Industrial Control
Technology
Industrial Control
Technology
A Handbook for
Engineers and Researchers
Peng Zhang
Beijing Normal University, People’s Republic of China
N o r w i c h , NY, USA
Copyright © 2008 by William Andrew Inc.
No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system,
without permission in writing from the Publisher.
ISBN: 978-0-8155-1571-5
Library of Congress Cataloging-in-Publication Data
Zhang, Peng.
Industrial control technology : a handbook for engineers and researchers / Peng Zhang.
p. cm.
Includes bibliographical references and index.
1. Process control--Handbooks, manuals, etc. 2. Automatic control--Handbooks, manuals, etc. I.
Title.
TS156.8.Z43 2008
670.42--dc22
2008002701
Printed in the United States of America
This book is printed on acid-free paper.
10 9 8 7 6 5 4 3 2 1
Published by:
William Andrew Inc.
13 Eaton Avenue
Norwich, NY 13815
1-800-932-7045
www.williamandrew.com
Environmentally Friendly
This book has been printed digitally because this process does not use any plates, ink,
chemicals, or press solutions that are harmful to the environment. The paper used in
this book has a 30% recycled content.
NOTICE
To the best of our knowledge the information in this publication is accurate; however the Publisher
does not assume any responsibility or liability for the accuracy or completeness of, or consequences
arising from, such information. This book is intended for informational purposes only. Mention of
trade names or commercial products does not constitute endorsement or recommendation for their
use by the Publisher. Final determination of the suitability of any information or product for any use,
and the manner of that use, is the sole responsibility of the user. Anyone intending to rely upon any
recommendation of materials or procedures mentioned in this publication should be independently
satisfied as to such suitability, and must meet all applicable safety and health standards.
Contents
Preface
xix
1 Sensors and Actuators for Industrial Control
1.1 Sensors
1.1.1
Bimetallic Switch
1.1.1.1 Operating Principle
1.1.1.2 Basic Types
1.1.1.3 Application Guide
1.1.1.4 Calibration
1.1.2
Color Sensors
1.1.2.1 Operating Principle
1.1.2.2 Basic Types
1.1.2.3 Application Guide
1.1.2.4 Calibration
1.1.3
Ultrasonic Distance Sensors
1.1.3.1 Operating Principle
1.1.3.2 Basic Types
1.1.3.3 Application Guide
1.1.3.4 Calibration
1.1.4
Light Section Sensors
1.1.4.1 Operating Principle
1.1.4.2 Application Guide
1.1.4.3 Specifications
1.1.4.4 Calibration
1.1.5
Linear and Rotary Variable Differential
Transformers
1.1.5.1 Operating Principle
1.1.5.2 Application Guide
1.1.5.3 Calibration
1.1.6
Magnetic Control Systems
1.1.6.1 Operating Principle
1.1.6.2 Basic Types and Application Guide
1.1.7
Limit Switches
1.1.7.1 Operating Principle
1.1.7.2 Basic Types and Application Guide
1.1.7.3 Calibration
1
1
1
1
2
4
5
6
6
7
9
9
10
10
11
12
12
15
15
16
19
20
22
22
25
26
27
28
33
38
38
40
43
vii
Zhang_Prelims.indd vii
5/24/2008 9:42:07 AM
viii
CONTENTS
1.1.8
1.2
Zhang_Prelims.indd viii
Photoelectric Devices
1.1.8.1 Operating Principle
1.1.8.2 Application Guide
1.1.8.3 Basic Types
1.1.9
Proximity Devices
1.1.9.1 Operating Principle
1.1.9.2 Application Guide
1.1.9.3 Basic Types and Specifications
1.1.10 Scan Sensors
1.1.10.1 Operating Principle
1.1.10.2 Basic Types
1.1.10.3 Technical Specifications
1.1.11 Force and Load Sensors
1.1.11.1 Operating Principle
1.1.11.2 Basic Types
1.1.11.3 Technical Specifications
1.1.11.4 Calibration
Actuators
1.2.1
Electric Actuators
1.2.1.1 Operating Principle
1.2.1.2 Basic Types
1.2.1.3 Technical Specification
1.2.1.4 Application Guides
1.2.1.5 Calibrations
1.2.2
Pneumatic Actuators
1.2.2.1 Operating Principle
1.2.2.2 Basic Types and Specifications
1.2.2.3 Application Guide and Assembly
on Valve
1.2.3
Hydraulic Actuators
1.2.3.1 Operating Principle
1.2.3.2 Basic Types and Specifications
1.2.3.3 Application Guide
1.2.3.4 Calibration
1.2.4
Piezoelectric Actuators
1.2.4.1 Operating Principle
1.2.4.2 Basic Types
1.2.4.3 Technical Specifications
1.2.4.4 Calibration
1.2.5
Manual Actuators
44
44
47
49
52
53
56
56
63
63
65
69
72
72
77
79
86
87
88
88
90
94
96
98
100
100
104
106
111
111
115
119
123
125
126
132
136
137
141
5/24/2008 9:42:07 AM
CONTENTS
ix
1.3
142
142
143
149
150
155
155
161
165
166
170
171
172
173
175
175
177
177
179
180
2
Valves
1.3.1
Control Valves
1.3.1.1 Basic Types
1.3.1.2 Technical Specifications
1.3.1.3 Application Guide
1.3.2
Self-Actuated Valves
1.3.2.1 Check Valves
1.3.2.2 Relief Valves
1.3.3
Solenoid Valves
1.3.3.1 Operating Principles
1.3.3.2 Basic Types
1.3.3.3 Technical Specifications
1.3.4
Float Valves
1.3.4.1 Operating Principle
1.3.4.2 Specifications and Application Guide
1.3.4.3 Calibration
1.3.5
Flow Valves
1.3.5.1 Operating Principle
1.3.5.2 Specifications and Application Guide
1.3.5.3 Calibration
Computer Hardware for Industrial Control
2.1 Microprocessor Unit Chipset
2.1.1
Microprocessor Unit Organization
2.1.1.1 Function Block Diagram of a
Microprocessor Unit
2.1.1.2 Microprocessor
2.1.1.3 Internal Bus System
2.1.1.4 Memories
2.1.1.5 Input/Output Pins
2.1.1.6 Interrupt System
2.1.2
Microprocessor Unit Interrupt Operations
2.1.2.1 Interrupt Process
2.1.2.2 Interrupt Vectors
2.1.2.3 Interrupts Service Routine (ISR)
2.1.3
Microprocessor Unit Input/Output Rationale
2.1.3.1 Basic Input/Output Techniques
2.1.3.2 Basic Input/Output Interfaces
2.1.4
Microprocessor Unit Bus System Operations
2.1.4.1 Bus Operations
2.1.4.2 Bus System Arbitration
Zhang_Prelims.indd ix
187
187
190
191
192
201
201
205
207
207
208
210
210
213
213
216
218
219
222
5/24/2008 9:42:08 AM
x
CONTENTS
2.1.4.3 Interrupt Routing
2.1.4.4 Configuration Registers
Programmable Peripheral Devices
2.2.1
Programmable Peripheral I/O Ports
2.2.2
Programmable Interrupt Controller Chipset
2.2.3
Programmable Timer Controller Chipset
2.2.4
CMOS Chipset
2.2.5
Direct Memory Access Controller Chipset
2.2.5.1 Idle Cycle
2.2.5.2 Active Cycle
Application-Specific Integrated Circuit (ASIC)
2.3.1
ASIC Designs
2.3.1.1 ASIC Specification
2.3.1.2 ASIC Functional Simulation
2.3.1.3 ASIC Synthesis
2.3.1.4 ASIC Design Verification
2.3.1.5 ASIC Integrity Analyses
2.3.2
Programmable Logic Devices (PLD)
2.3.3
Field-Programmable Gate Array (FPGA)
2.3.3.1 FPGA Types and Important Data
2.3.3.2 FPGA Architecture
2.3.3.3 FPGA Programming
223
224
226
226
229
233
235
235
238
239
240
242
242
243
244
246
247
248
250
251
252
255
System Interfaces for Industrial Control
3.1 Actuator–Sensor (AS) Interface
3.1.1
Overview
3.1.2
Architectures and Components
3.1.2.1 AS Interface Architecture: Type 1
3.1.2.2 AS Interface Architecture: Type 2
3.1.3
Working Principle and Mechanism
3.1.3.1 Master–Slave Principle
3.1.3.2 Data Transfer
3.1.4
System Characteristics and Important Data
3.1.4.1 How the AS Interface Functions
3.1.4.2 Physical Characteristics
3.1.4.3 System Limits
3.1.4.4 Range of Functions of the Master
Modules
3.1.4.5 AS Interface in a Real-Time
Environment
259
259
259
260
261
263
266
267
269
275
275
275
276
2.2
2.3
3
Zhang_Prelims.indd x
277
277
5/24/2008 9:42:08 AM
CONTENTS
xi
3.2
279
280
280
289
291
309
319
327
328
333
335
339
344
345
347
351
351
353
3.3
3.4
Zhang_Prelims.indd xi
Industrial Control System Interface Devices
3.2.1
Fieldbus System
3.2.1.1 Foundation Fieldbus
3.2.1.2 PROFIBUS
3.2.1.3 Controller Area Network (CAN bus)
3.2.1.4 Interbus
3.2.1.5 Ethernets/Hubs
3.2.2
Interfaces
3.2.2.1 PCI, ISA, and PCMCIA
3.2.2.2 IDE
3.2.2.3 SCSI
3.2.2.4 USB and Firewire
3.2.2.5 AGP and Parallel Ports
3.2.2.6 RS-232, RS-422, RS-485, and RS-530
3.2.2.7 IEEE-488
Human–Machine Interface in Industrial Control
3.3.1
Overview
3.3.2
Human–Machine Interactions
3.3.2.1 The Models for Human–Machine
Interactions
3.3.2.2 Systems of Human–Machine Interactions
3.3.2.3 Designs of Human–Machine Interactions
3.3.3
Interfaces
3.3.3.1 Devices
3.3.3.2 Tools
3.3.3.3 Software
Highway Addressable Remote Transducer (HART)
Field Communications
3.4.1
HART Communication
3.4.1.1 HART networks
3.4.1.2 HART Mechanism
3.4.2
HART System
3.4.2.1 HART System Devices
3.4.2.2 HART System Installation
3.4.2.3 HART System Configuration
3.4.2.4 HART System Calibration
3.4.3
HART Protocol
3.4.3.1 HART Protocol Model
3.4.3.2 HART Protocol Commands
3.4.3.3 HART Protocol Data
353
365
368
371
371
375
376
377
378
378
383
387
387
397
400
402
406
406
409
411
5/24/2008 9:42:08 AM
xii
CONTENTS
3.4.4
4
HART Integration
3.4.4.1 Basic Industrial Field Networks
3.4.4.2 Choosing the Right Field Networks
3.4.4.3 Integrating the HART with Other
Field Networks
Digital Controllers for Industrial Control
4.1 Industrial Intelligent Controllers
4.1.1
Programmable Logic Control (PLC) Controllers
4.1.1.1 Components and Architectures
4.1.1.2 Control Mechanism
4.1.1.3 PLC Programming
4.1.1.4 Basic Types and Important Data
4.1.1.5 Installation and Maintenance
4.1.2
Computer Numerical Control (CNC) Controllers
4.1.2.1 Components and Architectures
4.1.2.2 Control Mechanism
4.1.2.3 CNC Part Programming
4.1.2.4 CNC Controller Specifications
4.1.3
Supervisory Control and Data Acquisition
(SCADA) Controllers
4.1.3.1 Components and Architectures
4.1.3.2 SCADA Protocols
4.1.3.3 Functions and Administrations
4.1.4
Proportional-Integration-Derivative (PID)
Controllers
4.1.4.1 PID Control Mechanism
4.1.4.2 PID Controller Implementation
4.1.4.3 PID Controller Tuning Rules
4.1.4.4 PID Control Technical Specifications
4.2 Industrial Process Controllers
4.2.1
Batch Controllers
4.2.1.1 Batch Control Standards
4.2.1.2 Control Mechanism
4.2.2
Servo Controllers
4.2.2.1 Components and Architectures
4.2.2.2 Control Mechanism
4.2.2.3 Distributed Servo Control
4.2.2.4 Important Servo Control Devices
4.2.3
Fuzzy Logic Controllers
4.2.3.1 Fuzzy Control Principle
4.2.3.2 Fuzzy Logic Process Controllers
Zhang_Prelims.indd xii
415
415
420
420
429
429
429
429
437
440
454
455
462
463
466
474
483
488
488
498
512
519
519
520
524
526
532
532
534
536
539
540
544
547
550
558
559
564
5/24/2008 9:42:08 AM
CONTENTS
xiii
5 Application Software for Industrial Control
5.1 Boot Code for Microprocessor Unit Chipset
5.1.1
Introduction
5.1.2
Code Structures
5.1.2.1 BIOS and Kernel
5.1.2.2 Master Boot Record (MBR)
5.1.2.3 Boot Program
5.1.3
Boot Sequence
5.1.3.1 Power On
5.1.3.2 Load BIOS, MBR and Boot
Program
5.1.3.3 Initiate Hardware Components
5.1.3.4 Initiate Interrupt Vectors
5.1.3.5 Transfer to Operating System
5.2 Real-Time Operating System
5.2.1
Introduction
5.2.2
Task Controls
5.2.2.1 Multitasking Concepts
5.2.2.2 Task Types
5.2.2.3 Task Stack and Heap
5.2.2.4 Task States
5.2.2.5 Task Body
5.2.2.6 Task Creation and Termination
5.2.2.7 Task Queue
5.2.2.8 Task Context Switch and Task
Scheduler
5.2.2.9 Task Threads
5.2.3
Input/Output Device Drivers
5.2.3.1 I/O Device Types
5.2.3.2 Driver Content
5.2.3.3 Driver Status
5.2.3.4 Request Contention
5.2.3.5 I/O Operations
5.2.4
Interrupts
5.2.4.1 Interrupt Handling
5.2.4.2 Enable and Disable Interrupts
5.2.4.3 Interrupt Vector
5.2.4.4 Interrupt Service Routines
5.2.5
Memory Management
5.2.5.1 Virtual Memory
5.2.5.2 Dynamic Memory Pool
569
570
570
570
571
573
575
575
575
Zhang_Prelims.indd xiii
577
577
578
578
579
579
579
579
581
582
585
586
586
588
589
593
595
597
597
597
598
599
601
601
608
609
610
612
613
616
5/24/2008 9:42:08 AM
xiv
CONTENTS
5.3
Zhang_Prelims.indd xiv
5.2.5.3 Memory Allocation and
Deallocation
5.2.5.4 Memory Requests Management
5.2.6
Event Brokers
5.2.6.1 Event Notification Service
5.2.6.2 Event Trigger
5.2.6.3 Event Broadcasts
5.2.6.4 Event Handling Routine
5.2.7
Message Queue
5.2.7.1 Message Passing
5.2.7.2 Message Queue Types
5.2.7.3 Pipes
5.2.8
Semaphores
5.2.8.1 Semaphore Depth and Priority
5.2.8.2 Semaphore Acquire, Release and
Shutdown
5.2.8.3 Condition and Locker
5.2.9
Timers
5.2.9.1 Kernel Timers
5.2.9.2 Watchdog Timers
5.2.9.3 Task Timers
5.2.9.4 Timer Creation and Expiration
Real-Time Application System
5.3.1
Architecture
5.3.2
Input/Output Protocol Controllers
5.3.2.1 Server or Manager
5.3.2.2 I/O Device Module
5.3.3
Process
5.3.3.1 Process Types
5.3.3.2 Process Attributes
5.3.3.3 Process Status
5.3.3.4 Process and Task
5.3.3.5 Process Creation, Evolution, and
Termination
5.3.3.6 Synchronization
5.3.3.7 Mutual Exclusive
5.3.4
Finite State Automata
5.3.4.1 Models
5.3.4.2 Designs
5.3.4.3 Implementation and Programming
616
618
618
619
621
621
622
622
622
625
626
629
630
632
634
638
639
640
645
646
647
647
650
650
652
653
654
654
655
656
656
657
658
659
660
665
667
5/24/2008 9:42:08 AM
CONTENTS
xv
6 Data Communications in Distributed Control System
6.1 Distributed Industrial Control System
6.1.1
Introduction
6.1.1.1 Opened Architectures for Distributed
Control
6.1.1.2 Closed Architectures for Distributed
Control
6.1.1.3 Similarity to Computer Network
6.1.2
Data Communication Model for Distributed
Control System
6.1.2.1 Data Communication Models for
Open Control Systems
6.1.2.2 Data Communication Models for
Closed-Control Systems
6.2 Data Communication Basics
6.2.1
Introduction
6.2.1.1 Data Transfers within an IC Chipset
6.2.1.2 Data Transfers over Medium Distances
6.2.1.3 Data Transfer over Long Distances
6.2.2
Data Formats
6.2.2.1 Bit
6.2.2.2 Byte
6.2.2.3 Character
6.2.2.4 Word
6.2.2.5 Basic Codeword Standards
6.2.3
Electrical Signal Transmission Modes
6.2.3.1 Bit-Serial and Bit-Parallel Modes
6.2.3.2 Word-Parallel Mode
6.2.3.3 Simplex Mode
6.2.3.4 Half-Duplex Mode
6.2.3.5 Full-Duplex Mode
6.2.3.6 Multiplexing Mode
6.3 Data Transmission Control Circuits and Devices
6.3.1
Introduction
6.3.2
Universal Asynchronous Receiver Transmitter
(UART)
6.3.2.1 Applications and Types
6.3.2.2 Mechanism and Components
6.3.3
Universal Synchronous Receiver Transmitter
(USRT)
675
675
675
Zhang_Prelims.indd xv
676
678
680
680
681
690
691
691
691
693
693
695
695
696
696
697
698
700
700
701
701
702
703
703
705
705
706
706
707
708
5/24/2008 9:42:08 AM
xvi
CONTENTS
6.3.4
6.4
6.5
Zhang_Prelims.indd xvi
Universal Synchronous/Asynchronous
Receiver Transmitter (USART)
709
6.3.4.1 Architecture and Components
709
6.3.4.2 Mechanism and Modes
712
6.3.5
Bit-Oriented Protocol Circuits
718
6.3.5.1 SDLC Controller
719
6.3.5.2 HDLC Controller
721
6.3.6
Multiplexers
721
6.3.6.1 Digital Multiplexer
722
6.3.6.2 Time Division Multiplexer (TDM)
723
Data Transmission Protocols
725
6.4.1
Introduction
725
6.4.2
Asynchronous Transmission
726
6.4.2.1 Bit Synchronization
726
6.4.2.2 Character Synchronization
727
6.4.2.3 Frame Synchronization
730
6.4.3
Synchronous Transmission
733
6.4.3.1 Bit Synchronization
734
6.4.3.2 Character-Oriented Synchronous
Transmission
737
6.4.3.3 Bit-oriented Synchronous Transmission 739
6.4.4
Data Compression and Decompression
740
6.4.4.1 Loss and Lossless Compression and
Decompression
741
6.4.4.2 Data Encoding and Decoding
741
6.4.4.3 Basic Data Compression Algorithms
742
Data-Link Protocols
749
6.5.1
Framing Controls
749
6.5.1.1 High-Level Data Link Control (HDLC) 750
6.5.1.2 Synchronous Data Link Control (SDLC) 752
6.5.2
Error Controls
753
6.5.2.1 Error Detection
754
6.5.2.2 Error Correction
755
6.5.3
Flow Controls
758
6.5.3.1 Stop-and-Wait
758
6.5.3.2 Sliding Window
759
6.5.3.3 Bus Arbitration
760
6.5.4
Sublayers
760
6.5.4.1 Logic Link Control (LLC)
760
6.5.4.2 Media Access Control (MAC)
762
5/24/2008 9:42:08 AM
CONTENTS
xvii
6.6
Data Communication Protocols
6.6.1
Client–Server Model
6.6.1.1 Two and Three-Tier Client–Server
6.6.1.2 Message Server
6.6.1.3 Application Server
6.6.2
Master–Slave Model
6.6.2.1 Master
6.6.2.2 Slave
6.6.3
Producer–Consumer Model
6.6.3.1 Designs
6.6.3.2 Implementations
6.6.4
Remote Procedure Call (RPC)
763
763
764
765
766
766
767
767
768
768
769
770
System Routines in Industrial Control
7.1 Overview
7.2 Power-On and Power-Down Routines
7.2.1
System Hardware Requirements
7.2.1.1 Low Voltage Power Supply Circuit
(LVPSC)
7.2.1.2 Basic Input and Output System
(BIOS)
7.2.2
System Power-On Process
7.2.3
System Power-On Self Tests
7.2.3.1 When Does the POST Apply?
7.2.3.2 What does the POST do?
7.2.3.3 Who Does the POST?
7.2.4
System Power-Down Process
7.3 Install and Configure Routines
7.3.1
System Hardware Requirements
7.3.1.1 PCI Address Spaces
7.3.1.2 PCI Configuration Headers
7.3.1.3 PCI I/O and PCI Memory Addresses
7.3.1.4 PCI-ISA Bridges
7.3.1.5 PCI-PCI Bridges
7.3.1.6 PCI Initialization
7.3.1.7 The PCI Device Driver
7.3.1.8 PCI BIOS Functions
7.3.1.9 PCI Firmware
7.3.2
System Devices Install and Configure Routine
7.3.3
System Configure Routine
775
775
776
778
7
Zhang_Prelims.indd xvii
778
780
781
783
783
783
785
785
788
789
790
790
791
791
792
795
796
798
800
802
803
5/24/2008 9:42:08 AM
xviii
7.4
7.5
Index
Zhang_Prelims.indd xviii
CONTENTS
Diagnostic Routines
7.4.1
System Hardware Requirements
7.4.2
Device Component Test Routines
7.4.3
System NVM Read and Write Routines
7.4.4
Faults/Errors Log Routines
7.4.5
Change System Mode Routines
7.4.5.1 System Modes List
7.4.5.2 System Modes Transition
7.4.6
Calibration Routines
7.4.6.1 Calibration Fundamentals
7.4.6.2 Calibration Principles
7.4.6.3 Calibration Methodologies
Simulation Routines
7.5.1
Modeling and Simulation
7.5.1.1 Process Models
7.5.1.2 Process Modeling
7.5.1.3 Control Simulation
7.5.2
Methodologies and Technologies
7.5.2.1 Manufacturing Process Modeling
and Simulation
7.5.2.2 Computer Control System Modeling
and Simulation
7.5.3
Simulation Program Organization
7.5.3.1 Simulation Routines for Single
Microprocessor Control Systems
7.5.3.2 Simulation Routines for Distributed
Control Systems
7.5.3.3 Simulation Routine Coding Principles
7.5.4
Simulators, Toolkits, and Toolboxes
7.5.4.1 MATLAB
7.5.4.2 SIMULINK
7.5.4.3 SIMULINK Real-Time Workshop
7.5.4.4 ModelSim
7.5.4.5 Link for ModelSim
804
805
806
807
808
809
810
811
813
813
814
816
817
818
818
821
826
831
833
836
840
840
840
841
841
841
844
846
847
849
853
5/24/2008 9:42:08 AM
Preface
Objectives
Industrial control consists of industrial process control and industrial
production automation. This book applies to both industrial process control and industrial production automation, and it covers the technology in
three branches: theory, design, and technology.
In recent years, there has been a technical revolution in the semiconductor
industry and in the electronics industry, which has significantly advanced
the existing technologies in industrial control. The recent technical developments in the semiconductor and electronics industries are mainly represented as these seven aspects:
(1) The microprocessor chipsets have been very capable in interrupt
handling, data passage, and interface communication.
(2) The operating speeds of both microprocessors and programmable integrated circuits have become much faster.
(3) The enhancements in the register arrays and the instruction set of
microprocessor units have made multitasking or multithreads
possible.
(4) The sizes of various semiconductor chips are being increased
and their production costs are going lower and lower.
(5) The controllers of intelligent functionalities are more and more
designed to perform various control strategies and protocols. For
example, Programmable Logic Control (PLC) controllers implement Ladder Logic, and fuzzy logic controllers operate in terms
of fuzzy control theory; the Controller Area Network (CAN) is a
very powerful automatic system used even in aerospace. These
industrial intelligent controllers are being increasingly used in
industrial control so that the establishment of industrial control
systems is becoming more and more feasible.
(6) The various development tools for both hardware and software
are becoming more and more feasible and powerful, which is
largely shortening the time for developing software and hardware and is significantly enhancing their quality.
(7) The programmable application-specific integrated circuits (ASIC)
have now approached an intelligence similar to that of microprocessors, so that they are performing a more important functional
role in various control systems.
xix
Zhang_Prelims.indd xix
5/24/2008 9:42:08 AM
xx
PREFACE
These technical developments in both the semiconductor industry and
the electronics industry have advanced industrial control into both realtime control and distributed control. Real-time control requires controllers to capture all the significant target activities and to deliver their
responses as swiftly as possible so that system performance is never
degraded. Distributed control indicates that controls are performed by
a number of microprocessor controllers and executed in a group of
independent agents or units that are physically and electronically connected and communicate with each other. This tendency in industrial
control has led to the future continuation of both real-time control and
distributed control. Consequently, industrial control has been gradually
extended from device and machine control to plant and enterprise and
industry.
To demonstrate that these technical developments satisfy the new
industrial control requirements, this book provides comprehensive technical details, including the necessary rationales, methodologies, types,
parameters, and specifications, for the devices of industrial control. As a
technical handbook for engineers, a technical reference, and an academic
textbook for students, this book particularly emphasizes the following
seven areas:
(1) the sensors, actuators, and valves currently existing in all kinds
of industrial control systems;
(2) the electronic hardware resident on the microprocessor chipset
system;
(3) the system interfaces including devices, Fieldbuses, and techniques used for all kinds of industrial control;
(4) the digital controllers performing the written programs and the
given protocols;
(5) the embedded software on a microprocessor chipset for real-time
control applications;
(6) the data-transmission hardware and protocols between independent agents or units of their own microprocessors;
(7) the routines, containing special hardware and software, which
are very useful to any kind of industrial control system.
All these seven areas are crucial for accomplishing both real-time control and distributed control in industry. This book, therefore, provides the
key technologies applied to modern industrial control so that it will be
widely available to all the engineers and researchers as well as students
who are working in industrial control and its relative disciplines.
Zhang_Prelims.indd xx
5/24/2008 9:42:08 AM
PREFACE
xxi
Readership
This book has been written primarily as an engineering handbook for
those engineers working in the research and development of all kinds of
control systems. However, the faculties and postgraduates in universities
or colleges will also find this book a useful technical reference for their
projects related to control and computer engineering. For university students, this book can be taken as a textbook in classes such as automation,
control, computer network, and other related technical subjects.
As an engineering handbook, this book will help professionals to design,
deploy, and make both manufacture control equipment and production
process control systems. Modern industrial control technologies involve
three essential phases: machinery, hardware, and software. However, no
matter what phase a control engineer is working with, he or she will find
that this book is very helpful.
As a reference, this book will aid the faculties and postgraduates in universities and colleges to understand all the technical details involved in
their research projects on controls. The wide coverage of this book allows
it to bridge the gap between theory and technique in control. In addition, it
is suitable for practicing postgraduates who wish or need to gain an engineering knowledge of the control topics.
This book is also intended to be a course textbook for students studying
the subjects of automatic control, computer hardware and electronics, computer network, as well as data communication. Typically, the students will
be in electronic engineering, computer control, control systems, or industrial automation courses.
Synopsis
This book has been organized into chapters, sections, and titled
graphs, etc.
The first of its seven chapters, “Sensors and Actuators for Industrial
Control,” lists the typical sensors, meters, actuators, and valves that are
crucial devices located between the front and the rear of industrial control
systems. This chapter provides the mechanism concepts, working principles, device types, technical data, and the guides to enable engineers to
design and develop industrial control systems.
The second chapter, “Computer Hardware for Industrial Control,”
provides a detailed list of the types of electronic devices resident on the
Zhang_Prelims.indd xxi
5/24/2008 9:42:08 AM
xxii
PREFACE
system given by a microprocessor chipset. These are the microprocessor,
programmable peripheral devices, and ASIC. The architecture of the electronic components on a computer motherboard is also plotted so that engineers are able to see how the microprocessor chipset is populated. This
chapter provides engineers with an explanation of how microprocessors
operate, and also all the necessary technical data for microprocessors to
perform.
The third chapter, “System Interfaces for Industrial Control,” discusses
four types of interfaces: actuator–sensor interface, control system interfaces, human–machine (or human–controller) interfaces, and highway
addressable remote transducer (HART) field interfaces. These four interfaces basically cover all the interface devices and technologies existing in
various industrial control systems. The actuator–sensor interface is located
at the front or rear of the actuator–sensor level to bridge the gap between
this level and the controllers. The control system interfaces include the
Fieldbus and microprocessor chipset interfaces that are used for connecting and communicating with controllers. The human–machine interfaces
contain both the tools and technologies to provide humans with easy and
comfortable methods of handling the devices. The HART field communications include the HART protocol and HART interface devices used for
field communications in industrial process control.
The fourth chapter is entitled “Digital Controllers for Industrial Control.”
A controller, similar to a computer, is a system with its own hardware and
software capable of performing independent control. This chapter lists the
controllers necessary for both industrial production control and industrial
process control: they are PLC controllers, CNC controllers, SCADA
system, PID controller, batch controllers, servo controllers, and the fuzzy
controllers.
The title of the fifth chapter is “Application Software for Industrial
Control.” The real-time control works with the microprocessor chipset
installed on a motherboard or a daughter board. Any microprocessor chipset, except for the inherent microcode and BIO to the CPU, must have a
software package consisting of three program systems: boot code, operating system, and application system. This chapter provides engineers with
the basic rationale, semantics, principles, work sequence, and program
structures for each of these three systems. With reference to this chapter,
an experienced software engineer should be capable of programming the
design of the whole software package of a microprocessor controller board.
The sixth chapter is “Data Communications in Distributed Control
System.” Several independent units, each of which will probably have
their own microprocessor to monitor a number of mechanical systems, are
Zhang_Prelims.indd xxii
5/24/2008 9:42:08 AM
PREFACE
xxiii
physically and electronically connected together. These communicate with
each other electronically in an interactive manner so as to form a distributed control system. In distributed control, to set up this type of industrial
control system the engineers must understand the connection methodologies and communication rationales between the independent units. This
chapter contains all the technical information, data, methodologies, and
some theories necessary for the implementation of a distributed control
system.
The seventh chapter is for the complex topics. These are the topics dealing with subjects that are over and above the basics when compared with
the first six chapters. Chapter 7 explains system routines that make the
control systems more user friendly and safe to operate. With the Power-up
and Power-down routines, the system is able to safely establish and terminate when power is switched ON and OFF, respectively. The installation
and configuration routines permit the system devices to communicate with
each other through both the software and the hardware. The diagnostic
routines then allow engineers to determine the root reasons when a system
suffers from a malfunction.
Bibliography
Writing this book has involved reference to a large number of sources
including academic books, journal articles, and in particular industry technical manuals and company introductory or demonstrational materials displayed on web sites of various dates and locations. The number and the
scale of the sources are such that it would be practically impossible to
acknowledge each source individually in the body of the book. The sources
are, therefore, alphabetically ordered and placed at the end of each chapter.
This method has two benefits. It enables the author to acknowledge the
contribution of other individuals and institutions whose scholarship or
products have been referred to in this book. It also provides the reader the
convenience of tracing more relevant sources.
Acknowledgments
My most sincere thanks go to my family: to my wife, Minghua and my
son Huza for their unwavering understanding and support. I would also
like to thank my younger brother Zhang wei for the ideas he contributed to
this book.
Zhang_Prelims.indd xxiii
5/24/2008 9:42:08 AM
xxiv
PREFACE
I am extremely grateful to Professor D. T. Pham, director of the
Manufacturing Engineering Center, Cardiff University, United Kingdom,
who granted me some very valuable suggestions in his review of some
portions of this manuscript, and to Dr. Chunquian Ji in Cardiff University
and to Dr. Wenbo Mao in Hewlett Packard Research Laboratory, United
Kingdom, for their valuable comments and suggestions on the manuscripts. Without the support of these people, there would not have been this
book.
Peng Zhang
Beijing
May 2008
Zhang_Prelims.indd xxiv
5/24/2008 9:42:08 AM
1
Sensors and Actuators for
Industrial Control
1.1 Sensors
1.1.1
Bimetallic Switch
Bimetallic switches are electromechanical thermal sensors or limiters
that are used for automatic temperature monitoring in industrial control.
They limit the temperature of machines or devices by opening up the
power load or electric circuit in the case of overheating or by shutting off
a ventilator or activating an alarm in the case of overcooling.
Bimetal switches can also serve as time-delay devices. The usual technique is to pass current through a heater coil that eventually (10 s or so)
warms the bimetal elements enough to actuate. This is the method employed
on some controllers such as cold-start fuel valves found on automobile
engines.
1.1.1.1
Operating Principle
A bimetallic switch essentially consists of two metal strips fixed together.
If the two metals have different expansibilities, then as the temperature of
the switch changes, one strip will expand more than the other, causing the
device to bend out of the plane. This mechanical bending can then be used
to actuate an electromechanical switch or be part of an electrical circuit
itself, so that contact of the bimetallic device to an electrode causes a circuit to be made. Figure 1.1 is a diagrammatic representation of the typical
operation of temperature switches. There are two directional processes
given in this diagram, which cause the contacts to change from open to
close and from close to open, respectively:
(1) Event starts; the time is zero, the temperature of switch is T1, and
the contacts are open. As environment temperature increases, the
switch is abruptly heated and reaches the temperature of T2 at
some moment causing the contacts to close.
(2) When the temperature of the switch is T2 and the contacts are
closed, if the environment temperature keeps decreasing, the
switch is abruptly cooled, which takes the temperature to T1 at
some instance causing the contacts to open again.
1
Zhang_Ch01.indd 1
5/13/2008 5:45:16 PM
2
INDUSTRIAL CONTROL TECHNOLOGY
Temperature
Contacts closed
T2
Differential
Decreasing temperature
Increasing temperature
Environment temperature
T1
Contacts open
0
Time
Figure 1.1 Typical operation of temperature switches.
1.1.1.2
Basic Types
Bimetallic switches and thermal controls basically fall into two broad
categories: (1) Creep action devices with slow make and slow break
switching action and (2) snap action devices with quick make and quick
break switching action.
Creep action devices are excellent in either a temperature-control application or as a high-limit control. They have a narrow temperature differential between opening and closing, and generally have more rapid cycling
characteristics than snap action devices.
Snap action devices are most often used for temperature-limiting applications, as their fairly wide differential between opening and closing temperature provides slower cycling characteristics.
Although in its simplest form a bimetallic switch can be constructed
from two flat pieces of metal, in practical terms a whole range of shapes is
used to provide maximum actuation or maximum force during thermal
cycling. As given in Fig. 1.2, the bimetallic elements can be of three configurations in a bimetallic switch:
(1) In Fig. 1.2(a), two metals make up the bimetallic strip (hence the
name). In this diagram, the black metal would be chosen to
expand faster than the white metal if the device were being used
in an oven, so that as the temperature rises the black metal
expands faster than the white metal. This causes the strip to bend
downward, separating from contact so that current is cut off. In a
Zhang_Ch01.indd 2
5/13/2008 5:45:17 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
3
Contact
Rivet
Bimetallic strip
(a)
Base
Set-point adjustment screw
Insulating connection spring
Switch contact
Bimetallic strip
(b)
Electrical connections
Base
Contact
(c)
Bimetallic disc
Figure 1.2 The operating principle for bimetallic switches: (a) basic bimetallic
switch, (b) adjustable set-point switch, and (c) bimetallic disc switch.
refrigerator you would use the opposite setup, so that as the temperature rises the white metal expands faster than the black metal.
This causes the strip to bend upward making contact so that current can flow. By adjusting the size of the gap between the strip
and the contact, you can control the temperature.
(2) Another configuration uses a bimetallic element as a plunger or
pushrod to force contacts open or closed. Here the bimetal does
not twist or deflect, but instead is designed to lengthen or travel
as a means of actuation as illustrated by Fig. 1.2(b). Bimetallic
switches can be designed to switch at a wide range of temperatures. The simplest devices have a single set-point temperature
determined by the geometry of the bimetal and switch packaging. Examples include switches found in consumer products.
Zhang_Ch01.indd 3
5/13/2008 5:45:17 PM
4
INDUSTRIAL CONTROL TECHNOLOGY
More sophisticated devices of industrial usages may incorporate
calibration mechanisms for adjusting temperature sensitivity or
switch-response times. These mechanisms typically set the separation between contacts as a means of changing the operating
parameters.
(3) Bimetal elements can also be disc shaped as in Fig. 1.2(c). These
types often incorporate a dimple as a means of producing a snap
action (not appropriately plotted in this figure). Disc configurations tend to handle shock and vibration better than cantilevered
bimetallic switches.
1.1.1.3 Application Guide
Bimetallic devices are generally specified for temperatures from –65°F
to several hundred degrees Fahrenheit. Specialized devices can handle
upward of 2000°F. Set-point tolerance and repeatability is generally on the
order of ±5°F, and set-point drift is usually negligible.
(1) Choosing the right thermal control. The rate of temperature rise,
location of the thermal control, the electrical load, and the mass
of the application can each greatly affect cycling (operational)
characteristics of a thermal control. Because of these variables, it
is strongly recommended that you conduct testing of the switches
specifically in your application. Certain aspects should be taken
into consideration when applying both creep and snap action
devices. Careful attention must be paid to input voltage, load
currents, and the characteristics of the load. Final design criteria
should be based upon results of the testing of our devices in your
application, at your facility.
(2) Choosing the right bimetallic switches. It has been realized that
each application for thermal controls is unique in one form or
another. Because of this, there is no standard product. A wide
range of options is offered, including the calibration temperature
range and tolerances. The length of the lead wires and the type of
insulation material also require deliberate consideration. You
should require samples for your application testing before deciding to use bimetallic switches.
(3) Snap action configurations. Snap action bimetal elements are
used in applications where an action is required at a threshold
temperature. As such, they are not temperature-measuring
devices, but rather temperature-activated devices. The typical
temperature change to activate a snap action device is several
degrees and is determined by the geometry of the device. When
Zhang_Ch01.indd 4
5/13/2008 5:45:17 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
5
the element activates, a connection is generally made or broken,
and a gap between the two contacts exists for a period of time.
For a mechanical system, there is no problem; however, for an
electrical system, the gap can result in a spark that can lead to
premature aging and corrosion of the device. Having the switch
activate quickly, hence the use of snap action devices, reduces
the amount and duration of spark. Snap action elements also
incorporate a certain amount of hysteresis into the system, which
is useful in applications that would otherwise result in an oscillation about the set point. It should be noted, however, that special
design of creep action bimetals can also lead to different ON/
OFF points, such as in the reverse lap-welded bimetal.
(4) Sensitivity and accuracy. Modern techniques are more useful
where sensitivity and accuracy are concerned for making a temperature measurement; however, bimetals find application in
industrial temperature control where an action is required without external connections. Evidently, geometry is important for
bimetal systems as the sensitivity is determined by the design,
and a mechanical advantage can be used to yield a large movement per degree temperature change.
1.1.1.4
Calibration
Temperature range calibration can be conducted with the following two
methods:
(1) The ice method. Immerse the temperature probe at least 2 in. into
a glass of finely crushed ice. Add cold tap water to remove air
pockets. Wait at least 1 min. The gauge should read 32°F. If it
does not, turn the adjustment nut on the back of the reading dial
with a pair of pliers until the dial reads 32°F. Wait at least 1 min
to verify correct adjustment.
(2) The boiling method. Submerge the probe into boiling water. Wait
until the needle stops moving, then adjust the calibration nut
until the dial reads 212°F. Since the boiling point of water
decreases as altitude increases, this method may not be as accurate as the ice method at altitudes above sea level unless the exact
boiling point temperature is known.
Calibration is a broad topic and includes the ultimate reference sources,
such as the national metrology laboratories, which are the custodians of
the International Temperature Scale, and those services that are directly
traceable to the national standards. For example, this is the scale that
Zhang_Ch01.indd 5
5/13/2008 5:45:17 PM
6
INDUSTRIAL CONTROL TECHNOLOGY
the national labs, or those affiliated to those labs, refer to in the calibration
certificates of reference devices that may be used in corporation or
university or other measurement laboratories that provide a more local
service, such as to working instruments in a process plant or experimental
apparatus.
1.1.2
Color Sensors
Color sensors that can operate in real time under various environmental
conditions can benefit many applications, including quality control, chemical sensing, food production, medical diagnostics, energy conservation,
and monitoring of hazardous waste, etc. Analogous applications can also
be found in other fields of economy, for example, in the electric industry
for recognition and assignment of colored cords, in the electronic industry
for the automatic test of mounted LED arrays or matrices, in the textile
industry to check coloring processes, or in the building materials industry
to control compounding processes.
These color sensors are generally advisable wherever color structures,
color processes, color nuances, or colored body rims must be recognized
in homogeneously continuous processes over a long period and have influence on the process control or quality protection as measuring or controlled variables.
1.1.2.1
Operating Principle
The color detection occurs at the color sensors according to the threefield procedure. Color sensors cast light on the objects to be tested, calculate the chromaticity coordinates from the reflected or transmitted radiation,
and compare them with previously stored reference tristimulus (red, green,
and blue) values. If the tristimulus values are within the set tolerance range,
a switching output is activated. Color sensors can detect both the color of
opaque objects through their reflections (incident light) and of transparent
materials in transmitted light, whereby a reflector is mounted opposite the
sensor.
In Fig. 1.3, the color sensor can sense eight colors: red, green, and blue
(primary colors); magenta, yellow, and cyan (secondary colors); and black
and white. The ASIC chipset of the color sensor is based on the fundamentals of optics and digital electronics. The object whose color is to be
detected should be placed in front of the system. The light rays reflected
from the object will fall on the three convex lenses that are fixed in front of
Zhang_Ch01.indd 6
5/13/2008 5:45:17 PM
7
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
Light
Color sensor
Blue
Green
Object
Red
LDR1
Magenta
LDR2
ASIC chipset
LDR3
Cyan
Yellow
Black
White
Figure 1.3 Operating principle of an assumed color sensor.
the three LDRs. The convex lenses are used to cause the incident light rays
to converge. Red, green, and blue glass filters are fixed in front of LDR1,
LDR2, and LDR3, respectively. When reflected light rays from the object
fall on the gadget, the filter glass plates determine which of these three
LDRs would get triggered. When a primary color light ray falls on the
system, the glass plate corresponding to that primary color will allow that
specific light to pass through. But the other two glass plates will not allow
any light to pass through. Thus, only one LDR will get triggered and the
gate output corresponding to that LDR will indicate which color it is.
Similarly, when a secondary color light ray falls on the system, the two
primary glass plates corresponding to the mixed color will allow that light
to pass through while the remaining one will not allow any light ray to pass
through it. As a result two of these three LDRs get triggered and the gate
output corresponding to these indicates which color it is. When all three
LDRs get triggered or none of them are triggered, you will observe white
and black light indications, respectively.
1.1.2.2
Basic Types
In accordance with the working processes and application features,
color sensors can be categorized into three-field color sensors and structured color sensors.
(1) Three-field color sensors. The sensor works based on the tristimulus (standard spectral) value function and identifies colors with
absolutely unerring precision and 10,000 times faster than the
human eye could. It provides a compact and dynamic technical
Zhang_Ch01.indd 7
5/13/2008 5:45:17 PM
8
INDUSTRIAL CONTROL TECHNOLOGY
solution for general color detection and color measurement jobs.
It is capable of detecting, analyzing, and measuring minute differences in color, for example, as part of LED testing, calibration of
monitors, or where mobile equipment is employed for color measurement. These color sensors can have these advantages:
(a) dielectric color filters adaptable for individual customers;
(b) MHz signal frequency for time critical measurements;
(c) profitable and fast signal processing;
(d) compact design—without optical beam guidance;
(e) high aging resistance;
(f) temperature stability and environmental resistance.
Based on the techniques used, there are two kinds of threefield color sensors:
(i) Three-element color sensor. This sensor includes special
filters so that their output currents are proportional to the
function of standard XYZ tristimulus values. The resulting absolute XYZ standard spectral values can thus be
used for further conversion into a randomly selectable
color space. This allows for a sufficiently broad range of
accuracies in color detection—from “eye accurate” to
“true color,” that is, standard-compliant colorimetry to
match the various application environments
(ii) Integral color sensor. This kind of sensor accommodates
integrative features including (1) detection of color
changes; (2) recognition of color labels; (3) sorting colored objects; (4) checking of color-sensitive production
processes; and (5) control of the product appearance.
(2) Structured color sensors. Structured color sensors are used for
the simultaneous recording of color and geometric information
including (1) determination of color edges or structures and
(2) checking of industrial mixed and separation processes. These
color sensors can have the following advantages:
(a) high selection of the sensor during applications in fast continuous processes of manufacture;
(b) high signal sequence and parallel data transfer;
(c) implementation of integral colorimetry;
(d) applications with high-measuring frequencies;
(e) adapted receiver geometries;
(f) specific color adaptation.
Based on the techniques used, there are two kinds of structured color sensors:
(i) Row color sensor. The row color sensor has been developed for detecting and controlling the color codes
and color sequences in the continuous measurement of
Zhang_Ch01.indd 8
5/13/2008 5:45:18 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
9
moving objects. These color sensors are designed as
PIN-photo-diode arrays. The photo diodes are arranged
in the form of honeycombs for each three rhombi. The
diode lines consist of two honeycomb lines displaced to
each other half-line. As a result, it is possible on least
expanse to implement a high resolution. In this case, the
detail to be controlled is determined by the choice of
suitable focusing optics.
(ii) Hexagonal color sensor. This kind of sensor generates
information for the subsequent electronics about the
three-field chrominance signal (intensity of the threereceiving segments covered with the spectral filters) as
well as about the structure and the position.
1.1.2.3 Application Guide
In industrial control, color sensors are selected by means of the applicationoriented principle. Although having many factors is a required consideration, the following technical items are of primary reference: (1) operating
voltage, (2) rated voltage, (3) output voltage, (4) residual ripple maximum,
(5) no-load current, (6) spectral sensitivity, (7) size of measuring dot minimum, (8) limiting frequency, (9) color temperature, (10) light spot minimum,
(11) permitted ambient temperature, (12) enclosure rating, (13) control
interface type, etc.
1.1.2.4
Calibration
Calibration is used to establish the link between the output signals (voltage, digital numbers) of the color sensor and their absolute physical values
at color sensor input, which describe the overall transfer function of the
color sensor. Calibration is a part of the color sensor characterization.
Different applications require calibrating different parameters. For example,
calibrating an ocean color sensor for remote sensing particularly concerns
with spectral and radiometric parameters if geometric calibrations are not
considered.
Many companies or organizations in the world provide color sensor
calibration services with their (1) software, (2) platform, and (3) evaluation boards. For example, SONY Corporation provides the ARTISANTM
COLOR REFERENCE SYSTEM; DMN Digital Inc. provides Pantone
ColorPlus Color Calibrator, etc.
Zhang_Ch01.indd 9
5/13/2008 5:45:18 PM
10
INDUSTRIAL CONTROL TECHNOLOGY
1.1.3
Ultrasonic Distance Sensors
There are numerous applications for ultrasonic distance sensors in industrial control. Ultrasonic distance sensors are used in all industries for measuring the distance to or size of material objects. That covers almost any
size and type of object that can be measured in most industrial sectors.
(1) Machine builders. Whether retrofitting existing machines or building new ones, ultrasonic distance sensors are used for motion
control, or level control, or dimensioning, or proximity sensing.
These are common applications in the converting, pulp and
paper, printing, rubber, metals, textile, and other manufacturing
industries.
(2) Automation. Ultrasonic distance sensors reduce automation costs
by providing a simple and effective means of monitoring the size
or position of objects in production processes. Sensor information is used to accept or reject objects based on size, position, or
fill level; make decisions on the routing of packages based on
size or position; control the flow of liquid, solid, or granular
materials; indicate when an object is nearby or in position, in/out
of tolerance, or provide alarms when objects are in/out of position, near full/empty, or indicate process completion.
(3) Process controls. Common applications include measuring the
level of bulk materials in a tank or bin (inventory) or controlling
the amount of material dispersed from a vessel (batching). Tank
levels can be locally displayed or reported to a remote computer
by a data network. Alarms can warn of low level, order levels,
high level, or other conditions.
1.1.3.1
Operating Principle
Ultrasonic distance sensors measure the distance or presence of target
objects by sending a pulsed ultrasound wave at the object and then measuring the time for the sound echo to return. Knowing the speed of sound,
the sensor can determine the distance of the object.
As displayed in Fig. 1.4, the ultrasonic distance sensor regularly emits a
barely audible click. It does this by briefly supplying a high voltage either
to a piezoelectric crystal or to magnetic fields of ferromagnetic materials.
In the first case, the crystal bends and sends out a sound wave. A timer
within the sensor keeps track of exactly how long it takes the sound wave
to bounce off something and return. This delay is then converted into a
voltage corresponding to the distance of the sensed object. In the second
case, the physical response of a ferromagnetic material in a magnetic field
Zhang_Ch01.indd 10
5/13/2008 5:45:18 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
11
Figure 1.4 Operating principle of ultrasonic distance sensor.
is due to the presence of magnetic moments. Interaction of an external
magnetic field with the domains causes a magnetostrictive effect. Controlling
the ordering of the domains through alloy selection, thermal annealing,
cold working, and magnetic field strength can optimize this effect. The
generated magnetostrictive effects are the use of magnetostrictive bars to
control high-frequency oscillators and to produce ultrasonic waves in
gases, liquids, and solids.
Applying converters based on the reversible piezoelectric effect makes
one-head systems possible where the converter serves both as transmitter
and as receiver. The transceivers work by transmitting a short burst ultrasonic packet. An internal clock starts simultaneously, measuring propagation time. The clock stops when the object reflects the sound packet back
to the sensor. The time elapsed between transmitting the packet and receiving the echo is the basis for calculating distance. Complete control of the
process is realized by an integrated microcontroller, which allows for
excellent output linearity.
1.1.3.2
Basic Types
The ultrasonic distance sensor can be operated in two different modes.
The first mode, referred to as continuous (or analog) mode, involves the
sensor continuously sending out sound waves at a rate determined by the
manufacturer. The second mode, called clock (or digital) mode, involves
the sensor sending out signals at a rate determined by the user. This rate
Zhang_Ch01.indd 11
5/13/2008 5:45:18 PM
12
INDUSTRIAL CONTROL TECHNOLOGY
can be several signals per second with the use of a timing device, or it can
be triggered intermittently by an event such as the press of a button.
1.1.3.3 Application Guide
The major benefit of ultrasonic distance sensors is their ability to measure difficult targets—solids, liquids, granulates, powders, and even transparent and highly reflective materials that cause problems for optical
sensors. In addition, analog output ultrasonic sensors offer comparatively
long ranges, in many cases >3 m. They can be made very small too; some
tubular models are only 12 mm in diameter, and 15 mm × 20 mm × 49 mm
square-bodied versions are available for limited space applications.
Ultrasonic devices do have some limitations. Foam and other attenuating surfaces may absorb most of the sound, significantly decreasing measuring range. Extremely rough surfaces may diffuse the sound excessively,
decreasing range and resolution. However, an optimal resolution is usually
guaranteed up to a surface roughness of 0.2 mm. Ultrasonic sensors emit a
wide sonic cone, limiting their usefulness for small target measurement
and increasing the chance of receiving feedback from interfering objects.
Some ultrasonic devices offer a sonic cone angle as narrow as 6°, permitting detection of much smaller objects and sensing of targets through narrow spaces such as bottlenecks, pipes, and ampoules.
For various distance-measuring sensors, there are four types of technical
definitions emphasized in their applications: (1) resolution, (2) linearity,
(3) repeat accuracy, and (4) reaction time. Figure 1.5 provides graphic
illustrations of these four technical definitions for measuring distance
sensors.
1.1.3.4
Calibration
Ultrasonic distance sensors in either continuous or clock mode can be
calibrated by means of some uncomplicated method. Once calibrated,
switching between modes will not affect calibration. To calibrate an ultrasonic distance sensor, a voltmeter is essential.
First, fasten the sensor to a surface such that there will be nothing but
the target in front of it, at the same time leaving plenty of room behind it
to adjust the small screws on the potentiometers by hand. Apply power to
the sensor to begin warming it up. Allow several minutes for warm-up
before starting the calibration process. The device should start clicking
once power is applied. Figure 1.6 shows location of pots on an ultrasonic
sensor.
Zhang_Ch01.indd 12
5/13/2008 5:45:18 PM
13
∇
Output
Output
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
Deviation
Distance
∇
Distance
Distance
(b)
Output
Distance
(a)
Output
∇
Response time
Time
Time
(c)
(d)
Figure 1.5 Technical definitions of distance-measuring sensors: (a) Resolution
corresponds to the smallest possible change in distance that causes a detectable
change in the output signal. (b) Linearity is the deviation from a proportional linear
function or a straight line, given as a percentage of the upper limit of the measuring
range (full scale). (c) Repeat accuracy is the difference between measured values
in successive measurements within a period of 8 h at an ambient temperature of
23 ± 5°C. (d) Reaction time is the time required by the sensor’s signal output to
rise from 10% to 90% of the maximum signal level. For sensors with digital signal
processing, it is the time required for calculation of a stable measured value.
Zero pot
Scale adjust pot
Full-scale pot
No connector
Analog output
Clock out
Trig-enable
Ext-trigger
Common
8–16 V DC
Gain
Figure 1.6 Location of pots on an ultrasonic sensor.
Zhang_Ch01.indd 13
5/13/2008 5:45:18 PM
14
INDUSTRIAL CONTROL TECHNOLOGY
The gain control must now be set to 50%. It is the potentiometer with
the screw head near the bottom of the device. It should rotate between
what would correspond to 8 and 4 on the face of a clock, but no further. Set
it to the “12” position
The positive lead of a voltmeter is put on pin 6 and the negative lead on
pin 2. Try to find a way to fasten these leads in place, or get someone to
hold them there. A screwdriver with a very small flat head is required to do
this. Rotate the zero and full-scale potentiometer screws fully counterclockwise, or about 12 turns. The screws will not stop turning after 12
turns, so you have to keep track.
Now place the target at the maximum distance from the sensor of interest. The literature claims a maximum distance of 10 ft, which we found to
be quite accurate. The sensor will work for objects at least 13 ft away, but
the objects must be very sound reflective and/or large (such as a refrigerator door) to obtain a usable reading.
Once the target is in place, adjust the scale adjust potentiometer. This
potentiometer compensates for the varying voltages that may be supplied
to the sensor. Rotate its screw until the voltmeter reads +5 V. Rotate the
full-scale potentiometer clockwise until its voltage ceases to change, and
then slowly rotate it counterclockwise until the +5 V reading is obtained
again.
Place the target at the minimum distance from the sensor. Do not put it
any closer than 6 in. Slowly rotate the zero adjust potentiometer until a
reading of 0 V is attained. We could not get a reading below 0.034 V, but
anything from +0.5 to –0.5 V is acceptable.
At this point it would be useful to move the target back and forth
between the minimum and maximum distances while watching the voltmeter. It should read +5 V when the target is at its maximum distance and
0 V when at the minimum. Make sure to keep the target within the sensor’s line of sight, which can be thought of as an imaginary cone emanating from the sensor to a width of 2.6 ft when standing 10 ft away. The
more you step outside the cone, the less likely you are to get a good
reading.
To adjust the gain, put the target at maximum distance. Rotate the gain
screw fully counterclockwise, and then slowly rotate it clockwise until
detection occurs. Rotate it an additional 1/16th turn after this. It is always
best to keep the gain setting as low as possible, since higher gain settings
increase the likelihood of false target detection.
Once all the above steps have been completed, the sensor should be calibrated and ready for detection.
Zhang_Ch01.indd 14
5/13/2008 5:45:19 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
1.1.4
15
Light Section Sensors
Light section methods have been utilized as a three-dimensional measurement technique for many years as noncontact geometry and contour
control. The light section sensor is primarily used for automating industrial production processes and testing procedures in which system-relevant
positioning parameters are generated from profile information. Twodimensional height information can be collected by means of laser light,
the light section method, and a high resolution of receiver array. Height
profiles can be monitored, filling level detected, magazines counted, and
the presence of objects checked.
1.1.4.1
Operating Principle
The light section can be simply achieved in many cases with the laser
light section method. The laser light section method is a three-dimensional
procedure to measure object profiles in one-sectional plane. The principle
of laser triangulation (see Fig. 1.7) requires a camera position orthogonal
to the object’s surface area to measure the lateral displacement or the
deformation of a laser line projected at an angle q onto the object’s surface
(see Fig. 1.8). The elevation profile of interest is calculated from the deviation of the laser line from the zero position.
A light section sensor consists of one camera and laser projector also
called laser line generator. The measurement principle of a light section
sensor is based on active triangulation (Fig. 1.7). Its simplest realization is
Camera
Laser projector
q
h
Figure 1.7 Laser triangulation (optical scheme), where h is elevation measurement range and q is the angle between the plane of the laser line and the axis of
the camera. A high resolution can be obtained by increasing h and decreasing q
and vice versa.
Zhang_Ch01.indd 15
5/13/2008 5:45:19 PM
16
INDUSTRIAL CONTROL TECHNOLOGY
A
Figure 1.8 Laser light sectioning is the two-dimensional extension of the laser
triangulation. By projecting the expanded laser line an elevation profile of the
object under test is obtained. Inset A: Image recorded by the area camera.
The displacement of the laser line indicates the object elevation at the point
of incidence.
scanning a scene by a laser beam and detecting the location of the reflected
beam. A laser beam can be spread by passing it through a cylindrical lens,
which results in a plane of light. Its profile can be measured in the camera
image thus realizing triangulation along one profile. In order to generate
dense range images, one has to project not only one but also many light
planes (Fig. 1.8). This can be achieved either by moving the projecting
device or by projecting many stripes at once. In the latter case the stripes
have to be encoded somehow; this is referred to as the coded-light approach.
The simplest encoding is achieved by assigning different brightness to every
projection direction, for example, by projecting a linear intensity ramp.
Measuring range and resolution are determined by the triangulation
angle between the plane of the laser line and the optical axis of the camera
(see Fig. 1.7). The more grazing this angle, the larger is the observed lateral displacement of the line. The measured resolution is increased, but the
measured elevation range is reduced. Criteria related to objects’ surface
characteristics, camera aperture or depth of focus, and width of the laser
line might reduce the achievable resolution.
1.1.4.2 Application Guide
In applications, the laser light section method has some technical details
requiring particular attention.
Zhang_Ch01.indd 16
5/13/2008 5:45:19 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
17
(1) Object surface characteristics. One requirement for the utilization
of the laser light section method is an at least partial diffuse reflecting surface. An ideal mirror would not reflect any laser radiation
into the camera lens, and the camera cannot view the actual
reflecting position of the laser light on the object surface. With a
complete diffuse reflecting surface, the angular distribution of
the reflected radiation is independent of the angle of incidence of
the incoming radiation as it hits the object under test (Fig. 1.8).
Real technical surfaces usually provide a mixture of diffuse
and reflecting behavior. The diffuse reflected radiation is not
distributed isotropic, which means that the more grazing the
incoming light, the lesser the radiation reflected in an orthogonal
direction to the object’s surface. Using the laser light section
method, the reflection characteristics of the object’s surface
(depending on the submitted laser power and sensitivity of the
camera) limit the achievable angle of triangulation q (Fig. 1.7).
(2) Depth of focus of the camera and lens. To ensure widely constant
signal amplitude on the sensor, the depth of focus of the camera
lens as well as the depth of focus of the laser line generator have
to cover the complete measurement elevation range.
By imaging the object under test onto the camera sensor the
depth of focus of the imaging lens increases proportional to the
aperture number k, the pixel distance y, quadratic to the imaging
factor @ (= field of view/sensor size).
The depth of focus 2z is calculated by
2z = 2yk@ (1+@).
In the range ±z around the optimum object distance, no reduction in sharpness of the image is evident.
Example:
Pixel distance y = 0.010 mm
Aperture number k = 8
Imaging factor @ = 3
2z = 2 × 0.010 × 8 × 3 × (1 + 3) = 1.92 mm
With fixed imaging geometry a fading aperture of the lens
increases its depth of focus. A larger aperture number k cuts the
signal amplitude by a factor of 2 with each aperture step; it
decreases the optical resolution of the lens and increases the negative influence of the speckle effect.
(3) Depth of focus of a laser line. The laser line is focused to a fixed
working distance. With actual working distances diverging from
the setting the laser line widens and the power density of the
radiation decreases.
The region around the nominal working distance, where line
width does not increase by more than a given factor, is according
Zhang_Ch01.indd 17
5/13/2008 5:45:19 PM
18
INDUSTRIAL CONTROL TECHNOLOGY
to agreement characterized as the depth of focus of a laser line.
There are two types of laser line generators: laser micro line generators and laser macro line generators.
Laser micro line generators create thin laser lines with
Gaussian intensity profile orthogonal to the laser line. The depth
of focus of a laser line at wavelength L and of width B is
given by the so-called Rayleigh range 2ZR:2ZR = (pB2)/(2L),
where p = 3.1415926.
Laser macro line generators create laser lines with increased
depth of focus. At the same working distance macro laser lines
are wider than micro laser lines.
Within the two design types, macro and micro line generators,
the respective line width is proportional to the working distance.
Due to the theoretical connection between line width and depth
of focus, the minimum line width of the laser line is limited by
the application due to the required depth of focus.
(4) Basic setback: Laser speckle. Laser speckling is an interference
phenomenon originating from the coherence of the laser radiation,
for example, the laser radiation reflected by a rough-textured
surface.
Laser speckle disturbs the edge sharpness and the homogeneity
of the laser lines. Orthogonal to the laser line, the center of intensity is displaced stochastically. The granularity of the speckle
depends on the setting of the lens aperture viewing the object.
With a small aperture number the arising speckles have a high
spatial frequency; with a large k number the speckles are rather
rough and particularly disturbing. Because a diffuse reflective
and thus an optically rough-textured surface is essential for the
utilization of the laser light section method, laser speckling cannot be avoided in principle.
A reduction of the disturbing effect is possible by (1) utilizing
laser beam sources with decreased coherence length, (2) a relative movement between object and sensor, possibly using a necessary or existent movement of the sensor or the object (e.g., the
profile measurement of railroad tracks while the train is running),
(3) diminishing the speckle pattern by choosing large lens apertures (small aperture numbers), as long as the requirements of
depth of focus tolerate this.
(5) Dome illuminator for diffuse illumination. The introduced application requires simultaneously with the three-dimensional profile measurement, control of the object outline, and surface.
For this purpose, the object under test is illuminated homogeneously and is diffused by a dome illuminator. An LED ring lamp
generates the illumination that propagates, scattered by a diffuse
Zhang_Ch01.indd 18
5/13/2008 5:45:19 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
19
reflecting cupola, to the object of interest. In the center of the
dome an opening for the camera is located. There is no radiation
falling onto the object from this direction.
Shadow and glint are widely avoided. Because the circumstances
correspond approximately to the illumination on a cloudy day, this
kind of illumination is also called “cloudy day illumination.”
(6) Optical engineering. Using a laser light section application with
high requirements, the design of the system configuration is of
great importance. This “optical engineering” implies the choice
and the contractive design of the utilized components like camera, lens, and laser line generator from the optical point of view.
Considering the optical laws and their interactions, an optimum
picture recording within the given physical boundary conditions
is accomplished. Elaborate picture preprocessing algorithms are
avoided.
Arranging first steps to measure objects with largely diffuse
reflecting surfaces or with reduced requirements in resolution,
cameras and laser line generators from the electronic mail order
catalog may be utilized for first system testing (e.g., school
practicum, etc).
These simple laser line generators utilize mostly a glass rod
lens to produce a Gaussian intensity profile along the laser line
(as mentioned in the Operating Principle section). With increased
requirements, laser lines with largely constant intensity distribution and line width have to be utilized.
1.1.4.3
Specifications
The specifications of the light section sensors are routinely documented
with these technical data:
(1) supply voltage which is the voltage of the direct current power
supplied to the sensor;
(2) voltage ripple which defines the maximum tolerance for the
ratio of the maximum voltage bias to the supply voltage;
(3) reverse polarity protected which indicates whether or not the
sensor has functionality protected from reverse polarity;
(4) short circuit protected which indicates whether or not the sensor can be protected from damage if short circuit occurs;
(5) power consumption which represents the power consumption
of the sensor;
(6) maximum output load which is the output power value of the
sensor;
Zhang_Ch01.indd 19
5/13/2008 5:45:20 PM
20
INDUSTRIAL CONTROL TECHNOLOGY
(7) maximum operation frequency which is the permitted working
frequency of the sensor;
(8) response time tON/tOFF , where tON is the time interval from when
the sensor is turned on to when it is ready to be loaded for tasks
and tOFF is the time interval from when the sensor is turned off
to when it completely stops;
(9) hysteresis which is the sensing delay of the sensor;
(10) length of light line which is maximum working laser length, etc.
In addition to these technical data, the specifications of the light section
sensors normally also include some environmental data as below:
(1) vibration which is the allowed environmental vibration;
(2) shock which is the allowable energy of exterior air shocks;
(3) operation temperature that is the tolerant environmental temperature for the sensors, etc.
1.1.4.4
Calibration
In the process of calibration, camera and projector parameters are estimated simultaneously using the nonlinear least squares estimation model.
The three-dimensional coordinates of checkpoints are introduced as additional unknowns, and thus they are estimated simultaneously with the model
parameters. This estimation model was chosen because its output contains
not only the estimated parameters but also their accuracy. Obtaining the
checkpoint accuracy is the major topic of this work.
The estimation of camera and projector parameters requires a threedimensional calibration standard with well-defined target points. These
control points are characterized by their three-dimensional coordinates, the
two-dimensional coordinates of their images, and their one-dimensional
projector coordinates. Given n control points as input, the maximum likelihood estimation of unknown parameters can be derived by the general
nonlinear least squares estimation.
Here are some important requirements for the calibration standard popularly used in industrial control.
(1) The control points have to be regularly distributed in threedimensional workspace.
(2) The images of the control points have to have a good contrast
and an extent of at least 10 pixels. In this case they can be measured very accurately using least squares template matches.
Zhang_Ch01.indd 20
5/13/2008 5:45:20 PM
21
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
(3) Highly accurate three-dimensional coordinates of control points
as well as their errors have to be available.
For example, assume an aluminum plate with white squares printed on
a black background is used as a calibration standard. To achieve a regular
distribution of control points in work space, the calibration plate can be
precisely displaced along the third axis. Figure 1.9 shows the section sensor setup as it was used in this work. Both camera and projector are inclined
to facilitate the measurement of not only horizontal, but also vertical,
and even of overhanging surfaces for our application of grasping threedimensional objects. The position of the RSP projected onto the XY plane
of the world coordinate system is also given in Fig. 1.9(b).
From Fig. 1.9(a) it should be clear that the angle q between the optical
axes of the camera and the projector was chosen to be relatively small; it
is approximately 15o. This choice has the advantage of making it possible
to measure surfaces of larger orientation range while the achievable accuracy is not ideal as it would be theoretically with q. = 90o. Notice that the
optical axis of the section sensor ar is defined as the symmetry axis of ac
and ap in Fig. 1.9(a). The work space in this example (200 × 200 × 100 mm3)
is mainly constrained by the depth of focus of the projector and the camera, which is about 200 mm at a distance of 1300 mm.
Weights of observations depend on the measurement process. For light
section sensor calibration, three types of observations are used: (1) a priori
knowledge about the three-dimensional coordinates of white squares
on the calibration plate, (2) image measurements of square centers, and
(3) measurements of the projector coordinates of square centers.
(a)
(b)
Y
Projector
Work
space
RSP
Camera
ap
ac
Z
ai
X
1300 mm
q
RSP
200 mm
Z
100 mm
Work space
XY plane
200 mm
Figure 1.9 Experimental range sensor setup: (a) relative position of camera,
projector, and work space in the range sensor plane (RSP); (b) position of the
RSP and the work space projected onto the XY–Z plane of the world coordinate
system.
Zhang_Ch01.indd 21
5/13/2008 5:45:20 PM
22
INDUSTRIAL CONTROL TECHNOLOGY
1.1.5 Linear and Rotary Variable Differential
Transformers
The linear variable differential transformer (LVDT) is a well-established
transducer design which has been used throughout many decades for the
accurate measurement of displacement and within closed loops for the
control of positioning. The LVDT design lends itself for easy modification
to fulfill a whole range of different applications in both research and
industry:
(1) pressurized versions for hydraulic cylinder applications;
(2) materials suitable for sea water and marine services;
(3) dimensions to suit specific application requirements;
(4) multichannel, rack amplifier–based systems;
(5) automotive suspension system.
The rotational variable differential transformer (RVDT) is also a wellestablished transducer design used to measure rotational angles and operates under the same principles as the LVDT sensor. Whereas the LVDT
uses a cylindrical iron core, the RVDT uses a rotary ferromagnetic core.
Some of the typical applications of RVDT are
(1) flight control/navigation
(2) flap actuators
(3) fuel control
(4) cockpit control
(5) automation assembly equipment.
1.1.5.1
Operating Principle
(1) An LVDT is much like any other transformer in that it consists
of a primary coil, secondary coils, and a magnetic core as illustrated in Fig. 1.10(a). The primary coils (the upper coil in
Fig. 1.10(a)) are energized with constant amplitude alternating
current. This produces an alternating magnetic field in the center
of the transducer which induces a signal into the secondary coils
(the two lower coils in Fig. 1.10(a)) depending on the position of
the core.
Movement of the core within this area causes the secondary
signal to change (Fig. 1.10(b)). As the two secondary coils are
positioned and connected in a set arrangement (push–pull mode),
when the core is positioned at the center, a zero signal is derived.
Zhang_Ch01.indd 22
5/13/2008 5:45:20 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
23
Primary coil
Movable core
(a)
Secondary coil
Secondary #1
Secondary coil
Primary
Secondary #2
Lead wires
Displacement
(b)
Moveable core
Figure 1.10 The working principles of an LVDT: (a) the three coils and the
movable core and (b) the displacement system of a sensor.
Movement of the core from this point in either direction causes
the signal to increase. As the coils are wound in a particular precise manner, the signal output has a linear relationship with the
actual mechanical movement of the core.
The secondary output signal is then processed by a phasesensitive demodulator which is switched at the same frequency as
the primary energizing supply. This results in a final output which,
after rectification and filtering, gives direct current output proportional to the core movement and also indicates its direction, positive or negative, from the central zero point (Fig. 1.10(b)).
As with any transformer, the voltage of the induced signal in
the secondary coil is linearly related to the number of coils. The
basic transformer relation is
Vout/Vin = Nout/Nin
where Vout is the voltage at the output, Vin is the voltage at the
input, Nout is the number of windings of the output coil, and Nin
is the number of windings of the input coil.
The distinct advantage of using an LVDT displacement transducer is that the moving core does not make contact with other
electrical components of the assembly, as with resistive types,
and so offers high reliability and long life. Further, the core can
Zhang_Ch01.indd 23
5/13/2008 5:45:20 PM
24
INDUSTRIAL CONTROL TECHNOLOGY
be so aligned that an air gap exists around it, which is ideal for
applications where minimum mechanical friction is required.
(2) An RVDT is an electromechanical transducer that provides a
variable alternating current output voltage. This output voltage is
linearly proportional to the angular displacement of its input
shaft. When energized with a fixed alternating current source, the
output signal is linear within a specified range over the angular
displacement. RVDT utilizes brushless, noncontacting technology
to ensure long life, reliability, and repeatable position sensing
with infinite resolution. Such reliable and repeatable performance ensures accurate position sensing under the most extreme
operating conditions.
As diagrammed in Fig. 1.11, rotating a ferromagnetic-core
bearing supported within a housed stator assembly provides a
basic RVDT construction and operation. The housing is passively stainless steel. The stator consists of a primary excitation
coil and a pair of secondary output coils. A fixed alternating current excitation is applied to the primary stator coil that is electromagnetically coupled to the secondary coils. This coupling is
proportional to the angle of the input shaft. The output pair is
structured so that one coil is in phase with the excitation coil, and
the second is 180° out of phase with the excitation coil. When the
rotor is in a position that directs the available flux equally in both
the in phase and out of phase coils, the output voltages cancel
Rotary
ferromagnetic
core
Vin
Vout
Figure 1.11 The working principles of a typical RVDT.
Zhang_Ch01.indd 24
5/13/2008 5:45:20 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
25
and result in a zero value signal. This is referred to as the electrical zero position. When the rotor shaft is displaced from the electrical zero position, the resulting output signals have a magnitude
and phase relationship proportional to the direction of rotation.
Because the performance of an RVDT is essentially similar
to that of a transformer, excitation voltage changes will cause
directly proportional changes to the output (transformation ratio).
However, the voltage out to excitation voltage ratio will remain
constant. Since most RVDT signal conditioning systems measure signal as a function of the transformation ratio, excitation
voltage drift beyond 7.5% typically has no effect on sensor accuracy and strict voltage regulation is not typically necessary.
Excitation frequency should be controlled within ±1% to maintain accuracy.
1.1.5.2 Application Guide
The following factors should be considered when selecting an LVDT
(AC, alternating current; DC, direct current):
(1) measurement range
(2) armature type
(3) AC–AC vs DC–DC
(4) environment.
LVDTs are available with ranges from ±0.01" to ±18.5". An LVDT with
a ±18.5" range can be used in one direction to measure up to 37". If accuracy
is important, the range selected should not be any larger than necessary.
Three armature types are available: free unguided armatures, captive
guided spring return armatures, and captive guided armatures. Free unguided
armatures are recommended for applications in which the target being
measured moves parallel to the transducer body as well as those which
require frequent or continuous measurements. This armature type is well
suited for dynamic applications. When using a free unguided armature, the
armature and the LVDT body must be mounted so that their correct relative positions are maintained. This type of LVDT features an armature/
threaded push rod assembly which is completely separable from the LVDT
body. Since the free unguided armature involves no mechanical coupling
between the armature and the LVDT body, there are no springs or bearings
to fatigue. This unit has a virtually unlimited fatigue life. Captive guided
spring return armatures are well suited for those applications requiring the
measurement of multiple targets or applications in which the target moves
Zhang_Ch01.indd 25
5/13/2008 5:45:21 PM
26
INDUSTRIAL CONTROL TECHNOLOGY
transverse to the armature and changes in the structure’s surface are to be
measured. In this type of LVDT, the armature moves over bearings in the
LVDT body. The armature is biased by an internal spring so that the ballended probe bears against the surface of the target whose displacement is
being measured. The LVDT is held in position by clamping the body alone.
The armature is not attached to the target being measured. Captive guided
armatures are designed for applications requiring a longer working range.
The armature moves freely over machined bearings but cannot be removed
from the body. The LVDT body has a threaded mounting hole, and the
armature is attached to the structure being measured. The armature end is
threaded so that special adapters such as spherical bearings or rollers can
be attached.
The major advantages of DC–DC LVDTs are the ease of installation,
the ability to operate from dry cell batteries in remote locations, and lower
system cost, while the advantages of AC–AC LVDT include greater accuracy and a smaller body size. An AC–AC LVDT can be equipped with
more sophisticated electronics such as SENSOTEC SC instrumentation.
The SC instrument provides an AC power supply, a phase-sensitive demodulator, a scaling amplifier, and a DC output. The AC–AC LVDT system
has less residual noise at minimum readings than DC–DC units which
utilize internal electronics.
For applications involving very high humidity or requiring submersion
of the LVDT, a submersible LVDT is required. Submersible units are available for either AC–AC or DC–DC operation and with free unguided or
captive spring return armatures. The unit selected should also operate and
survive at the temperatures dictated by the application. Note that AC–AC
units will operate at higher temperatures (up to 257°F) than the DC–DC
units (up to 158°F) which have internal electronics. Side loads must be
kept to a minimum since they will cause rubbing between the armature and
the LVDT body. This friction will cause excessive wear of bearings, and in
extreme cases the armature may bend. At a minimum, side loads will
reduce the unit’s life and accuracy.
1.1.5.3
Calibration
The manual calibration procedure for both the LVDT sensors and the
RVDT sensors is to check and adjust the zero and gain settings of the signal conditioner, which includes these steps: (1) First, the signal conditioner
at zero displacement should be made to be zero. (2) Second, the micrometer should be traversed to a maximum displacement that can be anticipated in the calibration experiment. (3) The gain setting should be adjusted
Zhang_Ch01.indd 26
5/13/2008 5:45:21 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
27
to attain a transducer “high” output value. (4) Finally, the “reference values” of the zero, maximum displacement, and the corresponding transducer output must be recorded.
In further experiments, it may be necessary to set the zero and gain of the
signal conditioner to these “reference values” and use the calibration results
or curves obtained from this laboratory for displacement measurements.
The laboratory assignments are executed with the following procedure:
(1) Take at least five different settings to cover the displacement
range of 0–1 in. For each displacement setting, repeat five times
for increasing displacement (or, low to high) and repeat five
times for decreasing displacement (or, high to low) to check for
hysteresis.
(2) Establish a spread sheet and calculate the average voltage for
each displacement.
(3) Plot the averaged voltage output of the signal conditioner vs displacement and obtain a linear correlation between the averaged
voltage output vs displacement using a least squares technique
(e.g., in the format of y = a + bx ± c).
1.1.6
Magnetic Control Systems
Magnetic control systems are dominated with magnetic field sensors,
magnetic switches, and instruments that measure magnetic fields and or
magnetic flux by evaluating a potential, current, or resistance change due
to the field strength and direction. They are used to study the magnetic
field or flux around the Earth, permanent magnets, coils, and electrical
devices in displacement measurement, rail inspection system, and the linear potentiometer replacement, etc. Magnetic field sensors and switches
can measure and control these properties without physical contact and
have become the eyes of many industrial and navigation control systems.
Magnetic field sensors and switches are typically applied in industrial control for proximity detection, displacement sensing, rotational reference
detection, current sensing, and vehicle detection.
Magnetic field sensors indirectly measure properties such as direction,
position, rotation, angle, and current by detecting the magnetic field and its
changes. The first application of a permanent magnet was a third-century
bc Chinese compass, which is a direction sensor. Compared to other direct
methods such as optical or mechanical sensor, most magnetic file sensors
require some signal processing to get the property of interest. However,
they provide reliable data without physical contact even in adverse conditions such as dirt, vibration, moisture, hazardous gas and oil, etc. At present,
Zhang_Ch01.indd 27
5/13/2008 5:45:21 PM
28
INDUSTRIAL CONTROL TECHNOLOGY
there are two kinds of magnetic switches: magnetic reed switches and
magnetic level switches. Magnetic switches are suitable for applications
requiring a switched output for proximity, linear limit detection, logging
or counting, or actuation purposes.
Flux gate and coil instruments perform a continuous measurement of
the differences in the magnetic field at the ends of a vertical rod and plot
these differences on a grid of the area. Hall effect describes a device that
converts the energy stored in a magnetic field to an electrical signal by
means of the development of a voltage between the two edges of a current
carrying conductor whose faces are perpendicular to a magnetic field.
1.1.6.1
Operating Principle
A unique aspect of using magnetic sensors and switches is that measuring magnetic fields is usually not the primary intent. Another parameter is
usually desired such as wheel speed, presence of a magnetic ink, vehicle
detection, or heading determination, etc. These parameters cannot be measured directly but can be extracted from changes or disturbances in magnetic fields. The most widely used magnetic sensors and switches are Hall
effect sensor switches, magnetoresistive (MR) series of sensors switches,
magnetic reed switch, magnetic level switch, etc.
(1) Hall effect sensors and switches. The Hall effect is a conduction
phenomenon which is different for different charge carriers. In
most common electrical applications, the conventional current is
used partly because it makes no difference whether positive or
negative charge is considered to be moving. But the Hall voltage
has a different polarity for positive and negative charge carriers,
and it has been used to study the details of conduction in semiconductors and other materials which show a combination of
negative and positive charge carriers.
The Hall effect can be used to measure the average drift velocity of the charge carriers by mechanically moving the Hall probe
at different speeds until the Hall voltage disappears, showing
that the charge carriers are now not moving with respect to the
magnetic field. Other types of investigations of carrier behavior
are studied in the quantum Hall effect. An alternative application
of the Hall effect is that it can be used to measure magnetic fields
with a Hall probe.
As shown in Fig. 1.12, if an electric current flows through a
conductor in a magnetic field, the magnetic field exerts a transverse force on the moving charge carriers, which tends to push
Zhang_Ch01.indd 28
5/13/2008 5:45:21 PM
29
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
VH
Fm = Magenetic
force on
negative charge
carriers
I
Magenetic
field B
+
−
Fe
−
−
−
d
I
+
− F
m
−
+
+
+
+
Fe = Electric force
from charge
buildup
Direction of conventional
electric current
Figure 1.12 The Hall effect.
them to one side of the conductor. This is most evident in a thin,
flat conductor as illustrated. A build up of charge at the sides of
the conductors will balance this magnetic influence, producing a
measurable voltage between the two sides of the conductor. The
presence of this measurable transverse voltage is called the Hall
effect, after E. H. Hall who discovered it in 1879.
Note that the direction of the current I in the diagram is that of
conventional current, so that the motion of electrons is in the
opposite direction. This further confuses all the “right-hand rule”
manipulations you have to go through to get the direction of the
forces.
As displayed in Fig. 1.12, the Hall voltage VH is given by VH =
IB/(ned), where I is the induced electric current, B is the strength
Input signal
Sensor
Magnetic
flux
HAL
Out signal
VOUT
Figure 1.13 The functional principle of a Hall sensor.
Zhang_Ch01.indd 29
5/13/2008 5:45:21 PM
30
INDUSTRIAL CONTROL TECHNOLOGY
of the magnetic filed, n is the density of mobile charges, e is the
electron charge, and d is the thickness of the film.
In the Hall sensor, the Hall element with its entire evaluation
circuitry is integrated on a single silicon chip. The Hall plate
with the current terminals and the taps for the Hall voltage are
arranged on the surface of the crystal. This sensor element detects
the components of the magnetic flux perpendicular to the surface
of the chip and emits a proportional electrical signal which is
processed in the evaluation circuits integrated on the sensor chip.
The functional principle of a Hall sensor is, as displayed in
Fig. 1.13, that the output voltage of the sensor and the switching
state, respectively, depend on the magnetic flux density through
the Hall plate.
(2) Magnetoresistive sensors and switches. Magnetoresistance is the
property of some conductive materials to gain or lose some of their
electrical resistance when placed inside a magnetic field. The resistivity of some materials is greatly affected when the material is
subjected to a magnetic field. The magnitude of this effect is known
as magnetoresistance and can be expressed by the equation:
MR = (r(H)–r(0))/r(0)
where MR is the magnetoresistance, r(0) is the resistivity at zero
magnetic field, and r(H) is the resistivity in an applied magnetic
field.
The magnetoresistance of conventional materials is quite small,
but materials with large magnetoresistance have been synthesized now. Depending on the magnitude, it is either called giant
magnetoresistance (GMR) or colossal magnetoresistance (CMR).
Magnetoresistive sensor or switch elements are magnetically
controllable resistors. The effect whereby the electric resistance
of a thin, anisotropic ferromagnetic layer changes through a
magnetic field is utilized in these elements. The determining factor for the specific resistance is the angle formed by the internal
direction of magnetization (M) and the direction of the current
flow (I). Resistance is largest if the current flow (I) and the direction of magnetization run parallel. The resistance in the base
material is smallest at an angle of 90° between the current flow
(I) and the direction of magnetization (M).
In addition, highly conductive material is applied below an
angle of 45°. The current passing the sensor element takes the
shortest distance between these two ranges. This means that it
flows at a preferred direction of 45° against the longitudinal axis
of the sensor element. Without an external field, the resistance of
the element is then in the medium range. An external magnetic
Zhang_Ch01.indd 30
5/13/2008 5:45:23 PM
31
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
field with field strength (H) influences the internal direction of
magnetization, which causes the resistance to change as a factor
of the influence. Figure 1.14 gives an example using Permalloy
(NiFe) film to illustrate that the resistance of a material depends
upon the angle between the internal direction of magnetization
M and the direction of the current flow I.
The actual sensor element is often designed with four magnetic field sensitive resistors interconnected to form a measuring
bridge (Fig. 1.15). The measuring bridge is energized and
Current I
Magnetization M
q
Easy axis
No applied field
I
M
HApplied
Figure 1.14 As an example, the Permalloy (NiFe) film has a magnetization
vector, M, that is influenced by the applied magnetic field being measured. The
resistance of the film changes as a function of the angle between the vector M
and current flow, I, flowing through it. This change in resistance is known as the
magnetoresistive effect.
Bias
current
Permalloy
Out−
Shorting bars
Vb
GND
Easy axis
Out+
Sensitive
axis
Figure 1.15 As an example, the magnetoresistive bridge is made up of four
Permalloy parallel strips. A crosshatch pattern of metal is overlaid onto the strips
to form shorting bars. The current then flows through the Permalloy, taking the
shortest path, at a 45° angle from shorting bar to shorting bar. This establishes
the bias angle between the magnetization vector M of the film and the current I
flowing through it.
Zhang_Ch01.indd 31
5/13/2008 5:45:23 PM
32
INDUSTRIAL CONTROL TECHNOLOGY
supplies a bridge voltage. A magnetic field which influences
the bridge branches in different degrees leads to a voltage difference between the bridge branches which is then amplified and
evaluated.
The sensor detects the movement of ferromagnetic structures
(e.g., in gearwheels) caused by the changes in the magnetic flow.
The sensor element is biased with a permanent magnet. A tooth
or a gap moving past the sensor influences the magnetic field at
different degrees. This causes changes in the magnetic field
dependent on resistance values in a magnetoresistive sensor. The
changes in the magnetic field can therefore be converted into an
electric variable and can also be conditioned accordingly. The
output signal from the sensor is a square wave voltage which
reflects the changes in the magnetic field.
Changes in the magnetic field cause the bridge voltage to be
deflected. This voltage is amplified and supplied to a Schmitt
trigger after conditioning. If the effective signal reaches an adequate level, the output stage is set accordingly.
The sensor is used for the noncontact rotational speed detection on ferromagnetic sensing objects such as gearwheels. The
distance between the sensed object and the surface of the active
sensor is described as air gap. The maximum air gap is dependent on the geometry of the object. The measurement principle
dictates the direction-dependent installation. The magnetoresistive sensor is sensitive to changes in the external magnetic field.
For this reason the sensed objects should not have different
degrees of magnetization.
(3) Magnetic switches. Most magnetic switches actually work with
two mechanisms: the magnetic reed switches and the magnetic
level switches.
Magnetic reed switches normally consist of two overlapping
flat contacts which are sealed into a glass tube filled with inert
gas. When approached by a permanent magnet the contact ends
attract each other and make contact. When the magnet is removed,
the contacts separate immediately (Fig. 1.16).
For magnetic level switches, the operation is achieved using
the time-proven principle of repelling magnetic forces. One permanent magnet forms part of a float assembly which rises and
falls with changing liquid level. A second permanent magnet is
positioned within the switch head so that the adjacent poles of
the two magnets repel each other through a nonmagnetic diaphragm. A change in liquid level moves the float through its permissible range of travel causing the float magnet to pivot and
repel the switch magnet. The resulting snap action of the repelling magnets actuates the switch.
Zhang_Ch01.indd 32
5/13/2008 5:45:24 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
Flat contacts
33
Magnet
Glass tube
Figure 1.16 Operating principle of a simple magnetic reed switch.
1.1.6.2
Basic Types and Application Guide
One way to classify the various magnetic sensors and switches is by the
magnetic field sensing range. These sensors and switches can be arbitrarily
divided into three categories: (1) low field, (2) medium field, and (3) high
field sensing. Sensors that detect magnetic fields less than 1 µG will be
classed as low field sensors. Sensors with a range of 1 µG to 10 G will be
considered Earth’s field sensors, and those sensors that detect fields above
10 G will be considered bias magnet field sensors in this book. Only those
magnetic field sensors and switches used in industrial control are listed
below so that the low field magnetic sensors are neglected because these
are primarily used for medical applications and laboratory research.
(1) Earth’s field sensors. (1 µG to 10 G) The magnetic range for the
medium field sensors lends itself well to using the Earth’s magnetic field. Several ways to use the Earth’s field are to determine
compass headings for navigation, detect anomalies in it for vehicle detection, and measure the derivative of the change in field to
determine yaw rate.
(a) Fluxgate. Fluxgate magnetometers are the most widely used
sensors for compass navigation systems. Fluxgate sensors
have also been used for geophysical prospecting and airborne magnetic field mapping. The most common type of
fluxgate magnetometer is called the second harmonic device.
This device involves two coils, a primary and a secondary,
wrapped around a high-ability ferromagnetic core. The magnetic induction of this core changes in the presence of an
external magnetic field. Another way of looking at the fluxgate operating principle is to sense the ease, or resistance, of
saturating the core caused by the change in its magnetic flux.
The difference is due to the external magnetic field.
Zhang_Ch01.indd 33
5/13/2008 5:45:24 PM
34
INDUSTRIAL CONTROL TECHNOLOGY
(b) Magnetoinductive. Magnetoinductive magnetometers are
relatively new with the first patent issued in 1989. The sensor
is simply a single winding coil on a ferromagnetic core that
changes permeability within the Earth’s field. The sense coil
is the inductance element in an L/R relaxation oscillator. The
frequency of the oscillator is proportional to the field being
measured. A static DC current is used to bias the coil in a
linear region of operation.
(c) Anisotropic magnetoresistive (AMR). Magnetoresistive (MR)
sensors come in a variety of shapes and forms. The newest
market growth for MR sensors is high density read heads for
tape and disk drives. Most AMR sensors are made of
Permalloy (NiFe) thin film deposited onto a silicon substrate
and patterned to form a Wheatstone resistor bridge.
AMR sensors provide an excellent means of measuring both
linear and angular position and displacement in the Earth’s magnetic field. Permalloy thin films of a sensor deposited on a silicon
substrate in various resistor bridge configurations provide highly
predictable outputs when subjected to magnetic fields. Low cost,
high sensitivity, small size, noise immunity, and reliability are
advantages over mechanical or other electrical alternatives.
Highly adaptable and easy to assemble, these sensors solve a
variety of problems in custom applications.
(2) Bias magnet field sensors. (above 10 G) Most industrial sensors
use permanent magnets as a source of the detected magnetic field.
These permanent magnets magnetize, or bias, ferromagnetic
objects close to the sensor. The sensor then detects the change in
the total field at the sensor. Bias field sensors must not only detect
fields which are typically larger than the Earth’s field, but they
must also not be permanently affected or temporarily upset by a
large field. Sensors in this category include reed switches, InSb
magnetoresistors, Hall devices, and GMR sensors.
(a) Reed switch. The reed switch can be considered the simplest
magnetic sensor used to produce a usable output for industrial control. Reed switches are maintenance free and highly
immune to dirt and contamination. Rhodium-plated contacts
ensure long contact life. Typical capabilities are 0.1–0.2 A
switching current and 100–200 V switching voltage. Contact
life is measured at 106–107 operations at 10 mA. Reed
switches are available with normally open, normally closed,
and class C contacts. Latching reed switches are also available. Mercury-wetted reed switches can switch currents as
high as 1 Å and have no contact bounce.
Zhang_Ch01.indd 34
5/13/2008 5:45:24 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
35
Low cost, simplicity, reliability, and zero power consumption make reed switches popular in many applications. The
addition of a separate small permanent magnet yields a simple proximity switch often used in security systems to monitor the opening of doors or windows. The magnet, affixed to
the movable part, activates the reed switch when it comes
close enough. The desire to sense almost everything in cars
is increasing the number of reed switch sensing applications
in the automotive industry.
(b) Lorentz force devices. There are several sensors that utilize
the Lorentz force, or Hall effect, on charge carriers in a semiconductor. The Lorentz force equation describes the force FL
experienced by a charged particle with charge q moving with
velocity v in a magnetic field B:
FL = q (v × B).
The Hall effect is a consequence of the Lorentz force in
semiconductor materials. The following sensors work with
the Lorentz force and the Hall effect.
(i) Magnetoresistors. The simplest of Lorentz force devices
are magnetoresistors using semiconductors with high
room temperature carrier mobility. If a voltage is applied
along the length of a thin slab of semiconductor material,
a current will flow and a resistance can be measured.
When a magnetic field is applied perpendicular to the
slab, the Lorentz force will deflect the charge carriers. If
the width of the slab is greater than the length, the charge
carriers will cross the slab without a significant number
of them collecting along the sides. The effect of the magnetic field is to increase the length of their path and,
therefore, the resistance. An increase in resistance of several hundred percentage is possible in large fields. In
order to produce sensors with hundreds to thousands of
ohms of resistance, long, narrow semiconductor stripes a
few microns wide are produced using photolithography.
The required length to width ratio is accomplished by
forming periodic low-resistance metal shorting bars
across the traces. Each shorting bar produces an equipotent across the semiconductor stripe. The result is, in
effect, a number of small semiconductor elements with
the proper length to width ratio connected in series.
(ii) Hall effect sensors. The second type of sensor, which utilizes the Lorentz force on charge carriers, is a Hall sensor.
Zhang_Ch01.indd 35
5/13/2008 5:45:24 PM
36
INDUSTRIAL CONTROL TECHNOLOGY
The Hall resistance and Hall voltage increase linearly
with applied field to several teslas (10 s of kG).
The temperature dependence of the Hall voltage and
the input resistance of Hall sensors are governed by the
temperature dependence of the carrier mobility and that
of the Hall coefficient. The Hall voltage is measured
between electrodes placed at the middle of each side.
This differential voltage is proportional to the magnetic
field perpendicular to the slab. It also changes sign
when the sign of the magnetic field changes. The ratio
of the Hall voltage to the input current is called the Hall
resistance, and the ratio of the applied voltage to the
input current is called the input resistance.
(iii) Integrated Hall sensors. Hall devices are often combined with semiconductor elements to make integrated
sensors. By adding comparators and output devices to a
Hall element, manufacturers provide unipolar and bipolar digital switches. Adding an amplifier increases the
relatively low voltage signals from a Hall device to produce radiometric linear Hall sensors with an output
centered on one-half the supply voltage.
(c) Giant magnetoresistive (GMR) devices. Large magnetic field
dependent changes in resistance are possible in thin-film ferromagnetic and/or nonmagnetic metallic multilayers. This
phenomenon was first observed in 1988. Changes in resistance with magnetic field of up to 70% were observed.
Compared to the few percentage change in resistance
observed in anisotropic magnetoresistance (AMR), this phenomenon was truly giant magnetoresistance (GMR). GMR
devices involve a sandwich of two outer layers of a ferromagnetic material, such as cobalt or iron, with a center of a
nonmagnetic metal. One of the ferromagnetic layers is kept
under a constant magnetic field, while the other layer is
exposed to the variable magnetic field to be sensed. Maximum
current flows when both ferromagnetic layers are magnetized in the same direction; minimum current flows when the
layers are magnetized in reverse directions. GMR sensors
are currently used in read-write heads for disk drives.
(i) Unpinned sandwich GMR. These materials consist of
two soft magnetic layers of iron, nickel, and cobalt
alloys separated by a layer of a nonmagnetic conductor
such as copper. With magnetic layers 4–6 nm (40–60 Å)
thick separated by a conductor layer typically 3–5 nm
thick, there is relatively little magnetic coupling between
Zhang_Ch01.indd 36
5/13/2008 5:45:24 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
37
the layers. For use in sensors, sandwich material is usually patterned into narrow stripes. The magnetic field
caused by a current of a few milliamperes per micron of
stripe width flowing along the stripe is sufficient to
rotate the magnetic layers into antiparallel or high resistance alignment. An external magnetic field of 3–4 kA/m
applied along the length of the stripe is sufficient to
overcome the field from the current and rotate the magnetic moments of both layers parallel to the external
field. A positive or negative external field parallel to
the stripe will produce the same change in resistance.
An external field applied perpendicular to the stripe will
have little effect due to the demagnetizing fields associated with the extremely narrow dimensions of these
magnetic objects. The value usually associated with the
GMR effect is the percentage change in resistance normalized by the saturated or minimum resistance.
Sandwich materials have values of GMR typically 4–9%
and saturate with 2.4–5 kA/m applied field. Figure 1.21
shows a typical resistance vs field plot for sandwich
GMR material.
(ii) Antiferromagnetic multilayer. These materials consist
of multiple repetitions of alternating conducting magnetic layers and nonmagnetic layers. Since multilayer
has more interfaces than do sandwiches, the size of the
GMR effect is larger. The thickness of the nonmagnetic
layers is less than that for sandwich material (typically
1.5–2.0 nm) and the thickness is critical. Only for certain thicknesses, the polarized conduction electrons
cause antiferromagnetic coupling between the magnetic
layers. Each magnetic layer has its magnetic moment
antiparallel to the moments of the magnetic layers on
each side—exactly the condition needed for maximum
spin-dependent scattering. A large external field can
overcome the coupling which causes this alignment and
can align the moments so that all the layers are parallel—
the low-resistance state. If the conducting layer is not of
proper thickness, the same coupling mechanism can
cause ferromagnetic coupling between the magnetic
layers resulting in no GMR effect.
(iii) Spin valves. These materials, or antiferromagnetically
pinned spin valves, are similar to the unpinned spin
valves or sandwich materials described earlier. An additional layer of an antiferromagnetic material is provided
Zhang_Ch01.indd 37
5/13/2008 5:45:24 PM
38
INDUSTRIAL CONTROL TECHNOLOGY
on the top or the bottom. The antiferromagnetic material
couples to the adjacent magnetic layer and pins it in a
fixed direction. The other magnetic layer is free to rotate.
These materials do not require the field from a current to
achieve antiparallel alignment.
(iv) Colossal magnetoresistance (CMR). CMR occurs in
crystals of manganese oxide known as “magnates.”
Under certain conditions these mixed oxides undergo a
semiconductor to metallic transition with the application
of a magnetic field of a few teslas (10 s of kG). The size
of the resistance ratios, measured at 103–108%, has generated considerable excitement even though they required
high fields and liquid nitrogen temperatures. The MR of
these crystals actually decreases in the presence of a
magnetic field. CMR requires cryogenic cooling.
1.1.7
Limit Switches
A limit switch is an electromechanical device that can be used to determine the physical position of equipment. For example, an extension on a
valve shaft mechanically trips a limit switch as it moves from open to shut
or shut to open. The limit switch is designed to give a signal to an industrial control system when a moving component like an overhead door or
piece of machinery has reached the limit (end point) for its travel or just a
specific point on its journey. The primary purpose of the limit switch is to
control the intermediate or end limits of a linear or rotary motion. In industrial control systems, the limit switch is often used as a safety device to
protect against accidental damage to equipment.
1.1.7.1
Operating Principle
A linear limit switch is an electromechanical device that requires the
physical contact of an object with the activator of the switch to make the
contacts change state. As an object (target) makes contact with the actuator
of the switch, it moves the activator to the “limit” where the contacts
change state. Limit switches can be used in almost any industrial environment because of their typically rugged design. However, the device uses
mechanical parts that can wear over time and the device is “slower” when
compared to noncontact, electrical devices such as proximity sensors and
photoelectric sensors.
Rotary limit switches are similar to relays in that they are used to allow
or prevent current flow when in closed or open position. The groupings
Zhang_Ch01.indd 38
5/13/2008 5:45:24 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
39
within this family are usually defined based on the manner in which the
switch is actuated (e.g., rocker, foot, read, lever, etc.). Switches can range
from simple push button devices, usually used to delineate between ON
and OFF, to rotary and toggle devices, for varying levels, through to multiple entry keypads, for multiple control functions. In addition to maintaining or interrupting flow, and maintaining flow levels, switches are used in
safety applications as security devices (locker switches) and as functionary actuators when controlled by sensors or computer systems.
Normally, the limit switch gives ON/OFF output that corresponds to
valve position. Limit switches are used to provide full open or full shut
indications as illustrated in Fig. 1.17 which gives a typical linear limit
switch operation. Many limit switches are of the push button variety. In
Fig. 1.17, when the valve extension comes in contact with the limit switch,
the switch depresses to complete, or turn on, the electrical circuit. As the
valve extension moves away from the limit switches, spring pressure opens
the switch, turning off the circuit. Limit switch failures are normally
mechanical in nature. If the proper indication or control function is not
achieved, the limit switch is probably faulty. In this case, local position
indication should be used to verify equipment position.
Full open
limit switch
Indicating
and control
circuite
Stem
Full closed
limit switch
Flow
Figure 1.17 Operations of a linear limit switch.
Zhang_Ch01.indd 39
5/13/2008 5:45:24 PM
40
INDUSTRIAL CONTROL TECHNOLOGY
1.1.7.2
Basic Types and Application Guide
Limit switches come in many different forms, from small, enclosed
switches to large, heavy-duty multicircuit switches and are divided into
several types as given below. The basic types of limit switches are (1) linear
limit switches where an object will move a lever (or simply depress a
plunger) on the switch far enough for the contact in the switch to change
state; (2) rotary limit switches where a shaft must turn a preset number of
revolutions before the contact changes state, used in cranes, overhead
doors, etc.; and (3) magnetic limit switches, or reed switches, where the
object is not touched but sensed. One heavy-duty application would be in
mine hoists where a stationary switch will sense a strong magnet mounted
to a car going up or down the mine shaft.
(1) Linear limit switches. Linear limit switches basically have these
three types:
(a) Safety limit switches. Safety limit switches are designed for
use with moveable machine guards/access gates, which must
be closed for operator safety and for any other presence, and
position-sensing application normally addressed with conventional limit switches. Their positive opening network
connection contacts provide a higher degree of reliability
than conventional spring-driven switches whose contacts
can weld or stick shut. These limit switches are of multiple
actuator styles and four 90° head position so that they provide application versatility.
(b) Comprehensive range limit switches. Comprehensive range
limit switches are provided in the most popular industrial
sizes, shapes, contact configurations, as well as an innovative series of encapsulated limit switches, available with a
connection cable or plug.
Their output contacts are offered in snap action, slow
action, or overlapping configurations. Various types of push
button and roller actuators, roller levers, and multidirectional
levers are available. These switches are ideal for manufacturers of material handling, packaging, conveying, and
machine tool equipment.
(c) Mechanical limit switches. Mechanical limit switches are
frequently used in position detection on doors and machinery and parts detection on conveyors and assembly lines.
Their range comprises various housing and actuator styles,
choice of slow or fast action contacts, and various contact
arrangements.
Zhang_Ch01.indd 40
5/13/2008 5:45:25 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
41
(2) Rotary limit switches. Rotary switches move in a circle and can
stop in several positions in their range, which provide singledeck rotary limit and multiple-deck rotary limit.
(a) Single-deck rotary switches. Single-deck rotary switches can
control several circuits at a time. Actuator choices for singledeck rotary switches include flush actuator, bare shaft actuator, knobbed shaft, and locker. In a flush actuator configuration
the actuator does not project above the switch body. Typically
it requires a screwdriver for operation. In a bare shaft actuator configuration the shaft has no knob, but may be notched
to accept various knob configurations. A knobbed shaft
comes with an integral knob. In a locker configuration the
actuation is done with a key or other security or tamperproof
method.
Important physical switch specifications to consider when
searching for single-deck rotary switches include angle
between positions, mechanical life, number of decks, number of poles per deck, and number of poles. The angular distance (in degrees) exists between positions. For example, for
a 4-position switch, the angle of throw is 90°, and for a 100position switch, the angle of throw is 3.6°. The mechanical
life is the maximum life expectancy of the switch. Often,
electrical life expectancy is less than mechanical life (please
consult manufacturer). The number of decks specifies the
maximum number of decks that can be attached to a common actuating shaft. The number of poles per deck refers to
the number of separate circuits that can be activated through
a switch at any given time per deck. The number of poles
refers to the number of separate circuits that can be activated
through a switch at any given time. Important electrical
switch specifications to consider include maximum current
rating, maximum AC voltage rating, and maximum DC voltage rating. Common materials of construction for the base
and actuator include plastics and metals.
Other specifications to consider for single-deck rotary
switches include stop style, contact style, actuator features,
terminal type, features, and environmental parameters. Stop
style choices include fixed stop, adjustable stop, and continuous or no stops. Contact style choices are shorting or nonshorting. Actuator features include integral potentiometer,
actuator detents, tease proof, and guarded positions. Terminal
type choices include wire leads, solder terminals, screw terminals, and PCB pins. An important environmental parameter to consider is the operating temperature.
Zhang_Ch01.indd 41
5/13/2008 5:45:25 PM
42
INDUSTRIAL CONTROL TECHNOLOGY
(b) Multiple-deck rotary switch. Multiple-deck rotary switches
can control several circuits simultaneously. Actuator choices
for multiple-deck rotary switches include flush actuator, bare
shaft actuator, knobbed shaft, and locker. In a flush actuator
configuration the actuator does not project above the switch
body, and typically requires a screwdriver for operation. In a
bare shaft actuator configuration the shaft has no knob, but
may be notched to accept various knob configurations. A
knobbed shaft comes with an integral knob. In a locker configuration the actuation is done with a key or other secure or
tamperproof method.
Important physical switch specifications to consider when
searching for multiple-deck rotary switches include angle
between positions, mechanical life, number of decks, and
number of poles per deck. The angular distance (in degrees)
exists between positions. For example, for a 4-position
switch, the angle of throw is 90°; for a 100-position switch,
the angle of throw is 3.6°. The mechanical life is the maximum life expectancy of the switch. Often, electrical life
expectancy is less than mechanical life (please consult manufacturer). The number of decks specifies the maximum
number of decks that can be attached to a common actuating
shaft. The number of poles per deck refers to the number of
separate circuits that can be activated through a switch at any
given time per deck. Important electrical switch specifications to consider include maximum current rating, maximum
AC voltage rating, and maximum DC voltage rating. Common
materials of construction for the base and actuator include
plastics and metals.
Other specifications to consider for multiple-deck rotary
switches include stop style, contact style, actuator features,
terminal type, features, and environmental parameters. Stop
style choices include fixed stop, adjustable stop and continuous or no stops. Contact style choices are shorting or nonshorting. Actuator features include integral potentiometer,
actuator detents, tease proof, and guarded positions. Terminal
type choices include wire leads, solder terminals, screw terminals, and PCB pins. Common features for multiple-deck
rotary switches include optional coded outputs, momentary
on, wiping contacts, CE certification, CSA certification,
UL listed, dustproof, and weather resistant or waterproof.
An important environmental parameter to consider is the
operating temperature.
Zhang_Ch01.indd 42
5/13/2008 5:45:25 PM
43
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
(c) Magnetic limit switches. Magnetic limit switches work based
on the operation of reed switches (see Section 1.2.8), and are
more reliable than the read switches because of their simplified construction. The switches are constructed of flexible
ferrous strips (reeds) and are placed near the intended travel
of the valve stem or control rod extension.
When using reed switches, the extension used is a permanent magnet. As the magnet approaches the reed switch, the
switch shuts. When the magnet moves away, the reed switch
opens. This ON/OFF indicator is similar to a mechanical
limit switch. By using a large number of magnetic reed
switches, incremental position can be measured.
Failures are normally limited to a reed switch, which is
stuck open or stuck shut. If a reed switch is stuck shut, the
open (closed) indication will be continuously illuminated. If
a reed switch is stuck open, the position indicator for that
switch remains extinguished regardless of valve position.
1.1.7.3
Calibration
Limit switches are not adjusted at the factory. Limit switches should be
set during installation according to the specifications from the factories
and vendors. In most cases, turning some screws performs these adjustments. Figure 1.18 is an example of a rotary limit switch that gives
eight positions. By turning some screws, each of these positions can be
fitted in installations.
Position
1
Position
2
Position
3
Position
4
Position
5
Position
7
Position
6
Position
8
Figure 1.18 The positions of a rotary limit switch.
Zhang_Ch01.indd 43
5/13/2008 5:45:25 PM
44
1.1.8
INDUSTRIAL CONTROL TECHNOLOGY
Photoelectric Devices
Photoelectric sensors and switches represent perhaps the largest variety
of problem-solving choices in the industrial sensor market. Today, photoelectric technology has advanced to the point where it is common to find a
sensor or a switch that can detect a target less than 1 mm in diameter while
other units have a sensing range up to 60 m. These factors make them
extremely adaptable in an endless array of applications. Photoelectric sensors and switches and their future successors will continue to be key detection devices that can be depended on for these applications.
