Metrology in Industry
The Key for Quality
French College of Metrology
Series Editor
First published in Great Britain and the United States in 2006 by ISTE Ltd Translated into English by Jean Barbier
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address:
ISTE Ltd ISTE USA
6 Fitzroy Square 4308 Patrice Road
London W1T 5DX Newport Beach, CA 92663
UK USA
www.iste.co.uk
© ISTE Ltd, 2006
The rights of the French College of Metrology to be identified as the authors of this work has been asserted by them in accordance with the Copyright, Designs and Patents Act 1988.
____________________________________________________________________ Library of Congress Cataloging-in-Publication Data
Metrology in industry : the key for quality / edited by French College of Metrology. p. cm.
Includes bibliographical references and index. ISBN-13: 978-1-905209-51-4
1. Quality control. 2. Metrology. I. Collège français de métrologie. TS156.M485 2006
620'.0045--dc22
2006003530
British Library Cataloguing-in-Publication Data
A CIP record for this book is available from the British Library ISBN 10: 1-905209-51-7
ISBN 13: 978-1-905209-51-4
Preface . . . 15
Foreword. . . 17
Chapter 1. Analysis of the Metrological Requirements Needed to Ensure Quality . . . 19
Jean-Yves ARRIAT and Klaus-Dieter SCHITTHELM 1.1. Introduction. . . 19
1.2. Definition of the objectives . . . 21
1.3. Choice of the method of measurement . . . 22
1.3.1. Accounting for the selection of the method . . . 22
1.3.2. Defining the method and the principle to implement . . . 23
1.4. Choice of the means of measurement . . . 24
1.4.1. Introduction . . . 24
1.4.2. Analysis of what is already available . . . 25
1.4.3. Assessment and acquisition of material . . . 26
1.4.4. Technical criteria . . . 27
1.4.4.1. Basic characteristics . . . 27
1.4.4.2. Comportment towards influence quantities. . . 27
1.4.4.3. Durability of the instruments used . . . 27
1.4.4.4. Homogeneity of the supply of instruments . . . 28
1.4.4.5. Quality of the supplier’s service . . . 28
1.4.4.6. Adaptation of the instrument . . . 28
1.4.4.7. Possibility of traceability . . . 29
1.4.4.8. Computerization and the speed of taking measurements . . . 29
1.4.4.9. Ergonomics . . . 29
1.4.4.10. Capability of measuring instruments. . . 29
1.4.5. Economic criteria . . . 30
1.4.6. Grid of the analysis of the choice . . . 31
1.4.6.1. Stage 1: primary technical requirements (unavoidably necessary) . . . 31
1.4.6.2. Stage 2: secondary technical requirements (desirable) . . . 31
1.4.7. Technical assistance for users of measuring instruments. . . 33
1.4.7.1. The EXERA (France) . . . 33
1.4.7.2. VDI/VDE-GMA (Germany) . . . 34
1.5. The traceability of the measurements . . . 36
1.5.1. The necessity of traceability of the measurements . . . 36
1.5.2. Calibration requirements . . . 38
1.5.3. The selection of standards . . . 39
1.6. Conclusion . . . 42
Chapter 2. Organization of Metrology: Industrial, Scientific, Legal. . . 43
Luc ERARD, Jean-François MAGANA, Roberto PERISSI, Patrick REPOSEUR and Jean-Michel VIRIEUX 2.1. A metrological organization: why? . . . 43
2.2. Metrology: how?. . . 45
2.3. Scientific and technical metrology . . . 47
2.3.1. The BIPM . . . 48
2.3.2. Results of the international activities . . . 50
2.3.3. Regional organizations. . . 51
2.3.3.1. EUROMET . . . 51
2.3.3.2. European Cooperation for Accreditaton (EA) . . . 54
2.3.3.3. Accreditation procedure . . . 58
2.3.4. Organization at the national level . . . 59
2.3.4.1. The Laboratoire National de Métrologie et d’Essais (LNE) . . . 59
2.3.4.2. The Italian national calibration system (SNT) . . . 63
2.3.4.3. The Swiss national calibration system . . . 65
2.4. Legal metrology . . . 67
2.4.1. Scope of legal metrology . . . 67
2.4.2. The International Organization of Legal Metrology (OIML) . . . . 68
2.4.3. The European level . . . 71
2.4.3.1. European Union harmonization . . . 71
2.4.3.2. WELMEC . . . 71
2.4.3.3. Other regional bodies . . . 73
2.4.4. At national level. . . 73
2.4.4.1. Legal metrology in Italy . . . 73
2.4.4.2. Legal metrology in Switzerland . . . 74
Chapter 3. Mastering Measurement Processes Approach to the Setting
up of a Metrology Function . . . 79
Marc PRIEL and Patrick REPOSEUR 3.1. What to do at the beginning? . . . 79
3.2. Goals and role of the measurement management system – metrological function. . . 80
3.3. The measurement processes . . . 86
3.3.1. Conception and development of a new measurement process. . . . 86
3.3.1.1. Analysis of the requirements . . . 86
3.3.1.2. Transcription of the characteristics of the product in “measurand” form or “characteristics to be measured” form . . . 87
3.3.1.3. The development of a measurement process can be managed as a project . . . 87
3.3.2. Exploitation of a valid process . . . 88
3.3.3. Continuous improvement of measurement processes . . . 88
3.4. Management of the measuring equipment (metrological confirmation) . . . 89
3.4.1. Analysis of the requirement and selection of the measuring equipments . . . 91
3.4.1.1. Technical requirements . . . 91
3.4.1.2. Economic and commercial conditions. . . 93
3.4.1.3. Assessment of the measuring equipment . . . 93
3.4.2. Receiving the measuring equipment and putting it into service . . . 93
3.4.2.1. Compliance with the order . . . 94
3.4.2.2. Identification of the measuring equipment . . . 94
3.4.2.3. Inventory (description). . . 94
3.4.2.4. Technical dossier of the equipment . . . 94
3.4.2.5. Technical documentation . . . 94
3.4.2.6. Basic definitions . . . 94
3.4.3. Calibration and verification operations . . . 97
3.4.3.1. Calibration or verification program . . . 99
3.4.3.2. Calibration or verification intervals . . . 99
3.4.3.3. Supervision of the measuring equipment . . . 100
3.4.4. Fitness for use of measuring equipment. . . 100
3.4.4.1. Freedom from bias, repeatability, stability . . . 100
3.4.4.2. Maximum permissible errors . . . 101
3.4.4.3. Demands for an assurance of the quality . . . 101
3.5. Setting up a metrological structure within the firm . . . 102
3.5.1. Analysis of the metrological requirements and setting up standards . . . 102
3.5.2. Traceability of the measuring instrument(s) to the firm’s reference standards . . . 104
3.6. Suggested approach for setting up a metrology function . . . 105
3.7. Bibliography . . . 106
Chapter 4. Handling of a Bank of Measuring Instruments . . . 109
Jean-Yves ARRIAT 4.1. Acquaintance with the bank . . . 110
4.1.1. Inventory . . . 110
4.1.2. Identification . . . 110
4.2. Metrological policy of the firm . . . 113
4.2.1. Objective and commitment of the firm’s management . . . 113
4.2.2. Plan of actions to launch. . . 113
4.2.3. Awareness, training and vocabulary . . . 113
4.2.4. Selection of the material to be followed periodically . . . 114
