Temperature Measurement Second Edition

L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee

Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

Temperature Measurement
Second Edition
Temperature Measurement
Second Edition

L. Michalski
K. Eckersdorf
J. Kucharski
Technical University ofLodz
Poland

J. McGhee
University of Strathclyde
Glasgow, UK

JOHN WILEY & SONS, LTD

Chichester " Weinheim " New York " Brisbane " Singapore " Toronto
Copyright @ 2001 by John Wiley & Sons Ltd,
Baffins Lane, Chichester,
West Sussex P019 IUD, England

National 01243 779777

International (+44) 1243 779777

e-mail (for orders and customer service enquiries) : cs-books@wiley .co .uk

Visit our Home Page on http ://www .wiley.co .u k

or
http ://www.wiley.com

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

in a retrieval system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, recording, scanning or otherwise, except under the terms
of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued
by the Copyright Licensing Agency, 90 Tottenham Court Road, London, UK W 1 P 9 HE,
without the permission in writing of the publisher.

Other Wiley Editorial Offices

John Wiley & Sons, Inc ., 605 Third Avenue,

New York, NY 10158-0012, USA

Wiley-VCH Verlag GmbH, Pappelallee 3,

D-69469 Weinheim, Germany

Jacaranda Wiley Ltd, 33 Park Road, Milton,

Queensland 4064, Australia

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01,
Jin Xing Distripark, Singapore 0512

John Wiley & Sons (Canada) Ltd, 22 Worcester Road,

Rexdale, Ontario M9W 1L1, Canada

Library of Congress Cataloging-in-Publication Data

Temperature measurement / L . Michalski . . . [et al.) . - 2"d ed .
p . cm . - (Wiley series in measurement science and technology)
Rev . ed . of: Temperature Measurement / L. Michalski, K . Eckersdorf and J . McGhee . c 1991 .
Includes bibliographical references and indexes .
ISBN 0 471 86779 9
1 . Temperature Measurements . . I . Michalski, L . II Michalski, L .Pomiary temperatury.

QC271 . M483 2000

.536' . 5'0287 - dc21 2000-043373

British Library Cataloguing in Publication Data

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

ISBN 0 471 86779 9

Produced from Word files supplied by the author

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham, Wilts
This book is printed on acid-free paper responsibly manufactured from sustainable
forestry, in which at least two trees are planted for each one used for paper production .
Contents

PREFACE xiii
LIST OF PRINCIPAL SYMBOLS xiv

1. Temperature Scales and Classification of Thermometers 1

1 .1 Temperature - Historical Background 1
1 .2 Thermodynamic Temperature Scale 3
1 .3 International Temperature Scales 6
1 .3 .1 From the Normal Hydrogen Scale to EPT-76 6
1 .3.2 The International Temperature Scale of 1990 (ITS-90) 9
1 .4 Classification of Thermometers 12
1 .4.1 Temperature measuring chains 12
1 .4.2 General principles for thermometer classification 13
1 .4 .3 The non-electrical contacting temperature sensors 14
1 .4.4 The electrical contacting temperature sensors 15
1 .4.5 The non-contacting group of temperature sensors 15
1 .4 .6 Temperature measuring ranges oftemperature sensors 17
1 .5 References 18

2. Non-Electric Thermometers 19
2.1 Liquid-in-Glass Thermometers 19
2.2 Thermometers Using Expansion of Solids 23
2 .2.1 Dilatation thermometers 23
2 .2.2 Bimetallic thermometers 25
2.3 Manometric Thermometers 28
2 .3.1 Liquid-filled thermometers 28
2 .3.2 Vapour-pressure thermometers 31
2 .3.3 Gas thermometers 32
2 .3.4 Summary of properties of manometric thermometers 32
2.4 Temperature Indicators 32
2.5 References 35

3. Thermoelectric Thermometers 37
3 .1 Physical Principles 37
3 .1 .1 Thermoelectric force 37
3 .1 .2 Law ofthe third metal 39
3 .1 .3 Law of consecutive metals 41
3 .1 .4 Law of consecutive temperatures 43
3.2 Thermocouples 44
3 .2 .1 General information 44

v
vi CONTENTS

3 .2.2 Properties of commonly used thermocouples 46

3 .2.3 Measuring junctions 50
3.3 Thermocouple Sensors 52
3 .3.1 Construction 52
3 .3.2 Sheath materials 53
3 .3.3 Review ofstructures 57
3.4 High Temperature Thermocouples 65
3.5 Compensating Cables 67
3.6 Reference Temperature 72
3 .6.1 Calculation of correction 72
3 .6.2 Stabilisation and correction techniques 74
3.7 Measuring Circuits 76
3 .7.1 Introduction 76
3 .7.2 Deflection type measuring circuits 77
3 .7.3 Potentiometers 78
3 .8 References 82

4. Resistance Thermometers 85
4 .1 General and Historical Background 85
4.2 Physical Principles 86
4.3 Resistance Thermometer Detectors (RTDs) 88
4.3.1 General information 88
4.3.2 Properties of different metals 89
4.3.3 Construction 90
4.4 Resistance Thermometer Sensors 94
4.5 Measuring Circuits 96
4.5.1 Voltage divider circuits 96
4.5.2 Bridge circuits 98
4.5 .3 Quotient instruments 101
4.6 References 102

5. Semiconductor Thermometers 103

5.1 Classification of Semiconductor Thermometers 103
5.2 Thermistor Thermometers 104
5.2 .1 Principles of operation 104
5.2.2 Thermistor sensors 109
5.2 .3 Correction and linearisation of thermistor characteristics 111
5.2 .4 Measuring circuits 112
5.3 Silicon Resistance Thermometers 114
5.4 Diode and Transistor Thermometers 116
5.4.1 Principles of operation 116
5.4.2 Diode thermometers 118
5.4 .3 Transistor thermometers 119
5.5 Integrated Circuit Thermometers 121
5.6 References 122
CONTENTS vii

6. Fibre Optic Thermometers 125

6.1 Properties of Optical Fibres 125
6.2 Classification of Fibre Optic Thermometers 126
6 .3 Extrinsic Sensing Thermometers 127
6.3 .1 Thermometers with GaAs semiconductor sensors 127
6.3 .2 Thermochromic thermometers 128
6.3 .3 Fluorescent thermometers 129
6.3 .4 Thermometers with black body sensors 132
6.3 .5 Thermometers with Fabry-Perot sensors 133
6.3 .6 Optical coupling thermometers 134
6 .4 Intrinsic Sensing Thermometers 134
6.4 .1 Raman scattering thermometers 134
6.4.2 Refractive index thermometers 135
6.5 Summary 136
6 .6 References 137

7. Quartz, Ultrasonic and Noise Thermometers and Distributed Parameter

Sensors 139
7.1 Quartz Thermometers 139
7.2 Ultrasonic Thermometers 140
7 .3 Noise Thermometers 144
7.4 Distributed Parameter Sensors 148
7.5 References 149

S. Pyrometers Classification and Radiation Laws 151

8.1 Classification ofPyrometers 151
8.2 Radiation, Definitions and Laws 154
8.2.1 Absorption, reflection and transmission of radiation 154
8 .2.2 Radiation laws 155
8 .2.3 Total emissivity and spectral emissivity 159
8.2 .4 Radiant heat exchange 160
8.3 References 162

9. Manually Operated Pyrometers 163

9.1 Disappearing Filament Pyrometers 163
9.1 .1 Principle ofoperation 163
9.1 .2 Red filter 164
9 .1 .3 Scale defining equation for black bodies 166
9 .1 .4 Temperature measurement ofnon-black bodies 167
9.1 .5 Extension of measurement range 169
9 .1 .6 Applications and construction 170
9.2 Two-Colour Pyrometers 172
9.2.1 General information 172
9 .2.2 Scale defining equation 172
9 .2.3 Temperature measurement of non-grey bodies 174
9 .3 References 175
Viii CONTENTS

10. Automatic Pyrometers 177

10.1 Optical Systems 177
10.1.1 Lenses 177
10.1 .2 Light guides 179
10.1 .3 Mirrors 180
10.2 Radiation Detectors 181
10.2 .1 Thermal detectors 181
10.2.2 Photoelectric detectors 182
10.3 Total Radiation Pyrometers 184
10.3 .1 General information 184
10 .3.2 Scale defining equation for black bodies 185
10.3.3 Temperature measurement of non-black bodies 186
10.3 .4 Influence of housing temperature 189
10 .3.5 Influence of target distance 189
10.3.6 Extension of measurement range 189
10.3.7 Review of construction 190
10.4 Photoelectric Pyrometers 192
10.4.1 General information 192
10.4.2 Scale defining equation for black bodies 195
10.4.3 Temperature measurement of non-black bodies 196
10.4.4 Review of construction 198
10.5 Two-Wavelength Pyrometers 201
10 .5.1 General information 201
10.5 .2 Review ofconstruction 202
10.6 Multi-Wavelength Pyrometers 204
10.6.1 General information 204
10.6.2 Review ofconstruction 206
10 .7 References 207

11 . Practical Applications of Pyrometers 209

11 .1 Introduction 209
11.2 Influence ofTarget Emissivity 210
11 .2 .1 Calculation oftrue temperature 210
11 .2.2 Methods of approaching black-body conditions 212
11 .2.3 Other methods 213
11 .3 Influence of Surrounding Walls 216
11 .4 Influence of Atmosphere 220
11 .4.1 Methods of eliminating and limiting the errors caused by atmosphere
absorption 222
11 .5 Influence of Other Factors 22
11 .6 Summary of Properties and Applications 223
11 .7 References 228

12. Conditioning of Temperature Sensor Output Signals 229

12.1 Introduction 229
12.2 Methods of Signal Processing in Temperature Measurement 230
12.2.1 Transformation of signal nature 231
12.2.2 Transformation of signal scale 234
CONTENTS ix

12.2.3 Transformation of signal shape 235

12 .3 Transmission of Measured Signals 244
12 .4 Multi-Channel Temperature Measurements 245
12 .5 Transmitters 247
12 .6 Indicating Instruments and Recorders 250
12.6.1 Indicating instruments 250
12.6.2 Recorders 252
12.7 Temperature Controllers 256
12 .8 References 257

13. Computerised Temperature Measuring Systems 259

13 .1 Trends in Process Instrumentation 259
13.2 Hardware for Computerised Measurement 260
13 .2 .1 External measuring modules 260
13 .2 .2 Plug-in data acquisition cards 264
13.3 Software for Computerised Measurement 269
13.4 References 272

14. Imaging of Temperature Fields of Solids 273

14 .1 Introduction 273
14.2 Surfacial Systems 273
14.3 Linear Systems 276
14 .4 Applications 277
14.5 References 278

15. Dynamic Temperature Measurement 279

15.1 General Information 279
15 .1 .1 Dynamic errors 280
15 .1 .2 Dynamic properties of temperature sensors 281
15.2 Idealised Sensor 284
15 .2.1 Transfer function 285
15 .2 .2 Measurement of time varying temperature 287
15.3 Real Sensors 291
15 .3.1 Sensor design 291
15 .3.2 Changing heat transfer coefficient 293
15 .3.3 Equivalent transfer function 294
15 .3.4 Calculation of dynamic properties of sensors 296
15.4 Experimental Determination of the Dynamic Properties
of Sensors 299
15 .4.1 Classification and application of the methods 299
15 .4 .2 External input, time domain testing with convective heat
transfer 300
15 .4.3 External input, frequency domain testing with convective heat
transfer 307
15 .4.4 External input with radiative heat transfer 308
15 .4.5 Internal input method 311
15.5 Dynamic Properties of Chosen Sensors 313
15.6 Correction of Dynamic Error s 320
x CONTENTS

15 .6.1 Principles of correction 320

15 .6.2 Analogue correction 321
15 .6.3 Digital correction 325
15.7 References 329

16. Temperature Measurement of Solid Bodies by Contact Method 333

16.1 Introduction 333
16.2 Theory of the Contact Method 333
16.2.1 Disturbing temperature field 334
16 .2 .2 Heat flux entering the sensor 338
16.2.3 Method errors and their reduction 342
16.2.4 Influence of thermal properties of bodies on errors 347
16.3 Sensors for Surface Temperature Measurement
16.3 .1 Portable contact sensors 348
16.3.2 Fixed contact sensors 351
16.3.3 Performance comparison of different sensors 352
16.4 Quasi-Contact Method 354
16.5 Extrapolation Method 355
16.6 Measurement of Internal Temperature of Solid Bodies 357
16.7 References 359

17. Temperature Measurement of Fluids 361

17.1 Low Velocity Gas 361
17.1 .1 Contact sensors 361
17.1 .2 Methods of reducing errors in contact measurements 365
17.1 .3 Indirect pyrometric measurements 371
17.1 .4 Direct pyrometric measurements 371
17.2 High Velocity Gas 372
17.3 Still Air 374
17.4 Liquids 375
17.5 High Temperature Gas and Plasma 375
17.6 References 378

18. Temperature Measurement of Transparent Solid Bodies 381

18.1 Pyrometric, Contactless Method 381
18.2 Contact Methods 386
18.3 References 386

19. Temperature Measurement of Moving Bodies 387

19.1 Introduction 387
19 .2 Pyrometric Contactless Methods 387
19.3 Sliding Contact Method 390
19.4 Inductive Circuits 390
19.5 Wireless Systems 391
19.6 Friction Sensors and `Quasi-Contactless' Method 392
19.7 Other Methods 394
19.8 References 396
CONTENTS xi

20. Temperature Measurement in Industrial Appliances 397

20.1 Introduction 397
20.2 Chamber Furnaces 397
20.2.1 General information 397
20 .2.1 Internal furnace temperature 398
20.2.3 Charge temperature 399
20.2.4 Measurements for temperature control 399
20.3 Continuous Furnaces 401
20.4 Salt-Bath Furnaces 403
20.5 Glass Tank Furnaces 404
20.6 Induction Heated Charges 405
20 .7 Dielectric Heated Charges 407
20.8 Molten Metals 408
20.9 Temperature Measurement in Hazardous Areas 409
20.10 References 410

21. Temperature Measurement in Medicine 413

21 .1 Introduction 413
21 .2 Measurement for Diagnosis 413
21 .3 Therapeutic and Diagnostic Measurement Using High Frequency Magnetic
Fields 416
21 .4 Other Measurements 417
21 .5 References 418

22. Calibration and Testing of Temperature Measuring

Instruments 419
22.1 Definitions and Terminology 419
22.2 Fixed Points of ITS-90 420
22.2.1 General information 420
22.2.2 Realisation of fixed points 421
22.3 Primary Standards 425
22.4 Working Standards 429
22.5 Testing ofIndustrial Thermometers 431
22.5.1 Introduction 431
22 .5.2 Variable volume thermometers 431
22.5.3 Resistance thermometer sensor 432
22.5.4 Thermocouples 433
22 .5.5 Disappearing filament pyrometers 436
22 .5 .6 Total radiation, photoelectric, two-colour and
two-wavelength pyrometers 437
22.6 Auxiliary Equipment 437
22.6.1 Fixed points 437
22.6.2 Thermometer testing baths 439
22.6.3 Testing furnaces 441
22 .6.4 Black bodies 445
22.7 Calibrators 447
22.8 References 447
xii CONTENTS

AUXILIARY TABLES 453

AUTHOR AND ORGANISATION INDEX 471
SUBJECT INDEX 479
Preface to the 2nd Edition
Temperature influences all natural physical phenomena as well as physiological,
technological and thermal processes . It is one of the most important parameters in any kind
of research .
From the vast and ever growing number of possible methods of temperature
measurement, that best suited for each application should be chosen in order to attain
readings which are as precise as possible . As the errors of the method are more important in
most cases than those of the instrumentation, this book concentrates on the analysis of the
different temperature measurement methods and sources of errors .
The scope of the book covers all of the principal temperature measurement methods and
instruments . Special stress is placed on such problems as temperature measurement of
solids, liquids and gases as well as temperature measurement in industrial heating
appliances.
Many new achievements, additions and developments, as well as literature references,
are included to update a most of the material contained in the 1-st Edition of 1991 . This
updating has also lead to eliminating the descriptions of some of the obsolete and
obsolescent, or less popular methods and techniques of temperature measurement .
Some chapters have been extended or reorganised in view of the developing perception
of the grouping in temperature measurement. Fibre optic and noise thermometry are
extensively described together with distributed parameter sensors . Major reorganisation of
the presentation of the theory and applications of optical pyrometry has been undertaken. A
new chapter, devoted to practical applications of pyrometers, has been introduced .
Other chapters are thoroughly revised versions of those from the 1st Edition or perhaps
completely new . One completely new chapter considers the application of computer
technique in temperature measurement . A second completely new chapter examines and
describes all of the problems associated with signal conditioning, hardware components,
hazardous area measurements and recording methods . Fuzzy logic together with software
methods for temperature measurement and the influence of the techniques of Virtual
Instrumentation are also described. Temperature measurement in medicine is an important
addition to this 2nd Edition.
A large number of numerical examples, tables and diagrams are given to provide
assistance in choosing and implementing the temperature measuring system best suited for a
particular application. Many references enable the reader to find supplementary information
regarding those aspects which could not be treated in detail in the book.
This book is intended for engineers, pure and applied scientists and student readerswho
wish to master the beautiful art of temperature measurement.
The authors extend their thanks to Prof W . Nawrocki and gratefully acknowledge his
contribution on noise thermometry .

Autumn 2001
The Authors
X1L
List of Principal Symbols
A amplitude, area
a thermal diffusivity
C radiation constant, electrical capacitance
c specific heat
D, d diameter
E thermal emf
e thermal emf in a junction
f frequency, function
G(s), F(s) transfer function
G (jco), F(jco) frequency response
I electric current
K gain
k general coefficient
L time lag also called dead time
I length
N time constant
P power
Q energy
q heat flux density
R resistance
r radius
s Laplace operator
T temperature in K
t time, temperature in °C
V voltage, volume
v velocity
W thermal resistance, radiant intensity
a heat transfer coefficient, coefficient of linear thermal expansion,
temperature coefficient of resistance, absorptivity
,li coefficient of cubic thermal expansion
A error, difference, amplitude, sampling time
8 relative error, penetration depth
s emissivity
9 temperature in °C or °F
O excess temperature over a reference temperature such as ambient or
original value
wavelength, thermal conductivity
P density, reflectivity, resistivity
heat flux or rate of heat flow
phase angle
xiv
LIST OF PRINCIPAL SYMBOLS xv

angular frequency
T transmissivity

Subscripts
a adjustable, ambient, average
C correction, corrector
C compensating, conduction
d disturbance
e effective, end-value, equivalent
gr grey body
i indicated, input value
k convection
l leads, limit value, loop
M measuring instrument
n nominal value
o black body, output value
r radiation, reference, reflection, relative
s set-point value, shield, solid
T temperature sensor
t true value
w wall
spectral
Subject Index
absolute temperature, 4 two-wavelength pyrometers, 153
absorptivity See two-wavelength pyrometers
band, 158 auxiliary testing equipment, 437
definition, 154 black bodies, 445
spectral, 158 cylindrical cavity radiation sources, 445
ADC, 263 flat plate radiation sources, 445
See also analogue-to-digital converters heat pipes, 446
analogue-to-digital conversion, 236 high temperature furnace, 446
analogue-to-digital converters middle and high temperature sources,
in a signal transmitter, 248 445
in strip chart recorders, 255 pyrometer fixed points calibration, 446
anti-Stokes light, 134 spherical cavity furnace, 446
associative temperature measurements, 4 calibrators, 447
automatic pyrometers, 177 applications, 447
application, 154 automatic reference junction
block diagram, 153 compensation, 447
distance ratio, 179 portable type, 447
fibreoptic fixed points, 437
operating principle, 180 boiling point of water, 437
lens materials for thermocouple tersting, 438
fluorite, 178 ice point, 437
KRS-5, 179 most commonly used (various), 438
pyrex glass, 178 laboratory type calibrators, 447
quartz, 178 testing furnaces, 441
silicon, 179 heat pipe, 444
spectral transmission of plates, 178 heat pipe materials, 444
synthetic sapphire, 178 spherical, 443
multi-wavelength, 154 tubular furnaces, 441
See multi-wavelength pyrometers whole instrumentation, 442
mirrors thermometer testing baths, 439
specific spectral reflectivities of metals, fluidised bed thermostat, 440
180 for higher temperature range, 439
optical systems, 177 liquid baths, 439
distance ratio examples, 179 water bath, 439
lens focusing, 179
lens transmission, 177
lenses, 177 bimetallic strips
light guides, 179 cylindrical helical, 28
mirrors, 180 bimetallic thermometers, 25
sighting window, 177 errors, 28
transmission limits of lens materials, 178 flat shaped strip, 26
photoelectric, 153. for surface temperature measurement, 28
See photoelectric pyrometers helix shaped strip, 27
ratio, 153 materials, 27
total radiation, 153 numerical example, 27
479
48 0 SUBJECT INDEX

structure, 25 ideal reversible, 4

typical forms, 26 Celsius temperature
U-shaped strip, 26 unit of, 9
using cylindrical helical strip, 28 t9o, 9
black body, 11 CGPM, 7,9
concept of, 155 chamber furnaces
models, 155 charge temperature
radiance, 162 calculated, 399
viewing angle, 168 charge temperature, 399
radiant charge temperature measurement
flux distribution, 156 contact methods, 399
intensity, 156, 157, 161 fibre optic thermometer with black body
radiation, 155 sensor, 399
concepts of, 6 pyrometric method, 399
constant, 159 emissivity of walls, 399
low velocity gas, 362 general information, 397
scale defining equation, 166 internal heat transfer, 398
spectral intensity, 156, 165 internal temperature measurement, 398
spectral radiance, 165 errors, 398
technical radiation constant, 159 measurement and control
total power radiated, 157 block diagram, 398
total radiant intensity, 165 temperature control
boiling point of water, 2, 5, 8 equivalent sensor transfer function, 400
boiling points, 7 influence of sensor dynamics, 400
Boltzmann's constant, 4, 128 measurement, 399
Boyle's law, 4 on-off, 400
Boyle-Mariotte law, 4 on-off quality, 400
P, PI and PID, 400
transfer function, 400
calibration and testing, 419 characteristic temperatures, 7
auxiliary equipment. See auxiliary testing Charles' law, 4
equipment classification
definitions and terminology, 419 heat transfer methods
calibration, 419 contacting, 13
industrial thermometers, 419 non-contacting, 13
laboratory thermometers, 419 semiconductors, 15
primary standards, 419 temperature signal conditioning, 231
testing, 419 thermometers, 12
transfer standards, 419 general principles, 13
working standards, 419 colour temperature
national laboratories (various), 420 indicated
primary standards. See primary standards numerical calculation, 175
sensors compensating cables
capsule type resistance, 420 circuit with thermocouple, 68
high temperature platinum resistance, colour codes, 69
420 national, 70
long stem platinum resistance, 420 composition, 69
pyrometers, 420 correct use of, 68
working standards. See working standards design of, 70
capillary, 33 electrical insulation, 70
axial, 22 errors, 68
vacuumised, 22 for Type K thermocouples, 70
Carrot cycle, 6 standard specifications, 69
SUBJECT INDEX 481

use of, 67 conversion

various types analogue-to-digital .
specifications, 71 See analogue-to-digital conversion
compensating leads converter
emf across, 68 4 to 20 mA
errors continuous semiconductor sensors, 149
from inadequate matching, 69 digital-to-digital, 233
Computer networks
ETHERNET, 260
TOKEN RING, 260 data acquisition, 259
computerised measurement computer based, 261, 269
hardware, 260 computerised, 259
external modules, 260 conversion rate of ADC, 261
signal conditioning modules, 261 distributed, 261
software, 269 high-speed, 269
diversity of packages, 271 PCMCIA, 269
functions, 269 external temperature module, 269
large scale processing systems, 270 plug-in cards, 264
modern packages, 269 application example, 268
visual programming environments, 270 block diagram, 267
temperature measuring general purpose, 266
main groups of modules, 260 software, 266
conditions specialised cards, 260
normal, 4 data loggers.
contact temperature measurement See recorders
induction heated charges, 405 detectors
continuous furnaces, 401 groups of, 182
charge temperature measurement, 401 thermal radiation.
drawn-through thermocouples, 402 See radiation detectors
microprocessor based furnace tracker, deviation functions, 9, 11
402 correction of, 9
moving charge, 401 dielectric heated charges, 407
thermally insulated storage device, 402 internal temperature
multi-zone, 401 contact sensors,difficulties, 407
applications, 401 MI thermocouple
salt-bath furnaces, 403 instrumentation filter, 407
typical types, 401 surface temperature
continuous sensors, 148 pyrometers, 407
application, 148 digital interfaces, 260
fibre optic, 148 IEEE-488, 260, 263
semiconductor, 148 RS-232, 260, 261, 263
application, 148 RS-232/RS-422, 263
construction, 148 RS-232/RS-485 converter, 261
correct operation, 149 RS-422, 263
with 4 to 20 mA converter, 149 RS-485, 261, 263
thermoelectric, 148, 149 digital noise filtering, 235, 236, 263
application of, 149 dilatation thermometers, 23
hot spot, 149 ambient temperature effects, 25
connecting leads amplifying transmission, 24
resistance, 100 cross-section, 24
continuous thermocouple sensors. errors, 24
See continuous sensors:thermoelectric materials used for, 24
convective, 15 materials, 25
48 2
principles of, 24
diode thermometers, 118
compound semiconductor, 118
correct operation, 118
errors, 119
forward voltage/temperature curve, 118
measuring circuit, 119
high impedance voltmeter, 119
measuring range, 118
non-linearity errors, 118
self-heating error, 118
disappearing filament pyrometers
applications, 170
block diagram, 163
calibrated for black body, 166
construction, 170
correction to readings, 169
direct ammeter calibration, 167
effective wavelength, 164
emitted radiant intensity
total, 164
errors
correction, 168
influence of atmosphere, 221
extension of measuring range, 169
filament temperature, 166
grey filter, 163, 169
radiance reducing factor, 170
spectral transmissivity, 170
lamp current 167
manual
practical applications, 223
operating principle, 163
radiance, 163
viewing angle, 168
red filter, 163
application, 164
reason for use, 164
reasons for application, 165
relative spectral sensitivity, 164
spectral transmissivity, 164, 166
relative spectral sensitivity
human eye, 164
spectral emissivity of filament, 166
sublimation of tungsten filament, 169
temperature scale divisions, 167
true body temperature
numerical example, 168
division
scale, 3
dynamic error correction
adaptive
microprocessor-based block diagram,
SUBJECT INDEX
329
analogue
active, 323
adaptive, 323
adaptive, block diagram, 324
adaptive, design of, 324
adaptive, principle of, 324
block diagrams, 324
passive, 321
sensitivity factor, 323
comparison of dynamic errors, 321
correction coefficient, 321
corrector performance indices, 321
corrector time constant, 321
corrector transfer function, 320
digital
microprocessor based, 325
PASCAL procedure, 327
simulation and prototyping, 327
principles, 320
dynamic properties of sensors
frequency response, 317
MI thermocouples, 317
dynamic temperature measurement
dynamic error
amplitude, 281
definition, 280
phase, 281
relative, 281
transfer function, 281
dynamic temperature measurement, 279
dynamic errors, 279, 280
systematic, 280
two necessary cases, 280
dynamic properties
need for knoweldge of, 279
dynamic properties of thermocouples, 317
properties of sensors, 281
step input, 280
emissivity
band, 158
clean materials, 160
different materials, 159, 160.
See Tables XIX and XX
non-conductors, 160
specific spectral, 159
spectral, 159
total, 157, 159
non-metals, 160
values for different surfaces, 160
energy domain
SUBJECT INDEX 483

thermal, 12 second generation, 130

EPTS-78, 7 grouping, 126
equilibrium, 7 indirect use. See fibre optic
hydrogen, 10 thermometers :extrinsic
points intrinsic, 126
of various working substances, 10 intrinsic sensing, 134
states intrinsically safe, 409
reproducible, 9 with black body sensor, 399
thermal, 5 modulating sensors, 126
thermodynamic, 6 Raman scattering, 134
equipartition theorem, 144 block diagram, 135
errors, 1 operating principle, 134
relative emissivity, 196 summary, 136
self-heating, 94 thermochromic thermometers, 128
excitation signal, 17 block diagram, 128
Extended Practical Temperature Scale (EPT- principle of operation, 128
76), 9 system design, 128
extension (compensating) leads. fixed points, 2, 8, 428, 446
See compensating leads defining, 10
definition of, 68 of IPTS-68, 9
of true centigrade scale, 8
temperature, 3, 5, 9
Fahrenheit scale, 2 fluids temperature measurement, 361
filter, 13 high temperature gas.
fibre optic sensors. See high temperature gas measurement
See fibre optic thermometers high velocity gas.
fibre optic thermometers, 125 See high velocity gas temperature
applications, 137 measurement
black body sensors, 132 liquids.
application range, 133 See liquid temperature measurement
block diagram, 132 low velocity gas.
indication error, 133 See low velocity gas temperature
change of refractive index type measurement
structure, 135 conduction, convection and radiation
change of refractive index type, 135 effects, 361
classification, 127 plasma .
continuous, 148 See plasma temperature measurement
extrinsic, 126 still air.
GaAs sensors, 127 See still air temperature measurement
extrinsic optical coupling, 134 freezing points, 7, 439
structure, 134 of various materials, 11
Fabry-Perot sensors, 133 function, 5, 15
block diagram, 133
fluorescent thermometers
material, 130 gas, 1
fluorescent thermometers, 129 compressed inert, 22
applications, 131 hydrogen, 7
block diagram, 129 ideal, 4, 6
construction, 131 moles of, 4
external excitation, 129 thermometer, 6, 7, 8, 11
first generation, 129 3He, 10
principle of operation, 129
properties of fluoro-germanate, 130 4He, 10
measurement range of, 6
484 SUBJECT INDEX

law recovery factor, 373

ideal, 4 static gas temperature, 372
temperature measurement total gas temperature, 372
ultrasonic thermometers, 142 typical sensor construction, 374
General Conference on Weights and hydrogen gas, 7
Measures (CGPM), 5
germanium resistance thermometers, 115
glass bulb, 1 ideal
glass-tank furnaces, 404 gas, 4, 6
furnace partition temperatures, 405 gas law, 4
pyrometers, photoelectric, 405 thermometric working substance, 4
furnace roof temperature indicating instruments, 250
pyrometers, photoelectric, 404 indicators, 250, 259
pyrometers, proper choice of, 404 analogue
pyrometers, total radiation, 404 electric, 250
thermocouples, 404 non-electric, 250
thermocouples, errors, 404 communication with other instruments,
glass outflow temperature, 405 250
molten glass digital
types of thermocouples used, 404 block diagram, 250
outside wall temperature process monitor, 251
infrared imagers, 404 general purpose, 251
portable contact thermometers, 404 multimeters, 252
pyrometers, 404 intrinsically safe, 250
thermocouples, 404 specialised, 251
pyrometers with digital interface, 251
photoelectric, 404 induction heated charges, 405
grouping contact temperature measurement, 405
See classification penetration depth, 405
Gummel number, 117 problems using thermocouples, 405
pyrometers, 406
photoelectric fibre optic, reasons for
Harrison cones, 33 use, 406
hazardous areas, 409 two-wavelength lens, 406
intrinsically safe design . two-wavelength lens, errors, 407
See intrinsically safe design two-wavelength/ratio, 406
heat resistiviy of thermocouple wires, 405
flux thermocouples
decomposition illustration, 154 parasitic emfs, 406
transfer precautions, 406
radiant, 162 industrial appliance temperature
high temperature gas measurement, 375 measurement, 397
pulsed thermocouple, 375 chamber furnaces .
operating reuirements, 378 See chamber furnaces
performance wave form, 376 continuous furnaces .
principle of operation, 378 See continuous furnaces
sensor requirements, 375 dielectric heated charges.
high velocity gas temperature measurement, See dielectric heated charges
372 for control, 397
application of different thermocouples, glass tank furnaces .
373 See glass tank furnaces
sensor hazardous areas.
effect, 372 See hazardous areas
SUBJECT INDEX 485

induction heated charges. colour temperature scale, 274

See induction heated charges commutation, 274
molten metals . Focal Plane Array systems
See molten metals photovoltaic, 274
salt-bath furnaces . thermopile, 274
See salt-bath furnaces light infrared cameras, 275
industrial thermometers linear systems, 276
disappearing filament pyrometers, 436 principle of operation, 276
precision of readings, 436 opto-mechanical scanning, 273
use of tungste strip lamp, 436 photoelectric detector array, 273
visual examination, 436 solar radiation, 275
liquid-in-glass variable volume surfacial systems, 273
accuracy, 431 thermovision systems
visual examination, 431 principle of operation, 273
manometric variable volume, 432 infrared thermometers .
accuracy and hysteresis, 432 See pyrometers
visual examination, 432 instrumentation, 12
other pyrometer types, 437 intelligent, 263
pyrometers virtual, 270
comparison method testing, 431 See also process instrumentation
resistance sensors, 432 integrated circuit thermometer, 121
break-down strength, 433 sensitivity, 122
insulation resistance, 433 International Bureau of Weights and
self-heating error, 433 Measures (BIPM), 7
stability checking, 432 International Committee of Weights and
testing of thermometric characteristic, Measures (BIPM), 7
432 International Electrotechnical Commission
visual examination, 432 (IEC), 45
testing International Temperature Scale of 1990
fixed points method, 431 (ITS-90), 9
testing, 431 interpolating limit
commparison method, 431 PRT, 8
thermocouples, 433 interpolation
comparison testing of various types, between fixed points, 3
433 intrinsically safe design
differential comparison method, 434 circuit of intrinsically safe barrier, 410
emf vs temperature, 433 classification of instrumentation
in situ testing, 435 non-simple devices, 409
reference junction test temperature, 434 simple devices, 409
visual examination, 433 electric temperature sensors
thermometric characteristic, 431 intrinsic safety barrier, 409
variable volume fibre optic thermometer, 409
indication variations, 432 isolation
variable volume thermometers, 431 electric, 409
industrial measurements, 9 mechanical, 409
industrial temperature sensors IPTS-48(60), 9
various types, 57 IPTS-68, 7
information bearing signal basis of, 8
conversion of, 12 extension at low temperatures, 9
infrared thermal imaging, 273 ITS-27, 7, 8
applications, 277 inadequacies of, 8
linear systems, 277 ITS-47
surfacial systems, 277 true centigrade scale, 8
486 SUBJECT INDEX

ITS-48, 7, 8 capillary, 19
ITS-90, 7 common thermometric glasses, 21
defining fixed points, 147 common thermometric liquids, 21
fixed points, 420 correct immersion depth, 20
construction, 420 cubic expansion of glass, 19
cryogenic, 420 coefficient of, 19
ranges (various), 420 cubic expansion of liquid
fixed points realisation coefficient of, 19
details of various, 421 domestic, 22
reproducing, 420 emergent liquid column  20
enclosed scale, 22
etched stem, 22
junction etched stem, 22
of two dissimilar metals, 37 example of calculation of correction, 21
heating and cooling of, 37 industrial, 22
inner capillary, 19
laboratory, 22
Kelvin, 12 maximum, 22
kelvin (unit of ITS-90), 5, 9 max-min, 22
Kelvin Thermodynamic Scale, 8 nominal immersion depth, 20
Kelvin's proposal, 5 sensitivity, 19
Kirchhoffs law, 158, 221 standardised laboratory, 20
correct use of, 20
structure of, 19
Lambert's directional law, 161 systematic error, 20
Lambert's law, 161 correction of, 20
law thermometric glasses
Boyle's, 4 properties of, 21
Boyle-Mariotte, 4 thermometric liquids, 19, 21
Charles', 4 with elongated bulb, 21
Kirchhoffs, 158, 221 zero depression, 22
Lambert's directional, 161, 162 low-temperature source, 4
Newton's low velocity gas temperature measurement
cooling, 285 conductive heat flux loss
of third metal methods of reducing, 366
surface temperature measurement, 352 contact sensors, 361
Ohm's, 37 convection heat flux, 362
Planck's, 6, 8, 9, 11,128, 132, 156, 157, convective heat flux from sensor, 364
173, 195, 213 direct pyrometric method, 371
Stefan-Boltzmann, 159, 160, 164, 214 error
Wien's, 8, 156, 157, 166, 172, 174, 204 analysis, 365
Wien's displacement law, 157, 177 from radiant heat transfer, 364
lead resistance, 37 reducing methods, 365
linearisation of characteristics. extrapolation method, 369
See temperature signal graphical estimation of radiation error, 364
transformation :shape increase of convective heat flux, 365
liquid temperature measurement, 375 indirect pyrometric method, 371
ultrasonic thermometers, 142 non-uniform cross-sectional temperature,
liquid-in-glass thermometers 370
axial capillary pyrometric methods
etched stem, 22 spectral ranges, 372
bulb volume radiation to tube wall, 362
small changes of, 19 radiative heat flux loss
SUBJECT INDEX 48 7

methods of reducing, 367 errors, 414

radiation shield, 368 electronic
sensor errors, 414
sheathed, 361 range, 414
sensor conductive heat flux, 362 elementary scale division, 413
differential equation, 362 permissible error, 413
errors, 363 fibre optic thermometers
sensor to tube wall radiant heat flux, 364 prostate hypertbermia, 417
sensors use in hf fields, 416
finned, 365 fluorescent thermometer, 416
methods of pipe-line installation, 366 infrared, 414
suction thermometer, 366 applications, 415
luminance ear canal, 414
definition, 161 liquid-in-glass, 413
luminosity, 162 mercury-in-glass, 413
definition, 161 ranges, 413
other measurements, 417
heated skeletal tissue, 417
manometric thermometers, 28 other measurments
gas, 32 thermistors in cardiac pacemakers, 417
constant volume, 32 skin surface sensors, 414
constant volume, errors, 32 thermistor based, 413
constant volume, specialised laboratory, thermistors
32 glass coated small bead, 417
constant volume, structure, 32 specialised applications, 414
liquid-filled, 28 use of hf fields, 416
error sources, 31 benign prostatic hyperplasia, 416
error values, 31 diagnostic NMR, 416
sensitivity, 30 hyperthermic oncology, 416
specification, 30 melting point of ice, 2, 5
structure, 28 melting points, 7
structure, basic elements of, 29 mercury
non-electric recorders, 253 solidification of, 4
continuous recording, 253 mercury-in-glass thermometer, 2, 3, 8
summary of properties, 32 application of, 4
vapour pressure, 31 capillary
application range, 31 vacuumised, 22
errors, 31 solid body temperature measurement, 352
operating principle, 31 meters .
structure, 31 See indicators
variable pressure, 28 modifier
variable volume, 28 amplifier, 13
material data converter, 13
thermometric, 7 filter, 13
measurements molten metals, 408
fundamental laboratory, 8 application areas, 408
measuring temperature measuring methods, 408
ranges contact immersion, 408
of standard interpolating sensors, 7 pyrometric, 408
signal, 13 pyrometric with immersion tube, 408
temperature, 4 moving body temperature measurement
medical thermometers applying a peak-picker, 388
contact character of movement, 387
488 SUBJECT INDEX

friction sensors, 392 emissivity

thermocouples, bow-band, 392 influence, 205
thermocouples, convex-band, 392 general information, 204
inductive, 390 main use, 204
block diagram, 391 resolution, 205
compensating circuits, 391 transmissivity, 205
Hall effect sensor, 391
infrared imaging, 388
oscillatory motion, 387 Newton's law of cooling, 285
photodiode arrays, 388 NHS, 7
other methods, 394 noble gas, 6
fluorescent fibre optic thermometer, 394 noise thermometers
temperature indicators, 394 application range, 146
temperature indicators, errors, 394 applications, 148
pyrometers, 387 block diagram, 145
application dedicated, 388 combined correlation and substitution
band photoelectric, 387 method, 148
errors, other sources, 388 comparison method, 146
errors, remedies, 388 block diagram, 146
low inertia detectors, 388 correlation method, 147
multi-wavelength, 387 block diagram, 147
photoelectric, 388 influencing factors, 148
photoelectric, fibre optic, 388 error, 145
pressure proof sight glass, 388 numerical example, 145
total radiation, 387 operating principle, 144
turbine applications, 388 output voltage, 145
two-wavelength, 387 reference temperature, 146
quasi-contactless, 392 substitution method, 146
errors vs peripheral speed, 394 thermal noise theory, 144
rotating cylinder surfaces, 393 noise thermometry, 15
schematic diagram, 394 non-black bodies
rotational motion, 387 spectral radiant intensity, 157
sensor mounting method, 387 temperature
sliding contact, 389 measurement, 167
errors, 389 true, 168
mercury, 390 non-contact thermometers .
parasitic emfs, 389 See pyrometers
Pt-RTDs, 389 non-electric thermometers .
thermistors, 389 See specific types under thermometers
thermocouple and slip rings, 389 non-grey bodies
wireless systems, 391 temperature errors, 175
errors, 392 normal direction, 160
thermistors, 391 Normal Hydrogen Scale (NHS), 7
transmitter shielding, 392
typical block diagram, 392
wireless transmission, 387 Ohm's law, 37
multi-wavelength pyrometers, 204 optical fibre
application in aluminium extrusion, 206 high temperature, 126
automatic light guide, 125
See also automatic pyrometers properties of,, 125
practical applications, 228 quartz glass, 126
construction, 206 refractive index of core, 125
effective wavelength, 204 transmitted wavelength, 125
SUBJECT INDEX 489

wavelength range, 125 plasma temperature measurement, 375

output signal platinum
desired, 13 bulb, 3
desirable properties of, 11
history in temperature measurement, 85
PCMCIA temperature dependence of resistance, 85
in recorders, 256 noble metal, 8
Peltier effect, 37 properties of, 85
personal computer, 259 resistance sensor, 11
Personal Computer Memory Card RTD, 86
International Association . Platinum Resistance Thermometer, 8, 10
See PCMCIA platinum resistors
photoelectric pyrometers, 192 Callendar equation, 85
absorption free atmospheric window, 194 numerical example on wire length, 86
automatic Van Dusen equation, 85
See also automatic pyrometers potentiometric recorders, 80
practical applications, 227 practical temperature scale, 8
band, 193 primary standards, 9
basic diagram, 193 pyrometers, 427
construction, 198 disappearing filament, 428
Drude theory, 198 heat flux comparison, 428
effective wavelength, 195, 198 photoelectric spectropyrometer, 428
errors resistance sensors
influence of atmosphere, 221 capsule, 426
general information, 192 for interpolation in ITS-90, 427
modulated radiant flux high temperature, 426
basic diagram, 193 long stem, 426
monochromatic, 193 long stem (construction), 427
non-black bodies, 196 long stem (contemporary design), 426
calculation of corrections, 196 resistance thermometers, 425
emissivity errors, 197 tungsten strip lamps, 427
reference temperature, 197 calibration, 428
relative error, 196 thermometric characteristic, 428
portable, 199 process instrumentation
relative spectral detector sensitivity, 194 current, 259
relative spectral transmissivity, 194 hardware elements, 260
RS-232 interface, 200 three level network, 260
scale defining equation for black bodies, trends, 259
195 processing systems
specialised, 200 computerised
spectral transmission of atmosphere, 194 components, 259
spectral transmission of atmosphere layer, PRT, 8, 9
194 interpolating equations, 9
stationary, 199 PTC silicon resistors. See silicon resistance
wavelength thermometers
effective, 194 Pt-RTD, 115
reference, 194 Pt thermometers, 10
working band temperature sub-ranges, 10
correct choice, 193 pyrometers, 152, 190
working wavelength band, 193 automatic.
Planck's See automatic pyrometers
constant, 128, 194 classification, 151
law, 6, 8, 9, 11,132, 156, 157, 173, 195 by operating method, 152
49 0 SUBJECT INDEX

by spectral response, 152 total radiation.

disappearing filament, 153 See total radiation pyrometers
See disappearing filament pyrometer pyrometric cones, 33
hand operated, 152 pyrometric temperature measurement
induction heated charges, 406 calculation of correction, 159
influence of atmosphere, 220
atmospheric windows, 220
disappearing filament pyrometers, 221 quartz thermometers, 139
effects of gases, 221 coefficients, 139
error elimination/limitation (various crystal
methods), 222 cross-section, 139
gas mixtures, 221 properties, 139
total radiation pyrometer errors, 221 errors, 140
influence of other factors (various), 222 app, 140
influence of surrounding walls, 216 operating principle, 139
chamber furnace, 217 reference temperature, 139
cooled sighting tube, 218
eliminating reflected radiation, 218 radiance
errors conditions, 217 definition, 161
numerical example of true temperature, reducing factor, 170
218 spectral concentrations of, 11
various methods to reduce, 217 total, 165
influence of target emissivity, 210 radiant
approaching black body conditions, 212 exitance, 155
automatic compensation, 216 heat
calculation of measurement errors, 211 exchange, 160
errors, 214 transfer, 162
true temperature calculation, 210 radiant intensity, 6, 155, 156, 158, 165
band photoelectric type, 211 band, 156
disappearing filament type, 210 black body, 156, 161
multi-wavelength type, 211 of an element of area, 161
total radiation type, 211 spectral, 156
two-colour type, 211 maxima, 157
various techniques, 212 total, 159
influencing factors, 209 radiation
main error components, 210 absorption, 154
photoelectric black body, 155
salt-bath furnaces, 404 constant
proper choice first, 156
glass-tank furnaces, 404 second, 156
real operating conditions, 210 definition, 154
summary of properties/applications, 223 direction, 161
total error, 210 emitting thermal, 160
total radiation heat flux density, 156
salt-bath furnaces, 404 incident, 156
manually operated, 152, 153 intensity, 6
structure, 153 laws, 154
modulated radiant flux, 193 reflection, 154
operating principle, 152 source, 437
practical applications, 209 thermal, 6
ratio, 153 transmission, 154
two-colour, 153 transparent body, 155
See two-colour pyrometers wavelength, 155
SUBJECT INDEX 491

white body, 155 deflection, 253

radiation detectors, 181 potentiometric, 253
comparative tabulation, 184 automatic potentiometric, 253
photodiodes, 182 circular chart, 255
photoelectric, 182 digital, 253, 254
application, 181 miniature, 255
photoconductors/photoresistors, 182 hybrid, 253, 255
spectral sensitivity of various types, 183 strip chart, 255
photomultipliers, 183 reference
photovoltaic cells, 182 function, 9, 11
properties, 183 system, 6
thermal reference junction
metal bolometers, 181 stabilisation techniques
properties, 181 various types, 77
pyroelectric, 182 reference temperature, 72
thermistors, 181 correcting devices (various), 75
thermopiles, 181 correction
thermopiles, hair-pin type, 181 automatic reference junction, 74
thermopiles, ribbon type, 181 calculation of, 72
thermopiles, thin film type, 181 graphical calculation of, 72
thermopiles, wire type, 181 numerical example, 73
vacuum photocells, 182 techniques, 74
radiation pyrometry errors, 72
errors, 156 calculation of correction, 72
relative errors, 156 nominal, 72
radiation thermometers . stabilisation
See pyrometers various conditions, 74
Raman scattering, 134 reference temperature stabilisation.
ratio pyrometers . See reference temperature correction
See two-colour pyrometers reflectivity
Rayleigh scattering light, 134 definition, 154
real industrial sensors . refractive index, 135, 160
See real temperature sensors relations
real temperature sensor kinship, 13
changing heat transfer ascent, 13
equivalent transfer function, 294 collaterality, 13
heat tranfer coefficient, 293, 294 descent, 13
radiative heat exchange, 293 likeness, 13
sensor emissivity, 293 resistance
design of, 291 four wire connections, 9
equivalent transfer functions (various), 295 lead, 37
heat transfer coefficient, 293 loop, 78
response(s) of a wire, 4
central, 292 ratio
numerical classification of, 293 thermometer, 10
qualitative explanation of, 292 ranges, 11
structural classification of, 292 resistance thermometer
surfacial, 292 construction, 90
volumetric, 292 error
recorders, 252, 259 self-heating, 91
analogue, 253 materials (various)
non-electric, 253 properties of, 89
analogue electric, 253 measuring
492
circuit, 91
self heating
error
numerical calculation, 94
thin film, 91
typical characteristics, 92
wire-wound, 90
resistance thermometer detectors .
See also resistance thermometers
general information, 88
metallic materials
desirable properties of, 88
structure, 88
resistance thermometer sensors, 94
construction, 94
dynamic behaviour, 95
measuring circuit errors
parasitic emfs, 101
measuring circuits, 96
bridges, 98
bridges, deflection type, 98
bridges, manual Wheatstone, 99
bridges, three wire, 98
Digital indicating instruments, 99
effect of connecting leads, 98
four-wire bridges, 100
self-balanced bridges, 100
three-wire Wheatstone bridge, 100
two wire Wheatstone bridge, 99
voltage dividers, 96
mineral insulated (MI), 95
special purpose, 95
thermowell structures . See Table 3.3
resistance thermometers
See also resistance thermometer detectors
Callendar equation, 88
Callendar's interpolation, 88
conversion of temperature to resistance, 87
deflection type bridge circuits
digital indicating instruments, 101
general and historical background, 85
linear temperature dependence, 87
limitation of, 87
measuring circuits
quotient instruments, 101
physical principles, 86
polynomial coefficients of dependence, 87
principle of operation, 86
resistance at reference temperature, 87
resistance temperature coefficient 87
smallest measured temperature, 101
temperature dependence, 87
Van Dusen equation, 88
SUBJECT INDEX
Romer scale, 2, 3
Romer's laboratory, 2
RS-232 interface
photoelectric pyrometers, 200
RTD.
See resistance thermometers
Pt, Ni and Cu
permissible tolerances .
See Tables XVI and XVII
temperature relationship .
See Tables XVI and XVII
salt-bath furnaces
charge temperature measurement
emissivity of salt, 404
K-type immersion thermocouples, 403
metal sheaths, 403
pyrometers, ceramic directing tube, 404
pyrometers, photoelectric, 404
pyrometers, total radiation, 404
slag layer errors, 404
scale, 2, 3, 5, 6, 8
division, 3
Romer, 2
temperature, 8
See specific types
thermodynamic, 6, 8
Seebeck effect, 37
Seger cones, 33
semiconductor
normal region resistivity, 114
bolometers, 186
semiconductor diode thermometers .
See diode thermometers
semiconductor junction thermometers
forwardjunction current, 116
principles of operation, 116
reverse saturation current, 117
temperature sensitivity, 117
semiconductor thermometers, 103
bulk effect two-electrode, 103
See silicon resistance thermometers
germanium
See germanium resistance thermometers
groups of, 103
junction device sensors, 103.
See also integrated circuit
thermometers .
See also transistor thermometers .
See also diode thermometers
resistance
See silicon resistance thermometers
SUBJECT INDEX 493

semiconductors ceramic-metals, 57
carrier mobility, 103 materials, 53
classification of devices, 103 correct choice of, 53
cryogenic freeze out, 103 influences, 53
doped, 103 properties of, 53
extrinsic, 103 metal, 53
intrinsic region, 103 different structures, 53
normal region, 103 uses, 53
n-type protecting, 53
carrier mobility, 104 signal transformation
resistivity, 104 initial signal, 13
temperature dependence signal transmitters
bulk material, 103 See also temperature signal transmission
junction, 103 analogue, 247
temperature measurement, 103 conditioning operations, 247
thermal properties of, 103 current loop, 247
semi-transparent solid bodies basic block diagram, 247
radiation mechanism, 381 digital, 247
sensor(s) four-wire
classification rail/wall mounted, 248
by conditioning circuits, 13 various signal transformations, 248
by energy form, 13 four-wire system
by function, 13 isolation from power supply, 247
by structure, 13 head mounted, 248
by thermal expansion, 15 infrared thermometry, 249
electrical output signal, 15 intelligent panel mounted, 248
key diagram, 13 microprocessor based, 248
non-electrical output signal, 15 multi-channel, 249
tree diagrams, 14 output signal formats, 250
collectivity of, 13 panel mounted
composition of, 13 block diagram, 249
contact, 15 standard 4-20 mA, 247
conductive heat transfer, 15 standardised process signals, 248
convective heat transfer, 15 two-wire system, 247
development of, 13 with RS-232 interface, 248
distributed parameter. silicon resistance thermometer detectors.
See continuous sensors See silicon resistance thermometers
electric silicon resistance thermometers
dynamic parameters of. acceptance testing, 115
See dynamic temperature measurement important parameters, 116
evolution of, 13 manufacture, 114
generality of, 13 measuring circuits, 115
groups of, 13 resistance/temperature equation, 114
modulators, 12 silicon-RTDs .
platinum resistance, I 1 See silicon resistance thermometers
resemblance of, 13 silistors.
self-supporting cross-converters, 12 See silicon resistance thermometers
structures for IPTS-68, 9 Si-RTDs.
various families of, 14 See silicon resistance thermometers
sheaths, 53 software, 261
ceramic specialised, 260
porous, 57 solders
uses, 57 properties of, 51 .
49 4 SUBJECT INDEX

