Frequency Domain Spectroscopy

Abstract Frequency domain spectroscopy (FDS) is a non-

destructive method of assessing the insulation condition of
high voltage equipment, especially oil-paper insulation
system of transformers. It is a reliable tool for predictive
maintenance of such equipment in the field. In this method
each test object is considered as a black box accessible only
through its termi-nals. Consequently this method identifies
the global changes in the insulation condition and is not
suitable for identifying any local defect in the insulation
system. FDS results are affected by the geometry of the
insulation system, i.e. the relative dimensions of spacers,
barriers and oil ducts. The external factors that affect FDS
most are operating temperature and weather condi-tions like
rain. For on-site testing significant errors can be introduced
in FDS measurement, if certain precautions are not taken,
particularly when different equipments are connected in
parallel with the transformer. Mathematical modeling is
important for the interpretation of FDS results for the
estimation of moisture content in solid insulation.
Basic Theory of Frequency Domain Spectroscopy

Effect of Moisture in Insulation

FDS results are significantly influenced by the presence of
moisture in insulation. In gen-eral, a higher dielectric
dissipation factor is obtained with the increase of moisture
content in cellulose insulation. Fig. 1.5(a) shows that at
frequencies lower than 0.01Hz a sharp rise of real part of
complex permittivity occurs in the case of pressboard
samples having moisture content >2%. From 1.5(b) it may be
seen that the dielectric losses in impregnated pressboard
increases considerably with the increase in moisture content
particularly for frequencies low-er than 100Hz. For
frequencies higher than 10kHz, the difference between the
loss curves of pressboards having different moisture content
becomes insignificant.
Researchers have found that the dielectric dissipation factor
as well as the capacitance val-ues increases with aging.
Higher capacitance means higher permittivity and hence
poor insu-lation condition. Thus from FDS results different
aging conditions can be detected particular-ly at low
frequencies. This fact can be seen from Fig 1.6(a) and (b)
where the tan and capacitance values of oil impregnated
paper insulation have been found to increase with ag-ing at
frequencies below 0.1Hz.
Advantages and Disadvantages of Dielectric Response
Measurements in Time-Domain and Frequency-
Domain
Dielectric response measurement techniques in time domain,
viz. PDC and RVM, as well as that in frequency domain
provide useful information on the condition of transformer
insula-tion. Practical experience shows that all the three
techniques rank the transformers in the same order. Salient
features of the three measurement techniques are discussed
below. RVM being a high input impedance method, leakage
current, e.g. surface leakage over the bushings, affect the
measurement. RVM has been found to be more sensitive to
systematic er-rors and the interpretation of RVM results is
not always unambiguous, especially in the pres-ence of
multiple peaks in RVM spectrum for old transformers.
Moreover, RVM considers in-sulation.

