High Voltage Products, Switchgear

High-Voltage On-Site Testing

Field Testing for Switchgear Integrity

The quality of our products is es-

tablished by design, type testing
and routine testing. However, to
ensure that the complete gas
insulated switchgear has the di-
electric integrity and to detect
potential irregularities, it is tested
after the installation.
Introduction
This paper explains the purpose and methods for high-voltage onsite
testing of gas-insulated metal-enclosed switchgear.
Recommendations are based on: Carrying out such site tests is not always practicable and deviations
from the standard tests may be accepted.
• IEC 62271-203 [1]
• Experience gained on GIS plants during more than 50 years in all
Since the aim of these tests is a final check before energizing, it is
voltage levels
very important that the chosen test procedure does not jeopardize
• Technical, economical and practical aspects of field testing
sound parts of the switchgear.

Purpose of on-site testing The appropriate test method and procedure has to be adjusted to
each individual case and special agreements may be necessary in
The purpose of on-site testing (field testing) is well defined in [1] as
the interest of practicability and economy. Points to consider are
summarized below:
for example the electrical power requirements, the dimensions, the
weight and the interface of the high-voltage test equipment.
After installation, and before being put into service, the gas-insulated
switchgear shall be tested with the aim of checking the dielectric
A detailed test program for the dielectric tests on site, as recom-
integrity of the completed installation and of detecting irregularities.
mended in this paper, shall be agreed between customer and
Such irregularities might include wrong fastening, damage during
manufacturer.
handling, transportation, storage and installation, presence of foreign
bodies, etc. and might in the future give rise to an internal fault.
Normally the dielectric test shall be made after the gas-insulated
switchgear has been fully installed and gas-filled at the rated filling
density preferably at the end of all site tests.

Having this specific purpose, these tests do not replace the type
tests or the routine tests carried out on gas-insulated switchgear
transport units in the factory. They are supplementary to the dielec-
tric routine tests.
Test arrangement Test procedures
Before testing, the gas-insulated switchgear shall be installed One of the following test procedures are recommended by
com-pletely and gas-filled at its rated filling density. IEC62271-203:

• PROCEDURE “A” (recommended for the rated voltage of 170

Some parts shall be disconnected for the test, either because of their
kVand below)
high charging current or because of their effect on voltage limitation,
– Power-frequency voltage test for a duration of 1 min
such as:
• High-voltage cables and overhead lines • PROCEDURE “B” (recommended for the rated voltage of 245
• Power transformers kVand above)
• Surge arresters – Power-frequency voltage test for a duration of 1 min
– Partial discharge measurements
The inductive voltage transformer can be tested together with the
• PROCEDURE “C” (recommended for the rated voltage of 245kV
gas-insulated switchgear. For this purpose a minimum permissible
and above, alternative to procedure “B”)
testing frequency is given by the voltage transformer suppliers. Only
– Power-frequency voltage test for a duration of 1 min
in case that the minimum testing frequency cannot be fulfilled, the
– Lightning impulse tests with three impulses of each polarity
voltage transformer has to be disconnected in order to avoid core
saturation and damage Manufacturer recommends using test procedures “A” or “B” as
shown on page 7 in figure 3 “Procedures for on-site testing accord-
The choice of the gas-insulated switchgear section to be tested and ing IEC 62271-203 as recommended from manufacturer”.
connection of the test source should take the following factors into
account:
Test voltages
• Facility of localizing electric discharges
The test voltages for dielectric tests on site shall be according to Ta-
• Limitation of the capacitive load on the test voltage source
ble 7 of IEC 62271-203, as shown on table 1 in section References
• Limitation of the disruptive energy in case of a flashover
[page 4].
• Limitation of the overvoltage amplitude due to reflections in case
of a flashover
Short-duration power-frequency with-
For these reasons it is often convenient to divide the whole instal-
stand tests
lation of gas-insulated metal enclosed switchgear into sections
byopening switching devices. Power-frequency voltage tests are especially sensitive in detecting
contaminations (for example conducting particles), and are, in most
The sections which, in such cases, are not being tested, and which cases, also sufficient in detecting abnormal dielectric field configu-
are isolated by a circuit-breaker or a disconnector from the section rations.
under test, shall be earthed. Unless dismantled after routine test, no Normally a resonant AC generator (see fig. 1) will be used for this
dielectric test across the open switching devices need to be carried test, with operating frequency of 10 to 300 Hz. The natural frequency
out on site. is determined by the inductance of the high-voltage reactor (L) and
the capacitance of the respective gas-insulated switchgear section
In the case of extensions, the preferred procedure is to perform the (CP) and is calculated by the following equation:
high-voltage test before connecting the extension to the existing
gas-insulated switchgear. If the coupling is done before the high
voltage test, the adjacent existing part of the switchgear must be
de-energized and earthed during the dielectric test.

