitions of the second law. δQ =0 T (2.9) (2. No matter which deﬁnition is used to describe the second law it will end in a mathematical form. This observation leads to the deﬁnition of entropy and designated as S and the derivative of entropy is ds ≡ δQ T rev (2. Hence. These states are independent of the path the system goes through. choosing any point in time will make it correct. Bf . is constant with time like in the case of gravity equation (2.12) The last integral can go though several states. If there is no lost.

the process in which it is reversible and adiabatic. which is the combination of already deﬁned properties.17) (2.17) in mass unit is dP T ds = du + P dv = dh − (2.10) results in T dS = d EU + P dV (2. Thus. For reversible process equation (2. h.21) ρ .15) (2. H = EU + P V The speciﬁc enthalpy is enthalpy per unit mass and denoted as. the entropy remains constant and referred to as isentropic process. the enthalpy of the system.18) the (2. It can be noted that there is a possibility that a process can be irreversible and the right amount of heat transfer to have zero change entropy change.18) For isentropic process. Thus. it still remail valid for all situations. equation (2.17) is reduced to dH = V dP .12) can be written as δQ = T dS and the work that the system is doing on the surroundings is δW = P dV Substituting equations (2.20) (2.19) (2.17) yields (one form of) Gibbs Equation T dS = dH − V dP (2. the reverse conclusion that zero change of entropy leads to reversible process. Furthermore. The equation (2.14) One of the conclusions that can be drawn from this analysis is for reversible and adiabatic process dS = 0.16) into (2. Enthalpy It is a common practice to deﬁne a new property. REVIEW OF THERMODYNAMICS Performing integration between two states results in 2 S2 − S1 = 1 δQ = T rev 2 dS 1 (2. it can be shown that it is valid for reversible and irreversible processes. Or in a diﬀerential form as dH = dEU + dP V + P dV Combining equations (2.16) (2.15) Even though the derivation of the above equations were done assuming that there is no change of kinetic or potential energy.36 CHAPTER 2. isn’t correct.

that ”all gases at the same pressures and temperatures have the same number of molecules per unit of volume. and it is deﬁned as P = ρRT (2.22) And since the change of the enthalpy involve some kind of work is deﬁned as Spesiﬁc Pressure Heat Cp ≡ ∂h ∂T (2. The simplest equation of state referred to as ideal gas.26) . 37 Speciﬁc Heats The change of internal energy and enthalpy requires new deﬁnitions. the ratio of the speciﬁc heats is almost 1 and therefore the diﬀerence between them is almost zero. k. Spesiﬁc Heats Ratio Cp k≡ Cv (2.2. The ratio the speciﬁc heat of gases is larger than one. BASIC DEFINITIONS when the density enters through the relationship of ρ = 1/v. pressure. and speciﬁc volume deﬁne the equation of state for gases. Commonly the diﬀerence for solid is ignored and both are assumed to be the same and therefore referred as C.23) The ratio between the speciﬁc pressure heat and the speciﬁc volume heat is called the ratio of the speciﬁc heat and it is denoted as.1.25) Application of Avogadro’s law. The ﬁrst change of the internal energy and it is deﬁned as the following Spesiﬁc Volume Heat Cv ≡ ∂Eu ∂T (2.” This constant to match the standard units results in ¯ R = 8. Equation of state Equation of state is a relation between state variables.24) For solid.” allows the calculation of a “universal gas constant. Normally the relationship of temperature. This approximation less strong for liquid but not by that much and in most cases it applied to the calculations.3145 kj kmol K (2.

7354 0.29637 2.38 CHAPTER 2.14304 0.07 28.29683 0.091 1.7662 1. REVIEW OF THERMODYNAMICS Thus.299 1.20813 0.12418 0.7448 1.667 1.48152 1. Table -2.01 28.28700 0.5734 0.0413 1. the speciﬁc gas can be calculate as R= ¯ R M (2.186 1.8418 1.27650 0.1.29680 0.2091 2.400 1.015 R kj KgK CP kj KgK Cv kj KgK k Air Argon Butane Carbon Dioxide Carbon Monoxide Ethane Ethylene Helium Hydrogen Methane Neon Nitrogen Octane Oxygen Propane Steam 0.51835 0.016 16.7164 0.18892 0.25983 0.948 58.25) of state for perfect gas it follows d(P v) = RdT For perfect gas dh = dEu + d(P v) = dEu + d(RT ) = f (T ) (only) (2.28) .289 1.124 44.7113 0.29) (2.970 39.4897 1.0035 0. Properties of Various Ideal Gases [300K] Gas Chemical Formula Ar C4 H10 CO2 CO C 2 H6 C 2 H4 He H2 CH4 Ne N2 C8 H18 O2 C 3 H8 H2 O Molecular Weight 28.07703 4.5482 5.237 1.5203 1.6529 0.999 44.04 20.07279 0.409 1.3122 1.0299 1.1926 14.0416 1.013 114.1.327 From equation (2.6385 0.41195 0.044 1.667 1.01 30.27) The speciﬁc constants for select gas at 300K is provided in table 2.003 2.7445 1.4108 1.393 1.0849 1.183 28.230 31.6179 0.7165 0.6618 1.18855 0.6794 1.097 18.400 1.4909 1.400 1.8723 0.2518 3.667 1.2537 1.9216 1.1156 10.054 4.126 1.

35) Or using speciﬁc heat ratio equation (2.1). k value ranges from unity to about 1.2.30) Utilizing equation (2.30) and dividing by dT yields Cp − Cv = R This relationship is valid only for ideal/perfect gases.667.32) (2. The entropy for ideal gas can be simpliﬁed as the following 2 s2 − s1 = 1 dh dP − T ρT (2.1.” The values of several gases can be approximated as ideal gas and are provided in Table (2. These values depend on the molecular degrees of freedom (more explanation can be obtained in Van Wylen “F.37) as Ideal Gas Isontropic Relationships T2 = T1 P2 P1 k−1 k = V1 V2 k−1 (2.34) Using the identities developed so far one can ﬁnd that 2 s2 − s1 = 1 Cp dT − T 2 1 R dP T2 P2 = Cp ln − R ln P T1 P1 (2.35) transformed into k T2 P2 s2 − s1 = ln − ln R k − 1 T1 P1 For isentropic process.28) and subsisting into equation (2. BASIC DEFINITIONS From the deﬁnition of enthalpy it follows that d(P v) = dh − dEu 39 (2.33) The speciﬁc heat ratio.36) (2. the following is obtained T2 ln = ln T1 P2 P1 k−1 k (2.37) There are several famous identities that results from equation (2.31) Cp to Spesiﬁc Heats Ratio Cp = kR k−1 (2. ∆s = 0. of Classical thermodynamics.38) . The ratio of the speciﬁc heats can be expressed in several forms as Cv to Spesiﬁc Heats Ratio Cv = R k−1 (2.

This correction factor referred as the compressibility factor and deﬁned as Z deviation from the Ideal Gas Model PV Z= RT (2.39) . The real gas has a correction factor to account for the deviations from the ideal gas model. REVIEW OF THERMODYNAMICS The ideal gas model is a simpliﬁed version of the real behavior of real gas.40 CHAPTER 2.

CHAPTER 3 Review of Mechanics This author would like to express his gratitude to Dan Olsen (former Minneapolis city Engineer) and his friend Richard Hackbarth. t in a location. R . The velocity is derivative of the change of the location and using the chain role (for the direction and one for the magnitude) results. These concepts and deﬁnitions will be used in this book and a review is needed.1: A water jet is supposed be used to extinguish the ﬁre in a building as depicted in Figure 41 . It also can be noticed that this derivative is present derivation of any victory.1) Notice that ω can have three dimensional components. This chapter provides a review of important deﬁnitions and concepts from Mechanics (statics and dynamics).1 Kinematics of of Point Body A point body is location at time. change in R direction change in perpendicular to R U = dR = dt dR dt + R ω ×R (3. 3.2) Example 3. The acceleration is the derivative of the velocity “regular acceleration” angular acceleration centrifugal acceleration Coriolis acceleration a= dU = dt d2R dt2 + R× R dω dt +ω × R × ω +2 dR dt ×ω R (3.

Calculate what is the veFig.I.a).42 CHAPTER 3.I.I.c) b= and equation (3.c) These nonlinear equations (3.d) End Solution 1 While the simple example does not provide exact use of the above equation it provides experience of going over the motions of kinematics.a) becomes √ ag −g a 0= + U cos θ =⇒ U = U cos θ cos θ Substituting (3. and time. Description of the extinguish locity so that the jet reach the winnozzle aimed at the building window. t and three equations.I.e) into (3.b) The velocity for the y coordinate at the window is zero u(t) = 0 = −g t + U sin θ (3. θ.11 .e) −g a2 + a tan θ 2 U 2 cos2 θ (3.I.a) The distance for y equation for coordinate (zero is at the window) is 0=− g t2 + U sin θ t − b 2 (3.c) can be solved explicitly.I. For given velocity.I. REVIEW OF MECHANICS 3. What is the angle that jet has to be aimed.I. dow. (3. Isolating t from (3.1.I.I.a) and substituting into equations (3. U sin θ U cos θ θ Solution The initial velocity is unknown and denoted as U which two components.b) and (3.I.f) (3. at what angle the jet has to be shot so that velocity will be horizontal at the window.d) results in tan θ = 1 b + a 2 (3. -3. it b proposed to calculate the velocity a of the point particle to toward the window.I. Assume that gravity is g and the distance of the nozzle from the building is a and height of the window from the nozzle is b. U . .b) and (3.I. To simplify the calculations. The equation for the x coordinate is a = U cos θ t (3.I.I. The velocity at x is Ux = U cos θ and the velocity in y direction is Uy = U sin θ. There there are three unknowns.

CENTER OF MASS 43 3. It turns out that this concept is very useful in calculating rotations. x. In same is calculated. fashion the center of mass can be deﬁned in the other directions as following xi of Center Mass 1 xi = ¯ xi ρ(xi )dV m V (3. Suppose that the body has a distribution of the mass (density.Fig. 3. . Thus. This concept is derived from the fact that a body has a center of mass/gravity which interacts with other bodies and that this force acts on the center (equivalent force).3. The density “normally” deﬁned as mass per volume. ρ(xi ) is the line density as function of xi . Note. center of area (two–dimensional body with equal distribution mass). the body will not rotate. the center will be deﬁned as 1 x= ¯ m dm y x ρ(x)dV V (3.2.1 Actual Center of Mass In many engineering problems. m is the total mass of the object. center of the mass and two. y or z. moment of inertia.4) where xi is the direction of either.2.2. In x coordinate. the the line density is referred to density mass per unit length in the x direction. rho) as a function of the location. Description of how the center of mass tion in the y nor by z directions. if the body isn’t be held through the center of mass. if a body will be held by one point it will be enough to hold the body in the direction of the center of mass. The center of mass doesn’t depend on the coordinate system and on the way it is calculated. It is convenient to use the Cartesian system to explain this concept. It can be noticed that center of mass in the x–direction isn’t aﬀected by the distribu. the center of mass is required to make the calculations. the x mass. even for solid and uniform density the line density is a function of the geometry.2. etc. The density. the dV element has ﬁnite dimendV sions in y–z plane and inﬁnitesimal dimension in x direction see Figure 3. ﬁrst.2 Center of Mass The center of mass is divided into two sections. then a moment in additional to force is required (to prevent the body for rotating). Here. The physical meaning of the center of mass is that if a straight line force acts on the body in away through the center of gravity. Also. In other words.3) z Here. -3.

6) when the integral now over only the area as oppose over the volume. it turns out that the moment of inertia has much wider applicability.5) Fig. the moment of inertia is divided into moment of inertia of mass and area.3. Thus. Moment of inertia of mass is deﬁned as Moment of Inertia Irr m = ρr2 dm m (3. are constant and can be canceled.3 which has density. 3.5) can be transferred into Aproxiate xi of Center Mass xi = ¯ 1 A xi dA A (3. Thus equation (3.2 Aproximate Center of Area t dA Y In the previous case. the body was a three dimensional shape.2. schematic. equation (3. Finding the centroid location should be done in the most convenient coordinate system since the location is coordinate independent. t.8) . Furthermore.7) If the density is constant then equation (3.3. REVIEW OF MECHANICS 3. ρ. Consider a uniform thin body with constant thickness shown in Figure 3. -3.3) can be transferred into 1 x= ¯ tA ρ V dm z x x ρ t dA V (3.3 Moment of Inertia As it was divided for the body center of mass. Thin body center of mass/area The density. There are cases where the body can be approximated as a twodimensional shape because the body is with a thin with uniform density.7) can be transformed into Irr m = ρ r2 dV V (3. ρ and the thickness.1 Moment of Inertia for Mass The moment of inertia turns out to be an essential part for the calculations of rotating bodies.44 CHAPTER 3. 3.

2 3.3. .12) can be written as Ixx = A y +z 2 2 dA (3. Any point distance can be calculated from axis x as y (3. -3. but dependent on the location of axis of rotation relative to the body.2.14) ∆y x z’ ∆x x’ In the same fashion for other two coordinates as Iyy = A x2 + z 2 dA Fig.1 Moment of Inertia for Area General Discussion For body with thickness.12) x= y2 + z2 (3.10) 3. The schematic that explains the sum- (3.4. equation (3.3. MOMENT OF INERTIA 45 The moment of inertia is independent of the coordinate system used for the calculation.15) mation of moment of inertia.3.3.9) The body has a diﬀerent moment of inertia for every coordinate/axis and they are Ixx = Iyy = Izz = V V V rx 2 dm = ry 2 dm = rz 2 dm = (3.13) y’ C z Thus. t and uniform density the following can be written moment of inertia for area Ixx m = r2 dm = ρ t m A r2 dA (3. rk = Im m (y 2 + z 2 ) dm (x2 + z 2 ) dm V (x2 + y 2 ) dm V V (3.11) The moment of inertia about axis is x can be deﬁned as Moment of Inertia Ixx m Ixx = r2 dA = ρt A where r is distance of dA from the axis x and t is the thickness. Some people deﬁne the radius of gyration as an equivalent concepts for the center of mass concept and which means if all the mass were to locate in the one point/distance and to obtain the same of moment of inertia.

-3.5.20) The moment of inertia of several areas is the sum of moment inertia of each area see Figure 3.2 The Parallel Axis Theorem The moment of inertial can be calculated for any axis.21) x If the same areas are similar thus n Fig.17) can be expended as Ixx =0 Ix x = A y 2 + z 2 dA + 2 A (y ∆y + z ∆z) dA + A (∆y) + (∆z) 2 2 dA (3.2.16) 3.17) equation (3.19) Hence. The third term is a new term and can be written as constant r2 2 A 2 A 2 2 (∆y) + (∆z) A dA = (∆y) + (∆z) dA = r2 A (3. REVIEW OF MECHANICS Izz = A x2 + y 2 dA (3. The knowledge about one axis can help calculating the moment of inertia for a parallel axis.5 and therefore. n 2 1 y Ixx = i=1 Ixx i (3.18) on the right hand side is the moment of inertia about axis x and the second them is zero.46 CHAPTER 3. the relationship between the moment of inertia at xx and parallel axis x x is Parallel Axis Equation Ix x = Ixx + r2 A z (3. The schematic to explain the summation of moment of inertia. Ixx = i=1 Ixxi = n Ixxi (3.3. The second therm is zero because it integral of center about center thus is zero.22) . The moment of inertia for axis x is Ix x = A r dA = A 2 y 2 +z 2 dA = A (y + ∆y) + (z + ∆z) 2 2 dA (3.18) The ﬁrst term in equation (3. Let Ixx the moment of inertia about axis xx which is at the center of mass/area.

The material is with an uniform density and homogeneous.6. End Solution Fig.3 Examples of Moment of Inertia Example 3. dr r 3.7. Solution The moment of inertia is calculated utilizing equation (3.6 as Irr = ρ V r dm = ρ 0 2 r0 dV r h 2 π r dr = ρ h 2 π 2 r0 4 1 1 = ρhπr0 4 = m r0 2 4 2 2 The radius of gyration is rk = 1 2 mr0 2 r0 =√ m 2 End Solution Example 3.3: Calculate the moment of inertia of the rectangular shape shown in Figure 3.3.6. -3. as shown in Figure 3.3. -3.7 around x coordinate.14) as following 0 y z b dx a x Ixx = A 2 2 y +z dA = 0 a dA z 2 bdz = a b 3 3 This value will be used in later examples. . r0 . Solution The element can be calculated using cylindrical coordinate. Here the convenient element is a shell of thickness dr which shown in Figure 3. The moment of inertia can then move the center of area. Description of rectangular in x–y plane for calculation of moment of inertia. Cylinder with an element for calculaof the tion moment of inertia. the moment of inertial of half a circle is half of whole circle for axis a the center of circle.3. For example. MOMENT OF INERTIA h 47 Equation (3. Fig.22) is very useful in the calculation of the moment of inertia utilizing the moment of inertia of known bodies.2: Calculate the moment of inertia for the mass of the cylinder about center axis which height of h and radius.

consider a simple shape to see the eﬀects of this assumption.23) dz dIx x m b a3 = ρdy + z2 12 2 r r A 2 ba A a b (3.8.4: To study the assumption of zero thickness. b has no eﬀect on the error. A square element for the calculations of inertia of two-dimensional to three– dimensional deviations.26) = 2 = 3 t2 Ixx m t ba + ba 1 + a2 Ixx Ixxm It can be noticed right away that equation (3. Ad. Solution The moment of inertia of transverse slice about y (see Figure mech:ﬁg:squareEll) is Ixx t dIxx m = ρ dy The transformation into from local axis x to center axis. I can be noticed that the error is signiﬁcant very fast even for small values of t/a while the with of the box. February 28. 2008 t a End Solution ratio is a dimensionless number that commonly has no special name.26) indicates that ratio approaches one when thickness ratio is approaches zero. -3. This author suggests to call this ratio as the B number. -3. Ixx m = ρ −t/2 b a3 a b t 2 + a3 b + z 2 b a dz = ρ t 12 12 (3. REVIEW OF MECHANICS Example 3.48 CHAPTER 3.9.25) Comparison with the thin body results in Ixx ρ t b a3 1 (3. x can be done as following Ixx b a3 12 (3. 2 This .24) to write as t/2 Fig. Ixx m (t → 0) → 1. The results are present in Figure 3. The ratio of the moment of inertia of ditionally it can be noticed that the ratio two-dimensional to three–dimensional.9.Fig. Calculate the moment of inertia about the center of mass of a square shape with a thickness. a2 /t2 is the only contributor to the error2 . t compare the results to a square shape with zero thickness.24) The total moment of inertia can be obtained by integration of equation (3.

V. There are several ways to approach the calculation (diﬀerent inﬁnitesimal area). First the area inside the parabola calculated as √ b/α dA/2 A=2 0 2(3 α − 1) (b − αξ )dξ = 3 2 b α 3 2 .3.11.10. -3. y = αx2 . Parabola for calculations of moment of inertia. Hint. MOMENT OF INERTIA Example 3. Hint.a) The second integration ( no need to use (3. Solution dx dy y r x 49 2b 2a Fig.10 is a r2 d Izz |dy = y 2 + x2 dy dx = −a 2 3 a y 2 + a3 dy 3 (3. why?) is b Izz = −b 2 3 a y 2 + a3 dy 3 4ab (3.5: Calculate the rectangular moment of Inertia for the rotation trough center in zz axis (axis of rotation is out of the page).11. calculate the area ﬁrst. -3.V. Using this method calculate the entire rectangular.V. Solution For y = b the value of x = Fig.6: Calculate the center of area and moment of inertia for the parabola. Use this area to calculate moment of inertia. depicted in Figure 3. Rectangular Moment of inertia. construct a small element and build longer build out of the small one. b/α. The moment of inertia for a long element with a distance y shown in Figure 3.b) Results in Izz = Or a 2 a b3 + 2 a3 b = A 3 End Solution (2a)2 + (2b)2 12 (3.c) Example 3.20).3.

12. Utilizing equation (3. Triangle for example 3. .7. What is the moment inertia when h −→ 0.7: Calculate the moment of inertia of strait angle triangle about its y axis as shown in the Figure on the right.27) The moment of inertia of the area about the center can be found using in equation (3.6). What is the moment when a symmetrical triangle is attached on left.14) and doing the integration from 0 to maximum y provides dA b Ix Utilizing equation (3. What is the moment inertia when a −→ 0.50 CHAPTER 3. What is the moment when a symmetrical triangle is attached on bottom.20) x =4 0 ξ2 ξ 2 b7/2 dξ = √ α 7 α Ix x A (∆x=xc )2 Ixx = Ix x − A ∆x2 = 4 b7/2 3 α − 1 √ − 3 7 α √ b α 3 2 3αb 15 α − 5 2 or after working the details results in Ixx = b 20 b3 − 14 b2 √ 35 α End Solution Example 3. Solution The right edge line equation can be calculated as x y = 1− h a Y h dy X a Fig. Assume that base is a and the height is h. α ξ 2 + b−αξ the element area is used before and therefore 2 √ 1 xc = A 0 xc b/α (b − αξ 2 ) αξ + 2 2 dA (b − αξ 2 )dξ = 3αb 15 α − 5 (3. -3. REVIEW OF MECHANICS The center of area can be calculated utilizing equation (3. The center of every 2 element is at.27) can be done in two steps ﬁrst calculate the moment of inertia in this coordinate system and then move the coordinate system to center.

End Solution a3 h 2 3. Ix y = A x y dA = A (x + ∆x) (y + ∆y)dA (3. The product of inertia deﬁned as Ix i x j = xi xj dA A (3.3. Here only the product of the area is deﬁned and discussed.30) results in 0 0 Ixy ∆y A x dA x ∆ydA + ∆x A y dA ∆x ydA + ∆x ∆y A Ix y = A x ydA + A ∆x ∆ydA A (3.4 Product of Inertia In addition to the moment of inertia.28) For example.31) A . the product of inertia for x and y axises is Ixy = A x ydA (3. The units of the product of inertia are the same as for moment of inertia.3. MOMENT OF INERTIA or x y = 1− a h 51 Now using the moment of inertia of rectangle on the side (y) coordinate (see example (3. the product of inertia is commonly used.29) Product of inertia can be positive or negative value as oppose the moment of inertia.3)) y 3 h a 1− dy a3 h h = 3 4 0 For two triangles attached to each other the moment of inertia will be sum as The rest is under construction. The calculation of the product of inertia isn’t diﬀerent much for the calculation of the moment of inertia.30) expanding equation (3.3. Transfer of Axis Theorem Same as for moment of inertia there is also similar theorem.

End Solution 3. The total product of inertia is ∆x ∆y A b ′ y y x a ′ x Ix y =0+ a 3 b 3 ab 2 = a2 b2 18 Fig. .34) can be transform into Ix x 0 0 0 Iy y 0 0 0 Iz z (3.35) System which creates equation (3.35) referred as principle system.34) In linear algebra it was shown that for some angle equation (3.5 Principal Axes of Inertia The inertia matrix or inertia tensor is Ixx −Iyx −Izx −Ixy Iyy −Izy −Ixz −Iyz Izz (3. REVIEW OF MECHANICS = Ixy + ∆x ∆y A (3. -3.52 The ﬁnal form is Ix y CHAPTER 3.32) There are several relationships should be mentioned Ixy = Iyx (3.33) Symmetrical area has zero product of inertia because integration of odd function (asymmmertial function) left part cancel the right part. Solution The equation of the line is a y = x+a b The product of inertia at the center is zero. Example 3.8: Calculate the product of inertia of straight edge triangle.3.13. Product of inertia for triangle.

3. ω×(r × ω).36) F= Dt It can be noted that D replaces the traditional d since the additional meaning which be added.38) The velocity. Yet this examination provides a tool to study what happened in the ﬂuid during operation of the forces. The acceleration is divided into three categories: Centrifugal. NEWTON’S LAWS OF MOTION 53 3. The surface forces are forces that act on the surface of the body (pressure. The external forces are typically divided into two categories: body forces and surface forces.41) . The body forces are forces that act from a distance like magnetic ﬁeld or gravity. stresses).4 Newton’s Laws of Motion These laws can be summarized in two statements one. Yet. F= ρ rdV V (3.36) can be transformed to a continuous form as D (ρ U ) F= dV (3. 2 (Ur × ω). for every action by body A on Body B there is opposite reaction by body B on body A.4. is deﬁned as L = r × Udm (3. The radial velocity is denoted as Ur . These small “bodies” when became small enough equation (3.40) The angular momentum of the entire system is calculated by integration (summation) of all the particles in the system as Ls = m r × U dm (3. Two. Coriolis. ˙ 3. the derivative can be taken out of the integral and the alternative form can be written as F= D Dt D2 Dt2 ρ U dV V (3.37) Dt V The external forces are equal to internal forces the forces between the “small” bodies are cancel each other.39) where r is the location of the particles from the origin. The same as in the dynamic class. Since the derivative with respect to time is independent of the volume. U is a derivative of the location with respect to time. dm.5 Angular Momentum and Torque The angular momentum of body. the system acceleration called the internal forces. it can be treated as the regular derivative. thus. This law apply to any body and any body can “broken” into many small bodies which connected to each other. which can expressed in mathematical form as D (m U ) (3. r × ω. Angular.

The force can be written. The torque of entire system is Tτ s = D DL = Dt Dt (r × Udm) m (3. in analogous to the momentum change of time which is the force.48) In the same way the component in y and z can be obtained.47) The torque is a vector and the various components can be represented as Tτ x = ˆ • i D Dt r × U dm m (3. in the same fashion. as F = Fxˆ + Fy ˆ Utilizing equation i j. (3.45) u v 0 Utilizing equation (3. it can be proved utilizing vector mechanics) that Tτ = D D Dr D2 r (r × U) = (r × )= Dt Dt Dt Dt2 (3. consider a particle moving in x–y plane. Tτ = DL D = (r × Udm) Dt Dt (3. .44) To understand these equations a bit better. REVIEW OF MECHANICS The change with time of angular momentum is called torque.40) provides ˆ ˆ k i j ˆ ˆ L = r × U = x y 0 = (x v − y u)k (3.1 Tables of geometries Th following tables present several moment of inertias of commonly used geometries. The velocity can be written as U = uˆ + vˆ and the location from the origin can be written as r = xˆ + yˆ i j i j.43) m It can be noticed (well.42) to calculate the torque as ˆ ˆ k i j ˆ ˆ Tτ = r × F = x y 0 = (x Fx − y Fy )k Fx Fy 0 (3.54 CHAPTER 3.46) Since the torque is a derivative with respect to the time of the angular momentum it is also can be written as xFx − yFy = D [(xv − yu) dm] Dt (3. A force is acting on the particle in the same plane (x–y) plane.5. 3.42) where Tτ is the torque.

5.3. 2 2 ab ab3 12 XX Triangle b b/3 a a 3 ab 3 ab3 36 XX Circle a=b b b/2 b 2 π b2 4 πb4 64 a Ellipse XX a>b b b/2 b b 2 2 π ab 4 Ab2 64 a y = αx2 Parabola a XX b xc 3αb 15 α−5 6α−2 3 × 3 b 2 α √ b (20 b3 −14 b2 ) √ 35 α .1. yc A Ixx XX Rectangle b b/2 a b a . ANGULAR MOMENTUM AND TORQUE 55 Table -3. Moments of Inertia for various plane surfaces about their center of gravity (full shapes) Shape Name Picture description xc .

REVIEW OF MECHANICS Table -3.56 CHAPTER 3. yc A Ixx Quadrant of Circle XX 4r 3π 4r 3π π r2 4 4 π r 4 ( 16 − 9π ) r Ellipsoidal Quadrant XX b 4b 3π 4b 3π πab 4 π 4 a b3 ( 16 − 9π ) a Half of Elliptic XX b 4b 3π 4b 3π πab 4 π 4 a b3 ( 16 − 9π ) a Circular Sector XX α α 0 2α r2 r4 4 (α− 1 sin 2α) 2 r XX Circular Sector α α 2 r sin α 3 α 2 r sin α 3 α Ix 2α r2 r4 4 x = r (α+ 1 sin 2α) 2 .2. Moment of inertia for various plane surfaces about their center of gravity Shape Name Picture description r xc .

and dz is motionless in the accelerated system. Later. y. Fig. gG (x. The system is in a body force ﬁeld. a = 0. a as shown in Figure 4. The changes of the second derivative pressure are not signiﬁcant compared to the ﬁrst derivative (∂P/∂n × d >> ∂ 2 P/∂n2 ). the methods discussed here will be expanded to more complicated dynamics situations. that pressure can increase and later decrease.1.1 Introduction The simplest situation that can occur in the study of ﬂuid is when the ﬂuid is at rest or queasy rest.1) can be reduced and simpliﬁed for the case of no acceleration. several assumptions must be made. -4. The combination of an acceleration and the body force results in eﬀective body force which is gG − a = geﬀ y P P+ ∂P dy dxdz ∂y P+ dy ∂P dz dxdy ∂z P+ ∂P dx dydz ∂x dz dx z P x (4. There is no requirement that the pressure has to be a monotonous function e. Furthermore.1) erated system under body forces.g. the student will be exposed to stability analysis probably for the ﬁrst time. In these derivations. where n is the steepest 57 . dy. However. 4.2 The Hydrostatic Equation A ﬂuid element with dimensions of DC. z). here this topic will be more vigorously examined.1.CHAPTER 4 Fluids Statics 4. Description of a ﬂuid element in accel- Equation (4. This topic was introduced to most students in previous study of rigid body. with acceleration. The ﬁrst assumption is that the change in the pressure is a continuous function.

2) In the same fashion. The net pressure force on the faces in the x direction results in dF = − ∂P ∂x dydx ˆ i (4. now. If the pressure. FLUIDS STATICS direction of the pressure derivative and d is the inﬁnitesimal length. P . the eﬀective gravity force is utilized in case where the gravity is the only body force and in an accelerated system.58 CHAPTER 4.7) . Even though. it has a direction). as a scalar function (there no reference to the shear stress in part of the pressure) the gradient is a vector. the dot product of the following is i · gradP = i · P = ∂P ∂x (4.6) Hence.8) (4. the calculations of the three directions result in the total net pressure force as F =− surface ∂P ˆ ∂P ˆ ∂P ˆ i+ j+ k ∂x ∂y ∂y (4. the utilizing the above derivations one can obtain −gradP dx dy dz + ρ geﬀ dx dy dz = 0 or Pressure Gradient gradP = P = ρ geﬀ (4. if the coordinates were to “rotate/transform” to a new system which has a diﬀerent orientation the dot product results in in · gradP = in · P = ∂P ∂n (4. As before. This mathematical statement simply requires that the pressure can deviate in such a way that the average on inﬁnitesimal area can be found and expressed as only one direction.3) referred to in the literature as the pressure gradient (see for more explanation in the Mathematics Appendix). the pressure is treated.4) In general. The body (element) is in rest and therefore the net force is zero F= total surface F+ body F (4. was a two–dimensional height (that is only a function of x and y) then the gradient is the steepest ascent of the height (to the valley). This mathematical operation has a geometrical interpretation.3) The term in the parentheses in equation (4. The second point is that the gradient is a vector (that is.5) where in is the unit vector in the n direction and ∂/∂n is a derivative in that direction. For example.

it will be used. the z coordinate is used as the (negative) direction of the gravity1 .12) (4. The eﬀective body force is ˆ geﬀ = −g k (4. These equations are ∂P ∂P = =0 ∂x ∂y and Pressure Change ∂P = −ρ g ∂z Equations (4. if at point z0 the pressure is P0 then the equation (4. For this reason sometime there will be a deviation from the above statement.12) can be absorbed by the integration of equation (4.3 Pressure and Density in a Gravitational Field In this section.13) becomes P (z) − P0 = −ρg(z − z0 ) (4. 4.3.10) and constant in equation (4.14) 1 This situation were the tradition is appropriated.3. For example. . a discussion on the pressure and the density in various conditions is presented. pressure. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD 59 Some refer to equation (??) as the Fluid Static Equation. ρ. 4. z) = −ρgz + constant (4.11) (4. P . and temperature. First.10) can be integrated to yield P (x. a discussion on a simple condition and will continue in more challenging situations.9) Utilizing equation (4.9) and substituting it into equation (??) results into three simple partial diﬀerential equations.4. T (in a way no function of the location) are constant. there are several physical implications to this equation which should be discussed and are presented here. This equation can be integrated and therefore solved. Traditionally.11) and therefore P (x.13) The integration constant is determined from the initial conditions or another point. However. There are ﬁelds where x or y are designed to the direction of the gravity and opposite direction. y) = constant (4. y.1 Constant Density in Gravitational Field The simplest case is when the density.

60

CHAPTER 4. FLUIDS STATICS

Constant Pressure Lines

Fig. -4.2. Pressure lines in a static ﬂuid with a constant density.

It is evident from equation (4.13) that the pressure depends only on z and/or the constant pressure lines are in the plane of x and y. Figure 4.2 describes the constant ρgh pressure lines in the container under the gravity body force. The pressure lines are continuous even in area where there is a discontinuous ﬂuid. The reason that a a solid boundary doesn’t break the continuity of the pressure lines is because there is always a path to some of the planes. It is convenient to reverse the direction of z to get rid of the negative sign and Fig. -4.3. A schematic to explain the measure to deﬁne h as the dependent of the ﬂuid of the atmospheric pressure. that is h ≡ −(z − z0 ) so equation (4.14) becomes Pressure relationship P (h) − P0 = ρgh (4.15)

In the literature, the right hand side of the equation (4.15) is deﬁned as piezometric pressure.

4.3.2

4.3.2.1

Pressure Measurement

Measuring the Atmospheric Pressure

One of the application of this concept is the idea of measuring the atmospheric pressure. Consider a situation described in Figure 4.3. The liquid is ﬁlling the tube and is brought into a steady state. The pressure above the liquid on the right side is the vapor pressure. Using liquid with a very low vapor pressure like mercury, will result in a device that can measure the pressure without additional information (the temperature).

4.3. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD

61

Example 4.1: Calculate the atmospheric pressure at 20◦ C. The high of the Mercury is 0.76 [m] and the gravity acceleration is 9.82[m/sec]. Assume that the mercury vapor pressure is 0.000179264[kPa]. The description of the height is given in Figure 4.3. The mercury density is 13545.85[kg/m3 ]. Solution The pressure is uniform or constant plane perpendicular to the gravity. Hence, knowing any point on this plane provides the pressure anywhere on the plane. The atmospheric pressure at point a is the same as the pressure on the right hand side of the tube. Equation (4.15) can be utilized and it can be noticed that pressure at point a is Pa = ρ g h + Pvapor The density of the mercury is given along with the gravity and therefore, Pa = 13545.85 × 9.82 × 0.76 ∼ 101095.39[P a] ∼ 1.01[Bar] The vapor pressure is about 1 × 10−4 percent of the total results.

End Solution

(4.16)

The main reason the mercury is used because its of large density and the fact that it is in a liquid phase in most of the measurement range. The third reason is the low vapor (partial) pressure of the mercury. The partial pressure of mercury is in the range of the 0.000001793[Bar] which is insigniﬁcant compared to the total measurement as can be observed from the above example.

Gas The pressure, P valve

2

1

Example 4.2: A liquid2 a in amount Ha and a liquid b in amount Hb in to an U tube. The ratio of the Fig. -4.4. Schematic of gas measurement liquid densities is α = ρ1 /ρ2 . The width of the utilizing the “U” tube. U tube is L. Locate the liquids surfaces. Solution The question is to ﬁnd the equilibrium point where two liquids balance each other. If the width of the U tube is equal or larger than total length of the two liquids then the whole liquid will be in bottom part. For smaller width, L, the ratio between two sides will be as ρ1 h1 = ρ2 h2 → h2 = α h1

2 This

example was requested by several students who found their instructor solution unsatisfactory.

h

62 The mass conservation results in

CHAPTER 4. FLUIDS STATICS

Ha + Hb = L + h 1 + h 2 Thus two equations and two unknowns provide the solution which is h1 = Ha + Hb − L 1+α

When Ha > L and ρa (Ha − L) ≥ ρb (or the opposite) the liquid a will be on the two sides of the U tube. Thus, the balance is h1 ρb + h2 ρa = h3 ρa where h1 is the height of liquid b where h2 is the height of “extra” liquid a and same side as liquid b and where h3 is the height of liquid b on the other side. When in this case h1 is equal to Hb . The additional equation is the mass conservation as Ha = h 2 + L + h 3 The solution is h2 = (Ha − L) ρa − Hb ρb 2 ρa

End Solution

4.3.2.2

Pressure Measurement

The idea describes the atmoh1 P1 P2 A1 A1 spheric measurement can be exρ1 ρ1 tended to measure the pressure gas chambers. Consider a chamber ﬁlled with gas needed to be ρ1 measured (see Figure 4.4). One A2 h2 ρ2 technique is to attached “U” tube ρ2 to the chamber and measure the ρ2 pressure. This way, the gas is prevented from escaping and its pressure can be measured with a min- Fig. -4.5. Schematic of sensitive measurement device. imal interference to the gas (some gas enters to the tube). The gas density is signiﬁcantly lower than the liquid density and therefore can be neglected. The pressure at point “1” is P1 = Patmos + ρg h (4.17)

Since the atmospheric pressure was measured previously (the technique was shown in the previous section) the pressure of the chamber can be measured.

4.3. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD 4.3.2.3 Magniﬁed Pressure Measurement

63

For situations where the pressure diﬀerence is very small, engineers invented more sensitive measuring device. This device is build around the fact that the height is a function of the densities diﬀerence. In the previous technique, the density of one side was neglected (the gas side) compared to other side (liquid). This technique utilizes the opposite range. The two sides of the densities are very close to each other so the height became large. Figure 4.5 shows a typical and simple schematic of such an instrument. If the pressure diﬀerences between P1 and P2 is small this instrument can “magniﬁed” height, h1 and provide “better” accuracy reading. This device is based on the following mathematical explanation. In steady state, the pressure balance (only diﬀerences) is P1 + g ρ1 (h1 + h2 ) = P2 + g h2 ρ2 (4.18)

It can be noticed that the “missing height” is canceled between the two sides. It can be noticed that h1 can be positive or negative or zero and it depends on the ratio that two containers ﬁlled with the light density liquid. Additionally, it can be observed that h1 is relatively small because A1 >> A2 . The densities of the liquid are chosen so that they are close to each other but not equal. The densities of the liquids are chosen to be much heavier than the measured gas density. Thus, in writing equation (4.18) the gas density was neglected. The pressure diﬀerence can be expressed as P1 − P2 = g [ρ2 h2 − ρ1 (h1 + h2 )] (4.19)

If the light liquid volume in the two containers is known, it provides the relationship between h1 and h2 . For example, if the volumes in two containers is equal then −h1 A1 = h2 A2 −→ h1 = − h2 A2 A1 (4.20)

Liquid volumes do not necessarily have to be equal. Additional parameter, the volume ratio, will be introduced when the volumes ratio isn’t equal. The calculations as results of this additional parameter does not cause signiﬁcant complications. Here, this ratio equals to one and it simplify the equation (4.20). But this ratio easily can be inserted into the derivations. With the equation for height (4.20) equation (4.18) becomes P1 − P2 = g h2 ρ2 − ρ1 1 − or the height is h2 = P1 − P2 g (ρ2 − ρ1 ) + ρ1 A2 A1 (4.22) A2 A1 (4.21)

For the small value of the area ratio, A2 /A1 << 1, then equation (4.22) becomes h2 = P1 − P2 g (ρ2 − ρ1 ) (4.23)

64

CHAPTER 4. FLUIDS STATICS

Some refer to the density diﬀerence shown in equation (4.23) as “magniﬁcation factor” since it replace the regular density, ρ2 . Inclined Manometer One of the old methods of pressure measurement is the inclined manometer. In this method, the tube leg is inclined relatively to gravity (depicted in Figure 4.6). This method is attempt to increase the accuracy by “extending” length visible of the tube. The equation (4.17) is then

Poutside P1 dy θ dℓ

Fig. -4.6. caption.

P1 − Poutside = ρ g d

(4.24)

If there is insigniﬁcant change in volume (the area ratio between tube and inclined leg is signiﬁcant), a location can be calibrated on the inclined leg as zero3 . Inverted U-tube manometer The diﬀerence in the pressure of two diﬀerent liquids is measured by this manometer. This idea is similar to “magniﬁed” manometer but in reversed. The pressure line are same for the both legs on line ZZ. Thus, it can be written the pressure on left is equal to pressure on the right legs (see Figure 4.7).

right leg left leg

Z

Z

h

P2 − ρ2 (b + h) g = P1 − ρ1 a − ρ h) g Rearranging equation (4.25) leads to P2 − P1 = ρ2 (b + h) g − ρ1 a g − ρ h g

(4.25)

a

b

2

(4.26)

1

For the similar density of ρ1 = ρ2 and for a = b equation (4.26) becomes P2 − P1 = (ρ1 − ρ) g h (4.27)

Fig. -4.7. Schematic of inverted manometer.

As in the previous “magniﬁed” manometer if the density diﬀerence is very small the height become very sensitive the change of pressure.

4.3.3

Varying Density in a Gravity Field

There are several cases that will be discussed here which are categorized as gases, liquids and other. In the gas phase, the equation of state is simply the ideal gas model

3 The personal experience of this author while working in ship that this manometer is signiﬁcantly inaccurate (to be the ﬁrst thing to be replaced on the ship). Due to surface tension cause air entrapment especially in rapid change of the pressure or height.

4.3. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD

65

or the ideal gas with the compressibility factor (sometime referred to as real gas). The equation of state for liquid can be approximated or replaced by utilizing the bulk modulus. These relationships will be used to ﬁnd the functionality between pressure, density and location. 4.3.3.1 Gas Phase under Hydrostatic Pressure

Ideal Gas under Hydrostatic Pressure The gas density vary gradually with the pressure. As ﬁrst approximation, the ideal gas model can be employed to describe the density. Thus equation (4.11) becomes ∂P gP =− ∂z RT (4.28)

Separating the variables and changing the partial derivatives to full derivative (just a notation for this case) results in dP g dz =− P RT Equation (4.29) can be integrated from point “0” to any point to yield ln P g =− (z − z0 ) P0 RT

„ «

(4.29)

(4.30)

It is convenient to rearrange equation (4.30) to the following P = P0

e

−

g(z−zo ) RT

(4.31)

Here the pressure ratio is related to the height exponentially. Equation (4.31) can be expanded to show the diﬀerence to standard assumption of constant pressure as

−

h ρ0 g P0

P (z − z0 ) g (z − z0 ) g =1− + + ··· P0 RT 6RT Or in a simpliﬁed form where the transformation of h = (z − z0 ) to be correction factor h2 P ρ0 g h − + ··· =1+ 6 P0 P0

2

(4.32)

(4.33)

**Equation (4.33) is useful in mathematical derivations but should be ignored for practical use4 .
**

4 These derivations are left for a mathematical mind person. These deviations have a limited practical purpose. However, they are presented here for students who need to answer questions on this issue.

66

CHAPTER 4. FLUIDS STATICS

Real Gas under Hydrostatic Pressure The mathematical derivations for ideal gas can be reused as a foundation for the real gas model (P = ZρRT ). For a large range of P/Pc and T /Tc , the value of the compressibility factor, Z, can be assumed constant and therefore can be swallowed into equations (4.31) and (4.32). The compressibility is deﬁned in equation (2.39). The modiﬁed equation is P = P0 Or in a series form which is P (z − z0 ) g (z − z0 ) g =1− + + ··· P0 Z RT 6Z RT

2

e

„

−

g (z−zo ) Z RT

«

(4.34)

(4.35)

Without going through the mathematics, the ﬁrst approximation should be noticed that the compressibility factor, Z enter the equation as h/Z and not just h. Another point that is worth discussing is the relationship of Z to other gas properties. In general, the relationship is very complicated and in some ranges Z cannot be assumed constant. In these cases, a numerical integration must be carried out. 4.3.3.2 Liquid Phase Under Hydrostatic Pressure

The bulk modulus was deﬁned in equation (1.28). The simplest approach is to assume that the bulk modulus is constant (or has some representative average). For these cases, there are two diﬀerential equations that needed to be solved. Fortunately, here, only one hydrostatic equation depends on density equation. So, the diﬀerential equation for density should be solved ﬁrst. The governing diﬀerential density equation (see equation (1.28)) is ρ = BT ∂ρ ∂P (4.36)

**The variables for equation (4.36) should be separated and then the integration can be carried out as
**

P ρ

dP =

P0 ρ0

BT

dρ ρ

(4.37)

The integration of equation (4.37) yields P − P0 = BT ln ρ ρ0 (4.38)

Equation (4.38) can be represented in a more convenient form as Density variation ρ = ρ0

e

P −P0 BT

(4.39)

8. the solution is presented as P BT = ln P0 P0 g ρ0 z +1 +1 BT (4. is a typical length of the problem.3. If at z = 0 the pressure is P0 and the density is ρ0 then the constant is Constant = This constant. The exponent can be expanded as piezometric corrections pressure (P − P0 ) + BT 2 P − P0 BT 2 + BT 6 P − P0 BT 3 + ··· = z g ρ0 (4. Additional discussion will be presented in the dimensionless issues chapter (currently under construction).8.45) An approximation of equation (4. The solution becomes BT g ρ0 BT g ρ0 (4.41) It can be noted that BT has units of pressure and therefore the ratio in front of the exponent in equation (4. with units of length.43) March 11.43) and is plotted in Figure 4. -4.39) in equation (4.4. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD 67 Equation (4. Hydrostatic pressure when there is The solution is presented in equation (4. can be evaluated at any speciﬁc point.44) is presented for historical reasons and in order to compare the constant density assumption.39) is the counterpart for the equation of state of ideal gas for the liquid phase. Utilizing equation (4.41) has units of length. Fig.40) e P −P0 BT = z + Constant (4.42) e P −P0 BT P −P0 BT −1 =z (4. Sometimes.11) transformed into ∂P = −gρ0 ∂z Equation (4.40) can be integrated to yield BT g ρ0 e P −P0 BT (4.44) compressibility in the liquid phase. The integration constant. 2008 P −P0 BT Or in a dimensionless form Density in Liquids g ρ0 z BT e −1 = z g ρ0 BT (4. The solution is a reverse function (that is not P = f (z) but z = f (P)) it is a monotonous function which is easy to solve for any numerical value (that is only one z corresponds to any Pressure).46) . BT /g ρ0 .

52) 5 A colleague asked this author to insert this explanation for his students. A bit more complicate case is when the gas is a function of the pressure and another parameter. it can be observed that the correction is on the left hand side and not as the “traditional” correction on the piezometric pressure side.4 4. Here.4.48) where the Constant is the integration constant which can be obtained by utilizing the initial condition.68 CHAPTER 4.49) with (4. it is commonly assumed that the temperature is a linear function of the height. For the atmosphere. P/BT is small (<< 1).50) (4. when the temperature ﬁeld is not uniform. the temperature is T0 and using it leads to Temp variations T = T0 − Cx h Combining equation (4.51) P R (T0 − Cx h) Deﬁning a new variable5 as ξ = (T0 − Cx h) for which ξ0 = T0 − Cx h0 and d/dξ = −Cx d/dh.49) Separating the variables in equation (4. the temperature–distance function can be written as T = Constant − Cx h (4.50) and changing the formal ∂ to the informal d to obtain dP g dh =− (4. a simple case is examined for which the temperature is a linear function of the height as dT = −Cx (4. Using these deﬁnitions results in dP g dξ = P RCx ξ (4. Additionally.11) results in ∂P gP =− ∂h R (T0 − Cx h) (4. just ignore it. Hence. If you feel that it is too simple.3.). For h = 0. please. For example.47) dh where h here referred to height or distance. 4. FLUIDS STATICS It can be noticed that equation (4. in the atmosphere the temperature is assumed to be a linear with the height under certain conditions. . Air can be a function of the temperature ﬁeld and the pressure.3.1 The Pressure Eﬀects Due To Temperature Variations The Basic Analysis There are situations when the main change of the density results from other eﬀects.46) is reduced to the standard equation when the normalized pressure ratio. the density is aﬀected and thus the pressure is a location function (for example.

The simplest assumption to combine these 6 These concepts are very essential in all the thermo–ﬂuid science. . Equation (4. inﬂuences at only large values of height.e. This situation is roughly representing the pressure in the atmosphere and results in a temperature decrease. The correction factor occurs only at the third term which is important for larger heights. It has to be noted that these equations (4.4.3. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD 69 After integration of equation (4. It can be recalled that the following expansions are g h ρ0 P0 correction factor P Cx = lim 1 − h h−>0 P0 T0 g R Cx =1− R g C x − g 2 h2 gh − − . For engineering purposes. negative direction).53) Or in a more convenient form as Pressure in Atmosphere P = P0 g T0 − Cx h ( R Cx ) T0 (4.56) shows that the correction factor (lapse coeﬃcient). This relationship can be obtained from combining equations (4.55) shows that the ﬁrst two terms are the standard terms (negative sign is as expected i. it is sometimes important to obtain the density ratio.54) and (4. this kind of analysis will be presented in the dimensional analysis chapter6 .51) and reusing (the reverse deﬁnitions) the variables transformed the result into ln P g T0 − Cx h = ln P0 R Cx T0 (4.. However.54) can be approximated by two approaches/ideas.49). It has to be inserted to make the physical signiﬁcance clearer.54) is a monotonous function which decreases with height because the term in the brackets is less than one. Equation (4. (4.55) Equation (4. I am grateful to my adviser E. T0 R 2 T0 2 R2 (4.54) represents only the pressure ratio. and the second approximation for a small temperature gradient. Eckert who was the pioneer of the dimensional analysis in heat transfer and was kind to show me some of his ideas.. The second approximation for small Cx is P Cx = lim h 1− Cx −>0 P0 T0 g R Cx = e gh −RT 0 − g h2 Cx 2 T0 2 R e gh −RT 0 − . It can be observed that Cx has a “double role” which can change the pressure ratio..R..56) are not properly represented without the characteristic height.55) and (4. The ﬁrst approximation for a small distance. Cx . It is worth to point out that the above statement has a qualitative meaning when additional parameter is added. h.G.54) It can be noticed that equation (4.56) Equation (4.

70 CHAPTER 4. the stability question is whether the slab density from layer h. That is. FLUIDS STATICS equations is by assuming the ideal gas model. -4. A weak wind or other disturbances can make the unstable system to move to a new condition. to yield P P0 T0 T g R Cx ρ P T0 = = ρ0 P0 T 1− Cx h ( T0 ) 1+ Cx h T (4. Clearly. . If ρ (h) > ρ(h + dh) then the situation is stable. The term ρ (h) is slab from layer h that had undergone the free expansion. Under equilibrium. the surroundings “pressure” forces (buoyancy forces) are equal to gravity forces. The two processes that occurred here are thermal and the change of pressure (at the speed of sound).9. The buoyancy forces are proportional to the ratio of the density of the slab to surrounding layer density. this shock is insigniﬁcant (check book on Fundamentals of Compressible Flow Mechanics by this author on the French problem). In reality. a small slab of material moves from a layer at height. The ﬁrst case is referred to as the stable condition and the second case referred to as the unstable condition.57) Advance material can be skipped 4. h. main possibilities one: the slab could return to the original layer or two: stay at the new layer (or even move further. The reason that the free expansion is chosen to explain the process that the slab undergoes when it moves from layer h to layer h + dh is because it is the simplest. This question is determined by the net forces acting on the slab. the slab is in equilibrium with its surroundings before the movement (not necessarily stable). Thus.3. the body forces that acting on the slab are equal to zero. the pressure process is about thousands times faster then the thermal process.2 The Stability Analysis It is interesting to study whether h + dh this solution (4. The second issue that occurs during the “expansion” is the shock (in the reverse case [h + dh] → h). Suppose that h for some reason. The thermal process is in the range of [cm/sec] while the speed of sound is about 300 [m/sec]. equation (2.54) is stable and if so under what conditions. However. The two forces that act on the slab are the gravity force and the surroundings pressure (buoyant forces). The whole system falls apart and does not stay if the analysis predicts unstable conditions. Whether these forces are toward the original layer or not. ρ (h) undergoing a free expansion is higher or lower than the density of the layer h + dh.9) What could happen? There are two Fig. to layer at height h + dh (see Figure 4.25). Two adjoin layers for stability analysis. higher heights).4. the free expansion is not far way from the actual process. That is.

PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD 71 The slab density at layer h+dh can be obtained using equation (4.3. From a mathematical point of view.63) and (4.. Pk 2 P 2 k2 (4.4.60) Again using the ideal gas model for equation (4.58) The pressure and temperature change when the slab moves from layer at h to layer h + dh.59) When the symbol denotes the slab that moves from layer h to layer h + dh. The process.57) as following ρ(h + dh) P T0 = = ρ(h) P0 T 1− g Cx dh ( R Cx ) Cx dh 1+ T0 T (4.61) in taylor series results in 1− ρ gdh P 1/k =1− g 2 ρ2 k − g 2 ρ2 dh2 g ρ dh − − .54) but can be approximated by equation (4.61) Expanding equation (4..58) and then it is expanded in taylor series as ρ(h + dh) = ρ(h) g Cx dh ( R Cx ) Cx dh 1− 1+ T0 T ∼1− g ρ Cx − P T dh + · · · (4.63) The comparison of the right hand terms of equations (4. under the above discussion and simpliﬁcations.25)) ρ (h + dh) = ρ(h) P (h + dh) P (h) 1/k (4.62) provides the conditions to determine the stability. The pressure ratio is given by equation (4. no signiﬁcant heat transfer occurs in the short period of time). The little slab undergoes isentropic expansion as following for which (see equation (2.61) transformed into ρ (h + dh) = ρ(h) 1− ρ gdh P 1/k (4.64) .55) and thus ρ (h + dh) = ρ(h) 1− gdh T (h) R 1/k (4.62) The density at layer h + dh can be obtained from (4. to keep the inequality for a small dh only the ﬁrst term need to be compared as g ρ Cx gρ > − Pk P T (4. can be assumed to be adiabatic (that is.

