Partial discharge measurements and IEC standards:
Justification of the use for their inclusion in
afterlaying test for extruded cable systems
Mirza Batalović, Dejan Bešlija, Mirsad Kapetanović
Myoung-Hoo Kim, Kyong-Hoe Kim
Faculty of Electrical Engineering
Department of Power Engineering
Sarajevo, Bosnia and Herzegovina
E-mail: [email protected]
ILJIN Electric Co. Ltd
Suwon, South Korea
E-mail:[email protected]
Abstract—Partial discharges, as their name states, only
partially bridge a small portion of electrical insulation in the form
of a tiny electrical arcs, which burn inside the defects that could
appear in insulation system. Because of the fact that extruded
cable system insulation is very sensitive on partial discharge
activities detection vise, partial discharge measurements could be
used as a powerful diagnostic tool in evaluating the actual
condition of cable system through measuring procedures during
afterlaying tests. If such procedures would be included in
standards, they would provide an effective way to identify and
detect the defects that might appear during the cable system
installation and to forestall their appearance during exploitation,
ultimately reducing the probability of failure. Very first aim of this
paper is to address some shortcomings of current IEC standards
related to analyses of cable systems with polymer insulation (IEC
60840 and IEC 62067). In order to justify these statements, a
review of a recent alignment between IEC 60840 and IEC 62067,
simulation support, using the contemporary software tool
(COMSOL Mph), backed up with experimental results for two
artificially induced defects in cable accessories, are provided in
this paper.
Keywords—Partial discharges; IEC standards; Afterlaying
test
I. INTRODUCTION
The population growth in urban areas represents a direct and
consequent relation with growth of energy consumption, which
is the reason why the use of power cables is nowadays steadily
increasing. Another reason for the power cable implementation
enhancement lays in the fact that a large number of subsistent
cable networks has almost reached the end of their life cycle and
therefore must be replaced [1].
In order to ensure a greater level of reliability of correct
operation and lower life cycle costs, various tests for medium
and high voltage cables are performed [2]. The implementation
of these tests starts from the early design phase of the cable/cable
accessory, up to the installation phase. The installation phase
covers prequalification tests, type and routine tests, as well as a
test after installation, also known as an afterlaying test.
In current international standards that cover after-installation
tests of extruded cable systems: IEC 60840:2011 (fourth edition,
for cables of rated voltages from 30 kV (Um = 36 kV) up to 150
DELING Tuzla – The smart way to energy – Designing Engineering
Production Assembly
kV (Um =170 kV) [3] and IEC 62067:2011 (second edition, for
rated voltages above 150 kV up to 500 kV (Um = 550 kV) [4],
partial discharge (PD) measurements are actually not required,
even though they are recommended by appropriate IEEE and
CIGRE documents [5,6].
The research proposed inside the frames of this work through
a simulation of two types of defects in cable accessories,
resulting from bad design and improper handling, backed up
with experimental results, have the main goal to present the
benefits of PD testing and use of PD measurements in
afterlaying test for extruded cable systems.
II. RELEVANT STANDARDS
During the last revision, the above mentioned IEC standards
have been aligned. The following text gives an overview of the
issues covered by these standards and latest adjustments [2].
A. IEC 60840
This international standard considers extruded cable systems
and their accessories concerning voltages in the range from 30
to 150 kV. It describes the various tests to be performed for
routine, sample and type tests [3]. Latest alignment between IEC
60840 and IEC 62067 lies in the fact that some tests that used to
exist only in IEC 62067 have been introduced in IEC 60840.
Those are prequalification tests intended for cable designs with
high electrical field stressing. Also, cables/cable accessories,
which do not pass these procedures regarding high-field stress,
can now only be tested on the basis of a system approach. On
the other hand, cables/cable accessories, which passed this highstress testing, can still be type tested. Type tests for both these
categories, high-stressed cable systems and normal-stressed
cable systems and accessories, consist of checking of insulation
thickness, a bending test, semi-conductive resistivity screen
measurements, PD tests, tanδ measurements, heating cycles
under voltage application, impulse voltage tests and an AC
voltage withstand test. The non-electrical type tests are
considering material characteristics for various material types
used in cable systems.
