CISPR TR 16-4-1:2003/AMD1:2004

TECHNICAL
CISPR
REPORT 16-4-1
AMENDMENT 1
2004-12
INTERNATIONAL SPECIAL COMMITTEE ON RADIO INTERFERENCE
Amendment 1
Specification for radio disturbance and immunity
measuring apparatus and methods –
Part 4-1:
Uncertainties, statistics and limit modelling –
Uncertainties in standardized EMC tests

 IEC 2004 Copyright - all rights reserved
International Electrotechnical Commission, 3, rue de Varembé, PO Box 131, CH-1211 Geneva 20, Switzerland
Telephone: +41 22 919 02 11 Telefax: +41 22 919 03 00 E-mail: inmail@iec.ch Web: www.iec.ch
PRICE CODE
S
Commission Electrotechnique Internationale
International Electrotechnical Commission
МеждународнаяЭлектротехническаяКомиссия
For price, see current catalogue

– 2 – CISPR 16-4-1 Amend. 1  IEC:2004(E)

FOREWORD
This amendment has been prepared by CISPR subcommittee A: Radio interference

measurements and statistical methods.

The text of this amendment is based on the following documents:

DTR Report on voting
CISPR/A/496/DTR CISPR/A/516/RVC

Full information on the voting for the approval of this amendment can be found in the report
on voting indicated in the above table.
The committee has decided that the contents of this amendment and the base publication will
remain unchanged until the maintenance result date indicated on the IEC web site under
"http://webstore.iec.ch" in the data related to the specific publication. At this date, the
publication will be
• reconfirmed,
• withdrawn,
• replaced by a revised edition, or
• amended.
_____________
Page 38
6 Voltage measurements
Renumber Figures 6-1 to 6-8 as Figures 10 to 17.
Renumber the existing references to Figures 6-1 to 6-8 in Clause 6 accordingly.
Page 57
7 Absorbing clamp measurements

Replace the existing text by the following subclauses:
7.1 General
7.1.1 Objective
The primary goal of this clause is to provide information and guidance for the determination of
uncertainties associated with the absorbing clamp measurement and calibration methods.
This clause gives rationale for the various uncertainty aspects described in several parts of
CISPR 16 related to the absorbing clamp, i.e.:
• the absorbing clamp calibration method (see Clause 4 of CISPR 16-1-3);.
• the absorbing clamp measurement method (see Clause 7 of CISPR 16-2-2).

CISPR 16-4-1 Amend. 1  IEC:2004(E) – 3 –

The rationale given in this clause is background information for the above-mentioned parts of
CISPR 16 related to the absorbing clamp and it may be useful in the future when modifying

these parts. In addition, this clause provides useful information for those who apply the

absorbing clamp measurement and calibration method and who have to establish their own

uncertainty estimates.
7.1.2 Introduction
This clause provides information on the uncertainties associated with the absorbing clamp

test method (ACTM) described in CISPR 16-2-2, and with the absorbing clamp calibration

methods described in CISPR 16-1-3. The uncertainty budgets on the ACTM as described in

CISPR 16-4-2 or in LAB 34 [15] are not suitable for actual compliance tests in accordance
with the CISPR specification given in CISPR 16-2-2. The reason is that this uncertainty
budget is limited to the measurement instrumentation uncertainties (MIUs). Uncertainties due
to the set up of the equipment under test (EUT) including the lead under test (LUT), and due
to the measurement procedure are not taken into account. In this clause however, for the
uncertainty considerations of the absorbing clamp measurement method, all the uncertainty
sources that are relevant for the compliance test in accordance with the standard (the
standards compliance uncertainty (SCU)) are considered. For these uncertainty calculations it
is assumed that the EUT is the same. In other words, we consider the uncertainty of an ACTM
using the same EUT that is measured by different test laboratories, using different
measurement instrumentation, a different test site, different measurement procedures and
different operators. Consequently, the reproducibility of this ‘same’ EUT may become a
significant uncertainty source. Also the length of the LUT and the type of the cable can be
slightly different if a test laboratory has to extend the lead by a cable of the ‘same’ type.
The uncertainty assessment described in this clause is performed in accordance with the
basic considerations on uncertainties in emission measurements given in Clause 4.
Subclause 7.2 gives the uncertainty considerations related to the calibration of the absorbing
clamp, while 7.3 gives the uncertainty considerations related to the absorbing clamp
measurement method.
7.2 Uncertainties related to the calibration of the absorbing clamp
CISPR 16-1-3 specifies three different calibration methods for the absorbing clamp, i.e., the
original method, the jig method and the reference device method.
This section describes the determination of the uncertainty budgets for the original clamp
calibration method. The budgets for the jig and reference calibration methods will be included
at a later stage.
For convenience a schematic overview of the original clamp calibration method is given in
Figure 18.
7.2.1 The measurand
For a clamp calibration using the original (org) method, the measurand is the clamp factor
CF in dB(pW/μV).
org
The original clamp calibration method is in fact an insertion loss measurement (see Clause 4
of CISPR 16-1-3,):
CF = A − 17 in dB(pW/μV)
org org
where
(20)
A = the measured insertion loss in dB
org
– 4 – CISPR 16-4-1 Amend. 1  IEC:2004(E)