See Table XIV surface heat transfer coefficient, 336

solid body temperature measurement . survey of sensor errors, 353
See also ultrasonic thermometers :solid thermal contact resistance, 343, 345
body thermal resistance of sensor, 340, 342,
contact method, 333 343
application of peak-picker, 349 thermally compensated sensors, 346
bow-band thermocouples, 350 thermistors, 351
ceramic surface errors, 354 thermocouples, 349
convective heat transfer, 333 thin-film thermocouple, 353
convex-band thermocouples, 351 thin-wire resistance detector, 354
copper surface errors, 354 thin-wire thermocouple, 354
differential equation of heat conduction, third partial error, 334, 346
335 internal, 358
disk sensor, 340 partial error elimination, 360
disk thermocouple, 350 partial errors, 359
disk thermocouple sensor, 342 rod sensor, 358
disturbing heat flux, 337 ultrasonic, non-invasive sensor, 360
disturbing temperature field, 335 partial error reduction, 359
disturbing temperature filed, 337 quasi-contact method, 333
double thermocouple sensor, 342 principe of operation, 355
dynamic errors, 347 simplified sensor diagram, 356
effect of sensor, 334 rough surface estimation .
elimination of all partial errors, 346 See temperature indicators
errors of different materials, 355 semi-contact method, 333
extrapolation method, 356 specific spectral emissivity, 160
extrapolation method errors, 357 spectral
first partial error, 333 absorptivity, 158
first partial error calculation, 338 distribution, 6
first partial error, calculation of, 343 emissivity, 157, 159
first partial error, reduction of, 343 radiant intensities
first partial error, reduction of, 344 of a black body, 6
first partial error, total elimination of, standard thermometers
344 temperature ranges of, 18
fixed contact sensors, 352 Stefan-Boltzmann law, 159, 160, 160, 214
flat-band thermocouples, 351 still air temperature measurement, 374
infinitely large plate, 338 correct sensor design, 375
influence of body thermal properties, Stokes light, 134
348 surface temperature measurement
mercury-in-glass thermometer, 352 plastics .
numerical calculation of method errors, See plastics :surface temperature
347 measurement
peened thermocouple, 352
point contact thermocouples, 350
portable contact sensors, 349 taxonomy
rod sensor, 341 classical
second partial error, 334, 344 application of, 13
second partial error, reduction of, 346 principles of, 13
sensor heat flux, 339 See classification
sensor heat flux models, 339 temperature, 2, 6, 15, 66, 129, 135
sensor sensitive point, 334 absolute, 156
soldered thermocouple, 352 zero of, 4
spiral thermocouple, 350 ,. ambient, 1
surface heat flow effects, 333 concept of, 1
SUBJECT INDEX 495

correlation with radiated colour, 151 temperature measurement

definition, 1, 5 channels
dependence, 135 status and trends, 230
fixed points, 1 dynamic.
historical background, 1 See dynamic temperature measurement
human body, 3 in industrial appliances .
intermediate, 1 See industrial appliance temperature
lowest attainable, 2 measurement
measure of, 6 in medicine, 413 .
measurement, 135 See medical thermometers
neutral, 1 diagnostic, 413
normal, 2 moving bodies.
of boiling water, 2, 3, 4 See moving body temperature
reference, 72 measurement
scales multi-channel, 245
See specific type non-black bodies, 167
scale of, 3 non-grey bodies, 174
signal transformation, 14 process, 229
thermodynamic, 5 real time .
water triple point, 8 See infrared thermal imaging
distribution sensor thermal transient.
uniform, 6 See dynamic temperature measurement
fixed point, 5 signal(s)
defined value of, 5 See temperature signal
fixed points, 1, 2, 3, 5, 9 solid body .
of ITS-90, 10 See solid body temperature
instrumentation chains, 12 measurement
measurement, 4 systems
broad view of, 13 classification of, 230
using metal conductor, 85 distributed, 244
scale, 3, 4 time varying, 287
development of, 8 time varying dynamic error
practical, 8 step input illustration, 287
suitable, 4 traditional equipment, 231
scales, 1, 4 transparent solid bodies .
conversion of, 12 See transparent solid bodies
comparison of, 7 temperature measuring
practical realisations of, 7 chains, 12
temperature controllers, 256 instruments, 12, 419
analogue, 256 calibration and esting .
analogue electric, 256 See calibration and testing
digital, 256, 257 temperature sensor dynamics
microprocessor-based, 257 temperature sensor dynamics, 280
non-electric, 256 calculation of dynamic properties
temperature indicators, 32 differential equation, 296
crayons, 34 various assumptions, 296
liquid crystals, 35 experimental determination
moving body temperature application of methods, 299
measurement, 394 classification of methods, 299
self-adhesive, 35 external input
therrnochromic paints, 34 equivalent models and responses, 303
applications, 34 external input frequency domain, 307
transition temperature, 34 multifrequency binary signals, 308
496
rectangular periodic input, 307
sinusoidal input, 307
external input methods, 300
convective heat transfer, 300
convective step input, 300
external input methods
logarithmic time constant
determination, 304
external radiative input, 308
heating-up curve method, 308
heating-up/cooling-down, 309
sinusoidal input, 310
step input, 310
influence on chamber furnace control, 400
internal input, 308, 311
multifrequency binary sequence, 312
transfer function, 311
MBS methods, 300
microprocessor based instrumentation, 300
parameters of chosen sensors, 313
parameters of MI thermocouples, 313
surfacial response
dynamic properties, 305
volumetric response
dynamic response, 305
temperature sensor modelling, 281
temperature sensors, 12, 15, 306
calculation of dynamic properties
differential equation, 296
various assumptions, 296
classification, 13
by measuring ranges, 17
construction similarities, 14
contacting
genus, 15
non-electrical, 14
correction of dynamic properties, 240
dynamic properties
See temperature sensor dynamics
dynamics of, 280
See temperature sensor dynamics
electric
block diagram, 282
electrical contacting, 15
classification of, 16
electric and thermal analogy, 297
electrical conversion stage, 282
frequency response
amplitude characteristic, 284
Cartesian form, 283
imaginary part, 284
phase characteristic, 284
polar form, 284
SUBJECT INDEX
rational polynomial, 283
real part, 284
function of, 13
idealised, 284
differential equation, 285
frequency response, 286
heat transfer coefficient, 285
initial conditions, 285
time constant, 286
tranfer function, 285
transfer function, 286
idealised response
calculation of dynamic error, 291
dynamic error us immersion time, 289
half-value time, 288
nine-tenth value time, 288
various input signals, 290
non-contacting
classification of, 15, 17
group, 15
non-electrical
classification of, 16
real .
See real temperature sensor
self-sustaining cross-converters, 15
surfacial response
transfer function, 305
thermal conversion stage, 282
temperature signal(s)
conditioning
See temperature signal conditioning
data, 229
electromechanical, 231
indicators, 231, 259
initial transformation
sensor, 230
measured, 230
non-electric, 231
processing, 229
various techniques, 230
recorders, 231
transformation, 229
See temperature signal transformation
temperature signal conditioning, 229
classification, 231
electric
analogue, 231
digital, 231
temperature signal transformation
alarming, 243
analogue-to-digital converter, 245
characteristic value detection, 241
averager, 241
SUBJECT INDEX 497

illustration, 242 serial, 244

peak-picker, 241 temperature signal transmitters .
solid body temperature measurement, See signal transmitters
349 temperature source
peak-to-peak holder, 241 higher, 5
valley-picker, 241 temperatures
comparison of shielding and analogue characteristic, 7
filtering, 237 thermal
converter location, 244 behaviour, 4
data acquisition equilibrium, 5
sampling rate calculation, 236 expansion
system, 236 of a copper rod, 4
initial energy transformers, 231 of another metal, 4
manometric thermometer, 233 hysteresis, 85
nature, 230, 231 imaging
between physical quantities, 231 medical application, 415
digital, 231 radiation, 6, 154
digital-to-digital, 233 See also radiation
non-electric to electric, 231 radiator, 156
scale, 230, 234 state, 1, 5
analogue amplifier, 234 thermal electromotive force, 37
unified, 234 See thermoelectric force
shape, 230, 235 thermal emf, 37
active filter, 235 See thermoelectric force
analogue filtering, 235 thermal noise voltage, 147
analogue linearisation, 238 amplifier specific, 147
anti-aliasing filter, 236 thermistors, 104
averaging filter, 236 advantages of, 108
Butterworth filter (maximally flat applications, 111
amplitude), 235 temperature ranges, 109
conversion aliasing, 236 compared with RTD, 105
digital correction
filtering, 235, 263 array configuration, 111
linearisation, 239 post-conversion techniques, 112
noise filtering, 235, 236, 263 disadvantages, 108
filter effectiveness, 236 heat dissipation constant, 106
linearisation of sensor characteristics, in medical thermometry, 111
238 interchangeability, 109
shielding, 236 linear
Tchebishev filter, 235 output assemblies, 112
Thomson filter (maximally linear resistance type, 112
phase), 235 voltage type, 112
signalling, 243 linearisation
temperature signal transmission, 244 analogue, 111
analogue, 244 digital, 112
multiplexing, 245 circuits, 112
digital, 244 long time instability, 111
multiplexing, 246 manufacture
parallel, 244 bead type, 109
IEEE-488,244 disk type, 109
RS-232C serial, 244 portable type, 109
RS-422 serial, 244 production control, 111
RS-485 serial, 244 measuring circuits, 111
498 SUBJECT INDEX

deflection bridge, 112 low thermal inertia, 60

digital indicating instruments, 113 thermocouples, 46
series connected, 113 90%Pt-10%Rh/Pt, 9
various, 114 after welding, 51
non-linearity errors, 112 bow-band
normal applications, 105 moving body temperature
NTC type, 104 measurement, 392
permissible measuring current, 107 circuits, 37
production, 104 application of superposition, 44
PTC type, 104 law of consecutive metals, 41, 43
as binary temperature sensors, 108 law of consecutive temperatures, 43
reference temperature, 105 law of third metal, 39
resistance temperature coefficient, 105 chromel / alumel, 47
self-heating, 105 Chromel / Constantan, 47
self-heating errors code B, 46, 47
estimation, 106 code E, 46, 47
numerical example, 107 code J, 46, 47
value, 106 code K, 46, 47
sensitivity code N, 46, 50
compared with thermocouple, 112 code R, 46, 47
signal, 111 code T, 46, 47
sensors, 109 convex-band
solid body temperature measurement, 351 compensating cables .
stationary type, 111 See compensating cables
still air characteristic, 105 drawn-through
still water characteristic, 105 continuous furnaces, 402
temperature emf vs temperature
dependent resistance, 104 commonly used, 45
sensitivity, 105 high temperature, 65
typical types general information, 44
various, 109 hard soldered, 51
thermocouple measuring circuits high temperature, 65
potentiometeric, 79 applications (various), 66
variable current, 81 Graphite / SiC, 67
variable voltage, 79 Graphite / W, 66
variable voltage bridge, automatic, 80 Graphite / graphite with 0.1 % to
thermocouple reference temperature 0.2% Be, 67
analogue compensation, 240 Ir / IrRh, 66
automatic compensation, 239 most popular, 65 . See Table XII
digital compensation, 240 non-metallic, 67
thermocouple sensors, 52 other types, 67
conical well, 61 W / Mo, 65
construction, 52 W / MoW, 65
cylindrical screw-in, 60 W / WRe, 66
furnace mounting, 60 WRe / Re, 66
high-pressure, 60 highest working temperature, 53
industry standard, 58 identification of types, 50
mineral insulated, 61 industrial assemblies
applications, 63 common, 52
metal sheaths, 63 double, 52
terminal enclosures, 64 typical, 52
various structures, 62 interpolating
structure 90%Pt-10%Rh/Pt, 8
SUBJECT INDEX 499

junctions 6% rhodium, 46,47

various, 51 standardised wires, 50 .
laboratory, 64 See Tables XIII
material properties, 45 properties of, 53
measuring circuits, 76 structure, 44
deflection type, 77 review, 57
manual variable voltage, 80 straight, 57
potentiometers . See thermocouple various, 57
measuring circuits type-K immersion
measuring junction, 44 with thermowell, 58
measuring junctions, 50 reverse polynomial. See Tables XI
correctly formed, 51 value tolerances . See Table X
Type J, 51 thermodynamic
Type K, 51 equilibrium, 6
Type T, 51 Kelvin Scale, 4
various types, 50 scale, 3, 6, 8
mineral insulated temperature, 5, 9
concentric, 61 temperature scale (TTS), 3
needle, 64 temperatures, 6
unprotected laboratory, 65 thermoelectric circuit, 38
Pallaplat type, 50 thermoelectric force
parasitic emfs, 51 application of double subscript notation,
physical principles, 37 38
thermoelectric force, 37 circuit of two metals, 39
platinum-10% rhodium / dependence on electron density, 37
platinum (90%Pt-10%Rh/Pt), 8 real source of, 37
platinum as reference metal, 41 reference junction, 39
properties of commonly used, 46 reference temperature, 39
properties of metals and alloys, 46, 48 resultant, 38
rare metal, 51 summing thermal emfs, 38
laboratory, 65 value of, 38
reference thermoelectric series, 41
calibration tables, 46 . emf value of various metals, 41
See Tables II to IX thermoelectric thermometry
junction, 44, 68 foundation of modern, 39
temperature, 42 thermometers
right angle, 60 bimetallic .
salt-bath furnaces, 403 See bimetallic thermometers
standard emf, 8 dilatation .
standardised, 46 See dilatation thermometers
code S, 46 dynamic errors .
copper / copper-nickel, 46, 47 See dynamic temperature measurement
Iron / copper-nickel, 46, 47 principal cause, 279
Nickel-chromium / copper-nickel, 46, fibre optic.
47 See fibre optic thermometers
Nickel-chromium / nickel-aluminium, first, 1
46,47 Florentine, 2
Nickel-chromium-silicon / nickel- gas, 6, 8
silicon, 46, 50 primary standard, 8
Platinum-10 % rhodium / platinum, 46 liquid-in-glass.
Platinum-13 % rhodium / platinum, 46, See liquid-in-glass thermometers
47 manometric .
Platinum-30 % rhodium / platinum - See manometric thermometers
50 0 SUBJECT INDEX

mercury-in-glass, 2, 3, 8 pairs, 120

See mercury-in-glass thermometers
quartz.
See quartz thermometers
resistance, 1 l
ratio, 10
using expansion of solids, 23
See bimetallic thermometers
See dilatation thermometers
modulators
metallic resistance thermometer, 15
semiconducting resistance, 15
thermo-electric cross-conversion, 15
thermometric
liquid, 2
material, 7
working substance, 3, 4
ideal, 4
thermoscope, l, 2
Thomson effect, 37
in tungsten strip lamps, 429
total emissivity, 159
total radiation pyrometers, 184
automatic
See also automatic pyrometers
practical applications, 227
basic diagrams, 184
construction, 190
microprocessor based, 191
corrections to readings vs temperature, 188
detector sensitivity decrease
cause of, 189
emissivity corrections, 188
errors
influence of atmosphere, 221
extension of measurement range, 189
general information, 184
grey filter, 190
imprecision of corrections, 188
influences
housing temperature, 189
target distance, 189
measuring ranges, 192
non-black body measurement, 186
portable type, 191
scale defining equation for black bodies,
185
simplified design, 185
true temperature calculation, 187
transistor thermometers, 119
errors
non-linearity, 120
measuring circuits, 120
transmissivity
definition, 154
transparent body radiation, 155
transparent solid bodies
contact methods, 386
errors, 386
definition, 381
equivalent emissivity, 383
errors
causes, 381
general conditions, 381
glass, 383
cooling processes, 385
deeper layers, 385
emissivity, 383
pyrometer wavelength for 1 mm thick
layer, 384
surface and internal simultaneous, 385
surface temperature measurement, 384
through heating, 385
industrial practice, 383
influence of thickness
spectral emissivity, 385
internal temperature differences, 383
layer thickness more than 10 mm, 384
logarithmic absorption coefficient, 381
natrium-silica glass
relative spectral reflectance, 384
relative spectral transmission, 384
pyrometric method, 381
radiation notation, 382
surface reflectivity, 382
triple point of water, 5, 9, 11
triple points, 7, 448
various, 11
TTS, 6
two-colour pyrometers, 172
errors, 172
influence of atmosphere, 221
general information, 172
manual
practical applications, 223
modern, 172
principle of operation, 172
ratio of spectral intensities, 173
ratio of spectral radiant intensities vs
wavelength, 174
red filter, 173
effective wavelength, 173
scale defining equation, 172
body emissivity, 172
simplified diagram, 172
SUBJECT INDEX 501

spectral radiant intensity, 172 water

vs wavelength, 174
upper application limit, 174
two-wavelength pyrometers, 201
construction, 202
electro-mechanical self-balancing, 201
errors, 201
red filter effects, 201
operating principle, 201
principles, 201, 202
ultrasonic thermometers, 140
applications, 143
block diagram, 141
errors, 143
gas temperature measurement, 142
liquid temperature measurement, 142
non-invasive technique, 142
operating principle, 140
solid body temperature measurement, 143
internal, 360
summary of various, 143
wires and rods, 141
vapour pressure temperature, 10
boiling point of, 2
freezing point of, 3
mixture, 3
triple point temperature, 8
wavelength, 6
of radiation, 155
Wien's displacement law, 157, 177
Wien's law, 8, 156, 157, 166, 172, 174, 204
working standards, 429
disappearing filament pyrometers
temperature range, 430
function, 429
photoelectric pyrometer
silicon detector, 430
resistance sensors
application range, 429
testing baths
comparison testing, 429
thermocouple sensors, 430
accuracy, 430
calibration, 430
tungsten strip lamps
temperature range, 430
working substance, 4, 5, 6
thermometric, 3, 4
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

6
Fibre Optic Thermometers
6.1 Properties of Optical Fibres
Dielectric opticalfibre, which is also referred to as a light guide, is composed of a rod core
surrounded by a sheath as shown in Figure 6.1 . The core, which conducts the
electromagnetic wave, has a refractive index, nj , while the sheath, which contains the wave
within the core, has a refractive index, n2 . To ensure that the wave is contained in the core
it is necessary that the condition nj > n2 is satisfied. The electromagnetic waves incident
on the front part of the optical fibre, within the angle cone, 2a, corresponding to the critical
angle, 9c,. , enter the fibre core. After multiple reflections from the core-sheath interface
these waves leave the other end of the fibre . The waves from outside the 2a-cone leave the
sheath to environment . Smolinski (1985) has shown that the value of critical angle, Bcr,, and
of aare given by:

cose cr = n2 / nj (6 .1)

sin a = ni - n2 (6 .2)

where n j is the refractive index of the core material and n2 is the refractive index of the
sheath material.
In addition to the condition n 1 / n2 > 0, optical fibre materials, whose core material has a
low absorption coefficient at the transmitted wavelength, should be chosen. Optical fibres
are used for wavelengths in the range from ultraviolet through visible up to infrared
radiation . The most commonly used optical fibres have a core and sheath made of quartz

CORE SHEATH PROTECTIVE TUBE

Figure 6.1 Propagation of electromagnetic waves in optical fibre

126 FIBRE OPTIC THERMOMETERS

glass with different chemical composition . Usually, sheath materials are made entirely of
plastics . High temperature optical fibres have cores made of quartz and sapphire . Some
optical fibres are covered by layers which reduce optical losses and provide good
mechanical strength for protective purposes . In Table 6 .1, which gives parameters of some
optical fibres, it can be seen that quartz optical fibres have the lowest absorption coefficient
and thus the lowest transmission losses at the wavelengths 0 .85 gm, 1 .3 lm and 1 .55 l.tm .

6.2 Classification of Fibre Optic Thermometers

It is important to examine the grouping of fibre optic thermometers . McGhee and
Henderson (1989) give a general classification of optical fibre sensors based upon previous
groupings given by Culshaw (1982), Pitt et al . (1985) and Medlock (1986 ; 1987) . Further
classification of optical fibre sensors is given by Ning et al . (1991).
The general classification of optical fibre sensors provides the foundation for the
grouping of optical fibre thermometers . From Grattan (1995) and Wickersheim (1992), it
can be seen that there are two groups of fibre optic thermometers as shown in Figure 6 .2 .

" Extrinsic sensing, also termed, indirect use thermometers, which are more popular, use
the optical fibre for transmission of the electromagnetic waves between the sensor and
the electronic temperature indicating system . In these sensing systems the light itself is
modulated by some process outside the optical transmission system .
" Intrinsic sensing, also referred to as direct use thermometers, where the optical fibre
itself is used as the temperature sensor, also exhibit the characteristics of modulating
sensors described in Chapter 1 . In these sensors a light transmitting property of the
optical fibre is modulated by the temperature . They indicate either the average
temperature of the fibre or the temperature distribution along its length .
Optical fibres are also applied in the construction of pyrometers as described in
Chapter 10 .

Table 6 .1 Types of optical fibre

Fibre type Properties Core Sheath

Quartz Material Quartz glass Si02 doped Quartz glass

Refractive index n j = 1 .48 n 2 = 1 .46
Diameter 50 to 300 pm 12 .5 to 400 pm

Quartz Material Quartz glass Polymers

Refractive index n, = 1 .46 n 2 = 1 .40
Diameter 200 to 1500 Itm 300 to 2000 pm

Plastic Material Polymer e .g . polymethyl-methacrylate Polymers

Refractive index n j = 1 .49 n 2 = 1 .40
Diameter 300 to 1500 Itm 500 to 2000 pm
EXTRINSIC SENSING THERMOMETERS 127

FIBER OPTIC THERMOMETERS

EXTRINSIC SENSORS INTRINSIC SENSORS

GaAs sensor absorption type Raman scattering type

thermochromic reflection type change of refractive index

fluorescent
L optical coupling of two fibres

black body sensor

thermal radiation type

"Fabry-Perot" sensor
optical interference type

Figure 6.2 Classification of fibre optic thermometers

6 .3 Extrinsic Sensing Thermometers

6.3.1 Thermometers with GaAs semiconductor sensors

The operating principle of these thermometers, whose structure is given in Figure 6.3, is
based on the dependence of the absorption coefficient, A, of a semiconductor material upon
its temperature and upon the frequency of the incident radiation. Samulski (1992) has
shown that in the case of gallium arsenide (GaAs) this dependence is empirically given by :

A = Ao exp[a(h v - Eg ) l kT] (6 .3)

(b)
1,0
P2s,-_ _ ___
(a) LED GaAs SENSOR z 0'8 II
r ----- ~-~ TRANSMITTING OPTICAL FIBRE L __ i
NwA "' 0,6 P4u _ _-40 C
I L~ N I

I ~ a 25°C I lED EMITTING

I Nyyh a 0,4- I LIGHT
RECEIVING OPTICAL FIBRE .., i

L --------- I PHOTODIODE DETECTOR 0.2 ~I l

w 1 l`
0
0.88 0,90 0,92
WAVELENGTH 5,, Nm
Figure 6.3 A fibre optic thermometer with the basic operating structure of (a) uses a GaAs
semiconductor sensor with the relative radiation transmitted through a 0.25 mm thick prism in (b)
128 FIBRE OPTIC THERMOMETERS

where Ao is a constant, a is a material constant, h is the Planck constant, v is the frequency

ofincident radiation, Eg is the energy gap, k is Boltzmann constant and Tis the temperature .
As the energy, hv , approaches Eg, for a chosen wavelength, equation (6.3) shows that the
variation in absorption coefficient, A = f(T), really exhibits a very steep dependence . In this
thermometer, a narrow-band LED light source, as shown in Figure 6.3(a), emits the light
along an input light guide to a GaAs prism sensor (Samulski, 1992) . After passing the prism
the radiation returns by an output light guide to the diode photo-detector . The GaAs
detecting prism exhibits the phenomenon of a shift of the radiation band border. Thus, the
resulting output signal depends upon the thermally dependent optical transmission of the
GaAs prism as shown in Figure 6 .3(b) . Fibre optic thermometers with GaAs semiconductor
sensors are used in the temperature range below 50 °C usually in medicine and biology .

6.3.2 Thermochromic thermometers

These thermometers are based on the temperature dependence of the reflection factor of
liquid crystals . The reflection factor also depends upon the frequency of the incident visible
radiation. Since they appear to change colour with temperature, liquid crystals are called
thermochromic materials. In a thermometer described by Brenci et al (1984) and Grattan
(1987), a thermochromic material consisting of a cobalt chloride solution in water/alcohol is
used. This substance, which shows intensive colour change between 25 °C and 75 °C, has a
spectral absorption with temperature given in Figure 6.4(b). The system design shown in
Figure 6.4(a) uses a halogen lamp whose light is chopped and transmitted by a silica optical
fibre to a temperature sensor . The sensor is a small probe containing the thermochromic
solution . Incoming light is reflected from the bottom of the mirrored probe to another
optical fibre where it is split into two light beams at At = 0.655 pm and Az = 0.800 pm. As
can be seen in Figure 6.4(b) the light beam at Az = 0.800 um, which is not absorbed by the
cobalt chloride solution, serves as a reference value while the light absorption at
At = 0.655 pm is temperature dependent . Both signals are re-detected by photodiodes,

(b!
(a) TRANSMITTING OPTICAL CHOPPER HALOGEN
75 °C
FIBRE LAMP
1_
FILTER 0,655pm
a 60°C
OPTICAL D
U o
COUPLER ~,
a
D SO°C
RECEIVING
FIBRE
MICROPROCESSOR
a
/
PHOTODIODES a ~~ 40°
'SENSOR FILTER O,BOOym ~5°
=5°C
0,4 0,5 0,6 0.7 O,8
WAVELENGTH a, jim

Figure 6.4 A tbermochromic optical fibre thermometer with the basic block diagram of (a) uses the
spectral absorption of cobalt chloride solution in water-alcohol at different temperatures in (b)
EXTRINSIC SENSING THERMOMETERS 129

amplified and ac/dc converted . Subsequently their ratio, which is formed in a

microprocessor, is a measure of temperature . The readings are independent of source
intensity fluctuations and of transmission losses because both signals are transmitted along
one common optical path. Thermochromic thermometers are applied in the temperature
range from 25 °C to 75 °C in medicine and biology, which is considered in Chapter 21 .

6.3.3 Fluorescent thermometers

In this type of thermometer a fluorescent material, which is placed at the end of the light
guide, is excited by radiation of a given wavelength . The external excitation, which is
essentially an interrogative stimulation, causes the sensor to fluoresce at a different
wavelength from the exciting light source . The fluorescent radiation and the exciting
radiation are subjected to monochromatic filtering, which allows easy separation of the
radiation from the two sources .
First generation fluorescent thermometers were based on temperature dependence of
monochromatic repartition of emitted radiation:

e,~
- .f(T) (6 .4)
Ee,

where Ee, A and Ee,,;,2 are radiation intensities respectively at wavelength ~l and Xz and T

is temperature .
The operating principle of a fluorescent thermometer offered by ABB (Sweden) is
shown in Figure 6 .5 (Grattan, 1987) . The GaAs sensor, which is placed at the end of the
light guide, is excited by a LED modulated light when it emits fluorescent radiation . The
combined light from the visible fluorescent light emitted by the sensor and the exciting light
is transmitted by the same light guide to an optical bifurcator which splits the light into two
beams . Each beam then passes through a filter. The output signal of each filter, which is
proportional to the respective intensities, Ee, A and E, , ~2 , are applied to detecting

CHOPPER I
LED ---
I U I
I
I
I

II D
L ---- -
D
--_-- _J
I
I

OPTICAL PHOTODIODES DIVIDER

FILTERS SENSOR GaAs

OPTICAL FIBRE

Figure 6.5 Fluorescent optical fibre thermometer

130 FIBRE OPTIC THERMOMETERS

photodiodes before being amplified . Taking the ratio gives a signal which is a function of
the measured temperature . The readings are independent of the exciting radiation intensity,
nevertheless they may be influenced by bending ofthe light guide (Grattan, 1987) .
Second generation fluorescent thermometers are based on the decay-time concept of a
periodically excited probe . Mei Sun (1992) and Samulski (1992) state that the emitted
radiation intensity as a function oftime Ee/t) is given by:

Ee (t) = Ee (0)exp(-t /'Z) (6 .5)

where Ee (0) is the radiation intensity at the instant the exciting impulse stops, z is a so
called decay time and t is time. The decay time is usually defined, as shown in Figure 6.6, as
the time difference t2 - t l such that Ee(t2) = E e (tt ) / e , where e is the base of the natural
logarithms. Grattan and Zhang (1995) provide a survey of other measurement methods of
fluorescent lifetime.
Figure 6.7(a) shows decay time, z, as a function of temperature, S, of magnesium fluoro-
germanate activated with tetravalent manganese. The wavelength bands of exciting and

j I EXCITING IMPULSE

E Q (O) - t
I
z
E e (t 1=E, (0) exp (-t 1e)
Eelt'1
I 1
z II I

a E . (t,)
0
- i -_--t._------_-----

1 I

0 t, tz
TIME t

Figure 6.6 Impulse excited radiation intensity of a fluorescent material as a function oftime
(6)
(a) EXCITATION- _ EMISSION
_ REGION ~ REGION

tr S z_ 0,9
W z
4 20,6- t
a
v
>0
3 1
~0"4 \ :1
02 ,
> i
1 0.2
w
0 0
-200 -100 0 100 200 300 400 200 300 400 S00 600 700
TEMPERATURE 4 . °C WAVELENGTH 7, , mm
Figure 6.7 Fluorescent properties of magnesium fluoro-germanate activated with tetravalent Mn.
(a) decay time, r as a function of temperature, S,. (b) relative intensity of exciting and emitted
radiation as a function of wavelength, A
EXTRINSIC SENSING THERMOMETERS 131

emitted radiation shown in Figure 6 .7(b) are quoted by Mei Sun (1992) and Ballico (1997) .
Properties of different materials used for fluorescent temperature sensors in the temperature
range from -190 °C to 1300 °C are given by Fernicola and Galleano (1997) and Grattan and
Zhang (1995) . Fernicola and Galleano (1997) and Zhang et al (1997) state that red and blue
laser light are used as exciting radiation.
An example of a second generation fluorescent thermometer is shown in Figure 6 .8 . In
this Model 3000 thermometer by Luxtron, the sensor is excited periodically by the
microsecond long pulses of a xenon flash lamp (Samulski, 1992) . The radiation emitted by
the sensor is conveyed by a light guide and optical system to the photodiode, whose output
signal corresponds to decaying radiation intensity . After the transformation, this signal is a
measure of decay time, z and thus also of measured temperature . Neither variations of
radiation intensity nor bending of the light guide influence the readings .
Figure 6 .9 illustrates an interesting construction of a fluorescent thermometer for use in
aviation as described by Phillips and Tilstra (1992) . When the sensor is excited by a LED
radiation of wavelength 0 .65 to 0 .69 lun it subsequently emits radiation in the wavelength
range 0 .7 to 0 .98 pm transmitted to the photodiode . The temperature measure is decay time,
r, of the photodiode signal . The thermometer is used in the temperature range from -75 °C
to 300 °C and is intended for temperature measurement of rapidly flowing gas with the
sensor structure shown in Figure 6 .9(b) . The sensor cavity, which slows down the gas flow,
is shaped so that the sensor temperature is as close as possible to that of the gas .
Further details of temperature measurement of rapidly flowing gas are discussed in
Chapter 17 . An interesting application of the phenomenon of fluorescence is an arrangement

TO OTHER CHANNELS
= 3PUMPFIBER

FLAS=A

TRANSFORMATION PHOTODIODE BEAM

SYSTEM SPUTTER

SENSOR OPTICAL FIBRE

Figure 6 .8 Optoelectronic arrangement of a fluorescent thermometer by Luxtron, USA . (Samulski,

1992)

(bl
la! LED FLUORESCENT CAVITY
MATERIAL E-. GAS
SENSOR
SHEATH
d=1, 6 mm
OPTICAL FIBRE OPTICAL
OPTICAL COUPLER FIBRE
SIGNAL PHOTODIODE
PROCESSOR

Figure 6 .9 Optoelectronic arrangement of a fluorescent thermometer TRD (Phillips and Tilstra,

1992) in (a) has the sensor arrangement in (b)
132 FIBRE OPTIC THERMOMETERS

to measure the surface temperature of a turbo-generator rotor in a Canadian 540 MW hydro-

electric power station (Mannik and Brown, 1992) described in Chapter 19 .
Second generation fluorescent thermometers, which are the most popular of all fibre
optic thermometers, give high measurement precision and do not need any periodical
calibration . The fundamental principles of these thermometers with all the important aspects
of the subject are given by Grattan and Zhang (1995) . Their typical applications are
concerned with temperature measurement :

" in medicine and biology, which is considered in more depth in Chapter 21,
" in high voltage appliances (Wickersheim, 1991),
" of rotating bodies, which is discussed in Chapter 9,
" of microwave and dielectric heated bodies,
" in chemical and physical research .

6.3.4 Thermometers with black body sensors

The operating principle of these thermometers is based on the temperature dependence of
spectral radiant intensity emitted by a black body, as given by the Planck's law in
equation (8 .7) :

Cl ~ -s
Wok - e c 2 /,,T _
1

Win2,
where cl =3 .7415xl0-16 C2 =14 388 pm K, A is the wavelength in pm and T is the
temperature in K .
The basic diagram of a fibre optic thermometer with black body sensor is shown in
Figure 6 .10 (Adams, 1992) . At the end of the high-temperature optical fibre, the cavity,
which is covered by a noble metal layer, makes a sensor exhibiting the properties of a black
body . Its radiation is sent by a high-temperature fibre optic light guide, with a low-
temperature extension to a photodiode . The measured temperature is obtained from the
photodiode output signal . The high-temperature light guide, which is made of mono-
crystalline sapphire operating up to 2000 °C (Grattan and Zhang, 1995), is resistant to the
majority of oxidising media . The low-temperature light guide is made of quartz with a

HIGH TEMPERATUREE LOW TEMPERATURE TO SIGNAL PROCESSOR

OPTICAL FIBRE OPTICAL FIBRE
"BLACK BODY"
SENSOR

OPTICAL COUPLER
CAVITY CERAMIC PROTECTIVE FILM
LENS / PHOTODIODE
NARROW BAND FILTER

THIN FILM METAL COATING

Figure 6 .10 Fibre optic thermometer with a `black body' sensor

EXTRINSIC SENSING THERMOMETERS 133

polymer sheath . Depending on the application range of the thermometer, the noble metals
platinum, rhodium or iridium are used for the black body sensor cavity . This thermometer
which is used in the range from 300 to 1900 °C, has an indication error below ±0 .2 % at a
level of 1000 °C and a resolution up to 0 .01 °C . The thermometer sensor follows
temperature variations up to the frequency of 10 kHz. Adams (1992) describes the
applications of this small diameter thermometer sensor, which is immune to the presence of
electromagnetic fields, has a large temperature range, high precision and resolution .
Application examples are quoted as :

" semiconductor production,

" ceramic products sintering, -
" hard soldering,
" plasma etching,
" deposition of diamond film .

6.3.5 Thermometers with Fabry-Perot sensors

The temperature dependent spectral reflection coefficient of a thin mono-crystalline Si film
provides the operating principle of this thermometer . The arrangement of the thermometer
using a LED radiation source is shown in Figure 6 .11 (Saaski and Hartl,1992) . At the end of
the optical fibre, a thin Si monocrystal layer is tightly enclosed between two glass layers .
The radiation of a LED is transmitted to thin film Si sensor, where it is reflected with a
reflection coefficient depending on the wavelength . The reflected radiation is transmitted by
the same optical fibre to a semi-transparent mirror splitting the incoming radiation into two
radiation beams of wavelengths A1 and AZ . The temperature of the thin film sensor is related
to the ratio of the radiation intensity at these two wavelengths, which is directly proportional
to the resulting ratio of the two electrical signals from the respective photodetectors .
These thermometers, which are immune to electromagnetic fields, exhibit high
resolution and operate in the temperature range from 0 to 400 °C . Saaski and Hard (1992)
report the applications of, Fabry-Perot thermometers as :

" cross-linking of polymer composites,

" microwave drying and food processing,
" research on thermal effects of induced electromagnetic fields in aircraft .

} OPTICAL COUPLER

LED OPTICAL FIBRE

SPECTRAL
SIGNAL MIRROR
PROCESSOR LJ
'FABRY-PEROT°
a SENSOR

c LIGHT SHIELD
' a2 LENS
GLASS
Si MONOCRYSTAL
PHOTODETECTORS PEDESTAL

Figure 6.11 Fibre optic thermometer with "Fabry-Perot" sensor

134 FIBRE OPTIC THERMOMETERS

6.3.6 Optical coupling thermometers

A thermometer, based on extrinsic optical coupling of two light guides, is shown in
Figure 6 .12 . Two sheathed optical fibres, which are aligned in parallel, have their sheaths
partially removed at the measuring point . The measuring part is immersed in the liquid,
which makes one common sheath around the two bared cores . If the liquid refractive index,
n2 , is smaller than that of the core, n 1 , no coupling between the two optical fibres occurs . As
the temperature changes, the refractive index, n 2 , becomes larger than n l , so initiating the
optical coupling, which allows transmission of the signal from the light source to the
detector. It can be seen that this thermometer is especially suited to temperature limit
detection, thus approaching the idea of temperature indicators .

6.4 Intrinsic Sensing Thermometers

6.4.1 Raman scattering thermometers
Optical fibres may be applied for the direct measurement of the temperature distribution or
average temperature along its length. When a light pulse is propagated through an optical
fibre, temperature dependent Raman scattering of the light intensity components occurs . As
a result the Raman scattering of two components of a back-scattering light pulse, of
wavelengths different from that of the incident light, are observed . These are called, Stokes
light and Anti-Stokes light. Also a third component called Rayleigh scattering light is
observed as given in Figure 6 .13 .
Iida et al (1992) describe how back-scattering Stokes and Anti-Stokes lights are used in
the measuring arrangement on their return to the incident light point of the optical fibre . The
ratio of their intensities is a function of temperature, T, of that part of the optical fibre at
which the scattering occurs . The path length covered by the input pulses in the optical fibre
up to the point of scattering, is measured by a reflectometcr.
The block diagram of a fibre optic thermometer based on Raman scattering, for
temperature distribution measurement, is shown in Figure 6 .14 (Iida et al, 1992) . The
thermometers based on Raman scattering are usually designed for a certain well specified
problem . Their main application range is to measure the temperature distribution and the
average temperatures on large surfaces and along long objects like pipe-lines (Sandberg and
Haile, 1987) .

FROM LIGHT TO DETECTOR

SOURCE
FIBRE SHEATH
SENSING FIBRE CORE
PART

_ LIQUID

Figure 6 .12 Optical fibres with variable optical coupling

INTRINSIC SENSING THERMOMETERS 135

A special optical fibre construction for temperature measurement along the inside lining
of hot stove blast furnace, as given in Figure 6.15, was described by Iida et al. (1992) . The
a
optical fibre, protected by two special layers, is placed inside a steel tube in nitrogen
atmosphere. The measuring arrangement records the temperature distribution along the
20 m long optical fibre every one meter at temperatures up to 500 °C.

6.4 .2 Refractive index thermometers

The temperature dependence of the refractive index ofoptical fibres is also used for direct
temperature measurement . For example Gottlieb and Brandt (1979) describe the
temperature induced change of refractive index for measuring the average temperature
along a length of fibre . Temperature dependent optical losses at bends in the fibre can
guarantee an average measured value of temperature along a uniformly wound fibre as
shown in Figure 6.16(a) . The fibre can also be bent lightly over short regions connected by
straight sections as given in Figure 6.16(b) . Since the sensitive points are connected by low
loss straight sections, a very long length can be used. Another possibility, also described by
Gottlieb and Brandt (1979), is to use fibres whose bending radius changes with temperature
by using various thermo-mechanical effects, such as a bimetal or others .
Thermometers based on change of refractive index similar to those based on Raman
scattering are usually intended for the solution of certain well defined problems . Grattan
(1987) describes the application of this thermometer for detecting leaks in a low
temperature liquid container . A glass core light guide, covered by a plastic sheath, is placed
below the container . The appearance of cold liquid stops the light wave from being
transmitted because below 25 °C the refractive indices of core and sheath are equal .

STOKES LIGHTANTI-STOKES LIGHT

has / ~a RAYLEIGH SCATTERING

LIGHT
INCIDENT
OPTICAL FIBRE
LIGHT

THE OBJECT OF MEASUREMENT

Figure 6.13 Raman scattering in an optical fibre

SEMICONDUCTOR OPTICAL WAVELENGTH
LASER DIVISION DEMULTIPLEXER

- SYNCHRONISING OPTICAL FIBRE

SIGNAL

SIGNAL STOKES LIGHT

AVERAGER DETECTOR
PROCESSOR

ANT1-STOKES LIGHT DETECTOR

MEASURED TEMPERATURE DISTRIBUTION

Figure 6.14 Thermometer based on Raman scattering for measurement of temperature distribution
136 FIBRE OPTIC THERMOMETERS

(a) (bl
STAINLESS TUBE CERAMICS COATING

NITROGEN *,CARBON COATING

QUARTZ CORE
d=200ym
QUARTZ DOPED F
- OPTICAL SHEATH d=250ym
FIBRE

Figure 6 .15 Optical fibre to measure the temperature in the range up to 500 °C . (a) optical fibre in a
steel tube, (b) optical fibre cross-section

lal Ihl
THE OBJECT OF MEASUREMENT

c-
OPTICAL
FIBRE

Figure 6 .16 Optical fibre with temperature dependent refractive index, (a) measurement of average
temperature, (b) measurement of temperature distribution

6 .5 Summary
Fibre optic thermometers present a rapidly developing branch of thermometry . The most
widespread applications of extrinsic optical fibre sensors are given in Table 6 .2
(Wickersheim, 1992) . Intrinsic types of fibre optic thermometers are mainly used to
measure the temperature distribution of large surfaces and long objects (Sandberg and
Haile, 1987) .
In many applications fibre optic thermometers supersede commonly applied
thermoelectric, resistance and semiconductor thermometers and open new applications of
measuring so far not measurable temperatures (Grattan and Zhang, 1995) .The main merits of
fibre optic thermometers are :

" immunity to electric and magnetic fields and at the same time not influencing them, thus
making possible temperature measurement in induction, dielectric and microwave
heating,
" small diameter (below 1 mm) and low thermal conductivity does not deform the existing
temperature fields,
" being made of isolating material may contact live metallic elements,
" small diameter and flexibility make them especially convenient for temperature
measurement of living organisms,
" immunity to chemical and mechanical influences,
" possibility of coupling several optical fibres with one measuring arrangement
(multichannel measurement),
" low thermal inertia,
" large measured temperature range from about -200°C to +2000°C,
" large fibre optic length ranging from a dozen or so metres in extrinsic sensing
thermometers to some kilometres in the intrinsic types.
REFERENCES 137

Table 6.2 Application of extrinsic sensing fibre optic thermometers

Field of application Examples of applications Temperature

range
Aircraft and aerospace Temperature measurement of engine inlet air Low and medium
Monitoring and temperature control in High
combustion engines, space craft and cermetal
production
Automobile industry Temperature measurement in construction and High
testing of engines
Chemical industry Temperature measurement and control in Low and medium
hazardous areas
Electronic component Temperature measurement in both production Low and medium
production and inspection of components
High voltage appliances Monitoring the temperature of transformer Low and medium
and generator windings and of electrical
supply lines
Laboratories and Combustion process analysis and materials High
science research
Medicine and biology Temperature measurement in therapy using Low and medium
microwave and radio frequencies
Power engineering Temperature monitoring in gas turbines High
Various other industrial Temperature measurement in heat treatment High
processes and production of metals, glass and ceramics

6.6 References
Accufiber, Canada (1987) Catalog. 2000E Columbia Way, Vancouver.
Adams, B.E . (1992) Optical fiber thermometry for use at high temperatures . Temperature : Its
Measurement and Control in Science and Industry. 6(2), American Institute of Physics, New York,
739-744.
Ballico, M.J . (1997) Thermoluminescence of Europium doped Yttria as a precision temperature
sensor in the range 600 °C-1100°C. Proc . TEMPMEKO'96, 6th International Symposium on
Temperature and Thermal Measurements in Industry and Science, Levrotto & Bella, Torino, 437-
442.
Brenci, M. (1984) Thermochromic transducer optical fibre temperature sensor . 2nd Optical Fibre
Sensors Conference, Stuttgart, 155-160.
Culshaw, B . (1982) Optical fibre transducers, Radio and Electronic Engineer, 52, 283-290.
Fernicola, V. and Galleano, R. (1997) Optical fibre thermometry by laser-induced fluorescence Proc .
TEMPMEKO 96, 6th International Symposium on Temperature and Thermal Measurements in
Industry and Science, Levrotto and Bella, Torino, 427-436
Gottlieb, M, and Brandt, G.B . (1979) Measurement of temperature with optical fibres . Fibre Optic
Conference, Chicago, 236-242.
Grattan, K.T .V . (1987) The use of fibre optic techniques for temperature measurement . Measurement
and Control, 20(6), 32-39.
Grattan, K.T .V . and Zhang, Z.Y . (1995) Fiber opticfluorescence thermometry, Chapman and Hall .
138 FIBRE OPTIC THERMOMETERS

Iida, O., Iwamura, T., Hashiba, K. and Kurosawa, Y. (1992) A fibre optic distributed temperature
sensor for high temperature measurements. Temperature : Its Measurement and Control in Science
and Industry, 6(2), American Institute of Physics, New York, 745-750.
Luxtron, USA (1992) Models 1000, 2000, 3000 Fluoroptic Thermometer, Catalog Information .
Mannik, L., and Brown, S .K. (1992) Electrical industry application of fibre optic thermometry
measurement of generator rotor temperature. Temperature: Its Measurement and Control in
Science and Industry, 6(2), American Institute of Physics.
McGhee, J. and Henderson, I.A. (1989) Holistic perception in measurement and control: Applying
keys adapted from classical taxonomy, IFAC Proc Series (1989) No .5 pp . 85-90.
Mei Sun (1992) Fiberoptic thermometry based on photoluminescent decay times, Temperature : Its
Measurement and Control in Science and Industry, 6(2), American Institute of Physics, New
York,. 715-720.
Medlock, R. (1986) Review of modulating techniques for fibre optic sensors, Measurement and
Control, 19(1), 6-13 .
Medlock, R. (1987) The present and future status of fibre optic sensors in industry, Measurement and
Control, 20(3), 14-17 .
Ning, Y.N ., Grattan, K.T .V ., Wang, W.M . and Palmer, A.W. (1991) A systematic classification and
identification of optical fibre sensors, Sensors and Actrs A-29, 21 .
Phillips, R.W . and Tilstra, S .D . (1992) Design of fiber optic temperature sensor for aerospace
application. Temperature: Its Measurement and Control in Science and Industry. 6(2), American
Institute of Physics, New York, 721-724.
Pitt, G.D ., Extance, P., Neat, R.C . and four others (1985) Optical-fibre transducers, Proc IEE, 132(J),
214-248.
Saaski, E.W . and Hard, J.C . (1992) Thin-film Fabry Perot temperature sensor . Temperature : Its
Measurement and Control in Science, and Industry, 6(2), American Institute of Physics New
York 711-714.
Samulski, T.V . (1992) Fiberoptic thermometry: medical and biomedical applications,
Temperature : Its Measurement and Control in Science and Industry, 6(2), American Institute of
Physics, New York, 1185-1190.
Sandberg, C., and Haile, L. (1987) Fiberoptic application in pipes and pipelines. IEEE Transactions
on Industry Applications, 1A-23(6) .
Smolinski, A. (1985) Light guide optoelectronics (in Polish) WKiL, Warsaw
Wickersheim, K.A . (1991) Application of fibre optic thermometry to the monitoring of winding
temperatures in medium and large power transformers SPIE Proceedings 1584, 3-14 .
Wickersheim, K.A . (1992) Fiberoptic thermometry : an overview. Temperature : Its Measurement and
Control in Science, and Industry, 6(2) . American Institute of Physics, New York, 711 - 714.
Zhang, Z.Y ., Grattan, K.T .V ., Hu, Y.L., Palmer, A.W ., and Meggitt, B.T . (1997) Use of blue LED
excitation source for Alexandrite fluorescence thermometry over region from -100 to 600 °C . Proc.
TEMPMEKO'96, 6th Intenational Symposium on Temperature and Thermal Measurement in
Industry and Science, Levrotto and Bella, Torino, 433-436.
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

8
Pyrometers Classification and
Radiation Laws
8.1 Classification of Pyrometers
The simplest and oldest non-contact way of estimating the temperature of a radiating body
is by observing its colour. Table 8 .1 summarises the relationship between temperature and
colour . Using this method, experienced practitioners can estimate temperatures over about
700 °C, with a precision sufficient for simpler heat-treatment processes . This is shown in a
witty way in Figure 8.1, which is taken from Forsythe's paper (Forsythe, 1941) . It was
presented at the historical Symposium on Temperature in November 1939, a symposium
that was a milestone in further development of thermometry .
Table 8.1 Temperature correlation with colours of radiating bodies

Temperature (°C) Colour Temperature (°C) Colour

550-580 Black/purple 830-880 Dark orange
580-650 Brown/purple 880-1050 Orange
650-750 Purple 1050-1150 Yellow/orange
750-780 Dark carmine 1150-1250 Yellow
780-800 Carmine 1250-1320 White/yellow
800-830 Orange/c armine

Figure 8.1 First pyrometric temperature measurement

152 PYROMETERS CLASSIFICATION AND RADIATION LAWS

Pyrometers, also known as infrared thermometers, or radiation thermometers, are non-

contact thermometers, which measure the temperature of a body based upon its emitted
thermal radiation, thus extending the ability of the human eye to sense hotness . No
disturbance of the existing temperature field occurs in this non-contact method . In
pyrometry the most important radiation wavelengths which are situated between from 0 .4 to
20 ltm belong to the visible and infrared (IR) radiation bands .

