Mechanical Engineer's Reference Book

Mechanical Engineer’s Reference Book

Twelfth Edition

Edward H. Smith, BSc, MSc, PhD, CEng, FIMechE

Head of Computing Services, University of Central Lancashire

Table of Contents

Cover image

Title page

Copyright

Preface

Contributors

Chapter 1: Mechanical engineering principles

Publisher Summary

1.1 Statics of rigid bodies

1.2 Strength of materials

1.3 Dynamics of rigid bodies

1.4 Vibrations

1.5 Mechanics of fluids

1.6 Principles of thermodynamics

1.7 Heat transfer

1.7.3 Analysis of heat transfer14–16

1.7.4 Use of computers

1.7.5 Heat transfer: nomenclature

Chapter 2: Electrical and electronics principles

Publisher Summary

2.1 Basic electrical technology

2.2 Electrical machines

2.3 Analogue and digital electronics theory

2.4 Electrical safety

Chapter 3: Microprocessors, instrumentation and control

Publisher Summary

3.1 Summary of number systems

3.2 Microprocessors

3.3 Communication standards

3.4 Interfacing of computers to systems

3.5 Instrumentation

3.6 Classical control theory and practice

3.7 Microprocessor-based control

3.8 Programmable logic controllers

3.9 The z-transform

3.10 State variable techniques

Chapter 4: Computers and their application

Publisher Summary

4.1 Introduction

4.2 Types of computer

4.3 Generations of digital computers

4.4 Digital computer systems

4.5 Categories of computer systems

4.6 Central processor unit

4.7 Memory

4.8 Peripherals

4.9 Output devices

4.10 Terminals

4.11 Direct input

4.12 Disk storage

4.13 Digital and analogue input/output

4.14 Data communications

4.15 Computer networks

4.16 Data terminal equipment

4.17 Software

4.18 Database management

4.19 Language translators

4.20 Languages

Chapter 5: Computer-integrated engineering systems

Publisher Summary

5.1 CAD/CAM: Computer-Aided Design and Computer-Aided Manufacturing

5.2 Industrial robotics and automation

5.3 Computer graphics systems

Acknowledgement

Appendix: Bresenham’s line algorithm

Chapter 6: Design standards

Publisher Summary

6.1 Standardization in design

6.2 Drawing and graphic communications

6.3 Fits, tolerances and limits

6.4 Fasteners

6.5 Ergonomic and anthropometric data

6.6 Total quality-a company culture

Chapter 7: Materials, properties and selection

Publisher Summary

7.1 Engineering properties of materials

7.2 The principles underlying materials selection

7.3 Ferrous metals

7.4 Non-ferrous metals

7.5 Composites

7.6 Polymers

Appendix: Worked examples of design of plastic components

7.7 Elastomers

7.8 Engineering ceramics and glasses

7.9 Corrosion

7.10 Non-destructive testing

Acknowledgements

Chapter 8: Mechanics of solids

Publisher Summary

8.1 Stress and strain

8.2 Experimental techniques

8.3 Fracture mechanics

8.4 Creep of materials

8.5 Fatgue

Chapter 9: Tribology

Publisher Summary

9.2 Lubricants (oils and greases)

9.3 Bearing selection

9.4 Principles and design of hydrodynamic bearings

9.5 Lubrication of industrial gears

9.6 Rolling element bearings

9.7 Materials for unlubricated sliding

9.9 Fretting

9.10 Surface topography

Chapter 10: Power units and transmission

Publisher Summary

10.1 Power units

10.2 Power transmissions

Acknowledgements

Chapter 11: Fuels and combustion

Publisher Summary

11.1 Introduction

11.2 General fuel types

11.3 Major property overview

11.4 Major fuel groupings

11.5 Combustion

11.6 Conclusions

Chapter 12: Alternative energy sources

Publisher Summary

12.1 Introduction

12.1.1 The Earth’s energy flows

12.1.2 Energy from solar radiation

12.1.3 Energy from the Earth’s interior

12.1.4 Energy from the tides

12.1.5 The energy flow system

12.2 Solar radiation

12.3 Passive solar design in the UK

12.4 Thermal power and other thermal applications

12.5 Photovoltaic energy conversion

12.6 Solar chemistry

12.7 Hydropower

12.8 Wind power

12.9 Geothermal energy

12.10 Tidal power

12.11 Wave power

12.12 Biomass and energy from wastes

12.13 Energy crops

Chapter 13: Nuclear engineering

Publisher Summary

13.1 Introduction

13.2 Nuclear radiations and energy

13.3 Mechanical engineering aspects of nuclear power stations and associated plant

13.4 Other applications of nuclear radiation

13.5 Elements of health physics and shielding

Chapter 14: Offshore engineering

Publisher Summary

14.1 Historical review

14.2 Types of fixed and floating structures

14.3 Future development

14.4 Hydrodynamic loading

14.5 Structural strength and fatigue

14.6 Dynamics of floating systems

14.7 Design considerations and certification

Chapter 15: Plant engineering

Publisher Summary

15.1 Compressors, Fans and Pumps

15.2 Seals and Sealing

Acknowledgements

15.3 Boilers and waste-heat recovery

15.4 Heating, ventilation and air conditioning

15.5 Refrigeration

15.6 Energy management

15.7 Condition monitoring

15.8 Vibration isolation and limits

15.9 Acoustic noise

Chapter 16: Manufacturing methods

Publisher Summary

16.1 Large-chip metal removal

16.2 Metal forming

16.3 Welding, soldering and brazing

16.4 Adhesives

16.5 Casting and foundry practice

Chapter 17: Engineering mathematics

Publisher Summary

17.1 Trigonometric functions and general formulae

17.2 Calculus

17.3 Series and Transforms

17.4 Matrices and Determinants

17.5 Differential Equations

17.6 Statistics

Chapter 18: Health and safety

Publisher Summary

18.1 Health and safety in the European Community