7.2.2 Uncertainty sources
This subclause gives the uncertainty sources associated with the clamp factor measurement.

The uncertainty of the clamp factor is equal to the uncertainty of the measured insertion loss

(see Equation 20).
The uncertainty sources for the insertion loss are given by the uncertainty sources of the

measurement chain. The measurement chain-related uncertainty sources are the EUT

(=clamp under test in this case), the measurement instrumentation, the set-up, the

measurement procedure and the environmental conditions. Figure 19 gives a schematic

overview of all relevant uncertainty sources using a fish-bone diagram. The fish-bone diagram
indicates the categories of uncertainty sources that contribute to the overall uncertainty of the
clamp factor.
7.2.3 Influence quantities
For most of the qualitative uncertainty sources given in Figure 19, one or more influence
quantities can be used ‘to translate’ the uncertainty source in question. Table 7 gives the
relation between the uncertainty source and the influence quantity. If no influence quantity
can be given, then in the uncertainty budget, the original uncertainty source will be used.
For each of the uncertainty sources/influence quantities some explanation is now given.
7.2.3.1 EUT-related
• Stability clamp
The absorbing clamp is a mechanically rigid device that typically is quite stable over time.
Nonetheless, aging effects may lead to poor contact between the ferrite cores which degrades
the functions of the current probe and the decoupling. This may result in a ‘degradation’ of the
clamp factor and may also cause a degradation of the decoupling factor. This is especially
important if the test laboratory for quality assurance reasons repeats the clamp calibration. If
the manufacturer calibrates new clamps, aging is not an issue. If the manufacturer performs a
type test, then the manufacturer may repeat the calibration using different samples of the
same type of clamp. Depending on the number of samples used, this Type-A uncertainty must
be entered in the uncertainty budget. If the manufacturer performs a unit-specific calibration,
then the calibration result is valid for that specific unit only, and consequently no uncertainty
due to type testing shall be incorporated.
7.2.3.2 Set-up related
a) Cross section lead under test
For calibration of the clamp, a 4 mm diameter wire shall be used. The tolerance of the wire
diameter is not specified. The resulting uncertainty is however considered negligible.
b) Length of lead under test
The length of the lead under test shall be 7 m, of which 6 m runs over the clamp slide and
1 m is routed downwards to the CDN on the reference plane. Due to the application of the
secondary absorbing device, the uncertainty due to variation in length and routing of the
lead under test is considered to be low.
c) Height of lead under test above reference plane
The LUT is running at a height of 0,8 m above the reference on top of the clamp slide with
a tolerance of 5 cm. At the end of the clamp slide the LUT is routed to the CDN. The
uncertainty due to residual routing variations is considered to be minor.