4.3. Drafting of the documents . . . 115
4.3.1. Codification of the documents . . . 115
4.3.2. Work instructions . . . 116
4.3.3. Result-recording documents . . . 117
4.3.4. Other documents . . . 118
4.4. Physical handling of the measuring instruments . . . 119
4.4.1. Receipt . . . 119
4.4.2. Transfer. . . 120
4.4.2.1. Traceability . . . 120
4.4.2.2. Transfer. . . 120
4.4.2.3. Precautions. . . 121
4.4.3. Storing and environment. . . 121
4.4.4. Maintenance . . . 122
4.5. Follow-up of the measuring instruments over time . . . 123
4.5.1. Periodicity of the follow-up . . . 123
4.5.2. Campaign of recall . . . 124
4.5.3. Follow-up of the results . . . 125
4.6. Software for the handling of the means of measurements . . . 125
Chapter 5. Traceability to National Standards . . . 127
Luc ERARD and Patrick REPOSEUR 5.1. Introduction. . . 127 5.2. Definitions . . . 127 5.2.1. Traceability . . . 127 5.2.2. Calibration . . . 128 5.2.3. Verification . . . 129 5.3. Traceability chains . . . 129 5.4. Traceability . . . 131
5.5. Calibration . . . 132
5.5.1. Calibration in an accredited laboratory . . . 132
5.5.2. Calibration in a non-accredited laboratory . . . 132
5.6. Verification . . . 133
5.6.1. Verification in an accredited laboratory and in its accreditation scope . . . 133
5.6.2. Verification in a non-accredited laboratory or out of the accreditation scope. . . 133
5.7. Use of calibration and verification results . . . 133
5.7.1. Use of the results of a calibration . . . 134
5.7.2. Use of the results of a verification . . . 134
5.8. Particular cases. . . 135
5.8.1. “Self-calibrating” or “self-gauging” measuring instruments. . . 135
5.8.2. Complex instruments in which components/equipments and software are narrowly combined and large measurement ranges are covered for complex quantities. . . 136
5.9. Metrology in chemistry and physical methods of chemical analysis . . . 136
5.9.1. Traceabilty in metrology in chemistry. . . 137
5.9.2. Influence of the principle of the method . . . 139
5.9.2.1. Absolute methods. . . 139
5.9.2.2. Relative method. . . 140
5.9.2.3. Comparative method . . . 140
5.9.3. “Documentary” traceability . . . 141
5.9.4. Control of the reference materials . . . 143
5.9.5. Conclusion . . . 145
5.10. Assessment of traceability . . . 145
5.11. Bibliography . . . 146
Chapter 6. Calibration Intervals and Methods for Monitoring the Measurement Processes . . . 149
Patrizia TAVELLA and Marc PRIEL 6.1. Normative requirements . . . 149
6.2. Methods for monitoring the instruments in use – general criteria . . . . 150
6.2.1. First method: metrological redundancies . . . 150
6.2.2. Second method: checking the coherence of the results . . . 151
6.2.3. Third method: “monitoring standards” and statistical supervision of the measurement processes . . . 152
6.2.3.1. Statistical control of the measurement processes . . . 152
6.2.3.2. Control charts . . . 154
6.2.3.3. Use of the monitoring methods . . . 157
6.3. The determination of the calibration intervals . . . 158
Chapter 7. Measurements and Uncertainties . . . 163
Marc PRIEL 7.1. Introduction. . . 163
7.2. Measurement of physical quantity . . . 164
7.3. Analysis of the measurement process . . . 166
7.3.1. The cause and effect diagram method . . . 166
7.3.2. Using the list published in the GUM (section 3.3.2) . . . 167
7.3.3. Errors . . . 168
7.3.4. Cutting down the errors . . . 169
7.3.4.1. Cutting down random errors by repeating measurements . . . . 170
7.3.4.2. Cutting down systematic errors by applying corrections . . . 171
7.4. Modeling of the measurement process . . . 172
7.4.1. Measurement procedure and model of the measurement process . . 172
7.4.2. An essential stage for the assessment of uncertainty: modeling the measurement . . . 173
7.5. Assessment of the uncertainty of the input quantities . . . 174
7.5.1. Type A methods. . . 175
7.5.2. Type B methods. . . 176
7.5.3. Comparing the Type A and Type B methods . . . 179
7.6. Calculating the combined uncertainty on the result . . . 180
7.6.1. Situation when all the input quantities are independent . . . 180
7.6.1.1. Situation when the input quantities are independent and the model is a sum. . . 180
7.6.1.2. Situation when the model is a product . . . 181
7.6.2. Situation when the input quantities are dependent . . . 181
7.6.2.1. Assessment of the covariances by assessing a coefficient of correlation r ,
( )
xixj . . . 1817.6.2.2. Assessment of the covariances by calculating the terms of covariance . . . 181
7.6.2.3. Assessment of the covariances by considering the terms common to two input quantities . . . 181
7.7. Use of the performances of the method (repeatability and freedom of bias) to assess the uncertainty of the measurement result . . . . 183
7.7.1. Intra- or interlaboratory approaches . . . 184
7.7.2. Intra-laboratory approach . . . 185
7.7.3. Interlaboratory approach. . . 186
7.7.4. Data processing for intra- and interlaboratory approaches . . . 187
7.7.4.1. Assessment of the repeatability and the reproducibility . . . 187
7.7.4.2. Assessment of the freedom of bias (trueness) . . . 188
7.7.4.3. Evaluation of the linearity . . . 189
7.7.4.4. The terms i ( )i i x u c 2
∑
. . . 1897.8. Reporting of the measurement result . . . 189
7.9. Example . . . 190
7.10. Bibliography . . . 193
Chapter 8.The Environment of Measuring . . . 195
Jean-Yves ARRIAT and Marc PRIEL 8.1. The premises . . . 196
8.1.1. Ambient temperature . . . 197
8.1.2. Relative humidity . . . 198
8.1.3. Handling of the air conditioning systems . . . 199
8.1.4. Power network . . . 199
8.1.5. Radioelectric disturbances. . . 199
8.1.6. Measurements on-site . . . 200
8.2. The personnel . . . 200
8.2.1. The connection of metrology function . . . 200
8.2.2. Staff involved in the metrology function . . . 201
8.2.3. The qualification of the personnel . . . 202
8.3. The documentation . . . 202
8.3.1. Filing of the documents . . . 202
8.3.1.1. Documents dealing with the quality system . . . 202
8.3.1.2. Records regarding quality . . . 203
8.3.2. Management of the documents . . . 204
8.4. Bibliography . . . 205
8.5. Appendix . . . 206
Chapter 9. About Measuring . . . 209
Claude KOCH 9.1. Preliminary information . . . 209
9.1.1. Physical quantity . . . 209
9.1.2. The object to be measured. . . 210
9.1.3. Field of measurement . . . 210
9.1.4. Four types of uses of measuring instruments. . . 211
9.1.5. Influencing quantities . . . 212
9.2. Choice of a measuring principle. . . 213
9.2.1. Differential measurement . . . 214
9.2.2. Direct measurement . . . 214
9.3. Practicing in metrology . . . 215
9.3.1. Implementing the instruments . . . 216
9.3.2. Precautions before measuring. . . 216
9.3.3. Measurements . . . 216
9.3.4. Variations and their sign. . . 217
9.3.5. The time factor . . . 218
9.4. Expression of the results . . . 218