Photoelectric sensors and switches are used extensively on packaging
machinery, automatic door systems, automotive and metal industries, and
food processing industry. For example, they are used for detecting the
presence of automobile wheel rims and tire presence on an assembly line
in automotive and metal industries, or for detecting the part cases on an
assembly line in food processing and packaging.
A very familiar application of photoelectric switches is that they can
turn the trap on at dark and off at daylight. This enables the user to set the
trap out earlier and retrieve it later in the morning while reducing wear on
the battery. Quite often, a garage door opener has a through beam photoelectric sensor mounted near the floor, across the width of the door. This
sensor makes sure nothing is in the path of the door when it is closing. A
more industrial application for a photoelectric device is detecting objects
on a conveyor. An object will be detected any place on a conveyor running
between the emitter and the receiver as long as there is a gap between the
objects and the sensor’s light does not “burn through” the object. This is
more a figurative term than a literal one. It refers to an object that is thin or
light in color and allows the light emitted from the emitter to penetrate the
target so the receiver never detects the object.
1.1.8.1
Operating Principle
Almost all photoelectric sensors and switches contain an emitter, which
is a light source such as a light emitting diode or laser diode, a photodiode,
or a phototransistor receiver to detect the light source, as well as the supporting electronics designed to amplify the signal relayed from the receiver.
Photoelectric sensing uses a beam of light to detect the presence or absence
of an object: the emitter transmits a beam of light either visible or infrared,
which in some fashion is directed to and detected by the receiver.
All photoelectric sensors and switches identify their output as “dark
on” and “light on,” which refer to output of the sensor or switch in
Zhang_Ch01.indd 44
5/13/2008 5:45:26 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
45
relation to when the light source is hitting the receiver. If an output is
present while no light is received, this would be called a “dark-on” output. In reverse, if the output is ON while the receiver is detecting the light
from the emitter, the sensor or switch would have a “light-on” output.
Either way, a light-on or dark-on output needs to be selected prior to purchasing the sensor unless it is user adjustable. In this case it can be decided
upon during installation by either flipping a switch or wiring the sensor
accordingly.
The method in which light is emitted and delivered to the receiver is the
way to categorize the different photoelectric configurations. They can be
categorized into three main categories: through beam, retroreflective, and
proximity.
(1) Through beam photoelectric sensors or switches. are configured
with the emitter and detector opposite the path of the target and
sense presence when the beam is broken.
(2) Retroreflective photoelectric sensors or switches. are configured
with the emitter and detector in the same housing and rely on a
reflector to bounce the beam back across the path of the target.
This type may be polarized to minimize false reflections.
(3) Proximity photoelectric sensors or switches. have the emitter
and detector in the same housing and rely upon reflection from
the surface of the target. This mode can include presence sensing
and distance measurement via analog output. The proximity category can be further broken down into five submodes: diffuse,
divergent, convergent, fixed field, and adjustable field. With a
diffuse sensor the presence of an object is detected when any
portion of the diffuse reflected signal bounces back from the
detected object. Divergent beam sensors and switches are shortrange, diffuse-type sensors or switches without collimating
lenses. Convergent, fixed focus, or fixed distance optics (such as
lenses) are used to focus the emitter beam at a fixed distance
from the sensor or switch. Fixed-field sensors or switches are
designed to have a distance limit beyond which they will not
detect objects, no matter how reflective. Adjustable field sensors
or switches utilize a cut-off distance beyond which a target will
not be detected, even if it is more reflective than the target. Some
photoelectric sensors and switches can be set for multiple different optical sensing modes. Reflective properties of the target and
environment are important considerations in the choice and use
of photoelectric sensors and switches.
Zhang_Ch01.indd 45
5/13/2008 5:45:26 PM
46
INDUSTRIAL CONTROL TECHNOLOGY
Diffuse sensors or switches operate under a somewhat different style than retroreflective and through-beams although the
operating principle remains the same: diffuse photoelectric sensors and switches actually use the target as the “reflector,” such
that detection occurs upon reflection of the light off the object
back onto the receiver as opposed to an interruption of the beam.
The emitter sends out a beam of light. Most often it is a pulsed
infrared, visible red, or laser beam, which is reflected by the target when it enters the detectable area. The beam is diffused off of
the target in all directions. Part of the beam will actually return
to the receiver inside the same housing from which the sensor or
switch originally emitted it. Detection occurs and the output will
either turn on or off (depending upon if it is light on or dark on)
when sufficient light is reflected to the receiver. This can be commonly witnessed in airport washrooms, where a diffuse photo
will detect your hands as they are placed under the faucet and the
attending output will turn the water on. In this application, your
hands act as the reflector.
To ensure repeatability and reliability, photoelectric sensors and switches
are available with three different types of operating principles: fixed-field
sensing, adjustable field sensing, and background suppression through triangulation. In the simplest terms, these sensors and switches are focused on
a specific point in the foreground, ignoring anything beyond that point.
(1) Standard fixed-field sensors and switches. operate optimally at
their preset “sweet spot,” the distance at which the foreground
receiver will detect the target. As a result, these sensors and
switches must be mounted within a certain fixed distance of the
target. In fixed-field technology, when the emitter sends out a beam
of light, two receivers sense the light on its return. The shortrange receiver is focused on the target object’s location. The
long-range receiver is focused on the background. If the longrange receiver detects a higher intensity of reflected light than
the short-range receiver, the output will not turn on. If the shortrange receiver detects a higher intensity of reflected light than the
long-range receiver, an output occurs and the object is detected.
(2) Adjustable field sensors and switches. operate under the same
principle as fixed-field sensors or switches, but the sensitivity of
the receivers can be electrically adjusted using a potentiometer.
By adjusting the level of light needed to trigger an output, the
range and sensitivity of the sensor or switch can be altered to fit
the application.
Zhang_Ch01.indd 46
5/13/2008 5:45:26 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
47
(3) Background suppression by triangulation. also emits a beam of
light that is deflected back to the sensor or switch. Unlike fixedand adjustable field sensors and switches, which rely on the
intensity of the light reflected back to them, background suppression sensors rely completely on the angle at which the beam of
light returns. Like fixed- and adjustable field sensors or switches,
background suppression sensors or switches feature short-range
and long-range receivers in fixed positions. In addition, background suppression sensors or switches have a pair of lenses that
are mechanically adjusted to precisely focus the reflected beam
to the appropriate receiver, changing the angle of the light
received. The long-range receiver is focused through the lens on
the background. Deflected light returning along that focal plane
will not trigger an output. The short-range receiver is focused,
through a second lens, on the target. Any deflected light returning along that focal plane will trigger an output—an object will
be detected.
1.1.8.2 Application Guide
In industrial control systems which require photoelectric sensors and
switches, it is important to check whether the parameters and principle of
the sensors and switches satisfy your applications.
(1) Choosing the right parameters. Important parameters to consider
when looking for photoelectric sensors and switches include
sensing mode, detecting range, position measurement window,
minimum detectable object, and response time. Sensing modes
can be used for detecting presence or absence and position measurement. With a presence or absence sensor, the sensor detects
presence or absence in an ON or OFF mode. In position measurement, the sensing mode of the sensor can detect position in a
linear region by the intensity of reflected light. Analog output is
linear with position in the measurement range. The detecting
range is the range of sensor detection. For presence sensors, this
goes up to the maximum distance for which the signal is stable.
For position measurement sensors, this is the distance or range
over which the position vs output response is linear and stable.
The position measurement window is the width of linear region
for the sensor. For example, if the sensor could measure between
14 and 24 cm, this window would be 10 cm. The minimum
detectable object is the smallest sized object detectable by the
sensor. The response time is the time from target object entering
detection zone to the production of the detection signal.
Zhang_Ch01.indd 47
5/13/2008 5:45:26 PM
48
INDUSTRIAL CONTROL TECHNOLOGY
Common configuration features for photoelectric sensors and
switches include beam visibility, light-on or dark-on modes, light
and dark programmability, adjustable sensitivity, self-teaching,
laser source, fiber-optic glass, and fiber-optic plastic. The body
style of the sensor can be threaded barrel, cylindrical, limit
switch, rectangular, slot, ring, and window or frame. The sensor
may be self-contained and may have a remote head.
Repeatability and reliability are critical to the overall performance of a photoelectric sensor or switch in an industrial processing line. One of the most common type of sensor and switch
used for object detection is the diffuse photoelectric sensor which
performs well in a wide range of industrial processing applications. However, these sensors and switches can experience problems with some target or background materials.
The accuracy of diffuse sensors is often at the mercy of the
surface properties of the target and background materials. A nonreflective target, such as a matte or dull black material, is difficult for diffuse photoelectric sensors to detect, because it reflects
much less light than a brightly colored or white target. Similarly,
if the target is presented against a light-colored or reflective
background, the sensor can be falsely triggered by light reflected
from the background material rather than the target object.
To overcome these challenges, a variety of technologies have
been developed to allow the sensors to see an object while ignoring background materials.
(2) Choosing the right principle. Photoelectric sensors and switches
are reliable, versatile, and able to sense and take on or off action
with objects of almost any material, size, and shape at distances
ranging from 5 mm to 40 m depending on type and configuration
used. The addition of fiber cables considerably extends application opportunities for photoelectric sensors, allowing their installation in confined areas, as well as in areas in which intrinsic
safety would normally disallow the use of electronics.
For applications where backgrounds are not within sensing
range and target color is consistent, a standard diffuse sensor or
switch is completely sufficient for object detection. These sensors
and switches are also quite effective for detecting large objects.
Similarly, if a background within the sensing range is not particularly reflective and the color and reflectivity of the target will
remain relatively constant, a fixed- or adjustable field sensor will
likely provide trouble-free performance. These sensors are also
appropriate for smaller targets than standard diffuse sensors.
When reliable sensing is challenging due to shiny backgrounds and targets, shifting colors, and reflectivity, background
Zhang_Ch01.indd 48
5/13/2008 5:45:26 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
49
suppression by triangulation is the most repeatable and reliable
solution. These sensors and switches, especially laser varieties,
are very effective for detecting ultra-small targets. The precisely
focused, finely collimated beam allows them to detect extremely
small items.
When an application requires the durability and performance
of a full-featured photoelectric sensor or switch but distances are
shorter and space is tight, you can take a 90° turn to right sight of
photoelectric sensors and switches. Right sight of photoelectric
sensor and switches takes many of the features of the Series 9000
and puts them in a smaller, more adaptable package to deliver
excellent detection capabilities where size and shape matter.
These photoelectric sensors and switches combine universal
voltages (24–264 V AC/DC) with short-circuit protection across
the full voltage range.
1.1.8.3
Basic Types
(1) Diffuse photoelectric sensors
(a) Standard diffuse sensors. The emitter and receiver are in the
same housing. The emitter sends out a beam of pulsed red or
infrared light that is reflected directly by the target. When the
beam of light hits the target at any angle, it is diffused in
all directions and some light is reflected back. The receiver
sees only a small portion of the original light, switching
the sensor when a target is detected within the effective scan
range.
The simplest diffuse photoelectric sensors use the target
object as the reflective surface for object detection. Detection
occurs when a beam of infrared, visible red, or laser light
emitted from the sensor is deflected off the target material in
all directions and detected by the receiver.
Standard diffuse sensors have these features: (1) the sensing range depends largely on the reflective properties of
the target’s surface, (2) suitable for distinguishing between
black and white targets, (3) relatively large active range, and
(4) positioning and monitoring with only one sensor.
The typical applications of standard diffuse sensors are
(1) distinguishing and sorting of objects according to their
volume or degree of reflection, (2) counting of objects, and
(3) presence detection of boxes.
(b) Diffuse sensors with background suppression. These sensors
are a special development of the diffuse sensor. The beam of
light is closely focused, and therefore able to distinguish a
Zhang_Ch01.indd 49
5/13/2008 5:45:26 PM
50
INDUSTRIAL CONTROL TECHNOLOGY
target within a precisely defined scan range and ignore targets outside the range.
Diffuse sensors with background suppression have these
features: (1) sensing range largely independent of the color
and surface of the target and (2) they detect small objects.
They obtain the following typical applications: (a) sorting
objects without concern for the background color, purely on
their distance from the sensor and (b) sensing contents within
transparent packaging.
(2) Retroreflective photoelectric sensors
(a) Retroreflective sensors. With the emitter and receiver in the
same housing, this sensor transmits a pulsed infrared or red
light beam which is reflected back from a “triple prism” reflector or reflective tape. The sensor switches when the light beam
is interrupted. These devices recognize objects independent of
their surface qualities, as long as they are not too shiny.
Retroreflective sensors offer these features: (1) large sensing range and (2) matte-finished objects are recognized independent of their surface properties.
The typical applications are (1) height detection of stacked
objects and (2) control of randomly positioned objects on a
conveyor.
(b) Retroreflective sensors with polarization filter. Retroreflective
sensors with polarization filters correctly recognize highly
reflective objects. The polarizing filter prevents false switching with shiny objects. Only the stray and nonpolarized light
from the reflector actuates the sensor.
Features are similar to retroreflective sensors, but with the
added advantage of being able to accurately distinguish
shiny objects.
Typical application is for monitoring shiny cans on a conveyor belt.
(3) Through-beam photoelectric sensors. Emitter and receiver are in
two separate housings facing each other. The sensor switches
whenever the light beam is interrupted.
The features of the through-beam photoelectric sensors are
(1) through-beam sensors offer the largest sensing ranges, (2) the
switching point is independent of the surface nature of the object,
and (3) due to the narrow effective beam, through-beam sensors
have excellent repeatability.
Typical applications are (1) monitoring doors and gates and
(2) counting and monitoring of objects over large distances.
(4) Fiber-optic photoelectric sensors. The front surfaces of the fibers,
which are set and glued into the sensing head, are precisely
Zhang_Ch01.indd 50
5/13/2008 5:45:26 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
51
ground and polished to provide outstanding optics. Even difficult
multiple sensing problems can be solved by using fiber optics.
There are two types of fibers: Reflective fiber types may be
applied using the same guidelines as diffuse sensors. Throughbeam type fiber optics may be applied using the same guidelines
as through-beam sensors.
Typical applications are (1) sorting various objects, (2) measuring of diameters and heights, (3) checking for double sheets,
(4) detecting small objects, (5) monitoring flow of bread in an
oven, and (6) detecting absence/presence of lids on a process filling line.
(5) Light barrier photoelectric sensors. The infrared light barrier
photoelectric sensors are a series of infrared light through beams,
mounted in a tower-type emitter and receiver housing. The beam
arrays can be used to recognize objects or to make continuous
measurements.
The system consists of individual emitters and receivers separately mounted in two towers, placed opposite each other. In
addition, a micro controller is also mounted in one of the housings to control the light beams. During a measuring cycle, individual emitting diodes are activated in sequence. At the same
time the corresponding receiving diode is enabled. Each light
beam is defined as interrupted as soon as the imaginary line from
transmitter to receiver is blocked. The same principle is followed
for all subsequent beams; hence a light barrier is created. During
a measuring cycle all interrupted beams are registered. Light grid
sensors with retroreflective technology are simple in installation
and setup.
(6) Level monitoring photoelectric sensors. Levels can be measured
simply and accurately using infrared light, without the need for
any electrical or thermal connection between the target medium
and sensor. The ratio of reflective indices changes depending on
whether the tip of the sensor is surrounded by liquid or air. If the
sensor tip is immersed in liquid, the light rays will be deflected
into the liquid and the electronics of the receiver changes its
switching status.
The operating principle remains the same irrespective of
whether the liquid medium can conduct electricity or not. The
medium can also be clear or cloudy.
(7) Line photoelectric sensor. Line photoelectric sensors can easily
detect positions, edges, or widths of objects. The accurate information can be read out either as a 4–20 mA or via serial interface.
The adopted integrated illumination allows a reliable function
without spending too much time on maintenance and installation.
Zhang_Ch01.indd 51
5/13/2008 5:45:26 PM
52
INDUSTRIAL CONTROL TECHNOLOGY
The three different measuring principles (width, edge, middle)
are reachable by a button—the current measuring mode, if the
sensor is powerless for a certain time.
(8) Photoelectric safety switches. Single-beam or through-beam photoelectric safety switches are used as noncontact access protection to hazardous zones. Photoelectric safety switches consist of
testable sender and receiver units in combination with a safety
evaluation unit, which are mainly used on robot systems and processing machines.
(9) Photoelectric proximity switches. Photoelectric proximity switches
are sought for reliable detection of objects within a defined scanning range on a conveyor system. Objects which reduce distance
from scanning plane to sensor are detected. Adjusting sensitivity
can set scanning range and switching point. In the case of photoelectric switches with fiber-optic cable, the sender and receiver
are contained in a single housing. A separate fiber-optic cable is
used for the sender and the receiver for operation as a throughbeam system. For use as a proximity switch the sender and
receiver fiber-optic cables are combined in one cable.
1.1.9
Proximity Devices
Proximity sensing is the ability of a device to tell when it is near an object
or when something is near it. This sense keeps a device from running into
things. It can also be used to measure the distance from a device to some
object. Proximity sensors detect the presence of an object without physical
contact, and proximity switches execute necessary responses when sensing
the presence of the targets or some critical positions located by the targets. A
position sensor determines an object’s coordinates (linear or angular) with
respect to a reference; displacement means moving from one position to
another for a specified distance (or angle). In effect, a proximity sensor is a
threshold version of a position sensor. Proximity sensing is the technique of
detecting the presence or absence of an object using a critical distance.
Typical applications include the control, detection, position, inspection,
and automation of machine tools and manufacturing systems. They are
also used in the following machinery: packaging, production, printing,
plastic merging, metal working, and food processing, etc. The measurement of proximity, position, and displacement of objects is essential in
many different applications: valve position, level detection, process control, machine control, security, etc.
Special purpose proximity sensors perform in extreme environments
(exposure to high temperatures or harsh chemicals) and address specific
Zhang_Ch01.indd 52
5/13/2008 5:45:26 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
53
needs in automotive and welding environments. Inductive proximity sensors are ideal for virtually all metal sensing applications, including detecting all metals or nonferrous metals only.
1.1.9.1
Operating Principle
In terms of physics, proximity sensors and switches operate with capacitive, inductive, photoelectric, ultrasonic, and magnetic mechanisms to
achieve both the proximity sensing and proximity switching in industrial
control. The sensing principle with ultrasonic waves is given in Section 1.1.3;
the operating principle with magnetism is given in Section 1.1.6 and with
photoelectric physics is given in Section 1.1.8; this subsection therefore
introduces the working principle for capacitive and inductive proximity
sensors and switches.
(1) Capacitive proximity sensors and switches make use of the
variation of the parasitic capacitance that develops between the
sensor and the object to be detected. When the object is at a predetermined distance from the sensitive side of the sensor, an
electronic circuit inside the sensor begins to oscillate. A capacitive sensor can detect metallic and nonmetallic objects like wood,
plastic, and liquid materials. The operating distance can be
trimmed, making the sensor useful for each specific application.
Capacitive proximity sensors are housed in smooth or threaded
cylindrical metallic cases. Capacitive proximity sensors sense
“target” objects, owing to the target’s ability to be electrically
charged. Since even nonconductors can hold charges, this means
that just about any object can be detected with this type of sensor.
Figure 1.19 demonstrates the working principle of capacitive
proximity sensing. As given in Fig. 1.19, a capacitive proximity
sensor or switch normally contains four essential components:
an electrode assembly, an oscillator circuit, an evaluation circuit,
and an output circuit. When the increase in capacitance is large
enough, an oscillation is set up. This oscillation is detected by
the evaluation circuit, which then changes the state of the output
circuit.
An electrode assembly is designed so that an electrostatic field
is formed between the active electrode and the earth electrode.
Any object entering this field will increase the capacitance. The
increase in capacitance depends on the following factors: the distance and position of the object in front of the active electrode,
the dimensions of the object, and the dielectric constant of the
object.
Zhang_Ch01.indd 53
5/13/2008 5:45:26 PM
54
INDUSTRIAL CONTROL TECHNOLOGY
Internal capacitor plate
Effective
capacitor plate
in target
Current sensor
Oscillator
+
DC supply
DC output
−
Air
(dielectric)
+
Figure 1.19 Capacitive proximity sensor and its sensing principle.
(2) Inductive proximity sensors and switches comprise an oscillating
circuit, a signal evaluator, and a switching amplifier. The coil of
this oscillating circuit generates a high-frequency electromagnetic alternating field. This field is emitted at the sensing face of
the sensor. If a metallic object (switching trigger) nears the sensing face, eddy currents are generated. The resultant losses draw
energy from the oscillating circuit and reduce the oscillations.
The signal evaluator behind the oscillating circuit converts this
information into a clear signal. Figure 1.20 demonstrates its
operating principle. The supply DC is used to generate AC in an
internal coil, which in turn causes an alternating magnetic field.
If no conductive materials are near the face of the sensor, the
only impedance to the internal AC is due to the inductance of the
coil. If, however, a conductive material enters the changing
magnetic field, eddy currents are generated in that conductive
material, and there is a resultant increase in the impedance to the
Oscillator
(generates AC)
Induction coil
(generates
changing
magnetic
field)
+
Magnetic field
(metal sensing region)
DC supply
−
+
DC output
(NO or NC available)
Current sensor
Figure 1.20 Inductive proximity sensor and its sensing principle.
Zhang_Ch01.indd 54
5/13/2008 5:45:26 PM
55
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
AC in the proximity sensor. A current sensor, also built into the
proximity sensor, detects when there is a drop in the internal AC
current due to increased impedance. The current sensor controls
a switch providing the output.
The operating principle for both capacitive and inductive is based on a
high-frequency oscillator that creates a field in the close surroundings of
the sensing surface. The presence of a metallic object (inductive) or any
material (capacitive) in the operating area causes a change in the oscillation amplitude. The rise or fall of such oscillation is identified by a threshold circuit that changes the output state of the sensor. The operating
distance of the sensor depends on the shape and size of both capacitive and
inductive proximity sensing devices and is strictly linked to the nature of
the material.
Table 1.1 gives the sensitivity of the capacitive proximity sensing
devices with respect to several typical metals, and Table 1.2 gives the sensitivity of the inductive proximity sensing devices with respect to several
typical metals. Normally, both a capacitive and inductive proximity sensor
and switch have a screw that allows regulation of the operating distance.
This sensitivity regulation is useful in applications such as detection of full
containers and nondetection of empty containers.
Table 1.1 The Sensitivity of Capacitive Proximity Sensors
for Different Metals
Metal
Water
Plastic
Glass
Wood
1 ⫻ Sn
1 ⫻ Sn
0.5 ⫻ Sn
0.5 ⫻ Sn
0.4 ⫻ Sn
Note: Sn is the operating distance.
Table 1.2 The Sensitivity of Inductive Proximity Sensors
for Different Metals
Fe37 (iron)
Stainless steel
Brass–bronze
Aluminum
Copper
1 ⫻ Sn
0.9 ⫻ Sn
0.5 ⫻ Sn
0.4 ⫻ Sn
0.4 ⫻ Sn
Note: Sn is the operating distance.
Zhang_Ch01.indd 55
5/13/2008 5:45:27 PM
56
INDUSTRIAL CONTROL TECHNOLOGY
1.1.9.2 Application Guide
The most important parameter to consider when specifying proximity
sensors is the operating distance. The rated operating distance is the distance at which switching takes place. Common body styles for proximity
sensors are barrel, limit switch, rectangular, slot style, and ring. Important
dimensions to consider when specifying proximity sensors include barrel
diameter, length, width, and height.
Proximity sensors can be a sensor element or chip, a sensor or transducer, an instrument or meter, a gauge or indicator, a recorder or totalizer,
or a controller. A sensor element or chip denotes a “raw” device such as a
strain gage, or one with no integral signal conditioning or packaging. A
sensor or transducer is a more complex device with packaging and/or signal conditioning that is powered and provides an output such as DC voltage, a current loop, etc. An instrument or meter is a self-contained unit that
provides an output such as a display locally at or near the device. Typically
it also includes signal processing and/or conditioning. A gauge or indicator is a device that has a (usually analog) display and no electronic output
such as a tension gauge. A recorder or totalizer is an instrument that
records, totalizes, or tracks force measurement over time. It includes simple data logging capability or advanced features such as mathematical
functions, graphing, etc.
Load configurations are also important parameters to consider. Proximity
sensors may switch an AC load or a DC load. DC load configurations can
be NPN or PNP. NPN is a transistor output that switches the common or
negative voltage to the load; load is connected between sensor output and
positive voltage supply. PNP is a transistor output that switches the positive voltage to the load; load is connected between sensor output and voltage supply common or negative. Wire configurations are 2-wire, 3-wire
NPN, 3-wire PNP, 4-wire NPN, and 4-wire PNP. Switch types can be normally open or normally closed.
1.1.9.3
Basic Types and Specifications
Proximity sensors and switches can have one of many physics and technology types. The physics types of proximity sensors and switches include
capacitive, inductive, photoelectric, ultrasonic, and magnetic proximity
sensors or switches. Common terms for technology types of proximity
sensors and switches include these kinds of proximity sensing devices:
eddy current, air, capacitance, infrared, fiber optics, etc. Proximity sensors
and switches can be of the contact or noncontact type.
Zhang_Ch01.indd 56
5/13/2008 5:45:27 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
57
(1) Physics types of proximity sensors and switches
(a) Capacitive proximity sensors and switches. Capacitive sensing devices utilize the face or surface of the sensor as one
plate of a capacitor and the surface of a conductive or dielectric target object as the other. Capacitive proximity sensors
can be a sensor element or chip, a sensor or transducer, an
instrument or meter, a gauge or indicator, a recorder or totalizer, or a controller.
Common body styles for capacitive proximity sensors are
barrel, limit switch, rectangular, slot style, and ring. A barrel
body style is cylindrical in shape, typically threaded. A limit
switch body style is similar in appearance to a contact limit
switch. The sensor is separated from the switching mechanism and provides a limit of travel detection signal. A rectangular or block body style is a one-piece rectangular or
block-shaped sensor. A slot style body is designed to detect
the presence of a vane or tab as it passes through a sensing
slot, or “U” channel. A ring-shaped body style is a doughnutshaped sensor, where the object passes through the center of
the ring. Electrical connections for capacitive proximity sensors can be fixed cable, connector(s), and terminals. A fixed
cable is an integral part of a sensor and often includes “bare”
stripped leads. A sensor with connectors has an integral connector for attaching into an existing system. A sensor with
terminals has the ability to screw or clamp down.
Important specifications for capacitive proximity sensors
include operating distance, repeatability, and switching frequency. Rated operating distance is the distance at which
switching takes place. Repeatability is the distance within
which the sensor repeatably switches. It is a measure of precision. The switching frequency is the frequency at which
the switch may be turned ON and OFF.
Other important parameters to consider when specifying
capacitive proximity sensors include housing materials,
dimensions, whether or not the sensor is shielded and
intrinsically safe, and environmental operating condition
parameters.
(b) Inductive proximity sensors and switches. Inductive proximity sensors are noncontact proximity devices that set up a
radio frequency field with an oscillator and a coil. The presence of an object alters this field and the sensor is able to
detect this alteration.
The body style of inductive proximity sensors can be barrel,
limit switch, rectangular, slot, or ring. Electrical connections
Zhang_Ch01.indd 57
5/13/2008 5:45:27 PM
58
INDUSTRIAL CONTROL TECHNOLOGY
for inductive proximity sensors can be fixed cable, connector(s),
and terminals. A fixed cable is an integral part of the sensor
and often includes “bare” stripped leads. A sensor with connectors has an integral connector for attaching into an existing system. A sensor with terminals has the ability to screw
or clamp down.
Important specifications for inductive proximity sensors
include operating distance, repeatability, field adjustability,
and minimum target distance. Rated operating distance is the
distance at which switching takes place. Repeatability is the
distance within which the sensor repeatably switches. It is a
measure of precision. Field adjustable sensors can be adjustable while in use. Depending on the sensor’s technology,
there can be minimum target size requirements.
Other important parameters to consider when specifying
inductive proximity sensors include power requirements,
housing materials, dimensions, special features, and environmental operating conditions.
(c) Photoelectric proximity sensors and switches. These sensors
utilize photoelectric emitters and receivers to detect distance,
presence, or absence of target objects. Proximity photoelectric sensors have the emitter and detector in the same housing and rely upon reflection from the surface of the target.
This mode can include presence sensing and distance measurement via analog output. The proximity category can be
further broken down into five submodes: diffuse, divergent,
convergent, fixed field, and adjustable field. A diffuse sensor
presence is detected when any portion of the diffuse reflected
signal bounces back from the detected object. Divergent
beam sensors are short-range, diffuse-type sensors without
any collimating lenses. Convergent, fixed focus, or fixed distance optics (such as lenses) are used to focus the emitter
beam at a fixed distance from the sensor. Fixed-field sensors
are designed to have a distance limit beyond which they will
not detect objects, no matter how reflective. Adjustable field
sensors utilize a cutoff distance beyond which a target will
not be detected, even if it is more reflective than the target.
Some photoelectric sensors can be set for multiple different
optical sensing modes. Reflective properties of the target and
environment are important considerations in the choice and
use of photoelectric sensors.
Important parameters to consider when looking for photoelectric sensors include sensing mode, detecting range, position measurement window, minimum detectable object, and
Zhang_Ch01.indd 58
5/13/2008 5:45:27 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
59
response time. The modes can be presence or absence sensing and position measurement. With a presence or absence
sensor, the sensor detects presence or absence in an ON/OFF
mode. In a position measurement sensing mode, the sensor
can detect position in a linear region by the intensity of
reflected light. Analog output is linear with position in the
measurement range. The detecting range is the range of sensor detection. For presence sensors, this goes up to the maximum distance for which the signal is stable. For position
measurement sensors, this is the distance range over which
the position vs output response is linear and stable. The position measurement window is the width of the linear region
for the sensor. The minimum detectable object is the smallest sized object detectable by the sensor. The response time
is the time from target object entering detection zone to the
production of the detection signal.
Common configuration features for photoelectric proximity
sensors include beam visibility, light-on or dark-on modes,
light and dark programmability, adjustable sensitivity, selfteaching, laser source, fiber-optic glass, and fiber-optic plastic.
The body style of the sensor can be threaded barrel, cylindrical,
limit switch, rectangular, slot, ring, and window or frame. The
sensor may be self-contained and may have a remote head.
(d) Ultrasonic proximity sensors and switches. Ultrasonic proximity sensing can be a sensor element or chip, a sensor or
transducer, an instrument or meter, a gauge or indicator, a
recorder or totalizer, or a controller.
The body style of the ultrasonic proximity sensors can be
barrel, limit switch, rectangular, slot, or ring. Electrical connections can be fixed cable, connector(s), or terminals. Intrinsically
safe ultrasonic proximity sensors are incapable of releasing
sufficient electrical or thermal energy under normal or abnormal conditions to cause ignition of a specific hazardous atmospheric mixture in its most ignited concentration.
Important performance specifications to consider when
searching for ultrasonic proximity sensors include maximum
operating distance, repeatability, sonic cone angle, impulse
frequency, and transmitter frequency. Load configurations
are also important parameters to consider. Ultrasonic proximity sensors may switch an AC load or a DC load.
Additional parameters that are important to consider when
searching for ultrasonic proximity sensors include switch
types, housing materials, dimensions, and environmental
operating parameters.
Zhang_Ch01.indd 59
5/13/2008 5:45:27 PM
60
INDUSTRIAL CONTROL TECHNOLOGY
(e) Magnetic proximity sensors and switches. These noncontact
proximity devices utilize inductance, Hall effect principles,
variable reluctance, or magnetoresistive technology. Magnetic
proximity sensors are characterized by the possibility of
large switching distances, available from sensors with small
dimensions. They detect magnetic objects (usually permanent
magnets), which are used to trigger the switching process.
As the magnetic fields are able to pass through many nonmagnetic materials, the switching process can also be triggered
without the need for direct exposure to the target object. By
using magnetic conductors (e.g., iron), the magnetic field can
be transmitted over greater distances so that, for example,
the signal can be carried away from high-temperature areas.
Important specifications for magnetic proximity sensors
include operating distance, repeatability, field adjustable,
and minimum target distance. Rated operating distance is the
distance at which switching takes place. Repeatability is the
distance within which the sensor repeatably switches. It is a
measure of precision. Field adjustable sensors can be adjustable while in use. Depending on the sensor’s technology,
there can be minimum target size requirements.
Other important parameters to consider when specifying
magnetic proximity sensors include power requirements,
housing materials, dimensions, special features, and environmental operating conditions.
(2) Technical types of proximity sensors and switches
(a) Eddy current proximity sensor or switch. In an eddy current
proximity sensor, electrical currents are generated in a conductive material by an induced magnetic field. Interruptions
in the flow of the electric currents (eddy currents), which are
caused by imperfections or changes in a material’s conductive properties, will cause changes in the induced magnetic
field. These changes, when detected, indicate the presence of
change in the test object. Eddy current proximity sensors and
switches detect the proximity or presence of a target by sensing the magnetic fields generated by a reference coil. They
also measure variations in the field due to the presence of
nearby conductive objects. Field generation and detection
information is provided in the kilohertz to the megahertz
range. They can be used as proximity sensors to detect presence of a target, or they can be configured to measure the
position or displacement of a target.
Target materials for eddy current proximity sensors can be
magnetic, nonmagnetic, ferrous, and nonferrous. Magnetic
Zhang_Ch01.indd 60
5/13/2008 5:45:27 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
61
target materials are magnetized, usually with a permanent
magnet component. Nonmagnetic detection targets do not
require magnetization. Ferrous targets for position detection
include iron or iron-based materials such as steel, stainless
steel, etc. Nonferrous target materials are metallic but are not
iron- or steel-based, such as aluminum, brass, and copper.
Electrical connections for eddy current proximity sensors
can be fixed cable, connector(s), and terminals. A fixed cable
is an integral part of a sensor and often includes “bare”
stripped leads. A sensor with connectors has an integral connector for attaching into an existing system. A sensor with
terminals has the ability to screw or clamp down.
Other important parameters to consider include switched
output types, position or distance output type, housing materials, dimensions, and environmental operating parameters.
(b) Air proximity sensor or switch. The air proximity sensor is a
noncontact, no moving part sensor. In the absence of an
object, air flows freely from the sensor resulting in a near
zero output signal. The presence of an object within the sensing range deflects the normal air flow and results in a positive output signal. At low supply pressure, flow from the
sensor exerts only minute forces on the object being sensed
and is consequently appropriate for use where the object
is light weight or easily marred by mechanical sensors.
Since there are no moving mechanical parts in the air proximity sensor, there are no inherent wear mechanisms or life
limitations.
In this regard, the sensor is not cycle dependent and is
particularly appropriate for applications requiring large
numbers of cycles. Also, the air proximity sensor is inherently explosion-proof and self-purging. Consequently, it is
suitable for many adverse environments.
(c) Capacitance proximity sensor or switch. Many industrial
capacitance sensors or switches work by means of the physics of capacitance. The physics of capacitance claims that
the capacitance varies inversely with the distance between
capacitor plates in this arrangement, and a certain value can
be set to trigger target detection. Note that the capacitance is
proportional to the plate area but is inversely proportional to
the distance between the plates. When the plates are close to
each other, even a small change in distance between the
plates can result in a sizeable change in capacitance.
Some of the capacitance proximity switches are tiny 1-in.
cube electronic modules that operate using a capacitance
Zhang_Ch01.indd 61
5/13/2008 5:45:27 PM
62
INDUSTRIAL CONTROL TECHNOLOGY
change technique. These sensors or switches contain two
switch outputs: a latched output, which toggles the output
ON and OFF with each cap input activation, and a momentary output, which will remain activated as long as the sensor
input capacitance is higher than the level set by the module’s
adjustment screw.
(d) Infrared proximity sensor or switch. Infrared light is beyond
the light range visible to the naked human eye and falls
between the visible light and microwave spectra (the wavelength is longer than visible light). The longest wavelengths
are red, which is where infrared got its name (“beyond red”).
Infrared waves are electromagnetic waves and may be detected
as also heat; heat from campfires, sunlight, etc., is actually
infrared radiation.
Infrared proximity sensors work by sending out a beam of
infrared light, and then computing the distance to any nearby
objects employing the characteristics of the returned signal.
(e) Fiber-optic proximity sensor or switch. Fiber-optic proximity sensors are used to detect the proximity of target objects.
Light is supplied and returned via glass fiber-optic cables.
Fiber-optic cables can fit in small spaces, are not susceptible
to electrical noise, and exhibit no danger of sparking or
shorting. Light is supplied and returned via glass fiber-optic
cables. Glass fiber exhibits very good optical qualities and
typically carries high-temperature ratings. Plastic fiber can
be cut to length in the field and can be flexible enough to
accommodate various routing configurations.
Important parameters to consider when specifying fiberoptic proximity sensors include detecting range, position measurement window, minimum detectable object, and response
time. The detecting range is the range of sensor detection.
For presence sensors, this goes up to the maximum distance
for which the signal is stable. For position measurement
sensors, this is the distance range over which the position vs
output response is linear and stable. The position measurement window is the width of linear region for the sensor. For
example, if the sensor could measure between 14 and 24 cm,
this window would be 10 cm. The minimum detectable
object is the smallest sized object detectable by the sensor.
The response time is the time from target object entering
detection zone to the production of the detection signal.
Other important parameters to consider include output
options, dimensions, electrical connections, and environmental operating conditions.
Zhang_Ch01.indd 62
5/13/2008 5:45:27 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
1.1.10
63
Scan Sensors
Scan sensors, also called image sensors or vision sensors, are built for
industrial applications. Common applications for these sensors in industrial control include alignment or guidance, assembly quality, bar or matrix
code, biotechnology or medical, color mark or color recognition, container
or product counting, edge detection, electronics or semiconductor inspection, electronics rework, flaw detection, food and beverage, gauging, scanning and dimensioning, ID detection or verification, materials analysis,
noncontact profilometry, optical character recognition, parcel or baggage
sorting, pattern recognition, pharmaceutical packaging, presence or absence,
production and quality control, seal integrity, security and biometrics, tool
and die monitoring, and web inspection.
1.1.10.1
Operating Principle
A scan or vision or image sensor can be thought of as an electronic input
device that converts analog information of a document like a map, photograph, or an overlay into an electronic image of a digital format that can be
used by the computer. Scanning is the main operation of a scan or vision or
image sensor, which automatically captures document features, text, and
symbols as individual cells, or pixels, and produces an electronic image.
While scanning, a bright white light strikes the image and is reflected
onto the photosensitive surface of the sensor. Each pixel transfers a gray
value (values given to the different shades of black in the image ranging
from 0 (black) to 255 (white), that is, 256 values to the chipset (software).
The software interprets the value in terms of 0 (black) or 1 (white), thereby
forming a monochrome image of the scanned portion. As the sensor moves
ahead, it scans the image in tiny strips and the sensor continues to store the
information in a sequential fashion. The software running the scanner
pieces together the information from the sensor into a digital form of the
image. This type of scanning is known as one-pass scanning.
Scanning a color image is slightly different as the sensor has to scan
the same image for three different colors: red, green, and blue (RGB).
Nowadays most of the color scans or vision or image sensors operate in
one-pass scanning all the three colors in one go by using color filters. In
principle, a color sensor works in the same way as a monochrome sensor.
But in this each color is constructed by mixing red, green, and blue as
given in Fig. 1.21. Thus, a 24-bit RGB sensor presents each pixel by
24 bits of information. Usually, a sensor using these three colors (in full
24 RGB mode) can create up to 16.8 million colors.
Zhang_Ch01.indd 63
5/13/2008 5:45:27 PM
64
INDUSTRIAL CONTROL TECHNOLOGY
Red
85
Green 43
Blue
6
“Brown”
Figure 1.21 The scan sensor operation of scanning a color image. In this figure,
a pixel of red = 85, and green = 43, and blue = 6 is being scanned which is
identified as “brown.”
A new technology: full width, single-line contact sensor array scanning
has emerged in which the document to be scanned passes under a line of
chips which captures the image. In this technology, a scanned line could be
considered as the cartography of the luminosity of points on the line
observed by the sensor. This new technology enables the scanner to operate at previously unattainable speeds.
(1) CCD image sensors. A charge-coupled device (CCD) gets its
name from the way the charges on its pixels are read after an
exposure. After the exposure the charges on the first row are
transferred to a place on the sensor called the read out register.
From there, the signals are fed to an amplifier and then on to an
analog-to-digital converter. Once the row has been read, its
charges on the readout register row are deleted, the next row
enters, and all of the rows above march down one row. The
charges on each row are “coupled” to those on the row above so
when one moves down, the next moves down to fill its old space.
In this way, each row can be read one row at a time. Figure 1.22
is an elucidation of the CCD scanning process.
(2) CMOS image sensors. A complementary metal oxide semiconductor (CMOS) typically has an electronic rolling shutter design.
In a CMOS sensor the data is not literally passed from bucket to
bucket. Instead, each bucket can be read independently to the
output. This has enabled designers to build an electronic rolling
slit shutter. This shutter is typically implemented by causing a reset
to an entire row and then, some time later, reading the row out.
Zhang_Ch01.indd 64
5/13/2008 5:45:27 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
65
Last row
read
First row
read
To output amplifier
Figure 1.22 The CCD image sensor shifts one whole row at a time into the readout register. The readout register then shifts one pixel at a time to the output
amplifier.
The readout speed limits the speed of a wave that passes over the
sensor from top to bottom. If the readout wave is preceded by a
similar wave of resets, then uniform exposure time for all rows is
achieved (albeit, not at the same time). With this type of electronic rolling shutter there is no need for a mechanical shutter
except in certain cases. With these advantages, CMOS image
sensors are used in some of the finest industrial control devices
or finest cameras. Figure 1.23 gives a typical architecture of
industrial control devices with CMOS image sensor.
1.1.10.2
Basic Types
All scan (or image or vision) sensors can be monochrome or color sensors. Monochrome sensing sensors present the image in black and white or
grayscales. Color sensing sensors are able to read the spectrum range using
varying combinations of different discrete colors. One common technique
is sensing the red, green, and blue components (RGB) and combining them
to create a wide spectrum of colors. Multiple chip colors are available on
some scan (image or vision) sensors. In a widely used method, the colors
are captured in multiple chips, each of them being dedicated to capturing
part of the color image, such as one color, and the results are combined to
generate the full color image. They typically employ color separation
devices such as beam-splitters rather than having integral filters on the
sensors.
Zhang_Ch01.indd 65
5/13/2008 5:45:27 PM
66
INDUSTRIAL CONTROL TECHNOLOGY
WR
RD
Address
Data
RAM
Address
Pixel data
CMOS image sensor
Pixel clock
Line valid
Frame valid
Clock out
Programmable logic
Data
RD
CS
12c data
12c clock
Figure 1.23 Block diagram of an industrial control device having CMOS image
sensor.
The imaging technology used in scan or image or vision sensors includes
CCD, CMOS, tube, and film.
(1) CCD image sensors (charge-coupled device) are electronic
devices that are capable of transforming a light pattern (image)
into an electric charge pattern (an electronic image). The CCD
consists of several individual elements that have the capability of
collecting, storing, and transporting electrical charges from one
element to another. This together with the photosensitive properties of silicon is used to design image sensors. Each photosensitive
element will then represent a picture element (pixel). With semiconductor technologies and design rules, structures are made
that form lines or matrices of pixels. One or more output amplifiers at the edge of the chip collect the signals from the CCD. An
electronic image can be obtained in this way: after having
exposed the sensor with a light pattern, apply a series of pulses
that transfer the charge of one pixel after another to the output
amplifier, line after line. The output amplifier converts the charge
into a voltage. External electronics will transform this output signal
into a form suitable for monitors or frame grabbers.
Zhang_Ch01.indd 66
5/13/2008 5:45:27 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
67
CCD image sensors have extremely low noise figures and can
also be a color sensor or a monochrome sensor.
Choices for array type include linear array, frame transfer area
array, full-frame area array, and interline transfer area array.
Digital imaging optical format is a measure of the size of the
imaging area. Optical format is used to determine what size lens
is necessary for use with the imager. Optical format refers to the
length of the diagonal of the imaging area. Optical format choices
include 1/7, 1/6, 1/5, ¼, 1/3, ½, 2/3, ¾, and 1 in. The number of
pixels and pixel size are important to consider. Horizontal pixels
refer to the number of pixels in a row of the image sensor. Vertical
pixels refer to the number of pixels in a column of the image sensor. The greater the number of pixels, the higher the resolution of
the image.
Important image sensor performance specifications to consider when searching for CCD image sensors include spectral
response, data rate, quantum efficiency, dynamic range, and
number of outputs. The spectral response is the spectral range
(wavelength range) for which the detector is designed. The data
rate is the speed of a data transfer process, normally expressed in
megahertz.
Quantum efficiency is the ratio of photon-generated electrons
that the pixel captures to the photons incident on the pixel area.
This value is wavelength dependent so the given value for quantum efficiency is generally for the peak sensitivity wavelength
for the CCD. Dynamic range is the logarithmic ratio of well
depth to the readout noise in decibels, the higher the number, the
better.
Common features for CCD image sensors include antiblooming and cooling. Some arrays for CCD image sensors offer an
optional antiblooming gate designed to bleed off overflow from
a saturated pixel. Without this feature, a bright spot, which has
saturated the pixels, will cause a vertical streak. Some arrays are
cooled for lower noise and higher sensitivity. An important environmental parameter to consider is the operating temperature.
(2) CMOS image sensors operate at lower voltages than CCD image
sensors, reducing power consumption for portable applications.
In addition to their lower power consumption when compared
with CCD image sensors, CMOS image sensors are generally of
a much simpler design: often just a crystal and decoupling. For
this reason, they are easier to design with, generally smaller, and
require less support circuitry.
Each CMOS active pixel sensor cell has its own buffer amplifier and can be addressed and read individually. A commonly
Zhang_Ch01.indd 67
5/13/2008 5:45:28 PM
68
INDUSTRIAL CONTROL TECHNOLOGY
used cell has four transistors and a photo-sensing element. The
cell has a transfer gate separating the photo sensor from a capacitive “floating diffusion,” a reset gate between the floating diffusion and power supply, a source-follower transistor to buffer the
floating diffusion from readout-line capacitance, and a row-select gate to connect the cell to the readout line. All pixels on a
column connect to a common sense amplifier. In addition to
being a color sensor or a monochrome sensor, CMOS sensors
have two categories as defined by their manner of output: analog
and digital. Analog sensors feed their encoded signal in a video
format that can be fed directly to standard video equipment.
Digital CMOS image sensors provide digital output, typically
via a 4/8- or 16-bit bus. The digital signal is direct not requiring
transference or conversion via a video capture card.
CMOS image sensors can offer many advantages over CCD
image sensors. Just some of the technical advantages of CMOS
sensors are (1) no blooming, (2) low power consumption, ideal
for battery-operated devices, (3) direct digital output (incorporates ADC and associated circuitry), (4) small size and little support circuitry, and (5) simple to design.
(3) Tube camera is also an electronic device in which the image is
formed on a fluorescent screen. It is then read by an electron
beam in a raster scan pattern and converted to a voltage proportional to the image light intensity.
(4) Film technology exposes the image onto photosensitive film,
which is developed to play or store. The shutter, a manual door
that admits light to the film, typically controls exposure.
CCD and CMOS are the important types of image sensors. A comparison of CCD and CMOS features is given in Table 1.3.
Table 1.3 Comparison of CCD and CMOS Image Sensors Features
CCD
Smallest pixel size
Low noise
Lowest dark current
–100% fill factor for full-frame CCD
Established technology market base
Highest sensitivity
Electron shutter without artifacts
Zhang_Ch01.indd 68
CMOS
Single power supply
Single master clock
Low power consumption
X,Y addressing and subsampling
Smallest system size
Easy integration of circuitry
5/13/2008 5:45:28 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
69
1.1.10.3 Technical Specifications
Inspection functions include object detection, edge detection, image
direction, alignment, object measurement, object position, bar or matrix
code, optical character recognition (OCR), and color mark or color recognition. Other parameters to consider when specifying scan or vision or
image sensors include performance features, physical features, lens mounting, shutter control, sensor specifications, dimensions, and operating environment parameters.
Sensor specifications to consider when searching for scan or image or
vision sensors include number of images stored and maximum inspection
rate. The number of images stored represents captured images that can be
stored into on-board memory or nonvolatile storage. The maximum inspection rate is the maximum number of parts or process steps that can be
inspected or evaluated per unit time. This is usually given in units of
inspections per second. Other important parameters of the sensor specifications include
(1) Image sensor resolution. Image resolution is a way of expressing how sharp or detailed images are. There are two kinds of
resolution: optical and interpolated. The optical resolution of an
image sensor is an absolute number because an image sensor’s
pixels or photo-elements are physical devices that can be
counted. To improve resolution in certain limited respects, the
optical resolution can be increased using software. This process, called interpolated resolution, adds pixels to the image to
increase the total number of pixels. To do so, software evaluates
those pixels surrounding each new pixel to determine what its
color should be. For example, if all of the pixels around a newly
inserted pixel are red, the new pixel will be made red. It is
important to keep in mind that interpolated resolution does not
add any new information to the image; it just adds pixels and
enlarges the file. This same thing can be done in a photo-editing
program such as Photoshop by resizing the image.
(2) Color depth. Resolution is not the only factor governing the
quality of your images. Equally important is color. When you
view a natural scene, or a well done photographic color print, you
are able to distinguish millions of colors. Digital images can
approximate this color realism, but whether they do so depends
on their capabilities and settings. For example, almost all newer
computer systems can display what’s called 24-bit True Color.
It’s called True Color because these systems display 16 million
colors, about the number the human eye can distinguish.
Zhang_Ch01.indd 69
5/13/2008 5:45:28 PM
70
INDUSTRIAL CONTROL TECHNOLOGY
(3) Shutter control. There is much confusion about the need for a
mechanical shutter with scan and image sensors. This discussion below clarifies the differences.
Most progressive scan and image sensors can be considered
to be long chains of pixels or buckets that are each sensitive to
light. Sensing begins by clearing the sensor, and then exposing
it to light. After the proper exposure time, the buckets pass from
pixel to pixel in a bucket brigade fashion to the output pin. If the
sensor is being exposed to light while it is passing the data, the
pixels will be polluted with additional light that confuses the
image. Therefore, this type of a progressive scan sensor requires
a mechanical shutter to block the sensor while it is shifting the
data out. Alternatively, some other progressive scan and image
sensors have an entire frame store. This frame store is a place to
save the contents of each pixel in a shielded bucket that is insensitive to light. The shielded buckets are then passed to the output pin and are, by their nature, not corrupted by additional
light. Unfortunately, these sensors must be almost twice the size
of their counterparts without the shielded frame storage area.
So, it is important to compare the cost of the larger sensor to the
savings associated with not needing a mechanical shutter. Both
of these implementations can be used for flash photography since
the flash can be fired when the sensor is in integration mode.
Interlaced scan and image sensors are similar to progressive
sensors except that there is a shielded row store located between
each pair of odd and even rows. For normal use, alternating
frames copy either the odd or even rows into the shielded store
prior to shifting while the other row (even or odd) is being
exposed. In this way there is the advantage of not requiring a
mechanical shutter. However, note that each frame has only half
the vertical resolution of the entire sensor. This is the type of
sensor commonly found in television cameras. When the flash
fires only one-half of the rows will be in integration mode. So,
this type of sensor cannot be used for flash photography unless
you are willing to limit the vertical resolution.
(4) Sensitivity. An International Organization for Standardization
(ISO) number that appears on the film package specifies the
speed, or sensitivity, of a silver-based film. The higher the number the “faster” or more sensitive the film is to light. Each doubling of the ISO number indicates a doubling in film speed so
each of these films is twice as fast as the next fastest.
Image sensors are also rated using equivalent ISO numbers.
Just as with film, an image sensor with a lower ISO needs more
light for a good exposure than one with a higher ISO. All things
Zhang_Ch01.indd 70
5/13/2008 5:45:28 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
71
being equal, it is better to get an image sensor with a higher ISO
because it will enhance freezing motion and shooting in low
light. Typically, ISOs range from 100 (fairly slow) to 3200 or
higher (very fast).
Some cameras have more than one ISO rating. In low-light
situations, you can increase the sensor’s ISO by amplifying the
image sensor’s signal (increasing its gain). Some cameras even
increase the gain automatically. This not only increases the sensor’s sensitivity, it also increases the noise or “grain,” making
the images softer and less sharp.
(5) Aspect ratio. Image sensors have different aspect ratios—the
ratio of image height to width. The ratio of a square is 1:1 (equal
width and height) and that of 35 mm film is 1.5:1 (1.5 times
wider than it is high). Most image sensors fall in between these
extremes. The aspect ratio of a sensor is important because it
determines the shape and proportions of the photographs you
create. When an image has a different aspect ratio than the
device it’s displayed or printed on, it has to be cropped or
resized to fit. Your choice is to lose part of the image or waste
part of the paper. To imagine this better, try fitting a square
image on a rectangular piece of paper.
The aspect ratio of an image sensor determines the shape of
your prints. An image will only perfectly fill a sheet of paper if
both have the same aspect ratio. If the ratios are different, you
have to choose between losing part of the image or leaving
some white space on the paper.
To calculate the aspect ratio of any camera, divide the largest
number in its resolution by the smallest number. For example,
if a sensor has a resolution of 3000 × 2000, divide 3000 by
2000. In this case the aspect ratio is 1.5, the same as 35 mm film.
(6) Dynamic range. Dynamic range is the ratio of signal to noise of
an image sensor.
(7) Package. The sensor package is often neglected in sensor selection. It is common in smaller sensors for the package cost to be
at least half of the total cost of the product. Sensor packages are
costly because they are produced in relatively low volumes,
have optical quality glass tops, must be dirt free, have low
humidity, and require precise die positioning.
(8) Image quality. The size of an image file depends in part on the
resolution of the image. The higher the resolution, the more
pixels there are to store, so the larger the image file becomes. To
make large image files smaller and more manageable most
cameras store images in a format called JPEG after its developer, the Joint Photographic Experts Group. This file format not
Zhang_Ch01.indd 71
5/13/2008 5:45:28 PM
72
INDUSTRIAL CONTROL TECHNOLOGY
only compresses images, but it also allows specifying how
much they are compressed. This is a useful feature because
there is a trade-off between compression and image quality.
Less compression gives better images but cannot store as many
images. More compression allows storing more images, but the
only problem is that image quality would not be as good.
(9) Frame rate. Most digital cameras have automatic exposure controls. There are two delays built into digital cameras that affect
the ability to respond to fast action when taking pictures.
The first brief delay is between pressing the shutter button
and actually capturing the image. This delay, called the refresh
rate, occurs because the camera clears the image sensor, sets
white balance to correct for color, sets the exposure, and focuses
the image. Finally, it fires the flash if it is needed and takes the
picture.
The second delay, the recycle time, occurs when the captured
image is processed and stored. This delay can range from a few
seconds to half a minute.
Both of these delays affect how quickly a series of photos
can be taken one after another, called the frame rate, shot-toshot rate, or click-to-click rate.
(10) Clocking and power supply design. This specifies the requirement for a wider variety of power supply voltages and clocks or
single power supply voltage and includes clocking.
1.1.11
Force and Load Sensors
Force and load sensors cover electrical sensing devices that are used to
measure tension, compression, and shear forces. Tension cells are used for
measurement of a straight line force “pulling apart” along a single axis,
typically annotated as positive force. Compression tension cells are used
for measurement of a straight line force “pushing together” along a single
axis, typically annotated as negative force. Shear is induced by tension or
compression along offset axes. They are manufactured in many different
packages and mounting configurations. In industries, force and load sensors are used for security devices, packaging devices, production automation, etc.
1.1.11.1
Operating Principle
The most common technologies for the operation of force and load sensors
are various types of strain gauges. Strain gauges are measuring elements
Zhang_Ch01.indd 72
5/13/2008 5:45:28 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
73
that convert force, pressure, tension, etc., into an electrical signal. They are
the most universally used measuring devices for electrical measurement of
mechanical quantities. A strain gauge is a resistive elastic sensor whose
resistance is a function of applied strain (unit deformation).
Many types of strain gauges exit, depending on the electrical resistance
to strain. These types include piezoresistive or semiconductor, carbonresistive, bonded metallic wire, and foil gauges. The most widely used
characteristic that varies in proportion to strain is electrical resistance.
In these gauges the electrical resistance varies linearly with strain.
The resistance of an electrically conductive material changes with dimensional changes that take place when the conductor is deformed elastically.
When such a material is stretched, the conductors become longer and narrower, which causes an increase in resistance. A Wheatstone bridge then
converts this change in resistance to an absolute voltage. The resulting
value is linearly related to strain by a constant called the gauge factor.
Capacitance devices, which depend on geometric features, can be used to
measure strain. Changing the plate area or the gap can vary the capacitance. The electrical properties of the materials used to form the capacitor
are relatively unimportant, so capacitance strain gauge materials can be
chosen to meet the mechanical requirements. This allows the gauges to
be more rugged, providing a significant advantage over resistance strain
gauges.
(1) The strain gauge. Strain is the amount of deformation of a body
due to an applied force. More specifically, strain is defined as the
fractional change in length. While there are several methods of
measuring strain, the most common is with a strain gauge, a
device whose electrical resistance varies in proportion to the
amount of strain in the device. The most widely used gauge,
however, is the bonded metallic strain gauge.
The metallic strain gauge consists of a very fine wire or, more
commonly, metallic foil arranged in a grid pattern. The grid pattern maximizes the amount of metallic wire or foil subject to
strain in the parallel direction (Fig. 1.24). The cross-sectional
area of the grid is minimized to reduce the effect of shear strain
and Poisson strain. The grid is bonded to a thin backing, called
the carrier, which is attached directly to the test specimen.
Therefore, the strain experienced by the test specimen is transferred directly to the strain gauge, which responds with a linear
change in electrical resistance. Strain gauges are available commercially with nominal resistance values from 30 to 3000 Ω,
with 120, 350, and 1000 Ω being the most common values.
Zhang_Ch01.indd 73
5/13/2008 5:45:28 PM
74
INDUSTRIAL CONTROL TECHNOLOGY
Alignment marks
Solder tabs
Active grid
length
Carrier
Figure 1.24 Bonded metallic strain gauge.
It is very important that the strain gauge be properly mounted
onto the test specimen so that the strain is accurately transferred
from the test specimen, through the adhesive and strain gauge
backing, to the foil itself. Manufacturers of strain gauges are the
best source of information on proper mounting of strain gauges.
(2) Strain gauge measurement. In practice, the strain measurements
rarely involve quantities larger than a few millistrain ε × 10−3.
To measure such small changes in resistance, and compensate
for the temperature sensitivity, proper selection and use of the
bridge, signal conditioning, wiring, and data acquisition components are required for reliable measurements.
(a) Bridge completion. Unless you are using a full-bridge strain
gauge sensor with four active gauges, you will need to complete the bridge with reference resistors. Therefore, strain
gauge signal conditioners typically provide half-bridge completion networks consisting of two high-precision reference
resistors. Figure 1.25 diagrams the wiring of a half-bridge
strain gauge circuit to a conditioner with completion resistors R1 and R2. The nominal resistance of the completion
resistors is less important than how well the two resistors are
matched. Ideally, the resistors are well matched and provide
a stable reference voltage of VEX/2 to the negative input lead
of the measurement channel. For example, the half-bridge
Zhang_Ch01.indd 74
5/13/2008 5:45:28 PM
75
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
EC+
R1
VEX
+
In+
+
–
–
R2
RG
In–
EC–
RG
Strain
gauges
Signal conditioner
Figure 1.25 Connection of half-bridge strain gauge circuit.
completion resistors provided on the SCXI-1122 signal
conditioning module are 2.5 kΩ resistors, with a ratio tolerance of 0.02%. The high resistance of the completion resistors helps minimize the current drawn from the excitation
voltage.
(b) Bridge excitation. Strain gauge signal conditioners typically
provide a constant voltage source to power the bridge.
While there is no standard voltage level that is recognized
industry-wide, excitation voltage levels of around 3 and 10 V
are common. While a higher excitation voltage generates a
proportionately higher output voltage, the higher voltage can
also cause larger errors due to self-heating. Again, it is very
important that the excitation voltage be very accurate and
stable. Alternatively, one can use a less accurate or stable
voltage, and accurately measure, or sense, the excitation
voltage so the correct strain can be calculated.
(c) Excitation sensing. If the strain gauge circuit is located at a
distance away from the signal conditioner and excitation
source, a possible source of error is voltage drop caused by
resistance in the wires connecting the excitation voltage to
the bridge. Therefore, some signal conditioners include a
feature called remote sensing to compensate for this error.
There are two common methods of remote sensing. With
feedback remote sensing, you connect extra sense wires to
the point where the excitation voltage wires connect to the
bridge circuit. The extra sense wires serve to regulate the
excitation supply to compensate for lead losses and deliver
the needed voltage at the bridge. This scheme is used with
the SCXI-1122.
An alternative remote sensing scheme uses a separate measurement channel to measure directly the excitation voltage
Zhang_Ch01.indd 75
5/13/2008 5:45:29 PM
76
INDUSTRIAL CONTROL TECHNOLOGY
delivered across the bridge. Because the measurement channel leads carry very little current, the lead resistance has negligible effect on the measurement. The measured excitation
voltage is then used in the voltage-to-strain conversion to
compensate for lead losses.
(d) Signal amplification. The output of strain gauges and bridges
is relatively small. Therefore, strain gauge signal conditioners usually include amplifiers to boost the signal level to
increase measurement resolution and improve signal-to-noise
ratios. SCXI signal conditioning modules, for example,
include configurable gain amplifiers with gains up to 2000.
(e) Bridge balancing, offset nulling. When a bridge is installed,
it is very unlikely that the bridge will output exactly 0 V when
no strain is applied. Rather, slight variations in resistance
among the bridge arms and lead resistance will generate
some nonzero initial offset voltage. There are a few different
ways that a system can handle this initial offset voltage:
(i) Software compensation. The first method compensates
for the initial voltage in software. With this method, you
take an initial measurement before strain input is applied.
This initial voltage is then used in the strain equations
listed at the end of this application note. This method is
simple, fast, and requires no manual adjustments. The
disadvantage of the software compensation method is
that the offset of the bridge is not removed. If the offset
is large enough, it limits the amplifier gain you can
apply to the output voltage, thus limiting the dynamic
range of the measurement.
(ii) Offset nulling circuit. The second balancing method uses
an adjustable resistance, or potentiometer, to physically
adjust the output of the bridge to zero. For example,
Fig. 1.26 illustrates the offset nulling circuit of the
SCXI-1321 terminal block. By varying the position of
R1
+
–
VEX
RPOT
R4
–
RNULL
R2
VO
+
R3
Figure 1.26 Offset nulling circuit of SCXI-1321 terminal block.
Zhang_Ch01.indd 76
5/13/2008 5:45:29 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
77
the potentiometer (RPOT), you can control the level of
the bridge output and set the initial output to 0 V. The
value of RNULL sets the range that the circuit can balance. On the SCXI-1321, this resistor is socked for easy
adjustment of the balancing range.
(iii) Buffered offset nulling. The third method, like the software method, does not affect the bridge directly. With
buffered nulling, a nulling circuit adds an adjustable DC
voltage to the output of the instrumentation amplifier.
For example, the SC-2043-SG strain gauge accessory
uses this method. The SC-2043-SG includes a useradjustable potentiometer that can add ±50 mV to the
output of an instrumentation amplifier that has a fixed
gain of 10. Therefore, the nulling range, referred to
input, is ±5 mV.
(f) Shunt calibration. The normal procedure to verify the output
of a strain gauge measurement system relative to some predetermined mechanical input or strain is called shunt calibration. Shunt calibration involves simulating the input of
strain by changing the resistance of an arm in the bridge by
some known amount. Shunting, or connecting, a large resistor of known value accomplishes this across one arm of the
bridge, creating a known drift resistance. The output of the
bridge can then be measured and compared to the expected
voltage value. The results can then be used to correct span
errors in the entire measurement path, or to simply verify
general operation to gain confidence in the setup.
1.1.11.2
Basic Types
(1) Force and load sensors. Force and load sensors can be devices of
many different types including sensor element or chip, sensor or
transducer, instrument or meter, gauge or indicator, and recorder
and totalizers. A sensor element or chip denotes a “raw” device
such as a strain gauge, or one with no integral signal conditioning or packaging. A sensor or transducer is a more complex
device with packaging and/or signal conditioning that is powered and provides an output such a DC voltage, a 4–20 mA current loop, etc. An instrument or meter is a self-contained unit that
provides an output such as a display locally at or near the device.
Typically it also includes signal processing and/or conditioning.
A gauge or indicator is a device that has a (usually analog) display and no electronic output such as a tension gauge. A recorder
Zhang_Ch01.indd 77
5/13/2008 5:45:29 PM
78
INDUSTRIAL CONTROL TECHNOLOGY
or totalizer is an instrument that records, totalizes, or tracks force
measurement over time. It includes simple data logging capability or advanced features such as mathematical functions, graphing, etc.
Features common to force and load sensors include biaxial
measurement, triaxial measurement, and temperature compensation. Biaxial load cells can provide load measurements along
two, typically orthogonal, axes. Triaxial load cells can provide
load measurements along three, typically orthogonal, axes.
Temperature compensated load cells provide special circuitry to
reduce/eliminate sensing errors due to temperature variations.
Other parameters to consider include operating temperature,
maximum shock, and maximum vibration.
(2) Strain gauges. The most common technology used by force and
load sensors is the principle of strain gauges.
In a photoelectric strain gauge a beam of light is passed
through a variable slit, actuated by the extensometer, and directed
to a photoelectric cell. As the gap opening changes, the amount
of light reaching the cell varies, causing a varying intensity in the
current generated by the cell. Semiconductor or piezoelectric
strain gauges are constructed of ferroelectric materials. In ferroelectric materials, such as crystalline quartz, a change in the
electronic charge across the faces of the crystal occurs when
the material is mechanically stressed. The piezoresistive effect
is defined as the change in resistance of a material due to an
applied stress, and this term is used commonly in connection
with semiconducting materials. Optical strain gauge types include
photoelastic, moiré interferometer, and holographic interferometer strain gauges. In a fiber-optic strain gauge, the sensor
measures the strain by shifting the light frequency of the light
reflected down the fiber from the Bragg grating, which is embedded inside the fiber itself.
The gauge pattern refers cumulatively to the shape of the grid,
the number and orientation of the grids in a multiple grid (rosette)
gauge, the solder tab configuration, and various construction features that are standard for a particular pattern. Arrangement types
include uniaxial, dual linear, strip gauges, diaphragm, tee rosette,
rectangular rosette, and delta rosette. Specialty applications for
strain gauges include crack detection, crack propagation, extensometer, temperature measurement, residual stress, shear modulus gauge, and transducer gauge.
The three primary specifications when selecting strain gauges
are operating temperature, the stat of the strain (including
Zhang_Ch01.indd 78
5/13/2008 5:45:29 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
79
gradient, magnitude, and time dependence), and the stability
required by the application. The operating temperature range is
the range of ambient temperature where the use of the strain
gauge is permitted without permanent changes of the measurement properties. Other important parameters to consider include
the active gauge length, the gauge factor, nominal resistance, and
strain sensitive material. The gauge length of a strain gauge is the
active or strain-sensitive length of the grid. The end loops and
solder tabs are considered insensitive to strain because of their
relatively large cross-sectional area and low electrical resistance.
The strain sensitivity, k, of a strain gauge is the proportionality
factor between the relative changes of the resistance. The strain
sensitivity is a figure without dimension and is generally called
gauge factor. The resistance of a strain gauge is defined as the
electrical resistance measured between the two metal ribbons or
contact areas intended for the connection of measurement cables.
The principal component that determines the operating characteristics of a strain gauge is the strain-sensitive material used in
the foil grid.
1.1.11.3 Technical Specifications
Important parameters for force and load sensors include the force
and load measurement range and the accuracy. The measurement range is
the range of required linear output. Most force sensors actually measure
the displacement of a structural element to determine force. The force is
associated with a deflection as a result of calibration. There are many form
factors or packages to choose from: S-beam, pancake, donut or washer,
plate or platform, bolt, link, miniature, cantilever, canister, load pin, rod
end, and tank weighing. Shear cell type can be shear beam, bending beam,
or single point bending beam. Force and load sensors can have one of
many output types. These include analog voltage, analog current, analog
frequency, switch or alarm, serial, and parallel.
(1) Force and load sensor specifications
(a) Force to measure
(i) Force/load measurement range. The range required of
linear output.
Search Logic: User may specify either, both, or neither of the “At Least” and “No More Than” values.
Zhang_Ch01.indd 79
5/13/2008 5:45:29 PM
80
INDUSTRIAL CONTROL TECHNOLOGY
Products returned as matches will meet all specified
criteria.
(ii) Accuracy. The accuracy required of the device.
Search Logic: All matching products will have a
value less than or equal to the specified value.
(b) Force sensor type
(i) Tension. Tension cell for measurement of a straight
line force “pulling apart” is along a single axis; typically annotated as positive force.
(ii) Compression. Tension cell for measurement of a
straight line force “pushing together” is along a single
axis; typically annotated as negative force.
(iii) Shear. Shear is induced by tension or compression
along offset axes.
Search Logic: All products with ANY of the selected
attributes will be returned as matches. Leaving all
boxes unchecked will not limit the search criteria for
this question; products with all attribute options will
be returned as matches.
(c) Load cell package. Most force sensors actually measure the
displacement of a structural element to determine force. The
force is associated with a deflection as a result of calibration.
There are many form factors or packages to choose from:
S-beam, pancake, donut or washer, plate or platform, bolt,
link, miniature, cantilever, canister, load pin, rod end, and
tank weighing. Your choices are
(i) S-beam. S-beam units are shaped like a squared-off S.
Variable resistors (whose resistance is a function of
strain induced by the load, e.g., strain gauges or
piezoresistive elements) are bonded to the regions of
maximum strain and change resistance as a load is
applied. These resistance changes are typically measured in a Wheatstone bridge circuit.
(ii) Pancake. Pancake cells are similarly instrumented
short, low-profile cylinders. They are quite popular
and are capable of measuring very small through very
large loads.
(iii) Donut/washer. Donut cells are like pancake cells but
with a through bore or hole.
(iv) Plate/platform
(v) Bolt. Load sensor with one or two threaded ends for
attachment to measured system; measures force along
the long axis of the bolt.
Zhang_Ch01.indd 80
5/13/2008 5:45:29 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
81
(vi) Link. Sensor for inline measurement of tension or
compression.
(vii) Miniature. Miniature is a characteristic assigned by
the supplier. If the supplier considers the load cell
to be miniature, it will be carried in the database as
such.
(viii) Cantilever. Cantilever units are designed to have the
load applied to a cantilever that is typically instrumented at the base.
(ix) Canister. Canister cells are instrumented cylinders.
They are of considerably higher aspect ratio than pancake cells.
(x) Load pin. Load pins are typically instrumented shear
elements that undergo a strain when a load is applied.
A typical application is an instrumented wrist pin that
attaches a hook to a cable on a crane.
(xi) Rod end—male. Rod end—male is an instrumented
rod with male threads.
(xii) Rod end—female. Rod end—female is an instrumented
rod with female threads.
(xiii) Tank weighing. Tank weighing load cells are specially
designed to support tanks.
Search Logic: All products with ANY of the selected
attributes will be returned as matches. Leaving all
boxes unchecked will not limit the search criteria for
this question; products with all attribute options will
be returned as matches.
(d) Sensor output. It includes the type of electrical signal that
will be produced.
(i) Analog voltage. Output voltage is a simple (usually
linear) function of the measurement, including voltage
ranges such as 0–10 V, 5 V, and voltage ratios dependent upon excitation, typically expressed as millivolts
per volt.
(ii) Analog current. Often called a transmitter, a current is
imposed on the output circuit proportional to the measurement; typical ranges are 4–20 mA, 0–50 mA, etc.
Feedback is used to provide the appropriate current
regardless of line noise, impedance, etc. Current outputs are often useful when sending signals over long
distances.
(iii) Analog frequency. The output signal is encoded via
amplitude modulation (AM), frequency modulation
Zhang_Ch01.indd 81
5/13/2008 5:45:29 PM
82
INDUSTRIAL CONTROL TECHNOLOGY
(FM), or some other modulation scheme; the signal is
analog in nature.
(iv) Switch/alarm. Sensor triggers on a sensed force level
to close or open a switch, or provide a signal to an
alarm or interlock.
(v) Serial is a standard serial digital output protocol such
as RS232, RS422, RS485, USB, etc.
(vi) Parallel is a standard parallel digital output protocol
such as IEEE 488, a Centronics or printer port, etc.
(vii) Other. Any digital outputs other than the standard
serial or parallel signals. Simple TTL logic signals are
an example.
Search Logic: All products with ANY of the selected
attributes will be returned as matches. Leaving all
boxes unchecked will not limit the search criteria for
this question; products with all attribute options will
be returned as matches.
(e) Category of devices. You can think of products as belonging
to general categories based on what they are designed to do
and what you have to do to use them. The “category” criteria
attempts to distinguish “unpacked” sensors that might be
used as part of a larger sensor from, say, a gauge which can
be read just by looking at it.
(i) Sensor element/chip denotes a “raw” device such as a
strain gauge or one with no integral signal conditioning or packaging.
(ii) Sensor/transducer. A more complex device with packaging and/or signal conditioning that is powered and
provides an output such a DC voltage, a 4–20 mA current loop, etc.
(iii) Instrument/meter is a self-contained unit that provides
an output such as a display locally at or near the device.
Typically it also includes signal processing and/or
conditioning.
(iv) Gauge/indicator. A device that has a (usually analog)
display and no electronic output such as a tension
gauge.
(v) Recorder/totalizer is an instrument that records, totalizes, or tracks force measurement over time including
simple data logging capability or advanced features
such as mathematical functions, graphing, etc.
Search Logic: All products with ANY of the selected
attributes will be returned as matches. Leaving all boxes
unchecked will not limit the search criteria for this
Zhang_Ch01.indd 82
5/13/2008 5:45:29 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
83
question; products with all attribute options will be
returned as matches.
(f) Sensor technology
(i) Piezoelectric. For piezoelectric devices, a piezoelectric material is compressed and generates a charge that
is conditioned by a charge amplifier.
(ii) Strain gauge. For strain gauge devices, strain gauges
(strain-sensitive variable resistors) are bonded to parts
of the structure that deform when making the measurement. These strain gauges are typically used as
elements in a Wheatstone bridge circuit, which is used
to make the measurement. Strain gauges typically
require an excitation voltage and provide output sensitivity proportional to that excitation.
Search Logic: All products with ANY of the selected
attributes will be returned as matches. Leaving all
boxes unchecked will not limit the search criteria for
this question; products with all attribute options will
be returned as matches.
(g) Features
(i) Biaxial measurement. Biaxial load cells can provide
load measurements along two, typically orthogonal,
axes.
Search Logic: “Required” and “Must Not Have” criteria limit returned matches as specified. Products with
optional attributes will be returned for either choice.
(ii) Triaxial measurement. Triaxial load cells can provide
load measurements along three, typically orthogonal,
axes.
Search Logic: “Required” and “Must Not Have” criteria limit returned matches as specified. Products with
optional attributes will be returned for either choice.
(iii) Temperature compensation. Temperature compensated
load cells provide special circuitry to reduce/eliminate
sensing errors due to temperature variations.
Search Logic: “Required” and “Must Not Have” criteria limit returned matches as specified. Products with
optional attributes will be returned for either choice.
(2) Strain gauge specifications
(a) Construction
(i) Electrical resistance. The resistance of an electrically
conductive material changes with dimensional changes
that take place when the conductor is deformed elastically. When such a material is stretched, the conductors
Zhang_Ch01.indd 83
5/13/2008 5:45:29 PM
84
INDUSTRIAL CONTROL TECHNOLOGY
become longer and narrower, which causes an increase
in resistance. A Wheatstone bridge then converts this
change in resistance to an absolute voltage. The resulting value is linearly related to strain by a constant
called the gauge factor.
(ii) Capacitance. Capacitance devices, which depend on
geometric features, can be used to measure strain.
Changing the plate area or the gap can vary the capacitance. The electrical properties of the materials used
to form the capacitor are relatively unimportant, so
capacitance strain gauge materials can be chosen to
meet the mechanical requirements. This allows the
gauges to be more rugged, providing a significant
advantage over resistance strain gauges.
(iii) Photoelectric. A beam of light is passed through
a variable slit, actuated by the extensometer, and
directed to a photoelectric cell. As the gap opening
changes, the amount of light reaching the cell varies,
causing a varying intensity in the current generated by
the cell.
(iv) Semiconductor (piezoresistive). In ferroelectric materials, such as crystalline quartz, a change in the
electronic charge across the faces of the crystal occurs
when the material is mechanically stressed. The piezoresistive effect is defined as the change in resistance of
a material due to an applied stress, and this term is
used commonly in connection with semiconducting
materials.
The resistivity of a semiconductor is inversely proportional to the product of the electronic charge, the
number of charge carriers, and their average mobility.
The effect of applied stress is to change both the
number and average mobility of the charge carriers.
By choosing the correct crystallographic orientation
and dopant type, both positive and negative gauge factors may be obtained. Silicon is now almost universally used for the manufacture of semiconductor strain
gauges.
(v) Optical photoelastic strain gauges. When a photoelastic material is subjected to a load and illuminated with
polarized light from the measurement instrumentation
(called a reflection polariscope), patterns of color
appear which are directly proportional to the stresses
Zhang_Ch01.indd 84
5/13/2008 5:45:29 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
85
and strains within the material. The sequence of colors
observed as stress increases is black (zero stress),
yellow, red, blue, green, etc. The transition lines seen
between the red and green bands are known as
“fringes.” The stresses in the material increase proportionally as the number of fringes increases. A closely
spaced fringe means a steeper stress gradient, and uniform color represents a uniformly stressed area. Hence,
the overall stress distribution can easily be studied
by observing the numerical order and spacing of the
fringes. Furthermore, a quantitative analysis of the
direction and magnitude of the strain at any point on
the coated surface can be performed with the Polaris
reflection and a digital strain indicator.
(vi) Moire interferometry strain gauges. Moire interferometry is an optical technique that uses coherent laser
light to produce a high contrast, two-beam optical
interference pattern. Moire interferometry reveals
planar displacement fields on a part’s surface, which
is caused by external loading or other source deformation. It responds only to geometric changes of
the specimen and is effective for diverse engineering
materials. Contour maps of planar deformation fields
can be generated from x and y components of
displacements.
(vii) Holographic interferometry strain gauges. Holographic
interferometry allows the evaluation of strain, rotation,
bending, and torsion of an object in three dimensions.
Since holography is sensitive to the surface effects
of an opaque body, extrapolation into the interior of
the body is possible in some circumstances. In one
or more double-exposure holograms, changes in the
object are recorded. From the fringe patterns in the
reconstructed image of the object, the interference
phase-shifts for different sensitivity vectors are measured. A computer is then used to calculate the strain
and other deformations.
(viii) Fiber optic. The sensor measures the strain by shifting
the light frequency of the light reflected down the fiber
from the Bragg grating, which is embedded inside the
fiber itself. Since it is possible to put several sensors on
the same fiber, the amount of cabling required is reduced
significantly compared to other types of strain gauges.
Zhang_Ch01.indd 85
5/13/2008 5:45:29 PM
86
INDUSTRIAL CONTROL TECHNOLOGY
Also, since the signal is optical rather than electronic, it
is not affected by electromagnetic interference.
(ix) Other. Other is unlisted, specialized, or proprietary
strain gauge construction.
Search Logic: All products with ANY of the selected
attributes will be returned as matches. Leaving all
boxes unchecked will not limit the search criteria for
this question; products with all attribute options will
be returned as matches.
(b) Physical specifications
(i) Active gauge length (grid length). The gauge length of
a strain gauge is the active or strain-sensitive length of
the grid. The end loops and solder tabs are considered
insensitive to strain because of their relatively large
cross-sectional area and low electrical resistance.
Search Logic: User may specify either, both, or neither of the “At Least” and “No More Than” values.
Products returned as matches will meet all specified
criteria.
(ii) Number of gauges in gauge pattern. The total number
of strain gauges in the gauge pattern.
Search Logic: User may specify either, both, or neither
of the “At Least” and “No More Than” values. Products
returned as matches will meet all specified criteria.
(iii) Operating temperature. The operating temperature
range is the range of ambient temperature where the
use of the strain gauge is permitted without permanent
changes of the measurement properties.
Search Logic: User may specify either, both, or neither
of the limits in a “From–To” range; when both are specified, matching products will cover entire range. Products
returned as matches will meet all specified criteria.
(iv) Gauge factor. The strain sensitivity, k, of a strain gauge
is the proportionality factor between the relative changes
of the resistance. The strain sensitivity is a figure without dimension and is generally called gauge factor.
Search Logic: All matching products will have a
value greater than or equal to the specified value.
1.1.11.4
Calibration
Calibration and temperature compensation of strain gauge–based weight
sensors using the precision sensor signal conditioner integrated circuits is
Zhang_Ch01.indd 86
5/13/2008 5:45:30 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
87
becoming increasingly popular. Leading automotive suppliers of safety
products are increasingly turning to the use of force sensors in their quest
to optimize airbag deployment forces appropriate for the mass of the occupant and severity of the deployment situation.
A typical weight sensor characterization requires the use of a deadweight test stand (also sometimes called a creep tester) in order to obtain
accurate and repeatable sensor loading. Several points must be observed in
order to achieve the desired results.
(1) fixture orientation for positive and negative force loads on the
test stand;
(2) sensor and cable orientation;
(3) torque of the sensor mounting bolts;
(4) weight numbering;
(5) shaft-to-weight clearance (when using an automatic weight lift
for removing weights from load);
(6) preconditioning sensor and fixtures after sensor mounting;
(7) monotonic application and removal of weights;
(8) golden samples and data tracking.
1.2 Actuators
An actuator is simply defined as a device that produces a linear, rotary
motion from a source of power under the action of a source of control. The
sources of power driving actuators can be from electric, pneumatic, fluid,
and piezoelectric, or some others such as our hands. Actuators are accordingly classed into electric actuators, pneumatic actuators, hydraulic actuators, piezoelectric actuators, as well as manual actuators, based on the
applied sources of power.
Basic actuators are used to move valves or switches to either fully
opened or fully closed positions. Actuators for control or position regulating valves or switches are also given a positioning signal to move to
any intermediate position with a high degree of accuracy. Although
the most common and important use of an actuator is to open and close
valves or switches, current actuator designs go far beyond the basic open
and close function to implement more and more positioning functions.
Therefore, in some instances of industrial control technology, actuators
are also termed as positioners. In addition to directly positioning, an actuator can be packaged together with position sensing equipment, torque
sensing, motor protection, logic control, digital communication capacity,
Zhang_Ch01.indd 87
5/13/2008 5:45:30 PM
88
INDUSTRIAL CONTROL TECHNOLOGY
and even PID control to play a role as a position detector or position
indicator.
1.2.1
Electric Actuators
Electric actuators, utilizing the simplicity of electrical operation, provide the most reliable means of positioning a valve in a safe condition
including fail-safe to close or open, or lock in position on power or system
failure. However, electric actuators are not restricted to open or close
applications; with the addition of one or more of the available kit options,
the requirements for fully fledged control units can often be met. For
example, with both weatherproof and flameproof models, the range simplifies process automation by providing true electronic control from process variable to valve and supplies a totally electric system for all
environments. The unit can be supplied with the appropriate electronic
controls to match any process control system requirement. Electric actuators are actively marketed and considered as replacements for pneumatic
actuators. Although pneumatic actuators are still an important method of
actuation, more and more often, electric actuators provide a superior solution, especially when high accuracy, high duty cycle, excellent reliability,
long life expectancy, and low maintenance are required by extra switches,
speed controllers, potentiometers, position transmitters, positioners, and
local control station. These options may be added to factory-built units, or
supplied in kit form. When supplied as kits all parts are included together
with an easy to follow installation sheet.
1.2.1.1
Operating Principle
Architecture for an electric actuator is given in Fig. 1.27, which consists
basically of gears, a motor, and switches. In these components, the motor
plays a key role. In most applications, the motor is the primary torquegenerating component. Motors are available for a variety of supply voltages, including standard single-phase alternating current, three-phase and
DC voltages. In some applications, three-phase current for the asynchronous is generated by means of the power circuit module in the electronic,
regardless of the power supply (one- or three phase). Frequency converters
and microcontrollers allow different speeds and precise tripping torques to
be set (no overtorque). When an electric actuator is running, the phase
angle is checked and automatically adjusted so that the rotation is always
correct. To prevent heat damage due to excessive current draw in a stalled
Zhang_Ch01.indd 88
5/13/2008 5:45:30 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
Gears
89
Motor
Hand
wheel
Switches
Figure 1.27 Basic components for electric actuator (courtesy of Emerson).
condition, or due to overwork, electric actuator motors usually include a
thermal overload sensor or switch embedded in the winding of the stator.
The sensor or switch is installed in series with the power source, and opens
the circuit when the motor is overheated, and then closes the circuit once
it has cooled to a safe operating temperature.
Electric actuators rely on a gear train (a series of interconnected gears)
to enhance the motor torque and to regulate the output speed of the actuator. Some gear styles are inherently self-locking. This is particularly
important in the automation of butterfly valves or when an electric actuator
is used in modulating control applications. In these situations, seat and
disc contact, or fluid velocity, act upon the closure element of the valve
and cause a reverse force that can reverse the motor and camshaft. This
causes a reenergization of the motor through the limit switch when the
cam position is changed. This undesirable cycling will continue to occur
unless a motor brake is installed, and usually leads to an overheated motor.
Spur gears are sometimes used in rotary electric actuators, but are not selflocking. They require the addition of an electromechanical motor brake for
these applications.
A few of the self-locking gear styles include the worm and wheel and
some configurations of planetary gears. A basic worm gear system operates as follows. A motor applies a force through the primary worm gear to
the worm wheel. This, in turn, rotates the secondary worm gear which
applies a force to the larger radius of the secondary worm wheel to increase
the torque.
Zhang_Ch01.indd 89
5/13/2008 5:45:30 PM
90
INDUSTRIAL CONTROL TECHNOLOGY
An electric actuator system replacing the traditional hydraulic piston
system is given in Fig. 1.28: it is made up of a motor, a reducer, and a ball
screw. The motor receives its power from an electrical unit (no longer generated hydraulically) situated in the aircraft’s avionics bay. Electric cables
that carry the power to the motor have replaced hydraulic pipes. The control unit is able to directly and individually transmit the braking order to
each brake, thus optimizing both braking and the use of each brake in
operation. Electric brake technology offers aircraft manufacturers and airline companies significant gains in mass, installation costs (optimized aircraft assembly line integration), and operating costs (maintenance costs).
1.2.1.2
Basic Types
Electric actuators are divided into two different types: rotary and linear.
Rotary electric actuators rotate from open to closed using butterfly, ball,
and plug valves. With the use of rotary electric actuators, the electromagnetic power from the motor causes the components to rotate, allowing for
numerous stops during each stroke. Either a circular shaft or a table can be
used as the rotational element. When selecting an electric rotary actuator,
important factors to consider include actuator torque and range of motion.
The actuator torque refers to the power that causes the rotation, while the
Reduction gear
Stator
Ball screw and nut
Electric motor
Rotor
Figure 1.28 Working principle of an electric actuator system.
Zhang_Ch01.indd 90
5/13/2008 5:45:30 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
91
full range of motion can be either nominal, quarter-turn, or multiturn.
Linear electric actuators, in contrast, open and close using pinch, globe,
diaphragm, gate, or angle valves. Linear electric actuators are often used
when tight tolerances are required. These electric actuators use an acme
screw assembly or motor-driven ball screw to supply linear motion. Within
linear electric actuators, the load is connected to the end of a screw that is
belt or gear driven. Important factors to consider when selecting linear
electric actuators include the number of turns, actuating force, and the
length of the valve stem stroke.
(1) Linear electric actuators provide linear motion via a motordriven ball screw or screw assembly. The linear actuator’s load is
attached to the end of a screw, or rod, and is unsupported. The
screw can be direct, belt, or gear driven. Important performance
specifications to consider when searching for linear actuators
include stroke, maximum rated load or force, maximum rated
speed, continuous power, and system backlash. Stroke is the distance between fully extended and fully retracted rod positions.
The maximum rated load or force is not the maximum static
load. The maximum rated speed is the maximum actuator linear
speed, typically rated at low or no load. Continuous power is
sustainable power; it does not include short-term peak power ratings. Backlash is position error due to direction change. Motor
choices include DC, DC servo, DC brushless, DC brushless
servo, AC, AC servo, and stepper. Input power can be specified
for DC, AC, or stepper motors.
Drive screw specifications to consider for linear actuators
include drive screw type and screw lead. Features include selflocking, limit switches, motor encoder feedback, and linear position feedback. Screw choices include acme screws and ball
screws. Acme screws will typically hold loads without power but
are usually less efficient than ball screws. They also typically
have a shorter life but are more robust to shock loads. If backlash
is a concern, it is usually better to select a ball screw. Ball screws
exhibit lower friction and therefore higher efficiency than “lead
screws.” Screw lead is the distance the rod advances with one
revolution of the screw. Other features for linear actuators to
consider include holding brakes, integrated overload slip clutch
or torque limiters, water-resistant construction, protective boot,
and thermal overload protection. Design units can be English or
metric. Some manufacturers specify both. Dimensions to consider when specifying linear actuators include retracted length,
width, height, and weight. The housing can have flanges, rear
Zhang_Ch01.indd 91
5/13/2008 5:45:30 PM
92
INDUSTRIAL CONTROL TECHNOLOGY
clevis, side angle brackets, side lugs, tapped holes, and spherical
bearings. Rod ends can be clevis, female eye, female thread,
male thread, and spherical bearing. An important environmental
parameter to consider is the operating temperature.
(2) Rotary electric actuators provide incremental rotational movement of the output shaft. In its most simple form, a rotary actuator consists of a motor with a speed reducer. These AC and DC
motors can be fabricated to the exact voltage, frequency, power,
and performance specified. The speed reducer is matched with
the ratio to the speed, torque, and acceleration required. Life,
duty cycle, limit load, and accuracy are considerations that
further define the selection of the speed reducer. Hardened, precision spur gears are supported by antifriction bearings as a
standard practice in these speed reducers. Compound gear reduction is accomplished in compact, multiple load path configurations, as well as in planetary forms. The specifications for rotary
actuator include angular rotation, torque, and speed, as well
as control signals and feedback signals, and the environment
temperature.
Rotary actuators can incorporate a variety of auxiliary components such as brakes, clutches, antibacklash gears, and/or special
seals. Redundant schemes involving velocity or torque summing
of two or more motors can also be employed. Today the linear
motion in actuators is converted to a rotary one in many applications. By delivering the rotary motion directly, some fittings
can be saved in the bed. This enables the bed manufacturer to
build in a rotary actuator far more elegantly than a linear actuator. The result is a more “pure” design because the actuator is not
experienced as a product hanging under the bed, but as a part of
the bed.
Those rotary electric actuators are used for modulating valves,
which are divided based on the range from multiturn to quarterturn. Electrically powered multiturn actuators are one of the most
common and dependable configurations of actuators. A single or
three-phased electric motor drives a combination of spurs and/or
level gears, which in turn drive a stem nut. The stem nut engages
the stem of the valve to open or close it, frequently via an acme
threaded shaft. Electric multiturn actuators are capable of quickly
operating very large valves. To protect the valve, the limit switch
turns off the motor at the ends of travel. The torque sensing
mechanism of the actuator switches off the electric motor when
a safe torque level is exceeded. Position indicating switches are
utilized to indicate the open and closed position of the valve.
Zhang_Ch01.indd 92
5/13/2008 5:45:31 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
93
Typically a declutching mechanism and hand wheel are included
so that the valve can be operated manually should a power failure
occur. The main advantage of this type of actuator is that all of
the accessories are incorporated in the package and are physically and environmentally protected. It has all the basic and
advanced functions incorporated in a compact housing which
can be water tight, explosion proof, and in some circumstances,
submersible. The primary disadvantage of an electric multiturn
actuator is that should a power failure occur, the valve remains in
the last position and the fail-safe position cannot be obtained
easily unless there is a convenient source of stored electrical
energy.
Electric quarter-turn actuators are very similar to electric multiturn actuators. The main difference is that the final drive element is usually in one quadrant that puts out a 90° motion. The
newer generation of quarter-turn actuators incorporates many of
the features found in most sophisticated multiturn actuators, for
example, a nonintrusive, infrared, human machine interface for
set up, diagnostics, etc. Quarter-turn electric actuators are compact and can be used on smaller valves. They are typically rated
to around 1500 foot pounds. An added advantage of smaller
quarter-turn actuators is that because of their lower power
requirements, they can be fitted with an emergency power source
such as a battery to provide fail-safe operation.
Thrust actuators can be fitted to valves which require a linear
movement. Thrust actuators transform the torque of a multi-turn
actuator into an axial thrust by means of an integrated thrust
unit. The required (switch-off) actuating force (thrust and traction) can be adjusted continuously and reproducibly. Linear
actuators are mainly used to operate globe valves. Thrust units,
fitted to the output drive of a multiturn actuator, consist mainly
of a threaded spindle, a metric screw bolt to join the valve shaft,
and a housing to protect the spindle against environmental influences. The described version is used for “direct mounting” of
the actuator to the valve. However, thrust actuators version “fork
joint” (indirect mounting) can also operate butterfly valves or
dampers, when direct mounting of a part turn actuator is not
possible or efficient. The thrust units of the thrust actuators for
modulating duty also comply with the high demands of the
modulating duty. Also, for these thrust units, high-quality materials and accurate tolerances secure perfect function for many
years of operation. The thrust units are operated by modulating
actuators.
Zhang_Ch01.indd 93
5/13/2008 5:45:31 PM
94
INDUSTRIAL CONTROL TECHNOLOGY
1.2.1.3 Technical Specification
When selecting an electric actuator the specification including power
source, correct type, and size can be found utilizing the following criteria:
(1) Power source. When electric power is selected, a three-phase supply depends upon the valve driven by the actuator. It is usually
required for large valves; however, small valves can be operated
on a single-phase supply. Usually an electric valve actuator can
accommodate any of the common voltages. Sometimes a DC supply is available. This is often an emergency back-up power supply.
Variations of fluid power are much greater. First there is a variety
of fluid media such as compressed air, nitrogen, hydraulic fluid,
or natural gas. Then, there are the variations in the available pressures of those media. With a variety of cylinder sizes, most of the
variations can be accommodated for a particular valve size.
(2) Type of valve. Whenever sizing an actuator for a valve, the type
of valve has to be known, so that the correct type of actuator can
be selected. There are some valves that need multiturn input,
whereas others need quarter-turn. This has a great impact on the
type of actuator that is required. When combined with the available power supply, the size and type of actuator quickly come
into focus.
Generally multiturn fluid power actuators are more expensive
than multiturn electric actuators. However, for rising nonrotating
stem valves a linear fluid power actuator may be less expensive.
A definitive selection cannot be made until the power requirements of the valve are determined. After that decision has been
made, the torque requirement of the valve will be the next selection criterion.
The next task is to calculate the torque required by the valve.
For a quarter-turn valve, the best way of determining the torque
required is by obtaining the valve maker’s torque data. Most
valve makers have measured the torque required to operate their
valves over the range of operating line pressures. They make this
information available for customers.
The situation is different for multiturn valves. These can be
subdivided into several groups: the rising rotating, rising nonrotating, and nonrising rotating valves. In each of these cases the
measurement of the stem diameter together with the lead and
pitch of the valve stem thread is required in order to size the
automation for the valve. This information coupled with the size
of the valve and the differential pressure across the valve can be
used to calculate torque demand.
Zhang_Ch01.indd 94
5/13/2008 5:45:31 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
95
(3) Type and size of the electric actuator. The type and size of the
actuator can be determined after the power supply, the type of
valve, and the torque demand of that valve have been defined.
Once the actuator type has been selected and the torque
requirement of the valve has been determined, then the actuator
can be sized using one of the actuator manufacturer’s sizing programs or tables. A further consideration in sizing the actuator is
the required speed of operation of the valve. As speed has a direct
relationship to the power required from the actuator, more horsepower would be needed to operate a valve at a faster speed.
The electric motor operators of the three-phase type have a
fixed speed of operation. Smaller, quarter-turn actuators utilize
DC motors and may have adjustable speed of operation.
(4) Predictive maintenance. Motor operators can utilize built-in data
loggers coupled with highly accurate torque sensing mechanisms
to record data on the valve as it moves through its stroke. The
torque profiles can be used to monitor changes in the operating
conditions of the valve and to predict when maintenance is
required. They can also be used to troubleshoot valves.
Forces on a valve can include the following: (1) valve seal or
packing friction; (2) valve shaft, bearing friction; (3) valve
closure element seat friction; (4) closure element travel friction;
(5) hydrodynamic forces on closure elements; (6) stem piston
effect; (7) valve stem thread friction.
Most of these are present in all types of valves, but in varying
degrees of magnitude. For example, closure element travel friction in a butterfly valve is negligible, whereas a nonlubricated plug
valve has significant travel friction. Valve actuators are designed
to limit their torque to a preset level using a torque switch, usually
in a closing direction. An increase in torque above this level will
stop the actuator. In the opening direction, the torque switch is
frequently bypassed for the initial unseating operation. The resulting torque profile is useful in analyzing the valve condition.
Different types of valves have different profiles. For example,
a wedge gate valve has significant torque at the opening and
closing positions. During the remaining portion of the stroke
the torque demand is made up of packing and thread friction on
the acme threaded shaft. On seating, the hydrostatic force on the
closure element increases the seating friction, and finally the
wedging effect of the closure element in the seat causes a rapid
increase in torque demand until seating is completed. Changes in
torque profile can, therefore, give a good indication of pending
problems and can provide valuable information for an effective
predictive valve maintenance program.
Zhang_Ch01.indd 95
5/13/2008 5:45:31 PM
96
INDUSTRIAL CONTROL TECHNOLOGY
1.2.1.4 Application Guides
There are many applications where an electric actuator may be considered for the process control. Although electric actuators may be used anywhere a power source (electricity) is available, there are many applications
where they are particularly well suited. For instance, in many remote
installations, it may be impractical to run an air supply and to maintain it.
Air lines that freeze up may clog and render the equipment inoperable or
damage more delicate instruments. If only a few actuators are to be installed
in an area, electric actuators offer a simple means of automation for these
smaller systems.
Perhaps one of the most important reasons for the trend toward using
electric actuators has been the decreasing cost of using computers as system controllers and the ease and economy with which the actuators can be
interfaced to such systems. This trend can be expected to accelerate with
the advent of new microprocessors, or smart controllers, based on the versatile, low-cost microprocessor chips. Because of the increased speed and
decision- and control-making capability that the computer adds to a process system, there is less need for final control elements, which have high
control capability, such as characterized by globe and plug valves. As a
result, the simpler and less expensive electric actuators and ball and butterfly valves have become more acceptable and are proving more than
adequate for many applications. Probably the most important reason for
the widespread use of electric actuators is their control circuit versatility.
As an electric device, electric actuators naturally lend themselves for use
as an enclosure for a variety of control and feedback devices. Furthermore,
the switch and cams may be set and wired for almost any contact development for process and valve control.
It is more economical to install and maintain electric valve actuators
than pneumatic ones. A pneumatic system includes not only actuators, but
also compressors, piping, filters, air lubrication systems, and dryers.
However, electric actuators eliminate the need for air, an expensive source
of energy, and do not require energy when not in motion. There is no need
to be concerned with compressor noise, housing to shield against noise, air
venting, or other operating restrictions associated with pneumatic systems.
One important advantage of an electric actuator is its ability to manually
“jog” the valve position when it is used in filling operations. By installing
a feedback potentiometer an operator can monitor the exact position of the
actuators and stop it at any point between open and closed with a manual
control switch.
Zhang_Ch01.indd 96
5/13/2008 5:45:31 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
97
A few examples of the control circuit versatility of electric actuators are
three-position control and interposing relay. Three-position control is
especially useful for the automation of multiported (three- or four-way)
valves. A simple ON/OFF switch powers an interposing relay. This circuit
is similar to the type used in energizing single-acting solenoid valves in
pneumatic actuator installations.
(1) Controls. The great advantage of having an automated valve is
that it can be remotely controlled. This means that operators can
sit in a control room and control a process without having to
physically go to the valve and give it an open or close command,
the most basic type of control for an automated valve. The ability
to remotely control a valve is easily achieved by running a pair
of wires out to the actuator from the control room. Applying
power across the wires can energize a coil, initiating motion in
an electric or fluid power actuator. Positioning a valve in an intermediate position can be done using this type of control. However,
feedback would be needed to verify whether the actuator is at the
desired position. A more common method of positioning an
actuator is to feed a proportional signal to the actuator, such as
4–20 mA, so that the actuator, using a comparator device, can
position itself in direct proportion to the received signal.
(2) Modulating control. In some workplaces where an actuator is
required to control a level, flow, or pressure in a system, it may
be required to move frequently. Modulating or positioning control can be achieved using the same 4–20 mA signal. However,
the signal would change as frequently as the process required. If
very high rates of modulation are required then special modulating control valve actuators are needed that can accommodate the
frequent starts required for such duty.
Where there are many actuators on a process, the capital cost
of installation can be reduced by utilizing digital communication
over a communicating loop that passes from one actuator to
another (Fig. 1.29). A digital communication loop can deliver
commands and collect actuator status rapidly and cost effectively. There are many types of digital communication such as
Foundation Fieldbus, Profibus, DeviceNet, Hart, as well as proprietary communication systems custom designed for valve actuator use such as Pakscan. Digital communication systems have
many advantages over and above the saving in capital cost. They
are able to collect a lot of data about the condition of the valve,
and as such can be used for predictive maintenance programs.
Zhang_Ch01.indd 97
5/13/2008 5:45:31 PM
98
INDUSTRIAL CONTROL TECHNOLOGY
Figure 1.29 Digital communication systems for remote control of valves
(courtesy of Rotork Controls, Inc.).
1.2.1.5
Calibrations
Here, VA-7202 Electric Valve Actuator (Johnson Controls, Inc.) is taken
as an example to introduce a method and a process for calibrating electric
actuators.
(1) Set the stroke jumper to approximate the stroke of the valve.
See Fig. 1.30 for jumper location that includes (1) Jumper 8:
5/16 in. or 8 mm; (2) Jumper 10: 3/8 in. or 10 mm; (3) Jumper
13: 1/2 in. or 13 mm; (4) Jumper 19: 3/4 in. or 19 mm.
Stroke
jumpers
Stroke
adjustment
Input signal
selection
Span
value
8
10
13
19
Action
selection
DA
RA
V
mA
*Supply
disconnect
Starting
point
Down
Up
1 2 3 4
Fail-safe
input signal
Down
Up
LED
* Disconnects power supply to the circuit
must be in place for actuator operation.
Figure 1.30 VA-7202 electric actuator (Johnson Controls, Inc.) components.
Zhang_Ch01.indd 98
5/13/2008 5:45:31 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
99
(2) Set the direct/reverse action jumper so that the valve stem travels in the desired direction (per changes in control signal)
including (1) DA (top jumper) = stroke down on signal increase;
(2) RA (bottom jumper) = stroke up on signal increase.
(3) Set the input signal selection jumper for voltage input or current (mA) input to match the controller output; see Fig. 1.30. (If
the input signal selection jumper is removed, the actuator
defaults to voltage input.) However, if mA input is selected,
multiply the start and span scales by 2.
(4) Set the signal fail position jumper to select default position of
fully up or fully down. If the signal is lost at the actuator (open
connection), the actuator will default to the predesignated position of full up or full down. However, if mA input is selected,
the actuator will default to the low input signal position.
(5) Adjust the potentiometers to the nominal values. Set the stroke
adjustment to the midpoint as shown in Fig. 1.30. Set the starting point (offset) to the low input signal using the scale printed
on the circuit board as a reference. Set the span value to the
high input signal minus the offset and then use the scales for
reference.
(6) Apply voltage specified by application (RA/DA) requirements
to drive the actuator to the full up position. If mA input is
selected, multiply all values by 2.
(7) Slowly turn the starting point potentiometer (shown in
Fig. 1.30) CW (clockwise) until the valve stem reaches the end
of stroke to ensure that the valve stem is in the full-up position.
LED will be on; there should be no gear movement.
(8) Slowly turn the starting point potentiometer counterclockwise
(CCW); stop when the LED flashes or goes out. If the LED
does not flash or go out, verify Dimension “A” gaps. Excessive
gap may not allow full-up calibration. The actuator circuit contains a time-out feature. If calibration takes longer than 3–10 min,
the LED will go out giving a false satisfied condition. If this
occurs, cycle the power to the actuator and readjust the starting
point.
(9) Apply the input voltage specified by application (RA/DA)
requirements to drive the valve stem to the full-down position
per chart in Step 6.
(10) To ensure that the valve stem is in the full-down position, slowly
turn the stroke potentiometer CW until the valve stem reaches
the end of stroke. LED will be on, and there should be no gear
movement.
(11) Slowly turn the stroke potentiometer CCW until the LED
goes off.
Zhang_Ch01.indd 99
5/13/2008 5:45:33 PM
100
INDUSTRIAL CONTROL TECHNOLOGY
(12) If the full-down position cannot be reached, return the stroke
potentiometer to the nominal position and slowly turn the span
potentiometer CCW until full down is reached. Then, repeat
Step 11.
(13) Adjust voltage to drive the actuator to the full-up position.
Verify starting point adjustment.
(14) Check for proper operation using the desired minimum and
maximum operating voltages. Allow the actuator to operate
through several complete cycles. The LED will remain on for
3–10 min after the actuator has completed the operation cycle.
(15) Replace the cover and secure with the screws. At this point, the
unit is ready for operation.
1.2.2
Pneumatic Actuators
Pneumatics have obtained a variety of applications in manufacturing to
control industrial processes, in automotive and aircraft settings to modulate valves, even in medical equipment such as dentistry drills to actuate
torque movements, etc. In contrast with other physics like electrics and
hydraulics, the operating torque of pneumatics makes possible a compact
actuator that is economical to both install and operate. Pneumatic devices
are also used where electric motors cannot be used for safety reasons and
where no water is supplied, such as mining applications where rock drills
are powered by air motors to preclude the need for electric motors deep in
the mine where explosive gases may be present. In many cases, it is easier
to use a liquid or gas at high pressure rather than electricity to provide
power for the actuator. Pneumatic actuators provide a very fast response
but little power, whereas hydraulic systems can provide very great forces,
but act more slowly. This is partly because gases are compressible and
liquids are not. The pneumatic actuators (Fig. 1.31) offer the latest technology; a premium quality ball valve, a quality actuator designed to meet
the torque requirements of the valve, and a mounting system which ensures
alignment and rigidity.
1.2.2.1
Operating Principle
Industrial pneumatics may be contrasted with hydraulics that use
uncompressible liquid media such as oil, or water combined with soluble
oil, instead of air. Air is compressible, and therefore it is considered to be
a fluid. The pneumatic principles conclude that the pressure formed in
compressible liquids can be harnessed to a high potential of power. This
gives us new potential for several pneumatic-powered operations and
Zhang_Ch01.indd 100
5/13/2008 5:45:33 PM
101
H0
H0
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
D
D
H
H
Air inlet
Air inlet
Air to open
Air to close
Figure 1.31 Components of a spring-type pneumatic actuator (courtesy of
Forbes Marshall).
hence creates many new developments. Both pneumatics and hydraulics
are applications of fluid power.
Both pneumatic linear and rotary actuators use pressurized air to drive
or rotate mechanical components. The flow of pressurized air produces the
shift or rotation of moving components via a stem and spring, rack and
pinion, cams, direct air or fluid pressure on a chamber or rotary vanes, or
other mechanical linkage. A valve actuator is a device mounted on a valve
that, in response to a signal, automatically moves the valve to the desired
position using an outside power source. Pneumatic valve actuators convert
air pressure into motion.
