Đang tạo bản xem trước...
SENSORS AND SIGNAL CONDITIONING Second Edition RAMON PALLA¡ S-ARENY Universitat Polite¡cnica de Catalunya JOHN G. WEBSTER University of Wisconsin–Madison A Wiley-Interscience Publication JOHN WILEY & SONS, INC. New York . Chichester . Weinheim . Brisbane . Singapore . Toronto
This book is printed on acid-free paper.zy Copyright ( 2001 by John Wiley & Sons. All rights reserved. Published simultaneously in Canada. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, (508) 750-8400, fax (508) 750-4744. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 605 Third Avenue, New York, NY 10158-0012, (212) 850-6011, fax (212) 850-6008, E-Mail: PERMREQ@WILEY.COM. For ordering and customer service, call 1-800-CALL-WILEY. Library of Congress Cataloging-in-Publication Data: Palla¡s-Areny, Ramon. Sensors and signal conditioning /Ramon Palla¡s-Areny, John G. Webster.–2nd ed. p. cm. ``A Wiley-Interscience publication.'' Includes bibliographical references. ISBN 0-471-33232-1 (cloth : alk. paper) 1. Transducers. 2. Detectors. 3. Interface circuits. I. Webster, John G., 1932± II. Title. TK7872.T6 P25 2000 621.3815–dc21 00-028293 Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1
CONTENTS Preface xi 1 Introduction to Sensor-Based Measurement Systems 1 1.1 General Concepts and Terminology, 1 1.1.1 Measurement systems, 1 1.1.2 Transducers, sensors and actuators, 2 1.1.3 Signal conditioning and display, 4 1.1.4 Interfaces, data domains, and conversion, 4 1.2 Sensor ClassiÆcation, 6 1.3 General Input±Output ConÆguration, 7 1.3.1 Interfering and modifying inputs, 7 1.3.2 Compensation techniques, 11 1.4 Static Characteristics of Measurement Systems, 12 1.4.1 Accuracy, precision, and sensitivity, 13 1.4.2 Other characteristics: Linearity and resolution, 15 1.4.3 Systematic errors, 17 1.4.4 Random errors, 18 1.5 Dynamic Characteristics, 21 1.5.1 Zero-order measurement systems, 22 1.5.2 First-order measurement systems, 23 1.5.3 Second-order measurement systems, 26 1.6 Other Sensor Characteristics, 31 1.6.1 Input characteristics: Impedance, 33 1.6.2 Reliability, 34 1.7 Primary Sensors, 36 v
vi CONTENTS 1.7.1 Temperature sensors: Bimetals, 37 1.7.2 Pressure sensors, 38 1.7.3 Flow velocity and Øow-rate sensors, 41 1.7.4 Level sensors, 48 1.7.5 Force and torque sensors, 50 1.7.6 Acceleration and inclination sensors, 51 1.7.7 Velocity sensors, 52 1.8 Materials for Sensors, 54 1.8.1 Conductors, semiconductors, and dielectrics, 57 1.8.2 Magnetic materials, 59 1.9 Microsensor Technology, 62 1.9.1 Thick-Ælm technology, 63 1.9.2 Thin-Ælm technology, 64 1.9.3 Micromachining technologies, 65 1.10 Problems, 68 References, 70 2 Resistive Sensors 73 2.1 Potentiometers, 73 2.2 Strain Gages, 80 2.2.1 Fundamentals: Piezoresistive e®ect, 80 2.2.2 Types and applications, 85 2.3 Resistive Temperature Detectors (RTDs), 88 2.4 Thermistors, 94 2.4.1 Models, 94 2.4.2 Thermistors types and applications, 102 2.4.3 Linearization, 106 2.5 Magnetoresistors, 109 2.6 Light-Dependent Resistors (LDRs), 114 2.7 Resistive Hygrometers, 119 2.8 Resistive Gas Sensors, 121 2.9 Liquid Conductivity Sensors, 126 2.10 Problems, 129 References, 131 3 Signal Conditioning for Resistive Sensors 133 3.1 Measurement of Resistance, 133 3.2 Voltage Dividers, 139 3.2.1 Potentiometers, 141 3.2.2 Application to thermistors, 146 3.2.3 Dynamic measurements, 147 3.2.4 AmpliÆers for voltage dividers, 149 3.3 Wheatstone Bridge: Balance Measurements, 152 3.4 Wheatstone Bridge: DeØection Measurements, 154