CISPR 16-4-1 Amend. 1  IEC:2004(E) – 5 –

d) Displacement tolerance of lead under test in clamp

For the calibration procedure, a centering guide shall be used to control the position of the

LUT within ±1 mm of the centre position at the location of the clamp reference point

(CRP). The uncertainty figures reported in [16] are used.

e) Start and stop position tolerance

The start position of the CRP is 100 mm from the vertical reference plane (= equal to the

SRP). The stop position of the CRP is 5,1 m from the vertical reference plane (SRP). The

tolerance of the start position determines the uncertainty. A tolerance of ±5 mm is

assumed. The resulting uncertainty is considered to be minor.

f) Guidance and routing of the measurement cable

The guidance and routing of the measurement cable to the receiver is specified. Still some
degree of freedom remains which contributes to uncertainty.
7.2.3.3 Measurement procedure related
Clamp scanning step size
The scanning speed and the frequency step size is specified. Still a residual uncertainty is
expected due to the limited scanning step size.
7.2.3.4 Environment related
a) Temperature and humidity tolerances
These environmental influence quantities are considered to have a negligible impact on
the result of the measurement if the calibration is performed using an indoor test site. For
outdoor test sites, the influence of temperature and humidity on the uncertainty shall be
incorporated.
b) Signal to ambient ratio
For calibration, the measured signal levels shall be 40 dB above ambient levels. In this
situation, the resulting uncertainty may be neglected. For lower signal to noise ratios, an
additional uncertainty shall be taken into account.
c) Distance between operator and set-up
It is assumed that the scanning of the clamp is automated by some means (e.g., by a rope
and pulley arrangement), and that the operator is not in the vicinity of the set-up.
However, if an operator is needed to scan the clamp by hand, then the consequent
uncertainty may be significant, especially below 100 MHz [16]. Such an operator-induced
uncertainty can be investigated experimentally by measuring the clamp output signal at
certain fixed position of the clamp, while the operator is approaching and touching the
clamp from different sides (e.g., from the left and right side of the clamp slide). This can

be repeated for a number of positions of the clamp. The maximum variation due to
presence of the operator and touching the clamp can be determined for instance by using
the maximum-hold and minimum-hold functions of a spectrum analyzer. This maximum
variation can be used as a type-B input for the uncertainty budget.
7.2.3.5 Measurement instrumentation related
a) Generator stability
The stability of the generator of the spectrum or network analyzer system is of importance
for the uncertainty of the measured site attenuation.
b) Receiver/analyzer linearity
This uncertainty is obtained from information on the calibration of the measuring system.
The uncertainty depends on the sweep mode or stepped mode of the analyzer.

– 6 – CISPR 16-4-1 Amend. 1  IEC:2004(E)

c) Mismatch at the input
The attenuator in the input cable shall be at least 10 dB. Resulting mismatch uncertainties

are taken from [16].
d) Mismatch at the output
The attenuator in the measuring cable shall be at least 6 dB. Resulting mismatch

uncertainties are taken from [16].

e) Attenuator (optional)
If a separate generator is used for the clamp factor measurement, then during the direct

measurement of the generator output, an additional attenuator may be used to avoid

overload and consequent non-linear effects in the receiver. In this case, the absolute

value of the attenuator and its uncertainty shall be taken into account in Equation 20 and
in the uncertainty budget respectively.
f) Measuring system reading
Receiver reading uncertainties depend on receiver noise, meter scale interpolation errors.
The latter should be a relatively insignificant contribution to the uncertainty for measuring
systems with electronic displays (least significant digit fluctuation). For classical analogue
meter displays this uncertainty contribution needs to be considered.
g) Signal to noise ratio
For clamp calibrations, the noise floor is usually sufficiently below the measured signal
levels for calibration. The impact of the noise depends on the type of measuring system
used (network analyzer versus spectrum analyzer).
h) Absorbing clamp test site deviation
The clamp calibration result is sensitive to the surrounding environment. The test site
performance depends on the floor material and nearby obstacles.
The test site that is used for the calibration shall be validated in accordance with the
specified validation procedure. Consequently, the pass/fail criterion for the deviation
between the test site attenuation and the reference site attenuation given in CISPR 16-1-3
can be used in the uncertainty budget.
i) Clamp slide material
Typically the same clamp slide is used for clamp site validation and for clamp calibration
procedure. If the clamp slide material is not RF-transparent, then the possible perturbing
effects of the clamp slide material shall be taken into account.
j) SAD decoupling factor
The decoupling performance of the SAD specifies the decoupling of the far end of the LUT
from the near end of the LUT. A minimum requirement for the SAD decoupling factor is
given.
k) CDN impedance tolerance
For the clamp calibration, a CDN is specified to terminate the LUT near the reference
plane. In the lower frequency range (30 MHz – 230 MHz) this gives a common-mode
termination impedance of approximately 150 Ω. Beyond 230 MHz, the common-mode
termination impedance of CDNs is not specified. The tolerance of the common-mode
impedance of the CDN will affect the common-mode current in the LUT. However this
effect will also depend on the common-mode impedance contributions from the EUT, LUT
and the SAD. Quantitative information on the resulting uncertainty is not available. It is
estimated that the effect due to the CDN common-mode impedance tolerance is minor.
7.2.3.6 Repeatability of measurement
‘Measurement system repeatability’ is an influence quantity that is often a generic part of
uncertainty budgets.
CISPR 16-4-1 Amend. 1  IEC:2004(E) – 7 –