9.4.1. Graphs . . . 220
9.4.2. Histograms . . . 220
9.5. What qualities does a metrologist require? . . . 221
9.5.1. Be inquisitive . . . 222
9.5.2. Be tidy and methodical . . . 222
9.5.3. Be open to doubt . . . 222
9.5.4. Be observant . . . 222
9.5.5. Be honest. . . 223
Chapter 10. Organization of Metrology at Solvay Research and Technology. . . 225
José MONTES 10.1. Presentation of the company . . . 225
10.2. Organization of the metrology sector . . . 226
10.2.1. Creation . . . 226
10.2.2. Missions . . . 226
10.2.3. Organization . . . 226
10.2.4. Geographic localization of the activities . . . 227
10.2.5. Composition of the bank of measuring equipment. . . 227
10.3. Metrology . . . 228
10.3.1. Identification . . . 228
10.3.2. Connection of the standards . . . 228
10.3.3. Periodicity of the calibrations . . . 229
10.3.4. Calibration operations . . . 229
10.3.5. Documentation of the calibration results . . . 230
10.3.6. Verdict of the metrological confirmation . . . 231
10.3.7. Indication of the state of the calibrations . . . 231
Chapter 11. Metrology within the Scope of the ISO 9001 Standard . . . 233
Philippe LANNEAU and Patrick REPOSEUR 11.1. Introduction . . . 233
11.2. Introduction to the evolution of the standard . . . 234
11.2.1. The concept of continuous improvement . . . 234
11.2.2. The process approach. . . 235
11.3. Measurement control process . . . 236
11.4. The ISO 9001 (2000) standard step-by-step . . . 238
11.5. Conclusion . . . 245
Chapter 12. Training for the Metrology Professions in France . . . 247
Bernard LARQUIER 12.1. The metrology function in a firm’s strategy . . . 247
12.2. Metrology profession . . . 248
12.2.1. Metrological engineer . . . 249
12.2.2. Metrological technician . . . 249
12.2.3. Metrological operator. . . 250
12.3. Initial training. . . 250
12.3.1. Schools for engineers . . . 250
12.3.2. Courses for higher level technicians . . . 251
12.3.3. Vocational high schools . . . 251
12.4. Continuing education . . . 251
12.5. Long-lasting training courses . . . 253
12.6. The teaching of metrology in secondary schools . . . 265
12.7. Prospects for the development of long-lasting training courses . . . . 265
12.8. Bibliography . . . 266
The Authors . . . 267
Metrology is an essential part of the infrastructure of today’s world. It enters into our lives in a multitude of ways, some direct and some indirect. National and international trade increasingly require demonstrated conformity to written standards and specifications and mutual recognition of measurements and tests. The economic success of most manufacturing industries is critically dependent on how well its products are made, a requirement in which measurement plays a key role. Navigation and telecommunications require the most accurate time and frequency standards. Human health and safety depend on reliable measurements in diagnosis and therapy and in the production and trade in food and food products. The protection of the environment from the short-term and long-term destructive effects of industrial activity can only be assured on the basis of accurate and reliable measurements. Global climate studies depend on reliable and consistent data from many disciplines often over long periods of time and this can be assured only on the basis of measurements traceable to measurement standards that are themselves linked to fundamental and atomic constants.
Metrology is not an activity that is only carried out in specialized institutes or calibration laboratories. In order to meet the needs of society for accurate and reliable measurements in all its many applications, a strong spirit of metrology must also exist in companies and enterprises that make the instruments and that use them to make measurements.
For this reason I welcome this book. It gives a clear outline of the basic ideas of metrology, why we need it and how, in an enterprise it can be practiced. I wish it every success.
T.J. Quinn, Director of BIPM
Foreword
Technically, economically, commercially and, sometimes, statutorily speaking, having relevant and reliable results of measurements, analyses and tests is a real asset for a firm which wishes to make efficacious decisions.
You cannot achieve such an end if you do not have firm control over the processes of measurement, analysis and testing. Nowadays, however, the measuring techniques, the normative and statutory requirements, the methods of measurement uncertainty assessment or those to secure the traceability of measurements are all complex and it is more necessary than ever to integrate them into a network of competent bodies so as to exchange experience and information. It is on this fundamental principle that the Metrology College was created in 1986, which became the French College of Metrology in 2002. The purpose of this association is obviously much wider:
– to identify which firms and organisms’ needs are to be met from the angle of metrology;
– to spread metrological culture and knowledge through the industrial, scientific and economic fabric;
– to be a form of exchange between people involved in metrology;
– to contribute to make the collective national and regional actions coherent in this sphere;
– to perform any action likely to contribute to the development and promotion of metrology.
The permanent evolution of metrology, together with the willingness to impart all the knowledge acquired so far, have led a working party of the French College of Metrology to write a second edition of the book Metrology in the Firm. Metrologists from various callings (national metrology laboratories, accrediting organisms,
industrial concerns and consulting firms) and from different nationalities make up this working party. This broad range of authors gives the book a pragmatic characteristic and enables it to answer the questions and concerns of organizations, whether they be principals, small or medium firms, laboratories, etc.
The contribution from foreign authors gives the book an unquestionable international aspect which accurately reflects the current reality. More than ever, as a matter of fact, metrology contributes to the free circulation of goods between countries, thanks to the international organization of metrology and thanks to the international agreements between national metrology laboratories and between accrediting organisms.
Moreover, most of the authors belong to different national or international standardization committees. As a result, the latest normative evolutions are to be found in this book, whether it is the concept of firm certification developed in the 2000 version of standard ISO 9001, or the approach concerning the competence of activities of measurement, testing or analysis as expounded in standard ISO 17025.
