NCRP report no 112 calibration of survey instruments used in radiation protection for the assessment of ionizing radiati~1

11 2 CALIBRATION OF SURVEY INSTRUMENTS USED IN RADIATION PROTECTION FOR THE ASSESSMENT OF IONIZING RADIATION FIELDS AND RADIOACTIVE SURFACE CONTAMl NATION Recommendations of the NATI

NCRP REPORT No 11 2

CALIBRATION OF SURVEY INSTRUMENTS USED IN

RADIATION PROTECTION FOR THE ASSESSMENT OF IONIZING RADIATION

FIELDS AND RADIOACTIVE SURFACE CONTAMl NATION

Recommendations of the

NATIONAL COUNCIL O N RADIATION

PROTECTION AND MEASUREMENTS

Issued December 31,1991

National Council on Radiation Protection and Measurements

7910 WOODMONT AVENUE 1 Bethesda, MD 20814

LEGAL NOTICE This report was prepared by the National Council on Radiation Protection and Measurements (NCRP) The Council strives to provide accurate, complete and useful information in its reports However, neither the NCRP, the members of NCRP, other persons contributing to or assisting in the preparation of this report, nor any person acting on the behalf of any of these parties: (a) makes any warranty or representation, express or implied, with respect to the accuracy, completeness or usefulness of the information contained in this report, or that the use of any information, method or process disclosed in this report may not infringe on privately owned rights; or (b) assumes any liability with respect to the use of, or for damagesresulting from the use

of any information, method or process disclosed in this report, under the Civil Rights

Act of 1964, Section 701 et seq as amended 42 U.S.C Section 2000e et seq (Title VZZ)

or any other statutory or common law theory governing liability

Library of Congress Cataloging-in-Publication Data

National Council on Radiation Protection and Measurements

Calibration of survey instruments used in radiation protection for the

assessment of ionizing radiation fields and radioactive surface contamination: recommendations of the National Council on Radiation Protection and

Bibliography: p

Includes index

Copyright 0 National Council on Radiation Protection and Measurements 1991 All rights reserved This publication is protected by copyright No part of this publica- tion may be reproduced in any form or by any means, including photocopying, or utilized by any information storage and retrieval system without written permission from the copyright owner, except for brief quotation in critical articles or reviews

Contents

1 Introduction

1.1 General

1.2 Scope and Structure

1.3 Need and Intent

1.4 Review of Current Efforts/Recommendation

2 Considerations i n the Calibration Process

2.1 General

2.2 Level of Calibration

2.2.1 General

2.2.2 Full Characterization

2.2.3 Calibration for Specific Acceptance

2.2.4 Routine Calibration

2.3 Performance Check

2.4 Precalibration Check

2.5 Qchnical Considerations of Source Selection

2.5.1 Radiation Type

2.5.2 Field Intensity and Source Strength

2.5.3 Source-Detector Geometry

2.5.4 Traceability of Source Calibration

2.5.5 Accuracy of Calibration Source for Field Intensity Determination

2.5.6 Incidental and Spurious Radiations

2.6 Instrument Response Considerations

2.6.1 General

2.6.2 Energy Dependence

2.6.3 Directional or Angular Response

2.6.4 Detector Wall Effect

2.6.5 Geotropism

2.6.6 Environmental Effects

2.6.7 Influence of Other Ionizing Radiations

2.6.8 Linearity Measurements in Calibration

2.6.9 Calibration on Selected Scales and Limited

Ranges 2.7 Uncertainty in the Calibration Process

2.7.1 Genera.1

2.7.2 Uncertainty Associated with Random

Variations

2.7.3 Uncertainty Associated with Systematic Errors

2.7.4 Instrument Stability

2.7.5 Applying the Accuracy Criterion in the Calibration Process

2.8 Frequency of Calibration

2.9 Record Requirements

2.10 Summary of Recommendations

3 Calibration Facility

3.1 General

3.2 Background Radiation

3 3 Scattering 3.4 Equipment Requirements

3.5 The Physical Facility

3.6 Staffing 4 Calibration of Photon Measuring Instruments for External Radiation Field Evaluation

4.1 Introduction

4.2 Source Selection

4.2.1 General

4.2.2 Energy Requirements

4.2.3 Source Strength

4.2.4 Source Output Characteristics

4.2.5 Source Geometry

4.2.5.1 Sources in Free Air 4.2.5.2 Collimated or Enclosed Fields

4.2.5.3 Calibration Boxes

4.3 Characterization of Radiation Field

4.3.1 General

4.3.2 Selection and Use of Transfer-Standard Instruments

4.3.3 Field Uniformity Over Detector Volume 4.3.4 Energy Spectral Quality

4.3.5 Effects of Scatter

4.3.6 Incidental and Spurious Radiations

4.4 Instrument Response Considerations

4.4.1 General 4.4.2 Energy Dependence

4.4.3 Mixed Radiation Fields

4.4.4 Pulsed Radiation Fields

4.4.5 Time Constant

4.5 Accuracy and Acceptance Criteria

4.6 Frequency of Calibration

4.7 Calibration Examples

vii

32

33

viii 1 CONTENTS

5 Calibration of Beta Dose Measuring Instruments for

External Radiation Field Evaluation

5.1 Introduction

5.2 Source Selection

5.2.1 Energy Requirements

5.2.2 Source Strength

5.2.3 Source Geometry

5.3 Characterization of Radiation Field

5.3.1 Dose Rate

5.3.2 Field Uniformity

5.3.3 Energy Spectral Quality and Incidentall Spurious Radiations

5.4 Instrument Response Considerations

5.4.1 Linearity and Stability

5.4.2 Energy Dependence and Geometry Effects 5.4.3 Mixed Radiation Fields

5.5 Accuracy and Acceptance Criteria 5.6 Frequency of Calibration and Conditions of Recalibration

5.7 Calibration Examples-Determination of Point Source and Distributed Source Calibration Factors

5.7.1 Calibration with Point Sources

5.7.2 Calibration with Distributed Sources 5.7.3 Calibration Factor Application for Field Measurement Geometries

6 Calibration of Portable Instruments for the

Assessment of Neutron Radiation Fields

6.1 Introduction 6.2 Source Selection

6.2.1 General

6.2.2 Energy Requirements

6.2.3 Source Strength

6.2.4 Source Geometry

6.3 Characterization of Radiation Field

6.3.1 Fluence Rate and Dose Equivalent Rate

6.3.2 Field Uniformity over Detector Volume

6.3.3 Energy Spectral Quality

6.3.4 Effects of Scatter

6.3.5 Incidental and Spurious Radiations

6.4 Survey Instrument Response Considerations

6.4.1 General

6.4.2 Energy Dependence

6.4.3 Mixed Radiation Fields

6.4.4 Pulsed Radiation Fields

7 Calibration of Field Instrumentation for the

Assessment of Surface Contamination 105

Appendix A-2 Photon Measuring Instrument

Appendix A-3 Examples of Calibrations in Photon

Appendix B-1 Calibration of a Source Using an

Appendix B-2 Example of E,, Determination 144

Appendix B-3 Example of Instrument Calibration for

X 1 CONTENTS

Appendix C-1 Neutron Source Measurements 151

(2.1.1 Manganese Sulfate Technique 151

(2.1.2 Long Counter Application 151

C.1.3 Activation Techniques for Thermal Neutrons 152

Appendix C-2 Estimation of Dose Equivalent Rates from Moderated 23aPu-Be and

Moderated 252Cf Sources 154 Appendix C-3 Calibration of an Anderson-Braun Type

Neutron Survey Meter 157 C.3.1 General 157

(2.3.2 Example 158

Appendix D Examples of Calibration of a Thin Window G-M Detector for Assessment of Surface

Contamination 162 D-1.1 Example 1 - Calibration of a Thin End Window G-M Counter with a Reference Point Source in a "Weightless" Source Mount 162

D-1.2 Example 2 - Calibration of a Thin End Window G-M Counter with a Reference Point Source on a Thick Disc Mount 165