18.2 Health and safety at work-law and administration in the USA

18.3 UK legislation and guidance

18.4 The Health and Safety at Work etc. Act 1974

18.5 The Health and Safety Executive

18.6 Local Authorities

18.7 Enforcement Notices

18.8 Control of Substances Hazardous to Health Regulations 1988

18.9 Asbestos

18.10 Control of lead at work

18.11 The Electricity at Work Regulations 1989

18.12 The Noise at Work Regulations 1989

18.13 Safety of machinery

18.14 Personal protective equipment

18.15 Manual handling

Chapter 19: Units, symbols and constants

Publisher Summary

19.1 SI units

19.2 Conversion of existing imperial terms

19.3 Abbreviations

19.4 Physical and chemical constants

Index

Copyright

Butterworth-Heinemann Ltd

Linacre House, Jordan Hill, Oxford OX2 8DP

A member of the Reed Elsevier group

OXFORD LONDON BOSTON

MUNICH NEW DELHI SINGAPORE SYDNEY

TOKYO TORONTO WELLINGTON

First published as Newnes Engineer’s Reference Book 1946

Twelfth edition 1994

© Butterworth-Heinemann Ltd 1994

All rights reserved. No part of this publication may be reproduced in any material form (including photocopying or storing in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written permission of the copyright holder except in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London, England W1P 9HE. Applications for the copyright holder’s written permission to reproduce any part of this publication should be addressed to the publishers

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication Data

A catalogue record for this book is available from the Library of Congress

ISBN 0 7506 1195 2

Typeset by TecSet Ltd, Wallington, Surrey

Printed in Great Britain by The Bath Press, Avon

Preface

I was delighted when Butterworth-Heinemann asked me to edit a new edition of Mechanical Engineer’s Reference Book. Upon looking at its predecessor, it was clear that it had served the community well, but a major update was required. The book clearly needed to take account of modern methods and systems.

The philosophy behind the book is that it will provide a qualified engineer with sufficient information so that he or she can identify the basic principles of a subject and be directed to further reading if required. There is a blurred line between this set of information and a more detailed set from which design decisions are made. One of my most important tasks has been to define this distinction, so that the aims of the book are met and its weight is minimized! I hope I have been able to do this so that the information is neither cursory nor complex.

Any book of this size will inevitably contain errors, but I hope these will be minimal. I will be pleased to receive any information from readers so that the book can be improved.

To see this book in print is a considerable personal achievement, but I could not have done this without the help of others. First, I would like to thank all the authors for their tremendous hard work. It is a major task to prepare information for a book of this type, and they have all done a magnificent job. At Butterworth-Heinemann, Duncan Enright and Deena Burgess have been a great help, and Dal Koshal of the University of Brighton provided considerable support. At the University of Central Lancashire, Gill Cooke and Sue Wright ensured that the administration ran smoothly.

I hope you find the book useful.

Ted Smith,     University of Central Lancashire, Preston

Christmas Eve, 1993

Contributors

Dennis H. Bacon, BSc(Eng), MSc, CEng, MIMechE,     Consultant and technical author

Neal Barnes, BSc, PhD,     Formerly Manager, Pumping Technology, BHR Group Ltd

John Barron, BA, MA(Cantab),     Lecturer, Department of Engineering, University of Cambridge

Christopher Beards, BSc(Eng), PhD, CEng, MRAeS, MIOA,     Consultant and technical author

Jonh S. Bevan, IEng, MIPlantE, ACIBSE,     Formerly with British Telecom

Ronald J. Blaen,     Independent consultant

Tadeusz Z. Blazynski, PhD, BSc(Eng), MIMechE, CEng,     Formerly Reader in Applied Plasticity, Department of Mechanical Engineering, University of Leeds

James Carvill, BSc(MechE), BSc(ElecEng),     Formerly Senior Lecturer in Mechanical Engineering, University of Northumbria at Newcastle

Trevor G. Clarkson, BSc(Eng), PhD, CEng, MIEE, Senior Member IEEE,     Department of Electronic and Electrical Engineering, King’s College, University of London

Paul Compton, BSc CEng, MCIBSE,     Colt International Ltd, Havant, Hants

Vince Coveney, PhD,     Senior Lecturer, Faculty of Engineering, University of the West of England

Roy D. Cullum, FIED,     Editor, Materials and Manufacture

A. Davies,     National Centre of Tribology, Risley Nuclear Development Laboratory

Raymond J.H. Easton, CEng, MIMechE,     Chief Applications Engineer, James Walker & Co Ltd

Philip Eliades, BSc, AMIMechE,     National Centre for Tribology, UKAEA, Risley, Warrington

Duncan S.T. Enright, BA, MA(Oxon), CertEd,     GradlnstP Commissioning Editor, Butterworth-Heinemann, Oxford

Charles J. Fraser, BSc, PhD, CEng, FIMechE,     MInstPet Reader in Mechanical Engineering

Eric M. Goodger, BSc(Eng), MSc, PhD, CEng, MIMechE, FInstE, FInstPet, MRAeS, MIEAust,     Consultant in Fuels Technology Training

Edward N. Gregory, CEng, FIM, FWeldl,     Consultant

Dennis R. Hatton, IEng, MIPlantE,     Consultant

Tony G. Herraty, BTech, MIMechE, CEng,     SKF (UK) Service Ltd, Luton, Bedfordshire

