The Concise Valve V II

EBOOKS The Concise Valve Handbook
CRABTREE
FOR THE Actuation, Maintenance, and Safety Relief, AUTOMATION AND CONTROL
ENGINEERING Volume II COLLECTION
LIBRARY
Michael A. Crabtree
Create your own
Research studies within the process industry routinely indicate
Customized Content
that the fluid control valve is responsible for 60 to 70% of poor-
Bundle — the more
functioning control systems. Furthermore, valves in general are
books you buy, consistently wrongly selected, regularly misapplied, and often
the higher your incorrectly installed.
discount! This two-volume book comprises a comprehensive up-to-date
body of knowledge that provides a total in-depth insight into valve
THE CONTENT and actuator technology—looking not just at control valves, but a
• Manufacturing whole host of other types including: check valves, shut-off valves,
The Concise Valve

Engineering solenoid valves, and pressure relief valves.
A methodology is presented to ensure the optimum selection of
The Concise Valve Handbook, Volume II

• Mechanical
& Chemical size, choice of body and trim materials, components, and ancillaries.
Handbook
Engineering Whilst studying the correct procedures for sizing, readers will also
• Materials Science learn the correct procedures for calculating the spring ‘wind-up’ or
& Engineering ‘bench set’.
• Civil & Maintenance issues also include: testing for deadband/
Environmental
Engineering
hysteresis, stick-slip and non-linearity; on-line diagnostics; and
signature analysis.
Actuation, Maintenance,
• Advanced Energy
Technologies
Written in a detailed but understandable language, the two
volumes are presented in a form suitable for both the beginner,
and Safety Relief
with no prior knowledge of the subject, and the more advanced
THE TERMS specialist.
Volume II
• Perpetual access for For the last sixteen years, ‘Mick’ Crabtree, who holds an MSc in
a one time fee industrial flow measurement, has been involved in technical training
• No subscriptions or and consultancy—running workshops on industrial instrumentation
access fees and networking throughout the world covering the fields of process
• Unlimited control (loop tuning), process instrumentation, data communications,
concurrent usage fieldbus, safety instrumentation systems (according to both ISA S84
• Downloadable PDFs and IEC 61508/61511), project management, on-line analysis, and
• Free MARC records technical writing and communications.
This book represents some thirty years wealth of experiential
For further information,
a free trial, or to order,
contact:
knowledge gleaned by the author working within a wide variety of
industries and from more than 6000 technicians and engineers who Michael A. Crabtree
have attended the author’s workshops.
sales@momentumpress.net
ISBN: 978-1-94708-369-1
The Concise
Valve Handbook
The Concise
Valve Handbook
Actuation, Maintenance, and
Safety Relief
Volume II
Michael A. Crabtree
MOMENTUM PRESS, LLC, NEW YORK

The Concise Valve Handbook: Actuation, Maintenance, and Safety
Relief, Volume II
Copyright © Momentum Press®, LLC, 2018.
All rights reserved. No part of this publication may be reproduced, stored

in a retrieval system, or transmitted in any form or by any means—
electronic, mechanical, photocopy, recording, or any other—except for
brief quotations, not to exceed 400 words, without the prior permission
of the publisher.
First published by Momentum Press®, LLC

222 East 46th Street, New York, NY 10017
www.momentumpress.net
ISBN-13: 978-1-94708-369-1 (print)

ISBN-13: 978-1-94708-368-4 (e-book)
Momentum Press Automation and Control Collection
Cover and interior design by Exeter Premedia Services Private Ltd.,

Chennai, India
10 9 8 7 6 5 4 3 2 1
Printed in the United States of America

To my wife Pam—for her love and patience
Abstract
Volume II: Actuation, Maintenance, and Safety Relief, takes an in-depth

look at actuators and positioners. This volume also explores a variety
of maintenance and diagnostic issues including: testing for dead-band/
hysteresis, stick-slip and non-linearity; on-line diagnostics; signature
analysis; and correct procedures for calculating the spring “wind-up” or
“bench set.”
A complete section is also devoted to the whole field of safety relief
devices.
Lastly, this volume covers a number of topics which are all too often
ignored: acoustics; water hammer; and even classification of stainless
steel.
Keywords
acoustics, actuators, block and bleed, diagnostics, fail-safe, maintenance,

positioners, safety relief valves, stainless steel classification, transfer
mechanisms, water hammer
Contents
List of Figures xv
List of Tables xxiii
Foreword xxv
Volume I
1 Basic Principles 1
1.1 The Final Control Element as Part of the Control Loop 2
1.2 Basic Theory 3
1.3 Equation of Continuity 3
1.4 Bernoulli’s Equation 5
1.5 Choked Flow 8
1.6 Pressure Recovery 9
1.7 Turndown Ratio and Rangeability 11
1.8 Velocity Profiles 12
1.9 Reynolds Number 13
1.10 Flashing and Cavitation 14
1.11 Flashing 15
1.12 Cavitation 16
1.13 Leakage Classification 18
1.14 Isolation Valve Leakage Classification 21
2 Liquid Valve Sizing 23
2.1 Practical Considerations 23
2.2 Application of Formulae 24
2.3 Sizing Example 1 27
x • Contents
2.4 Piping Geometry Factor 29

3 Gas Valve Sizing 33
3.1 Pressure Drop Mechanism 33
3.2 Specific Heat Ratio Factor 38
3.3 Gas Expansion Factor 40
3.4 Valve Sizing 41
4 Valve Construction 47
4.1 Globe Valve 48
4.2 Bonnet Assembly 49
4.3 PTFE (Teflon) 49
4.4 Laminated Graphite 50
4.5 Extended Bonnet 52
4.6 Bellows Seal Bonnet 52
4.7 Valve Trim 54
4.8 Guiding 55
4.9 Post-guiding 55
4.10 Top- and Bottom-guided Double Seat 56
4.11 Single-ported Balanced Globe Valve 57
4.12 Cage-guiding 58
4.13 Split Body Globe 59
4.14 Angle Is 60
4.15 Needle Valve 61
4.16 Bar Stock Body Valve 61
4.17 Gate Valve 61
4.18 Wedge Gate 62
4.19 Slab Valve 64
4.20 Expanding Gate Valve 65
4.21 Knife Edge Gate Valve 67
4.22 Pinch Valve 69
4.23 Diaphragm Valve 72
4.24 Rotary Control Valves 74
Contents • xi
4.25 Ball Valve 75
4.26 Trunnion Ball Valve 77
4.27 Characterized Ball Segment Valve 81
4.28 Butterfly Valve 82
4.29 Plug Valve 84
4.30 Eccentric Plug Valve 86
4.31 Check Valves 88
4.32 Valve Sizes and Pipe Schedules 88
4.33 Material Selection 90
4.34 Corrosion 90
4.35 Erosion 94
4.36 End Connections 94
4.37 Screwed End Connections 94
4.38 Flanged End Connections 95
4.39 Hub End Body 96
4.40 Welded End Connections 96
4.41 Lap Joint Flange 98
4.42 Flangeless Connections 99
4.43 Grayloc® Connector 100
5 Valve Trim and Characterization 103
5.1 Inherent Characteristics 103
5.2 Linear Inherent Flow Characteristic 103
5.3 Equal Percentage Inherent Flow Characteristic 104
5.4 Quick Opening Inherent Flow Characteristic 104
5.5 Modified Percentage Inherent Flow Characteristic 105
5.6 Characteristic Profiling 105
5.7 Installed Characteristics 105
5.8 Cavitation Control 108
5.9 Reducing Cavitation 110
5.10 Eliminating Cavitation 112
5.11 Noise Sources 113
5.12 Mechanical Noise 115
5.13 Hydrodynamic Noise 116
xii • Contents
5.14 Aerodynamic Noise 117

5.15 Noise Prediction 117
5.16 Noise control 117
5.17 Path Treatment 118
5.18 Insulation 119
5.19 Silencers 120
5.20 Source Treatment 120
5.21 Velocity Control 120
6 Valve Selection 123
Glossary 129
Bibliography 131
About the Author 133
Index 135
Volume II
7 Valve Actuators and Positioners 141
7.1 Pneumatic Control 141
7.2 Flapper–Nozzle Assembly 141
7.3 I/P Converter 142
7.4 Diaphragm Actuators 144
7.5 Springless Diaphragm 145
7.6 Advantages and Disadvantages of Diaphragm Actuators 147
7.7 Cylinder Actuators 147
7.8 Spool Block 149
7.9 Electro-Hydraulic Actuation 149
7.10 Electric Actuation 150
7.11 Torque Limiting 152
7.12 Hammer-Blow Mechanism 153
7.13 Solenoid Valve 153
7.14 Digital Actuators 155
7.15 Transfer Mechanisms 157
7.16 Valve Positioners 161
7.17 Positioner Guidelines 163
Contents • xiii
8 Valve Testing and Diagnostics 167

8.1 Deadband and Hysteresis 167
8.2 Testing Procedures 169
8.3 Online Diagnostics 173
8.4 Electronic Torque Monitoring 176
9 Valve Maintenance and Repair 179

9.1 In-Line Repairs 180
9.2 Repairs Under Pressure 180
9.3 Repairs on Drained Systems 181
9.4 Packing Replacement 181
9.5 Replacing or Refinishing Seat Rings 181
9.6 Other In-Line Repairs 182
9.7 In-Line Post-Repair Procedures 182
9.8 Shop Repairs 182
9.9 Actuator Bench Set 183
9.10 Spring Calculations 184
10 Safety Relief Valves 187

10.1 History 187
10.2 Definitions 190
10.3 Weight-Loaded Pressure/Vacuum Relief Valves 191
10.4 Spring-Loaded Relief Valves 192
10.5 Applications 194
10.6 Limitations 194
10.7 Safety Valves 194
10.8 Basic Operation: Lifting 196
10.9 Basic Operation: Reseating 198
10.10 Conventional Safety Relief Valves 200
10.11 Balanced Safety Relief Valves 204
10.12 Bellows-Type Balanced Safety Valve 204
10.13 Piston-Type Balanced Safety Valve 206
10.14 Non-Reclosing Pressure Relief Devices 211
10.15 Conventional Rupture Disc 215
10.16 Scored Tension-Loaded Rupture Disc 215
xiv • Contents
10.17 Composite Rupture Disc 216

10.18 Graphite Rupture Disc 217
10.19 Burst Disc Applications and Installation Practices 218
10.20 Performance Tolerance 219
10.21 Maximum Operating Pressure 221
10.22 Cyclic/Pulsating Duties 221
10.23 Case A: 276 kPa (g) or Higher 222
10.24 Case B: Lower than 276 kPa (g) 223
10.25 Standards 223
Appendix A: J–T Valve 225
Appendix B: Basic Acoustics 227
Appendix C: Block and Bleed 241
Appendix D: Water Hammer 245
Appendix E: Stainless Steel 255
Glossary 263
Bibliography 265
About the Author 267
Index 269
List of Figures
Figure 7.1. The flapper–nozzle assembly converts a

small physical displacement into a pressure change. 142
Figure 7.2. As the flapper moves away from the nozzle, the airflow
will increase and the output pressure will fall. Although
this produces a non-linear output, over the normal range
of interest, 0.2 to 1 bar (3 to 15 psi), the relationship is
normally considered to be a straight line. 143
Figure 7.3. The 4–20 mA current signal is applied to a coil that
is physically attached to a spring diaphragm on which is
mounted the flapper. As the current varies, its magnetic
field interacts with the permanent magnet field,
deflecting the diaphragm by an amount proportional
to the control signal current. 143
Figure 7.4. Typical configuration of an I/P converter
(courtesy Foxboro). 144
Figure 7.5. A direct-acting diaphragm actuator. 145
Figure 7.6. A reverse-acting diaphragm actuator. 146
Figure 7.7. Springless diaphragm actuator uses a
differential air input. 146
Figure 7.8. The cylinder or piston-type actuator. 148
Figure 7.9. Size comparison between a diaphragm actuator
(left) and cylinder actuator (right) mounted on two
comparable valves (courtesy Valtek International). 148
Figure 7.10. Pneumatic spool assembly in which the incoming
air is supplied to either one or other side of the actuator,
while at the same time, simultaneously exhausting air
from the opposite side (courtesy Mitech). 149
Figure 7.11. Swing jet controller. 150
xvi • List of Figures
Figure 7.12. Basic electrically operated actuator. 151

Figure 7.13. (a) The worm drive is free to move longitudinally on
a spline that is held in its central position by means of
pre-loaded torque springs. (b) If the valve reaches its
end position, the tangential force on the driven wheel
rises and displaces the worm gear axially on its shaft
against the pressure of the holding springs. 153
Figure 7.14. The worm wheel and output shaft are connected
via a dog coupling with backlash. When the rotation
direction is reversed, the backlash first has to be
covered, and the motor can run up to its nominal
output speed without load before the valve is
unseated (hammer blow). 154
Figure 7.15. A direct-acting solenoid valve. 154
Figure 7.16. A three-way solenoid valve might be used to switch
air from one side of an actuator diaphragm to the other
(a) de-energized position (b) energized position. 155
Figure 7.17. P&ID representation of a three-way solenoid valve
(dot indicates normally open (N.O.) port and a solid
indicates normally closed (N.C.) port. 155
Figure 7.18. Typical construction of a ‘four-phase’ stepping
motor with a basic step of 1.8°. 156
Figure 7.19. Rack and pinion transfer mechanism (courtesy
Mitech).157
Figure 7.20. Double-piston arrangement (a) air is supplied f orcing
the pistons away from each other and rotating the
pinion anticlockwise (b) air exhaustion (loss of pressure)
allows compressed springs to force the pistons toward
each other and rotate the pinion clockwise (courtesy
Spirax Sarco). 158
Figure 7.21. Double crank transfer mechanism (courtesy Mitech). 159
Figure 7.22. In the double-crank mechanism, the run torque
(the torque in a mid-position) is higher than the
end torque. 159
Figure 7.23. Scotch yoke transfer mechanism (courtesy Mitech). 160
Figure 7.24. Vector diagrams showing reaction force on the
rocker arm. 161
Figure 7.25. Vector diagrams showing moment of the reaction
force.161
List of Figures • xvii
Figure 7.26. In the scotch yoke mechanism, the end torques are
twice as high as the run torque (the torque in a
mid-position).161
Figure 7.27. Basic principle of operation of a pneumatic positioner. 162
Figure 8.1. Deadband as a result of mechanical play within
a gear-train. 168
Figure 8.2. An illustration of hysteretic error. 168
Figure 8.3. Hysteresis: A combination of deadband and
hysteretic error. 168
Figure 8.4. Determining the effects of hysteresis/deadband as a
result of an input change of two steps up, three down
and one up (courtesy Michael Brown Control
Engineering).169
Figure 8.5. The effects of stick-slip, without hysteresis and
deadband (courtesy Michael Brown Control
Engineering).170
Figure 8.6. As the PD increases in regular step, the PV increases
in a series of steps that gradually become smaller,
showing a marked non-linearity that is typical of an
oversized valve. 171
Figure 8.7. Testing connections for a complete pneumatically
operated final control assembly. 172
Figure 8.8. Negative hysteresis: one of the effects of an
undersized actuator (courtesy Michael Brown
Control Engineering). 173
Figure 8.9. Plotting the ‘valve signature’ plot with the actuator
pressure plotted on the y-axis and the travel plotted
along the x-axis. The separation between the opening
(red) and closing (blue) lines is the result of the
friction band (courtesy Fisher-Emerson). 174
Figure 8.10. The packing friction is approximately twice that of
the previous example. Typically, this might be due to
errantly over-tightening the packing, resulting in the
excessive friction (courtesy Fisher–Emerson). 175
Figure 8.11. An example of a valve signature showing several
revealed faults (courtesy Fisher–Emerson). 176
Figure 8.12. Opening torque characteristics of a typical wedge gate
valve in which the valve position (travel) is plotted on
the x-axis and the torque demand is plotted on the
y-axis (courtesy Rotork). 177
xviii • List of Figures
Figure 9.1. Although not generally recommended, in-line repair

can be carried out while the line is still under
pressure, on gate and globe valves having back seats. 180
Figure 9.2. Forces developed on a nominal 50-mm valve plug. 185
Figure 9.3. Unbalance of forces. 186
Figure 10.1. Family of pressure relief devices, classified as either
reclosing or non-reclosing. 188
Figure 10.2. Papin’s safety valve was kept closed by means of a
lever and movable weight. Sliding the weight along
the lever kept the valve in place and regulated the
steam pressure. 188
Figure 10.3. The weight of the seat assembly (or pallet) keeps
the valve closed until the pressure acting on the
underside equals this weight. 191
Figure 10.4. When associated with tank breathing (1 to 4 in WC),
weight-loaded valves are often referred to as a
‘conservation valves’ and provide IN- and
OUT-breathing.192
Figure 10.5. Spring-loaded pressure relief valve. 193
Figure 10.6. The valves have closed bonnets to prevent the release
of corrosive, toxic, flammable, or expensive fluids
and can be supplied with lifting levers. 194
Figure 10.7. The spring of a safety valve is usually fully exposed
and has a lifting lever for manual opening. 195
Figure 10.8. Illustration of the standard defined areas. 196
Figure 10.9. Typical disc and shroud arrangement used on rapid
opening safety valves. 197
Figure 10.10. Operation of a conventional safety valve. 197
Figure 10.11. Relationship between pressure and lift for a typical
safety valve. 198
Figure 10.12. Blowdown ring is threaded around the valve nozzle
and positioned to form a huddling chamber with the
disc skirt. 199
Figure 10.13. When the blowdown ring is adjusted up, the forces
required to lift the seat disc occur very close to set
pressure and the blowdown is long. When the ring is
adjusted down, the seat lift does not occur until the
pressure under the seat disc is considerably higher
and the blowdown is short. 199
List of Figures • xix
Figure 10.14. Schematic diagram of a valve with the spring housing

vented to the discharge side of the valve. 201
Figure 10.15. Schematic diagram of a valve with spring housing
vented to the atmosphere. 202
Figure 10.16. Bellows-type balanced safety valve. 205
Figure 10.17. Block schematic of a bellows-type balanced safety
valve showing force balancing. 205
Figure 10.18. Block schematic of piston-type balanced safety valve
showing force balancing. 207
Figure 10.19. High-pressure pilot-operated valve incorporating an
unbalanced piston and an integrally mounted pilot. 208
Figure 10.20. Alternative seating arrangements available for a
pilot-operated piston-type safety relief valve. 209
Figure 10.21. Low-pressure diaphragm-type pilot-operated valve. 209
Figure 10.22. Similar in construction to a spring-loaded valve,
but using a shear-pin in place of a spring. 212
Figure 10.23. The basic buckling pin valve comprises a pin of a
precise length that holds a piston on its seat. As the
pressure increases and the axial force on the pin
subsequently also increases, the pin will buckle. 212
Figure 10.24. Also known as a rupture disc, bursting disc, or burst
diaphragm, the disc is designed to rupture at a
predetermined pressure and, once ruptured, will not
re-seal (courtesy Oseco). 214
Figure 10.25. Bursting discs may be forward- or reverse-acting. 215
Figure 10.26. Typical rupture disc holders (courtesy Oseco). 215
Figure 10.27. Conventional domed rupture discs are pre-bulged
solid metal discs designed to burst when operating
conditions are 70% or less of the rated burst
pressure.216
Figure 10.28. Scored tension-loaded rupture discs allow a closer
ratio (generally 85%) of system operating pressure
to disc burst pressure. 216
Figure 10.29. Composite rupture disc (courtesy Continental
Disc Corp.). 217
Figure 10.30. Graphite rupture disc manufactured from graphite
impregnated with a binder material and designed to
burst by bending or shearing (courtesy Svi Carbon
Private Limited). 217
xx • List of Figures
Figure 10.31. Bursting disc installed on a safety valve. 219

Figure 10.32. Performance tolerances. 220
Figure 10.33. A zero-manufacturing range is the tightest, meaning
the average of the burst tests in the factory must
equal the nominal burst pressure at the
coincident temperature. 221
Figure 10.34. Determining the upper and lower maximum operating
ranges according to whether the pressure lies above
or below 276 kPa (g). 223
Figure B.1. Graphic representation of a sound wave showing
frequency and amplitude. 228
Figure B.2. Addition of harmonics to the fundamental:
(a) second harmonic; (b) third harmonic. 229
Figure B.3 Two identical loudspeakers connected across
an amplifier. 231
Figure B.4. Logarithmic response of the human ear. 232
Figure B.5. The 10-fold power ratio increase is designated a
Bel with each power increment of 26% being
one-tenth of a Bel—called a decibel (dB). 233
Figure B.6. Simple circuit showing power developed
by a resistor. 235
Figure B.7. Doubling the power is achieved by only a 1.414
increase in voltage. 236
Figure B.8. Threshold of hearing (lower) and of pain (upper). 238
Figure B.9. Equal loudness contours. 239
Figure B.10. A- B-, and C-weighted responses required for
measuring sound pressure levels. 240
Figure C.1. (a) Under normal operation, the valves are set
with the isolation valves 1 and 2 open and the bleed
valve 3 closed. (b) When isolating the downstream
equipment, the valves are set with isolation valves
1 and 2 closed and bleed valve 3 open. 242
Figure C.2. Typical construction of a single double
block-and-bleed valve (courtesy Habonim). 242
Figure C.3. The bleed often takes the form of a cap or plug. 243
List of Figures • xxi
Figure D.1. If a moving column of liquid (a) is slowed down

suddenly by, for example, a quick-closing valve,
the sudden change in liquid velocity in the delivery
line creates a pressure wave (b). 246
Figure D.2. The pressure wave travels back up the line (a) at
between 1,000 and 1,300 m/s, to the end of the pipe
where it will reverse direction and travel back toward
the valve (b). 247
Figure D.3. Depending on the valve size and system conditions,
a valve closing in 1.5 s or less can produce a pressure
spike five times the system working pressure. 248
Figure D.4. Hydraulic shock wave produced as a result of
accumulated condensate in steam piping. 249
Figure D.5. A pulsation dampener or surge suppressor. During a
surge, the fluid pressure displaces the bladder and
compresses the trapped gas. 251
Figure D.6. A Daniel gas-loaded axial flow style valve in which
nitrogen gas is used to pressurize the valve piston
to keep it in the closed position (courtesy Emerson). 252
Figure D.7. As the pipeline pressure increases, the combined
force of the spring and nitrogen gas pressure is
overcome and the valve opens (courtesy Emerson). 253
Figure E.1. In ferritic stainless steel, the atoms are arranged in a
body-centered cubic structure. 256
Figure E.2. Ferritic stainless steel family. 257
Figure E.3. With the addition of nickel, the atoms in austenitic
stainless steels are arranged on the corners of the
cube and also in the center of each of the faces. 257
Figure E.4. The relationship between the various 300 series
austenitic grades. 258
Figure E.5. The martensitic Grade 400 series. 261
Figure E.6. Relationship between the complete family of
stainless steels. 262
List of Tables
Table 10.1. Typical operating pressure to burst pressure ratios

(courtesy BS&B) 222
Table 10.2. Safety valve performance summary 224
Table B.1. The wavelengths of several frequencies travelling
in air 230
Table B.2. Gain and attenuation ratios expressed in dBs 234
Table B.3. Sound intensities of various sources 238
Table D.1. Some typical velocities in various liquids 246
Table E.1. Difference in the properties of ferritic and austenitic
stainless steels 258
Table E.2. Chemical composition of standard grades (courtesy
International Stainless Steel Forum) 260
Foreword
In this book, ‘The Concise Valve Handbook—Part 2. Actuation, Maintenance,

and Safety Relief,’ I have made use of a building-block approach, present-
ing material in a form suitable for two distinct classes of reader: the
beginner, with no prior knowledge of the subject and the more advanced
specialist.
The complete text is suitable for the advanced reader. However,
those parts of the text that involve a mathematical treatment, which are
not required by the beginner, are indicated by a mark ► at the beginning
and ◄ at the end. Consequently, for the beginner, the text may be read,
with full understanding, by ignoring the marked sections.
I offer no apologies for my preference for metric-based measurement—
the SI system. Apart from the United States, only two other countries in
the world still adhere to the fps system (foot-pound-second)—the so-called
imperial system first defined in the British Weights and Measures Act of
1824—Burma and Liberia.
I have tried to mix it up as far as possible, and I have got a units con-
version table right in the front of the book. But, for the moment, just try for
the following:
1 bar = 100 kPa ≈ 1 atmosphere ≈ 14.7 psi