Effect of Insulation Aging

geometry and oil conductivity only to a limited extent. It
takes very long time to rec-ord the complete polarization
spectrum. A more practical approach has been suggested in
in which three recovery voltage measurements are
recommended to be taken with charging times 10s, 100s and
1000s.
Interpretation of PDC results is straightforward and it
provides moisture content in solid in-sulation and both oil
and paper conductivities. Other diagnostic parameters like
tan and po-larization spectra can be calculated from PDC
results. But, drift and jitter free voltage source is required for
accurate analysis. Since the currents are low, the
measurements are affected by offset currents and
interference in the field. But it is possible to implement
guarded meas-urement to eliminate the surface leakage
currents from measurements. Another important as-pect of
PDC measurement is the switching transients at the start of
polarization and depolari-zation which means that the first
few seconds are difficult to measure, which in turn makes the
lower limit of the equivalent frequency about 1Hz.
Interpretation of FDS results is also straightforward. It could
separate real and imaginary parts of complex capacitance
and complex permittivity. It also provides the moisture
content in solid insulation and diagnostic quantity like C-
ratio. FDS method is relatively insensitive to interference due
to its inherent small bandwidth.
Terminology
Factor of Earthing: This is the ratio of the highest r.m.s.
phase-to-earth power frequency voltage on a sound phase
during an earth fault to the r.m.s. phase-to-phase power
frequency voltage which would be obtained at the selected
location without the fault. This ratio characterises, in general
terms, the earthing conditions of a system as viewed from
the selected fault location. Effectively Earthed System : A
system is said to be effectively earthed if the factor of
earthing does not exceed 80%, and non-effectively earthed if
it does.
[Note: Factor of earthing is 100% for an isolated neutral
system, while it is 57.7% (corresponding to 1/%3) for a solidly
earthed system. In practice, the effectively earthed
condition is obtained when the ratio x0/x1 < 3 and the ratio
r0/x1 < 1.
Insulation Level: For equipment rated at less than 300 kV, it
is a statement of the Lightning impulse withstand voltage and
the short duration power frequency withstand voltage.
For equipment rated at greater than 300 kV, it is a statement
of the Switching impulse withstand voltage and the power
frequency withstand voltage.
Conventional Impulse Withstand Voltages: This is the peak
value of the switching or lightning impulse test voltage at
which an insulation shall not show any disruptive discharge
when subjected to a specified number of applications of this
impulse under specified conditions. Conventional Maximum
Impulse Voltage: This is the peak value of the switching or
lightning overvoltage which is adopted as the maximum
overvoltage in the conventional procedure of insulation co-
ordination.
Statistical Impulse Withstand Voltage: This is the peak value
of a switching or lightning impulse test voltage at which
insulation exhibits, under the specified conditions, a 90%
probability of withstand. In practice, there is no 100%
probability of withstand voltage. Thus the value chosen is
that which has a 10% probability of breakdown.Statistical
Impulse Voltage: This is the switching or lightning
overvoltage applied to equipment as a result of an event of
one specific type on the system (line energising, reclosing,
fault occurrence, lightning discharge, etc), the peak value of
which has a 2% probability of being exceeded.
Rated Short Duration Power Frequency Withstand Voltage:
This is the prescribed r.m.s. value of sinusoidal power
frequency voltage that the equipment shall withstand during
tests made under specified conditions and for a specific time,
usually not exceeding one minute.
Protective Level of Protective Device: These are the highest
peak voltage value which should not be exceeded at the
terminals of a protective device when switching impulses and

lightning impulses of standard shape and rate values are
applied under specific conditions.
In order to avoid insulation failure, the insulation level of
different types of equipment connected to the system has to
be higher than the magnitude of transient overvoltages that
appear on the system. The magnitude of transient over-
voltages are usually limited to a protective level by protective
devices. Thus the insulation level has to be above the
protective level by a safe margin. Normally the impulse
insulation level is established at a value 15-25% above the
protective level.
Consider the typical co-ordination of a 132 kV transmission
line between the transformer insulation, a line gap (across an
insulator string) and a co-ordinating gap (across the
transformer bushing). In co-ordinating the system under
consideration, we have to ensure that the equipment used
are protected, and that inadvertent interruptions are kept to
a minimum. The co-ordinating gap must be chosen so as to
provide protection of the transformer under all conditions.
However, the line gaps protecting the line insulation can be
set A typical set of characteristics for insulation co-ordination
by conventional methods, in which lightning impulse voltages
are the main source of insulation failure, is shown in the
figure 1.3.
For the higher system voltages, the simple approach used
above is inadequate. Also, economic considerations dictate
that insulation co-ordination be placed on a more scientific

Conventional method of insulation co-ordination Statistical Method of Insulation Co-ordination

