Asme Ptc 19.3 Temp Msrmt

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

REAFFIRMED 2004 FOR CURRENT COMMITTEE PERSONNEL PLEASE E-MAIL [email protected]

Copyright ASME International Provided by IHS under license with ASME No reproduction or networking permitted without license from IHS

Licensee=OMAN & THAILAND LOCATION/5940240002 Not for Resale, 03/04/2007 23:48:04 MST

PART

Temperature Measurement

8

INSTRUMENTS

I

AND APPARATUS

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



Library

of Congress

Catalog

No. 74-76612

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

No part of this document may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publisher.

Copyright,

Q 1974, by

The American Society of Mechanical ‘ Engineers Rinted in the United States of America



FOREWORD The Scope of the work of Technical Committee No. 19 on Instruments andApparatus is to describe the various types of instruments and methods of measurement likely to be prescribed in any of the ASUE Performance Test Codes. Such details as the limits and sources of error, method of calibration, precautions, etc., as will determine their range of application are given. Only the methods of measurement and instruments, including instructions for their use, specified in the individual test codes are mandatory. Other methods of measurement and instruments, that may be treated in the Supplements on Instruments and Apparatus, shall not be used unless agreeable to all the parties to the test. This Supplement on Instruments and Apparatus, Part 3 on Temperature Measurement, replaces an older one published during the period from 19521961. Since that time the technology of temperature measurement has so changed and broadened that the earlier material has become obsolete. This necessitated a complete revision on the Supplement resulting in the currently expanded and more comprehensive document. In accordance with the established policy of the American Society of Mechanical Engineers concerning the inclusion of metric (SI or International System) units in all ASME publications, this document includes an Appendix of appropriate conversion factors which will enable the user to utilize both systems. These conversions are listed in the Appendix as they first appear throughout the Supplement. Extensive use was made of the “ASME Orientation and Guide forUseof Metric Units, Third Edition” and The ASTM Metric Practice Guide E380-92.” These two publications should be consulted for additional material concerning conversions from the US system to SI units. This Edition was approved by the Performance Test Codes Committee on July 12, 19’73. It was approved and adopted by the Council of the Society by action of the Board on Codes and Standards on May 29, !974.

. ..

111

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



PERSONNEL OF PERFORMANCE TEST CODES COMMITTEE NO. 19.3 ON INSTRUMENTS AND APPARATUS

R. F. Abrahamsen, Chairmm K. W. Woodfield, Secretary

Il.

F. Abrahamsen, oratory,

Manager,

Technical and Administrative Services, Kreisinger Research Engineering Inc., 1000 Prospect Hill Road, Windsor, Ct. 06095

Combustion

R. P- Benedict, Fellow Engineer, STDE, 9175, Philadelphia, Pa. 19113 J. T. Callahan,

Research

sion, Applied G. 0. Nutter,

Mechanical

Physics

Assistant

Engineer,

Department,

Director,

Professor

of Mechanical

Electric

Corp., Lester

Ship Engineering

Branch Post Office

Center,

Philadelphia

Divi-

Pa. 19112

Systems Center,

University

of Wisconsin - Madi-

Wi. 53706 Engineering,

General Moton

Institute,

1700 West Third

Mi. 48502

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

Avenue, Flint,

Naval

Philadelphia,

Instrumentation

son, 1500 Johnson Drive, Madison, K. W. Woodfield,

Westinghouse

Lab-

V



Personnel

of Performance

K. C. Cotton, J. H. Fernandes,

Test Codes Committee

Chairman Vice Chairman

Hilke Knoedler Leung Light

L. C. Neate

C. A. Dewey

J. L. E. L. Paul F. H.

V. F. Estcourt

S. W. Lovejoy

C. B. Scharp

A. S. Grimes

W. G. McLean

J. F. Sebald

K. G. Grothues

S. L. Morse

J. C. Westcott

R. P. Benedict W. A. Crandall R. C. Dannettel

J. W. Murdock

vi

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



W. C. Osborne W. A. Pollock J. H. Potter

ASME

Performance Supplement Instruments and Part

Temperature

Test Codes on Apparatus

3 Measurement

TABLE OFCONTENTS Chapter 1 2 3

Pages 1 GENERAL ...................................................................... 12 RADIATION THERMOMETERS .................................. 17 THERMOCOUPLE THERMOMETERS ........................ Section A, Thermocouples ........................................ 17 Section B, Instrumentation ........................................ 27 36 RESISTANCE THERMOMETERS ................................ 44 LIQUID-IN-GLASS THERMOMETERS ........................ FILLED SYSTEM THERMOMETERS .......................... 55 70 OPTICAL PYROMETERS ............................................ 86 BIMETALLIC THERMOMETERS ................................ 91 CALIBRATION OF INSTRUMENTS ............................ APPENDIX ..................................................................... 134

CHAPTER 1, ,GENERAL CONTENTS

GENERAL Par.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

GENERAL: scope ....................................................................... Introduction ........................................................... TEMPERATURE SCALES ....................................... INSTRUMENTS ........................................................... ACCESSORIES: Wells ........................................................................ Other Accessories ............................................... INSTALLATION ......................................................... SOURCES OF ERROR: Introduction ........................................................... Conduction Error ................................................... Radiation Error.. ..................................................... Heat Transfer at Low Velocity ........................... Aerodynamic Heating Effect ................................ Heat Transfer at High Velocity.. ......................... Gradient Error ....................................................... Dynamic Error ....................................................... CONCLUSIONS ........................................................... REFERENCES ...........................................................

Scope 1 The purpose of this chapter is to present a summary discussion of temperature measurement as related to Performance Test Code work with particular emphasis on basic sources of error and means for

8 20 21

coping with them.

25

Introduction

:8 28 29 31 34 36 38 39

2 Measurement of temperature is generally considered to be one of the simplest and most accurate measurements

performed in engineering.

cidedly a misconception. measurement

Accurate

under some conditions

with our present knowledge.

1



This is de-

temperature is impossible

Under many of the con-

ASME ditions

met in Performance

sired accuracy

Test

cautions

in the selection, measuring

of suitable

installation

instruments;

5

pre

and use of

available

(o)

Radiation

temperature

measurements

encountered

in installation

centrate a definite

accurate

Specific

directions

system,

of the body. They conused to intercept

from a body whose temperature

is being measured; a

temperature sensitive

usually

element,

force measuring

and pre-

lb)

are given in

A Thermocouple

temperature

measuring

Thermometer (Chapter 3) is a system comprising

converting

scales

electrical

(centi-

the boiling

and freezing

ard atmospheric

the interval

scale,

and electrical

point is

formulas:

Thermometers (Chapter 4) are tem-

expansive

C = 5/9 (F - 32) where F = reading in deg Fahrenheit C = reading in deg Celsius.

(e)

conductors for operatively

connecting

with an

liquid. System Thermometers (Chapter 6) are measuring

instruments

in which the

change in volume of a liquid, a change in pressure of a gas, or the change in vapor pressure of a volatile liquid is used as a means of temperature meas-

FACTORS

temperature

Filled

temperature

readings

urement. They consist

from

of an all metal assembly

comprised of a bulb, capillary tube and Bourdon tube, provided with a temperature responsive fill.

one scale to another are given in the Appendix.* *Whenever U.S. Customary units are used in this suppliment the SI equivalent may be calculated by using the conversion factors listed in the Appendix.

(f) Optical Pyrometers (Chapter 7) are temperature measuring instruments in which the brightness

2 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


in which the electri-

is used as a means of temperature

with the bulb and a portion of the stem filled

C t 32

for converting

the

ferential expansion of a liquid in a closed glass systern is used as a means of temperature measurement. They consist of a thin-walled glass bulb attached to a glass capillary stem closed at the opposite end,

temperature on one scale may be converted to the corresponding reading on the other scale by use

Tables

and

ass Thermometers (Chapter 5) are (d) Liquid&G1 temperature measuring instruments in which the dif-

scales is called a degree. The reading for a given

CONVERSION

units,

connecting

the two.

marked 100, and the freezing point is marked 0. Each of the 180 or 100 divisions in the respective

F = 9/5

temperature

measurement. They consist of a sensing element called a resistor, a resistance measuring instrument,

between the same fixed points is di-

of the following

for

3.1A.l

Resistance

cal resistance

point is marked 212, and the

vided into 100 equal parts; the boiling

emf to equivalent

perature measuring instruments

into 180 equal

point is marked 32. In the Celsius

the interval

(c)

between

points of water at stand-

pressure is divided

parts; the boiling freezing

scale,

which

a printed scale for

conductors for operatively

two. (See Fig.

grade) temperature scales. A detailed discussion of these and other scales is given in Chapter 9. In the Fahrenheit

a tempera-

force (emf), a device

sensing emf which includes

and the Celsius

instru-

ture sensing element called a thermocouple

SCALES

4 There are in general use two temperature

device,

ment.

produces an electromotive

known as the Fahrenheit

a thermo-

and a measuring

such as an electromotive

subsequent chapters for each of the various types of temperature measuring instruments. TEMPERATURE

and con-

portion of the energy radiated

couple or a thermopile;

is

2) are tem-

in which the intensity

emitted from a body is used as a

sist of an optical

or use of the temperature

in usage of the instruments

Measurement.

measure of the temperature

tempera-

with such instruments

are availabIe

The chapter

Thermometers (Chapter

perature measuring instruments

for tempera-

in obtaining

conditions.

tus, Part 3, Temperature

ture to a closer degree of accuracy than is required in some of the tests considered in the Performance Test Codes. The difficulty

types of instruments

numbers refer to chapters in the ASME Performance Test Codes, Supplement on Instruments and Appara-

and in the proper

are capable of indicating

instruments.

following

of the radiation

3 Some of the instruments

cautions

The

for use under appropriate

under such conditions.

measuring

CODES INSTRUMENTS

of temperature

interpretation of the results obtained with them. In some cases an arbitrarily standardized method is prescribed in the Performance Test Codes which is to be followed in making temperature measurements

ture measurement

TEST

Code work, the de-

in the measurement

can be obtained only by observance temperature

PERFORMANCE


INSTRUMENTS of radiation

AND

APPARATUS ACCESSORIES

in a very narrow band of wavelengths

emitted by a source, the temperature of which is to be measured photometrically matched against the brightness

of a calibrated

source,

Wells*

is used as a

8 Introduction.

means of temperature measurement. They consist of a telescope, a calibrated lamp, a filter to provide for viewing nearly monochromatic readout device,

and usually

radiation,

an absorption

a

glass

instruments

differential

of two metals

expansion

means of temperature

in which the is used as a

measurement.

They consist

of

9 Thermometer

wells

an indicating or recording device, a sensing element called a bimetallic thermometer bulb, and a means

temperature

for operatively

ties of 300 fps or less,

connecting

are used in measuring

of a moving fluid in a conduit,

the stream exerts an appreciable

the two.

the

where

force. For veloci-

tapered thermometer wells

of the design shown in Fig. given in Table

6 The above instruments are those which are recommended for ASME Performance Test Code work for the measurement of temperature under appropriate conditions.

ele-

a well may be used, which by definition is a pressure tight receptacle adapted to receive a temperature sensing element and provided with external threads or other means for tight pressure attachment to a vessel [l].**

Bimetallic Thermometers(Chapter 8) are

(g)

measurements

temperature

Test Code work the sensitive

ment cannot be placed directly into the medium whose temperature is to be measured. In such cases

filter.

temperature measuring

In many

in Performance

1.1, and of dimensions

1.3, shall be used. For velocities

in

excess of 300 fps, a fixed beam type thermometer well is recommended [7].

when used

7 The recommended ranges of use for these temperature measuring instruments when properly installed are indicated in Table 1.1:

TABLE

l.lRECOMMENDEDTEMPERATURE

I

Chtz+er

Ran e of Use,

Tee (a)

RANGES

I Beg F

.

Radiation thermometers

2

Ambient and above

3

-300 to t4500

4

450 to t1950

(b)

Thermocouple

(c)

Resistance

(d)

Liquid-in-Glass thermometers

5

-328 to t1110

(e)

Filled System thermometers

6

400 to t1200

(f)

Optical pyrometers

7

(g)

Bimetallic

thermometers thermometers

thermometers

Above 1300

8

-200 to t 800 FIG.

1.1

PERFORMANCE

TEST

CODE

THERMOMETER WELLS 10 Attachment *At the time of the current revision, ASME Ad Hoc &;;rittee PR 51 is writing a new standard for thermo-

l *N;mbers

iu b a kets designate

References


may be made in any

Vessel

or Piping

Codes. Any material

approved by

these Codes for the intended service may be used. Where materials are specified for the purposes of

at end

of chapter, thus [Ij.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

to the vessel

manner approved by the ASME Boiler and Pressure

3 Licensee=OMAN & THAILAND LOCATION/5940240002 Not for Resale, 03/04/2007 23:48:04 MST

ASME illustrating that 11

the example,

these materials

no inference

For the experimental

and theoretical

Strength

Versus

Ref. [2] should

required to produce adequate well strength tend to

Table

1.2 is not all inclusive,

Design

= length of well as given in Fig.

= modulus of elasticity

y

use temperature, psi = specific weight of well material

procedure is to enable the user to determine well selected for thermometry considerations strong enough to withstand

specific

at use

lb per cu in.

f,

1.4

= 2.64;

(2)

where

if a is

application

at

The wake or Strouhal frequency is given by:

but indicates

f,

= wake frequency,

V

= fluid velocity, fps = diameter at tip (Fig.

The purpose of this design

Procedure.

1.1, in.

of well material

A, = a constant obtained from Table

that thermometer.well design methods must carefully balance these factors so that accuracy is compromised a minimum when using a well of adequate strength. 14

L

E

temperature,

reduce the accuracy and response of the temperature measurement, as shown in Table 1.2 below. I3

CODES

f n -- natural frequency of the well at use temperature, cycles per set

bases of

Those factors

Measurement.

TEST where

is intended

are preferred.

the design procedure set forth herein, be consulted. I2

PERFORMANCE

B con-

cycles per set l.l),

in.

The ratio of wake to natural frequency (f /f ) shall not exceed 0.8, and when this conditYon’is

ditions of temperature, pressure, velocity and vibration. Well failures are caused by forces imposed bv

met, the Magnification

static pressure, steady state flow, and vibration. Separate evaluations of each of the above effects

namic to static

Factor,

amplitude

relationship

of dy-

is given by:

should be made in order to determine the limiting effects

FMz

This design procedure does not allow for

due to corrosion

or erosion.

15 The natural frequency

i=mm)2=iJ2

For r 5 0.8

of a well desieed

where

in

accordance with Fig. 1.1 and of the dimensions given in Table 1.3 is given by the following equa-

FM = magnification factor, dimensionless r = frequency ratio, (f,Jf,,), dimensionless

tion:

16 Stress Analysis.

The maximum pressure that

a thermometer well can withstand rial at a given temperature TABLE

1.2 FACTORS

Factor

THAT

INFLUENCE

STRENGTH

AND MEASUREMENT

Ideal for Strength

Long

Short

Conductivity

errors reduced. Active

tion of thermometer must be in flow stream.

por-

Impingement

Reduced conductivity

force reduced.

Higher natural frequency.

Thick

Thin

Thickness

Faster

shall be computed from

Ideal for Measurement

Length

for a given mate-

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

condition.

Greater moment of inertia,

loss.

less stress.

Higher natural frequency.

response.

Mass Low

High

Velocity Increased Faster

Reduces

heat transfer.

impingement

4 Copyright ASME International Provided by IHS under license with ASME No reproduction or networking permitted without license from IHS

forces.

Karman trail vortex frequency.

response.


Lower

INSTRUIIENTS

AND

APPARATUS

the following:

TABLE P =K,S

P = maximum allowable

Nominal

(4) Stress Constant

static gage pressure,

psi

K,

= a stress

Vessel

l/4

Kl

S = allowable stress for material at operating temperature as given in the ASME Boiler and Pressure

1.5 VALUES OF STRESS CONSTANTS

3/8

0.412 37.5 0.116

K2

KS

Size of Sensing Element

0.334 42.3 0.205

9/16

11/16

0.223 46.8 0.389

0.202 48.7 0.548

7/8 0.155 50.1 0.864

or Piping Codes, psi

constant obtained from Table

1.5. other limitation is one of steady state stress considerations, as given by the following equation:

TABLE

1.3 WELL DIMENSIONS, IN IN. Nominol


_&

L mox-

Dimension l/4

3/8

9/16

1 l/16

7/a

13/16 5/8 0.254 0.262

15/16 3/4 0.379 0.387

l-1/8 15/16 0.566 0.575

l-1/4 l-1/16 0.691 0.700

l-7/16 l-1/4 0.879 0.888

v

Jto-&PO) + 1

(5)

FM

where A (minimum) B (minimum) d (minimum) d (maximum)

TABLE

2-l/2 4-l/2 7-l/2 10-l/2 16 24

V = fluid

velocity, fps II= specific volume of the fluid, cu ft per lb S = allowable stress for material at operat-

1.4 VALUES OF Kf

Nominal we! ,Llengrh n.

value of L (as shown in Fig. 1.1) for a given service, in.

L mox = maximum

ing temperature Boiler


Codes, l/4

3/8

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

2.06 2.07 2.08 2.09 2.09 2.09

9/16

2.42 2.45 2.46 2.47 2.47 2.47

1 l/16

2.97 3.01 3.05 3.06 3.07 3.07

7/8

3.32 3.39 3.44 3.46 3.47 3.48

P 0 = static

as given in the ASME

and Pressure

Vessel

or Piping

psi

operating gage pressure, psi factor as computed from

FM = magnification

3.84 3.96 4.03 4.06 4.08 4.09

Eq. (3) K,,

Ks = stress

18

Example

determined

constants

Problem.

obtained from Table

Assume

that

1.5

it has been

on the basis of thermometry

considera-

tions that a 4% in. well is required to accommodate a 9/16

17 The maximum length that a thermometer well

in. sensing

ture of superheated

can be made for a given service is dependent upon both vibratory and steady state stress. The neces-

element to measure the temporasteam at 2400 psig,

1050°F,

sity for keeping the frequency ratio at 0.8 or less

flowing at a velocity of 300 fps. If the well material is to be Type 321 stainless steel, will the well be

imposes one limitation

safe?

on maximum length. The

Solution:

Step l-Obtain

the necessary

data as follows:

V

E 70

F

Specific

volume of superheated

Modulus of elasticity Specific

s’

Value

Item

Symbol

Allowable

(a)

0.3353

cu ft per lb

28.0 x 106 psi 0.290 lb per cu in.

weight of metal at 7OoF

13,100

stress at 1050°F

Ratio of frequency Step 2-Frequency

steam

at 70??

psi

0.918

at 1050 to 709

Calculations Natural

frequency

(Eq. (1))



Reference

131 [41 (41 [51

[61

ASME

PERFORMANCE

TEST

CODES

f, = 0.918 x 1461 = 1341 cps (b)

Strouhal frequency f

(c)

uI =

2.64 x 300 = 845 cps 15/16

Magnification r = f,/f, FM =

Step 3-Stress

(Eq. (211

factor (Eq. (311

= 845/1341 (0.63012

1 -

= 0.630 < 0.8 (satisfactory)

_ o 658

(0.63012

’

Calculation (a)

Maximum pressure (Eq. (41)

P P 0.223 x 13,100 = 2921 > 2400 (satisfactory) (b) Maximum stress length (Eq. (5))

L

max=

46.8 300

.3353 (13,100 - 0.389 x 24001 1 + 0.658

= 7.7 in. > 4% in. b?isfuctory~ Result:.

All the requirements

eter well is satisfactory

19 Thermometer

wells

expected to satisfy

for frequency

and stress have been met, therefore,

as shown in Fig.

l-l,

Several

are

Increased

of central

stations

temperature-pressure

(a) By surrounding the sensing junction with one or more coaxial tubes mounted in the direction of

ratings

now await advances in met-

Th is arrangement screens the gas flow [8,9,101. sensor from radiation exchange with the surrounding surfaces.

allurgy. Such materials when available will also increase the ratings of the thermometer wells. For services

where these thermometer wells

are not

the use of a fixed beam type of well

(b) By increasing

described in Ref. [7] is recommended. Interest now centers on velocities of 300 fps or less. Higher velocities

give rise to considerable

ferences between staguation peratures.

schemes may be employed as follows:

95 percent of the present well

problems.

now suitable,

For velocities

and static

in excess ‘

dif-

plished

tem-

of 300 fps the

A special

(c)

radiation

higher or

arrangement has been found

effective,in

minimizing

radiation

losses where

prevent the use of the coaxial-

6 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


onto

1141. This

For a more complete discussion see Ref. [151.

under such conditions.

with a low

shield mounted directly

the junction

space limitations type screens.

lower temperatures than that of the medium in which it is immersed, accessories may be used to minimize from radiation

is em-

1133.

By covering the sensing junction

emissivity

20 When it is necessary to place a temperature sensing element in a gas or vapor at a location

errors arising

type pyrom-

case of this principle

ployed in the sonic-flow thermocouple pyrometer in which the gas flow over the sensor is maintained at

Other Accessories

at materially

heat transfer

by using a suction or aspirating

eter 111,121.

sonic velocity

surfaces

the convective

rate from the gas to the sensor thereby minimizing the effect of radiation losses. This may be accom-

fixed beam type of well is recommended.

where it can “see”

this thermom-

for the intended service.


of this subject,

INSTRUMENTS

AND

measurements on piping, the leads should be wound around the eipe for at least four turns ad-

INSTALLATION

21 Details of installation applying specifically to a given type of instrument are treated in a subsequent chapter for the particular general considerations 22

instrument.

APPARATUS

jacent to the junction.

Certain

are given here.

SOURCES

Where the sensing element is immersed in the

substance

whose temperature

is to be measured,

it

should be so located as to acquire and maintain

as

nearly as possible If the possibility gradients exists, a sufficient install

the temperature

Introduction 25

of the substance.

of the measurement

sensing

element indicates

and, even under steady-state

its

con-

ditions, the temperature of the element may not be that.of the fluid or solid with which it is in contact.

number of elements and to locate and according

Any temperature

own temperature

of stratification, stagnation, or care must be exercised to choose

them properly,

OF ERROR

In general, temperature

to the requirements

errors are more pronounced

when dealing with the gaseous phase as compared

to be made.

with the liquid or solid phase. However,

23 Wherever possible, methods which do not involve the use of thermometer wells between medium and instrument shall be used. Unfortunately, not always possible to do this. Thermometer

or moving with relatively low velocity, the temperature indicated by the temperature sensing element for steady-state conditions is a result of a balance

shall be designed lined in Pars.

peratures in liquids

it is wells

in accord with the principles

8 to 19. In addition,

out-

The part of the well projecting

to eliminate (b)

heat transfer

or solids.

If a fluid is at rest

the fluid,

precautions must be observed, particularly when the temperature being measured differs by more than SOoF from that of the surroundings: (a)

when measuring tem-

of convective heat transfer between the element and and heat transfer by conduction and radi-

the following

ation between the element and its surroundings. a gas stream moving at high velocity

For

(above Mach

0.3), however, the temperature determination becomes more difficult because of the aerodynamic

beyond or out-

side the vessel must be as small as possible

insignificant

errors are

not necessarily

heating effect. The following paragraphs describe typical sources of error and means by which errors

so as

to or from surroundings.

can be determined

or reduced.

The exposed parts of the well shall be cov-

ered with a suitable

thermal insulating

The vessel wall shall be insulated

material.

tance from the thermometer well if the vessel already insulated, materially

and if such insulation

affect the temperature

Conduction

for some dis-

26 Conduction error, commonly called immersion error, may be present whenever a temperature gradient exists in the temperature sensor (e.g., in the wires of a thermocouple between its measuring

is not

will not

of the medium to

be measured.

junction and point of attachment). Recommended installation practice for thermometer wells is described in Par. 23 which, if followed, will reduce the conduction error. The following relation may be used with good accuracy for determining the extent of the conduction error if this is the only error of

(c) The sensing element should be in intimate thermal contact with the well. This may be accom--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

plished by direct contact, as with thermocouples, by heat-transfer filling media or metallic sleeves other thermometers

that may be inserted

Error

for

in wells.

significance:

24 In the measurement of surface temperature the extent of insertion of the sensing element will obviously

be limited

whose temperature

by the thickness

of the material

is being measured.

couples are generally

used for this purpose. To

aid in bringing the measuring perature of the material

(6)

Thermo-

junction

surface,

where

to the tem-

the junction

should

be peened into position and a portion of the thermocouple leads should be in intimate contact with the material

surface.

T,,

= static temperature

Ti

= temperature

of the gas, deg F

indicated

sensor, deg F

In the case of surface temperature



by temperature

[161

ASME

T,

z

PERFORMANCE

= immersion length of temperature sensor, ft

m

= (hp/kd

h

I

Btu/hr

the

be simplified

as follows:

ft2 deg F

= perimeter of temperature

k

I thermal conductivity

qr = 0.1714

sensor, ft

ft deg F

Heat

area of tempera-

28

ture sensor, ft2 Since the hyperbolic

cs

(%);1’

(8)

cosine of the

Transfer

at Low

Velocity

Consider the case of a temperature sensor

exposed to a low velocity (i.e., no aerodynamic heating) gas stream with the sensor experiencing

mL product in-

both radiation

creases as the product itself increases, it follows that the larger mL becomes the closer the indicated temperature, Ti, approaches the static temperature of the gas, Tsg, (i.e., the conduction error is reduced). As a consequence,

A, [go)‘-

of temperature sensor,

= conduction cross-sectional

a

by the surrounding surfaces,

configuration factor (FA) from the sensor to its surroundings is equal to unity. As a result, Eq. (7) may

of heat transfer,

P

Btu/hr

sor is intercepted

ft”

coefficient

(c,) of the sen-

sor. Also, since all the energy radiated by the sen-

L

convective

CODES

equal to the normal total emissivity

temperature at point of attachment (e.g. vessel wall), deg F

l/2,

TEST

steady-state

and conduction

effects.

For the

condition between the flowing gas and

the sensor, heat transfer by convection must equal the rate of heat transfer by radiation and conduction.

any means of increasing

mL product will result in a decreased conduction

This equilibrium

error.

follows:

condition may be written as

hA, (T,, - Ti) = 0.1714 es A, Radiation 27

If the temperature

Error

sensor can “see”

surfaces

which are at either higher or lower temperatures than the sensor itself,

net radiant interchange

take place. The sensor will

experience

or loss of heat by radiation

and, therefore,

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

nificant

will

a net gain

It can be seen from the above expression

a sig-

radiation

error in temperature measurement may result.

and conduction

effects

that as the

are reduced, the

Ti, will approach the

temperature of the sensor,

static temperature of the gas, T,,. Means of reducing the radiation effect are described in Par. 20. There is alsb a more complete discussion of radia-

The net radiant interchange may be determined by means of the following relation:

tion and related factors in Chapter 2, Radiation qr = 0.1714

Fc FA A, [(g&(

&)j

(7)

El71

Thermometers.

where

Aerodynamic

qr c net rate of radiant

interchange,

F c = effective emissivity,

Btu/hr

29

Aerodynamic

Heating

heating

Effect

is caused by localized

stagnation of the moving gas stream in the immediate vicinity of the temperature sensor. As a result,

dimensionless

FA= configuration factor, dimensionless

the temperature

AS” surface area of temperature sensor, ft2

be higher than the static temperature of the gas stream. Static temperature is defined as the temper-

Ti = temperature indicated by temperature

as indicated

ature of the gas,stream as indicated by a temperature sensing element moving with the same velocity

sensor, R

as the gas and with isentropic Tr = mean temperature

conditions

existing

of surrounding surfaces, R *Esuations (8). (9) and (12) following, are valid only for th’e case where there are no radiatik absorbing gases present. For the case where absorbing gases such as water vapor or carbon dioxide are present, see Ref. [ 181.

If the surface area of the sensor is small with respect to the area of the surrounding surfaces, is the usual case, the effective

emissivity

as

(FJ is


by the sensor tends to


INSTRUMENTS at the temperature

sensing element.

AND

ation effects, the temperature, Ti, indicated by the sensor may differ from the adiabatic temperature. In

In the case

where radiation, conduction, and aerodynamic heating occur simultaneously, the temperature indicated by the temperature sensor will be dependent upon the corresponding transfer effects.

magnitudes

localized

called “total,”

of these three heat

at the temperature temperature

sensor, the

generally

have different

values.

For this case, ,the

applicable

steady-state

relation

is as follows:

heAs (To- T,) = 0.1714 ESAS

of the gas stream is

or “stagnation,”

temperature,

T,. [(s)’

The total temperature would be higher than the static temperature because of the conversion of kinetic

energy to internal

manner:

T,, = Vz/2Jgcc,,

I201

heat transfer coefficient,

[201

(10)

The effective coefficient, he, which is primarily dependent upon flow regime, geometric configuration,

V = gas velocity, = mechanical

and orientation may be calculated through use of ap propriate convection correlations. Further infonnation may be obtained from Refs. [22, 241. When cal-

ft/sec equivalent

of heat=778ft

lbf/Btu

culating gc = dimensional

constant = 32.1740

lb,ft/lbrsece

the temperature

the adiabatic

the static 30 Whenever the kinetic energy of the gas stream is reduced, the conversion of kinetic energy to

in heat transfer

2200ft/sec, ference

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

The rise in temperature,

with velocities

(T, - Tsg), of the stag-

of 3OOft/sec

such as stagnation

or less.

thermocouple

Gradient

34

[21, 22, 23, 261.

at High

heating at high velocities.

In temperature

there are velocity,

thermometers,

Error

measurements density,

in a system where

or temperature

gradients,

will be taken into account if the tem-

perature measurement

Velocity

31 In the case where aerodynamic heating occurs and the temperature sensor has conduction and radi-

is used to determine the

* See footnote under Eq. 8.


A recovery

should be used, 17, 8, 211.

such gradients Heat Transfer

the dif-

the signif-

at this velocity or lower. For higher velocities, a fixed beam type of well or temperature measuring devices designed expressly for high velocity flow,

1201

means of Eq. (10). Th e recovery factor is primarily dependent on geometric configuration, orientation, and Mach number. For a more complete discussion of see Refs.

Mach 2 at 40%‘,

factor for air of 0.65 should be used for these wells

nated portion of the gas stream during isentropic slowing of the gas stream may be calculated by

this subject

approximately

is 400 deg F, thus emphasizing

in temof

33 The standardized Performance Test Code wells dimensioned in Table 1.3 are recommended for use

the

(11)

to 40 deg F. At avelocity

icance of aerodynamic

converted kinetic energy of the gas stream is defined in terms of a “recovery factor,” r, as follows:

- Ts+$ / (Tt - T,,)

of llOOft/sec,

Mac.h 1 at 4Q%‘, the difference

perature increases

transfer, an adiabatic condition, the temperature which the sensor then assumes is defined as the “adiabatic temperature,” T,. For convenience, the

r=(T,

of the gas.

is only 7 deg F, but at a velocity

from the local-

sensor to “recover”

temperature

approximately

ized region to the surrounding gas stream, as well as to the sensor. If the sensor experiences no heat

of a temperature

being the unknown quan-

32 At the relatively low velocity of 300 ft/sec the difference between the total and static temperatures

internal energy is manifested by a localized rise in gas temperature at the temperature sensor. This temperature rise results

error for the above case,

temperature,

tity, is determined through use of Eq. (12). Equations (10) and (11) may then be used for determining

cp t specific heat at constant pressure, Btu/lb, deg F

ability

@+)‘(12)

]+h

h, = effective convective Btu/hr ftz OF.

where

J

_(cJ

where

energy. These two temper-

atures are related in the following

Tt -

Ti, Tsg, Tt and Ta

other words, all four temperatures

If the moving gas stream is brought

to rest isentropically resulting

APPARATUS


ASME

PERFORMANCE

37

The ability

of a temperature

sensing element

to respond to a change in temperature given in terms of its “time

de-

posed step-change.

is related to the

Therefore,

change in temperature

equation:

stant. As a result,

Z (C, T VA) C (p

temperature

VA)

following

(a)

where

error due to dynamic

measurement

Ratio of sensor surface area to sensor mass. As

transfer

= local stagnation temperature,

P

= local stream density,

lbm/ft’

V

= local stream velocity,

ft/sec

coefficient.

increases,

As the heat

the time constant

decreases.

(cl

Thermal conductivity of the sensor material. As the thermal conductivity increases, the time constant decreases.

(d

“F.

Specific

(e)

Mechanical accessory

Further

ftz

characteristics

are considered

al chapters dealing with various

Error

of the

equipment.

is given in Refs.

features

As the spe-

the time constant also in-

and electrical measuring

information

and specific Dynamic

heat of the sensor material.

cific heat increases, creases.

= local area represented by measuring station,

heat transfer

coefficient

“F

T

A

the time constant decreases.

“F

Cp = specific heat at constant pressure, evaluated at the local stagnation temperature, Btu/lbm

system depends on the

“F

= specific heat at constant pressure, evaluated at the bulk temperature, Btu/lbm

for elements

The response of a

five major parameters:

(b) Convective (Cp)b

sens-

than if it had a long time con-

the ratio increases,

T,, = bulk stagnation temperature,

is the

a temperature

temperature

having long time constants.

(13)

=

This

ing element having a short time constant for a given physical situation will respond more rapidly to a

response becomes more significant Tb (Cp)b

is typically

constant.”

time required for the element to change in temperature an amount equal to 63.2 percent of the im-

sired. For each station, the local temperature, density, velocity, and flow area shall be evaluated. though the following

will be

dition to being out of phase it will also be of differ-

35 A number of measuring stations and sensing elements shall be selected depending on the rela-

local quantities

in general,

ent magnitude.

tice the desired result can be obtained by making sets of measurements at a sufficient number of points to permit numerical integration.

The value of the bulk temperature

of the sensor,

out of phase with the temperature of the medium being measured during a transient condition. In ad-

temperature realized if the flow could be interrupted and the material thoroughly mixed without gain or loss in energy (i.e ., an adiabatic situation). In prac-

and the accuracy

CODES

the temperature

energy of the medium. In such cases the value of the bulk temperature is desired, which would be the

tive magnitude of the gradients

TEST

[23, 24, 251, in the individu-

temperature

meas-

uring instruments.

36 It cannot be too strongly emphasized that Performance Test Code measurements are to be taken under steady-state conditions. However, failure of a

CONCLUSIONS

38

In the measurement

of temperature

it is im-

temperature sensor to indicate a change in temperature of the medium being measured may not in itself be positive proof that no change has taken place.

portant that the instrument

The sensor and its accessory may be so slow in responding

erned by required accuracy, accessibility to the material to be measured, types of available equip-

lar problem be selected.

measuring equipment to a change that it

may serve to obscure the actual

conditions.

be conducted as described temperature

of the thermal capacity of the sensor itself. A finite time interval is required for the sensor to absorb, or heat during a transient.

Because

in Chapter 9. However,

test data obtained from calibrated

in-

struments should not be taken for granted as necessarily being accurate. The possibility of tempera-

of this,

10 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


The choice will be gov-

ment, and economic factors. When the proper selection of equipment has been made, calibration shall

It is

impossible for any temperature sensor to instantaneously undergo a step-change in temperature because

dissipate,

best suited to the particu-


INSTRUMENTS ture errors occurring as a result of conduction, ation, aerodynamic

heating,

should be investigated

AND

radi-

[15] “Gas Temperature Measurement and the High Velocity Thermocouple,” H. F. Mullikin; “Temperature, Its Measurement and Control in Science and Industry,” Reinhold, New York, pp. 175-802, 1941. [161 “I nt rod UCt’ton to Heat and Mass Transfer,” E. R. G Eckert and J. F. Gross, McGraw-Hill, New York, pp. 34-36, 1963. [17] “Introduction to Heat Transfer,” A. I. Brown and S. V. Marco, McGraw-Hill, New York, third edition, p. 63, 1958. 1181 “Heat, M ass, and Momentum Transfer,” W. M. Rohsenow and H. Y. Choi, Prentice-Hall, Englewood Cliffs, N. J., first edition, second printing, l%l. [lP] “H eat Transmission,*’ W. H. McAdams, McGraw-Hill, New York, third edition, p. 261, 1954. 1201 “Determine the Static and Total Temperatures of a High Temperature, High Velocity Gas Stream,” K. W. Woodfield and R. E. Bloomfield, General Motors Engineering Journal, vol. 5, nos. 3 and 4,1958. 1211 “Temperature Measurement in Moving Fluids,” R. P. Benedict, ASME Paper 59-A-257, 1959. [22] “Steady-State Thermal Analysis cf a Thermometer Well,” R. P. Benedict and J. W. Murdock, Trans.

and dynamic response

and, if significant,

evalu-

ated. REFERENCES --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

39

Throughout

the text Reference

closed in brackets,

numbers are en-

thus [l].

!i’, ::SpMA Standard.”

[3] [4]

[S] [6] [7]

[S]

APPARATUS

PMC 17, 1%3. ower Test Code Thermometer Wells,” J. W. Murdock. I. of Engrg. for Power, Trans. ASME, vol. 81, 1959. “1967 ASME Steam Tables.” “Metal Properties,” ed. by S. L. Hoyt; “ASME Handbook,” McGraw-Hill, New York, first edition, p. 62, 1954. “Unfired Pressure Vessels,‘* Section VIII, ASME Boiler and Pressure Vessel Code, 1968. “American Standard Code for Pressure Piping,” ASA B31.1-1955, pp. 93-94, 1955. “Measurement of Temperature in High Velocity Steam,” J. W. Murdock and E. F. Fiock, Trans. ASME, vol. 72, pp. 1155-1161, 1950. “Measurement of High Temperatures in High Velocity vol. 65, Gas Streams,” W. J. King, Tmss. AWE,

ASME, J. of Engrg. for Power, July, 1963, p. 235. [23] “Recovery and Time-Response Characteristics of Six Thermocouple Probes in Subsonic and Supersonic Flow,” T. M. Stickney, National Advisory Committee on Aeronautics Technical Note 3455, Lewis Flight Propulsion Laboratory, Cleveland, Ohio, July, 1955. 1241 “Experimental Determination of Time Constants and Nusselt Numbers for Bare-Wire Thermocouples in High Velocity Air Streams and Analytic Approximation of Conduction and Radiation Errors,” M. D. Scadron and Warshawsky, National Advisory Committee for Aeronautics, Technical Note 2599, 1952. 1251 “The Dy namic Response of Industrial Thermometers in Wells,” T. C. Linahan, Trans. ASXE, vol. 78, pp. 759-763, 1956. 1261 “Design of Th ermometer Pockets for Steam Mains,” J. E. Roughton, The Institution of Mechanical Engineers, Procedings 1965-66, vol. 180, part 1, no. 39. [27] “Manual on the Use of Thermocouples in Temperature Measurement,” ASTM STP 470.

[9] &421. 1943: eterminatton of the Thermal Correction for a Single-Shielded Thermocouple,” W. M. Rohsenow and J. C. Hnnsaker, Trans. ASME, vol. 69, p. 699, 1947. [lo] “Multiple-Shielded High Temperature Probes,” E. M. Moffatt, Trass. SAE, vol. 6, p. 567, 1952. [ll] “The Design of Suction Pyrometers,” T. Land and R. Barter, Trans. Sot. of Instrwsent Technology, vol. 6. No. 3, p. 112, 1954. [12] “The Measurement of Gas Temperatnre by lmmersionType Instruments,‘* E. F. Fiock and A. I. Dahl, Iour. Amer. Rocket Society, vol. 23, p. 155, 1953. Pyrometer for Measuring Gas Temper 1131 “A So&-Flow atures,” G. T. Lalos, Jour. Res. NBS, vol. 47, p. 179, 1951. [14] “Shielded Thermocouples for Gas Turbines,‘* E. F. Fiock and A. I. Dahl, Trans. ASME, vol. 71, p. 1153, 1949.

11



CHAPTER 2, RADIATION THERMOMETERS

the body.

Par.

4 A selective

GENERAL: scope ........................................................................ .............................................................. : Definitions

PRINCIPLES ()F OPERATION ................................ CLASSIFICATION: Description .............................................................. Materials of Construction ...................................... CHARACTERISTICS: Range ...................................................................... .............................................................. Sensitivity . . P rectston ................................................................ Accuracy .................................................................. Response ................................................................ ............................................................ ACCESSORIES APPLICATION AND INSTALLATION: Sources of Error .................................................... Essential Considerations .................................... Treatment of Data .................................................. AND DISADVANTAGES: ADVANTAGES

radiation

7 5 A blackbody incident 6

20 21 22

is one that absorbs all radiation

uppn it, reflecting or transmitting none.

Emissivity

is the ratio of the radiant energy

emitted per unit time and per unit area by a body, to that emitted by a blackbody at the same temperature. Total emissiuity refers to radiation of all

222 25

wavelengths, and monochromatic or spectral emissiuity refers to radiation of a particular wavelength.

28 34 35

PRINCIPLES

OF OPERATION

7 The operation of a radiation

bears a definite

relation

body. The temperature

to the temperature of the of a body may be determined

from a mea’surement of the intensity of the radiation emitted. This measurement may involve the radia-

scope

tion of all wavelengths formation which will guide the user in the selection, and use of radiation

thermometers.

consists

the emissive proximately

a thermocouple or a thermousually

power or emissivity

an emf meas-

of carbon is ap

three times that of platinum.

having the highest

theoretically

is known as a blackbody

uring instrument.

term “blackbody.”

radiator,

By definition,

possible

A material emissivity

or by a single the emissivity

a blackbody is unity. All materials have an emissivity less than unity. A blackbody is experimen-

3 A total radiation thormomoter is one which utilizes as an index of the temperature of a body all

12


in

approximately three times as bright as glowing platinum when both are at the same temperature. This is technically expressed by the statement that

of an optical

system, used to intercept and concentrate a definite portion of the energy rczdiuted from a body whose temperature is being measured; a temperature sensi-

pile; and a measuring device,

or the radiation

but also on the particular material constituting the source. Thus, glowing carbon appears to the eye

Definitions

tive element, usually

emitted,

a restricted portion or portions of the spectrum. However, in general, the intensity of radiation depends not only on the temperature of the source,

1 The purpose of this chapter is to present in-

Thomometer

thermometer de-

pends on the phenomenon that a body at elevated temperature emits radiation, the intensity of which

GENERAL

2 A Radiation

thermometer is one which

utilizes as an index of the temperature of a body the energy from only a narrow wavelength band or bands.

Advantages ............................................................ 36 Disadvantages ........................................................ 37

installation,

by

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

the energy per unit area per unit time radiated

CONTENTS


of

INSTRUMENTS tally realized

by uniformly

AND

heating a hollow en-

11 Newton’s Law of Cooling states “The rate of cooling of a body under given conditions is propor-

closure and observing the radiation from a small openina in its wall. The intensity of the radiation

tional to the temperature difference between the body and its surroundings.” This law is applicable

emitted from this opening depends only on the temperature

of the walls,

which the walls

and not on the material

energy radiated

to radiation temperature

of

are constructed.

8 Ths Stefan-Boltrmann

law expresses

APPARATUS

thermometry in that any change in the of the source results in a change in

the intensity of its radiation and there is a resultant change in the temperature of the receiver, but a

the total

by a blackbody.

much smaller one. To restore

equilibrium

when the

temperature of the source increases from T, to Tz in absolute temperature, the increase in tempera-

W,= uT4

ture of the receiver

where

should be proportional

to

T24 - T,4. Wb = total energy radiated per unit time, by a blackbody of unit area o = Stefan-Boltzmann

T = absolute Planck’s

9

12 The receiver must be sensitive to small changes in temperature. It is blackened for maxi-

constant

mum absorption

radiation

of enerp;y. The absorbed energy is

transduced to a measuring instrument as an emf. This emf is a function of the difference between

temperature. law expresses

the temperature

the distribu-

tion of energy in the spectrum of blackbody

of the receiver

and its surroundings.

radia-

tion.

CLASSIFICATION

Darcription

where 13 Radiation Wb = radiant energy per unit time, per unit of wave length interval, by a black-

= constants

in the Planck of radiant

radiation

h

= wavelength

e

= base of natural or Napierian

energy

and the characteristics

before its temperature surroundings, candescent.

the rate of heat loss

15 Single Mirror Type.

is reached

radiation

is much above that of its

even when the source is brightly

and fundamental

thermometers

In the single mirror type,

from the source enters the optical

system

through the aperture in the “front diaphrap.” It is reflected from a concave mirror at the other end of

in-

an enclosing

tube and is focused on the receiver

13 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


clas-

is that based

the lens.

of energy from the source.

is such that equilibrium

of radiation

itself.

on their means of collecting; the radiation and focusing it on the receiver. There are three types in use-the single mirror, the double mirror, and

rises until its rate of heat loss to its surroundings by conduction, convection, and radiation is equal thermometers

of the receiver

14 The most conspicuous sification

ceiver, which is heated by the incident radiation absorbed by it. The temperature of this receiver

from the receiver

is not capable of making

degrees of absorption of the emitted wavelength by the media which transmit or reflect the radiation,

area, the re-

In most radiation

instrument

measurements throughout this range because of the varying amounts of energy available, the varying

source by measuring the intensity of the radiation that it emits. This measurement is accomplished by focusin energy radiated from a source at a uni-

to its rate of absorption

of

from the region of absolute zero up to the highest temperatures found in the sun and stars. A single

1op;arithms.

10 As a consequence of the Stefan-Boltzmann law, it is possible to measure the temperature of a

on an absorbing

of the various

adapted. At least theoretically, rate of radiation energy may be used as a measure of temperature

law

measuring

form temperature,

have taken many forms,

designers, and partly because of the diversity of the uses to which the thermometers were to be

body of unit area C,,G

pyrometers

partly as a result of the preferences


of

ASME the temperature sensitive

PERFORMANCE

TEST Calcium

element which is placed

COBES Fluoride

is usually

ture applications,

between the diaphragm and the mirror. See FiE. 2.1.

used for low-tempera-

quartz glass for medium-range

applications, and or Pyrex for high-range applications. All of these materials absorb radiation in varying amounts.

CHARACTERISTICS FIG. 2.1 SINGLE MIRROR RADIATION

16 Double Mirror Type. eter, radiation

In this radiation

enters the instrument

dow, and is reflected

19 The radiation thermometer is generally designated as a total radiation thermometer. It occupies

THERMOMETER

a position

thermom-

through a win-

by a concave mirror which The image of the

emf and the temperature

portion of the source to be measured is made to

ceiver,

where an image of the aperture is formed, 2.2.

with the resistance or the optical thermometer

of the body under observa-

l-

thermometer;

pyrometer.

the thermocouple,

Therefore,

cannot be classified

the radiation

as a primary

laboratory instrument, but rather as an industrial instrument, which is empirically calibrated by determining

-cc

its emf for a number of temperatures

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

DOUBLE MIRROR RADIATION

THERMOMETER

all types

This curve is not .the same for

ofradiation

thermometers,

although for a

given type the curves will be similar. perature-emf

17 Lens Typo. radiation

furnished

A lens is used in this type of

thermometer to form an image of the por-

20

relationship

The tem-

for each thermometer

Range.

The practical

useful range of the

radiation thermometer extends from ambient to ~500~ although it is not possible to use one

receiver, diameter

thermometer to cover this entire span.

the aperture being very slightly than the receiver. See Fig. 2.3.

larger in

21

Because radiation

Sensivity.

based on the Stefan-Boltzmann

sovacr

6‘ ”s

LENS TYPE RADIATION

Materials

is

at the low end of each range is very over the upper third of each range

the sensitivity

is excellent.

22

Precision.

If properly designed

the radiation

the same temperature

and con-

thermometer will

indicate

under the same conditions.

Absolute temperatures can be determined if the emissivity of the source is known and corrections

of Construction

applied for the absorption

18 The window and lens materials used in radiation thermometers are usually determined by the

which the radiation

range over which it is desired to operate.

pyrometer itself.

for the absorption

Mica or

of the media through

thermometer sights as well as of the lens or window of the

14


thermometry

fourth power law,

the sensitivity poor; however, ‘

structed,

THERMOMETER

is

by the instrument manufacturer.

tion of the source to be sighted on which covers an aperture in a diaphragm closely in front of the

FIG. 2.3

in

its range and drawing a smooth curve through the points determined.

FIG. 2.2

field over-

tion is dependent upon the design and materials of construction of the thermometer. This relationship cannot be expressed by a general formula, such as those used for interpolations and extrapolations

cover the aperture and the radiation passing through is focused by a second concave mirror on the reSee Fig.

measuring

ducing any part of the International Practical Temperature Scale. The relationship between output

forms an image of the source on a diaphragm in which there is a small aperture.

in the temperature

lapping and extending beyond that of the thermocouples. It is not recognized as a means of repro-


INSTRUMENTS 23

Accuracy.

Accuracy

of measurement ‘

AND

is de-

slide-wire S, is graduated in terms of drop potential through end coil A and slide-wire S, with its contact set at the highest position. Slide-wire St provides means for adjusting the drop in potential through the AS branch to any set of values depend-

pendent upon tbe aame factors noted in Par. 21 and iu addition to the accuracy of calibration of the thermometer. To attain the utmost in accuracy, the radiation thermometer should be certified by NBS

ing on the output of the radiation pyrometer. A and E are selected to provide the proper emf at the low

over the range in which it will be used. U Response. Radiation thermometers are available having 99 percent response times of less than ous second, and up to 30 set wben the receiver

end of the range. 27

is

Accessories

include mounting brackets,

sigbt-

ing or target tubes with supporting flanges, nozzles for directing air or a nonabsorbing gas in the target or sighting tubes, and cooling jackets (either

lagged to reduce fluctuations that might be caused by flame radiations or reflections. Those most generally

APPARATUS

used have the lower response times.

air or water). ACCESSORIES 35

Potentiometric

connection

recorders are generally

with radiation

thermometers.

used in

APPLICATION

from tbose used with thermocouples in that refer ence junction compensation is not generally provided. As it is required for Performance purposes to make adjustment provide this adjustment which includes

in the potentiometer

the proper emissivity

28 The radiation thermometer must be used with due precautions to minimize the sources of error. Some of these sources of error are beyond the con-

of

available

so that a direct reading of temperature

Sources of Error

Test Code

for emissivities

less than one, there are instruments

that

trol of the user of the instrument,

circuit

be minimized

can be made

A schematic

circuit

correction.

used iq radiation

mometer potentiometric

recorders

24.

type potentiometer

A branched circuit

by attention

while others can

to details

of installation

aud use. 29

26

AND INSTALLATION

They differ

Errors due to uncertain

or variable

of the object under observation

ther-

ent when sighting

is shown in Fig.

when sighting

is used,

on a nonblackbody.

cavity

in a heated object that emissivity

is dependent upon the position of the contact on the slide-wire S,. Each adjustment of St requires an adjustment of rheostat H to keep the total cur

may be neglected. Absorption

of radiation

gases which have

rent constant at the value required to make the

absorption bands in the infrared results

drop of potential through the standardizing resister F equal to the emf of standard cell K. The scale of

These are apt to be overlooked vapor is invisible

errors

by carbon dioxide,

water vapor, or other invisible

less likely

pres-

It is only

into a closed end tube or other deep

in which the ratio of the currents in the two branches

30

emissivity

are inevitably

in errors.

since the gas or

to the eye. Smoke and fumes are

to be disregarded

because they are

visible, but are no more serious as a source of error. Closed-end tubes may fill with gas or fumes, giviag erroneous temperature readings. it is sometimes necessary to purge closed-end tubes continuously with a gentle stream of air to remove absorbing gases. When sighting

on a hot body in a fur

nece gases absorb some of the radiation.