1 inch = 25.4 mm
20 °C = 68 °F
100 °C = 212 °F
And lastly, while I have made some compromises (analog instead

of analogue; program instead of programme), I reserve my right to spell
according to the British system:
xxvi • Foreword
English United States

Metre Meter
Litre Liter
Fibre Fiber
Colour Color
Unit Conversions
Quantity SI United States customary

Distance 25.5 mm 1 in
1 millimetre 0.03937 in
1m 39.37 in
1m 3.281 ft
0.9144 m 1 yd
Area 1 square metre (m2) 1550 in2
1 square metre (m2) 10.76 ft2
1 square metre 0.00155 in2
millimetre (mm2)
Volume 1 cubic metre (m3) 61.02 in3
1 cubic metre (m3) 35.31 ft3
0.02832 m3 1 ft3
1 litre 61.02 in3
1 litre 0.03531 ft3
1 litre 0.2642 gal
3.785 litres 1 gal
Mass 1 kg 2.205 lb
454 g 1 lb
Force 1N 0.2248 lbf
4.448 N 1 lbf
Pressure 1 bar 14.504 lbf/in2 (psi)
1 kPa (kN/m2) 0.145 lbf/m2 (psi)
6.895 kPa 1 psi
1 psi 0.0361 inches H2O
(in WC)
Foreword • xxvii
Quantity SI United States customary

Temperature K 1.800 °R
°C 1.8 °C + 32 = °F
Flow rate 1 m3/h 4.403 gal/min (gpm)
1 kg/h 2.205 lb/h
CHAPTER 7
Valve Actuators and

Positioners
In any flow control loop, a primary sensing flow device is used to produce
a signal, which ultimately controls a valve: either to open or close, in an
ON/OFF mode, or to provide proportional control. The actuator, therefore,
is that part of the final control element that moves the control valve—
either in a linear manner (for the control of a globe valve) or in a rotary
manner (for control of butterfly or ball valves).
An actuator may be powered electrically, pneumatically, or hydrau-
lically. However, despite the trend away from pneumatically controlled
instrumentation and toward electronics, the actuator still remains predom-
inantly pneumatic.
7.1 Pneumatic Control
In the process control instrumentation field, the pneumatically controlled

actuator is still used for four main reasons:
• users feel that there is little, if any, improvement in the performance

of electric actuators;
• the cost of electric actuators is higher than pneumatic actuators;
• the power dissipation on electric actuators is considered excessive,
giving rise to thermal problems; and
• users feel that pneumatic control is more reliable and can provide a
FAIL-OPEN or FAIL-CLOSE operation.
7.2 Flapper–Nozzle Assembly
At the heart of most pneumatic process control systems lies the flapper–
nozzle assembly (Figure 7.1)—a device that converts a small physical
142 • The Concise Valve Handbook
Supply
pressure
Flapper Pressure
reducing
restriction
Output
pressure
Nozzle
b a
Figure 7.1. The flapper–nozzle assembly converts a

small physical displacement into a pressure change.
displacement into a pressure change. An air supply is applied to the nozzle

via a pressure reducing restriction, such that the output pressure will be
lower that the supply pressure by an amount determined by the flow of air
from the nozzle.
The outflow of air from the nozzle varies according to the position of
the flapper. With the nozzle covered, the pressure approaches the supply
pressure, but as the flapper moves away from the nozzle, the airflow will
increase and the output pressure will fall. This is shown in Figure 7.2.
The input displacement is applied to the flapper, which increases or
decreases the distance from the nozzle, and thus varies the output pres-
sure. It should be noted that the displacement range is quite small, of the
order of micrometres and produces a non-linear output. However, over
the normal range of interest, 0.2 to 1 bar (3 to 15 psi), the relationship is
normally considered to be a straight line.
7.3 I/P Converter
A common application of the flapper–nozzle device is the electro-

pneumatic current-to-pneumatic converter, normally referred to as an I/P
converter, which, typically converts a standard 4–20 mA process signal
into a pneumatic output varying linearly over the range 20 to 100 kPa (3
to 15 psi). In a practical arrangement, the flapper is physically attached to
a spring diaphragm on which is mounted a coil system (Figure 7.3). The
coil is arranged within a magnetic field that is produced by a permanent
magnet. As the current in the coil varies from 4 to 20 mA, it produces
Valve Actuators and Positioners • 143
2.0
1.8
1.6
1.4
Pressure (bar)
1.2
1.0
0.8
0.6
0.4
0.2
0
a b
Displacement (µm)
Figure 7.2. As the flapper moves away from the nozzle,

the airflow will increase and the output pressure will
fall. Although this produces a non-linear output, over the
normal range of interest, 0.2 to 1 bar (3 to 15 psi), the
relationship is normally considered to be a straight line.
Permanent magnet
4 – 20 mA
Coil
Nozzle Flapper
Spring diaphragm
Figure 7.3. The 4–20 mA current signal is applied

to a coil that is physically attached to a spring
diaphragm on which is mounted the flapper. As the
current varies, its magnetic field interacts with the
permanent magnet field, deflecting the diaphragm by
an amount proportional to the control signal current.
a magnetic field that interacts with the permanent magnet field. The
diaphragm, thus, deflects by an amount proportional to the control signal
current, to produce a change in the flapper–nozzle gap.
The fully packaged I/P converter (Figure 7.4) comprises the flapper/
nozzle assembly together with a downstream volume booster that acts as a
pilot-operated regulation device.
Permanent magnet
Coil
4 – 20 mA
Spring diaphragm
Exhaust to
Flapper atmosphere
Nozzle Diaphragm
Pilot air
Exhaust to
atmosphere
Control diaphragm
Supply air Control air
Figure 7.4. Typical configuration of an I/P converter (courtesy

Foxboro).
The supply air is applied to the lower chamber of the volume booster
where a certain amount, determined by the position of the control dia-
phragm, flows to the output.
When the flapper moves closer to the nozzle, the dynamic back-
pressure increases until it corresponds to the input pressure and pushes
both the diaphragm and the control diaphragm downward, causing the
output pressure to increase until a new state of equilibrium is reached in
the diaphragm chambers.
When the output pressure decreases, the diaphragm moves upward,
allowing the output pressure to vent until the forces on the diaphragms are
balanced again.
7.4 Diaphragm Actuators
The diaphragm actuator is the most widely used pneumatic actuator for
proportional control. As shown in Figure 7.5, the variable operating air
is applied to one side of a flexible diaphragm. In this form, the lower
chamber is vented to atmosphere and the operating air, thus, moves the
diaphragm downward, against the force of the ‘ranging’ spring.
It is important, at this point, to consider the effect of air pressure fail-
ure. Since the plug stem needs to move upward to open the valve and
increase the flow, the direct-actuating diaphragm would be closing the
valve against the range spring pressure—and the pressures acting on the
plug of the valve. Thus, in the event of air pressure failure, the valve would
go to a fully open position—fail-open.
Travel stop Control air input
Diaphragm Spring
Spring
flange
Actuator
stem
Travel indicator
plate
OPEN
Figure 7.5. A direct-acting diaphragm

actuator.
In many cases, the fail-open mode is highly desirable. There are,

however, also many more cases where a fail-close mode of operation is
required. As shown in Figure 7.6, in the reverse-acting diaphragm actua-
tor, the variable operating air is applied to the lower sealed chamber with
the upper chamber vented to atmosphere. In this case, failure of the air
supply would result in the actuator stem moving downward under the
action of the spring.
7.5 Springless Diaphragm
In the springless diaphragm actuator (Figure 7.7), the control air is applied
differentially to both sides of the diaphragm. This arrangement allows a
much higher actuating force to be applied, for emergency on/off control,
since one side can be bled, and there is no restraining opposition, other
than the valve itself.
The springless diaphragm may also be used for proportional control
with the signal air pressure fed to one side of the diaphragm and a separate
regulated supply fed to the other side.
Spring
Diaphragm
Travel stop
Control
air input
Actuator
stem
Travel indicator
plate
OPEN
Figure 7.6. A reverse-acting diaphragm actuator.
Air
Air
Diaphragm
OPEN
Figure 7.7. Springless diaphragm actuator

uses a differential air input.
7.6 Advantages and Disadvantages of

Diaphragm Actuators
The main advantage of the diaphragm-type actuator is its cost since it

is the least expensive method of applying proportional control. In addi-
tion, because it is the most widely used type of actuator, a wide choice of
devices is available to suit virtually any type of valve. Furthermore, by
using a characterized spring, it is possible to obtain rough control by feed-
ing a 0.2 to > 1 bar (3–15 psi) signal directly onto the diaphragm.
Further advantages of the diaphragm actuator are: it is essentially
fail-safe; the speed is adjustable from fast to slow, with reasonably close
control; it is easily adapted for use in explosion proof areas; and it is easy
to maintain.
One of the main problems of the diaphragm actuator becomes
apparent when high-thrust forces are required, for example, to obtain tight
shut-offs on certain types of valves.
To obtain high thrusts, either the diaphragm area or the control pres-
sure must be increased, both of which place increased restraints on the
casing. Thus, for example, the Samson type 271 pneumatic actuator can
have an effective diaphragm area of up to 2,800 cm2, but with a maximum
pressure of 3 bar to provide a thrust of some 84 kN. Already, however, the
outside diameter of the casing is some 600 mm, a bulky and top-heavy
structure when used with smaller valves.
Other disadvantages include: stiffness is low, and so, precise control
is not always possible; the need for a supply of clean air; and, if required,
hand-wheel overrides are expensive and large.
7.7 Cylinder Actuators
The cylinder or piston-type actuator (Figure 7.8) makes use of a cast cyl-
inder much better able to withstand high pressures than the diaphragm
type (up to 1 MPa) and may be hydraulically or pneumatically operated.
To generate a thrust of 84 kN (as in the previous example), the piston
area must only be 840 cm2—almost half the diameter. Figure 7.9 shows a
size comparison between a diaphragm actuator (left) and cylinder actuator
(right) mounted on two comparable valves.
Although many cylinder actuators are spring-opposed, they are gen-
erally used either in a differential mode or make use of a constant load air
cushion restraint.
Control
input
Piston
Spring
Cylinder
Air Neoprene
cushion boot
Yoke Travel
indicator
Figure 7.8. The cylinder or piston-type

actuator.
Figure 7.9. Size comparison between a diaphragm

actuator (left) and cylinder actuator (right) mounted
on two comparable valves (courtesy Valtek
International).
Generally, spring-and-diaphragm actuators contribute less friction to

the control valve assembly than piston actuators, and their frictional char-
acteristics are more uniform with age.
Piston actuator friction will probably increase significantly with use
as guide surfaces and the O-rings wear, lubrication fails, and the elastomer
degrades. Thus, to ensure continued good performance, maintenance is

required more often for piston actuators than for spring-and-diaphragm
actuators. If that maintenance is not performed, process variability can
suffer dramatically without the operator’s knowledge.
7.8 Spool Block
In the springless piston-type actuator, in which the air must be applied dif-
ferentially, the pneumatic amplifier often takes the form a spool assembly
(Figure 7.10). Movement of the spindle switches the incoming air supply
to either one or other side of the actuator, while at the same time, simulta-
neously exhausting air from the opposite side. The spindle, which moves
inside the spool block, must be virtually frictionless to ensure that the
spindle will move for small changes in the input signal.
7.9 Electro-Hydraulic Actuation
The increasing acceptance of electronics in the process control industry

has led to the need for direct electronic control of actuators. One such
development is the electro-hydraulic actuator used to control, for exam-
ple, a cylinder actuator. One such operating device, the swing jet, is illus-
trated in Figure 7.11. High-pressure oil flows through a pivoted needle jet,
which may be deflected by coils to one side or the other. In the quiescent
Spool block
Spindle
Supply top
of actuator
Exhaust
top of
actuator
Supply in Supply in Supply in
Supply
bottom of
Exhaust actuator
bottom of
actuator
Figure 7.10. Pneumatic spool assembly in which the incoming air is

supplied to either one or other side of the actuator, while at the same time,
simultaneously exhausting air from the opposite side (courtesy Mitech).
High pressure
oil input
Pivot
Deflection
coils
Swinging
needle jet
Pick-up
block
Outlet
A B
Figure 7.11. Swing jet

controller.
state, the high-pressure oil impinges in the center of the pick-up block and
the outlet oil pressure from A and B is equal.
Using electronic control, the jet may be deflected to the left or right
to increase the flow into either of the output pick-ups A or B. Such a dif-
ferential output would, therefore, result in movement of the piston-type
actuator, with position feedback ensuring that the needle is returned to its
central quiescent position when the valve stem reaches its demanded posi-
tion. The main drawback of this design is that, due to the pressure losses
within the swinging jet itself, the full power of the hydraulic pressure is
not realized at the actuator. Although this problem may be overcome by
using a hydraulic servo, this serves to increase the cost further, a cost that
is already higher than pneumatically operated diaphragm actuators.
7.10 Electric Actuation
One of the major disadvantages of the electro-hydraulic-type system is

the need for a constant source of pressure, entailing the constant use of
electric power: the constant running of motor and pumps. This problem is
overcome in the electrical actuator (often referred to as a Motor-operated
(MOV)) where an electric motor drives the valve stem through a worm
gear assembly (Figure 7.12). Voltage requirements are generally in the
range 110/230 V AC and 24 V AC or 24 V DC.
One of the fundamental requirements of any actuator/valve combina-
tion is that it should be non-reversing. The motor should drive the valve
and the forces on a butterfly valve, for example, must not feed back and
drive the motor. Traditionally, this non-reversing characteristic has been
accomplished using a simple worm gear system in which a worm drive on
the motor shaft drives a worm gear.
In this arrangement, the wheel cannot drive back through the worm as
long as the worm crosses the wheel at an acute angle (less than 45°). Such
gearing also provides a speed reduction of as much as 100:1—with a cor-
responding increase in torque—providing a reasonably compact solution
for even large valves.
Despite these advantages, electric actuators suffer from a number of
drawbacks that preclude their use in all, but a few applications.
One of the most serious limitations of an electric actuator is the speed
of valve movement, which can be as low as 4 s/mm, and generally rules
out their use for modulating control.
Another serious a drawback is that MOVs generally have a ‘fail-in-
place,’ rather than a ‘failsafe’ action. For this reason, most are equipped
with a mechanically operated hand-wheel that allows the valve to be
manually operated to its open or closed position in the event of power
Worm
gear drive
Motor
Figure 7.12. Basic electrically operated

actuator.
or mechanical failure. A number of solutions are also offered in which a

spring or battery-backed UPS drives the valve to its open or closed posi-
tion in the event of power failure neither of which is entirely satisfactory.
Apart from being, generally, both cumbersome and expensive,
springs require the electric motor to be up to three times the size otherwise
required, and batteries do not deliver power quickly enough to provide
the rapid shutdown needed on an electric actuator to replicate spring clo-
sure. Another solution lies in the use of electric actuators incorporating
advanced ‘super’ capacitors that can be easily configured to drive a valve
to any position (open, close, or any intermediate position) on loss of power
or control signal.
A further problem with electric actuators is that there are only a few
that are certified for hazardous areas.
Lastly, electric actuators are generally more expensive than their
equivalent pneumatically operated actuators, and their complexity makes
them more difficult to maintain.
7.11 Torque Limiting
The basic role of the actuator is to move the valve to either a mechanically
limited end position or an intermediate position. At the same time, so as
not to overload the valve, the actuator must avoid producing any excess
torque either during the travel or at the end positions. Thus, an important
consideration in the design in any actuator is to ensure that the torque
drive is discontinued when the valve reaches its end limits (fully opened
or fully closed).
Many designs of electrical actuator accomplish this by limit switches.
However, if faced with the possibility of a limit switch failure, precau-
tions must be taken to ensure that, when an end limit is reached and the
torque starts to rise, that it does not increase to a point where the valve is
damaged.
In order not to overload the valve, the actuator must avoid producing
any excess torque either during the travel or at the end positions.
In one method, the worm drive is free to move longitudinally on a
spline and is held in its central position by means of pre-loaded torque
springs (Figure 7.13). If now, while the drive is running the valve should
reach its end position (or become jammed), then the tangential force on the
driven wheel will rise considerably. This rise in torque displaces the worm
gear axially on its shaft against the pressure of the holding springs. This
movement is detected by means of a lever that operates the torque switch.
Splined shaft Worm drive

Pre-loaded
torque springs
Worm gear
(a) (b)
Figure 7.13. (a) The worm drive is free to move longitudi-

nally on a spline that is held in its central position by means
of pre-loaded torque springs. (b) If the valve reaches its
end position, the tangential force on the driven wheel rises
and displaces the worm gear axially on its shaft against the
pressure of the holding springs.
7.12 Hammer-Blow Mechanism
Often, rotary valves that are seldom operated become jammed or sticky
and are difficult to open or close. In many cases, this problem may be
overcome by the application of a quick, focused blow (similar to that of
a hammer striking an anvil). One method of applying such a controlled
blow, used by Auma Riester GmbH & Co. is shown in Figure 7.14.
The worm wheel and output shaft are connected via a dog coupling
with backlash. When the direction of rotation is reversed, the backlash first
has to be covered and the motor can run up to its nominal output speed
without load before the valve is unseated (hammer blow).
7.13 Solenoid Valve
The solenoid valve is essentially an on–off control element comprising an

electromagnetically operated plunger (or core) that is attached directly to
the valve stem (Figure 7.15).
Available in normally open or normally closed configurations, sole-
noid valves are widely used for emergency shut-off service or for opening
a valve simultaneously with the operation of, for example, a pump.
Driven shaft
Driven plate
Drive plate Drive dog

Drive shaft attached to
drive plate
Driven shaft
Driven plate
‘Backlash’ slot
in driven plate
Figure 7.14. The worm wheel

and output shaft are connected
via a dog coupling with backlash.
When the rotation direction is
reversed, the backlash first has to
be covered, and the motor can run
up to its nominal output speed
without load before the valve is
unseated (hammer blow).
Figure 7.15. A direct-acting solenoid valve.
When used in safety shut-off applications, the valve is normally held

in its energized position and would open or close to allow air to bleed
from, or be supplied to, the actuator.
Spool Common Flow Path when Return Solenoid

de-energized spring
Normally Normally
open closed
Flow Path when

de-energized
Normally Normally
open closed
Figure 7.16. A three-way solenoid valve might be used to switch air from
one side of an actuator diaphragm to the other (a) de-energized position
(b) energized position.
Figure 7.17. P&ID representation of a

three-way solenoid valve (dot indicates nor-
mally open (N.O.) port and a solid indicates
normally closed (N.C.) port.
A three-way solenoid valve (Figures 7.16 and 7.17) might be used to

switch air from one side of an actuator diaphragm to the other.
7.14 Digital Actuators
Digital type actuators are centered around the stepping motor, which is,
essentially, a DC motor in which the output shaft can be made to move
in a series of discrete angular steps. This is achieved through a spe-
cial motor design combined with an electronic drive circuit that applies
current pulses to a series of windings in a rotating sequence. The actual

methods of construction are many and varied, but the most commonly
available motor is known as ‘four-phase’ and has a basic step angle
of 1.8° (200 steps/r). This is achieved by using a laminated iron rotor
incorporating a permanent magnet that has 50 teeth, which are accu-
rately machined (Figure 7.18). The stator has eight toothed pole pieces,
each with its own winding, and the clearance between stator and rotor is
extremely small.
By means of a Vernier arrangement, the stator windings can be
sequentially energized to produce 200 steps in a revolution. Motors are
also available with other step angles, for example, 32, 48, 100, and 500
steps per revolution.
As an actuator, the stepping motor may be applied in three ways:
direct mechanical drive, stepper motor control of a pneumatic or hydrau-
lic servo valve, or stepping motor hybrid positioner in conjunction with a
piston actuator.
To date, little progress has been made in digital actuators with a suf-
ficiently low price tag, sufficient power, and sufficient speed to capture a
large share of the market.
5-tooth pole
piece
Stator
winding
50-tooth rotor
lamination
stack
Figure 7.18. Typical construction of a ‘four-phase’ stepping

motor with a basic step of 1.8°.
7.15 Transfer Mechanisms
Most rotary actuators use a linear cylinder in conjunction with a trans-

fer mechanism to translate the linear movement of the piston into rotary
motion of the shaft. Three mechanisms are commonly used:
7.15.1 Rack and Pinion
The rack and pinion transfer mechanism is effectively a car-steering

mechanism in reverse in which (Figure 7.19) a pinion gear is attached to
the drive shaft, the rack is attached to the pistons (most of these designs
use a double-piston arrangement), and the movement of the pistons causes
the shaft to rotate.
While this is a compact and neat arrangement, there are several dis-
advantages: only one tooth is fully engaged; the size is limited; there is no
adjustable end stop in the closed position of spring return units; the unit
comprises a complex multi-spring pack, and therefore maintenance is not
easy; no hand-wheel override is possible; and the mechanism is prone to
wear due to continuous reversals, hence backlash.
Figure 7.19. Rack

and pinion transfer
mechanism (courtesy
Mitech).
Pistons
Pinion driven
anticlockwise
(a) Air in
Air out
Pinion driven
clockwise
(b)
Figure 7.20. Double-piston arrangement (a) air

is supplied forcing the pistons away from each
other and rotating the pinion anticlockwise
(b) air exhaustion (loss of pressure) allows com-
pressed springs to force the pistons toward each
other and rotate the pinion clockwise (courtesy
Spirax Sarco).
The disadvantage of single-gear engagement is overcome in the

double-piston arrangement (Figure 7.20). However, these are even more
complex.
7.15.2 Double Crank
Figure 7.21 shows how a rocker plate is attached rigidly to the drive
shaft and an arm connects the piston shaft to the rocker plate. This arm
rotates to take up the lateral movement of the rocker plate pivot joint as
the shaft rotates.
Major advantages are that the run torque (i.e., the torque in a mid-
position) is higher than the end torque (Figure 7.22).
Figure 7.21. Double crank transfer mechanism

