Improvement of Insulation Performance of Solid/Gas Composite Insulation With Embedded Electrode
Improvement of Insulation Performance of Solid/Gas Composite Insulation With Embedded Electrode
ABSTRACT
Solid insulated switchgear (SIS) has been developed as a substitution for medium
voltage sulfur hexafluoride (SF6) gas insulated switchgears. Its main circuit including a
vacuum interrupter is coated with epoxy resin. Therefore, a solid/gas insulation system
is composed. For further miniaturization and higher stress design of SIS, improvement
of the insulation performance is required. In the solid/gas insulation system, the
breakdown strength along the surface is lower than that of solid. Then, the authors
investigated the surface insulation performance of the solid/gas insulation system
containing an embedded electrode. In the present study, the sizes of the embedded
electrode and the solid insulation material were varied and the partial discharge
inception voltage and the surface breakdown voltage were measured. As a result, it was
found that the surface breakdown voltage varied considerably with the radius of the
embedded electrode. From the result, it was deduced that the surface insulation
performance could be improved by choosing the size of the embedded electrode
appropriately. Moreover, the effects of the thickness of the insulation material and of
the surface distance on the surface insulation performance were clarified.
Index Terms - Composite insulation, electric breakdown, stress control, surface
discharge, switchgear.
DOI: 10.1109/TDEI.2015.005407
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
788 T. Yamashita et al: Improvement of Insulation Performance of Solid/Gas Composite Insulation with Embedded Electrode
interrupter are coated with epoxy resin, has been developed Figure 2 shows the cross section of a test sample of the
[2]. However, improvement of the surface insulation solid/gas composite insulation system. This test sample was
performance of the solid/gas composite insulation system is the model of insulation system in Figure 1 and was composed
required for the development of higher stress and compact of a disc-shaped epoxy resin plate with back-side electrode
design. and disc-shaped embedded electrode. The back-side electrode
Therefore, the authors investigated the surface insulation was formed by painting conductive layer on the underside of
performance of the solid/gas composite insulation system. the insulation plate. The embedded electrode was made of
Metal electrode was embedded into the insulation material of copper and its thickness was 2 mm. It has round shape at the
epoxy resin. Then, the effect of the embedded electrode on the edge. The radius of curvature was 1 mm. The distance
insulation performance was investigated. In general, coating between the bottom of the embedded electrode and the back-
metal electrode with solid insulation materials is well known side electrode was 4 mm. The embedded electrode was
as a method of the improvement of the breakdown connected to the center rod electrode which was made of
characteristics [7]. In addition, the stress control effect of copper with a radius of 5 mm. By embedding electrode,
embedding an electrode is known. However, little is known modification of the electric stress around the triple junction is
about the effect of the embedded electrode on the surface expected.
breakdown voltage in air. In the present study, the sizes of the
insulation material and the embedded electrode were varied
and the partial discharge inception voltage and the surface embedded electrode
breakdown voltage were measured. As a result, it was found rod electrode
that the surface breakdown voltage varied considerably with epoxy resin
x
the radius of the embedded electrode. Moreover, the effects of
the thickness of the insulation material and of the surface t
distance were clarified. re
ground
rp
2 EXPECTED EFFECT OF EMBEDDED
ELECTRODE Figure 2. A model of composite insulation system with embedded electrode.
conductive layer
operation rod
H. V. shield
air
ground Figure 3. Calculation result of the potential distribution when the static
voltage was applied to the rod electrode and the back-side electrode was
grounded. r p = 100 (mm), r e = 50 (mm) and t = 11 (mm).
the surface to be considered
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
IEEE Transactions on Dielectrics and Electrical Insulation Vol. 23, No. 2; April 2016 789
center rod electrode. Incidentally, the electric field at the triple and r e was varied from 5 to 70 mm. The specifications of the
junction without embedded electrode becomes infinite test samples in this case were listed in Table 1.
theoretically.
Table 1. Specifications of test samples used for the test of the effect of the
radius of the embedded electrode.
14 Radius of the
Radius of the Thickness of the
re=5 mm Embedded Electrode
12 Plate (mm) Plate (mm)
Electric field strength [kV/mm]
(mm)
10 100 11 5
100 11 15
8 100 11 25
re=25 mm re=50 mm re=75 mm
100 11 50
6
100 11 70
4
Compared with the case when r e = 5 (mm), the maximum Table 2. Specifications of test samples used for the test of the effect of the
electric field strength became much smaller when r e was thickness of the insulation plate.
larger than 25 mm. The position where the electric field Radius of the
Radius of the Thickness of the
Embedded Electrode
strength along the surface was the maximum appeared just Plate (mm) Plate (mm)
(mm)
above the tip of the embedded electrode. The maximum 100 9 50
electric field strength was almost the same regardless of r e in 100 11 50
the range of 25 to 75 mm. The position x with the maximum 100 15 50
100 21 50
electric field strength was almost equal to r e.