Assuming that the pressure is aﬀected by this gravity/body force. r → ∞ the pressure is about zero. Thus. 4.11) can be used 7 The same issue of the ﬂoating ice. ρ and the body force. See example for the ﬂoating ice in cup. This assumption must be deviated when the distance from the body source is signiﬁcantly change.will not be introduced here.5 Gravity Variations Eﬀects on Pressure and Density Until now the study focus on the change of density and pressure of the ﬂuid. The body force was assumed rb g ∝ r2 until now to be constant. Thus. it was explained that the gravity is a function of the distance from the center of the plant/body. around this value additional analysis is needed 7 . g. The gravity force can be assumed that for inﬁnity. . Fig. The varying gravity eﬀects Thus. the density.72 CHAPTER 4. Again. it is said that situation is neutral. The issues of magnetohydrodynamics are too advance for undergraduate student and therefore. One of the common question this author has been asked is about the forces of continuation. while the gravity force source in liquid can be the liquid itself.3. At ﬁrst glance. would the situation become stable now? One has to remember that temperature gradient forces continuous heat transfer which the source temperature change after the movement to the new layer. the body force is independent of the ﬂuid. The gravity force is reversely proportional to r2 . it transformed to Cx (k − 1) g ρ > T kP k−1 g Cx < k R (4. the unstable situation is continuously unstable. Equation (4. one has to bear in mind that this analysis only provides a range and isn’t exact. The source of the gravity force in gas is another body. 4.5.3.10. equation (4. the discussion is separated into two diﬀerent on density and pressure. However.11) has two r P b ρb terms on the right hand side. -4.1 Ideal Gas in Varying Gravity In physics. What is the source of the force(s) that make this situation when unstable continue to be unstable? Supposed that the situation became unstable and the layers have been exchanged. issues. FLUIDS STATICS After rearrangement of the inequality (4. When lapse rate Cx is equal to the right hand side of the inequality.64) and using the ideal gas identity.65) The analysis shows that the maximum amount depends on the gravity and gas properties. It should be noted that this value should be changed a bit since the k should be replaced by polytropic expansion n.

It can be observed that for r → rb the pressure is approaching P → Pb . 4..71) With the same process as before for ideal gas case.67) is transformed into P Pb dP G =− P Z RT r rb dr r2 (4. The integration of equation (4.68) e G r−r − RT r r b b (4.2 Real Gas in Varying Gravity The regular assumption of constant compressibility.69) demonstrates that the pressure is reduced with the distance. PRESSURE AND DENSITY IN A GRAVITATIONAL FIELD (semi one directional situation) when r is used as direction and thus ∂P G = −ρ 2 ∂r r 73 (4. Thus.5.3.3.69) can be expanded in taylor series as standard correction f actor ρ P 2 2 G R T + G2 rb (r − rb ) G (r − rb ) (4. This equation conﬁrms that the density in outer space is zero ρ(∞) = 0.70) = = 1− − + . is employed. ρb Pb 2 RT 2 rb (R T ) Notice that G isn’t our beloved and familiar g and also that G rb /RT is a dimensionless number (later in dimensionless chapter about it and its meaning). It can be noticed that for r → rb the pressure is approaching P → Pb . Z..72) Equation (4.67) results in ln Or in a simpliﬁed form as ρ P = = ρb Pb P G =− Pb RT 1 1 − rb r (4. numerical integration is a possible solution.4. As before. The regular method of separation is employed to obtain P Pb dP G =− P RT r rb dr r2 (4.67) where the subscript b denotes the conditions at the body surface. equation (4.69) Equation (4.66) where here G denotes the general gravity constant. one can obtain P ρ = = ρb Pb e G r−r −Z RT r r b b (4. It has to remember when this assumption isn’t accurate enough.69) demonstrates that the pressure is reduced with the distance. This equation conﬁrms . equation (4.

.76) • + 4 π Gρ = 0 (4. If the liquid “equation of state” (4. there are situations where the spherical coordinates must be considered and used.39) is used with the hydrostatic ﬂuid equation results in ∂P = −ρ0 ∂r which the solution of equation (4.74) e P0 −P BT = Constant − B T g ρ0 r (4.3 Liquid in Under Varying Gravity For comparison reason consider the deepest location in the ocean which is about 11.74 CHAPTER 4.75) Since this author is not aware to which practical situation this solution should be applied.3. ρb Pb 2 Z RT 2 rb (Z R T ) (4. FLUIDS STATICS that the density in outer space is zero ρ(∞) = 0.74) is e P −P0 BT G r2 (4.5.000 [m].3. As before taylor series for equation (4..77) . 4.6 Liquid Phase While for most practical purposes.69) is standard correction f actor ρ P 2 2 G Z R T + G2 rb (r − rb ) G (r − rb ) = = 1− − + . the Cartesian coordinates provides suﬃcient treatment to the problem. Derivations of the ﬂuid static in spherical coordinates are Pressure Spherical Coordinates 1 d r2 dP + 4 π Gρ = 0 r2 dr ρ dr Or in a vector form as 1 P ρ (4. it is left for the reader to apply according to problem. if applicable.73) It can be noted that compressibility factor can act as increase or decrease of the ideal gas model depending on whether it is above one or below one. 4.

This question is one of the traditional question the ﬂuid static and is straight forward. Consider the following example to illustrate the situation.79) and the angle/direction can be obtained from tanβ = a g (4.4. what happened to the liquid surface? What is the relative angle of the liquid surface for a container in an accelerated system of a = 5[m/sec]? Solution 27. The change of the acceleration form the right to left is . Equation (??) can be transformed into a diﬀerent coordinate system where the main coordinate is in the direction of the eﬀective gravity. When the acceleration is changing from the right to the left. 4. The general. FLUID IN A ACCELERATED SYSTEM 75 4.4. in a two dimensional system. The eﬀective angle is obtained by adding vectors. -4. for the eﬀective gravity (4. As before. Thus. the constant pressure plane is perpendicular to the direction of the eﬀective gravity. the linear acceleration have three components as opposed to the previous case of only one. body forces were considered as one-dimensional.1 a 5 m sec g geﬀ Fig.11. the previous derivations can be easily extended.3: A tank ﬁlled with liquid is accelerated at a constant acceleration.80) Perhaps the best way to explain the linear acceleration is by examples. the previous method can be used and there is no need to solve new three (or two) diﬀerent equations. Example 4.4.4 Fluid in a Accelerated System Up to this stage.1 Fluid in a Linearly Accelerated System ˆ gef f = a ˆ + g k i For example.78) where the magnitude of the eﬀective gravity is |gef f | = g 2 + a2 (4. However. The eﬀective gravity is for accelerated cart. Generally the acceleration is divided into two categories: linear and angular and they will be discussed in this order. The solution is obtained by ﬁnding the eﬀective angle body force.

81) ) F (a β The eﬀective body force is acting perpendicu.015[m/sec2 ] End Solution Example 4. End Solution (b) In case of resistance force (either of friction due to the air or resistance in the wheels) reduces the acceleration of the cart. -4.12.812 = 11. lar to the slope. is now tan α = net g − Fm g cosβ Fnet m (4.81 The magnitude of the eﬀective acceleration is |gef f | = 52 + 9.4: Cart partially is ﬁlled with liquid and is sliding on an inclined plane as shown in Figure 4.Fig. Thus. the acceleration in the cart direction is determined from a = g sin β (4. A cart slide on inclined plane. Calculate the shape of the surface. the liquid surface is parallel to the surface of the inclination surface. In that case the eﬀective body moves closer to the gravity forces.01◦ g 9.76 CHAPTER 4. α < β. This angle/direction can be found using the following a 5 tan−1 β = tan−1 = ∼ 27. If there is a resistance what will be the angle? What happen when the slope angle is straight (the cart is dropping straight down)? Solution (a) The angle can be found when the acceleration of the cart is found.12. If there is no resistance.83) (c) . FLUIDS STATICS like subtracting vector (addition negative vector). The net body force depends on the mass of the liquid and the net acceleration is a=g− The angle of the surface.82) (4.

The pressure is uniform in the tank and no pressure diﬀerence can be found.85) can be integrated as ω2 r2 z − z0 = 2g (4. FLUID IN A ACCELERATED SYSTEM In the case when the angle of the inclination turned to be straight (direct falling) the eﬀective body force is zero. -4. Expresses the relationship between the diﬀerent parameters of the problem.84) z r unit mass ω2 r g geﬀ center of circulation Equation (4. So. the ﬁrst case to deal with a rotation in a perpendicular to the gravity. Notice that the integration constant was substituted by z0 . The constant pressure will be along Angular Acceleration System ω2 r2 (4. the pressure at any point in the liquid is the same and equal to the atmospheric pressure. Because the asymmetrical nature of the problem there is diﬀerence in the heights in the U tube arms of S as shown in Figure 4. That is the eﬀective body force can be written as ˆ gef f = −g k + ω 2 r r ˆ The lines of constant pressure are not straight lines but lines of parabolic shape. The center of rotation is a distance. Schematic to explain the angular angle.15. Solution .86) Fig.4. For simpliﬁcation reasons. -4.14. Example 4.4. The angle of the line depends on the radius as dz g =− 2 dr ω r (4. a rf su ce wi th fr o ti ic n 77 α a β g sin β − Fnet m β g geﬀ 4. example 4.13. L from the “left” hand side. Forces diagram of cart sliding on inclined plane.5: A “U” tube with a length of (1 + x)L is rotating at angular velocity of ω.2 Angular Acceleration Systems: Constant Density Fig.4.5 is provided.87) P − P0 = ρg (z0 − z) + 2g To illustrate this point.85) (4.

6: Assume the diameter of the U tube is Rt .15 shown the inﬁnitesimal area used in these calculations. The pressure at the interface at the two sides of the tube is same.86) represent the pressure line. The distance of the inﬁnitesimal area from the rotation center is ?.7: In the U tube in example 4.15.5. Taking the “left” wing of U tube change in z direction zl − z0 The same can be said for the other side zr − z0 = ω 2 x2 L2 2g = change in r direction ω 2 L2 2g Thus subtracting the two equations above from each each other results in zr − zl = L ω 2 1 − x2 2g It can be noticed that this kind equipment can be used to ﬁnd the gravity. End Solution Example 4.5 is rotating with upper part height of . At what rotating nt . equation (4. Thus.78 Calculation of the correction factor dA Rotation center ns ta su pr es re lin e CHAPTER 4. Notice that the curvature in the two sides are diﬀerent from each other. It ﬁrst assumed the height is uniform at the tube (see for the open question on this assumption). What will be the correction factor if the curvature in the liquid in the tube is taken in to account. How would you suggest to deﬁne the height in the tube? Solution In Figure 4. Schematic angular angle to explain example 4. End Solution Example 4. -4. FLUIDS STATICS S ω L co xL Fig. The volume above the lower point is ? which is only a function of the geometry. The height of the inﬁnitesimal area is ?.

8: Consider a rectangular shape gate as shown in Figure 4.4.1 Fluid Forces on Straight Surfaces A motivation is needed before going through the routine of derivations.” The element of moment is a dξ for the width of the gate and is dF "0" β = 50 h ◦ A-A ξ ℓ = 5[m] ξ A-A a[m] dξ dM = P a dξ ( + ξ) dA F2 F1 b[m] The pressure. Later. a simple case will be examined.5. These calculations are divided into two categories. M= 0 g ρ ( + ξ) sin β a dξ( + ξ) The integral can be simpliﬁed as b M = g a ρ sin β 0 ( + ξ)2 dξ (4.88) .16. Calculate the minimum forces. FLUID FORCES ON SURFACES 79 velocity liquid start to exit the U tube? If the rotation of U tube is exactly at the center. what happen the rotation approach very large value? 4. 4. Initially. F1 and F2 to maintain the gate in position. This section deals with these calculations. Rectangular area under pressure. how the calculations can be simpliﬁed will be shown. P can be expressed as a function ξ as the following P = g ρ ( + ξ)sinβ The liquid total moment on the gate is b Fig. Example 4. Solution The forces can be calculated by looking at the moment around point “O. -4.5. straight surfaces and curved surfaces. Assuming that the atmospheric pressure can be ignored.16.5 Fluid Forces on Surfaces The forces that ﬂuids (at static conditions) extracts on surfaces are very important for engineering purposes.

These concepts have been introduced in Chapter 3. which is b Ftotal = 0 g ρ ( + ξ) sin β a dξ The total force integration provides b Ftotal = g ρ a sin β 0 ( + ξ)dξ = g ρ a sin β 2 b + b2 2 The forces on the gate have to provide F1 + F2 = g ρ a sin β 2 b + b2 2 Additionally. it can be observed that there is a moment of area in equation (4.17. "O" β ξ ξ ℓ0 dξ ℓ1 ξ Symmetrical Shapes . Additional equation is needed.88) and also a center of area.Fig. Looking at the above calculations. Several rep. It is the total force. -4. FLUIDS STATICS 3 b l2 + 3 b2 l + b3 3 This value provides the moment that F1 and F2 should extract.80 The solution of the above integral is M = g ρ a sin β CHAPTER 4. These tabulated values can be used to solve this kind of problems. the moment of forces around point “O” is F1 + F2 ( + b) = g ρ a sin β The solution of these equations is F1 = F2 = (3 + b) a b g ρ sin β 6 (3 + 2 b) a b g ρ sin β 6 End Solution 3 b l 2 + 3 b2 l + b3 3 The above calculations are time consuming and engineers always try to make life simpler. and center of area have been tabulated in Chapter 3. Schematic of submerged area to resented areas for which moment of inertia explain the center forces and moments.

-4.5. the atmospheric pressure can include any additional liquid layer above layer “touching” area. is about the axis through point “O” into the page. The total force and moment that the liquid extracting on the area need to be calculated. First. the force is h(ξ) 1 F = A P dA = (Patmos + ρgh)dA = A Patmos + ρg 0 (ξ + 0 ) sin β dA (4. The integral in equation (4. Ix ξ1 x My = Patmos ξ0 ξdA +g ρ sin β ξ0 ξ 2 dA (4. The symmetry is around any axes parallel to axis x.95) .91) "O" y ξ0 β a F1 b My = ξ0 P (ξ)ξdA (4.17.89) In this case.94) The moment of inertia.4. The boundaries of the integral of equation (4. Equation (4. FLUID FORCES ON SURFACES 81 Consider the two–dimensional symmetrical area that are under pressure as shown in Figure 4.94) can be written in more compact form as Total Moment in Inclined Surface My = Patmos xc A + g ρ sin βIx x (4. The “atmospheric” pressure can be set to zero.18.90) ξdA (4.89) can be further developed as Ftotal = A Patmos + ρ g sin β In a ﬁnal form as Total Force in Inclined Surface Ftotal = A [Patmos + ρ g sin β ( 0 + xc )] The moment of the liquid on the area around point “O” is ξ1 0A+ 0 xc A 1 (4.89) refer to starting point and ending points not to the start area and end area.93) F2 Or separating the parts as xc A ξ1 Fig. Ix x .92) ξ1 ξ1 ξ sin β My = ξ0 (Patmos + g ρ h(ξ) )ξdA (4. The general forces acting on submerged area.

101) The integral in equation (4. in non–symmetrical area there two diﬀerent moments and therefor three forces are required. If the “atmospheric pressure” can be zero or include additional layer of liquid.91). and the distance a or b can be positive or negative. Consider the general symmetrical body shown in ﬁgure 4. Ix x ρ sin β (4. the forces can be negative or positive. However.98) ρ sin β − ρ sin β − Patmos ga xc + 0 ρ sin β + Patmos g aA g (b − a) (4.102) The compact form can be written as Moment in Inclined Surface Mx = Patmos A yc + ρ g sin β Ix (4. Equations (4. Thus.8 can be generalized to solve any two forces needed to balance the area/gate. The forces balance reads F1 + F2 = A [Patmos + ρ g sin β ( and moments balance reads F1 a + F2 b = Patmos xc A + g ρ sin βIx The solution of these equations is F1 = and F2 = Ix x x 0 + xc )] (4.95) can be combined the moment and force acting on the general area.103) y .100) Substituting the components for the pressure transforms equation (4.91) and (4. Additionally.97) ρ sin β − Patmos gb xc + 0 ρ sin β + g (b − a) Patmos g b A−.96) (4. additional equation is required.19). In symmetrical area only two forces are required since the moment is one dimensional. The moment around the x axis (which was arbitrary chosen) should be Mx = A y P dA (4.18 which has two forces that balance the body. The moment around the y axis is given by equation (4.82 CHAPTER 4.99) In the solution.95) and the total force is given by (4.100) can be written as A yc Ix y Mx = Patmos A y dA +ρ g sin β A ξ y dA (4.100) into Mx = A y (Patmos + ρ g ξ sin β) dA (4. This equation is for the additional moment around the x axis (see for explanation in Figure 4. FLUIDS STATICS Example 4. the atmospheric pressure can contain either an additional liquid layer above the “touching” area or even atmospheric pressure simply can be set up to zero.

ξ (4.91).9: Calculate the forces that are required to balance the triangular shape shown in the Figure 4.105) + F2 (a + 0) + F3 0 = Mx (4.103) provide the base for solving any problem for straight area under pressure with uniform deny dA sity. Example to illustrate the use of Fig. moment of inertia about its center is Ixx = b( 1 − 0 )3 /36.4. the moment of ∆x2 2 1 A2 Ixx 2 = b[a−( 1− 36 0 )] + b[a−( 1 − 0 )] 3 + [a−( 1 − 0 )] 2 3 .20) for triangle 1 and 2).20.104) (4. These equations (4.103).g. Example 4.95) and (4.5. The moment of inertia of the triangle around x is made of two triangles (as shown in the Figure (4. -4. There are many combinations of x problems (e. two forces and moment) but no general solution is provided.91). The height of triangle 1 is ( 1 − 0 ) and its width b and thus. (4.19. Triangle 1 can be calculated as the moment of inertia around its center which is 0 +2∗( 1 − 0 )/3. Solution The three equations that needs to be solved are F1 + F2 + F3 = Ftotal The moment around x axis is F1 b = My The moment around y axis is F1 1 (4.106) The right hand side of these equations are given before in equations (4. The moment of inertia for triangle 1 about y is A1 ∆x1 2 0 2 Ixx 1 = b( 3 1− 0) 36 + 1 b( 1− 0) 3 0) + 2( 1− 0) 3 The height of the triangle 2 is a − ( inertia about its center is 3 − and its width b and thus.95) and (4. metrical straight area. The general forces acting on non symthese equations is provided. FLUID FORCES ON SURFACES 83 The product of inertia was presented y in Chapter 3.

” Thus. F2 ab 3 =− F3 ab 3 − 72 „„ « « 24 1 48 0 Patmos a −24 + a . pressure centers are commonly deﬁned.20.1.95) and equation (4. The deﬁnition is derived or obtained from equation (4.1 Pressure Center In the literature.107) In the same way.108) . 72 „„ « „ « « 15 12 12 2 a− a 1 + 0 27− a 1 + a 0 g ρ sin β „„ 24 a 72 « 48 1 +24 + a 72 End Solution 1 −14 a)− 0 « « 12 2 −27 + a 0 g ρ sin β = + « 0 Patmos 4.5. 2 a b2 2 1 +2 a b 0 +a 2 + 0 +a Ixy = 0 b 1 − 0 )x + b 0 x y dx dy = b2 24 The solution of this set equations is A F1 = a b (g (6 3 „ (3 1 + 3 a) + 6 g „ 12 a 0) ρ sin β + 8 Patmos 24 1 . the pressure center in the x direction is xp = 1 F x P dA A (4. the pressure center in the y direction is deﬁned as yp = 1 F y P dA A (4. The lower line of the b 0 triangle is y = ( 1 − b −a)x + 0 + a. The pressure center is the distance that will create the moment with the hydrostatic force on point “O. b ( ( 1 − 0 −a)x CHAPTER 4. The general forces acting on non symmetrical straight area. It can be noticed that upper line of the triangle is y = ( 1 − 0 )x + 0 . -4.84 and the total moment of inertia Ixx = Ixx 1 + Ixx 2 The product of inertia of the triangle can be obtain by integration. FLUIDS STATICS ℓ1 y b 1 ℓ0 F3 a F1 2 F2 x Fig.103). These deﬁnitions are mathematical in nature and has physical meaning of equivalent force that will act through this center.

4. FLUID FORCES ON SURFACES 85 To show relationship between the pressure center and the other properties. In fact. . the force that is acting on the body is Ftotal = A g ρ h dA ∼ ρ ¯ A g h dA (4.110) and in the same fashion in y direction yp = Ixy + yc yc A (4.109) according to equation (3. Why? Because of the buoyancy issue.17) results in xp = Ixx + xc xc A (4. the density was assumed to be constant. Thus.5. Consider straight/ﬂat body that is under liquid with a varying density8 . It also means that the density can be a non-continuous function. However. it can be found by setting the atmospheric pressure and 0 to zero as following xp = Expanding Ix x g ρ sin β Ix x A ρ g sin β xc (4. the reader can ﬁnd that direct calculations can sometimes simplify the problem.111) It has to emphasis that these deﬁnitions are useful only for case where the atmospheric pressure can be neglected or canceled and where 0 is zero. if the variation of the density is within 10% and/or the accuracy of the calculation is minimal.5. the reasonable average can be used.114) Ftotal = g sin β ρ1 ξ dA +ρ2 A1 A2 ξ dA + · · · + ρn An ξ dA 8 This statement also means that density is a monotonous function. but constant in segments. In cases where density is non–continuous. the integral has be carried out.2 Multiply Layers In the previous sections. the following can be written x c A1 xc A2 1 2 xc n An (4. 4. these limitations diminish the usefulness of pressure center deﬁnitions. 9 A qualitative discussion on what is reasonably is not presented here.1.113) As before for single density. the following can be said Ftotal = A g ρ h dA = A1 g ρ1 h dA + A2 g ρ2 h dA + · · · + An g ρn h dA (4.112) In cases where average density cannot be represented reasonably9 . For non constant density the derivations aren’t “clean” but are similar. If density can be represented by average density.

ρi is the density of the layer i and Ai and xc i are geometrical properties of the area which is in contact with that layer. FLUIDS STATICS Or in a compact form and in addition considering the “atmospheric” pressure can be written as Total Static Force n Ftotal = Patmos Atotal + g sin β i=1 ρi xc i Ai (4.10: Consider the hypothetical Figure 4. h2 = 2[m]. h3 = 3[m]. the atmospheric pressure can include all the layer(s) that do(es) not with the “contact” area.21. and b1 = 4.115) where the density.and h4 = 4[m]. Moreover. one can ﬁnd that n My = g sin β i=1 ρi Ix x i (4.86 CHAPTER 4.118) In the same fashion one can obtain the moment for x axis as Total Static Moment n Mx = Patmos yc Atotal + g sin β i=1 ρi I x y i (4. My under the same considerations as before is My = A g ρ ξ 2 sin β dA (4. ρ3 = 850[kg/m3 ]. a2 = 1. .116) After similar separation of the total integral.75[m].117) If the atmospheric pressure also enters into the calculations one can ﬁnd that Total Static Moment n My = Patmos xc Atotal + g sin β i=1 ρi I x x i (4. The densities are ρ1 = 500[kg/m3 ]. Calculate the forces at points a1 and b1 . The atmospheric pressure can be entered into the calculation in the same way as before. Assume that the layers are stables without any movement between the liquids. Also neglect all mass transfer phenomena that may occur. Example 4.5[m]. The heights are: h1 = 1[m]. The angle of inclination is is β = 45◦ . The moment around axis y.119) To illustrate how to work with these equations the following example is provided. The last layer is made of water with density of 1000[kg/m3 ]. ρ2 = 800[kg/m3 ].5[m]. and ρ4 = 1000[kg/m3 ]. The forces distances are a1 = 1.

only two equations are needed. Equation (4. the ﬁrst equation is Atotal 3 F1 + F2 = Patmos (b2 − a2 ) +g sin β i=1 ρi+1 xc i Ai The second equation is (4. -4. FLUID FORCES ON SURFACES Solution ρ1 y "O" 87 β h1 a2 Since there are only two unh4 knowns. which are (4. of using the regular atmospheric forces. The solution method of this example is applied for cases with less layers (for example by setting the speciﬁc height difρ4 ference to be zero).118) and (4. the following equations are obtained Thus.118) to be written for the moment around the point “O” as xc Atotal 3 F1 a1 + F2 b1 = Patmos (b2 + a2 ) (b2 − a2 ) +g sin β ρi+1 Ix 2 i=1 x i The solution for the above equation is 2 b1 g sin β P3 i=1 F1 = ρi+1 xc i Ai −2 g sin β 2 b1 −2 a1 P3 i=1 ρi+1 Ix x i − (b2 2 −2 b1 b2 +2 a2 b1 −a2 2 ) 2 b1 −2 a1 Patmos . as can be noticed that instead Fig.118) can be used by modifying it.115).21. pressure the new “atmospheric” pressure can be used as ρ2 ρ3 ρ4 h3 h2 b2 b1 F2 F1 a1 ℓ The eﬀects of multi layers density on static Patmos = Patmos + ρ1 g h1 The distance for the center for each area is at the middle of each of the “small” rectangular. The geometries of each areas are „ xc 1 = xc2 = xc3 = h2 a2 + sin β 2 h2 +h3 2 sin β h3 +h4 2 sin β A1 = A2 = A3 = h2 sin β sin β sin β − a2 Ix x 1 = x 2 x 3 h2 sin β −a2 36 «3 + (xc 1 ) A1 + (xc2 ) A2 + (xc3 ) A3 2 2 2 (h3 − h2 ) (h4 − h3 ) Ix Ix = = (h3 −h2 )3 36 sin β (h4 −h3 )3 36 sin β After inserting the values.5.4.

88 2 g sin β P3 i=1 CHAPTER 4. [m2 ].696 A2 = 3.92[N ] End Solution 4. A mini source code for the calculations is provided in the the text source. the conventional notation is used which is to denote the area. The total force on the area will be the integral of the unit force F=− A y dAz x Fig. is simply the integral of the area perpendicular to y as . the pressure is treated as a scalar function.2892 xc2 = 3. The forces on curved area.121) The result of the integral is a vector. dA. for example.122) From this analysis (equation (4. [m4 ]) are: xc1 = 2.5355 xc3 = 4.292 Ix x 3 = 86. only a dot product is needed as dFy = dF • ˆ j (4.5. At this stage. outward as positive. if the y component of the force is needed. So.535 Ix x 1 = 14. -4.718 The ﬁnal answer is F1 = 304809.22.9497 A1 = 2. The element force is d F = −P n dA ˆ (4.122)) it can be observed that the force in the direction of y.535 A3 = 3.120) z dAy dAx dA Here.215 Ix x 2 = 44.79[N ] and F2 = 958923. It is simpler to compute the terms separately. FLUIDS STATICS ρi+1 Ix x i F2 = −2 a1 g sin β 2 b1 −2 a1 P3 i=1 ρi+1 xc i Ai + (b2 2 +2 a1 b2 +a2 2 −2 a1 a2 ) 2 b1 −2 a1 Patmos The solution isn’t provided in the complete long form since it makes things messy.2 Forces on Curved Surfaces The pressure is acting on surfaces perpendicular to the direction of the surface (no shear forces assumption). The intermediate results in SI units ([m]. P n dA ˆ (4.

For example. Thus. Cut–Out Shapes Eﬀects There are bodies with a shape that the vertical direction (z direction) is “cut– out” aren’t continuous. However. Figure 4. Additional point that is worth mentioning is that the depth where the cut–out occur is insigniﬁcant (neglecting the change in the density).124) implicitly means that the net force on the body is z direction is only the actual liquid above it.123) The same can be said for the x direction. “y z” is tan θzy = (4. -4. Inside the slot. This force component can be combined with the other components in the other directions to be Ftotal = Fz 2 + Fx 2 + Fy 2 (4. the atmospheric pressure with it piezometric pressure is canceled by the upper part of the slot. only the net force is the actual liquid in the slot which is acting on the body.124) The force which acting on the z direction is the weight of the liquid above the projected area plus the atmospheric pressure.127) body.126) and the angle in the other plane. for some geometries there are readily calculated center of mass and when combined with two other components provide the moment (force with direction line). The force in the z direction is Fz = A h g ρdAz (4. Equation (4. Schematic of Net Force on ﬂoating The moment due to the curved surface require integration to obtain the value.125) only the liquid above the body affecting the body And the angle in “x z” plane is tan θxz = Fz Fx Fz Fy (4. Fig. FLUID FORCES ON SURFACES 89 Fy = A P dAy (4.5.11: . The atmospheric pressure acts on the area with continuous lines.4.23. There are no readily made expressions for these 3–dimensional geometries.23 shows a ﬂoating body with cut–out slot into it. Example 4.

The diﬀerential area that will be used is. -4. b r dθ where b is the width of the dam (into the page). You can assume that the liquid density is constant and equal to 1000 [kg/m3 ]. Solution The force in the x direction is CHAPTER 4. The atmospheric pressure does cancel itself (at least if the atmospheric pressure on both sides of the dam is the same. The pressure is only a function of θ and it is P = Patmos + ρ g r sin θ The force that is acting on the x direction of the dam is Ax × P . . why? Fig.25) as Ax xc A△ = r2 sin θ cos θ Aarc = r θ r2 2 Fx = ρ g b r sin θ0 r sin θ0 ρgbr = sin2 θ 2 2 Notice that dAx (cos θ) and Ax (sin θ) are different. The gravity is 9.). dAx Fx = A P r cos θ dθ (4. Calculations of forces on a circular shape dam.90 Calculate the force and the moment around point “O” that is acting on the dam (see Figure (4. -4.25.128) Note that the direction of the area is taken into account (sign). direct and indirect.24)). Ax (see Figure 4.8[m/sec2 ] and width of the dame is b = 4[m]. When the area Ax is b r dθ cos θ. The dam is made of an arc with the angle of θ0 = 45◦ and radius of r = 2[m]. Area above the dam arc subtract triangle. Compare the diﬀerent methods of computations. The net force will be θ0 P dAx Fx = 0 ρ g r sin θ b r cos θ dθ results in The Fx = integration ρ g b r2 1 − cos2 (θ0 ) 2 Alternative way to do this calculation is by calculating the pressure at mid point and then multiply it by the projected area.24. FLUIDS STATICS δθ θ θ0 θ Y 4[m] x direction A θ Ax Ay Fig.

348[m] 10 Well. -4.26.26) is at yc arc = 4 r sin2 3θ θ 2 All the other geometrical values are obtained from Tables 3. just a demonstration! .65174 ∼ 0.4.1 and 3. The arc center (see Figure 4.2.65174[m] The result of the arc center from point “O” (above calculation area) is Fig. Some mathematics are required because the shift in the arc orientation.216[N ] 2 2 The center area ( purple area in Figure 4. clearly the discussion earlier was right (not a good proof LOL10 ). V A θ r2 r2 sin θ0 cos θ0 0 − Fy = − b g ρ ∼ 22375. The force in the y direction is the area times width.0[N ] 2 Since the last two equations are identical (use the sinuous theorem to prove it sin2 θ + cos2 = 1).25) should be calculated as yc = yc Aarc − yc Atriangle A The center area above the dam requires to know the center area of the arc and triangle shapes. FLUID FORCES ON SURFACES 91 The values to evaluate the last equation are provided in the question and simplify subsidize into it as Fx = 1000 × 9. Area above the dam arc calculation for the center. and substituting the proper values results in Aarc θ r2 2 yc yc Atriangle 4 r sin θ 2 cos θ 2 3θ 4 r sin yc r = 2 r cos θ sin θ r2 − 3θ 3 2 θ r2 r2 sin θ cos θ − 2 2 cos Aarc Atriangle θ 2 θ 2 θ 4 r sin 3θ θ 2 This value is the reverse value and it is yc r = 1.8 × 4 × 2 sin(45◦ ) = 19600.5. yc = r − yc r = 2 − 1.

FLUIDS STATICS Mv = yc Fy ∼ 0.21[N × m] 9 The total moment is the combination of the two and it is Mtotal = 23191.31759[N × m] The center pressure for x area is Ixx b ¡ (r cos θ0 ) r cosθ0 Ixx 5 r cos θ0 36 = + xp = xc + = r cosθ0 xc A 2 9 b ¡ (r cos θ0 ) 2 xc 3 The moment due to hydrostatic pressure is Mh = xp Fx = 5 r cosθ0 Fx ∼ 15399.348 × 22375. Moment on arc element around Point “O.92 The moment is CHAPTER 4.” or M= 0 ρ g sin θ b r 2 r sin θ 2 cos θ 2 dθ The solution of the last equation is M= g r ρ (2 θ0 − sin (2 θ0 )) 4 θ0 The vertical force can be obtained by Fv = 0 P dAv .2 ∼ 7792.27.5[N × m] For direct integration of the moment is done as following θ0 O dF = P dA = 0 ρ g sin θ b r dθ θ θ/2 θ/2 π−θ 2 ℓ = 2 r sin and element moment is θ 2 θ 2 θ 2 π 2 dM = dF × = dF 2 r sin and the total moment is θ0 cos dF θ/2 M= 0 dM θ0 Fig. -4.

Solution o y= b n i=1 ai x i dA dy y x dx Fig. The calculations are done per unit depth (into the page) and do not require the actual depth of the dam. it can be shown that dx2 + dy 2 = dx 1+ dy dx 2 .12: For the liquid shown in Figure 4. The function of the dam shape is n y = i=1 ai xi and it is a monotonous function (this restriction can be relaxed somewhat). The reason that a polynomial function was chosen is that almost all the continuous functions can be represented by a Taylor series. -4. It is much simpler now to use the second method. The element force (see Figure 4. this example provides for practical purposes of the general solution for curved surfaces. consider a more general case of a polynomial function.calculate the moment around point “O” and the force created by the liquid per unit depth.4. there are many programs or hand held devices that can carry numerical integration by inserting the function and the boundaries.28) in this case is P h dA dF = (b − y) g ρ dx2 + dy 2 The size of the diﬀerential area is the square root of the dx2 and dy 2 (see Figure 4. Also calculate the horizontal and vertical forces. and the direct approach was presented second. the traditional approach was presented ﬁrst.28).28. and thus. FLUID FORCES ON SURFACES or θ0 P dAv 93 Fv = 0 ρ g r sin θ r dθ cos θ g r2 ρ 2 1 − cos (θ0 ) 2 Here. Fv = End Solution To demonstrate this point further. It can be noticed that the diﬀerential area that is used here should be multiplied by the depth.28 . Example 4. Polynomial shape dam description for the moment around point “O” and force calculations. From mathematics.5. In fact.

consider the speciﬁc case of y = 2 x6 . In this case.29). -4. only one term is provided and xb can be calculated as following xb = Notice that 6 6 b 2 b 2 is measured in meters. in this case describing the dam function is 1+ dy dx 2 n 2 dy dx ℓ dF b θ y = 1+ i=1 i a (i) x (i) i−1 x x The value of xb is where y = b and can be obtained by ﬁnding the ﬁrst and positive root of the equation of n 0= i=1 ai x − b i Fig. The distance is = (b − 2 2 x6 ) + 6 b −x 2 2 . The derivative at x is and the derivative is dimensionless (a dimensionless number). The diﬀerence between the slop and the direction angle.94 CHAPTER 4.29. expression of the distance and angle to point “O” are needed (see Figure 4. For example. The distance between the point on the dam at x to the point “O” is (x) = (b − y)2 + (xb − x)2 The angle between the force and the distance to point “O” is θ(x) = tan−1 dy dx − tan−1 b−y xb − x The element moment in this case is dF 2 dM = (x) (b − y) g ρ 1+ dy dx cos θ(x) dx To make this example less abstract. FLUIDS STATICS O y The right side can be evaluated for any given function. The number “2” is a dimensional number with dy = 12 x5 dx units of [1/m5 ]. To evaluate the moment.

This omission saves considerable time.4. for a given value b this integral can be evaluate.30. While the question of the stability was not scientiﬁcally examined in the past.6.Fig. -4. In fact. However. While Archimedes did not know much about integrals. ture the essence. End Solution 4. Archimedes princib ple is related to question of density and volume. ship builders who know how to calculate GM but weren’t aware of scientiﬁc principles behind it. To understand this issue. If the reader wonders why such a class is taught in a high school.6 Buoyancy and Stability h r One of the oldest known scientiﬁc rea search on ﬂuid mechanics relates to buoyancy due to question of money was carried by Archimedes. more advance mathematics will be used. The total forces the liquid exacts on a body are considered as a buoyancy issue. This author ﬁnd this method to be simpler for complicated geometries while the indirect method has advantage for very simple geometries. he was able to cap. trying to ﬁnd the center of the area will double the work. Here. Schematic of Immersed Cylinder. . the ﬂoating vessels structure (more than 150 years ago) show some understanding11 . 150 years ago and more. It was taught by people like these. The horizontal force is Fh = b ρ g b ρ g b2 = 2 2 The vertical force per unit depth is the volume above the dam as √ 6 b Fv = 0 b − 2 x6 ρ g dx = ρ g 5 b6 7 7 In going over these calculations. perhaps the name can explain it: Sea Oﬃcers High School. the calculations of the center of the area were not carried out. because this material is presented in a diﬀerent era. consider a cubical and a cylindrical body that is immersed 0 0 11 This topic was the author’s class name in high school. BUOYANCY AND STABILITY The angle can be expressed as θ = tan−1 12 x5 − tan The total moment is √ 6 b 95 b − 2 x6 −1 6 b 2 −x M= 0 (x) cos θ(x) b − 2 x6 g ρ 1 + 12 x5 dx This integral doesn’t have analytical solution.

130) V The force on the immersed body is equal to the weight of the displaced liquid. This analysis can be generalized by noticing two things. the total force is made of the sum of all the small rectangles which is the weight of the sum of all volume. In illustration of this concept. For any two rectangle bodies. Another way to look at this point is by approximation. The solution of equation (4. The force on every rectangular shape is made of its weight of the volume. The force due to the liquid pressure per unit depth (into the page) is F = ρg ((h0 − a/2) − (h0 + a/2)) b = −ρ g a b = −ρgV In this case the be (4. any shape is made of many small rectangles. Rearranging equation (4.134) The negative sign indicate that the force acting upwards.129) represents a depth (into the page). On the lower surface the pressure is ρg(h0 + a/2). the imaginary pressure make it so that they cancel each other. All the horizontal forces are canceled.30. While the horizontal force is Fv = 0 (h0 − r sin θ) cos θ dθ = 0 (4. The force per area (see Figure 4. Any body that has a projected area that has two sides. The forces on square geometry body are made only of vertical forces because the two sides cancel each other. However.132) Rearranging equation (4.131) transforms it to 2π F = rgρ 0 (h0 − r sin θ) sin θ dθ (4. on the vertical direction. Thus even these bodies are in contact with each other. On the other hand.133) Fig. h0 as shown in Figure 4. the pressure on the two surfaces are diﬀerent. On the upper surface the pressure is ρg(h0 − a/2). consider the cylindrical shape in Figure 4.131) r θ The total force will be the integral of the equation (4.129) to F = ρg (4.135) .31) is P dAvertical h0 dF = ρ g (h0 − r sin θ) sin θ r dθ 2π (4. the horizontal forces are canceling each other.133) is F = −π r2 ρ g 2π The ﬂoating forces on Immersed Cylinder.30.31.131) F = 0 ρ g (h0 − r sin θ) r dθ sin θ (4. FLUIDS STATICS in liquid and center in a depth of. (4. The force to hold the cylinder at the place must be made of integration of the pressure around the surface of the square and cylinder bodies. -4. Thus. those will cancel each other.96 CHAPTER 4.

ρl and the surroundings air temperature. calculate the pressure inside the container.32.13: A cylindrical body.32 . T1 for the body to ﬂoat. function of the density of the wall. Express the maximum wall thickness. ρs liquid density. The container diameter is w. Assume that the wall thickness is small compared with the other dimensions (t << w and t << h). the height. ρ π w2 h ρl g hin + Patmos = Pin = Patmos R T1 R T1 hin π w2 . Note that for the maximum thickness. it can be inserted into the above equation. as a ﬂoating body.is ﬂoating in liquid with density. the pressure at the interface from the air point of view (ideal gas model) should be mair R T1 Pin = hin π w2 V Since the air mass didn’t change and it is known.4. Since there are no better examples. In the case where thickness is half the maximum. these examples are a must. The second example of the speed of the ﬂoating bodies. shown in Figure 4. BUOYANCY AND STABILITY Typical examples to explain the buoyancy are of the vessel with thin walls put upside down into liquid. Schematic of a thin wall in it. h1 has to be zero. Thus. The body was inserted into liquid in a such a way that the air had remained Fig. -4. the pressure at the interface can be written as Pin = ρl g hin On the other hand. Solution The air mass in the container is V ρair mair = π w2 h The mass of the container is Patmos RT A mcontainer = π w2 + 2 π w h t ρs The liquid amount enters into the cavity is such that the air pressure in the cavity equals to the pressure at the interface (in the cavity). ρl . t.6. h1 t 97 w hin h Example 4.

FLUIDS STATICS h Patmos hin 4 g h Patmos ρl + Patmos 2 2 g ρl 4 g h Patmos ρl + Patmos 2 − Patmos 2 g ρl The solution must be positive. When the ﬂoating is under vacuum condition.14: Calculate the minimum density an inﬁnitely long equilateral triangle (three equal sides) has to be so that the sharp end is in the water. End advance material Archimedes theorem states that the force balance is at displaced weight liquid (of the same volume) should be the same as the container. Example 4. Thus. Advance material can be skipped Extreme Cases The solution demonstrates that when h −→ 0 then hin −→ 0. so that the last solution is the only physical solution. the air.98 The last equation can be simpliﬁed into ρl g hin + Patmos = And the solution for hin is hin = − and hin = Patmos + CHAPTER 4. net displayed water π w2 (h − hin ) g container = π w 2 + 2 π w h t ρs g + π w 2 h air Patmos R T1 g . the following height can be expanded into hin = h Patmos Patmos + + ··· g ρl 2 g ρl which shows that the large quantity of liquid enters into the container as it is expected. When the gravity approaches zero (macro gravity) then hin = h 2 ρl g 2 h 3 ρl 2 g 2 Patmos 5 h 4 ρl 3 g 3 +h− + − + ··· ρl g Patmos Patmos 2 Patmos 3 This “strange” result shows that bodies don’t ﬂoat in the normal sense.

BUOYANCY AND STABILITY If air mass is neglected the maximum thickness is tmax = 2 g h w ρl + Patmos w − w 4 gh Patmos ρl + Patmos 2 (2 g w + 4 g h) ρl ρs 99 The condition to have physical value for the maximum thickness is 2 g h ρl + Patmos ≥ The full solution is tmax = − “ ” √ w R 4 gh Patmos ρl +Patmos 2 −2 g h w R ρl −Patmos w R T1 +2 g h Patmos w ρl (2 g w+4 g h) R ρl ρs T1 4 gh Patmos ρl + Patmos 2 In this analysis the air temperature in the container immediately after insertion in the liquid has diﬀerent value from the ﬁnal temperature.6. The pressure at the interface (after long time) is ρl g hin + Patmos = which can be simpliﬁed to h Patmos hin + h1 The second equation is Archimedes’ equation. The equation of state is Pi = mair R Ti Vi The new unknown must provide additional equation which is Vi = π w2 hi Thickness Below The Maximum For the half thickness t = tmax the general solution for any given thickness below 2 maximum is presented.4. but the liquid displacement is still unknown. It is reasonable as the ﬁrst approximation to assume that the process is adiabatic and isentropic. The thickness is known. which is ρl g hin + Patmos = π w2 (h − hin − h1 ) = π w2 + 2 π w h) t ρs g + π w2 h End Solution atmos π w2 h PR T1 R T1 (hin + h1 ) π w2 Patmos R T1 g . Thus. the temperature in the cavity immediately after the insertion is Ti = Tf Pi Pf The ﬁnal temperature and pressure were calculated previously.

it can be observed that only the density ratio is important. where α is ratio between the body density to the liquid density and (0 < α < 1).100 CHAPTER 4.16: In some situations. Calculate acceleration and time for a body to reach the surface. End Solution Example 4. FLUIDS STATICS Example 4. Find equivalent cylinder that has the same diameter that Solution The force act on the half sphere can be found by integrating the forces around the . the acceleration is a=g 1−α α If the object is left at rest (no movement) thus time will be (h = 1/2 a t2 ) t= If the object is very light (α −→ 0) then tmin = 2hα + g √ 2 g h α2 3 + 2g 3 2 hα g(1 − α) √ 2 g h α2 5 + 8g 5 √ 2 g h α2 + ··· 16 g 7 From the above equation. it is desired to ﬁnd equivalent of force of a certain shape to be replaced by another force of a “standard” shape. This idea can lead to experiment in “large gravity” because the acceleration can be magniﬁed and it is much more than the reverse of free falling. Consider the force that acts on a half sphere. The body’s density is α ρl . h0 and left at rest. Is the body volume important? Solution The net force is liquid weight body weight F = V g ρl − V g α ρl = V g ρl (1 − α) But on the other side the internal force is m F = m a = V αρl a Thus.15: A body is pushed into the liquid to a distance.

it was assumed that above liquid is a gas with inconsequential density.17: In the introduction to this section. ρl < ρs < rhoh is ﬂating between the two liquids.6. This force is balance according to above explanation by the two liquid as ρc ¨A = ¨h (α ρl + (1 − α)ρh ) h¨ A¨ .4. BUOYANCY AND STABILITY sphere. In this case there are two diﬀerent liquid densities. Develp the relationship between the densities of liquids and solid and the location of the solid cubical. Where h is the height of the body and A is its cross section. Solution In discussion to this section it was shown that met force is the body volume times the the density of the liquid. The element force is dAx h 2 dA 101 dF = (ρL − ρS ) g r cos φ cos θ cos θ cos φ r dθ dφ The total force is then π π 0 Fx = 0 (ρL − ρS ) g cos2 φ cos2 θ r3 dθ dφ The result of the integration the force on sphere is Fs = The force on equivalent cylinder is Fc = π r2 (ρL − ρS ) h These forces have to be equivalent and thus $ ! π $$− ρS ) r¡ (ρL $$ 3 $ = & $$− ρS ) h π r2 (ρL $$ 4 2 £ 1 π 2 (ρL − ρS ) r3 4 Thus the height is π h = r 4 End Solution Example 4. Body with density between the two liquids. There situations where density is a function of the depth. What will be location of solid body if the liquid density varied parabolically. The net force down is the weight of the body ρc h A. the body can be separated into two which one in ﬁrst liquid and second in the second liquid. Suppose that the above layer is another liquid which has a bit lighter density. In the same vein.

This arrangement has mass centroid close to the middle of the sphere. The density as a function of x coordinate along h starting at point ρh is ρ(x) = ρh − x h 2 (ρh − ρl ) Thus the equilibration will be achieved.33. x1 = h ρl + h ρh − 6 ρc 2 ρl − 2 ρh In many cases in reality the variations occur in small zone compare to the size of the body. -4. x1 . will be at √ 3 3 h2 ρl 2 + (4 ρc − 6 h2 ρh ) ρl + 3 h2 ρh 2 − 12 ρc ρh + 3 h ρl − 3 h ρh X1 = 6 ρh − 2 ρl For linear relationship the the following results can be obtained. Schematic of ﬂoating bodies. Thus. if body is small compare the zone of variation. The buoyant center is below the middle of the balloon. when x1 +h ρc h = x1 ρh − x h 2 (ρh − ρl ) dx After the integration the equation transferred into ρc h = (3 ρl − 3 ρh ) x12 + (3 h ρl − 3 h ρh ) x1 + h2 ρl + 2 h2 ρh 3h And the location where the lower point of the body (the physical). they have to accounted for.102 CHAPTER 4. the balloon will be on the top and sphere on the bottom.6. . End Solution 4. If this arrangement is inserted into liquid and it will be ﬂoating. After rearrangement it became α= ρc − ρh ρl − ρh the second part deals with the case where the density varied parabolically. However.33 shows a body made of hollow balloon and a heavy sphere connected by a thin and light rod. the calculations can be carried out under assumption sharp change. FLUIDS STATICS Where α is the fraction that is in low liquid. Tilting the body with a small angle from gravity center Full a b c Fig. A is canceled on both sides.1 Stability Empty buoyancy center Figure 4.

BUOYANCY AND STABILITY 103 its resting position creates a shift in the forces direction (examine Figure 4. The cubic mass (gravity) centroid is in the middle of the cubic. Schematic of ﬂoating cubic. categories. However. shown in Figure 4. The cubic is stable in six positions (every cubic has six faces). This analysis doesn’t violate the second law of thermodynamics. If the new immerse volume creates a new center in such way that the couple forces (gravity and buoyancy) try to return the body. the cubic results in returning to the original position. So. HowG ever the buoyant center is the middle of the volume under the water (see Figure 4. That is. When tilting a larger amount than π/4 . should to be calculated. -4.35 is displaced by the same . any experiment of this cubic wood shows that it is stable locally. when given a tilted position.6. This B situation is similar to Figure 4. the body is in situation like in 4. These points are raised from the buoyant G dA force analysis. the body stability must be based on the diﬀerence between the body’s local positions rather than the “absolute” stability. B’. the original state is referred as the stable body and vice versa. Moving bodies from an unstable position is in essence like a potential.34). it results in a ﬂipping into the next stable position. The center of the mass (gravity) is still in the old location since the body did not change. Stability analysis of ﬂoating body. If one draws the stability (later about this criterion) as a function of the rotation angle will show a sinusoidal function with four picks in a whole rotation.33b). in any of these six positions. When the body is tilted at δF ∆F GM a small angle.35. These forces create a moment which wants to return the body to the resting (original) position. The stability of the body is divided into three Fig. In fact. A wooden cubic (made of pine. for example) is inserted into water. The body.35. When the body is at the position shown in Figure 4.33c. The right green area (volume) in Figure 4. B’ as shown in Figure 4.the body is unstable and any tilt from the original position creates moment that will further continue to move the body from its original position. the immersed part of the B B’ body center changes to a new location. it is referred to as the neutral stable. -4.33c. This analysis is based on the diﬀerence of the displaced liquid.4.33c . Small amount of tilting of Fig. the body is “stable” in some β M δF ∆F points more than others in their vicinity. The reason for this local stability of the cubic is that other positions are less stable.35. β. Part of the block ﬂoats above water line. The third state is when the couple forces do have zero moment.34. move to a new buoyant center. B. This deviation of the buoyant center from the old buoyant center location.

The body weight creates opposite moment to balance the moment of the displaced liquid volume. . BB W = M (4. BB .137) M= A g ρl x β dA x = g ρl β dV A x2 dA (4.104 CHAPTER 4. FLUIDS STATICS area (really the volume) on left since the weight of the body didn’t change12 so the total immersed section is constant.136) Where M is the moment created by the displaced areas (volumes). The displacement of the buoyant center can be calculated by examining the moment these forces are creating.138) is referred to as the area moment of inertia and was discussed in Chapter 3.138) as BB = g ρl Ixx ρs Vbody (4. BB is the distance between points B and point B’. For small angle.141) lim (4. However.140) yields BM = For small angle (β ∼ 0) β→0 g ρl βIxx ρl Ixx ¡ = g ρs sin β Vbody ρs Vbody ¡ sin β ∼1 β (4. W referred to the weight of the body. should be BB = The moment M can be calculated as δF M W (4.139) The point where the gravity force direction is intersecting with the center line of the cross section is referred as metacentric point. The distance. and.138) The integral in the right side of equation (4.140) And combining equations (4. The location of the metacentric point can be obtained from the geometry as BM = BB sin β (4. the moment is calculated as the integration of the small force shown in the Figure 4. when the body is not symmetrical. M. So the perpendicular distance.). It can be noticed that the distance BB is an approximation for small angles (neglecting the vertical component. BB can be written from equation (4.35 as ∆F . the analysis is still correct because the volume and not the area is used.142) 12 It is correct to state: area only when the body is symmetrical.139) with (4. β.

Stability of cubic body inﬁnity long.6 1.1 and is Ixx = La .5 -0.4 1.0 ρs 1− ρl -0.2 0.2 α = 0.18: A solid block of wood of uniform density.8 1.0 0.2 BG = h ρs 1 h − h = 2 ρl 2 2 1− GM h ρs ρl (4. the distance BG is (see Figure 4. Fig. Where L is the length into 12 the page.2 1. is insigniﬁcant for this analysis.3 1. Show that the block’s length.5 L a g ρl ¡ 12 − h GM = g ρs a h 2 L ¡ V 3 α = 0. Looking at Figure 4. 2008 Simplifying the above equation provides Fig.4 α = 0.143) a To understand these principles consider the following examples.4.9 0. Construct a graph that shows the relationship of the GM as a function of ratio height to width.145) 2 . The distance BG obtained from Archimedes’ theorem and can be expressed as immersed V volume ρs W = ρs a h L = ρl a h1 L =⇒ h1 = h ρl Thus.143) requires that several quantities should be expressed.8 2.4 Ixx α = 0.1 1. 1 GM = h 12 α a h 2 1 − (1 − α) (4.36.2 -0.5 0.2 1. the geometrical quantities can be related as BM 105 h h1 L GM = ρl Ixx −BG ρs Vbody (4.37.0 1. BUOYANCY AND STABILITY It is remarkable that the results is independent of the angle.6 0.0 0.1 1.8 0. Solution Equation (4.144) 2.9 1.4 0.34) 3.0 α = 0.0 a h April 16.35.5 0. The moment of 3 inertia for a block is given in Table 3. Cubic body dimensions for stability analysis.7 0.6.0 -0.0 0.3 0.3 1.5 α = 0.7 α = 0. ρs = α ρl where ( 0 ≤ α ≤ 1 ) is ﬂoating in a liquid. L.6 0.1 2.0 1. -4. -4. Example 4.8 0.1 Stability of Square Block h1 0.