Appendix E Determination of Average Fluence Rate in a Detector Volume Relative to the Fluence Rate at the Center of the Detector Volume for Unattenuated Radiation from a Point Isotropic Source 168 E-1 General 168

E-2 Mean-Value Calculations 169

Appendix F Systematic Uncertainties in the Calibration Process 172

F-1 General 172

F-2 Systematic Uncertainties Associated with Specific

Aspects of Calibration 172 F.2.1 The Instrument Being Calibrated 172

F.2.2 The Transfer Standard Instrument 173

F.2.3 The Radiation Source 174

F.2.4 Associated Measuring Instruments 174 F.2.5 Environmental Influences 174

F-3 Example of the Influences of Systematic Uncertainties in the Calibration Process 175

Appendix G Glossary 178

References 182

The NCRP 190

NCRP Publications 198

Index 209

1 Introduction

1.1 General

The NCRP has provided recommendations for the protection of workers and the public from the harmful effects of radiation from occupational or other sources Implementation of these recommenda- tions a s well a s demonstration of compliance with the requirements

of regulatory agencies requires instrumentation and techniques for the measurement and evaluation of radiation fields and radioactive contamination

Instruments designed to detect and evaluate radiation andlor to assess radioactivity in the workplace provide information necessary

to control the radiological hazards For situations in which personnel dosimetry is not available to provide acceptably accurate estimations

of dose equivalent, evaluations based on portable instrument mea- surements may be helpful The major applications ofportable instru- ments, however, are for purposes of radiation dose control (In this Report the phrases portable instruments and survey instruments are used synonymously to refer to hand-held instruments used for the assessment of radiation fields and/or radioactive surface centami- nation.)

Proper calibration procedures are an essential requisite toward providing confidence in measurements made for these purposes This Report provides guidance and includes recommendations with respect to the calibration of portable instruments used in dose equiv- alent assessment and evaluation of surface contamination For an instrument intended to measure dose equivalent or dose equivalent rate related quantities, calibration is the determination of the instru- ment response in a specified radiation field delivering a known dose equivalent (rate) at the instrument 1ocation;calibration normally involves the adjustment of instrument controls to read the desired dose (rate) and typically requires response determination on all instrument ranges For instruments designed to measure radioactive surface contamination, calibration may be the determination of the detector reading per unit surface activity (uniformly distributed) or the reading per unit radiation emission rate per unit surface area,

or the reading per unit activity Because of the NCRP's concern

of portable instruments currently available for radiation measure- ments (e.g., the discrepant responses of thin end window detectors to point and distributed sources of beta radiation) References to, or discussions of, the operational use of instruments are included, and observations are made that an acceptably accurate laboratory cali- bration does not guarantee the same level of accuracy operationally

In view of these considerations, some recommendations with respect

to the accuracy required of calibrations differ from earlier recommen- dations of the NCRP and other groups In addition, it is noted that

it may not be possible to achieve the level of accuracy in operational measurements sometimes recommended by such groups None of this is intended to excuse any reasonable attempt at eliminating controllable sources of error in the calibration process, but only

to recognize that real and difficult problems do exist in radiation measurements, and these necessarily affect our ability to make accu- rate measurements

The Report provides considerable discussion of various problems, complicating factors, and uncertainties in the calibration process Awareness of such considerations is necessary in order not only to understand the impact of various influencing factors on the calibra- tion process but also to encourage attempts to reduce sources of error and uncertainty

1.2 Scope and Structure

This Report is concerned with the calibration of radiation survey instruments The objectives are to establish the technical guidance, the techniques and the procedures to characterize the desired responses of various types of survey instrumentation through appro- priate calibration techniques Dosimetry and techniques for radio- logical hazards control in the workplace are not discussed

For purposes of this Report, instruments will be categorized accord- ing to intended measurements, as follows:

1.2 SCOPE AND STRUCTURE / 3

1) radiation field measuring instruments-values are generally

reported in terms of dose equivalent rate with units, e.g., Sv h-l,

rem h-' or in terms of units of absorbed dose rate, air kerma rate or exposure rate that can be related to dose equivalent rates In order

to facilitate the use of the international system of units (SI) , the quantity air kerma can be substituted for exposure The quantity air kerma is used in the discussions that relate to calibration of photon- measuring instruments, although the quantity exposure is com- monly used in the United States, and it is referred to at times Appendix A provides details on photon-measuring instrument cali- brations and in the examples the quantity exposure rate is used in relation to instruments that read out in exposure rate units Air kerma is the product of the photon energy fluence and the average (weighted according to the photon energy spectral distribution) value

of the mass energy transfer coefficient in air at a point of concern Under conditions of secondary charged particle equilibrium and insignificant electron energy loss by bremsstrahlung, one roentgen

of exposure corresponds to an air kerma of about 8.7 mGy (NCRP,

1985) The instruments dealt with are those the readings of which provide a direct measure of, or may be used to determine, absorbed dose or dose rate or dose equivalent or dose equivalent rate in radia- tion fields comprised in whole or in part of x and gamma rays, beta particles and neutrons

2) instruments for measuring surface-distributed radioactivity- values are generally reported in Bq [disintegrations per second (dpsll

or [disintegrations per minute (dpm)] commonly referred to a speci- fied surface area The instruments discussed are those intended for measurement of alpha, beta and gamma contamination levels on personnel, accessible surfaces and/or equipment

The uses of portable instruments can be categorized as follows: detectionlsearch for this use, instruments are designed with

maximum sensitivity in order to permit detection of low levels quickly; response pri- orities in order of importance are sensitivity, precision, and accuracy;

relative response this use requires evaluation of existing radi-

ation fields to determine changes from previ- ous survey values; response priorities in order of importance are precision, sensitiv- ity, and accuracy;

exposure control for this use, survey instrumentation must

provide accurate results which are consis- tent with personnel dosimetry results;

4 1 1 INTRODUCTION

response priorities in order of importance would typically be accuracy, precision, and sensitivity

This Report is intended primarily for those who deal with applied radiation protection Therefore, portable survey instruments of the hand-held type are emphasized I t may be useful to instrument designers and manufacturers/suppliers as well as to dosimetrists and metrologists Much of the discussion also applies to calibration of fixed monitors for detection of external radiation with some modify- ing considerations a s discussed briefly in Section 2.1 There are no discussions or recommendations regarding calibration of field-use spectrometers for the assessment of the energy distribution associ- ated with photons, neutrons, or charged particles

Sections 2 and 3 include subject matter applicable to calibration

of most portable instruments The remaining four sections relate

to concerns and recommendations specific to the particular type of calibration being performed In order to provide an appreciation of the actual implementation of these concepts in the calibration pro- cess, specific examples of selected calibrations are noted a t the end

of each section and are presented in detail in the appendices

1.3 Need and Intent

Characteristics of the ionizing radiation fields in work places vary depending upon the radioactive materials being handled, radiation- producing devices in use, and the facility design The radiation field can consist of particles and photons, individually or in combination The energies present are characteristic of the particular radionu- clides or devices that produce the radiations and can be modified by radiation interactions

Each instrument has a response characteristic for the various types of ionizing radiation that is determined by its design However, this response may be different for each instrument design In addi- tion, a given design may show variable response with radiation energy as well as with radiation type As a result, there may exist

a n inconsistency of response among instruments and uncertainty regarding the response of a given instrument This produces a num- ber of concerns, which can be summarized as follows:

1) limited ability to relate the reading of a survey meter to that of

a n alternative dose-measuring instrument or device; proper calibra-

1.3 NEED AND INTENT / 5

tion of the instrument and a thorough understanding of its response characteristics can reduce such discrepancies;

2) different responses of differently designed instruments in the same radiation field;

3) inconsistent response of a given instrument in fields of different intensity (see Section 2.5.2 for definition of intensity)