At the higher transmission voltages, the length of insulator
strings and the clearances in air do not increase linearly with
voltage but approximately to V1.6 The required number of
suspension units for different overvoltage factors is shown.
It is seen that the increase in the number of disc units is only
slight for the 220 kV system, with the increase in the
overvoltage factor from 2.0 to 3.5, but that there is a rapid
increase in the 750 kV system. Thus, while it may be
economically feasible to protect the lower voltage lines up to
an overvoltage factor of 3.5 (say), it is definitely not
economically feasible to have an overvoltage factor of more
than about 2.0 or 2.5 on the higher voltage lines. In the
higher voltage systems, it is the switching overvoltages that is
predominant. However, these may be controlled by proper
design of switching devices.
Evaluation of Risk Factor
The aim of statistical methods is to quantify the risk of failure
of insulation through numerical analysis of the statistical
nature of the overvoltage magnitudes and of electrical
withstand strength of insulation.
The risk of failure of the insulation is dependant on the
integral of the product of the overvoltage density function
f0(V) and the probability of insulation failure P(V). Thus the
risk of flashover per switching operation is equal to the
area under the curve I f0(V)*P(V)*dV.
Length of Overhead Shielding Wire
For reasons of economics, the same degree of protection is
not provided throughout a transmission line. Generally, it is
found sufficient to provide complete protection against
Surge Protection
direct strikes only on a short length of line prior to the
substation. This can be calculated as follows.
Consider a surge e approaching the terminal equipment.
When the surge magnitude exceeds the critical voltage e0,
corona would occur, distorting the surge wavefront, as it
travels. The minimum length of earth wire should be chosen
such that in traversing that length, all voltage above the
maximum surge that can arrive att the terminal has been
distorted by corona. [The maximum permissible surge
corresponds to the incident voltage that would cause
insulation failure at the terminal equipment.]
Modification of Waveshape by Corona
When a surge voltage wave travelling on an overhead line
causes an electric field around it exceeding the critical stress
of air, corona will be formed. This corona formation
obviously extracts the energy required from the surge. Since
the power associated with corona increases as the square of
the excess voltage, the attenuation of the waveform will not
be uniform so that the waveform gets distorted. Further,
corona increases the effective radius of the conductor giving
rise to a greater capacitance for the outer layers. Since the
line inductance remains virtually a constant, the surge
associated with the outer layers of corona would have a
lower wave velocity than in the conductor itself. These
effects in practice give rise to a wavefront distortion and not
a wavetail distortion, as shown in figure 10.7.
An overhead earth wire provides considerable protection
against direct strikes. They also reduce induced
overvoltages. However, they do not provide protection
against surges that may still reach the terminal equipment.
Such protection may either be done by diverting the major
part of the energy of the surge to earth (surge diverters), or
by modifying the waveform to make it less harmful (surge
modifiers). .
Spark gaps for surge protection
The simplest and cheapest form of protection is the spark
gap. The selected gap spacing should no only be capable of
withstanding the highest normal power frequency voltage
but should flash-over when overvoltages occur, protecting
the equipment. Also, once a gap flashes over under a surge
voltage, the ionised gap allows a power frequency follow
through current, leading to a system outage. Thus rod gaps
are generally used as a form of back up protection rather
than the main form of protection.
Expulsion Tube Lightning Arrestor
An expulsion tube arrestor consists essentially of a spark gap
arranged in a fibre tube, and another series external rod gap.
A typical arrangement for a 33 kV expulsion tube, with the
external gap of the order of 50 mm and the internal gap of
about 180 mm is shown in figure 10.9. The purpose of the
external gap is to isolate the fibre tube from normal voltages
thus preventing unnecessary deterioration. The high
temperature of the arc rapidly vaporises the organic
materials of the wall of the tube and causes a high gas
pressure (up to 7000 p.s.i.) to be built up. The high Figure
10.9 - Expulsion Tube pressure and the turbulence of the gas
extinguishes the arc at a natural current zero, and the hot
gasses are expelled through the vent in the earthed
electrode. The power frequency follow current is interrupted
within one or two half cycles so that protective relays would
not operate causing unnecessary interruptions.
Surge diverters (or lightning arrestors) generally consist of
one or more spark gaps in series, together with one or more
non-linear resistors in series. Silicon Carbide (SiC) was the
material most often used in these non-linear resistor surge
diverters. However, Zinc Oxide (ZnO) is being used in most
modern day surge diverters on account of its superior volt-
ampere characteristic. In fact the ZnO arrestor is often used
gapless, as its normal follow current is negligibly small. The
volt-ampere characteristics of SiC and of ZnO non-linear
elements are shown for comparison with that of a linear
resistor.
the voltage does not collapse to zero instantly due to the
voltage across the non-linear resistor. When the surge
voltage increases, there is a corresponding but rapid
decrease of the resistance discharging the surge energy to
earth. Once the surge passes through, the power frequency
voltage remaining is insufficient to maintain a sufficient
current for the arc to continue. Thus the arcs extinguish and
the gaps reseal. In the case of the ZnO arrestor, due to the
negligible continuous power frequency current even in the
absence of a series gap, the series gap is sometimes
eliminated simplifying construction.