4 April 16. 2008 α 0. the equation is Stability of Solid Blocks GM g = h 64 α b h 2 − 1 (1 − α) 2 a h 3. Notice that GM /h isn’t a function of the depth.0 2.0 square circle And the condition for maximum height for stability is b ≥ h 32 (1 − α) α 1.0 0. For each of these axises there is a diﬀerent moment of inertia.5 0. In cylindrical shape equivalent equation to equation (4. -4. The diﬀerence between the previous calculation and the moment of inertia around the diagonal is I diagonal axis √ ∆Ixx = 2a √ 3a 2 3 “normal axis 6 − a4 12 ∼ 0. Thus.5 0. Principle Main Axises Any body has inﬁnite number of diﬀerent axises around which moment of inertia can be calculated. it is enough to ﬁnd if the body is stable around the smallest moment of inertia.5 1.7 0.2 0. this problem is reduced to ﬁnd the stability for principle axis.6 0. For cylinder (circle) the moment of inertia is Ixx = π b4 /64.8 0. For example.146) End Solution One of the interesting point that the above analysis was that there is a point above which the ratio of the height to the body width is not stable any more. a square shape body has larger moment of inertia around diagonal. The distance BG is the same as for the square shape (cubic) (see above (4.3 0.0 0.146) can be expressed.5 2.38.144)).07 a4 Which show that if the body is stable at main axises.0 0. these two shapes in Figure 4. .0 This kind of analysis can be carried for difFig.106 CHAPTER 4. Some analysis of ﬂoating bodies are done by breaking the rotation of arbitrary axis to rotate around the two main axises. it must be stable at the “diagonal” axis. With the exception of the circular shape. The maximum height reverse as a ferent shapes and the results are shown for function of density ratio.38. This equation leads to the condition where the maximum height above which the body is not stable any more as a ≥ h 6 (1 − α)α (4. every geometrical shape has an axis in which the moment of inertia is without the product of inertia. It can be noticed that the square body is more stable than the circular shape body. For stability analysis. Thus. L. This axis is where the main rotation of the body will occur.1 0.9 1. FLUIDS STATICS where α is the density ratio.

Assume that density ratio is ρl /ρs = ρ. Denote the liquid density as ρl and solid density as ρs . the fuel is stored at the bottom of the ship and thus the mass center (point G) is changing during the voyage.4. The point B can be expressed as B= And thus the distance BG is BG = a 2 1− ρs ρl a ρs 2 ρl . explanation in example (3. -4.7) is a h3 Ixx = 2 And the volume is Vbody = a2 h2 − a2 = a2 h 4 1− 1 a2 4 h2 The point B is a function of the density ratio of the solid and liquid. ¯ Solution h The answer to the question is that the limiting case where GM = 0. but rather practical. Suppose that two opposite sides triangle (prism) attached to each other to create a long “ship” see Figure 4. This happens when a ship is overloaded with containers above the maximum height. On the other side if the a/h −→ ∞ the body is very stable. the ship that was stable (positive GM ) leaving the initial port might became unstable (negative GM ) before reaching the destination port. The Volume of the body is V =2 a2 h 2 = a2 h Fig. In commercial ships. What is the ˜ minimum ratio of a/h that keep the body stable at half of the volume in liquid (water).39. a a Stability of two triangles put The moment of inertia is triage (see for tougher.6. Example 4. Supposed that a/h −→ ˜ the body will be 0 unstable. BUOYANCY AND STABILITY Unstable Bodies 107 What happen when one increases the height ratio above the maximum height ratio? The body will ﬂip into the side and turn to the next stable point (angle). So.143) have to found. To ﬁnd this ratio equation terms in (4.19: One way to make a ship to be a hydrodynamic is by making the body as narrow as possible.39. This is not a hypothetical question.

a ship that carries Gc wheat grains where the cargo is not propG G′ erly secured to the ship. After the about manipulation and selecting the positive value and to keep stability as √ x< 64 ρ4 −64 ρ3 +ρ2 −2 ρ+1 ¯ ¯ ¯ ¯ ρ ¯ + 2 √ 1 ρ ¯ −1 2ρ ¯ End Solution 4. Exact analysis requires taking into the GM . here. The movement of the load (grains. the slow reaction of the load.1. the extreme case where the load reacts in the same speed as the tilting of the ship/ﬂoating body is examined. account these shifting mass speeds. FLUIDS STATICS The limiting condition requires that GM = 0 so that ρl Ixx = BG ρs Vbody Or explicitly ρl ρs a2 h a h3 2 1− 1 a 4 h2 2 = a 2 1− ρs ρl After rearrangement and using the deﬁnitions of ξ = h/a ρρl /ρs results in ¯ ρ ξ2 ¯ 1− ξ 4 2 = 1− 1 ρ ¯ The solution of the above solution is obtained by squaring both sides and deﬁning a new variable such as x = ξ 2 .Fig.1 Stability of Body with Shifting Mass Centroid Ships and other ﬂoating bodies carry liquid or have a load which changes the M mass location during tilting of the ﬂoating body. For example. and/or liquid) B does not occur in the same speed as the B′ body itself or the displaced outside liquid. There are situations where the real case approaches to this extreme. Sometimes.108 CHAPTER 4. The eﬀects of liquid movement on nored. furniture. These situations involve liquid with a low viscosity (like water.6. it is used as a limit for the stability analysis. for stability analysis. -4.40. However. For practical purposes. alcohol) and ship with . is enough to be ig.

151) The distance that was used before GM is replaced by the criterion for stability by Gc M and is expressed as Gc M = g ρA Ixx A 1 Ixx b − BG − ρs Vbody mtotal Vb (4. When the body is given a tilting position the body displaces the liquid on the outside. G Gc = GG sin β (4. This point is the intersection of the center line with the vertical line form G . Moreover. The total change of the vessel is then calculated similarly to center area calculations.150) A new point can be deﬁned as Gc . This quantity.148) Equation (4. the liquid inside is changing its mass centroid. A body is loaded with liquid “B” and is ﬂoating in a liquid “A” as shown in Figure 4.4.153) . The change in the mass centroid of the liquid “A” then is Ixx B g ρl¨βIxx B G1 G1 = ¡ ¨B = g VB ¨B ρl¨ VB ¡ Inside liquid weight (4. in this analysis. G1 G1 .6. X0 g mtotal GG = $mbody + g mf G1 G1 g $$$ ¡ ¡ For more than one tank.148) shows that GG is only a function of the geometry. The moment created by the inside displaced liquid is Min = g ρl B βIxx B (4. it can be written as GG = g Wtotal n (4. the general formula is g ρA Ixx A 1 Gc M = − BG − ρs Vbody mtotal n i=1 Ixx bi Vb i (4.147) Note that Ixx B isn’t the same as the moment of inertia of the outside liquid interface.152) If there are more than one tank partially ﬁlled with liquid. At the same time. BUOYANCY AND STABILITY 109 low natural frequency (later on the frequency of the ships). is similar for all liquid tanks on the ﬂoating body.149) Gi Gi ρl i Vi = i=1 g Wtotal n i=1 Ixxb i Vbi (4. the dynamics are ignored and only the statics is examined (see Figure 4.40).40.

The angle. The change in height of G is g mtotal Gnew = g mship Gactual + g T h ¡ ¡ ¡ Combining equation (4.156) results in GM a ctual = GM new T mtotal −h mship mship (4.2 Metacentric Height.41. The metacentric height is GM new = Td Wtotal θ (4. Movement of the liquid (mostly the fuel and water) provides way to control the stability. of the ship. equation (4. is measured as the diﬀerence in the orientation of the ﬂoating body. can be achieved by having some tanks spanning across the entire body with tanks spanning on parts of the body.158) (4. Increasing the number of tanks reduces the moment of inertia. GM . The moment of inertial of the combine two tanks is smaller than the moment of inertial of a single tank. Mrighting = Wtotal GM new θ (4. .6. This control of the stability. is the total weight of the ﬂoating body including measuring weight. Measurement The metacentric height can be measured by ﬁnding the change in the angle when a weight is moved on the ﬂoating body.154) T d h G Fig. d then the moment created is Mweight = T d This moment is balanced by (4. Measurement of GM of ﬂoating body.156) provides the solution.155) Where.110 CHAPTER 4.157) The weight of the ship is obtained from looking at the ship depth. GM . FLUIDS STATICS One way to reduce the eﬀect of the moving mass center due to liquid is done by substituting a single tank with several tanks. GM . 4. Wtotal . θ. The engineer could design the tanks in such a way that the moment of inertia is operationally changed.1. The calculation of GM can be improved by taking into account the eﬀect of the measuring weight. Moving the weight. -4.157) with equation (4.156) If the change in the GM can be neglected. T a distance.

.6.4. When the body is submerged in a single ﬂuid layer.6. 4. BUOYANCY AND STABILITY 4. then none of the changes of buoyant centroid occurs. The moment created by change in the displaced liquid (area) act in the same fashion as the before. the mass centroid must be below the buoyant centroid in order to have stable condition. the stability analysis must take into account the changes of the displaced liquids of the two liquid layers. Fig. is when b = 3 a. In cases where the density changes signiﬁcantly. However. there are situations where the body has a “strange” ∆F M δβ and/or un-symmetrical body. This point is intersection point liquid with lower body and it is needed to be calculated. For the case where b < 3 a x some part is under the liquid. the upper part of the body is above the liquid or part of the body is submerged under the water.1. The calculations for such cases are a bit more complicated but based on the similar principles. -4. For an example of such a case is an object ﬂoating in a solar pond where the upper layer is made of water with lower salinity than the bottom layer(change up to 20% of the density). this density change helps to increase the stability of the ﬂoating bodies. Calculations of GM for abrupt The mathematical condition for the border shape body.6. There are two situations that can occur.4 Stability of None Systematical or “Strange” Bodies While most ﬂoating bodies are symmeta rical or semi–symmetrical. Thus.42).42.42. The amount of area under the liquid section depends on the tilting angle. the moment to return the body is larger than actually was calculated and the bodies tend to be more stable (also for other reasons). The moment of inertia is calculated around this point (note the body is “ended” at end of the upper body).1. However. For the case of b < 3 a the calculation of moment of inertia are similar to the previous case. The body ∆F GM weight doesn’t change during the rotation that the green area on the left and the B’ B green area on right are the same (see Figb ure 4. Generally. When the ﬂoating object is immersed in the two layers. This point is the intersection of the liquid line with the brown middle line. The moment of inertia should be calculated around this axis. it must be taken into account. all ﬂuids have density varied in some degree. The center of the moment is needed be found. Consider the ﬁrst strange body that has an abrupt step G change as shown in Figure 4. After the tilting.3 Stability of Submerged Bodies 111 The analysis of submerged bodied is diﬀerent from the stability when the body lay between two ﬂuid layers with diﬀerent density. These calculations are done as if none of the body under the liquid. This analysis is out of the scape of this book (for now).

159) Where here is length of the rode (or the line/wire) connecting the mass with the rotation point. In this section. the frequency of pendulum is 21π g which measured in Hz. If the ﬂoating body is used for transport humans and/or other creatures or sensitive cargo it requires to reduce the GM so that the traveling will be smother. Thus. in this case.6.6. However. End Solution . coating. etc. only simpliﬁed topics like constant value will be discussed.5 Neutral frequency of Floating Bodies CHAPTER 4. Increase in GM increases the frequency of the ﬂoating body. Newton’s law states when there is unbalanced force. the larger GM the more stable the ﬂoating body is. the liquid is not in motion. Example 4. the common explanation is wrong.161) In general. 4.1. FLUIDS STATICS This case is similar to pendulum (or mass attached to spring).2 Surface Tension The surface tension is one of the mathematically complex topic and related to many phenomena like boiling. the body should be accelerate. Similar situation exists in the case of ﬂoating bodies. Thus.112 4.162) V ρs GM Ibody (4.20: In interaction of the molecules shown in Figure ? describe the existence of surface tension. The basic diﬀerential equation is used to balance and is rotation rotating moment ¨ Iβ − V ρs GM β =0 (4. The period of the cycle is 2 π /g. The governing equation for the pendulum is ¨ β −gβ = 0 (4. Explain why this description is erroneous? Solution The upper layer of the molecules have unbalanced force towards the liquid phase.160) In the same fashion the frequency of the ﬂoating body is 1 2π and the period time is 2π Ibody V ρs GM (4.

4. above lower ﬂuid with lower density. ρL . the surface tension between the needle and the liquid hold the needle above the liquid. The original Rayleigh’s paper deals with the dynamics and density variations. Solution Under Construction End Solution 4.21: Needle is made of steel and is heavier than water and many other liquids. Taylor. Calculate the maximum diameter needle that can be inserted into liquid without drowning. For example. This instability deals with a dense. some of heavy liquid moves down. liquid metal is injected in a cavity ﬁlled with air. heavy ﬂuid that is being placed above a lighter ﬂuid in a gravity ﬁeld perpendicular to interface. After certain diameter. Supposed that a liquid density is arbitrary function of the height. This disturbance can grow or returned to its original situation.3. If the surface will disturbed.7. the needle cannot be held by the liquid. This situation has engineering implications in several industries. This analysis is referred to the case of inﬁnite or very large surface.3. some air is not evacuated and stay in small cavity on the edges of the shape to be casted. This distortion can be as a result of heavy ﬂuid above the lighter liquid. ρG . For example a heavy ﬂuid density. The simpliﬁed case is the two diﬀerent uniform densities. RAYLEIGH–TAYLOR INSTABILITY 113 Fig. or water over air(gas–liquid). However. This condition is determined by competing . A heavy needle is ﬂoating on a liquid. In poor designs or other situations. -4. Thus. Example 4. the heavy ﬂuid will stay above the lighter ﬂuid. I.43. density variations according to the bulk modulus (see section 4. There are situations where a heavy liquid layer is placed over a lighter ﬂuid layer. For perfectly straight interface.2) are always stable but unstable of the density is in the reversed order. This analysis asks the question what happen when a small amount of liquid from above layer enter into the lower layer? Whether this liquid continue and will grow or will it return to its original conditions? The surface tension is the opposite mechanism that will returns the liquid to its original place. it can create a situation where the liquid metal is above the air but cannot penetrate into the cavity because of instability. For example in die casting. Example for such systems are dense water over oil (liquid–liquid).7 Rayleigh–Taylor Instability RayleighTaylor instability (or RT instability) is named after Lord Rayleigh and G.

167) . -4. The radius of any equation is expressed by equation (1. example of a cosine function will be examined. and the buoyancy forces.44). The depression is returned to its h original position if the surface forces are L large enough. As usual there is the neutral stable when the forces are equal.47) can be approximated as 1 d2 h = 2 R dx For equation (4. Thus.38) the pressure diﬀerence or the pressure jump is due to the surface tension at this point must be PH − PL = 4 hmax σ π 2 L2 (4. The ﬂuid above the depression is in equilibrium with the sounding pressure since the material is extending to inﬁnity.163) where hmax is the maximum depression and L is the characteristic length of the depression. if the surface forces (surface tension) are not suﬃcient. the situation is Fig. Description of depression to explain unstable and the heavy liquid enters into the Rayleigh–Taylor instability. the liquid ﬂuid zone and vice versa.166) The pressure diﬀerence due to the gravity at the edge of the disturbance is then PH − PL = g (ρH − ρL ) hmax (4. Any continues function can be expanded in serious of cosines.114 CHAPTER 4. the force that acting to get the above ﬂuid down is the buoyancy force of the ﬂuid in the depression. equation (1. Thus. The weakest point is at x = 0 because symmetrical reasons the surface tension does not act against the gravity as shown in Figure (4. the surface density.44. The conditions that required from this function will be required from all the other functions. The depression has diﬀerent radius as a function of distance from the center of the depression. this situation x σ σ is considered to be stable. if the center point of the depression can “hold” the intrusive ﬂuid then the whole system is stable. The disturbance is of the following h = −hmax cos 2πx L (4.165) (4. On the other hand. Thus. x. FLUIDS STATICS forces. In that case.163) the radius is 1 4 π 2 hmax =− R L2 (4.47). Thus.164) According to equation (1. The ﬁrst derivative of cos around zero is sin which is approaching zero or equal to zero.

the error is not signiﬁcant. -4. Additionally when the depression occurs. The θ “extra” lines of the depression should be ignored. To analyze it. This acts Fig. The heavier liquid needs to move in one side and the lighter liquid in another location. the force is ρ g h × A. The horizontal forces around the control volume are canceling each other. At the bottom. In reality. At the top. If all the heavy liquid “attempts” to move straight down.166) and (4.7. In this process the heavier liquid “enter” the lighter liquid in one point and creates a depression as shown in Figure 4.45. Thus. the force at the top is the same force at the bottom of the cylinder.16) This value is exact if the shape is a perfect half sphere. cylinder to be in equilibrium with its surroundings if the pressure at bottom is indeed ρ g h. the liquid level is reduced a bit and the lighter liquid is ﬁlling the missing portion.167) show that if the relationship is 4 σ π2 > g (ρH − ρL ) L2 115 (4. For the depression. The ﬁrst control volume is made of a cylinder with a radius r and the second is the depression below it. The lighter liquid needs to move up at the same time but in a diﬀerent place. RAYLEIGH–TAYLOR INSTABILITY Comparing equations (4.171) .170) Fbottom ∼ π r2 (ρL − ρG ) g (4. the lighter liquid will “prevent” it.45. the force at the bottom is σ σ Fbottom ∼ π r2 The net force is then πr + h (ρL − ρG ) g + Patmos 4 πr 4 (4.4. At the cylinder bottom. they are not part of the control volume. Consider the situation described in Figure 4. the force is atmospheric pressure times the area. the force is the integral around the depression. It can be approximated as a ﬂat cylinder that has depth of r π/4 (read the explanation in the example 4.169) An alternative approach to analyze this instability is suggested here.168) It should be noted that hmax is irrelevant for this analysis as it is canceled. The point where the situation is neutral stable Lc = 4 π2 σ g (ρH − ρL ) (4. considered two control volumes bounded by the blue lines in 2r Figure 4.45. Description of depression to explain against the gravity force which make the the instability.45.

173) (4. FLUIDS STATICS The force that hold this column is the surface tension.46).45.” The maximum possible radius of the depression depends on the geometry of the container. For the cylindrical geometry. As shown in Figure 4. θ = π/2. The cross section of the interface. the radius is r∼ 2πσ (ρL − ρG ) g (4.176) . the total force is then Fσ = 2 π r σ cos θ The forces balance on the depression is then 2 π r σ cos θ ∼ π r2 The radius is obtained by r∼ 2 π σ cos θ (ρL − ρG ) g (4. -4.175) Fig. it can be written that the minimum radius is rmin tube = 2 2πσ g (ρL − ρG ) (4.172) The maximum surface tension is when the angle. At that case. This radius is limited because the lighter liquid has to enter at the same time into the heavier liquid zone. Thus.46. the maximum depression radius is about half for the container radius (see Figure 4. Since the “exchange” volumes of these two process are the same.116 CHAPTER 4.174) πr 4 (ρL − ρG ) g (4. The purple color represents the maximum heavy liquid raising area. the speciﬁc radius is limited. The yellow color represents the maximum lighter liquid that are “going down.

81 The minimum radius is r ∼ 0. The density of the aluminum is 2400kg/m3 . In Figure 4. End Solution Z L3 L2 L1 Fig.4 2400 × 9. RAYLEIGH–TAYLOR INSTABILITY 117 The actual radius will be much larger.46. These two scenarios should be inserting into equation 4. The actual value of this angle is about π/4 to π/3 and in only extreme cases the angle exceed this value (considering dynamics).4. this angle is never obtained.175) can be used.22: Estimate the minimum radius to insert liquid aluminum into represent tube at temperature of 600[K]. The heavier liquid can stay on top of the lighter liquid without being turned upside down when the radius is smaller than the equation 4. σ r∼ 8 π 0.02[m] which demonstrates the assumption of h >> r was appropriate.168 by introducing experimental coeﬃcient.176. Assume that the surface tension is 400[mN/m]. . The density of air is negligible as can be seen from the temperature compare to the aluminum density. it was shown that the depression and the raised area are the same. However. This analysis introduces a new dimensional number that will be discussed in a greater length in the Dimensionless chapter.176) the angle was assumed to be 90 degrees. Solution The depression radius is assume to be signiﬁcantly smaller and thus equation (4.47.the depression is larger for square area. The actual area of the depression is only a fraction of the interfacial cross section and is a function. -4. In equation (4.7. Example 4. Three liquids layers under rotation with various critical situations. For example.

23: A canister shown in Figure 4.S. Is there a diﬀerence if the ﬂuids are compressible? Where is the maximum pressure points? For the case that the ﬂuids are compressible. You can assume that the process is isothermal. Example 4.118 CHAPTER 4. the canister top center is connected to another tank with equal pressure to the canister before the rotation (the connection point). for known geometries of the ﬂuids. FLUIDS STATICS Open Question by April 15. dollars and your name will be associated with the solution in this book. Is there diﬀerence if the process is isentropic? If so what is the diﬀerence? . What happen after the canister start to be rotated? Calculated the volume that will enter or leave. Describe the interface of the ﬂuids consider all the limiting cases. Assume that the ﬂuids do not mix. You the ideal gas model.47 has three layer of diﬀerent ﬂuids with diﬀerent densities. ω. The canister is rotate with circular velocity. 2010 The best solution of the following question will win 18 U.

Part I Integral Analysis 119 .

.

These differential equations will be used in Fig. the Lagrangian system turned out to be diﬃcult to solve and to analyze.1 Introduction This chapter presents a discussion on the control volume and will be focused on the conservation of the mass. -5. This name is in honored J. This method applied and used in very few cases. Even though this system looks reasonable. Langrange (1736–1813) who formulated the equations of motion for the moving ﬂuid particles. one wants to ﬁnd or predict the velocities in the system. L.motion.1. The main diﬃculty lies in the fact that every particle has to be traced to its original state. The main target of such analysis is to ﬁnd the value of certain variables. This kind of analysis is reasonable and it referred to in the literature as the Lagrangian Analysis. When the ﬂuid system moves or changes. Leonard Euler (1707–1783) suggested an alternative approach. The use of the Eulerian methods leads to a set diﬀerentic control ation equations that is referred to volume as Navier–Stokes equations which are commonly used. Control volume and system before and after the later part of this book. Ad. The Eulerian method focuses on a deﬁned area or locaa system tion to ﬁnd the needed informab tion. In Euler’s approach the focus is on a deﬁned point or a deﬁned volume.CHAPTER 5 The Control Volume and Mass Conservation 5. This methods is referred as Eulerian method. 121 .

the control gains some material which is marked as “c”. entropy etc. The material that remained in the control volume is marked as “b”. This methods has its limitations and for some cases the Lagrangian is preferred (and sometimes the only possibility). energy. The red dotted lines are the control volume. Every control volume is the focus of the certain interest and will be dealt with the basic equations. 5. After a certain time. When a piston pushing gases a good choice of control volume is a deformable control volume that is a head the piston inside the cylinder as shown in Figure 5. The control volume is a deﬁned volume that was discussed earlier.piston with in and out ﬂow. trol volume. Deformable control volume is a volume having part of all of its boundaries in motion during the process at hand. This coordinate system may be in a relative motion to another (almost absolute) coordinate system. The Eulerian method plays well with the physical intuition of most people. non–deformable and deformable. The coordinate system could be ﬁxed to the conduit. MASS CONSERVATION ditionally. The control volume is diﬀerentiated into two categories of control volumes.122 CHAPTER 5. .2. mass.2 Control Volume The Eulerian method requires to deﬁne a control volume (some time more than one). momentum. Two examples of control volume are presented to illustrate diﬀerence between a deformable control volume and non–deformable control volume. The control volume should be chosen so that the analysis should be simple and dealt with as less as possible issues which are not in question. The green lines in Figure ?? represent the system. Control volume of a moving The control volume chosen is non-deformable con. -5. some of the mass in the system exited the control volume which are marked “a” in Figure ??. The diﬀerence between the system and the control volume is shown in Figure ??. At certain time the system and the control volume are identical location. Non–deformable control volume is a control volume which is ﬁxed in space relatively to an one coordinate system. the Eulerian system leads to integral equations which are the focus of this part of the book. Therefore a limited discussion on the Lagrangian system will be presented (later version). Fig.2. Lagrangian equations are associated with the system while the Eulerian equation are associated with the control volume. At the same time. the control volume is a system. Flow in conduits can be analyzed by looking in a control volume between two locations. In the case where no mass crosses the boundaries.

The mass ﬂow out of the control volume is the system mass that is not included in the control volume. The actual velocity of the ﬂuid leaving the control volume is the relative velocity (see Figure 5. is made of msys = mc.3. The system mass change is D msys D = Dt Dt ρdV = 0 Vsys (5. is perpendicular to the surface is − → Urn = −ˆ · Ur = Ur cos θ n (5.5.7) . The change with time (time derivative of equation (5. + ma − mc (5. In this chapter.3).v. The velocity component that velocities at the interface.3.3) is the derivative of the mass in the control volume and at any given time is d mc.v.3 Continuity Equation In this chapter and the next three chapters. -5. CONTINUITY EQUATION 123 5. (5. d ma d mc = + − Dt dt dt dt (5. (t) d = dt dt ρ dV Vc.4) Control Volume and is a function of the time.1) The system mass after some time.3). the conservation equations will be applied to the control volume. The interface of the control volume can move.3) The ﬁrst term in equation (5. The convention of direction ˆ is taken positive if ﬂow out the control volume and negative if the ﬂow is into the control volume. Thus.v.2)) results in 0= D msys d mc.v.2) The change of the system mass is by deﬁnition is zero. the ﬂow out is d ma = dt ρs Urn dA Scv (5.6) Where n is an unit vector perpendicular to the surface.5) Ub n ˆ θ Uf −Ub Uf − Ub Where Uf is the liquid velocity and Ub is the boundary Fig. according Figure ??. The relative velocity is − → − → − → Ur = Uf − Ub (5. Schematics of velocity (see Figure 5. the mass conservation will be discussed.

ρ0 L t0 . can be approximated as x 2 t ρ(x.124 CHAPTER 5. -5. X dx L Fig.7) are similar and can be combined.9) into equation (5. Again notice the negative sign in surface integral.v.1.1: The density changes in a pipe.3) results in d dt ρs dV = − c.8) and (5.10) is essentially accounting of the mass.8) It can be noticed that the two equations (5. The next example is provided to illustrate this concept. (5. Scv ρ Urn dA (5. Schematics of ﬂow in in pipe with varying density as a function time for example 5. In the same manner. MASS CONSERVATION It has to be emphasized that the density is taken at the surface thus the subscript s. t) = 1− cos . The negative sign is because ﬂow out marked positive which reduces of the mass (negative derivative) in the control volume.4.9) applying negative value to keep the convention. the ﬂow rate in is d mb = dt ρs Urn dA Sc. Example 5. due to temperature variation and other reasons.10) Equation (5. taking the positive or negative value of Urn with integration of the entire system as d mb d ma − = dt dt ρs Urn dA Scv (5. Substituting equation (5.v. The change of mass change inside the control volume is net ﬂow in or out of the control system.

and the mass in the conduit as function of time. (5.10) can enter the integral since the boundaries are ﬁxed in time and hence. 5.2 Constant Density Fluids Further simpliﬁcations of equations (5.3. d dt ρ0 1 − c.10) can be obtained by assuming constant density and the equation (5. x 1− L 2 cos t t0 π R2 dx The density is not a function of radius.v.v. the derivative in equation (5. .10). θ and they can be taken out the integral as d dt which results in A ρdV = π R2 c. Vc. Express the mass ﬂow in and/or out.10).3.10) become conservation of the volume. x L 2 cos t t0 dx Flow Out = π R2 d dt L ρ0 1 − 0 x L 2 cos t π R2 L ρ0 dx = − sin t0 3 t0 t t0 The ﬂow out is a function of length.4 length is L and its area is A. Solution Here it is very convenient to choose a non-deformable control volume that is inside the conduit dV is chosen as π R2 dx.v. Write the expression for the mass change in the pipe.v. Using equation (5. and time. ρ0 c. L.11) Equation (5.1 Non Deformable Control Volume When the control volume is ﬁxed with time. dρ dV = − dt ρ Urn dA Sc.v. and is the change of the mass in the control volume. t. r and angle. CONTINUITY EQUATION 125 The conduit shown in Figure 5.3. End Solution 5.5.v. the ﬂow out (or in) is ρ(t) dV d dt d ρdV = dt c.11) is simpler than equation (5.

=0 =0 d dt ρ dV = c. And where Aj is the area of jet when touching the liquid boundary in bucket.v.2 Deformable Control Volume The left hand side of question (5.15) The meaning of the equation (5.3. 5.5. Ubn dA = Sc.v. the meaning is that volume ﬂow rate in and the volume ﬂow rate out have to equal.14) where Ub is the boundary velocity and Ubn is the normal component of the boundary velocity.v. Urn dA (5.15) is the net growth (or decrease) of the Control volume is by net volume ﬂow into it.v. Hence. The average velocity of the liquid at the exit of the ﬁlling pipe is Up and cross section of the pipe is Ap .2: Liquid ﬁlls a bucket as shown in Figure 5.126 5. Example 5.2 illustrates this point. Sc. and hence the mass change of the control volume is zero. Physically.2. The liquid ﬁlls a bucket with cross section area of A and instantaneous height is h. Sc. c. (5. Assume that the density is constant and at the boundary interface Aj = 0.2.12) Vrn dA = Sin Sout Vrn dA = 0 (5. The last assumption is result of the energy equation (with .1 CHAPTER 5. dρ dV +ρ dt n · Ub dA = ρ ˆ Sc. MASS CONSERVATION Non Deformable Control Volume For this case the volume is constant therefore the mass is constant. This condition can be written mathematically as =0 d −→ dt or in a more explicit form as Vrn dA = 0 Sc. Example 5. Ubn dA (5. the net ﬂow (in and out) is zero.13) Notice that the density does not play a role in this equation since it is canceled out.10) can be examined further to develop a simpler equation by using the extend Leibniz integral rule for a constant density and result in thus.3. Find the height as a function of the other parameters.7 Ap .v.v.v.

v. The relationship is function of the distance of the pipe from the boundary of the liquid.v.v. The control volume around the jet is deformable because the length of the jet shrinks with the time. This assumption is a strong assumption for certain conditions but it will be not discussed here since it is advance topic. there are no evaporation or condensation processes. several assumptions must be made. Urn dA where Ubn is the perpendicular component of velocity of the boundary.2. the ratio is determined by height of the pipe from the liquid surface in the bucket. (Uj + Ub ) dA . c. Fourth. Filling of the bucket and choices of the deformable control volumes for example 5. Calculate the bucket liquid interface velocity. In reality. some inﬂuence of momentum equation). ﬂow in = c. Solution This problem requires two deformable control volumes.5. this eﬀect can be neglected for this range which this problem. The mass conservation of the liquid in the bucket is boundary change Ubn dA c.3. However. CONTINUITY EQUATION 127 Up Ap Ub Aj h Uj A Fig. In this analysis. First. Substituting the known values for Urn results in Urn Ub dA = c. the air eﬀects are negligible. Third.v. -5. the liquid in the bucket has a straight surface. Second. The ﬁrst control is around the jet and second is around the liquid in the bucket. no liquid leaves the jet and enters the air.5.

the velocity will be Ub = Up in the limiting case and not inﬁnity.II.7 Ap results Up Ap − Ub A = −0.II. End Solution Example 5. e. mi . the side. c.128 CHAPTER 5.b) The above two equations (5.II.v. When examining solution there are two limits.II.a) To ﬁnd the jet velocity. The other limit is that and Ap /A −→ 0 then Ap Ub = A which is the result for the “intuitive” solution. Solution The applicable equation is Ubn dA = c.v.3: Balloon is attached to a rigid supply in which is supplied by a constant the mass rate. The reason for this diﬀerence is that the liquid already ﬁll the bucket and has not to move into bucket. It also interesting to point out that if the ﬁlling was from other surface (not the top surface).7 Ap Ub (5.II.c) The solution of equation (5.c) is Ub = Ap A − 0. the second control volume around the jet is used as the following ﬂow in ﬂow out boundary change Up Ap − Aj (Ub + Uj ) = −Aj Ub (5.II.b) and using the ratio of Aj = 0. MASS CONSERVATION The integration can be carried when the area of jet is assumed to be known as Ub A = Aj (Uj + Ub ) (5. Urn is Urn = −Up 0 @ the valve every else . Calculate the velocity of the balloon boundaries assuming constant density.a) into (5.7 which is Ap Ub = =∞ 0 The physical meaning is that surface is ﬁlled instantly. The ﬁrst limit is when Ap = A/0. Uj .II.b) are enough to solve for the two unknowns. (5.II.g. Substituting the ﬁrst equation.7 Ap It is interesting that many individuals intuitively will suggest that the solution is Ub Ap /A. Urn dA The entrance is ﬁxed. thus the relative velocity.a) and (5.

(Ubr r) · n dA ˆ ˆ The ﬁrst integral is zero because it is like movement of solid body and also yield this value mathematically (excises for mathematical oriented student). (Ux x) · n dA + ˆ ˆ Sc.v. Substituting into the general equation yields A ρ 4 π r2 Ubr = ρ Up Ap = mi Hence.2.3.5.v.v.3. The total velocity of boundary is Ut = mi (ˆ + r) x ˆ ρ 4 π r2 It can be noticed that the velocity at the opposite to the connection to the rigid pipe which is double of the center velocity. The right side of equation (5. End Solution 5. The main assumption made in this model is that the proprieties in the across section are only function of x coordinate . CONTINUITY EQUATION 129 Assume equal distribution of the velocity in balloon surface and that the center of the balloon is moving. Ubr = mi ρ 4 π r2 The center velocity is (also) exactly Ubr .16) . This assumptions leads d ρ2 U2 dA − ρ1 U1 dA = dt A2 A1 dV ρ(x) A(x) dx V (x) (5. The second integral (notice n = r) yields ˆ ˆ (Ubr r) · n dA = 4 π r2 Ubr ˆ ˆ Sc. Sc. thus the velocity has the following form Ub = Ux x + Ubr r ˆ ˆ Where x is unit coordinate in x direction and Ux is the velocity of the center and where ˆ r is unit coordinate in radius from the center of the balloon and Ubr is the velocity in ˆ that direction.v.3 One–Dimensional Control Volume Additional simpliﬁcation of the continuity equation is of one dimensional ﬂow. This simpliﬁcation provides very useful description for many ﬂuid ﬂow phenomena.15) is the net change due to the boundary is center movement net boundary change (Ux x + Ubr r) · n dA = ˆ ˆ ˆ Sc.

MASS CONSERVATION When the density can be considered constant equation (5. h as function of the time. The mass ﬂow rate out is √ function of the height. min h Fig. continuity equation is at its minimum form of U1 A1 = A2 U2 (5. Assume that the height at time zero is h0 . What happen if the h0 = 0? .16) to become ρ2 U2 dA = A2 A1 ρ1 U1 dA (5.6.18) For steady state and uniform density and velocity equation (5. For the ﬁrst case.16).19) For incompressible ﬂow (constant density). determine the height.20) The next example is of semi one–dimensional example to illustrate equation (5.18) reduces further to ρ1 A1 U1 = ρ2 A2 U2 (5. Example 5. -5. Is there a critical value and then if exist ﬁnd the critical value of the system parameters.4: Liquid ﬂows into tank in a constant mass ﬂow rate of a.16) is reduced to U2 dA − A2 A1 U1 dA = d dt A(x)dx (5.4. Height of the liquid for example 5. First assume that qout = b h second Assume as qout = b h.17) For steady state but with variations of the velocity and variation of the density reduces equation (5.130 CHAPTER 5.

For case one the right hand side term in equation (5. the critical ratio state if the ﬂow in is larger or lower than the ﬂow out determine the condition of the height. However. It can be noticed that control volume satisfy the demand of one dimensional since the ﬂow is only function of x coordinate. the height will increase.16) is ﬂow out ﬂow in √ dh ρL = b h − mi dt with the general solution of √ hb ln −1 mi √ √ mi hb hb + − 1 = (t + c) ρL mi 2ρL . If the reverse case appeared.3. b t − ρ1L e e h b1 = m1 e − ρ1L b t +c e b1 t ρL With the initial condition that at h(t = 0) = h0 the constant coeﬃcient can be found as h0 b 1 h0 b1 = 1 − c =⇒ c = 1 − m1 mi which the solution is h b1 = m1 e − ρ1L b t + 1− h0 b1 mi e b1 t ρL 0 0 It can be observed that if 1 = hmb1 is the critical point of this solution.16) is d L dh ρ h dx = ρ L dt 0 dt Substituting into equation equation (5.5. The solution can rearranged to a new form (a discussion why this form is preferred will be provided in dimensional chapter). If the term hmb1 i i is larger than one then the solution reduced to a negative number. CONTINUITY EQUATION Solution 131 The control volume for both cases is the same and it is around the liquid in the tank. negative number for height is not possible and the height solution approach zero.16) is dh ρL = dt solution is h= ﬂow out b1 h ﬂow in − mi private solution homogeneous solution b1 t mi + c1 ρ L b1 The solution has the homogeneous solution (solution without the mi ) and the solution of the mi part. the governing equation (5. Essentially. For second case.

To make the previous derivation clearer. a proof will be provided and the physical meaning will be explained. y) dx = x1 (y) x1 (y) dx2 dx1 ∂f dx + f (x2 . the Reynolds Transport Theorem will be reproofed and discussed. MASS CONSERVATION The constant is obtained when the initial condition that at h(t = 0) = h0 and it left as exercise for the reader. y) = ∂G ∂x (5.4 Reynolds Transport Theorem It can be noticed that the same derivations carried for the density can be carried for other intensive properties such as speciﬁc entropy. This author ﬁnd material just given so no questions will be asked.archive. Nevertheless. c. The change of accumulative property will be then D Dt f ρdV = sys d dt f ρdV + c. The ideas are the similar but extended some what. For its derivative of equation (5. y) ∂y dy dy (5.org/details/papersonmechanic01reynrich.21) This theorem named after Reynolds.132 CHAPTER 5.22) Initially. y) − f (x1 . it will be provided. if there will be a demand for such. 3 There was a suggestion to insert arbitrary constant which will be canceled and will a provide rigorous proof. speciﬁc enthalpy. Suppose that g is intensive property (which can be a scalar or a vector) undergoes change with time. Osborne.24) diﬀerentiating (chain rule d uv = u dv + v du) by part of left hand side of the Leibniz integral rule (it can be shown which are identical) is 1 2 3 4 d [G(x2 . End Solution 5. . Leibniz integral rule2 is an one dimensional and it is deﬁned as d dy x2 (y) x2 (y) f (x.v f ρ Urn dA (5. y) = f (α. y) dα (5.23) Notice that lower boundary of the integral is missing and is only the upper limit of the function is present3 . This is engineering book and thus.v. Assume that there is a function that satisfy the following x G(x. material is not necessarily but is added her for completeness. y) dy ∂x2 dy ∂y ∂x1 dy ∂y 1 These 2 This (5.25) papers can be read on-line at http://www. the exact mathematical proof is not the concern here. y) − G(x1 .23) is f (x. y)] ∂G dx2 ∂G ∂G dx1 ∂G = + (x2 . y) − − (x1 . (1842-1912) which is actually a three dimensional generalization of Leibniz integral rule1 .

28) For one dimensional situation the change with time is DF D = Dt Dt ρ f A(x)dx sys (5. The same can be said for the other side.26) The ﬁrst term (1) in equation (5.31) The change with time of the accumulative property. The interesting information that commonly needed is the change of the accumulative property. F .30) Fout = f2 ρ Urn (5. Since the change is very short (diﬀerential). is then F1 dx1 dt Fin = f1 ρ Urn The accumulative ﬂow of the property out.25) is dx2 ∂G dx2 = f (x2 . the mass will be diﬀerent and it will not be a system. Urn = U1 − Ub . F . between the boundaries is d dt ρ(x) f A(x) dA c.4.29) If two limiting points (for the one dimensional) are moving with a diﬀerent coordinate system. The above “proof” is mathematical in nature and physical explanation is also provided. The accumulative ﬂow of the property in. F . REYNOLDS TRANSPORT THEOREM 133 The terms 2 and 4 in equation (5. with time. This limiting condition is the control volume for which some of the mass will leave or enter.25) are actually (the x2 is treated as a diﬀerent variable) x2 (y) x1 (y) ∂ f (x. (5. Thus this explanation is a proof the Leibniz rule.5. Suppose that a ﬂuid is ﬂowing in a conduit. The intensive property. F . F . The change with time is DF D = Dt Dt ρ f dV sys (5.27) The same can be said for the third term (3). is then F2 dx2 dt (5. y) dx ∂y (5. y) ∂x2 dy dy (5. f is investigated or the accumulative property.v. the ﬂow in (or out) will be the velocity of ﬂuid minus the boundary at x1 .32) .

Reynolds Transport theorem is a generalization of the Leibniz rule and thus the same arguments are used. What is the magical averaged velocity at the exit? Assume no–slip condition.5: Liquid enters a circular pipe with a linear velocity proﬁle as a function of the radius with maximum velocity of Umax . the boundary condition is U (r = R) = 0 and U (r = 0) = Umax Therefore the velocity proﬁle is r U (r) = Umax 1 − R Where R is radius and r is the working radius (for the integration).a) is R2 = Uave π R2 6 The solution of equation (b) results in average velocity as Umax π Uave = Umax 6 (5.V.v Sc.b) (5.a) The integration of the equation (5. The above discussion is one of the physical meaning the Leibniz rule. is arbitrary and it can be replaced by any letter. t. Solution The velocity proﬁle is linear with radius. Thus. The only diﬀerence is that the velocity has three components and only the perpendicular component enters into the calculations.13).V. D DT d dt f ρdV = sys f ρ dV + c. Additionally.v. Since the time variable. later a discussion on relationship between velocity at interface to solid also referred as the (no) slip condition will be provided. It will be assumed that the velocity at the interface is zero. This assumption is good for most cases with very few exceptions. For which R Umax 1 − 0 r R 2 π r dr = Uave π R2 (5. Example 5.33) 5. After magical mixing.V. the velocity became uniform. The magical averaged velocity is obtained using the equation (5.5 Examples For Mass Conservation Several examples are provided to illustrate the topic. f ρ Urn dA (5.c) End Solution . Write the equation which describes the velocity at the entrance. MASS CONSERVATION When put together it brings back the Leibniz integral rule.V.134 CHAPTER 5.

No mass can cross the lower surface (solid boundary). calculate the relationship of the mass transfer across the control volume. and upper. The slowed liquid is slowing the layer above it. The control volume has three surfaces that mass can cross. Boundary Layer control mass. In this case.” The integral simply multiply by negative one. A common boundary layer analysis uses the Reynolds transform theorem.13) results in x direction y direction out in δ δ U0 dy − 0 0 U0 y dy = δ L U xdx 0 It can be noticed that the convention used in this chapter of “in” as negative is not “followed. -5.7.5. Solution Assuming the velocity proﬁle is linear thus. Compare the two diﬀerent velocity proﬁles aﬀecting on the mass transfer. The assumption is that ﬂuid attaches itself to surface. The boundary layer is growing with x because the boundary eﬀect is penetrating further into ﬂuid. Example 5. The situation is steady state and thus using equation (5. the left. For simplicity assume slowed ﬂuid has a linear velocity proﬁle.6: Experiments have shown that a layer of liquid that attached itself to the surface and it is referred to as boundary layer. The above integrals on the . Then assume parabolic velocity proﬁle as Ux (y) = 2 U0 y 1 + δ 2 y δ 2 and calculate the mass transfer across the control volume.5. Where δ is the height of the boundary layer at exit point of the ﬂow as shown in Figure 5.7. (to satisfy the boundary condition) it will be U0 y Ux (y) = δ The chosen control volume is rectangular of L × δ. right. EXAMPLES FOR MASS CONSERVATION (1) (2) 135 U0 ge Ed of B n ou yL dr er ay L Fig.

The tank also has three emptying valves of 5 hours. The next example deal with such reversed mass ﬂow rate.1 ) 5.1 m2 with velocity of 500 m/sec. Example 5. However sometime. Thus.1 kg/sec.7: Air ﬂows into a jet engine at 5 kg/sec while fuel ﬂow into the jet is at 0.1 kg/sec = = 1.02kg/m3 AU 0.01 m2 500 m/sec End Solution The mass (volume) ﬂow rate is given by direct quantity like x kg/sec. 7 hours. the ﬂow out is ( 5 + 0. Is there a combination of valves that make the tank at steady state? .13) is used.136 right hand side can be combined as δ CHAPTER 5. calculate the time for tank reach empty or full state when all the valves are open. the mass (or the volume) is given by indirect quantity such as the eﬀect of ﬂow.1 kg/sec The density is ρ= m ˙ 5. and 8 hours. MASS CONSERVATION U0 1 − 0 y δ L dy = 0 U xdx the integration results in U0 δ = 2 or for parabolic proﬁle δ δ L U xdx 0 U0 dy − 0 0 δ U0 y y + δ δ y y − δ δ L 2 L dy = 0 U xdx or 0 U0 1 − the integration results in 2 dy = U0 U0 δ = 2 U xdx 0 End Solution Example 5. The tank is 3/4 fulls.8: The tank is ﬁlled by two valves which one ﬁlled tank in 3 hours and the second by 6 hours. What is the density of the gases at the exhaust? Solution The mass conservation equation (5. The burned gases leaves at the exhaust which has cross area 0.

For simplicity. dρ dV dt ρ Ub dV ρUrn dA Every term in the above equation is analyzed but ﬁrst the equation of state and volume to pressure relationship have to be provided.v.5. Solution The applicable equation is increase pressure Vc. Calculate the cylinder boundaries velocity. The gas inside the cylinder obeys the ideal gas law. For example valve of 3 hours can be converted to 1/3 tank per hour.v boundary velocity + Sc. Thus.v. End Solution Example 5.9: Inﬂated cylinder is supplied in its center with constant mass ﬂow. assume that the process is isothermal. EXAMPLES FOR MASS CONSERVATION Solution 137 Easier measurement of valve ﬂow rate can be expressed as fraction of the tank per hour. ρ= P RT and relationship between the volume and pressure is P = f π Rc 2 . Assume that the cylinder inﬂated uniformly and pressure inside the cylinder is uniform. The time to completely ﬁlled the tank is 1 70 4 = hour 159 1 131 − 2 280 The rest is under construction. in or out ﬂow rate = Sc.5. Assume that the gas mass is supplied in uniformed way of mi [kg/m/sec]. mass ﬂow rate in is min = 1/3 + 1/6 = 1/2tank/hour ˙ The mass ﬂow rate out is mout = 1/5 + 1/7 + 1/8 = ˙ 131 280 Thus. The pressure inside the cylinder is linearly proportional to the volume. if all the valves are open the tank will be ﬁlled.

Where fv is a coeﬃcient describing the balloon physical characters. Combining the above two equations results in f π Rc 2 ρ= RT Where f is a coeﬃcient with the right dimension. MASS CONSERVATION Where Rc is the instantaneous cylinder radius. 4 The proof of this idea is based on the chain diﬀerentiation similar to Leibniz rule. Assume that balloon volume is a linear function of the pressure inside the balloon such as P = fv V . Assume that gas obeys the ideal gas law.138 CHAPTER 5. When the derivative of the second part is dUb /dRc = 0.v. A f π 3 Rc 2 RT Ub substituting in the governing equation obtained the form of f π 2 Rc 3 4 f π 2 Rc 3 Ub + Ub = mi RT 3RT The boundary velocity is then Ub = mi 3 mi R T G= 7 f π 2 Rc 3 7 f π 2 Rc 3 3RT End Solution Example 5. .10: A balloon is attached to a rigid supply and is supplied by a constant mass rate.v 2 f π Rc Ub RT 2 π Rc dRc dV = 4 f π 2 Rc 3 Ub 3RT The integral can be carried when Ub is independent of the Rc 4 The second term is ρ f π Rc 2 ρ Ub dA = Ub 2 πRc = RT Sc. mi . Calculate the velocity of the balloon boundaries under the assumption of isothermal process.v Vc. It also can be noticed that boundary velocity is related to the radius in the following form Ub = dRc dt The ﬁrst term requires to ﬁnd the derivative of density with respect to time which is Ub dρ d = dt dt Thus the ﬁrst term is dρ dV = dt 2 π Rc f π Rc RT 2 = 2 f π Rc dRc RT dt Vc.

v The right hand side of the above equation is ρUrn dA = mi Sc.v. The ideal gas law is ρ= P RT 139 The relationship between the pressure and volume is P = fv V = 4 fv π Rb 3 3 The combining of the ideal gas law with the relationship between the pressure and volume results 4 fv π Rb 3 ρ= 3RT The applicable equation is dρ dV + dt ρ (Uc x + Ub r) dA = ˆ ˆ Sc. Sc. .v.5.v. EXAMPLES FOR MASS CONSERVATION Solution The question is more complicated than Example 5.10. The density change is Ub dρ 12 fv π Rb dRb = dt RT dt The ﬁrst term is =f (r) Rb 0 2 12 fv π Rb 2 16 fv π 2 Rb 5 Ub 4 π r2 dr = Ub RT 3RT dV The second term is 8 fv π 2 R b 5 4 fv π R b 3 4 fv π R b 3 Ub dA = Ub 4 π Rb 2 = Ub 3RT 3RT 3RT A A Subsisting the two equations of the applicable equation results Ub = 1 mi R T 8 fv π 2 R b 5 Notice that ﬁrst term is used to increase the pressure and second the change of the boundary. ρUrn dA Vc.5.

140

**CHAPTER 5. MASS CONSERVATION
**

End Solution

Open Question: Answer must be received by April 15, 2010 The best solution of the following question will win 18 U.S. dollars and your name will be associated with the solution in this book. Example 5.11: Solve example 5.10 under the assumption that the process is isentropic. Also assume that the relationship between the pressure and the volume is P = fv V 2 . What are the units of the coeﬃcient fv in this problem? What are the units of the coeﬃcient in the previous problem?

**5.6 The Details Picture – Velocity Area Relationship
**

The integral approach is intended to deal with the “big” picture. Indeed the method is used in this part of the book for this A purpose. However, there is very little written about the usability of this approach to provide way to calculate the average quantities in the control system. Sometimes hℓ z y Ae it is desirable to ﬁnd the averaged velocx Ue ity or velocity distribution inside a control volume. There is no general way to provide these quantities. Therefore an example will be provided to demonstrate the use Fig. -5.8. Control volume usage to calculate of this approach. local averaged velocity in three coordinates. Consider a container ﬁlled with liquid on which one exit opened and the liquid ﬂows out as shown in Figure 5.8. The velocity has three components in each of the coordinates under the assumumption that ﬂow is uniform and the surface is straight5 . The integral approached is used to calculate the averaged velocity of each to the components. To relate the velocity in the z direction with the ﬂow rate out or the exit the velocity mass balance is constructed. A similar control volume construction to ﬁnd the velocity of the boundary velocity (height) can be carried out. The control volume is bounded by the container wall including the exit of the ﬂow. The upper boundary is surface parallel to upper surface but at Z distance from the bottom. The mass balance reads dρ dV + dt Ubn ρ dA +

A A

Urn ρ dA = 0

(5.34)

V

5 The liquid surface is not straight for this kind of problem. However, under certain conditions it is reasonable to assume straight surface which have been done for this problem.

5.6. THE DETAILS PICTURE – VELOCITY AREA RELATIONSHIP For constant density (conservation of volume) equation6 and (h > z) reduces to Urn ρ dA = 0

A

141

(5.35)

In the container case for uniform velocity equation 5.35 becomes Uz A = Ue Ae =⇒ Uz = − Ae Ue A (5.36)

It can be noticed that the boundary is not moving and the mass inside does not change this control volume. The velocity Uz is the averaged velocity downward. The x component of velocity is obY control Volume Volume tained by using a diﬀerent control volume. Ax− X controlpage into the page into the The control volume is shown in Figure 5.9. The boundary are the container far from Ay − y the ﬂow exit with blue line projection into x page (area) shown in the Figure 5.9. The mass conservation for constant density of this control volume is −

A

Ue

Ae

Ubn ρ dA +

A

Urn ρ dA = 0 (5.37)

Fig. -5.9. Control volume and system before and after the motion.

Usage of control volume not included in the previous analysis provides the velocity at the upper boundary which is the same as the velocity at y direction. Substituting into (5.37) results in Ae Ue ρ dA + A Ux ρ dA = 0

Ayz

(5.38)

Ax

−

Where Ax − is the area shown the Figure under this label. The area Ayz referred to area into the page in Figure 5.9 under the blow line. Because averaged velocities and constant density are used transformed equation (5.38) into Ae − Ax Ue + Ux Y (x) h = 0 A

Ayz

(5.39)

Where Y (x) is the length of the (blue) line of the boundary. It can be notice that the velocity, Ux is generally increasing with x because Ax − increase with x. The calculations for the y directions are similar to the one done for x direction. The only diﬀerence is that the velocity has two diﬀerent directions. One zone is right to the exit with ﬂow to the left and one zone to left with averaged velocity to right. If the volumes on the left and the right are symmetrical the averaged velocity will be zero.