CONTENTS vii 3.4.1 Sensitivity and linearity, 154 3.4.2 Analog linearization of resistive sensor bridges, 158 3.4.3 Sensor bridge calibration and balance, 158 3.4.4 Di®erence and average measurements and compensation, 159 3.4.5 Power supply of Wheatstone bridges, 165 3.4.6 Detection methods for Wheatstone bridges, 168 3.5 Di®erential and Instrumentation AmpliÆers, 170 3.5.1 Di®erential ampliÆers, 170 3.5.2 Instrumentation ampliÆer based on two op amps, 177 3.5.3 Instrumentation ampliÆers based on three op amps, 179 3.6 Interference, 184 3.6.1 Interference types and reduction, 184 3.6.2 Signal circuit grounding, 188 3.6.3 Shield grounding, 190 3.6.4 Isolation ampliÆers, 193 3.7 Problems, 198 References, 205 4 Reactance Variation and Electromagnetic Sensors 207 4.1 Capacitive Sensors, 207 4.1.1 Variable capacitor, 207 4.1.2 Di®erential capacitor, 216 4.2 Inductive Sensors, 220 4.2.1 Variable reluctance sensors, 220 4.2.2 Eddy current sensors, 225 4.2.3 Linear variable di®erential transformers (LVDTs), 229 4.2.4 Variable transformers: Synchros, resolvers, and Inductosyn, 238 4.2.5 Magnetoelastic and magnetostrictive sensors, 250 4.2.6 Wiegand and pulse-wire sensors, 254 4.2.7 Saturation-core (Øux-gate) sensors, 256 4.2.8 Superconducting quantum interference devices (SQUIDs), 258 4.3 Electromagnetic Sensors, 260 4.3.1 Sensors based on Faraday's law, 260 4.3.2 Hall e®ect sensors, 267 4.4 Problems, 272 References, 274 5 Signal Conditioning for Reactance Variation Sensors 277 5.1 Problems and Alternatives, 277 5.2 ac Bridges, 281
viii CONTENTS 5.2.1 Sensitivity and linearity, 281 5.2.2 Capacitive bridge analog linearization, 285 5.2.3 ac ampliÆers and power supply decoupling, 286 5.2.4 Electrostatic shields and driven shields, 292 5.2.5 ac/dc signal converters, 294 5.3 Carrier AmpliÆers and Coherent Detection, 299 5.3.1 Fundamentals and structure of carrier ampliÆers, 299 5.3.2 Phase-sensitive detectors, 306 5.3.3 Application to LVDTs, 311 5.4 SpeciÆc Signal Conditioners for Capacitive Sensors, 313 5.5 Resolver-to-Digital and Digital-to-Resolver Converters, 316 5.5.1 Synchro-to-resolver converters, 317 5.5.2 Digital-to-resolver converters, 319 5.5.3 Resolver-to-digital converters, 321 5.6 Problems, 322 References, 326 6 Self-Generating Sensors 329 6.1 Thermoelectric Sensors: Thermocouples, 329 6.1.1 Reversible thermoelectric e®ects, 329 6.1.2 Common thermocouples, 334 6.1.3 Practical thermocouple laws, 339 6.1.4 Cold junction compensation in thermocouple circuits, 341 6.2 Piezoelectric Sensors, 345 6.2.1 The piezoelectric e®ect, 345 6.2.2 Piezoelectric materials, 348 6.2.3 Applications, 350 6.3 Pyroelectric Sensors, 357 6.3.1 The pyroelectric e®ect, 357 6.3.2 Pyroelectric materials, 359 6.3.3 Radiation laws: Planck, Wien, and Stefan± Boltzmann, 360 6.3.4 Applications, 362 6.4 Photovoltaic Sensors, 363 6.4.1 The photovoltaic e®ect, 363 6.4.2 Materials and applications, 365 6.5 Electrochemical Sensors, 366 6.6 Problems, 369 References, 373 7 Signal Conditioning for Self-Generating Sensors 375 7.1 Chopper and Low-Drift AmpliÆers, 376 7.1.1 O®set and drifts in op amps, 376 7.1.2 Chopper ampliÆers, 383