The repeatability of the calibration is determined by deriving the standard deviation of a series
of repeated calibration measurements using the same set up and measurement equipment. In

this way statistical information is gained about a number of influence quantities together, i.e.,

stability of the clamp, stability of the analyzer generator, measuring system reading, start/stop

position tolerance, clamp scanning. Consequently, if ‘repeatability of measurement’ is

included as a generic item of the uncertainty budget, then it is important to be sure that

certain influence quantities that are part of this ‘repeatability of measurement’ category, are

not included twice.
7.2.4 Application of the uncertainty budget

In general, the expanded uncertainty figure of the clamp factor is used by a test laboratory as
an input to derive the expanded uncertainty of its clamp measurement method. Note that for
this purpose, the standard uncertainty has to be derived from the expanded uncertainty. If we
assume that the uncertainty of the clamp factor has a normal distribution, then the expanded
uncertainty value of the clamp factor has to be divided by a factor k = 2. Consequently, the
clamp manufacturer may also directly provide the standard uncertainty instead of the
expanded uncertainty.
As already discussed in the previous section, the uncertainty figure of the clamp factor may
be a unit-specific figure or it may be a figure that is applicable to that type of clamp. The
uncertainty that is related to a type calibration is generally larger than the unit specific
uncertainty. The reason is that for type testing a limited number of samples of the same type
of clamp is used and the average of the individual clamp sources is taken as clamp factor of
that particular type. Consequently the uncertainty due to the spread of this average clamp
factor will result in an increased uncertainty.
7.2.5 Typical examples of an uncertainty budget
Tables C.1 and C.2 of Annex C give a typical uncertainty budget for the original clamp
calibration method in the two frequency bands 30 MHz – 300 MHz and 300 MHz – 1 000 MHz
respectively. The uncertainty budgets for the jig calibration method and the reference device
calibration method are still under consideration.
The uncertainty budgets are calculated in accordance with the procedure given in Clause 4.
Each budget contribution can be determined by using the Type A and Type B methods of
evaluation. Type A evaluations of uncertainty are done by using statistical analysis of
repeated measurement, and Type B evaluations of uncertainty are done by other than
statistical analysis.
In practice, EMC compliance measurements are typically executed once for a certain type of
EUT. Repeated measurements using the same EUT are not common practice. Therefore, the

uncertainty budget contributions are mostly determined using the Type B method of
evaluation.
This is also the case for the budgets presented in Annex E, i.e., most of the budget
contributions are Type B evaluations and use data from calibration certificates,
instrumentation manuals, manufacturers’ specifications, previous measurements or from
models or generic understanding of the measurement method. The probability distributions
and uncertainty values for the various uncertainty sources/influence quantities that are given
in Annex C are derived from various sources of information [16] [17] [20].
Unfortunately no model is available for the relation between the measurand and the various
influence quantities. All that can be said is that the measurand is a function of the influence
quantities given in Table 7. Most standard uncertainty values of each influence quantity must
be derived from specifications or from experimental data. Further, it is assumed that all
sensitivity coefficients are equal to one. However, due to the absence of a realistic model, the
true value of the sensitivity coefficients is unknown.

– 8 – CISPR 16-4-1 Amend. 1  IEC:2004(E)

From the clamp calibration uncertainty budgets given in Annex C it can be concluded that the

expanded uncertainty is approximately 3 dB for the frequency band of 30 MHz – 1 000 MHz.