4) energy and geometry dependence, and

5) the limited ability to repeat accurately surveys for comparative purposes due to inappropriate changes in response with changing field conditions, including intensity and radiation type

Thus, the selection and use of radiation detectors zind instruments require detailed knowledge of their response characteristics as well

as judgment in their application Traditionally, radiation protection personnel, on the basis of their experience, have developed "rules- of-thumb", "favorite instruments", and unique techniques for specific situations However, because instrument responses can vary widely with radiation type or energy and with source-detector geometry,

it is not unusual in complex, mixed-field situations for personnel dosimeter results to differ considerably from what is expected on the basis of instrument measurements This uncertainty may lead protection personnel to apply the most dose-restrictive interpretation

to instrument readings, and this results in significant conservatism

in the application of radiation exposure control techniques Recent recommendations of the American National Standards Institute (ANSI, 1989a;1989b) deal with performance specifications for instru- mentation and should have a beneficial impact on the design and operation of portable instruments

In view of the large number and variety of instruments available and the sometimes specialized applications of these instruments, there will be situations in which the recommendations given in this report will not apply or will not be inclusive, or will require modifications Absolute calibration requirements are not recom- mended This is to recognize specialized needs and to allow for the fact that, with due attention to the response characteristics of a particular instrument in a particular situation, acceptable calibra- tions can be performed using approaches different from those recom- mended in the Report

This Report provides means for achieving greater consistency in the evaluation of instrument response Improved calibration should provide improved knowledge of instrument response, which will allow for a better choice of instrument, better determination of effec- tive dose equivalent, and reduction of unnecessary exposure

6 1 1 INTRODUCTION

Various groups and organizations have made recommendations regarding instrument calibration; their work forms the basis for many of the recommendations given

1.4 Review of Current Efforts/Recommendations

Various national and international standards and handbooks have been written to establish performance specifications and calibration requirements for health physics instrumentation; among those cited

a s references for this Report are the following:

1) ANSI Report No N323, Radiation Protection Instrumentation

Test and Calibration, 1978;

2) ANSI Report No N320, Performance Specifications for Reactor

Emergency Radiological Monitoring Instrumentation, 1979;

3) ANSI Report No N42.17A, Performance Specifications for Health Physics Instrumentation-Portable Instrumentation for Use

in Normal Environmental Conditions, 1989;

4) ANSI Report No N421.17C, Radiation Znstrumentation Perfor-

mance Specifications for Health Physics Instrumentation-Portable Instrumentation to Use in Extreme Environmental Conditions, 1989;

5) IAEA Technical Report No 133, Handbook on Calibration of

Radiation Protection Monitoring Instruments, 1971;

6) IAEA Technical Report No 285, Burger, G and Schwartz, R.B.,

Guidelines on Calibration of Neutron Measuring Devices, 1988; 7) IS0 Report No 4037, X and Gamma Reference Radiations for

Calibrating Dosimeters and Dose Ratemeters and for Determining their Response as a Function of Photon Energy, 1979;

8) I S 0 Report No 6980, Reference Beta Radiations for Calibrat-

ing Dosimeters and Dose Ratemeters and for Determining Their Response as a Function of Beta Radiation Energy, 1984;

9) I S 0 Report No 7503-1, Evaluation of Surface Contamination- Part 1: Beta Emitters (Maximum Beta Energy Greater than 0.1 5 MeV) and Alpha Emitters, 1988;

10) IS0 Report No 8529, Neutron Reference Radiations for Cali- brating Neutron-Measuring Devices Used for Radiation Protection Purposes and for Determining Their Response as a Function of Neu- tron Energy, 1989;

11) IS0 Report No 8769, Reference Sources for the Calibration

of Surface Contamination Monitors-Beta Emitters (Maximum Beta Energy greater than 0.15 MeV) and Alpha Emitters, 1989; and

12) Lalos, G (Ed.), Calibration Handbook: Ionizing Radiation

Measuring Instruments, 1983; Calibration Coordinating Group,

1.4 REVIEW OF CURRENT EFFORTS/RECOMMENDATIONS / 7 Department of Defense Joint Coordinating Group for Metrology and Calibration (The Lalos reference is a comprehensive treatment of many aspects of calibration Unfortunately, as of this writing the document is no longer in print, and only a limited number of copies are available.)

The literature contains many additional papers and reports appli- cable to various aspects of radiation monitoring and calibration NCRP Report No 57, Instrumentation and Monitoring Methods for Radiation Protection, pertains to personnel monitoring and the use

of radiation survey instruments (NCRP, 1978) It includes some recommendations regarding measurement accuracy and survey instrument calibration NCRP Report No 47, Tritium Measurement Techniques (NCRP, 1976), relates exclusively to techniques for mea- suring tritium and provides guidance on the calibration of tritium monitors

The International Commission on Radiation Units and Measure- ments (ICRU) has published a large number of reports that relate to measurement and evaluation of ionizing radiation dose Many of these pertain to various aspects of calibration Among these are the following:

1) ICRU Report 12, Certification of Standardized Radioactive Sources, 1968;

2) ICRU Report 14, Radiation Dosimetry: X Rays and Gamma Rays with Maximum Photon Energies Between 0.1 and 50 MeV, 1969; 3) ICRU Report 20, Radiation Protection Instrumentation and Its Application, 1971

4) ICRU Report 26, Neutron Dosimetry for Biology and Medicine,

1977;

5) ICRU Report 34, The Dosimetry of Pulsed Radiation, 1982; 6) ICRU Report 39, Determination of Dose Equivalents Resulting from External Radiation Sources, 1985; and

7) ICRU Report 43, Determination of Dose Equivalents Resulting from External Radiation Sources-Part 2, 1988

Details of the above references can be found at back of Report The latter two reports provide useful information not only on characteristics of radiation protection instrumentation and some considerations in calibration, but also on the relationships among quantities important in dose assessment Some of the new quantities

(e.g., ambient dose equivalent and directional dose equivalent) which ICRU has defined for monitoring purposes are reviewed, and particu- lar interrelationships among quantities are described The informa- tion is important to individuals who are calibrating instruments in accordance with the ICRU recommended quantities These quanti-

of instrument calibration laboratories The American Association of Physicists in Medicine has been concerned with instrument calibra- tion for many years and oversees calibration accreditation of partici- pating laboratories The Health Physics Society has also initiated a calibration accreditation program that should provide needed ser- vices to the radiation protection community Requirements on dosim- etry and survey instruments to provide information from which organltissue doses can be estimated are becoming more severe Such information, obtained from instrument measurements, may consti- tute the only substantial basis for implementing sound radiation dose control procedures This serves to emphasize the need for better calibration and more complete knowledge of survey instrument responses to all radiations encountered in the workplace

digital) and human factor design features ( e g , weight, balance, size)

which affect the selection or desirability of particular instruments, while important, are not covered here

Most of the considerations in the Report apply to fixed radiation monitors as well as to portable instruments Fixed area detectors are frequently located on walls or other surfaces and may be mounted

in proximity to sources of radiation or in areas of generally high background radiation It may be difficult or impossible to carry out calibration of a fixed monitor in-situ; the detector may have to be removed from its normal location to a more convenient one for cali- bration If such a detector is normally cable-connected to a remote readout station, the same or equivalent cable and readout system should be used in the calibration process Because of the presence of potential radiation scattering materials close to a fixed monitor in the field, such a monitor may be exposed to both primary and scatter- degraded radiations during actual use If the detector in question exhibits an energy-dependent response, calibration in a laboratory setting may not assure accurate performance in the field if the energy

or angular distribution in the two situations are different Other features specific to the calibration of these monitors are not elabo- rated in this Report

2.2 Level of Calibration

2.2.1 General

Calibration refers to the determination and adjustment of instru-

ment response in a particular radiation field of known intensity

10 / 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

Some obvious factors which affect response, such as meter zero adjustment and battery condition, are necessarily considered in the overall calibration procedure Additional influencing factors, such

as energy dependence and environmental conditions, may require consideration in the calibration process, depending on the conditions

of use of the instrument Thus, the procedures required for calibra- tion may be more or less complex, depending on the need to assess the impacts of these influencing factors Three levels of calibration are defined; these are discussed below and identified as full character- ization, characterization for specific acceptance, and routine calibra- tion