The problem is generally overcome by having high equal

resistances shunting the series gaps, ensuring a uniform
distribution.
Selection of Surge Diverters
Surge Diverters When a surge appears at a surge diverter terminal, within a
Surge diverters for a particular purpose are selected as
short time the breakdown voltage of the series gap is
follows.
reached, and the arrestor discharges. Unlike in the rod gap,

(a) Rated Voltage
The designated maximum permissible r.m.s. value of power
frequency voltage between line and earth terminals.
This is generally selected corresponding to 80% of the system
phase-to-phase voltage for effectively earthed systems and
corresponding to 100% of the system phase-to-phase voltage
for non-effectively earthed systems. [Note: A surge diverter
of a higher rating may sometimes have to be chosen if some
of the other required criteria are not satisfied by this
diverter].
(b) Discharge Current
The surge current that flows through the surge diverter after
spark over.
Nominal discharge current: This is the discharge current
having a designated crest value and waveshape, which is
used to classify a surge diverter with respect to durability and
protective characteristics.
The standard waveform for the discharge current is taken as
8/20 µs). he nominal value of discharge current is selected
from the standard values 10 kA (station type), 5 kA
(intermediate line type), 2.5 kA (distribution type) and 1.5 kA
(secondary type), depending on the application. The highest
ratings are used for.
99 % of discharge currents are less than 10 kA 95 % of
discharge currents are less than 5 kA 90 % of discharge
currents are less than 3 kA 70 % of discharge currents are
less than 1 kA 50 % of discharge currents are less than 0.5 The power frequency spark over voltage should generally be
kA greater than about 1.5 times the rated voltage of the
(C) Discharge Voltage (or Residual voltage) arrestor, to prevent unnecessary sparkover during normal
switching operations.
The Discharge voltage is the voltage that appears between
the line and earth terminals of the surge diverter during the (e) Impulse spark over voltage
passage of discharge currents. The impulse spark over voltage is the highest value of voltage
The discharge voltage of the selected arrestor should be attained during an impulse of a given waveshape and
below the BIL of the protected equipment by a suitable polarity, applied between the line and earth terminals of a
margin (generally selected between 15% and 25%). surge diverter prior to the flow of discharge current.
The impulse spark over voltage is not a constant but is
The discharge voltage of an arrestor at nominal discharge
current is not a constant, but also depends on the rate dependant on the duration of application. Thus it is common
to define a wavefront impulse sparkover voltage in addition
of rise of the current and the waveshape. Typically, an to the impulse spark-over voltage.
increase of the rate of rise from 1 kA/µs to 5 kA/µs would
increase the discharge voltage by only about 35 %.
The dependence of the discharge voltage on the discharge
current is also small. Typically, an increase of discharge
current from 5 kA to 10 kA would increase the discharge
voltage by about 15% for Silicon Carbide arrestors and by
about 2% for Zinc Oxide arrestors. {The discharge voltage is
more often referred to as the residual discharge voltage].
(d) Power frequency spark over voltage ;- The power
frequency spark over voltage is the r.m.s. value of the lowest
power frequency voltage, applied between the line and earth
terminals of a surge diverter, which causes spark-over of all
the series gaps.
 at the terminal increases with separation for typical rates of
rise.
rise to what is called “displacement current” in addition to
normal conduc-tion current. In other words, the nature of
this displacement current is dependent on polariza-tion of
the dielectric material which in turn depends on the
condition of the material under consideration.