If the

gases are hotter then the surface sighted on, they radiate into the radiation too high a reading.

thermometer,

resulting

tube may be extended through the furnace wall, nearly to the hot surface,

and, e gentle stream of

air or an inert, nonabsorbing FIG. 2.4 POTENTIOMETER

gas may be passed

through the tube.

CIRCUIT

15 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


in

To avoid this error, an open-end


ASME

PERFORMANCE

ADVANTAGES AND DISADVANTAGES

36 Advantages. In the range of the thermocouples, the radiation thermometer has the following advantages:

(a) They

are not subject to deterioration by furnace atmosphere and high temperatures. (b) There is no contact with hot body required, making it possible to measure the temperature of moving objects. They can be used for measurement of surface (cl temperatures. M They can be used for measurement of high temperatures. (e) They can be so located as not to be subject to vibration or shock. (f) They can be used where fast response and long, useful life are required.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

32 Errors caused by fluctuations in ambient temperature are sometimes encountered. Most manufacturers provide ambient temperature compensation built into the radiation thermometer which conpensates for these fluctuations to within stated limits. 33 When sighting into a closed-end tube, it is important that the bottom of the tube is focused on the diaphragm of the radiation thermometer, and that its image is centered on the aperture in the diaphragm. Otherwise the temperature reading may be affected by variations in temperature along the walls of the tube.

In the range of the optical pyrometer, the radiation thermometer has the following advantages: (a) Recording and/or controllinK of temperatures is obtainable. (b) The personal element does not enter into the measurement. (c) Lower temperatures can be measured.

Considerations

34 An examination of the site where the temper ature measurement is to be made will determine the focusing distance required. Manufacturers’ literature or representatives should aid in determining the desired components.

freatmont

CODES

absorption, and calibration. Corrections at other than calibration temperatures should be determined by linear interpolation.

31 It is sometimes desired to measure the temperature of a body inside a furnace which is sealed to maintain an atmosphere of hydrogen or other gas. Windows provided for temperature measurements introduce large errors, whether they are made of glass or quartz. Such errors must be determined experimentally if possible and corrected. Closely related to this type of error is the error caused by deposits of dirt or foreign material on the windows, lenses, or mirrors of the radiation thermometer. This error can be eliminated by periodic cleaning.

Essential

TEST

37 Disadvantages. In the range of the optical pyrometer or thermocouple thermometer, the disadvantages of the radiation thermometer are: (a) It has a higher first cost. (b) When sighting from a distance, large sources are required. (c) In general, the accuracy is lower.

of Data

35 The observed temperature readings should be corrected by adding corrections for emissivity,

16



CHAPTER 3, THERMOCOUPLE THERMOMETERS CONTENTS

ACCESSORIES APPLICATION AND INSTALLATION: Sources of Error . . . . .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . Essential Considerations . . . . . . . . . . . . . . ...................... Treatment of Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ADVANTAGES AND DISADVANTAGES: Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ... ... .. .. .. .. . ... . .. ... .. .. ... ... .. . ..... .. ... .. . ... . Disadvantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . REFERENCES

Section A, Thermocouples Par.

GENERAL: Scope ........................................................................ .............................................................. Definitions .................................. PRINCIPLES OF OPERATION CLASSIFICATION: Description .............................................................. Materials of Construction ......................................

GENERAL .................................................................... ................................ PRINCIPLES OF OPERATION DESCRIPTION: Potentiometer Circuits .......................................... ...................... Reference Junction Compensation Automatic Null-Blancing Mechanism .................... Types of Indicating Potentiometers .................... ............................ Reference Junction Apparatus

cept where joined together to form junctions. are necessarily corresponding

Scope 1

elements. subiected

The purpose of this chapter is to present in-

tion, installation,

guide the user in the selec-

and use of thermocouple

perature.

Definitions

system comprising

emf which includes emf to equivalent

sens-

conductors for operatively See Fig. 3

a device for sensing

324:

a printed scale for converting

temperature

units,

and electrical

connecting

(Note: For practical

Extension

Wires

wires having such temperature-emf

con-

to the thermocouple

are a pair

of

characteristics

with which the wires

are intended to be used that, when properly con-

sensing ele-

nected to the thermocouple,

ments, electrically

of the wires.

ence iunction

from each other ex-

the thermocouple

refer

is in effect extended to the other end

17 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


the refer-

Wires are a pair of electrical

ment of a thermocouple thermometer comprising two dissimilar electrical conductors called thermoeleinsulated

reasons,

formed in two parts, as

(ice point).

relative is the temperature

of the thermo-

3.lA.I

5 Thermocouple

the two.

3.1A.

A Thermocouple

There

to each thermocouple

to the two extremities

4 Connecting

ing element called a thermocouple which produces an electromotive force (emfl,

2; 67

ductors which connect the thermocouple to the emf measuring device. These are generally of copper when the reference iunctions of the thermocouple are maintained at some fixed temperature such as

is a temperature

a temperature

9 28

The mea.surinR junction is that which is to the temperature to be measured. The

shown in Fig.

measuring

two junctions

ence junction is usually

Thermometer

1 7

reference junction is that which is at a known tem-

ther-

mometers.

2 A Thermocouple

41 42 43

A, THERMOCOUPLES

GENERAL

formation which will

i: 40

Section 6, Instrumentation

CHARACTERISTICS: 25 Range and Accuracy .............................................. .............................................................. Sensitivity ................................................................ :$ Precision 28 Response .................................................................

SECTION

29


ASME THERMOCOUPLE

PERFORMANCE

WIRES

TEST

CODES

has no influence good electrical

EXTENSION

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

MEASURlNG

WIRES

on the emf developed, contact is attained.

providing

The most widely

used method for making the iunction is uutogenous welding, in which the thermoelements are fused to-

JUNCTION

gether by a torch or by electrical using any other material

means without

to form the junction.

The

noble metals should be welded without using a flux. For torch welding ATHERMOCOUPLE

base metals it is advanta-

geous to use a flux to minimize

WIRES

should be exercised

oxidation.

in the application

the ends of the thermoelements

Care

of heat to

to avoid overheat-

ing. All traces of flux should be removed after the welding process. Further details on the welding of thermoelements may be found in ISA Recommended

JUNCTION

Practice INSTRUMENT

FIG.

3.1A

THERMOCOUPLE

RP l.l-.7,

Measurement

THERMOMETER

8 The thermocouple

SYSTEMS

soldering PRINCIPLES

and in the ANSI Temperature

Thermocouples

with the thermoelements

temperatures.

the production

in 1821 that an electric

to prevent the solder or brazing material

of two dis-

of the emf which actuates

9 The mechanical

the cup be-

increased

metals placed in contact,

the

turns at the iunction

is’ heated at one end, an emf between

the hot and cold ends of the wire, the magnitude of

is the sum of the Peltier

of two dissimilar

of the junction may be

the wires together for a few end. The twist should be

The temperature

measured is that at the first point

of the electrical

contact proceeding

10

of the ends. The total emf

actinK in a closed circuit

strength

by twisting

ence iunction to the measuring

which depends upon the metal and upon the differences in temperature

from

All traces of flux

omitted whenever the thermocouple is to be used where a temperature gradient exists at the iunction.

if a wire of homo-

known as the Thomson emf is developed

Care should be taken

should be removed.

emf exists

and upon thk metals used. Also,

at the temperatures

to

magnitude of which depends upon the temperature geneous material

in service.

running back from the iunction.

of the metals are

Two causes contribute

rent. An emf known as the Peltier tween two different

that is corn-

OF OPERATION

metals when the iunctions

at different

may be formed by

patible

current will flow in a closed circuit similar

junction

or brazing with a material

to be encountered 6 Seebeck discovered

C %.l.

from the refer=

junction.

In measuring the temperature

of a metal sur-

face, it is often advantageous to attach the thermoelements separately to the metal. This may be done

metals

emf at each iunction and

by spot welding

or peening the wires to the metal.

the Thomson emf over each wire, consideration being Riven, of course, to the algebraic signs of

The points of attachment should be close enough together that there will be no significant difference

these four emfs. The total emf acting in such a cir-

in temperature

cuit thus depends upon the temperatures junctions.

If the temperature

between

them.

of the two

of one junction

is

11 insulations

and Protection

Tubes.

fixed at some known value such as that of the room

mocouple wires must be electrically

or of the ice point (32W,

all points other than the measuring

the temperature

other junction

can be determined

emf developed

in the circuit.

principle

of thermoelectric

This

of the

by measuring the

ous refractory

is the basic

materials

CLASSIFICATION

lower temperature

Description Measuring

Junction.

enamels,

are used as thermocouple

at Vari-

alumina

fiberglass, asbesand various plastics

insulations.

The thermo-

couple, extension, and connecting wires should be maintained dry over their entire length.

The method employed for the thermoelements

18


such as porcelain,

applications,

tos, rubber, fabrics, 7

junction.

and magnesia in the form of beads, tubes and powder encased in metallic sheaths serve as means of insulating and supporting the thennoelements. For

thermometry.

making the junction between

The ther-

insulated


INSTRUMENTS

AND

12 Although most of the metals and alloys used

APPARATUS 16 When it is inconvenient

to use an ice bath, a

as thermoelements exhibit a relatively high degree of mechanical stability, the majority of them are

thermally

subject

In such case, the temperature

to change in calibration

contaminating

and corrosive conditions.

tubes which are impervious therefore

Platinum

thermocouples tamination

Protection

to such conditions

required if high accuracy

are required.

silver,

when exposed to

are particularly

are

mechanical

strength,

to con-

by ceramic

a refractory

of the reference

junc-

instrument

17 Frequently some accuracy is sacrificed for convenience by eliminating the constant tempera-

tubes which are impervious to gases and vapors at temperatures within the working range. To provide additional

or a stirred liquid bath may be substituted.

and taken into proper account.

and long life

susceptible

block of copper, aluminum or

tion must be measured with an auxiliary

versus platinum rhodium

and should be protected

insulated

ture reference

junction

ence junction

compensation

in favor of automatic refer-

or recording instrument

built into an indicating

used to measure the emf de-

veloped by the thermocouple.

metal

Recently,

automatic

reference junction controls, external from the emf measuring instrument, have become available.

tube is often placed over the ceramic tube.

These provide either the ice point temperature 13 Protecting

wells

are employed when the ther-

mocouple is used in liquids

various elevated

and gases at high

18 Thermocouple

welding. They are usually fabricated of carbon steel, stainless steel (188) or 14 percent chro-

highest

mium iron. Thermocouple

or reference

in molten metals, and chemical

stand the particular vailing

conditions

or

within

junction

Where the

thermocouples

apparatus.

istics

and hazards pre-

identical

character-

to those of the thermocouple.

this is not feasible,

in the installation.

eliminate

by the use of ex-

tension wires not having temperature-emf

to with-

should

to the instrument

This will

errors which might be introduced

salt baths

must be selected

Wires.

Extension

accuracy is desired,

be long enough to ccnnect directly

tubes for use

furnace atmospheres,

processes

usually

?O.SV.

pressure. They are made of metal, and may be turned and drilled from bar stock or built up by

protection

temperatures,

extension

Where

wires from the ther-

mocouple reference temperature may be used. For base-metal thermocouple installations, the exten14 In providing protection for the thermocouple, however, one must not lose sight of the fact that a

sion wires are either the same or nominally

thermocouple

wires must have the equivalent

conditions

can perform its function

of heat transfer

same as the thermocouple

only when the

are such that the meas-

uring junction attains or at least approaches temperature to be measured.

lation over the temperature

the

Reference

and easily

reproducible

platinim

The most satisfactory

Junction.

method of reference

generally

junc-

nickel-copper

The actual reference

copper wire, electrical

a thermoelement

each insulated

contact between

it ap-

matched pair

of a copper wire and a

and the copper-nickel thermoelement.

wires do not match the individual

These

thermoelements,

but when used together they compensate reasonably well over the range specified above. It is important

junc-

that the junctions

and a

thermoelement

in such a manner that

20

them is made only below

Switches

of the extension

wire and the

be at the same temperature. and Terminal

Blocks.

All switches

used with thermocouples should be of a rugged construction and designed so that both wires are

the surface of the mercury. The copper connecting wires are extended to the emf measuring instrument.

19 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


are

wire for thermocouples

alloy wire, have found general use

alloy wire to the platinum

proximately 1/2in. The tubes are inserted so that the mercury is several inches below the surface of mixture.

and

materials

A specially

consisting

num rhodium thermoelement

through a hole in the cork.

tion is made by inserting

re-

as extension wires for the platinum vs platinum rhodium thermocouple over the temperature range 32 to 400%. The copper wire is joined to the plati-

small glass tube, closed at the bottom and flared at

the ice-water

substitute

used as extension

of conductors,

wide-mouth Dewar jar provided with a cork will serve to maintain the ice bath over several hours. For each junction to be kept at the ice point, a

Enough mercury is placed in each tube to fill

emf-temperature

range to be encountered

of the high cost of platinum

rhodium alloys,

made from these metals.

tion control is realized by an ice bath, consisting of an intimate mixture of shaved ice and water. A

the top, is inserted

the

Extension

in service. 19 Because

15

materials.


switched

*hen

switching

PERFORMANCE

from one thermocouple

22

to the next, such that thermocouples not in use are entirely disconnected from the measuring instrument. The switches

TEST

(a)

Platinum-10

national (freezing

similarly

studs and jacks be of thermocouple

21

of Construction

Common thermocouple

materials

the approximate

limits

are available of -300

to

and chemical

6)

properties,

standard material

the conventional

thermoelements

inertness

and stability

at high temperatures

in

atmospheres.

Both thermoelement

ma-

Platinum

-13

percent

R)

produces a slightly

Cc) Platinum

- 30 percent

Rhodium versus Plati-

num - 6 percent Rhodium. (Type B) This combination of platinum-rhodium alloys is useful

versus platinum

800

1000

1200

TEMPERATURE

EMFS OF VARIOUS

1400

1600

I800

, OF

MATERIALS

VERSUS

PLATINUM

20


greater emf for a given

to somewhat higher temperatures

FIG. 3.2A

-

type. It

temperature.

to which

600

is similar

to the Platinum

10 percent Rhodium versus Platinum

3.2A.

400

table to 20.3

Rhodium versus Plati-

This thermocouple

in general characteristics

the thermoelectric characteristics of other materials are referred. The emf-temperature relation of are shown in Fig.

and

oxidizing

num. (Type

requirements, but each possesses characteristics desirable for selected applications. Platinum is accepted

to 1064.43%

by a high degree of chemical

percent of the measured emf.

melting point, thermoelectric properties, reproducibility and cost. No single thermocouple meets all

the generally

point of antimony)

will match the standard reference

thermometry.

These few have been chosen on the basis of such factors as mechanical

the Inter-

Scale from 630.74%

percent-Rhodium versus Platinum thermocouples, as procured from a reputable source,

t3200eF. Of the vast number of possible combinations of metals and alloys, only a limited number are in actual use in thermocouple

Temperature

serves as the

for defining

terials are ductile and can be drawn into fine wires. The thermocouple is widely used in industrial laboratories as a standard for the calibration of base metal thermocouples and other temperature sensing instruments. Platinum -10

materials.

for use within

Rhodium versus Platinum.

instrument

is characterized

from rapid or large temperature fluctuations. It is recommended that wherever practical, all connectors

Materials

percent

S) This th ermocouple

interplating

be subjected to temperature fluctuations due to air currents or radiation from hot sources. Terminal

and terminal

Noble Metals

(Type

should be located so as not to

blocks and panels should be protected

CODES


2000

2200

and shows

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

AWE

INSTRUMENTS slightly

greater mechanical

conventional

Platinum

AND

strength than the

24

versus Platinum

(a)

used for special

purposes.

versus stainless-steel, denum, and platinum

Base Metals n ta n. (Type T) Constantan

Copper-Con&a

The copper-constantan

couple is widely used in industrial

for particular

25

Range and Accuracy.

The upper temperature

mocouples are used. Table 3.1A lists the recommended upper temperature limits for thermocouples

range -200

a sacrifice

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

protected

pheres. The ranges of applicability

wire sizes erally

For the higher temperatures, For temperatures

standard sizes

up to l@OT,

responding

iron-constantan thermocouples show good calibration stability in nonoxidizing atmospheres.

(c) Chromel-Alumel.

(Type

alloy of approximately percent Chromium. ganese,

2 percent

This thermocouple, range -200

couples,

1 percent Silicon.

emf versus temperature,

The following

to oxidation

combination.

table lists

thermocouples.

It must

reducing atmospheres. and reducing atmospheres

exposed to the temperature

(d) Chromel-Constantan. tion of thermoelements

and temperature

the average thermoelectric

power for the conventional

are particularly destructive. Both thermoelements are mechanically strong and are often directly ment.

3.2A. The corand emf for the

The sensitivity of a thermocouple, 26 Sensitivity. that is dE/dT, varies somewhat with temperature.

and higher for short

is more resistant

be protectedagainst Alternate oxidizing

are given in Table

of

wires of

versus emf, deg F and deg C.

usable over the temperature

than any other base-metal

and extension

values of temperature

atmos-

and limits

3 percent ‘Man-

Aluminum,

to 23OOT,

time intervals,

10

Alumel has a composition

of about 94 percent Nickel,

or contaminating

various types of thermocouples are given in Table 3.3A. See National Bureau of Standards ‘Monograph 125 for expanded reference tables of these thermo-

K) Chrome1 P is an 90 percent Nickel,

from corrosive

error for thermocouples

of number 8 gage or larger are gen-

employed.

applica-

limits for the various thermocouples depend on the wire sizes and the environment in which the ther-

to t 14OOdF but may be used up to 1800%’ at of life.

include chrome1

nickel versus nickel molybrhodium versus gold palla-

CHARACTERISTICS

(Type J) This thermocouple

to the temperature

above are sometimes

thermo-

is probably the most widely used of all thermocouples in industrial thermometry. It is genlimited

of

and lab-

or;atory applications over the temperature range - 300 to t 7ooq;‘.

(b) Iron-Constantan.

Combinations

These

dium. Each has advantages tions.

is an

55 percent Copper, 45

alloy of approximately percent Nickel.

erally

Special Purpose Metals.

metals other than those listed

Rhodium thermocouples. 23

APPARATUS

(Type

E) This

develops

Average dE/dl, Microvolts per Deg F for the Range Specified

Thermocouple

environ-

combina-

the highest

thermoelectric output of any of the conventional thermocouples, namely, about 34 PV per

Copper-Constantan Iron Constantan

28.0 (32-650% 32.0 (32-14OO?)

Chromel-Alumel Platinum-10% Rhodium -Platinum Chromel-Constantan

23.0 (322200%) 6.3 ( lOOO-2650°F) 42.0 (32-1400%‘)

deg F at normal ambient temperature and increasing to about 45 PV per deg F at 1000%‘. This high output has led to the use of ChromelConstantan

as sensing elements

for radiation thermocouple

detection systems.

The thermocouple

The precision

with thermocouples

the temperature has

techniques

for temperature

up to about 14OOv.

Precision.

tainable

and in differential

also found general application measurements

27

in thermopiles

It is char-

Small diameter

All thermocouples

show a

with operating service.

thermoelements

(less than 0.010 in.

diameter) are particularly susceptible to change in calibration when used near their upper temperature limit.

21


atupon

range and upon the experimental

employed.

gradual drift in calibration

acterized by a high degree of calibration stability when used at temperatures not exceeding 1OOOv.

of measurement

depends primarily


ASME

TABLE

3.1A

PERFORMANCE

R&commended Upper femperoture Voriour

Ng $j5G”,g”

Thermocouple Type

.

(0.128

for Protected

Wire Sires (Awg),

CODES

(0.032

(O.d64 in.))

in.)) I

Limit

for

F)

Nq&;i;ge (0.020

in.))

I

T

Upper Temperoture

Thermocouples

DEG C (DEG

Nii Q ‘ 3’ mC,age NTbsFl($ge

In

I

Limits

TEST

Nyb.::,c,“ge

in.))

(0.013

L

in.))

I

371 (700)

260 (500)

204 (400)

204 (400)

J

760 t 1400)

593 (1100)

482 (900)

371 (700)

371 (700)

E

871 (1600)

649 (1200)

538 (1000)

427 (800)

427 (800)

K

1260 (2300)

1093 ( 2000)

982 (1800)

87 1 (1600)

87 1 ( 1600)

R&S

1482 (2700)

B

1705 (3100)

TABLE

3.2A

LIMITS

OF ERROR

OF THERMOCOUPLES

AND

EXTENSION

WIRES FOR STANDARD

WIRE SIZES

l Thermocouple

Limits

Temp. Range, *

Type

I T

J

E

K

R&S

B

Std.

of Error Special

Extension Wire Type

I

I

-184

to -59

---

?l%

-101

to -59

22%

+l%

-

59 to t93

~0.8~

kO.4C

t

93 to+371

+3/4%

?3/8%

0 to 277

+2.2oc

+1.1c

277 to 760

+3/4%

?3/8%

0 to 316

k1.7Qc

--

316 to 871

?1/2%

----

0 to 277

?2.2oc

+1.1c

277 to 1260

?3/4%

+3/8X

0 to 538

of Error Special

I

to i-93

+0.80c

+0.4oc

JX

0 to 204

+2.2oc

+l.loc

EX

0 to 204

t1.7oc

-__-

KX

0 to 204

+2.2c

---

TX

-69

?1.4oc

---

m-e

--

W-N

538 to 1482

fl/4%

--

e-e

__-

__I

-___

871 to 1705

?1/2%

I--

o--m

_-

--

____

22 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


Limits Std.

Temp. Range, y


_-

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

TABLE 3.3A

TEMPERATURE-EMF

RELATIONSHIP

FOR THERMOCOUPLES

Emf-Millivolts Temperature, Deg F

R(?-13pRlh)

b(?-3~~h)

K(c,;u:g,>

.J(thztan)

0.00

E~~n~~an

0.

32 200 400 600 700

0.000 0.595 1.474 2.458

0.000 0.596 1.504 2.547

0.00 0.02 0.18 0.47

0.00 3.82 8.31 12.86

4.91 11.03 17.18

5.8 13.7 22.2

800 1000 1200

3.506 4.5% 5.726

3.677 4.868 6.125

0.89 1.43 2.09

17.53 22.26 26.98

23.32 29.52 36.01

31.0 40.0 49.0

1400 1600

6.897 8.110

7.436 8.809

2.85 3.72

31.65 36.19

42.96

57.9

1800 2000 2200 2400

9.365 10.662 11.989 13.325

10.237 11.726 13.255 14.798

4.68 5.72 6.84 8.03

40.62 44.91 49.05 53.01

2600

14.656

16.340

9.28

2800 3000

15.979 17.292

17.875 19.394

10.56 11.85

3300


S~~-‘Opqh)

13.84


ASME 28

The time response

Response.

PERFORMANCE

ings.

between the junc-

32 Inasmuch as the curves giving the relation between emf and temperature are not, in general,

tion and its surroundings. The greater the surface area to mass ratio of the junction, the greater will be the speed of response condition.

linear,

for a given heat transfer

When a thermocouple

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

well or protecting

sheath,

indicated

by the probe will be governed by the combined effects of heat transfe.r from the medium to the well and that of the well to the enclosed

junction.

The minor accessories used in conjunction with thermocouples, such as protection tubes, ex-

are provided with junction

temperature

from that on which standard calibrations

are based.

of the reference

instrument has

automatic reference

junction

mocouple extension

wires should be used to con-

34

Essential

compensation,

to the instrument. Regardless

Considerotions.

or the techniques

31

Sources

mary consideration is that the temperature indicated by a thermocouple is that of the measuring junc-

Instrumentation.

tion. The accuracy obtained

there is a definite between

AND INSTALLATION

of Error.

Experience

and reproducible

the difference

uring and reference

junctions

upon how closely thermocouple

has shown that

in measuring

is definitely

of the meas-

of a thermocouple

35

and

measuring junction

of the instrument.

However,

junction

ference in temperature depends upon the rate of heat transfer and the thermal resistance between and substance.

pose it is desired

must be electri-

As an illustration,

to measure the temperature

metal plate which is heated from within

cept at the measuring junction. Whena thermocouple is mounted along a conductor, such as a pipe or a metal structure, special care should be

is brought into contact with the metal plate.

to ensure good electrical

insulation

junction The

junction will receive some heat from the plate by thermal conduction. The junction will lose heat by conduction along the thermocouple wires, and by

be-

convection,

wires and the conductor to

conduction

24


measuring

sup of a

by some

means. The bare thermocouple

exercised

of

and the

from conductors on which they may be mounted, ex-

tween the thermocouple

in

substance, then a difference in temperature will exist between the two. The magnitude of this dif-

the junction from each other, from ground and

accuracy

heat between the thermocouple

temperature in each case must be known. Temperature emf tables for thermocouples are usually based on a reference junction temperature of 324;‘ (ice

cally insulated

to that of the object or space.

imbedded in a solid or immersed in a

calibration

tures. It is not necessary to maintain the reference junction temperature, during use, the same as durina; calibration. However, the reference junction

of a thermocouple

related

of

which

many applications this may not be the case. If under steady conditions there is a net exchange

tempera-

point). The elements

of the

liquid will attain equality in temperature with the substance and will, therefore, indicate the true temperature of the solid or liquid to within the

couple can be determined as a function of the temperature of the measuring junction. Thus the device and used for measuring

junction

can be brought to the temperature

A small size thermocouple

suitably

the emf developed. If the reference junction is maintained at some known temperature, such as 32oF (ice point), the emf developed by the thermo-

may be calibrated

the measuring

the tem-

depends

the object or space, or to some temperature

relationship

in temperature

of the

employed

perature of any object or space usually APPLICATION

ther-

in carrying out the measurements, there are certain basic factors which must be considered. The pri-

instrumen-

is given in Section B,

Thermometry

with thermocouples

type of thermocouple

tation employed to measure and record the emf sigThermocouple

conjunction

nect the thermocouple

tension wires, switches and terminal blocks have been discussed in earlier paragraphs of this chapter.

nals from thermocouples

observed in applying reference junction Many commercial instruments used in

33 In the event the emf measuring

covering

do not cor-

of emf. This should be

particularly corrections.

for deviations

29

discussion

of temperature

either manual or automatic means for compensating

ACCESSORIES

30 A detailed

equal increments

respond to equal increments

is encased in a

the response

CODES

prevent stray currents in the conductor from entering the thermocouple circuit and vitiating the read-

of a thermo-

couple to a temperature change is a function of the mass and geometry of the sensing junction and the mode or modes of heat transfer

TEST


and radiation

to the sur-

INSTRUMENTS Obviously

the junction

will

be at a

M~z&-J~@

lower temperature than the plate. This difference in temperature can be reduced by: (1)

Improving the thermal contact by: (a) Flattening the junction to obtain a larger or welding

thermocouple

practical

FIG.

dicated

so as to reduce the tempera-

ture gradient in the wires near the junction. the temperature of the space sur-

36

Various

by use of insulation

and special

CONNECTED

circuits

widely

the indicated

in Fig.

type thermo-

------

3.4A.

COPPER

INSTRUMENT

Coaxial-tube

(2)

111,* 121. Aspirating or high-velocity

(3)

Thermocouple

(4)

shield 141. Sonic flow thermocouple

(5)

Radiation

(6)

[71. Thermocouples

for steam temperature

measure-

(7)

ments [81. Thermocouples

shielded

thermocouple

for gas measurements

in two-

thermocouple

with low emissivity

REFERENCE JUNCTIONS

[3].

radiation

pyrometer

FIG. 3.4A

[5,61.

thermocouple

39

pyrometer

The installation

ment technicians.

CONNECTED

of extensive the services

Special

IN PARALLEL

thermocouple

of qualified

attention

to extension wires, reference and terminal assemblies.

instru-

should be given

junctions,

switches,

40 Treatment of Data. Although the calibration accuracies of the conventional thermocouples are

37 Thermocouples may be joined in series. A series-connected thermocouple assembly is generally referred to as a thermopile and is used in measuring

THERMOCOUPLES

equipment requires

phase flow [9].

primarily

WIRES

where detailed

(1)

compensated

be the

mean of the remaining thermocouples. A schematic diagram of a parallel thermocouple circuit is shown

on each may be found: radiation

become

reading will

applications and operatis a list of the most

used methods and references

information

average

are of equal resistance.

couple assemblies have been employed in making gas temperature measurements. Each has certain advantages for particular ing conditions. Following

IN SERIES

be the true arithmetic

Should one or more of the thermocouples open-circuited,

source of heat.

procedures

and it will

if all thermocouple

(c) Raising

or an auxiliary

THERMOCOUPLES

38 In a parallel connected thermocouple circuit, a mean value of the individual thermocouples is in-

(b) Keeping the wires close to the plate for

rounding the junction

3.3.4

by:

wire of the smallest

diameter.

some distance

REFERENCE JUNCTIONS

the junction

to the plate. Reducing the heat loss from the junction

62)Using

..-. - - .~_

JUNCTIONS

area of contact (b) Soldering, brazing, (2)

APPARATUS

small temperature

usually

which the output is the arithmetic

well within the tolerances

listed

in Table

3.1A, it is recommended that sample thermocouples from each lot of material be checked to determine compliance to specifications. Usually it is suffi-

differ-

ences, for example, as the sensing element of a radiation receiver. The series connection, in

cient to check samples from each end and the center of a spool or coil and consider

sum of the emfs

calibration

applicable

the average

to the entire lot. Corrections

of the individual thermocouples, may be used to obtain greater measurement sensitivity. A sche-

for deviations of the average calibration from the established reference table should be applied in

matic diagram of a series thermocouple

reducing observed data. For installations

is shown

where the

in Fig. 3.3A.

highest

*Numbers

should be calibrated separately and corrections applied for deviations from the established refer-

chapter,

in brackets designate

References

at end of

accuracy

is required,

ence table in the reduction

thus [I].

the thermocouples

of the data. Since the

25


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

roundings.

4ND


ASME instruments

on which the thetiocouple

measured are also subject tions,

PERFORMANCE

Moffatt,

to scale reading correc-

AND DISADVANTAGES

Advantages Simple in basic design and operation. Small in size,

flexible,

capable of installation

in relatively inaccessible spaces. Suitable for remote indication; sipal used to indicate, Primary

may be

record or control temperature.

elements are relatively

All components of the measuring

low in cost. system are

individually replaceable. Suitable for wide range temperature

applica-

tions. High accuracy 42

Trans.

attainable.

Disadvantages

(a)

A relatively small signal output is produced requiring sensitive measuring equipment.

Other references couples:

(b)

Knowledge of or compensation for reference junction temperature is required.

“Thermoelectric

(c)

Subject to calibration

REFERENCES

on the general subject

numbers are

26 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


Probes,”

E. !+I.

6, p. 567, 1952.

of thermo-

Thermometry,” Wm. F. Roeser, lour. Applied Physics, vol. 11, no. 6, p. 388, 1940. “Thermoelectric Thermomefry,” Paul H. Dike, Leeds & Northrup Co., 1954. “Thermoelecrric Effects,” R. P. Benedict, Electrical Manufacturing, p. ,103, Feb. 1960. “General Principles of Thermoelectric Thermomerry,” D. I. Finch, vol. 3, part 2, of “Temperature - Its Measurement and Control in Science and Industry,” Reinhold, New York, 1962. “Manual on the Use of Thermocouples in Temperature Measurement,” STP 470.

changes with use.

43 Throughout the text Reference enclosed in brackets, thus [II.

High Temperature SAE, vol.

[2] “‘Determination of Thermal Correction for a SingleShielded Thermocouple,” W. M. Rohsenow and J. P. Trans. ASME, vol. 69, p. 699, 1947. Hunsacker, [3] “The Design of Suction Pyrometers,” T. Land and R. Barber, Trans. Sot. Instr. Tech., p. 112, Sept. 1954. A. I. [4] “Shielded Th ermocouples for Gas Turbines,” Dahl and E. F. Fiock, Trans. ASME, vol. 71, p. 153, 1949. [S] “A Sonic-Flow Pyrometer for Measuring Gas Temperatures,” George T. Lalas, Joum. Res. Nat. Bur. Stds., vol. 47, no. 3. p. 179, 1951. [6] “A Py rometer for Measuring Total Temperature in Low Density Gas Streams,” S. Allen and J. R. Ramm. Trans. ASME, vol. 72, p. 851, 1950. [7] “Gas Temperature Measurement,” W. L. SeveringMay 1937, p. 334. haus, Mechanical Engineering, [8] “Measurement of Temperature in High-Velocity Steam.” J. W. Murdock and E. F. Fiock, Trans. ASME, vol. 72, p. 1155, 1950. [9] “High R es p onse Aerosol Probe for Sensing Gaseous Temperature in a Two-Phase. Two-Component Flow,” R. P. Benedict, Trans. ASME, 1. Engrg. for Power, vol. 85, p. 245, 1963.

in reducing the observed data.

41

CODES

[1] “Multiple-Shielded

emfsare

proper account of this factor must be made

ADVANTAGES

TEST


INSTRUMENTSAND APPARATUS SECTION

B, INSTRUMENTATION

GENERAL

1 The basic principle

of thermoelectric

etry is that a thermocouple is a function measuring

develops

of the difference

(hot) and reference

the temperature

thermom-

an efm which

in temperature

of its

(cold) junctions.

If

of the reference

8 The potentiometer, as the term is used here, serves a similar function in the measurement of voltage, and in fact may be called a “voltage balance,” the standard voltage being furnished by a standard cell, the “lever” being resistance ratios, and the galvanometer serving as the balance indi-

junction

is known,

cator. Since no current is drawn from the standard

the temperature of the measuring junction can be determined by measuring the emf generated in the

urement is independent

circuit. The use of a thermocouple in temperature measurements therefore requires the use of an in-

vanometer or balancing

strument capable

of measuring

cell or the measured source at balance,

ance, except to the extent that this affects

Potentiometer

gal-

sensitivity.

Circuits *

Test Code work because of 9 One-Dial

limitations.

3 Recording industrial

mechanism

resist-

DESCRIPTION

in-

struments in common use in industry, millivoltmeters and potentiometers, the former are rarely their inherent

the meas-

circuit

emf.

2 While there are two types of emf measuring

used in Performance

of external

potentiometers

are widely

process temperature

used for

measurement

Potentiometer.

Fig.

and

that it has a range of 0 to 100 mV.

control. However, in general, they are not well suited to Performance Test Code work. Inaccura-

BATTCRV

RHEOSTAT

OATTERT

cies in charts are caused by printing limitations and humidity effects. Practical limits on chart widths and scale lengths

3.1B shows the

circuit of an elementary potentiometer having one measuring dial and a single range. Let us assume

result in inadequate

readability. 4 Indicating

potentiometers

all Performance

are recommended

able with the required readability 5 Thermocouples and the calibration to the instrument 6 Thermocouples nect directly

for

Test Code work as they are availand accuracy.

should be calibrated separately corrections should be applied readings. should be long enough to con-

to the instrument

or reference

junc-

tion apparatus. This will eliminate errors which might be introduced by the use of extension wires which may not have temperature-emf identical

FIG. 3.18

Range

7 Accurate

10 Slide-wire

is usually

Accurate

often accomplished

weighing,

by direct

for example,

comparison

selected potential

is

lever arms may be used to multiply

so that when standardized, the drop of across resistance A is equal to the volt-

age of the standard cell which is nominally

against

standard weights using a mechanical balance. If the measured weights are too heavy for direct comparison,

1.0190 V.

*The circuits shown in this section do not represent the actual circuits of the instruments listed in Table 3.18. They are intended only to illustrate basic principles.

the

forces.

27 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


A are con-

of the emf of the standard cell SC, and the current through slide-wire C and fixed resistance A is

a matter of

comparing an unknown quantity against a known quantity or standard-the more direct the comparison the better.

C and fixed resistance

nected in series with the battery and battery rheostat. The resistance A is made a simple multiple

OF OPERATION

measurement

POTENTIOMETER

O-100 mV

characteristics

to those of the thermocouple. PRINCIPLES

SINGLE-DIAL


ASME 11

The resistance

of the slide-wire

so that with this same standardized drop of potential

PERFORMANCE

potentiometer

S to position

the current through resistance

C is standardized

A

which taps resistance

of the battery rheostat

13 The measurement

on the single dial

to improve the pre-

19 Fig. 3.2B

but with a dial-switch

C at various points.

shows the circuit

of such an instru-

ment. Note that it now takes a double pole switch

that is, that substantially no current flows from the standard cell. The voltage drop across resistance A then matches the emf of the standard cell SC to within extremely close tolerances, the limit being of the galvanometer

it is necessary

using another slide-wire,

by throwing switch

1, and the battery current is adjusted

the adjustability

possible

cision with which it is possible to adjust the value of resistance M. This is done most simply by converting connection Pt to a movable contact, not by

by means of the battery rheostat until the galvanometer G shows that balance has been attained,

sensitivity

To improve the ac-

Potentiometer.

curacy and readability

across it will be 100 mV. The

12 In operation,

CODES

18 Two-Dial

C is selected current, the

slider Pz is arranged so that any value of voltage from 0 to 100 mV may be picked from the slide-wire.

and slide-wire

TEST

to shift the galvanometer from the standardizing circuit to the measuring circuit. 20 If M is the resistance between sliders P, and Pz and C is the total resistance of Dials I and II,

and the

the operation and explanation of the circuit of Fig. 3.2B is the same as for Fig. 3.1B, except that the

G.

of the unknown emf connect-

measurement

requires

adjustment

of two dials.

ed to terminals “EMF” is then accomplished by throwing switch S to position 2 and sliding movable contact P, on measuring slide-wire C to a point where the galvanometer

G again gives a null indi-

cation. The voltage drop across resistance M, that is, from connection P, to slider P, on the slide-

t-A--T----c-

wire, then matches the measured emf to within close tolerances, limited again only by the adjustability

of the calibrated

vanometer sensitivity. on the slide-wire

slide-wire The position

C and the galof slider

scale is then directly

P,

proportion-

al to the measured emf. I4 Since the same current flows through resistances M and A, the measured emf is compared to ratio M/A

with an accuracy very close to that with which this

SC

ratio is known, provided of course that the battery

FIG. 3.28

current remains constant during the time interval between the standardizing

and measuring

operations.

15 The calibration of an instrument of this type is stable, since resistors and slide-wires can be made with a high degree of stability and the emf of

21

coeffi-

of an elementary

potentiom-

eter of this sort having its entire range across the slide-wire, is limited by the resolution possible Setting and reading the position of slider P,. 17 The smallest an instrument

practicable

scale division

fore, the slide-wire

in

now covers only 10 mV. The

smallest practicable scale division is of the order of 0.05 mV or 0.05 percent of span. Or, to put it another way, the two.-dial instrument

on

can be read

to one more place than can the single-dial

of this sort is of the order of 0.5mV

ment.

of 0.5 percent of span.

28


ratio M/A equals the

ments of 1 ohm each, and Dial II is reduced to 1 ohm, the total resistance C is 10 ohms as before. But if the total range remains 0 to 100 mV as be-

cient of emf of very nearly zero. 16 The usefulness

Again the resistance

voltage ratio measured-emf/standard-emf, but resistance M can now be adjusted to allow a much closed subdivision of voltage than with only one dial. If Dial I has 10 taps and 9 resistance incre-

an unsaturated standard cell as used for potentiometric work does not change more than about 0.01 percent per year. It also has a temperature

TWO-DIAL POTENTIOMETER Range 0.100 mV

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

the standard cell emf by the resistance


instru-

INSTRUMENTS

AND

22 Note that the two dial readings are made by

29

taking the first figure from Dial I and two or three additional figures from Dial II.

Commercially

available

thermocouples

are of

several types, each type having a temperature-emf relationship represented by standardized tables. These tabular values are commonly stated in terms

23 Three-Dial Potentiometer. A further improvement over the two-dial instrument can be achieved by making a three-dial instrument. Fig. 3.3B shows the circuit

APPARATUS

of a reference 30

of such an instrument.

junction

For Performance

temperature

of 32%‘.

Test Code work it is recom-

mended that the reference junction of the thermocouple be maintained at 32T by means of an ice bath as described in the paragraphs on Reference Junction Apparatus in this Section. When such an

A-c-

t-

ice bath is used, the thermocouple

emf as meas-

ured by a potentiometer

in millivolts

all those recommended

calibrated

in this section

(as

are) can be

converted directly to the equivalent temperature of the measuring junction by referring to the proper temperature-emf table. 31

Certain

of the potentiometers

recommended

this Supplement have no means for correcting readings

for reference

than 32%‘,

junction

in

their

temperatures

other

and should always be used with an ice

bath. SC

32 THREE-DIAL POTENTIOMETER Range O-100 mV

This

designated

pensator, maintains a constant resistance battery circuit as Dial II is operated. 26

The smallest

scale division

or subtract,

as comin the

places than the one-dial

General.

junction

tion temperature and known value,

temperature,

is the

or correction

by the thermocouple.

corrects

reading to a value equiv-

ture-emf

the instrument

This

reference

junction

been at

table and added to that generated

by the

thermocouple in order to convert its readings to a reference junction temperature of 32oF, and thus permit direct use of the temperature-emf 34

junc-

For example,

thermocouple

at some fixed

table.

if the emf generated by a

is 27.83 mV when its reference

Type K junc-

tion temperature is 2OO”F, 3.82 mV, the equivalent of 2000F in the temperature-emf table based on a

for its temperature

must be made.


to or

from the emf generated

if the reference junction were at 32oF. The emf equivalent of the reference junction temperature of 2oooF must then be determined from the tempera-

the emf generof the differ-

the reference

must be maintained

to add

as the need may be, a voltage

emf generated when the measuring junction is at a given temperature will be less than that generated

ence in temperature between its junctions. Therefore, to use the generated emf to evaluate the measuring

whereby the set-

33 If the reference junction temperature of a thep mocouple is, say 2OOoF, rather than 32oF, the

Compensation

As stated previously, ated by a thermocouple is a function 28

with manual reference

facilities

32%‘.

instrument.

Junction

recomused

alent to that which would have been generated,

and to two more

27 The actual reading of the instrument sum of the settings of Dials I, II, and III. Reference

potentiometer is sometimes

causes the circuit

had the thermocouple

is of the order of

instrument

is provided

compensation

ting of proper dials

0.005 mV or 0.005 percent of span, making it passible to read the instrument accurately to one more place than the two-dial

instrument

junction

and III, Dial II covering the range 0 to 9 mV and Dial III, the slidewire, covering the range 0 to 1mV. The third tapped resistor

precision

where the highest degree of accuracy is not required and/or it is undesirable to use an ice bath.

24 As in the case of a two-dial instrument, Dial I covers the range 0 to 90 mV while the remainder of the range of 0 to 10 mV is divided between Dials II

25

The portable

mended in this Supplement


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

FIG. 3.38

ASME reference

junction

temperature

PERFORMANCE

of 32’F,

TEST 39

must be

CODES

Slider P,

can now be moved along slide-wire

added to convert to the equivalent value with a reference junction temperature of 32q; (27.83 + 3.82 =

1) to add or subtract a voltage to or from the emf generated by the thermocouple. If slider P, is

31.55 mV). Referring

is moved toward resistor

to the temperature-emf

for Type K thermocouple, temperature 35

junction

junction

36

compensation

When using manual reference

pensating

facilities,

the reference

ature must be measured accurately

calibration

is 1400°F.

made in this manner, it is inconvenient readings are being taken.

can be

40

junction

comtemper-

The principle

in Fig.

3.4B.

If reference

Reference

D, connection

P, can be

in branch 1 and the

for use with any type of thermocouple.

41 Method 2 for Manual

Reference

Junction

Another method of obtaining

pensation.

reference Junction

be in-

required for any reference within its range.

Since the main slide-wire

suitable

with a liquid-in-

set to this value.

37 Method 1 for Manual pensation.

will

reference junction compensating slide-wire in branch 2 are both calibrated in millivolts, they are

glass thermometer, the equivalent emf found from the appropriate temperature-emf table based on 32oF, and the dials

on slide-wire

set to the correction junction temperature

when many

junction

E, the voltage

creased and vice versa. With a suitable millivolt

it will be found that the

of the measuring

While reference

table

junction

compensation

Com-

manual

is shown in prin-

ciple in Fig. 3.5B. Here again the upper portion of

Com-

of this method is shown is made to Fig.

3.lB

lob mV

FIG.

3.48

SINGLE-DIAL

POTENTIOMETER

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

Range O-100 mV with junction compensator

manual

reference

FIG.

showing the elementary circuit of the one-dial potentiometer, it will be seen that the upper sec-

3.58

SINGLE-DIAL Range junction

tion of this circuit and that of Fig. 3.4B are identical. A second branch has been added in the cir-

POTENTIOMETER

O_lOO mV with

cuit of Fig. 3.4B consisting of resistances B and E and slide-wire D connected in series, which in turn are connected in parallel with branch 1 and

the circuit

the battery

series with the thermocouple.

circuit.

therefore,

currents

one-dial

circuit,

resistance

however,

resistance

A in

42

P, is moved to the slider

3.1B,

potentiometer.

that

In this

B has been placed in An auxiliary

battery

C , and C, supply current through

B, and the drop of potential

Adjustment

ence junction

on slide-wire D and placed in its mid-position, it will be seen that for any specific value of emf, the two circuits will be in balance for the same position of slider Pe on the main slide-wire.

of this circuit

temperature

across it is

for a given refer-

is made by short circuit-

ing the emf terminals and setting the main slidewire scale to the millivolt equivalent of the refelc ence junction temperature and then adjusting rheostats C, and C, until a balance is obtained. Fol-


reference

added to the emf of the thermocouple.

it and iz are equal. If the resist-

branch 1, and connection

with that of Fig.

of the elementary

and two rheostats

38 For sake of illustration, let us assume that the resistance of each branch is the same and, ance of (B + %D) is made equal to resistance

is identical

manual

compensator


INSTRUMENTS lowing this, the emf read directly

AND

from the main

leads that of the line phase by 90 deg and the motor rotates in a direction to increase the magni-

slide-wire scale will be corrected for a reference junction temperature of 32oF and may be used directly

in the temperature-emf

APPARATUS

tude of the known voltage.

tables.

49

When the unknown voltage

known voltage, Automatic

43

Automatic

Null-Balancing

potentiometers

quently employ an electro-mechanical nism which detects

known emf and known voltage

fre-

and actuates

in the opposite

direction

to re-

duce the magnitude of the known voltage.

between an un-

to drive the slider on a slide-wire --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

of a typical

the motor rotates

servomecha-

the difference

the two are equal. Fig.

lags the current in the line phase by 90 deg and

Mechanism

null-balancing

is lower than the

the current from the power amplifier

50 Thus the motor is always driven in a direction such that the difference between the unknown emf and the known voltage is reduced to zero, at which point the motor ceases to rotate, the instru-

a motor

to a point where

3.6B shows a block diagram

ment being in balance.

system.

MOTOR

TC

FIG.

44

As in a manually

3.6B

balancing

BLOCK DIAGRAM-AUTOMATIC MECHANISM

instrument,

the

Types

unknown emf of the thermocouple is connected in opposition to the drop of potential across a slide-

51

istics

wire. 45 The difference

between

which is a d-c voltage,

of Indicating

Potentiometers

Table 3.1B lists the significant characterof three types of precision indicating poten.

tiometers presently available, which are used for Performance Test Code work. They are (1) labora-

these two voltages,

is converted

NULL-BALANCING

tory high precision,

to an a-c volt-

age whose phase depends upon whether the unknown

portable precision.

voltage

Laboratory

is higher or lower than the known voltage.

High Precision

dial automatic

If the unknown emf is higher, the converted alternating voltage is in phase with the line voltage. If

balancing,

(2) plant precision,

and (3)

There are three subtypes

balancing,

Instruments, (1B) three-dial

and (1C) two-dial

of the

(1A) threemanual

manual balancing.

it is lower, it is 180 deg out of phase. 52 46

The alternating

voltage

from the converter

applied to the input of a multistage fier where its amplitude sufficient

is increased

is

voltage ampli-

47 The power amplifier supplying

serves

however, herein.

as a phase dis-

53

current to one phase of a two-

Type

be at least equivalent

1A Laborato.ry

High

Precision

Potentiom-

null-balancing indicator. It has a range of O-70.1 mV, is readable to 0.001 mV(lpV) and has a limit of

When the unknown emf is higher than the

error of 3 PV or 0.02 percent, It is suitable

the current from the power amplifier

whichever

for panel mounting.


should,

to those described

Laboratory High Precision Potentiometer Type 1A is a self-contained, three-dial, automatic

capacitance. 48

instruments from Perform-

eter.

phase reversible induction motor, whose other phase is supplied from the line voltage through a

known voltage,

that suitable

in the future be excluded

ance Test Code work. Such instruments

to a value

to drive a power amplifier.

criminator,

It is not intended

developed


is greater.

_ .

_ “.

.._.. .~_ ASME 54

The balancing

PERFORMANCE

operation is ‘ automatic

the two dial switches

are set manually

TEST 57

once

CODES

Type 1B is a three-dial

Type 1C is a two-dial

to place the

instrument while

instrument.

unknown emf within the scale range. This automatic feature facilitates

58

the rapid measurement of

skill on the part of the operator, and with a high degree of accuracy.

dials and have corresponding limits of error of 1, 3, and 30 PV or 0.015, 0.015, and 0.01 percent, whichever is greater.

55 This instrument is used primarily in the laboratory for calibrating thermocouples, and thermocouple wire differentially against standardized wire of the same type. It is available with selfcontained push-button switches multiplicity of thermocouples. Types

High

same performance characteristics.

Potenthe

setup with a number of accessories, laboratory-type

including

to Type 1A. It has a range of 0 to 40.1 mV, is readable to 0.0025 mV (2.5 bV) and has a limit of error of 0.5 percent below 6 mV and 0.33 percent

a

galvanometer

and

above 6 mV. It is suitable

standard cell.

TABLE

3.18

TYPES

OF

INDICATING

POTENTIOMETERS

Laboratory

Type

High

1A

Range(s) mV

for panel mounting.