Martin Hodskinson, BSc, PhD, CEng, FIMechE, MIED, REngDes,     Senior Lecturer, Department of Engineering and Product Design, University of Central Lancashire

Allan R. Hutchinson, BSc, PhD, CEng, MICE,     Deputy Head, Joining Technology Research Centre, School of Engineering, Oxford Brookes University

Jeffery D. Lewins, DSc(Eng), FINucE, CEng,     Lecturer in Nuclear Engineering, University of Cambridge and Director of Studies in Engineering and Management, Magdalene College

Michael W.J. Lewis, BSc, MSc,     Senior Engineering Consultant, National Centre of Tribology, AE Technology, Risley, Warrington

R.Ken Livesley, MA, PhD, MBCS,     Lecturer Department of Engineering, University of Cambridge

J. Cleland McVeigh, MA, MSc, PhD, CEng, FIMechE, FInstE, MIEE, MCIBSE,     Visiting Professor, School of Engineering, Glasgow Caledonian University

Gordon M. Mair, BSc, DMS, CEng, MIEE, MIMgt,     Lecturer, Department of Design, Manufacture and Engineering Management, University of Strathclyde

Fraidoon Mazda, MPhil, DFH, DMS, MIMgt, CEng, FIEE,     Northern Telecom

Bert Middlebrook,     Consultant

John S. Milne, BSc, CEng, FIMechE,     Professor, Department of Mechanical Engineering, Dundee Institute of Technology

Peter Myler, BSc, MSc, PhD, CEng, MIMech,     Principal Lecturer, School of Engineering, Bolton Institute

Ben Noltingk, BSc, PhD, CPhys, FInstP, CEng, FIEE,     Consultant

Robert Paine, BSc, MSc,     Department of Engineering and Product Design, University of Central Lancashire

John R. Painter, BSc(Eng), CEng, MRAes, CDipAF,     Independent consultant (CAD/CAM)

Minoo H. Patel, BSc(Eng), PhD, CEng, FIMechE, FRINA,     Kennedy Professor of Mechanical Engineering and Head of Department, University College, London

George E. Pritchard, CEng, FCIBSE, FInst, FIPlantE,     Consulting engineer

Donald B. Richardson, MPhil, DIC, CEng, FIMechE, FIEE,     Lecturer, Department of Mechanical and Manufacturing Engineering, University of Brighton

Carl Riddiford, MSc,     Senior Technologist, MRPRA, Hertford

Ian Robertson, MBCS,     Change Management Consulatnt, Digital Equipment Corporation

Roy Sharpe, BSc, CEng, FIM, FInstP, FIQA, HonFInstNDT,     Formerly Head of National Nondestructive Testing Centre, Harwell

Ian Sherrington, BSc, PhD, CPhys, CEng, MInstP,     Reader in Tribology, department of Engineering, and Product Design, University of Central Lancashire

Edward H. Smith, BSc, MSc, PhD, CEng, FIMechE,     Head of Computing Services, University of Central Lancashire

Keith T. Stevens, BSc(Phy),     Principle scientist

Peter Tucker, BSc(Tech), MSc, CEng, MIMechE,     Formerly Principal Lecturer, Department of Mechanical and Production Engineering, Preston Polytechnic

Robert K. Turton, BSc(Eng), CEng, MIMechE,     Senior Lecturer in Mechanical Engineering, Loughborough University of Technology and Visiting Fellow, Cranfield University

Ernie Walker, BSc CEng, MIMechE,     Formerly Chief Thermal Engineer, Thermal Engineering Ltd

Roger C. Webster, BSc, MIEH,     Roger Webster & Associates, West Bridgford, Nottingham

John Weston-Hays,     Managing Director, Noble Weston Hays Technical Services Ltd, Dorking, Surrey

Leslie M. Wyatt, FIM, CEng,     Independent consultant and technical author

1

Mechanical engineering principles

Robert Paine, (Sections 1.1–1.4.2), Christopher Beards , (Section 1.4.3), Peter Tucker, (Section 1.5) and Dennis H. Bacon, (Sections 1.6 and 1.7)

Publisher Summary

This chapter discusses the mechanical engineering principles of materials. The study of mechanics may be divided into two distinct areas. These are static, which involves the study of bodies at rest, and dynamic, which is the study of bodies in motion. When a set of forces act on a body they give rise to a resultant force or moment or a combination of both. The situation may be determined by considering three mutually perpendicular directions on the free body diagram and resolving the forces and moment in these directions. When using analysis, the moment of each element of weight within the body, about a fixed axis, is equal to the moment of the complete weight about that axis. Circular motion is a special case of curvilinear motion in which the radius of rotation remains constant. In this case, there is acceleration toward the center of ω²r. This gives rise to a force toward the center known as the centripetal force. This force is reacted to by the centrifugal reaction. The principle of balancing is that by the addition of extra masses to the system the out-of-balance forces may be reduced or eliminated.

Contents

Statics of rigid bodies

Strength of materials

Dynamics of rigid bodies

Basic definitions

Linear and angular motion in two dimensions

Circular motion

Linear and angular motion in three dimensions

Balancing

Balancing of rotating masses

Vibrations

Single-degree-of-freedom systems

Multi-degree-of-freedom systems

Further reading

British Standards

Random vibrations

Further reading

Mechanics of fluids

Introduction

Fluid statics

Fluid flow

Flow measurement

Open-channel flow

Boundary layer flow

Pressure transients (water hammer)

Gas flow

Ideal fluid flow

Conclusion

Further reading

Principles of thermodynamics

Introduction

The laws of thermodynamics

Thermoeconomics

Work, heat, property values, process laws and combustion

Cycle analysis

Heat transfer

Introduction

Basic principles of heat transfer

Analysis of heat transfer

Use of computers

Heat transfer: nomenclature

References

In general, the study of mechanics may be divided into two distinct areas. These are statics, which involves the study of bodies at rest, and dynamics, which is the study of bodies in motion. In each case it is important to select an appropriate mathematical model from which a ‘free body diagram’ may be drawn, representing the system in space, with all the relevant forces acting on that system.

