SIST IEC/TR 60816:1998

SLOVENSKI STANDARD
SIST IEC/TR 60816:1998
01-september-1998
Guide on methods of measurement of short duration transients on low voltage
power and signal lines
Guide on methods of measurement of short duration transients on low-voltage power
and signal lines
Guide sur les méthodes de mesure des transitoires de courte durée sur les lignes de
puissance et de contrôle basse tension
Ta slovenski standard je istoveten z: IEC/TS 60816
ICS:
29.240.01 2PUHåMD]DSUHQRVLQ Power transmission and
GLVWULEXFLMRHOHNWULþQHHQHUJLMH distribution networks in
QDVSORãQR general
SIST IEC/TR 60816:1998 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

---------------------- Page: 1 ----------------------

SIST IEC/TR 60816:1998

---------------------- Page: 2 ----------------------

SIST IEC/TR 60816:1998
RAPPORT CEI
TECHNIQUE IEC
816
TECHNICAL
édition
Première
REPORT
First edition
1984
Guide sur les méthodes de mesure des transitoires
de courte durée sur les lignes de puissance et de
contrôle basse tension
Guide on methods of measurement of short
duration transients on low voltage power and
signal lines
© CEI 1984 Droits de reproduction réservés — Copyright - all rights reserved
Aucune partie de cette publication ne peut être reproduite ni No part of this publication may be reproduced or utilized
utilisée sous quelque forme que ce soit et par aucun procédé, in any form or by any means, electronic or mechanical,
électronique ou mécanique, y compris la photocopie et les including photocopying and microfilm, without permission
microfilms, sans l'accord écrit de l'éditeur. in writing from the publisher
Bureau central de la Commission Electrotechnique Internationale 3, rue de Varembé Genève Suisse
CODE PRIX
Commission Electrotechnique Internationale
X
PRICE CODE
International Electrotechnical Commission
IEC
Me ayHapogHae 3nenTporexHH4ecrtaalioMHCCHa
Pour prix, voir catalogue en vigueur
•
For price, see current catalogue

---------------------- Page: 3 ----------------------

SIST IEC/TR 60816:1998
— 3
816 © I E C 1984
CONTENTS
Page
5
FOREWORD
5
PREFACE
7
INTRODUCTION
Clause
1. Scope 7
2. Characteristics of transients 9
2.1 9
Environment-produced transients
2.2 Appliance-produced transients 9
2.3 Parameters to be measured 9
3.
Characteristics of mechanisms of coupling between transient sources and potentially
15
susceptible devices
3.1 Propagation modes 17
4. Susceptibility/Immunity 19
4.1 Damage effects 19
4.2 Malfunction effects 21
5. Instrumentation 23
23
5.1 Obtaining statistical data on parameters of transients
5.2 Transient counter 25
5.3 Peak voltmeter 25
27
5.4 Other parameters
27
5.5 Waveform recording and analysis
37
5.6 Transient energy measurements
5.7 Frequency domain measurement 39
45
5.8 Special inexpensive devices
45
Measurement techniques 6.
47
6.1 Measurement of conducted transients
61
6.2 Measurement of radiated transients
62
FIGURES
77
APPENDIX A — Method for measuring transient conducted emissions
83
APPENDIX B — Equipment input impedance
APPENDIX C — Example of a monitoring probe 89
Bibliography and references 90

---------------------- Page: 4 ----------------------

SIST IEC/TR 60816:1998
1984 — 5 —
816 © IEC
INTERNATIONAL ELECTROTECHNICAL COMMISSION
GUIDE ON METHODS OF MEASUREMENT
OF SHORT DURATION TRANSIENTS
ON LOW VOLTAGE POWER AND SIGNAL LINES
FOREWORD
1) The formal decisions or agreements of the I EC on technical matters, prepared by Technical Committees on which all
the National Committees having a special interest therein are represented, express, as nearly as possible, an
international consensus of opinion on the subjects dealt with.
2) They have the form of recommendations for international use and they are accepted by the National Committees in
that sense.
In order to promote international unification, the IEC expresses the wish that all National Committees should adopt
3)
the text of the IEC recommendation for their national rules in so far as national conditions will permit. Any
divergence between the IEC recommendation and the corresponding national rules should, as far as possible, be
clearly indicated in the latter.
PREFACE
has been prepared by IEC Technical Committee No. 77: Electromagnetic
This report
Compatibility between Electrical Equipment including Networks.
The text of this repo rt is based on the following documents:
Six Months' Rule Report on Voting
77(CO)21
77(CO)20
Further information can be found in the Repo rt on Voting indicated in the table above.