2.2.2 Full Characterization

Full characterization of an instrument involves more than what

is normally required by users of instruments Routine calibration (See Section 2.2.4) often requires simply the determination of reading linearity when an instrument is exposed to a single radiation type

of specified energy Manufacturers of instruments and others may, however, have the need to characterize fully an instrument being supplied to users in the field Such characterization should include the following:

1) evaluation of the energy-dependence of the response of the instrument to the radiation types to which the instrument is intended to respond; note that response, as it applies to instrument calibration, is the quotient of the instrument reading by the true value of the quantity being measured;

2) evaluation of linearity of instrument readings;

3) evaluation of the effects of other ionizing radiation types which may be encountered in field use on the instrument reading;

4) evaluation of the effects of environmental influences, such as temperature, pressure, and humidity, on the instrument reading;

5 ) evaluation of the effects of nonionizing radiations, particularly

RF radiations, on the instrument reading;

6) evaluation of geotropic effects;

7) evaluation of the ability of the instrument to survive mechani- cal shock as might be encountered in field use;

8) evaluation of the dose rate-dependence of the response andlor dead-time characteristics; this is particularly important to avoid significant exposure when an instrument's response is depressed at high dose rates;

9) evaluation of the effects of other influencing factors, such as magnetic and electrostatic fields, and

2.2 LEVEL OF CALTBRATION / 11 10) evaluation of the angular response of the instrument, prefera- bly at an energy close to the minimum useful energy for the instru- ment

Presently, most manufacturers provide information relating to item (1) above for portable instruments used in air kermaldose mea- surements and for some instruments used in assessing alpha- and beta-emitting surface contamination A user may have to arrange for characterization with respect to additional items from the list given above

2.2.3 Calibration for Specific Acceptance

It may be necessary to use an instrument under specific conditions

of a non-routine nature, and calibration specific to that objective may

be required An example would be the intended use of an instrument

at temperatures higher than those encountered in general use Such

an application would require evaluation of the instrument response

at the anticipated temperatures Calibration might be carried out at the elevated temperature and, if the adjusted response is acceptable, the instrument approved for such use As an alternative to calibrat- ing the instrument at the elevated temperature, if the temperature dependence of response is known, the calibration reading at a lower temperature may be used to adjust to what would be expected at the higher temperature In these cases, a label should be applied to the instrument noting that it may not be suitable for other uses if this

is the case Alternatively, the instrument may be calibrated for routine use and its response then evaluated under the proposed use conditions If responses under routine and proposed use conditions are significantly different, a correction factor or chart should be supplied with the instrument for use under the proposed conditions ANSI, in report number N42.17C (ANSI, 1989a), discusses perfor- mance specifications for portable instruments that are to be used under extreme environmental conditions

2) use of the instrument for radiation energies within the range

of energies for which the instrument is designed;

12 1 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

3) use under environmental conditions for which the instrument

6) use of the instrument in a manner that will not subject the

instrument to mechanical stress beyond that for which designed Routine calibration commonly involves the use of one or more sources of a specific radiation type and energy (e.g., 137Cs or 6 0 C ~

photon-emitting sources for many photon air kerma- or exposure- or dose-measuring instruments) and of sufficient activity to provide adequate field intensities for calibration on all ranges of concern

2.3 Performance Check

Calibrations need to be carried out periodically as discussed in Section 2.8 In the interval between calibrations, however, the instrument user should validate acceptable operation by carrying out

a performance check This is merely intended to establish whether or not the instrument is operatinglfunctioning within certain specified, rather large, uncertainty limits Although the performance check may range from a crude determination that the instrument is responding to a source, to a more detailed determination, deviations

of + 20 percent from the expected reading are generally considered acceptable for a performance check The initial performance check should be carried out in the calibration laboratory following calibra- tion; the source should be held at a fixed and reproducible location and the instrument reading recorded The source should be identified along with the instrument, and the same check source should be employed in the same fashion to demonstrate the instrument's opera- bility on a daily basis when the instrument is in use

Beta- or gamma-radiation-emitting radionuclides are commonly used in sources for performance checking of beta- and/or gamma- radiation-measuring instruments The sources are often no more than a few hundred kBq in activity and produce a reasonable reading

on the instrument when held very close to the detector Some instru- ments use internally mounted sources that can be moved close to the detector by means of an external control Alpha-emitting radionu- clides are used as check sources for alpha radiation detectors Porta- ble neutron sources in fixed geometries or, at times, well-character- ized beams a t reactor facilities, are useful as check sources for neu-

2.4 PRECALIBRATION CHECK 1 13

tron-measuring instruments Tissue-equivalent proportional counters (TEPC) often use an internally mounted alpha-emitting source which serves as both a check source and a calibration source

It is sometimes convenient to have available more than one check source for use with a given instrument or with several instruments

of the same type In such situations, the reading of the instrument, when exposed to each such check source,should be evaluated in the calibration laboratory As above, the specific source must be identi- fied along with the appropriate reading of a given instrument

2.4 Precalibration Check

Before an attempt is made to calibrate an instrument, a series of simple operations should be completed to ensure proper condition of the instrument for calibration Although the exact checks to be made will vary with the design of the particular instrument, a number of these are common to most instruments These include checking for radioactive contamination, condition of the batteries, loose or broken parts, proper operation of the switches, and that the instrument zero can be adjusted in accordance with the manufacturer's instructions

2.5 Technical Considerations of Source Selection

Photon sources of the required energy spectra are provided by x- ray machines with specified filters or K-fluorescence radiators (below

300 keV) and isotopic sources, e.g., 137Cs and 6 0 C ~ for energies greater than a few hundred keV Beta radiation fields are not monoenergetic,

14 / 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

and the calibration sources are generally radionuclides mounted with thin coverings Recently, electron accelerators have been used

in an attempt to provide mono-energetic electron calibration fields for defining better the instrument response characteristics Neutron fields of particular energy distributions may be difficult to obtain, and the selection of sources may involve a combination of neutron generators, fission sources, and isotopic sources Sources appropriate for calibration of instruments to be used in surface contamination assessment include a variety of beta- and/or alpha-emitting radionu- clides

Specific sources and their characteristics are discussed in the sec- tions of the Report treating source selection

2.5.2 Field Intensity and Source Strength

For purposes of this report, field intensity is defined as radiation fluence (rate), radiation energy fluence (rate), or quantities derived from these, such as absorbed dose (rate) and dose equivalent (rate) Radiation field intensities necessary to evaluate instruments in the calibration process may require sources that yield absorbed dose rates or kerma rates from less than 0.1 Gy h-' to greater than 100

Gy h-' Source activity may range from about 10 MBq to more than

10 PBq While a source as large as 10 PBq would likely not be applied

to portable survey instrument calibration, it may be required for calibration of fixed area monitors intended for use in accident dosime- try Calibration sources for instruments intended to measure surface contamination commonly range in activity from 100 Bq to greater than 10 kBq Choice of the source andlor the calibration facility arrangement must take the intensities into account In addition, high enough intensities must be provided to evaluate instrument linearity and saturation characteristics

2.5.3 Source-Detector Geometry

A number of considerations must be taken into account in choosing

a source either to reduce or evaluate geometrical dependencies These considerations include whether to select point or distributed sources, the significance of angular response variations of the instru- ment, and the ability of the source to provide uniform irradiation over the detector volume With regard to the latter point, calibrations are often performed using sources that produce penetrating radiation fields whose intensities decrease with the inverse square of the dis-