After minor repair or maintenance work, for technical or practical

reasons, dielectric tests on-site may be carried out with reduced
voltage values (for example with service voltage, applied through
the largest possible impedance in order to reduce damage caused
by a possible disruptive discharge; the test period should be at least
*
30min). With very small installations (below 2000 pF) or for plant extensions
a power frequency operated test transformer may be used, however
the possible saturation of connected voltage transformers should be
considered.
Fig. 1: Example of resonant frequency circuit for high-voltage on-site testing

ITr Insulating transformer FC Frequency converter

ETr Exciter transformer L Reactor unit
D Damping resistor C1 HV capacitor
C2 LV Capacitor CP Test object / gas-insulated switchgear
I Resonant circuit current Vm AC peak voltmeter

Impulse voltage tests - Oscillating light- Test voltage should be 80 percent of switching insulation level (SIL)
as shown in table 1 in section References [page 4].
ning impulse (OLI)
Switching impulse testing should only be chosen if AC voltage or
Tests with lightning impulse voltages are effective for detection of lightning impulse tests cannot be applied.
sharp protrusions and abnormal field configurations (for example
damaged electrodes). DC voltage tests
Based on the existing experience oscillating lightning impulse voltag- Tests with DC voltages are not recommended. The voltage distribu-
es with front times up to about 15 μs are suitable [1]. tion across the insulators differs from that under power-frequency
However, it must be emphasized, that in case of a flashover there voltage stress and does not represent the condition in service cor-
may be reflections resulting in significant over voltages and subse- rectly.
quent dangerous secondary flashovers. Therefore, the amplitude Manufacturer therefore only allows very restricted use of DC test
chosen for the lightning impulse must be sufficiently low in order to voltages, for example when insulators are unavoidably involved
avoid secondary flashovers at all times; maximum 90 percent of BIL during testing of joined cables.
(basic insulation level).

Normally, tests with lightning impulse voltages should be selected

Partial discharge detection
only when an AC voltage test with partial discharge measurement is Partial discharge measurements combined with AC testing may be
not applicable. helpful in detecting certain kinds of faults like voids in insulating
materials, damages of shield electrodes and presence of particles.
Oscillating switching impulse (OSI) Partial discharge measurements are therefore a useful complement
to dielectric tests on site but are often difficult to perform because of
Tests with switching impulse voltages are useful to detect the pres-
ambient disturbances.
ence of contaminations as well as abnormal field configurations with
relatively simple test equipment, but they are not as sensitive as AC Partial discharge measurements on-site performed according to
voltage tests. IEC 60270 (conventional method) are not recommended for high
voltage gas-insulated switchgear as this method is especially sensi-
Based on existing experience, switching impulses with oscillating
tive to electrical disturbances and it is difficult to keep the noise level
waveforms and with a time to crest in the range of 150 μs to 10 ms
below 5 pC as required by IEC.
are suitable.
Instead, electrical VHF/UHF or acoustic methods are recommended Therefore, partial discharge detection is normally performed on site
for partial discharge detection on site. Both methods are less sen- by applying UHF narrow band measurements (see figure 2) com-
sitive to noise than the conventional method. However, with both bined with acoustic partial discharge detection for the localization of
methods calibration is not possible as the sensitivity depends on the partial discharge sources. Suitable UHF sensors are integrated into
distance between the defect and the sensor. Instead a sensitivity the earthing switches and, where needed, additional UHF sensors
check must be performed on site (IEC 62271-203, Annex C.7.5), veri- are installed in busduct elements.
fying that any defect causing 5 pC or more will be detected [1,5].