In addition to the methods outlined in Chapter 1 it is also possible to classify pyrometers

according to their spectral response and operating method, as shown in Figure 8 .2 and
described in more detail later.

Manually operated, or hand operated, pyrometers : In manually operated pyrometers

the human operator is an indispensable part of the measuring channel . Figure 8 .3 illustrates
that the operator's eye acts as a comparator . An eye comparison is made between the

one wavelength (0.65 Eun) disappearing

M O
filament A
pyrometers N E
U R
A A
0.55 pin
L T
two-colour L E
two wavelengths
pyrometers
0.65 ,um
H
E
R
M
A --___--____--___-________-__
total radiation
total
pyrometers
A

I one wavelength
AT photoelectric A
I wavelength band pyrometers + U
O ~E T

M
A
two wavelength i T
two wavelength bands
pyrometers 1
c

:multi-wavelength .
several wavelength bands pyrometers

Figure 8.2 Classification of pyrometers by wavelength and operating method

CLASSIFICATION OF PYROMETERS 153

OPTICAL OPERATOR'S
TARGET SYSTEM EYE

MEASURING
OPERATOR
v INSTRUMENT

REFERENCE
UNIT

Figure 8 .3 Structure of a manually operated pyrometer

radiation from the source with a signal from a reference unit whereupon the operator
activates the read-out instrument.
The following two types belong to the group of manually operated pyrometers :

" Disappearing filament pyrometers based upon matching the luminance of the object and
of the filament, by adjusting the lamp current . The observer's eye is the detector. Their
operating wavelength band is so narrow as to allow them to be regarded as
monochromatic pyrometers of A e = 0 .65 Vim .
" Two-colour pyrometers or ratio pyrometers deduce the temperature from the ratio of the
radiation intensity emitted by the object in two different spectral wavebands, which are
most commonly 0 .55 and 0 .65 pm .

Automatic pyrometers : A simplified block diagram of an automatic pyrometer, which is

shown in Figure 8 .4, is composed of the following main parts :

" optical system concentrating the radiation on radiation detector,

" radiation detector which may be either a thermal or a photoelectric sensor,
" signal converter, conditioning the detector output signal before being displayed,
" measuring instrument, which may have an additional analogue or digital output .

The following four types belong to the group of automatic pyrometers:

" Total radiation pyrometers using thermal radiation detectors, which are heated by the
incident radiation . In reality the wavelength band used is about 0 .2 to 14 Pin resulting
from transmissivity of the optical system.
" Photoelectric pyrometers operate in chosen wavelength bands in which the signal is
generated by photons bombarding a photoelectric detector .
" Two-wavelength pyrometers, also called ratio pyrometers, in which the emitted radiation
intensity in two wavelength bands is compared by photoelectric detectors.

OPTICAL
TARGET SYSTEM
DETECTOR
SIGNAL MEASURING
CONVERTER INSTRUMENT

Figure 8 .4 Block diagram of an automatic pyrometer

154 PYROMETERS CLASSIFICATION AND RADIATION LAWS

" Multi-wavelength pyrometers, where the source radiation, which is concentrated in some
wavelength bands, is incident upon photoelectric detectors . They are used for measuring
the temperature of bodies with low emissivity .

Automatic pyrometers are produced for use in stationary or portable applications .

However, the technical parameters of both types are nearly identical in practice . Stationary
pyrometers, which are usually more robust, can withstand higher ambient temperatures .

8.2 Radiation, Definitions and Laws

8.2.1 Absorption, reflection and transmission of radiation
Thermal radiation is a part of electromagnetic radiation . Let us assume that a radiant heat
flux, (P, defined as a quantity of heat in a unit time, is incident on the surface of a solid . Of
this heat flux, the portion, (Da , is absorbed, whilst (D is reflected and (D T is transmitted .
P

The following definitions are introduced :

" absorptivity, a = (Da /(D

" reflectivity, p = q) /(D (8 .l)
P

" transmissivity, r = (D, /(D

Applying the principle of energy conservation shows that for every solid :

a+p+r=1 (8 .2)

In the case of transparent bodies, as represented in Figure 8 .5, many internal reflections
cause additional absorption . For example, Harrison (1960) notes that the total reflected heat
flux, (D , is composed of the primary heat flux (DpI , and a secondary one `f p .
P

L. _
REFLECTED
FLUX TRANSMITTED
FLUX
OF~~Df1 P2 Owl
~T

m ms ABSORBED
INCIDENT / FLUX
`
HEAT FLUX ="

Figure 8 .5 Decomposition of the heat flux, (1), in a transparent body

RADIATION, DEFINITIONS AND LAWS 155

There are three specific cases :

I . a =1, p = 0, ,r = 0 the body is a black body, which totally absorbs all incident radiation .
2. a = 0, p =1, r = 0 the body is a white body, which totally reflects all incident radiation .
3 . a = 0, p = 0, z =1 the body is a transparent body as all of the incident radiation is
completely transmitted .

The concept of a black body is very important in pyrometry . Figure 8 .6 presents some
configuration properties approaching those of a black body. Heinisch (1972) shows that in
the cavities presented in Figure 8 .6, total absorption of the incident radiation is reached by
its multiple internal reflection .
Similarly to the factors, a, p and z, which are valid for total radiation, the spectral
properties, ax , pX and z?, at the wavelength A, may also be introduced :

as = (DA, /)

pa =(DAP /(' (8 .3)

zX = (DXt /(D

Equation (8 .2) then becomes :

ak + pa + TX =1 (8 .4)

The values of a, p and z depend upon the material, its surface state and temperature while
ax , p) and z), additionally depend upon the wavelength, A.

8.2 .2 Radiation laws

The radiant intensity W or the radiant exitance is the heat flux per unit area expressed as
the ratio of the heat flux dD, emitted from the infinitesimal element of the surface dA, to the
surface area dA itself:

(8
W = ~ W/m 2 .5)

(a) (b) (c) (d)

r

1 II, ~I

Figure 8.6 Models of a black body

156 PYROMETERS CLASSIFICATION AND RADIATION LAWS

In the same units as the radiant intensity, the heat flux density, q, of the incident radiation is
given by:

q = ~ W/m2 (8 .5a)

This also takes account of the conduction and convection heat flux in addition to the
radiation heat flux .
The spectral radiant intensity, Wk, is defined as :

=
dW
W~ W/m2 pin (8 .6)

Planck's law gives the radiant flux distribution of a black body as a function of the
wavelength and of the body's temperature by the relation :

WOA (8 .7)
°~ - C C 2 IRT -1

where W°A is the spectral radiant intensity of a black body, W/m2 pm (the suffix `o' will be
used in future to indicate a black body), A is the wavelength, pin, T is the absolute
temperature of the thermal radiator, K, c, is the first radiation constant whose value is

c i = 3.7415 x 10 -16 W m 2 and c2 is the second radiation constant with a value of

c2 = 14 388 pin K .
For a given wavelength range, from Xi to )L2 , equation (8 .7) can be evaluated as :

elf-s
Az

where W°, ~ _,~2 is the band radiant intensity of a black body .

Hackforth (1960) has shown that if AT << c2, Planck's law of equation (8 .7) can be
replaced, using the same notation, by a simpler Wien's law :

el V
eC2 11T

The spectral radiant intensity W°A of a black body as a function of wavelength A, at

different temperatures, calculated from Planck's law, is shown in Figure 8 .7 At all
temperatures of importance in radiation pyrometry, the errors, which result from replacing
Planck's law by Wien's law, are negligibly small . The relative errors may be calculated
from the relation :
RADIATION, DEFINITIONS AND LAWS 157

ay
AWo; _ W.X,w - W.R,P1 = _ /XT
e 8 .10
W W.,~,w

where Wo3,W is the spectral radiant intensity calculated from Wien's law and W X , p l is O
calculated as above from Planck's law .
The relative errors calculated from equation (8 .10) are presented in Table 8 .2 as a
function of the values of the product AT.

Table 8 .2 Relative errors resulting from replacing Planck's law in equation (8 .7) by
Wien's law in equation (8 .9) as a function of the value of the product AT

2x10
3x10
AT (m.K) 1 .25 x 10 -3 1 .5 x1 0 -3 -3

W.),/ W.?, (%) 0 .001 0 .007 0 .08 0 .8

Figure 8 .7 shows that the maxima of the spectral radiant intensity are displaced towards

the shorter wavelengths with increasing temperature . At the given temperature, T, where the

maximum is reached, the wavelength Amax , may be easily calculated from Wien's

displacement law to obtain :

Amax T = 2 896 ltm K (8 .11)

For any given temperature, the area under the corresponding curve is a measure of the total
power radiated at all wavelengths by a black body so that :

W o = fWa
,'-0 Ad1 (8 .12)

The ratio of the spectral radiant intensity, Wj, at the wavelength, A, of a non-black body to
the spectral radiant intensity of a black body, Wo j, at the same temperature is called the
spectral emissivity cA.

W;L
E~ = (8 .13)
Wo;

If the spectral emissivity el of a given body is constant for each wavelength (i.e.
al =constant) such a body is called a grey body . Similarly to equation (8 .13), if all
wavelengths from 0 to oc, are taken into consideration, the term total emissivity, s, is used:

W
W
e= (8 .14)
0
158 PYROMETERS CLASSIFICATION AND RADIATION LAWS

where W is the radiant intensity of any given body and Wo is the radiant intensity of a black
body at the same temperature .
Following Kirchhoff's law, the spectral absorptivity, ak, of all opaque bodies equals
their emissivity, cX, so that :

For a given wavelength band, from A, to ~2 , Kirchhoff's law is expressed by :

- band ii i band

11

10- V

' - -_
1

' , ., ,

-'.
.; . ,

140 K

0 1 2 3 4

WAVELENGTH 1. jim

. .. . . lack body, W~, versus wavelength at different temperatures

in accordance with Planck's law in equation

RADIATION, DEFINITIONS AND LAWS 159

When all wavelengths from .11 --> 0 to '12 --> oo are taken into consideration, the
corresponding form for equation (8 .15a), which is also valid, then becomes :

(8 .15b)

where a is the total absorptivity, and s is the total emissivity.

The Stefan-Boltzmann law, which represents the dependence of the total radiant
intensity, W., of a black body upon the temperature, T, is expressed as :

a
(8
W,, Wo,d;, = aO T .16)
= JA o

where W,,
; is the spectral radiant intensity of a black body as given by Forsythe (1941),

The radiation constant of a black body, a,, has a value a, = 5 .6697x10_ 8 W/m 2 K4 .
Equation (8 .16) can be expressed in a more readily usable form as :

4
Wo (8 .16a)
C°(100)

where Co is the technical radiation constant of a black body, with the value :

Co = 6 o x 10 8 = 5.6697 W/m 2 K4

For grey bodies equation (8 .16a) becomes :

4
( T
W = C°£ (8 .17)
100 )

where Co is as before, and 8 is the total emissivity .

In technical practice the majority of real bodies may be regarded as grey ones .

8.2.3 Total emissivity and spectral emissivity

Spectral emissivity e A and total emissivity, e were defined by equations (8 .13) and (8 .14) .
Knowledge of the values of c and e,, , especially at A = 0 .65 Vm, for different materials, is
necessary, to be able to calculate the corrections to be introduced when making pyrometric
temperature measurements . The emissivity of different materials, which depends heavily
upon the surface state, its homogeneity and temperature, may only be determined
approximately . Worthing (1941) describes methods for the measurement of emissivity .
Comparison of the properties of different materials, independent of their surface state
may be made using the specific total emissivity, e', and the specific spectral emissivity,
E'. . The values of e' and e ;L are determined for the direction normal to surface for flat
160 PYROMETERS CLASSIFICATION AND RADIATION LAWS

samples, which should be polished and sufficiently thick. This last condition allows semi-
transparent bodies to be regarded as totally opaque . The values of E and E,~ are also
determined for the direction normal to the surface. Approximate values for the emissivity of
different materials are given in Tables XIX and XX.
It must be stressed that uneven, rough and grooved surfaces may have much higher
values of emissivity than are their specific emissivities.
Using the Maxwell theory of electromagnetism, Considine (1957), following Drude,
have proposed an approximate formula to calculate the specific spectral emissivity, E;1, of
metals as:

E;1 = K (8.18)

where K = 0.365 S2-'"z, p is the resistivity in S2cm, and A is the wavelength in cm .

Equation (8.18) which is valid for A > 2 pm, uses the original units of Drude. The emissivity
of non-conductors, which is a function of the material refractive index, n,l, is given in
BS 1041, p . 5 by the formula :

4nj
E,~ = (8.19)
+1)2
(nX

where nA which is the refractive index of the material, has a value in the range of 1 .5 to 4
for most inorganic compounds and in the range 2.0 to 3.0 for metallic oxides. For most
clean metals the emissivity is low, with a value of about 0.3 to 0 .4, falling sometimes to 0.1
for aluminium . Spectral emissivities of metals become lower at lower temperatures where
the wavelengths are longer. Non-metallic substances have emissivities of about 0 .6 to 0.96,
which do not vary greatly with temperature . It should be borne in mind, that the appearance
of non-metals in visible light cannot be a basis for predicting their emissivities. Most non-
metals, such as wood, brick, plastic and textiles at 20 °C have a value of total emissivity
nearly equal to unity.

8.2.4 Radiant heat exchange

Consider two parallel surfaces, having identical areas A and the respective temperatures and
emissivities TI , T2, E l , 02, emitting thermal radiation towards each other with the
intensities given by the Stefan-Boltzmann law in equation (8.16a) . The heat flux (power)
(D12 exchanged between these surfaces, for T1 > T2, is given by:

2
CO
012 (8 .20)
(1/El )+(/ E2)-1 (100)4 -(100)4

where Co is the technical radiation constant, and A is the radiating area.

RADIATION, DEFINITIONS AND LAWS 161

If one of the bodies of area A I is placed inside another one of area A 2 and with A 1 < A2,
then equation (8 .20) becomes :

)4-(100)4
,
biz (8 .20a)
(1/s,)+(A~~A2 )[(1/EZ)-1] [( 100

In the very important practical case when A 2 > 3A 1, equation (8 .20a) becomes :

(101)4 00\4
X12 = A1E1Co -~ (8 .20b)
[

Lambert's directional law which describes the radiant intensity of a black body as a
function of the radiation direction, is given by :

Woe = Way_ cos (p (8 .21)

where Wo,, is the radiant intensity of an element of area under the angle rP between the

radiation direction and the direction normal to the surface, and Woe is the radiant intensity
as before but in the direction normal to the surface .
Radiant intensity, W,1, in the direction normal to the surface is 7T times smaller than the
total radiant intensity .

W
Wo1 = . (8 .22)

Equation (8 .21) is only partially valid for non-black bodies . Large deviations from
Lambert's law, which can be observed especially for polished metals when rp > R/4, are
caused by the dependence of the emissivity upon the observation angle . Some definitions,
taken from illumination technique, are also used in optical pyrometry, in the case when the
thermal radiation takes place in the visible wavelength range . Luminosity, I,,, is the radiant

flux propagated in an element of solid angle . Radiance, L, also called luminance, which is a
density of luminosity of a surface in a given direction, is expressed as :

dig
L=- (8 .23)
cos (P dA

where dA is the area of an element of the radiating surface and T is the angle between the
radiant flux direction and the direction normal to the surface .
Radiance is a deciding factor in the subjective impression of the body's brightness .
162 PYROMETERS CLASSIFICATION AND RADIATION LAWS

Lambert's law of equation (8.21) which is also valid for the luminosity, is :

1(P = I 1 cos tp (8 .24)

where Il is the luminosity in the direction normal to the surface .

Combining equations (8 .23) and (8 .24), yields :

L _ d1 l cos tp = d-[1
.25)
dA cos (p dA (8

From equation (8 .25) it follows that the radiance of a black body is independent of the
viewing angle and is always the same as in the direction normal to the surface . For the
majority of non-black bodies, the radiance is nearly constant for ~0 in the range from 0 to
n/4 . A detailed description of radiant heat transfer is given by Jakob (1959) and Gr6ber et
al. (1963) .

8.3 References
Considine, D . (1957) Process Instruments and Controls Handbook . McGraw-Hill, New York .
Forsythe, W .E. (1941) Optical Pyrometry. Temperature : Its Measurement and Control in Science and
Industry, Reinhold Publishing Corporation, New York, 1115-1131 .
Gr6ber, H ., Erk, S . and Grigull, U . (1963) Die Grundgesktze der Wlirmeubertragung, Springer
Verlag, Berlin
Hackforth, H .L . (1960) Infrared Radiation, McGraw-Hill, New York .
Harrison, T .R . (1960) Radiation Pyrometry and its Underlying Principles of Radiant Heat Transfer,
John Wiley and Sons, New York.
Heinisch, R .P . (1972) The emittance of black body cavities . Temperature : Its Measurement and
Control in Science and Industry, 4(1) . Instrument Society of America, Pittsburgh, 435-448 .
Jakob, M . (1957) Heat Transfer, John Wiley and Sons, New York.
Worthing, A .G. (1941) Temperature radiation emissivities and emittances, Temperature: Its
Measurement and Control in Science and Industry, Reinhold Publ. Co ., New York, 1164-1187 .
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

14
Imaging of Temperature Fields
of Solids
14 .1 Introduction
Infrared thermal imaging of temperature fields has become an extremely versatile and
popular method of real time temperature measurement and thermal condition monitoring in
all industrial and research applications. In the near future, the rapid further development of
the method is more than guaranteed, as it will replace many existing temperature measuring
methods, offering exciting new opportunities with the added most important benefit of a
two-dimensional look at problems.
The first marketed infrared imagers were based on the use of an opto-mechanical
scanning system to sequentially scan the target surface . The resulting series of radiated
image signals is transformed into a series of electrical signals in an appropriate detector. In
the subsequent step in their development, the opto-mechanical scanning system was
replaced by a matrix of mainly photoelectric detectors. However, since the detectors had to
be cooled, the device was big and heavy . The majority ofcontemporary imagers is based on
an array of un-cooled micro-bolometers, which can operate at ambient room temperature .
Infrared thermal imaging systems can be divided into two main groups. In surfacial
systems two-dimensional imaging of temperature fields is produced using scanning and
matrix systems . In the second group, which may be referred to as linear systems, the aim is
to produce a temperature field image of continuously cooled or heated moving targets .
These linear systems present a certain simplification of two-dimensional systems .

14.2 Surfacial Systems

Opto-mechanical scanning infrared systems . Among the first commercially available
infrared systems were the two-dimensional opto-mechanical scanning systems, produced
since 1965 by AGEMA Infrared Systems . As shown in Figure 14.1 all of the target surface
points were scanned sequentially in these systems, which were appropriately called
Thermovision systems . Rotating or oscillating mirrors or prisms were used in the scanning
system. At a scanning frequency of 25 Hz, a spatial resolution of 100 to 175 measuring
elements per line is achievable . Detectors of InSb or (HgTe),, (CdTe) 1_X which are cooled
either thermoelectrically or by liquid nitrogen, ensure a temperature resolution of 0 .1 °C in
274 IMAGING OF TEMPERATURE FIELDS OF SOLIDS

IR-DETECTOR
w

I
LENS

LENS VERTICAL HORIZONTAL

ROTATING ROTATING
PRISM PRISM

Figure 14.1 Thermovision scanning system

a measuring range of 30 °C . The measurement ranges are between -20 and +1500 °C
having a sensitivity of 0 .07 to 0 .1 °C and accuracy of ±2% or ±2 °C . The resulting series of
signals was transformed in the detector into electrical signals, displayed on a monitor screen
as a visible image of the temperature field, with a reference colour scale . Chosen isotherms
could also be marked . Microprocessor based systems vastly extended the imaging
possibilities by making the whole assembly a really universal and precise measuring tool .
Although this whole family of Thermovision systems is no longer produced, it is still widely
used by their still numerous owners. In spite of their high price they were of a high quality
and commensurate robustness .

Matrix infrared systems . Surfacial systems, based on a Focal Plane Array (FPA) matrix,
are now the most popular since they do not need any moving parts in the scanning system .
There are two groups of FPA . In the first one photon detectors based on photovoltaic
(HgCdTe and InSb) or photo-conductive (PbSe, PbS) cells are mainly used . In the second
the detectors are either thermopiles or micro-bolometers whose operation is based on the
absorption of thermal energy . The signals from particular detectors, corresponding to
different target points, are scanned by a contactless commutator, amplified if needed before
forming a colour picture on a screen .
FPA systems formed by monolithic PtSi micro-bolometers are now the most popular,
since they allow the integration of many detectors in one integrated matrix of 256x256 or
even 320x240 pixels . Moreover, these micro-bolometer FPAs have a considerable benefit
since they do not require any cooling . Compensation for small ambient temperature
variations is achieved by temperature sensors inside the camera . The size of each FPA pixel
is about 30 ~tm square (Santa Barbara Research Center) .
The ThermaCAM PM 695 by FUR Systems (2001) shown in Figure 14 .2, is a good
technical example of a surfacial matrix system . This imager is equipped with an uncooled
MicroIR solid stated detector matrix, incorporating 320x240 pixel micro-bolometers,
connected by a contactless commutator . The coloured target picture on the view-finder or an
LCD screen, which is obtained from the matrix, is quite stationary due to the non-existent
inertia of the micro-bolometers and rapid commutation of the scanning electronics . At the
same time a colour-temperature scale is provided for use by the operator. The new
electronic techniques and processor speed provides rapid operation and operates in
SURFACIAL SYSTEMS 275

~ i
=

Figure 14 .2 Infrared ThermaCAM PM 695 system (Courtesy of FUR Systems)

wavelength range 7.5 to 13 .0 pun, thus eliminating the influence of solar radiation on the
readings. The system is especially useful for detecting hot points in different technical
installations . Digital voice recording, which allows detailed annotations for each stored
image to be entered, is also included in the system.
The ThermaCAM PM 695 is the first infrared system having an integrated digital
camera. It also includes ThermaCAM Reporter software . This enables the production of
automatic reports, which include all field inspection data such as thermal, visual,
measurement analysis as well as voice and text. The operator can store at the touch of one
button both infrared and visible images .
The technical data of ThermaCAM PM 695 are as follows :

" field of view : 24°x 18° at a minimum distance of 0.5 m,

" detector : uncooled micro-bolometer MicroIR,
" spectral range: 7.5-13 pm,
" colour image in viewfinder, LCD screen or video output,
" temperature range: -40 to 120 °C or 0 to 500 °C with options up to 1500 °C or 2000 °C,
" accuracy: ±2 % of range or ±2 °C,
" possible correction of target emissivity,
" measurement functions:
- a cross-point which is movable over entire image,
- horizontal or vertical temperature profile with movable spot measurement,
- isotherms,
- automatic reading of max, min and average values in the view field,
- difference temperature relative to a reference temperature,
" weight with battery 2.4kg, size 220xl33xl4Omm,
" remote operation via RS 232 data output .

Light infrared cameras. Some producers offer these for applications, where they can replace
much more expensive thermal imaging systems. They are used for observing and recording
thermal images of stationary objects. An example of such a camera is digiCAM-IR by Ircon
Inc. (1999) . This camera is equipped with an array of uncooled thermopile detectors, giving
an image scan time below 1 .5 s. Image display is on 4 .0 inch active matrix colour LCD
276 IMAGING OF TEMPERATURE FIELDS OF SOLIDS

screen, with RS232 communication and an accuracy within 2 °lo of reading . A "Hot Spot
Mode", which is unique to this camera, allows automatic production of an image
concentrating on the hot spot of the image . Four buttons are used by the operator to
manoeuvre the cursor over the image displaying the corresponding temperature values . The
spectral range is 8 to 12 gm. Up to 140 images can be stored on smart card . The camera,
which has an overall weight of 2 .0 kg including the lens and battery, has a physical size of
240x 100x 130 mm.

14 .3 Linear Systems
Two-dimensional colour pictures of temperature fields of continuously heated moving
charges is easily possible using the FPA matrix systems described in Section 14 .2 . However
it is much simpler to apply what are called linear systems . There is no need for electronic
scanning of the second dimension since this is replaced by the charge movement as shown
in Figure 14 .3 .
The Landscan Infrared Linescan System by Land Infrared (1997) is a representative
example of this technique . The system consists of a sensor head, performing the temperature
measurement, and a dedicated Landscan software program for a PC or a dedicated LPU 1
microcomputer . Ethernet capabilities, which are provided using the LPU 2E model of this
system, can provide data over a network to a process control computer . The sensing head in
the six different models of the camera, has a Si or Ge detector, which is scanned by a
rotating mirror to produce the line scanned image . The scanning occurs along a line,
perpendicular to the charge movement .
For different models the technical data of the system are :

" temperature range : 70-350 °C to 800-1400 °C,

" wavelength range : 1 pm to 3-5 pm,
" scan angle : 60 °C,
" emissivity adjustable from 0 .2 to 1 .0 by internal switch or by 4 to 20 mA isolated input,
distance ratio 11d: 120/1 or 30011,

LINE SCANNER

SCANNING
ANGLE

MOVING
CHARGE

SCANNING
LINE

Figure 14.3 Principle of linear scanning of moving charges

APPLICATIONS 277

" focus : 300 to 3000 mm,

" detector time constant : 6 to 25 ps,
" dimensions : 430x230x165 mm,
" scan speed : 5 to 25 scans/s, having 10 000 points in one cycle,
" accuracy : ±5 °C,
" additional laser aiming,
output : up to 14 analogue outputs 0/4-20 mA .
There are four basic display modes :

1. Two-dimensional colour picture (map) with a colour temperature scale .

2. Temperature profile across the moving charge .
3. Zones of given temperature .
4. Three-dimensional, colour temperature relief.

As in the Land Landscan and Ircon ScanlR II systems, a pixel line with an output signal
commutator could be used instead of one detector and a rotating mirror for scanning .

14.4 Applications
The range of applications of thermal imaging systems is extremely large including the
analysis of thermal problems, monitoring operation of industrial installations, investigation
and optimisation of different prototypes and all types of research .
Some typical applications of surfacial systems are :

" Monitoring and detection of hot points of moving or rotating equipment like bearings,
gears, clutches, shafts, chains, transporters, pumps and blowers among others .
" Detection of
- hot points in electrical installations such as in fuses, contactors, power
networks, transformers, motors, isolators and cables .
- overheated points in the thermal insulation of different furnaces, dryers,
pipelines and boilers .
- heat leaks in central heating installations and cold storage plants .
- insufficient thermal insulation and moisture in buildings .
- overheating in store houses, grain silos or dumps .
" Temperature imaging of electronic circuits .
" Tracing of steel reinforcements in concrete structures after their preliminary induction or
resistance heating .
" Checking the heating of car tyres while running .
" Medical and veterinary diagnostics .
" Surveying the temperature fields of land and waters from the air .

Linear System can be used for the following continuously heated charges :

" tape and plate rolling,

" heating in continuous furnaces,
278 IMAGING OF TEMPERATURE FIELDS OF SOLIDS

" rod and wire drawing,

" continuous casting,
" continuous induction heating,
" glass heat treatment, hardening and annealing,
" float glass manufacture,
" galvanising .

14.5 References
FLIR Systems (1999) Catalog, ThermaCam PM 695 .
FUR News, (1999) Summer 1999.
IRCON Inc . (1999) Catalog, digiCam-IR .
Land Infrared (1997) Industrial Non-contact Temperature Measurement.
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

15
Dynamic Temperature
Measurement
15.1 General Information
The term dynamic temperature measurement covers all measurements during which thermal
transients occur in a sensor irrespective of whether the transient is caused by temperature
variations in the medium, whose temperature is to be measured, or in the temperature
sensor, itself. Thus, unavoidable dynamic errors occur during the measurement of any
temperatures, changing with time. Errors also arise during the temperature measurement of
a medium at a constant temperature using a temperature sensor immersed in the medium.
Determination of the dynamic errors of a thennometer, requires knowledge of its dynamic
properties. In many non-electric thermometers where the sensor and indicator form one
inseparable unit, the dynamic properties to be described must refer to the whole device .
Electric thermometers are mostly used when it is essential to know the dynamic error so
that it can be taken into consideration . Consequently, the dynamic parameters of electric
sensors will be the main topic for discussion in this chapter. It must be stressed that dynamic
errors in temperature measurement are principally caused by the sensor. For this reason, any
influence of the dynamic properties of indicating instruments may be neglected in most
cases.
Knowledge of the dynamic properties of a temperature sensor is necessary for the
following main cases:

" to determine the necessary immersion time, while measuring a constant medium
temperature,
" to determine the dynamic errors while measuring temperatures changing with time,
" to compare the dynamic properties of different temperature sensors, so that the one best
suited for a specific application, may be chosen,
" to determine the true temperature variations of temperatures changing in time by
correcting known indicated values,
" to describe the dynamics of a sensor when it is part of a closed loop temperature control
system as described by Michalski and Eckersdorf (1987),
" to choose the type and optimum settings of a corrector of dynamic errors .
280 DYNAMIC TEMPERATURE MEASUREMENT

As the problem of the dynamics of temperature sensors is approached in different ways,

the number of existing references concerned with these dynamics is very large .
Consequently only concise principles are presented in this chapter.

15.1 .1 Dynamic errors

Dynamic error can be defined as the difference between the sensor temperature, t9T (t), and
the temperature, 0(t) , measured by another inertia free sensor, exhibiting the same static
errors . Consequently the dynamic error is that part of the systematic error which varies with
time . To simplify the problem, assume that the static error equals zero . Using Figure 15 .1,
the dynamic error is then defined as :

AOdy. (t) = OT (t) - t9(t) (15 .1)

Measuring the stepwise changing temperature, the relative dynamic error is defined as :

A6dy ( t) t9T(t)-6(t)
Sz9dyn(t)= - (15 .2)
A6 A6

where A9 is the value of the temperature step .

The response time, tr, after which the relative dynamic error does not exceed a certain
value, is closely connected with the relative dynamic error . For instance at t >_ tr,5 % the

absolute value of the relative error is 60d,(t) <<-5% . The dynamic error can also be

defined for other, non-periodic temperature variations . Its value is then mostly related to the
maximum change of the measured temperature .
In dynamic temperature measurement Hofmann (1976) asserts that it is necessary to
determine the dynamic errors in two cases . The first occurs when the measured, indicated
temperature and the sensor's dynamic properties are known . Another occurs when the
medium temperature, the input, is known as a function of time as well as the dynamic
properties of the sensor . In both cases it is convenient to represent the dynamic error using
the Laplace transform to obtain :

A6dyn (s) = 6r (s) - 6(s) (15 .3)

TEMPERATURE "Tit) tok G~d ,it)=4T ( tl-~(t)

SENSOR ~ -

Figure 15 .1 Definition of dynamic error

GENERAL INFORMATION 281

When measuring a sinusoidally changing temperature 6(CV) = A6sin 0ot +tip the

dynamic errors consist of amplitude and phase errors . Amplitude error, AA(o)), is given by
the difference of the amplitudes of the sensor temperature 019T (c)) and its true value A6,
so that :

AA(o)) = Otfr((o) - A6 (15 .4)

or the relative error, 8A(Co), related to the amplitude of the measured temperature may be
written as :

AOT(o))-019
SA(o))= AA ( co) = (15 .5)
AO A6

It is also given by the amplitude ratio :

11(co) = A QUO)) (15 .6)

Phase error is defined as the phase shift between the sensor temperature, OT (t), and the
measured temperature 6(t) . The cut-off frequency, f, of a temperature sensor defines a
frequency below which the amplitude error does not exceed a given value (e.g . iff <fc,5
then 1,5AJ < 5 %) when a state of stationary oscillations exists .

15.1.2 Dynamic properties of temperature sensors

Modelling temperature sensors consists of four main steps decribed by Jackowska-Strumillo
et al. (1997) . Applying this process eventually allows the dynamic properties of a
temperature sensor to be described by the following equation :

. .. ...
F[Yn(t),Yn-1(t) Y(t) ;6m (t),6 m-1 (t) . . . . . .t9(t)] = 0 (15 .7)

where y(t) . . . . . . y" (t) are the sensor output signal and its time derivatives, and

13(t) . . . . . . 0 m (t) are the measured temperature and its time derivatives .
In the case when the dynamic behaviour is linear, equation (15 .7) becomes :

~ tg(t)
ja i
d l Y(t)
_ Ybj d (15 .8)
dt` dti
t-o j-o

where a i and bj are constant coefficients and m < n .

The transfer function, G 7 (s), can be used to describe and present the dynamic properties
of a temperature sensor . It is the ratio of the Laplace transform of the sensor output signal,
282 DYNAMIC TEMPERATURE MEASUREMENT

y(s), to the Laplace transform of the measured temperature signal, 9(s), when the initial
conditions are zero so that:

GT(s)= (15 .9)

As)

Taking equation (15.8) into account the transfer function can then be given by:

Ebjsj
m

= J 0 = L(s) (15.10)
GT (s)
Eals i M(s)
n

i=0

The operational transfer function, GT(s), is thus presented as the ratio of the
polynomials L(s) and M(s). Poles of the transfer function are the roots of the
equation M(s) = 0 and the zeros of the transfer function are the roots of the
equation L(s) = 0. If the temperature input signal, 9(s), is known, a knowledge of the
transfer function, GT(s), of a temperature sensor, enables its output signal to be determined
as a function of time (Doetsch, 1961).
Each electric temperature sensor may be regarded as composed of a thermal conversion
stage and an electrical conversion stage as shown in Figure 15.2 . In the thermal conversion
stage, the temperature, 9(t), of the medium whose temperature is being measured is
converted into the sensor's temperature, t9T(t) . The sensor's temperature, OT (t), is
converted into the electrical output signal y(t) (e.g. thermal emf) in the electrical conversion
stage. This second conversion stage has a purely static character. Thus, the sensor transfer
function GT(s) of equation (15 .9) can be expressed as a product of the transfer function of
the thermal conversion stage, FT(s), and ofthe coefficient KT, representing the properties of
the electrical conversion stage, and called the sensor gain so that equation (15.11) is
obtained:

(a)

4(t) THERMAL (t) ELECTRICAL y (t)

CONVERSION CONVERSION
STAGE STAGE

t-DOMAIN PRESENTATION
(b)

~(s) 4T (s) y(s)

FT(s) KT

s-DOMAIN } PRESENTATION

Figure 15.2 Block diagram of a temperature sensor

GENERAL INFORMATION 283

GT (s) = KTFT(s) (15 .11)

where :

KT = (15 .11 a)
d~T
(s))
FT = (15 .11 b)

This approach to the presentation of sensor dynamics, makes it possible to limit further
discussions of the dynamics to those ofthe thermal conversion stage.
In the case of steady-state periodic variations of measured temperature, the frequency
response, GT (jco) , of the sensor may be considered instead of the sensor transfer function,

GT(s) . The sensor frequency response is the ratio of the phasor values of the output signal

y(jro) to the phasor value of the variable component of the sinusoidally changing
measured temperature 6(jo))

y(jw) (15 .12)

GT(jCO) =
Nja))

where co is the angular frequency and j = .

The frequency response ofa sensor can be obtained by substitutingjco in equation (15 .9)
in place of the operator, s. Consequently from equation (15 .10) the frequency response
becomes:

m
Y,bi(jo)) j
L(jo)) (15 .13)
GT(jco)= j=0 =
M(j~)
~)
nYai (jco)i
i=0

In a similar fashion, the transfer function of equation (15 .11) can be rewritten as the
frequency response function :

GT(jO)) = KTFT(jo)) (15 .14)

where KT is the sensor gain, and FT(Iw) is the frequency response of the thermal stage.
Another way of expressing the frequency response of the sensor is:

GT (jo)) = KT [P(o)) + jQ(o))) (15 .15)

284 DYNAMIC TEMPERATURE MEASUREMENT

where

P(o)) = J2e [FT (jco)] (15 .15 a)

Q(co) = Jm [FT(j co)] (15.15b)

where JZe means "take the real part of and Dm means "take the imaginary part of'. Using
the polar form for the complex variable

AY((D) (15.16)
GT(j(O) = exp[j(VO)]
Ad

where Ay((o) is the amplitude of the output signal y(jw), AO is the amplitude of the first
harmonic of measured temperature O(jco) and (p(o)) is the phase shift between y(jO))
and 6(j))
.
Equations (15 .15) and (15 .16) are related as :

Ay ) 2
= JGT(jp)j = KT P2 (p)+Q (0 ) (15 .17)
A~

(p(co) = argGT Go)) _ -arctan Q (15.18)

( )
Ay((o) l AO of equation (15 .17) and (p(o-)) of equation (15 .18) are respectively called the
amplitude and phase characteristics of a temperature sensor. Sometimes it is more
convenient to use the amplitude of the sensor temperature AOT((O) instead of the amplitude
ofthe output signal Ay((o) .
The ratio :

) PZ((0)+QZ((0) (15.19)
A = =IFT(j(o)I
0~CO

is called the amplitude characteristic of the thermal conversion stage of the temperature
sensor.

15.2 Idealised Sensor

The existence of an idealised temperature sensor, designed as a homogeneous cylinder
made of a material having infinitely great thermal conductivity ~ , will be assumed . Let this
sensor have the mass m, specific heat c and a surface area, A, for heat exchange with its
surroundings . During the temperature measurement of a liquid or gaseous medium, the
sensor is completely immersed, so that it does not exchange heat with any other medium
with a different temperature . As an example, any electric temperature sensor can be taken,
provided that it is connected with the indicating instrument by extremely thin wires. It is
IDEALISED SENSOR 285

further assumed, that the thermal capacity, mc, of the sensor is negligibly small compared
with the total thermal capacity of the medium and that the heat transfer coefficient a ,
between the sensor and the medium is constant .

15.2.1 Transfer function

To set up the differential equation, which describes the sensor's dynamics and thus its
transfer function, the method of heat balances will be used. Assume, at the time t = 0-, an
infinitesimally small time before zero, that the sensor is in a steady state, with its
temperature equal to the ambient temperature OT = 0, At t = 0+ immerse the sensor in the
medium at temperature 0 , higher than the ambient temperature so that 0 > Oa . For the
temperature excess over the given reference value the notation O is introduced. In this
book O is also simply referred to as temperature . The initial conditions at t = 0- are given
by:

OT =ZT -Oa =0 and O=6-6a >0

According to Newton's law, when the sensor is immersed in the medium, the heat
transferred to the sensor in the time interval dt will be:

dQ = a4(E) -OT )dt (15.20)

where a is the heat transfer coefficient between sensor and medium and A is the heat
exchange area.
The heat stored in the sensor is :

dQ=mcdOT (15.21)

where m is the mass of the sensor, and c is the specific heat of the sensor material.
From equations (15.20) and (15 .21) it follows that

(15.22)
a4(0- OT)dt = mcdOT

or

nic deT
+OT =O (15.23)
a4 dt

Introducing the notation:

me = NT
(15.24)
aA

Equation (15 .23) can be expressed as :

286 DYNAMIC TEMPERATURE MEASUREMENT

d
NT T + OT = O (15.25)
d

where NT , which is called the sensor time constant at the given heat transfer conditions
(a = constant), is expressed in time units.
Taking the Laplace transform, equation (15 .25) becomes:

NTSO T (s)+OT (s)=0(s) (15 .26)

where s is the Laplace operator . Defining the transfer function of the thermal stage of the
sensor as:

FT(s) = O(sj) (15 .27)

Equation (15 .26) becomes:

(15.28)
FT (s) 1+sNT

The sensor transfer function is:

GT(s)= (15 .29)

O(s)

or

GT(s) = KT FT (s) = KT 1 (15 .30)

+ sNT

Equation (15 .30) shows that the transfer function of an idealised temperature sensor is
that of a first order inertia.
The frequency response of the thermal stage of an idealised sensor may be written as :

OT(jCO) = 1
FT(j(o) = (15.31)
O(jco) 1 + j(oNT

and of the sensor as a whole :

GT(j(O) _ O(jco) - KT 1 + jcoNT (15.32)

IDEALISED SENSOR 287

15.2.2 Measurement of time varying temperature

If the changes of measured time varying temperature, 0(t), can be described by elementary
functions, the temperature indicated by the sensor 6r (t) , and the dynamic measurement
error can be determined in a simple way by using the Laplace transform. As an example,
consider a step temperature input from an initial temperature, 6b, to a final temperature
oe . In practice this case corresponds to the immersion of a temperature sensor with a
temperature, Ob, into a medium with a temperature, 6e , which is written mathematically
as:

fort Ob <_ 0
~9(t)
- {6 e ort>0

or

z9(t) = (Oe - Ob )1 (t) + Ob where 1(t) is a unit step at t = 0 (15 .33)

As it is necessary to obtain zero initial conditions for the Laplace transform, the excess
temperature O = Oe - 6b , will be used, to obtain:

O(t) = Oa 1(t) (15 .34)

for which the Laplace transform is:

O(s) = Oe (15 .35)

s

From (15.27) it follows that:

OT(s) = FT(s)O(s) (15 .36)

Inserting FT (s) from equation (15.28) and O(s) from equation (15 .35) into
equation (15 .36) the Laplace transform of the sensor temperature will be:

O
(15 .37)
OT(s) s(1_ NT)

After the inverse-transformation in accordance with Doetsch (1961) the temperature

time dependence will be:

OT (t) =-C-1 [aT(s)l (15 .38)

288 DYNAMIC TEMPERATURE MEASUREMENT

- e -t/N T) (15 .39)

OT(t) = OX

The final result is then:

_tIAI T) +
OT (t) = (Ve - Ob)(I - e Ob (15 .40)

From equations (15 .38) and (15.40) it follows that the step input response of an
idealised temperature sensor is an exponential curve, having the time constant NT as shown
in Figure 15 .3 . From this curve, the time constant NT can be found in a graphical way from
the tangent to the curve OT(t)=f(t) at any point, or as the time after which
O T (t = N T ) = 0.6320, .
Also the half-value time, or 50 % rise time to .5, which is the time when O T = 0.50, can
be used to determine the time constant. From equation (15 .39) at t = to.5, it is clear that:

0.50, = 0,(I - e to 5 I NT )

After some transformations the time constant is obtained as:

1 I
(15 .41)
2 tO.5 = 0 .693
NT = tO.5

The nine-tenth value time, t o .9 , is also a characteristic value of the temperature step
response. From the exponential function it can be shown that:

to-5
= 3.32 (15 .42)
to.9

la) (b
TIME NT CONSTANT

'e Ae
e4d,"' (t) *T = f (t)
m 0 .99,
ej U, T =f(t)

a 0,6328,
0,59,
W
n
E
w
H
W N
H
- - - -- - _ - - -- W
F x 0
t
4' to s TIME
NT
to.s
0 TIME t

Figure 15.3 Step response of an idealised temperature sensor

IDEALISED SENSOR 289

From equation (15.1) (15 .39) and (15 .40) the dynamic error will be:

AOdyn (t) =OT(t) - 0e =-(tie-f%)e-t/NT (15 .43)

or

AOdyn(t) = OT (t) - Oe = -pee-t / NT (15.43a)

Thus it is clear that:

AOdyn (t) = AOdy (t) (I5 .43b)

From equation (15 .2) (15.39) and (15.40) the relative dynamic error will be:

T(t)1e =-e-t/NT
S6dyn (t) _ (15 .44)
e
or

6Odyn (t) = OT (O - pe = -e-t l NT (I 5.44a)

e
As before :
t50dyn (t) = 60dyn (t) (I 5.44b)

The dependence of 829dyn upon the immersion time t is given in Table 15.1 . This
relative error as a function of time, tends to zero in an exponential manner.

Table 15.1 Dynamic error as a function of immersion time, t. (NT is the sensor time constant)

Time, t 0 .631VT NT 3 NT 3 .9 NT 4 .6 NT 6 .9 NT 9 .2 NT

S9dyn -0.5 -1/e -0 .05 -0 .02 -0 .01 -0 .001 -0 .0001

The true and measured temperatures O(t) and pT(t) as well as the dynamic errors,
AOdyn , and relative dynamic errors, 8t9dyn , respectively described by equations (15 .45) to
(15.62) are given in Table 15 .2 for the following four cases:

" Ramp input response,

" Exponential input response,
" Sinusoidal input response,
" Periodic non-sinusoidal input response.
290 DYNAMIC TEMPERATURE MEASUREMENT

Table 15 .2 Measurement of time-varying temperature by an idealised temperature sensor with the

transfer function given in equation (15 .30)

Input Mathematical expression s Eq'n no. Input and output comparison

a. 0(t) = kt (15 .45) g!
Aq,Y~ ct) NT

(x OT(t) =k[t-N T (1-e -tiNT)j (15 .46)

em flt)
O T (t) = k(t - N T ) for t >> N T (15 .47)
eT =f(t)
(15
AOdy (t)=-kNT fort»NT .48)

' t
0
NT

O(t) = 0,(I - e -t/'v ) (15 .49)

a
o
OT(t)=0e1- I (Ne-"N-NTe-tlNT)1 (15 .50) 6°
N - NT 8=f(t)
W l
eedY~(t)
OT(t)=0 1 - N e-"N~forN>>NT(15 .51)
e1 N-NT
eT =f(t)

AOdy. (t) = -Oe N

N
N e-uN
T (15 .52) t
-Oe ALL e-'IN for t » NT ;N >> NT
0
N
VT e-t/N for t >> NT ;N >> NT
S Odyn(t) =-- (15 .53)

0(t)=AOsin" (15 .54) 9

b
(15 .55) a e
. OT(t)=AO T ((o)sin[cot+tp((o)]with (p=2tc~t 6(t)
. To
CA t
AOT (0))=AO I (15 .56) 0
e(t)
1 + (coN T ~
AOT (c)) -
1 r
(15 .57)
AO I + (,NT )2

At
(15
cp =-2tc-=-arctanCONT .58)
TO

1 I . e
.C O(t )-
-AOL sinax--sin 3ux- . . . .--smncot- .. . . (15 .59)
n2
I HARMONIC 9(t)
o 9 6(t) "~ m
a; a a
8
AE), = AO (15 .60) _.
n 0 -. t
a
8 (15 .61) HARMONIC9TIt)
IHARMONIC At
AOT,1((o)= A0 1
7t ( .NT ) Z
L

t
(15
cp i =-27c -=-arctancoNT .62)
TO
REAL SENSORS 291

Numerical example
The medium temperature, 9e = 220 °C, is to be measured by a temperature sensor with a time
constant, NT = 30 s and an initial ambient sensor temperature of 9b = 20 °C . How long should the
immersion time be, to ensure that the indication error is less than 2 °C?
Solution:

A9 d, <_ 2 °C corresponds to < 2 = 0.01

Is1~dy°I 220 - 20

From Table 15 .1 for c59d, =- 0.01, the necessary immersion time is t= 4 .6NT , so that:

t > 4 . 6NT = 4 .6x30 = 138 s

Numerical example
A sinusoidally varying medium temperature was measured by a temperature sensor with NT = 30 s.
The period of temperature oscillations was TO =100 s, while the amplitude of the sensor
temperature, AO T = 5 °C . Determine the true amplitude of the temperature oscillations of the
medium .
Solution:
From equation (15 .56) in Table (15.2), the amplitude of the medium temperature may be
calculated to obtain :

AO =AOT 1+(coNT)2 =10.65°C

15.3 Real Sensors

The dynamic properties of real industrial sensors differ from those of the idealised sensors
described in Section 15 .2 . Materials and specific structures as well as working conditions,
which must be considered in detail for the description of real sensor behaviour will be dealt
with in Sections 15 .3 .1 and 15 .3 .2 .

15.3.1 Sensor design

The temperature indications of an idealised sensor depend upon the average temperature of
its whole mass . As the thermal conductivity, A, , of the material was assumed to be infinitely
high, the sensor temperature is the same all over its volume . In real industrial sensors,
which are mostly cylindrical, the sensitive part of the sensor is not always the whole of the
sensor. Figure 15 .4 gives a way of relating the dynamic step response behaviour of the
sensor to its sensitive physical structure.
292 DYNAMIC TEMPERATURE MEASUREMENT

1-SENSOR SENSITIVE
PA RT
2-SHEATH
EATH
3-INSULATING
3 MATERIAL

a
\ 2
E ~' 1 1
1
(a) (b) (c)

0 TIME t

Figure 15 .4 Step response of real temperature sensors

Each of the three general kinds of response is due to one of the following sensitive parts
of the sensor, relevant for the measurement :

" in volumetric response sensors the whole mass, as in mercury-in-glass thermometers

neglecting the extremely thin glass layer, as shown in response (a) of Figure 15 .4,
" in surfacial response sensors the surface of the sensor, as in bare resistance temperature
detectors with the resistance wire wound on the surface, as in (b) of Figure 15 .4,
" in central response sensors the centre of the cross-section, as in sheathed thermocouple
sensors represented in (c) of Figure 15 .4 .