2.5 TECHNICAL CONSIDERATIONS OF SOURCE SELECTION 1 15 tance from the respective source to the point of interest The question commonly arises as to how close a given detector may be to such a source and still yield a response equal to that estimated from the fluence rate at a point in the center of the detector volume In order

to provide a t least a partial answer to this question, the data of Table 2.1 should be useful The geometry factor G, given in the last column

of the table, represents the ratio of the average radiation fluence rate throughout the detector volume to the fluence rate at a point a distance L from the point isotropic source and at the center of the detector volume Both the diameter and detector height for the cylin- drical detector, and the diameter, for the spherical detector, are expressed in units of L The calculations done to obtain the table values are described in Appendix E; no radiation attenuation was assumed in the calculations The factor G represents a correction by which the fluence rate (or fluence rate-dependent quantity such as dose rate) at distance L should be multiplied to obtain the fluence rate (or fluence rate-dependent instrument reading) averaged over the detector volume, the latter result being the true value appro- priate for the calibration The variation of the value of G from unity provides an estimate of the magnitude of the systematic error expected in the calibration process if the fluence rate at distance L

is assumed to be representative of the fluence rate throughout the detector volume The tabulated G-values would apply to typical ion- ization chambers They would not apply to certain detectors that use spherical or cylindrical shells for purposes of modifying the incident radiation so that an enclosed detector would yield a particular response (eg., neutron dose-equivalent-measuring instruments with spherical or cylindrical moderators surrounding a thermal neutron detector)

The data in Table 2.1 show that for a right-circular-cylindrical detector irradiated with penetrating radiation from a point isotropic source on the central longitudinal axis of the detector so that radia- tion is incident on the flat detector face, the average fluence rate over the detector volume will be within 1 percent of the fluence rate at the detector center, if neither the detector diameter nor the detector height is more than 20 percent of the distance from the source to the detector center Similarly, for the cylindrical detector irradiated on its curved surface by a source on the transverse central axis, the average fluence rate and that a t the detector center will not differ by more than about 0.5 percent if neither the detector diameter nor height exceeds 20 percent of the distance from the source to the detector center; about the same agreement exists for the spherical detector

Additional corrections may be appropriate in the calculations for

particular detectors (e.g., corrections for volume occupied by the

16 / 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

TABLE 2.1-Ratio of average primary radiation fluence mte in detector volume to primary radiation fluence rate at center of detector volume (Distance from source to

center of detector volume = L)."

Results for Cylindrical Detector

Detector surface of Cylinder Cylinder = Average fluence rate in volume radiation incidence DiameteriL heightll Fluence rate at diatance L

"See Appendix E for additional description

collecting electrode in some detectors) It is recommended that the source-to-detector-center distance should be at least five times the maximum dimension of the detector for calibrations using sources of primary penetrating radiation whose intensity follows an inverse square relationship with distance from the source If a source dimen-

2.5 TECHNICAL CONSIDERATIONS OF SOURCE SELECTION 1 17

sion is larger than the maximum detector dimension (as might be the case when dealing with high activity planar sources used for some high-level calibrations) , the uniformity of the field over the detector volume depends on both the detector dimensions and the source dimensions For these situations the calibrator may have to make measurements to demonstrate acceptable uniformity over the detector volume However, if a given detector is placed at a fixed

distance from the surface of a distributed source of unattenuated radiation, the ratio of average fluence rate in the detector volume to the fluence rate at the fixed distance will be closer to unity than the same ratio for a point isotropic source exposing the same detector at the same fixed distance (This assumes usual calibration sources and detector geometries; the statement would not hold for an unusual source configuration such as a curved surface, concave toward the detector.) This observation is based on the fact that for distributed sources and detectors of common geometries, the distributed source has relatively more of its activity further removed from a reference point in the detector volume (e.g., the center point) compared to the point source The greater such distance is, the smaller will be the difference between the fluence rate to that point and any other point within the detector volume

2.5.4 Traceability of Source Calibration

It is common practice to make use of a recognized standards labora- tory to provide necessary references for establishing the calibration fields This is accomplished in several ways:

1) sources are sent to the National Institute of Standards and Technology1 (NIST) for calibration:

2) instruments are sent to NIST for calibration; these instruments are then used to calibrate the facility sources/fields;

3) sources or instruments are sent to a Secondary Calibration Laboratory for calibration

NIST is the Primary Calibration (Standards) Laboratory in the U.S.A Other countries maintain and operate similar laboratories Secondary Calibration Laboratories are laboratories which partici- pate in formal programs involving comparative measurements with the primary laboratory; these programs are used to establish and demonstrate an acceptable degree of quality and consistency of per- formance on the part of the secondary laboratories Secondary cali-

'Formerly known as the National Bureau of Standards

18 / 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

bration laboratories may exist among the private, federal, and state sectors and may offer sewices to various groups within their respec- tive domains It is likely that tertiary calibration laboratories will also be established in the near future Such laboratories would be accredited through cooperation with secondary calibration labora- tories and would have demonstrated a satisfactory level of compe- tence and equipment to perform valid instrument calibrations Natu- rally, the further traceability is removed from the Primary Calibra- tion Laboratory, the greater will be the uncertainty associated with calibration accuracy Figure 2.1 is a schematic diagram of the tri- level measurement support system common in the U.S.A (Eisen- hower,1982) Tertiary-level laboratories would lie between second- ary-level and field-use level on the figure Figure 2.lb includes a description from Lalos (1983) of the hierarchy of standards

The International Atomic Energy Agency has discussed the estab- lishment, development, status and future trends of Secondary Stan- dards Dosimetry Laboratories in the IAEAIWHO network (IAEA, 1985)

2.5.5 Accuracy of Calibration Source for Field Intensity

Determinution

Measurement uncertainties may be introduced at every step in the calibration NIST typically provides standards of radioactivity, calibrated in terms of radioactivity or radiation emission rate, with uncertainties on the order of one to two percent Similar uncertaint- ies apply to NIST sources of x rays and gamma rays calibrated

in terms of exposurelair kerma rate Uncertainties in NIST beta- emitting sources calibrated in terms of absorbed dose are typically

5 to 15 percent Uncertainties in calibrations made at Secondary Standards Laboratories will likely be two or more times greater than those of NIST A facility laboratory dependent on a secondary laboratory for calibration will commonly operate with uncertainties

in its standards which are greater than those of the secondary labora- tory Thus, uncertainties on the order of 10 percent are not uncom- mon for such facilities although uncertainties on the order of 4 to 8 percent may be achievable (See Figure 2.lb)

Except for national standard sources maintained by the NIST, all other standard sources or instruments fall into a category denoted

as transferred standards This implies that standardization (calibra- tion) has been performed through a transfer process in which the instrument or source of concern has been standardized through a measurement made using a standard maintained by NIST Second-

2.5 TECHNICAL CONSIDERATIONS OF SOURCE SELECTION 1 19

(b)

Typical

1 National Standard x 1-2 Includes uncertainty on

y 1-2 physical constants necessary

p 1-2") to determine the quantity; 5-10'~' represents latest state-of-the-

a 1-2'') art measurements

n 1-5

2 1) Primary standard of x 2-5 Instrument and source

Secondary Standards -y 2-5 manufacturers desiring

2) Primary standard of a 2-10 comparable to the Seconda organizations needing n 5-15 Level labs in the Federal an? the highest level in- state sectors will need the

available to them

3 1) Primary standard for x 3-15 These standards measured

Field Level labs -y 3-15 by Secondary Standards labs 2) Working standard for p 10-20 could serve as their working Secondary Standards a 3-20 standards if they did not wish

standards for routine calibration

Constancy standard x 10-50 The absolute value of these

-y 10-50 standards is not as important

p 15-50 as being able to use them in

a 15-50 a stable manner, i.e.,

n 20-50 instrument position,

scattering, etc., remaining the same It may be necessary to make corrections for source decay These sources may be used

by personnel at any level to monitor equipment

~erformance

8 -

"'For radioactive sources calibrated in terms of activity or emission rate

or sources calibrated in terms of absorbed dose measured with an

extrapolation chamber

Fig 2.1 (a) Tri-level national measurement support system (Eisenhower, 1982)

and (b) hierarchy of standards (Lalos, 1983)