(1) Linear pneumatic actuators. A simplified diagram of a pneumatic
linear actuator is shown in Fig. 1.32. It operates with a combination of force created by air and spring force. The actuator shifts
the positions of a control valve by transmitting its motion through
the stem. A rubber diaphragm separates the actuator housing into
two air chambers. The left or upper chamber receives a supply of
air through an opening in the top of the housing. The right or bottom chamber contains a spring that forces the diaphragm against
mechanical stops in the upper chamber. Finally, a local indicator
is connected to the stem to indicate the position of the valve.
Zhang_Ch01.indd 101
5/13/2008 5:45:33 PM
102
INDUSTRIAL CONTROL TECHNOLOGY
Positioner
Stem
Spring
Air
Chamber
Figure 1.32 Operation principle of a simplified linear pneumatic actuator.
The position of the valve is controlled by varying supply air
pressure in the left or upper chamber, which results in a varying
force on the top of the diaphragm. At the beginning, with no
supply air, the spring forces the diaphragm upward against the
mechanical stops and holds the valve fully open. As supply air
pressure is increased from zero, its force on top of the diaphragm
begins to overcome the opposing force of the spring. The causes
the diaphragm to move rightward or downward and the control
valve to close. With increasing supply air pressure, the diaphragm will continue to move rightward or downward and compress the spring until the control valve is fully closed. Conversely,
if supply air pressure is decreased, the spring will force the
diaphragm leftward or upward and open the control valve.
Additionally, if supply pressure is held constant at some value
between zero and maximum, the valve will position at an intermediate point.
The valve can hence be positioned anywhere between fully
open and fully closed in response to changes in supply air
pressure. A positioner is a device that regulates the supply air
pressure to a pneumatic actuator. It does this by comparing the
position demanded by the actuator with the control position of
the valve. The requested position is transmitted by a pneumatic
or electrical control signal from a controller to the positioner.
The controller generates an output signal that represents the
requested position. This signal is sent to the positioner. Externally,
the positioner consists of an input connection for the control
signal, a supply air input connection, a supply air output connection, a supply air vent connection, and a feedback linkage.
Internally, it contains an intricate network of electrical transducers, air lines, valves, linkages, and necessary adjustments.
Zhang_Ch01.indd 102
5/13/2008 5:45:34 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
103
Other positioners may also provide controls for local valve positioning and gauges to indicate supply air pressure and controller
pressure.
(2) Rotary pneumatic actuators. Pneumatic rotary actuators may
have fixed or adjustable angular strokes and can include such
features as mechanical cushioning, closed-loop hydraulic dampening (oil), and magnetic features for reading by a switch.
When the compressed air enters the actuator from the first
tube nozzle, the air will push the double pistons toward both ends
(cylinder end) for straight line movement. The gear on the piston
drives the gear on the rotary shaft to rotate counterclockwise,
and then the valve can be opened. This time, the air at both ends
of the pneumatic actuator will drain with another tube nozzle.
Conversely, when the compressed air enters the actuator from
the second tube nozzle, the gas will push the double pistons
toward the middle for straight line movement. The rack on the
piston drives the gear on the rotary shaft to rotate clockwise and
then the valve can be closed. This time, the air at the middle of
the pneumatic actuator will drain with the first tube nozzle.
Above is the standard-type driving principle. According to users’
requirement, the pneumatic actuator can be assembled into the
driving principle beyond the standard type, which means opening the valve while rotating the rotary shaft clockwise and closing the valve while rotating the rotary shaft counterclockwise.
Figure 1.33(a) is for the double-acting type. For clockwise
operation, Port 2 (P2) is open to atmosphere, and air pressure is
directed to Port 1 (P1). As the pistons move apart, the pinion
rotates clockwise. The linear movement of the pistons is converted to rotary motion by the piston racks and the output
pinion gear. For counterclockwise operation, Port 1 is open to
atmosphere and air pressure is directed to Port 2. The pressure
differential moves the pistons together, rotating the pinion
counterclockwise.
Figure 1.33(b) is for the spring-return type. For clockwise
operation, Port 2 is open to atmosphere and air pressure is
directed to Port 1. The air pressure compresses the springs and
moves the pistons outward. As the pistons move apart, the pinion
rotates clockwise. The linear movement of the pistons is converted to rotary motion by the piston racks and the output pinion
gear. For counterclockwise operation, Port 1 is open to atmosphere and air pressure is directed to Port 2. High pressure and/
or spring force moves the pistons inward, rotating the pinion
counterclockwise.
Zhang_Ch01.indd 103
5/13/2008 5:45:34 PM
104
INDUSTRIAL CONTROL TECHNOLOGY
P1 P2
Clockwise operation
P1 P2
Clockwise operation
P1 P2
Counterclockwise operation
P1 P2
Counterclockwise operation
(a)
(b)
Figure 1.33 Operation principle of rotary pneumatic actuators: (a) double acting
and (b) spring return.
1.2.2.2
Basic Types and Specifications
In terms of the actuated movements, pneumatic actuators can be two
types: linear and rotary.
(1) Pneumatic linear motion specifications:
(a) breakaway torque (in. lb)
(b) actuation type: spring return or double acting
(c) fail safe: no or yes
(d) positioner: no or yes; if yes: pneumatic or electropneumatic
(e) limit switch: no, two, or four.
(2) pneumatic rotary valve specifications:
(a) output torque (in. lb)
(b) pipe size (specify)
(c) product flowing in pipe (specify)
(d) temperature (oF)
(e) pressure (psig)
(f) flow (GPM)
(g) valve material (specify)
(h) connections (specify).
Zhang_Ch01.indd 104
5/13/2008 5:45:34 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
105
In terms of operators, pneumatic actuators can be of four types:
(1) Piston-style pneumatic actuators. Piston-type, air-operated valves
offer a unique, reliable design providing for a long and dependable life. These valves are more compact than diaphragm valves
and are appropriate for applications such as high-flow gas and
liquid delivery systems to reactors and mixers and vaporizers. The
technical data for the piston-style pneumatic actuators include
(a) area of piston (sq. in.)
(b) maximum allowable working pressure (psi or bar)
(c) allowable piston temperature range (oF)
(d) approximate air usage and air cycle ( SCF per 100 psi)
(e) tested to 100,000 cycles at 100 psi (6.89 bar) with no leakages or signs of wear or fatigue.
(2) Diaphragm-style pneumatic actuators. Diaphragm-type, airoperated valves are an efficient and economical means for “remote
ON–OFF” of a wide range of process requirements. Diaphragmtype actuators are designed to provide a dependable alternative
to piston-type actuators. The technical data for the diaphragmstyle pneumatic actuators include
(a) area of diaphragm (sq. in.)
(b) maximum allowable working pressure (psi or bar)
(c) allowable diaphragm temperature range (oF)
(d) approximate air usage and air cycle ( SCF per 100 psi).
(3) Solenoid valve packages. Solenoid valves are used to supply
ON–OFF control of air to the valve actuator. They are normally
mounted on the valve actuator, but they can be mounted remotely
as well. Solenoid valves can be supplied in different voltages and
configurations if required. Solenoid manifolds are available and
used when a large number of valve actuators require control or
contain power to a local area reducing assembly time. Remotemounted solenoids also permit the usage of actuated valves in
hazardous sensitive locations. These indices are assigned to solenoid valve packages of actuators:
(a) Piston and diaphragm operators: light and heavy-light duty
(b) piston and diaphragm operators: medium and heavy duty
(c) piston and diaphragm operators: extra heavy duty.
(4) Servo pneumatic actuator package. The servo pneumatic actuator
has an integral displacement transducer and is designed to operate on standard compressed air available in most factories and
laboratories. The package is aimed at engineers building specialpurpose systems including
(a) Displacement control. The servo pneumatic actuator can be
used in systems configured for displacement control using
Zhang_Ch01.indd 105
5/13/2008 5:45:38 PM
106
INDUSTRIAL CONTROL TECHNOLOGY
feedback from the integral displacement transducer and signal
conditioning card supplied.
(b) Load control. Load control may be implemented by providing a load cell and signal conditioning.
(c) Strain control. Strain control may be implemented by providing a clip gauge and signal conditioning.
1.2.2.3 Application Guide and Assemble on Valve
(1) Preparation for installation
(a) Actuator checks. When the actuator arrives, the actuator
checks that determine whether it is mounted on the valve
should be carried out: If the actuator arrives already assembled onto the valve, the setting of the mechanical stops and
of the electric limit switches (if existing) has already been
made by the person who assembled the actuator onto the
valve. If the actuator arrives separately from the valve, the
settings of the mechanical stops and of the electric limit
switches (if existing) must be checked and, if necessary, carried out while assembling the actuator onto the valve.
Furthermore, it is often necessary to check that the actuator has not been damaged during transport. If necessary,
repair all damages to the paintcoat, etc.
Thereafter, it is often required to check that the model, the
serial number of the actuator, and the performance data written on the data plate are in accordance with those described
on the order acknowledgment, test certificate, and delivery
note.
Then, check that the fitted accessories comply with those
listed in the order acknowledgment and the delivery note.
(b) Store checks. The actuators leave the factory in excellent
working condition and with an excellent finish (these conditions are guaranteed by an individual inspection certificate);
in order to maintain these characteristics until the actuator is
installed, it is necessary to observe a few rules and take
appropriate measures during the storage period.
It should be ensured that plugs are fitted in the air connections and in the cable entries. The plastic plugs which close
the inlets do not have a weatherproof function, but are only a
means of protection against the entry of foreign matter during transport. If long-term storage is necessary and especially if the storage is out of doors, the plastic protection
plugs must be replaced by metal plugs, which guarantee a
complete weatherproof protection.
Zhang_Ch01.indd 106
5/13/2008 5:45:38 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
107
If the actuators are supplied separately from the valves,
they must be placed onto a wooden pallet so as not to damage the coupling flange to the valve. In case of long-term
storage, the coupling parts (flange, drive sleeve, insert bush)
must be coated with protective oil or grease. If possible,
blank off the flange by a protection disk.
In case of long-term storage, it is advisable to keep the
actuators in a dry place or to provide some means of weather
protection. If possible, it is also advisable to periodically
operate the actuator with filtered, dehydrated, and lubricated
air; after such operations all the threaded connections of the
actuator and the valves of the control panel (if present)
should be carefully plugged.
(2) Installation and start-up
(a) Pneumatic connections. Pneumatic connections connect the
actuator to the pneumatic feed line with fittings and pipes in
accordance with the plant specifications. They must be sized
correctly in order to guarantee the necessary air flow for the
operation of the actuator, with pressure drops not exceeding
the maximum allowable value. The shape of the connecting
piping must not cause excessive stress to the inlets of the
actuator. The piping must be suitably fastened so as not to
cause excessive stress or loosening of threaded connections,
if the system undergoes strong vibrations.
Every precaution must be taken to ensure that any solid or
liquid contaminants which may be present in the pneumatic
pipe work to the actuator are removed to avoid possible damage to the unit or loss of performance. The inside of the pipes
used for the connections must be well cleaned before use,
which includes washing with suitable substances and blowing through them with air or nitrogen. The ends of the tubes
must be well debarred and cleaned.
Once the connections are completed, operate the actuator
and ensure that it functions correctly, that the operation times
meet the plant requirements, and that there are no leaks in
the pneumatic connections.
(b) Electrical connections. Electric connections connect the
electrical feed, control, and signal lines to the actuator, by
linking them up with the terminal blocks of the electrical
components. In order to do this, the housing covers must
be removed without damaging the coupling surfaces, the
O-rings, or the gaskets.
Then, the plugs should be removed from the cable entries.
For electrical connections, we can use components (cable
Zhang_Ch01.indd 107
5/13/2008 5:45:38 PM
108
INDUSTRIAL CONTROL TECHNOLOGY
glands, cables, hoses, conduits), which meet the requirements
and codes applicable to the plant specifications (mechanical
protection and/or explosion-proof protection). The cable
glands should be screwed tightly into the threaded inlets so
as to guarantee weatherproof and explosion-proof protection
(when applicable). The connection cables are inserted into
the electrical enclosures through the cable glands, and the
cable wires are connected to the terminals according to the
applicable wiring diagram. If conduits are used, it is advisable to carry out the connection to the electrical enclosures
by inserting hoses so as not to cause anomalous stress on the
housing cable entries. Replace the plastic plugs of the unused
enclosure entries with metal ones to guarantee perfect weatherproof tightness and to comply with the explosion-proof
protection codes where applicable.
Once the connections are completed, check that the controls and signals work properly.
After the installation is complete, it is time for the start-up
of the actuator that proceeds as follows:
(i) Check that the pressure and quality of the air supply
(filtering degree, dehydration) are as prescribed. Check
that the feed voltage values of the electrical components (solenoid valve coils, micro switches, pressure
switches, etc.) are as prescribed.
(ii) Check that the actuator controls work properly (remote
control, local control, emergency controls, etc.)
(iii) Check that the required remote signals (valve position,
air pressure, etc.) are correct.
(iv) Check that the setting of the components of the actuator control unit (pressure regulators, pressure switches,
flow control valves, etc.) meet the plant requirements.
(v) Check that there are no leaks in the pneumatic connections. If necessary, tighten the nuts of the pipe
fittings.
(vi) Remove all rust and, in accordance with the applicable
painting specifications, repair paint coat that has been
damaged during transport, storage, or assembly.
(3) Mounting the actuator onto the valve. The actuator can be assembled onto the valve flange either by using the actuator housing
flanges with threaded holes, or by the interposition of an adaptor
flange or a spool piece. The actuator drive sleeve is generally
connected to the valve stem by an insert bush or a stem extension. The assembly position of the actuator, with reference to the
valve, must comply with the plant requirements (cylinder axis
Zhang_Ch01.indd 108
5/13/2008 5:45:38 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
109
parallel or perpendicular to the pipeline axis). To assemble the
actuator onto the valve, proceed as follows:
(a) Check that the coupling dimensions of the valve flange and
stem, or of the relevant extension, meet the actuator coupling
dimensions.
(b) Bring the valve to the “closed” position.
(c) Lubricate the valve stem with oil or grease in order to make
the assembly easier. Be careful not to pour any of it onto the
flange.
(d) Clean the valve flange and remove anything that might prevent a perfect adherence to the actuator flange, especially all
traces of grease, since the torque is transmitted by friction.
(e) If an insert bush or stem extension for the connection to the
valve stem is supplied separately, assemble it onto the valve
stem and fasten it by tightening the proper stop dowels.
(f) Bring the actuator to the “closed” position.
(g) Connect a sling to the support points of the actuator and lift
it: make sure the sling is suitable for the actuator weight.
When possible, it is easier to assemble the actuator to the
valve if the valve stem is in the vertical position. In this case
the actuator must be lifted while keeping the flange in the
horizontal position.
(h) Clean the actuator flange and remove anything that might
prevent a perfect adherence to the valve flange, especially all
traces of grease.
(i) Lower the actuator onto the valve in such a way that the
insert bush, assembled on the valve stem, enters the actuator
drive sleeve. This coupling must take place without forcing
and only with the weight of the actuator. When the insert
bush has entered the actuator drive sleeve, check the holes of
the valve flange. If they do not meet with the holes of the
actuator flange or the stud bolts screwed into them, the actuator drive sleeve must be rotated; feed the actuator cylinder
with air at the proper pressure or actuate the manual override, if existing, until coupling is possible.
(j) Tighten the nuts of the connecting stud bolts evenly with the
torque prescribed in the table.
(k) If possible, operate the actuator to check that it moves the
valve smoothly.
It is important that the mechanical stops of the actuator
(and not those of the valve) stop the angular stroke at
both extreme valve positions (fully open and fully closed),
except when otherwise required by the valve operation
Zhang_Ch01.indd 109
5/13/2008 5:45:38 PM
110
INDUSTRIAL CONTROL TECHNOLOGY
(e.g., metal-seated butterfly valves). The setting of the open
valve position is performed by adjusting the travel stop screw
in the left wall of the mechanism housing, or in the end
flange of the manual override, if that exists. The setting of
the closed valve position is performed by adjusting the travel
stop screw in the cylinder end flange. Proceed as follows:
(i) Loosen the lock nut.
(ii) If the actuator angular stroke is stopped before reaching the end position (fully open or closed), unscrew
the stop screw by turning it counterclockwise, until the
valve reaches the correct position. When unscrewing
the stop screw, keep the lock nut still with a wrench so
that the sealing washer does not withdraw together
with the screw.
(iii) Tighten the lock nut.
(iv) If the actuator angular stroke is stopped beyond the
end position (fully open or closed), screw the stop
screw by turning it clockwise until the valve reaches
the correct position.
(v) Tighten the lock nut.
(4) Maintenance. Before carrying out any maintenance operation, it
is necessary to close the pneumatic feed line and exhaust the
pressure from the actuator cylinder and from the control unit, to
ensure the safety of maintenance staff.
(a) Routine maintenance. Most actuators have been designed to
work for long periods in the severest conditions with no need
for maintenance. It is, however, advisable to periodically
check the actuator as follows:
(i) Check that the actuator operates the valve correctly
and with the required operating times. If the actuator
operation is very infrequent, we can carry out a few
opening and closing operations with all the existing
controls (remote, local, emergency controls, etc.), if
this is allowed by the plant conditions.
(ii) Check that the signals to the remote control desk are
correct.
(iii) Check that the air supply pressure value is within the
required range.
(iv) If there is an air filter on the actuator, bleed the condensed water accumulated in the cup by opening the
drain cock. Disassemble the cup periodically and wash
it with soap and water; disassemble the filter: if this is
made of a sintered cartridge, wash it with nitrate solvent and blow through it with air. If the filter is made
of cellulose, it must be replaced when clogged.
Zhang_Ch01.indd 110
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
111
(v) Check that the external components of the actuator are
in good condition.
(vi) Check all the paint coat of the actuator. If some areas
are damaged, repair the paint coat according to the
applicable specification.
(vii) Check that there are no leaks in the pneumatic connections. If necessary tighten the nuts of the pipe fittings.
(b) Special maintenance. If there are air leaks in the pneumatic
cylinder or a malfunction in the mechanical components,
or in case of scheduled preventative maintenance, the actuator must be disassembled and seals must be replaced with
reference to the sectional drawing, adopting the following
procedures:
(i) disassemble the actuator correctly;
(ii) carry out the seal replacement in the actuator;
(iii) reassemble the actuator correctly.
1.2.3
Hydraulic Actuators
Pneumatic actuators are normally used to control processes requiring
quick and accurate response, as they do not require a large amount of
motive force. However, when a large amount of force is required to operate
a valve such as the main steam system valves, hydraulic actuators are normally used. A hydraulic actuator receives pressure energy and converts it
to mechanical force and motion. Fluid power systems are manufactured by
many organizations for a very wide range of applications, which often
embody differing arrangements of components to fulfill a given task.
Hydraulic components are manufactured to provide the control functions
required for the operation of systems.
1.2.3.1
Operating Principle
(1) Hydraulic cylinders. A cylinder is a hydraulic actuator that is
constructed of a piston or plunger that operates in a cylindrical
housing by the action of liquid under pressure. Cylinder housing
is a tube in which a plunger (piston) operates. In a ram-type cylinder, a ram actuates a load directly. In a piston cylinder, a piston
rod is connected to a piston to actuate a load. At the end of a
cylinder from which a rod or plunger protrudes is a rod end. Its
opposite end is the head end. The hydraulic connections are a
head-end port and a rod-end port (fluid supply).
(a) Single-acting cylinder. This cylinder has only a head-end
port and is operated hydraulically in one direction. When oil
Zhang_Ch01.indd 111
5/13/2008 5:45:39 PM
112
INDUSTRIAL CONTROL TECHNOLOGY
is pumped into a port, it pushes on a plunger, thus extending
it. To return or retract a cylinder, oil must be released to a
reservoir. A plunger returns either because of the weight of a
load or from some mechanical force such as a spring. In
mobile equipment, a reversing directional valve of a singleacting type controls flow to and from a single-acting
cylinder.
(b) Double-acting cylinder. This cylinder must have ports at the
head and rod ends. Pumping oil into the head end moves a
piston to extend a rod while any oil in the rod end is pushed
out and returned to a reservoir. To retract a rod, flow is
reversed. Oil from a pump goes into the rod end, and the
head-end port is connected to allow return flow. The flow
direction to and from a double-acting cylinder can be controlled by a double-acting directional valve or by actuating
control of a reversible pump.
(c) Differential cylinder. In a differential cylinder, the areas
where pressure is applied on a piston are not equal. On a
head end, a full piston area is available for applying pressure. At a rod end, only an annular area is available for applying pressure. The area of a rod is not a factor, and what space
it does take up reduces the volume of oil it will hold.
Two general rules about a differential cylinder: with an
equal GPM delivery to either end, a cylinder will move faster
when retracting because of a reduced volume capacity. With
equal pressure at either end, a cylinder can exert more force
when extending because of the greater piston area. In fact, if
equal pressure is applied to both ports at the same time, a
cylinder will extend because of a higher resulting force on a
head end.
(d) Nondifferential cylinder. This cylinder has a piston rod
extending from each end. It has equal thrust and speed either
way, provided that pressure and flow are unchanged. A nondifferential cylinder is rarely used on mobile equipment.
(e) Ram-type cylinder. A ram-type cylinder is a cylinder in which
the cross-sectional area of a piston rod is more than one-half
the cross-sectional area of the piston head. In many cylinders
of this type, the rod and piston heads have equal areas. A
ram-type actuating cylinder is used mainly for push rather
than pull functions.
(f) Piston-type cylinder. In this cylinder, a cross-sectional area
of a piston head is referred to as a piston-type cylinder. A
piston-type cylinder is used mainly when the push and pull
functions are needed.
Zhang_Ch01.indd 112
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
113
A single-acting, piston-type cylinder uses fluid pressure to
apply force in one direction. In some designs, the force of
gravity moves a piston in the opposite direction. However,
most cylinders of this type apply force in both directions.
Fluid pressure provides force in one direction and spring
tension provides force in the opposite direction.
Most piston-type cylinders are double acting, which means
that fluid under pressure can be applied to either side of a
piston to provide movement and apply force in a corresponding direction. This cylinder contains one piston and pistonrod assembly and operates from fluid flow in either direction.
The two fluid ports, one near each end of a cylinder, alternate
as inlet and outlet, depending on the directional-control valve
flow direction. This is an unbalanced cylinder, which means
that there is a difference in the effective working area on the
two sides of a piston. A cylinder is normally installed so that
the head end of a piston carries the greater load; that is, a
cylinder carries the greater load during a piston-rod extension stroke.
(g) Cushioned cylinder. To slow an action and prevent shock at
the end of a piston stroke, some actuating cylinders are constructed with a cushioning device at either or both ends of a
cylinder. This cushion is usually a metering device built into
a cylinder to restrict the flow at an outlet port, thereby slowing down the motion of a piston.
(h) Lockout cylinders. A lockout cylinder is used to lock a suspension mechanism of a tracked vehicle when a vehicle
functions as a stable platform. A cylinder also serves as a
shock absorber when a vehicle is moving. Each lockout cylinder is connected to a road arm by a control lever. When
each road wheel moves up, a control lever forces the respective cylinder to compress. Hydraulic fluid is forced around a
piston head through restrictor ports causing a cylinder to act
as a shock absorber. When hydraulic pressure is applied to
an inlet port on each cylinder’s connecting eye, an inner control-valve piston is forced against a spring in each cylinder.
This action closes the restrictor ports, blocks the main piston’s motion in each cylinder, and locks the suspension
system.
(2) Hydraulic motors. Hydraulic motors convert hydraulic energy
into mechanical energy. In industrial hydraulic circuits, pumps
and motors are normally combined with a proper valve and pipe
to form a hydraulic-powered transmission. A pump, which is
mechanically linked to a prime mover, draws fluid from a reservoir
Zhang_Ch01.indd 113
5/13/2008 5:45:39 PM
114
INDUSTRIAL CONTROL TECHNOLOGY
and forces it to a motor. A motor, which is mechanically linked
to the workload, is actuated by this flow so that motion or torque,
or both, are conveyed to the work.
(a) Gear-type motors. Both gears are driven gears, but only one
is connected to the output shaft. Operation is essentially the
reverse of that of a gear pump. Flow from the pump enters
chamber A and flows in either direction around the inside surface of the casing, forcing the gears to rotate as indicated. This
rotary motion is then available for work at the output shaft.
(b) Vane-type motors. Flow from the pump enters the inlet, forces
the rotor and vanes to rotate, and passes out through the outlet.
Motor rotation causes the output shaft to rotate. Since no centrifugal force exists until the motor begins to rotate, something, usually springs, must be used to initially hold the vanes
against the casing contour. However, springs usually are not
necessary in vane-type pumps because a drive shaft initially
supplies centrifugal force to ensure vane-to-casing contact.
Vane motors are balanced hydraulically to prevent a rotor
from side-loading a shaft. A shaft is supported by two ball
bearings. Torque is developed by a pressure difference as oil
from a pump is forced through a motor. On the trailing side
open to the inlet port, the vane is subject to full system pressure. The chamber leading the vane is subject to the much
lower outlet pressure. The difference in pressure exerts the
force on the vane that is, in effect, tangential to the rotor.
This pressure difference is effective across vanes. The other
vanes are subject to essentially equal force on both sides.
Each will develop torque as the rotor turns. The body port is
the inlet, and the cover port is the outlet. Reverse the flow
and the rotation becomes clockwise; otherwise the rotation
is counterclockwise.
In a vane-type pump, the vanes are pushed out against a
cam ring by centrifugal force when a pump is started up. A
design motor uses steel-wire rocker arms to push the vanes
against the cam ring. The arms pivot on pins attached to the
rotor. The ends of each arm support two vanes that are 90°
apart. When the cam ring pushes vane A into its slot, vane B
slides out. The reverse also happens. The pressure plate of a
motor functions the same as a pump’s. It seals the side of a
rotor and ring against internal leakage, and it feeds system
pressure under the vanes to hold them out against a ring.
This is a simple operation in a pump because a pressure plate
is right by a high-pressure port in the cover.
(c) Piston-type motors. Although some piston-type motors are
controlled by directional-control valves, they are often used in
Zhang_Ch01.indd 114
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
115
combination with variable-displacement pumps. This pump–
motor combination (hydraulic transmission) is used to provide a transfer of power between a driving element, such as
an electric motor, and a driven element.
Piston-type motors can be inline-axis or bent-axis types:
(i) Inline-axis piston-type motors. These motors are almost
identical to the pumps. They come in built-in, fixed-, and
variable-displacement models in several sizes. Torque is
developed by a pressure drop through the motor. Pressure
exerts a force on the ends of the pistons, which is translated into shaft rotation. Shaft rotation of most models
can be reversed anytime by reversing the flow direction.
Oil from a pump is forced into the cylinder bores through
a motor’s inlet port. Force on the pistons at this point
pushes them against a swash plate. They can move only
by sliding along a swash plate to a point further away
from a cylinder’s barrel, which causes it to rotate. The
barrel is then splined to a shaft so that it must turn.
The displacement of a motor depends on the angle of
the swash plate. At maximum angle, displacement is at
its highest because the pistons travel at maximum length.
When the angle is reduced, piston travel shortens, reducing displacement. If flow remains constant, a motor
runs faster, but torque is decreased. Torque is greatest at
maximum displacement because the component of piston force parallel to the swash plate is greatest.
(ii) Bent-axis piston-type motors. These motors are almost
identical to the pumps. They are available in fixed- and
variable-displacement models, in several sizes. Variabledisplacement motors can be controlled mechanically or
by pressure compensation. These motors operate similar to inline motors except that the piston thrust is against
a drive-shaft flange. A parallel component of thrust
causes a flange to turn. Torque is the maximum displacement; speed is at a minimum. This design of piston
motor is very heavy and bulky, particularly the variabledisplacement motor. Use of these motors on mobile
equipment is limited.
1.2.3.2
Basic Types and Specifications
Table 1.4 gives the basic types of hydraulic actuators in three columns
which are cylinder, valve, and motor, and the valve controlling source in
three rows which are electro, servo, and piezoelectric. The motion type for
Zhang_Ch01.indd 115
5/13/2008 5:45:39 PM
116
INDUSTRIAL CONTROL TECHNOLOGY
Table 1.4 Basic Types of Hydraulic Actuators
Types of Valve
Controller
Hydraulic
Cylinder
Actuators
Hydraulic
Valve
Actuators
Hydraulic
Motor
Actuators
Electro
Servo
Piezoelectric
Linear
Linear
Linear
Linear Rotary
Linear Rotary
Linear Rotary
Rotary
Rotary
Rotary
each type of hydraulic actuator device is specified in each cell of this table,
which indicates that cylinders have linear motion only, valves can have
both linear and rotary motions, and motors rotary motion only.
An actuator can be linear or rotary. A linear actuator gives force and
motion outputs in a straight line. It is more commonly called a cylinder but
is also referred to as a ram, reciprocating motor, or linear motor. A rotary
actuator produces torque and rotating motion. It is more commonly called
a hydraulic motor.
(1) Hydraulic cylinders and linear actuators. Hydraulic cylinders
are actuation devices that utilize pressurized hydraulic fluid to
produce linear motion and force. Hydraulic cylinders are used in
a variety of power transfer applications. Hydraulic cylinders can
be single action or double action. A single action hydraulic cylinder is pressurized for motion in only one direction. A double
action hydraulic cylinder can move along the horizontal (x-axis)
plane, the vertical (y-axis) plane, or along any other plane of
motion.
Operating specifications, configuration or mounting, materials
of construction, and features are all important parameters to consider when searching for hydraulic cylinders.
Important operating specifications for hydraulic cylinders
include the cylinder type, stroke, maximum operating pressure,
bore diameter, and rod diameter. Choices for cylinder type
include tie-rod, welded, and ram. A tie-rod cylinder is a hydraulic cylinder that uses one or more tie-rods to provide additional
stability. Tie-rods are typically installed on the outside diameter
of the cylinder housing. In many applications, the cylinder tierod bears the majority of the applied load. A welded cylinder is a
smooth hydraulic cylinder that uses a heavy-duty welded cylinder housing to provide stability. A ram cylinder is a type of
hydraulic cylinder that acts as a ram. A hydraulic ram is a device
in which the cross-sectional area of the piston rod is more than
Zhang_Ch01.indd 116
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
117
one-half the cross-sectional area of the moving component.
Hydraulic rams are primarily used to push rather than pull and
are most commonly used in high-pressure applications.
Stroke is the distance that the piston travels through the cylinder. Hydraulic cylinders can have a variety of stroke lengths,
from fractions of an inch to many feet. The maximum operating
pressure is the maximum working pressure the cylinder can sustain. The bore diameter refers to the diameter at the cylinder
bore. The rod diameter refers to the diameter of the rod or piston
used in the cylinder. Choices for cylinder configuration are simple configuration or telescopic configuration. A simple configuration hydraulic cylinder consists of a single cylindrical housing
and internal components. A telescopic configuration hydraulic
cylinder uses “telescoping” cylindrical housings to extend the
length of the cylinder. Telescopic configuration cylinders are
used in a variety of applications that require the use of a long
cylinder in a space-constrained environment.
Choices for mounting method include flange, trunnion,
threaded, clevis or eye, and foot. The mount location can be cap,
head, or intermediate. Materials of construction include steel,
stainless steel, and aluminum. Common features for hydraulic
cylinders include integral sensors, double end rod, electrohydraulic cylinders, and adjustable stroke.
(2) Hydraulic valves. Hydraulic valve actuators convert a fluid pressure supply into a motion. A valve actuator is a hydraulic actuator mounted on a valve that, in response to a signal, automatically
moves the valve to the desired position using an outside power
source. The hydraulic actuators in hydraulic valves can be either
linear like cylinders or rotary like motors. The hydraulic actuator
operates under servo-valve control; this provides regulated
hydraulic fluid flow in a closed loop system having upper and
lower cushions to protect the actuator from the effects of high
speed and high mass loads. Piston movement is monitored via a
linear voltage displacement transducer (LVDT), which provides
an output voltage proportional to the displacement of the movable core extension to the actuator.
The outside power sources used by hydraulic valves are normally in these types:
(a) electronics
(b) servo
(c) piezoelectricity.
(3) Hydraulic motors and rotary actuator. Hydraulic motors are
powered by pressurized hydraulic fluid and transfer rotational
kinetic energy to mechanical devices. Hydraulic motors, when
Zhang_Ch01.indd 117
5/13/2008 5:45:39 PM
118
INDUSTRIAL CONTROL TECHNOLOGY
powered by a mechanical source, can rotate in reverse direction
and act as a pump.
Hydraulic rotary actuators use pressurized fluid to rotate
mechanical components. The flow of pressurized hydraulic fluid
produces the rotation of moving components via a rack and pinion, cams, direct fluid pressure on rotary vanes, or other mechanical linkage. Hydraulic rotary actuators and pneumatic rotary
actuators may have fixed or adjustable angular strokes and can
include such features as mechanical cushioning, closed-loop
hydraulic dampening (oil), and magnetic features for reading by
a switch.
Motor type is the most important consideration when searching for hydraulic motors. Choices for motor type include axial
piston, radial piston, internal gear, external gear, and vane. An
axial piston motor uses an axially mounted piston to generate
mechanical energy. High-pressure flow into the motor forces the
piston to move in the chamber, generating output torque. A radial
piston hydraulic motor uses pistons mounted radially about a
central axis to generate energy. An alternate-form radial piston
motor uses multiple interconnected pistons, usually in a star pattern, to generate energy. Oil supply enters the piston chambers,
moving each individual piston and generating torque. Multiple
pistons increase the displacement per revolution through the
motor, increasing the output torque. An internal gear motor uses
internal gears to produce mechanical energy. Pressurized fluid
turns the internal gears, producing output torque. An external
gear motor uses externally mounted gears to produce mechanical
energy. Pressurized fluid forces the external gears to turn, producing output torque. A vane motor uses a vane to generate
mechanical energy. Pressurized fluid strikes the blades in the
vane, causing it to rotate and produce output torque.
Additional operating specifications to consider for hydraulic
motors include operating torque, operating pressure, operating
speed, operating temperature, power, maximum fluid flow, maximum fluid viscosity, displacement per revolution, and motor
weight. The operating torque is the torque that the motor is capable of delivering. Operating torque depends directly on the pressure of the working fluid delivered to the motor. The operating
pressure is the pressure of the working fluid delivered to the
hydraulic motor. Working fluid is pressurized by an outside
source before it is delivered to the motor. Working pressure
affects operating torque, speed, flow, and horsepower of the
motor. The operating speed is the speed at which the hydraulic
motors’ moving parts rotate. Operating speed is expressed in
Zhang_Ch01.indd 118
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
119
revolutions per minute or similar terms. The operating temperature is the fluid temperature range the motor can accommodate.
Minimum and maximum operating temperatures are dependent
on the internal component materials of the motor and can vary
greatly between products. The power the motor is capable of
delivering is dependent on the pressure and flow of the fluid
through the motor. The maximum volumetric flow through the
motor is expressed in terms of gallons per minute or similar units.
The maximum fluid viscosity the motor can accommodate is a
measure of the fluid’s resistance to shear, and is measured in
centipoises (cP), a common metric unit of dynamic viscosity
equal to 0.01 poise or 1 mP. The dynamic viscosity of water at
20°C is about 1 cP (the correct unit is cP, but cPs and cPo are
sometimes used). The fluid volume displaced per revolution of
the motor is measured in cubic centimeters (cc) per revolution,
or similar units. The weight of the motor is measured in pounds
or similar units.
1.2.3.3 Application Guide
(1) Proactive maintenance for hydraulic cylinders. Damaged hydraulic cylinder rods and wiper seals are an eternal problem for users
of hydraulic machinery. Dents and gouges on the surface of
hydraulic cylinder rods reduce seal life and give dust and other
contaminants an easy path into the hydraulic system. These siltsized particles act like lapping compound, initiating a chain of
wear in hydraulic components.
The top four causes of hydraulic seal failure in cylinders are
the following:
(a) Improper installation is a major cause of hydraulic seal failure. The important things to watch during seal installation
are (1) cleanliness, (2) protecting the seal from nicks and
cuts, and (3) proper lubrication. Other problem areas are
over tightening of the seal gland where there is an adjustable
gland follower or folding over a seal lip during installation.
Installing the seal upside down is a common occurrence, too.
The solution to these problems is commonsense and taking
reasonable care during assembly.
(b) Hydraulic system contamination is another major factor in
hydraulic seal failure. It is usually caused by external elements such as dirt, grit, mud, dust, ice, and internal contamination from circulating metal chips, breakdown products of
fluid, hoses, or other degradable system components. As
most external contamination enters the system during rod
Zhang_Ch01.indd 119
5/13/2008 5:45:39 PM
120
INDUSTRIAL CONTROL TECHNOLOGY
retraction, the proper installation of a rod wiper and scraper
is the best solution. Proper filtering of system fluid can prevent internal contamination. Contamination is indicated by
scored rod and cylinder bore surfaces, excessive seal wear
and leakage, and sometimes, tiny pieces of metal imbedded
in the seal.
(c) Chemical breakdown of the seal material is most often the
result of incorrect material selection in the first place, or a
change of hydraulic system fluid. Misapplication or use of
noncompatible materials can lead to chemical attack by fluid
additives, hydrolysis, and oxidation–reduction of seal elements. Chemical breakdown can result in loss of seal lip
interface, softening of seal durometer, excessive swelling, or
shrinkage. Discoloration of hydraulic seals can also be an
indicator of chemical attack.
(d) Heat degradation is to be suspected when the failed seal
exhibits a hard, brittle appearance and/or shows a breaking
away of parts of the seal lip or body. Heat degradation results
in loss of sealing lip effectiveness through excessive compression set and/or loss of seal material. Causes of this condition may be use of incorrect seal material, high dynamic
friction, excessive lip loading, no heel clearance, and proximity to outside heat source. Correction of heat degradation
problems may involve reducing seal lip interference, increasing lubrication, or a change of the seal material. In borderline situations, consider all upper temperature limits to be
increased by 50°F in hydraulic cylinder seals at the seal
interface due to running friction caused by the sliding action
of the lips.
In response to this problem, a protective cylinder rod cover
called Seal Saver has been developed and patented. Seal
Saver is a continuous piece of durable material, which wraps
around the cylinder and is closed with Velcro. It is then
clamped onto the cylinder body and rod end. This makes
installation simple with no disassembly of hydraulic cylinder components required.
Seal Saver forms a protective shroud over the cylinder rod
as it strokes and prevents build-up of contaminants around
the wiper seal that is a common cause of rod scoring, seal
damage, and contaminant ingress. Research has shown that
the cost to remove contaminants is 10 times the cost of exclusion. This, combined with the benefits of extended hydraulic
cylinder rod and seal life, makes Seal Saver a cost-effective,
proactive maintenance solution.
Zhang_Ch01.indd 120
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
121
(2) Hydraulic cylinder rods maintenance. As a product group,
hydraulic cylinders are almost as common as pumps and motors
combined. They are less complicated than other types of hydraulic components and are therefore relatively easy to repair. As a
result, many hydraulic equipment owners or their maintenance
personnel repair hydraulic cylinders in-house. An important step
in the repair process that is often skipped is the checking of rod
straightness.
Bent rods load the rod seals causing distortion, and ultimately
premature failure of the hydraulic cylinder seals. Rod straightness should always be checked when hydraulic cylinders are
being resealed or repaired. This is done by placing the rod on
rollers and measuring the run-out with a dial gauge. The rod
should be as straight as possible, but a run-out of 0.5 mm per
linear meter of rod is generally considered acceptable. In most
cases, bent rods can be straightened in a press. It is sometimes
possible to straighten hydraulic cylinder rods without damaging
the hard-chrome plating; however, if the chrome is damaged, the
rod must be either rechromed or replaced.
Black nitride is a relatively recent alternative to the hard
chrome-plated hydraulic cylinder rod. With reports of achieved
service life three times that of conventional chrome, longer seal
life, and comparable cost, black nitride rods for hydraulic cylinders are an option that all hydraulic equipment users should be
aware of. Black nitride is an atmospheric furnace treatment
developed and patented in the early 1980s. It combines the high
surface hardness and corrosion resistance of nitride with additional corrosion resistance gained by oxidation. The process
begins with the cleaning and superpolishing of the material to a
surface roughness of 6–10 Ra. The steel bars or tubes are then
fixed vertically and lowered into an electrically heated pit
furnace.
(3) Other maintenances for hydraulic cylinders. Hydraulic cylinders
are compact and relatively simple. The key points to watch are
the seals and pivots. The following lists service tips in maintaining cylinders:
(a) External leakage. If the end caps of a cylinder are leaking,
tighten them. If the leaks still do not stop, replace the gasket.
If a cylinder leaks around a piston rod, replace the packing.
Make sure that a seal lip faces toward the pressure oil.
(b) Internal leakage. Leakage past the piston seals inside a cylinder can cause sluggish movement or settling under load.
Piston leakage can be caused by worn piston seals or rings or
scored cylinder walls. The latter may be caused by dirt and
Zhang_Ch01.indd 121
5/13/2008 5:45:39 PM
122
INDUSTRIAL CONTROL TECHNOLOGY
grit in the oil. When repairing a cylinder, replace all the seals
and packing before reassembly.
(c) Creeping cylinder. If a cylinder creeps when stopped in middle stroke, check for internal leakage. Another cause could
be a worn control valve.
(d) Sluggish operation. Air in a cylinder is the most common
cause of sluggish action. Internal leakage in a cylinder is
another cause. If an action is sluggish when starting up a
system, but speeds up when the system is warm, check for
oil of too high a viscosity (see the machine’s operating manual). If a cylinder is still sluggish after these checks, test the
whole circuit for worn components.
(e) Loose mounting. Pivot points and mounts may be loose. The
bolts or pins may need to be tightened, or they may be worn
out. Too much slop or float in a cylinder’s mountings damages the piston-rod seals. Periodically check all the cylinders
for loose mountings.
(f) Misalignment. Piston rods must work in-line at all times. If
they are side-loaded, the piston rods will be galled and the
packing will be damaged, causing leaks. Eventually, the piston rods may be bent or the welds broken.
(g) Lack of lubrication. If a piston rod has no lubrication, a rod
packing could seize, which would result in an erratic stroke,
especially on single-acting cylinders.
(h) Abrasives on a piston rod. When a piston rod extends, it can
pick up dirt and other material. When it retracts, it carries the
grit into a cylinder, damaging a rod seal. For this reason, rod
wipers are often used at the rod end of a cylinder to clean the
rod as it retracts. Rubber boots are also used over the end of
a cylinder in some cases. Piston-rod rusting is another problem. When storing cylinders, the piston rods are always
retracted to protect them.
(i) Burrs on a piston rod. Exposed piston rods can be damaged
by impact with hard objects. If a smooth surface of a rod is
marred, a rod seal may be damaged. Clean the burrs on a rod
immediately, using crocus cloth. Some rods are chromeplated to resist wear. Replace the seals after restoring a rod
surface.
(j) Air vents. Single-acting cylinders (except ram types) must
have an air vent in the dry side of a cylinder. To prevent dirt
from getting in, use different filter devices. Most are selfcleaning, but inspect them periodically to ensure that they
operate properly.
Zhang_Ch01.indd 122
5/13/2008 5:45:39 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
1.2.3.4
123
Calibration
Many applications in modern industrial control require numerous
hydraulic actuators that direct the flow of hydraulic fluid between the system components when necessary. For example, a typical antilock brake
system in cars can include several hydraulic actuators to control the fluid
pressure in the individual components such as a master cylinder and a
plurality of wheel cylinders. The memory of an actuator control system
includes numerous look-up tables which allow the control system to know
what electric signals, such as current values, to apply to the actuators in
order to yield specific actuation pressures. Typically, these look-up tables
are generic tables that are not tailored to the individual actuators in the
fluid system. These generic tables are created to account for worst-case
part-to-part variances, manufacturing variances, and system variances.
Thus, the tolerances of the values contained in the tables are relatively
large and result in less than optimal performance of the actuators. Although
very expensive actuators can be used to decrease part-to-part variances,
the overall system tolerances remain larger. However, less tolerance can be
obtained by individually calibrating less expensive actuators to create customized look-up tables for each actuator.
The following introduces two calibration examples for hydraulic
actuators:
(1) Stroke calibration for actuator valve. The first example is stroke
calibration for a hydraulic actuator for valves. Figure 1.34 illustrates the working block for this kind of hydraulic actuator that
consists of a piston and spring. The spring, which can be preloaded, trends to keep the piston at the initial position. As pressure applied to the piston develops enough force to overcome the
spring preload, the piston moves to the opposition until it reaches
its maximum stroke.
A
p
s
Valve
Figure 1.34 The working block of a hydraulic actuator for valve (courtesy of
Siemens).
Zhang_Ch01.indd 123
5/13/2008 5:45:39 PM
124
INDUSTRIAL CONTROL TECHNOLOGY
To determine the stroke positions 0 and 1 in the valve, calibration is required when the valve and actuator are commissioned
for the first time. For this purpose, the actuator must be mechanically connected to the valve and supplied with a standard voltage
of electric power. The calibration procedure can be repeated as
often as necessary. Normally, there is a slot on the printed circuit
boards of many actuators. In most cases, to initiate the calibration procedure, the contacts inside this slot must be shortcircuited by, say, a screwdriver. The calibration thus can proceed
automatically:
(a) actuator runs to the zero stroke position: valve closes, green
LED flashes;
(b) actuator then runs to the 100 stroke position;
(c) measured values are stored;
(d) the calibration procedure is finished, and green LED flashes;
(e) the actuator now moves to the position defined by control
signal of its controller.
(2) Resistance calibration for hydraulic pump. Figure 1.35 is the
schematic diagram of a hydraulic actuator for pumping oil.
S
2
1
C1
Pump
C2
3
3
R
4
4
5
Figure 1.35 Schematic diagram of a hydraulic actuator. The pump (1) supplies a
steady flow of oil to the supply point of a hydraulic Wheatstone bridge labeled as
S. The oil continuously flows through the bridge to the return point labeled as R
and is finally returned to the pump station. The four variable flow restrictors in this
bridge are contained in a valve unit. In this diagram, 2 indicates a pneumatic
valve; 4 gives two bellows; 5 is the actuator plate.
Zhang_Ch01.indd 124
5/13/2008 5:45:40 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
125
The goal of the calibration is for all four nozzles to have a nominal flow resistance measured by pascals per second per cubic
meter. The test should be done at the nominal pressure and flow
rate and with the fluid running from the supply to the return
(Fig. 1.35). Before starting calibration, ensure that (1) the resistance of each nozzle is equal; (2) the pressure at the two control
ports is the same when the electric drive is 0 A; (3) at 0 A drive,
the pressure of the control port is half the pressure drop between
the supply and the return; (4) the flow through the two sides
of the bridge is the same; (5) the valves must be all the same;
(6) for each side of the valve, calibration proceeds independently.
1.2.4
Piezoelectric Actuators
Piezoelectric actuators represent an important new group of actuators
for active control of mechanical systems. Although the magnitudes of
piezoelectric voltages, movements, or forces are small, and often require
amplification (e.g., a typical disc of piezoelectric ceramic will increase or
decrease in thickness by only a small fraction of a millimeter), piezoelectric materials have been adapted to an impressive range of applications
requiring small amounts of displacement (typically less than a few thousandths of an inch of displacement).
Today, modern polycrystalline piezoelectric ceramic is mass produced
for applications including underwater transducers, point level sensors,
medical products, ultrasonic cleaners, actuators, fish finders, and motors.
The piezoelectric effect of the piezoelectric ceramic is used in sensing
applications, such as accelerometers, sensors, flow meters, level detectors,
and hydrophones as well as in force or displacement sensors. Its inverse
piezoelectric effect is used in actuation applications, such as in motors and
devices that precisely control positioning, and in generating sonic and
ultrasonic signals.
Piezoelectric actuators can be used for the conversion of electrical
energy to mechanical movement, for accurate positioning down to nanometer levels, for producing ultrasonic energy and sonar signals, and for the
conversion of pressure and vibration into electrical energy. Piezoelectric
actuators can also be manufactured in a variety of configurations and fabrication techniques. The industry recognizes these devices as monomorphs,
bimorphs, stacks, cofired actuators, and flexure elements. Piezoelectric
actuators are found in telephones, stereo music systems, and musical
instruments such as guitars and drums. The use of piezoelectric actuators
is beginning to appear in endoscope lenses used in medical treatment.
Piezoelectric actuators are also being used for valves in drilling equipment
Zhang_Ch01.indd 125
5/13/2008 5:45:40 PM
126
INDUSTRIAL CONTROL TECHNOLOGY
at offshore oil fields. Piezoelectric actuators are also used to control
hydraulic valves, act as small-volume pumps or special-purpose motors,
and in other applications.
The use of piezoelectric actuators is thus steadily growing in advanced
fields where conventional actuators are no longer effective. At present,
however, this is no more than just a beginning. Piezoelectric actuators that
combine a number of superior characteristics will continue to evolve into
powerful devices that support our society in the future.
1.2.4.1
Operating Principle
(1) Piezoelectricity. In 1880, Jacques and Pierre Curie discovered an
unusual characteristic of certain crystalline minerals: when subjected to a mechanical force, the crystals became electrically
polarized. Tension and compression generated voltages of opposite polarity, and in proportion to the applied force. Subsequently,
the converse of this relationship was confirmed: if one of these
voltage-generating crystals was exposed to an electric field it
lengthened or shortened according to the polarity of the field,
and in proportion to the strength of the field. These behaviors
were called the piezoelectric effect and the inverse piezoelectric
effect, respectively.
The findings by Jacques and Pierre Curie have been more and
more confirmed since then. Many polymers, ceramics, and molecules such as water are permanently polarized; some parts of
the molecule are positively charged, whereas other parts of the
molecule are negatively charged. This behavior of piezoelectric
materials is depicted in Fig. 1.36(a). When the material changes
dimensions as a result of an imposed mechanical force, a permanently polarized material such as quartz (SiO2) or barium titanate (BaTiO3) will produce an electric field. This behavior of
piezoelectric materials when subject to an imposed force is
depicted in Fig. 1.36(b). Furthermore, when an electric field is
applied to these materials, the polarized molecules within them
will align themselves with the electric field, resulting in induced
dipoles within the molecular or crystal structure of the material
as illustrated in Fig. 1.36(c).
Piezoelectricity involves the interaction between the electrical
and mechanical behaviors of the material. Static linear relations
between two electrical and mechanical variables have approximated this interaction:
S = SET + dE,
D = dT + eTE,
Zhang_Ch01.indd 126
5/13/2008 5:45:40 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
Domain
Dipole
127
Stress
–
Voltage
+
Electric
field
Strains
Electric
strains
Stress
(a)
(b)
(c)
Figure 1.36 Behaviors of piezoelectric materials: (a) nonpolarized state when
no force and no electricity are applied, (b) polarized state when compression
stresses are imposed, and (c) polarized state when electric field is applied after
poling.
where S is the strain tensor, T is the stress tensor, E is an electric
field vector, D is the electric displacement vector, SE is the elastic
compliance matrix when subjected to a constant electric field
(the superscript E denotes that the electric field is constant), d is
the matrix of piezoelectric constants, and eT is the permittivity
measured at constant stress.
The piezoelectric effect is, however, very nonlinear in nature.
Piezoelectric materials exhibit, for example, a strong hysteresis
and drift that is not included in the above model. It should be
noted, too, that the dynamics of the material are not described by
the two equations above.
(2) Piezoelectric actuator. The three basic types of piezoelectric
actuators are stacks, linear motors, and benders.
(a) Piezoelectric stack actuators. The linear motion produced
by the piezoelectric effect has been used for making a stack
actuator, which is a multilayer construction: each stack
is composed of several piezoelectric layers, as depicted in
Fig. 1.37. The required dimensions of the stack can be easily
determined from the requirements of the application in question. The height is determined with respect to the desired
movement and the cross-sectional area with respect to the
desired force.
The main problem of stack actuators is the relatively small
strain (0, 1–0, 2%) obtained. Using, for example, levers or
hydraulic amplifiers can increase the movement. It is noticeable that, in addition to the desired longitudinal movement,
Zhang_Ch01.indd 127
5/13/2008 5:45:40 PM
128
INDUSTRIAL CONTROL TECHNOLOGY
–
+
Figure 1.37 Structure of a piezoelectric stack.
some lateral movement typically also occurs, which causes
the bias of the piezoelectric stack to be away from its straight
line. Therefore, a guide has to be used if only longitudinal
motion is desired. Figure 1.38 illustrates deviations from
straight line accuracy.
(b) Linear motors. Since the strain of piezoelectric ceramics is
relatively small, displacement amplifiers or hybrid structures
are needed. There are many amplification techniques such as
levers and hydraulic systems, and piezoelectric motors. In
the lever system, amplification is achieved with lever arms
that magnify the displacement. The output force of the lever
system is significantly smaller than the actuator force.
Hydraulic systems generally use a piston for amplification.
The principle of the piezohydraulic actuator is illustrated in
Fig. 1.39, which develops a hydraulic amplifier based on the
use of bellows. This kind of piezohydraulic motor uses a linear piezoelectric actuator to control the liquid input to the
fluid chamber which drives the bellows.
Piezoelectric motors increase displacement by providing
many small steps. There are many different types of linear
piezoelectric motors: the main categories are linear stepper
motors and ultrasonic motors. The linear steppers include an
inchworm motor, a stick and slip actuator, and an impact
V
Straightness
H
Flatness
Figure 1.38 Straight line accuracy of piezoelectric stack.
Zhang_Ch01.indd 128
5/13/2008 5:45:41 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
129
Fluid chamber
Piezoelectric actuator
Bellows
Figure 1.39 Schematic of piezohydraulic actuator.
drive motor. The ultrasonic motors can be divided into
standing wave and traveling wave ultrasonic motors. The
operating principles of the inchworm motor, the stick and
slip actuator, and the traveling wave ultrasonic motor are
described below.
(i) Inchworm motors. Inchworm motors are a kind of linear motor in which the linear movement is achieved by
using three piezoelectric elements. The operation principle is illustrated in Fig. 1.40. The outer piezoelectric
elements work as clamps. The contractions and expansions of the middle element generate the movement of
the motor rod.
(ii) Stick and slip actuators. The stick and slip actuator is a
type of an inertia device that uses inertia of the moving
mass. The actuator consists of particular legs and a
slider. Each step consists of a slow deformation of the
legs and fast jump backward. In slow deformation of the
legs, the moving mass follows the legs due to friction
(the frictional force is higher than the force caused by
the slider inertia). In the sudden jump backward, the
slider cannot follow the legs due to its inertia. Figure 1.41
shows the operating principle of stick and slip actuator.
(iii) Traveling wave ultrasonic motors. A voltage having two
phases drives the traveling wave ultrasonic motor. The
voltage is applied to the piezoelectric element at the
resonance frequency. The resonance frequency produces
a traveling wave. The particles on the surface move
along the elliptical trajectories. The motion of the
Zhang_Ch01.indd 129
5/13/2008 5:45:41 PM
130
INDUSTRIAL CONTROL TECHNOLOGY
2
1
3
OFF
Clamp element 1
Extend element 2
Camp element 3
Uncamp element 1
Contract element 2
Camp element 1
Uncamp element 3
Figure 1.40 Operation processes of inchworm motor.
Zhang_Ch01.indd 130
5/13/2008 5:45:42 PM
131
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
Slider
Voltage
2
Leg
Stick::1
Time
1
Slip::2
Figure 1.41 Operation principle of the stick and slip actuator.
particles is on the opposite direction of the wave. When
a moving body (rotor) is placed in contact with the surface, it moves in the same direction as the particles due
to the frictional force produced between the moving
body and the elastic body. The ultrasonic piezoelectric
motor’s faster response times, higher precision, hard
brake with no backlash, high power-to-weight ratio,
and smaller packaging envelope more than compensate
for the lack of brute horsepower and speed associated
with its electromagnetic motor counterparts.
(c) Piezoelectric benders. Piezoelectric bending actuators (or
piezoelectric cantilevers or piezoelectric bimorphs) bear a
close resemblance to bimetallic benders. The application of
an electric field across the two layers of the bender results in
the expansion of one layer while the other contracts. The net
result is a curvature much greater than the length or thickness deformation of the individual layers. With a piezoelectric bender, relatively high displacements can be achieved,
but at the cost of force and speed. There are some benders
that have only one piezoelectric layer on top of a metal layer,
but generally there are two piezoelectric layers and no metal.
Two piezoelectric layers make the displacement double in
comparison to a single layer version. If the number of piezoelectric layers exceeds two, the bender is referred to as a
multilayer. With thinner piezoelectric layers, a smaller voltage is required to produce the same electric field strength,
and the benefit of the multilayer benders is, therefore, lower
operating voltage. Multilayer benders can be built into one
of these two types: a serial or parallel bender. In a serial
Zhang_Ch01.indd 131
5/13/2008 5:45:42 PM
132
INDUSTRIAL CONTROL TECHNOLOGY
bender, there are two piezoelectric layers with an antiparallel
polarization connected to each other, and two surface electrodes. In this arrangement, one of the electrodes is connected to the ground and the other to the output of a bipolar
amplifier. Figure 1.42 gives the schematic of a parallel bender
in operation. Parallel benders can be distinguished from
serial benders by their three electrodes. In between the two
parallel-polarized piezoelectric layers is a middle electrode
to which the actual control signal is supplied. The two surface electrodes are connected to the ground and to a fixed
voltage. The control voltage is applied to the middle electrode,
and it varies between zero and a fixed voltage (Fig. 1.42).
The parallel bender can also be connected in such a way that
the two surface electrodes are connected to the ground and a
bipolar signal is applied to the middle electrode.
1.2.4.2
Basic Types
Piezoelectric devices make use of direct and inverse piezoelectric
effects to perform a function. Both these piezoelectric effects are found in
the crystal structures of some materials. For example, ceramics acquire a
charge when being compressed, twisted or distorted and produced physical displacements when electric voltages are imposed. Several types of
devices are available. Some of the most important are listed here:
(1) Piezoelectric actuators. Piezoelectric actuators are devices that
produce a small displacement with a high force capability when
voltage is applied. There are many applications where a piezoelectric actuator may be used, such as ultra-precise positioning
and in the generation and handling of high forces or pressures in
static or dynamic situations.
+Ur
Displacement
Electrodes
Pr
PZT layers
0
Figure 1.42 Schematic of a parallel bender in operation.
Zhang_Ch01.indd 132
5/13/2008 5:45:42 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
133
Actuator configuration can vary greatly depending on application. Piezoelectric stack or multilayer actuators are manufactured
by stacking up piezoelectric disks or plates, the axis of the stack
being the axis of linear motion when a voltage is applied. Tube
actuators are monolithic devices that contract laterally and longitudinally when a voltage is applied between the inner and outer
electrodes. A disk actuator is a device in the shape of a planar
disk. Ring actuators are disk actuators with a center bore, making the actuator axis accessible for optical, mechanical, or electrical purposes. Other less common configurations include block,
disk, bender, and bimorph styles.
These devices can also be ultrasonic. Ultrasonic actuators are
specifically designed to produce strokes of several micrometers
at ultrasonic (>20 kHz) frequencies. They are especially useful
for controlling vibration, positioning applications, and quick
switching. In addition, piezoelectric actuators can be either direct
or amplified. The effect of amplification is not only larger displacement, but it can also result in slower response times.
The critical specifications for piezoelectric actuators are displacement, force, and operating voltage of the actuator. Other
factors to consider are stiffness, resonant frequency, and capacitance. Stiffness is a term used to describe the force needed to
achieve a certain deformation of a structure. For piezoelectric
actuators, it is the force needed to elongate the device by a certain amount. It is normally specified in terms of newtons per
micrometer. Resonance is the frequency at which the actuators
respond with maximum output amplitude. The capacitance is a
function of the excitation voltage frequency.
(2) Piezoelectric motors. Piezoelectric motors use a piezoelectric
ceramic element to produce ultrasonic vibrations of an appropriate type in a stator structure. The elliptical movements of the
stator are converted into the movement of a slider pressed into
frictional contact with the stator. The consequent movement may
either be rotational or linear depending on the design of the structure. Linear piezoelectric motors typically offer one degree of
freedom, such as in linear stages. However, these devices can be
combined to provide more complex positioning factors. Rotating
piezoelectric motors are commonly used in submicrometric positioning devices. Large mechanical torque can be achieved by
combining several of these rotational units.
Piezoelectric motors have a number of potential advantages
over conventional electromagnetic motors. They are generally
small and compact when compared with their power output, and
provide greater force and torque than their dimensions would
Zhang_Ch01.indd 133
5/13/2008 5:45:43 PM
134
INDUSTRIAL CONTROL TECHNOLOGY
seem to indicate. In addition to a very positive size to power
ratio, piezoelectric motors have high holding torque maintained
at zero input power, and they offer low inertia from their rotors,
providing rapid start and stop characteristics. Additionally, they
are unaffected by electromagnetic fields, which can hamper other
motor types.
Piezoelectric motors usually do not produce magnetic fields
and they are not affected by external magnetic fields either.
Because they operate at ultrasonic frequencies, these motors do
not produce sound during operation. However, piezoelectric
motors do have some disadvantages.
These disadvantages include the need for high voltage, highfrequency power sources, and the possibility of wear at the rotor/
stator interface, which tends to shorten their service life.
Piezoelectric motors have been in industrial use for years, but
have not been popular due to what was perceived as an exorbitant
cost of production and use. However, recent advances have significantly reduced the channel cost of this technology for closedloop systems that require high positioning accuracy. With the use
of a wide range of controllers and/or position sensors, the list of
piezoelectric motor product applications is constantly growing.
Some of the common applications for piezoelectric motors include
camera focus systems, computer disk drives, material handling,
robotics, and semiconductor testing and production systems.
(3) Multilayer piezoelectric benders. High efficiency, low-voltage
multilayer benders have been developed to meet the growing
demand for precise, controllable, and repeatable deflection devices
in the millimeter and micrometer range. Multilayer piezoelectric
ceramic benders are devices capable of rapid (<10 ms) millimeter
movements with micrometer precision. They utilize the inverse
piezoelectric effect, in which an electric field creates a cantilever
bending effect. By making the ceramic layers very thin, between
20 and 40 µm, deflections can be generated with low power consumption at operating voltages from –10 to +60 V. With an electrical field of <3 kV/mm large deflections per unit volume can be
achieved with high reliability.
Note: A single bender cannot combine maximum deflection
and maximum blocking force! Typical applications: proportional
valves, low-energy consuming switches, or pumps.
(4) Piezoelectric drivers and piezoelectric amplifiers. Piezoelectric
drivers and piezoelectric amplifiers are developed to match the
requirements for driving and controlling piezoelectric actuators
and stages in some applications. Standard linear amplifier products are simple voltage followers that amplify a low-voltage
Zhang_Ch01.indd 134
5/13/2008 5:45:43 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
135
input signal, and others are recommended for use as integrated
or stand-alone systems in applications that require advanced
capabilities for closed-loop servo control requirements. A voltage
amplifier is typically needed to control piezoelectric actuators
due to the high operating voltage needed for the former. In other
words, before the computer, through a DA converter, provides
the control signal, it must be amplified. This section describes
the most important piezoelectric amplifier characteristics such as
voltage range, peak and average currents, a slew rate, power efficiency, and noise.
For bench top products, menu-driven user interfaces through
front panel LCD and input dial enable amplifier setup, monitoring, and configuration. A built-in sinusoidal function generator
capability is available for these models. Standard amplifier
bandwidth is <1.2 kHz (–3 dB). Their features include RS-232
serial communication, digital I/O, analog or digital feedback
for closed-loop input signals, a 24-bit analog input, and userconfigurable PID gain parameter settings. The serial communication capability allows the user to configure and monitor system
parameters, to command the desired target position, and to query
actual position.
The output voltage range is perhaps the most important
property of the amplifier because it either limits the range of
displacement when being too small or decreases the displacement resolution when being too large. In addition to the supply
voltage range, an important property is the current-driving capability of the amplifier. This together with the capacitance of
the piezoelectric actuator determines the maximum operating
frequency.
For most amplifiers, both the peak and the average current
limits are given. With capacitive loads, such as piezoelectric
actuators, the peak current is more important but average current
cannot be forgotten. The required peak and average currents ratio
show a fixed ratio of approximately 3:1 for a sine oscillation, for
example. One aspect to consider is the power efficiency of the
supplied power. This is important especially in portable devices,
in devices that have wireless power supply, and in devices operating on high frequencies.
Piezoelectric actuators have theoretically an unlimited resolution. Therefore, every infinitely small voltage step caused by the
noise of the amplifier, for example, is transformed into an infinitely small mechanical shift. Therefore, an important property
of the amplifier when designing a precision positioning system is
the noise characteristics of the amplifier.
Zhang_Ch01.indd 135
5/13/2008 5:45:43 PM
136
INDUSTRIAL CONTROL TECHNOLOGY
1.2.4.3 Technical Specifications
(1) Piezoelectric actuator configuration. Your choices are
(a) Stack. Piezoelectric stack actuators are manufactured by
stacking up piezoelectric disks or plates. These disks are
electrically connected. The stack axis is the axis of the linear
motion. When a voltage is applied, the thickness of the layers increases and thereby the total stack lengthens.
(b) Tube. Piezoelectric tube actuators are monolithic devices
that contract laterally and longitudinally when a voltage is
applied between the inner and outer electrodes. With quadrature electrodes the tubes can be operated as XY scanners.
(c) Disk. A disk actuator is a device in the shape of a planar disk.
(d) Ring. Ring actuators are disk actuators with a center bore.
This makes the actuator axis accessible for optical, mechanical, or electrical purposes.
(e) Other. Other unlisted configurations are specialized or proprietary configurations, such as block, disk, bender, bimorph, etc.
(2) Performance specifications
(a) Maximum displacement. The maximum elongation (normally specified in meters) that the actuator will produce
when the maximum operating voltage is applied.
(b) Blocked force. The maximum force (normally specified in
newtons, N) that the actuator will produce when the maximum operating voltage is applied.
(c) Maximum operating voltage. The maximum voltage that can
be applied to the actuator without impairing its functionality.
(d) Stiffness. Stiffness is a term used to describe the force needed
to achieve a certain deformation or deflection of a structure.
For piezoelectric actuators, it is the force needed to elongate
the device by a certain amount. It is normally specified in
terms of newtons per meter, or N/m.
(e) Resonance frequency. Resonance is the frequency at which
the actuator responds with maximum output amplitude.
(f) Capacitance. The capacitance is that which the actuator
exhibits. The capacitance is a function of the excitation voltage frequency.
(3) Ultrasonic operation. Ultrasonic actuators are specifically
designed to produce strokes of several micrometers at ultrasonic
(>20 kHz) frequencies.
(4) Electrical connectors. Your choices are
(a) DB-9. Similar in appearance to a CAPITOL D, D-subminiature
connectors are generally referred to by the number of pins or
sockets they have, for example, DB-9, DB-25, etc. The design
Zhang_Ch01.indd 136
5/13/2008 5:45:43 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
137
of these connectors varies little among manufacturers, except
for the color of the shell.
(b) BNC. The BNC is essentially a miniature version of the C
connector that is a bayonet version of the N connector. BNC
connectors are available in both 50 and 75 Ω versions; both
versions will mate together. The 50 Ω designs operate up to
a frequency of 4 GHz. BNC connectors are used in many
applications, some of which are flexible networks, instrumentation, and computer peripheral interconnections.
(c) Two wires AWG26. Two 15.9-mils diameter (AWG26) wires.
(d) Two wires AWG30. Two 10.0-mils diameter (AWG30) wires.
(e) LEMO connector. LEMO is a precision push–pull locking
connector for demanding applications. LEMO is a registered
trademark of LEMO.
(f) Other. Other includes those unlisted, specialized, or proprietary connector types.
1.2.4.4
Calibration
Under normal environments, piezoelectricity is often stable in physics.
Those devices working with piezoelectricity, including actuators, sensors,
and motors, are therefore stable, and their calibrated performance characteristics do not change over time under normal environmental conditions.
However, often these devices are exposed to harsh environmental conditions, like mechanical shock, temperature changes, humidity, etc., which
basically generate three groups of errors in piezoelectric devices:
(1) Sensitivity. Errors that include calibration errors, linearity errors,
frequency and phase response errors, aging errors, temperature
coefficients.
(2) Coupling. Errors that include influence of transducer weight,
quality of the coupling surfaces, transverse sensitivity.
(3) Noise and environmental influences that include noise, base
strain, magnetic fields, temperature transients, sound pressure,
cable motion, electromagnetic interference in cables.
In order to correct these errors it is necessary to establish a recalibration
cycle. For applications where high accuracy is required, we recommend
recalibrating the piezoelectric devices every time after use under severe
conditions or at least every 2 years. Nevertheless, in some less critical
applications, for example, in machine monitoring, recalibration may be
unnecessary.
Zhang_Ch01.indd 137
5/13/2008 5:45:43 PM
138
INDUSTRIAL CONTROL TECHNOLOGY
To recalibrate these piezoelectric devices, many companies choose to
purchase their own calibration equipment to perform recalibration themselves. This may save calibration cost, particularly if a large number of
piezoelectric devices is used. When no calibrator is at hand, a measuring
chain can be calibrated by one of the following techniques:
(1) adjusting the amplifier gain to the required sensitivity of the
piezoelectric devices;
(2) typing in the stated sensitivity when using a computer-based data
acquisition system;
(3) replacing the piezoelectric device by a generator signal and measuring the equivalent magnitude.
However, due to the limitations of calibration, the uncertainty of calibration may not be better than ±2% by means of these three techniques.
These errors also cause systematic errors in industrial control. For
the evaluation of systematic errors it is very important to assess their
contribution from all relevant error sources. This is of particular importance for unknown and undetectable systematic errors. Most errors, however, will occur accidentally in an unpredictable manner. They cannot be
compensated for by a simple mathematical model since their amount and
their process of formation are unknown. For practical measurements, systematic errors and accidental errors are combined in one quantity called
measuring uncertainty.
The following example illustrates the contribution of several error components and their typical amounts:
(1) Accelerometer:
(a) calibration error 2%
(b) frequency error (band limits at 5% deviation) 5%
(c) linearity error 2%
(d) external influences 5%.
(2) Instrument with mathematical model calculation:
(a) basic error 1%
(b) frequency error (band limits at 5% deviation) 5%
(c) linearity error 1%
(d) waveform error 1%.
Piezoelectric actuators have exceptional linearity when properly mounted,
with typical 0.5% value of full scale output (FSO). New users may sometimes become confused about how to mount and calibrate the piezoelectric
actuators. Piezoelectric actuators may be used at multiple incremental ranges
Zhang_Ch01.indd 138
5/13/2008 5:45:43 PM
139
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
up to their maximum measuring range. Therefore, selecting the proper actuator really depends on the size and mechanical constraints of the apparatus
under test. In general, there are two types of piezoelectric actuator:
(1) internally preloaded actuators
(2) ring-style actuators that require external preloading.
Internally preloaded actuators do not require any preloading, whereas
ring-style actuators require preloading during installation. Ring-style actuators must be preloaded to approximately 20% or more of their measuring
range in order to obtain the best possible linearity. This linearity is achieved
by tightly clamping the internal components (piezoelectric material and
housing) together (see Fig. 1.43). The preload also acts to limit slippage of
the actuator caused by side loads experienced during use. Tension measurements are also possible if the actuator has been mounted with proper
preload. This allows the ability to measure tensile and compressive loads
with one actuator. This is how actuators are constructed.
Preloading is also required for shear force measurements using threeaxis actuators. The preload generates the required friction between the
actuator and mating surfaces in order to transmit the shear forces. The
required preload force is calculated as:
Force of preload = Force of shear/the coefficient of friction.
A typical value for the coefficient of friction is 0.13. Thus, the required
preload is at least 7.7 times the desired shear force. Some manufacturers of
Sensor output
Linear portion of output
Applied load
20% of FSO
FSO
Figure 1.43 Internally preloaded actuators (sensor) do not require any preloading
(courtesy of PCB PIEZOTRONICS, Inc.).
Zhang_Ch01.indd 139
5/13/2008 5:45:43 PM
140
INDUSTRIAL CONTROL TECHNOLOGY
piezoelectric actuators recommend 10 times the desired shear force. Force
rings that require preload are calibrated and shipped with a standard
mounting stud. This stud is specially designed to stretch, yet still maintain
a very high tensile strength beyond the force ring measurement range. The
stretching action of the stud has been designed to allow the force ring to
maintain the best possible sensitivity. The stiffer the stud, the more force it
takes away from the actuator, thus effectively reducing the force ring output. The standard stud is normally made of beryllium copper and shunts
approximately 5% of the force. Steel bolts can take away approximately
20–50% of the applied force. Different bolt materials may be used, but the
actuator requires recalibration with the new bolt. A properly mounted and
preloaded force ring is depicted in Fig. 1.44. Proper alignment and orientation of the actuator is also critical to long-term performance and calibration values. The general guidelines are to mount the actuator between
flat, parallel, and rigid supports that are at least twice the thickness of the
piezoelectric actuator. This aligns the actuator and contact surfaces to prevent edge loading or bending moments, resulting in better dynamic measurements. Loading the entire force-sensing surface is also important for
good measurements. This can be difficult if the surface being brought into
contact with the actuator is not parallel to the actuator surface.
F
Preload stud
Antifriction
washer
Force ring
sensor
Pilot bushing
(for centering)
Figure 1.44 A properly mounted force ring (courtesy of PCB PIEZOTRONICS,
Inc.).
Zhang_Ch01.indd 140
5/13/2008 5:45:43 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
141
The unique quasistatic nature of piezoelectric actuators allows static calibrations to be performed. As long as the calibration engineer applies the
following three rules, the results will closely match the factory calibration:
(1) The actuator must have a discharge time constant of at least 50 s,
or be a charge mode actuator.
(2) The signal conditioning must be DC coupled.
(3) For force rings, the factory-supplied beryllium copper preload
stud must be used.
The calibration engineer simply needs to place a known weight on the
actuator and wait for the signal to decay to zero (or reset the charge amplifier). The next step is to remove the weight from the actuator and record the
voltage output. This produces a negative voltage output. This value, divided
by the applied weight, equals the actuator sensitivity in volts per pound.
1.2.5
Manual Actuators
A manual actuator employs levers, gears, or wheels to facilitate movement. There are some manual gear actuators that are recommended for
use on ball and butterfly valves. For the sake of convenience, and where
lever handles present space problems, gear actuators are the ideal choice.
Fully enclosed weatherproof, all cast iron and carbon steel construction
manual gear actuators are factory lubricated for their lifetime, requiring no
future lubrication. Each unit is supplied with a pointer to indicate valve
position.
A manual actuator, by definition, is an actuator that requires no outside
power source. Handwheels, chainwheels, and levers are examples of manual actuators. A handwheel or lever is utilized to drive a series of gears
(typically worm gears) whose ratio results in a higher output torque compared to input (manual) torque. The manually operated actuators are of
screw mechanism, and are of large diameter wheels or long levers for use
in high-pressure applications equipped with a reduction gear to ease the
operation. Manual actuators can also be fitted with a chain wheel and
extended stem that are used for tank outlet valves or process valves where
lack of space demands extended handles.
An automatic actuator has an external power source to provide the force
and motion to operate a valve remotely or automatically. Power actuators
are a necessity on valves in pipelines located in remote areas: they are also
used on valves that are frequently operated or throttled. Valves that are
particularly large may be impossible or impractical to operate manually
Zhang_Ch01.indd 141
5/13/2008 5:45:44 PM
142
INDUSTRIAL CONTROL TECHNOLOGY
simply by the sheer horsepower requirements. Some valves may be located
in extremely hostile or toxic environments, which preclude manual operation. Additionally, as a safety feature, certain types of power actuators may
be required to operate quickly, shutting down a valve in case of emergency.
1.3 Valves
The valve is one of the most basic and indispensable component of our
modern industries. It is essential to industrial control technology that is
included in virtually all manufacturing processes and every energy production and supply system.
The valve is one of the oldest products known to man, with a history of
thousands of years. The modern history of the valve industry parallels the
Industrial Revolution, which began in 1705 when Thomas Newcomen
invented the first industrial steam engine. Because steam built up pressures
that had to be contained and regulated, valves acquired a new importance.
As Newcomen’s steam engine was improved upon by James Watt and
other inventors, designers and manufacturers also improved the valves for
these steam engines. Their interest, however, was in the whole project, and
the manufacture of valves as a separate product was not undertaken on a
large scale for a number of years.
A valve is a device that controls not only the flow of a fluid, but also the
rate, the volume, the pressure or the direction of liquids, gases, slurries, or
dry materials through a pipeline, chute, or similar passageway. With valves,
the flow of a fluid in various passageways can be turned ON and OFF,
regulated, modulated, or isolated for the range in size from a fraction of an
inch to as large as 30 ft in diameter, which can vary in complexity from a
simple brass valve available at the local hardware store to a precisiondesigned, highly sophisticated coolant system control valve, made of an
exotic metal alloy in a nuclear reactor. Valves can control flow of all types
of fluid, from the thinnest gas to highly corrosive chemicals, superheated
steam, abrasive slurries, toxic gases, and radioactive materials. They can
handle temperatures from cryogenic region to molten metal and pressures
from high vacuum to thousands of pounds per square inch.
1.3.1
Control Valves
Final control element stands for the device that implements the control
strategy determined by the output of the controller. While the final control
element can be a damper or a variable speed drive pump or an ON–OFF
Zhang_Ch01.indd 142
5/13/2008 5:45:44 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
143
switching device, the most common final control element in the process
control industries is the control valve. The control valve is this kind of
final element that manipulates a flowing fluid, such as gas, steam, water,
or chemical compounds, to compensate for the load disturbance and keep
the regulated process variable as close as possible to the desired set point.
Control valves or valves really refer to a control valve assembly. The
control valve assembly typically consists of the valve body, the internal
trim parts, an actuator to provide the motive power to operate the valve,
and a variety of additional valve accessories that can include positioners,
transducers, supply pressure regulators, manual operators, snubbers, or
limit switches.
The control valve regulates the rate of fluid flow as the position of the
valve plug or disk is changed by force from the actuator. To do this, the
valve must contain the fluid without external leakage; have adequate
capacity for the intended service; be capable of withstanding the erosive,
corrosive, and temperature influences of the process; and incorporate
appropriate end connections to mate with adjacent pipelines and actuator
attachment means to permit transmission of actuator thrust to the valve
plug stem or rotary shaft.
1.3.1.1
Basic Types
Many styles of control valve bodies have been developed through the
years. Some have found wide application; others meet specific service
conditions and are used less frequently. The following summary describes
some popular control valve body styles in use today, some special application valves, stem conditioning valves, and the ancillary devices to the control valve including valve actuators, positioners, and accessories.
(1) Linear globe valves. Linear globe valves are those valves with a
linear motion closure member, one or more ports, and a body
distinguished by a globular-shaped cavity around the port region.
Globe valves can be further classified as single-ported valve bodies, balance-plug cage-guided bodies, high capacity cage-guided
valve bodies, port-guided single-port valve bodies, double-ported
valve bodies, and three-way valve bodies.
(2) Rotary shaft valves. Rotary shaft valves are those valves of the
style in which the flow closure member (full ball, partial ball, disk,
or plug) is rotated in the flow stream to control the capacity of the
valve. Rotary shaft valves can be further classified as butterfly
valve bodies, V-notch ball control valve bodies, eccentric-disk
control valve bodies, and eccentric-plug control valve bodies.
Zhang_Ch01.indd 143
5/13/2008 5:45:44 PM
144
INDUSTRIAL CONTROL TECHNOLOGY
(3) Special valves. Standard control valves can handle a wide range
of control applications. Certainly, corrosiveness and viscosity of
the fluid, leakage rates, and many other factors demand consideration even for standard applications. The following discusses
some special control valve modifications useful in severe controlling applications.
(a) High capacity control valves. The following covers the special valve category of globe style valves larger than 12-in.,
ball valves over 24-in., and high-performance butterfly
valves larger than 48-in. As valve sizes increase arithmetically, static pressure loads at shutoff increase geometrically.
Consequently, shaft strength, bearing loads, unbalance forces,
and available actuator thrust all become more significant
with increasing valve size. Normally, maximum allowable
pressure drop is reduced on large valves to keep design and
actuator requirements within reasonable limits. Even with
lowered working pressure ratings, the flow capacity of some
large-flow valves remains tremendous.
(b) Low flow control valves. Many applications exist in laboratories and pilot plants in addition to the general processing
industries where control of extremely low flow rates is
required. These applications are commonly handled in one
of two ways. First, special trims are often available in standard control valve bodies. The special trim is typically made
up of a seat ring and valve plug that have been designed and
machined to very close tolerances to allow accurate control
of very small flows. These types of control valves are specially designed for the accurate control of very low-flowing
liquid or gaseous fluid applications.
(c) High temperature control valves. These designed control
valves are special for service at temperatures above 450oF
(232oC). They are frequently at elevated temperatures, such
as may be encountered in boiler feed water systems and
superheater bypass systems.
(d) Cryogenic service valves. Cryogenic service valves are for
dealing with materials and processes at temperatures below
–150oF (–101oC). For control valve applications in cryogenic services, many of the same issues need consideration
as with high temperature control valves. Packing is a concern in cryogenic applications. Plastic and electrometric
components often cease to function appropriately at temperatures below 0oF (–18oC). In these temperature ranges, components such as packing and plug seals require special
consideration. For plug seals, a standard soft seal will become
Zhang_Ch01.indd 144
5/13/2008 5:45:44 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
145
very hard and less pliable, thus not providing the shut-off
required from a soft seat. Special electrometrics have been
applied in these temperatures but require special loading to
achieve a tight seal.
(4) Stem conditioning valves. A stem conditioning valve is used for
the simultaneous reduction of steam pressure and temperature to
the level required for a given application. Frequently, these applications deal with high inlet pressures and temperatures and
require significant reductions of both properties. They are, therefore, best manufactured in a forged and fabricated body that can
better withstand steam loads at elevated pressures and temperatures. Forged materials permit higher design stresses, improved
grain structure, and an inherent material integrity over cast valve
bodies. The forged construction also allows the manufacturer to
provide up to Class 4500, as well as intermediate and special
class ratings, with greater ease vs cast valve bodies.
Due to frequent extreme changes in steam properties as a
result of the temperature and pressure reduction, the forged
and fabricated valve body design allows for the addition of an
expanded outlet to control outlet steam velocity at lower pressure. Similarly, with reduced outlet pressure, the forged and fabricated design allows the manufacturer to provide different
pressure class ratings for the inlet and outlet connections to more
closely match the adjacent piping.
The latest versions of the stem conditioning valves have these
designs: feed-forward design, manifold design, pressure reduction–
only design, and turbine bypass design, etc.
The turbine bypass system has evolved over the past few
decades as the mode of power plant operations has changed. It is
employed routinely in utility power plants where operations
require quick response to wide swings in energy demands. A
typical power plant operation might start at minimum load,
increase to full capacity for most of the day, rapidly reduce back
to minimum output, and then up again to full load—all within a
24-h period. Boilers, turbines, condensers, and other associated
equipment cannot respond properly to such rapid changes without some form of turbine bypass system. The turbine bypass system allows operation of the boiler independent of the turbine. In
the start-up mode, or rapid reduction of generation requirement,
the turbine bypass not only supplies an alternate flow path for
steam, but conditions the steam to the same pressure and temperature normally produced by the turbine expansion process.
By providing an alternate flow path for the steam, the turbine
bypass system protects the turbine, boiler, and condenser from
Zhang_Ch01.indd 145
5/13/2008 5:45:44 PM
146
INDUSTRIAL CONTROL TECHNOLOGY
damage that may occur from thermal and pressure excursions.
For this reason, many turbine bypass systems require extremely
rapid open/close response times for maximum equipment protection. This is accomplished with an electrohydraulic actuation
system that provides both the forces and controls for such operation. Additionally, when commissioning a new plant, the turbine
bypass system allows start-up and check out of the boiler separately from the turbine. This means quicker plant start-ups, which
results in attractive economic gains. It also means that this closed
loop system can prevent atmospheric loss of treated feed water
and reduction of ambient noise emissions.
(5) Valve actuators. Pneumatically operated control valve actuators
are the most popular type in use, but electric, hydraulic, and
manual actuators are also widely used. All these actuators are
introduced in Section 1.2 of this chapter.
The spring-and-diaphragm pneumatic actuator is most commonly specified due to its dependability and simplicity of design.
Pneumatically operated piston actuators provide high stem force
output for demanding service conditions. Adaptations of both
spring-and-diaphragm and pneumatic piston actuators are available for direct installation on rotary shaft control valves. Electric
and electrohydraulic actuators are more complex and more expensive than pneumatic actuators, which offer advantages where no
air supply source is available, where low ambient temperatures
could freeze condensed water in pneumatic supply lines, or where
unusually large stem forces are needed.
Pneumatically operated diaphragm actuators use air supply
from controller, positioner, or other sources. Various styles include
direct action of increasing air pressure pushes down the diaphragm and extends actuator stem; reverse action of increasing
air pressure pushes up the diaphragm and retracts actuator stem;
reversible actuators that can be assembled for either direct or
reverse action; direct acting unit for rotary valves of increasing
air pressure pushes down on the diaphragm, which may either
open or close the valve, depending on orientation of the actuator
lever on the valve shaft.
Piston actuators are pneumatically operated using high-pressure
plant air to 150 psig, often eliminating the need for a supply
pressure regulator. Piston actuators furnish maximum thrust output and fast stroking speeds. Piston actuators are double acting to
give maximum force in either direction, or spring return to provide fail-open or fail-closed operation.
Electrohydraulic actuators require only electrical power to
the motor and an electrical input signal from the controller.
Zhang_Ch01.indd 146
5/13/2008 5:45:44 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
147
Electrohydraulic actuators are ideal for isolated locations where
pneumatic supply pressure is not available but where precise control of valve plug position is needed. Units are normally reversible by making minor adjustments and might be self-contained,
including motor, pump, and double-acting hydraulically operated piston within a weatherproof or explosion-proof casing.
Rack and pinion designs provide a compact and economical
solution for rotary shaft valves. Because of backlash, they are
typically used for ON–OFF applications or where process variability is not a concern.
Traditional electric actuator designs use an electric motor and
some form of gear reduction to move the valve. Through adaptation, these mechanisms have been used for continuous control with
varying degrees of success. To date, electric actuators have been
much more expensive than pneumatic for the same performance
levels. This is an area of rapid technological change, and future
designs may cause a shift toward greater use of electric actuators.
Manual actuators are useful where automatic control is not
required, but where ease of operation and good manual control is
still necessary. They are often used to actuate the bypass valve in
a three-valve bypass loop around control valves for manual control of the process during maintenance or shutdown of the automatic system. Manual actuators are available in various sizes for
both globe style valves and rotary shaft valves. Manual actuators
are much less expensive than automatic actuators.
(6) Positioners. Positioners are used for pneumatically operated
valves that depend on a positioner to take an input signal from a
process controller and convert it to a valve travel. Positioners are
mostly available in three configurations:
(a) Pneumatic. A pneumatic signal (usually 3–15 psig) is supplied to the positioner. The positioner translates this to a
required valve position and supplies the valve actuator with
the required air pressure to move the valve to the correct
position.
(b) Analog I/P. This positioner performs the same function as
the one above, but uses electrical current (usually 4–20 mA)
instead of air as the input signal.
(c) Digital. Although this positioner functions very much like
the analog I/P described above, it differs in that the electronic signal conversion is digital rather than analog. The
digital products cover three categories:
(i) Digital noncommunicating. A current signal (4–20 mA)
is supplied to the positioner, which both powers the
electronics and controls the output.
Zhang_Ch01.indd 147
5/13/2008 5:45:44 PM
148
INDUSTRIAL CONTROL TECHNOLOGY
(ii) HART. This is the same as the digital noncommunicating
but is also capable of two-way digital communication
over the same wires used for the analog signal.
(iii) Fieldbus. This type receives digitally based signals and
positions the valve using digital electronic circuitry
coupled to mechanical components.
(7) Valve accessories. Valve accessories include
(a) Limit switches. Limit switches operate discrete inputs to a
distributed control system, signal lights, small solenoid valves,
electric relays, or alarms. An assembly that mounts on the
side of the actuator houses the switches. Each switch adjusts
individually and can be supplied for either alternating current or direct current systems. Other styles of valve-mounted
limit switches are also available.
(b) Solenoid valve manifold. The actuator type and the desired
fail-safe operation determine the selection of the proper solenoid valve. The solenoids can be used on double-acting pistons or single-acting diaphragm actuators.
(c) Supply pressure regulator. Supply pressure regulators, commonly called airsets, reduce plant air supply to valve positioners and other control equipment. Common reduced air
supply pressures are 20, 35, and 60 psig. The regulator
mounts integrally to the positioner or nipple-mounts or bolts
to the actuator.
(d) Pneumatic lock-up systems. Pneumatic lock-up systems are
used with control valves to lock in existing actuator loading
pressure in the event of supply pressure failure. These devices
can be used with volume tanks to move the valve to the
fully open or closed position on loss of pneumatic air supply. Normal operation resumes automatically with restored
supply pressure. Functionally similar arrangements are
available for control valves using diaphragm actuators.
(e) Fail-safe systems for piston actuators. In these fail-safe systems, the actuator piston moves to the top or bottom of the
cylinder when supply pressure falls below a predetermined
value. The volume tank, charged with supply pressure,
provides loading pressure for the actuator piston when
supply pressure fails, thus moving the piston to the desired
position. Automatic operation resumes, and the volume tank
is recharged when supply pressure is restored to normal.
(f) PC diagnostic software. PC diagnostic software provides a
consistent, easy to use interface to every field instrument
within a plant. For the first time, a single resource can be
used to communicate and analyze field electronic “smart”
Zhang_Ch01.indd 148
5/13/2008 5:45:44 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
149
devices such as pressure xmtrs, flow xmtrs, etc., not pneumatic positioners, boosters.
Users can benefit from reduced training requirements and
reduced software expense. A single purchase provides the
configuration environment for all products. Products and
services are available that were not possible with stand-alone
applications. The integrated product suite makes higher level
applications and services possible.
(g) Electropneumatic transducers. The transducer receives a
direct current input signal and uses a torque motor, nozzle
flapper, and pneumatic relay to convert the electric signal to
a proportional pneumatic output signal. Nozzle pressure
operates the relay and is piped to the torque motor feedback
bellows to provide a comparison between input signal and
nozzle pressure.
1.3.1.2 Technical Specifications
Control valves handle all kinds of fluids at temperatures from the cryogenic range to well over 1000°F (538oC). Selection of a control valve body
assembly requires particular consideration to provide the best available
combination of valve body style, material, and trim construction design
for the intended service. Capacity requirements and system operating pressure ranges must also be considered in selecting a control valve to ensure
satisfactory operation without undue initial expense.
Reputable control valve manufacturers and their representatives are
dedicated to helping in selecting the control valve most appropriate for the
existing service conditions. Because there are often several possible correct choices for an application, it is important that all the following information be provided:
(1) type of fluid to be controlled;
(2) temperature of fluid;
(3) viscosity of fluid;
(4) specific gravity of fluid;
(5) flow capacity required (maximum and minimum);
(6) inlet pressure at valve (maximum and minimum);
(7) outlet pressure (maximum and minimum);
(8) pressure drop during normal flowing conditions;
(9) pressure drop at shutoff;
(10) maximum permissible noise level, if pertinent and the measurement reference point;
Zhang_Ch01.indd 149
5/13/2008 5:45:44 PM
150
INDUSTRIAL CONTROL TECHNOLOGY
(11) degrees of superheat or existence of flashing, if known;
(12) inlet and outlet pipeline size and schedule;
(13) special tagging information required;
(14) body material;
(15) end connections and valve rating;
(16) action desired on air failure;
(17) instrument air supply available;
(18) instrument signal.
The following information will require the agreement of the user and
the manufacturer depending on the purchasing and engineering practices
being followed:
(1) valve type number;
(2) valve size;
(3) valve body construction (angle, double port, butterfly, etc.);
(4) valve plug guiding (cage style, port guided, etc.);
(5) valve plug action (push down to close or push down to open);
(6) port size (full or restricted);
(7) valve trim materials required;
(8) flow action (flow tends to open valve or flow tends to close
valve);
(9) actuator size required;
(10) bonnet style (plain, extension, etc.);
(11) packing material (laminated graphite, environmental sealing
systems, etc.);
(12) accessories required (positioner, handwheel, etc.).
The following steps need to be undertaken for the selection of a valve:
(1) determine service condition;
(2) calculate preliminary cv required;
(3) select trim type;
(4) select valve body and trim size;
(5) select trim materials;
(6) other consideration such as shutoff, stem packing, etc.
1.3.1.3 Application Guide
The performance of the control valves can be affected by the following
factors:
(1) Dead band. Dead band is a range or band of controller output
values that fails to produce a change in the measured process
Zhang_Ch01.indd 150
5/13/2008 5:45:44 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
151
variable when the input signal reverses direction. When a load
disturbance occurs, the process variable deviates from the set
point; the deviation initiates a corrective action through the controller and back through the process. However, an initial change
in controller output can produce no corresponding corrective
change in the process variable. Only when the controller output
has been changed enough to progress through the dead band does
a corresponding change in the process variable occur.
Any time the controller output reverses direction, the controller
signal must pass through the dead band before any corrective
change in the process variable will occur. Therefore, dead band is
a major contributor to excess process variability, and control valve
assemblies can be a primary source of dead band in an instrumentation loop due to a variety of causes such as friction, backlash,
shaft windup, relay or spool valve dead zone, etc. Its presence in
the process ensures that the process variable deviation from the
set point will have to increase until it is big enough to get through
the dead band. Only then can a corrective action occur.
Some of most common causes affecting dead band are friction
and backlash in the control valve, along with shaft wind-up in
rotary valves, and relay dead zone. Because most control actions
for regulatory control consist of small changes (1% or less), a
control valve with excessive dead band might not even respond
to many of these small changes. 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 dead band as great as 5% or more. In a recent plant
audit, 30% of the valves had dead bands in excess of 4%. Over
65% of the loops audited had dead bands greater than 2%.
Friction is a major cause of dead band in control valves. Rotary
valves are often very susceptible to friction caused by the high
seat loads required 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 translate motion to the
control element. As a result, an improperly designed rotary valve
can exhibit significant dead band that clearly has a detrimental
effect on process variability. Manufacturers usually lubricate
rotary valve seals during manufacture, but after only a few hundred cycles this lubrication wears off. In addition, pressureinduced loads also cause seal wear. As a result, the valve friction
can increase by 400% or more for some valve designs. This illustrates the misleading performance conclusions that can result
from evaluating products using bench-type data before the torque
has stabilized. Packing friction is the primary source of friction
Zhang_Ch01.indd 151
5/13/2008 5:45:45 PM
152
INDUSTRIAL CONTROL TECHNOLOGY
in sliding stem valves. In these types of valves, the measured
friction can vary significantly between valve styles and packing
arrangements. Actuator style also has a profound impact on control valve assembly friction. Generally, spring-and-diaphragm
actuators contribute less friction to the control valve assembly
than piston actuators. Piston actuator friction probably will
increase significantly with use as guide surfaces and the O-rings
wear, lubrication fails, and the elastomer degrades. Backlash is
the name given to slack or looseness of a mechanical connection.
This slack results in a discontinuity of motion when the device
changes direction. Backlash commonly occurs in gear drives of
various configurations. Rack-and-pinion actuators are particularly prone to dead band due to backlash. Some valve shaft connections also exhibit dead band effects. Spline connections
generally have much less dead band than keyed shafts or double-D
designs.
While friction can be reduced significantly through good valve
design, it is a difficult phenomenon to eliminate entirely. A wellengineered control valve should be able to virtually eliminate
dead band due to backlash and shaft wind-up. For best performance in reducing process variability, the total dead band for the
entire valve assembly should be 1% or less. Ideally, it should be
as low as 0.25%.
(2) Actuator/positioner design. Both the actuator and positioner
designs greatly affect static performance (dead band), as well as
the dynamic response of the control valve assembly and the overall air consumption of the valve instrumentation.
Static gain is related to the sensitivity of the device to the
detection of small (0.125% or less) changes of the input signal.
Unless the device is sensitive to these small signal changes, it
cannot respond to minor upsets in the process variable. This high
static gain of the positioner is obtained through a preamplifier,
similar in function to the preamplifier contained in high fidelity
sound systems. In many pneumatic positioners, a nozzle flapper
or a similar device serves as this high static gain preamplifier.
Once the high static gain positioner preamplifier has detected a
change in the process variable, the positioner must then be capable of making the valve closure member move rapidly to provide
a timely corrective action to the process variable. This requires
much power to make the actuator and valve assembly move
quickly to a new position, which means that the positioner
must rapidly supply a large volume of air to the actuator to
make it respond promptly. The ability to do this comes from the
high dynamic gain of the positioner. Although the positioner
Zhang_Ch01.indd 152
5/13/2008 5:45:45 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
153
preamplifier can have high static gain, it typically has little ability
to supply the power needed. Thus, the preamplifier function must
be supplemented by a high dynamic gain power amplifier that
supplies the required air flow as rapidly as needed. A relay or a
spool valve typically provides this power amplifier function.
In summary, high-performance positioners with both high
static and dynamic gain provide the best overall process variability performance for any given valve assembly.
(3) Valve response time. It is important that the valve reach a specific
position quickly in control. A quick response to small signal
changes (1% or less) is one of the most important factors in providing optimum process control. Valve response time is measured
by a parameter called T63 (Tee-63). T63 is the time measured from
initiation of the input signal change to when the output reaches
63% of the corresponding change. It includes both the valve assembly dead time, which is a static time, and the dynamic time of the
valve assembly. The dynamic time is a measure of how long the
actuator takes to get to the 63% point once it starts moving.
Dead band, whether it comes from friction in the valve body
and actuator or from the positioner, can significantly affect the
dead time of the valve assembly. It is important to keep the dead
time as small as possible. Generally dead time should be no more
than one-third of the overall valve response time. However, the
relative relationship between the dead time and the process time
constant is critical.
If the valve assembly is in a fast loop where the process time
constant approaches the dead time, the dead time can dramatically affect loop performance. On these fast loops, it is critical to
select control equipment with dead time as small as possible
because some valve assembly designs can have dead times that
are 3–5 times longer in one stroking direction than the other.
Once the dead time has passed and the valve begins to respond,
the remainder of the valve response time comes from the dynamic
time of the valve assembly. This dynamic time will be determined primarily by the dynamic characteristics of the positioner
and actuator combination. These two components must be carefully matched to minimize the total valve response time. This
dynamic gain comes mainly from the power amplifier stage in
the positioner. However, this high dynamic gain power amplifier
will have little effect on the dead time unless it has some intentional dead band designed into it to reduce static air consumption. The design of the actuator significantly affects the dynamic
time. For example, the greater the volume of the actuator air
chamber to be filled, the slower the valve response time.
Zhang_Ch01.indd 153
5/13/2008 5:45:45 PM
154
INDUSTRIAL CONTROL TECHNOLOGY
To minimize the valve assembly dead time, minimize the dead
band of the valve assembly, whether it comes from friction in the
valve seal design, packing friction, shaft wind-up, actuator, or
positioner design. As indicated, friction is a major cause of dead
band in control valves. On rotary valve styles, shaft wind-up can
also contribute significantly to dead band. Actuator style also has
a profound impact on control valve assembly friction. On the
impact from the actuators, generally say, spring-and-diaphragm
actuators contribute less friction to the control valve assembly
than piston actuators over an extended time. As mentioned, this
is caused by increasing friction from the piston O-ring, misalignment problems, and failed lubrication. Having a positioner design
with a high static gain preamplifier can make a significant difference in reducing dead band. This can also make a significant
improvement in the valve assembly resolution. Valve assemblies
with dead band and resolution of 1% or less are no longer adequate for many process variability reduction needs. Many processes require the valve assembly to have dead band and
resolution as low as 0.25%, especially where the valve assembly
is installed in a fast process loop.
Selecting the proper valve, actuator, and positioner combination is not easy. It is not simply a matter of finding a combination
that is physically compatible. Good engineering judgment must
go into the practice of valve assembly sizing and selection to
achieve the best dynamic performance from the loop.
(4) Valve types and sizing. The style of valve used and the sizing of
the valve can have a large impact on the performance of the control valve assembly in the system. Although a valve must be of
sufficient size to pass the required flow under all possible contingencies, a valve that is too large for the application is a detriment
to process optimization.
Flow capacity of the valve is also related to the style of valve
through the inherent characteristic of the valve. The inherent
characteristic is the relationship between the valve flow capacity
and the valve travel when the differential pressure drop across
the valve is held constant. The best process performance occurs
when the required flow characteristic is obtained through changes
in the valve trim rather than through use of cams or other methods. Proper selection of a control valve designed to produce a
reasonably linear installed flow characteristic over the operating
range of the system is a critical step in ensuring optimum process
performance.
Oversizing of valves sometimes occurs when trying to optimize
process performance through a reduction of process variability.
Zhang_Ch01.indd 154
5/13/2008 5:45:45 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
155
This results from using line-size valves, especially with highcapacity rotary valves, as well as the conservative addition of
multiple safety factors at different stages in the process design.
Oversizing the valve hurts process variability in two ways.
First, the oversized valve puts too much gain in the valve, leaving
less flexibility in adjusting the controller. Best performance results
when most loop gain comes from the controller. The second way
oversized valves hurt process variability is that an oversized valve
is likely to operate more frequently at lower valve openings where
seal friction can be greater, particularly in rotary valves.
Because an oversized valve produces a disproportionately
large flow change for a given increment of valve travel, this phenomenon can greatly exaggerate the process variability associated with dead band due to friction. When the valve is oversized,
the valve tends to reach system capacity at relatively low travel,
making the flow curve flatten out at higher valve travels.
When selecting a valve, it is important to consider the valve
style, inherent characteristic, and valve size that will provide the
broadest possible control range for the application.
1.3.2
Self-Actuated Valves
Self-actuated valve stands for those valves that use fluid or gas existing
in a system to position the valve. Check valves and relief valves are two
important examples of self-actuated valves. In additional to check valves
and relief valves, safety valves and steam traps are also defined as selfactuated valves. All of these valves are being actuated with the system
fluid or gas; no source of power outside the system fluid or gas energy is
necessary for operation of these valves.
1.3.2.1
Check Valves
Check valves are self-activating safety valves that permit gases and liquids
to flow in only one direction, preventing process flow from reversing. When
open and under flow pressure, the checking mechanism will move freely in
the medium, offering very little resistance and minimal pressure drop. Check
valves are classified as one-way directional valves: fluid flow in the desired
direction opens the valve, while backflow forces the valve to close.
(1) Operating principle. A check valve is a one-way valve for fluid
flow. There are many ways to achieve one-way flow. Most check
valves contain a ball that sits freely above the seat, which has
Zhang_Ch01.indd 155
5/13/2008 5:45:45 PM
156
INDUSTRIAL CONTROL TECHNOLOGY
only one through-hole. The ball has a slightly larger diameter
than that of the through-hole. When the pressure behind the seat
exceeds that above the ball, liquid is allowed to flow through the
valve; however, once the pressure above the ball exceeds the
pressure below the seat, the ball returns to rest in the seat, forming a seal that prevents backflow.
Figure 1.45 is used here to describe briefly the principles of
check valves. Device A consists of a ball bearing retained by a
spring. Fluid flowing to the right will push the ball bearing
against the spring, and open the valve to permit flow. This device
requires some pressure to compress the spring and open the
valve. If an attempt is made to flow fluid to the left, the ball bearing seals against the opening and no flow is allowed. This is a
modern design that requires round balls. Device B is simply a
flapper that is anchored on one side. The flapper can be a hinged
metallic door, a thin piece of metal, or a piece of rubber or polymer. This is the simplest design and was used in early pumps.
Two methods of incorporating check valves into pumps are
also shown in Fig. 1.46. In schematic A, the check valves permit
expulsion of the fluid to the right on the downward stroke while
denying flow to the left. On the upward stroke, the pump fills
from the left, while denying reverse flow from the right. The
design in schematic B is somewhat different. The piston has one
or more holes drilled though with a check valve on each hole.
(One hole is illustrated.) On the downward stroke, the fluid
moves from below the piston to the chamber above the piston
and is denied exit to the left. On the upward stroke, fluid is pushed
Flow
(a)
Flow
(b)
Figure 1.45 Device A consists of a ball bearing retained by a spring; device B is
simply a flapper anchored on one side (courtesy of Michigan State University).
Zhang_Ch01.indd 156
5/13/2008 5:45:45 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
157
Flow
(a)—Downward stroke forces fluid out in this device
Flow
Hole drilled
in piston
(b)—Upward stroke forces fluid out in this device
Figure 1.46 Two methods of incorporating check valves into pumps (courtesy of
Michigan State University).
out the exit on the right and simultaneously more fluid is drawn
from the entrance on the left. Case B is the design illustrated by
Watt in his patent and described as the air pump since it pumps
air as well as water.
(2) Basic types. Check valves use a variety of technologies to allow
and stop the flow of liquids and gases. Single-disk swing valves
are designed with the closure element attached to the top of the
cap. Double-disk or wafer check valves consist of two half-circle
disks hinged together that fold together upon positive flow and
retract to a full circle to close against reverse flow. Lift-check
valves feature a guided disk. Spring-loaded devices can be
mounted vertically or horizontally. Silent or center guide valves
are similar to lift check valves, with a center guide extending
from inlet to outlet ports. The valve stopper is spring and bushing
actuated to keep the movement “quiet.” Ball check valves use a
free-floating or spring-loaded ball resting in a seat ring as the
closure element. Cone check valves use a free-floating or springloaded cone resting in the seat ring as the closure element.
Although there are many types of check valves, two basic types
are most popular in industrial control: swing check valves and
ball check valves. Both types of valves may be installed vertically or horizontally.
(a) Swing check valves. Swing check valves are used to prevent
flow reversal in horizontal or vertical upward pipelines (vertical pipes or pipes in any angle from horizontal to vertical
Zhang_Ch01.indd 157
5/13/2008 5:45:45 PM
158
INDUSTRIAL CONTROL TECHNOLOGY
with upward flow only). Swing check valves have disks that
swing open and closed. The disks are typically designed to
close on their own weight and may be in a state of constant
movement if velocity pressure is not sufficient to hold the
valve in a wide open position. Premature wear or noisy operation of the swing check valves can be avoided by selecting
the correct size on the basis of flow conditions.
The minimum velocity required to hold a swing check
valve in the open position is expressed by the empirical formula given by Fig. 1.47, where V is the liquid flow measured
in m/s or ft/s; v is the special volume of the liquid measured
in m3/N or ft3/lb; j equals 133.7 (35) for Y-pattern, or = 229.1
(60) for bolted cap, or 381.9 (100) for U/L listed.
Tilting disk check valves are pivoted circular disks
mounted in a cylindrical housing. These check valves have
the ability to close rapidly, thereby minimizing slamming
and vibrations. Tilting disk checks are used to prevent reversal in horizontal or vertical-up lines similar to swing check
valves. The minimum velocity required for holding a tilting
check valve wide open can be determined by the empirical
formula given in Fig. 1.47, where V is the liquid flow measured in m/s or ft/s; v is the special volume of the liquid
measured in m3/N or ft3/lb; j equals 305.5 (80) for 5 times by
disk angle (typically for steel), or = 114.6 (30) for 15 times
by disk angle (typical for iron).
Lift check valves also operate automatically by line pressure. They are installed with pressure under the disk. A lift
check valve typically has a disk that is free floating and is
lifted by the flow. Liquid has an indirect line of flow, so the
lift check restricts the flow. Because of this, lift check valves
are similar to globe valves and are generally used as a companion to globe valves.
(b) Ball check valves. A ball check valve is a type of check valve
in which the movable part is a spherical ball as illustrated in
Fig. 1.48. Ball check valves are used in spray devices, dispenser spigots, manual and other pumps, and refillable dispensing syringes. Ball check valves may sometimes use a
free-floating or spring-loaded ball.
Ball check valves are generally simple, inexpensive metallic parts, although specialized ball check valves are also
Vi j v
Figure 1.47 The minimum velocity formula of swing check valves.