CONTENTS ix 7.1.3 Autozero ampliÆers, 384 7.1.4 Composite ampliÆers, 386 7.1.5 O®set and drifts in instrumentation ampliÆers, 387 7.2 Electrometer and Transimpedance AmpliÆers, 388 7.2.1 Transimpedance ampliÆers, 391 7.2.2 Current measurement by integration, 394 7.2.3 Cautions in designing electrometer circuits, 395 7.3 Charge AmpliÆers, 397 7.4 Noise in AmpliÆers, 403 7.4.1 Noise fundamentals, 403 7.4.2 Noise in op amps, 407 7.4.3 Noise in transimpedance ampliÆers, 416 7.4.4 Noise in charge ampliÆers, 418 7.4.5 Noise in instrumentation ampliÆers, 419 7.5 Noise and Drift in Resistors, 421 7.5.1 Drift in Æxed resistors, 421 7.5.2 Drift in adjustable resistors (potentiometers), 424 7.5.3 Noise in resistors, 425 7.6 Problems, 427 References, 432 8 Digital and Intelligent Sensors 433 8.1 Position Encoders, 433 8.1.1 Incremental position encoders, 434 8.1.2 Absolute position encoders, 441 8.2 Resonant Sensors, 445 8.2.1 Sensors based on quartz resonators, 447 8.2.2 SAW sensors, 451 8.2.3 Vibrating wire strain gages, 453 8.2.4 Vibrating cylinder sensors, 455 8.2.5 Digital Øowmeters, 456 8.3 Variable Oscillators, 458 8.3.1 Sinusoidal oscillators, 459 8.3.2 Relaxation oscillators, 460 8.3.3 Variable CMOS oscillators, 463 8.3.4 Linearity in variable oscillators, 465 8.4 Conversion to Frequency, Period, or Time Duration, 467 8.4.1 Voltage-to-frequency conversion, 468 8.4.2 Direct quantity-to-frequency conversion, 470 8.4.3 Direct quantity-to-time duration conversion, 474 8.5 Direct Sensor±Microcontroller Interfacing, 476 8.5.1 Frequency measurement, 476 8.5.2 Period and time-interval measurement, 478 8.5.3 Calculations and compensations, 482
x CONTENTS 8.5.4 Velocity measurement. Digital tachometers, 484 8.6 Communication Systems for Sensors, 486 8.6.1 Current telemetry: 4 to 20 mA loop, 487 8.6.2 Simultaneous analog and digital communication, 489 8.6.3 Sensor buses: Fieldbus, 490 8.7 Intelligent Sensors, 492 8.8 Problems, 494 References, 498 9 Other Sensing Methods 501 9.1 Sensors Based on Semiconductor Junctions, 501 9.1.1 Thermometers based on semiconductor junctions, 502 9.1.2 Magnetodiodes and magnetotransistors, 508 9.1.3 Photodiodes, 509 9.1.4 Position-sensitive detectors (PSDs), 518 9.1.5 Phototransistors, 519 9.1.6 Semiconductor-junction nuclear radiation detectors, 521 9.2 Sensors Based on MOSFET Transistors, 522 9.3 Charge-Coupled and CMOS Image Sensors, 525 9.3.1 Fundamentals, 525 9.3.2 Types of CCD and CMOS imaging sensors and applications, 529 9.4 Fiber-Optic Sensors, 533 9.4.1 Fiber-optic basics, 533 9.4.2 Fiber-optic sensor technologies and applications, 535 9.5 Ultrasonic-Based Sensors, 538 9.5.1 Fundamentals of ultrasonic-based sensors, 539 9.5.2 Ultrasonic-based sensing methods and applications, 541 9.6 Biosensors, 544 9.7 Problems, 546 References, 550 Appendix: Solutions to the Problems 553 Index 571
