HVPD Longshot PD Mapping System User Manual Sep 2009

High Voltage Partial Discharge Ltd
Empress Business Centre

380 Chester Road
Manchester
M16 9EA, UK
HVPD On-line Cable PD Mapping

Expansion System
User Manual
PD Map of Circuit
Stuart St - Queens Park (tee Eastlands)
9,000
All
8,000
All Phases PD
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
0
10 15 20
25 30 35
40 45 50 55 60 65 70 75
Location (% along cable)
80 85 90
95 100 105
HV Cable PD Mapping (location) using the HVPD-Longshot PD Spot

Tester, PTT 2000-CT Transponder & PD Map Software
PD Map Software Manual Version 4 - September 2009

Prepared by:
High Voltage Partial Discharge Ltd.
www.hvpd.co.uk
HVPD-Longshot - PD Map Software Manual v3 May 2009
Copyright HVPD Ltd.
CONTENTS
1.0
Introduction to PD Mapping Methodology and Hardware... 3
1.1
Requirements for Portable Transponder (Trigger and Pulse Booster).. 4
1.2
Portable Transponder Hardware (PD Trigger and Pulse Booster Units)... 6

1.2.1 PTT 2000-CT Discharge Trigger Unit. 7
1.2.2 PTT 2000-CT Portable Pulse Generator Unit 8
1.2.3
Transponder & HFCT Connections For Cable Mapping..11
1.3
Tips for using the Portable Transponders for PD location...12
1.4
Hardware Specification......13
2.0
PD Map Software for use with the HVPD-Longshot..........14
2.1
General...14
2.2
PD Map Software..15
2.3
PD Map Software - Main Capture Window..16

2.3.1 Operation.16
2.3.2 Selection of the Other Windows in PD Map..16
2.3.3 Setting up the Longshot Oscilloscope....17
2.3.4 Capturing the Set-up Waveform (Cable Return Time).17
2.3.5 Start and Re-start Waveform Capture..18
2.4
PD Map Software - Data Input Window.19

2.4.1 Entering and Setting channels..19
2.4.2 Entering data about Joint Positions on the cable20
2.4.3 Entering Default Data about joint sections on the cable...20
2.4.4 Deleting a Cable Section from the Inputted Data.20
2.4.5 Entering Calibration Data for Direct and HFCT Tests.20
2.5
PD Map Software - Data Processing Window.21
2.5.1
2.5.2
2.5.3
Display and Data Processing of the collected data21

Displaying a Preview of the Cable Map... .23
Scrolling through the records.23
2.6
PD Map Software - PD Map Window23
2.6.1
2.6.2
Display and Analysis of the PD Map..23

Print and Preview of Main PD Map.24
Appendix 1: Case Study of use of PD Map Software and Transponder..25
Appendix 2: Velocity of Cable PD Pulses for different types of HV Cables..36
1.0
Copyright HVPD Ltd.
Introduction to PD Mapping Methodology and Hardware
The HVPD-Longshot PD Spot Tester, when combined with the PD Map Software,
Portable Transponder and HFCT Sensors can be successfully applied to the accurate
location of PD sites along the length of high voltage cables. The technique utilises the
principle of Time-Of-Flight (TOF) measurement of the PD pulses, in power-line carrier mode,
along the cable. The PD pulse emanates from within the cable insulation passing along both
the earth screen and the core of the cable in this way. By measuring the time difference
between a directly received pulse and a reflected pulse (which has been reflected from the far
end of the cable) it is possible to locate the position of the PD site on the cable to less than
1% of the cable length.
IPEC High Voltage have successfully located PD sites within cables with this method to an
accuracy of less than 1% of the cable length in cables up to 4km in length (refer to Appendix
1 for Case Study of this application). The PD location method is based on the concept
illustrated below in Figure 1.
When a PD event occurs, the PD pulses travel outwards in both directions along the cable
earth screen (and cable core) from the originating PD site as illustrated in Figure 1 below.
The first pulse (Direct Pulse) to arrive at the measurement end of the cable is the pulse
which has travelled directly to that end. The pulse which then allows the PD site to be located
is the Reflected Pulse which originally sets off in the opposite direction, and is then reflected
back from the remote end back to the measurement end. This technique is called SingleEnded PD Location and is, when possible, the simplest and quickest way to provide PD
mapping of cables.
Direct Pulse
Reflected Pulse
PD event
Measurement End
Remote End
Figure 1: Single-Ended PD Site Location Method