PERFORMANCE

Precision

TEST

Plant

Precision

1C

Three-Dial

CODE

WORK Portable P recision

2

Two-Dial

Two-Dial

3 Two-Dial

Automatic

Manual

Manual

Automatic

Manual

Balancing

Balancing

Balancing

Balancing

Balancing

O-70.1

o-16/0-160/

O-16/0-160/ o-1600

O-40.1

O-16/0-80.5

o-15/0-150/

o-15/0-150/

o-35

o-l 5/o-75

O-1500

O-1500 ck5.1

o-1.1/0-5.5

O-1600

Dial 1, mV

FOR

1B

Three-Dial Description

of thermocouple ther-

60 Plant Precision Potentiometer Type 2. The Plant Precision Potentiometer is a self-contained, two-dial automatic null-balancing indicator similar

They are of the

manually balancing laboratory type, not self-contained, requiring a more or less permanent bench storage battery, high sensitivity

are used for the calibra-

and thermocouple materials,

mometer instruments.

Precjsion

Laboratory High Precision 1B and 1C have essentially

tiometers Types

These instruments

tion of thermocouples

and checking the.calibration

16 and 1C Laboratory

Potentiometers.

59

a

O-60

-

Dial 2, mV

019

O-l/O-10/~100

Dial 3, mV

O-1.1

0.1/1.0/10.0

1.0/10.0/100.0

Limit of Error

3 pv or* 0.02%

1 &V/3 rlv/ 30 ,uv or* 0.015%/0.015%/ 0.01%

0.015% or* 0.05 mV

0.5% below 6 mV 0.33% above 6 mV

0.01 mV/O.OS mV

Manual Ref. Junction Compensation

No

No

No

No

Yes

range mV

o-1/0-5

Method of Balancing

Automatic

Manual

Manual

Automatic

Manual

Portable

No

No

No

No

Yes

Self-Contained

Yes

No

No

Yes

Yes

Self-Contained point Switch *Whichever

MultiOptional

Optional

is greater



--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

56

for selecting

These instruments have three ranges, 0 to

16,0 to 160, and 0 to 1600 mV. They are readable to 0.0005, 0.005, and 0.05 mV on their respective

a large number of emfs with a minimum of effort and

INSTRUMENTS

AND

its measuring junction

61 The balancing operation is automatic once the dial switch is set manually to place the unknown emf within the scale range.

the temperature

63

Portable

Portable

Precision

Precision

contained,

for selecting

Potentiometer

Potentiometer

two-dial

The low range is readable and 0.015 percent,

3.

68

Most thermocouple

applications

instrument.

thermometers

are furnished

reference junction is normally located in the instrument and varies in temperature with ambient condi-

of error are 0.01

respectively.

tions. Such compensated their scale calibrated

64 This instrument has a manual reference junction compensator with two ranges, O-l and O-5 mV,

of the measuring

instruments

junction.

65 This instrument, while not recommended for the most precise Performance Test Code work, is

tion provides an instrument perature,

used where convenience

with only one type of thermocouple

66

Analog

Level

Input.

conversion signals

to Digital

Converters

-

from thermocouples

requires

which are not normally necessary balance potentiometric

Millivolt

millivolt

stability,

precautions

70

tion, errors may be introduced

zero offset, exist.

tion is allowed

by noise caused by

introduced

usable

and for only one

Test Code work there are two

or the temperature

of the junc-

to vary, and a compensating

into the circuit

or accounted

emf is

for by cal-

culation. 71 Under fixed temperature reference junctions can be listed: triple point of water cells, ice baths,

frequency. The conversion milliseconds

seconds.

is generally

versatility and accuracy for Test Code work.

In Performance

fixed temperature,

In addi-

series and common mode voltage, gain accuracy, quantization and fluctuations in supply voltage and

erally

compensa-

in terms of tem-

basic methods for providing suitable reference junctions. Either the junction is maintained at a

types of equipment.

and sensitivity

junction

reading

such an instrument

type lack sufficient use in Performance

level

when using null-

The usual errors due to nonlinearity, resolution,

reference

range of temperatures. Furthermore, the accuracy of the automatic reference junction compensation is limited. Generally speaking, instruments of this

analog to digital

equipment for measuring

on this type of compensation.

59 While automatic

degree of accuracy.

The use of electronic

have

See Pars. 28-42 for more

details

are more

usually

in terms of the temperature

either of which can be used with either range .. of the instrument.

important than the highest

for industrial

with means for automat-

ically compensating the reading of the instrument for the actual reference junction temperature. The

to 0.001 mV and the

and portability

of the

tion is measured can be no ,yreater than the accu-

It has two ranges, 0 to 16.1 mV and 0 to 80.5 mV. high range to 0.005 mV. Limits

the temperature

acy with which the temperature of the reference junction is known.

Type 3 is a self-

manual balancing

junction,

junction must be

measuring junction. It is axiomatic that the accuracy with which the temperature of the measuring iunc-

a multi-

Type

and its reference

of the reference

known in order to determine

62 This instrument is used in taking multiple readings throughout the plant. It is available with self-contained key switches plicity of thermocouples.

APPARATUS

usually

occurs very rapidly;

and in some instances

This introduces

one possible

gen-

automatic

micro-

ice baths and constant

These are described

temperature

ovens.

below.

source error. 72

All of these errors may be minimized by proper selection of primary sensors, careful attention to

Triple

Point

of Water.

A cell can be con-

structed in which there is an equilibrium

between

ice, water, and water vapor. The temperature of this triple point is t 0.01% on the IPTS-68, and it is reproducible to about O.OOOl”c.

intercabling practices, and selection of hardware having the capability of the required accuracy. Systems are available having stated conversion errors of less than 0.01 percent.

73 Single

Ice

Bath.

Fig. 3.7B shows a method

for providing an ice bath reference junction. A wide-mouth vacuum bottle is partially filled with Reference

67

Junction

Since the emf generated

a function

of the difference

shaved ice and water, mixed to give a thick slush. The lid of the vacuum bottle is drilled to accommo-

Apparatus

by a thermocouple

in temperature

date two test tubes for each thermocouple

is

33 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


to be

used. Clean mercury is placed in the bottom of each

between


ASME tube to a depth of approximately

one-half

PERFORMANCE inch.

In-

TEST 75

CODES

It is strongly recommended that a mercury-in-

sulated copper wires are attached to the ends of the thermocouple wire from the measuring junction, and each joint is placed in a test tube and bottomed so that the junctions of the wires are submerged in

the thermometer should be close to the bottoms of the test tubes. Refer to Chapter 6, Liquid-in-Glass

the mercury. The assemblage

Thermometers,

glass thermometer be installed cate the stability

of tubes is then slow-

ly and gently pressed into the slush.

in the bath, to indi-

of its temperature.

for emergent-stem

The bulb of

correction

pro-

cedure. 76 Multiple

Ice

Bath

With Master

Ice

Bath.

When

the number of thermocouples used on an installation is large enough to require a number of ice baths, it is recommended that a reference

junction

master ice bath be used to enable the operator to readily determine whether all the baths are maintaining

a temperature

pictorial

of 32%‘.

and schematic

Fig.

3.813 shaws in

form a master ice bath in-

stallation. JUNCTION

77

OF WlftES

A checking

thermocouple

is installed

in each

of the ice baths and run to the master ice bath. There its wires are joined to copper wires and the junctions are immersed in mercury contained in test tubes in a manner similar to that used for the measuring thermocouples in the individual ice baths. 78 The copper wires from the test tubes are run FIG. 3.78

THERMOCOUPLE ICE BATH

REFERENCE

to individual points on the indicator switch. This enables the operator to check the temperature of

JUNCTION

each ice bath to make sure it remains at 32%‘. 73

Automatic

Practical

thermoelectric

74 It is important that enough ice be maintained in the bath so that the level of water does not come closer than 1 in. to the bottom of the test tubes.

refrigerator equilibrium

devices are available in which all between ice and water is constantly

When setting up the bath, periodic

maintained.

The change of volume of water in freez-

observations

should be made to determine the time required for the water level to reach this point. Thereafter shaved ice should be added within

Ice

Bath.

ing is used to control the heat transfer. Some commercially available devices provide wells into which the user may insert reference junctions

this time.

THERNOCOUPLE WIRE m MEASURIffi JUNCTION COPPER WIAE TO

FIG. 3.88

REFERENCE

JUNCTION

MASTER

ICE

BATH

34 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



INSTRUMENTS formed from his own calibrated

AND

81 Electrical Compensation. A compensating circuit containing a source of current and a combi-

wire. Others are

provided with many reference junction

pairs brought

out to terminals which the user may connect into his system. The error introduced into a system by these devices

APPARATUS

nation of fixed resistors and a temperature sensitive resistor can be obtained which will have a

may be as small as 0.1~.

variation

of emf similar

to that of the reference

junction of the thermocouple is allowed to change. 80

Constant

ically

Temperature

controlled

Ovens.

82 Zone Box. An alternative method of dealing with many thermocouples makes use of a single

A thermostat-

oven provides a means of holding

a reference junction at an approximately temperature. To use reference junctions

reference

constant held at

two ovens are available

junction.

tions in turn as they are switched

which make this correction

uring instrument.

COPPER

are routed

Through copper wires, the emf of the reference junction is added to each of the measuring junc-

using

automatically.

THERMOCOUPLE

All the thermocouples

to a zone of uniform temperature. A single reference thermocouple also is routed to the same zone box.

elevated temperatures with tables based on 0°C reference junction temperature, a constant amount must be added to the thermal emf. Devices

when its temperature

to the emf meas-

See Pig. 3.9B.

CONNECTING

F

cu

sELECTOR

9

0

b

7

(COPPER)

POTENTIOMETER

COPPER CONNECTING WIRES

UNIFORM TEMPERATURE ZONE

I

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

MEASURING JUNCtlONS

J

1 ICE BATH (REFERENCE JUNCTION)

FIG.

3.98

A ZONE-BOX

CIRCUIT

INVOLVING

ONLY

ONE

REFERENCE

JUNCTION

35



CHAPTER 4, RESISTANCE THERMOMETERS 4

CONTENTS

A Thermistor

is a special

type of resistor

prised of a mixture of metallic

GENERAL:

less

Scope .. .................. .................................. ....................

that of metals but greater than that of insulators. Semiconductors have a high

than

typical

. . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . Definitions . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . PRINCIPLES OF OPERATION CLASSIFICATION: .......... Description and Materials of Construction CHARACTERISTICS: Range .......................................................................... 29 .................................................................. Sensitivity 30 Precision .................................................................... 31 Accuracy .................................................................... 32 Response .................................................................... 33 ............................................................ ACCESSORIES 34 APPLICATION AND INSTALLATION: Sources of Error ........................................................ 40 Consideration .......................................... 41 Essential Treatment of Data .................................................... 43 ADVANTAGES AND DISADVANTAGES: ................................................................ Advantages 44 .......................................................... Disadvantages 45 REFERENCES ............................................................ 46

negative temperature coefficient in contrast with most metals which have a positive coefficient. PRINCIPLES 5

The basis for resistance

reproducible

expressed

as a simple mathematical

any temperature

thermometer,

can be represented

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

R, = resistance is to present in-

and use of resistance

thermom-

3 A Resistor ducible

consists

at various, freezing

of a metallic

temperature-resistiuity

at temperature,

T of

known, stable

and boiling

temperatures

such as the

points of certain pure materials.

CLASSIFICATION

sensing

ele-

charac-

7

teristics.

Description

and Materials

Resistance

thermometers

ing to the material

of Construction

are classified

used in the resistor.

36


+ AT + BT2)

of thermometer

the two.

in wire form, having known repro-

and stable

at

by the equation:

6 Temperature is measured by measuring the resistance of a calibrated resistor at the temperature to be measured. Pure metal resistors are calibrated

2 A Resistanca Thcrmomefer is one consisting of a sensing element called a resistor, a resistance measuring instrument, and electrical conductors

ment, usually

For a

R, = resistance of thermometer at 0°C A, B = constants, dependent on characteristics the platinum wire.

guide the user in the selec-

connecting

formula.

the resistance

where

Definitions

operatively

is the

manner. It has been found that the

platinum resistance

scope

tion, installation, eters.

thermometry

changes of resistivity with temperature of some metals follow a definite relationship which can be

GENERAL

which will

OF OPERATION

fact that most metals and some semiconductors change in resistivity with temperature in a known,

R, = R,(l

1 The purpose of this chapter

com-

known as

semiconductors which are substances whose electrical conductivity at or near room temperature is

Par.

formation

oxides


accord-

The plati-

INSTRUMENTS

AND

num resistance thermometer is further delineated according to usage as precision or industrial. thermometers

are usually

ing spaced with mica washers, The assembly is evacuated to one-half atmosphere of dry air and then sealed. This construction is usable over the range -190 to + 500°C (-310 to 932oF). See Fig.

cylindri-

cal but special purpose resistance thermometers can be obtained in a flat construction. Some resistors are available

as a webbing

eled copper or nickel silk or other textile 9

Platinum

in which an enam-

wire is woven into a warp of material.

Resistance

accuracy thermometers The relation

material

for the following

between resistivity

ture is very simple; perature range.

(b) Its resistivity

(c)

Physically,

for high

FIG. 4.1(a)

reasons:

and tempera-

of resistivity

(a)

is

The protecting

contain water of crystallization. At temperatures above 500°C (932v) water is

by heating

to high

driven off and the crystalline

in air.

Platinum

Resistance

The precision

platinum resistance

used to define

the International

ature Scale from gen) to 630.74% (-297.3

182.96’C

(melting

to 1168.3v).

In making precision

thermometer

Practical

high-temperature

is

of which the thermometer that the ratiot

R

condition

point of oxy-

of precision

connection

precision

wire is wound helically form consists

in detail

the helically

wound platinum and then adjusted

is frequently

in the range of -50

of mica

is annealed

and filled

thermometer

has a

to +482@F).

platinum resistance

used in calorimetry

to t 100°C (-58

work

to t212W.

In

this form, the platinum wire is wound on a flat mica form, notched to retain the platinum

to

25.5 ohms + 0.1 ohm at 0°C. Four gold leads are

ing is protected

welded to the platinum (two to each end of the

bly is placed in a thin-walled

winding) and the assembly mounted in a Pyrex tube 7 mm OD. The leads are insulated with glass tub-

*Throughout brackets,

by flat mica sheets,

the text Reference

wire; the windand the assem-

metal tube. After

numbers are enclosed

thus 111.

37


the

is placed in is used to in-

to t250°C (-452

13 Another form of precision thermometer

on a mica form. The mica

wire. The winding

Helium

platinum resistance

useful range of -269 See Fig. 4.1(b).

platinum

notched to receive

is

crease the speed of response, by virtue of it having a higher thermal conductivity than air. This form of

1.3925 for

a relatively spiraled platinum

of two crossed pieces

winding

tube; the tube is evacuated

with helium prior to sealing.

precision of a strainfree

a second form

thermometer

head. In this form of thermometer

a platinum

T = 100°C. the construction

calorimetry,

platinum resistance

standard thermometer resistor

is made should be such

resistance thermometer occupying small volume in which a helically

a

mica must be used to

frequently used in which the thermometer is part of the calorimeter, thereby eliminating the need for a

of the platinum

Meyers [3] describes

thermom-

above 500°C (932%‘)

in a quartz tube.

12 In low temperature

with the defi-

shall not be less than

platinum resistance

resistant

must be enclosed

R, A paper by C.H.

and crumble.

make the winding form and the platinum winding

Temper-

nition of the International Practical Temperature Scale, [1,21* th e p 1a t’lnum resistance thermometer must meet the following specifications: The purity and physical

distort,

eters for use at temperatures

point of antimony),

In accordance

is de-

by reducing

Thermometer.

(boiling

structure

stroyed; the mica forms which support the platinum winding swell,

Platinum is subiect to contamination atmospheres and metallic vapors. Precision

to

higher than about

(6) Mica crystals

It can be drawn to very fine wires.

IO

platinum resistance

tube is Pyrex which yields

stresses at temperatures 5oo”c (932W.

to corrosion.

It can be stress relieved temperature

the upper limit of the

thermometer.

it is very stable.

It is resistant

establish

range of the standard precision

high.

coefficient

TYPE OF PRECISION PLATINUM RESISTANCE THERMOMETER

11 Two factors

it holds over a wide tem-

is relatively

Its temperature satisfactory.

Platinum

Thermometer.

has been chosen as the resistor

(4

4.1(a).

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

S Resistance

APPARATUS


in

ASME closing

PERFORMANCE

TEST

16 The differences

the end, the portion of the tube over the

resistance

winding is flattened

speed of response.

to increase

CnDES

standard precision ance thermometers

the

See Fig. 4.1(c).

in construction

Stondard Precision

Component I

Wire

between

the

and industrial platinum resistare tabulated below.

Industrial

I

I

Bare

Insulated-high temperature

Winding Form

Mica

Metal

Protection Tube Insulation

Pyrex

Metal-(usually stainless Ceramic

Glass Tube and Mica

(of leads)

steel)

Spacers --

FIG.

4.1(b)

17 Nickel


Resistance

Thermometer.

Nickel

has

been used satisfactorily as a resistance thermometer material. Its low cost as compared with the standard platinum thermometer has been the determining factor in its adoption for industrial

measure-

ments in its range of temperatures about -75 to +150°C (-100 to +3OOoF). It is less stable in its characteristics

than platinum.

The upper limit is

imposed by the materials used in insulating the nickel wire-enamel, silk, or cotton. The limit of error is dependent

upon the measuring

circuit

used,

and on the range. With a balanced bridge method of measurement, it is of the order of +0.5 deg F. FIG. 4.1(c)

18 Copper an excellent


Resistance

material

Thermometer.

mometry. Its temperature 14 industrial

Platinum

Resistonce

The platinum requirements limit of error in resistance

Thermometers.

for reproducibility and thermometers for in-

ficult

to observe

important.

Larger measuring

teristic

15 Platinum

is used in industrial

because of its stability high temperatures

and ability

to withstand

work.

19 The disadvantages thermometer are:

The upper

limit of the thermometer range is a result of the limitation

of the copper resistance

6~) Its tendency to oxidize at higher temperatures. (b) Its low resistivity compared to platinum or nickel.

of the mountings and methods of conof the

38 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


where

ments to within 20.1 deg F can be made for temperature-difference

applications

struction rather than of the characteristics platinum.

measurements

a compensator is used to match the resistances of the two thermometers so that accurate measure-

is required.

without deterioration.

temperature-resis-

Because of this linear charactwo copper resistance thermometers can be

used for temperature-difference

currents are permissible

a more robust construction

is slightly

It can be secured so that it is not dif-

Rt = R, (1 t AT).

some of the precautions

required for the construction of standard thermometers. Strain-free support of the windings is less and usually

to match an established

ther-

tivity table. The resistivity-temperature curve is straight within narrow limits from about -60 to &OOT. That is, for copper the curve becomes

dustrial temperature measurements are, in general, less severe than for primary standards. Usually a precision of +O.l deg C is adequate. Consequently, it is unnecessary

coefficient

greater than that of platinum. commercially in a pure state,

is

Copper

for use in resistance


INSTRUMENTS

(c)

Its windings siveness,

must be of fine wire to avoid mas-

and consequent slowness

to temperature sistance

thermometers

of 10 ohms at 77‘%.

bration is excellent, Copper resistance upon to maintain

usually

their accuracy

limitation 20

of cali-

27

over a long period

row temperature

is not exdeeded.

circuits

In a common method of construction

‘of both

thermometers the in- ’ on a metal bobbin and

three leads are attached-two

to one end of the

28

of tube-by

sponse is greatly over the nickel

The protecting

or copper resistance

coefficient

thermometer

insertion

thermometers

in the windings

of electrical

22

intended

machinery

thermometers

in the determination

inum resistance thermometers would be require-d to cover this span-the thermometer described in Par. 10 plus a low temperature calorimetry thermometer are

useful over the range of -269 t48 2%?. The industrial

of relative

humidity.

23

Nonmetallic

Resistance

Thermometer

(Ther-

small size,

and calibra-

24

and higher over-all

The high resistance

the effect

cost.

of thermistors

of lead length variations

30 Sensitivity.

minimizes

with a uniformity

thermistors

while the high

of resistance

at a given temperaon a run-of-

mill basis.

on the order of

2 percent

By selection, can be realized.

tolerances

Although

the coefficient

proximately

0.022 ohm per deg F; 100 ohms of cop-

per, 0.215 ohm pei deg F; and 100 ohms of nickel produces a sensitivity

This lack of interchange-

of 0.186 to 0.213 ohm per

deg F.

39 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


of re-

tivity of about 0.1 ohm per°C. Comparatively, 10 ohms of copper produces a sensitivity of ap-

cannot be obtained

ture to better than about 20 percent,

thermometer

sistivity of platinum is lower than that for the base metal resistance thermometers, approximately 25 ohms in the resistance winding produces a sensi-

temperature coefficient of resistivity permits the design of circuits having high sensitivity. 25 Unfortunately,

to

although the 100 ohm type is preferred for negative temperature measurements. Nickel resistance thermometers have a useful range of -40 to t121.1°C C-100 to t3OOW.

tion stability requirements exceed those attainable with other types of thermometers. Thermistors have limited use, however, because of short ranges, noninterchangeability

platinum resistance

(-452.2

cumulation shorting out the leads. Copper resistance thermometers can be used over the range of -195.6 to 121.l”C (-320 to +250??)

In general, thermistor resistance thermometers are used where sensitivity, accuracy, speed ruggedness,

to t250°C

can be used over ranges of 0 to 1064’C (32 to 195OoF), and -182.97 to 0°C (-297.3 to 32%‘) the latter type is sealed to prevent moisture ac-

mistor).

of response,

than pure metals.

Precision resistance thermometers are 29 Range. used for measuring temperatures from -269 to 630.74OC (-452.2 to 1168.3v). Two precision plat-

for

used in pairs for wet bulb and dry bulb temperature measurements

the

not be sub-

CHARACTERISTICS

strips.

Ten ohm copper resistance

their output.

To be useful as a resistor,

of resistivity

is

are made of grids of copper imbedded in flexible plastic

Metals.

.nar-

designed

tube

sealed. 21 Copper resistance

spans unless specially

Pure metals are generally preferable. Alloys usually have lower and less reproducible temperature

contact with

so doing the speed of re-

increased.

the useful

to relatively

are used to “linearize”

Other

in shape.

ject to permanent change, and not have any critical temperature; i.e., transformation point where crystalline structure will change when heated to any temperature within the intended operating range.

and one to the other end to compensate

the inside

limits

instrument

metal must have stable characteristics,

for ambient temperature changes. The bobbin is inserted in a thin-walled metal tube, closed at one end. The bobbin must make intimate

upon replacement

is exponential

This extreme nonlinearity

range of a calibrated

temperature

copper and nickel resistance sulated wire is wound bifilar &inding,

the curve for a thermistor

can be depended

of time provided the manufacturer’s

adjustment

eter materials such as copper, nickel and platinum depart to only a minor degree from a straight line,

have a re-

much superior to that of nickel. thermometers

circuit

26 While the shape of the temperature-resistivity curves for the commonly used resistance thermom-

thermometers.

The stability

APPARATUS

ability requires of units.

in response

changes if it has the same re-

as platinum or nickel

Copper resistance sistance

AND


ASME 31

Resistance

Precision.

utmost in precision

PERFORMANCE

thermometers

deg C; the industrial

ACCESSORIES

34

metal resistance

bridge is used for measuring of precision

The Mueller

to 20.05 deg C.

[41 circuit

over a comparatively

provides

extreme accuracy

narrow range. Designed

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

provide a standard of temperature 32

Accuracy.

thermometer certified.

The precision

platinum

deg C when

thermometer

precision

has

with special

limits

guaranteed

thermometers

are

to +0.25 deg C; they can be adjusted

of measuring

The response time of resistance varies considerably

their construction.

depending

resistances

ten thousandths ever

33 Response.

for

resistance

quired, the Type G-l may be used. It is capable

to +O.l deg C over a limited span when so specified and at a premium price.

thermometers

thermometer

by the electrical

method. This combination has a limit of only ti.01 deg C over its working range. Where the extreme accuracy of the Type G-2 Mueller bridge is not re-

of error of 20.75 deg

C. The base metal resistance

platinum resistance

exact measurements

a standard limit of error of 21.5 deg C but can be obtained

to

throughout the

range of -190 to t500°C, the Type G-2 %eller bridge, reading directly in ohms, combines with the

platinum resistance

has an accuracy of ~01

The industrial

platinum resist-

ance thermometers. This bridge is an advanced modification of the conventional Wheatstone bridge.

to 20.15 deg C; the base

thermometers

A Mueller

changes in resistance

type of platinum resistance

reproduces

CODES

offer the

over their useful range. Tem-

peratures as measured by the precision platinum resistance thermometer are reproducible to +O.OOl thermometer

TEST

35

The instrument

resistors

The values shown in Table 4.1

the limit of a few which-

is larger.

in resistance

upon

within

of an ohm, or _+0,02 percent

used in measuring

of base-metal

usually

the changes

and industrial

platinum

employs some form of Wheatstone

are based on the time of the resistance thermometers to detect 90 percent of any temperature

bridge circuit and may be either an indicator or a recorder. The bridge may be of the balanced or un-

change in stirred water moving at approximately

balanced type. A potentiometric method of measuring the resistance is used occasionally.

one foot per second.

TABLE

4.1

TYPICAL

CHARACTERISTICS

OF RESISTANCE

THERMOMETERS

Noble Metal Precision

industrial

Sensitivity

0.1 ohm/deg C

0.22 ohm/deg

Precision

9.00

1 deg C

9.3

Accuracy

9.01

deg C

F

deg F

f3.0 deg F Std +l.SdegFSpec

0.22 ohm/deg F

0.22 ohm/deg F 0.186 ohm/deg F

9.1

+O.l deg F

deg F

a.1

deg F

k0.5 deg F Std 20.5 deg F Std a.5 deg F Std N.2 deg F Spec HI.2 deg F Spec fO.2 deg F Spec

KJ.02de F up to 200 $ 9.5 deg F Std fO.2 deg F Spec

Response-Bare

15 set

20 set

40 set

40 set

Response-w/well

30 set

60 set

90 set

90 set

1Sb5Gh””

25 ohms at 32oF

10 ohms at 77°F

100 ohms at 779

100 ohms at 779

Varies

70. 1°c/500c

70.1oc/50oc

excellent

excellent

excellent

Exponential

span

span

-452.2 to 1168.3’=F (-269 to 630.74%)

-297.3 to 195ooF

-100 to 3009

-325 to 3oooF

-100 3009

-100 (-75

Resistance Linearity

Range

40



to

Fast

with units

to SOOOF to 260°C)

INSTRUMENTS 36

Balanced

Bridge

Methods

AND

APPARATUS

well as those resulting

FiR. 4.2 shows a

from unequal leads, may be

typical diagram of a Wheatstone bridge used for

reduced by usinK a resistor

resistance thermometer measurement; (a) and (b) are ratio arms of equal resistance; (r) is a variable resistance, the value of which can be adjusted to

resistance in the thermometer. Where, in the interest of speed of response, this resistance is of the order of 100 ohms or less, a circuit shown in Fig. 4.3 is often used in resistance thermometer instruments.

balance the bridge so that, except for lead resistance, (r) = (x), (x1 being the resistance thermometer resistor.

of several hundred ohms

of the

Copper wires vary in resist-

ance with temperature,

having a temperature

coef-

ficient of the same order of magnitude as that of the thermometer resistor, and, if their resistance is appreciable

in comparison with that of the thermom-

eter resistor,

may introduce large and uncertain

errors into the measurement of temperature.

Since

the thermometer resistor usually must be placed at a considerable distance from the bridge, the resist4.2

FIG. 4.3 FIG. 4.2

SCHEMATIC

OF WHEATSTONE

BRIDGE

CIRCUIT

(x1. Of these, A

neously and by equal distance along the slide-wide

and C should be identical in size, length, and material, and should be placed side by side through-

to balance the bridge. The variable

contacts are

thus placed in the battery and detector circuits

out their length, so as to be alike in temperature. The B wire, which is one of the battery wires,

FOR

S and S, are uniform slide-wires of equal length; S has twice the resistance of S,. The contacts of the detector and battery leads are moved simulta-

illustrates one method of accomplishing this result. Three wires A, B, C connect the measuring instrument and the thermometer resistor

WHEATSTONE BRIDGE CIRCUIT INDICATORS OR RECORDERS

where they can have no effect on the balance point of the bridge, while a one-to-one bridge ratio is

need

not be similar to the others, but it is common prac-

constantly

tice to form the three wires into a cable and make them all alike. A and C are in the thermometer re-

maintained.

A scale associated

with the

slide-wire may be graduated in degrees (C or F), or in ohms. If the scale is graduated in degrees, a

sistor arm (xl, and the variable resistance arm (r-1, respectively. Their resistance remains equal al-

winding properly adjusted to match resistance

and

temperature coefficient

of the material

and hence, with a one-to-one bridge ratio, such

the bridge is calibrated

must be used as the sensi-

changes have no effect on the bridge reading.

tive element.

though their temperature

conditions may change,

38 37

It is desirable

sistances

to have no variable

in the bridge arms, because the variations

The effect of these variations,

Bridge

Method.

eter resistor located at the point where the temperature is to be measured. D and E are equal ratio arms. P is a fixed resistor of a value equal to that

as


This method is

shown schematically in Figs. 4.4 and 4.5. A, B, C, represent the terminals of the resistance thermom-

contact re-

in bridge balance introduced at the contacts may be sufficient to seriously effect the reliability of the measurements.

Unbalanced

for which


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

ance of the wires must be compensated. Fig.

ASME PERFORMANCE of the winding at the highest

temperature

TEST

CODES

on the

indicator scale, while y is a fixed resistor equal in resistance to the winding at the lowest temperature on the scale. switch

To standardize

the current the

is thrown to the side marked “Std,” and the is adjusted until the indi“R.H.”

battery rheostat

to the lowest

temperature

reading on

made at intervals

depending

upon the stability

PII

of

P

the current source, The method has the advantage that the temperature

is constantly

scale without requiring

indicated

manual or automatic

on the

of the bridge. It may be used with indicating, cording,

or controlling

instruments.

the usual limitations is not adaptable

of deflection

to precision

FIG. 4.6

POTENTIOMETRIC

METHOD

balance re-

It is subject instruments

spectively, to

tiometer

and

are measured by means of the poten-

P. The resistance

R, is determined

from

the relation

measurements.

Rt=Rs.%

Or, if Rs is adjustable,

es may be

es made equal to et by adjusting

R,,

in which case

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

cator deflects

the scale. After standardizing, the switch is thrown to the measuring position. This check should be

Rr = R, and R, is read directly from the calibrated dials of R,. It is essential that the current through R s and R, remain constant during the two potentiometer balances required. Consequently, in the second method, it is necessary

C

change resulting

0

insignificant.

to check that the current

from adjustment

Rr, this condition will be more easily satisfied. The resistance measurement by means of this

*E

FIG. 4.4

UNBALANCED DC SUPPLY

BRIDGE

METHOD-

method is independent

of the lead resistance.

APPLICATION

AND INSTALLATION

Sources 40

radiation

-,,

A

Ar

FIG. 4.5

thermometer

errors than the thermocouple

Essential UNBALANCED AC SUPPLY

Potentiometric

large sensing

area,

is more susceptible

to

when the

temperature of a gas is to be measured in a pipe or duct with the walls at a considerably different temperature from that of the gas.

‘ s l

of Error

Because of its inherently

the resistance

39

or R, has been

If r, is large compared with Rs and

Method.

BRIDGE

METHOD-

Considerations

41 The proper value of current to be used in a given apparatus will usually be specified by the manufacturer. Since the thermometer resistor is in-

The potentiometric

sulated electrically,

and to some extent,

thermally,

method of resistance measurement is applicable to four-lead resistance thermometers only, Fig. 4.6.

its temperature is raised by the measuring current to something above that of its surroundings. There-

The current,

fore, the current must be kept small enough to avoid causing a rise in temperature which is more than a small fraction of the limit of error of the

adjusted

to a suitable

value by resis-

tor r,, flows through a standard resistor

(or resist-

ance box) R, and the thermometer resistor RI. The potential drops e, and er, through R, and Rt re-

measuring equipment.

42



To test the degree of com-

INSTRUMENTS

AND APPARATUS ADVANTAGES AND DISADV’ANTAGES

pliance with this condition, the thermometer may be placed in melting ice and the bridge balanced, using a measured bridge current. When a steady reading is obtained,

44

b-4)By

the current should be reduced

to half of its original

value and, if necessary,

the

-450

Ii (4

should be reduced to such a value that further recation of the instrument.

resistance

thermometers from

to 195oV.

(b) High accuracy.

made. The current

duction produces no observable

proper selection,

may be used to cover ranges extending

bridge rebalanced. If a significant change in resistance is observed, the current is again reduced by half and a new measurement

Advantages

change in the indi-

For the sake of sensitivity,

Excellent stability and reproducibility. Interchangeable. Can be matched to close tolerances for temperature difference measurements.

the bridge current used should be as large as permissible.

45

42 The indicating

or recording

instrument

may be

Disadvantages

located at a distance from the point at which the temperature is to be measured. Connection is made

More mass than thermocouple. Relatively slow response. Subject to mechanical damage if not properly

by means of a three or four conductor copper cable

handled.

covered with rubber or lead. The conductors are

Higher cost than thermocouple.

usually

16 or 18 AWG copper individually

with rubber or more temperature

insulated

resistant

material.

Each wire must be of the same length in order to assure equality

of resistance

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

temperature

REFERENCES

of the compensating

leads even though they may be subject

to uniform 46 Throughout the text Reference enclosed in brackets, thus [II.

conditions. Treatment

[ 11“The

of Data

International Practice-Temperature Scale of 1963,” Comite International des Poids et Measures, Metrologie vol. 5, No. 2 p. 35, Apr. 1%9. [2] “International Practical Temperature Scale.of 1968.” Benidict, RP, L&N Technical Journal, Spring 1%9. [3] Published in 1932 as National Bureau of Research, Paper 508, (vol. 9. p. 8071. [4] “Wh ea t a t one Bridges and Some Accessory Apparatus For Resistance Thermometry,” by E.F. Mueller, Scientific Papers of the Bureau of Standards, vol. 13, p. 547, 1916.

43 The observed temperature readings need not be corrected provided the resistance thermometer is

immersed, tions,

in accordance

with manufacturer’s

in the medium whose temperature

measured.

Corrections

evaluated

by periodic

thermometer

direc-

is being

for drift in calibration checking

may be

of the resistance

at the ice point and comparing these

checks to the original

ice point resistance.


numbers are


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

CHAPTER 5, LIQUID-IN-GLASS THERMOMETERS CONTENTS

3 Par.

GENERAL: Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . PRINCIPLES OF OPERATION . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CLASSIFICATION: Description . . . . . . . . . ..... .......................... ................ .......... Materials of Construction .., . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . CHARACTERISTICS: Range . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sensitivity Precision . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Accuracy .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... . . . ... . ... .. . . .. . ... . .. . . ... . ... .. ... ... .. .. .. ... . ... . .. .. .. .. .. Response ACCESSORIES . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . APPLICATION AND INSTALLATION: Sources of Error . . . . . . . . . .. . . . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . . . . . . . . . . .......................................... Essential Considerations Treatment of Data ............ ...................................... .... ADVANTAGES AND DISADVANTAGES: Advantages . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . Disadvantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . .. . . . . . REFERENCE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ............. ....................

A Partial

Immersion

one which is designed

(Fig. 5.1) is temperature cor-

Thermometer

to indicate

rectly when used with the bulb and a specified 1 2 6

part of the liquid

column in the stem exposed to the

temperature being measured, the remainder of the liquid column and the gas above the liquid exposed

7

to a temperature

?O

4 A Total

25 26 27 28 31 32

which may or may not be different.

Immersion

one which is designed

(Fig.

Thermometer

to indicate

5.1) is

temperature COP

rectly when used with the bulb and the entire liquid column in the stem exposed to the temperature

34 45 46 49 50 51

GENERAL Scope

lmmrrrion Linr 1 The purpose of this chapter is to present infor

mation which will installation,

guide the user in the selection,

and use of liquid-in-glass

thermom-

eters. Definitions 2

A Liquid-in-

Gloss

ing of a thin-walled capillary

Thermometer

glass

is one consist-

bulb attached to a glass

stem closed at the opposite end, with the

bulb and a portion of the stem filled

with an ex-

pansive liquid, the remaining part of the stem being filled with the vapor of the liquid or a mixture of this vapor and an inert gas. Associated stem is a scale in temperature

with the

degrees so arranged

that when calibrated the reading corresponding to the end of the liquid column indicates the tempera-

FIG. 5.1

PARTIAL, IMMERSION

ture of the bulb.


be-

ing measured, and the gas above the liquid exposed to a temperature which may or may not be different.


TOTAL

AND COMPLETE

THERMOMETER

INSTRUMENTS 5

A Complete

Immersion

(Fig.

Thermometer

AND APPARATUS

5.1)

is one which is designed

correctly

to indicate temperature when used with the bulb, entire liquid

in the stem, and gas above the liquid ex-

column

posed to the temperature being measured. PRINCIPLES

6 The operation

OF OPERATION

of a liquid-in-glass

depends upon the coefficient

thermometer

of expansion

of the

liquid being greater ‘ .than that of the bulb glass. As a consequence, an increase in the temperature of the bulb causes liquid to be expelled from the bulb, resulting in a rise in position of the end of the liquid column. The capillary

stem attached to the

bulb serves to magnify this change in volume on a scale.

CLASSIFICATION

Description 7

Etched

Stem Loborotory

Thermometer.

As the

name suggests, the scale is marked directly on the stem by etching. The etched stem marks are made legible

by filling

with a pigment material

of such

composition as to adhere to the etched surfaces or be chemically bonded as by fusion. A bare, total immersion thermometer, A partial

immersion

is illustrated

in Fig.

5.2.

thermometer mounted in an open

face armor is illustrated

in Fig.

5.3. FIG. 5.2

8

Industrial

Type Thermometer.

in a

10 The case-stem

metal tube while the scale section is contained in an attached metal case. The scale is engraved or

thermometers

are available

closed by

ature is being measured, socket)

nut

and union bushing connection. 12

Tube-and-Scale

Type

form of tube-and-scale

or it may be inserted in a

inscribed

which in turn is immersed.

a threaded swivel

angles are the 180 deg or

a 90 deg back angle thermometer with swivel

in a vari-

within

Thermometer.

thermometer,

a protecting

13 A special

Union bushing and flange connections are also available as alternative means of mounting. Some

or milk

attached to the stem and mounted glass sheath. In the better

grades, the bulb is not contained within

nut connection.

In one

the scale is

on a piece of paper, cardboard,

glass suitably

Where the thermometer is mounted in an essentially permanent manner, the extension of the bulb assembly incorporates

THERMOMETER

11 Fig. 5.4 shows a straight form thermometer with swivel nut mounted in a well. Fig. 5.5 shows

ety of stem lengths, case sizes, and case-stem angles. The bulb chamber or sensitive portion may be immersed directly in the medium whose temperwell (separable

IMMERSION

straight, the 90 deg back angle, 90 deg right and left side angles, and various oblique angles.

to the inside of

the case. The case opening is generally a glass window. 9 Industrial

TOTAL

In this type, the

bulb and a portion of the stem are enclosed

printed on metal plates fastened

BARE,

the sheath.

form of this type is the Beckman

types are used at various immersions and are termed plain bulb style. NO threaded connection is included

Differential Thermometer usually made with a short range such as 5 deg C and a very open scale haviug, for example, 0.01 deg C subdivisions. The

with this type.

range can be varied at will by changing the amount

45 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



ASME PERFORMANCE

TEST

CODES

FIG. 5.4

15

STRAIGHT INDUSTRIAL THERMOMETER WITH SWIVEL NUT, MOUNTED IN A WELL

A further form is the so-called

window or

wall type, in which the scale is printed on a wood, metal or plastic

back to which the tube is fastened.

This type is commonly used for measuring FIR.

5.3

PARTIAL MOUNTED

peratures

IMMERSION THERMOMETER IN AN OPEN FACE ARMOR

air tem-

both indoors and outdoors.

16 Registering

Type

form of liquid-in-glass

Thermometer.

thermometer

The common is nonregister-

of mercury in the bulb, any excess being retained in a reservoir at the top. This type is used fre-

ing and must be read while immersed in the medium whose temperature is being measured. Thermom-

quently in calorimetry,

eters of a registering type are used for the measurement of temperature in locations where the ther-

in the bomb

14 Another form of tube-and-scale is the tin-case

or cup-case.

mometer can be observed only after it has been taken from the medium whose temperature is being

thermometer

In this form the tube

measured. These thermometers etched stem type.

is mounted either on an engraved metal scale, which in turn is held in place in an open face

17 The maximum-registering

case, or the tube and scale are independently mounted on a suitable

cury (or mercury-thallim

support, often made of wood,

indicates

to which a cup is attached in which the thermometer is immersed. The cup type is used for measur ing temperatures of liquids in inaccessible points such as in large storage tanks. The thermometer is

the maximum temperature

Built

which allows

on rising

temperatures,

from returning

nary force is applied,

the bulb at

being measured.

mer-

to which the to resetting.

mercury to squeeze but prevents

through

the mercury

to the bulb except when extraordi-

bulb.


contains

into the bore just above the bulb is a con-

striction

the liquid

the temperature

type

of the

under vacuum and

bulb has been exposed subsequent

dipped into the tank and is allowed to come to thermal equilibrium. As it is withdrawn for reading, in the cup helps to maintain

alloy)

are generally


as by shaking toward the

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

particularly

method for fuels.

INSTRUMENTSAND APPARATUS vented by the action of the spring. be reset by a magnet. Materials

The index can

of Construction

20 Glasses. In order to obtain optimum performance characteristics, various criteria should be considered glasses

in the selection

of materials.

The

of bulb and stem should be compatible

permit uniting with a well knit, strong joint.

to

They

should not soften or become excessively brittle in the temperature range of the thermometer. The bulb glass in particular should be of a formulation which lends itself to dimensional stabilization or aging by heat treatment. The stem glass should lend itself to drawing into capillary degree of uniformity from optical

tube form with a high

of bore diameter

and freedom

distortion.

21 The range of liquid-in-glass thermometers is limited by the glass and the liquid used. The safe upper limit of several shown in Table

TABLE FIG.

5.5

90 DEG BACK ANGLE INDUSTRIAL THERMOMETER WITH SWIVEL NUT UNION

BUSHING

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

18 The minimum-registering

been exposed subsequent

Exposure

alCorning Normal 7560

to resetting.

Kimble R 6 Jena 16 III Corning Borosilicate 8800 Jeno Borosilicate 2954 Corning 1720 Jena Supremax 2955

It contains This

index

is carried toward the bulb by surface forces at the end of the liquid column on falling in position

temperatures,

on rising temperatures.

aFrom Reference [l]. b405OC or 760°F if Corning glass is used for the stem.

19 A combination maximum-minimum thermometer, identified as a Six’s Thermometer (invented by is built in a U shape and contains

limits

Continuous

Intermittent

OC 370 360 365 400 420 540 535

“C b430 420 425 460 480 600 595

OF 700 680 690 750 790 1005 995

OF baoS 790 795 860 900 1110 1100

The

index is reset by tipping the bulb upward.

James Six),

TEMPERATURE EXPOSURE LIMITS FOR VARIOUS THERMOMETER GLASSES’

AND

type is usually

index submerged in the liquid.

but remains

in common use is

CONNECTION

cohol filled and is used horizontally. It indicates the minimum temperature to which the bulb has a glass

5.1

glasses

5.1.

Standard

Thermometer

0041

a

maximum and a minimum index. The temperature 22

Liquids.

The liquid should be of a high state

sensing bulb at the end of one limb is filled with a beechwood creosote-alcohol mixture. The U is

of purity to insure constancy

closed by mercury which positions

teristics.

index on changes in temperature. is partially

filled

one or the other

serves as a lubricant

bute. If organic liquids

which

cases whe.re visibility

for the maximum registering

alloy have this attri-

are used, as for example

in

of a mercury thermometer

is

with a minimum amount of “film

holdup”

and

on falling

temperatures. Such liquids should be chemically stable and lend themselves to coloring with light-

is pre-

47


charac-

poor, the liquid should wet the bore uniformly

index. Each index consists of a closed glass tube containing a piece of iron wire. Attached to one end of the index is a glass spring. As the mercury moves, it forces the index upward. Falling

of expansion

the liquid should not wet the bore.

Mercury and mercury-thallium

The other limb

with the creosote-alcohol

Ideally


ASMEPERFORMANCETESTCODES fast dyes. The working ranges of several common use are shown in Table 5.2.

TABLE

5.2

WORKING TEMPERATURE FOR LIQUIDS COMMONLY

Working Range -38

27 Precision of measurement of temperature with liquid-in-glass thermometers depends upon the thermometer design and the application conditions

in

as well as the care exercised

RANGE USED

28

Liquid

to 250’F to 450°F

Accuracy

in reading. measurement

is de-

pendent upon the same factors which affect

of temperature

preci-

sion, and in addition standardization

Mercury Mercury-thallium Organic liquids

to 1150°F

-56 -328

liquids

the accuracy

and periodic

of calibration

evaluation

changes in the bulb glass. With well designed mometers the accuracy

of calibration

or

of secular ther-

is a function

of range and graduation interval. National Bureau of Standards certification tolerances for such thermometers are listed

23 Gases. When gas is used above the liquid, the gas should be inert to the liquid. Nitrogen and carbon dioxide

are commonly used with mercury,

while hydrogen

is usually

thallium.

29

the choice with mercury-

For most organic liquids

24 Metals. semblies,

air is sufficient-

stainless

steel,

fication

and

used for the case. Heat transfer

normally included

in the bulb chamber. The most are mercury, copper dust, graph-

dust. Selection

range of the thermometer

is determined

and the material

limits

by the

limited

of the materials

of the

by the application

of construction,

of accuracy

re-

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

26

will

Selection

of ther-

Sensitivity

be helpful

is determined

industrial

in choosing

it is preferable such as a

if a higher order

Because

of the uncertainty of an emergent

thermometer

manufacture

are significant

establishing;

minimum bore diameters.

factors

equivalent

design total immer-

The etched stem form is to be of the highest

and tube-and-scale

accuracy.

forms are affected

The by

31 Response of liquid-in-glass thermometers is a function of thermometer design and use conditions.

by the cross-section

In most cases,

the response

time-temperature

rela-

tionship is exponential and a single value, most commonly the 63 percent response, is used in evalu-

tions frequently limit the length of scale and the size of bulb. Practical limitations of tubing and

ation of this variable.

in

Sensitivities

or accuracy

ACCESSORIES

of meas-

urement are misleading and generally are costly, particularly if specially designed thermometers are

32 Wells or sockets are the major accessories for liquid-in-glass thermometers. For a general discussion of wells refer to Chapter 1, Pars. 8 through 19.

being used.

48


the

heat conduction of the parts other than the glass tube. Such heat transfer is difficult to measure accurately and result in uncertainties in temperature measurement.

area of the capillary bore and the proportion of that area to the volume of the bulb. Application condi-

of the precision

to detail 5.3 and 5.4

device,

of the temperature

relied upon for results

ranges.

far in excess

is required.

sion thermometers.

broad or narrow ranges or

frequently

but usually

thermometer,

curacy as otherwise

characteristics

very fine degree of subdivision, will add significantly to the cost. Review of manufacturers standspecial

in Tables

column, partial immersion thermometers generally cannot be expected to give results of the same ac-

by the physical

mometers with extremely

listed

platinum resistance of measurement

Range is determined

ard Listings

of accuracy

to use another type of measuring

CHARACTERISTICS

25

are rounded off.

might be made smaller,

bulb chamber.

quirements

values

30 With extreme care and attention

media are

common materials ite, and silver

acceptable

exercised in the use of the thermometers. “Corrections stated to” are the limits to which NBS certi-

brass are the most common. Brass and aluminum are generally

5.3 and 5.4 for

represent

proper attention to such details as maintenance of correct immersion, avoidance of parallax, etc., are

metals are used for bulb as-

although steel,

shown in Tables

in degrees”

of error of uncertified thermometers. “Accuracy in degrees” is the limit of error to be anticipated when corrections are applied, and when

in the operating range.

Various

5.3 and 5.4.

limits

ly inert. The pressure of the gas should be high enough to minimize the vaporization of the liquid at any temperature

The values

“Tolerance

in Tables


INSTRUMENTSANDAPPARATUS TABLE

Temperature

5.3

TOLERANCES

FOR

LABORATORY

THERMOMETERS

FAHRENHEIT

Graduation

Range

Interval

in Degrees

in

Degrees

to 32

-35

to 32

1

Tolerance

Accuracy in Degrees

for Low

Not

0.1

-0.2

0.02

Above

300

Dog

1

0.2

-0.5

0.2

1

or 0.5

1

0.1

-0.2

0.1

0.2

or 0.1

0.5

0.02-0.05

Thermometers

2

I

Corrections Stated to

0.05

Graduated

2

32 up to 212 Above 212 up to 600

0.1

0.5

Thermometers

IMMERSION

Temperatures

1

or 0.5 0.2

32 up to 300 32 up to 300 32 up to 212

TOTAL

in Degrees

Thermometers -35

MERCURIAL

Not

Groduoted

i

or 1 Thermometers

600

Dog 0.2

1 2

0.2

-0.5

0.2

0.5

Graduated

32 up to 600

Above

0.02

Above

600

Dog

4

0.5

-1.0

0.5

5

Above 600 up to 950

t

7

1

-2

0.5

32 up to 600

I

3

0.2

-1.0

0.2

6

0.5

-1.0

0.2

2

Above 600 up to 950

TABLE

Temperature

5.4

or1

TOLERANCES

FOR

LABORATORY

THERMOMETERS

FAHRENHEIT

Graduation

Range

Interval

in Degrees

in

Degrees

Tolerance

Accuracy in Degrees

for Low

Thermometers

Not

Graduated

0.3-0.5 Above

300

Thermometers

Not

Graduated

Corrections Stoted

Above

600

0.1

Dog

0.2-1.0

2

2 or 1

32 up to 300

IMMERSION

Temperatures

1

1

to 32

PARTIAL

in Degrees

Thermometers

-35

MERCURIAL

0.2

Dog

32 up to 212

2 or 1

2

0.2-0.5

0.2

Above 212 up to 600

2 or 1

3

1

0.5

Thermometers

Groduoted

Above

32 up to 600

600

-2

Deo

1

-2

1

2

-3

1

5 or 2 Above 600 up to 950

i

I

10

49 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



to

ASME PERFORMANCE 33

As an aid to accurate

reading,

of parallax

CODES

t, = temperature* deg F

small magnifiers

are sometimes used. These are usually clipped on the stem, but must be positioned carefully to avoid the introduction

TEST

indicated

by the thermometer in

t, = average temperature in deg F of the exposed mercury column. Values ,of t, are usually

error.

measured by means of an auxiliary APPLICATION

eter mounted as shown in Fig.

AND INSTALLATION

Sources

of Error

IN SERVICE

-_---------_ --_ --: --_---4 E ---------4_---_---_ ---_ -----_ --E-----. _--_ ----. ---_ -----------_ ---3 ---_ ----. _--_ ----__ ---1---_

34 Liquid-in-glass thermometers are subject to variations in manufacture which necessitate the determination and application of instrument corrections for accurate

results.

ment for such calibration

Techniques

and equip

are discussed

in detail

CALIBRATION

DURING

in

Chapter 9.