1.1 Statics of rigid bodies

When a set of forces act on a body they give rise to a resultant force or moment or a combination of both. The situation may be determined by considering three mutually perpendicular directions on the ‘free body diagram’ and resolving the forces and moment in these directions. If the three directions are denoted by x, y and z then the sum of forces may be represented by ∑Fx, ∑Fy and ∑Fz and the sum of the moments about respective axes by ∑Mx, ∑My and ∑Mz. Then for equilibrium the following conditions must hold:

(1.1)

(1.2)

If the conditions in equations (1.1) and (1.2) are not satisfied then there is a resultant force or moment, which is given by

The six conditions given in equations (1.1) and (1.2) satisfy problems in three dimensions. If one of these dimensions is not present (say, the z direction) the system reduces to a set of coplanar forces, and then

are automatically satisfied, and the necessary conditions of equilibrium in a two-dimensional system are

(1.3)

If the conditions in equation (1.3) are not satisfied then the resultant force or moment is given by

The above equations give solutions to what are said to be ‘statically determinate’ systems. These are systems where there are the minimum number of constraints to maintain equilibrium.¹

1.2 Strength of materials

Weight: The weight (W) of a body is that force exerted due to gravitational attraction on the mass (m) of the body: W = mg, where g is the acceleration due to gravity.

Centre of gravity: This is a point, which may or may not be within the body, at which the total weight of the body may be considered to act as a single force. The position of the centre of gravity may be found experimentally or by analysis. When using analysis the moment of each element of weight, within the body, about a fixed axis is equated to the moment of the complete weight about that axis:

where δm are the positions of the centres of gravity from these axes. Table 1.1 shows the position of the centre of gravity for some standard shapes. (See reference 2 for a more comprehensive list.)

Table 1.1

Centres of gravity and moments of inertia or second moments of area for two-dimensional figures

Shear force and bending moment: If a beam subject to loading, as shown in Figure 1.1, is cut, then in order to maintain equilibrium a shear force (Q) and a bending moment (M) must be applied to each portion of the beam. The magnitudes of Q and M vary with the type of loading and the position along the beam and are directly related to the stresses and deflections in the beam.

Figure 1.1

Relationship between shear force and bending moment If an element of a beam is subjected to a load w then the following relationship holds:

Table 1.2 shows examples of bending moments, shear force and maximum deflection for standard beams.

Table 1.2

Bending equation If a beam has two axes of symmetry in the xy plane then the following equation holds:

where Mz is the bending moment, Rz is the radius of curvature, Iz the moment of inertia, E the modulus of elasticity, y the distance from the principal axis and a is the stress.

Torsion equation: If a circular shaft is subject to a torque (T

where J is the polar second moment of area, G the shear modulus, L the length, θ the angle of twist, τ the shear stress and r the radius of the shaft.

1.3 Dynamics of rigid bodies

1.3.1 Basic definitions

1.3.1.1 Newton’s Laws of Motion

First Law A particle remains at rest or continues to move in a straight line with a constant velocity unless acted on by an external force.

Second Law The sum of all the external forces acting on a particle is proportional to the rate of change of momentum.

Third Law The forces of action and reaction between interacting bodies are equal in magnitude and opposite in direction.

Newton’s law of gravitation, which governs the mutual interaction between bodies, states

where F is the mutual force of attraction, G is a universal constant called the constant of gravitation which has a value 6.673 × 10−11 m³ kg−1 s−2, m1 and m2 are the masses of the two bodies and x is the distance between the centres of the bodies.

Mass (m) is a measure of the amount of matter present in a body.

Velocity is the rate of change of distance (x) with time (t):

Acceleration is the rate of change of velocity (v) with time (t):

Momentum is the product of the mass and the velocity. If no external forces are present then the momentum of any system remains constant. This is known as the Conservation of Momentum.

Force is equal to the rate of change of momentum (mv) with time (t):

F = d(mv)/dt

F = m · dv/dt + v · dm/dt

If the mass remains constant then this simplifies to F = m · dv/dt, i.e. Force = mass × acceleration, and it is measured in Newtons.

Impulse (I) is the product of the force and the time that force acts. Since I = Ft = mat = m(v2 − v1), impulse is also said to be the change in momentum.

Energy There are several different forms of energy which may exist in a system. These may be converted from one type to another but they can never be destroyed. Energy is measured in Joules.

Potential energy (PE) is the energy which a body possesses by virtue of its position in relation to other bodies: PE = mgh, where h is the distance above some fixed datum and g is the acceleration due to gravity.

Kinetic energy (KE) is the energy a body possesses by virtue of its motion:

Work (W) is a measure of the amount of energy produced when a force moves a body a given distance: W = F · x.

Power (P) is the rate of doing work with respect to time and is measured in watts.

Moment of inertia (I): The moment of inertia is that property in a rotational system which may be considered equivalent to the mass in a translational system. It is defined about an axis xx as

where x is the perpendicular distance of an element of mass δm from the axis xx and kXX is the radius of gyration about the axis xx. Table 1.1 gives some data on moments of inertia for standard shapes.