20 1 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

ary standards laboratories use transfer standards obtained through measurements performed against NIST standards; thus, calibration services provided by such laboratories are one step removed from NIST calibrations and will necessarily incur somewhat greater uncertainties in the results(See Section 2.7) Some of the calibration services provided by NIST are discussed in Special Publication 250 (NBS, 1985)

2.5.6 Incidental and Spurious Radiations

In the manufacture of sources it is possible to have contaminants For example, 134Cs is a common contaminant in 137Cs sources, and 146Pm is a common contaminant of 147Pm sources The different energy radiations from the contaminants can change or distort the calibration energy spectra even if such contaminants are present in small percentage amounts If the contaminant radionuclide is longer lived than the desired radionuclide, its relative importance increases with time Corrections of source strength through use of the half-life

of the presumed major radionuclide can also lead to error

Decay of a parent radionuclide to a radioactive daughter may also result in the production of undesired radiation Thus, 137mBa, the short-lived daughter of 137Cs, is a photon emitter which might pres- ent some interference if 137Cs was being used for its beta decay characteristics Yttrium-90 is the high energy beta-emitting daugh- ter of 90Sr;90Sr + sources are frequently used to provide high energy beta radiation for instrument calibration In this case, the lower energy beta particles from 'OSr are not desirable, and such sources are often covered with a n appropriate absorber to remove this interference Standards organizations such as IS0 commonly specify source coverings or encapsulations that will eliminate certain interfering radiations (ISO, 1984)

Reactor- and accelerator-produced radiation fields frequently con- tain undesired radiations Multiple energies of the radiation of inter- est may be present as well as radiations of different types, as exempli- fied by the common occurrence of gamma radiation in both accelera- tor- and reactor-produced neutron fields

2.6 Instrument Response Considerations

2.6.1 General

Each instrument exhibits a unique response to the particular radi- ation in question, depending upon the design of the detector and

2.6 INSTRUMENT RESPONSE CONSIDERATIONS 1 21 associated electronic readout systems It is essential to understand the factors that affect response of the instrument and its intended usage

Instruments that are read out in terms of air kerma rate, exposure rate or dose rate are calibrated typically in a field of known intensity

by adjusting the instrument to yield the proper reading An instru- ment that reads out in integral air kerma, exposure or dose units is calibrated by placing the instrument in a known radiation field for

a fixed time period; adjustments of the instrument are made until such an exposure produces the proper integrated reading

This Report does not address in detail the relationship between instrument reading and the dose equivalent (rate) a t particular depths of concern Section 5 dealing with calibration of beta-dose- measuring instruments discusses calibration for dose interpretation

at a depth of 0.07 mm below the surface of the body Most photon measuring instruments for dose assessment measure air kerma (rate) or exposure (rate) Current recommendations (ANSI,1983; ICRU, 1985) call for evaluation of the dose equivalent from penetrat- ing radiation at a depth of 10 mm; the shallow or superficial dose is

to be evaluated at 0.07 mm If dose to the lens of the eye is a concern, the assumed depth below the surface is 3 mm With respect to testing

of personnel dosimetry devices, ANSI Standard N13.11 (1983) includes a table for converting exposure units to dose equivalent at respective depths of 0.07 mm, 3 mm, and 10 mm for photon fields ranging in energy from 15 keV to 662 keV The values are based on calculations performed for unidirectional photon fields incident on a tissue equivalent 30-cm-diameter sphere For neutron testing with

a D20-moderated 252Cf source, the same ANSI standard uses compu- tations by Schwartz and Eisenhauer (1980) and calculated spectral data from IAEA (1978) and fluence-to-dose equivalent conversion factors from NCRP Report No 38 (NCRP, 1971)

The ICRU (1985) defines a quantity H * ( 4 called the ambient dose equivalent and H' (d), the directional dose equivalent which, for instrument measurement purposes, are intended to provide a link between the external radiation field and the effective dose equivalent and shallow dose equivalent, respectively The recommended values

of d, the appropriate depth in the body, are 10 mm and 0.07 mm, respectively for effective dose equivalent and shallow dose equiva- lent assessment (See glossary definitions and Section 6.3.1 for addi- tional discussion of effective dose equivalent, ambient and direc- tional dose equivalent.) It should be kept in mind that the dosimetric quantity of major interest when the body is irradiated with penetrat- ing radiation is the effective dose equivalent HE which is obtained

22 1 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

by multiplying the dose to each significantly irradiated tissue HT by its respective weighting factor wT and summing all such products:

This quantity cannot be measured, and other quantities such as H*(10) are used as practical alternatives to HE in measurement situations Such alternatives are acceptable if the dose equivalent values of these quantities are not less than nor significantly greater than the respective effective dose equivalents Radiation attenuation and geometry effects expectedly reduce the dose equivalent as a function of penetration depth in the body, and values of the dose equivalent at the 1 cm depth often overestimate the effective dose equivalent ICRU Report 39 (1985) and ICRU Report 43 (1988) pro- vide detailed information on the relationship between H*(10) and

HE when anthropomorphic phantoms are irradiated with photons or neutrons of varying energies When a significant portion of the body

is irradiated more or less uniformly with low penetrating radiations, instrument measurements of dose equivalent a t a depth of 0.07 mm often provide good estimates of the dose equivalent to the skin Instruments calibrated in terms of the quantity H'(0.07) are com- monly acceptable in such applications

The calibration procedures recommended for instruments intended for measurements of air kerma (rates) or exposure (rates)

in photon fields involve establishing the instrument response with respect to these quantities The instrument user has the option of transforming field measurements to dose equivalent values at spe- cific depths using available information such as that noted above Instruments which read out in dose equivalent units should be cali- brated to read dose equivalent Popular neutron dose-measuring instruments fall in this category; in Section 6 the fluence-to-dose equivalent conversion factors evaluated by NCRP (1971; 1987) are recommended

Examples of instrument response characteristics which should

be considered and understood for each instrument as calibration is planned and performed are given below Some of these are discussed

in more detail in later sections of the Report

2.6.2 Energy Dependence

A complete calibration to define the response characteristics of an instrument includes data within the entire anticipated energy range Most instruments and detectors tend to exhibit the greatest energy

2.6 INSTRUMENT RESPONSE CONSIDERATIONS 1 23

dependence to lower energy radiations (eg., photons below 200 keV, beta particles below 500 keV and neutrons below 1 MeV) Knowledge

of the magnitude of the energy dependence is essential when the same instrument may be used to assess various source fields or when scattering and shielding in the workplace can change the spectra of the radiation from location to location Such knowledge is of limited use if the spectral quality of the radiation field is unknown

2.6.3 Directional or Angular Response

Many instruments and detectors exhibit responses which are dependent on the angle of incidence of radiation on the instrument

or detector The size, shape, and chamber construction each affect the angular response A complete calibration includes evaluation

of the response of each instrument to radiation incident upon the chamberldetector from different angles This information is essential

in evaluating the response of the instrument to various field situa- tions

Angular responses are frequently presented as normalized values, referred to the response of the instrument in a given orientation in the radiation field While, in principle, any angular orientation may

be selected as the reference orientation, a logical choice is the most likely orientation of the instrument in actual field use when the instrument user approaches a radiation source For example if, in common usage, an instrument with a cylindrical detector is exposed

to radiation incident normally on the flat face of the detector such

an orientation might be preferred as the reference orientation in the calibration field This position would be the 0" orientation and other angular responses would be evaluated by proper rotation of the instrument in the calibration field

2.6.4 Detector Wall Erects

The design of the instrument chamber can have significant effects

on the response characteristics For photon fields, the wall thickness may be dictated by the requirement that equilibrium of the second- ary electrons be achieved in the chamber walls; the wall thickness should be at least as great as the range of the most energetic electron likely to be produced within it Since the energy of the secondary electrons is directly related to the energy of the incident radiation, the equilibrium thickness of the wall for the radiation to be measured should be evaluated as well as the magnitude of effects resulting

