0 Votes +0 Votes -

6 vues54 pagesarticulo del proceso del calculo de incertidumbre de las mediciones

Jun 12, 2018

© © All Rights Reserved

PDF, TXT ou lisez en ligne sur Scribd

articulo del proceso del calculo de incertidumbre de las mediciones

© All Rights Reserved

6 vues

articulo del proceso del calculo de incertidumbre de las mediciones

© All Rights Reserved

- SQL All-in-One For Dummies
- Networking All-in-One For Dummies
- Structural Elements Design Manual
- Project Management, Planning and Control: Managing Engineering, Construction and Manufacturing Projects to PMI, APM and BSI Standards
- Total Quality Management: Key Concepts and Case Studies
- Chemical Engineering Design: Principles, Practice and Economics of Plant and Process Design
- ISO 9001:2015: A Pocket Guide
- 5G NR: The Next Generation Wireless Access Technology
- Bridge Engineering: Classifications, Design Loading, and Analysis Methods
- Learn C++
- Gerenciamento de Projetos – 9ª Edição
- Construction Process Planning and Management: An Owner's Guide to Successful Projects
- Mechanical Design
- Learn Jazz Piano Book 3
- ISO 9001:2015 Audit Guide and Checklist
- Piping and Pipeline Calculations Manual: Construction, Design Fabrication and Examination
- Process Plant Layout
- Basic Wifi Hacking
- Manual of Engineering Drawing: Technical Product Specification and Documentation to British and International Standards
- HTML, CSS, and JavaScript Mobile Development For Dummies

Vous êtes sur la page 1sur 54

Compatibility—Computations and

Treatment of Measurement

Uncertainty

C63®

Sponsored by the

Accredited Standards Committee C63® — Electromagnetic Compatibility

Sponsored by the

Accredited by the American National Standards Institute

IEEE

3 Park Avenue ANSI C63.23-2012

New York, NY 10016-5997

USA

13 March 2013

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

American National Standard

Guide for Electromagnetic

Compatibility—Computations and

Treatment of Measurement

Uncertainty

accredited by the

American National Standards Institute

Secretariat

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Abstract: Methods for estimating measurement uncertainty of emissions measurement results

are provided, for use in conjunction with the basic methods of ANSI C63.4. Included in this

document are both Type A and Type B uncertainty evaluation methods.

Keywords: ANSI C63.23, electromagnetic compatibility (EMC), emissions, immunity,

measurement uncertainty, metrology, Type A evaluation, Type B evaluation

3 Park Avenue, New York, NY 10016-5997, USA

All rights reserved. Published 13 March 2013. Printed in the United States of America.

IEEE is a registered trademark in the U.S. Patent & Trademark Office, owned by The Institute of Electrical and Electronics

Engineers, Incorporated.

Print: ISBN 978-0-7381-8251-3 STDPD98135

IEEE prohibits discrimination, harassment and bullying. For more information, visit http://www.ieee.org/web/aboutus/whatis/policies/p9-26.html.

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

of the publisher.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

American National Standard

An American National Standard implies a consensus of those substantially concerned with its scope and

provisions. An American National Standard is intended as a guide to aid the manufacturer, the consumer,

and the general public. The existence of an American National Standard does not in any respect preclude

anyone, whether he has approved the standard or not, from manufacturing, marketing, purchasing, or using

products, processes, or procedures not conforming to the standard. American National Standards are

subject to periodic review and users are cautioned to obtain the latest editions.

CAUTION NOTICE: This American National Standard may be revised or withdrawn at any time. The

procedures of the American National Standards Institute require that action be taken to reaffirm, revise, or

withdraw this standard no later than five years from the date of publication. Purchasers of American

National Standards may receive current information on all standards by calling or writing the American

National Standards Institute.

Authorization to photocopy portions of any individual standard for internal or personal use is granted by

the Institute of Electrical and Electronics Engineers, Inc., provided that the appropriate fee is paid to

Copyright Clearance Center. To arrange for payment of licensing fee, please contact Copyright Clearance

Center, Customer Service, 222 Rosewood Drive, Danvers, MA 01923 USA; (978) 750-8400. Permission to

photocopy portions of any individual standard for educational classroom use can also be obtained through

the Copyright Clearance Center.

Errata

Users are encouraged to check the IEEE Errata URL (http://standards.ieee.org/findstds/errata/index.html),

and the one for ASC C63® at http://www.c63.org/explanations_interpretations_request.htm, for errata

periodically.

Interpretations

Current interpretations can be accessed at the following URLs: http://www.c63.org/explanations_

interpretations_request.htm.

For more information about the committee that produced and maintains this standard, visit the ANSI

Accredited Standards Committee C63® web site at http://www.c63.org.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Notice to users

Users of these documents should consult all applicable laws and regulations. Compliance with the

provisions of this guide does not imply compliance to any applicable regulatory requirements.

Implementers of the guide are responsible for observing or referring to the applicable regulatory

requirements. IEEE does not, by the publication of its standards, intend to urge action that is not in

compliance with applicable laws, and these documents may not be construed as doing so.

Copyrights

This document is copyrighted by the IEEE. It is made available for a wide variety of both public and

private uses. These include both use, by reference, in laws and regulations, and use in private self-

regulation, standardization, and the promotion of engineering practices and methods. By making this

document available for use and adoption by public authorities and private users, the IEEE does not waive

any rights in copyright to this document.

Users of IEEE Standards documents should be aware that these documents may be superseded at any time

by the issuance of new editions or may be amended from time to time through the issuance of amendments,

corrigenda, or errata. An official IEEE document at any point in time consists of the current edition of the

document together with any amendments, corrigenda, or errata then in effect. In order to determine whether

a given document is the current edition and whether it has been amended through the issuance of

amendments, corrigenda, or errata, visit the IEEE-SA Website at http://standards.ieee.org/index.html or

contact the IEEE at the address listed previously. For more information about the IEEE Standards

Association or the IEEE standards development process, visit the IEEE-SA Website at

http://standards.ieee.org/index.html.

Errata

Errata, if any, for this and all other standards can be accessed at the following URL:

http://standards.ieee.org/findstds/errata/index.html. Users are encouraged to check this URL for errata

periodically.

iv

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Patents

Attention is called to the possibility that implementation of this guide may require use of subject matter

covered by patent rights. By publication of this guide, no position is taken by the IEEE with respect to the

existence or validity of any patent rights in connection therewith. If a patent holder or patent applicant has

filed a statement of assurance via an Accepted Letter of Assurance, then the statement is listed on the

IEEE-SA Website http://standards.ieee.org/about/sasb/patcom/patents.html. Letters of Assurance may

indicate whether the Submitter is willing or unwilling to grant licenses under patent rights without

compensation or under reasonable rates, with reasonable terms and conditions that are demonstrably free of

any unfair discrimination to applicants desiring to obtain such licenses.

Essential Patent Claims may exist for which a Letter of Assurance has not been received. The IEEE is not

responsible for identifying Essential Patent Claims for which a license may be required, for conducting

inquiries into the legal validity or scope of Patents Claims, or determining whether any licensing terms or

conditions provided in connection with submission of a Letter of Assurance, if any, or in any licensing

agreements are reasonable or non-discriminatory. Users of this guide are expressly advised that

determination of the validity of any patent rights, and the risk of infringement of such rights, is entirely

their own responsibility. Further information may be obtained from the IEEE Standards Association.

v

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Participants

At the time this guide was published, Accredited Standards Committee C63®—Electromagnetic

Compatibility had the following membership:

Jerry Ramie, Secretary

Organization Represented Name of Representative

Alcatel–Lucent Technologies ............................................................................................................... Dheena Moongilan

Alliance for Telecommunications Industry Solutions (ATIS).........................................................................Mel Frerking

.............................................................................................................................................................. James Turner (Alt.)

American Council of Independent Laboratories (ACIL) ................................................................................ Harry Hodes

............................................................................................................................................................... John Repella (Alt.)

American Radio Relay League (ARRL) .................................................................................................... Edward F. Hare

........................................................................................................................................................... Kermit Carlson (Alt.)

Apple, Inc. ..................................................................................................................................................... Fraidun Akhi

.............................................................................................................................................................. Indrandil Sen (Alt.)

AT&T ........................................................................................................................................................ David Chapman

Bureau Veritas ......................................................................................................................................... Jonathan Stewart

........................................................................................................................................................... Mairal Hussain (Alt.)

Cisco Systems ................................................................................................................................................. Andy Griffin

................................................................................................................................................................. David Case (Alt.)

Dell Inc. ..................................................................................................................................................... Richard Worley

Electric Power Research Institute ....................................................................................................................Phil Keebler

ETS-Lindgren .................................................................................................................................................. Zhong Chen

........................................................................................................................................................ Michael Foegelle (Alt.)

Federal Communications Commission (FCC) .............................................................................................. William Hurst

Food and Drug Administration (FDA) ................................................................................................ Jeffrey L. Silberberg

................................................................................................................................................................ Don Witters (Alt.)

Information Technology Industry Council (ITIC) .......................................................................................... John Hirvela

....................................................................................................................................................... Joshua Rosenberg (Alt.)

Institute of Electrical and Electronics Engineers, Inc. (IEEE) ............................................................. Donald N. Heirman

IEEE-Electromagnetic Compatibility Society (EMCS) ......................................................................... H. Stephen Berger

........................................................................................................................................................ Donald Sweeney (Alt.)

Motorola Mobility.......................................................................................................................................... Tom Knipple

Motorola Solutions ...................................................................................................................................... William Elliott

........................................................................................................................................................ Deanna Zakharia (Alt.)

National Institute of Standards and Technology (NIST) .............................................................................. Dennis Camell

Polycom .......................................................................................................................................................... Jeff Rodman

............................................................................................................................................................ Tony Griffiths (Alt.)

Research in Motion (RIM) ............................................................................................................................. Masud Attayi

.............................................................................................................................................................. Daoud Attayi (Alt.)

Samsung Telecommunications .......................................................................................................................Tony Riveria

............................................................................................................................................................. Kendra Green (Alt.)

Society of Automotive Engineers (SAE) ..................................................................................................... Poul Andersen

............................................................................................................................................................... Gary Fenical (Alt.)

Sony Ericsson Mobile Communications ....................................................................................................... Mats Hansson

.................................................................................................................................................................. Jon Kenny (Alt.)

Telecommunication Certification Body (TCB) Council ........................................................................................Art Wall

.......................................................................................................................................................... William Stumpf (Alt.)

Telecommunications Industry Association (TIA) ................................................................................... Stephen Whitesell

TUV-America, Inc. .......................................................................................................................................... Chip Fleury

........................................................................................................................................................... David Schaefer (Alt.)

UL LLC ......................................................................................................................................................... Robert Delisi

............................................................................................................................................................. Jeffrey Moser (Alt.)

U.S. Department of Defense—Joint Spectrum Center ............................................................................. Marcus Shellman

....................................................................................................................................................Michael Duncanson (Alt.)

U.S. Department of the Navy—SPAWAR ....................................................................................................... Chris Dilay

...............................................................................................................................................................David Hilton (Alt.)

vi

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Individual Members .................................................................................................................................. Daniel Hoolihan

.........................................................................................................................................................................John Lichtig

.................................................................................................................................................................. Werner Schaefer

............................................................................................................................................................... Ralph M. Showers

.............................................................................................................................................................. David Zimmerman

Members Emeritus ................................................................................................................................ Warren Kesselman

..................................................................................................................................................................... Herbert Mertel

.......................................................................................................................................................... H. R. (Bob) Hofmann

At the time this guide was completed, ASC C63® Subcommittee 1 had the following membership:

Harry Hodes, Vice Chair

Janet O’Neil, Secretary

Masud Attayi Daniel Hoolihan Werner Schaefer

Zhong Chen Greg Kiemel Ralph M. Showers

Bob DeLisi Victor Kuczynski Dan Sigoiun

Dean Ghizzone Rick Lombardi Jeffrey L. Silberberg

Andy Griffin Randy Long William Stumpf

Tim Harrington Doug Parker Art Wall

Donald Heirman Ghery Pettit Stephen Whitesell

At the time this guide was completed, the ASC C63® Subcommittee 1 Measurement Uncertainty Working

Group had the following membership:

Dennis Camell, Vice Chair

Donald Heirman Doug Parker Werner Schaefer

Daniel Hoolihan Michael J. Windler

vii

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Introduction

This introduction is not part of ANSI C63.23-2012, American National Standard Guide for Electromagnetic

Compatibility—Computations and Treatment of Measurement Uncertainty.

ANSI C63.23 is intended to provide measurement laboratories with guidelines and generally accepted

laboratory practices in the determination of EMI measurement uncertainties. The primary application of

this edition of ANSI C63.23 is for use with ANSI C63.4.

viii

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Contents

1. Overview .................................................................................................................................................... 1

1.1 Scope ................................................................................................................................................... 1

1.2 Purpose ................................................................................................................................................ 2

2. Normative references.................................................................................................................................. 2

3. Definitions .................................................................................................................................................. 3

4.1 Introduction to measurement uncertainty ............................................................................................ 5

4.2 Concepts of uncertainty ....................................................................................................................... 5

4.3 Application of uncertainty values in conformity assessment ............................................................... 7

5.1 General ................................................................................................................................................ 8

5.2 Uncertainty contributors common to all conducted and radiated emission EMC measurements—

Type B evaluation ................................................................................................................................ 9

5.3 Uncertainty contributors for conducted emission measurements at a mains port using an LISN—

Type B evaluation .............................................................................................................................. 12

5.4 Uncertainty contributors for radiated emission measurements in the frequency range 30 MHz to

1000 MHz on an open-area test site (OATS) or in a semianechoic chamber (SAC)—Type B

evaluation ........................................................................................................................................... 13

5.5 Uncertainty contributors for radiated emission measurements in the frequency range above 1 GHz—

Type B evaluation .............................................................................................................................. 17

5.6 Type A evaluations of contributors ................................................................................................... 19

5.7 Example spreadsheets ........................................................................................................................ 26

ix

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

American National Standard

Guide for Electromagnetic

Compatibility—Computations and

Treatment of Measurement

Uncertainty

IMPORTANT NOTICE: IEEE Standards documents are not intended to ensure safety, health, or

environmental protection, or ensure against interference with or from other devices or networks.