Angular velocity (ω) is the rate of change of angular distance (θ) with time: =

Angular acceleration (α) is the rate of change of angular velocity (ω) with time:

Both angular velocity and accleration are related to linear motion by the equations v = ωx and a = αx (see Figure 1.2).

Figure 1.2

Torque (T) is the moment of force about the axis of rotation:

T= I0α

A torque may also be equal to a couple, which is two forces equal in magnitude acting some distance apart in opposite directions.

Parallel axis theorem: if IGG is the moment of inertia of a body of mass m about its centre of gravity, then the moment of inertia (I) about some other axis parallel to the original axis is given by I = IGG + mr², where r is the perpendicular distance between the parallel axes.

Perpendicular axis theorem. If IXX, IYY and IZZ represent the moments of inertia about three mutually perpendicular axes x, y and z for a plane figure in the xy plane (see Figure 1.3) then IZZ = IXX + IYY.

Figure 1.3

Angular momentum (HO) of a body about a point O is the moment of the linear momentum about that point and is ωIOO. The angular momentum of a system remains constant unless acted on by an external torque.

Angular impulse is the produce of torque by time, i.e. angular impulse = Tt = Iα · t = I(ω2 − ω1), the change in angular momentum.

Angular kinetic energy about an axis O is given by

Work done due to a torque is the product of torque by angular distance and is given by Tθ.

Power due to torque is the rate of angular work with respect to time and is given by Tdθ/dt = Tω).

Friction: Whenever two surfaces, which remain in contact, move one relative to the other there is a force which acts tangentially to the surfaces so as to oppose motion. This is known as the force of friction. The magnitude of this force is μR, where R is the normal reaction and μ is a constant known as the coefficient of friction. The coefficient of friction depends on the nature of the surfaces in contact.

1.3.2 Linear and angular motion in two dimensions

Constant acceleration: If the accleration is integrated twice and the relevant initial conditions are used, then the following equations hold:

Variable acceleration: If the acceleration is a function of time then the area under the acceleration time curve represents the change in velocity. If the acceleration is a function of displacement then the area under the acceleration distance curve represents half the difference of the square of the velocities (see Figure 1.4).

Figure 1.4

Curvilinear motion is when both linear and angular motions are present.

If a particle has a velocity v and an acceleration a then its motion may be described in the following ways:

1. Cartesian components which represent the velocity and acceleration along two mutually perpendicular axes x and y (see Figure 1.5(a)):

Figure 1.5

2. Normal and tangential components: see Figure 1.5(b):

3. Polar coordinates: see Figure 1.5(c):

1.3.3 Circular motion

Circular motion is a special case of curvilinear motion in which the radius of rotation remains constant. In this case there is an acceleration towards the cente of ω²r. This gives rise to a force towards the centre known as the centripetal force. This force is reacted to by what is called the centrifugal reaction.

Velocity and acceleration in mechanisms: A simple approach to determine the velocity and acceleration of a mechanism at a point in time is to draw velocity and acceleration vector diagrams.

Velocities: If in a rigid link AB of length l the end A is moving with a different velocity to the end B, then the velocity of A relative to B is in a direction perpendicular to AB (see Figure 1.6).

Figure 1.6

When a block slides on a rotating link the velocity is made up of two components, one being the velocity of the block relative to the link and the other the velocity of the link.

Accelerations: If the link has an angular acceleration α then there will be two components of acceleration in the diagram, a tangential component αl and a centripetal component of magnitude ω²l acting towards A.

When a block slides on a rotating link the total acceleration is composed of four parts: first, the centripetal acceleration towards O of magnitude ω²l; second, the tangential acceleration αl; third, the acceleration of the block relative to the link; fourth, a tangential acceleration of magnitude 2vω known as Coriolis acceleration. The direction of Coriolis acceleration is determined by rotating the sliding velocity vector through 90° in the direction of the link angular velocity ω.

1.3.4 Linear and angular motion in three dimensions

1.3.4.1 Motion of a particle in a moving coordinate system

xyz is a moving coordinate system, with its origin at O which has a position vector Rand an angular velocity vector ω relative to a fixed coordinate system XYZ, origin at O’. Then the motion of a point P whose position vector relative to O is ρ and relative to O’ is r is given by the following equations (see Figure 1.7):

Figure 1.7

where ρr is the velocity of the point P relative to the moving system xyz and ω × ρ is the vector product of ω and ρ:

where ρr is the sum of:

2. The absolute velocity R of the moving origin O;

3. The velocity ω × ρ due to the angular velocity of the moving axes xyz.

is the sum of:

of the moving origin O;

due to the angular acceleration of the moving axes xyz;

4. The centripetal acceleration ω × (ω × ρ) due to the angular velocity of the moving axes xyz;

5. Coriolis component acceleration 2ω × ρr due to the interaction of coordinate angular velocity and relative velocity.

In all the vector notation a right-handed set of coordinate axes and the right-hand screw rule is used.

1.3.4.2 Gyroscopic effects

Consider a rotor which spins about its geometric axis (see . When this is observed it is the effect of gyroscopic reaction torque that is seen, which is in the opposite direction to the gyroscopic torque.⁴

Figure 1.8

1.3.5 Balancing

In any rotational or reciprocating machine where accelerations are present, unbalanced forces can lead to high stresses and vibrations. The principle of balancing is such that by the addition of extra masses to the system the out-of-balance forces may be reduced or eliminated.