A polarization current measurement is usually stopped if the

Separation limit for lightning arrestors
Best protection is obtained for terminal equipment by placing
the arrestor as near as possible to that equipment. When
arrestors must be separated from the protected equipment,
additional voltage components are introduced, which add
instant by instant to the discharge voltage. The maximum
voltage at the terminal of a line as a result of the first
reflection of a travelling wave may be expressed
mathematically as
Ea is the discharge voltage at the arrestor, and de/dt is the
rate of rise of the wavefront. When the value of $ is not
known, it may generally be assumed as equal to 1 without
much loss of accuracy. Figure 10.12 shows how the voltage
1.2 Basic Theory of Time Domain Dielectric Response current becomes either stable due to 0 or becomes very
low. If the test object is now short-circuited at t = tc, the
Measurement
depolari-zation or discharging current flows as idepol(t).
For proper utilization of time-domain dielectric response
According to the superposition principle, sudden reduction of
measurement, one should under-stand the physics behind
the voltage U0 to zero can be regarded as a negative step
the processes. However, the physical mechanism of dielectric
voltage at time t = tc . Then we get for t (t0 + tc)
re-sponse of oil-paper insulation of transformer is very
complex, as it includes moisture migra-tion through cellulose
depending on moisture potential and moisture equilibrium
within transformer, adsorption of water vapour in cellulose
material and mechanisms of bubbles evolution. Any dielectric
material, under the influence of an electric field, results in a
relative shift of positive and negative charges in the material.
This phenomenon is called dielectric polariza-tion which gives