6 The

point where (z = h) the boundary term is substituted the ﬂow in term.

142

CHAPTER 5. MASS CONSERVATION

Example 5.12: Calculate the velocity, Ux for a cross section of circular shape (cylinder). Solution The relationship for this geometry needed to be expressed. The length of the line Y (x) is Y (x) = 2 r 1− 1− x r

2

Ax− Y(x)

(r − x) x α

y

r

(5.XII.a)

Ue

Ae

This relationship also can be expressed in the term of α as Y (x) = 2 r sin α (5.XII.b)

Fig. -5.10. Circular cross section for ﬁnding Ux and various cross sections.

Since this expression is simpler it will be adapted. When the relationship between radius angle and x are x = r(1 − sin α) (5.XII.c) The area Ax − is expressed in term of α as Ax − = Thus the velocity, Ux is Ae A α− 1 sin(2α) r2 Ue + Ux 2 r sin α h = 0 2 Ae r α − 1 sin(2α) 2 Ue A h sin α 1 S (5.XII.e) (5.XII.f) 1 α − , sin(2α) r2 2 (5.XII.d)

Ux = Averaged velocity is deﬁned as

Ux =

U dS

S

(5.XII.g)

Where here S represent some length. The same way it can be represented for angle calculations. The value dS is r cos α. Integrating the velocity for the entire container and dividing by the angle, α provides the averaged velocity. Ux = which results in Ux = 1 2r

π 0

**Ae r α − 1 sin(2α) 2 Ue r dα A h tan α (π − 1) Ae r Ue 4 A h
**

End Solution

(5.XII.h)

(5.XII.i)

Example 5.13:

5.7. MORE EXAMPLES FOR MASS CONSERVATION Calculate the velocity, Uy for a cross section of circular shape (cylinder). What is the averaged velocity if only half section is used. State your assumptions and how it similar to the previous example. Solution

X(y)

x

143

y

r (r − x)

Ay − Ue Ae

Fig. -5.11. y velocity for a circular shape

The ﬂow out in the x direction is zero because symmetrical reasons. That is the ﬂow ﬁeld is a mirror images. Thus, every point has diﬀerent velocity with the same value in the opposite direction. The ﬂow in half of the cylinder either the right or the left has non zero averaged velocity. The calculations are similar to those in the previous to example 5.12. The main concept that must be recognized is the half of the ﬂow must have come from one side and the other come from the other side. Thus, equation (5.39) modiﬁed to be Ae − Ax Ue + Ux Y (x) h = 0 A The integral is the same as before but the upper limit is only to π/2 Ux = which results in Ux = (π − 2) Ae r Ue 8 A h

End Solution

Ayz

(5.40)

1 2r

π/2 0

Ae r α − 1 sin(2α) 2 Ue r dα A h tan α

(5.XIII.a)

(5.XIII.b)

**5.7 More Examples for Mass Conservation
**

Typical question about the relative velocity that appeared in many ﬂuid mechanics exams is the following. Example 5.14:

144 CHAPTER 5. MASS CONSERVATION A boat travels at speed of 10m/sec upstream in a river that ﬂows at a speed of 5m/s. The inboard engine uses a pump to suck in water at the front Ain = 0.2 m2 and eject it through the back of the boat with exist area of Aout = 0.05 m2 . The water absolute velocity leaving the back is 50m/sec, what Fig. -5.12. Schematic of the are the relative velocities entering and leaving the boat for example 5.14 boat and the pumping rate?

Us = 5[m/sec] Uo = 50[m/sec] Ub = 10[m/sec]

Solution The boat is assumed (implicitly is stated) to be steady state and the density is constant. However, the calculation have to be made in the frame of reference moving with the boat. The relative jet discharge velocity is Urout = 50 − (10 + 5) = 35[m/sec] The volume ﬂow rate is then Qout = Aout Urout = 35 × 0.05 = 1.75m3 /sec The ﬂow rate at entrance is the same as the exit thus, Urin = Aout 0.05 Urout = 35 = 8.75m/sec Ain 0.2

End Solution

**CHAPTER 6 Momentum Conservation for Control Volume
**

6.1 Momentum Governing Equation

6.1.1 Introduction to Continuous

In the previous chapter, the Reynolds Transport Theorem (RTT) was applied to mass conservation. Mass is a scalar (quantity without magnitude). This chapter deals with momentum conservation which is a vector. The Reynolds Transport Theorem (RTT) is applicable to any quantity and the discussion here will deal with forces that acting on the control volume. Newton’s second law for single body is as the following F = U d(mU ) dt (6.1)

It can be noticed that bold notation for the velocity is U (and not U ) to represent that the velocity has a direction. For several bodies (n), Newton’s law becomes

n n

Fi =

i=1 i=1

U d(mU )i dt

(6.2)

The ﬂuid can be broken into inﬁnitesimal elements which turn the above equation (6.2) into a continuous form of small bodies which results in

n

Fi =

i=1

D Dt

element mass U ρ dV

sys

(6.3)

145

146

CHAPTER 6. MOMENTUM CONSERVATION

Note that the notation D/Dt is used and not d/dt to signify that it referred to a derivative of the system. The Reynold’s Transport Theorem (RTT) has to be used on the right hand side.

6.1.2

External Forces

First, the terms on the left hand side, or the forces, have to be discussed. The forces, excluding the external forces, are the body forces, and the surface forces as the following F total = F b + F s (6.4)

In this book (at least in this discussion), the main body force is the gravity. The gravity acts on all the system elements. The total gravity force is

element mass

Fb =

sys

g ρ dV

(6.5)

which acts through the mass center towards the center of earth. After inﬁnitesimal time the gravity force acting on the system is the same for control volume, hence, g ρ dV =

sys cv

g ρ dV

(6.6)

The integral yields a force trough the center mass which has to be found separately. In this chapter, the surface forces are with the divided into two categories: one perpendisurface n ˆ perpendicular to cular to the surface and one with the surthe surface face direction (in the surface plain see Figure 6.1.). Thus, it can be written as Fs =

c.v.

Sn dA +

c.v.

τ dA (6.7)

Fig. -6.1. The explaination for the direction relative to surface perpendicular and with the surface.

Where the surface “force”, Sn , is in the surface direction, and τ are the shear stresses. The surface “force”, Sn , is made out of two components, one due to viscosity (solid body) and two consequence of the ﬂuid pressure. Here for simplicity, only the pressure component is used which is reasonable for most situations. Thus,

∼0

Pˆ S n = −P n + Sν

(6.8)

Where Sν is perpendicular stress due to viscosity. Again, n is an unit vector outward ˆ of element area and the negative sign is applied so that the resulting force acts on the body.

In Cartesian coordinate. c. is ρ U dV = sys ρ U dV + c. etc.v.v.v. is measured in the frame of reference and U rn is the liquid relative velocity to boundary of the control volume measured in the same frame of reference.v.v.11) is a vector equation which can be broken into its three components.11) ρ U dV + c. t = dt c. c.6.10) is transformed to Integral Momentum Equation & External Forces F ext + g ρ dV − P · dA + τ · dA = c.v. 6.v. c. Thus.v.v.1. the components are Fx + c. (6.12) t dt ρ U x dV + c.9) The liquid velocity. ρ U Urn dV The external forces. c.v. (6.v. according Reynolds Transport Theorem (RTT).13) . c. P cos θx dA + c.v. ducts. supporting solid structures. These external forces are commonly associated with pipe. are the forces resulting from support of the control volume by non–ﬂuid elements. where θx is the angle between n and ˆ or (ˆ · ˆ ˆ i n i).v. Fext .10) ρ U Urn dV ρ U dV + c.3 Momentum Governing Equation D Dt t dt The right hand side. for example in the x coordinate. ρ U U rn dA (6.v. c. the general form of the momentum equation without the external forces is Integral Momentum Equation g ρ dV − P dA + τ · dA c. U . g · ˆ ρ dV i c. With external forces equation (6. t dt c. Equation (6. MOMENTUM GOVERNING EQUATION 147 6.1. τ x · dA = ρ U x · U rn dA (6.4 Momentum Equation in Acceleration System For accelerate system. the right hand side has to include the following acceleration r ˙ a acc = ω × (r × ω) + 2 U × ω + r × ω − a 0 (6.v.v.1.v. friction (non-ﬂuid).

1.148 CHAPTER 6. τ dA = c. g ρ dV − c.5 Momentum For Steady State and Uniform Flow The momentum equation can be simpliﬁed for the steady state condition as it was shown in example 6. Thus. ρ U Urn dA (6.3.1.17) 1 A [U (r)] dA A 2 (6. equation (??) is further reduced to Urn Urn F = out U U ˆ ρU (U · n) dA − in U U ˆ ρU (U · n) dA (6. ﬂow is exposed to the atmosphere and thus (almost) uniform pressure surrounding the control volume.16) In situations where the velocity is provided and known (remember that density is constant) the integral can be replaced by F = mU o − mU i ˙U ˙U The average velocity is related to the velocity proﬁle by the following integral U = 2 (6.1 Momentum for For Constant Pressure and Frictionless Flow Another important sub category of simpliﬁcation deals with ﬂow under approximation of the frictionless ﬂow and uniform pressure. This situation is a similar idea that was shown in Section 4.v. the mass ﬂow rate in and out are equal. .v. The unsteady term (where the time derivative) is zero.v. In this case the resulting force due to the pressure is zero to all directions. Integral Steady State Momentum Equation F ext + c.v. 6. a acc ρ dV (6. P dA + c.18) Equation (6.14) Integral of Uniform Pressure on Body In this kind of calculations.6. it common to obtain a situation where one of the term will be an integral of the pressure over the body surface. in these situations.v. This kind of situations arise when friction (forces) is small compared to kinetic momentum change.18) is applicable to any velocity proﬁle and any geometrical shape.5. Additionally.15) 6. MOMENTUM CONSERVATION Where r is the distance from the center of the frame of reference and the add force is F add = Vc. In this situation.

2).1: Calculate the average velocity for the given parabolic velocity proﬁle for a circular pipe. Solution The velocity proﬁle is U r R = Umax 1 − r R 2 (6.2. Neglect the friction. Fig. -6.2. Umax U= √ 6 End Solution 2 1 2 π R2 1 0 R 0 [U (r)] 2 π r dr 2 (6.a) into equation (6. What is the angle for which maximum force will be created? .I. MOMENTUM GOVERNING EQUATION 149 Example 6. Schematics of maximum angle for impinged by a jet. Example 6. Schematics of area impinged by a jet for example 6.b) 2 2 1 − r2 ¯ 2 rd¯ = ¯ r 1 2 (Umax ) 6 (6.I.6.2: A jet is impinging on a stationary surface by changing only the jet direction (see Figure 6.c) y x Uo Ui F Fig a. Uo θ Ui F Fig b. calculate the force and the angle which the support has to apply to keep the system in equilibrium.I.18) U = results in U = (Umax ) Thus.I.a) Substituting equation (6.1. Schematics of area impinged by a jet and angle eﬀects.

The equation for the x coordinate is Fx = m (cos θ Uo − Ui ) ˙ or since Ui = Uo Fx = m Ui (cos θ − 1) ˙ It can be observed that the maximum force. the mass ﬂow rate remains constant.a) can be explicitly written for the two coordinates. For small angle analysis is important in the calculations of ﬂow around thin wings. to Urn Urn F = out U U ˆ ρU (U · n) dA − in U U ˆ ˙U ˙U ρU (U · n) dA = mUo − mUi (6.II.3 with a mass . MOMENTUM CONSERVATION Equation (6.150 Solution CHAPTER 6.3: Liquid ﬂows through a symmetrical nozzle as shown in the Figure 6. Hence Fx |max = −2 m Ui ˙ and the force in the y direction is Fy = m Ui sin θ ˙ the combined forces are Ftotal = Which results in Ftotal = m Ui sin (θ/2) ˙ with the force angle of tan φ = π − Fy π θ = − Fx 2 2 Fx 2 + Fy 2 = m Ui ˙ (cos θ − 1) + sin2 θ 2 For angle between 0 < θ < π the maximum occur at θ = π and the minimum at θ ∼ 0. Fx occurs when cos θ = π. It can be proven by setting dFx /dθ = 0 which yields θ = 0 a minimum and the previous solution.11) can be reduced. End Solution Example 6.II. because it is a steady state. Equation (6.a) It can be noticed that even though the velocity change direction.

The exit pressure is 1[Bar].6. Solution U2 =? P2 = 1[Bar] A2 = 10[cm2] 151 z P2 = 3[Bar] A1 = 50[cm2] U1 = 5[m/sec] Fig. The entrance pressure is 3[Bar] and the entrance velocity is 5 [m/sec]. The exit velocity is uniform but unknown. The control volume does not cross any solid body (or surface) there is no external forces. -6.1. Hence.v.III.3. Nozzle schematic for the discussion on the forces and for example 6.v.3. forces on the nozzle Fnozzle solid surface P cos θz dA + c.v. =0 c.v. c.v. First.v. τ z dA = (6. MOMENTUM GOVERNING EQUATION ﬂow rate of 0.III.3. ρ U z · U rn dA .0015 [m3 ].v. The entrance area is 0.v. This situation is a steady state for constant density.0001[cm2 ]. =0 liquid surface Fz + c. τ z dA = c.b) c.v. What is the exit velocity? What is the force acting the nozzle? Assume that the density is constant ρ = 1000[kg/m3 ] and the volume in the nozzle is 0. ˆ g · k ρ dV + P cos θz dA + (6. ˆ g · k ρ dV + t dt P cos θz dA + c. c. Then A1 U1 = A2 U2 and after rearrangement.01 [gk/sec]. the exit velocity is U2 = A1 0.a) ρ U z · U rn dA ρ U z dV + c.v.0001 Equation (6.0005[m2 ] and the exit area is 0.0005 U1 = × 5 = 25[m/sec] A2 0. The chosen control volume is shown in Figure 6.12) is applicable but should be transformed into the z direction which is Fz + c. the velocity has to be found.

4.v. The part of the pressure which act on the nozzle in the z direction is − c. Of course it is only approximation but is provided a crude tool.v. fan etc. In the Figure 6. Combining all transform equation (6. ρ U z · U rn dA (6.2 Momentum Equation Application Momentum Equation Applied to Propellers The propeller is a mechanical devise that is used to increase the ﬂuid momentum.v. .v.III.III. P cos θz dA = c.v. ˆ g · k ρ dV + c. Here.c) into Fz = −g ρVnozzle + P A|2 − P A|1 + ρ U2 2 A2 − U1 2 A1 Fz = 9.III.III. MOMENTUM CONSERVATION All the forces that act on the nozzle are combined as Fnozzle + c.8 × 1000× End Solution (6. this analysis is used for academic purposes. Or there is a line (or surface) in which the ﬂuid outside changes only the ﬂow direction. Either it is assumed that some of the ﬂuid enters into the container and ﬂuid outside is not aﬀected by the propeller. A2 U2 (U2 ) dA − A1 U1 (U1 ) dA which results in ρ U z · U rn dA = ρ U2 2 A2 − U1 2 A1 c. The propeller can be stationary like in cooling tours. P dA = 1 P dA − 2 P dA = P A|1 − P A|2 The last term in equation (6.c) The second term or the body force which acts through the center of the nozzle is Fb = − c.152 CHAPTER 6. This surface is called slip surface. g · n ρ dV = −g ρVnozzle ˆ Notice that in the results the gravity is not bold since only the magnitude is used.v. ships and other devices (thrust) as shown in Figure 6. The other common used of propeller is mostly to move ﬂuids as a pump. For a steady state the analysis is simpler and used here to provide an example of steady state.4 the ﬂuid ﬂows from the left to the right.d) 6. The propeller analysis of unsteady is complicated due to the diﬃculty in understanding the velocity ﬁeld.v. Improvements can be made to this analysis. Many times it is used for propulsion purposes of airplanes.c) is ρ U z · U rn dA = c.

6.4: A sled toy shown in Figure ?? is pushed by liquid jet. The diﬀerence between the jets propulsion and propellers is based on the energy supplied. Hence.19) 153 1 U1 3 4 2 U2 Liquid Combining the control Fig.2. pressure drops from the calculation. Jet Propulsion Jet propulsion is a mechanism in which the air planes and other devices are propelled. Notice that Example 6. Thus. The propellers are moved by a mechanical work which is converted to thrust.20) This analysis provide way to calculate the work needed to move this propeller. the thermal energy is converted to thrust. The inlet area and exit area are diﬀerent for most jets and if the mass of the fuel is neglected then F = ρ A2 U2 2 − A1 U1 2 (6. Here.4. in many case more eﬃcient. 3 with (note that there are no external forces) with points 4 and 2 results in ρ U2 2 − U1 2 = P4 − P3 (6. Calculate the friction force on the . this direct conversion can be.21) An academic example to demonstrate how a steady state calculations are done for a moving control volume. -6. The analysis of such device in complicated and there is a whole class dedicated for such topic in many universities. the air is sucked into engine and with addition heating (burning fuel) the velocity is increased. MOMENTUM EQUATION APPLICATION As ﬁrst approximation. as it will be shown in the Chapter on compressible ﬂow it allows to achieve velocity above speed of sound. Further increase of the exit area with the increased of the burned gases further increase the thrust. and is. Propeller schematic to explain the change of movolume between points 1 and mentum due to velocity. a major obstacle in the past. Note that in this analysis it was assumed that the ﬂow is horizontal that z1 = z2 and/or the change is insigniﬁcant. a very limited discussion related to the steady state is oﬀered. Essentially. The one dimensional momentum equation is reduced F = ρ U2 2 − U1 2 (6. Furthermore. In Jet propulsion. the pressure around control volume is the same.

154

CHAPTER 6. MOMENTUM CONSERVATION

toy when the toy is at steady state with velocity, U0 . Assume that the jet is horizontal and the reﬂecting jet is vertical. The velocity of the jet is uniform. Neglect y the friction between the liquid (jet) and control x volume the toy and between the air and toy. U0 Uj Calculate the absolute velocity of the Ff jet exit. Assume that the friction between the toy and surface (ground) is Fig. -6.5. Toy Sled pushed by the liquid relative to the vertical force. The dyjet in a steady state for example 6.4. namics friction is µd .

2 1

Solution The chosen control volume is attached to the toy and thus steady state is obtained. The frame of reference is moving with the toy velocity, U 0 . The applicable mass conservation equation for steady state is A1 U1 = A2 U2 The momentum equation in the x direction is Ff +

c.v.

g ρ dV −

c.v.

P dA +

c.v.

τ dA =

c.v.

ρ U U rn dV

(6.IV.a)

The relative velocity into the control volume is U 1j = (Uj − U0 ) x ˆ The relative velocity out the control volume is U 2j = (Uj − U0 ) y ˆ The absolute exit velocity is U 2 = U0 x + (Uj − U0 ) y ˆ ˆ For small volume, the gravity can be neglected also because this term is small compared to other terms, thus g ρ dV ∼ 0

c.v.

**The same can be said for air friction as τ dA ∼ 0
**

c.v.

**The pressure is uniform around the control volume and thus the integral is P dA = 0
**

c.v.

6.2. MOMENTUM EQUATION APPLICATION The control volume was chosen so that the pressure calculation is minimized. The momentum ﬂux is ρ Ux Ui rn dA = A ρ U1j 2

Sc.v.

155

(6.IV.b)

The substituting (6.IV.b) into equation (6.IV.a) yields Ff = A ρ U1j 2 The friction can be obtained from the momentum equation in the y direction mtoy g + A ρ U1j 2 = Fearth According to the statement of question the friction force is Ff = µd mtoy g + A ρ U1j 2 The momentum in the x direction becomes µd mtoy g + A ρ U1j 2 = A ρ U1j 2 = A ρ (Uj − U0 ) The toy velocity is then U0 = Uj − µd mtoy g A ρ (1 − µd )

2

(6.IV.c)

**Increase of the friction reduce the velocity. Additionally larger toy mass decrease the velocity.
**

End Solution

6.2.1

Momentum for Unsteady State and Uniform Flow

The main problem in solving the unsteady state situation is that the control volume is accelerating. A possible way to solve the problem is by expressing the terms in an equation (6.10). This method is cumbersome in many cases. Alternative method of solution is done by attaching the frame of reference to the accelerating body. One such example of such idea is associated with the Rocket Mechanics which is present here.

FR

mf mR UR

Ug

Fig. -6.6. A rocket with a moving control volume.

156

CHAPTER 6. MOMENTUM CONSERVATION

6.2.2

Momentum Application to Unsteady State

Rocket Mechanics A rocket is a devise similar to jet propulsion. The diﬀerence is the fact that the oxidant is on board with the fuel. The two components are burned and the gases are ejected through a nozzle. This mechanism is useful for speciﬁc locations because it is independent of the medium though which it travels. In contrast to other mechanisms such as jet propulsion which obtain the oxygen from the medium which they travel the rockets carry the oxygen with it. The rocket is accelerating and thus the frame for reference is moving the with the rocket. The velocity of the rocket in the rocket frame of reference U is zero. However, the derivative with respect to time, dU /dt = 0 is not zero. The resistance of the medium is Denote as FR . The momentum equation is

FR 0

τ dA +

c.v. c.v.

g ρ dV +

c.v.

P dA − d dt

ρ a0 dV = ρUy dV +

Vc.v. c.v.

ρUy Urn dA

(6.22)

There are no external forces in this control volume thus, the ﬁrst term FR , vanishes. The pressure term vanish because the pressure essentially is the same and the diﬀerence can be neglected. The gravity term is an instantaneous mass times the gravity times the constant and the same can be said for the acceleration term. Yet, the acceleration is the derivative of the velocity and thus ρ a0 dV = dU (mR + mf ) dt (6.23)

The ﬁrst term on the right hand side is the change of the momentum in the rocket volume. This change is due to the change in the volume of the oxidant and the fuel. d dt ρUy dV =

Vc.v.

d [(mR + mf ) U ] dt

(6.24)

Clearly, the change of the rocket mass can be considered minimal or even neglected. The oxidant and fuel ﬂow outside. However, inside the rocket the change in the velocity is due to change in the reduction of the volume of the oxidant and fuel. This change is minimal and for this analysis, it can be neglected. The last term is ρUy Urn dA = m (Ug − UR ) ˙

c.v.

(6.25)

Combining all the above term results in −FR − (mR + mf ) g + dU (mR + mf ) = m (Ug − UR ) ˙ dt (6.26)

6.2. MOMENTUM EQUATION APPLICATION

157

Denoting MT = mR + mf and thus dM/dt = m and Ue = Ug − UR . As ﬁrst approx˙ imation, for constant fuel consumption (and almost oxidant), gas ﬂow out is constant as well. Thus, for constant constant gas consumption equation (6.26) transformed to −FR − MT g + dU ˙ MT = MT Ue dt (6.27)

Separating the variables equation (6.27) yields dU = ˙ −MT Ue FR − − g dt MT MT (6.28)

Before integrating equation (6.28), it can be noticed that the friction resistance FR , is a function of the several parameters such the duration, the speed (the Reynolds number), material that surface made and the medium it ﬂow in altitude. For simplicity here the part close to Earth (to the atmosphere) is assumed to be small compared to the distance in space. Thus it is assume that FR = 0. Integrating equation (6.28) with limits of U (t = 0) = 0 provides

U 0

˙ dU = −MT Ue

0

t

dt − MT

t

g dt

0

(6.29)

˙ the results of the integration is (notice M = M0 − t M) U = Ue ln M0 ˙ M0 − t M −gt (6.30)

The following is an elaborated example which deals with an unsteady two dimensional problem. This problem demonstrates the used of control volume to ﬁnd method of approximation for not given velocity proﬁles1 Example 6.5:

1 A variation of this problem has appeared in many books in the literature. However, in the past it was not noticed that a slight change in conﬁguration leads to a constant x velocity. This problem was aroused in manufacturing industry. This author was called for consultation and to solve a related problem. For which he noticed this “constant velocity.”

158

**CHAPTER 6. MOMENTUM CONSERVATION
**

UT y x

h

A tank with wheels is ﬁlled with liquid is depicted in Figure 6.7. The tank upper part is opened to the atmosphere. At initial time the valve on the tank is opened and the liquid ﬂows out with an uniform velocity proﬁle. The tank mass with the wheels (the solid parts) is known, mt . Calculate the tank velocity for two cases. One the wheels have a constant resistance with the ground and two the resistance linear function of the weight. linear function of the height. Solution

Uo

FR

Fig. -6.7. Schematic of a tank seating on wheel for unsteady state discussion

Assume that the exit velocity is a

This problem is similar to the rocket mechanics with a twist, the source of the propulsion is the potential energy. Furthermore, the ﬂuid has two velocity components verse one component in the rocket mechanics. The control volume is shown in Figure 6.7. The frame of reference is moving with the tank. This situation is unsteady state thus equation (6.12) for two dimensions is used. The mass conservation equation is d dt ρ dV +

Vc.v. Sc.v.

ρdA = 0

(6.V.a)

Equation (6.V.a) can be transferred to dmc.v. = −ρ U0 A0 = −m0 dt (6.V.b)

Where m0 is mass ﬂow rate out. Equation (6.V.b) can be further reduced due to constant density to d (A h) (6.V.c) + U0 A0 = 0 dt It can be noticed that the area of the tank is almost constant (A = constant) thus A dh dh U0 A0 + U0 A0 = 0 =⇒ =− dt dt A (6.31)

The relationship between the height and the ﬂow now can be used. U0 = B h (6.V.d)

Where B is the coeﬃcient that has the right units to mach equation (6.V.d) that represent the resistance in the system and substitute the energy equation. Substituting equation (6.V.d) into equation (6.V.c) results in dh B h A0 + =0 dt A (6.V.e)

6.2. MOMENTUM EQUATION APPLICATION

159

Equation (6.V.e) is a ﬁrst order diﬀerential equation which can be solved with the initial condition h(t = 0) = h0 . The solution (see for details in the Appendix A.2.1 ) is h(t) = h0 e

−

t A0 B A

(6.V.f)

UT To ﬁnd the average velocity in the L x direction a new control volume is used. y The boundary of this control volume are x the tank boundary on the left with the U straight surface as depicted in Figure 6.8. F The last boundary is variable surface in a distance x from the tank left part. The tank depth, is not relevant. The mass con- Fig. -6.8. A new control volume to ﬁnd the servation for this control volume is velocity in discharge tank for example 6.5.

h x

o R

w &x

dh = −& h Ux w dt

(6.V.g)

Where here w is the depth or width of the tank. Substituting (6.V.f) into (6.V.g) results ¨ t A0¨ B h0 x A0 B − ¨ ¨ A = x A0 B (6.V.h) e Ux (x) = ¨ A Ah ¡ The average x component of the velocity is a linear function of x. Perhaps surprising, it also can be noticed that Ux (x) is a not function of the time. Using this function, the average velocity in the tank is Ux = 1 L

L 0

x A0 B L A0 B = A 2A

(6.V.i)

It can be noticed that Ux is not function of height, h. In fact, it can be shown that average velocity is a function of cross section (what direction?). Using a similar control volume2 , the average velocity in the y direction is Uy = dh h0 A0 B − =− e dt A t A0 B A (6.V.j)

It can be noticed that the velocity in the y is a function of time as oppose to the x direction. The applicable momentum equation (in the tank frame of reference) is (6.11) which is reduced to

acceleration

F −F R − (mt + mf ) g − a (mt + mf ) =

R

d [(mt + mf ) U r ] + U0 mo dt

(6.V.k)

2 The boundaries are the upper (free surface) and tank side with a y distance from the free surface. R Ubn dA = Urn dA =⇒ Ubn = Urn .

160

CHAPTER 6. MOMENTUM CONSERVATION

Where U r is the relative ﬂuid velocity to the tank (if there was no tank movement). mf and mt are the mass of the ﬂuid and the mass of tank respectively. The acceleration of the tank is a = −ˆ 0 or ˆ · a = −a. And the additional force for accelerated system ia i is −ˆ · i a ρdV = mc.v. a

Vc.v.

The mass in the control volume include the mass of the liquid with mass of the solid part (including the wheels). mc.v. = mf + mT because the density of the air is very small the change of the air mass is very small as well (ρa << ρ). The pressure around the control volume is uniform thus P cos θx dA ∼ 0

Sc.v.

**and the resistance due to air is negligible, hence τ dA ∼ 0
**

Sc.v.

**The momentum ﬂow rate out of the tank is ρ Ux Urn dA = ρ Uo 2 Ao = mo Uo
**

Sc.v.

(6.32)

In the x coordinate the momentum equation is −Fx + (mt + mf ) a = d [(mt + mf ) Ux ] + U0 mf ˙ dt (6.V.l)

Where Fx is the x component of the reaction which is opposite to the movement direction. The momentum equation in the y coordinate it is Fy − (mt + mf ) g = d (mt + mf ) Uy dt (6.V.m)

There is no mass ﬂow in the y direction and Uy is component of the velocity in the y direction. The tank movement cause movement of the air which cause momentum change. This momentum is function of the tank volume times the air density times tank velocity (h0 × A × ρa × U ). This eﬀect is known as the add mass/momentum and will be discussed in the Dimensional Analysis and Ideal Flow Chapters. Here this eﬀect is neglected. The main problem of integral analysis approach is that it does not provide a way to analysis the time derivative since the velocity proﬁle is not given inside the control volume. This limitation can be partially overcome by assuming some kind of average. It

n) Noticing that the derivative with time of control volume mass is the ﬂow out in equation (6. Ux Vf Thus. d d Ux Ux + mc.V.p) into a= Fx − m0 mf + mt L A0 B + 2 A U0 (mf + mt ) 2 A (mf + mt ) (6.V.6. if the diﬀerence between the actual and averaged momentum is neglected then ∼0 d dt ρ Ux dV ∼ Vf d mc.2. From equation (6. However. d Ux Ux + mc.v.d) the mass ﬂow out is U0 h m0 (t) = B h0 e − t A0 B A A0 ρ (6.V. if it is the opposite the toy start to move.q) A0 If the Fx ≥ m0 L 2 A B + U0 the toy will not move.v.V. Ux = mc. The velocity in the y direction does not contribute to the momentum in the x direction. The ﬁrst component is downward (y) direction and the second in the exit direction (x).V. The average velocity in the tank (because constant density and more about it later section) is Ux = 1 Vt Ux dV Vf Because the integral is replaced by the average it is transferred to ρ Ux dV ∼ mc.v. dt dt L A0 B Ux = −m0 2A (6.o) Combining all the terms results in −Fx + a (mf + mt ) = −m0 L A0 B − U0 m0 2A (6.n) becomes mass rate out ˙ = − m0 d mc.v.v.p) Rearranging and noticing that a = dUT /dt transformed equation (6.v. MOMENTUM EQUATION APPLICATION 161 can be noticed that the velocity in the tank has two components.V.V.V. dt dt dt (6.r) .

r) and (6. There are situations where actual velocity proﬁle is not known but is function can be approximated. ρ Ux dV (6. For example.v U = ρ V c.1 relationship between momentum of maximum velocity to average velocity was presented.5 can be estimated even by hand sketching. the velocity proﬁle can be estimated using the ideal ﬂuid theory but the actual values are not known. relationship between momentum for the average velocity to the actual velocity is presented.v. End Solution Average Velocity Estimates In example 6.34) The actual momentum for control volume is Mc = c.V.162 The mass in the control volume is CHAPTER 6. the ﬂow proﬁle in example 6. MOMENTUM CONSERVATION V mf = ρ A h0 e − t A0 B A (6.35) .V. For example.V.V.33) The y component of the average velocity is function of the time.s) into equation (6. This correction factor can be calculated by ﬁnding the relation between the two cases. Here. The momentum for average velocity is Ma = mc.V.V.q) transforms it to a diﬀerential equation which is integrated if Rx is constant.s) The initial condition is that UT (t = 0) = 0. Substituting equations (6. For the second case where Rx is a function of the Ry as Rx = µ Ry (6.u) The actual results of the integrations are not provided since the main purpose of this exercise to to learn how to use the integral analysis.t) (mf ) Uy = mf Uy + dt dt dt The reason that mf is used because the solid parts do not have velocity in the y direction. For these cases a correction factor can be used. The change in the accumulative momentum is d dmf dUy (6.v U dV (6. Rearranging the momentum equation in the y direction transformed m f t A0 B − A g + 2 Fy = mt + ρ A h0 e ρ h0 A0 2 B2 A 2 e − t A0 B A (6.

38) Now the left hand side has to be transformed into the control volume as M= d dt r ρ (r × U ) dV + Vc.6.40) . A pump impeller is shown in Figure 6. However.37) The angular momentum then will be obtained by calculating the change of every element in the system as M = r ×F = D Dt ρ r × U dV Vsys (6.v 163 U dV = c. ρ Ux dV (6. To demonstrate this idea.v r ρ (r × U ) U rn dA (6. -6. the following discussion is provided. The impeller increases the velocity of the ﬂuid by increasing the radius of the particles.3.39) The angular momentum equation.v.3 Conservation Moment Of Momentum The angular momentum can be derived in the same manner as the momentum equation for control volume. The pressure on the outer side is uniform thus does not create a moment. Sc.Fig. The inside particle is obtained larger velocity and due to centrifugal forces is moving to outer radius for which ad. the coeﬃcient is not equal to one. CONSERVATION MOMENT OF MOMENTUM These two have to equal thus. C ρV c. applying equation (6. 6. The impeller of the centrifugal pump and the velocities diagram at the exit.39) to uniform and steady state ﬂow with neglected pressure gradient is reduced to M = m (r2 × U2 + r2 × U1 ) ˙ Introduction to Turbo Machinery The analysis of many turbomachinary such as centrifugal pump is fundamentally based on the angular momentum.v. ditionally increase of velocity occur.9 commonly used in industry. if the density is not constant.9. The ﬂow is Um2 Ulr2 U2 Un2 Ut2 (6. The force F = D Dt U ρU dV Vsys (6.36) If the density is constant then the coeﬃcient is one (C ≡ 1).

one the tangential velocity. Example 6.41) M ω = m r2 ω Ut2 ˙ The shaft work is given by the left side and hence. r.6: A centrifugal pump is pumping 600 2[m3 /hour]. The relative exit velocity is Ulr2 and the velocity of the impeller edge is Um2 . and h. M = m r2 Ut2 ˙ Multiplying equation (6.4 More Examples on Momentum Conservation Example 6. Un2 .p. Ut2 and radial component. h is 2[cm] and the exit diameter is 0. Ut2 is not equal to the impeller outer edge velocity Um2 . Notice that tangential liquid velocity. It is assumed that required torque is function U2 . Estimate what is the minimum energy required by the pump.43) The diﬀerence between Um2 to Ut2 is related to the eﬃciency of the pump which will be discussed in the chapter on the turbomachinary. Here it is assumed that ﬂuid is incompressible (ρ = constant). 6.40[m]. The angular velocity is 1200 r. The thickness of the impeller. ˙ W = m Um2 Ut2 ˙ (6.42) (6.41) results in Um2 (6. Assume that angle velocity is leaving the impeller is 125◦ .m.164 CHAPTER 6. The exit liquid velocity. The height of the impeller is h. MOMENTUM CONSERVATION assumed to enter the impeller radially with average velocity U1 . U2 has two components.7: A design of a rocket is based on the idea that density increase of the leaving jet increases the acceleration of the rocket see Figure .

-6.10. Assume that ter rocket for the discussion on the the gas is obeying the perfect gas model. The gas obeys the ideal gas law. hg Gas Select the control volume in such a way that provides the ability to ﬁnd the rocket Liquid acceleration.2. What is the instantaneous vehℓ locity of the rocket at time zero? Develop Uexit the expression for the pressure (assuming no friction with the walls). Once the gas pressure reduced to be equal or below the outside pressure the rocket have no power for propulsion. The mass conservation is similar to the rocket hence it is dm = −Ue Ae dt (6. Liquid ﬁlls the lower part of the rocket tank. The following is not the solution but rather the approach how to treat this problem. In this problem the energy source is the pressure of the gas which propels the rocket. The mixing between the liquid and gas is negligible.7 What are the parameters that eﬀect the problem. hypotherical volume height Solution Under construction for time being only hints3 In the solution of this problem several assumptions must be made so that the integral system can be employed. Nozzle schematics wapression for rocket velocity. The upper part of the rocket tank is ﬁlled with compressed gas. The surface remained straight at the times and no liquid residue remains behind. the initial take oﬀ is requires a larger pressure.6. The gas mass is negligible in comparison to the liquid mass and/or the rocket. The cross section of the liquid is constant.VII. Several people ask to provide a solution or some hints for the solution.4. forces for example 6. No resistance to the rocket (can be added).a) 3 This problem appeared in the previous version (0.3) without a solution. The process is isothermal (can be isentropic process). Additionally. Develop exFig. No gas leaves the rocket. Assume that this idea has a good enUrocket gineering logic. MORE EXAMPLES ON MOMENTUM CONSERVATION 165 6. .10.

VII. The minus sign is to account for change of “direction” of the liquid height.d) mg R T V0g (6.VII.VII.VII.f) It must be point out that integral is not function of time since the height as function of time is known at this stage. Using the above deﬁnition.b) becomes P0 = The relationship between the gas volume Vg = h g A (6.e) =A = −A dt dt dt The last identify in the above equation is based on the idea what ever height concede by the liquid is taken by the gas. Furthermore.h) (6. The total change of the gas volume can be obtained by integration as Vg = A (hg0 − ∆h ) (6.166 CHAPTER 6.i) m (t) = m 0 − 0 Ue Ae dt (6. The change of the gas volume is dVg dhg dh (6.VII.b) Per the assumption the gas mass remain constant and is denoted as mg . The initial pressure now can be expressed as P0 = The pressure at any time is P = Thus the pressure ratio is P hg0 hg0 = = = hg0 P0 hg hg0 − ∆h Equation (6.c) The gas geometry is replaced by a virtual constant cross section which cross section of the liquid (probably the same as the base of the gas phase). MOMENTUM CONSERVATION The mass conservation on the gas zone is a byproduct of the mass conservation of the liquid. it can be observed that the gas pressure is a direct function of the mass ﬂow out.a) can be written as t mg R T hg0 A mg R T hg A 1 ∆h 1− hg0 (6.VII.VII.g) (6. equation (6. The gas pressure at the initial point is P0 = ρ0 R T (6.VII.j) .VII.VII.VII.

5 it was mentioned that there are only two velocity components.l) d −g (mR + m ) − a (mR + m ) = (mR + m ) U +bc + (UR + U ) m dt Where bc is the change of the liquid mass due the boundary movement.m) Example 6.VII. At a speciﬁc moment the valve is opened and the rocket is allowed to ﬂy.1 Qualitative Questions Example 6.9: In Example 6.10: For each following ﬁgures discuss and state force direction and the momentum that act on the control volume due to .VII. What was the assumption that the third velocity component was neglected. Example 6.a) it also can be written that dh Ue Ae = dt ρe A 167 (6.8: A rocket is ﬁlled with only compressed gas.6. Ue = f (P ) + g h rho Where ζ here is a constant which the right units.VII. MORE EXAMPLES ON MOMENTUM CONSERVATION From equation (6.4. End Solution (6. The liquid momentum balance is =0 f (P ) = ζ P (6. Example 6. What is the minimum pressure which make the rocket ﬂy.4.k) According to the assumption the ﬂow out is linear function of the pressure inside thus. 6.VII. What are the parameters that eﬀect the rocket velocity. Develop an expression for the rocket velocity.11: .

consider the unsteady eﬀects.11 is built with a exit located in uneven distance from the the right and the left and is ﬁlled with liquid. Flow out of un symmetrical tank for example 6.168 Situations CHAPTER 6.11. The exit is located on the left hand side at the front. -6.11 . What are the direction of the forces that keep the control volume in the same location? Hints. Look at the directions which the unsteady state momentum in the tank change its value. Fig. MOMENTUM CONSERVATION Explanations Uout F U Uin Flow in and out of Angle β θ Flow in and out at angle from a tank A similar tank as shown in Figure 6.

1) Equation (7. as all phases and materials. obeys this law which creates strange and wonderful phenomena such as a shock and choked ﬂow. However. The ﬂuid.10) for a system is D EU DU Dz ˙ ˙ Q−W = + mU + mg Dt Dt Dt This equation can be rearranged to be D ˙ ˙ Q−W = Dt EU + m U2 + mgz 2 (7. Moreover.CHAPTER 7 Energy Conservation 7. the relationship between height and ﬂow rate was assumed previously. which is external force. Clearly this topic is very important and will be extensively discussed here.2) (7. this law allows to solve problems. For example.1 The First Law of Thermodynamics This chapter focuses on the energy conservation which is the ﬁrst law of thermodynamics1 . It was shown in Chapter 2 that the energy rate equation (2.3) in which the right hand side has to be interpreted and the left hand side interpolated using the Reynold’s Transport Theorem (RTT)2 . during time of the constructing this book only a simple skeleton by Potto standards will be build. the above view is wrong when the heat convection. The right hand side is very complicated and only some of the eﬀects will be discussed (It is only an introductory material). is included on the right hand side. here it will be derived. 1 Thermodynamics is the favorite topic of this author since it was his major in high school. 2 Some view the right hand side as external eﬀects while the left side of the equation represents the internal eﬀects.2) is similar to equation (6. which were assumed in the previous chapters. Additionally a discussion on various energy approximation is presented. 169 . This simplistic representation is correct only under extreme conditions. For example.

The heat derivative is normalized into area direction.5) dw d S S = − (Sn + τ ) · dA = − (Sn + τ ) · U dA dt dt The total work for the system including the shaft work is ˙ W =− Ac. the discussion here will be restricted to convection and conduction. In most problems.170 CHAPTER 7. The ﬁrst kind work is by the friction or the shear System at t + dt stress and the second by normal force.v. conduction. There are three modes of heat transfer. ENERGY CONSERVATION The energy transfer is carried (mostly3 ) by heat transfer to the system or the control volume. The work on the control volume is are divided into two categories: one per. There are two kinds of works that the system does on the surroundings. the surface forces Fig.4) System at t Sn The work done on the system is dℓ more complicated to express than the heat transfer. Hence. 3 There 4 When (7. As in the previous chapter. The total heat transfer to the control volume is ˙ Q= Acv k dT dA dn τ (7.1.3) Where dq is heat transfer to an inﬁnitesimal small area per time and kT is the heat ˙ conduction coeﬃcient. dealing with convection. actual mass transfer must occur and thus no convection is possible to a system by the deﬁnition of system. Conduction for most simple cases is governed by Fourier’s Law which is dq = kT ˙ dT dA dn (7. The main heat transfer mode on the left hand side is conduction. Issues related to radiation are very complicated and considered advance material and hence will be left out. The work done by system on the surroundings (see Figure 7.done by two diﬀerent mechanisms pendicular to the surface and one with the surface direction.1) is F dF dV S A S dw = −S dA ·d = − (Sn + τ ) · d dA The change of the work for an inﬁnitesimal time (excluding the shaft work) is U (7.7) are other methods such as magnetic ﬁelds (like microwave) which are not part of this book.6) S (Sn + τ ) U dA − Wshaf t (7. convection4 and radiation. The issues of convection are mostly covered by the terms on the left hand side. -7. . the radiation is minimal.

8) for more details).9) the velocity appears twice.8) Equation (7.1.10) The ﬁrst term on the right hand side is referred to in the literature as the ﬂow work and is Urn P n · U dA = ˆ S S P (U − Ub ) n dA + ˆ S P Ubn dA (7. As it was discussed in the previous chapter the normal stress component is replaced by the pressure (see equation (6.2) for system is kT Asys 171 dT dA+ dn S (Sn + τ ) dV D ˙ +Wshaf t = Dt Asys ρ Vsys EU + m U2 + g z dV 2 (7.8) and thus yields kT Acv dT dA+ dn Energy Equation ˙ S (Sn + τ ) dA + Wshaf t = d dt + Acv Acv (7.9) ρ Vcv Eu + m Eu + m U2 +gz 2 U2 + g z dV 2 ρ Urn dA From now on the notation of the control volume and system will be dropped since all equations which deals with the control volume.11) can be further manipulated to become work due to the ﬂow work due to boundaries movement P n · U dA = ˆ S S P ρ Urn dA + ρ P Ubn dA S (7.11) Equation (7. THE FIRST LAW OF THERMODYNAMICS The energy equation (7.7. In the last term in equation (7. The system can contain solid parts as well several diﬀerent kinds of ﬂuids.12) .8) does not apply any restrictions on the system. The work rate (excluding the shaft work) is ﬂow work ˙ = W ∼ S P n · U dA − ˆ S τ · U n dA ˆ (7. Now Reynolds Transport Theorem can be used to transformed the left hand side of equation (7. Note that U is the velocity in the frame of reference while Urn is the velocity relative to the boundary.

2.15) describes the energy conservation for the control volume in stationary coordinates.14) The new term P/ρ combined with the internal energy. surface tension eﬀects are negligible and the liquid surface is straight6 . The enhℓ Ae ergy equation with mass conservation will Ue be utilized for this analysis. assumption is appropriated only under certain conditions which include the geometry of the tank or container and the liquid properties. and exit area is relatively with a small diameter.15) Equation (7. small.13) Substituting all these terms into the governing equation yields d U2 ˙ ˙ ˙ Q − Wshear − Wshaf t = Eu + + g z dV + dt V 2 2 P U Eu + + + g z Urn ρ dA + P Urn dA ρ 2 S S (7. -7. so the velocity can be assumed uniform (not a function of the opening height)5 . Flow Out From A Container In the previous chapters of this book. In this analysis several assumptions are made which includes the following: constant density. which was discussed on page 36. With these deﬁnitions equation (7. The example of ﬂow from a tank or container is presented to demonstrate how to treat some of terms in equation (7. ENERGY CONSERVATION The second term is referred to as the shear work and is deﬁned as ˙ Wshear = − S τ · U dA (7. Additionally. the gas density is very small compared to Fig.172 CHAPTER 7. a discussion about the height opening eﬀects will be discussed. A discussion about this issue will be presented in the Dimensional Chapter and is out of the scope of this chapter. the ﬂow rate out of a tank or container was assumed to be a linear function of A the height. the temperature is assumed to constant.15).14) transformed ˙ ˙ Q − Wshear + Simpliﬁed Energy Equation d U2 ˙ Wshaf t = Eu + + g z dV + dt V 2 2 U + g z Urn ρ dA + P Urn dA h+ 2 S S (7. Discharge from a Large Container liquid density. Eu is referred to as the enthalpy. Also note that the straight surface assumption is not the same surface tension eﬀects zero. h. 6 This 5 Later . The ﬂow out is related to the height but in a more complicate function and is the focus of this discussion.

Second reason for this exercise the surface velocity has only one component is to avoid dealing with Bar-Meir’s instability.17) Equation (7. 7 For the mass conservation analysis. However.16) which also can be written (because = 0) as Urn dA = 0 A Ubn dA + A (7. Since in this case the geometry is assumed to be symmetrical one side Fig.1. However. the velocity is zero for symmetrical geometry and some other geometries. . for simplicity.7.20) In this analysis. How to compensate and estimate the kinetic energy when averaged Velocity is zero. in this discussion it is assumed that surface has only one component in z direction.3. The kinetic energy of the tank or container is based on the half part Uy = 0 as shown in Figure 7. Hence it requires that velocity proﬁle in x y to be parabolic. Similar estimate that was done for x direction can be done Ue to every side of the opening if they are not symmetrical. However.17) provides the relationship between boundary velocity to the exit velocity as A Ub = Ae Ue (7. THE FIRST LAW OF THERMODYNAMICS 173 The control volume is chosen so that all the liquid is included up to exit of the pipe. is suﬃcient as (π − 2)r dh Uy ∼ = 8h dt (7. for the energy analysis the averaged velocity cannot be considered zero.21) Also notice that the surface velocity is not zero.19) The x component of the averaged velocity is a function of the geometry and was calculated in Example 5. The conservation of the mass is d dt V ρ dV + ¡ dρ dt A ρ Urn dA = 0 ¡ (7. The averaged velocity in the y direction is zero because the ﬂow is symmetrical7 .18) Note that the boundary velocity is not the averaged velocity but the actual velocity. The surface has three velocity components which non have them vanish. this quantity will be used.3. the change of the kinetic Uy 1 Uy 1 energy due to the change in the velocity 2 2 ﬁeld isn’t zero. The averaged velocity in z direction is same as the boundary velocity Ub = Uz = dh Ae = Ue dt A (7.12 to be larger than Ux 2 r Ae 2r 2 r dh Ue =⇒ Ux ∼ Ub = = h A h h dt (7. -7.

G. Pa is the external pressure and Pe is the exit pressure10 . Additionally.27) V A 8 This approach is a common approximation. Note that the exit velocity on the upper surface is zero Urn = 0. The dissipation can be neglected for small Ec number. .24) is reduced to d dt U2 + g z ρ dV − 2 Ue 2 2 Ue ρ dA = 0 (7. the dissipation creates a loss that has temperature component. this change is a function of Eckert number. the internal shear work is assumed negligible. the shear stresses at the exit are normal to the ﬂow direction hence the shear work is vanished. The pressure terms in equation (7.24) Where Ub is the upper boundary velocity. Ec number is named after this author’s adviser. why this approach is correct in most cases is not explained here. Combining all these information results in energy ﬂow out internal energy change energy in and out upper surface work d dt V U + g z ρ dV + 2 2 A Pe Ue + ρ 2 2 Ue ρ dA − A Pa Ub dA = 0 (7.23) Now the energy equation deals with no “external” eﬀects.24) are Pe Ue ρdA − ρ Pa Ub dA = Pe A A Ue dA − Pa A Ub dA (7. The temperature is constant8 . Clearly. E.25) A It can be noticed that Pa = Pe hence =0 Pa A Ue dA − A Ub dA =0 (7. Furthermore.26) The governing equation (7.15) which is applicable to this case.22) The boundary shear work is zero because the velocity at tank boundary or walls is zero. 10 It is assumed that the pressure in exit across section is uniform and equal surroundings pressure. ˙ ˙ Wshear = Wshaf t = 0 (7. In this case.R. ENERGY CONSERVATION The energy balance can be expressed by equation (7. Yet. In this light. Ec which is very small. A discussion about this eﬀect will be presented in the dimensional analysis chapter. the following approximation can be written Eu ˙ = hin − hout = 0 Q= dt (7. Eckert. 9 It is only the same assumption discussed earlier.174 CHAPTER 7. At the free surface the velocity has only normal component9 and thus shear work vanishes there as well.

Similarly to the previous chapter which the integral momentum will be replaced by some kind of average.30) Substituting the results of equation (7. A discussion on the correction factor is presented to provide a better “averaged” velocity. The inside integral can be evaluated as h zdz = 0 h2 2 (7.29) yields d gρ dt V h2 d h dh dA = g ρ hA = g ρAh 2 dt 2 dt (7. Even the averaged velocity is zero the kinetic energy is not zero and another method should be used.s) of equation (7.h.30) into equation (7.28) V V The second integral (in the r. A comparison between the actual kinetic energy and the kinetic energy due to the “averaged” velocity (to be called the averaged kinetic energy) provides a correction coeﬃcient.1.33) . The averaged velocity is Uave = 1 V U dV V (7.31) A The kinetic energy related to the averaged velocity with a correction factor which depends on the geometry and the velocity proﬁle.32) The total kinetic energy for the averaged velocity is ρ Uave 2 V = ρ 1 V 2 2 U dV V V =ρ V U dV (7. The terms under the time derivative can be divided into two terms as d dt d U2 + g z ρdV = 2 dt U2 d dV + 2 dt g z ρ dV V (7.29) V Where h is the height or the distance from the surface to exit. The ﬁrst integral can be estimated by examining the velocity proﬁle eﬀects. THE FIRST LAW OF THERMODYNAMICS 175 The minus sign is because the ﬂow is out of the control volume.7.28) is d dt d g z ρ dV = g ρ dt h A 0 dV z dz dA (7. Furthermore.

12 The advantage is described in the Dimensional Analysis Chapter.34) Here. The relationship between Umax to the averaged velocity. Uave is obtained by using equation (7. More about this issue in the Diﬀerential Analysis Chapter. the velocity is U r R = U (¯) = Umax 1 − r2 = 2 Uave 1 − r2 r ¯ ¯ (7.36) It can be noticed that the velocity is presented as a function of the reduced radius12 .35) V V This integral can be evaluated for any given velocity proﬁle.36) into equation (7. In this case. For circular shape.35) results Uave 2 V U 2 dV V V = Uave 2 V 2 Uave 1 − r2 ¯ 2 = dV 3 Uave 2 V = 4 4 Uave 2 π L R2 3 (7. The ﬂow is parabolic only when the ﬂow is driven by pressure or gravity. For a pipe geometry. Calculations according to this point can improve the accurately based on the above discussion. ENERGY CONSERVATION The general correction factor is the ratio of the above value to the actual kinetic energy as 2 ρ U dV CF = V ρ U 2 dV V ρ (Uave ) V = ¡ ρ U 2 dV ¡ V 2 (7. Substituting equation (7. a good guess number is about 1.32) which yields 1/2. it is assumed that the averaged velocity indeed represent the energy in the tank or container.176 CHAPTER 7.37) The correction factor for many other velocity proﬁles and other geometries can be smaller or larger than this value. for simplicity reason. Note.1. diﬀerent directions cancel each other out. A large family of velocity proﬁles is laminar or parabolic (for one directional ﬂow)11 . . The correction factor for a constant density ﬂuid is 2 2 ρ U dV CF = V = ρ U 2 dV ρ ¡ ρ ¡ U dV V = U 2 dV Uave 2 V U 2 dV V (7. 11 Laminar ﬂow is not necessarily implies that the ﬂow velocity proﬁle is parabolic. The diﬀerence between the “averaged momentum” velocity and the “averaged kinetic” velocity is also due to the fact that energy is added for diﬀerent directions while in the momentum case. the inequality sign because the density distribution for compressible ﬂuid. CF is the correction coeﬃcient.