1.3.6 Balancing of rotating masses

1.3.6.1 Single out-of-balance mass

One mass (m) at a distance r from the centre of rotation and rotating at a constant angular velocity ω produces a force mω²r. This can be balanced by a mass M placed diametrically opposite at a distance R, such that MR = mr

1.3.6.2 Several out-of-balance masses in one transverse plane

If a number of masses (m1, m2, …) are at radii (r1, r2, …) and angles (θ1 θ2, …) (see Figure 1.9) then the balancing mass M must be placed at a radius R such that MR is the vector sum of all the mr terms.

Figure 1.9

1.3.6.3 Masses in different transverse planes

If the balancing mass in the case of a single out-of-balance mass were placed in a different plane then the centrifugal force would be balanced. This is known as static balancing. However, the moment of the balancing mass about the original plane would lead to what is known as dynamic unbalance.

To overcome this, the vector sum of all the moments about the reference plane must also be zero. In general, this requires two masses placed in convenient planes (see Figure 1.10).

Figure 1.10

1.3.6.4 Balancing of reciprocating masses in single-cylinder machines

The acceleration of a piston as shown in Figure 1.11 may be represented by the equation⁵

Figure 1.11 *This equation forms an infinite series in which higher terms are small and they may be ignored for practical situations.

where n = 1/r. If n is large then the equation may be simplified and the force given by

The term mω2rcos θ is known as the primary force and (l/n)mω2rcos 2θ as the secondary force. Partial primary balance is achieved in a single-cylinder machine by an extra mass M at a radius R rotating at the crankshaft speed. Partial secondary balance could be achieved by a mass rotating at 2ω. As this is not practical this is not attempted. When partial primary balance is attempted a transverse component Mω2Rsin θ is introduced. The values of M and R are chosen to produce a compromise between the reciprocating and the transverse components.

1.3.6.5 Balancing of reciprocating masses in multi-cylinder machines

When considering multi-cylinder machines account must be taken of the force produced by each cylinder and the moment of that force about some datum. The conditions for primary balance are

and

The addition of extra masses to give secondary balance is not attempted in practical situations.

1.4 Vibrations

1.4.1 Single-degree-of-freedom systems

The term degrees of freedom in an elastic vibrating system is the number of parameters required to define the configuration of the system. To analyse a vibrating system a mathematical model is constructed, which consists of springs and masses for linear vibrations. The type of analysis then used depends on the complexity of the model.

Rayleigh’s method: Rayleigh showed that if a reasonable deflection curve is assumed for a vibrating system, then by considering the kinetic and potential energies* an estimate to the first natural frequency could be found. If an inaccurate curve is used then the system is subject to constraints to vibrate it in this unreal form, and this implies extra stiffness such that the natural frequency found will always be high. If the exact deflection curve is used then the natural frequency will be exact.

1.4.1.1 Transverse vibration of beams

Consider a beam of length (l), weight per unit length (w), modulus (E) and moment of inertia (I). Then its equation of motion is given by

where ω is the natural frequency. The general solution of this equation is given by

y = Acos βx + B sin βx + C cosh βx D sinhβx

where β⁴ = wω²/gEI.

The four constants of integration A, B, C and D are determined by four independent end conditions. In the solution trigonometrical identities are formed in β which may be solved graphically, and each solution corresponds to a natural frequency of vibration. Table 1.3 shows the solutions and frequencies for standard beams.⁶

Table 1.3

Dunkerley’s empirical method is used for beams with multiple loads. In this method the natural frequency (f1) is found due to just one of the loads, the rest being ignored. This is repeated for each load in turn and then the natural frequency of vibration of the beam due to its weight alone is found (f0).

* Consider the equation of motion for an undamped system (Figure 1.13):

Figure 1.13

(1.4)

but

Therefore equation (1.4) becomes

Integrating gives

the term ½m(dx/dt)² represents the kinetic energy and 1½kx² the potential energy.

Then the natural frequency of vibration of the complete system (f) is given by

(see reference 7 for a more detailed explanation).

Whirling of shafts : If the speed of a shaft or rotor is slowly increased from rest there will be a speed where the deflection increases suddenly. This phenomenon is known as whirling. Consider a shaft with a rotor of mass m such that the centre of gravity is eccentric by an amount e. If the shaft now rotates at an angular velocity ω then the shaft will deflect by an amount y due to the centrifugal reaction (see Figure 1.12). Then

Figure 1.12

mω²(y + e) = ky

where k is the stiffness of the shaft. Therefore

When (k/mω²) = 1, y is then infinite and the shaft is said to be at its critical whirling speed ωc. At any other angular velocity ω the deflection y is given by

When ω < ωc, y is the same sign as e and as ω increases towards ωc the deflection theoretically approaches infinity. When ω > ωc, y is opposite in sign to e and will eventually tend to –e. This is a desirable running condition with the centre of gravity of the rotor mass on the static deflection curve. Care must be taken not to increase ω too high as ω might start to approach one of the higher modes of vibration.⁸

Torsional vibrations: The following section deals with transverse vibrating systems with displacements x and masses m. The same equations may be used for torsional vibrating systems by replacing x by θ the angular displacement and m by I, the moment of inertia.

1.4.1.2 Undamped free vibrations

, where m is the mass, k , which is the natural frequency of vibration of the system (see Figure 1.13). The solution to this equation is given by

x = A sin(ωnt + α)

where A and a are constants which depend on the initial conditions. This motion is said to be simple harmonic with a time period T = 2π/ωn.