Implementers of IEEE Standards documents are responsible for determining and complying with all

appropriate safety, security, environmental, health, and interference protection practices and all

applicable laws and regulations.

This IEEE document is made available for use subject to important notices and legal disclaimers.

These notices and disclaimers appear in all publications containing this document and may

be found under the heading “Important Notice” or “Important Notices and Disclaimers

Concerning IEEE Documents.” They can also be obtained on request from IEEE or viewed at

http://standards.ieee.org/IPR/disclaimers.html.

1. Overview

1.1 Scope

In standardized emissions compliance measurements, the emission level of an equipment under test (EUT)

is measured, after which compliance with the associated limit is determined. The measured emission level

is an approximation of the true level due to uncertainties induced by various contributors. In emissions

compliance testing, major relevant contributors related to the EUT or to the measurement procedure happen

to be unspecified and no quantitative information is available about their values, for example, operator

skills, quality system of the laboratory performing the test, environmental conditions, production tolerance

of the EUT model, operation of the EUT, and so on. This document concentrates on the measurement

instrumentation uncertainty, which is a subpart of the total uncertainty of the measurement, and it includes

only the effects of those contributors that are related to the measurement instrumentation.

This application guide provides methods for determining the uncertainty of measurement for

electromagnetic interference (EMI) measurement results. This guide provides information on the

application of Type A statistical evaluations. For Type B applications, this guide also provides information

on where to obtain specified published information that can lead to an evaluation of uncertainty. The

1

Copyright © 2013 IEEE. All rights reserved

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

current document provides information on the range 150 kHz to 30 MHz for conducted emissions on main

lines, and 30 MHz to 18 000 MHz for radiated emissions measurements.

1.2 Purpose

This document provides recommended methods for estimating the uncertainty of EMI measurements as

performed in accordance with C63® standards (e.g., ANSI C63.41), for the purposes of this edition, or

similar standards that do not already contain recommended methods for estimating measurement

uncertainty.

2. Normative references

The following referenced documents are indispensable for the application of this document (i.e., they must

be understood and used, so each referenced document is cited in text and its relationship to this document is

explained). For dated references, only the edition cited applies. For undated references, the latest edition of

the referenced document (including any amendments or corrigenda) applies.

ANSI C63.2, American National Standard for Electromagnetic Noise and Field Strength Instrumentation,

10 Hz to 40 GHz Specifications.2

ANSI C63.4, American National Standard for Methods of Measurement of Radio-Noise Emissions from

Low-Voltage Electrical and Electronic Equipment in the Range of 9 kHz to 40 GHz.

ANSI C63.14, American National Standard Dictionary of Electromagnetic Compatibility (EMC) including

Electromagnetic Environmental Effects (E3).

CISPR 16-1-1:2007, Specification for Radio Disturbance and Immunity Measuring Apparatus and

Methods—Part 1-1: Radio Disturbance and Immunity Measuring Apparatus—Measuring Apparatus.3

CISPR 16-1-2:2006, Specification for Radio Disturbance and Immunity Measuring Apparatus and

Methods—Part 1-2: Radio Disturbance and Immunity Measuring Apparatus—Ancillary Equipment—

Conducted Disturbances.

CISPR 16-1-4:2010, Specification for Radio Disturbance and Immunity Measuring Apparatus and

Methods—Part 1-4: Radio Disturbance and Immunity Measuring Apparatus—Antennas and Test Sites for

Radiated Disturbance Measurements—Edition 3.0.

CISPR 16-4-2, Specification for Radio Disturbance and Immunity Measuring Apparatus and Methods—

Part 4-2: Uncertainties, Statistics and Limit Modeling—Uncertainty in EMC Measurements.

ISO 21748:2004, Guidance for the Use of Repeatability, Reproducibility and Trueness Estimates in

Measurement Uncertainty Estimation.4

and Nested Experiments.

1

Information on references can be found in Clause 2.

2

ANSI publications are available from the American National Standards Institute (http://www.ansi.org/).

3

CISPR documents are available from the International Electrotechnical Commission (http://www.iec.ch/). They are also available in

the United States from the American National Standards Institute (http://www.ansi.org/).

4

ISO publications are available from the ISO Central Secretariat (http://www.iso.org/). ISO publications are also available in the

United States from the American National Standards Institute (http://www.ansi.org/).

2

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

3. Definitions

For the purposes of this document, the following terms and definitions apply. ANSI C63.14 and the IEEE

Standards Dictionary Online should be consulted for terms not defined in this clause.5 General terms and

definitions used in the expression of uncertainty are provided in ISO/IEC Guide 98-3:2008 [B16]6 (GUM),

and important general terms and definitions related to measurement are contained in ISO/IEC Guide

99:2007 [B17] (VIM).

analysis of variance (ANOVA): A collection of statistical models and their associated procedures in

which the observed variance is partitioned into components due to different explanatory variables. ANOVA

is a general technique that can be used to test the hypothesis that the means among two or more groups are

equal, under the assumption that the sampled populations are normally distributed.

a priori distribution: A distribution in which the probability curve is an isosceles triangle symmetrical

about the center of the base [e.g., the result of casting (rolling) a pair of fair dice]. Syn.: triangular

distribution.

combined standard uncertainty, uc(y): Standard uncertainty of the result of a measurement when that

result is obtained from the values of a number of other quantities, equal to the positive square root of a sum

of terms, the terms being the variance or covariance of these other quantities weighted according to how the

measurement result varies with changes in these quantities.

conformity assessment: Demonstration that specified requirements relating to a product, process, system,

person, or body are fulfilled. [ISO/IEC 17000:2004 [B15], 2.1 and 0.5, MOD]

NOTE 1—The subject field of conformity assessment includes activities such as testing, inspection, and certification,

as well as the accreditation of conformity assessment bodies.7

NOTE 2—The expression “object of conformity assessment” or “object” is used in ISO/IEC 17000:2004 [B15] to

encompass any particular material, product, installation, process, system, person, or body to which conformity

assessment is applied.

NOTE 3—ISO/IEC 17000:2004 [B15] does not include a definition of “conformity” on the grounds that it is not

necessary to do so. “Conformity” does not feature in the definition of “conformity assessment.” The concept of

“conformity assessment” is concerned with “fulfillment of specified requirements,” not with the wider concept of

“conformity.” A definition of the term “specified requirement” is included in ISO/IEC 17000-2004 [B15]. In English,

the term “compliance” is used to distinguish the action of doing what is required (e.g., an organization “complies” by

making something conform or by fulfilling a regulatory requirement).

correction factor: Numerical factor by which the uncorrected result of a measurement is mathematically

adjusted to compensate for an assumed systematic error (e.g., in EMC measurements), the loss in a signal

cable that is usually measured and stated in decibels and added to an uncorrected result that is also

measured and stated in decibels.

NOTE—Because the systematic error cannot be known perfectly, the compensation cannot be complete.

coverage factor, k: Numerical factor used as a multiplier of the combined standard uncertainty in order to

obtain an expanded uncertainty. When k = 2, the confidence level approximates 95% (in case of normal,

Gaussian, probability distribution).

5

The IEEE Standards Dictionary Online subscription is available at http://www.ieee.org/portal/innovate/products/standard/

standards_dictionary.html.

6

The numbers in brackets correspond to those of the bibliography in Annex B.

7

Notes in text, tables, and figures of a standard are given for information only and do not contain requirements needed to implement

this standard.

3

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

NOTE—The coverage factor k is typically in the range of 2 to 3, but different values may be used depending on the

desired confidence level and on the type of probability distribution of the combined standard uncertainty.

degrees of freedom—νi,νeff: The degrees of freedom νi of the standard uncertainties based on Type B

evaluation are, by definition, infinite. If the uncertainty is obtained from a limited number of data points or

observations, as in the case of a Type A evaluation, then νi = n – 1 (see ANSI/NCSL Z540-1-1994 [B3],

CISPR/TR 16-4-1:2009 [B7], UKAS LAB 34:2002 [B21], UKAS M 3003:2007 [B22], and ISO/IEC Guide

98-3:2008 [B16]). The effective degrees of freedom νeff are the “combined” degrees of freedom of the

combined standard uncertainty when it is made up of a combination of Type B and Type A evaluations.

drift: The unwanted change of the value of a measurable characteristic of a device over a specified period

of time when the values of all other characteristics of the device are kept constant. For example, drift may

be the change in the device response with use or age, or since the last calibration or measurement. This may

take the form of a change in readout, usually gradual, without associated change in any influence quantity.

error (of measurement): Result of a measurement minus the true value of the measurand.

NOTE 1—The quantity is sometimes called absolute error of measurement when it is necessary to distinguish it from

relative error.

NOTE 2—Because a true value cannot be determined, in practice a conventional true value is used, if available.

expanded uncertainty, U: Quantity defining the interval about the result of a measurement within which

the values that could reasonably be attributed to the measurand may be expected to lie with a high-specified

level of confidence. Expanded uncertainty is obtained by multiplying the combined standard uncertainty by

a coverage factor k.

influence quantity: Quantity that is not necessarily included in the specification of the measurand but that

nonetheless affects the result of the measurement (e.g., frequency in the measurement of an alternating

electric potential difference) or temperature of a micrometer used to measure length.

measurement uncertainty: Non-negative parameter characterizing the dispersion of the quantity values

being attributed to a measurand based on the information used. Syn.: uncertainty of measurement;

uncertainty.

NOTE 1—Measurement uncertainty includes components arising from systematic effects, such as components

associated with corrections and the assigned quantity values of measurement standards, as well as the definitional

uncertainty. Sometimes estimated systematic effects are not corrected for, but instead, associated measurement

uncertainty components are incorporated.

NOTE 2—The parameter may be, for example, a standard deviation called standard measurement uncertainty (or a

specified multiple of it), or the half-width of an interval, having a stated coverage probability.

NOTE 3—Measurement uncertainty comprises, in general, many components. Some of these may be evaluated by

Type A evaluation of measurement uncertainty from the statistical distribution of the quantity values from series of

measurements and can be characterized by standard deviations. The other components, which may be evaluated by

Type B evaluation of measurement uncertainty, can also be characterized by standard deviations, evaluated from

probability density functions based on experience or other information.

NOTE 4—In general, for a given set of information, it is understood that the measurement uncertainty is associated

with a stated quantity value attributed to the measurand. A modification of this value results in a modification of the

associated uncertainty (for example, different measurement uncertainty values for a radiated emissions measurement

result may apply in different frequency ranges).

4

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

NOTE 5—This definition of measurement uncertainty (i.e., as repeated from 2.26 of ISO/IEC Guide 99:2007 [B17]),

and for situations where uncertainty components arising from sampling are not considered, corresponds to the term and

definition for standards compliance uncertainty given in 3.1.16 of CISPR 16-4-1:2009 [B7].

4. Basic concepts

Measurement uncertainty is the best estimated quantity by which a measured value differs from the true

value of a parameter under evaluation. Correction factors (or biases; see the definition in A.1.4) are

typically used to improve correlation between measurement devices and systems and the reference quantity

to which they have been calibrated. Additionally, a correction factor is always accompanied by a

measurement uncertainty. The need for such factors is an indication that the true value cannot be directly

and completely obtained from the instrumentation and with the method of measurement used. Correction

factors determined during equipment calibration processes determine the bias to be applied to the

measurement result, and they are subject to measurement uncertainty.

Determining the operational characteristics of the EUT with respect to electromagnetic compatibility

compliance requires both emission and immunity measurements. This guide, however, only focuses on

emission measurement, which is covered in, for example, ANSI C63.4. The measurements or evaluations

involve the use of various instrumentation and techniques, requiring operator interaction and decisions. The

measurement (or evaluation) process produces both random and systematic effects that influence the

ultimate outcome of the measurement. Examples of systematic effects are bias of the measurement result

by cable loss, parallax when reading a D’Arsonval meter, or habits and preferences of the operator. An

example of a random effect is the influence of noise contribution to the measured amplitude if the

amplitude of the measurand is close to the system noise floor. Additional papers on the treatment of

uncertainty components (influence quantities) are listed in an informative bibliography (i.e., see Annex B).

An evaluation of each effect included in a measurement enables the identification of influence factors (i.e.,

elements that contribute to the uncertainty of the measurement), and a thorough description of each effect

can assist in assignment of an appropriate value and weighting factor indicating how the overall

measurement uncertainty budget may be affected. Each element of the measurement process involves an

uncertainty of measurement that can be assigned a value and described by the owner of the process. The

combinations of the elements of uncertainty form the basis of an uncertainty budget for a particular

measurement process.

4.2.1 General

Measurement uncertainty is a parameter associated with the result of a measurement that characterizes the

dispersion of the values that could reasonably be attributed to the true value of the measurand. It shows the

spread of values above and below the measurement result within which the true value of the measurand

may be expected to lie. It can be considered to be a measure of the possible error in the value of the

measurand provided by the result of a measurement; however, the true value of a measurand can never be

known. Measurement uncertainty should not be confused with correction or correlation factors. Correction

and correlation factors are quantities added to measurement values to create agreement between established

standard values and measurement device readings (these can also be described as biases: see A.1.4).

5

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Two types of evaluations are defined to determine the values for an uncertainty budget: Type A and Type

B. The type of evaluation used depends on what input and reference data is available. Other

electromagnetic compatibility (EMC) measurement uncertainty documents, such as CISPR 16-4-2 (an

international standard) and UKAS LAB 34:2002 [B21] (a UKAS guideline document), describe primarily

the Type B analyses and may use Type A methods to obtain only a few contributors, whereas Clause 5 of

this guide describes how to apply either method.