However. 14 The solution not the derivation is about one page.38) The relationship between the boundary velocity to the height (by deﬁnition) is Ub = dh dt (7. But in the same vain. THE FIRST LAW OF THERMODYNAMICS The unsteady state term then obtains the form 2 U2 d U d gh ∼ρ ρ + g y dV = + dt V 2 dt 2 2 177 V hA (7. .41) Combining all the three components of the velocity (Pythagorean Theorem) as 2 2 2 2 U ∼ Ux + Uy + Uz = (7. It must be remembered that is eﬀect extremely important in the later stages of the emptying of the tank. The reader is encouraged to do it to get acquainted with this concept. some other eﬀects have to be taken into account which were neglected in construction of this model such as upper surface shape.44) It can be noticed that f (G) is a weak function of the height inverse.7. However.43) f (G) dh U∼ = dt (π − 2) r 8h 2 + (π − 1) r 4h 2 + 12 (7. 13 A similar point was provided in mass conservation Chapter 5. it easy can be proved by construction the same control volume. the mathematical complication are enormous14 and this eﬀect is assumed neglected and the function to be constant. Analytical solution of the governing equation is possible including this eﬀect of the height.42) 2 U ∼ = (π − 2) r dh 8h dt 2 + (π − 1) r dh 4h dt 2 + dh dt 2 (7.40) Ue = A dh dh = −Ub Ae dt dt (7.1. the velocity in the z direction13 is Uz = dh dt (7.39) Therefore.

27) results in V 2 2 2 d U gh 1 dh A ρ + hA − Ue ρ Ae = 0 ¡ dt ¡ 2 2 2 dt Ae taking the derivative of ﬁrst term on l.178 The last term is CHAPTER 7.50) f (G)2 h 15 This d2 h g h f (G)2 + + dt2 2 2 dh dt 2 + gh 1 − 2 2 dh dt 2 A Ae 2 =0 (7.h. results in d U gh + dt 2 2 2 (7. Advance material can be skipped Dividing equation (7. ENERGY CONSERVATION A Ue 2 Ue 2 Ue ρ dA = Ue ρ Ae = 2 2 dh A dt Ae 2 Ue ρ Ae (7.47) can be rearranged and simpliﬁed and combined with mass conservation 15 .51) part can be skipped to end of ”advanced material”.46) hA + U gh + 2 2 2 A dh 1 − dt 2 dh dt 2 A Ae 2 Ue Ae = 0 (7.45) Combining all the terms into equation (7.47) Equation (7.s.40) A Ae A Ue d U gh + dt 2 2 2 hA U gh + + Ue Ae 2 2 2 dh 1 A − dt 2 dh dt 2 A Ae 2 $ Ue A $$ e = 0 (7. .46) by Ue Ae and utilizing equation (7.49) Further rearranging to eliminate the “ﬂow rate” transforms to 1 U dh 1 A B ¨ gh dU Ub A¨ f (G)2 dt ¨ + f (G) h + ¨ dt ¨Ue Ae 2 e Ae U 2 dh dt 2 + gh 1 − 2 2 dh dt 2 A Ae 2 =0 (7.48) Notice that U = Ub f (G) and thus f (G) Ub U dU h A g dh h A U gh 1 + + + − dt Ue Ae 2 dt Ue Ae 2 2 2 2 dh dt 2 A Ae 2 =0 (7.

1.52) by f (G)2 and using this parameter. as Te = A f (G) Ae 2 (7.54) The solution can either of these equations16 dh − = t + k2 (k1 Te − 2 k1 ) eln(h) T e + 2 g h2 h (T e − 2) f (G) or - (7.56) The solution with the positive solution has no physical meaning because the height cannot increase with time.7.52) after minor rearrangement transformed to h d2 h g Ae 2 + dt2 Te A2 + 1 2 dh dt 2 [1 − Te ] = 0 (7. Dividing equation (7.57) (k1 Te − 2 k1 ) eln(h) T e + 2 g h2 h (T e − 2) f (G) 16 A discussion about this equation appear in the mathematical appendix.53) This parameter represents the characteristics of the tank which controls the emptying process. Te . . equation (7.52) Deﬁning a new tank emptying parameter. Thus deﬁne function of the height as dh f (h) = − (7. THE FIRST LAW OF THERMODYNAMICS 179 End Advance material Combining the gh terms into one yields f (G)2 h d2 h 1 +gh+ dt2 2 dh dt 2 f (G)2 − A Ae 2 =0 (7.55) dh (k1 Te − 2 k1 ) eln(h) T e + 2 g h2 h (T e − 2) f (G) = t + k2 (7.

62) [1 − Te ] = 0 (7. if this phenomenon is ignored this solution is correct.58) This condition pose a physical limitation17 which will be ignored.61) (7. ENERGY CONSERVATION The initial condition for this case are: one the height initial is h(0) = h0 The initial boundary velocity is dh =0 dt (7.54) into h g Ae 2 Te A2 + 1 2 dh dt 2 (7. The ﬁrst condition yields k2 = −f (h0 ) The second condition provides dh =0= dt (k1 Te − 2 k1 ) eln(h0 ) T e + 2 g h0 2 h0 (T e − 2) f (G) (7.180 CHAPTER 7.63) While equation (7. The initial velocity 17 For the initial condition speed of sound has to be taken into account. However.60) The complication of the above solution suggest a simpliﬁcation in which d2 h g Ae 2 << dt2 Te A2 which reduces equation (7. the information about opening of the valve did not reached to the surface.64) It can be noticed that Te “disappeared” from the equation. Thus for a very short time.59) (7. And taking the “positive” branch √ dh 2gh = (7. . the non linear element can be removed by taking negative branch (height reduction) of the equation as dh dt 2 = 2gh −1 + A Ae 2 (7.63) is still non linear equation. This information travel in characteristic sound speed which is over 1000 m/sec. This initial condition is the initial height of the liquid.65) 2 dt A 1 − Ae The nature of ﬁrst order Ordinary Diﬀerential Equation that they allow only one initial condition.

7.68) 1− 1− U∼ = 2gh (7.2 Limitation of Integral Approach Some of accuracy issues to enhance the quality and improvements of the integral method were suggested in the analysis of the emptying tank. LIMITATION OF INTEGRAL APPROACH 181 ﬁeld was eliminated by the approximation (remove the acceleration term). Equation (7.65) integrated to yield 1− A Ae 2 h h0 dh √ = 2gh t dt 0 (7.69) Equation (7. . He was student (not formal) and follower of Galileo Galilei. 1608 October 25. Today the exact reference to his work is lost only “sketches” of his lecture elude work.66) The initial condition has been inserted into the integral which its solution is 1− A Ae √ 2 h − h0 √ =t 2gh A = 2 Ae √ 2gh Ae 2 A (7.2. It is correct only for large ratio of h/r and the error became very substantial for small value of h/r. There are problems that the integral methods even with these enhancements simply cannot tackle. this analysis demonstrates the usefulness of the integral analysis to provide a reasonable solution. He derived this equation based on similar principle to Bernoulli equation (which later leads to Bernoulli’s equation). It seems that Torricelli was an honest man who gave to others and he died at young age of 39 while in his prime. 1647) was an Italian physicist and mathematician.67) dh A Ue = = dt Ae If the area ratio Ae /A << 1 then 2gh A Ae (7.70) 7. The improvements to the integral methods are the corrections to the estimates of the energy or other quantities in the conservation equations.69) is referred in the literature as Torricelli’s equation18 This analysis has several drawbacks which limits the accuracy of the calculations. This analysis can be improved by experimental investigating the phenomenon. Yet. The experimental coeﬃcient can be added to account for the dissipation and other eﬀects such dh ∼ =C dt 2gh (7. In the calculations 18 Evangelista Torricelli (October 15. Thus it is assumed that the initial velocity is not relevant at the core of the process at hand.

the integral does not provide a sufH equilibrioum level ﬁcient tool to analyze the issue at hand. The second type of corrections is the understanding that averaged of the total ﬁeld is diﬀerent from the averaged of diﬀerent zooms. too crude to handle issues of free interface. It turn out that these simpliﬁcations can provide reasonable results and key understanding of the physical phenomena and yet with less work.3 Approximation of Energy Equation The emptying the tank problem was complicated even with all the simpliﬁcations that were carried. In fact. Hence diﬀerent techniques are required. 7. the problems can be solved. Additionally. In the analysis of the tank it was assumed that the dissipation can be igD nored. The ﬁrst type is the prediction of the velocities proﬁle (or the concentration proﬁle). In cases that dissipation play major air air role. . The description H(t) as a function of time requires exact knowledge of the velocity ﬁeld. A liquid in manometer is disturbed from a rest by a distance of H0 . The integral method is a powerful and yet simple method but has has to be used with the limitations of the method in mind. the integral methods is Fig. Flow in an oscillating manometer. -7. The following sections provides further explanation. the averaged energy in the x direction contributes or eﬀects the energy equation. H For example. These problem were minor for the emptying the tank but for the oscillating manometer it is the core of the problem. the knowledge of whether the ﬂow is laminar or turbulent (later on this issue) has to come from diﬀerent techniques. The accuracy issues that integral methods intrinsically suﬀers from no ability to exact ﬂow ﬁeld and thus lost the accuracy as was discussed in the example. two such corrections were presented. In the case of the tank. the analysis of the oscillating manometer cannot be carried by the intelowest level for the liquid gral methods. Furthermore. The discussion on the limitations was not provided to discard usage of this method but rather to provide a guidance of use with caution. the averaged velocity in x direction is zero yet the averaged velocity in the two zooms (two halves) is not zero. ENERGY CONSERVATION of the exit velocity of a tank. The integral method does not handle the problems such as the free surface with reasonable accuracy.182 CHAPTER 7. Hence the prediction can skew the actual predictions. Engineers in order to reduce the work further simplify the energy equation.4.

For a stationary ﬁx control volume the energy equation.21) which can be written as dqrev = T ds = dEu + P dv (7. Fix m & uniform ˙ q − wshear − wshaf t = ˙ ˙ ˙ h+ U2 +gz 2 − out h+ U2 +gz 2 (7. APPROXIMATION OF ENERGY EQUATION 183 7.74) in 7.7.71) is reduced to Steady State Equation & uniform ˙ ˙ ˙ Q − Wshear − Wshaf t = h+ h+ U2 + g z Urn ρAout − 2 (7.72) for non-deformable control volume does not vanished.73) Dividing equation the mass ﬂow rate provides Steady State Equation. The reason is that while the velocity is constant. Equation (2.1 Energy Equation in Steady State The steady state situation provides several ways to reduce the complexity.75) .3.3.72) U2 + g z Urn ρAin + P Ubn Aout − P Ubn Ain 2 It can be noticed that last term in equation (7. Hence the energy equation is reduced to Steady State Equation ˙ ˙ ˙ Q − Wshear − Wshaf t = S h+ U2 + g z Urn ρ dA + 2 P Ubn dA S (7. This imaginary ﬂuid reduces the amount of work in the calculations and Ideal Flow Chapter is dedicated in this book.2 Energy Equation in Frictionless Flow and Steady State In cases where the ﬂow can be estimated without friction or where a quick solution is needed the friction and other losses are illuminated from the calculations.71) If the ﬂow is uniform or can be estimated as uniform. equation (7. The time derivative term can be eliminated since the time derivative is zero.3. The acceleration term must be eliminated for the obvious reason. under this simpliﬁcation transformed to ˙ ˙ ˙ Q − Wshear − Wshaf t = h+ U2 + g z Urn ρAout − 2 U2 + g z Urn ρAin h+ 2 (7. The second low is the core of “no losses” and can be employed when calculations of this sort information is needed. the pressure is different.

81) As before equation (7.80) Using the Reynolds Transport Theorem to transport equation to control volume results in d ˙ Qrev = dt h ρ dV + V A h Urn ρ dA + D Dt V dP ρ ρ dV (7.79) becomes h D ˙ Qrev = Dt Eu + V P ρ ρ dV − D Dt V dP ρ ρ dV (7.78) Integration over the entire system results in h Qrev = V Eu + P ρ ρ dV − V dP ρ ρ dV (7. ENERGY CONSERVATION Using the multiplication rule change equation (7.83) in Subtracting equation (7.79) Taking time derivative of the equation (7.74) results in change in pressure energy change in kinetic energy change in potential energy 0 = wshaf t + dP ρ − 2 dP ρ + 1 U2 2 − U1 2 + g (z2 − z1 ) 2 (7.77) P ρ − v dP (7.82) in (7.76) qrev = Eu + P ρ − dP ρ (7.80) can be simpliﬁed for uniform ﬂow as ˙ Qrev = m (hout − hin ) − ˙ or qrev = (hout − hin ) − ˙ dP ρ − out dP ρ − out dP ρ dP ρ (7.75) dqrev = dEu + d (P v) − v dP = dEu + d integrating equation (7.84) .184 CHAPTER 7.76) yields dqrev = dEu + d P ρ − v dP (7.83) from equation (7.

General Acceleration can be broken into a linear acceleration and a rotating acceleration. linear and rotational. The gravity force for ﬂuid element in small distance then is g dz dm. In many cases.89) The reference was set to inﬁnity.4. In building the gravity potential it was assumed that the gravity is a conservative force. = − ref F ·d (7. it was assumed that the control volume is at rest.86) 7. The accelerations are referring to two kinds of acceleration.88) r2 Where G is the gravity coeﬃcient and M is the mass of the Earth. 7. r and m are the distance and mass respectively.90) The total work or potential is the integral over the whole mass. The gravity potential is then F =− r P Egravity = − ∞ − GM m dr r2 (7.4 Energy Equation in Accelerated System In the discussion so far.84) for constant density is 0 = wshaf t + P2 − P1 U2 2 − U1 2 + + g (z2 − z1 ) ρ 2 185 (7.7. ENERGY EQUATION IN ACCELERATED SYSTEM Equation (7.E.87) In Chapter 3 a discussion about gravitational energy potential was presented. It was pointed earlier in this book that accelerated forces can be translated to potential force.85) reduced to 0= U2 2 − U1 2 P2 − P1 + + g (z2 − z1 ) ρ 2 (7. For example.85) For no shaft work equation (7. the control volume is moving in accelerated coordinates. The only acceptation to the above statement.4. However. The work this element moving from point 1 to point 2 is 2 g dz dm = g (z2 − z1 ) dm 1 (7. . These accelerations will be translated to potential energy. the mathematical treatment is somewhat diﬀerent which is the reason for the separation. There is no conceptional diﬀerence between these two accelerations. for the gravity force is GM m (7.1 Energy in Linear Acceleration Coordinate 2 The potential is deﬁned as P. is the gravity that was compensated by the gravity potential.

4. ENERGY CONSERVATION 7. Using this trick the notion of the ax (x1 − x0 ) can be replaced by ax x.95) The Reynolds Transport Theorem is used to transferred the calculations to control volume as Energy Equation in Linear Accelerated Coordinate d ˙ ˙ Q−W = dt + cv Eu + cv U2 + ax x + ay y + (az + g)z ρ dV 2 (7.94) Equation can be added to the energy equation as D ˙ ˙ Q−W = Dt Eu + sys U2 + ax x + ay y + (az + g)z ρ dV 2 (7. The “potential” of moving the mass in the ﬁeld provides the energy. y = 0. The linear acceleration “creates” a conservative force of constant force and direction. the element of the potential is d P Ea = a · d dm The total potential for element material (1) (7.93) The change of the potential with time is D D P Ea total = Dt Dt (ax x + ay y + az z) dm sys (7. The potential of unit material is P Ea total = (ax x + ay y + az z) ρ dV sys (7.186 CHAPTER 7.2 Linear Accelerated System The acceleration can be employed in similar fashion as the gravity force. The same can be done for the other two coordinates.96) h+ U2 + ax x + ay y + (az + g)z Urn ρ dA 2 + cv P Ubn dA 7. and z = 0. Thus.91) P Ea = (0) a · d dm = (ax (x1 − x0 ) ay (y1 − y0 ) az (z1 − z0 )) dm (7.92) At the origin (of the coordinates) x = 0.4. There are two kinds of acceleration due this rotation one is the .3 Energy Equation in Rotating Coordinate System The coordinate system rotating around ﬁx axis creates similar conservative potential as the linear system. The Force due to the acceleration of the ﬁeld can be broken into three coordinates.

The ˆ ˆ potential is then ˆ ˆ P E = ω 2 r r + 2 U × ω · drˆ + r dθ θ + dz k dm ˆ r (7. Hence.97) (7. However.103) P Ubn dA . θ.4.7. The second part is (2 U × ω) · d dm (7. the most important direction is the direction of the velocity. the ﬂux of this property is important only in the direction of the velocity.101) This multiplication does not vanish with the exception of the direction of U . and k are units vector in the coordinates r. θ and z respectively. The cross product is zero of U ×ω×U =U ×ω×ω = 0 because the ﬁrst multiplication is perpendicular to the last multiplication. ENERGY EQUATION IN ACCELERATED SYSTEM 187 centrifugal and second the Coriolis force. consider a particle which moves with the our rotating system. From physical point of view. To understand it better.99) ˆ where r. The net change of the potential energy due the centrifugal motion is 2 P Ecentrif ugal = − 1 ω 2 r2 dr dm = ω 2 r1 2 − r2 2 dm 2 (7. The forces acting on particles are centrifugal Coriolis F = ω 2 r r + 2 U × ω dm ˆ The work or the potential then is P E = ω 2 r r + 2 U × ω · d dm ˆ The cylindrical coordinate are ˆ ˆ d = drˆ + r dθ θ + dz k r (7.100) The ﬁrst term results in ω 2 r2 (see for explanation in the appendix 271 for vector explanation).98) (7.102) Inserting the potential energy due to the centrifugal forces into the energy equation yields Energy Equation in Accelerated Coordinate d ˙ ˙ Q−W = dt + cv U2 ω 2 r2 + ax x + ay y + (az + g)z − ρ dV 2 2 cv ω2 r2 U2 + ax x + ay y + (az + g) − z Urn ρ dA h+ 2 2 Eu + + cv (7. this term canceled and does not contribute to the potential. This multiplication creates lines (surfaces ) of constant values.

103) is to assume uniform ﬂow. ENERGY CONSERVATION 7. .188 CHAPTER 7.4 Energy Equation in Accelerated Coordinate with Uniform Flow One of the way to simplify the general equation (7.4.

Part II Diﬀerential Analysis 189 .

.

the solution is not unique. Barr´ de e Saint Venant (1843) and George Gabriel Stokes (1845) derived these equation based on the relationship between stress and rate–of–strain (this approach is presented in this book). As usual Simon-Denis Poisson independently. One of the approaches is to reduce the equations by eliminating the viscosity eﬀects. with a presentation of the “non–regular” solutions will be presented with the associated issues of stability.g.CHAPTER 8 Diﬀerential Analysis 8. derived these equations in 1831 for the same arguments as Navier. Like many equations they were independently derived by several people. A discussion about the “regular” solution is present and a brief discussion about limitations when the solution is applicable. Later in the Chapters on Real Fluid and Turbulence. Navier-Stokes equations are non–linear and there are more than one possible solution in many cases (if not most cases) e. The concepts of Add Mass and 1 Which can be view as complementary analysis to the integral analysis. which leads to a diﬀerent approach of differential analysis. The diﬀerential analysis allows the investigation of the ﬂow ﬁeld in greater detail. In diﬀerential analysis. 191 . First these equations were derived by Claude– Louis–Marie Navier as it is known in 1827. This analysis leads to partial diﬀerential equations which are referred to as the Navier-Stokes equations. the emphasis is on inﬁnitesimal scale and thus the analysis provides better accuracy1 . These equations are named after Claude–Louis Navier–Marie and George Gabriel Stokes. However even for the “regular” solution the mathematics is very complex.1 Introduction The integral analysis has limited accuracy. as he done to many other equations or conditions. The equations without the viscosity eﬀects are referred to as the ideal ﬂow equations (Euler Equations) which will be discussed in the next chapter. The foundations for their arguments or motivations are based on a molecular view of how stresses are exerted between ﬂuid layers.

the complexity of the equations. DIFFERENTIAL ANALYSIS Add Force.1. At a speciﬁc time this control volume can be viewed as a system. The connection between these two ideas or ﬁelds was done via introduction of the boundary layer theory by Prandtl which will be discussed in a separate chapter. 8. The mass balance on the inﬁnitesimal control volume. The mass conservation for this inﬁnitesimal small system is zero thus D Dt ρdV = 0 (8. leads to approximations and consequently to ideal ﬂow approximation (equations) and on the other hand experimental solutions of Navier–Stokes equations. -8. there are cases when the complying with the boundary conditions leads to a discontinuity (shock or choked ﬂow).2 Mass Conservation Fluid ﬂows into and from a three dimensional inﬁnitesimal control volume depicted in Figure 8.2) For a constant control volume the derivative can enter into the integral (see also for the divergence theorem in the appendix A. The choked ﬂow is associated with single phase ﬂow (even the double choked ﬂow) while the Interfacial instability associated with the Multi–Phase ﬂow. Sometimes. Historically.2) and hence dρ dt dV dρ dV + dt Urn ρ dA = 0 (8. It has to be pointed out that the Add Mass and Add Force appear regardless to the viscosity. (8.1. the following can be written D Dt ρdV = d dt ρdV + Urn ρ dA = 0 Fig. These issues are discussed in Open Channel Flow and Compressible Flow chapters. the boundary conditions create instability which alters the boundary conditions itself which is known as Interfacial instability. and will be presented in the Ideal Flow chapter.1. These equations cannot satisfy the boundary conditions in other cases and in way the ﬂuid pushes the boundary condition(s) further downstream (choked ﬂow). on one hand. This phenomenon is presented in Multi–phase chapter and in this chapter. Even for simple situations.192 CHAPTER 8.1) x A ρ Ux dy dz ρ+ dρ dz Uz + dUz dz dx dy E F dρ y ρ+ d U dU + dy y y dx dz B ρ+ dρ dx Ux + dUx dx dy dz G dx dz H ρU y C ρ Uz dx dy D However for a control volume using Reynolds Transport Theorem (RTT). which are easier to discuss when the viscosity is ignored.3) .

The net mass change. It also can be noticed that. in the control volume is ∂ρ dr dz r dθ dm = ˙ ∂t 2 Note (8. MASS CONSERVATION 193 The ﬁrst term in equation (8. produces additional dx thus a inﬁnitesimal volume element dV is obtained for all directions.7) Combining the ﬁrst term with the second term results in the continuity equation in Cartesian coordinates as Continuity in Cartesian Coordinates ∂ρ ∂ρ Ux ∂ρ Uy ∂ρ Uz + + + =0 ∂t ∂x ∂y ∂z Cylindrical Coordinates The same equation can be derived in cylindrical coordinates. .4) The second term in the LHS of equation (8. The combination can be divided by dx dy dz and simpliﬁed by using the deﬁnition of the partial derivative in the regular process to be Urn ρ dA = − ∂(ρ Ux ) ∂(ρ Uy ) ∂(ρ Uz ) + + ∂x ∂y ∂z (8.6) The same can be said for the y and z coordinates.2. the operation.3) for the inﬁnitesimal volume is expressed.2. in the x coordinate. as ∼0 dρ dρ dV = dx dy dz + f dt dt dV d ρ dt2 2 + ··· (8.2) is expressed2 as dAyz Urn ρ dA = dy dz (ρ Ux )|x − (ρ Ux )|x+dx + dAxz dAxz (8. as depicted in Figure 8.9) that some time the notation dAyz also refers to dAx . for example.8.8) dv (8.5) dx dz (ρ Uy )|y − (ρ Uy )|y+dy + dx dy (ρ Uz )|z − (ρ Uz )|z+dz The diﬀerence between point x and x + dx can be obtained by developing Taylor series as (ρ Ux )|x+dx = (ρ Ux )|x + ∂ (ρ Ux ) ∂x dx x (8. neglecting higher order derivatives.

The mass ﬂow at r + dr is ρ Ur r dθ dz|r + d/dr (ρ Ur r dθ dz) dr + · · · .11)–(8.11) Note3 that the r is still inside the derivative since it is a function of r. Expansion to Taylor serious ρ U r dθ dz| r r r+dr is obtained by the regular procedure.g.14) 3 The mass ﬂow is ρ U r dθ dz at r point.12) Combining equations (8. dr r dθ dz results in total net ﬂux =− 1 ∂ (ρ Ur r) ∂ρ Uz r ∂ρ Uθ + + r ∂r ∂z ∂θ (8. the r is “trapped” in the derivative.194 ρ U r z CHAPTER 8.2. the change of r with r. DIFFERENTIAL ANALYSIS + ∂ (ρ Ur r) dz dθ dr ∂z ρ U θ ∂ (ρ Uθ ) + dθ dr dz ∂θ dz ρ Uz r) d r r ρU ∂ ( ∂z r+ dθ dz ρU r rd θd z rd z θ dr ρ Uθ dr dθ y θ x ρ Uz r dr dθ Fig. Hence. The net mass ﬂow out or in the r direction has an additional term which is the area change compared to the Cartesian coordinates. The mass conservation in cylindrical coordinates. -8.13) ∂ (ρ Uz ) dz ∂z (8. This change creates a diﬀerent diﬀerential equation with additional complications. In a similar fashion the net ﬂux in the z coordinate be written as net ﬂux in z direction = r dθ dr The net change in the θ direction is then net ﬂux in θ direction = dr dz ∂ρ Uθ dθ ∂θ (8.13) and dividing by inﬁnitesimal control volume. The change is ﬂux in r direction = dθ dz r ρ U r − r ρ Ur + ∂ρ Ur r dr ∂r (8. r .10) The net ﬂux in the r direction is then ∂ρ Ur r net ﬂux in the = dθ dz dr ∂r r direction (8. e.

this writing reduce the amount of writing and potentially can help think about the problem or situation in more conceptional way.15) and (8.17) + · (ρ U ) = 0 ∂t Advance material can be skipped The mass equation can be written in index notation for Cartesian coordinates.18) Compare to equation (8. t0 . There are very few cases where this equation is used on its own merit. The index notation really does not add much to the scientiﬁc understanding. the continuity equation becomes Continuity in Spherical Coordinates ∂ρ 1 ∂ r 2 ρ Ur 1 ∂ (ρ Uθ sin θ) 1 ∂ρ Uφ + 2 + + =0 ∂t r ∂r r sin θ ∂θ r sin θ ∂z (8. MASS CONSERVATION 195 Combining equation (8. However. The mass equation (see in the appendix for more information on the index notation) written as ∂ρ ∂ (ρ U )i + =0 ∂t ∂xi (8.15) Carrying similar operations for the spherical coordinates.16) The continuity equations (8. the upper surface is exposed to temperature T1 .9) divided by inﬁnitesimal control volume.1 Mass Conservation Examples Example 8.8. At time. For academic purposes. dr r dθ dz yields Continuity in Cylindrical Coordinates ∂ρ Uz ∂ρ 1 ∂ (r ρ Ur ) 1 ∂ρ Uθ + + + =0 ∂t r ∂r r ∂θ ∂z (8. Again remember that the meaning of repeated index is summation.1.8). Hence. the continuity equation can be written in a general vector form as Continuity Equation ∂ρ (8. End Advance material The use of these equations is normally combined with other equations (momentum and or energy equations). several examples are constructed here. It can be noticed that the second part of these equations is the divergence (see the Appendix A.2.2 page 274).8). .1: A layer of liquid has an initial height of H0 with an uniform temperature of T0 . 8.16) can be expressed in diﬀerent coordinates. (8.14) with the change in the control volume (8.2.

I.I.d) ρ1 − ρ0 H0 Equation (8. it is treated as a constant.b) results in ∂ρ ∂t ∂ρ Uy ∂y 0 ∂ Uy α HH−y 1 − e−β t 0 αβ H0 − y H0 (8.I. -8. Assume that the velocity at the lower boundary is zero at all times.d) relates the temperature with the time and the location was given in the question (it is not the solution of any model). Mass ﬂow due to temperature tom. Assume that the velocity is only a diﬀerence for example 8.I. ρ. as a function of time into the governing equation (8. The expression that connects the temperature with the space for the ﬁnal temperature as T − T0 H0 − y =α T1 − T0 H0 (8.I. DIFFERENTIAL ANALYSIS Assume that the actual temperature is ρ T1 exponentially approaches to a linear temperature proﬁle as depicted in Figure 8.b) with the boundary condition of zero velocity at the lower surface Uy (y = 0) = 0.3.196 CHAPTER 8. Neglect the mutual dependency of the temperature and the height.I.I.I. Substituting the density. Calculates the velocity of the liquid.e) =0 e−β t + ∂y Equation (8. For this question.c) The exponential decay is 1 − e−β t and thus the combination (with equation (8. Solution The situation is unsteady state thus the unsteady state and one dimensional continuity equation has to be used which is ∂ρ ∂ (ρUy ) + =0 ∂t ∂y (8.3.1 function of the y coordinate.a) T(t = 0) T(t > 0) T(t = ∞) where ρ1 is the density at the surface and where ρ0 is the density at the botFig.I.e) is ﬁrst order ODE with the boundary condition Uy (y = 0) = 0 which can be arranged as 0 ∂ Uy α HH−y 1 − e−β t 0 ∂y = −α β H0 − y H0 e−β t (8.a)) is ρ − ρ0 H0 − y =α 1 − e−β t (8.f) . H0(t) The density is a function of the temperT0 y ature according to ρ 1 0 T − T0 =α T1 − T0 ρ − ρ0 ρ1 − ρ0 (8.I. It can be noticed that the height H0 is a function of time.

Calculate the ﬁlm velocity ﬁeld if the density is a function of the temperature.I.19) If the ﬂuid is incompressible then the governing equation is a volume conservation as ·U = 0 Note that this equation appropriate only for a single phase case.I. it can be treated for the solution of equation (8. asH0 T0 T(x) T∞ sume that no mass transfer occurs or can x be neglected and the main mechanism is x heat transfer.2. MASS CONSERVATION 197 Uy is a function of the time but not y. The relationship between the density and the temperature is linear as T − T∞ ρ − ρ∞ =α (8. .I.4. Mass ﬂow in coating process perature is only a function of the distance for example 8. Example 8. 4 Since (8.2.f) holds for any time and thus.II.2.20) the time can be treated as constant for y integration. from the extraction point. Hence.a) ρ0 − ρ∞ T0 − T∞ State your assumptions.f) as a constant4 . the integration with respect to y yields Uy α H0 − y 1 − e−β t H0 = −α β 2 H0 − y 2 H0 e−β t y + c (8. Assume that the ﬁlm temFig. -8. End Solution 8.I.g) Utilizing the boundary condition Uy (y = 0) = 0 yields Uy α H0 − y 1 − e−β t H0 = −α β 2 H0 − y 2 H0 e−β t (y − 1) (8.2: In many coating processes a thin ﬁlm is created by a continuous process in which liquid injected into a moving belt that carries the material out as exhibited in Figure 8.2 Simpliﬁed Continuity Equation A simpliﬁed equation can be obtained for a steady state in which the transient term is eliminated as · (ρ U ) = 0 (8. Equation (8.8.h) It can be noticed that indeed the velocity is a function of the time and space y. For this example.4. The temperature and mass transfer takT0 ing place which reduces (or increases) the thickness of the ﬁlm.

However. See the resembles to equation (8. this author Master thesis is extension Higbie’s equation. Hence equation (8.II.II.b) can be written as Ux ∂ (ρ Uy ) ∂ρ =− ∂x ∂y (8. the discussion about this point is beyond the scope of this section. One of the early pioneers who suggest this idea is Higbie which Higbie’s equation named after him. ∂ρ α ∂F (x) = (ρ0 − ρ∞ ) ∂x Ux ∂x (8. 6 In reality this assumption is correct only in a certain range.b) + =0 ∂x ∂y At ﬁrst.198 Solution CHAPTER 8. Higbie suggested this idea which was rejected by the scientiﬁc establishment. however it can be considered as steady state.a) and thus Uy = α ∂F (x) (ρ0 − ρ∞ ) y + c ρ Ux ∂x (8.II.d) into the governing equation results in α ∂F (x) ∂Uy ρ = (ρ0 − ρ∞ ) ∂y Ux ∂x The density is expressed by equation (8. it can be assumed that the material moves with at the belt in the x direction in the same velocity. The solution is similar to the previous Example 8.c) Where Ux is the belt velocity.g) 5 The presentation of one dimension time dependent problem to two dimensions problems can be traced to heat and mass transfer problems.d) Substituting this relationship in equation (8. Applying the boundary condition Uy (t = 0) = 0 results in Uy = α ∂F (x) (ρ0 − ρ∞ ) y ρ(x) Ux ∂x End Solution (8. At any point the governing equation in coordinate system that moving with the belt is ∂ (ρ Ux ) ∂ (ρ Uy ) (8.II.II. . This assumption is consistent with the ﬁrst solution (no stability issues).f) (8.I. DIFFERENTIAL ANALYSIS This problem is somewhat similar to Example 8.1 for the general function T = F (x).b).e) Notice that ρ could “come” out of the derivative (why?) and move into the RHS.15 .II. If the frame of reference was moving with the belt then there is only velocity component in the y direction6 . On personal note.II.II. He spend the rest of his life to proof it and ending only to suicide.II.

Ux = a x2 + b y 2 Next.f) and switching sides to be ∂ Uy b + Uy = −a 2 x + x2 + y 2 (8. The velocity.8. Thus.19) is applicable and used as ∂ a x2 + b y 2 (m ex+y ) ∂ Uy (m ex+y ) + =0 ∂x ∂y (8.III. Ux has also arbitrary function in the y component. Assume that the density is constant and calculate the vertical velocity (y component) for the following x velocity component.III.III.III.2.III. assume the density is also a function of the location as ρ = m ex+y Where m is constant. It can be noted that x should be treated as a constant parameter for the y coordinate.20) thus 2ax + ∂Uy =0 ∂y (8. The homogeneous equation is ∂ Uy + Uy = 0 (8.III.g) is ﬁrst order ODE that can be solved by combination of the homogeneous solution with the private solution (see for explanation in the Appendix).h) ∂y .c) (8.c) is an ODE with constant coeﬃcients.III.3: The velocity in a two dimensional ﬁeld is assumed to be in a steady state. Notice the symmetry of the situation.e) Taking the derivative of the ﬁrst term and second part move the other side results in a 2 x + x2 + b 2 y a ex+y = − ex+y ∂ Uy + Uy ∂y (8.III.III. Calculate the velocity ﬁeld in this case.d) The integration constant in this case is not really a constant but rather an arbitrary function of x.f) The exponent can be canceled to simplify further the equation (8.b) (8. MASS CONSERVATION 199 Example 8. Solution The ﬂow ﬁeld must comply with the mass conservation (8.g) ∂y a Equation (8. Uy = − 2 a x + f (x) = −2 x y + f (x) (8.a) Equation (8. For the second part equation (8.III.III.

Check if the ﬂow is incompressible for this continuity equation should be used for constant density. Solution This problem is one dimensional unsteady state and for a compressible substance. Ux = (x t) z 2 Uy = (x t) + (y t) + (z t) Uz = (x t) + (y t) + (z t) (8. must be compressible ﬂow. the mass conservation is reduced only for one dimensional form as ∂ρ ∂ (Ux ρ) + =0 ∂t ∂x (8.i) End Solution Example 8.IV.V. The private solution is Uy |private = −b y 2 − 2 y + 2 − a x2 − 2 a x The total solution is Uy = c e−y + −b y 2 − 2 y + 2 − a x2 − 2 a x (8. Thus. DIFFERENTIAL ANALYSIS The solution for (8. This ﬂow can exist only for a limit time since over time the divergence is unbounded. ∂Ux ∂x = t2 z ∂Uy ∂y =t ∂Uz ∂z =t (8. Hence.5: Find the density as a function of the time for a given one dimensional ﬂow of Ux = x e5 α y (cos (α t)).a) Mathematically speaking. if it exist. this ﬂow ﬁeld is not steady state. However physically there are velocity components in y and z directions.III.h) is Uy = c e−y (see for explanation in the appendix). The initial density is ρ(t = 0) = ρ0 .200 CHAPTER 8. it can be observed that the velocity contains time component. End Solution Example 8.j) (8. In this problem these physical . this kind of presentation is possible.III.b) Or the combination of these derivatives is U = t2 z + 2 t (8.III.IV.c) The divergence isn’t zero thus this ﬂow.a) Is the ﬂow is incompressible? Is the ﬂow in a steady state condition? Solution To check whether the solution is in a steady state.IV.4: Can the following velocities co-exist.

V. MASS CONSERVATION 201 components are ignored for academic reasons.V. c3 .d) (8.a) and noticing that the density.V.V.8.V. Equation (8. Ux . and independent of x thus ρx + ∂ρ c1 = 5 α y = c3 ∂x e (8.V. ∂Ux = e5 α y (cos (α t)) ∂x (8.2. c1 vanishes as well and the solution contain only the homogeneous part and the private solution is dropped ρ = c2 e− x2 2 (8. ρ.e) is reduced to ODE and its solution is ρ= c1 sin (α t) + c2 α ∂ρ 5 α y e = c1 ∂x (8.j) .g) The term e5 α y is always positive.f) (8.e) The same can be done for the right hand side as ρ x e5 α y + (8. is a function of x results in ∂ρ ∂ρ 5 α y = −ρ x e5 α y (cos (α t)) − e (cos (α t)) ∂t ∂x Equation (8.V.V.c) can be separated to yield 1 ∂ρ ∂ρ 5 α y = −ρ x e5 α y − e cos (α t) ∂t ∂x (8. In that case the left hand side is 1 ∂ρ = c1 cos (α t) ∂t The solution of equation (8.b) into equation (8.h) Equation (8.i) which indicates that the solution is a complex number thus the constant.V.V.V.V. real value.c) Possible solution is when the left and the right hand sides are equal to a constant.a) is ﬁrst order partial diﬀerential equation which can be converted to ordinary diﬀerential equation when the velocity component.V.V. Hence.V.V.b) Substituting equation (8. The solution of (8. is substituted. must be zero and thus the constant.h) is given by impossible solution ρ=e −x 2 2 √ π i c3 erf √ c − 2 ix √ 2 (8.h) is constant coeﬃcients ﬁrst order ODE which its solution discussed extensively in the appendix.

the derivative can enter into the integral and thus combining the two integral on the RHS results in D Dt φ ρ dV = sys cv d (φ ρ) + dt · (ρ φ U ) dV (8. c2 is arbitrary function of the y coordinate. . DIFFERENTIAL ANALYSIS The solution is the multiplication of equation (8.24) Substituting equation (8. vector or tensor.g.V.23) yields D Dt φ ρ dV = sys d dt φ ρ dV + cv cv · (ρ φ U ) dV (8.V.3.24) into equation (8.33)) yields D Dt φ ρ dV = sys d dt φ ρ dV + cv A ρ φ U · dA (8.21) Where Φ is the total quantity of the system has a volume V and a surface area of A which are a function of time.23) The last term on the RHS can be converted using the divergence theorem (see the appendix7 ) from a surface integral into a volume integral (alternatively. t).V. The total amount of quantity that exist in arbitrary system is Φ= sys φ ρ dV (8. the volume integral can be changed to the surface integral) as ρ φ U · dA = A V · (ρ φ U ) dV (8.22) Using RTT to change the system to control volume (see equation (5. are presented.k) Where the constant. A change with time is DΦ D = Dt Dt φ ρ dV sys (8.202 CHAPTER 8.25) Since the volume of the control volume remains independent.1 Generalization of Mathematical Approach for Derivations In this section a general approach for the derivations for conservation of any quantity e. 8.f) is ρ = c2 e− x2 2 c1 sin (α t) + c2 α End Solution (8.j) by (8. Suppose that the property φ is under a study which is a function of the time and location as φ(x. scalar.3 Conservation of General Quantity 8. Basically the divergence theorem relates the ﬂow out (or) in and the sum of the all the changes inside the control volume. z. y.26) 7 These integrals are related to RTT.

30) · (ρ U ) = 0 (8.32) .29) can be rearranged as ∂ρ +U ∂t ·ρ+ρ ·U = 0 (8.31) relates the density rate of change or the volumetric change to the velocity divergence of the ﬂow ﬁeld. In that case D Φ = D ρ and hence equal to zero as Dt Dt φ d dV 1 ρ φ (8. z. z. y.1 Examples of Generalized of Quantities The General Mass Time Derivative For example. y. y. t) j + Uz (x. The integral is carried over arbitrary system. A ﬂuid particle velocity is a function of the location and time. t) = Ux (x.28) dx dy dz = 0 + · ρ 1 U dt - The integral is over arbitrary volume which mean that integrand is zero as ∂ρ + ∂t Equation (8.30) can be further rearranged so derivative of the density is equal the divergence of velocity as 1 ρ ∂ρ +U ∂t ·ρ =− ·U (8.3. Acceleration Direct Derivations One of the important points is to ﬁnd the particles acceleration of the ﬂuid. x. The term in the bracket LHS is referred in the literature as substantial derivative. CONSERVATION OF GENERAL QUANTITY 203 By the deﬁnition of equation (8. for the last derivations using φ = 1 which is the same for mass conservation. z.29) Equation (8.27) 8.8.3.2. For an inﬁnitesimal control volume the change is DΦ ∼ = Dt d (φ ρ) + dt dV · (ρ φ U ) dx dy dz (8. t) i + Uy (x.21) LHS can be change to simply to derivative of Φ. it can be written that U (x.3. t) k (8. Therefore.2 8. The substantial derivative represents the change rate of the density at a point which moves with the ﬂuid.31) Equation (8. y.

The area has a direction or orientation which control the results of this division. This division creates a tensor which the physical meaning will be explained here (the mathematical explanation can be found in the mathematical appendix of the book).38) It was shown that in static case (or in better words.35) The same can be developed to the other two coordinates and when combined to be (in a vector form) ∂U dU U = + (U · dt ∂t or in more explicit form as local acceleration convective acceleration )U (8. So it can be written that F τ = f (F . DIFFERENTIAL ANALYSIS U dU d Ux d Uy d Uz = i+ j+ k dt dt dt dt The velocity components are a function of four variables and hence =1 Ux Uy Uz (8. 8. While the ﬂow is in a steady state there is acceleration of the ﬂow.33) ∂ Ux d t ∂ U x d x ∂ U x d y ∂ U x d z d Ux = + + + dt ∂t d t ∂x d t ∂y d t ∂z d t The acceleration in the x can be written as ∂ Ux ∂ Ux d Ux ∂ Ux ∂ Ux ∂ Ux U = + Ux + Uy + Uz = + (U · dt ∂t ∂x ∂y ∂z ∂t ) Ux (8. when the shear stresses are absent) it was written τ = −P n (8. The stress is relationship between the force and area it is acting on or force divided by the area (division of vector by a vector).34) (8. The ﬂow in a nozzle is an example to ﬂow at steady state but yet has acceleration which ﬂow with low velocity can achieve a supersonic ﬂow.204 The acceleration will be CHAPTER 8.37) The time derivative referred in the literature as the local acceleration which disappear when the ﬂow is steady state.4 Momentum Conservation The relationship among the shear stress various components have to be established.36) ∂U ∂U dU ∂U ∂U = + U +U +U dt ∂t ∂x ∂y ∂z (8. A ) (8.39) .

δAx is the surface area of the tetrahedron in the x diFig.4. Stress diagram on a tetrahedron rection and δAz is the surface area of the shape. -8. The notation τ (xi ) is used to denote the stresses on xi surface. MOMENTUM CONSERVATION 205 It also was shown that the pressure has to be continuous.5. The opposing forces which acting on the slanted surface in the x are Fx = δAn τnn n · i − τn Where here ℵ. It can be noticed that no mathematical symbols are written between the components. and τxy is the stress acting on surface x in the y direction. However. The forces balance in the x direction excluding the slanted surface is Fx = −τyx δAy − τxx δAx − τzx δAz (8. The reason for this omission is that there is no physical meaning for it. these stresses that act on every point and have three components on every surface and depend on the surface orientation. A common approach is to collect the stress in a “standard” orientation and then if needed the stresses can be reorientated to a new direction.40) where τxx is the stress acting on surface x in the x direction. Similar “vectors” exist for the y and z coordinates which can be written in matrix forms τxx τxy τxz τ = τyx τyy τyz (8. The transformation n·i n·j n·k ·i ·j ·k ℵ·i ℵ·j ℵ·k When the tetrahedron is shrunk to a point relationship of the stress on the two sides can be expended by Taylor series and a diﬀerence is related to the ﬁrst derivative. If Y τn ℵ τnℓ X τ τ xx xy τxz τ yx τy y τyz (8.5.43) and n are the local unit coordinates on n surface. similarly for τxz .42) T0 Z τnn where δAy is the surface area of the tetrahedron in the y direction.8. The transformation is available because the “standard” surface can be transformed using trigonometrical functions. In Cartesian coordinates on surface in the x direction the stresses are τ (x) = τxx τxy τxz (8. tetrahedron in the z direction.41) τzx τzy τzz Suppose that a straight angle tetrahedron is under stress as shown in Figure 8.44) . matrix is then · i − τzℵℵ · i (8.

can create torque.46) Where Mz is the cubic moment around the cubic center and Izz 8 is the moment of inertia around that center. DIFFERENTIAL ANALYSIS the ﬁrst derivative is neglected (tetrahedron is without acceleration) the two sides are related as −τyx δAy − τxx δAx − τzx δAz = δAn τnn n · i − τn · i − τzℵℵ · i (8. for cases that body force. τyx τ However. the shear stress at point x + dx is τxy |x+dx = τxy + dτxy dx dx (8. The areas are related to each other through angles. However. dx For simplicity and generality. Hence. The momentum can be accessed by the shear stresses that act on it. The cubical rotation can involve two parts one distortion and one rotation. it is assumed x that the external body force exert a torque GT per unit volume at the speciﬁc location. Diagram to analysis the shear stress due to the fact that the body force is not tensor.47) The same can be said for τyx for y τyy direction. Advance material can be skipped 8 See for derivation in Example 3.45) The same can be done for y and z directions. uniform and hence act not through the mass center. These relationships provide the transformation for the diﬀerent orientations which depends only angles of the orientations.6. This matrix is referred to as stress tensor and as it can be observed has nine terms. The shear stress at point x is τxy . A ﬁnite angular distortion of inﬁnitesimal cube requires requires an inﬁnite shear.6 is Mz = Izz dθ dt (8.206 CHAPTER 8. The body force can exert torque is Fig. Balance of momentum around the z direction shown in Figure 8. . -8. the rotation of the inﬁnitesimal ﬂuid cube can be viewed as it is done almost as a solid body. such yy as the magnetic ﬁelds. The clarity of this analysis can y τyx be improved if additional terms are taken but the results will be the same. The Symmetry of the Stress Tensor A small liquid cubical has three possible rotation axes and here only one will be discussed.5 for moment of inertia. The τxy τxx dy τ normal body force (gravity) acts through τxx xy the cubic center of gravity and can be neglected (the changes are insigniﬁcant).

4. MOMENTUM CONSERVATION 207 The torque due to the shear stress in the surface direction results in a change due to the shear stress9 .53) This analysis can be done on the other two directions and hence the general conclusion is that GT + τij = τji (8.52) which means since that dx −→ 0 and dy −→ 0 that GT + τxy = τyx (8.48) into (8. The moment that results from this shear force (clockwise positive) is y+dy τxx (η) y η− dy 2 dη (8. It can be ignored.51) 2 2 dθ dt dx dy dz 2 = ρ dx dy dz (dx) + (dy) The actual components that contribute to the moment are =0 (dx)2 + (dy)2 =0 GT + τxy − τxy + ∂ (τyx − τxy ) =ρ ∂y 12 dθ dt (8.54) 9 This point bother this author in the completeness of the proof.49) results y+dy y τxx |y + dτxx dy η y η− dy 2 dη (8. . For example. The reason that this term neglected because the other face of the cubic contributes an identical term but in the opposing direction. τxx at x can be expended as a linear function τxx = τxx |y + dτxx dy η y (8.8. This point is for self convincing since it deals with a “strange” and problematic “animals” of integral of inﬁnitesimal length.50) The integral of (8.48) where η is the local coordinate in the y direction stating at y and mostly used between y < η < y + dy. but provided to those who wonder why body forces can contribute to the torque while pressure does not.50) isn’t zero (non symmetrical function around the center of integration).49) Substituting (8. End Advance material The net torque in the z-direction around the particle’s center would then be (τyx ) dx dy dz − 2 τxy + τyx + ∂τxy ∂x ∂τxy ∂x dx dy dz 2 + (τxy ) dx dy dz − 2 Izz (8.

gravitation forces.11) is also applicable for the small inﬁnitesimal cubic.7. z 10 . this change. for almost all practical purposes.55) 8. since Potto Project books are used extensively and numerous people asked to include this notation it was added. z and the j is any of the other x. . Previously it was shown that equation (6.208 CHAPTER 8. y. y.56) 10 The index notation is not the main mode of presentation in this book. The shear stress at diﬀerent surfaces. If the body forces eﬀect is neglected or do not exist in the problem then regardless the coordinate system τij = τji (i = j) (8. and internal forces. Later it will be used and generalized. All shear stress shown in surface x and x + dx. However.11) is equivalent to Newton second law for ﬂuids. For the case of GT = 0 the stress tensor become symmetrical. DIFFERENTIAL ANALYSIS where i is one of x. It is believed that this notation should and can be used only after the physical meaning was “digested. One direction of the vector equation will be derived for x Cartesian coordinate (see Figure ??). The magnetic body forces on the other hand is signiﬁcant and has to be included in the calculations. However. Equation (6.5 Derivations of the Momentum Equation τzz + ∂τzz dz ∂z ∂τ y y + ∂ y dy τ yy Z τxz τxx τxy τxx + ∂τxz dx ∂x ∂τxx dx ∂x τxz + τxy + τyy ∂τxy dx ∂x y τzz x Fig. can be neglected11 . The body force that acting on inﬁnitesimal cubic is i · f B = f B x dx dy dz (8.” 11 In the Dimensional Analysis a discussion about this eﬀect hopefully will be presented. The gravity is a body force that is considered in many kind of calculations and this force cause a change in symmetry of the stress tensor. For surface forces that acting on the cubic are surface forces. -8.

59) where i in this case is x.57) fxy = τyx |y+dy × dx dz − τyx |y × dx dz (8. The dot product yields the force in the directing of x. The surface forces in x direction on the x surface on are dAx dAx fxx = τxx |x+dx × dy dz − τxx |x × dy dz The surface forces in x direction on the y surface on are dAy dAy (8.64) .57) and (8.58) becomes internal forces surface forces body forces DUx & & ρ& & = dx dy dz Dt ∂τxx ∂τyx ∂τzx + + ∂x ∂y ∂z & & & & dx dy dz dx dy dz & & + fG x ρ & & (8.62) The same can be obtained for the z component and generally it is as ρ DUi = Dt ∂τii ∂τji ∂τki + + ∂i ∂j ∂j + ρ fG i (8.61) for y coordinate is ρ DUy = Dt ∂τxy ∂τyy ∂τzy + + ∂x ∂y ∂z + ρ fG y (8.63) Advance material can be skipped Where i is the balance direction and j and k are two other coordinates. The shear stresses can be expanded into Taylor series as τix |i+di = τix + ∂ (τix ) di + · · · ∂i i (8. or z. Hence.61) equation (8. The vector form is ρ U DU = Dt · τ (i) + ρ fG (8.63) can be written in a vector form which combined all three components into one equation.60) after rearrangement equations such as (8. Equation (8. the total net force results from the shear stress in the x direction is fx = ∂τxx ∂τyx ∂τzx + + ∂x ∂y ∂z dx dy dz (8. The advantage of the vector from allows the usage of the diﬀerent coordinates.5.8. y. DERIVATIONS OF THE MOMENTUM EQUATION 209 Where B is in the body force direction and f B is the body force per unit mass.58) The same can be written for z direction.

At time t + dt the control volume undergoes three diﬀerent changes. there is no left over stresses (In over words when the “no shear stress” situation exist the rate of deformation or strain is zero).8. when applying the shear stress in ﬂuids.8 (by the blue color). In contrast. rotates and changes the shape (the blow color in in Figure (8. There is no preference in the orientation (also call isentropic ﬂuid). Control volume at t and t + dt under continuous angle deformation. the deformation can be viewed as a function of the velocity ﬁeld. DIFFERENTIAL ANALYSIS τ (i) = τix i + τiy j + τiz k is part of the shear stress tensor and i can be any of the x. As engineers do in general. y. Additionally.65) End Advance material Equations (8. In solid material the shear stress yields a ﬁx amount of deformation. the result is a continuous deformation. a.64) requires that stress tensor be deﬁned. or z. b. The translational movement is referred to a movement of body without change of the body and without rotation.65) or (8. At time t. Or in index (Einstein) notation as ρ DUi ∂τji = + ρ fG i Dt ∂xi (8. Notice the three combinations of the deformation shown by purple color relative to blue color. The control volume moves to a new location. In this model the (shear) stresses and rate of strains are assumed to be linearly related. The similarity to solids the increase shear stress in ﬂuids yields larger deformations. and c. Furthermore. The rotation is the second movement that referred to change in of the relative orientation inside the control x’ 45◦ . the simplest model is assumed which referred as the solid continuum model. reduction of the shear stress does not return the material to its original state as in solids. a linear relationship between the shear stress to the rate of shear strain. Thus this “solid” model is a tions: = y B + + ∂Uy U + dt y ∂y D @ t + dt Uy dt A A @t Uxdt Uy + ∂Uy dt ∂x C x y’ linear relationship with three main assump- Fig.8)). -8. The relationship between the stress tensor and deformation depends on the classes of materials the stresses acts on.210 where here CHAPTER 8. the control volume is at a square shape and location as depicted in Figure 8.