(courtesy Mitech).
800
700
600
Torque (N.m)
500
400
300
200
100
10 20 30 40 50 60 70 80 90
Shaft angle (°)
Figure 7.22. In the double-crank mechanism, the run torque ( the torque
in a mid-position) is higher than the end torque.
This is an advantage for modulating control applications and for

large valves on liquid applications where dynamic torques are signifi-
cant. In addition, backlash is negligible due to the fact that all move-
ment takes place at pivot points in bearings, unlike other designs that
incorporate gears or sliding surfaces. This is an advantage for modulating
control duties.
However, it must be recognized that this is a more expensive design
due to the larger number of components, and that in most applications, the
actuator is sized on the end torque that is produced by the unit. For this
reason, it is often necessary to use a larger model for the double-crank

design than for the scotch yoke.
7.15.3 Scotch Yoke
With this design (Figure 7.23), a pin is assembled through the piston shaft
and a slot is machined in the rocker arm to take up the relative lateral
movement of the rocker arm and the piston shaft as the arm rotates.
While this design does not appear to differ significantly from the
double-crank mechanism, the characteristic is surprisingly different.
The reason is that, in the end positions (Figure 7.24), the pin acts like
a wedge in the slot, which in those positions is at approximately 45° to the
piston shaft. Since there can be no force on the rocker arm in the direction
of the slot, the force produced by the piston must be opposed by a force
perpendicular to the slot. The vertical component of the reaction must be
the same as the piston force, and so, the reaction force must equal F/cos
45 when the rocker arm is at 45° to the piston shaft. Note that there must
also be a reaction in the bearings equal to F.
If the horizontal distance between the center line of the piston shaft
and the center line of the drive shaft is M, then the moment arm that the
above force works on is M/cos 45° (Figure 7.25).
Thus, the torque produced at the end position is:
(F/cos 45°) . (M/cos 45°) = 2 FM
This means that, in theory, this type of actua-
tor transfer mechanism produces twice the torque
at the end (Figure 7.26) than it does in the center
position.
The high-end torque characteristic of the
scotch yoke is ideal for on–off duties for ball
and butterfly valves where the greatest torque is
required to seat and unseat the ball or disc. This
usually results in a smaller unit than any of the
other mechanisms. In addition, large actuators of
this design are possible.
A major disadvantage is that the run torques
are low compared with the end torque, and so,
this type of actuator is not suitable for modu-
Figure 7.23. Scotch
lating control. In addition, more wear can be yoke transfer mech-
expected on the piston shaft bushes due to the anism (courtesy
high side thrusts. Mitech).
Piston center
Piston line
Pin
F Drive shaft
Pin St center line
Ef
F op
fe
ns
ct
ce
iv
o nt
cti
el
Sl er
ea
en
45° ot (M
gt
R ax /c
h
is os
at
en
F/cos 45° 45
d
°)
45° Pivot Pivot
Effective length in
center (M)
Figure 7.24. Vector diagrams
showing reaction force on the Figure 7.25. Vector diagrams
rocker arm. showing moment of the reaction
force.
600
Torque (N.m)
500
400
300
200
100
10 20 30 40 50 60 70 80 90
Shaft angle (°)
Figure 7.26. In the scotch yoke mechanism, the end torques are twice as
high as the run torque (the torque in a mid-position).
7.16 Valve Positioners
In any proportional control system, it is assumed that a given signal out-

put represents a corresponding control action. In the case of an actuator,
a given signal air input should result in a specific opening of the valve.
Thus in the linear range 20–100 kPa, 60 kPa represents a 50% control
action, which should be reflected in the valve moving to a 50% open posi-
tion. However, the valve may not assume its correct position for a number
of reasons. These may include valve stem friction, unbalanced forces on
valve plug, variations in the process fluid pressure, etc.
As shown earlier, I/P transducers generally provide operating pres-
sure in the 20–100 kPa range, and thus 100 kPa is the maximum force
that could be applied to an actuator diaphragm. By contrast, the valve
positioner acts as a pneumatic relay that is capable of applying the full

force of the supply air (from 550 up to as much as 700 kPa) to drive the
valve, and thus overcome the various forces that prevent the valve from
reaching its correct position.
Figure 7.27 illustrates the basic principle of operation. The signal
control air is fed to the proportioning bellows that creates a signal force
on one end of a pivoted lever: the flapper. The resulting change in the
flapper–nozzle gap will raise or lower the control pressure applied to the
diaphragm of the air relay.
This action positions the pilot valve to either feed or vent the supply
air to the actuator, which moves the valve stem up or down. This action,
in turn, is fed back via the feedback positioner to balance the force at the
opposite end of the flapper lever from the proportioning bellows. Many
devices include a cam that can be used to characterize the system and
compensate for non-linearity in the process and/or final control devices.
It should be noted that, although positioner cams can be used to mod-
ify the characteristic of the final control element, its effect is limited in
most cases. This is because the cams can also dramatically change the
positioner loop gain, which severely limits its dynamic response. However,
while characterizing the valve trim is far more effective, the use of cams
is always better than no characterization at all and is often the only choice
with rotary valves.
Many modern electronic positioners incorporate valve characteri-
zation that electronically shapes the input signal ahead of the positioner
Control
diaphragm
Control input
Actuator
diaphragm
Proportioning bellows
Nozzle
Restriction
Flapper Supply
Feedback
cam
Figure 7.27. Basic principle of operation of a pneumatic positioner.

loop. Such devices use a pre-programmed table of values to produce the

valve input required to achieve the desired valve characteristic. This is
sometimes referred to as ‘forward path’ or ‘set-point characterization.’
Because characterization occurs outside the positioner’s feedback loop, it
avoids the problem of changes in the positioner loop gain.
7.17 Positioner Guidelines
The following have been adapted from a number of guidelines suggested

by John Egnew, of Emerson–Fisher.
7.17.1 All Positioner Types
• Reducing control valve deadband caused by friction

Most control valves without positioners, even those using “low
friction” packing material, may demonstrate a 5% deadband. Dead-
band greater than 1% can produce loop controllability problems.
Use of a positioner can reduce deadband caused by friction to less
than 1%.
• Reducing the effects of frictional stick/slip
Stick/slip occurs when a valve first sticks and then jerks (slips) to a
new position. Frequently, the stick/slip action causes process vari-
ables to overshoot the setpoint. Control action reverses the signal to
the control valve and the stick/slip action is repeated in the opposite
direction. A positioner’s stem feedback can reduce stick/slip effects
and maintain the process variable closer to the setpoint.
• Split ranged control elements
Processes requiring extended flow rangeability may use two con-
trol valves. The first valve operates over the first half of the range.
When the first valve is nearly 100% open, the control action begins
to open the second valve. This is called control valve split ranging.
When split ranging is performed on the pneumatic signal to the
valve actuators, positioners on each valve are used to achieve full
valve travel over the reduced input range. Most applications need
a predictable overlap region in the middle of the signal to avoid a
zone of no control. Positioners provide the accuracy to ensure the
correct overlap exists.
• Increasing the seating force and improving shutoff
A positioner will drive the output pressure to either zero or full
supply pressure whenever the valve reaches a physical travel stop.
An air-to-close actuator can use the full supply pressure to provide

greater valve seat loading. Completely removing the air signal from
air-to-open actuators allows the full force of the spring to load the
valve seat.
• Double-acting actuators
All double-acting piston actuators must have a positioner
because the pressure on both sides of the piston must be precisely
controlled. This can only be accomplished with the stem fed back
in a positioner.
7.17.2 Digital Positioners
Intelligent digital positioners can extend capabilities and benefits of tradi-

tional analog positioners. Using digital communications, such as HART,
Profibus, and Foundation Fieldbus, digital positioners provide the features
listed earlier, but can also provide the following additional benefits.
• Automatic calibration
Digital positioners can perform remote and automatic zero and
span calibration in a few minutes—a task that can take a few hours
with non-digital positioners.
• Characterization
The output signal can be characterized to match the system to
achieve a linear process with constant gain. The main advantage
over cam linearization is that it is performed on the output signal,
not the feedback from the valve stem.
• Digital noise filtering
While filters should be applied carefully, where appropriate, a dig-
ital filter time constant can be applied to minimize the effects of
excessive process noise. Users should remember filters add to pro-
cess response times. Applying a filter will likely require retuning
the control loop.
• Alarm generation
Users can assign positioner-based alarms, such as valve travel devi-
ation from the input signal, travel beyond a certain point, and oth-
ers. These alarms can be displayed on operator graphics.
• Maintenance-related data
Digital positioners can track valve reversal and total stem travel
data that can be correlated with time and actual maintenance events
to improve predictive maintenance forecasting.
• Valve stroke speed control

In applications where hydraulic hammering might occur a digital
positioner can be used to slow the valve stroke.
• Limiting the valve travel
Digital positioners allow the application of travel limits in cases
where a valve should never reach a fully closed position.
• Automation of installed valve performance testing
Digital positioners communicate well with control valve perfor-
mance software. Following a pre-defined test, data curves and
calculated results can be compared with previous tests to help
determine whether a control valve requires maintenance. Being
able to compare control valve performance can save time and
money during planned outages by focusing maintenance activities
on the control valves needing maintenance.
• Less susceptibility to vibration influences
The solid-state electronics in digital positioners provide a device
with few moving parts and help maintain positioner performance in
high-vibration installations.
CHAPTER 8
Valve Testing and

Diagnostics
In Chapter 1, we looked at the process gain: the percentage change in the

PV (Process Variable) as a result of a given percentage change in the PD
(Process Demand).
Another consideration is not by how much the PV changes, but how it
changes. A step change of the output will not necessarily result in the same
step change in the final control element. This may be as a result of a number
of factors that are usually described under the terms deadband and hysteresis.
8.1 Deadband and Hysteresis
Deadband is defined as the range through which an input can be varied with-
out initiating an observable response. In a mechanical system, deadband may
be, for example, as a result of mechanical play within a gear-train (Figure 8.1).
As the input is increased from point A, there will be no change in output until
the ‘slack’ is taken up and point B is reached—the extent of the deadband.
A well-engineered valve should respond to signals of 1% or less
to provide effective reduction in process variability. However, it is not
uncommon for some valves to exhibit deadband as great as 5% or more.
In a recent plant audit, 30% of the valves had deadbands in excess of 4%.
Over 65% of the loops audited had deadbands greater than 2%.
In a linear system, the output will now increase proportionally while
the input increases. However, when the input (points C to D) is reversed,
the output will again not change until the slack or deadband is taken up.
Imagine now, a gear-train system in which there is no play, but in which
there is an element of elasticity within the gears. As shown in Figure 8.2,
the system is no longer linear, and the output does not increase proportion-
ally with the input. This non-linearity is because of energy absorption that
D C
Output
Deadband
Output
Hysteretic
error
A B
Input
Figure 8.1. Deadband as a result Input

of mechanical play within a gear- Figure 8.2. An illustration of hysteretic
train. error.
Output
Hysteresis
Input
Figure 8.3. Hysteresis: A combination

of deadband and hysteretic error.
appears as heat. This is called the hysteretic error. When the hysteretic error
is summed with deadband (Figure 8.3), it is called hysteresis.
Hysteretic error normally manifests itself in a mechanical system that
is subject to a cyclic mechanical force or in a magnetic system that is
subject to a cyclic magnetizing force. In most pure electronic systems,
the hysteretic error can be effectively ignored, and thus the deadband and
hysteresis are one and the same.
It should also be noted that, in most mechanical-based systems, espe-
cially valves, the hysteretic error is referred to as hysteresis, and reference
is, therefore, made to ‘deadband and hysteresis.’
Friction is a major cause of deadband in control valves. Rotary valves
are often very susceptible to friction caused by the high seat loads required
Valve Testing and Diagnostics • 169
to obtain shut-off with some seal designs. Because of the high seal friction
and poor drive-train stiffness, the valve shaft winds up and does not trans-
late motion to the control element. As a result, an improperly designed
rotary valve can exhibit significant deadband that clearly has a detrimental
effect on process variability.
8.2 Testing Procedures*
*Some of the following information has been gleaned from notes pro-
duced by Michael Brown from Michael Brown Control Engineering
CC using the Protuner Loop Tuning software.
8.2.1 Deadband and Hysteresis
In order to determine the amount of hysteresis plus deadband in a control

valve, it is necessary to perform a series of step changes to the input and
to observe the changes in the stem position. The method illustrated here is
carried out using a Protuner.
The results are shown in Figure 8.4 and are the result of two steps
up, three down and one up. The first step up overcomes the effects of any
hysteresis or deadband to obtain a starting point for the testing.
Equal input
step changes
Valve stem
position indication % offset in valve
position
Figure 8.4. Determining the effects of hysteresis/deadband as a

result of an input change of two steps up, three down and one up
(courtesy Michael Brown Control Engineering).
The generally acceptable limits are:
• Spring and diaphragm 3%

• Spring and diaphragm with positioner 1%
• Piston with positioner 1%
• Variable speed drive 1%
8.2.2 Stick-Slip
‘Stick-slip’ response in a control valve is the result of a difference in the

static and sliding friction in the valve assembly. First, in order to move the
valve, the applied force is increased to a level where the static friction is
overcome. This is the ‘stick’ phase. Once the static frictional forces are
overcome, the sliding frictional forces are much smaller than the static
frictional forces and the valve ‘pop’ to a new position. This is the ‘slip’
phase. Stick-slip in the final control element can result in a continuous
limit cycle that can destabilize the process. The effects of stick-slip, with-
out hysteresis and deadband, are shown in Figure 8.5.
It should be noted that stick-slip is a completely different phenome-
non to hysteresis and deadband.
8.2.3 Non-Linearity
In an ideal control loop, there should be a linear relationship between the

PD and the PV—in other words, the process gain should be constant.
Slip
Stick
Figure 8.5. The effects of stick-slip, without

hysteresis and deadband (courtesy Michael
Brown Control Engineering).
50.0
37.5
PV
25.0
12.5
PD
0.0
Figure 8.6. As the PD increases in regular step, the PV

increases in a series of steps that gradually become smaller,
showing a marked non-linearity that is typical of an
oversized valve.
Testing for non-linearity is fairly easily accomplished by applying

a series of equal-interval steps to the PD as shown in Figure 8.6. Here,
PV is shown as a series of steps that gradually become smaller as the PD
increases and shows a marked non-linearity that is typical of an oversized
valve. Or, is it? The assumption here, of course, is that the PV is a linear
measurement. In the vast majority of cases, this would undoubtedly be
true, but it need not necessarily be so.
8.2.4 Testing a Complete Assembly
It should be appreciated that, in testing a complete final control element, it

is not only seal and packing friction within the valve itself that contributes
to deadband and hysteresis. Other factors include:
• inadequate air supply;

• loose or worn linkages in actuator connector;
• defective or improper calibration of I/P converter;
• loose or worn linkages in positioner;
• defective or improper calibration of positioner; and
• undersized actuator.
In order to test a complete pneumatically operated final control assem-

bly, an accurate valve stem position transmission should be connected to
140 kPa 2 140 kPa 4

20 – 100 kPa
1 3
Positioner
PD I/P 5 Mechanical
4 – 20 mA converter 20 – 100 kPa linkage
Valve stem
6 position
4 – 20 mA transmitter
Figure 8.7. Testing connections for a complete pneumatically operated final

control assembly.
the valve. In addition, pressure transducers should be connected to the

airlines, via quick-disconnect pneumatic fittings, at points 3, 4, and 5 as
shown in Figure 8.7. All six measuring points should then be connected to
a six channel analyzer/recorder.
Two series of tests should be carried out. The first test should involve
several steps in both directions, with a minimum of two steps up, three
down, and one up again. The valve should then be slowly ramped up and
down to test for stick-slip.
In evaluating the valve-installed dynamic operation, you should look
for the following points:
• Examine the change in the supply pressure during changes in the

valve position. If it drops during position changes, the supply line
size should be increased.
• Determine both the repeatability and linearity of the I/P converter
by recording both its input and output.
• Compare the PD signal (input to the I/P converter) with the valve
stem position to determine linearity, repeatability, and the response
time constant.
• Determine whether the valve is stroking correctly.
In normal modulating control applications, the valve should never be

operated either fully open or at flows of less than 20% of maximum. The
evaluations should, therefore, be carried out within these two limits.
One of the effects of an undersized actuator is shown in Figure 8.8. If
the actuator is unable to cope adequately with the static friction (stiction),
Equal input
step changes
Valve stem
position indication Overshoot indicates
negative hysterisis
Figure 8.8. Negative hysteresis: one of the effects

of an undersized actuator (courtesy Michael Brown
Control Engineering).
then, on reversal, the valve is effectively stuck. As a result, the valve posi-
tioner starts to force excess air pressure into the actuator. Once it moves
and the static friction is overcome, overshoot occurs. This characteristic is
likely to result in continuous cycling.
8.3 Online Diagnostics
The tests outlined above have been used for many years to perform anal-
yses on final control elements as a part of routine maintenance. Now,
however, using modern Fieldbus communication systems, together with
intelligent Smart positioners, a wide variety of information can be moni-
tored on a continuous basis to provide a real-time continuous process and
instrument diagnostic system.
At the simpler level, both the total valve stem travel (travel accumu-
lation) and the number of stem travel reversals (cycles) can be monitored
in order to determine ‘usage.’ Further, should a problem show up, users
can define how the instrument reacts to the problem. For example, if the
pressure sensor fails, should the instrument be shut down? Users can also
select which problems will cause the instrument to shut down. Such indi-
cators maybe reported as alerts to give an instant indication of any prob-
lem with the instrument valve or process.
Advanced diagnostics also include signature analysis that allows users
to determine the valve/actuator friction, Bench Set, spring rate, and seat load.
8.3.1 Signature Analysis
In the ‘valve signature’ plot (Figure 8.9), the actuator pressure (input) is
plotted on the y-axis, while the travel (output) is plotted along the x-axis.
Examination of this plot shows spikes (change of slope) at each end of
the curve, which verifies that a solid stop had been reached at both ends
of travel.
Next, the opening and closing lines should be parallel and linear
throughout the full stroke. The separation of these lines is the result of the
friction band—the higher the friction, the wider the separation. Because
friction opposes motion, in both directions, the separation between these
lines is actually double the friction (friction opposing the up stroke plus
friction opposing the down stroke).
The primary source of friction on a good valve is the valve packing.
Packing materials that have a high coefficient of friction, such as graphite,
will produce a greater amount of friction, and thus a wider bandwidth than
the low-coefficient materials, such as PTFE. Regardless of the packing
material, the separation (friction band) should remain constant throughout
the full travel.
The slope indicates that the actuator contains an opposing spring. If
there were no spring, the opening and closing lines would be nearly flat
Top stop
1.6
1.4 2 × Friction
Actuator pressure (bar)
1.2
Closed
ning ing
1.0 Ope Clos
0.8
0.6
0.4
0.2 Bottom stop
0
10 0 10 20 30
Travel (mm)
Figure 8.9. Plotting the ‘valve signature’ plot with the

actuator pressure plotted on the y-axis and the travel plot-
ted along the x-axis. The separation between the opening
(red) and closing (blue) lines is the result of the friction
band (courtesy Fisher-Emerson).
(horizontal), and thus the actuator spring and the effective area of the actu-
ator diaphragm govern the slope’s angle.
Examining this data allows a full analysis to be performed. For exam-
ple, by looking at individual pairings of adjacent upstroke/downstroke
data points between 10% and 90% of the travel range, it is possible to
derive the minimum, maximum, and average friction values.
The minimum friction value should never be less than 25% of the
expected friction value (20% if PTFE packing), while the maximum value
should never exceed 100% of the expected value.
Fairly obviously, there should not be a large difference between the
minimum and maximum values, which is the result of calculating the dif-
ference in actuator pressure between the upstroke and downstroke, times
the effective area of the actuator, divided by two.
Figure 8.10 shows that the packing friction is approximately
twice that of the previous example. Typically, this might be due to
errantly over-tightening the packing, resulting in the excessive friction.
Because the total amount of available actuator force is limited via the
installed spring, diaphragm area, and air supply, any additional force
required to travel the valve through any excess friction must come from
some area within this limitation. The only force available is the one
reserved for seat loading, and thus, any increase in friction will dimin-
ish seat load.
Top stop
1.6
1.4 2 × Friction
1.2
Closed
1.0 ning
Ope
0.8 ing
Clos
0.6
0.4
0.2 Bottom stop
0
10 0 10 20 30
Travel (mm)
Figure 8.10. The packing friction is approximately twice

that of the previous example. Typically, this might be due
to errantly over-tightening the packing, resulting in the
excessive friction (courtesy Fisher–Emerson).
Top stop
1.6
1.4
1.2
Closed
1.0 ning sin
g
Ope Clo
0.8
0.6
0.4
0.2
Bottom stop
0
10 0 10 20 30
Travel (mm)
Figure 8.11. An example of a valve signature showing

several revealed faults (courtesy Fisher–Emerson).
Figure 8.11 shows an example of a valve signature in which the large

increase in friction, as the valve travels towards the open position, is prob-
ably due to some form of galling.
8.4 Electronic Torque Monitoring
Generally, we have confined ourselves to diagnostics within a pneumatic

system. However, as indicated in Chapter 5, modern electrical actuators
can monitor both torque and position, allowing data relating to these
parameters to be used to compare the footprint torque characteristic of a
valve during installation to subsequent torque profiles.
As shown in Figure 8.12, the valve position (travel) is plotted on the
x-axis, while the torque demand is plotted on the y-axis. The plot illus-
trates the opening torque characteristics of a typical wedge gate valve.
The significant torque requirements for opening or closing a wedge
gate valve occur during the final travel going closed and the initial travel
going open. During the remaining portion, the torque demand is essen-
tially due to packing and thread friction. As the valve seats, the hydrostatic
force on the closure element (the disk) increases the seating friction, and
finally, the wedging effect of the disc into the seat causes a rapid increase
in torque until seating is affected.
Similarly, when under seating the valve, the disk has to be unwedged,
and the hydrostatic force of the differential pressure across the valve has
to be overcome as the valve is opened. Once the valve is cracked open and
Stem requires
Galled lubrication
seat
Torque demand
Packing
too tight
Footprint
torque profile
Closed Valve Position Open
Figure 8.12. Opening torque characteristics of

a typical wedge gate valve in which the valve
position (travel) is plotted on the x-axis and the
torque demand is plotted on the y-axis (courtesy
Rotork).
the differential pressure has dissipated, then the torque demand drops off
significantly. Should, for example, the valve stem packing be over-tight-
ened, then an immediate increase in torque profile would be recorded.
Should lubrication on the thread to deteriorate over time, then there would
be an incremental increase in overall torque. Alternatively, should the
valve seat become galled or deteriorate, then there would be an increase in
the unseating torque required.
CHAPTER 9
Valve Maintenance and

Repair
Valves are dynamic devices that exist in a dynamic environment.
• They have moving parts that wear.

• They have packing and sealing that age and lose their effectiveness.
• They are subject to the abrasive and corrosive effects of the fluids.
• They are subject to fluctuations in pressure and temperature of the
process of media.
• They are subject to fluctuations in the environmental temperature
and vibration.
Valves, therefore, need to be maintained, prepared, and some-

times replaced.
While maintenance on most valves is limited, repair possibilities are
extensive—limited only by economic considerations.
Apart from lubrication, maintenance consists mainly of correcting
external fluid leakage at the stem or shaft of gate, globe, ball, and butterfly
valves. In most cases, stem or shaft leakage can be stopped by tightening
the packing flange and nuts, which compresses the packing and forces it
tighter against the stem.
Any partial disassembly of the valve is termed a repair.
Repairs are required when:
• there is external leakage at the stem, shaft, or body joints, which can-
not be stopped by tightening packing nuts, flanges, or body bolting;
• closing the valve does not stop fluid flow;
• opening the valve does not allow flow to start;
• fluid leakage occurs through the valve shell due to erosion or corrosion;
• there is excessive deadband/hysteresis; and

• there is excessive slip-stick.
9.1 In-Line Repairs
The decision to make an in-line repair or send it out to a shop depends on

the nature and urgency of the repair and the ease of removal. In general,
repair in a shop is preferred over in-line repair.
In-line repairs are carried out when:
• repairs must be done promptly;

• if the valve is large or difficult to handle;
• the valve is in an awkward position; and
• the valve is welded into the line.
The extent of in-line repair is limited by the type and design of the
valve and whether the line has been pressurized drained.
9.2 Repairs Under Pressure
The only in-line repair that can be done while the line is still under pres-
sure is the replacement of stem packing on gate and globe valves having
back seats (Figure 9.1).
Back-seat Back-seat
Figure 9.1. Although not generally recommended, in-line repair can be

carried out while the line is still under pressure, on gate and globe valves
having back seats.
Valve Maintenance and Repair • 181
After tightly back-seating the stem, the valve packing nut is

unscrewed, or the gland flange nuts removed and the gland flange and the
gland lifted. The old packing can then be removed using mechanical pack-
ing pullers or high-pressure water sprays to blow out the packing. When
using mechanical pullers, take care not to scratch the stem or wall of the
stuffing box because scratches in these areas produce potential leak paths.
Although the old packing is usually either a continuous length of
material or solid rings, the new packing must be in the form of split rings,
so that it can be fitted around the stem, with the splits staggered around the
stem to prevent a direct leak path.
After installing the new packing, the packing or gland flange nuts are
reattached and tightened. The stem can then be carefully unseated from
the back seat and the packing or gland flange nuts tightened further to stop
any leakage.
Repacking a valve while it is under pressure is not a universally accepted
procedure. Some manufacturers recommend against it, stating that the back
seat should be used only as a temporary measure to minimize leakage until
repacking can be done in a depressurized condition. Furthermore, many
technicians refuse to carry out on-line repacking, particularly on valves car-
rying steam or other high-pressure or high-temperature fluids.
9.3 Repairs on Drained Systems
If the pipeline and valve have been drained, many valves allow the valve
bonnet cover to be removed, allowing removal of the flow control element
and exposing the body seating services.
9.4 Packing Replacement
Because the stem or shaft can be removed from the bonnet or cover, replac-
ing the stem packing is greatly facilitated, with the new packing either con-
tinuous material or solid rings. Whenever a valve is disassembled for any
repair, it is considered good practice to replace packing, seals, and gaskets.
9.5 Replacing or Refinishing Seat Rings
For top-entry valves, the removal and replacement of seat rings is often
easily accomplished, especially if they are screwed in. However, special
spanner wrenches may be required.
If the seat rings are tack welded to the body, the tack weld must be
cut away.
In the case of fully welded-in seat rings or integral seating surfaces,
these can be refinished, still with the valve in place, using special machin-
ery that clamps onto the valve. It may be necessary to perform further
grinding or lapping of the seat rings to get a good leak-tight fit.
9.6 Other In-Line Repairs
The balls in a ball valve are not easily refinished and are usually replaced.
However, this is seldom necessary because the soft seat rings experience
almost all the wear.
Diaphragm valve diaphragms are easily replaced and their fit with
their seating services is not a concern.
9.7 In-Line Post-Repair Procedures
After an in-line repair, the valve should be pressure-tested to check the

integrity of the body–bonnet or body–cover joint and seat tightness.
Because this would involve all or at least part of the pipeline, it may not
be possible to test to the pressures that would be used to test the valve
alone (normally 1.5 times the Maximum Working Pressure (MWP) of
the valve).
Consequently, the recommended practice is to test at 1.5 times the
design pressure of the pipeline.
Checking for seat leakage is also difficult after an in-line repair
because the seating services are not visible. As a result, seat leakage must
be checked at a point downstream of the valve.
9.8 Shop Repairs
Because there is much better control in a shop, the quality is better and
the valve can be more effectively tested for shell and seat tightness.
Furthermore, if the valve can be replaced with a spare one, the down-
time of the piping system may be less than with an in-line repair and the
repaired valve can then be become a spare.
During periodic shutdowns, it is a common practice to remove all
the large bore valves and send them to the shops for disassembly, inspec-
tion, and if necessary, repair. For economic reasons, small bore piping
systems, including their valves, are often scrapped and replaced, rather
than repaired.
Work performed in a valve shop should be controlled by a valve
repair specification that defines the repairs permitted and the quality of
work required. These specifications address the following topics:
1. Disassembly and cleaning.