---------------------- Page: 5 ----------------------

SIST IEC/TR 60816:1998
— 7
816 © IEC 1984
GUIDE ON METHODS OF MEASUREMENT
OF SHORT DURATION TRANSIENTS
ON LOW VOLTAGE POWER AND SIGNAL LINES
INTRODUCTION
Transients appearing on power and signal lines are capable of producing a variety of effects
ranging from minor equipment performance degradation to catastrophic insulation breakdown.
They have a wide variety of waveforms, which depend upon the mechanism of generation.
Furthermore, those that originate from switching a.c. power on and off will have a form that
depends upon the exact moment in the power cycle at which switching takes place, but in
addition can have very complicated micro (detailed) and macro (overall) waveform
characteristics.
Because of this variety and the frequently random time of occurrence, there is considerable
difficulty in making a suitable measurement of a transient. The advent of new technologies in
device design and manufacture has increased concern for identifying more precisely the effects
of transients.
icular, a solid-state device can be susceptible even to an overvoltage of very sho rt
In part
(nanosecond) time duration. Furthermore, because of va riations in the waveforms, to have a
precise measurement of any given transient would require the measurement of a large number
of parameters. Even if one measures the exact waveform of a transient, for control purposes,
one must then describe the transient with a finite number of parameter values.
The choice of these parameters and their expected range of values is still a matter of some
speculation, and the proper method of measurement is still considered by some to be an open
question. Modern types of test equipment provide measurement capabilities not available
previously, but they must be used with particular care.
Accordingly, there is a need for well-defined and accepted methods of measuring transients
for two major reasons, namely so that:
a) measurements made by different laboratories may be compared;
b) meaningful limits may be placed on transients generated by particular types of equipment
and on the susceptibility of particular equipment to transients.
This guide has been prepared to assist in meeting these requirements. Note that in this
guide the concern is with transient phenomena which are not line-frequency related and are of
duration no greater than 40 ms. It is also not concerned with sustained voltage changes or
fluctuations.
1. Scope
rt duration
is intended to give guidance on methods of measurement of sho
This report
transients on low voltage power and signal lines.

---------------------- Page: 6 ----------------------

SIST IEC/TR 60816:1998

— 9
816 © I EC 1984
2. Characteristics of transients
Transients may be classified according to their origin as follows:
a) those produced by the environment, that is to say, by lightning;
those produced by electrical switching or faults;
b)
those produced internally within the circuits of particular equipment.
c)
2.1 Environment produced transients
These transients a rise from lightning and are most severe on overhead and unscreened
cable sections. At the point closest to the point at which the transient is generated, the
rise time can be sho rt and the amplitude high. The rise time and fall time can be
considerably lengthened and the amplitude reduced as the transient propagates along the
network. Typically, such transients have rise times of the order of microseconds and fall
times from 50 µs to 50 ms and may be oscillatory. The effects on inner conductors are
reduced in the case of screened cables and cables buried in areas of low ground
resistivity.
2.2 Appliance produced transients
Transients produced by appliances arise from three basic causes:
the operation of a mechanical or semiconductor switch;
a)
b) turn-on currents associated with the saturation properties of an iron-core transformer
or starting currents in motors;
c) faults within equipment.
The transient produced by a switch or fault can range from a simple surge or dip
(sag) to a very complex waveform caused by repeated "restriking" of an arc as the
contacts of a mechanical switch separate. The most serious transients usually arise as a
result of breaking an inductive circuit, for example, the blowing of a fuse. In many
cases, special techniques, such as placing capacitors across the contacts, will reduce the
magnitude of the transients generated, and in other cases suppression can be obtained by
the use of semiconductor devices. The transients can have rise times of the order of a
few nanoseconds in the immediate vicinity of the switch, that is to 'say, within a fraction
of a metre; however, at distances of several metres from the switch, the rise time will be
considerably increased due to attenuation of the line of the higher frequency components.
Switching of transformers produces transients which may be of the order of several times
the peak line voltage but will have rise times of the order of tens of microseconds.
Parameters to be measured
2.3
Because of the complex and variable nature of transients, it is difficult to specify
which parameters should be measured. Under such circumstances, it is useful to examine
the susceptibility characteristics of the equipment under consideration and to divide these
into several categories in order to determine the parameters to be measured (see Clause 4):
a) those which are susceptible to a restricted band of frequencies, such as radio or
carrier frequency receivers;