-Z.-r---_-----_-----

35

Sources-of

__--_---- -- 4 __----_-----.-----.---_ _ --- _----.------.--------_- -_--_---_--- --._---_-----_-----_--____------_------_----_-_--_---_---. - i _-------i -------_-- _--_----

error may be present in the use of

liquid-in-glass thermometers which do not enter into either the original manufacture or subsequent calibration. These should be taken into consideration in Performance Test Codes work and the corrections

determined

should be applied when sig-

The most frequently

encountered

source of

error is the misuse of the emergent-stem correction. This correction

derives

from use of the thennom-

eter with a portion of the stem exposed to a different temperature

from that of calibration.

example is the use at partial mometer calibrated

--e-_-1 ----_-------_----

A common This

cor-

(a‘

rection may be quite large if the number of degrees emergent and the difference

!

j

--_--_ ----_ -_--F---.

between the tempera-

FIG. 5.6

ture of the bath and the space above it are large. For example, at’ a bath temperature of 7500F a total immersion etched stem thermometer used at 3 in.

-----------_ _--------_--_ .-- _---. ------_----------_-----------_ _------YzI-2: _----_-_----_ _----i -----. _--__--_----. _---_----_--_ --_ -------_-- -- _--__---i------------_---- ----_------_- ----_ _-----_----_ _--__---------. _--_--------_----__--_ --_---__-------_----. ---i --_--_ _---2

A

1

---_ ----

---- -. _---1F -----_-_ ---_

----_----_-_----. ----_---

immersion of a ther-

for total immersion.

-------. -----------_-Lx-~

A

nificant. 36

thermom-

5.6.

----_-------_ -------_ -_ b -----.

1

----_---_--_-_---_-_---_--_---_

---

--------_ ---_

a-_--------. ----. -----

~

THERMOMETER CALIBRATED FOR TOTAL IMMERSION AND USED FOR PARTIAL IMMERSION

immersion may be in error by as much as 35v. emergent-stem

correction

The correction K for various values of D, t,, and t, may also be obtained from the chart, Fig. 5.7.

is to be added

algebraically to the indicated temperature of a liquid-in-glass thermometer. For a total immersion mercury-in-glass thermometer it can be calculated from the following

38 If a partial immersion thermometer is calibrated for specified emergent-stem temperatures and is used under other conditions,

equation:

K = 0.00009 D (t,

D=

using LI + K as the new value for tl, the new correction K will be nearer correct than the first value. Further re-

in deg F

calculation with tl corrected for the new value of K will result in a more correct value for K. Seldom are more than two recalculations necessary and then only for high temperatures and long emergent-stems.

emergent-stem, which is the length of ex-_ posed mercury column, expressed in depl F on thermometer stem


should be deter-

*Inasmuch as the t( is not the true temperature of the bulb of the immersed thermometer, the correction K is only approximate upon the first substitution in the above equation. If a new substitution in the equation is made

-t,)

where

K = correction

a correction


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

37 Ihe

14H1 SIllI

--

I

II-

I -

---

I

0 bserved

Temperature

FIG. 5.7

EMERGENT-STEM

M CORRECTIONS

FOR LIQUID-IN-GLASS

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



THERMOMETERS

ASME PERFORMANCE mined for this difference

in emergent-stem

ture. For a mercury-in-glass calculated

thermometer

from the following

TEST

CODES

tempera-

separations

are not apparent in the scale section.

it can be

Comparison

against

a standardized

permits ready detection

equation:

43

K = 0.00009 D (t, -to)

When high temperature

range thermometers

are heated at the upper limits

where K and D have the same significance

glass may become plastic

as in

stretch.

Par. 37 and average temperature exposed mercury column

in deg F of the

44

of the range, the

and allow the bulb to

Such damage can be detected

calibration

t, = specified

at any convenient

exposed

in deg F of the

The changes which occur in thermometer

within

mercury column.

The general

account. The diameter

The techniques

of the medium being

considerations

Pars. 36 and 37 of Chapter

of good manufacture

given in

of use. The achievement

of the

conditions

of perfect

of use, however,

stability

for all

is not possible

in ther-

thermometer bulb should be as small as possible consistent with other features of instrument design

mometer manufacture

and performance.

changes observed in scale readings

The filling

liquid

readings

in the thermom-

eter should have as low a heat capacity as possible recognizing however that for most accurate

so that changes in ice point

with time and use are observed.

The

at the ice point

reflect changes of the same magnitude and sign at all points on the scale since they are the result of

work mercury should be used.

changes in bulb volume; changes in the stem have very little

40

are designed

to produce in the thermometer glass a state which will result in maximum stability at the temperature

1 should be taken into

and wall thickness

of time

perature and the rate of cooling.

is the dynamic error or the lag of the thermometer measured.

function

and temperature and will depend upon the thermal history of the glass, both during manufacture and previous use, the time of exposure to the high tem-

39 A second factor which should be considered the temperature

bulb

high but still

its intended range of use, and subsequent

cooling to ambient are an involved

The correction so determined should be added algebraically to the indicated temperature.

in indicating

only by re-

temperature.

glass on heating to a temperature,

t, = observed averaae temperature

thermometer

of such faults.

effect.

A third factor which may be of significance

as a source of error is the effect

of external

sure. If the bulb of the thermometer pressures

appreciably

different

pressure,

the bulb volume will

elasticity

of the glass. Experience

The changes in bulb volume are of two kinds re-

pres-

sulting

is exposed to

(a)

change due to the

from the behavior

of glass.

it should be evaluated

Temporary Changes. Upon heating to high temperature the bulb expands from its initial

.

has shown that

state and, after a short period of time, appears

for bulb diameters of 5 to 7 mm as commonly used in liquid-in-glass thermometers, the pressure coefficient is approximately 0.2 deg F per atmosphere. If the conditions of test are such that this factor is significant,

naturally

from atmospheric

to reach an equilibrium to that particular

condition

corresponding

high temperature.

If the ther-

mometer is then cooled sufficiently through critical

and a correction

temperature

slowly

regions,

the glass

applied.

will return to close to its initial state, and the ice point reading will show no change on this

41 On most industrial and many tube-and-scale types of thermometers, it is possible to effect rela-

is cooled rapidly as, for example, naturally

tive displacement of the tube with respect to the scale. Hence, for maximum accuracy a reference

portion of its expanded condition, and the icepoint reading will be lower than its reading be-

account. If, on the other hand, the thermometer

to a specific inspection

scale graduation. and adjustment

This

fore the heating. This phenomenon is known as “zero, or ice-point depression.” Thermometers

permits ready

if such displacement

which have been heated to high temperatures recover from this ice-point depression in an un-

has taken place. 42

If industrial

ly, separations

thermometers

predictable

are handled rough-

way, and frequently

no significant

of the liquid may occur and such

recovery

52


cooling

air, the bulb will retain a

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

mark should be engraved on the tube corresponding

in still


there will

after a year’s

be

time at

INSTRUMENTSAND APPARATUS room temperature.

The icepoint

depression

has

have been repeatedly

at low tempera-

tures, for exam.ple between

taken from time-to-time immediately (within about 1 hr) following cooling in this manner,

ally progresses more rapidly at first, but, with continued heating, tends toward lower rate of

may be used reliably

change with time. The rate of secular change will be dependent upon the kind of glass used in the thermometer bulb and the particular

At high temperatures

to show changes in ther-

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

mometer bulb volume with time and use. On the other hand, thermometers used only up to about 100 deg C will usually exhibit a relatively rapid recovery

from the ice-point

-30

and +25’C.

the secular

change usu-

heat treatment given the thermometer in manufacture. Thermometers manufactured according

depression,

to good practices will evidence only small secular changes but thermometers made of

and the original bulb volume will be recovered within the equivalent of 0.01 or 0.02 deg C in about 3 days. This phenomenon has an important

glass unsuitable

bearing on the precision

improperly annealed, may show changes as large as 20 deg C (36 deg F) after continued

cury thermometers

attainable

with mer-

and must be taken into con-

for the use temperature,

sideration in precision thermometry, especially in the interval 0 to 100 deg C. Thus, if a thermom-

ing at high temperature

;eter is used to measure a given temperature, will read lower than it otherwise would if it

care must be taken to avoid overheating.

has a short time previously higher temperature. thermometric

In the use of high-temperature

it

been exposed to a

will

the error resulting not exceed

thermometer

has recently

with the best glasses

erratic

at temperatures

the bulb resulting

Essential

to which the 45

only a few thousandths

Considerations

When using a liquid-in-glass

following (a)

Decide where to place the bulb of the thennometer, considering: (1) Is the temperature at the place selected

much above 100 deg C. For

representative

of the information

and (2) Can the thermometer

scale correction based upon an icepoint reading taken immediately after the temperature

temperature

measurement. (b)

A second type of change

wanted,

attain or assume the

of the medium if so placed.

Select the thermometer considering ing factors: (1) Range and graduation

in thermometer glasses, known as the “secudelar change,” results in a nonrecoverable

the follow-

interval.

(2) I mmersion. (3) Sensitivity and response time.

crease in bulb volume which may progress with time even at room temperature, but which

(4) Form, i.e., etched stem, industrial etc. (5) If a well is to be used, it should be of the

is markedly accelerated at high temperatures. This type of change is evidenced by an increase in the ice-point reading. At low to moderate

right material, shape, and size. (6) The thermometer should be capable of

temperatures there may be a gradual change which will continue for years. With better

giving results of the accuracy desired but

grades of thermometer glasses the change will not exceed 0.1 deg C in many years, provided the thermometer has not been heated to temperatures above about 150°C. In addition, permanent changes in bulb volumes have somewhich

not appreciably (c)

better.

Install the thermometer properly and use care in reading, avoiding the following: (1) External heat sources or sinks near the thermometer which might affect its indication.

53


thermometer the

must be observed:

The

become somewhat

times been observed with thermometers

of

in lower thermometer indica-

tions.

the reasons briefly set forth above it is customary, in precision thermometry, to apply a

Permanent Changes.

In

column may cause a permanent distortion

been exposed and

errors due to this hysteresis

thermometers

eter, the built-up gas pressure above the liquid

from

(in the inter-

of a degree for each 10 deg difference.

heat-

higher than the intended range of the thermom-

val 0 to 100) 0.01 of a degree for each 10 deg difference between the temperature being measured and the higher temperature

or

only a few minutes of heating at a temperature

With the better grades of

glasses

this hysteresis

(b)

cycled

a reproducible value, however, for a thermometer cooled in still air, so that the ice point,


ASME PERFORMANCE (2) Parallax

TEST

t2 = 125

in reading.

(3) Inadequate

CODES

K, = (0.00009)(287.7)(352.7-125)

illumination.

= (0.00009)(287.7)(227.7) Treatment

= 5.8 deg F

of Data

46 The observed temperature readings should be corrected for instrumental errors using the calibration correction

values.

other than calibration termined

Corrections

in correction

352.7 t 5.8 = 358.59:

should be de-

Corrections

47

for

Emergent-stem,

+K, = 352,7+5.8

= (0.00009)(287.7)(233.5) = 6.0 deg F

pressure

and applied

= 352.7 t 6.0 = 358.7oF

True temperature

when

In this installation

the external

Example:

A total immersion

effect

imADVANTAGES

corrections are -0.5 deg F at 32 deg F, + 1.5 deg F at 300 deg F and 0.0 deg F at 4OO?F. An auxiliary thermometer

used to measure the temperature

49

Advantages.

The advantages

(a)

Available tivities,

with wide variety

(b) Simple to use.

the test well of 65?‘.

ice point reading on the

taken after the test was 33.0%.

Ice point correction:

32.0-33.0

= -1.0

-LO-(-0.5)

= -0.5

Interpolated

power supply required.

corrections: +1.5-0.5

= t1.0

deg F at 300°F

0.0-0.5

= -0.5

deg F at 400°F

correction

50

Disadvantages.

in-glass (a)

at 352.5 OF:

The disadvantages

thermometers

Relatively

+l.O-(-0.5) tl.O-0.8

1

fragile. in superheated

steam tempera-

ture range.

= 0.8 deg F

= t0.2

of liquid-

are

(b) Least reliable $$$-

except for over-

ranging at high temperatures.

deg F.

No auxiliary calibration

check.

Relatively inexpensive. Damage readily apparent,

Change in ice point correction:

sensi-

Calibration constant, except for drift in range span which can be measured readily by reference temperature

deg F.

of ranges,

and accuracies.

(c)

main thermometer

of liquid-in-

are

of

immersed and it had a scale reading at the top of The

AND DISADVANTAGES

glass thermometers

the emergent stem read 125 deg F. The total immersion etched stem thermometer was not totally

(c) deg F

Upper range limits tures encountered

reading:

(d)

work. Not adaptable

do not include in Performance

all temperaTest Code

to remote reading.

352.5 + 0.2 = 352.7 REFERENCE

Emergent

is

etched stem ther-

when partially

mersed and mounted in a well in a steam line indicated a temperature of 352.5T. The calibration

Corrected

pressure

negligible.

mometer range 30 to 400eF,

Revised

= 358.5

L, = 125 K = (0.00009)(287.7)(385.5-125)

to all other cor-

lag, and external

should be calculated

of emergent stem correction:

D = 287.7 tt

values.

corrections necessary. 48

at this temperature

to true temperature

Second calculation

drift in calibration may be evaluated by periodic checking of the ice point or other convenient reference temperature and applying the observed change rection

approximation

at temperatures

temperatures

by linear interpolation.

First

stem correction:

D = 352.7-65 t, = 352.7

51

= 287.7

[I] James F. Swindells, “Monograph 90,” National Bureau of Standards, Feb. 1965.

54 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



CHAPTER 6, FILLED SYSTEM THERMOMETERS changes in shape when internal

CONTENTS PZU.

pressure or volume

changes are applied.

GENERAL: Scope . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Definitions . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 .................................. 4 PRINCIPLES OF OPERATION CLASSIFICATION: General Classification . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 5 Subclassification . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Description Materials of Construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 28 CHARACTERISTICS: ........................ 31 Maximum and Minimum Temperature Range . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Accuracy . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .................... 41 Temperature Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Response . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 ACCESSORIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 APPLICATION AND INSTALLATION: Sources of Error . . . . . . . . . . . . . . . . . . . . . . . . . . ,............................... 55 Essential Considerations . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 ADVANTAGES AND DISADVANTAGES: Advantages . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 Disadvantages REFERENCES . . . .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

PRINCIPLES

OF OPERATION

4 The sensing element (bulb) contains which changes in physical temperature.

a fluid

characteristics

with

This change is communicated

Bourdon through a capillary

movement provides an essentially motion through mechanical

to the

tube. The Bourdon linear pointer

linkages

ments. Bourdon motion is directly

in some instrurelated

to:

GENERAL Scope 1 The purpose of this chapter is to present information which will guide the user in the selection,

installation

and use of filled

system thermometers.

Definitions 2

A Filled

assembly

System

consisting

Thermometer is an all metal of a bulb, capillary tube and

Bourdon tube,* containing a temperature responsive fill. A mechanical device associated with the Bourdon is designed to provide an indication or record of temperature.

See Fig.

6.1.

FIG. 6.1

FILLED

SYSTEM

THERMOMETER

consists of a closed and flat3 A Bourdon [I]** tened tube formed into a spiral, helix or arc, which (a) * In the interest of brevity, hereafter referred to as Bourdon. ** Numbers Chapter,

in brackets

t11.

designate

References

Volume change of a liquid within

(b) Pressure (c)

at end of

change of a gas within the bulb.

Vapor pressure change of a volatile within the bulb.

55 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


the bulb.


liquid

ASME PERFORMANCE

TEST

CODES AUXILIARY COMPENSATING CAPILLARY

CLASSIFICATION

General

Clossificotion

5 Filled system thermometers may be separated into two fundamental types: those in which the Bourdon responds to volume changes and those which respond to pressure changes.

The systems

that respond to volume changes are completely filled with a liquid. The liquid in the bulb expands with temperature

to a greater degree than does the

bulb metal, thereby producing a net volume change which is communicated to the Bourdon. An internal system pressure change is always associated with the Bourdon volume change, but this effect

is not

of primary importance.

FIG. 6.2

6 The system that responds to pressure

changes

FULLY COMPENSATED LIQUID, MERCURY OR GAS FILLED THERMAL SYSTEMCLASS IA, IIIA, OR VA

is either filled with a gas, or is partially filled with a volatile liquid. Changes in gas or vapor pressure with changes in bulb temperature are communicated

to the Bourdon. The Bourdon will

crease in volume with increase this effect

in-

in pressure,

but

is not of primary importance.

7 Based on these two fundamental principles of operation, filled system thermometers have been classified [2] as follows: Volumetric

Principle:

Class I, Liquid-Filled Class

System

V, Mercury-Filled

Pressure Principle: Class II, Vapor-Filled Class III, Gas Filled

System

COMPENSATED CAPILLARY, ALSO

WITH

System System

Subclossificotion Liquid-Filled

Thermol

System

(Class

I):

A

The system is usually compensated temperature effects either:

FIG. 6.3

CASE COMPENSATED LIQUID, MERCURY OR GAS FILLED THERMAL SYSTEMCLASS

IB, 1118, OR VB

for ambient liquid and operating

on the principle

of vapor pres-

sure. Four types are employed: (a)

With full compensation pensating

(Class

IA),

tem minus the bulb, or equivalent compensation. (6)

9

See Fig.

With compensating (Class

Vapor

IB).

the com(a)

means being a second thermal sys-

means within

See Fig.

Pressure

filled

System

Designed to operate with the measured temperature above the temperature of the rest of the thermal system (Class

(b)

the case only

6.3.

Thermol

thermal system partially

means of

6.2.

(Class

II):

(c)

A

Designed

See Fig. 6.4.

to operate with the measured temper-

ature below the temperature

of the rest of the

thermal system (Class

See Fig.

Designed

IIB).

6.5.

to operate with the measured temper-

56


IIA).

ature above and below the temperature

with a volatile

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

8

thermal system completely filled with a liquid (other than metals such as mercury) and operating on the principle of liquid expansion.


of the

ANDAPPARATUS INSTRUMENTS SHOWING POSITION OF W)LATILE LIQUID WHEN BULB TEMPERATURE IS nJi THAN TEMPERATURE OF REST OF SYSTEM

y

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

PWWINC PDSITION >F VOLATILE LIQUID NHEN BULB TEMPERATURE

VOLATILE

FIG. 6.6 FIG. 6.4 CLASS

VAPOR

PRESSURE

THERMAL

VAPOR

PRESSURE

THERMAL

SYSTEM-CLASS

IIC

SYSTEM-

IIA

VAPOR \

VOLATILE

LIQUID>

NONVOLATILE VOLATILE FIG. 6.5 CLASS

VAPOR

PRESSURE

LIQUID THERMAL

FIG. 6.7

SYSTEM-

VAPOR

IIB

PRESSURE

THERMAL



SYSTEM-CLASS

IID

TABLE

Low temp limit High temp limit Longest span Shortest span Bulb size-Long span -Short span Dial or chart divisions

Equal

6.1

COMPARISON

-3OO’F 600°F 600 deg F 25 deg F smallest intermediate Equal

OF THERMAL

SYSTEMS

-4 60 30

-38OF.

-65OF Hg-Th eutectic 1200OF 600OF 1000 deg F 600 deg F

Equal

40 deg F intermediate large Equal

Larger

interm interm at rang

Maximum standard capillary length (Approximate) Capillary temperature compensation

200 ft

15 ft

200 ft

25 ft

with separate 200 ft

Dual capillary and Bourdons

None

Yone

None necessary

Case

Second Bourdon

Bimetal

Compensated capillary or dual capillary and Bourdons Second Rourdon

Generally small 100% of range

Yegligible 100% of range

Frequently larg Generally sma

Intermediate slowest Negligible

in water in air Negligible

Fastest

to inte

Usually

small

temperature

compensation

Bulb elevation error Over-range capacity

Speed of response (see Fig. 81 Barometric errors

strip

Negligible Negligible 100% of range Varies with length 200% to 0% range Slowest in water intermediate in air Negligible Negligible

---_ Note: Dimensions see Table 6.2

Rimetal

-___are to be used as a guide only and will

vary in OD and length with manufacturer.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



strip

None necessary

___-_____--_._-_ Functional

values

also

AND APPARATUS

rest of the thermal system (Class IX). This type normally requires a larger sensitive portion than Class HA or IIB. See Fig. 6.6. (d)

Designed

above, below and at the temperature of the rest

and capillary

of the thermal system (Class

determining

the sensitive

See Fig. 6.7.

liquid is confined to

10 Gas-Filled

Thermal

mal system filled

cury filled

III):

bient temperature

effects

either:

BIB).

means within

systems (Class of

proportional

to tempera-

larger temperature

spans will

require smaller bulbs. Since the temperature span of a liquid-filled system (Class I) or a mercury-

With a second thermal system minus the bulb,

(Class

in mer-

in the bulb, have a bulb internal

ture span. Therefore,

or equivalent means of compensation 111.4). See Fig. 6.2. With compensating

and mercury-filled

volume which is inversely

for am-

vari-

manu-

V), which operate on the principle

liquid expansion

of pressure change with temperature compensated

among the various

thermometer bulbs are shown in Fig. 6.8.

I and Class

with a gas and operating on the

change. The system is usually

length. Table 6.2 provides a guide in the size of the bulb. Considerable

13 Liquid-filled

A ther-

ther-

100 to 1) span

facturers. The basic reasons for the variations bulb size are briefly described below. Typical

device.

System (Class

The bulb size of the various

ation in bulb size exists

liquid is used to transmit the vapor

pressure to the expansible

principle

IID).

portion and a second relatively

non-volatile

(b)

12 Bulb Size.

mal systems varies greatly (approximately depending upon system class, temperature

to operate with the bulb temperature

In this type the volatile

(a)

Description

(Class

filled

system (Class

V) may vary by 25 to 1, the

bulb size will vary accordingly.

A few manufacturers

have designed mercury thermometers in the 400 to 12CKl°F range with 3/8 in. bulb diameters and 3 in.

the case only

See Fig. 6.3.

bulb lengths. 11 Mercury-Filled

Thermal System (Class

thermal system completely mercury-thallium principle

eutectic

pensating

A

14 The bulb size for all types of vapor systems (Class

on the

reasons.

The system is usual-

for ambient temperature

With full compensation

V):

with mercury or

amalgam operating

of liquid expansion.

ly compensated either:

(a)

filled

(Class

VA),

With compensating (Class

means within

but for different the system to which

interface of the liquid and vapor. The interface must always be located in the bulb. The fill will

the com-

always be in a liquid state at the coolest parts of the system. The system must be filled so that the liquid in the bulb will not completely vaporize nor

means of

compensation. See Fig. 6.2. The note of Fig. 6.3 shows an equivalent means of compensation.

(6)

The pressure within

the Bourdon responds is the vapor pressure at the

effects

means being a second thermal sys-

tem minus the bulb, or equivalent

II) also varies greatly,

completely

fill the bulb, under any conditions

of

bulb or ambient temperature.

the case only

15 The Class IIB system requires

VB). See Fig. 6.3.

that liquid

exist only in the bulb. Since the vapor density in the capillary and Bourdon is affected only slightly

In the equivalent means of compensation, an invar wire is drawn through the capillary. Thus, the volume of mercury in the annular space between the

by ambient temperatures, the bulb may be very small, as indicated by the dimensions in Table 6.2.

wire and the inner wall of the capillary

Whereas the bulb may be as small as illustrated,

is very

is frequently supplied in a large size, reduce the number of bulb sizes.

small compared with the mercury volume in the bulb. This reduces ambient temperature considerably

since volumetric

bient temperature

effects

changes due to am-

changes are reduced. A bimetallic

element is used to compensate

16 The Class IIA system bulb must be somewhat larger in order to accommodate the liquid expan-

for mercury volume

changes in the Bourdon which are caused by changes in case ambient temperature, and for any change in the modulus of elasticity tube with temperature.

sion within the capillary and Bourdon resulting from ambient temperature variations. The standard bulbs of a particular manufacturer may be larger or

of the Bourdon

smaller than specified

in Table


it

in order to


6.2.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

INSTRUMENTS

ASME PERFORMANCE TABLE

6.2

APPROXIMATE DIMENSIONS

BULB

SENSITIVE

TEST

CODES

17 The Class DC system bulb needs to accommodate the entire capillary

and Bourdon volume when

the bulb temperature

becomes equal to the tempera-

ture of the capillary

and Bourdon (See Fig. 6.6.)

The bulb size therefore is generally larger than that of the Class IIA system; it is also dependent upon --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

IA & B

Liquid

9/16

3

50 deg F span

IA & B

Liquid

3/8

2-l/2

275 deg F or

IIA IIB IIC IID

Vapor Vapor Vapor Vapor

9/16 3/8 9/16 9/16

4 2 6 4

IIIA & B

Gas

7/8

VA&

Mercury

9/16

greater

VA&

B B

Mercury

11/16

10

the capillary

span

length.

18 The Class IID system bulb must have an internal trap of such dimensions that the volatile liquid will not enter the capillary under all values of ambient temperature (i.e., the trap must accommodate the non-volatile

Based on 75 ft. capillary

2-l/2

500 deg F or

4

greater span 100 deg F span

liquid expansion

values of ambient temperature). 19 The gas-filled

See Fig.

system (Class

bulb sizes

on the capillary.

for various temperature

spans and capil-

FLANGE

-A-

EXTENSION

_/

BULB

.A

UNION CONNECTION

--------I b. FLANGED

BULB

WELL KEY: MEDIUM

ABCD-

BULB LENGTH SENSITIVE PORTION INSERTION LENGTH IMMERSION LENGTH

c. THREADED WELL ( UNION BULB 1 FIG. 6.8

TYPICAL

MERCURY FILLED


This

lary lengths are specified in Table 6.2. A further restriction based on the bulb temperature may be

r

a. PLAIN

re-

errors

error is also increased as the span is reduced and as the bulb temperature is raised. Approximate

Note 1: Dimensions vary between manufacturers and can often be reduced below those shown to meet particular requirements. Note 2: When faster response is desired and space and strength requirements permit, a longer sensitive of a smaller diameter is generally available.

BULB EXTENSION

6.7.

III) generally

quires a large bulb in order to minimize caused by ambient variations

under all


INSTRUMENTS

AND APPARATUS

obtained from the manufacturer. Several manufacturers have been able to design very small bulb

22 Liquid systems (Class IB) generally are provided with overrange protection of 100 percent of

gas thermometers.

temperature span. Some thermometers are provided with greater overrange protection, depending on the

3- to 3%in.

Bulb size in these designs

long and 3/8 in. in diameter.

ture ranges vary from -40

is

Tempera-

to 180°F to 400 to

manufacturer.

12ooT.

23

In vapor pressure thermal systems

overrange protection 20 Overrange Protection. Overrange protection is defined as the maximum temperature to which the bulb of a filled

in other systems because

It is usually

pressed in per cent of temperature

in Fig.

span above the

21

The overrange

protection

in Table

6.1.

of liquid systems

span for short systems.

For long sys-

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

range temperature

change neutralizes

tionship

for systems 200 ft long.

more reading sensitivity

VAPOR

24 The nonlinear

for

vapor pressure-temperature

is an advantage

PRESSURE-TEMPERATURE

rela-

where the user desires toward the top of range.

CURVES


are shown

can be increased.

the overrange possibilities of the Bourdon, thus reducing the overrange protection to essentially zero

FIG. 6.9

relationships

overrange temperature

exhausted from a bulb at a bulb temperature above the instrument range, in which case the safe over-

tems, because the capillary volume generally approaches the bulb volume, the capillary volume change with ambient temperature

rate of

rise. Typical

limit of range is near the aritical point of the fluid fill, the overrange protection may be extended because of the fill being a vapor above the critical point. Under some limited conditions, it is possible to fill the system so that all of the liquid will be

(Class IA) varies with capillary length. Generally, it is in the region of 100 to 200 percent of the temperature

6.9. A specific

II), than

each range offered is usually specified by the manufacturer. This is generally appreciably less than 100 percent of temperature span. If the upper

upper limit of the range. A summary of extent of overrange protection for filled system thermometers of the various types is specified

of the increasing

vapor pressure temperature

ex-

(Class

more limited

vapor pressure rise with temperature

system may be exposed indefinitely

without damage to the system.

is generally


ASME PERFORMANCE

varies

Since the protection

considerably

pressures

29 Well Materials.

offered

materials

for various system ranges and

various manufacturers, should be obtained

the overrange protection

Bourdons.

30 l/16

tempera-

the temperature

dicating

instruments.

ally used in industrial other devices frequently

Although filled

pointer are eliminated.

and diaphragms

Pointer

mentioned

small

conventional

designs,

Thus, any failure

the pointer.

31

eutectic

since

special

facturers

supply

by

The upper temperby

limited to in vapor pres-

is operated up to ISOOOF.

and Types

materials. as

low

Materials

as -430cF.

by the filling

are available

for temperatures

The maximum temperature

limited by chemical instability to approximately 600%‘.

Some manu-

of the Vapor Pres-

II) is limited

is

of organic liquids

35 The minimum temperature of the Gas System (Class III) must be above the critical temperature

finned


by

used to lower the

system is not limited

34 The minimum temperature

Monel, Inconel,

bulbs made of externally

device

sure System (Class

Copper and

are available.

I sys-

sure above this temperature. However, some manufacturers supply systems to IZOO’F, and one

SAE alloy steels,

such as nickel,

in Class

I) is limited

chemical instability, but is usually 1000eF b ecause of a rapid increase

bronze are vulnerable to mercury attack and accordingly are not available with Class V systems. Op and silver

to -650F.

ature of a mercury-filled

Bulb Materiols. Among standard bulb materials as listed in most manufacturers’ catalogs, are

tional materials,

employed

amalgam is frequently

minimum temperature

28

Hastelloy,

by the freezing

33 The mercury system (Class V) is limited its freezing point to -3S°F. A mercury-thalium

bulb. In

mechanism

steel.

I and V) is limited

600%‘.

of Construction

304, 347 and 316 stainless

The and mercury-filled

the upper temperature at which the organic liquid remains chemically stable, which is approximately

it.

bronze, copper, steel,

6.1)

Temperatures.

of liquid

the organic liquid system (Class

case can be

to be replaced

(Class

(SEE TABLE

and Minimum

32 The organic liquids

the bourdon is in the case.

of the indicating

Materials

in. OD are frequently

tems freeze between -100 and -3OOoF, depending upon the liquid used. The maximum temperature of

the thermal system cannot be severed without destroying

of 3/16

(Types in. OD

point of the fluid fill.

This design

the thermometer

Maximum

systems

and indicators

Thus, a faulty

causes the entire instrument

armor of bronze,

or by an armor covered

armor protection.

minimum ,temperature

drive is either direct

is unique in that thermal elements without removing

steel,

is com-

(approximately

by a flexible

stainless

CHARACTERISTICS

are

In this design, a bulb, capillary and piston-cylinder head make up the thermal system. Piston motion in the cylinder is transmitted directly to the linkage

can be interchanged.

in.), protected

employed without

in in-

or accomplished through magnetic coupling. One mercury design completely eliminates the Bourdon.

replaced

The capillary

Materials.

and copper capillaries

signal.

Bourdon motion to the

in the case which rotates

are also

copper or stainless steel. Stainless steel 304 or 316) and Inconel capillaries of l/8

links and gears used in

to transmit

and silver

with a plastic (such as polyethylene) for corrosion resistance. Capillary materials normally consist of

the Bourdon is in the form of a

coil. The conventional

Optional

Bourdons are gener-

used. In the previously

most designs

Hastelloy,

steel.

cast iron, nickel,

system thermometers,

such as bellows

bulb gas designs,

Capillary

plated steel,

amplified by a mechanical linkage or gear system to drive a pointer for temperature indication or to A Bourdon may be used without amplification

to brass, steel

monly of a small outside diameter

The Bourdon motion is normally

drive a pen for recording

such as aluminum,

Monel, Inconel,

ture. 27

confined

available.

ture span overrange protection; with some manufactures exceeding this limit. A top limit of 1000%’ for overrange

to tem-

Standard bushing and well

are generally

material,

from the manufacturer.

specified

the response

and Types 304 and 316 stainless

26 Mercury systems (Class V) are normally provided with a minimum of 100 percent of tempera-

is also generally

to increase

perature changes.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

of these systems.

CODES

tubing designed

25 The overrange protection of a gas system (Class III) will be reduced for short range temperature spans because of the higher internal

TEST


AND APPARATUS

of the gas employed, which for commonly used Th e upper temperature is usually helium is -451°F. limited

to lOOO?‘,

but gas systems have been made

to operate successfully 36

forms Bourdon motion into pointer or pen motion. The backlash in mechanical gearing and links usually is greater than 0.1 percent; thus, the sensitivity

up to 15OOoF.

The minimum range of the organic

Range.

of expansivity

and compressibility

of the

tivity

organic liquids employed, the maximum range is frequently limited to 200 to 400 deg F’, because of differences

in manufacture

so that a specified

41

The range of a vapor system (Class

Par. 34. However,

the nonlinear

ranges and the range is therefore to approximately 250 deg F.

Filled system thermometers are regarded as 1.0 percent instruments. This

Accuracy.

conditions

of

instruments

are calibrated

to higher accuracy

and

in indoor applications the maximum error is frequently specified as 0.5 percent of temperature

by greater

normally

are

case or capillary ambients the error will not exceed 1 percent of temperature span. However, many

II) is

vapor pressure-

is accentuated

linkages

is no better than above.

means that under most environmental

limited by low and high temperatures of -430 and 600°F, respectively, for the reasons discussed in temperature relationship

in these designs

normally

ac-

curacy may be met with linear dials or charts. 37

where mechanical

reduced or eliminated, sensitivity may be affected by highly viscous oil which is put on the Bourdon coil to damp shock and vibration effects. Sensi-

liquid system (Class I) is limited by maximum bulb size to approximately 25 deg F. Because of nonlinearity

may be on the order of 0.25 percent of range

span. In the designs

span. Accuracy

limited

thermometers

may be only 2 or 3 percent

are used in environments

case and capillary

38 The minimum range of a mercury system

temperature

from usual room temperature lary temperature

(Class V) is limited by maximum bulb size to approximately 50 deg F. For amercury system the bulb

when

where the

vary considerably

(e.g.,

case or capil-

can be as low as 60°F and as

high as 16OoF in some power plant applications).

size is larger for a particular comparative range than it is for an organic liquid system, because the expansion rate of mercury is less than that of or-

The reduction ability

of accuracy

is caused by the in-

of the compensation

devices

to completely

ganic liquid by a factor of approximately six. The maximum range is limited only by the upper useful

compensate for ambient temperature changes. Direct reading thermometer cases, which are attached directly to the bulb, are exposed to heat

temperature, generally lOOO’%‘, and the freezing point of -38oF. Mercury filled system thermometers

conducted along the thermometer

stem and also to

heat radiated

Capillaries

are made, however,

with ranges as low as -40

to

wound around boiler casings are also subjected

180 deg F and as high as 400 deg F to 1200 deg F.

high ambient temperatures. temperatures. mechanical friction.

absolute pressure is proportional to absolute temperature) it is characteristic of this system that the

maximum range is limited only by the upper temperature, usually lOOOoF, although longer ranges

perature.

generally require larger bulbs to provide an adequately linear output. Small bulb and large bulb gas thermometers capable of operating up to 1200°F designs to eliminate 48

Sensitivity.

The Bourdon of a filled

bulb temperature.

the output for small

temperature

Therefore,

changes is affected

loose fits in the mechanical

perature,

system errors will result bevariations

unless

means are employed.

and is usually

44 The liquid-,

or

(Class

apparatus which trans-

ignored.

gas- and mercury-filled

systems

I, III and V) are provided with full compen-


by

and fluid

The only temperature error observed in this system is of small magnitude; it is caused by change of elastic modulus of the Bourdon material with tem-

change in

only by friction

and mechanical

43 The vapor-filled system (Class II), as an exception, is not subject to errors from the fluid fill.

system

measurable

Therefore,

compensation

output effect.

will respond to the smallest

backlash

to very low ambient

is also affected

cause of ambient temperature

scale is used in some

the nonlinear

Accuracy

to

used

42 Temperature Compensation. Since the capillaries and Bourdons as well as the bulbs of thermal systems are filled with actuating fluid, these portions of the system are sensitive to ambient tem-

shorter the range the higher will be the internal system operating pressure. This condition limits the minimum range to approximately 120 deg F. The

A nonlinear

Thermometers

outdoors could be subjected

39 Because the pressure within a gas system (Class III) essentially follows Charles’ Law, (i.e.

are available.

by unlogged pipes.


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

INSTRUMENTS

ASME PERFORMANCE

TEST

CODES

sation (the capillary and Bourdon compensated) by means of an auxiliary system less the bulb (see

is in the range of 0.003

Fig. 6.2).

The length of these systems is normally limited

The capillary

the auxiliary

and Bourdon volumes of

system are made essentially

20 ft. Likewise,

equal to

fied by SAMA standards as IA, IIIA, tems, respectively error tolerance

are classi-

fore recommended that this information

by

48

the manufacturer and is usually equal to or less than +l percent of range for an ambient tempera-

cussed previously,

sponse time for the bulbs of the various types of thermal systems in water, with a velocity of 2.5 ft

as dis-

per set,

of these systems

45

III)

velocities

change of +5O deg

by the curves of Fig. 6.10.

for typical

bulb sizes is given by the

nomograph of Fig. 6.11. See Chapter 1 for installation procedures to obtain optimum response.

length of 100 ft or less.

Because the capillary

(Class

is approximated

The 63 percent response time in air at various

should be equal to or less than 1 percent of range F and for a capillary

The response of a thermal system

determined by the response of the bulb

because the lag in the capillary is generally equal to or less than one second, The 63 percent re-

are also used and are probably

span for an ambient temperature

Response.

is usually

ture change of +50 deg F and for a capillary length of 100 ft (i.e., 0.0002 percent of range per ft per

more common. The error tolerance

49 A bulb will respond faster if the following three fundamental design factors are employed:

error of a gas system

is reduced as the bulb size is increased,

the Class IIIA system which has full compensation is rarely built. Gas-filled systems are therefore

(a)

Increase the external area relative ternal volume.

generally limited to the Class IIIB the case is compensated.

(b) (c)

Lower the heat capacity. Increase the thermal conductivity

type, where only

walls 46

and internal

to the in-

of the bulb

fill.

The mercury system with full compensation

(Class

VA) is frequently

capillary

The gas system (Class III)

supplied with a single

which is continuously

temperature

favorable

com-

with a relatively

This is achieved by employing a capillary with a precision bore enclosing a precision drawn Invar

the internal

However,

of the

tional

larger bulb gas thermometers or mercury

filled

thermometers.

gas, and mercury systems with

case compensation

only (Classes

IB, IIIB

and VB)

50

are frequently employed because of the simplicity of construction. The capillary bore size is reduced to a point where system response is not seriously

the heat capacity

range is sufficiently

system (Class

of the volatile

and IIC)

fluids employed is

somewhat slower response because of the presence of the internal bulb trap and in some cases, also because of the increased viscosity of the nonvolatile liquid fill.

the IB)

64 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


IIB,

response because

evaporation of the fill will take place on the internal bulb walls. The (Class IID) vapor system has

small that

error may be ignored. In practice,

error of a liquid-filled

favorable

low and the thermal conductivity high. This is particularly true for small temperature changes because under these conditions condensation and

6.3). These systems are employed when the capillary length may be sufficiently short or when the ambient temperature

The vapor systems (Class IIA,

have almost equally

affected in order to minimize the capillary temperature error. The Bourdon of these systems is compensated by means of a bimetallic strip (see Fig.

capillary

thin wall and the heat capacity

gas is almost negligible.

eters are somewhat slower in response than conven-

capillary.

the capillary

the most be made

large bulb size frequently required tends to offset this natural advantage. The small bulb gas thermom-

wire so that the expansion of the Invar wire and mercury equals the expansion of the surrounding

The liquid,

is frequently

because the bulb can usually

pensated along its entire length (see Fig. 6.3).

47

be obtained

from the manufacturer.

(see Pars. 8, 10 and 11). The

deg F). Other methods of compensation,

to

error of a mercury-

length is normally limited to 50 ft. The capillary temperature error of the gas system (Class IIIB) varies with length, range and bulb size. It is there-

and VA sys-

of these systems is specified

the capillary

filled system (Class VB) is in the range of 0.0008 to 0.0016 percent range per ft per deg F, and the

the corresponding volumes of the primary system. This arrangement permits the erroneous response to be opposed by an equal erroneous response, thus providing full temperature compensation. These systems with full compensation

to 0.005 percent range per

ft per deg F, depending upon the manufacturer.


INSTRUMENTS

AND APPARATUS

51 The liquid and mercury systems (Classes I and V) have the slowest response because of the increased mass and poorer conductivity of the fluid fill. Whereas the liquid system (Class I) is slower than the mercury system (Class

ACCESSORIES

53 The Bourdon motion of filled system thermometers is usually amplified by a simple linkage as shown in Fig. 6.1, in order to drive the pointer of an indicator or the pen of a recorder. In dial gages,

V) for any specific

bulb diameter, the fact that the bulb size of the former will

be smaller

for a particular

greater angular motion of the pointer,

range, fre-

deg angular displacement,

quently more than offsets this disadvantage. Some manufacturers incorporated fin-like copper disks on liquid-filled bulbs to enhance heat transfer. 52 l/16

is achieved

usually

270

by a “move-

ment.”

The most common movement employs a geared sector to drive a pinion to achieve angular

amplification.

Capillary bulbs with outside diameters of to 3/16 in. are sometimes employed to pro-

54 In temperature

transmitters

the temperature

vide rapid response. Response is nearly directly proportional to bulb outside diameter. These bulbs

signal

are frequently

recorder or other readout device. The transmitters provide the means of transmitting temperature in-

capillary

long capillary

bulb form is shown in Fig. 6.12.

A

formation

bulb (up to 200 ft in length) is some-

times left uncoiled

to measure average temperature

along its installed

length.

tems may be employed

Class

I, III,

pneumatic

and V sys-

FIG. 6.10

BULB

BULB

RESPONSE

(Velocity

in turn is communicated

over long distances. transmitter

In the case of the

the Bourdon is usually

DIAMETER

(: IN INCHES )

VERSUS

BULB

OD IN WATER

of 2.5 fps)

65 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


to a

re-

placed by a diaphragm which will exert a force responsive to bulb temperature. This force in turn is balanced by a feedback force of a pneumatic

in this manner.

OUtSlDE

to a pneumatic or electrical

signal and this signal

coiled by the manufacturer (preformed bulb) p roviding compactness for installa-

tion. A typical

is converted


ASME PERFORMANCE servo, the feedback transmitter transmitter

force being generated

pressure. the filled

by the

Similarly, in an electrical system force is balanced

TEST

CODES

up or down, which may be corrected by a simple screw adjustment. However, a severe shock could

by

cause a permanent set in the Bourdon or misalign-

a force which is generated by an electrical current. In other electrical transmitters the Rourdon motion

ment of the linkage.

directly

the entire range.

operates

the core of a differential

calibration

trans-

former for a-c output or the force from a Hourdon actuates

would not necessarily

56 Conduction

a strain gage for d-c output.

a filled

In this case, the change in

and Immersion

system thermometer

be uniform over

Error.

The

bulb of

must be completely

mersed in the medium in which the temperature APPLICATION

being measured.

AND INSTALLATION

If this is not done, a significant

portion of the filling Sources 55

Zero

Shift

Filled

Error.

tremely large. Actually,

system thermometers

which may cause an error in the calibration. user, therefore, should check the instrument A calibration

the bulb should be im-

mersed so that not only the filling

abuse during shipment,

bration and make corrections.

medium volume can be at a dif-

ferent temperature than that which is being measured. Errors due to improper immersion can be ex-

of Error

are subject to mechanical

is immersed but also a sufficient

The cali-

bulb extension

to prevent heat conduction

IO’8 6

4

IO= 8 6 4 *AM M-40

AT

S-0

AT

AThi

7O’F

ro-

FIG. 6.11

BULB

RESPONSE

RATE

IN AIR AT VARIOUS

VELOCITIES

66 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


fluid reservoir amount of the to or

from the sensitive portion. The amount of extra immersion varies as the heat transfer and temperature environment varies. For a thermometer with a 3-in.

error

associated with shipment is usually confined to a “zero” shift, in which the entire range is shifted

2

imis


INSTRUMENTS sensitive

AND APPARATUS

portion length and being used to measure

temperature

as high as lOOO”F, for example,

user to specify this elevation

the

manufacturer

or depression

so that he will calibrate

to the

the instru-

bulb should be immersed about 5 in. Heat transfer

ment accordingly.

from unimmersed portions of the thermometer should

will have liquid within the capillary

only over part

be reduced when measurements

of the range span and is, therefore,

not recommend-

are being made in a

medium having low heat transfer capabilities. The entire sensitive portion should be immersed in the

57

Capillary

Immersion

Error.

The capillary

temperature sensitive.

Dual capillary

ferent elevations. the capillary

conditions filled

for the liquid-filled

system

system (Class

tive to barometric pressure changes by the ratio of barometric pressure change to the internal pressure change corresponding to the range. These

If the immersion

systems therefore are designed to have a minimum pressure change of 100 psi for the range of the

Small bulb liquid

thermometer. Since the maximum barometric sure change is approximately

e.g., on a mer-

pres-

20.4 psi, this error

will be equal to or less than 0.4 percent of range.

cury filled thermometer with a 8-in. by 8/8-in. bulb and a range of 400 to 12oOoF, no more than 2 in. of capillary 58

Bulb

should be immersed. Elevation

Error.

When the Bourdon ele-

vation of a Hquid or mercury system (Class is changed relative

I or V)

to the bulb, a pressure head

caused by the column of the fluid fill is generated within the system. This pressure redistribution causes a small volume change of the fluid and of the bulb and capillary thereby causing a system error. If the bulb is to be elevated more than 25 ft above the case, it is desirable for the manufacturer to know the above elevation to increase the pressure of the system so that the bulb pressure will not drop to zero after installation. 59 The elevation system (Class 60

error is nonexistent

in a gas

III).

If the Bourdon is above the bulb in a vapor

system (Class

IIA or Class IID)

the pressure with-

in the Bourdon equals the vapor pressure in the bulb minus the liquid pressure head in the capillary. This means that the bulb elevation error is equal to the ratio of the liquid head to the internal vapor pressure change across the temperature span. This further confirms that it is advantageous the manufacturer tively relative

for

to provide systems having a rela-

large internal

pressure.

If the bulb elevation

to the case is 20 ft, it is advisable

FIG. 6.12

for the

PREFORMED

CAPILLARY

6’7 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


V).

Vapor systems (Class II) and gas systems (Class III) operating on the pressure principles are sensi-

thermometers are more affected by capillary

immersion than larger bulb designs,

i.e.,

(Class I) and the mercury-filled

by the user under the

of the application.

case temperature.

metric principle,

systems fre-

length is greater than 8 in., the immersion length should be specified to the manufacturer or the instrument should be adjusted

dif-

cannot read cor-

This error is essentially Error. 61 Barometric non-existent for systems operating on the volu-

of II) is

(Class VA). These compensating means are imperfect and the instrument output reading will vary immersion.

The instrument

rectly for bulb temperatures both above and below

quently are used in liquid systems (Class IA) and compensated capillary is used in mercury systems

with length of capillary

IIC)

ed when the bulb and case are at appreciably

flowing fluid when thermometers are used in forced convection applications.

all system types except vapor systems (Class

The vapor system (Class


BULB

ASMEPERFORMANCETESTCODES Essential 62

65 A relatively inexpensive mounting into equipment where pressure tightness is not important is

Considerations

A thermal system is generally

installed

in a

that illustrated by Form 3. The flange is generally split so that it may be attached to or removed from

vessel by means of a union connection, a flange, or combination of union connection and flange.

the completely

See Fig. 6.13. 63

Form 1 is employed when it is desired to at-

fabricated

thermal system.

66 The pressure rating of the fittings is specified by the manufacturer and is generally 100 psig.

tach to the equipment by means of a bushing or flange (in which case the bulb is exposed) or when

If higher ratings are necessary

it is desired that the bulb be protected by insertion into a well. If it is not necessary to have the union

must be supplied. The well ratings follow no established code. The manufacturer’s well ratings will

connection adjustable

vary from 1000 to 5000 psi, depending upon material and design.

along the extension,

union is attached to the extension

the

by soldering,

brazing or welding to provide a pressure-tight

joint.

to have the union ADVANTAGES

adjustable along the extension, particularly when mounting in a bushing to provide a pressure tight seal, the union connection is provided with an additional 64

67

pressure seal as shown in Fig. 6.13.

Extension

bulb sensitive

(a)

(b)

portion and the union in Form 2 or

ATTACHMENT VESSELS.

Advantages

System construction Low initial

(c) Instrument

between the external threads of the bushing or well and the hex nut. The external threads of bushings and wells have been standardized as % NPT, ?4 NPT and 1 NPT.

FIG. 6.13

AND DISADVANTAGES

keep is generally

stems may be provided between the

is rugged. Amount of up

minor.

cost. can be located up to 250 ft from

point of measurement. (d)

Instrument needs no auxiliary less an electric

OF THERMAL

SYSTEMS

power supply un-

chart drive is employed.

TO


fittings


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

However, when it is necessary

special

INSTRUMENTS 68 (a)

AND APPARATUS in brackets,

Disadvantages

[II ASME

Paper 53-IRD-1, “Bibliography of Bourdon Tubes and Bourdon-Tube Gages,” which gives reference to 142 papers on this subject. [zI Scientific Apparatus Makers of America (SAMA) Standard PMC 6-lO-1%3. An additional suggested reference is: “Process Instruments and Control Handbook,” Douglas M. Considine, Section 2, McGraw-Hill Book CO. N.Y. 1957.

Bulb size may be too large for some applications.

(b) Minimum ranges.are (c)

limited.

Maximum temperature

is limited.

REFERENCES

69

In the text Reference

thus [l].

numbers are enclosed

69 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



CHAPTER 7, OPTICAL PYROMETERS Definitions

CONTENTS

2 An Optical

Par. GENERAL: Scope ............................................................................ 1 .................................................................. 2 Definitions .................................. 8 PRINCIPLES OF OPERATION CLASSIFICATION: ................................................................ 12 Description 22 Materials of Construction .......................................... CHARACTERISTICS: 23 Range ............................................................................ Precision ......................................................................25 Accuracy ......................................................................27 31 .............................................................. ACCESSORIES: A;FnW;;4TION AND INSTALLATION: 32 .......... .......... .................................................... 47 Sources of Error ........................................................ ........................................ 63 Essential Considerations 65 Treatment of Data ...................................................... 66 ADVANTAGES AND DISADVANTAGES .................... ..............................................................68 REFERENCES

and usually

radiance.

will

guide

Optical

(images)

pyrometers

are distinguished

instruments

in that two sources

of equal radiance are compared.