Zhang_Ch01.indd 158
5/13/2008 5:45:45 PM
159
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
7
3
5
2
6 4
1
8
4
E
EG
Figure 1.48 The working block of a ball check valve. In this figure, 1 is the seat
body, 2 is the cap, 3 is the ball, 4 is the angle body, 5 is the body clamp, 6 is the
body gasket, and 8 is the cap gasket (courtesy of VNE Corporation).
available. For example, ball check valves in high-pressure
pumps used in analytical chemistry have a ball of synthetic
ruby, a hard and chemically resistant substance. A ball check
valve is not to be confused with a ball valve, a quarter-turn
valve similar to a butterfly valve in which the ball acts as a
controllable rotor.
(3) Specifications. The following contents are the basic specifications of check valves:
(a) Technical types of check valves include
(i) Ball and cone check valves. Ball and cone check valves
use a free-floating or spring-loaded ball resting in a
seat ring as the closure element. Upon reverse flow,
the ball is forced back into its seat preventing
backflow.
(ii) Double check valves. Double check valves are assemblies that contain two distinct check valves.
(iii) Duckbill check valves. Duckbill valves are flowsensitive, variable-area, check valves. They get their
name from their shape, which consists of two flaps
shaped like a duck’s bill. In zero flow conditions, the
valve remains closed. As the flow increases, the pressure on the flaps increases and the valve opens.
(iv) Foot check valves. Foot valves are a type of check
valve with a built-in strainer. They are used at the point
of liquid intake to retain liquid in the system.
Zhang_Ch01.indd 159
5/13/2008 5:45:46 PM
160
INDUSTRIAL CONTROL TECHNOLOGY
(v) Lift check valves. Lift check valves use a free-floating
closure element, consisting of a piston or poppet and a
seat ring. The piston or poppet actuates either horizontal or vertical to the flow, depending on the valve
construction.
(vi) Swing check valves. Swing check valves are designed
with the closure element attached to the top of the cap.
The closure element can be pushed aside by the flow,
but swings back into the close position upon flow
reversal.
(vii) Umbrella check valves. Umbrella check valves are
elastomer self-activating devices. These valves simply
press into a hole and can be designed to function within
a specified pressure range. The valve gets its name
from the umbrella-like shape of the device.
(viii) Wafer/split disc check valves. Wafer or split disk check
valves have two half-circle disks hinged together that
fold together upon positive flow and retract to a full
circle to close against reverse flow.
(b) Technical parameters of check valves include
(i) Valve size. It is the designated size of the valve as
specified by the manufacturer, which typically represents the size of the passage opening.
(ii) Pressure rating. It is the maximum safe pressure value
for which the valve is rated.
(iii) Media temperature. It is the maximum temperature of
media the valve is designed to accommodate.
(iv) Flow. The valve flow coefficient is the number of U.S.
gallons per minute of 60° F water that will flow through
a valve at a specified opening with a pressure drop of
1 psi across the valve. It is used to predict flow rates.
(c) Connection methods for the check valves can be
(i) Threaded. The valve has internal or external threads
for inlet or outlet connection(s).
(ii) Compression fitting. It is a sealed pipe connection
without soldering or threading. As the nut on one fitting is tightened, it compresses a washer around the
second pipe, forming a watertight closure.
(iii) Bolt flange. The valve has a bolt flange(s) for inlet or
outlet connection.
(iv) Clamp flange. The valve has a clamp flange(s) for inlet
or outlet connection.
(v) Union. The valve has a union connection for inlet or
outlet connection(s).
Zhang_Ch01.indd 160
5/13/2008 5:45:46 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
161
(vi) Tube fitting. The valve has a connection for directly
joining tubing at the inlet and or outlet connections.
(vii) Butt weld. The valve has a butt weld–sized connection
for inlet or outlet connection.
(viii) Socket weld/solder. The valve has a socket weld connection for inlet or outlet connection.
1.3.2.2
Relief Valves
The relief valve is a valve mechanism that ensures system fluid flow
when a preselected differential pressure across the filter element is
exceeded; the valve allows all or part of the flow to bypass the filter element. Relief valves are used on oil and gas production systems, compressor stations, gas transmission (pipeline) facilities, storage systems, in all
gas processing plants, and whenever there is a need to exhaust the overpressure volume of gas, vapor, and/or liquid.
(1) Operating principle. Figure 1.49 illustrates the working blocks
of a relief valve. Relief valves operate on the principle of unequal
areas exposed to the same pressure. When the relief valve is
closed, the system pressure pushes upward against the piston
seat seal on an area equal to the inside diameter of the seat.
Simultaneously, the same system pressure, passing through the
Normal
pressure
Figure 1.49 The working blocks of a relief valve (courtesy of P.C. McKenzie
Company).
Zhang_Ch01.indd 161
5/13/2008 5:45:46 PM
162
INDUSTRIAL CONTROL TECHNOLOGY
pilot, exerts a downward force on the piston acting on an area
approximately 50% greater than the inside diameter of the seat.
The resulting differential force holds the valve tightly closed. As
the system reaches the discharge set pressure of the valve, the
piston seal becomes tighter until the system pressure reaches the
relief valve discharge set pressure. At that moment, and not
before, the pilot cuts off the supply of system pressure to the top
of the piston and vents that system pressure which is located in
the chamber above the piston of the relief valve. At the same
instant, the relief valve pops open. When the predetermined blow
down pressure is reached (either fixed or adjustable), the pilot
shuts off the exhaust and reopens flow of system pressure to the
top of the piston, effectively closing the relief valve.
Figure 1.50 is a drawing of a direct operating pressure relief
valve. This pressure relief valve is mounted at the pressure side
of the hydraulic pump that is located on its bottom. The task of
this pressure relief valve is to limit the pressure in the system on
an acceptable value. In fact a pressure relief valve has the same
construction as a spring-operated check valve. When the system
gets overloaded the pressure relief valve will open and the pump
flow will be led directly into the hydraulic reservoir. The pressure in the system remains on the value determined by the spring
T
P
Figure 1.50 A drawing of a direct operating pressure relief valve.
Zhang_Ch01.indd 162
5/13/2008 5:45:47 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
163
on the pressure relief valve. In the pressure relief valve, the pressure that is equal to the system energy will be converted into
heat. For this reason pressure relief valves should not be operated
for long durations.
(2) Basic types and specifications. Table 1.5 lists several important
types of relief valve, categorized in terms of their applications. In
industrial control, typical features for relief valves include the
following: (1) Pressure settings of relief valves are externally
adjustable while the valve is in operation. Most vendors of relief
valves offer eight different spring ranges to provide greater system sensitivity and enhanced performance. (2) Manual override
option with positive stem retraction is available for pressures.
This option permits the user to relieve upstream pressure while
maintaining the predetermined cracking pressure. (3) Colorcoded springs and labels indicate spring cracking range. (4) Lock
wire feature secures a given pressure setting.
Typical specifications of a relief valve include
(a) Working pressure. It can be up to 6000 psig (414 bars) or up
to 8000 psig (552 bars) during relief with no internal seal
damage.
Table 1.5 Important Types of Relief Valves
Temperature and
pressure relief valves
Reseating temperature
and pressure relief
valves
Pressure relief valves
Poppet style relief
valves
Zhang_Ch01.indd 163
Temperature and pressure relief valves are used in
water heater and hot water storage tank applications to provide automatic temperature and pressure protection to hot water supply tanks and hot
water heaters
Automatic reseating temperature and pressure
relief valves are used in commercial water heater
applications to provide automatic temperature and
pressure protection to domestic hot water supply
tanks and hot water heaters
Pressure relief valves are used in hot water heating
and domestic supply boiler applications to protect
against excessive pressures on all types of hot
water heating supply boiler equipment.
Calibrated pressure relief valves are used in commercial, residential, and industrial applications to
protect against excessive pressure in systems containing water, oil, or air.
5/13/2008 5:45:47 PM
164
INDUSTRIAL CONTROL TECHNOLOGY
(b) Cracking pressure. For eight springs, it is normally from 50
to 6000 psig in the following ranges: 50–350, 350–750, 750–
1500, 1500–2250, 2250–3000, 3000–4000, 4000–5000, and
5000–6000 psig.
(c) Temperature rating. For Buna-N Rubber it can be –30°F to
+225°F (–34°C to +107°C); for highly fluorinated fluorocarbon rubber, it can be –20°F to +200°F (–29°C to +93°C); for
ethylene propylene rubber, it can be –70°F to +275°F (–57°C
to +135°C); for fluorocarbon rubber, it can be –10°F to +400°F
(–23°C to +204°C); for neoprene rubber, it can be –45°F to
+250°F (–43°C to +121°C).
The American Society of Mechanical Engineers (ASME)
has already issued the code for valves (please refer to ASME
Code Section I or Section VIII for the proper code).
(a) Pressure setting points. ASME Code stipulates that the pressure setting of a safety relief valve does not exceed 10 psig
or 20% of the operating pressure of the system or vessel.
(b) Capacity guidelines
(i) ASME Code—Section I. The total relieving capacity of the
valve shall not be less than the maximum operating pressure of the vessel of line, as designed by the manufacturer.
(ii) ASME Code—Section VIII. The minimum relieving
capability of the valve shall discharge the total amount of
the maximum operating pressure of the system or vessel,
without a rise in the pressure vessel in the event of overpowering the system.
(c) About sizing. It is important not to oversize a relief valve.
Typically, this will result in valve chatter or rapid opening
and closing of the valve seating and disk. If chattering is
present, it would be more economical to use two valves.
(3) Installation and maintenance for pressure relief valves. It should
be noted that only a qualified engineer who is familiar with pressure relief valves be allowed to perform all installations and
maintenances. These steps below are routinely followed in all
the installations and maintenances for pressure relief valves:
(a) First, you should turn off the operations system and enable
all pressure to bleed through prior to installation.
(b) Then, you can remove all thread protectors and plugs from
the valve.
(c) Note that valves should only be installed in an upright position, allowing for correct reseating of the valve disc upon
opening or popping.
(d) You must clean the connecting area of all dirt and grime;
then apply a small amount or piping compound to the valve
inlet side.
Zhang_Ch01.indd 164
5/13/2008 5:45:47 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
165
(e) Keeping away from the first few threads, the valve can be
tightened by hand to ensure proper thread alignments.
(f) By using the proper sized wrench on the hex shaped valve
body, you can tighten the system to a firm or snug setting by
a padded wrench as recommended.
(g) Please note that discharge piping must be equal to or larger
than the outlet size of the valve, ensuring that flow-rated
characteristics are not compromised.
(h) Note that discharge piping must be designed to anchor and
must be secured in a manner to prevent swaying, rattling, or
vibration.
(i) If the valve is venting to atmosphere, you should take all the
necessary steps to ensure that the outlet (discharge) is pointed
in a direction away from personnel or critical equipment.
(j) When testing a valve, the lift lever is designed to be opened
only when the system pressure is at 80% of the set pressure
(popping or cracking) point. The valve is kept in an open
position for a period of time, long enough to ensure the
cleansing of the seating area.
1.3.3
Solenoid Valves
A solenoid control valve is a kind of isolation valve that is an electromechanical device allowing for an electrical device to control the flow of gas
or liquid. The electrical device causes a current to flow through a coil
located on the solenoid valve, which in turn results in a magnetic field that
causes the displacement of a metal actuator. The actuator is mechanically
linked to a mechanical valve inside the solenoid valve. This mechanical
valve then opens or closes to allow a liquid or gas either to flow through or
be blocked by the solenoid valve. In this control system, a spring is used to
return the actuator and valve back to their resting states when the current
flow is removed. Figure 1.51 gives an application of the typical control
system with a solenoid valve. A coil inside the solenoid valve generates a
magnetic field once an electric current is flowing through. The generated
magnetic field actuates the ball valve that can change states to open or
close the device in the fluid direction indicated by the arrow.
Solenoid valves are used wherever fluid flow has to be controlled automatically. Factory automation makes a typical example of frequent use of
solenoid valves. A computer device running a factory automation program
to fill a container with some liquid can send a signal to the solenoid
valve to open, allowing the container to fill, and then remove the signal to
close the solenoid valve and stop the flow of liquid until the next container
is in place. A gripper for grasping items on a robot is frequently an
Zhang_Ch01.indd 165
5/13/2008 5:45:47 PM
166
INDUSTRIAL CONTROL TECHNOLOGY
Solenoid valve
S
Ball valve
Figure 1.51 A typical flow control system with solenoid valve (courtesy of Z-Tide
Valves).
air-controlled device. A solenoid valve can be used to allow air pressure to
close the gripper, and a second solenoid valve can be used to open the gripper. If a two-way solenoid valve is used, two separate valves are not needed
in this application. Solenoid valve connectors are used to connect solenoid
valves and pressure switches.
1.3.3.1
Operating Principles
(1) Solenoid. Solenoid valves are control units which, when electrically energized or deenergized, either shut off or allow fluid flow.
The actuator inside a solenoid valve takes the form of an electromagnet. When energized, a magnetic field builds up which pulls
a plunger or pivoted armature against the action of a spring.
When deenergized, the plunger or pivoted armature is returned
to its original position by the spring action.
According to the mode of actuation, a distinction is made
between direct-acting valves, internally piloted valves, and externally piloted valves. A further distinguishing feature is the number of port connections or the number of flow paths or “ways.”
With a direct-acting solenoid valve, the seat seal is attached to
the solenoid core. In the deenergized condition, a seat orifice is
closed, which opens when the valve is energized. With direct-acting
Zhang_Ch01.indd 166
5/13/2008 5:45:47 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
167
valves, the static pressure forces increase with increasing orifice
diameter, which means that the magnetic forces required for
overcoming the pressure force become correspondingly larger.
Internally piloted solenoid valves are, therefore, employed for
switching higher pressures in conjunction with larger orifice
sizes; in this case, the differential fluid pressure performs the
main work in opening and closing the valve.
The two-way solenoid valves are shut-off valves with one inlet
port and one outlet port as shown in Fig. 1.52(a). In the deenergized condition, the core spring, assisted by the fluid pressure,
holds the valve seal on the valve seat to shut off the flow. When
energized, the core and seal are pulled into the solenoid coil and
the valve opens. The electromagnetic force is greater than the
combined spring force and the static and dynamic pressure forces
of the medium.
The three-way solenoid valves have three port connections
and two valve seats. One valve seal always remains open and the
other closed in the deenergized mode. When the coil is energized, the mode reverses. The three-way solenoid valve shown in
Fig. 1.52(b) is designed with a plunger-type core. Various valve
operations are available according to how the fluid medium is
connected to the working ports in Fig. 1.52(b). The fluid pressure
builds up under the valve seat. With the coil deenergized, a conical spring holds the lower core seal tightly against the valve seat
and shuts off the fluid flow. Port A is exhausted through R. When
the coil is energized the core is pulled in, and the valve seat at
Port R is sealed off by the spring-loaded upper core seal. The
fluid medium now flows from P to A.
Unlike the versions with plunger-type cores, pivoted-armature
solenoid valves have all port connections in the valve body. An
isolating diaphragm ensures that the fluid medium does not come
into contact with the coil chamber. Pivoted-armature valves can
be used to obtain any three-way solenoid valve operation. The
basic design principle is shown in Fig. 1.52(c). Pivoted-armature
valves are provided with manual override as a standard feature.
Internally piloted solenoid valves are fitted with either a twoway or a three-way pilot solenoid valve. A diaphragm or a piston
provides the seal for the main valve seat. The operation of such a
valve is indicated in Fig. 1.52(d). When the pilot valve is closed,
the fluid pressure builds up on both sides of the diaphragm via a
bleed orifice. As long as there is a pressure differential between
the inlet and outlet ports, a shut-off force is available by virtue of
the larger effective area on the top of the diaphragm. When the
pilot valve is opened, the pressure is relieved from the upper side
Zhang_Ch01.indd 167
5/13/2008 5:45:48 PM
168
INDUSTRIAL CONTROL TECHNOLOGY
R
A
P
P
A
(a)
(b)
P
P
A
(c)
A
R
(d)
P
A
(e)
(f)
Figure 1.52 The operating principle of solenoid valve (courtesy of OMEGA).
Zhang_Ch01.indd 168
5/13/2008 5:45:48 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
169
of the diaphragm. The greater effective net pressure force from
below now raises the diaphragm and opens the valve. In general,
internally piloted valves require a minimum pressure differential
to ensure satisfactory opening and closing.
Internally piloted four-way solenoid valves are used mainly in
hydraulic and pneumatic applications to actuate double-acting
cylinders. These valves have four port connections: a pressure
inlet P, two cylinder port connections A and B, and one exhaust
port connection R. An internally piloted four/two-way poppet
solenoid valve is shown in Fig. 1.52(e). When deenergized, the
pilot valve opens at the connection from the pressure inlet to the
pilot channel. Both poppets in the main valve are now pressurized and switch over. Now port connection P is connected to A,
and B can exhaust via a second restrictor through R.
With these types an independent pilot medium is used to actuate the valve. Figure 1.52(f) shows a piston-operated angle-seat
valve with closure spring. In the unpressurized condition, the
valve seat is closed. A three-way solenoid valve, which can be
mounted on the actuator, controls the independent pilot medium.
When the solenoid valve is energized, the piston is raised against
the action of the spring and the valve opens. A normally open
valve version can be obtained if the spring is placed on the opposite side of the actuator piston. In these cases, the independent
pilot medium is connected to the top of the actuator. Doubleacting versions controlled by four/two-way valves do not contain
any spring.
(2) Manifold. The manifold of the solenoid valves consists of a matrix
of solenoid valves mounted in modules on a skid with adjustable
legs along one direction (Fig. 1.53). The quantity of the mounted
solenoid valves depends on the elements to be connected as tanks
or lines, and on the functions of each of these elements. A plurality of solenoid valves is arranged and placed on a solenoid valve
installing face of the manifold, and a board formed with an electric circuit for feeding these solenoid valves (Fig. 1.53). Each
solenoid valve includes a valve portion containing a valve member and a solenoid operating portion for driving the valve member. The board is mounted on the first side face of the manifold
under the solenoid operating portion. The board can be attached
and detached while leaving the solenoid valves mounted on the
manifold, feeding connectors and indicating lights being respectively provided in positions on the board corresponding to the
respective solenoid valves. Each the feeding connector is disposed in such a position that is connected to a receiving terminal
of the solenoid valve in a plug-in manner simultaneously with
Zhang_Ch01.indd 169
5/13/2008 5:45:50 PM
170
INDUSTRIAL CONTROL TECHNOLOGY
Figure 1.53 Several types of manifold of solenoid valves (courtesy of KIP Inc.).
mounting of the solenoid valve to the manifold. Each the indicating light is disposed in such a position that can be visually recognized from above the solenoid valve while leaving the solenoid
valve mounted on the manifold.
This manifold allows centralizing the functions of one or various tanks in a modular way, enhancing the efficiency of the system and control over the process. Manifold of the solenoid valves
is an automated alternative to the flexible hoses and the flow
divert panels with changeover bends. As many valves as the
number of functions the element has to perform are connected to
the tank or working line. No manual operation is required. The
operation is automated, preventing any risk of accidents.
1.3.3.2
Basic Types
Solenoid valves are opened and closed via a solenoid activated by an
electrical signal including all types of flow paths and proportional solenoid
valves. In most industrial applications, solenoid valves are arranged as the
following five types:
(1) Two-way solenoid valves. This type of solenoid valve normally
has one inlet and one outlet and is used to permit and shut off
fluid flow. The two types of operations for this type are “normally closed” and “normally open.”
Zhang_Ch01.indd 170
5/13/2008 5:45:50 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
171
(2) Three-way solenoid valves. These valves normally have three
pipe connections and two orifices. When one orifice is open, the
other is closed and vice versa. They are commonly used to alternately apply pressure to an exhaust pressure from a valve actuator or a single-acting cylinder. These valves can be normally
closed, normally open, or universal.
(3) Four-way solenoid valves. These valves have four or five pipe
connections, commonly called ports. One is a pressure inlet port,
and two others are cylinder ports providing pressure to the double-acting cylinder or actuator, and one or two outlets exhaust
pressure from the cylinders. They have three types of construction: single solenoid, dual solenoid, or single air operator.
(4) Direct mount solenoid valves. These series are two-way, threeway, and four-way solenoid valves that are designed for gang
mounting into different quantities of valves. Any combination of
normally closed, normally open, or universal valves may be grouped
together. These series are standard solenoid valves whose pipe
connections and mounting configurations have been replaced by
mounting configuration that allows each valve to be mounted
directly to an actuator without the use of hard piping or tubing.
(5) Manifolds. Manifolds are fluid distribution devices. They range
from simple supply chambers with several outlets to multichambered flow control units including integral valves and interfaces
to electronic networks. Manifolds are generally configured for
several outlets sharing one inlet or supply chamber; exhaust
manifolds can have several inlets sharing one exhaust port. They
may have one or more shared supply chambers and any number
of outlets. The manifold circuit style can be series or parallel. In
a series manifold the pressure supply is ported through one valve
to get to the next. In a parallel manifold the inlet ports all share
common pressure supply. Valve specifications to consider for
manifolds include integral manifold valves, integral valve types,
and solenoid valve power input. Integral valves are integrally
assembled with manifold, as opposed to a base or subplate to
which separate valves are attached. Integral valve choice types
include manual, solenoid, and air pilot. Manual valves are manually adjusted or actuated via knob, lever, or other manual device.
1.3.3.3 Technical Specifications
Solenoid valves are composed of several parts such as the solenoid coil,
electrical connector, bonnet nut, seal cartridge, O-rings, end connector,
body, and union nut. All these components are critical to the overall
Zhang_Ch01.indd 171
5/13/2008 5:45:50 PM
172
INDUSTRIAL CONTROL TECHNOLOGY
performance of solenoid valves. If there is any malfunction, it will affect
the entire operation of the automotive starter system, as well as the industrial air hammer and the electric bell assembly. That is why solenoid valves
should always be maintained and regularly checked in order to keep them
functioning at their best.
Performance specifications for solenoid valve connectors include connection voltage, nominal power, jacket material, conductor size, insulation
group, clamping voltage, bend radius.
Captive screw solenoid valve connectors prevent the attachment screw
from being lost. Low-profile solenoid valve connectors and right angle
solenoid valve connectors allow for installation in tight spaces. Options
for solenoid valve connectors include indicator lights and surge suppression. Solenoid valve connectors can also be part of a molded assembly,
which saves installation time. Solenoid valve connectors vary in terms of
applications and approvals. Some products are suitable for applications
such as steam, air, gas, water, pure water, light oil, heavy oil, or hightemperature fluids. Others are designed for cryogenics or corrosive fluids.
Complex pneumatic and hydraulic circuits can utilize manifolds with
interfaces to sophisticated electronic networks. Applications, port specifications, flow and pressure specifications, manifold circuit style, and valve
specifications are all important parameters to consider when searching for
manifolds. Additional specifications to consider for manifolds include communication network, body materials, features, and operating temperature.
Common applications for manifolds include general purpose, gas, pneumatic or compressed air, pneumatic or vacuum, water, steam, marine, coolant, refrigerant, cryogenic, high temperature, hydraulic fluid, oil or fuel,
slurry, high viscosity, general chemical, corrosive or solvent chemical, sanitary, food processing, and medical or pharmaceutical. Important port specifications to consider when searching for manifolds include supply ports,
outlet ports, and port types. Supply ports are the number of independent
fluid supplies that can be interfaced with the manifold. Outlet ports specify
the number of outlets. This is frequently specified as number of ports or
valves that are or can be attached to the manifold. Port type choices include
quickly connect and metric thread. Flow and pressure specifications are
important to consider; when selecting manifolds include maximum flow for
gas or air, maximum flow for liquid, and maximum pressure.
1.3.4
Float Valves
Float control systems monitoring the liquid or powder levels of containers such as tanks and reservoirs are installed with two kinds of sensors:
Zhang_Ch01.indd 172
5/13/2008 5:45:50 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
173
float switches and float valves. In a float control system, float switches are
used to detect liquid or powder levels, or interfaces between liquids; float
valves control the liquid or powder level in elevated tanks, standpipes, or
storage reservoirs and modulate the reservoir flow to maintain a constant
level.
Figure 1.54 is an assumed float control system monitoring the liquid
level of a high-temperature or high-pressure tank with both float switches
and float valves. Three switches are located along the right side of this tank
to detect three different levels of liquid by turning on either alarms or signals once the liquid reaches them, respectively. The two valves set at the
top and bottom of this tank are used to input or output the liquid according
to the control requirements to maintain the appropriate levels in this tank.
1.3.4.1
Operating Principle
Float control systems require both float switches and float valves to
accomplish their functions. Figure 1.55(a) is the diagram of a sample float
switch, and Fig. 1.55(b) is the diagram of a sample float valve.
(1) Float switch. (Fig. 1.55(a)) Float switches can be used either as
alarm devices or as control switches, turning something ON or
OFF, such as a pump, or sending a signal to a valve actuator.
What makes level switches special is that they have a switched
Emergency
cut out
High temperature or
high pressure tank
High level
Low level
Figure 1.54 An assumed float control system for a tank installed with three float
switches along the right side and two float valves at the top and the bottom.
Zhang_Ch01.indd 173
5/13/2008 5:45:50 PM
174
INDUSTRIAL CONTROL TECHNOLOGY
Inner chamber
Outer chamber
Magnet
Reed switch
and wiring
Rod
Approx.
level
Float
pilot
Float
Fingertrip
controller
(a)
(b)
Figure 1.55 (a) A float switch and (b) a float valve.
output and can be either electromechanical or solid state, either
normally open or normally closed.
Float switches provide industrial control for motors that pump
liquids from a sump or into a tank. For the tank operation, a float
operator assembly is attached to the float switch by a rod, chain,
or cable. The float switch is actuated based on the location of the
float in the liquid. The float switch contacts are open when the
float forces the operating lever to the Up position. As the liquid
level falls, the float and operating lever move downward. The
contacts can directly activate a motor or provide input for a logic
system to fill the tank. As the liquid level rises, the float and operating lever move upward. When the float reaches a preset high
level, the float switch contacts open, deactivating the circuit and
stopping the motor. However, sump operation is exactly the
opposite of tank operation.
(2) Float valve. (Fig. 1.55(b)) A float valve is mounted on the tank
or reservoir inlet, below or above the requested water level. The
float pilot can either be assembled on the main valve (for abovelevel installation) or be connected to the main valve by a command tube. The valve closes when the water level rises by filter
discharge pressure acting with the tension spring on the top of
the diaphragm in the valve cover chamber, thus raising the float
to its closed position. It opens when the float descends due to a
Zhang_Ch01.indd 174
5/13/2008 5:45:50 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
175
drop in water level by filter discharge pressure acting under the
valve disk. The difference between maximum and minimum levels is very small and is affected by the length of the float pilot
arm. In practice, the water level is maintained continuously at the
maximum point as long as the upstream flow exceeds the downstream flow.
1.3.4.2
Specifications and Application Guide
The typical options for poles and throws are available. Most float switches
and valves have either one or two poles or one or two throws, but some
manufacturers will produce custom level switches for specifications.
The measuring range is probably the most important specification to
examine when choosing switches and valves. Also of critical concern are
the ratings for current and voltage the switches require.
Depending on the needs of the application, float switches and valves
can be mounted different ways. These switches and valves can be mounted
on the top, bottom, or side of the container holding the substance to be
measured. Among the technologies for measuring level is air bubbler technology, capacitive or RF admittance, differential pressure, electrical conductivity or receptivity, mechanical or magnetic floats, optical units,
pressure membrane, radar or microwave, radio frequency, rotation paddle,
ultrasonic or sonic, and vibration or tuning fork technology. Analog outputs from level switches can be current or voltage signals. Also possible is
a pulse or frequency. Computer signal outputs that are possible are usually
serial or parallel.
Float switches and valves can have displays that are analog, digital, or
video displays. Control for the devices can be analog with switches, dials,
and potentiometers; digital with menus, keypads, and buttons; or controlled by a computer. Some features that can make float switches and
valves more desirable are being programmable, having controller, recorder,
or totalizer functions and a built-in alarm indicator, whether audible or
visible. Also important for some applications are sanitary ratings and the
ability to handle slurries with suspended solids, such as wastewater or
sewage.
1.3.4.3
Calibration
Many calibration laboratories are finding themselves facing more stringent accuracy requirements. Local gravity, the effects of air buoyancy on
the piston gauge, and masses and temperature all affect the accuracy of the
Zhang_Ch01.indd 175
5/13/2008 5:45:51 PM
176
INDUSTRIAL CONTROL TECHNOLOGY
results as well as the uncertainty of the pressure being generated and
should be calibrated or measured. The error may in fact be greater—in
some cases much greater—than anticipated. The quality of the calibration
output is dependent on the skill and knowledge of the operator.
A digital transfer standard can improve accuracy in these cases, because
it does not generate pressure and is not affected by local gravity, the effects
of air buoyancy, or the age and condition of masses. If a high-accuracy
digital transfer standard is used, these error sources are not present.
The pressure generated for the calibration is more likely closer to the
true pressure, resulting in better calibration. Automated digital transfer
standards deliver equal or higher precision than that provided by many
industrial dead-weight gauges. The precision of digital transfer standards
usually ranges from 0.01% F.S. (Frequency Series or Fourier series) to
0.003% F.S., with total accuracies depending on the calibration standard
used. The total accuracy of the digital transfer standard includes the accuracy of the primary standard used to calibrate it.
Interfaces to PCs can be either through calibration software and gauge
monitoring devices, or through the PC interfaces found in digital transfer
standards. The interfaces to PCs dramatically improve the timeliness
of reports and simplify the reporting process. Templates created in offthe-shelf word-processing or spreadsheet programs and accessed by calibration systems can include organization logos, addresses and contact
information, and calibration standard information (e.g., the pressure range
and the serial number and other identifying information for the the device
under test). Data from calibrations can be automatically incorporated into
the template files. This process allows managers to report as-found/as-left
data and a host of other parameters and then save the files for retrieval later.
A calibration sequence (a set of pressure points) is sent to the Model
2492 either via the IEEE 488 interface using a remote computer interface
or locally via the system keyboard. The system’s microprocessor calculates the masses required generating a requested pressure, automatically
correcting for environmental factors (e.g., local gravity, air density, head
pressure, and temperature). The automated mass loading system selects
and loads the amount of mass needed to generate the desired pressure.
Once the masses are loaded, a self-recharging pump pressurizes the system to float the masses. When the masses reach the proper float position,
the pump is turned off to ensure that a static pressure condition is met.
Float position indicator and resistance temperature devices evaluate float
position, sink rate, and temperature in determining a valid float condition.
When conditions are met, the system signals the user (or host PC in remote
operations), and the balance is maintained until a new command is entered.
Zhang_Ch01.indd 176
5/13/2008 5:45:51 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
177
If desired, the remote host computer can automatically acquire data from
the device under test without an operator present.
1.3.5
Flow Valves
Measuring the flow of liquids and gases is a critical need in many industrial productions such as chemical packaging, wastewater treatments, and
aircraft manufacturing. In some industrial operations, the ability to conduct accurate flow measurements is so important that it can make the difference between making a profit and taking a loss. In other cases, failure in
flow control can cause serious (or even disastrous) results. Flow control
systems are designed specifically for operation with a wide range of liquids
and gases in both safe and hazardous areas, in hygienic, high-temperature,
and high-pressure environments, and for use with aggressive media. Some
applications for this purpose are pump protection, cooling circuit protection, high and low flow rate alarm, general flow monitoring, etc.
1.3.5.1
Operating Principle
The technologies for gas flow meters and liquid flow meters vary widely.
The most common types of operation principles are inferential flow measurement, positive displacement measurement, velocity measurement, true
mass flow measurement, and thermodynamic loss measurement. The main
types of operating principles for gas flow meters and liquid flow meters are
given below:
(1) Gas flow switches and liquid flow switches, velocity. Gas flow
switches and liquid flow switches, velocity, are used to measure
the flow or quantity of a moving fluid in terms of velocity, such
as feet per minute. The most common types are inferential flow
measurement, positive displacement, velocity meters, and true
mass flow meters.
Inferential measurement refers to the indirect measurement of
flow by directly measuring another value and inferring the flow
based on well-known relationships between the directly measured value and flow. The use of differential pressure as an
inferred measurement of a liquid’s rate of flow is the most common type of unit in use today.
Positive displacement meters take direct measurements of liquid flows. These devices divide the fluid into specific increments
and move it on. The total flow is an accumulation of the measured increments, which can be counted by mechanical or electronic techniques. They are often used for high-viscosity fluids.
Zhang_Ch01.indd 177
5/13/2008 5:45:51 PM
178
INDUSTRIAL CONTROL TECHNOLOGY
Velocity-type gas flow switches and liquid flow switches are
devices that operate linearly with respect to volume flow rate.
Because there is no square-root relationship, as with differential
pressure devices, their range ability is greater.
(2) Liquid flow switches and gas flow switches, mass. Liquid flow
switches and gas flow switches, mass, are devices used for measuring the flow or quantity of a moving liquid or gas, respectively, in terms of unit of mass per unit time, such as pounds per
minute.
These may be sensors with electrical output or may be standalone instruments with local displays and controls. The most
common types of liquid flow switches and gas flow switches,
mass, are true mass flow meters.
True mass flow meters are devices that measure mass rate of
flow directly, such as thermal meters or Carioles meters. The
most important specifications for mass gas flow meters and
liquid flow meters and sensors are the flow range to be measured
and whether liquids or gases will be the measured fluids. Also
important are operating pressure, the fluid temperature, and
accuracy. Typical electrical outputs for mass gas flow meters
and liquid flow meters are analog current, voltage, frequency, or
switched output. Computer output options can include serial and
parallel interfaces. These sensors can be mounted either as inline
or insertion devices. Inline sensors can be held in place by using
flanges, threaded connections, or clamps. Insertion style sensors
are typically threaded through a pipe wall and stick directly in
the process flow.
(3) Gas flow switches and liquid flow switches, volumetric. Gas flow
switches, volumetric, provide output based on the measured flow
of a moving gas in terms of volume per unit time, such as cubic
feet per minute. Liquid flow switches, volumetric, are devices
with a switch output used for measuring the flow or quantity of a
moving fluid in terms of a unit of volume per unit time, such as
liters per minute.
The basis of volumetric gas flow switch and liquid flow switch
selection is a clear understanding of the requirements of the particular application. With most liquid flow measurement instruments, the flow rate is determined inferentially by measuring the
liquid’s velocity or the change in kinetic energy. Velocity depends
on the pressure differential that is forcing the liquid through a
pipe or conduit. Because the pipe’s cross-sectional area is known
and remains constant, the average velocity is an indication of the
flow rate. The basic relationship for determining the liquid’s flow
rate in such cases is Q = V × A, where Q is the liquid flow through
Zhang_Ch01.indd 178
5/13/2008 5:45:51 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
179
a pipe, V is the average velocity of the flow, and A is the crosssectional area of the pipe. Other factors that affect liquid flow
rate include the liquid’s viscosity and density. Specific applications should be discussed with a volumetric gas flow switch
manufacturer before purchasing to ensure proper fit, form, and
function.
Both volumetric gas flow switches and volumetric liquid flow
switches are available with four different meter types: inferential
flow meters, positive displacement meters, velocity meters, and
true mass flow meters. The basic operating principle of differential pressure flow meters is that the pressure drop across the meter
is proportional to the square of the flow rate. The flow rate is
obtained by measuring the pressure differential and extracting
the square root. Direct measurements of liquid flows can be made
with positive-displacement flow meters. These devices divide the
liquid into specific increments and move it on. The total flow is
an accumulation of the measured increments, which can be
counted by mechanical or electronic techniques. They are often
used for high-viscosity fluids.
(4) Pneumatic relays. Pneumatic relays control output air flow and
pressure in response to a pneumatic input signal. They can perform simple functions such as boosting or scaling the output, or
complex reversal, biasing, and math function operators.
1.3.5.2
Specifications and Application Guide
The most common specifications for flow control systems can be pressure drop, system efficiency, and process modifications.
(1) Pressure drop. Pressure drop is defined by the following items:
(a) Reduce piping system pressure loss by increasing the line
size and rerouting the pipes.
(b) Optimize process equipment pressure loss in the flow lines.
(c) Reduce or eliminate control valve pressure loss.
(2) System efficiency. System efficiency includes the following
operations:
(a) Operate rotating equipment at or close to its best efficiency
point.
(b) Split a single system into more than one to achieve higher
aggregate O and M efficiency.
(c) Downsize the pump (or trim impeller or modify compressor
wheels) and motors. This is only an end result of system
improvement or optimization.
Zhang_Ch01.indd 179
5/13/2008 5:45:51 PM
180
INDUSTRIAL CONTROL TECHNOLOGY
(d) Modify a process control scheme, including loading and
unloading and spillback controls on positive displacement
compressors, and location of throttling control.
(e) Install variable frequency drives (VFD)—investigate needs
and implications carefully. The variable frequency drive is
not always the best solution!
(f) Improve operating efficiency by implementing an energy
recovery strategy. For example, the recovery heat of compression from heating a building or from replacing low pressure steam heating.
(g) Replace standard efficiency motors with premium efficiency
motors.
(3) Process modifications. Process modifications include the following items:
(a) Lower the delivery pressure (without impacting process
requirement) at user(s).
(b) Lower the pressure profile of the system.
(c) Incorporate an advanced process control scheme and algorithm to eliminate operator intervention.
(d) Reroute or resequence the process streams.
(e) Reduce the compressor inlet temperature.
(f) Regenerate or replace catalyst (same for inline filters) to
eliminate an excessive pressure drop due to carbon build-up.
(g) Reduce or eliminate the minimum flow bypass recirculation
flow.
1.3.5.3
Calibration
The calibration of a flow control system is facilitated with the following
equipments:
(1) Gas flow calibrator. The gas flow calibrator (GFC) is an automated, sonic nozzle, scalable, state-of-the-art test system providing exceptional metrology for all gas metering technologies. The
customizable, turnkey system includes hardware, software, and
accessories, and supports testing on different meter types at pressures ranging from atmospheric to over 100 psig with closed
loop control.
(2) Flow controller. The flow controller (FC) was the first commercially available flow computer that accommodates combinations
of sonic nozzles configured in a multiple sonic nozzle array. This
unique functionality allows the flow controller to act as a standalone flow controller. The flow controller can interface with a
single sonic nozzle and/or subsonic Venturi, or up to several
sonic nozzles, thus offering a ratio range in flow.
Zhang_Ch01.indd 180
5/13/2008 5:45:51 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
181
(3) Sonic nozzles. The sonic nozzle, also known as a “critical flow
Venturi” or “critical flow nozzle” has rapidly gained acceptance
as a flow measurement standard and flow meter. Sonic nozzles
are now utilized in many diverse applications by the aerospace,
automotive, energy, and metrology industries. Sonic nozzles can
be used as a calibration standard for gas flow meters or any flow
measurement device. By design, sonic nozzles are a constant
volumetric flow meter. However, with the use of a regulated pressure supply, the sonic nozzle becomes a state-of-the-art mass
flow meter.
(4) Flow computer. Flow computers work with a single sonic nozzle
and/or subsonic Venturi and can also accommodate combinations of sonic nozzles configured in a multiple sonic nozzle array.
With the increasing popularity of using multiple, binary throat
areas, sonic nozzles installed in a common inlet plenum. Flow
Systems introduces the flow computer to meet the challenges of
this application. Flow computers are able to interface with single,
interchangeable, sonic nozzles and/or subsonic Venturis, or up to
some sonic nozzles in a multiple sonic nozzle array. This unique
solution (hardware and software) combines instrumentation, data
acquisition and control, computation, monitoring, and data logging to meet the needs of the high-end flow measurement user.
Bibliography
AccesIO (http://www.accesio.com). 2000. http://www.accesio.com/manuals/lvdt-8
.pdf. Accessed date: April 2005.
Alibaba (http://www.alibaba.com). 2005. http://www.alibaba.com/productsearch/
Ultrasonic_Distance_Meter.html. Accessed date: April.
AQUATIC (http://www.aquaticeco.com). 2005. http://www.aquaticeco.com/index
.cfm/fuseaction/listings.categories/ssid/353. Accessed date: May.
Asahi-America (http://asahi-america.com). 2005a. http://asahi-america.com/
documents/documents/Pneumatic%20Actuator%20Intro.pdf. Accessed date:
May.
Automation Direct (http://web5.automationdirect.com). 2005. http://web5
.automationdirect.com/adc/Overview/Catalog/Sensors_-z-_Encoders/
Ultrasonic_Sensors. Accessed date: April.
BEAMEX (http://www.beamex.com). 2005. http://www.beamex.com/services/
services.html. Accessed date: May.
Calibrator Depot (http://www.calibratordepot.com). 2005. Calibrators and Calibration
Services. http://www.calibratordepot.com/index.asp. Accessed date: June.
Carl T. Lira (lira@egr.msu.edu). 2001. Check Valve and Pump Description. http://
www.egr.msu.edu/~lira/supp/Check Valves and Pumps.mht. Accessed date:
September 2007.
Zhang_Ch01.indd 181
5/13/2008 5:45:51 PM
182
INDUSTRIAL CONTROL TECHNOLOGY
Chris Warnett (Rotork Controls Inc.). 2004. A descriptive definition of valve actuators. http://www.valve-world.net/actuation/ShowPage.aspx?pageID=557.
Accessed date: September 2007.
COMPACT (http://www.compact4.com). 2005a. http://www.compact4.com/
technical.html. Accessed date: May.
Crane Valve (http://www.cranevalve.com/index.htm). 2007. Electric Actuators.
http://www.cranevalve.com/Act_Elec.htm. Accessed date: September.
DALSA (http://www.dalsa.com). 2005a. http://www.dalsa.com/markets/ccd_vs_
cmos.asp. Accessed date: April.
Direct Industry (http://www.directindustry.com). 2005a. http://www.directindustry
.com/industrial-manufacturer/magnetic-sensor-70932.html. Accessed date: April.
Direct Industry (http://www.directindustry.com). 2005b. http://www.directindustry
.com/industrial-manufacturer/magnetic-switch-72332.html. Accessed date:
April.
Direct Industry (http://www.directindustry.com). 2005c. http://www.directindustry
.com/industrial-manufacturer/hall-effect-sensor-71708.html. Accessed date:
April.
Direct Industry (http://www.directindustry.com). 2005d. http://www.directindustry
.com/industrial-manufacturer/reed-switch-74782.html. Accessed date: April.
DMN Digital (http://www.digitalmedianet.com). 2005. http://cad.digitalmedianet
.com/articles/viewarticle.jsp?id=25291. Accessed date: April.
EFLOTS-INFO (http://efloatswitch.info/). 2005. Accessed date: May.
EMERSON (http://www.emersonprocess.com). 2005a. http://www.documentation
.emersonprocess.com/groups/public/documents/book/cvh99.pdf. Accessed
date: May.
EMERSON (http://www.emersonprocess.com). 2005b. http://www.emersonprocess
.com/valveautomation/bettis/Downloads/Hydraulic_Manuals.htm. Accessed
date: May.
FineMech (http://www.finemech.com). 2005. http://www.finemech.com/
Sensors%20category%202.shtml?gclid=COuHoPvhsosCFQPilAodU1qWvw.
Accessed date: April.
Forbes Marshall Inc. (http://www.forbesmarshall-inc.com). 2007. Pneumatic
Actuator
Manual.
http://www.forbesmarshall-inc.com/ProductPage
.asp?ProdGrpId=29&ProdId=79. Accessed date: September.
Fraunhofer-ILT (http://www.ilt.fraunhofer.de). 2005. http://www.ilt.fraunhofer.de/
eng/ilt/pdf/eng/products/HZ_LTS-2D.pdf. Accessed date: December.
Free Study (http://www.freestudy.co.uk). 2005a. http://www.freestudy.co.uk/
control/t2.pdf. Accessed date: May.
GlobalSpec (http://www.globalspec.com). 2005a. http://sensors-transducers
.globalspec.com/SpecSearch/Suppliers?QID = 8900153 & Comp = 106 .
Accessed date: April.
GlobalSpec (http://www.globalspec.com). 2005b. http://search.globalspec.com/
productfinder/findproducts?query=color%20sensor. Accessed date: April.
GlobalSpec (http://www.globalspec.com). 2005c. http://search.globalspec.com/
productfinder/findproducts?query = Ultrasonic%20distance%20sensors .
Accessed date: April.
GlobalSpec (http://www.globalspec.com). 2005d. http://sensors-transducers
.globalspec.com/LearnMore/Sensors_Transducers_Detectors/Linear_
Position_Sensing/LVDT_Position_Sensors. Accessed date: April.
Zhang_Ch01.indd 182
5/13/2008 5:45:51 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
183
GlobalSpec (http://www.globalspec.com). 2005e. http://sensors-transducers
.globalspec.com/LearnMore/Sensors_Transducers_Detectors/Rotary_
Position_Sensing/Rotary_Position_Sensors. Accessed date: April.
GlobalSpec (http://www.globalspec.com). 2005f. http://search.globalspec.com/
Search?query=hydraulic&show=total. Accessed date: May.
GlobalSpec (http://www.globalspec.com). 2005g. http://flow-control.globalspec
.com/Specifications/Flow_Transfer_Control/Valve_Actuators_Positioners/
Electric_Electronic_Motor_Actuators. Accessed date: June.
GlobalSpec (http://www.globalspec.com). 2005h. http://flow-control.globalspec
.com/LearnMore/Flow_Control_Flow_Transfer/Valves/Check_Valves .
Accessed date: June.
Grainger (http://www.grainger.com). 2005. http://www.grainger.com/production/
info/limit-switch.htm. Accessed date: April.
Hamamatsu (http://www.sales.hamamatsu.com). 2005. http://www.sales.hamamatsu
.com/en/products/solid-state-division/color_sensors.php. Accessed date: April.
HASCO (http://www.hascorelays.com). 2003. http://www.hascorelays.com/reed_
switches.asp. Accessed date: April 2005.
Honey Well (http://hpsweb.honeywell.com). 2005a. http://hpsweb.honeywell.com/
Cultures/en-US/IndustrySolutions/PulpPaperPrinting/PaperMeasurement/
ColourMeasurement/default.htm. Accessed date: April.
Honey Well (http://hpsweb.honeywell.com). 2005b. http://content.honeywell.com/
sensing/prodinfo/solidstate/technical/chapter2.pdf. Accessed date: April.
Honey Well (http://hpsweb.honeywell.com). 2005c. http://www.ssec.honeywell
.com/magnetic/datasheets/sae.pdf. Accessed date: April.
Honey Well (http://hpsweb.honeywell.com). 2005d. http://sensing.honeywell.com/.
Accessed date: April.
Honey Well (http://hpsweb.honeywell.com). 2005e. http://www.honeywell.com
.pl/pdf/automatyka_domow/palniki_do_kotlow/golden/Silowniki/V4062.pdf.
Accessed date: April.
Hydraulic-Tutorial (http://www.hydraulicsupermarket.com). 2005a. http://www
.hydraulicsupermarket.com/technical.html. Accessed date: May.
IQS Directory (http://www.iqsdirectory.com). 2007. Actuators. http://www
.iqsdirectory.com/actuators/. Accessed date: September.
Johnson Controls, Inc. 1997. Product/Technical Bulletin for VA-7200 Electric
Valve Actuator.
KEYENCE-America (http://www.keyance.com). 2005. http://www.keyence.com/
products/sensors/rgb/rgb.php. Accessed date: April.
KIP Inc. (http://www.kipinc.com). 2005. Solenoid Valves. http://www.norgren
.com/kip/. Accessed date: June.
LESLIECONTROLS (http://www.lesliecontrols.com). 2005. http://www
.lesliecontrols.com/Handbooks/Handbooks.html. Accessed date: May.
Leuze Electronics (http://www.leuze.de). 2003. http://www.leuze.de/downloads/
los/03/brusds_e.pdf. Accessed date: April of 2005.
Liteon-Semi (http://www.liteon-semi.com). 2005. http://www.liteon-semi.com/_
en/02_cis/00_overview.php. Accessed date: May.
Lvdt.co.uk (http://www.lvdt.co.uk). 2005a. http://www.lvdt.co.uk/howtheywork
.html. Accessed date: April.
Lvdt.co.uk (http://www.lvdt.co.uk). 2005b. http://www.lvdt.co.uk/selection.html.
Accessed date: April.
Zhang_Ch01.indd 183
5/13/2008 5:45:51 PM
184
INDUSTRIAL CONTROL TECHNOLOGY
Machine Design (http://www.machinedesign.com). 2003. http://www.machinedesign
.com/Switch Tips Bimetallic switches.mht. Accessed date: February 2005.
Machine Design (http://www.machinedesign.com). 2005a. http://productsearch
.machinedesign.com/browse/Motion_Controls/Limit_Switches. Accessed date:
April.
Machine Design (http://www.machinedesign.com). 2005b. http://productsearch
.machinedesign.com/browse/Motion_Controls/Linear_Rotary_Motion_
Components. Accessed date: May.
Macro Sensors (http://www.macrosensors.com). 2005a. http://www.macrosensors
.com/lvdt_macro_sensors/lvdt_tutorial/why_use_lvdt.html. Accessed date:
April.
Macro Sensors (http://www.macrosensors.com). 2005b. http://www.macrosensors
.com/ms-lvdt_faq-tutorial.html. Accessed date: April.
Maintenance-World (http://www.maintenanceworld.com). 2005. http://www
.maintenanceworld.com/Articles/wmaorg/Valve.pdf. Accessed date: May.
Making Things (http://www.makingthings.com). 2005. http://www.makingthings
.com/teleo/products/acc_datasheets/acc_usd_001.htm. Accessed date: April.
Metra Mess- und Frequenztechnik (http://www.mmf.de). 2005. Piezoelectric
Accuracy and Calibration. http://www.mmf.de/introduction.htm. Accessed
date: June.
Microchip (http://www.microchip.com). 2005. http://ww1.microchip.com/
downloads/en/DeviceDoc/39757a.pdf. Accessed date: May.
Migatron
(http://www.migatron.com).
2005.
http://www.migatron.com/
understanding_ultrasonics.htm. Accessed date: April.
MORGAN (http://www.morganelectroceramics.com). 2005. http://www.morgan
electroceramics.com/tutorials/piezoguide1.html. Accessed date: May.
NewArk (http://www.newark.com). 2005. http://www.newark.com/pdfs/techarticles/
pepperl.pdf. Accesses date: April.
NEWPORT (http://www.newport.com). 2005. http://www.newport.com/
Manual-Actuator-Selection-Guide/168530/1033/catalog.aspx. Accessed
date: May.
NI (http://zone.ni.com). 2005. http://zone.ni.com/devzone/cda/tut/p/id/3602.
Accessed date: May.
Nook Industries (http;//www.nookindustries.com). 2004. http://www.nookindustries
.com/pdf/CylinderLimitSwitch.pdf. Accessed date: April 2005.
NPL (http://www.npl.co.uk). 2005. http://www.npl.co.uk/pressure/guidance/
nonprescals.html. Accessed date: May.
OMEGA (http://www.omega.com). 2005a. http://www.omega.com/techref/pdf/
STRAIN_GAGE_TECHNICAL_DATA.pdf. Accessed date: May.
OMEGA (http://www.omega.com). 2005b. http://www.omega.com/techref/
flowmetertutorial.html. Accessed date: May.
OMEGA (http://www.omega.com). 2005c. http://www.omega.com/techref/
techprinc.html. Accessed date: May.
OMRON (http://www.omoron.com). 2005a. http://www.simcotech.com/sensors/
omronrgb.pdf. Accessed date: April.
OMRON (http://www.omron.com). 2005b. http://www.mikrokontrol.co.yu/katalog/
datasheets/senzori/foto/e3mc.pdf. Accessed date: April.
OMRON (http://www.omron.com). 2005c. http://www.sti.com/switches/swdatash
.htm. Accessed date: April.
Zhang_Ch01.indd 184
5/13/2008 5:45:51 PM
1: SENSORS AND ACTUATORS FOR INDUSTRIAL CONTROL
185
PAControl (http://electricalequipment.pacontrol.com). 2005a. http://electricalequip
ment.pacontrol.com/proximitysensors.html. Accessed date: April.
PAControl (http://electricalequipment.pacontrol.com). 2005b. http://electricalequip
ment.pacontrol.com/capacitiveproximitysensors.html. Accessed date: April.
PAControl (http://electricalequipment.pacontrol.com). 2005c. http://electricalequip
ment.pacontrol.com/inductiveproximitysensors.html. Accessed date: April.
PAControl (http://electricalequipment.pacontrol.com). 2005d. http://electricalequip
ment.pacontrol.com/magneticproximitysensors.html. Accessed date: April.
PCB PIEZOTRONICS (http://www.pcb.com). 2005. Mounting_force_sensors.pdf.
Accessed date: June.
PEPPERL+FUCHS (http://www.am.pepperl-fuchs.com). 2005a. http://www.am
.pepperl-fuchs.com/products/productfamily.jsp?division=FA&productfamily_
id=1455. Accessed date: April.
PEPPERL+FUCHS (http://www.am.pepperl-fuchs.com). 2005b. http://www.am
.pepperl-fuchs.com/products/productfamily.jsp?division=FA&productfamily_
id=1575. Accessed date: April.
PHILIPS. 2000. http://www.nxp.com/acrobat_download/various/SC17_GENERAL_
MAG_2-1.pdf. Accessed date: April 2005.
Physics-psu.edu. 2004. http://class.phys.psu.edu/p457/experiments/html/hall_
effect_2004.htm. Accessed date: April 2005.
PI (http://www.physikinstrumente.com). 2005. http://www.physikinstrumente
.com/en/products/piezo_tutorial.php. Accessed date: May.
PIEZO (http://www.piezo.com). 2005. http://www.piezo.com/tech2intropiezotrans
.html. Accessed date: May.
Plant-Maintenance (http://www.plant-maintenance.com). 2005. http://www
.plant-maintenance.com/maintenance_articles_valves.shtml. Accessed date: May.
Red Soft (http://www.redsofts.com). 2005a. http://www.redsofts.com/articles/
read/386/70578/Electric_Linear_Actuators_The_Rotary_Motion_Producer
.html. Accessed date: May.
Red Valve (http://www.redvalve.com). 2005. http://www.redvalve.com/control.html.
Accessed date: May.
Rheodyne Corporation (http://www.rheodyne.com). 2007. Operating Instruction for
Pneumatic Actuators. http://www.rheodyne.com/support/product/instructions/.
Accessed date: September.
Robotica (http://www.robotica.co.uk). 2005. http://www.robotica.co.uk/robotica/
ramc/products/sensors/ultra_sensor.htm. Accessed date: April.
SENSORS (http://www.sensorsmag.com). 2005a. http://www.sensorsmag.com/
articles/1298/mag1298/main.shtml. Accessed date: April.
SENSORS (http://www.sensorsmag.com). 2005b. http://www.sensorsmag.com/
sensors/article/articleDetail.jsp?id=179165. Accessed date: May.
Short Courses (http://www.shortcourses.com). 2005. http://www.shortcourses
.com/choosing/sensors/05.htm. Accessed date: April.
SICK (http://www.sick.com). 2005a. http://www.sick.com/home/factory/catalogues/
industrial/coloursensors/en.html. Accessed date: April.
SICK (http://www.sick.com). 2005b. http://englisch.meyle.de/contract_partner/
sick_magnetic_proximity_sensors.php. Accessed date: April.
SIEMENS (http://www.sbt.siemens.com). 2005. Actuators and Valves. http://www
.sbt.siemens.com/hvp/components/products/damperactuators/default.asp.
Accessed date: June.
Zhang_Ch01.indd 185
5/13/2008 5:45:52 PM
186
INDUSTRIAL CONTROL TECHNOLOGY
SONY (http://www.sony.com). 2005. http://www.docs.sony.com/release/
GDMC520Kguide.pdf. Accessed date: April.
SukHamburg (http://www.SukHamburg.de). 2005. http://www.silicon-software
.de/download/archive/Laser_Light_Section_e.pdf. Accessed date: December.
VNE (http://www.vnestainless.com/default.aspx). 2005. Valves. http://www
.vnestainless.com/default.aspx. Accessed date: September 2007.
WATTS (http://www.watts.com). 2005. http://www.watts.com/pro/_products_sub
.asp?catId=69&parCat=125. Accessed date: May.
Young (http://www.youngcalibration.co.uk). 2005. http://www.youngcalibration
.co.uk/calibration.htm?gclid=CMSGqLDX9IsCFQ7dlAodBHewVQ. Accessed
date: May.
Z-Tide Valves (http://www.z-tide.com.tw). 2005. Valves. http://www.z-tide.com.tw/
multi/index-e.htm. Accessed date: June.
Zhang_Ch01.indd 186
5/13/2008 5:45:52 PM
2
Computer Hardware for
Industrial Control
2.1 Microprocessor Unit Chipset
The microprocessor within personal computers and industrial controllers has evolved continuously, with each newer version being compatible
with the previous ones. The major producer of microprocessors has been
Intel Corporation (Intel is a contraction of Integrated Electronics) since the
early 1970s. Intel marketed the first microprocessor in 1971, named the
4004, which caused a revolution in the electronics industry. With this processor, the functionality started to be programmed by software. However,
it could only handle 4 bits of data at a time (a nibble), contained 2000
transistors, had 46 instructions, and allowed 4 kB of program code and
1 kB of data. However, from this, personal computers and industrial controllers have evolved with the use of Intel microprocessors.
(1) First generation. The next generation of Intel microprocessors
arrived in 1974, which could handle 8 bits (a byte) of data at a
time and were named the 8008, 8080, and 8085. The 8008 had a
14-bit address bus and can thus address up to 16 kB of memory;
the 8080 had a 16-bit address bus giving it a 64 kB limit.
(2) Second generation. The next generation came with the launch of
the 16-bit processors. Intel released the 8086 microprocessor,
which was mainly an extension to the original 8080 processor,
and thus retained a degree of software compatibility. It had a
16-bit data bus and a 20-bit address bus, and thus had a maximum
addressable capacity of 1 MB. The 8086 could handle either 8 or
16 bits of data at a time, although in a messy way. A strippeddown, 8-bit external data bus version called as the 8088 was also
available. This stripped-down processor allowed designers to
produce less complicated systems. An improved architecture
version, the 80286, was launched in 1982 and was used in the
IBM Advanced Technology.
(3) Third generation. In 1985, Intel introduced its first 32-bit microprocessor, the 80386DX. This device was compatible with the
previous 8088/8086/80286 (80x86) processors and gave excellent performance, handling 8, 16, or 32 bits at a time. It had full
32-bit data and address buses and could thus address up to 4 GB
of physical memory. A stripped-down 16-bit external data bus
187
Zhang_Ch02.indd 187
5/13/2008 5:53:42 PM
188
INDUSTRIAL CONTROL TECHNOLOGY
and 24-bit address bus version called 80386SX was released in
1988, which could only access up to 16 MB of physical memory.
(4) Fourth generation. In 1989, Intel introduced the 80486DX, which
was basically an improved 80386DX with a memory cache and
math coprocessor integrated onto the chip. It had an improved
internal structure making it around 50% faster than a compatible
80386. The 80486SX was also introduced, which was merely an
80486DX with the link to the math coprocessor broken. As processor speeds increased, there was a limiting factor for the
system clock speed, thus the system clock was doubled or tripled to
produce the processor clock. Typically, a system with clock double
processors is around 75% faster than the compatible nondoubled
processors. Intel has also produced a range of 80486 microprocessors, which run at three or four times the system clock speed
and are referred to as DX4 processors. These include the Intel
DX4-100 and Intel DX4-75, both with a 25 MHz clock.
(5) Fifth generation. The Pentium (or P-5) is a 64-bit superscalar
processor. It can execute more than one instruction at a time and
has a full 64-bit (8-byte) data bus and a 32-bit address bus. In terms
of performance, it operates almost twice as fast as the equivalent
80486. It also has improved floating-point operations (roughly
three times faster) and is fully compatible with previous 80 × 86
processors.
(6) Sixth generation. The Pentium II/III and Pentium Pro (or P-6)
are enhancements of the P-5 and have a bus that supports up to
four processors without extra supporting logic, with clock multiplying speeds of over 1 GHz. They also have major savings of
electrical power and the minimization of electromagnetic interference. There is a great enhancement of the P-6 bus in that it
detects and corrects all single-bit data bus errors and also detects
multiple bit errors on the data bus.
(7) Seventh generation. New features added with the AMD K7 Athlon
include (1) ultrahigh clock speeds of over 1 GHz, (2) 128 kB
level 1 cache and up to 8 MB for level 2 cache, (3) capability to use
and rearrange up to 72 instructions simultaneously, (4) ability to
execute up to nine instructions simultaneously.
Processors come and go, but most manufacturers know the important
differences are often processor clock speeds and cache sizes. The future is
likely to involve an increase in real-time audio and video over the Internet.
Although the performance of today’s processors continues to improve,
existing architectures based on an out-of-order execution model require
increasingly complex hardware mechanisms and are impeded increasingly
by performance limiters such as branches and memory latency. Some new
Zhang_Ch02.indd 188
5/13/2008 5:53:42 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
189
architectures, like IA-64 architecture and x84-64 architecture, are a unique
combination of innovative features, including explicit parallelism, prediction, and speculation, which are described below:
(1) Parallelism. In today’s processor architectures, the compiler
creates sequential machine codes that attempt to imply parallelism to hardware. The processor’s hardware must then reinterpret this machine code and try to identify opportunities for
parallel execution, which is the key to higher performance. This
process is inefficient not only because the hardware does not
always interpret the compiler’s intentions correctly, but also
because it uses a valuable die area that could be better used to
do real work like executing instructions. Even today’s fastest
and most efficient processors devote a significant percentage of
hardware resources to this task of extracting more parallelism
from software code.
The use of explicit parallelism enables far more effective parallel execution of software instructions. In the new architecture
models, the compiler analyzes and explicitly identifies parallelism in the software at compile time. This allows the most optimal
structuring of the machine code to deliver the maximum performance before the processor executes it, rather than potentially
wasting valuable processor cycles at run time. The result is
significantly improved processor utilization. Also, there is no
wasting of precious die area for the hardware reorder engine
used in out-of-order reduced instruction set computer (RISC)
processors.
(2) Prediction. Simple decision structures, or code branches, are a
hard performance challenge to out-of-order RISC architectures.
In the simple if-then-else decision code sequence, traditional
architectures view the code in four basic blocks. In order to continuously feed instructions into the processor’s instruction pipeline, a technique called branch prediction is commonly used to
predict the correct path. With this technique, mispredicts commonly occur 5–10% of the time, causing the entire pipeline to be
purged and the correct path to be reloaded. A misprediction rate
of just 5–10% can slow processing speed as much as 30–40%.
To address this problem and to improve performance, the new
architectures use a technique known as prediction. Prediction
begins by assigning special flags called predicate registers to both
branch paths—p1 to “then” path and p2 to the “else” path. At run
time, the compare statement stores either a true or a false value
in the 1-bit predicate registers. The processor then executes both
paths but only the results from the path with a true predicate flag
Zhang_Ch02.indd 189
5/13/2008 5:53:42 PM
190
INDUSTRIAL CONTROL TECHNOLOGY
are used. Branches, and the possibility of associated mispredicts,
are removed, the pipeline remains full, and performance is
increased accordingly.
(3) Speculation. Memory latency is another big problem for current
processors’ architectures. Because memory speed is significantly
slower than processor speed, the processor must attempt to load
data from memory as early as possible to ensure that data is
available when needed. Traditional architectures allow compilers and processor to schedule loads before data is needed, but
branches act as barriers to this load hoisting.
These new architectures employ a technique known as “speculation” to initiate loads from memory earlier in the instruction
stream, even before a branch. Because a load can generate exceptions, a mechanism to ensure that exceptions are properly handled is needed to support speculation that hoists loads before
branches. The memory load is scheduled speculatively above the
branch in the instruction stream so as to start the memory access
as early as possible. If an exception occurs, this event is stored
and the “checks” instruction causes the exception to be processed.
The elevation of the load allows more time to account for memory latency, without stalling the processor pipeline. Branches
occur with great frequency in common software code sequences.
The unique ability of these architectures to schedule loads before
branches increases significantly the number of loads that can be
speculated relative to traditional architectures.
2.1.1
Microprocessor Unit Organization
The microprocessor plays a significant role in the functioning of industries everywhere. Nowadays, the microprocessor is being used in a wide
range of devices or systems as a digital data processing unit or a computing unit of an intelligent controller or a computer to control processes or
turn ON or OFF devices. The microprocessor is a multipurpose, programmable, clock-driven, register-based electronic device that reads binary
instructions from a storage device called memory, accepts binary data as
input and process data according to those instructions, and provides results
as output. At a very elementary level, an analogy can be drawn between
microprocessor operations and the functions of the human brain that process information according to understandings (instructions) stored in its
memory. The brain gets input from the eyes and ears and sends processed
information to output “devices,” such as the face with its capacity to register expression, the hands, or the feet.
Zhang_Ch02.indd 190
5/13/2008 5:53:42 PM
191
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
A typical programmable machine can be represented with five components: microprocessor, memory, input, output, and bus. These five components work together and interact with each other to perform a given
task; thus they comprise a system. The physical components are called
hardware. A set of instructions written for the microprocessor to perform
a task is called a program, and a group of programs is called a software.
Assume that a program and data are already entered in the memory; the
microprocessor executes a program by reading all the data including
instructions from the memory via the bus, and then processing these
data in terms of instructions, then writing the result into memory via
the bus.
2.1.1.1
Function Block Diagram of a Microprocessor Unit
Figure 2.1 depicts the function block diagram of the Intel486 GX processor, which gives the microarchitecture of this Intel processor. Figure 2.2
is the block diagram for the microarchitecture of the Intel Pentium-4
processor.
64-bit interunit transfer bus
32-bit data bus
32-bit data bus
32
Linear address
32
Core
clocks
32
PCD,PWT
Barrel shifter
Base/
index
bus
Register file
32
Segmentation
unit
Descriptor
registers
Decoded
Control ROM instruction
path
Prefetcher
Request
sequencer
32 byte code
queue
Burst Bus
control
2 × 16 bytes
Cache
control
32
Instruction
decode
24
Data bus
transceivers
Bus control
Displacement bus
Control and
protection test
unit
32
32
Code
stream
A2-A31
BEO#-BE3#
Write buffers
4 × 32
8 kbyte
cache
128
Microinstruction
CLK
Address
drivers
32
20
Physical
address
Translation
look aside
buffer
Limit and
attibute PLA
ALU
Bus interface
Cache unit
2
Paging
unit
Clock
control
Boundary scan
control
DO-D31
ADS# W/R# D/C# M/IO# PCD
PWT RDY# LOCK# PLOCK#
BOFF# A20M# BREQ HOLD
HLD ARESET SRE SET INTR
NMI SMI# SMIACT# STP CLK#
BRDY# BLAST#
KEN# FLUSH# AHOLD EADS#
TCK
TMS
TDI
TDO
Figure 2.1 The function block diagram of the Intel486 GX Processor (courtesy of
Intel Corporation).
Zhang_Ch02.indd 191
5/13/2008 5:53:42 PM
192
INDUSTRIAL CONTROL TECHNOLOGY
Figure 2.2 The function block diagram of the Intel Pentium-4 processor (courtesy
of Intel Corporation).
2.1.1.2
Microprocessor
The Pentium series of microprocessors made in Intel, up to now, includes
the Pentium Pro, the Pentium II, the Pentium III, and the Pentium 4, etc.
As given in Fig. 2.3(a), the Pentium Pro consists of the following basic
hardware elements:
(1) Intel Architecture registers. The Intel Architecture register set
implemented in the earlier 80x86 is extremely small. The small
number of registers permits the processor (and the programmer)
to keep only a small number of data operands close to the execution units where they can access them quickly. Rather, the programmer is frequently forced to write back the contents of one or
more of the processor’s registers to memory when he or she needs
to read additional data operands from memory to be operated on.
Later, when the programmer requires access to the original set of
data operands, they must again be read from memory. This juggling of data between the register set and memory takes time and
exacts a penalty on the performance of the program. Figure 2.3(b)
illustrates the Intel Architecture general register set.
(2) External bus unit. This unit performs bus transactions when
requested to do so by the L2 cache or the processor core.
(3) Backside bus unit. This unit interfaces the processor core to the
unified L2 cache.
Zhang_Ch02.indd 192
5/13/2008 5:53:42 PM
193
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
Unified L2 cache
(SRAM)
External bus
Backside
External bus unit
bus
Backside bus unit
L1 Code
cache
(SRAM)
Branch
prediction
L1 Data cache
(SRAM)
Prefetcher
APIC bus
Local
APIC
Address bus
Data bus
Instruction
decoder
Retire
unit
Execution unit
Instruction pool
(a)
31
23
15
8 7
0
EAX
AH
AX
AL
EBX
BH
BX
BL
ECX
CH
CX
CL
EDX
DH
DX
DL
EBP
BP
ESI
SI
EDI
DI
ESP
SP
(b)
Figure 2.3 Intel Pentium II processor: (a) simplified processor block diagram,
(b) Intel Architecture general register set (courtesy of Intel Corporation).
(4) Unified L2 cache. It services misses on the L1 data and code caches.
When necessary, it issues requests to the external bus unit.
(5) L1 data cache. It services data load and stores requests issued by
the load and store execution units. When a miss occurs, it forwards a request to the L2 cache.
Zhang_Ch02.indd 193
5/13/2008 5:53:43 PM
194
INDUSTRIAL CONTROL TECHNOLOGY
(6) L1 code cache. It services instruction fetch requests issued by the
instruction prefetcher.
(7) Processor core. The processor logic is responsible for the following: (1) instruction fetch, (2) branch prediction, (3) parsing of Intel
Architecture instruction stream, (4) decoding of Intel Architecture
instructions into RISC instructions that are referred to as microops or uops, (5) mapping accesses for Intel Architecture register
set to a large physical register set, (6) dispatch, execution, and
retirement of micro-ops.
(8) Local Advanced Programmable Interrupt Controller (APIC) unit.
The APIC is responsible for receiving interrupt requests from
other processors, the processor local interrupt pins, the APIC
timer, APIC error conditions, performance monitor logic, and
the IO APIC module. These requests are then prioritized and forwarded to the processor core for execution.
(a) Processor startup. Refer to Fig. 2.4. At startup, upon the
desertion of reset and the completion of a processor’s BIST
bit of the configuration register, the processors within the
cluster must negotiate amongst themselves to select the processor that will wake up and start fetching, decoding, and
executing the Power-On Self Test (POST) code from the
ROM. This processor is referred to as the BootStrap processor, or BSP. After the BSP is identified, the other processors,
referred to as the Application processors, or APs, remain
dormant until they receive a startup message from the BSP
via the APIC bus.
Processor cluster
APIC bus
CPU3
(BSP)
CPU2
(AP)
Pentium Pro
Host/PCI
bridge
(compatibility PB)
CPU1
(AP)
CPU0
(AP)
Bus
Main
DRAM
memory
Host/PCI
bridge
(auxiliary PB)
PCI bus
IO APIC
module
E/ISA bridge
E/ISA bus
Figure 2.4 Pentium multiprocessor system block diagram (courtesy of Intel
Corporation).
Zhang_Ch02.indd 194
5/13/2008 5:53:43 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
195
The Intel multiprocessing specification (available for
download at the Intel developers’ web site) dictates that the
startup code executed by the BSP is responsible for detecting
the presence of processors other than the BSP. When the
available APs have been detected, the startup code stores this
information as a table in nonvolatile memory. According to
the Intel multiprocessor specification (available for download at the Intel developers’ web site), both the BIOS code
and the POST code are responsible for detecting the presence of and initializing the APs. Intel recommends that this
be accomplished in the following manner:
(i) Both the POST code and the BIOS code executing on
the BSP initialize a predefined RAM location to 1hex (to
represent the fact that one processor, the BSP, is known
to be present and functioning). This location is referred
to as the Central Processing Unit (CPU) counter.
(ii) Both the POST code and the BIOS code executing
on the BSP clear a memory semaphore location to 00hex
to permit one of the APs to execute the body of the
FindAndInitAllCPU routine.
(iii) Both the POST code and the BIOS code executing on
the BSP broadcast a startup message to all APs (assuming any are present). The vector field in this message
selects a slot in the interrupt table that points to the
FindAndInitAllCPU routine.
(iv) Upon receipt of this message, all of the APs simultaneously request ownership of the Pentium Pro bus to
begin fetching and executing the FindAndInitAllCPU
routine.
(v) Both the POST code and the BIOS code executing on
the BSP then wait for all the APs that may be present
to complete execution of the FindAndInitAllCPU routine. The wait loop can be implemented using a long,
software enforced delay, the chipset’s Timer 0 (refer to
the Intel relevant specification available for download
on the Intel web site), or using the timer built into the
BSP’s local APIC. Alternatively, the following wait
procedure can be used: (1) Using a locked read, both
the POST code and the BIOS code executing on the
BSP examine the CPU counter RAM location every 2 s
and compare the counter value read 2 s ago. If the value
has not changed (i.e., been incremented), all APs have
completed their execution of the FindAndInitAllCPU
routine. This assumes that an AP takes less than 2 s to
Zhang_Ch02.indd 195
5/13/2008 5:53:43 PM
196
INDUSTRIAL CONTROL TECHNOLOGY
complete execution of routine. (2) The counter value
at the end of wait indicates the total number of processors in the system including the BSP.
(vi) Once all of the APs have completed execution of the
FindAndInitAllCPU routine, they have all made entries
in the Multiprocessor Table in the CMOS memory.
(vii) Both the POST code and the BIOS code executing on
the BSP read the user-selected setup parameters from
CMOS to determine how many of the available processors to utilize during this Operating System (OS)
session. Both the POST and the BIOS codes then
complete building the multiprocessor table in CMOS,
removing or disabling the entries associated with the
processors not to be used in this session. A new checksum value is computed for the adjusted multiprocessor
table and its length and number of entry fields are
completed.
(viii) As each of the APs complete execution of the
FindAndInitAllCPU routine, they either halt or enter a
program loop. Both the POST and the BIOS codes
instruct the BSP’s local APIC to broadcast an INIT
message to the APs. This causes them to enter (or
remain) in the halted state and await receipt of a startup
message that will be issued by the multiprocessor OS
once it has been loaded and control passed to it.
(b) The fetch, decode, execute engine. At the heart of the processor are the execution units that execute instructions. As given
in Fig. 2.3(a), the processor includes a fetch engine that
attempts to properly predict the path of program execution
and generates an ongoing series of memory-read operations
to fetch the desired instructions.
The high-speed (e.g., 150 or 200 MHz) processor execution engine would then be bound by the speed of external
memory accesses. It should be obvious that it is extremely
advantageous to include a very high-speed cache memory on
board so that the processor keeps copies of recently used
information of both code and data. Memory-read requests
generated by the processor core are first submitted to the
cache for a lookup before being propagated to the external
bus in an event of cache miss.
The Pentium processors include both a code and a data cache
in the level 1 cache. In addition, they include a level 2 cache
tightly coupled to the processor core via a private bus. The
processors’ caches are disabled at power-up time, however.
Zhang_Ch02.indd 196
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
197
In order to realize the processors’ full potential, the caches
must be enabled.
The steps for the Pentium processors to execute instructions are briefly described below:
(i) fetch Intel Architecture instructions from memory in
strict program order;
(ii) decode, or translate, them in strict program order into
one or more fixed-length RISC instructions known as
micro-ops or u-ops;
(iii) place the micro-ops into an instruction pool in strict
program order;
(iv) until this point, the instructions have been kept in original program order. This part of the pipeline is known as
the in-order front end. The processor then executes the
micro-ops in any order possible as the data and execution units required for each micro-op become available.
This is known as the out-of-order portion of pipeline;
(v) finally, the processor commits the results of each micro-op
execution to the processor’s register set in the order of
the original program flow. This is the in-order rear end.
The new Pentium processors implement a dynamic execution microarchitecture, a combination of multiple branch
prediction, speculation execution, and data flow analysis.
These Pentium processors execute MMX (will be detailed in
a subsequent paragraph) technology instructions for enhanced
media and communication performance.
Multiple branch predicts the flow of the program through
several branches: using a branch prediction algorithm, the
processor can anticipate jumps in instruction flow. It predicts
where the next instruction can be found in memory with a
90% or greater accuracy. This is made possible because,
while the processor is fetching instructions, it is also looking
at instructions further ahead in the program.
Data flow analysis analyzes and schedules instructions
to be executed in an optimal sequence, independent of the
original program order; the processor looks at decoded software instructions and determines whether f they are available for processing or whether they are dependent on other
instructions.
Speculative execution increases the rate of execution by
looking ahead of the program counter and executing instructions that are likely to be needed later. When the processor
executes several instructions at a time, it does so using
speculative execution. The instructions being processed are
Zhang_Ch02.indd 197
5/13/2008 5:53:44 PM
198
INDUSTRIAL CONTROL TECHNOLOGY
based on predicted branches and the results are stored as
speculative results. Once their final state can be determined,
the instructions are returned to their proper order and committed to permanent machine state.
(c) Processor cache. Figure 2.3(a) also provides an overview of
the processor’s cache, which shows that the processor cache
mainly contains these two types: data cache and code cache.
The L1 code cache services the requests for instructions generated by the instruction prefetcher (the prefetcher is the
only unit that accesses the code cache and it only reads from
it, so the code cache is read only), whereas the L1 data cache
services memory data read and write requests generated by
the processor’s execution units when they are executing any
instruction that requires a memory data access. The unified
L2 cache resides on a dedicated bus referred to as the backside bus. It services misses on the L1 caches, and, in the
event of an L2 miss, it issues a transaction request to the
external memory. The information is placed in the L2 cache
and is also forwarded to the appropriate L1 cache for
storage.
The L1 data cache in processor services memory data read
and writes requests initiated by the processor execution units.
The size and structure of the L1 data cache is processor
implementation–specific. As processor core speeds increase,
the cache sizes may also be increased because the faster core
can process code and data faster. Each of the data cache’s
cache banks, or ways, is further divided into two banks.
When performing a lookup, the data cache views memory as
divided into pages equal to the size of one of its cache banks
(or ways). Furthermore, it views each memory page as having the same structure as one of its cache ways. The target
number is used to index into the data cache directory and
select a set of two entries to compare against. If the target
page number matches the tag field in one of the entries in the
E, S, or M state given below, it is a cache hit. The data cache
has a copy of the target line from the target page. The action
taken by the data cache depends on whether or not the data
access is a read or a write, the current state of the line, and
the rules of conduct defined for this area of memory. Each
line storage location within the data cache can currently be
in one of four possible states:
(i) invalid state (I); there is no valid line in the entry.
(ii) exclusive state (E); the line in the entry is valid, is still
the same as memory, and no other processor has a copy
of the line in its caches.
Zhang_Ch02.indd 198
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
199
(iii) shared state (S); the line in the entry is valid, still the
same as memory, and one or more processors may also
have copies of the line or may not because the processor cannot discriminate between reads by processors
and reads performed by other, noncaching entries such
as a host/PCI bridge.
(iv) modified state (M); the line in the entry is valid, has
been updated by this processor since it was read into the
cache, and no other processor has a copy of the line in
its caches. The line in memory is stable.
The L1 code cache exists for only one reason: to supply requested code to the instruction prefetcher. The
prefetcher issues only read requests to the code cache,
so it is a read-only cache. A line stored in the code cache
can only be one of two possible states, valid or invalid,
implemented as the S and I states. When a line of code
is fetched from memory and is stored in the code cache,
it consists of raw code. The designers could have chosen to prescan the code stream as it is fetched from
memory and store boundary markers in the code cache
to demark the boundaries between instructions within
the cache line. This would preclude the need to scan the
code line as it enters the instruction pipeline for decode
so each of the variable-length Intel Architecture instructions can be aligned with the appropriate decoder.
However, this would bloat the size of the code cache.
Note the Pentium’s code cache stores boundary markers. When performing a lookup, the code cache views
memory as divided into pages equal to the size of one
of its cache banks (or ways). Furthermore, it views each
memory page as having the same structure as one of its
cache ways.
(d) MMX technology. Intel’s Matrix Math Extensions (MMX)
technology is designed to accelerate multimedia and communication applications. The MMX technology retains its
full compatibility with the original Pentium processor. It
contains five architectural design enhancements:
(i) New instructions
(ii) Single Instruction Multiple Data (SIMD). The new
instructions use a SIMD model, operating on several
values at a time. Using the 64-bit MMX registers, these
instructions can operate on eight bytes, four words, or two
double words at once, greatly increasing throughout.
(iii) More cache. Intel has doubled on-chip cache size to 32k.
That way, more instructions and data can be stored on
Zhang_Ch02.indd 199
5/13/2008 5:53:44 PM
200
INDUSTRIAL CONTROL TECHNOLOGY
the chip, reducing the number of times the processor has
to access slower, off-chip memory area for information.
(iv) Improved branch prediction. The MMX processor
contains four prefetch buffers that can hold up to four
successive code streams.
(v) Enhanced pipeline and deeper write buffers. An
additional pipeline stage has been added and four write
buffers are shared between the dual pipelines to improve
memory write performance.
MMX technology uses general-purpose basic instructions
that are fast and easily assigned to the parallel pipelines in
Intel processors. By using this general-purpose approach,
MMX technology provides performance that will scale
well across current and future generations of Intel processors. The MMX instructions cover several functional areas
including:
(i) basic arithmetic operations such as add, subtract, multiply arithmetic shift, and multiply add;
(ii) comparison operations;
(iii) conversation instructions to convert between the new
data types—pack data together and unpack from small
to larger data types;
(iv) logical operations such as AND, NOT, OR, and XOR;
(v) shift operations;
(vi) data transfer (MOV) instructions for MMX register-toregister transfers, or 64-bit and 32-bit load and store to
memory.
The principal data type of the MMX instruction set is the
packed, fixed-point integer, where multiple integer words
are grouped into single 64-bit quantities. These 64-bit quantities are moved to the 64-bit MMX registers. The decimal
point of the fixed-point values is implicit and is left for the
programmer to control for maximum flexibility. Arithmetic
and logical instructions are designed to support the different
packed integer data types. These instructions have a different
op code for each data type supported. As a result, the new
MMX technology instructions are implemented with 57 op
codes. The supported data types are signed and unsigned
fixed-point integers, bytes, words, double words, and quad
words. The four MMX technology data types are
(i) packed bytes: 8 bytes packed into one 64-bit quantity;
(ii) packed word: four 16-bit words packed into one 64-bit
quantity;
Zhang_Ch02.indd 200
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
201
(iii) packed double word: two 32-bit double words packed
into one 64-bit quantity;
(iv) quad word: one 64-bit quantity.
From the programmer’s view, there are eight new MMX registers
(MM0–MM7) along with new instructions that operate on these registers.
But to avoid adding new states, these registers are mapped onto the existing floating-point registers (FP0–FP7). When a multitasking operating
system (or application) executes an FSAVE instruction, as it does today to
save state, the contents of MM0–MM7 are saved in place of FP0–FP7 if
MMX instructions are in use.
Detecting the existence of MMX technology on an Intel microprocessor
is done by executing the CPUID instruction and checking a set bit. Therefore,
when installing or running, the software can query the microprocessor to
determine whether MMX technology is supported and install or execute
the code that includes, or does not include, MMX instructions based on the
result.
2.1.1.3
Internal Bus System
Figures 2.1, 2.3(a), and 2.4 show us that the internal bus systems of an
Intel microprocessor, based on their functions, comprises these types
below, each of them monitored by a corresponding bus controller or bus
unit:
(1) Backside bus.
(2) Displacement bus.
(3) APIC bus.
(4) Cache buses that are divided into data bus and address bus.
(5) CPUs’ cluster bus.
2.1.1.4
Memories
Memory can be classified into two groups: prime (system or main)
memory and storage memory. The R/WM and ROM are examples of prime
memory; this is the memory the microprocessor uses in executing and
storing programs. This memory should be able to respond fast enough to
keep up with the execution speed of the microprocessor. Therefore,
it should be random access memory, meaning that the microprocessor
should be able to access information from any register with the same speed
Zhang_Ch02.indd 201
5/13/2008 5:53:44 PM
202
INDUSTRIAL CONTROL TECHNOLOGY
(independent of its place in the chip). The size of a memory chip is specified
in terms of bits. For example, a 1k memory chip means it can store 1k (1024)
bits (not bytes). On the other hand, memory in a system such as a PC computer is specified in bytes. For example, 4M memory means it has 4 megabytes of volume. The other group is the storage memory, such as magnetic
disks and tapes (see Fig. 2.5). This memory is used to store programs and
results after the completion of program execution. Information stored in
these memories is nonvolatile, meaning information remains intact even if
the system is turned off. The microprocessor cannot directly execute or
process programs stored in these devices; programs need to be copied into
the R/W prime memory first. Therefore, the size of the prime memory,
such as 512k or 8M (megabytes), determines how large a program the
system can process. The size of the storage memory is unlimited; when
one disk is full, the next one can be used.
Figure 2.5 also shows two groups in storage memory: secondary storage
and backup storage. The secondary storage and backup storage include
devices such as disks, magnetic tapes, etc. Figure 2.5 shows that the prime
(system) memory is divided into two main groups: read/write memory
(R/WM) and read-only memory (ROM); each group includes several different types of memory, as discussed below.
(1) Read/write memory (R/WM). As the name suggests, the microprocessor can write into or read from this memory; it is popularly
Memory
Prime memory
Storage memory
Readonly
memory
ROM
Read/write
memory
R/WM
Static R/WM
Zero power
RAM
Nonvolatile
RAM
Dynamic
R/WM
Integrated
RAM
Secondary
storage
Semi-random
access
Erasable
memory
Permanent
memory
EPROM
PROM
EE-PROM
Masked
ROM
Flash
memory
Disk
Backup
storage
Serial
access
Magnetic
tape
CCD
Floppy
CD-ROM; Zip disk
Figure 2.5 The classification of microprocessors’ memories.
Zhang_Ch02.indd 202
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
203
known as Random Access Memory (RAM). It is used primarily
for information that is likely to be altered, such as writing programs or receiving data. This memory is volatile, meaning that
when the power is turned off, all the contents are destroyed. Two
types of R/W memories, static and dynamic, are available; they
are described in the following paragraphs.
(a) Static memory (SRAM). This memory is made up of flipflops, and it stores the bit as a voltage. Each memory cell
requires six transistors; therefore, the memory chip has low
density but high speed. SRAM, known as cache memory, is
included on the processor chip. In addition, high-speed cache
memory is also included external to the processor to improve
the performance of a system.
(b) Dynamic memory (DRAM). This memory is made up of
MOS transistor gates, and it stores the bit as a charge. For the
DRAM, stored information needs to be read and then written
again every few milliseconds. It is generally economical to
use dynamic memory when system memory is at least 8k;
for small systems, the static memory is appropriate. To
increase the speed of DRAM, various techniques are being
used. These techniques have resulted in the production of
high-speed memory chips, such as Extended Data Out
(EDO), Synchronous DRAM (SDRAM), and Rambus
DRAM (RDRAM).
(2) Read-only memory (ROM). The ROM is a nonvolatile memory;
it retains stored information even if the power is turned off. This
memory is used for programs and data that need not be altered.
As the name suggests, the information can be read only, which
means once a bit pattern is stored, it is permanent or at least
semi-permanent. The permanent group also includes two types
of memory: masked ROM and PROM. The semi-permanent group
also includes two types of memory: EPROM and EE-PROM, as
shown in Fig. 2.5. Five types of ROM—masked ROM, PROM,
EPROM, EE-PROM, and flash memory—are described in the
following paragraphs.
(a) Masked ROM. In this ROM, a bit pattern is permanently
recorded by the masking and metallization process. Memory
manufacturers are generally equipped to do this process. It is
an expensive and specialized process, but economical for
large production quantities.
(b) Programmable read-only memory (PROM). This memory
has nichrome or polysilicon wires arranged in a matrix; these
wires can be functionally viewed as diodes or fuses. This
memory can be programmed by the user with a special PROM
Zhang_Ch02.indd 203
5/13/2008 5:53:44 PM
204
INDUSTRIAL CONTROL TECHNOLOGY
programmer that selectively burns the fuses according to the
bit pattern to be stored. The process is known as “burning the
PROM,” and the information stored is permanent.
(c) Erasable programmable read-only memory (EPROM). This
memory stores a bit by charging the floating gate of an FET.
Information is stored by using an EPROM programmer,
which applies high voltages to charge the gate. All the information can be erased by exposing the chip to ultraviolet
light through its quartz window, and the chip can be reprogrammed. Because the chip can be reused many times, this
memory is ideally suited for product development and experimental projects. The disadvantages of EPROM are (1) it
must be taken out of the circuit to erase it, (2) the entire chip
must be erased, and (3) the erasing process could take 15
or 20 min.
(d) Electrically erasable PROM (EE-PROM). This memory is
functionally similar to EPROM, except that information can
be altered by using electrical signals at the register level
rather than erasing all the information. This has an advantage in field and remote control applications. In microprocessor systems, software update is a common occurrence. If
EE-PROMs are used in the systems, they can be updated
from a central computer by using a remote linkage via serial
cable bus. This memory also includes Chip Erase mode,
whereby the entire chip can be erased in 10 ms rather than
the 20 min taken to erase an EPROM.
(e) Flash memory. This is a variation of EE-PROM that is
becoming popular. The major difference between the flash
memory and EE-PROM is in the erasure procedure: The
EE-PROM can be erased at a register level, but the flash
memory chip must be erased either in its entirety or at the
sector (block) level. These memory chips can be erased and
programmed at least a million times.
In a microprocessor-based device, programs are generally written in
ROM, and data that are likely to vary are stored in R/WM. Memory technology has advanced considerably in recent years. In addition to static
and dynamic R/W memory, other options are also available in memory
devices. Examples include zero power RAM, nonvolatile RAM, and integrated RAM.
The zero power RAM is a CMOS read/write memory with battery backup
built internally. It includes lithium cells and voltage-sensing circuitry.
Zhang_Ch02.indd 204
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
205
When the external power supply voltage falls below 3 V, the power-switching
circuitry connects the lithium battery; thus, this memory provides the
advantages of R/W and read-only memory.
The nonvolatile RAM is a high-speed static R/W memory array backed
up, bit for bit, by EE-PROM array for nonvolatile storage. When the power
is about to go off, the contents of R/W memory are quickly stored in the
EE-PROM by activating the store signal or the memory chip, and stored
data can be read into the R/W memory segment when the power is again
turned on. This memory chip combines the flexibility of static R/W memory
with the nonvolatility of EE-PROM.
The integrated RAM (iRAM) is a dynamic memory with the refreshed
circuitry built on a chip. For the user, it is similar to the static R/W memory.
The user can derive the advantages of dynamic memory without having to
build the external refresh circuitry.
2.1.1.5
Input/Output Pins
To allow for easy upgrades and to save space, the 80486 and Pentium
processors are available in a pin-grid array (PGA) form. For all the Intel
microprocessors, their PGA pin-out lists are provided in the corresponding
Intel specifications. A 168-pin 80486 GX block is illustrated in Fig. 2.1; it
can be seen that the 80486 processor has a 32-bit address bus (A0-A31)
and a 32-bit data bus (D0-D31). Table 2.1 defines how the 80486 control
signals are interpreted.
Table 2.1 Intel 80486 Processor Control Signal
M/IO
D/C
W/R
Description
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
Interrupt acknowledge sequence
STOP/special bus cycle
Reading from an I/O port
Writing to an I/O port
Reading an instruction from memory
Reserved
Reading data from memory
Writing data to memory
Zhang_Ch02.indd 205
5/13/2008 5:53:44 PM
206
INDUSTRIAL CONTROL TECHNOLOGY
The main 80486 pin connections are as follows:
(1) A2-A31 (I/O)
(2) A20M (I)
(3) ADS (O)
(4) AHOLD (I)
(5) BE0–BE3 (O)
(6) BLAST (O)
(7) BOFF (I)
(8) BRDY (I)
(9) BREQ (O)
(10) BS16, BS8 (I)
(11) DP0–DP3 (I/O)
(12) EADS (I)
(13) FERR (O)
(14) FLUSH (I)
(15) HOLD, HOLDA (I/O)
(16) IGNNE (I)
(17) INTR (I)
(18) KEN (I)
Zhang_Ch02.indd 206
The 30 most significant bits of the address bus
When active low, the processor internally
masks the address bit A20 before every
memory access
Indicates that the processor has valid control
signals and valid address signals
When active, a different bus controller can have
access to the address bus. This is typically used
in a multiprocessor system
The byte enable lines indicate which of the
bytes of the 32-bit data bus is active
It indicates that the current burst cycle will end
after the next BRDY signal
The backoff signal informs the processor to
deactivate the bus on the next clock cycle
The burst ready signal is used by an addressed
system that has sent data on the data bus or
read data from the bus
It indicates that the processor has internally
requested the bus
The BS16 signal indicates that a 16-bit data
bus is used; the BS8 signal indicates that an
8-bit data bus is used. If both are high, then a
32-bit data bus is used
The data parity bits give a parity check for
each byte of the 32-bit data bus. The parity bits
are always even parity
Indicates that an external bus controller has put
a valid address on the address bus
Indicates that the processor has detected an
error in the internal floating-point unit
When it is active the processor writes the
complete contents of the cache to memory
The bus hold (HOLD) and acknowledge
(HOLDA) are used for bus arbitration and allow
other bus controllers to take control of the buses
When active the processor ignores any numeric
errors
External devices to interrupt the processor use
the interrupt request line
This signal stops caching of a specific address
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
(19) LOCK (O)
(20) M/IO, D/C, W/R (O)
(21) NMI (I)
(22) PCHK (O)
(23) PLOCK (O)
(24) PWT, PCD (O)
(25) RDY (I)
(26) RESET (I)
2.1.1.6
207
If it is active, the processor will not pass
control to an external bus controller when it
receives a HOLD signal
See Table 2.1
The nonmaskable interrupt signal causes an
interrupt 2
If it is set active then a data parity error has
occurred
The active pseudo lock signal identifies that
the current data transfer requires more than
one bus cycle
The page write-through (PWT) and page cache
disable (PCD) are used with cache control
When it is active, the addressed system has sent
data on the data bus or read data from the bus
If the reset signal is high for more than 15
clock cycles, then the processor will reset itself
Interrupt System
In the Intel microprocessors, the interrupt lines are interrupt request pin
(INTR), nonmaskable interrupt request pin (NMI), and system reset pin
(RESET), all of which are high signals. The INTR pin is activated when
an external device, such as a hard disk or a serial port, wishes to communicate with the processor. This interrupt is maskable and the processor
can ignore the interrupt if it wants. NMI pin is a nonmaskable interrupt
and is always acted on. When it becomes active the processor calls the
nonmaskable interrupt service routine. The RESET pin signal causes a
hardware reset and is normally made active when the processor is powered up.
2.1.2
Microprocessor Unit Interrupt Operations
The interrupt I/O is a process of data transfer whereby an external device
or a peripheral can inform the processor that it is ready for communication
and it requests attention. The process is initiated by an external device and
is asynchronous, meaning that it can be initiated at any time without reference to the system clock. However, the response to an interrupt request is
directed or controlled by the microprocessor. Unlike the polling technique,
an interrupt processing allows a program or an external device to interrupt
the task currently being executed by the microprocessor. The generation of
Zhang_Ch02.indd 207
5/13/2008 5:53:44 PM
208
INDUSTRIAL CONTROL TECHNOLOGY
an interrupt can occur by hardware (hardware interrupt) or by software
(software interrupt). When an interrupt occurs an interrupt service routine
(ISR) is called. For a hardware interrupt, the ISR then communicates with
the device and processes data. When it has finished the program execution,
it then returns to the original program. A software interrupt causes the
program to interrupt its execution and goes to an ISR. Software interrupts
include the processor-generated interrupts normally occurring either when
a program causes a certain type of error or if it is being used in a debug
mode. In debug mode the program can be made to break from its execution
when a breakpoint occurs. It seems that software interrupts, in most cases,
do not require the program to return back when the ISR task is complete.
Apart from this difference between them, both the software interrupts and
the hardware interrupts use the same mechanisms, methodologies, and
processes to handle interrupts.
The interrupt requests are classified in two categories: maskable interrupt and nonmaskable interrupt. The microprocessor can ignore or delay a
maskable interrupt request if it is performing some critical task; however,
it must respond to a nonmaskable interrupt immediately.
2.1.2.1
Interrupt Process
(1) The operation of a real mode interrupt. When the microprocessor completes executing the current instruction, it determines
whether an interrupt is active by checking the following:
(1) instruction executions, (2) single step, (3) NMI pin, (4) coprocessor segment overrun, (5) INTR pin, and (6) INT instruction,
in the order presented. If one or more of these interrupt conditions are present, the following sequence of events occurs:
(a) The contents of the flag register are pushed onto the stack.
(b) Both the interrupt (IF) and trap (TF) flags are cleared. This
disables the INTR pin and the trap or single-step feature.
(c) The contents of the code segment register (CS) are pushed
onto the stack.
(d) The contents of the instruction pointer (IP) are pushed onto
the stack.
(e) The interrupt vector contents are fetched and then placed
into both IP and CS so that the next instruction executes the
ISR addressed by the vector.
Whenever an interrupt is accepted, the microprocessor
stacks the contents of the flag register, CS and IP; clears both
IF and TF; and jumps to the procedure addressed by the
interrupt vector. After the flags are pushed onto the stack,
Zhang_Ch02.indd 208
5/13/2008 5:53:44 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
209
IF and TF are cleared. These flags are returned to the state prior
to the interrupt when the IRET instruction is encountered at
the end of the ISR. Therefore, if interrupts were enabled
prior to the ISR, they are automatically reenabled by the
IRET instruction at the end of the interrupt service routine.
The return address (stored in CS and IP) is pushed onto
the stack during the interrupt. Sometimes, the return address
points to the next instruction in the program; sometimes it
points to the instruction or point in the program where the
interrupt occurred. Interrupt type numbers 0, 5, 6, 7, 8, 10,
11, 12, and 13 push a return address that points to the offending instruction, instead of to the next instruction in the program. This allows the ISR to possibly retry the instruction
crashed in certain error cases. Some of the protected mode
interrupts (type 8, 10, 11, 12, and 13) place an error code on
the stack following the return address. The error code identifies the selector that caused the interrupt. In case no selector
is involved, the error code is 0.
(2) The operation of a protected mode interrupt. In the protected
mode, interrupts have exactly the same assignments as in the real
mode, but the interrupt vector table is different. In place of interrupt vectors, protected mode uses a set of 256 interrupt descriptors that are stored in an interrupt descriptor table (IDT). The
interrupt descriptor table is normally 256 × 8 (2k) bytes long,
with each descriptor containing 8 bytes. The IDT is located at
any memory location in the system by the IDT address register
(IDTR). Each entry in the IDT contains the address of the ISR in
the form of a segment selector and a 32-bit offset address. It also
contains the P bit (present) and DPL bits to describe the privilege
level of the interrupt.
Real mode interrupt vectors can be converted into protected
mode interrupts by copying the interrupt procedure addresses
from the interrupt vector table and converting them to 32-bit offset addresses that are stored in the interrupt descriptors. A single
selector and segment descriptor can be placed in the global
descriptor table that identifies the first 1M byte of memory as the
interrupt segment.
Other than the IDT and interrupt descriptors, the protected
mode interrupt functions like the real mode interrupt. They return
from both interrupts by using the IRET or IRETD instruction.
The only difference is that in protected mode the microprocessor
accesses the IDT instead of the interrupt vector table.
(3) Interrupt flag bits. The interrupt flag (IF) and trap flag (TF) are
both cleared after the contents of the flag register are stacked
Zhang_Ch02.indd 209
5/13/2008 5:53:45 PM
210
INDUSTRIAL CONTROL TECHNOLOGY
during an interrupt. When the IF bit is set, it allows the INTR pin
to cause an interrupt; when the IF bit is cleared, it prevents the
INTR pin from causing an interrupt. When IF = 1, it causes a trap
interrupt (interrupt type number 1) to occur after each instruction
executes. This is why we often call trap a single step. When TF = 0,
normal program execution occurs. The interrupt flag is set and
cleared by the STI and CLI instructions, respectively. There are
no special instructions that set or clear the trap flag.
2.1.2.2
Interrupt Vectors
The interrupt vectors and vector table are crucial to the understanding of
hardware and software interrupts. Interrupt vectors are addresses that
inform the interrupt handler as to where to find the ISR (also called interrupt service procedure). All interrupts are assigned a number from 0 to
255, with each of these interrupts being associated with a specific interrupt
vector.
The interrupt vector table is normally located in the first 1024 bytes
of memory at addresses 000000H–0003FFH. It contains 256 different
interrupt vectors. Each vector is 4 bytes long and contains the starting
address of the ISR. This starting address consists of segment and offset
of the ISR. Figure 2.6 illustrates the interrupt vector table used for the
Intel microprocessors. Remember that in order to install an interrupt
vector (sometimes called a hook), the assembler must address absolute
memory.
In an interrupt vector table, the first five interrupt vectors are identical in
all Intel microprocessor family members, from the 8086 to the Pentium.
Other interrupt vectors exist for the 80286 that are upward-compatible to
80386, 80486, and Pentium to Pentium 4, but not downward-compatible to
the 8086 or 8088. Intel reserves the first 32 interrupt vectors for its use in
various microprocessor family members. The last 224 vectors are available as user interrupt vectors.
2.1.2.3
Interrupts Service Routine (ISR)
The interrupts of the entire Intel family of microprocessors include two
hardware pins that request interrupts (INTR pin and NMI pin), and one
hardware pin (INTA) that acknowledges the interrupt requested through
INTR. In addition to the pins, the Intel microprocessor also has software
interrupt instructions: INT, INTO, INT 3, and BOUND. Two flag bits, IF
(interrupt flag) and TF (trap flag), are also used with the interrupt structure
Zhang_Ch02.indd 210
5/13/2008 5:53:45 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
0003FFH
211
Type 32 – 255:
User interrupt vectors
Type 14 – 31:
Reserved
Type 18:
Machine check
Type 17:
Alignment check
Type 16:
Coprocessor error
Type 15:
Unassigned
Type 14:
Page fault
Type 13:
General protection
Type 12:
Stack segment overrun
Type 11:
Segment not present
Type 10:
Invalid task state segment
Type 9:
Coprocessor segment overrun
Type 8:
Double fault
Type 7:
Coprocessor not available
Type 6:
Undefined opcode
Type 5:
BOUND
Type 4:
Overflow (INTO)
Type 3:
1-byte breakpoint
Type 2:
NMI pin
Type 1:
Single-step
Type 0:
Divide error
000000H
(a)
Segment (low)
Offset (high)
Offset (low)
(b)
Figure 2.6 (a) The interrupt vector table for the Intel microprocessor, and (b) the
contents of an interrupt vector.
Zhang_Ch02.indd 211
5/13/2008 5:53:45 PM
212
INDUSTRIAL CONTROL TECHNOLOGY
and with a special return instruction IRET (or IRETD in the 80386, 80486,
or Pentium-Pentium 4).
(1) Software interrupts. Intel microprocessors provide five software
interrupt instructions: BOUND, INTO, INT, INT 3, and IRET.
Of these five software interrupt instructions, INT and INT 3 are
very similar, BOUND and INTO are conditional, and IRET is a
special interrupt return instruction.
The INT n instruction calls the ISR that begins at the address
represented in vector number n. The only exception to this is the
“INT 3” instruction, a 1-byte instruction. The INT 3 instruction
is often used as breakpoint-interrupt, because it is easy to insert
a 1-byte instruction into a program. As mentioned previously,
breakpoints are often used to debug faulty software.
The BOUND instruction, which has two operands, compares
a register with two words of memory data. The INTO instruction
checks the overflow flag (OF); If OF = 1, the INTO instruction
calls the ISR whose address is stored in interrupt vector type
number 4. If OF = 0, then the INTO instruction performs no
operation and the next sequential instruction in the program
executes.
The IRET instruction is a special return instruction used to
return for both software and hardware interrupts. The IRET
instruction is much like a “far RET” because it retrieves the
return address from the stack. It is unlike the “near return”
because it also retrieves a copy of the flag register from the stack.
An IRET instruction removes six bytes from the stack: two for
the IP, two for CS, and two for flags. In the 80386 to Pentium 4,
there is also an IRETD instruction because these microprocessors can push the EFLAG register (32 bit) on the stack, as well as
the 32-bit EIP in the protected mode. If operated in the real mode,
we use the IRET instruction with the 80386 to Pentium 4
microprocessors.
(2) Hardware interrupts. The microprocessor has two hardware
inputs: nonmaskable interrupt (NMI) and interrupt request
(INTR). Whenever the NMI input is activated, a type 2 interrupt
occurs because NMI is internally decoded. The INTR input must
be externally decoded to select a vector. Any interrupt vector can
be chosen for the INTR pin, but we usually use an interrupt type
number between 20H and FFH. Intel has reserved interrupts 00H
through 1FH for internal and future expansion. The INTA signal
is also an interrupt pin on the microprocessor, but it is an output
that is used in response to the INTR input to apply a vector-type
number to the data bus connections D7–D0.
Zhang_Ch02.indd 212
5/13/2008 5:53:45 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
213
The NMI is an edge-triggered input that requests an interrupt
on the positive edge (0-to-1 transition). After a positive edge, the
NMI pin must remain logic 1 until it is recognized by the microprocessor. The NMI input is often used for parity errors and other
major system faults, such as power failure. Power failures are
easily detected by monitoring the AC power line and causing an
NMI interrupt whenever AC power drops out.
The interrupt request input (INTR) is level sensitive, which
means that it must be held at logic 1 level until it is recognized.
The INTR pin is set by an external event and cleared inside the
ISR. This input is automatically disabled once it is accepted by
the microprocessor and reenabled by the IRET instruction at the
end of the ISR. The microprocessor responds to the INTR input
by pulsing the INTA output in anticipation of receiving an interrupt vector-type number on data bus connection D7–D0. There
are two INTA pulses generated by the system that are used to
insert the vector-type number on the data bus.
2.1.3
Microprocessor Unit Input/Output Rationale
The I/O devices can be interfaced with a microprocessor using both techniques: isolated I/O (also called peripheral-mapped I/O) and memorymapped I/O. The process of data transfer in both is identical. Each device is
assigned a binary address, called a device address or port number, through
its interface circuit. When the microprocessor executes a data transfer
instruction for an I/O device, it places the appropriate address on the address
bus, sends the control signals, enables the interfacing device, and then
transfers data. The interface device is like a gate for data bits, which is
opened by the microprocessor whenever it intends to transfer data.
2.1.3.1
Basic Input/Output Techniques
As previously mentioned, there are two main methods of communicating with external equipment: either the equipment is mapped into the
physical memory and given a real address on the address bus of the microprocessor (memory mapped I/O), or it is mapped into a special area of
input/output memory (isolated I/O). Devices mapped into memory are
accessed by reading or writing to the physical address of the memory.
Isolated I/O provides ports that are gateways between the interface device
and the processor. They are isolated from the system using a buffering
system and are accessed by four machine code instructions: IN, INS,
OUT, OUTS. The IN (INS) instruction inputs a byte, or a word, and the
Zhang_Ch02.indd 213
5/13/2008 5:53:45 PM
214
INDUSTRIAL CONTROL TECHNOLOGY
OUT (OUTS) instruction outputs a byte, or a word. A high-level compiler
interprets the equivalent high-level functions and produces machine code
that uses these instructions.
Figure 2.7 shows the two methods. This figure also tells us that devices
are not directly connected onto the address and data bus because they may
use part of the memory that a program uses or they could cause a hardware
fault. This device interprets the microprocessor signals and generates the
required memory signals. Two main output lines differentiate between a
read and a write operation (R/W) and between direct and isolated memory access (M/IO). The R/W line is low when data is being written to
memory and high when data is being read. When M/IO is high, direct memory access is selected, and when low, the isolated memory is selected.
(1) Isolated I/O. The most common I/O transfer technique used in
the Intel microprocessor-based system is isolated I/O. The term
“isolated” describes how the I/O locations are isolated from the
memory system in a separate I/O address space. The addresses
for isolated I/O devices, called ports, are separate from the
memory. Because the ports are separate, the user can expand the
memory to its full size without using any of the memory space
for I/O devices. A disadvantage of isolated I/O is that the data
transferred between I/O and the microprocessor must be accessed
Bus controller
Read/write
R/W
Memory/isolated
M/IO
Address bus
Microprocessor
Data bus
Memory
mapped
I/O
Interface
device
Address bus
Data bus
Isolated
I/O
Figure 2.7 Access memory-mapped and isolated I/O.