---------------------- Page: 7 ----------------------

SIST IEC/TR 60816:1998
816 © I EC 1984 - 11 —
those which are susceptible to a broad band of low radio frequencies (for example, a
b)
mains rectifier). For such devices the peak voltage is usually the critical parameter;
but energy may also be an important parameter;
those which are susceptible to a broad band of frequencies in the higher frequency
c)
bands. The critical quantity is the rate of rise of the pulse. Digital equipment is often
susceptible to this parameter, and destruction of devices may also occur.
Some general measurement capabilities can be desirable but one may not be able to
measure all parameters with a single instrument. For convenience, these parameters may
be classified according to whether they give information in the time or frequency
domains.
Figure 1, page 62, illustrates the possible complex nature of a typical transient and
some of the time domain parameters that may be used to describe it. In addition,
effective pulse strength (voltage x time) and energy content may be significant.
The most common frequency domain parameter used to describe a transient is the
spectrum amplitude. The frequency vs. phase characteristic may also be impottant but is
not usually measured because of difficulties in both measurement and use of the data.
Where the interference is discontinuous in nature, time weighting techniques such as
those used in the C.I.S.P.R. instrument may also be applied. The unweighted component
is of interest in any case.
2.3.1 Relation between time domain and frequency domain parameters
Figure 2 a), page 63, shows a representative waveform of one type of transient
disturbance produced during a switching-off operation of a 220 V auxiliary conductor.
Figure 2 b), page 63, shows a spectrum amplitude representation of such a waveform.
The relationship between the spectrum amplitude plot and the time domain waveform is
best explained by comparing the relevant characteristics for a trapezoidal pulse.
The spectrum amplitude of a symmetrical trapezoidal pulse with the mean pulse time
T is, in the frequency range below f = 1/7t T, independent of frequency (this po rtion of
the spectrum amplitude curve is parallel to the abscissa) and has a magnitude equal to
the envelope of the
the amplitude-time area of the pulse. Above the frequency f = 1/n T
spectrum varies as 1/f. If the trapezoidal pulse has rise and fall times t, the envelope of
the spectrum amplitude above the frequency 1/1tt varies as 1/f2.
Note that on Figure 2 b) the abscissa is marked in megahertz on a logarithmic scale
6 uV
and the ordinate is given in decibels with respect to 1 µVs. (1 µVs corresponds to 10
in 1 MHz.) The spectrum amplitude representation can be calculated using standard
Fourier integral techniques. When the pulses are repeated at regular intervals, a discrete
spectrum rather than a continuous spectrum is obtained. In that case, a representation
corresponding to Figure 2 b) can be used, but the curve shown corresponds to the
amplitude of the discrete components (envelope curve) which are spaced on the
frequency scale at a distance corresponding to the repetition rate.