3 Radiance

is the amount of energy

tion, per unit projected radiance is radiance at a particular tral radiance

per

radiating

unit time, per unit solid angle in a particular

direc-

area of a source. Spectral

per unit waveleneh

wavelength;

interval

total radiance

is spec-

summed over all wavelengths.

4 A blackbody

is one that absorbs all radiation

incident upon it, reflecting or transmitting none; the spectral radiance of a blackbody is a known function

is to present in-

of its absolute

temperature.

the user in the selection,

and use of optical

5

pyrometers.

Emissivity*

is the ratio of the radiance

body to that of a blackbody

of a

at the same tempera-

ture. Total emissivity refers to radiation of all wavelengths, and monochromatic or spectral emissivity refers to radiation of a particular wavelength. Total emissivity is the average value of spectral

*The nomenclature used here is consistent with that prescribed in the International Lighting Vocabulary. Some confusion in terminology exists in the literature, in which a number of authors use the term “emittance” to distinguish between the emissivity as defined above and the emissivity of an ideally flat surface of the same material. In that usage, emissivity is assumed to be an intrinsic property of the material, and is taken to be the limiting value of emittance as the effects of surface roughness are reduced to zero. The term “emittance,” as just defined, has not been accepted by standards-setting organizations primarily because of its similarity to the which is defined to be radiant term “radiant emittance,” power per unit area emitted from a surface. In the terminology used herein, no distinction is made between these two definitions of “emissivity” and “emittance”; values reported in the literature and defined in of “emittance” the manner described above are treated as being inter changeable with the values of emissivity using the definition in the above tabulation.

emissivity

weighted

distribution

6

The

with respect to the blackbody

and summed over the entire spectrum.

spectral

radiance

temperature

of a source

is the temperature of a blackbody having the same spectral radiance as the source, wavelength(approximately

at a specified

0.65 pm, in the case of

optical pyrometry). It has also been commonly known in the past as the brightness temperature, or (currently)

as the luminance temperature when

the eye is used as the sensor.

70


The spectral

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

which

installation,

of a telescope,

an absorption glass filter.

from other similar

Scope

formation

consists

radiance of a body whose temperature to be measured is compared to that of a standard source of

GENERAL

1 The purpose of this chapter

Pyrometer

a calibrated lamp, a filter to provide for viewing nearly monochromatic radiation, a readout device,


INSTRUMENTS AND APPARATUS 7 The target is the source of radiation whose temperature

Planck’s

as seen by the op-

is to be measured,

tical pyrometer. The calibration

of an optical

effect of the emissivity target temperature.

pyrometer is based on

e

OF OPERATION

Radiation.

The operation

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

and spectral relation

distribution

of an optical

c, 4T

-1

c,

= a constant in the Planck radiation

c,

=: 0.014388

h

= wavelength

T

= absolute temperature,

e

= base of the natural or Napierian

The function

the intensity

temperature

(1)

Nbh = spectral radiance of a blackbody at wavelength law

m-K of radiant energy, in meters

of the optical

in kelvins logarithms

pyrometer is to determine

the ordinate NbA of the Pl anck radiation

of which bear a definite

to the absolute

A’=

.I

pyrometer depends upon the phenomenon that a body (most noticeably at elevated temperature) at all wavelengths,

Law

where the index of refraction of the surrounding medium is assumed to be unity, and

of the target to obtain the

PRINCIPLES

emits radiation

c,

Nbh =

the radiance of a blackbody at the temperature at which pure gold freezes (or melts), and its calibration at other temperatures is in terms of blackbody radiation; the temperatures it indicates, called “radiance temperatures,” must be corrected for the

8 Thermal

Radiation

distribution;

at a (nearly) constant wavelength, NbA becomes a measure of T. The Planck radiation distribution is illustrated in Fig. 7.1.

of the body.

The temperature of a body may be determined from a measurement of its radiance. This measurement may involve the total radiance or the spectral radiance; in the case of optical pyrometers, radiance is

10 A blackbody

is experimentally

realized

by

uniformly heating a hollow enclosure and observing the radiation from a small opening in the wall of

measured in the visible portion of the spectrum, conventionally at a (red) wavelength of about 0.65

the enclosure.

p,

depends not

ted from this opening depends almost entirely

but also on

the temperature

of the walls,

on the material

of which the walls are constructed.

However,

in general,

only on the temperature

the radiance of the source,

the’particular material constituting the source, and on the character of its surface roughness. Thus, glowing carbon radiates

approximately

Design

three times

are at the same temperature.

This

expressed by the statement or emissivity

11

is technically

that the emissive

of carbon is approximately

that of platinum in the neighborhood

considerations

Blackbodies”

as much power per unit area in the visible red portion of the spectrum as Rlowing platinum when both

Blackbody

in effect,

Radiation.

of 0.65 p.

Kirchhoff’s

that the emissivity

law is an approximation

law; it is mathematically

function

temperatures

is numer-

below 6000%.

error is detectable, temperature.

equilibrium with its surroundings. A perfect absorber absorbs all radiation incident upon it, reflecting nothing; such a surface is said to be black.

-&/AT h’bh = C, )I-’ e

12

ator, or simply as a blackbody. By definition, the emissivity of a blackbody is unity. For the unique the spectral

ture is known exactly,

distribu-

of its tempera-

and is described

and rises rapidly

above that

(2)

Radiance

Comparison-lamp

(Disappearing

Variable

Filament

Optical

The essential

elements of the instrument

*Numbers in brackets

chapter, thus

as follows:

designate

[1I.


at

At about 7OOOT the

Description

a surface having the highest theoretically possible emissivity is therefore known as a blackbody radi-

radiator,

instead of

CLASSIFICATION

A perfect absorber must also be a perfect emitter;

case of a blackbody

and is ordinarily

at 0.65 pm is negligible

ically equal to its absorptivity, a condition that guarantees that the surface can exist in thermal

tion of radiant energy as a function

function

of the Planck much more con-

due to using the Wien function

the Planck

law states,

of a surface

on

and almost negligibly

sufficiently accurate to be used in calculations for calibration and application of optical pyrometers. The-error

9

emit-

are treated under “Practical

venient than the Planck

three times

of the radiation

(Par. 35).

Wien’s

radiation

power

The intensity


Pyromerer)

Type

[1,21*. are typi-

Reference at end of

A

ASME PERF’ORMANCE

TEST CODES filament

are magnified

for the observer by a micro-

scope lens and an ocular lens. The eyepiece is focused first to provide a sharp image of the standard lamp filament, and then the target image is focused by adjusting the objective lens. 13 The red filter

between the eyepiece

and lamp

serves to produce approximately monochromatic light to the viewer. In making an observation, the current through the lamp filament

is adjusted by a

rheostat until the image of reference portion of the filament (opposite an index if the filament is straight

or at the apex if the filament

is of the same luminance

'1632%

is U-shaped)

as the image of the target

viewed. The outline or detail of the reference section of the filament is indistinguishable from the surrounding field and “disappears” when the current in the lamp is properly adjusted. The value of the current in the lamp may be measured by means of a milliameter, the scale of which is ordinarily

RADIATION

-

MEASURED BY PYROMETER

I I I 4 5 2 3 WAVELENGTH, A o&ad

L I

FIG. 7.1

graduated in terms of temperature, or, alternatively, a potentiometric measurement of the current may be made; in models employing a built-in potentiometer, I

the potentiometer scale is graduated in terms of temperature, Standardized absorption glass filters

6

are interposed

PLANCK’S BLACKBODY RADIATION DISTRIBUTION FUNCTION, SHOWING SPECTRAL BAND UTILIZED BY AN AUTOMATIC OPTICAL PYROMETER AT 0.65 pm.

thus permitting

between the target and the lamp, a wide range of temperature

measured without requiring

high filament

to be

tempera-

tures. Optical pyrometers of this type are available covering the temperature range 1400 to 18,000°F;

[The spectral bandwidth for a disappearing filament optical pyrometer at the same wavelength

however, the majority of applications are below 45000F, with applications above 7000°F being rare.

is somewhat greater.] 14 Constont tally

arranged as illustrated

tive lens focuses

in Fig.

7.2. An objec-

The

a real image of the target in the

plane of a standard lamp filament.

Both image and

OBJECTIVE

essential

Radioncs

Comparison-lamp

elements of the instrument

trated in Fig.

7.3.

lamp filament

operates

Type.

are illus-

In this type of instrument, at a constant radiance

the ob

ABSDRPTION FILTER (USED FOR TEMP. ABOVE ISOO*C)

TARGET

OBJECTIVE APERTURE

APERTURE STOP

-III FIG. 7.2

SCHEMATIC

DIAGRAM

OF AN OPTICAL

CURRENT MEASURING INSTRUMENT

PYROMETER

72 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



111

OCULAR

INSTRUMENTS

AND APPARATUS

:AL WEDGE

OBJECTIVE LENS

,--

GROUND GLASS

SPOT

FILTER

_-_

AXIS OF ROTATIONOF OPTICALWEDGE CALIBRATED

SCALE /

FIXED INDEX TO INDICATE J TEMPERATURE ON THE SCALE FIG. 7.3

CONSTANT

RADIANCE

COMPARISON-LAMP

[The ground glass spot in the mirror M, can be illuminated by the pyrometer lamp, with the diffusely transmitted

radiation

visible

is imaged on the ground glass

spot, which appears spot in the field of

view. The optical transmittance tion.Thewedge the fraction

wedge is a gray filter

varies

as a function

can therefore

of radiation

The angular position

of the wedge may then be taken as a measure of the target radiance directly

whose

temperature,

which is indicated

on a scale attached to the wedge. The

lamp is operated at a constant preset radiance

of angular rota-

which may be adjusted

be rotated to adjust

transmitted

PYROMETER.

tions the spot will disappear.

to the eye. The target

to the viewer as an illuminated

OPTIC4L

get until the target image and ground glass spot image have the same radiance, under which condi-

milliameter

from the tar-

indicated

by a rheostat

filament.1

73 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


to a preset

current through the lamp


ASME PERFORMANCE

image

tained by setting the current through the lamp to a standard value by means of a rheostat and milliammeter. The radiance of the target is matched against that of the lamp filament by means of a polarizing prism, an iris diaphragm, or an absorbing device such as a gray glass circular wedge, interposed between the target and the lamp. A circular wedge is used to reduce the luminance

spot on a ground glass screen (illuminated lamp) in the field of view of the eyepiece.

of the target.

The temperature

glass

to the wedge. A red filter

in the eyepiece

the wavelengths

15

restricts

Self-Balancing

Lamp

Type

essential

of a

elements

used to a

of this type are available range from 1400 to 52000F.

Variable

(Automatic

in Fig. 7.4 and Fig.

by the The

is read direct-

ly from a scale attached narrow band. Pyrometers covering the temperature

of the light

emitted from the target to match the luminance

TEST CODES

Radiance

Optical

Comparison-

The

Pyrometer).

of the instrument 7.5. Operation

are illustrated of the instru-

ment [3I is similar in principle to that of the disappearing filament optical pyrometer; it differs in that

lamp current is adjusted to the fixed value specified for the particular lamp, and the instrument is focused on the target and the wedge is rotated

detection of radiation is accomplished with a photodetector (usually a photomultiplier tube) rather than

until the image of the spot disappears

by eye, the lamp current is adjusted

against the

r

by an electron-

PYROMETER LAMP

ROTATABLE POLARIZING FILTER

POLARIZING FILTER SMALL APERTURE IN FIRST SURFACE MIRROR OEFINES TARGET AREA

OCULAR

FIG. 7.4

SCHEMATIC RADIANCE

OPTICAL SYSTEM COMPARISON-LAMP

OF AUTOMATIC TYPE.

E OPTICAL

spectral

[Radiation from the target is focused on a small aperture in the mirror M. The portion that does not go through the aperture is reflected into the view-

PYROMETERS

bandwidth

tion arriving

just ahead of the detector, matched pair of such filters

of the modulator,

There

tween lenses absorbing

manner as radiation

the objective an absorbing

passing

between

through the mirror

aperture. The modulator allows the detector to receive radiation alternately from the pyrometer lamp and from the detector,

filter

F,limits

be-

Bl and Cl.

locations

glass (range changing)

for the

F,, one B and C, and the other between

filters

lens A and the aperture mirror M. If glass filter is not used between lens

located

the

between the lens E and the ocular,

permit variable

is to

dimming of the target image.1


a

may be located

A and aperture mirror M, a crossed polarizer

but not from both at

the same time. An interference

or alternatively,

B and C and between

lenses

of radia-

F, may be located

are also two alternative

C is focused on the detector. Radiation from the pyrometer lamp filament is treated in the same

-

VARIABLE

and peak wavelength

at the detector:

ing system. The portion that goes through the aperture is focused by lenses B and C at the plane after which the image of the lens

-


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

LENS

INSTRUMENTS ic null-balancing the spectral stantially

system rather than manually,

bandwidth

employed

AND APPARATUS

and

is usually

sub-

narrower than in the disappearing

fila-

integrated signal controls the lamp current, driving it up or down to achieve a zero amplitude square wave from the photomultiplier, at which time the

ment type. A further difference is that the pyrometer lamp is not mounted in the plane of the target image,

lamp current is said to be in null-balance. The pyrometer has the same range and is calibrated in essentially the same manner as the conventional disappearing filament optical pyrometer, i.e., the pyrometer lamp current is determined as a function of blackbody target temperature.

resulting in the use of two separate optical trains, one for pyrometer lamp radiation and one for target radiation;

the lamp filament

is therefore

not neces-

sarily at the same radiance as the source, although the radiant flux arriving at the detector through one optical train is equal to that arriving through the other optical train. 16 In the automatic lator alternately

optical

tronics system, most totally

pyrometer,

passes radiation

Because of the null-balance

at the detector

and is determined

from the target

of the elec-

repeatability

is al-

of normal aging effects

in the electronic almost entirely

components,

by the stability

of the pyrometer lamp calibration.

and then from the pyrometer standard lamp at some --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

frequency

independent

and other variations

the modu-

operation

the calibration

such as 90 Hz or 400 Hz; if the photo-

17 Other Automatic

Pyrometers.

The automatic

multiplier receives unequal amounts of radiant energy from the two sources, its response is a

optical

square wave signal,

spect to the modulator driver) determines whether the lamp current is too high or too low. This signal

operated optical pyrometers, stemming from the use of photodetectors other than the eye, together with appropriately selected spectral bandpass

is synchronously

filters.

sible design variations

the phase of which (with re-

demodulated,

then integrated.

pyrometer permits a large number of pos-

The

OPTICAL

not available

MODULATOR

FILTER 7

-

PREAMPLIFIER

-

7 PHOlOMULTlPLlER

AUTOMATIC GAIN CONTROL 1 VARIOUS METHODS USED)

I

REFERENCE

MODULATOR DRIVER -1

AMPLIFIER

1 -

SYNCHRONOUS DEMODULATOR

FIG. 7.5

-

MILLIAMMETER

-

INTEGRATOR

ELECTRONIC PYROMETER

-

LAMP DRIVER

RECORDER

SYSTEM BLOCK DIAGRAM FOR AUTOMATIC - VARIABLE RADIANCE COMPARISON-LAMP

OUTPUT

OPTICAL TYPE.



in manually

ASME PERFORMANCE 18 Some of the practical variations, such as the use of a wavelength significantly different from 0.65 /.nn, may involve the basic instrument design

TEST

CODES

over the next few years. In the interim, some instruments in this category should be considered

tion of the sources of error discussed following Par. 47, especially of the transmissive properties

suitable for ASME Performance Test Code work. Their use in that capacity may yield a significant advantage over either manually operated or automatic optical pyrometers operating at about 0.65 p,

of the medium between the pyrometer and the tar-

but such use should be undertaken only after care-

get. Water vapor and carbon dioxide absorption

in

ful consideration of their suitability to the particular problem at hand; such consideration should be

taken into account. Spectral bandwidth as well as

primarily with respect to the measurement accuracy associated with any particular application.

as described above, but require extensive

considera-

certain spectral regions of the infrared must be wavelength

then becomes inportant.

vantage derives from the ability

Special ad-

to select a spectral Materials

region in which the target emissivity is known to be high, and in being able to make measurements at lower temperatures

22

by operating in the infrared

pyrometer is an optical

in-

strument, it is essential that all compcnents be of high quality and properly aligned and assembled.

region of the spectrum. 19 Some automatic

Since the optical

of Construction

The lenses and other transmitting materials must be free from imperfections which will cause distortion or scattering of the light rays. Particular at-

pyrometers used in the infra-

red region of the spectrum use a standard source of radiance (such as a lamp or a blackbody operated at a fixed temperature) as a reference, but also depend in part for their accuracy upon the stability of

tention should be given to the pyrometer lamp to

active electronic

luminance

components. Other automatic

rometers use no reference

py-

standard of radiance,

depend for their accuracy entirely of active electronic

insure that the tungsten filament

that is used as a radiance reference. Solid-state electronic components used in automatic or semi-

but

upon the stability

components, including

automatic

the

photodetector. called the two-color ratio

to reduce or eliminate

the effect of emissivity

through windows,

atmosphere,

measuring the ratio of the target radiance

etc., by

of the low radiance

at two

measuring,

eter is advisable

in

(or a suitable

factor) has been well established application.

able lighting

the use of a ratio pyrom-

only when the validity

assumption

The low sensitivity

While the state-of-the-art

of the

24

correction

for the particular

are difficult

to

Some automatic

pyrometers operating

in the

errors may occur.

of automatic pyrom-

Where the human eye serves as 25 Precision. the detector in the use of an optical pyrometer, the

etry (other than automatic optical pyrometers such as those described in Pars. 15 and 16) is too much

precision

a

of setting (repeatability)

depends to a

considerable extent on the experience and skill of the observer. The average observer can usually detect a “mismatch” in luminance equivalent to 0.1

comprehensive codification of existing instruments, it is probable that the situation will clarify itself

76 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


readings

infrared can be utilized effectively down to near room temperature; in these instruments, special care must be exercised (see Pars. 47-61), or large

of ratio pyrometers

in a state of flux to permit or make advisable

conditions,

obtain below about 1400c’F.

generally restricts their use to higher temperatures than is the case with “monochromatic” pyrometers. 21.

since one or more absorbing glass

filters may be interposed between the target and the lamp filament, so as to reduce the apparent radiance of the target. For some observers using visual optical pyrometers, especially under unfavor-

terms of temperature. However, emissivity and transmittance of materials can vary markedly with wavelength; since this can lead to large errors if a

emissivity

of bodies below this tempera-

ture. There is basically no upper limit to the temperature that the optical pyrometer is capable of

wavelengths; if the product of emissivity and transmittance at each of the two wavelengths has nearly the same value, the product cancels out in the ratio

correction is not applied,

to

122’V.

23 Range. The optical pyrometer has a low temperature limit of approximately 1300°F because

and

measurement and the instrument reads directly

above approximately CHARACTERISTICS

pyrometer, or simply the ratio pyrometer, attempts transmission

pyrometers should not be subjected

temperature

20 One design concept,

is of uniform

over that portion in the field of view


INSTRUMENTS

2OOOT and a bove, depending tions. The precision creases noticeably

In industrial

for values of

upon viewing

of photometric

as tempera-

ture is reduced below 1600 F, due to decreasing radiance coupled with decreasing visual discrimination of contrast, due to flattening temperature

while the sensitivity

of the lamp current-versusoptical

pyrometer,

variations

termine the temperature of a blackbody. When the temperature of a nonblackbody is being determined

the tem-

a correction should be applied to account for the target emissivity, and allowance must be made for

perature resolution is generally higher than in the manually operated optical pyrometer by an order of magnitude or more [3,4I, and is a function of sever-

uncertainty

al design parameters.

and of the mean-effective-wavelength

noise, in the energy ature *

The resolution

is limited

by

both in the detection system and ultimately randomness of the rate of emission of radiant from the target; the noise-equivalent-temperis reduced below 1600%, due to decreasing

(typically

time, and varies

target temperature

as a function

circuit

design employed.

of

automatic

gain control minimizes

higher resolution)

at higher temperatures.

method of automatic

29

An al-

balancing

system,

the reciprocal

the response

it is nearly constant,

proximately be increased

by increasing

grating time; in practice, to integrating

time.

resolution

30

may

pyrom-

in

is not yet known. The accuracy

of

be determined by the

A good lamp will not change calibration

by

more than a few tenths of a degree over a period of

the inte-

several

corresponding

hundred hours of use (a poor lamp may

change calibration

times of more than a few seconds

is rarely needed and is presently

optical

stability of the pyrometer lamp, and will therefore vary somewhat from one instrument to another.

On

the resolution

of the automatic

the pyrometer usually will

being fixed at ap-

1 sec. In principle, indefinitely

seconds.

be known to per42, 49).

of the mean-effective-wavelength

such instruments

time is adjust-

able from about 0.2 set to several others,

term stability

self-

to an amount that varies with

square root of the integrating

On some pyrometers,

The accuracy

(Pars.

The mean-effective-wavelength may be determined to within approximately 0.2 rnp; however, the long

tains the noise-equivalent-temperature at a somewhat higher but more nearly constant value. The in the electric

computation

eter may be increased substantially [SI by having it calibrated by the National Bureau of Standards.

gain control main-

noise is reduced by integration

which must therefore

mit numerical

thenoise-equiv-

causing it to be lower (giving

ternative

wavelength,

One method of

alen t-temperature,

(Par. 41)

[l] of the py-

sivity or window transmission loss are usually small or moderately small in magnitude (Tables 7.3, 7.4) and are propbrtional to the mean-effective-

in a way that depends upon the

particular

in the value of the emissivity

rometer (Par. 49). Uncertainty in the value of the emissivity must be determined on a case-by-case basis. Uncertainties in the corrections for emis-

0.2 to 0.9 deg F at 1948.OoF) using a 1

set response

in emissivity,

28 Inaccuracies quoted in manufacturers’ specifications (unless specifically stated otherwise) ordinarily apply when the pyrometer is used to de-

decreases

curve.

26 In the automatic

where errors due to

radiation,

observer fatigue, or other unfavorable working conditions may exist, the tolerance will depend on severity of conditions of measurements.

matching de-

and progressively

measurements,

fumes, reflected

condi-

by several

degrees in that

amount of time). Most lamps drift at a rate of ap-

not readily

proximately

0.02 deg F per hour at the gold point,

achievable.

but are subject

27 Accuracy. Th e accuracies attainable [l, 51 in measuring temperatures with an optical pyrometer

deg F when the filament temperature is cycled slowly; they can change by a similar amount when

depend primarily

the lamp is turned off for a few days and then turned on again. Accuracies appropriate to the use

on the optical

system of the in-

to hysteresis

strument (especially the stability of the pyrometer lamp calibration discussed in Par. 29), the condi-

of any other type of automatic

tions under which observations

carefully

certainty

of the emissivity

cations of a high-grade

are made, and unoptical

pyrometer must be

in terms of the particular

pyrometer ACCESSORIES

when used by an experienced observer under favorable conditions may be relied upon to 6 deg F at 2000 deg F and approximately 10 deg F at 3200%.

31 Optical pyrometers are generally supplied complete with optical system, lamp current meas-

uring device, and associated electrical or electronics system. A number of special accessories

*Noise-equivalent temperature is the noise-induced fluctuation in the indicated temperature. 77 Copyright ASME International Provided by IHS under license with ASME No reproduction or networking permitted without license from IHS

of 0.2 to 0.3

application.

of the target. The indi-

portable

evaluated

effects


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

or 0.2 percent of the temperature

AND APPARATUS

ASME PERFORMANCE

or steel or any other material

that is important, such as in some manufacturing processes, it is often adequate to use the radiance temperature ivithout correction. To determine when this can be done it is necessary to have some understanding of the factors that

having an emissivity

short-focus

objective

CODES

actual temperature

are available for certain commercially available units, including: (a) emissivity compensating glass filter to be used for direct temperature measurement (requiring no emissivity correction) of molten iron of 0.4; (b) special

TEST

influence

lens for

emissivity.

Similar

information

is neces-

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

measuring objects of less than l/32-in. diameter. For some models, temperatures of targets whose diameters are as small as 0.005 in. may be meas-

sary when corrections are to be applied to convert radiance temperature to temperature. This informa-

ured. In general, for disappearing filament optical pyrometers, precision suffers when the target width

the Surface of Real ‘Materials”

is less than about four times the filament seen through the optical APPLICATION

32

General.

calibrated

pyrometer

AND

Optical

35

as

cavity

[21.

pyrometers

are ordinarily

hole,

on a black-

this is the preferred

way in

by creating a cavity a hole in its sur-

face, and viewing

emerging from the

is presented

Blackbodies”

regarding

for simulated

design criteria blackbodies

in the following

section

at a uniform

radiation

from

41). enclosed tempera-

characteristic

of

or aperture,

in the enclosure approximates

will

emit radia-

that of a black-

body at that temperature.

to simulate blackbody radiation in the body, such as by drilling

fectiveness

material

blackbody

on “Radiation (Par.

A completely

Blackbadies.

in any opaque

tion that very closely

which to use them. When the temperature of a nonblackbody is to be determined, it is often possible

hole. Information

in the section

the temperature of the cavity walls, but independent of the materials of their construction. A small

when sighted

the radiation

Ractical

ture contains

INSTALLATION

to read correctly

body and, in general,

width,

tion is presented

36 When the cavity is at room temperature, the aperture will appear visually to be very black. However, a small fraction of light incident aperture from outside will be reflected

upon the back out the

aperture after a number of reflections from the cavity walls; since the reflectance of the aperture

and ef-

of this type

is slightly

on “Practical

therefore

(Par. 35).

greater than zero, its absorptance its effective

emissivity)

(and

must be slightly

less than unity (Par. 43). 33

Many furnaces

approximate

tions very satisfactorily, venient

quantitative

fective

emissivity.*

blackbody

condi37 Anything that will reduce the reflectance of the aperture will increase its effective emissivity.

although there is no con-

method for estimating In a perfect

their ef-

blackbody,

This can be done by increasing the number.of times an incident ray is reflected by the cavity walls before it emerges from the aperture (usually

the

details of the inside of the furnace vanish and a piece of steel, for example, that is being heated

by making the dimensions

cannot be distinguished from the background. If the objects in the furnace can be distinguished, but only on close observation,

and if much of the de-

so that more radiation

tail is lost after they have been in the furnace for some time, it is not likely that the temperature measurement

will be seriously

are of a lower radiance.

is possible

in error. If in error

The latter

38 The namber of reflections a ray will make before finally finding its way back out through the

shut off and the furnace allowed 34 When blackbody

conditions

aperture will depend upon the general shape of the cavity and on the detailed character of the surface

condition

roughness

or if it is

to cool.

perfectly specular,

cannot be simu-

rough surface

if the heat supply is variable

materials

lated, it is necessary to account for the effect of emissivity. Where it”is repeatability rather than

of the cavity

walls.

Reflection

from a

smooth surface is described as being while the reflection from a perfectly is described

have surfaces

as being diffuse.

Real

that are characterized

by

a mixture of specular and diffuse components of reflected radiation. Effective emissivities near unity are more easily attainable in cavities with specu-

*Since the emissivity of a blackbody is unity by definition, it is a contradiction in terms to speak of a blackbody with emissivity less than unity. The most common solution to this problem, in the case of a simulated blackbody, is to speak of its “effective” emissivity.

larly reflecting materials;

than with diffusely

however,

considerable

taken when the interior 78


is absorbed at each reflec-

tion.

at all, the observed temperature will be too high when the furnace walls are of higher radiance than the material being heated and too low when the walls

of the aperture small

compared to those of the cavity), and by constructing the cavity walls of a low reflectance material


surface

reflecting

wall

care must be of the cavity

is

INSTRUMENTS specular, fective

because under these conditions

emissivity

seriously

is significantly

AND APPARATUS

the ef-

directional,

=

c

large errors may result from viewing the

determined parameters,

= area of aperture

S

= area of interior surface of blackbody including the aperture

n ,=

shape in terms

the solid angle of radiation

various formulas

cance. The method of DeVos [6] is generally

con41

and is

from the Surface of Real Materials. may be desig-

nated as a real body, to distinguish it from a blackbody. The interior of an opaque real body is totally absorbing, and when at a uniform temperature must

however, it is mathemati-

of radiation

Radiation

A body made of an actual material

cally very cumbersome and is not recommended for routine engineering applications unless high accuracy is mandatory. A review of the subject from the viewpoint

emerging from

back wall of the cavity.

commonly used as a reference against which other formulas are evaluated;

cavity,

the cavity aperture, having its apex at the intersection of the viewing axis with the

or graphs have been devised [6-101 that are applicable in most special cases of practical signifisidered to be valid and of great generality,

forming the blackbody

s

39 While there is as yet no simple formula for accurately expressing the effective emissivity of of easily

of materials

interior surface

cavity from the wrong direction.

an aperture in a cavity of arbitrary

emissivity

and

therefore radiate as a blackbody. When blackbody radiation from the interior approaches the surface, part of it is reflected back into the interior; the re-

thermometry has recent-

ly been made by Bedford [i’].

mainder passes through the surface and is emitted. Gouffe’s

computation arbitrary

Method [8]. of effective

yeilds

method for

emissivity

for cavities

shape assumes perfectly

from the cavity walls; cavities

Gouffe’s

effective

to increase

emissivity

diffuse reflection

flected

it is exact for spherical

(with diffusely

reflecting

emissivity

low for cavities

The fraction that is emitted is defined to be the of

walls),

values

but

that are slightly

fective

emissivity

It is

because it is concisely

lated for easy application;

tral bandwidth)

ef-

more nearly exact

values of effective emissivity may be obtained from the references cited above [9,10] for certain

l/T,

commonly used cavity shapes and for wall materials having both specular and diffuse components of reflectance.

where

where

(3)

t

,

c,, = c [l -

and k = (1 - 0

w91 + (s/s) [(s/s)

-

(#)I

, a small negative number, tendcavity shape approaches that of

emissivity

of blackbody

T, and To is as

= he/c2 loge q, , To > Tr

l/T,

T,

=

absolute temperature

T,

= absolute temperature of the target as in-

(41

of the target in kelvins

dicated by the pyrometer,

called

tral radiance (luminance) kelvins

temperature,

mean-effective-wavelen@h,

in meters

the specin

of the target surface

43 The blackbody radiation incident upon-the surface from the interior must either be emitted or

aperture


at a lower temper-

-

spectral emissivity

a sphere. effective

between

0.0 14388 mK

ing to zero as the

co I

to be a blackbody

ature T,; the relationship follows [ll:

formu-

Eg = c; (1 + It)

the sur-

solute temperature TO and having an emissivity 6~ will appear to the pyrometer (having a narrow spec-

the use of

presented here for use as a guide in estimating

which has

42 The spectral emissivity l A is the fraction by which blackbody radiation is reduced in the process of being emitted from the surface. A surface at ab-

the further the cavity shape departs

the method for many common applications.

approaching

face from either side.

of other shapes. The error tends

of cavity shapes to justify

The fraction that is re-

the same value for radiation

from that of a sphere, but is small enough for a wide variety

of the surface.

is defined to be the reflectance,


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

40

ASME PERFORMANCE internally

reflected,

for every wavelength;

this may

TEST

CODES

under consideration

be expressed as follows:

should be used, rather than

published values such as those in Tables

7.1 and

7.2; an often used method is to drill a small hole +

where

RA

=

in the surface of the material

1

CA = spectral

emissivity

of the surface

RA = spectral reflectance

in question to provide

a blackbody cavity. From an optical pyrometer determination of the radiance temperature of both the blackbody

of the surface

cavity and the surface of the material

adjac.ent to it in the temperature 44

From the above expression

it can be seen that

the spectral

a good emitter is a poor reflector, and vice versa. Thus, carbon has a high emissivity and a low re-

Eq. (4)_or the A-value

Since reflectance slightly direction the emissivity. A constant fraction

47

effect on emissivity .

ized only as an approximation.

T4BLE

Actual

The

sources

of error

in op-

SPECTRAL EMISSIVITY OF MATERIALS, SMOOTH SURFACE, UNOXIDIZED Wovelength

materials

= 0.65 pn (red light)

Motoriol

Solid

Liquid

regions. Beryllium Carbon Ch romium Cobalt Columbium

flectance, and will therefore have its lowest possible emissivity. If the surface is roughened, its reflectance

will be reduced because of increased

absorption due to multiple

reflections

(within

the

small cavities constituting the surface roughness), but ita emissivity will be increased by a like amount. Emissivity is thus seen to be dependent upon the state of surface roughness of the radiating body. Tables 7.1 and 7.2 list the spectral emisof the more common engineering

materials

(Refs. 11-14 contain an exhaustive compilation and evaluation of emissivity data on a large number of materials),

and Table

7.3 presents

that must be added to indicated

to

for pyrometers

operating at 0.65 pm. A more complete tabulation of corrections may be found in NBS Monograph 30 [15]. In using Tables 7.1 and 7.2 it is necessary to take into account the state of surface roughness, which will tend to increase the emissivity over the values listed

in the table.

will also influence

often causes difficulty exhibit

The extent of oxidation

the emissivity, in practice,

surface oxidation

0.61 0.80-0.93 0.34 0.36 0.37

an effect that where heated

0.10 0.55 0.14 0.30 0.35

0.15 0.38 0.22 ... 0.37

Manganese Molybdenum Nickel Palladium Platinum

0.59 0.37 0.36 0.33 0.30

0.59 0.40 0.37 0.37 0.38

Rhodium Tantalum Thorium Titanium

0.24 0.07 0.49 0.36 0.63

0.30 0.07 ... 0.40 0.65

Tungsten Uranium Vanadium Yttrium Zirconium

0.43 0.54 0.35 0.35 0.32

... 0.34 0.32 0.35 0.30

Steel Cast Iron Constantan Monel Chrome1 P (90 Ni-10

0.35 0.37 0.35 0.37 0.35

0.37 0.40 . ..

0.35 0.36 0.37 0.27

... ... ... . ..

80 Ni-20 Cr 60 Ni-24 Fe-16 Cr Alumel (95 Ni;Bal. 90 Pt-10 Rh

changing with

46 When possible, a measured value of the spectral emissivity of the particular piece of material

Cr)

Al, Mn, Si)

*From the “Handbook of Chemistry Rubber Publishing Company.


0.61 .. .. 0.39 0.37 0.40

C ower Erbium Gold Iridium Iron

Silver

corrections

temperatures

correct for the effect of emissivity

materials time.

using using

(Roarer and Wensel, Notional Bureau of Standards)’

45 If the surface of a particular material is perfectly smooth, it will have its highest possible re-

sivities

7.1

real-

may be considered to be gray only in restricted spectral

Sources of Error.

a perfect graybody

that can be experimentally

may be calculated,

tical pyrometry may be broadly grouped into three categories: (a) those associated with the pyrom-

is wavelength dependent and and temperature dependent, so is material that reflects and emits a at all wavelengths is said to be a

l’kI e a blackbody,

“graybody”;

is an idealization

range of interest,

may be calculated,

Eq. (6), (Par. 55).

flectance, while platinum has a low emissivity and a high reflectance. Anything that affects reflectance must have a corresponding

emissivity


and Physics,”

. .. ...

Chemical

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

ch

INSTRUMENTSANDAPPARATUS eter, (bl those associated

with the media between

radiance measured by the pyrometer, ally correctable.

the pyrometer and the source, and (cl those asso-

and are usu-

ciated with the source. 7.2

SPECTRAL EMISSIVITY SMOOTH SURFACES Wavelength

(Roeser

= 0.65

and Wensel,

p

National

OF

(red

OXIDES

WITH

49 The mean-effective-wavelength varies somewhat from one model to another, but for optical pyrometers it is usually between 0.636 q and 0.662

ligh’t)

For the type shown in Fig.

p.

typically

Bureau of Standards)*

the mean-effective-wavelength Range of Observed Values

Material

1

_-L

Aluminum Beryllium

0.22 0.07 0.58 0.60

oxide

to to to to

0.40 0.37 0.80 0.80

oxide Cerium oxide Chromium oxide Cobalt oxide

. .. . . . . .

Columbium oxide C opper oxide Iron oxide Magnesium oxide Nickel oxide

0.55 0.60 0.63 0.10 0.85

Thorium oxide Tin oxide Titanium oxide Uranium oxide Vanadium oxide

0.20 to 0.57 0.32 to 0.60

Yttrium oxide Zirconium oxide Alumel (oxidized) Cast Iron (oxidized) Chrome1 P (90 Ni-10 (oxidized)

........

bration of the reference

filter,

and the accuracy

the mean effective

in disappear-

cause variations

to values supplied

by the

in the value of the meansffective-

since the uncertainty

50

Effect.

Size-of-Source

due to differing will rarely ex,

Radiation

from outside

of the target area but from the immediate neighborhood of the target is found to influence the pyrometer indication at least to a small extent; this is called the size-of-source effect. For visual and automatic optical pyrometers, the effect is most noticeable

for small targets.

It is caused primarily

of radiation

within

the pyrom-

0.87

eter optical

........ ........

0.90 0.83

eters) by heating of the pyrometer lamp filament the incident radiation; on the upper temperature

........

0.78 0.75

........

0.84

. . . . . . . .

0.80

. . . . . . . . 0.25 to 0.50

0.85 ...

and Physics,”

Chemical

the spectral

of the absorption

optical

Rubber

trans-

and some

pyrometers;

small

below,

source area is,at nominally

same temperature

as the target,

the effect I,

absolute

the

as the source area is increased.’

If the extraneous

is proportional (target)

temperature

fective-

wavelength;

herently

larger in infrared

the magnitude

the of

to the square of the and to the mean-ef-

the effect

is therefore

sensing


by

of the two possible locations of F, in Fig. 7.4) reduces the filter heating effect to a negligible level [3]; in this configuration, an automatic optical pyrometer with clean optical surfaces has a size-of-

higher temperature

glass

the extent to which they reduce (or increase)

pyrom-

source effect usually not greater than 0.6 deg F at the gold point (194&O%‘), tending to indicate a

of

these were discussed in the section on “accuracy.’ Most of the other errors can be treated in terms of

pyrometers

optical

optical

51 In the automatic optical pyrometer, location of the range filters behind the mirror aperture (one

except for

discussed

and (in visual

transmission changes in the range filters can occur if the filters are heated somewhat by absorbed radiation.

of cali-

lamp, the determination

wavelength,

ranges of visual

with the pyrom-

with the stability

of calibration;

system,

models of automatic

........

of

wave-

........

the mean-effective-wavelength, characteristics

among filters

pyrometers

of the

from the

percent in the mean-effective

by the scattering

48 Sources of error associated eter have to do primarily

Variations

available

visual responses among observers ceed 0.2 percent.

Cr)

of Chemistry

is ordinarily

optical

as much as +l

wavelength,

0.60 0.40 0.87 0.70

........ ........

ing filament

nant uncertainty

0.50 ... 0.50’ 0.30 0.70

0.18 to 0.43

manufacturer.

A curve of

as a function

manufacturer, giving rise to corresponding uncertainties in computed emissivity and window transmission corrections; this will usually be the domi-

0.70 0.70 0.70 0.20 0.90

0.80 0.98 0.43 0.96

........

Carbon Steel (oxidized) Stainless Steel (18-8) (oxidized) Porcelain *From the ‘*Handbook Publishing Company.

to 0.71 to to to to

target temperature

length 1161, relative

0.30 0.35 ... 0.70 0.75

........ ........

80 Ni-20 Cr (oxidized) 60 Ni-24 Fe-16 Cr (oxidized) 55 Fe-37.5 Cr-7.5 Al (oxidized) 70 Fe-23 Cr-5 Al-2 Co (oxidized) Constantan (55 Cu-45 Ni) (oxidized)

mittance

Proboble Value for the Oxide Formed on Smooth Metal

7.2, it is

assumed to be about 0.65 pm.


in-

pyrometers.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

TABLE

ASME PERFORMANCE 52

Very few data are available

of this error in disappearing

on the magnitude

filament

optical

X,, the spectral

or if a much higher temperature

TX, and the second

because the

mean-effective-wavelength

slightly

temperature),

varies

with

and may thus be used to relate

other

T and 7,.

The window transmittance, and thus the A-value, are dependent upon the di-

to the target is at

a much higher radiance temperature

transmittance

To

constant c2 are known. It is usually more to experimentally determine the value of

nearly constant (it varies slightly

values of

53 The size-of-source effect can cause a very large error, especially in pyrometers operating in if the area adjacent

transmittance

A by direct measurement of T and To; A is very

pyrometer is of the

order of 2 to 4 deg F (at the gold point) difference between viewing very small and very large targets.

the infrared,

same as that of a window of spectral

radiation practical

mize the effect of scattered radiation. The small amount of available data [4] suggests that the effilament

CODES

7A* For any measured value of T, the value of may be obtained if the mean-effective-wavelength

py-

rometers; the pyrometer lamp filament is heated slightly by the radiation of the target image, and no special precautions have been taken to mini-

fect in a disappearing

TEST

than the target,

source in the back-

rection of the transmitted

radiation.

tance is highest in the direction

The transmit-

normal to the

surface.

ground (behind the target) lies near the line of sight.

56 A tabulation of the difference between T and To appears in Table 7.4 1151 for the case of a thin

54 The effect can be almost entirely eliminated for a particular application by calibrating the py-

piece of clear window glass. Although surface reflectance accounts for almost all of the reduction in transmittance for very thin windows, the trans-

rometer against a target of the same size and shape, and at the same distance as is to be used

mittance

in the intended application.

advisable

Absorption. Sup55 Windows and Atmospheric pose a pyrometer sighted on a blackbody indicates

transmittance

a temperature To (expressed on the Kelvin scale), but indicates a lower temperature 7’ when viewing

57 The effect of atmospheric transmission is analogous to that of window transmission. Rowever,

the same blackbody

if atmospheric attentuation is not visually the atmosphere is sufficiently transparent

transmittance

rh.

through a window having a

The relationship

between the

temperatures indicated with and without the window in place is given [l] to a close approximation by l/T,

- l/T

= he/c2

Note the similarity

logerA

=

-A,

To> T

(6)

between Eqs. (4) and (6); the

role of the spectral

emissivity

CA is exactly

of a window tends to decrease with in-

creasing thickness especially

it is usually

experimentally,

for windows that are thick or if reduced is visually

detectable.

apparent, that the

correction is negligible for any pyrometer using only visible red wavelength radiation. Where it is not negligible,

atmospheric

to be so variable

transmission

is likely

as a function of time as to render

computed corrections mon practice

the

due to absorption;

to determine the A-value

impractical.

The most com-

in such cases is to sight the pyrom-

TABLE 7.3 EMISSIVITY AND TRANSMITTANCE CORRECTIONS For Addition To Observed Temperatures for Optical Pyrometer Using Red Light at Wavelength = 0.650 /&II Indicated Temperature

0.1

Spactrol

Emissivity

0.2

0.3

0.4

0.5

or Transmittance 0.6

0.7

0.8

0.9

1.0

41oc 50 60 71 83

31-z 37 45 53 62

22% 27 33 39 45

16% 19 23 27 31

10% 12 14 17 19

5% 6 7 8 9

0 0 0 0 0

7oooc 800 900 1000 1100

IlOoc 135 163 194 229

74% 91 109 130 152

54oc 66 80 95 111

1200 1300 1400 1500 1600 1700

266 308 352 401 453 510

177 203 232 262 295 330

128 147 168 189 212 237

96 110 124 140 157 176

71 81 92 104 117 130

52 59 67 76 85 94

36 41 46 52 58 65

22 25 29 32 36 40

IO 12 13 15 17 19

0 0 0 0 0 0

1800 1900 2000

570 634 704

368 408 450

263 291 321

195 215 236

144 159 174

104 115 126

72 79 86

44 49 53

21 23 25

ii 0

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



INSTRUMENTS TABLE

7.4

WINDOW CORRECTIONS

such as by the edge of a misaligned

Corrections to be added to observed temperatures to correct for (reflectance corresponding to a refractive index of 1.57) loss of light transmitted through a single thin window of uncoated nonabsorbing glass. Absorption losses, dependent on the character and thickness of the glass, are likely to increase the corrections for thickness greater than about 1 mm. Wavelength

AND APPARATUS

appearing filament

pyrometers,

filament

disappearance

such obstruction

Correction to be Added OC

800

6

1000

8

1200

10

1400

13

1600

17

1800

21

2000

25

2200

29

2400

34

2600

40

2800

46

3000

52

59

is a good indication

When an optical

Emissivity.

radiation),

its temperature

apply a correction

pyrometer is in the open

indication

will be low

emissivity.

for the effect

Failure

to

of the emissivity

of a nonblackbody, or a large uncertainty in the emissivity value used, will lead to a significantly large error or uncertainly in the measured temperature; this is usually the largest potential source of error associated with optical pyrometry. The error, or a correction

for it, may be computed from Eq. (4)

or an estimate

[ll-141

Tables

of it may be obtained from

7.1, 7.2, and 7.3.

60 Reflected Radiation. A radiating target also reflects radiation from its surroundings. The error caused by such reflection

[S] depends especially

strongly on the specular

component of reflectance,

eter through a tube through the offending region

i.e.,

another radiating

gas is slowly

When a sight tube is used, precautions

that

amounts of extraneous

due to the effect of spectral

through which a clean transparent

in effect a transparent

obstruct-

does not exist.

(unexposed to significant

purged, creating

a partially

ed entrance aperture will make it impossible to ob tain good disappearance of the filament; complete

sighted on the surface of a material

= 0.650 pm

Observed Temperature oc

window or

sight tube, the pyrometer will read low. In the dis-

the tendency of the target to form an image of body, as seen by the pyrometer.

window.

If the specular

component of reflectance

must be ob

cant and especially

if the radiance

is signifi-

temperature

of

served so as not to restrict the entrance aperture of the pyrometer, as discussed in the next para-

the other radiating

graph. For automatic

sighted on by the pyrometer, large errors may result; even if the target is diffuse, the resulting er-

pyrometers

operating

infrared, the errors caused by atmospheric tion may be severe at certain wavelengths,

in the absorpand are

a function of absolute humidity and of the distance between the pyrometer and the target.

body (such as an incandescent

lamp) is high compared to that of the target being

ror may be significant. This effect is most pronounced when the target temperature is low. Screen-. ing the radiation

from the offending

ing the target from a direction 58

Peepholes

and Sight

Tubes.

When a pyrometer

is used to view through a peephole, sight tube, or any other small diameter opening, care must be

taken to assure that th,e entrance aperture of the pyrometer is not obstructed. The entrance aper-

source or view-

where the reflected

component is not likely to be seen is recommended. The magnitude of the error is larger in infrared sensing pyrometers. 61

Nonuniformity

of Cavity

Wall

Temperature.

ture of the pyrometer may be thought of as that por-

The

tion of the objective lens through which radiation must pass to be measured; it is ordinarily a circular area slightly smaller in diameter than the ob

practical blackbodies by computation only of the reflectance contain the assumption that the cavity walls

jective

ten not very well met in practice.

lens. The cone of radiation

having the en-

methods of determining

are at a uniform temperature,

trances aperture as a base and a point on the target

distribution

(at which the measurement is made) as the apex, which may be called the entrance cone, must be free from obstructions to assure that the entrance aperture is unobstructed. If an obstruction occurs,

the walls

a condition

by a single

emissivity

few special

temperature,

has been calculated

cases [lo].

of-

in which cannot and the for only a

In the determination


of

The radiation

from an aperture in a cavity

be characterized effective

emissivity

are not at a uniform temperature

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


effective

of ra-

ASMEPERFORMANCETEST CODES diance temperature using monochromatic optical pyrometers, this condition will lead to errors that are less than the maximum wall-temperature difference. Temperature uniformity of the cavity is

an eyeshield so as to screen the observer’s from extraneous light during an observation.

dependent,

of the

ing, such as a photographer’s

The

making pyrometer readings

in part, on the thermal diffusivity

cavity walls,

and the heat transfer

geometry.

use of highly conducting

materials

small cavity dimensions,

and thick cavity

tend to promote temperature

64

pyrometers are usually

be advantageous

(where possible),

uniformity.

Optical

to make use of additional

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

a nonblackbody

Radiation.

Radiation

surface is slightly

at angles approaching

in a brilliantly

lighted

of Data. Table 7.3 shows the cor65 Treatment rections that must be added to the readings obtained with an optical pyrometer for various emissivities

by the

emitted from

polarized,

to obtain the true tempera-

tures [15] , or the appropriate calculated

usu-

directly

corrections

may be

from Eqs. (4) or (6). Corrections

for reflected light from extraneous sources can sometimes be made; the experimental and analytical techniques required, however, are outside of the

ally being negligibly polarized in a direction normal to the surface and somewhat strongly polarized

screen-

black cloth, when

walls all

or window transmittances Polarized

eye It may

area.

The dimen-

sion, however, is necessarily compromised minimum acceptable aperture diameter.

62

supplied with

the tangent. In a

scope of this discussion.

pyrometer using reflection optics, some polarization of the reflected radiation will occur which, if uncompensated within the pyrometer, may cause an error when viewing polarized radiation. This error can ordinarily

be made negligible

target in a direction

within

by viewing

the

ADVANTAGES

66

its surface.

(4

63 Essential Considerations. Since the optical pyrometer is basically an optical instrument, all precautions

related

to handling

and to cleanliness and reflecting

The instrument

has a long life expectancy,

high temperature

of transmitting

proper alignment

Advantages

since no part of it is exposed to destructive

of the pyrometer

of the surfaces

components,

AND DISADVANTAGES

15’ to the normal of

(b) The indicated

and

effects.

temperature

is that of the body

under test and not that of some adjacent

focusing should be exercised. Regular cleaning of the objective lens prior to use of the pyrometer is

(for example, a thermocouple) to be at the same temperature

especially

der test; the temperature

important;

lens tissue

is recommended

for this purpose. A thin film of foreign material the objective

lens is often more detrimental

curacy than several lens surface,

on

is not disturbed

to ac-

(c)

small opaque specks on the

such as fine droplets

metal. If the accumulation

of splattered

readings

responding to about 1 deg F error at a target temperature of 2000%‘), however, the lens should be replaced. Care should also be exercised so as not

peratures

Temperatures

(0

the lamp filament may eventually

at its natural

frequency,

cause the filament

manufacturer’s

recording

the

conditions,

optical

method for de:

optical

pyrometer is suitable

and controlling

for

temperature.

(h) The temperature of moving objects can be

with

measured,

recommendations.

since direct contact is not required.


influencing

high temperatures.

The automatic

which

to break. Main-

tenance should be performed in accordance

When used under suitable termining

of

such as thin

of the body.

pyrometry is the most accurate

Care should vibration

of small bodies,

can be measured without

temperature

If higher lamp tem-

detectable

in photometric

(e)

wires,

are required to match the source, the

be taken to avoid visually

with moderate accuracy

radiance temperature

The method is rapid under most circumstances.

to the maximum of

next higher range should be employed.

permits accurate

(d)

to operate the pyrometer lamp at temperatures the low range scale calibration.

