Advanced Materials Research Vols.

1070-1072 (2015) pp 1039-1043

Online available since 2014/Dec/11 at www.scientific.net
© (2015) Trans Tech Publications, Switzerland
doi:10.4028/www.scientific.net/AMR.1070-1072.1039

Testing and Diagnosis of Extra High Voltage Power Cables Using

Damped AC Voltages combined with Distributed Partial Discharge
Measurement

Biao Yan1, a, Li Zhou1,b, Jie Chen1,c, Fengbo Tao1,d and Jian Zhang1,e
1
Jiangsu Electric Power Company Research Institute Nanjing 210013, China
a
fengchuiguolai@126.com, bzl-jtt@163.com, c2008840320@163.com

Keywords: on-site testing and diagnosis, extra high voltage (EHV), damped AC voltages (DAC),
distributed measurement, partial discharges (PD), monitored testing system.

Abstract. State assessment method of cables with extruded insulation and their accessories, DAC
withstand test combined with a diagnostic test (such as PD measurement) has two shortcomings, the
output voltage is not high enough for EHV cables, it cannot detect and locate PD effectively when
defects are far away from signal acquisition point. This paper focus on DAC voltages up to 250kV
and a kind of distributed measurement of PD. Defects existing in arbitrary of tested cable can be
accurately measured in theory by installing sensor at intervals along the tested cable. This method has
been validated on a long extra high voltage cross-linked polyethylene (XLPE) insulated underground
cable circuit and the filed PD test has been completed successfully.

Introduction
As an alternative to overhead power lines, power cables especially super long ones become the main
carrier of big cities’ electrical power network [1]. Insulation failures of power cables always occur
during operation at normal voltage or as a result of a transient voltage spike due to lighting or
switching surges, the failures also can occur when the bulk dielectric materials degrade to the point
where it cannot withstand the applied voltage. For on-site testing of extruded cables, withstand stress
were developed such as classic continuous AC or very-low-frequency on-site test [2-4]. It is know
that an overvoltage test will not always uncover all manufacturing and installing problems, what also
should be mentioned that a test may damage the insulation ever if a failure does not occur, also
failures will happen after a successful withstand stress testing at overvoltage within in a few months
in service, there is a possibility of a failure occurring if the test were to be continued for another few
minutes. Consequently a comprehensive state assessment of cables and their accessories following
installation or repair, known as monitored testing is now common practice, which means withstand
test combined with a diagnostic test, such as PD measurement [5-8].
It is the aim of this paper to contribute to the effective DAC monitored testing for super long and
large cross sectional area EHV power cables in China and, in particular of XLPE cables. Due to
frequencies of PD is very high, PD signal decays quickly in the cable, whose propagation distance is
limited in a small range, at the same time it is also affected by on-site external interference, it cannot
be detected accurately when the defects are far away from the PD signal acquisition devices,
conventional monitored testing cannot provide a comprehensive reflection of insulation condition of
cables and their accessories; conventional PD signal acquisition method using coupling capacitor is
not appropriate [9-12]. These defects not found by conventional PD detection method eventually
develop into an insulation breakdown later in service. A kind of distributed measurement system of
PD is presented in this paper and an example of DAC test system is shown in Figure 1.

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,
www.ttp.net. (ID: 130.126.162.126, University of Illinois, Urbana, USA-17/03/15,22:41:47)
1040 Applied Energy and Power Engineering IV

Figure 1. Diagnostic test of an 110kV (a) and a 220kV (b) XLPE power cable circuit.

Damped AC voltages monitored testing system

DAC was introduced more than 10 years ago as an alternative to the classic continuous AC or
very-low-frequency voltages, using continuous AC or DAC voltages on oil-related defect samples
and polymeric cable samples in the laboratory shows no major differences for the PD inception
voltage, PD magnitudes and PD patterns [5,13]. DAC testing can be used as withstand test or in
combination with PD measurements, it is in compliance to some related international standards
[11,12,14,15].
A test system has been developed to generate DAC up to 250kV with duration of a few tens of
cycles of AC voltage at frequency up to a few hundreds hertz, see figure 2, figure 3.