B. IEC 62067
This international standard treats extruded cable systems and
describes the various tests to be performed in routine, sample
and type tests [4]. IEC 62067 has been aligned with 60840. The
only contrast lies in the fact that IEC 60840 differentiates
between high-stressed and normal-stressed cable designs, while
this standard recognizes only high-stressed cable designs and for
that reason exclusively accepts testing on a system basis. The
electrical type tests covered by this standard are in line with the
corresponding ones in IEC 60840. Also, a pre-qualification test
is mandatory in addition to type tests. This test is needed for a
satisfactory long term performance of cable systems
(cables/cable accessories).
Regarding type test for both IEC 60840 and IEC 62067,
alignment of the shape of suitable AC test voltage and its
duration time has been accomplished [7]:

substantially sinusoidal waveform;

frequency between 20 and 300 Hz;

time of voltage application equal to 1 hour.
While IEC 60840 and IEC 62067 provide basic guidance on
the waveform, frequency and defined voltage to be applied
during the overvoltage test, there is still no standard defined
procedure for PD measurements concerning after-installation
tests. [8].
III. THE BENEFITS OF ON-SITE PD MEASUREMENTS
Even though each single cable/cable accessory has passed
the procedure of typical and routine tests in the factory, and is
then considered to be "healthy", free from any kind of defects,
there is still the risk of defects that could appear as a result of
damages generated during transport, storing and installation, or
damages that result from imperfection of an on-site cable
accessory assembling process [9-13]. For all these reasons
appropriate tests are being performed on cable systems after
their installation [3-5, 14]. Due to the fact that the polymer cable
insulation is very sensitive on PD activities, after-installation
tests of the insulation can focus on defects in cable accessories,
e.g. interfacial problems, improper positioning of the joint, cuts
or scratches, contaminations in the form of foreign conductive
particles, dust etc. Usually this type of defects does not lead to
breakdown within testing time, which instead leads to a greater
risk of breakdown later in service. In order to reduce the risk,
sensitive on-site PD measurements would be desirable [15-17].
In respect of entire power system networks, from the data
provided from investigations in some countries [9], it is obvious
that the distribution network is responsible for the great
percentage of outages in power supply. The majority of time
during which customers are without supply relates to failures of
medium voltage cables, more precise on the cable accessory
(cable joints and terminations). Statistics show that more than 60
% of breakdowns of the cable system network are caused by
internal defects appearing inside the insulation of power cable
network [9]. Accordingly, two typical insulation defects,
resulting from poor workmanship during the installation and
assembling process of cable accessories, are simulated:

Missing of a part of the semi-conductive screen in the
cable joint - Defect 1;

Electrode bounded void - Defect 2.
The main purpose of these simulations is to show the
conditions in the electrical insulation by comparing the defect
and "healthy" cable state in regard of electrical field stresses.
IV. ELECTRIC FIELD SIMULATION AND LABORATORY
TESTING RESULTS
A. Missing of a part of semi-conductive screen in the cable
joint - Defect 1
The defect in the form of a missing part of the semiconductive screen is illustrated in Fig.1. The missing part of the
semi-conductive screen inside the cable joint results in an
electric field increase on the edge of the removed semiconductive screen [6]. This causes surface discharges that occur
at the edge of the outer semi-conductive screen and across the
dielectric surface made of XLPE (Cross Linked Polyethylene).
Surface discharges present on the interface air/XLPE could lead
to degradation of XLPE. This process is associated with
chemical changes of the polymer surface resulting with
formation of crystals on the polymer insulation [9]. In the area
of crystal clusters, higher electric field strength results in higher
PD activity. After a certain period of time, the treeing process
would be triggered and such activity would consequently lead
to erosion of the insulating material surface.