The high rate of change of spectral

match.

pies as much as 0.5 percent of the lens area (cop-

higher than that corresponding

of the body under test

by the pyrometer.

with temperature

of such droplets occu-

body

that is assumed as the body un-


INSTRUMENTS 67

AND APPARATUS Science

Disadvantages

(cd Errors

may be caused by the presence of win-

dows, smoke, observer tion

incandescent

and the object

from the target

of radiation

gases

under

reflected

(especially

between

test,

the

or by radia-

sources

from other

if the extraneous

source is at a higher temperature than the tarSuch errors may be difficult to prevent or

get).

to otherwise

correct.

(b) It is necessary to simulate blackbody tions or to have a knowledge

pyrom-

eter.

(d) The visual optical pyrometer does not lend itself readily

(4

(0

to automatic operation.

Line-of-sight Initial

viewing is usually

capital

investment

required.

may be relatively

high, depending upon the circumstances.

Thermophysical Properties of Matter, IFI/Plenum Data Corp. [13] “Thermal Radiative Properties of Non-Metallic TPRC Series on the Thermophysical Prop Solids.” - , erties of Matter, IFI/Plenum Data Corp., (in press). Radiative Properties of Coatings,” [141 “Thermal TPRC Series on the Thermophysical Properties of Matter, IFI/Plenum Data Corp., (in preparation). Optical Pyrometer Readings,” National 1151 “Corrected Bureau of Standards Monograph 30, U.S. Government Printing Office, Washington, D.C. “Recent Advances in’optical Pyrometry,” D. R. “Temperature, Its Measurement and ConLovejoy,

REFERENCES

68 Throughout

the text Reference numbers are en-

closed in brackets,

thus 111.

[1I “Theory

and Methods of Optical Pyrometry,” H. J. Kostkowski and R. D. Lee, “Temperature, Its Measurement and Control in Science and Industry.” vol. 3, part 1, p. 449, Reinhold, New York, 1962. P. H. Dike, W. T. Gray and [21 “Optical Pyrometry,” F. K. Schroyer, Leeds and Northrup Company, 1%6. [3] “A High P recision Automatic Optical Pyrometer,” G. D. Nutter, “Temperature, Its Measurement and Control in Science and Industry,” vol. 4, ISA Pittsburgh.

[161

trol in Science and Industry,” 506, ISA Pittsburgh.

vol. 3, part I, 487-

Thermometry, Additional Reference: “Radiation Part 1 and 2,” G. D. Nutter, Mechanical Engineering, June and July, 1972.

[4] “The NBS Photoelectric Pyrometer of 1961,” R. D. Lee, “Temperature, Its Measurement and Control in

85 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


Reinhold,

A. Gouffd, Revue d’optique, 24, 1, 1945. [9] “Radiation Heat Transfer,” E. M. Sparrow and R. D. Cess, ch. 3 and 6, Brooks-Cole Publishing Company, Belmont, Calif., 1966. [lo] “A Note on the Numerical Evaluation of Thermal Radiation Characteristics of Diffuse Cylindrical and Conical Cavities,” B. A. Peavy, Journal of Research of the National Bureau of Standards, 70 C, 139, 1966. Ill] “Theory and Measurement of Emittance Properties for Radiation Thermometry Applications,” D. P. Dewitt and R. S. Hernicz, “Temperature, Its Measurement and Control in Science and Industry,” vol. 4, ISA Pittsburgh. [12] “Thermal Radiative Properties of Metallic Elements and Alloys, ‘* vol. 7, TPRC Series on the

The human element enters as an important source of error with the visual optical

vol. 3, part 1, 511,

[8] “Corrections de’ouverture des corps-noirs artificiels compte tenu des diffusions multiples internes,”

condi-

of the spectral

emissivity of the material sighted upon in order to obtain its temperature accurately.

(c)

and Industry,”

New York, 1962. Pyrometer and Its Use in [51 “The NBS Ph otoelectric Realizing the International Practical Temperature Scale above 1%3°C.” R. D. Lee, Metrologia, vol. 2, no. 4, 150, Oct. 1966. [6] “Evaluation of the Quality of a Blackbody,” J. C. [7] ~;;os;.Physico, 28,.669, 1954. ec tve Emissivtttes of Blackbody Cavities A review with Applications to Pyrometry,” R. E. Bedford, “Temperature, Its Measurement and Control in Science and Industry,“vol. 4, ISA Pittsburgh.


CHAPTER 8, BIMETALLIC THERMOMETERS tates a shaft and attached

CONTENTS

Par.

tion is essentially GENERAL: Scope ............ .................................... ........ ...... ............ Definitions

. .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 2

.................. ................

4

PRINCIPLES OF OPERATION CLASSIFICATION:

A scale gradu-

linearly

related

to A T, the

scale is graduated linearly. A calibration must be performed over the entire range span, however, because the actual pointer motion may not be linear. Mechanical

Description . . . . . . .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials of Construction .. . ... .. .. .. .. .. .... ... .. . . . .... .. . .. . CHARACTERISTICS: Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............. ................................................................ Sensitivity ...................................................... ............ Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . Response Mechanical Stability .................... ...... ...... .... ............ . .. .... . .. .. . . .. .. .. .. .. .. .. .. . ... .. ... . . ... .. ... . .. Thermal Stability . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ACCESSORIES APPLICATION AND INSTALLATION .. . ... ... . .... .. . .. .. . .... .... .. ... . .. ... .. . .. .. . ..... .. .. .. Other Considerations Treatment of Data . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ADVANTAGES AND DISADVANTAGES: Advantages . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . Disadvantages .. . ... .. .. ... . .. .. . .. . .. .. .. .. .. .. ... . . . .. .... . ... .. . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . REFERENCES

pointer.

ated in temperature units is used to relate pointer motion to temperature. Since bimetal element rota-

5 10

affect 3

11 12 13 14 15 16 17 18 19 20

friction

and geometrical

alignment

can

linearity.

The

Immersion

Length

immersed in the flowing

is the length

fluid.

immersed in a stagnant fluid.

of the bulb

It is also the length Insertion

Length

is

the distance from the thermometer bulb tip to the highest possible point of fluid immersion. This distance is from the bulb tip to the base of the threads in Fig. 8.3. If the thermometer bulb was inserted

in a properly fitted

well,

Insertion

length

would be from the well tip to the base of the well

21 22 23

GENERAL

scope INDICATING DEVICE 1 The purpose of this chapter

formation

is to present

in-

INSTRUMENT CA%

which will guide the user in the selec-

tion, installation ters.

and use of bimetallic

thermomeBULB

Definitions 2 A Bimetallic an indicating ment called

Thermometer

or recording a bimetallic

means for operatively A bimetallic

differential

of

ele-

bulb and a

the two (Fig.

bulb is comprised

and its protective

coil rotation,

a sensing

thermometer

connecting

thermometer

metal element The bimetal

is one consisting

device,

sheath (Fig.

8.1).

of a bi8.2).

which is caused by the

thermal expansion

of the metals,

FIG. 8.1 BIMETALLIC THERMOMETER (Courtesy of Scientific Apparatus Makers Association)

ro-

86 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



INSTRUMEW'SANDAPPARATUS threads. Insertion length can be distinguished immersion length with the following example.

from In a

ternal standard pipe thread connections. diameter varies

from approximately

The bulb

l/8

to 3/8

in.

immersion length fully filled pipe fl ow application, is the portion of the thermometer bulb exposed to

depending on the model, bulb length, and manu-

the flowing

to 5 ft are available.

fluid.

However,

a relatively

stagnant

facturer.

Bulb lengths from approximately

2-l/2

in.

fluid column surrounds the remainder of the thermometer bulb between the inner pipe surface and the base of the threads. Length is the distance

The Sensitive

from the thermometer

tip to the upper end of the bimetallic

bulb

element.

See

Fig. 8.3. PRINCIPLES

4 The operation

6

Laboratory

thermometers

Portion

Type

Thermometer.

cies than the industrial

These

by higher accura-

type and the absence

of

threaded connections. Bulb diameters and lengths are available in the sizes given for Industrial thermometers.

OF OPERATION

of a bimetallic

pends upon the difference

thermometer de-

in thermal expansion

r

of

two metals. The most common type of bimetallic thermometer used in industrial applications is one in’which

or Test

are characterized

a strip of composite

material

I

is wound in

the form of a helix or helices. The c’omposite material consists of dissimilar metals which have been fused together to form a laminate.

The differ-

ence in thermal expansion of the two metals produces a change in curvature of the strip with changes in temperature. is used to translate

The helical

construction

this change of curvature

to

-

rotary motion of a shaft.

-

CLASSIFICATION FIG. 8.3

Description 5 Industrial Type Thermometer. These thermometers are generally supplied with l/2 or 3/4 in. ex-

(Courtesy

of Scientific

NOMENCLATURE Apparatus

One method of increasing

Makers

accuracy

Association)

is to increase

the number of coils in the bimetal element thus increasing the angular motion for a given temperature change. 7 Straight and angle forms of bimetallic eters are available. spring is generally

In the straight

thermom-

form, a helical

employed to transmit

the rotary

motion of the shaft through an angle to the pointer. EXTENSION

See Fig. 8.4. In the angle form, the pointer is attached to the shaft. See Fig. 8.5. 8 Case diameters

ELEMENT

J

range from 1 to 6 in. with ef-

fective

scale

lengths from 2 to 12 in. Graduationc-

usually

cover 270 to 300 angular degrees.

9 Thermometers are also available with frictionrestrained extra pointers that indicate maximum or

FIG. 8.2 BIMETALLIC THERMOMETER BULB (Courtesy of Scientific Apparatus Makers Association)

minimum temperatures. Some manufactureis offer thermometers with fume-proof casings. Thermometers are also manufactured which have silicone oil damping in the stem for protection against shock and vibration.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



ASME PERFORMANCE

TEST

CODES

Silicone oil damping cannot be used above approximately 5OOoF. Damping is also accomplished by placing a high viscosity oil in the clearance of the upper shaft bearing. Thus, damping can be accomplished on high range thermometers.

FIG. 8.5 (Courtesy

SECTIONAL INDUSTRIAL

of Weston Instruments-Division

of the laminated

FIG. 8.4 (Courtesy

STRAIGHT FORM INDUSTRIAL BIMETALLIC THERMOMETER of Weston Instruments-Division of Daystrom, Inc.)

Materials

VIEW OF ANGLE FORM BIMETALLIC THERMOMETER of Daystrom,

Inc.)

glass can soften at high ambient

temperatures (15OoF or more). Therefore, tempered glass may be a better choice for high range shock resistant thermometer applications where the case temperatures facturers’

will be at or above l5OoF. Manu-

literature

should be consulted

before

selection asthe above statement may not apply in every case. Plastic windows are adequate for shock resistance but also soften at ambient tem-

of Construction

peratures above 150%‘. On hermetically sealed thermometers, the internal air pressure increases

rosion resistant

as temperature increases and can cause a softened plastic window to blow out.

and will withstand

pressures up to

2OOOpsi, Where pressure or corrosive indicate

conditions

the need for greater protection,

corrosion resistant

materials

wells

are available.

of

Suit-

CHARACTERISTICS

able plastic or lead coatings may be applied directly to the protective shell to overcome some corrosive conditions. Case windows are available plastic,

in plain glass,

or safety glass.

The safety glass employed is either laminated glass or tempered glass. The bond between layers


11 Range. Bimetallic thermometers are available in temperature ranges from - 200 to looo°F;

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

10 Bimetallic thermometers are available with 18-g type stainless protective shells which are cor-

however, they are not recommended for continuous operation above 800%‘. Range changes are made by use of different bimetal

materials

or by changing the

element length (number of coils).

is shortened as span is increased.


Length

INSTRUMENTS

AND APPARATUS ACCESSORIES

12 Sensitivity (Bimetal element angular motion for a given temperature change) is determined by

17 Wells are the major accessories for bimetallic thermometers. Where pressure, corrosion, or erosion indicate the need for greater protection of the bi-

the physical characteristics of the bimetallic element and the dimensions of the helix used. A maximum sensitivity of approximately three angular degrees displacement per Fahrenheit degree may be expected. 13 Accuracy bimetallic

of temperature measurement

metallic element than offered by the bulb protective shell, wells should be used. Care should be taken

with a

under which the measurement

of the well filling

avoid materials

detrimental

material

to

to the protective

shell

of the thermometer bulb being used. For a general discussion of wells refer to Chapter 1.

thermometer depends on thermometer de-

sign, environment

in the selection

is

taken, proper immersion, accuracy of the thermometer’s calibration, bimetal element thermal stability and observer errors such as parallax. 14

Response of bimetallic

function

APPLICATION

thermometers

AND INSTALLATION

18 In order to obtain an accurate

is a

temperature

measurement, the thermometer stem must be immersed so that the sensitive portion reaches the

of thermometer design and use conditions.

The 63 percent time constant is commonly used in the evaluation of response. Response characteristics of high quality bimetallic thermometers, both

medium temperature

laboratory and industrial types, are somewhat similar to those of liquid-in-glass thermometers. High

thermometer must be kept small enough to prevent a significant change in indicated temperature. The

quality

a time constant of three to four seconds in a well

proper depth of immersion depends on the stem material, bimetal element length and the temperature

stirred water bath. However,

and heat transfer environment.

bimetallic

industrial

thermometers

will have

the manufacturer

can

vary the response by the size of the bimetallic element, by the care he exercises in fitting the bimetallic shell,

Some bimetallic

(a)

15 Mechanical Stability of the bimetallic thermometer is affected by severe shock or vibrations the bimetallic

producing errors in indication. affect

element,

the thermal stability

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

ternal calibration

adjustment

of the ele-

not move freely but jumps from point to point with changes in temperature.

with ex-

(c) 20

16 Thermal Stability of the bimetallic thermometer is an inherent characteristic of the metals used. Certain bimetallic thermometers may be used up to lOOOoF, but prolonged exposure to such temstable continued

operation

The thermometer should be tapped lightly fore taking any reading. Treatment

of Dato.

be-

The observed temperature,

readings should be corrected for instrument errors using the calibration at temperatures

correction

values.

other than standardization

atures should be determined

in calibration;

Corrections ternpep

by linear interpolation.

Do not apply freezing point or other single point corrections to all points on the scale.

cannot be assured above

800%.

89


that pulsa-

(b) Do not use the thermometer if the pointer does

screws which rotate

may produce shifts

realizing

mental as motion of the entire unit?

the dial are available.

perature levels

is subjected,

tion of the fluid on the stem may be as detri-

with original accuracy provided the element is not deformed to the point where friction has been inThermometers

must be ob-

Decide where to place the bulb of the thermom-

instrument

does

ment and the thermometer may be reset to perform

troduced into the system.

thermometer the following

eter considering: Does the location selected minimize the shock and vibration to which the

thereby

This distortion

When using a bi-

19 Other Considerations. metallic served:

of 8 to 15 set are not uncommon.

not usually

De-

element and the shell.

Therefore, bimetallic thermometers may have longer response times than specified above. Time

which may distort

environments.

signs exist which are capable of operating under vibration beyond 100 Hz and 10 g acceleration.

and by the type of heat transfer material

constants

Heat

thermometer designs are suitable

for use in shock and vibration

element to the inside of the protective

used between the bimetallic

which is being measured.

transfer from or to the unimmersed portion of the


ASMEPERFORMANCETESTCODES ADVANTAGES

22

AND DISADVANTAGES

(a)

2 1 Advantages

may not be

evident. Easily

read.

(b) No ambient temperature Low maintenance.

[I?)

Liquids

(b) In the event of excessive

correction.

the thermometer is difficult

(f) No extensive

(Id Low

purchase

mechanical

23 The ‘

linkages.

following

are suggested

references:

“Bimetallic Thermometers,*’ SAMA Standard, PMC-Cl1962. “Master Test Code for Temperature Measurement Of AIEE No. 551, August 1950 Electrical Apparatus,”

cost.

(h) Calibration easily adjusted. (i) Not extremely position sensitive.

Catalog

Low weight.

T-B-K,

Weston Electrical


pointer vibrations, to read.

REFERENCES

and gases are not required in the sens-

ing elements.

(i)

due to shock or vibration

Damage


Instrument

Corporation

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

(a)

Disadvantages

CHAPTER 9, ,CALIBRATlON OF INSTRUMENTS

GENERAL: Scope ..............................................................................

Par. 1

TEMPERATURE SCALES .............................................. Thermodynamic Temperature Scale ............................ Ideal Gas Scale ............................................................ International Practical Temperature Scale ................ Fixed Points .............................................................. Interpolation Means ..................................................

2 4 6 7 9 10

METHODS OF CALIBRATION: Fixed Points .................................................................. 13 Comparison With Primary Standards .......................... 14 Primary StandardsStandard Platinum Resistance Thermometer .......... 16 Standard Platinum-Rhodium Versus Platinum Thermocouple .......................................................... 25 Optical Pyrometer ...................................................... 31 Comparison With Secondary Standards ...................... 35 Secondary StandardsLiquid-in-Glass Thermometers ................................ 38 Base Metal Thermocouples ...................................... 39 Optical Pyrometers .................................................... 41 APPARATUS .................................................................... ComparatorsComparators-Metal Block.. ........................................ Comparators-Bath Type ............................................ Mueller Bridge .............................................................. Laboratory Standard Potentiometers ..........................

43 44 47 50

RADIATION

52

THERMOMETERS

....................................

RESISTANCE THERMOMETERS: Comparison With Standard Platinum Resistance Thermometer ...............................................................

LIQUID-IN-GLASS THERMOMETERS: GeneralGeneral Purpose Liquid-in-Glass Thermometers .a.. 143 General Purpose Partial immersion Thetmometers .................................................................. 144 Special Use Partial Immersion Thermometers ........ 145 Visual Inspection: General ........................................................................ 146 Gas Bubbles .............................................................. 147 Globules of Liquid ...................................................... 151 Foreign Matter ............................................................ I52 Glass Faults .............................................................. 154 Test for Permanency of Pigment .............................. 155 Test for Permanency of Range .................................... 156 Calibration: General Considerations ............................................ 158 Calibration at Ice Point ............................................159 Calibration at Other Fixed Points ............................ 164 Calibration at Temperatures Other Than Fixed Points ........................................................................ 165 Checking for Changes In Bulb Volume ...................... 172 Treatment of Data ........................................................ 173

42

THERMOCOUPLE THERMOMETERS .......................... 57 General ConsiderationsGeneral Methods ........................................................ 58 Working Standards ........................................ -. ........... 59 Resistance Thermometers ........................................ 60 Liquid-in-Glass Thermometers ................................ 61 Types E and T Thermocouples ................................ 62 Types R and S Thermocou les ................................ 63 High Temperature Standar Bs .................................... 64 Annealing ................................................ .._ ................ Measurement of Emf .................................................. 6695 Homogeneity ................................................................ 70 General Calibration Methods .................................... 75 Calibration Uncertainties .......................................... 80 Uncertainties Using Fixed Points .......................... 82 Uncertainties Using Comparison Methods .............. 84 Freezing Points .......... _. ............................................. 87

FILLED OPTICAL

SYSTEM THERMOMETERS PYROMETERS

BIMETALLIC REFERENCES BIBLIOGRAPHY

............................

..............................................

THERMOMETERS

..................................

177 184 191

................................................................

199

. ............................................................

200


140


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

Melting Points ............................................................ 89 Calibration Using Comparison Methods .................. 90 Laborato Furnaces .................................................. 91 Platinum- f‘; hodium Versus Platinum Thermocouples ........................................................ 92 Base-Metal Thermocouples ...................................... 103 Stirred Liquid Baths .................................................. 104 Fixed Installations .................................................... 107 Interpolation Methods ................................................ 113 ................................ 119 Reference-Junction Corrections C;llibtr;ti; of Thermocouple Materials 123 .. .... .......... ............... ............................................................. 129 I31 Plitinum-Rhodium Alloy ............................................ Base-Metal Thermocouple Materials ...................... 134 .............................. Reference-Junction Corrections 135 137 Accuracies Obtainable ................................................

CONTENTS

AS\lE PERFORVANCE

TEST CODES Reversible

GENERAL

Scope 1 The purpose of this chapter

formation which will tion of instruments work.

It includes

is to present in-

industrial

and are therefore

test conditions,

for general everyday

used in Performance apparatus

TEMPERATURE

Test Code

International Through

Practical

SCALES

when

one body loses heat to another,

If

with its subdivisions

it is necessary

units,

just as it is

or the yard with its subdivisions,

This

scale was adopted at that time

adopted by the Ninth General

on Weights

Temperature

the foot and the

inch.

of the text of the International

Scale wa? suggested

by the Eleventh

General

1960 the International

The ideal temperature scale.

scale is known as the

Kelvin

scale to be such that “the

Measures Practical

Scale

has designed

absolute

values of two

revision

independent

scale defined

of the physical

in a reversible

on this

of the scale of 1948 but merely a revision

In October 1968, the International

as recommended

on Thermometry.

of any spe-

on

adopted eight major

changes in the international scale,

Committee

as empowered by the Thir-

teenth General Conference,

empirical

temperature

by the Advisory

Committee

The changes incorporated

in

IPTS - 68 are:

cific substance.

(a) Ideal

Gas Scale

6 Theory shows that the thermodynamic identical

values of temperature

Weights and Measures,

in this manner is

properties

Also in

on Weights and

of its text.

thermodynamic engine working with a source and refrigerator at the higher and lower temperature, respectively.” 5 The temperature

Committee

scale are the same as in 1948, this scale is not a

are to one another in the proportion of

the heat taken in to the heat rejected

and adopted in

Conference.

approved a new name “International Temperature Scale of 1948.” Inasmuch

as the numerical

this

Conference

in 1948.

A further refinement

and millimeter,

Temperature

and in 1948 the

was

l%O

Thermodynamic

to the Seventh General

perature Scale of 1927, and this revision and Measures

length to have the meter

of centimeter

Kingdom, and in 1927,

Committee

Advisory Committee on Thermometry and Calorimetry proposed revisions in the International Tem-

the first is at a

to have a scale with appropriate

have been held

of Germany, the

but further study was continued

higher temperature. 3 In order to measure temperature

States and the United

ature Scale.

there

is no thermal (heat) flow from one to the other.

laboratories

Scale

Conference on Weights and Measures recommended what was then known as the International Temper-

is a measure of thermal potential.

in measuring

Temperature

discussions

the years,

the International

Two bodies are at the same temperature

temperatures

not

use.

required and methods

United

4

and

let alone under

guide the user in the calibra-

7

thermodynamic

to construct

to construct

conditions

among the national

2 Temperature

are impossible

are difficult

use under ideal laboratory suitable

to be used.

necessary

heat engines

and gas thermometers

with that defined

The name kelvin designate

scale is

and the symbol K are taken to

the unit of thermodynamic

tempera-

ture .

(b) All values assigned

by the ideal gas equa-

are changed (except

tion of state

to the defining

of water) to conform as closely

PV=RT

the corresponding

where

(c)

fixed points

for that of the triple point

thermodynamic

as possible

to

temperatures.

The lower limit of Range 1 is changed from the

P = absolute pressure

boiling point of oxygen to the triple point of hy-

V = specific

drogen.

volume

R = gas constant T = absolute temperature

(d)

The standard instrument and 2 is the platinum

92 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



to be used in Ranges 1

resistance

thermometer

INSTRUMENTS

(e)

R,,,/R,,

tight ened from 1.3920 to 1.3925.

Range 1 is now divided Part 1 extends

las establishing

into four parts:

from the triple point to the boil-

points.

Part 2 extends from the boiling point of hydrogen from the triple point to the boil-

given in Table 10

Dusen equation

is no longer used, but interpolation

The reference

Interpolation

assigned

11

)’

as detailed

It should be understood

and their units are arbitrary

ratio represented

to them are

The means available

Means.

1,I( (9.1)

A. (In R/,,

resistance

WREF is defined

values

except where

fixed points and

9.1.

ments for interpolation

i=j’

fixed

is a document not a device.

by

in Table

9.2.

that temperature considerations.

IPTS-68

While the JPTS-68 is

scale having 180 units or degrees between

these

two points is commonly used in engineering

and A

WMis the measured resistance W is a deviation defined by a

nominal interpolation

equation,

RT;‘Ro,

ratio, specific

tice.

The relationship

given in Chapter

poly-

temperatures;

where t’ is temperature

12

A precise

Points

method of calibration

is that of de-

termining the reading of an instrument

i.e.,

(9.2)

At by the Callendar

tion; and At is a correction

OF CALIBRATION

Fixed

more of the defining

t 68= t’+

is

1.

METHODS

In Range 2, the Callendar equation is modified by a correction term so that interpolated values of temperature will conform more closely with thermodynamic

prac-

between these two scales

one being given

for each part of Range 1.

k)

scales

the basic scale having 100 units or degrees between the ice point and the steam point, the Fahrenheit

by

W~F=WM-AW where

for

interpolating temperatures led to the division of the scale into four parts, using three different instru-

is by a new

function

273.15 +y

by

each of which is under a

The defining

the exact numerical

In Range 1, the Callendar-Van

=

states,

noted differently.

ing point of oxygen.

T,,

to the defining

pressure of 1 standard atmosphere,

Part 4 extends from the boiling point of oxygen to the freezing point of water.

reference

calibrated

fixed points are defined by speci-

fied equilibrium

to the triple point of oxygen.

(0

These

between temperature

of instruments

means of values assigned

ing point of hydrogen.

Part 3 extends

the relation

and the indications

in Table

9.1.

TABLE

equa-

at one or

fixed points which are listed

9.1

SUMMARY

OF FIXED-POINT

VALUES

term given by

I PTS-68 At = 0.045 (&)

(Go

Defining

-1)

Fixed-Points OC

(4&

-1>

(6&

-1)

-259.34

13.81

-256.108

17.042

b.p. hydrogen

-252.87

20.28

b.p. neon

-246.048

27.102

t.p. oxygen

-218.789

54.36 1

b.p. oxygen

-182.962

90.188

t.p. hydrogen

(h) In Range 4, the second radiation is changed

0.014388

constant,

from 0.01438 meter kelvins

b.p.

C,,

to

meter kelvins.

8 By design,

JPTS-68 has been chosen in such a

way that temperatures

K

(9.3)

measured on it closely

ap-

hydrogen, 25/76

atm

f.p. water

0.0

273.15

t.p. water

+O.Ol

273.16

proximate thermodynamic temperatures; i.e. differ ences are within the limits of the present accuracy

b.p. water

100

373.15

of measurement.

f.p. zinc

419.58

692.73

9

Fixed

Points.

IPTS-68

is based on 11 repro-

ducible temperatures (defining fixed points), to which numericalvalues are assigned, and on formu-

b.p. sulfur

444.674

f.p. silver

916.93

1235.08

f.p. gold

1064.43

1337.58

93



7 17.824

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

with

AND APPARATUS

ASMEPERFORMANCETESTCODES TABLE

9.2

INTERPOLATION

as practicable

MEANS

use.

and will remain so during continued

The most suitable

platinum

wire is that drawn

from a fused ingot, not from forged sponge.

Range 1

18

Temperature

limits

13.81 to 273.15K

Interpolating

relation

Reference

interpolating

instruments

Pk;kztrnrresistance

function

Eq. (9.1)

Satisfactory

thermometers

have been made with

wire as small as 0.05 mm and as large as 0.5 mm in

ther-

diameter,

a short portion of each lead adjacent

the resistor Range 2

being made of platinum

to

and continuing

with gold wire through the region of temperature

Temperature

limits

0 to 630.74?

Interpolating

relation

Modified

Interpolating

instruments

Platinum resistance mometer

Temperature

limits

630.74

Interpolating

relation

Parabola

Interpolating

instruments

Platinum 10% rodium vs. platinum thermocouple

gradient.

Callendar

Eq. (9.2)

The completed

ther-

est temperature to 1064.43’C

protecting

Above

relation

Planck’s

Interpolating

instruments

Optical

A useful

19

resistor

is usually

The tube filled

some oxygen.

criterion,

which serves as a safe-

guard against inferior construction of the completed thermometer and against errors in the calibrations

1064.43’C

Interpolating

at which it is to be used.

the completed

with gas containing

Range 4 limits

of the thermometer

in air at a temperature not lower than about 840??, or if it is to be used at temperatures above 8400F, at a temperature higher than the high-

Range 3

Temperature

resistor

is annealed

at the fixed points,

law

is that

pyrometer

(Rs - R,) / (RB- RF) 13 Calibration

at fixed

points requires

equipment and painstaking resistance

thermometers

platinum-rhodium

techniques.

specialized

thermocouples

are calibrated

fixed points for use as primary standards.

vices

It is good practice calibrated

14

at

Primary

Standards

With

The common method of calibrating

where

an instru-

with those of

primary or secondary

at temperatures

standards

.parators described 15

The primary standards

sistance platinum

thermometer,

of resistance

thermocouple

at a reference

criterion

of the thermometer

20 Also,

vs.

the ratio

optical

the platinum

16 Standard

mercially 17

Platinum

Resistance

temperature

platinum resistance thermometers are comavailable from a number of sources.

They are so designed

wire of the platinum

resistor

and coustructed

is defined

that the *All

temperatures

in OC.

94 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


is used

In this range the

by the formula:

R, = R, (1 + At

is as nearly strain-free

thermometer

in the range 32oF to the freezing

point of antimony at 1167.3or;. Thermometer.

and the

in service.

shall be of such purity that

21 The standard resistance Standards

before and

is also a valuable

R&R~ is greater than 1.3925.

for comparison Primary

The constancy such as the

of the adequacy of the annealing

reliability

pyrometer.

Standard

point,

triple point of water (or the ice point),

are the platinum re-

and the narrow-band

if

at the oxygen point,

after use at other temperatures,

43 to 46, inclusive.

the platinum-rhodium

Ro2, the resistance

should be between 6.143 and 6.144.

which are produced in com-

in Pars.

Similarly,

for use in the range

(R, - R, ) / (R, - R,) 2

ment is to compare its readings other than fixed points,

at ice point at steam point

the thermometer is calibrated below 32”F, the ratio,

Bureau of Standqualified.

at sulfur point

should be between 4.2165 and 4.2180.

to have these two de-

similarly

= resistance

RF = resistance RB = resistance

These

of other instru-

by the National

ards or other laboratory Comparison

Rs

Standard

and standard platinum-

are then used for the calibration ments.

where


+ Btz)*

INSTRUMENTS

AND APPASATUS

where Rr is the resistance, at temperature t, of the platinum resistor between the branch points formed

than 0.65 mm in diameter.

by the junctions of the current and potential leads of a standard resistance thermometer. The constant,

calibration,

mounted so as to avoid all mechanical

R0 is the resistance at 0% (32°F), and the constants, A and B, are to be determined from measured values of

Before

the

wires of the couple are annealed in air for an hour at about 2000°F. The wire of the thermocouple is in the region where steep temperature likely

constraints gradients

are

to occur.

Rt at the steam and zinc points.

22 From the oxygen point perature, 23

to 0°C (32oF),

29 The primary standard platinum-rhodium versus platinum thermocouple is used in the calibration of

the temp-

t, is defined by Eq. (9.1).

other thermocouples

For easier use in temperature

the interpolation

formulae,

to the freezing

calibration

measurement, (9.2)

thermocouples

should be calibrated

resistance

periodically

Bureau of Standards or equally 25

Standard

at 1167.34:

by the National

Platinum-Versus

agency.

Platinum

From the freezing

Thermocouple.

Rhodium

to the gold point at 1947.97%‘,

versus platinum thermocouple

31 Optical

the

thermometer

thermocouple,

tional Practical 26

Temperature

E = a+ bt +

at the freezing

and platinum a,

b, and c,

The platinum

10 percent

wire of the standard thermoand of such purity that the ratio

rhodium by weight.

= 1.4388 cm K h = a wavelength

c2

e = base of Naperian

such that

logarithm.

which defines

IPTS-68

a narrow-band

optical

above the gold point, pyrometer

must be calibrated

at a number of points using this formula, EAcr = 10,300

spec-

point of

antimony (1167.3°F), silver, or gold, the completed thermocouple has electromotive forces, in microvolts,

_ of the visible

trum, cm

When one junction

and the other at the freezing

ATI-1

of a blackbody at temperature T, and at the gold point TAa, respectively.

Rz120~/R,20~is greater than 1.3920. The alloy wire shall consist nominally of 90 percent platinum and is at 32oF,

1

length interval at wavelength, A, emitted per unit time by unit area

and gold points. 27

exp (c2/

-

I, and IA,, = the radiant energies per unit wave-

from the measured values of E

point of antimony and at the silver

couple is annealed

exp (c2 / h TA”)

in which

is at 32oF and the other is

t. The constants,

are to be calculated

Jt = JAU

force of a standard

of platinum-rhodium

alloy when one junction

formula leads to

ctz,

E is the electromotive

at the temperature

that of the InternaScale.

t, is defined by the formula:

The temperature,

thermocouple

there is

which when

to a set of specifications,

approaching

32 While Planck’s

where

the standard plati-

pyrometer,

such as an optical according

by

quali-

and the standard plati-

no device

has a calibration

is used.

periodically or equally

num-rhodium versus platinum constructed

10 percent

Unlike

Pyrometer.

num resistance

point of antimony

standard platinum 90 percent-rhodium

versus platinum

should be calibrated

the National Bureau of Standards fied laboratory.

thermometers

qualified

and

below a tempera-

30 Standard platinum-rhodium

and (9.3). 24 Standard platinum

pyrometers

ture of 2750°F.

for the range 0°C (32 F)

point of antimony are in Eqs.

and also in the checking

of optical

produce a means for extrapolation

+ 5OpV

in order to

above the gold

point. E,,

- EAg = 118.3 + 0.158 (EAT - 10,300)

EA,, - Esb = 476.6 + 0.631 (RA,, - 10,300)

2 41~V

33

Calibration

on a blackbody

f 8pV

is done by sighting maintained

ing sectors of known transmission 28 Satisfactory

standard thermocouples

have been

are then interposed

made of wires not less than 0.35 mm and not more

pyrometer,

Rotat-

characteristics

between the blackbody

and calibration

and the

at points higher than the

95 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


the pyrometer

at the gold point.


ASME PERFORMANCE gold point is calculated energy received

based on the ratio of the

TEST

ASTM Thermometer Number

through the sector disks to that re-

ceived at the Kold point. 34 Standard optical brated periodically ards or equally

qualified

Comparison 35

pyrometers

Secondary standards

calibration

glass thermometers,

accuracy

liquid-in-

thermometers,

base-metal

pyrometers,

for the

particularly

bimetallic


Standards

are satisfactory

of many devices,

and optical

Bureau of Stand-

laboratory.

With Secondary

thermometers, pyrometers.

standards

base-metal

in Pars.

pyrometers

by compar-

(b)

as described

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

or thermocouple

38

Liquid-in-Glass

secondary

as described

for testing

eters to be tested will

gOVem

For thermometers

5 deg divisions, cable corrections. thermometers

is used.

The change

complete

are

of the emergent column for indications

In such cases,

of the ther-

standardization

is

carried out for the emergent column prevailing

(c)

Special use partial

ways.

with the standardization

immersion thermometers may

One method involves

thermometers

into two

mersion standards.

purpose

comparison

at total immersion

of the

with total im-

The number of degrees of

scale which will be in the emergent column when in actual use shall then be measured. From these data the corrections ified

calculated.

the choice of

the liquid

thermom-

may then be

In the case of organic liquid-filled

thermometers

graduated in 1, 2 or

under the spec-

emergent column temperatures the coefficient

should be obtained

of expansion

of

by experiment

or

from the manufacturer in order to perform these computations. This method has the advantage

when used with appligraduated

set is recommended W.*

that the standard may be selected greater sensitivity tested,

in brackets indicate References at end of chapter, thus [I]. *Numbers

ment.

to have

than the thermometer

thus increasing

the accuracy

being

of measure-

A second method, which is the one best

96


with means for

equipment being employed.

in Pars.

general

For fractionally

the following

is provided

to their own specifications,

temperatures

when used as

a set of well-made

eters will be adequate

for total im-

General purpose partial immersion thermometers, as commonly listed in manufacturers’ catalogs

usually

In the case of general purpose total immersion thermometers, the sensitivity of the thermomstandard.

set is standardized

at the ice point, which should be done

mometers.

total immersion thermometers and those intended for checking partial immersion thermometers. (a)

380

have their emergent mercury column or stem tempe;atures specified. Cognizance of those specified temperatures may be taken in various

may be classified

groups, those intended

1.0

the various temperature

Standards

Thermometers,

standards,

563 to 761

of the temperatures

readings

184 to 190, inclusive. Secondary

70 F

normally bought and sold without specification

other than fixed points,

pyrometer

380 380

according

are compared with those of the secondary

standard optical

380

0.5 1.0

all readings. Normally, a single standardization is adequate.

which are produced in comparators described in Pars. 43 to 46, inclusive. Test optical pyrometer

readings

0.5

293 to 401 383 to 581

in ice point reading should then be applied to

are then compared with those of the secondary

standard at temperatures

203 to 311

68 F 69 F

each time the thermometer

resistance

184 to 190, inclusive.

37 The test thermometer

67 F

F F

checking

thermometers or standard platinum-rhodium versus platinum thermocouples at temperatures which are produced in comparators described in Pars. 43 to 46, inclusive. Optical pyrometers are compared with primary standard optical

380 380 380 380 380

to to to to to

eters in this series

and optical

The first two are calibrated

ing; them with primary standard platinum

0.2 0.2 0.2 0.2 0.2

-36 18 77 122 167

mersion. With the exception of the first two, each thermometer is provided with an auxiliary scale including 32??, thus each of the thermom-

are liquid-in-glass

thermocouples,

t35 89 131 176 221

F

F F

The foregoing

degree of

is not required.

36 The secondary

Length (mm)

Range, OF

filled

thermocouples,

where the highest

Divisions, Deg F

62 63 64 65 66

should be cali-

by the National

CODES


INSTRUMENTS suited to large-quantity

testing,

involves

AND APPARATUS

com-

Optical

pyrometers

can be standardized

parison of the thermometers with standards similar in all details in construction above the

other optical

pyrometers

that have been calibrated

immersion point, but differing

calibrate

completely

Bureau of Standards, definitely

when 42 American

may then be employed in-

for standardization

purposes

ard E77-70

if peri-

Individual

ary standards

thermocouple

elements

for the calibration

43

platinum-

Block.

can also be

OVEN

test is illustrated

thermo-

in Fig.

9.2.

Type. Comparators for use 44 Comparators-Bath in standardization are of two types, fixed-point or

FOR PERMANENCY

OF PIGMENT

TESl


An oven suitable

for foreign matter in the bore and for

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

9.1

Stand-

in this sec-

permanency of pigment is illustrated in Fig. 9.1. An air bath suitable for the permanency of range

and used as secondof similar

Comparators-Metal

for testing

couple wire comparison.

FIG.

Materials

Comparators

thermocouples.

against platinum

Society for Testing

is the basis for the material

tion.

Bureau of Standards or by the user against

40

and used to

by comparison.

APPARATUS

39 Base Metal Thermocouples may be standardized for use as secondary standards by the National

standardized

pyrometers

as by the National

odic ice point checks are made.

rhodium versus platinum

Bureau of Standards

other optical

against

an auxil-

Such thermometers,

standardized,

by the National

below the immer-

sion point to the extent of including iary ice point scale.

41


ASME PERFORMANCE

ll ll ContactTtmnometef

TEST

CODES

various

capacities.

Other types of voltage regu-

lators,

such as the so-called

used.

While generally

induction

type, may be

more expensive,

they are

capable of finer adiustment. 46

In all test baths a properly

other suitable platinum

provision

resistance

located

well or

should be made for using a

thermometer

as the ultimate

primary standard.

FIG.

9.2

AIR BATH RANGE

variable.

Certain

FOR PERMANENCY

general requirements

met by all such comparators equipment.

All readings

eters should be observed, suitable

OF

TEST

for observing

should be

and their accessory

of liquid-in-glass

thermom-

using an optical

vertical

motions.

device

Focusing

range should start from not more than 20 cm, magnification

should be about 5 to 10 diameters,

field about 1.5 cm.

The eyepiece

and the

should be pro-

vided with 90 deg cross hairs and may include erecting

prism.

proximately

Vertical

movement should total ap-

30 cm and may be obtained

rough and delicate

adjustment.

ment is illustrated

in Fig.

available

both with

A suitable

instru-

9.3; component parts are

FI G. 9.4

APPARATUS

commercially. of accurate

tinuous control to meet the requirement rising temperature

con-

of a very

at the test point.

For best

(a)

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

results an alternating current power supply is recommended in conjunction with variable transformers. Such transformers

FOR STANDARDIZATION

OF ICE POINT

45 Heat input should be capable slowly

an

are available

commercially

Comparators,

Fixed Point.

The most common

and also most useful of the fixed point type comparators is the ice point apparatus. Fig. 9.4 illustrates a typical setup, consisting of a

in

Dewar flask, telescope

a thermometer

holder,

of 10X magnification,

a viewing

and the neces-

sary supports and siphon tube for withdrawing excess

water.

is a desirable

A suitable adjunct.

ice-shaving

For readings

tenth of a division

on liquid-in-glass

eters,

telescope

the viewing

observing

the temperature

is satisfactory

machine to onethermom-

is necessary,

but

with the unaided eye

provided poorer precision

or ac-

curacy is satisfactory. Other fixed-point triple

comparators

point of water,

and at secondary

for use at the

steam point,

sulfur point,

points such as the benzoic

acid points, are described in publications of the National Bureau of Standards and the Massachusetts FIG.

9.3

MAGNIFIER

FOR READING

Institute

THERMOMETER

of Technology.



INSTRUMENTS

(6)

Variable

Comparators,

ardization

able temperature-type Figs.

Temperature.

at other than fixed points, comparators

AND APPARATUS

For stand-

mation point of solid

the vari-

nitrogen is used as the cooling

are used.

9.5, 9.6, 9.7, 9.8, and 9.9, illustrate

an isolated vari-

carbon dioxide,

bath filled

hydrocarbon

with some low-boiling A suitable

such as isopentane.

bath for this purpose is illustrated

-256

and is described

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

ous types of such units for use in the range to t1000~.

consist provided

of a well-stirred, with suitable

the temperature creasing. viewing one-tenth

All of these units basically

either

insulated controls

liquid bath

telescope

is necessary

of a scale division

thermometers.

The liquids

characteristics

low viscosity,

a

for precision

of

in Fig.

re-

It must extend above the enough so that nitrogen spilling

covered with isopentane nitrogen.

temperature

A loose fitting

condensation

and convection.

the desired temperature,

-103OF. For comparison in the range from -256 to -103%, which is below the subli-

evacuated

with liquid

When the

iluminum Bloq inner Evogg;ble-

Unritvered Outer

Evocuoted Florh--r Silvrrrd Heating Coil * Clara Wool Xurhion Met01 Flask ~-Holder

FOR TEMPERATURE

RANGE

FROM -256

TO -103OF


above

the inner flask is

by the pump to stabilize

Platinum -Rosistoncr Thermometer


filled

cardboard shield

system has cooled to a few degrees

(1) Comparators for the range from -2.56 to

COMPARATOR

and the space be-

with holes is placed over the top to reduce

ranges in which they are to be used.

FIG. 9.5

into the iso-

The aluminum block is completely

tween the two Dewar flasks

and freedom from other offensive

9.5 is positioned

by three cork wedges

nitrogen container pentane.

nonflammability,

in the various

in Fig.9.5

in ASTM Method

strung on a wire.

may be added without

used in the comthe following

flask

in the larger container

on liquid-in-glass

parators are chosen to fulfill quirements: nontoxicity,

The evacuable

in-

As with the ice point equipment,

in detail

E77-70.

for maintaining

constant or uniformly

liquid

medium for

the tem-

ASME PERFORMANCE

TEST

CODES (b) Liquid

air or liquid oxygen or a mixture

of them, should never be used as the refrigerant. (cl At all t imes the level of the pentane should be slightly above the surface of the aluminum block.

Whenever fresh

chilled pentane is added, care must be taken not to spill

any on the heating

coil. (d) The heating coil should be attached to the variable thoroughly (2) Comparators

by means of

transformer insulated

leads.

for the Range from -112

to

+41’F. For temperature testing in the range from -112 to t 41°F, as indicated in Fig. 9.6, two baths are recommended-one elements

requiring

for primary

a long depth of immersion

and the other for short immersion. case the bath shall consist vacuum flask,

the walls

In either

of a Dewar-type

of which may or may

not be silvered as required. Normally the unsilvered is preferred. These baths are commercially detail

available

and are described

in

in ASTM E’77-70.

The bath medium may be either alcohol, COMPARATOR FROM -112

FOR TEMPERATURE

RANGE

light hydrocarbon,

TO t4fF

50.6 percent perature equilibrium testing

or other organic solvent,

such as the carbon tetrachloride-chloroform eutectic (49.4 percent carbon tetrachloride, in the test tube.

in the range of -250°F,

(When

chloroform)

If a water miscible

this evacua-

or trichlorethylene.

solvent

is used, the

water content must be not more than 5 percent.

tion is not necessary.) At the desired test temperature, heat transfer is exactly bal-

In use the proper bath is partly filled

anced by means of the heating

the bath medium.

be necessary raising

to agitate

and lowering

It will

the isopentane

the thermometer,

any vigorous stirring will

coil.

cent.

by

to prevent bubbling over.

but

of proper equilibrium.

a precooling

Cer-

to

bath, is similarly

chilled

in a

second container. When the two are near the proper temperature, the testing bath is filled and brought to temperature. The thennom-

tain safety precautions must be carefully observed. Some of the more important are as follows:

eters to be tested and the standard may be precooled

(a) Because liquid nitrogen has a lower abquantities

phere moisture, within the walls Therefore,

condense

of the evacuable

(3) Comparators for the Range from 41 to ZOOOF.

being vented to the at-

In the range from 41 to 2OOoF a comparator

mosphere or under vacuum and should

as illustrated

in Fig.

always

may be used.

It consists

be vented

through a drying tube.

100


before test-

shorten the time required.

flask.

the flask should never be

warmed without

temperature

ing. If large batches of thermometers are to be tested, this preCOOhg will materially

of liquid air or atmos-

or both, will

in the standby bath to approxi-

mately the desired

solute boilinK point than liquid air, substantial

sufficient

fill the bath and an extra amount to serve as

generate enough heat to make difficult

the maintaining

At the same time

another portion of the liquid,

at low temperatures

with

Dry ice is added slowly

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

FIG. 9.6


9.7, or its equivalent, of a heavy wall

INSTRUMENTS

AND APPARATIJS

Top Plate showing Position of Holes ond showing onr Ring and Bronrr 8alls in Ploco

Ring Segment showing Boll Groove ond Method of Holding to Plato 2 Required - Eros8

Spring Clip 26 Req. Phos. Brontr

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

Section of lop Plato showing Position of Groovr on under ridr

FIG. 9.7

Pyrex

jar suitably

with heaters, equipment. provided

COMPARATOR

supported

cooling

coil,

A commercially

with controls

or very slowly the testing

rising

FOR TEMPERATURE

and equipped available

temperature

FROM 41 to 200OF

illustrated

and stirring

to maintain

RANGE

in Figs.

9.8(a)

or 9.8(b) may be

used. bath

Suitable high-flash

constant

point oils should be used.

At the higher temperatures

throughout

great care must

be taken to avoid dangerous flash fires which

range and with a top cover hav-

may occur, particularly

on removal of ther-

ing a large opening is shown in Fig. 9.7. For liquid-in-glass thermometer testing an

mometers or thermometer

insert equipped with two rotary holders made as shown in Fig. 9.7 is provided. Each of

er as well as adequate fire protection ment of the carbon dioxide

these holders

provided.

is a flat plate resting

number of ball bearings may easily

circulation

on a

so that the plate

into view.

Positive

heat dis-

of cooling and heating

tube arrangement

by a unique assembly coils

commercial

in a cylindrical

in Fig.

housing which also surrounds the stirring propeller.

The bath should be filled

distilled

water.

An alternative

lustrated

in Fig. 9.8(a).

holdequip-

9.9.

A type having a

immersed

heat treating

in the pot of a

bath is illustrated

Welded or riveted

pots should

under no circumstances be used. External electrical heating is commonly practiced,

with

design is il-

although gas fired units may be purchased and have been used successfully. Care should be exercised,

(4) Comparators for the Range from 190 to TOOOF. In the range 190 to 7004: a comparator as

752oF, to avoid bringing

particularly

above

any organic matter

101


A solid

type, should be

to 1000°F is a salt bath.

to all parts of the bath and thorough are obtained

holder.

the thermometer

(5) Comparators for the Range from 450 to 1000°F. A satisfactory comparator for the range 450

be spun to bring any particular

thermometer tribution

cover plate to replace


ASME PERFORMANCE

TEST

CODES

_.__ Coil -

Tubr boorin-

Support

FIG.

9.8

ALTERNATIVE

or low boiling

liquid,

DESIGNS

(b) StirredLiquidBathwithh ‘ o TuberConnectedst TopandBottom

lb0 caukl

OF COMPARATORS

FOR TEMPERATURE

FROM 190 to 700OF

avoided by using steel wells

such as water, in con-

tact with the molten salt,

RANGES

immersed in

the molten salt into which the thermometers

since dangerous

or thermocouples

fires or explosions may occur. The bath should be heated up slowly at the start to avoid the formation of pockets of molten salt which have a tendency to blow out the solid mass with disastrous results. The

are placed.

Thin walled

steel tubes closed at one end are suitable for this purpose. Comparators with molten tin as the bath liquid have been used successfully

bath should be covered while heating until approximately 25 percent of the salt has

temperature

range.

Details

in this

of design,

con-

struction, and use may be obtained from the National Bureau of Standards,

melted. A salt comprised of 50 percent sodium nitrate and 50 percent sodium nitrite

is sug-

Mueller

Rested. At the lower temperatures or thermocouples in the salt,

47

the thermometers

may be immersed directly

but at the higher temperatures

attack of the glass may occur.

This

When calibrating

comparison

can be

resistance

thermometers

with a standard platinum

thermometer,

a bridge of the Mueller

monly used.

The circuit

102


Bridge


by

resistance type is com-

is shown in simplified

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

(8) stirrod Liqldd~uritb *

INSTRUMENTSAND APPARATUS thermometer arm, which is equivalent sistance

into the measuring arm.

to putting re-

Contact

resist-

ances in these decades are in series with the high resistance ratio arms Q andQ, and their variations are, therefore, negligible. Contact resistance in the three decades D, E, and F (0.01, 0.0001 ohm decades)

have a negligible

cause the resistance

variation

0.001,

and

effect,

is secured

be-

by chang-

inK a comparatively high resistance in series with the contact, shunting a resistor of low value. The bridge as shown is adapted for use with three-lead thermometers, but to make full use of its available accuracy a four-lead thermometer with a commutator for connecting

its leads three at a time to the bridge

should be used.