24 / 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

from a design that does not provide equilibrium For radiations of limited penetrating power such as beta radiation, detector walls may produce significant attenuation effects Geiger-Mueller detectors intended for dose-related measurements in photon fields often employ rather thick metallic shields to reduce an inherently exces- sive response to low energy gamma or x rays Such shields may be the cause of high readings, compared to expected values, when the detectors are placed in high-energy photon fields, and pair-produc- tion interactions in the wall-shields are significant

2.6.5 Geotropism

The design of some meter movements can result in deviations of the readings based on meter orientation Significant deviation is considered a design deficiency, but should be evaluated prior to cali- bration Criteria considered in the ANSI Standard N42.17A include specifications for acceptable changes in instrument readings due to changes in orientation of the instrument, independent of the radia- tion field (ANSI, 1989b)

For most portable instruments, the geotropic effects criteria of ANSI would result in a change of not more than a few percent in the instrument reading If this information is not available from the manufacturer, tests should be performed to confirm that the geo- tropic effects do not exceed the requirements for the intended use of the instrument

2.6.6 Environmental Effects

The typical instrument is designed to function within specific per- formance criteria through a range of environmental conditions A complete evaluation of an instrument will include the evaluation of its response under environmental conditions even outside the design ranges Several environmental conditions that can have significant effects on the operation of survey instruments are temperature, humidity, ambient pressure, RF and microwave fields, magnetic fields, and electrostatic fields

For many of the instruments currently in use, information regard- ing the effect of environmental conditions may not be available from the manufacturer If a n instrument is to be used under unusual environmental conditions, the calibration facility must evaluate instrument response under conditions which will simulate those expected in the field

2.6 INSTRUMENT RESPONSE CONSIDERATIONS 1 25

2.6.7 Influence of Other Ionizing Radiations

Though each instrument will typically be designed for a specific application and/or service with specific radiation types and energies

in mind, other radiations can have an effect on the accuracy in mixed fields The contribution to the reading from non-design radiations may be small, but it may have a major confusing effect in some applications, particularly if such contribution is unknown and/or unanticipated Examples of this are:

1) neutron-induced reading of an instrument designed primarily for gamma radiation;

2) gamma-induced reading of an instrument designed primarily for neutrons;

3) alpha-induced reading of a thin-window detector used to mea- sure beta radiation, and

4) beta-induced reading of a n instrument designed for gamma or other penetrating radiation

2.6.8 Linearity Measurements in Calibration

A knowledge of the response of each instrument for a wide range

of dose rates is important The ideal relationship between instrument reading and dose rate is linear, and deviations from linearity should

be known Evaluation of linearity should be carried out for all scales

on which a particular instrument will be used Characteristics of the detector and of the associated electronics can affect linearity Non- linear readings may be outside the acceptable accuracy limits and make an instrument unsuitable for general use

For instruments with linear readout scales, calibration should include response evaluations and adjustment for a t least two points

of each scale to be calibrated The response points should be separated

by at least 40 percent ofthe full scale range and should be represented

by points approximately equidistant from the mid-point of the scale Acceptable results at the two points provide reasonable assurance of

a linear response over the range of values covered by the selected scale This procedure is reasonable for both analog display instru- ments and digital display instruments which have selectable or auto- matic scale-switching

For analog instruments with multiple-decade-log-scale displays and for digital instruments with no scale selection, a t least one point

on each response decade should be used in calibration The end point

of a decade may be easier to read than a mid-point and such would

be acceptable for calibration purposes

26 / 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

2.6.9 Calibration on Selected Scales and Limited Ranges

Some instruments have selectable scales or single ranges which represent intensities known to be greater than those which will be encountered in field use In such cases, it is considered acceptable to perform calibrations a t intensities which include the highest intensi- ties that could be encountered in field use and to exclude calibration

a t higher levels The instrument should have a label affixed to it to inform the user that the instrument is not calibrated on specific ranges or above a particular intensity

2.7 Uncertainty in the Calibration Process

2.7.1 General

The accuracy of an instrument undergoing calibration is a measure

of how close the reading is to the expected (true) value of the quantity being measured The accuracy attainable in a given calibration pro- cedure depends on the characteristics of the radiation source(s)/ field(s) used and the response characteristics of the instrument being evaluated

The ultimate aim is to provide a calibration with sufficient accu- racy that when the instrument is put into field use, its reading will yield an acceptably accurate estimate of the desired quantity ( e g , air kerma (rate), contamination level) For a n instrument whose response is independent of energy and fieldldetector geometry, rou- tine use may yield measurements with accuracies close to those demonstrated in calibration In other instances, a n accurate calibra- tion may not guarantee an acceptably accurate measurement in field use

In Report No 57, the NCRP recommended that instruments used for radiation protection purposes be calibrated to an accuracy of

f 5 percent (NCRP, 1978) Because of uncertainties in calibration standards and because of possible adverse effects of certain influenc- ing factors attendant to the use of these standards, such accuracies may not be achievable This is particularly true of beta- and neutron- dose-responding instruments Recommendations of accuracy to be achieved in particular calibrations are presented in subsequent spe- cific sections of this Report and are summarized in Table 2.3 a t the end of Section 2

NCRP Report No 57 also recommended that when projected doses are near the level of the maximum permissible dose, a field measure-

2.7 UNCERTAINTY IN THE CALIBRATION PROCESS 1 27

ment accuracy of -e 30 percent should be achieved; for projected

doses less than 25 percent of the dose limit, inaccuracies on the order

of 100 percent are acceptable; for projected doses significantly above

the dose limits, accuracies of + 20 percent are recommended (NCRP,

1978) Again, these accuracies may not be easily achievable, depend- ing on the calibration accuracy achieved and the influences of the radiation field and other physical factors on the instrument response The dose limits which apply to non-occupational exposures are con- siderably more restrictive than occupational limits [e.g., NCRP

(1987) recommends an annual dose limit of 1 mSv for members of the public and a monthly limit of 0.5 mSv to a fetus] Measurements

using portable instruments for dose projections with those limits will

have large uncertainties; inaccuracies exceeding 100 percent would

not be unusual but may be acceptable for radiation control purposes

It may be possible to improve field measurement accuracies by altering or extending the calibration procedure For example, cali- bration with larger area sources may be a technique for reducing the error in the interpreted dose rate from distributed beta radiation sources

The user must be aware of the response characteristics of an instru- ment in order to make reasonable estimates of the expected measure- ment accuracy based on the calibration accuracy and the differences between calibration and actual field conditions Such estimates in the workplace can be difficult, especially if the radiation field condi- tions are not well known or are variable If such factors as instrument energy dependence and fieldldetector geometry dependence are known, it may be possible to make estimates of the maximum errors likely in field measurements when the calibration accuracy is speci- fied under known conditions For example, if an air kerma rate-

measuring instrument is calibrated with 137Cs 662 keV photons to

a n accuracy of + 10 percent on all ranges, and it is known that the instrument photon response does not vary by more than + 20 percent

(relative to the response a t 662 keV) for the range of energies to be

encountered in the field, and if it is known that no significant field/ detector geometry dependence exists, field measurements are likely

to be accurate to within + 30 percent

Within the major sections of this Report, recommendations are made with respect to accuracies applicable to calibration, and some discussion of field measurement accuracies is included At times, the user will have to assess the accuracy which might be expected in a given field situation If measurements are being made for dose con- trol purposes, such assessments should err on the conservative side

to ensure that personnel exposures are below specified limits Results

of experimental testing have been reported (Swinth et al., 1988)

28 1 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

that show that photon-measuring instruments which pass specific performance criteria discussed in ANSI N42.17A (ANSI, 1989b) can achieve accuracies within + 30 percent

The entire subject of accuracy in field measurements is very impor- tant and could be the subject of a separate report The present Report provides limited discussion of this topic in relation to the calibration process, but no attempt is made to treat the topic in detail