A reasoned understanding of the step responses of all three structures, shown in

Figure 15 .4, is possible using a qualitative explanation of the heat transfer processes
involved.
Volumetric units exhibit a response which is closest to the response of the idealised
sensor . This arises from their whole volume acting as their sensitive part . A good example
of a sensor which exhibits this kind of behaviour is a mercury-in-glass thermometer.
In surfacial types, such as bare RTDs, the surface of the unit is the sensor sensitive part.
During the initial part of the response, it heats up quickly. Some time later, the central
regions of the assembly start to absorb more and more of the heat input. As it obviously
slows down the rate of temperature increase at the surface, the sensitive part of the sensor
"slows" to its final steady value.
In the central response type, it takes a finite time for heat to diffuse into the central
region of a sensor structure such as that of a sheathed thermocouple . Hence, with central
response sensors, there is a noticeable delay before the centrally located sensor starts to heat
up . After the main volume of the assembly has been heated the sensitive part starts to absorb
heat fairly quickly because of its relatively low thermal mass .
As proposed by Lieneweg (1975) all of the sensors can be classified by the ratio of their
response times, t O .9 / tO .5 , so that for :

1 . sensors with volumetric response, tO.9 / tO .5 = 3 .32 ,

2 . sensors with surfacial response, tO .9 / tO.5 > 3 .32 ,

3 . sensors with central response, tO.9 / tO.5 < 3 .32 .

REAL SENSORS 293

Defining a sensor to be within one of the above three groups, depends on its step
response . From this reasoning, it can be seen that the step response of any real sensor
depends upon its design, its working temperature and the heat transfer conditions at this
temperature .

15 .3 .2 Changing heat transfer coefficient

In the derivation of the differential equation (15 .25), which describes the dynamic
properties of an idealised sensor, it was assumed that both the sensor time constant and the
heat transfer coefficient between the sensor surface and the surrounding medium were
constant and temperature independent . In reality, the overall heat transfer coefficient, which
must include convection, conduction and radiation, is a function of the medium temperature
as well as of the instantaneous sensor temperature . Hence, time constant of the sensor will
vary with the varying heat transfer conditions .
These problems are especially apparent while measuring temperatures due to a
predominantly radiative heat exchange, as occurs inside chamber furnaces working above
600 °C (Hackforth, 1960; Michalski, 1966). The radiant heat flux between the chamber
walls at temperature, T2, and the temperature sensor at Ti , is given by equation (8.24a)) as:

T
T2
X21 Ale1Co 100)4-Co1\4
[( ]

where A 1 is the sensor heat exchange surface and E1 is the sensor emissivity.
This formula which is valid for walls with surface A2 > 3A 1, can be rewritten as :

(D21 - Ajar LT2 -T1~ (15 .63)

where ar is the radiant heat transfer coefficient given by:

)a )a
[( 00 - (100
a r -ciCo (15 .64)
(T2 -Ti)
)2
-z )3 T2 TI )2
e1 Co 10_
[(100 + 100 (100 + 100 (100 + (100 )3

If the sensor temperature, T1, does not differ from T2 by more than ±10 % of T2,
corresponding to :

T1
0.9 < < 1.1
T2
294 DYNAMIC TEMPERATURE MEASUREMENT

then equation (15 .64) can be replaced by the following approximate dependence of
accuracy to within 1% .

3
a r = e l Co x 0 .5 x 10-z (TI + T (15 .65)
100 100)

According to Eijkman (1955) and Lieneweg (1975) the coefficient, a r , increases as the
temperatures of the furnace walls and the sensor increase, as shown in Figure 15 .5,
achieving far higher values than the convective heat exchange coefficient, ak . The
temperature dependence of a r causes a considerable difference between the step responses
of bare thermocouples in an electric furnace and exponentially curved responses . On the
other hand high values of A means that they closely approach ideal sensors .

15.3.3 Equivalent transfer function

The application of the concept of the transfer function to real temperature sensors leads to
the idea of an equivalent transfer function, which is only valid under the following
necessary simplifying assumptions :

" The dynamic properties of the sensor are linearised within the given temperature range .
" A mean time constant is used as a value to describe the bi-directional heat flow between
the sensor and the surrounding medium as described by Skoczowski (1982) .
" The sensor is represented by a lumped parameter model .

The most commonly used sensor transfer functions are given in Table 15 .3 . These
models, which take the design of the sensor and the heat transfer conditions into account,
have been considered by Bliek and Fay (1979), Eckersdorf (1980), Hofmann (1965, 1966,
1967b), Rubin and Feldman (1968), Schwarze (1964) and Souksounov (1970) . The

600
T2 =1500K
z
W
vx Tj ~~T2
400
03 T2 =1200K

i 200
T2 = 900K

T2 =600K
0 -
300 600 900 1200 1500
SENSOR TEMPERATURE T, K

Figure 15 .5 Radiant heat transfer coefficient, %, versus sensor temperature, Ti , for a sensor of
a i = 0 .8, at given temperature, TZ , of the furnace walls
Table 15.3 Equivalent transfer function ofelectrical temperature sensors

Response Equivalent transfer function Equation Examples of sensor design

type number
v 1 THERMOCOUI
GT(S) = KT (15 .66)
1+SNT

0
SHEATH

1 THERMOCOUPLES
GT(s)=KT (15 .60
(1+SNTI)(1+SNT2)

SHEATHS
-SL
G T(S) - KT (15 .68)
1+sNT

e-SL Ps
GT(S)=KT (15 .69)
(1+SNTI)(1+SNT2) MI-THERMOCOUPLES

1 +SNT3 M1-THERMOCOUPLES
GT(s)=KT (1 (15 .70)
+SNTI)(1+SNT2)
w
d<4mm ~ d< 4mm
a-MEDIUM oC-MEDIUM
d' 4mm & BIG
d'-SMALL

d
296 DYNAMIC TEMPERATURE MEASUREMENT

majority of industrial thermocouples and resistance thermometer sensors may be simulated

by the second order system given by equation (15 .67) in Table 15 .3 . Low inertia sensors are
usually represented by a first order system model in accordance with equation (15 .66) in
Table 15 .3 .

15.3.4 Calculation of dynamic properties of sensors

It is assumed that the temperature sensor can be represented by a homogeneous cylinder of
thermal conductivity, A, specific density, p, and specific heat, c . The sensor is totally
immersed in the medium whose temperature is to be measured . Its thermal capacity is
infinitesimally small compared with that of the medium, while the heat transfer coefficient
between the sensor and the medium, as well as the sensor parameters, are constant. Internal
heat sources are also assumed to be non-existent in the sensor . Under the above
assumptions, Hofman (1976) and Jakob (1957, 1958) show that the Fourier differential
equation of an infinitesimally long cylinder is :

dOa(r,t) = [d 2 OYr,t) + I dO(r,t)

a (15 .71)
2

where a, which is the thermal diffusivity of the cylinder material, is equal to A/pc.
For real temperature sensors of radius R, the solution of this equation follows that for the
boundary condition of the third kind which have the general form :

d O(R,t) -- a
-[O(R,t)-OJ (15 .71 a)
ar

a0(O,t) - 0
(15 .71b)
ar

and also with zero initial conditions given by :

O(r,O) = 0 (15 .71 c)

This solution, considered by Hofmann (1976), gives the temperature any chosen point of
the cylinder at a distance, r, from the cylinder axis (0 < r < R) for a step change of the
ambient temperature from 0 to O e .
For some sensors it is also possible to use other simple models such as that of a sphere
or plate . The corresponding solutions of the Fourier equation for boundary conditions of the
third kind are given in Grober et al. (1963), Hofmann (1976) and Jakob (1957, 1958) . The
above theory, which is valid for a simple sensor, can also be extended to multi-layer sensors
as shown by Hofmann (1976), Lieneweg (1938a, 1938b, 1941, 1962) and Yarishev (1967) .
To determine the dynamic properties of real sensors, it is necessary to set up their
equivalent models first . As temperature sensors with tubular sheaths are widely used in
practice, most publications concern this type of sensors . The majority of authors, such as
REAL SENSORS 297

Caldwell et al. (1959), Eijkman (1955), Eijkman and Verhagen (1958), Meyer-Witting
(1959) and Yarishev (1967), use equivalent circuits based on a second order system, as
shown in Figure 15.6 . The thermal capacity, m1cl, of the sheath is represented by a
capacitor, C , and that of the sensitive part of the sensor, m2c2, by C . The thermal
l 2

resistances between the sheath and its environment, Ilk,, and between the sheath and the
temperature sensitive part, 1/k2, are similarly represented by the resistors R and R 2 1

respectively.
All these capacitances and resistances are given as relative values per unit of the sensor
length provided that the sensor is sufficiently long to neglect any heat exchange along its
length. The thermal resistance across the sheath wall has also been neglected . Element pairs
of the analogous quantities of the thermal and electric models are given in Table 15 .4
(Gr6ber et al., 1963). The input voltage, V, corresponds to the measured mean temperature,
O, the output voltage, VT, to the temperature, O , of the sensor's sensitive part, so that the
T

transfer function of the analogue circuit, given in Figure 15.6, is :

G(s) - VT(S) = 2 1 (15 .72)

V(S) S R C R2C2 +s(R C,+R2C +R
I I I 2 I C2)+1

1/kj] I R, R2

t Cz I VT

U
e T e
6 mi ci

m ~z

Figure 15.6 Real temperature sensor and its electric analogue circuit

Table 15 .4 The analogy between electric and thermal systems

Thermal system Electrical system Scale factor

Temperature above ambient O,(°C) Voltage V, (V) of voltage K,, = VIO (V/°C)
Thermal resistance 1/k, (°C/W) Resistance R, (S2) of resistance K = Rk, (f2W/°C)
R

Thermal capacitance mc, (J/°C) Capacitance C (F) of capacitance K = C/mc, (F°C/J)

C

Time t, (s) Model time tej , (s) of time K t = te 1lt, (s/s)

298 DYNAMIC TEMPERATURE MEASUREMENT

Denoting

RI C, = Nt (15 .72a)

R2 C2 = N2 (15 .72b)

NIN2 = N 2 and N= NIN2 (I 5 .72c)

and

Nt + N2 + RI C2 = 2~N

- 1 Nl + N2 + RIC2
(15 .72d)
2 NIN2

allows equation (15 .72) to be written in a standard second order form as follows :

G(s) = 1 (15 .73)

s 2 N2 + s2~N + 1

The step response of the circuit as given by Doetsch (1961) is then :

Z-t
h(t) = sI G(s)
_t

e- N sin 1- 2 . t -0
N
=1+ (15 .74)
1-5 2

where

(D = arctan 1- (15 .74a)

To obtain the transfer function of a sensor or its step response, the electric quantities in
equations (15 .73) and (15 .74) should be replaced by the analogous thermal quantities given
in Table 15 .4 .
As this rather complicated solution is of only limited practical use, a temperature sensor
can be modelled for simulation purposes by two, non-interacting RC-elements . Such a
model is normally used in the experimental determination of the step responses of a
temperature sensor. Since the analytical method is not very precise it is rarely used .
EXPERIMENTAL DETERMINATION OF THE DYNAMIC PROPERTIES OF SENSORS 299

15.4 Experimental Determination of the Dynamic

Properties of Sensors
15.4.1 Classification and application of the methods
Dynamic properties of real sensors are usually determined experimentally by adapting step
response and frequency response methods (Rake, 1980) . When testing for operational
integrity and performance checking, the dynamic parameters of temperature sensors should
be found in the same environment and at the same temperature where the sensor will be
used . Nevertheless, it may be just as important to measure the parameters for design or
model validation purposes . Before the development of digital techniques it was essential to
ensure that analogue recorders used for recording the measurements, had a response time of
at least five times shorter than the sensor's response time . This condition is easily met using
any of the various digital data acquisition systems described in Chapter 13 .
McGhee et al. (1992a, 1992b) provide a review of the dynamic properties and testing of
contact temperature sensors . A classification of experimental identification methods, which
is given in Figure 15 .7, shows that the two main methods rely upon external or internal

electnc two-section low inertia

furnace furnace furnace
EXTERNAL or
two furnaces
INPUT W
METHODS a N
W
F-. +
AT HIGHER
TEMPERATURES ~
-a
a)--
(radiant heat a$
exchange)

INTERNAL "----- ~~ STEP TEMPERATURE

INPUT heating \ ~~, SINUSOIDAL,
METHODS `, current ;SQUARE WAVE,
MSS
------" SENSOR

EXTERNAL O
a m
INPUT N
METHODS air channel air channel air channel Glower
or or
AT LOWER water pipe-line water pipe-line
TEMPERATURES or or
(convective heat two pipe-lines double pipe-line
exchange) or
liquid bath

Figure 15 .7 Classification of methods for the experimental determination of sensor dynamics

300 DYNAMIC TEMPERATURE MEASUREMENT

stimulation by an interrogating signal . As shown in Section 15.4.4, the results obtained by

both methods are not identical . Moreover, the aperiodic or periodic nature of the signal
respectively determines whether the method used belongs to the time domain or frequency
domain group of testing procedures .
External temperature input methods are widely used to measure the dynamic properties
of temperature sensors, which are subjected to known changes of the temperature of the
surrounding medium . For example, for step input response the sensor should be immersed
in a medium at a different temperature from its initial temperature . As this means that the
sensor is not in its real operating conditions, this method is used for the purposes of design
or model validation. At low temperatures, below about 100 °C, where heat exchange by
radiation is negligibly small, a convective heat exchange coefficient , a k , can explicitly
describe the working conditions of the sensor. Most methods are based on measurements in
either flowing water or air . In practice, at temperatures over 600 °C, with free convection
conditions, there is only radiant heat exchange characterised by the source temperature as a
final value. In all other cases it is necessary to specify the heat exchange conditions
precisely, by giving the values of the heat exchange coefficient a and of the final
temperature 9,.
In most of the older publications such as those of Higgins and Keim (1954), Hofmann
(1976), Huhnke (1973), Jakob (1958) and Kondratiev (1947), the experimental determin-
ation of the sensor dynamic properties is performed at low temperatures using well known
methods, although the associated instrumentation is now largely obsolete . More
contemporary, microprocessor based instrumentation was used by McGhee et al. (1989) in
the first application of MBS signals to temperature sensor testing. Further development of
this method appear in McGhee et al. (1993) and Jackowska-Strumillo et al. (1992, 1996,
1997) . These low temperature applications for external input testing are described in
Section 15.4.2 for the time domain approach, while external low temperature frequency
domain testing is presented in Section 15.4 .3. Only a few publications deal with external
input in the temperature range above 600 °C (Rubin and Feldman, 1968; Bernhard, 1979 ;
Eckersdorf, 1980; Eckersdorf and Michalski, 1984). These methods are presented in
Section 15 .4.4.
Dynamic testing of electric sensors in real operating conditions uses in situ testing by the
internal input method . An electric current, flowing in the electrical part of the sensor,
causes direct ?R self-heating in the resistance winding of an RTD or in the lead wires and
measuring junction of a thermocouple . These methods are described in Section 15.6.3 .

15.4.2 External input, time domain testing with convective heat

transfer
Step input method . The simplest and most popular method of measuring the step
response of a temperature sensor is to immerse the sensor by plunging it into a well-stirred
water bath at a temperature in the range of 20 to 60 °C above the ambient temperature . This
method belongs to the time domain group of testing methods . Berger and Balko (1972)
advise the mechanisation of the immersion procedure to ensure that the transition in the
external temperature of the sensor is as fast as possible. The heat transfer coefficient, a,
between the sensor and the medium is some thousands of W/m °C. This method gives
explicit and reproducible measuring conditions, because any occurring variations of the
EXPERIMENTAL DETERMINATION OF THE DYNAMIC PROPERTIES OF SENSORS 301

high values of a are usually so small that they do not exert a significant influence on the
sensor dynamics (Kondratiev, 1947 ; Yarishev, 1967).
To measure sensor dynamics with different values of a, some special channels and
pipe-lines are constructed, in which a gas, mostly air, or liquid, mostly water, flows at an
adjustable velocity. This velocity is measured in order to enable a comparison of the results

2 2
obtained. Hofmann (1976) describes an air channel in which the flow velocity may be
adjusted between 0 and 20 m/s. It is intended for sensors, having sheath diameter, D, up to
10 mm and an immersion length up to 100 mm . The highest achievable values of a were
150 W/m °C, for D = 10 mm, and 400 W/m °C, for D = 1 nun . Chohan and Natour
(1988) describe an air channel, which was designed for sensor testing at air velocities up to
2000 m/s .
In the water pipe-lines described by Hofmann (1976) and Huhnke (1973), the water
velocity is adjustable between 0 .1 and 2 m/s . The values of a at the lowest water flow
velocity are comparable with those at the highest air flow velocity, while the values of a for
wet air approach the values for liquid (Chohan and Natour, 1988) .
Measurements in air and water proceed in a similar way. The tested sensor is washed by
hot water coming from a pipe inside a shield, as shown in Figure 15 .8 . Protection of the
sensor from contact with the flowing medium is ensured by the shield . The hot-water
temperature is 10 °C above the medium temperature . At a given moment, the springy shield
is pushed in the direction of the flowing medium and simultaneously the flow of hot-water
stops . A deciding factor is the step input duration, which has to be shorter than 0 .1 of the
half-value time, to .5 , of the tested sensor. In this set-up a mechanical device ensures that the
95 % step response time, tO .95, is shorter than 20 to 100 ms .
A step change in temperature can also be realised using a low-inertia mesh heating
element . Such a mesh is mostly used for gas heating, because in liquid heating far greater
heating power would be needed . The air-channel shown in Figure 15 .9 (Huhnke, 1973) has
a main construction consisting of a thin wire mesh heating element placed parallel to the
tested temperature sensor and perpendicular to the direction of the air stream . A step change
in the temperature of the flowing medium is achieved by switching the heating power on
and off. In most cases temperature changes of about 10 °C, which are applied, should be
regarded as exponential approximations to step temperature changes characterised by a
time-constant, N (Huhnke, 1973) . For example, at an air velocity of v = 2 to 15 m/s, the

TESTED
SENSOR

FLOWING
300 TESTED 160
MEDIUM SHIELD
SENSOR
-'~ f I I
I C===- O
I I SHIELD ~------- ----
MESH
I
MOVEMENT f 1 1
HEATING
` ELEMENT t f } ,°oo

-D -
HOT WATER
SUPPLY

BLOWER

Figure 15.8 Pipe line with a movable shield Figure 15.9 Air channel with mesh heating element
302 DYNAMIC TEMPERATURE MEASUREMENT

time constant of the heating wires of d = 0 .02 mm, is N = 16 to 7 ms, corresponding to a

response time, t o .95 = 50 to 20 ms . Application of a low inertia heating element, also makes
it possible to generate other temperature test signals, especially sinusoids . The method of
interpreting the recorded step-input response of a temperature sensor depends upon its
character and upon the necessary precision and may be as follows :

" determination of t r , considered in Section 15 .1 .1, or to .5 and to .9 , from Section 15 .2 .2,

directly from the step input response,
" determination of the equivalent transfer function choosing the right sensor model from
Table 15 .3 based on the to.9/t0 .5 ratio .

All of the following variations of sensor response are summarised in Table 15 .5 .

Volumetric response sensors, with to .9 1to.5 z 3 .32, which also have dynamic properties
similar to those of the ideal sensor of Section 15 .2 .2, are regarded as first order inertia
elements . Their equivalent transfer function is given by equation (15 .66) of Table 15 .3 .
Central response sensors, with to .9 lt o .5 < 3,32, which are characterised by s-shaped step
response curves, are regarded as a series connection of a first order inertia element and a
pure lag . Their equivalent transfer function is given by equation (15 .68) in Table 15 .3 . The
values of the pure lag parameter, L T, the time constant, NT, and thus LTINT, can be found in
the way shown in Figure 15 .10 . In some cases, an appropriate alternative is to use a model
with a second order inertia element with an equivalent transfer function given by
equation (15 .67) of Table 15 .3 . In this case, the values of NT i and NT2 can be found from
Figure 15 .11 .
A more precise method of determining N TI and NT2 , when NTI >> NT2, is the
Yarishev (1967) logarithmic technique shown in Figure 15 .12 . When the step-input
response of equation (15 .77) of Table 15 .5, is given as the relative unit-step response :

O (t)
h(t) = (15 .79a)
Te

it allows a definition of the function, h * (t) , given by :

h * (t) =1- h(t) (15 .79b)

The relations in equations (I 5 .79a) and (15 .79b) are shown in Figure 15 .12(a) .
When t >> NT2 , h * (t) will be transformed into hi (t)

(t) = NTI e-ONTI

h1 (15 .80)
NT 1 - NT2
Taking logarithms on each side of equation (15 .80) gives the straight line :

NTI _ t
lnhj (t) =1n (15 .81)
NT I -NT2 NT I
Table 15.5 Dynamic models oftemperature sensors
Equivalent block Equivalent transfer function, GT(s) Step response &f Q)
diagram of a sensor (Equation numbers from Table 15 .3) Graphical Mat

Volumetric response
(Figure 15 .4(a))
1-st order inertia element 1
G T(s) -KT (15 .66) Or(t)- O(I _ e-IINT
I+sNT

t 0 NT
Central response NT
(Figure 15 .4(c))
1-st order inertia element
+ pure lag _ e-%/.T Or(t) = fl in t LT
GT(s)-KT (15 .68)
1+sNT
Or(t) - O (1 _ e-(r-t
-z~ 4-r
t 0 LT
in
Central response
(Figure 15 .4(c)) Oe _ N
2-nd order inertia element Or(t) - ~[ 1 -
- 1 NTI _
GT(s) (15 .67)
KT(,+sNTt)(1 +sNTZ)

0 t
Surfacial response O
(Figure 15 .4(b)) Ob NT, -
2-nd order inertia element Or(t) ° Oe 1- -
+ derivative element _ I +SNT3 NTI -
GT(s)-KT (1+SNTt)(1+sNTZ) (15 .70)

0 t
304 DYNAMIC TEMPERATURE MEASUREMENT

Plots . In h*(t) are shown in Figure 15 .12(b) . Solving equation (15.80) for
the instants tj and t2, shown in Figure 15 .12(b) gives the value of the time constant, NTI, as :

t2 -4 (15 .82)
NTI -

O_rL_~_=

um-nnN
0,06

niri~~


0,008
11N11_III_111
, ..
__ ,
, ...
. . ., ......
, . . . .8 ., ..

i - . . response . D. . .--
temperature sensor with central response constants NTI and NT2 when LTINT is known and
has .--n found using the method of Figure 15 . 10

Kh N2
h(t)'' EXPERIMENTAL STEP-INPUT tr,(t) LOGARITHMIC DISPLAY

~(t)- - - ~
~(t , RESPONSE OF STEP-INPUT RESPONSE
e'*"'

--- - - - - - - - - - - -- - - -
1,0
.:
..

0,2 I
.

(t)
..

TIME t TIME t

Figure 15 .12 Logarithmic method of determining the time constants, NT t and NT2
EXPERIMENTAL DETERMINATION OF THE DYNAMIC PROPERTIES OF SENSORS 305

The time constant, NT2 , can be determined by reading the value of h*(O) from
Figure 15 .12(b) and inserting it into equation (15 .80) at t = 0, to obtain :

h1(0 ) -- NTI (15 .83)

NTI - NT2

Since N TI has already been found it can be used in equation (15 .83) to obtain N T2 from :

hI (0) -1
NT2 = (0) NTI (15 .84)
hl

The time constant NT2 can also be found in a similar way as NTI by drawing the function
h2* (t) = hl (t) - h * (t) with a semi-logarithmic scale . This method is then specially useful

for checking if the second-order inertia system gives a sufficiently precise model of the
sensor. If the function h2 (t) is not a straight line, the sensor should be approximated by a
higher order system .
Sensors with surfacial response, where t0.9/t0 .5 > 3,32, are mostly approximated by first-
order inertia systems, as described by equation (15 .66) in Table 15 .3 . If a more precise
approximation is necessary, a transfer function as given in equation (15 .70) from Table 15 .5
is used :

(I+sNT3)
GT(s)=KT
(1 + SNT I )(1 + sNT2 )

The transfer function zero, at s = - 1/NT3, takes account of the thermal layer from the inside
of the sensitive sensor surface to the core of the sensor .
A method to find the time constants, NTI, NT2 and NT3, which is based on the recorded
step input response, is described in Hofmann (1976) and Yarishev (196/) . However, for
sensors with surfacial response it is usually precise enough to represent their dynamic
properties by an equivalent transfer function with a single inertia :

GT( s) =K
- T I
I+sNTI

Subsequently, the value of NTI is determined in the same way as for sensors with volumetric
response.
To determine the parameters from a step input response of sensors with volumetric
response under changing heat-transfer coefficient the Lieneweg method is used to find a
characteristic parameter, Y, from Figure 15 .13 . This method is based on experimentally
determined step input responses of the tested sensor in two different media . The values of
the surfacial heat transfer coefficients, a, must be known in each case . In most cases the
measurements are conducted in water and air to find either the half-value times, tO .5,w, in
 "" .won . mom
.""
1 on

awO rm
ONE
MEN
Wasamm

aoffixo

level

w - i . 11

1 : 11
EXPERIMENTAL DETERMINATION OF THE DYNAMIC PROPERTIES OF SENSORS 307

step input response of the sensor and that of an infinite solid-cylinder. This method includes
an algorithm, which enables calculation of the transfer function of the sensor for any other
value of ax based on measured values of to .5 and to.9 and a known value of the coefficient,
a.

15.4 .3 External input, frequency domain testing with convective

heat transfer
Although the time domain step input response to determine temperature sensor dynamics is
easy to implement, it does not give sufficiently high precision for low-inertia sensors with
high heat-transfer coefficients . For this type of sensor, it is preferable to use periodic input
signals to obtain its frequency response. An experimentally determined frequency response
of a temperature sensor, which is a type of non-parameterc model, is the most precise way of
presenting its dynamics, because no simplifying assumptions of any kind are needed .
Moreover, the use of contemporary computer based data acquisition systems and digital
signal processing techniques remove all of the laborious hand calculations which
characterised the classical application of the techniques .

Rectangular, periodic input signals. Rectangular periodic input signals are easy to
generate in water pipe lines or air channels . More details of this method can be found in
Bliek and Fay (1979), Huhnke (1973) and Woschni (1980) . In the early use of this method
interpretation of the results was difficult . Nowadays, as described in Chapter 13, the power
of digital signal processing methods greatly simplifies this task. Square-waves are
practically very useful because they are easily generated and applied in industrial systems.

Sinusoidal input signals. Application of sinusoidal input signals for determining sensor
dynamics requires rather complicated instrumentation . Hence it is not very popular . Because
each experiment at each different frequency is conducted separately, taking the
measurements is extremely time consuming, even though the latter may be automated using
the methods described in Chapter 13 . The design of a special air channel with low-inertia
heating elements, described by Huhnke (1973) decisively simplifies the application of this
method . Recording the temperatures, 9, of the air and, 9r, of the tested sensor at different
frequencies of the sinusoidal temperature oscillations allows the determination of a non-
parametric frequency response of the sensor, which is usually drawn in the form of a Bode
diagram like that shown in Figure 15 .14 (Michalski et al., 1981) . If the sensor is
approximated by a first order inertia element, Figure 15 .14 shows an easy graphical method
of finding its time constant, NT , using the asymptotes of the frequency response to
determine the corner angular frequency co, as N T =1 / coy . Presentation of the frequency
response of a sensor by a Bode diagram is especially convenient in the synthesis of an
automatic temperature control system . (Michalski and Eckersdorf, 1987) . The frequency
response is very useful in selecting the best sensor model and its characteristic parameters
(Michalski et al., 1981) . The experimentally determined sensor frequency response under
known heat transfer conditions between the sensor and surrounding medium, can form the
basis for its conversion to a model for other operating conditions (Kocurov, 1963).
308 DYNAMIC TEMPERATURE MEASUREMENT

(a)

1,0 ---~ -------

I \
o ~_IF, (jw)I
a
I
w 0,1
0

CORNER
=1
I FREQUENCY
a ~~=11NT
0,01
0,001 0,01 0,1 1,0
(b) ANGULAR FREQUENCY
0,00 0,01 1 0,1 1,0
0
o I

9- -1/2 __________
-?r 314
n
a
x
0-

Figure 15 .14 Bode diagram of frequency response of a temperature sensor

Multifrequency binary signals NIBS) . Multifrequency testing is a well known

identification method using multifrequency perturbation signals (Godfrey, 1993) . The
precision of this method is comparable with the application of single sinusoidal signals,
while the total experiment time is much shorter . It is correct to regard all frequency domain
testing methods as applications of frequency analysis by the correlation method .
Multifrequency Binary Signals, sometimes called MBS or MFBS signals, which are the
most easily generated multifrequency signals because they are based upon step signals
(Henderson at al., 1987), are deterministic, discrete interval, periodic signals whose
predominant power is concentrated in a few dominant harmonics . Generating MBS signals
for testing RTD sensors using a two-channel blower arrangement was originally described
by McGhee et al. (1989) . The MBS method is also used in the internal input method, which
is described in Section 15 .4 .5 .

15.4.4 External input with radiative heat transfer

Heating-up curve method The heating-up curve method of determining the dynamics of
a temperature sensor can be obtained by placing the sensor in a furnace chamber at the
temperature 9e . To give explicit and reproducible measuring conditions, the furnace should
be constructed so that the emissivity of the furnace chamber is as near to unity as possible,
its thermal capacity is far greater than that of the sensor with the furnace heated to a uniform
temperature over a zone much longer than the immersion depth of the sensor . In addition
the furnace temperature has to be controlled precisely in the range from 500 °C up to the
maximum temperature of application of the sensor. In that temperature range, the sensor
EXPERIMENTAL DETERMINATION OF THE DYNAMIC PROPERTIES OF SENSORS 309

cannot be approached as a linear element and therefore it should not be described by a

transfer function . The experimental heating-up curve of a sensor, can only be used for really
rough, approximate estimation ofthe sensor dynamics . Commonly, the tO.5 and t0 .9 times are
used with their values referred to the final value of the temperature 9e of the heating-up
curve .
Heating-up and cooling-down curves method An easy way to determine the equivalent
transfer function, given by equation 15.67 form Table 15.5, for temperature sensors
exhibiting volumetric response (t0,9/t0 .5 < 3 .32), which is based on the heating-up curve and
the first part of the cooling-down curve, is known as the heating-up and cooling-down
curves method (Eckersdorf and Michalski, 1984). Afurnace, similar to that for the heating-
up curve method described above, should be used for the method which is based on the step
response equation (15 .77) from Table 15.5 .
Assuming that NT2 < 0.5NTI it follows that the latter part of the heating-up curve is only
determined by the bigger time constant, NTI. In the temperature interval, A6, below
100 °C, the sensor can be regarded as a linear element so that the graphically found time
constant, NTI, in Figure 15.15(a) has a correct value, relevant to the final temperature value
9e. From the cooling-down curve of Figure 15.15(b), the equivalent lag, LT, of the sensor
may be found . This is then used to find its second time constant NT2 as shown in

la) (b)
N T, LT

W DETERMINATION OF LT

a
Q
w
. DETERMINATION OF Nr
W
1
E
W W
F-
H

~0 TIME t 0 TIME t

(c) 1
_NT, NT2 FOUND FROM NTZINT,RATIO
NT1

0,1

.
001
0 0,1 0,2 0,3 L T INT1

Figure 15 .15 Determination of the time constants, NTI and NT2, of a sensor by (a) the heating-up
curve method and (b) the cooling-down curve method
310 DYNAMIC TEMPERATURE MEASUREMENT

Figure 15 .15(c) . The diagram (NT2/NTI) =f(L T1NTd is based on the data from Figure 15 .11 .

The determined values of NTI and NT2 are referred to the temperature, 9e . It has been
proved by Eckersdorf and Michalski (1984) that the proposed method is sufficiently precise
for those industrial sensors typically applied in electric furnaces in the temperature range
from 600 °C to 1200 °C .

Step input response method To realise a step input precisely at a given temperature
level for use in the step input response method, the sensor should be transferred from
temperature (9e - 0 .5A9) to temperature (9e + 0 .50,9) as quickly as possible . After each
transition, the sensor should be immersed up to its normal immersion depth . To increase the
measurement precision the step increase and the step decrease of the temperature should be
repeated twice, taking the average values as the final result . The value of the temperature
step, A9, is of the utmost importance. To ensure that the sensor response remains mainly
linear, the step should not be too big, while a step which is too small may result in large
random errors. In practice, the value should be about 60 to 100 °C . Either two electric
furnaces standing side-by-side or one two-section furnace can be used . In the first case, in
transferring the sensor from one furnace to another, the temperature versus time is far from
a pure step-input, thus resulting in additional errors . In the second case, although generation
of a step input gives a much higher precision, its application range is restricted, depending
upon the complexity of furnace construction . In particular, the furnace, which should have
sufficiently long non-interacting heating zones, will be limited only to testing very long
sensors . The application range and precision of the step input response method of
(9e ± 0 .5A9) are similar to the heating-up and cooling-down curves method . Interpretation
of measured responses is performed in the same ways as given in Table 15 .5 .

Sinusoidal input response method The methods of determining the dynamics of sensors
given before do not assure sufficient precision in the case of low-inertia sensors, such as MI
thermocouples, in the temperature range above 500 °C . It is then advisable to measure their
frequency responses at given temperature levels using the sinusoidal input response method .
To achieve this aim Eckersdorf (1986) has constructed a special low-inertia furnace, shown
in Figure 15 .16, which enables the generation of sinusoidal temperature variations from
0 .001 to 0 .5 Hz in the range from 600 to 1000 °C .
The furnace chamber, 1, made of a temperature resistant steel tube with closed ends is
heated by halogen radiators, 2, whose radiation is concentrated by parabolic mirrors, 3 .

3 2 2 1 3

1
1-FURNACE CHAMBER
2-INFRARED HALOGEN RADIATORS
3-PARABOLIC MIRRORS

Figure 15.16 Special electric furnace for measuring the frequency response of low-inertia sensors in
conditions of radiant heat transfer
EXPERIMENTAL DETERMINATION OF THE DYNAMIC PROPERTIES OF SENSORS 311

Sinusoidal temperature variations are generated in a closed-loop system applying a

generator, a temperature controller and a thyristor final control element, with thin
thermocouples for temperature measurement and control welded to the chamber tube, 1 .
The temperature amplitudes for the measurements are 10 to 40 °C . Normal presentation of
the results is similar to that described in Section 15 .4 .3 .

15 .4 .5 Internal input method

Warshawski (1974) first described the internal input method of identifying the dynamic
properties of electric temperature sensors . The method is especially useful for in-situ
identification . This idea was further developed in USA (Kerlin et al., 1978, 1981, 1982 ;
Hashemian et al., 1990) . The method, which was initially applied to RTDs and later also to
thermocouple sensors, satisfied the need for continuous monitoring of changes in the
dynamics of temperature sensors operating in extremely difficult and dangerous conditions
such as occur in nuclear power stations . The internal input is realised by direct electric
current flow in the RTD winding or in the electrodes and measuring junction of a
thermocouple sensor . By measuring the sensor response to this input, it is possible to
determine the operational transfer function for the internal input GT (s) as :

(s)
GT (s) (15 .85)
P,, (S)

where, OT (s) , is the Laplace transform of the sensor temperature and Ph (s) is the Laplace
transform of the applied internal heating power both at the zero initial conditions .
Comparing the equations (15 .85) and (15 .11), it can be seen that the transfer function,

GT (s) , does not correspond to the transfer function GT(s) . Assuming a unidirectional heat
flow in the sensor and very small heat capacity of the sensitive part of the sensor, the
operational transfer function of the sensor is given by ;

GT (s)=KT n 1 (15 .86)

rl(1+sNTi )
i=1

and by internal input, by :

m
rl(1+sN Tj )
j
GT (s) = KT n 1 (15 .87)
~(1+sNTi )
i=1

Comparing equations (15 .86) and (15 .87) it is clear that the inertia time constants, NTi, are
identical in both cases . Thus it is possible to determine NTi by the internal input method and
312 DYNAMIC TEMPERATURE MEASUREMENT

eventually also the sensors transfer function GT(s) (Bernhard and Noack 1981, Kerlin et al .,
1978, 1981, 1982) .
In the measuring circuit for RTDs, which is shown in Figure 15 .17, the bridge supply
voltage is modulated by an additional signal, generating the internal temperature input
(Taylor and Navarro, 1987) . The resulting changes in sensor temperature are compared with
the internal input signal and enable the sensor dynamic properties to be determined . A step-
input signal as an additional signal is usually applied so that the time constants, NTt, and the
transfer function GT(s), in equation (15 .86), are determined from its output . Using MBS to
determine the frequency response of RTDs by the internal input method is now well
established approach (Jackowska-Strumillo et al., 1992, 1996, 1997 ; McGhee et al ., 1992,
1993) . After some signal processing and other calculations, the time constants, NTi, and
transfer function GT (s) can be obtained .
Application of internal input method for testing of thermocouple sensors, or simply
thermocouples, is more complicated . The heating current, circulating in the thermocouple,
has to be an alternating current to avoid any additional changes in thermocouple
temperature, caused by Peltier and Thomson effects, described in Section 3 .1 .1 . The heating
current is the source of a far greater voltage drop across the thermocouple than the
generated thermoelectric force . Thus the measurement of this emf during the internal input
test is impossible . The first attempts of applying this method were based on the cooling
down-curve, immediately after switching off the heating current (Kerlin et al . (1981) .
Michalski and Eckersdorf (1990) introduced the circuit, shown in Figure 15 .18 to measure
the sensor output signal, nearly continuously, using a switching device . The emf was
measured during the very short intervals when the heating current is interrupted . These
interruptions had no marked influence on the total heating power . The sinusoidal heating
power input, applied to the sensor, was then replaced by a computerised application of MBS
signals (Eckersdorf et al ., 1995 ; Kucharski and Sankowski, 1997 ; Kucharski, et al., 1997) .
The internal input method described above is especially useful for determining the
dynamics of those sensors, which can be regarded as first order inertia systems . It gives

THERMOCOUPLE COMPENSATION REFERENCE

R, Rs CABLE TEMPERATURE
R i .R2 .R3 - RATIO
RESISTORS i
RT R T -RTD -
R3

4
O .C. VOLTAGE
SOURCE A . C . POWER
SOURCE
V=coast

TEST SIGNAL TEMPERATURE TEST SIGNAL TEMPERATURE

GENERATION MEASUREMENT CONTROL
GENERATION MEASUREMENT

SENSOR DYNAMI CS EVALUATION SENSOR DYNAMICS EVALUATION

Figure 15 .17 Block diagram of the system for Figure 15 .18 Block diagram of the system for
determining RTD dynamics by the internal determining thermocouple dynamics by the
input method internal input method
 MUM

I X 11' I 1 1

MEN A . :
BOOM
v

_ KOAA= 1

Q~-
v v v v v ~ v v v v v
Table 15 .6 Characteristic dynamic parameters of temperature sensors determined from step respot
(VDE/VDI-Richtlinien 5522, 1979)

Temperature sensors
Air at v =1 mh
to.5 (s) to .9 (s) to.91

I Liquid-in-glass thermometers
Mercury, bulb d = 6 mm, 1= 12 mm 48 150 3
Mercury, bulb d = 6 mm, 1= 38 mm; steel sheath d = 10 mm 121 350 2
Alcohol, bulb d = 6 mm, 1= 38 mm; steel sheath d = 10 mm 200 500 2
II Manometric liquid-filled thermometers
Bulb d = 17 mm 180 600 3
Bulb d = 17 mm; steel sheath d = 22 mm 340 1090 3

III Thermocouple sensors used at atmospheric pressure

Type K, chromium iron alloy sheath d = 22 mm 390 1030 2.
Type K, chromium iron alloy sheath d= 15 mm 190 490 2.
Type K, ceramic sheath d= 15 mm 220 580 2
Type S, ceramic sheath d= 15 mm 180 455 2
Type S, two ceramic sheaths d = 15 mm, d = 26 mm 235 570 2

IV Thermocouple sensors used at medium and high pressure

Cylindrical thermowell d = 9 mm, thermocouple insert d = 6 mm 92 300 3
(max 60 MPa, max 400 °C)
Cylindrical thermowell d = 14 mm, thermocouple insert d = 8 mm 148 460 3.
(max 10 MPa, max 500 °C)
Conical thermowell d = 24 mm, thermocouple insert d = 6 mm 225 800 3.
(max 60 MPa, max 540 °C)
Conical thermowell d = 18 mm, MI thermocouple d = 3 mm 190 750 3.
(max 16 MPa, max 540 °C)
Table 15 .6 (continued)

Temperature sensors
Air at v = 1 m/s
to .5 (s) to .9 (s) NY

V Resistance thermometer detectors (RTD)

Glass platinum RTD, d = 4 mm 11 .5 47 4.
Glass platinum RTD, d = 3 mm 8 28 3
Glass platinum RTD, d = 1 mm 1 .3 4 .1 3.
Ceramic platinum RTD, d= 2 .8 mm 10 39 3
Ceramic platinum RTD, d= 1 .5 mm 6 20 3
Ceramic nickel RTD, d = 5 .2 mm 21 67 3.
Resistance thermometer insert d= 6 mm 72 220 3
Resistance thermometer insert d = 8 mm 115 335 2
VI Resistance thermometer sensors used in medium and highpressure environments
Cylindrical thermowell d = 9 mm, RT insert d = 6 mm 140 390 2
(max 4 MPa, max 400 °C)
Cylindrical thermowell d = 14 mm, RT insert d = 8 mm 240 720 3
(max 10 MPa, max 500 °C)
Conical thermowell d = 24 mm, RT insert d = 6 mm 315 1070 3
(max 60 MPa, max 540 °C)
VII Low inertia resistance thermometers
Metal sheath d= 6 mm (max 500 °C) 32-40 118-130
Metal sheath d = 3 .5 mm, thinned down to d = 2 .3 mm (max 600 °C) 4 .5 14 3.
High pressure cylindrical thermowell d = 12 mm, thinned down to d = 4 .5 mm 80 280 3.
(max 10 MPa, max 400 °C)
High pressure conical thermowell d = 24 mm, thinned down to d = 4 .5 mm 80 250 3.
(max 60 MPa, max 400 °C)
316 DYNAMIC TEMPERATURE MEASUREMENT

Table 15 .7 Half value time, tp . s, of MI thermocouples determined from the step input response
from 20 to 40 °C

Sheath diameter, d (mm)

Medium 0.5 1 1.5 3 4.5 6
Water at v = 0.2 mls 0.06 0.15 0.21 1 .2 (0 .6)* 2.5 4 (4)*
Air at v = 0.2 m/s 1 .8 5 8 23 (26)* 37 60 (55)*
* Value for MI resistance sensors

Table 15 .8 Heat Transfer coefficient, a, (WIm2 K) for cylindrical temperature sensors

of diameter, d, (mm)

Medium Temp . (°C) Flow velocity (m/s)

0 2
d= d=
1 4 10 20 1 4 10 20
Water 30 840 700 580
50 2400 1390 1130 950
70 1900 1530 1390

Air 50 37 16 .0 11 .3 10 .2
100 20 14 .7 12 .7 140 49 42 38
200 25 .3 18 .8 16.2
Medium Temp. (°C) Flow velocity (m/s)
5 10 20 50
d= d= d= d=
4 10 20 1 4 10 20 4 10 20 1 4 10 20
Water 30
50 1100-11000
70

Air 50
100 80 70 64 300120 105 93 168 147 133 520 285 248 212
200

Overheated 300 (0 .5
104 156 261 162 226 360 226 320 500 360 505 800
Mpa) 290 383 630 383 545 860 560 780 1200 870 1230 1950
water 400 (2 Mpa) 755 1100 1800 1100 1570 2550 1570 2260 3600 2500 3550 4200
vapour 300 (5 Mpa) 930 1330 2030 1330 1850 2900 1850 2610 4070 2960 4100 6600
500 (10 Mpa)
Medium Temp . (°C) Flow velocity (m/s)
0 0 .01 0.1
d= d= d=
4 10 20 4 10 20 4 10 20
Oil 70 I 52 36 29 174 116 87 520 337 232
Condensed, saturated watervapour:- a= 5800 - 10 500; condensing overheated water vapour :- a= 400;
condensing liquids:- a= 1750 - 35 000
DYNAMIC PROPERTIES OF CHOSEN SENSORS 317

The dynamic properties of thermocouple sensors, intended for application over 600 °C
at atmospheric pressure, are displayed against temperature, in Figure 15 .21 . In this figure
two time constants, NTI and NT2 , or one equivalent time constant, NTe , versus temperature
are shown for two different industrial sensors . Determination of the equivalent time
constant, NT .' is based on the measured half-value time, tO .5, in accordance with Eckersdorf
and Michalski (1984) . Experimental frequency responses for the same sensors obtained by
the same authors are shown in Figure 15 .22 . Figures 15 .23 and 15 .24 show the frequency
responses of MI thermocouples and of bare, unprotected thermocouples, determined
experimentally as described in Section 15 .4 .

(al

THERMOCOUPLE TYPE K, WIRE DIAMETER d,=3mm

300
HEAT-RESISTANT STEEL SHEATH, d,=20mm

250 -

200
z

150 FT Is)
z
r
z
0
100 N Te to,51 0,693
f

Nn
50

N rz
0 F (s)
500 600 700 800 900 1000 T
TEMPERATURE , °C

(b)

THERMOCOUPLE TYPES, WIRE DIAMETER d =Q5mm

200 CERAMIC SHEATH, d s =16mm
H

z 150 FT(s)
z
a
z 100 NT. = to.s/0,693
0
V
4, NTT
50
F

NT2 y FT (s)
D
300 600 700 800 900 1000
TEMPERATURE - °C

Figure 15 .21 Characteristic parameters of thermocouples, measured in an electric chamber furnace in

still air
 MEN a- I 1 ~~ 1

ME i=

w = '

w"%

II
11'~,\\ 11' " IW

~-~~ 11
11
DYNAMIC PROPERTIES OF CHOSEN SENSORS 319

N 10 ~- =0,35mm
6 d=4,6mm~ 1mm
3mm
2
- .- BARE THERMOCOUPLES
c 102
6 - MI-THERMOCOUPLES 3,2mm
WITH INSULATED 1,6 mm
a MEASURING JUNCTION 1mm
0,002
4 6 10-3 2 4 6 10 2 4 6 10, 2 4 8
FREQUENCY f , Hz

Figure 15.24 Comparison of amplitude frequency response of some thermocouples at 700 °C

Approximating the dynamics of MI thermocouples by the equivalent transfer functions

of first order inertia elements gave relevant time constants (Figure 15 .25) which were found
from the as frequency responses, as NT = 1/cJ c shown in Figure 15 .14 . Data from
Figure 15 .26, can be used for approximate estimation and comparison of the dynamics of
different thermocouple sensors. The equivalent time constants NT, have been found from
the heating-up curves from ambient temperature to 0e as NT, = tO.5/0.693 .

(a) Ibl

50 --- BARE THERMOCOUPLES 50 500°C

MI-THERMOCOUPLES
WITH INSULATED
MEASURING JUNCTION
40 40
d =3mm 500°C
H

\\ z
z 30 30 /
r
z \~ / 7000C

a 20 \\ d=4,6mm
o / 7000C
Uj 900K

1mm~\ / /
10 3,2 mm 1O
w
1,6mm
0,35 mm tmm
D ____ I I I f-T I
500 700 900 0 1 2 3 4 5
TEMPERATURE 4, °C DIAMETER d , d mm

Figure 15.25 Time constants versus temperature in (a), and diameter in (b), of some thermocouples
320 DYNAMIC TEMPERATURE MEASUREMENT

350 THERMOCOUPLES TYPE K

d=2mm IN STEEL SHEATH d=22/16mm

300 d =0,54mm IN DOUBLE CERAMIC
SHEATH d=1016 mm . d=22118 mm
250 d=2mm IN STEEL SHEATH d=15112mm
d =0,54mm IN CERAMIC SHEATH
` 200 d=14/10mm
d=3mm BARE
i
'~dy =0,54mm IN CERAMIC SHEATH
z 150 d=1016mm
d 2mm BARE

131 100 d w,=3,2mm MI-THERMOCOUPLE

-d ,=0,54mm BARE
E
'' 50 d=0,9mm BARE

0 _ ~
200 400 600 800
TEMPERATURE , °C

Figure 15 .26 Approximate values of time constants of Type-K thermocouples in an electric chamber
furnace in still air . The time constants were found from a step input response from ambient to
temperature, .4e

15.6 Correction of Dynamic Errors

15.6.1 Principles of correction
To eliminate the dynamic error of a temperature sensor, a corrector with a transfer function
GC(s), shown in Figure 15.27, can be used. The dynamic error will be zero if:

GTC (s) = GT (s)Gc (s) = KTKC (15.88)

Thus the transfer function of a perfect corrector should be :

_ KTKC
GC(s) (15 .88a)
GT(s)

where Kc is the corrector gain. If the sensor is approximated by a first order inertia
element, with the transfer function given from equation (15 .66) in Table 15 .3, the transfer
function of the corrector should be:

GC (s)=KC (1+sNC ) (15 .89)

where Nc= NT .
Since each practical corrector has some inertia, which has to be considered in the
transfer function of equation (15.89), the final corrector transfer function is:
CORRECTION OF DYNAMIC ERRORS 321

1 + sNC
GC (s) = KC (15 .90)
NC
l+s
k

where k is the correction coefficient, k > 1 .