1) Simulation results
The simulation tool used to address the conditions regarding
the presence of defects inside the electric insulation and its
comparison with a "healthy" cable, free from any kind of
defects, was COMSOL Mph version 4.3. The model was
prepared using the AC/DC module, covering electrostatic
issues and stationary problems. The governing equation of the
used simulation model is:
𝑑𝑖𝑣𝑫 = 𝜌
(1)
where D is the electric displacement field and ρ is the free charge
density. If non-dispersive, linear and isotropic dielectric
insulation is assumed, exposed to slowly varying fields, (1) can
be written in the following form:
−∇dε0 εr ∇𝑉 = 𝑑𝜌
(2)
where V is the applied electric potential, d represents the
thickness of the sample, and ε0 and εr are dielectric constants of
vacuum and the used insulating material.
The cross-section of the cable/cable accessory part is modeled
in the RZ plane, which means that the cross-section of the
structure is rotating around an axis of symmetry, Fig.2.
Fig. 1. Schematic view of the missing part of the semi-conductive layer next
to the cable joint.
Fig. 4. Electric field distribution for the cable with Defect 1, V0 = 35 kV
Graph A in Fig. 5 represents the electric field strength for the
"healthy" cable, in the direction of the potential defect site. It
can be seen that the electric field magnitude reaches the value
of 3.6 MV/m at potential defect site, while in the case of a defect
that occurred in the same direction, the electric field magnitude
increases by approx. 1.2 MV/m, as shown in Fig. 6. The
epilogue of these conditions is higher electrical stress at a
location with weaker insulating strength, which could trigger
PD activity.
Fig. 2. Cross-section of the cable accessory with presence of Defect 1
For the purpose of comparison of conditions regarding
electrical field stresses inside the insulation, the electrical field
strength distribution for the "healthy" cable and the one with a
defect in the form of a missing part of the semi-conductive
layer, are shown in Fig. 3 and Fig. 4, respectively. It is clear
that the insulation is more stressed in the region of the defect
site compared with the "healthy" cable insulation.
2) Experimental testing results
Laboratory testing results in case of Defect 1 are given in
Fig. 7. As can be noticed, for this artificially induced defect, the
PD inception voltage equals 22 kV. At this value of applied
voltage, the basic level of PD activity is always present and it
does not change significantly with voltage rise. This type of
defect allows to be tested up to the value of 2U0, when
compared to insulation with certain types of defects, which
experience breakdown before the voltage could be raised up to
the maximum testing value.
B. Electrode bounded void - Defect 2
Fig. 3. Electric field distribution for the “healthy” cable, V0 = 35 kV
The presence of gas-filled voids in the form of spherical or
elliptical shapes is frequent [6]. By subjecting the insulation
system to the voltage source, voids become more electrically
stressed than the surrounding insulating media. The reason for
that lies in the fact that gas has a lower dielectric constant than
the surrounding insulation material. Also, shape and location of
the void have an impact on the field enhancement factor inside
the void. Accordingly, there are two types of voids: dielectric
bounded cavity with two dielectric walls, as shown in Fig. 8a,
and electrode bounded cavity with one dielectric and one
electrode wall, Fig. 8b. An electrode bounded void can appear
in the area between cable insulation and the semi-conductive
screen. This type of void is orriented in the direction of the
tangential electric field component and also produces a
noticeable field concentration after breakdown occurs.
The presence of this type of defect in the insulation can result
in total insulation breakdown due to degradation of insulating
material, which depends on electric field strength, type of
insulating material and PD magnitude [18].
1) Simulation results
According to previous investigations [19], Paschen’s curve
is completely sufficient for determining of dielectric strength of
small voids in solid insulation, as a function of the product of
distance and pressure. According to Paschen’s curve, the
dielectric strength of a gas-filled void equals to 3 kV/mm or
peak value of 4.24 kV/mm. For that reason, it is important to
know the electric field distribution inside the void, or in the
direction of the defect site, in order to confirm the electric field
strength increase.