Zhang_Ch02.indd 214
5/13/2008 5:53:45 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
215
by the IN, INS, OUT, and OUTS instructions. Separate control
signals for the I/O space are developed (using M/IO and R/W),
which indicate an I/O read (IORC) or an I/O write (IOWC) operation. These signals indicate that an I/O port address, which
appears on the address bus, is used to select the I/O device. In the
personal computer, isolated I/O ports are used for controlling
peripheral devices such as direct memory access (DMA) controller, NMI reset, game I/O adaptor, floppy disk controller, second
serial port (COM2), and primary serial port (COM1). An 8-bit
port address is used to access devices located on the system
board, such as the timer and keyboard interface, while a 16-bit
port is used to access serial and parallel ports as well as video
and disk drive system.
(2) Memory-mapped I/O. Interface devices can map directly onto
the system address and data bus. Unlike isolated I/O, memorymapped I/O does not use the IN, INS, OUT, or OUTS instructions. Instead, it uses any instruction that transfers data between
the microprocessor and memory. A memory-mapped I/O device
is treated as a memory location in the memory map. The main
advantage of memory-mapped I/O is that any memory transfer
instruction can be used to access the I/O device. The main disadvantage is that a portion of the memory system is used as the I/O
map, which reduces the amount of the usable memory volumes
for applications.
In a PC-compatible system the address bus is 20 bits wide,
from address 00000h to FFFFFh (1MB). Figure 2.8 gives a typical memory allocation in PC.
FFFFFFFFh (4 GB)
Extended
memory
00FFFFFFh (16 MB)
Extended
memory
Video
graphic text
display
Application
programs
(640 kB)
Interrupt
vectors
BIOS
000FFFFFh (1 MB)
0009FFFFh (640 kB)
00000600h
00000000h
Figure 2.8 Typical PC memory map.
Zhang_Ch02.indd 215
5/13/2008 5:53:45 PM
216
INDUSTRIAL CONTROL TECHNOLOGY
2.1.3.2
Basic Input/Output Interfaces
The basic input device is a set of three-state buffers. The basic output
device is a set of data latches. The term IN refers to moving data from the
I/O device into the microprocessor and the term OUT refers to moving
data out of the microprocessor to the I/O device.
Many I/O devices accept or release information at a much slower rate
than the microprocessor. Another method of I/O control, called “handshaking” or “polling,” synchronizes the I/O device with the microprocessor. An example device that requires handshaking is a parallel printer that
prints 100 characters per second (CPS). It is obvious that the microprocessor can send more than 100 CPS to the printer, so a handshaking must be
used to slow the microprocessor down to match speeds with the printer.
(1) The basic input interface. Three-state buffers 74ALS244 are
used to construct the 8-bit input port depicted in Fig. 2.9(a). The
external TTL data (simple toggle switches in this example) are
connected to the inputs of the buffers. The outputs of the buffers
connect to the data bus. The exact data bus connections depend
on the version of the microprocessor. For example, the 8088 has
data bus connections D7–D0, the 80486 has D31–D0, and the
Pentium to Pentium 4 have D63–D0. The circuit of Fig. 2.9(a)
allows the microprocessor to read the contents of the eight
switches that connect to any 8-bit section of the data bus when
the select signal SEL becomes logic 0. Thus, whenever the IN
instruction executes, the contents of the switches are copied into
the AL register.
When the microprocessor executes an IN instruction, the I/O
port address is decoded to generate the logic 0 on SEL. A 0
placed on the output control inputs (1G and 2G) of the 74ALS244
buffer causes the data input connections (A) to be connected to
the data input (Y) connections. If a logic 1 is placed on the output
control inputs of the 74ALS244 buffer, the device enters the
three-state high-impedance mode that effectively disconnects the
switches from the data bus.
The basic input circuit is not optional and must appear any
time that input data are interfaced to the microprocessor.
Sometimes it appears as a discrete part of the circuit, as shown in
Fig. 2.9(a); sometimes it is built into a programmable I/O
device.
It is possible to interface 16- or 32-bit data to various versions
of the microprocessor, but this is not nearly as common as using
8-bit data. To interface 16 bits of data, the circuit in Fig. 2.9(a) is
doubled to include two 74ALS244 buffers that connect 16 bits of
Zhang_Ch02.indd 216
5/13/2008 5:53:46 PM
217
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
10 k
1 2
1
16
3
4 5 6 7
8
2
2
15
4
3
4
5
14
13
12
6
8
11
6
7
8
11
10
9
13
15
17
1
19
1A1
!A2
1Y1
1Y2
1A3
1A4
2A!
1Y3
1Y4
2Y1
2A2
2A3
2A4
2Y2
2Y3
2Y4
Data
bus
18
16
14
12
9
7
5
3
1G
2G
74ALS244
.
SEL
(a)
Data
bus
D0
D1
D2
Q0
Q1
Q2
D3
D4
D5
Q3
Q4
Q5
D6
D7
Q6
Q7
OC
CLK
SEL
(b)
Figure 2.9 The basic input and output interfaces. (a) The basic input interface
illustrating the connection of eight switches. Note that the 74ALS244 is a three
state that controls the application of switch data to the data bus. (b) The basic
output interface connected to a set of LED displays.
Zhang_Ch02.indd 217
5/13/2008 5:53:46 PM
218
INDUSTRIAL CONTROL TECHNOLOGY
input data to the 16-bit data bus. To interface 32 bits of data, the
circuit is expanded by a factor of 4.
(2) The basic output interface. The basic output interface receives
data from the microprocessor and must usually hold it for some
external device. Its latches or flip-flops, like the buffers found in
the input device, are often built into the I/O device.
Figure 2.9(b) shows how eight simple light-emitting diodes
(LEDs) connect to the microprocessor through a set of eight data
latches. The latch stores the number output by the microprocessor from the data bus so that the LED can be lit with any 8-bit
binary number. Latches are needed to hold the data because when
the microprocessor executes an OUT instruction, the data are
only present on the data bus for less than 1.0 µs. Without a latch,
the viewer would never see the LED illuminate.
When the OUT instruction executes, the data from AL, AX, or
EAX are transferred to the latch via the data bus. Here, the D
inputs of a 74ALS374 octal latch are connected to the data bus to
capture the output data, and the Q outputs of the latch are attached
to the LED. When a Q becomes a logic 0, the LED lights. Each
time that the OUT instruction executes, the SEL signal to the
latch activates, capturing the data output to the latch from any
8-bit section of the data bus. The data are held until the next OUT
instruction executes. Thus, whenever the output instruction is
executed in this circuit, the data from the AL register appears on
the LED.
2.1.4
Microprocessor Unit Bus System Operations
This subsection uses the Peripheral Component Interconnect (PCI) bus
to introduce the microprocessor unit bus system operations. The PCI bus
has been developed by Intel for its Pentium processors. This technique can
be populated with adaptors requiring fast accesses to each other and/or
system memory and that can be accessed by the processor at speeds
approaching that of the processor’s full native bus speed. A PCI physical
device package may take the form of a component integrated onto the system board or may be implemented on a PCI add-in card. Each PCI package (referred to in the specification as a device) may incorporate from one
to eight separate functions. A function is a logical device. Each function
contains its own, individually addressable configuration space, 64 double
words in size. Its configuration registers are implemented in this space.
Using these registers, the configuration software can automatically detect
the presence of a function, determine its resource requirements including
memory space, I/O space, interrupt lines, etc., and can then assign resources
Zhang_Ch02.indd 218
5/13/2008 5:53:46 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
219
to the function that are guaranteed not to conflict with the resources
assigned to other devices.
2.1.4.1
Bus Operations
The PCI bus operates in the multiplexing mode (also called normal
mode) and/or in the burst mode. In multiplexing mode, the address and
data lines are used alternatively. First, the address is sent, followed by a
data read or write. Unfortunately, this mode requires two or three clock
cycles for a single transfer that is an address followed by a read or write
cycle. The multiplex mode obviously slows down the maximum transfer
rate. Additionally, a PCI bus can be operated in burst mode. A burst transfer is one consisting of a single address phase followed by two or more
data phases. In the burst mode, the bus master only has to arbitrate for bus
ownership one time. The start addresses and transaction type are issued
during the address phase. All devices on the bus latch the address and
transaction type and decode them to determine which the target device is.
The target device latches the start address into an address counter and is
responsible for incrementing the address from data phase to data phase.
Figure 2.10 shows an example of the burst data transfer.
There are two participants in every PCI burst transfer: the initiator and
the target. The initiator, or bus master, is the device that initiates a transfer.
The target is the device currently addressed by the initiator for the purpose
of performing a data transfer. PCI initiator and target devices are commonly referred to as PCI-compliant agents in the specifications. It should
be noted that a PCI target may be designed such that it can only handle
single data phase transactions. When a bus master attempts to perform a
burst transaction, the target forces the master to terminate the transaction
at the completion of the first data phase. The master must rearbitrate for
Address and
command
Data
Data
Data
Data
Data
Address and
command
Data
Data
………….
Figure 2.10 Example of the burst data transfer.
Zhang_Ch02.indd 219
5/13/2008 5:53:46 PM
220
INDUSTRIAL CONTROL TECHNOLOGY
the bus to attempt resumption of the burst when the next data phase completes. Each burst transfer consists of the following basic components:
(1) the address and the transfer type are output during the address phase;
(2) a data object may then be transferred during each subsequent data phase.
Assuming that neither the initiator nor the target device inserts wait
states in each data phase, a data object may be transferred on the rising
edge of each PCI clock cycle. At a PCI bus clock frequency of 33 MHz, a
transfer rate of 132 MB/s may be achieved. A transfer rate of 264 MB/s
may be achieved in a 64-bit implementation when performing 64-bit transfers during each data phase.
(1) Address phase. Refer to Fig. 2.11. Every PCI transaction (with
the exception of a transaction using 64-bit addressing) starts off
with an address phase one PCI clock period in duration. During
the address phase, the initiator identifies the target device and
the type of transaction (also referred to as command type). The
target device is identified by driving a start address within its
assigned range onto the PCI address and data bus. At the same
time, the initiator identifies the type of transaction by driving
the command type onto the 4-bit wide PCI Command/Byte
Enable bus. The initiator also asserts the FRAME# signal to
indicate the presence of a valid start address or transaction type
on the bus. Since the initiator only presents the start address and
Initiator
starts
transaction
by
inserting
Target latch
and decode
address and
command
Turnaround
cycle,
initiator
stops
driving
AD bus
Address
phase
1
2
Data phase 1
Wait state
3
4
Data phase 2
Wait state
5
6
Data phase 3
Wait state
7
Data 2
Data 3
8
9
CLK
Target begins to drive data
back to initiator
FRAME#
AD
C/BE#
IRDY#
Address
Bus
cmd
Data 1
Byte enables
Byte enables
Byte enables
Wait states
TRDY#
Data transfer
DEVSEL#
Target
device
asserts
DEVSET#
GNT#
Figure 2.11 Typical PCI bus transactions.
Zhang_Ch02.indd 220
5/13/2008 5:53:46 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
221
command for one PCI clock cycle, it is the responsibility of
every PCI target device to latch the address and command on
the next rising edge of the clock so that it may be decoded
subsequently.
By decoding the address latched from the address bus and the
command type latched from the Command/Byte Enable bus, a
target device can determine if it is being addressed and the type
of transaction in progress. It is important to note that the initiator
only supplies a start address to the target during the address
phase. Upon completion of the address phase, the address or data
bus becomes the data bus for the duration of the transaction and
is used to transfer data in each of the data phases. It is the responsibility of the target to latch the start address and to autoincrement it to point to the next group of locations during each
subsequent data transfers.
(2) Data phase. Refer to Fig. 2.11. The data phase of a transaction is
the period during which a data object is transferred between the
initiator and the target. The number of data bytes to be transferred during a data phase is determined by the number of
Command/Byte Enable signals that are asserted by the initiator
during the data phase. Each data phase is at least one PCI clock
period in duration. Both the initiator and the target must indicate
that they are ready to complete a data phase, or the data phase is
extended by a wait state one PCI CLK period in duration. The
PCI bus defines ready signal lines used by both the initiator
(IRDY#) and the target (TRDY#) for this purpose. The initiator
does not issue a transfer count to the target. Rather, in each data
phase it indicates whether it is ready to transfer the current data
item and, if it is, whether it is the final data item. FRAME# is
inserted at the start of the address phase and remains inserted
until the initiator is ready (inserts IRDY#) to complete the final
data phase. When the target samples IRDY# are inserted and
FRAME# are not inserted, it realizes that this is the final data
phase.
Refer to Fig. 2.11. The initiator indicates that the last data
transfer (of a burst transfer) is in progress by uninserting
FRAME# and inserting IRDY#. When the last data transfer has
been completed, the initiator returns the PCI bus to the idle state
by uninserting its ready line (IRDY#). If another bus master had
previously been granted ownership of the bus by the PCI bus
arbiter and was waiting for the current initiator to surrender the
bus, it can detect that the bus has returned to the idle state by
detecting FRAME# and IRDY# both uninserted on the same rising
edge of the PCI clock.
Zhang_Ch02.indd 221
5/13/2008 5:53:47 PM
222
2.1.4.2
INDUSTRIAL CONTROL TECHNOLOGY
Bus System Arbitration
Bus masters are devices on a PCI bus that are allowed to take control of
the bus. A component named bus arbiter works for this purpose. An arbiter
is usually integrated into the PCI chip set; specifically, it is typically integrated into the host/PCI or the PCI/expansion bus bridge chip. Each master
device is physically connected to the arbiter via a separate pair of lines,
with each of them being used as REQ# (request) signal or GNT# (grant)
signal, respectively. Ideally, the bus arbiter should be programmable by the
system. If it is, the startup configuration software can determine the priority to be assigned to each member by reading from the maximum latency
(Max_Lat) configuration register associated with each bus master (see
Fig. 2.12). The bus designer hardwires this register to indicate, in increments of 250 ns, how quickly the master requires access to the bus in order
to achieve adequate performance.
At a given instant in time, one or more PCI bus master devices may
require use of the PCI bus to perform a data transfer with another PCI
device. Each requesting master asserts its REQ# output to confirm to the
bus arbiter its pending request for the use of the bus. In order to grant the
PCI bus to a bus master, the arbiter asserts the device’s respective GNT#
signal. This grants the bus to a bus master for one transaction as given in
Fig. 2.11. If a master generates a request, it is subsequently granted the bus
and does not initiate a transaction by asserting FRAME# signal within 16
PCI clocks after the bus goes idle; the arbiter may assume that this bus
master is malfunctioning. In this case, the action taken by the arbiter would
depend upon the system design. If a bus master has another transaction to
31
0
Unit ID
Status
Class code
BIST
Header
64-byte header
Man. ID
Command
Latency
Revision
CLS
Base address register
Reserved
Reserved
Expansion ROM base address
Reserved
Reserved
Max_Lat
Min_GNT
INT-pin
INT-line
Figure 2.12 PCI configuration space.
Zhang_Ch02.indd 222
5/13/2008 5:53:47 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
223
perform immediately after the one it just initiated, it should keep its REQ#
line asserted when it asserts FRAME# signal to begin the current transaction. This informs the arbiter of its desire to maintain ownership of the bus
after completion of the current transaction. In the event that ownership is
not maintained, the master should keep its REQ# line asserted until it is
successful in acquiring bus ownership again.
However, at a given instant in time, only one bus master may use the
bus. This means that no more than one GNT# line will be asserted by the
arbiter during any PCI clock cycle. On the other hand, a master must only
assert its REQ# output to signal a current need for the bus. This means that
a master must not use its REQ# line to “park” the bus on itself. If a system
designer implements a bus parking scheme, the bus arbiter design should
indicate a default bus owner by asserting the device’s GNT# signal when
no request from any bus masters are currently pending. In this manner,
signal REQ # from the default master is granted immediately once no other
bus masters require the use of the PCI bus.
The PCI specification does not define the scheme used by the PCI bus
arbiter to decide the winner of the competition when multiple masters
simultaneously request bus ownership. The arbiter may utilize any scheme,
such as one based on fixed or rotational priority or a combination of these
two, to avoid deadlocks. However, the central arbiter is required to implement a fairness algorithm to avoid deadlocks. Fairness means that each
potential bus master must be granted access to the bus independent of other
requests. Fairness is defined as a policy that ensures that high-priority
masters will not dominate the bus to the exclusion of lower priority masters when they are continually requesting the bus. However, this does not
mean that all agents are required to have equal access to the bus. By requiring a fairness algorithm there are no special conditions to handle when the
signal LOCK# is active (assuming a resource lock) or when cacheable
memory is located on PCI. A system that uses a fairness algorithm is still
considered fair if it implements a complete bus lock instead of resource
lock. However, the arbiter must advance to a new agent if the initial transaction attempting to establish a lock is terminated with retry.
2.1.4.3
Interrupt Routing
The host/PCI bus bridge will transfer the interrupt acknowledgment
cycle from the processor to the PCI bus, which requires the microprocessor chipset having an interrupt routing functionality. This router for the
interrupt routing could be implemented using an Intel APIC I/O module as
given in Figs 2.3(a) and 2.4. The APIC I/O module can be programmed to
Zhang_Ch02.indd 223
5/13/2008 5:53:47 PM
224
INDUSTRIAL CONTROL TECHNOLOGY
assign a separate interrupt vector (interrupt table entry number) for each of
the PCI interrupt request lines. It can also be programmed so that it realizes that one of its inputs is connected to an Intel programmable interrupt
controller. If a system does not have this kind of programmable interrupt
controller, the microprocessor chipset should incorporate a software programmable interrupt routine device. In this case, the startup configuration
software of the microprocessor attempts to program the router to distribute
the PCI interrupt in an optimal fashion.
Whenever any of the PCI interrupt request lines is asserted, the APIC
I/O module supplies the vector (see Fig. 2.6 for an interrupt vector table)
associated with that input to the processor’s embedded local APIC I/O
module. Whenever this programmable interrupt controller generates a
request, the APIC I/O informs the processor that it must poll this programmable interrupt controller to get this vector. In response, the Intel processor can generate two back-to-back Interrupt Acknowledge transactions.
The first Interrupt Acknowledge forces this programmable interrupt controller to prioritize the interrupts pending, while the second Interrupt
Acknowledge requests that the interrupt controller send the vector to the
processor. For a detailed discussion of APIC operation, refer to the
MindShare book entitled Pentium Processor System Architecture (published by Addison-Wesley). For a detailed description of the Programmable
Interrupt Controller chipset, refer to Section 2.2.2.
Figure 2.11 can also be used to explain an interrupt acknowledgment
cycle on the PCI bus, where a single byte enable is asserted. The PCI bus
performs only one interrupt acknowledgment cycle per interrupt. Only one
device may respond to the interrupt acknowledgment; that device must
assert DEVSEL# indicating that it is claiming the interrupt acknowledgment. The sequence is as follows:
(1) During the address phase, the AD signals do not contain a valid
address; they must be driven with stable data so that parity can be
checked. The C/BE# signals contain the interrupt acknowledge
command code (not shown).
(2) IRDY# and the BE#s are driven by the host/PCI bus bridge to
indicate that the bridge (master), is ready for response.
(3) The target will drive DEVSEL# and TRDY# along with the vector on the data bus (not shown).
2.1.4.4
Configuration Registers
Each PCI device has 256 bytes of configuration data, which is arranged
as 64 registers of 32 bits. It contains a 64-byte predefined header followed
Zhang_Ch02.indd 224
5/13/2008 5:53:47 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
225
by an extra 192 bytes which contain extra configuration data. Figure 2.12
shows the arrangement of the header. The definitions of the fields in this
header are as follows:
(1) Unit ID and Man. ID. A Unit ID of FFFFh defines that there is
no unit installed, while any other address defines its ID. The PCI
SIG, which is the governing body for the PCI specification, allocates a Man. ID. This ID is normally shown at BIOS start-up.
(2) Status and command
(3) Class code and revision. The class code defines PCI device type.
It splits into two 8-bit values with a further 8-bit value that defines
the programming interface for the unit. The first defines the unit
classification, followed by a subcode which defines the actual
type.
(a) BIST, header, latency, CLS. The built-in-self test (BIST) is an
8-bit field, where the most significant bit defines if the device
can carry out a BIST, the next bit defines if a BIST is to be
performed (a 1 in this position indicates that it should be performed), and bits 3–0 define the status code after the BIST has
been performed (a value of zero indicates no error). The header
field defines the layout of the 48 bytes after the standard
16-byte header. The most significant bit of the header field
defines whether the device is a multifunction device or not. A
1 defines a multifunction unit. The cache line size (CLS) field
defines the size of the cache in units of 32 bytes. Latency indicates the length of time for a PCI bus operation, where the
amount of time is the latency + 8 PCI clock cycles.
(b) Base address register. This area of memory allows the device
to be programmed with an I/O or memory address area. It
can contain a number of 32- or 64-bit addresses. The format
of a memory address is
(i) Bit 64-4: base address;
(ii) Bit 3: PRF. Prefetching, 0 identifies not possible, 1 identifies possible;
(iii) Bit 2, 1: Type. 00—any 32-bit address, 01—less than
1 MB, 10—any 64-bit address, and 11—reserved;
(iv) Bit 0: 0. Always set a 0 for a memory address.
For an I/O address space it is defined as:
(i) Bit 31-2: base address;
(ii) Bit 1, 0: 01. Always set to a 01 for an I/O address.
(c) Expansion ROM base address. It allows a ROM expansion to
be placed at any position in the 32-bit memory address area.
(d) Max_Lat, Min_GNT, INT-pin, INT-line. The Min_GNT and
Max_Lat registers are read-only registers that define minimum
Zhang_Ch02.indd 225
5/13/2008 5:53:47 PM
226
INDUSTRIAL CONTROL TECHNOLOGY
and maximum latency values. The INT-line field is a 4-bit
field that defines the interrupt line used (IRQ0–IRQ15). A
value of 0 corresponds to IRQ0 and a value of 15 corresponds
to IRQ15. The PCI bridge can then redirect this interrupt to
the correct IRQ line. The 4-bit INT pin defines the interrupt
line that the device is using. A value of 0 defines no interrupt
line, 1 defines INTA, 2 defines INTB, and so on.
2.2 Programmable Peripheral Devices
A programmable peripheral device is designed to perform various interface functions. Such a device can be set up to perform specific functions
by writing an instruction (or instructions) in its internal register, called the
control register. Furthermore, function can be changed any time during
execution of the program by writing a new instruction in the control register. These devices are flexible, versatile, and economical; they are widely
used in microprocessor-based products.
In a programmable device, on the other hand, functions are determined
through software instructions. A programmable peripheral device can
not only be viewed as a multiple I/O device, but it also performs many
other functions, such as time delay, interrupt handling, and graphic user–
machine interactions, etc. In fact, it consists of many devices on a single
chip, interconnected through a common bus. This is a hardware approach
through software control to performing the I/O functions, discussed earlier
in this chapter. This approach, a trade-off between hardware and software,
should reduce programming efforts.
This section describes five typical programmable peripheral devices
that are the programmable I/O ports: interrupt, controller, timer, CMOS,
and DMA.
2.2.1
Programmable Peripheral I/O Ports
The programmable peripheral interface (PPI), especially the 8255
Programmable Peripheral I/O Interface, is a very popular and versatile
input and output chip that is easily configured to function in several different configurations. The 8255 is used on several ranges of cards that plug
into an available slot in controllers or computers. This chip allows the use
of both digital input and output (DIO) with controllers or computers.
As illustrated in Fig. 2.13, each 8255 Programmable Peripheral I/O
Interface has three off 8-bit TTL-compatible I/O ports that will allow the
Zhang_Ch02.indd 226
5/13/2008 5:53:47 PM
227
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
D0-D7
Data
bus
buffer
RD
WR
A0
A1
Reset
Read/write
control
logic
GA control
GB control
I
n
t
e
r
n
a
l
d
a
t
a
PA
PA0-PA7
PCU
PC4-PC7
PCL
PC0-PC3
CS
b
u
s
PB
PB0-PB7
Figure 2.13 8255 Programmable peripheral I/O interface functional blocks.
control of up to 24 individual outputs or 24 individual inputs or both input
or output. For example, they can be attached to a robotic device to control
movement by use of motors to control motion and switches to detect position, etc.
Addressing ports is different from addressing memory. Ports have port
addresses and memory has memory addresses; port address 1234 is different from memory address 1234. The 8255 Programmable Peripheral I/O
Interface cards use port addresses and cannot be set to use memory addresses
(see Table 2.2). The 8255 Programmable Peripheral I/O Interface cards
plug into any available 8- or 16-bit slot (also known as an AT or ISA slot)
on the motherboard of a controller or a computer, just like a sound card or
disk drive controller card does with personal computers. The CPU of the
motherboard communicates with cards by knowing the card’s address and
sending data to it. By physically using jumpers on the card, we can assign
a set of addresses to the card; then in software, we can tell the CPU what
these addresses are (more about this in the programming section).
The first thing that must be done, before the chip can be used, is to tell
it which configuration is required. The configuration tells the 8255 whether
ports are input or output and even some strange arrangements called bidirectional and strobed. The 8255 allows for three distinct operating modes
(modes 0, 1, and 2) as follows:
(1) Mode 0—Basic input/output. Ports A and B operate as either
inputs or outputs and Port C is divided into two 4-bit groups
either of which can be operated as inputs or outputs.
Zhang_Ch02.indd 227
5/13/2008 5:53:47 PM
228
INDUSTRIAL CONTROL TECHNOLOGY
Table 2.2 DC-0600 Addresses
Options address
8255 8255Port
Port 1A
Port 1B
Port 1C
Port 1 Control register
Port 2A
Port 2B
Port 2C
Port 2 Control register
Option 1: Default
(JP2 linked)
Option 2
(JP2 open)
Address [Hex (decimal)]
300H (768)
301H (769)
302H (770)
303H (771)
304H (772)
305H (773)
306H (774)
307H (775)
Address [Hex (decimal)]
360H (864)
361H (865)
362H (866)
363H (867)
364H (868)
365H (869)
366H (870)
367H (871)
(2) Mode 1—Strobed input/output. Same as Mode 0 but Port C is
used for handshaking and control.
(3) Mode 2—Bidirectional bus. Port A is bidirectional (both input and
output) and Port C is used for handshaking. Port B is not used.
For most applications using this range of cards, mode 0 will be used.
Each of the 3 ports has 8 bits, and each of these bits can be individually set
ON or OFF, somewhat like having three banks of eight light switches.
These bits are configured in groups to be inputs or outputs allowing their
function to either read data into the computer or control data out of the
computer. The various modes can be set by sending a value to the control
port. The control port is Base Address + 3 (i.e., 768 + 3 = 771 decimal).
Table 2.3 shows the different arrangements that can be configured and the
values to be sent to the configuration port.
Table 2.3 8255 Control Register Configuration (Mode 0)
Controlword [Hex(Dec)]
Port A
Port B
80H (128)
82H (130)
85H (133)
87H (135)
88H (136)
8AH (138)
8CH (140)
8FH (143)
OUT
OUT
OUT
OUT
IN
IN
IN
IN
OUT
IN
OUT
IN
OUT
IN
OUT
IN
Zhang_Ch02.indd 228
5/13/2008 5:53:48 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
229
As mentioned, the control port is Base Address + 3. Port A is always at
Base Address; Port B is Base Address + 1; Port C is Base Address + 2.
Thus, in our example Ports A, B, and C are at 768, 769, and 770 (decimal),
respectively. By writing, say, 128, the control port will then configure the
8255 to have all three ports set for output.
2.2.2
Programmable Interrupt Controller Chipset
Both microcontroller and microcomputer system designs require that
I/O devices such as keyboards, displays, sensors, and other components
receive servicing in an efficient manner so that large amounts of the total
system tasks can be assumed by the microcomputer with little or no effect
on throughput. As mentioned in Section 2.1.2, there are two common
methods of servicing such devices: the first method is the polled approach;
the second is a more desirable method called interrupt that allows the
microprocessor to continue executing its main program and only stop to
service peripheral devices when it is told to do so by the device itself.
The programmable interrupt controller (PIC) functions as an overall
manager in an interrupt-driven system environment. It accepts requests
from the peripheral equipment, determines which of the incoming requests
is of the highest importance (priority), ascertains whether the incoming
request has a higher priority value than the level currently being serviced,
and issues an interrupt to the CPU based on this determination. The PIC,
after issuing an interrupt to the CPU, must somehow input information into
the CPU that can “point” the program counter to the service routine associated with the requesting device. This “pointer” is an address in a vectoring
table and will often be referred to, in this document, as vectoring data.
The 8259A is taken as an example of the PIC. It manages eight levels or
requests and has built-in features for expandability to other 8259As (up to
64 levels). It is programmed by the system’s software as an I/O peripheral.
A selection of priority modes is available to the programmer so that the
manner in which the requests are processed by the 8259A can be configured to match system requirements. The priority modes can be changed or
reconfigured dynamically at any time the main program is executing. This
means that the complete interrupt structure can be defined on the requirements of the total system environment.
Figure 2.14 gives the block function diagram of 8259A PIC, which
includes these function blocks and pins:
(1) Interrupt request register (IRR) and in-service register (ISR).
The interrupts at the IR input lines are handled by two registers
Zhang_Ch02.indd 229
5/13/2008 5:53:48 PM
230
INDUSTRIAL CONTROL TECHNOLOGY
INTAn
INT
Control logic
D[7..0]
RDn
WRn
CSn
A0
RESET
Data bus
buffer
Read/write
logic
In-service
register
Priority
resolver
Interrupt
request
register
IR0
IR1
IR2
IR3
IR4
IR5
IR6
IR7
CAS0
CAS1
CAS2
Cascade
buffer/
comparator
Interrupt mask register
SPnENn
Figure 2.14 8259A PIC block diagram.
in cascade, the IRR and the ISR. The IRR is used to store all the
interrupt levels that are requesting service, and the ISR is used
to store all the interrupt levels that are being serviced.
(2) Priority resolver. This logic block determines the priorities of
the bits set in the IRR. The highest priority is selected and strobed
into the corresponding bit of the ISR during INTA pulse.
(3) Interrupt mask register (IMR). The IMR stores the bits that
mask the interrupt lines to be masked. The IMR operates on the
IRR. Masking of a higher priority input will not affect the interrupt request lines of lower priority.
(4) INT (interrupt). This output goes directly to the CPU interrupt
input. The VOH level on this line is designed to be fully compatible with the 8080A, 8085A, and 8086 input levels.
(5) INTA (interrupt acknowledge). INTA pulses will cause the 8259A
to release vectoring information onto the data bus. The format of
this data depends on the system mode (mPM) of the 8259A.
(6) Data bus buffer. This three-state, bidirectional 8-bit buffer is
used to interface the 8259A to the system data bus. Control
words and status information are transferred through the data
bus buffer.
(7) Read/write control logic. The function of this block is to accept
OUTPUT commands from the CPU. It contains the initialization command word (ICW) registers and operation command
word (OCW) registers that store the various control formats for
device operation. This function block also allows the status of
the 8259A to be transferred onto the data bus.
Zhang_Ch02.indd 230
5/13/2008 5:53:48 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
231
(8) CS (chip select). A LOW on this input enables the 8259A. No
reading or writing of the chip will occur unless the device is
selected.
(9) WR (write). A LOW on this input enables the CPU to write
control words (ICWs and OCWs) to the 8259A.
(10) RD (read). A LOW on this input enables the 8259A to send
the status of the IRR, ISR, IMR, or the interrupt level onto the
data bus.
(11) A0. This input signal is used in conjunction with WR and RD
signals to write commands into the various command registers,
as well as reading the various status registers of the chip. This
line can be tied directly to one of the address lines.
(12) The cascade buffer/comparator. This function block stores and
compares the IDs of all 8259A’s used in the system. The associated three I/O pins (CAS0-2) are outputs when the 8259A is
used as a master and are inputs when the 8259A is used as a
slave. As a master, the 8259A sends the ID of the interrupting
slave device onto the CAS0 ± 2 lines. The slave thus selected
will send its preprogrammed subroutine address onto the data
bus during the next one or two consecutive INTA pulses.
The powerful features of the 8259A in a microcomputer system are its programmability and the interrupt routine addressing capability. The latter allows direct or indirect jumping to the
specific interrupt routine requested without any polling of the
interrupting devices. The normal sequence of events during an
interrupt depends on the type of CPU being used. The events
occur as follows in an MCS-80/85 system:
(a) One or more of the INTERRUPT REQUEST lines (IR7 ± 0)
are raised high, setting the corresponding IRR bit(s).
(b) The 8259A evaluates these requests and sends an INT to
the CPU, if appropriate.
(c) The CPU acknowledges the INT and responds with an
INTA pulse.
(d) Upon receiving an INTA from the CPU group, the highest
priority ISR bit is set, and the corresponding IRR bit is reset.
The 8259A will also release a CALL instruction code
(11001101) onto the 8-bit data bus through its D7 ± 0 pins.
(e) This CALL instruction will initiate two more INTA pulses
to be sent to the 8259A from the CPU group.
(f) These two INTA pulses allow the 8259A to release its preprogrammed subroutine address onto the data bus. The
lower 8-bit address is released at the first INTA pulse and
the higher 8-bit address is released at the second INTA
pulse.
Zhang_Ch02.indd 231
5/13/2008 5:53:48 PM
232
INDUSTRIAL CONTROL TECHNOLOGY
(g) This completes the 3-byte CALL instruction released by
the 8259A. In the AEOI mode the ISR bit is reset at the end
of the third INTA pulse. Otherwise, the ISR bit remains set
until an appropriate EOI command is issued at the end of
the interrupt sequence.
The events occurring in an 8086 system are the same until
the fourth step; from the fourth step onward: (d) Upon receiving an INTA from the CPU group, the highest priority ISR bit
is set and the corresponding IRR bit is reset. The 8259A does
not drive the data bus during this cycle. (e) The 8086 will
initiate a second INTA pulse. During this pulse, the 8259A
releases an 8-bit pointer onto the data bus where it is read by
the CPU. (f) This completes the interrupt cycle. In the AEOI
mode the ISR bit is reset at the end of the second INTA pulse.
Otherwise, the ISR bit remains set until an appropriate EOI
command is issued at the end of the interrupt subroutine. If
no interrupt request is present at step (d) of either sequence
(i.e., the request was too short in duration) the 8259A will
issue an interrupt level 7. Both the vectoring bytes and the
CAS lines will look like an interrupt level 7 was requested.
When the 8259A PIC receives an interrupt, INT becomes
active and an interrupt acknowledge cycle is started. If a
higher priority interrupt occurs between the two INTA
pulses, the INT line goes inactive immediately after the
second INTA pulse. After an unspecified amount of time
the INT line is activated again to signify the higher priority
interrupt waiting for service. This inactive time is not specified and can vary between parts. The designer should be
aware of this consideration when designing a system that
uses the 8259A. It is recommended that proper asynchronous design techniques be followed.
Advanced programmable interrupt controllers (APICs) are designed to
attempt to solve interrupt routing efficiency issues in multiprocessor computer systems. There are two components in the Intel APIC system: the
local APIC (LAPIC) and the IOAPIC. The LAPIC is integrated into each
CPU in the system, and the IOAPIC is used throughout the system’s
peripheral buses. There is typically one IOAPIC for each peripheral bus in
the system. In the original system designs, LAPICs and IOAPICs were
connected by a dedicated APIC bus. Newer systems use the system bus for
communication between all APIC components.
LAPICs manage all external interrupts for the processor that it is part of.
In addition, they are able to accept and generate interprocessor interrupts
(IPIs) between LAPICs. LAPICs may support up to 224 usable IRQ vectors
Zhang_Ch02.indd 232
5/13/2008 5:53:48 PM
233
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
from an IOAPIC. Vectors numbers 0–31, out of 0–255, are reserved for
exception handling by x86 processors.
IOAPICs contains a redirection table, which is used to route the interrupts it receives from peripheral buses to one or more LAPICs.
2.2.3
Programmable Timer Controller Chipset
The programmable timer controller provides a programmable interval
timer and counter that are designed for use with microcomputer systems to
solve one of the most common problems in any microcomputer system:
the generation of accurate time delays under software control. Instead of
setting up timing loops in software, the programmer configures the programmable timer controller to match the requirements and programs one
of the counters for the desired delay. After the desired delay, the programmable timer controller will interrupt the CPU. Software overhead is minimal and variable length delays can easily be accommodated.
Some of the other computer and timer functions common to microcomputers that can be implemented with it are real-time clock, event counter,
digital one shot, programmable rate generator, square wave generator, binary
rate multiplier, complex waveform generator, and complex motor controller.
Figure 2.15 gives the typical function blocks for an 82C54 programmable interval timer controller, which has these main blocks:
(1) Data bus buffer. This three-state, bidirectional 8-bit buffer is
used to interface the 82C54 to the system bus.
CLK 0
D7–D0
8
Counter
0
Data bus
buffer
OUT 0
RD
WR
Read/
write
logic
Control
word
register
Internal bus
A0
A1
GATE 0
Counter
1
CLK 1
GATE 1
OUT 1
CLK 2
Counter
0
GATE 2
OUT 2
Figure 2.15 82C54 Programmable timer controller function blocks.
Zhang_Ch02.indd 233
5/13/2008 5:53:48 PM
234
INDUSTRIAL CONTROL TECHNOLOGY
(2) Read/write logic. The read/write logic accepts inputs from the
system bus and generates control signals for the other functional
blocks of the 82C54. A1 and A0 select one of the three counters
on the control word register to be read from or written into. A
“low” on the RD input tells the 82C54 that the CPU is reading
one of the counters. A “low” on the WR input tells the 82C54
that the CPU is writing either a control word or an initial count.
Both RD and WR are qualified by CS; RD and WR are ignored
unless the 82C54 has been selected by holding CS low.
(3) Control word register. The control word register is selected by
the read/write logic when A1, A0 = 11. If the CPU then does a
write operation to the 82C54, the data is stored in the control
word register and is interpreted as a control word used to define
the counter operation. The control word register can only be
written to.
(4) Counter 0, Counter 1, Counter 2. These three functional blocks
are identical in operation, so only a single counter will be
described. The counters are fully independent. Each counter may
operate in a different mode.
The programmable timer is normally treated by the system
software as an array of peripheral I/O ports; three are counters
and the fourth is a control register for mode programming.
Basically, the select inputs A0, A1 connects to the A0, A1 address
bus signals of the CPU. The CS can be derived directly from the
address bus using a linear select method or it can be connected to
the output of a decoder.
After power-up, the state of the programmable timer is undefined. The mode, count value, and output of all counters are
undefined. How each counter operates is determined when it is
programmed. Each counter must be programmed before it can be
used. Unused counters need not be programmed. Counters are
programmed by writing a control word and then an initial count.
All control words are written into the control word register, which
is selected when A1, A0 = 11. The control word specifies which
counter is being programmed. By contrast, initial counts are
written into the counters, not the control word register. The A1,
A0 inputs are used to select the counter to be written into. The
format of the initial count is determined by the control word used.
(a) Write operations. A new initial count may be written to a
counter at any time without affecting the counter’s programmed mode in any way. Counting will be affected as
described in the mode definitions.
The new count must follow the programmed count format.
If a counter is programmed to read and write 2-byte counts,
Zhang_Ch02.indd 234
5/13/2008 5:53:49 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
235
the following precaution applies. A program must not transfer
control between writing the first and second byte to another
routine that also writes into that same counter. Otherwise,
the counter will be loaded with an incorrect count.
(b) Read operations. There are three possible methods for reading
the counters. The first is through the read-back command.
The second is a simple read operation of the counter, which
is selected with the A1, A0 inputs. The only requirement is
that the CLK input of the selected counter must be inhibited
by using either the GATE input or external logic. Otherwise,
the count may be in the process of changing when it is read,
giving an undefined result.
2.2.4
CMOS Chipset
The complementary metal oxide semiconductor (CMOS) chip is battery-powered and stores the hard drive’s configuration and other information. In a microcomputer and a microcontroller, CMOS chips normally
provide two functions: real-time clock (RTC) and CMOS memory.
The real-time clock provides the board with time-of-day clock, periodic
interrupt, and system configuration information to a microcomputer or a
microcontroller. With respect to personal computers, CMOS chipset typically contains 64 (00hex-3Fhex) 8-bit locations of battery-backed up
CMOS RAM (random access memory). The split is (1) 00hex-0Ehex, used
for real-time clock functions (time of day), (2) 0Fhex-35hex, used for system configuration information, for example, hard drive type, memory size,
etc., and (3) 36hex-3Fhex, used for power-on password storage.
The CMOS memory is an accessible set of memory locations on the
same chip as the RTC and has its own battery backup so that it retains both
functions, even when the computer is turned off. Battery-powered CMOS
and RTCs did not originally exist and the current time was entered manually every time the system was turned on. This memory location in CMOS
is separate and apart from the RTC registers and there are several ways to
update it. Specifically, the BIOS can update the century information, as
can many operating systems, network time systems, and applications, or
the user can set it using the right commands.
2.2.5
Direct Memory Access Controller Chipset
Direct memory access (DMA) is an I/O technique commonly used for
high-speed data transfer among internal memories, I/O ports, and peripherals, and also between the memories and I/O devices on different chipsets.
Zhang_Ch02.indd 235
5/13/2008 5:53:49 PM
236
INDUSTRIAL CONTROL TECHNOLOGY
DMA technique allows the microprocessor to release the control of the
buses to a device called a DMA controller. The DMA controller manages
data transfer between memory and a peripheral under its control, thus
bypassing the microprocessor. The microprocessor communicates with
the controller by using the chip select line, buses, and control signals.
However, once the controller has gained control, it plays the role of a
microprocessor for data transfer. For all practical purposes, the DMA controller is a microprocessor capable only of copying data at high speed from
one location to another location. As an illustration, a programmable DMA
controller, the Intel 8237A programmable DMA controller, is described
below.
The 8237A block diagram given in Fig. 2.16 includes the major logic
blocks and all of the internal registers. The data interconnection paths are
also shown. Not shown are the various control signals between the blocks.
The 8237A contains 344 bits of internal memory in the form of registers.
Table 2.4 lists these registers by name and shows the size of each. The
8237A contains three basic blocks of control logic. The Timing Control
block generates internal timing and external control signals for the 8237A.
The Program Command Control block decodes the various commands
RESET
CLK
CSN
Ready
IORNIN
IOWNIN
EOPNIN
AEN
ADSTB
EOPNOUT
MEMRN
MEMWN
IORNOUT
IOWNOUT
16-bit
incrementor/
decrementor
temp address reg
16-bit
decrementor
temp word
count reg
Timing
and
control
Channel-3
Channel-2
AOUT [7..0]
C8237REG
Channel-1
Channel-0
Read/write
State
machine
Current word
count register
Current word
address register
Write
Base word
address register
Base word
count register
HLDA
DBOUT [7..0]
DREQ [3.0]
HRQ
DACK [3..0]
Fixed
priority
and
rotating
priority
logic
Command
register
DBIN [7..0]
Mask
register
AIN [3..0]
Request
register
Mode
register
Status
register
Temporary
register
Figure 2.16 Intel 8237A DMA controller block diagram.
Zhang_Ch02.indd 236
5/13/2008 5:53:49 PM
237
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
Table 2.4 8237A DMA Controller Internal Registers
Name
Base address registers
Base word count registers
Current address registers
Current word count registers
Temporary address register
Temporary word count register
Status register
Command register
Temporary register
Mode registers
Mask register
Request register
Size (bits)
Number
16
16
16
16
16
16
8
8
8
6
4
4
4
4
4
4
1
1
1
1
1
4
1
1
given to the 8237A by the microprocessor prior to servicing a DMA
request. It also decodes the mode control word used to select the type of
DMA during the servicing. The Priority Encoder block resolves priority
contention between DMA channels requesting service simultaneously.
To perform block moves of data from one memory address space to
another with a minimum of program effort and time, the 8237A includes a
memory-to-memory transfer feature. Programming a bit in the command
register selects channels 0 and 1 to operate as memory-to-memory transfer
channels. The transfer is initiated by setting the software DREQ for channel 0. The 8237A requests a DMA service in the normal manner. After
HLDA is true, the device, using four state transfers in block transfer mode,
reads data from the memory. The channel 0 current address register is the
source for the address used and is decremented or incremented in the normal manner. The data byte read from the memory is stored in the 8237A
internal temporary register. Channel 1 then performs a four-state transfer
of the data from the temporary register to memory using the address in its
current address register and incrementing or decrementing it in the normal
manner. The channel 1 current word count is decremented. When the word
count of channel 1 goes to FFFFH, a TC is generated causing an EOP output terminating the service. Channel 0 may be programmed to retain the
same address for all transfers. This allows a single word to be written to a
block of memory. The 8237A will respond to external EOP signals during
memory-to-memory transfers. Data comparators in block search schemes
may use this input to terminate the service when a match is found.
Zhang_Ch02.indd 237
5/13/2008 5:53:49 PM
238
INDUSTRIAL CONTROL TECHNOLOGY
The 8237A will accept programming from the host processor any time
that the HLDA is inactive; this is true even if HRQ is active. The responsibility of the host is to ensure that programming and HLDA are mutually
exclusive. Note that a problem can occur if a DMA request occurs on an
unmasked channel while the 8237A is being programmed. For instance,
the CPU may be starting to reprogram the 2-byte address register of channel 1 when channel 1 receives a DMA request. If the 8237A is enabled
(bit 2 in the command register is 0) and channel 1 is unmasked, a DMA
service will occur after only 1 byte of the address register has been reprogrammed. This can be avoided by disabling the controller (setting bit 2 in
the command register) or masking the channel before programming any
other registers. Once the programming is complete, the controller can be
enabled and unmasked. After power-up it is suggested that all internal
locations, especially the mode registers, be loaded with some valid value.
This should be done even if some channels are unused. An invalid mode
may force all control signals to go active at the same time.
The 8237A is designed to operate in two major cycles. These are called
idle and active cycles. Each device cycle is made up of a number of states.
The 8237A can assume seven separate states, each composed of one full
clock period. State I (SI) is the inactive state. It is entered when the 8237A
has no valid DMA requests pending. While in SI, the DMA controller is
inactive but may be in the program condition, being programmed by the
processor. State S0 (S0) is the first state of a DMA service. The 8237A has
requested a hold but the processor has not yet returned an acknowledge.
The 8237A may still be programmed until it receives HLDA from the
CPU. An acknowledge from the CPU will signal that DMA transfers may
begin. S1, S2, S3, and S4 are the working states of the DMA service. If
more time is needed to complete a transfer than is available with normal
timing, wait states (SW) can be inserted between S2 or S3 and S4 by the
use of the ready line on the 8237A. Eight states are required for a single
transfer. The first four states (S11, S12, S13, S14) are used for the readfrom-memory half and the last four states (S21, S22, S23, S24) for the
write-to-memory half of the transfer.
2.2.5.1
Idle Cycle
When no channel is requesting service, the 8237A will enter the idle
cycle and perform “SI” states. In this cycle the 8237A will sample the
DREQ lines every clock cycle to determine whether any channel is requesting a DMA service. The device will also sample CS, looking for an attempt
by the microprocessor to write or read the internal registers of the 8237A.
Zhang_Ch02.indd 238
5/13/2008 5:53:49 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
239
When CS is low and HLDA is low, the 8237A enters the program condition. The CPU can now establish, change, or inspect the internal definition
of the part by reading from or writing to the internal registers. Address
lines A0 ± A3 are inputs to the device and select which registers will be
read or written. The IOR and IOW lines are used to select and time reads
or writes. Special software commands can be executed by the 8237A in the
program condition. These commands are decoded as sets of addresses with
the CS and IOW. The commands do not make use of the data bus.
Instructions include Clear First/Last Flip-Flop and Master Clear.
2.2.5.2 Active Cycle
When the 8237A is in the idle cycle and a nonmasked channel requests
a DMA service, the device will output an HRQ to the microprocessor and
enter the active cycle. It is in this cycle that the DMA service will take
place, in one of four modes:
(1) Single transfer mode. In single transfer mode the device is programmed to make one transfer only. The word count will be decremented and the address decremented or incremented following
each transfer. When the word count “rolls over” from 0 to FFFFH,
a terminal count (TC) will cause an autoinitialize if the channel
has been programmed to do so. DREQ must be held active until
DACK becomes active in order to be recognized. If DREQ is
held active throughout the single transfer, HRQ will go inactive
and release the bus to the system. It will again go active and,
upon receipt of a new HLDA, another single transfer will be performed. Details of timing between the 8237A and other bus
control protocols will depend upon the characteristics of the
microprocessor involved.
(2) Block transfer mode. In block transfer mode the device is activated by DREQ to continue making transfers during the service
until a TC, caused by word count going to FFFFH, or an external
end of process (EOP) is encountered. DREQ need only be held
active until DACK becomes active. Again, an autoinitialization
will occur at the end of the service if the channel has been programmed for it.
(3) Demand transfer mode. In demand transfer mode the device is
programmed to continue making transfers until a TC or external
EOP is encountered or until DREQ goes inactive. Thus, transfers
may continue until the I/O device has exhausted its data capacity.
After the I/O device has had a chance to catch up, the DMA service is reestablished by means of a DREQ. During the time
Zhang_Ch02.indd 239
5/13/2008 5:53:49 PM
240
INDUSTRIAL CONTROL TECHNOLOGY
between services when the microprocessor is allowed to operate,
the intermediate values of address and word count are stored in
the 8237A current address and current word count registers. Only
an EOP can cause an autoinitialize at the end of the service. EOP
is generated either by TC or by an external signal. DREQ has to
be low before S4 to prevent another transfer.
(4) Cascade mode. This mode is used to cascade more than one
8237A together for simple system expansion. The HRQ and HLDA
signals from the additional 8237A are connected to the DREQ
and DACK signals of a channel of the initial 8237A. This allows
the DMA requests of the additional device to propagate through
the priority network circuitry of the preceding device. The priority chain is preserved and the new device must wait for its turn to
acknowledge requests. Since the cascade channel of the initial
8237A is used only for prioritizing the additional device, it does
not output any address or control signals of its own. These could
conflict with the outputs of the active channel in the added device.
The 8237A will respond to DREQ and DACK but all other outputs except HRQ will be disabled. The ready input is ignored.
Each of the three active transfer modes above can perform three different types of transfers. These are read, write, and verify. Write transfers
move data from an I/O device to the memory by activating MEMW and
IOR. Read transfers move data from memory to an I/O device by activating MEMR and IOW. Verify transfers are pseudotransfers. The 8237A
operates as in read or write transfers generating addresses, and responding
to EOP, etc. However, the memory and I/O control lines all remain inactive. The ready input is ignored in verify mode.
2.3 Application-Specific Integrated
Circuit (ASIC)
ASIC is basically an integrated circuit designed specifically for a special purpose or application. Strictly speaking, this also implies that an
ASIC is built for one and only one customer. The opposite of an ASIC is a
standard product or general purpose IC, such as a logic gate or a general
purpose microcontroller, both of which can be used in any electronic application by anybody. ASICs are usually classified into one of three categories: full custom, semi-custom, and structured as listed below.
(1) Full-custom ASICs are those that are entirely tailor-fitted to a
particular application from the very start. Since ultimate design
Zhang_Ch02.indd 240
5/13/2008 5:53:49 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
241
and functionality are prespecified by the user, they are manufactured with all the photolithographic layers of the device already
fully defined, just like most off-the-shelf general purpose ICs. A
full-custom ASIC cannot be modified to suit different applications, and it is generally produced as a single, specific product
for a particular application only.
(2) Semi-custom ASICs, on the other hand, can be partly customized
to serve different functions within their general area of application. Unlike full-custom ASICs, semi-custom ASICs are designed
to allow a certain degree of modification during the manufacturing process. A semi-custom ASIC is manufactured with the
masks for the diffused layers already fully defined, so the transistors and other active components of the circuit are already fixed
for that semi-custom ASIC design. The customization of the final
ASIC product to the intended application is done by varying the
masks of the interconnection layers, for example, the metallization layers.
(3) Structured or platform ASICs, which belong to a relatively new
ASIC classification, are those that have been designed and
produced from a tightly defined set of (1) design methodologies,
(2) intellectual properties, and (3) well-characterized silicon,
aimed at shortening the design cycle and minimizing the development costs of the ASIC. A platform ASIC is built from a group
of “platform slices,” with a platform slice being defined as a premanufactured device, system, or logic for that platform. Each
slice used by the ASIC may be customized by varying its metal
layers. The reuse of premanufactured and precharacterized platform slices simply means that platform ASICs are not built from
scratch, thereby minimizing design cycle time and costs.
There are two types of programmable ASIC: programmable logic
devices (PLD) and field-programmable gate arrays (FPGA). The distinction between the two is blurred. The only real difference is their heritage.
In this section, these two types of programmable ASICs are discussed,
respectively.
ASICs have been widely used in various industrial control applications.
Examples of ASICs include (1) an IC that encodes and decodes digital data
using a proprietary encoding and decoding algorithm, (2) a medical IC
designed to monitor a specific human biometric parameter, (3) an IC
designed to serve a special function within a factory automation system,
(4) an amplifier IC designed to meet certain specifications not available in
standard amplifier products, (5) a proprietary system-on-a-chip, and (6) an
IC that is custom-made for particular automated test equipment.
Zhang_Ch02.indd 241
5/13/2008 5:53:49 PM
242
INDUSTRIAL CONTROL TECHNOLOGY
2.3.1
ASIC Designs
ASICs are used to design entire systems on a single chip. ASICs are
interconnects of standard cells that have been standardized by fabrication
houses. With the integration of more and more system components on a
single IC, the complexity of IC fabrication has increased. An advanced
system design involves complex layout issues. Specifications of cells are
provided by the vendors in the form of a technology library that contains
information about geometry, delay, and power characteristics of cells.
Design flow of ASIC is highly automated. These automation tools
provide reasonable performance and cost advantage over manual design
process. Broadly, ASIC design flow can be divided into these phases given
below in this subsection, which is also illustrated in Fig. 2.17.
2.3.1.1 ASIC Specification
ASIC design specifications are written by designers at different levels of
abstraction. Most common hardware specification languages used by
designers are Verilog and VHDL. Both these languages are equally capable of providing complex constructs to describe complex functionality.
Behavioral modeling forms the highest level of abstraction.
(1) Behavioral specification. At the initial stage of the design
process, the designer provides a behavioral specification of
the functionality intended. The behavioral model does not care
about the structure of the design, combinational, or sequential
elements used in the design, clock signal, and the timing
Design specification
HDL (Verilog and VHDL)
Technology
library
Design rule
constraints
Syntheses
Simulation
Place and Route
Extraction
Manufacture
Post layout analysis
Figure 2.17 ASIC design flow.
Zhang_Ch02.indd 242
5/13/2008 5:53:50 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
243
constraints involved. It captures the intended behavior of the
design. It is important to note that this specification does not
capture timing information.
(2) RTL specification. RTL stands for register transfer level. In this
model the entire design is split into registers with flow of information between these registers at each clock cycle. RTL specification captures the change in design at each clock cycle. All the
registers are updated at the same time in a clock cycle. Typically
an RTL specification divides a design into registers and the logic
blocks that join those registers together. RTL captures the data
flow but fails to give a good specification of control flow.
(3) Structural specification. Structural specification consists of a
network of instances of logic gates and registers described by a
technology library. Technology library is provided by fabrication
houses. Technology library is a specification of simple AND,
OR, NOT, and complicated multiple functionality cells. The
specification of a cell includes its geometry, delay, and power
characteristics. Structural modeling describes circuits in the form
of instances of cells and interconnects between those cells.
2.3.1.2 ASIC Functional Simulation
Logic simulation is an essential part of digital circuit design. Logic simulation and verification are used to verify the functionality described by a
design specification against output values expected at the output ports of a
digital integrated circuit. There are mainly three classes of logic simulators
given below:
(1) Compiled code logic simulator. Compiled code logic simulation
algorithm evaluates every logic element in the design at each
time step. The earliest logic simulators were compiled code simulators. In a compiled code simulator a combinational circuit is
topologically ordered and an equation is generated for each gate
output in terms of its inputs using Boolean operators AND, OR,
and NOT. A compiled code simulator evaluates every circuit element for every new input pattern. Two limitations of compiled
code simulators are (1) their inability to handle asynchronous
feedback and (2) lack of accurate timing and delay information
in their models. Also since circuit activity is very low at each element, run time of such algorithm is huge for big circuits. A large
part of the design cycle is taken up by circuit simulation.
(2) Interpretive event-driven logic simulator. An event-driven logic
simulator works on the principle that output of a logic element
Zhang_Ch02.indd 243
5/13/2008 5:53:50 PM
244
INDUSTRIAL CONTROL TECHNOLOGY
changes only when one of its inputs changes. An event-driven
logic simulator evaluates a logic element when an event occurs
on one of its inputs. Statistical data shows that event activities in
large circuits are very low, and with increase in size of the circuit
percentage activity on a logic element decreases.
Each concurrent process is converted to an abstract syntax
tree. When a process is executed its tree is traversed, and nodes
of the tree are interpreted and acted upon till the process suspends. A process may suspend when it finishes or due to nonavailability of one of the inputs. A suspended process is scheduled
by the scheduler to wait for some event to wake it up. Execution
of some statements in a process may resume other processes.
There are two major classes of event-driven logic simulation
algorithms: (1) Synchronous simulation algorithms: These algorithms are centralized-timed and are used for single processor
machines. The algorithm follows the path of events in the circuit.
Simultaneous events are handled through centralized control of
time. In this scheme simulation does not advance until all the
events that occurred on current simulation time are processed. To
implement this algorithm one needs to store events in a global
ordered queue of events that is circular in nature. Each slot in this
queue represents simulation time and it stores a linked list
of events that occur at that simulation time. As the events of
the current time slot are processed, the output of those events is
compared with the previous output of corresponding logic
elements, and if they differ new events are generated on logic
elements whose input is driven by the output of current event.
(2) Asynchronous simulation algorithms: In asynchronous simulation there is no global centralized time. Instead each data item
carries a time stamp that is indicative of time up to which data is
valid. The evaluation of an event depends on availability of a
token. An asynchronous algorithm can process events that occur
at different time instances. Hence it can extract more parallelism
compared to synchronous simulation algorithms.
(3) Compiled-code event-driven simulation. Modern simulators
employ the best of both of the above. Compiled-code eventdriven simulators compile the hardware description language
(HDL) description of the design to machine code. The generated
code is then linked with the simulator kernel.
2.3.1.3 ASIC Synthesis
A design specification that is functionally correct is fed to a synthesis
tool that extracts finite state machine from the design and performs data
Zhang_Ch02.indd 244
5/13/2008 5:53:50 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
245
path optimizations on the finite state machine. The resultant hardware
specification is mapped to gates, flip-flops, and nets to get gate-level netlist.
User supplied timing constraints are used to perform timing optimization
on gate-level netlist. The first step in synthesis process is to convert a given
RTL into a finite state machine. Many transformations can be applied to
finite state machine in order to reduce the number of states. Some of the
common transformations applied to FSM are constant propagation, gate
merging, dead code elimination, and arithmetic merging. The next step is
to generate hardware.
RTL synthesis involves three major steps: transition from RTL description into gates and flip-flops, optimization of logic, and placement and
routing of optimized netlist. Most of the intelligence resides in the optimization stage but modern synthesis tools apply many smart techniques while
converting RTL description into gates in order to reduce the number of
gates in the design. There are broadly two types of optimizations: technology
independent optimizations that are carried out once netlist has been mapped
into technology cells provided by fabrication house and technologydependent optimization.
Timing and area constraints are provided by the designer. Slack is defined
as the difference between the expected arrival time and actual arrival time
of the signal at a particular output port. Slack is calculated for input to
output paths. The aim of timing optimization is to reduce slack on critical
paths. Certain timing optimizations might lead to area escalation. Area
reclamation algorithms try to reclaim the area that does not affect timing
on critical paths. Other design rule constraints are maximum fan-out for a
logic element, maximum capacitance, and the slope of signal from 20% of
target to 80% of target.
The technology library provided by the fabrication house contains basic
components like sequential gates: AND, OR, NOT, NAND, NOR, XOR,
BUFFER, and sequential elements like latches, flip-flops, and memories.
Information about cell characteristics includes cell delay and area. There
are three major quality metrics: area, time, and power. Designer’s quality
metric for an IC is driven by specific application.
(1) Area. With shrinking system size, ASIC should be able to accommodate maximum functionality in minimum area. The designer
can specify area constraint and the synthesis tool will optimize
for minimum area. Area can be optimized by having lesser number of cells and by replacing multiple cells with a single cell that
includes both functionalities.
(2) Timing. Designer specifies maximum delay between primary
input and primary output. This is taken as maximum delay across
Zhang_Ch02.indd 245
5/13/2008 5:53:50 PM
246
INDUSTRIAL CONTROL TECHNOLOGY
any critical path. There are four types of critical paths: path
between a primary input and primary output, path from any primary input to a register, path from a register to a primary output,
and path from a register to another register.
(3) Power. Development of handheld devices has led to reduction of
battery size and hence low power consuming systems. Low power
consumption has become a big requirement for lots of designers.
2.3.1.4 ASIC Design Verification
The biggest challenge in IC design is verification because the cost of a
single error is huge. Verification is time consuming and requires a large
amount of resources. The types of verification tasks can be classified into
two categories: (1) Functional verification that checks the functionality of
synthesized and optimized design against golden representation of design.
(2) Implementation verification that occurs once placement and routing is
over. In implementation verification, the design is checked for functional
correctness once again. Timing and power constraints are also verified.
Formal verification methods are used to test the functional correctness
of the gate-level netlist. Testing functional correctness involves testing an
optimized design against a golden design specification. There are two
methods of performing the verification: (1) Black box verification methods include simulation, emulation, and hardware acceleration. (2) White
box verification methods involve the use of formal methods, for example,
assertion-based verification.
(1) Assertion. Assertion-based verification is aimed at digital designers. It is a white-box verification technique. Unlike simulation, it
is not applied on the block level once the design is complete.
Assertion-based verification can be applied alongside design process. In fact assertion-based verification entities reside in the HDL
description of the design. Assertions are active comments embedded within the design. Assertions turn design specifications into
verification objects. Assertions can be used to (1) monitor signals on interfaces that connect different blocks; (2) track expected
behavior of a gate, flip-flop, or module; (3) watch for forbidden
behavior within a design block.
Assertions can be implicit or explicit: (1) Implicit assertions
are supported by HDL like Verilog and VHDL. These assertions
are added at the time of design analysis, synthesis, and HDL
analysis. (2) Explicit assertions are user-defined assertions. Such
assertions are provided by EDA vendors in the form of library
Zhang_Ch02.indd 246
5/13/2008 5:53:50 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
247
such as OVL. Academic languages like CTL, LTL, and automata
provide a way to define explicit assertions.
(2) Emulation. The emulator is a hardware device that can be used to
emulate a piece of hardware functionality. It is commonly used
as a debugging tool to test a system under development for functional correctness. Emulation is a faster solution to a verification
problem. In emulation a portion of the design is synthesized and
optimized. The compiled design is then loaded onto an emulator.
The rest of the design is simulated by the workstations that
are connected to the emulator. Remember only the portion of the
design that is being tested resides on the emulator. Emulators are
able to provide execution speed close to real time. This allows
verification engineers to reduce verification time. An emulation
system typically consists of a small number of large FPGA. This
provides multimillion ASIC-equivalent gate capacities. Such
an emulation system comes as a separate box. An emulation
box can be connected to a collection of workstations using PCI
card. The workstations are connected via emulation network
architecture.
A complex IC is typically divided into a number of different modules.
Each module is developed by a separate team of designers. Each team
verifies the functionality of its own module. The modules then go to an
integration team that integrates all the modules and carries out verification.
With emulation providing faster methods of design verification, last minute changes can be incorporated in the design. This significantly reduces
time to market.
2.3.1.5 ASIC Integrity Analyses
Due to an increase in signal speed, miniaturization of features, smaller
chip sizes, and lower power supply voltages, there has been greater interconnect signal integrity problems. Signal delay due to interconnect delay
is more significant compared to gate delay. As a result, more powerful
automation tools are required for layout parameter extraction, timing delay
and crosstalk simulation, and power analysis.
(1) Parasitic extraction. Accurate extraction of on-chip parasitic is
crucial due to shrinking size and the increasing contribution of
interconnect delay. The parasitic consists of resistance, inductance,
and capacitance. Inductance is not critical for signal propagation
until transmission line effect occurs. Resistance is easy to compute using algorithms like square counting and two-dimensional
Zhang_Ch02.indd 247
5/13/2008 5:53:50 PM
248
INDUSTRIAL CONTROL TECHNOLOGY
finite-difference approach. Another reason for easy resistance
estimation is that one has to consider only one conductor trace at
a time. On the other hand, capacitance extraction requires that
neighborhood conductors be considered for electromagnetic
coupling effect.
(2) Signal integrity. The design for a high speed integrated circuit
depends on understanding and predicting interconnect parasitic
effects and behavior. Increasing switching speed and complexity
of VLSI circuits are becoming crucial factors in determining reliability and performance of an electronic system. Estimation is
complex due to increase in metallization layers, increasing material complexity, and higher operating frequencies. The various
aspects of signal integrity include
(a) Technology scale down. As technology dips into the deep
submicrometer range, lateral coupling effects between interconnects dominate compared to vertical coupling effects
in micrometer technology. Aluminum has been used until
recently to manufacture interconnects but the increasing contribution of interconnects in signal propagation has forced IC
manufacturers to replace it with materials like copper with
lower receptivity. As a result, gain in propagation delay is
almost double. Technology scale down has introduced some
new problems like complex resistance, three-dimensional
capacitance, and inductance.
(b) Propagation delay. With decrease in the size of technology
interconnect, delay increases.
(c) Crosstalk. When two wire segments are closer to each other
than a minimum threshold, they will interfere in each other’s
functioning. A signal on one wire may weaken due to electromagnetic effects of a signal carried by the other wire. This
interference with each other’s signal is called crosstalk. With
diminishing technology size, crosstalk is a major contributor
to high speed IC defects.
(d) Crosstalk delay. Crosstalk delay is a major reason for timing
uncertainty. The simultaneous switching of the victim and
the affecting signals may lead to a wide variety of phenomena. Among the most important is delay increase when the
victim and aggressor signals switch in opposite directions,
starting with the victim signal followed by the aggressor.
2.3.2
Programmable Logic Devices (PLD)
PLD are designed with configurable logic and flip-flops linked together
with programmable interconnect. PLDs provide specific functions, including
Zhang_Ch02.indd 248
5/13/2008 5:53:50 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
249
device-to-device interfacing, data communication, signal processing, data
display, timing and control operations, and almost every other function a
system must perform. Memory cells control and define the function that
the logic performs and how the various logic functions are interconnected.
Logic devices can be classified into two broad categories: fixed and
programmable.
As the name suggests, the circuits in a fixed logic device are permanent;
they perform one function or a set of functions and once manufactured,
they cannot be changed. With fixed logic devices, the time required to go
from design to prototypes to a final manufacturing run can take from several months to more than a year, depending on the complexity of the device.
If the device does not work properly, or if the requirements change, a new
design must be developed.
On the other hand, PLD are standard, off-the-shelf parts that offer
customers a wide range of logic capacity, features, speed, and voltage
characteristics—these devices can be changed at any time to perform any
number of functions. With programmable logic devices, designers use
inexpensive software tools to quickly develop, simulate, and test their
designs. Then, a design can be quickly programmed into a device and
immediately tested in a live circuit. The PLD that is used for this prototyping is the exact same PLD that will be used in the final production of a
piece of end equipment, such as a network router, a DSL modem, a DVD
player, or an automotive navigation system. There are no NRE costs and
the final design is completed much faster than that of a custom, fixed logic
device. Another key benefit of using PLD is that during the design phase
customers can change the circuitry as often as they want until the design
operates to their satisfaction. That is because PLDs are based on rewriteable memory technology: to change the design, simply reprogram the
device. Once the design is final, customers can go into immediate production by simply programming as many PLDs as they need with the final
software design file.
PLDs can be described as being one of three different types: simple
programmable logic devices (SPLD), complex programmable logic
devices (CPLD), or field-programmable gate arrays (FPGA). The FPGA is
individually discussed in Section 2.3.3. The distinction between CPLD
and FPGA is often a little fuzzy, with manufacturers designing new,
improved architectures, and frequently muddying the waters for marketing
purposes. Together, CPLD and FPGA are often referred to as high-capacity programmable logic devices (HCPLD).
The programming technologies for PLD devices are actually based on
the various types of semiconductor memory. As new types of memories
Zhang_Ch02.indd 249
5/13/2008 5:53:50 PM
250
INDUSTRIAL CONTROL TECHNOLOGY
have been developed, the same technology has been applied to the creation
of new types of PLD devices. Today, SPLD are devices that typically contain the equivalent of 600 or fewer gates, while HCPLD have thousands
and hundreds of thousands of gates available. SPLDs are often used for
address decoding, where they have several clear advantages over the 7400series TTL parts that they replaced. First, of course, is that one chip requires
less board area, power, and wiring than several do. Another advantage is
that the design inside the chip is flexible, so a change in the logic does not
require any rewiring of the board. Rather, the decoding logic can be altered
by simply replacing that one PLD with another part that has been programmed with the new design. Hardware designs for these simple PLDs
are generally written in languages like ABEL or PALASM (the hardware
equivalents of assembly).
Inside each PLD is a set of fully connected macro cells. These macro
cells are typically comprised of some amount of combinatorial logic
(e.g., AND, OR gates) and a flip-flop. In other words, a small Boolean
logic equation can be built within each macro cell. This equation will combine the state of some number of binary inputs into a binary output and, if
necessary, store that output in the flip-flop until the next clock edge. Of
course, the particulars of the available logic gates and flip-flops are specific to each manufacturer and product family. But the general idea is
always the same.
Because these chips are rather small, they do not have much relevance
to the remainder of this discussion. But you do need to understand the
origin of programmable logic chips before we can go on to talk about
the larger devices. At the low end of the spectrum are the original programmable logic devices. These were the first chips that could be used to implement a flexible digital logic design in hardware. In other words, you could
remove a couple of the 7400-series transistor-transistor logic parts (AND,
OR, and NOT) from your board and replace them with a single PLD. Other
names you might encounter from this class of device are programmable
logic array (PLA), programmable array logic (PAL), and generic array
logic (GAL).
2.3.3
Field-Programmable Gate Array (FPGA)
Field-programmable gate arrays (FPGA) are integrated circuits (ICs)
that contain an array of logic cells surrounded by programmable I/O
blocks. FPGAs contain as many as tens of thousands of logic cells and an
even greater number of flip-flops. Because of cost, FPGAs do not provide
a 100% interconnection between logic cells; however, FPGAs still provide
Zhang_Ch02.indd 250
5/13/2008 5:53:50 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
251
significantly higher capacities than programmable logic devices (PLD) that
are interconnected through a central global routing pool. Often, design engineers use FPGA to program electrical connections through several iterations in order to minimize nonrecurring costs. FPGA are used in applications
ranging from data processing and storage, to instrumentation, telecommunications, and digital signal processing. Other terms for FPGA include
logic cell array and programmable application-specific integrated chip.
2.3.3.1
FPGA Types and Important Data
Selecting field-programmable gate arrays requires an analysis of
memory, performance, and I/O interface requirements. Available memory
types include content addressable memory; Flash, random access memory
(RAM); dual-port RAM; read-only memory (ROM); electrically erasable
programmable read-only memory (EEPROM); first-in, first-out; and last-in,
last-out. Performance considerations include internal frequency, the
number of integrated phase-locked loops and delay-locked loops with
clock-frequency-synthesis capabilities, and the total number of I/O ports.
I/O interfaces include accelerated graphics port, bus low voltage differential signaling, and peripheral component interconnect (PCI).
Field-programmable gate arrays are available with different numbers of
system gates, shift registers, logic cells, and lookup tables. Logic blocks or
logic cells do not include I/O blocks, but generally contain a lookup table
to generate any function of inputs, a clocked latch (flip-flop) to provide
registered outputs, and control logic circuits for configuration purposes.
Logic cells are also known as logic array blocks, logic elements, and
configurable logic blocks. Lookup tables or truth tables are used to implement a single logic function by storing the correct output logic state in a
memory location that corresponds to each particular combination of input
variables.
Field-programmable gate arrays are available with many logic families.
Transistor-transistor logic and related technologies such as Fairchild
advanced Schottky use transistors as digital switches. By contrast, emittercoupled logic uses transistors to steer current through gates that compute
logical functions. Another logic family, CMOS, uses a combination of
P-type and N-type metal-oxide-semiconductor field effect transistors to
implement logic gates and other digital circuits. Logic families for fieldprogrammable gate arrays include crossbar switch technology, gallium
arsenide, integrated injection logic, and silicon on sapphire. Gunning
with transceiver logic and gunning with transceiver logic plus are also
available.
Zhang_Ch02.indd 251
5/13/2008 5:53:50 PM
252
INDUSTRIAL CONTROL TECHNOLOGY
Field-programmable gate arrays are available in a variety of IC package
types and with different numbers of pins and flip-flops. Basic IC package
types for field-programmable gate arrays include ball grid array, quad flat
package, single in-line package, and dual in-line package. Many packaging variants are available.
2.3.3.2
FPGA Architecture
The typical basic architecture consists of an array of logic blocks and
routing channels. Multiple I/O pads may fit into the height of one row or
the width of one column. Generally, all the routing channels have the same
width (number of wires). Figure 2.18 illustrates the FPGA structure.
(1) FPGA logic block. An application circuit must be mapped into
an FPGA with adequate resources. The typical FPGA logic block
consists of a 4-input lookup table (LUT) and a flip-flop, as shown
in Fig. 2.19.
There is only one output in each logic block, which can be
either the registered or the unregistered LUT output. As shown in
Figure 2.20, the logic block has four inputs for the LUT and
a clock input. Since clock signals (and often other high-fanout
Routing
channel
I/O pad
Logic block
Figure 2.18 FPGA structure.
Inputs
4-input
LUT
Clock
D Flip
flop
Out
Figure 2.19 A FPGA logic block.
Zhang_Ch02.indd 252
5/13/2008 5:53:51 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
253
In3
In4
In2
Out
In1
Out
Figure 2.20 FPGA logic block pin locations.
signals) are normally routed via special-purpose dedicated routing networks in commercial FPGA, they are accounted for separately from other signals.
Each input is accessible from one side of the logic block, while
the output pin can connect to routing wires in both the channel to
the right and the channel below the logic block. Each logic block
output pin can connect to any of the wiring segments in the channels adjacent to it. Figure 2.21 should make the situation clear.
Similarly, an I/O pad can connect to any one of the wiring segments in the channel adjacent to it. For example, an I/O pad at the
top of the chip can connect to any of the W wires (where W is the
channel width) in the horizontal channel immediately below it.
(2) FPGA routing. Generally, the FPGA routing is unsegmented
(Fig. 2.22). That is, each wiring segment spans only one logic
block before it terminates in a switch box. By turning on some of
the programmable switches within a switch box, longer paths
can be constructed. For higher speed interconnect, some FPGA
architectures use longer routing lines that span multiple logic
blocks.
Potential
connection
Logic
block
pin
Routing wire
Figure 2.21 Logic block pin to routing channel interconnect.
Zhang_Ch02.indd 253
5/13/2008 5:53:51 PM
254
INDUSTRIAL CONTROL TECHNOLOGY
Logic block
Switch block
Wire segment
Figure 2.22 Unsegmented FPGA routing mechanism.
Whenever a vertical and a horizontal channel intersect there is
a switch box. In this architecture, when a wire enters a switch
box, there are three programmable switches that allow it to
connect to three other wires in adjacent channel segments. The
pattern, or topology, of switches used in this architecture is the
planar or domain-based switch box topology. In this switch box
topology, a wire in track number 1 connects only to wires in
track number 1 in adjacent channel segments, wires in track
number 2 connect only to other wires in track number 2, and so
on. Figure 2.23 illustrates the connections in a switch box.
Modern FPGA families expand upon the above capabilities to
include higher level functionality fixed into the silicon. Having
these common functions embedded into the silicon reduces the
area required and gives those functions increased speed compared to building them from primitives. Examples of these
include multipliers, embedded processors, high speed I/O logic,
and embedded memories.
Wire
segment
Programmable
switch
Figure 2.23 FPGA switch box topology.
Zhang_Ch02.indd 254
5/13/2008 5:53:51 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
2.3.3.3
255
FPGA Programming
To define the behavior of the FPGA the user provides a hardware
description language (HDL) or a schematic design. Common HDLs are
VHDL and Verilog. Then, using an electronic design automation tool, a
technology-mapped netlist is generated. The netlist can then be fitted to
the actual FPGA architecture using a process called place-and-route,
usually performed by the FPGA company’s proprietary place-and-route
software. The user will validate the map, place and route results via timing
analysis, simulation, and other verification methodologies. Once the design
and validation process is complete, the binary file generated (also using
the FPGA company’s proprietary software) is used to reconfigure the
FPGA device.
In an attempt to reduce the complexity of designing in HDLs, which
have been compared to the equivalent of assembly languages, there are
moves to raise the abstraction level of the design. Companies are promoting SystemC as a way to combine high level languages with concurrency
models to allow faster design cycles for FPGAs than is possible using
traditional HDL. Approaches based on standard C or C++ (with libraries
or other extensions allowing parallel programming) are found in the
Catapult C tools from Mentor Graphics, and in the Impulse C tools from
Impulse Accelerated Technologies. Languages such as SystemVerilog,
SystemVHDL, and Handel-C (from Celoxica) seek to accomplish the
same goal, but are aimed at making existing hardware engineers more
productive vs making FPGAs more accessible to existing software
engineers.
To simplify the design of complex systems in FPGAs, there exist libraries of predefined complex functions and circuits that have been tested and
optimized to speed up the design process. These predefined circuits are
commonly called intellectual property blocks and are available from FPGA
vendors, from third-party IP suppliers, and in the public domain via
OpenCores.org and other sources.
In a typical design flow, an FPGA application developer will simulate
the design at multiple stages throughout the design process. Initially, the
RTL description in VHDL or Verilog is simulated by creating test benches
to stimulate the system and observe results. Then after the synthesis engine
has mapped the design to a netlist, the netlist is translated to a gate-level
description where simulation is repeated to confirm the synthesis proceeded without errors. Finally, the design is laid out in the FPGA at which
point propagation delays can be added and the simulation run again with
these values back-annotated onto the netlist.
Zhang_Ch02.indd 255
5/13/2008 5:53:52 PM
256
INDUSTRIAL CONTROL TECHNOLOGY
Bibliography
Altera (http://www.altera.com). 2005. C8237 DMA Controller. http://www.cast-inc
.com/cores/c8237/cast_c8237-a.pdf. Accessed date: June.
Best-Microprocessor (http://www.best-microcontroller-projects.com). 2005. http://
www.best-microcontroller-projects.com/hardware-interrupt.html. Accessed
date: June.
Buchanan, William and Wilson, Austin. 2001. Advanced PC Architecture. Reading,
England: Addison-Wesley.
Chipx (http://www.chipx.com). 2007a. Structured ASIC. http://www.chipx.com/
about/index.asp. Accessed date: May.
Chipx (http://www.chipx.com). 2007b. Embedded ASIC. http://www.chipx.com/
embedded/index.asp. Accessed date: May.
Datorarkitektur, I. 2004. Input/Output Device. http://www.ida.liu.se/~TDTS57/
info/lectures/lect4.frm.pdf. Accessed date: June 2005.
DCD (http://www.dcd.pl). 2005. http://www.dcd.pl/dcdpdf/xil/d8255_ds.pdf.
Accessed date: June.
DDJ (http://www.ddj.com). 2005. http://www.x86.org/intel.doc/686manuals.htm.
Accessed date: June.
Decision Cards (http://www.decisioncards.com). 2005. http://www.decisioncards
.com/io/index.php?style=isol. Accessed date: June.
DELORIE (http://www.delorie.com). 2005. http://www.delorie.com/djgpp/doc/
ug/interrupts/inthandlers1.html. Accessed date: June.
Design-Reuse (http://www.us.design-reuse.com). 2007. FPGA Design Articles.
http://www.us.design-reuse.com/articles/fpga-0.html. Accessed date: May.
FPGA-ASIC Engineering (http://www.ge-research.com/index.html). 2007. ASIC
design Flow. http://www.ge-research.com/atmel_design.html. Accessed date:
May.
FtdiChip (http://www.ftdichip.com). 2007. ASIC Design Service. http://www
.ftdichip.com/DesignServices/ASICDesign.htm. Accessed date: May.
Fujitsu (http://www.fujitsu.com). 2007. Standard Cell ASIC. http://www.fujitsu
.com/emea/services/microelectronics/asic/standardcell/index.html. Accessed
date: May.
Gaonkar, Ramesh S. 2002. Microprocessor Architecture, Programming, and
Applications with the 8085. Fifth Edition. New Jersey: Prentice Hall.
IECI (http://www.ieci.com). 2005. http://www.ieci.com.au/products/data_
acquisition_index.asp. Accessed date: June.
InfoMit (http://www.informit.com). 2005. http://www.informit.com/articles/article
.asp?p=482324&seqNum=3&rl=1. Accessed date: June.
Intel (http://www.intel.com). 2005a. http://www.intel.com/pressroom/kits/
quickreffam.htm. Accessed date: June.
Intel (http://www.intel.com). 2005b. 8259A Programmable Interrupt Controller.
http://bochs.sourceforge.net/techspec/intel-8259a-pic.pdf.gz. Accessed date:
June.
Intel (http://www.intel.com). 2005c. 82C54 Programmable Interval Timer. http://
bochs.sourceforge.net/techspec/intel-82c54-timer.pdf.gz. Accessed date: June.
Intel (http://www.intel.com). 2005d. 8237A Programmable DMA Controller. http://
bochs.sourceforge.net/techspec/intel-8237a-dma.pdf.gz. Accessed date: June.
Zhang_Ch02.indd 256
5/13/2008 5:53:52 PM
2: COMPUTER HARDWARE FOR INDUSTRIAL CONTROL
257
Interface Bus (http://www.interfacebus.com). 2007. Programmable Logic Device
Definition. http://www.interfacebus.com/Programmable_Logic.html. Accessed
date: May.
Intersil (http://www.intersil.com). 2005. http://www.ortodoxism.ro/datasheets/
intersil/fn2970.pdf. Accessed date: June.
MITRE (http://www.mitre.org). 2005. Real-Time Clock and CMOS. http://www
.mitre.org/tech/cots/RTC.html. Accessed date: June.
Netrino (http://www.netrino.com). 2007. SPLD CPLD PLD. http://www.netrino
.com/Publications/index.php. Accessed date: May.
Patterson, David A. and Hennessey John L. 1998. Computer Organization and
Design. Second Edition. San Francisco: Morgan Kaufmann.
Real Time Devices (http://www.rtdfinland.fi). 2005. DM5854HR/DM6854HR
Isolated Digital I/O-Module User’s Manual. Available from this company’s
web site.
Rieker, Mike 2005. Advanced Programmable Interrupt Controller. http://www
.osdever.net/tutorials/pdf/apic.pdf. Accessed date: June.
Silicon-Fareast (http://www.siliconfareast.com). 2005. Application-Specific
Integrated Circuits. http://www.siliconfareast.com/asic.htm. Accessed date:
June.
Styles, Lain 2004. Lecture 11: I/O & Peripherals. http://www.cs.bham.ac.uk/
internal/courses/sys-arch/current/lecture11.pdf. Accessed date: June 2005.
Tutorial-Report (http://www.tutorial-reports.com). 2007a. ASIC Design Guide.
http://www.tutorial-reports.com/hardware/asic/tutorial.php. Accessed date: May.
Tutorial-Report (http://www.tutorial-reports.com). 2007b. ASIC Simulation
and Synthesis. http://www.tutorial-reports.com/hardware/asic/synthesis.php.
Accessed date: May.
Tutorial-Report (http://www.tutorial-reports.com). 2007c. FPGA Tutorials. http://
www.tutorial-reports.com/computer-science/fpga/tutorial.php?PHPSESSID=
c3c8bde5dfc3cd38afdc175639f587c7. Accessed date: May.
Wikipedia (http://en.wikipedia.org/wiki/Main_Page). 2005. Standard Cell ASIC.
http://en.wikipedia.org/wiki/Standard_Cell. Accessed date: June.
Xilinx (http://www.xilinx.com). 2007a. PLD Specifications. http://www.altium
.com/Forms/libraries/dxp2002/library_contents.asp?zip=Xilinx%5F050503
%2Ezip&lib=Xilinx+PLD+XC7000%2EIntLib. Accessed date: May.
Xilinx (http://www.xilinx.com). 2007b. PLD-SPLD CPLD. http://www.xilinx
.com/index.shtml. Accessed date: May.
Zilog (http://www.zilog.com). 2005. http://www.zilog.com/docs/z180/ps0140.pdf.
Accessed date: June.
Zhang_Ch02.indd 257
5/13/2008 5:53:52 PM
Zhang_Ch02.indd 258
5/13/2008 5:53:52 PM
3
System Interfaces for
Industrial Control
3.1 Actuator–Sensor (AS) Interface
3.1.1
Overview
Industrial automation systems require large amounts of control devices,
and the number of binary actuators and sensors on a typical system has
increased over the years. Conventional input and output (I/O) methods for
wiring include point-to-point connection or bus systems. For example,
typical batching valve wiring networks attach each of the I/O to a central
location resulting in multiple wire runs for each field device. Large expenditures are needed for cabling conduit, installation, and I/O points. Space
for I/O racks and cabling must be accommodated in order to attach only a
few field devices. These methods can prove to be too complex for networking simple binary devices and, therefore, too slow for the interactions
between the controllers and the controlled devices. Point-to-point wiring
is the most common method of wiring in the industry, but large wire bundles take up valuable space, installation is time consuming, and troubleshooting is complex.
Actuator–sensor interface, or AS interface, was developed by a group of
sensor manufacturers and introduced into the market in 1994. Since that
time, it has become the standard for discrete sensors in process industries
throughout the world. AS interface is also a bus system for low-level field
applications in industrial automation to communicate with small binary
sensors and actuators using the AS interface standard. The AS interface
modernizes automation systems effectively and eliminates wire bundles
completely, with only one wire cable required, compared to one cable
from each device with point-to-point wiring. Junction boxes are also eliminated and the size of the control cabinet needed is significantly reduced.
The plug and play wiring supports all typologies. Figure 3.1 gives the
locations of the AS interface in industrial control networks.
In comparison with conventional I/O wiring methods, the AS interface
has many advantages. The most important ones are given below:
(1) Minimum wiring and cost saving. AS interface offers a single
cable, which uses simple serial connection to the controller,
instead of parallel with a multitude of cables.
259
Zhang_Ch03.indd 259
5/13/2008 5:41:14 PM
260
INDUSTRIAL CONTROL TECHNOLOGY
Control level
Master
PLC, PC, SCADA, PID, ……
Field level
(Profibus, Foundation Fieldbus, CAN bus,
Ethernet, etc.)
AS interface level
Slave
device
Slave
device
Slave
device
Slave
device
……
Actuators, sensors level
Figure 3.1 The functionality of AS interface in industrial control networks.
The AS interface can be at two locations in an industrial control network:
between the controllers and the actuators–sensors and between the field level
and actuators–sensors.
(2) Fast and safe installation. Sensors and actuators are simply
installed with modules on the AS interface cable. Contact pins in
the modules penetrate the insulation of the cable and establish
contact with the copper wire. Incorrect connections are practically impossible because of the design of the cable and the special piercing method.
(3) Flexible configuration. Owing to the distributed and modular
design, plant sections can be tested in parallel even before the
overall solution is finished. This permits flexible modification
and expansion.
(4) Open system. AS interface is an open system, which means that
it is independent of manufacturer and future-proof.
3.1.2
Architectures and Components
In industrial control networks, as displayed in Fig. 3.1, there are two
types of AS interface architectures.
Zhang_Ch03.indd 260
5/13/2008 5:41:14 PM
261
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
3.1.2.1 AS Interface Architecture: Type 1
In the first type of AS interface architecture, a controller such as PLC,
SCADA, or PC applies the controls of sensors and actuators via the field
level buses including Fieldbus, PROFIBUS, etc.
As displayed in Fig. 3.2, the AS interface has gateways directly connected to the field level bus and the I/O module. The I/O module is the
device of this AS interface architecture to contact the sensors and the actuators. A field level bus may be able to support several AS interface gateways depending on the system designs, each of which profiles a segment
of an industrial control system.
In this type of architecture, AS interface requires the following
components:
(1) Gateways. Gateways are interface modules between the AS
interface and a higher level bus system. They are used when
more complex applications are to be solved using standard products. AS interface gateways are the core of the wiring system,
which handles the complete data transfer, cyclically polling
(master/slave) all participants connected to the wiring system.
The AS interface gateway can be placed anywhere in the AS
interface segments. One gateway can handle 124 inputs and 124
outputs over 31 addressable I/O modules. For gateways, setup is
accomplished through the setup tools for the respective system.
.
Controller or PC
Field level bus
AS interface
Gateway
AS interface
I/O module
AS interface
Gateway
AS interface
I/O module
AS interface
Power
supply/repeater
AS interface
Power
supply/repeater
AS interface
I/O module
. . . . . .
Sensors or actuators
Figure 3.2 AS interface architecture: Type 1.
Zhang_Ch03.indd 261
5/13/2008 5:41:15 PM
262
INDUSTRIAL CONTROL TECHNOLOGY
(2) I/O modules. I/O modules are the interface between standard
sensors and actuators and AS interface. I/O modules are available
for any kind of application, including flat modules for limited
space applications, compact modules for a variety of mounting
options, field modules that use cord grips instead of quick disconnects, standard modules that use both the mechanically keyed
AS interface cable and the standard 16 AWG round cable. For
enclosures and junction boxes, enclosure modules connect AS
interface bus to a power rail system, and junction box modules
for use within junction boxes.
(3) Power supplies and repeaters. With AS interface, one single
cable transmits both power and data. Power supplies contain
internal data separation coils so that the capacitive filtering of the
supply does not interfere with the data stream. Adding to the
high interference immunity of AS interface is the power supply
data isolation coil between the voltage transformer and the output so that the data signals are isolated from line noise. Repeaters
extend AS interface networks up to 100 additional meters, and
by using two in series, an AS interface network can be up to 300 m
long. Repeaters do not require a network address and allow I/O
modules to be placed anywhere along the network.
(4) AS interface safety at work. This extension of AS interface allows
for safety equipment to be wired together on a two-wire cable
rather than hardwired back to a panel. A maximum of 31 category 4 inputs, for example, E-Stops, can be put on one cable. The
parts included in this section are safety slaves, safety monitor,
and configuration software.
In many applications safety relevant functions are to be guaranteed. These are in the form of emergency stop buttons near
process lines or the implementation of safe sensors (e.g., safe
photo grits and locking of safety related doors) to automatically
stop machines. Depending on the safety category, there are
requirements of differing severity. Typically, a separate wiring is
necessary as well as redundancy or increased protection for the
cables.
With the integration of the safety technique into the AS interface line under the terminology “AS interface safety at work,”
the additional costs can be drastically reduced. The concept designates connection of the safety related switches by safe AS
interface module. There is also a safety monitor on the AS interface line permanently observing the communication. The communication happens through a given and predetermined pattern
by a dynamic code table with 8 times 4 bits sequence. The safe
monitor continuously checks “must” and “actual” values of the
Zhang_Ch03.indd 262
5/13/2008 5:41:15 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
263
communication through comparison. In case of the bit sequence
“0000,” the safe monitor switches off the safe relay in less than
40 ms. Several safe monitors can be operated in one AS interface
line arranged on any position.
Clear benefits of AS interface safety at work include the following: only one AS interface line is required for the communication of safe and nonsafe data; full compatibility with all
standard AS interface devices; no specific communication mechanisms required; mixed applications on one and the same AS
interface line possible; diagnosis of the safe modules via the
standard AS interface master possible.
(5) AS interface encoders. In order to be able to meet the real-time
requirements of many applications, a “multislave” solution was
achieved. The position value, up to 16 bits in length, is transferred
to the gateways within a single cycle, via the four integrated AS
interface chips used for control purposes. AS interface rotary
encoders include 13-bit-Singleturn and 16-bit-Multiturn.
(6) Accessories. To make AS interface perfect and to make the
installation as easy as possible, various accessories ranging from
hand-held addressing devices to mounting bases to simulators
for higher level bus systems are offered: sealing for flat cable and
adaptor to round cable.
3.1.2.2 AS Interface Architecture: Type 2
In the second type of AS interface architecture, as shown in Fig. 3.3, the
AS interface master module resides inside a controller such as PLC,
SCADA, or PC. In this type of AS interface architecture, the AS interface
master terminal enables the direct connection of AS interface slaves.
The AS interface compliant interface supports digital and analog slaves.
The AS interface master does not manage the sensors and actuators via the
field level buses, but rather via the AS interface slave modules or cables.
The slave modules are connected to each other by means of the AS interface cable which can be branched with the cable branch device. Power
supply and repeater are used too. A group of slave modules frames a segment of an industrial control network with one interface cable. The AS
interface master module is able to support several segments depending on
the designed system capabilities.
In this type of architecture, AS interface requires the following
components:
(1) AS interface masters. The AS interface master automatically
controls all communication over the AS interface cable without
Zhang_Ch03.indd 263
5/13/2008 5:41:15 PM
264
INDUSTRIAL CONTROL TECHNOLOGY
Controller or PC
AS interface cable
AS interface
Master module
AS
interface
Power
supply
/repeater
AS interface
AS interface
Slave
module
Sensors/actuators
Sensors/actuators
Segment A
Cable
. . . . .
Cable
Cable
Branch
Cable
AS interface
Slave module
Segment B
Figure 3.3 AS interface architecture: Type 2.
the need for special software. The master can connect the system
to a controller such as PLC, SCADA, or PC, act as a standalone
controller, or serve as a gateway to higher level bus systems.
There exist the following AS interface masters in the current
markets:
(a) Standard AS interface master. Up to 31 standard slaves or
slaves with the extended addressing mode can be attached to
standard AS interface masters.
(b) Extended AS interface masters. The extended AS interface
masters support 31 addresses that can be used for standard
AS interface slaves or AS interface slaves with the extended
addressing mode. AS interface slaves with the extended
addressing mode can be connected in pairs (programmed as
A or B slaves) to an extended AS interface master and can
use the same address. This increases the number of addressable AS interface slaves to a maximum of 62. Due to the
address expansion, the number of binary outputs is reduced
to three per AS interface slave on slaves using the extended
addressing mode.
Zhang_Ch03.indd 264
5/13/2008 5:41:16 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
265
(2) AS interface slaves. All the nodes that can be addressed by an AS
interface master are defined as AS interface slaves.
(a) AS interface slave assembly system. AS interface slaves with
the following assembly systems are available:
(i) AS interface modules. AS interface modules are AS
interface slaves to which up to four conventional sensors
and up to four conventional actuators can be connected.
The standard coupling module, which is the lower section of a standard device, connects the user module to
the yellow AS interface cable. The user module connects the sensors and actuators, while the application
modules connect via screw terminals or connectors.
Sensors and actuators with a built-in AS interface chip
can be directly connected to the AS interface cable.
(ii) Sensors/actuators with an integrated AS interface connection. Sensors/actuators with an integrated AS interface
connection can be connected directly to the AS interface.
(b) Addressing mode. AS interface slaves are available with the
following addressing modes:
(i) Standard slaves. Standard slaves each occupy one
address on the AS interface. Up to 31 standard slaves
can be connected to the AS interface.
(ii) Slaves with an extended addressing mode (A/B slaves).
Slaves with an extended addressing mode can be operated in pairs at the same address with an extended AS
interface master. This doubles the number of addressable AS interface slaves to 62.
One of these AS interface slaves must be programmed
as an A slave using the addressing unit and the other as
a B slave. Due to the address expansion, the number of
binary outputs is reduced to three per AS interface slave.
Slaves can also be operated with a standard AS interface
master. For more detailed information about these functions, refer to the AS interface master discussion in the
previous paragraphs.
(c) Analog slaves. Analog slaves are special AS interface standard
slaves that exchange analog values with the AS interface
master. Analog slaves require special program sections in the
user program (drivers, function blocks) that execute the
sequential transfer of analog data. Analog slaves are intended
for operation with extended AS interface masters. The
extended AS interface masters handle the exchange of analog data with these slaves automatically. No special drivers
or function blocks are required in the user program.
Zhang_Ch03.indd 265
5/13/2008 5:41:16 PM
266
INDUSTRIAL CONTROL TECHNOLOGY
(3) Further AS interface system components. The further AS interface components include AS interface cable, AS interface power
supply unit, addressing unit, and SCOPE for AS interface.
(a) AS interface cable. The trapezoidal AS interface cable is recommended over standard two-wire round cable for quick
and simple connection of slaves. The AS interface cable is
available in different colors to signify its voltage rating with
color assignments as follows:
(i) Yellow. The yellow AS interface cable is used for data
and control power between the master and its slaves.
(ii) Black. It is the external output power cable up to
60 VDC.
(iii) Red. It is the external output power cable up to 240 VAC.
The AS interface cable, designed as an unshielded
two-wire cable, transfers signals and provides the power
supply for the sensors and actuators connected using
AS interface modules. Networking is not restricted to
one type of cable. If necessary, appropriate modules or
“T pieces” can be used to change to a simple two-wire
cable.
(b) AS interface power supply unit. The AS interface power
supply unit supplies power to the AS interface nodes connected
to the AS interface cable. For actuators with particularly high
power requirements, the connection of an additional load
power supply may be necessary (e.g., using special application modules). Data and control power are normally transmitted simultaneously via the AS interface cable. Power for
the electronics and inputs is supplied by a special AS interface power supply that feeds a symmetrical supply voltage
into the AS interface cable via a data-decoupling device.
(c) Addressing unit. The addressing unit allows simple programming of AS interface slave addresses.
(d) SCOPE for AS interface. SCOPE AS interface is a monitoring
program for Windows that can record and evaluate the data
exchange in AS interface networks during the commissioning phase and during operation. SCOPE AS interface can be
operated on a PC under Windows in conjunction with an AS
interface master communications processor.
3.1.3 Working Principle and Mechanism
AS interface utilizes a single, trapezoidal, unshielded two-wire cable,
which eliminates the extensive parallel control wiring required with most
installations. In a network with AS interface, a simple gateway interfaces
Zhang_Ch03.indd 266
5/13/2008 5:41:16 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
267
the network into the field communication bus. Data and power are transferred over the two-wire network to each of the AS interface compatible
field devices. The existing controller sees AS interface as remote I/O;
therefore, AS interface connects to the existing network with minimal
programming changes. The AS interface system utilizes only one master
per network to control the exchange of data. This allows the master to
interrogate up to 31 slaves and update all I/O information within 5 ms
(10 ms for 62 slaves). For slave connection, an insulated two-wire cable is
recommended to prevent reversing polarity. The electrical connection is
made using contacts that pierce the insulation of the cable, contacting the
two wires, thus eliminating the need to strip the cable and wire to screw
terminals. For data exchange to occur, each slave must be programmed
with an address that is stored internally in nonvolatile memory and remains
even after power is removed.
The tasks and functions of an AS interface master are described below,
which is important for understanding the functions, modes, and interfaces
available with the AS interface master modules.
3.1.3.1
Master–Slave Principle
The AS interface operates on the master–slave principle. This means
that the AS interface master connected to the AS interface cable controls
the data exchange with the slaves via the interface to the AS interface
cable.
Figure 3.4 illustrates the two interfaces of the AS interface master communication processor.
(1) The process data and parameter assignment commands are transferred via the interface between the master CPU and the master
communication processor. The user programs have suitable function calls and mechanisms available for reading and writing via
this interface.
(2) Information is exchanged with the AS interface slaves via the
interface between the master communication processor and AS
interface cable.
(1) Tasks and functions of the AS interface master. The AS interface
master specification distinguishes masters with different ranges
of functions known as a “profile.” For standard AS interface masters and extended AS interface masters, there are three different
master classes (M0, M1, M2 for standard masters, and M0e,
M1e, M2e for extended masters). The AS interface specification
Zhang_Ch03.indd 267
5/13/2008 5:41:16 PM
268
INDUSTRIAL CONTROL TECHNOLOGY
PLC/PC
CPU
User
program
AS interface
master
communication
processor
AS interface
slave
I/O
Configuration
Parameters
Interface
Address
Figure 3.4 AS interface operation.
stipulates the functions a master in a particular class must be able
to perform.
The profiles have the following practical significance:
(a) Master profile M0/M0e. The AS interface master can
exchange I/O data with the individual AS interface slaves.
The station configuration on the cable, called the “expected
configuration,” is used to configure the master.
(b) Master profile M1/M1e. This profile covers all the functions
according to the AS interface master specification.
(c) Master profile M2/M2e. The functionality of this profile corresponds to master profile M0/M0e, but in this profile the AS
interface master can also assign parameters to the AS interface slaves. The essential difference between extended AS
interface masters and standard AS interface masters is that
they support the attachment of up to 62 AS interface slaves
using the extended addressing mode. Extended AS interface
masters also provide particularly simple access for AS interface analog slaves complying with profile specifications.
However, if standard operation (master profile M0) is chosen for use, the following contents can be skipped.
(2) How an AS interface slave functions
(a) Connecting to the AS interface cable. The AS interface slave
has an integrated circuit (AS interface chip) that provides the
attachment of an AS interface device (sensor/actuator) to the
common bus cable to the AS interface master. The integrated
circuit contains these components: four configurable data
inputs and outputs; four parameter outputs.
Zhang_Ch03.indd 268
5/13/2008 5:41:16 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
269
The operating parameters, configuration data with I/O
assignment, identification code, and slave address are stored
in additional memory (e.g., EEPROM).
(b) I/O data. The useful data for the automation components
that were transferred from the AS interface master to the AS
interface slave are available at the data outputs. The values at
the data inputs are made available to the AS interface master
when the AS interface slave is polled.
(c) Parameter. Using the parameter outputs of the AS interface
slave, the AS interface master can transfer values that are not
interpreted as simple data. These parameter values can be
used to control and switch over between internal operating
modes of the sensors or actuators. It could, for example, be
possible to update a calibration value during the various
operating phases. This function is possible with slaves with
an integrated AS interface connection providing they support the function in question.
(d) Configuration. The input/output configuration (I/O configuration) indicates which data lines of the AS interface slave
are used as inputs, outputs, or as bidirectional outputs. The
I/O configuration (4 bits) can be found in the description of
the AS interface slave. In addition to the I/O configuration,
the type of the AS interface slave is described by an identification code; with newer AS interface slaves it is identified by
three identification codes (ID code, ID1 code, ID2 code).
For more detailed information on the ID codes, refer to the manufacturer’s description.
3.1.3.2
Data Transfer
(1) Information and data structure. Before introducing the operating
phases and the functions during these operating phases, a brief
outline of the information structure of the AS interface master/
slave system is necessary.
In Fig. 3.5, the data fields and lists of the system are configured in the system structure diagram as given in Fig. 3.4.
The following structures are found on the AS interface master:
(a) Data images. These contain temporarily stored information:
(i) actual parameters that are an image of the parameters
currently on the AS interface slave;
(ii) actual configuration data that contains the I/O configurations and ID codes of all connected AS interface
Zhang_Ch03.indd 269
5/13/2008 5:41:16 PM
270
INDUSTRIAL CONTROL TECHNOLOGY
PLC/PC
CPU
AS interface master
communication
processor
User
program
Data images
I/O data
AS interface
slave
I/O data
Parameters
Active parameters
Active
configuration
data
Configuration
Address
LDS
LAS
Configuration data
Expected
configuration
data
Parameters
LPS
Figure 3.5 Data transfer between the AS interface master and the AS interface
slave.
slaves once these data have been read from the AS interface slaves;
(iii) the list of detected AS interface slaves (LDS) that specifies which AS interface slaves were detected on the AS
interface bus;
(iv) the list of activated AS interface slaves (LAS) that specify which AS interface slaves were activated by the AS
interface master. I/O data are only exchanged with activated AS interface slaves.
(b) I/O data. The I/O data are the process input and output
data.
Zhang_Ch03.indd 270
5/13/2008 5:41:17 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
271
(c) Configuration data. These are nonvolatile data (e.g., stored
in an EEPROM), which are available unchanged even following a power failure.
(i) Expected configuration data that are selectable comparison values which allow the configuration data of the
detected AS interface slaves to be checked.
(ii) List of permanent AS interface slaves (LPS) that specifies the AS interface slaves expected on the AS interface
cable by the AS interface master. The AS interface master checks continuously whether all the AS interface
slaves specified in the LPS exist and whether their configuration data match the expected configuration data.
The AS interface slave has the following structures:
(a) I/O data
(b) Parameters
(c) Actual configuration data. The configuration data include
the I/O configuration and the ID codes of the AS interface
slave.
(d) Address. The AS interface slaves have address “0” when
installed. To allow a data exchange, the AS interface slaves
must be programmed with addresses other than “0.” The
address “0” is reserved for special functions.
(2) The operating phases. Figure 3.6 illustrates the individual operating phases.
(a) Initialization mode. The initialization mode, also known as the
offline phase, sets the basic status of the master. The module is
initialized after switching on the power supply or following a
restart during operation. During the initialization, the images
of all the slave inputs and the output data from the point of
view of the application are set to the value “0” (inactive).
After switching on the power supply, the configured parameters are copied to the parameter field so that subsequent
activation uses the preset parameters. If the AS interface
master is reinitialized during operation, the values from the
parameters field that may have changed in the meantime are
retained.
(b) Start-up phase.
(i) Detection phase. Detection of AS interface slaves in the
startup phase.
During startup or after a reset, the AS interface master runs through a startup phase during which it detects
which AS interface slaves are connected to the AS interface cable and what type of slaves these are. The “type”
of the slaves is specified by the configuration data stored
Zhang_Ch03.indd 271
5/13/2008 5:41:17 PM
272
INDUSTRIAL CONTROL TECHNOLOGY
Initialization
Offline phase
Startup phase
Detection phase
Activation phase in the
protected mode
Activation phase in the
configuration mode
“Startup with configured data”
“Startup without configured
data/obtain configuration data”
Normal operation
Data exchange phase
Management phase
Inclusion phase
Figure 3.6 How do the individual operating phases work in the data transfer
through an AS interface.
permanently on the AS interface slave when it is manufactured and can be queried by the master. Configuration
files contain the I/O assignment of an AS-I slave and
the slave type (ID codes). The master enters detected
slaves in the list of detected slaves (LDS).
(ii) Activation phase: Activating AS interface slaves. After
the AS interface slaves are detected, the master sends a
special call which activates them. When activating individual slaves, a distinction is made between two modes
on the AS interface master:
Master in the configuration mode: All detected stations (with the exception of the slave with address “0”)
are activated. In this mode, it is possible to read actual
values and to store them for a configuration.
Master in the protected mode: Only the stations corresponding to the expected configuration stored on the
AS interface master are activated. If the actual configuration found on the AS interface cable differs from this
Zhang_Ch03.indd 272
5/13/2008 5:41:17 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
273
expected configuration, the AS interface master indicates this.
The master enters activated AS interface slaves in the
list of activated slaves (LAS).
(iii) Normal mode. On completion of the startup phase, the
AS interface master switches to the normal mode.
(iv) Data exchange phase. In the normal mode, the master
sends cyclic data (output data) to the individual AS
interface slaves and receives their acknowledgment
messages (input data). If an error is detected during the
transmission, the master repeats the appropriate poll.
(v) Management phase. During this phase, all existing jobs
of the control application are processed and sent.
Possible jobs are, for example, as follows:
Parameter transfer: Four parameter bits (three parameter bits with AS interface slaves with the extended
addressing mode) are transferred to a slave and are
used, for example, for a threshold value setting.
Changing slave addresses: This function allows the
addresses of AS interface slaves to be changed by the
master if the AS interface slave supports this particular
function.
(vi) Inclusion phase. In the inclusion phase, newly added
AS interface slaves are included in the list of detected
AS interface slaves and, providing the configuration
mode is selected, they are also activated (with the exception of slaves with address “0”). If the master is in the
protected mode, only the slaves stored in the expected
configuration of the AS interface master are activated.
With this mechanism, slaves that were temporarily out
of service are also included again.
(3) Interface functions. To control the master and slave interaction
from the user program, there are various functions available on the
interface. The possibilities are explained below. The possible operations and the direction of data flow are illustrated in Fig. 3.7.