Whether you are involved in your firm’s metrology function, or are simply interested in a concrete matter of measurement, analysis or testing, I am confident you will find here some clues which will help you progress and improve your processes.
The growing interest you have shown in this book has encouraged us in our intention of producing this English version. It is my sincere wish that whatever your need and country may be, you can get as much out of it as our French colleagues do.
May you enjoy reading it.
P. LEBLOIS, President of the French College of Metrology
Analysis of the Metrological Requirements
Needed to Ensure Quality
Anybody with a mind to implement (or improve) a metrology function might feel a bit panicky at the thought of all the work to be done if they read this book unwarned, and more particularly this chapter. Let the reader’s mind be put at ease first. All the content is not, fortunately, to be carried out literally. All we want to do is to offer as broad as possible a survey of the subject by pointing out practically all the items that require consideration.
And then, is it not normal to start wondering what one really needs?
Experience has taught us, too often alas, that this is not a natural process. Many industrial difficulties, or many costs, grow out of the inadequacy “means of measurement/real need”.
1.1. Introduction
Before we start any concrete action, it is primordial to analyze the metrological needs carefully. There are two kinds:
– The organizational needs for the management of metrology. Are those needs great enough to require the introduction of full-scale metrology? Are premises or qualified personnel needed permanently? What possibilities are there in the region?
Chapter written by Jean-Yves ARRIAT – Ascent Consulting – and Klaus-Dieter SCHITTHELM – Expert in Metrology, Germany.
Does someone want to manage metrology on his or her own, with the help of a someone else, or to handle it to a subcontractor?
– The material needs for the realization of the measurements. In order to realize measurements correctly, it is necessary to have appropriate means; these means are found after analysis of the objectives and the possibilities of the instruments and the connection. In order to define the firm’s needs, it is necessary to answer the following questions:
1. What are my industrial needs?
– What do I have to measure and what accuracy shall I expect? 2. How can I meet my needs?
– What are the possible measuring methods? – Which method and principle will be used? 3. Which measuring instruments can be used?
– Which instrument shall I use?
– Can the selected instrument ensure the required accuracy? 4. How is to be used the selected instrument?
– What assembly is to be set up and what procedure is to be followed? – What technical competence do you have to have to use it?
Then a question of a very different magnitude arises: how am I going to guarantee the quality of my measurements?
Setting up a metrological function
The three key components of a metrological function have to be under control (see Chapter 4):
– adequacy of means to needs;
– traceability of the means of measurement to international standards;
– administrative management of the equipment (measuring instruments, standards, etc.).
The preliminary analysis of the needs will produce a first set of specifications. There is a good chance that these analysis are going to be a bit theoretical and take little heed of the notions of profitability. You have to accept the principle which says that the specifications will evolve and obtain agreement from the major actors taking part in the drafting of the specifications.
For a new measuring instrument, all the stages from conception to utilization must be taken into account by the specifications. This is fundamentally the concern
of the manufacturers, but potential users may sometimes take part in the elaboration of the specifications.
The specifications for a new measuring laboratory must ignore all of the environmental characteristics of the measurement (see Chapter 8), and take into consideration the problems of maintainability (for instance, the maintenance of air conditioning), of access to the personnel, of user-friendliness, etc.
However big or small the problem is, one must always begin by analyzing one’s real meterological need.
1.2. Definition of the objectives
The metrological function must be approached as soon as you start thinking
about problems of measurement. Its role may depend on each particular firm (see Chapter 3), but its chief role is to act as a consultant. It examines the need in a logical process based on:
– the functional analysis of the measurement (drafting of specifications);
– the analysis of the achievement of the measurement results (and of the level of
accuracy reached);
– the analysis of the risks related to the selected means;
– the analysis of the non-conformities which could be encountered.
This process makes it possible to identify and quantify the means (personnel and material) to be implemented to take the intended measurements.
It is during these phases that the “tools of quality” will be used. Let us point out that the analysis of the value (fundamental at the outset) is among the most useful tools. In order to clearly define the objective, we strongly recommend to use “brainstorming”, cause/effect diagrams, Pareto, etc., which make analysis and collective participation easier.
So as to guarantee the quality of its measurements (i.e. a process of management
by quality), the firm sets up a real management of the means of measurement. For this purpose, the metrological function conducts the management of these means according to needs that are clearly defined and regularly updated. This involves examining a large number of actions in order to start up and maintain the supply of measuring instruments necessary to meet the firm’s needs.
The first thing to do regarding the analysis of the supply of material is to work out:
– the list of physical quantities (e.g. temperature, length, electric resistance, etc.); – the ranges which need to be covered for each physical quantity (e.g. length from 0.1 mm to 1,000 mm);
– the permissible uncertainty for each quantity and each range (the uncertainty in the 0.1 mm to 0.5 mm range will be different from the one which is expected between 100 mm and 1,000 mm).
Then, for each separate case, it will be necessary to consider and define: – the analysis of the needs and the choice of the means of measurement; – the acquisition, the reception and the implementation of these means;
– the traceability of the material of measurement (in the case where materials of measurement are assigned);
– the traceability of the measurements (which material do they come from?); – the calibration or the verification of the means and the decisions they entail; – the exploitation of the calibration results;
– the operations related to the moving of these means (protection, authorization, etc.);
– the updating of the inventory of these means.
The outcome of this is that the intended objectives must not be mixed up to satisfy: – the needs for the management of metrology with;
– the needs for the realization of the measurements.
1.3. Choice of the method of measurement 1.3.1. Accounting for the selection of the method
You have to justify the choice of the selected method. It is to be understood by this that the criteria have to possess as little subjectivity as possible.
This choice must take possible restraints of qualification into consideration. The fact is that within the scope of some contracts (notably related to safety, public security, health, etc.) you may have to qualify the method of measurement. This means it must be subjected to an authenticated description, officially certified tests, etc., in accordance with the relevant program and by a very precise process. Besides, the ISO/QS 9000 or TS 16 949 certification process also involves a description of the selected method.
Fortunately, it is often possible to hang on to the methods which are known and officially accepted. You must not forget that the great metrology laboratories can be a great help in this area. In France, for example, these are the laboratories of the LNE (Laboratoire National de Métrologie et d’Essais), and in Germany, those of the PTB (Physikalisch-Technische-Bundesanstalt), or calibration laboratories accredited by the DKD (Deutscher Kalibrierdienst).
Whether the method is qualified or not, it is important, after the metrological objectives have been set, to make the methodology of the measurement explicit. The different stages, the conditions of the material and the environment, the operations that make it possible to get the measurement, i.e. everything related to the carrying out of these measurements, must be written in a document and will be taken into account particularly when choosing the operators.
One of the very first principles of quality assurance is to write down what is being done. This process is simple and allows people to think further about the choice of the method. There must be a clear distinction between chosing a method and chosing a measuring instrument. For example, you may want to measure a dimension on a rubber part: you happen to be close to a three-dimensional measuring machine and your instant reaction may be to go to this machine without thinking whether there may be a more suitable method than this one.