Satisfying the condition NC =NT, the transfer function ofa sensor-corrector set will be:

1 1
= (15 .91)
GTC (s) KTKC = KTC
kC
l+s l+s
kT
From the above equation it follows, that a corrected system, like the sensor on its own, is
also a first order inertia element, but with a time constant k times smaller . In practice
corrector types described by equation (15.90) are also used to correct the dynamics of
sensors described as second order inertia elements, having a transfer function like
equation (15 .67) in Table 15 .3 . If NC = NTt, the sensor-corrector set approaches a double
inertia sensor, whose larger time constant, NTI, is reduced k times . The optimum corrector
parameters, NC and k, depend on the type of sensor transfer function, the range of
variations of sensor dynamics and the time dependence of the measured temperature .
Optimisation criteria are usually based on the comparison ofthe step input response and
the amplitude versus frequency characteristic of the sensor and of the sensor-corrector
assembly. Further consideration of this question appears in Bainbridge and Kaltner (1984),
Eckersdorf (1980), Hofmann (1967a, 1970, 1976), Kraus and Woschni (1979) and
Michalski and Eckersdorf (1978) . Performance indices, typically based upon the response
time tr,5%, and the cut-off frequency of f5%, are used to judge the correction efficiency.
Figure 15 .28 presents the dynamic errors of the step response of a sensor, having the
transfer function of equation (15 .66) connected to a corrector having the transfer function of
equation (15 .90) for different corrector parameters. For the same set, Figure 15 .29 also
presents the influence of corrector parameters on corrector performance indices, defined as
the ratio of response times of sensor-corrector, tr,si,TC to the response time of the sensor
alone tr,5%,T . It is apparent that the deciding influence on assembly performance is the
correct choice of the ratio of corrector time constant NC to that of sensor NT.

15.6.2 Analogue correction

Passive correctors The simplest passive correctors, having the transfer function of
equation (15 .90) are shown in Figure 15.30. A characteristic feature of all passive
correctors is their gain KC which is the reciprocal of the correction coefficient k. It follows
that the steady-state value of the corrector's output signal is k times smaller than the steady-
state value of the input signal . Thus, a preliminary amplification of the sensor output
signalis necessary. In passive correctors it is thus difficult to get large values of the time
constant NC because of the limited range of the values of R, L and C which may be used .
As high values of correction coefficient, k, give increasing noise level, the value of k cannot
be increased indefinitely.
322 DYNAMIC TEMPERATURE MEASUREMENT

In practice, Souksounov (1970) asserts that passive correctors can be applied, when
NC <5sandk<30 .

0,5 r., \
N INT =1,5 1 1+SNC
c
\T_ OTI(s)
=KT1+SNTKCN
1.
N~ /N T=1

NC IN T =0,5
i

' k=100
CORRECTION
z -0' S
r
k=5 }COEFFICIENT
° ' TEMPERATURE SENSOR ALONE
NT -SENSOR TIME CONSTANT
NC -CORRECTOR TIME CONSTANT
-1
0 Q5 1,0 1,5 t1N 2,0

Figure 15.28 Dynamic errors of sensor-corrector set having a transfer function, GTC(s), with a step
input

tus%,TC k=5
tu5%,OT,B
OTC (s)=KTI+SNT KC k=10
i
1+5NC i
k k=20 ./

~k=100
0,6 ~~.

k=50

0,4-

k- CORRECTION
COEFFICIENT
0,2

0
0,5
L-1 ~
1,0
' I I I
1,5 NC /N T

Figure 15.29 Ratio of 5 % response times of sensor-corrector set to sensor alone, versus the ratio of
the corrector time constant, Nc , to the sensor time constant, NT

(a) (b)
Ri
C
NC R, C
_ II L
k R L NC-
Rz
Ri Rz k= R+ +R z
R R2
Rz KC= Rz
RI+Rz
K = R2
0- 0 C RI+Rz

Figure 15 .30 Types of passive correctors

CORRECTION OF DYNAMIC ERRORS 323

Active correctors By combining proportional, derivative and integral elements in

different configurations, based on operational amplifiers, it is easy to design a corrector of
any required transfer function . The same transfer function can be achieved from many
different circuits . To characterise the usability of different circuits, and thus to help in a
correct choice, two indices proposed by Hofmann (1967a, 1976) are used for the circuits
described by equation (15 .90) . The first is a sensitivity factor, Q 1 , written as :

Q1 = ~ KcI = .f(k)

where KC is the corrector gain and k is the correction coefficient .

The second is a ratio of time constants, Q2 , given by :

NC
Q2 =
NCe

where NC is the time constant in the corrector transfer function and NC,, is the time constant
of the applied derivative or inertia element .
For sensors with large time constants, correctors should have large values of Q2 . The
value of the index, Q 1 , is of decisive importance if the sensor output signals are small, when

Ql should be as high as possible .

Some chosen active corrector circuits, having transfer functions as in equation (15 .90)
are presented in Figure 15 .31 (Hofmann, 1976). More sophisticated circuits are described
by Hofmann (1976), Praul and Hmurcik (1973) and Souksounov (1970) .
In correctors, based on operational amplifiers, it is necessary to amplify the sensor
output signal to the voltage levels needed by operational amplifiers . As active correctors
based on operational amplifiers will also be influenced by amplifier noise, circuits which do
not contain differentiating elements as in Figure 15 .31(b), are preferred to those in
Figure 15 .31(a) .
A detailed analysis of the quality factors of operational amplifiers, described by the
influence of their noise, drift, internal resistance and dynamic behaviour is given in
Souksounov (1970) . In a similar way to the passive corrector, the maximum values of NC
and k are also limited in active types by the quality of the R and C elements and of the
operational and measuring amplifiers . Nevertheless, the maximum achievable values of NC
and k of active correctors are much higher than in passive ones . For example, Souksounov
(1970) states that some of them even reach NC 1 = 800 s and k = 500 .

Adaptive correctors Under changing measuring conditions, adaptive or self-tuning

correctors should be used . The performance of a corrector critically depends upon good
matching of its parameters to the transfer function of the sensor . In changing measuring
conditions the equivalent transfer function of the sensor also changes . Eckersdorf (1980)
has shown that if the main time constant of the sensor, NT or NTl respectively, changes by
more than 30 to 40 % in relation to the value for which the time constant of the corrector
has been chosen, then the application of correctors with constant parameters is not advised .
324 DYNAMIC TEMPERATURE MEASUREMENT

(a) Cz

C INC -R, C Q J<

k R2IRT<
k
Rz k-RiC1
0,=k R z IRl =Ilk
RZCz
R1
K, =- R 0, >> k R2/R,>>
k
(b) C 0z=1

Rz R, RZ < .L
Ri R NC = C 01
R1-R,

R R3 R'
k = R'7R2 0'=k
z

-0 -0 Kc = R3 (R,-Rz )
RT (R+R3 )
C K"
(C)
NC- KKK 
CR 01»
k
R K' KTr~KK~ .
k= 02=k

Kc = K'

Figure 15 .31 Types of active correctors

To design a corrector, which adapts its parameters to changing sensor dynamics, it is

necessary to identify the sensor parameters continuously during the measurements . Different
ways of solving this problem have been presented by Souksounov (1970) and Zubov and
Semenistyj (1973, 1974) .
The principle of an adaptive corrector, in which two temperature sensors are used, is
shown in Figure 15 .32 . An auxiliary sensor, T2, with a time constant, NT2 , is placed, side-by
-side with the measuring sensor, T 1 , having a time constant NTI,(NT2 > NT1) . Considering,
that both sensors are subject to the same heat transfer conditions to and from the
surrounding medium it is valid to assure that :

NT2 = C =constant (15 .92)

NT I

MEASURING SENSOR MAIN CORRECTOR RECORDER

G T1 (s) G, (s)

T1

FINAL CONTROL
AUXILIARY SENSOR AUXILIARY CORRECTOR ELEMENT

G TZ (s) EGc,(s) E

T2 - _ .- I

Figure 15 .32 Block diagram of an adaptive corrector

CORRECTION OF DYNAMIC ERRORS 325

It is also assumed, that the transfer functions of both sensors, GTI (s) and GT2(s), are as
in equation (15 .66)and that the transfer functions of the correctors, GCI(s) and GC2(s), are
as in equation (15 .90) . Assume further, that the correction coefficient k2 of the auxiliary
corrector is k2 = C and that the time constants of the correctors are such, that NU N I = C.
It may be easily proven, that the condition of optimum correction NCI = NTI is satisfied,
when the difference signal, s, equals zero . Assume a change of the time constant of the main
sensor from NTI to NT I and that of the auxiliary sensor from NT2 to NTZ . The difference
signal, s , driving the final control element results in a new setting of both correctors until a
new equilibrium state is reached . This state occurs when :

N'C
N' I = NT', (when = C)
NCI

As the adaptive correctors described above are rather complicated and expensive they
are rarely used in an industrial environment . However, they are easy to implement using
microprocessor systems, and are applied in research .

15.6.3 Digital correction

Correcting action may be implemented using computers or more frequently microprocessor
based systems . These popular devices offer many new possibilities such as :

" Optimisation of correcting system parameters in given operation conditions due to easy
construction of numerical models of temperature sensors .
" Possibility of easy design of programming prototypes for correcting systems, coupled
with real temperature sensors, for experimental verification of the described methods .
" Mass production of easily tuned miniaturised correctors, mounted in the head of
industrial sensors .
" Realisation of self-contained correcting systems adapting their structure and settings to
actual operating conditions .

Because computer systems operate in discrete time, realisation of digital correction is

based on representing the sensor and corrector as discretised mathematical models, as
shown before in Section 15 .6 .1 in equation (15 .90) and in Table 15 .3 . The main point is to
transform from derivatives of function df/dt in differential equations to the corresponding
differential quotient AflAt. The simplest way to evaluate the differential quotient of a
function, f(t), at instant, t, is to determine the rate of variations of the function in the
neighbourhood of instant, t, such that :

O.f
= f(t) - .f (t-0) (15 .93)
At A

where A is the sampling period .

326 DYNAMIC TEMPERATURE MEASUREMENT

Although several methods for the evaluation of differential quotients are known, only
the reverse difference can be used practically in real-time systems . Such a method is based
on the present reading of the signal and the sample read one step before (Takahasi et al .,
1976) . Moreover, the influence of measurement noise can be reduced by using a larger
number of samples for the calculations .
As the temperature sensor can be represented by a first order inertia transfer function
given by equation (15 .66), the relation of Laplace transforms of its input signal, 9(s), and
output signal, YT(s), is given by :

YT(t) = 1
KNT 1)(s) (15 .94)

After transformations, the time form of equation (15 .94) can be written as :

YT (t) = -NT ddt + KT 6(t) (15 .95)

Transforming from derivative to differential quotient in equation (15 .95), the discrete time
form of the sensor output signal at instant, tn , has the form :

T
(15 .96)
Yn = A NT (KTO, + 4 Yn-1)

where 6, and yn are the n-th samples of the input and output signals respectively and

value of the output signal at the instant t - A.

Yn -1 is the
Similar analysis allows the equation related to the corrector transfer function of
equation (15 .90) to be derived to obtain :

Yn = LKC + KA T 01- K T 6,-1+ Yn -1 I (15 .97)

A
A+k
NT (1 kA J

Discretisation of continuous functions makes the sample time, A, an additional

parameter which may influence the properties of a digital correcting procedure . Although
the quantisation step, resulting from ADC conversion, can also introduce an additional
error, it is pointed out in Chapter 13 that this influence can be minimised by using an ADC
with sufficient resolution .
Exemplary PASCAL source code of a correcting procedure, written in accordance with
equation (15 .97), is shown in Figure 15 .33 . Such a procedure can be useful for modelling,
simulation and also for practical implementation of the real time correcting system .
CORRECTION OF DYNAMIC ERRORS 327

procedure corrector (Kc,Nt,K : real ; (corrector parameters)

del : real (sampling period}
var in,out : array [-1 ...0] of real) ;
(input and output signal arrays)
begin
out[0] :=Kk*(l+Kc*Nt/del)*in[[0]-Kc*Nt/(k*del)*in[1]
+Nt/(k*del)*out[-1] ;
out[-1]= :=out[0] ;
in[-1] :=in[0] ;
end
Figure 15.33 PASCAL procedure for a correcting function

Simulation and prototyping of correcting systems can also be performed effectively on a

PC using professional software packages like MATLAB-SIMULINK offered by Mathworks
Inc. (Mathworks, 1994, 1996) . This software allows computer models of various dynamic
systems to be built by object-oriented programming within a user-friendly graphical
environment . Figure 15 .34 shows the window containing a sensor-corrector system built
using this package . Such a computer model allows many simulations and detailed
investigations of correcting systems to be performed easily. Consideration of the typical
operating conditions, like measurement noise, uncertainty as to the knowledge of sensor
dynamics, various changes of real temperature and others can also be taken into account in a
straightforward manner . For example the simulation results of the correcting action for a
thermocouple of equivalent time constant NT = 10 s, coupled with a corrector with the
parameters, NC = 10 s and k = 50, are shown in Figure 15 .35 . Three periodic input signals
have been investigated and noise of standard deviation of 0 .1 °C has been added .

_ p x
_File iadit View simulation Format Tools

NOISE

Ks + (Kc*Nt)s+Kc
(Nt)s+t + (NUk)s+1
Sum
GENERATOR RECORDER
SENSOR CORRECTOR

. .__. ._. _. ._ ._ ._ ,
Ready -,,o de45- .

Figure 15 .34 Digital model of sensor-corrector system using MATLAB-SIMULINK package

328 DYNAMIC TEMPERATURE MEASUREMENT

lal (b)
400 400
TRUE TEMPERATURE TRUE TEMPERATURE
SENSOR+CORRECTOR+FILTER SENSOR+CORRECTOR FILTER
380 380-
V V
O O
360 , 360-

340- 340
W

f- 320 320
SENSOR SENSOR
300 300
60 70 80 90 100 60 70 80 90 100
TIME s TIME s
(c)
400 TRUE TEMPERATURE
SENSOR+CORRECTOR+FILTER
FILTER
380
0
360-

340-

320-
- SENSOR
300
60 70 80 90 100

Figure 15 .35 Correction of periodic sensor signals : (a) sinusoidal, (b) square wave and (c) triangular

Figure 15 .36 shows how the MATLAB-SIMULINK package may be configured for the
rapid prototyping of correcting systems using Real-Time Workshop (Mathworks, 1995) as
an interface between real temperature sensors and the software .
Microprocessor based systems are of paramount importance for the development of
adaptive correction (Michalski et al ., 1996), whose idea is presented in Figure 15 .37 .
Adaptive correcting systems, which are mostly realised by computer software, consist of
the corrector and adaptation mechanism. They comprise the combination of an in-situ
identification procedure and the calculation and setting of the corrector parameters . When

SENSOR IN REAL
OPERATING CONDITIONS

DIA RTW AID

v
MATLAB OPERATING
FURNACE SIMULINK
SYSTEM

Figure 15 .36 Prototype correction system using PC based MATLAB-SIMULINK package and a
Real-Time-Workshop
CORRECTION OF DYNAMIC ERRORS 329

-
,F-SOFTWARE - - - - - - - -
I SENSOR DYNAMICS DATA I
I I
IN SITU SETTING OF
I IDENTIFICATION CORRECTOR I
I PROCEDURE PARAMETERS I
ACTIVATION I
OF ADAPTATION -f ------ --- I
PROCEDURE L- --
INTERNAL I
HEAT I I
GENERATION I I
I
SENSOR YT CORRECTOR
I I

Figure 15.37 Block diagram of microprocessor-based adaptive correcting system

the adaptation is activated by the operator, the in-situ identification procedure is performed
by internal heating of the sensor as described in Section 15 .4 .5 . After the identification is
completed, newly calculated corrector parameters are used to obtain a properly tuned
correcting system for use in real measurements .

15 .7 References
Bainbridge, B.L . and Kaltner, N.R . (1984) Application of transient response characterisation and
enhancement to thermal instrument . Temperature Measurement Symposium, Knoxville, TN, 21 .01-
21 .18 .
Berger, R.L . and Balko, B . (1972) Thermal sensor coatings suitable for rapid response biomedical
application. Temperature : Its Measurement and Control in Science and Industry, 4(3), Instrument
Society of America, Pittsburgh . 2169-2192.
Bernhard, F. (1979) Umrechenbarkeit dynamischer Kennwerte von Beriihrungsthermometem auf
andere Betriebsbedingungen . m.s .r ., 22(8), 448-449 .
Bernhard, F. and Noack, G. (1981) Possibilities and borders of the identification of dynamical
parameters of thermometers under working conditions . First Symposium of the IMEKO TC-12
Temperature Measurement in Industry and Science, Karlovy Vary, 219-231 .
Bliek, L., and Fay, E. (1979) Ein Messverfahren zur einfachen Ermittlung des Ubertragungsver-
haltens von Temperaturaufnehmem . Techn. Messen, 46(7/8), 283-286, 291-292.
Caldwell, W. I., Coon, G. A. and Zoss, L. M. (1959) Frequency Response for Process Control,
McGraw-Hill, New York .
Chohan, R.K. and Natour, M. (1988) Temperature sensor testing in air. J. Phys . E: Sci., Instrum., 21,
550-553.
Doetsch, C. (1961) Anleitung zum praktischen Gebrauch der Laplace-Transformation, R.
Oldenbourg, Miinchen. .
Eckersdorf, K. (1980) Optimisation of correctors for correction of dynamics of electric temperature
sensors, PhD Thesis. Technical University Lodz (in Polish).
Eckersdorf, K. (1986) Experimental determination of dynamic properties of temperature sensors
installed in electroheat plants . 31 Int. Wiss Koll, T. H. Illmenau, 27-31 .
Eckersdorf, K. and Michalski, L. (1984) Dynamics of the low inertia temperature sensors under the
conditions of the radiant heat exchange . Temperature Measurement in Industry, and Science,
TEMPMEKO-84, 2nd Symposium IMEKO TC-12, Kammer der Technik, Suhl, 305-314.
330 DYNAMIC TEMPERATURE MEASUREMENT

Eijkman, E.G.J . (1955) Temperatuurmetingen, Fouten bij dynamische Temperamurmetingen. De

Ingenieur, 67(2), 0.15-0.30
Eijkman, E.G.J. and Verhagen, D.M . (1958) Response and phase-lag of thermometers . Process
Control, 158-166
Godfrey, K. (1993) Perturbation signals for system identification, Prentice Hall, University Press,
Cambridge.
Gr6ber, H., Erk, S. and Grigull, U. (1963) Die Grundgesktze der Wlirmeiibertragung . Springer-
Verlag, Berlin .
Hackforth, H.L. (1960) Infrared Radiation, .McGraw-Hill, New York.
Hashemian, H.W., Petersen, K.M ., Mitchell, D.W., Hashemian, M., and Beverly, D.D . (1990) In situ
response time testing of thermocouples, ISA Trans. 29(4) .
Henderson, I.A ., Ibrahim, A.A ., McGhee, J. and Sankowski, D. (1987) Assembler generated binary
sequences for process identification . IFAC Proc Series, 1987(7), Pergamon Press, Oxford . 77-82.
Higgins, S.P . and Keim, J.R . (1954) A thermal sine wave apparatus for testing industrial
thermometers . ASME, 76, Paper 54-SA-20, 1-9
Hofmann, D. (1965) Mepdynamik elektrischer Industriethermometer, m.s.r., 8(12), 407-410
Hofmann, D. (1966) Me(3dynamik elektrischer Industriethermometer, m.s .r ., 9(1) 17-20.
Hofmann, D. (1967a) Zur elektrischer Korrektur des dynamischen Verlaltens von tragen,
Me(3wandlem, m.s .r., 10(1), 20-26.
Hofmann, D. (1967b) Zur Me(Jdynamik von elektrischen Beruhrungsthermometem, IMEKO IV,
Warszawa, DDR-160.
Hofmann, D. (1970) Erh6hung der Genauigkeit von Me(3systeme mittels Korrektur IMEKO V,
Versailles, B-404.
Hofmann, D. (1976) Dynamische Temperaturmessung. Verlag Technik, Berlin
Huhnke, D. (1973) Neue Apparaturen zur Aufnahme der Ubergangsfunktion elektrischer
Berahrungsthermometem in str6menden Medien . VDI-Berichte, 1987, Technische Temperatur-
messung, 103-109.
Jackowska-Strumillo, L., Kucharski, J., Sankowski, D., (1996) On line and in-situ identification of
resistance temperature sensors dynamics by use of multifrequency signals. 3"d International
Symposium on Methods and Models in Automation and Robotics . MMAR'96, Miedzyzdroje, 661-
664 .
Jackowska-Strumillo, L., McGhee, J., Henderson, I.A . and. Al Muhaisni, M. (1992) : Instrumen-
tation and experimentation for identification, monitoring and diagnosis of temperature sensors,
Temperature: Its Measurement and Control in Science and Industry, 6(2) . American Institute of
Physics, New York, 1043-1048 .
Jackowska-Strumillo, L., Sankowski, D., McGhee, J. and Henderson, I .A. (1997) Modelling and
MBS experimentation for temperature sensors. Measurement, 20(1)_49-60 .
Jakob, M. (1957) Heat Transfer, Vol. 2, John Wiley and Sons, New York.
Jakob, M. (1958) Heat Transfer, Vol. 1 (6th edition), John Wiley and Sons, New York .
Kerlin, T.W ., Miller, L.F . and Hashemian, N.M . (1978) In situ response time testing of platinum
resistance thermometers ISA Transactions, 17(4), 71-88.
Kerlin, T.W ., Hashemian, N.M. and Peterson, K.M . (1981) Time response of temperature sensors.
ISA Transactions, 20(1), 65-67.
Kerlin, T. W., Hashemian, H. H., and Peterson, K. M. (1982) Response characteristics of temperature
sensors installed in processes. ACTA IMEKO III, North-Holland, 95-103 .
Kocurov, W.I . (1963) Determination of dynamic characteristics of temperature sensors. Energetika,
14(5), 91-97 (in Russian) .
Kondratiev, G.M . (1947) Universal method of determination of the index of thermal inertia . Trudy
WNIIM, 4,62-83 (in Russian) .
Kraus, M. and Woschni, F.G . (1979) Melinformationssysteme, EB Verlag, Berlin .
Kucharski, J., Eckersdorf, K. and Michalski, L. (1997) Dynamic properties of high pressure
thermocouples, IMEKO TC-12 International Seminar on Low Temperature Thermometery and
Dynamic Temperature Measurement, Wroclaw-Ladek Zdr6j, D7-D12 .
CORRECTION OF DYNAMIC ERRORS 331

Kucharski, J. and Sankowski, D. (1997) Practical aspects of MBS test signal generation for in-situ
thermocouple identification, IMEKO TC-12 International Seminar on Low Temperature
Thermometry and Dynamic Temperature Measurement, Wroclaw - Ladek Zdr6j, D13-1318 .
Lieneweg, F. (1938a) Bestimmung der Anzeigeverzogerung von Thermometern . ATM,
Lieneweg, F . (1938b) Anzeigedampfung von Thermometern bei zeitlichen Temperaturanderungen. A
TM, 44, T 143 -T 144.
Lieneweg, F . (1941) Bestimmung der Anzeigeverz6gerung von Thermometern. ATM, 6, T79-T80 .
Lieneweg, F . (1962) Die Ubergangsfunktion beim Abkuhlung and Erhitzen fester K6rper in
beliebigen Mitteln, einschliesslich der Anzeigeverz6gerung von Thermometern . Regelungstechnik,
10(4), 159-165 .
Lieneweg, F. (1975) Handbuch, Technische Temperaturmessung, F. Vieweg, Braunschweig.
Mathworks Inc. (1994) Matlab User's Guide.
Mathworks Inc. (1995) Real-Time Workshop User's Guide.
Mathworks Inc. (1996) Simulink User's Guide.
McGhee, J . (1989) The impact of electronics and information technology upon measurement and
control in thermal systems. Keynote paper IMEKO TC-12, Symp . Microprocessors in Temperature
and Thermal Measurement, Lodz.
McGhee, J., Fisher, G. and Henderson, I. A. (1987) A fast DFT algorithm for on-line MBS process
identification . Proc. CSCS7, 7th Int. Conf. Con. Sys. Comp . Sc ., 1, Bucharest, 111-119.
McGhee, J., Henderson, I.A., Ibrahim, A.A. and Sankowski, D. (1986) Identification : Systems,
Structures, Signals, Similarities, Systems Science IX, Int. Conf. on Systems Science . Wroclaw.
McGhee, J., Henderson, I.A . and Jackowska-Strumillo, L (1993) Temperature sensor identification
with multifrequency binary sequences, in Godfrey, K R. Perturbation signals for system
identification . Prentice-Hall, University Press, Cambridge, 277-297
McGhee, J., Henderson, I.A . and Mackie, S. (1991) Simulation of MBS idenification of temperature
sensors, IV International Symposium on Temperature and Thermal Measurement in Industry and
Science, TEMPMEKO 90, Helsinki, Finland, 201-206.
McGhee, J., Henderson, I.A ., Michalski, L., Eckersdorf, K. and Sankowski, D. (1992a) Dynamic
properties of contact temperature sensors: I. Thermokinetics, modelling and the idealised
temperature sensor . Temperature: Its Measurement and Control in Science and Industry, 6(2)
American Institute of Physics, New York, 1157-1162.
McGhee, J., Henderson, I.A ., Michalski, L., Eckersdorf, K. and Sankowski, D. (1992b) Dynamic
properties of contact temperature sensors: II . Characteristics modelling and testing of real
temperature sensors. Temperature : Its Measurement and Control in Science and Industry, 6(2)
American Institute of Physics, New York, 1163-1168.
McGhee, J., Sheppard, D. and Henderson, I.A . (1989) Immersion testing of temperature sensors using
a microprocessor based MBS generator/analyser Proc . IMEKO TC-12 Symposium.
Microprocessors in Temperature and Thermal Measurement, Lodz .
Meyer-Witting, O. (1959) Berechnungsverfahren zur Bestimmung der Anzeigeverz6gerung von
Temperaturfiihlern bei periodischen Temperaturanderungen . PhD thesis, T. H. Braunschweig.
Michalski, L. (1966) Temperature eines Kammerofens. Elektrotechniek (Netherlands), 44(20).
Michalski, L. and Eckersdorf, K. (1978) Electrical correction of dynamics of temperature sensors,
having temperature dependent properties . Mereni Teploty, Praha, 1978, Dum Techniky CSVTS, .
110-119 .
Michalski, L. and Eckersdorf, K. (1987) Temperature sensors in a closed loop temperature control.
TEMPMEKO 87, 3rd International Symposium on Temperature and Thermal Measurement in
Science and Industry . Sheffield UK, 127-136 .
Michalski, L. and Eckersdorf, K. (1990) In situ determination of dynamics of temperature sensors,
TEMPMEKO 90, 4th International Symposium on Temperature and Thermal Measurement in
Industry and Science, Helsinki, 193-200.
Michalski, L., Eckersdorf, K., Kucharski, J., Sankowski, D. and Urbanek, P. (1993) Experimental
computerised in-situ identification of thermocouple sensors. TEMPMEKO 93, 5th International
Symposium on Temperature and Thermal Measurement in Industry and Science, Prague, 107-111 .
33 2 DYNAMIC TEMPERATURE MEASUREMENT

Michalski, L., Eckersdorf, K., Kucharski, J. and Sankowski, D. (1996) Adaptive correctors for
dynamic temperature measurement. TEMPMEKO 96, 6th International Symposium on
Temperature and Thermal Measurement in Industry and Science. Torino, 475-480.
Michalski, L. Eckersdorf, K. and Sankowski, D. (1995) . Dynamic properties of temperature sensors.
Metrologia i Systemy Pomiarow.. 2(4), 289-302.
Michalski, L., Kuzminski, K. and Sadowski, J. (1981) Temperature Control in Electroheat, WNT,
Warsaw (in Polish).
Praul, S.H : and Hmurcik, L.V . (1973) Instantaneous temperature measurement . Rev . Sci., Instrum .,
14(9),1363-1364.
Zubov, W.G . and Semenistyj, K.C . (1973) Active corrector for the correction of inertia of
temperature sensors, Priborostroyenye, 14/15, 107-109 (in Russian) .
Rake, H. (1980) Step response and frequency response methods, Automatica, 16, 519-526.
Reed, R.P . (1972) The transient response of embedded thin film temperature sensors. Temperature :
Its Measurement and Control in Science and Industry, 4(3), Instrument Society of America,
Pittsburgh, 2193-2196.
Rubin, G. and Feldman, 1. (1968) Radiant heat exchange and dynamics of thermal processes in
resistance furnaces . V1. Int. Congress on Electroheat, Brighton, paper 314.
Sankowski, D., Michalski, L., Eckersdorf, K. and Kucharski, J. (1995) The use of MBS test signals
for determination of thermocouple sensor dynamics . Proc IMTC 95, IEEE Instrumentation
Measurement Technology Conference, Massachusetts, . 619-623.
Schwarze, G. (1964) Neue Erghebnisse zur Bestimmung der Zeitkonstanten im Zeitbereich, m.s .r .,
7(1), H 1 .
Skoczowski, S. (1982) Me(3fehler verursacht durch die dynamische Nichtlinearitat des Me/3fiihlers .
Techn . Messen, 49(1), 27-30 .
Souksounov, V.E . (1970) Correcting elements in instruments for measuring non-stationary
temperatures. Biblioteka po automatike, 33, Ind, Energiya, Moscow (in Russian)
Taylor, H.R. and Navarro, H.A. (1987) A method to determine and compensate for the frequency
response of a platinum resistance thermometer under practical conditions . TEMPMEKO-87, .
Thermal and Temperature Measurement in Science and Industry, Sheffield, 137-147.
Woschni, E.G . (1979a) Vergleich der statistischen Linearitats and dynamischen Korrektur von
Systemen mit Mikrorechnern, insbesondere in der Prozesstechnik . 24 Int. Wiss . Koll . Ilmenau, T.
H. Ilmenau, 3, 55-58.
Woschni, E.G. (1979b) Beziehungem zwischen den verschiedenen. Definitionen von dynamischen
Fehlern in der Me(3technik and Schlu(3folgerungen fiir praktische Anwednung, m.s.r, 22(5), 259-
261 .
Woschni, E. (1980) Die Me(3technik als Teilgebiet der Informationstechnik Techn., Messen, 47(1), 7-
13 .
Yarishev, N.A . (1967) Theoretical principles of measurement of non-stationary temperatures . lzd.
Energiya, Leningrad (in Russian) .
Zubov, W.G. and Semenistyj, K.C . (1973) Active corrector for the correction of inertia of
temperature sensors, Priborostrojenie, 14/15, 107-109, (in Russian)
Zubov, W.G. and Semenistyj, K.C . (1974) Self tuned measuring system with dynamic correction of
inertia of temperature sensor: . Odbor i Peredaca Inf, 40-45 (in Russian) .
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

16
Temperature Measurement of
Solid Bodies by Contact Method
16.1 Introduction
One of the most frequently encountered problems in temperature measurement is measuring
the temperature of solid bodies on their surfaces in contact with a surrounding gas or liquid .
This may be solved by either contact or non-contact methods. Semi-contact or quasi-contact
methods may also be used. Roeser and Mueller (1930) have previously discussed the
problem .
The non-contact, or pyrometric, methods are described in detail in Chapters 8 to 11 . For
a rough estimation of surface temperatures, temperature indicators, described in Section 2.5
are also used. The problem of measuring the internal temperatures of solid bodies,
considered in Section 16.6, is also similar to surface temperature measurement .

16.2 Theory of the Contact Method

It is assumed that a solid body in contact with a surrounding gaseous medium, as shown in
Figure 16.1, remains in a thermal steady-state . A surfacial heat source is placed inside the
solid body, whose true surface temperature, Ot, is higher than the ambient gas temperature,
Oa . The surface temperature, Ot , is to be measured by a contact sensor, which is a bare
thermocouple having a flat-cut measuring junction, in contact with the surface.
Figure 16.1(a) illustrates the original isotherms of the undisturbed thermal field. When the
contact sensor is introduced, the thermal field is deformed, as shown in Figure 16.1(b).
Assume that the heat transfer between the investigated surface and the surrounding gaseous
medium takes place through convection and conduction and that the isotherms in the gas are
deformed in the vicinity of the sensor . The corresponding temperature distribution in the
direction normal to the surface is shown in Figures 16.1(c) and (d) .
While making contact with the investigated surface, the sensor causes a more intense
heat flow from the surface, resulting in a drop in the surface temperature from its original
value, Ot, to a new temperature, 0', as shown in Figure 16.1(d) . The temperature
difference, Ad, = 0' - Ot , which is called the first partial error of the measurement, is
caused by the deformation of the original temperature field . Between the flat cut measuring
334 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

ORIGINAL STATE STATE AFTER APPLYING CONTACT SENSOR

SOLID BODY q GAS SOLID BODY q GAS
(a) (b)

A.

SENSOR

(c)~ (d) AA4 '% \ S-SENSOR SENSITIVE POINT ('8T)

Wc -THERMAL CONTACT RESISTANCE

.S, > .S 4r

'9 'T
"o A.
t ~ l

Figure 16.1 Surface temperature measurement ofa solid body by a contact sensor. The isotherms and
heat flux density lines, q, for the undisturbed conditions without the sensor are given in (a) with the
corresponding temperature distribution in the direction normal to the surface shown in (c) . When the
sensor is introduced the isotherms and heat flux density become distorted as in (b) and the
temperature distribution becomes that in (d)

junction of the thermocouple and the investigated surface there is always a thermal contact
resistance, Wc , caused by a non-ideal contact. The temperature drop across this contact
resistance, A62 = 6" - 0' , is called the second partial error.
It is further assumed, as in all sensors, that there is also a sensitive point in a contact
sensor determining the thermometer readings, OT . From Figure 16.1(b) this point, S, in a
contact thermocouple is placed at the distance, l' , from the investigated surface. The
temperature, OT, at the point S differs from 0" by a value A03 = OT - 0", called the
third partial error, which depends on the sensor design. All of the differences A?),, A02
and A63 are systematic errors of the contact method of surface temperature measurement
of a solid body in the thermal steady-state . To determine their values the temperature field
of a solid body in contact with a sensor and the heat flux entering the sensor will be
analysed. In Sections 16.2.1 and 16.2.2, the temperature excess, O, over ambient will be
used.

16.2 .1 Disturbing temperature field

The investigated temperature field of a solid body in contact with a sensor, according to
Kulakov and Makarov (1969), can be regarded as a superposition of two fields. Firstly,
there is the original temperature field in the body without the sensor, described by
THEORY OF CONTACT METHOD 335

Ob = f(x, r)and secondly, the disturbing temperature field Od = f(x, r) . The disturbing
temperature field is caused by the disturbing heat flux density, resulting from the difference
between the density, qT, of the heat flux conducted along the sensor and the density, q b, of
the heat flux transferred from the body to its ambient surrounding . This density, q d, of
disturbing heat flux is given by

qd = qT - qb =f(x,r) (16.1)

Using the semi-infinite body in Figure 16 .2 as an example, gives an explanation of the

manner of determining the disturbing flux and also shows the surfacial temperature
distribution. The medium value of disturbing temperature Od,m , shown in Figure 16 .2(c), at

the contact surface between sensor and body permits determination of the first partial error,
061 .
The differential equation of heat conduction, describing the disturbing temperature field
in a semi-infinite cylindrical body, is

2 -y-~ 2
zd + ~d
+
°d
=o (16 .2)
r a

with the boundary conditions :

jdE)d
L ax - ~ Od]x- _- ~ (16 .3a)
-0 r<<-RT

Od] x_0< °° ; Od]x-- -0 ; Od],.--=0 (16 .3b)

(a) BODY AND SENSOR (b) EOUIVALENT MODEL

2R
SENSOR
GAS ZRT
a -
qT qo qT qb a

XIl
qb _
//Tl 1' I'

X'b
-ql
I
SOLID
x BODY x
I I
r
(c) DISTURBING
TEMPERATURE I l
FIELD ON SURFACE I B d.m
OF SOLID BODY
Ba(x=O)

Figure 16 .2 Disturbing temperature field on the surface of a semi-infinite body, resulting from the
application of a contact temperature sensor
336 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

where A'b is the thermal conductivity of the body, and ab is the heat transfer coefficient at
the surface of the body.
Solving equation (16 .2) with the boundary conditions of equation (16 .3) gives:

= _ gdRT e-kxyh(y)IO(kry),, v (16.4)

od - v+B

x aRT
where v is the variable of integration, k,, = ; kr = r ; B = and ID (v) and
RT RT kb
I, (krv) are Bessel functions ofthe first kind and of order zero and first, respectively .
From equation (16 .4), let the value of the integral, a function of k,,, kr and B, be
described by F(kX ,kr ,B), so that equation (16 .4) becomes :

od =-gT (16.5)
~ F(kXkr,B)

To help with the practical use of equation(16 .5) Kulakov and Makarov (1969)
graphically display the values of the function F versus kX in Figure 16.3 and versus kr in
Figure 16 .4 as well as the mean value Fm of the function F, at the contact surface between
the sensor and the body, versus parameter B as in Figure 16.5.

F(r=O, B=0)
1,0 B= a RT
p,8 ~n
0,6
0,4
kx=
R T
Q2- II
01 1 ,.

I ~0 1 2 3 4 5 ~ kx
i I

11 5 6 7 8 9 10

Figure 16.3 Function, F, from equation (16 .5) versus parameter, kX, for r = 0, B = 0

F(x =0, B=01

1,0 B-_ 1Rr
Fm(0 ~ r < RTI
0,8-
0,6- k,=
0,4 I RT
I
0,2 I II

1 -~0~ 1 2 3 4 5 ~k,

11 5 6 7 8 9 10

Figure 16.4 Function, F, from equation (16.5) versus parameter, kr , for x = 0, B = 0

THEORY OF CONTACT METHOD 337

Fm (x=0 ; 0 < r -<R T )

1,0 a'RT
B=
~b
0,8

0.6

.4-
0

0,2-I

0 0.5 1,0 1,5 2,0 B

Figure 16 .5 Mean value, Fm, from equation (16 .5) versus parameter, B, for x = 0, 0<r<_
_ RT

Figure 16.6 presents a case in which the disturbing heat flux density, q d, diffuses into an
infinitely large plate of limited thickness, through a surface limited by a circle of radius R T
The differential equation of heat conduction, characterising the disturbing temperature field,
is the same as that given in equation (16 .2) for the semi-infinite body with the boundary
conditions :

d9d __ _ _
9d

Xb Od Jx-0 '113 lr<-RT

aod _ 0 (16 .6)

C od
~x-!b

Od Jr-O < °° ; Od Jr-- - 0

The solution ofequation (16 .2) for the boundary conditions (16 .6) is:

°" (v + Bl )e [v(k, -k .)] + (v - )el (k, -k .)]

- gdRT B2 -v
Od = )ek)v jl (v)Io (kr dV
Ab 0 (v + Bj)(v + B2 - (v - Bl)(v - B2 )e-k`v ) (16 .x)
T H(kx,kr,kl,Bl,B2)
--q~b

(a) PLATE AND SENSOR (b) EQUIVALENT MODEL

2R
SENSOR
2R
-
a, }qdqT qb °b,
qb qT -qe

r j
i ,Zy / Ab

` PLATE

qd -DISTURBING HEAT
x FLUX x

Figure 16.6 Infinitely large plate with a surface temperature sensor

338 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

x r
where k,, = , kr = , kl = lb , B1 = atRT and B2 = azRT .
RT RT RT 4 Ab

To calculate the value of the first partial error, At9 1 , which equals the medium value of
the disturbing temperature, O d,m , at the contact surface between the body and sensor, the
function H from equation (16 .7) must be known for k, varying in the range
0<kr <_1(0<_rSRT) and for kx =0.
As the function H depends upon several parameters it is difficult to display it graphically
in an universal way . Figure 16 .7 presents values of H, for kr = 0 and k, = 0, as a function of
relative plate thickness k l for some chosen values of B 1 = Bz = B. This corresponds to a case
when the heat transfer coefficients on both sides of the plate are the same (a1 = a2 = a) .
Figure 16 .7 permits the maximum value of disturbing temperature in the centre of the
contact surface to be found . From the curves of H = f(kv, it follows that the disturbing
temperature values decrease with increasing plate thickness . If the plate is sufficiently thick
it can be regarded as a semi-infinite body and equation (16 .5) can be applied . Kulakov and
Makarov (1969) discuss some bodies with finite dimensions and other shapes .

16.2.2 Heat flux entering the sensor

As follows from equation (16 .1), determination of the density, qd, of the disturbing heat
flux, requires knowledge of the density, qT , of the heat flux entering the sensor. The four
simple sensor models, given in Figure 16 .8, will be considered under some simplifying
assumptions . It will be assumed that the heat transfer coefficients have constant values and
that the temperature field in the cross-section of the rod or plate is uniform, with a uniform
density of heat flux all over the front surface of the sensor as well . In addition it is assumed
that the rods shown in Figure 16 .8(b), Figure 16 .8(c) and Figure 16 .8(d) are infinitely long .

k= b
7 B=10 R
6
'B=10, BI =82 =B
5
(Cci=oc2=a)
B=103
4
Y -2 .
0 3 B=10 B= a Rr
,. ~n
2

0 1 2 3 4 5 6 7 8 9 10 11
ki -

Figure 16 .7 Function, H, from equation (16 .7) versus relative plate thickness, kj, for different B
values
THEORY OF CONTACT METHOD 339

x 2R,
(a) DISK x~ GAS (b) ROD

JT
2RP 1%
2
a,
P
7-
XP
r r
`qT qT OT 4T
~ SOLID BODY
x
OT ~Pi

(c) THERMOCOUPLE SENSOR (d) DISK THERMOCOUPLE SENSOR

x x

2R,i
2R
2R, 12 , 2Rc2

oc c am ac2
' as X,2 '% P2

Y
'~C1 Xc2 a:

77/71=/
~'l 1-777
, r r
qci q,2 0,2 ' qT 2RP ~P1
l J
0T = 0 c1 +0,2
T

Figure 16.8 Sensor models for evaluating the heat flux, (DT, entering the sensor with the associated
heat flux density, qT

For each model, both the heat flux, (DT, entering the sensor, as well as the heat flux
density, qT , at the sensor surface, AT, in contact with the investigated body, will be
determined. These two are related by:

qT = AT (16 .8)
T

Also to be determined is the thermal resistance, WT, of the sensor, defined by

Mackiewicz (1976a) in terms of, OT , the temperature of the surface of the sensor in
contact with the investigated body as,

WT = (16 .9)
~T
T

Disk sensor: To simplify the problem of the disk, shown in Figure 16.8(a), it is
assumed that the disk (plate) only transfers heat to the environment from its upper
340 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

surface (1p << 2Rp ), and that Op I = Opt = 6T . This corresponds to a thin disk, with a
large value of thermal conductivity Xp. The heat flux (DT is:

(DT =KRpapWT -6a)=IrRpapOT (16.10)

with the heat flux density given by:

9T = -apOT (16.11)
icR2
P

The thermal resistance of the disk, following equation (16.9), is:

WT = 1 (16.12)
7rR2ap

Rod sensor: The differential equation describing the temperature distribution along the
rod (conductor) of the sensor in Figure 16.8(b) is:

d 2 O c (x) - 2ac
O c (x)=0 (16.13)
dx 2 AC R,

with the boundary conditions :

Oc(x)],=o=OT ; Oc(x)]x--=0 (16.14)

The solution of equation (16.13) for the boundary conditions in equation (16 .14) is:

F27,
Oc(x)=0T exp - x (16.15)
( Ac Rc )

The heat flux, (DT, entering the sensor equals the total heat flux transferred to the
environment from the side surface of the rod in accordance with:

(DT = f ~21rR c acOc (x)dx (16.16)

0

Substituting the expression for Oc(x) from equation (16 .15) into equation (16 .16),
yields:
THEORY OF CONTACT METHOD 341

q'T = IrR, 2a cAc Rc OT (16 .17)

The heat flux density, q T, is then

qT = z (16.18a)
IrRc

and finally :

2a° A°
qT - OT (16.18b)
RC

Following the definition of equation(16 .9), the thermal resistance, WT, of the rod is
given by:

WT (16 .19)
7rR c 2acX~Rc

Double conductor thermocouple sensor : Figure 16.8(c) shows that this sensor may be
regarded as two independent rods from the model in Figure 16.8(b) . In most cases it can
be assumed that:

Rcl = Rc2 = R c > ac, = ace = ac ; OTi = OT2 = OT

and then the total heat flux, (DT, entering the sensor, is

(DT = Rcir 2acRc (XI + jc2 )E)T (16 .20)

and the thermal resistance of the sensor, following equation (16 .9) will be:

WT = I (16.21)
R c ir 2a c Rc ( jci + jc2 )

The heat flux density of each conductor is given by equation (16 .18).

Disk thermocouple sensor : The simplifications for Figure 16.8(d), are the same as for
the models of Figure 16.8(a) and 16 .8(b) but with the assumption that Rc << RP . The
total heat flux, (D,,,, entering the sensor then equals the sum of the heat fluxes of both
conductors and of the disk (plate) :
342 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

(PT = [irR2a,
P +xR, 2a~R, + Jz ) JOT (16.22)

The appropriate heat flux density in this case is :

gT =
7rRzP
= aP + Ri 2a~R, ( ~~, + ~~ z) OT (16 .23)
P

and the thermal resistance, WT, conforming with equation (16 .9) is :

WT = 1 (16 .24)
it[Reap +Rc 2a~R, (X,, + A2

16.2.3 Method errors and their reduction

The first partial error, A6 1 , agreeing with the definition from Section 16 .2 is :

A61 = 6'- 6t (16 .25)

This error equals the medium value of the disturbing temperature Od,m, at the contact
surface between the sensor and the body . For a semi-infinite body, from equation (16 .5), the
error, A6 1 , is described by

= Od .m]x=0 gdRT ~ (16 .26)

001 Fn
0_<r_RT /~b x=0
0<r<RT

The density, q d , of the disturbing heat flux in equation (16 .26), is calculated from
equation (16 .1) as :

qd = qT - qb (16 .27)

Determination of, q T , for the four sensor models is described in Section 16 .2 .2 . To

simplify the problem it is assumed that 6T = t9' (Wc = 0) . The value of qb is calculated
from q b = abO T , while the value of Fm in equation (16 .26) is found from Figure 16 .5 .
Calculation of the error, A6 1 , is accomplished in a step by step iterative way or
graphically . This will be explained in a numerical example . To calculate these A6 1 errors,
a ready formula can also be used . This is derived for a semi-infinite body by Mackiewicz
(1976a), taking into consideration the thermal contact resistance, Wc , in the manner:
THEORY OF CONTACT METHOD 343

+W _ (WT ' )/W.-1 O

A~ - (16 .28)
` 1+(;rRT Ab /F.)(WT +W) `

where Ab is the thermal conductivity of the investigated body, WT is the thermal resistance
for each sensor described in Section 16 .2 .2, W, is the thermal contact resistance between the
sensor and the investigated body and Wa is the thermal resistance of heat transfer from the
surface of the body to the environment, before applying the sensor, given by the relation :

Wa = 1 (16.29)
aRTtc

In a similar way, based on the theory of disturbing heat flux for an infinitely large plate,
the error, A6 1 , can be found with the calculations based on equation (16 .7) . The appli-
cation range of this method is limited to the case, when the heat transfer coefficients on both
plate surfaces are the same (al = a2 ) and the plate itself is composed of only one layer .
There are different ways to reduce the first partial error A6 1

1. Error A01 can be reduced by increasing the contact surface area between the sensor
and the investigated body. In this way the heat flux density is diminished at the contact
area and thus the deformation of the original temperature field is also reduced . One of
the ways to achieve this is to apply an additional metal disk of high thermal
conductivity as shown in Figure 16 .9(a) .
2. Reduction of the value of A0 1 can be obtained by decreasing the heat flux conducted
from the measuring point along the sensor or its conductors . For this purpose the
thermocouple conductors should be as thin as possible and should initially be led for a
short distance, parallel to the investigated surface which is along the isotherms . Any
influence of the conductor radius, R C , on the heat flux entering the sensor is seen from
the relation of equation (16 .17) .
3. Total elimination of the error, A61 , can be achieved by applying a thin disk sensor,
made from a high conductivity material with the same emissivity as the investigated
surface and fastened permanently to it . In such a solution the heat flux densities qb and
q T are equal (qb = q T) as well as thermal resistances Wa = WT , (ab = ap ) . For a disk

fastened permanently to the surface the thermal contact resistance, between the sensor
and the surface, is nearly zero (Wc ~L, 0) . From equations (16 .26) and (16 .28) it follows
that A61 = 0 in the case described above .

The second partial error, A6 2 , is defined, in agreement with Figure 16 .1, as :

A62 =6"- 0' (16 .30)

344 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

SOLID BODY SOLID BODY q

r
(a) (b) GAS
q
GAS
' 9r
i

4:z ni t
a

I SENSOR

4
4>4t>
t a

(a) THICK SENSOR CONDUCTORS (b) THIN SENSOR CONDUCTORS,

PERPENDICULAR TO SURFACE, PARALLEL TO SURFACE, GOOD !
FALSE !

Figure 16.9 Surface temperature measurement of a solid body by a disk thermocouple

This error results from the existence of a thermal contact resistance, Wc, at the interface
between the sensor and the body. It can be regarded as a temperature drop across the
resistance, Wc, under the influence of heat flux, (DT, entering the sensor . Thus, At9 2 is
given by:

At92 =-(DTWc (16 .31)

In the case ofa semi-infinite body, A0 2 is given by

(Wc /W)+(7rRT 2b /F.)Wc

A62 = ot (16 .32)
1 + (1rRT b / F.)(WT + We )

where all symbols are as in equation (16.28) .

It is rather difficult to calculate the thermal contact resistance, Wc, which depends upon
many factors such as the smoothness and cleanliness of the surface, the force with which the
sensor is pressed to the investigated surface, the elasticities of the sensor and the surface
materials, and so on. Tye (1969) and Michalski (1978) show how it can be found
experimentally using techniques similar to that illustrated in Figure 16 .10. Here, the thermal
contact resistance between a copper rod and a steel surface is displayed against the contact
force at different contact temperatures ec . Based on these results Michalski (1978) has
advised the application of a force of about 30 N for copper-plate sensors of diameter 8 mm.
Reduction of the thermal contact resistance is possible by cleaning the contacting surfaces
to remove oxides, fats and other impurities followed by the application of a paste of high
thermal conductivity (Tye, 1969 ; Michalski, 1978).
 4
THEORY OF CONTACT METHOD 345

P
Cu
20 8 .$c
°, 1 BRASS
20
3
w
09 =89°C
a 15
N O
N
w
=110°C
r c
v x
a li
z 10
° 4=136°C
J
a
f
.,y =163°C
5 O O p

0 10 20 30 40 50
CONTACT FORCE , NIm'

Figure 16.10 Thermal contact resistance, Wc , of copper-brass contact versus contact force, P, at
different temperatures

The third partial error, A63, is defined, conforming to Figure 16.1, as

Ot93 = OT - 0" (16 .33)

The temperature distribution along the sensor must be known for it to be calculated .
However, for really small distances between the sensor sensitive point and its front surface,
the heat flux from the side surface of the sensor along the length, P, can be neglected . Thus
the third partial error is given by:

A63 =-9T (16 .34)

T

where AT is the thermal conductivity ofthe sensor material.