Type A evaluations of uncertainty are those obtained by using statistical methods where multiple

observations of the same event are recorded. These observed values are used to calculate the standard

deviation of the results. The standard deviation is then used to obtain the contribution of the process under

observation to the uncertainty budget.

Type B evaluations of uncertainty are those by any method other than a statistical evaluation. Knowledge

of previous performance of instruments, specifications, instrumentation reference data, and uncertainty data

provided with calibrations are examples of this evaluation type.

Either a Type A or a Type B analysis shall be carried out for each contributor to the uncertainty budget. A

Type A analysis has the advantage of providing a more representative evaluation because it is based on the

actual test setup/equipment used during measurement and, as such, generally results in a smaller

contributor value than a Type B analysis. However, a Type A analysis is more demanding in terms of time

and resources because it usually involves testing or analyzing extensive amounts of data.

A Type A evaluation of a contributor is determined by following the guidance for that particular

contributor (see 5.6) and finding the standard deviation for the data collected. For example, antenna factor

accuracy (also called antenna factor calibration) can be determined by taking the calibration value at each

frequency over the past five calibrations and determining the worst-case standard deviation over the

antenna’s usable frequency range. This could result in a smaller contributor than the Type B contributor. If

the Type B contributor for antenna factor accuracy was 0.6 dB, and the Type A analysis of the data yielded

a contributor of 0.2 dB, then it would be advantageous to use the smaller contributor.

Annex A includes a detailed presentation of measurement uncertainty concepts and methods of calculation

of contributors’ contribution to the measurement uncertainty budget based on their known or estimated

probability distribution.

Table 1 summarizes the formulas to be used for each type of probability distribution and provides cross-

references to corresponding subclauses from Annex A.

6

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Evaluation Probability Divisor, k Contributor, uS Annex A

type distribution subclause

Type A Normal (Gaussian) 1 Experimental standard A.2.1.1

deviation of the mean,

s (q )

Type A t-distribution 1 Adjusted experimental A.2.1.2

standard deviation of the

mean, η (ν ) s( q )

Type B Normal (Gaussian) Depends on expanded uncertainty A.2.2.2

confidence level

(usually k = 1 or 2; k

see Table A.1)

Type B t-distribution kp (Table A.3) expanded uncertainty A.2.2.3

k

p

Type B Rectangular A.2.2.4

3 a 3

Type B U-shaped A.2.2.5

2 a 2

Type B Triangular A.2.2.6

6 a 6

NOTE 1— s (q ) is the standard deviation of the mean, q , of the population of repeated observations, q .

j

()

NOTE 2— η ν is an expansion coefficient, used for t-distributed Type A analysis, which depends on the

number of degrees of freedom, ν . See Table A.2.

NOTE 3—kp is the coverage factor for the level of confidence p. For the t-distribution, see Table A.3.

NOTE 4—a is the half-width of the probability distribution, a = (a+ − a-) / 2, where a+ and a– define the

interval given by the prescribed limits of the individual uncertainty contributor.

Several approaches are available to the application of measurement uncertainty when determining

compliance of a EUT with EMC standards (e.g., limits). Some documents, like CISPR 16-4-2, include a

process for the application of calculated measurement uncertainty to the process of determining compliance

with emission limits; this procedure is used in documents like CISPR 11-2010 [B10]. Other documents,

like CISPR 22-2010 [B11], state that uncertainty must be calculated and reported but not used in adjusting

measured emission values. Other guidance documents have also been used by laboratories and

accreditation bodies. For example, Figure 1 in UKAS LAB 34:2002 [B21] shows various situations for

evaluating whether measurement results comply with a limit. Please refer to UKAS LAB 34:2002 [B21] for

more details on the application of measurement uncertainty to limits or measurements.

This approach is not considered herein because this document is intended to address only measurement

instrumentation uncertainty and associated parameters within the measurement chain. Adding uncertainty

to limits (other than the uncertainty of the measurement instrumentation chain) would need to be based on

either the interference potential of the device measured or the immunity characteristics of given products.

Such interference and immunity analyses are beyond the scope of this document.

7

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

5.1 General

This clause describes uncertainty contributors for the conducted and radiated emission EMC test methods,

and it illustrates EMC measurement instrumentation uncertainty estimations using example numeric values.

The contributors described in this clause are based on measurements performed as per ANSI C63.4. In the

case of other measurement standards, the list of contributors shall be adjusted accordingly to reflect the

measurement method therein.

⎯ Descriptions of contributors and equations used to calculate standard, combined, and expanded

uncertainties, and example Type B uncertainty template

⎯ Type A uncertainty statistical estimation

The Type B uncertainty for each test method is addressed in 5.2 through 5.5. The Type A uncertainties

apply to all methods and are addressed in 5.6. Example spreadsheets are included in 5.7.

Numerical examples are included for some contributors in the following subclauses. These are given only

for illustration purposes, and they are not meant to be used indiscriminately when preparing the

measurement uncertainty budget, as they may not be appropriate for a particular test laboratory. Each

laboratory shall consider the information available about its own particular measuring system, including

equipment characteristics, actual validation data of the test sites, the specific calibration data, the known or

likely probability distributions, and internal measurement procedures.

Table 2 lists all contributors to the measurement uncertainty budget, for both conducted and radiated

emissions test methods.

8

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Contributor Test method Subclause Subclause

(Type A (Type B

evaluation) evaluation)

Receiver reading Conducted and radiated 5.6.2.1 5.2.1

emissions

Attenuation of the connection between signal Conducted and radiated 5.6.2.2 5.2.2

transducer and receiver emissions

Receiver sine-wave voltage accuracy Conducted and radiated 5.6.2.3 5.2.3

emissions

Receiver pulse amplitude response Conducted and radiated 5.6.2.4 5.2.4

emissions

Receiver pulse response variation with repetition Conducted and radiated 5.6.2.5 5.2.5

frequency emissions

Receiver noise floor proximity Conducted and radiated 5.6.2.6 5.2.6

emissions

Mismatch effects between the receiver port of the Conducted and radiated 5.6.2.7 5.2.7

signal transducer and receiver emissions

Preamplifier gain variation Conducted and radiated 5.6.2.8 5.2.8

emissions

Instability of preamplifier Conducted and radiated 5.6.2.9 5.2.9

emissions

Line impedance stabilization network (LISN) Conducted emissions 5.6.3.2 5.3.2

voltage division factor

LISN impedance Conducted emissions 5.6.3.3 5.3.3

Antenna factor accuracy Radiated emissions 5.6.4.2 5.4.2 and 5.5.2

Antenna factor frequency interpolation Radiated emissions 5.6.4.3 5.4.3 and 5.5.3

Antenna factor variation with height Radiated emissions 5.6.4.4 5.4.4

Antenna directivity Radiated emissions 5.6.4.5 5.4.5 and 5.5.4

Antenna phase center location Radiated emissions 5.6.4.6 5.4.6 and 5.5.5

Antenna cross-polarization response Radiated emissions 5.6.4.7 5.4.7 and 5.5.6

Antenna balance Radiated emissions 5.6.4.8 5.4.8

Test site contribution Radiated emissions 5.6.4.9 5.4.9 and 5.5.7

Separation between equipment under test and Radiated emissions 5.6.4.10 5.4.10 and

measurement antenna 5.5.8

Height of table supporting the equipment under test Radiated emissions 5.6.4.11 5.4.11

Effect of setup table material supporting the Radiated emissions 5.6.4.12 5.4.12 and

equipment under test 5.5.9

5.2 Uncertainty contributors common to all conducted and radiated emission EMC

measurements—Type B evaluation

The receiver reading will vary for different reasons, such as input attenuation switching uncertainty,

detector nonlinearity, and marker resolution (round-off error). Estimates of these contributors can be taken

from manufacturer specifications or from the calibration certificate of the receiver. A rectangular

probability distribution shall be used in either case. The use of calibration certificate data will significantly

reduce the values of these contributors; however, in this case, the uncertainties of measurement for each

contributor, as reported by the calibration laboratory, must be included in the overall uncertainty budget.

9

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

EXAMPLE 1—The calibration certificate reports a worst-case deviation from linearity (detector nonlinearity) of

+0.12 dB/−0.11 dB. Using a rectangular distribution, the contributor value to be used in the overall budget will be [0.12

− (−0.11)] / (2 √3) = 0.066 dB. See A.2.2.4 for details on how to calculate the contributor value in case of rectangular

probability distribution.

EXAMPLE 2—The calibration laboratory reported an expanded measurement uncertainty for detector nonlinearity

measurement of ±0.078 dB, also stating that this is based on a normal distribution with a coverage factor k = 2. This

must be included as a separate contributor in the overall budget as [0.078 − (−0.078)] / (2k) = 0.039 dB. See A.2.2.2 for

details on how to calculate the contributor value in case of normal (Gaussian) probability distribution.

When evaluating the contributors affecting the receiver reading, the user must be careful not to double-

count specific contributors. For example, a frequency response will be included in the receiver sine-wave

voltage accuracy, which is separately considered in 5.2.3. For more details, refer to Medler [B18] and

Stecher [B20].

An estimate of the attenuation of the connection between the receiver and the signal transducer (LISN,

current probe, or antenna) is assumed to be available from a calibration report, along with an expanded

uncertainty and a coverage factor. If the estimate of attenuation is obtained from manufacturer’s data for a

cable or attenuator, then a rectangular probability distribution having a half-width equal to the

manufacturer’s specified tolerance on the attenuation may be assumed. If the connection is a cable and an

attenuator, with the manufacturer’s data available on each, then this contribution has two components, each

with its own rectangular probability distribution.

An estimate of the magnitude for receiver sine-wave voltage accuracy is assumed to be available from a

calibration report, along with an expanded uncertainty and a coverage factor. If a calibration report states

only that the receiver sine-wave voltage accuracy meets the ANSI C63.2 specification (i.e., ±2 dB), the

estimate of the contributing factor should be taken as zero with a rectangular probability distribution having

a half-width of 2 dB.

In general, it is not practical to correct for receiver pulse response characteristics. A verification report

stating that the receiver pulse amplitude response complies with the CISPR 16-1-1:2007 tolerance of

±1.5 dB (at the reference pulse repetition frequency) for peak, quasi-peak, or average detection is assumed

to be available. The magnitude is estimated to be zero with a rectangular probability distribution having a

half-width of 1.5 dB.

If an emission is narrowband in nature [e.g., continuous wave (CW) signal], then pulse response corrections

need not be considered.

In general, it is not practical to correct for receiver pulse response characteristics. The CISPR 16-1-1

tolerance for pulse repetition rate response varies with repetition rate and detector type. A verification

report stating that the receiver pulse repetition rate responses comply with the CISPR 16-1-1 tolerances is

assumed to be available. The correction is estimated to be zero with a rectangular probability distribution

having a half-width of 1.5 dB, a value considered to be representative of the various CISPR 16-1-1

tolerances.

10

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

If an emission is narrowband in nature (e.g., CW signal), then pulse response corrections need not be

considered.

The noise floor of an EMI receiver meeting CISPR requirements is usually sufficiently below the emission

voltage or current limit that its effect is negligible on measurement results close to these limits. However,

for radiated emissions measurements, the proximity of the receiver noise floor may influence measurement

results near the radiated emissions limit. For radiated emissions measurements, the magnitude is estimated

to be zero with an expanded uncertainty of 0.5 dB and a coverage factor of 2 (normal distribution).

5.2.7 Mismatch effects between the receiver port of the signal transducer and the receiver

In general, the receiver port of a signal transducer (LISN, current probe, or antenna) will be connected to

port 1 of a two-port network whose port 2 is terminated by a receiver of reflection coefficient Γr. The two-

port network, which might be a cable, an attenuator, a combination of attenuator and cable, or some other

combination of components, can be represented by its S-parameters. The mismatch correction is then (see

UKAS LAB 34:2002 [B21], Carpenter [B5], and Carpenter [B6]):

[

δM = 20 log10 ( 1 − Γe S11 )( 1 − Γr S 22 ) − S 21

2

Γe Γr ] (1)

where Γe is the reflection coefficient observed looking into the receiver port of the LISN with the EUT

connected, or looking into the output port of the antenna when it is set up for disturbance measurement. All

parameters are with respect to 50 Ω.

When only the magnitudes, or extremes of magnitudes, of the parameters are known, it is not possible to

calculate δM, but its extreme values δM± are not greater than:

[ (

δM ± = 20 log10 1 ± Γe S11 + Γr S 22 + Γe Γr S11 S 22 + Γe Γr S 21

2

)] (2)

The probability distribution of δM is approximately U-shaped, with width not greater than (δM+ − δM−) and

standard deviation not greater than the half-width divided by √2.

It is usually assumed that the connection to the receiver from the transducer is a well-matched cable (i.e.,

S11 << 1 and S22 << 1 ) of negligible attenuation (i.e., S 21 ≈ 1 ), and that the receiver radio frequency (RF)

attenuation is set to 10 dB or more, for which the CISPR 16-1-1:2007 voltage standing wave ratio (VSWR)

tolerance of VSWR ≤ 1.2:1 can be assumed for the calculation, as a worst-case estimate.

In radiated emission measurements, the VSWR of the antenna can be a major contributor to the

measurement uncertainty, due to the resulting mismatch effect. The VSWR of antenna used shall be

known, and well-matched attenuators should be used at the antenna connector, as necessary, to meet

antenna VSWR requirements per CISPR 16-1-4:2010.

If the receiver RF attenuation is set to 0 dB, then the corresponding VSWR must be used for mismatch

calculations. Likewise, if the receiver input attenuation is set to 10 dB or more, then the associated lower

VSWR is used. The VSWR specifications per CISPR 16-1-1:2010 are shown in Table 3.