Figure 2. Schematic diagram of DAC circuit. Figure 3. Generation of one DAC excitation.
A high voltage power supply is used to energize the equivalent capacitive load of the cable circuit
being tested until the wanted maximum peak voltage Ump is measured [10], immediately Ump is
reached, a specially designed high switch connects an specially designed inductor to earth within 1µs,
the series of AC voltage cycles then starts at the resonance frequency of the circuit f0=1/(2•π•√(L•C)),
where L is the fixed inductance of the inductor and C is the overall cable capacitance, at the same time
PD is measured, the inductor generate almost no PD under extra high voltage. A predetermined
number of DAC excitations at maximum applied DAC level are applied, and one additional
diagnostic properties of the cable, e.g., PD levels are measured and used to determine whether the
cable circuit passes or fails the DAC test [10-14]. On-site testing and diagnosis process of
service-aged or after-laying cable circuits is shown in figure 4.

Distributed PD measurement
Conventional measurement of PD applies only to short power cable circuits; distributed PD
measurement method can solve this problem.
For long power cable circuits up to ten or several tens kilometers, an arbitrary number of sensors of
PD detection are installed on the cable being tested from space to space. Signal acquisition and digital
transformation are executed in each sensor on the spot; all of data generated by sensors are uploaded
to the data analysis terminal synchronously by the help of optical fibers, external interference can be
 Advanced Materials Research Vols. 1070-1072 1041

avoided by this means except for those generated by the test power source, see figure 5. Test results
obtained in this way may differ from those obtained by conventional PD measurement.

Figure 4. Testing and diagnosis process of cable circuits. Figure 5. Distributed PD measurement.

Practical verification
Field measurements have been performed in the field for comparisons between conventional PD
measurement and distributed PD measurement.
A breakdown of a joint occurred during operation on a service-aged 15.9km long, 220kV
XLPE-insulated underground power cable circuit with 2500mm2 cross sectional area, state
assessment has been decided to perform monitored testing on this cable circuit using DAC, see figure
6. The cable circuit has 34 groups of joints, 34 sensors are placed on these joints. PD activity has been
observed in a joint at 4.28km location starting from 0.5U0 (U0=220kV/√3=127kV) during increase of
the test voltage, increasing of the test voltage has shown increase of PD activity and at 0.8U0 test
voltage another breakdown during the DAC period has occurred, and based on the PD measurements
the failed joint has been localized, see figure 6.

Figure 6. Joints breakdown occur at: (a) normal voltage, (b) 0.8U0, (c) 1.0U0.
Figure 7 shows the PD activity for both diagnostics. The PD measurements results show that no
further PD activity or damage was detected of the cable circuit in figure 7 a) and b) through the
method of conventional PD measurement, but figure 7 c) and d) shows that PD activity could be
detected in another two joints (the 14th one and the 28th one) through the method of distributed PD
measurement, these two joints were further away from the PD signal acquisition point than the
damaged one. the test voltage was increased up to 1.0U0, the third breakdown of joint occurred and
was located at 5.96km location, it is consistent with the PD patterns shown in figure 7 c). All of these
damaged joints and the other suspected one whose PD activity is showed in figure 7 d) have been
collected for laboratory analysis. Anatomy analysis of the suspected one shows that creepage traces
can be observed clearly on the inner face of the suspected joint’s insulator as shown in figure 8.
1042 Applied Energy and Power Engineering IV

Figure 7. Partial patterns measured at: (a) 0.7U0, (b) 0.8U0, (c) 0.8U0, (d) 0.8U0.

Figure 8. Creepage traces found on the inner face of the suspected joint.