1.3.2. Defining the method and the principle to implement
When there are several methods of measurement, it is often difficult to determine which one will best fit your need if you are not able to classify them.
Our advice is to keep only the two (maybe three) most important criteria in mind and to draw a table. Let us consider the example of Table 1.1. It makes it possible to analyze the different methods of measurement that lead to the assessment of the characteristics of industrial robots.
Two criteria have been selected:
– the principle of measurement (two groups of them here); – the characteristics measured (two families of them here).
As a rule, there are in metrology three great principles of measurement; the three of them have advantages and drawbacks. They are:
– differential measurement; – direct measurement; – indirect measurement. See Chapter 9 for more details.
Positioning characteristics Trajectory characteristics Local
methods
Measurement terminal with cubes (Peugeot SA and LNE) Measurement terminal on measuring machine (IPA) Different realizations based on the
same principles have been developed (IBM, General Motors,
etc.)
Measurement terminal with materialized trajectories
(rule and circle) (LNE) Measurement terminal with trajectory
(broken line) (Peugeot SA)
Big base
methods Method of the two theodolites (Renault)
Theodolites with automatic data (LNE)
Selspine system Photogrammetry (University of Dresden, NEL and SETP-LNE)
Devices with three sensors and wire (Peugeot)
Sweep of two laser beams (University of Surrey, England)
Selspine system Robotest (Polytech, FRG) Stroboscoped photogrammetry (University of Dresden, NEL and
SETP-LNE) IPA: Institute for Production techniques and Automation, Germany
LNE: National Testing Laboratory NEL: National Engineering Laboratory, England SETP: Photogrammetric Studies and Works Society Table 1.1. “Classification of the methods of measurement”
(Reproduced with the kind permission of Techniques de l'ingénieur – France)
1.4. Choice of the means of measurement 1.4.1. Introduction
The choice of the material and/or the equipment must be based on specifications. To make this choice, you must take into consideration:
– the technical needs;
– the possibilities of calibration; – the assessments already made;
– the economic conditions (last, for the technical specifications have to be seen first).
Practicing metrology is not simply doing plain measurements. To begin with, a special material has to be used, which means that you do not simply use any dimensional comparator lying about on a shelf, you do not borrow a frequency meter from a colleague and you do not hire a “lowborn” multimeter. On the contrary, you use instruments which are well-known and well-regarded, which come with documents and certificates, so as to be sure of their traceability and, therefore, to better guarantee the quality of the measurements.
These instruments (said to be “reference instruments”) have to be acquired after you have seriously studied the criteria of choice. It is known that:
– the ideal instrument does not exist;
– the instrument closest to what is ideal is too expensive; – each buyer limits the claims of technical applicants.
Moreover, the choice of an instrument depends on its type of use. Four types of utilization can be distinguished:
– for a study (you must look for an instrument that can evolve); – for a site (robustness ought to be favored);
– in manufacturing (the “cost” factor will probably prevail);
– for a laboratory (your preference will go to a very reliable, strong and proven instrument).
For further information, see Chapter 9.
1.4.2. Analysis of what is already available
The first thing to do will be to see if there is not already in the firm some available material which can meet your needs. This requires:
– good communication between the various parties concerned with the measurements; and
– a good knowledge of the material available.
The latter point is all the more important when there is a risk of technological obsolescence (using a state-of-the-art instrument to its maximum capacity justifies its acquisition and it makes it easier to get new ones), or when the material is very expensive (when you increase the duration of its productive use, you make its amortization easier).
1.4.3. Assessment and acquisition of material
Speaking of compromise about the choice was actually slightly simplistic. Of course, the economic requirements are obviously taken into account; few are the cases when the material is selected without the price being considered (either before or after the purchase!). As for the assessments which are otherwise made, they quite simply depend on the competence and professionalism of the person in charge of the metrological function. He must indeed be on a permanent technological watch. Furthermore, he must make an inventory of what is in store (material and tested material), in order not to have to repeat work endlessly.
The companies which take the trouble to check all the electric and electronic material they buy admit that a far from negligible proportion of the instruments delivered is partly defective or does not comply with tolerances on delivery. A few years ago a survey showed that the percentage of rejected instruments could reach 50%. This is partly explained by the fact that the stated characteristics are obtained by the manufacturers, in a laboratory and in ideal conditions of use; and this situation is very remote from the user’s reality. Tests of assessment preliminary to purchase would be greatly recommended. However, in frequent cases, the instruments that can perform the same function are many in number, the parameters of each of them are numerous and, consequently, the tests are long and expensive.
So, before launching into testing, any person who is interested in purchasing an instrument is entitled to ask the salesman the following questions:
– Have any tests been done? If the answer is yes, when? Where? By whom? In which domain? Is a report of the tests available?
– How long has the instrument been manufactured? How many copies of it have been produced?
– Has stopping its production been considered? – Who has bought it? Is it possible to consult users?
Once you have got this information, and if tests seem necessary, you have to choose between doing them yourself or subcontracting them to a better-equipped organization whose results cannot be questioned. A distinction must be made between learning about a instrument which is presented by a salesman and having its characteristics verified by a specialized laboratory.
Once again, evidence arises of the importance of good relationships (partnership even) with the manufacturers of the instrument and of their obligation to pass on information in a transparent and unrestricted way. However, the role of the buyer is not simple. He must estimate whether the supplier is capable of keeping to the agreed times in general: time of delivery, time of assistance after the sale. Besides, it
seems to be of paramount importance that the team responsible for maintaining the instruments, as well as the users, should be involved in choosing the instruments they need for their activities. In essence there are three reasons for this:
– Because of their experience, the user and the maintenance team know the little details, which make all the difference (and those which mostly “hinder” the smooth progress of their work).
– They get used more easily to equipment they have helped to choose (working and utilizing conditions are improved: that is what is called communicating without demagogy!).
– They are not so easily influenced by attractive advertising, or by purely economic criteria, which makes the overall analysis more objective.
So, economic conditions and assessments generally being what they are, we find ourselves left with technical criteria. The following are those that seem to be the most important.
1.4.4. Technical criteria
1.4.4.1. Basic characteristics
For a measuring instrument (whether used as a standard or not) this most often means that its necessary accuracy is in one certain domain of the studied quantity in ideal conditions, said to be reference conditions: a temperature of 20°C or 23°C, 230V/50Hz power from the mains, no mechanical and electrical perturbations, etc. 1.4.4.2. Comportment towards influence quantities
This concerns the way the basic characteristics change with time according to external constraints: variation of the temperature or the electric power, electromagnetic perturbations, vibrations, etc. The way instruments react over a period of time is often undetermined. As a rule, on-off cycles are more harmful than a long, uninterrupted, working period. Contrary to a widespread opinion, all instruments (even the very accurate ones, the expensive ones, etc.) are liable to drift in time. They have to be recalibrated or reset regularly.