This assumption is referred as isotropic viscosity. In general equation (8. For the assumption of linear ﬂuid12 τxy = µ Dγxy =µ Dt dUy dUx + dx dy B (8. It can be noticed at this stage. .9. ﬁrst assumption is mentioned above. The deformation of the control volume has several components. Shear stress at two coordinates in 45◦ orientations.66) The total angle deformation (two sides x and y) is dUy dUx Dγxy = + Dt dx dy dU (8.69) D y τxx τxy A τx y ’ ’ τx x ’ ’ C τyx τyy x y’ τyz = µ Dγyz =µ Dt dUz dUy + dy dz (8. 12 While 13 The Fig. the symmetry dxy = dUx was not assumed and or required because dy rotation of the control volume. In a similar fashion it can be written to other directions for x z as τxz = µ Dγxz =µ Dt dUz dUx + dx dz (8. The only missing thing.5. under isentropic material it is assumed that the contribution of all the shear stresses contribute equally. at this stage. Dγij =µ Dt dUj dUi + di dj x’ and for the directions of y z as 45◦ (8. The third change is the misconﬁguration or control volume deformation. The angle between x to the new location of the control volume can be approximate for a small angle as dγx = tan dt Uy + dUy dx dx − Uy dx = tan dUy dx ∼ dUy = dx (8.70) Note that the viscosity coeﬃcient (the linear coeﬃcient13 ) is assumed to be the same regardless of the direction. -8. the relationship between the two of stress tensor are found.71) not marked as important equation this equation is is source of the derivation. The shear stress is related to the change in angle of the control volume lower left corner. However. is the diagonal component which dealt below. µ is the “normal” or “ordinary” viscosity coeﬃcient which the linear coeﬃcient of proportionality the shear and it is assumed to be a property of the ﬂuid.67) In these derivatives. DERIVATIONS OF THE MOMENTUM EQUATION 211 volume.68) can be written as τij = µ where i = j and i = x or y or z.8.68) where.

The small deformations a . y) is frame related to the strain rates in the (x’ . The normal shear stress relates to the change in the diagonal line length change.72) dividing by dx and some rearrangements utilizing the identity τxy = τyx results in τxx + τyy + τyx = τx’ x’ 2 Setting the similar analysis in the y’ results in τxx + τyy − τyx = τy’ y’ 2 Subtracting (8. z) appears in shear matrix diagonal.73) Equation (8.76) (8. b.74) (8. c. The rate of strain in the x direction is c (8. In addition. and d in the Figure are related to the incremental linear strains. The force balance in the x’ is Ax cos θx dy τxx √ 1 1 1 1 √ + dx τyy √ + dx τyx √ + dy τxy √ = dx 2 τx’ x’ 2 2 2 2 Ay cos θy Ay cos θy Ax cos θy Ax’ (8.73) results in 2 τyx = τx’ x’ − τy’ y’ or dividing by 2 equation (8. Figure 8. on the “x” surface (lower surface) and the “y” (left) surface. The dx is construct so it equals to dy. y’ coordinates) and the angular strain rate in the regular (x.75) (8. The linear deformations in the x’ and y’ directions which are rotated 45◦ relative to the x and y axes can be expressed in both coordinates system. The forces acting in the direction of x ’ are combination several terms.10(a) depicts the deformations of the triangular particles between time t and t + dt. The diagonal lines (line BC and line AD in Figure 8. √ shear the stresses are acting in this direction. y’ ) frame.77) d x= dx .74) from (8.75) becomes τyx = 1 (τx x − τy’ y’ ) 2 ’’ (8. y coordinates). the sides AB and AC rotate in unequal amount which make one diagonal line longer and one diagonal line shorter. It can be noticed that “dx’ ” surface is 2 times larger than dx and dy surfaces. τii (where i is either . To ﬁnd the main (on the diagonal) stress the coordinates are rotate by 45◦ . y.75) relates the diﬀerence between the normal shear stress and the shear stresses in x’ . DIFFERENTIAL ANALYSIS Normal Stress The normal stress.8) in the control volume move to the new locations.9. For example. The angular strain rate in the (x. The relationship can be obtained by changing the coordinates orientation as depicted by Figure 8.x.212 CHAPTER 8.

in both sides (d/dx+b/dy) which in turn is related to combination of the two sides angles. -8.80) can be interpreted as (using equations (8. DERIVATIONS OF THE MOMENTUM EQUATION b 213 b y’ y x’ a 45◦ y a d+a c+b d c 45◦ y’ x’ x (a) Deformations of the isosceles triangular. The hypotenuse of the triangle is oriented in the x’ direction (again observe Figure 8. the ratio strain in the x direction is 2 2 d x’ = (c + b) + (a + d) √ 2dx (c + b) (c + b) √ + √ + f (dx’ ) 2 √ 2 2dx ∼0 (8.77).5. Deformation of the diﬀerent triangles for the calculations of the normal stress.10. Fig.79) Here.81) .10(b)). d x is the linear strain (increase in length divided by length) of the particle in the x direction. The change in the hypotenuse length is (c + b) + (a + d) . It ◦ can be approximated that the change is about 45 because changes are inﬁnitesimally small. The linear angular deformation in xy direction is dγxy = b+d dx (8.10(b).78). and d y is its linear strain in the y-direction. The linear strain in the x direction can be computed by observing Figure 8. Hence. and (8.8. Thus. The original length of the √ 2 2 hypotenuse 2dx. x (b) Deformation of the straight angle triangle.78) The total change in the deformation angle is related to tan θ.79)) d x’ = 1 2 a+b+c+d dx = 1 (d 2 y +d y + dγxy ) (8. (8. The rate of the strain in y direction is d y = a dx (8.80) Equation (8. cos 45◦ or sin 45◦ times the change contribute as ﬁrst approximation to change.

∂Uy’ D y’ = Dt ∂y’ (8.85) From equation (8.87) y’ (8.214 CHAPTER 8.88) changes into τx’ x’ − τy’ y’ = 2µ ∂Ux’ ∂Uy’ − ∂x’ ∂y’ Fig. The linear deformation is the diﬀerence between the two sides as D x’ ∂Ux’ = Dt ∂x’ (8. For example in y’ –z’ plain one can obtained τx’ x’ − τz’ z’ = 2µ ∂Ux’ ∂Uz’ − ∂x’ ∂z’ (8.83) Equation (8.87) and (8. (8. -8.89) Similar two equations can be obtained in the other two plains. Dashed squares denotes the movement without the linear change.11 depicts the approximate linear deformation of the element.68) it can be observed that the right hand side can be replaced by τxy /µ.87) can be written in the y’ and is similar by substituting the coordinates.85) can be continue and replaced as D x’ D y’ 1 − = (τx x − τy’ y’ ) Dt Dt 2µ ’ ’ Figure 8.90) . Changing it to the Eulerian coordinates and location diﬀerential transform equation (8.83) describing in Lagrangian coordinates a single particle. The rate of strain relations can be substituted by the velocity and equations (8.11.86) Uy’ + ∂Uy ’ ’ dy dt ∂y ’ Uy’dt The same way it can written for the y’ coordinate.82) Combining equation (8.82) results in d x’ −d y’ = dγxy (8.81) with equation (8.88) ∂Ux ’ ’ Ux ’ + dx dt ∂x ’ x’ Equation (8.83) into D x’ D y’ Dγxy − = Dt Dt Dt D x’ D y’ τxy − = Dt Dt µ (8.75) τxy be substituted and equation (8. DIFFERENTIAL ANALYSIS In the same fashion the strain in y’ coordinate can be interpreted to be d y’ = 1 (d 2 y +d y − dγxy ) (8. (8. Linear strain of the element purple denotes t and blue is for t + dt.84) From (8.

96) Advance material can be skipped 14 It 15 G. 1967.71).93) must be valid in any coordinate system thus equation (8. Batchelor.141. identical only in the limits not in mechanical measurements.93) can be written as τxx = −Pm + 2 µ ∂Ux 2 + µ ∂x 3 ·U (8. This pressure is a true scalar value of the ﬂow ﬁeld since the propriety is averaged or almost14 invariant to the coordinate transformation.94) Where Pm is the mechanical pressure and is deﬁned as Pm = − τxx + τyy + τzz 3 (8.90) results in 2 4 215 (3 − 1) τx’ x’ − τy’ y’ − τz’ z’ = (6 − 2) µ ∂Ux’ − 2µ ∂x’ ∂Uy’ ∂Uz’ + ∂y’ ∂z’ (8.5. Cambridge University Press. . DERIVATIONS OF THE MOMENTUM EQUATION Adding equations (8.91) transforms it into 3 τx’ x’ = τx’ x’ + τy’ y’ + τz’ z’ + 6 µ ∂Ux’ − 2µ ∂x’ ∂Uy’ ∂Uz’ ∂Ux’ + + ∂x’ ∂y’ ∂z’ (8. With this deﬁnition and noticing that the coordinate system x’ –y’ has no special signiﬁcance and hence equation (8. The mechanical pressure can be deﬁned as averaging of the normal stress acting on a inﬁnitesimal sphere. p.92) The further rearranging the results by dividing by 3 so that “mechanical pressure τx’ x’ = τx’ x’ + τy’ y’ + τz’ z’ ∂Ux’ 2 +2 µ − µ 3 ∂x’ 3 ∂Ux’ ∂Uy’ ∂Uz’ + + ∂x’ ∂y’ ∂z’ (8.95) It can be observed that the non main (diagonal) terms of the stress tensor are represented by an equation like (8.91) rearranging equation (8. An Introduction to Fluid Mechanics.89) and (8. Commonality engineers like to combined the two diﬀerence expressions into one as 2 τij = − Pm + µ 3 ·U δij + µ ∂Ui ∂Uj + ∂xj ∂xi (8.93) The “mechanical” pressure is the (negative) average value of pressure in directions of x’ –y’ –z’ . In situations where the main diagonal terms of the stress tensor are not the same in all directions (in some viscous ﬂows) this property can be served as a measure of the local normal stress.8. K. It can be shown that this two deﬁnitions are “identical” in the limits15 .

Equation (8. According to second law of thermodynamic derivations (not shown here and are under construction) demonstrate that λ must be positive. However.97) End Advance material where δij is the Kronecker delta what is δij = 1 when i = j and δij = 0 otherwise.100) 16 Since the publishing the version 0. The dimension of the bulk viscosity. it does not add any additional information. The thermodynamic pressure always tends to follow the mechanical pressure during a change.” Here the term bulk viscosity will be adapted. While this expression has the advantage of compact writing.98) The value of λ is obtained experimentally. This correction results in P = Pm + λ ·U (8. Second Viscosity Coeﬃcient The coeﬃcient 2/3µ is experimental and relates to viscosity.2. Then the association between normal stress and perpendicular stress was constructed.99) can be written in terms of the thermodynamic pressure P . an additional correction will be needed. This coeﬃcient is referred in the literature by several terms such as the “expansion viscosity” “second coeﬃcient of viscosity” and “bulk viscosity. it will be provide before version 0.9. connection between τxy and the deformation was built. The expansion rate of change and the ﬂuid molecular structure through λ control the diﬀerence. Using the coordinates transformation.216 or index notation 2 τij = − Pm + µ 3 CHAPTER 8. With God help.99) ·U (8. First.3 . are similar to the viscosity µ. λ. The linkage was established between the stress int he rotated coordinates to the deformation. this association was established. DIFFERENTIAL ANALYSIS ·U δij + µ ∂Ui ∂Uj + ∂xj ∂xi (8.0 several people ask me to summarize conceptually the issues.61) the relationship between the stress tensor and the velocity were to be established. To reduced one unknown (the shear stress) equation (8. This expression suggests a new deﬁnition of the thermodynamical pressure is 2 P = Pm + µ 3 Summary of The Stress Tensor The above derivations were provided as a long mathematical explanation16 . as τij = − P + 2 µ−λ 3 ·U δij + µ ∂Ui ∂Uj + ∂xj ∂xi (8. if the derivations before were to include additional terms.

Clearly for incompressible ﬂow. λ is large (3 times µ) but the net eﬀect is small because in that cases · U −→ 0.107) 17 The reason that the eﬀect vanish is because · U = 0.104) ∂Uy ∂y (8. For example. Only in micro ﬂuids and small and molecular scale such as in shock waves this eﬀect has some signiﬁcance. neglecting this eﬀect results in τij = −P δij + µ ∂Ui ∂Uj + ∂xj ∂xi (8. DERIVATIONS OF THE MOMENTUM EQUATION 217 The signiﬁcance of the diﬀerence between the thermodynamic pressure and the mechanical pressure associated with ﬂuid dilation which connected by · U . Thus. can be over 100 times larger than µ. is τxy = τyx = ∂Uy ∂Ux + ∂x ∂y (8. it can be written for spesiﬁc coordinates. In material such as water.106) Por in index form as ρ D Ui ∂ =− Dt ∂xi P+ 2 µ−λ 3 ·U + ∂ ∂xj µ ∂Ui ∂Uj + ∂xj ∂xi + f Bi (8. . λ. for the τxx it can be written that τxx = −P + 2 and the y coordinate the equation is τyy = −P + 2 however the mix stress. For complex liquids this coeﬃcient. For simple gas (dilute monatomic gases) it can be shown that λ vanishes.61) which results in ρ DUx Dt =− ∂ P+ 2 3µ −λ ∂x ·U +µ ∂ 2 Ux ∂ 2 Ux ∂ 2 Ux + + ∂x2 ∂y 2 ∂z 2 f +f B x (8. substitute equation (8. this coeﬃcient or the whole eﬀect is vanished17 .8.105) or in a vector form as ρ U DU =− P + Dt 1 µ+λ 3 ( ·U) + µ 2 U +fB (8.102) For the total eﬀect.103) ∂Ux ∂x (8.5.101).101) To explain equation (8. τxy . In fact this eﬀect is so insigniﬁcant that there is diﬃculty in to construct experiments so this eﬀect can be measured.100) into equation (8. In most cases. The physical meaning of · U represents the relative volume rate of change. the total eﬀect of the dilation on the ﬂow is very small.

A common velocity condition is that the liquid has the same value as the solid interface velocity. For this discussion.1 Boundary Conditions Categories The governing equations that were developed earlier requires some boundary conditions and initial conditions. In the literature. thus equation (8. the shear tensor will be separated into two categories.106) is reduced to 2 U DU =− P + Dt U +fB (8.6 Boundary Conditions and Driving Forces 8.108) or in the index notation it is written ρ ∂P ∂ 2U D Ui =− +µ + f Bi Dt ∂xi ∂xi ∂xj (8.110) + ρgx g Where gx is the the body force in the x direction (i ·g ).112) + ρgz 8. pressure (at the interface direction) and shear stress (perpendicular to the area).218 For incompressible ﬂow the term ρ CHAPTER 8.109) The momentum equation in Cartesian coordinate can be written explicitly for x coordinate as ρ ∂Ux + ∂t ∂Ux ∂Uy ∂Uz + Uy + Uz = ∂x ∂y ∂z 2 2 ∂P ∂ Ux ∂ Ux ∂ 2 Ux − +µ + + ∂x ∂x2 ∂y 2 ∂z 2 Ux (8. In y coordinate the momentum equation is ρ ∂Uy + ∂t ∂Uy ∂Uy ∂Uz + Uy + Uz = ∂x ∂y ∂z ∂P ∂2v ∂2v ∂2v − +µ + 2 + 2 + ρgy 2 ∂y ∂x ∂y ∂z Ux (8. The solid surface is rough thus the liquid participles (or molecules) are slowed to be at the solid surface velocity. These conditions described physical situations that are believed or should exist or approximated. pressure. These conditions can be categorized by the velocity.111) in z coordinate is ρ ∂Uz + ∂t ∂Uz ∂Uz ∂Uz + Uy + Uz = ∂x ∂y ∂z ∂P ∂ 2 Uz ∂ 2 Uz ∂ 2 Uz − +µ + + ∂z ∂x2 ∂y 2 ∂z 2 Ux (8. DIFFERENTIAL ANALYSIS · U vanishes. . or in more general terms as the shear stress conditions (mostly at the interface). this condition is referred as the “no slip” condition.6.

In another view.12. when the ﬂow is with a strong velocity ﬂuctuations.113) can be mathematically represented in direction another way for free surface conditions. For example. the free surface in the two dimensional case is represented as f (t. x. t pendicular velocity at the interface must be zero.113) is given in a vector form.115) As oppose to a given velocity at particular point. a requiret n ment on the acceleration (velocity) is given in unknown posiflow tion. Another condition which aﬀects whether the slip condition exist is who rapidly of the velocity change. y).113) where n is referred to the area direction (perpendicular to the area). The slip condition cannot be ignored in some regions. The location of the (free) moving boundary can be given as r f (r . The ”slip” condition is written in similar fashion to equation (8. the “no slip” condition is applicable to the ideal ﬂuid (“inviscid ﬂows”) because this kind of ﬂow normally deals with large scales. t) = 0 (8.117) . t) = 0 as the equation which describes the bounding surface.116) Dt This condition is called the kinematic boundary condition. -8. etc) (8.113) as U t · (U f luid − U boundary ) = f (Q.8. The diﬀerence between the small scale and the large scale is that the slip can be neglected in the large scale while the slip cannot be neglected in the small scale.Fig. the diﬀerence in the velocities vanishes as the scale increases. The condition becomes as 0= ∂f ∂f ∂f + Ux + Uy ∂t ∂x ∂y (8. While this condition (8.6. 1–Dimensional free surface describing n and b. scale. BOUNDARY CONDITIONS AND DRIVING FORCES 219 This boundary condition was experimentally observed under many conditions yet it is not universal true. To make sure that all the x material is accounted for in the control volume (does not cross b the free surface) the relative per. The condition (8. Mathematically the “no slip” condition is written as U t · (U f luid − U boundary ) = 0 (8. The slip condition is dealing with a diﬀerence in the velocity between the solid (or other material) and the ﬂuid media.114) Note. The perpendicular relative velocity at the surface must be zero and therefore f (x) y Df r = 0 on the surface f (r . The slip condition (as oppose to “no slip” condition) exist in situations where the scale is very small and the velocity is relatively very small. it is more common to write this condition as a given velocity at a certain point U( ) = U (8.

this constant is determined from the volume conservation.12) and R1 and R2 are principal radii. This condition is expressed mathematically equating the shear stress diﬀerence to the forces results due the surface tension. In same instances. Oﬀ course. The jump in shear stress can appear when the density has a jump in density. the jump in a shear stress (without a jump in density) does break a physical law. DIFFERENTIAL ANALYSIS The solution of this condition sometime is extremely hard to handle because the location is not given but the derivative on unknown location. it just for completeness and can be ignored for most purposes. While a jump in density does not break any physical laws (at least those present in the solution). The free surface is a special case of moving surfaces where the surface between two distinct ﬂuids.119) is written18 as τij nj + σ ni (1) 1 1 + R1 R2 = τij nj (2) (8. In this book. Furthermore. There are situations where the transition should be analyzed as a continuous transition between two phases. The jump in the density (between the two ﬂuids) creates a surface tension which oﬀset the jump in the shear stress. . The shear stress diﬀerence is ∆τ (n) = 0 = ∆τ (n) upper − ∆τ (n) lower surface surface (8. this condition will not be discussed (at least not plane to be written). there situations where the ﬂuid (above one of the sides) should be considered as weightless material. the moving surface condition) is that integration constant is unknown). In other cases. One of results of the free surface condition (or in general.121) 18 There is no additional beneﬁt in this writing.220 CHAPTER 8. and the density has a jump while the shear stress are continuous (in some case continuously approach zero value). If the surface is straight there is no jump in the shear stress. the transition is idealized an almost jump (a few molecules thickness).120) t · τ (t) = −t · where n is the unit normal and t is a unit tangent to the surface (notice that direction pointed out of the “center” see Figure 8.119) (8. n · τ (n) = σ 1 1 + R1 R2 σ (8. In reality the interface between these two ﬂuids is not a sharp transition but only approximation (see for the surface theory).118) where the index (n) indicate that shear stress are normal (in the surface area). In these cases the assumptions are that transition occurs in a sharp line. A jump in the shear stress creates inﬁnite force on the adjoin thin layer. this condition cannot be tolerated since inﬁnite velocity (acceleration) is impossible. In index notation equation (8. The condition with curved surface are out the scope of this book yet mathematically the condition is given as without explanation.

Gravity as Driving Force The body forces in general and gravity as a particular body force driving the ﬂow beside the velocity. The equation is given by ∂f ∂f + Ux = Uy ∂t ∂x (8. 197–217. Free surface of the steady ﬂow of a Newtonian ﬂuid in a ﬁnite channel.13. Rational Mech. The shear stress carry the material as if part of it was a solid Fig. The gravity can be considered as a constant force in most case (see for dimensional analysis for the reasons).122) 2 the unit vector is given as two vector in x and y and the radius is given by equation (1. Consider for example the case shown in Figure 8.13. material. The gravity is a common body force which is considered in many ﬂuid mechanics problems. Kerosene lamp. -8. 3. For example in one dimensional19 n= t= (−f (x). For example. The last condition is similar the pressure condition is of prescribed shear stress or some relationship to it. In this category include the boundary conditions with issue of surface tension which were discussed earlier. In ﬂuid (mostly liquid) shear stress pulling side (surface) will have limited eﬀect and but sometime is signiﬁcant and more rarely dominate. Shear Stress and Surface Tension as Driving Force If the ﬂuid was solid material. The static pressure is measured perpendicular to the ﬂow ﬂow direction.6. It can be noticed that the boundary conditions that involve the surface tension are of the kind that where the something is given on boundary but no at speciﬁc location. BOUNDARY CONDITIONS AND DRIVING FORCES 221 where 1 is the upper surface and 2 is the lower surface.8. 1) 1 + (f (x)) (1. M.123) The Pressure Condition The second condition that commonality prescribed at the interface is that the static pressure at speciﬁc value.47). . pulling the side will pull all the material. The liquid does not move up due the gravity (actually it is against the gravity) but because the surface tension move the liquid up. 19 A one example of a reference not in particularly important or signiﬁcant just a random example. in the kerosene lamp the burning occur at the surface of the lamp and the liquid is at the bottom. f (x)) 1 + (f (x)) 2 (8. 74 (1980). Anal. Arch. no. shear stress (including the surface tension) and the pressure. Jean.

6: Incompressible liquid ﬂows between two inﬁnite plates from the left to the right (as shown in Figure 8. For further enhance the understanding some of the derivations are repeated.15. The control volume shown in darker colors. One radius toward the center of the rode while the other is inﬁnite (approximated). -8. Even in that case.15).222 CHAPTER 8. The velocity at U(ri) = 0 mix zone the outer surface is unknown. The surface tension is a function of the temperature therefor the gradient in surface tension is results of temperature gradient. U which is deﬁned as positive when it move with the ﬂow. examples with two phase are presented. Schematic of kerosene lamp. example dealing won one phase are present. In the case. The upper surface is moving in Velocity. the physical condition in Figure 8. The static pressure per length is given as ∆P 20 . book. However. First. Example 8. the surface tension gradient remains. However the diameter at the depend on the surface tension. } } } 8. The ﬂuid surrounds the rode temperature and ﬂows upwards. somewhere downstream the temperature gradient is insigniﬁcant. the contribution due to the curvature is zero in the direction of the ﬂow. the velocity at the gradent surface of the inner rode is zero. Later. The distance between the plates is . 20 The diﬀerence is measured at the bottom point of the plate. this eﬀect is not discussed.13 can be used and idealized as a ﬂow around an inner rode.7 Examples for Diﬀerential Equation (Navier-Stokes) Examples of an one-dimensional ﬂow driven by the shear stress and pressure are presented. Uℓ y flow direction dy x z Fig. The only (almost) propelling source of the ﬂow is the surface gradient. -8. Flow between two plates top moving at U to the right (as positive). DIFFERENTIAL ANALYSIS To understand how to apply the surface tension boundary condition. In that case. In this Fig. The boundary condition at outer surface given by a jump of the shear ∂U ∂σ constant = µ T ∂r ∂h stress. It can be notice that under the assumption here there are two principal radii.14. .

127) On the upper surface is diﬀerent by Taylor explanation as ∼ =0 dU d2 U d3 U 2 τ xy = µ + dy + dy + · · · dy 2 3 dy dy The net eﬀect of these two will be diﬀerence between them µ dU d2 U dU ∼ d2 U + dy − µ = µ 2 dy dy dy 2 dy dy (8.126) The shear stress on the lower surface based on Newtonian ﬂuid is τ xy = −µ dU dy (8.125) Thus.124) The momentum is not accumulated (steady state and constant density). The only diﬀerence in the pressure is in the x direction and thus P− P+ dP dx dx =− dP dx dx (8. the velocity in and out are the same (constant density). the ﬂow in and the ﬂow out are equal.7.129) The assumptions is that there is no pressure diﬀerence in the z direction. Furthermore.8.128) (8.131) . EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) Solution In this example. Further because no change of the thus ρ Ux Urn dA = 0 A (8. the mass conservation yields =0 223 d dt ρdV = − cv cv ρ Urn dA = 0 (8.130) The momentum equation in the x direction then results (no gravity eﬀects) in − dP d2 U =µ 2 dx dy (8. The momentum conservation is − cv P dA + cv τ xy dA = 0 (8.

No assumption was imposed on the pressure distribution.75 Ψ = 0.tween -1. The boundaries conditions are 1.25 Ψ = 0.75 Ψ = −0. The problem is still one dimensional because the ﬂow velocity is a function 21 A discussion about the boundary will be presented.0 0.134) (8. In that case.8 0.25 Ψ = 1.2 0.133) (8.75 green line to 3 the blue line. the left hand side is equal to constant.132) Applying the boundary conditions results in =Ψ Ux (y) = y y y 2 dP 1− + U0 2µ dx For the case where the pressure gradient is zero the velocity is linear as was discussed earlier in Chapter 1.0 0.224 CHAPTER 8.75 Ψ = 1. However.75 Ψ = 2.3 0.75 Ψ = −1.16.6 0.25 Ψ = −0.4 0.5 0.131) is a partial diﬀerential equation but can be treated as ordinary diﬀerential equation in the z direction of the pressure diﬀerence is uniform.2 Ψ = −1.9 1. DIFFERENTIAL ANALYSIS Equation (8. Newtonian ﬂuid. -8.25 Ψ = 2. . The “standard” boundary conditions is non–vanishing pressure gradient (that is the pressure exist) and velocity of the upper or lower surface or Fig.135) Cylindrical Coordinates Similarly the problem of one dimensional ﬂow can be constructed for cylindrical coordinates.7 0. One dimensional ﬂow with a shear both.6 0.0 y ℓ October 4.1 0. dition21 . It is common to assume that the between two plates when Ψ change value be“no slip” condition on the boundaries con.131) is Ux = − 1 dP 2 y + c2 y + c3 2 dx (8.2 0. 2010 Ux (y = 0) = 0 Ux (y = ) = U The solution of the “ordinary” diﬀerential equation (8.8 Ux Uℓ 0.25 1.4 0.131) was constructed under Velocity distributions in one dimensional ﬂow several assumptions which include the direction of the ﬂow. Equation (8.0 0. if the plates or the boundary conditions do not move the solution is Ux (y) = dP y 1− U0 2µ dx End Solution 2 + y (8.

-8. The analysis can be carried out by two diﬀerent approaches: one is a kind of short cut and one for the full analysis. The full analysis will be presented to review the the previous analysis of building the equation. Thus. The control volume of liquid element in “short cut” Fig b.17. the last term is ρ Uz Urn dA = 0 (8.138) Uz Urn dA = − z Uz+dz 2 dA Uz 2 dA =ρ z Uz+dz 2 − Uz 2 dA (8.136) The shear stress in the front and back surfaces do no act in the z direction.8. Poiseuille study ﬂow in a small diameters (he was not familiar with the concept of Reynolds numbers). The control volume of liquid element in cylindrical coordinates.17a is − P dA + τ dA = ρ Uz Urn dA (8. The momentum equation for the control volume shown in the Figure 8. of (only) radius.137) Pz + ∂P dz − Pz ∂z π r2 = ∂P dz π r2 ∂z (8.7. The control volume of liquid element in cylindrical coordinates.140) . EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) 225 r θ dz r θ dz r θ flow Directi o r z n r θ flow Directi o z n dr Fig a. This ﬂow referred as Poiseuille ﬂow after Jean Louis Poiseuille a French Physician who investigated ﬂow of blood in veins. The shear stress on the circumferential part small dark blue shown in Figure 8. Fig. Hence.139) The term Uz+dz − Uz is zero because Uz+dz − Uz as can be shown by conservation of the mass for any element.17a is dUz τ dA = µ 2 π r dz dr The pressure integral is P dA = (Pzd z − Pz ) π r2 = The last term is ρ Uz Urn dA = ρ ρ z+dz 2 2 dA (8.

144) It can be noticed that asymmetrical element22 was eliminated due to the smart short cut.145) (8.146) The trick in the construction of control volumes such as the above can shorten the solution of problems. DIFFERENTIAL ANALYSIS Substituting equation (8.138) into equation (8. it has several deﬁciencies which include the ability to incorporate diﬀerent boundary conditions such as pipe within a pipe.147) The shear stress can be expressed in a Taylor series as τ (z + dz) = µ 22 Asymmetrical dU dr + r d2 U dr2 dr + · · · r (8.17a. The governing equation can be constructed by this smaller control volume for various boundary condition. However. While the above analysis provides a solution. The “no slip” condition is assumed Uz (r = R) = 0 (8.148) element or function is −f (x) = f (−x) .136) results in µ Which shrinks to 2 µ dUz ∂P =− r dr ∂z (8.142) is a ﬁrst order diﬀerential equation for which only one boundary condition is needed.137) and (8.136) since the condition did not change. The shear stress in the z direction can be on the control volume (again the front and the back do not contribute here) is expressed τ dA = µ dU dA dr (8. The applicable equation is still (8.141) Equation (8.142) results in Uz = − 1 ∂P 2 r + c1 µ ∂z (8.142) dUz ∂P & & 2&¡ & = − π r dz dz π 2 & &r £ dr ∂z (8.226 CHAPTER 8. Integrating equation (8. in this analysis the control volume that will be used is of Figure 8. However. this chapter deals with construction of diﬀerential analysis more general analysis is provided. The integration constant obtained via the application of the boundary condition which is c1 = − The solution is Uz = 1 ∂P 2 r R 1− µ ∂z R 2 1 ∂P 2 R µ ∂z (8.143) Where R is the outer radius of pipe or cylinder.

150) Substituting equation (8. The same result is obtained.150) and equation (8. The only diﬀerence is the boundary conditions for demonstration of this point see the following example 8.152) is the governing equation only the z direction.136) results in µ Or µ d2 U ∂P =− dr2 dz (8.148) remembering that velocity is canceled into (8. The double integration results in Uz = − 1 ∂P 2 r + c1 r + c2 2 µ dz (8. the governing equation (8. The boundary conditions “no slip” at the outer radius and symmetry at the center which are Uz (r = 0) = 0 dU dr (r = 0) = 0 (8.152) can be integrated since the left right side is a function of r and right is a function of x.8.154) The symmetry requirement (derivative at r = 0 is equal zero force the coeﬃcient c1 to be zero.7.151) Equation (8. However.7: .152) d2 U ∂P 2 π r dz dr = − dz 2 π r dr dr2 dz (8.152) or its solution (8.154) can be used to solve problem that are not symmetrical.149) (8.7. EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) The net change in the shear stress is d2 U τ dA = µ 2 2 π r dz dr dr The pressure diﬀerence P dA = ∂P dz 2 π r dr dz dA A 227 (8.153) Equation (8. Example 8.

That is. For this mode the ﬂow is assumed to be one dimensional. DIFFERENTIAL ANALYSIS r θ θ r Fig.VII. In fact there is no physical reason why the ﬂow should be only one dimensional.b) Dividing equation (8. It can be noticed that it this case the change with time is zero but also the there is no acceleration (convective acceleration) is zero =f (t) =0 ∂U ∂Uz Uφ z ρ + Ur + ∂r r ∂t =0 Uz =f (φ) =0 ∂Uz ∂Uz +Uz =0 ∂φ ∂z ro flow Directi o ut r in z n (8. Solution CHAPTER 8. or z coordinate.VII.VII.228 A liquid with constant density is ﬂowing between two cylinders as shown in Figure 8. Thus only equation in z coordinate is needed. It turn out that the “simple” solution is the ﬂow ﬁrst mode that appear in reality. the appropriate version of the Navier–Stokes equation will be used.18. However. the velocity isn’t a function of the angle. Calculate the ﬂow rate for a given pressure gradient.a) required boundary conditions which are Uz (r = ri ) = Uz (r = ro ) = Integrating equation (8. The situation is best suitable to solved in cylindrical coordinates.a) The PDE above (8.18. Assume that the velocity at the surface of the cylinders is zero calculate the velocity proﬁle.VII.155) The steady state governing equation is ρ ¡ =0=− 0 ∂P +µ ∂z 1 ∂ r ∂r r ∂Uz ∂r =0 + ··· ρ gz + $$ (8. After the previous example. Flow of liquid between concentric cylinders for example 8. Build the velocity proﬁle when the ﬂow is one directional and viscosity is Newtonian.a) once results in r ∂Uz 1 ∂P 2 = r + c1 ∂r 2 µ ∂z (8.c) and second integration results in ∂Uz 1 ∂P c1 = r+ ∂r 2 µ ∂z r (8.d) .VII.7.VII. it is possible to satisfy the boundary conditions. In this solution will be discussing the ﬂow ﬁrst mode. One of the solution of this problems is one dimensional. -8.VII.c) 0 0 (8.

1957 in Munchen 24 Extensive discussion can be found in this author master thesis. This problem is related to many industrial process and is fundamental in understanding many industrial processes.VII.h) ∂P ln(ri ) ro 2 − ln(ro ) ri 2 dz 1 ∂P 2 ro + c1 ln ro + c2 4 µ ∂z (8.VII. .j) The next example deals with the gravity as body force in two dimensional ﬂow.VII.g) = The solution is then Uz (r) = 1 ∂P 2 1 r + ln 4 µ ∂z 4µ 1 + ln 4µ The ﬂow rate is then Q= ri End Solution ro ri ∂P ro 2 − ri 2 ln r dz (8.f) with the second boundary condition yields 0= The solution is c1 c2 = 1 ln 4µ 1 ln 4µ ro ri ro ri ∂P ro 2 − ri 2 dz (8.8: In many situations in nature and many industrial processes liquid ﬂows downstream 23 German mechanical engineer. Example 8.7.VII. this analysis is a building bloc for heat and mass transfer understanding24 . This problem study by Nusselt23 which developed the basics equations.8.d) results in Uz = 1 ∂P 2 r + c1 ln r + c2 4 µ ∂z 229 (8.VII.VII. Comprehensive discussion about this problem can be found this author Master thesis.e) Applying the ﬁrst boundary condition results in 0= 1 ∂P 2 ri + c1 ln ri + c2 4 µ ∂z (8. Furthermore. 1882 September 1. EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) Integration of equation (8.i) ro ri ∂P ln(ri ) ro 2 − ln(ro ) ri 2 dz ro Uz (r)dA (8.VII. Ernst Kraft Wilhelm Nusselt born November 25.

pump Solution This problem is satiable to Cartesian coordinates in which x coordinate is pointed in the ﬂow direction and y perpendicular to ﬂow direction (depicted in Figure 8. For this system. Assume that the ﬂow obtains a steady state after some length (and the acceleration vanished). The surface has no curvature and hence the pressure at liquid side similar to the gas phase and the only change in liquid is in the y direction. Calculate the velocity proﬁle.19. The second and the third terms in the convective acceleration are zero because the velocity at that direction is zero (Uy = Uz = 0). DIFFERENTIAL ANALYSIS on inclined plate at θ as shown in Figure 8. Write the governing equations for this situation.a) − g sin θ ∂ 2 Ux ∂ 2 Ux ∂P ∂ 2 Ux + ρ gx +µ + + ∂x2 ∂x ∂y 2 ∂z 2 The ﬁrst term of the acceleration is zero because the ﬂow is in a steady state.VIII. the gravity in the x direction is g sin θ while the direction of y the gravity is g cos θ. Assume that the ﬂow is one Fig.230 CHAPTER 8. -8. The dominate force is the gravity. Hence the pressure at the gas phase is almost constant hence the pressure at the interface in the liquid is constant. The governing in the x direction is =f (t) ∂U x ρ + ∂t =0 −0 ∂Ux ∂Uy ∂Uz Ux + Uy + Uz = ∂x ∂y ∂z =0 ∼0 =0 =0 (8.1 dimensional in the x direction. For this example. The pressure is almost constant along the x coordinate.19. assume that the y gas density is zero (located outside the liqx uid domain). assume that the ﬂow is two dimenθ g cos θ g θ sional. Mass ﬂow due to temperature diﬀerence for example 8. The ﬁrst term of the convective acceleration is zero under the assumption of this example ﬂow is fully developed and hence not a function of x (nothing to be “improved”). the pressure loss in the gas phase (mostly air) is negligible.19). As it will be shown later. Fully developed ﬂow means that the ﬁrst term of the velocity Laplacian is zero ( ∂Ux ≡ 0). For simplicg sin θ ity. . The last term of the velocity Laplacian is zero because no velocity in ∂x the z direction. Assume that “scale” is large h enough so that the “no slip” condition prevail at the plate (bottom).

VIII.VIII.VIII.g) The second integration applying the second boundary condition yields c2 = 0 results in τair g sin θ 2 y h − y2 − Ux = (8.VIII.f) τair 1 g sin θ h + µ ν µ ρ (8.e) ∂y µ The integration constant can be obtain by applying the condition (8.VIII. the velocity proﬁle is Ux = The ﬂow rate per unit width is Q = W h g sin θ 2 h y − y2 ν (8.VIII.j) Where W here is the width into the page of the ﬂow.VIII.VIII.VIII.i) Ux dA = A 0 g sin θ τair 2 h y − y2 − ν µ dy (8.VIII.7.VIII.c) The boundary at the interface is simpliﬁed to be ∂Ux ∂y = τair (∼ 0) y=0 (8.k) .d) If there is additional requirement. equation (8.VIII.b) yields ρ ∂Ux = g sin θ y + c1 (8. EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) Thus.a) is reduced to 0=µ ∂ 2 Ux + ρ g sin θ ∂y 2 231 (8. the governing equation can not be suﬃcient from the mathematical point of view.b) With boundary condition of “no slip” at the bottom because the large scale and steady state Ux (y = 0) = 0 (8. Integration of equation (8.8. Which results in Q g sin θ 2 h3 τair h = − W ν 3 µ (8.VIII.h) ν µ When the shear stress caused by the air is neglected.d) as τair Solving for c1 results in c1 = ∂Ux =µ ∂y y = −ρ g sin θ h +c1 µ h (8. such a speciﬁc velocity at the surface.

Calculate the minimum shear stress that required to operate the lump (alternatively. the maximum height). The Fig.1 Interfacial Instability In Example 8. The phenomenon is explained by the fact that there is somewhere instability which is transferred into the ﬂow. For the liquid. two of the boundary conditions for these equations are the identical and thus the six boundary conditions are really only 4 boundary conditions. However.l) Note the shear stress at the interface can be positive or negative and hence can increase or decrease the ﬂow rate and the averaged velocity. Assume that radios of the liquid and solid core are given and the ﬂow is at steady state. solution to the diﬀerential governing equations provides only two constants. The experiments were conducted on a solid concrete laboratory and the ﬂow was in a very stable system. End Solution In the following following example the issue of driving force of the ﬂow through curved interface is examined. the boundary condition of “no slip” at the bottom surface of liquid must be satisﬁed. 8.7. Example 8. No matter how low ﬂow rate was small and big occurred. If the air is considered two r( liq uid governing equations must be solved one ) ah for the air (gas) phase and one for water h (liquid) phase. .VIII. Thus. This explanation bothered this author. The second domain (the gas phase) provides another equation with two constants but again three boundary conditions need to satisﬁed. 25 The author was hired to do experiments on thin ﬁlm (gravity ﬂow). These experiments were to study the formation of small and big waves at the interface.232 The average velocity is then CHAPTER 8.9: A simpliﬁed ﬂow version the kerosene lump is of liquid moving up on a solid core.8 no requirement was made sa as for the velocity at the interface (the upm e air so ve (g lu loc per boundary). -8. The ﬂow in the kerosene lamp is depends on the surface tension. Two boundary conditions must be satisﬁed at the interface. DIFFERENTIAL ANALYSIS Q 2 W = g sin θ 2 h − τair Ux = h ν 3 µ (8. The ﬂow surface is curved and thus pressure is not equal on both sides of the interface.20. thus current explanation was developed to explain the wavy phenomenon occurs. there is total of three boundary conditions25 to be satisﬁed. The vanishing shear stress as tio it ) n y y at the interface was the only requirement x w ate was applied. Flow of liquid in partially ﬁlled duct.

7.157) The gas velocity at the upper interface is vanished thus Ux g [(1 + a) h] = 0 At the interface the “no slip” condition is regularly applied and thus Ux g (h) = Ux (h) Also at the interface (a straight surface). EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) The governing equation solution26 for the gas phase (h ≥ y ≥ a h) is Ux g = g sin θ 2 y + c1 y + c2 2 νg 233 (8.163) µg µ g sin θ h + c1 µg = g sin θ h + c3 µ νg ν Combining boundary conditions equation(8.165) equation results from double integrating of equation (8.161) which leads c4 = 0 Applying equation (8.159) with (8.161) The same can be said for boundary condition (8.160) yields ρg ρ (8.158) results in 0= g sin θ 2 h (1 + a)2 + c1 h (1 + a) + c2 2 νg (8. the constants c1 and c2 are dimensional which mean that they have physical units (c1 −→ [1/sec] The governing equation in the liquid phase (0 ≥ y ≥ h) is Ux = g sin θ 2 y + c3 y + c4 2ν (8. .158) Assuming “no slip” for the liquid at the bottom boundary as Ux (0) = 0 The boundary condition (8.159) (8.8.162) (8.VIII. the shear stress must be continuous µg ∂Uxg ∂Ux =µ ∂y ∂y (8.162) results in g sin θ 2 g sin θ 2 h + c1 h + c2 = h + c3 h 2 νg 2ν 26 This (8.156) Note.160) (8.b) and subtitling ν = µ/ρ.164) (8.

169) Or rearranging equation (8.171) .168) 2 νg h c1 ¡ 2 h£ g sin θ + 2 νg c2 νg 2 νg h c3 ¡ = + 2 sin θ h2 g sin θ ν g h£ (8. DIFFERENTIAL ANALYSIS Advance material can be skipped The solution of equation (8.164) 1 2 2 2 νg c1 2 c2 νg + g h sin θ g h2 sin θ (8.165) is obtained by computer algebra (see in the code) to be sin θ (g h ρg (2 ρg ν ρ + 1) + a g h ν ) ρg (2 a ν + 2 ν ) (8.234 CHAPTER 8.167) can be written as (1 + a) = C1 + C2 2 (8.165) 1+ νg c1 ρ µ νg c3 = + g h sin θ ρg µg g h sin θ (8.170) This presentation provide similarity and it will be shown in the Dimensional analysis chapter better physical understanding of the situation.162).166) c1 = − c2 = sin θ g h2 ρg (2 ρg ν ρ + 1) − g h2 ν 2 ρg ν sin θ (g h ρg (2 a ρg ν ρ − 1) − a g h ν ) ρg (2 a ν + 2 ν ) End Advance material c3 = When solving this kinds of mathematical problem the engineers reduce it to minimum amount of parameters to reduce the labor involve.164) and (8.169) C1 C2 C3 νg 2 νg c1 2 νg c2 2 νg c3 −1= + 2 − ν h g sin θ h g sin θ g h sin θ (8. (8.162) transformed by some simple rearrangement to be C1 C2 (1 + a) = And equation (8.167) C1 1 µ 2 µg C3 1+ and equation (8. Equation (8. So equation (8.

The boundary conditions of ﬂow with inﬁnite depth is that ﬂow at the interface is zero. The requirement of the shear stress in the inﬁnite is zero as well. EXAMPLES FOR DIFFERENTIAL EQUATION (NAVIER-STOKES) Further rearranging equation (8.168) ρ C1 µ C3 −1= − ρg 2 µg 2 and equation (8. ﬂow at inﬁnite is zero.7.175) − C2 = µg ρ +a µ ρg 2 (8. The point here the understanding issue related to boundary condition not per se solution of the problem.8. mostly the gas.172) (8. This is one of the source of the instability at the interface.173) to be µg ρ − 1 = C1 + C2 − C3 µ ρg (8. What happen when the lighter ﬂuid.176) C3 = − µg ρ + a2 + 2 a + 2 µ ρg (8.174) The set of equation can be solved for the any ratio of the density and dynamic viscosity. 28 This 27 Later . The ratio of the dynamics viscosity can be eliminated from equation (8. is inﬁnite long.177) The two diﬀerent ﬂuids28 have ﬂow have a solution as long as the distance is ﬁnite reasonable similar.170) νg − 1 = C1 + C2 − C3 ν 235 (8. Combining inﬁnite size domain of one ﬂuid with ﬁnite size on the other one side results in unstable interface. There is no way obtain one dimensional solution for such case and there is a component in the y direction. The solution for the constant is C1 = ρg µg µg − 2 + a2 + 2 a +2 ρ µ µ µg µg −2 +3 + a2 µ µ µg µ µg −1 −2 µ (8. it will be move to the Dimensional Chapter topic will be covered in dimensional analysis in more extensively.173) This process that was shown here is referred as non–dimensionalization27 .

236 CHAPTER 8. DIFFERENTIAL ANALYSIS .

who did not consider the ﬂow as two–phase ﬂow and ignoring the air. Thus. the nature of multiphase ﬂow requires solving many equations. Calculations of many kinds of ﬂow deals with more than one phase or material ﬂow1 . As result. Initially the models were using simple assumptions. only the trends and simple calculations are described. there is not a consensus to the exact map of many ﬂow regimes. 2 Or when the scientiﬁc principles simply dictate.2 History The study of multi–phase ﬂow started for practical purposes after World War II. For simple models.D. In the past. This book attempts to describe these issues as a fundamentals of physical aspects and less as a literature review. The author believes that the trends and eﬀects of multiphase ﬂow could and should be introduced and considered by engineers.CHAPTER 9 Multi–Phase Flow 9. there was a Ph. 237 .1 Introduction Traditionally. For many engineers. Additionally. is required for many engineering problems. this class will be the only opportunity to be exposed to this topic. In fact. It is recognized that multiphase ﬂow is still evolving. books on multiphase ﬂow were written more as a literature review or heavy on the mathematics. in many books the representations is by writing the whole set governing equations. 9. it is believed that the interactions/calculations requires a full year class and hence. the topic of multi–phase ﬂow is ignored in an introductory class on ﬂuid mechanics. Here. This chapter provides information that is more or less in consensus2 .there are two possibilities (1) the ﬂuids/materials are ﬂowing in well homogeneous mixed (where the main problem 1 An example. The knowledge in this topic without any doubts. working for the government who analyzed ﬁling cavity with liquid metal (aluminum). his analysis is in the twilight zone not in the real world.

They postulate that there is a relationship between the pressure loss of a single phase and combine phases pressure loss as a function of the pressure loss of the other phase. .3 What to Expect From This Chapter As oppose to the tradition of the other chapters in this book and all other Potto project books. If the pressure loss was linear then the total loss will be the summation of the two pressure losses (of the lighter liquid (gas) and the heavy liquid). Researchers that followed Lockhart and Martinelli looked for a diﬀerent map for diﬀerent combination of phases. 9. double choking phenomenon (hopefully).238 CHAPTER 9. will know and understand the trends. and actual calculation of pressure of the diﬀerent regimes. It is an attempt to explain and convince all the readers that the multi–phase ﬂow must be included in introductory class on ﬂuid mechanics3 . Also the researchers looked at the situation when the diﬀerent regimes are applicable. and calculation of pressure drop of simple homogeneous model. This chapter will introduce these concepts so that the engineer not only be able to understand a conversation on multi-phase but also. However. this chapter will explain the core concepts of the multiphase ﬂow and their relationship. researchers started to look for diﬀerent ﬂow regimes and provided diﬀerent models. Under this assumption the total is not linear and experimental correlation was made. partial discussion on speed of sound of diﬀerent regimes. When it became apparent that speciﬁc models were needed for diﬀerent situations. (2) the ﬂuids/materials are ﬂowing separately where the actual total loss pressure can be correlated based on the separate pressure loss of each of the material. Hence. ﬂow parameters eﬀects on the ﬂow regimes. phase change or transfer processes during ﬂow. MULTI–PHASE FLOW to ﬁnd the viscosity). Taitle and Duckler suggested a map based on ﬁve non-dimensional groups which are considered as the most useful today. the concept of ﬂow regimes. The ﬂow patterns or regimes were not considered. and more importantly. This chapter will provide: a category of combination of phases. For example. multi–phase ﬂow parameters deﬁnitions. This was suggested by Lockhart and Martinelli who use a model where the ﬂow of the two ﬂuids are independent of each other. and importance to real world. this chapter will not provide a discussion of transient problems. Taitle–Duckler’s map is not applicable for microgravity. Taitle and Duckler’s map is not universal and it is only applied to certain liquid–gas conditions. a description of what to expect in this chapter is provided. Which leads to the concept of ﬂow regime maps. However. It turned out this idea provides a good crude results in some cases.

However. The creation 3 This author feels that he is in an unique position to inﬂuence many in the ﬁeld of ﬂuid mechanics. consider air ﬂow that was discussed and presented earlier as a single phase ﬂow. -9. many layers (inﬁnite) of diﬀerent materials). The number of the downloads of the book on Fundamental of compressible ﬂow has exceed more than 100. Hence. The results of the calculations of a mixture do not change much if it is assumed that the air ﬂow as stratiﬁed ﬂow 4 of many concentration layers (thus. In fact. this assumption will not be appropriate when the air is stratiﬁed because of large body forces. For example. or a large acceleration. it is assumed that air is made of only gases. It also provides an opportunity to bring the latest advances in the ﬁelds since this author does not need to “sell” the book to a publisher or convince a “committee. the homogeneous assumption is enough and suitable. Adopting this assumption might lead to a larger error.4. This fact is due to the shear number of the downloaded Potto books. . While the diﬀerence of the concentration between the top to button is insigniﬁcant. nonetheless it exists. many proprieties of air are calculated as if the air is made of well mixed gases of Nitrogen and Oxygen.9. Diﬀerent ﬁelds of multi phase ﬂow. 9. KIND OF MULTI-PHASE FLOW Gas Liquid Liquid Solid Gas Liquid Liquid Liquid Liquid 239 Gas Solid soid Liquid Solid Solid Soid Fig. there are situations when air ﬂow has to be considered as multiphase ﬂow and this eﬀect has to be taken into account.1. The discussion in the previous chapters is only as approximation when multiphase can be “reduced” into a single phase ﬂow.” 4 Diﬀerent concentration of oxygen as a function of the height. Air is not a pure material but a mixture of many gases.4 Kind of Multi-Phase Flow All the ﬂows are a form of multiphase ﬂow. In our calculation. Practically for many cases.000 in about two and half years.

plasma and river ﬂow with live creatures (small organisms to large ﬁsh) ﬂow of ice berg. The same can be said for gas–gas ﬂow. 9. Open Channel ﬂow is. The category of liquid–gas should be really viewed as the extreme case of liquid-liquid where the density ratio is extremely large. 360070 issued to George Westinghouse. There are many more categories. living bodies (bio–ﬂuids). The body inhales solid particle with breathing air. and industries. When the . This notion eliminates many other ﬂow categories that can and should be included in multiphase ﬂow. it is more common to categorize the ﬂow by the distinct phases that ﬂow in the tube.. Thus. One way to categorize the multiphase is by the materials ﬂows. although important. However. liquid–gas and solid–liquid–gas ﬂow. Many industries are involved with this ﬂow category such as dust collection. March 29. the cleanness of air can reduce the speed of sound. The multiphase is an important part of many processes. for example. is only an extreme case of liquid-gas ﬂow and is a sub category of the multiphase ﬂow.5 Classiﬁcation of Liquid-Liquid Flow Regimes The general discussion on liquid–liquid will be provided and the gas–liquid ﬂow will be discussed as a special case. Jr. In a four (4) miles long train. the breaks would started to work after about 20 seconds in the last wagon. For example. the density is a strong function of the temperature and pressure. The liquid–solid. spray casting. the ﬂow of oil and water in one pipe is a multiphase ﬂow. Flow of air is actually the ﬂow of several light liquids (gases). Generally. hydraulic with two or more kind of liquids. The multiphase can be found in nature. they can be solid–liquid. a 10% change of the speed of sound due to dust particles in air could reduce the stopping time by 2 seconds (50 meter diﬀerence in stopping) and can cause an accident. solid–gas. MULTI–PHASE FLOW of clean room is a proof that air contains small particles. the breaks in long trains were activated by reduction of the compressed line (a patent no. Many natural phenomenon are multiphase ﬂow. there are two possibilities for two diﬀerent materials to ﬂow (it is also correct for solid–liquid and any other combination). the cleanness of the air or the fact that air is a mixture is ignored. and avalanches. Gas–solid can be found in sand storms. Since there are three phases.240 CHAPTER 9. The engineering accuracy is enough to totally ignore it. rain. Liquid–liquid ﬂow is probably the most common ﬂow in the nature. and river ﬂow. 1887). The “average” viscosity is meaningless since in many cases the water follows around the oil. This category should include any distinction of phase/material. The materials can ﬂow in the same direction and it is referred as co–current ﬂow. In almost all situations. solid propellant rocket. For example. This ﬂow is used by engineers to reduce the cost of moving crude oil through a long pipes system. sand and grain (which are “solids”) ﬂow with rocks and is referred to solid–solid ﬂow. in nature can be blood ﬂow. Yet. This ﬂow also appears in any industrial process that are involved in solidiﬁcation (for example die casting) and in moving solid particles. mud ﬂow etc. for an example. paint spray. there are situations where cleanness of the air can aﬀect the ﬂow. In the past. ﬂuidized bed. Many industrial process also include liquid-liquid such as painting. For the gas. The water ﬂow is the source of the friction.