The commutation

inates errors resulting lead resistances.

from slight

of the leads eliminequalities

With these precautions

of

the re-

sistance measurements can be made within a limit of error of _+O.OOOlohm, or +0.02 percent of the measured resistance, SALT BATH COMPARATOR FOR TEMPERATURE RANGE FROM 450 TO 1000°F

whichever

thermometer

the limit

is larger.

With a

of error is depend-

ent on the care with which the A and C leads are

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

FIG. 9.9

three-lead

adjusted to equality, but is approximately +O.OOl ohm or to.03 percent of the measured resistance, form in Fig.

9.10; Q and Q, are equal ratio arms of

from 250 to 1000 ohms each, adjustable by means of the sliding cuit on the slide-wire and B are permanently ometer circuit. resistance

S.

The resistance

connected

decades

ically

A

into the measuring

ing the setting of B takes resistance

of A inserts arm.

49

Increas-

FIG. 9.10

DIAGRAM

qualified

“Notes

Bureau of Standards January 1, 1949”

BRIDGE

103


or

publication

to Sup.plement Resistance

mometer Certificates,

OF MUELLER

period-

Bureau of Standards

laboratory.

The National

entitled

out of the

bridges should be calibrated

by the National

equally

into the galvan-

the setting

is larger.

48 Mueller

contact of the battery cir-

Increasing

directly

whichever

to equality


Therincludes

de-

ASME PERFORMANCE tailed

instructions

of calibration

for checking

of Mueller

Laboratory 50

Laboratory

and thermocouple

of Testing

Potentiometers

are

ture Measurement”, Testing

General 58

52

Calibration

in measuring

B of this Supple-

of operation

and types of Labora-

temperatures

of a radiation

thermometer

apparent temperature

has been measured with a standardized which follows

the radiation

thermometer

53

It is not necessary

environment standard

that the source be a black

in its emissivity;

throughout its spectrum.

that is, have the

59

It is,

tion of thermocouples. the temperature

rotating

are simulated

sector disk between thermometer,

expected

24OOoF

by interposing

a

erence of the calibrating

temperature

measured by means of a standard radiation

60

thermom-

eter. To make this method of calibration valid, the two radiation thermometers being compared must follow

the same power law, because the spectral

distribution

of the radiation

is quite different

from that for a black body at the

55 In all cases the dimensions of the furnace opening should be such that the field-of-view reof the radiation

thermometer

are satis-

to 630.5’C

Liquid-in-Glass

upon National

Bureau of Standards

Circular

Thermometers.

This

type of -183OC

or even higher with

where instability

may require frequent ASTM liquid-in-glass

is based

ASTM El-71.

590,

of the bulb glass

calibration.

104 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


-253OC

In cases where an

may be used from approximately

to 400°C (752%?,

300°C (572v)

THERMOMETERS

in this section

(1167oF).

special types. Generally, the accuracy of these thermometers is less below -56OC (-69(F), where organic thermometric fluids are used, and above

equipment if requested.

57 Much of this material

The standard plati-

is the most accurate

for use from approximately

61

supply data for checking

THERMOCOUPLE

Thermometers.

thermometer

standard

thermometer

fied. will

or pref-

laboratory.

t-423%?

t-297’%?

56 Manufacturers

or in cases where more the convenience

uncertainty approaching 0.1% is necessary at temperatures below -56OC (-69OF) or above about 200°C (392T) there are few alternatives to the use of resistance thermometers as standards.

from a source at 2400°F.

measured apparent temperature.

quirements

Resistance

num resistance

whether a laboratory

liquid bath is used, the accuracy

of the calibration,

than one type will suffice,

the source and the

and the apparent

The choice will depend upon

range covered,

furnace or a stirred

54 The source is a furnace which is maintained for example

Any one of several types calibrated in terms of the IPTS,

Standards.

may be used as a working standard for the calibra-

in routine calibrations.

radiation

Working

of thermometers,

which has been found to be convenient

Lower temperatures

and the

can be brought to the same temperature.

this section.

to make use of a method of

at a constant temperature;

and a controlled

in which the thermocouple

but it should

same emissivity calibration

means of

Some of the more commonly used techniques for accomplishing such calibrations will be discussed in

be nonselective

permissible

number of known

so that, with some accepted

uring the emf of the thermocouple,

of which

radiation

under test.

valid,

The calibration of a thermoof the determination of its electo-

Methods.

process requires a standard thermometer to indicate temperatures on a standard scale, a means for meas-

consists

the same power law as

body to make this calibration

therefore,

for

interpolation, its emf will be known over the entire temperature range in which it is to be used. The

the emf which it produces when fo-

thermometer

Society

Considerations

motive force (emf) at a sufficient

THERMOMETERS

cused on a source,the

General

couple consists

available.

RADIATION

STP 470, American

They should be cali-

51 Refer to Chapter 3, Section tory Potentiometers

and Thermo6, 1958, and the

and Materials.

brated periodically by the National Bureau of Standards or other laboratory similarly qualified.

ment for principles

Thermocouples dated February

calibration methods appearing in Chapter 8, of the “Manual On The Use of Thermocouples In Tempera-

of all types of thermocouples

materials.

CODES

couple Materials,”

Potentiometer

High Precision

used in the calibration

“Methods

the self-consistency

bridges.

Standard

TEST


thermometers

Specifications are given in

for

INSTRUMENTS 62

Types

perature of -183’C tainable

Either

E and 1 Thermocouples.

types of thermocouples

AND APPARATUS

of these

electrically

may be used down to a tem-

(GXV°F)

or lower,

accuracy may be limited

but the at-

in air.

which should be

close together,

so that the tension

in the wires and

while hot are kept at a minimum.

of the emf measurements and the inhomogeneity of the wire at low temperatures. The stability of the

temperature

larger sizes

able type of optical

wires under the same conditions.

Twenty-four

wire is a useful compromise between stability

for this purpose.

larger wire.

upper limits

of

easily

are 425’C

The ordinary port-

is very satisfactory

on an object

thermocouples,

by lengthening

or using an objective

as small as but this is

the telescope

Types

is the most satisfactory

constant

up

about 12OO’C (2192%.

to

characteristics

whether annealing

Its use may

as to the optimum

and length of time at which such thermo-

couples should be annealed

work-

ing standard for use in the range from 630.5% (1167%

There are some questions

temperature

The Type S or

R and S Thermocouples.

Type R thermocouple

to produce the most

in later use, and as to

for more than a few minutes is

be extended down to room temperature if it is desired to use the same standard over a wide, range,

harmful or beneficial. strains

are relieved

but its sensitivity falls off appreciably as temperatures below 200°C (392T) are reached. Twenty-four

heating

at 1400 to 1500°C (2552 to 2732oF),

ards. High

Temperoture

The IPTS above

Standards.

1064.43OC (1946%) is defined in terms of ratios of radiant energy, the ratios usually being measured by means of an optical rometer,

pyrometer.

sighted on a blackbody

the calibration

furnace,

The optical cavity

therefore,

On the other hand, thermocouples, optical

pyrometer

standards.

scale,

The Type B thermocouple

to about 16OO’C (2912%‘).

Tungsten

can be used to higher temperatures, pyrometer

py-

wires are heated for several

hours before calibra-

tion and use.

The principal

objection

thermocouples

for a long time at high temperatures,

Rrain growth. temperatures

above 1064.43OC.

calibrated

to annealing

mechanically

on the

be used as

It has been found that annealing

rhenium alloys

68

but the optical

the wire very

for 1 hr at 1450% (2642°F).

It has not been demonstrated

conclusively

longed heating in air, although it is logical

considered

is it found advisable

pose that certain through oxidation,

sufficient,

and seldom

to further anneal the wire be-

new platinum-rhodium

num thermocouple ers is already

versus plati-

tice in many laboratories

accuracy

used to measure the emf.

of the instrument

to anneal all Types

whether new or previously

before attempting

an accurate

accomplished

calibration.

by heating

whose performance

accuracy-of

R and

highest

used, This

in most instances,

racy of the calibration

it has become regular prac-

S thermocouples, usually

rendered less detrimental. One of the factors in the of a thermocouple is the

available

wire as sold by some manufactur-

annealed,

an instrument

need not be limited

the emf measurements.

accuracy

is

is such that the accu-’

it is advisable

by the

For work of the

to use a potenti-

ometer of the type in which there are no slidewires

is

and in which all the settings are made by means of dial switches. However, for most work, in which an

the thermocouple

105 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


to sup-

can be driven off or,

69 Meosurement of Emf. accuracy of the calibration Fortunately,

Although

impurities

Such treat-

fore testing. 66

that

Types R and S thermocouples after contamination can be materially improved in homogeneity by pro-

65 Annealing. Practically all base-metal thermocouple wire is annealed or given a “stabilizing ment is generally

at

much above 1500°C (2732oF) produces

rapid changes in the emf and leaves

R and S thermocouples

is useful up

by the manufacturer.

is that

as a result of

weak mechanically. The National Bureau of Standards has adopted the procedure of annealing Types

is more commonly used.

heat treatment”

but it

aside from the changes in emf taking place,

built into

can themselves

during, the first few minutes of

the wires are weakened

can serve as a

working standard for all temperatures

Most of the mechanical

has been claimed that the changes in the thermal emf of a couple in later use will be smaller if the

gage wire is most commonly used for these stand-

64

tube

lens of shorter focal length.

for the Type E and 200°C (392oF) for Type T. 67

63

remedied

deter-

As commonly used, the magnifica-

the wires of noble-metal

(greater required depth of immersion) Recommended

pyrometer

tion is too low for sighting

of smaller wire and the greater thermal

conduction (797%‘)

gage

the lesser

The

of the wire is most conveniently

mined with an optical pyrometer.

of wire is greater than that of smaller

is sup-

two binding posts,

stretching

by the accuracy

The entire thermocouple

ported between


ASME PERFORMANCE accuracy of 5 /LV will

suffice,

eters of the laboratory Portable

slidewire

potentiom-

73

are also available. 70 Homogeneity. The emf developed by a thermocouple made from homogeneous wires will be a function

of the temperature

measuring

difference

and the reference

between

junction.

the wires are not homogeneous, extraneous

exists,

emf’s

will

and the output of the thermocouple factors

in addition

to the temperature

tween the two junctions. thermocouple

difference

The homogeneity

wire, therefore,

is an important

The resulting

emf at various depths

furnace.

Other similar

Tests

such as those described

the uncertainty

ally sufficiently

element

of a platinum-rhodium

various

parts of the wire to 600%

homogeneous

in chemical

Occasionally

compo-

inhomoge-

in the wires during tests or use.

ally is not necessary, thermocouples

therefore,

the emf developed

to examine new

for inhomogeneity,

before an accurate

thermoelectric

calibration

dent that corrections ticable

it or the resulting

for

no satis-

errors in the measure-

General

employed

in determining

two ends of the wire to a sensitive

type of thermocouple,

wire.

the length of the wire. may be detected

such as a

furnace,

gradual changes in the thermoelectric Inhomogeneity

as before.

of this nature

between

vanometer

gradient,

the emf determined

thermoelectric

properties

whatever

the reference

difference

junctions.

method of calibration

is used,

must be maintained

constant

junction

at some known temperature

the

must be stated as a necessary

and this temperature part of the calibra-

tion results.

so that

76

Thermocouple

calibrations

are required

with

various degrees of accuracy ranging from 0.1 to 5 or 10 deg C.

in the

For an accuracy of 0.1 deg, agreement

with the IPTS and methods of interpolating

of the wire at these two

the calibration

points.

portance,

between

points become problems of prime im-

but for an accuracy

106 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


at a specified

and reference

to a

with the gal-

is a measure of the difference

ther-

range, accuracy

depends upon the temperature

its measuring

Therefore,

heated furnace,

two points on the wire 20 in. apart are in the temperature

temperature

power along

gradient

Consequent-

the choice of which depends upon the

couple with its measuring junction

doubled end of the wire is immersed 10 in. in a furnace with a sharp temperature

this relation.

temperature

If, for example,

The temperature-

Methods.

for detecting

the two ends of the wire being connected galvanometer

it is eviare imprac-

required, size of wires, apparatus available and personal preference. However, the emf of a thermo-

along the

by doubling the wire and inserting

it to various depths in a uniformly

and, as wire de-

ly, there are numerous methods of calibrating

galvanometer

This method is not satisfactory

distribution,

for inhomogeneity

Calibration

mocouples,

Bunsen burner or small electric

or subtractive

emf relation of a homogeneous thermocouple is a definite physical property and therefore does not depend upon the details of the apparatus or method

for quantitatively

moving a source of heat,

along an

if not impossible.

ment of temperatures. Abrupt changes in the thermoelectric power may be detected by connecting the and slowly

of

is attempted.

inhomogeneity,

factory method has been devised determining

measure-

along an inhomogeneous

pends upon the temperature

but thermocouples

While rather simple methods are available

detecting

(11126F),

to an uncertainty

if an emf of 10 PV is detected

wires may be either additive

It usu-

75 72

couple by heating

ject to an uncertainty of the order of 0.2 deg at this temperature. The effects of inhomogeneity in both

that have been used for some time should be so examined

if

along either

element of a base-metal couple with a source of heat at lOO”c, measurements made with it are sub-

neity in a thermocouple may be traced to the manufacturer, but such cases are rare. More often it is introduced

For example,

the order of 1 deg to SOO’C, or 2 deg at 1200°C.

wire now being produced is usu-

sition for most purposes.

in the wires.

in-

measurements

in emf of 1OpV is detected

Similarly, Thermocouple

above will

in temperature

ments made with it are subject

factor

in accurate measurements. 71

methods have been described

inhomogeneity.

due to inhomogeneity

be-

into a

may be measured by any convenient

a difference

of the

wires by welding

heated

74

depend upon

of similar

the junction

dicate

be developed,

will

of one sample

it may be used in

and inserting

for detecting

If, however,

gradient

the homogeneity

method.

and the inhomoge-

in a region where a temperature

homogeneity

the two together of immersion

the

neity is present

After reasonable

testing

within 40 to 1OOpV

accurate

CODES

of wire has been established,

type are sufficiently-accurate.

potentiometers

TEST


of about 10 deg com-

INSTRUMENTS paratively

simple methods of calibration

suffice.

The most accurate calibrations (-297%‘) to 300°C (572%‘) fan ge -183’C by comparing the couples directly

AND APPARATUS

will usually

a calibration

in the are made

room temperature

Other thermocouples

with a standard

(and below if a platinum-resistance

thermometer thermocouples

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

liquid

bath is not available)

are most accurately

calibrated

at the freezing

rately as the calibration

and

suitable

or

with the resistance to yielding

thermocouple.

gold and silver,

serves to define

and

certain

are most accurately

may be calibrated

temperatures

78

or optical

on the IPTS as the Type S thermocouples

The calibration

classes,

points may depart slightly from the IPTS. Above 1063’C (1945”F), the most basic calibrations are

and (2) calibration by comparison struments such as thermocouples,

made by observing, the emf when one junction

mometers,

is in a blackbody

furnace,

the temper-

ature of which is measured with an optical

79

pyrom-

of instruments 77

Although

serves to

-

ACCURACIES

I

in detail

USING

ATTAINABLE

FIXED

POINT

b

Paints, Paints”

O-1100

Zn, Sbb, Ag, Au

R

C-1100

Sn, Zn, Al,

Cu-Ag,

E

O-870

Sn, Zn, Al,

Cu-Ag

J

O-760

Sn, Zn, Al

K

O-1100

Sn, Zn, Al,

Metal

freezing points.

a

Temperature

Uncertainties.

as determined

At Observed

Temp. Range

s

measured

Cu-Ag,

by standard

Tbe

several

by calibration

TECHNIQUES

Deg

Cu

platinum

Of Interpolated Values,

C

0.2 Cu

Uncertainty

Deg

0.3

0.2

0.5

0.2

0.5

0.2

1.0

0.2

1.0

resistance

C

thermometer.

107


but

need not be employed.

I

Calibration

precautions

paragraphs,

of about 5 deg C the more elaborate

Calibration

thermocouple

I

“C

at

certain pre-

fac-

tors which contribute to the uncertainties in the emf versus temperature relationship for a particular

Calibration

Type

to follow

in the following

apparatus to be described 80

referred to

with calibrations

it is necessary

for an accuracy

define the IPTS only in the range 630.5 to 1063”C, this type of thermocouple calibrated at fixed points is used extensively both above and below this range as a working standard in the calibration of For most industrial purposes other thermocouples.

9.3

associated

scribed methods and to take the special described

by no means a simple matter.

TABLE

at fixed points with standard inresistance ther-

In order toobtain the high accuracies

fixed points,

of these two types

the Type S thermocouple

(1) calibration

upon

of the

etc.

above and usually

encountered in eter. However, the difficulties bringing a blackbody furnace to a uniform temperature make the direct comparison

junction,

then may be

depending

the temperature

measuring

are defi-

measurements.

of thermocouples

into two general

curately at the fixed points as the Type S thermocouple, but interpolated values at intermediate

thermocouple

pyrom-

In fact, at the lower temperatures

the method of determining

of the

to agree

has as much claim

better adapted for precise

divided

iust as ac-

ranges,

types of base-metal

nitely

calibrated in this range by direct comparison with a standard thermocouple calibrated as specified. Other thermocouples

thermometer

eter in their respective

points of

the IPTS,

of the standard is known.

for the purpose, and calibrated

calibrated

other types of thermocouples

by compari-

almost as accu-

However, it might be pointed out that outside the range 630.5 to 1063’C any type of thermocouple

boiling points of pure substances. Between 630.5 and 1063OC (1167 and 1945%‘), Type S thermocouple at 630.5OC and the freezing

can be calibrated

son with such working standards

platinum-resistance thermometer. in a stirred liquid bath. In the range 300 to 630.5’C (572 to 1167°F) stirred

accurate to 2 or 3 deg C in the range to 1200°C (2192oF) is sufficient.


may be

ASME PERFORMANCE grouped into two kinds; those influencing servations

at calibration

points,

from any added uncertainty tion between

reduced,

points.

within limits,

care is exercised

Errors from

calibration

rate results depending

can be mate-

crude measurements phasized

understood

Estimates

calibration

attainable

of homogeneous thermocouples

TABLE

are sufficient,

that inadequate

9.4

the important

in the

various

by dif-

ACCURACIES

ATTAINABLE

USING

IN LABORATORY R OR S STANDARD)

(TYPE

Temp. Range Points

Dcg

0 to 1100 0 to 870

About every 100 deg C ,I

J

0 to 760 0 to 1100

11 II

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

ACCURACIES IN STIRRED

-1%

-56 -1%

-56

ATTAINABLE LIQUID BATHS

Calibration

Type of Standard’

Points

to 425 About every 100 deg C ” About every 50 deg C

Of Interpolated Values,

C

Dog C

0.3

0.5

0.5

0.5

0.5

1.0

0.5

1.0

COMPARISON

TECHNIQUES

Of Interpolated

Points,

Values,

Deg

Dog C

C

0.2

”

0.1 II

0.1

PRT

”

,I

1: or T

0.2

0.2

II

LlG

0.1

0.1

to 250

About every 100 deg C

,t

50deg

I,

50 deg C

to 200

50 deg C

= Standard platinum resistance =

LIG

Liquid-in-glass

C

PRT

0.1

0.2

I‘

0.1

0.1

E or T

0.2

0.2

LIG

0.1

0.1

thermometer.

Type E or T thermocouple. thermometer.

108


Uncertainty

At Observed

to 200

E or T =

USING

associated

methods are briefly

Coli brotion

Temp. Range OC

TYPO

o PRT

Calibration

In the following

Uncertainty

Points,

oc

9.5

to possible

COMPARISON

At 0 bserved

R or S E

TABLE

it should be em-

FURNACES

Calibration

K

attention

considerations

calibration

TECHNIQUES

TYPO

used.

sources of error is more often found to be the prac-

facilities.

of the accuracies

More or less accu-

care is a waste when relatively

tice than the converse. 81

9.3,

using the same methods,

upon soundness of the techniques

techniques;

hence the

9.3, 9.4

assume that reasonable

in the work.

are possible

through use of well de-

accuracy should be clearly

when choosing

are given in Tables

The estimates

While excessive

signed equipment and ‘careful required

9.6 and 9.7.

as a result of interpola-

the calibration

CODES

ferent techniques

and those arising

either of these sources of uncertainty rially

the ob-

TEST


some of with the emphasized.

INSTRUMENTS ANDAPPARATUS TABLE

9.6

TUNGSTEN-RHENIUM

TYPE

THERMOCOUPLES

(Maximum calibration uncertainties for ronge 1000 to 2000 C using melting points by wire or disc method)

Calibration

At Observed Cold f 1063°C)

Uncertainty

Points

Of Interpolated

Values“

1000 +9.5 deg C

+2.7 deg C l?53oc

Nickel ( 1453°c)

1453 k3.5 deg C

f4.6

deg C

lt:52°C Palladium (1552oc)

1552 k3.0 deg C

24.0 deg C lt;69% 1769

+3-O deg C

k7.0 deg C 2b”ooocI

Rhodium ( 1960cC1

+5.0 deg C

a These values are taken.

TABLE

9.7

ACCURACIES IN SPECIAL

IrRh vs. Ir b #1 ” W vs. WRe’ 11 ” 30 VS. 6 d

apply only when all five observed

ATTAINABLE USING FURNACES (OPTICAL

COMPARISON PYROMETER

points

TECHNIQUES STANDARD)

1000 to 1300 1300 to 1600 1600 to 2000 1000 to 1300 1300 to 1600 1600 to 2000 loo0 to 1550 1500 to 1750

a Using difference curve from reference table with calibration points spaced every 200 deg C. b 40MORh vs. Ir, 50 IrSORh vs. Ir, or 60Ir40Rh vs. Ir. c W vs. 74W26Re, 97W3Re vs. 74W26Re, or 95We vs. 74W26Re. d 70Pt30Rh vs. 94Pt6Rh.

109 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



ASME PERFORMANCE TABLE

(The

9.8

pressure

SECONDARY

TEST

REFERENCE

is 1 standard atmosphere, point

CODES

of benzaic

POINTS

except

for the triple

acid.)

Boiling point of helium Boiling point of equilibrium hydrogen Sublimation point of carbon dioxide

-268.935 -252.883 -78.5

-452.083 -423.189

-38.86

-37.95

Freezing point of mercury Freezing point of water

0.00

32.00 252.25

Freezing point of indium

122.36 156.61

Freezing point of tin

231.91

449.44

Freezing point of bismuth Freezing point of cadmium

271.37 321.03

520.47 609.86

Triple point of benzoic acid

Uncertainties

313.90

Freezing point of lead

327.43

Freezing point of antimony

630.5

1166.9

660.1

1220.2

Freezing point of aluminum

82

-109.3

621.37

Freezing point of copper

1083.0

1981.0

Freezing point of palladium

1552.0

2826.0

Freezing point of platinum

1769.0

32 16.0

Using

Fixed

cial problems provided that sufficient

The equi-

Points.

immersion

is

librium temperatures listed in Table 9.8 (with the possible exception of the sublimation point of car-

used. Because of the high thermal conductivity of copper, special attention should be given to the

bon dioxide)

problem of immersion when calibrating

are sufficiently

als are readily

available

exact,

and the materi-

in high enough purity,

mocouples.

that

accurate work can be done using these fixed points with no significant

error being introduced

ing the temperatures

listed.

good designs of freezing

furnaces

are important for controlling

and for providing mocouple,

sufficient

points,

point cells

and the standard at the same measured temperature

and

are magnified

the freezes

immersion

if the full potential

85 As higher and higher temperatures are used the difficulties of maintaining the test thermocouple

by accept-

Using freezing

however,

moderating

for the ther-

Although

uncertainties

of palladium

to the freezing:

possible

to the melting

insulators

and platinum,

in only a minor way to the overall

84

using freezing

Uncertainties

accuracy

In addition, and

attained

the comparison

Using

of

Comparison

racy of the standard used.

Methods.

liquid baths usually

When an optical measuring

is used as the

standard,

a good blackbody

blackbody. 87

measurements

Freezing

homogeneous

Points.

The emf developed

thermocouple

at the freezing

a metal is constant and reproducible

no spe-

110 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


pyrometer

are

and the accupresent

from the

tube and from electrical

must be used and the design must be such that the, test thermocouple is at the same temperature as the

The

point using

Comparison

or protection

temperature

method will depend upon the degree

at the same temperature

made in stirred

86

point techniques.

at each calibration

errors arising from contamination

leakage.

these contribute uncertainties

to which the standard and the test thermocouple maintained

temperature.

of about 1500°C (2732oF),

the choice of insulating materials becomes Special attention must be paid to very important.

of the order of 21 degC

are assigned

points (and hence by implication

calibrations

the desired

at temperatures

an oven with

means is used for

higher,

in the temperatures points)

whether a tube furnace,

block, or whatever

maintaining

of the method is to

be realized. 83

Type T ther-


by a point of

if all of the

INSTRUMENTS

AND APPARATUS

following

conditions are fulfilled: (1) the thermo(21 the couple is protected from contamination; thermocouple is immersed in the freezink-point

by optical

sample sufficiently

by the oxide, atmosphere.

far to eliminate

heating

or cool-

ing of the junction by heat flow along the wires and protection tube; (3) th e reference junctions are maintained ture; (5)

at a constant and reproducible

(4) the freezing-point

88

at essentially

these conditions

Many of the metals listed

table.

point temperatures

It is essential,

however,

in

observer

the pure metals will

not be contaminated.

must be protected

cibles

the freezing

heating

temperature

crufurnace

available

commercially.

crucible, Freezing

purchased

Bureau of Standards.

the same apparatus as that described

91

Melting

vantage,

however,

material

is available

experimental difficult.

where

thermocouple

remains

can give results

93

trically

that with which the IPTS

is realized

Thermo-

platinum

(572%?

of these thermocouples

These

thermocouples

and

because

decreases

are usually

up to 1200%

the

rapid-

calibrated

by comparison

heated furnaces.

Above

temperatures.

This thermocouple

with in elec-

1200°C (2192TJ

is a preferred

standard because of its greater

stability

working at high

may be used to

16OO’C @9120F) or higher.

the method

94

than

One method for the comparison

of two such

thermocouples is based upon the simultaneous reading of the emf of the standard and the test there

above 1063OC

111


and

are seldom used for accu-

below 300%

the Type B thermocouple

steady for a few minutes and

With good technique

Platinum

employing

either a Type S or Type R working standard

the emf of the

with no greater uncertainty

Versus

Thermocouples

at temperatures

then drops to zero as the fused metal drops away from the junction.

range covered,

ly at low temperatures.

When the

melting point of the metal is reached,

points will depend on the type

Platinum-Rhodium

sensitivity

points are

raised.

calibration

The number and

the temperature

platinum rhodium alloys

heated furnace the

of which is slowly

the temperature

rate measurements

tween the end of the two wires of the thermocouple in an electrically

at selected

of each point being meas-

92

metal whose melting point is known is joined beand placed

couple being calibrated points,

couples.

amount of

with freezing

pro-

the emf of the thermo-

the accuracy required.

To apply this method, a short length of

temperature

and the

The calibration

Furnaces.

of measuring

choice of calibration

is usually

or at high temperatures

Laboratory

of thermocouple,

points are used to ad-

when only a limited

techniques

type of standard,

ured with a working standard.

above for

but the use of the freeze

more satisfactory.

junction

must be known, but this can be controlled.

cedure consists

The emf of a thermocouple at 89 Melting Points. the melting; point of a metal may be determined with points,

The ac-

by the accuracy

method of heating.

and furnace are point standards

and copper may be

freezing

or the bulb of a

thermometer.

point sample,

of tin, lead, zinc, aluminum, from the National

of the ther-

as the actuating

Of course, the reference

the type of thermocouple,

The furnace must provide uniform

point sample,

of

of the

The method of bringing the measuring junction of the thermocouple to the same temperature as that of the actuating element of the standard depends upon

and have adequate controls to bring the sample Complete units consisting slowly into its freeze. of freezing

The success

is further limited

of the standard.

point metals with

in the region of the freezing

for most

in most in-

such as the measuring

or liquid-in-glass

curacy obtained

aluminum

Tl le choice of a suitable

graphite.

is also important.

that

Copper

this is done by using covered

and covering

powdered

to protect

laboratories.

of a standard thermocouple

resistance

from oxygen contamina-

tion, and it is also advisable and antimony;

accurate

to bring the measuring junction

junction

The

with a

depends upon the ability

element of the standard,

that protection

be chosen of such material

and silver

Methods.

by comparison

mocouple to the same temperature

given in the

tubes and crucibles

and technical

this method usually

Table 9.8 are available commercially in high purity (ca 99.999 percent or better) and can be used assuming the freezing

Comparison

purposes and can be done conveniently

for achieving

are well developed.

Using

of a thermocouple

working standard is sufficiently

a uniform

dustrial

Techniques

This method is not well

the metal should be melted in a neutral

Calibration

calibration

during freezing.

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

temperature

98

tempera-

sample is pure; and

the metal is maintained

pyrometry.

adapted to metals that oxidize rapidly, and if used with materials whose melting; temperature is altered


ASME PERFORMANCE

lize

waiting

for the furnace to stabi-

at any given temperature.

tions are maintained

The measuring

always

iunc-

A separate potentiometer

ure each emf, one connected and each potentiometer ing galvanometer. flected

onto a single

with a reflect-

the galvanometers

being at the

taneous readings are obtained

good insulation

are open,

by setting

The reference

99

the other

of the furnace

found, several

those due to differences

junctions

in the periods

Variations

calibration of thermocouples lected points. For example, termine the temperature connected

being calibrated

and the

standard

are of the same type.

100

reduced.

thermocouple

between

ing to this temperature

observed

101 welding

If two potentiometers

the from

heating 97

as the heating calibration 1063’C

IO2

tube as the

ratios of radiation

of the furnace is to caliof the ther

as possible

temperature

before taking

There are a number of other methods of heat-

properly

to approximately

for example, protected

inserting

the

into a bath of in a 1arKe metal

block. The block of metal may be heated in a muffle furnace or, if made of a good thermal conductor such

above

the IPTS

is defined

in terms of

as copper, may be heated electrically.

Tin,

usually

measured

with an optical

has a low melting point, 232’C

and low

112


in

In this

that the furnace be brought to

molten metal or into holes drilled

carbide tube

At temperatures

both

without removing them from

a constant

the same temperature,

element can be used to extend the

range upward.

together,

or advisable

ing and of bringing the junctions thermocouples

(1945?‘)

by

heated portion of the furnace.

approximately observations.

in one furnace

furnace using a silicon

are calibrated

tubes, then the junctions

case it is necessary

directly

elkment.

A similar

on

should be brought as close together a uniformly

that it will heat or cool

design which employs a nickel-chromium

If it is necessary

the protection

To reduce the time

is obtained

and

mocouple being tested and that of the standard

by this method the furnace

Fast response

temperature

read alternately

to be close to the same temper-

brate the thermocouples

to deter-

is observed

a constant

or wrapping the junctions

changing;.

to the standard and the emf of the

with the second potentiometer.

for

the furnace may be

ature even when the temperature

is set up on the potentiom-

being calibrated

are not available

readings,

When the thermocouples

would be expected

as de-

obtained

If it is desired

should be so constructed

the standard and

one instrument.

corres-

to the thermocouple,

the emf of the standard.

rapidly.

If the

data may be recorded automatically.

the emf of each thermocouple

this emf is set up on the

above, and the temperature

to calibrate

method

when the thermocouple

taking simultaneous

mine the emf of a thermocouple corresponding to 1000°C (1832cF), th e emf of the standard correspond-

required

at a known

process

the calibration

at any number of seif it is desired to deter-

emf of the standard thermocouple

eter connected

are maintained

the unknown are read with the second potentiometer

adapted to the

of a thermocouple

ponding to 10.0 millivolts,

thermocouple

except at

of the two potentiometer

and the emf difference

of the gal-

or greatly

This method is particularly

scribed

circuits

are welded together.

minor errors such as

etc., are eliminated

potentiometer

between the two po-

may be used to automate the calibration

brought to essentially 96

against

emf of the standard is read with one potentiometer

95 By making; observations first with a rising: and then with a falling temperature, the rates of rise and fall being approximately equal, and taking the mean

vanometers,

can be used as the

temperature.

is raised or lowered.

of the results

be maintained

and thermocouple

the point where the iunctions

so that both spots of light pass across the zero of as the temperature

may be

Alterna-

one poten-

to a desired value and adjusting

the scale together

the Type B thermocouple

tentiometers

and therefore, also when the potentiometers are set to balance the emf of each thermocouple. Simultiometer

a pyrometer

as the workina standard.

98 The thermocouples are insulated and protected by suitable ceramic tubes. It is essential that

that the spots coincide

zero point on the scale when the circuits

built into the furnace,

a pyrometer.

to each thermocouple,

scale,

If the test thermo-

working standard after it has been calibrated

The two spots of light are re-

set in such a position

into the back of a blackbody

tively,

is used to meas-

is provided

pyrometer.

couple is inserted used directly

temperature by welding them into a common bead or by wrapping them together with platinum wire or ribbon.

or a photoelectric cavity

at close to the same

CODES

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

tnocouple without

TEST


(450cF),

which

volatility,

makes a satisfactory

thermocouples

bath material.

appear to be good. More than one base-metal thermocouple may be welded together and the hole

The

should be immersed to the same

depth with the junctions tubes are sufficient

close together.

protection,

drilled

Ceramic

when immersed in molten

metal it is preferable

to place them inside

ary tubes of iron, nickel-chromium,

main in contact.

couples are not brought into direct

important that the depth of immersion to eliminate

cooling or heating

heat flow along the thermocouple and protecting

This

tubes.

it is

cally

constant during observations.

insulators,

by

but if a

and protection

When wires,

tubes are large,

tests

should be made to insure that the depth of immer

and the insulating

can be determined

of

single instrument is used for measuring the emf, the furnace temperature should be maintained practi-

be sufficient

of the junctions

the temperature

grees per minute during an observation,

of the thermo-

contact,

readings,

re-

are used for

the furnace may be changing as much as a few de-

or simi-

In all of these methods, particularly

in those cases in which the junctions

The thermocouple

If two potentiometers

taking simultaneous

of second.

graphite,

in the composite junction.

should be clamped in place so that the junctions

but to avoid break-

age by thermal shock

lar material.

AND APPARATUS

sion is sufficient.

by

observing the change in the emf of the thermocouple .as the depth of immersion is changed slightly: If

104 Stirred Liquid Baths. At temperatures below 620°C (11488oF) stirred liquid baths provide an effi-

proper precautions are taken, the accuracy yielded by any method of heating or bringing the junctions

perature

to the same temperature

cient medium for bringing a thermocouple

may be as great as that ob-

tained by any other method. 103

Base-Metal

IO5 The methods of

Thermocouples.

for testing

the exception, bringing

noble-metal

in some cases,

the junctions

thermocouples

platinum-rhodium

ards from contamination.

One arrangement

ing the junction

of a platinum-rhodium

the same temperature

for accurate

the junction

of the standard

0.06 in. in diameter) the base-metal

and

stand-

calibration

standard to

drilled

of

The platinum-rhodium

exception

cement.

This prevents

of the standard thermocouple,

it is) contamination

cause any error. brittle

at this point will

is convient

Vitreous

to 620°C

silica

(1148%.

bare wire in such

glass tubing, for use

up

such as

to 538OC

or ceramic tubing may be The tube should be closed

Unfavorable

close fitting

heat transfer

large diameter

tube.

being calibrated, con-

diameter

conditions

tube will

in

require a

If a bare wire thermocouple the wires must be provided

electrical

insulation

protection

tube.

over the length inserted

Sheathed thermocouples

immersed directly

ther

in the bath liquid

the sheath material

is

with in the

may be

in cases where

will not be attacked

by the

liquid. Salt baths for use at high temperature must be provided with suitable wells into which the ther

not

mocouple protection

If the wire of the standard becomes

at the junction,

tin

greater depth of immersion in the bath than would a

mocouple. If the furnace is uniformly heated in this region (and it is of little value to make such a test unless

Borosilicate

an unnecessarily

with the

in contact with the base-metal

bath.

Pyrex glass,

necessary.

of the small length of about 0.1 in., which

is necessarily

either

calibrated

at the immersed end and of an internal

couple by Pyrex glass or by a small amount of tamination

thermocouples,

may be accurately

or liquid

such as to permit easy insertion of the thermocouple or thermocouples to be calibrated, but no longer than

standard is protected by ceramic tubes to within a few hundredths of an inch of the hot junction, and the end of the ceramic tube is sealed to the thermokaolin and water-glass

Base-metal

(IOO@F9.

is to insert

in the hot junction

higher than are suitable

used

into a small hole (about

thermocouple.

are used at temperatures

liquid

of bring-

as that of a large base-metal

thermocouple

Molten salts

baths. Usually no special preparation of the thermocouple will be required other than to insert it to the bottom of a protection tube for immersion in the

of the standard and the ther-

the methods of protecting

range, are com-

monly used bath media.

or insulated,

of the methods of

or other organic

depending upon temperature

for oil.

with

mocouple being tested to the same temperature

Water, petroleum oils,

liquids,

testing base-metal thermocouples above room temperature are generally the same as those just described

and a tem-

standard to the same temperature.

eters may be inserted

this part of the wire may be

tubes and standard thermomfor protection

from the molten

salt.

cut off and enough wire drawn through the softened seal to form a new junction. examined

106

The seal should be

The standard thermometer

couple standard inserted

after each test and remade if it does not


may be a thermo-

in the protection


tube with

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

INSTRUMENTS

ASME PERFORMANCE

the thermocouple being tested,

or it may be a

thermometer or resistance

eter will

tube.

to the same depth.

uncertainty IO7

The choice of a standard thermom-

be governed principally

110

by the degree of

several

comparisons

temperatures,

permanently

After thermocouples

Installations.

If desired,

can be made, prefer-

ably by either of the first or second methods at

which can be tolerated.

Fixed

then remove this

thermocouple and insert the standard thermocouple

thermom-

eter immersed in the bath close to the thermocouple protection

CODES

and a curve obtained for each

installed

thermocouple

have been used for some time at high temperatures,

necessary

it is difficultsif

Although testing a thermocouple

not impossible,

much the calibrations from an installation nace.

to determine how

are in error by removing them and testing in a laboratory

The thermocouples

are usually

temperature distribution sible,

which it is used.

that the emf

Although

thermocouple

pos-

111

by testing the

couple.

The calibration

is accomplished

dustrial

stand-

ard. IO8

processes,

In this case, as in the calibration

jective

large temperature

perature as that of the thermocouple

thermocouple

is at the same temperature

thermocouple

being tested.

checking thermocouples. thermocouple

flue,

the temperature

the hot junction

112

The standard

113

Interpolation

the base-metal

measurements

thermo-

which is permanently

consists

and indi-

An experimental

ce-

100 temperatures

Frequent-

be little

ther-

of a series of voltage number of

If a test thermocouple

compared with a standard temperature

tube of iron, fire clay, carborundum,

or some other refractory

wires,

determined at a finite

known temperatures.

tube are mounted inside an-

mented or fastened into the furnace wall.

in which they are used is extension

Methods.

mocouple calibration

other protecting

as

that of the

a potenti-

couple. In many installations

the

as closely

cator are tested as a unit and under the conditions of use.

ometer should be used with the standard thermo-

couple and protecting

the

Another advantage of checking thermocouples

that the thermocouple,

tubes as

Preferably

close together,

of the latter represents

in the same installation

being tested with

ends of the protection

clo’se together as possible.

located

furnace.

of the

is inserted through this hole to the

same depth as the thermocouple

how-

in temperature between two similarly

temperature of the fixed thermocouple

The hole is kept plugged,

except when tests are being made.

as the

objection,

reading of the standard thermocouple represents

permanently

large enough to permit insertion

This

because if temperature gradi-

mounted thermocouples

being tested.

etc., at the side of each thermocouple installed,

difference

to the same tem-

One method is to drill a hole in the furnace,

gradients

that the standard

ents do exist of such a magnitude as to cause much

of any

by comparison methods, the main ob-

is to bring the hot junction

is un-

in most furnaces used in in-

ever, is not serious,

thermocouple

IO9

under working conditions

because,

exist and there is no certainty

by com-

with a thermocouple

of 9 deg

had changed only

It may be thought that the method of check-

satisfactory

of the behavior of the thermo-

For

of 6 deg C at 1100°C (11 deg F at

ing thermocouples

the result is far more useful in the sense of

paring the thermocouple

or to the emf.

(16 deg F at 599’%)

the equivalent 20 12OF).

in

in place as is obtained in laboratory

being representative

to the temperature

which had changed in use by the equivalent

If pos-

it is not usually

at one temperature

it is not safe to assume

it has been observed that a thermocouple

C at 315’C

and in the same installation

sible to obtain as high a precision tests,

proportional

such a thermocouple should be tested under

the same conditions

some information,

example,

depends upon the

along the wires.

showing the

to be applied to its readings.

that the changes in the emf of the thermocouple are

fur-

heterogeneous

after such use and in such a condition developed by the thermocouples

yields

corrections

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

the thermocouple being calibrated, liquid-in-glass

TEST

were

instrument at

within a 10 deg F range, there would

need for interpolation However,

between the calibra-

if from 4 to 10 calibration

ly there is room to insert a small test thermocouple

tion points.

in this outer tube alongside

points are all that can be afforded in a given range

couple.

of the fixed thermo-

A third method, much less satisfactory,

of interest,

is

to wait until the furnace, flue, etc., have reached a constant temperature and make observations with

individual

then what is needed to characterize thermocouple

means of which temperatures



an

is a continuous relation, can be approximated

by

AND APPARATUS

10

15

Voltage FIG.

9.11

with a minimum uncertainity a continuous

appear thwarted

OF RAW CALlBRATldN THERMOCOUPLE

calibrating

from voltage measure-

levels.

relation

(E couple) mV

TEMPERATURE EMF PLOT FOR AN IRON-CONSTANTAN

ments at intermediate

20

DATA

the thermocouple

point of zinc and using an equation

from the start

c = c, +

because of the small number of discrete calibration However, interpolation between

extended down to 400% certainty

the calibrati&r points is possible since the emf changes only slowly and smoothly with temperature.

to 1063.OOC. By calibrating

114

One can present raw calibration

represented the highest

accuracy

data directly

equation.

calibration

in the range 630.5

general,

9.11) or by

For example,

the- thermocouple

at

for the range 400 to llOO°C,

is obtained

this practice

thermocouple

for

to 0.5 deg C. However,

of directly

characteristics

within the required

limits

in

representing

does not yield results

of uncertainty.

to 1063.Ooc

115 A better method is based on the use of dif-

with the Type S thermocouple, the method is that prescribed in the NBS C ircular No. 590. An equa-

ferences between observed values

tion of the form e = a + bt + ct2, is used where a, b,

tained from standard reference

and c are constants determined by calibration freezing points of gold, silver, and antimony.

ence tables are presented

at the By

and values ob-

tables.

Such refer-

in ASTM Standard E23O-

72, Temperature-Electromotive


an un-

of more than 0.1 deg C in the range 630.5

which agrees with IPTS

appears well

by a single curve (see Fig.

a simple mathematical

without introducing

freezing points of gold, antimony, and zinc and using an equation of the form e = (2 t bt t ct2, a

T and voltage EC-h, on a

scale so chosen that the information

of the form

bt + ct 2 + dt3, the temperature range can be

points available.

in terms of temperature

also at the freezing

to obtain such

Efforts


Force (EMF)

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

INSTRUMENTS

ASMEPERFORMANCETESTCODES 0.01

>

Oqr\

E

i; ‘ x 2 U

e

Probable “Bert’ Value

,,n

\‘ -0.01

I

7 t 0.5 dea F

I

I*

I 1

I

3.4v !8 w

+ Rl -

Run 3 21

v-

Run 4

-0.05 t0

5

10

15

20

Voltage FIG. 9.12

25

30

(E couple) mV

DIFFERENCE PLOT OF RAW CALIBRATION AN IRON-CONSTANTAN THERMOCOUPLE

I

DATA

1 I

I

I

I\ i

0

5

10 Voltage

FIG.

19.13

15

20

(E couple),

25

30

mV

TYPICAL DETERMINATION ENVELOPE (FROM DATA

OF UNCERTAINTY OF FIG. 9.12)

116 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



FOR

INSTRUMENTS Tables for Thermocouples. are replotted

AND APPARATUS

The data of Fig. 9.11

minimized by use of the least squares technique, one starts the search for the most probable interpolation equation by passing a least squares equa-

in Fig. 9.12 in terms of differences

from the proper reference

table.

The maximum

tion of the first degree through the experimental

spread between points taken at the same level (replication),

but obtained

data.

in random order with

respect to time and level (randomization)

A check is then made to ascertain

all experimental

is taken

points are contained

as the uncertainty envelope. This information, taken from Fig. 9.12 is plotted in Fig. 9.13, and

interpolation

constitutes

ceeds, according

particular

a vital

bit of information

thermocouple

certainty,

determination

system.

only a single

which satisfies

the many possible thermocouple

equation

set of calibration

9.14 together

to the results

characteristic.

is required

thermocouple in one-half

the

difference

Although

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

most probable 117

the uncertainty

interpolation

is centered

experimental

)

points

number of distinct

interval

i; ‘

x. i u

on the

will

together

uncertainties

with

and on the temperature It is recommended

calibration

one which is separated,

are

The factor 2 is

temperature-wise,

Fifth degree least square* interpolating equation LaGrange

e :_ -0.03 e r 6 “, -0.04

First

fit

least

interpolating

equation

l

,o

Experimental

**Polygonal

-0.05 5

10 Voltage

FIG.

(n -1)

degree

squares

9.14

15

25

20

(E couple),

CALI

BRATION

Data Fit

30

mV

VARIOUS POSSIBLE EMPIRICAL THERMOCOUPLE CHARACTERISTIC

REPRESENTATIONS OF BASED ON A SINGLE

THE

RUN


range

points available

-0.01

?! UJ -0.02

equa-

that the


(Run

2)

A as

from all

other points in the set by as much as one-tenth

0

the

on the amount of cali-

Dotted portions of LaGrange fit indicate that curve extends beyond scale

0

be with-

Generally,

arrived at from numerical analysis reasoning. distinct calibration point is defined arbitrarily

I

3

By obtaining

and the degree of

should be at least 2 (degree + 1).

0.01

E .

For

squares interpolation

dependent

bration data available

points,

interval.

envelope

the most probable least

equation.

Making use of this principle,

the fact that overall

9.16).

characteristic

form of the uncertainty

under consideration.

interval

stop-

requirements.

(see Fig.

the uncertainty

determinate from a single set of calibration it is an important fact that all experimental when the uncertainty

degree equation,

the uncertainty

tion are strongly

within

One pro-

of the foregoing

degree least squares equation

at first it appears that the most probable relation characterizing a given thermocouple is sensibly in-

must be contained

on the linear

9.15).

voltage differences from the least squares fit of any set of calibration points, the uncertainty in the

with four of

methods for representing

difference

(see Fig.

the un-

the example given here, a third degree interpolation

points is available. Typical points would be those taken from one run shown in Fig. 9.11 or 9.12 and these are shown in Fig.

which is centered

equation

ping at the lowest

for this information.

Usually,

envelope

check, to the next highest

of the un-

one must rely on judgment or on the cur-

rent literature 116

about the

and the calibration

In lieu of an experimental

certainty

whether

within

the

ASMEPERFORMANCETESTCODES

w -0.01

_-

I

/

c-

,_1

,

0

Uncertainty

-ii b

&I

-0.02

0 c” f -0.03

First

z 5

I!-__LL__/L_L_l .

,o = > -0.05

5

10

15

Voltage

9.15

(E

UNCERTAINTY OF

LEAST

20

couple),

25

RUN

2)

30

mV

ENVELOPE SQUARES

CALIBRATION

METHOD

FOR

INTERPOLATING

DETERMINING

EQUATION

DEGREE

FOR

A SINGLE

(LINEAR)

-0.01

E . 7 x

I % w’

Data (Run

Experimental

0

: w

least

interpolating

equation

g -0.04

>

degree

squaws

FIG.

envelope

(does not include all experimental data; therefore, most probable interpolating equation has not been obtained)

Third degree least squares interpolating equation

0

-0.01

Experimental Uncertainty

equation best this calibration

4 c’ -0.03

-0.04

H 2 ’

-0.05 0

5

10 Voltage

FIG.

9.16

15

20

25

(E couple),

RUN

represents run)

30

mV

UNCERTAINTY ENVELOPE METHOD OF LEAST SQUARES INTERPOLATING CALIBRATION

FOR DETERMINING EQUATION FOR

(CUBIC)

118 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


(Run

(includes all experimenntal data; therefore, the third degree interpolating

-0.02

z b0

Data envelope


DEGREE A SINGLE

2)

INSTRUMENTS in temperature

minimum temperatures

of the particular

choice of one-tenth presupposes cal degree of four for the least equation,

run.

The

erence-junction

a maximum practi-

ment of numerical

analysis.

squares interpolation

equation,

Indeed,

val and start the fitting Thus,

the junction

if the data can-

the uncertainty

procedure

in general,

of the calibration

curve,

data alone.

119

It is not always

erence junctions at a desired

temperature

thermocouple, junctions

possible

to maintain

during the calibration

tion with the desired

OF

of a

of the reference

is measured it is possible

tions to the observed

to apply correc-

emf which will yield a calibrareference-junction

temperature.

of t, by adding to the observed

junction

junctions at t.

percent

1832oF and the reference

rhodium thermocouple

junction

is at

is at 77OF, and

to develop the emf value of the couple

when the measuring junction is at 1832v and the reference junctions are at 32oF. Because the emf is at 77%

and the reference

are at 32oF, the desired and 0.143, 120

-4

-0.103

-0.77

-1.00

-0.75

-0.079

-0.58

-0.75

-0.57

14

-0.054

-0.39

-0.50

-0.38

23

-0.027

-0.19

-0.25

-0.19

0.000

0.00

0.00

to.25

+0,20

to.028

0.00 +0.19

50

0.056

0.40

0.50

0.39

59

0.084

0.60

0.76

0.59

68

0.113

9.08

1.02

0.79

77

0.143

1.00

1.28

0.99

86

0.173

1.20

1.54

1.19

0.204

1.40

1.80

1.40

104

0.235

1.61

2.06

1.61

113

0.266

1.81

2.32

1.82

122

0.299

2.58

2.03

2.02

average temperature-emf does not exceed 2pV. the

junctions

relationships

10 or 13 between

in this range

value is the sum of9.427

or 9.670 mV.

121

The sign of the corrections

ered when applying

mV

--

* The values in this column apply for either the percent rhodium thermocouples. The difference

of the couple is 0.143 mV when the measuring junction

mV

95

suppose the observed

is 9.427 mV when the measuring junction it is desired

mV

were at t, and the measuring

For example,

emf of a piatinum-10

Co perCpOnStanton

Iron-ConStanton

t5

41

value the emf which the couple would give if the reference

ChromelAlumel

mV

32

If the emf of the couple is measured with the refer ence junctions at temperature t, and a calibration is desired with these junctions at temperature to, the measured emf may be corrected for a reference-junction temperature

Force

the ref-

cold junctions)

but if the temperature

PlatinumRhodium* Platinum Versus

Temperoture

Corrections

(commonly called

greater.