• The match marking of parts to ensure proper reassembly and the
methods permitted for cleaning without damaging them.
2. Inspection.
• Define the visual and dimensional standards to be used to check
cleaned parts.
3. Evaluation of inspection results.
• Defining the criteria for the rejection of inspected parts and the
valve as a whole.
• A limit is usually set on the allowable cost of repair of a valve.
• A percentage of the cost of a new valve is usually established.
• If the cost of repair exceeds the percentage, the valve is scrapped.
4. Permitted repairs.
• Define the repairs allowed for specific valve parts.
• Should include requirements for the welding of parts, refinish-
ing of weld overlay of seating services, machining of stems and
shafts, and machining of flange faces.
5. Reassembly.
• Define the need to replace packing gaskets, seals, and bolting
and their requirements.
• Specify dimensional requirements for reassembled valves.
6. Testing.
• The need to pressure test each repaired valve is stated and the
specific testing requirements are defined. Such standards (e.g.,
ANSI) define the different tests required, test media, pressure
duration, and the allowable leakage.
7. Preparation for shipment.
• In painting and tagging requirements, the records necessary to
document the work performed or established.
9.9 Actuator Bench Set
Although definitions vary, at its simplest, the term ‘Bench Set’ entails
selecting the correct actuator spring. More formally, it is ‘the calibration
of the actuator spring range of a control valve to account for the in-ser-
vice process forces.’ For a given actuator/valve assembly, this involves
selecting the optimum actuator spring characteristics to achieve a mechan-
ical force equal to, or greater than, all the forces acting against the valve
throughout its rated travel range while in service. The forces acting against
the valve include: process forces, frictional forces, seating force, and
forces due to special assemblies (such as those with multiple piston rings).
Bench Set is, thus, a predetermined value that is established during
the actuator sizing procedure. The name, Bench Set, stems from the fact
that this test is usually performed on a workbench in the instrument shop,
prior to placing the valve into service.
Factors that influence the Bench Set span are: actuator size, spring
characteristics, and rated valve travel. For a given actuator/valve assem-
bly, the valve travel and diaphragm size are known, and thus, the Bench
Set establishes the required spring characteristics: the spring compression
ratio—sometimes referred to as the spring windup.
Often, setting the spring windup is carried out with the actuator dis-
connected from the valve so as not to introduce any forces, particularly
frictional forces. Typically, valves use two stops: one being the valve seat
and the other the top actuator stop. Since, with the actuator removed from
the valve, there is no way of knowing where the bottom stop (seat) will
be, the only common reference point for establishing the Bench Set would
be the upper stop of the actuator. With this in mind, the Bench Set spring
windup and span are established from the upper actuator stop down to the
rated travel of the valve on which the actuator will be installed.
9.10 Spring Calculations
But first, a recap on a few basics.

Pressure is defined as the ‘force per unit area’ with units of psi, mm
Hg, bar, and kPa.
One of the most well-known is lbs/inch2. So, the question if this is
correct, what about kg/m2?
The answer is no! Why? Because the term ‘pound’ is a unit of force,
while the term kilogram is unit of mass. In reality, the SI unit of force is
the Newton, which is defined as the force that gives a 1 kg mass an accel-
eration of 1 m/s2.
Consequently, SI pressure is defined as N/m2, or more commonly, the
Pascal (Pa). Practically, because the Pascal is so small, use is made of the
kilopascal (kPa) in which 100 kPa = 1 bar.
Now, some idea of the parameter’s values involved can be gained by

examining Figure 9.2: a nominal 50-mm-diameter valve.
Assume a process inlet pressure (P1) of 3 bar, and in its fully open
position, an outlet pressure (P2) of 2 bar, resulting in a ∆P of 1 bar (10 N/
cm2). For a 50 mm valve, the port area is 19.6 cm2, and thus, there is a
force of 19.6 ×10 = 196 N acting along the valve stem, which is tending
to open the valve, and which is assumed to be constant over the full length
of travel.
To start closing the valve, therefore, the actuator would need to exert
a downward force of 196 N at the minimum operating air input of 20 kPa
(2 N/cm2), necessitating a diaphragm area of 196/2 = 98.2 cm2.
Assuming that the actuator is ranged correctly, then, at the full instru-
ment air input of 100 kPa (10 N/cm2), the valve should be fully closed. At
that point, the closing force is 98.2 ×10 = 982 N; an excess of 982–196 =
786 N above that required to close the valve (Figure 9.3).
The absorption of this surplus force is the function of the ‘ranging’
spring. With, for example, a valve having a full travel of 1.75 cm, the
spring would need to have a compression rate of 786/1.75 = 450 N/cm.
In this manner, it is possible to choose the diaphragm size and the spring
compression ratio from a number of standard values in order to satisfy the
requirements of the valve, to within around 3 to 4 kPa.
From the foregoing, it is important to note that, before reusing a
control valve in a different application, mixing and matching valves and
EN
OP
P1 = 300 kPa P2 = 200 kPa
∆P = 100 kPa
Figure 9.2. Forces developed on a nominal

50-mm valve plug.
196 N 982 N
EN
OP
Figure 9.3. Unbalance of

forces.
actuators, adding or removing a positioner, changing components that

influence hysteresis, or adjusting spring compression, it is best to discuss
the planned changes with the original valve manufacturer.
CHAPTER 10
Safety Relief Valves
The primary function of a pressure relief device is to prevent overpres-

sure by automatically opening and releasing a volume of fluid (gas or
liquid) from within the vessel when a predetermined maximum pressure
is reached. Acting as a ‘last resort,’ these fully mechanical devices are
designed to open when an overpressure situation occurs within a process
pressure system, protecting not only life, but safeguarding the investment
and plant itself.
Such devices have been around since the 1600s and may be classified
as either reclosing or non-reclosing (Figure 10.1).
Pressure safety relief valves should be taken very seriously.
Manufactured from castings, they may not look very sophisticated, but in
their design, accuracy, and function, they resemble a delicate instrument,
while at the same time, performing an essential role. Self-contained and
self-operating, they respond to system conditions and prevent catastrophic
failure when other instruments and control systems fail to control process
limits adequately.
10.1 History
It is usually supposed that the Frenchman Papin was the inventor of the
safety valve, which he first applied about in 1682 to a digester. The valve
was kept closed by means of a lever and movable weight; sliding the
weight along the lever enabled Papin to keep the valve in place and regu-
late the steam pressure (Figure 10.2).
However, it now appears that safety valves were already in use some
50 years earlier by the German Glauber and that Papin only improved
on Glauber’s device. In Glauber’s treatise on philosophical furnaces,
Pressure Relief Devices
Pressure Relief Non-Reclosing

Valves Pressure Relief Devices
Weight Loaded Spring Loaded Pilot Operated Buckling/Shear

Pressure/Vacuum Pressure Relief Pressure Relief Rupture Disc
Pin
Relief Valve Valve Valve
Safety Valve Conventional Conventional Scored Tension
Relief Valve Balanced Composite Reverse Action
Safety Relief Graphite

Valve
Figure 10.1. Family of pressure relief devices, classified as either reclosing or

non-reclosing.
Figure 10.2. Papin’s safety valve was

kept closed by means of a lever and
movable weight. Sliding the weight
along the lever kept the valve in place
and regulated the steam pressure.
translated into English in 1651, he describes how he prevented retorts and

stills from bursting from an excessive pressure. A conical valve was fitted,
being ground air-tight to its seat, and loaded with a ‘cap of lead,’ so that
when the vapor became too ‘high,’ it slightly raised the valve and a portion
escaped. The valve then closed again on itself, ‘being pressed down by the
loaded cap and so kept closed.’
Safety Relief Valves • 189
The idea was followed up by others, and we find in the art of distilla-
tion, by John French, published soon afterward in London, the following
concerning the action of such safety valves:
Upon the top of a stubble (valve) there may be fastened some lead,
that if the spirit be too strong, it will only heave up the stubble and
let it fall down.
It should be realized that the word ‘steam’ was unknown at the time
and was only coined sometime later. In its place, we find the words ‘vapor,’
‘spirit,’ ‘smoke,’ and even, ‘ghost.’
In the early 1800s, there were literally thousands of boiler explosions
in the United States and Europe. However, there were no legal codes for
boilers in the United States. During the five years between 1905 and 1911,
there were 1,700 boiler explosions resulting in 1,300 deaths in the New
England region of the United States alone.
Boiler failure in Brockton, Massachusetts, on March 10, 1905, at the
Brockton Shoe Factory resulted in 58 deaths and 117 injuries, and com-
pletely leveled the factory.
In 1906, Massachusetts established a five-man Board of Boiler Rules,
whose charge was to write a boiler law for the state. This was published in
1908. And, in 1911, the State of Ohio enacted a boiler law.
By 1911, the year in which the ASME Council appointed a commit-
tee to formulate a boiler code, there were laws and regulations in effect
in 10 states and 19 metropolitan areas. The individual state require-
ments differed greatly from one another, so a boiler built in one state
could not be installed in another state. Consequently, the ASME Council
established the Boiler Code Committee to prepare a standard that could
be accepted by all states. The committee’s mission was to formulate a
standard specification for the construction of steam boilers and other
pressure vessels.
In 1916, the ASME Council approved the formation of the Conference
Committee to provide technical input, as it sees fit, to the additions and
revisions to the Boiler and Pressure Vessel Code.
All the provinces in Canada, 48 of the 50 states of the United
States, and various regulatory agencies around the world have
adopted, by law or regulation, various sections of the Boiler and
Pressure Vessel Code.
The American Society of Mechanical Engineers was asked to formu-
late a design code. The boiler and pressure vessel committee was formed,
and hence, the ASME Section 1 for fired vessels was formulated, becom-
ing a mandatory requirement for all states that recognized the need for
regulation. The sole purpose of a pressure-relieving device (safety relief

valve) is to protect life and property.
10.2 Definitions
The terms ‘safety valve’ and ‘safety relief valve’ are generic terms that
describe a variety of pressure relief devices designed to prevent excessive
internal fluid pressure build-up. One of the first problems encountered in
the field of pressure safety relief valves lies in the differences in terminol-
ogy used between the United States and Europe.
One of the most important is that a valve referred to as a ‘safety valve’ in
Europe is referred to as a ‘safety relief valve’ or ‘pressure relief valve’ in the
United States. Furthermore, the term ‘safety valve’ in the United States gen-
erally refers specifically to the full-lift type of safety valve used in Europe.
The European standards (BS 6759 and DIN 3320) provide the follow-
ing definition:
Safety valve: A valve that automatically discharges a certified amount

of fluid to prevent a predetermined safe pressure being exceeded.
When normal pressure conditions have been restored, the valve will
close and prevent further fluid flow.
In the United States, the ASME/ANSI PTC25.3 standards define the
following generic terms:
Pressure relief valve is a generic term that includes safety valves, relief
valves, and safety relief valves. In essence, a pressure relief valve
describes any spring-loaded device designed to open and relieve
excess pressure—re-closing after normal conditions have been
restored to prevent further fluid flow. All three varieties are similar
in design and operation, but have different applications.
Safety valves are primarily used with compressible gases, particularly
for steam and air services, and are characterized by a rapid-opening
‘pop’ action.
Relief valves are commonly used in liquid systems, especially for lower
capacities and thermal expansion duty, and are characterized by a
gradual lift, generally proportional to the increase in pressure over
opening pressure.
Safety relief valves may be used for either liquid or compressible fluid
applications and will perform as a safety valve when used in a com-
pressible gas system (characterized by a ‘pop’ action) and as a relief
valve when used in liquid systems (characterized by a proportional
opening action).
10.3 Weight-Loaded Pressure/Vacuum
Relief Valves
Direct-acting, weight-loaded valves are the simplest and least complex

type of pressure relief devices. Commonly referred to as weighted pallet
valves, they are frequently used on storage vessels requiring breather vents.
As illustrated in Figure 10.3, the weight of the seat assembly (or
pallet) keeps the valve closed until the pressure acting on the underside
equals this weight.
Normally designed for set pressures of less than 0.14 bar (2 psig),
weight-loaded valves are not ASME-coded.
A 30 cm (12 inch) valve (commonly used on large storage tanks) typ-
ically has a nozzle area of approximately 580 cm2 (90 inch2).
In order to obtain a set pressure of only 0.07 bar (1.0 psig), a mass
of 40 kg (90 lb) weight would be required. Consequently, large valves are
usually limited to pressures of 0.03 bar (0.5 psig) or less because of the
excessive weight required to obtain the set pressure.
When associated with tank breathing (1 to 4 in WC (0.04 to 0.15 psi)),
weight-loaded valves are often referred to as a ‘conservation valves’ and
comprise a combination vacuum plus a pressure valve to provide IN- and
OUT-breathing (Figure 10.4).
At these very low pressures, the relief valve/vent will almost always
be weight-loaded since it is not economically feasible to manufacture a
spring for such low pressures.
The main advantages of weight-loaded valves are their low cost,
their simple operation, and the ability to set very low pressures (down to
0.5 ounce/in2).
Force due to seat

weight Stem guide assembly
Seat stem
Seat assembly/pallet
Nozzle
Medium
pressure
Figure 10.3. The weight of the seat assembly (or

pallet) keeps the valve closed until the pressure acting
on the underside equals this weight.
Tank Connection Air Inlet
Figure 10.4. When associated with tank breathing

(1 to 4 in WC), weight-loaded valves are often
referred to as a ‘conservation valves’ and provide
IN- and OUT-breathing.
However, there are a number of limitations:
• The set pressure is not readily adjustable.

• Extremely long simmer and poor tightness.
• High overpressure necessary for full lift (100% or more in some
cases).
• The seat is easily frozen closed at low (cryogenic) temperatures.
• The weights required for higher set pressures become prohibitively
large.
• At higher pressures, many weighted pallet pressure relief valves
exhibit oscillation, resulting in seat plate damage.
• Accumulation of liquid can occur at the top of the plate.
10.4 Spring-Loaded Relief Valves
A relief valve is a direct spring-loaded pressure relief valve actuated by

the static pressure upstream of the valve and characterized by a gradual
lift proportional to the increase in pressure. Figure 10.5 illustrates a typical
relief valve where a spring force opposes the system pressure acting on the
valve disc. When the system pressure rises above the level of the spring
force, the valve opens.
The basic elements comprise a right-angle-pattern valve body with
the valve inlet connection, or nozzle, mounted on the pressure-containing
Cap
Stem
Adjusting screw
Spring
Bonnet
Seating
surface
Disk
Body
Nozzle
Figure 10.5. Spring-loaded pressure

relief valve.
system. The outlet connection may be screwed or flanged for connection

to a piped discharge system. However, in some applications, such as com-
pressed air systems, the safety valve will not have an outlet connection,
and the fluid is vented directly to the atmosphere.
The valves have closed bonnets to prevent the release of corrosive,
toxic, flammable, or expensive fluids and can be supplied with lifting levers
(Figure 10.6), balancing bellows, and soft seats as needed. The ASME
Code requires that liquid service relief valves installed after January 01,
1986, have their capacity certified and stamped on the nameplate.
The closing force on the disc is provided by a spring, typically made
from carbon steel. The amount of compression on the spring is usually
adjustable, using the spring adjuster, to alter the pressure at which the disc
is lifted off its seat.
A relief valve begins to open when the static inlet pressure reaches set
pressure. When the static inlet pressure overcomes the spring force, the disc
begins to lift off the seat, allowing flow of the liquid. The value of the clos-
ing pressure is less than that of the set pressure and will be reached after the
blowdown phase is completed. Relief valves usually reach full lift at either
10% or 25% overpressure, depending on the type of valve and trim.
Spring
adjusting screw
Test lever
Spring
Stem
Nozzle
Figure 10.6. The valves have

closed bonnets to prevent the release
of corrosive, toxic, flammable, or
expensive fluids and can be supplied
with lifting levers.
10.5 Applications
Relief valves are normally used for incompressible fluids (see Part I of
API RP 520).
10.6 Limitations
Relief valves should not be used as follows:
• In steam, air, gas, or other vapor services.

• In services piped to a closed header unless the effects of any con-
stant or variable back-pressure have been accounted for.
• As pressure control or bypass valves.
10.7 Safety Valves
A safety valve is a direct spring-loaded pressure relief valve that is actu-

ated by the static pressure upstream of the valve and characterized by rapid
opening or pop action. When the static inlet pressure reaches the set pres-
sure, it will increase the pressure in the huddling chamber and overcome
the spring force on the disc. This will cause the disc to lift and provide full
opening at minimal overpressure. The closing pressure will be less than the
set pressure and will be reached after the blowdown phase is completed.
The spring of a safety valve is usually fully exposed, outside of the
valve bonnet to protect it from degradation due to the temperature of the
relieving medium. A typical safety valve (Figure 10.7) has a lifting lever
for manual opening to ensure the freedom of the working parts. Open bon-
net safety valves are not pressure-tight on the downstream side.
Standards that govern the design and use of safety valves generally
only define the three dimensions that relate to the discharge capacity of
the safety valve, namely, the flow (or bore) area, the curtain area, and the
discharge (or orifice) area (Figure 10.8).
• Flow area: The minimum cross-sectional area between the inlet

and the seat, at its narrowest point. The diameter of the flow area is
represented by dimension ‘d’ in Figure 10.8.
Spring adjusting
screw
Test lever
Spring
Stem
Upper adjusting
Lower adjusting ring
ring
Figure 10.7. The spring of a safety valve is usually

fully exposed and has a lifting lever for manual
opening.
D
Curtain area L
Flow area
d
Flow
Flow
Figure 10.8. Illustration of the standard defined areas.
• Curtain area: The area of the cylindrical or conical discharge open-

ing between the seating surfaces created by the lift of the disc above
the seat. The diameter of the curtain area is represented by dimen-
sion ‘d1’ in Figure 10.8.
p ⋅ d2
Flow area = (10.1)
4
Curtain area = p ⋅ d1 ⋅ L (10.2)
• Discharge area: This is the lesser of the curtain and flow areas,
which determines the flow through the valve.
10.8 Basic Operation: Lifting
When the inlet static pressure rises above the set pressure of the safety
valve, the disc will begin to lift off its seat. However, as soon as the spring
starts to compress, the spring force will increase; this means that the pres-
sure would have to continue to rise before any further lift can occur and for
there to be any significant flow through the valve. The additional pressure
rise required before the safety valve will discharge at its rated capacity is
called the overpressure. The allowable overpressure depends on the stan-
dards being followed and the particular application. For compressible flu-
ids, this is normally between 3% and 10%, and for liquids, between 10%
and 25%.
In order to achieve full opening from this small overpressure, the disc
arrangement has to be specially designed to provide rapid opening. This is
usually done by placing a shroud, skirt, or hood around the disc (Figures
10.9 and 10.10). The volume contained within this shroud is known as the
control or ‘huddling’ chamber.
As lift begins (Figure 10.10 (b)) and fluid enters the chamber, a larger
area of the shroud is exposed to the fluid pressure. Since the magnitude
of the lifting force (F) is proportional to the product of the pressure (P)
and the area exposed to the fluid (A), (F = P ¥ A), the opening force is
increased. This incremental increase in opening force overcompensates
for the increase in spring force, causing rapid opening. At the same time,
the shroud reverses the direction of the flow, which provides a reaction
force, further enhancing the lift.
These combined effects allow the valve to achieve its designed lift
within a relatively small percentage overpressure. For compressible fluids,
an additional contributory factor is the rapid expansion as the fluid volume
increases from a higher to a lower pressure area. This plays a major role
in ensuring that the valve opens fully within the small overpressure limit.
Huddling Disc
chamber Shroud
Figure 10.9. Typical disc and shroud

arrangement used on rapid opening
safety valves.
(a) (b) (c)
Figure 10.10. Operation of a conventional safety valve.

For liquids, this effect is more proportional, and subsequently, the over-
pressure is typically greater; 25% is common.
10.9 Basic Operation: Reseating
Once normal operating conditions have been restored, the valve is

required to close again, but since the larger area of the disc is still exposed
to the fluid, the valve will not close until the pressure has dropped below
the original set pressure. The difference between the set pressure and this
reseating pressure is known as the ‘blowdown,’ and it is usually specified
as a percentage of the set pressure (Figure 10.11). For compressible flu-
ids, the blowdown is usually less than 10%, and for liquids, it can be up
to 20%.
The design of the shroud must be such that it offers both rapid opening
and relatively small blowdown, so that, as soon as a potentially hazardous
situation is reached, any overpressure is relieved, but excessive quantities
of the fluid are prevented from being discharged. At the same time, it is
necessary to ensure that the system pressure is reduced sufficiently to pre-
vent immediate reopening.
In order to achieve a significant lifting force without an extremely
long blowdown, a ring is threaded around the valve nozzle and positioned
to form a huddling chamber with the disc skirt (Figure 10.12).
Although this ring, shown as the ‘nozzle ring,’ is commonly called
a blowdown ring, its function is also very important for controlling the
valve opening.
Pressure is generated in the huddling chamber when gas or vapor
flows past the seat. This pressure, acting over a larger area than the seat
sealing area, increases, and thus creates an instantaneous amplification of
the upward force, and the seat disc rapidly lifts off the nozzle.
100% Maximum
discharge
% lift Closing
Opening
Set
pressure Pop
action
Reseat
10% Blowdown Overpressure 10%
Figure 10.11. Relationship between pressure and lift for a

typical safety valve.
Spring
Disc
Huddling
chamber Nozzle ring
Nozzle
Figure 10.12. Blowdown ring is

threaded around the valve nozzle and
positioned to form a huddling chamber
with the disc skirt.
This initial lift of the seat disc is enough to establish 60–75% full-
rated flow, driving the seat disc up to the change in momentum and the
expansion of the gas can sustain lift.
When the blowdown ring is adjusted up (Figure 10.13), the forces
required to lift the seat disc off the nozzle occur at a pressure very close to
set pressure. The reason for this is that the huddling chamber is restricted
and gas flowing into the chamber quickly pressurizes it. However, in this
position, the blowdown is long because the pressure between the seat disc
skirt and the ring remains high, preventing the seat disc from losing lift
until the pressure under the disc reduces to a much lower value.
Increases blowdown
Reduces simmer
Blowdown
ring
Decreases blowdown
Increases simmmer
Figure 10.13. When the blowdown ring is

adjusted up, the forces required to lift the seat
disc occur very close to set pressure and the
blowdown is long. When the ring is adjusted
down, the seat lift does not occur until the
pressure under the seat disc is considerably
higher and the blowdown is short.
When the ring is adjusted down, the forces required to lift the seat
disc off the nozzle do not occur until the pressure under the seat disc is
considerably higher. This is because the huddling chamber exit area is
less restricted and considerably more gas must flow into the chamber to
pressurize it. With the ring in this position, the blowdown is short since the
pressure between disc holder skirt and ring quickly decreases when the lift
of the seat disc is decreased.
10.10 Conventional Safety Relief Valves
A conventional safety relief valve is a direct spring-loaded valve whose

operational characteristics (opening pressure, closing pressure, and reliev-
ing capacity) are directly affected by changes in any back pressure in the
discharge system.
Conventional safety relief valves can be used in refinery and pet-
rochemical processes that handle flammable, hot, or toxic material.
However, the effect of temperature and back-pressure on the set pressure
must be considered when using them.
10.10.1 Limitations
Conventional safety relief valves should not be used in the following

applications:
• Where any built-up back pressure exceeds the allowable overpressure.