1 INTRODUCTION TO SENSOR-BASED MEASUREMENT SYSTEMS Measurements pervade our life. Industry, commerce, medicine, and science rely on measurements. Sensors enable measurements because they yield electric sig- nals with embedded information about the measurand. Electronic circuits pro- cess those signals in order to extract that information. Hence, sensors are the basis of measurement systems. This chapter describes the basics of sensors, their static and dynamic characteristics, primary sensors for common quantities, and sensor materials and technology. 1.1 GENERAL CONCEPTS AND TERMINOLOGY 1.1.1 Measurement Systems A system is a combination of two or more elements, subsystems, and parts necessary to carry out one or more functions. The function of a measurement system is the objective and empirical assignment of a number to a property or quality of an object or event in order to describe it. That is, the result of a measurement must be independent of the observer (objective) and experimen- tally based (empirical). Numerical quantities must fulÆll the same relations ful- Ælled by the described properties. For example, if a given object has a property larger than the same property in another object, the numerical result when measuring the Ærst object must exceed that when measuring the second object. One objective of a measurement can be process monitoring: for example, ambient temperature measurement, gas and water volume measurement, and clinical monitoring. Another objective can be process control: for example, for temperature or level control in a tank. Another objective could be to assist 1
2 1 INTRODUCTION TO SENSOR-BASED MEASUREMENT SYSTEMS Figure 1.1 Functions and data Øow in a measurement and control system. Sensors and actuators are transducers at the physical interface between electronic systems and pro- cesses or experiments. experimental engineering: for example, to study temperature distribution inside an irregularly shaped object or to determine force distribution on a dummy driver in a car crash. Because of the nature of the desired information and its quantity, computer-aided design (CAD) does not yield complete data for these experiments. Thus measurements in prototypes are also necessary to verify the results of computer simulations. Figure 1.1 shows the functions and data Øow of a measurement and control system. In general, in addition to the acquisition of information carried out by a sensor, a measurement requires the processing of that information and the presentation of the result in order to make it perceptible to human senses. Any of these functions can be local or remote, but remote functions require infor- mation transmission. Modern measurement systems are not physically arranged according to the data Øow in Figure 1.1 but are instead arranged according to their connection to the digital bus communicating di®erent subsystems (Sec- tions 8.6 and 8.7). 1.1.2 Transducers, Sensors, and Actuators A transducer is a device that converts a signal from one physical form to a corresponding signal having a di®erent physical form. Therefore, it is an energy converter. This means that the input signal always has energy or power; that is, signals consist of two component quantities whose product has energy or power dimension. But in measurement systems, one of the two components of the measured signal is usually so small that it is negligible, and thus only the re- maining component is measured.