To reduce the third partial error, A03, the distance P in Figure 16.1 should be kept as
small as possible . This can be achieved by using thin flat band-thermocouples, thin plates,
of well-conducting materials or thermocouples having non joined, pointed conductors, such
as in Figure 16.14(a), which can only be used on metallic surfaces.
Thermally compensated sensors A proper elimination of all partial errors could be
achieved by applying thermally compensated sensors, which have additional heating by
low power small heating elements (Michalski et al.,1991) . Due to their complicated
construction and measuring circuit, they are no longer produced.
346 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

Dynamic errors: In contact surface temperature measurements, especially when using

hand-held sensors, dynamic errors are also observed . A theory of dynamic errors in contact
temperature measurements has not been properly developed so far (Mackiewicz, 1976b;
Znichenko, 1969). These errors occur when readings are taken before the contact sensor has
reached a thermal steady state after making contact with the investigated surface, . This error
can be eliminated by a sufficiently long contact time before the readings are taken. In
practice, this time has to be below about 1 min, to prevent tiresome working for the
operator . Also, the probability of ensuring the correct sensor position on the surface
decreases with increasing time . It is possible to reduce dynamic errors by applying a peak-
picker device of the type described in Section 12 .2 .3 .

Numerical example
Calculate the method error of contact temperature measurement of a chromium-nickel steel
surface, using a disk thermocouple. It is assumed that the thermal contact resistance, Wc, between
the sensor and investigated surface is null .
Data : original surface temperature, Ot = 120 °C ; ambient temperature, Oa = 20 °C,
body thermal conductivity, fb =10 W/m °C,
heat transfer coefficient at body's surface, ab = = 10 W/m2 °C .
Sensor : type K thermocouple, with copper disk of:
Ip - 1 .5 mm; Rp - 7.5 mm, Ab = 372 W/m °C,
positive conductor: R cl = 1 .5 mm, Ac , = 13 W/m °C,
negative conductor: R c2 = 1 .5 mm, Act = 58 W/m °C,
heat transfer coefficient on disk surface, ap = 10 W/m2 °C,
heat transfer coefficient on side surfaces of conductors, ac = 50 W/m 2 °C .
Solution :
1. Calculation of q T =f(Ar41) conforming to equation (16.23) :

1.5 x10 -3
qT= 10+ 2x50Xl.5X10-3 ( 13+ 58) OT
z
(7 .5 x 10-3

=127 .30T W/mz

Assuming : A02 = 0 (Wc = 0) and A63 = 0 , the method error A O = Af91 .

Consequently : 6T =6t +Ar91 or OT = 13T -Aria = (OT -61)+Ana .
Thus : qT = 127.3[(120 - 20) + A61 ] =12 730 +127 .3A6,

and finally: Az1 = ( qT -12 730) °C

11
273
2. Calculation of A61 = f(qT)
From equation (16.4): B = (10 x 7.5 x 10-3 )/15=5x10-'
From the diagram in Figure 16 .5, for B = 5 x 10 -3 , Fm = 0-9,
and thus, A61 = ( q d x 7.5 x 10 -3 / 15) x 0.9 = 0.45 x 10 -3 qd
THEORY OF CONTACT METHOD 347

From equation (16 .27): qb =10(120-20) = 1000 W/m2

Substituting qb in equation (16.1) yields : qd = (qT - 1000) W/m 2

and then A0 1 =-0.45x10 -3 (qT-1000)=0.45-0.45x10-3 q T
3 . Determination of At91 by a graphical method.
The functions : A61 = (qT -12 730)(1 / 1273) and

A61 =(0.45-0.45 x 10 -3 q T are displayed in one diagram (Figure 16.11) .

The intersection point of both lines determines the error Atg1 = -5 ° C

HEAT FLUX DENSITY q , Wle?

5-103 10-103

u -1 4,9

a -2-
o " 04S .
0,4S.7 .a
),
o
_4
w AA, -_I,,g .C

_6 A'$1 Iq09
T -12730) 1
127,3

Figure 16 .11 Auxiliary diagram for the numerical example

16.2.4 Influence of thermal properties of bodies on errors

The main physical property characterising a body, whose temperature is to be measured, is
its specific thermal conductivity, Ab . With increasing Ab, the disturbing temperature, Ood,
decreases in agreement with equation (16 .5) and thus the first partial error, A t91, also
decreases as in equation (16 .26) . This dependence is not strictly inversely proportional,
because in equation (16 .26) the function, Fm, also depends upon the thermal conductivity
ab , hidden in the constant B, which is illustrated in Figure 16 .5 . For low values of the heat
transfer coefficient, ab , at the surface of the body, and for normally applied dimensions of
the end part of the sensor, the error, 0 01, is a hyperbolically decreasing function of thermal
conductivity, Ab. This explains the difficulties encountered in the use of contact methods for
the temperature measurement of non-metallic bodies whose thermal conductivity is many
times smaller than that of metals . The thermal conductivity, Ab, of a body also influences the
response time of the contact sensor, which is far longer for non-metals . A decisive influence
is exercised by the size of the investigated body, as compared with the sensor dimensions,
upon the precision of the readings .
The disturbing temperature values in a semi-infinite body also decreases in a direction
perpendicular to the surface, the dimension, x, in Figure 16 .3, as along it, the dimension, r,
in Figure 16 .4, as a function of distance from the contact point . At a depth, x, and a
distance, r, about five times greater than the radius, R T , of the contacting area, the
disturbing temperature falls below 10 % of its maximum value . It follows that, any body
348 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

having dimensions x and r more than five times greater than the radius of the contacting
area (x > 5R T ; r > 5R T), can be regarded as a semi-infinite body . For any other bodies, or
characteristic dimensions below those given above, the deformation of the original
temperature field and so the errors, A6 1 , will be far greater .
The surface state of the investigated body influences the thermal contact resistance, W.,
between sensor and body and thus also the precision of the readings . Errors, resulting from
the presence of thermal contact resistance, often exceed all other partial errors .

16.3 Sensors for Surface Temperature Measurement

16.3.1 Portable contact sensors
Portable contact sensors, or probes, pressed to the investigated surface, give good readings,
known as spot readings, when a thermal steady-state is reached . Many such thermometers
have exchangeable measuring tips, adapted to the condition, shape and material of the
surface . All of the sensor tips used should fit the surface in the best possible way.
Besides the method errors, previously described in Section 16 .2, large random errors
may also sometimes occur . They are mostly caused by incorrect positioning of the sensor tip
relative to the surface . The greater the sensor response time the higher is the probability of
occurrence of random errors . To reduce them it is advisable to use sensors of low thermal
inertia or to repeat the measurements several times . A marked improvement is obtained by
the application of a peakpicker device described in Chapter 12, which is essential when
using digital indicators . Without a peak-picker the readings of a digital instrument are a
series of random, erroneous values .
Most surface sensors are thermocouples . For temperature measurement of smooth, flat
surfaces, disk thermocouples of the type in Figure 16 .12, are mostly used . The
thermocouple conductors are soldered into a copper (up to 400'C) or silver (up to 600 °C)
disk . A detailed discussion of its constructional parameters is given in Section 16 .2 .3 . As an
example, a type K thermocouple with a silver disk of diameter dp = 10 mm and thickness
lp = 0 .8 mm, having conductors of dc = 1 mm, gives readings for metal surfaces which are
about 10 to 20 °C low . Its 98 °lo response time to .98 ;:t 10 s. This type is of no use for non-
metals . Application of heat conducting paste improves the precision by causing smaller
A0 2 . A survey of special surface sensors applied in scientific instruments is given by
Gatowski et al. (1989) . A better solution for steel and other low conductivity metals is
ensured by flat, spiral, elastic, as in Figure 16 .13, which are mostly type K, combining thin,
flat measuring junction, to give small A0 3 , with conductors led along isotherms, giving
small A0 1 . They can be used up to about 500 °C, with errors of about -5 to -12 °C for
metallic surfaces and -10 to -30 °C for ceramics . In the case of 0 .8 mm spiral thickness the
to .9g response time is about 15 s for metals and about 2 min for ceramics .
Thermocouples with a non-soldered point tip, similar to that in Figure 16 .14(a), are used
to measure the temperature of clean ., non-ferrous metallic surfaces such as Al or Cu. The
conductors, which are pressed with a force, P, to the surface, form the measuring junction
with the body under measurement so eliminating the third partial error, A03 . Type K and
SENSORS FOR SURFACE TEMPERATURE MEASUREMENT 349

\CERAMIC
INSULATION

CONDUCTORS

II CERAMIC
INSULATION
II II
II III THERMOCOUPLE
WIRES
II II
II II
SOLDER Cu-or Ag-DISK

MEASURING
!UNCTION

Figure 16.12 Disk thermocouple Figure 16.13 Flat spiral thermocouple

(o) CONFIGURATION 100 (b) INDICATED TEMPERATURE

=100-C dc =0, 92mm

98
IP IP
1t____J_____1 4 ='
96 -
d, o1,27mm -_-
X---
o
--

do 'C F- d -3,Omm
~ 94
a ~I
I w X/ TYPE K THERMOCOUPLE
r-
^ft =100-C 92 /

90
0 0,5 1,0 2,0 3,0 4,0
COPPER
CONTACT FORCE P , N

Figure 16.14 Point-contact tip thermocouples

type T thermocouples are used as they are the only types hard enough to remain sharp for
long. Typical dependence ofthe indicated temperature, OT , from a type K thermocouple, as
a function of the contact force, P, with conductor diameter as parameter is shown in
Figure 16 .14(b) . The necessary pressing force is secured by springs in the conductors
mounting. In the case of conductors with d. = 3 mm, the to .9g response time is about 1 s.
For temperature measurement of ferromagnetic surfaces, contact sensors with a magnet,
pressing it to the surface are used (Omega Eng . Inc ., 1999).
For cylindrical surfaces bow-band thermocouples stretched across elastic yokes as in
Figure 16.15, are used. In this way, a perfect contact with the cylindrical surface is ensured .
Due to the tangential position of the band, no deformation of the original temperature field
occurs so that A19 2 = 0 . This thermocouple can be used for metallic and non-metallic
surfaces . For example, when measuring the temperature of a copper tube of 30 mm
diameter, the overall error was about 2 °C at 100 °C and the response time was about 3 s.
350 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

THERMOCOUPLE WIRES
CERAMIC INSULATION
m
II '
II II
ELASTIC YOKE

CL

THERMOCOUPLE BAND
MEASURING JUNCTION

Figure 16 .15 Bow band thermocouple

For flat non-metallic surfaces flat band thermocouples, like that shown in Figure 16 .16,
are used . With a similar construction as for the bow band type, the band with the measuring
junction is pressed to the surface by a non-metallic bar with a coil-spring .
Measuring errors may be negative or positive . They are positive in the case when the
band emissivity is smaller than the surface emissivity .
For temperature measurement of small metallic or non-metallic bodies of flat, concave
or convex surfaces convex-band thermocouples, shown in Figure 16 .17, are used . The
elastic, type E or type K, convex band thermocouple, is flattened, while touching the
investigated surface, thus ensuring good thermal contact . This flattening is limited by a
buffer, to prevent damage . The longitudinal incisions permit the shape of the thermocouple
to adjust to any surface irregularities . Band thickness has a deciding influence on measuring
errors . The different sizes of convex-band thermocouples produced, are typically applied in
the temperature measurement of foundry forms, metal plates, walls, glass-ware, ball-
bearings and so on, up to about 500°C . The smallest sensors are used in temperature
measurements of electronic components . The to .9 response time of such a sensor on metallic
surfaces is 0 .2-0 .5 s and on non-metals is 3-10 s .
Thermistors which are also used for surface contact measurements, are mounted in a
small silver plate to intensify the heat transfer between the thermistor and the investigated
surface . They are used mainly for metallic surfaces up to 250 °C and have short response
times .

THERMOCOUPLE WIRES

CERAMIC INSULATION
li I I ~ CERAMIC
I I ELASTIC YOKE INSULATION

BAR BUFFER THERMOCOUPLE

BAND

[~ o
THERMOCOUPLE BAND
MEASURING JUNCTION O MEASURING
JUNCTION

Figure 16.16 Flat band thermocouple Figure 16.17 Convex-band thermocouple

SENSORS FOR SURFACE TEMPERATURE MEASUREMENT 351

For less precise surface temperature measurement when mercury-in-glass thermometers

are also used, they are mounted directly in special thermometric pockets or holes or in
additional copper or brass metallic stands as in Figure 16.18 . A thermometer placed in a
thermometric hole may give readings which differ from the surface temperature .

THERMOMETER

METALLIC FOIL
OR OIL

(a) IN THERMOMETRIC (b) IN METALLIC

POCKET STAND

Figure 16.18 Mercury-in-glass thermometers for surface temperature measurement

16.3.2 Fixed contact sensors

For stationary, continuous surface temperature measurement the temperature sensors are
either pressed in a hole or glued or soldered to the surface. Their indications are more
precise than those of portable types, especially as there are no thermal contact resistances .
Figure 16.19 presents a typical method of fixing thermocouple conductors in a hole using a
special peening tool. Each conductor is hammered lightly after it is inserted into the hole in
the body through the axial hole of the tool shown in Figure 16.19(c) . Following the law of
the third metal (see Section 3.1 .2) and for a small distance between both conductors, the
existence of the third metal does not change the readings . Leading the conductors along the
metal surface, ensures that all three partial errors are negligibly small . Instead of peening,
the thermocouple is sometimes soldered to metallic surfaces or glued to non-metallic ones
by a refractory cement as generally illustrated in Figure 16.20. A detailed description of the
ways of fixing thermocouples is given by Baker et al. (1953) .
Thin-film thermocouples, similar to that shown in Figure 16.21, are produced using thin-
film technology through vacuum deposition on thin glass or ceramic laminate plate, which is
subsequently glued to the investigated surface. (Moeller, 1963a,b ; Browning and Hemphill,

(a) ARRANGEMENT (b) DETAIL (c) PEENING TOOL

THERMOCOUPLE WIRES WIRES
A SOLDER OR BRAZING INSULATION
A
MATERIAL

,
METAL
; ~76 __ M_

Figure 16.19 Peened thermocouple Figure 16.20 Soldered thermocouple

352 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

LAMINATE
PLATE

MEASURING
JUNCTION

v
THERMCOUPLE LEADS

Figure 16.21 Thin-film thermocouple

1962 ; Kinzie, 1973 ; Bransier, 1975) . A typical application is the case of fixed contact
temperature measurement of electrical machinery . Although this technology can be applied
for all thermocouple materials, best results are obtained with pure metals . It should be taken
into consideration that the thermoelectric characteristics of thermocouples of film thickness
below 0 .25 pm can essentially deviate from those of the standardised ones (Bransier, 1975) .
Film-thermocouples, thin as they are, give very small measurement errors . They also have a
very low thermal inertia, with low time constants of some milliseconds.
The temperature of tubes and cylindrical surfaces is easily measured by thin-wire sensor
or thin-band thermocouples strung by springs as shown in Figure 16 .22 . Thin-wire sensors
are more commonly constructed with the form of Figure 16 .23 using Pt or Ni wire of
diameter 0 .03 to 0 .1 mm . After the wire is wound round a surface insulated tube it is then
fixed by an encapsulation of epoxy or silicon resin .

16.3.3 Performance comparison of different sensors

A survey of the errors resulting from the use of different types of contact sensors is given in
this section and illustrated in Figures 16 .24 for aluminium, Figure 16 .25 for steel surfaces
and Figure 16 .26 for ceramic surfaces . All of the results in this section are average values of
about a dozen readings taken by an experienced operator. In Sections 16 .2 .3 and 16 .2 .4 the
importance of correctly choosing the type of sensor used for each application was stressed .
The measurement results given below confirm these discussions . The measurements were
made by the following types of thermocouples :

Type K, silver disk, as in Figure 16 .12, with dP = 10 mm, lP = 0 .8 mm,

Type K, spiral, as in Figure 16 .13, with a spiral thickness of 0 .8 mm,
Type K, non-soldered, pointed, as in Figure 16 .14, with dc = 3 mm,
Type K, flat band, as in Figure 16 .17, with a band thickness 0 .1 mm .

Measuring errors, applying a commonly used silver disk sensor of diameter dP = 10 mm

and lP = 0 .8 mm, are shown in Figure 16 .27, and their dependence upon the thermal
conductivity of the body is shown in Figure 16 .28 . Another set of measurements, shown in
Table 16 .1, were obtained from temperature measurements of a copper sheet 1 mm thick, in
contact from below with boiling water .
SENSORS FOR SURFACE TEMPERATURE MEASUREMENT 353

MEASURING JUNCTION RESISTANCE WINDING INSULATION

TUBE

TUBE THERMOCOUPLE WIRE

Figure 16 .22 Thin wire thermocouple for Figure 16.23 Thin-wire resistance detector
tubes and cylindrical surfaces wound round a tube

, TEMPERATURE ,A , -C TEMPERATURE %tt °C

0 100 200 300 400 500 0 100 200 300 400 500 600

_10 e _10

a -20
a -30
-20 oO
(Fig, 16 .13)
IFig .16 .14) o -30
w (Fig .16 .121
(Fig . *12)
(Fig 16 .13) ~+

Figure 16 .24 Errors, A&, of the temperature Figure 16.25 Errors, A6, of the temperature
measurement of an aluminium surface vs object measurement of a clean steel surface vs object
temperature, r~t temperature, 6,

TEMPERATURE , °C ,tm (Fig . 16 .16)

10 %$t
AFZER
2 m1~
vo10 ~} 300

` 200
-10
AFTER 2min.
~z -20
(Fig . 16 .13)
a -30
a \ ` 1mi0.
0
-40 \
W
-50 A TER
\ _F Zmln .
-60 ~l (Fig .16 .12)
1min.

Figure 16.26 Errors, A6, of the temperature measurement of a ceramic surface versus object
temperature, 81, after 1 and 2 min .

Table 16 .1 Errors, AO, of the temperature measurement of a copper surface at zgt = 100 °C

Temperature sensor Figure Error (°C)

Thermocouple
Type K, silver disk with dp = 10 mm, lP = 0 .8 mm . 16 .12 -5

Thermocouple
Type K, non-soldered, pointed with d o = 3 mm . 16 .14 -7

Thermistor -7
Mercury-in-glass thermometer in copper stand 16 .18(b) -10
354 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

TEMPERATURE 4, , °C
0 100 200 300 400 500

-10 ALUMINIUM Xy-220W/m-°C

-20
STEEL 9, e ~law/ m'C
dZ -30-
a
_40-
0
-50-
w CERAMICS Xy ~Q2W/m-°C
-60

Figure 16 .27 Errors, AO, of the temperature measurement of the surface of different materials, by a
disk thermocouple, vs object temperature, t9t

THERMAL CONDUCTIVITY 'X y ,Wlm'C

0,1 0,2 0,5 1 2 5 10 20 50 100 200 500 1000
0 _ _
AI,Co
-10 CERAMICS -~ STEEL
k' -20-
_30- Ag
,?t = 250 ° C 0,8 mm
0
10mm
W -50-
-60

Figure 16 .28 Errors, Ad, of the temperature measurement of a semi-infinite body by a disk
thermocouple vs body thermal conductivity, 1b , at object temperature, ~1 = 250 °C

16 .4 Quasi-Contact Method

The quasi-contact method comprises the properties of both the pyrometric and the contact
methods . A cup is the main part of a quasi-contact sensor as shown in Figure 16 .29 . The
inside of the cup is a polished gold mirror, which is placed at a distance smaller than 1 mm
from the investigated surface . Thermal radiation emitted from the surface irradiates a
thermopile radiation detector through a fluorite window. Although oxidised metal surfaces
have an emissivity factor well below unity, the presence of the hemispherical golden mirror
means that the whole system approaches black-body conditions . In this manner, the readings
which are emissivity independent, are not influenced by any outside radiation due to the
presence of the protecting hemisphere .
One of the drawbacks of the semi-contact method is the changed heat-transfer conditions
from the estimated surface to the environment after applying the mirror . To eliminate this
influence, the mirror is sometimes made with a reflectivity which is a bit lower, or readings
are taken during a short time of application of the sphere (Drury et al ., 1951 ; Burton, 1953) .
A measuring device by Land Infrared Ltd operates in the temperature ranges from 200
to 900 °C, 500 to 1300 °C and 100 to 400 °C . The sphere diameter is 50 mm and the
EXTRAPOLATION METHOD 355

CONDUCTORS

THERMOPILE-_ FLUORITE
WINDOW
CUP MIRROR
F

INVESTIGATED
SURFACE I

Figure 16.29 Simplified diagram of a sensor for surface temperature measurement by the "quasi-
contact" method

response time ofthe device is about 5 to 6 s. The overall error for oxidised metals and non-
metals is about ±5 °C This instrument, which can also be used for concave surfaces with a
radius of curvature over 50 mm, is typically applied for short-time temperature
measurements of ingots, cylinders, foundry forms, furnace surfaces and so on. The same
instrument can be used to measure surface emissivities . For this purpose, the mirror is
temporarily covered by a black insert and comparative readings are taken . From the
readings with and without the black insert, a set of diagrams permits the surface emissivity
to be found.
A further development of the idea, which is proposed by Land Infrared Ltd . and called
the Emissivity Enhancer, is designed for joint use with photoelectric pyrometers of the types
described in Section 11 .2.
Convective methods also belong to the class of semi-contact methods . The forced
convection intensifies the heat transfer between the surface and the sensor . Although
convectioe compensating methods help in attaining more precise readings, they need
additional heating of the sensor . In both these methods, the sensor itself does not touch the
investigated surface. These two methods are mainly applied in the temperature measurement
of moving bodies . They will be described in Section 19 .6 .

16.5 Extrapolation Method

The extrapolation method, which is one of the most precise ways to determine the surface
temperature of solids, is based upon the principles shown in Figure 16.30. Usually, thin bare
or MI thermocouples are placed along the isotherms inside a solid body, although only MI
thermocouples with insulated measuring junctions are used in metals and semiconductors .
Based on the indications of particular thermocouples, the surface temperature, Ot , can
be determined, without any deformation of the original surface temperature distribution.
Because of the necessity of introducing n thermocouples and drilling the holes, Wilcox and
Rohsenow (1970) assert that the extrapolation method can only be used in some chosen
cases, most commonly in some specialised measuring instruments . This is also the case in
devices for the standardisation of the contact sensor, or in devices for measuring the thermal
contact resistance, as pointed out by Michalski (1978) . Whereas the simple graphical
extrapolation of Figure 16.30(b) is applied in sporadic measurements, microcomputers are
used in continuous measurements. Michalski and Borowik (1985) show that the most
356 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

THERMOCOUPLES
T1 T2 T3 T4 T5
fat

` ARRANGEMENT
OF SENSORS
t

GAS

(b)
I'fII
i
A's
'fc GRAPHICAL
.9.t EXTRAPOLATION

EXTRAPOLATED
^fn VALUE -4t

Figure 16 .30 Extrapolation method for surface temperature measurement : (a) sensor arrangement and
(b) graphical extrapolation

suitable method then uses polynomial extrapolation following the algorithm of Neville and
Aitken (Stoer, 1972) . In most technical problems the application of thermocouples is
normally adequate .
In determining the temperature of rather thick bodies, which may be regarded as semi-
infinite bodies, the two-sensor method of Yarishev and Minin (1969) is sufficiently precise.
In this method, based on the indications 6 1 and 732 of two sensors, at the respective
distances x 1 and x2 from the investigated surface, the surface temperature is calculated from
the relation :

3
dt (t) = 61(t)C - x2x2 3 (16 .35)
X2 -XI
- XI xl 01(t)
With a small range of temperature variations and small temperature differences within a
solid body, such as plates of finite thickness, the thermal conductivity, Xb, can be assumed
constant so that the temperature drop in the body is also nearly linear. In this case, the
simplest possible linear extrapolation, based on two measured values, has the form :

01(1)x2 -02(1)x1
Ot(t)= (16 .36)
x2 - x1

Extrapolation methods which are used in principle for steady-state surface temperature
determination, can only rarely be used for transients . Emelyanenko (1960) and Hollander
(1959) illustrate their use to determine temperature transients of the edges of cutting tools .
Thomas (1975) points out that the main error sources of extrapolation methods are due
to thermal non-homogeneity and anisotropy of the solid body material and the deformation
MEASUREMENT OF INTERNAL TEMPERATURE OF SOLID BODIES 357

of the existing temperature field by the introduction of sensors inside the solid body . To
avoid this last source of errors it is advisable to use thermocouples, which are as thin as
possible, introduced along the isotherms .

16 .6 Measurement of Internal Temperature of Solid

Bodies

Similarly as in the measurement of surface temperature the sensor which is introduced

inside a solid body, deforms the original temperature field . It is assumed, as in normal
practice, that the measurement is taken with a thermocouple, which for further analysis is
replaced by an equivalent homogeneous cylinder.
Figure 16 .31 (a) presents the isotherms in a solid body in contact with a gaseous medium .
To measure temperature at a depth l from the surface, a sensor, T, in Figure 16 .31(b), is
inserted through a hole . The sensitive point S of the sensor is at a distance, h , from its end.

SOLID BODY GAS SOLID BODY GAS

Ia) I (b)
q q
0

T-SENSOR

nA ~A
i 4> ° s~2nf~ ,g. >,g
(c) (d) I t'
I i~
I
4%~ SENSOROPINTITI~E
)

'S'
tt
4, I 1 I

4T

,ya

x x
Figure 16 .31 Internal temperature measurement in a solid body by a rod sensor, T. The isotherms and
heat flux density, g, are shown before the sensor is inserted in (a) and after the sensor is inserted in
(b) . The temperature distribution before and after is shown in (c) and (d) respectively . The original
true object temperature at the depth, l, from the surface is Ot, the temperature at the bottom of the
hole after inserting the sensor is 0' and the temperature at the end of the sensor is 0 "
358 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

Figure 16 .31 (b) also shows the isotherms and heat-flux lines after inserting the sensor, T,
assuming that its thermal conductivity is ?,r, which is greater than that of the body, fb .
The overall measurement error is composed of three partial errors :

1. A6 l = 01 - 01 is the error resulting from the deformation of the original

temperature field,
6"
2. A 19 2 = - 0' is the error resulting from the temperature drop across the thermal
contact resistance, and across the air gap, together,
3. A03 = OT - 6" is the error resulting from the temperature drop across the distance 1 1
between the end of sensor and its sensitive point .

The following comments are appropriate for the above partial errors :

1. To reduce A61, thin thermocouple conductors made of materials of low thermal

conductivity should be used. The sensor should be inserted as deep as possible into the
solid body . A t91 errors increase with increasing difference in the thermal conductivities
of the solid body and the thermocouple materials . This is shown in Figure 16 .32 which
gives the A191 errors versus the thermal conductivity of the investigated body of 'ib for
four different insertion depths . Bare type T thermocouples in varnish insulation were

-60
d, =0,6mm
#2000
d, =0,6mm

-50 TYPE T THERMOCOUPLE d=2 . .

~, -40
0

SA ~At 120-C
T
-30 A4-A41 _ - 4,
4T
h«1, A-$Z CO, At=0
w

-2 0 $120 0

-10
~r x1200C, l=3cm

e 1200C,1=7cm -31,120-C, 1=5cm

t
0
0,03 0,05 0,1 0,2 0,3 0,5 1 2 3 4 6 8 10 20 30 4050
CERAMICS STEEL
THERMAL CONDUCTIVITY 1b ,Wlm''C

Figure 16 .32 Errors of internal temperature measurement in a solid body by a bare type T sensor
versus thermal conductivity of the body, Xb ' for different insertion depths
MEASUREMENT OF INTERNAL TEMPERATURE OF SOLID BODIES 359

used . The measured temperature was Ot = 120 °C each time . At91 errors are negligibly
small when measuring metal temperatures (X'b > 40 W/m °C) at insertion depth
1 > 3 cm. According to Hunsinger (1966) it is advisable to ensure 1/dT > 5 for metals
and 1/dT > 15 for non-metals .
2. To reduce A1) 2 , the shape of the end of the sensor should match the shape of the
bottom of the hole with the sensor pressed fully to the bottom .
3. Reduction of A0 3 can be achieved by choosing the appropriate sensor design .

Although all three partial errors can be avoided by inserting the sensor along isotherms,
as shown in Figure 16 .33, this is not always easy to realise for technical reasons . However,
skewed insertion of the sensor frequently helps to reduce the errors . Only thin MI
thermocouples and preferably, especially in non-metals, those with non-insulated measuring
junction are presently used for internal temperature measurements . More details concerning
the measurement of the internal temperature of solids can be found in Baker et al. (1953)
and Chyu and Bergles (1989),
Determination of the internal temperature of solids, without introducing a sensor inside
the body is possible by using ultrasonic thermometers described in Section 7 .2 or by
models .

PLATE CYL(NDER
(a) _%(b) ,g.
a b-2 a b-1
a-1 a-2

qJ' 
I' I'll

~q
~~q
q~~~q
q q

Figure 16.33 Internal temperature measurement in a solid body . a - 1 and b - 1, sensor normal to
surface ; a - 2 and b - 2 sensor along isotherms

16.7 References
Baker, H .D ., Ryder, B .A . and Baker, N .H . (1953) Temperature Measurement in Engineering, 1, John
Wiley and Sons, New York .
Bransier, J . (1975) Les temperature superficielles . Quelques methods de m6sure par contact direct .
Mesures, 40(10), 54-61 .
Browning, W .E . and Hemphill, R.L . (1962) Thermocouples for measurement of the surface
temperature of nuclear fuel elements . Temperature: Its Measurement and Control in Science and
Industry, 3(2), Reinhold Publ . Co ., New York, .723-733 .
Burton, E .J . (1953) Recent advances in radiation and immersion pyrometry . Instruments, 26(10),
1524-1525 .
Chyu, M.C . and Bergles, A.E . (1989) Locating method for temperature sensing elements in solid
bodies . Experimental Thermal and Fluid Science, 2(2), 247-249.
360 TEMPERATURE MEASUREMENT OF SOLID BODIES BY CONTACT METHOD

Drury, M.D ., Perry, K.P . and Land, T. (1951) Pyrometers for surface measurements . J. Iron Steel
Inst., No . 11, 245-250.
Emelyanenko, W.O . (1960) Surface temperature measurement of solids by heated thermocouples .
Inz.-Fiz. Journal, 3(10), 54-56 (in Russian) .
Gatowski, J.A ., Smith, M.K . and Alkidas, A.C . (1989) An experimental investigation of surface
thermometry and heat flux . Experimental Thermal and Fluid Science, 2(3), 280-292.
Hollander, M. (1959) An experimental measurement of the temperature distribution in the workpiece
during metal-cutting . Ph .D . thesis, Columbia University, New York .
Hunsinger, W. (1966) Temperaturmessung. Handbuch der Physik. Springer-Verlag, Berlin .
Kinzie, P.A . (1973) Thermocouple Temperature Measurement . John Wiley and Sons, New York .
Kulakov, M .W . and Makarov, B .I . (1969) Surface Temperature Measurement of Solid Bodies. -
Energiya, Moscow (in Russian) .
Mackiewicz, E. (1976a) Dynamic compensating method of surface temperature measurement of
solids . Archiwum Budowy Maszyn, 23(3), 443-451 (in Polish).
Mackiewicz, E. (1976b) Static compensating method of surface temperature measurement of solids .
Archiwum Budowy Maszyn, 23(5), 549-557 (in Polish).
Michalski, L. (1978) Experimental determination of thermal contact resistance of metallic surfaces .
Archiwum Termodynamiki i Spalania, 9(1), 109-122.
Michalski, L. and Borowik, L. (1985) Continuous determination of the temperature at the inaccessible
parts of electroheat plants . Elektrow&rme International, 43, B4, 189-193.
Michalski, L., Eckersdorf, K. and McGhee, J., (1991) Temperature Measurement (1st Edition), John
Wiley and Sons Ltd., Chichester .
Moeller, C.E. (1963a) Special surface thermocouples . Instr. Contr. Syst., 36(5), 97-98 .
Moeller, C .E . (1963b) Thermocouples for the measurement of transient surface temperatures .
Temperature: Its Measurement and Control in Science and Industry, 3(2), Reinhold Publ . Co .,
New York, 617-623.
Omega Engineering Inc . (1999) The Temperature Handbook, Omega Engineering Incorporation,
USA.
Roeser, W.F . and Mueller, E.F . (1930) Measurement of surface temperatures . J. Res. Nat. Bur.
Stand., 9(4), 793-802.
Stoer, J. (1972) Einfuhrung in die numerische Mathematik., Springer-Verlag, Berlin, Heidelberg.
Thomas, R.T . (1975) Extrapolation errors in thermal contact resistance measurements. J. Heat
Transfer, 97(5), 305-307.
Tye, R.P. (1969) Thermal Conductivity, Academic Press, London .
Wilcox, S.J . and Rohsenow, W.M . (1970) Film condensation of potassium using copper block for
precise wall temperature measurement . J. Heat Transfer, 92(8), 359-371 .
Yarishev, N.A . and Minin, O.W . (1969) Extrapolation method of measurement of temperature and
heat-flux. Prihorostroyenye, 8(1), 19-24 (in Russian).
Znichenko, W.M . (1963) Dynamic characteristics of thermocouples in surface temperature
measurement. Prihorostroyenye, 2(9), 112-115 (in Russian) .
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

17
Temperature Measurement of
Fluids
17 .1 Low Velocity Gas
17.1.1 Contact sensors

Assume that a gas, with a temperature, T., flows in a tube with a flow-velocity below about
20 m/s, as shown in Figure 17.1(a) . In the tube there is a sheathed sensor, whose sensitive
part, at temperature, TT , is placed in the sheath bottom. This case corresponds to a
thermocouple with the measuring junction welded into the sheath bottom . Also let the
temperature ofthe internal surface ofthe tube wall be TW , while the length to diameter ratio
of the sensor sheath is assumed to be so large that heat transfer through the sheath bottom
can be neglected . It is further assumed that the gas temperature, Tg , is higher than the
temperature of the tube wall, T, Consider a cylindrical sheath element of length dx, placed
at a distance, x, from the sheath end as represented in Figure 17.1(b) . In the thermal steady-

EJ"
state it is apparent that the heat balance must take account of all conduction, convection and
radiation effects .

(a) ARRANGEMENT (b) SHEATH

ELEMENT
D
T9 > TS - " -- TUBE
d

T= '` T'
v 0~ +d0,

GAS
T9 d Op c d0~

-~ ' `SENSOR
TT SHEATH

HEAT FLUX : 0k - CONVECTION

O r - RADIATION
0, - CONDUCTION

Figure 17.1 Gas temperature measurement in a tube by a sheathed sensor

362 TEMPERATURE MEASUREMENT OF FLUIDS

The convection heatflux from the gas to an element of length dx is :

d'Dk = ak'rDdx(Tg - Tx ) (17 .1)

where ak is the convective heat transfer coefficient, Tg is the gas temperature, and Tx is
the sheath temperature at a distance x from its end .
Assuming that the gas transmissivity equals unity, the heat flux dissipated from this
element by radiation to the tube wall is given by:

d(')r=ETCo7SDdx (17 .2)

(100)4-( 00)4

where ET is the emissivity of the sheath surface, Co is the radiation constant of a black
body and T,, is the temperature of the tube wall and the other symbols are as in
equation (17 .1) . Equation (17 .2) is valid under the assumption that the sheath surface is
many times smaller than that of the tube as considered in equation (8.21a), while
equation (17 .1) is valid for ak =constant.
The conductive heatflux along the sensor is described by the relation:

dam, =-.1,A d2T dx (17 .3)

where A is the thermal conductivity ofthe sheath material,

A = )rD 2 nd 2
4 4

and Tx is the temperature at a distance x from the sheath end .

A heat balance equation of element dx is then:

(17.4)
d(Dk = d0c + d(D r

Substituting the corresponding values from (17 .1), (17 .2) and (17 .3) into (17.4) yields :

)4 )4
d2Tx
~1A+ak~cD(Tg-Tx)-ETC070 Tx ( Tw (17 .5)
_(
dx2 100 100 =d

This nonlinear differential equation, which cannot be solved by direct integration, can be
solved in some cases by a graphical method or by introducing some application specific
simplifications . In most cases, similarly to the convective heat transfer coefficient, ak , an
LOW VELOCITY GAS 363

equivalent heat transfer coefficient by radiation, ar , with a constant value is introduced .

Equation (17 .5) then becomes :

~2 AA +ak7rD(Tg - TX ) - ar ;rD(T,, - Tw)=0 (17 .6)

(ak + ar )7rD
Making the substitution n = this equation has the solution :
AA

Tg -Tw _ ak +a r cosh nl
(17 .7)
cosh
TX - Tw ak nl-1

Considering that the sensitive point of the sensor is mostly located at its end, where
x = 0, equation (17 .7) is simplified to:

Tg - Tw _ ak + ar cosh nl
(17 .8)
cosh
aTT -Tw k nl -cosh nx

or after some transformations :

_ _ _ ak (cosh nl -1)
ATc,r = TT Tg - (Tw Tg )Cl - (ak (17 .9)
+ar )coshnl

where AT,,,r is the overall error resulting from heat conduction along the sensor and from
radiant heat exchange .
Ifradiant heat transfer can be neglected, so that ar = 0 , equation (17 .9) becomes:

T
OTC =TT -Tg = w-Tg
(17.10)
cosh ml

where ATE is the error resulting from heat conduction along the sensor and

a k~D
m=
N AA

If the sensor is sufficiently long so that any error resulting from heat conduction along
the sensor could be neglected, or if the product, .l.A, is sufficiently small and substituting
Tx = TT , equation (17 .5) is simplified to:

ak'rD(Tg -TT)-ETC,,7D 0 (17 .11)

L- 4
[( 100)-(00)4
364 TEMPERATURE MEASUREMENT OF FLUIDS

From equation (17.11), the measuring error, AT, resulting from radiant heat transfer, is
expressed by:

4 4
ETC°
AT,r = TT - Tg =
ak T"' TT (17 .12)
(100) - (100)

The simplest way of solving this equation is a graphical one, as shown in Figure 17.2,
where the curves q k =fl(TT) and qr = f2(TT) are drawn. The convective heat flux density
from the gas to the sensor surface is q k = ak(Tg - TT) while the radiant heat flux density, q r,
from the sensor surface to the tube wall is:

00)4
-~
qr = ETCo
0J4

In the thermal steady-state both these quantities are equal so that:

qk = 9r (17.13)

The point where both curves intersect in Figure 17.2, indicates the sensor temperature,
TT , and the radiant measuring error, ATr . Equations (17 .9), (17 .10) and (17 .12) enable
calculation of the measuring errors and the introduction of any necessary corrections in
particular cases. This latter possibility is rarely used. Mostly based on these equations,
conclusions may be drawn for the design of sensors, so that indication errors can be kept as
small as possible.
Sometimes more complex models of a temperature sensor in a tube, through which gas
flows, are also applied . In these models described by Blumroder (1981), Haas (1969) and
Rudolphi (1969), the heat exchange between the part of the sensor sheath and sensor head,
protruding from the wall into the environment, is considered.

TUBE WALL TEMPERATURE T =const

v
GAS TEMPERATURE T =const 9
r
r

o
l

aw
Y
T9
0 T Ti ATr
TEMPERATURE

Figure 17.2 Graphical method of determining the error, ATr, of equation (17 .2)
LOW VELOCITY GAS 365

Error analysis of gas temperature measurement permits the practical establishment of the
dependence of errors upon the sensor insertion depth into the medium or to establish the
dependence of errors upon the values of the heat transfer coefficients between the sensor
sheath and the gas .

17.1.2 Methods of reducing errors in contact measurements

From the preceding section, it follows that reduction ofthe measuring errors requires:

" increased heat flux, (Dk, gained by convection,

" decreased heat flux, (DC, lost by conduction,
" decreased heat flux, (D,., lost by radiation .

Increase of heat flux, Ok, by convection can be achieved by:

1 . Increase of the surface of the convective heat transfer by using afinned sensor as shown
in Figure 17 .3 . This means that the fins should be made of a material with a high thermal
conductivity and low emissivity, to reduce the radiant heat losses .
2 . Increase of the convective heat transfer coefficient by applying high gas velocity in the
tube using a sensor sheath with as small a diameter as possible and by placing the sensor
at an angle of about n/2 relative to the direction of the gas flow.
3 . Applying suction thermometers, in which the gas velocity is increased only in the
surrounding of the sensor . This method is especially appropriate in those cases where an
increase ofthe gas velocity in the tube is not possible . A suction thermometer, also called
aspiration thermometer, was first proposed by R. Assmann (1892) . Its operating
principle, shown in Figure 17.4 (Wenzel and Schulze, 1926), has the end of a suction
tube with a thermocouple inserted into a pipe-line or gas-filled enclosure, in which the
gas temperature is to be measured. The compressed air produces suction of the gas in the
nozzle . In this way, the gas, whose temperature is to be measured, streams past the
thermocouple at high velocity. Thermocouple readings, which are a function of gas
velocity, have the dependence on rate of gas flow shown in Figure 17.5 for a gas
temperature of 100 °C . These thermometer readings are practically independent of the
position of the measuring junction in the suction tube. Industrial suction thermometers
are often equipped with one or a number of concentric radiation shields to reduce the
radiant heat exchange . Suction thermometers, which can be used up to about 1600 °C,
are often water or air cooled, using MI thermocouples as temperature sensors. More

TRANSVERSE FINS LONGITUDINAL FINS

Figure 17.3 Finned sensors

366 TEMPERATURE MEASUREMENT OF FLUIDS

t
COMPRESSED AIR

MEASURING 1'
"1
"

NOZZLE
OV
. 1

v
1

THERMOCOUPLE
JUNCTION

1
1 1 0 1 40 50
Ei
SUCKED GAS
Z RATE 1 1

Figure 17 .4 Suction thermometer by Wenzel Figure 17.5 Thermocouple readings of the

and Schulze (1926) thermometer from Figure 17 .4, versus gas flow
rate at 'gg = 100 -C

detailed information " " found " publications " "4

"" " 41) and Ribaud et al . (1959) . It is also important to note that they can only be
used when the quantity of gas sucked by the thermometer is negligibly small compared
with the rate of gas flow in the pipe-line itself.

Decrease I q)C9 1 conduction 1 ' obtained 1

Application " / I cross-sectional - . " " from materials with

low thermal conductivity . Figure 17 .6 presents some ways of installing rather long
sensors in pipe-lines . It is advisable that the ratio of the sensor length, 1, to its diameter,
I should " " I 2: 6 to 10 in flowing gas or Ild 2! 12 to 15 in still gas.

2. Laying the sensor from the measuring point, along isotherms, as shown in
Figure 17 .6(a). Similar results can be obtained in arrangements like those in
Figures

(a) IN ELBOW (b) OBLIQUELY (c) ALONG PIPELINE AXLE

IbI (CI
I

Figure 17 .6 Methods of installing temperature sensors in a pipe-line

LOW VELOCITY GAS 367

3. Thermally insulating or heating of the sensor head to increase its temperature in the
manner shown in Figure 17.7.

Decrease of heat flux, (Dr, lost by radiation, can be obtained by:

1 . Covering the sensor surface with materials such as gold, silver or platinum which have a
low emissivity, ET .
2 . Application of radiation shields . This is the most popular method of reducing radiant
heat exchange between sensor and surrounding walls . A shield of low emissivity, E,, is
placed between the sensor and tube wall as shown in Figure 17.8. Assuming that the
shield is sufficiently long, to be able to neglect the influence of its open ends on the
sensor heat balance, and that the shield inner surface is many times larger than that ofthe
sensor, the measurement error due to radiant heat exchange can be calculated from
equation (17 .12). In that relation the shield temperature TS is substituted instead of the
wall temperature, TW . The relevant error, ATr,s , is given by:

.
ETC
AT"s = TT -T9 = (17 .14)
k (lOiO)4 -(100)4

The shield temperature, Ts, is determined from the shield heat balance :

O k,s = (Dr,s (17 .15)

where (Dk,s is the heat flux gained by convection and (Dr,s is the heat flux lost by radiation .

Heat flux 4)k,, is given by:

(Dk,s =270s ls ak's (Tg -TS ) (17 .16)

where ak,s is the convectioe heat transfer coefficient of the shield and the other symbols
are as in Figure 17.8 .

(a) INSULATED SENSOR HEAD (b) HEATED SENSOR HEAD

INSULATION ELECTRIC HEATING

ELEMENT

T Ts s

Figure 17.7 Methods of reducing heat-flux conducted along the sensor

368 TEMPERATURE MEASUREMENT OF FLUIDS

PIPELINE SENSOR

IT Tw
E

EE 

0 0
Ty
Tr s Es SHIELD

Figure 17.8 Sensor with a radiation shield

In equation (17 .16) double, internal and external shield surfaces, have been considered .
The heat flux, (Dr,S, is described by:

cl?r,s _9GDsls"Co (17.17)

[ (100)4-(100)4 1

where es is the shield surface emissivity, and the other symbols are as in Figure 17.8.
Equation (17 .17), which is based on equation (8.21 a), is valid under the assumption that
the external shield surface is many times smaller than that of the surrounding surface of
the tube wall.
After some simplifications, the heat balance equation in steady-state, becomes :

2aks(Tg-Ts)=EsCO (17.18)
(100)4- ( 00)4

The value of shield temperature, Ts, can be found graphically as in Figure 17.2 .
Assuming that Tg > Tw the shield temperature, Ts, is always higher than that of the wall,
TW. Comparing equations (17 .12) and (17 .14) it is clearly seen that error due to the
radiant heat exchange between the sensor and its surroundings decreases due to the
application of the shield . Analogous reasoning can be made for two, three and more
shields . Each consecutive shield decreases the error resulting from the radiant heat
exchange with a progressively smaller and smaller influence. According to King (1943)
this error, OTr, ns , for n shields is given by the approximate relation :

ATim = iT, (17.19)

A more detailed analysis of the influence of shields on the readings of a gas measuring
thermometer is presented by Moffat (1952) . The distances between the shields should be
large enough to enable a free gas flow.
3 . Application of heated radiation shields (Mullikin, 1941) . In this method a thermocouple
is placed on the axis of a shield which is heated by an additional low-power heating
LOW VELOCITY GAS 369

element . A second thermocouple, which measures the shield temperature, allows

adjustment of the heating power until the readings of both thermocouples are the same.
When this state is reached, the thermocouple which measures the gas temperature, does
not exchange any energy with the surroundings by radiation so that its readings are
correct . It is a rather time consuming method unless it is automated.

The simultaneous fulfilment of all three conditionsfor reducing measuring errors may be
realised with the use of a bare thermocouple of very small wire diameter. The diameter of
the thermocouple wire and its surface area are prime influences which determine radiation
heat exchange . As a thin gas film always exists around a wire, the convective heat exchange
depends upon the wire diameter plus double film thickness . With wire diameter approaching
zero the convective heat exchange is mainly determined by the double film thickness while
the radiant heat exchange disappears. Thus, as only convective heat exchange remains, any
radiant measuring errors, AT, disappear . At the same time, with wire diameter approaching
zero any conductive measuring errors, AT, also disappear . In practice it is advisable to use
bare or MI thermocouples which are as thin as possible. The lower diameter limit is
imposed by the mechanical strength and the corrosion resistance of the wires . This method
is suitable for both laboratory and industrial applications.
The extrapolation method is another very precise method of gas temperature
measurement . Gas temperature is simultaneously measured by a number of bare or MI
thermocouples of different diameters . The results, which are displayed graphically as a
function of the thermocouple diameters are extrapolated to the zero diameter. This value is
then the true gas temperature . As an example, Figure 17.9 presents results obtained while
using bare Type K thermocouples to measure the temperature of hot air with velocity 8 m/s
flowing through a tube having a wall temperature of 15 °C. The extrapolated value was
175 °C. Before the measurements it is very important to check the identity of the
thermoelectric characteristics ofthe thermocouples used. MI thermocouples, as described in
Section 3.3 .3, which are produced with a diameter as thin as 0.2 mm, are very convenient in
the extrapolation method .

-GAS TEMPERATURE
EXTRAPOLATED VALUE
190
TYPE K THERMOCOUPLE
0
170

4T

150

~w --15°C
a
E 130 -
w
r

110 ' -
n r r ~ r

0 0,5 1 1,5 2 2,5 3

WIRE DIAMETER d , mm

Figure 17.9 Temperature indicated, 9T, by bare thermocouples in flowing air of velocity, v = 1 .8 m/s,
versus wire diameter, d
370 TEMPERATURE MEASUREMENT OF FLUIDS

An important source of errors occurs in the temperature measurement of flowing gas

when a non-uniform distribution of gas temperature occurs across the tube section . To get
readings approaching the average gas temperature a number of sensors is used as given in
Figure 17.10. If thermocouples are used, they have to be connected in series . The measured
thermal emf must then be converted into a temperature value . A correct average temperature
can be obtained directly, using parallel connected thermocouples, provided all of the
thermocouples used have precisely the same resistance. A merit of the parallel connection is
that correct readings are obtained even in the case of a broken circuit in one of the
measuring loops.
For gas temperature measurement in the temperature range of from about 1000 to
3000 °C, only thermocouples of the metal group, Pt, Rh and Ir, whose properties have been
described in Section 3.4, are used. However, besides the changes in their characteristics,
described before, these thermocouples can also act as catalysts in oxidisation and
combustion processes . These phenomena result in additional heating of the thermocouples
causing important additional errors. A detailed analysis of these phenomena, which can also
occur in any other temperature sensors with protecting shields made of the previously
mentioned metals, is presented in Ash and Grossmann (1972) and Thomas and Freeze
(1972) .
Many references and methods for the experimental detection of these phenomena are
also given by these authors . As an example, in measuring the temperature of not completely
burned exhaust gases of internal combustion engines or of gas turbines at about 1000 to
1500 °C, the measuring error can be as high as about 400 °C (Thomas and Freeze, 1972) .
To prevent additional catalytic heating, thermocouples or their protective sheaths as pointed
out by Kinzie (1973), have to be covered by a layer of BeO, ZrO, Si203, Cr203 or other
appropriate materials.
Detailed information on methods of gas temperature measurement and many references
can be found in papers by Baas and Mai (1972), Benson and Brundrett (1962), Moffat
(1952), Mullikin (1941), Mullikin and Osborn (1941) and Torkelsson (1980) .