The latter value is also applied in the tables of Annex D. Note that this value is also used in

the disturbance power uncertainty budget given in Table A.3 of CISPR 16-4-2.

7.2.6 Verification of the uncertainty budget

Two round robin tests (RRTs) have been carried out as part of the CISPR work on modifying

the clamp calibration method. The results of the last RRT are reported in [18]. Six test
laboratories contributed to this RRT. The standard deviation was less than approximately

1 dB over the frequency band of 30 MHz to 1 000 MHz, resulting in an expanded uncertainty

of approximately 2 dB.
7.3 Uncertainties related to the absorbing clamp measurement method
This section describes the determination of the uncertainty budgets for the absorbing clamp
test method (ACTM) described in Clause 7 of CISPR 16-2-2.
For convenience a schematic overview of the clamp measurement method is given in
Figure 20.
7.3.1 The measurand
For a clamp measurement, the measurand is the disturbance power. The disturbance power
P corresponding to the measured voltage V at each measurement frequency is calculated by
using the clamp factor CF obtained from the absorbing clamp calibration procedure described
in CISPR 16-1-3.
P = V +CF
where
P = the disturbance power in dB(pW)
(21)
V = the measured voltage in dB(µB()
CF = the clamp factor in dB(pW/µB)

7.3.2 Uncertainty sources
This section gives the uncertainty sources associated with the clamp measurement. From
equation 21 we see that the uncertainty is determined by the uncertainty of the voltage
measurement and the uncertainty of the clamp factor.

The uncertainty of the voltage measurement is determined by the uncertainties induced by the
EUT, the set-up, the measurement procedure, the measurement instrumentation and the
environment.
Figure 20 gives a schematic overview of all the relevant uncertainty sources. This fish-bone
diagram indicates the categories of uncertainty sources that contribute to the overall
uncertainty of the disturbance power. From this diagram we see that most set-up related
uncertainty sources are the same as the sources that were applicable for the clamp
calibration. An important set-up uncertainty source that has been added is the reproducibility
of the set up of the EUT. For the measurement instrumentation uncertainty, now the absolute
uncertainty of the receiver and the uncertainty of the clamp factor are important uncertainty
sources that were not relevant for the clamp calibration.

CISPR 16-4-1 Amend. 1  IEC:2004(E) – 9 –

7.3.3 Influence quantities
For most of the uncertainty sources given in Figure 20, no real influence quantities can be

defined to translate the qualitative uncertainty source in question. Table 8 gives the relation

between the uncertainty source and the influence quantity. If no influence quantity can be

given, then in the uncertainty budget, the original uncertainty source will be used.

For each of the uncertainty sources or influence quantities that are new or that deviate from

the calibration situation (see 7.2.3) some explanation is given in the following subclauses.

7.3.3.1 EUT-related
a) Size of EUT
Various influence quantities depend on the type of the EUT, i.e., large EUTs, small EUTs,
EUTs with just one, or with many cables. The electromagnetic behavior of these different
types may cause different magnitudes of uncertainty.
b) Signature of disturbance
The signature of the disturbance (wide band, narrow band) may affect the magnitude of
uncertainties induced by the receiver.
c) Product sampling (optional)
This is especially important if the measurement is repeated by the manufacturer for quality
assurance reasons or if the 80 %/80 % rule is to be applied. If the manufacturer performs
a type test, then the manufacturer may repeat the measurement using different samples of
the same type of EUT. In case of market control by an authority using different samples of
the same type of EUT, then also the 80 %/80 % rule may be applied.
d) Set up unit(s) and cables
Despite the specification of the EUT set-up in product standards, this influence quantity
may give rise to significant uncertainties if the same EUT is prepared and set up by
different operators and test laboratories. Especially if the EUT consists of different units
and several interconnecting cables, the uncertainty due to the many degrees of freedom of
setting up the EUT may be significant. Also EUT cables have to be extended using
representative cables, to make clamp measurements possible. Different types
(diameter/shield performance etc) of extension cables may introduce also differences in
results.
e) Modes of operation EUT
During the measurement, meaningful modes of operation shall be selected. If the test
mode of operation is not specified, then different operators/test laboratories may select
different modes in conjunction with different receiver settings and scan speeds.
...