1.1 GENERAL CONCEPTS AND TERMINOLOGY 3 When measuring a force, for example, we assume that the displacement in the transducer is insigniÆcant. That is, that there is no ``loading'' e®ect. Other- wise it might happen that the measured force is unable to deliver the needed energy to allow the movement. But there is always some power taken by the transducer, so we must ensure that the measured system is not perturbed by the measuring action. Since there are six di®erent kinds of signals–mechanical, thermal, magnetic, electric, chemical, and radiation (corpuscular and electromagnetic, including light)–any device converting signals of one kind to signals of a di®erent kind is a transducer. The resulting signals can be of any useful physical form. Devices o®ering an electric output are called sensors. Most measurement systems use electric signals, and hence rely on sensors. Electronic measurement systems provide the following advantages: 1. Sensors can be designed for any nonelectric quantity, by selecting an appropriate material. Any variation in a nonelectric parameter implies a variation in an electric parameter because of the electronic structure of matter. 2. Energy does not need to be drained from the process being measured because sensor output signals can be ampliÆed. Electronic ampliÆers yield (low) power gains exceeding 1010 in a single stage. The energy of the ampliÆer output comes from its power supply. The ampliÆer input signal only controls (modulates) that energy. 3. There is a variety of integrated circuits available for electric signal con- ditioning or modiÆcation. Some sensors integrate these conditioners in a single package. 4. Many options exist for information display or recording by electronic means. These permit us to handle numerical data and text, graphics, and diagrams. 5. Signal transmission is more versatile for electric signals. Mechanical, hydraulic, or pneumatic signals may be appropriate in some circum- stances, such as in environments where ionizing radiation or explosive atmospheres are present, but electric signals prevail. Sensor and transducer are sometimes used as synonymous terms. However, sensor suggests the extension of our capacity to acquire information about physical quantities not perceived by human senses because of their subliminal nature or minuteness. Transducer implies that input and output quantities are not the same. A sensor may not be a transducer. The word modiÆer has been proposed for instances where input and output quantities are the same, but it has not been widely accepted. The distinction between input-transducer (physical signal/electric signal) and output-transducer (electric signal/display or actuation) is seldom used at pres- ent. Nowadays, input transducers are termed sensors, or detectors for radiation,
4 1 INTRODUCTION TO SENSOR-BASED MEASUREMENT SYSTEMS and output transducers are termed actuators or e®ectors. Sensors are intended to acquire information. Actuators are designed mainly for power conversion. Sometimes, particularly when measuring mechanical quantities, a primary sensor converts the measurand into a measuring signal. Then a sensor would convert that signal into an electric signal. For example, a diaphragm is a pri- mary sensor that stresses when subject to a pressure di®erence, and strain gages (Section 1.7.2 and Section 2.2) sense that stress. In this book we will designate as sensor the whole device, including the package and leads. We must realize, however, that we cannot directly perceive signals emerging from sensors unless they are further processed. 1.1.3 Signal Conditioning and Display Signal conditioners are measuring system elements that start with an electric sensor output signal and then yield a signal suitable for transmission, display, or recording, or that better meet the requirements of a subsequent standard equipment or device. They normally consist of electronic circuits performing any of the following functions: ampliÆcation, level shifting, Æltering, impedance matching, modulation, and demodulation. Some standards call the sensor plus signal conditioner subsystem a transmitter. One of the stages of measuring systems is usually digital and the sensor output is analog. Analog-to-digital converters (ADCs) yield a digital code from an analog signal. ADCs have relatively low input impedance, and they require their input signal to be dc or slowly varying, with amplitude within speciÆed margins, usually less than G10 V. Therefore, sensor output signals, which may have an amplitude in the millivolt range, must be conditioned before they can be applied to the ADC. The display of measured results can be in an analog (optical, acoustic, or tactile) or in a digital (optical) form. The recording can be magnetic, electronic, or on paper, but the