(a) Read/write. When writing, parameters are transferred to the
slave and the parameter images on the communication processor; when reading, parameters are transferred from the
slave or from the communication processor parameter image
to the CPU.
(b) Read and store (configured) configuration data. Configured
parameters or configuration data are read from the nonvolatile memory of the communication processor.
Zhang_Ch03.indd 273
5/13/2008 5:41:17 PM
274
INDUSTRIAL CONTROL TECHNOLOGY
AS interface
master
CPU
AS interface
slave
Communication processor
Process
image
IO data
Data images
I/O data
I/O data
1.Read/write
User
program
Activation
parameters
Parameters
Activation
configuration data
Configuration data
4. Supply slaves with
configuration parameters
(activation)
LDS
LAS
2.
Read/store configuration data
3. Configure actual
Configuration data (EEPROM)
Expected configuration data
Parameters
LAS
Figure 3.7 How does the AS interface function.
(c) Configure actual. When reading, the parameters and configuration data are read from the slave and stored permanently
on the communication processor; when writing, the parameters and configuration data are stored permanently on the
communication processor.
(d) Supply slaves with configured parameters. Configured parameters are transferred from the nonvolatile area of the communication processor to the slaves.
(4) Operating extended AS interface slaves with standard AS interface masters. The following information is about operating
extended AS interface with standard AS interface masters.
(a) Slaves are connected to standard masters. The most significant slave bit (bit 4) of each A slave must be set to “0.” The
most significant parameter bit (bit 4) must also be set to “1”
(default value). Without these settings, the A slave cannot be
operated with a standard master.
(b) B slaves must not be connected to standard AS interface
masters.
Zhang_Ch03.indd 274
5/13/2008 5:41:18 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
275
3.1.4 System Characteristics and Important Data
3.1.4.1 How the AS Interface Functions
The AS interface or AS interface system operates as outlined below:
(1) Master–slave access techniques. The AS interface is a “single
master system.” This means that there is only one master per AS
interface network that controls the operations of process. This
polls all AS interface slaves one after the other and waits for a
response.
(2) Electronic address setting. The address of an AS interface slave
is its identifier. This only occurs once within an AS interface
system. The setting can either be made using a special addressing unit or by an AS interface master. The address is always
stored permanently on the AS interface slave.
(3) Operating reliability and flexibility. The transmission technique
used (current modulation) guarantees high operating reliability.
The master monitors the voltage on the cable and the transferred
data. It detects transmission errors and the failure of slaves and
sends a message to the controller such as a PLC or PC. The user
can then react to this message. Replacing or adding AS interface
slaves during normal operation does not affect communication
with other AS interface slaves.
3.1.4.2
Physical Characteristics
The most important physical characteristics of the AS interface and its
components are as follows:
(1) The two-wire cable for data and power supply. A simple twowire cable can be used. Shielding or twisting is not necessary.
Both the data and the power are transferred on this cable. The
power available depends on the AS interface power supply unit
used. For optimum wiring, the mechanically coded AS interface
cable is available preventing the connections from being reversed
and making simple contact with the AS interface application
modules using the penetration technique.
(2) Tree structure network with a cable. The “tree structure” of the
AS interface allows any point on a cable section to be used as the
start of a new branch.
(3) Direct integration. Practically all the electronics required for a
slave have been integrated on a special integrated circuit. This
allows the AS interface connector to be integrated directly in
binary actuators or sensors.
Zhang_Ch03.indd 275
5/13/2008 5:41:18 PM
276
INDUSTRIAL CONTROL TECHNOLOGY
(4) Increased functionality, more uses for the customer. Direct
integration allows devices to be equipped with a wide range of
functions. Four data and four parameter lines are available. The
resulting “intelligent” actuators/sensors increase the possibilities, for example, monitoring, parameter assignment, wear or
pollution checks, etc.
(5) Additional power supply for higher power requirements. An
external source of power can be provided for slaves with a higher
power requirement.
3.1.4.3
System Limits
(1) Cycle time
(a) Maximum 5 ms with standard AS interface slaves.
(b) Maximum 10 ms with AS interface slaves using the extended
addressing mode. AS interface uses constant message lengths.
Complicated procedures for controlling transmission and
identifying message lengths or data formats are not required.
This makes it possible for a master to poll all connected standard slaves within a maximum of 5 ms and to update the data
both on the master and slave.
If only one AS interface slave using the extended addressing mode is located at an address, this slave is polled at least
every 5 ms. If two extended slaves (A and B slave) share an
address, the maximum polling cycle is 10 ms. (B slaves can
only be connected to extended masters.)
(2) Number of connectable AS interface slaves
(a) Maximum of 31 standard slaves.
(b) Maximum of 62 slaves with the extended addressing mode.
AS interface slaves are the input and output channels of the AS
interface system. They are only active when called by the AS
interface master. They trigger actions or transmit reactions
to the master when commanded. Each AS interface slave
is identified by its own address (1–31). A maximum of 62
slaves using the extended addressing mode can be connected
to an extended master. Pairs of slaves using the extended
addressing mode occupy one address; in other words, the
addresses 1–31 can be assigned to two extended slaves.
If standard slaves are connected to an extended master,
these occupy a complete address; in other words, a maximum of up to 31 standard slaves can be connected to an
extended master.
(3) Number of inputs and outputs
(a) A maximum of 248 binary inputs and outputs with standard
modules.
Zhang_Ch03.indd 276
5/13/2008 5:41:18 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
277
(b) A maximum of 248 inputs and 186 outputs with modules
using the extended addressing mode. Each standard AS interface slave can receive 4 bits of data and send 4 bits of data.
Special modules allow each of these bits to be used for a
binary actuator or a binary sensor. This means that an AS
interface cable with standard AS interface slaves can have a
maximum of 248 binary attachments (124 inputs and 124
outputs). All typical actuators or sensors can be connected to
the AS interface in this way. The modules are used as distributed inputs/outputs.
If modules with the extended addressing mode are used, a
maximum of 3 inputs and 3 outputs is available per module; in
other words a maximum of 248 inputs and 186 outputs can be
operated with modules using the extended addressing mode.
3.1.4.4
Range of Functions of the Master Modules
The functions of the AS interface master modules are stipulated in the AS
interface master specification. An overview of these functions can be found
in the master module manual provided by the vendor or manufacturer.
The AS interface protocol was created in Germany in 1994 by a consortium of factory automation suppliers. Originally developed to be a lowcost method for addressing discrete sensors in factory automation
applications, AS interface has since gained acceptance in process industries due to its high power capability, simplicity of installation and operation, and low cost adder for devices.
Each AS interface segment can network up to 31 devices. This provides
for 124 inputs and 124 outputs, giving a maximum capacity of 248 I/O per
network on a v2.0 segment. The AS interface v2.1 specification doubles
this to 62 devices per segment, providing 248 inputs and 186 outputs for a
total network capacity of 434 I/O points.
Both signal and power are carried on two wires. Up to 8 A at 30 VDC of
power are available for field devices such as solenoid valves.
3.1.4.5 AS Interface in a Real-Time Environment
The system characteristics listed below can offer the AS interface the
capability to work in a real-time environment:
(1) Optimized system for binary sensors and actuators and for simple
analog elements.
(2) Master–slave principle with cyclic polling.
Zhang_Ch03.indd 277
5/13/2008 5:41:18 PM
278
INDUSTRIAL CONTROL TECHNOLOGY
(3) Tree structure of the network.
(4) Both data and power by means of one unshielded two-wire
cable.
(5) Flat cable for contacting by piercing technology.
(6) Modules as remote I/O ports for conventional sensors and
actuators.
(7) Integrated slaves with their own AS interface capabilities.
(8) No communication software in the slaves, only firmware in the
self-configuring master.
(9) Low costs, simple installation, easy handling, flexible networks,
high reliability in an industrial environment, open and internationally accepted system with many manufacturers and products.
There are three aspects of the AS interface that are of particular importance in real-time applications: connectivity, cycle time, and availability.
(1) Connectivity. AS interface has two distinct ways to be connected
to the first control level.
The first and most important way is a direct connection as the
type 2 of AS interface architecture given in Section 3.1.2. In that
case, the system’s master is part of a controller such as PLC,
SCADA, or PC, running at its own cycle time. As the AS interface is an open system, any kind of controller such as PLC,
SCADA, or PC manufacturer can build a master for their own
system. There are masters available to a lot of systems already,
with several more in development.
The second way is to connect AS interface via a coupler to a
higher Fieldbus and to use it as a subsystem, which has been
given as the type 1 of AS interface architecture in Section 3.1.2.
In that case, all data from the AS interface network is handled in
one node of the Fieldbus and it is connected to the above lying
host together with other components of the higher Fieldbus. The
application program has to handle all data as usual for the particular Fieldbus. For real-time applications, an analysis of the
cycle time and the availability of the combination of the two
systems has to be done. AS interface is definitely open to such
solutions and offers couplers to most known higher Fieldbus
like PROFIBUS, CAN, etc., with others (e.g., LON, Fieldbus
Foundation) being in preparation. Together with its tree structure, AS interface thus offers the most flexible networking solution to any application in automation.
(2) Cycle time. AS interface is a single-master system with cyclic
polling. Thus, any slave is addressed in a definite time.
Zhang_Ch03.indd 278
5/13/2008 5:41:18 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
279
For a complete net with 31 slaves, the cycle time is 5 ms. It
may be shorter with fewer slaves. (With very few slaves the cycle
time can be shortened to less than 500 µs.) Analog data with
more than 4 bits needs several cycles depending on its length, but
without affecting the basic cycle time for binary sensors and
actuators.
The cycle time includes all steps from and to the interface to
the host system and even includes one repetition. The data
exchange with the host happens from here via process I/O images
at the end of each cycle stored in, for example, a dual-ported
memory at the interface. Therefore no other steps have to be
taken into account for a direct connection to the control device.
This is asynchronous coupling, and in real-time applications
of the cycle times of both the network and the controller this may
present a restriction, but for many systems and applications this
is short enough.
(3) Availability. Availability in this context means that a system will
deliver reliable data and diagnostic values continuously and in
time under all specified conditions, especially under severe electromagnetic noise. The answers to three questions are of special
importance for real-time applications:
(a) Can electromagnetic noise or other faults disturb the reliability of data?
(b) How much time is necessary for the correction of a faulty
transmission?
(c) How often does such a fault happen and can this affect the
whole system?
3.2 Industrial Control System Interface Devices
In reference to Fig. 3.1, there exist two kinds of interface between the
control level and the actuator/sensor level in industrial control systems,
which are AS interface and field level interface. In additional to these two
kinds of interface, the interface between controller and either AS interface
or field level interface is also of importance in industrial control systems.
The interface between controller and either AS interface or field level
interface normally resides in the controller’s microprocessor unit or chipset to bridge the central processing unit (CPU) with exterior environments,
which therefore can be defined as controller interface, or simplified as
interfaces hereafter. Section 3.1 gives a detailed discussion on AS interface. This section concentrates on the field level in Section 3.2.1, and interfaces in Section 3.2.2.
Zhang_Ch03.indd 279
5/13/2008 5:41:18 PM
280
INDUSTRIAL CONTROL TECHNOLOGY
3.2.1
Fieldbus System
In recent years, there have emerged literally hundreds of Fieldbuses developed by different companies and organizations all over the world. The term
Fieldbus covers many different industrial control protocols. The following
lists some typical Fieldbuses with their applications as shown in Fig. 3.1.
3.2.1.1
Foundation Fieldbus
The Foundation Fieldbus can be flexibly used in process automation
applications. The specification supports bus-powered field devices as
well as allows application in hazardous areas. The Fieldbus Foundation, an
independent not-for-profit organization which aims at developing and
maintaining an internationally uniform and successful Fieldbus for automation tasks, claimed to establish an international, interoperable Fieldbus
standard to replace the expensive, conventional 4–20 mA wiring in the
field and enables bidirectional data transmission. The entire communication between the devices and the automation system as well as the process
control station takes place over the bus system, and all operating and device
data are exclusively transmitted over the Fieldbus. The communication
between control station, operating terminals, and field devices simplifies
the start-up and parameterization of all components. The communication
functions allow diagnostic data, which are provided by up-to-date field
devices, to be evaluated. The essential objectives in Fieldbus technology
are to reduce installation costs, save time and costs due to simplified planning, as well as improve the operating reliability of the system due to additional performance features. Fieldbus systems are usually implemented in
new plants or existing plants that must be extended. To convert an existing
plant to Fieldbus technology, the conventional wiring can either be modified into a bus line or be replaced with a shielded bus cable, if required.
(1) Performance features. The Foundation Fieldbus provides a broad
spectrum of services and functions compared to other Fieldbus
systems:
(a) Intrinsic safety for use in hazardous environments
(b) Bus-powered field devices
(c) Line or tree topology
(d) Multimaster capable communication
(e) Deterministic (predictable) dynamic behavior
(f) Distributed data transfer (DDT)
(g) Standardized block model for uniform device interfaces
(h) Flexible extension options based on device descriptions.
Zhang_Ch03.indd 280
5/13/2008 5:41:18 PM
281
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
The characteristic feature of distributed data transfer
enables single field devices to execute automation tasks so
that they are no longer just sensors or actuators, but contain
additional functions. For the description of a device’s
function(s) and for the definition of a uniform access to the
data, the Foundation Fieldbus contains predefined function
blocks. The function blocks implemented in a device provide information about the tasks the device can perform.
Typical functions provided by sensors include the following:
analog input or discrete input (digital input). Control valves
usually contain the following function blocks: analog output
or discrete output (digital output). The following blocks exist
for process control tasks: Proportional and Derivative (PD
controller) or Proportional and Integral and Derivative
(PID controller). If a device contains such a function block,
it can control a process variable independently.
The shift of automation tasks—from the control level
down to the field—results in the flexible, distributed processing of control tasks. This reduces the load on the central process control station which can even be replaced entirely in
small-scale installations. Therefore, an entire control loop
can be implemented as the smallest unit, consisting only of
one sensor and one control valve with integrated process
controller, which communicates over the Foundation Fieldbus
(see Fig. 3.8). The enhanced functionality of the devices
leads to higher requirements to be met by the device hardware and comparably complex software implementation and
device interfaces.
User 1
User 2
Switch
Bridge
High speed ethernet
H1 bus
2
1
Junction
box
3
5
6
Figure 3.8 Foundation Fieldbus control network.
Zhang_Ch03.indd 281
5/13/2008 5:41:18 PM
282
INDUSTRIAL CONTROL TECHNOLOGY
(2) Layered communications model. The Foundation Fieldbus
specification is based on the layered communications model
and consists of three major functional elements as illustrated
in Fig. 3.9:
(a) Physical layer
(b) Communication “stack”
(c) User application is made up of function blocks and the
device description. It is directly based on the communication stack. Depending on which blocks are implemented in
a device, users can access a variety of services. System
management utilizes the services and functions of the user
application and the application layer to execute its tasks
(Fig. 3.9(b) and (c)). It ensures proper cooperation between
the individual bus components as well as synchronizes the
measurement and control tasks of all field devices with
regard to time.
The Foundation Fieldbus layered communications model
is based on the ISO/OSI reference model. As is the case for
most Fieldbus systems, and in accordance with an IEC specification, layers 3–6 are not used. The comparison in Fig. 3.9
shows that the communication stack covers the tasks of layers 2 and 7 and that Layer 7 consists of the Fieldbus Access
Sublayer and the Fieldbus Message Specification.
User
application
(a)
Communication
stack
Physical layer
User
application
(b)
Application layer
Presentation layer
Session layer
Transport layer
Network layer
Data link layer
Physical layer
Function
block
model
Device
description
(c)
Fieldbus message
specification
Fieldbus access
sublayer
Presentation layer
Session layer
Transport layer
Network layer
Data link layer
Physical layer
Figure 3.9 Structure and description of the Foundation Fieldbus communication
model.
Zhang_Ch03.indd 282
5/13/2008 5:41:19 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
283
(3) Physical layer. The Foundation Fieldbus model solves pending
communication tasks by using two bus systems, the slow, intrinsically safe H1 bus and the fast, higher level H2 bus as given in
Fig. 3.8.
(a) H1 bus. The following summary gives a brief overview of
the basic values and features of the H1 bus.
(i) Manchester coding is used for data transfer. The data
transfer rate is 31.25 kbit/s.
(ii) Proper communication requires that the field devices
have enough voltage. Each device should have minimum 9 V. To make sure that this requirement is met,
software tools are available which calculate the resulting currents and terminal voltages based on the network topology, the line resistance, and the supply
voltage.
(iii) The H1 bus allows the field devices to be powered
over the bus. The power supply unit is connected to the
bus line in the same way (parallel) as a field device.
Field devices powered by supply sources other than
the bus must be additionally connected to their own
supply sources.
(iv) With the H1 bus it must be ensured that the maximum
power consumption of current consuming devices is
lower than the electric power supplied by the power
supply unit.
(v) Network topologies used are usually line topology or,
when equipped with junction boxes, star, tree, or a
combination of topologies. The devices are best connected via short spurs using tee connectors to enable
connection/disconnection of the devices without interrupting communication.
(vi) The maximum length of a spur is limited to 120 m and
depends on the number of spurs used as well as the
number of devices per spur.
(vii) Without repeaters, the maximum length of a H1 segment can be as long as 1900 m. By using up to four
repeaters, a maximum of 5 × 1900 m = 9500 m can be
achieved. The short spurs from the field device to the
bus are included in this total length calculation.
(viii) The number of bus users per bus segment is limited to
32 in intrinsically safe areas. In explosion-hazardous
areas, this number is reduced to only a few devices due
to power supply limitations.
Zhang_Ch03.indd 283
5/13/2008 5:41:19 PM
284
INDUSTRIAL CONTROL TECHNOLOGY
(ix) Various types of cables are useable for Fieldbus. Type A
is recommended as preferred Fieldbus cable, and only
this type is specified for the maximum bus length of
1900 m.
(x) Principally, there need to be two terminators per bus
segment, one at or near each end of a transmission line.
(xi) It is not imperative that bus cables be shielded, however, it is recommended to prevent possible interferences and for best performance of the system.
The H1 bus can be designed intrinsically safe (Ex-i)
to suit applications in hazardous areas. This requires
that proper barriers be installed between the safe and
the explosion-hazardous area. In addition, only one
device, the power supply unit, must supply the Fieldbus
with power. All other devices must always, that is, also
when transmitting and receiving data, function as current sinks.
Since the capacity of electrical lines is limited in
intrinsically safe areas depending on the explosion
group—IIB or IIC—the number of devices that can be
connected to one segment depends on the effective
power consumption of the used devices. Since the
Foundation Fieldbus specification is not based on
the FISCO model, the plant operator must ensure
that intrinsic safety requirements are met when planning and installing the communications network. For
instance, the capacitance and inductance of all line
segments and devices must be calculated to ensure
that the permissible limit values are observed.
(b) High speed Ethernet (HSE). The HSE is based on standard
Ethernet technology. The required components are therefore
widely used and are available at low cost. The HSE runs at
100 Mbit/s and cannot be equipped not only with electrical
lines, but also with optical fiber cables.
The Ethernet operates by using random (not deterministic)
CSMA bus access. This method can only be applied to a limited number of automation applications because it requires
real-time capability. The extremely high transmission rate
enables the bus to respond sufficiently fast when the bus load
is low and devices are only few. With respect to process
engineering requirements, real-time requirements are met in
any case.
If the bus load must be reduced due to the many connected
devices, or if several HSE partial networks are to be combined
Zhang_Ch03.indd 284
5/13/2008 5:41:19 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
285
to create a larger network, Ethernet switches must be used
(see Fig. 3.8). A switch reads the target address of the data
packets that must be forwarded and then passes the packets
on to the associated partial network. This way, the bus load
and the resulting bus access time can be controlled to best
adapt it to the respective requirements.
(c) Bridge to H1–HSE coupling. A communications network
that consists of a H1 bus and a HSE network results in a
topology as illustrated in Fig. 3.8. To connect the comparatively slow H1 segments to the HSE network, coupling components, so-called bridges, are required. A bridge is used to
connect the individual H1 buses to the fast high speed
Ethernet. The various data transfer rates and data telegrams
must be adapted and converted, considering the direction of
transmission. This way, powerful and widely branched networks can be installed in larger plants.
(4) Communication stack. The field devices used with the Foundation
Fieldbus are capable of assuming process control functions. This
option is based on distributed communication which ensures that
each controlling field device can exchange data with other devices
(e.g., reading measuring values, forwarding correction values),
all field devices are served in time (“in time” meaning that
the processing of the different control loops is not negatively
influenced), and two or more devices never access the bus
simultaneously.
To meet these requirements, the H1 bus of the Foundation
Fieldbus uses a central communication control system.
(a) Link active scheduler (LAS). The LAS controls and schedules the communication on the bus. It controls the bus activities using different commands which it broadcasts to the
devices. Since the LAS also continuously polls unassigned
device addresses, it is possible to connect devices during
operation and to integrate them in the bus communication.
Devices that are capable of becoming the LAS are called
Link Masters. Basic devices do not have the capability to
become LAS. In a redundant system containing multiple
Link Masters, one of the Link Masters will become the LAS
if the active LAS fails (fail-operational design).
(b) Communication control. The communication services of the
FF specification utilize scheduled and unscheduled data
transmission. Time-critical tasks, such as the control of process variables, are exclusively performed by scheduled services, whereas parameterization and diagnostic functions are
carried out using unscheduled communication services.
Zhang_Ch03.indd 285
5/13/2008 5:41:19 PM
286
INDUSTRIAL CONTROL TECHNOLOGY
(i) Scheduled data transmission. To solve communication
tasks in time and without access conflicts, all time-critical
tasks are based on a strict transmission schedule. This
schedule is created by the system operator during the
configuration of the Foundation Field system. The
LAS periodically broadcasts a synchronization signal
(TD: Time Distribution) on the Fieldbus so that all
devices have exactly the same data link time. In scheduled transmission, the point of time and the sequence
are exactly defined. This is why it is called a deterministic system.
(ii) Unscheduled transmission. Device parameters and diagnostic data must be transmitted when needed, that is, on
request. The transmission of this data is not time critical. For such communication tasks, the Foundation
Fieldbus is equipped with the option of unscheduled
data transmission.
Unscheduled data transmission is exclusively
restricted to the breaks in between scheduled transmission. The LAS grants permission to a device to use the
Fieldbus for unscheduled communication tasks if no
scheduled data transmission is active.
Permission for a certain device to use the bus is granted
by the LAS when it issues a pass token (PT command) to
the device. The pass token is sent around to all devices
entered in the Live List which is administrated by the
LAS. Each device may use the bus as long as required
either until it returns the token or until the maximum
granted time to use the token has elapsed.
The Live List is continuously updated by the LAS.
The LAS sends a special command, the Probe Node
(PN), to the addresses not in the Live List, searching for
newly added devices. If a device returns a Probe
Response (PR) message, the LAS adds the device to the
Live List where it receives the pass token for unscheduled communication according to the order submitted
for transmission in the Live List. Devices which do not
respond to the PT command or return the token after
three successive tries are removed from the Live List.
Whenever a device is added or removed from the Live
List, the LAS broadcasts these changes to all devices.
This allows all Link Masters to maintain a current copy
of the Live List so that they can become the LAS without the loss of information.
Zhang_Ch03.indd 286
5/13/2008 5:41:19 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
287
(c) Communication schedule. The LAS follows a strict schedule
to ensure that unscheduled communication using the token
as well as the TD or PN commands do not interfere with the
scheduled data transmission.
Before each operation, the LAS refers to the transmission
list to check for any scheduled data transmissions. If this is
the case, it waits (idle mode) for precisely the scheduled time
and then sends a Compel Data (CD) message to activate the
operation.
In case there are no scheduled transmissions and sufficient
time is available for additional operations, the LAS sends
one of the other commands. With PN it searches for new
devices, or it broadcasts a TD message for all devices to have
exactly the same data link time, or it uses the PT massage to
pass the token for unscheduled communication. Following
this, the sequence starts all over again with the abovementioned check of the transmission list entries.
It is obvious that this cycle gives scheduled transmission
the highest priority and that the scheduled times are strictly
observed, regardless of other operations.
(5) User application layer. The Fieldbus Access Sublayer (FAS) and
Fieldbus Message Specification (FMS) layer form the interface
between the data link layer and the user application (see Fig. 3.9).
The services provided by FAS and FMS are invisible for the user.
However, the performance and functionality of the communication system considerably depends on these services.
(a) Fieldbus access sublayer (FAS). FAS services create Virtual
Communication Relationships (VCR) which are used by the
higher level FMS layer to execute its tasks (Figure 3.10).
VCRs describe different types of communication processes
and enable the associated activities to be processed more
quickly. Foundation Fieldbus communication utilizes three
different VCR types as follows:
(i) The Publisher/Subscriber VCR type is used to transmit
the input and output data of function blocks. As described
above, scheduled data transmission with the CD command is based on this type of VCR. However, the
Publisher/Subscriber VCR is also available for unscheduled data transmission; for instance, if a subscriber
requests measuring or positioning data from a device.
(ii) The Client/Server VCR type is used for unscheduled,
user-initiated communication based on the PT command. If a device (client) requests data from another
device, the requested device (server) only responds
Zhang_Ch03.indd 287
5/13/2008 5:41:20 PM
288
INDUSTRIAL CONTROL TECHNOLOGY
Client/Server
Operator
communication
Report distribution
Event notification,
alarms, trend reports
Publisher/subscriber
Data publication
Set point changes
Mode and device
data changes
Send process alarms
to operator consoles
Send actual value of a
transmitter to PID
block and operator
console
Upload/download
Adjusting alarm
values
Access display views
Remote diagnostics
Send trend reports to
data
historians
Figure 3.10 Virtual Communication Relationships of the FAS.
when it receives a PT from the LAS. The Client/Server
communication is the basis for operator-initiated
requests, such as set point changes, tuning parameter
access and change, diagnosis, device upload and download, etc.
(iii) Report distribution communication is used to send alarm
or other event notifications to the operator consoles or
similar devices. Data transmission is unscheduled when
the device receives the PT command together with the
report (trend or event notification). Fieldbus devices
that are configured to receive the data await and read
this data.
(b) Fieldbus message specification (FMS). FMS provides the
services for standardized communication. Data types that
are communicated over the Fieldbus are assigned to certain
communication services. For a uniform and clear assignment, object descriptions are used.
Object descriptions not only contain definitions of all
standard transmission message formats, but also include
application-specific data. For each type of object there are
special, predefined communication services.
Object descriptions are collected together in a structure
called an object dictionary. The object description is identified
by its index. (1) Index 0, called the object dictionary header,
provides a description of the dictionary itself. (2) Indices
between 1 and 255 define standard data types that are used
to build more complex object descriptions. (3) The User
Application object descriptions can start at any index above
255. The FMS defines Virtual Field Devices (VFD) which
are used to make the object descriptions of a field device as
Zhang_Ch03.indd 288
5/13/2008 5:41:20 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
289
well as the associated device data available over the entire
network. The VFDs and the object description can be used to
remotely access all local field device data from any location
by using the associated communication services.
3.2.1.2
PROFIBUS
PROFIBUS is the largest Fieldbus in the world with cost-saving solutions in factory automation and process automation plus safety, drives, and
motion control coverage. This Fieldbus approach produces significant cost
savings in design, installation, and maintenance expenses over the old
approach of point-to-point wiring. For many years, PROFIBUS development has continued with undiminished enthusiasm and energy. With more
than 13,000,000 nodes installed there are significantly more PROFIBUS
nodes installed than of any other Fieldbus. Current PROFIBUS activities
are targeted at system integration, PROFIBUS engineering development,
and application profiles. Because of these application profiles, PROFIBUS
today is the only Fieldbus that provides robust engineering solutions for
both factory and process automation.
(1) Working mechanism. PROFIBUS is suitable for both fast,
time-critical applications and complex communication tasks.
PROFIBUS communication is in the international standards
IEC 61158 and IEC 61784. The application and engineering
aspects are specified in the generally available guidelines of the
PROFIBUS User Organization. This fulfills user demand for
manufacturer independence and openness and ensures communication between devices of various manufacturers.
PROFIBUS can handle large amounts of data at high speed
and can serve the needs of large installations. Based on a realtime capable asynchronous token bus principle, PROFIBUS
defines multimaster and master–slave communication relations,
with cyclic or acyclic access, allowing transfer rates of up to
500 kbit/s. The physical layer (two-wire RS485), the data link
layer, and the application layer are all standardized. PROFIBUS
distinguishes between confirmed and unconfirmed services, allowing process communication with both broadcast and multitasking
protocols. PROFIBUS DP is a master/slave polling network with
the ability to upload/download configuration data and precisely
synchronized multiple devices on the network. Multiple masters
are possible in PROFIBUS, but the outputs of any device can
only be assigned to one master. There is no power on the bus.
Zhang_Ch03.indd 289
5/13/2008 5:41:20 PM
290
INDUSTRIAL CONTROL TECHNOLOGY
(2) Basic types. PROFIBUS encompasses several Industrial Bus
Protocol Specifications, including PROFIBUS-DP, PROFIBUSPA, PROFIBUS-FMS, PROFInet, PROFIBUS-safe, and
PROFIBUS for motion control.
(a) PROFIBUS-DP. PROFIBUS-DP is the main emphasis for
factory automation; it uses RS485 transmission technology,
one of the DP communications protocol versions, and has
widespread usage for such items as remote I/O systems, motor
control centers, and variable speed drives. PROFIBUS-DP
communicates at speeds from 9.6 kbps to 12 Mbps over distances from 100 to 1200 m. PROFIBUS-DP does not natively
support intrinsically safe installations. More than 2500
PROFIBUS-compliant products are available from which
you can select best-in-class devices to suit your individual
needs, with alternative sources usually available.
(b) PROFIBUS-PA. PROFIBUS-PA is the main emphasis for
process automation, typically with MBP-IS transmission
technology, the communications protocol version DP-V1,
and the application profile PA devices. PROFIBUS-PA is a
full-function Fieldbus that is generally used for process level
instrumentation. PROFIBUS-PA communicates at 31.25 kbps
and has a maximum distance of 1900 m per segment.
PROFIBUS-PA is designed to support intrinsically safe
applications. PROFIBUS is also tailored to process automation requirements. It is of modular design and comprises the
communication protocol PROFIBUS-DP, different transmission technologies, numerous application profiles, and structured device integration tools. Typical PROFIBUS-PA
applications are formed by combining modules suited for or
required by the respective applications.
(c) PROFIBUS-FMS. PROFIBUS-FMS is designed for communication at the cell level according to Fieldbus message specification. At this level programmable controllers (e.g., PLC
and PC) communicate primarily with each other. In this
application area a high degree of functionality is more important than fast system reaction times. FMS services are a
subset of the services (MMS = Manufacturing Message
Specification, ISO 9506) which have been optimized for
Fieldbus applications and to which functions for communication object administration and network management have
been added. Execution of the FMS services via the bus is
described by service sequences consisting of several interactions which are called service primitives. Service primitives
describe the interaction between requester and responder.
Zhang_Ch03.indd 290
5/13/2008 5:41:20 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
291
(d) PROFInet. PROFInet is the leading Industrial Ethernet standard for automation that includes plant-wide Fieldbus communication and plant-to-office communication. PROFInet is
designed to work from I/O to MES and hence can simultaneously handle standard Ethernet transmissions and real-time
transmissions at 1 ms speeds. PROFInet embraces industry
standards like TCP/IP, XML, OPC, and ActiveX. Because of
the integrated proxy technology it connects other Fieldbuses
in addition to PROFIBUS, thus protecting the existing investment in plant equipment and networks (whether that is
PROFIBUS or another Fieldbus).
(e) Motion control with PROFIBUS. Motion control with
PROFIBUS is the main emphasis for drive technology using
RS485 transmission technology, the communications protocol version DP V2, and the application profile PROFI drive.
The demands of motion control propelled the implementation of functionalities such as clock cycle synchronization or
slave-to-slave communication. Decentralized drive applications can be realized economically by means of intelligent
drives, since PROFIBUS now also permits the highly dynamic
distribution of the technological signals among the drives.
(f) PROFI-safe. PROFI-safe is the main emphasis for safetyrelevant applications (universal use for almost all industries),
using RS485 or MBP-IS transmission technology, one of the
available DP versions for communication, and the application profile PROFI-safe. PROFIBUS is the very first Fieldbus
in merging standard automation and safety automation in one
technology, running on the same bus, using the same communication mechanisms, and thus providing highest efficiency to the user. This supports simple and cost-effective
installation and operation.
3.2.1.3
Controller Area Network (CAN bus)
CAN is a serial bus system, which was originally developed for automotive applications in the early 1980s. The CAN protocol was internationally
standardized in 1993 as ISO 11898-1 and comprises the data link layer of
the seven layer ISO/OSI reference model. CAN bus system can theoretically link up to 2032 devices (assuming one node with one identifier)
on a single network. However, due to the practical limitation of the
hardware (transceivers), it can only link up to 110 nodes (with 82C250,
Philips) on a single network. It offers high-speed communication rate
up to 1 Mbit/s thus allowing real-time control. In addition, the error
Zhang_Ch03.indd 291
5/13/2008 5:41:20 PM
292
INDUSTRIAL CONTROL TECHNOLOGY
confinement and the error detection feature make it more reliable in noise
critical environment.
CAN bus systems provide the following: (1) A multimaster hierarchy,
which allows building intelligent and redundant systems. If one network
node is defective the network is still able to operate. (2) Broadcast communication. A sender of information transmits to all devices on the bus.
All receiving devices read the message and then decide if it is relevant to
them. This guarantees data integrity as all devices in the system use the
same information. (3) Sophisticated error detecting mechanisms and
retransmission of faulty messages. This also guarantees data integrity.
The CAN serial bus system is used in a broad range of embedded as well
as automation control systems. It usually links two or more microcontrollerbased physical devices. The original equipment manufacturers (OEM)
design embedded control systems; the end user has no or only some knowledge of the embedded network functions and is therefore not responsible
for the CAN communication system. However, automation control systems are specified by the end user. The system design including the CAN
network services may be implemented by the end users themselves or by
a system house.
The main CAN application fields include (1) passenger cars, (2) trucks
and buses, (3) off-highway and off-road vehicles, (4) maritime electronics;
(5) aircraft and aerospace electronics, (6) factory automation, (7) industrial machine control, (8) lifts and escalators, (9) building automation,
(10) medical equipment and devices, (11) nonindustrial control, (12) nonindustrial equipment.
(1) CAN basic working mechanism.
(a) Principles of data exchange. When data are transmitted by
CAN, no stations are addressed, but instead, the content of
the message (e.g., rpm or engine temperature) is designated
by an identifier that is unique throughout the network. The
identifier defines not only the content but also the priority of
the message. This is important for bus allocation when several stations are competing for bus access.
If the CPU of a given station wishes to send a message to
one or more stations, it passes the data to be transmitted and
their identifiers to the assigned CAN chip (“Make ready”).
This is all the CPU has to do to initiate data exchange. The
message is constructed and transmitted by the CAN chip.
As soon as the CAN chip receives the bus allocation (“Send
Message”) all other stations on the CAN network become
receivers of this message (“Receive Message”). Each station
Zhang_Ch03.indd 292
5/13/2008 5:41:20 PM
293
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
CAN
station 1
CAN
station 2
Accept
Prepare
Select
CAN
station 3
CAN
station 4
Accept
Select
Select
Receive
message
Receive
message
Send
message
Receive
message
Figure 3.11 Broadcast transmission and acceptance filtering by CAN nodes.
in the CAN network, having received the message correctly,
performs an acceptance test to determine whether the data
received are relevant for that station (“Select”). If the data
are of significance for the station concerned they are processed (“Accept”), otherwise they are ignored. Figure 3.11
illustrates this scenario.
A high degree of system and configuration flexibility
is achieved as a result of the content-oriented addressing
scheme. It is very easy to add stations to the existing CAN
network without making any hardware or software modifications to the existing stations, provided the new stations are
purely receivers. Because the data transmission protocol
does not require physical destination addresses for the individual components, it supports the concept of modular
electronics and also permits multiple reception (broadcast,
multicast) and the synchronization of distributed processes:
measurements needed as information by several controllers
can be transmitted via the network, in such a way that it is
unnecessary for each controller to have its own sensor.
(b) Nondestructive bitwise arbitration. For the data to be processed in real time, they must be transmitted rapidly. This
not only requires a physical data transfer path with up to
1 Mbit/s but also calls for rapid bus allocation when several
stations wish to send messages simultaneously (Fig. 3.12).
In real-time processing the urgency of messages to be
exchanged over the network can differ greatly: a rapidly
changing dimension (e.g., engine load) has to be transmitted
Zhang_Ch03.indd 293
5/13/2008 5:41:20 PM
294
INDUSTRIAL CONTROL TECHNOLOGY
Recessive
Dominant
Bus line
1
1
2
3
2
3
1 loses
3 loses
Figure 3.12 Principle of nondestructive bitwise arbitration.
more frequently and therefore with fewer delays than other
dimensions (e.g., engine temperature) which change relatively slowly.
The priority at which a message is transmitted compared
with another less urgent message is specified by the identifier of the message concerned. The priorities are laid down
during system design in the form of corresponding binary
values and cannot be changed dynamically. The identifier
with the lowest binary number has the highest priority. Bus
access conflicts are resolved by bitwise arbitration on the
identifiers involved by each station observing the bus level
bit for bit. In accordance with the “wired” mechanism, by
which the dominant state (logical 0) overwrites the recessive
state (logical 1), the competition for bus allocation is lost by
all those stations with recessive transmission and dominant
observation. All “losers” automatically become receivers of
the message with the highest priority and do not reattempt
transmission until the bus is available again.
(c) Destructive bus allocation. Simultaneous bus access by more
than one station causes all transmission attempts to be
aborted and, therefore, there is no successful bus allocation.
More than one bus access may be necessary in order to allocate the bus at all. The number of attempts before bus allocation is successful being a purely statistical quantity (examples:
CSMA/CD, Ethernet).
In order to process all transmission requests of a CAN
network while complying with latency constraints at as low
a data transfer rate as possible, the CAN protocol must
implement a bus allocation method that guarantees that there
Zhang_Ch03.indd 294
5/13/2008 5:41:21 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
295
is always unambiguous bus allocation even when there are
simultaneous bus accesses from different stations. The
method of bitwise arbitration using the identifier of the messages to be transmitted uniquely resolves any collision
between a number of stations wanting to transmit, and it
does this at the latest within 13 (standard format) or 33
(extended format) bit periods for any bus access period.
Unlike the message-wise arbitration employed by the CSMA/
CD method, this nondestructive method of conflict resolution ensures that no bus capacity is used without transmitting
useful information.
Even in situations where the bus is overloaded, the linkage
of the bus access priority to the content of the message proves
to be a beneficial system attribute compared with existing
CSMA/CD or token protocols: in spite of the insufficient bus
transport capacity, all outstanding transmission requests are
processed in order of their importance to the overall system
(as determined by the message priority). The available transmission capacity is utilized efficiently for the transmission
of useful data since “gaps” in bus allocation are kept very
small. The collapse of the whole transmission system due to
overload, as can occur with the CSMA/CD protocol, is not
possible with CAN. Thus, CAN permits implementation of
fast, traffic-dependent bus access which is nondestructive
because of bitwise arbitration based on the message priority
employed.
Nondestructive bus access can be further classified into
centralized bus access control or decentralized bus access
control depending on whether the control mechanisms are
present in the system only once (centralized) or more than
once (decentralized). A communication system with a designated station (inter alia for centralized bus access control)
must provide a strategy to take effect in the event of a failure
of the master station. This concept has the disadvantage that
the strategy for failure management is difficult and costly to
implement and also that the takeover of the central station by
a redundant station can be very time consuming. For these
reasons and to circumvent the problem of the reliability of
the master station (and thus of the whole communication
system), the CAN protocol implements decentralized bus
control. All major communication mechanisms, including
bus access control, are implemented several times in the
system because this is the only way to fulfill the high requirements for the availability of the communication system.
Zhang_Ch03.indd 295
5/13/2008 5:41:21 PM
296
INDUSTRIAL CONTROL TECHNOLOGY
In summary it can be said that CAN implements a trafficdependent bus allocation system that permits, by means of a
nondestructive bus access with decentralized bus access control, a high useful data rate at the lowest possible bus data
rate in terms of the bus busy rate for all stations. The efficiency of the bus arbitration procedure is increased by the
fact that the bus is utilized only by those stations with pending transmission requests. These requests are handled in the
order of the importance of the messages for the system as a
whole. This proves especially advantageous in overload situations. Since bus access is prioritized on the basis of the
messages, it is possible to guarantee low individual latency
times in real-time systems.
(d) Message frame formats. The CAN protocol supports two
message frame formats, the only essential difference being
in the length of the identifier (ID). In the standard format the
length of the ID is 11 bits, and in the extended format the
length is 29 bits. The message frame for transmitting messages on the bus comprises seven main fields (Fig. 3.13).
A message in the standard format begins with the start bit
“start of frame,” followed by the “arbitration field,” which
contains the identifier and the remote transmission request
(RTR) bit, which indicates whether it is a data frame or a
request frame without any data bytes (remote frame). The
“control field” contains the IDE (identifier extension) bit,
which indicates either standard format or extended format, a
bit reserved for future extensions and—in the last 4 bits—a
count of the data bytes in the data field. The “data field”
ranges from 0 to 8 bytes in length and is followed by the
“CRC field,” which is used as a frame security check for
detecting bit errors. The “ACK field” comprises the ACK
slot (1 bit) and the ACK delimiter (1 recessive bit). The bit in
the ACK slot is sent as a recessive bit and is overwritten as a
dominant bit by those receivers which have at this time
received the data correctly (positive acknowledgment).
Correct messages are acknowledged by the receivers regardless of the result of the acceptance test. The end of the
Arbitration field
S
O
F
11-bit identifier
Control
field
R I r
T D 0 DLC
R E
Data field
CRC
field
0–8 bytes
15-bit CRC
Ack
field
End of
frame
Int
Bus idle
Figure 3.13 Message frame for standard format (CAN Specification 2.0A).
Zhang_Ch03.indd 296
5/13/2008 5:41:21 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
297
message is indicated by “end of frame.” “Intermission” is the
minimum number of bit periods separating consecutive messages. If there is no following bus access by any station, the
bus remains idle (“bus idle”).
(e) Detecting and signaling errors. Unlike other bus systems,
the CAN protocol does not use acknowledgment messages
but instead signals any errors that occur. For error detection,
the CAN protocol implements three mechanisms at the message level:
(i) Cyclic redundancy check (CRC). The CRC safeguards
the information in the frame by adding redundant check
bits at the transmission end. At the receiver end these
bits are recomputed and tested against the received bits.
If they do not agree, there has been a CRC error.
(ii) Frame check. This mechanism verifies the structure of
the transmitted frame by checking the bit fields against
the fixed format and the frame size. Errors detected by
frame checks are designated “format errors.”
(iii) ACK errors. As mentioned above, frames received
are acknowledged by all recipients through positive
acknowledgment. If no acknowledgment is received by
the transmitter of the message (ACK error) this may
mean that there is a transmission error which has been
detected only by the recipients, that the ACK field has
been corrupted, or that there are no receivers.
The CAN protocol also implements two mechanisms
for error detection at the bit level:
(i) Monitoring. The ability of the transmitter to detect
errors is based on the monitoring of bus signals: each
node which transmits also observes the bus level and
thus detects differences between the bit sent and the bit
received. This permits reliable detection of all global
errors and errors local to the transmitter.
(ii) Bit stuffing. The coding of the individual bits is tested at
bit level. The bit representation used by CAN is NRZ
(nonreturn-to-zero) coding, which guarantees maximum
efficiency in bit coding. The synchronization edges are
generated by means of bit stuffing, that is, after five
consecutive equal bits the sender inserts into the bit
stream a stuff bit with the complementary value, which
is removed by the receivers. The code check is limited
to checking adherence to the stuffing rule.
If one or more errors are discovered by at least one
station (any station) using the above mechanisms, the
Zhang_Ch03.indd 297
5/13/2008 5:41:22 PM
298
INDUSTRIAL CONTROL TECHNOLOGY
current transmission is aborted by sending an “error
flag.” This prevents other stations from accepting the
message and thus ensures the consistency of data
throughout the network.
After transmission of an erroneous message has been
aborted, the sender automatically reattempts transmission (automatic repeat request). There may again be
competition for bus allocation. As a rule, retransmission will be begun within 23 bit periods after error
detection; in special cases the system recovery time is
31 bit periods.
However effective and efficient the method described
may be, in the event of a defective station it might lead
to all messages (including correct ones) being aborted,
thus blocking the bus system if no measures for selfmonitoring were taken. The CAN protocol, therefore,
provides a mechanism for distinguishing sporadic errors
from permanent errors and localizing station failures
(fault confinement).
This is done by statistical assessment of station error
situations with the aim of recognizing a station’s own
defects and possibly entering an operating mode where
the rest of the CAN network is not negatively affected.
This may go as far as the station switching itself off to
prevent messages erroneously recognized as incorrect
from being aborted.
(f) Extended format CAN messages. The SAE “Truck and Bus”
subcommittee standardized signals and messages as well as
data transmission protocols for various data rates. It became
apparent that standardization of this kind is easier to implement when a longer identification field is available.
To support these efforts, the CAN protocol was extended
by the introduction of a 29-bit identifier. This identifier is
made up of the existing 11-bit identifier (base ID) and an
18-bit extension (ID extension). Thus, the CAN protocol
allows the use of two message formats: StandardCAN
(Version 2.0A) and ExtendedCAN (Version 2.0B). As the
two formats have to coexist on one bus, it is laid down which
message has higher priority on the bus in the case of bus
access collisions with differing formats and the same base
identifier: the message in standard always has priority over
the message in extended format.
CAN controllers that support the messages in extended
format can also send and receive messages in standard format.
Zhang_Ch03.indd 298
5/13/2008 5:41:22 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
299
When CAN controllers that only cover the standard format
(Version 2.0A) are used in one network, then only messages
in standard format can be transmitted on the entire network.
Messages in extended format would be misunderstood.
However, there are CAN controllers that support only standard format but recognize messages in extended format and
ignore them (Version 2.0B passive).
The distinction between standard format and extended format is made using the IDE bit (Identifier Extension Bit) which
is transmitted as dominant in the case of a frame in standard
format. For frames in extended format it is recessive.
The RTR bit is transmitted dominant or recessive depending on whether data are being transmitted or whether a specific message is being requested from a station. In place of
the RTR bit in standard format the substitute remote request
(SRR) bit is transmitted for frames with extended ID. The
SRR bit is always transmitted as recessive, to ensure that in
the case of arbitration the standard frame always has priority
bus allocation over an extended frame when both messages
have the same base identifier.
Unlike the standard format, in the extended format the
IDE bit is followed by the 18-bit ID extension, the RTR bit,
and a reserved bit (r1).
All the following fields are identical with standard format.
Conformity between the two formats is ensured by the fact
that the CAN controllers which support the extended format
can also communicate in standard format.
(g) Implementations of the CAN protocol. Communication is
identical for all implementations of the CAN protocol. There
are differences, however, with regard to the extent to which
the implementation takes over message transmission from
the microcontrollers which follow it in the circuit.
CAN controllers with intermediate buffer (formerly called
basicCAN chips) have implemented as hardware the logic
necessary to create and verify the bit stream according to
protocol. However, the administration of data sets to be sent
and received, acceptance filtering in particular, is carried out
to only a limited extent by the CAN controller.
Typically, CAN controllers with intermediate buffer have
two reception and one transmission buffer. The 8-bit code
and mask registers allow a limited acceptance filtering (8 MSB
of the identifier). Suitable choice of these register values
allows groups of identifiers or in borderline cases all IDs to
be selected. If more than the 8 ID-MSBs are necessary to
Zhang_Ch03.indd 299
5/13/2008 5:41:22 PM
300
INDUSTRIAL CONTROL TECHNOLOGY
differentiate between messages, then the microcontroller
following the CAN controller in the circuit must complement acceptance filtering by software. CAN controllers with
intermediate buffer may place a strain on the microcontroller
with the acceptance filtering, but they require only a small
chip area and can therefore be produced at lower cost. In
principle they can accept all objects in a CAN network.
CAN objects consist mainly of three components: identifier, data length code, and the actual useful data. CAN
controllers with object storage (formerly called FullCAN)
function like CAN controllers with intermediate buffers, but
also administer certain objects. Where there are several
simultaneous requests they determine, for example, which
object is to be transmitted first. They also carry out acceptance filtering for incoming objects. The interface to the following microcontroller corresponds to a RAM. Data to be
transmitted are written into the appropriate RAM area, and
data received are read out correspondingly. The microcontroller has to administer only a few bits (e.g., transmission
request).
CAN controllers with object storage are designed to take
as much strain as possible off the local microcontroller.
These CAN controllers require a greater chip area, however,
and are therefore more expensive. In addition to this, they
can only administer a limited number of chips.
CAN controllers are now available which combine both
principles of implementation. They have object storage, at
least one of which is designed as an intermediate buffer. For
this reason there is no longer any point in differentiating
between basicCAN and fullCAN.
As well as CAN controllers which support all functions of
the CAN protocol, there are also CAN chips which do not
require a following microcontroller. These CAN chips are
called serial link I/O (SLIO). CAN chips are CAN slaves and
have to be administered by a CAN master.
(2) CAN physical layer.
(a) Physical CAN connection. Data rates (up to 1 Mbit/s) necessitate a sufficiently steep pulse slope, which can be implemented only by using power elements. A number of physical
connections are basically possible. However, the users and
manufacturers group, CAN in Automation, recommends
the use of driver circuits in accordance with ISO 11898.
Integrated driver chips in accordance with ISO 11898 are
available from several companies (Bosch, Philips, Siliconix,
Zhang_Ch03.indd 300
5/13/2008 5:41:22 PM
301
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
and Texas Instruments). The international users and manufacturers group, CAN in Automation (CiA), also specifies several
mechanical connections (cable and connectors) (Fig. 3.14).
(b) Physical media. The basis for transmitting CAN messages
and for competing for bus access is the ability to represent a
dominant and a recessive bit value. This is possible for electrical and optical media so far. For electrical media the differential output bus voltages are defined in ISO 11898-2 and
ISO 11898-3, in SAE J2411, and ISO 11992. With optical
media the recessive level is represented by “dark” and the
dominant level by “light.”
The physical medium most commonly used to implement
CAN networks is a differentially driven pair of wires with
common return. For vehicle body electronics, single wire
bus lines are also used. Some efforts have been made to
develop a solution for the transmission of CAN signals on
the same line as the power supply.
The parameters of the electrical medium become important when the bus length is increased. Signal propagation,
Microcontroller
CAN controller
T×0 T×1
R×0 R×1
T×D
R×D Ref
+6 V
Rs
Vcc
100 nF
CAN transceiver
Gnd
CAN_L CAN_H
Bus termination
Bus termination
CAN_H
RT
CAN bus lines
RT
CAN_L
Figure 3.14 Physical CAN connection according to ISO 11898.
Zhang_Ch03.indd 301
5/13/2008 5:41:22 PM
302
INDUSTRIAL CONTROL TECHNOLOGY
the line resistance, and wire cross-sections are factors when
dimensioning a network. In order to achieve the highest possible bit rate at a given length, a high signal speed is required.
For long bus lines the voltage drops over the length of the
bus line. The wire cross-section necessary is calculated by
the permissible voltage drop of the signal level between the
two nodes farthest apart in the system and the overall input
resistance of all connected receivers. The permissible voltage drop must be such that the signal level can be reliably
interpreted at any receiving node.
(c) Network topology. Electrical signals on the bus are reflected
at the ends of the electrical line unless measures against that
have been taken. For the node to read the bus level correctly,
it is important that signal reflections are avoided. This is done
by terminating the bus line with a termination resistor at both
ends of the bus and by avoiding unnecessarily long stub lines
of the bus. The highest possible product of transmission rate
and bus length line is achieved by keeping as close as possible to a single line structure and by terminating both ends
of the line. Specific recommendations for this can be found
in the applicable standards (i.e., ISO 11898-2 and -3).
It is possible to overcome the limitations of the basic line
topology by using repeaters, bridges, or gateways. A repeater
transfers an electrical signal from one physical bus segment
to another segment. The signal is only refreshed and the
repeater can be regarded as a passive component comparable
to a cable. The repeater divides a bus into two physically
independent segments. This causes an additional signal propagation time. However, it is logically just one bus system.
A bridge connects two logically separated networks on
the data link layer (OSI Layer 2). This is so that the CAN
identifiers are unique in each of the two bus systems. Bridges
implement a storage function and can forward messages or
parts thereof in an independent time-delayed transmission.
Bridges differ from repeaters since they forward messages,
which are not local, whereas repeaters forward all electrical
signals including the CAN identifier. A gateway provides the
connection of networks with different higher layer protocols.
It therefore performs the translation of protocol data between
two communication systems. This translation takes place on
the application layer (OSI Layer 7).
(d) Bus access. For the connection between a CAN controller
chip and a two-wire differential bus, a variety of CAN transceiver chips according to different physical layer standards
are available ( ISO 11898-2 and -3, etc.).
Zhang_Ch03.indd 302
5/13/2008 5:41:22 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
303
This interface basically consists of a transmitting amplifier and a receiving amplifier transceiver = transmit and
receive). Aside from the adaptation of the signal representation between chip and bus medium, the transceiver has to
meet a series of additional requirements. As a transmitter it
provides sufficient driver output capacity and protects the
on-controller-chip driver against overloading. It also reduces
electromagnetic radiation. As a receiver the CAN transceiver
provides a defined recessive signal level and protects the
on-controller-chip input comparator against overvoltages on
the bus lines. It also extends the common mode range of the
input comparator in the CAN controller and provides sufficient input sensitivity. Furthermore, it detects bus errors such
as line breakage, short circuits, shorts to ground, etc. A
further function of the transceiver can also be the galvanic
isolation of a CAN node and the bus line.
(e) Physical CAN protocols. The CAN protocol defines the data
link layer and part of the physical layer in the OSI model,
which consists of seven layers. The International Standards
Organization (ISO) defined a standard, which incorporates
the CAN specifications as well as a part of physical layer:
the physical signaling, which comprises bit encoding and
decoding (Non-Return-to-Zero (NRZ)) as well as bit timing
and synchronization.
(i) Bit encoding. In the chosen NRZ bit coding the signal
level remains constant over the bit time and thus just one
time slot is required for the representation of a bit (other
methods of bit encoding are, e.g., Manchester or pulsewidth modulation). The signal level can remain constant
over a longer period of time; therefore, measures must
be taken to ensure that the maximum permissible interval between two signal edges is not exceeded. This is
important for synchronization purposes. Bit stuffing is
applied by inserting a complementary bit after five bits
of equal value. Of course the receiver has to unstuff the
stuff bits so that the original data content is processed.
(ii) Bit timing and synchronization. On the bit level (OSI
level one, physical layer), CAN uses synchronous bit
transmission. This enhances the transmitting capacity
but also means that a sophisticated method of bit synchronization is required. While bit synchronization in a
character-oriented transmission (asynchronous) is performed upon the reception of the start bit available with
each character, a synchronous transmission protocol is
just one start bit available at the beginning of a frame.
Zhang_Ch03.indd 303
5/13/2008 5:41:22 PM
304
INDUSTRIAL CONTROL TECHNOLOGY
To enable the receiver to read the messages correctly,
continuous resynchronization is required. Phase buffer
segments are, therefore, inserted before and after the
nominal sample point within a bit interval.
The CAN protocol regulates bus access by bitwise
arbitration. The signal propagation from sender to
receiver and back to the sender must be completed within
one bit time. For synchronization purposes, a further
time segment, the propagation delay segment, is needed
in addition to the time reserved for synchronization, the
phase buffer segments. The propagation delay segment
takes into account the signal propagation on the bus as
well as signal delays caused by transmitting and receiving nodes.
Two types of synchronization are distinguished: hard
synchronization at the start of a frame and resynchronization within a frame. After a hard synchronization the
bit time is restarted at the end of the sync segment.
Therefore, the edge, which caused the hard synchronization, lies within the sync segment of the restarted bit
time. Resynchronization shortens or lengthens the bit
time so that the sample point is shifted according to the
detected edge.
(iii) Interdependency of data rate and bus length. Depending
on the size of the propagation delay segment, the maximum possible bus length at a specific data rate (or the
maximum possible data rate at a specific bus length) can
be determined. The signal propagation is determined by
the two nodes within the system that are farthest apart
from each other. It is the time that it takes a signal to
travel from one node to the one farthest away (taking
into account the delay caused by the transmitting and
receiving node), synchronization and the signal from
the second node to travel back to the first one. Only then
can the first node decide whether its own signal level
(recessive in this case) is the actual level on the bus or
whether it has been replaced by the dominant level by
another node. This fact is important for bus arbitration.
(3) CAN application layer protocols. In the CAN world there are
different standardized application layer protocols. Some are very
specific and related to specific application fields. Examples of
CAN-based application layer protocols are given below:
(a) CANopen. CANopen is a CAN-based higher layer protocol.
It was developed as a standardized embedded network with
Zhang_Ch03.indd 304
5/13/2008 5:41:22 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
305
highly flexible configuration capabilities. CANopen was predeveloped in an Esprit project under the chairmanship of
Bosch. In 1995, the CANopen specification was handed over
to the CAN in Automation (CiA) international users’ and
manufacturers’ group. Originally, the CANopen communication profile was based on the CAN Application Layer
(CAL) protocol. Version 4 of CANopen (CiA DS 301) is
standardized as EN 50325-4.
The CANopen specifications cover application layer and
communication profile (CiA DS 301), as well as a framework
for programmable devices (CiA 302), recommendations for
cables and connectors (CiA 303-1), and SI units and prefix
representations (CiA 303-2). The application layer as well as
the CAN-based profiles is implemented in software.
Standardized profiles (device, interface, and application
profiles) developed by CiA members simplify the system
design job of integrating a CANopen network system. Offthe-shelf devices, tools, and protocol stacks are widely available at reasonable prices. For system designers, it is very
important to reuse application software. This requires not
only communication compatibility, but also interoperability
and interchange ability of devices. In the CANopen device and
interface profiles, defined application objects exist to achieve
the interchangeability of CANopen devices. CANopen is
flexible and open enough to enable manufacturer-specific
functionality in devices, which can be added to the generic
functionality described in the profiles.
CANopen unburdens the developer from dealing with
CAN-specific details such as bit-timing and implementationspecific functions. It provides standardized communication
objects for real-time data (Process Data Objects, PDO), configuration data (Service Data Objects, SDO), and special
functions (Time Stamp, Sync message, and Emergency message) as well as network management data (Boot-up message, NMT message, and Error Control).
(b) CAN Kingdom. CAN Kingdom unleashes the full power of
CAN. It gives system designers maximum freedom to create
their own systems, which is not bound to the CSMA/AMP
multimaster protocol of CAN but can create systems using
virtually any type of bus management and topology. CAN
Kingdom opens the possibility for a module designer to
design general modules without knowing which system they
will finally be integrated into and what type of higher layer
CAN protocol it will have. As the system designer can allow
Zhang_Ch03.indd 305
5/13/2008 5:41:22 PM
306
INDUSTRIAL CONTROL TECHNOLOGY
only specific modules to be used in the system, the cost
advantage of an open system can be combined with the security of a proprietary system!
Since the identifier in a CAN message not only identifies
the message but also governs the bus access, a key factor is
the enumeration of the messages. Another important factor
is to see to it that the data structure in the data field is the
same in both the transmitting and receiving modules. By
adopting a few simple design rules these factors can be fully
controlled and communication optimized for any system.
This is done during a short setup phase at the initialization of
the system. Including some modules not following the rules
of the CAN Kingdom into a CAN Kingdom system is even
possible. CAN Kingdom also enforces a conform documentation of modules and systems.
(c) DeviceNet. DeviceNet is a low-cost communications link
to connect industrial devices (such as limit switches, photoelectric sensors, valve manifolds, motor starters, process
sensors, bar code readers, variable frequency drives, panel
displays, and operator interfaces) to a network and eliminate
expensive hard wiring. The direct connectivity provides
improved communication between devices as well as important device-level diagnostics not easily accessible or available through hard wired I/O interfaces. DeviceNet is a
simple, networking solution that reduces the cost and time to
wire and install factory automation devices, while providing
interchangeability of “like” components from multiple
vendors. DeviceNet specifications have been developed
by the Open DeviceNet Vendor Association (ODVA) and
are internationally standardized. Buyers of the DeviceNet
Specification receive an unlimited, royalty-free license to
develop DeviceNet products.
(d) J1939-based higher layer protocols. A J1939 network connects electronic control units (ECU) within a truck and trailer
system. The J1939 specification—with its engine, transmission, and brake message definitions—is dedicated to diesel
engine applications. It is supposed to replace J1587/J1708
networks. Other industries adopted the general J1939 communication functions, in particular the J1939/21 and J1939/31
protocol definitions—they are required for any J1939compatible system. They added other physical layers and
defined other application parameters. The ISO standardized
the J1939-based truck and trailer communication (ISO
11992) and the J1939-based communication for agriculture
Zhang_Ch03.indd 306
5/13/2008 5:41:22 PM
307
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
and forestry vehicles (ISO 11783). The National Maritime
Electronics Association (NMEA) specified the J1939-based
communication for navigation systems in marine applications (NMEA 2000). One reason for the incorporation of
J1939 specifications into others is the fact that it makes sense
to reinvent the basic communication services. An industryspecific document defines the particular combination of layers for that industry.
CiA has developed several CANopen interface profiles for
J1939-based networks (CiA DSP 413). Gateways are defined
according to ISO 11992-2 and ISO 11992-3. In addition, the
CANopen profile family includes a framework for gateways
according to SAE J1939/71.
(4) CAN standards. The original specification is the Bosch specification. Version 2.0 of this specification is divided into two parts:
(a) Standard CAN (Version 2.0A). Uses 11 bit identifiers.
(b) Extended CAN (Version 2.0B). Uses 29 bit identifiers.
The two parts define different formats of the message frame,
with the main difference being the identifier length. There are
two ISO standards for CAN. The difference is in the physical
layer, where ISO 11898 handles high speed applications up to
1Mbit/s. ISO 11519 has an upper limit of 125 kbit/s.
(a) Part A and Part B compatibility. There are three types of CAN
controllers: Part A, Part B passive, and Part B (Table 3.1).
They are able to handle the different parts of the standard as
follows:
Most 2.0A controllers transmit and receive only standard
format messages, although some (known as 2.0B passive)
will receive extended format messages but then ignore them.
2.0B controllers can send and receive messages in both formats. Note that if 29 bit identifiers are used on a bus that
contains part A controllers, the bus will not work!
(b) CAN bus physical layer. The physical layer is not part of the
Bosch CAN standard. However, in the ISO standards transceiver characteristics are included. CAN transmits signals on
Table 3.1 CAN Part A and Part B Compatibility
Message Format\CAN
Chip Type
Part A
Part B Passive
Part B
11 bit ID
29 bit ID
Ok
Error!
Ok
Tolerated on the bus,
but ignored
Ok
Ok
Zhang_Ch03.indd 307
5/13/2008 5:41:22 PM
308
INDUSTRIAL CONTROL TECHNOLOGY
the CAN bus which consists of two wires, a CAN-High and
CAN-Low. These two wires operate in differential mode, that
is, they carry inverted voltages (to decrease noise interference). The voltage levels, as well as other characteristics of
the physical layer, depend on which standard is being used.
(i) ISO 11898. The voltage levels for a CAN network
which follows the ISO 11898 (CAN High Speed) standard are described in Table 3.2.
Note that for the recessive state, nominal voltage for
the two wires is the same. This decreases the power
drawn from the nodes through the termination resistors.
These resistors are 120 Ω and are located on each end of
the wires. Some people have played with using central
termination resistors (i.e., putting them in one place on
the bus). This is not recommended since that configuration will not prevent reflection problems.
(ii) ISO 11519. The voltage levels for a CAN network
which follows the ISO 11519 (CAN Low Speed) standard are described in Table 3.3.
ISO 115519 does not require termination resistors.
These are not necessary because the limited bit rates
(maximum 125 kB/s) make the bus insensitive to reflections. The voltage level on the CAN bus is recessive
when the bus is idle.
Table 3.2 ISO 11898 Parameters for CAN
Signal
CAN-High
CAN-Low
Recessive State (V)
Dominant State (V)
Min
Nominal
Max
Min
Nominal
Max
2.0
2.0
2.5
2.5
3.0
3.0
2.75
0.5
3.5
1.5
4.5
2.25
Table 3.3 ISO 11519 Parameters for CAN
Signal
CAN-High
CAN-Low
Zhang_Ch03.indd 308
Recessive State (V)
Dominant State (V)
Min
Nominal
Max
Min
Nominal
Max
1.6
3.1
1.75
3.25
1.9
3.4
3.85
0
4.0
1.0
5.0
1.15
5/13/2008 5:41:22 PM
309
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
(iii) Bus lengths. The maximum bus length for a CAN network depends on the bit rate used. It is required that the
wavefront of the bit signal has time to travel to the most
remote node and back again before the bit is sampled.
This means that if the bus length is near the maximum
for the bit rate used, one should choose the sampling
point with utmost care—on the other hand, one should
always do that!
Table 3.4 gives the different bus lengths and the corresponding maximum bit rates.
(iv) Cable. According to the ISO 11898 standard, the impedance of the cable shall be 120 ± 12 Ω. It should be
twisted par, shielded or unshielded. Work is in progress
on the single-wire standard SAE J2411.
3.2.1.4
Interbus
Interbus was one of the very first Fieldbuses to achieve widespread popularity. It continues to be popular because of its versatility, speed, diagnostic and autoaddressing capabilities. Physically, it has the appearance of
being a typical line-and-drop-based network, but in reality it is a serial ring
shift register. Each slave node has two connectors, one which receives data
and one which passes data onto the next slave.
Interbus technology provides an open Fieldbus system, which embraces
all the process I/Os required for almost any control system. Interbus is able
to fulfill essential requirements of high-performance control concepts, as
it is (1) a cost-effective solution with bus systems, which transmits data
serially and reduces the amount of parallel cabling required; (2) an open
and manufacturer-independent networking system, which can be easily
connected with existing control systems; (3) flexible with regard to future
modifications or expansions.
Table 3.4 CAN Bus Length
Bus Length (m)
Maximum Bit Rate (bit/s)
40
100
200
500
6 km
1 Mbit/s
500 kbit/s
250 kbit/s
125 kbit/s
10 kbit/s
Zhang_Ch03.indd 309
5/13/2008 5:41:23 PM
310
INDUSTRIAL CONTROL TECHNOLOGY
With its special features and an extensive product range, Interbus has
established itself successfully in all sectors of industry. Its traditional field
of application is the automotive industry, but Interbus is also increasingly
being used as an automation solution in other areas such as materials
handling and conveying, the paper and print industry, the food and beverage industry, building automation, the wood-processing industry, assembly and robotics applications, general mechanical engineering, and, more
recently, in process engineering. In addition to standard applications
for connecting a large number of sensors and actuators in the field to the
higher level control system via a serial bus system, Interbus can also
be used to fulfill a variety of special application requirements such as
(1) driving synchronically a control loop application in a mill train and
(2) alternative and changing bus configuration in a machining center.
(1) Operation mechanism. Interbus works with a master/slave access
method, in which the master also establishes the connection to
the higher level control or bus system. In terms of topology,
Interbus is a ring system with an active connection to communication devices. Starting at the Interbus master, the controller
board, all devices are actively connected on the ring system.
Each Interbus device (slave) has two separate lines for data transmission: one for forward data transfer and one for return data
transfer. This eliminates the need for a return line from the last to
the first device, necessary in a simple ring system.
The forward and return lines run in one cable. From the installation point of view, Interbus is similar to bus or linear structures,
as only one bus cable connects one device with the next. To
enable the structuring of an Interbus system, subring systems
(bus segments) can be formed on the main ring, the source of
which is the master. These subring systems are connected with
bus couplers (also known as bus terminal modules). Figure 3.15
illustrates the basic structure of an Interbus system with one main
ring and two subring systems.
The remote bus is installed from the controller board. Remote
bus devices and bus couplers are connected to the remote bus.
Each bus coupler connects the remote bus with a subring system.
There are two different types of subring system, which are available in different installation versions:
(a) The local bus (formally known as the I/O bus) is responsible
for local management, connects local bus devices, and is
typically used to form local I/O compact stations, for example, in the control cabinet. It is also available as a robust version for direct mounting on machines and systems.
Zhang_Ch03.indd 310
5/13/2008 5:41:23 PM
311
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
Host system
(IPC, PC, PLC)
256 remote bus devices, maximum
Controller board
(bus master)
4096 I/O points, maximum
512 devices in total
Localbus
Bus coupler
LB device
LB device
400 m
Remote bus
RB device
Remote bus branch
Bus coupler
RB device
RB device
400 m (1312.34 ft.)
RB device
RB—Remotebus
LB—Local bus
Figure 3.15 Basic structure of an Interbus system.
(b) The remote bus branch connects remote bus devices and
connects distributed devices over large distances. Remote
bus branches can be used to set up complex network topologies, which are ideal for complex technical processes distributed over large distances.
The Interbus remote bus cable forms an RS-485 connection and, because of the ring structure and the additional
need for an equalizing conductor between two remote bus
devices, it requires five cables.
Due to the different physical transmission methods, the local
bus is available with nine cables and TTL levels for short
distances (up to 1.5 m) and as a two-wire cable with a TTYbased current interface for medium distances (up to 10 m).
Due to the integrated amplifier function in each remote bus
device, the total expansion of the Interbus system can reach
13 km. To ensure that the system is easy to operate, the number
of Interbus devices is limited to a maximum of 512.
Zhang_Ch03.indd 311
5/13/2008 5:41:23 PM
312
INDUSTRIAL CONTROL TECHNOLOGY
Interbus works as a shift register, which is distributed
across all bus devices and uses the I/O-based summation
frame method for data transmission. Each bus device has
data memories, which are combined via the ring connection
of the bus system to form a large shift register. Figure 3.16
illustrates the data transmission principle. A data packet in
the summation frame is made available in the send shift register by the master. The data packet contains all data that is
to be transmitted to the bus devices (OUT data). The corresponding data registers in the bus devices contain the data to
be transmitted to the master (IN data) (Fig. 3.16a).
The OUT data is now transferred from the master to
the device and the IN data is transferred from the devices to
the master in one data cycle. The master starts by sending the
loop-back word through the ring. At the end of the data cycle,
Master
Loopback
OUT data 4
OUT data 3
OUT data 2
OUT data 1
Slave 1
IN data 1
Slave 2
IN data 2
Slave 4
IN data 4
IN data 3
Slave 1
OUT data 1
Slave 2
OUT data 2
Slave 4
OUT data 4
Slave 3
OUT data 3
Slave 3
(a)
Master
IN data 4
IN data 3
IN data 2
IN data 1
Loopback
(b)
Figure 3.16 Principle of data transmission on Interbus: (a) distribution of
data before a data cycle and (b) distribution of data after a data cycle.
Zhang_Ch03.indd 312
5/13/2008 5:41:23 PM
313
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
the master receives the loop-back word. The loop-back word
“pulls” the OUT data along behind it while “pushing” the IN
data along in front of it. This is called full duplex data transmission (Fig. 3.16b).
The devices do not have to be addressed explicitly as
the physical position of a device in the ring is known and the
master can position the information to be transmitted at this
point in the summation frame telegram. In the example,
the first data word after the loop-back word is addressed to
slave 4, for example.
The amount of user data to be pushed through the ring
corresponds to the total data length of all bus devices. The
bus couplers are integrated into the ring but do not provide
any user data. Data widths between 1 bit and 64 bytes per
data direction are permitted in one Interbus device.
Unlike local bus segments, whose components only really
differ in terms of installation technology, Interbus loop (sensor loop, IP 65 local bus) offers a new physical transmission
method. The individual devices are connected via a simple
two-wire unshielded cable to form a ring. The data and the
24 V power supply for up to 32 sensors are also supplied via
the cable. Figure 3.17 shows the configuration of an Interbus
loop segment.
Data is transmitted as load-independent current signals,
which have a higher level of immunity to interference than
the voltage signals normally used. The data to be transmitted
is modulated using Manchester code on the 24 V supply voltage (Interbus usually uses the NRZ code). The physical bus
Remote
bus
IN
UL
US
Interbus Loop segment
Loop device
1
10 m
(32.81 ft.)
Loop device
2
UA
Loop device
64
...
...
Bus coupler
Remote
bus
OUT
100 m (328.08 ft.)
Figure 3.17 Interbus loop segment (UL, power supply for the bus logic; US, power
supply for the Interbus loop; UA, local power supply for actuators).
Zhang_Ch03.indd 313
5/13/2008 5:41:23 PM
314
INDUSTRIAL CONTROL TECHNOLOGY
characteristics are converted by an appropriate bus terminal
module, which can be connected to the Interbus ring at any
point in a remote bus segment.
One of the main fields of application of Interbus loop is
the connection of individual devices with IP 65 and IP
54 connections directly in the system. An extensive range of
functions and devices is available as bus devices. The Interbus
protocol is not converted in any way in an Interbus loop,
which means that complex gateways are not required and an
Interbus loop segment can be used in conjunction with any
other type of Interbus device. Data scanning is absolutely
synchronous in all parts of the Interbus system. Despite this,
the high scanning speed is maintained.
An Interbus system is configured by connecting the bus
devices one after the other in a ring. Bus couplers segment
the ring according to the application requirements. With
Interbus G4 (Generation 4) and later, it is possible to set up
complex network topologies, which can be optimized for the
structure of the automation system, by integrating bus couplers with an additional bus connection. There are two ways
of structuring the configuration of this type of Interbus
network:
(i) divide the entire network into various levels;
(ii) assign segment-specific device numbers.
Both configuration methods are explained using the example of an Interbus network configuration with four levels, as
illustrated in Fig. 3.18. The network is split into four different levels starting with the bus master on the main remote
bus line as the first level. The branching secondary lines are
now assigned a second level. The devices connected to these
lines can form additional substructures, etc. In this way, a
nesting depth of up to 16 levels can be achieved. The sequence
is such that a local bus (formally known as the I/O bus) in a
remote bus segment is always assigned to the next level.
Segment-specific device numbers are assigned either automatically according to the physical configuration or they can
be freely specified by the user. The numbering comprises
two components:
<Device number> = <Bus segment number> • <Position
number in bus segment>.
According to this pattern, the second digit of the device
number for all remote devices is zero, for example, 1.0. The
second digit is only used by the local bus devices (e.g., I/O
Zhang_Ch03.indd 314
5/13/2008 5:41:27 PM
315
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
Bus
master
1.0
...
1.1
1.2
3.0
3.1
3.2
5.0
5.1
1.8
2.0
BK
...
3.8
4.0
BK
5.2
...
5.8
6.0
BK
8.0
7.0
8.1
Level 1
Level 2
Level 3
Level 4
BK—Bus coupler with additional remote bus
branch, e.g., IBS ST 24 BK RB-T
—Remotebus
—Localbus
Figure 3.18 Interbus network configuration with four levels.
modules) connected downstream of the remote device, for
example, 1.1.
Bus couplers with an additional remote bus branch appear
as two separate remote bus devices with one local bus/remote
bus branch, for example, bus coupler 1.0/2.0. When physically assigning this type of remote bus device, the remote
bus branch is assigned the next consecutive number, for
example, 3.0. Any additional subbranches on this branch are
assigned the next consecutive number, for example, 4.0, 5.0,
etc. The outgoing remote bus from the branch is counted as
the last component, for example, 8.0.
Device numbering is a structuring tool and should not be
confused with device addressing. Although the device numbers can be used for addressing purposes, this is not absolutely necessary.
(2) Interbus system devices
(a) Protocol chip. The most important element in the electrical
configuration of an Interbus device is the Interbus protocol
chip, which manages the complete summation frame protocol and provides the physical interface to the Interbus ring.
The bus master and Interbus slave devices use different
protocol chips according to their function in the Interbus
Zhang_Ch03.indd 315
5/13/2008 5:41:27 PM
316
INDUSTRIAL CONTROL TECHNOLOGY
system. Hardware solutions tailored to meet specific technical requirements are available for both Interbus master and
slave solutions.
The parts of the Interbus protocol that correspond to layers
1 and 2 of the OSI reference model are processed entirely in
the protocol machine. This means that basic devices require
additional software or processing power. The protocol machine
also provides physical access to the incoming (IB IN) and outgoing (IB OUT) Interbus interface.
Both shift registers—the ID register and data register—
operate as send and receive buffers in the ID and data cycle.
The application and/or higher protocol layers can access
this buffer via the MPM interface (Multifunction Pin,
MFP). The MFP interface can be set according to interface
requirements.
The data registers can be expanded with external registers
(ToExR, FromExR). The Interbus register chip SRE 1,
which, if required, can expand the shift register width of an
Interbus device to 64 bytes, is used for register expansion.
By default, the register width of the SUPI 3 is 8 bytes.
The diagnostic and report manager constantly monitors the
operation (on chip diagnostics). Any error descriptions that
are received, such as CRC errors, transient loss of medium,
voltage dips, etc., are saved to the ID send buffer and can be
read from there by the master at any time. This means that
unique error locations can even be identified for sporadic
errors that are difficult to diagnose.
The Interbus slave chip enables all Interbus device
variants for the remote and local bus to be implemented,
with the exception of those for Interbus loop. Interbus
loop also works with the Interbus protocol but uses a different physical transmission medium, which requires the
protocol chip on the physical interface to be of the same
format.
The standard master protocol chip for Interbus masters is
the IPMS microcontroller. The IPMS is designed to work
with a wide range of different processors. The master chip is
often used together with the Motorola CPU 68332. The master firmware, which manages the Interbus functions, is stored
in the EPROM.
Only actual bit transmission (Layer 1, parts of Layer 2)
takes place via the IPMS. The IPMS is connected to the relevant host system via a shared memory area, which, in its
simplest format, is a Dual Port Memory (DPM) or a Multiport
Zhang_Ch03.indd 316
5/13/2008 5:41:28 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
317
Memory (MPM). Interbus masters with IPMS are available
in various formats depending on the functions required.
(b) Local bus devices. An I/O bus interface for the two-wire
protocol with the SUPI 3 is an interface example used to
configure an ST local bus. The ST local bus operates with
four transmission signals, which, due to the ring format of
Interbus, are available twice at the incoming and outgoing
local bus interface as IN and OUT signal lines. In addition,
one incoming and one outgoing reset line are also available
in the local bus. The bus signals can be connected directly to
the local bus connectors, as the SUPI 3 meets the Interbus
specification for the local bus even without external drivers
and receivers.
(c) Remote bus devices. If it is used as a remote bus device, the
drivers and receivers required for differential signal transmission to RS-485 must be added to the SUPI. On the remote
bus, transmission takes place via two twisted pair cables
(DO+/DO–, DI+/DI–). Unlike the local bus, remote bus
devices require a dedicated power supply for the device
logic, as this is no longer provided via the bus cable.
(d) Interbus loop devices. Although Interbus loop devices also
operate with the standardized Interbus protocol, they do not
transmit voltage signals to RS-485, which is usually the case
on Interbus. Instead, they use load-independent current signals and Manchester coding to transmit the data and supply
voltage on one and the same bus line (loop). Due to the different physical transmission medium, a special protocol
chip, the IBS LPC, is available for Interbus loop. This chip is
an ASIC with approximately 7000 gate equivalents and is
supplied in QFP-44 housing. Special loop diagnostics are
integrated into the LPC 2 to extend the familiar diagnostic
functions of the SUPI 3.
(3) Protocol structure. The Interbus protocol, which has been optimized specifically for the requirements of automation technology, transmits single-bit data from limit switches or to switching
devices (process data) and complex programs or data records to
intelligent field devices (parameter data) with the same level of
efficiency and safety. Process data is transmitted in the fixed and
cyclic time slot in real-time conditions, while parameter data
comprises the acyclic transmission of larger volumes of data as
and when required. The continuity of an Interbus network for
very different tasks within an automation system—ensured in
essence by the standard protocol—is supported by additional
measures: (1) the adaptation of the physical transmission method
Zhang_Ch03.indd 317
5/13/2008 5:41:28 PM
318
INDUSTRIAL CONTROL TECHNOLOGY
“downward,” making it easy to install and connect individual
sensors and actuators; (2) the provision of “upward” interface
couplers to connect Interbus networks directly with factory and
or company networks (Ethernet networks); (3) the guarantee of
easy configuration, project planning, and diagnostics with uniform software tools.
The Interbus protocol is based on the OSI reference model and
for reasons of efficiency only takes into account layers 1, 2, and
7 (Fig. 3.19). Certain functions from layers 3 to 6 have been
included in application Layer 7.
The physical layer (Layer 1) defines both the time conditions
(such as the baud rate, permissible jitter, etc.) and the formats for
encoding information. The data link layer (Layer 2) ensures data
integrity and manages cyclic data transfer via the bus using the
summation frame protocol. The transmission methods and protocols on layers 1 and 2 can be found in DIN 19 258.
Following on from the data link layer, data access to the
Interbus devices takes place in the application layer as required
via two different data channels:
(a) The process data channel serves the primary use of Interbus
as a sensor and actuator bus. The cyclic exchange of I/O data
between the higher level control system and the connected
sensors/actuators takes place via this channel.
(b) The parameter channel supplements cyclic data exchange
with individual I/O points in connection-oriented message
exchange. This type of communication requires additional
data packing, as large volumes of information are being
exchanged between the individual communication partners.
Data is transmitted using communication services based on
the client/server model.
Interbus devices almost always have one process data channel.
A parameter channel can also be fitted as an optional extra.
Application layer
Parameter
data channel
Process
data
channel
Layer 2
Data link layer
Layer 1
Physical layer
Network management
Layer 7
Figure 3.19 Interbus protocol structure.
Zhang_Ch03.indd 318
5/13/2008 5:41:28 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
319
During operation, an Interbus system requires settings to be
made and provides a wide range of diagnostic information. This
information is processed by the network management on each
layer. More detailed information about readiness for operation,
error states, and statistical data can also be accessed and evaluated, and network configuration settings can be made.
The hybrid protocol structure of Interbus for the two different
data classes (process data and parameter data) and its independent data transmission via two channels is a decisive factor in the
performance of the Interbus protocol. The protocol enables the
creation of a seamless network comprising control systems and
intelligent field devices right down to individual sensors and
actuators.
3.2.1.5
Ethernets/Hubs
Ethernet is the major local area network (LAN) technology in use today,
and is widely used for the LAN-connected PCs and workstations. Ethernet
refers to the family of LAN products covered by the IEEE 802.3 standard,
and the technology can run over both optical fiber and twisted-pair cables.
Over the years, Ethernet has steadily evolved to provide additional performance and network intelligence. More than 300 million switched Ethernet
ports have been installed worldwide. Ethernet technology enjoys such wide
acceptance because it is easy to understand, deploy, manage, and maintain.
Ethernet is low cost and flexible, and supports a variety of network topologies. Although traditional, non-Ethernet-based industrial solutions have a
data rate of between 500 kbps to 12 Mbps, Ethernet technology can deliver
substantially higher performance. Because it is based on industry standards,
it can run and be connected over any Ethernet-compliant device from any
vendor.
This continual improvement has made Ethernet an excellent solution for
industrial applications. Today, the technology can provide four data rates.
(1) 10BASE-T Ethernet delivers performance of up to 10 Mbps over
twisted-pair copper cable.
(2) Fast Ethernet delivers a speed increase of 10 times the 10BASE-T
Ethernet specification (100 Mbps) while retaining many of
Ethernet’s technical specifications. These similarities enable organizations to use 10BASE-T applications and network management
tools on Fast Ethernet networks.
(3) Gigabit Ethernet extends the Ethernet protocol even further,
increasing speed 10-fold over Fast Ethernet to 1000 Mbps, or
Zhang_Ch03.indd 319
5/13/2008 5:41:28 PM
320
INDUSTRIAL CONTROL TECHNOLOGY
1 Gigabit/s. Because it is based upon the current Ethernet standard and compatible with the installed base of Ethernet and Fast
Ethernet switches and routers, network managers can support
Gigabit Ethernet without needing to retrain or learn a new
technology.
(4) 10 Gigabit Ethernet, ratified as a standard in June 2002, is an
even faster version of Ethernet. It uses the IEEE 802.3 Ethernet
media access control (MAC) protocol, the IEEE 802.3 Ethernet
frame format, and the IEEE 802.3 frame size. Because 10 Gigabit
Ethernet is a type of Ethernet, it can support intelligent Ethernetbased network services, interoperate with existing architectures,
and minimize users’ learning curves. Its high data rate of
10 Gigabits/s makes it a good solution to deliver high bandwidth
in wide area networks (WANs) and metropolitan area networks
(MANs).
(1) Industrial Ethernet. Recognizing that Ethernet is the leading networking solution, many industry organizations are porting the
traditional Fieldbus architectures to Industrial Ethernet. Industrial
Ethernet applies the Ethernet standards developed for data communication to manufacturing control networks (Fig. 3.20). Using
IEEE standards-based equipment, organizations can migrate all
or part of their factory operations to an Ethernet environment at
the pace they wish. Instead of using architectures composed of
multiple separate networks, Industrial Ethernet can unite a company’s administrative, control-level, and device-level networks
to run over a single network infrastructure. In an Industrial
Ethernet network, Fieldbus-specific information that is used to
control I/O devices and other manufacturing components are
embedded into Ethernet frames. Because the technology is based
Device profile
Valves
Drivers
Robotics
Application object library
Application
Layer 2 (Data link)
Layer 1 (Physics)
Explicit messages, I/O messages
Message routing, connection management
TCP
UDP
IP
Ethernet
MAC/LLC
Physical layer
QoS
parameter
Layer 4(Transport
Layer 3 (Network)
Data management services
Other
Fieldbus—Spec
Semiconductor
Figure 3.20 Using intelligent Ethernet for automation control.
Zhang_Ch03.indd 320
5/13/2008 5:41:28 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
321
on industry standards rather than on custom or proprietary standards, it is more interoperable with other network equipment and
networks.
Although Industrial Ethernet is based on the same industry
standards as traditional Ethernet technology, the implementation
of the two solutions is not always identical. Industrial Ethernet
usually requires more robust equipment and a very high level of
traffic prioritization when compared with traditional Ethernet
networks in a corporate data network.
The primary difference between Industrial Ethernet and traditional Ethernet is the type of hardware used. Industrial Ethernet
equipment is designed to operate in harsh environments. It
includes industrial-grade components, convection cooling, and
relay output signaling. And it is designed to operate at extreme
temperatures and under extreme vibration and shock. Power
requirements for industrial environments differ from data networks, so the equipment runs using 24 V of DC power. To maximize network availability, it also includes fault-tolerant features
such as redundant power supplies.
Industrial Ethernet environments also differ from traditional
Ethernet networks in their use of multicasting by hosts for certain applications. Industrial applications often use producer–
consumer communication, where information “produced” by
one device can be “consumed” by a group of other devices. In a
producer–consumer environment, the most important priority for
a multicast application is to guarantee that all hosts receive data
at the same time. A traditional Ethernet network, on the other
hand, focuses more on the efficient utilization of bandwidth in
general, and less on synchronous data access. To help optimize
synchronous data access, Industrial Ethernet equipment must
include the intelligence and Quality of Services (QoS) features
needed to enable organizations to appropriately prioritize multicast transmissions.
Ethernet technology can provide not only excellent performance for manufacturing applications, but a wide range of network security measures to provide both confidentiality and data
integrity. Confidentiality helps ensure that data cannot be
accessed by unauthorized users. Data integrity protects data from
intentional or accidental alteration.
These network security advantages protect manufacturing
devices like programmable logic controllers (PLCs) as well as
PCs, and apply to both equipment and data security. Manufacturers
can use many methods to help ensure network confidentiality and
integrity. These network security measures can be grouped into
Zhang_Ch03.indd 321
5/13/2008 5:41:28 PM
322
INDUSTRIAL CONTROL TECHNOLOGY
several categories, including access control and authentication,
and secure connectivity and management.
Access control is commonly implemented using firewalls or
network-based controls protecting access to critical applications,
devices, and data so that only legitimate users and information
can pass through the network. However, access-control technology is not limited to dedicated firewall devices. Any device that
can make decisions to permit or deny network traffic, such as an
intelligent switch, is part of an integrated access-control solution.
When designing an access-control solution, network administrators can set up filtering decisions based on a variety of criteria,
such as an IP address or TCP/UDP port number. Intelligent
switches can provide support for this advanced filtering to limit
network access to authorized users. At the same time, they can
enable organizations to enforce policy decisions based on the IP
or MAC address of a laptop or PLC.
Virtual LANs (VLANs) are another access-control solution,
providing the ability to create multiple IP subnets within an
Ethernet switch. VLANs provide network security and isolation
by virtually segmenting factory floor data from other data and
users. VLANs can also be used to enhance network performance,
separating low-priority end devices from high-priority devices.
Access controls can also include a variety of device or userauthentication services. Authentication services determine who
may access a network and what services they are authorized to
use. For example, the 802.1x authentication protocol provides
port-based authentication so that only legitimate devices can
connect to switch ports. Authentication services are an effective
complement to other network security measures in a manufacturing environment. To provide additional protection for manufacturing networks, organizations can take several approaches to
authenticate and encrypt network traffic. Using virtual private
network (VPN) technology, Secure Sockets Layer (SSL) encryption can be applied to application-layer data in an IP network.
Organizations can also use IP Security (IPSec) technology to
encrypt and authenticate network packets to thwart network
attacks such as sniffing and spoofing.
VPN client software, together with dedicated VPN network
hardware, can be used to encrypt device monitoring and
programming sessions, and to support strong authentication.
Manufacturers can also use Secure Shell (SSH) Protocol encryption for remote terminal logins to network devices. Version 3 of
Simple Network Management Protocol (SNMP) also offers
Zhang_Ch03.indd 322
5/13/2008 5:41:28 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
323
support for encryption and authentication of management commands and data.
(2) Network topology. Because factory floor applications run in real
time, the network must be available to users on a continuous
basis, with little or no downtime. Manufacturers can help ensure
network reliability using effective network design principles, as
well as intelligent networking services. Manufacturers deploying an Ethernet solution should design networks with redundant
paths to ensure that a single device outage does not take down
the entire network. Two network topologies most often used are
ring and hub-and-spoke. In hub-and-spoke designs (Fig. 3.21),
three layers of switches are usually installed. The first layer is
often referred to as the access layer. These switches provide connections for end-point devices like PLCs, robots, and Human–
Machine Interfaces (HMIs). A second layer called the distribution
layer provides connectivity between the access-layer switches.
And a third layer called the core layer provides connectivity to
other networks or to the Internet service provider (ISP) via routers. The distribution layer may include switches with routing
functions to provide inter-VLAN routing. Access-layer switches,
on the other hand, generally provide only Layer 2 (data link)
forwarding services. For optimum performance, network equipment at each of these layers must be aware of the information
contained within the Layer 2 through Layer 4 packet headers.
In ring topologies (Fig. 3.22), all devices are connected in a ring.
Each device has a neighbor to its left and right. If a connection
on one side of the device is broken, network connectivity can still
be maintained over the ring via the opposite side of the device. In
some situations, manufacturers install dual counterrotating rings
Figure 3.21 Hub-and-spoke network topology.
Zhang_Ch03.indd 323
5/13/2008 5:41:28 PM
324
INDUSTRIAL CONTROL TECHNOLOGY
Figure 3.22 Ring topology.
to maximize availability. In a ring topology, each switch functions as both an access-layer and a distribution-layer switch.
(3) Network protocols. To prevent loops from being formed in the
network when devices are interconnected via multiple paths,
some organizations use the Spanning Tree Protocol. If a problem
occurs on a network node, this protocol enables a redundant
alternative link to automatically come back online. The traditional Spanning Tree Protocol has been considered too slow for
industrial environments. To address these performance concerns,
the IEEE standards committee has ratified a new Rapid Spanning
Tree Protocol (802.1w). This protocol provides subsecond convergence times that vary between 200 and 800 ms, depending on
network topology. Using 802.1w, organizations can enjoy the
benefits of Ethernet networks, with the performance and reliability that manufacturing applications demand. Another spanningtree option is Multiple Spanning Tree Protocol (802.1s). This
enables VLANs to be grouped into spanning-tree instances. Each
instance has a spanning-tree topology that is independent from
other spanning-tree instances. This architecture provides multiple forwarding paths for data traffic, enables load balancing, and
reduces the number of spanning-tree instances needed to support
a large number of VLANs.
Ethernet switches provide excellent connectivity and performance; however, each switch is another device that must be managed on the factory floor. To make switched Ethernet networks
easy to support and maintain, intelligent switches include built-in
management capabilities. These intelligent features make it easy
to connect manufacturing devices to the network, without creating additional configuration tasks. And they help minimize network downtime if part of the network should fail. One of the
most useful intelligent features in a switched Ethernet network is
Option 82. In an Ethernet network, Dynamic Host Configuration
Zhang_Ch03.indd 324
5/13/2008 5:41:29 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
325
Protocol (DHCP) lets devices dynamically acquire their IP
addresses from a central server. The DHCP server can be configured to give out the same address each time or generate a dynamic
one from a pool of available addresses.
Because the interaction of the factory floor devices requires
specific addresses, Industrial Ethernet networks usually do not
use dynamic address pools. However, static addresses can have
drawbacks. Because they are linked to the MAC address of the
client, and because the MAC address is often hard-coded in the
network interface of the client device, the association is lost
when a client device fails and needs to be replaced. Extended
fields in the DHCP packet can be filled in by the switch, indicating the location of the device requesting an IP address. The 82nd
optional field, called Option 82, carries the specific port number
and the MAC address of the switch that received the DHCP
request. This modified request is sent on to the DHCP server. If
an access server is Option 82-aware, it can use this information
to formulate an IP address based on the Option 82 information.
Effective use of Option 82 enables manufacturers to minimize
administrative demands and maintain maximum network uptime
even in the event of the failure of individual devices.
Because manufacturing processes depend on the precise
synchronization of processes, network determinism must be
optimized to deliver the best possible performance. Data must
be prioritized using QoS to ensure that critical information is
received first. The multicast applications that are prevalent in
manufacturing environments must be well managed using
Internet Group Management Protocol (IGMP) snooping.
Many Industrial Ethernet applications depend on IP multicast
technology. IP multicast allows a host, or source, to send packets
to another group of hosts called receivers anywhere within the
IP network using a special form of IP address called the IP multicast group address. While traditional multicast services, such as
video or multimedia, tend to scale with the number of streams,
Industrial Ethernet multicast applications do not. Industrial Ethernet
environments use a producer–consumer model, where devices
generate data called “tags” for consumption by other devices. The
devices that generate the data are producers and the devices
receiving the information are consumers. Multicast is more efficient than unicast, because consumers will often want the same
information from a particular producer. Each device on the network can be both a producer and a consumer of traffic.
While most devices generate very little data, networks with a
large number of nodes can generate a large amount of multicast
Zhang_Ch03.indd 325
5/13/2008 5:41:29 PM
326
INDUSTRIAL CONTROL TECHNOLOGY
traffic, which can overrun end devices in the network. Using
mechanisms like QoS and IGMP snooping, organizations can control and manage multicast traffic in manufacturing environments.
Many manufacturing applications depend on multicast traffic,
which can introduce performance problems in the network. To
address these challenges in an Industrial Ethernet environment,
organizations can deploy IGMP snooping. IGMP snooping limits the flooding of multicast traffic by dynamically configuring
the interfaces so that multicast traffic is forwarded only to interfaces associated with IP multicast devices. In other words, when
a multicast message is sent to the switch, the switch forward the
message only to the interfaces that are interested in the traffic.
This is very important because it reduces the load of traffic
traversing through the network. It also relieves the hosts from
processing frames that are not needed. In a producer–consumer
model used by Industrial Ethernet, IGMP snooping can limit
unnecessary traffic from the I/O device that is producing, so that it
only reaches the device consuming that data. Messages delivered
to a given device that were intended for other devices consume
resources and slow performance, so networks with many multicasting devices will suffer performance issues if IGMP snooping
or other multicast limiting schemes are not implemented.
The IGMP snooping feature allows Ethernet switches to “listen” to the IGMP conversation between hosts. With IGMP snooping, the Ethernet switch examines the IGMP traffic coming to the
switch and keeps track of multicast groups and member ports.
When the switch receives an “IGMP join” report from a host for
a particular multicast group, the switch adds the host port number to the associated multicast forwarding table entry. When it
receives an IGMP “leave group” message from a host, it removes
the host port from the table entry. After the switch relays the
IGMP queries, it deletes entries periodically if it does not receive
any IGMP membership reports from the multicast clients. A
Layer 3 router normally performs the querying function.
When IGMP snooping is enabled in a network with Layer 3
devices, the multicast router sends out periodic IGMP general
queries to all VLANs. The switch responds to the router queries
with only one “join” request per MAC multicast group. The
switch then creates one entry per VLAN in the Layer 2 forwarding table for each MAC group from which it receives an IGMP
join request. All hosts interested in this multicast traffic send
“join” requests and are added to the forwarding table entry.
Layer 2 multicast groups learned through IGMP snooping
are dynamic. However, in a managed switch, organizations
Zhang_Ch03.indd 326
5/13/2008 5:41:29 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
327
can statically configure MAC multicast groups. This static setting supersedes any automatic manipulation by IGMP snooping.
Multicast group membership lists can consist of both userdefined settings and settings learned via IGMP snooping.
(4) Quality of service (QoS). An Industrial Ethernet network may
transmit many different types of traffic, from routine data to critical control information, or even bandwidth-intensive video or
voice. The network must be able to distinguish among and give
priority to different types of traffic.
To address these issues, organizations can implement QoS
using several techniques. QoS involves three important steps.
First, different traffic types in the network need to be identified
through classification techniques. Second, advanced buffermanagement techniques need to be implemented to prevent highpriority traffic from being dropped during congestion. Finally,
scheduling techniques need to be incorporated to transmit highpriority traffic from queues as quickly as possible.
In Layer 2 switches on an Ethernet network, QoS usually prioritizes native, encapsulated Ethernet frames, or frames tagged
with 802.1p class of service (CoS) specifications. More advanced
QoS mechanisms take this definition a step further. For example,
advanced Ethernet switches can study and interpret the flow of
QoS traffic as it is processed through the switch.
A switch can be configured to prioritize frames based on given
criteria at different layers of the OSI reference model. For example, traffic could be prioritized according to the source MAC
address (in Layer 2) or the destination TCP port (in Layer 4).
Any traffic traveling through the interface to which this QoS is
applied is classified, and tagged with the appropriate priority.
Once a packet has been classified, it is then placed in a holding
queue in the switch, and scheduled based on the scheduling algorithm desired.
In an Industrial Ethernet application, real-time I/O control
traffic would share network resources with configuration (FTP)
and data-collection flows, as well as other traffic, in the upper
layers of the OSI reference model. By using QoS to give high
priority to real-time UDP control traffic, organizations can prevent delay or jitter from affecting any control functions.
3.2.2
Interfaces
Devices connect to the microprocessor using an interface bus. The specification of this bus defines speed between the microprocessor and the
connected device that greatly affects the performance of the microprocessor.
Zhang_Ch03.indd 327
5/13/2008 5:41:29 PM
328
INDUSTRIAL CONTROL TECHNOLOGY
Peripherals can connect to the controller (or computer) using either an
internal or an external interface. This subsection lists main interfaces popularly used in industrial control.
3.2.2.1
PCI, ISA, and PCMCIA
(1) PCI bus. Intel has developed a standard interface, named the PCI
(Peripheral Component Interface/Interconnect) local bus for
microprocessors. This technology allows fast memory, disk and
video access. The PCI bus is now the main interface bus used in
most industrial controllers, and is rapidly replacing the ISA bus
for internal interface devices. It is a very adaptable bus and most
external buses, such as SCSI and USB connect to the processor
via the PCI bus.
The PCI bus transfers data using the system clock and can
operate over a 32- or 64-bit data path. The high transfer rates
used in PCI architecture machines limit the number of the PCI
bus interfaces to two to three, normally the graphics adapter and
hard disk controller. If data is transferred at 64 bits at a rate of
33 MHz, then the maximum transfer rate is 264 MB/s.
Detailed descriptions for the PCI bus system architecture, the
PCI operation, the bus arbitration, the bus configuration, and PCI
bus interrupt handling are given in both Sections 2.1.4 and 7.3.1.
(2) ISA bus. IBM developed the Industry Standard Architecture or
ISA bus for their 80286-based AT (advanced technology) computer. It had the advantage of being able to deal with 16 bits of
data at a time. An extra edge connector gives compatibility with
the PC bus. This gives an extra 8 data bits and four address lines.
Thus, the ISA bus has a 16-bit data and a 24-bit address bus. This
gives a maximum of 16 MB of addressable memory and, like the
PC bus, it uses a fixed clock rate of 8 MHz. The maximum data
rate is thus 2 bytes (16 bits) per clock cycle, giving a maximum
throughput of 16 MB/s.
IBM’s PC/AT was designed with an expansion bus which not
only provided for taking advantage of the new technology, but
also remained compatible with the older style 8-bit XT add-in
boards. Anticipating that advances in processors would again
outpace advances in bus technology, the PC/AT was designed
with two separate oscillators. In this way, the microprocessor
and expansion bus could be run on different clocks with different
speeds. Therefore, a controller or computer running a newer processor with 33 MHz clock speed could also run its expansion bus
at an 8 MHz clock rate. ISA cards are more cumbersome to
Zhang_Ch03.indd 328
5/13/2008 5:41:29 PM
329
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
install than other cards because I/O addresses, interrupts, and
clock speed must be set using jumpers and switches on the card
itself. The other bus options which use software to set these
parameters are called Plug & Play. While there is nothing inferior about using jumpers and switches, it can be more intimidating for novice users. The ISA system, however, does not have
a central registry from which to allocate system resources.
Consequently, each device behaves as though it has sole access
to system resources such as DMA, I/O ports, IRQs, and memory.
Obviously, this can cause problems when using multiple add-in
boards in a single system.
In practice there is no speed difference between running many
serial communication peripherals using a PCI or an ISA bus,
though the PCI advantage is obvious for high-speed devices such
as video cards. Thus, there is no reason to convert your current
ISA serial communication systems to PCI, as ISA will provide
equivalent functionality, generally at a lower price. However, if
you are starting a new installation using a PC with few or no (as
is increasingly the case today) ISA slots, or you prefer using
Plug & Play cards, then you should consider using PCI adapters.
Figure 3.23 shows a typical connection to the ISA bus. The
ALE (sometimes known as BALE) controls the address latch;
when active low, it latches the address lines A2–A19 to the ISA
bus. The address is latched when ALE goes from a high to a low.
The Pentium’s data bus is 64 bits wide, whereas the ISA expansion bus is 16 bits wide. It is the bus controller’s function to steer
D0–D31
Processor
Data
latch
D0–D15
Memory
Address
latch
BE0
–BE3
A2–A19
ALE
ISA bus
A2–A31
A0
Bus
controller
A1
SBHE
M16
IO16
Figure 3.23 ISA bus connections.
Zhang_Ch03.indd 329
5/13/2008 5:41:29 PM
330
INDUSTRIAL CONTROL TECHNOLOGY
data between the processor and the slave device for either 8-bit
or 16-bit communications. The following are the descriptions of
the ISA signals.
(a) SA19 to SA0. System Address bits 19:0 are used to address
memory and I/O devices within the system. These signals
may be used along with LA23 to LA17 to address up to
16 MB of memory. Only the lower 16 bits are used during
I/O operations to address up to 64K I/O locations. SA19 is
the most significant bit. SA0 is the least significant bit. These
signals are gated on the system bus when BALE is high and
are latched on the falling edge of BALE. They remain valid
throughout a read or write command. These signals are normally driven by the system microprocessor or DMA controller, but may also be driven by a bus master on an ISA
board that takes ownership of the bus.
(b) LA23 to LA17. Unlatched Address bits 23:17 are used to
address memory within the system. They are used along
with SA19 to SA0 to address up to 16 MB of memory. These
signals are valid when BALE is high. They are “unlatched”
and do not stay valid for the entire bus cycle. Decodes
of these signals should be latched on the falling edge of
BALE.
(c) AEN. Address Enable is used to de-gate the system microprocessor and other devices from the bus during DMA transfers. When this signal is active, the system DMA controller
has control of the address, data, and read/write signals. This
signal should be included as part of ISA board select decodes
to prevent incorrect board selects during DMA cycles.
(d) BALE. Buffered Address Latch Enable is used to latch the
LA23 to LA17 signals or decodes of these signals. Addresses
are latched on the falling edge of BALE. It is forced high
during DMA cycles. When used with AEN, it indicates a
valid microprocessor or DMA address.
(e) CLK. System Clock is a free-running clock typically in the
8–10 MHz range, although its exact frequency is not guaranteed. It is used in some ISA board applications to allow
synchronization with the system microprocessor.
(f) SD15 to SD0. System Data serves as the data bus bits for
devices on the ISA bus. SD15 is the most significant bit.
SD0 is the least significant bit. SD7 to SD0 are used for
transfer of data with 8-bit devices. SD15 to SD0 are used
for transfer of data with 16-bit devices. Sixteen-bit devices
transferring data with eight-bit devices convert the transfer
into two eight-bit cycles using SD7 to SD0.
Zhang_Ch03.indd 330
5/13/2008 5:41:30 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
331
(g) –DACK0 to –DACK3 and –DACK5 to –DACK7. DMA
Acknowledge 0 to 3 and 5 to 7 are used to acknowledge
DMA requests on DRQ0 to DRQ3 and DRQ5 to DRQ7.
(h) DRQ0 to DRQ3 and DRQ5 to DRQ7. DMA Requests are
used by ISA boards to request service from the system DMA
controller or to request ownership of the bus as a bus master
device. These signals may be asserted asynchronously. The
requesting device must hold the request signal active until
the system board asserts the corresponding DACK signal.
(i) –I/O CH CK. I/O Channel Check signal may be activated by
ISA boards to request that a nonmaskable interrupt (NMI)
be generated to the system microprocessor. It is driven active
to indicate an incorrect error has been detected.
(j) I/O CH RDY. I/O Channel Ready allows slower ISA boards
to lengthen I/O or memory cycles by inserting wait states.
This signals normal state is active high (ready). ISA boards
drive the signal inactive low (not ready) to insert wait states.
Devices using this signal to insert wait states should drive it
low immediately after detecting a valid address decode and
an active read or write command. The signal is released high
when the device is ready to complete the cycle.
(k) –IOR. I/O Read is driven by the owner of the bus and
instructs the selected I/O device to drive read data onto the
data bus.
(l) –IOW. I/O Write is driven by the owner of the bus and
instructs the selected I/O device to capture the write data on
the data bus.
(m) IRQ3 to IRQ7 and IRQ9 to IRQ12 and IRQ14 to IRQ15.
Interrupt Requests are used to signal the system microprocessor that an ISA board requires attention. An interrupt
request is generated when an IRQ line is raised from low to
high. The line must be held high until the microprocessor
acknowledges the request through its interrupt service routine. These signals are prioritized with IRQ9 to IRQ12 and
IRQ14 to IRQ15 having the highest priority (IRQ9 is the
highest) and IRQ3 to IRQ 7 having the lowest priority (IRQ7
is the lowest).
(n) –SMEMR. System Memory Read instructs a selected memory
device to drive data onto the data bus. It is active only when
the memory decode is within the low 1 MB of memory
space. SMEMR is derived from MEMR and a decode of the
low 1 MB of memory.
(o) –SMEMW. System Memory Write instructs a selected memory device to store the data currently on the data bus. It is
Zhang_Ch03.indd 331
5/13/2008 5:41:30 PM
332
INDUSTRIAL CONTROL TECHNOLOGY
active only when the memory decode is within the low 1 MB
of memory space. SMEMW is derived from MEMW and a
decode of the low 1 MB of memory.
(p) –MEMR. Memory Read instructs a selected memory device
to drive data onto the data bus. It is active on all memory
read cycles.
(q) –MEMW. Memory Write instructs a selected memory device
to store the data currently on the data bus. It is active on all
memory write cycles.
(r) –REFRESH. Memory Refresh is driven low to indicate a
memory refresh operation is in progress.
(s) OSC. Oscillator is a clock with a 70 ns period (14.31818
MHz). This signal is not synchronous with the system clock
(CLK).