The cross-section of the cable/cable accessory part for this
type of defect is modeled analogue to Defect 1, Fig. 9. Even in
case of such small dimensions of a defect, an increase of electric
field stress is present during normal operating conditions in the
direction of the defect site.
Fig. 7. Results from PD measurements for Defect 1
Fig. 8 a – dielectric bounded void; b – electrode bounded void
Fig. 10 depicts visually these conditions, while Fig. 11
graphically confirms it. As can be noticed, in the direction of the
potential site for this type of defect, the electrical stress was 2.9
MV/m, Graph B in Fig. 5. In this case, the electrical stress in the
direction of the defect site approaches 4.9 MV/m. That means
an increase of the electric field by the amount of 2 MV/m inside
the weak spot of insulation.
Fig. 5. Electric field distribution in a potential defect site along referent line,
Defect 1 - blue line, Defect 2 - dashed green line, V0 = 35 kV
Fig. 9. Cross-section of the cable accessory with presence of Defect 2
Fig. 6. Electric field distribution in the direction of the Defect 1 site along the
referent line – black dashed line, Fig.4, V0= 35 kV
Fig. 10. Electric field distribution for Defect 2, V0 = 35 kV
Fig. 11. Electric field distribution in the direction of the Defect 2 site along the
referent line – black line, Fig. 10, V0= 35 kV
2) Experimental testing results
Fig. 12 shows the results of PD measurements for Defect 2
in the form of an electrode bounded void. It can be noticed, that
for this type of defect, the PD magnitude of 10 pC when only 5
kV of the testing voltage 35 kV is applied. The base value of
the PD magnitude does not fall below 1000 pC. It must be
emphasized that the voids are very small in volume, with a large
pressure building up inside them in a short period of time,
which is the reason why there are discharges in the amount of
1000 pC at only 7 kV of the applied voltage. If the time
distribution of these PD pulses in relation to the applied
sinusoidal voltage signal is observed, Fig. 13, it is obvious that
there are more PD activities detected during the positive period
of the sinusoidal voltage wave. The reason for that lies in the
fact that once PD occurs inside a void, it becomes ionized and
for the next discharge to occur, less time and a lower value of
inception voltage is needed.
Fig. 12. Results from PD measurements for Defect 2
Fig.13. Visual display of the time distribution of generated PD pulses during
laboratory testing for Defect 2
It should be emphasized that in practice small voids represent a
common occurrence (manifestation).
V. CONCLUSIONS
According to the performed investigation and results
presented inside the frames of this paper the following
conclusions are drawn:
1) It is known that certain types of defects can occur inside
the insulation of cables/cable accessories during the
process of storage, transport and poor on-site
workmanship. Consequently, PD can occur inside or in the
direction of the defect site. PD's cause insulation
degradation and early failures of the cable system. These
defects could be detected if the cable system would be
tested on PDs in an afterlaying test. Thereby, more
information about the actual condition of new installed
cable systems would be provided and cable network
managers would be able to give more precise estimation of
the reliability of power cable networks. International
standards IEC 60840 and IEC 62067 still do not define
procedures for PD measurements during afterlaying test.
2) Two typical installation defects are investigated in this
paper. The investigation is performed using simulation and
laboratory testing. The results of the numerical calculations
for the electric field strength for both models, undoubtedly
point out that analyzed defects cause increased electric
stresses inside the cable system insulation. This fact is
confirmed by PD measurements in laboratory tests
performed on models with analogue artificially induced
defects.
[6]
Future work on this issue should involve further study of
these types of defects regarding their dimensions, as well as
more different types of defects, in order to obtain more
information and knowledge about their nature.
The main intention of the investigation presented in this
paper was to emphasize the benefits of PD testing and
especially the use of PD measurements in afterlaying tests for
extruded cable systems.