Time Domain Response Measurements

During recovery voltage measurement, the current density is
zero as this voltage needs to be measured by open circuit
condition. Thus equation 1.4 can be solved with j(t) = 0. If the
con-ductivity 0, permittivity 0 are known, then the
recovery voltage depends on the response function f(t,)
which in turn depends on the characteristics of the dielectric
material. Hence recovery voltage measurement is another
important parameter in time domain dielectric spec-
troscopy.
Polarization and depolarization current measurements
Two resistors of this circuit represent the protection
resistances, which will not influence the recorded currents.
The sample under test has two electrodes marked HV and LV.
LV is per-manently connected with electrometer, a type of
precision ammeter (and also voltmeter on requirement).
During charging or polarization, the switch S connects at
position 2 due to which the polarization current ipol(t) flows Concept of RV spectra and the central time constant
under the influence of the applied step voltage U0. Recovery
According to the methodology of RV measurement, as stated
Voltage Measurement
in Section 1.3.2, the charging time is increased gradually with
As per the basic theory of recovery voltage measurement, the corresponding increase in discharging time from a small
the charging (polarization) and dis-charging (depolarization) initial value and a different peak value of recovery voltage,
process can be performed using the same circuit as that of Vrecovery(peak) is obtained in each step. Peak values
Fig. 1.3. The difference is in the time where the discharging obtained for different charging times (tch) can be plotted to
time is much less than the charging time according to the yield a Vrecov-ery(peak) vs. tch curve, which is called
requirement. A typical case may be discharging time ( recovery voltage spectrum. This recovery voltage spec-trum
tdischarge) = half of charging time (i.e. tch/2). To measure is also analyzed for condition monitoring. The charging time
the voltage, the electrometer should be able to change its corresponding to the peak
mode from ammeter to voltmeter at the proper instant of
time.
Time Domain Response of Transformer Insulation insulation by preparing oil impregnated pressboard samples
with accelerated aging. Due to aging, the slope of the initial
In electric utilities around the world, significant proportion of
part of both polarization and depolarization current becomes
power transformers are being operating beyond their design
lower compared to unaged insu-lation. One of the
life. Sudden failure of such transformers operating at a
explanations may be that, due to aging, the migration of
pivotal position may cause total power disruption resulting in
interfacial charges becomes sluggish through the oil and
substantial loss of money and time. At present, emphases are
consequently, the overall rate of polarization becomes
given on non-invasive methods for condition assessment of
slower. This results in lower slope in the obtained current. An
oil-paper insulation of power transformers. As explained in
approximate assessment of the dielectric condition can be
the aforesaid sections, the methodology of PDC and RVM is
made by visualization of the curves. The closer the curve
best suited in assessment of condition in transformer
peak (CTC) is towards the small time constant, the worse is
insulation by time domain response measurement in
the condition of the dielectric. It may be noted here that, the
dielectric spectroscopy. Henceforth, in the subsequent sec-
degeneration of cellulose insulation produce water as an end
tions, results of PDC and RVM will be dealt with composite
product of the chemical reaction, which contributes to the
oil-paper insulation as dielectric under test. Before
moisture content of the cellulose insulation (mainly paper).
performing the test, HV and LV is short circuited for a sub-
Hence, the paper moisture content reflects the aging
stantial period of time such that the test object is completely
condition of insulation.
“discharged”. Otherwise some polarization will exist in the
insulation that may adversely affect the obtained results. To
prove the effectiveness of the PDC and RVM methodologies,
transformer models are con-structed in the laboratory using
CIGRE standards. Various parameters like temperature, oil
conductivity, moisture content in paper, artificial aging effect
are varied in the model for a comprehensive study of the
phenomena involved.
Effect of Insulation Aging
Nature of polarization and depolarization current varies
significantly with overall aging of the oil-paper insulation. J. L.
Wei et al studied the aging characteristics of oil-paper