While the accuracy of any instrument measurement is judged by the extent to which the mean reading deviates from the true value

of the quantity being measured, the precision associated with the measurement provides an indication of the reproducibility of the measurement The precision associated with a group of repetitive measurements made under the same conditions refers to the close- ness of agreement among the measurements High precision is asso- ciated with tightly grouped measurement values while low (poor) precision implies a wide spread of measurement values Random variations embodied in the measurement process are considered in quantifying precision Such random variations may be treated by standard statistical techniques as discussed below

Other uncertainties that are not estimable by usual statistical methods can affect the overall uncertainty in a calibration result Such uncertainties, frequently referred to as systematic, may result from a number of causes, such as a miscalculation or an erroneous measurement of field strength, certain errors in reading a n instru-

ment scale ( e g , parallax reading error), and errors in measuring a

source-to-detector distance because of a misjudgment as to the posi- tion of the source or the center of the detector Systematic errors

of this type individually result in measurements that are either consistently high or low To the extent possible, such errors should

be eliminated by thorough investigation and correction of the errors Even when these biases are eliminated, if they can be, other uncer- tainties of a systematic nature may persist Uncertainty in the half- life of a source radionuclide, inability to read a distance scale or instrument scale exactly or to measure time without error, in the case of a n integral measurement, will introduce uncertainties in the calibration Uncertainties of this type may just as likely be positive

as negative In general, it may be impossible or unrealistically diffi- cult to evaluate exactly all systematic errors that might affect a measurement; in such cases, the maximum values of such errors should be estimated

The overall uncertainty " of a reported value refers to its likely inaccuracy in terms of credible limits, and combines both (1) compo- nents based on data that are amenable to statistical treatments and (2) components due to systematic errors that cannot be treated

2.7 UNCERTAINTY IN THE CALIBRATION PROCESS 1 29 statistically" (NCRP Report No 58,1985) The appropriate method

to be used for combining random and systematic errors is not always clear and is a subject of some debate The ICRU suggested that the overall uncertainty be expressed as "the arithmetic sum of the uncertainties due to random and assessable systematic errors ." (ICRU,1968) In 1980, representatives from eleven national stand- ardizing laboratories met at the International Bureau of Weights and Measures (BIPM) as a Working Group on the Statement of Uncertainties This group concluded that persistent systematic uncertainties behave as do random uncertainties and, with sufficient methods, their stochastic nature would be evident It was recom- mended that systematic uncertainties be measured by quantities u; which are interpreted as estimates of the respective variances; the quantities uj are treated as if they are standard deviations The general laws governing propagation of errors are assumed to apply

to both random and systematic uncertainties in the same fashion There is an ongoing effort by a working group of the International Standards Organization to promote international adoption of this approach In this Report the NCRP has adopted some of the recom- mendations of the BIPM working Group with respect to the treat- ment of systematic uncertainties in calibration The assumption is made that all identifiable biases in the calibration process have been corrected The use of confidence levels and, by inference therefore, the quantities of concern, associated with calculated uncertainties, are assumed to have values that are normally distributed (The BIPM Working Group on the statement of uncertainties has recommended against the use of confidence levels when quantities are not normally distributed.) With respect to systematic uncertainties, as they are discussed below, the value of u (the apparent "standard deviation" representative of the systematic uncertainty) has been estimated as

113 the value of the estimated maximum systematic uncertainty For purposes of defining the "95 percent confidence level", -+ 2 u has been selected as the range of uncertainty about the mean This selection is somewhat arbitrary but, given the frequently indefinite magnitudes of systematic uncertainties, more precise specification

is not necessary The use of quotation marks around the phrases

"standard deviation" and "-confidence interval" will be used in this report to indicate that the quantity referred to includes systematic uncertainties as described above

The random error is assumed to be the value tul where ul is the standard deviation in the mean value (also known as the standard error), and t is the Student's t-factor The table value oft for a given number of measurements is associated with a particular confidence

30 1 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

level, as given in Table 2.2 For example, for a set of 10 measurements (n = 10) there is a 95 percent confidence that the true mean value falls in the interval I & 2.2% (Note that here and in the discussion below, the symbol a is used in relationship to errors associated with

a finite number of measurements; it represents an estimate of the standard deviation and is sometimes denoted by 6 or s) If 6 is the estimated maximum magnitude of the systematic error, the value of

u is estimated by

and the overall uncertainty is given by

The value of k, at the "95 percent confidence level", has been taken

as 2

2.7.2 Uncertainty Associated with Random Variations

In order to assess the overall random uncertainty associated with the determination of the magnitude of a particular quantity, it is necessary to consider the random errors which contribute to that uncertainty For example, in determining the accuracy of a calibra- tion, evaluate the reading of an instrument relative to the estimated true value of the quantity being measured by dividing the reading

TABLE 2.2-Value of Student's t-factor to yield a giwn probability that the true

value X will be included in the confidence interval xk t a;

Probability

Degrees of

Freedom, n-1 0.50 0.90 0.95 0.99

2.7 UNCERTAINTY IN THE CALIBRATION PROCESS 1 31

by the true value The response ratio R and its overall random uncertainty, UR, may be expressed

-

R 2 uR=(I 2 ui)/(T r uT), (2.1) where R is the ratio of the mean instrument reading, 1, divided by the true (calibration) value, T UR is calculated from propagation

of errors (Equation 2.4), ul is the standard deviation in the mean instrument response, UT is the standard deviation in the calibration value, and UR is the standard deviation in the ratio, R

The ratio R is commonly defined as the instrument response The value of uI can be determined by making several measurements with the instrument positioned at a fixed point at which T has been determined The estimate of the experimental standard deviation of

a single measurement based on n measurements is

(Ii - fi2/(n - I), (2.2)

and the standard deviation in the mean of n measurements is esti- mated as

The appropriate response value to report then is

The standard deviation in the mean is calculated and

32 1 2 CONSIDERATIONS IN THE CALIBRATION PROCESS

implicit assumption has also been made that the distribution of R is normal; if this were not true, it would not be appropriate to specify the confidence limits a s has been done above

2.7.3 Uncertainties Associated with Systematic Errors

In reporting overall uncertainty a s discussed in Section 2.7.1, sys- tematic uncertainties associated with a particular quantity are treated here similarly to the random uncertainties The implicit assumption being made is that all identified biases have been elimi- nated and that remaining systematic uncertainties behave as if they had a random nature

For demonstration, consider t h e example discussed i n Section

2.7.2 Suppose that, for the instrument being calibrated and the

2.7 UNCERTAINTY IN THE CALIBRATION PROCESS 1 33

conditions of the calibration, a maximum relative systematic error

of + 4 percent in the instrument reading was estimated If there was also a systematic error assessed a t the time that the transfer standard was obtained, such an error would be included in the deter- mination of overall uncertainty For this example, assume a maxi- mum relative systematic error of + 3.3 percent in the use of the transfer standard employed to assess the true air kerma rate The estimated relative "standard deviation7' in the instrument reading and that associated with determination of the true value by use of the transfer instrument are denoted as uilf and %IT, respectively, and are estimated as (see Section 2.7.1)

The relative "standard deviation" in the response R is obtained by the usual method for handling the propagation of errors thus,

The value of uR is

An uncertainty of 2 2uR has been chosen to be associated with the

"95 percent confidence level" (See Section 2.7.1)

From Section 2.7.2 the random error contribution to uncertainty

in the value of R was 0.072 a t the "95 percent confidence level" In the determination of R, then, the overall uncertainty a t the "95 percent confidence level7' is reported as

Appendix F provides information on sources of systematic uncertain- ties and includes a n example illustrating the propagated influence

of such uncertainties on the value of R

2.7.4 Instrument Stability

The quantity R represents a measure of the accuracy with which the test instrument reads in the calibration field [Frequently the quantity R - 1 (i.e., the deviation from the ideal response) is quoted