Discharge inception on the surface is considered to be
related to the electric field strength along the surface, the (3) Effect of the radius of the surface distance
increase of the discharge inception voltage is expected by In order to examine the effect of the radius of the surface
embedding the electrode more than 25 mm in radius. The distance, t and r e were fixed at 11 and 25 mm, respectively,
uniformity of the electric field is also improved. Then, the and r p was varied from 50 to 250 mm. The specifications of
breakdown voltage is expected to increase. On the other hand, the test samples in this case were listed in Table 3.
by moving the position of the maximum electric field strength
towards the edge of the insulation plate, the distance between
the position of the maximum electric field strength and the Table 3. Specifications of test samples used for the test of the effect of the
radius of the insulation plate.
ground electrode along the surface decreases. Then, the
Radius of the
breakdown voltage is considered to vary with the radius of the Radius of the Thickness of the
Embedded Electrode
Plate (mm) Plate (mm)
embedded electrode. The reason is that the surface potential (mm)
distribution is varied with r e and the propagation of the 50 11 25
surface discharge, that is the elongation of the partial 100 11 25
250 11 25
discharge, will depend on the surface potential distribution.
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
790 T. Yamashita et al: Improvement of Insulation Performance of Solid/Gas Composite Insulation with Embedded Electrode
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
IEEE Transactions on Dielectrics and Electrical Insulation Vol. 23, No. 2; April 2016 791
80 breakdown voltage can be increased if the radius of the
120
the point of the maximum electric field strength along the
surface. 100
Voltage [kV]
employed to observe the partial discharge. As a result, the 60
100
40
80
20
BDV
60
Voltage [kV]
P D IV
0
0 20 40 60 80 100
40
R a dius of em be dded electro de, r e [m m ]
0
BDV and PDIV had maximum values when r e = 50 (mm). 0 50 10 0 15 0 20 0 250 300
R adius o f plate, r p [m m ]
The maximum BDV was almost 1.5 times higher compared
with the case r e = 5 (mm) which was the same as the radius of Figure 11. Relation between ac breakdown voltage and the radius of the
the center rod electrode. Therefore, it is deduced that ac plate. r e = 25 (mm) and t = 11 (mm).
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
792 T. Yamashita et al: Improvement of Insulation Performance of Solid/Gas Composite Insulation with Embedded Electrode
Voltage [kV]
ac. The maximum breakdown voltage was measured when the 60
radius was 25 mm. In addition, the polarity effect can be seen
slightly. Impulse BDV of positive polarity was slightly lower 40
than that of negative polarity.
20
BDV
100 P D IV
0
0 5 10 15 20 25
80 T hick ness o f pla te, t [m m ]
Figure 14. Relation between positive impulse breakdown voltage and the
60 thickness of the insulation plate. r p = 100 (mm) and r e = 50 (mm).
Voltage [kV]
40 120
100
20
BDV
P D IV 80
0
Voltage [kV]
0 20 40 60 80 100
60
R adius of em bedded e lectrode, r e [m m ]
40
Figure 12. Relation between positive impulse breakdown voltage and the
radius of the embedded electrode. r p = 100 (mm) and t = 11 (mm).
20
BDV
PD IV
100
0
0 5 10 15 20 25
T hick ness o f pla te, t [m m ]
80
Figure 15. Relation between negative impulse breakdown voltage and the
thickness of the insulation plate. r p = 100 (mm) and r e = 50 (mm).
60
Voltage [kV]
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
IEEE Transactions on Dielectrics and Electrical Insulation Vol. 23, No. 2; April 2016 793
100 5
80 4
40 2
20 1
B DV
P D IV P D IE
0 0
0 50 100 150 200 250 300 0 20 40 60 80 100
R a dius of pla te, r p [m m ] R adius of em be dded electro de, r e [m m ]
Figure 16. Relation between positive impulse breakdown voltage and the Figure 18. Relation between the partial discharge inception field strength
radius of the plate. r e = 25 (mm) and t = 11 (mm). and the radius of the embedded electrode under ac voltage application, r p =
100 (mm) and t = 11 (mm).
100
Figure 17. Relation between negative impulse breakdown voltage and the 80
radius of the plate. r e = 25 (mm) and t = 11 (mm).
70
60
5 DISCUSSION
50
Voltage [kV]
same and was around 3 kV/mm. The value of PDIE is R adius o f em bedded e lectrode, r e [m m ]
reasonable because the discharge inception electric field in air Figure 19. Relation between insulation performance and the radius of the
is known around 30 kV/cm. embedded electrode.