11

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Frequency range Input attenuation Maximum VSWR

0 dB 2.0:1

9 kHz to 1 GHz

10 dB or higher 1.2:1

0 dB 3.0:1

1 GHz to 18 GHz

10 dB or higher 2.0:1

If a preamplifier external to the receiver is used, then its contribution to the mismatch uncertainty due to the

input and output VSWR is also to be included in the overall uncertainty calculation. In this case, two

mismatch uncertainties have to be considered: between antenna/LISN and preamplifier input port and

between the preamplifier output port and the receiver input port. Please note that the values used for these

calculations must be supported by calibration of the instruments and accessories.

Calibrated preamplifiers are used either at the input of the measuring receiver or built into the measuring

receiver. Usually, external preamplifiers are only used for radiated emissions above 1 GHz; however, they

may be occasionally also used for radiated emissions within 30 MHz to 1000 MHz or even for conducted

emissions measurements.

If an external preamplifier is used in the test setup, then the accuracy of its gain calibration shall be

considered as an additional uncertainty contribution. It shall be obtained from a calibration certificate,

which has to provide the expanded uncertainty and a coverage factor. The magnitude is estimated to be

0.4 dB (expanded uncertainty) with a coverage factor of 2 (normal distribution).

Any preamplifier gain variations that are not taken into account by the internal built-in calibration process

of the receiver need to be considered in the uncertainty estimate. An estimate of the preamplifier gain is

assumed to be available from a calibration report, along with an expanded uncertainty and a coverage factor

(see 5.2.8). Any additional gain deviations from the calibrated frequency response (e.g., due to temperature

changes in the test environment or aging) have to be taken into account as an uncertainty contribution. The

estimate of the magnitude for this uncertainty component is ±0.7 dB with a rectangular distribution.

port using an LISN—Type B evaluation

5.3.1 General

In addition to the contributors included in 5.2, uncertainty quantities to be considered for conducted

disturbance measurements at a mains port using an LISN are as follows.

An estimate of the LISN voltage division factor is assumed to be available from a calibration report, along

with an expanded uncertainty and a coverage factor.

12

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The impedance tolerance in ANSI C63.4 for a 50 Ω/50 μH + 5 Ω AMN or a 50 Ω/50 μH LISN requires the

impedance to be within 20% of the magnitude of the nominal impedance when the receiver port is

terminated in 50 Ω. The estimate of the correction is zero with a probability distribution bounded by the

extremes from all combinations of the constrained LISN impedance and unconstrained EUT impedance

over the defined frequency range. A triangular probability distribution is assumed because there is a low

probability of encountering the particular combinations of frequency, LISN impedance, and EUT

impedance needed to produce those extremes.

range 30 MHz to 1000 MHz on an open-area test site (OATS) or in a semianechoic

chamber (SAC)—Type B evaluation

5.4.1 General

In addition to the contributors included in 5.2, uncertainty quantities to be considered for radiated emission

measurements on an OATS or in a SAC, in the frequency range 30 MHz to 1000 MHz, are as follows.

An estimate of the free-space antenna factor is assumed to be available from a calibration report, along with

an expanded uncertainty and a coverage factor. The uncertainty of the antenna calibration may vary,

dependent on the antenna type, the applied calibration method, and the frequency range.

When an antenna factor value is calculated by linear interpolation between frequencies at which calibration

data are available, the uncertainty associated with that antenna factor depends on the frequency interval

between calibration points and the variability of the antenna factor with frequency. Plotting the calibrated

antenna factor versus frequency can help to estimate the actual magnitude of this contribution. The

uncertainty can be reduced by requesting a large number of frequency points (e.g., 1601) from the

calibration laboratory, which can easily be achieved by using a vector network analyzer for the antenna

calibration process.

The antenna factor changes due to mutual coupling of the antenna with its image in the ground plane. When

an antenna is scanned in height above a well-conducting ground plane, the average antenna factor is close

in magnitude to the free-space antenna factor. The height scan must be at least half a wavelength with

readings taken at intervals of one eighth of a wavelength or less, with the lowest height greater than a third

of a wavelength. The change in mutual impedance is most sensitive for tuned dipoles. Assuming that the

longest dipole is tuned at 80 MHz, a height of no more than 4 m is required. The uncertainty contribution is

the possible deviation from the free-space antenna factor. If this uncertainty contribution varies

significantly over the frequency range, then a correction is applied to the measurement data for each

frequency range. In general, this contribution decreases as the frequency increases, and it becomes

negligible above 300 MHz.

13

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

CISPR 16-1-4:2010 requires the responses of a complex antenna in the direction of the direct ray and in the

direction of the ground-reflected ray to be within 1 dB of the maximum response. To meet this requirement,

the main lobe of the complex antenna may need to be tilted downward, particularly at separations of less

than 10 m. The uncertainty contribution for the effect of directivity is 0 dB for an antenna having a uniform

pattern in the vertical plane, and between 0 dB and +xi dB for an antenna having a nonuniform pattern in

the vertical plane, where xi dB is given in Table 4 through Table 7 for horizontally aligned and for tilted

antennas. CISPR 16-1-4:2010 gives guidance on the maximum gain allowable for a biconical or

logarithmic-periodic antenna for which the values of xi in Table 4 through Table 7 apply.

A horizontally polarized biconical antenna is assumed to have a uniform pattern in the vertical plane. A

vertically polarized biconical antenna, and a horizontally or vertically polarized log-periodic antenna, is

assumed to require a correction of up to +xi/2 dB at separations of 3 m and 10 m, but not more than +0.15

dB at a separation of 30 m.

The estimate of the correction is +xi/2 dB with a rectangular probability distribution of the uncertainty

having half-width of xi/2 dB.

NOTE 1—The reduction in signal strength caused by reduced directivity at angles off antenna bore sight is a

systematic error and therefore can be corrected; a nonzero estimate with reduced uncertainty can be evaluated from the

known pattern of the measuring antenna and applied as a function of frequency and separation.

NOTE 2—In Table 4 through Table 7, the uncertainty contribution is +xi/2 dB for vertical polarization with an EUT

height of 1.5 m. The value is positive because it represents a reduction in field strength.

using a biconical antenna at distances of 3 m, 10 m, or 30 m

Magnitude Probability Normalized

Uncertainty contribution

(dB) distribution magnitude (dB)

Directivity difference at 3 m ±0.0 Rectangular 0.00

at 10 m ±0.0 Rectangular 0.00

at 30 m ±0.0 Rectangular 0.00

Table 5 —Vertically polarized radiated emissions from 30 MHz to 200 MHz using a

biconical antenna at distances of 3 m, 10 m, or 30 m

Magnitude Probability Normalized

Uncertainty contribution

(dB) distribution magnitude (dB)

Directivity difference at 3 m (<130 MHz) +1.0/–0.0 Rectangular 0.29

at 3 m (>130 MHz) +2.0/–0.0 Rectangular 0.58

at 3 m with tilting +1.0/–0.0 Rectangular 0.29

at 10 m +0.5/–0.0 Rectangular 0.14

at 30 m +0.2/–0.0 Rectangular 0.06

14

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

using a logarithmic-periodic antenna at a distance of 3 m, 10 m, or 30 m

Magnitude Probability Normalized

Uncertainty contribution

(dB) distribution magnitude (dB)

Directivity difference at 3 m +2.0/–0.0 Rectangular 0.58

Directivity difference at 3 m with tilting +1.0/–0.0 Rectangular 0.29

at 10 m +0.7/–0.0 Rectangular 0.20

at 30 m +0.2/–0.0 Rectangular 0.06

using a logarithmic-periodic antenna at a distance of 3 m, 10 m, or 30 m

Magnitude Probability Normalized

Uncertainty contribution

(dB) distribution magnitude (dB)

Directivity difference at 3 m +2.5/–0.0 Rectangular 0.73

Directivity difference at 3 m with tilting +1.2/–0.0 Rectangular 0.35

at 10 m +1.0/–0.0 Rectangular 0.29

at 30 m +0.3/–0.0 Rectangular 0.09

The uncertainty contribution due to phase center location variation is negligible for a biconical antenna, but

the change in phase-center location with frequency for a log-periodic antenna causes a deviation from the

required measurement distance.

For a log-periodic antenna, the magnitude is determined by considering the effect of an error of ±0.35 m in

the separation and assuming that field strength is inversely proportional to separation.

NOTE 1—If a dipole is the measuring antenna, then this uncertainty contribution is negligible.

NOTE 2—For hybrid antennas, the magnitude will increase if no correction for the phase center variation is applied for

this systematic effect.

The cross-polarization response of a biconical antenna is considered to be negligible. The estimate of the

cross-polarization response of a log-periodic antenna is assumed to be ±0.9 dB with a rectangular

distribution, corresponding to the CISPR 16-1-4:2010 cross-polarization response tolerance of ±20 dB.

y−x

20 log10 1 ± 10 20 (3)

The cross-polarization error of the antenna used for measurements can be measured by placing it in a plane

wave. The antenna under test is set to the same polarization as the transmit antenna (vertical polarization is

better because it minimizes ground reflections), and the value of x is recorded. Then, without changing the

transmitting antenna, the receive antenna is changed to the opposite polarization, and the value of y is

recorded. This test must be performed at a sufficient number of frequencies evenly spread across the

30 MHz to 1000 MHz range. The magnitude of the worst-case cross-polarization error thus measured is

considered as the contributor’s value, with a rectangular distribution.

15

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

NOTE 1—At the receive antenna, a plane wave must be realized (no ground or wall reflections).

NOTE 2—If a dipole is used as the measuring antenna, then the cross-polarization response is negligible.

The effect of an unbalanced antenna is greatest when the input coaxial cable is aligned parallel to the

antenna elements. Antenna balance is predominantly an issue for biconical antennas. It is caused by the

imbalance of the balun used to feed the elements of the biconical antenna. This imbalance causes the

coupling of surface currents on to the cable. The estimate of the magnitude is based on the performance of

commercially available antennas, with a rectangular distribution. Table 8 summarizes the values.

Antenna Polarization Magnitude Normalized

magnitude

Biconical Horizontal ±0.3 0.17

Vertical ±0.9 0.52

Logarithmic-periodic Horizontal ±0.0 0.0

Vertical ±0.3 0.17

The magnitude Δmax of the maximum difference between theoretical site attenuation, on the one hand, and

measured site attenuation increased by the site attenuation measurement uncertainty, on the other hand,

provides an indication of the effect that site imperfection may have on a disturbance measurement. The

tolerance for this difference per ANSI C63.4 is ±4 dB. However, the measurement uncertainty associated

with the site-attenuation measurement method is usually large and is dominated by the two antenna factor

uncertainties. Therefore, a site that meets the 4 dB tolerance is unlikely to have imperfections sufficient to

cause errors of 4 dB in disturbance measurements. In recognition of this fact, a triangular probability

distribution is assumed for the determination of the impact of this uncertainty component. Therefore, the

worst-case estimate for the magnitude of this contribution is ±4 dB with a triangular probability

distribution.

NOTE—If Δmax is less than 4 dB, then the estimate can be assumed as the actual Δmax with a triangular probability

distribution.

The error in separation is due to errors in determining the perimeter of the EUT and distance measurement.

The magnitude of this uncertainty contribution is based on assuming a maximum separation error of ±0.1 m

and on the assumption that field strength is inversely proportional to separation over that distance margin.

This uncertainty is caused by the placement of the equipment under test on a table of a height other than the

nominal height of 0.8 m. The magnitude estimate is based on a maximum deviation of ±0.01 m from the

nominal table height, resulting in an expanded uncertainty of 0.1 dB. A normal distribution is assumed and

a coverage factor k = 2 is used, for approximately 95% level of confidence.

16

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

CISPR 16-1-4:2010 describes a method for the evaluation of the effect of the table material up to 1 GHz.

No tolerance is given for this effect. Below 200 MHz, the magnitude of this uncertainty contribution is

assumed to be zero. Above 200 MHz, the estimate of the magnitude is ±0.5 dB with a rectangular

distribution function. The actual magnitude for a specific setup must be derived from measurements, using

the procedure outlined in CISPR 16-1-4:2010.

range above 1 GHz—Type B evaluation

5.5.1 General

In addition to the contributors included in 5.2, uncertainty quantities to be considered for radiated emission

measurements above 1 GHz in a quasi–free-space environment are as follows.

An estimate of the free-space antenna factor is assumed to be available from a calibration report, along with

an expanded uncertainty and a coverage factor. The stated uncertainty for the antenna factor is to be used in

the measurement uncertainty calculation, not the antenna factor itself. The uncertainty of the antenna

calibration may vary, dependent on the antenna type, the applied calibration method, and the frequency

range.

When an antenna factor value is calculated by interpolation between frequencies at which calibration data

are available, the uncertainty associated with that antenna factor depends on the frequency interval between

calibration points and the variability of antenna factor with frequency. Plotting calibrated antenna factor

versus frequency may help to estimate the actual magnitude of this contribution. The uncertainty can be

reduced by requesting a large number of frequency points (e.g., 1601) from the calibration laboratory,

which can easily be achieved by using a vector network analyzer for the antenna calibration process.

The directivity of the receive antenna determines the value w [i.e., per Equation (10) of

CISPR 16-2-3:2006-2007 [B11], reproduced below], which is used to determine the need for an antenna

height scan. The parameter w is calculated assuming that the far-field criterion is valid. At close

measurement distances, measurements are made in the Fresnel zone and not in the far field. The actual

dimension w is different from the value obtained from Equation (4):

(

w = 2d tan 0.5θ 3 dB ) (4)

where

q3 dB is the receive antenna beam width at the 3 dB points

17

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The impact of the receive antenna properties on the uncertainty is also determined by frequency, the size of

the EUT, and the measurement distance. The resulting value of the uncertainty is not easily determined. At

higher frequencies, some receive antennas have multiple lobes instead of one main lobe. This may cause

additional instrumentation uncertainties, which are not considered here. The magnitude estimate of this

uncertainty parameter is +3/–0 dB with a rectangular probability distribution, assuming that the EUT

dimension is larger than the parameter w, based on the antenna radiation pattern.

The uncertainty contribution due to phase center location for a horn antenna causes a deviation from the

required measurement distance.