1.4.4.3. Durability of the instruments used
The durability is the interval of time during which the instrument remains capable of meeting your normal need of it. It must not be mistaken for the longevity, which defines the lifespan, generally speaking, of the instrument, even if the instrument no longer meets your need.
A material is durable if it is both reliable (few breakdowns) and maintainable (easy to repair). The information provided by the maintenance teams allows us to have good facts upon which to make a decision.
The most accurate metrological instruments are expensive and, as such, you have to be able to use them for a sufficient length of time. So, you should prefer the makes with good durability; higher investments having sometimes to be considered. You have to estimate how much longer the instrument will be manufactured or maintained. In addition, is this instrument “open” to future evolution? Is there any assurance that it will be compatible with the next generation of equipment?
1.4.4.4. Homogeneity of the supply of instruments
You must avoid having too many different types of equipment and material: if you have equipment of similar types, maintenance will be less costly, you will know your material better, the supplies of spare parts will be cheaper, there will be a possibility of interchangeability in case of a breakdown, periodic calibrations or verifications can be automated, etc.
1.4.4.5. Quality of the supplier’s service
Your relationship with the supplier of instrument must not stop with the purchase. You must analyze the technical assistance the supplier can provide. Have provisions been made for the setting up of the instrument, for clear explanatory documents (utilization, maintenance, intervention, etc.), in the language of the country where it will be used, or at least in English? How is the supplier able to help if problems occur, and how long, on average, will he make you wait? The more sophisticated the instrument is, the more these questions matter.
Placing an order with a instrument dealer may, sometimes, save time, but there is actually nothing that can replace communication with the manufacturer. As a matter of fact, there are few dealers who have a good knowledge of the instrument they sell, or who attend to the training of the users. It is very often difficult to go beyond the stage of purely commercial advertising.
1.4.4.6. Adaptation of the instrument
It is advisable to get instruments which have been conceived with a “metrological” outlook; i.e. instruments adapted in their principle and realization to the needs of metrologists. For example, all metrologists who work in the time-frequencies scope have “major oscillators”, which are excellent sources of 5 or 10MHz. It is therefore eminently desirable that any synthesizer or frequency meter should be able to work either on its internal oscillator or on an external signal. The best-equipped “frequency” laboratories possess a caesium clock, or at least a
rubidium clock, from which a 10MHz signal is drawn and distributed in the firm in order to synchronize frequency meters and synthesizers.
1.4.4.7. Possibility of traceability
When you buy a measuring instrument, you have to raise the question of traceability to national or international standards before you eventually make up your mind to proceed with the purchase. Is it or is not possible to relate your measurements validly to the accepted standards at the national or international level? The question of traceability is developed in 1.5 below.
1.4.4.8. Computerization and the speed of taking measurements
There is a technical parameter which has a direct consequence on the cost of quality to the firm: how fast it will be to take a measurement? The question, and its answer, is as much about how quickly the instrument can provide the necessary information as about the transcription of the measurement in a simple form. A digital display offers ease of reading and can, in the case of the vernier calliper for example, reduce by a factor of five the time it takes to take measurements.
In addition, it may be important to computerize the measurement. Computerization makes it possible:
– to increase the speed at which measurements are obtained by decreasing the input time;
– to increase the Quality Assurance by reducing the risk of making mistakes while, for example, writing the results out by hand;
– to incorporate the measuring instrument into a computerized “Statistic Process Control” (SPC).
Of course, computerization is possible on adaptable instruments, for instance digital display instruments which have an outlet to connect to an RS 232 plug. These remarks refer, in particular, to those instruments which are used on sites or in production, and also in metrology laboratories.
1.4.4.9. Ergonomics
Several types of instruments can be selected for a specific measurement. However, some will turn out to be less “handy” to implement. The ergonomic aspect of the utilization must not be forgotten: ease of handling, utilization by a left-handed person, integration into the work surface, bulk and weight, etc.
1.4.4.10. Capability of measuring instruments
This is a very important parameter that people in charge of metrology and people who use measuring instruments must keep in mind. The “capability” of the
measuring instrument is an indication which is the extent to which the instrument makes it possible to assess whether, and to what extent, the measuring system fits with the tolerance that is being checked. The measuring system includes the measuring instruments (the material), applied measurement processes (the methods) and the personnel who do the measuring, that is to say the users (the person). To put it another way, it is about whether the prescribed interval of tolerance properly fits with the overall uncertainty of measurement.
Choosing too effective a means would result in a superquality which would lead to too high a price. On the other hand, a lack of effectiveness would bring about an unacceptable percentage of defective parts being manufactured. Who amongst us has not had to struggle with too strict intervals of tolerance, which are hard to comply with in manufacture, and also in measurement? What is the good of striving to get a result to the hundredth of a unit (0.01 volt for example) when the dispersion of a series of measurements is already equal to one tenth of this unit? You need to take into consideration the limits (and the cost) of the measuring instruments to be used to check the technical specifications (intervals of tolerance) when you choose the instruments.
Consequently, the choice of the instrument depends on the tolerance to be verified. You have to clearly delimit the uncertainties of measurement that will appear when you use the material. The French standard NF E02204 (which concerns mechanical engineering, but which can serve as a basis for other purposes) provides very useful supplementary information and definitively repeals the widespread “10%” rule.
In production, the capability index (whole or by centering) is given by the following formula:
Cp = [upper tolerance - lower tolerance]/6 s
with s = standard deviation of the series produced
Cpk = MIN [ (upper tolerance - mean)/3 and (lower tolerance - mean)/3 s]
In metrology, the capability index of the means of measurement (Cmm) is often: Cmm = IT/6 Ig with IT = interval of tolerance (from specifications) Ig = overall uncertainty of the measurement
1.4.5. Economic criteria
For reasons that are the very bases of the metrological function, it is necessary to practice metrology with well-known measuring equipment. It is possible to reckon
how much a measurement costs, but this does not mean anything unless all the parameters of the cost are taken into account:
– the purchase price of the material and its resale price after it has been used a certain number of years;
– the costs of operation (expenses for operating the material, usually the lowest cost), electric power, recording paper, accessories;
– the cost of maintenance (including calibration, and preventive and corrective maintenance);
– the cost of lack of availability: will a replacement material be needed while it is being maintained? Will there be any financial consequence?
These different parameters are interdependent; automation increases the purchase price, but it reduces the operating cost. High reliability also increases the purchase price, but it cuts down the cost of maintenance.