• Where the Cold Differential Test Pressure (CDTP) cannot be
reduced to account for the effects of variable back pressure (see
API RP 520 Part I).
• On ASME Section I steam boiler drums or ASME Section I
super-heaters.
10.10.2 Back-Pressure
It is important to note that the total back-pressure is generated from two

components:
• Superimposed back-pressure. The static pressure that exists on the

outlet side of a closed valve.
• Built-up back-pressure. The additional pressure generated on the

outlet side when the valve is discharging.
Subsequently, in a conventional safety valve, only the superimposed

back-pressure will affect the opening characteristic and set value, but the
combined backpressure will alter the blowdown characteristic and re-seat
value.
The ASME/ANSI standard makes the further classification that
a conventional valve has a bonnet that encloses the spring and forms a
pressure-tight cavity, with the bonnet cavity vented to the discharge side
of the valve.
If the spring housing is vented to atmosphere, any superimposed
back-pressure will still affect the operational characteristics. This is illus-
trated in Figure 10.14, which shows schematic diagrams of a valve whose
spring housing is vented to the discharge side of the valve (an ASME
conventional safety relief valve.)
Consider the forces acting on the disc (with area AD), the required
opening force (equivalent to the product of inlet pressure (PV), and the
nozzle area (AN)) is the sum of the spring force (FS) and the force due to
the back-pressure (PB) acting on the top and bottom of the disc:
Vented to
discharge
PB
Spring
FS
Disk area
AD PB PB
Back pressure
PB PB PB
Nozzle area
AN Nozzle
pressure
PN
Figure 10.14. Schematic diagram of a valve with

the spring housing vented to the discharge side of
the valve.
PV ⋅ A N = FS + PB ⋅ A D − PB ⋅ ( A D − A N ) (10.3)
where:
PV = Fluid inlet pressure
AN = Nozzle area
FS = Spring force
PB = Back-pressure
which simplifies to:
PV ⋅ A N = FS + PB ⋅ A N (10.4)
This shows that any superimposed back-pressure will tend to increase

the closing force and the inlet pressure required to lift the disc is greater.
Figure 10.15 show a valve whose spring housing is vented to the
atmosphere.
Here, the required opening force is:
PV ⋅ A N = FS − PB ⋅ ( A D − A N ) (10.5)
This shows that the superimposed back-pressure acts with the vessel
pressure to overcome the spring force and the opening pressure will, as a
Vented to
atmosphere
Spring
FS
Disk area
AD
Back pressure
PB PB PB
Nozzle area
AN Nozzle
pressure
PN
Figure 10.15. Schematic diagram of a valve with

spring housing vented to the atmosphere.
result, be less than expected. In both cases, if a significant superimposed

back-pressure exists, its effects on the set pressure need to be considered
when designing a safety valve system.
Once the valve starts to open, the effects of built-up back-pressure
also have to be taken into account.
In a conventional valve, with the spring housing vented to the dis-
charge side, once the valve starts to open, the inlet pressure (PV) now
becomes the sum of the set pressure, PS, and the overpressure, PO. This
modifies the opening force, which now becomes:
( PS + PO ) ⋅ A N = FS + PB ⋅ A N (10.6)
which is simplified to:
PS ⋅ A N = FS + A N ⋅ ( PB − PO ) (10.7)
where:
PS = Set pressure of safety valves
AN = Nozzle area
FS = Spring force
PB = Back-pressure
PO = Overpressure
From this, it can be seen that, if the backpressure is greater than the
overpressure, the valve will tend to close, reducing the flow. This can lead
to instability within the system and can result in flutter or chatter of the
valve.
In general, if a conventional safety valve is used in applications,
where there is an excessive built-up back-pressure, it will not perform as
expected. According to the API 520 Recommended Practice Guidelines:
A conventional pressure relief valve should typically not be used

when the built-up backpressure is greater than 10% of the set
pressure at 10% overpressure. A higher maximum allowable
built-up backpressure may be used for overpressure greater
than 10%.
The British Standard BS 6759, however, states that the built-up

back-pressure should be limited to 12% of the set pressure when the valve
is discharging at the certified capacity.
For many applications, the back-pressure can be maintained within
these limits by carefully sizing of the discharge pipes. If this is not feasi-
ble, then it may be necessary to use a ‘balanced’ safety valve.
10.11 Balanced Safety Relief Valves
A balanced safety relief valve is a direct spring-loaded pressure device

that incorporates a bellows or piston arrangement to minimize the effect
of back-pressure on the operational characteristics.
Balanced safety relief valves are normally used in refinery and
petrochemical process industries that handle flammable, hot, or toxic
material, where high back-pressures are present at the valve discharge.
This typically occurs where material from the valve is routed to a collec-
tion system. They are used as follows:
• In gas, vapor, steam, air, or liquid services.

• In corrosive service to isolate the spring, bonnet cavity, and dis-
charge side of the valve from process material.
• When the discharge from the valves must be piped to remote
locations.
10.11.1 Limitations
Balanced safety relief valves should not be used as follows:
• On ASME Section I steam boiler drums or ASME Section I

super-heaters.
There are two basic designs of balanced safety relief valves: bellows
type and piston type.
10.12 Bellows-Type Balanced Safety Valve
In the bellows-type balanced safety valve (Figure 10.16), a bellows with

an effective area (AB) equivalent to the nozzle seat area (AN) is attached
to the upper surface of the disc and to the spindle guide (Figure 10.17).
The bellows prevents back-pressure acting on the upper side of the
disc within the area of the bellows. The disc area extending beyond the
bellows and the opposing disc area are equal, and so the forces acting on
the disc are balanced, and the back-pressure has little effect on the valve
opening pressure. The bellows vent allows air to flow freely in and out of
the bellows as they expand or contract.
Cap
Stem
Adjusting screw
Bonnet
Spring Bonnet vent
Bellows vent
Bellows
Disk holder
Disk
Body
Nozzle
Figure 10.16. Bellows-type balanced safety

valve.
Spring FS
Bonnet vent
Spindle guide Bellows vent
AB AB
Disk
Bellows
AN
PV
Figure 10.17. Block schematic of a bellows-

type balanced safety valve showing force
balancing.
Balanced-type valves require vented bonnets. A bellows failure allows

process media from the discharge side of the valve to discharge from the
bonnet vent. Consider the nature of the process media (e.g., liquid/vapor,
toxicity, and flammability) when evaluating the bonnet vent disposition.
Bonnet vents are typically routed to a drain, a closed system, or atmo-

sphere, depending on the process media involved.
Bellows failure is an important concern when using these valves since
this may affect the set pressure and capacity of the valve. It is important,
therefore, that there is some mechanism for detecting any uncharacteristic
fluid flow through the bellows vents. In addition, some bellows balanced
safety valves include an auxiliary piston that is used to overcome the
effects of back-pressure in the case of bellows failure. This type of safety
valve is usually only used in critical applications in the oil and petrochem-
ical industries.
In addition to reducing the effects of back-pressure, the bellows also
serves to isolate the spindle guide and the spring from the process fluid;
this is important when the fluid is corrosive.
Since balanced pressure relief valves are typically more expensive
than their unbalanced counterparts, they are commonly only used where
high-pressure manifolds are unavoidable, or in critical applications where
a very precise set pressure or blowdown is required.
10.13 Piston-Type Balanced Safety Valve
Although there are several variations of the piston valve, they generally
comprise a piston-type disc whose movement is constrained by a vented
guide. The area of the top face of the piston, AP, and the nozzle seat area,
AN, are designed to be equal. This means that the effective area of both
the top and bottom surfaces of the disc exposed to the back-pressure are
equal, and therefore, any additional forces are balanced. In addition, the
spring bonnet is vented such that the top face of the piston is subjected to
atmospheric pressure, as shown in Figure 10.18.
By considering the forces acting on the piston, it is evident that this
type of valve is no longer affected by any back-pressure:
PV ⋅ A N = FS + PB ⋅ ( A D − A P ) − PB ⋅ ( A D − A N ) (10.8)
where:
PV = Fluid inlet pressure
AN = Nozzle area
FS = Spring force
PB = Back-pressure
AD = Disc area
AP = Piston area
Spring
FS Bonnet vent
AP
Piston
PB PB PS
AD
Vent
Disk
PB PB
AN AP = AN
PV
Figure 10.18. Block schematic of piston-

type balanced safety valve showing force
balancing.
Since AP equals AN, the last two terms of the equation are equal in
magnitude and cancel out. This simplifies to:
PV ⋅ A N = FS (10.9)
10.13.1 Pilot-Operated Pressure Relief Valve
A pilot-operated safety relief valve is a pressure relief valve in which the

major relieving device or main valve is combined with, and controlled by,
a self-actuated auxiliary pressure relief valve (pilot). As with the spring-
loaded valve, many unique models exist. However, some common design
features include: the sensing line, the pilot valve, and the main valve.
Depending on the design, the pilot valve (control unit) and the main valve
may be mounted on either the same connection or separately.
Pilot-operated safety relief valves offer a number of advantages over
conventional safety relief valves including: good overpressure and blow-
down performance and use in applications where a large relief area and/or
high set pressures are required.
The pilot is a spring-loaded valve that operates when its inlet static
pressure exceeds its set pressure. This causes the main valve to open and
close according to the pressure. Process pressure is either vented-off by
the pilot valve to open the main valve or applied to the top of the unbal-
anced piston, diaphragm, or bellows of the main valve to close it.
The two most commonly used pilot-operated safety valves are the pis-
ton and diaphragm types. Figure 10.19 shows a high-pressure pilot-operated
valve that uses an unbalanced piston and has an integrally mounted pilot.
The piston and seating arrangement in the main valve is designed so
that the bottom area of the piston, exposed to the inlet fluid, is less than
the top area of the piston. Since both ends of the piston are exposed to the
fluid at the same pressure, under normal system operating conditions, the
closing force, resulting from the larger top area, is greater than the inlet
force. The resultant downward force holds the piston firmly on its seat
(Figure 10.20).
If the inlet pressure rises, the net closing force on the piston also
increases, ensuring that a tight shut-off is continuously maintained.
However, when the inlet pressure reaches the set pressure, the pilot valve
will pop open to release the fluid pressure above the piston. With much less
fluid pressure acting on the upper surface of the piston, the inlet pressure
generates a net upward force, and the piston will leave its seat. This causes
the main valve to pop open, allowing the process fluid to be discharged.
When the inlet pressure has been sufficiently reduced, the pilot valve
will reclose, preventing the further release of fluid from the top of the
piston, thereby re-establishing the net downward force and causing the
piston to re-seat.
Set pressure
adjustment screw
Pilot piston
External blowdown
adjustment
Seat
Optional pilot
filter
Pilot
supply
Outlet Piston line
Seat
Internal pressure
pickup
Main valve
Inlet
Figure 10.19. High-pressure pilot-operated valve incorporating

an unbalanced piston and an integrally mounted pilot.
Cover
Spring
Soft disc
Metal disc
Guide Disc
Sliding rings
Disc holder
Body
Nozzle
Figure 10.20. Alternative seating arrangements available for a

pilot-operated piston-type safety relief valve.
Figure 10.21 shows a diaphragm-type pilot-operated valve that is,

typically, only available for use in low-pressure applications and pro-
duces a proportional type action—characteristic of relief valves used in
liquid systems.
Figure 10.21. Low-pressure diaphragm-type

pilot-operated valve.
Pilot-operated safety valves offer good overpressure and blowdown

performance (a blowdown of 2% is attainable). For this reason, they are used
where a narrow margin is required between the set pressure and the system
operating pressure. Pilot-operated valves are also available in much larger
sizes, making them the preferred type of safety valve for larger capacities.
One of the main concerns with pilot-operated safety valves is that the
small bore, pilot-connecting pipes are susceptible to blockage by foreign
matter, or due to the collection of condensate in these pipes. This can lead
to the failure of the valve, either in the open or closed position, depending
on where the blockage occurs.
The British Standard BS 6759 states that all pilot-operated safety
valves should have at least two independent pilot devices, which are con-
nected individually and arranged such that failure of either of the pilot will
still enable the safety valve to continue to operate effectively.
10.13.2 Applications
Pilot-operated safety relief valves are generally used as follows:
• Where a large relief area and/or high set pressures are required,
since pilot-operated valves can usually be set to the full rating of
the inlet flange.
• Where a low differential exists between the normal vessel operating
pressure and the set pressure of the valves.
• On large low-pressure storage tanks (see API Standard 620).
• Where very short blowdown is required.
• Where back-pressure is very high and balanced design is required,
since pilot-operated valves with the pilots either vented to the atmo-
sphere or internally balanced are inherently balanced by design.
• Where process conditions require sensing of pressure at one loca-
tion and relief of fluid at another location.
• Where inlet or outlet piping frictional pressure losses are high.
• Where in-situ, in-service, set pressure verification is desired.
10.13.3 Limitations
Pilot-operated safety relief valves are not generally used as follows:
• In service where fluid is dirty, unless special provisions are taken

(such as filters, sense line purging, etc.)
• In viscous liquid service, as pilot-operated valve operating times

will increase markedly due to flow of viscous liquids through rela-
tively small passages within the pilot.
• With vapors that will polymerize in the valves.
• In service where the temperature exceeds the safe limits for the
diaphragms, seals, or O-rings selected.
• Where chemical compatibility of the lading fluid with the dia-
phragms, seals, or O-rings of the valves is questionable.
• Where corrosion build-up can impede the actuation of the pilot.
10.14 Non-Reclosing Pressure
Relief Devices
While pressure relief valves are designed to automatically reset to their

previous condition once the overpressure condition has ceased, non-re-
closing pressure relief devices generally require the device to be replaced.
In essence, non-reclosing pressure relief devices are available in two
forms: buckling/shear pin devices and rupture discs.
10.14.1 Buckling/shear pin devices
10.14.1.1 Shear-Pin Safety Valve
The shear-pin safety valve is similar in construction to a spring-loaded

valve, but instead of a spring, use is made of a shear-pin (Figure 10.22). As
shown, the valve spindle slides in a guide, with both drilled to accept the
shear-pin. When the set pressure is reached, the force exerted on the disc
equals the failure shear force of the pin. The use of two pins can extend the
set pressure range up to 5,000 psi.
The shear-pin failure force is dictated by the pin’s material strength
and cross-sectional area. Because of the lack of precision in determining
the shear failure rate, this form of device is not ASME-approved.
10.14.1.2 Buckling Pin Safety Valve
The basic buckling pin valve (Figure 10.23) comprises a pin of a precise
length that holds a piston on its seat. The pin ends are restrained for
precise, repeatable operation.
Spindle guide
Shear pin
Spindle
Figure 10.22. Similar in construction

to a spring-loaded valve, but using a
shear-pin in place of a spring.
Buckling pin
Figure 10.23. The basic buckling pin valve comprises a pin of a precise
length that holds a piston on its seat. As the pressure increases and the
axial force on the pin subsequently also increases, the pin will buckle.
As the pressure increases and the axial force on the pin subsequently
also increases, the pin will buckle. This ‘buckling’ point is based on
Euler’s law of Compressed Columns that states:
E ⋅ d4
F≈ (10.10)
L2
where:
F = axial force causing the pin to buckle
E = pin modulus of elasticity
d = pin diameter
L = pin length
When the set pressure is reached, the pin buckles and the piston rap-
idly moves off its seat to relieve the pressure.
The pin has only two stable conditions: straight or buckled and cannot
fail early due to fatigue or pulsation.
The buckling point is accurately repeatable with no adverse buckling
point variation. The standard buckling point is ±5% of set point with ±2%
available with valve test certificates.
Features include:
• The pin is external from aggressive system fluid.

• Proximity detector may be used to detect buckling.
• Buckling pin can be changed in minutes.
• Buckling pin valve accepted by:
 ASME Section VIII Division I Code Case 2091-3
 API RP-520 Part One, Section 2.4
 The National Board of Boiler and Pressure Vessel Inspectors
10.14.2 Burst Disc
Also known as a rupture disc, bursting disc, or burst diaphragm, the disc
is usually made out of metal and designed to rupture at a predetermined
pressure. Once the disc has ruptured, it will not re-seal (Figure 10.24).
Apart from their low cost and almost instantaneous (milliseconds)
response, rupture discs provide a number of other advantages that are spe-
cific to a wide range of applications:
• Protection of the upstream side of a pressure relief valve against

corrosion by the system fluid.
• Protection of the upstream side of a pressure relief valve against
plugging or clogging by viscous fluids or polymerization p roducts.
• In place of a pressure relief valve if the protected system can
tolerate process interruptions or loss of fluids in case the disc
ruptures.
• In place of a pressure relief valve if extremely fast response is
desirable.
Figure 10.24. Also known as a rupture disc, burst-

ing disc, or burst diaphragm, the disc is designed
to rupture at a predetermined pressure and, once
ruptured, will not re-seal (courtesy Oseco).
• As a secondary pressure-relieving device when the difference

between the operating pressure and the rupture pressure is large,
depending on the type of rupture disc selected.
• To protect the downstream side of a pressure relief valve against
downstream corrosion from headers or against atmospheric
corrosion.
• To minimize process/product leakage and reduce fugitive emissions.
Bursting discs may be forward- or reverse-acting, depending on

whether the pressure forces are acting on the concave (forward-acting) or
convex (reverse-acting) faces (Figure 10.25).
The operating pressure of forward-acting discs is usually limited to
65–85% of the disc’s predetermined bursting pressure. Reverse-acting
discs are able to support pressures much closer to their rated burst pres-
sure, typically up to 90% of the disc design bursting pressure.
Rupture discs are usually held in place by a rupture disc holder
(Figure 10.26), although some discs are designed to be installed between
standard flanges without holders.
Rupture discs are available in several configurations that include:
• Conventional rupture disc (pre-bulged)

• Scored tension-loaded rupture disc
Reverse
Forward acting
acting
Figure 10.25. Bursting discs may be forward- or
reverse-acting.
Rupture disc
Disc tab
FLOW
DIRECTION
Pressure
Figure 10.26. Typical rupture disc holders (courtesy Oseco).
• Composite rupture disc (low pressure)

• Graphite rupture discs (corrosion-resistant)
10.15 Conventional Rupture Disc
A conventional domed rupture disc (Figure 10.27) is a pre-bulged solid

metal disc designed to burst when it is overpressured on the concave side.
They generally provide satisfactory service life when operating conditions
are 70% or less of the rated burst pressure of the disc. Special designs are
available for back-pressures exceeding normal atmospheric (15 psi).
Conventional domed rupture discs will fragment on bursting.
10.16 Scored Tension-Loaded
Rupture Disc
Designed to open along scored lines, scored tension-loaded rupture discs

(Figure 10.28) allow a closer ratio (generally 85%) of system operating
pressure to disc burst pressure.
Figure 10.27. Conventional domed rupture discs are pre-

bulged solid metal discs designed to burst when operating
conditions are 70% or less of the rated burst pressure.
Figure 10.28. Scored tension-loaded

rupture discs allow a closer ratio (generally
85%) of system operating pressure to disc
burst pressure.
10.17 Composite Rupture Disc
Composite rupture disks (Figure 10.29) are designed for applications

where lower rupture pressures are required than can be achieved with
standard rupture discs. Typically, a composite bursting disc comprises
a slotted metal top section together with a plastic membrane located
on the concave or pressure side. The top section is the pressure zone
and controls the bursting rating of the disc. Since the top section has
open slots, the membrane isolates it from process media and prevents
leakage.
Composite rupture disks provide better corrosion resistance and are
often used as corrosion barriers.
Figure 10.29. Composite rupture disc

(courtesy Continental Disc Corp.).
10.18 Graphite Rupture Disc
A graphite rupture disc is manufactured from graphite impregnated with a

binder material and designed to burst by bending or shearing (Figure 10.30).
Graphite rupture discs are resistant to most acids, alkalis, and organic
solvents. Operation to 70% of the rated burst pressure is generally permis-
sible. A support may be required for discs that are rated 15 psi or less and
for conditions of higher back-pressure.
Figure 10.30. Graphite rupture

disc manufactured from graphite
impregnated with a binder material
and designed to burst by bending
or shearing (courtesy Svi Carbon
Private Limited).
Graphite rupture discs fragment upon rupture, and thus, provisions

for capturing fragments may be required in certain applications.
The service life of pre-bulged metal rupture discs under normal
operating conditions is usually one year. They are subject to relatively
rapid creep stress failure, especially at high operating temperatures. If not
replaced periodically, they may fail without warning at normal operating
pressures.
Higher operating pressures (up to 90% of the disc design bursting
pressure) are possible with the reverse-buckling disc. Less fatigue due
to pulsating and cyclic operating pressure results in a longer service life
than would be expected if the disc were installed with the pressure acting
against the concave side. Most reverse-buckling discs should not be used
in liquid full service. However, if an assured gas pocket rests against the
disc and the disc manufacturer is consulted, liquid service can be consid-
ered. With finite lives, these discs should be replaced periodically. Consult
the manufacturer for recommended replacement times.
10.19 Burst Disc Applications and

Installation Practices
Impervious graphite rupture discs offer nearly the same advantages and
disadvantages as the reverse-buckling, metal type. However, with imper-
vious graphite rupture discs, the piping arrangement may be more com-
plicated, and uneven flange bolt loads or thermal strains in the piping may
crack the disc.
Rupture discs that tend to fragment, such as conventional and graph-
ite discs, are typically not installed beneath pressure relief valves unless
a means of protecting the pressure relief valve inlet from the fragments is
provided.
Caution: When rupture discs are removed for inspection or when an
accompanying relief valve is serviced, the discs can easily be damaged
and can fail prematurely if reused. Replacement of discs at every mainte-
nance interval will minimize the chance of damage and premature failure.
The proper receipt, storage, handling, and installation of a rupture
disc are critical to its successful performance. The manufacturer’s instal-
lation instructions must be adhered to, especially those concerning limits
on bolt torque.
Some rupture discs using knife blades to open have failed to open
properly. Consultation with the manufacturer concerning proper installa-
tion and maintenance of these kinds of rupture discs may be beneficial.
Tell-tale pressure
indicator
Rupture
disk
Figure 10.31. Bursting disc

installed on a safety valve.
A pressure gauge, a try cock, a free vent, or a suitable tell-tale indica-

tor must be inserted between a rupture disc device installed at the inlet of a
pressure relief valve and the valve (Figure 10.31), permitting the detection
of disc rupture or leakage.
Since rupture discs are designed to burst at a specified differential
pressure, pressure build-up on the downstream side of the disc may inhibit
the disc’s ability to provide overpressure protection. A rupture disc device
must have full pipe area and must not fragment into the pressure relief
valve inlet after the disc bursts.
When a rupture disc device is used with a pressure relief valve, con-
sult the ASME Code for the capacity reduction and installation details.
10.20 Performance Tolerance
A simple way of understanding and comparing rupture disc specifications

is to recognize the definition given by AS1358-1989 for Performance
Tolerance (section 1, definition 1.2.8, as follows):
‘Performance tolerance—a range of pressure in positive and nega-
tive quantities or percentages which include both manufacturing range and
bursting tolerance at a coincident temperature, which is applied directly to
the specified bursting pressure.’
A rupture disc is usually specified using an min–max range of pres-
sure at a specified temperature. When the min–max range is given by the
supplier, the other specific details necessary to specify are:
• The nominal burst pressure

• The manufacturing range
• The burst tolerance
In some cases, suppliers try to confuse the issue by stating the min–
max, but applying it to the manufacturing range only. This should be asked
of the supplier to ensure their complete understanding of the true min–
max as the true min–max incorporates not only the manufacturing range,
but also all burst tolerances (Figure 10.32).
The inclusion of manufacturing ranges and tolerances in the perfor-
mance tolerance means that the batch of discs being ordered today and
future batches will not burst outside this range at the at the customer’s
specified coincident temperature.
The testing carried out in the factory determines the actual burst pres-
sure of the batch. If the customer nominates ASME VIII certification, two
tests are made in an oven at the customer’s coincident burst pressure. The
average of these tests must lie in the manufacturing range and is stamped
on the disc tab in accordance with ASME VIII.
If ASME certification is needed, then the stamping on the disc tab
cannot vary. If stamping is needed to be min/max, users can specify the
same stringent testing as the ASME code by stating that the testing shall
include at least two tests at the elevated, coincident temperature that fall
in the manufacturing range, e.g., they can then specify in accordance with
ISO 6718.
Essentially, if a user asks for a tighter min/max, he or she is asking
for a tighter manufacturing range. Manufacturing ranges are specified in
the manufacturers’ literature. A zero-manufacturing range is the tightest,
meaning the average of the burst tests in the factory must equal the nomi-
nal burst pressure at the coincident temperature (Figure 10.33).
The burst tolerance is always ±5% in accordance with all the rupture
disc codes for stamped burst pressure equal to or greater than 276 kPa (g)
Min
+Z burst tolerance
Performance
Nominal burst pressure (High end

tolerance
of manufacturing range)
Manufacturing range
Low end of manufacturing range
−Z burst tolerance
Max
Figure 10.32. Performance tolerances.