---------------------- Page: 8 ----------------------

SIST IEC/TR 60816:1998
816 © I EC 1984 — 13 —
Accordingly, the following interpretation can be placed on Figure 2 b), page 63:
a) the low-frequency or flat po rtion of the curve is a level determined by the effective
area under the voltage-time curve shown in Figure 2 a), page 63;
b) the high-frequency po rtion, above about 20 MHz, falls off at a rate inversely
proportional to the square of the frequency, and the points at which this rate of
fall-off begins are determined by the rate of rise of the initial pa rt of the waveform
(that is to say to the amplitude U,);
c) the peak in the spectrum amplitude curve appears at a frequency equal to the
frequency of oscillation of the transient. Thus, if one is given a spectrum
rtant
corresponding to that in Figure 2 b), one can interpret it in terms of the impo
characteristics of the originating transient waveform.
Furthermore, as shown in Figure 2 b), at point p which is the point of intersection of
the actual curve with the low frequency (horizontal) po ion of the curve, by extending
rt
from this point a line with slope proportional to 1/f (shown dotted on Figure 2 b)) and
one with a slope proportional to 1/f 2 (the actual spectrum curve shown in the solid
line), one can obtain, from the scales shown on the right-hand po rtion of Figure 2 b) the
actual maximum voltage dB(V) and the rate of rise dB(kV/µs) [10]*.
Measuring practice sets limits on the viewing time in time domain measurements and
on bandwidth in frequency domain measurements. Therefore, if transients of unknown
characteristics (amplitude, rise time, duration, repetition frequency) are to be measured,
the measurements should be performed both in the time domain and the frequency
domain. In this way, maximum information about the transients can be obtained.
2.3.2 Importance of various transient parameters
a) Rise time
The rise time characterizes the transient in its amplitude-frequency relation (see
Fourier series development). The shorter the rise time, the more extensive is the
disturbing action in the frequency spectrum. Normally one would expect the risks of
performance degradation of a susceptible device to be dependent on its acceptance
bandwidth, among other factors. It has been reported that, in practice, the rise
time/amplitude relation shows that 5% of the disturbances have significant components
above 10 MHz and only 1% above 30 MHz. (However, even very low level
components at VHF may interfere with radio reception.)
Amplitude
b)
The amplitude is especially significant for long transients (for example, >1 µs). It can
be the most significant quantity relating to performance degradation or semiconductor
device destruction.
c) Energy
The energy of the transient, although related to the amplitude, is also dependent on
ant parameter with
the internal impedance of the disturbance source and is an impo rt
regard to component destruction.
* The figures in square brackets refer to "Bibliography and references", page 90.

---------------------- Page: 9 ----------------------

SIST IEC/TR 60816:1998
816 © I EC 1984 — 15 —
d) Duration
The importance of the parameter depends on the time constant of the susceptible
equipment in question. For logic systems, the chances of release of the circuits
controlled by the synchronization clock may be increased.
e) Range of frequencies
As mentioned in paragraph a), the spectrum of the disturbance may not be significant
above 10 MHz to 30 MHz (maximum).
Repetition frequency
f)
In general, a knowledge of the repetition frequency is impo rtant for estimating the
disturbing effect of transients. For analogue systems, its impo rtance depends on the
time constant of the susceptible equipment and can involve an integration
phenomenon. For logic systems, the risks of failure may be most severe if the
transient and control signal are in phase.
3. Characteristics of mechanisms of coupling between transient sources and potentially susceptible
devices
The transients of concern here are assumed to be coupled to the susceptible device
primarily by conduction. They are usually initiated by some switching action on the
connected power line. The switching action could be at any point either locally (on the
immediate low-voltage dist ribution circuit) or at a more remote point on a high-voltage
transmission line. Transients may also originate from atmospheric effects, for example,
lightning, either as a direct strike on a high-voltage line or from a ground stroke by
induction into a high-voltage or low-voltage dist ribution circuit. When the susceptible
device is located close to the o riginal disturbance, the coupling is primarily by induction.
In such a path, the effect of the coupling is described in terms of three basic
parameters:
a) the attenuation characteristic as a function of frequency of the line;
b) the nature of the loading on the line;
c) the geometry in relation to the ground plane.
Since power lines are very rarely loaded in their characteristic impedance, one can
expect multiple reflections to occur on the line at each discontinuity, for example,
rtance in
wherever a load is connected. Reflection characteristics are of considerable impo
shaping transients, especially those produced by switching operations. The consequence is
ringing, at a frequency usually in the range of tens of kilohe rtz to tens of megahertz,
which causes the spectrum amplitude of the transient to have a peak at that particular
frequency.
Similar ringing is also possible as a result of conducted transients produced by
appliances; however, the separation between discontinuities is smaller and therefore the
ringing frequency can be much higher. However, it should be noted that the attenuation
of the line increases with frequency, so that the ringing would usually be observed only
for transients which are measured at positions relatively close to the source.