1.4.6. Grid of the analysis of the choice
There are two stages when you select a measuring instrument. 1.4.6.1. Stage 1: primary technical requirements (unavoidably necessary)
The point is to determine the quantities, the ranges of measurement and the uncertainties which should be found in the instrument so that you can get the expected quality of instrument. The outcome of this stage will be a list of the instruments available on the market which meeting the technical requirements. 1.4.6.2. Stage 2: secondary technical requirements (desirable)
It will be possible at this stage to make a decision based on the results of outside evaluations, and taking commercial and economic conditions into account.
Here is a tool to help thinking with the decision-making: a good mind of “Management of Quality” will always try to use practical tools. We suggest that you make a list of the criteria to consider when choosing an instrument, then to attribute to each criterion a coefficient depending on how important each criterion appears to be, and then a mark. The items on this grid should come from the analysis of the criteria undertaken by the manager of the metrological activities (the person in charge of the metrological function in the firm), the user, the buyer and the personnel responsible for the maintenance. Each person’s opinion will thus be taken into account. The important thing is to make a careful list of questions and provide an answer to each one. It is true that experience is not easy to weigh, but the object of this method is just to provide a starting point to work out a decision (Table 1.2).
The various people who are concerned with the instrument should meet to determine the values of the weightings. The role of these weightings is to give more weight to one or several items of the grid which, according to the group, have a certain importance.
The final mark for each item is obtained by multiplying the mark of the item by the associated weighting (n*c). The weightings c (Σc) are added, then the products c*n (Σc*n) are added. The evaluation of the measuring instrument is obtained by the division: Σc*n --- Σc ∑c = ________ ∑c*n = ________ Identification = Type = Manufacturer = Coef. c Note n c*n Technical needs
– homogeneity of the supply of instruments – risk of rapid obsolescence
– documents from the supplier – technical assistance
– adaptation of the instrument to technological requirements
– etc. Outside
evaluations
– evaluation from a centre accredited by the COFRAC or the DKD
– evaluation by users (EXERA, etc.) – experience gained on similar material of the same make
– press-cuttings from the specialized press – etc.
Economic and commercial conditions
– cost/price of the competitor’s range – possibilities of purchase or loan – required time for delivery – time allowed for repair – etc.
1.4.7. Technical assistance for users of measuring instruments
In some countries organizations have established themselves to provide users of measuring instruments with technical assistance. Two examples of such organizations are given below.
1.4.7.1. The EXERA (France)
This chapter, which deals with the analysis of metrological requirements, would be left unfinished if no mention was made of the EXERA, one of the few associations which work to support industrial metrology.
The EXERA is a non-profit-making association, an amalgamation of companies and organizations that are major users of instruments and systems of measurement, regulation and automation. Since its foundation, in 1970, its purpose has been to produce and circulate original information and to provide its members with assistance when they need to express their requirements, to choose, to install and to operate materials and systems.
The EXERA is first and foremost a club; it is a privileged meeting place for users, where specialists (over 500) can freely exchange what information about what experience has taught them, as well as information about instruments and systems.
This club acts, in essence, through its members by organizing the technical evaluation of materials. It also initiates the writing of guides about the choice of material in the different technical areas and, at the same time, does its best to develop a constructive dialogue with manufacturers.
In a spirit of partnership, groups of users are constituted so that they can take responsibility for their needs and they can better express and defend them in front of manufacturers. This enables the users and the manufacturers to obtain more elements of explanation on investments and technological trends. There are technical commissions about automation, instruments, analyzers, measurements and systems for the tests, etc.
In 1982, the EXERA signed an agreement of international cooperation with two other organizations of users:
– the SIREP (Britain); and – the WIB (the Netherlands).
These two other associations have members in other industrialized countries, for example, the USA, Canada, Japan, Finland, Sweden, Belgium, Switzerland, etc.
The main features of the agreement, which ultimately concerns more than 100 large companies, are:
– the full-scale and well-balanced exchange of assessments of instruments and surveys, which are all written in English;
– the acceptance of common principles regulating the procedures of evaluation and the presentation of the documents;
– the harmonization of the work programs;
– the gradual adjustment of the formalities regulating the testing of materials. Altogether, there are about 90 members in the three associations; 40 of the members belong to the EXERA, among them are: CEA, CGE, EDF, GIAT, IFP, L'OREAL, PECHINEY, RENAULT, TOTAL, etc. At present, approximately 80 reports are distributed annually by the three associations. In December 1991, the SIREP, the WIB and the EXERA were officially recognized by the European Organisation for Conformity Assessment (EOTC) as “Agreement group”. For more information, see www.eotc.bc or www.exera.com.
1.4.7.2. VDI/VDE-GMA (Germany)
In Germany an organization similar to EXERA is the Society for Measurement and Automatic Control GMA (Gesellschaft Mess- und Automatisierungstechnik). This organization is a joint organization of the Association of German Engineers VDI (Verein Deutscher Ingenieure) and the Association for Electrical, Electronic and Information Technologies VDE (Verband der Elektrotechnik, Elektronik und
Informationstechnik).
The GMA is a network of technical competence in metrology and other fields of activity. It combines expertise of institutions such as the German National Metrology Institute (PTB), the German Calibration Service (DKD), the German Institute for Standardisation (DIN), the International Organization for Standardization (ISO) and several industry associations and societies.
GMA activities include:
– the promotion of the exchange of information between industry, public authorities and scientific institutions;
– the organization of congresses, conferences, symposiums, etc. to promote the flow of information concerning new processes and developments;
– the preparation of publications, recommendations, guidelines, etc. to improve understanding;
– the national and international representation in the field of measurement and automation controls;
– the publication and promotion of technical and scientific literature; – the support of education and post-graduate training.
In common technical committees, honorary experts of industry, research and science cooperate in different fields of metrology. Each committee is focused on specific branches of metrology. These committees produce newly-developed or up-dated technical documents. These documents are first presented as drafts. Views and comments of potential users are evaluated and the documents are modified before they are definitively published.
The guidelines published by VDI/VDE-GMA describe standards, e.g. in metrology. These metrology documents define procedures for users of measurement instruments (see the following table).
Metrological level Guidelines, documents and standards National Metrology Institute
(PTB) DKD accredited calibration laboratory
National DIN standards or DKD guidelines International EN or ISO standards
EA documents Optional:
In-house calibration laboratory Measurement and testing
equipment Product
VDI/VDE guidelines DKD guidelines
EA documents Table 1.3. Metrology literature used in Germany
In the VDI/VDE guidelines there are three series dealing with the treatment of measuring equipment:
The series VDI/VDE/DGQ 2618, “Inspection of measuring and test equipment – instructions to inspect measuring and test equipment for geometrical quantities”, contains general considerations and determinations, as well as information on the expression of uncertainty in measurement. In separate documents there are procedures for calibration and surveillance of specific-measurement instruments. An example of such papers is the paper about the procedures for “Callipers for external, internal and depth dimensions”.