1.4.1.3 Damped free vibrations

+ cx˙ + kx = 0 (see . The solution to this equation and the resulting motion depends on the amount of damping. If c > 2mωn the system is said to be overdamped. It will respond to a disturbance by slowly returning to its equilibrium posi tion. The time taken to return to this position depends on the degree of damping (see Figure 1.15(c)). If c = 2mωn the system is said to be critically damped. In this case it will respond to a disturbance by returning to its equilibrium position in the shortest possible time. In this case (see Figure 1.15(b))

Figure 1.14

Figure 1.15

x = e−(c/2m)t(A+Bt)

where A and B are constants. If c < 2mωn the system has a transient oscillatory motion given by

where C and D are constants. The period

(see Figure 1.15(a)).

1.4.1.4 Logarithmic decrement

A way to determine the amount of damping in a system is to measure the rate of decay of successive oscillations. This is expressed by a term called the logarithmic decrement (δ), which is defined as the natural logarithm of the ratio of any two successive amplitudes (see Figure 1.16):

Figure 1.16

δ = loge(x1/x2)

where x is given by

Therefore

where τ is the period of damped oscillation. If the amount of damping present is small compared to the critical damping, τ approximates to 2π/ω, and then

δ = cπ/mωn

1.4.1.5 Forced undamped vibrations

The equation of motion is given by (see Figure 1.17)

Figure 1.17

or

The solution to this equation is

where ω is the frequency of the forced vibration. The first two terms of the solution are the transient terms which die out, leaving an oscillation at the forcing frequency of amplitude

or

The term

is known as the dynamic magnifier and it gives the ratio of the amplitude of the vibration to the static deflection under the load F0. When ω = ωn the amplitude becomes infinite and resonance is said to occur.

1.4.1.6 Forced damped vibrations

The equation of motion is given by (see + kx = F0 sin ωt

or

The solution to this equation is in two parts: a transient part as in the undamped case which dies away, leaving a sustained vibration at the forcing frequency given by

The term

ωn. As the damping is increased the value of ω for which resonance occurs is reduced. There is also a phase shift as CD increases tending to a maximum of π radians. It can be seen in Figure 1.18(a) that when the forcing frequency is high compared to the natural frequency the amplitude of vibration is minimized.

Figure 1.18

1.4.1.7 Forced damped vibrations due to reciprocating or rotating unbalance

Figure 1.19 shows two elastically mounted systems, (a) with the excitation supplied by the reciprocating motion of a piston, and (b) by the rotation of an unbalanced rotor. In each case the equation of motion is given by

Figure 1.19

The solution of this equation is a sinusoid whose amplitude, X, is given by

In representing this information graphically it is convenient to plot MX/me against ω/ωn for various levels of damping (see Figure 1.20(a)). From this figure it can be seen that for small values of ω the displacement is small, and as ω is increased the displacement reaches a maximum when ω is slightly greater than ωn. As ω is further increased the displacement tends to a constant value such that the centre of gravity of the total mass M remains stationary. Figure 1.20(b) shows how the phase angle varies with frequency.

Figure 1.20

1.4.1.8 Forced damped vibration due to seismic excitation

If a system as shown in Figure 1.21 has a sinusoidal displacement applied to its base of amplitude, y, then the equation of motion becomes

Figure 1.21

The solution of this equation yields

where x is the amplitude of motion of the system.

When this information is plotted as in Figure 1.22 it can be seen that for very small values of ω the output amplitude X is equal to the input amplitude Ythe curves intersect and the effect of damping is reversed.

Figure 1.22

The curves in Figure 1.22 may also be used to determine the amount of sinusoidal force transmitted through the springs and dampers to the supports, i.e. the axis (X/Y) may be replaced by (Ft/F0) where F0 is the amplitude of applied force and Ft is the amplitude of force transmitted.

1.4.2 Multi-degree-of-freedom systems

1.4.2.1 Normal mode vibration

The fundamental techniques used in modelling multi-degree-of-freedom systems may be demonstrated by considering a simple two-degree-of-freedom system as shown in Figure 1.23. The equations of motion for this system are given by

Figure 1.23

or in matrix form:

Assuming the motion of every point in the system to be harmonic then the solutions will take the form

x1 = A1 sin ωt

x2 = A2 sin ωt

whereA1 and A2 into the original equations the values of the natural frequencies of vibration may be found along with the appropriate mode shapes. This is a slow and tedious process, especially for systems with large numbers of degrees of freedom, and is best performed by a computer program.

1.4.2.2 The Holtzer method

When only one degree of freedom is associated with each mass in a multi-mass system then a solution can be found by proceeding numerically from one end of the system to the other. If the system is being forced to vibrate at a particular frequency then there must be a specific external force to produce this situation. A frequency and a unit deflection is assumed at the first mass and from this the inertia and spring forces are calculated at the second mass. This process is repeated until the force at the final mass is found. If this force is zero then the assumed frequency is a natural frequency. Computer analysis is most suitable for solving problems of this type.

Consider several springs and masses as shown in Figure 1.24. Then with a unit deflection at the mass m1 and an assumed frequency ω there will be an inertia force of m1ω² acting on the spring with stiffness k1 This causes a deflection of m1ω²/k1, but if m2 has moved a distance x2 then m1ω²/ k1 = 1 − x2 or x2 = 1 − m1ω²/k1. The inertia force acting due to m2 is m2ω²x2, thus giving the total force acting on the spring of stiffness k2 as [m1ω² + m2ω²x2]/k2. Hence the displacement at x3 can be found and the procedure repeated. The external force acting on the final mass is then given by

Figure 1.24

If this force is zero then the assumed frequency is a natural one.