T2
9

T1 T3

T1 , T2, T3 -THERMOCOUPLES + -

TO MEASURING
INSTRUMENT

Figure 17.10 Series thermocouple connection for measuring the average temperature of a flowing gas
LOW VELOCITY GAS 371

17.1.3 Indirect pyrometric measurements

Gas Temperature measurement by the contact method is sometimes very difficult, especially
when no measuring lag is acceptable, as occurs, for example, in closed loop temperature
control. In addition, at high temperatures and in corrosive atmospheres, when the sensors
have to be equipped with heavy protective sheaths, indirect pyrometric measurements are
used. A photoelectric or total radiation pyrometer could be directed at the inner ceramic
lining of the channel, although the temperature drop near the surface or the thermal inertia
of the lining could result in measuring errors . Therefore, as this method is excluded when
the walls are cooled, it is advisable to place in the channel, a special ceramic element at
which the pyrometer would be directed as shown in Figure 17.11 . This element should be
hollow inside to increase the surface area of convective heat transfer and to reduce the
amount of heat conducted to the channel walls. Hence such an element should be made of a
material with a high surface emissivity and a low thermal conductivity . The part of the
element at which the pyrometer is directed must be placed in a zone of highest gas
temperature and biggest gas velocity, which usually occurs as far away from the wall as
possible . The application of a closed-end ceramic tube, placed in flowing gas, with
pyrometer directed at the tube bottom from inside, is based on the same principle .
Sufficiently long tubes can be regarded as black bodies .

17.1.4 Direct pyrometric measurements

Direct pyrometric measurement of gas temperature is possible using a pyrometer of
effective wavelength corresponding to the absorption wavelength range of the gas, whose
layer thickness has to be large enough to eliminate any background influence . As an
interesting example of the necessary precision of choosing the effective wavelength is the
application of a pyrometer operating at the wavelength, Xe = 4.6 pin . Due to dependence of
the absorptivity of COZ upon temperature, this pyrometer can be used to measure the
temperature of a hot COZ layer through a layer of cool COZ. In the chosen wavelength range
this absorptivity at 1800 K is about 1000 times larger than at 300 K. Similar problems also
concern the temperature measurements of flames .

PYROMETER
r

CHANNEL
__ w4All
CERAMIC
ELEMENT

A
Figure 17.11 Indirect pyrometric measurement of a flowing gas with a radial temperature
distribution, 9(r)
372 TEMPERATURE MEASUREMENT OF FLUIDS

Some producers offer pyrometers with a spectral operating range, designed especially
for temperature measurement of some specified gases . For example Heitronics (1999)
advises the application ofthe following wavelengths :

CO2 above 300 °C - Xe = 4.26 ±0.13 pm

CO2 and CO above 300 °C - Xe = 4.5 f 0 .1 gm
CO above 300 °C - ~e = 4.66 ± 0.1 um
NO2 above 300 °C - Xe = 5 .3 f 0.1 Etm.

As such measurements are rather complicated they are outside the scope of this book.
Interested readers will find further details in Baker et al. (1953) and Green (1987) .

17.2 High Velocity Gas

New problems arise in measuring the temperature ofgases flowing at velocities over 20 m/s .
A sensor placed in a high-velocity gas stream causes the gas movement velocities to slow
down resulting in the sensor being heated to a temperature, Otot, which is higher than the
gas temperature, Og . At high speed, the kinetic energy of the gas is high. During adiabatic
expansion, when the gas velocity reduces to zero, this kinetic energy is transformed into
heat. According to the first principle of thermodynamics, when the total kinetic energy of
the gas is transformed into heat in ideal conditions, an energy balance gives:

mv2 =
MCP(Vtot - z9g) (17 .20)
2

or after some transformations :

v2
Otot - Og = (17 .21)
2c P

where m is the mass ofthe gas, v is the velocity of the gas, cP is the specific heat of the gas
at constant pressure, 09 is the temperature of gas flowing at a velocity, v, and Otot is the
gas temperature at the adiabatic point where the velocity is reduced to zero.
Figure 17.12 presents the difference (Otot - Og ) versus flow velocity v in which 0,()t is
called the total gas temperature and 6 9 the true or static gas temperature . The
temperature, z9g , can also be defined as the gas temperature, which would be indicated by a
minute, point-like thermometer moving along in the same direction as the gas flow and with
the same velocity .
HIGH VELOCITY GAS 373

80

o 70

60
~a

~` 50
W
Z
w 40--
w
0
w
30 --
w
E,
20-
!a
W
lo-

0 100 200 300 400

FLOW VELOCITY v, mls

Figure 17.12 Difference oftotal, 9tot, and gas, L1g, temperatures, versus flow velocity, v

At the surface of a real sensor the gas velocity does not fall to zero and there are also
heat losses . For these two reasons the sensor temperature, OT, is lower than Otot given by
equation (17 .21). Using the symbols in equation (17 .21) the average sensor temperature,
OT , indicated by a measuring instrument connected to it is given by:

2
OT = 299 + r
2cp (17 .22)

The factor, r, in (17 .22), which is called the recovery factor of a given sensor, depends upon
the sensor design and the gas parameters. If the indicated sensor temperature, OT, and
factor, r, are known, it is possible to calculate the true gas temperature, 299 .
Gas temperature is often measured by thin, bare thermocouples . The r-values for
different wire diameters and thermocouple designs, which are given in many publications
such as Baker et al. (1961), Jakob (1957) and Hottel and Kalitinsky (1945), unfortunately
differ from one author to another. Figure 17 .13 gives the experimental data by Hottel and
Kalitinsky (1945), for bare thermocouples . Data for MI thermocouples can be found in
Breitkopf et al. (1980). As there is uncertainty in determining the value of the recovery
factor, r, due to its dependence on the measuring conditions, some special sensors having
high values for r are constructed . The principal idea of their design is based on the
application of a protecting shield, which serves to slow down the gas and at the same time
also as a shield to prevent any radiant heat exchange with the environment . It is not possible
to achieve a value of r= 1, as this would correspond to a velocity of zero around the
measuring junction and consequently also to a value of zero for convective heat transfer
coefficient, ak, on the junction surface. Practically, values of the recovery factor, r, lie
mostly between 0.95 and 0.98 .
374 TEMPERATURE MEASUREMENT OF FLUIDS

( b)
0 .9
1
la) ~ 13 mm
_ 0,9
d=0,25=0,50 -y
Q8 d-
Q25 -
d=0,25 --~ `- `
0,8
JUNCTION d=1,8 -' o o
0,7-
d=0,08+015 ~
y - 0,7 } AIR .-20"C
d=0,25
}w
JUNCTION d'--1,8 -` o W 0.6 d=0,25
"13mm

0,6 w
d=0,1=0,5 y ~ AIR-20°C

0,5 0,5 I
0,1 0,2 0,3 01 0,2 Q3 0,4
MACH NUMBER MACH NUMBER

Figure 17 .13 Recovery factor, r, of bare Type J thermocouples versus Mach number in air, at 20 °C
(Hottel and Kalitinsky, 1945)

Figure 17.14(a) shows a typical sensor construction for temperature measurement of

high-velocity gases by Pratt-Whitney Aircraft (USA), described by Baker et al. (1961) .
Figure 17 .14(b) presents values of the recovery factor, r, as a function of air velocity . Some
more detailed information can be found in Baker et al. (1961), and Moffat (1952) .

PLASTIC d, =Q6 F VENT-HOLES

( a)

. .
r
4
TYPE J O6 STAINLESS STEEL TUBE
THERMOCOUPLE

(b)

0,98

0
V Q94
Q

0,90
W

w 0,86
0 150 300
AIR VELOCITY V, m/s

Figure 17 .14 Sensor for the temperature measurement of high velocity gas (Baker et al., 1961)

17.3 Still Air

A very special case of the temperature measurement of slowly moving gas is that of gas
moving under the influence of natural convection. This happens among other occasions,
while measuring the temperature of air in rooms. Considering only normal room
temperature it is noted that the intensity of heat transfer between the sensor and air is very
LIQUIDS 375

low . At the same time, as the sensor and its surrounding are at nearly the same temperature,
errors owing to conductive and radiant heat exchange are very small . Increase in the sensor
time constant resulting from this low intensity of heat transfer may cause dynamic errors .
Correct sensor design involves the following considerations :

" small sensor mass, large surface of convective heat exchange and if needed an
arrangement to force local air circulation,
" radiation shield of vertical axis to enhance free air circulation,
" low emissivity of the sensor surface,
" placing the sensor at a distance from room walls and insulating it from them.

Correct location of the sensor in the room is extremely important for measuring average
room temperature, which is needed in the majority of cases . Placing the sensor near
windows, doors, heaters and lamps should be avoided .
When measuring the temperature of out-of-door air, sensors should be protected from
direct solar exposure . Strictly prescribed measuring conditions which are described by
Jakob (1958), are defined for meteorological purposes. Often special suction thermometers
are used . Liquid-in-glass thermometers, resistance temperature detectors and thermistors
may all be used for still-air temperature measurements .

17.4 Liquids
In this book, considerations of the temperature measurement of liquids is limited to
problems related to either slowly flowing or still liquids which may be water, oil or other
well known types. The temperature range is also limited below about 200 °C, where the
conditions for only convective heat transfer occur so that equation (17 .10) is relevant. In
liquids, the convective heat transfer coefficient is many times higher than in gases . As a
result, the necessary immersion depths for a given error are much smaller. An example of
this dependence for a Pt-100 0 sensor in a steel sheath is given in Figure 17 .15 . The
permissible errors, AT, are below 0.1 % or 2 %. Typical ranges of convective heat transfer
coefficients are also indicated.

17.5 High Temperature Gas and Plasma

Gas and plasma temperatures of several thousands of °C, which far exceeds the application
range of normal and high-temperature thermocouples, often need to be measured. A
method, in which a pulsed thermocouple is periodically inserted into the high temperature
medium, shown in Figure 17 .16, can be used for this purpose. Kretschmer et al. (1977) have
described how its temperatures do not exceed its normal application range provided its
immersion time periods are short enough . Such a sensor, which should be chemically inert
and should have no catalytic properties, takes up the heat by convection and radiation. Ifthe
actual sensor temperature is kept sufficiently low, there is no need to calculate any
corrections for radiant heat exchange with the surrounding .
376 TEMPERATURE MEASUREMENT OF FLUIDS

T
1 .

0,2

AT= 2%
01

1
1 1 11 11 11 111 111 111

1 LIQUID
1
"1 "

" _ , 00 0 sensor in a steel sheath for

permissible_ . _ ,_ I

ii~
q.ii

Figure 17 .16 Junction temperature, *, of a pulsed thermocouple versus time, t, at sampling time, r,
and two different immersion time, ti, values

In those , . . withdrawn . .
temperature medium, it cools down . The mean sensor temperature, 6 Tm , corresponds to a
thermal steady-state, in which the thermal energy, or heat, taken up during the time, ti,
-" _ _ " " , corresponding equation

.. " .. ,
HIGH TEMPERATURE GAS AND PLASMA 377

where t; is the time duration of the immersion, r is the sampling period, g is the heat
(D -T

flux taken up from the medium during the time, ti, and (DT-a is the heat flux given up to the
ambient during the withdrawal time, to.
Assuming that both heat fluxes, (D g -T , and, (DT-a, are linear functions at the relevant
temperature difference, equation (17 .23) becomes:

kg-T (T9g - 6Tm )ti = kT-a(OTm - 6a)(.r - ti) (17 .24)

where kg-T is the equivalent heat transfer coefficient between the gas and the sensor, kT- a
is the equivalent heat transfer coefficient between the sensor and ambient of withdrawal, 6.
is the gas temperature which is to be determined, 6Tm is the mean sensor temperature, 6a
is the ambient temperature, r is the sampling period and t; is the immersion time.
Assume, for simplicity that a linear variation of sensor temperature with time occurs
during heating and cooling . From equation (17 .24) the gas temperature can be expressed as:

4g = kTa_ (OTm - ~a) - til + 6Tm1

Z
(17 .25)
kg-T ti 1

For brevity, substitute the relations :

k' = kT--a and t; _ (17 .26)

- t'
kg-T ti

Holding the sampling time, r, the gas temperature, Og , and the ambient temperature, Oa ,
all constant, the experiments are conducted for two different immersion times, t; = til , and,
ti = tie, with the corresponding mean temperature readings, 6Tm1 and OTm2 . From
equations (17 .25) and (17 .26) two different formulae for gas temperature are derived, as
follows :

$g = k'ti1(OTml -Oa)+OTml (17.27)

z9g = k'tiz (~Tmz - 6a ) + eTm2 (17.28)

The gas temperature, which can be determined by eliminating k' from equations (17.27) and
(17.28), is given by:

1
til(OTml - Oa) 1- til(OTml -6a) (17.29)
g _ ~Tml -~Tmz
ti2 (OTm2 - 6a) 1[ tit (OTm2 - 601 _
378 TEMPERATURE MEASUREMENT OF FLUIDS

or after some transformations :

t5g ^ OTmlti2( 19`rm2 - Oa) - .

OTm2til(OTml -Oa) (17 .30)
ti2( 6Tm2 -6a) - til(OTml -0a)

Thus, the gas temperature, 0. , may be determined from equation (17 .30) from the two
mean sensor temperatures, OTml and OTm2, corresponding to the two respective values of
relative time, t il and tie . To ensure that the amplitudes of the sensor temperature variations
are small compared with the mean sensor temperature, oTm , the sampling time, r, has to be
short enough . When the sampling frequency is too low the amplitudes of the sensor
temperature are so high that they cause a decrease in the precision of the measurement of
the mean sensor temperature, I)Tm , and hence of the gas temperature, 09' On the other
hand, if the immersion frequency is too high, the resulting mechanical accelerations
required may cause some structural damage to the sensor. According to Roeser and Olsen
(1962), the sampling frequency should be about 0.5 to 5 Hz.
The unsheathed end thermocouples with bare measuring junction, which have a water-
jacket as in Figure 17 .17, should protrude by a length of about 6 to 10 wire diameters . As
thermocouples containing platinum are too brittle for the mechanical stresses they have to
stand, it is advisable to use Type K thermocouples . The junction movement should be very
rapid, so that the time needed for transfer from the gas to the water jacket is negligibly small
compared with the immersion time, t i . One of the proven solutions is the application of an
electromagnetic drive by two coils and a ferromagnetic core as shown in Figure 17.17 .
Choice of the ratio (z- t i )/ti should be so that the mean sensor temperature, t9Tm , does
not exceed about 700 °C. This corresponds to the value from which the cooling down curve
of the sensor may still be approximated by a straight line.

HIGH TEMPERATURE GAS OR PLASMA COILS

THERMOCOUPLE WATER JACKET MAGNETIC CORE
IN 'OUT
0000d ~000

000
0 T TO RECORDER

Figure 17.17 Principle ofpulsed thermocouple

17.6 References
Ash, R.L. and Grossmann, G.R . (1972) Catalytic considerations in temperature measurement.
Temperature : Its Measurement and Control in Science and Industry, 4(3), Instrument Society of
America, Pittsburgh, 1663-1670 .
Assmann, R. (1892) Das Aspirationsthermometer. Abhandl. kgl. preuss . Meteorolog. Inst., 5, 117 .
REFERENCES 379

Baas, P.B . and Mai, K (1972) Trends of design in gas turbine temperature sensing equipment.
Temperature : Its Measurement and Control in Science and Industry, 4(3), Instrument Society of
America, Pittsburgh, 1811-1822.
Baker, H.D ., Ryder, E.A . and Baker, N.H. (1953) Temperature Measurement in Engineering, 1, John
Wiley and Sons, New York .
Baker, H.D ., Ryder, E.A . and Baker, N.H . (1961) Temperature Measurement in Engineering, 2, John
Wiley and Sons, New York .
Benson, R.S. and Brundrett, G.W. (1962) Development of resistance wire thermometer for measuring
transient temperatures in exhaust system of internal combustion engines. Temperature: Its
Measurement and Control in Science and Industry, 3(2), Reinhold Publ . Co ., New York, 631-654 .
Blumr6der, G. (1981) Beitrag zur Beschreibung and Ermittlung des statischen thermischen
Fehlerverhaltens industrieller Beriihrungsthermometer, PhD Thesis, T. H. Ilmenau.
Breitkopf, G., Witting, S. and Kim, S. (1980) Recovery factor des frontal angestr6mten zylindrischen
Mantelthermoelementes mit ebener Stirnflache. Warme and Stoffubertragung, 13(4), 287.
Green, S.F . (1987) Temperature in flames and gases. Measurement and Control, 20(6), 19-22.
Haas, A. (1969) Einfluss des Thermometerkopfes auf die statischen and dynamischen Eigenschaften
industrieller Thermometern, m.s .r, 12(12), 141 - 142.
Heitronics Infrarot Messtechnik GmbH (1999) Infrared Radiation Pyrometer.
Hottel, H.C . and Kalitinsky, A. (1945) Temperature measurement in high velocity air streams. J.
Appl. Mechanics, 12(3), A25-A32.
Jakob, M. (1957) Heat Transfer, Vol. 2, John Wiley and Sons, New York .
Jakob, M. (1958) Heat Transfer, Vol. 1, John Wiley and Sons, New York .
King, W.J . (1943) Measurement of high temperatures in high velocity gas streams. Trans. ASME, 65,
421-425 .
Kinzie, P .A. (1973) Thermocouple Temperature Measurement, John Wiley and Sons, New York .
Kretschmer, D., Odgers, J. and Schlader, A.F . (1977) The pulsed thermocouple for gas turbine
applications . Trans. ASME, 99 .
Moffat, E.M . (1952) Multiple shielded high temperature probes, comparison of experimental and
calculated errors . SAE, T-13, No I.
Mullikin, H.F . (1941) Gas temperature measurement and the high velocity thermocouple .
Temperature : Its Measurement and Control in Science and Industry, Reinhold Publ . Co ., New
York, . 775-805 .
Mullikin, H.F . and Osborn, W.J . (1941) Accuracy tests of the high velocity thermocouple .
Temperature : Its Measurement and Control in Science and Industry, Reinhold Publ . Co ., New
York, 805-830.
Ribaud, G. (Editor) (1959) Etudes de pyrometrie pratique, Editions Eyrolles, Paris.
Roeser, S.D. and Olsen, H.L . (1962) The intermittent thermometer a new technique for the
measurement of extreme temperatures . Temperature : Its Measurement and Control in Science and
Industry, 3(2), Reinhold Publ . Co ., New York, 901-906.
Rudolphi, H. (1969) Einfluss der Warmedbergangsbedingungen auf die Messgenauigkeit eines
Temperaturfiihlers in Luftkanalen. ATM, Blatt V2165-5/6.
Thomas, D.B . and Freeze, P.D . (1972) The effects of catalysis in measuring the temperature of
incompletely-burned gases with noble-metal thermocouples. Temperature: Its Measurement and
Control in Science and Industry, 4(3), Instrument Society of America, Pittsburgh, . 1671-1676.
Torkelsson, S.A . (1980) A new type of resistance thermometer for accurate temperature measurement
in high gradient thermal boundary layers. J. Sci. Instrum., 57(5), 549-552.
380 TEMPERATURE MEASUREMENT OF FLUIDS

Wenzel, M. and Schulze, E. (1926) Versuche mit Durchflusspyrometem. Mitteilungen der

Wdrmestelle No. 92 des Vereins Deutscher Eisenhuttenleute.
West, W.E . and Westwater, J.W . (1953) Radiation-conduction correction for temperature
measurements in hot gases. Industrial and Eng. Chemistry, 45(10), 2152-2156.
 Temperature Measurement Second Edition
L. Michalski, K. Eckersdorf, J. Kucharski, J. McGhee
Copyright © 2001 John Wiley & Sons Ltd
ISBNs: 0-471-86779-9 (Hardback); 0-470-84613-5 (Electronic)

22
Calibration and Testing of
Temperature Measuring
Instruments
22.1 Definitions and Terminology
The following main terms are used in the calibration and testing of temperature measuring
instruments, necessary for the maintenance and dissemination of ITS-90 .

" Calibration of a thermometer is the sum of activities concerned with the determination
of its thermometric characteristics. These characteristics define the function correlating
the chosen property of the thermometer with the temperature. If a thermometer directly
indicates the measured temperature, its calibration depends on correlating certain
numerical values with the scale graduation . For example, this concerns liquid-in-glass
thermometers .
" Testing a thermometer is the sum of activities concerned with verifying that the
thermometer complies with the relevant regulations.
" Primary standards are thermometers used for reproduction of ITS-90, as well as for
international comparisons .
" Transfer standards are thermometers used for the transfer of temperature units to other
thermometers, which thus have lower accuracy than these standards . They comprise
secondary, tertiary and other standards, which occupy important transfer levels in what
is called the chain of traceability of standards .
" Working standards are thermometers destined for the calibration of other working
standards, situated lower in the traceability hierarchy . They are also used in the
calibration of industrial thermometers .
" Industrial thermometers are thermometers used in the day-to-day practice of temperature
measurement .
" Laboratory thermometers are thermometers used in laboratories .

Calibration and testing procedures for thermometers comprise a general scheme which
defines the hierarchy of thermometers. They also determine the methods for and accuracy
420 CALIBRATION AND TESTING

of, transferring the temperature unit from primary standards to industrial thermometers
(Richardson, 1962; Gray and Chandon, 1972; Gray and Finch, 1972) .
Most industrialised countries have established national standard laboratories which are
equipped for reproducing ITS-90 through the calibration and testing of standard
thermometers.
The following are among the more well known Laboratories :

" Institut National de M6trologie (INM), France,

" Institute of Metrology "Mendeleiew", Russia,
" Istituto di Metrologia "G. Collonetti" (IMGC), Italy,
" National Institute of Standards and Technology (NIST), USA,
" National Institute of Metrology (NIM), China,
" National Physical Laboratory (NPL), UK,
" National Research Council of Canada (NTC), Canada,
" Physikalisch-Technische Bundesanstalt (PTB), Germany .

Industrial thermometers are tested in regional and industrial laboratories .

22 .2 Fixed Points of ITS-90

22.2.1 General information
Fixed points which define ITS-90 are given in Chapter 1 .
Cryogenic fixed points in the range from 13.8033 K (-259 .3467 °C) to 83 .8058 K
(-189.3442 °C) are used for calibration of capsule-type platinum resistance sensors . They
consist of two boiling points of H2 and the four triple points of H2, Ne, OZ and Ar.
Although the triple point of Ar is often used in the calibration of long-stem platinum
resistance sensors, the triple point of OZ is very rarely used. Cryogenic fixed points, below
the Ar triple point, are not dealt with, as this book does not cover extremely low
temperatures .
Fixed points in the range from 234.3156 K (- 38 .8344 °C) to 1234 .93 K (961 .78 °C) are
used for calibration of long-stem and high temperature platinum resistance sensors. In this
range there are two triple points, of Hg and H2O, one melting point, of Ga, and five freezing
points, of In, Sn, Zn, Al and Ag.
Fixed points in the range from 961 .78 °C to 1084.62 °C, which are used for calibration
of pyrometers, consist ofthe three freezing points ofAg, Au and Cu.
Construction of a fixed point has to be adapted to the kind of calibrated sensor. When
calibrating resistance thermometer sensors, it is extremely important to provide as good a
heat transfer as possible between the sensor and the medium applied in the fixed point. In
the calibration of pyrometers, the fixed points used should have all the properties
of a black body.
FIXED POINTS OF ITS-90 421

22.2.2 Realisation of fixed points

Some of the more important and commonly used fixed points of Table 1 .1 will now be
described .
The triple point of argon 83 .8058 K (-189 .3442 °C), for the calibration of long-stem
platinum sensors is shown in Figure 22 .1 (Bonnier, 1987) . Argon of 99 .9999 % purity is
kept in a sealed cell of stainless steel, immersed in liquid nitrogen . This cell should be able
to withstand pressures up to a maximum of 10 MPa, which occurs at maximum ambient
temperature . The triple point of argon is attained in the melting process from the solid state .
After supplying each pulse of energy to the cell with frozen argon, the cell temperature is
measured. Until the melting process commences, only smaller and smaller steps of energy
are supplied. Pavese et al. (1984) point out that the temperature of the triple point of argon
can then be determined by observing the temperature as a function of time during the
melting process . The protecting tube, in which the calibrated sensor is placed, is in direct
contact with the liquid and solid phase of argon .
The triple point of mercury, 234 .3156 K (-38 .8344 °C), is realised in a glass cell as
shown in Figure 22 .2 (Preston-Thomas et al., 1990) . A cell with mercury is thermally
insulated and placed in a vacuum stainless steel sheath . The degree of vacuum between the
cell and the sheath is controlled to achieve the desired insulation between the cell and the
surroundings . For calibration the RTD sensor is placed in the protecting tube, which is filled
with ethyl alcohol to enhance the thermal contact between both of them. Very high purity
mercury (1 to 10 8 ratio) allows the triple point of mercury to be obtained during either
melting or solidifying. The realisation of the Hg triple point, during solidification, proceeds
as follows . The steel sheath is cooled down by a mixture of solidified COZ and ethyl
alcohol . At the moment when the ampule temperature attains the mercury solidifying point,
the air contained between the cell and its sheath is pumped off. A rod, cooled in liquid

LONG-STEM PLATINUM
HELIUM GAS INLET RESISTANCE SENSOR
^/ VALVE
MANOMETER \
FILLING TUBE FOR
LIQUID NITROGEN

SENSOR TUBE

POLYURE-
THANE
FOAM -
STAINLESS STEEL
VESSEL

-- LIQUID NITROGEN
- - SOLID -LIQUID
-- ARGON
- CRYOSTAT
77 7 -

Figure 22 .1 A cell for the triple point of argon (Bonnier, 1987)

422 CALIBRATION AND TESTING

VACUUM SYSTEM VALVE

RING TUBE SEAL

SENSOR TUBE

ETHYL ALCOHOL

INDIUM SEAL

PAPER INSULATION

STAINLESS STEEL
JACKET

CONNECTION FOR
CLEANING AND FILLING

PAPER INSULATION

Cu FOIL CYLINDER

BOROSILICATE
GLASS CELL

MERCURY

QUARTZ WOOL

STAND

AL-SILICATE
INSULATION

50mm

Figure 22 .2 Triple point of mercury (Preston-Thomas et al., 1990)

nitrogen, is then introduced in place of the sensor . This creates a layer of solidified mercury
in the ampule . In the final stage of this procedure, the calibrated sensor, which has been
cooled down beforehand in a mixture of solidified C0 2 and ethyl alcohol, is then introduced
in place of the rod. The triple point of mercury in solidification is reproduced with a
precision better than f0 .1 mK . Different ways of realising the Hg triple point as well as
their precision are given by Hermier and Bonnier (1992) and by Furukawa(1992) .
The triple point of water, 273 .16 K (0 .01 °C) , which is shown in Figure 22 .3, is
realised in a sealed glass cell filled with distilled water under vacuum. After cooling the
water down to about 0 °C a layer of some millimetres of ice is formed around the inside of
the tube by means of the powdered solid C02 . With the solid C02 removed, the inner tube
is filled with water at about +20 °C for a short time, until a thin layer of ice is melted and
replaced by a thin layer of water . The inner tube is then filled by water at 0 °C to enhance
the heat transfer to the calibrated thermometer. Keeping the cell in the ice-water mixture
maintains the triple point temperature with an accuracy better than t0 .1 mK to f0 .3 mK for
FIXED POINTS OF ITS-90 423

at least 24 hours. A more detailed description is given by Stimson (1956), Hall and Barber
(1964) and Furukawa and Bigge (1982) .
The melting point of gallium 29.7646 °C, is realised in the apparatus built in NPL
(Chattle and Pokhodun, 1987) shown in Figure 22.4. Owing to an increase in the volume of
Ga on solidification, it is placed in an elastic teflon container. Standardised thermometers
are introduced into a nylon tube with a lining of Al . The sealed cell which is immersed in
melting Ga, is filled with an atmosphere of pure argon . Each cell is supplied with
recommendations from the manufacturer specifying the necessary immersion depth in the
bath. An accuracy in the reproduction of the melting point of gallium of about f0.4 mK can
be attained (Chattle and Pokhodun, 1987). A commercially produced gallium melting point
cell, which can be removed for freezing and then returned to the apparatus, holds the
melting temperature for many hours, ensuring a precision of about fl mK.
The freezing point of indiuM 156 .5985 °C, consists of high purity indium (better than
99 .999 %) in either a graphite crucible (Chattle and Pokhodun, 1987) or in a teflon cell
(Mangum, 1989). The indium container is placed in a special furnace . After the ingot is
melted, the furnace temperature is stabilised about a degree below the freezing point . When
the temperature indicated by a resistance sensor in a protecting sheath, placed in the cell,
has fallen close to the freezing point, the sensor is withdrawn and allowed to cool for up to
1 min before being put into the cell again . The loss of heat is sufficient to form a thin layer
of solid indium around the sensor well. The plateau corresponding to the freezing point
temperature is then quickly reached . The freezing point of 99.999 % pure indium is
reproducible to about f0.1 mK (Mangum, 1989; Hanafy et al., 1982) .
The freezing point of tiny 231 .928 °C, when based on tin of very high, 99 .9999 %,
purity, is reproducible to about 0.1 mK (Preston-Thomas, 1990). In realising this point, the
phenomenon of large supercooling at the beginning of freezing, should be taken into

PUMPING TUBE

NYLON CAP

- - TEFLON CONTAINER

OUTER NYLON
CASE

GALLIUM

- - NYLON TUBE
GLASS CELL

WATER VAPOUR -

WATER
Al SENSOR TUBE

ICE LAYER

,~ WATER

1\ ICE-WATER - -
MIXTURE _

hti^1 DEWAR VESSEL - -

U_

Figure 22.3 Triple point of water Figure 22.4 Melting point of gallium (Chattle
and Pokhodun, 1987)
424 CALIBRATION AND TESTING

consideration (McLaren and Murdock, 1960). The construction of this freezing point, which
is described by Marcarino (1992), is similar to the freezing point of zinc, shown in Figure
22.5 . It is important to achieve a high degree of supercooling (over 4 K) to attain the plateau
temperatures by means at outside slow freezing. To attain this supercooling the ingot should
be kept in an inert atmosphere and not topped with graphite powder.
The freezing point of zing 419 .527 °C, is realised with a reproducibility of about 2 mK
if zinc of very high purity (99 .9999 %) is used (Preston-Thomas et al., 1990; McLaren,
1958, Ma and Lawlor, 1992; Furukawa et al., 1981 ; Marcarino, 1992) . The apparatus used
to produce the fixed point should ensure the necessary zinc purity as well as a uniform
temperature distribution during its solidification, with equality of the temperatures of the
sensor and the metal. A tubular furnace with a copper block, which is shown in Figure 22.5,
can ensure this uniformity of temperature . Inside the block there is a graphite crucible with a
lid, through which a sheath for carrying calibrated sensors is inserted . At the beginning of
metal solidification, the observable under-cooling by some hundredths of a degree may be
eliminated by withdrawing the calibrated sensor for a short time. After reaching ambient
temperature the sensor is reinserted. The temperature which then quickly rises to the
freezing point, stays constant at the freezing point temperature for a long period of time.
Zinc purity should be periodically tested.
The freezing point of aluminiuiA 660 .323 °C, is produces in a similar way as for zinc.
Graphite fibre insulation and a graphite crucible are used combined with an atmosphere of
argon gas to prevent oxidation . The reproducibility of this point is about 1 mK with the
temperature stability at the beginning of freezing of 0 .2 mK (Ancsin,1992 ; McAllan and
Ammar,1972 ; Furukawa, 1974). No contact of molten aluminium with moisture, oxygen
and silicon ceramic can exist .
The freezing point of silver, 961 .78 °C, is realised in a similar apparatus to that shown
in Figure 22.5, but with a higher temperature furnace . To get high temperature uniformity in
the middle part of furnace, where the crucible with the silver is placed, it is advisable to
install a heat pipe screen around the crucible. Even though silver does not oxidise easily, it
should still be protected from air contact, since it absorbs oxygen in the molten condition,

SENSOR TUBE
A12 03 -POWDER

ELECTRIC FURNACE

LID

GRAPHITE CRUCIBLE

MOLTEN ZINC

COPPER BLOCK

CERAMICS
THERMAL INSULATION

Figure 22.5 The freezing point ofzinc

PRIMARY STANDARDS 425

resulting in depression ofthe freezing point. As oxygen absorption starts at a temperature of

about 30 °C above the melting point, any unnecessary overheating of the metal should be
avoided. If the ingot is kept in the molten state in an inert gas, the surrounding graphite will
effect the complete removal of the oxygen within a few hours . The constant temperature
level, corresponding to the freezing point of silver, is reached in a time period, covering 20
to 60 % of the total solidifying time, with an accuracy of 1 mK (Preston-Thomas et al.,
1990).
The freezing point of silver, 961 .78 °C, gold, 1064 .18 °C, and copper, 1084.62 °C
which are intended for the calibration of pyrometers, are each realised in water cooled
electric resistance furnaces with a graphite chamber inside as shown in Figure 22.6
(Ohtsuka and Bedford, 1982). The chamber has all the properties of a black body. The
graphite black body and crucible containing the metal are made ofgraphite ofhighest purity
so that an attainable emissivity of about 0.99999 is achieved (Lee, 1966) . The crucible is
placed in a sodium heat-pipe liner, giving a uniform temperature of the crucible walls.
Additional blocks and rings of graphite keep oxygen from reaching the crucible . The
thermal insulation of the furnace is made of quartz wool and inconel heat shields . During
operation nitrogen or argon flows slowly (0.1 litre/min) along the furnace length to inhibit
graphite oxidation . The temperature is measured by a type S thermocouple . The calibrated
pyrometer is aimed at the bottom of the graphite chamber.

CERAMIC SODIUM GRAPHITE FREEZING METAL(A9,Au,Cu)

TUBE HEAT PIPE CRUSIBLE
THERMOCOUPLE

r r r r ~ ~ r
r
r ~ ^ r r z
r r
r ~ ^ t

_ _
r
r
r 2 ~r 1-*ARGON
r

r 1 r

Z t 1 J r r

^.
1 t r rr I t QUARTZ
WOOL

GRAPHITE BLACK INCONEL

WATER COOLING BODY CAVITY RADIATION SHIELD
60 an

Figure 22.6 Metal freezing points for calibration ofpyrometers (Ohtsuka and Bedford, 1982)

22.3 Primary Standards

Primary standards are instruments specified in the text of ITS-90 for interpolation between
the fixed points .
Standard resistance thermometers are used for interpolation from -259.3467 °C to
961 .78 °C, which is the freezing point of silver. Following ITS-90 the thermometer resistor
must be strain free, annealed, pure platinum and wound from 0.05 to 0 .5 mm Pt wire . It is
advisable to use resistors of 25 S2, at 0 °C . In the upper temperature range, 0.1 to 2.5 S2
resistors are recommended . The resistor should be enclosed in a hermetic sheath filled by
426 CALIBRATION AND TESTING

dry, neutral gas with an addition of oxygen. At the lower temperature range, up to 13 K, it
should be helium filled. The resistor should be annealed at a temperature higher than the
highest expected working temperature, but in any case never below 450 °C (except for
cryosensors) . The quality of a sensor, its design and annealing are verified during
calibration, determining the constants from interpolation equations and checking the
stability of the resistance (Curtis, 1972; Foster, 1972).
Depending on the working temperature range, there are three types of resistance
temperature sensors:

" -260 °C to 0 °C - low temperature capsule-type sensors,

" -190 °C to 600 °C - normal long-stem sensors,
" +600 °C to 960 °C - high temperature sensors .

Capsule-type resistance sensors, which are beyond the scope ofthis book, are described by
Curtis (1972), Hust (1970) and Sparks and Powell (1972) .

Long-stem-type resistance sensors, which are used as interpolation standards of ITS-90

from -190 °C to 600 °C, have undergone many modifications to increase their accuracy and
stability, to reduce their size and to intensify the heat transfer between the resistor and
sheath and between the sheath and environment.
A typical contemporary design is presented in Figure 22 .7. Platinum wire, wound in a
spiral of about 1 mm diameter is placed in a thin-walled Pyrex tube matching the spiral
diameter and shaped as shown in Figure 22.7. Platinum terminals, which are soldered to
both spiral ends, are sealed in glass in such a way that the spiral is totally strain free. It is
also important for the spiral to remain strain free during its subsequent working life. These
terminals are extended by low resistance gold wires in ceramic insulation.
The whole assembly, which is encapsulated in a glass sheath, is hermetically sealed after
careful drying at about 400 °C. The resistance of the sensor is about 25 0 at 0 °C. As an
example, the standard Pt resistance sensor produced by Rosemount Inc . (USA) (Berry,
1982) is hermetically sealed in a metal sheath containing a helium-oxygen atmosphere. Its
stability within the specified temperature range of -200 °C to +650 °C is better than 0.01 °C
per year. The self-heating increase in temperature is less than 0.002 °C with an insulation
resistance from the resistor to the outside sheath greater than 5000 MQ at 100 V dc, while
its nominal resistance is about 25 S2 at 0 °C.

High temperature resistance sensors, operating from 600 °C to 960 °C replace the
S-type thermocouple as a primary standard ofITS-90 . Design ofhigh temperature resistance
sensors is the subject of many publications (Arai, 1997 ; Anderson, 1972, Evans and Burns,

Au-WIRES Pt-WIRES GLASS SHEATH d=6mm

CERAMIC INSULATOR Pt-SPIRAL PYREX TUBE

Figure 22.7 Standard long-stem-type resistance sensor

PRIMARY STANDARDS 427

1962; Chattle, 1972; Curtis, 1972; Evans, 1972; Furukawa et al., 1981; Strouse et al.,
1992) . One of the designs proposed by Nubbemeyer (1992) is shown in Figure 22.8.
The resistor is composed of a bipolar spiral winding of 0.4 mm diameter platinum wire,
supported by a notched quartz blade. Two Pt wires of the same diameter, which are welded
to both ends of the spiral, are extended by two 75 cm long, 0.35 mm diameter Pt wires .
These wires are insulated by quartz tubes passing through 9 quartz disks placed along the
sensor. After the sensor has been annealed at 700 °C, the external 7 mm diameter quartz
sheath is hermetically sealed and the tube filled with a gas mixture of 90 % Ar and 10 % Oz.
The sensor resistance is 0.25 SZ at 0 °C.

Resistance sensors for interpolation in ITS-90, whose resistance is measured by an ac

bridge of highest precision in a four-wire circuit, are calibrated at relevant fixed points
within the sensor application range. An example of the bridge is the F 18 bridge produced by
Automatic Systems Laboratories Ltd (1999) having the following technical data:

" accuracy better than + 0.25 mK,

" resolution : 0.75 pK,
" measuring range: 0-390 fl (for Ra = 2.5-100 S2 in the temperature range : 13 K-960 °C),
" frequencies: 25/75 Hz or 30/90 Hz,
" possibility of measuring temperature difference,
" automatic or manual balancing,
" interface : IEEE-488 .

Standard pyrometers and tungsten strip lampsare used for reproducing ITS-90 above
the freezing point of silver, 961 .78 °C. They are calibrated at the three fixed points of Ag,
Au and Cu, simulating black body radiation

Pt 0, 35 mm

QUARTZ TUBE
QUARTZ
4-HOLE DISK QUARTZ
SHEATH d=7mm

WINDING
Pt 0,4mm

QUARTZ BLADE QUARTZ BLADE

QUARTZ DISK

JUNCTION OF
TWO PARTS
OF WINDING

Figure 22.8 Standard high temperature resistance sensor (Nubbemeyer, 1992)

428 CALIBRATION AND TESTING

Standard pyrometers and their calibration methods have undergone many modifications in
recent years . The photoelectric spectropyrometer, which was the early type, is based on the
principle of the disappearing filament pyrometer, where a photoelectric detector replaces
the human eye, as described by Hahn et al. (1992), Kandyba and Kowalewski (1956), Lee
(1966), Lee et al. (1972) and Nutter (1972) . This early type has now been replaced by a
narrow-band photoelectric pyrometer with either a photomultiplier detector or by the
increasingly popular Si detector, which is characterised by high sensitivity, high stability
and good linearity (Coslovi and Righini, 1980 ; Jung 1979) .

Standard photoelectric pyrometers, which have an optical system like that shown in
Figure 22 .9 (Rosso and Righini, 1985), also have an accuracy better than 0 .1 K in the
temperature range of 800 to 1400 K and about 1 K at a level of 2000 K similar to others
(Zhao et al., 1990, 1992) . The silicon detector of the pyrometer, which is placed in a
thermally stabilised housing, gives an equivalent stability of 0 .1 K per month and some
tenths of kelvin per year. Methods of pyrometer calibration at the fixed points are
considered by Bussolino et al. (1987) .
Although pyrometers are mainly calibrated for measuring temperature, some pyrometers
may also be used as instruments for comparing heat fluxes (Preston-Thomas et al., 1990 ;
Zhao et al., 1992) .
Standard tungsten strip lamps are used for interpolation in the temperature range from 1337
to 2600 K. Vacuum lamps can be used up to about 2000 K, whereas above this temperature
the use of gas-filled lamps, shown in Figure 22 .10, is advised . The strip length must be big
enough to prevent any substantial influence of ambient temperature on the strip temperature .
A `place' is also marked on the strip where the measurements should be made . The sighting
angle of the pyrometer is given by two points . One point is on the sighting window and
another is on the strip. These two points should coincide during measurements . To prevent
any reflection of the radiation, which might be a source of errors, both lamp windows are
situated at an angle of 5° to the lamp axis . The dependence of the temperature of the lamp's
strip upon the lamp current is called the thermometric characteristic of the lamp . To achieve
high stability of this characteristic the lamps are degassed many times during the production
process before being finally glued and annealed (Quinn and Lee, 1972) . Calibration of
tungsten strip lamps is made by a comparison method based on the readings of a
photoelectric spectropyrometer and simultaneous measurement of the lamp current at
different strip temperatures .

OBJECTIVE SHUTTER IN TEM E RA R

PETURE
LENS MIRROR APERTURE CONTROLLED
"' DIAPHRAGM STOP ENCLOSURE
TARGET

INTERFERENCE
EYE PIECE MIRROR

Ir
MIRROR LL------------- -
Figure 22 .9 Standard photoelectric pyrometer - optical arrangement (Rosso and Righini, 1985)
WORKING STANDARDS 429

PYREX WINDOW

PYREX BULB

TUNGSTEN
( STRIP

Ni-CONDUCTORS

LAMP HOLDER

Figure 22 .10 Standard tungsten strip lamp

Tungsten strip lamps must be fed by a direct current, maintaining the correct polarity
during the measurements, because the temperature distribution along the strip depends upon
the current direction through the Thomson effect:.

22.4 Working Standards

The primary standards described above are used both for the realisation of ITS-90 as well
as for international comparisons . Working standards are usedfor calibration and testing of
other thermometers.
Resistance sensors are used in the temperature range from -190 to +960 °C for the
calibration and testing of resistance sensors of lower accuracy . Below 0 °C they are used for
the calibration and testing of type T thermocouples and semiconductor thermometers. The
nominal value ofthe reference resistance, Ro (at 0 °C), ofthese sensors is about 10 0, 25 0
or 100 S2. These values do not apply to first order sensors (transfer standards) from 0 to
650 °C, for which only 10 S2 and 25 0 are permitted . This is also true for high-temperature
sensors of lower resistance. General outlines of the design of working standard resistors are
also applicable to the design of standard resistance thermometers. Calibration of resistance
sensors, can be made at fixed points. In many laboratories the ice-point is used instead of
the triple point of water. The comparison method in testing baths is also used, comparing
the readings of the thermometer to be calibrated with those of a thermometer of higher
accuracy. Calibration thermostats to be used are described in Section 22 .6 .2 . To avoid
errors arising from conduction losses along the sheath and leads, the recommended
minimum depth of immersion of sensors is about 20 cm. All sensors should be annealed
prior to calibration at 500 °C for about 30 min. Self-heating errors are usually less than
0.004 °C, provided the sensor resistance for which Ro (at 0 °C) is about 25 S2 are generally
measured at a current of 1 mA dc.
Mercury-in-glass thermometers (-30 to +630°C) and mercury-thallium-in-glass
thermometers (-55 to +30°C) are used as transfer standards for calibration and testing of
other glass thermometers, as well as that of manometric, resistance and thermocouple
thermometers. Different liquid-in-glass thermometers are used. Their permissible errors are
±0 .01 °C to ±3 °C. To increase the accuracy of measurement by mercury-in-glass
thermometers, errors due to zero changes, external pressure variations and variations in
430 CALIBRATION AND TESTING

temperature of the mercury column should be eliminated . Detailed information on the

design of precision mercury-in-glass thermometers, including consideration ofthe glass and
quartz used, as well as measuring technique and errors is given by Hall and Barber (1964) .
Calibration of glass thermometers is conducted at fixed points or by the comparison
method in baths .
Thermocouple sensors used as working standards are types S, R and B and some other
thermocouples of non-rare metals . They are calibrated against a resistance thermometer and
at higher temperatures against standard optical pyrometers . Types S and R working standard
thermocouples are used in the temperature range from 300 to 1300 °C for the calibration
and testing of the same types of thermocouples of lower accuracy as well as other
thermocouples. They have to be annealed carefully, in most cases by direct current flow.
They are mounted in twin-hole insulation of pure A1z03. The thermocouple conductors
range in diameter from 0.35 mm to 0.65 mm.
The accuracy of different working standard thermocouples, such as those produced by
Isothermal Technology Ltd (1999a), are :

" rare metal types:

±1 °C in the range 0 to 1100 °C,
±2 °C in the range 1100 to 1300 °C,
" other types:
±0.3 °C in the range -80 to +250 °C,
±4 °C in the range 1100 to 1300 °C.

Type B working standard thermocouples can be used in the temperature range from
650 °C to 1770 °C.
Working standard thermocouples which are calibrated at fixed points or by the
comparison method have their reference junction usually kept at 0 °C, the ice-point
(Roberts, 1980) .

Tungsten strip lamps are used in the temperature range from about 850 to 2800 °C for
the calibration and testing of similar lamps of lower accuracy as well as for calibration and
testing of pyrometers. These working standard lamps which are constructed in a similar way
to standard tungsten strip lamps, are calibrated by standard photoelectric pyrometers .

Disappearing fllament pyrometers are used in the temperature range from about 800
to 2600 °C for the calibration of working standard tungsten strip lamps and for testing of
similar pyrometers of lower accuracy. Grey filters are used for expanding their measuring
range. These pyrometers are calibrated against standard tungsten strip lamps, based on their
thermometric characteristics . Now they have been almost entirely replaced by the Si
detector pyrometer.

Photoelectric silicon detector pyrometers, operating in a similar way to the standard

pyrometers shown in Figure 22 .9, are used for temperatures over 960 °C. Recently a new
pyrometer with a detector of InGaAs as well as a number ofnew black-body radiators have
been described . All of them are to be used for calibration in temperature range of 275-
960 °C (McEvoy et al., 1997; Fischer and Gutschwager, 1997).
TESTING OF INDUSTRIAL THERMOMETERS 431

22 .5 Testing of Industrial Thermometers

22.5.1 Introduction
During normal operation industrial thermometers are subject to various factors, such as high
temperatures, mechanical and chemical influence and so on . Under the impact of such
factors, their thermometric characteristics can significantly vary. Thus they have to be
checked periodically and recalibrated or repaired if needed . The frequency of such
periodical checking, which depends upon the working conditions, should be determined by
the maintenance service .
Industrial thermometer testing is achieved by the fixed points method or by the
comparison method .

Fixed points testing uses some fixed points of ITS-90, described in Section 22 .2 .1, but
realised with lower accuracy. These are combined with some additional fixed points listed
in Auxiliary Table XXI, as well as freezing points of some chemical compounds . The most
popular instruments used in this method are described in Section 22 .6 .1 . Testing at fixed
points, especially at the freezing points of metals, is precise and free from the subjective
judgement of operators . However, because of the expensive equipment required and time
consuming nature of the measurements, it is not so popular for industrial thermometers .

Comparison methods for testing industrial thermometers is based on a comparison of the

readings of tested and reference thermometers, while ensuring that the sensitive sensor parts
are always kept at exactly the same temperature . This condition is difficult to satisfy
especially at higher temperatures . The tested and reference thermometers are placed in a
testing bath, fluidised bed thermostat or a testing furnace, depending on the necessary
temperature range . The testing accuracy depends upon the accuracy of the standard
thermometer, on uniformity and constancy of temperature and, to a great extent, on the
ability and experience of the operator .

Industrial pyrometer testing by the comparison method requires that both the standard and
testing pyrometers are directed at the same radiation source . In Section 22 .6 .4 some black
body radiation sources are described.

22.5.2 Variable volume thermometers

Testing of liquid-in-glass thermometers comprises visual examination and testing of
thermometer accuracy.

Visual examination covers control of the scale and of its position relative to the capillary . In
mercury-in-glass thermometers the mercury column should neither be broken nor contain
observable impurities . If dirt is observed or if moisture traces exist in the capillary of any
thermometer it must be rejected . Before testing, breaks in the mercury column must be
rejoined.

Thermometer accuracy is tested by the comparison method using a stirred-liquid bath and
two standard liquid-in-glass thermometers described in Section 22 .6 .2 . Readings are taken
while the temperature is slowly rising, completing two measuring cycles. During each cycle
432 CALIBRATION AND TESTING

the routine of measurements should be : standard thermometer No. 1, tested thermometers,

standard thermometer No 2, standard thermometer No. 2 again, tested thermometers in
reverse succession and standard thermometer No. l . Calculated mean values of the four
readings for each tested thermometer correspond to the mean temperature values
determined by the standard thermometers.
In the range -200 to +90 °C, the thermometers must be tested while totally immersed.
Over +90 °C they are tested, while partially immersed, with the average value of the
temperature of the emergent liquid column also being measured .