Fig. 2: Schema of partial discharge measurement with UHF Narrow Band Method

Assessment of the test References

The switchgear shall be considered to have passed the test if each • [1] IEC 62271-203, High-voltage switchgear and control gear –
section has withstood the specified test voltage without any disrup- Gas-insulated metal-enclosed switchgear for rated voltages
tive discharge. above 52 kV
• [2] IEC 60060-3, High-voltage test techniques –Part 3: Definition
In the event of a disruptive discharge occurring during di-electric
and requirements for on-site testing
tests on site, a repetition test should be done. Recommendations
• [3] 2GHV074551, General Technical Information, Partial dis-
are given in Annex C.6 of IEC 62271-203.
charge concepts for GIS type ELK
In the event that this repetition test is successfully withstood, it is • [4] 2GHV075119, General Technical Information, On-site PD
assumed that the disruptive discharge was self-restoring. de-tection using UHF techniques
• [5] CIGRE WG D1.25 TB 654; UHF Partial Discharge Detection
However, if further disruptive discharges occur at considerably lower
Systems for GIS: Application Guide for Sensitivity Verification;
voltage levels, the solid insulation involved has to be inspected and
April 2016
replaced (if necessary) and the test shall be repeated.
PROCEDURE “A”
AC voltage test
Uds Test voltage
(for 170 kV and below)

Conditioning

(15 to 20 min)

1 min

Voltage Duration (min)

PROCEDURE “B”
Test voltage
AC voltage test + PD measurement Uds
(for 245 kV and above) Conditioning
(15 to 20 min)

PD measurement

1 min
Voltage

Fig. 3: Procedures for on-site testing according IEC 62271-203 as recommended from manufacturer Duration (min)

Rated voltage for On-site short duration power On-site switching impulse On-site lightning impulse
equipment frequency withstand voltage withstand voltage withstand voltage

Ur Uds Uss Ups

kV (r.m.s.) kV (r.m.s. value) kV (peak value) kV (peak value)

(1) (2) (see note 1) (3) (4)

72.5 120 - 260

100 165 - 360

123 200 - 440

145 235 - 520

170 270 - 600

245 380 - 840

300 380 680 840

362 425 760 940

420 515 840 1140

550 550 940 1240

800 760 1140 1680

Table 1: On-site test voltages according Table 7 of IEC 62271-203 and as recommended from Manufacturer

NOTE 1: Values of column (2) are only applicable for SF6 insulation or when SF6 is a major part of the gas mixture. For other insulation refer to tables 1 and 2 of IEC 62271-1, applying a factor 0,8 on column (2).
NOTE 2: The on-site test voltages have been calculated as follows: Uds (on-site test value) = Up × 0,45 × 0,8 (column (2)) Uss (on-site test value) = Us × 0,8 (column (3)). All values have been rounded
up to the next higher modus 5 kV. All values have been rounded up to the next higher modus 5 kV.
NOTE 3: If other insulation levels than the preferred values of tables 2 and 3 (example the lower insulation levels of tables 1 and 2 in IEC 62271-1) are specified,
Then the on-site test voltage should be calculated according to note 2.

Hitachi Energy Ltd We reserve the right to make technical chang- We reserve all rights in this document and in
Service, High Voltage Products es or modify the contents of this document the subject matter and illustrations contained
Brown Boveri Strasse 5 without prior notice. With regard to purchase therein. Any reproduction, disclosure to third
CH-8050 Zurich orders, the agreed particulars shall prevail. parties or utilization of its contents – in whole
Switzerland Hitachi Energy Ltd. does not accept any or in parts – is forbidden without prior written
responsibility whatsoever for potential errors or consent of Hitachi Energy Ltd.
possible lack of information in this document.

hitachienergy.com