information to be recorded should always be in electrical form. 1.1.4 Interfaces, Data Domains, and Conversion In measurement systems, the functions of signal sensing, conditioning, pro- cessing, and display are not always divided into physically distinct elements. Furthermore, the border between signal conditioning and processing may be indistinct. But generally there is a need for some signal processing of the sensor output signal before its end use. Some authors use the term interface to refer to signal-modifying elements that operate in the electrical domain, even when changing from one data domain to another, such as an ADC. A data domain is the name of a quantity used to represent or transmit information. The concept of data domains and conversion between domains helps in describing sensors and electronic circuits associated with them [1]. Figure 1.2 shows some possible domains, most of which are electrical. In the analog domain the information is carried by signal amplitude (i.e.,
1.1 GENERAL CONCEPTS AND TERMINOLOGY 5 Figure 1.2 Data domains are quantities used to represent or transmit information [1]. (From H. V. Malmstadt, C. G. Enke, and S. R. Crouch, Electronics and Instrumentation for Scientists, copyright 1981. Reprinted by permission of Benjamin/Cummings, Menlo Park, CA.) charge, voltage, current, or power). In the time domain the information is not carried by amplitude but by time relations (period or frequency, pulse width, or phase). In the digital domain, signals have only two values. The information can be carried by the number of pulses or by a coded serial or parallel word. The analog domain is the most prone to electrical interference (Section 1.3.1). In the time domain, the coded variable cannot be measured–that is, converted to the numerical domain–in a continuous way. Rather, a cycle or pulse duration must elapse. In the digital domain, numbers are easily displayed. The structure of a measurement system can be described then in terms of domain conversions and changes, depending on the direct or indirect nature of the measurement method. Direct physical measurements yield quantitative information about a physical object or action by direct comparison with a reference quantity. This compari- son is sometimes simply mechanical, as in a weighing scale. In indirect physical measurements the quantity of interest is calculated by applying an equation that describes the law relating other quantities measured with a device, usually an electric one. For example, one measures the mechan- ical power transmitted by a shaft by multiplying the measured torque and speed of rotation, the electric resistance by dividing dc voltage by current, or the traveled distance by integrating the speed. Many measurements are indirect.
6 1 INTRODUCTION TO SENSOR-BASED MEASUREMENT SYSTEMS 1.2 SENSOR CLASSIFICATION A great number of sensors are available for di®erent physical quantities. In order to study them, it is advisable Ærst to classify sensors according to some criterion. White [10] provides additional criteria to those used here. In considering the need for a power supply, sensors are classiÆed as modu- lating or self-generating. In modulating (or active) sensors, most of the output signal power comes from an auxiliary power source. The input only controls the output. Conversely, in self-generating (or passive) sensors, output power comes from the input. Modulating sensors usually require more wires than self-generating sensors, because wires di®erent from the signal wires supply power. Moreover, the presence of an auxiliary power source can increase the danger of explosion in explosive atmospheres. Modulating sensors have the advantage that the power supply voltage can modify their overall sensitivity. Some authors use the terms active for self-generating and passive for modulating. To avoid confusion, we will not use these terms. In considering output signals, we classify sensors as analog or digital. In analog sensors the output changes in a continuous way at a macroscopic level. The information is usually obtained from the amplitude, although sensors with output in the time domain are usually considered as analog. Sensors whose output is a variable frequency are called quasi-digital because it is very easy to obtain a digital output from them (by counting for a time). The output of digital sensors takes the form of discrete steps or states. Digital sensors do not require an ADC, and their output is easier to transmit than that of analog sensors. Digital output is also more repeatable and reliable and often more accurate. But regrettably, digital sensors cannot measure many physical quantities. In considering the operating mode, sensors are classiÆed in terms of their function in a deØection or a null mode. In deØection sensors