Max
manufacture range
tolerance for zero
+Z burst tolerance
Performance
rupture disk
Nominal burst pressure
= Stamped burst pressure
−Z burst tolerance
Min
Figure 10.33. A zero-manufacturing range is the

tightest, meaning the average of the burst tests in the
factory must equal the nominal burst pressure at the
coincident temperature.
at 22ºC. The burst tolerance varies according to the particular disc design
for stamped burst pressures below 276 kPa (g) at 22ºC.
The burst tolerance refers to the accuracy of each disc in the batch
received.
Essentially, once it has been confirmed what manufacturing range
(ask the manufacturer to specify low end and high end) and what burst
tolerance apply to the high end and low end of the manufacturing range,
then the performance tolerance is worked out. If the supplier cannot pro-
vide this detail, then they should confirm that all manufacturing ranges
and tolerances are included in the min/max they have given.
10.21 Maximum Operating Pressure
Any rupture disc can be operated to any desired pressure. However, at

what pressure should users operate for reliable performance, so that the
life of the disc is not reduced?
From more than 65 years’ experience in the field, BS&B (a leading
manufacturer) has established the following operating pressure to burst
pressure ratios (expressed as %) (Table 10.1).
10.22 Cyclic/Pulsating Duties
The next question is: which discs can be operated reliably in cyclic/pul-
sating duties?
The answer is S90, JRS, RLS, MRB, and ECR. All others are ten-
sion-loaded discs that will have a limited cycle, life but are very good in
static conditions.
Table 10.1. Typical operating pressure to burst pressure ratios (courtesy

BS&B)
Operate from
full vacuum
to Y% of the
BS&B types Description stamped BP
S90, JRS, CSR, Single-section reverse-buckling 90%
RLS, MRB, ECR disc
GFN Tension-loaded disc that is 85%
scored after disc crowned
XN, LCN, DV Composite tension-loaded disc 80%
EXP/DV Tension-loaded, domed, 80% of
composite disc with vacuum nominal
support
AV, AVV Flat composite disc with 60% of
gaskets minimum
BV Tension-loaded, solid metal 70%
pre-bulged disc with vacuum
support
MBV Integral disc and holder of 80%
high-grade impregnated
graphite with vacuum support
Hence, the operating ratio can only be qualified depending on the type
of service that the disc is used in. This optimum value must, therefore, be
seen with caution.
On the specification sheet, the next step is to specify the maximum
positive operating pressure that the disc should see, at the coincident burst
temperature to ensure maximum service life.
To specify this, for the worst case condition, users must establish first
whether the low end of the manufacturing range will be above or below
276 kPa (g) (Figure 10.34).
10.23 Case A: 276 kPa (g) or higher
In this case, users may operate to Y% of the stamped burst pressure (at
worse case Y% of the minimum of the manufacturing range).
Min
+Z burst tolerance
Performance
tolerance Nominal burst pressure
Manufacturing range
Y% × Min of low end of

−Z burst tolerance manufacturing range is
> 276 kPa(g)
Max
Y% × Min of low end of Maximum operating

manufacturing range is pressure
< 276 kPa(g)
Maximum operating
pressure
Figure 10.34. Determining the upper and lower maximum operating
ranges according to whether the pressure lies above or below 276 kPa (g).
10.24 Case B: Lower than 276 kPa (g)
In this case, you may operate to Y% of the min, and hence, the burst toler-
ance must be deducted from the stamped burst pressure (at worse case, the
burst tolerance is deducted from the low end of the manufacturing range
and the min is calculated. Then Y% is applied to the min).
The performance of rupture discs at temperatures other than the coin-
cident disc temperature cannot be guaranteed unless the user is prepared
to pay for extra testing. Estimates can be given for various disc materials,
although these should not be taken for granted.
10.25 Standards
Commensurate with that of the different definitions described, there are

also a wide number of standards. Furthermore, national standards define
many varying types of safety valve.
The performance of different types of safety valve, set out by the var-
ious standards, is summarized in Table 10.2.
Table 10.2. Safety valve performance summary

Standard Fluid Overpressure Blowdown
Steam Standard 10% 10%
full lift 5%
A.D. Merkblatt A2
Standard 10%
Air or gas
10% full
Liquid
lift 5%
10% 20%
I Steam 3% 2–6%
Steam 10% 7%
ASME VIII Air or gas 10% 7%
Liquid 10% (see note
3)
Part 1 Steam Standard 10% 10%
full lift 5%
BS 6759
Part 2 Air or gas 10% 10%
Part 3 Liquid 10–25% 2.5–20%
APPENDIX A
J–T Valve
The J–T (Joule–Thomson) valve is frequently used in the p etrochemical

industry to liquefy gases and is essentially an expansion valve, often
referred to as a ‘choke’ or ‘choke valve.’
The Joule–Thomson effect describes how, when a gas is forced
through a restriction and then expands, the average distance between the
molecules increases. Because of intermolecular attractive forces, the gas,
thus, undergoes an increase in the potential energy, and assuming no heat
is transferred during this process, the total energy of the gas remains the
same. The increase in potential energy is matched by a corresponding
decrease in the kinetic energy, and therefore in temperature.
There is also a reverse process (limited to hydrogen, helium, and neon
gases) in which the temperature actually increases as the gas expands.
Called the ‘reverse Joule-Thompson effect,’ as the gas expands and the
intermolecular distances increase, the number of collisions will fall and
cause a decrease in the average potential energy. Again, since the total
energy of the gas remains the same, there will be an increase in the kinetic
energy, and therefore in temperature. It should also be noted that, at very
high pressures, in the order of 600 bar, many naturally occurring hydro-
carbon gases are subject to the ‘reverse Joule-Thompson effect’ and will
heat on expansion.
J–T Valve Construction
The term ‘J-T valve’ refers to the application, rather than the valve itself.
That being said, there are certain physical constraints of any valve used in
cryogenic applications.
226 • J–T Valve
Although use may be made of any throttling valve, angle needle

valves have proved popular in the oil industry, although straight cage/
globe valves are also frequently used.
Probably, the first requirement is to select the right materials of con-
struction. At cryogenic temperatures, valve body materials would exclude
carbon steel and include austenitic stainless steel, bronze, and Monel.
A second area of concern in cryogenic applications is packing. While
PTFE is restricted to around –40°C, graphite packing generally caters for
temperatures down to –200°C. An area of more general concern, however,
is that moisture from the atmosphere condenses on the colder surfaces and
forms a layer of frost, and even ice, on the bonnet and stem areas of the
valve. As the stem is stroked, frost may be drawn into the packing area
and cause damage. This problem may be overcome through the use of
an extended valve stem and bonnet that places the packing box area far
enough away from the cold area of the valve to prevent freeze-up of the
packing and minimize damage. The bonnet would generally be made of
austenitic stainless steel to minimize heat transfer.
Some form of insulating ‘cold box’ surrounding the valve and its pip-
ing, to minimize heat exchange with the environment, is also helpful in
combating the effects of both moisture build-up and noise.
Another basic problem associated with these J–T valves is as a result
of water or other contaminants freezing out of the gas. The formation of
both water ice and CO2 dry ice can plug the valve. This may be overcome
by dehydrating the gas well below the recommended dewpoint.
APPENDIX B
Basic Acoustics
The term ‘acoustics’ applies to a branch of physics concerned with the

properties of sound. And, ‘sound’ implies that it is something we hear―
that it is detected by the ear.
All sound originates from a vibrating body and is generated when
we displace the normal random motion of air molecules. If you pluck the
string of a musical instrument, the air molecules are displaced to produce
alternate rarefaction (expansion) and compression.
Pitch (Frequency)
One of the most important properties of sound is its pitch or tone. As we

have seen, sound is basically a vibration of molecules, and it is the rate of
vibration that determines the pitch (how high or low) of the resultant note.
The normal method of illustrating the alternate rarefaction and
compression of sound is in the form of a waveform, called a sine wave
(Figure B.1).
The number of vibrations or cycles or repeats of an event each second
is called the frequency and is represented by the unit hertz (Hz), named
after the German physicist Heinrich Hertz.
Thus, 100 cycles or vibrations per second would be designated 100
Hz. It is also normally customary to abbreviate 10,000 Hz to 10 kHz,
where the k signifies kilo or 1,000.
The range of vibration audible to human beings ranges from 20
to 20,000 Hz (20 kHz), encompassing all musical instruments plus the
human voice. The range of 20 to 20 kHz normally applies only to a young
healthy person of about 25 years. As people age, so their high frequency
response falls off rapidly until at the age of 60 years, few are able to detect
frequencies above 10 kHz.
228 • Basic Acoustics
Amplitude
1 cycle
Figure B.1. Graphic representation of a sound wave showing frequency and

amplitude.
Timbre
If a Middle C is played on a piano, on a violin, and on a trumpet, we

have learned that it will produce the same tone of 256 Hz in each case.
Nonetheless, the tone produced by the piano is easily distinguishable from
that played on the violin, and in turn from that played on the trumpet.
What distinguishes the three instruments is their timbre.
If a violin string is plucked in the middle of its length, it will pro-
duce a tone that is determined by the tension and the mass of the string.
By adjusting the tension, this frequency can be adjusted to, for example,
Middle C (256 Hz). This is called its fundamental frequency.
If the string is now lightly touched in the middle and plucked at a
distance that is a quarter of its length, it will vibrate in two segments and
give rise to a frequency that is twice that of the original or fundamental
frequency.
And, by lightly touching a point one-third of its length and plucking
half-way between that point and the end, the string will vibrate in three
segments and produce a frequency that is three times higher than the orig-
inal or fundamental.
These higher frequencies are all a characteristic of the same length of
string and are called harmonics or overtones of the fundamental.
When these harmonics are added to the fundamental, they give rise to
a waveform that, although containing the tone of the note, also makes an
individual sound.
Basic Acoustics • 229
2nd harmonic Fundamental
(a)
(b)
3rd harmonic Fundamental
Figure B.2. Addition of harmonics to the fundamental: (a) second harmonic;

(b) third harmonic.
Figure B.2(a), for example, illustrates a second harmonic added to

the fundamental, while Figure B.2(b), illustrates a third harmonic added
to the fundamental.
Velocity
The third property of sound that we need to look at is its velocity through
the medium in which it is traveling.
The velocity of sound in gases is given by the equation:
g ⋅ R ⋅ ( 273 + T )
C =
M
where:
C = velocity of ultrasonic wave (m/s)
R = universal gas constant 8,314.3 (J/K.mol)
T = temperature (°C)
M = molecular weight (kg/K.mol)
γ = adiabatic component
In air, the most common medium used for the propagation of sound,
the velocity, may be calculated by:
20.048 ( 273.15 + T)
Therefore, at 20°C:
C = 20.048 293.15 ≈ 343 m / s
This means that the sound heard at a distance of 343 m from its source
reaches the ear after 1 second. Put another way, the sound from a lightning
flash takes about 3 seconds to travel each kilometre distance.
In most media, other than air, sound travels faster. In water, for
example, sound travels four times faster than in air, and in steel, it travels
some 14 times faster.
Wavelength
Assume a sound wave traveling in air at a frequency of 1 kHz. Thus, in

one second, there will be 1,000 peaks during which the sound will have
traveled 343 metres. So, on one full cycle (peak to peak), the wave will
have travelled only 1/1,000th the distance i.e., 0.343 metre.
This distance is called the wavelength (denoted by the symbol λ) and
is obtained by dividing the velocity by the frequency. Thus:
V
l=
f
The wavelengths of several frequencies traveling in air are shown in

Table B.1.
Table B.1. The wavelengths of several frequencies travelling in air

Frequency Wavelength
500 Hz 686 mm
1 kHz 343 mm
2 kHz 171.5 mm
4 kHz 85.8 mm
8 kHz 42.88 mm
16 kHz 21.44 mm
Intensity
If a violin string is plucked with varying degrees of force, the pitch of the
note remains the same, but its intensity (or loudness) increases. Referring
to the waveform of Figure B.1, the intensity can be represented by the
height or amplitude. The greater force used to pluck the string, the louder
the sound and the higher the amplitude.
Logarithmic Characteristic of Ear
A remarkable ability of the human ear is its ability to cope with a huge
range of sound intensities Ñ ranging from the fall of a pin through to the
scream of a modern jet engine. This ability of the ear to cope with this
range is due to its non-linear characteristic.
Figure B.3 shows two identical loudspeakers, each connected across
an amplifier. The amplifiers are driven by a 1 kHz tone generator that is
switched so that it is connected, alternately, to each.
If the same power is supplied to each speaker, e.g., 100 mW, both
their notes will be of equal intensity, and a listener will perceive them
to be the same. Initially, as the power to one of the speakers is gradually
increased, the listener, listening to each of them one at a time, will per-
ceive no difference in intensity. Only when one loudspeaker receives 26%
more power will it in fact sound louder. At this point, 126 mW is being fed
to one loudspeaker and 100 mW to the other.
Figure B.3 Two identical loudspeakers connected across an amplifier.

If the balance is now restored, by bringing both loudspeakers up to

126 mW, the intensities will again be equal. Once more, if the power to
the one loudspeaker is increased, no perceptible difference will be noticed
until it receives 26% more power, e.g., 26% of 126 mW = 32 mW, thus
bringing the higher loudspeaker output to:
l26 + 32 = 158 mW.
Once again, if the balance is restored by bringing the other loud-

speaker up to 158 mW, the intensities will again be equal. Now, again, the
intensity of one speaker will need to be 26% higher, i.e., 26% of 159 mW
= 41 mW, before the listener perceives a difference. This brings the higher
loudspeaker output to:
159 + 41 = 200 mW.
If this procedure is repeated in 10 stages, it will be found that the

intensity has also increased by 10 times to 1,000 mW. The perceived
increase in the sound intensity is, thus, obtained by raising the power level
in a given ratio—not by adding specific amounts. The important point to
note is that this 10-fold increase has not been reached by 10 equal linear
steps, but by 10 non-linear steps, the first being quite small and the last
quite large (Figure B.4). This response is called logarithmic.
1000
900
800
700
600
mW
500
400
300
200
100
1 2 3 4 5 6 7 8 9 10
Step number
Figure B.4. Logarithmic response of the human ear.

10 dB 1000 mW
9 dB 794 mW
8 dB 630 mW
mW
7 dB 501 mW
6 dB 398 mW
5 dB 316 mW
4 dB 251 mW
3 dB 200 mW
158 mW
126 mW
0 dB 100 mW
Figure B.5. The 10-fold power ratio
increase is designated a Bel with each
power increment of 26% being one-tenth
of a Bel—called a decibel (dB).
As illustrated in Figure B.5, the 10-fold power ratio increase is desig-

nated a Bel with each power increment of 26% being one-tenth of a Bel,
called a decibel (dB). It must be appreciated that the dB is only a ratio,
and that the ear hears the same small difference between 1 W and 2 W, as
between 100 W and 200 W.
Definition of Decibel
The mathematical definition of the decibel is:
P2
10 log10
P1
where:
P2 = output power
P1 = input power
In practice, fortunately, it is not necessary to make use of this equa-

tion, as tables are given for any ratio that is required (Table B.2).
Table B.2. Gain and attenuation ratios expressed in dBs

Gain Attenuation
Voltage, current, Voltage, current,
Power and pressure dB Power and pressure
ratios ratios +- ratios ratios
(A) (B) ‹ﬁ (C) (D)
1 1 0 1 1
1.26 1.122 1 0.79 0.89
1.585 1.26 2 0.63 0.79
2 1.413 3 0.5 0.7
2.5 1.585 4 0.4 0.63
3.162 1.778 5 0.32 0.56
3.98 2 6 0.25 0.5
5 2.24 7 0.2 0.45
6.31 2.5 8 0.16 0.4
7.943 2.82 9 0.13 0.355
10 3.162 10 0.1 0.32
12.6 3.55 11 0.08 0.28
15.9 3.98 12 0.063 0.25
20 4.47 13 0.05 0.224
25.1 5 14 0.04 0.2
31.6 5.62 15 0.032 0.18
39.8 6.31 16 0.025 0.16
50.1 7.08 17 0.02 0.14
63.1 7 943 18 0.016 0.126
79.4 8 91 19 0.013 0.112
100 10 20 0.01 0.1
1 000 31.6 30 0.001 0.0316
104 100 40 10 −4
10−2
105 316 50 10 −5
3.16 × 10−3
106 103 60 10−6 10−3
107 3.16 × 103 70 10 −7
3.16 × 10−4
108 104 80 10−8 10−4
109 3.16 × 104 90 10 −9
3.16 × 10−5
1010 105 100 10−10 10−5
1011 3.16 × 105 110 10 −11
3.16 × 10−6
1012 106 120 10−12 10−6
1013 3.16 × 106 130 10 −13
3.16 × 10−7
1014 107 140 10−14 10−7
Voltage Ratios
So far, we have discussed power ratios, and thus an amplifier having a

3 dB gain will double the power. The question arises, what would happen
to the voltage in the same amplifier?
In this circuit (Figure B.6), the power developed across the resistor
is given by:
V 2 l00
P= = = 10 W
R 10
How much would we need to increase the voltage if we wanted to

double the power to 20 W (i.e., increase of 3 dB)?
Since:
V2
P =
R
V2 = P ⋅ R
V= P⋅R = 20 ⋅10 = 14.4V
As illustrated in Figure B.7, doubling the power is achieved by only

a 1.414 increase in voltage.
Using a similar reasoning, it can be shown that:
3 dB = 2 × power but only 1.414 × voltage
10 V 10 Ω 10 W
Figure B.6. Simple circuit showing

power developed by a resistor.
14.4 V
20 W
10 V 10 Ω
10 W
Figure B.7. Doubling the power is achieved

by only a 1.414 increase in voltage.
6 dB = 4 × power but only 2 × voltage

10 dB = 10 × power but only 3.16 × voltage
20 dB = 100 × power but only 10 × voltage
Absolute Levels
As stressed earlier, the decibel is only a ratio. Nonetheless, it can be

used to express absolute values if a reference is given. If the reference
chosen was 1 W, then 3 dB would be 2 W and 6 dB would be 3.98 W
and so on.
In fact, one of the common references used is the milliwatt = 10−3 W
and is designated dBm.
Thus:
20 dBm = 100 ¥ 1 mW = 100 mW.
Another reference that is used is in the measurement of sound
pressure levels.
Sound Pressure Level (SPL)
Previously we have seen that sound is a movement or vibration of

molecules giving rise to increases or decreases of pressure. How do we
measure air pressure?
Pressure is defined as the force per unit area, and for many, the defi-
nition of pounds per square inch (lbs/in2 or psi), still prevalent in the
United Kingdom and United States, seems very much more descriptive of
pressure rather than, for example, the Pascal. However, from physics, we
learned that another useful definition is the bar:
1 atmosphere = 14.7 lbs/in2 = 29.92 inches of mercury (Hg)
= 760 mm Hg
= 1 bar.
Now clearly, 1 bar is far too large a unit in which to express changes
in sound pressure, and therefore the microbar (µbar) is often used where
1 µbar is one millionth of a bar:
1 µbar = 1/1,000,000 bar
= 10−6 bar.
With the introduction of the metric system, the microbar was replaced
by the Pascal where the Pascal (Pa) is one-hundred-thousandth of a bar:
1 Pa = 1/100,000 bar
= 10−5 bar.
From this, it follows that:
1 Pa = 10 µbar
Standards
The absolute standard that has been chosen for measuring the sound pres-
sure level (SPL) is called the threshold of hearing and is the average sound
pressure that is barely perceptible to a young human, with undamaged
hearing, at 1 kHz. This level is generally described as 0.0002 µbar or 0 dB
(SPL).
In reality (as shown in Figure B.8), the actual threshold is about 4
dB (0.003 µbars) higher at 1 kHz and is very frequency-dependent such
that the lower the frequency the more energy required to make it audible.
At 50 Hz, for example, the average person cannot hear SPLs below 40 dB
(0.02 mbars)
As the SPL is increased, we reach a point just short of being painful
to the ear, called the threshold of pain which, using the 1 kHz datum,
corresponds to 200 µbars.
Since the absolute reference of 0 dB is 0.0002 µbar, the value for the
threshold of pain is 200/0.0002 = 106 = 1 million times greater. This may
be expressed in dBs by referring to Table B.3, column B where:
106 = 120 dB (SPL)
130
120
110 Threshold of pain
100
Sound pressure level (SPL dB)
90
80
70
60
50
40
30
20 Threshold of hearing
10
0
–10
10 Hz 100 Hz 1 kHz 10 kHz 100 kHz
Frequency (Hz)
Figure B.8. Threshold of hearing (lower) and of pain (upper).
Table B.3. Sound intensities of various sources

Source Intensity level dB (SPL)
Jet plane at 30 m 140
Threshold of pain 120
Indoor rock concert 120
Siren at 30 m 100
Male voice shouting 87
Motor car interior moving at 90 k/h 75
Busy street traffic 70
Average male voice at 1 metre 67
Average car 60
Quiet radio 40
Ticking clock 30
Whisper 20
Rustle of leaves 10
Threshold of hearing 0
Other important levels are:

1 mbar = 74 dB (SPL)
1 Pa = 10 mbar = 94 dB (SPL)
As discussed, the energy level at the threshold of hearing varies with

frequency. Thus, it would require 54 dB (SPL) at 30 Hz to produce the
same ‘loudness’ (as perceived by the listener) as 0 dB (SPL) at 1 kHz. The
contour for the threshold of hearing represents only the bottom limit of a
series of ‘equal loudness contours’ called phons (Figure B.9).
Two factors are very much evident from the equal loudness contours
of Figure B.9. Firstly, very much more energy is needed to produce a
bass note of a given loudness, when compared with a 2–3 kHz note—an
important consideration in any music system.
The second point is that, if a broad spectrum noise level of, say, 20 dB
(SPL) is introduced, the listener will have a certain sound ‘image’ accord-
ing to the equal loudness contour corresponding to 20 dB. If the noise
level is now raised, another sound image is received, of the same noise,
due to the mechanism of the changing curves of equal loudness. In other
words, all frequencies are present in the signal, but depending on the level,
will be heard in different relationships.
In order to reproduce this subjective perception characteristic of the
ear, a sound level meter should incorporate automatic switching filters
that represent the phon curves at every level. Because this would increase
the cost of such instruments to an unaffordable level, standard weighting
curves have been determined (Figure B.10):
130
120 Threshold of pain
110
100
Sound pressure level (SPL dB)
90
80
70
60
50
40
30
20
10 Threshold of hearing
0
–10
10 Hz 100 Hz 1 kHz 10 kHz 100 kHz
Frequency (Hz)
Figure B.9. Equal loudness contours.