Another series, the VDI/VDE/DGQ/DKD 2622 guidelines, deals with “Calibration of measuring equipment for electrical quantities”. Again, along with a general introduction including information on the measurement uncertainty, there are separate documents defining the calibration of specific, electrical measurement instruments. An example is the calibration procedure for electrical oscilloscopes.
The third series, VDI/VDE 2617, is entitled “Accuracy of coordinate measuring machines – parameters and their reverification”. The calibration, the acceptance and the surveillance of coordinate measuring equipment is defined in separate documents. This series is used as a base for the development of a new ISO standard on coordinate measuring machines.
More detailed information is available at the GMA secretariat in Düsseldorf, Germany (e-mail: [email protected]).
1.5. The traceability of the measurements
It has to be said repeatedly: the calibration requirements and the traceability define the quality of the measurements. The metrological function is responsible for the management of the quality of the measurements. This has to be taken into consideration from the beginning of the process that leads to the selection of the method of measurement, and then the means of measurement.
1.5.1. The necessity of traceability of the measurements
Traceability is the very basis of metrology. What good is it to take measurements if the measurements do not mean the same thing to everybody? For example, let us look at the measurement of the value of the “foot” in the past. Until about the 18th century (and even later), the “foot” was used as a unit to measure distances. Everyone used the same word. A worthy sample of this quantity was available to avoid arguments, such as “is it a child’s foot, or a woman’s, or a man’s?” The problem was that when the value was translated into the metric system, it gave the following results:
– foot of the King of France 32.48 cm – Roman foot 29.63 cm – foot from Bordeaux (South of France) 35.70 cm – foot from Lorraine (East of France) 28.60 cm – foot from Vienna (Austria) 31.50 cm
These discrepancies resulted from the lack of a national reference (let us not even talk of a European one), and of local comparisons to each reference. Just imagine the Airbus today manufactured from all parts of the world.
It is therefore indispensable to have metrological references in one’s firm and to have them compared to national reference quantities by calibration. Comparisons between accredited laboratories are made by national accreditation bodies (the COFRAC in France, the Deutscher Kalibrierdienst (DKD) in Germany) and there are programs of comparison that make it possible to ensure that the standards of different countries are related.
It is to be regretted that all the industrialized countries are not at the same level of progress in metrology. However, such European countries as Britain, France, Germany, Italy, Spain, etc. are the leaders.
It has been said above that it is important to have reference standards in one’s firm and to have them calibrated in accredited calibration centers or laboratories. However, a choice must be made between having the metrology integrated in the firm and having it subcontracted. As some providers of calibration services propose to calibrate the measuring instruments with standards of their own, you need to be careful.
You must absolutely make sure that:
– their standards are periodically calibrated in a competent laboratory (whose organization complies with the ISO 17025) accredited by a national organization (COFRAC, DKD, UKAS, etc.);
– the provider of the service can guarantee the quality of the measurements provided. An audit of the provider’s system of management of the quality will probably be necessary.
You have to be able to demonstrate full traceability of the measurement that has been made, the relationship between the measurement and the instrument used, and also the traceability of the firm’s instrument, in order to show that the chain of calibration has not been broken. In addition, do not forget to verify that at every stage the uncertainties of measurement are not too large.
Bringing in a provider of services who has their own accredited laboratory is not a must. However, if the provider has one, it is further evidence of his seriousness and commitment to his job. There is every reason to think, that a provider with an accredited laboratory knows better what the word “metrology” means than a competitor who does not have any accredited laboratory. The provider with the
accredited laboratory can grasp the primary technical needs of the client: quantities, scope of measurement and uncertainties.
Furthermore, the investment required for launching an accredited laboratory excludes the “transitory” type of company that starts up in “commercial niches” and then vanishes as quickly as it has appeared.
Stability is a key word in metrology. It is important not to change your provider
too regularly when you decide to subcontract the calibrations; for example, do not consider only the price and have a yearly competition.
Nevertheless, let us point out that what has been said so far applies to movable measurement, control, test or analysis instruments. In the case of equipment such as heavy machinery (traction, compression, hardness, etc.), scales, air conditioning chambers, etc., the verification can only be done on-site. It is not necessary for the provider to have their own laboratory since the whole intervention is carried out on-site. However, the provider must use working standards which are related to the calibration chains.
1.5.2. Calibration requirements
Several problems come to mind when thinking of calibration. First of all, how can a particular measuring instrument be calibrated? If it is a calliper, you will think about using gauge blocks. Has anyone even considered measuring rods for a micrometer? What is to be done with dynamometric spanners, balances, etc.? If you go into physical chemistry, etc. it gets even more complex! Some methods of measurement demand equivalent methods of calibration. Fortunately, some manufacturers of materials provide tips.
When you look deeper into the matter, you realize that quite often you talk about calibration, but what you actually need is a verification, perhaps even a metrological confirmation (see ISO 10012 standard). Therefore, it might be necessary to proceed to an internal checking between two interventions, which is just a simplified examination of good working order.
Calibration must be done intelligently, which means doing just what is necessary; it is not only a means to avoid auditor’s critical views. How many firms, which work in mechanical engineering and have their sets of gauge blocks calibrated in an accredited calibration laboratory simply open their calibration certificate?
Never must it be forgotten that the major purpose of calibration is to verify the measuring instrument and calculate the uncertainties that go with the results of the measurements taken with that instrument.
The question of the interval of the calibrations inevitably arises quickly. The answer, which should make everybody happy, is that it depends.
Some methods of measurement meet a few demands, particularly in the field of physical chemistry. In any event, measuring instruments should be calibrated reasonably frequently, so as to detect and prevent any possible drift, but not too often because of the overall cost involved.
On the question of follow-up interval, the reader’s attention is drawn to Chapter 6, as well as to the handbook of documentation published by AFNOR on the subject of the surveillance intervals.
The reader should wary of any person who claims that they can tell which intervals are the right ones. As a matter of fact, you always start quite randomly and then, with experience, you define the necessary intervals more accurately.
There is the question of subcontracting the calibration; it is not cheap regardeless of whether you do it yourself or subcontract it.
It is our opinion that a compromise can be considered. In fact, even though the metrology is not the firm’s chief activity, it is a part of the “Management of Quality”. If you retain part of it in the firm, it makes it possible to maintain the user’s awareness of the importance of the measuring instruments, of the notion of connected uncertainty, etc. However, a firm cannot excel in everything and it must avoid spreading its resources too thinly. It is always possible to ascertain whether there are any local providers of services in metrology and, if so, their charges.
1.5.3. The selection of standards
The content of this technical paragraph does not concern all firms; the small- or medium-sized firms that do not use many standards (merely a set of gauges or masses for example) need not worry. What is presented here is a practically complete line of thought which can reveal useful for the firms with a metrology service. However, let us first recall the definition of the word “standard” in the “International Vocabulary of basic and general terms in Metrology” (ISO document, 1993):