Conclusions
Based on the results above several conclusions can be drawn:
DAC can be applied as a good alternative to AC voltage or very-low frequency voltage in testing
and diagnosis of EHV transmission power cable circuits.
Combination with PD measurement, monitored testing is a better choice for conditional
assessment of EHV power cables.
Distributed measurement of PD can detect and locate more discharging defects and provide a more
comprehensive reflection of insulation condition of cables and their accessories than conventional PD
detection.

References
[1] SGCC: The 2009 annual concluding report of cable profession of SGCC (PTD of SGCC
Publications, China 2009).
[2] P. Grönefeld, R. von Olshausen: 4th Int. Symp. High Voltage. Vol. 21 (1983).
[3] W. Boone, G.C. Damstra, and W.J. Jansen: 5th Int. Symp. High Voltage. Vol. 62 (1987).
[4] E. Lemke, R. Röding and W. Weißenberg: CIGRE Symp. Vol. 1020 (1987).
[5] C. Aucourt, W. Boone, and W. Kalkner: CIGRE Symp. Vol. 105 (1990).
[6] E. Gulski: IEEE Trans. Dielectr. Electr. Insul. Vol. 2 (1995), p. 630
[7] J. Densley: IEEE Electr. Insul. Mag. Vol. 17 (2001), p. 14
 Advanced Materials Research Vols. 1070-1072 1043

[8] E. Gulski, P. Cichecki, and F. J. Wester: IEEE Trans. Dielectr. Electr. Insul. Vol. 15 (2008), p.
1691
[9] S. Boggs, A. Pathak, and P. Walker: IEEE Electr. Insul. Mag. Vol. 12 (1996), p. 9
[10] GB/T22078.1: 500kV rated voltage (Um=550kV) XLPE-insulated power cables and their
accessories. Part 1: 500kV rated voltage (Um=550kV) XLPE-insulated power cables and their
accessories-test methods and requirements (2008).
[11] IEC62067: Power cable with extruded insulation and their accessories for rated voltages above
150kV (Um=170kV) up to 500kV (Um=525kV)-test methods and requirements (2011).
[12] IEEE400: Guide for field testing and evaluation of the insulation of shielded power cable
systems rated 5kV and above (2012).
[13] F.J. Wester, E. Gulski, and J.J. Smith: IEEE Electr. Insul. Mag. Vol. 23 (2007), p. 28
[14] IEC60060-3: High-voltage test techniques-Part 3: Definitions and test requirements for on-site
testing (2006).
[15] IEC60840: Power cables with extruded insulation and their accessories for rated voltages above
30kV (Um=36kV) up to 150kV (Um=170kV) -Test methods and requirements (2011).
Applied Energy and Power Engineering IV
10.4028/www.scientific.net/AMR.1070-1072

Testing and Diagnosis of Extra High Voltage Power Cables Using Damped AC Voltages Combined
with Distributed Partial Discharge Measurement
10.4028/www.scientific.net/AMR.1070-1072.1039

DOI References
[6] E. Gulski: IEEE Trans. Dielectr. Electr. Insul. Vol. 2 (1995), p.630.
http://dx.doi.org/10.1109/94.466701
[7] J. Densley: IEEE Electr. Insul. Mag. Vol. 17 (2001), p.14.
http://dx.doi.org/10.1109/57.901613
[8] E. Gulski, P. Cichecki, and F. J. Wester: IEEE Trans. Dielectr. Electr. Insul. Vol. 15 (2008), p.1691.
http://dx.doi.org/10.1109/TDEI.2008.4712673
[9] S. Boggs, A. Pathak, and P. Walker: IEEE Electr. Insul. Mag. Vol. 12 (1996), p.9.
http://dx.doi.org/10.1109/57.484104
[13] F.J. Wester, E. Gulski, and J.J. Smith: IEEE Electr. Insul. Mag. Vol. 23 (2007), p.28.
http://dx.doi.org/10.1109/MEI.2007.386481