If both the Direct Pulse and the Reflected Pulse are identifiable, then location of the site of
the PD event is relatively easy with the Single-Ended Location Method. Results would look
like:-
Direct pulse
Reflected pulse
pulse
L
T
L
Figure 2: PD Pulse Trains as seen from the Measurement End
Copyright HVPD Ltd.
With reference to Figure 2, the time difference between the first two pulses (the direct pulse
and the reflected pulse) T, locates the site of the PD event. It can be noted from Figure 2
that the two pulses will continue to travel up and down the cable, until they become too small
to be seen above the noise level. During this time, the pulses are reflected at exactly a cable
return time = L away from the previous arrival at the measurement end. This gives rise to
sets of pulses of diminishing size, each spaced at the cable return time, L. If L is the return
time of the cable length (this can be easily measured with the HVPD-Longshot with Cable
Map Software) then the location of the PD event is:
Location from Measurement End (in % Cable Length) = 100*(1-T/L)
Or alternatively,
Location from Remote End (in %) = 100*T/L
1.1
Requirements for Portable Transponder (Trigger and Pulse Booster)
Whilst the single-ended location method is possible in an ideal environment it has been
shown through practical testing (both on-line and off-line) using PD location methods on high
voltage cables, that the Single-Ended mapping methods can be made more difficult if the
cables are long or other circuit constraints apply. Difficulties can be encountered in the
following cases:
Signal Attenuation is high - long cables with high attenuation can reduce the magnitude of
the reflected pulse so that it is lost in the background noise.
PD Waveforms are difficult to interpret due to interference such as switching noise from
motors attached to the feeder.
Teed or jointed cables producing attenuation and reflections.
Cables with many Ring Main Units (RMUs) producing attenuation and (part) reflections of
pulses.
Cables with no change in impedance at the far end.
The solution to this problem is to apply the 2xUnit Portable Transponder Type: PTT 2000-CT
(shown below in Figure 3) which has been designed for use in PD location in the above
cases.
Figure 3: Portable Transponder Unit Type PTT 2000-CT

Top Discharge Trigger Unit
Bottom -Pulse Generator Unit
Copyright HVPD Ltd.
The concept of operation is that if a pulse is received by the Signal HFCT connected to the
Transponder Trigger Unit which exceeds the Transponders adjustable trigger level, then the
Trigger Unit will emit a signal to the Pulse Generator Unit which outputs a large, 100V pulse
(into 50 Ohms) to a Pulse-Injection HFCT which sends a large pulse back down the cable.
This process essentially converts the Single-Ended location system into a Double-Ended
location system.
The major advantage which the Double-Ended location system has over the Single-Ended
system, is that no waveform interpretation is required by the user for the Double-Ended case
as this is done by the Transponders Discharge Trigger Unit. It has been shown in practice
that it is often difficult to carry out the locations using the simple, Single-Ended method for the
reasons mentioned above (attenuation, high noise etc). Typically the reflected pulse is either
too small or it is confused with a lot of other pulses which may be present due to other
reflections, noise, or some other distortion of the waveform. Hence if the second pulse can
be made to stand out, then location is much simpler and more efficient. This is the primary
task of the Transponder, which boosts the size of the reflected pulse so that it can easily be
distinguished. A typical waveform, as seen at the measurement end of the cable, would then
look like:-
Reflected, Transponded pulse