the measured quantity produces a physical e®ect that generates in some part of the instru- ment a similar but opposing e®ect that is related to some useful variable. For example, a dynamometer to measure force is a sensor where the force to be measured deØects a spring to the point where the force it exerts, proportional to its deformation, balances the applied force. Null-type sensors attempt to prevent deØection from the null point by ap- plying a known e®ect that opposes that produced by the quantity being mea- sured. There is an imbalance detector and some means to restore balance. In a weighing scale, for example, the placement of a mass on a pan produces an imbalance indicated by a pointer. The user has to place one or more calibrated weights on the other pan until a balance is reached, which can be observed from the pointer's position. Null measurements are usually more accurate because the opposing known e®ect can be calibrated against a high-precision standard or a reference quan- tity. The imbalance detector only measures near zero; therefore it can be very
1.3 GENERAL INPUT±OUTPUT CONFIGURATION 7 TABLE 1.1 Sensor ClassiÆcations According to Di®erent Exhaustive Criteria Criterion Classes Examples Power supply Modulating Thermistor Self-generating Thermocouple Output signal Analog Potentiometer Digital Position encoder Operation mode DeØection DeØection accelerometer Null Servo-accelerometer sensitive and does not require any calibration. Nevertheless, null measurements are slow; and despite attempts at automation using a servomechanism, their response time is usually not as short as that of deØection systems. In considering the input±output relationship, sensors can be classiÆed as zero, Ærst, second, or higher order (Section 1.5). The order is related to the number of independent energy-storing elements present in the sensor, and this a®ects its accuracy and speed. Such classiÆcation is important when the sensor is part of a closed-loop control system because excessive delay may lead to oscillation [6]. Table 1.1 compares the classiÆcation criteria above and gives examples for each type in di®erent measurement situations. In order to study these myriad devices, it is customary to classify them according to the measurand. Conse- quently we speak of sensors for temperature, pressure, Øow, level, humidity and moisture, pH, chemical composition, odor, position, velocity, acceleration, force, torque, density, and so forth. This classiÆcation, however, can hardly be exhaustive because of the seemingly unlimited number of measurable quanti- ties. Consider, for example, the variety of pollutants in the air or the number of di®erent proteins inside the human body whose detection is of interest. Electronic engineers prefer to classify sensors according to the variable electrical quantity–resistance, capacity, inductance–and then to add sensors generating voltage, charge, or current, and other sensors not included in the preceding groups, mainly p±n junctions and radiation-based sensors. This approach reduces the number of groups and enables the direct study of the associated signal conditioners. Table 1.2 summarizes the usual sensors and sensing methods for common quantities. 1.3 GENERAL INPUT±OUTPUT CONFIGURATION 1.3.1 Interfering and Modifying Inputs In a measurement system the sensor is chosen to gather information about the measured quantity and to convert it to an electric signal. A priori it would be unreasonable to expect the sensor to be sensitive to only the quantity of interest
TABLE 1.2 Usual Sensors and Sensing Methods for Common Quantities Acceleration Flow Rate Humidity Sensor type Vibration Point velocity Force Moisture Resistive Mass±springá Anemometer Strain gage Humistor strain gage Thermistor Targetástrain gage Capacitive Mass±springávaria- Capacitive strain Dielectric- ble capacitor gage variation capacitor Inductive and Mass±springáLVDT Faraday's law Load celláLVDT electro- RotameteráLVDT Magnetostriction magnetic Self-generating Mass±springápiezo- Thermal transportá Piezoelectric sensor electric sensor thermocouple Digital Impeller, turbine SAW sensor Positive displacement Vortex shedding PN junction Optic, Æber Laser anemometry Chilled mirror optic Ultrasound Doppler e®ect Travel time Vortex Other Di®erential pressure Variable areaálevel sensor (open channel) Variable areaádis- placement Coriolis e®ectáforce