From the above, it is deduced that the partial discharge will
be initiated when the electric field reaches to be 3 kV/mm and Generally, the surface breakdown voltage (BDV) can be
the partial discharge inception voltage, PDIV can be increased decided based on the surface distance, distributed electrostatic
with choosing the radius of embedded electrode appropriately. capacity per unit area and the voltage polarity. However, it is
AC PDIV is quite important as the reason why the solid complicated in the present case. The surface discharge
insulation material such as epoxy resin will be deteriorated by propagates from the partial discharge inception point in both
the partial discharge. In the actual switchgear, the partial directions (towards the center electrode and the ground
discharge free design under the service voltage is necessary. electrode). The polarity of each surface discharge will be
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.
794 T. Yamashita et al: Improvement of Insulation Performance of Solid/Gas Composite Insulation with Embedded Electrode
different and the electrostatic property of each surface is also [5] T. Shioiri, J. Sato, T. Ozaki, O. Sakaguchi, T. Kamikawaji, M.
Miyagawa, M. Homma, K. Suzuki, “Insulation Technology for Medium
different. Moreover, the potential at the starting point varies
Voltage Solid Insulated Switchgear”, IEEE Conf. Electr. Insul. Dielectr.
with the configuration of the test sample. Therefore, detailed Phenomena, pp.341-344, 2003.
discussion is necessary to decide the BDV. The estimation of [6] J. Sato, O. Sakaguchi, N. Makishima, S. Kinoshita, T. Shioiri, T.
BDV will be reported precisely in another paper. Yoshida, M. Miyagawa, M. Homma and E. Kaneko, “New Technology
for Medium Voltage Solid Insulated Switchgear”, IEEE PES T&D Conf.,
Vol.3, pp.1791-1796, 2002.
6 CONCLUSION [7] e.g. T. Yoshida, H. Nishikawa, K. Takagi, S. Imakoma and N. Masaki,
“Air Gap Flash-over Voltage Rise Due to Inserting a Barrier Plate”,
Solid insulated switchgear (SIS) has been developed as Trans. IEE Japan, Vol.B-100, No.1, pp.57-64, 1980 (in Japanese).
substitution for medium voltage SF6 gas insulated switchgear.
For further miniaturization and higher stress design,
Takahiko Yamashita (M’00) was born in 1957 in
improvement of insulation performance is required. In SIS, Fukuoka, Japan. He received B.E., M.E. and D.E.
solid/gas insulation system is composed. The breakdown degrees from Kyushu University, Japan, in 1980, 1982
strength along the surface is lower than that of solid. Then, the and 1985, respectively. He has been working in
authors investigated the surface insulation performance using Nagasaki University, Japan, since 1985. He is a
Professor of the Graduate School of Engineering and a
the test sample, which was the disc-shaped epoxy resin plate Vice President of Nagasaki University. He is a senior
with embedded electrode. The sizes of embedded electrode member of Institute of Electrical Engineers Japan.
and the plate were varied and the partial discharge inception
voltage and the surface breakdown voltage were measured.
Kazuhisa Iwanaga was born in Nagasaki, Japan, in
The partial discharge inception position was also observed. 1991. He received B.S., degree from Nagasaki
The results are summarized as follows. University in 2014, where he is currently pursuing on
(1) Partial discharge is initiated at the point where the M. S. degree with the Graduate School of Engineering.
He is mainly engaged in development of gas/solid
electric field strength along the surface has the maximum. composite insulation system under high stress
(2) The partial discharge inception electric field is almost condition.
the same regardless of the sizes of the plate and
embedded electrode except for some conditions. The
partial discharge inception electric field strength for ac Tomohiro Furusato (M’11) was born in Kagoshima,
voltage is almost the same as 3 kV/mm. Japan, in 1988. He received B.S., M.S., and Ph.D.
degrees from Kumamoto University, Japan, in 2011,
(3) In case the radius of embedded electrode is varied, the 2012, and 2014, respectively. He was with the JSPS,
surface breakdown voltage has a peak value. However, Kumamoto University, from 2013 to 2014, as a Research
the radius where the breakdown voltage has the Fellow. Since 2014, he has been an Assistant Professor in
maximum is different between ac and impulse. Nagasaki University, Japan. His research interests are
pulsed-power, creeping discharge, and discharge
(4) The surface breakdown voltage increases with the phenomena in supercritical fluids.
thickness of the plate.
Hiroyuki Koreeda was born in Miyazaki, Japan, 1985.
(5) The effect of the surface distance on the surface He received B.S. and M.S. Eng. degrees in electrical
insulation performance is limited. engineering from Nagasaki University, Nagasaki Japan,
in 2008 and 2010. He was with the Mitsumi Electric Co.
From the above, it is deduced that the surface insulation Ltd.. from 2010 to 2013. Since 2014, he has been
performance can be improved by choosing the size of working as a technician in Nagasaki University.
embedded electrode appropriately. Furthermore, increasing
the thickness of the solid material is considered to be effective.
Authorized licensed use limited to: UNIVERSITY TEKNOLOGI MALAYSIA. Downloaded on March 11,2020 at 05:13:53 UTC from IEEE Xplore. Restrictions apply.