British Standards

BS 3318: Locating the centre of gravity of earth moving equipment and heavy objects

BS 3851: 1982: Glossary of terms used in mechanical balancing of rotary machines

BS 3852: 1986: Dynamic balancing machines

BS 4675: 1986: Mechanical vibrations in rotating and reciprocating machinery

BS 6414: 1983: Methods for specifying characteristics of vibration and shock absorbers

1.4.3 Random vibrations

1.4.3.1 Introduction

If the vibration response parameters of a dynamic system are accurately known as functions of time, the vibration is said to be deterministic. However, in many systems and processes responses cannot be accurately predicted; these are called random processes. Examples of a random process are turbulence, fatigue, the meshing of imperfect gears, surface irregularities, the motion of a car running along a rough road and building vibration excited by an earthquake (Figure 1.25).

Figure 1.25 Example random process variable as f(t)

A collection of sample functions x1(t), x2(t), x3(t),…, xn(t) which make up the random process x(t) is called an ensemble (Figure 1.26). These functions may comprise, for example, records of pressure fluctuations or vibration levels, taken under the same conditions but at different times.

Figure 1.26 Ensemble of a random process

Any quantity which cannot be precisely predicted is non-deterministic and is known as a random variable or a probabilistic quantity. That is, if a series of tests are conducted to find

the value of a particular parameter, x, and that value is found to vary in an unpredictable way that is not a function of any other parameter, then x is a random variable.

1.4.3.2 Probability distribution

If n experimental values of a variable x are x1, x2, x3, …, xn, the probability that the value of x will be less than x′ is n′/n, where n′ is the number of x values which are less than or equal to x′. That is,

Prob(x ≤ x′) = n′/n

When n approaches ∞ this expression is the probability distribution function of x, denoted by P(x), so that

The typical variation of P(x) with x is shown in Figure 1.27. Since x(t) denotes a physical quantity,

Figure 1.27 Probability distribution function as f(x)

Prob(x < −∞) = 0, and Prob(x < +∞) = l

The probability density function is the derivative of P(x) with respect to x and this is denoted by p(x). That is,

where P(x + Δx) − P(x) is the probability that the value of x(t) will lie between x and x + Δx (Figure 1.27). Now

so that

Hence

so that the area under the probability density function curve is unity.

A random process is stationary if the joint probability density

p(x(t1), x(t2), x(t3), …)

depends only on the time differences t2 − t1, t3 − t2 and so on, and not on the actual time instants. That is, the ensemble will look just the same if the time origin is changed. A random process is ergodic if every sample function is typical of the entire group.

The expected value of f (x), which is written

is

so that the expected value of a stationary random process x(t) is

E[x(t1)] = E[x(t1 + t2)]

for any value of t.

If f(x) = x, the expected value or mean value of x, E[x] , is

In addition, if f(x) = x², mean square value of x, x¯² is

The variance of x, σ² is the mean square value of x about the mean, that is,

σ is the standard deviation of x, hence

Variance = (Standard deviation)²

= {Mean square − (Mean)²}

If two or more random variables x1 and x2 represent a random process at two different instants of time, then

and if t1 and t2 are the two instants of time,

E[x(t1,x(t2)] = R(t1,t2)

which is the autocorrelation function for the random process (Figure 1.28). For random processes which are stationary,

Figure 1.28 Random processes

E[x(t1), x(t2)] = R(t1, t2) = R(t2 − t1) = R(τ), say,

since the average depends only on time differences. If the process is also ergodic, then

It is worth noting that

which is the average power in a sample function.

1.4.3.3 Random processes

The most important random process is the Gaussian or normal random process. This is because a wide range of physically observed random waveforms can be represented as Gaussian processes, and the process has mathematical features which make analysis relatively straightforward.

The probability density function of a Gaussian process x(t) is

may vary with time for a non-stationary process but are independent of time if the process is stationary.

One of the most important features of the Gaussian process is that the response of a linear system to this form of excitation is usually another (but still Gaussian) random process. The only changes are that the magnitude and standard deviation of the response may differ from those of the excitation.

A Gaussian probability density function is shown in , and the standard deviation σ controls the spread.

Figure 1.29 Gaussian probability density function

= 0,

Figure 1.30 shows the Gaussian probability density function with zero mean. This integral has been calculated for a range

Figure 1.30 Gaussian probability density function with zero mean

of values of λ and the results are given in Table 1.4. The probability that x(t) lies outside the range −λσ to +λσ is 1 minus the value of the above integral. This probability is also given in Table 1.4.

Table 1.4

1.4.3.4 Spectral density

The spectral density S(ω) of a stationary random process is the Fourier transform of the autocorrelation function R(τ), and is given by

The inverse, which also holds true, is

That is, the mean square value of a stationary random process x is the area under the S(ω) against frequency curve. A typical spectral density function is shown in Figure 1.31.

Figure 1.31 Typical spectral density function

A random process whose spectral density is constant over a very wide frequency range is called white noise. If the spectral density of a process has a significant value over a narrower range of frequencies, but one which is nevertheless still wide compared with the centre frequency of the band, it is termed a wide-band process (Figure 1.32). If the frequency range is narrow compared with the centre frequency it is termed a narrow-band process (Figure 1.33). Narrow-band processes frequently occur in engineering practice because real systems often respond strongly to specific exciting frequencies and thereby effectively act as a filter.

Figure 1.32 Wide-band process

Figure 1.33 Narrow-band process

1.5 Mechanics of fluids

1.5.1 Introduction

Fluid is one of the two states in which matter can exist, the other being solid. In the fluid state the matter can flow; it will, in general, take the shape of its container. At rest a fluid is not able to sustain shear forces.

Some ‘solids’ may flow over a long period (glass window panes thicken at the