pulse
Direct pulse
T
Figure 4: Example of waveform with Transponded Pulse

In most cases the HVPD-Longshot Spot Tester will be set by the user to see the size of the
direct (small amplitude) PD pulse, so that the amplified, transponded pulse is off the scale of
the OSM Tester screen in normal circumstances. Under these conditions, the OSM
oscilloscopes amplifiers become overloaded, and so the waveform which follows the
transponded pulse is not truly representative of the cable reflections which might exist in the
absence of the transponded pulse. To avoid this problem the Portable Transponder allows
for the transponded pulse to be delayed, so that the original waveform can be preserved and
measured in the conventional Single-Ended way, but still be triggered using the transponded
pulse. The HVPD-Longshot Spot Tester has sufficient memory to able to allow this
function enabling the same sample rate as would be used in the Single-Ended method, and
with a memory depth which allows either 20Sec, 50Sec or 100Sec delays in addition to
the cable time L. In this way the whole of the original waveform can be preserved, but will be
triggered with the delayed transponded pulse.
Copyright HVPD Ltd.
The location of the PD site with the transponder is calculated in exactly the same way as the
standard Single-Ended method, using the times T and L.
It can be seen from Figure 5 below that the original waveform can be preserved with this
method. This allows the Single-Ended method to be used, and in addition, a confirmation of
the position of the reflected pulse using the known delay of the transponded pulse.
A typical waveform would look like:-
Transponded pulse
pulse
Direct pulse
Reflected pulse
pulse
Transponded delay
Figure 5: Using the Manual Transponder with Transponder Delay

The PD Map Software (see section 2 of this Manual) for PD mapping has been designed to
take into account the Transponder Delay which may be programmed by the user and the
software automatically calculates the PD site location.
1.2 Portable Transponder Hardware ( 2x Units)
The concept behind the Transponder operation is to allow the PD signal from the remote end
of the cable (the position of the Transponder) to be first detected by the Discharge Trigger
Unit. Once a PD signal has been detected the Trigger Unit passes a signal through to the
Pulse Generator Unit which is used to boost (amplify) an outgoing pulse which is injected
back onto the cable screen/core and sent back along the cable to the measurement end. This
then allows the Reflected Pulse from the far end of the cable to be detected by the HVPDLongshot Spot Tester system at the measurement end. Hence the essence of the equipment
is a PD pulse trigger system coupled to a large output pulse generator (100V). Details of
these 2x units are given in the following sections.
1.2.1
Copyright HVPD Ltd.
PTT 2000-CT Discharge Trigger Unit
This instrument is designed to act as a Trigger unit for partial discharge pulses passing along
the screen/core of HV cables. The unit can also be used as a rough level guide for PD
activity, and is in the Transponder System to trigger the Pulse Generator unit which can be
used to locate partial discharges on cable networks. The Trigger unit is powered from a
rechargeable battery, which has the capacity to last 6-8 hours of continual usage between
charges. The Front Panel of the Discharge Trigger Unit is shown below in Figure 6.
Figure 6:
PTT 2000-CT -Discharge Trigger Unit
Power on/off
This switch turns on the power to the unit, which is battery operated.
ON
A Green LED is continuously on while the battery voltage is at a satisfactory level.
When the battery voltage is insufficient for the unit to work properly, the ON LED will flash as
a warning. At this point, switch off the unit and recharge the battery.
Input and input gain control
The BNC input for the PD signal (from the HFCT Sensor) is shown on the far left of the front
panel (marked INPUT). The switch to the right of the BNC input determines whether the
impedance is 50 Ohms or High Impedance. This choice is useful in cases where the PD
trigger unit is in a parallel connection with another device which has a 50 Ohm termination. In
such a case, the 50 Ohm switch should be set to OUT. Where the input is the only device
on the end of the sensor, then the 50Ohm switch should be set to IN.
Gain Control
The gain control is a single rotating knob which varies the gain of the Trigger Units amplifier
which feeds the trigger circuits. The signal levels are represented on this instrument by a
series of 10x LEDs running vertically on the front panel (6x Yellow and 4x Red). The signal
received will trigger the device if it reaches the first red LED (Level 7 LED out of 10). The
units gain level can be adjusted so that this trigger level can be obtained with the particular
signal levels available.
Copyright HVPD Ltd.
Trigger Light
Each time the device is triggered, the red Trigger LED will flash. If the light is on
continuously, this means that the trigger circuits are being continuously triggered which is not
normally required. If this occurs then the level of triggering should be altered (increased) by
adjusting the gain control so that the unit triggers only intermittently.
BNC Output (PULSE OUT)