For a horn antenna, the magnitude is determined by considering the effect of an error of ±0.10 cm in the

separation and assuming that field strength is inversely proportional to separation.

NOTE 1—If a dipole is the measuring antenna, then this uncertainty contribution is negligible.

NOTE 2—For hybrid antennas, the magnitude will increase if no correction for the phase center variation is applied for

this systematic effect.

The cross-polarization response of a dipole type antenna is considered to be negligible. The estimate of the

cross-polarization response of a log-periodic antenna is assumed to be ±0.9 dB with a rectangular

distribution, corresponding to the CISPR 16-1-4:2010 cross-polarization response tolerance of 20 dB. The

cross-polarization response for a horn antenna is considered negligible.

The measured site voltage standing wave ratio SVSWR provides an indication of the effect that site

imperfection may have on an emission measurement. CISPR 16-1-4:2010 states a tolerance for the SVSWR

of 6 dB. However, a site that meets the 6 dB SVSWR tolerance will not cause errors of 6 dB in disturbance

measurements. A useful comparison between SVSWR and deviation from the reference transmission loss for

a 3 m site has been made in document CISPR/A/838/INF (2009) [B7]. In this document, it is shown that a

SVSWR of 6 dB corresponds roughly to a 4 dB deviation from transmission loss. Assuming that the

transmission loss has a Gaussian probability distribution and because the value of 4 dB is not exceeded in

the entire frequency range, a value of 4 dB is assumed as the corresponding expanded uncertainty with a

coverage factor k = 3 (i.e., the standard uncertainty is 1.33 dB).

NOTE—If SVSWR is less than 6 dB, then the estimate of the uncertainty contribution may be determined as

[4 × (SVSWR / 6)] dB, assuming a normal distribution and a coverage factor k = 3.

The error in separation is due to errors in determining the perimeter of the EUT and distance measurement.

The magnitude of this uncertainty contribution is based on assuming a maximum separation error of ±0.1 m

and on the assumption that field strength is inversely proportional to separation over that distance margin.

18

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

CISPR 16-1-4:2010 describes a method for the evaluation of the effect of the table material above 1 GHz.

No tolerance is given for this effect. From 1 GHz to 6 GHz, the estimated magnitude of this uncertainty

contribution is ±1.5 dB with a rectangular distribution function. Above 6 GHz, the estimate of the

magnitude is ±2.0 dB with a rectangular distribution function. The actual magnitude for a specific setup

must be derived from measurements, using the procedure outlined in CISPR 16-1-4:2010.

5.6.1 General

Type A evaluation of individual contributors can be used to assess uncertainty as long as the evaluation

methodology is sound and based on a statistical analysis of data. The following are examples of the

methods that can be employed to assess a contributor. These examples are by no means the only approach.

The specific contributors listed in this subclause are not the only areas where a Type A evaluation approach

can be used. These are merely examples of how to apply such an approach. Type A evaluations of test

systems, as opposed to contributors, are also possible. For further information on using repeatability and

reproducibility studies including interlaboratory comparisons in Type A studies, see ISO 21748:2004 and

ISO 21749:2005.

When the number of observations made is sufficiently large, a normal distribution can be assumed and the

experimental standard deviation of the mean is used as the uncertainty value. However, if a reduced

number of observations is carried out, then a Student’s t-distribution shall be used, and the calculated

experimental standard deviation of the mean shall be adjusted based on the number of observations made

(see A.2.1.2). For both normal distribution and t-distribution, a coverage factor (divisor) k = 1 shall be used

for computing the contributor’s value (see A.2.1.1 and A.2.1.2).

For a given influence quantity, the decision between using normal or t-distribution for the Type A analysis

depends on the acceptable level of error in the resulting uncertainty budget. The expanded uncertainty shall

be first calculated using the contributor’s value derived by assuming a normal distribution (1 standard

deviation; see A.2.1.1) and then by using the standard deviation adjusted with the correction factor

corresponding to the number of observations performed (see Table A.2 and A.2.1.2). The difference

between the two, expressed as a percentage of the later, represents the error in the expanded uncertainty

budget.

For less than five observations, it is recommended not to use a Type A estimation, but to perform a Type B

analysis instead.

In the context of a Type A evaluation, repeated observations may consist of a set of repeated measurements

or of using a number of previous calibration certificates of a given measurement equipment to calculate the

standard deviation of a specific parameter. Care shall be taken that only the contributor evaluated is varied

during the repeated observations, while all others are kept constant, to avoid double or multiple counting of

contributors in the resulting overall uncertainty budget. As such, a Type A analysis based on previous

calibration certificates cannot usually be used for contributors that are or may be susceptible to aging

and/or drift (except when the aging and/or drift effects are not accounted for through other contributors in

the overall uncertainty budget).

19

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

5.6.2 Uncertainty contributors common to all conducted and radiated emission EMC

measurements—Type A evaluation

Receiver readings will vary for different reasons, such as measuring system instability and meter scale

interpolation errors. The estimate of this parameter is the mean of many readings, with a standard

uncertainty given by the experimental standard deviation of the mean (k = 1).

The attenuation of the cable used to connect the signal transducer (LISN, current probe, or antenna) to the

receiver can be measured using a separate set of instruments such as a network analyzer or a spectrum

analyzer with a signal generator. The standard deviation of the repeated measurements is the contribution

value.

The Type B evaluation of this contributor is discussed in 5.2.3. An estimate of the magnitude for receiver

sine-wave voltage accuracy can be found in the calibration certificate. Multiple receiver calibrations will

result in multiple results for the sine wave accuracy. These calibration results can be used as the basis for a

Type A evaluation having a normal distribution. Because the sample size is very likely to be relatively

small, a Student’s t-distribution should be used. An alternative approach would be for a test laboratory to

conduct an analysis of variance (ANOVA) of the sine wave accuracy, using the calibration process defined

in the calibration manual of the instrument manufacturer and, if possible, including multiple units of the

same equipment type in the study to avoid biasing.

NOTE—If the test laboratory does not use multiple units for the ANOVA, then a bias may be present in the Type A

evaluation and not be truly representative of the contributor. Caution should be used if the ANOVA is used such that

multiple units of the same device are used.

The Type B evaluation of this contributor is covered in 5.2.4. It is important to note that this factor is only a

contributor when measuring a noncontinuous wave. During the calibration, continuously repeated pulses

are applied and the response measured for each quasi-peak (QP) and average (AVG) detector bandwidth

setting. The value is recorded. Then, the difference is calculated between the measured amplitude value and

the target for each bandwidth of interest. This procedure is repeated for a number of frequency points per

each detector per each bandwidth setting to verify the receiver compliance with the pulse amplitude

response requirements of CISPR 16-1-1:2007. If each one of these measurements is repeated a number of

times, then the results will form a small statistical data set, which can then be used in a Type A analysis to

compute the value of this contributor.

Because the receiver pulse response is subject to aging and temperature drift, this procedure can only be

used if the repeated calibration cycles are all performed within a relatively short period of time (not by

using previous years’ calibration reports).

20

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The Type B evaluation of this contributor is covered in 5.2.5. It is important to note that this is only a

contributor to the measurement uncertainty of broadband signals and that it is not a factor in uncertainty of

measurements of narrowband emissions. Similar to the pulse amplitude response, this contributor is

assessed during the normal instrument calibration. The response of the instrument to different pulse

repetition rates over numerous calibration cycles can be used as small sample Type A evaluation. The

calibration report for the instrument will provide the amplitude error at each pulse repetition rate for each

QP and AVG bandwidth setting. These data, taken from multiple calibration cycles, form a small statistical

data set and the standard deviation can be computed at each pulse rate for each bandwidth. The greatest

resulting standard deviation (i.e., of the different pulse rates) should be used for each bandwidth of interest.

Because the receiver pulse response is subject to aging and temperature drift, this procedure can only be

used if the repeated calibration cycles are all performed within a relatively short period of time (not by

using previous years’ calibration reports).

This contributor is negligible if the receiver noise floor is more than 10 dB below the signal level that

corresponds to the applicable emissions limit. If this margin is less than 10 dB, then the following

procedure can be employed for estimating the contribution to the conducted or radiated emissions

measurement uncertainty budget due to the receiver noise floor proximity. The example provided is given

for radiated emissions within 30 MHz to 1000 MHz, but a similar procedure can be used for conducted

emissions (below 30 MHz) or for radiated emissions above 1 GHz.

To perform this analysis, a reference EUT is required, which exhibits stable (in both level and frequency)

radiated emissions at or close to the applicable limit, at a set of frequencies evenly spread out across the

30 MHz to 1000 MHz band.

Repeated measurements shall be performed at all frequencies radiated by the reference EUT, with each

measurement (at each frequency) taken twice: with the EUT ON and with the EUT OFF. For each

measurement, the ratio of [signal + noise (EUT ON)] to the [noise (EUT OFF)] shall be computed. The error

in radiated emissions measurement due to the receiver noise floor proximity can be found using the

calculated (S + N)/N value and the plots from Moulthrop and Muha [B19]. This procedure shall be repeated

for a minimum of five iterations, and the standard deviation shall be calculated for each test frequency. The

highest standard deviation thus obtained, from across all investigated frequency points, is the contribution

value.

There is no efficient means to assess this contributor with a Type A method because of the following

reasons:

⎯ For conducted emissions, the impedance at the LISN receiver port will be affected by each

unique product connected for test.

⎯ For radiated emissions, the impedance at the antenna port will be affected by the test site.

21

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

This contributor needs to be considered if a preamplifier external to the receiver is used during the

measurements. It also needs to be considered when a preamplifier internal to the receiver is used and when

that preamplifier’s gain variation is not taken into account during the receiver calibration routine (or when

the receiver calibration routine is not performed before the measurements).

A Type A evaluation of this contributor can be performed by repeatedly measuring the preamplifier gain,

computing the standard deviation of the obtained results at each measured frequency, and taking the worst-

case (largest) such standard deviation across the applicable frequency range as the contributor’s value.

Alternatively, the Type A analysis can be performed by using the previous year’s calibration certificates. A

minimum of 5 previous years’ (most recent) calibration reports shall be used; otherwise, a Type B analysis

shall be performed. If the Type A analysis is based on old calibration reports and not on repeated

measurements, then the effect of aging on the preamplifier gain will also be accounted for, and in this case,

a separate contributor for the aging effect shall not be used in the uncertainty budget.

This contributor needs to be considered if a preamplifier external to the receiver is used during the

measurements. It also needs to be considered when a preamplifier internal to the receiver is used and when

that preamplifier’s gain variation is not taken into account during the receiver calibration routine (or when

the receiver calibration routine is not performed before the measurements).

Instability of preamplifier gain is due to two contributors: aging and temperature drift.

The first one can only be evaluated through a Type A analysis of the previous years’ calibration reports

(see 5.6.2.8) and not through repeated measurements. In this case, only one contributor will be used in the

uncertainty budget to account for both preamplifier gain variation and instability of preamplifier gain due

to aging.

The temperature drift effect need only be considered if a large variation in ambient temperature exists

between different measurements (usually only for radiated emissions measurement on an OATS). This

contributor cannot be evaluated from previous calibration certificates. A Type A analysis of the

preamplifier gain temperature drift may be performed by taking repeated measurements, at various

temperatures within the expected temperature range and at different frequencies within the applicable

frequency range, calculating the standard deviation of the obtained results at each measured frequency and

taking the worst-case (largest) such standard deviation across the applicable frequency range as the

contributor’s value. Note that a minimum of five measurements must be taken at each frequency, at

different temperatures, and the selected temperature values must be the same for all frequency points and

must be evenly spread across the expected temperature range.

When a statistically representative number of measurements (i.e., minimum five) is taken at each frequency

and temperature combination, then the resulting contributor’s value will account for both preamplifier gain

variation and instability of preamplifier gain due to drift.

22

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

using an LISN—Type A evaluation

5.6.3.1 General

In addition to the contributors included in 5.6.2, uncertainty quantities to be considered for conducted

disturbance measurements at a mains port using an LISN are as follows.

This contributor can be assessed by measuring the voltage division factor repeatedly and computing the

standard deviation of those repeated measurements.

This contributor can be assessed by repeating the calibration procedure as specified in ANSI C63.4 and

computing the standard deviation of those repeated measurements.

5.6.4 Uncertainty contributors for radiated emission measurements in the frequency range

30 MHz to 1000 MHz on an OATS or in an SAC—Type A evaluation

5.6.4.1 General

In addition to the contributors included in 5.6.2, uncertainty quantities to be considered for radiated

emission measurements in the frequency range 30 MHz to 1000 MHz on an OATS or in a SAC are as

follows.

A Type A evaluation of the contribution of the antenna factor uncertainty can be made by computing the

standard deviation of the previously measured antenna factors (minimum five measurements).

This analysis is to be performed for each frequency of calibration. The maximum standard deviation

obtained from across all analyzed frequency points shall be used as the contributor value.

Because the antenna factor is subject to temperature drift, this procedure can be used only if the repeated

calibration cycles are all performed within a relatively short period of time (not by using previous years’

calibration reports). However, if measurements are always performed within a SAC, then a Type A analysis

based on previous years’ calibration reports can be used instead of the repeated measurements because the

temperature inside a SAC is controlled and relatively constant.

There is no efficient means to assess this contributor with a Type A method. A Type B evaluation of this

contributor should be used instead.

23

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The antenna calibration processes for Standard Site Method (SSM) and Reference Antenna Method (RAM)

include this variable. Consequently, a series of at least five antenna calibrations of the same type (e.g., SSM

or RAM) will provide a sufficient data set for a Student’s t-distribution.

NOTE 1—Repeated calibrations will account not only for this contributor but also for the antenna factor (AF)

accuracy, directivity, and phase center location contributors. As such, if this Type A analysis is performed, then the

resulting contributor's value shall be used in the overall budget to represent all these contributors. It should be noted

that the other contributors should not be included in the overall budget to ensure the contributor is not counted more

than one time.