(t) RESET DRV. Reset Drive is driven high to reset or initialize
system logic upon power up or subsequent system reset.
(u) TC. Terminal Count provides a pulse to signal a terminal
count has been reached on a DMA channel operation.
(v) –MASTER. Master is used by an ISA board along with a
DRQ line to gain ownership of the ISA bus. Upon receiving
a –DACK a device can pull –MASTER low which will allow
it to control the system address, data, and control lines. After
–MASTER is low, the device should wait one CLK period
before driving the address and data lines, and two clock
periods before issuing a read or write command.
(w) –MEM CS16. Memory Chip Select 16 is driven low by a
memory slave device to indicate it is capable of performing
a 16-bit memory data transfer. This signal is driven from a
decode of the LA23 to LA17 address lines.
(x) –I/O CS16. I/O Chip Select 16 is driven low by an I/O slave
device to indicate it is capable of performing a 16-bit I/O
data transfer. This signal is driven from a decode of the
SA15 to SA0 address lines.
(y) –0WS. Zero Wait State is driven low by a bus slave device to
indicate it is capable of performing a bus cycle without
inserting any additional wait states. To perform a 16-bit
memory cycle without wait states, –0WS is derived from an
address decode.
(z) –SBHE. System Byte High Enable is driven low to indicate
a transfer of data on the high half of the data bus
(D15–D8).
(3) PCMCIA (PC Card). The Personal Computer Memory Card
International Association (PCMCIA) interface allows small, thin
cards to be plugged into laptop, notebook, or palm computers.
Zhang_Ch03.indd 332
5/13/2008 5:41:30 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
333
The interface was originally designed for memory cards (Version
1.0), but it has since been adopted for many other types of adapters (Version 2.0), such as fax/modems, sound cards, local area
network cards, CD-ROM controllers, digital I/O cards, and so
on. Most PCMCIA cards comply with either PCMCIA Type II or
Type III. Type I cards are 3.3 mm thick, Type II take cards up to
5 mm thick, and Type III allows cards up to 10.5 mm thick. A
new standard, Type IV, takes cards that are thicker than 10.5 mm.
Type II interfaces can accept Type I cards, Type III accept Type I
and Type II, and Type IV interfaces accept Type I, II, and III.
The PCMCIA standard uses a 16-bit data bus (D0–D15) and a
26-bit address bus (A0–A25), which gives an addressable memory of 226 bytes (64 MB). The memory is arranged as: (1) Common
memory and attribute memory, which gives a total addressable
memory of 128 MB. (2) I/O addressable space of 65,536 (64K)
8-bit ports.
The PCMCIA interface allows the PCMCIA device to map
into the main memory or into the I/O address space. For example, a modem PCMCIA device would map its registers into the
standard COM port addresses (such as 3F8h–3FFh for COM1 or
2F8h–2FF for COM2). Any accesses to the mapped memory
area will be redirected to the PCMCIA rather than the main memory or I/O address space. These mapped areas are called windows. A window is defined with a START address and a LAST
address.
3.2.2.2
IDE
The most popular interface for hard disk drives is the integrated drive
electronics (IDE) interface. Its main advantage is that the hard disk controller is built into the disk drive and the interface to the motherboard consists simply of a stripped-down version of the ISA bus. The most common
standard is the ANSI-defined ATA-IDE standard. It uses a 40-way ribbon
cable to connect to 40-pin header connectors. The standard allows for the
connection of two disk drives in a daisy-chain configuration. This can
cause problems because both drives have controllers within their drives.
The primary drive is assigned as the master, and the secondary driver is the
slave. A drive is set as a master or a slave by setting jumpers on the disk
drive. They can also be set by software using the cable select pin on the
interface. The specifications for the IDE are
(1) Maximum of two devices (hard disks)
(2) Maximum capacity for each disk of 528 MB
Zhang_Ch03.indd 333
5/13/2008 5:41:30 PM
334
INDUSTRIAL CONTROL TECHNOLOGY
(3) Maximum cable length of 18 in.
(4) Data transfer rates of 3.3, 5.2, and 8.3 MB/s.
A new standard called as enhanced IDE (E-IDE) allows for higher
capacities than IDE has
(1) Maximum of four devices (hard disks)
(2) Uses two ports (for master and slaves)
(3) Maximum capacity for each disk of 8.4 GB
(4) Maximum cable length of 18 in.
(5) Data transfer rates of 3.3, 5.2, 8.3, 11.1, and 16.6 MB/s.
The PC (Personal Computer, in this chapter) is now a highly integrated
system. The main elements of modern systems are the processor, the systems controller and the PCI IDE/ISA accelerator, as illustrated in Fig. 3.24.
The system controller provides the main interface between the processor
and the level-2 cache, the DRAM, and the PCI bus. It is one of the most
important devices in the system and allows data to flow to and from the
processor in the correct way. The PCI bus links to interface devices and also
the PCI IDE/ISA accelerator (such as PIIX4 device). The PCI IDE/ISA
device then interfaces to other buses, such as IDE and ISA. The IDE interface has separate signals for both the primary and secondary IDE drives;
these are electrically isolated, which allows drives to be swapped easily
without affecting the other port.
The PCI IDE/ISA accelerator is a massively integrated device (the
PIIX4 has 324 pins) and provides for an interface to other buses, such as
Processor
Level 2
cache
System
bus
System
controller
(PAC/ MTXC)
Cache bus
DRAM
DRAM bus
PCI bus
DMA
PCI IDE/ISA accelerator
(PIIX 4)
interrupts
ISA bus
Reset
USB bus
Power
management
Primary IDE
Secondary IDE
X-bus
SDIORDY#
SDIOW#
SDIOR#
SDDREQ
SDDACK#
SDD [15:0]
SDA [2.0
SDCS3#
SDCS1#
PDIORDY#
PDIOW#
PDIOR#
PDDREQ
PDDACK#
PDD [15:0]
PDD [2:0
PDS3#
PDCS1#
Figure 3.24 IDE system connections.
Zhang_Ch03.indd 334
5/13/2008 5:41:30 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
335
USB and X-Bus. It also handles the interrupts from the PCI bus and ISA
bus. It thus has two integrated 82C59 interrupt controllers, which support
up to 15 interrupts. The PCI IDE/ISA accelerator also handles DMA transfers (on up to 8 channels), and thus has two integrated 82C37 DMA controllers. Along with this, it has an integrated 82C54, which provides for the
system timer, DRAM refresh signal, and the speaker tone output.
The IDE (or AT bus) is the de facto standard for most hard disks in PCs.
It has the advantage over older type interfaces that the controller is integrated into the disk drive. Thus, the computer only has to pass high-level
commands to the unit and actual control can be achieved with the integrated controller. Several companies developed a standard command set
for an AT attachment (ATA). Commands include
(1) read sector buffer—reads contents of the controller’s sector buffer
(2) write sector buffer—writes data to the controller’s sector buffer
(3) check for active
(4) read multiple sectors
(5) lock drive door.
The control of the disk is achieved by passing a number of high-level
commands through a number of I/O port registers.
3.2.2.3
SCSI
Small computer systems interface (SCSI) has an intelligent bus subsystem and can support multiple devices cooperating concurrently. Each
device is assigned a priority. The main types of SCSI are
(1) SCSI-I. SCSI-I transfers at rate of 5MB/s with an 8-bit data bus
and seven devices per controller.
(2) SCSI-II. SCSI-II supports for SCSI-I and with one or more of the
following:
(a) Fast SCSI, which uses a synchronous transfer to give 10 MB/s
transfer rate. The initiator and target initially negotiate to see
if they can both support synchronous transfer. If they can,
they then go into a synchronous transfer mode.
(b) Fast and wide SCSI-2, which doubles the data bus width to
16 bits to give 20 MB/s transfer rate.
(c) Fifteen devices per master device.
(d) Tagged command queuing (TCQ), which greatly improves
performance and is supported by Windows, NetWare, and
OS-2.
Zhang_Ch03.indd 335
5/13/2008 5:41:30 PM
336
INDUSTRIAL CONTROL TECHNOLOGY
(e) Multiple commands sent to each device.
(f) Commands executed in whatever sequence will maximize
device performance.
(3) Ultra SCSI (SCSI-III). Ultra SCSI operates either as 8-bit or
16-bit with either 20 or 40 MB/s transfer rate (Table 3.5).
SCSI standard uses a 50-pin header connector and a ribbon cable to connect to up to eight devices. It overcomes the problem existing in IDE,
where devices have to be assigned as a master and a slave. SCSI and
fast SCSI transfer one byte at a time with a parity check on each byte.
SCSI-II, wide SCSI, and ultra SCSI use a 16-bit data transfer and a 68-pin
connector.
An SCSI bus is made of an SCSI host adapter connected to a number of
SCSI units via SCSI bus. As all units connect to a common bus, only two
units can transfer data at a time, either from one SCSI unit to another or
from one SCSI unit to the SCSI host. The great advantage of this transfer
is that it does not involve the processor.
Each unit on an SCSI unit is assigned a SCSI ID address. In the case of
SCSI-I, this ranges from 0 to 7 (where 7 is normally reserved for a tape
drive). The host adapter takes one of the addresses; thus a maximum of
seven units can connect to the bus. Most systems allow the units to take on
any SCSI ID address, but older systems required boot drives to be connected to a specific SCSI address. When the system is initially booted, the
host adapter sends out a Start Unit command to each SCSI unit. This allows
each of the units to start in an orderly manner (and not overload the local
power supply). The host will start with the highest priority address (ID = 7)
and finishes with the lowest address (ID = 0). Typically, the ID is set with
a rotating switch selector or by three jumpers.
Table 3.5 SCSI Types
Indices
Types
Data
Bus
(bits)
Transfer
Rate
(MB/s)
Tagged
Command
Queuing
Parity
Checking
Maximum
Devices
Pins on
Cable and
Connector
SCSI-I
SCSI-II
fast
SCSI-III
fast/wide
Ultra
SCSI
8
8
5
10
(10 MHz)
20
(10 MHz)
40
(20 MHz)
No
Yes
Option
Yes
7
7
50
50
Yes
Yes
15
68
Yes
Yes
15
68
Zhang_Ch03.indd 336
16
16
5/13/2008 5:41:30 PM
337
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
SCSI defines an initiator control and a target control. The initiator
requests a function from a target, which then executes the function, as
illustrated in Fig. 3.25, where the initiator effectively takes over the bus for
the time to send a command and the target executes the command and then
contacts the initiator and transfers any data. The bus will then be free for
other transfers. Table 3.6 gives the definitions of main SCSI signals. Each
of the control signals can be true or false. They can be either OR-tied
driven or Non-OR-tied driven. In OR-tied driven, the driver does not drive
the signal to the false state. In this case, the bias circuitry of the bus terminators pulls the signal false whenever it is released by the drivers at every
SCSI device. If any driver is asserted, then the signal is true. The BSY, SEL,
and RST signals are OR-tied. In the ordinary operation of the bus, the BSY
and RST signals may be simultaneously driven true by several drivers.
However, in Non-OR-tied driven, the signal may be actively driven false.
Initiator
SCSI bus
Function
request
Target
Function
execution
Figure 3.25 Initiator and target in SCSI.
Table 3.6 SCSI Main Signals
Signals
Definitions
BSY
ACK
Indicates that the bus is busy, or not (an OR-tied signal).
Activated by the initiator to indicate an acknowledgment
for an REQ information transfer handshake.
When active (low) resets all the SCSI devices (an
OR-tied signal).
Activated by the initiator to indicate the attention state.
Activated by the target to indicate the message phase.
Activated by the initiator; it is used to select a
particular target device (an OR-tied signal).
Activated by the target to identify whether there is data
or control on the SCSI bus.
Activated by the target to acknowledge a request for an
ACK information transfer handshake.
Activated by the target to show the direction of the
data on the data bus. Input defines that data is an input
to the initiator, else it is an output.
RST
ATN
MSG
SEL
C/D (control/data)
REQ
I/O (input/output)
Zhang_Ch03.indd 337
5/13/2008 5:41:30 PM
338
INDUSTRIAL CONTROL TECHNOLOGY
No signals other than BSY, RST, and D(PARITY) are driven simultaneously by two or more drivers.
The SCSI bus allows any unit to talk to any other unit, or the host to talk
to any unit. Thus there must be some way of arbitration where units capture the bus. The main phases that the bus goes through are as follows:
(1) Free bus. In this state, there are no units that either transfer data
or have control of the bus. It is identified by deactivate SEL and
BSY (both will be high). Thus, any unit can capture the bus.
(2) Arbitration. In this state, a unit can take control of the bus and
become an initiator. To do this, it activates the BSY signal and
puts its own ID address on the data bus. After a delay, it tests the
data bus to determine whether a high-priority unit has put its own
address on the bus. If it has, then it will allow the other unit
access to the bus. If its address is still on the bus, then it asserts
the SEL line. After a delay, it then has control of the bus.
(3) Selection. In this state, the initiator selects a target unit and gets
the target to carry out a given function, such as reading or writing
data. The initiator outputs the OR value of its SCSI-ID and the
SCSI-ID of the target onto the data bus (e.g., if the initiator is 2
and the target is 5 then the OR-ed ID on the bus will be 00100100).
The target then determines that its ID is on the data bus and sets
the BSY line active. If this does not happen within a given time,
then the initiator deactivates the SEL signal, and the bus will be
free. The target determines that it is selected when the SEL signal
and its SCSI ID bit are active and the BSY and I/O signals are
false. It then asserts the BSY signal within a selection abort time.
(4) Reselection. When the arbitration phase is complete, the wining
SCSI device asserts the BSY and SEL signals and has delayed at
least a bus clear delay plus a bus settle delay. The wining SCSI
device sets the DATA BUS to a value that is the logical OR of its
SCSI ID bit and the initiator’s CSI ID bit. Sometimes, the target
takes some time to reply to the initiator’s request. The initiator
determines that it is reselected when the SEL and I/O signals and
its SCSI ID bit are true and the BSY signal is false. The reselected
initiator then asserts the BSY signal within a selection abort time
of its most recent detection of being reselected. An initiator does
not respond to a reselection phase if other than two SCSI ID bits
are on the data bus. After the target detects that the BSY signal is
true, it also asserts the BSY signal and waits a given time delay
and then releases the SEL signal. The target may then change the
I/O signal and the data bus. After the reselected initiator detects
the SEL signal is false, it releases the BSY signal. The target continues to assert the BSY signal until it gives up the SCSI bus.
Zhang_Ch03.indd 338
5/13/2008 5:41:31 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
339
(5) Command. The command phase is used by the target to request
command information from the initiator. The target asserts the
C/D signal and negates the I/O and MSG signals during the REQ/
ACK handshake(s) of this phase. The format of the command
descriptor block for 6-byte commands is: Byte 0—operation code;
Byte 1—logical unit number (MSB, if required); Byte 2—logic
block address; Byte 3—logic block address (LSB, if required);
Byte 4—transfer length (if required)/parameter list length (if
required)/allocation length (if required); Byte 5—control.
(6) Data. The data phase covers both the data-in and data-out phases.
In the data-in phase, the target requests that data be sent to the
initiator from the target. For this purpose, the target asserts the
I/O signal and negates the C/D and MSG signals during the REQ/
ACK handshake(s) of the phase. In the data-out phase, the target
requests that data be sent from the initiator to the target. The
target negates the C/D, I/O, and MSG signals during the REQ/
ACK handshake(s) of this phase.
(7) Message. The message phase covers both the message-out and
message-in phases. The first byte transferred in either of these
phases can be either a single-byte message or the first byte of a
multiple-byte message. Multiple-byte messages are contained
completely within a single message phase. The message system
allows the initiator and the target to communicate over the interface connection. Each message can be one, two, or more bytes in
length. In a single message phase, one or more messages can be
transmitted (but a message cannot be split between multiple message phases).
(8) Status. The status phase allows the target to request that status
information be sent from the target to the initiator. The target shall
assert the C/D and I/O signals and negate the MSG signal during
the REQ/ACK handshake(s) of this phase. The status phase normally occurs at the end of a command (although in some cases it
may occur before transferring the command descriptor block).
3.2.2.4
USB and Firewire
(1) Universal serial bus (USB). USB is mainly used for the connection of medium bandwidth peripherals such as keyboards, scanner, modem, video, game or graphic controller, music interface,
etc. The great advantage of USB is that it allows for peripherals
to be added and deleted from the system without causing any
system upsets. The system will also automatically sense the
connected device and load the required driver. Basically, USB
provides these features: (1) easy to use; (2) self-identifying
Zhang_Ch03.indd 339
5/13/2008 5:41:31 PM
340
INDUSTRIAL CONTROL TECHNOLOGY
peripherals with automatic mapping of function to driver and
configuration; (3) dynamically attachable and reconfigurable
peripherals; (4) low-speed and medium-speed transfer rate of
1.5 Mbps or 12 Mbps.
USB is a balanced bus architecture that hides the complexity
of the operation from the devices connected to the bus. The USB
host controller controls system bandwidth. Each device is
assigned a default address when the USB device is first powered
or reset. Hubs and functions are assigned a unique device address
by USB software. All USB devices are attached to the USB via a
port on specialized USB devices known as hubs. Hubs indicate
the attachment or removal of a USB device in its per port status.
Figure 3.26 shows an example connection of the USB 2.0 system. In this example, a memory hub is used to provide a fast data
transfer (GB/s), while the Firewire connection provides ultrahigh-speed connection for video transfers. The USB connection
provides low-high and full-speed connections to most of the
peripherals that connect to the system. The USB connections can
be internal or can connect to an external hub. In the implementation of USB, there are two main ways, as given below.
(a) Open host controller interface (OHCI). This method defines
the register level interface that enables the USB controller to
communicate with the host computer and the operating
system. OHCI is an industrial standard hardware interface for
operating systems, device drivers, and the BIOS to manage the
USB. It optimizes performance of the USB bus while minimizing CPU overhead to control the USB with a scatter/gather
Processor
AGP
graphic
controller
RDRAM
AGP 2.0
DRAM bus
Host
bus
Memory controller
hub (MCH)
Hub interface
Floppy disk
USB 2.0
hub
Hard disk
Video device
Firewire
USB Port 1
USB 2.0
controller
hub
USB Port 2
Audio/modem
USB 2.0
hub
Scanner
Adaptor
Figure 3.26 An example connection of the USB 2.0 system.
Zhang_Ch03.indd 340
5/13/2008 5:41:31 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
341
bushmaster hardware support. It has efficient isochronous
data transfers allowing for high USB bandwidth without
slowing down the host CPU. Furthermore, it ensures full
compatibility with all USB devices.
(b) Universal host controller interface (UHCI). This method
defines how the USB controller talks to the host computer
and its operating system. It is optimized to minimize host
computer design complexity and uses the host CPU to control the USB bus. This method has the advantages of simple
design which reduces the transistor count required to implement the USB interface on the host computer, thus reducing
the system cost. Furthermore, it can provide full compatibility with all USB devices.
In data transmission, USB supports two types of transfers:
stream and message. A stream has no defined structure,
whereas a message does. At start-up, one message pipe, control pipe 0, always exits as it provides access to the device’s
configuration, status, and control information. USB optimizes large data transfers and real-time data transfers. When
a pipe is established for an endpoint, most of the pipe’s transfer characteristics are determined and remain fixed for the
lifetime of the pipe. Each bus transaction of USB involves
the transmission of up to three packets which can be
(1) token packet transmission, (2) data packet transmission,
and (3) handshake packet transmission. With these transfer
characteristics, USB defines four transfer types:
(i) Control transfers. This is bursty, nonperiodic, hostsoftware-initiated request/response communication
typically used for command/status operations.
(ii) Isochronous transfers. This is periodic, continuous
communication between host and device typically used
for time-relevant information. This transfer type also
preserves the concept of time encapsulated in the data.
This does not imply, however, that the delivery needs of
such data are always time-critical.
(iii) Interrupt transfers. This is small data, nonperiodic, low
frequency, bounded latency, device-initiated communication typically used to notify the host of device service
needs.
(iv) Bulk transfers. Nonperiodic, large bursty communication
typically used for data that can use any available bandwidth and also is delayed until bandwidth is available.
As mentioned earlier, a major advantage of USB is the
hot attachment and detachment of devices. USB does
Zhang_Ch03.indd 341
5/13/2008 5:41:31 PM
342
INDUSTRIAL CONTROL TECHNOLOGY
this by sensing when a device is attached or detached.
When this happens, the host system is notified, and system software interrogates the device. It then determines
its capabilities, and automatically configures the devices.
All the required drivers are then loaded and applications
can immediately make use of the connected device.
(1) Attachment of USB devices. All USB devices are
addressed using the USB default address when initially connected or after they have been reset. The
host determines whether the newly attached USB
device is a hub or a function and assigns a unique
USB address to the USB device. The host establishes a control pipe for the USB device using
assigned USB address and endpoint number zero. If
the attached USB device is a hub and USB devices
are attached to its ports, then the above procedure is
followed for each of the attached USB devices. If
the attached USB device is a function, then attachment notifications will be dispatched by USB software to interested host software.
(2) Removal of USB devices. When a USB device has
been removed from one of its ports, the hub automatically disables the port and provides an indication of device removal to the host. Then the host
removes knowledge of the USB device. If the
removed USB device is a hub, then the removal
process must be performed for all of the USB
devices that were previously attached to the hub. If
the removed USB device is a function, removal
notifications are sent to interested host software.
(2) Firewire. The main competitor to USB is the Firewire standard
(IEEE 1394–1995 buses), which is a high-speed serial bus
typically for video transfers, whereas USB supports low-tomedium-speed peripherals. Firewire supports rates of approximately 100, 200, and 400 Mbps, known as S100, S200, and S400,
respectively. The future standard promises higher data rates, and
ultimately it is envisaged that rates of 3.2 Gbps will be achieved
when optical fiber is introduced into the system. It uses pointto-point interconnect with a tree topology: 1000 buses with
64 nodes gives 64,000 nodes. Firewire also can be automatic
configuration and hot plugging. In additional to asynchronous
transfer, Firewire is able to be isochronous data transfer, where a
fixed bandwidth is dedicated to a particular peripheral. However,
it has the limit of the maximum cable length as 4.5 m. This should
Zhang_Ch03.indd 342
5/13/2008 5:41:31 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
343
subsequently reduce the costs of production of controller interfaces and peripheral connectors, as well as simplifying the
requirements placed on users when setting up their devices.
Firewire is a more economical interface bus standard that performs fast and high-bandwidth data transmissions.
There are two bus categories in the Firewire:
(a) Cable. This is a bus that connects external devices via a
cable. This cable environment is a noncyclic network with
finite branches consisting of bus bridges and nodes (cable
devices). Noncyclic networks contain no loops and result in
a tree topology, with devices daisy-chained and branched
(where more than one device branch is connected to a device).
Devices on the bus are identified by node IDs. Configuration
of the node IDs is performed by the self ID and tree ID
processes after every bus reset. This happens every time a
device is added to or removed from the bus, and is invisible
to the user.
(b) Backplane. This type of topology is an internal bus. An internal IEEE-1394 device can be used alone, or incorporated
into another backplane bus. Implementation of the backplane
specification lags the development of the cable environment,
but one could image internal IEEE-1394 hard disks in one
computer being directly accessed by another IEEE-1394
connected computer.
One of the key capabilities of IEEE-1394 is isochronous
data transfer. Both asynchronous and isochronous are supported, and are useful for different applications. Isochronous
transmission transmits data like real-time speech and video,
both of which must be delivered uninterrupted, and at the
rate expected, whereas asynchronous transmission is used to
transfer data that is not tied to a specific transfer time. With
IEEE-1394, asynchronous transmission is the conventional
transfer method of sending data to an explicit address, and
receiving confirmation when it is received. Isochronous,
however, is an unacknowledged guaranteed bandwidth transmission method, useful for just-in-time delivery of multimedia type data.
An isochronous “talker” requests an amount of bandwidth
and a channel number. Once the bandwidth has been allocated, it can transmit data preceded by a channel ID. The
isochronous listeners can then listen for the specified
channel ID and accept the data following. If the data is not
intended for a node, it will not be set to listen on the specific
channel ID. Up to 64 isochronous channels are available,
Zhang_Ch03.indd 343
5/13/2008 5:41:31 PM
344
INDUSTRIAL CONTROL TECHNOLOGY
and these must be allocated, along with their respective bandwidths, by an isochronous resource manager on the bus.
By comparison, asynchronous transfers are sent to explicit
addresses on the 1394 bus. When data is to be sent, it is preceded by a destination address, which each node checks to
identify packets for itself. If a node finds a packet addressed
to itself, it copies it into its receive buffer. Each node is identified by a 16-bit ID, containing the 10-bit bus ID and 6-bit
node or physical ID. The actual packet addressing, however,
is 64 bits wide, providing a further 48 bits for addressing a
specific offset within a node’s memory.
3.2.2.5 AGP and Parallel Ports
(1) AGP. The accelerated graphic port (AGP) is a major advancement in the connection of three-dimensional graphics applications, and is based on an enhancement of the PCI bus. One of the
major improvements is the speed of transfer between the main
system memory and the local graphic card. This reduces the need
for large areas of memory on the graphics card.
The main gain in moving graphics memory from the display
buffer (on the graphic card) to the main memory is the display of
text information because (1) it is generally read-only, and does
not have to be displayed in any special order. (2) Shifting text
does not require a great deal of data transfer and can be easily
cached in memory, thus reducing data transfer. A shift in text can
be loaded from the cached memory. (3) It is dependent on the
graphics quality of the application, rather than the resolution of
the display. (4) It is not persistent, as it resides in memory only
for the duration that it is required. When it has completed the
main memory, it can be assigned to another application. A display buffer, on the other hand, is permanent.
The 440LX was the first AGP set product designed to support
the AGP interface. The HOST BRIDGE AGP implementation is
compatible with the accelerated graphics port specification 1.0.
HOST BRIDGE supports only a synchronous AGP interface,
coupling with the host bus frequency. The AGP 1.0 interface can
reach a theoretical 528 MB/s transfer rate and AGP 2.0 can
achieve a theoretical 1.056 GB/s transfer rate. The actual bandwidth will be limited by the capability of the HOST BRIDGE
memory subsystem.
(2) Parallel port. The parallel port is hardly the greatest technology.
In its standard form, it allows only for simple communications
Zhang_Ch03.indd 344
5/13/2008 5:41:31 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
345
from the PC outwards. However, like the RS-232, the parallel
port is a standard port of the PC, and it is cheaper.
All parallel ports use a bidirectional link in either a compatible, nibble, or byte mode. These modes are relatively slow as the
software must monitor the handshaking lines (up to 100 kbps).
To allow high speeds, the enhanced parallel port and extended
capabilities port protocol modes allow high-speed data transfer
using automatic hardware handshaking.
3.2.2.6
RS-232, RS-422, RS-485, and RS-530
(1) RS-232. RS-232 (Recommended Standard-232) is a TIA/EIA
standard for serial transmission between DTE and DCE. Using a
25-pin DB-25 or 9-pin DB-9 connector, its normal cable limitation of 50 ft can be extended to several hundred feet with highquality cable.
RS-232 defines the purpose and signal timing for each of the
25 lines; however, many applications use less than a dozen.
RS-232 transmits positive voltage for a 0 bit, negative voltage for
a 1. In 1984, this interface was officially renamed TIA/EIA232-E standard (E is the current revision, 1991), although most
people still call it RS-232. Table 3.7 lists the some RE-232
specifications.
(2) RS-422 and RS-485. RS-422 (Recommended Standard-422) is a
balanced serial interface for the transmission of digital data. The
advantage of a balanced signal is the greater immunity to noise.
The EIA describes RS-422 as a DTE to DCE interface for pointto-point connections.
RS-422 was designed for greater distances and higher baud
rates than RS-232. In its simplest form, a pair of converters from
RS-232 to RS-422 (and back again) can be used to form an
“RS-232 extension cord.” Data rates of up to 100 kbps and distances up to 4000 ft can be accommodated with RS-422. RS-422
is also specified for multidrop (party-line) applications where
only one driver is connected to, and transmits on, a “bus” of up
to 10 receivers.
RS-485 (Recommended Standard-485) is standard for sending serial data. It uses a pair of wires to send a differential signal
over distances up to 4000 ft without a repeater. The differential
signal makes it very robust; RS-485 is one of the most popular
communications methods used in industrial applications where
its noise immunity and long-distance capability are a perfect fit.
RS-485 is capable of multidrop communications—up to 32 nodes.
Zhang_Ch03.indd 345
5/13/2008 5:41:32 PM
346
INDUSTRIAL CONTROL TECHNOLOGY
Table 3.7 RS-232 and RS-422 Specifications
Specifications
Mode of operation
Total number of drivers and receivers on
one line
Maximum cable length (ft)
Maximum data rate
Maximum driver output voltage (V)
Driver output signal level
Loaded
(loaded minimum) (V)
Driver output signal level
Unloaded
(unloaded maximum) (V)
Driver load impedance (Ω)
Maximum driver current in
Power on
high Z state
Maximum driver current in
Power off
high Z state
Slew rate (maximum)
Receiver input voltage range (V)
Receiver input sensitivity
Receiver input resistance (Ohms)
RS-232
RS-422
Single-ended
1 Driver
1 Receiver
50
20 kbits/s
+/−25
+/−2.0
Differential
1 Driver
10 Receiver
4000
10 Mbits/s
−0.25 to +6
+/−5 to +/−15
+/–6
+/−25
3–7 k
N/A
100
N/A
+/−100 µA
+/−6 mA @
+/−2 V
N/A
−10 to +10
+/−200 mV
4k min
30 V/µs
+/−15
+/−3 V
3–7 k
RS-485 can be configured for their half or full duplex. Half
duplex typically uses one pair of wires; full duplex requires two
pairs.
Both RS-422 and RS-485 use a twisted-pair wire (i.e., 2 wires)
for each signal. They both use the same differential drive with
identical voltage swings: 0 to +5 V. The main difference between
RS-422 and RS-485 is that while RS-422 is strictly for point-topoint communications (and the driver is always enabled), RS-485
can be used for multiple drop systems. Since the basic differential receivers of RS-423-A and RS-422-A are electrically identical, it is possible to interconnect equipment using RS-423-A
receivers and generators on one side of the interface with equipment using RS-422-A generators and receivers on the other side
of the interface, if the leads of the receivers and generators are
properly configured to accommodate such an arrangement and
the cable is not terminated.
Table 3.7 lists some specifications for RS-422. The data is
coded as a differential voltage between the wires. The wires are
named A (negative) and B (positive). When B > A then the output
Zhang_Ch03.indd 346
5/13/2008 5:41:32 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
347
is a mark (1 or off) and when A > B then it is counted as a space
(0 or on). In general, a mark is +1 VDC for the A line and +4 VDC
for the B line. A space is +1 VDC for the B line and +4 VDC for
the A line. At the transmitter end the voltage difference should
not be less than 1.5 VDC and not exceed 5 VDC. At the receiver
end the voltage difference should not be less than 0.2 VDC. The
minimum voltage level is –7 VDC and maximum +12 VDC.
(3) RS-530. RS-530 (Recommended Standard-530) employs differential signaling on its send, receive, and clocking signals, as well
as its control and handshaking signals. The differential signals
for RS-530 are labeled as either “A or B.” At both connectors,
wire A always connects to A, and B connects to B.
The RS-530 transmitter sends a data 0 (or logic ON) by setting
the potential on the A signal to 0.3 V (or more) higher than the
voltage on the B signal. The transmitter sends a data 1 (or logic
OFF) by setting the potential on the B signal to 0.3 V or more
than the voltage on the A signal. The voltage offset (from ground
reference) is not to exceed 3 V, however, most receivers can handle much more; check the receiver data sheet for exact limits.
This approach is relatively immune to noise when the cable is
constructed so that the A and B signal wires are a twisted pair.
Shielding the cable is generally not required.
Data 0 = A > B + 0.3 V; Data 1 = B > A + 0.3 V
Example: Data 0: A = 2 V, B = 1 V; Data 1: A = 1 V, B = 2 V.
Most receivers can handle both + and – voltages; again check
the data sheet on the part used to be sure. If you have the correct
receivers it is possible for the older V.35 (+/–5 V) signaling to be
wired to RS-530 or V.11. This is how Cisco and others get many
different interfaces on their Smart Serial connectors, and you
thought it was magic!
3.2.2.7
IEEE-488
Hewlett-Packard originally developed the IEEE-488 bus called the
HP-IB to connect and control programmable instruments, and to provide a
standard interface for communication between instruments from different
sources. The interface quickly gained popularity in the computer industry.
Because the interface was so versatile, the IEEE committee renamed it
General Purpose Interface Bus (GPIB).
Almost any instrument can be used with the IEEE-488 specification,
because it says nothing about the function of the instrument itself, or about
the form of the instrument’s data. Instead, the specification defines a
separate component, the interface, which can be added to the instrument.
Zhang_Ch03.indd 347
5/13/2008 5:41:32 PM
348
INDUSTRIAL CONTROL TECHNOLOGY
The signals passing into the interface from the IEEE-488 bus and from the
instrument are defined in the standard. The instrument does not have complete control over the interface. Often the bus controller tells the interface
what to do. The active controller performs the bus control functions for all
the bus instruments.
(1) IEEE-488 standards. The IEEE-488.1 standard greatly simplified
the interconnection of programmable instruments by clearly
defining mechanical, hardware, and electrical protocol specifications. For the first time, instruments from different manufacturers
were connected by a standard cable. This standard does not address
data formats, status reporting, message exchange protocol, common configuration commands, or device specific commands.
The IEEE-488.2 standard enhances and strengthens the IEEE488.1 standard by specifying data formats, status reporting, error
handling, controller functionality, and common instrument commands. It focuses mainly on the software protocol issues and
thus maintains compatibility with the hardware-oriented IEEE488.1 standard. IEEE-488.2 systems tend to be more compatible
and reliable.
The IEEE-488 standard allows up to 15 devices to be interconnected on one bus. Each device is assigned a unique primary
address, ranging from 0 to 30, by setting the address switches
on the device. A secondary address may also be specified, ranging from 0 to 30. See the device documentation for more information on how to set the device primary and optional secondary
address.
The IEEE-488 bus specifies a maximum total cable length of
20 m with no more than 20 devices connected to the bus and at
least two-thirds of the devices powered on. A maximum separation of 4 m between devices and an average separation of 2 m
over the full bus should be followed. Bus extenders and expanders are available to overcome these system limits. The standard
IEEE-488 cable has both a plug and receptacle connector on both
ends. Special adapters and nonstandard cables are available for
special interconnect applications.
(2) Interface signals and data lines. At power-up time, the IEEE488 interface that is programmed to be the System Controller
becomes the Active Controller in charge. The System Controller
has several unique capabilities including the ability to send
Interface Clear (IFC) and Remote Enable (REN) commands.
IFC clears all device interfaces and returns control to the System
Controller. REN allows devices to respond to bus data once they
are addressed to listen. The System Controller may optionally
Zhang_Ch03.indd 348
5/13/2008 5:41:32 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
349
pass control to another controller, which then becomes Active
Controller.
There are three types of devices that can be connected to the
IEEE-488 bus (listeners, talkers, and controllers). Some devices
include more than one of these functions. The standard allows
a maximum of 15 devices to be connected on the same bus. A
minimum system consists of one controller and one talker or listener device (i.e., an HP 700 with an IEEE-488 interface and a
voltmeter).
It is possible to have several controllers on the bus but only
one may be active at any given time. The Active Controller may
pass control to another controller which in turn can pass it back
or on to another controller. A listener is a device that can receive
data from the bus when instructed by the controller and a talker
transmits data on to the bus when instructed. The controller can
set up a talker and a group of listeners so that it is possible to
send data between groups of devices as well.
The IEEE-488 interface system consists of 16 signal lines and
8 ground lines. The 16 signal lines are divided into 3 groups (8 data
lines, 3 handshake lines, and 5 interface management lines).
The lines DIO1 through DIO8 are used to transfer addresses,
control information, and data. The formats for addresses and
control bytes are defined by the IEEE 488 standard. Data formats
are undefined and may be ASCII (with or without parity) or
binary. DIO1 is the Least Significant Bit (note that this will correspond to bit 0 on most computers).
(3) Handshake lines and handshaking. The three handshake lines
(NRFD, NDAC, DAV) control the transfer of message bytes
among the devices and form the method for acknowledging the
transfer of data. This handshaking process guarantees that the
bytes on the data lines are sent and received without any
transmission errors and is one of the unique features of the IEEE488 bus.
(a) The NRFD (Not Ready for Data) handshake line is identified
by a listener to indicate it is not yet ready for the next data or
control byte. Note that the controller will not see NRFD
released (i.e., ready for data) until all devices have released it.
(b) The NDAC (Not Data Accepted) handshake line is identified
by a listener to indicate it has not yet accepted the data or
control byte on the data lines. Note that the controller will
not see NDAC released (i.e., data accepted) until all devices
have released it.
(c) The DAV (Data Valid) handshake line is identified by the
talker to indicate that a data or control byte has been placed on
Zhang_Ch03.indd 349
5/13/2008 5:41:32 PM
350
INDUSTRIAL CONTROL TECHNOLOGY
the data lines and has had the minimum specified stabilizing
time. The byte can now be safely accepted by the devices.
The handshaking process is outlined as follows. When the
controller or a talker wishes to transmit data on the bus, it
sets the DAV line high (data not valid), and checks to see that
the NRFD and NDAC lines are both low, and then it puts the
data on the data lines. When all the devices that can receive
the data are ready, each releases its NRFD (not ready for
data) line. When the last receiver releases NRFD, and it goes
high, the controller or talker takes DAV low indicating that
valid data is now on the bus. In response each receiver takes
NRFD low again to indicate it is busy and releases NDAC
(not data accepted) when it has received the data. When the
last receiver has accepted the data, NDAC will go high and
the controller or talker can set DAV high again to transmit the
next byte of data. Note that if after setting the DAV line high,
the controller or talker senses that both NRFD and NDAC
are high, an error will occur. Also if any device fails to perform its part of the handshake and releases either NDAC or
NRFD, data cannot be transmitted over the bus. Eventually a
timeout error will be generated. The speed of the data transfer is controlled by the response of the slowest device on the
bus; for this reason it is difficult to estimate data transfer rates
on the IEEE-488 bus as they are always device dependent.
(4) Interface management lines. The five interface management
lines (ATN, EOI, IFC, REN, and SRQ) manage the flow of control and data bytes across the interface.
(a) The ATN (Attention) signal is chosen by the controller to
indicate that it is placing an address or control byte on the
data bus. ATN is released to allow the assigned talker to
place status or data on the data bus. The controller regains
control by reasserting ATN; this is normally done synchronously with the handshake to avoid confusion between control and data bytes.
(b) The EOI (End or Identify) signal has two uses. A talker may
assert EOI simultaneously with the last byte of data to indicate end-of-data. The controller may assert EOI along with
ATN to initiate a parallel poll. Although many devices do not
use parallel poll, all devices should use EOI to end transfers
(many currently available ones do not).
(c) The IFC (Interface Clear) signal is selected only by the
System Controller in order to initialize all device interfaces
to a known state. After releasing IFC, the System Controller
is the Active Controller.
Zhang_Ch03.indd 350
5/13/2008 5:41:32 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
351
(d) The REN (Remote Enable) signal is selection only by the
System Controller. Its selection does not place devices into
remote control mode; REN only enables a device to go into
remote mode when addressed to listen. When in remote
mode, a device should ignore its local front panel controls.
(e) The SRQ (Service Request) line is like an interrupt: it may
be used by any device to request the controller to take some
action. The controller must determine which device is calling for the SRQ by conducting a serial poll. The requesting
device releases SRQ when it is polled.
3.3 Human–Machine Interface
in Industrial Control
3.3.1
Overview
The term “user interface” refers to the methods and devices that are
used to accommodate interaction between machines and the human beings
who use them. The user interface of a mechanical system, a vehicle, or an
industrial installation and so on is often referred to as the human–machine
interface. In any industrial control system, the human–machine interface
can be used to deliver information from machine to people, which allows
people to control, monitor, and record the system through devices such as
image, keyboard, Ethernet, screen, video, radio, software, etc. Actually, the
human–machine interface can take many forms. Although there are many
techniques and methods used in industry, the human–machine interface
always accomplishes two fundamental tasks: communicating information
from the machine to the user, and communicating information from the
user to the machine.
Two industrial applications of the human–machine interface are given
below to demonstrate its importance in industry and industrial control.
However, in view of the fact that the human–machine interface becomes
more and more essential in industry, these two examples are far from representing its applications in industry.
A robot control is a good example that requires working with the human–
machine interface. This robot control is based on human speech and gestures instead of keyboard or joystick control. In turn, the robot can also
respond to the human control orders by using speech and gestures. When
both the operator and the robot understand the environment and the work
objects, they can communicate easier and the work task can be completed
collaboratively. In most robots, the human–machine interface is a key
Zhang_Ch03.indd 351
5/13/2008 5:41:32 PM
352
INDUSTRIAL CONTROL TECHNOLOGY
component in any work cell. The human–machine interface must allow
operators to run the equipment and cell in an intuitive manner. It must be
configurable to give each level of personnel appropriate access to various
layers of functionality. An autonomous service robot operates in the user’s
own environment, performing independent tasks to reach user goals.
Applications include, for instance, delivery agents in hospitals and factories, and cleaning robots in the home or in supermarkets. The latest robot
controllers now are beginning to offer built-in human–machine interface
functionality, complete with touch-screen interfaces, status indicators,
program selection switches, part counters, and various other functions,
which can be seen in Fig. 3.27.
Another example in this aspect is the SCADA system in which the
human–machine interface is an essential component. In industry, the
human–machine interface in SCADA was born out of a need for a userfriendly front-end to a control system containing PLC. While a PLC does
provide automated, preprogrammed control over a process, they are usually distributed across a plant, making it difficult to gather data from them
manually. Additionally, the PLC information is usually in a crude userunfriendly format. The SCADA of human–machine interface gathers
information from the PLCs via some form of network, and combines and
formats the information. A sophisticated human–machine interface may
also be linked to a database, to provide instant trending, diagnostic data,
Figure 3.27 A touch-screen shot of the human–machine interface for a robot
(courtesy of Siemens).
Zhang_Ch03.indd 352
5/13/2008 5:41:32 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
353
scheduled maintenance procedures, logistic information, detailed schematics for a particular sensor or machine, and expert-system troubleshooting
guides. Since about 1998, many companies, especially all major PLC
manufacturers, have offered integrated SCADA and human–machine
interface systems for its comprehensive range of facilities for industrial
automation and process control. These companies recognized the benefits
of using the same reliable monitoring and control software throughout
their business, from shop floor through to top management. Many of them
used open and nonproprietary communications protocols. Numerous specialized third-party SCADA packages with the human–machine interface,
offering built-in compatibility with most major PLCs, have also entered
the market, allowing mechanical engineers, electrical engineers, and technicians to configure the human–machine interface by themselves, without
the need for a custom-made program written by a software developer.
Figure 3.28 is a diagram including an SCADA system and its control
screen in Website.
3.3.2
Human–Machine Interactions
Automated systems have penetrated virtually every area of our private
life and our work environments. Development work on technical products
is accelerating, and the products themselves are becoming increasingly
complex and powerful. Human–machine interaction is already playing a
vital role along the entire production process, from planning individual
links in the production chain right through to designing the finished product.
Innovative technology is made for humans, used and monitored by humans.
The optimum form for this interaction depends on whether a technical
innovation is reliable in operation, is safe, is accepted by personnel, and,
last but not least, is cost-effective. This interplay between technology and
user, known as human–machine interaction, is hence at the very heart of
industrial automation, automated control, and industrial production.
3.3.2.1 The Models for Human–Machine Interactions
Modeling the human–machine interaction is done simply to depict how
human and machine interact in a system. The human–machine interaction
model illustrates a typical information flow (or process context) between
the “human” and “machine” components of a system. Figure 3.29 provides
the components involved in each side of the human–machine interaction.
The environment side has three components: Machine display component,
Zhang_Ch03.indd 353
5/13/2008 5:41:33 PM
354
INDUSTRIAL CONTROL TECHNOLOGY
Figure 3.28 A SCADA system and its Website-control screen (Courtesy of
Siemens).
Human musculoskeletal
component
Human cognitive component
Human sensory component
Machine I/O device component
Machine CPU component
Machine display component
Environment
Interaction
Human
Figure 3.29 The components involved in the human–machine interaction.
Zhang_Ch03.indd 354
5/13/2008 5:41:33 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
355
Machine CPU component, and Machine I/O device component. The human
side has another three components: Human sensory component, Human
cognitive component, and Human musculoskeletal component.
In modern control systems, a model is a common architecture for grouping several machine configurations under one label. The set of models
in a control system corresponds to a set of unique machine behaviors. The
operator interacts with the machine by switching among models manually,
or monitoring the automatic switching triggered by the machine. However,
our understanding of models and their potential contribution to confusion
and error is still far from complete. For example, there is widespread disagreement among user interface designers and researchers about what
models are, independent of how they affect users. This blurred vision,
found not only in the human–machine interaction domain, impedes
our ability to develop methods for representing and evaluating human
interaction with control systems. This limitation is magnified in high-risk
systems such as automated cockpits, for which there is an urgent need
to develop methods that will allow designers to identify, early in the
design phase, the potential for error. The errors arising from modeling the
human–machine interaction are thus an important issue that cannot be
ignored.
(1) Definition and construct. The constructs of the human–machine
interaction model that will be discussed below are measurable
aspects of human interaction with machines. As such, they can
be used to form the foundation of a systematic and quantitative
analysis.
(a) Models’ behaviors. One of the first treatments of models
came in the early 1960s from the science of cybernetics, the
comparative study of human control systems and complex
machines. The first treatments set forth the construct of a
machine with different behaviors. The following is a simplified description of the machine behaviors’ construct: A given
machine may have several components (e.g., X1, X2, X3).
For each component there is a finite set of states. On “Startup,”
for example, the machine initializes itself so that the active
state of component X1 is “a,” X2 is “f,” and X3 is “k” (see
Table 3.8). The vector of states (a, f, k) thus defines the
machine’s configuration on “Startup.” Once a built-in test is
performed, the machine can move from “Startup” to “Ready.”
The transition to “Ready” can be defined so that for component X1, state “a” undergoes a transition and becomes state
“b”; for X2 state “f” transitions to “g”, and for X3, “k”
Zhang_Ch03.indd 355
5/13/2008 5:41:34 PM
356
INDUSTRIAL CONTROL TECHNOLOGY
changes to “l”. The new configuration (b, g, l) defines
the “Ready” model of the machine. Now, there might be a
third set of transitions, for example, to “Engaged” (c, h, m),
and so on.
The set of configurations labeled “Startup,” “Ready” and
“Engaged,” if embedded in the same physical unit, corresponds to a machine with three different ways of behaving.
A real machine whose behavior can be so represented is
defined as a machine with “input.” The input triggers the
machine to change its behavior. One source of input to the
machine is manual; the user selects the model, for example,
by turning a switch, and the corresponding transitions take
place. But there can be another type of input: If some other
machine selects the model, the input is “automatic.” More
precisely, the output of the other machine becomes the input
to our machine. For example, a separate machine performs
a built-in test and outputs a signal that causes the machine
in Table 3.8 to transition from “Standby” to “Ready,”
automatically.
Here in this book, we first define a model as a machine
configuration that corresponds to a unique behavior. This is
a very broad definition of the term. Later on, we constrain
this definition from our special perspective: user interaction
with control systems that employ models.
(b) Model’s error and ambiguity. Model errors fall into a class
of errors that involve forming and carrying out an intention.
That is, when a situation is falsely classified, the resulting
action may be one that was intended and appropriate for a
perceived or expected situation, but inappropriate for the
actual situation. However, there remains an open question as
to what kind of situations lead to model error. This issue can
be addressed by an example of a word processor. Specifically,
we looked at situations in which the user’s input has different interpretations, depending on model. For example, in one
Table 3.8 A Machine with Different Behaviors
Startup
Ready
Engaged
Zhang_Ch03.indd 356
X1
X2
X3
a
b
c
f
g
h
k
l
m
5/13/2008 5:41:34 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
357
word processing application, the keystroke d can be interpreted as either (1) the literal text “d,” or (2) as the command
“delete.” The interpretation depends on the word processor’s
active model: either Text or Command.
Model error can be linked to model ambiguity with interjecting the notion of user expectations. In this view, “model
ambiguity” will result in model error only when the user has
a false expectation about the result of his or her actions.
There are two types of ambiguity: one that leads to model
error, and one that does not lead to model error. An example
of model ambiguity that does lead to model error is timesharing operating systems in which keystrokes are buffered
until a “Return” or “Enter” key is pressed. However, when
the buffer gets full, all subsequent keystrokes are ignored
accordingly. This feature leads to two possible outcomes; all
or only a portion of the keystrokes will be processed. The
two outcomes depend on the state of buffer which is either
“not full” or “full.” Since the state of the buffer is unknown
to the user, false expectations may occur. The user’s action:
hitting the “Return” key and seeing only part of what was
keyed on the screen is therefore a “model error” because the
buffer has already filled up.
An example of model ambiguity that does not lead to
model error is a common end-of-line algorithm which determines the number of words in a line. An ambiguity is introduced because the criteria for including the last word on the
current line, or wrapping to the next line, are known to the
user. Nevertheless, as long as the algorithm works reasonably well, the user will not complain because he or she has
not formed any expectation about which word will stay on the
line or scroll down, and either outcome is usually acceptable.
Therefore, model error will not take place, even though
model ambiguity does indeed exist.
(c) User factors: Task, knowledge, and ability. One important
element that constrains user expectations is the task at hand.
If discerning between two or more different machine configurations is not part of the user’s task, model error will
not occur. Consider, for example, the radiator fan of your car.
Do you know what configuration (OFF, ON) it is in? The
answer, of course, is no. There is no such indication in most
modern cars.
The fan mechanism changes its mode automatically depending on the output of the water temperature sensor. Model
Zhang_Ch03.indd 357
5/13/2008 5:41:34 PM
358
INDUSTRIAL CONTROL TECHNOLOGY
ambiguity exists because at any point in time, the fan mechanism can change its model or stay in the current model. The
configuration of the fan is completely ambiguous to the
driver. But does such model ambiguity lead to model error?
The answer is obvious; not at all because monitoring the fan
configuration is not part of the driver’s task. Therefore, the
user’s task is an important determinant of which machine
configurations must be tracked by the user and which machine
configurations need not be tracked.
The second element that is part of the assessment of user
expectations is user knowledge about the machine’s behaviors. By this, we mean that the user constructs some mental
model of the machine’s “response map.” This mental model
allows the user to track the machine’s configuration, and
most importantly, to anticipate the next configuration of the
machine. Specifically, our user must be able to predict what
the new configuration will be following a manually triggered
event or an automatically triggered event. The problem of
reliably anticipating the next configuration of the machine
becomes difficult when the number of transitions between
configurations is large. Another factor in user knowledge is
the number of conditions that must be evaluated as TRUE,
before a transition from one model to another takes place.
For example, the automated flight control systems of modern
aircraft can execute a fully automatic (hands-off) landing.
Several conditions (two engaged autopilots, two navigation
receivers tuned to the correct frequency, identical course set,
and more) must be TRUE before the aircraft will execute
automatic landing. Therefore, in order to reliably anticipate
the next model configuration of the machine, the user must
have a complete and accurate model of the machine’s behavior, including its configurations, transitions, and associated
conditions. This model, however, does not have to be complete in the sense that it describes every configuration and
transition of the machine. Instead, as discussed earlier, the
details of the user’s model must be merely sufficient for the
user’s task, which is a much weaker requirement.
The third element in the assessment of user expectations
is the user ability to sense the conditions that trigger a transition. Specifically, the user must be able to first sense the
events (e.g., a flight director is engaged; aircraft is more than
400 ft above the ground) and then evaluate whether or not
the transition to a model (say, a vertical navigation) will take
place. These events are usually made known to the user
Zhang_Ch03.indd 358
5/13/2008 5:41:34 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
359
through an interface. Nevertheless, there are more than a few
control systems in which the interface does not depict the
necessary input events. Such interfaces are said to be incorrect. In large and complex control systems, the user may
have to integrate information from several displays in order
to evaluate whether the transition will take place or not. For
example, one of the conditions for a fully automated landing
in a two-engine jetliner is that two separate electrical sources
must be online, each one supplying its respective autopilot
among the two autopilots. This information is external to the
automatic flight control system, in the sense that it involves
another aircraft system. The user’s job of integrating events,
some of which are located in different displays, is not trivial.
One important requirement for an efficient design is for the
interface to integrate these events and provide the user with
a succinct cue.
In summary, we have discussed three elements that help to
determine whether a given model ambiguity will or will not
lead to false expectations. First is the relationship between
model ambiguity and the user’s task. If distinguishing between
models (e.g., radiator fan is “ON” or “OFF”) is not part of
the user’s task, no meaningful errors will occur. Second, in a
case where model ambiguity is relevant to the user’s task, we
assess the user’s knowledge. If the user has an inaccurate and
or incomplete model of the machine’s response map, he or
she will not be able to anticipate the next configuration and
model confusion will occur. Third, we evaluate the user’s
ability to sense input events that trigger transitions. The interface must provide the user with all the necessary input events.
If it does not, no accurate and complete model will help; the
user may know what to look for but will never find it. As a
result, confusion and model error will occur.
(2) Classifications and types. A classification of the human–machine
interaction models is proposed here to encompass three types of
models in automated control systems: (1) “Interface models”
that specify the behavior of the interface, (2) “Functional models” that specify the behavior of the various functions of a
machine, and (3) “Supervisory models” that specify the level of
user and machine involvement in supervising the process. Before
we proceed to discuss this classification, we shall briefly describe
a modeling language, “State Charts,” that will allow us to represent these models.
In Chapter 5 of this book, Finite State Machine Model is given
as a natural medium for describing the behavior of a model-based
Zhang_Ch03.indd 359
5/13/2008 5:41:34 PM
360
INDUSTRIAL CONTROL TECHNOLOGY
system. A basic fragment of such a description is a state transition
which captures the states, conditions or events, and transitions in
a system. The State Chart language is a visual formalism for
describing states and transitions in a modular fashion by extending the traditional Finite State Machine to include three unique
features: “hierarchy,” “concurrency,” and “broadcast.” The “hierarchy” is represented by substates encapsulated within a super
state. The “concurrency” is shown by means of two or more independent processes working in parallel. The “broadcast” mechanism allows for coupling of components, in the sense that an
event in one end of the network can trigger transitions in another.
These features of the State Chart are further explained in the following three examples.
(a) Interface models. Figure 3.30 is a modeling structure of an
interface model. It has three concurrently active processes
(separated by a broken line): speed knob behavior, speed
knob indicator, and speed window display. The behavior
of the speed knob (middle process) is either “normal” or
“pushed-in.” (These two states are depicted, in the State
Chart language, by two rounded rectangles.) The initial state
of the speed knob is normal (indicated by the small arrow
above the state), but when momentarily pushed, the speed
knob engages or disengages the Speed Intervene submode of
the vertical navigation (VNAV, hereafter) model. The transition between normal and pushed-in is depicted by a solid
arrow and the label “push” describes the triggering event.
The transition back to normal occurs immediately as the
pilot lifts his or her finger (the knob is spring loaded).
Speed knob
(behavior)
Speed knob
(indicator)
Normal
Speed window
(display)
Closed
d1 [disengaging VNAV
.or. (in VNAV and b)]
Blank
Push/b
d2 [engaging VNAV
.or. (in VNAV and b)]
Pushed-in
Open
Figure 3.30 An interface model.
Zhang_Ch03.indd 360
5/13/2008 5:41:34 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
361
The left-most process shown in Fig. 3.30 is the speed knob
indicator. In contrast to many such knobs that have indicators, the Boeing-757 speed knob itself has no indicator and
therefore is depicted as a single (blank) state. The right-most
process is the speed window display that can be either closed
or open. After VNAV is engaged, the speed window display
is closed (implying that the source of the speed is from
another component, the flight management computer). After
VNAV is disengaged, and a semiautomatic model such as
vertical speed is active, the speed window display is open,
and the pilot can observe the current speed value and enter a
new speed value. This logic is depicted in the speed knob
indicator process in Fig. 3.30: transition d1 from closed to
open is conditioned by the event “disengaging VNAV,” and
d2 will take place when the pilot is “engaging VNAV.”
When in vertical navigation model, the pilot can engage
the Speed Intervene submodel by pushing the speed knob.
This event, “push” (which can be seen in speed knob behavior process), triggers event b, which is then broadcast to
other processes. Being in VNAV and sensing event b (“in
VNAV and b”) is another (.OR.) condition on transition d1
from closed to open. Likewise, it is also the condition on
transition d2 that takes us back to closed. To this end, the
behavior of the speed knob is circular; the pilot can push the
knob to close and push it again to open, ad infinitum.
As explained above and seen in Fig. 3.30, there are two
sets of conditions on the transitions between close and open.
Of all these conditions, one, namely “disengaging VNAV,” is
not always directly within the pilot’s control; it sometimes
takes place automatically (e.g., during a transition from
VNAV to the altitude hold model). Manual reengagement of
VNAV will cause the speed parameter in the speed window
to be replaced by economy speed computed by the flight
management computer. If the speed value in the speed
window was a restriction required by the American Transport
Council, the aircraft will now accelerate/decelerate to the
computed speed and the American Transport Council speed
restriction will be ignored!
(b) Functional models. When we survey the use of models in
devices, an additional type of model emerges: the “functional
model” which refers to the active function of the machine
that produces a distinct behavior. An automatic gearshift
mechanism of a car is one example of a machine with different models, each one defining different behaviors.
Zhang_Ch03.indd 361
5/13/2008 5:41:35 PM
362
INDUSTRIAL CONTROL TECHNOLOGY
As we move to discussion of functional models and their
uses in machines that control a timed process, we encounter
the concept of “dynamics.” In dynamic control systems, the
configuration and resulting behavior of the machine are a
combination of a model and its associated parameter (e.g.,
speed, time, etc.). Referring back to our car example, the
active model is the engaged gear that is Drive, and the associated parameter is the speed that corresponds to the angle of
the accelerator pedal (say, 65 miles/h). Both model (Drive)
and parameter (65 miles/h) define the configuration of the
mechanism.
Figure 3.31 depicts the structure of a functional model in
the dynamic automated control system of a modern airliner.
Two concurrent processes are depicted in this modeling
structure: (1) models, and (2) parameter sources.
Three models are depicted in the vertical models superstate in Fig. 3.31: vertical navigation, altitude hold, and vertical speed (the default model). All are functional models
related to the vertical aspect of flight. The speed parameter
can be obtained from two different sources: the flight management computer or the model control panel. The default
source of the speed parameter, indicated by the small arrow
Vertical autopilot
(models)
Speed
(parameter)
Vertical
navigation
m2 / rv1
Altitude
hold
m1
Flight
management
computer
rv1
Model control
panel
Vertical
speed
Figure 3.31 A functional model.
Zhang_Ch03.indd 362
5/13/2008 5:41:35 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
363
in Fig. 3.31, is the model control panel. As mentioned in the
discussion on interface models, engagement of vertical navigation via the model control panel will cause a transition to
the flight management computer as the source of speed. This
can be seen in Fig. 3.31 where transition m2 will trigger
event rv1, which, in turn, triggers an automatic transition
(depicted as a broken line) from “model control panel” to
“flight management computer.” In many dynamic control
mechanisms, some model transitions trigger a parameter
source change while others do not. Such independence appears
to be a source of confusion to operators.
(c) Supervisory models. The third type of model we discuss here
is “supervisory models” that sometimes are also referred to
as “participatory” or “control” models. Modern automated
control mechanisms usually allow the user flexibility in
specifying the level of human and machine involvement in
controlling the process. That is, the operator may decide to
engage a manual model in which he or she is controlling the
process; a semiautomatic model in which the operator specifies target values, in real time, and the machine attempts to
maintain them; or fully automatic models in which the operator specifies in advance a sequence of target values, that is
parameters, and the machine executes these automatically,
one after the other.
Figure 3.32 is an example of a supervisory model structure that can be found in many control mechanisms, such as
the automated flight control system, cruise control of a car,
and robots on assembly lines. The modeling structure consists
of hierarchical layers of superstates, each with its own set of
models. The supervisory models in the Automated Flight
Control System are organized hierarchically. Three main
levels are described in Fig. 3.32. The highest level of automation is the vertical navigation model (level 3), depicted as
a superstate at the top of the models pyramid. Two submodels are encapsulated in the vertical navigation model; VNAV
Speed and VNAV Path; each one exhibiting a somewhat different control behavior. One level below (level 2) are two
semiautomatic models: vertical speed and altitude hold.
One model in the Automated Flight Control System, altitude capture, can only be engaged automatically; no direct
manual engagement is possible. This model engages automatically when the aircraft is beginning the level-off maneuver to capture the selected altitude. When the aircraft is
Zhang_Ch03.indd 363
5/13/2008 5:41:35 PM
364
INDUSTRIAL CONTROL TECHNOLOGY
Automated flight control system models
Semiautomatic models
Vertical navigation
VNAV speed
VNAV path
3
Vertical speed
Altitude hold
2
m3
1
m4
Altitude capture
Figure 3.32 A supervisory model.
several hundred feet from the selected altitude, an automatic
transition from any climb model to altitude capture takes
place (m3). In this aspect, an example can be that a transition
from vertical navigation or vertical speed to altitude capture
takes place (m3). Finally, when the aircraft reaches the
selected altitude, a transition back from altitude capture to
altitude hold model also takes place automatically (m4).
In summary, we have illustrated a modeling language, Start Charts, for
representing human interaction with control systems, and proposed a classification of three different types of models that are employed in computers, devices, and supervisory control systems. The “Interface models”
change display format. The “Functional models” allow for different functions and associated parameters. Last are “Supervisory models” that specify the level of supervision (manual, semiautomatic, and fully automatic)
in the human–machine system. The three types of models described here
are essentially similar, in that they all define the manner in which a certain
component of the machine behaves. The component may be the interface
Zhang_Ch03.indd 364
5/13/2008 5:41:35 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
365
only, a function of the machine, or the level of supervision. This commonality brings us back to our general working definition of the term “model,”
a machine configuration that corresponds to unique behavior.
3.3.2.2
Systems of Human–Machine Interactions
There are three architectures of human–machine systems that are currently popular in industrial control: (1) adaptive human–machine interface,
(2) supervisory human–machine interface, (3) distributed human–machine
interface.
(1) Adaptive human–machine interface. In a complex control system,
human–machine interface is attempted to give users the means to
perceive and manipulate easily huge quantities of information
under resource constraints such as time, cognitive workload,
devices, etc. The intelligence in the human–machine interface
makes the control system more flexible and more adaptable. One
subset of intelligent user interface is adaptive interfaces. An adaptive interface modifies its behavior according to some defined
constraints in order to best satisfy all of them, and varies in the
ways and means that are used to achieve adaptation.
In human–machine interface of an industrial control, operators, system, and context are continuously changing and are
sometimes in contradiction with one another. Thus, this kind of
interface can be viewed as the result of a balance between these
three components and their changing relative importance and
priority. An adaptive interface aims at assisting the operator in
acquiring the most salient information in any context, in most
appropriate form, and at the most opportune time. In any industrial control systems, two main factors are considered: the system that generates the information stream and the operator to
which this stream is presented. The system and the operator share
a common goal: to control the process and to solve any problems
that may arise. This common objective makes them cooperate
although they may both have their own goals. The role of the
interface is to integrate these different goals with different levels
of importance and the various constraints that come from the task,
the environment, or the interface itself in order to produce an
information presentation that best harmonizes the set of all these
parameters. Specifically, an industrial process control should react
consistently, in a timely fashion without disturbing the operator
needlessly in his task. However the most salient pieces of information should be presented in the most appropriate way.
Zhang_Ch03.indd 365
5/13/2008 5:41:35 PM
366
INDUSTRIAL CONTROL TECHNOLOGY
The two main adaptation triggers used to modify the human–
machine interface could be
(a) The process. When the process moves from a normal state to
a disturbed state, the streams of information may become
denser and more numerous. To avoid any cognitive overload
problems, the interface acts thus as a filter that channels the
streams of information. To this end, it adapts the presentation of the pieces of information in order to help the operator
identify and solve the problem more quickly, more easily,
and more efficiently.
(b) The operator. An operator is much more difficult and trickier
to drive the adaptation on the user state. As a matter of fact,
the interface has to infer whether the operator reacts incorrectly and needs help based on his actions. Then, it may
decide to adapt itself to highlight the problem and suggest
solutions to assist the user.
The aim of the adaptation triggered by the process or the
operator is to adapt the composition of the streams of information, that is, to adapt the organization and the presentation
of the pieces of information on the interface in the best possible way, according to the state of the process and the
inferred state of the operator. What is expected is to improve
the communication between the system and the user. The
means proposed in an adaptive human–machine interface to
reach this are the following:
(i) Highlight relevant pieces of information. The importance of a piece of information depends on its relevance
according to the particular goals and constraints of each
of the entities that participates in the communication
between the system and the operator.
(ii) Optimize space usage. According to the current usage
of the resources, it may turn out that it is necessary to
reorganize the display space to cope with new constraints and parameters.
(iii) Select the best representation. According to the piece of
information, its importance, the resources available,
and the media currently in usage, the most appropriate
media to communicate with the operator should be used.
(iv) Timeliness of information. The display of a particular
piece of information should be timely regarded the process and the operator. This adaptation should follow the
evolution of the process over time, but it should also
adapt the timing of the displayed information to the
inferred needs of the operator.
Zhang_Ch03.indd 366
5/13/2008 5:41:36 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
367
(v) Perspectives. In traditional interfaces, the operator has
to decide what, where, when, and how the information
should be presented. Thus, an operator can wonder
whether he or she needs an adaptive interface to achieve
their task or whether it will constitute more a bother
than a help to him or her. This raises at least four questions: (1) From the client’s point of view, whether the
cost of developing an adaptive human–machine interface is justifiable. (2) From the human–machine interface designer’s point of view, whether this kind of
interface is usable and how to evaluate its usability.
(3) From the developer’s point of view, what are the best
technical solutions to implement an efficient system
within the required time? (4) From the operators’ point
of view, whether they consider such an adaptive interface as a collaborator or a competitor.
(2) Supervisory human–machine interface. Supervisory human–
machine interface can be used in such systems and similar ones
where there is a considerable distance between the control room
and the machine house in a plant. It is from this machine house
that the controller such as SCADA or PLC controls the objects
which are, for example, pumps, blowers and purification monitors, etc. To provide the data communication, the supervisory
software of the human–machine interface is linked with the controllers over a network such as Ethernet, Control Area Network,
and so on. The supervisory software is such that only one person
is needed at any one time to monitor the whole plant from a single master device. The generated graphics show a clear representation on screen of the current status of any part of the system. A
number of alarms are automatically activated directly on screen
if parameters deviate from their tight tolerance band. This ensures
extremely rapid updating of the control room screen contents.
All the calculations for the controllers are calculated by the control software, using constant feedback from sensors throughout
the production process.
According to many applications, the supervisory human–
machine interface is indeed an ideal software package for these
cases above. Thanks to its interactive configuration and its
setting assistants, supervisory human–machine interface is able
to straightforwardly get the system up and running and tested
out. Supervisory human–machine interface is an open architecture which offers all the functions and options necessary for data
collection and graphical representation of data on the operator
screen. The system provides comprehensive logging of all
Zhang_Ch03.indd 367
5/13/2008 5:41:36 PM
368
INDUSTRIAL CONTROL TECHNOLOGY
measured values with databases. By accessing this database and
by using real-time measurements, a wide variety of reports and
trend curves can be viewed on the screen or output to a printer.
(3) Distributed human–machine interface. The distributed human–
machine interface is a component-based approach. In a system
of such an approach, the human–machine interface could directly
access any controller component, which also means that each
controller exposes the human–machine interface. Since all the
system components are location transparent the human–machine
interface can bind to a component anywhere, be it in-process,
local-process, or remote process. The most likely case is remote
binding because it would be assumed that the human–machine
interface and the controller would reside on different platforms.
In the distributed human–machine interface, the multiple servers are typically used to provide the systems with the flexibility
and power of a peer-to-peer architecture. Each controller can
have its own human–machine interface server. The assigned server
to a controller or a proxy server of a controller is perfect for each
controller to manage expansion, frequent system changes, maintenance, and replicated automation lines within or across plants.
Instead of a single data server, each controller component provides its own data services through a proxy server.
The primary drawback to decentralized components is the
uncertainty of real-time controller performance generally resulting from poorly designed proxy agent use, for example, if the human–
machine interface samples controller data at too high a frequency.
However, the distributed human–machine interface is ideal for
SCADA applications. Its distributed peer-to-peer architecture,
reusable components, and remote deployment and maintenance
capabilities make supporting SCADA applications remarkably
efficient. The software’s network services have been optimized
for use over slow and intermittent networks, which significantly
enhance application deployment and communications.
3.3.2.3
Designs of Human–Machine Interactions
The design for a human–machine interface is important, because
the human–machine interface of an application will often make or break
this application. Although the functionality that an application provides
to users is important, the way in which it provides that functionality is
of the same importance. An application that is difficult to use will not
be used. So, the value of human–machine interface design should not be
underestimated.
Zhang_Ch03.indd 368
5/13/2008 5:41:36 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
369
(1) Design principles. The following describes a collection of principles for improving the quality of human–machine interface
design.
(a) The structure principle. The human–machine interface design
should organize the interface purposefully, in meaningful and
useful ways based on clear, consistent models that are apparent
and recognizable to users, putting related things together and
separating unrelated things, differentiating dissimilar things
and making similar things resemble one another. The structure
principle is concerned with overall interface architecture.
(b) The simplicity principle. The human–machine interface
design should make simple, common tasks simple to do,
communicating clearly and simply in the user’s own language, and providing good shortcuts that are meaningfully
related to longer procedures.
(c) The visibility principle. The human–machine interface design
should keep all needed options and materials for a given task
visible without distracting the user with extraneous or redundant information. Good designs do not overwhelm users
with too many alternatives or confuse them with unneeded
information.
(d) The feedback principle. The human–machine interface
design should keep users informed of actions or interpretations, changes of state or condition, and errors or exceptions
that are relevant and of interest to the user through clear,
concise, and unambiguous language familiar to users.
(e) The tolerance principle. The human–machine interface
design should be flexible and tolerant, reducing the cost of
mistakes and misuse by allowing undoing and redoing, while
also preventing errors wherever possible by tolerating varied
inputs and sequences and by interpreting all reasonable
actions.
(f) The reuse principle. The human–machine interface design
should reuse internal and external components and behaviors,
maintaining consistency with purpose rather than merely
arbitrary consistency, thus reducing the need for users to
rethink and remember.
(2) Design process.
(a) Phase one. The design process begins with a task analysis in
which we identify all the stakeholders, examine existing
control or production systems and control and production
processes, whether they are paper-based or computerized,
and identify ways and means to streamline and improve
the control or production process. Tasks at this phase are to
Zhang_Ch03.indd 369
5/13/2008 5:41:36 PM
370
INDUSTRIAL CONTROL TECHNOLOGY
conduct background research, interview stakeholders, and
observe people conducting tasks.
(b) Phase two. Once having an agreed-upon objective and set of
functional requirements, the next step should go through a
design phase to generate a design that meets all the requirements. The goal of the design process is to develop a coherent, easy to understand software front-to-end that makes
sense to the eventual users of the system. The design and
review cycle should be iterated until we are satisfied with our
design.
(c) Phase three. The next phase is implementation and test.
We are also developing and implementing any performance
support aids as necessary, for example, on-line help, paper
manuals, etc.
(d) Phase four. Once a functional system is complete, we
move into the final phase. What constitutes “success” is that
people using our system, whether it be an intelligent tutoring
system or online decision aid, are able to see solutions
that they could not see before and/or better understand the
constraints that are in place. We generally conduct formal
experiments, comparing performance using our system with
perhaps different features turned on and off to make contributions to the literature on decision support and human–
machine interaction.
(3) Design evaluation. An important aspect of human–machine
interaction is the methodology for evaluation of user interface
techniques. Precision and recall measures have been widely used
for comparing the ranking results of noninteractive systems, but
are less appropriate for assessing interactive systems. The standard evaluations emphasize high recall levels. However, in many
interactive settings, users require only a few relevant documents
and do not care about high recall to evaluate highly interactive
information access systems. Useful metrics beyond precision
and recall include: time required to learn the system, time
required to achieve goals on benchmark tasks, error rates, and
retention of the use of the interface over time.
Empirical data involving human users is time consuming to
gather and difficult to draw conclusions from. This is due in part
to variation in users’ characteristics and motivations, and in part
to the broad scope of information access activities. Formal
psychological studies usually only uncover narrow conclusions
within restricted contexts. For example, quantities such as the
length of time it takes for a user to select an item from a fixed
Zhang_Ch03.indd 370
5/13/2008 5:41:36 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
371
menu under various conditions have been characterized empirically, but variations in interaction behavior for complex tasks
like information access are difficult to account for accurately. A
more informal evaluation approach is called a heuristic evaluation in which user interface affordances are assessed in terms of
more general properties and without concern about statistically
significant results.
3.3.3
Interfaces
For all industrial control systems, a high degree of user friendliness at
the interface between human and machine is a decisive prerequisite for
being accepted by the general public. Ambient Intelligence applications
are characterized by multimodal interfaces as well as by the pro-active
behavior of the controller system. Therefore, various interfaces must be
sensibly combined with each other, and the interaction with humans must
be perfectly adapted to the individual situation of the human. Specific
challenges include, among others, the selection of suitable interfaces for
specific applications, the dynamic changes of interfaces based on changes
in the state of the human such as “experiences gained” or “accident,” as
well as the experience-based optimization of such interfaces.
Regarding selection, methods are currently being developed that can
suggest suitable interfaces based on a comprehensive characterization of
the requirements. This methodology is very comprehensive and complex,
since the requirements involve human properties such as their desire for
information or personal preferences. Further evaluation and optimization
of the methodology are absolutely indispensable. This work urgently
requires the collaboration of psychologists. With respect to the dynamic
changing of interfaces, this must be supported at least by semiautomatic
generation. Experience-based patterns may be a suitable approach for this.
Concerning the optimization of interfaces, an increase of acceptance
through experience-based optimization can be envisioned. Such assistance
systems already exist in vehicles, where, for example, the type of acceleration can be adapted to the style of driving of the respective driver.
3.3.3.1
Devices
(1) Operator interface terminals. These human–machine interfaces
are operator interface terminals with which users interact in order
to control other devices. Some human–machine interfaces include
Zhang_Ch03.indd 371
5/13/2008 5:41:36 PM
372
INDUSTRIAL CONTROL TECHNOLOGY
knobs, levers, and controls. Others provide programmable function keys or a full keypad. Devices that include a processor or
interface to personal computers are also available. Many human–
machine interfaces include alphanumeric or graphic displays.
For ease of use, these displays are often backlit or use standard
messages. When selecting human–machine interfaces, important
considerations include devices supported and devices controlled.
Device dimensions, operating temperature, operating humidity,
and vibration and shock ratings are other important factors.
Many human–machine interfaces include flat panel displays
(FPD) that use liquid crystal display (LCD) or gas plasma technologies. In LCD, an electric current passes through a liquid
crystal solution that is trapped between two sheets of polarizing
material. The crystals align themselves so that light cannot pass,
producing an image on the screen. LCD can be monochrome or
color. Color displays can use a passive matrix or an active matrix.
Passive matrix displays contain a grid of horizontal and vertical
wires with an LCD element at each intersection. In active matrix
displays, each pixel has a transistor that is switched directly on
or off, improving response times. Unlike LCD, gas plasma displays consist of an array of pixels, each of which contains red,
blue, and green subpixels. In the plasma state, gas reacts with the
subpixels to display the appropriate color.
These human–machine interfaces differ in terms of performance specifications and I/O ports. Performance specifications
include processor type, random access memory (RAM), and
hard drive capacity, and other drive options. I/O interfaces allow
connections to peripherals such as mice, keyboards, and modems.
Common I/O interfaces include Ethernet, Fast Ethernet, RS-232,
RS-422, RS-485, small computer system interface (SCSI), and
universal serial bus (USB). Ethernet is a local area network
(LAN) protocol that uses a bus or star typology and supports data
transfer rates of 10 Mbps. Fast Ethernet is a 100 Mbps specification. RS-232, RS-422, and RS-485 are balanced serial interfaces
for the transmission of digital data. Small computer systems
interface (SCSI) is an intelligent I/O parallel peripheral bus with
a standard, device-independent protocol that allows many peripheral devices to be connected to the SCSI port. Universal serial
bus (USB) is a four-wire, 12-Mbps serial bus for low-to-medium
speed peripheral device connections.
These human–machine interfaces are available with a variety of
features. For example, some devices are web-enabled or networkable. Others include software drivers, a stylus, and support for a
keyboard, mouse, and printer. Devices that provide real-time
Zhang_Ch03.indd 372
5/13/2008 5:41:36 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
373
clock support use a special battery and are not connected to the
power supply. Power-over-Ethernet (PoE) equipment eliminates
the need for separate power supplies altogether. These human–machine interfaces that offer shielding against electromagnetic
interference (EMI) and radio frequency interference (RFI) are
commonly available. Devices that are designed for harsh environments include enclosures that meet standards from the
National Electronics Manufacturers’ Association (NEMA).
(2) Operator interface monitors. Machine controllers and monitors
use electronic numeric control and a monitoring interface for
programming and calibrating computerized machinery. This
product area includes general-purpose machine controllers,
embedded machine controllers, machine monitors, CNC stepper
motors, and CNC router controllers. A machine controller is a
programmable, automatic, and computer numerically controlled
(CNC) device. An embedded machine controller is part of a
larger system. A machine monitor is used to collect and display
production data from production equipment such as presses. A
CNC stepper motor is used to drive a machine tool with power
and precision. A CNC router controller is used to cut tool paths.
Many other types of machine controllers and monitors are also
available.
Machine controllers and monitors consist of many different
components. A machine controller uses a microprocessor to perform predetermined control and logical operations. Memory is
added to the processor in order to record data from the machine.
Often, an input device is used to provide menus or options. Some
embedded machine controllers provide 16-axis pulse motion
control capabilities. Others include antivibration design mechanisms. Machine monitors track a machine’s uptime, downtime,
and idle time. They also allow operators to enter a reason for
downtime or nonproductive activities. In some cases, a machine
monitor can be programmed to require the entry of a reason code
after each downtime event. In this way, machine controllers and
monitors can be configured to meet the needs of specific machinery and industries.
Machine controllers and monitors are used in many different
applications. Some machine control products are used to regulate medical equipment such as respirators. Others are used in
aerospace, automotive, or military applications. An embedded
machine controller can be used in a printing machine, pipe bending equipment, CNC stepper motor, or CNC router controller.
Embedded machine controllers are also used in the manufacture
of semiconductors and electronic devices. Machine controllers
Zhang_Ch03.indd 373
5/13/2008 5:41:36 PM
374
INDUSTRIAL CONTROL TECHNOLOGY
and monitors with integral software are used in industries where
reliability, quality, and cost are important considerations.
(3) Industrial control pendants. Industrial control pendants are
sophisticated, hand-held terminals that are used to control robot
or machine movements from point to point, within a determined
space. They consist of a hanging control console furnished with
joysticks, push buttons, or rotary cam switches. A type of industrial control pendant, teach pendants are the most popular robotics teaching method, and are used widely with all types of robots,
in many industries. As the robot moves within this determined
space, the various points are recorded into its memory banks,
and can be located later on through subsequent playback. There
are a number of teach pendant types available, depending on the
type of application for which they will be used. If the goal is
simply to monitor and control a robotics unit, then a simple control box style is suitable. If additional capabilities such as on the
fly programming are required, more sophisticated boxes should
be used.
Industrial control pendants are equipped with switches, dials,
and pushbuttons through which data is relayed to the robotics
unit, and additional monitoring systems if necessary. The relationship between industrial control pendants and their subservient unit is generally established via an interconnected cabling
system. However, more advanced wireless devices are also
available.
During use, the operator actuates the switches on manual pendants in a specific order. This, in turn, causes the robot, end effectors, or machine, to move to and from the desired points. As the
end effector reaches the desired point, the operator uses the
record pushbutton to enter the location into the robot, or robot
controller’s memory banks. This is the most common programming method for playback robots.
The usage of industrial control pendants is common; however
it has a significant disadvantage in that the operator must divert
his and her attention away from the movement of the machine
during programming in order to locate the appropriate pushbutton to move the robot. The use of a joystick provides a solution
to this problem as the movement of the stick in a certain direction propels the robot or machine in that direction. This option is
available on more advanced industrial control pendant types.
(4) SCADA HMI (human–machine interface) devices. Distributed
control systems (DCS) and supervisory control and data acquisition (SCADA) systems are system architectures for process control applications. A distributed control system (DCS) consists of
Zhang_Ch03.indd 374
5/13/2008 5:41:36 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
375
a programmable logic controller (PLC) that is networked both to
other controllers and to field devices such as sensors, actuators,
and terminals. A DCS may also interface to a workstation. A
SCADA system is a process control application that collects data
from sensors or other devices on a factory floor or in remote
locations. The data is then sent to a central computer for management and process control. SCADA systems provide shop floor
data collection and may allow manual input via bar codes and
keyboards. Both distributed control systems (DCS) and supervisory control and data acquisition (SCADA) systems often include
integral software for monitoring and reporting.
There are several parts to a supervisory control and data
acquisition system or SCADA system. To control SCADA, a
SCADA system integrator, SCADA security, and SCADA HMI
are required. A SCADA system integrator is used to interface a
SCADA system to an external application. SCADA security uses
one or more computers at a remote site to monitor and control
sensors or shop floor devices. SCADA security includes remote
terminal units (RTU), a communications infrastructure, and a
central control room where monitoring devices such as workstations are housed. SCADA HMI is a human machine interface
that accounts for human factors in engineering design.
Distributed control systems (DCS) and supervisory control
and data acquisition (SCADA) systems are used in a variety of
industries. Distributed control systems or DCS systems are used
to control traffic lights and manage chemical processing, pharmaceutical, and power generation facilities. Supervisory control
and data acquisition systems or SCADA systems are used in
warehouses, petrochemical processing, iron and steel production, food processing, and agricultural applications. Providers
of distributed control systems and supervisory control and data
acquisition systems are located across the United States and
around the world.
3.3.3.2 Tools
Operator interface mounts and arms are articulating components used
to hold and position industrial computer monitors, keyboards, or other
operator interfaces. Operator interface mounts and arms are designed to
improve the physical and spatial relationships between machines and the
humans that operate them. The science of these relationships, called
ergonomics, is the study of human–machine interactions. Ergonomically
compatible products are designed to maximize productivity and minimize
Zhang_Ch03.indd 375
5/13/2008 5:41:36 PM
376
INDUSTRIAL CONTROL TECHNOLOGY
operator fatigue, discomfort, and injury. The goal of using operator interface mounts and arms as part of an ergonomics program is to reduce injuries, illnesses, and musculoskeletal disorders in the workplace.
Several of the most common types of ergonomic operator interface
mount and arm products include computer accessories (e.g., keyboard
drawer, mouse tray, glare screen, wrist rest, and monitor support arm) and
workstation accessories (e.g., instrumentation booms, articulating supports, foot rests, chairs, document stands). A monitor support arm is a type
of operator interface mount that is designed to support computer screens
or monitors in work stations, control centers, and operating theaters. A
support arm should combine stability and full adjustability to meet the
operator’s needs. A support arm can be a desk, wall, ceiling, or mobile
mounting arm. Articulating supports are movable support arms that a user
can readjust for the height or location of monitors and equipment in relation to the user’s eyes or hands. Keyboard drawers are used to store unused
keyboards. A monitor drawer mounts an LCD and keyboard within a rack
frame or enclosure so that a monitor can be folded down and stored when
not in use. Instrumentation booms are another type of operator interface
mounts and arms. This type of operator interface mount is used to support
various types of equipment, including computer or industrial monitors,
video equipment, and manufacturing equipment.
The U.S. Occupational Safety and Health Administration (OSHA) has a
four-pronged, comprehensive approach to ergonomics, including operator
interface mounts and arms, that is designed to quickly and effectively
address musculoskeletal disorders in the workplace. The OSHA approach
includes industry or task specific guidelines, enforcement actions, outreach and assistance activities, and a national advisory committee.
3.3.3.3
Software
The human–machine interface (HMI) software enables operators to
manage industrial and process control machinery via a computer-based
graphical user interface (GUI). There are two basic types of HMI software:
supervisory level and machine level. The supervisory level is designed for
control room environment and used for system control and data acquisition (SCADA), a process control application which collects data from sensors on the shop floor and sends the information to a central computer for
processing. The machine level uses embedded, machine-level devices
within the production facility itself. Most human–machine interface software is designed for either supervisory level or machine level; however,
Zhang_Ch03.indd 376
5/13/2008 5:41:36 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
377
applications that are suitable for both types of HMI are also available.
These software applications are more expensive, but can eliminate redundancies and reduce long-term costs.
Selecting human–machine interface software requires an analysis
of product specifications and features. Important considerations include
system architectures, standards and platforms; ease of implementation,
administration, and use; performance, scalability, and integration; and
total costs and pricing. Some human–machine interface software provides
data logging, alarms, security, forecasting, operations planning and control (OPC), and ActiveX technologies. Others support data migration from
legacy systems. Communication on multiple networks can support up to
four channels. Supported networks include ControlNet and DeviceNet.
ControlNet is a real-time, control-layer network that provides high-speed
transport of both time-critical I/O data and messaging data. DeviceNet is
designed to connect industrial devices such as limit switches, photoelectric cells, valve manifolds, motor starters, drives, and operator displays to
programmable logic controllers (PLC) and personal computers (PC).
Some human–machine interface software runs on Microsoft Windows
CE, a version of the Windows operating system that is designed for
hand-held devices. Microsoft and Windows are registered trademarks of
Microsoft Corporation. Windows CE allows users to deploy the same
human–machine interface software on distributed HMI servers, machinelevel embedded HMI, diskless open-HMI machines, and portable or pocketsized HMI devices.
3.4 Highway Addressable Remote Transducer
(HART) Field Communications
HART is an acronym for “Highway Addressable Remote Transducer”
that represents a two-way digital communication simultaneously with the
4–20 mA analog signaling used by traditional instrumentation equipment
in industrial process control. HART was developed in the early 1980s by a
company named Rosemount Inc. for the host to perform the management
of the field devices in industrial systems. In July 1993, the HART
Communication Foundation was established to provide worldwide support for application of this technology. HART Specifications continue to
be updated to broaden the range of HART applications. A recent HART
development, the Device Description Language (DDL), provides a universal software interface to new and existing devices.
Zhang_Ch03.indd 377
5/13/2008 5:41:36 PM
378
INDUSTRIAL CONTROL TECHNOLOGY
3.4.1
HART Communication
Most industrial control systems include numerous field system functions. The host controller, therefore, should instantly communicate with
all the field instruments and devices while control processes are running
for (1) device configuration or reconfiguration, (2) device diagnostics,
(3) device troubleshooting, (4) reading the values of additional measurements provided by the device, (5) device health and status, and other
requirements. A host in the system can be a Distributed Control System,
PLC, and Asset Management System, Safety System, or a hand-held device.
By fully using HART communication, industrial control can be benefited
in many aspects. Utilizing the full capabilities of HART-enabled devices
and systems reduces costs by improving plant operations and increasing
efficiency and helps to avoid the high cost of process disruptions and
unplanned shutdowns. Properly utilized, the intelligent capabilities of
HART-smart devices are a valuable resource for keeping plants operating
at maximum efficiency. Real-time HART integration with plant control,
safety, and asset management systems unlocks the value of connected
devices and extends the capability of systems to detect any problems with
the device, its connection to the process, or interference with accurate
communication between the device and system.
The world’s leading process automation control systems and instrumentation suppliers all support HART Communication in their field device and
system products. Most automation system suppliers offer direct HARTenabled I/O and PC-based software applications to leverage the intelligence in HART-smart field devices for continuous device condition
monitoring, real-time diagnostics, and multivariable process information.
3.4.1.1
HART networks
(1) Wired HART networks. There are two kinds of wired HART networks available in the industrial control systems. Figure 3.33
displays the architectures of these two wired HART networks;
the first (Fig. 3.33(a)) is a point-to-point HART network, and the
second (Fig. 3.33(b)) is a multiple-dropped HART network. As
shown in Fig. 3.33, wired HART networks include the Host
Controller and some field devices that can be transmitters;
between the Host and the field devices, this system has an I/O
system that can be the system interface with the HART, and a
hand-held terminal or hand-held communicator.
The type of network with a single Field Instrument that does
both HART network functions and analog signaling is probably
Zhang_Ch03.indd 378
5/13/2008 5:41:36 PM
379
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
Analog
HART interface
Digital data
(2–3 updates
per second)
Primary master:
control system
or other host
application
Power
supply
Transmitter
Secondary master
(a)
Control system or other host
application
Handheld terminal
Input/output (I/O)
system
Field devices
(b)
Figure 3.33 The architecture of HART networks: (a) the point-to-point HART
network and (b) the multiple-dropped HART network.
Note: Intrument power is provided by an internal or external power source that is not shown.
the most common type of wired HART network and is called a
point-to-point network. In some cases the point-to-point network
might have a HART Field Instrument but no permanent HART
Master. This might occur, for example, if the user intends primarily analog communication and Field Instrument parameters
Zhang_Ch03.indd 379
5/13/2008 5:41:36 PM
380
INDUSTRIAL CONTROL TECHNOLOGY
are set prior to installation. A HART user might also set up this
type of network and then later communicate with the Field
Instrument using a hand-held communicator (HART Secondary
Master). This is a device that clips onto device terminals (or other
points in the network) for temporary HART communication with
the Field Instrument.
A HART Field Instrument is sometimes configured so that it
has no analog signal, only HART function. Several such Field
Instruments can be connected together (electrically in parallel)
on the same network, as in Fig. 3.34. These Field Instruments are
said to be multiple-dropped. The Master is able to talk to and
configure each one, in turn. When Field Instruments are multidropped there cannot be any analog signaling. The term “current
loop” ceases to have any meaning. Multiple-dropped Field
Instruments that are powered from the network draw a small,
fixed current (usually 4 mA) so that the number of devices can be
maximized. A Field Instrument that has been configured to draw
a fixed analog current is said to be “parked.” Parking is accomplished by setting the short-form address of the Field Instrument
to some number other than 0. A hand-held communicator might
also be connected to the network of Fig. 3.34.
There are few restrictions on building wired HART networks.
The topology may be loosely described as a bus, with drop
attachments forming secondary busses as desired, which are
illustrated in Fig. 3.35. The whole collection is considered a
single network. Except for the intervening lengths of cable,
all of the devices are electrically in parallel. The Hand-Held
Communicator (HHC) may also be connected virtually anywhere. As a practical matter, however, most of the cable is inaccessible and the HHC has to be connected at the Field Instrument,
in junction boxes, or in controllers or marshalling panels. In
Control
area
Field area
HART
Master
Current
loop
(network)
. .
.
Field instrument 1 and 2 and Nth
Figure 3.34 HART network with multiple-dropped field instruments.
Zhang_Ch03.indd 380
5/13/2008 5:41:38 PM
381
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
F1
F1
Control Field
area
area
Single
twisted pair
Secondary bus
F1
F1
Master
Main bus
F1
Field
instrument
HHC
F1
Figure 3.35 HART network showing free arrangement of devices.
intrinsically safe (IS) installations there will likely be an IS barrier separating the control and field areas.
A Field Instrument may be added or removed or wiring
changes made while the network is live (powered). This may
interrupt an on-going transaction. However, if the network is
inadvertently short-circuited, this could reset all devices. The
network will recover from the loss of a transaction by retrying a
previous communication. If Field Instruments are reset, they will
eventually come back to the state they were in prior to the reset.
No reprogramming of HART parameters is needed.
Digital signaling brings with it a variety of other possible
devices and modes of operation. For example, some Field
Instruments are HART only and have no analog signaling. Others
draw no power from the network. In still other cases the network
may not be powered (no DC). There also exist other types of
HART networks that depart from the conventional one described
here. These are covered in another section.
(2) Wireless HART networks. Wireless HART is the first open and
interoperable wireless communication standard designed to
address the critical needs of the process industry for reliable,
robust, and secure wireless communication in real world industrial plant applications.
A Wireless HART Network consists of Wireless HART field
devices, at least one Wireless HART gateway, and a Wireless
HART network manager. These components are connected into
Zhang_Ch03.indd 381
5/13/2008 5:41:39 PM
382
INDUSTRIAL CONTROL TECHNOLOGY
a wireless mesh network supporting bidirectional communication from the HART host to field device and back. Figure 3.36
gives the typical Wireless HART network architecture with the
principal devices plotted:
(a) Network manager. The Network Manager is an application
that manages the mesh network and Network Devices.
The Network Manager performs the following functions:
(1) Forms the mesh network, (2) Allows new devices to connect to the network, (3) Sets the communication schedule of
the devices, (4) Establishes the redundant data paths for all
communications, (5) Monitors the network.
(b) Gateway devices. The gateway device connects the mesh
network with a plant automation network, allowing data to
flow between the two network devices. The gateway device
provides access to the Wireless HART devices by a system
or other host application.
(c) Network devices. A network device is a node in the mesh
network. It can transmit and receive Wireless HART data
and perform the basic functions necessary to support network formation and maintenance. Network devices include
Plant
automation
network
n
m
Host appllication
(e.g., Asset management)
l
Gateway
g
l
m
Field devices
Network manager
e
g
c
f
Handheld
Process automation
controller
Gateway
Existing HART
devices
a
Adapter
Figure 3.36 Typical wireless HART architecture (courtesy of the HART
Communication Foundation).
Zhang_Ch03.indd 382
5/13/2008 5:41:39 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
383
field devices, router devices, gateway devices, and mesh
hand-held devices.
(d) Field devices. The field device may be a process connected
instrument, a router, or hand-held device. The Wireless HART
network connects these devices together.
(e) Router device. A router device is used to improve network
coverage (to extend a network) so that it is capable of forwarding messages from other network devices.
(f) Process connected instrument. Typically a measuring or
positioning device used for process monitoring and control,
it is also capable of forwarding messages from other network
devices.
(g) Adapter. An adapter is a device that allows a HART instrument without wireless capability to be connected to a
Wireless HART network.
(h) Hand-held support device. Hand-held devices are used in the
commissioning, monitoring, and maintenance of network
devices; they are portable and operated by the plant personnel.
Wireless HART networks can be configured in a number of
different topologies to support various application requirements
including the following:
(a) Star network. Star networks have just one router device that
communicates with several end devices. This is one of the
simplest network topologies. A star network may be appropriate for small applications.
(b) Mesh network. Mesh networks are formed by network
devices that are all router devices. Mesh networks provide a
robust network with redundant data paths which is able to
adapt to changing RF environments.
(c) Star mesh network. Star mesh networks are a combination of
the star network and mesh network.
3.4.1.2
HART Mechanism
HART communication occurs between two HART-enabled devices,
typically a field device and a control or monitoring system. To perform the
communication between the host and the field instruments, the analog
measurement signal is used to transmit digital information. For this purpose, an additional signal is modulated to the measurement signal using
the Frequency Shift Keying (FSK) process. The two frequencies of the
additional signal, 1200 and 2200 Hz, represent the bit values 1 and 0. This
makes it possible to transfer additional information without affecting the
analog measurement signal. As indicated by Fig. 3.37, HART provides
Zhang_Ch03.indd 383
5/13/2008 5:41:40 PM
384
INDUSTRIAL CONTROL TECHNOLOGY
+0.5 mA
HART signal
0
20 mA
–0.5 mA
1200 2200
Hz
Hz
“1”
“0”
C
Analog
signal
R
C
R
R
C
C
R
C = Command
R = Response
4 mA
0
1
Time (s)
2
Figure 3.37 HART signaling (digital and analog).
two simultaneous communication channels: the 4–20 mA analog signal
and a digital signal. The 4–20 mA signal communicates the primary measured value (in the case of a field instrument) using the 4–20 mA current
loop, the fastest and most reliable industry standard. Additional device
information is communicated using a digital signal that is superimposed
on the analog signal. The digital signal contains information from the
device including device status, diagnostics, additional measured or calculated values, etc. Together, the two communication channels provide a
complete field communication solution that is easy to use and configure, is
low cost and is very robust.
The HART signal path from the microprocessor in a sending device to
the microprocessor in a receiving device is displayed in Fig. 3.38.
Amplifiers, filters, and the network between these two interfaces have been
omitted for simplicity in Fig. 3.38. At this level the diagram is the same,
regardless of whether a Master or Slave is transmitting. Notice that, if the
signal starts out as a current, the FSK is a voltage. But if it starts out a voltage it stays a voltage.
The transmitting device begins by turning on its carrier and loading the
first byte to be transmitted into its interface circuits. It waits for the byte to
be transmitted and then loads the next one. This is repeated until all the
Zhang_Ch03.indd 384
5/13/2008 5:41:40 PM
385
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
FSK (current
or voltage)
Receiver’s interface (circuits and modem)
Sender’s interface (circuits and modem)
Sender’s
microprocessor
Receivers’
microprocessor
Figure 3.38 HART signaling path.
bytes of the message (these messages are always defined as commands
that are of predefined format) are exhausted. The transmitter then waits for
the last byte to be serialized and finally turns off its carrier. With minor
exceptions, the transmitting device does not allow a gap to occur in the
serial stream, the start and stop bits are used for synchronization, and the
parity bit is part of the HART error detection.
The serial character stream is applied to the modulator of the sending
modem. The Modulator operates such that a logic 1 applied to the input
produces a 1200 Hz periodic signal at the Modulator output. Logic 0 produces 2200 Hz. The type of modulation used is called Continuous Phase
Frequency Shift Keying (CPFSK). “Continuous Phase” means that there is
no discontinuity in the modulator output when the frequency changes.
When the sender’s interface output (modulator input) switches from logic
1 to logic 0, the frequency changes from 1200 to 2200 Hz with just a
change in slope of the transmitted waveform. A moment’s thought reveals
that the phase does not change through this transition. Given the chosen
shift frequencies and the bit rate, a transition can occur at any phase.
At the receiving end, the demodulator section of a modem in its interface converts FSK back into a serial bit stream at 1200 bps. Each character
is converted back into an 8-bit byte and parity is checked. The receiving
microprocessor reads the incoming bytes from its interface and checks
parity for each one until there are no more or until parsing of the data
stream indicates that this is the last byte of the message. The receiving
Zhang_Ch03.indd 385
5/13/2008 5:41:40 PM
386
INDUSTRIAL CONTROL TECHNOLOGY
processor accepts the incoming message only if its amplitude is high
enough to cause carrier detect to be asserted. In some cases, the receiving
processor will have to test an I/O line to make this determination. In others, the carrier detect signal gates the receive data so that nothing (no transitions) reaches the receiving interface unless carrier detect is asserted.
HART protocol puts most of the responsibility (such as timing and arbitration) into the Masters. This eases the Field Instrument software development and puts the complexity into the device that is more suited to deal
with it. A Master typically sends a command and then expects a reply. A
Slave waits for a command and then sends a reply. The command and
associated reply are called a transaction. There are typically periods of
silence (no device is allowed communicating) between transactions. A
Slave accesses the network as quickly as possible in response to a Master.
Network access by Masters requires arbitration. Masters arbitrate by
observing who sent the last transmission (a Slave or the other Master) and
by using timers to delay their own transmissions. Thus, a Master allows
time for the other Master to start a transmission. The timers constitute
dead time when no device is communicating and therefore contribute to
“overhead” in HART communication.
Each HART field instrument (in normal cases, a field instrument plays
a role of Slave) must have a unique address. Each command sent by a
Master contains the address of the desired Field Instrument. All Field
Instruments examine the command. The one that recognizes its own
address sends back a response. This address is incorporated into the command message sent by a Master and is echoed back in the reply by the
Slave. Addresses are either 4 bits or 38 bits and are called short and long
or “short frame” and “long frame” addresses, respectively. A Slave can
also be addressed through its tag (an identifier assigned by the user).
Each command or reply is a message that starts with the preamble and
is ended with the checksum. The preamble is allowed to vary in length,
depending on the requirements in the Slave end. Different Slaves can have
different preamble length requirements, so that a Master might need to
maintain a table of these values. A Master will use the longest possible
preamble when talking to a Slave for the first time. Once the Master reads
the Slave’s preamble, it first checks the length requirement (a stored HART
parameter), then will subsequently use this new length when talking to that
Slave. The checksum at the end of the message is used for error control. It
is the exclusive-OR of all of the preceding bytes, starting with the start
delimiter. The checksum, along with the parity bit in each character, creates a message matrix having so-called vertical and longitudinal parity. If
a message is in error, this usually necessitates a retry.
Zhang_Ch03.indd 386
5/13/2008 5:41:41 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
387
One more feature, available in some Field Instruments, is burst mode. A
Field Instrument that is burst-mode capable can repeatedly send a HART
reply without a repeated command. This is useful in getting the fastest possible updates (about 2–3 times per second) of process variables. If burst
mode is to be used, there can be only one bursting Field Instrument on the
network. A Field Instrument remembers its mode of operation during
power down and returns to this mode on power up. Thus, a Field Instrument
that has been parked will remain so through power down. Similarly, a Field
Instrument in burst-mode will begin bursting again on power up.
3.4.2
HART System
HART communication in industrial control comprises two folders:
HART connection system and HART protocol. This section focuses on the
HART system, and the next section will be on the HART protocol. HART
system devices work to support the HART protocol by communicating
their data over the transmission lines of the 4–20 mA connections. This
enables the field devices to be parameterized and initialized in a flexible
manner or to read measured and stored data (records). All these tasks
require field devices based on microprocessor technology. These devices
are frequently called smart devices.
For building and maintaining a HART-enabled system, the technical
kernel will be choosing the HART-compatible devices, installing the system’s devices, configuring the system’s devices, and calibrating the system’s devices. The key is to make sure that the engineers or designers are
requesting or specifying devices or systems that are fully compliant with
the HART protocol specification and are tested and registered with the
HART Communication Foundation (the HCF). The engineers or designers
are required to be assured of these aspects: interoperability with other
HART-compatible devices and HART-enabled systems; getting a device
that will provide the powerful features of HART technology; specifying a
product that will fully integrate into your HART-enabled applications.
3.4.2.1
HART System Devices
The devices constructing a HART connection system have several features that significantly reduce the time required to fully commission a
HART-enabled network. When less time is required for commissioning,
substantial cost savings are hence achieved. Devices which support the
HART protocol are grouped into master (host) and slave (field) devices.
Master devices include communicator or hand-held terminals as well as
Zhang_Ch03.indd 387
5/13/2008 5:41:41 PM
388
INDUSTRIAL CONTROL TECHNOLOGY
PC-based workplaces that stay in a control room. HART slave devices, on
the other hand, include sensors, transmitters, and various actuators. The
variety ranges from two-wire and four-wire devices to intrinsically safe
versions for use in hazardous environments. Field devices and communicators as well as compact hand-held terminals have an integrated FSKmodem, whereas computers or workstations have a serial interface to
connect the modem externally. HART communication is often used for
such simple point-to-point connections (Fig. 3.33(a)). Nevertheless, many
more connection variants are possible. In extended systems, the number of
accessible devices can be increased by using a multiplexer. In addition to
that, HART enables the networking of devices to suit special applications.
Network variants include multiple-dropped (Fig. 3.39), FSK-bus, and networks for split-range operation.
(1) HART Communicator. The HART Communicator is the most
widely used communicator across the world in industrial control.
The HART Communicators are portable devices; their weights
have been evenly distributed for comfortable one-handed operation in the field. The result is the universal, user upgradeable,
intrinsically safe, rugged and reliable Field Communicator. In
HART-enabled systems, the HART Communicators are often
PC host
Modem
Multiplexer
HART field device
Controller
HART signals
Address 0
4–20 mA
Address 0
Address 0
Figure 3.39 An HART connecting system including the FSK-modem and
HART-multiplexer, and HART-buses (courtesy of the SAMSON, Inc.).
Zhang_Ch03.indd 388
5/13/2008 5:41:41 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
389
defined by engineers as the Second Master (Fig. 3.33(a)) or
Hand-held Terminals (Fig. 3.33(b)).
With a memory and a microprocessor or some applicationspecific integrated circuits, the HART Communicator provides a
complete solution for configuring and monitoring all HART
devices and all Fieldbus devices of an industrial system.
It is comprised of three main components plus accessories.
These parts consist of the hand-held; the HART interface hardware, and the application software Suite. The Communicator
runs on a robust, real-time, operating system. This trio of hardware and software comprises a complete HART field communicator that can be powerful, multifaceted, and portable all in one.
The hardware for the HART Communicators primarily includes
the HART interface and the pinch connectors. The HART interface is designed to interface to the multiple connectors located
on the bottom of the hand-held, allowing communication between
the Palm and the HART network. The pinch connectors easily
connect to any HART network for instant communication. Most of
the HART interface requires no batteries, running solely off the
hand-held’s internal power supply. Its compact size and low power
consumption makes the interface an ideal solution for portability.
The software suite for the HART Communicators includes
some distinct applications. Each of these applications is preloaded onto the hand-held and designed for a particular function.
The main application of the suite allows communication, monitoring, and configuration of HART-compatible devices. The software is based upon manufacturer device description files (DDL)
and thus allows access to all menus and parameters as designed
by the manufacturer. This software application allows for the
logging of device variable values over time. A wide range of
variables can be logged automatically at a user selectable sample
time, or manually one by one. These logs can be saved and transferred to a PC for further analysis.
This graphing application allows device variables to be trended
over time in an easy to view graphical format. Device parameters
can be simultaneously graphed in various colors for easy identification. The display makes it easy to read in both bright sunlight
and in normal lighting. To make sure all conditions are covered,
a multilevel backlight is added, allowing the display to be viewed
in those areas of your plant with dim light. The touch sensitive
display and large physical navigation buttons provide for efficient use both on the bench and in the field.
User upgradeable HART and Fieldbus devices, as well as
functional updates to existing devices are introduced continually
Zhang_Ch03.indd 389
5/13/2008 5:41:41 PM
390
INDUSTRIAL CONTROL TECHNOLOGY
by device vendors. Keeping up-to-date with the required Device
Description (DD) drivers for all the devices in plant can be a
real challenge. Nowadays, with the Easy Upgrade option,
keeping communicators updated with the most current Device
Descriptions (DDs) is an easy job.
(2) FSK-Modem. There are often two kinds of modem required in
the HART-enabled industrial control networks: USB (Universal
Serial Bus) Modem and FSK (Frequency Shift Keying) Modem.
These two kinds of modem are all connected with the host PC
(Personal Computer) in the HART-enabled networks. This USBModem is just an ordinary PC modem used for computer networks, without special design for HART functions. However, the
FSK-Modem should be particularly designed for HART functions. The following will focus on the FSK-Modem.
The FSK-modem is designed to provide HART communication capabilities for the implementation of Frequency Shift
Keying (FSK) techniques to transfer data. The FSK-modem is
also required to conform to the HART network’s physical layer.
For this purpose, most FSK-Modems operate at the Bell 202
standard and are made into a chipset containing some integrated
circuits. As shown in Fig. 3.37, the FSK is the frequency modulation of a carrier of digital capability. For Simplex or Half
Duplex operation, the FSK-modem uses a single carrier in which
the communication can only be transmitted in one direction
at a time. For Full Duplex, the FSK-modem uses multiple carriers so that data communication can be simultaneous in both
directions.
The basic block diagram for the FSK-modem chipset is
depicted in Fig. 3.40, which illustrates the mechanism of a data
modulation and demodulation. This chip is divided into three
main parts: receive, transmit, and clock recovery. Both receive
Audio
signal input
Low-pass
filter
Received digital data
Descrambler
Clock
recovery
Recovered clock signal
TX clock
Transmitted
digital data
Scrambler
FIR
Anti-imaging
filter
Audio signal
output
Figure 3.40 Block diagram of an FSK-modem chipset.
Zhang_Ch03.indd 390
5/13/2008 5:41:41 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
391
and transmit blocks are separated and data can be processed in
each direction independently.