20
10
–10
Gain (dB)
A-weighting
–20 B-weighting
C-weighting
–30
–40
–50
10 Hz 100 Hz 1 kHz 10 kHz 100 kHz
Frequency (Hz)
Figure B.10. A- B-, and C-weighted responses required for measuring

sound pressure levels.
A-curve
Designated dBA, this weighting was originally intended only for the mea-
surement of low-level sounds (around 40 phon).
B-curve
This curve is applied for levels between 40 dB and 70 dB.
C-curve
Used at levels higher than 70 dB.
Note that all three curves cross at 1 kHz.
A-weighting is now mandated for the measurement of environmental
noise and industrial noise. In reality, it is badly suited for these purposes,
since it tends to moderate the effects of low frequency noise.
Appendix C
Block and Bleed
The primary function of a double block-and-bleed system is to isolate or

block the flow of the upstream process medium from reaching the down-
stream equipment while carrying out maintenance, repair, or component
replacement.
As illustrated in Figure C.1, it typically comprises two block valves
(Valves 1 and 2) and a bleed valve (Valve 3) that vents to a relief or safe
disposal location. When delivering the process medium to the downstream
equipment, the valves are set with the isolation valves 1 and 2 open and
the bleed valve 3 closed. When isolating the downstream equipment from
the process fluid, the valves are set with isolation valves 1 and 2 closed
and bleed valve 3 open.
By monitoring the outlet of the bleed, users may determine whether
the downstream system is, in fact, properly isolated and whether there is
any leakage past either of the two block valves.
Usually, a double block-and-bleed would comprise two separate
block valves and a separate bleed valve assembled on a tee. Block valves
can take the form of virtually any high-integrity shut-off valve (e.g., ball
valve, expanding gate valve, or plug valve), while the bleed is usually a
ball valve, or in many cases, a cap or plug.
Although such multi-block valve systems work effectively, they can
be expensive to install and maintain, especially when dealing with large-
sized valves or working with automated systems.
Consideration must be given not only to the capital cost of the two
block valves; bleed valve; actuation, four or three valves; control sys-
tem, type T; flange bolting; and the flange gaskets, but also the design
and installation costs. Consequently, several manufacturers have designed
single manifold systems, such as that shown in Figure C.2, which illus-
trates the typical construction of a single double block-and-bleed manifold
242 • Block and Bleed
Process
OPEN OPEN
medium in
Valve 1 Valve 2
Process
(a) Valve 3 CLOSED
vessel
Process CLOSED CLOSED

medium in
Valve 1 Valve 2
(b) Process
Valve 3 OPEN
vessel
To relief system
or safe disposal
Figure C.1. (a) Under normal operation, the valves are set with the isolation
valves 1 and 2 open and the bleed valve 3 closed. (b) When isolating the
downstream equipment, the valves are set with isolation valves 1 and 2 closed
and bleed valve 3 open.
Figure C.2. Typical construction of a single double block-

and-bleed valve (courtesy Habonim).
Block and Bleed • 243
Valve
body
Process
fluid
Figure C.3. The bleed often takes the form of a cap or plug.
system comprising two large balls acting as blocks (both shown closed)
and a small ball serving as the bleed (ball is shown in the open position).
In addition, specially designed trunnion-mounted ball valves,
equipped with a valve body bleed between the seats, provide a satisfac-
tory substitute for separate individual double block-and-bleed valves (see
Chapter 4. ‘Valve Construction,’ Section 4.9).
The bleed often takes the form of a cap or plug as illustrated in
Figure C.3.
APPENDIX D
Water Hammer
Because liquid is essentially incompressible, any energy applied to it

is transmitted instantly. If a moving column of liquid is slowed down
suddenly by, for example, a quick-closing valve, the sudden change in
liquid velocity in the delivery line creates a pressure wave (Figures D.1(a)
and (b)).
Despite the frequent assumption that liquids are incompressible, in
reality, most substances diminish in volume when exposed to a uniform
externally applied pressure. The Bulk Modulus describes the compress-
ibility of a fluid as the ratio of the very small decrease in volume resulting
from an applied external pressure. A large bulk modulus indicates a rela-
tive incompressible fluid.
Assuming that the walls of the pipe are sufficiently thick for it to be
approximated as rigid, the velocity of the pressure wave in a rigid pipe is
given by:
K
C= (D.1)
ρ
where:
C = velocity of pressure wave (m/s)
K = bulk modulus (GPa)
ρ = density (kg/m3)
Thus, for example, given that water has a density (ρ) of 1,000 kg/m³
and a bulk modulus of 2.2 GPa, the velocity is:
2.2 ⋅109
C= = 1483m/s (D.2)
1000
Some typical sonic velocities in various liquids are shown in
Table D.1.
246 • Water Hammer
Large diameter riser

Open valve
(a)
Branch
Flow
Normal flow
Valve closed
(b)
Shock
Quick closure
Figure D.1. If a moving column of liquid (a) is slowed down suddenly by,
for example, a quick-closing valve, the sudden change in liquid velocity in
the delivery line creates a pressure wave (b).
Table D.1. Some typical velocities in various liquids

Sound velocity at 25°C
Liquid (m/s) (ft/s)
Kerosene (paraffin) 1,324 4,344
Gasoline (petrol) 1,250 4,101
Water 1,483 4,862
Crude oil: Light 1,347 4,420
Crude oil: Medium 1,401 4,598
Crude oil: Heavy 1,441 4,729
Crude oil: Extra heavy 1,480 4,856
In practice, the pressure rise may be sufficient to deform the pipe,

increasing its cross-section. And, since the pipe thus absorbs strain energy,
the velocity of the pressure wave is reduced:
Ke
C= (D.3)
ρ
where:
Ke = the effective bulk modulus, given by:
1 1 D
= + (D.4)
Ke K E T
Water Hammer • 247
where:
K = bulk modulus of the liquid (GPa)
D = internal diameter of the pipe (mm)
E = bulk modulus of the pipe (GPa)
T = thickness of the pipe (mm)
Again, assuming water (density (ρ) 1,000 kg/m³ and a bulk modulus
of 2.2 GPa) flowing in a pipe having a bulk modulus (E) of 210 GPa, an
internal diameter of 200 mm, a wall thickness of 5 mm:
From equation (D.4):
1 1 200 1
= + = (D.5)
K e 2.2 ⋅109 210 ⋅109 ⋅ 5 6.45 ⋅1010
Therefore, the effective bulk modulus (Ke) is 1.55*109, and

substituting:
1.55 ⋅109
C= = 1244m/s (D.6)
1000
This corresponds to a substantial 16% reduction in the sonic velocity.

In practice, the pressure wave travels back up the line at between
1,000 and 1,300 m/s, to the end of the pipe where it will reverse direction
and travel back toward the valve (Figures D.2 (a) and (b)).
(a)
Pressure wave enlarges pipe
(b)
Reflected pressure wave
Figure D.2. The pressure wave travels back up the line (a) at between
1,000 and 1,300 m/s, to the end of the pipe where it will reverse direction
and travel back toward the valve (b).
Pressure wave reaches valve
Figure D.3. Depending on the valve size and system conditions, a valve clos-
ing in 1.5 s or less can produce a pressure spike five times the system working
pressure.
Depending on the valve size and system conditions, a valve closing

in 1.5 s or less can produce a pressure spike five times the system working
pressure (Figure D.3), leading to blown diaphragms, seals, and gaskets
and also catastrophic system component failure in transmitters, meters,
and gauges.
As intimated, the magnitude of the pressure spike in a given system is
very much determined on the speed at which the valve is closed. Although
there are many calculations available, a general rule of thumb is shown
as follows:
0.052 v L
P= + PI (D.7)
t
where:
P = increase in pressure (bar)
v = flow velocity (m/s)
t = valve closing time (s)
L = upstream pipe length (m)
PI = inlet pressure (bar)
In FPS terms, the equation becomes:
0.07 v L
P= + PI (D.8)
t
where:
P = increase in pressure (psi)
v = flow velocity (ft/s)
t = valve closing time (s)
L = upstream pipe length (ft)
PI = inlet pressure (psi)
To give you some idea of the magnitude of the spike, assume a sole-
noid valve having a closure time of approximately 40 to 50 ms, connected
to a 15-m-long upstream pipe. The water flow is 3 m/s and the inlet deliv-
ery pressure is 4 bar. What is the amplitude of the pressure spike?
From equation (D.7):
0.052 3 15
P= + 4 = 62.5 bar (D.9)
0.04
Water Hammer as a Result of Steam

Condensate
Water hammer also occurs as a result of condensate in steam pipes. Figures

D.4 (a) to (d) show accumulated condensate in a portion of horizontal
steam piping. As the steam flows over the condensate, it causes the surface
of the water to ripple and trap some of the condensates in the pipe.
At the same time, as a result of the Bernoulli effect (Figure D.4 (a)),
a wave is drawn up that effectively seals the pipe, producing an isolated
pocket of steam (Figure D.4 (b)).
The collapsing steam void (Figure D.4 (c)) creates an implosion
(Figure D.4 (d)) that produces a slug of condensate that is carried along
by the steam flow and that can travel at the speed of the steam (up to
160 km/hr).
The effect of this force, striking the first elbow in its path, is compara-
ble to a hammer blow and the damage sustained can be quite substantial.
Condensing steam Heat loss

Steam
(a)
Sub-cooled condensate Bernoulli effect

draws up wave
Isolated steam pocket Heat loss
(b) Steam
Sub-cooled condensate Wave seals pipe
Collapsing steam void
(c) 5 bar steam
Implosion
Rebounding wave
(d)
Figure D.4. Hydraulic shock wave produced as a result of accumulated

condensate in steam piping.
A second type of water hammer that occurs in steam piping is actually

cavitation. This is caused by a steam bubble forming or being pushed into
a pipe completely filled with water. As the trapped steam bubble loses its
latent heat, the bubble implodes, the wall of water comes back together,
and the force created can be severe. This condition can crush float balls
and destroy thermostatic elements in steam traps. This type of cavitation
usually occurs in wet return lines or pump discharge piping.
Pulsations
Of course, such shocks are not just produced by the closure of a valve. Other
causes include: starting or stopping a pump, closure of an ESD device, and
shut-off of a check valve. Pulsations are also often introduced through
the use of ‘Oval’ gear positive displacement flow meters or reciprocating
or peristaltic positive displacement pumps. The resultant acceleration and
deceleration of the pumped fluid produces pressure spikes of greater than
10 times the steady state flow pressure.
Prevention and Mitigation
The most obvious solution is, of course, prevention. Do not ever close a
valve, do not ever trip or start a pump, do not have an emergency discon-
nect of a hose, etc. Clearly, wishful thinking! However, it is possible, in
many cases, to close the valve under controlled conditions—increasing
the valve closure time.
In the example given previously, what would be the result of increas-
ing the closure time to 1.5 seconds?
0.052 3 15
P= + 4 = 5.56 bar (D.10)
1.5
There are of course many ‘rule of thumb.’ One frequently used approx-
imation is that the valve should not close faster than the acoustic round trip,
which gives a ballpark figure that closure should be no faster than 30 s.
Indeed, in U.S. waterways under the jurisdiction of the U.S. Coast
Guard, a discharge valve is not allowed to close faster than 30 s when load-
ing a tanker (Code of Federal Regulations 33CFR154—a USCG regulation).
If we look at a typical example of a discharge from an SPM (single-
point mooring) to the crude oil terminal through a 16 inch 5 km pipeline
and assuming a sonic velocity of 3,300 ft/s, the acoustic round trip is just
under 10 s—well below the accepted norm of 30 s.
However, what sort of overpressure surge are we still likely to expect?

Assume a flow velocity of 2 m/s (fairly slow) and a pipeline pressure
of 6 bar, then:
0.052 2 5000
P= + 6 = 23.3 bar (D.11)
30
With typical maximum allowable operating pressures (MAOPs) of
the order of only 18 to 20 bar, this is certainly exceeding operational lim-
its, even allowing for the fact that many codes allow for a 10% exceedance
of the MAOP in the event of hydraulic surge.
In very large pipelines, the use of motorised operated gate valves may
have closure times of up to 4 or 5 min. Even so, on a 300-km-long pipe-
line, the overpressure surge is likely to exceed the MAOP.
Since total prevention might well prove impossible to achieve, the
answer must lie with mitigation through surge relief systems.
In essence, there are three forms of relief devices available: pulsation
dampeners, rupture discs, and surge relief valves.
Pulsation Dampeners
A pulsation dampener or surge suppressor is a hydro-pneumatic dampener

comprising a pressure vessel containing a compressed gas, generally air
or nitrogen, separated from the process liquid by a bladder or diaphragm.
These devices are mainly used to minimize the pulsations resulting from
a pump’s stroking action. During the discharge stroke, fluid pressure dis-
places the bladder and compresses the trapped gas (Figure D.5). During
Air/gas
Bladder
Liquid
Figure D.5. A pulsation dampener or surge suppressor.

During a surge, the fluid pressure displaces the bladder
and compresses the trapped gas.
the following cycle, the momentary interruption of fluid flow causes the
compressed gas to expand, forcing the bladder or bellows to push the
accumulated fluid back into the discharge line.
An advantage of this type of system is that, it is ready for immediate
reuse after a pressure surge has occurred. On the negative side, a single
device can only relieve a small amount of fluid, and thus, on larger pipe-
line systems, a large bank of accumulators may be required.
Nitrogen-Loaded Surge Relief Valves
A typical gas-loaded axial flow style valve, from Daniel, is shown in

Figure D.6, in which nitrogen gas is used to pressurize the valve piston
to keep it in the closed position. The gas pressure less the 4 psi force
of the valve spring is the effective set-point of the valve. The oil acts as
a movable barrier between the gas and valve piston that eliminates the
possibility of gas bypassing the piston.
As the pipeline pressure increases, the combined force of the spring
and nitrogen gas pressure is overcome and the valve opens (Figure D.7),
with a response time typically under 100 ms.
Gas
pressure
Light oil
Check
valve
Spring
Piston
Flow
Figure D.6. A Daniel gas-loaded axial flow

style valve in which nitrogen gas is used to
pressurize the valve piston to keep it in the
closed position (courtesy Emerson).
Gas pressure
expelled
Sight
gauges
Light oil
Check
valve
Spring
Piston
Flow
Figure D.7. As the pipeline pressure increases,

the combined force of the spring and nitrogen
gas pressure is overcome and the valve opens
(courtesy Emerson).
APPENDIX E
Stainless Steel
Stainless steel is a family of corrosion-resistant steels containing chro-

mium in which chromium forms a passive film of chromium oxide (Cr2O3)
when exposed to oxygen. This phenomenon is called passivation and is
seen in other metals, such as aluminum and titanium.
The film layer is impervious to water and air and quickly reforms
when the surface is scratched. This protects the metal beneath, preventing
further surface corrosion. Since the layer only forms in the presence of
oxygen, corrosion resistance can be adversely affected if the component is
used in a non-oxygenated environment, e.g., underwater keel bolts buried
in timber.
Such passivation only occurs if the proportion of chromium is high
enough and is normally achieved with addition of at least 13% (by weight)
chromium. Progressively higher levels of corrosion resistance and strength
is achieved by the addition of other alloying elements, each offering spe-
cific attributes in respect of strength and corrosion resistance.
Classification Issues
The need to classify stainless steel has led to a fundamental problem of

which method to use. Probably, the best known system derives from of
the Society of Automobile Engineers (SAE), e.g., 316 Cr/Ni/Mo 17/12/2.
This is interpreted as stainless steel containing the proportions of 17%
chromium, 12% nickel, and 2% molybdenum.
However, the waters are somewhat muddied by a variety of interna-
tional and country-based systems that include EN (European Norm) and
UNS (Unified Numbering System). For example, SAE 304 Cr/Ni 18/10
stainless steel is EN 1.4301, which is UNS S30400.
256 • Stainless Steel
Stainless steels may also be graded into five basic families or phases
determined by their crystalline structure: the stable phases austenitic or
ferritic, a duplex mix of the two, the martensitic phase created when some
steels are quenched from a high temperature, and precipitation-hardenable.
Ferritic Stainless Steel
In ferritic stainless steel, the iron and chromium atoms are arranged in
what is termed a body-centered cubic structure in which the atoms are
arranged on the corners of the cube and one in the center (Figure E.1).
As well as being ferro-magnetic, ferritic stainless steel exhibits very high
stress corrosion-cracking resistance.
Ferritic stainless steels are plain chromium (10.5 to 18%) grades
characterized by moderate corrosion resistance and poor fabrication
properties. These characteristics may be improved with the addition of
molybdenum; some, aluminum or titanium.
The basic 430 grade is a simple corrosion and heat-resisting grade.
Alloying elements that tend to make the structure ferritic are called
ferrite formers and result in grades such as Grades 434 and 444 and in
the proprietary grade 3CR12. Common ferritic grades include: 18Cr-2Mo,
26Cr-1Mo, 29Cr-4Mo, and 29Cr-4Mo-2Ni (Figure E.2).
Austenitic Stainless Steel
With the addition of nickel, the properties change dramatically. As shown

(Figure E.3), the atoms are re-arranged so that they occur on the corners
of the cube and also in the center of each of the faces. In this manner, it
becomes what is termed austenitic stainless steel.
Ferrite
Body centred cubic crystal

Figure E.1. In ferritic stainless
steel, the atoms are arranged in a
body-centered cubic structure.
Stainless Steel • 257
444
Cr/Mo 17.7/2.1
Add more Mo for

further improved
corrosion resistance
434
Cr/Mo 17/1
Add Mo for improved

corrosion resistance
430 Added niobium for

Basic grade 436
increased corrosion
Cr 16.5 Cr/Mo/Nb 17.5/1.2/0.6
and heat resistance
Lower Cr plus Al prevents

hardening when cooled
from high temperatures.
405
Cr/Al 12/0.2
Lowest Cr
409
Cr/Ti 11/0.5
Figure E.2. Ferritic stainless steel family.
Ferrite Austenite
Add Nickel
Body centred cubic structure Face centred cubic structure
Figure E.3. With the addition of nickel, the atoms in austenitic

stainless steels are arranged on the corners of the cube and also
in the center of each of the faces.
It can, thus, be seen from Table E.1 that, unless you are specifically
looking for a ferro-magnetic material, austenitic stainless steel would be
the most obvious choice. Indeed, this is borne out by the fact that austenitic
stainless steels account for about 70% or more of all stainless steel used
worldwide, with ferritic stainless steels making up about 25%. The other
families each represent less than 1% of the total market.
Austenitic stainless steels are designated by numbers in the 200 and
300 series.
Table E.1. Difference in the properties of ferritic and austenitic stainless

steels.
Properties Ferritic Austenitic
Toughness Moderate Very high
Ductility Moderate Very high
Weldability Limited Good
Thermal expansion Moderate High
Stress corrosion cracking Very high Low
resistance
Magnetic properties Ferro Non-magnetic
magnetic
Series 300
The relationship between the 300 austenitic grades is shown in Figure E.4
The basic grade 304 contains about 18% chromium and 8% nickel
(often referred to as 18/8) and range through to the high alloy or ‘super
austenitics’ such as 904L and 6% molybdenum grades.
Ni-Cr-Fe
303, 303Se
Alloys
Add Ni for corrosion

resistance in high
temperature applications Add S or Se for
machinability 347
309, 310, 314, 330
Add Nb + Ta to
Add Cr and Ni for strength reduce sensitization
and oxidation resistance
304 Add Ti to
18Cr-8Ni 321
reduce sensitization
Add Mo for pitting

resistance 304L
Lower C to
316 reduce 316L
sensitization
Add more Mo for more
317L
pitting resistance
317 Add Ni, Mo, N for Super austenitic

corrosion resistance stainless steel
Figure E.4. The relationship between the various 300 series

austenitic grades.
Additional elements can be added, such as molybdenum, titanium, or

copper, to modify or improve their properties, making them suitable for
many critical applications involving high temperature, as well as corro-
sion resistance. This group of steels is also suitable for cryogenic applica-
tions because the effect of the nickel content in making the steel austenitic
avoids the problems of brittleness at low temperatures, which is a charac-
teristic of other types of steel.
Generally, the 300 grade alloys are subject to crevice and pitting cor-
rosion. The time it takes for this type of corrosion to occur is called the
‘incubation time.’ In seawater, the incubation time for machinable grades,
such as Type 303, is practically zero while that for the best alloys, such as
Type 316, the time ranges from six months to a year.
Low-carbon versions, (indicated by the letter suffix L) include 304L,
316L, and 317L, in which the carbon content of the alloy is below 0.03%.
This reduces the effect of ‘sensitization’ in which chromium carbides pre-
cipitate at the grain boundaries due to the high temperatures involved in
welding. The relatively high nickel content also inhibits the brittleness
exhibited by ferritic materials at low temperatures, and thus makes auste-
nitic steels suitable for cryogenic applications.
200 Series
We have seen earlier how the addition of nickel is used in the creation of
the classic chrome–nickel 300 series austenitic stainless steel.
The reduced nickel content of the 200 series chrome–manganese
grades makes them significantly cheaper. However, depending on their
chemistry, they also offer good formability (ductility) and/or strength.
Indeed, certain grades (201, 202, and 205 series) even offer about 30%
higher yield strength than the classic 304-series chrome–nickel grade,
allowing designers to cut weight (Table E.2).
Reducing nickel, on the other hand, reduces the maximum chromium
content possible in the alloy. Less chromium means less corrosion resis-
tance and a consequent narrowing of the range of applications for which
the material is suitable.
A word of warning comes from the International Stainless Steel
Forum (ISSF). Continuous pressure to cut costs, especially from the
Asian market, has resulted in the development of austenitic grades
ever lower in nickel and chromium, often not covered by international
codes or specifications. In fact, numerous chrome–manganese grades
are company-specific and identified simply by a title given to them by
the producer.
Table E.2. Chemical composition of standard grades (courtesy

International Stainless Steel Forum)
Chemical composition (wt. %)
Grades C Mn Cr Ni N Cu
201 0.15 max 5.50–7.50 16.0–18.0 3.50–5.50 0.25 max -
202 0.15 max 7.50–10.0 17.0–19.0 4.00–6.00 0.25 max -
204 0.15 max 6.50–9.0 15.5–17.5 1.5–3.5 0.05–0.25 2.0–4.0
205 0.12–0.25 14.0–15.5 16.5–18.0 1.0–1.75 0.32–0.40 -
Duplex Stainless Steels
Duplex stainless steels are a mixture of austenite and ferrite microstruc-

tures that combine some of the features of each class:
• resistance to stress corrosion cracking, but inferior to ferritic steel;

• superior toughness to ferritic steel, but inferior to austenitic steel;
• roughly twice the strength of austenitic steel;
• superior resistance to pitting, crevice corrosion, and stress corro-
sion cracking;
• high resistance to chloride ions attack; and
• high weldability.
These features are achieved by adding less nickel than would be nec-
essary for making a fully austenitic stainless steel. Thus, Grade 2304 com-
prises 23% chromium and 4% nickel, while Grade 2205 comprises 22%
chromium and 5% nickel—with both grades containing further minor
alloying additions.
On the negative side, austenitic–ferritic duplex stainless steels are
only usable between temperature limits of about –50°C and 300°C, out-
side which they suffer reduced toughness.
Martensitic Stainless Steel
Named after the German metallurgist, Adolf Martens, the martensi-

tic Grade 400 series (Figure E.5) are low-carbon (0.1–1%), low-nickel
(less than 2%) steels containing chromium (12 to 14%) and molybdenum
(0.2–1%).
414 416
Add Ni for improved Add P + S for

corrosion resistance improved machinability
410
Basic grade Add C to improve
420
0.15C-13.5Cr mechanical properties
Increased C to
improve toughness
Increased Cr for
increased corrosion 440
resistance
Figure E.5. The martensitic Grade 400 series.
Stainless steels hardened by transformation to martensite are tem-

pered to give the desired engineering properties. At high temperatures,
they have an austenitic structure that is transformed into martensitic struc-
ture upon cooling to room temperature. Unfortunately, this tempering can
influence corrosion susceptibility. For example, corrosion susceptibility
of type 420 stainless steel is at its maximum when the alloy is tempered
at temperatures in the range of 450° to 600°C. So, although not as
corrosion-resistant as the 200 and 300 classes, martensitic stainless steels
are magnetic, extremely strong (if not a little brittle), highly machinable,
and can be hardened by heat treatment.
Martensitic stainless steels are subject to both uniform and non-
uniform attack in seawater. And, the incubation time for a non-uniform
attack in even weak chlorides is often only a few hours or days.
Precipitation-Hardening Martensitic
Stainless Steels
These chromium- and nickel-containing steels can be precipitation-

hardened to develop very high tensile strengths. Precipitation-hardening
stainless steels are usually designated by a trade name, rather than by their
AISI 600 series designations.
The most common grade in this group is ‘17-4 PH,’ also known as
Grade 630, with a composition of 17% chromium, 4% nickel, 4% copper,
and 0.3% niobium. The main advantage of these steels is that they can
be supplied in the ‘solution-treated’ condition, in which state the steel
is just machinable. Following machining, forming, etc., the steel can be
hardened by a single, fairly low-temperature ‘ageing’ heat treatment that
causes no distortion of the component.
Precipitation-hardening generally results in a slight increase in
corrosion susceptibility and an increased susceptibility to hydrogen
embrittlement.
Figure E.6 shows the relationship between the complete family of
stainless steels.
Superferritic Ni-Cr-Fe
303, 303Se
stainless steel alloys
Add Ni for corrosion
Add Cr, Mo resistance in high Add S or Se for
temperature applications machinability
430 Duplex
stainless
309, 310, 314, 330
347 No Ni, Ferritic steel
Add Cr and Ni for Increase Cr,
Add Nb +Ta to Strength and lower Ni for
reduce sensitization oxidation resistance
higher strength
304 Precipitation
321 Add Ti to reduce Add Cu, Ti,
Fe-19Cr-10Ni hardening
sensitization Al, lower Ni
stainless steel
Add Mo for pitting
resistance Add Mn and N, lower Ni
304L
Lower C to for higher strength
316L reduce 316
No Ni addition, 201, 202
317L sensitization Add more Mo for lower Cr,
pitting resistance Martensitic
Superaustenitic Add Ni, Mo, N

stainless steel for corrosion 317 403, 410, 420
resistance
Figure E.6. Relationship between the complete family of stainless steels.