NOTE 2—The reference antenna method in ANSI C63.5-2006 [B1] does not require a receive antenna height scan. It is

recommended but not mandatory (see subclause 6.1 of ANSI C63.5-2006). If the antenna calibration does not include a

height scan, then repeating this method in a Type A analysis will not account for this contributor to uncertainty

resulting in the need for a Type B evaluation of this contributor.

If a complete data set is not available, then a Type B evaluation of this contributor should be used instead.

The antenna calibration processes for Standard Site Method and Reference Antenna Method include this

variable. Consequently, a series of at least five antenna calibrations of the same type (e.g., SSM or RAM)

will provide a sufficient data set for a Student’s t-distribution.

NOTE 1—Repeated calibrations will account not only for this contributor but also for the AF accuracy, directivity, and

phase center location contributors. As such, if this Type A analysis is performed, then the resulting contributor's value

shall be used in the overall budget to represent all these contributors. It should be noted that the other contributors

should not be included in the overall budget to ensure the contributor is not counted more than one time.

NOTE 2—The reference antenna method in ANSI C63.5-2006 [B1] does not require a receive antenna height scan. It is

recommended but not mandatory (see subclause 6.1 of ANSI C63.5-2006 [B1]). If the antenna calibration does not

include a height scan, then repeating this method in a Type A analysis will not account for this contributor to

uncertainty resulting in the need for a Type B evaluation of this contributor.

NOTE 3—The Standard Site Method does not include testing in the vertical polarization. Therefore, if this method is

used for antenna calibration, then either the laboratory needs to request data in the vertical polarization or a Type B

evaluation for the vertical polarization is to be used.

NOTE 4—The Reference Antenna Method does not specify a polarization. Therefore, if the laboratory desires to use a

Type A analysis of this contributor, then the laboratory must request data in both polarizations; otherwise, a Type B

evaluation is to be used.

If a complete data set is not available, then a Type B evaluation of this contributor should be used instead.

The antenna calibration processes for Standard Site Method and Reference Antenna Method include this

variable. Consequently, a series of at least five antenna calibrations of the same type (e.g., SSM or RAM)

will provide a sufficient data set for a Student’s t-distribution.

NOTE—Repeated calibrations will account not only for this contributor but also for the AF accuracy, directivity, and

phase center location contributors. As such, if this Type A analysis is performed, then the resulting contributor's value

shall be used in the overall budget to represent all these contributors. It should be noted that the other contributors

should not be included in the overall budget to ensure the contributor is not counted more than one time.

24

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

If a complete data set is not available, then a Type B evaluation of this contributor should be used instead.

There is no efficient means to assess this contributor with a Type A method. A Type B evaluation of this

contributor should be used instead.

Perform at least five measurements, which will provide a sufficient data set for a Student’s t-distribution, of

balance of the test antenna using the procedure described in ANSI C63.5. The standard deviation of the

error contribution shall be used as the Type A value.

If a complete data set is not available, then a Type B evaluation of this contributor should be used instead.

The contribution to measurement uncertainty from the test site and system can be estimated using a Type A

evaluation. This evaluation requires performing the volumetric normalized site attenuation as described in

ANSI C63.4. This evaluation requires five measurements that are taken at different locations on the ground

plane for each geometry and antenna. The standard deviation of these five values at each frequency

represents the worst-case contribution to uncertainty that can be associated with the site imperfections,

cable lay variations, distance measurement errors, antenna height start/stop errors, antenna travel

smoothness (directivity), and height versus frequency variations. Amplifier gain stability would also be

included in this value if the same amplifier were used during calibrations. The standard deviation in the

data of the five locations shall be calculated for each frequency. The contribution to measurement

uncertainty is then the maximum value of these standard deviations for the geometry of interest (for

example, 2 m horizontal transmitting height and 10 m distance for standard calibrations). A Student’s

t-distribution is assumed with n = 5 and a 95% confidence interval, yielding a coverage factor t = 2.78.

There is no efficient means to assess this contributor with a Type A method. A Type B evaluation of this

contributor should be used instead.

contributor should be used instead.

CISPR 16-1-4:2010 contains procedures for evaluation of the EUT support material. A Type A contributor

can be determined by conducting repeated measurements of this method. This analysis is to be performed

for each frequency range measured. The maximum standard deviation obtained from across all analyzed

frequency points shall be used as the contributor value.

25

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

5.6.5 Uncertainty contributors for radiated emission measurements in the frequency range

above 1000 MHz—Type A evaluation

The same uncertainty terms that are used below 1000 MHz can also be used above 1000 MHz. See 5.6.4

for a listing of these uncertainty terms.

Table 9 and Table 10 are based on example spreadsheets that can be accessed at the following URL:

http://standards.ieee.org/downloads/C63/. The example spreadsheets are the combined result of Type A and

Type B analyses of contributors. Each contributor is entered on a line in the spreadsheet. Then, based on

the particular parameter, the value is entered in column C, the type of contributor is entered in column B,

and the coverage factor (divisor) is assigned in column G. The remaining information is automatically

populated, and the standard uncertainty and the expanded uncertainty are calculated.

26

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement Uncertainty

Description of Type Parameter Parameter Distribution Normal Divisor Standard Effective degrees Sensitivity Computed Computed

uncertainty A/B uncertainty units distribution (k) deviation in of freedom (for coefficient uncertainty in uncertainty in

contributor limits level of parameter Type A enter the (normalization analysis units analysis units

confidence units number of factor) squared

measurements

minus 1, for Type

B leave blank)

Receiver reading A 2.60E-01 dB Std Dev 95.00% 1 2.60E-01 4 1 0.26 0.068

Attenuation of

the connection

B 3.80E-01 dB Rectangular 100.00% 1.732 2.19E-01 1 0.22 0.048

between antenna

and receiver

AMN voltage

B 2.50E-01 dB Normal 95.45% 2 1.25E-01 1 0.13 0.016

division factor

Receiver sine-

wave voltage B 2.00E+00 dB Rectangular 100.00% 1.732 1.15E+00 1 1.15 1.333

accuracy

Receiver pulse

amplitude B 1.50E+00 dB Rectangular 100.00% 1.732 8.66E-01 1 0.87 0.750

response

Receiver pulse

response

variation with B 1.50E+00 dB Rectangular 100.00% 1.732 8.66E-01 1 0.87 0.750

repetition

frequency

Noise Floor

B 0.00E+00 dB Normal 95.45% 2 0.00E+00 1 0.00 0.000

proximity

Mismatch

effects between

B 5.80E-01 dB U-Shaped 95.45% 1.414 4.10E-01 1 0.41 0.168

antenna port and

receiver

AMN

B 2.60E+00 dB Triangular 100.00% 2.449 1.06E+00 1 1.06 1.127

impedance

Expanded uncertainty [based on k factor of 2 (95% confidence level)] 4.13

27

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement Uncertainty

Description of Type Parameter Parameter Distribution Normal Divisor Standard Effective degrees of Sensitivity Computed Computed

uncertainty A/B uncertainty units distribution (k) deviation in freedom (for Type A coefficient uncertainty in uncertainty in

contributor limits level of parameter enter the number of (normalization analysis units analysis units

confidence units measurements minus factor) squared

1, for Type B leave

blank)

Receiver reading A 2.11 dB Std Dev 68.27% 1 2.11 4 1 2.11 4.464

Attenuation of the

connection between B 0.10 dB Normal 95.45% 2 0.05 1 0.05 0.003

antenna and receiver

Biconical antenna

B 0.40 dB Normal 95.45% 2 0.20 1 0.20 0.040

factor

Receiver sine-wave

B 2.00 dB Rectangular 100.00% 1.732 1.15 1 1.15 1.333

voltage accuracy

Receiver pulse

B 1.50 dB Rectangular 100.00% 1.732 0.87 1 0.87 0.750

amplitude response

Receiver pulse

response variation

B 1.50 dB Rectangular 100.00% 1.732 0.87 1 0.87 0.750

with repetition

frequency

Noise floor

B 0.50 dB Normal 95.45% 2 0.25 1 0.25 0.063

proximity

Mismatch effects

between antenna port B 0.36 dB U-Shaped 1 Std Dev 1.414 0.25 1 0.25 0.065

and receiver

Antenna factor

frequency B 0.25 dB Rectangular 100.00% 1.732 0.14 1 0.14 0.021

interpolation

Antenna factor

B 0.50 dB Rectangular 100.00% 1.732 0.29 1 0.29 0.083

variation with height

Antenna directivity B 0.50 dB Rectangular 100.00% 1.732 0.29 1 0.29 0.083

Antenna phase center B 0.00 dB 100.00% 0.00 0.000

Antenna cross-

B 0.00 dB 100.00% 0.00 0.000

polarization response

Antenna balance B 0.90 dB Rectangular 100.00% 1.732 0.52 1 0.52 0.270

Test site B 2.84 dB Triangular 100.00% 2.449 1.16 1 1.16 1.345

Separation between

EUT and

B 0.10 dB Rectangular 100.00% 1.732 0.06 1 0.06 0.003

measurement

antenna

Height of table

B 0.10 dB Normal 95.45% 2 0.05 1 0.05 0.003

supporting the EUT

Expanded uncertainty [based on k factor of 2 (95% confidence level)] 6.09

28

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Annex A

(informative)

A.1.1 General

Uncertainty is a parameter associated with the result of a measurement. It characterizes the dispersion of

the values that could be reasonably attributed to the measurand. The uncertainty of a measurement is

always an estimate. It may be arrived at by either computation or estimates based on experience. Because

the computations are statistical in nature, the uncertainty computed remains an estimate: a good estimate,

perhaps, but yet only an estimate.

⎯ Combined standard uncertainty, with symbol uc

⎯ Sensitivity coefficients, with symbol ci

⎯ Coverage factor, with symbol k

⎯ Expanded uncertainty, with symbol U

The combined standard uncertainty is calculated using the law of propagation of uncertainty (see ISO/IEC

Guide 98-3:2008 (GUM) [B16]). It represents the estimated standard deviation of a measurement result,

which is obtained from the values of several contributors. It is equal to the positive square root of a sum of

terms that are the weighted variances and covariances of all contributors. Equation (A.1) is an example (for

the case where all contributors are uncorrelated and independent from one another):

When selecting components or contributors, care must be taken to avoid double-counting of uncertainty

components. For example, if a component is already included in a Type A evaluation, then it should not be

included again separately as a Type B component [see ANSI/NCSL Z540-2:1997 [B4] or the

ISO/IEC Guide 98-3:2008 (GUM) [B16]]. Any bias that is known and used for correcting the measurement

result should not be included in the uncertainty budget.

A useful term, but one not defined officially in the GUM, is “uncertainty driver.” An uncertainty driver is

an influence quantity that is much larger (approximately two or more times) than most other contributors

affecting the value of the measurand. It effectively sets the minimum possible value of the calculated

combined standard uncertainty of the measurement [i.e., it “drives” (or dominates) the measurement

uncertainty more than all other contributors]. In this case, special consideration must be given to the

applicability of the calculation method described in this document and especially to the correct coverage

factor value k used in computing the expanded uncertainty. This is due to the fact that the central limit

29

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

theorem may no longer be applied, and the probability distribution of the “uncertainty driver” will also

“drive” the distribution of the resulted expanded uncertainty. See Annex C of UKAS M 3003:2007 [B22]

for details.

In some cases, the input quantity to the process may not be in the same units as the output quantity. For

example, the uncertainty in the measurement distance on an open site should be converted to the

uncertainty in the received signal level. In this case, the input quantity is length, but the output quantity is

electrical field strength. It is therefore necessary to relate the output quantity (y) to the input quantity (xi).

The conversion from one unit to the other is the mathematical expression relating the two quantities. If the

mathematical expression is linear (i.e., an algebraic expression with no terms raised to a power different

than unity), then the sensitivity coefficient is unity. In the earlier example, if the measurement distance is in

the far field of the source, then the field strength is inversely proportional to the distance, so that the

fractional error in measuring the separation distance results in a fractional error in the level of the field

strength. On the other hand, if the measurement distance is within the near field of the source, then the field

strength is inversely proportional to a higher order of the distance (e.g., the square of the distance or the

cube of the distance, or some fractional exponent greater than unity). In this case, the sensitivity coefficient

is equal to the exponent of distance that is applicable.

E = md n (A.2)

where

m is the proportionality constant

d is the measurement distance

n is the exponent of the change of E with respect to distance (n = −1 in the far field)

The exponent n of the distance is the sensitivity coefficient, usually shown as ci in uncertainty expressions.

Taking 20 times the logarithm of the Equation (A.2) gives:

E ± ΔE m d ± Δd

20 log = 20 log + 20n log

E m

d (A.5)

And simplifying:

ΔE Δd

20 log1 ± = 20n log1 ± (A.6)

E d

30

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The sensitivity coefficient is ci = n. At distances from the EUT that are in the far field, n = −1.

At 30 MHz or above, the 10 m measurement distance is in the far field so that n = −1; thus:

ΔE Δd

20 log1 ± = −20 log1 ± (A.7)

E d

Thus, if the error in setting the antenna at 10 m from the EUT is 5 cm, then the relative error in d is:

0.05

εd = = 0.005 (A.9)

10

− 0.043 dB

ε E = −20 log(1 ± ε d ) = (A.10)

+ 0.044 dB

dB

Bias or systematic effect is common in EMC measurements. It takes the form of correction factors (e.g.,

cable loss), conversion factors (e.g., antenna factor), and bias in specific instrumentation. Each systematic

effect has two elements, a known or approximately known or expected value µ and a random variation

about the expected value uµ. The expected value is used as a direct correction of the measurand and the

random variation is combined into the uncertainty of the measurement in Equation (A.1).

Expanded uncertainty defines an interval about the result of a measurement expected to encompass a large

fraction of possible values of the result. This interval is based on a coverage factor k. This coverage factor

is usually 2 or 3, but it may have other values. A few values of k related to the confidence level are shown

in Table A.1, assuming a normal (Gaussian) probability distribution for the expanded uncertainty (which,

due to the central limit theorem, is usually true, unless an “uncertainty driver” of other than normal

distribution exists).

