494
FLASHOVER STUDIES ON A SOLID DIELECTRIC SURFACE
*
A. N. SURESH , M. C . SIDDAGANGAPPA,
Department of High voltage Engineering,
Indian Institute of Science,
Bangalore-560012,INDIA.
*Department of Electrical Engineering,
Malnad College of Engineering,
Hassan-573201,INDIA.
INTRODUCTION
Solid-gas
interface
in
a
high
voltage
insulating system is often subjected to highly
localised critical electrical stresses leading
to
microdischarges and subsequent
surface
flashover. Such
a failure at the interface
is
of serious concern in a high voltage system.
This paper presents results of the
power
frequency and impulse flashover studies carried
out in air on a solid dielectric surface in a
non uniform field with air as the surrounding
media. The influence of humidity on the power
frequency
flashover
voltages
on a solid
dielectric surface is also discussed.
EXPERIMENT
Surface flashover studies were carried out on a
frosted glass plate (600mm x 450mm). Glass
plate was chosen because of its non hygroscopic
nature and flashover experiments could
be
repeated on the same solid dielectric specimen
with no change in surface conditions.
An
aluminium circular ground electrode (120 mm 0 )
with smoothly finished edge was fixed to one
of 5 mm 0 tapered to approximately 0.5 mm
one end was used as the high voltage
electrode.
1
single phase, 50 Hz, 2 KVA, 2 3 0 V / 2
transformer was used as high voltage so
The experiments relating to humidity eff
surface flashover studies were conducted i
a humidity and temperature controlled
(9OOmm x 600mm x 600mm). Impulse
vo
(1.4/70 v s wave) was obtained from a
generator.
A
The electrode gap system shown in figure
similar
to the one described by
Se
Hasegawa and Hideo Akagami [ 2 1 . The glass
was placed horizontally on an insulator
The gap length 'G' is the shortest distanc
the glass plane between the tip of the
electrode and the circular plate elect
Height 'H' is the vertical air gap maint
by the tip of the point electrode above
glass plate. The angle ' e ' is a measur
inclination of the point electrode with
glass plane, maintained at 6 0 " throughout
study.
RESULTS
Power
frequency
flashover
vo 1
experimentally obtained with and without
dielectric surface in the composite gap
and H are plotted against ' G ' in figure 2
has been observed that even in non un
field, presence of solid dielectric su
reduces the power frequency flashover vol
across the composite gap. Similar trend
496
Figure 3 shows the observed variations of power
frequency flashover voltage at different gap
lengths ' G ' and a constant vertical height H=O
for relative humidities of 60, 80 and 90 per
cent respectively. The flashover voltage in
the composite gap decreases with increase in
humidity.
Experiments were also conducted at a relative
humidity of 90 per cent for different 'G' and
'H' values. Figure 4 shows the flashover data
corresponding to H = 0, 0.4 cm and 0.8 cm over
a range of horizontal gap lengths ' G ' varying
from 0.5 cm to 4 cm in steps of 0.5 cm.
Positive impulse data of 50% flashover voltages
for different 'H' values are plotted against
horizontal gap lengths in figure 5. Increase in
air gap distance 'H' between point electrode
and
the surface of the solid
dielectric
resulted in increased flashover voltages €or a
particular gap length (GI.
DISCUSSION
It is well known that breakdown strength of any
by
effective
gaseous
gap
is controlled
ionisation coefficient (a-n), properties of the
electrode, and non uniformity of the applied
field. Apart from these conditions, inclusion
of solid dielectric surface even in a non
uniform field gap brings down the breakdown
strength of the gap. Further, reduction in
flashover voltage occurred due to fall in
leakage resistance of the solid dielectric at
high humidities.
With
the increase in vertical a i r aaD 'H'-
at
longer
gap
lengths
'G'
fla
characteristics of composite gap approach
path length characteristic (flashover v
versus 'G' characteristic for constant ve
height H = 0 ) . G to H ratio at which s
phenomenon occurs can be defined as "Cr
G/H ratio". Referring to figure 4, the cr
G/H ratio is 2 . 5 in both the cases of p
study. However, critical G/H ratio were 1
2 with the similar experiments conduct
Seiichi Hasegawa and Hideo Akagami 121
relative humidity of 90 per cent maintain
25°C. But, the present study though done
relative
humidity
of 90 per
cent,
temperature
was maintained at 36'C.
suggests that critical G / H ratios seem
dependent
on the ambient temperature
relative humidity.
REFERENCES
111
Craig Miller.H., "Surface flashove
insulators",
IEEE
Transactions
Electrical Insulation, vo1.24, No.5,
786,1989.
[21
Seiichi Hasegawa and Hideo Akaga
Discharge characteristics of a need
semicircular electrode gap with
dielectric surface under high humij
Japanese Journal of Applied Physic
27,N0.10,pp1942-1946,O~t.1988.
131
Cooke.C.M., "Surface flashover of
solid interfaces",Proceedings of the
G
international
symposium
on
498
Point electrode
(Sloinless s t e e l )
FIG.1 S C H E M A T I C ARRANGEMENT O F E L E C T R O D E S
40
H =O
25
- 0 :Wilh
30
-
15
-
70
-
9,
-
10
-
solid dielectric
: Wilhou1.solid dielectric
S l d . deviolion : 0 : l . L L 7
+
5 -
0 7
0
Fig. 2 Power frequency
G:
I -
1 -
-7
I
Iloshorcr
1
3
4
chorocleri>lics undcr o m b i e n l c o n d i l i o n r .
16
R H ‘90%
l5
0 H=O
14-
13 -
.*
C
CJ
12
-
+
H Z 0 tcm
o
H : o . B ~ ~
2
G C ~ C ~ I ~ I GC/ HO I~ O I I O
6
1 1 -
10
-
6
+
0
CP
Q -
0
8 -
0
8
>
0
,
C
0
-
U.
0
-I
I
I
I
1
t
2
1
0
I
3
Horbontol g a p length (C) in c m
F , g 4 P o w u r I r e q u a n c y Ilashovar choroclarislics w d l i solid diclcclric under
condltBons.
45
40
>,
-1
I
35 -
.-C
U
30-
0
%
25-
L
U
g
.c
200
0
G
6
In
Polorily :P O 5 i I i v Q
15-
0:H.O
+ : Hs04.cm
0 : HsOBcrn
Sld.dcviolion: 0 :1.509
+: 0.102
0:0.211
10-
.i
0
-I
0.0
I
I
2.0
2.5
I , - _ .
..
.
3.0
..
I-.
.