The geometries (even the boundaries) of open channel ﬂow are very diverse. -9. 9. the limits between the ﬂow regimes are considerably diﬀerent. . The ﬂow in inclined angle (that not covered by the word “near”) exhibits ﬂow regimes not much diﬀerent from the other two. Yet. up or down. and what ever between them. In general.2. It is common to diﬀerentiate between the vertical (and near vertical) and horizontal (and near horizontal). the ﬂow is referred to as open channel ﬂow. Open channel ﬂow appears in many nature (river) as well in industrial process such as the die casting process where liquid metal is injected into a cylinder (tube) shape. it is referred as counter–current. For example. The vertical conﬁguration has two cases. When the ﬂow rate of the lighter liquid is almost zero.1 Co–Current Flow In Co–Current ﬂow. the co-current is the more common.1.1 Horizontal Flow The typical regimes for horizontal ﬂow are stratiﬁed ﬂow (open channel ﬂow. The other characteristic that is diﬀerent between the gas ﬂow and the liquid ﬂow is the variation of the density. dispersed Heavy Liquid bubble ﬂow. Stratiﬁed ﬂow in horizontal tubes when bottom and lighter liquid ﬂows on the the liquids ﬂow is very slow. For low velocity (low ﬂow rate) of the two liquids. 5 top as depicted in Figure 9. For example. horizontal. The channel ﬂow will be discussed in a greater detail in Open Channel Flow chapter. water and air ﬂow as oppose to water and oil ﬂow. two liquids can have three main categories: vertical. and annular ﬂow.5. Additionally. Thus. the ﬂow of gas–liquid can have several ﬂow regimes in one situation while the ﬂow of liquid–liquid will (probably) have only one ﬂow regime. This issue of incline ﬂow will not be covered in this chapter. plug ﬂow. This deﬁnition (open channel ﬂow) continues for small amount of lighter liquid as long as the heavier ﬂow can be calculated as open channel ﬂow (ignoring the lighter liquid). The ﬂow regimes are referred to the arrangement of the ﬂuids. The main diﬀerence between the liquid–liquid ﬂow to gas-liquid ﬂow is that gas density is extremely lighter than the liquid density. a reduction of the pressure by half will double the gas volumetric ﬂow rate while the change in the liquid is negligible. the counter–current ﬂow must have special conﬁgurations of long length of ﬂow. 9.5.2.9. There is no exact meaning to the word “near vertical” or “near horizontal” and there is no consensus on the limiting angles (not to mention to have limits as a function with any parameter that determine the limiting angle).5. This kind of ﬂow regime is referred to as horizontal ﬂow. the heavy liquid ﬂows on the Fig. the counter–current ﬂow has a limited length window of possibility in a vertical ﬂow in conduits with the exception of magnetohydrodynamics. Light Liquid and non open channel ﬂow). 5 With the exception of the extremely smaller diameter where Rayleigh–Taylor instability is an important issue. Generally. CLASSIFICATION OF LIQUID-LIQUID FLOW REGIMES 241 materials ﬂow in the opposite direction.

All liquids are compressible Heavy Liquid to some degree.242 CHAPTER 9. Light Liquid Light Liquid Some referred to this regime as wavy stratiﬁed ﬂow Heavy Liquid Heavy Liquid but this deﬁnition is not accepted by all as a category by itself. as a stratiﬁed ﬂow will turned into a slug ﬂow or stratiﬁed wavy7 ﬂow after a certain distance depends on the heavy ﬂow rate (if 6 The 7 Well. At this stage. the ﬂow pattern is referred to as slug ﬂow or plug ﬂow. choking occurs relatively Fig. the ﬂow that starts liquids ﬂow is faster. These plugs are separated by large “chunks” that almost ﬁll the entire tube. Plug ﬂow in horizontal tubes when the closer/sooner.4. This kind of ﬂow regime is referred to as annular ﬂow. The plugs are ﬂowing in a succession (see Figure 9. when the lighter liquid ﬂow increases. liquid ﬂow rate. If the heavier liquid ﬂow rate is larger6 than the distance. For liquid which the density is a strong and primary function of the pressure. the friction between the phases increase. The superﬁcial velocity is referred to as the velocity that any phase will have if the other phase was not exist. If the heavier ﬂow rate is small.3.4). This friction is one of the cause for the instability which manifested itself as waves and changing the surface from straight line to a diﬀerent conﬁguration (see Figure 9. Choking occurs in compressible Light Liquid ﬂow when the ﬂow rate is above a certain point. . The wave shape is created to keep the gas and the liquid velocity equal and at the same time to have shear stress to be balance by surface tension. thus it is arbitrary deﬁnition. all the two phase ﬂow are categorized by wavy ﬂow which will proven later.3). At some point. The conﬁguration of the cross section not only depend on the surface tension. the possibility to go through slug ﬂow regime depends on if there is enough liquid ﬂow rate. Plug ﬂow is characterized by regions of lighter liquid ﬁlled with drops of the heavier liquid with Plug (or Slug) of the heavier liquid (with bubble of the lighter liquid). and other physical properties of the ﬂuids but also on the material of the conduit. liquid level is higher. As the lighter liquid velocity increases two things can happen (1) wave size increase and (2) the shape of cross section continue to deform. Fig. The slug ﬂow cannot be assumed to be as homogeneous ﬂow nor it can exhibit some average viscosity. for the wave to reach the conduit crown is smaller. The “average” viscosity depends on the ﬂow and thus making it as insigniﬁcant way to do the calculations. The pressure drop of this kind of regime is signiﬁcantly larger than the stratiﬁed ﬂow. Kind of Stratiﬁed ﬂow in There are two paths that can occur on the heavier horizontal tubes. Further increase of the lighter liquid ﬂow rate move the ﬂow regime into annular ﬂow. then the wave cannot reach to the crown and the shape is deformed to the point that all the heavier liquid is around the periphery. all the ﬂow is wavy. MULTI–PHASE FLOW As the lighter liquid (or the gas phase) ﬂow rate increases (superﬁcial velocity). Thus. the heavier liquid wave reaches to the crown of the pipe. -9. In fact. Thus. -9.

5. As in many ﬂuid mechanics and magnetic ﬁelds. The buoyancy . e. in normal gravitation. The reverse way is referred to the process where the starting point is high ﬂow rate and the ﬂow rate is decreasing.5. Hence. with the gravity or against it. The choking can occur before the annular ﬂow regime is obtained depending on the velocity and compressibility of the lighter liquid.2 Vertical Flow The vertical ﬂow has two possibilities. 9. This maximum is known as double choking phenomenon. the vertical ﬂow against the gravity is more common used. In engineering application.5. Modiﬁed Mandhane map for ﬂow regime in horizontal tubes. ﬂow that had slug ﬂow in its path can be returned as stratiﬁed wavy ﬂow. Flow that is under small angle from the horizontal will be similar to the horizontal ﬂow. etc. Qualitatively. CLASSIFICATION OF LIQUID-LIQUID FLOW REGIMES 243 this category is accepted). As many things in multiphase. The angle decreases with the length of the pipe. The results of the above discussion are depicted in Figure 9. weak to strong surface tension eﬀects (air/water in “normal” gravity). the ﬂow become annular or the ﬂow will choke. the return path is not move the exact same way. For example. the “near” angle depends on the length of the pipe.5. After a certain distance. There is even a possibility to return on diﬀerent ﬂow regime.” Dispersed Bubble Liquid Superficial Velocity Elongated Bubble Slug Flow Annular Flow Stratified Flow Wavy Stratified Open Channel Flow Gas Superficial Velocity Fig.1. this map is only characteristics of the “normal” conditions. However. other parameters can aﬀect the “near. This phenomenon is refer to as hysteresis.9. liquid–liquid ﬂow has a maximum combined of the ﬂow rate (both phases). there is no consensus how far is the “near” means.g. There is a difference between ﬂowing with the gravity and ﬂowing against the gravity. as in compressible ﬂow. Besides the length. -9.

When many bubbles collide. Flow Against Gravity For vertical ﬂow against gravity. The ﬂow is less stable as more turbulent ﬂow and several “super slug” or churn ﬂow appears in more chaotic way. . they create a large bubble and the ﬂow is referred to as slug ﬂow or plug ﬂow (see Figure 9. Notice. the diﬀerent mechanism in creating the plug ﬂow in horizontal ﬂow compared to the vertical ﬂow. Any further increase transforms the outer liquid layer into bubbles in the inner liquid. the ﬂow cannot start as a stratiﬁed ﬂow. Flow of near vertical against the gravity in two–phase does not deviate from vertical. Gas and liquid in Flow in verstical tube against the gravity. Thus. For the ﬂow against gravity. After additional increase of “super slug” . there are diﬀerent ﬂow regimes for these two situations. MULTI–PHASE FLOW Bubble Flow Slug or Plug Flow Churn Flow Annular Flow Dispersed Flow Fig. Again. is acting in two diﬀerent directions for these two ﬂow regimes. The increase of the lighter liquid ﬂow rate will increase the number of bubbles until some bubbles start to collide. the lighter liquid has a buoyancy that acts as an “extra force” to move it faster and this eﬀect is opposite for the heavier liquid. The opposite is for the ﬂow with gravity. The heavier liquid has to occupy almost the entire cross section before it can ﬂow because of the gravity forces. the ﬂow regime is referred as elongated bubble ﬂow. it can be noted the diﬀerence in the mechanism that create annular ﬂow for vertical and horizontal ﬂow.6. -9.6). Further increase of lighter liquid ﬂow rate will increase the slug size as more bubbles collide to create “super slug”. see Figure 9. The choking can occur at any point depends on the ﬂuids and temperature and pressure.244 CHAPTER 9. the ﬂow starts as a bubble ﬂow. all these “elongated slug” unite to become an annular ﬂow.6. Thus. The main reason that causes the diﬀerence is that the heavier liquid is more dominated by gravity (body forces) while the lighter liquid is dominated by the pressure driving forces.

the points where these transitions occur are diﬀerent from the horizontal ﬂow. -9. this ﬂow can starts with stratiﬁed ﬂow. A dimensional vertical ﬂow map considered as the “open channel ﬂow” of under very low gravity against the gravity. the ﬂow continues in a stratiﬁed conﬁguration. MULTI–PHASE FLOW VARIABLES DEFINITIONS 9. it is presented in a dimension form to explain the trends (see Figure 9. Figure 9. The ﬂow starts as disTrickling persed bubble (some call it as “gas conFlow Spray or tinuous”) because the gas phase occupies Mist Flow most of column. The common map is based on dimensionless parameters. for the convenience of the terms “gas and liquid” will be used to signify the lighter and heavier liquid. In the literature.7. it must be possible to construct a model that connects this conﬁguration with the stratiﬁed ﬂow where the transitions will be dependent on the angle of inclinations. Here. both will be united in this discussion. Flow With The Gravity As opposed to the ﬂow against gravity. A good example for this ﬂow regime is a water fall.7).9. the vertical conﬁguration.6 Multi–Phase Flow Variables Deﬁnitions Since the gas–liquid system is a speciﬁc case of the liquid–liquid system. In the medium range of the ﬂow rate of gas and liquid. While this author is not aware of an actual model. The transitions between the ﬂow regimes is similar to stratiﬁed ﬂow. all the gas phase change into tiny drops of liquid and this kind of regime referred to as mist ﬂow. However.6. and Weber numbers. The initial part for this ﬂow is more signiﬁcant. the regime liquid ﬁlls the entire void and the gas is in small bubble and this ﬂow referred to as bubbly ﬂow. When the ﬂow rate of the gas increases further. respectively. At a higher rate of liquid ﬂow and a low ﬂow rate of gas. Since the heavy liquid can be supplied from the “wrong” point/side. the initial part has a larger section compared to the ﬂow against the gravity ﬂow.3 Vertical Flow Under Micro Gravity 245 The above discussion mostly explained the Dispersed Dispersed ﬂow in a vertical conﬁguration when the Bubble Bubble surface tension can be neglected. The ordinate is a combination of ﬂow rate ratio and density ratio. Pulsing For example. The liquid–liquid (also .1.Reynolds. The abscissa is a function of combination of Froude . The liquid ﬂows through Gas Flow Rage a trickle or channeled ﬂow that only partially wets part of the tube. After the ﬂow has settled.7 presented in dimensionless coordinates. In cases where the surface tension is very important. the liquid becomes more turbulent and some parts enter into the gas phase as drops. Liquid Flow Rate 9. The interaction between the phases is minimal and can be Fig. there is pulse ﬂow in which liquid is moving in frequent pulses.5. As the gas ﬂow increases. out in space between gas and liquid (large density diﬀerence) the situaPulsing & Bubbling tion is diﬀerent. However.

7) . some variables are deﬁned so that the ﬂow can be described as a one-dimensional ﬂow. The gas mass velocity is GG = mG ˙ A (9. The liquid mass velocity is GL = The mass ﬂow of the tube is then G= m ˙ A (9.6.6) GG = UsG ρG (9.4) mL ˙ A (9.246 CHAPTER 9. the only serious missing point in this discussion is the change of the ﬂow along the distance of the tube. To simplify the descriptions of the problem and yet to retain the important features of the ﬂow. MULTI–PHASE FLOW gas–liquid) ﬂow is an extremely complex three–dimensional transient problem since the ﬂow conditions in a pipe may vary along its length. The volumetric ﬂow rate is not constant (since the density is not constant) along the ﬂow rate and it is deﬁned as QG = and for the liquid QL = GL ρL (9. the volumetric ﬂow rate can be considered as constant. and with time. over its cross section.2) Where A is the entire area of the tube. Perhaps. 9.1) It is common to deﬁne the mass velocity instead of the regular velocity because the “regular” velocity changes along the length of the pipe. It has to be noted that this mass velocity does not exist in reality.3) It has to be emphasized that this mass velocity is the actual velocity.1 Multi–Phase Averaged Variables Deﬁnitions The total mass ﬂow rate through the tube is the sum of the mass ﬂow rates of the two phases m = mG + mL ˙ ˙ ˙ (9.5) For liquid with very high bulk modulus (almost constant density). The total volumetric volume vary along the tube length and is Q = QL + QG (9. This method is the most common and important to analyze two-phase ﬂow pressure drop and other parameters.

The liquid fraction or liquid holdup is LH = 1 − α = AL A (9. Thus. It can be noticed that Um is not constant along the tube.8) In a similar fashion.14) Slip ratio is usually greater than unity. it can be noted that the slip velocity is not constant along the tube.13) Where Um is the averaged velocity. Thus. the value of (1 − X) is referred to as the “wetness fraction. MULTI–PHASE FLOW VARIABLES DEFINITIONS 247 Ratio of the gas ﬂow rate to the total ﬂow rate is called the ’quality’ or the “dryness fraction” and is given by X= GG mG ˙ = m ˙ G (9.9) This fraction is vary along tube length since the gas density is not constant along the tube length. The actual velocities depend on the other phase since the actual cross section the phase ﬂows is dependent on the other phase. The gas superﬁcial velocity is therefore deﬁned as UsG = The liquid superﬁcial velocity is UsL = GL (1 − X) m ˙ = = QL ρL ρL A (9. LH is not constant for the same reasons the void fraction is not constant. .” The last two factions remain constant along the tube length as long the gas and liquid masses remain constant. The average superﬁcial velocity of the gas and liquid are diﬀerent.12) GG Xm ˙ = = QG ρG ρG A (9. the ratio of these velocities is referred to as the slip velocity and is deﬁned as the following SLP = UG UL (9. a superﬁcial velocity is commonly deﬁned in which if only one phase is using the entire tube. Also.9.11) Since UsL = QL and similarly for the gas then Um = UsG + UsL (9.10) It must be noted that Liquid holdup. The ratio of the gas ﬂow cross sectional area to the total cross sectional area is referred as the void fraction and deﬁned as α= AG A (9.6.

17) Equation (9.16) results in mG ˙ mL ˙ ρaverage = ˙ X m + (1 − X) m ˙ ˙ X m + (1 − X) m ˙ = ˙ X m (1 − X) m ˙ QG + QL + ρG ρL QG QL (9. The average density of the material ﬂowing in the tube can be evaluated by looking at the deﬁnition of density.20) becomes X= ρG α ρL (1 − α) + ρG α (9. Substituting equations (9.15) This density represents the density taken at the “frozen” cross section (assume the volume is the cross section times inﬁnitesimal thickness of dx).1) and (9.21) .248 CHAPTER 9. the mixture density is deﬁned as ρm = α ρG + (1 − α) ρL (9.16) Where Q is the volumetric ﬂow rate.17) can be simpliﬁed by canceling the m and noticing the (1−X)+X = 1 ˙ to become + (1−X) ρL The average speciﬁc volume of the ﬂow is then X ρG ρaverage = 1 (9.7) into equation (9. thus equation (9. for the ﬂowing material it is ρ= m ˙ Q (9.18) vaverage = 1 ρaverage = X (1 − X) + = X vG + (1 − X) vL ρG ρL (9. MULTI–PHASE FLOW For the same velocity of phases (SLP = 1).20) mG + mL ˙ ˙ ρL UL A(1 − α) +ρG UG A α ρL UL (1 − α) + ρG UG α AL If the slip is one SLP = 1.19) The relationship between X and α is AG X= mG ˙ ρG UG A α ρG UG α = = (9. The density of any material is deﬁned as ρ = m/V and thus.

25) The energy equation can be approximated as dw d dq − =m ˙ dx dx dx hm + Um 2 + g x sin θ 2 (9. it is worthwhile to appreciate the complexity of the ﬂow.13)) is Um = QL + QG = UsL + UsG = Um A (9. Furthermore. it was assumed that the diﬀerent ﬂow regimes can be neglected at least for the pressure loss (not correct for the heat transfer). The average velocity (see also equation (9. this assumption has to be broken.26) . the simplest is to used it for approximation.24) or modifying equation (9.23) is correct? Solution Under construction End Solution The governing momentum equation can be approximated as m ˙ dUm dP = −A − S τw − A ρm g sin θ dx dx (9. The single phase was studied earlier in this book and there is a considerable amount of information about it.7 Homogeneous Models Before discussing the homogeneous models. HOMOGENEOUS MODELS 249 9.24) as − dP S m dUm ˙ = − τw − + ρm g sin θ dx A A dx (9.23) Example 9. the diﬀerent ﬂow regimes are examples of typical conﬁguration of segments of continuous ﬂow.7. Thus. these segments are not deﬁned but results of the conditions imposed on the ﬂow. For the construction of ﬂuid basic equations. it was assumed that the ﬂow is continuous.9. Initially. In fact.22) It can be noted that the continuity equation is satisﬁed as m = ρm Um A ˙ (9.1: Under what conditions equation (9. Now. and the ﬂow is continuous only in many chunks (small segments).

9.079 and n = 0. for small distances (dx) and some situations. from equation (9. the gravitation pressure loss reduce the pressure and thus the density must change and hence. These losses are non-linear and depend on each other. the total pressure loss can be written as f riction acceleration gravity dP dP dP dP + + = (9. the experimental data that was developed and measured for single ﬂow is used.25.27) dx dx f dx a dx g Every part of the total pressure loss will be discussed in the following section. acceleration. acceleration must occur. The friction factor is obtained by using the correlation f =C ρm Um D µm −n (9. Yet.1.7.28) Where S is the perimeter of the ﬂuid. For turbulent ﬂow C = 0. .29) The wall shear stress can be estimated by τw = f ρm Um 2 2 (9. For calculating the frictional pressure loss in the pipe is − dP dx = f 4 τw D (9. and body force(gravitation). However.1 Pressure Loss Components In a tube ﬂowing upward in incline angle θ.1 Friction Pressure Loss The frictional pressure loss for a conduit can be calculated as − dP dx = f S τw A (9. since there isn’t anything better. this dependency can be neglected.30) The friction factor is measured for a single phase ﬂow where the average velocity is directly related to the wall shear stress. There is not available experimental data for the relationship of the averaged velocity of the two (or more) phases and wall shear stress. this friction factor was not measured for the “averaged” viscosity of the two phase ﬂow. For example. MULTI–PHASE FLOW 9. In that case.250 CHAPTER 9. the pressure loss is aﬀected by friction loss.31) Where C and n are constants which depend on the ﬂow regimes (turbulent or laminar ﬂow).7.25). For laminar ﬂow C = 16 and n = 1. In fact.

Duckler suggest the following µm = µG QG µL QL + QG + QL QG + QL (9. In second case is where the mass ﬂow rates of gas and liquid is constant in which the derivative of X is zero. the friction loss can be estimated.37) a There are several special cases.18) average viscosity as µaverage = 1 X µG (1−X) µL + (9. .36) Or in an explicit way equation (9. For the last point.7. 9. Equation (9.35) The acceleration pressure loss (can be positive or negative) results from change of density and the change of cross section. The ﬁrst case where the cross section is constant.1. HOMOGENEOUS MODELS 251 There are several suggestions for the average viscosity.33) Or simply make the average viscosity depends on the mass fraction as µm = X µG + (1 − X) µL Using this formula.9. For example. the private case is where densities are constant for both phases.2 Acceleration Pressure Loss (9.7.36) becomes pressure loss due to pressure loss due to density change area change 2 =m ˙ 1 1 d 1 dA + 2 dx A dx ρm ρm A − dP dx (9.34) The acceleration pressure loss can be estimated by − dP dx =m ˙ a dUm dx (9. The third special case is for constant density of one phase only. dX/ dx = 0.32) Duckler linear formula does not provide always good approximation and Cichilli suggest similar to equation (9. dρL / dx = 0.35) can be written as − dP dx =m ˙ a d dx m ˙ A ρm (9. dA/ dx = 0.

42) SL f TP Where the SL denotes the pressure loss for the single liquid phase.. 8 This method was considered a military secret.4 Total Pressure Loss The total pressure between two points.39) and therefore f riction acceleration gravity ∆Pab = ∆Pab f + ∆Pab a + ∆Pab g (9. The density in equation (9. 9. (a and b) can be calculated with integration as b ∆Pab = a dP dx dx (9.7. Thus.3 Gravity Pressure Loss CHAPTER 9.38) is the density without the “movement” (the “static” density).41) SG f TP Where the T P denotes the two phases and SG denotes the pressure loss for the single gas phase.1. there are two parameters as shown below.7.2 Lockhart Martinelli Model The second method is by assumption that every phase ﬂow separately One such popular model by Lockhart and Martinelli8 .40) 9.252 9. Equivalent deﬁnition for the liquid side is φL = dP dx dP dx (9. MULTI–PHASE FLOW Gravity was discussed in Chapter 4 and is dP dx = g ρm sin θ g (9. Lockhart and Martinelli built model based on the assumption that the separated pressure loss are independent from each other. Lockhart Martinelli parameters are deﬁned as the ratio of the pressure loss of two phases and pressure of a single phase.7.1. Taitle . φG = dP dx dP dx (9.38) The density change during the ﬂow can be represented as a function of density. private communication with Y.

The ﬂow with the gravity and lighter density solid particles.46) Simpliﬁed model is when there is no interaction between the two phases. In solid–liquid.43) SG f SL where Ξ is Martinelli parameter. SOLID–LIQUID FLOW 253 The ratio of the pressure loss for a single liquid phase and the pressure loss for a single gas phase is Ξ= dP dx dP dx (9. cubic. To insert the Diagram. dP dx = SG dP dx (9.45) For the gas phase. It is also assumed that the “liquids” density does not change signiﬁcantly and it is far from the choking point. etc). In that case there are four possibilities for vertical ﬂow: 1. The ﬂow against the gravity and lighter density solid particles. For example. 9. it is assumed that the surface tension is insigniﬁcant compared to the gravity forces.9. 3.8. Consider the case where the solid is heavier than the liquid phase.8 Solid–Liquid Flow Solid–liquid system is simpler to analyze than the liquid-liquid system. and geometry. The ﬂow with the gravity and heavier density solid particles.44) SL The pressure loss for the liquid phase is dP dx = L 2 fL UL 2 ρl DL (9. the eﬀect of the surface tension are very minimal and can be ignored.” Therefor. in this discussion. 2.there will be a discussion about diﬀerent particle size and diﬀerent geometry (round. distribution. the pressure loss is dP dx = G 2 fG UG 2 ρl DG (9. Thus. . analysis of small coal particles in water is diﬀerent from large coal particles in water. The uniformity is categorizing the particle sizes. It is assumed that the pressure loss for both phases are equal. The density of the solid can be above or below the liquid. Diﬀerent combination of solid particle creates diﬀerent “liquid. The word “solid” is not really mean solid but a combination of many solid particles.

1 Solid Particles with Heavier Density ρS > ρL Solid–liquid ﬂow has several combination ﬂow regimes. the liquid cannot carry the solid particles because there is not enough resistance to lift up the solid particles. MULTI–PHASE FLOW 4. Re.48) Inserting equating (9. When the liquid velocity is very small. The ﬁrst set is similar because the solid particles are moving faster than the liquid velocity and vice versa for the second set (slower than the liquid). A particle in a middle of the vertical liquid ﬂow experience several forces. The force balance of spherical particle in ﬁeld viscous ﬂuid (creeping ﬂow) is gravity and buoyancy forces π D g (ρS − ρL ) 6 3 drag forces = CD ∞ π D2 ρL UL 2 8 (9. The ﬂow against the gravity and heavier density solid particles. However. 9. The discussion here is about the last case (4) because very little is known about the other cases. All these possibilities are diﬀerent.44 (9.8. it can be approximated for several regimes. The Reynolds number deﬁned as Re = UL D ρL µL (9.51) 24 Re (9. The ﬁrst regime is for Re < 1 where Stokes’ Law can be approximated as CD ∞ = In transitional region 1 < Re < 1000 CD ∞ = 24 Re 1+ 1 Re2/3 6 (9.49) Equation (9. the Newton’s Law region.49) relates the liquid velocity that needed to maintain the particle “ﬂoating” to the liquid and particles properties. possibility. is nearly constant as CD ∞ = 0.47) become CD ∞ (UL ) f (Re) UL 2 = 4 D g (ρS − ρL ) 3 ρL (9.254 CHAPTER 9.48) into equation (9. 1 and 4 and the second set is 2 and 3. there are two sets of similar characteristics. and D is the equivalent radius of the particles. The drag coeﬃcient. CD ∞ is complicated function of the Reynolds number. However.50) For larger Reynolds numbers.52) . CD ∞ .47) Where CD ∞ is the drag coeﬃcient and is a function of Reynolds number.

In very large range (especially for gas) the choking might be approached. in many cases the middle region is applicable. the only velocity that can be applied is Trasiton the ﬂuid velocity. Additional increase of the ﬂuid velocity will move all the particles and this Fig. partialy Fully the discussion will be focus on the ﬂuid solid fluidized velocity. the particle will sink into the liquid. Additional increase in the ﬂuid velocity causes large turbulence and the ordinary domes are replaced by churn type ﬂow or large bubbles that are almost empty of the solid particles. When the velocity of liquid is higher than the minimum velocity many particles will be ﬂoating.9). For the ﬁrst region. This regimes is referred to as Pneumatic conveying (see Figure 9. Consequently. Minimum velocity is the velocity when the particle is “ﬂoating”. US|avarge .53) When the subscript α is indicating the void. the minimum velocity is a range of velocity rather than a sharp transition point. When there are more than one particle in the cross section. the solid particles Packed can be supplied at diﬀerent rate. Thus. When the velocity is lower. additional increase create “tunnels” of empty almost from solid particles. The simplest assumption that the change of the cross section of the ﬂuid create a parameter that multiply the single particle as CD ∞ |α = CD ∞ f (α) (9.8.8. 9 It be wonderful if ﬂow was in the last range? The critical velocity could be found immediately. SOLID–LIQUID FLOW 255 In most cases of solid-liquid system. In the literature there are many functions for various conditions. Yet. the particle will drift with the liquid. the case of liquid. Further increase of the ﬂuid ﬂow increases the empty spots to the whole ﬂow. The terminal velocity that left the solid is referred to as fully ﬂuidized bed. It has to remember that not all the particle are uniform in size or shape. the velocity is small to lift the particle unless the density diﬀerence is very small (that very small force can lift the particles). then the actual velocity that every particle experience depends on the void fraction. the Reynolds number is in the second range9 .9. Thus. So far the discussion was about single particle. For small gas/liquid velocity. Thus. Increasing the ﬂuid velocity beyond a minimum will move the parti∆Ptube cles and it is referred to as mix ﬂuidized bed. the sparse solid particles are dispersed all over. particles flow the particles are what some call ﬁxed ﬂuidized bed. As the solid particles are not pushed by a pump but moved by the forces the ﬂuid applies to them. For particles. This slug ﬂow is when slug shape (domes) are almost empty of the solid particle. the function f (α) is not a linear function. If the velocity is larger. -9. For the case of gas. In that case. further increase will create a slug ﬂow.

8. it must be pointed out that even in solid–gas. Nevertheless. 9. the ﬂuid density can be higher than the solid (especially with micro gravity). There was very little investigations and known about the solid–liquid ﬂowing down (with the gravity). The ﬂow patterns in solid-liquid ﬂow. The buoyancy is accelerating the particle . There is no known ﬂow map for this kind of ﬂow that this author is aware of. Thus. the velocity √ of gas is limited when reaching the Mach somewhere between 1/ k and 1 since the gas will be choked (neglecting the double choking phenomenon).9. The ﬂow can have slug ﬂow but more likely will be in fast Fluidization regime. -9. there is very little knowledge about the solid–liquid when the solid density is smaller than the liquid density. The speed of sound of the liquid does not change much. The forces that act on the spherical particle are the buoyancy force and drag force. The issue of minimum terminal velocity is not exist and therefor there is no ﬁxed or mixed ﬂuidized bed. the speed of sound is reduced dramatically with increase of the solid particles concentration (further reading Fundamentals of Compressible Flow” chapter on Fanno Flow by this author is recommended).256 CHAPTER 9. One of the main diﬀerence between the liquid and gas ﬂow in this category is the speed of sound.2 Solid With Lighter Density ρS < ρ and With Gravity This situation is minimal and very few cases exist. Hence. several conclusions and/or expectations can be drawn. However. The ﬂow is fully ﬂuidized for any liquid ﬂow rate. Furthermore. MULTI–PHASE FLOW Fixed Bed Mixed Bed Slug or Plug Flow Turbulent Regimes Fast Fluidization Pneumatic Conveying Fig. the length of conduit is very limited. this limitation does not (eﬀectively) exist for most cases of solid–liquid ﬂow. In the gas phase. Hence.

for Pulse Flow Inpossible long tubes. the ﬂow will be in a pulse regime. The aﬀective body force “seems” by the particles can be in some cases larger than the gravity. When the liquid velocity increase and became turbulent. for large velocity of the ﬂuid it can be observed that UL /US → 1. the or Dripping Flow pressure diﬀerence in one phase can be f (D/L. In only certain conﬁgurations of the inﬁnite long pipes the counter–current ﬂow can exist. Thus. The ﬂow regimes will be similar but the transition will be in diﬀerent points. for a small ﬂuid velocity the velocity ratio is very large.54) 6 8 From equation 9.54. -9. Initially the solid particles will be carried by the liquid to the top. The pressure diﬀerence in the interface must Fig. In short tube. opening a can of milk or juice. In most cases. But in most cases. The two phase regimes “occurs” mainly in entrance to the cavity. for example. Annular Extented Flow In that case. this author have not seen any evidence that show the annular ﬂow does not appear in solid–liquid ﬂow.9. COUNTER–CURRENT FLOW and drag force are reducing the speed as 2 257 π D3 g(ρS − ρL ) CD ∞ π D2 ρL (US − UL ) = (9. The solid–liquid horizontal ﬂow has some similarity to horizontal gas–liquid ﬂow. However. Most people know that two holes are needed to empty the can easily and continuously. Hence. Flow The inertia components of the ﬂow. it can observed that increase of the liquid velocity will increase the solid particle velocity at the same amount. the pressure diﬀerence and Open Channel gravity (body forces) dominates the ﬂow.9 Counter–Current Flow This discussion will be only on liquid–liquid systems (which also includes liquid-gas systems). when cavity is ﬁlled or emptied with a liquid. Typically if only one hole is opened on the top of the can. The counter-current ﬂow occurs. For example.9. the possibility to have counter–current ﬂow is limited to having short length of tubes. be ﬁnite. physical properties) positive while the pressure diﬀerence in the other phase can be negative. This kind of ﬂow is probably the most common to be realized by the masses. Liquid Body Foreces . the liquid will ﬂow in pulse regime. For example.10. the heavy phase (liquid) is pushed by the gravity and lighter phase (gas) is driven by the pressure diﬀerence. some of the particles enter into the liquid core. UL /US → 0. 9. Further increase of the liquid velocity appear as somewhat similar to slug ﬂow. Otherwise. Counter–ﬂow in vertical tubes map. cannot compensate for the Flow pressure gradient. However. the counter–current ﬂow can have opposite pressure gradient for short conduit.

liquid ﬂows through one hole and the air through the second hole and the ﬂow will be continuous. in some cases. Even though the solid–gas ratio is smaller. 10 Caution! this statement should be considered as “so far found”.11 depicts emptying of can ﬁlled with liquid. Fig. Then.258 CHAPTER 9. Initially. Picture of Counter-current ﬂow in liquid–gas and solid–gas conﬁgurations.11. elongated pulse ﬂow was observed but measured. -9. the phases ﬂow turns into diﬀerent direction (see Figure 9. It also can be noticed that if there is one hole (oriﬁce) and a long and narrow tube. . Counter–current ﬂow in a can (the left ﬁgure) has only one hole thus pulse ﬂow and a ﬂow with two holes (right picture). it can be noticed that the solid–gas is faster than the liquid–gas ﬂow. The name pulse ﬂow is used to signify that the ﬂow is ﬂowing in pulses that occurs in a certain frequency.12). In this ﬂow regime. water solution.12. or small wood particles) by rotating the container. MULTI–PHASE FLOW Fig. Then the pressure in the can is reduced compared to the outside and some lighter liquid (gas)entered into the can. There are three ﬂow regimes10 that have been observed. For example. There must be other ﬂow regimes that were not observed or deﬁned. The air is “attempting” to enter the cavity to ﬁll the vacuum created thus forcing pulse ﬂow. The ﬁrst ﬂow pattern is pulse ﬂow regime. This ﬁeld hasn’t been well explored.). This is opposed to counter–current solid–gas ﬂow when almost no pulse was observed. the heavy liquid is leaving the can. If there are two holes. Figure 9. There are more things to be examined and to be studied. The container is made of two compartments. the liquid will stay in the cavity (neglecting other phenomena such as dripping ﬂow. -9. the pressure in the can increase. due to the gravity. The upper compartment is ﬁlled with the heavy phase (liquid.

A ﬂow in a very narrow tube with heavy ﬂuid above the lighter ﬂuid should be considered as a separate issue. Horizontal ﬂow is diﬀerent from vertical ﬂow from the stability issues.9. the pulsing ﬂow will start and larger diameter will increase the ﬂow and turn the ﬂow into annular ﬂow. the discussion was focused on the vertical tubes. if the can was on the sun (ignoring the heat transfer issue). The driving force is the second parameter which eﬀects the ﬂow existence. The ratio of the diameter to the Fig.1 Horizontal Counter–Current Flow Up to this point. -9. The cycle duration can be replaced by frequency. 9. the ﬂow regime in the can moves from pulse to annular ﬂow. Probably. In very small Steam Flow diameters of tubes the counter–current ﬂow is not possible because of the surface tension (see section 4. Annular Flow in Counter–current ﬂow The other ﬂow regime is annular ﬂow in which the heavier phase is on the periphery of Water the conduit (In the literature. The duration the cycle depends on several factors. The heavy liquid will ﬂow with the body forces (gravity). there is an additional ﬂow regime which is stratiﬁed . Further increase of the body force will move the ﬂow to be in the extended “open channel ﬂow. Consider the can in zero gravity ﬁeld. . In horizontal tubes.7) page 115) it can be considered stable for small diameters. When the driving (body) force is very small. A heavier liquid layer can ﬂow above a lighter liquid. This situation is unstable for large diameter but as in static (see section (4. no counter–current ﬂow possible. At this point. but somehow it contradicts with the experimental evidence. This process continue until almost the liquid is evacuated (some liquid stay due the surface tension). Example of such ﬂow in the nature is water falls in which water ﬂows down and air (wind) ﬂows up. In many situations. no counter–current ﬂow is possible. As opposed to the co–current ﬂow. COUNTER–CURRENT FLOW 259 and some heavy liquid will starts to ﬂow. However.13. Extended open channel ﬂow retains the characteristic of open channel that the lighter liquid (almost) does not eﬀect the heavier liquid ﬂow. ﬂow with gravity or against it.9.” In the vertical co–current ﬂow there are two possibilities. Thus it should be considered as non existent ﬂow. one or more of the assumptions that the analysis based is erroneous). the counter–current ﬂow has no possibility for these two cases.9. length with some combinations of the physical properties (surface tension etc) determines the point where the counter ﬂow can start. The analysis of the frequency is much more complex issue and will not be dealt here. Additional increase of the diameter will change the ﬂow regime into extended open channel ﬂow. the volume ﬂow rate of the two phase is almost equal. The analysis is provided. Flood in vertical pipe.7). there are someFlow one who claims that heavy liquid will be inside).

13. Thus. It is assumed that both ﬂuids are W ξ x y ﬂowing in a laminar regime and steady state.Fig. To analyze this situation consider a two dimensional conduit with a liquid inserted in the left side as depicted in Figure 9. This problem appears in nuclear engineering (or boiler engineering) where there is a need to make sure that liquid (water) inserted into the pipe reaching the heating zone. that is stratiﬁed ﬂow are discussed here. the ﬁre can be too large or/and the water supply failed below a critical value the water turn into steam. mass transfer. However.2 Flooding and Reversal Flow The limits of one kind the counter–current ﬂow regimes. A diagram to explain the As it will be shown later. The liquid velocity at very low gas velocity is constant but not uniform. The stratiﬁed counter ﬂow has a lower pressure loss (for the liquid side). closing the window of this kind of ﬂow. 11 The circular conﬁguration is under construction and will be appeared as a separated article momentarily. it is assumed that the entrance eﬀects L can be neglected. this increase terminates the two phase ﬂow possibility. the ﬂow will be stratiﬁed counter–current ﬂow. horizontal counter–current ﬂow. A simpliﬁed model for this situation is for a two dimensional conﬁguration where the liquid is D ﬂowing down and the gas is ﬂowing up as shown h in Figure 9. The change to pulse ﬂow increases the pressure loss dramatically. The ﬂow map of the horizontal ﬂow is diﬀerent f (D/L.9. wavy interface.15. MULTI–PHASE FLOW Single phase Flow or Dripping Flow When the ﬂow rate of both ﬂuids is very small. Fig. The model can be improved by considering turbulence.14. -9. Additional increase of the gas velocity will bring it to a point where the liquid will ﬂow in a reverse direction and/or disappear (dried out). both conditions cannot ﬂood in a two dimension geometry. When there is no water (in liquid phase). the ﬁre could melt or damage the boiler. Thus. The steam will ﬂow in the opposite direction. In some situations.14. The liquid ﬂow rate. is unknown. coexist. -9. etc11 .15. Further increase of the ﬂow will result in a single phase ﬂow regime. the pressure diﬀerence in the (x direction) is known and equal to zero. .260 CHAPTER 9. A ﬂow in an angle of inclination is closer to verti. Additionally. QL . The ﬂow will change to pulse ﬂow when the heavy liquid ﬂow rate increases. A ﬂow map to explain the cal ﬂow unless the angle of inclination is very small. Liquid Flow Rate Pulse Flow Straitified Flow 9. physical properties) from the vertical ﬂow and is shown in Figure 9. The boundLiquid Gas Flow ary conditions for the liquid is that velocity at the Flow wall is zero and the velocity at the interface is the same for both phases UG = UL or τi |G = τi |L . Further increase of the gas velocity will reduce the average liquid velocity.

The liquid ﬁlm thickness is unknown and can be expressed as a function of the above boundary conditions. Hence. the gravitational force has to be balanced by the shear forces as dτxy = ρL g dx The integration of equation (9.57) The liquid velocity at the wall. is zero and the integration coeﬃcient can be found to be C2 = 0 The liquid velocity proﬁle is then ρL g µL x2 τi x − hx + 2 µL (9. [U (x = 0) = 0].9. It is assumed that the last assumption does not contribute or change signiﬁcantly the results. COUNTER–CURRENT FLOW 261 This model is presented to exhibits the trends and the special features of counter-current ﬂow.63) . τi = ρL g h + C1 The integration constant is then Ci = τi − ρL g h which leads to τxy = ρL g (x − h) + τi Substituting the newtonian ﬂuid relationship into equation (9.59) (9.62) Uy = (9. Thus. C1 .).61) (9. The underline rational for this assumption is that gas density does not change signiﬁcantly for short pipes (for more information look for the book “Fundamentals of Compressible Flow” in Potto book series in the Fanno ﬂow chapter.60) dUy = ρL g (x − h) + τi dx (9.55) The integration constant.60) can be integrate to yield Uy = ρL g µL x2 τi x − hx + + C2 2 µL (9. On the liquid side. Assuming the pressure diﬀerence in the ﬂow direction for the gas is constant and uniform. can be found from the boundary condition where τxy (x = h) = τi .55) results in τxy = ρL g x + C1 (9.56) (9.58) (9.9. the liquid ﬂow rate is a function of the boundary conditions.58) to obtained µL or in a simpliﬁed form as dUy ρL g (x − h) τi = + dx µL µL Equation (9.

h.68) Where w is the thickness of the conduit (see Figure 9. The point where the liquid ﬂow rate is zero is important and it is referred to as initial ﬂashing point. Integration equation (9.69). The ﬁrst two solutions are identical in which the ﬁlm height is h = 0 and the liquid ﬂow rate is zero. also.69) It is interesting to ﬁnd the point where the liquid mass ﬂow rate is zero.69) is equated to zero. This request is identical to the demand in which 2 g h ρL 3 τi critical = (9.70) . the ﬂow rate is zero when 3 τi = 2 g h ρL .68) results in Q h2 (3 τi − 2 g h ρL ) = w 6 µL (9. This point can be obtained when equation (9. There are three solutions for equation (9. But.66) The maximum x value is limited by the liquid ﬁlm thickness. Q = w h h Uy dx = 0 0 ρL g µL x2 τi x − hx + dx 2 µL (9.67) If the shear stress is below this critical shear stress τi0 then no part of the liquid will have a reversed velocity. The ﬂow rate can be calculated by integrating the velocity across the entire liquid thickness of the ﬁlm. MULTI–PHASE FLOW τi h ρL g h2 − µL 2 µL (9.65) The solution for equation (9.64) The velocity can vanish (zero) inside the ﬁlm in another point which can be obtained from 0= ρL g µL τi x x2 − hx + 2 µL (9. The minimum shear stress that start to create reversible velocity is obtained when x = h which is 0= ρL g µL h2 τi h − hh + 2 µL h g ρL → τi0 = 2 (9. The notation of τi 0 denotes the special value at which a starting shear stress value is obtained to have reversed ﬂow.262 The velocity at the liquid–gas interface is Uy (x = h) = CHAPTER 9.65) is x|@UL =0 = 2 h − 2 τi µL g ρL (9.15).

72) the direction) 12 becomes (notice the change of the sign accounting for g h ρL 3 τL |@wall = (9. If the gas was compressible with an ideal gas equation of state then the pressure gradient is logarithmic. Utilizing the Newtonian relationship. Thus. the critical upward interface velocity is (2−1) 3 2 1 ρL g h2 (9.9. it is assumed that pressure gradient is linear. For this shear stress.73) Again. the linear equation is used.9. This assumption means or implies that the gas is incompressible ﬂow.72) Simplifying equation (9. reduces the ﬂow rate to zero or eﬀectively “drying” the liquid (which is diﬀerent then equation (9.76) noticing that equation (9. The shear stress on gas side is balanced by the pressure gradient in the y direction. Here. The momentum balance on element in the gas side is dτxy G dP = dx dy (9. For simplicity. the diﬀerential equation is ∆P d2 UG = dx2 µG L 12 Also (9.70) has to be equal g h ρL to support the weight of the liquid.74) can be rewritten as dτxy G ∆P ∆P = = dx ∆y L (9.67)).74) The pressure gradient is a function of the gas compressibility. the gas is assumed to be in a laminar ﬂow as well. COUNTER–CURRENT FLOW 263 This critical shear stress.71) 6 µL The wall shear stress is the last thing that will be done on the liquid side. for simplicity reasons.75) Where ∆y = L is the entire length of the ﬂow and ∆P is the pressure diﬀerence of the entire length. for a given ﬁlm thickness. In reality the logarithmic equation should be used ( a discussion can be found in “Fundamentals of Compressible Flow” a Potto project book). equation (9. . The wall shear stress is τi Ucritical |interf ace = τL |@wall = µL dU dx x=0 ρL g 2 g h ρL 1 B0 = µL 2¨ x µL ¨ − h + 3 µL x=0 (9.

80) (a) (b) or (9.83) (9. . in that case. (9. becomes ρL g h2 ∆P = h2 − D2 + C1 (h − D) 6 µL µG L The last integration constant. However.78) Which leads to UG = ∆P x2 − D2 + C1 (x − D) µG L (9. Mathematically these boundary conditions are UG (x = D) = 0 and UG (x = h) = UL (x = h) τG (x = h) = τL (x = h) Applying B. it is easy to show that the gas shear stress is not equal to the liquid shear stress at the interface (when the velocities are assumed to be the equal).77) results in UG = 0 = ∆P D2 + C1 D + C2 µG L ∆P → C2 = − D 2 + C1 D µG L (9. equation (9.264 CHAPTER 9.78) into equation (9.79) (9. The diﬀerence in shear stresses at the interface due to this assumption. cause this assumption to be not physical.64) when (x = h). the gas velocity proﬁle is UG = ∆P ρL g h2 (x − D) ∆P (h + D) (x − D) x2 − D2 + − µG L 6 µL (h − D) µG L (9. MULTI–PHASE FLOW Equation (9. The velocity at the interface is the same as the liquid phase velocity or the shear stress are equal.79)(a).84) The velocity in Equation (9.84) is equal to the velocity equation (9.77) This velocity proﬁle must satisfy zero velocity at the right wall. of the equal velocities.C.81) At the other boundary condition.82) With the integration constants evaluated.76) can be integrated twice to yield UG = ∆P 2 x + C1 x + C2 µG L (9. C1 can be evaluated as C1 = ρL g h 2 ∆P (h + D) − 6 µL (h − D) µG L (9.

9.89) This gas interface velocity is diﬀerent than the velocity of the liquid side. if no special eﬀects occurs. The shear stress at the interface must be equal. the two conditions can co–exist. Since there no possibility to have both the shear stress and velocity on both sides of the interface. The wall shear stress is τG |@wall = µG dUG dx = µG x=D ∆P 2 x + µG L 2 g h ρL 2 h ∆P − 3 µG µG L (9. COUNTER–CURRENT FLOW 265 The second choice is to use the equal shear stresses at the interface.91) The Required Pressure Diﬀerence . Then if the interface becomes wavy.85) The expressions for the derivatives are gas side liquid side 2 h ∆P 2 g h ρL + µG C1 = L 3 As result. This condition requires that µG dUG dUL = µL dx dx (9. It was assumed that the interface is straight but is impossible. The velocity at interface can have a “slip” in very low density and for short distances.87) (9. the integration constant is C1 = The gas velocity proﬁle is then UG = ∆P x2 − D2 + µG L 2 g h ρL 2 h ∆P − 3 µG µG L (x − D) 2 g h ρL 2 h ∆P − 3 µG µG L (9. diﬀerent thing(s) must happen.9.88) The gas velocity at the interface is then UG |@x=h = ∆P h2 − D 2 + µG L 2 g h ρL 2 h ∆P − 3 µG µG L (h − D) (9. condition (9.90) x=D or in a simpliﬁed form as τG |@wall = 2 ∆P (D − h) 2 g h ρL + L 3 (9.79)(b).86) (9.

16. In D ∆P the same fashion the pressure difference under the assumption the equal velocity can be calculated.95) .16) A/w G A/w L f orce due to pressure FgL + L τw = L τw + Substituting the diﬀerent terms into (9.94) Simplifying equation (9. However. this explanation is to show magnitudes and trends and hence it provided here.92) The total momentum balance is (see Figure 9. There are two forces that act against the gravity and two forces with the gravity. To calculate the required pressure that cause the liquid to dry.93) result in ρgLh + L 2 ∆P (D − h) 2 g h ρL + L 3 D ∆P (9.93) =L g h ρL + D ∆P 3 (9.96) (9. This model and its assumptions are too simplistic and the actual pressure diﬀerence is larger. Figure 9. Fig. However.16 describes the general forces that acts on the control volume. the total balance is needed.266 CHAPTER 9. as ρgLh L ﬁrst estimate the waviness of the Lτw |L Lτw |G surface can be neglected. The gravity force on the liquid is the liquid volume times the liquid volume as V olme/w FgL = ρ g hL (9.94) results in 4ρgLh = (2 h − D) ∆P 3 or ∆P = 4ρgLh 3 (2 h − D) (9. -9. The control volume include the gas and liquid volumes. The gravity force on the gas can be neglected in most cases. General forces diagram to calculated the in a The actual pressure diﬀerence can two dimension geometry. MULTI–PHASE FLOW The pressure diﬀerence to D create the ﬂooding (drying) has to take into account the fact that h W x y the surface is wavy. The estimation of the pressure diﬀerence under the assumption of equal shear stress can be applied. be between these two assumptions but not must be between them.

In other case where more accurate measurement are needed a speciﬁc model is required. the appropriate model for the ﬂow regime should be employed.10 Multi–Phase Conclusion For the ﬁrst time multi–phase is included in a standard introductory textbook on ﬂuid mechanics. . The interface between the two liquid ﬂowing together is wavy. There are several points that should be noticed in this chapter. Perhaps as a side conclusion but important.10. 9. There are many ﬂow regimes in multi–phase ﬂow that “regular” ﬂuid cannot be used to solve it such as ﬂooding.96) indicates that when D > 2 h is a special case (extend open channel ﬂow). the assumption of straight line is not appropriate when two liquid with diﬀerent viscosity are ﬂowing. MULTI–PHASE CONCLUSION 267 This analysis shows far more reaching conclusion that initial anticipation expected.9. In that case. Unless the derivations or assumptions are wrong. The homogeneous models or combined models like Lockhart–Martinelli can be employed in some cases. this analysis equation (9.