(a) Modulator for transmitting data. The data transmit part in
this chipset performs modulation, in which the scrambler is
to make a nearly flat spectrum of output signal. The output of
the scrambler is connected to a long digital FIR filter. This
FIR filter compensates distortion of transmission line and
removes sharp edges rising from High-Low or Low-High
logic transitions. The FIR filter makes the transmitted signal
spectrum narrow to fit bandwidth and compensates the signal for the receiver’s side. The transmit wave’s shapes are
stored in an EPROM (an electronic memory, see Chapter 2
of this book). In this way the transmitted waveform is
synthesized not only from the present bit’s state, but also
the four that preceded it and four to come. Data burnt
into EPROM represents filter response for each of 256
combinations.
(b) Demodulator for receiving data. At the receiving side, the
audio signal going from the discriminator of the transceiver
is passed through a low-pass filter to eliminate pertinent
higher frequencies, and remove out of band spurious noise
and residue. The signal is then limited and detected by sampling at the correct instant. At this point, the detected data,
still randomized, are passed through a descrambler, where
the original data are recovered and the result goes off to terminal. A descrambler, like a scrambler, is simply to provide
the invert function of the scrambler and perforce requires
some number of bits to synchronize.
(c) Clock recovery. The heart of the receiver is a digital phase
locked network (DPLL), which must extract a clock from
the received audio stream. It is needed to time the receiver
functions, including the all-important data detector. Each
waveform has a phase shift of 360/256° from another and is
made up of 16 samples. The received audio signal is limited,
and a zero crossing detector circuit generates one cycle of
9600 Hz for each zero-crossing (a proto-clock). This is compared with a locally generated clock in a phase detector
based on an up/down counter. The counter increments if
one clock is early, decrements otherwise. This count then
addresses an EPROM mentioned above. In this way, the
local clock slips rapidly into phase with that of the incoming
data. Local clock signal is derived from the output of
EPROM. Output data are converted to sine voltage with
maximum amplitude.
Zhang_Ch03.indd 391
5/13/2008 5:41:41 PM
392
INDUSTRIAL CONTROL TECHNOLOGY
Some FSK-modems have one more function block in
their chipsets; that is Carrier Detect. The carrier detect function in the FSK-modem is responsible for checking whether
or not the modulated or demodulated waves fall in some
range of frequencies. The carrier detect output is active low
whenever a valid carrier tone between some Hz (inclusive)
is detected. Detection occurs when timed transitions remain
within the band of these Hz periods for 10 nanoseconds to
1 bit time.
Some of the FSK-modem manufacturers use CMOS technologies to make the chipset. The FSK-modem is chipset
with pin-out specification. In a HART-enabled network, this
FSK-modem should be connected with the host microprocessor or the CPU (Central Processing Unit). Figure 3.41
illustrates how this modem’s pins connect with the CPU in
an FSK-modem.
(3) HART Multiplexer. In the HART-enabled industrial networks,
the HART Multiplexer acts as a gateway between the network
management computer and the HART-compatible field devices.
Many field devices are distributed over a wide area in industrial
process systems, and must be monitored and adapted to the
changes in the processing environment. The process system with
the HART Multiplexer enables on-line communication between
an asset management computer or workstation and those intelligent field devices that support the HART protocol so as to simultaneously fit into the changes in the process system. All actions
on the field device are parallel to the transmission of the 4–20 mA
MOD/DEMOD
I/O
Carrier detect
CPU
Bandpass
filter
INRTS
OCD
IRXA
HT2012
RX
TX
TX data
RX data
Waveshape
filter
ITXD
ORXD
OTXA
1460K
OLK
460.8 kHz
M
A
U
M
e
d
i
a
n
e
t
w
o
r
k
Figure 3.41 Typical hardware design of an FSK-modem.
Zhang_Ch03.indd 392
5/13/2008 5:41:41 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
393
measurement signals and have no influence on the measurement
value processing through the process system.
Each HART Multiplexer, regardless of whether it is a slave or
a master, provides a connection to a specified number of field
instruments. Up to thousands of field units can communicate and
exchange data with a computer or workstation. Working with a
hand-held terminal (HART-communicator) is also possible since
the HART protocol accepts two masters (computer and hand-held
terminal) in one system. Systems can be easily expanded and the
advantages of the HART communications can be exploited. At
the present technical level, the system consists of a maximum of
31 HART Multiplexer masters which are linked to the computer
with an RS485 interface. Each HART Multiplexer master controls up to 15 HART Multiplexer slaves.
At present, the HART Multiplexers are specified as the following three types:
(a) HART Multiplexer Master. This is a HART Multiplexer that
can operate up to 256 analog field instruments. The built-in
slave unit operates the first 16 loops. If more than 16 loops are
required, additional slave units can be connected. The slave
units are connected to the master with a 14-pin flat cable. The
connector for the ribbon cable is found on the same housing
side as the connectors for the interface and the power supply.
The analog signals are separately linked to a termination
board with a 26-pin cable for each unit. Sixteen leads are
reserved for the HART signal of the analog measurement
circuits. The remaining 10 leads are sent to ground. This unit
is designed with removable terminals and can be connected
to a Power Rail.
(b) HART Multiplexer Slave. This is a HART Multiplexer that
can operate up to 16 analog field instruments at the present.
In this case, the slave can only be operated with the HART
Multiplexer Master and is powered by the master across a
14-pin flat cable connection. Up to 15 slaves can be connected to the master. The slave address is set with a 16 position rotary switch (addresses 1–16). If only one slave is
connected to the master, then the slave address should be 1.
If multiple slaves are connected, slaves are to be assigned
addresses in ascending order. The analog signals are fed into
the slave by means of a 26-pin ribbon cable. Sixteen leads
are reserved for the HART signal of the analog measurement
circuits. The remaining 10 leads are assigned to ground.
(c) HART flexible interface. This is a flexible interface board
with a HART pick-up connector. This flexible interface
Zhang_Ch03.indd 393
5/13/2008 5:41:42 PM
394
INDUSTRIAL CONTROL TECHNOLOGY
board has 16 terminal blocks to connect up to 16 smart field
devices. This board can be used for general purpose applications or in conjunction with intrinsic safety barriers for hazardous area applications.
The specification of a HART Multiplexer should include
these important technical data:
(i) HART signal channels
(1) Leakage current (µA at some temperature range),
(2) Output termination External (measured by Ω),
(3) Output voltage (mVpp),
(4) Output impedance (measured by Ω, capacitively
linked),
(5) Input impedance per HART conventions,
(6) Input voltage range (mV-Vpp),
(7) Input voltage.
(ii) Power supply
(1) Nominal voltage (VDC),
(2) Power consumption (less and equal to some W).
(iii) Interface
(1) Type is RS-xxx (Some number of wire multidrops),
(2) Transmission speed 9600, 19200, 38400 baud,
(3) Address selection (32) possible RS-xxx addresses,
(4) The transmission speed (unit is “baud”) at the “ON”
and “OFF” state for every switch.
(iv) Mechanical data
(1) Mounting some mm DIN rail or wall mounted,
(2) Connection options some-pin ribbon cable for analog; some-pin ribbon cable for master–slave,
(3) Removable terminals, maximum some AWG for
interface and power supply.
(4) HART connecting buses. In industrial process systems, the
HART-enabled networks require several kinds of buses for connecting the HART-compatible devices and instruments. A brief
description of two of these buses is given below.
(a) Bus for split-range operation. In industrial process systems,
there are special applications which require that several (usually two) actuators receive the same control signal. A typical
example is the split-range operation of control valves. One
valve operates in the nominal current range from 4 to 12 mA,
while another valve uses the current range from 12 to 20 mA.
The split-range operation technique is a solution for this
case. In split-range operation, the control valves are connected in series in the current network. When both valves
have a HART interface, the HART host device must be
Zhang_Ch03.indd 394
5/13/2008 5:41:42 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
395
able to distinguish with which valve it must communicate.
To achieve this, the HART protocol revision 6 (anticipated
for autumn 1999) and later will be extended by one more
network variant. As is the case for multiple-dropped mode,
each device is assigned to an address from 1 to 15. The analog 4–20 mA signals preserve its device-specific function,
which is, for control valves, the selection of the required
travel.
(b) FSK-bus. The HART protocol can be extended by the
FSK-bus. Similar to a device bus, the FSK-bus can connect
approximately 100 HART-compatible devices and address
these devices with the technical level at the present. This
requires special assembly-type isolating amplifiers (e.g.,
TET 128). The only reason for the limited number of participants is that each additional participant increases the signal
noise. The signal quality is therefore no longer sufficient to
properly evaluate the telegram. The HART devices are connected to their analog current signal and the common FSKbus line with the isolating amplifier (Fig. 3.42). From the
FSK-bus viewpoint, the isolating amplifiers act as impedance converters. This enables devices with high load to be
integrated in the communication network. To address these
devices, a special, long form of addressing is used. During
the configuration phase, the bus address and the tag number
of each device are set with the point-to-point line. During
operation, the devices operate with the long addresses. When
using the HART command 11 (see the subsection below),
the host can also address the device via its tag. In this way,
the system configuration can be read and checked during the
start-up phase.
(5) HART system interface. HART communication between two or
more devices can function properly only when all communication participants are able to interpret the HART sine-wave signals correctly. To ensure this, not only must the transmission
lines fulfill certain requirements, but the devices in the current
network which are not part of the HART communication can
impede or even prevent the transmission of the data. The reason is
that the inputs and outputs of these devices are specified only for
the 4–20 mA technology. Because the input and output resistances
change with the signal frequency, such devices are likely to shortcircuit the higher frequency HART signals (1200–2200 Hz).
Where a HART communication system is connected with
other kinds of communication systems, gateways could be the
best interface devices to convert the HART protocol into the
Zhang_Ch03.indd 395
5/13/2008 5:41:42 PM
396
INDUSTRIAL CONTROL TECHNOLOGY
Host
Safe
area
PC
Hazardous area
3780-1
FSK isolating
amplifier (Ex-i)
Controller
FSK bus
3780-1
3780-1
•
• Up to max.
• 100 control loops
•
Figure 3.42 The connection architecture of an HART network with the FSK-bus
(courtesy of the SAMSON, Inc.).
protocols of the networks to be coupled. In most cases, when
complex communications must be performed, Fieldbus systems
would be the preferred choice. Even though there is no complex
protocol conversion, the HART-enabled system is capable of
communicating over long distances. Furthermore, the HART
data signals can be transmitted over telephone lines using HART/
CCITT converters, in which the Field devices directly connect to
dedicated lines owned by the telephone company, being able to
communicate with the centralized host located many kilometers
away. However, as already mentioned, within a HART-enabled
system, the HART-compatible field devices also require an
appropriate communication interface that could be, for example,
an integrated FSK-modem or a HART-Multiplexer.
As mentioned earlier, the HART signals are imposed on the
conventional analog current signal. Whether the devices in
the networks are designed in four-wire technique including an
Zhang_Ch03.indd 396
5/13/2008 5:41:42 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
397
additional power supply or in two-wire technique, HART communication can be used for both cases. However, it is important
to note that the maximum permissible load of a HART device is
fixed. The load of a HART device is limited by the HART specification. Another limitation is due to the process controller. The
output of the process controller must be able to provide the power
for the connected two-wire device.
The higher the power consumption of a two-wire device is,
the higher its load is. The additional functions of a HARTcommunicating device increase its power consumption, and
hence the load, compared to non-HART devices. When retrofitting HART devices into an already existing installation, the process controller must be checked for its ability to provide the
power required by the HART-compatible device. The process
controller must be able to provide at least the load impedance of
the HART device at 20 mA.
3.4.2.2
HART System Installation
The first task before installing a HART-enabled system is checking to
verify the HART-compatible devices. Manufacturers have different levels
of HART technical implementation in their devices and systems. In fact
the capabilities of the HART-compatible devices and HART-enabled system
vary widely, which requires when that engineers, when specifying HART
technology, consider such factors and parameters as the following:
(1) Registered device at the HART Communication Foundation
(the HCF).
(2) Registered Device-Description at the HART Communication
Foundation (the HCF).
(3) What is the number of variables this device can measure?
(4) Does this device comply with HART specification (in reference
of the HCF)?
(5) Does the device respond to HART Command 48 (in reference of
the HCF’ specifications)?
(6) What unique or special features does this device support?
(7) What diagnostic features does the device contain?
These questions below are for the suppliers of those I/O interface devices:
(1) How much HART capability is embedded into the I/O and how
smart is it?
(2) Can the I/O validate and secure the 4–20 mA signal?
Zhang_Ch03.indd 397
5/13/2008 5:41:42 PM
398
INDUSTRIAL CONTROL TECHNOLOGY
(3) Is there one HART modem per channel, or is the I/O multiplexed? How fast can it update the HART digital values?
(4) In what ways does the system support access to multivariable
HART data from multivariable devices?
(5) Can you merely “push a button” on the I/O to calibrate the
network current or check the range?
(6) Does the I/O support multiple-dropped networks?
(7) Does the I/O automatically scan and monitor the HARTcompatible field devices or is the scanning only possible using
“pass through”?
These questions below should be asked of the control system suppliers:
(1) Does your host use a “native” device description or does it
require a different file type?
(2) Does the system make it easy to use all HART capabilities?
(3) How much training is required to learn how to get and use
HART data?
(4) Review the configuration of a HART-compatible device using
the control system.
(5) Can the system use secondary digital process variables?
(6) Does it understand the HART-compatible device status change?
(7) Can the system detect configuration changes?
(8) Does the system do notification by exception?
(9) How does the system detect changes in configuration and status?
(10) How is the HART-compatible device status communicated to
the operators?
(11) How do you perform tests when there is an error in the device?
(12) How open is the system to third party software?
Before installation, it is also necessary to enter device tags and other
identification and configuration data into each field instrument. After
installation, the instrument identification (tag and descriptor) can be verified in the control room using a configuration tool such as hand-held
communicator or computer. Some field devices provide information on
their physical configuration (e.g., wetted materials). These and other configuration data can also be verified in the control room. The verification
process is important for safety.
Once a field instrument has been identified and its configuration data
confirmed, the analog network integrity can be checked using the network
test feature, which is supported by many HART-compatible devices. The
network test feature enables the analog signal from a HART transmitter to
be fixed at a specific value to verify network integrity and ensure proper
Zhang_Ch03.indd 398
5/13/2008 5:41:42 PM
399
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
connection to support devices such as indicators, recorders, and DCS displays. Use the HART protocol network test feature to check analog network integrity and ensure a proper physical connection among all network
devices. Additional integrity can be achieved if the analog value is compared to the digital value being reported in a device. For example, someone might have provided an offset to the 4–20 mA analog value that has
not been accounted for in the control system. By comparing the digital
value of the Primary Variable to the analog value, the network integrity can
be verified.
There are some ways to integrate HART data and leverage the intelligence in smart field devices. Several simple and cost-effective integration
strategies are listed below in order to get more from currently installed
HART-compatible devices and instruments (Fig. 3.43).
(1) Point-to-point integration. This is the most common way to use
HART. The communication capability of HART-compatible
devices allows them to be configured and set up for specific
applications, reducing costs and saving time in commissioning
and maintenance. With connection to the 4–20 mA wires, a
device can be integrated from remote locations by connecting
anywhere on the current network to obtain device status and
diagnostic information.
(2) HART-to-analog integration. Signal extractors communicate
with HART-compatible devices in real-time (simultaneously) to
convert the intelligent information in these devices into 4–20 mA
signals for input into an existing analog control system. Add this
Operator’s
computer
H
A
R
HART system/network controller
T
d
HART I/O interface devices
a
t
a
Filed
instruments
Figure 3.43 The dataflow diagram for the integration of the HART data.
Zhang_Ch03.indd 399
5/13/2008 5:41:43 PM
400
INDUSTRIAL CONTROL TECHNOLOGY
capability one device at a time to get more from the intelligent
HART-compatible devices.
(3) HART-plus-analog integration. New HART-multiplexer packaging solutions make it easy to communicate with HART-compatible
devices by replacing the existing I/O termination panels. The
analog control signal continues on to the control system as it
does today but the HART data is sent to a device asset management system providing valuable diagnostics information 24/7.
Although the control system is not aware of the HART data, this
solution provides better access to device diagnostics for asset
management improvements.
(4) Full HART integration. Upgrading a Field or Remote I/O system
provides an integrated path to continuously put HART data
directly into your control system. Most new control systems are
HART-capable and many suppliers offer software and I/O solutions to make upgrades simple and cost-effective. Continuous
communication between the field device and control system
enables problems with the devices, its connection to the process,
or inaccuracies in the 4–20 mA control signal to be detected
automatically so that corrective action can be taken before there
is negative impact to the process operation.
3.4.2.3
HART System Configuration
The purpose of the Device Configuration is for accessing its HART
Data. There are several methods of accessing the intelligent information in
the HART-compatible device on a temporary or a permanent basis. The
configuration of a HART-compatible device can be achieved by using the
software and hardware tools.
To configure a single device on a temporary basis, a universal hand-held
configuration tool is needed, with a power supply, a load resister, and a
HART-compatible device. Or, configuration can be achieved by using a
computer which is capable of running a device configuration application
and using a HART-modem.
(1) Universal hand-held communicators. HART hand-held communicators are available from major instrumentation suppliers across
the world and are supported by the member companies of the
HCF. Using Device Description (DD) files, the communicator
can fully configure any HART-compatible device for which it has
a DD installed. If the communicator does not have the DD for a
specific device, it will still communicate and configure the device
using the HART Universal and Common Practice commands.
Zhang_Ch03.indd 400
5/13/2008 5:41:43 PM
401
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
There are 35–40 standard data items in every registered HARTcompatible device. The data can be accessed by any approved
configuration tool such as a communicator. These items do not
require the use of a Device Description and typically include the
basic functionality for all devices. These are the Universal
and Common Practice commands required of every registered
HART device. To access the device specific data, a current Device
Description is required and provides the communicator with
the information it needs to fully access all the device specific
capabilities.
A HART hand-held communicator, if equipped, can also facilitate record keeping of device configurations. After a device is
installed, its configuration data can be stored in memory or on a
disk for later archiving or printing. There are many types of
hand-held communicators available today; their features and
ability should be compared to find those that meet your specific
requirements.
(2) Computer-based device configuration and management tools.
A HART-compatible device can be configured with a desktop
or laptop computer (or other portable models) by using a
computer-based software application and a HART interface
modem (Fig. 3.44). The advantages of using a computer include
an improved screen display and support for more Device
Descriptions and Device Configurations due to additional
computer-based memory storage capacity. Due to the critical
nature of device configurations in the plant environment, these
Handheld teminal
PC/host application
RS232 or USB HART
interface
250 Ω
resistor
Power supply
Field device
Figure 3.44 The connection of a computer with an HART-compatible device for
configuration (courtesy of the HCF).
Zhang_Ch03.indd 401
5/13/2008 5:41:43 PM
402
INDUSTRIAL CONTROL TECHNOLOGY
computers can also be used as backup storage for data from
hand-held communicators.
Software applications are available from many suppliers. It is
important to review their features to determine ease of use, ability to add or download the Device Descriptions, and general
functionality.
3.4.2.4
HART System Calibration
In order to take advantage of the digital capabilities of HART-compatible
devices and instruments, especially for precisely reporting the data values
of process control, it is essential that these devices and instruments should
be calibrated correctly. Like the calibration procedure of other devices, a
calibration procedure for HART-compatible devices and instruments consists of a verification test, an adjustment to within acceptable tolerance if
necessary, and a final verification test if an adjustment has been made.
Furthermore, data from the calibration is collected and used to complete a
report of calibration, documenting instrument performance over time.
(1) Functional parts of HART devices. For a HART-compatible
device, a multiple-point test between input and output does not
provide an accurate representation of its operation. Just like a
conventional device, the measurement process begins with a
technology that converts a physical quantity into an electrical
signal. However, the similarity to a conventional device ends
here. Instead of a purely mechanical or electrical path between
the input and the resulting 4–20 mA output signal, a HARTcompatible device has a microprocessor that manipulates the
input data. As shown in Fig. 3.45, there are typically three calculation sections involved, and each of these sections may be individually tested and adjusted in the calibration procedure.
Prior to the first box in Fig. 3.45, the microprocessor of this
device measures some electrical property that is affected by the
process variable of interest. The measured value may be voltage,
capacitance, reluctance, inductance, frequency, or some other
property. However, before it can be used by the microprocessor,
it must be transformed to a digital count by an analog to digital
(A/D) converter.
In the first box, the microprocessor of this device must rely
upon some form of equation or table to relate the raw count value
of the electrical measurement to the actual property (PV) of
interest such as temperature, pressure, or flow. The principal
form of this table is usually established by the manufacturer, but
Zhang_Ch03.indd 402
5/13/2008 5:41:44 PM
403
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
Counts
Input
section
PV may be
read digitally
High and low
output trim
mA
D/A
counts
Counts
PV
PV
A/D
counts
Range and
transfer function
mA
High and low
sensor trim
PV
mA
Conversion
section
Output
section
mA may be set
and read digitally
Figure 3.45 A functional block diagram of HART-compatible devices.
most HART-compatible devices and instruments include commands to perform field adjustments. This is often referred to as a
sensor trim. The output of the first box is a digital representation
of the process variable. When engineers read the process variable
using a communicator, this is the value that they can see.
The second box in Fig. 3.45 is strictly a mathematical conversion from the process variable to the equivalent milliamp representation. The range values of the instrument (related to the zero
and span values) are used in conjunction with the transfer function to calculate this value. Although a linear transfer function is
the most common, pressure transmitters often have a square-root
option. Other special instruments may implement common mathematical transformations or user defined break point tables. The
output of the second block is a digital representation of the
desired instrument output. When engineers read the network current using a HART-communicator, this is the value that they see.
Many HART-compatible instruments support a command which
puts the instrument into a fixed output test mode. This overrides
the normal output of the second block and substitutes a specified
output value.
The third box in Fig. 3.45 is the output section where the calculated output value is converted to a count value that can be
loaded into a digital to analog converter. This produces the actual
analog electrical signal. Once again the microprocessor must
rely on some internal calibration factors to get the output correct.
Adjusting these factors is often referred to as a current loop trim
or 4–20 mA trim.
(2) Basic steps of HART calibration. This analysis in (1) above tells
us why a proper calibration procedure for a HART-compatible
Zhang_Ch03.indd 403
5/13/2008 5:41:44 PM
404
INDUSTRIAL CONTROL TECHNOLOGY
instrument is significantly different from that for a conventional
instrument. The specific calibration requirements depend upon
the application. If the application uses the digital representation
of the process variable for monitoring or control, then the sensor
input section (the first box in Fig. 3.45) must be explicitly tested
and adjusted. Please note that this reading is completely independent of the milliamp output, and has nothing to do with the
zero or span settings. The PV as read with HART communication continues to be accurate even when it is outside the assigned
output range.
If the current network output is not used (i.e., the instrument is
used as a digital only device), then the input section calibration
is all that is required. If the application uses the milliamp output,
then the output section must be explicitly tested and calibrated.
Please note that this calibration is independent of the input section,
and again, has nothing to do with the zero and span settings.
The same basic multiple point test and adjust technique are
employed, but with a new definition for output. To run a test, use
a calibrator to measure the applied input, but read the associated
output (PV) with a communicator. Error calculations are simpler
because there is always a linear relationship between the input
and output, and both are recorded in the same engineering
units. In general, the desired accuracy for this test will be the
manufacturer’s accuracy specification. If the test does not pass,
then follow the procedure recommended by the manufacturer for
trimming the input section. This may be called a sensor trim and
typically involves one or two trim points. Pressure transmitters
also often have a zero trim, where the input calculation is adjusted
to read exactly zero (not low range). Do not confuse a trim with
any form of reranging or any procedure that involves using zero
and span buttons.
The same basic multiple point test and adjust technique is
employed again, but with a new definition for input. To run a test,
use a communicator to put the transmitter into a fixed current
output mode. The input value for the test is the mA value. The
output value is obtained using a calibrator to measure the resulting current. This test also implies a linear relationship between
the input and output, and both are recorded in the same engineering units (milliamps). The desired accuracy for this test should
also reflect the manufacturer’s accuracy specification. If the test
does not pass, then follow the procedure recommended by the
manufacturer for trimming the output section. This may be called
a 4–20 mA trim, a current loop trim, or a D/A trim. The trim
procedure should require two trim points close to or just outside
Zhang_Ch03.indd 404
5/13/2008 5:41:44 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
405
of 4 and 20 mA. Do not confuse this with any form of reranging
or any procedure that involves using zero and span buttons.
After calibrating both the Input and Output sections, a HARTcompatible device or instrument should operate correctly. The
middle block in Fig. 3.45 only involves computations. That is
why the range, units, and transfer function can be changed without necessarily affecting the calibration. Notice also that even if
the instrument has an unusual transfer function, it only operates
in the conversion of the input value to a milliamp output value,
and therefore is not involved in the testing or calibration of either
the input or output sections.
(3) Performance verification of HART calibration. If the goal of this
calibration is to validate the overall performance of a HARTcompatible device or instrument, it needs just to run a zero and
span test like that applied to a conventional instrument. However,
passing this test does not definitely indicate that the transmitter
is operating correctly, which is due to the following reasons.
Many HART-compatible instruments support a parameter
called damping. If this is not set to zero, it can have an adverse
effect on tests and adjustments. Damping induces a delay between
a change in the instrument input and the detection of that change
in the digital value for the instrument input reading and the corresponding instrument output value. This damping induced delay
may exceed the settling time used in the test or calibration. The
settling time is the amount of time the test or calibration waits
between setting the input and reading the resulting output. It is
advisable to adjust the instrument damping value to zero prior to
performing tests or adjustments. After calibration, be sure the
damping constant is returned to its required value.
There is a common misconception that changing the range of
a HART-compatible instrument by using a communicator somehow calibrates the instrument. Remember that a true calibration
requires a reference standard, usually in the form of one or more
pieces of calibration equipment to provide an input and measure
the resulting output. Therefore, since a range change does not
reference any external calibration standards, it is really a configuration change, not a calibration. Please note that in the block
diagram of HART-compatible devices (Fig. 3.45), changing the
range only affects the second box. It has no effect on the digital
process variable as read by a communicator.
Using only the zero and span adjustments to calibrate a HARTcompatible instrument (the standard practice associated with conventional instruments) often corrupts the internal digital readings.
As shown in Fig. 3.45, there is more than one output to consider.
Zhang_Ch03.indd 405
5/13/2008 5:41:44 PM
406
INDUSTRIAL CONTROL TECHNOLOGY
The digital PV and milliamp values read by a communicator are
also outputs, just like the analog current network.
The proper way to correct a zero drift condition is to use a zero
trim. This adjusts the instrument input block so that the digital
PV agrees with the calibration standard. If intending to use the
digital process values for trending, statistical calculations, or
maintenance tracking, then it should disable the external zero
and span buttons and avoid using them entirely.
3.4.3
HART Protocol
HART protocol is widely recognized as the industry standard for digitally enhanced 4–20 mA field instrument communication in process control. In industrial process control, the HART protocol provides a uniquely
backward compatible solution for filed instrument communication as both
4–20 mA analog and digital signals are transmitted simultaneously on the
same wiring.
3.4.3.1
HART Protocol Model
HART communication uses a master–slave protocol which means that a
field device as slave speaks only when it is spoken to by a device as master.
In every communication, a master device sends a “command” message
first; while receiving this command message, the slave device processes it
and then sends back a “response” message to that master device (Fig. 3.46).
Both the “command” and “response” messages include the HART-data
and must be formatted in accordance with the relevant HCF (HART
Communication Foundation) specifications.
Master
Slave
Command
Indication
Request
Time out
Response
Response
Confirmation
Figure 3.46 The HART master–slave protocol model.
Zhang_Ch03.indd 406
5/13/2008 5:41:44 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
407
The HART protocol can be used in various modes for communicating
information to and from smart field instruments and central control or
monitoring equipment, which includes analog plus digital signals, and
digital only signals. Digital master–slave communication simultaneous
with the 4–20 mA analog signals is the most common. This mode, depicted
in Fig. 3.46, allows digital information from the slave device to be updated
twice per second in the master. The 4–20 mA analog signals are continuous and can still carry the primary variable for control. Please note that
when a communication between the master and the slave is engaging,
“interrupt” is definitely not allowed.
“Burst” is an optional communication mode (Fig. 3.46) which allows a
single slave device to continuously broadcast a standard HART response
message. This mode frees the master from having to send repeated command requests to get updated process variable information. The same
HART response message (PV or other, see Fig. 3.45) is continuously
broadcast by the slave until the master instructs the slave to do otherwise.
Data update rates of 3–4 per second are typical with “burst” mode communication and will vary with the chosen command. Please note that the
“Burst” mode should be used only in single slave device networks.
Two masters (primary and secondary) can communicate with slave
devices in a HART network. Secondary masters, such as hand-held communicators, can be connected almost anywhere on the network and communicate with field devices without disturbing communication with the
primary master. A primary master is typically a PLC, or computer based
central control or monitoring system. A typical installation with two masters is shown in Fig. 3.33 and Fig. 3.44. From an installation perspective,
the same wiring used for conventional 4–20 mA analog instruments carries the HART communication signals. Allowable cable run lengths will
vary with the type of cable and the devices connected, but in general up to
3000 m for a single twisted pair cable with shield and 1500 m for multiple
twisted pair cables with a common shield. Unshielded cables can be used
for short distances. Intrinsic safety barriers and isolators which pass the
HART signals are readily available for use in hazardous areas.
The HART protocol also has the capability to connect multiple field
devices on the same pair of wires in a multiple-dropped network configuration as shown in Fig. 3.33(b). In multiple-dropped networks, communication is limited to master–slave digital only. The current through each
slave device is fixed at a minimum value to power the device (typically
4 mA) and no longer has any meaning relative to the process.
The HART protocol utilizes the OSI 7-layer reference model. As is the
case for most of the communication systems on the field level, the HART
Zhang_Ch03.indd 407
5/13/2008 5:41:44 PM
408
INDUSTRIAL CONTROL TECHNOLOGY
protocol implements only the layers 1, 2, and 7 of the OSI model. The
layers 3–6 remain empty since their services are either not required or
provided by the Application Layer 7 (see Fig. 3.47).
The Application Layer defines the commands, responses, data types,
and status reporting supported by the protocol. In addition, there are certain conventions in HART (e.g., how to trim the network current) that are
also considered part of the Application Layer. While the Command
Summary, Common Tables, and Command Response Code Specifications
all establish mandatory Application Layer practices (including data types,
common definitions of data items, and procedures), the Universal
Commands specify the minimum Application Layer content of all HARTcompatible devices.
OSI layer
Function
HART layer
Application
layer
Provides the user with
network capable
applications
Command oriented. Predefined data types
and application procedures
Presentation
layer
Converts application
data between network
and local machine
formats
Connection
management service
for applications
Session
layer
Transport
layer
Provides network
Autosegmented transfer of large datasets,
independent, transport reliable stream transport, negotiated
message transfer
segment sizes.
Network
layer
End to end routing of
packets. Resolving
network addresses
Data Link
layer
Establishes data
packet structure,
framing, error
detection, bus
arbitration.
Mechanical/electrical
connection. Transmits
raw bit stream
Physical
layer
A binary, byteoriented, token
passing, master–
slave protocol
Simultaneous
analog and digital
signaling, normal
4–20 mA copper
wiring.
Wired HART
Power-optimized,
redundant path, selfhealing wireless
mesh network
Secure and reliable,
time synched
TDMA/CSMA,
frequency agile with
ARQ
2.4 Hz wireless,
802.15.4 based
radios, 10 dBm T ×
power
Wireless HART
Figure 3.47 HART protocol implementing the OSI 7-layer model.
Zhang_Ch03.indd 408
5/13/2008 5:41:44 PM
409
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
3.4.3.2
HART Protocol Commands
In the communication routines of the application layer of the HART
protocol, the master devices and operating programs are based on HART
commands to give instructions or send messages plus data to a field device.
Once receiving the command message, the field devices immediately
process it and then respond by sending back a response message which
can contain requested status reports and/or the data of the field device.
Table 3.9 provides the classes of the HART commands, and Table 3.10
gives a summary of the HART commands.
Figure 3.48 is the standard format of both the HART command and
response messages. In accordance with the HART command specification,
this format includes the following:
(1) First, the preamble, of between 5 and 20 bytes of hex FF (all 1s),
helps the receiver to synchronize to the character stream.
Table 3.9 HART Commands Classes
Universal
Commands
All devices using the
HART protocol must
recognize and
support the universal
commands.
Universal commands
provide access to
information useful in
normal operations.
For example, read
primary variable and
units, read
manufacturer and
device type, read
current output and
percentage of range,
and read sensor
serial number and
limits
Zhang_Ch03.indd 409
Command Practice
Commands
Device-Specific
Commands
Common practice
commands provide
functions implemented
by many, but not
necessarily all, HART
communication devices.
The HART
specifications
recommend devices to
support these
commands when
applicable. Examples of
common practice
commands are read a
selection of up to four
dynamic variables,
write damping time
constant, write
transmitter range, set
fixed output current and
perform self-test
Device-specific
commands represent
functions that are unique
to each field device.
These commands access
setup and calibration
information as well as
information about the
construction of the
device. Information on
device-specific
commands is available
from device
manufacturers or in the
Field Device
Specification document.
Examples of devicespecific commands are
read or write sensor type;
start, stop, or clear
totalizer; read or write
alarm relay set point; etc.
5/13/2008 5:41:45 PM
410
INDUSTRIAL CONTROL TECHNOLOGY
Table 3.10 HART Commands Summary
Universal Commands
Command Practice
Commands
Device-Specific
Commands
(Example)
Read manufacturer and
device type
Read primary variable
(PV) and units
Read current output and
percentage of range
Read up to four predefined dynamic
variables
Read or write 8-character
tag, 16-character
descriptor, date
Read or write 32character message
Read device range values, units, and damping
time constant
Read or write final
assembly number
Write polling address
Read selection of up to
four dynamic variables
Write damping time
constant
Write device range
values
Calibrate (set zero, set
span)
Set fixed output current
Perform self-test
Perform master reset
Trim PV zero
Write PV unit
Trim DAC zero and
gain
Write transfer function
(square root/linear)
Write sensor serial
number
Read or write dynamic
variable assignments
Read or write lowflow cut-off
Start, stop, or clear
totalizer
Read or write density
calibration factor
Choose PV (mass,
flow, or density)
Read or write materials or construction
information
Trim sensor
calibration
PID enable
Write PID set point
Valve characterization
Valve set point
Travel limits
User units
Local display
information
PREAMBLE START ADDR COMM BCNT
[STATUS]
[DATA]
CHK
Preamble: 5 to 20 bytes, hex FF
Start character: 1 byte
Addresses: source and destination, 1 or 5 bytes
Command: 1 byte
Byte count (of status and data): 1 byte
Status: 2 bytes, only in slave response
Data: 0 to 25 bytes*
Checksum: 1 byte
*25 bytes is a recommended maximum data length
The maximum number of data bytes is not defined by the protocol specifications.
Figure 3.48 The standard format of the HART protocol command and response
frames.
Zhang_Ch03.indd 410
5/13/2008 5:41:45 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
411
(2) The start character may have one of several values, indicating
the type of message: master to slave, slave to master, or burst
message from slave; also the address format: short frame or long
frame.
(3) The address field includes both the master address (a single bit:
1 for a primary master, 0 for a secondary master) and the slave
address. In the short frame format, the slave address is 4 bits
containing the “polling address” (0–15). In the long frame format, it is 38 bits containing a “unique identifier” for that particular device. (One bit is also used to indicate if a slave is in burst
mode.)
(4) The command byte contains the HART command for this
message. Universal commands are in the range 0–30; commonpractice commands are in the range 32–126; device-specific
commands are in the range from 128 to 253.
(5) The byte count byte contains the number of bytes to follow in the
status and data bytes. The receiver uses this to know when the
message is complete. (There is no special “end of message”
character.)
(6) The status field (also known as the “response code”) is two bytes,
only present in the response message from a slave. It contains
information about communication errors in the outgoing message, the status of the received command, and the status of the
device itself.
(7) The data field may or may not be present, depending on the particular command. A maximum length of 25 bytes is recommended, to keep the overall message duration reasonable. (But
some devices have device-specific commands using longer data
fields.) See also the HART data field.
(8) Finally, the checksum byte contains an “exclusive-OR” or “longitudinal parity” of all previous bytes (from the start character
onward). Together with the parity bit attached to each byte, this
is used to detect communication errors.
3.4.3.3
HART Protocol Data
(1) HART data. There are several types of data or information that
can be communicated from a HART-compatible device. This
includes:
(a) Device data
(b) Supplier data
(c) Measurement data
(d) Calibration data.
Zhang_Ch03.indd 411
5/13/2008 5:41:45 PM
412
INDUSTRIAL CONTROL TECHNOLOGY
The following is a summary of these data items available for
communication between HART-compatible devices and a Host.
(a) Process variable values.
(i) Primary process variable (analog): 4–20 mA current
signals continuously transmitted to host.
(ii) Primary process variable (digital): Digital value in
engineering units, IEEE floating point, up to 24-bit
resolution.
(iii) Percent range: Primary process variable expressed as
percent of calibrated range.
(iv) Loop current: Loop current value in milliamps.
(v) Secondary process variable 1: Digital value in engineering units available from multivariable devices.
(vi) Secondary process variable 2: Digital value in
engineering units available from multivariable devices
(vii) Secondary process variable 3: Digital value in engineering units available from multivariable devices
(b) Commands from host to device.
(i) Set primary variable units
(ii) Set upper range
(iii) Set lower range
(iv) Set damping value
(v) Set message
(vi) Set tag
(vii) Set date
(viii) Set descriptor
(ix) Perform loop test: Force loop current to specific value
(x) Initiate self-test: Start device self-test
(xi) Get more status available information.
(c) Status and diagnostic alerts.
(i) Device malfunction: Indicates device self-diagnostic
has detected a problem in device operation.
(ii) Configuration changed: Indicates device configuration has been changed.
(iii) Cold start: Indicates device has gone through power
cycle.
(iv) More status available: Indicates additional device status data available.
(v) Primary variable analog output fixed: Indicates device
in fixed current mode.
(vi) Primary variable analog output saturated: Indicates
4–20 mA signal is saturated.
(vii) Secondary variable out of limits: Indicates secondary
variable value outside the sensor limits.
Zhang_Ch03.indd 412
5/13/2008 5:41:45 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
413
(viii) Primary variable out of limits: Indicates primary variable value outside the sensor limits.
(d) Device identification.
(i) Instrument tag: User defined, up to 8 characters.
(ii) Descriptor: User defined, up to 16 characters.
(iii) Manufacturer name (code): Code established by HCF
and set by manufacturer.
(iv) Device type and revision: Set by manufacturer.
(v) Device serial number: Set by manufacturer.
(vi) Sensor serial number: Set by manufacturer.
(e) Calibration information for 4–20 mA transmission of primary
process variable.
(i) Date: Date of last calibration, set by user.
(ii) Upper range value: Primary variable value in engineering units for 20 mA point that is set by user.
(iii) Lower range value: Primary variable value in engineering units for 4 mA point that is set by user.
(iv) Upper sensor limit: Set by manufacturer.
(v) Lower sensor limit: Set by manufacturer.
(vi) Sensor minimum span: Set by manufacturer.
(vii) PV damping: Primary process variable damping factor,
set by user.
(viii) Message: Scratch pad message area (32 characters),
set by user.
(ix) Loop current transfer function: Relationship between
primary variable digital value and 4–20 mA current
signal.
(x) Loop current alarm action: Loop current action on
device failure (upscale/downscale).
(xi) Write protect status: Device write-protect indicator.
(2) DDL device description. The HART commands in the application layer are based on the services of the lower layers and enable
an open communication between the master and the field devices.
All the HART-compatible devices, no matter their manufacturers, are capable for this openness and intercommunications as
long as the field devices operate exclusively with the universal
and common-practice commands. In a HART-enabled system,
the user does not need more than the simple HART standard
notation for the status and fault messages.
When the user wants the message to contain further devicerelated information or that special properties of a field device are
also used, the common-practice and universal commands are not
sufficient. Using and interpreting the data requires that the user
know their meaning. However, this knowledge is not available in
Zhang_Ch03.indd 413
5/13/2008 5:41:45 PM
414
INDUSTRIAL CONTROL TECHNOLOGY
further extending systems which can integrate new components
with additional options. To eliminate the adaptation of the master
device’s software whenever an additional status message is
included or a new component is installed, the device description
language (DDL) was accordingly developed.
The DDL is not limited to the HART applications. It was
developed and specified for all the Fieldbus, independent of the
HART protocol, by the Human–Machine Interface workshop of
the International Fieldbus Group (IFG).
The developers of the device description language (DDL)
aimed at achieving versatile usability. The DDL also finds use
in field networks. The required flexibility is ensured insofar as
the DDL does not itself determine the number and functions of
the device interfaces and their representation in the control
stations.
The DDL is simply a language, similar to a programming language, which enables the device manufacturers to describe all
communication options in an exact and complete manner. The
DDL allows the manufacturer to describe
(a) attributes and additional information on communication data
elements,
(b) all operating states of the device,
(c) all device commands and parameters,
(d) the menu structure, thus providing a clear representation of
all operating and functional features of the device.
Having the device description of a field device and being
able to interpret it, a master device is equipped with all necessary information to make use of the complete performance
features of the field device.
Device-specific and manufacturer-specific commands can
also be executed and the user is provided with a universally
applicable and uniform user interface, enabling him or her to
clearly represent and perform all device functions. Thanks to
this additional information, clear, exact and, hence, safer
operation and monitoring of a process is made possible.
The master device does not read the device description
as readable text in DDL syntax, but as short, binary-coded
Device Description data record specially generated by the
DDL encoder (or DDL compiler). For devices with sufficient storage capacity, this short form opens up the possibility
of storing the device description already in the firmware of
the field device. During the parameterization phase, it can be
read by the corresponding master device.
Zhang_Ch03.indd 414
5/13/2008 5:41:45 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
415
3.4.4 HART Integration
3.4.4.1 Basic Industrial Field Networks
There are many different field networks available in industrial control
for instrument engineers today. Understanding these networks’ classifications should allow us to choose the right tools for the control applications.
Some common industrial communication protocols will be introduced
below by focusing on the native applications and placing them in three
basic categories:
(1) Sensor networks are those protocols initially designed to support
discrete I/O.
(2) Device networks are those protocols originally focused on process instrumentation.
(3) Control networks are those protocols typically used to connect
controllers and I/O systems.
(1) Sensor networks. Sensor level protocols, with a principal focus
on supporting the digital communications for those discrete sensors and actuators. Sensor level protocols tend to have very fast
cycle times and, since they are often promoted as an alternative
to PLC discrete I/O, the cost of a network node should be relatively low. These protocols listed below are the simplest forms of
sensor networks available today.
(a) AS-i. AS-i acts as a network-based replacement for a discrete
I/O card. Consequently, AS-i offers perhaps the simplest network around consisting of up to 31 slave devices with 248
I/O bits and the following functionality: (1) the master polls
each slave, (2) the master message contains four output
bits, (3) the slave answers immediately with four input bits,
4) diagnostics are included in each message, (5) a worst-case
scan time of less than 5 ms.
(b) CAN. The Controller Area Network (CAN) defines only
basic, low level signaling and medium access specifications
which are both simple and unique. Even though CAN
medium access is technically CSMA/CD, this classification
can provide simple, highly reliable, prioritized communication between intelligent devices, sensors, and actuators in
automotive applications. Of these advantages, reliability is
paramount. Network errors while driving a car on a busy
interstate highway are unacceptable. Today, CAN is used in a
vast number of vehicles and in a variety of other applications.
Zhang_Ch03.indd 415
5/13/2008 5:41:45 PM
416
INDUSTRIAL CONTROL TECHNOLOGY
As a result: (1) a large number of different chips and vendors
support CAN, 2) the total chip volume is huge, 3) the parts
cost is small.
(c) DeviceNet. DeviceNet is a well-established machine and
manufacturing automation network supported by a substantial number of products and vendors in the semiconductor
industry. DeviceNet specifies physical (connectors, network
terminators, power distribution, and wiring) and application
layer operation based on the CAN standard. While all the
popular application layer hierarchies are supported (client
and server, master and slave, peer-to-peer, publisher and
subscriber), in practice most devices support only master–
slave operation which results in significantly lower costs.
In comparison with the CAN, DeviceNet provides configurable structure to the operation of the network, allowing
for the selection of polled, cyclic, or event driven network
operation.
(d) Interbus. Interbus is a popular industrial network that uses a
ring topology in which each slave has an input and an output
connector. Interbus is one of the few protocols that are full
duplex-data transmitted and received at the same time.
Interbus communication is cyclical, efficient, fast, and deterministic (e.g., 4096 digital inputs and outputs scanned in
14 ms). Of the ring topology, Interbus commissioning offers
us these advantages: (1) Node addresses are not required
because the master can automatically identify the nodes on
the network. (2) Slaves provide identification information
that allows the master to determine the quantity of the data
provided by the slave. (3) Using this data, the master explicitly maps the data to and from the slave into the bit stream as
it shifts through the network.
Interbus works as a large network-based shift register in
which a bus cycle begins with the network master transmitting a bit stream. As the first slave receives the bits, they are
echoed passing the data on to the next slave in the ring. This
process is simultaneous with the data being shifted from the
master to the first slave. In turn, the data from the first slave
is then being shifted into the master.
(2) Device networks. Device network protocols support process
automation that is fundamentally continuous and analog, more
complex transmitters, and valve-actuators. Transmitters typically
include pressure, level, flow, and temperature. The valve-actuators
can include those intelligent controllers, motorized valves, and
pneumatic positioners. Three device networks are prevalent in
Zhang_Ch03.indd 416
5/13/2008 5:41:45 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
417
the process automation industry: Foundation Fieldbus H1, HART,
and PROFIBUS-PA.
(a) Foundation Fieldbus H1. In August 1996, the Foundation
Fieldbus released its H1 Specifications that focus on “the
network is the control system.” Of the fundamental differences from the controller and I/O approach used in traditional systems, Foundation Fieldbus specifies not only a
communication network but also control functions. As a
result, the purpose of the communications is to pass data to
facilitate proper operation of the distributed control application. Its success relies on synchronized cyclical communication and on a well-defined applications layer.
In the Foundation Fieldbus H1, communications occur
within framed intervals of fixed time duration (a fixed repetition rate) and are divided into two phases and scheduled
cyclical data exchange and acyclic (e.g., configuration and
diagnostics) communication. The communication is controlled by polling the network and thereby prompting the
process data to be placed on the bus. This is done by passing
a special token to grant the bus to the appropriate device,
resulting in the cyclic data being generated at regular intervals. In the Foundation Fieldbus H1 networks, the application layer defines function blocks, which include analog in,
analog out, transducer, and the blocks. The data on the network is the transfer from one function block to another in the
network-based control system. The data is complex and
includes the digital value, engineering units, and status on
the data (to indicate the PID is manual or the measured value
is suspect).
(b) HART. HART is unique among device networks because it is
fundamentally an analog communications protocol. All the
other protocols use digital signaling, while HART uses modulated communications because the “HART digital communications” modulate analog signals centered in a frequency
band separated from the 4–20 mA signaling. HART enhances
smart 4–20 mA field devices by providing two-way communication that is backward compatible with existing installations, which allows HART to support two communications
channels simultaneously: a one-way channel carrying a single
process value (the 4–20 mA signal) and a bidirectional channel
to communicate digital process values, status, and diagnostics.
Consequently, the HART protocol can be used in traditional
4–20 mA applications, and allows the benefits of digital
communication to be realized in existing plant installations.
Zhang_Ch03.indd 417
5/13/2008 5:41:45 PM
418
INDUSTRIAL CONTROL TECHNOLOGY
HART is a simple, easy-to-use protocol because of these
facts: (1) Two masters are supported using token passing
to provide bus arbitration. (2) Allows a field device to
publish process data (“burst mode”). (3) Cyclical process
data includes floating-point digital value, engineering units,
and status. (4) Operating procedures standardized (for current loop reranging, loop test, and transducer calibration).
(5) Standardized identification and diagnostics provided.
(c) PROFIBUS-PA. PROFIBUS-PA (process automation) was
introduced to extend PROFIBUS-DP (decentralized peripherals) in order to support process automation. PROFIBUS-PA,
which operates over the same H1 physical layer as Foundation
Fieldbus H1, is essentially an LAN for communication with
process instruments.
PROFIBUS-PA networks are fundamentally master–slave
model, so sophisticated bus arbitration is not necessary.
PROFIBUS-PA also defines profiles for common process
instruments, including both mandatory and optional propertied (data items). When a field device supports a profile some
configuration of the device should be possible without being
device-specific.
(3) Control networks. Control networks are focused on providing a
communication backbone that allows integration of controllers,
I/O, and subnetworks. Control networks stand at the crossroads
between the growing capabilities of industrial networks and the
penetration of enterprise networks into the control system. As
such, the control networks are able to move huge chunks of heterogeneous data and operate at high data rates. A brief overview
of these four control networks is given below; they are ControlNet,
Industrial Ethernet, Ethernet/IP, and PROFIBUS-DP.
(a) ControlNet. ControlNet was developed as a high performance network suitable for both manufacturing and process
automation. ControlNet uses Time Division Multiple Access
(TDMA) to control the access to the network, which means
a network cycle is assigned a fixed repetition rate. Within the
bus cycle, data items are assigned a fixed time division for
transmission by the corresponding device. Data objects are
placed on ControlNet within a designated time slot and at
precise levels. Once the data to be published is identified
along with its time slot, any device on the network can be
configured to use the data. In the second half of a bus cycle,
acyclic communications occur. ControlNet is more efficient
than polled or token passing protocols because its data transmission is very deterministic.
Zhang_Ch03.indd 418
5/13/2008 5:41:45 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
419
(b) Industrial Ethernet. Many industrial communication protocols specify mechanisms to embed their protocols in Ethernet.
Ethernet addresses only the lower layers of communications
networks, but does not address the meaning of the data it
transports. Even though Ethernet effectively communicates
many protocols simultaneously over the same wire, it provides no guarantees that the data can be exchanged between
different protocols.
In Industrial Ethernet, the protocol being adopted is TCP/
IP and not Ethernet at all. TCP/IP is using two approaches to
support the session, presentation, and application layers of
the corresponding industrial protocol. First, the industrial
protocol is simply encapsulated in the TCP/IP, allowing the
shortest development time for defining industrial protocol
transportation over TCP/IP. The second approach actually
maps the industrial protocol to TCP and UDP services. While
this strategy takes more time and effort to develop, it results
in a more complete implementation of the industrial protocol
on top of the TCP/IP. UDP is a connectionless, unreliable
communication service that works well for broadcasts to
multiple recipients and fast, low-level signaling. UDP is used
by several industrial protocols (for time synchronization).
TCP is a connection oriented data stream that can be mapped
to the data and I/O functions in some industrial protocols.
(c) Ethernet/IP. Ethernet/IP is a mapping of the “Control and
Information Protocol (CIP)” used in both ControlNet and
DeviceNet to TCP/IP (not Ethernet). While all the basic
functionality of ControlNet is supported, the hard real-time
determinism that ControlNet offers is not present. CIP is
being promoted as a common, object-oriented mechanism
for supporting both manufacturing and process automation
functions.
Ethernet/IP is a good contribution to the growing discussion of Industrial Ethernet. However, Ethernet/IP is a recent
development and is basically the application of an existing
industrial network’s application layer to the TCP/IP.
(d) PROFIBUS-DP. PROFIBUS-DP (Decentralized Peripherals)
is a master–slave protocol used primarily to access remote
I/O. Each node is polled cyclically updating the status and
data associated with the node. Operation is relatively simple
and fast. PROFIBUS-DP also supports Fieldbus Message
Specification (FMS), which is a more complex protocol for
demanding applications that includes support for multiple
masters and peer-to-peer communication.
Zhang_Ch03.indd 419
5/13/2008 5:41:46 PM
420
3.4.4.2
INDUSTRIAL CONTROL TECHNOLOGY
Choosing the Right Field Networks
Before choosing the best suitable network for an industrial control
application, it is necessary to review all the open network types available
today. By carefully reviewing each of these types of open communication
networks, their respective strengths and weaknesses, in particular some
important technical factors, ought to be understood. Please note that the
specified application will be far more than the technology used. The specific, measurable benefits must drive this selection process.
The first factor that should be considered is the cost of each network.
The second is the network connectivity in which the bottom line is that you
want your data. It is very important to remember that all communications
networks cause changes. Although the actual changes can be very difficult
to foretell accurately, the effect of these changes must be realistically
considered.
In some instances, we have to make a decision on choosing between the
HART and other types of field networks. Before making such a decision,
it is important to understand the differences between the HART and other
field networks. For example, if all the choices are constrained between the
HART and the Foundation Fieldbus, comparisons of these two types in
every category are necessary. Table 3.11 is a list of the differences in elemental technical features between the HART and the Foundation Fieldbus,
which should be a help in choosing between the HART and the Foundation
Fieldbus.
3.4.4.3
Integrating the HART with Other Field Networks
The integration of the HART network with the Foundation Fieldbus
network is taken as an example here to demonstrate the strategies and
techniques of integrating HART with other field networks. As mentioned
above, one fundamental difference between HART and Fieldbus devices is
that with Fieldbus devices, you can implement control strategies inside the
field devices themselves; however, even if you can implement the same
control strategies with HART devices, the execution of actual control
algorithms would go on in the control system computer or PLC. A welldesigned control system will allow an integrated control strategy to use
devices independent of communication protocol.
HART and Fieldbus devices have the capability of providing a wide
range of diagnostics data about the device’s safety. In fact, the diagnostic
capabilities of HART and Fieldbus devices are nearly the same. The types
of device diagnostics change widely, depending on the type of device.
Measurement transmitters will have diagnostics related to the status of the
Zhang_Ch03.indd 420
5/13/2008 5:41:46 PM
421
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
Table 3.11 A Comparison between HART and Foundation Fieldbus
Feature
Technology
acceptance
Power limitation
Advances in silicon
power consumption
same for HART/FF;
thus FF will always
have capability for
more functionality
Communication
performance
Transmitter
diagnostics
Advanced
diagnostics
HART
Foundation Fieldbus (FF)
Well proven as Large
Installed base
Will continue to be sold
as a replacement unit
Simple for technicians’
competence
35 mW to 4 mA available for the HART
signal
Cannot “mirror”
Fieldbus, however may
provide an 80%
solution
Proven as Growing
Installed base
Training required
New investment to occur
increasingly in the FF
100 bits/s
Additional burden on
host
Device only
Includes predictive
No knowledge of other
devices
Does not have the processing power
Push or poll
Polled for HART status
periodically
Status can be missed
Two-way communication to other
devices
Multiple-dropped
No
Use in safety instrumented systems
Very limited
Theory 15 devices: real
around 3 slow series
loop
All devices wired
individually
FF minimum power
requirement of 8 mA
No spec limit;
Ultimate FF segment
power budget
FF devices order of more
magnitude than that
powering an Event for IS
FF H1 communicates at
31250 bits/s
Device + other devices
For example, Statistical
Process Monitoring and
Machinery Health
monitoring
Events are latched/time
stamped in the device
Sent by the device
There is no chance of
missing field problems
with FF
Yes
True multiple-dropped:
physically 32 devices realistically 12−16
Devices 2007 offers the
holding back technology
(Continued)
Zhang_Ch03.indd 421
5/13/2008 5:41:46 PM
422
INDUSTRIAL CONTROL TECHNOLOGY
Table 3.11 (Continued)
Feature
HART
Foundation Fieldbus (FF)
Control in the field
and advanced
applications
Does not support the
function block model
Multiple variables
In digital mode only
It is limited
No
Function block model
supports interoperable
control in the field where
blocks can reside in the
field device
Yes
Footprint and
hardware reduction
Future proof
devices—typical
upgrade capability
In the field
No
Full specifications
in the devices
No
Commissioning
speed
Hours for individually
wired devices
Renders obsolete all separate signal conditioners,
isolation amplifier cards,
output cards, CPU cards,
I/P converters, etc.
Ability to download new
version of firmware over
H1 link
None disconnected
device from the H1
segment
Communicate with this
devices conformant with
Fieldbus specification
Embedded at the factory
Travels with the instrument Upload directly to
“Smart Instrument”
software
Reduces commissioning
time Reduces time to
perform diagnostics
The networking capability
of FF allows the user to
commission a device in
10 s of seconds
transducer and measurement logic in the device. Control devices such as
valves will provide a lot of information about the mechanical condition of
the device. Both transmitters and valves will provide diagnostic information about the communication electronics in their respective devices.
Zhang_Ch03.indd 422
5/13/2008 5:41:46 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
423
Although the diagnostic data provided by HART and Fieldbus is very
similar, the way they get to the control system and the way they get to the
operator or technician to see it can be quite different. This has to do with
the speed and characteristics of the communication technology used by
these two protocols. Fieldbus basically uses point-to-point communication
technology. This means when a Fieldbus device detects a diagnostics condition it wants to report, it can send an event out on the bus with the related
information. The control system picks up the event and immediately displays or annunciates it on the console. HART devices, on the other hand,
have to continually undergo polling to see if there is anything to report.
Because the polling occurs at 1200 bps with HART, there are limitations
on how many devices it can poll for alters in a specific time frame. An
operator can poll a small number of critical devices for alters within
seconds or a large number of devices within minutes. However, it is possible to implement an effective diagnostic alert system with HART as long
as you understand the restrictions on response and device count.
Once the operator or maintenance engineer is aware of a problem in a
filed device either through an alert or some other means, the actual display
of the status information from HART and Fieldbus devices is very similar.
Usually, a record of this status event will automatically log into the control
system. The logging of HART and Fieldbus device problems should normally look the same on a well-integrated system.
The type of portable maintenance tools required in systems of HART
and Fieldbus devices is also an important factor for consideration. Portable
tools currently fall into two general categories. The first is intrinsically
safe hand-held devices. Several are available for HART only devices. A
combined HART and Fieldbus intrinsically hand-held device has become
available, too. This integrated tool allows the user to configure and diagnose HART and Fieldbus devices while in the field. The laptop computer
is the second type of portable tool. However, these types of computers cannot go in hazardous areas of a plant. As far as small hand-held computers
go, how practical these will be in a plant environment remains an open
question.
Bibliography
Alan R. Dewey in EMERSON. 2005. HART, Fieldbus Work Together in Integrated
Environment. http://www.emersonprocess.com/home/library/articles/protocol/
protocol0507_teamwork.pdf. Accessed date: October 2007.
Analog Services, Inc. (http://www.analogservices.com). 2006. HART Book. http://
www.analogservices.com/about_part0.htm. Accessed date: October 2007.
Zhang_Ch03.indd 423
5/13/2008 5:41:46 PM
424
INDUSTRIAL CONTROL TECHNOLOGY
AS-INTERFACE (http://www.as-interface.net). 2007a. AS-Interface System.
http://www.as-interface.net/System/. Accessed date: May.
AS-INTERFACE (http://www.as-interface.net). 2007b. AS-Interface Products.
http://www.as-interface.net/Products/. Accessed date: May.
CiA (http://www.can-cia.org). 2005a. Registered Free Download; CAN Physical
Layer Specification Version 2.0. http://www.can-cia.org/downloads/
ciaspecifications/?557. Accessed date: July.
CiA (http://www.can-cia.org). 2005b. Registered Free Download; CAN
Application Layer Specification Version 1.1. http://www.can-cia.org/
downloads/ciaspecifications/?1169. Accessed date: July.
CiA (http://www.can-cia.org). 2005c. Registered Free Download; CANopen
Specification Version 1.3. http://www.can-cia.org/downloads/ciaspeci
fications/?1136. Accessed date: July.
Commfront (http://www.commfront.com). 2007. RS232, 485,422,530 Buses.
http://www.commfront.com/CommFront-Home.htm. Accessed date: July.
Cyber (http://cyber.felk.cvut.cz). 2006. Supervisory Human Operation. http://
cyber.felk.cvut.cz/gerstner/biolab/bio_web/projects/iga2002/index.html.
Accessed date: October.
David Belohrad and Miroslav Kasal. 1999. FSK Modem with GALs. http://
www.isibrno.cz/~belohrad/radioelektronika99-fskmodem.pdf. Accessed date:
October 2007.
Degani, Asaf, Shafto, Michael, Kirlik, Alex. 2006. Modes in Human–Machine
Systems: Review, Classification, and Application. http://ic-www.arc.nasa
.gov/people/asaf/interface_design/pdf/Modes%20in%20Human-Machine
%20Systems.pdf. Accessed date: October.
Degani, Asaf. 2006. Modeling Human–Machine Systems: On Modes, Error, and
Patterns of Interaction. http://ase.arc.nasa.gov/people/asaf/hai/pdf/Degani_
Thesis.pdf. Accessed date: October.
ESD-Electronics (http://www.esd-electronics.com). 2005. Controller Area Network.
http://www.esd-electronics.com/german/PDF-file/CAN/Englisch/intro-e
.pdf. Accessed date: July.
Fieldbus (http://www.fieldbus.org). 2005a. FOUNDATION Technology. http://
www.fieldbus.org/index.php?option=com_content&task=view&id=45&Ite
mid=195. Accessed date: July.
Fieldbus (http://www.fieldbus.org). 2005b. Profibus Technology. http://www
.pepperl-fuchs.com/pa/interbtob/profibus/default_e.html. Accessed date: July.
Fieldbus Centre (http://www.knowthebus.org). 2005a. FOUNDATION Fieldbus.
http://www.knowthebus.org/fieldbus/foundation.asp. Accessed date: July.
Fuji Electric (http://web1.fujielectric.co.jp). 2007. Fuji AS-I Technologies.
http://www.fujielectric.co.jp/fcs/eng/as-interface/as_i/index.html. Accessed
date: May.
Grid Connect (http://www.industrialethernet.com). 2005. Industrial Ethernet.
http://www.industrialethernet.com/etad.html. Accessed date: July.
Groover, Mikell P. 2001. Automation, Production Systems, and ComputerIntegrated Manufacturing. Second Edition. New Jersey: Prentice Hall.
H. Kirrmann in ABB Research Center of Switzerland. 2006. The HART Protocol.
AI_411_HART.ppt. Accessed date: October 2007.
Zhang_Ch03.indd 424
5/13/2008 5:41:46 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
425
Hardware Secrets (http://www.hardwaresecrets.com). 2007a. PCI Bus Tutorial.
http://www.hardwaresecrets.com/article/190. Accessed date: May.
Hardware Secrets (http://www.hardwaresecrets.com). 2007b. AGP Bus Tutorial.
http://www.hardwaresecrets.com/article/155. Accessed date: July.
Harris, Don. 2006. Human–Machine Interaction. http://www.cranfield.ac.uk/soe/
postgraduate/hf_module9.htm. Accessed date: October.
HCF (HART Communication Foundation). 2007. HCF—Main Pages. http://www
.hartcomm2.org/index.html. Accessed date: October.
HIT (http://www.hit.bme.hu). 2007. GPIB Tutorial. http://www.hit.bme.hu/~papay/
edu/GPIB/tutor.htm. Accessed date: July.
HMS Industrial Networking (http://www.anybus.com). 2005a. AS-Interface
Technologies. http://www.anybus.com/technologies/asi.shtml. Accessed date:
July.
HMS Industrial Networking (http://www.anybus.com). 2005b. AS-Interface
Products. http://www.anybus.com/products/asinterface.shtml. Accessed date:
July.
HMS Industrial Networking (http://www.anybus.com). 2005c. Interbus Connectivity: http://www.anybus.com/products/interbus.shtml?gclid=CLXtoK7UjY
wCFT4GQgod3T_GBw. Accessed date: July.
Honey Well (http://hpsweb.honeywell.com). 2005a. http://hpsweb.honeywell.com/
Cultures/en-US/Products/Systems/ExperionPKS/FoundationFieldbus
Integration/default.htm. Accessed date: July.
Honey Well (http://hpsweb.honeywell.com). 2007. Wireless HART. http://hpsweb
.honeywell.com/Cultures/en-US/Products/wireless/SecondGeneration
Wireless/default.htm. Accessed date: October.
IBM (http://www.ibm.com/us). 2005. IDE Subsystem. http://publib.boulder.ibm
.com/infocenter/pseries/v5r3/index.jsp?topic=/com.ibm.aix.kernelext/doc/
kernextc/ide_subsys.htm. Accessed date: May.
Interbus Club (http://www.interbusclub.com). 2005a. Interbus Technology. http://
www.interbusclub.com/en/index.html. Accessed date: July.
Interbus Training in Web (http://pl.et.fh-duesseldorf.de/prak/prake/index.asp).
2007. Interbus Basic. http://pl.et.fh-duesseldorf.de/prak/prake/download/IBS_
grundlagen.pdf. Accessed date: May.
Interface Bus (http://www.interfacebus.com). 2005a. PCI Bus Pins. http://www
.interfacebus.com/Design_PCI_Pinout.html. Accessed date: May.
Interface Bus (http://www.interfacebus.com). 2005b. PCMCIA 16 bits Bus. http://
www.interfacebus.com/Design_Connector_PCMCIA.html. Accessed date:
May.
Interface Bus (http://www.interfacebus.com). 2005c. RS-485 Bus. http://www
.interfacebus.com/Design_Connector_RS485.html. Accessed date: May.
IO Tech (http://www.iotech.com). 2007. IEEE-488 Standard. http://www.iotech
.com/an06.html. Accessed date: July.
Israel, Johann Habakuk and Anja Naumann. 2006. http://useworld.net/ausgaben/
4-2007/01-Israel_Naumann.pdf. Accessed date: October.
IXXAT (http://www.ixxat.com). 2005. CAN Application Layer. http://www.ixxat
.com/can_application_layer_introduction_en,7524,5873.html. Accessed date:
July.
Zhang_Ch03.indd 425
5/13/2008 5:41:46 PM
426
INDUSTRIAL CONTROL TECHNOLOGY
Jaffe, David. 2006. Enhancing Human–Machine Interaction. http://ability.stanford
.edu/Press/rehabman.pdf. Accessed date: October.
Jim Russell. 2007. HART v Foundation Fieldbus. http://www.iceweb.com.au/
Instrument/FieldbusPapers/HART%20v%20FF%20PAPERfinal.pdf .
Accessed date: October.
Kenneth L. Holladay, P. E. in Southwest Research Institute of the USA.
1991. Calibrating HART Transmitters. http://www.transcat.com/PDF/Hart_
Transmitter_Calibration.pdf. Accessed date: October 2007.
Kvaser (http://www.kvaser.com). 2005. Controller Area Network. http://www
.kvaser.com/can/. Accessed date: July.
Microchip (http://ww1.microchip.com). 2005a. CAN Basics. http://ww1
.microchip.com/downloads/en/AppNotes/00713a.pdf. Accessed date: July.
Microchip (http://ww1.microchip.com). 2005b. CAN Physical Layer. http://ww1
.microchip.com/downloads/en/AppNotes/00228a.pdf. Accessed date: July.
MOXA (http://www.moxa.com). 2005a. Industrial Ethernet Technologies. http://
www.moxa.com/Zones/Industrial_Ethernet/Tutorial.htm. Accessed date: July.
MOXA (http://www.moxa.com). 2005b. Industrial Ethernet Products. http://www
.moxa.com/product/Industrial_Ethernet_Switches.htm. Accessed date: July.
Murray, Steven A. 2006. Human–Machine Interaction with Multiple Autonomous
Sensors. http://www.spawar.navy.mil/robots/research/hmi/ifac.html. Accessed
date: October.
PC Guide (http://www.pcguide.com). 2005a. IDE Guide. http://www.pcguide
.com/ref/hdd/if/ide/unstdIDE-c.html. Accessed date: May.
PC Guide (http://www.pcguide.com). 2005b. SCSI Guide. http://www.pcguide
.com/ref/hdd/if/scsi/. Accessed date: May.
PEPPERL + FUCHS (http://www.am.pepperl-fuchs.com). 2007. HART
Multiplexers. http://www.am.pepperl-fuchs.com/products/productsubfamily
.jsp?division=PA&productsubfamily_id=1343. Accessed date: October.
PEPPERL+FUCHS (http://www.am.pepperl-fuchs.com). 2005a. AS-Interface.
http://www.pepperl-fuchs.com/cgi-bin/site_search.pl. Accessed date: July.
PEPPERL+FUCHS (http://www.am.pepperl-fuchs.com). 2005b. Fieldbus
Technology; Foundation Fieldbus: AS-Interface. http://www.pepperl-fuchs
.com/pa/interbtob/communication/default_e.html. Accessed date: July.
PHM (http://www.phm.lu). 2007. IDE Pins Out. http://www.phm.lu/Documenta
tion/Connectors/IDE.asp. Accessed date: May.
PROFIBUS (http://www.profibus.com). 2005a. Profibus Technical Description.
http://www.profibus.com/pb/technology/description/. Accessed date: July.
PROFIBUS (http://www.profibus.com). 2005b. Profibus Specification. http://www
.profibus.com/pall/meta/downloads/. Accessed date: July.
QSI (http://www.qsicorp.com). 2006. Human–Machine Interface Devices. http://
www.qsicorp.com/product/industrial/?gclid=CK-apKS1l4wCFSQHE
god6iqP5w. Accessed date: October.
Quatech (http://www.quatech.com). 2005. ISA Bus Overviews. http://www.quatech
.com/support/comm-over-isa.php. Accessed date: July.
Samson (http://www.samson.de). 2005. FOUNDATION Fieldbus Technical
Information. http://www.samson.de/pdf_en/l454en.pdf. Accessed date: July.
SAMSON. 2005. SAMSON Technical Information—HART Communications.
http://www.samson.de/pdf_en/l452en.pdf. Accessed date: October 2007.
Zhang_Ch03.indd 426
5/13/2008 5:41:46 PM
3: SYSTEM INTERFACES FOR INDUSTRIAL CONTROL
427
Schneider-Electric (http://www.automation.schneider-electric.com). 2006. Human–
Machine Interface. http://www.automation.schneider-electric.com/as-guide/
EN/pdf_files/asg-8-human-machine-interface.pdf. Accessed date: October.
SCSI Library (http://www.scsilibrary.com). 2005. SCSI. http://www.scsilibrary
.com. Accessed date: May.
Semiconductors (http://www.semiconductors.bosch.de). 2005. CAN Specifications.
http://www.semiconductors.bosch.de/pdf/can2spec.pdf. Accessed date: July.
SIEMENS (http://www.automation.siemens.com). 2005a. Siemens AS-Interface
Technologies. http://www.automation.siemens.com/cd/as-interface/html_76/
asisafe.htm. Accessed date: July.
SIEMENS (http://www.automation.siemens.com). 2005b. Siemens AS-Interface
Products. http://www.automation.siemens.com/infocenter/order_form.aspx?
tab=3&nodekey=key_1994569&lang=en. Accessed date: July.
SIEMENS (http://www.automation.siemens.com). 2005c. Siemens Profibus:
http://www.automation.siemens.com/net/html_76/produkte/020_produkte.htm.
Accessed date: July.
SIEMENS (http://www.automation.siemens.com). 2005d. Industrial Ethernet
Technologies and Products. http://www.automation.siemens.com/net/html_76/
produkte/040_produkte.htm. Accessed date: July.
Simons, C. L. and Parmee, L. C. 2006. Human–Machine Interaction Software
Design.
http://www.ip-cc.org.uk/INTREP-COINT-SIMONS-2006.pdf.
Accessed date: October.
SMAR International Corporation (http://www.smar.com). 2006. HART Tutorial.
http://www.smar.com/PDFs/Catalogues/Hart_Tutorial.pdf. Accessed date:
October 2007.
Softing (http://www.softing.com). 2005a. FOUNDATION Fieldbus. http://www
.softing.com/home/en/industrial-automation/products/foundation-fieldbus/
index.php. Accessed date: July.
Softing (http://www.softing.com). 2005b. Profibus. http://www.softing.com/home/
en/industrial-automation/products/profibus-dp/index.php?navanchor=
3010004. Accessed date: July.
Tech Soft (http://www.techsoft.de). 2007. IEEEE-488 Tutorial. http://www
.techsoft.de/htbasic/tutgpibm.htm?tutgpib.htm. Accessed date: July.
Techfest (http://www.techfest.com). 2005. ISA Bus Technology. http://www
.techfest.com/hardware/bus/isa.htm. Accessed date: July.
Texas Instruments (http://sparc.feri.uni-mb.si). 2005. CAN Introduction. http://
sparc.feri.uni-mb.si/Sistemidaljvodenja/Vaje/pdf/Inroduction%20to%20
CAN.pdf. Accessed date: July.
The UK AS-i Expert Alliance (http://www.as-interface.com). 2007a. AS-Interface
Technologies. http://www.as-interface.com/asitech.asp. Accessed date: May.
The UK AS-i Expert Alliance (http://www.as-interface.com). 2007b. AS-Interface
Products. http://www.as-interface.com/asi_literature.asp. Accessed date: May.
Vector Germany (http://www.can-solutions.com). 2005. CAN Mechanism. http://
www.can-solutions.com/?gclid=CLWxnai6jIwCFQrlQgodnx3gBg.Accessed
date: July.
YOKOGAWA (http://www.yokogawa.com). 2007. HART Communicator. http://
www.yokogawa.com/us/mi/MetersandInstruments/us-ykgw-yhcypc.htm.
Accessed date: October.
Zhang_Ch03.indd 427
5/13/2008 5:41:46 PM
Zhang_Ch03.indd 428
5/13/2008 5:41:46 PM
4
Digital Controllers for
Industrial Control
4.1 Industrial Intelligent Controllers
4.1.1
Programmable Logic Control (PLC) Controllers
The development of Programmable Logic Controllers (PLCs) was
driven primarily by the requirements of automobile manufacturers who
constantly changed their production line control systems to accommodate
their new car models. In the past, this required extensive rewiring of banks
of relays—a very expensive procedure. In the 1970s, with the emergence
of solid-state electronic logic devices, several auto companies challenged
control manufacturers to develop a means of changing control logic
without the need to rewire the system totally. The PLC evolved from this
requirement. The PLCs are designed to be relatively “user-friendly” so
that electricians can easily make the transition from all-relay control to
electronic systems. They give users the capability of displaying and troubleshooting ladder logic that shows the logic in real time. The logic can be
“rewired” (programmed) and tested, without the need to assemble and
rewire banks of relays.
A PLC is a computer with a single mission. It usually lacks a monitor, a
keyboard, and a mouse, as it is programmed normally to operate a machine
or a system using but one program. The machine or system user rarely, if
ever, interacts directly with the PLC’s program. When it is necessary to
either edit or create the PLC program, a personal computer is usually
(but not always) connected to it. The information from the PLCs can be
accessed by supervisory control and data acquisition (SCADA) systems
and Human–Machine Interfaces (HMIs), to provide a graphical representation of the status of the plant. Figure 4.1 is a schematic of the PLC’s
control network resident in industrial systems.
4.1.1.1
Components and Architectures
PLC is actually an industrial microcontroller system (in more recent
years we meet microprocessors instead of microcontrollers) where you
have hardware and software specifically adapted to industrial environment.
Block schema with typical components that a PLC consists of is found in
Fig. 4.2. Special attention needs to be given to input and output, because
429
Zhang_Ch04.indd 429
5/13/2008 5:50:42 PM
430
INDUSTRIAL CONTROL TECHNOLOGY
SCADA system
Industrial network
(High level)
Other part
of factory
Industrial network
(Middle level)
Local PC
Visual and sound signals
Central PLC controller
Local PLC controller
Input sensing devices
Output load devices
Local process control system
Figure 4.1 Schematic of the PLC control network.
Screw terminals for input lines
PLC controller
Input port interface
Power
supply
Communi
cation
PC for programming
Extension
interface
Memories
CPU
Internal buses
Output port interface
Screw terminals for output lines
Figure 4.2 Basic elements of a PLC controller.
Zhang_Ch04.indd 430
5/13/2008 5:50:42 PM
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
431
most PLC models feature a vast assortment of interchangeable I/O modules that allow for convenient interfacing with virtually any kind of industrial or laboratory equipment. Program unit is usually a computer used for
writing a program (often in ladder diagram).
(1) Central Processing Unit (CPU). This unit contains the “brains”
of the PLC. It is often referred to as a microprocessor or sequencer.
The basic instruction set is a high-level program, installed in
Read-Only Memory (ROM). The programmed logic is usually
stored in Electrically Erasable Permanent Read-Only Memory
(EEPROM). The CPU will save everything in memory, even
after a power loss. Since it is “electrically erasable,” the logic can
be edited or changed as the need arises. The programming device
is connected to the CPU whenever the operator needs to monitor,
troubleshoot, edit, or program the system, but it is not required
during the normal running operations.
(2) Memory. System memory (today mostly implemented in FLASH
technology) is used by a PLC for a process control system. Aside
from this operating system, it also contains a user program translated from a ladder diagram to a binary form. FLASH memory
contents can be changed only in a case where the user program
is being changed. PLC controllers were used earlier instead of
FLASH memory and have had EPROM memory instead of
FLASH memory that had to be erased with UV lamp and programmed on programmers. With the use of FLASH technology
this process was greatly shortened. Reprogramming a program
memory is done through a serial cable in a program for application development.
User memory is divided into blocks having special functions.
Some parts of a memory are used for storing input and output
status. The real status of an input is stored either as “1” or as “0”
in a specific memory bit. Each input or output has one corresponding bit in memory. Other parts of the memory are used to
store variable contents for variables used in user programs. For
example, timer value, or counter value would be stored in this
part of the memory.
PLC controller memory consists of several areas given in
Table 4.1, some of these having predefined functions.
(3) Communication board. Every brand of PLC has its own programming hardware. Sometimes it is a small hand-held device,
which resembles an oversized calculator with a liquid crystal
display (LCD). However, most of the times it is the computerbased programmers. Computer-based programmers typically use
a special communication board, installed in an industrial terminal
Zhang_Ch04.indd 431
5/13/2008 5:50:44 PM
Zhang_Ch04.indd 432
Working
area
Output area
Input area
Timer/counter area
LR area
AR area
HR area
TR area
SR area
IR area
Data Area
IR 01000–IR 01915
(160 bits)
IR 20000–IR 23115
(512 bits)
SR23200–SR25515
(384 bits)
TR 0–TR 7 (8 bits)
IR 00000–IR 00915
(160 bits)
Bit(s)
HR 00–HR 19
HR0000–HR1915
(20 words)
(320 bits)
AR 00–AR 15
AR0000–AR1515
(16 words)
(256 bits)
LR 00–LR 15
LR0000–LR1515
(16 words)
(256 bits)
TC 000–TC 127 (timer/counter numbers)
IR 010–IR 019
(10 words)
IR 200–IR 231
(32 words)
SR 232–SR 255
(24 words)
–
IR 000–IR 009
(10 words)
Word(s)
Table 4.1 Memory Structure of PLC
Same numbers are used for both timers and
counters
1:1 connection with another PC
Temporary storage of ON/OFF states when jump
takes place
Data storage; these keep their states when power
is off
Special functions, such as flags and control bits
Working bits that can be used freely in the program. They are commonly used as swap bits
Special functions, such as flags and control bits
These bits may be assigned to an external I/O
connection. Some of these have direct output on
screw terminal (e.g., IR000.00–IR000.05 and
IR010.00–IR010.03 with CPM1A model)
Function
432
INDUSTRIAL CONTROL TECHNOLOGY
5/13/2008 5:50:44 PM
Zhang_Ch04.indd 433
PC setup
Read only
DM 6144–DM 6599
(456 words)
DM 6600–DM 6655
(56 words)
DM 0000–DM 0999
and DM 1022–DM
1023 (1002 words)
DM 1000–DM 1021
(22 words)
–
–
–
–
Storing various parameters for controlling the PC
Data of DM area may be accessed only in word
form. Words keep their contents after the power
is off
Part of the memory for storing the time and code
of error that occurred. When not used for this
purpose, they can be used as regular DM words
for reading and writing. They cannot be changed
from within the program
Notes:
1. IR and LR bits, when not used to their purpose, may be used as working bits.
2. Contents of HR area, LR area, counter, and DM area for reading/writing are stored within backup condenser. On 25C, condenser keeps the
memory contents for up to 20 days.
3. When accessing the current value of PV, TC numbers used for data have the form of word. When accessing the Completing flags, they are used
as data bits.
4. Data from DM6144 to DM6655 must not be changed from within the program, but can be changed by a peripheral device.
DM area
Error
writing
Read/write
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
433
5/13/2008 5:50:44 PM
434
INDUSTRIAL CONTROL TECHNOLOGY
or personal computer, with the appropriate software program
installed. Computer-based programming allows “offline” programming, where the programmers develop their logic, store it
on a disk, and then “down-load” the program to the CPU at their
convenience. In fact, it allows more than one programmer to
develop different modules of the program.
Programming can be done directly into the CPU if desired.
When connected to the CPU the programmer can test the system,
and watch the logic operate as each element is intensified in
sequence on a cathode ray tube (CRT) when the system is running. Since a PLC can operate without having the programming
device attached, one device can be used to service many separate
PLC systems.
(4) PLC controller inputs. Intelligence of an automated system
depends largely on the ability of a PLC controller to read signals
from different types of sensors and input devices. Keys, keyboards, and functional switches are a basis for human versus
machine relationship. On the other hand, to detect a working
piece, view a mechanism in motion, check pressure, or fluid level
you need specific automatic devices such as proximity sensors,
marginal switches, photoelectric sensors, level sensors, and so
on. Thus, input signals can be logical (ON/OFF) or analog.
Smaller PLC controllers usually only have digital input lines
while larger ones also accept analog inputs through special units
attached to a PLC controller. One of the most frequent analog
signals is a current signal of 4–20 mA and millivolt voltage signal
generated by various sensors. Sensors are usually used as inputs
for PLCs. You can obtain sensors for different purposes. They
can sense presence of some parts, measure temperature, pressure, or some other physical dimension, and so on (for instance,
inductive sensors can register metal objects).
Other devices also can serve as inputs to the PLC controller.
Intelligent devices such as robots, video systems, and so forth
often are capable of sending signals to PLC controller input
modules (robot, for instance, can send a signal to PLC controller
input as information when it has finished moving an object from
one place to the other.)
(5) PLC controller output. An industrial control system is incomplete if it is not connected with some output devices. Some of the
most frequently used devices are motors, solenoids, relays, indicators, sound signalization, and so forth. By starting a motor, or a
relay, PLC can manage or control a simple system such as a system
for sorting products all the way up to complex systems such as a
service system for positioning the head of a robotic machine.
Zhang_Ch04.indd 434
5/13/2008 5:50:44 PM
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
435
Output can be of analog or digital type. A digital output signal
works as a switch; it connects and disconnects lines. Analog output is used to generate the analog signal (for instance, a motor
whose speed is controlled by a voltage that corresponds to a
desired speed).
(6) Extension lines. Every PLC controller has a limited number of
input/output lines. If needed, this number can be increased
through certain additional modules by system extension through
extension lines. Each module can contain extension both of input
and output lines. Also, extension modules can have inputs and
outputs of a different nature from those on the PLC controller
(for instance, in case relay outputs are on a controller, transistor
outputs can be on an extension module). PLC has input and output lines through which it is connected to a system it directs.
This is a very important part of the story about PLC controllers
because it directly influences what can be connected and how it
can be connected to controller inputs or outputs. Two terms most
frequently mentioned when discussing connections to inputs or
outputs are “sinking” and “sourcing.” These two concepts are very
important in connecting a PLC correctly with the external environment. The briefest definition of these two concepts would be
Sinking = Common GND line (–)
Sourcing = Common VCC line (+),
The first thing that catches one’s eye is “+” and “–” supply DC
supply. Inputs and outputs that are either sinking or sourcing can
conduct electricity only in one direction, so they are only supplied with direct current. According to what we have discussed
so far, each input or output has its own return line, so 5 inputs
would need 10 screw terminals on a PLC controller housing.
Instead, we use a system of connecting several inputs to one
return line as in the following picture. These common lines are
usually marked “COMM” on the PLC controller housing.
(7) Power supply. Electrical supply is used in bringing electrical
energy to a CPU. Most PLC controllers work either at 24 VDC
or 220 VAC. On some PLC controllers, you will find electrical
supply as a separate module. Those are usually bigger PLC
controllers, while small and medium series already contain the
supply module. The user has to determine how much current to
take from the I/O module to ensure that the electrical supply
provides the appropriate amount of current. Different types of
modules use different amounts of electrical current.
This electrical supply is usually not used to start external
inputs or outputs. The user has to provide separate supplies in
Zhang_Ch04.indd 435
5/13/2008 5:50:44 PM
436
INDUSTRIAL CONTROL TECHNOLOGY
starting PLC controller inputs or outputs to ensure so called pure
supply for the PLC controller. With pure supply we mean supply
where industrial environment cannot affect it damagingly. Some
of the smaller PLC controllers supply their inputs with voltage
from a small supply source already incorporated into a PLC.
The internal logic and communication circuitry usually operates on 5 and 15 V DC power. The power supply provides filtering and isolation of the low voltage power from the AC power
line. Power supply assemblies may be separate modules, or in
some cases, plug-in modules in the I/O racks. Separate control
transformers are often used to isolate inputs and CPU from output devices. The purpose is to isolate this sensitive circuitry from
transient disturbances produced by any highly inductive output
devices.
(8) Timers and counters. Timers and counters are indispensable in
PLC programming. Industry has to number its products, determine a needed action in time, and so on. Timing functions are
very important, and cycle periods are critical in many processes.
There are two types of timers: delay-off and delay-on. First is
late with turn off and another runs late in turning on in relation to
a signal that activated timers. Example of a delay-off timer would
be staircase lighting. Following its activation, it simply turns off
after a few minutes.
Each timer has a time basis, or more precisely has several time
bases. Typical values are 1, 0.1, and 0.01 s. If the programmer
has entered 0.1 as time basis and 50 as a number for delay
increase, the timer will have a delay of 5 s (50 × 0.1 s = 5 s).
Timers also have to have the value SV set in advance. Value
set in advance or ahead of time is a number of increments that the
timer has to calculate before it changes the output status. Values
set in advance can be constants or variables. If a variable is used,
the timer will use a real time value of the variable to determine a
delay. This enables delays to vary depending on the conditions
during function. An example is a system that has produced two
different products, each requiring different timing during process itself. Product A requires a period of 10 s, so number 10
would be assigned to the variable. When product B appears, a
variable can change value to what is required by product B.
Typically, timers have two inputs. First is timer enable, or conditional input (when this input is activated, timer will start counting). The second input is a reset input. This input has to be in
OFF status in order for a timer to be active, or the whole function
would be repeated over again. Some PLC models require
this input to be low for a timer to be active; other makers require
Zhang_Ch04.indd 436
5/13/2008 5:50:45 PM
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
437
high status (all of them function in the same way basically).
However, if a reset line changes status, the timer erases accumulated value.
4.1.1.2
Control Mechanism
A programmable logic controller is a digital electronic device that uses
a programmable memory to store instructions and uses a CPU to implement specific functions such as logic, sequence, timing, counting, and
arithmetic to control machines and processes. Figure 4.2 shows a simple
schematic of a typical programmable logic controller. When running, the
CPU scans the memory continuously from top to bottom, and left to right,
checking every input, output, and instruction in sequence. The scan time
depends upon the size and complexity of the program, and the number and
type of I/O. The scan may be as short as a few milliseconds or less. A few
milliseconds per scan would produce tens of scans per second. This short
time makes the operation appear as instantaneous, but one must consider
the scan sequence when handling critically timed operations and sealing
circuits. Complex systems may use interlocked multiple CPUs to minimize total scan time.
The parts of the PLC that are quite different from the typical desktop
computer are the input and output modules. These modules allow the PLC
to communicate with the machine. The inputs may come from limit switches,
proximity sensors, temperature sensors, and so on. On the basis of the
software program and the combination of inputs, the CPU of the PLC will
set the outputs. These outputs may control motor speed and direction,
actuate valves, open or close gates, and control all the motions and activities of the machine.
(1) System address. The key to getting comfortable with any PLC is
to understand the total addressing system. We have to connect our
discrete inputs, pushbuttons, limit-switches, and so on, to our
controller, interface those points with “electronic ladder diagram”
(program), and then bring the results out through another interface to operate motor starters, solenoids, lights, and so forth.
Inputs and outputs are wired to interface modules, installed in
an I/O rack. Each rack has a two-digit address, each slot has its
own address, and each terminal point is numbered. Figure 4.3
shows a PLC product in which all of these addresses are octal.
We combine the addresses to form a number that identifies each
input and output.
Zhang_Ch04.indd 437
5/13/2008 5:50:45 PM
438
INDUSTRIAL CONTROL TECHNOLOGY
I/O Rack
Rack no. 00
Output module
7
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
6
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
5
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
4
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
3
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
2
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
Input module
I:000/04
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
Closed input
Slot numbers
1
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
0
Energized output
0:007/15
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
Output image
table
Word 0:007
17 16 15 14 13 12 11 10 07 06 05 04 03 02 01 00
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
Input image
table
Word 1:000
1:000
] [
04
User’s logic rung
Input, rack 00, slot 0, terminal 04
0:007
( )
15
Output, rack 00, slot 7, terminal 15
Figure 4.3 Solution of one line of logic.
Some manufacturers use decimal addresses. Some older systems are based on 8-bit words, rather than 16. There are a number of proprietary programmable controllers applied to special
industries, such as elevator controls, or energy management,
which may not follow the expected pattern, but they will use
either 8- or 16-bit word structures. It is very important to identify
the addressing system before you attempt to work on any system
that uses a programmable controller, because one must know the
purpose of each I/O bit before manipulating them in memory.
Zhang_Ch04.indd 438
5/13/2008 5:50:45 PM
439
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
If you know the address of the input or output, you can immediately check the status of its bit by calling up the equivalent
address on a cathode ray tube (CRT) screen for most of PLC
products.
(2) I/O addresses. Figure 4.4 gives an I/O address scheme, which
shows us that the I/O modules are closely linked with the Input
and Output image tables, respectively.
Figure 4.3 shows a very simple line of logic, where a pushbutton is used to turn on a lamp. The pushbutton and lamp “hardwiring” terminates at I/O terminals, and the logic is carried out
in software. We have a pushbutton, wired to an input module (I),
installed in rack 00, slot 0, and terminal 04. The address becomes
I:000/04. An indicating lamp is wired to an output module (O),
installed in rack 00, slot 7, and terminal 15. The address becomes
O:007/15. Our input address, I:000/04, becomes memory address
Word #
0
1
2
3
4
5
6
7
0
1
2
3
4
5
6
7
Output image table
I/O group designation
An I/O chassis containing 16-point modules
Note: Modules can also be
installed like this: I O O I
Input/output designation
I O I O I O IO IO IO IO IO
Input image table
Word
0
1
2
3
4
5
6
7
Figure 4.4 I/O addressing scheme.
Zhang_Ch04.indd 439
5/13/2008 5:50:45 PM
440
INDUSTRIAL CONTROL TECHNOLOGY
I:000/04, and the output address 0:007/15 becomes memory
address 0:007/15.
In other words, the type of module, the rack address, and the
slot position identifies the word address in memory. The terminal
number identifies the bit number.
(3) Image table addresses. An output image table is reserved in its IR
area of the memory (see Table 4.1) as File format, and an input
image table is reserved in the same way. A File in memory contains any number of words. Files are separated by type, according to their functions. In the same way, an input image table is
also reserved in its IR area of the memory (See Table 4.1) as File
format. Figure 4.4 illustrates the respective mapping relationship
of the I/O modules to both Output and Input image tables.
(4) Scanning. As the scan reads the input image table, it notes the
condition of every input, and then scans the logic diagram, updating all references to the inputs. After the logic is updated, the
scanner resets the output image table, to activate the required
outputs. Figure 4.4 shows some optional I/O arrangements and
addressing.
In Fig. 4.3, we show how one line of logic would perform
when the input at I:000/04 is energized, it immediately sets input
image table bit I:000/04 true (ON). The scanner senses this
change of state, and makes the element I:000/04 true in our logic
diagram. Bit 0:007/15 is turned on by the logic. The scanner sets
0:007/15 true in the output image table, and then updates the
output 0:007/15 to turn the lamp on.
4.1.1.3
PLC Programming
Programmable logic controllers use a variety of software programming
languages for control. These include sequential function chart (SFC),
function block diagram (FBD), ladder diagram (LD), structured text (ST),
instruction list (IL), relay ladder logic (RLL), flow chart, C, C++, and
Basic. Among these languages, ladder diagram is the most popular. Almost
every program for programming a PLC controller possesses various useful
options such as: forced switching on and off of the system inputs/outputs
(I/O lines), program follow up in real time as well as documenting a diagram. This documenting is necessary to understand and to define failures
and malfunctions. The programmer can add remarks, names of input or
output devices, and comments that can be useful when finding errors, or
with system maintenance. Adding comments and remarks enables any
technician (and not just a person who developed the system) to understand
a ladder diagram right away. Comments and remarks can even quote
Zhang_Ch04.indd 440
5/13/2008 5:50:47 PM
441
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
precisely part numbers if replacements would be needed. This would speed
up repair of any problems that come up due to bad parts. The old way was
such that a person who developed a system had protection on the program,
so nobody aside from this person could understand how it was done. A
correctly documented ladder diagram allows any technician to understand
thoroughly how the system functions.
(1) Relay ladder logic. Ladder logic is the main programming
method used for PLCs. As mentioned before, ladder logic has
been developed to mimic relay logic. Relays are used to let one
power source close a switch for another (often-high current)
power source, while keeping them isolated. An example of a
relay in a simple control application is shown in Fig. 4.5. In this
system, the first relay on the left is used as normally closed and
will allow current to flow until a voltage is applied to input A.
The second relay is normally open and will not allow current to
flow until a voltage is applied to input B. If current is flowing
through the first two relays, then current will flow through the
coil in the third relay, and close the switch for output C. This
circuit would normally be drawn in the ladder logic form. This
can be read logically as C will be on if A is off and B is on.
115 VAC wall plug
Relay logic
Input A
(Normally closed)
A
Input B
(Normally open)
B
Output C
(Normally open)
C
Ladder logic
Figure 4.5 A simple relay controller.
Zhang_Ch04.indd 441
5/13/2008 5:50:47 PM
442
INDUSTRIAL CONTROL TECHNOLOGY
The example in Fig. 4.5 does not show the entire control
system, but only the logic. When we consider a PLC there are
inputs, outputs, and the logic. Figure 4.6 shows a more complete
representation of the PLC. Here there are two inputs from pushbuttons. We can imagine the inputs as activating 24 VDC relay
coils in the PLC. This in turn drives an output relay that switches
115 VAC, which will turn on a light. Note, in actual PLCs inputs
are never relays, but outputs are often relays. The ladder logic in
the PLC is actually a computer program that the user can enter
and change. Note that both of the input pushbuttons are normally
open, but the ladder logic inside the PLC has one normally open
contact and one normally closed. Do not think that the ladder
logic in the PLC needs to match the inputs or outputs. Many
beginners will get caught trying to make the ladder logic match
the input types.
Many relays also have multiple outputs (throws) and this
allows an output relay to also be an input simultaneously. The
circuit shown in Fig. 4.7(a) is an example of this; it is called
a seal in circuit. In this circuit, the current can flow through
either branch of the circuit, through the contacts labeled A or B.
Power supply
+24 V
Push buttons
PLC
Inputs
C
A
B
Ladder
logic
Outputs
Light
115 VAC
power
Figure 4.6 A PLC illustrated with relays.
Zhang_Ch04.indd 442
5/13/2008 5:50:47 PM
443
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
A
B
B
X
Normally open
X
Normally closed
Normal output
Normally on output
IIT
X
Immediate inputs
One Shot Relay
OSR
X
Latch
L
U
IOT
(a)
(b)
X
Unlatch
X
Immediate Output T
(c)
Figure 4.7 Relay Ladder logic representations: (a) a seal-in circuit; (b) Ladder
logic inputs; and (c) Ladder logic outputs.
The input B will only be on when the output B is on. If B is off,
and A is energized, then B will turn on. If B turns on then the
input B will turn on and keep output B on even if input A goes
off. After B is turned on the output, B will not turn off.
PLC inputs are easily represented in ladder logic. Figure 4.7(b)
shows there are three types of inputs. The first two are normally
open and closed inputs, discussed previously. Normally open: an
active input x will close the contact and allow power to flow.
Normally closed: power flows when the input x is not open. The
IIT (Immediate InpuT) function allows inputs to be read after the
input scan, while the ladder logic is being scanned. This allows
ladder logic to examine input values more often than once every
cycle. Immediate inputs will take current values, but not those
from the previous input scan.
In ladder logic, there are multiple types of outputs, but these
are not consistently available on all PLCs. Some of the outputs
will be externally connected to devices outside the PLC, but it is
also possible to use internal memory locations in the PLC. Six
types of outputs are shown in Fig. 4.7(c). The first is a normal
output; when energized the output will turn on and energize an
output. The circle with a diagonal line through is a normally
on output. When it is energized, the output will turn off. This
type of output is not available on all PLC types. When initially
energized, the OSR (one shot relay) instruction will turn on for
Zhang_Ch04.indd 443
5/13/2008 5:50:47 PM
444
INDUSTRIAL CONTROL TECHNOLOGY
one scan, but then be off for all scans after, until it is turned off.
The L (latch) and U (unlatch) instructions can be used to lock
outputs on. When an L output is energized the output will turn on
indefinitely, even when the output coil is deenergized. The output can only be turned off using a U output. The last instruction
is the IOT (Immediate OutpuT) that will allow outputs to be
updated without having to wait for the ladder logic scan to be
completed. When power is applied (ON) the output x is activated
for the left output, but turned off for the output on the right. An
input transition on will cause the output x to go on for one scan
(this is also known as a one shot relay). When the L is energized,
x will be toggled on, and will stay on until the U coil is energized. This is like a flip-flop and stays set even when the PLC is
turned off. In some PLCs, all immediate outputs do not wait for
the program scan to end before setting an output.
For example, to develop (without looking at the solution) a
relay based controller that will allow three switches in a room to
control a single light, there are two possible approaches. The first
assumes that any one of the switches on will turn on the light, but
all three switches must be off for the light to be off. Figure 4.8(a)
displays the ladder logic of the first solution. The second solution
assumes that each switch can turn the light on or off, regardless
of the states of the other switches. This method is more complex
and involves thinking through all of the possible combinations of
switch positions. You can recognize this problem as an exclusive
or problem from Fig. 4.8(b).
(2) Programming. An example of ladder logic can be seen in Fig. 4.9.
To interpret this diagram, imagine that the power is on the
vertical line on the left-hand side; we call this the hot rail. On the
Switches
Switch 1
Light
Light
Switch 2
Switch 3
(a)
(b)
Figure 4.8 A case study: (a) solution 1 and (b) solution 2.
Zhang_Ch04.indd 444
5/13/2008 5:50:48 PM
445
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
Neutral
Hot
A
B
X
C
D
G
E
F
H
Inputs
Y
Outputs
Figure 4.9 A simple Ladder Logic diagram.
right-hand side is the neutral rail. In this figure there are two
rungs, and on each rung there are combinations of inputs (two
vertical lines) and outputs (circles). If the inputs are opened or
closed in the right combination the power can flow from the hot
rail, through the inputs, to power the outputs, and finally to the
neutral rail. An input can come from a sensor, switch, or any
other type of sensor. An output will be some device outside the
PLC that is switched ON or OFF, such as lights or motors. In the
top rung the contacts are normally open and normally closed,
which means if input A is ON and input B is OFF, then power
will flow through the output and activate it. Any other combination of input values will result in the output X being off.
The second rung of Fig. 4.9 is more complex; there are actually multiple combinations of inputs that will result in the output
Y turning on. On the left-most part of the rung, power could flow
through the top if C is OFF and D is ON. Power could also (and
simultaneously) flow through the bottom if both E and F are true.
This would get power half way across the rung, and then if G or
H is true the power will be delivered to output Y.
There are other methods for programming PLCs. One of the
earliest techniques involved mnemonic instructions. These
instructions can be derived directly from the ladder logic diagrams and entered into the PLC through a simple programming
terminal. An example of mnemonics is shown in Fig. 4.10. In
this example, the instructions are read one line at a time from top
to bottom. The first line 00000 has the instruction LDN (input
load and not) for input 00001. This will examine the input to the
PLC, and if it is OFF it will remember a 1 (or true); if it is ON it
will remember a 0 (or false). The next line uses an LD (input
load) statement to look at the input. If the input is OFF it remembers a 0; if the input is ON it remembers a 1 (note: this is the
Zhang_Ch04.indd 445
5/13/2008 5:50:48 PM
446
INDUSTRIAL CONTROL TECHNOLOGY
00000
00001
00002
00003
00004
00005
00006
00007
00008
LDN
LD
AND
LD
LD
AND
OR
ST
END
00001
00002
00003
00004
00107
The mnemonic code is equivalent to the ladder logic below
00001 00002
00107
00003 00004
END
Figure 4.10 A mnemonic program and equivalent Ladder Logic.
reverse of the LD). The AND statement recalls the last two numbers remembered and if they are both true the result is a 1; otherwise the result is a 0. This result now replaces the two numbers
that were recalled, and there is only one number remembered.
The process is repeated for lines 00003 and 00004, but when
these are done there are now three numbers remembered. The
oldest number is from the AND; the newer numbers are from the
two LD instructions. The AND in line 00005 combines the results
from the last LD instructions and now there are two numbers
remembered. The OR instruction takes the two numbers now
remaining and if either one is a 1 the result is a 1, otherwise the
result is a 0. This result replaces the two numbers, and there is
now a single number there. The last instruction is the ST (store
output) that will look at the last value stored and if it is 1, the
output will be turned on; if it is 0 the output will be turned off.
The ladder logic program in Fig. 4.10 is equivalent to the mnemonic program. Even if you have programmed a PLC with ladder
logic, it will be converted to mnemonic form before being used
by the PLC. In the past, mnemonic programming was the most
common, but now it is uncommon for users to even see mnemonic programs.
Sequential Function Charts have been developed to accommodate the programming of more advanced systems. These are
similar to flowcharts, but are much more powerful. The example
Zhang_Ch04.indd 446
5/13/2008 5:50:49 PM
447
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
seen in Fig. 4.11 is doing two different things. To read the chart,
start at the top where is says start. Below this there is the double
horizontal line that says follow both paths. As a result, the PLC
will start to follow the branch on the left- and right-hand sides
separately and simultaneously. On the left there are two functions; the first one is the power-up function. This function will
run until it decides it is done, and the power-down function will
come after. On the right-hand side is the flash function; this will
run until it is done. These functions look unexplained, but each
function, such as power-up will be a small ladder logic program.
This method is much different from flowcharts because it does
not have to follow a single path through the flowchart.
(3) Ladder diagram instructions. Ladder logic input contacts and
output coils allow simple logical decisions. Instructions extend
basic ladder logic to allow other types of control. Most of the
instructions will use PLC memory locations to get values, store
values, and track instruction status. Most instructions will normally become active when the input is true. But, some instructions, such as TOF timers, can remain active when the input is
off. Other instructions will only operate when the input goes
from false to true; this is known as positive edge triggered.
Consider a counter that only counts when the input goes from
false to true; the length of time the input is true does not change
the instruction behavior. A negative edge-triggered instruction
would be triggered when the input goes from true to false. Most
instructions are not edge-triggered: unless stated, assume instructions are not edge-triggered.
Instructions may be divided into several basic groups according
to their operation. Each of these instruction groups is introduced
with a brief description in Table 4.2.
Start
Power up
Flash
Multiple path execution flow
Power down
End
Figure 4.11 A sequential function chart.
Zhang_Ch04.indd 447
5/13/2008 5:50:49 PM
448
INDUSTRIAL CONTROL TECHNOLOGY
Table 4.2 Ladder Diagram Instructions
Group
Sequence Input
Instructions
Instruction
LOAD
LOAD NOT
AND
AND NOT
OR
OR NOT
AND LOAD
OR LOAD
Sequence
Output
Instructions
OUTPUT
OUT NOT
SET
RESET
KEEP
DIFFERENTIATE
UP
DIFFERENTIATE
DOWN
Sequence
Control
Instructions
NO OPERATION
END
INTERLOCK
Function
Connects an NO condition to the
left bus bar
Connects an NC condition to the
left bus bar
Connects an NO condition in
series with the previous condition
Connects an NC condition in
series with the previous condition
Connects an NO condition in parallel with the previous condition
Connects an NC condition in parallel with the previous condition
Connects two instruction blocks
in series
Connects two instruction blocks
in parallel
Outputs the result of logic to a bit
Reverses and outputs the result of
logic to a bit
Force sets (ON) a bit
Force resets (OFF) a bit
Maintains the status of the designated bit
Turns ON a bit for one cycle
when the execution condition
goes from OFF to ON
Turns ON a bit for one cycle
when the execution condition
goes from ON to OFF
—
Required at the end of the
program
It the execution condition for
IL(02) is OFF, all outputs are
turned OFF and all timer PVs
reset between IL(02) and the next
ILC(03)
(Continued)
Zhang_Ch04.indd 448
5/13/2008 5:50:49 PM
449
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
Table 4.2 Ladder Diagram Instructions (Continued)
Group
Instruction
INTERLOCK
CLEAR
JUMP
JUMP END
Timer/Counter
Instructions
Data
Comparison
Instructions
TIMER
COUNTER
REVERSIBLE
COUNTER
HIGH-SPEED
TIMER
COMPARE
DOUBLE
COMPARE
BLOCK COMPARE
TABLE COMPARE
Data Movement
Instructions
MOVE
MOVE NOT
BLOCK
TRANSFER
BLOCK SET
DATA EXCHAGE
SINGLE WORD
DISTRIBUTE
Function
ILC(03) indicates the end of an
interlock (beginning at IL(02))
If the execution condition for
JMP(04) is ON, all instructions
between JMP(04) and JME(05)
are treated as NOP(OO)
JME(05) indicates the end of a
jump (beginning at JMP(04))
An ON-delay (decrementing)
timer
A decrementing counter
Increases or decreases PV by one
A high-speed, ON-delay (decrementing) timer
Compares two four-digit hexadecimal values
Compares two eight-digit hexadecimal values
Judges whether the value of a
word is within 16 ranges (defined
by lower and upper limits)
Compares the value of a word to
16 consecutive words
Copies a constant or the content
of a word to a word
Copies the complement of a constant or the content of a word to a
word.
Copies the content of a block of
up to 1000 consecutive words to
a block of consecutive words
Copies the content of a word to a
block of consecutive words
Exchanges the content of two
words
Copies the content of a word to a
word (whose address is determined by adding an offset to a
word address)
(Continued)
Zhang_Ch04.indd 449
5/13/2008 5:50:49 PM
450
INDUSTRIAL CONTROL TECHNOLOGY
Table 4.2 Ladder Diagram Instructions (Continued)
Group
Instruction
Function
DATA COLLECT
Copies the content of a word
(whose address is determined by
adding an offset to a word
address) to a word
Copies the specified bit from one
word to the specified bit of a
word
Copies the specified digits (4-bit
units) from a word to the specified digits of a word
Copies the specified bit (0 or 1)
into the rightmost bit of a shift
register and shifts the other bits
one bit to the left
Creates a multiple-word shift register that shifts data to the left in
one-word units
Creates a shift register that
exchanges the contents of adjacent words when one is zero and
the other is not
Shifts a 0 into bit 00 of the specified word and shifts the other bits
one bit to the left
Shifts a 0 into bit 15 of the specified word and shifts the other bits
one bit to the right
Moves the content of CY into bit
00 of the specified word, shifts
the other bits one bit to the left,
and moves bit 15 to CY
Moves the content of CY into bit
15 of the specified word, shifts
the other bits one bit to the left,
and moves bit 00 to CY
Shifts a 0 into the rightmost digit
(4-bit unit) of the shift register
and shifts the other digits one
digit to the left
MOVE BIT
MOVE DIGIT
Shift
Instructions
SHIFT REGISTER
WORD SHIFT
ASYNCHRONOUS
SHIFT REGISTER
ARITHMETIC
SHIFT LEFT
ARITHMETIC
SHIFT RIGHT
ROTATE LEFT
ROTATE RIGHT
ONE DIGIT SHIFT
LEFT
(Continued)
Zhang_Ch04.indd 450
5/13/2008 5:50:49 PM
451
4: DIGITAL CONTROLLERS FOR INDUSTRIAL CONTROL
Table 4.2 Ladder Diagram Instructions (Continued)
Group