Electromotive

points than from the use

Reference-Junction

closer together and in-

TABLE 9.10 AVERAGE TEMPERATURE-EMF RELATIONS FOR THERMOCOUPLES FOR APPLYING REFERENCE-JUNCTION CORRECTIONS

greater precision in temperature determination by means of thermocouples can be obtained from a given number of calibration

temperatures

by bringing

inter-

again.

with a difference

or negative

if it is remembered

creased by making the difference

by using the proper refer-

ence table in conjunction

is positive

that the emf of the couple is lowered

by a fourth degree interpolation

one should increase

correction

should not cause any confusion

in keeping with the low degree require-

not be represented

I18

and when this is added to the observed emf the desired value 9.427 mV is obtained. Whether the ref-

between the maximum and

these corrections.

In the calibration

emf relation

must be consid-

is not always

of couples

the range of reference-junction

For example,

the type of couple may be used.

junctions

at 32oF is 9.570 mV and the emf of the couple with the measuring junction at 1832?? and

tions for the various

the cold junctions

average relations

and the measuring junction

in Table

is required. The emf of

the couple when the reference

junction

at 324:

9.10.

a particular

is at 77q:

is -0.143

relation

of

The average rela-

types of couples

instead

are specified

of the actual relation

couple are, in general,

F.

mV,

in

in

The errors caused by using these


determined

temperatures,

which case the average temperature-emf

suppose the observed emf of the couple with the measuring junction at 1832 “F and the reference

at 77q:

the temperature-

accurately


for

less than 2 deg

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

difference

AND APPARATUS

ASME PERFORMANCE I22

If the thermocouple

in the reference and subject

junctions

variations

ty in temperature,

it is usually

use thermocouple

leads to transfer

junctions reference

junctions

near the furnace.

made of the same materials

as the thermocouple

wires,

but in the case of platinum-rhodium couples a copper thermocouple

connected

to the platinum-rhodium

per-nickel

thermocouple

Thermocouple

leads,

instrument

thermocouple the individual

wire.

thermocouple

to which they are attached.

material.

emf of the working standard

couples,

Calibration 123

Thermocouples

the standard Pt 27, the relatively

quired.

working standard is of the same kind of material

lead and

that being tested,

Pt 27 is added to the small measured emf.

Materials

are ordinarily

125 Except in the case of constantan, two samples of a similar material which will develop more than 0.25 pV/deg F against one another are

made up to de-

each element

against some stable and reproducible

exceptional.

is re-

this is to de-

copper is sometimes

material.

exceed At

rial,

Two samples of platinum,

Table

the National

126

both of which are spec-

Bureau of Standards piece

(since

of platinum

the

mate-

to measure the temperature

The average thermal emf/deg F of platinum other thermocouple 9.11.

emf of these materials

in thermal To

materials

is specified

It is seen that in measuring

working standards,

directly

in

the thermal

against platinum

it is necessary

to measure an

emf which changes by a large amount for a small change in temperature. An accurate measurement of the emf corresponding to a given temperature, there-

1922) is designated

fore, requires

an accurate measurement

120 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


in determining

accurately.

avoid the ambiguity that might arise from this fact, the thermal emf of thermocouple materials tested at referred to an arbitrary

F. Th erefore,

between two samples of a similar

it is not necessary

against

slightly

1.5 FV/deg

difference

used for this

emf, but the same is true of any other metal.

In most cases the value is less than

0.1 pV/deg F. Even in the case of constantan, the thermal emf between two extreme samples does not

purpose, but platinum appears to be the most satisfactory because it can be used at any temperature up to its melting point, can be freed from all traces of impurities, and can be readily annealed in air. pure, may differ

against

the standard Pt 27 in this case, the relatively large emf of the working standard against the standard

One method of accomplishing

tro-chemically

as

the thermal emf measured is small.

To obtain the thermal emf of the material

termine the thermal emf of the various materials low temperatures

against

small emf of the

platinum working standard against the standard Pt 27 is added to the large measured emf. When the

wires.)

method of testing

some

the thermal emf measured is large.

To obtain the thermal emf of the material

velop a specified emf at one or more temperatures, and in order to select and match materials to do this, a convenient

When plati-

in testing

to the

are of the same

of Thermocouple

the standard

metals).

the two

leads are then referred

extension

sum of the emf

against

Pt 27 (the law of intermediate

in the case of base-

(The thermocouple

to as thermocouple

standard for

against the working standard and the

other material,

when each thermocouple

may be

any other

In order that the various laboratories

with the couple wires

couple wire to which it is attached

standard

standard is necessary.

of the material

couple should be kept at nearly the same temperametal couples

from setting up a laboratory

num is used as a working standard

Therefore,

is not necessary

nothing prevents

lead wires are not

where the leads are attached

ture. This

However,

the standard Pt 27 is the algebraic

the same as

versus platinum

it serves as a satisfactory

standard of the same material as that being tested. In any case, the thermal emf of a material against

the temperature-

lead wire is practically

and althat has

I24 Platinum is used as a working standard for testing thermocouple materials in some laboratories. It is generally more convenient to use a working

from all of the pyrometer

thermoelectrically

junctions

ultimate

versus

wire and a cop-

Although

is spectrochemi-

annealed,

and manufacturers may specify and express values of thermal emf on a common basis, a common and

of the copper versus copper-nickel

that of platinum-rhodium iden tical

laboratory

for any of the couples dis-

manufacturers.

to.

their own use.

lead is

lead to the platinum

cussed here, are available emf relation

referred

The thermo-

couples are usually

standard

to which the thermal emf of other materials

of the

couple leads of base-metal

platinum

been prepared,

to

and lower

than to measure the temperature

This

pure, has been thoroughly

though it may not be the purest platinum

the reference

to a region of more constant

temperature

tally

or uncertain-

more convenient

CODES

as standard Pt 27.

is very short, resulting being near the furnace

to considerable

TEST


of the tem-

INSTRUMENTS

AND APPARATUS

TABLE 9.11 AVERAGE THERMAL* EMF/DEG F OF PLATINUM AGAINST OTHER THERMOCOUPLE MATERIALS

Some of the precautions tain accurate results paragraphs.

that must be observed to ob

are discussed

in the following

129 Platinum. The thermal emf of thermocouple platinum against the standard Pt 27 is usually less than 20 /.LV at 2192oF and in testing one sample of

Matsri al

platinum

against

another it is not necessary

ure the temperature of the hot junction Platinum-10

percent rhodium

1832

6.4

Platinum-13

percent rhodium

1832

7.3

Chrome1

1652

Alumel

1652

4.8

1112

6.4

Constantan

1112

26.0

Constantan

212

20.8

Copper

212

5.2

The reference-junction curately

controlled.

accurate measurement of temperature, avoided when the measurements a working standard of material

ing platinumqhodium

tested,

temperature.

130

In many laboratories

the platinum standard

and the platinum element of the couple used to measure the temperature are one and the same. The sample or wire being tested is then welded to the junction of the couple and the emf of the couple and that between the two platinum simultaneously

to that being

wires are measured

with two potentiometers

or alter

nately with one instrument. Simultaneous readings of these electromotive forces should not be made

is small

even for large changes in

with a millivoltmeter or with a current flowing in either circuit, because one wire is common to both

In the latter method, the accurate

but merely shifted to the laboratory

Meas-

about 1112 and 2192oF,

the temperature.

however, is

measurement of temperature is not entirely

The wires

and protected.

are sufficient to develop the emf at any temperature as the emf is small and practically proportional to

for this

since in this case the emf developed

and changes very little

thermocouples. insulated

urements at two temperatures,

are made by using similar

standard (i.e.,

atures by any of the methods described for calibrat-

* Complete tables giving the average thermal emf of platinum-10 percent rhodium, and platinum-13 percent are aiven in NBS Circular 561. rhodium aaainst DhinUm The averaie theimal emf of chrome1 and of alumel against plitinum are given in NBS Research Paper 767, cower and constantan aaainst platinum in NBS Research .. Paper RP 1080, of iron and constantan against platinum in NBS Research Paper 2415.

The necessity

temperature need not be acThe platinum

the wire previously compared with the standard Pt 27) is welded to the wire being tested to form a couple and the emf measured at one or more temper

should be carefully

perature of the junctions.

to closer

than 9OoF to obtain a comparison accurate to 1 /LV.

17.5

Iroll

to meas-

avoided,

circuits

that determines

and in this case the potential

difference

the thermal emf of the working standards against the

measured by one instrument

standard Pt 27.

current flowing in the other circuit. However, this objection is not encountered in the method described

I27

The small thermal emf of a platinum --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

standard against

working

as accurately

as the emf can

131

be measured. These standards are subject to change during use but, if properly used and occasionally checked, can be relied upon to about 2 PV at 1832°F. The thermal emf of working standards rials is determined and certified

Platinum-Rhodium

platinum-rhodium platinum

Alloy.

The testing of

thermocouple wire directly

is exactly

against

the same as the calibration

of

platinum-rhodium thermocouples. Platinum against platinum-10 percent rhodium develops 6.5 pV/deg F

of other mate-

at the National

Bureau of Standards to the equivalent

by the

above in which the platinum standard is not the same wire as the platinum of the thermocouple.

the standard Pt 27 at any tempera-

ture can be determined

is influenced

and platinum

of +2 deg F at

against platinum-13

develops about 7.3 pV/deg

percent rhodium

F at 1832oF. Therefore,

high temperatures.

order to determine the thermal emf of a sample of

In any event the testing of a thermocouple material is essentially the determination of the emf

necessary

platinum-rhodium

128

of a thermocouple in which the material

against platinum to 220 /LV, it is

to measure the temperature to +2.7 deg

F. Such an accuracy in temperature measurements

being tested

is obtained only with a very homogeneous and ac-

is one element and a working standard the other.



in

ASME PERFORMANCE curately furnace,

calibrated couple in a uniformly heated but if the emf of one sample of wire is

known with this accuracy, measuring

into a hole drilled

it may be used to deter

the temperature.

CODES

welding

in the junction

the material

the junctions

mine the emf of other samples without the necessity of accurately

TEST

formed by

to platinum.

This

brings

to the same temperature.

In the use of platinum or platinum-rhodium for testing thermocouple materials, the wires are

For ex-

ample, the thermal emf per degree of any sample of

used a large number of times before checking

platinum-10

or scrapping.

percent rhodium against

any other

sample rarely exceeds 0.03 pV/deg I832oF).

Therefore,

F (5OpV at

Base-metal

used for testing

if the thermal emf of one

similar

thermocouple

materials

wires

should not be

used more than once if the highest

accuracy

sample against platinum is known to +2OgV at 1832cF, th e emf of other samples against the same

required,

change in

platinum

ture and if they are used repeatedly,

can be determined

these materials

to about the same ac-

curacy by comparing the samples of platinumrhodium and measuring junction

of the hot

to select

for

The working standard used to determine of platinum-rhodium,

measuring

the

13 percent rhodium develops

or either element

about 0.89 pV/deg

homogeneous

accurately standard,

F

as necessary

by. comparison

the coil against

for most purposes. similar

standard and the material of some materials

Moteriols

dardized, base-metal

thermocouple materials (alumel, chromel, constantan, copper, and iron) the procedure is very much the same as in calibrating thermocouples.

Although

measurements

are ultimately

it is not necessary rectly

temperature

In case

are small enough that

of 90 deg F is sufficient.

if ever, should it be necessary

Seldom,

to measure the

closer than 18 deg F.

base-metal

such thermal-emf

Annealed electrolytic (b) At LOW Temperatures. copper is very uniform in its thermoelectric

referred to platinum,

to measure each sample di-

against platinum.

between the

being tested.

that have been well stan-

the differences

an accuracy

The accuracy

must be measured

depends upon the difference

In testing

Any sample from

materials.

with which the temperature

A number of wires can be welded together

At High Temperatures.

from

this coil may then be used as a working stan-

and tested by any of these methods.

(a)

samples

the standard Pt 27 will apply

for the remainder of the coil with sufficient accuracy

dard for testing

Thermocouple

as

with a

the emf of which is known, against

thermal emf of the few selected

is known to 2% gV at

211 deg F.

Basa4otal

as found

one or more samples may be

the standard Pt 27. The average value for the

1832oF, the thermal emf of the other against the

134

and

samples.

taken from it and the thermal emf determined

same platinum can be determined to 230 PV by comparing the two and measuring the temperature to

I33

from different

the emf between the various

from such tests,

at 183201;’ so that if the thermal emf of one of these against platinum

samples

If the coil is sufficiently

may be a sample

of the thermocouple used in measuring the temperature. Platinum-10 percent rhodium against platinum-

materials

a coil of wire and test if for homoge-

parts of the coil, welding them all together,

thermal emf of the platinum-rhodium of platinum,

the wires

The procedure then is

neity by taking several

percent rhodium.

is

when heated to a high tempera-

become heterogeneous.

to 18 or 36 deg F. The same applies

platinum-I3 132

the temperature

because there is a slight

properties

When the measurements

and is often used as a standard for

thermoelectric

testing

at temperatures

below

are made against platinum (and this must frequently be done), the platinum wire should be

572oF. The th ermal emf of other materials

sealed through the end of a glazed porcelain

termined very accurately by using a stirred liquid bath or fixed points. The steam point is an excellent one for this purpose.

protecting

tube with Pyrex glass,

leaving

1 cm of the wire exposed for welding base-metal tainty

wire or wires.

in the measurements

certainty

The largest arises

in the determination

ture of the junction. platinumrhodium

about

to the uncer-

Table

from the un-

either copper or platinum

9.12 specifies

nealed electrolytic

of the tempera-

The junction

thermocouple

against

the’ thermal emf of ancopper against National

Bureau of Standards standard Pt 27 and may be

of a standard

used to convert values of the thermal emf of any material against one of these standard materials

may be inserted

122 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


may be de-


INSTRUMENTS

AND APPARATUS

TABLE 9.12 THERMAL EMF OF ANNEALED ELECTROLYTIC COPPER AGAINST NBS PLATINUM STANDARD PT 27

Temperature oF’ r

1 Emf, PV

ence-junction various

temperature

thermocouple

and, therefore,

in determining

materials

corrections

must be applied

1 TempeJ~ture~ 1 Emf,

thermocouples.

under the testing

The average temperature-emf

-328

-194

212

766

-238

-354

302

1265

tions for the various thermocouple

-148

-367

392

1831

platinum

-58

-242

482

2459

used for making reference-junction

0

572

3145

t340

662

3885

t122

136

are specified

necessary

against

in Table

against

9.13 and may be corrections.

at high temperatures, junctions.

ther-

it is not

to measure or control accurately

perature of the reference

the other standard material.

of

rela-

materials

In comparing two samples of a similar

mocouple material to values of emf of the same material

to ar-

rive at values for a common reference junction temperature. The method of applying these corrections is the same as that discussed

32

the emf of

against platinum

the tem-

The emf devel-

oped by two samples of platinum-rhodium, even the 10 against the 13 percent rhodium alloy, is prac135

Reference-Junction

convenient

TABLE

9.13

It is not

Corrections.

for everyone

tically independent of the temperature of the reference junctions between -4 and +122qF. In all other

to obtain the same refer-

AVERAGE TEMPERATURE.EMF RELATIONSHIPS OF VARIOUS AGAINST PLATINUM FOR APPLYING REFERENCE-JUNCTION

Electromotive

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

Temperature, OF

Platinum Versus PlatinumRhodium* mV

Chrome1 Versus

THERMOCOUPLE CORRECTIONS

MATERIALS

Force

Platinum

Alumel Versus Platinum

Constanton Versus Platinum

Iron Versus Platinum

C opper Versus Platinum

mV

mV

mV

mV

mV

-4

-0.103

-0.50

0.27

0.64

-0.36

-0.109

t5

-0.079

-0.38

0.20

0.48

-0.27

-0.084

14

-0.054

-0.25

0.14

0.32

-0.18

-0.057

23

-0.027

-0.13

0.07

0.16

-0.09

-0.029

0.00

0.00

32

0.000

0.00

0.000 t 0.030

t 0.13

-0.07

-0.16

t 0.09

50

0.056

0.26

-0.14

-0.33

0.18

0.060

59

0.084

0.40

-0.20

-0.49

0.27

0.091

68

0.113

0.52

-0.28

-0.66

0.36

0.124

77

0.143

0.66

-0.34

-0.83

0.45

0.158

86

0.173

0.79

-0.41

-1.00

0.54

0.193

95

0.204

0.93

-0.47

-1.17

0.63.

0.229

104

0.235

1.07

-0.54

-1.34

0.72

0.265

113

0.266

1.21

-0.60

-1.51

0.81

0.302

122

0.299

1.35

-0.67

-1.69

0.90

0.340

41

+0.028

0.00

* These values apply for either 10 or 13 percent rhodium.



TABLE

Type of Thermocouple

s

rhodium versus platinum

iG 6~%nk?rhSodium* versus platinum

SUMMARY

OF METHODS

Methods of Cali brotion

International temperature scale (fixed points) Fixed

P

9.14

points

NBS Standard fixed points

1166.9 to 1945.4

Comparison with standard thermocouple 9,

OBTAINABLE

at

Coli bration Points

32

to

2642

THER

Freezing points Sb, Ag, and Au

M Int

Observed Points, Dsg F

of

Freezing point of Zn, Sb, Ag, and Au

32 to 2642 32 to 2642

IN CALIBRATING

Accuracy

Temperature Range, OF

32 to 2642 samples,

AND ACCURAClES

0.36

Equation:

0.36

Difference reference

0.36 About every 200 deg F

0.54

About

0.54

1100 and 2000°F

(or more points) 32 to 2012

About every 200 deg F

0.9

L.inear

32 to 2012

Abouh 900, 1500. and 2000 F(or more points)

0.9


Comparison with standard resistance thermometer** or at fixed points

32 to 662

About

resistance Comparison thermometer** with standard

32 to -310

About every

thermocouple**with standard Comparison

32 to 1400


0.9

Linear

,.

32 to 1400

About 209, 600, and 1400 F

0.9


32 to 662


32 to -310

About every

thermocouple* with standard Comparison 1.

Type K Chromel-Alumel

Type S

resistance Comparison thermometer** with standard

every 200 deg F

100degF

900,

0.18

0.18

0.18

Iron-constantan or at fixed points Comparison with starldard resistance thermometer** --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



100 deg F

0.18

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

TABLE

9.14

Continued

Type of Thermocouple

h ‘e

T

Copper-constantan

Accuracy at Observed Points, Deg F

Methods of Calibration

In

Comparison with standard resistance thermometer** or at fixed points

32 to 572


0.18

Equation: or differenc reference

Compaiison with standard resistance thermometer**

32 to 212

About

0.09

E uation: dijference reference

Fixed

32 to 212

Boiling

0.09

Equation:

points

122, and 212’F

point of water

+ Comparison with standard resistance thermometer**

32 to -310

About every

100 deg F

0.18

F.quation: or differenc erence tab

Fixed

0 to -310

Sublimation point of CO, and boiling point of 0,

0.18

Difference ence table

*Either 10 or 13 percent ** In stirred liquid bath.


points

rhodium.


ASME PERFORMANCE cases,

with the possible

exception

mav be taken as proportional

of iron, the emf

to the difference

of the two junctions,

when the emf is small,

the corrections for the tem-

perature

changes of the reference

junctions

and

of the reference

junctions

are

protected stirred

than by

comparison

that the emf (320 ,V) developed by two samples of iron when one junction was at 1112oF and the other

137

container

of the

calibrating nitrogen.

The accuracies

in calibrating

These

accuracies

homogeneous

may be obtained

thermocouples

the

is exercised

in the test procedure.

curate results

can be obtained

In the case of chromel-alumel conplrs

at low temperatures,

fied in Table interpolated be greatly

9.14 are limited values.

couples at more points. with copper-constantan is usually

limited

in

junction I39

143

ards.

General

obtained

I44

and in

General

according

by em-

cases,

indications

standardization

are

I45

Platinum,

thermometers

nickel,

are usually

specified

THERMOMETERS

ways.

Platinum

Thermom-

catalogs

are normally of the temper-

is usually

tem-

In such

carried out for the

prevailing

with the

of Special

Use

Partial

Immersion

the emergent mercury column or stem may be specified.

temperatures

Cognizance

may be exercised

One method involves

comparison

will

calibrated

by comparison

shall then be measured.

thermometer,

rections

of those

in various of the therimmersion

The number of degrees of scale,

and copper resistance

which

be in the emergent column when in actual

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


use,

From these data the car-

under the specified


Immersion

in manufacturers’

mometers at total immersion with total

Thermometer

with a standard platinum resistance

Partiol

equipment being employed.

In the Case

standards. 140

stand-

and Pars.

of the thermometers.

st&d.ardization

temperatures

Resistonce

Purpose

to their own specifications,

Thermometers,

With Stondord

as secondary

14 to 24, inclusive

emergent column temperature

encountered.

Comporison

thermom-

thermometers

atures of the emergent column for the various

and leads or thermocouple, leads, and indicator as a unit by any of the methods described in the pre-

RESISTANCE

resistance

bought and sold without specification

(multiple-

difficulties

Thermom-

by comparing their readings

as commonly listed

eters,

to test a thermocouple

no additional

THERMOMETERS

Purpose Liquid-in-Gloss

Refer to Pars.

perature

ceding paragraphs,

of the

35 to 38 inclusive.

couples). When it is desired

in Par. 47

the resistance

which have been standardized

at low temperatures

a number of couples in series

air or liquid

eters or general purpose liquid-in-glass

the emf of the

suc!~ cases the accuracy may be improved ploying

bridge as described

with those of standard platinum

can

by the emf measurements

is cooled with liquid

Refer to Pars. 44 to 46, inclusive.

eters are calibrated

speci-

this uncertainty

The accuracy

lower

Generol

by the uncertainty

couples

For still

care

by the same methods.

reduced by observing

liquid

in a

may be used in which the

LIQUID-IN-GLASS

and iron-constantan

However,

resistance

standard bulb.

More or less ac-

the accuracies

a cryostat

with

when reasonable

of

below the ice point,

should be used in measuring

various tvpes of thermocouples by different methods and the uncertainty in the interpolated values by various methods are specified in Table 9.14. 138

of the

the limits

may be immersed in alcohol

A Mueller

142

Obtoinoble

obtained

in a thoroughly

The temperature

surrounded with dry ice.

temperatures,

junctions. Accuracies

winding

at temperatures

the thermometers

and 0.78pV

for each degree change in the temperature reference

for test

by immersing a suitably

with the standard platinum

thermometer

changed by 0.06 FV for each degree change of the hot junction

thermometer

water or oil bath.

with-

error of the calibration, may be measured by means of a standard platinum resistance thermometer. For

changing that of the hot junction by the same For example it was observed (in one case)

in the temperature

which is satisfactory

is obtained

bath, which should be constant within

amount.

at 77%

A calibration

141

code purposes

In comparing two samples of iron, the negligible. emf developed is changed more by changing the temperature

CODES

though the ice point value can be determined out the use of a standard thermometer.

be-

tween the temperatures

TEST

emergent column tem-

INSTRUMENTS peratures

may then be calculated.

organic liquid-filled expansion periment

In the case of

thermometers

the coefficient

of the liquid should be obtained or from the manufacturer

these computations.

to drive the liquid into the expansion

by ex-

overheat

in order to perform

the accuracy

testing,

thermometers

of constructions differing

comparison

similar

but

an auxiliary

cause the separated

ice point scale.

For separations VISUAL

146

General.

INSPECTION

Thermometers

bulb itself.

tected

and are more likely

during service. will

entirely

to occur in shipment

than that

prevent such displacement

If bubbles are observed

I49

of the gas.

mometer.

other convenient

spected

into the bulb.

Gentle

while held upright will the surface.

cooled in this process the liquid,

is drawn

tapping of the thermometer

below the freezing

point of

during the melting

no solidification

occurs in the stem; otherwise

bulb may burst or the capillary cause of the expansion

split

process

forces generated

should always

described

above.

in the bulb.

150

Organic liquids

contrast

to mercury,

should always particularly

the

be allowed

and slowly

heating

for drainage

It is frequently

appreciably

with an ex-

retarded

to occur, such thera good prac-

because drainage

if the capillary

is

is also cooled.

can be joined 151

Globules

separated

portion is driven into the expansion

from mechanical

chamber.

When the column itself

by heating

flows into the

of Liquid

separation

in the stem which result can normally

the bulb until the liquid

chamber, the separated portion usually will join with the main column. A slight tapping of the ther-

with the globules.

mometer against the palm of the hand will facilitate Th is method should not be employed this joining.

dicative

for high-temperature

mometer should be rejected.

If such globules

and then reappear upon cooling of oxidation

of obstructions

(above 500°F),

appear to unite

the bulb, this is in-

of the mercury,

127


be rejoined

column merges

or the presence

in the bore, and therefore,

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


in time

tice to immerse only the bulb of the thermometer

the bulb until the

thermometers

Sufficient

when using or standardizing

until drainage has taken place,

at the top of the

chamber, the liquid usually

to the

as used in thermometers,

wet the glass.

tion of the liquid has separated

by carefully

If the thermom-

of the chamber to drive the

columns, depending on the construction of the thermometer and the type of separation. If a small por-

pansion

be in-

which can be repaired

mometers below 32°F.

is provided

of the

the be-

I48 If gas bubbles are observed in the stem, several different ways are suggested for joining

column and the thermometer

distilla-

condensation

liquid into the bore whence it can be rejoined main body as described above.

so that

internally

Such thermometers for these separations

very careful heating

to warm the

stem sufficiently

thermometers

eter has an expansion chamber that is observed to be filled with liquid, the’column can be reunited by

that, if the bulb is

care should be exercised

should emerge

parent liquid in the upper part of the ther-

by the procedures

cause the bubbles to rise to

It is very important

Then the bulb should be al-

In organic-liquid-filled

ally be removed by cooling the bulb with dry ice or

into the

to bring the liquid

tion may occur, with subsequent colorless

it

on a pad or against

is possible

in the bulb.

in the bulb, they can gener-

coolant until all the liquid

By softly tapping

The liquid lowed to warm up slowly. into the bore with no separation.

de-

No method has been discovered

to join. to join,

to cool the bulb in dry ice to a

the hand it usually together

Gas bubbles are readily

usually

of the liquid

point low enough to bring all of the liquid

should be inspected

faults. Gas Bubbles.

portions

which are more difficult

may be necessary

for gas bubbles in the bulb or liquid column, globules of liquid in the stem, foreign matter, and glass

147

chamber beto develop

tapping the bulb on a pad of paper will

below the immersion point to the extent of

including

are likely

cooling the thermometer so that the separated portion as well as the main column both stand in the chamber. Tapping the tube against the hand, or

of the

point,

graduation

the ac-

the bulb.

separations either in the chamber or above it. It is frequently possible to join such separations by

in all details

above the immersion

that have a contraction

low the lowest

of measurement.

involves

with standards

of the pressure of the gas or destroy

Thermometers

being test-

A second method, which is the one best suited to large-quantity

result

chamber, may

the glass and either break the bulb as a

curacy of the thermometer by expanding

to have

than the thermometer

ed, thus increasing

because the heating of the bulb, which is necessary of

This method has the advan-

tage that the standard may be selected greater sensitivity

AND APPARATUS

the ther-

ASME PERFORMANCE 152

Foreign

detected

in the bore can sometimes

Matter

TEST

be

pearance

convenient

to use a magnifying

of low power for this examination.

of the tested and untested

the scale portion.

with the unaided eye, but it is desirable

and generally

CODES

chalking

glass

sections

Burning out, loosening,

of

or

of the pigment shall be cause for re-

jection.

The most com-

mon types of foreign matter which should be cause for rejection are glass chips, particles of dirt or lint, oxide of mercury either red, yellow, or black, products of glass weathering deposit,

Test

156

commonly called white

and stones or iron spots traceable

treatment

glass fabrication. I53

as filling

above the liquid,

may be treated example

specifications.

thermometers, by accident

thermometer

these conditions crystals

157

be

perature tions, 900°F

column

in Fig.

9.1.

by the production

quality

of

may be of various

types.

the glass as observed with a polarized gage near enlargements

are detrimental

severe that fire cracks may later occur. near the bulb are indicative

and are particularly

thermometers

for use above 320°F.

permanency

of range will normally

high range thermometers most significant.

158

General

(a)

Test

Conriderbtions.

ings at fixed points,

glass

comparison in

selected

The test for

The test for permanency to determine

the ability

(a)

Pigment

of pigment

such as the ice point,

with secondary

of the pigment

the exposure

in use without

graduation

and

exactly

con-

will

being oblit-

observing

may be eliminated

the thermometer

to be tested

the meniscus

hides its own image; the line of sight

then be normal to the stem at that point.

and compensate

of the type shown in Fig. 9.1. Heat for 3 hr at approximately 500°F. Allow it to cool slowly.

perature.

Inspect

parts.

one must realize

position

in ap-

for this in arriving

The best practice

128


width

at the tem-

is to consider

of the lines as defined

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


of

the line of sight so that the

of the scale nearest

that the scale lines are of appreciable

in an oven

for differences

by

in a

assure that the reflection

When reading thermometers, portion of the scale sec-

the thermometer

in

Reading Thermometers to Avoid Parallax.

(2) by adjusting

erated. tion of the thermometer

discussed

the scale can be seen in the mercury thread and

material

other markings to withstand

(b) Place any convenient

readand by

standard thermometers

in accord with the criteria

manner which will is designed

graduations

encountered

by observing

Pars. 35 to 38, inclusive.

is

used to fill the thermometer ditions

Liquid-in-glass

The error due to parallax of

to

of range.

serve to reject

in which this defect

for Permanency

standard-

the prescribed

shall be considered

may be calibrated

(1) carefully 155

is within

the thermometer

thermometers

Strains

objectionable

the thermom-

Calibration

if so

of incomplete

stabilization

is a measure of

If upon subsequent

the thermometer

light strain

in the stem or bore, or at

the top of the thermometer,

of specifica-

accorded

have passed the test for permanency

stones or striae that distort the bore or its appearance is sufficient cause for rejection. Strains in

Heat for

the ice point.

in ice point readings

of the heat treatment

tolerances,

Any

9.2.

bulb at least at the tem-

or, in the absence

eter in manufacture. ization

Faults

specified,

in Fig.

to

of range com-

to not over ‘700’F for normal glass bulbs or for borosilicate bulbs. Allow it to cool slow-

The difference

of red oxide of mercury after being heated

Glass

it in a permanency

ly, and after 72 hr again determine

Under

10 to 12 hr. 154

bulb will under

the ice point of the thermometer

24 hr with the thermometer

portion of the of the liquid

Determine

parator of the type illustrated

of the

of the mercury will occur

and will normally be evidenced

stabilized

at higher temperatures.

be tested and place

of 650 to ‘7OOoF, using the same oxidation

heat

bulb during manu-

with time which may be significant,

particularly

in

of air can readily

the gas filled

type of equipment as illustrated

An inadequately

go shrinkage

which may have

and a short section

to a temperature

of nitrogen

or in violation

The presence

by heating

the adequacy of the stabilizing

The most common

is in the use of air instead

mercury-in-glass

to be used

any other gas present

as foreign matter.

been introduced detected

gas is specified

of range is designed

accorded the thermometer

facture.

Where a specific

of Range

The test for permanency

to determine

to faulty

for Permanency

the

by their middle

INSTRUMENTS

AND APPARATUS

lb) Depth of Immersion for Total Immersion TherAlthough

mometers.

sion thermometers immersed,

by definition

should have the entire index

inconvenient

to do so.

portion of the index is exposed, mercury column correction No correction error; otherwise

should be applied.

mersion thermometers ditions

of complete

is standardized

described mometer.

of the standard-

162

the thermometer

under the same conditions

previous

as

of in-

mometers.

Thermometers

standardized

Immersion Ther-

corresponding

readings

to a

at all other tem-

will be correspondingly

increased

If the method involving

the taking of the

after heating

or de-

to a specified

is used, a note should appear in the for the thermometer,

as follows:

NOTE:The tabulated corrections apply for the condition of immersion indicated, provided the ice point reading taken after heating to . . . . . for not less than 3 min is . . . . .

using the

If the thermometer

are found to be

creased.

table of corrections

by one of the methods described comparators.

the ther-

peratures

temperature

of this type shall be

in Pars. 38 (b) and 38 (c) inclusive, appropriate

standardization,

ice point immediately

of the bulb.

Dept of immersion for Partial

raising

and compare with pre-

If the readings

higher or lower than the reading

creased gas pressure above the column produc(c)

may then be made as

however,

Record the readings

vious readings.

A significant

This is due to the effect

ing a distortion

Observations

above without,

total im-

may be used under conunless

of the

which is thus kept well below the general

level of the ice.

if it is

the proper correction

error may be introduced those in use.

meniscus

should be deter-

immersion.

some of the ice may be heaped

row channel formed to permit observations

the emergent

In some instances

Alternatively,

around the stem above the ice point and a deep nar-

If any

need be applied

found to be less than one-fifth ization

I61

both in use and in standardization,

it is frequently

mined.

ings taken at least 1 min apart should agree within one-tenth of a division,

total immer-

is

of the general purpose type, it should be standardized

by immersion to the specified

the readings ary

depth and

163

compared with those of the second-

If the ice point reading (taken in not less

than 2 min and not more than 5 min after removal of the thermometer from the heated bath) is found to be

standard.

higher (or lower) than stated, 159

Calibration

of ice, preferably Discard

ice with distilled

made from relatively

direct

Fill

precooled

packing

is not included, included,

the Dewar

in fractional

Massachusetts point,

is such that the ice point

may be made to the publica-

Institute

the triple

of Technology.

point of benzoic

fur point are particularly

worthy of mention.

graduated

should be

165

Calibration

at Temperatures

Other

it is specified

ondary standard and the thermometer

that the ice point be taken immedito a specified

temperature.

For

Points.

described

Determine

in Par. 127.

thermometers

graduated in single degrees or large

of the liquid-in-glass

subdivisions,

this waiting

room temperature

period may be omitted.

termination, 160

Raise

the thermometer

gently,

termined

a few millimeters

after at least 3 min have elapsed, and observe the reading.

Successive

standard

is

type, it should be held at the ice point was originally after heating

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


de-

to a specified

The thermometer to be tested

be treated in similar manner.

read-

to be tested as

If the secondary

for at least 72 hr before this de-

unless

immediately

temperature.

tap the stem

Than

the ice point of the sec-

129


The steam

acid, and the sul-

Fixed

after heating

For descripbest suited to.

Bureau of Standards and the

held at room temperature for at least 72 hr, unless ately

If the

Points.

it may be used to advantage.

tions of the National

the ice gently about the stem,

degrees the thermometer

Fixed

but one of the other fixed points is

the purpose, reference

Insert the

For thermometers

at Other

tion of the equipment and techniques

dis-

to a depth sufficient to cover the 32? graduation. As the ice melts, drain off some of the water and add more crushed ice.

Calibration

range of the thermometer

water to form a

slush, but not enough to float the ice. thermometer,

164

Rinse the

with the crushed ice and add sufficient and preferably

will

pure water.

contact with the hands

unclean objects.

all other readings

be higher (or lower) to the same extent.

clear pieces

water and shave or crush into

avoiding

or any chemically tilled

Select

Point.

any cloudy or unsound portions.

small pieces, vessel

at Ice

should

ASME PERFORMANCE Insert the thermometer

166

secondary

standard in the thermometer

adjust the temperature approximately perature.

Checking

for Chonges

nificant

graduated

though the thermometer

exposed portion of the stem will have attained

or aged.

thermal equilibrium

a result the thermometer

When the proper rate of temperature

167

been established, lowing

read the thermometers

order at equal time intervals:

mometer or thermometers thermometers ard.

are taken. rise has

in the fol-

standard,

to be tested,

to be tested in reverse

standard,

the standard should agree with the second. parison

of differences

also indicate

in successive

mometers.

the average readings

Apply the appropriate

reading of the standard. to be applied

For purposes of checking

I69

with specifications, ally sufficient.

For purposes

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

check determinations

173

minus one-tenth I70

Unless

are first received,

rection

for each thermometer

from a table.

are plotted

I74

Where corrections

175

tests should be

jarred,

tem-

the thermometer described

is

before taking a reading

176 in

with thermometers

having a capillary

diameter

to ensure that all portions

thermometer,

applying all corrections, is termed in degrees” of the thermomete:‘.

177

Filled

SYSTEM

THERMOMETERS

system thermometers

may be cali-

brated by any of the methods described


of any temperature

made with a standardized

FILLED

of the thermom-

by

stated to.”

of the to ad-

vantage in the rising temperature method of test as well. Sufficient time should elapse before taking readings

measured

and is generally

“corrections

The limit of reliability

after carefully the “accuracy

shall be

tables or charts shall

of the thermometer

by the phrase,

measurement

order to overcome any sticking of the mercury to the glass. Such tapping is particularly important order of 0.1 mm or less and may be employed

to which the correc-

shall be reported is a measure of

the sensitivity

does not vary more than the

as by tapping,

to which correction

for

in Chapter

temperatures

be made, or to which any temperature

perature in preference to the use of slowly rising temperatures. Such a procedure is satisfactory if of reading and if the thermometer

read-

are to be calculated

The limit of precision

recorded,

precision

In such a curve the cor-

emergent columns, the method described 5, Par. 37 shall be used.

plus or

apart throughout the range of the thermometer.

the bath temperature

a corthan

ings.

The results

may be made at constant

at tem-

against the temperature

made not less than 40 nor more than 100 divisions

Comparisons

standardized.

is to be used frequently

tions at the standardization

171

prepare a table

other than those of standardization,

interpolation

of standardization,

specified,

at the ice point and the

temperatures,

curve will be found more convenient

rections

is usu-

of Data

From the corrections

peratures

of a division. otherwise

When

make a monthly

if the changes are found to be

If the thermometer

for compliance

of at least three series should agree within

than it did originally.

read lower after use than originally.

of corrections

to the

under test.

should be made,

and as

in service

Treatment

for all ther-

one series of readings

slowly,

will read higher after it

the bulb expands and the thermom-

frequently

will

the corrections

to the thermometers

even

annealed

Less

may be lengthened insignificant.

of

rise has

corrections

Calculate

degreees,

has been placed

other standardization

Calculate

in fractional

has been carefully

check of the ice point: later these time intervals

been uniform. I68

sig-

short range

A com-

readings

if the rate of temperature

is especially

the bulb contracts

the thermometers

order, stand-

The average of the first and third readings

Usually

eter will

ther-

This

for large bulb and relatively

thermometers

point.

in Bulb Volume

Small changes in volume of the thermometer

172

bulb may occur during use.

at a

will ensure that any

before readings

is

remote

from room temperatures.

The rate should not exceed one scale of this requirement

This

at test temperatures

tem-

Apply suf-

slowly

in 3 to 10 min at the standardization

Fulfillment

thermal equilibrium.

important

errors of the

detected.

heat to raise the temperature

uniform rate.

particularly

to use two stand-

since observational

standard may then be readily

CODES

eter have attained

and

of the comparator to a value

It may be advantageous

division

holder,

10 deg F below the standardization

ard thermometers, ficient

to be tested and the

TEST


under

INSTRUMENTS “Liquid-in-Class specified 178

Thermometers,”

in some particular

unless

temperature

the average service

and in a descending

From these data a correction

should be deter

average service

by steps in an

Considerations,

sume that the error between cording to a straight-line

in a separable

Figs.

be necessary

necessary

[29].

If the bulb is used

it should be calibrated

socket.

to install

If properly

filled,

3

182

ac-

Errors may be introduced

temperature

law.

Any great variations

or V filled

with the temperature the result

of indications

ascending

of (a) not allowing

check point for establishment librium,

of the capillary

by variations

tubing of Classes


II, III or V filled

is

of temperature

(c) excessive

hysteresis

or (d) excessive

and the chart.

friction

If excessive

system thermometers.

I,

The magniin the follow-

ing manner. 183

While holding the bulb temperature

in a suitable

in the Bourdon

bath, the capillary

entire instrument

between the pen before final

in

of Classes

equi-

environmental

lost motion or friction

is found it should be corrected

in the I, III

or by variations

tude of these errors can be determined

time at any

(b) loose fits in the mechanism of the in-

strument, spring,

obtained

and descending

sufficient

in the

it is un-

the bulb in the same position

the case (Bourdon spring) temperature 179

I in-

to the

as when calibrated.

to as-

check points varies

pressure socket,

same separable

order, successively.

It will

During calibration,

strument should be as close as possible

curve, such as shown

in I & A, Part 1 on General and 4, should be drawn.

temperature.

the pressure of the air surrounding the Class

points on its scale while the

of the bulb is varied

ascending

otherwise

code.

The errors of the instrument

mined at successive

AND APPARATUS

should be placed

test chamber.

by the tubing and/or

calibra-

mined together

tion and use.

constant

tubing and/or

the

in a suitable

The error introduced

instrument

or separately

can then be deter-

at various

ambient

temperatures. 180

The pen movement of a recording

should be further checked by bringing such a position curved radial instruments

temperature

because

the temperature

184

The failure

son of the optical

to the

of a pen to travel to it indicates

platinum-rhodium

in the multiplying

pyrometer reading thermocouple

the comparison

perature indicates

temperature. eccentricity

conditions

before final

can be corrected

one not having

The

185

exist,

calibration.

they

bulb and the indicating

instruments

of this type, the

or recording

in the manner de-

To carry out the former method, the thermo-

of smoke and incandescent

Chart

only by substituting

this fault.

In calibrating

pyrometer

couple is heated in a closed furnace which is free of the thermo-

couple tube immediately

the measuring

“peep

part of the in-

gases and at such a

point in the furnace that that section junction

181

taken with a poten-

by revolv-

when the bulb is at a constant

If such faulty

with the simul-

should be made against the reading

scribed later.

should be eliminated

and

compari-

tiometer. If points above 27504: are to be checked,

sys-

loose fits in the mechanism.

checking

by direct

of the emf of a standard platinum-

of a standard optical

chart should be checked for concentricity

be,low 2750%,

may be accomplished

tem or a bent pen arm. The failure of the pen to trace the same curve with rising and falling tem-

ing it upon its spindle

PYROMETERS

At temperatures

calibration

taneous reading

along the curved line or parallel either a lack of adjustment

instrument

it may vary with them.

OPTICAL

of the

bulb should be varied so that the pen travels limit of the chart.

and/or

at the commonly used bulb

The others should move

While holding the chart

in this position,

of the capillary

On multi-pen

no more than one pen can be adjusted

to a time line.

stationary

The magnitude

error should be checked

that the pen lies on one of the

time lines of the chart.

to move along a time line. parallel

instrument

the chart to

surrounding

of the thermocouple

hole.”

is visible

through a

If the test is made in a laboratory,

tubular type electric

strument should be at the same relative elevations as when in use. The bulbs of Class I and III Ther

dimensions

mometers should be immersed

couple in this case is inserted at one end of the furnace tube to such a depth as will bring the end of the thermocouple tube to the middle of the fur-

the capillary

tubing of the thermometers

at a temperature

should be

which is as close as possible

are not less than 24 in. long and not

more than 1% in. in diameter.

to the same depth and to

131 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


a

furnace should be used whose


The standard thermo-

ASME PERFORMANCE nace, the other end of the furnace being left open so that the end of the couple is in sight. I86

After heating

perature,

the furnace to the desired

optical

and with the optical

pyrometer

between

thermocouple and the optical

sighted

to read black-body

190

As the tungsten

be a considerable

overlap

there will,

device

with resultant

is

blackening

in general,

so that the reading

may be applied to each scale and a check secured upon the accuracy

of the range shifting

filament

pyrometer

lamps

operated at filament

temper-

due to evaporation increase

of the tungsten

of filament

of the bulb.

resistance

and

Lamps which have had

years of service in industrial plants have shown no appreciable change in calibration. Lamp life is usua!ly limited only by mechanical defects in the Occasional checking is debase or by breakage.

between the two or several

the overlap,

the pyrometer un-

in the same terms-

atures higher than 2700%‘, they undergo practically

In such a case, it is well to check at a

point lying within

The

calibrated

OrigidlY

temperatures,

are never intentionally

is estab-

When, as is usual, a range shifting

depart from the true temby the same amount.

der check will be calibrated

no deterioration

a part of the optical pyrometer ranges.

of both will

lished. 187

compared are of the same type,

the same band of the spectrum in each,

perature of the filament

the standard

pyrometer

pyrometers

the readings

tem-

upon the end of the thermocouple tube, readings should be taken on each as nearly simultaneously as possible. Continue to take readings until equality or a constant difference

CODES

employing

it should be held as nearly constant as

possible,

TEST

sirable,

medium.

however,

to eliminate

metal fragments

on lenses,

errors due to dust or

screens,

and bulbs.

188 For checking at points beyond the range of a platinum-platinum-rhodium thermocouple, the same procedure as outlined lowed, substituting

in the foregoing

a standard optical

BIMETALLIC

may be folpyrometer

and

a fire clay target for the standard thermocouple.

191

The

Bimetallic

THERMOMETERS

thermometers

are usually

brated by comparing their readings

cali-

with those of

fire clay target is mounted in the furnace at the

primary or secondary

place normally occupied

mometers,

using variable

described

in Par. 46 (b) and its subparagraphs.

by the thermocouple

the standard and unknown alternately

sighted

and into

standard liquid-in-glass temperature

thell-

comparators

the same end of the furnace and read under conditions as described 189

A ribbon filament

convenient

192 Calibration at fixed points necessary because the accuracies

above. tungsten lamp is a very

means of checking

optical

the range of 0.5 to 10 deg F.

pyrometer I93

readings up to 41720F(2300°C) in laboratories or plants, where it is possible to maintain an optical pyrometer calibrated by the National Bureau of Standards or other qualified

is generally not involved are in

laboratory.

If a well is provided

with the thermometer,

I94

The fila-

Before

calibrating

a bimetallic

thermometer

ment should be at least 0.050 in. wide, and long

it should be checked for proper operation

enough to insure a uniform temperature over its central portion. A shallow notch in the filament

jecting

the filament,

indispensable.

eter on the flat filament, latter is adjusted reading. stituted

I95

the standard pyrom-

the desired

for the standard instrument

and a reading

made with it, the current in the flat filament

from point-to-point.

bath of the comparators.

suff,icient by increasing ing changes.

pyrometer

196

being

it to determine

The errors of the instrument

the temperature

of the bath(s)

and descending

order, successively.

While the

is not a black body, if the two

should be allowed

should be deter-

is varied in ascending Sufficient

at each check point to achieve

132 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


is

if the read-

mined at a minimum of five points on its scale while

during the standard pyrometer tungsten filament

A check

that the immersion

held constant at the value to which it was adjusted reading.

its

that the pointer moves freely

should be made to determine

The pyrometer to be checked is now sub-

by sub-

within

The bulb should be immersed at least 2 in.

in the liquid

the current through the

to develop

of temperature

without sudden movements

but it is not absolutely

While focusing

it to variations

range and observing

marks the point ‘at which settings are to be made. A flat window in the side of the bulb is desirable for viewing

it

should not be used during calibration.


time

IYSTRUMENTS

AND APPARATUS

stability of temperature and equilibrium between the bath, the standard thermometer and the bimetallic thermometer. From these data a correction curve,

REFERENCES

199

as shown in I 8~ A, Part 1 on General Considerations, Figs.

3 and 4, should be drawn.

It will be necessary

to assume that the error between check points is linear. 197

It is advisable

to tap the instrument

before taking each reading.

Excessive

ment upon tapping is indicative within the instrument

lightly

pointer move-

of excessive

friction

and is cause for its rejection.

198 Any great variation of indications with the temperature ascending and descending is the result of (11 not allowing sufficient time at the check points for establishment of temperature equilibrium, or (2) lost motion, in the mechanism the instrument

excessive

friction,

of the instrument.

should be repaired

or hysteresis If due to (21,

or discarded.

The following

references

“Standard Soecifications for ASTM Thermometers.” ASTM E-1-7 1. “Method for Verification and Calibration of Liquid lnGlass Thermometers.” ASTM E-77-70. “Manual on the Use of Thermocouples in Temperature STP 470 ASTM. Measurement,” “Fundamentals of Temperature Pressure and Flow Measurements,” Robert P. Benedict, John Wiley & Sons, Inc., N.Y., 1969. “Calibration of Liquid-In-Glass Thermometers NBS Monograph 90,” J-ames F. Swindells, 1965. “International Practical Temoerature Scale of 1968.” Robert P. Benedict, Leeds &d Northrup Technical Journal No. 6, 1969. “Temperature-Electromotive Force (EMF) Tables for Thermocounles,” ASTM E230-72. “Precision Measurement and Calibration,” J. F. Swindells. Editor. NBS Soecial Publication 300. Vol. ’ 2, August; i968. ’ “Theory and Methods of Optical Pyrometry,” H. J. Its MeasureKostkowski and R. D. Lee, “Temperature, ment and Control in Science and Industry,” Vol. 3. Part 1, p. 449, Reinhold, New York, 1962.

133 --`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---


are recommended:


ASME PERFORMANCE

TEST

CODES

APPENDIX

To Convert from

Factors

To

Multiply

I

degree Fahrenheit

degree Celsius

degree Fahrenheit

kelvin (K)

tK= ($

degree Fahrenheit

Rankine (R)

tR= (lF t 459.67)

degree Celsius

kelvin (K)

tK = tC t 273.15

foot/second

metre/second

foot inch pound-force/inch2

(psi)

(“C)

E - 01

metre fm)

3.048 000*

E -01

metre (m)

2.5.x) 000*

E -02

Pascal (Pa)

6.894

Btu in/s-fta

watt/metre-kelvin

Btu/h.ftz

deg F

t 459.67)/1.8

3.048 000*

kilogram/metrea

deg F

1~ = (lF -32)/1.8

(m/s)

pound-mass/footJ

(kg/ma)

watt/metrea-kelvin

(W/‘m-K) (w/rnz-K)

757 E t03

1.601846

E to1

5.192 204 E t02 5.678 263 E to0

foot2

metrea (ma)

9.290 304* E -02

Btu/hour

watt (W)

2.930 711 E -01

Btu/lbm.

deg F

joule/kilogram-kelvin

(J/kg-K)

4.186 800* Et0

*Relationships that are exact in terms of the base units are followed by an asterisk. The factors are written as a number greater than one and less than ten with six or less decimal places. The number is followed by the letter E (for exponent), a plus or minus symbol, and two digits which indicate to power of 10 by which the number must be multiplied to obtain the correct value. For example: 3.523 907 E-02

is 3.523 907 x lOA

or 0.035

239 07


by


--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---

Conversion

--`,````,,`,`,```,,,,`,,`,`,`-`-`,,`,,`,`,,`---



Asme Ptc 19.3 Temp Msrmt

Recommend Documents