Glossary
∆P Differential pressure
AChI American Chemical Institute
ANSI American National Standards Institute
API American Petroleum Institute
ASME American Society of Mechanical Engineers
ASTM American Society for Testing and Materials
AWG American Wire Gauge
BSI British Standards Institute
CO Controller output
CV Valve flow coefficient
DIN Deutsches Institit für Normung
DN Nominal diameter
E/P Voltage to pneumatic converter
FCI Fluid Controls Institute
FL Pressure recovery coefficient
I/P Current to pneumatic converter
IEC International Electrotechnical Commission
IEEE Institute of Electrical and Electronic Engineers
ISA International Society for Automation
ISO International Organization for Standardization
Note: ISO is not an acronym, but is based on the Greek
word isos meaning equal.
J-T Joule–Thomson (effect)
KV Valve flow coefficient (SI alternative = 0.865 × CV)
MAWP Maximum allowable working pressure
MOV Motor-operated valve
MV Manipulated variable
NAMUR Normen Arbeitsgen Mess Und Regeltechnik (loosely
interpreted as Standards Work Group for Instruments
and Controls.)
264 • Glossary
NEMA National Electrical Manufacturers Association

OP Output
PD Process demand
PV Process variable
PDR Pressure drop ratio
PN Nominal pressure
Q Volumetric flow rate
Qm Mass flow rate
Re Reynolds number
SG Specific gravity
SGf Specific gravity of fluid
SGg Specific gravity of gas
SPL Sound pressure level
x Pressure drop ratio
XT Choked value of pressure drop ratio
Y Gas expansion factor
Z Compressibility factor
Bibliography
“Control Valve Trims and Devices to Control Cavitation Damage and Excessive
Noise,” Mitech, Technical Product Bulletin No 1.
“Introduction to Safety Valves.” Spirax Sarco, at: http://spiraxsarco.com/resources/
steam-engineering-tutorials/safety-valves/introduction-to-safety-valves.as
“Pressure Relief Valve Engineering Handbook” Technical Document No.
TP-V300, Crosby Valve Inc.
“The Mitech Globe Control Valve Body,” Mitech, Technical Product Bulletin No
2.
“Valve Signature Analysis” at: http://www2.emersonprocess.com/enUS/brands/
fisher/DigitalValveControllers/FIELDVUESolutions/ValveDiagnostics/
Pages/ValveSignatureBasics.aspx
Bell, L.H., and D.H. Bell. 1994. Industrial Noise Control: Fundamentals and
Applications. Marcel Dekker Inc.
Boger, H., and L.Mazot. Why Most Control Valves Today are Throttling Around
60% Opening. Masoneilan-Dresser.
Borden Jr., G. 1998. Control Valves. ISA.
Campbell, J.M. February 2004. Gas Conditioning and Processing, Vol. 1: Basic
Principles, 8th ed.
Chris, W. 1999. A User’s Guide, Understanding Valve Actuators. Rotork Controls
Inc.
Chris, W. 2000. “New generation of valve actuators can provide important MOV
Predictive Maintenance Data.” Rotork Controls Inc., Valve Magazine.
Comparison of Different Valve Types. Crane Process Flow Technologies Ltd.
Control Valve Noise Reduction. Fisher Rosemount.
Dave, H. Understanding Control Valve Bench Set. Control Engineering.
Dave, H. Understanding Control Valves. Control Engineering.
Elonka, S., and A.R. Parsons. 1962. Standard Instrumentation Questions and
Answers For Production-Processes Control, Vol. 1. McGraw-Hill.
Emerson, G. 2005. Control Valve Handbook, 4th ed. Emerson Process Manage-
ment.
Herrmann, U.F. 1974. “Sound Reinforcement.” N.V. Philips’ Gloeilampenfab-
rieken, Eindhoven.
266 • Bibliography
Husu, M., I. Niemelä, J. Pyötsiä, and M. Simula. 1992. Flow Control Manual.
Neles-Jamesbury.
Hutchison, J.W. 1976. ISA Handbook of Control Valves. ISA.
John, E. Positioner Guidelines. Emerson-Fisher-Rosemount.
Mike Sessions, Cavitation Control in Control Valves.
Practical Industrial Process Measurement for Engineers and Technicians. IDC
Technologies.
Richard, R. Designing a Positioner for the South African Market. Mitech.
Sam, L. Control Valve Manual. Masoneilan.
Stojkov, B.T. 1997. The Valve Primer. Industrial Press Inc.
About the Author
Michael (Mick) Crabtree, Joining the Royal Air Force as an apprentice,

Mick Crabtree trained in aircraft instrumentation and guided missiles.
Completing his service career seconded to the Ministry of Defense as a
technical writer, he emigrated to South Africa in 1966 where he worked,
for many years, for a local manufacturing and systems integration com-
pany involved in industrial process control, SCADA, and PLC-based
systems.
Later, as an editor and managing editor of a leading monthly engi-
neering journal, Mick wrote and published hundreds of articles, as well
as eight technical resource handbooks on industrial process control:
‘Flow Measurement,’ ‘Temperature Measurement,’ ‘Analytical On-line
Measurement,’ ‘Pressure and Level Measurement,’ ‘Valves,’ ‘Industrial
Communications,’ and ‘The Complete Profibus Handbook.’
He subsequently founded his own PR and advertising company and
was retained by a number of leading companies involved in the process
control industry, including: Honeywell, Fisher-Rosemount, Krohne,
Milltronics, and AEG. Apart from producing all their press releases and
articles, he also undertook the conceptualization and production of a wide
range of advertisements and data sheets, as well as newsletters.
For the last 16 years, he has been involved in technical training
and consultancy and has run workshops on industrial instrumentation
and networking throughout the world (United States, Canada, United
Kingdom, France, Southern Africa, Trinidad, Middle East, Australia, and
New Zealand). During this period, he has led more than 6,000 engineers,
technicians, and scientists on a variety of practical training workshops
covering the fields of process control (loop tuning), process instrumen-
tation, data communications, fieldbus, safety instrumentation systems
(according to both ISA S84 and IEC 61508/61511), project management,
online liquid analysis, and technical writing and communications.
268 • About the Author
Completing his studies in Electrical, Electronic, and Instrumentation

engineering, he holds an MSc in Industrial Flow Measurement from
Huddersfield University.
His hobbies and pastime include: cycling, rambling, history, and
reading.
After nearly 35 years spent in South Africa, he now lives in Wales,
just outside Cardiff, having relocated to Britain some 18 years ago.
Index
A Alarm generation, digital

Acoustics positioners, 164
absolute levels, 236 Austenitic stainless steel
decibel, 233–234 300 series, 258–259
intensity, 231 atom rearrangement, 256, 257
logarithmic characteristic of ear, ferro-magnetic material, 257
231–233
pitch (frequency), 227–228 B
sound pressure level, 236–237 Balanced safety relief valves
standards, 237–240 bellows-type, 204–206
timbre, 228–229 limitations, 204
velocity, 229–230 piston-type, 206–211
voltage ratios, 235–236 Bellows-type balanced safety
wavelength, 230 relief valves
Actuators back-pressure, 204
bench set, 183–184 bellows failure, 205, 206
cylinder actuator, 147–149 block schematics, 205
diaphragm actuator, 144–147 bonnet vents, 205–206
digital actuators, 155–156 Bench Set, 183–184
electric actuation, 150–152 Block-and-bleed system
electro-hydraulic actuation, capital cost, 241
149–150 double, 241
flapper–nozzle assembly, function, 241
141–142 multi-block valve systems,
hammer-blow mechanism, 153, 241
154 single double block-and-bleed
I/P converter, 142–144 manifold system, 241, 243
pneumatic control, 141 trunnion-mounted ball valves,
solenoid valve, 153–154 243
spool block, 149 Buckling pin safety valve
torque limiting, 152–153 basic, 211, 212
transfer mechanisms, 157–161 buckling point, 213
270 • Index
Euler’s law of compressed reverse-acting, 145, 146

columns, 212–213 springless, 145, 146
features, 213 Digital actuators
stable conditions, 213 DC motor, 155
Burst disc four-phase stepping motor, 156
advantages, 213–214 stepping motor, 156
applications and installation vernier arrangement, 156
practices, 218–219 Digital noise filtering, digital
configurations, 214–215 positioners, 164
cyclic/pulsating duties, 221–223 Digital positioners, 164–165
holders, 214, 215 Direct-acting diaphragm actuator,
maximum operating pressure, 144, 145
221, 222 Direct-acting solenoid valve, 154,
performance tolerance, 219–221 155
Double-crank transfer mechanism
C advantage, 159
Composite rupture disks, 216–217 modulating control applications,
Conventional domed rupture disc, 159
215, 216 rocker plate movement, 158, 159
Conventional safety relief valves run torque, 158, 159
ASME/ANSI standard, 201 Duplex stainless steels, 260
back-pressure
API 520 Recommended E
Practice Guidelines, 203 Electric actuators
built-up, 201 drawbacks, 151
fluid inlet pressure, 202 non-reversing characteristics,
forces acting on disc, 201 151
spring housing, 202, 203 spring closure, 152
superimposed, 200, 202–203 worm gear assembly, 151
limitations, 200 Electro-hydraulic actuation
spring housing, 201 drawback, 150
Cyclic/pulsating duties, 221–223 electronic control, 150
Cylinder actuator swing jet controller, 149, 150
cast cylinder, 147 Electronic torque monitoring,
vs. diaphragm cylinder, 147, 148 176–177
friction, 148–149
F
D Ferritic stainless steel, 256
Decibel, 233–234 Flapper–nozzle assembly,
Diaphragm actuator 141–142
advantages, 147 Fundamental frequency, 228
air pressure failure, 144
direct-acting, 144, 145 G
disadvantages, 147 Graphite rupture disc, 217–218
Index • 271
H burst disc, 213–215

Hammer-blow mechanism, 153, shear-pin safety valve, 211
154
Harmonics, 228–229 P
High-pressure pilot-operated Pascal (Pa), 184
valve, 208 Passivation, 255
Hysteretic error, 168 Pilot-operated safety relief valve
advantages, 207
I applications, 210
In-line repairs, 180, 182 blockage, 210
Intensity, 231 high-pressure pilot-operated
valve, 208
J inlet pressure, 208
J–T valve limitations, 210–211
construction, 225–226 low-pressure diaphragm-type,
Joule–Thomson effect, 225 209
reverse process, 225 overpressure and blowdown
performance, 210
K piston and seating arrangement,
Kilogram, 184 208, 209
process pressure, 207
L self-actuated auxiliary pressure
Low-pressure diaphragm-type relief valve, 207
pilot-operated valve, 209 Piston actuator. See Cylinder
actuator
M Piston-type balanced safety valves
Maintenance-related data, digital back-pressure, 207
positioner, 164 force balancing, 206, 207
Martensitic stainless steel pilot-operated safety relief valve,
corrosion susceptibility, 261 207–210
grade 400 series, 260, 261 Pitch (frequency), 227–228
precipitation-hardening, Positioners
261–262 digital, 164–165
uniform and nonuniform attack, electronic positioners, 162–163
261 feedback positioner, 162
I/P transducers, 161–162
N principle of operation, 162
Newton, 184 proportional control system, 161
Nitrogen-loaded surge relief set-point characterization, 163
valves, 252–253 Precipitation-hardening martensitic
Non-reclosing pressure safety stainless steel, 261–262
relief valves Pressure, 184
buckling pin safety valve, Pressure safety relief valves
211–213 applications, 194
272 • Index
history, 187–190 Scotch yoke transfer mechanism,

limitations, 194 160–161
non-reclosing, 211–215 Shear-pin safety valve, 211
spring-loaded relief valves, Shop repairs, 182–183
192–194 Signature analysis
weight-loaded valves, 191–192 minimum and maximum friction
Pulsation dampener, 251–252 value, 175
Pulsations, 250 opening and closing lines, 174
packing friction, 175
R revealed faults, 176
Rack and pinion transfer valve packing, 174
mechanism ‘valve signature’ plot, 174
disadvantages, 157 Solenoid valve
double-piston arrangement, 158 direct-acting, 154, 155
Relief valves, 190 shut-off applications, 154
Reverse-acting diaphragm three-way solenoid valve, 155
actuator, 145, 146 Sound pressure level (SPL),
Reverse Joule–Thomson effect, 236–237
225 Spring calculations, 184–186
Springless diaphragm actuator,
S 145, 146
Safety relief valves, 190. See also Springless piston-type actuator,
Pressure safety relief valves 149
balanced, 204 Spring-loaded pressure relief
composite rupture disks, valves
216–217 closed bonnets, 193, 194
conventional, 200–203 closing force, 193
conventional domed rupture disc, elements, 192–193
215, 216 static inlet pressure, 193
graphite rupture disc, 217–218 static pressure, 192
scored tension-loaded rupture Stainless steel
discs, 217, 218 200 series, 259–260
standards, 223–224 300 grade alloys, 258–259
Safety valves austenitic stainless steel,
closing pressure, 195 256–260
curtain area, 196 classification issues, 255–256
discharge area, 196 duplex, 260
flow area, 195, 196 ferritic stainless steel, 256
lifting, 196–198 martensitic, 260–262
reseating, 198–200 passivation, 255
static inlet pressure, 194–195 Stick-slip response, 170
Scored tension-loaded rupture Superimposed back-pressure, 200
discs, 217, 218 Swing jet controller, 149, 150
Index • 273
T hysteretic error, 168

Threshold of hearing, 237, 238 linear system, 167
Threshold of pain, 237, 238 step changes, 169
Timbre, 228–229 electronic torque monitoring,
Torque limiting, 152–153 176–177
Transfer mechanisms non-linearity, 170–171
double-crank mechanism, online diagnostics
158–160 modern Fieldbus
rack and pinion, 157–158 communication systems,
scotch yoke mechanism, 173
160–161 signature analysis, 174–176
Trunnion-mounted ball valves, 243 stick-slip response, 170
Velocity, 229–230
U Voltage ratios, 235–236
Unbalance of force, 186
W
V Water hammer
Valve maintenance and repair bulk modulus, 245
actuator bench set, 183–184 effective bulk modulus,
drained system repair, 181 246–247
dynamic environment, 179 magnitude of pressure spike, 248
fluid leakage, 179 nitrogen-loaded surge relief
in-line repairs, 180, 182 valves, 252–253
packing replacement, 181 prevention and mitigation,
repairs under pressure, 180–181 250–251
seat rings replacement, 181–182 pulsation dampener, 251–252
shaft leakage, 179 pulsations, 250
shop repairs, 182–183 quick-closing valve, 245
spring calculations, 184–186 sonic velocities, 245–246
Valve stroke speed control, digital steam condensate, 249–250
positioner, 165 valve size and system condition,
Valve testing and diagnostics 248
complete assembly, 171–173 velocity of pressure wave, 245,
deadband and hysteresis 246
acceptable limits, 170 Wavelength, 230
friction, 168–169 Weight-loaded pressure/vacuum
gear-train system, 167, 168 relief valves, 191–192
OTHER TITLES IN OUR AUTOMATION AND
CONTROL COLLECTION
Measurement and Monitoring

by Vytautas Giniotis and Anthony Hope
Flexible Test Automation: A Software Framework for Easily Developing

Measurement Applications
by Pasquale Arpaia, Vitaliano Inglese, and Ernesto De Matteis
Resources Utilization and Productivity Enhancement Case Studies

by Anil Mital and Arunkumar Pennathur
Momentum Press is one of the leading book publishers in the field of engineering,
mathematics, health, and applied sciences. Momentum Press offers over 30 collections,
including Aerospace, Biomedical, Civil, Environmental, Nanomaterials, Geotechnical,
and many others.
Momentum Press is actively seeking collection editors as well as authors. For more
information about becoming an MP author or collection editor, please visit
http://www.momentumpress.net/contact
Announcing Digital Content Crafted by Librarians

Momentum Press offers digital content as authoritative treatments of advanced engineering top-
ics by leaders in their field. Hosted on ebrary, MP provides practitioners, researchers, faculty,
and students in engineering, science, and industry with innovative electronic content in sensors
and controls engineering, advanced energy engineering, manufacturing, and materials science.
Momentum Press offers library-friendly terms:
• perpetual access for a one-time fee

• no subscriptions or access fees required
• unlimited concurrent usage permitted
• downloadable PDFs provided
• free MARC records included
• free trials
The Momentum Press digital library is very affordable, with no obligation to buy in future years.
For more information, please visit www.momentumpress.net/library or to set up a trial in the US,
please contact mpsales@globalepress.com.
EBOOKS The Concise Valve Handbook
CRABTREE
FOR THE Actuation, Maintenance, and Safety Relief, AUTOMATION AND CONTROL
ENGINEERING Volume II COLLECTION
LIBRARY
Michael A. Crabtree
Create your own
Research studies within the process industry routinely indicate
Customized Content
that the fluid control valve is responsible for 60 to 70% of poor-
Bundle — the more
functioning control systems. Furthermore, valves in general are
books you buy, consistently wrongly selected, regularly misapplied, and often
the higher your incorrectly installed.
discount! This two-volume book comprises a comprehensive up-to-date
body of knowledge that provides a total in-depth insight into valve
THE CONTENT and actuator technology—looking not just at control valves, but a
• Manufacturing whole host of other types including: check valves, shut-off valves,
The Concise Valve

Engineering solenoid valves, and pressure relief valves.
A methodology is presented to ensure the optimum selection of
The Concise Valve Handbook, Volume II

• Mechanical
& Chemical size, choice of body and trim materials, components, and ancillaries.
Handbook
Engineering Whilst studying the correct procedures for sizing, readers will also
• Materials Science learn the correct procedures for calculating the spring ‘wind-up’ or
& Engineering ‘bench set’.
• Civil & Maintenance issues also include: testing for deadband/
Environmental
Engineering
hysteresis, stick-slip and non-linearity; on-line diagnostics; and
signature analysis.
Actuation, Maintenance,
• Advanced Energy
Technologies
Written in a detailed but understandable language, the two
volumes are presented in a form suitable for both the beginner,
and Safety Relief
with no prior knowledge of the subject, and the more advanced
THE TERMS specialist.
Volume II
• Perpetual access for For the last sixteen years, ‘Mick’ Crabtree, who holds an MSc in
a one time fee industrial flow measurement, has been involved in technical training
• No subscriptions or and consultancy—running workshops on industrial instrumentation
access fees and networking throughout the world covering the fields of process
• Unlimited control (loop tuning), process instrumentation, data communications,
concurrent usage fieldbus, safety instrumentation systems (according to both ISA S84
• Downloadable PDFs and IEC 61508/61511), project management, on-line analysis, and
• Free MARC records technical writing and communications.
This book represents some thirty years wealth of experiential
For further information,
a free trial, or to order,
contact:
knowledge gleaned by the author working within a wide variety of
industries and from more than 6000 technicians and engineers who Michael A. Crabtree
have attended the author’s workshops.
sales@momentumpress.net
ISBN: 978-1-94708-369-1

The Concise Valve V II

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Concise Valve V II

Uploaded by

Copyright:

Available Formats

EBOOKS The Concise Valve Handbook

The Concise Valve

The Concise Valve Handbook, Volume II

MOMENTUM PRESS, LLC, NEW YORK

Copyright © Momentum Press®, LLC, 2018.

All rights reserved. No part of this publication may be reproduced, stored

First published by Momentum Press®, LLC

ISBN-13: 978-1-94708-369-1 (print)

Momentum Press Automation and Control Collection

Cover and interior design by Exeter Premedia Services Private Ltd.,

Printed in the United States of America

Volume II: Actuation, Maintenance, and Safety Relief, takes an in-depth

acoustics, actuators, block and bleed, diagnostics, fail-safe, maintenance,

2.4 Piping Geometry Factor 29

5.14 Aerodynamic Noise 117

8 Valve Testing and Diagnostics 167

9 Valve Maintenance and Repair 179

10 Safety Relief Valves 187

10.17 Composite Rupture Disc 216

Figure 7.1. The flapper–nozzle assembly converts a

Figure 7.12. Basic electrically operated actuator. 151

Figure 9.1. Although not generally recommended, in-line repair

Figure 10.14. Schematic diagram of a valve with the spring housing

Figure 10.31. Bursting disc installed on a safety valve. 219

Figure D.1. If a moving column of liquid (a) is slowed down

Table 10.1. Typical operating pressure to burst pressure ratios

In this book, ‘The Concise Valve Handbook—Part 2. Actuation, Maintenance,

1 bar = 100 kPa ≈ 1 atmosphere ≈ 14.7 psi

And lastly, while I have made some compromises (analog instead

English United States

Quantity SI United States customary

Quantity SI United States customary

Valve Actuators and

In the process control instrumentation field, the pneumatically controlled

• users feel that there is little, if any, improvement in the performance

Figure 7.1. The flapper–nozzle assembly converts a

displacement into a pressure change. An air supply is applied to the nozzle

A common application of the flapper–nozzle device is the electro-­

Figure 7.2. As the flapper moves away from the nozzle,

Figure 7.3. The 4–20 mA current signal is applied

Supply air Control air

Figure 7.4. Typical configuration of an I/P converter (courtesy

Travel stop Control air input

Figure 7.5. A direct-acting diaphragm

In many cases, the fail-open mode is highly desirable. There are,

Figure 7.6. A reverse-acting diaphragm actuator.

Figure 7.7. Springless diaphragm actuator

7.6 Advantages and Disadvantages of

The main advantage of the diaphragm-type actuator is its cost since it

Figure 7.8. The cylinder or piston-type

Figure 7.9. Size comparison between a diaphragm

Generally, spring-and-diaphragm actuators contribute less friction to

degrades. Thus, to ensure continued good performance, maintenance is

The increasing acceptance of electronics in the process control industry

Supply in Supply in Supply in

Figure 7.10. Pneumatic spool assembly in which the incoming air is

Figure 7.11. Swing jet

One of the major disadvantages of the electro-hydraulic-type system is

Figure 7.12. Basic electrically operated

or mechanical failure. A number of solutions are also offered in which a

Splined shaft Worm drive

Figure 7.13. (a) The worm drive is free to move longitudi-

The solenoid valve is essentially an on–off control element comprising an

2.4 Piping Geometry Factor 29

5.14 Aerodynamic Noise 117

8 Valve Testing and Diagnostics 167

9 Valve Maintenance and Repair 179

10 Safety Relief Valves 187

10.17 Composite Rupture Disc 216

Figure 7.1. The flapper–nozzle assembly converts a

Figure 7.12. Basic electrically operated actuator. 151

Figure 9.1. Although not generally recommended, in-line repair

Figure 10.14. Schematic diagram of a valve with the spring housing

Figure 10.31. Bursting disc installed on a safety valve. 219

Figure D.1. If a moving column of liquid (a) is slowed down

Table 10.1. Typical operating pressure to burst pressure ratios

A common application of the flapper–nozzle device is the electro-

7.6 Advantages and Disadvantages of

7.15.1 Rack and Pinion

7.17.1 All Positioner Types

8.1 Deadband and Hysteresis

8.2.1 Deadband and Hysteresis

8.2.4 Testing a Complete Assembly

8.4 Electronic Torque Monitoring

9.2 Repairs Under Pressure