---------------------- Page: 10 ----------------------

SIST IEC/TR 60816:1998
— 17 —
816 © I E C 1984
The possibility of coupling as the result of induction (both inductive and capacitive
coupling) between a power line and a communication line must also be considered. This
is especially important in industrial plants where power cables and control or signal
cables run side by side over relatively large distances. Generally, such coupling
mechanisms can be reduced by using twisted pair or coaxial cables and by placing the
cables in screwed metal conduits.
Another source of coupling is a finite ground plane impedance. Many transients, for
example, are propagated in common mode on transmission lines and the return current
flows through the ground plane. If the ground return path is not of very low impedance
or the points of connection to the ground return path are close to similar return points
for a sensitive circuit, significant differences in potential can be produced. Balanced
symmetrical circuits can be used to minimize the effects of common-mode coupling, but
any small imbalance in the sensitive circuit may be critical.
In the case of direct coupling along the power cable from one equipment to another,
low-pass filters can often be used to suppress unwanted effects.
Propagation modes
3.1
The four general modes of propagation for power lines are shown in Figure 3, page
64. Similar modes of propagation exist on signal lines. As shown in Figure 3, there are
two main modes of conducted propagation: asymmetrical or common-mode (CM) and
symmetrical or differential/balanced mode (DM). Nearly all commercial products have a
protective conductor. In some domestic installations, a two-wire power system with no
protective conductor is used. Most low-voltage installations have the protective conductor
connected to earth at the service entrance.
For some purposes, measurements are made from each phase to earth. The relations
between phase A, B, common mode, U,M, and differential mode, open-circuit
U[)M+
voltages are shown in Figure 4, page 65.
A B, then
If the phase voltages are, respectively, and
A+ B
UCM =
2
A — B
UDM =
Measured impedances between phase and earth are shown in Figure 5,
page 66 [1]. This impedance plays a critical role in controlling the inse rtion loss between
the transient source and the point of measurement. Consider the various paths as
illustrated in Figure 6a), page 67, [2]. It has been found that the mean differential mode
insertion loss shown in Figure 6b), page 68, was controlled by the mismatch of the
various impedances. The method of signal injection used a current probe technique
shown in Figure 6c), page 69. Note that the differential mode loss is more or less
independent of frequency up to 30 MHz.
The differential mode impedance has a well-defined value, for example, 50 S2 for a
coaxial line and higher for a balanced line. In common mode, up to perhaps several tens
of kilohertz, the impedance can be expected to have a value approximately equal to the
reactance of a line of equivalent length and grounded by a low or zero impedance.

---------------------- Page: 11 ----------------------

SIST IEC/TR 60816:1998
O
816 O I E C 1984 — 19 —
4. Susceptibility/Immunity
Certain types of electrical equipment are potentially susceptible to transients unless
suitable preventive measures have been incorporated to provide immunity in the
environment. In general, long-term experience in the use of cables, connectors, capacitors,
insulating materials, transformers, switches, etc., has established the margins required to
enable transient overvoltages to be withstood, and for many components the appropriate
overvoltage tests are specified. However, for equipment incorporating semiconductor
devices there are various forms of susceptibility likely to occur, including catastrophic
damage and temporary malfunction. Some of these effects are discussed below, in
particular, because transient measuring and analysis equipment must not suffer these
effects.
4.1 Damage effects
Damage effects are largely confined to semiconductor devices although insulation
failure of other components can occur because of particularly high amplitude transients,
for example, nearby lightning strokes. Power semiconductor devices connected to the
supply lines are subjected to the full transient voltage, but devices of adequate rating are
selected for such applications based upon earlier experience of device failures.
Semiconductor devices in low-level signal and control circuits are only coupled indirectly
to the supply lines, but damage can occur since the devices in general have a fairly low
voltage and/or current rating. These coupling mechanisms involve high frequency
components of the transients and may be difficult to assess in many applications, so that
preventive measures to protect the devices require some consideration. Examples of
damage effects are given below.
4.1.1 Power semiconductor devices
These devices can be damaged by voltage transients (spikes) with durations as short as
about 1 ns. The likelihood of damage is a function of transient amplitude, duration,
polarity, rate of rise, position on the supply waveform, etc., as well as the device
parameters. The initial breakdown of the device is likely to be followed by a high
current discharge from the supply, which causes catastrophic damage. Typical devices
which have been found liable to damage are rectifier diodes in electronic equipment and
thyristors used for motor speed control.
4.1.2 Low-level signal and control circuits
While these circuits are not generally directly connected to the low-voltage supply
mains, there is coupling between them via the d.c. supply circuit and by induced effects
in signal and control cables, so that transients of reduced amplitude can be injected into
the circuits. Various semiconductor devices liable to damage by relatively low-level
rtain
transients are incorporated in these circuits, for example, integrated circuits, ce
discrete devices (such as field effect transistors and special purpose diodes (such as
tunnel diodes).