Quantity Position Distance Velocity Level Displacement Pressure Temperature Speed Floatá Magnetoresistor Bourdon tubeá RTD potentiometer potentiometer LDR Potentiometer Diaphragmástrain gage Thermistor Thermistor Strain gage Variable capacitor Di®erential Diaphragmávariable capacitor capacitor Magnetostriction Eddy currents DiaphragmáLVDT Eddy currents Magnetoresistive Hall e®ect Diaphragmá variable Hall e®ect reluctance Float á LVDT Inductosyn Faraday's law Eddy currents LVDT LVT Resolver, synchro Magnetostriction Piezoelectric sensor Pyroelectric sensor Thermocouple Vibrating rod Position encoder Bourdon tubeáencoder Quartz oscillator Incremental encoder Floatápulley Bourdon tube or bellowsáquartz resonator Diaphragmá vibrating wire Photoelectric Photoelectric sensor Diode Bipolar transistor T/I converter Diaphragmálight reØection Absorption Travel time Doppler e®ect Travel time Di®erential Liquid-based manometer pressure álevel sensor Microwave radar Nuclear radiation
10 1 INTRODUCTION TO SENSOR-BASED MEASUREMENT SYSTEMS Figure 1.3 E®ect of internal and external perturbations on measurement systems. xS is the signal of interest. yÖtÜ is the system output. xI is an interference or external pertur- bation. xM is a modifying input. (From E. O. Doebelin, Measurement Systems Applica- tion and Design, 4th ed., copyright 1990. Reprinted by permission of McGraw-Hill, New York.) and also to expect the output signal to be entirely due to the input signal. No measurement is ever obtained under ideal circumstances; therefore we must address real situations. We follow here the method proposed by Doebelin [2]. Figure 1.3 shows a general block diagram for classifying desired signal gains and interfering input gain for instruments. The desired signal xS passes through the gain block GS to the output y. Interfering inputs xI represent quantities to which the instrument is unintentionally sensitive. These pass through the gain block GI to the output y. Modifying inputs xM are the quantities that through GM;S cause a change in GS for the desired signal and through GM;I cause a change in GI for interfering inputs. The gains G can be linear, nonlinear, varying, or random. For example, to measure a force, it is common to use strain gages (Section 2.2). Strain gages operate on the basis of variation in the electric resistance of a conductor or semiconductor when stressed. Because temperature change also yields a resistance variation, we can regard any temperature variation as an interference or external disturbance xI with gain GI. At the same time, to mea- sure resistance changes as a result of the stress, an electronic ampliÆer is re- quired. Since any temperature change xM through GM;S a®ects the ampliÆer gain GS and therefore the output, it turns out that a temperature variation also acts as a modifying input xM. If the same force is measured with a capacitive gage (Section 4.1), a temperature variation does not interfere but can still modify the ampliÆer gain.
1.3 GENERAL INPUT±OUTPUT CONFIGURATION 11 Figure 1.4 (a) Negative feedback method to reduce the e®ect of internal perturbations. Block H may be insensitive to those perturbations because it handles lower power than block G. (b) Force-to-current converter that relies on negative feedback and a balance sensor. 1.3.2 Compensation Techniques The e®ects of interfering and modifying inputs can be reduced by changing the system design or by adding new elements to it. The best approach is to design systems insensitive to interference and that respond only to the desired signals. In the preceding example, it would have been best to use strain gages with a low temperature coeΩcient ÖGI à 0Ü. Thin, narrow, long magnetic sensors are only sensitive to magnetic Æelds parallel to their long dimension. In designing sensors for vector mechanical quantities, it would be best to obtain a unidirectional sensitivity and a low transverse sensitivity–that is, in directions perpendicular to the desired direction. In electronic circuits, low-drift components such as metal-Ælm resistors and NP0 capacitors are less sensitive to temperature. Nevertheless, this method is not always possible for obvious practical reasons. Negative feedback is a common method to reduce the e®ect of modifying inputs, and it is the method used in null measurement systems. Figure 1.4a shows the working principle. It assumes that the measurement system and the feedback are linear and can be described by their respective transfer functions GÖsÜ and HÖsÜ. The input±output relation is YÖsÜ GÖsÜ 1 à G Ö1:1Ü XÖsÜ 1 á GÖsÜHÖsÜ HÖsÜ where the approximation is valid when GÖsÜHÖsÜg1. If the negative feedback is insensitive to the modifying input, and it has been designed so that the system remains stable, then the output signal is not a®ected by the modifying input. The advantage of such a solution stems from the di®erent physical charac- teristics of the elements described by GÖsÜ and HÖsÜ. The probable insensitivity of H to a modifying input is a consequence of its lower power-handling capac-
Hiển thị 20 trang đầu. Tải tài liệu để xem đầy đủ.
Chưa có bình luận nào. Hãy là người đầu tiên!
Bình luận (0)
Chưa có bình luận nào. Hãy là người đầu tiên!