The BNC output connector (PULSE OUT) on the right hand side of the unit is connected to
the input of Pulse Generator Unit (see section below). Scope triggers, and other triggered
devices can also be driven with the output of the triggered BNC.
Trigger Unit Set-Up - General

The Trigger Unit has been designed to be easy-to-use, and the operation of the instrument is
carried out simply as follows:1)
2)
3)
4)
1.2.2
Plug the Signal BNC from the HFCT sensor into the INPUT BNC
Switch the internal 50 Ohms to ON or OFF as required (this should be set
at ON in standard mode)
Adjust the large gain control knob so that that the level is at about the 7
LED mark (+/- 1 LED). At this point the unit will trigger.
The Trigger Output is then fed into the pulse generator which outputs the
100V pulse back onto the cable as described in the next section.
PTT 2000-CT Portable Pulse Generator Unit
The Portable Pulse Generator is used in conjunction with the PD Trigger Unit as a source of
fixed amplitude, variable-width pulses for inductive coupling onto a power cable via a High
Frequency Current Transformer. The Pulse Generator has a variable pulse width output
which can be varied by the user between 100ns and 10us.
The user may choose whether to initiate the pulses from an internal 10Hz generator, or from
either the rising or falling edge of an external signal at TTL voltage levels. The output pulse
height is 200V nominal peak voltage into a high impedance, or 100V nominal peak when
matched into 50 Ohms. A low voltage pulse is also provided for scope triggering purposes
with an amplitude of 2V nominal. The unit is powered from a rechargeable battery, which has
the capacity to last 6-8 hours of continual usage between charges. The battery-powered unit
has significant advantages for testing when compared to other units which require a 3-phase
mains power supply. The Portable Pulse Generator Unit is shown in Figure 7 on the
following page.
Copyright HVPD Ltd.
Figure 7: PTT 2000-CT PORTABLE PULSE GENERATOR and DISCHARGE TRIGGER