---------------------- Page: 12 ----------------------

SIST IEC/TR 60816:1998
816 CO 1984 — 21 —
4.2 Malfunction effects
Various forms of equipment malfunction can be caused by transients generated on
low-voltage supply lines, which may or may not be coupled to signal lines, on a wide
range of types of equipment. Some of these effects could create safety hazards, for
example, fire or explosion in chemical manufacturing plants or sudden changes in motor
speed. However, the majority of the malfunctions likely to occur are relatively harmless,
possibly producing only a temporary effect which is quite acceptable to the user, for
example, a small transient change in a meter reading.
In practice, two different types of transients on the supply are found to be the cause
of 'most of the observed malfunctions, that is to say voltage spikes with durations of the
order of 1 µs and voltage dips or sags (reductions lasting for about 10 ms and longer).
Voltage dips are not covered in this guide, apart from the following note.
Note. — Voltage dips (sags), that is to say reductions in the supply voltage to electronic equipment lasting for
about 10 ms or longer, can upset the operation of the equipment because of the effects of reduction in
the internal stabilized d.c. voltage supply. The effects can be very drastic on certain types of equipment
and examples are given as follows:
a) Digital systems
Serious malfunctioning of digital systems will occur if the d.c. voltage supply is reduced significantly.
The effects produced can include corruption of data system "lock-up", loss of programme, etc.
b) Control systems
These systems are liable to suffer serious malfunctioning, causing disruption of the control function.
c) Instrumentation
Most types of instrumentation are likely to malfunction seriously as a result of voltage dips (sags).
d) Alarm and trip systems
False operation of these systems is likely to be caused by voltage dips (sags).
4.2.1 Effects of voltage spikes
a) Digital systems
Equipment which incorporates digital systems (such as computers, microprocessors, and
instrumentation) can be affected by voltage spikes which are coupled into the logic
circuits and corrupt the data. The effects may be overcome by various error correction
techniques but in extreme cases the corruption may cause serious effects (for example,
incorrect control function, systems "lock-up", unwanted change of programme, and
feeding incorrect data into a store).
b) Control systems
Control equipment can be affected by induced voltage spikes causing a malfunction of
the system.
Instrumentation
c)
Incorrect indication by some types of equipment can be produced by the effects of
spikes.
d) Alarm and trip systems
Undesired operation or failure of operation of these systems can be triggered by
voltage spikes.

---------------------- Page: 13 ----------------------

SIST IEC/TR 60816:1998
816 © I EC 1984 — 23 —
Equipment incorporating power semiconductor devices
e)
Motor speed control by semiconductors can be affected by voltage spikes and typically
takes the form of a sudden transient increase in speed. Heating controls are not so
drastically affected by single voltage spikes, but repetitive spikes could cause a large
change in temperature.
5. Instrumentation
In this clause various methods of measurement are described. In some cases the
instruments are available commercially, in others they have been constructed in
laboratories for particular experiments. The objective is to provide the user with guidance
on the significant characteristics of all such instruments.
In general, an instrument can be considered to consist of four basic pa rts as follows:
a) detector;
processor;
b)
c) output display;
d) control system.
These parts are related as shown in Figure 7, page 70.
The basic type of instrument is determined by the type of detector. Instruments having
similar detectors but produced by different manufacturers may differ principally in the
ways in which other functions are performed. Indeed, in some cases these functions may
be adjustable or in fact performed to various extents by auxiliary apparatus under the
control of the operator.
The instruments are described in several categories. In each case, fundamental
principles of operation are described along with the relationship between fundamental
parameters and those frequently stated in commercial literature.
5.1 Obtaining statistical data on parameters of transients
The fact that transients are so variable from one instance to the next, means that,
except where some particular significance can b
...