Testing of manometric thermometers consists of visual examination, testing of the

thermometer accuracy, testing ofthe hysteresis of the indications and testing of the variation
of the indications .
Visual examination includes verifying the correctness of the markings. The general state
of the thermometer, including the capillary, pointer, scale and elastic element, should also
be assessed.
Testing ofaccuracy and hysteresis is by the comparison method in a stirred-liquid bath.
Because of large time lags in the indications of manometric thermometers during their
testing, it is important to keep the sensor in the bath for as long as necessary until the
readings are stabilised. Meanwhile the bath temperature should be kept constant. For
accuracy tests, the mean values of readings at each temperature are considered . For
hysteresis testing, measurements are taken at several temperatures, starting at the lowest
one, going up to the highest and then reversing the procedure on the way back down.
The difference in the indications between increasing and decreasing temperature, is a
measure ofhysteresis.
Indication variations are determined by measuring the same temperature several times
under constant measuring conditions and observing differences in the readings . The
accuracy and variations of the indications should be evaluated from sensor and read-out
instrument, obtained at the same level.

22.5.3 Resistance thermometer sensor

Testing of resistance thermometer sensors requires visual examination, testing the
thermometric characteristic of the resistor, checking the stability of the resistance, testing
for self- heating errors and testing of both the break- down strength of the sensor insulation
and insulation resistance.
During visual examination of the sensor, its general state is checked and defects
observed. Correct marking of the working range, type of resistor, sheath material and so on,
are to be verified .
Testing ofthe thermometric characteristic has to verify that it conforms to the standard
EN 60751 with permissible tolerances. Usually the testing is performed by the comparison
method in liquid baths and in metal block calibrators .
Resistance stability checking (only for new sensor construction), which is performed if
the resistor is used over 300 °C is made by measuring the resistance R o at 0 °C before and
after 250 hours of heating at the lowest and the highest working temperature of the sensor
and before and after ten cycles of heating in both these temperatures. The changes of
resistance, R, should not exceed one-fourth of the permissible resistance tolerance, as
defined by EN 60751 .
TESTING OF INDUSTRIAL THERMOMETERS 433

Testing of the self-heating error (only for new sensor construction) is described in detail
in standard EN 60751 .
Testing of break-down strength and of resistance of the sensor insulation has to be done
to ensure conformity with relevant national standards .
To measure the resistance of thermometer detector calibrators (Section 22 .7), bridge or
voltage comparison systems equipped with digital read-out instruments as well as
computerised data acquisition and processing systems, as described in Chapter 13, are
increasingly popular .

22.5.4 Thermocouples
Testing of thermocouples (Roeser and Wensel, 1941 ; Standard EN 60584) involves visual
examination, testing of emf versus temperature characteristics and testing of sensor
insulation resistance
Visual examination is carried out after removing the thermocouple from its sheath and
removing the ceramic wire insulators . Thermocouples exhibiting stains and scale are
rej ected .
Before testing the emf versus temperature characteristic, rare metal thermocouples are
chemically cleaned by 50 % hydrochloric acid, carefully washed in distilled water and
annealed by direct current flow . Type S and R thermocouples are annealed at 1150±50 °C,
which corresponds to a current of about 1 l. A for a diameter of 0 .3 mm . Type B
thermocouples B are annealed at 1400±50 °C, which corresponds to about 13 .5 A for
0 .5 mm diameter wires .
The emf versus temperature characteristic, has to be compared with the standard
EN 60584 . Testing methods used are the comparison method, the differential comparison
method, the measurement of emf at fixed points and the measurement of emf at fixed points
by the wire method.

The comparison method is used for all of the following standardised industrial
thermocouples in the manner described below .

Type S, R and B thermocouples in ceramic insulators are placed, together with the standard
thermocouple in a tubular electrical furnace, having a normal working temperature of at
least 1200 °C . The measuring junction of the thermocouples to be tested, should be placed
simultaneously in the middle of the furnace length described in Section 22 .6 .3 and should
not exceed five in number. To ensure that the temperature of the measuring junctions of
standard and tested thermocouples is equal, they should be bound by platinum wire or
welded together .

Type K thermocouples are put in ceramic insulators, together with the standard Type K
thermocouples, and placed in a metal block in a tubular furnace described in Section 22 .6 .3,
so that the measuring junctions are in direct contact with the metal block . The working
temperature of the furnace should be at least 1000 °C . If a Type S thermocouple is used as a
standard thermocouple, it is also placed in one of the block holes . In that case the
thermocouple is not bare but is placed in a gas-tight glass or ceramic sheath .
To ensure better equalisation of the temperature of tested and standard thermocouples,
in more precise measurements, the measuring junction of the standard thermocouple is
placed in a hole, drilled in the junction of the tested thermocouple, which is always larger .
434 CALIBRATION AND TESTING

Contamination is prevented by protecting the wires of the standard thermocouple with

refractory cement . Inhomogeneity in the junction itself does not influence the results as long
as no temperature differences occur in it. When testing a number of thermocouples, they can
be welded together to form one common measuring junction .

Type J, T and E thermocouples are tested in baths, up to 300'C as described in

Section 22.6 .2, using a mercury-in-glass thermometer as the standard . From 300'C to
700 °C, Type J thermocouples are tested in the same way as Type K units. The number of
simultaneously tested thermocouples should not exceed six .
In all thermocouple testing, their reference junction temperature should be at
the ice point.
For measuring the thermoelectric force, the specialised digital voltmeters or calibrators
described in Section 22.7, are used. Computer programs also exist, enabling full
automisation of measurements (Isothermal Technology Ltd, 1997b). Thermocouple testing
by the comparison method is more and more frequently conducted using electric furnaces
with programmed temperature control (Automatic Systems Laboratories Ltd, 1999; Jones
and Egan, 1975; Kirby, 1982; Techne (Cambridge) Ltd, 1999a; TMS Europe Ltd, 1997) .

The differential comparison methodis only used for rare-metal thermocouples

with the same type of thermocouple used both as the standard and as the tested
thermocouples . Conductors with the same polarity for both the standard and tested
thermocouples, up to four at a time, are bound together by a platinum wire as near the
measuring junctions as possible to ensure as good a thermal and electrical contact as
possible. The same type of tubular furnace is used as in the comparison method .
Measurements of the differential emf values between conductors of the same polarity of the
standard and the tested thermocouples are taken from the first to the last thermocouple and
then in reverse succession. At the beginning and at the end of each cycle, the true-furnace
temperature is measured by the standard thermocouple. The relevant electric circuit for the
thermocouples is shown in Figure 22.11 .
Following the American National Standard ASTM E220-80, the differential comparison
method has the following advantages, relative to the comparison method :

" Measured differential emfs are small relative to the relevant thermocouple emf at the
given temperature and thus do not need to be measured very precisely.
" During testing, much higher rates of temperature increase can be applied, because the
differences of the thermoelectric characteristics of the tested and the standard
thermocouples vary insignificantly as a function of temperature . Also, the furnace
temperature does not need to be precisely measured
STANDARD THERMOCOUPLE DOUBLE-POLE SWITCH

90%Pt10%Rh = const.
b
MEASURING
INSTRUMENT

V/-

TESTED THERMOCOUPLES

Figure 22.11 Testing of S type thermocouples by differential comparison method

TESTING OF INDUSTRIAL THERMOMETERS 435

Emf measurement at fixed points is very popular, especially in USA (Richardson,

1962 ; Trabold, 1962) . The tested thermocouple is immersed consecutively in crucibles with
metals and salts of different freezing temperatures . The tested thermocouples should be
immersed to an adequate depth to prevent heat flow from the measuring junction along the
thermocouple conductors . Any small changes in immersion depth should not affect the
measured emf values provided this depth is adequate . In recent times, new types of
miniature slim cells, with the fixed points described in Section 22 .6 .1, have been developed .
These are mainly intended for testing rare-metal thermocouples. The emf values,
corresponding to relevant fixed point temperatures, are recorded as a function of time with
the horizontal part of the emf versus temperature curve determining the sought emf value .

Emf measurement at fixed points by the wire method (Hall and Barber, 1964 ;
Roberts, 1980 ; Trabold, 1962) is mainly used for rare-metal thermocouples . The measuring
junction of the tested thermocouple is cut in two with a short pure metal wire then soldered
to both thermocouple conductors as in Figure 22 .12(a) . After being prepared in this way,
the thermocouple is slowly heated up in a tubular furnace, till the inserted wire melts . The
recorded emf versus time value clearly indicates the constant temperature part of the curve,
till the circuit is broken . Instead of wire an Au, Pd or Pt plate can also be used as shown in
Figure 22 .12(b) and (c) .

In-situ testing of thermocouples is becoming increasingly more important.

Thermocouples should not be tested in a laboratory environment after a longer working life
or after being contaminated.
The emf of a non-homogeneous thermocouple depends upon the temperature
distribution along the thermocouple and thus the emf values measured by the laboratory
method do not correspond to in-situ readings . Testing is made by the comparison method in
which the standard thermocouple is placed alongside the tested one so that their measuring
junctions are at the same temperature . At a constant measured temperature, both
thermocouples can be placed alternately at the same place . If many thermocouples are used
in one installation, an additional empty sheath is introduced to hold the standard
thermocouple during testing . Such an empty ceramic sheath is sometimes placed together
with two other sheaths in one common outer ceramic tube . This third sheath which is kept empty
is only to house the standard thermocouple temporarily during testing (K6rtvelyessy, 1987) .
Testing should be carried out at some temperatures distributed over the working range .
Although the method described above is not as precise as the laboratory method, it is very
useful in many cases . If a temperature sensor is always connected to the same measuring
instrument then the whole installation is tested instead of only testing the thermocouple .
Readings of the whole installation are compared with those of a standard thermometer .

(a) WELDED OR IN WELDED (c) LOCKED

WRAPPED WIRE PLATE PLATE

Figure 22 .12 Testing of thermocouples by the wire method

436 CALIBRATION AND TESTING

The possibility of precise emf measurement ofa thermocouple at one fixed point, can be
done using a so-called self-calibrated thermocouple . Such a thermocouple has a "miniature
crucible" with a pure metal such as Ag, at its end (Lehmann and Bernhard, 1997; Ronsin
and Elgourdou, 1997). At the moment when this metal is just at the freezing state, the
measured emf should correspond to its freezing point.

22.5.5 Disappearing filament pyrometers

Testing of industrial disappearing filament pyrometers involves both visual examination,
and testing of the precision of readings .
Visual examination includes the optical system and filter as well as the smoothness of
movements ofthe objective.
Precision of readings of industrial pyrometers is tested against a tungsten-strip lamp
described in Section 22.3 .2 or a set of lamps, in the temperature range from 700 to 2000 °C.
The comparison method, where the readings are compared against the standard pyrometer
using the strip lamp as a transfer instrument, is also used.
Error in the measurement of tamp current, which should not exceed 0.1 % of the
measured value, is most commonly measured by a potentiometric method, where the voltage
drop across a standard resistor in series with the lamp is measured.
Testing of a pyrometer by the comparison method, using a standard pyrometer, is
explained in Figure 22.13 . Both the tested and standard pyrometers are aimed through
identical optical systems at the radiation source which is a tungsten strip-lamp. Its
temperature is set by means ofa regulating transformer. The applied standard pyrometer has
a known thermometric characteristic, given as the measured temperature versus voltage
drop across the pyrometer lamp. When the brightness of the tungsten strip lamp and ofboth
pyrometer filaments are made equal, the voltage drop across the standard pyrometer lamp is
measured by the potentiometric method. Simultaneously the temperature indicated by the
tested pyrometer is measured.

RADIANT
SOURCE

STANDARD
\ PYROMETER
TESTED
PYROMETER

DIGITAL T
VOLTMETER DECADE RESISTOR

Figure 22.13 Testing of a disappearing filament pyrometer by the comparison method

AUXILIARY EQUIPMENT 437

22.5.6 Total radiation, photoelectric, two-colour and

two-wavelength pyrometers
The above mentioned pyrometers are commonly tested by the comparison method using a
radiation source which approaches a black body as closely as possible. A precise
photoelectric Si-detector pyrometer, a disappearing filament pyrometer, a pyrometer of the
same type as the one tested, a thermocouple or a resistance thermometer may be used as a
standard instrument .
For very accurate testing, the fixed point method is used. Here the radiation sources
have the temperature of a given fixed point. In this case the thermometer is only used to
measure temperature as a function oftime. Pyrometer calibration is performed at a period of
time when the fixed point temperature is constant.While testing total radiation and
photoelectric pyrometers, their field of view should be totally filled by the radiation source
used. Following Ircon Inc. (1997) the diameter of the radiation source should be at least
twice as large as the diameter of the field of view of the pyrometer. Two-colour and two-
wavelength pyrometer calibration does not require either this condition or a high emissivity
value of the radiation source . Black body radiation sources for pyrometer testing are
described in Section 22 .6.4.

22.6 Auxiliary Equipment

The equipment of industrial testing laboratories depends upon the tasks they are required to
perform. Some auxiliary equipment is used for testing working standards, while the others
are for testing industrial and laboratory thermometers. Here their accuracy and cost is a
deciding factor .

22.6.1 Fixed points

Some fixed points of ITS-90 but of lower accuracy and also some other fixed points as
listed in Table XXI, are used in industrial testing laboratories.

The ice point (0 °C), shown in Figure 22 .14, is of great practical importance. It is used for
testing variable volume liquid thermometers and resistance thermometers. It also serves as a
reference temperature source in testing thermoelectric thermometers. Thermometers or
sensors to be tested are immersed in a mixture of finely shaved pure ice and instilled water
contained in a Dewar vessel. The water which should be air-saturated, should also be well
shaken with air at a temperature near 0 °C. If needed, some ice should be added and surplus
water removed (Hall and Barber, 1964). The constant temperature bath, Model ISOCAL6-
VENUS, offered by Isothermal Technology Ltd (1999b), ensures an accuracy of 0.001 °C.
In this apparatus the magnetic stirrer lets the water flow past the ice mantel, so it is brought
to 0 °C before coming in contact with the thermometers being tested . The calibration zone is
25 mm in diameter and 160 mm deep.
The boiling point of water (100 °C) shown. in Figure 22 .15 as an apparatus which is
open to the atmosphere, is simple in operation . Water in the boiler is kept boiling by an
electric heater . Saturated vapour flowing along the radiation shield, washes the calibrated
438 CALIBRATION AND TESTING

CONTAINER
COOLING
SENSOR ~WATER
SHEATS
RADIATION
SHIELD
WATER
DEWAR VESSEL
MANOMETER
CONDENSER
ICE-WATER
MIXTURE

WATER
HEATING
= ELEMENT

Figure 22 .14 Ice point Figure 22 .15 Boiling point of water in an open
system

sensors before eventually condensing in the water-cooled condenser and returning to the
boiler. A water manometer indicates the internal pressure . The open system used, which is
sufficient for an accuracy below 0 .005 °C, (Hall and Barber, 1964), is also suitable for
testing several thermometers simultaneously . This point is now only interesting from the
historical viewpoint .
The most commonly used fixed points are the melting and freezing points of metals and
salts such as those of Ga, In, Sri, Pb, Zn, Al, Ag and Sodium Chloride (801 °C) . They are
built in a similar way to that described in Section 22 .2 .2 . Different purity of metals is
needed, depending on the required accuracy . When testing industrial thermocouples a purity
of 99 .9 % is sufficient. Easily oxidising metals are melted in graphite crucibles while the
other kinds are melted in steel crucibles . Crucibles are covered by graphite or steel lids . The
tested thermocouple, which is placed in a protecting sheath, is immersed ata depth sufficient
to prevent any temperature gradients in the neighbourhood of its measuring junction .
Placing the crucible in a vertical tubular furnace, which can be used for many different
crucibles, ensures they all have a uniform temperature distribution .
For short and thin thermocouples and resistance sensors (d< 7 mm, 1 < 130 mm)
Isothermal Technology Ltd (1997a) offers Slim Cells shown in Figure 22 .16, which are
slimmer, shorter and lower in price than the standard fixed points cells . The same producer
also offers a range of furnaces and heat sources to allow Slim Cells to be utilised .

Figure 22 .16 Miniature fixed point cells Slim Cells (courtesy of Isothermal Technology Ltd)
AUXILIARY EQUIPMENT 439

Perhaps the most useful fixed point cell for most industrial laboratories is a slim gallium
cell realising the fixed point ofgallium (29 .7646 °C) . It is more robust, easier to use, less
expensive in running costs and can be used as a portable device, compared with the
neighbouring water triple point. Simply placing it into warm water gives a melting plateau
of several hours.
Miniature cells with the freezing points ofAg, Au and Pd are described in Edler (1997),
Ronsin and Elgourdou (1997) and Ronsin (1992) .

22.6.2 Thermometer testing baths

Thermometer testing baths are used for thermometer calibration and testing by the
comparison method. A uniform temperature distribution around both standard and tested
thermometers should be ensured .
In liquid testing baths this uniformity is ensured by efficient stirring and liquid
circulation . Although liquid heating is done by electric heating elements, previously cooled
liquid, or liquid presently cooled by the addition of solid C02 or by a cooling coil, can be
used for temperatures below ambient .
Liquid media which are used in testing baths should be characterised by good thermal
conductivity, high specific heat and low viscosity . Best suited is water, but its application
range is only 0 to 100 °C. Other liquids, applicable from -175 °C up to 630 °C are listed in
Table 22.1 (Hall and Barber, 1964).

Table 22.1 Media applied in testing baths and fluidised bed thermostats

Thermostat Medium Temperature rangeCC)

Liquid bath Freon -175 to -40
Alcohol -100 to 0
Water 0 to 100
Mineral* and Si oil 50 to 250
Salt mixture of
180 to 630
52 % KN03 + 44.8 % NaN03
Fluidised bed A1203 powder 70 to 1100
Zr powder
*Upper application limit should be about 50 °C below the inflammation point

A water bath, which is used from 0 to 100 °C shown in Figure 22.17 consists of two
cylindrical containers, connected at top and bottom by horizontal channels. The tested and
standard thermometers are placed in the larger of two containers, while the smaller one
contains the heating elements and a propeller stirrer . Water is continuously fed from above.
The whole assembly is thermally insulated .

Liquid testing baths for higher temperature range are similar in construction. For example,
the Model 915 liquid testing bath from Isothermal Technology Ltd (1997a) operates in the
range +40 to +300 °C (with silicon oil) and with an additional cooling system from -65 to
+40 °C. The bath is equipped with a precise temperature controls system, which can be set
440 CALIBRATION AND TESTING

WATER STIRRER
TESTED THERMAL
THERMOMETERS INSULATION

- SSti v
1
SIDE TUBE _ _ )S1~` S CALIBRATION
S~~ ti CHAMBER

C i~1
HEATING ti2 - \ i.,SS HOUSING
ELEMENTS

Figure 22 .17 Water bath

with an accuracy of ±0 .01 °C and giving a stability of ±0 .002 °C at a level of 50 °C . The

uniformity of temperature distribution around the tested thermometer is about ±0 .001 °C .
Using a Cu or Al metal block with the holes for tested thermometers, ensures a temperature
uniformity of 0 .0005 °C .
Other types of liquid baths, structured for temperatures over 100 °C, are designed as in
Figure 22 .18, where liquid circulation is achieved by pumping . The bath is composed of
three containers, a larger one of tested and standard thermometers and a smaller one for the
air supply . The air-inlet tube is periodically connected to an air pump . In one-half of the
cycle the compressed air lifts the valve 1 and allows the liquid to flow from B to A with
simultaneous overflow of the excess liquid into the outer tube C . The liquid level in C then
becomes higher than in B . In the second half of the cycle the air-inlet tube is disconnected
from the pump while the atmospheric pressure is restored in the tube B . The valve 2 opens
and the liquid levels in B and C become equal . Another design of a liquid bath, applicable
up to 2000 °C, is given by Marcarino et al. (198 1) .
In fluidised bed thermostats, temperature uniformity is reached by fluidisation of
powder particles (Staffin and Rim, 1972) . They can be used in the temperature range from
about 70 to 1100 °C . A schematic diagram of a fluidised bed, which is shown in
Figure 22 .19, illustrates how the standard and tested thermometers are placed in a container
with A1203 Or Zr powder . Air is allowed in from below, through a mesh which enforces a
uniform distribution of the air-stream . The air, whose flow rate is stabilised, is heated by
electric heating elements to give a temperature uniformity in the bed of about ±0 .1 to ±0 .5 °C .
Between the fluidised powder and the heated thermometers, the heat transfer coefficient
is about 300 W/m2 K . Although fluidised bed thermostats exhibit lower accuracy than liquid
baths, their main advantages are that one powder sort covers a very large temperature range,
no danger of liquid inflammation exists and no corrosion occurs .
Fluidised bed thermostat Type FB-08C from Techne (Cambridge) Ltd (1999a), shown in
Figure 22 .20, has an application range from 50 to 700 °C with temperature stability of
±0 .2 °C at 50 °C to ±0 .3 °C at 600 °C and a working volume with a diameter of 165 nun
and length of 385 mm . The thermostat is intended for use with a computer control system .
AUXILIARY EQUIPMENT 441

AIR
INLET
TESTED THERMOMETERS

AIR FLUIDISED
POWDER

HEATING
-~ A-INNER TUBE
ELEMENT
B-SIDE TUBE

_ THERMAL
_ INSULATION
MUSHROOM
- VALVES
2
AIR GRID
C-OUTER TUBE

Figure 22.18 Liquid bath for temperatures Figure 22.19 Fluidised bed thermostat
over 100 °C

v II c "

Figure 22.20 Fluidised bed thermostat FB-08LT (courtesy ofTechne Cambridge Ltd)

22.6.3 Testing furnaces

Tubular furnaces, with metal block and interchangeable metal inserts, are commonly used
for industrial thermometer testing by the comparison method. They are intended for less
exact applications which do not require larger better specified and more expensive testing
baths. These higher specification types are described in Section 22.6.2. Blocks and inserts
are made ofmetals such as aluminium or copper, for use up to about 500 °C, bronze, for use
up to 750 °C and nickel, up to about 900 °C. All of these have high thermal conductivity .
The furnaces, which are usually vertical, are also of light construction . Their comparatively
short heating-up and cooling-down times are combined with control systems giving
temperature stability of around ±0 .05 to ±0.1 °C. Temperature uniformity in the measuring
442 CALIBRATION AND TESTING

zone of the insert is of order 0.02 to 0.2 °C, while the temperature accuracy is ±0.2 to
±0.5 °C. Various models of these furnaces, which are called "Metal Block Baths"," Block
Calibrators" or "Dry Block Baths", cover the range from -30 to 1200'C (Automatic
Systems Laboratories Ltd, 1999; Isothermal Technology Ltd, 1999a; Techne (Cambridge)
Ltd, 1999a; TMS Europe Ltd, 1997). For example, the block calibrator TECAL 650 H from
Techne (Cambridge) Ltd (I999a), shown in Figure 22 .21, has the technical data:

" temperature range: 25 °C above 98 to 650 °C,

" temperature accuracy in the measuring zone: ±0 .2 °C,
" temperature uniformity in measuring zone: ±1 °C,
" measuring zone 0 to 50 mm from base of well,
" temperature stability, after 10 min : ±0 .09 °C,
" display resolution : 0.1 °C or 0.1 °F,
" testing rate 20 °C to 600 °C : 35 minutes,
" cooling rate 600 °C to 200 °C : 30 minutes,
" dimensions : 285(h)x190(w)x425(d) mm.

Available inserts have various holes for inserting of sensors such as, for example, 5
holes of d = 6 mm or holes of d = 10, 8, 6, 4.5, and 3 nun or other combinations . All holes
are 152 mm deep. One hole in the insert is usually used for the standard sensor.
In many furnaces the whole testing instrumentation is centred around a computer, which
includes metal block temperature measurement and control as well as automatic logging and
progress monitoring of the testing and calibration of all of the sensors . (Isothermal
Technology Ltd, 1999a ; Jones and Egan, 1975; Techne (Cambridge) Ltd, 1999b).
Jones and Egan (1975) constructed a furnace, represented in Figure 22.22, with several
heating zones, which gave a uniform longitudinal temperature distribution. Testing of the
thermocouples is performed during a continuous increase of temperature using a computer
system for temperature control and data recording. Testing or calibration is made in the

=
-2

.w+

00
Figure 22.21 Block calibrator TECAL 650 H (courtesy of Techne (Cambridge) Ltd)
AUXILIARY EQUIPMENT 443

STANDARD THERMOCOUPLE
TESTED THERMOCOUPLES
, . . ,. , ,.

\FJ
N
30
1
DIGITAL VOLTMETER

H
COMPUTER

EF1
; DATA
HEATING V, RECORDER
ZONES

VOLTAGE

VERTICAL TUBE FURNACE FINAL CONTROL UNIT

Figure 22 .22 Block , . , , thermocouple testing

,
falling temperature . Programme temperature control and sampling of tested thermocouples
is fully automatic . The particular points of the emf versus temperature characteristic are
recorded after being computed by the least squares method . By these means the total testing
time is significantly shortened. In addition, measuring errors are smaller than ±3 ~LV, which
- order
The spherical furnace (Isothermal Technology Ltd, 1999a), which is a new design for
. .,-
Eight to sixteen thermocouples to be tested at a time, are tightly inserted through ceramic
tubes to a depth of 180 mm into an equalising chamber. The equalising chamber, which is

?4

Figure 22 .23 Spherical furnace Saturn (courtesy of Isothermal Technology Ltd)

444 CALIBRATION AND TESTING

concentrically located inside a spherically shaped ceramic outer chamber, ensures excellent
temperature uniformity . This furnace has the technical data :

" temperature range : 100 to 1300 °C,

" temperature accuracy : ±0 .25 to ±0.1 °C at 1000 °C,
" heating up times : 1 hour to 700 °C and 3 hours to 1300 °C,
" stabilisation time : 1 hour to ±0 .25 °C .

Furnaces based upon the heatpipe principle (Busse et al., 1975 ; Coville and Laurencier,
1975 ; Neuer and Brost, 1975) allow excellent temperature uniformity to be achieved . The
heat-pipe systems, shown in Figure 22 .24, with sheath tubes containing the standard and
tested thermocouples are confined in the chamber of an electric furnace . A layer of capillary
structures covers the internal wall of the heat-pipe . The constructional materials and liquids
used, which are compiled in Table 22 .2, are chosen on the basis of the necessary

A-A

w
\
\
1

SHEATHS

e e HEATING ELEMENT
e
e
o -Al HEAT PIPE

° CAPILLARY STRUCTURE
u LAYER
0
THERMAL INSULATION
eosoe eeeeeo

Figure 22 .24 Basic diagram of a heat-pipe furnace

Table 22 .2 Liquid and wall materials in the construction of heat-pipes

Liquid Wall material Temperature range (°C)

Freon Copper, aluminium 120- 300
Ammonia Stainless steel, nickel, aluminium 230- 330
Acetone Copper, stainless steel 230- 420
Water Copper, nickel, titanium 300- 550
Organic liquids Stainless steel, steel 400- 600
Mercury with admixtures Stainless steel 450- 800
Potassium Stainless steel, nickel 700-1000
Sodium Stainless steel, nickel, inconel 800-1350
AUXILIARY EQUIPMENT 445

temperature range. Liquid, heated by heating elements, evaporates to condense in cooler

parts of the system, which is connected with heat dissipation . The condensed vapour returns
to the warmer heat-pipe parts, due to the capillary effect. This highly efficient heat transfer
process, which proceeds continuously, contributes to a very high temperature uniformity .
Heat-pipes are also used in the construction of black body models, used for calibration of
pyrometers.

22.6.4 Black bodies

For testing of total radiation and photoelectric. pyrometers used are the radiation sources
having the properties of black bodies described in Section 8 .2. Considering the very wide
temperature range in which the presently produced pyrometers operate, different radiation
sources are offered :

" low temperature flat plates,

" low temperature cylindrical cavities,
" middle and high temperature cylindrical cavities,
" high temperature spherical cavities .

Flat plate radiation sources, as for example, offered by Omega Engineering Inc. (1999)
have a flat square heating plate of dimension 160x160 mm, working in the temperature
range of 65 to 400 °C. The plate surface is grooved and covered by special matte black
varnish, exhibiting an equivalent emissivity of E = 0.99 . The arrangement is equipped with
an automatic temperature measuring and control system.

Cylindrical cavity radiation sources consist of a metal block, with cylindrical cavity, placed
in a furnace or a thermostat . The high emissivity, reaching even E = 0.994, is obtained due
to the correct length to diameter ratio and to the uniform wall temperature (Bedford, 1972) .
Low temperature cylindrical cavity sources of Isothermal Technology Ltd (1997a), which
have a cavity of dimensions 50(d)x 150(1) mm, are constructed for use in the range -10 to
+80 °C. A series of Peltier modules is used either to heat or to cool a metal block. A
specially programmed microprocessor based controller ensures a temperature accuracy of
±2 °C with a stability of±0.1 °C and temperature uniformity of±0.5 °C.

Middle and high temperature sources, shown in Figure 22 .25 of the same producer have the
following technical data:

Middle temperature High temperature

" temperature range 50-550 °C' 150-1200 °C
" cavity diameter, d 65 mm 20 mm
" cavity length, 1 164 mm 65 mm
" emissivity, e > 0 .995 > 0 .994
" temperature stability ±0.2 °C ±0.1 °C
" temperature uniformity ±1 °C ±0.3 °C
" measurement accuracy ±2 °C ±2 °C .
446 CALIBRATION AND TESTING

The high temperature furnace from Land Infrared Ltd (Land Infrared Ltd, 1999), which
uses silicon carbide heating elements to attain a maximum temperature of 1600 °C, also has
a silicon carbide cavity with dimensions 50(d)x300(1) mm and an emissivity e = 0 .998 .

The spherical cavity furnace described in Section 22 .6 .3 can also be used as a high
temperature black body radiator for pyrometer calibration . In this case the ceramic sphere
with tubes is replaced with a hollow black sphere . The tested pyrometer, which is aimed at
the interior of the cavity of the Cyclops furnace shown in Figure 22.26, is calibrated by the
comparison method based on the readings of either a thermometer installed in the chamber
or on the readings from a standard pyrometer.

Pyrometer calibration at fixed points can be achieved using middle and high temperature
cylindrical cavity furnaces . A crucible with pure metal is placed inside the cavity where the
freezing points of Ga (29 .76 °C), In (156 .60 °C), Sn (231 .93 °C), Zn (419 .53 °C), Al
(662 .32 °C), Ag (961 .8 °C) provide the calibration points . Radiators with fixed points,
produced by Isothermal Technology Ltd (1997a), have the following technical data:

Middle temperature High temperature

" fixed points Ga, In, Sn, Zn In, Sn, Al, Ag
" cavity diameter, d 65 mm 10 mm
" cavity length, 1 130 mm 65 mm
" emissivity, £ > 0 .999±0 .001 > 0 .9996±0 .0002
" melt plateau 1 to 6 hours 30 min to 1 hour
" testing uncertainty ±0,25 °C ±0 .1 °C

Heat pipes are also used in the construction of black body radiators for the calibration of
radiation pyrometers (Aa and Gelok, 1987 ; Fischer and Gutschwager, 1997 ; Hahn et al.,
1992 ; McEvoy et al., 1997 ; Neuer and Brost, 1975 ; Zhao et al., 1992) . A heat pipe radiator
used as the freezing point of gold is referred to in Section 22 .2 and shown in Figure 22.6 .

Figure 22 .25 Model 970 black body radiation Figure 22 .26 The Cyclops spherical furnace
source for the range 150 to 1200 °C with a (courtesy of Isothermal Technology Ltd)
tested pyrometer shown (courtesy of Isothermal
Technology Ltd)
CALIBRATORS 447

22 .7 Calibrators
Portable and laboratory calibrators are commonly used in either industry or laboratories
for checking or calibrating temperature sensors as well as all associated instrumentation
including measuring devices, transmitters, amplifiers, recorders, signal plotters, conditioners
and microprocessor systems .
Typical calibrator applications in temperature measurement are :

" measurement of output signals in temperature units of standardised thermocouples and

resistance sensors,
" generation of signals corresponding to the output signals of standardised thermocouple
and resistance sensors,
" output signal measurement of amplifiers, conditioners and signal transmitters .

Calibrators, which usually include automatic compensation of reference temperature in

thermocouple measurement, may also be used for resistance sensors in 2, 3 or 4 wire
configurations . Other available functions include the selection of sensor type and
measurement range at the turn of a knob . All measured and generated values are given on
LCD indicators .
The 744 Documenting Process Calibrator (Fluke Process Tools Selection Guide, Fluke
Corp ., 1999a) is a good representative example of available portable calibrators . It is
intended for temperature measurement using 11 types of thermocouples and 8 types of
RTDs, with an accuracy of 0 .3 °C at the mid range . When using the internal reference
junction compensation, the accuracy in thermocouple measurements is 0 .2 °C poorer.
Generation of precision electrical and temperature signals for sensor simulation, battery
operation using a NiMH battery, inclusion of PC utility software and HART communication
protocol make the 744 Calibrator a versatile self-contained unit .

Laboratory calibrators, which usually cover the wide range of do and low-frequency
electrical calibration workloads, are used to give higher precision. The Fluke Corp 5500A
Calibrator (Fluke Corp ., 1999b) is equipped with optional Microsoft Windows based
5500A/CAL software for documentation simplification . It also offers traceability in
conformance with ISO 9000 quality requirements . Applications cover source simulation and
measurement of 11 different types of thermocouples, with an accuracy of ±0 .14 °C and 8
types of RTD sources, including Pt, Cu and Ni, with a precision of ±0 .03 °C . Inclusion of
an RS-232 serial port allows its connection to other instruments for more sophisticated
instrumentation configurations .

22 .8 References
Aa, H .H .M . van der and Gelok, E .J . (1987) Design and use of a small, fixed temperature heat-pipe
black-body source . 3rd International Conference TEMPMEKO 87, Sheffield, The Institute of
Measurement and Control, London, 87-98 .
Ancsin, J . (1992) Freezing and melting points of aluminium and analysis of characteristics of a fixed-
point furnace temperature. Temperature : Its Measurement and Control in Science and Industry,
6(1), American Institute of Physics, New York, 343-348 .
448 CALIBRATION AND TESTING

Anderson, R.L. (1972) The high temperature stability of platinum resistance thermometers .
Temperature : Its Measurement and Control in Science and Industry, 4(2), Instrument Society of
America, Pittsburgh, 927-934.
Arai, M. (1997) A high-temperature platinum resistance thermometer for use as practical standard,
Proc. TEMPMEKO 96. 6th International Symposium on Temperature and Thermal Measurements
in Industry and Science, Levrotto &: Bella, Torino, 135-140.
Automatic Systems Laboratories Ltd, (1999) Precision Systems for Temperature Measurement,
Calibration, Control.
Bedford, R.E . (1972) Effective emissivities of black-body cavities - a review . Temperature: Its
Measurement and Control in Science and Industry, 4(1), Instrument Society of America,
Pittsburgh, 425-434.
Bedford, R.E ., Bonnier, G., Maas, H. and Pavese, F. (1990) Techniques fbr Approximating the
International Temperature Scale of 1990, Bureau International des Poids et Mesures, Sevres .
Berry, R.L . (1982) Oxidation, stability and insulation characteristics of Rosemount standard platinum
resistance thermometers . Temperature: Its Measurement and Control in Science and Industry, 6,
Instrument Society of America, Washington, 753-762.
Bonnier, G. (1987), Calibration of Temperature Sensors: Uncertainty in the Realisation of the Scale.
Proc. TEMPMEKO 87, 3rd International Symposium on Thermal and Temperature Measurement
in Science and Industry. Sheffield, 51-68.
Busse, C .A ., Labrande, J.P . and Bassani, C. (1975) The gas-controlled heat pipe - a temperature-
pressure transducer. Temperature Measurement, Conference Series, No . 26, The Institute of
Physics, London, 428-438
Bussolino, G.C ., Righini, F. and Rosso, A. (1987) Comparison of a transfer standard pyrometer and
of high stability lamps. Proc. TEMPMEKO 87, 3rd International Symposium on Thermal and
Temperature Measurement in Science and Industry, Sheffield, 77-86.
Chattle, M.V . (1972) Platinum resistance thermometry up to the gold point. Temperature : Its
Measurement and Control in Science and Industry, 4(2), Instrument Society of America,
Pittsburgh, 907-918 .
Chattle, M .V . and Pokhodun, A.I . (1987) An intercomparison between fixed-point cells made at
VNIIM (USSR) and NPL (UK) for the realisation of the melting and triple points of indium and
cadmium. Proc . TEMPMEKO 87, 3rd International Symposium on Thermal and Temperature
Measurement in Science andIndustry, Sheffield, 41-50.
Coslovi, L. and Righini, F. (1980) Fast determination of non-linearity of photodetectors . Appl . Opt.
19,3200-3203 .
Coville, P. and Laurencier, A (1975) Intercalibration of temperature transducer with a heat pipe
furnace. Temperature Measurement, Conference Series, No . 26, The Institute of Physics London,
439-445 .
Curtis, D.I . (1972) Platinum resistance interpolation standards. Temperature : Its Measurement and
Control in Science andIndustry, 4(2) Instrument Society of America Pittsburgh, 951-961 .
Edler, F. (1997) Miniature fixed points at the melting point of palladium. Proc; TEMPMEKO 96. 6th
International Symposium on Temperature and Thermal Measurement in Industry and Science,
Levrotto & Bella, Torino, 183-188.
Evans, J.P . (1972) High temperature platinum resistance thermometry. Temperature; Its
Measurement and Control in Science and Industry, 4(2), Instrument Society of America.
Pittsburgh, 899-906.
REFERENCES 449

Evans, J.P . and Burns, G.W. (1962) A study of high temperature platinum resistance thermometers.
Temperature; Its Measurement and Control in Science and Industry, 3(1), Instrument Society of
America. Pittsburgh, 313-318 .
Fischer, J. and Gutschwager, B. (1997) On the calibration of infrared radiation thermometers against
heat pipe black bodies at temperatures below the freezing point of silver. Proc . TEMPMEKO 96.
6th International Symposium on Temperature and Thermal Measurement in Industry and Science,
Levrotto & Bella, Torino, 251-256.
Fluke Corp . (1999a) Fluke Process Control Tools Selection Guide.
Fluke Corp . (1999b) Fluke Calibration, The Total Solution .
Foster, F.B . (1972) A fixed point calibration procedure for precision platinum resistance
thermometers. Temperature : Its Measurement and Control in Science and Industry, 4(2),
Instrument Society of America, Pittsburgh, 1403-1414.
Furukawa, G.T . (1974) Investigation of freezing temperatures at National Bureau of Standards,
Aluminium Standards. J. Research Natl. Bur. Stand. USA, 477-495.
Furukawa, G.T .: (1992) Realisation of the mercury triple point. Temperature: Its Measurement and
Control in Science and Industry, 6(l), American Institute of Physics, New York, 281-285.
Furukawa, G.T. Bums, G.W., et al. (1981) Temperature research above 0 °C at the National Bureau
of Standards. Temperature Measurement in Industry and Science, Second Symposium of IMEKO
TC-12 Committee, Czechoslovak Scientific and Technical Society, Praha, 39-47.
Furukawa, G.T., Bigge, W.R ., Riddle, J.L . and Reilly, M.L . (1975) The freezing point of aluminium
as a temperature standard . Temperature Measurement, Conference Series, No . 26, The Institute of
Physics, London, 389-397.
Furukawa, G.T . and Bigge W.R . (1982) Reproducibility of some triple point of water cells.
Temperature : Its Measurement and Control in Science and Industry, 5(1), American Institute of
Physics, New York, 291-297.
Gray, G.N . and Chandon H.C (1972) Development of a comparison calibbration capability.
Temperature : Its Measurement and Control in Science and Industry, 4(2), Instrument Society of
America, Pittsburgh, 1369-1378 .
Gray, W.T . and Finch, D.J. (1972) Accuracy of temperature measurement . Temperature: Its
Measurement and Control in Science and Industry, 4(2), Instrument Society of America,
Pittsburgh, 1381-1392.
Hahn, J.W., Park, S .N ., Lee, E.S . and Rhee, C. (1992) Realisation of ITS-90 above 961,78 °C and
study on the calibration system for radiation thermometers by using a standard optical pyrometer
with silicon detector, Temperature : Its Measurement and Control in Science and Industry, 6(1),
American Institute of Physics, New York, 227-23 :? .
Hall, J.A. and Barber, C.R . (1964) Calibration of temperature measuring instruments . Notes on
Applied Science, No . 12, National Physical Laboratory, London.
Hanafy, M., Moussa, M.R . and Omar, H.L . (1982) A small transportable indium cell for use as
temperature reference. Temperature: Its Measurement and Control in Science and Industry, 5(1),
Instrument Society of America, Pittsburgh, 347-350.
Hermier; Y. and Bonnier, G. (1992) The mercury point realisation: estimate of some uncertainties.
Temperature : Its Measurement and Control in Science and Industry, 6(1), American Institute of
Physics, New York, 287-292 .
Hust, J.G . (1970) Thermal anchoring of wires in cryogenic apparatus, Rev. Sci. Inst., 41, 622-624.
Ircon Inc. (1997) IR Answers and Solutions Handbook .
Isothermal Technology Ltd - ISOTECH (1997a) Reference Manual for Temperature Products and
Services, Ed. 2.
450 CALIBRATION AND TESTING

Isothermal Technology Ltd - ISOTECH (1997b) Metal Block, Baths & Data Acquisition, Next
Generation . Ed . 2.
Isothermal Technology Ltd - ISOTECH (1999a) Temperature calibration with Isotech block baths.
Isothermal Technology Ltd - ISOTECH (1999b) Isocal-6, Ed.2 .
Jones, T.P . and Egan, T.M . (1975) The automatic calibration of thermocouples in the range 0-
1 100 °C. Temperature Measurement, Conference Series, No . 26, The Institute of Physics, London,
211-218 .
Jung, H.J . (1979) Spectral nonlinearity characteristics of low-noise silicon detectors and their
application to accurate measurement of radiant flux ratios . Metrologia, 15, 173-181 .
Kandyba, V.V . and Kowaleskij, V.A. (1956) A photoelectric spectropyrometer of high precision.
Doklady Akad. Nauk. SSSR, 108, 633-670, (in Russian) .
Kirby, C.G .M . (1982) Automation of a thermometer calibration facility. Temperature : Its
Measurement and Control in Science and Industry, 5(2), Instrument Society of America,
Pittsburgh, 1293-1298.
Kortvelyessy, L. (1987) Thermoelement Praxis, 2nd Edition, Vulkan Verlag, Essen.
Land Infrared Ltd. (1999) Landcal Calibration Sources.
Lee, R.D., (1966) The NBS photoelectric pyrometer and its use in realising the International Practical
Temperature Scale above 1063 °C . Metrologia, 2(4), 150-162 .
Lee, R.D ., Kostkowski, H.J ., Quinn, T.J ., Chandler, P.R ., Jones, T.P ., Tapping, J. and Kunz, H.
(1972) Intercomparison of the IPTS-68 above 1064 °C by four National Laboratories .
Temperature : Its Measurement and Control in Science and Industry, 4(1), Instrument Society of
America, Pittsburgh, 377-393.
Lehmann, H. and Bernhard, F. (1997) Self calibrating thermocouples. Part I: modelling, design of
prototypes, Part II : testing of a prototype, temperature stability. Proc . TEMPMEKO 96, 6th
International Symposium on Temperature and Thermal Measurement in Industry and Science,
Levrotto & Bella, Torino, 195-206.
Ma, C .K. and Lawlor, D.J . (1992) A rugged zinc freezing-point cell for industrial applications .
Temperature. Its Measurement and Control in Science and Industry, 6(1), American Institute of
Physics, New York, 339-342.
Mangum, B.W . (1989) Determination of the indium freezing point and triple point temperatures,
Metrologia, 26, 211-218.
Marcarino, P. (Editor) (1992) An interlaboratory comparison of tin and zinc fixed points.
Temperature. Its Measurement and Control in Science and Industry, 6(I), American Institute of
Physics, New York, 333-338 .
Marcarino, P., Bongiovanni, G. and Frassinetti, G. (1981) An oil bath for high accuracy thermometer
comparison Temperature Measurement in Industry and Science, Second Symposium of IMEKO,
TC-12 Committee, Czechoslovak Scientific and Technical Society, Praha, 56-61 .
McAllan, J.V . and Ammar, M.M . (1972) Comparison of the freezing points of aluminium and
antimony . Temperature : Its Measurement and Control in Science and Industry, 4(1), Instrument
Society of America, Pittsburgh, 275-285 .
McEvoy, H.C ., Machin, G., Fox, N.P ., Theocharous, E. and Hassan, I .S . (1997) An InGasAs
radiation thermometer and fixed points blackbody sources for temperature scale realisation. Proc.
TEMPMEKO 96, International Symposium on Temperature and Thermal Measurement in Industry
and Science, Levrotto & Bella, Torino, 245-250.
McLaren, E.H . (1958) The freezing points of high purity metals as precision temperature standards
Can. J Phys ., 36, Part III (Zinc), 585-598, Part IV (Indium), 1131-1147.
REFERENCES 451

McLaren, E.H . and Murdock, E.G . (1960) The freezing points of high purity metals as precision
temperature standards, Can. J. Phy., 38, Part V (Tin), 100-118 .
Neuer, G. and Brost, O. (1975) Heat pipes for the realisation of isothermal conditions at temperature
reference sources. Temperature Measurement, Conference Series, No . 26 The Institute of Physics,
London, .446-452 .
Nubbemeyer, H.G. (1992) High-temperature platinum resistance thermometers and fixed-point cells
for the realisation of the ITS-90 in the range 0 °C to 961 .78 °C . Temperature : Its Measurement and
Control in Science and Industry, 6(1), American Institute of Physics, New York, 199-202.
Nutter, G.D . (1972) A high precision automatic optical pyrometer. Temperature: Its Measurement
and Control in Science and Industry, 4(l), Instrument Society of America, Pittsburgh, 519-530.
Ohtsuka, M. and Bedford, R.E . (1982) Measurement of the thermodynamic temperature interval
between the freezing points of silver and copper, Temperature : Its Measurement and Control in
Science and Industry, 5(1), American Institute of Physics, New York, 175-181.
Omega Engineering Inc . (1999) The Temperature Handbook.
Pavese, F., Ancsin, J., Astrov, D.N ., Bonhoure, J., Bonnier, G., Furukawa, G.T ., Kemp, R.C ., Maas,
H., Rusby, R.L ., Sakurai, H. and Ling Shan-Kang (1984) An international comparison on fixed
points by means of sealed cells in the range 13 .81 K to 90 .686 K, Metrologia, 20, 127-144.
Preston-Thomas, H. (1990) The International Temperature Scale of 1990 (ITS-90) . Metrologia,
27(1), 4-10 .
Preston-Thomas, H., Bloembergen, P. and Quinn, T.J . (1990) Supplementary Information for the
International Temperature Scale of 1990, Bureau International des Poids et Mesures, Sevres .
Quinn, T.J . and Lee, R.D . (1972) Vacuum tungsten strip lamps with improved stability as radiance
temperature standards. Temperature: Its Measurement and Control in Science and Industry, 4(1),
Instrument Society of America, Pittsburgh, 395-411 .
Richardson, S .C . (1962) Temperature standards and practices in a large industrial company.
Temperature: Its Measurement and Control in Science and Industry, 3(2), Reinhold Publ . Co .,
New York, 39-44.
Roberts, P .I . (1980) The importance of thermocouple calibrations . Measurement and Control, 13(6),
213-217.
Roeser, W.F . and Wensel, H.T . (1941) Methods of testing thermocouples and the thermocouple
materials. Temperature: Its Measurement and Control in Science andIndustry, Reinhold Publ . Co .,
New York, 284-314.
Ronsin, H and Elgourdou, M. (1997) Extension of the minicell thermocouple calibration technique to
the palladium fixed point. Proc. TEMPMEKO 96, 6th International Symposium on Temperature
and ThermalMeasurement in Industry and Science, Levrotto & Bella, Torino, 189-194.
Ronsin, H. (Editor) (1992) Assessment of minicrucible fixed points for thermocouple calibration
through an international comparison . Temperature : Its Measurement and Control in Science and
Industry, 6(2), Instrument Society of America, New York, 1061-1066.
Rosso, A. and Righini, F. (1985) A new transfer - standard pyrometer, Measurement, 3, 131-136.
Ruffino, C., Righini, F. and Rosso, A. (1972) A photoelectric pyrometer with effective wavelength in
the near infrared . Temperature: Its Measurement and Control in Science and Industry, 4(1),
Instrument Society of America, Pittsburgh, 531-537.
Sparks, L.L . and Powell, R.L . (1972) Calibration of capsule platinum resistance thermometers at the
triple point of water, Temperature: Its Measurement and Control in Science and Industry, 4(2),
Instrument Society of America, Pittsburgh, 1415-1421.
452 CALIBRATION AND TESTING

Staffin, H.K . and Rim, C.H . (1972) Calibration of temperature sensors between 538 °C (1000 °F) and
1092 °C (2000 °F) in air fluidised solids. Temperature: Its Measurement and Control in Science
and Industry, 4(2), Instrument Society of America, Pittsburgh, 1359-1368 .
Stimson, H.F . (1956) Precision resistance thermometry and fixed points . Temperature: Its
Measurement and Control in Science and Industry, 2, Reinhold, Publ, Co, New York, 141-168 .
Strouse, G.F ., Mangum, B.W., Pokhodun, A.I . and Moiseeva, N.P . (1992) Investigation of high-
temperature platinum resistance thermometers at temperatures up to 962 °C, and in some cases,
1064 °C . Temperature: Its Measurement and Control in Science and Industry, 6(1), American
Institute of Physics, New York, 389-394.
Techne (Cambridge) Ltd (1999a), Temperature Calibration Equipment.
Techne (Cambridge) Ltd (1999b), Tecal Series Block Calibrators .
TMS Europe Ltd (1997), Peaks ofMeasurement Perfection .
Trabold, W.G . (1962) An industrial thermocouple calibration facility. Temperature: Its Measurement
and Control in Science and Industry, 3(1), Instrument Society of America, Pittsburgh, 45-50.
Zhao Qi, Chen Biagquan, Yuan Zhundong and Duan Yuning (1992) The realisation of the aluminium
freezing point by pyrometry. Temperature : Its Measurement and Control in Science and Industry,
6(1), American Institute of Physics, New York, 353-355.
Zhao Qi, Yuan Zhundong and Duan Yuning (1990) The precise DC photoelectric pyrometer. Proc .
TEMPMEKO 90, 4th Symposium on Temperature and Thermal Measurement in Industry and
Science, Helsinki, 307-314.