[11]
REFERENCES
[14]
[1]
[2]
[3]
[4]
[5]
M. Batalović, K. Sokolija, M. Hadžialić, N. Batalović, “Partial discharges
and IEC standards 60840 and 62067: simulation support to encourage
changes,” Scientific Journal Technical Gazette, vol.23 No.2, pp. 589-598,
April 2016.
E. Pultrum, H. He, R. Gruntjes, “Type testing cable and accessories, a
must,” CIRED 23rd International conference on Electricity Distribution,
paper 0303, Lyon, 15-18 June 2015.
IEC 60840, Standard Power cables with extruded insulation and their
accessories for rated voltages above 30 kV (Um = 36 kV) up to 150 kV
(Um = 170kV) -Test methods and requirements, fourth edition, 2011.
IEC 62067, Standard Power cables with extruded insulation and their
accessories for rated voltages above 150 kV (Um = 170 kV) up to 500kV
(Um = 550 kV) - Test methods and requirements, second edition, 2011.
IEC 60060-3 High Voltage test techniques –Part 3: Definitions and
requirements for on-site testing
[7]
[8]
[9]
[10]
[12]
[13]
[15]
[16]
[17]
[18]
[19]
J. Setyawan:“Investigation of partial Discharge Occurance and
Defectability in High Voltage Power Cable Accessories“, MSc
Graduation thesis, Delft University of Technology, November 2009.
U. Herrmann, A. Kluge, R. Plath, “After-Installation Testing of HV/EHV
Extruded Cable Systems-Procedures and Experiences,” Jicable 07 7th
International conference on Power Insulated Cables, paper A104,
Versailles, 24-28 June, 2007.
INMR “Innovations in Type & Commissioning Testing of High Voltage
Cables”, Technical Journal, January 30, 2016.
F.J. Wester, “Condition Assessment of Power Cables Using PD
Diagnosis at Damped AC Voltages”, ISBN 90-8559-019-1, Ph.D. thesis
TU Delft, 2004.
N. Ahmed and N. Srinivas, “On-line Partial Discharge Detection in
Cables”, IEEE Trans. Dielectr. Electr. Insul., Vol. 5, pp.181-188, 1998.
B.R. Hamerling, F.J. Wester, E. Gulski, J.J. Smit and E.R.S. Groot,
“Fundamental aspects of on-line PD measurements on distribution power
cables”, IEEE 7th Intern. Conf. Solid Dielectrics, pp.408-411, 2001.
F.J. Wester, E. Gulski, E.R.S. Groot and M.G.A. van Vliet, “Sensitivity
of on-line PD detection for distribution power cables”, 13th Intern.
Sympos. High Voltage Engineering (ISH), Delft, The Netherlands, pp.
338, 2003.
F.J. Wester, E. Gulski, J.J. Smit and E.R.S. Groot, “Sensitivity Aspects of
On-line PD Diagnosis of MV Power Cables”, Cired, 17th Intern. Conf.
Electricity Distribution, Barcelona, Spain, pp 1-6, 2003.
IEEE 400 Guide for Field Testing and Evaluation of the Insulation of
Shielded power Cable Systems.
E. Pultrum, E.F. Steennis, M.J.M. Van Riet, “Test after laying, diagnostic
testing using partial discharge testing at site,” CIGRE Session, paper
15/2/33-12, 1996.
CIGRE WG 21.16, “Partial discharge detection in installed HV extruded
cable systems,” CIGRE technical report no. 182, 2001.
E. Pultrum, S.A.M. Verhoeven, “Testing of Extruded Cables: experience
in Type Testing, PQ Testing and Test After Installation. What do we learn
from it?” , CIGRE Session, paper B1-104I, 2004.
E. Gulski, „Computer-Aided Recognition of Partial Discharges Using
Statistical Tools“, Ph.D Thesis, Delft University Press, The Netherlands,
1991.
F.H. Krueger, „Industrial High Voltage Part I, Fields, dielectrics and
constructions“, Delft University Press, 1991.