31

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

for normal distribution

Confidence level, % 50 68.27 80 90 95.45 99.73 99.99

k 0.675 1 1.28 1.645 2 3 3.89

The normal (Gaussian or “bell curve”) distribution is commonly applied to Type A evaluations of additive

data.8

The first step in a Type A evaluation consists of calculating the arithmetic mean of the repeated

observations:

n

q = qj (A.11)

j =1

where:

qj is the value of the jth observation

n is the number of observations

(q )

n

( )

s qj =

1

n −1

j −q (A.12)

j =1

The standard uncertainty (the contribution to the measurement uncertainty budget due to an influence

quantity that is evaluated using a Type A statistical method) is expressed as 1 standard deviation (or error)

of the mean value:

us = s q =( ) s(qn ) j

(A.13)

The divisor used in the uncertainty budget calculation is in this case k = 1, because the contributor’s value

obtained from Equation (A.13) is already in terms of 1 standard deviation.

8

It is incorrect to use percentages with the normal distribution. Percentages are not additive unless converted to their base units, and

the correct mean value of a group of percentages is the harmonic mean, rather than the arithmetic mean of the normal distribution. The

errors or variations used with the normal distribution must be additive (e.g., decibels or raw linear variations such as ±50 µV/m)

(around a specific value such as 600 µV/m).

32

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

NOTE 1—The previous formulas can only be applied when the number of observations is sufficiently large such that

they may be considered to be a statistically representative sample set of the entire population of possible observations

and when the observations are independent from each other (no correlation exists between different observations).

NOTE 2—The quantity ν = n – 1 from the denominator of Equation (A.12) is called number of degrees of freedom, and

it represents the number of independent observations minus the number of parameters derived from these observations.

When the observations are not obtained from repeated measurements (for example, when old calibration data from a

number of previous years is used for the Type A analysis), then ν = n shall be used in Equation (A.12). See

IEC 61000-1-6:2012 [B13].

Student’s probability distribution, or the t-distribution, arises in the problem of estimating the mean of a

normally distributed population when the sample size is small. In practice, the standard deviation of the

population is always unknown and must be estimated from the data.

Suppose there is a random sample set of size n drawn from a normal-distributed population with mean μ

and standard deviation σ. Let x denote the sample mean and s the sample standard deviation. Then, the

quantity:

x−μ

t= (A.14)

s/ n

has a t distribution with n – 1 degrees of freedom. The variance (the standard deviation squared) is then:

( )

1 n

s2 = xi − x (A.15)

n − 1 i =1

Note that there is a different t distribution for each sample set size; in other words, it is a class of

distributions. When specifying a t distribution, the degrees of freedom must also be specified. The degrees

of freedom for this t statistics come from the sample standard deviation s in the denominator of

Equation (A.14).

The t density curves are symmetric and bell shaped like the normal distribution and have their peak at zero.

However, the spread is more than that of the standard normal distribution. This is due to the fact that in

Equation (A.14), the denominator is s rather than σ. Because s is a random quantity varying with various

samples, the variability in t is larger, resulting in a larger spread.

The larger the degrees of freedom, the closer the t-distribution is to the normal distribution. This reflects the

fact that the standard deviation s approaches σ for large sample set size n.

The contributor’s value of a t-distributed influence quantity (normal distributed quantity for which a

sufficiently large sample size cannot be obtained) can be calculated by using Equation (A.11) and

Equation (A.12), and then by adjusting the obtained experimental standard deviation by an appropriate

coefficient, h(n) , which accounts for the limited sample size:

h(ν) s (q j )

us = h(ν) s (q ) = (A.16)

n

The coefficient h(ν) depends on the number of degrees of freedom, ν = n – 1 (see IEC 61000-1-6:2012

[B13]):

33

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

ìï

ïï

ïï6.48 ν =1

ï

h(ν) = ïí2.20 ν=2 (A.17)

ïï

ïï ν

ïï ν³3

ïî ν - 2

The values of η (ν ) are calculated as the ratio between the upper critical values of the Student’s

t-distribution for ν degrees of freedom and an infinite degrees of freedom, respectively, for a 95 %

confidence level. As the number of observations increases, the number of degrees of freedom approaches

infinity and η (ν ) approaches unity. A selection of values for η (ν ) is given in Table A.2, for various

degrees of freedom.

The divisor used in the uncertainty budget calculation is in this case k = 1 because the contributor’s value

obtained from Equation (A.16) is already in terms of 1 standard deviation.

ν 1 2 3 4 5 6 7 8 9

η(ν) 6.48 2.20 1.73 1.41 1.29 1.22 1.18 1.15 1.13

ν 10 11 12 13 14 19 29 49 99

η(ν) 1.12 1.11 1.10 1.09 1.08 1.06 1.04 1.02 1.01

A.2.2.1 General

The distributions described in A.2.2.2 through A.2.2.7 are a method to put Type A and Type B evaluations

of uncertainty on the same “footing.” Their equivalent descriptive statistics may be estimated from

calculations using statistically sound approximations. This allows combining, with appropriate weighting,

uncertainties from various sources via the law of propagation of uncertainty. The resulting combined

standard uncertainty is considered normally distributed; thus, the expanded uncertainty can maintain its

prediction of confidence level.

The most common example of a normal-distributed influence quantity in a Type B analysis consists of

using data from a calibration certificate. For instance, the antenna factor, obtained from the calibration

certificate, is used in computing the radiated emission field strength level by adding its value (in dB/m) to

the cable loss (in dB) and to the signal amplitude measured by the receiver (in dBµV). The uncertainty of

calibration, also obtained from the calibration certificate, is then a contributor to the radiated emissions

measurement uncertainty budget.

The value of a normal-distributed uncertainty (such as that obtained from a calibration certificate) may be

reported either in terms of 1 standard deviation or as an expanded uncertainty. In the latter case, the

reported value must be accompanied by a statement of the confidence level and the value of the coverage

factor k used in computing the expanded uncertainty. The contributor’s value is then the reported

uncertainty value, and the divisor is k (for values reported as 1 standard deviation, the divisor is unity). See

Table A.1 for a set of more frequently used values of k.

34

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

As stated in A.2.1.2, when estimating the standard deviation of a normal-distributed quantity from a

reduced set of observations, a t-distribution must be used. Type B evaluations of normal-distributed

quantities are most likely based on data from calibration certificates, the vast majority of which report an

expanded uncertainty calculated based on a normal distribution. However, if the expanded uncertainty is

based on a t-distribution rather than a normal distribution, it has to be accompanied by additional

information, like the number of degrees of freedom (ν), confidence level (p), and the coverage factor (kp)

used in the calculation of the reported expanded uncertainty value. If any of these three parameters is not

reported, then it can be derived from the other two by using the t-distribution of Table A.3 (reproduced

from UKAS M 3003:2007 [B22] and ISO/IEC Guide 98-3:2008 [B16]).

If the reported number of degrees of freedom is not listed in Table A.3, then it will be necessary to

interpolate between the kp values given in the table. For ν > 3, linear interpolation is sufficient; for lower

values of ν, higher order interpolation should be used. Alternatively, the next lower value of ν listed in

Table A.3 may be used, which would give a worst-case estimation of the uncertainty value.

The contributor’s value to be used in the overall uncertainty budget is in this case the reported expanded

uncertainty, and the divisor is kp, as determined from Table A.3.

Degrees of kp as a function of the number of effective degrees of freedom and confidence level

freedom, ν p = 68.27% p = 90% p = 95% p = 95.45% p = 99% p = 99.73%

1 1.84 6.31 12.71 13.97 63.66 235.80

2 1.32 2.92 4.30 4.53 9.92 19.21

3 1.20 2.35 3.18 3.31 5.84 9.22

4 1.14 2.13 2.78 2.87 4.60 6.62

5 1.11 2.02 2.57 2.65 4.03 5.51

6 1.09 1.94 2.45 2.52 3.71 4.90

7 1.08 1.89 2.36 2.43 3.50 4.53

8 1.07 1.86 2.31 2.37 3.36 4.28

9 1.06 1.83 2.26 2.32 3.25 4.09

10 1.05 1.81 2.23 2.28 3.17 3.96

11 1.05 1.80 2.20 2.25 3.11 3.85

12 1.04 1.78 2.18 2.23 3.05 3.76

13 1.04 1.77 2.16 2.21 3.01 3.69

14 1.04 1.76 2.14 2.20 2.98 3.64

15 1.03 1.75 2.13 2.18 2.95 3.59

16 1.03 1.75 2.12 2.17 2.92 3.54

17 1.03 1.74 2.11 2.16 2.90 3.51

18 1.03 1.73 2.10 2.15 2.88 3.48

19 1.03 1.73 2.09 2.14 2.86 3.45

20 1.03 1.72 2.09 2.13 2.85 3.42

25 1.02 1.71 2.06 2.11 2.79 3.33

30 1.01 1.70 2.04 2.09 2.75 3.27

35 1.01 1.70 2.03 2.07 2.72 3.23

40 1.01 1.68 2.02 2.06 2.70 3.20

45 1.01 1.68 2.01 2.06 2.69 3.18

50 1.01 1.68 2.01 2.05 2.68 3.16

100 1.005 1.660 1.984 2.025 2.626 3.077

∞ 1.000 1.645 1.960 2.000 2.576 3.000

The rectangular distribution is used where all values in a continuum are equally likely (e.g., an instrument

manufacturer’s specified tolerance).

35

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The true value Xi may, with equal probability, lie anywhere in the interval a+ to a–. From a mathematical

analysis, u2(xi) = (a+ – a–)2 / 12. For symmetrical data, a+ – a– = 2a, and u2(xi) = a2 / 3, or u(xi) = a / √3

(i.e., weighting factor is 1 / √3). If the data are unsymmetrical, then understand why and/or perform a

further analysis as follows. For example, if the manufacturer specifies a tolerance of +2/−3 dB, then find

that the midpoint is −0.5 dB and that a+ – a- = 5 dB. Then the uncertainty is 2.5 / √3 dB with a bias of

−0.5 dB. This bias is additive with any other biases and is transferred to the final result. If the manufacturer

arrived at this unsymmetrical tolerance due to conversion of percentages to dB, then it is usually best

simply to use the largest absolute value as a in the equation for u(xi) and ignore the bias.

This distribution is applied to tolerances or errors stated as a range of possible values without any

indication of the statistics. For example, when an instruction manual gives the amplitude tolerance of an

instrument as ±2 dB, then the rectangular distribution must be applied because the actual measurand could

lie anywhere within the continuum from +2 dB to –2 dB with equal probability. The standard uncertainty

for a measurand with a tolerance of ±2 dB would be us = 2 / √3 ≈ 1.15 dB. This number would be entered

into Equation (A.1), the root sum of squares equation for the propagation of uncertainty, along with other

standard uncertainties of the measurand.

The following analysis shows how the statistics for the rectangular distribution arise. Similar, although

more complicated, analyses were performed for the remaining non-normal distributions.

The midpoint is Xi and the limits are Xi ± a. The area of the distribution must be unity (100% probability);

therefore, the height of the rectangle must be 1 / (2a), and the probability density function is P = 1 / (2a).

The variance is thus:

a

u 2 (xi ) = x 2 Pdx (A.18)

−a

Substituting P, then:

a

x2

a x3 a3 − a3 a2

u 2 (xi ) = dx = = − = (A.19)

− a 2a 6a −a 6a 6a 3

Therefore:

a

u (xi ) = (A.20)

3

A further example of the rectangular distribution is the result of repeated rolling or throwing of one fair die

as shown in Figure A.1.

0.25

Probability

0.2

0.15

0.1

0.05

0

0 1 2 3 4 5 6 7

Number

36

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

The U-shaped distribution is used where values tend to be clustered near the ends of a continuum (e.g., the

uncertainty associated with VSWR). The true value Xi is concentrated near the ends of the interval a+ to a–.

From a mathematical analysis, u2(xi) = (a+ – a–)2 / 8. For symmetrical data, a+ – a– = 2a, and u2(xi) = a2 / 2;

thus, u(xi) = a / √2 (i.e., weighting factor is 1 / √2).

Examples include 5% carbon-slug resistor values and the effects of VSWR on measurement uncertainty.

VSWR effects are symmetrical in linear terms but not symmetrical in decibels. In linear terms, mismatch

uncertainty is given by uM = ±2|ГG||ГL| / √2. Mismatch uncertainty, in decibels, is given by uM = [20

log(1+|ГG||ГL|) − 20 log(1−|ГG||ГL|)]/(2√2), dB. Note that this is a special case of Equation (2), where the

interconnecting network between the source and the load is assumed to be well matched (i.e., |S11|<<1 and

|S22|<<1) and of negligible attenuation (|S21|≈1).

If the VSWR for the connection between two devices in the measurement chain were 2.5:1 for the source

device and 1.8:1 for the load device, then the magnitudes of the reflection coefficients |ГG| (of the source)

and |ГL| (of the load) would be 0.428 and 0.286, respectively. The mismatch uncertainty would be

uM = [20 log(1 + 0.428 × 0.286) − 20 log(1 − 0.428 × 0.286)] / 2√2 = [1.003 – (−1.135)] / 2√2 = 0.756 dB.

One can carry both values into the propagation of uncertainty, but it is customary to use the largest of the

absolute values; in this case, 1.135 dB, which will result in uM = 1.135 / √2 = 0.802 dB.

The U-shaped distribution is shown in Figure A.2. The vertical dashed lines indicate the 1/√2 weighting

(±1 standard deviation around the expected value).