10.3 Impulse Voltage Withstand Test
10.3.1 The insulator shall be tested dry under the
conditions prescribed in Appendix B.
10.3.2 The impulse generator shall be adjusted to
produce a standard 1.2/50 impulse wave of peak value
equal to the specified value of the impulse withstand
voltage corrected for atmospheric conditions in
accordance with Appendix A.
10.3.3 Five such impulse voltage waves shall be applied
to the insulator. If there is no flashover or puncture, the
insulator shall be considered to have passed the test. If
during the application of these five waves puncture
occurs The insulator shall be considered to have passed
this test only if during this new series of tests there is no
flashover or puncture.
10.3.4 The insulator shall be capable of passing the
impulse voltage withstand' test with voltages of both
positive and negative polarity. However, when it is
evident which polarity will give the lower breakdown
voltage, it shall suffice to test with that polarity.
10.3.5.1 The impulse flashover voltages to be recorded shall
be the positive and negative 50 percent impulse flashover
voltage as measured above and corrected in accordance with
Appendix A.
10.3.5.2 The insulator shall not be damaged by these tests, but
slight marks on the surface of the insulating parts or chipping
of the cement or other material used for assembly shall be
permitted
2.9 Fifty-Percent Dry Impulse Flashover Voltage —
The impulse voltage which, under the conditions
prescribed in 10.3 has a 50 percent probability of
producing a flashover on the insulator.
2.10 Wet Power-Frequency Withstand Voltage — The
specified power-frequency voltage which the insulator
shall withstand (wet) under the conditions prescribed in
10.4 for the specified time (one minute) without
flashover or puncture.
2.11 Wet Power-Frequency Flashover Voltage — The
arithmetic mean value of the measured power frequency
voltages which cause flashover of the insulator under
the conditions prescribed in 10.4.
10.4 Wet Power-Frequency Voltage Withstand Test
10.4.1 The insulator shall be arranged as prescribed in
Appendix B.
10.4.2 Before the commencement of the test, the insulator
shall be exposed to the artificial rain produced in accordance
with 9.1 of IS : 2071 ( Part I )-1993* for at least one minute
before application of voltage and then throughout the test.
10.4.3 The test voltage to be applied to the insulator shall be
the specified value of the wet power-frequency withstand
voltage adjusted for atmospheric conditions at the time of
test (see Appendix A).
10.4.4 A voltage of about 75 percent of the test voltage as
determined in 10.4.3 shall be applied and then increased
gradually to reach the test voltage in a time not less than five
seconds. The test voltage shall be maintained at this value for
one minute [see IS : 2071 ( Part II )-1974†]. The insulator shall
not flashover or puncture during the application of the test
voltage.
10.4.5 To provide information when specially requested, the
wet flashover voltage of the insulator may be determined by
increasing the voltage gradually from about 75 percent of the
wet withstand voltage to reach the flashover voltage in not
less than five seconds. The flashover voltage shall be the
arithmetic mean of five consecutive readings and the value
after correction to standard atmospheric conditions (see
Appendix A) shall be recorded.
9. Wet tests
The preferred wet test procedure, described in 9.1. is
intended to simulate the effect of natural rain on external
insulation and is a revision of earlier test methods. It is
recommended for tests with all types of test voltages and on
all types of apparatus, but either of the alternative test
methods given below are permitted if specified by the
relevant Technical Committee.
Two earlier test methods. not intended to simulate natural
rain, are described in 9.2. They have been in use for many
years for tests with alternating voltages on apparatus having
Um up to 420 kV and many test data obtained by these
methods exist.
For a.c. apparatus of large dimensions, such as those having
Um higher than 800 kV, no appropriate wet test procedure is
available at present.
The relevant Technical Committee shall specify the
arrangement of the test object during the test procedure.
9.1 Standard wet test procedure
The test object shall be sprayed with water of prescribed
resistivity and temperature (see table 1) falling on it as
droplets (avoiding fog and mist) and directed so that the
vertical and horizontal components of the spray intensity are
approximately equal. These intensities are measured with a
divided collecting vessel having openings of 100 cm* to 750
cm*, one horizontal and one vertical, the vertical opening
facing the spray.
The position of the test object relative to the vertical and
horizontal rain components shall be specified
by the relevant Technical Committee.
In general, the reproducibility of wet test results is less than
that for other high voltage discharge or withstand tests. To
minimize the dispersion the following precautions shall be
taken:
B-2. METHODS OF SUPPORT
B-2.1 String Insulator Unit and Insulator String
B-2.1.1 The string insulator unit or insulator string shall be
suspended vertically by means of an earthed wire rope or
other suitable conductor from a supporting structure. The
distance between the uppermost Point of the insulator metal
work and the supporting structure shall be not less than 1
metreB-2.1.2 The diameter of the conductor shall be about
1.5 percent of the length of the string insulator unit or
insulator string with a minimum of 25 mm.
B-2.1.3 The length of the conductor shall be at least 1.5 times
that of the string insulator unit or insulator string, and it shall
extend at least 1 metre on each side of the vertical axis.
B-2.1.4 Precautions shall be taken to avoid flashover from the
ends of the conductor.B-2.1.5 The test voltage shall be
applied between the conductor and earth.B-2.2 Rigid
InsulatorB-2.2.1 The insulator shall be mounted on an
earthed metallic vertical pin, with a smooth surface and a
diameter not less than 16 mm.
B-2.2.3 If the rigid insulator is provided with means of
clamping, conductor shall be placed in the clamp.
B-2.2.4 The test voltage shall be applied between the
conductor and earth. NOTE — When to agreed, tests may
also be made under conditions reproducing service
conditions as closely as possible, taking account of all factors
which may influence flashover voltage.