268 CHAPTER 9. MULTI–PHASE FLOW .

1. This appendix condenses material that spread in many various textbooks some of which are advance. -A. Furthermore.APPENDIX A The Mathematics Backgrounds for Fluid Mechanics In this appendix a review of selected topics in mathematics related to ﬂuid mechanics is presented. A.1) z U Ux Uy y Uz x Vector can be normalized and in Cartesian coordi. Hence without additional reading.). Uy is the vector component in the y direction. and Uz is the vector component in the z direction. Vector in Cartesian coor- . this book on ﬂuid mechanics issues could be read by most readers.1 where Ux is the vector component in the x direction. The material is not presented in “educational” order but in importance order.dinates system. some of the material appears in specialty books such as third order diﬀerential equations (and thus it is expected that the student is not familiar with this material. The length of the vector in Cartesian coordinates (the coordinates system is relevant) is U = Ux 2 + Uy 2 + Uz 2 (A. There is very minimal original material which appears without proofs. nates depicted in Figure A. the 269 Fig. Thus.1 Vectors Vector is a quantity with direction as oppose to scalar. These topics are present so that one with some minimal background could deal with the mathematics that encompass within basic ﬂuid mechanics.

a (b U ) = a b U The multiplications and the divisions have somewhat diﬀerent meaning in a scalar operations. Then the following can be said U U V 1. V . U . -A. A. There are two kinds of multiplications for vectors. h2 .1. The “cross” product is deﬁned in an orthogonal coordinate (h1 . a (U + V ) = a U + a V 6. regular scalar multiplication angle between vectors W V U Fig. The right hand rule. U + V = V + U 3. The following are vectors. and W and for in this discussion a and b are scalars. V )) (A. Zero vector is such that U + 0 = U 4.3) Vectors have some what similar rules to scalars which will be discussed in the next section.4). MATHEMATICS FOR FLUID MECHANICS U Ux ˆ Uy ˆ Uz ˆ = i+ j+ k U U U U U = (A.1 Vector Algebra Vectors obey several standard mathematical operations which are applicable to scalars. Additive inverse U − U = 0 U 5. (U + V ) + W = (U + V + W ) = U + (V + W ) 2. V )) n (A.2) and general orthogonal coordinates U = U U1 U2 U3 = h1 + h2 + h3 U U U U (A.2.270 unit vector is APPENDIX A. U ·V = U V U |U | · |V | cos (∠(U . The ﬁrst multiplication is the “dot” product which is deﬁned by equation (A. The results of this multiplication is scalar but has no negative value as in regular scalar multiplication. and h3 ) as angle U V U U × V = |U | · |V | sin (∠(U .4) The second multiplication is the “cross” product which in vector as opposed to a scalar as in the “dot” product.5) . multiplication of U × V results in W .

that is scalar associated with cos θ vectors is associated with sin θ. Additionally note that order of multiplication is signiﬁcant.9) in matrix form as h1 U × V = U2 V2 h2 U2 V2 h3 U3 V3 (A. The reason that these current combinations. .6) (A. Multiplication combinations shows that there are at least four possibilities of combining the angle with scalar and vector. It turn out that these combinations have very little1 physical meaning. In right hand orthogonal coordinate system h1 × h2 = h3 h2 × h3 = h1 h3 × h1 = h2 The “cross” product can be written as U × V = (U2 V3 − U3 V2 ) h1 + (U3 V1 − U1 V3 ) h2 + (U1 V2 − U2 V1 ) h3 Equation (A.8) The most complex of all these algebraic operations is the division. The multiplication in vector world have two deﬁnition one which results in a scalar and one which results in a vector.10) (A.A. the result of the division can be a scalar combined or associated with the angle (with cos or sin). VECTORS 271 where θ is the angle between U and V . is that these combinations have physical meaning. Note that U and V are not necessarily orthogonal. and n is a unit vector perpendicular to both U and V which obeys the right hand rule.9) h2 × h1 = −h3 h3 × h2 = −h1 h1 × h3 = −h2 2 2 2 2 2 2 (A. or vector with the angle. these above four combinations are not the only possibilities (not including the left hand system). This multiplication has a negative value which means that it is a change of the direction. One of the consequence of this deﬁnitions in Cartesian coordinates is i =j =k =0 In general for orthogonal coordinates this condition is written as h1 × h1 = h1 = h2 = h3 = 0 where hi is the unit vector in the orthogonal system. For example.7) (A. The previous experience is that help to deﬁne multiplication help to deﬁnition the division. The number of the possible combinations of the division is very large. etc. The right hand rule is referred to the direction of resulting vector.1. However. Additional possibility is that every combination of one vector element 1 This author did ﬁnd any physical meaning these combinations but there could be and those the word “little” is used.

” Also note the multiplying matrices and inverse matrix are also available operation to these tensors. derivatives of diﬀerent directions can represented as a vector or vector function. For example. such as.13) .2 Diﬀerential Operators of Vectors Diﬀerential operations can act on scalar functions as well on vector and vector functions. The three vectors have a need for additional notation such of vector of vector which is referred to as a tensor.z). i. and z directions. the curl. A scalar function is a function that provide a valued based on the coordinates (in Cartesian coordinates x.y. One of these operations is nabla operator sometimes also called the “del operator. in Cartesian coordinates the operation is =ˆ i ∂ ∂ ˆ ∂ +ˆ j +k ∂x ∂y ∂z (A.272 APPENDIX A. Many of the operations of vector world. and the Laplacian are based or could be constructed from this single operator. y.” This operator is a diﬀerential vector. A. This tenser or the matrix can undergo regular linear algebra operations such as ﬁnding the eigenvalue values and the eigen “vectors. the temperature of the domain might be expressed as a scalar ﬁeld. In multivariate calculus.12) ˆ Where ˆ ˆ and k are denoting unit vectors in the x. MATHEMATICS FOR FLUID MECHANICS is divided by the other vector element. The following combination is commonly suggested U2 U3 U1 V V1 V1 1 U U1 U2 U3 = (A. divergence.11) V2 V V2 V2 U2 U3 U1 V3 V3 V3 One such example of this division is the pressure which the explanation is commonality avoided or eliminated from the ﬂuid mechanics books including the direct approach in this book. A compact presentation is a common way to handle the mathematics which simplify the calculations and explanations. It turned out that combination of three vectors has a physical meaning.1. For example. j. respectively. the gradient. Since every vector element has three possible elements the total combination is 9 = 3 × 3. There at least are two possibilities how to treat these elements. =ˆ i ∂T ˆ ∂T ˆ ∂T +j +k ∂x ∂y ∂z (A. Gradient This operation acts on a scalar function and results in a vector whose components are derivatives in the principle directions of a coordinate system. More diﬀerential operations can on scalar function can results in vector or vector function.

Laplacian The new operation can be constructed from “dot” multiplication of the nabla. Applying a divergence on the result creates a scalar ﬁeld again.1. and it deﬁned as 2 = 2 − 1 ∂2 c2 ∂ 2 t (A. This combined operations is known as the “div grad” which is given in Cartesian coordinates by · = ∂2 ∂2 ∂2 + 2+ 2 2 ∂x ∂y ∂z (A. y. in honor of Pierre-Simon Laplace (23 March 1749 – 5 March 1827). N . This operator also referred as the Laplacian operator.17) This combination is commonality denoted as 2 .14) The divergence results in a scalar function which similar to the concept of the vectors multiplication of the vectors magnitude by the cosine of the angle between the vectors. z)ˆ + Nz (x. y. The divergence is the similar to “dot” product which results in scalar. in Cartesian coordinate is results in div N = ·N= ∂Nx ∂Ny ∂Nz + + ∂x ∂y ∂z (A. VECTORS 273 Divergence The same idea that was discussed in vector section there are two kinds of multiplication in the vector world and two will be for the diﬀerential operators.18) . y. d‘Alembertian As a super–set for four coordinates (very minimal used in ﬂuid mechanics) and it reﬀed to as d’Alembertian or the wave operator. z)k i j The dot product of these two vectors. y. A vector domain (function) assigns a vector to each point such as velocity for example.16) Note that the result is a vector.15) (A. for Cartesian coordinates is ˆ N (x.A. A gradient acting on a scalar ﬁeld creates a vector ﬁeld. z)ˆ + Ny (x. Curl Similar to the “cross product” a similar operation can be deﬁned for the nabla (note the “right hand rule” notation) for Cartesian coordinate as curl N = ×N = ∂Nz ∂Ny − ∂y ∂z ∂Nx ∂Nz − ∂z ∂x ˆ i+ ˆ+ j ∂Ny ∂Nx − ∂x ∂y ˆ k (A. z) = Nx (x.

Many engineering class present this theorem as a theorem on its merit without realizing that it is a subset of Reynolds Transport Theorem. · · · )e2 + f3 (x1 . the dot product of operation 2 for more information http://math.edu/mathews/c2003/HarmonicFunctionMod. the harmonic functions can be added or subtracted because inside the volume these functions contributions is eliminated throughout the volume. The cross and dot products when the come under diﬀerentiation can be look as scalar.1.html . In Gauss notation it is written as ( V · N ) dV = A N · n dA (A.fullerton. z) then it can insert into identity to produce ( V · ( Φ)) dV = V 2 Φ dV = A Φ · n dA (A. Thus. x3 . The basic of these relationships is explored.20) V Note the strange notation of “Σ” which refers to the area. A.22) where e i is the unit vector in the i direction. A vector is made of the several scalar functions such as R = f1 (x1 . If N is a gradient of a scalar ﬁeld Π(x. x3 . · · · )e1 + f2 (x1 . x2 . MATHEMATICS FOR FLUID MECHANICS Divergence Theorem Mathematicians call to or refer to a subset of The Reynolds Transport Theorem as the Divergence Theorem.21) Since the deﬁnition of Φ = N . to a balance of the ﬂuxes through the surface. This subset can further produces several interesting identities.3 Diﬀerentiation of the Vector Operations The vector operation sometime fell under (time or other) derivative. Special case of equation (A. x2 . or called it Gauss’ Theorem (Carl Friedrich Gauss 30 April 1777 23 February 1855). For example.19) In Gauss-Ostrogradsky Theorem (Mikhail Vasilievich Ostrogradsky (September 24. 1862). y.274 APPENDIX A. · · · )e3 + · · · (A. 1801 – January 1. normally for steady state.21) for harmonic function (solutions Laplace equation see2 Harminic functions) then the left side vanishes which is useful identity for ideal ﬂow analysis. x2 . This results reduces equation. x3 . The notation is a bit diﬀerent from Gauss and it is written in Ostrogradsky notation as ∂Q ∂R ∂P + + ∂x ∂y ∂z dx dy dz = Σ (P p + Qq + Rr) dΣ (A. This theorem is applicable for a ﬁx control volume and the derivative can enters into the integral.

There are several orthogonal coordinates which appears in ﬂuid mechanics operation which include this list: Cartesian coordinates.1. Prolate spheroidal coordinates. it can done for the cross product. Cylindrical coordinates. Parabolic cylindrical coordinates Paraboloidal coordinates. the chain role is applicable. VECTORS R · S = (xˆ + y 2 ˆ · (sin xˆ + exp(y)ˆ can be written as i j) i j) R d (R · S ) d = dt dt It can be noticed that d x sin x + y 2 exp(y) R d (R · S ) = = dt dt dx d sin x d y 2 d y2 sin x + + exp(y) + exp(y) dt dt dt dt xˆ + y 2 ˆ · sin xˆ + exp(y)ˆ i j i j 275 It can be noticed that the manipulation of the simple above example obeys the regular chain role.25) (A.A. Yet.23) The ﬁrst part is zero because the cross product with itself is zero. and acceleration. R S d dR dS R (R · S ) = ·S + ·R dt dt dt And the the chain role for the cross operation is R S d dR dS R (R × S ) = ×S + ×R dt dt dt It follows that derivative (notice the similarity to scalar operations) of d dR R (R · R ) = 2 R dt at There are several identities that related to location. A. Parabolic coordinates. velocity.1.1 Orthogonal Coordinates These vectors operations can appear in diﬀerent orthogonal coordinates system. The results of operations of two vectors is similar to regular multiplication since the vectors operation obey “regular” addition and multiplication roles. Similarly. the most interesting is U d dU R (R × U ) = U × U + R × dt dt (A. The second part is zero because Newton law (acceleration is along the path of R). Elliptic . Oblate spheroidal coordinates. As in operation on scalar time derivative of dot or cross of constant velocity is zero. Spherical coordinates. Hence the chain role apply for dot operation. Ellipsoidal coordinates.24) (A.3.

ac. Cylindrical Coordinates The cylindrical coordinates are commonz ality used in situations where there is line of e1 symmetry or kind of symmetry. These three diﬀerent rs are some what similar to any of the Cartesian coordinate. Note the advantage of cylindrical (polar) coordinates in description of geometry or location relative to a center point.3.276 APPENDIX A. The new concept here is the length factor. θ.26) The reverse transformation is x = r cos θ y = r sin θ z=z (A.27) The line element and volume element are ds = dr2 + (r dθ) + dz 2 2 dr r dθ dz (A. Flat-disk cyclide coordinates. r. In Fig. The coordinate θ is angle. The meaning of − r and r are diﬀerent. Bipolar cylindrical coordinates Conical coordinates.uk/Extras/Coolidge Polars. Bi-cyclide coordinates and Cap-cyclide coordinates. and z. Toroidal coordinates. Bispherical coordinates.29) 3 Coolidge. Three common systems typical to ﬂuid mechanics will be presented and followed by a table and methods to present all the above equations. Because there are so many coordinates system is reasonable to develop these operations for any for any coordinates system.28) The gradient in cylindrical coordinates is given by =r ∂ 1 ∂ ∂ +θ +z ∂r r ∂θ ∂z (A. Julian (1952). → and z. These coordinates rer θ r duced the work.st-and. MATHEMATICS FOR FLUID MECHANICS cylindrical coordinates. The second coordinate θ has unite coordinate θ.3. because probθ x lem is reduced a two dimensions. http://www-history.html. In this book the dimensional chapter shows that in physics that derivatives have to have same units in order to compare them or use them. -A. x these coordinate were introduced for geometrical problems about 2000 years ago3 . Cylindrical Coordinate System. American Mathematical Monthly 59: 7885. The ﬁrst one represents the vector that is the direction of r while the second is the unit vector in the direction of the coordinate r. This kind sity uations occur in pipe ﬂow even if the pipe is not exactly symmetrical. . in this case. Flat-ring cyclide coordinates.mcs. Note that unite coordinates are denoted as r. The conversion between the Cartesian coordinate and the Cylindrical is r= x2 + y 2 θ = arctan y x y z=z (A. ”The Origin of Polar Coordinates”. Conversation of the angel to units of length is done by length factor which is. θ. in most cases. Historically. the ﬁgure shows that the coordinates are r. The cylindrical coordinates are shown in Figure A.

The second angle is between the positive y–axis and the vector as shown in Figure ?? The conversion between Cartesian coordinates to Spherical coordinates x = r sin φ cos θ The reversed transformation is r= x2 + y 2 + z 2 φ = arccos z r (A. Note that the ﬁrst angle (azimuth or longitude) θ range is between Fig. The ﬁrst angle between projection on x − y plane and the positive x–axis. VECTORS The curl is written ×N = 1 ∂Nz ∂Nθ − r ∂θ ∂z 1 r The Laplacian is deﬁned by · = 1 ∂ r ∂r r ∂ ∂r + 1 ∂2 ∂2 + 2 2 ∂θ 2 r ∂z r+ ∂Nr ∂Nz − ∂z ∂r ∂ (r Nθ ) ∂Nθ − ∂r ∂θ θ+ z 277 (A. In ﬂuid y mechanics such situations exist in bubble dynamics. sound wave propagax tion etc.35) The gradient is =r ∂ ∂ ˆ1 ∂ + φ 1 +θ ∂r r ∂θ r sin θ ∂φ (A. 0 < θ < 2 π while the second angle (colatitude) is only 0 < φ < π.A.37) .30) (A. -A.1.32) Spherical Coordinates z The spherical coordinates system is a φ r three-dimensional coordinates which is imθ provement or further modiﬁcations of the cylinφ r drical coordinates. Spherical Coordinate System. The radius is the distance between the origin and the location. Spherical system used for z y x θ cases where spherical symmetry exist.36) The divergence in spherical coordinate is ·N = 1 ∂ r2 Nr 1 ∂ (Nθ sin θ) 1 ∂Nφ + + r2 ∂r r sin θ ∂θ r sin θ ∂φ (A. A location is represented by a radius and two angles.31) (A. boom explosion.4.34) y = r sin φ sin θ z = r cos φ (A.33) Line element and element volume are ds = dr2 + (r cos θ dθ) + (r sin θ dφ) 2 2 dV = r2 sin θ dr dθ dφ (A.

-A. and z . The general Orthogonal is r. The coordinates distance coeﬃcient is in ase cre in e2 the change the diﬀerential to the actual distance.42) . The derivative quantity will be denoted by q superscript. the unit vectors are: r.2.39) General Orthogonal Coordinates There are several orthogonal system and general form is needed. The length of d d 2 = i=1 hk dq k 2 (A. In mathematics it is denoted as q while in engineering is denotes h. there is dispute what with unit vectors. θ is lengthens unit vector and the coordinate distance coeﬃcient in this case Fig. For example in cylindrical coordinates.41) Gradient The gradient in general coordinate for a scalar function T is the nabla operator in general orthogonal coordinates as T = T T T e1 ∂T e2 ∂T e3 ∂T + + h1 ∂q 1 h2 ∂q 2 h3 ∂q 3 (A. The notation for the presentation is required general notation of the units vectors is ei and coordinates distance coeﬃcient is hi where 1 e1 q i is 1.38) ˆ φ The Laplacian in spherical coordinates is 2 = r2 + sin θ + (A. θ. As in almost all cases. Since it is engineering book the h is adapted. The units r and z are units ˆ ˆ ˆ ˆ ˆ ˆ with length. However. Also note that the derivative of the coordinate in the case of cylindrical coordinate is ∂θ and unit ˆ vector is θ. While the θ is the same the meaning is diﬀerent and diﬀerent notations need.5.278 APPENDIX A.40) The nabla operator in general orthogonal coordinates is = e1 ∂ e2 ∂ e3 ∂ + + 1 2 h1 ∂q h2 ∂q h3 ∂q 3 (A. the proper notation for these coeﬃcients. MATHEMATICS FOR FLUID MECHANICS The curl in spherical coordinates is ×N = 1 r sin θ 1 r 1 ∂ r2 ∂r ∂ (Nφ sin θ) ∂Nθ − ∂θ ∂φ 1 ∂Nr ∂ (rNφ ) − sin θ ∂φ ∂r ∂ ∂r 1 ∂ 2 sin θ ∂θ r r+ ˆ 1 ˆ θ + r ∂ ∂θ ∂ (rNθ ) ∂Nr − ∂r ∂θ 1 ∂2 r2 sin2 θ ∂φ2 (A.3.

1. . 2 φ= + + Fig.A. Parabolic coordinates by user WillowW using Blender.45) The following table showing the diﬀerent values for selected orthogonal system.6. 1 h1 h2 h3 ∂q ∂q ∂q 279 (A.43) For general orthogonal coordinate system the curl is ∂ e1 ∂ (h3 N3 ) − 3 (h2 N2 ) + 2 h2 h3 ∂q ∂q ∂ ∂ e3 ∂ ∂ (h1 N1 ) − 1 (h3 N3 ) + (h2 N2 ) − 2 (h1 N1 ) 3 1 ∂q ∂q h1 h2 ∂q ∂q ×N = e2 h3 h1 (A.44) The Laplacian of a scalar equals 1 ∂ h1 h2 h3 ∂q 1 h2 h3 ∂φ h1 ∂q 1 ∂ ∂q 2 h3 h1 ∂φ h2 ∂q 2 ∂ ∂q 3 h1 h2 ∂φ h3 ∂q 3 (A. -A. VECTORS The divergence of a vector equals ·N = 1 ∂ ∂ ∂ (N1 h2 h3 ) + 2 (N2 h3 h1 ) + 3 (N3 h1 h2 ) .

it is possible to deﬁne fraction of derivative.2 Ordinary Diﬀerential Equations (ODE) In this section a brief summary of ODE is presented. A. . Ordinary diﬀerential equations are deﬁned by the order of the highest derivative.46) d An example of such linear operation L = dt + 1 acting on y is dy1 + y1 . there is no physical meaning to such a product according to this author believe.2.47) 4 Note that mathematically. If the highest derivative is ﬁrst order the equation is referred as ﬁrst order diﬀerential equation etc. It is not intent to be a replacement to a standard textbook but as a quick reference.g. the ﬁrst ODEs are easier to solve and they are the base for equations of higher order equation. Orthogonal coordinates systems (under construction please ignore) Orthogonal coordinates systems name Cartesian Cylindrical Spherical Paraboloidal Ellipsoidal Remarks 1 standard common common ? ? √ 1 1 1 u2 + v 2 √ h 2 1 r r u2 + v 2 3 1 1 r cos θ uv 1 x r r u λ q 2 y θ θ v µ 3 z z ϕ θ ν A. It is suggested that the reader interested in depth information should read “Diﬀerential Equations and Boundary Value Problems” by Boyce de–Prima or any other book in this area. MATHEMATICS FOR FLUID MECHANICS Table -A. The meaning of linear equation is that the operation is such that a L (u1 ) + b L (u2 ) = L (a u1 + b u2 ) (A. The most general form is f u. However. Note that the derivatives are integers e. du . ﬁrst derivative. The ﬁrst order equations have several forms and there is no one solution ﬁt all but families of solutions. ODE are categorized into linear and non-linear equations.280 APPENDIX A. second derivative etc4 .1. Or this dt dy2 operation on y2 is dt + y2 and the summation of operation the sum operation of +y L(y1 + y2 ) = y1dt 2 + y1 + y2 .t dt =0 (A.1 First Order Diﬀerential Equations As expect.

In that case it is said that F is spreadable and then equation (A.b) becomes du = u t dt =⇒ ln (u) + ln (c) = t2 2 (1.I.49) Equation can be integrated either analytically or numerically and the solution is du = U (u) X(t)dt (A. The initial condition is the value the function has at some points.I. (1.I.50) The limits of the integral is (are) the initial condition(s).48) can be written as F (t.I. The separation yields du = t dt u The integration of equation (1.48) can be written as du = X(t)dt U (u) (A.c) can be transferred to u = c et For the initial condition of u(0) = u0 then u = u0 et End Solution 2 2 (1.2 Variables Separation or Segregation In some cases equation (A. u) = X(t) U (u).e) .2.A. ORDINARY DIFFERENTIAL EQUATIONS (ODE) Sometimes equation (A.I. Example A.1: Solve the following equation du = ut dt with the initial condition u(t = 0) = u0 .b) (1.I.a) Solution The solution can be obtained by the variable separation method.48) A.d) (1.c) Equation (1.I. The name initial condition is used because the values are given commonly at initial time. u) dt 281 (A.2.47) can be simpliﬁed to the ﬁrst form as du = F (t.

59) N (x) A special case of g(t) = constant is shown next. MATHEMATICS FOR FLUID MECHANICS The Integral Factor Equations Another method is referred to as integration factor which deals with a limited but very important class of equations.53) This condition (note that the previous methods is employed here) requires that d N (x) d N (x) = N (x) g(x) =⇒ = g(x) dx dx N (x) Equation (A.57) by dx and integrating results in N (x) g(x) y = The solution is then N (x) m(x) dx y= g(x) N (x) m(x) dx (A.52) becomes d [N (x) g(x) y] = N (x) m(x) dx Multiplying equation (A.52) (A.53). The general form of the equation is dy + g(x) y = m(x) dx Multiplying equation (A.282 A.2. Thus equation (A.55) (A. .56) which indeed satisfy equation (A.51) by unknown function N (x) transformed it to N (x) dy + N (x) g(x) y = N (x)m(x) dx (A.54) Using the diﬀerentiation chain rule provides dv du du dx d N (x) =e dx g(x)dx g(x) (A.58) (A. This family is part of a linear equations.57) e R g(x)dx (A.1 APPENDIX A.51) What is needed from N (x) is to provide a full diﬀerential such as N (x) dy d [N (x) g(x) y] + N (x) g(x) y = dx dx (A.54) is integrated to be ln (N (x)) = g(x)dx =⇒ N (x) = e g(x)dx (A.2.

Many non linear equations can be transformed into linear equations and then solved with the linear equation techniques. The actual solution is obtained by reversed equation which transferred solution to u = v (p−1) (A.60) into a linear equation which is dv + (1 − p) m(t) v = (1 − p) n(t) dt (A.3 Non–Linear Equations Non-Linear equations are equations that the power of the function or the function derivative is not equal to one or their combination.60) The transformation v = u1−p turns equation (A.62) Example A.2: Find the solution for a typical problem in ﬂuid mechanics (the problem of Stoke ﬂow or the parachute problem) of dy +y =1 dx Solution Substituting m(x) = 1 and g(x) = 1 into equation (A. .59) provides y = e−x (ex + c) = 1 + c e−x End Solution A.a) to be confused with the Bernoulli equation without the s that referred to the energy equation.3: Solve the following Bernoulli equation du + t2 u = sin(t) u3 dt 5 Not (1. This equation is non–linear part du + m(t)u = n(t) dt up (A.2.2. One such equation family is referred in the literature as the Bernoulli Equations5 .A.61) The linearized equation can be solved using the linear methods.III. ORDINARY DIFFERENTIAL EQUATIONS (ODE) 283 Example A.

63) for any real positive a.64) In another words if the substitution u = v t is inserted the function f become a function of only v it is homogeneous function.65) Example A.a) . The solution is then ln |t| = dv +c f (1.1 Homogeneous Equations Homogeneous function is given as du = f (u.III.3. t) = f (a u.IV. v) − v (A.b) Using the deﬁnition (1.4: Solve the equation du u = sin + dt t u4 − t4 t4 (1. For this case.2.c) (1. a t) dt (A.b) equation (1.III.a) becomes dv −2 t2 v = −2 sin(t) dt The homogeneous solution of equation (1.d) (1.III. Example of such case u = u3 − t3 /t3 becomes u = v 3 + 1 .III.III. v) dt (A.III. the transformation of u = v t transforms equation (A.c) is u(t) = ce And the general solution is t3 − 3 −t3 3 1−p 1−p (1.e) private solution u= e e t3 3 sin (t) dt +c End Solution A. MATHEMATICS FOR FLUID MECHANICS v = u2 (1.284 Solution The transformation is APPENDIX A.63) into t dv + v = f (1.III.

5: Solve the following ODE du = −u2 t2 dt (1.66) The main point is that f (t) and be segregated from g(u).d) (1.V. End Solution 285 (1. ORDINARY DIFFERENTIAL EQUATIONS (ODE) Solution Substituting u = v T yields du = sin (v) + v 4 − 1 dt or dv dv + v = sin (v) + v 4 − 1 =⇒ t = sin (v) + v 4 − 1 − v dt dt Now equation (1.IV.IV.2.e) A.A.2.IV.67) g(u) Example A.a) Solution Segregating the variables to be du = u2 t2 dt (1.IV.c) (1. this kind of class of equations appears all over this book.c) can be solved by variable separation as t dv = t dt sin (v) + v 4 − 1 − v Integrating equation (1. In fact.IV. The solution of this kind of equation is du = f (t) dt (A.IV. For this sort equations. it can be written that du = f (t)g(u) dt (A.V.d) results in dv t2 = +c sin (v) + v 4 − 1 − v 2 The initial condition can be inserted via the boundary of the integral.2 Variables Separable Equations In ﬂuid mechanics and many other ﬁelds there are diﬀerential equations that referred to variables separable equations.b) .3.b) (1.

c) Rearranging equation (1. the solution of the quadratic equation is a complex number which means that the solution has exponential and trigonometric functions as u = c1 eα t cos(βt) + c2 eα t sin(βt) (A.d) End Solution A.b) transformed into − 1 t3 = + c1 u 3 (1.2.286 APPENDIX A.68) In a way.70) In the case of b2 > 4 a c.4 Second Order Diﬀerential Equations The general idea of solving second order ODE is by converting them into ﬁrst order ODE.c) becomes u= t3 −3 +c (1.3 Other Equations There are equations or methods that were not covered by the above methods. One such case is the second order ODE with constant coeﬃcients. MATHEMATICS FOR FLUID MECHANICS Integrating equation (1. transformation (like Laplace transform).V.V. A.71) (A. Many of these methods will be eventually covered by this appendix. it is done by substituting est where s is characteristic constant and results in the quadratic equation a s2 + b s + s = 0 (A. There are additional methods such numerical analysis.V. The simplest equations are with constant coeﬃcients such as a d2 u du +b + cu = 0 dt2 dt (A.V. variable substitutions. and perturbation methods. Practically.3. the second order ODE is transferred to ﬁrst order by substituting the one linear operator to two ﬁrst linear operators.72) .2.69) If b2 > 4 a c then there are two unique solutions for the quadratic equation and the general solution form is u = c1 es1 t + c2 es2 t For the case of b2 = 4 a c the general solution is u = c1 es1 t + c2 t es1 t (A.

74) Example A.a) Solution The characteristic equation is s2 + 7 s + 10 = 0 The solution of equation (1. Equations that not equal to zero in this form d2 u du (A.c) A. If the function on the right hand side is polynomial than the solution is will n utotal = uh + i=1 up i (A. . This fact can be used to solve non-homogeneous equation.VI.VI.A.VI.2.73) 4 a c − b2 2a (A.6: Solve the following ODE d2 u du +7 + 10 u = 0 dt2 dt (1.2.b) (1.4. ORDINARY DIFFERENTIAL EQUATIONS (ODE) Where the real part is α= and the imaginary number is β= √ −b 2a 287 (A. and −5. Thus.75) +b + c u = l(x) dt2 dt The solution of the homogeneous equation is zero that is the operation L(uh ) = 0. where L is Linear operator.b) are −2. the solution is u = k1 e−2 t + k2 e−5 t End Solution (1.77) The linearity of the operation creates the possibility of adding the solutions. The additional solution of L(up ) is the total solution as a =0 L (utotal ) = L (uh ) +L (up ) =⇒ utotal = uh + up (A.VI.76) Where the solution uh is the solution of the homogeneous solution and up is the solution of the particular function l(x).1 Non–Homogeneous Second ODE Homogeneous equation are equations that equal to zero.

79) .d) A.VII.b) 6t + 5 36 18 t2 + 30 t + 19 108 9 t2 + 24 t + 17 54 (1.7: Solve the non-homogeneous equation d2 u du −5 + 6 u = t + t2 dt2 dt Solution The homogeneous solution is u(t) = c1 e2 t + c1 e3 t the particular solution for t is u(t) = and the particular solution of the t2 is u(t) = The total solution is u(t) = c1 e2 t + c1 e3 t + End Solution (1. u.5 Non–Linear Second Order Equations Some of the techniques that were discussed in the previous section (ﬁrst order ODE) can be used for the second order ODE such as the variable separation. u) = 0 ˙ ¨ can be written or presented in the form f1 (u)u = f2 (u) u ˙ ˙ ¨ (A.78) then the equation (A.VII. ˙ ˙ ¨ equation (A. Hence. A.1 Segregation of Derivatives If the second order equation f (u.78) is referred to as a separable equation (some called it segregated equations).VII. The derivative of u can be treated as a new function v and v = u.288 APPENDIX A.c) (1.78) can be integrated u u ˙ v f1 (u)u = ˙ u0 u0 ˙ f2 (u) u = ˙ ¨ v0 f2 (u) v ˙ (A.2. MATHEMATICS FOR FLUID MECHANICS Example A.2.5.VII.a) (1.

The papers are still his drawer and waiting to a spare time.A.VIII. ﬁrst with initial integration and second with the second integration. 6 This . Note that the derivative initial condition is used once. This equation family is not well discussed in mathematical textbooks6 .VIII.b) After the integration equation (1.2.81) author worked (better word toyed) in (with) this area during his master but to his shame he did not produce any papers on this issue.VIII.VIII.VIII. It can be noticed that the function initial condition is used twice. ORDINARY DIFFERENTIAL EQUATIONS (ODE) 289 The integration results in a ﬁrst order diﬀerential equation which should be dealt with the previous methods.c) becomes 3 3 2 u 2 − u0 2 3 = cos (v0 ) − cos (v) = cos du0 dt − cos du dt (1.d) can be rearranged as du = arcsin dt t u 3 3 2 u0 2 − u 2 + cos (v0 ) 3 (A.c) du dt d dt & du dt du dt du dt (1.a) becomes u du = sin d (1.80) Using the ﬁrst order separation method yields dt = 0 u0 du 2 3 3 arcsin u0 2 −u 2 + cos (v0 ) 3 =0 =1 (A. Example A.a) du dt du dt d2 u =0 dt2 (t = 0) = 0 What happen to the extra Thus the extra dt is disappeared and equation (1.8: Solve the equation √ du u − sin dt With the initial condition of u(0) = 0 and “dt”? Solution Rearranging the ODE to be √ du u = sin dt & √ and transformation to v is √ u du = sin (v) dv (1.VIII. The physical reason is that the equation represents a strong eﬀect of the function at a certain point such surface tension problems.d) Equation (1.VIII.

87) It can be noticed that that c2 is actually two diﬀerent constants because the plus minus signs.2. MATHEMATICS FOR FLUID MECHANICS The solution (A. One solution is u = k1 and the second solution is obtained by solving 1 = a The transform of v = du dt du dt d2u d t2 (A.82) where a is constant.5.81) shows that initial condition of the function is used twice while the initial of the derivative is used only once. d dt u du dt =0 (A. u d2 u d t2 + du dt 2 =0 (A.2 Full Derivative Case Equations Another example of special case or families of second order diﬀerential equations which is results of the energy integral equation derivations as u − au du dt d2u d t2 =0 (A.3 Energy Equation ODE It is non–linear because the second derivative is square and the function multiply the second derivative.88) .290 APPENDIX A.86) A.83) results in 1 dv dt =v =⇒ = v dv a dt a (A.5.84) which can be solved with the previous methods.2. Bifurcation to two solutions leads t 1 du + c = v 2 =⇒ =± a 2 dt which can be integrated as u= ± a 2t + c1 dt = ± a 3 2t + c1 a 3 2 2t + c1 a (A.85) + c2 (A. End Solution A.

1999. Solving Third Order Linear Diﬀerential Equations in Terms of Second Order Equations Mark van Hoeij 8 “On 7 The .91) The solution is assumed to be of the form of est which general third order polonium. the general solution is depend on the solution of third order polonium.a) √ 3 u √ du 3 k1 − u3 √ = t + k2 2 u √ du 3 k − u3 √1 = t + k2 2 (1. Thus.c) A. the linear third-order diﬀerential equation” Springer Berlin Heidelberg. The general form for constant coeﬃcient is d3 u d2 u du +a 2 +b + cu = 0 3 dt dt dt (A.2. Solution to more complicate linear equations with non constant coeﬃcient (function of t) can be solved sometimes by Laplace transform or reduction of the equation to second order Olivier Vallee8 .IX.6 Third Order Diﬀerential Equation There are situations where ﬂuid mechanics7 leads to third order diﬀerential equation.b) (1. Example A.9: Show that the solution of u is − √ 3 d2 u d t2 + du dt 2 (A. This kind of diﬀerential equation has been studied in the last 30 years to some degree. ORDINARY DIFFERENTIAL EQUATIONS (ODE) after integration u du = k1 dt 291 (A.A.IX.90) +u=0 (1.89) Further rearrangement and integration leads to the solution which is u2 = t + k2 2 k1 For non–homogeneous equation they can be integrated as well.IX. Third unsteady energy equation in accelerated coordinate leads to a third order diﬀerential equation.2. The solution to constant coeﬃcients is relatively simple and will be presented here.

97) 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 (A. When the characteristic equation solution has three diﬀerent real roots the solution of the diﬀerential equation is u = c1 es1 t + c2 es2 t + c3 es3 t (A. if D > 0. Thus. From a mathematical point of view.98) (A. derivation of the leading equation (results of the ode) is reduced into quadratic equation and thus the same situation exist. all the roots are real and unequal. θ.101) . all the roots are real and at least two are identical. (A. For the case D = 0.92) (A.96) T = and where the D is deﬁned as 3 R− √ D (A. s3 + a1 s2 + a2 s + a3 = 0 The solution is 1 s1 = − a1 + (S + T ) 3 1 1 1 √ s2 = − a1 − (S + T ) + i 3(S − T ) 3 2 2 and 1 1 √ 1 s3 = − a1 − (S + T ) − i 3(S − T ) 3 2 2 Where S= 3 (A.292 APPENDIX A.100) Only three roots can exist for the Mach angle. one root is real and two roots are complex.94) (A. MATHEMATICS FOR FLUID MECHANICS order polonium has always one real solution.99) (A.93) (A.95) R+ √ D. In the last case where D < 0.

106) are the solutions of (A.106) and equation (A.2.102) Similarly derivations for the case of three identical real roots. the solution is u = (c1 sin b1 + c2 cos b1 ) ea1 t + c3 es3 t (A. For example.107) was discussed earlier.104).109) (A.107) (A. Thus the solution of d2 +1 u=0 dt2 with the solution of d2 −1 u=0 dt2 (A.103) Where a1 is the real part of the complex root and b1 imaginary part of the root. The solution of equation (A. the equation d4 u − u = 0 =⇒ dt4 can be written as combination of d2 −1 dt2 d2 +1 u=0 dt2 or d2 +1 dt2 d2 −1 u=0 dt2 (A. for constant coeﬃcients ODE can be transformed into multiplication of smaller order linear operations.2.A. Generally. ORDINARY DIFFERENTIAL EQUATIONS (ODE) In the case the solution to the characteristic has two identical real roots u = (c1 + c2 t) es1 t + c3 es2 t 293 (A.7 Forth and Higher Order ODE The ODE and partial diﬀerential equations (PDE) can be of any integer order. Sometimes the ODE is fourth order or higher the general solution is based in idea that equation is reduced into a lower order. The general procedure is based on the above concept but is some what simpler.104) The order of operation is irrelevant as shown in equation (A.105). For the case of only one real root.108) . A. Inserting es t into the ODE an u(n) + an−1 u(n−1) + an−2 u(n−2) + · · · + a1 u + a0 u = 0 yields characteristic equation an sn + an−1 sn−1 + an−2 sn−2 + · · · + a1 s + a0 = 0 (A.105) d4 −1 u=0 dt4 (A.

c) The roots are two pairs of complex numbers and one real number.4 s5 = = = 3 ± 3i 2±i 1 (1.X.a) (1.X. MATHEMATICS FOR FLUID MECHANICS If The Solution of Characteristic Equation all roots are real and diﬀerent e.X. sk+1 = sk+2 = sk+3 · · · = sn k/2 roots.g.g. roots are similar and some real and diﬀerent e. Thus the solution is u = c1 et + e2 t (c2 sin (t) + c3 cos (t)) + e3 t (c4 sin (3 t) + c5 cos (3 t)) End Solution (1. sk+1 = sk+2 = sk+3 · · · = sn k/2 roots. s1 = s2 = · · · = sk and some diﬀerent e.b) With the roots of the equation (1.g.d) . s1 = s2 = s3 = s4 · · · = sn all roots are real but some are identical e.X.10: Solve the ﬁfth order ODE d5 u d4 u d3 u d2 u du − 11 4 + 57 3 − 149 2 + 192 − 90 u = 0 5 dt dt dt dt dt Solution The characteristic equation is s5 − 11 s4 + 57 s3 − 149 s2 + 192 s − 90 = 0 (1.b) (these roots can be found using numerical methods or Descartes’ Rule) are s1. are pairs of conjugate complex numbers of si = ai ± bi and some real and diﬀerent e. are pairs of conjugate complex numbers of si = ai ± bi . sk+1 = sk+2 = sk+3 · · · = sn The Solution of Diﬀerential Equation Is u = c1 es1 t + c2 es2 t + · · · + cn esn t u = c1 + c2 t + · · · + ck tk−1 es1 t + ck+1 esk+1 t + ck+2 esk+2 t + · · · + cn esn t u = (cos(b1 t) + sin(b1 t)) ea1 t + · · · + (cos(bi t) + sin(bi t)) eai t + · · · + (cos(bk t) + sin(bk t)) eak t + ck+1 esk+1 t + ck+2 esk+2 t + · · · + cn esn t u = (cos(b1 t) + sin(b1 t)) ea1 t + · · · + (cos(bi t) + sin(bi t)) eai t + · · · + (cos(bk t) + sin(bk t)) eak t + ck+1 + ck+2 t + · · · + ck+ t −1 esk+1 t + ck+2 esk+2 t + ck+3 esk+3 t + · · · + cn esn t Example A.2 s3.294 APPENDIX A.g.X.g.

As one might expect PDE are harder to solve. sometimes similar deﬁnition can be applied for other order. The solution of elliptic equations depends on the boundary conditions The solution of parabolic equations depends on the boundary conditions but as well on the initial conditions. ux = ∂u . Partial diﬀerential equations are categorized by the order of highest derivative. the same can be done for a second-order PDE. Generally. PARTIAL DIFFERENTIAL EQUATIONS 295 A. the PDE solution is done by transforming the PDE to one or more ODE. Equation (A. The meaning for initial conditions is that of solution depends on some early points of the ﬂow (the solution). ayy might depend upon ”x” and ”y”.110) To be continue A. Note that ∂x partial derivative also include mix of derivatives such as ux y. Generally. ux .2. axy .) = 0 (A. The disxy criminant can be function of the x and y and thus can change sign and thus the characteristic of the equation. . The physical meaning of the these deﬁnition is that these equations have diﬀerent characterizations. The hyperbolic equations are associated with method of characteristics because physical situations depends only on the initial conditions. For example. . However. . Many situations in ﬂuid mechanics can be described by PDE equations.112) is similar to the equations for a conic geometry: axx x2 + axy x y + ayy y 2 + · · · = 0 (A. The general second-order PDE in two independent variables has the form axx uxx + 2axy uxy + ayy uyy + · · · = 0 (A. Example of such equation is F (ut .8 A general Form of the Homogeneous Equation dn u dn−1 u du + k1 tn−1 n−1 + · · · + kn−1 t + kn u = a x dtn dt dt The homogeneous equation can be generalized to k0 tn (A. One example of such equation is heat equation. this characterization is done for for second order. when the discriminant is zero the equation are called parabolic. When the discriminant . The nature of the solution is based whether the equation is elliptic parabolic and hyperbolic.111) Where subscripts refers to derivative based on it.113) In the same manner that conic geometry equations are classiﬁed are based on the discriminant a2 − 4 axx ayy .3.A.112) The coeﬃcients axx .3 Partial Diﬀerential Equations Partial Diﬀerential Equations (PDE) are diﬀerential equations which include function includes the partial derivatives of two or more variables. Normally.

In ﬂuid mechanics this kind equation appear in supersonic ﬂow or in supper critical ﬂow in open channel ﬂow. Hence. .116) Rearranging equation (A. the constant is denoted as λ and with this deﬁnition the PDE is reduced into two ODE.119) and (A. the physics dictates this character and will be used in the book.296 APPENDIX A.118) Equations (A.3..116) yields 1 ∂X(x) 1 ∂Y (y) + =0 X(x) ∂x Y (y) ∂y (A.115) The solution is assume to be u = Y (y) X(x) and substitute into the (A. However.114) The interpretation the equation characteristic is complicated.119) (A. The equations that not mentioned above are elliptic which appear in ideal ﬂow and subsonic ﬂow and sub critical open channel ﬂow. If it turned that assumption is wrong the ratio is not constant.115) results in Y (y) ∂X(x) ∂Y (y) + X(x) =0 ∂x ∂y (A. An example of ﬁrst order equation is ∂u ∂u + =0 ∂x ∂y (A. MATHEMATICS FOR FLUID MECHANICS is larger then zero the equation is referred as hyperbolic equations.117) can exist is that these two term equal to a constant.118) are ODE that can be solved with the methods described before for certain boundary condition.. A. ∂x ∂x (A.1 First-order equations First order equation can be written as u = ax ∂u ∂u + ay + .117) A possible way the equation (A. Is it possible that these terms not equal to a constant? The answer is no if the assumption of the solution is correct. The ﬁrst equation is X function 1 ∂X(x) =λ X(x) ∂x The second ODE is for Y 1 ∂Y (y) = −λ Y (y) ∂y (A.

b. .by ﬁnding the quadrant 2 2 α that lies in) 2 α 1 + cos α =± (same as above) 2 2 α 1 − cos α sin α 6. tan = = 2 sin α 1 + cos α for formulas 3-6. tan(α − β) = tan α + tan β 1 − tan α tan β tan α − tan β 1 + tan α tan β 1. c2 = a2 + b2 − 2 a b cos γ (Law of Cosines) 5. Area of triangle = 1 a b sin γ 2 6. -A.7. tan 2α = 1 − tan2 α 4. cos 2α = cos2 x − sin2 x = 2 cos2 x − 1 = 1 − 2 sin2 x 2 tan α 3. respectively 5. a+b+c (Heron’s Formula) where s = 2 β a c γ b α Fig. and c. and opposite angles α. sin(α − β) = sin α cos β − sin β cos α 3. consider the triangle with sides of length a. TRIGONOMETRY 297 A. Area of triangle = s(s − a)(s − b)(s − c). cos 1 − 2 cos(2α) 2 1 + 2 cos(2α) 2. cos2 α = 2 sin α sin β sin γ 3.4 Trigonometry These trigonometrical identities were set up by Keone Hon with slight modiﬁcation 1. The tringle angles sides. sin2 α = 4. cos(α + β) = cos α cos β − sin α sin β 4. = = (Law of Sines) a b c 1. sin α 1 − cos α =± (determine whether it is + or . β. and γ.A. cos(α − β) = cos α cos β + sin α sin β 5. sin 2α = 2 sin α cos α 2.4. sin(α + β) = sin α cos β + sin β cos α 2. tan(α + β) = 6.

MATHEMATICS FOR FLUID MECHANICS .298 APPENDIX A.

275 Deformable control volume. 77 Acceleration. 177 D D’Alembert paradox. 243. 164 Averaged kinetic energy. 99 buoyant force. 245 E Energy conservation. 62. 21 buoyancy. 90 Free expansion. 8. 276 Gravity varying Ideal gas. 11 Divergence Theorem. 230 Flow out tank. 257 C Co–current ﬂow. angular. 20. 12 Accelerated system. 261 Extended Open channel ﬂow. 171 Energy Equation Accelerated System. 185 Euler equations. 76 Conduction. 151 Average Velocity Integral Analysis. 193 dilettante. 187 Rotating Coordinate System. 61 Boundary Layer. 7. 162. 3. 172 Conservative force. 3 d‘Alembertian Operator. 59. 93 Archimedes. 187 Linear accelerate System. 67. 261 Cut–out shapes. 177 Counter–current Pulse ﬂow. 75 H Harmonic function. 187 Convection. 137 bulk modulus. 72–74 Fully ﬂuidized bed. 11. 124 Density ratio. 60. 242 Gauss-Ostrogradsky Theorem. 171 Fixed ﬂuidized bed. 97. 71. 79 Add Force. 240 Flow regimes in one pipe. 72. 11 Body force. 259 Annular ﬂow. 194 Add mass. 172 Convective acceleration. 177 Averaged momentum energy. 188 Energy equation Frictionless Flow. 74 Real gas. 185 Steady State. 108 Diﬀerential analysis. 62 Average Momentum. 105 F First Law of Thermodynamics. 74 eﬀective. 194 Add momentum. 260 Counter–current ﬂow. 72. 276 . 206 Correction factor. 174 Flow regime map. 3 Atmospheric pressure. 91 Cylindrical Coordinates. 148 B Bingham’s model. 244 Forces Curved surfaces. 71 Correction factor for averaged velocity. 195 G Gas–gas ﬂow. 193 External forces. 257 Flow ﬁrst mode. 276 Double choking phenomenon. 162. 243 Compressibility factor. 162 Arc shape.SUBJECTS INDEX 299 Subjects Index A absolute viscosity.

221 kinematic viscosity. 260 purely viscous ﬂuids. 134 Divergence Theorem. 206 Lockhart martinelli model. 185 . constant of integration. 283 Integral analysis big picture. 67 Return path for ﬂow regimes. 95 Pressure center. 94 I Ideal gas. 256 Mixed ﬂuidized bed. 74 Leibniz integral rule. 1. 147 Momentum conservation. 277 Oscillating manometer. 254 P Pendulum action. 62 Pneumatic conveying. 230 Polynomial function. 134 Limitation of the integral approach. 11 Pulse ﬂow. 8 No–slip condition. 124 Non–Linear Equations. 285 Normal stress. 69 Liquid–Liquid Regimes. 158 S Scalar function. 257 Momentum Conservation. 116. 222 Multi–phase ﬂow. 87 pseudoplastic. 248 Metacentric point. 86. MATHEMATICS FOR FLUID MECHANICS Moving surface.300 APPENDIX A. 276 Rocket mechanics. 213 N Navier-Stokes equations. 59. 77 Linear operations. 66 Mass velocity. 257 Poiseuille ﬂow. 183 Linear acceleration. 115. 74. 227 Concentric cylinders. 60. 219 Minimum velocity solid–liquid ﬂow. 90 Second Law of Thermodynamics. 172 Rayleigh–Taylor instability. 261 Horizontal ﬂow. 282 Liquid phase. 67 Inclined manometer. 242 Local acceleration. 245 Reynolds Transport Theorem. 194 Inverted manometer. 184 L Lapse rate. 66 Isotropic viscosity. 243 Hydrostatic pressure. 106 Micro ﬂuids. 142 Interfacial instability. 66 Initial condition. 142 small picture. 114 Piezometric pressure. 105. 246 horizontal counter–current ﬂow. 214 K Kinematic boundary condition. 149 index notation. 243 Orthogonal Coordinates. 221 Moving surface Free surface. 243 Real gas. 11 O Open channel ﬂow. 206 Momentum equation Accelerated system. 222 R Radiation. 211 Moving boundary. 105 Neutral stable. 11 M “Magniﬁcation factor”. 220 Non–deformable control volume. 239 Multiphase ﬂow against the gravity. 193 Neutral moment Zero moment. 117 Newtonian ﬂuids.

249 Void Fraction. 275 Westinghouse patent. 100 Turbomachinary. 62 Variables Separation 1st equation. 243 W Watson’s method. 290 shear stress. 218 Thermodynamical pressure. 205 Superﬁcial velocity. 242 T Tank emptying parameters.SUBJECTS INDEX Second viscosity coeﬃcient. 249 Quality of dryness. 72 Unsteady State Momentum. 157 V Vapor pressure. 207 substantial derivative. 207 symmetry. 272 Vertical counter–current ﬂow. 218 Segregated equations. 271 Vectors Algebra. 256 Thermal pressure. 261 Stable condition. 262 Slip velocity. 287 Vectors. 93. 210 transformation. 183 Total moment. 72. 97 cubic. 73. 255 Solid–solid ﬂow. 113 stratiﬁed ﬂow. 81 Transition to continuous. 76 Stability analysis. 208 Cartesian coordinates. 249 Wetness fraction. 242 Spherical coordinates. 6 Slip condition range. 249 Liquid holdup. 19 Wave Operator. 249 Reversal ﬂow. 181 Terminal velocity. 258 Solid–ﬂuid ﬂow. 72 stability analysis. 11 Torricelli’s equation. 221 Solid–ﬂuid ﬂow Gas dynamics aspects. 105 Stability in counter–current ﬂow. 147 Triangle shape. 208. 255 Solid–liquid ﬂow. 165 Two–Phase Gas superﬁcial velocity. 218 thixotropic. 12 301 U Unstable condition. 243 Stress tensor. 260 Vertical ﬂow. 244 Sutherland’s equation. 249 .

115 H Helmholtz. 4 K Kutta-Joukowski.. 115 Reynolds. 4 Mikhail Vasilievich Ostrogradsky. 134 Rose. Hermann von. 276 N Navier. 4 Martinelli. 4 E Evangelista Torricelli. 4 Blasiu.302 APPENDIX A. Claude–Louis. 4 de Saint Venant. 4 . 227 Poisson. MATHEMATICS FOR FLUID MECHANICS Authors Index B Bhuckingham. 193 e Duckler. 240 Meye. 4 Stokes. 276 T Taitle. 134 Lockhart. 193 Prandtl. 4 S Stanton. 4 D Darcy. 193 Nikuradse. 240. 183 F Fanning. 4 Blasius. 293 C Cichilli. 4 Nusselt. 4 Westinghouse.I. 253 R Rayleigh. 240 Taylor. Barr´. Ernst Kraft Wilhelm. 276 Poiseuille. 253 P Pierre-Simon Laplace. 4. G. Simon-Denis. 4 V von Karma. 231 O Olivier Vallee. 4 W Weisbach. 4 Gauss. Osborne. Carl Friedrich. George Gabriel. 240 M Manning. 193 G Ganguillet. 242 L Leibniz. Jean Louis. 4 Froude.

- Ashish Tewari Modern Control Design With MATLAB Org Www.eeeuniversity.com
- kalman.pdf
- RANSAC4Dummies.pdf
- Sisteme de Propulsie
- Understanding the Basis of the Kalman Filter
- M1 Montarea Si Utilizarea Instalatiilor de Joasa Tensiune
- 45127367 Manualul Instalatorului Sanitare 2010
- Masini Hidraulice
- Supernova Mecanica
- Tomescu_Ciohodaru
- Methode Numerique
- Cosmic Velocity
- Cosmic Velocities
- Cosmic Velocities
- Chimia Explozivilor Rotariu T 2010
- Hassel, Sven - Curtea Martiala
- Cadex Electronics - Batteries in a Portable World
- Suport Curs-Instalator ITSG
- Advances in Missile Guidance Theory
- Buzz Aldrin PHD Thesis
- Mathematical Modeling of Earth’s Magnetic Field
- Neural Network Based Edge Detection for Automated Medical Diagnos
- Atmel 2586 AVR 8 Bit Microcontroller ATtiny25 ATtiny45 ATtiny85 Datasheet
- Masini Electrice
- VOR Tutorial

Sign up to vote on this title

UsefulNot useful- Fluid Mechanics
- Mekanika Fluida
- Fluid Mechanics
- fluidMechanics
- Fluid Mechanics
- Fluid Mechanics
- Basics of FLUID MECHANICS
- Fluid Mechanics
- Fluid Mechanics
- Dieter A. Wolf-Gladrow- Lattice-Gas Cellular Automata and Lattice Boltzmann Models - An Introduction
- Optimal Control for Holonomic and Nonholonomic Mechanical
- Statistical mechanics lecture notes (2006), L4
- Nonholonomic Systems Caltech
- Level Set Methods - An Overview and Some Recent Results
- Derivations of Navier Stokes Equations
- Efect Non Zero
- Lattice Gas Cellular Automata and Lattice Boltzmann Models Chapter2
- BFM-Ch5
- Lecture 1
- Ucdavis Variation
- fea-bloch
- Lecture 3
- 4261 Advanced EM Theory Lecture notes 10 of 11
- Calculation of Densities From Cubic Equations Of
- CharacteristicsChapter
- Quantum Field Theory - J. Norbury 3
- Chaos&Dynamics
- g8 FeB 2 copies.docx
- An Drew Macro
- Nbm Sle Sim Naivegauss
- Fluid Mechanics

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd