Transcript Title should be like this A P Robinson1, P L Lewin1, S

Space Charge Measurement in Polymeric Insulation under AC Conditions
Z. Xu, J. Zhao and G. Chen
University of Southampton, Southampton, UK
Introduction
The space charge study under AC condition is limited because there are many factors which can affect the results such as varying magnitude of the sinusoidal voltage, charge
injection/extraction due to the polarity reversal, material degradation, the measurement intervals, etc. PEA technique is already widely used in the space charge measurement. It utilises the
interaction between high voltage pulse and the charge layer in the insulation to produce the acoustic pressure waves.
The new system gives an essential way to examine the role of space charge in electrical aging process under AC conditions. There are two features with the improved PEA system: 1) a pulse
generator with a 3kHz repetition rate is utilized to reduce the measurement time; 2)The Eclipse data acquisition system is used to achieve the high data acquisition rate, the faster acquisition the
better representation of charge dynamics.
Measurement system control
System setup
 Old system:
•
Mechanical switch pulse generator (500Hz, 2ns pulse
width, up to 600V pulse);
•
GPIB card data transferring;
•
P.O.W control.
Induced charge
on top electrode
Amplitude
N average pulses
Next point
Next point
Time
GPIB cable
Function
Generator
PC
High voltage Trek
power amplifier
2nd point N averges
Digital
Oscilloscope
Pulse generator
0
o
45
o
90
o
o
o
135 180
o
225
270
o
o
315
360
1st point N averges
Pulses
3rd point N averges
mth point N averges
PEA measurement cell
 Improved system:
•
•
•
HTS transistor switch pulse generator (3kHZ, 5ns width,
up to 4kV pulse);
The Eclipse signal average console( 2GHz effective
sampling rate);
PCI data transferring.
TTL from Function Generator to trigger the HP 8082A
pulse generator (2kHZ, 4V)
AC voltage waveform
(a)
Sine waveform from the Function Generator to the Amplifier (10Hz,
Time offset to suit the total delay time of the pulse generator)
(b)
for DC measurement
PC
STP5 cable
T/200
BNC cable
T/m
TTL trigger enable in
Trigger out
(m: the point number defined to measured in one cycle)
Eclipse Digital Signal
Averager
HP Pulse Generator
t
T/2
T/4
Pulse from HP 8082A Pulse generator, go to the “trigger enable
in” of the Eclipse (400ns width, 2kHz, 4V )
(c)
HV-Pulse generator
The HV Pulse apply to PEA cell, Trigged by the Eclipse
PEA signal
Pulse from Eclispe trigger output (60ns width, 2kHz, 2.5V)
(d)
PEA measurement
Cell
DC Supply
Pulse from FET switch HV Pulse generator
(5ns width, 2kHz, 500V-4kv )
HV pulse
N Pulses applied at the defined measurement point
t
(e)
Time
for AC Measurement
GPIB cable
PC
Function Generator
STP5 cable
BNC cable
TTL trigger enable in
Trigger out
Eclipse Digital Signal
Averager
HP Pulse Generator
HV-Pulse generator
PEA signal
High Voltage Trek
power amplifier
PEA measurement
Cell
HV pulse
Result and Discussions
30
10
40
0°
180°
30
45°
225°
90°
270°
135°
315°
5
10
0
51
101
151
201
251
301
-10
-20
Charge Density (C/m3)
3
Charge Density (C/m)
20
1
Results measured from the new system
(ƒac=50Hz, ƒpulse=500Hz, Vpp=8kV)
Results measured from the new system
(decay test after DC measurement )
0
1
51
101
151
201
Thickness (mm)
-5
Thickness (mm)
-30
-10
-40
-15
 8 points can be selected (every 45°in one cycle);
 1000 averages were taken and total measurement
time is 160 second;
 The hetero-charge can be found next to the lower
electrode area.
251
Series1
Series2
Series3
15°
30°
45°
60°
75°
90°
Series4
Series5
Series6
105°
120°
135°
150°
165°
180°
Series7
Series8
Series9
195°
210°
225°
240°
265°
270°
Series10
Series11
Series12
285°
300°
315°
330°
345°
360°
Series13
Series14
Series15
Series16
Series17
Series18
Series19
Series20
Series21
Series22
Series23
Series24
Series25
Series26
Series27
Series28
Series29
Series30
Series31
Series32
Series33
Series34
Series35
Series36
Series37
Series38
Series39
Series40
Series41
Series42
Series43
Series44
Series45
Series46
Series47
Series48
Series49
Series50
Series51
Series52
Series53
Series54
Series55
Series56
Series57
Series58
Series59
Series60
 60 data series data were captured in 12 seconds with
400 averages per data;
 The distinctive characteristic of the new PEA system is
capable of illustrating smooth changing of the charge
decay.
20
Charge Density (C/m3)
Results measured from the old system
(ƒac=50Hz, ƒpulse=500Hz, Vpp=10kV)
10
0
1
51
101
151
201
251
Thickness (mm)
-10
Heterocharge
-20
-30
 24 series data can be measured (every 15°in one cycle);
 80 averages were taken and total measurement time is
1second;
 The similar phenomenon of hetero-charge can be found in
the region close to the lower electrode.
Conclusions
 The high speed measurement and good resolution give more details of charge dynamic in a short period of time.
 The frequency range of the AC supply is less limitation due to the better resolving ability in the new system.
 The measurement time is dramatically shorted by the new system especially for the lower frequency; hundreds of averages can be achieved in one cycle.
 The good phase resolution can be found for both high and low frequency applied voltage.
Dr Zhiqiang Xu, [email protected]
Contact details :
University of Southampton, Highfield, Southampton, SO17 1BJ, UK