0.5

0.4

Probability

0.3

0.2

0.1

0

-1 -0.5 0 0.5 1

Number

The a priori or triangular distribution is used where values tend to be clustered near the center of a

continuum, as in the uncertainty associated with site attenuation measurements. The true value Xi is known

or reasonably believed to be concentrated near and around the center of the interval a+ to a–. From

mathematical analysis, u2(xi) = (a+ – a–)2/24. For symmetrical data, u(xi) = a/√6 (weighting factor is 1/√6).

Some examples are site attenuation data on a well-behaved open-area test site, and other processes that

produce tightly clustered, centrally located data.

The criterion in ANSI C63.6-1996 [B2] and ANSI C63.4 for the acceptance of an OATS is ±4 dB. Thus, in

finding the standard uncertainty for the OATS as an influence quantity of a radiated emission measurement,

us = 4 / √6 = 1.63 dB. This value is entered into the equation for the propagation of uncertainty as the

influence quantity corresponding to the OATS site attenuation compliance.

37

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Probability 1/a

of Xi

a– µx a+

µx – a / 2.45 µx + a / 2.45

Figure A.1 showed the distribution created by rolling or throwing one fair die. Figure A.4 shows the

distribution resulting from rolling two fair dice, and Figure A.5 shows the distribution of the results of

rolling three fair dice.

This process may be repeated for more and more dice, and as can be seen from the examples in Figure A.1,

Figure A.4, and Figure A.5, the more rectangular distributions one combines, the more nearly a normal

distribution results. This may help the reader to understand how the combined standard uncertainty

becomes a normal distribution even though it is made up of non-normal distributions (the central limit

theorem).

0.18

0.16

0.14

Probability

0.12

0.1

0.08

0.06

0.04

0.02

0

1 2 3 4 5 6 7 8 9 10 11 12 13

Number

38

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

25

20

15

Density

10

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Num ber

The normal (Gaussian) distribution was designed and is intended for the analysis of additive data. The log-

normal distribution was designed and is intended for the analysis of multiplicative data. The transformation

from the log-normal distribution to the normal distribution is f(Y) = ln g(X), where f(Y) is the normal

distribution and g(X) is the log-normal distribution. The transform of the log-normal distribution into the

normal distribution is shown in Figure A.6.

0.7

Y = Ln X (Norm.)

f(X) (Lognorm.) 0.6

f(X) Mode, Median, Mean

0.5

f f(X) Mean

0.3

f(X) Mode

0.2

0.1

0

–5 –4 –3 –2 –1 0 1 2 3 4 5

X

Y

locations of the principal log-normal statistics transformed to the normal distribution

39

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

EMC test data are multiplicative in nature. When one makes a measurement, for example, radiated field

strength, the instrument reading (units in µV) is multiplied by the antenna factor (units in 1/m), and then

multiplied by the cable loss (units in V/V). Usually, the measurements are made in terms of decibels, so the

previous example becomes one of addition of decibel values. To the instrument reading (units in dB[µV])

are added the antenna factor (units in dB[1/m]) and cable loss (units in dB). Because the transformation

from the log-normal distribution to the normal distribution is f(Y) = ln g(X), the direct transformation of

levels in linear units is via the same transformation. For example, the instrument reading X in µV is Y = ln

X nepers, and the conversion of nepers to decibels referenced to voltage is 8.686; that is, Y = 8.686 ln X =

20 log X, in dB(µV).

This transformation cannot be correctly used with percentages because they are neither multiplicative nor

additive!

40

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

Annex B

(informative)

Bibliography

Bibliographical references are resources that provide additional or helpful material but do not need to be

understood or used to implement this standard. Reference to these resources is made for informational use

only.

Emission Measurements in Electromagnetic Interference (EMI) Control—Calibration of Antennas (9 kHz

to 40 GHz).

[B2] ANSI C63.6-1996, American National Standard Guide for the Computation of Errors in Open-Area

Test Site Measurements.9

[B3] ANSI/NCSL Z540-1:1994, American National Standard for Calibration—Calibration Laboratories

and Measuring and Test Equipment—General Requirements.

[B4] ANSI/NCSL Z540-2:1997, American National Standard for Expressing Uncertainty — U.S. Guide

to the Expression of Uncertainty in Measurement.

[B5] Carpenter, D., “A demystification of the U-shaped probability distribution,” Proceedings of the

IEEE International Symposium on Electromagnetic Compatibility, Boston, MA, 2003.

[B6] Carpenter, D., “A further demystification of the U-shaped probability distribution,” Proceedings of

the IEEE International Symposium on Electromagnetic Compatibility, Chicago, IL, 2005.

[B7] CISPR/A/838/INF, January 2009, containing CISPR/A/WG1(Dunker-Riedelsheimer- Trautnitz)06-

01, Measurement of FAR similar to CISPR 16-1-4 and site VSWR in the Kolberg FAR of the Federal

Network Agency for Electricity, Gas, Telecommunications, Post and Railway, September 2006

(background material on an estimation of the uncertainty due to results of SVSWR measurements, to be

published before the FDIS).

[B8] CISPR/TR 16-4-1:2009, Specification for Radio Disturbance and Immunity Measuring Apparatus

and Methods—Part 4-1: Uncertainties, Statistics and Limit Modelling—Uncertainties in Standardized EMC

Tests.10

[B9] CISPR/TR 16-4-3:2007, Specification for Radio Disturbance and Immunity Measuring Apparatus

and Methods—Part 4-3: Uncertainties, Statistics and Limit Modelling—Statistical Considerations in the

Determination of EMC Compliance of Mass-Produced Products.

[B10] CISPR 11-2010, Industrial, Scientific, and Medical (ISM) Radio-Frequency Equipment—

Electromagnetic Disturbance Characteristics—Limits and Methods of Measurement.

[B11] CISPR 16-2-3:2006-2007, Specification for Radio Disturbance and Immunity Measuring Apparatus

and Methods—Part 2-3: Methods of Measurement of Disturbances and Immunity—Radiated Disturbance

Measurements.

[B12] CISPR 22-2010, Information Technology Equipment—Radio Disturbance Characteristics—Limits

and Methods of Measurement.

9

ANSI publications are available from the American National Standards Institute (http://www.ansi.org/).

10

CISPR documents are available from the International Electrotechnical Commission (http://www.iec.ch/). They are also available in

the United States from the American National Standards Institute (http://www.ansi.org/).

41

Copyright © 2013 IEEE. All rights reserved.

ANSI C63.23-2012

American National Standard Guide for Electromagnetic Compatibility—Computations and Treatment of Measurement

Uncertainty

[B13] IEC 61000-1-6:2012, Electromagnetic Compatibility (EMC)— Part 1-6: General—Guide to the

Assessment of Measurement Uncertainty, Ed. 1.11

[B14] ISO 10012-2003, Measurement Management Systems—Requirements for Measurement Processes

and Measuring Equipment.12

[B15] ISO/IEC 17000-2004, Conformity Assessment—Vocabulary and General Principles.13

[B16] ISO/IEC Guide 98-3:2008, Uncertainty of Measurement—Part 3: Guide to the Expression of

Uncertainty in Measurement (GUM:1995).

[B17] ISO/IEC Guide 99-2007, International Vocabulary of Metrology—Basic and General Concepts and

Associated Terms (VIM).

[B18] Medler, J., “Uncertainty contribution of the EMI test receiver in RF disturbance measurements,”

Proceedings IEEE Asia-Pacific International Symposium on EMC, Beijing, China, 2010.

[B19] Moulthrop, A. A. and Michael S. Muha,, “Accurate measurement of signals close to the noise floor

on a spectrum analyzer,” IEEE Transactions on Microwave Theory and Techniques, pp. 1182–1885,

Nov. 1991.

[B20] Stecher, M., “A detailed analysis of EMI test receiver measurement uncertainty,” Proceedings IEEE

International Symposium on EMC, Montreal, Canada, 2001.

[B21] UKAS LAB 34:2002, The Expression of Uncertainty in EMC Testing.14

[B22] UKAS M 3003:2007, The Expression of Uncertainty and Confidence in Measurement.

[B23] Warner, F. L., “New expression for mismatch uncertainty when measuring microwave attenuation,”

IEEE Proceedings, Part H—Microwaves, Optics and Antennas, vol. 127, Part H, no. 2, Apr. 1980.

11

IEC publications are available from the International Electrotechnical Commission (http://www.iec.ch/). IEC publications are also

available in the United States from the American National Standards Institute (http://www.ansi.org/).

12

ISO publications are available from the ISO Central Secretariat (http://www.iso.org/). ISO publications are also available in the

United States from the American National Standards Institute (http://www.ansi.org/).

13

ISO/IEC publications are available from the ISO Central Secretariat (http://www.iso.org/). ISO publications are also available in the

United States from the American National Standards Institute (http://www.ansi.org/).

14

UKAS publications are available from the United Kingdom Accreditation Service (http://www.ukas.com).

42

Copyright © 2013 IEEE. All rights reserved.

Authorized licensed use limited to: UNIVERSIDAD AUTONOMA DE NUEVO LEON. Downloaded on June 06,2018 at 18:22:56 UTC from IEEE Xplore. Restrictions apply.

- [Scheller G.] Error Analysis of da Temperature Meas(BookZZ.org)Transféré parTracey Morin
- ASME B89.7.3.1-2001 Guidelines for Decision Rules - Considering Measurement Uncertainty in Determining Conformance to SpecificationsTransféré parroberz27
- DKD-R 8-1 Guideline DKD-R 8-1 Calibration of Piston-operated Pipettes With Air CushionTransféré parFërnando Rodrîguëz Herrerä
- Novotny 2Transféré pardumitriu_irinaluciana6333
- 83158608-ISO-14253-1-1998-DAmd-1-E-Character-PDF-DocumentTransféré parBala Muru
- pdfTransféré parPankaj Sharma
- mgpg36Transféré parengrzia
- Draft for T-1Transféré parGabby Galvez
- Uncertainty In Measurement (1).pptxTransféré parDijo Paul
- JCGM_106_2012_ETransféré parRaymond Valdes
- Interpretation With Expanded Uncertainty 2007 v1Transféré parMaryuris Melo Pedraza
- H-9908-0482-01 Angular Measurement Error BudgetTransféré parkimikonon
- incertidumbre en química clínicaTransféré parEmmanuel Acevedp P
- Major Practical Report Guide 2016Transféré parjohn
- EA-10-08.pdfTransféré parrobert
- FAS625en_Error analysis of a temperature mesurement system.pdfTransféré parnadjib62
- Messurement AccuracyTransféré pargalati12345
- Stirling NatureTransféré parJack Stilgoe
- Decision Coaching ResourceTransféré parCarloss Valenzueela Ramiirez
- Lecture 4 - Risk & UncertaintyTransféré parIbrahim Munir
- How Do I Build My BusinessTransféré parAnonymous vCCKUda
- 00019-02R1Transféré parAshok Lenka
- 4th Lab Report Fixed FormulasTransféré parDaniel Garcia
- 213Transféré parMarcelo Protz
- Bim ReportTransféré parHien
- Class X EnglishTransféré parteachers rsr
- TLE july 15-19Transféré parMai Maehh Aznada
- IB Lab Report GuideTransféré parmgch99
- Evidence StatisticTransféré parediartmobe
- docTransféré parapi-384049835

- Definition_classification of Power System Stability IEEE-CIGRE Joint Task Force on Stability Terms and DefinitionsTransféré parlepton1892
- Kundur PaperTransféré parNando López
- Analysis Lightning Over Voltage 500kVTransféré parSteve Bootman
- Cigre PMU and FACTS.pdfTransféré parNando López
- Optimized DesigneTransféré parNando López
- Analysis of ResonanceTransféré parNando López
- Impedance Matching for HVDCTransféré parNando López
- IEEE9 Model Documentation R0Transféré parNando López

- ASTM A99 1992Transféré parNguyễn Duy Biên
- American FlourishTransféré parÉdipo
- 4-No-Fuss-Duo.docxTransféré parAhamed Imran Sheriff
- Synth Master Licence AgreementTransféré parKevinPereyra
- Lets Go Fly a Kite (Full Arrangement)Transféré parEmily Louise Teague
- DAP.pptTransféré parHans Mosquera
- Contracts for Engineers NotesTransféré parDouglas Wentzel
- PMDG-737NGX-Tutorial-1.pdfTransféré parReginaldo Lima
- Webpshere Upgration GuideTransféré parbonkuri
- Low Noise Analog MEMS Microphone and PreampTransféré partarpino
- As NZS 60745.1-2003 Hand-Held Motor-operated Electric Tools - Safety General RequirementsTransféré parSAI Global - APAC
- document.pdfTransféré parAnthony Kasena
- Partner Agency GuidelinesTransféré parEmily Fairbank
- 78EZNMTransféré parzqandah
- You-Are-The-Best-Thing-String-Quartet - fragment.pdfTransféré parlynnette
- fbxTransféré parNghiep Tran
- Hydraulic Design of Labyrinth WeirsTransféré parCarlos Mendez Perez
- Mage - Kevin Murphy - Penny DreadfulTransféré parOh Krill
- Patent in India Guideline PDFTransféré parPrasad Kulkarni
- Plan SketchTransféré parANGEL .
- AbridgedP6CodexV0p2Transféré parGrybel
- Lecture 7 - Leveraging Secondary Brand Associations to Build Brand EquityTransféré parPragash MuthuRajan
- 05.01 HIGHLAND Oil Water Separator EnTransféré parlonelystar19
- Gatioan v GaffudTransféré parNalyn Kis-ing
- xHarbour Language Reference GuideTransféré parLuiz Uberti
- Gil Hyatt USPTO Motion to StayTransféré parAnonymous VXeDekP
- Case-Study-3 (1)Transféré parFingiel Blu
- 240708017-GURPS-4th-Ed-Dungeon-Fantasy-Monsters-1.pdfTransféré parAlejandro Corona
- In 910 DS WebServicesGuide EnTransféré parEvrim AY
- Provrem 1Transféré parEd Nerosa

## Bien plus que des documents.

Découvrez tout ce que Scribd a à offrir, dont les livres et les livres audio des principaux éditeurs.

Annulez à tout moment.