UNIT
DISCHARGE TRIGGER UNIT Front Panel Indicators
ON
A Green LED is continuously on while the battery voltage is at a satisfactory level.
When the battery voltage is insufficient for the unit to work properly, the ON LED will flash as
a warning. At this point, switch off the generator and recharge the battery.
TRIGGERED
A Red LED is illuminated whenever a pulse is being generated by the unit i.e. when the unit
has been triggered externally (from the Trigger Unit) or alternatively by its internal 10Hz
generator.
INPUT BNC
A detection HFCT is applied to the cable under test and its signal lead is attached to this
input.
Output BNC's
A trigger out signal is taken via a short BNC lead to the input of the portable pulse generator
to externally trigger the pulse generator.
Portable Pulse Generator - Front Panel Controls
POWER
When this Power toggle switch is in the DOWN position, the generator is switched ON.
This Power switch should be switched to the UP position when the unit is not in use to avoid
draining the internal battery.
AUTO/EXT
When this toggle switch is in the UP or AUTO position, the unit is triggered from an internal
generator at a fixed 10Hz rate. This function can be useful for example when the user wishes
to send automatic pulses along the cable to measure the cable return time of the cable. For
all other applications of unit the toggle switch should be in the DOWN or EXT position so that
the unit can be triggered from the PD Trigger
Copyright HVPD Ltd.
Units output (or another external signal source with TTL voltage levels) which is connected to
the EXT IN BNC connector.
TRIGGER
This Trigger toggle only has effect when the unit is in the EXT mode. In the up position, the
unit triggers from the rising edge of the input waveform. When in the down position, it triggers
from the falling edge.
NB
This has significance when the generator is used in conjunction with the PD
TRIGGER UNIT, which has a precise output pulse width of 25 microseconds, with the rising
edge coincident with the discharge pulse which triggered it. By triggering from the rising
edge, the high voltage output pulse occurs at the same time as the discharge pulse; by
triggering from the falling edge, the high voltage output pulse triggers 25 microseconds later.
RANGE
This toggle controls the output pulse width. In the UP (10us) position, the pulse width can be
adjusted in the range 1 to 11 microseconds. In the DOWN (1us) position, the pulse width can
be adjusted in the range 0.1 to 1.1 microseconds.
PULSE WIDTH
This rotary control adjusts the output pulse width. When the desired range of output pulse
widths has been selected by the RANGE switch, turn this control to set up the required pulse
width.
BNC Input - EXT In
A short BNC lead from the discharge trigger unit is connected here to cause the pulse
generator to output a 100V pulse onto the cable via an HFCT when the discharge trigger unit
fires.
BNC Output 1 - LV OUT
When the generator is sending a high voltage pulse, a +2V pulse appears at this output. This
is a low power pulse, only intended as a trigger for an oscilloscope.
BNC Output 2 - HV OUT
The output from this BNC is a positive pulse with nominal amplitude 100V when matched into
a 50 ohms load. As a high voltage signal, suitable precautions must be taken to prevent
damage if it is comes into contact with personnel or sensitive electronic equipment. Ideally it
should be connected via good quality coaxial cable into a high frequency current transformer
designed for the purpose. The arrangement must provide sufficient screening to avoid
breaching any local electromagnetic compliance regulations (EMC). Under no circumstances
should this generator be used on an aircraft in flight or any other environment where the high
voltage output pulse could constitute a hazard.
WARNING Under No Circumstances should the PULSE OUT signal be connected to the
INPUT BNC on the PD Trigger Unit as the 100V pulse will damage the sensitive input
circuits of the Trigger Unit.
1.2.3
Copyright HVPD Ltd.
Transponder & HFCT Connections For On-Line Cable Mapping
The standard test set-up for locating the site(s) of PD activity in an HV cable feeder using the
HVPD-Longshot PD tester with Portable Transponder is shown below in Figure 8. In this
example measurements are made at Substation A with the HVPD-Longshot Unit with the
Trigger Unit and Pulse Generator positioned at Substation B to generate the amplified,
transponded reflected pulses which are passed back along the cable for detection at the
measurement end.
HV (100V Output)
A Yellow LED is illuminated when the Pulse Generator's internal high voltage supply is at an
adequate level to provide the nominal output voltage pulse. Under normal conditions this will
be illuminated continuously. When being externally triggered, it is possible (because of the
setting of an incorrect trigger level by the user) that the unit will attempt to generate pulses at
an excessive rate, which presents too great a current demand on the internal high voltage
generator. Under these conditions, the high voltage generator will safely reduce its output
voltage until the current demand is within its capability, and then the HV Yellow LED will go
out. When the pulse demand is reduced, the internal high voltage generator will revert to the
correct voltage, and the LED will illuminate again.
Input BNC
When in EXT mode, a pulse applied here at TTL voltage levels ie low < 0.5V, high > 2V will
trigger the generator.
Figure 8:
Standard Test Set-Up for Double-Ended On-Line PD Cable Mapping
1.3
Copyright HVPD Ltd.
Tips for using the Portable Transponders for PD location on HV Cables
1. When setting up the Portable Transponder locations at the remote end it is always
important to set the correct PD trigger level with the PD Trigger Unit. This will enable the
Pulse generator to be triggered by PD pulses only and not noise or some other pulse
type.
2. Try to ensure that the same signals of the PD events are being seen by both ends of the
cable. This will guarantee that the Transponder will trigger on pulses which will reach the
far end of the cable. For example, if the cable is very long, (e.g. greater than 3km say)
and the pulses on the cable are very small (e.g. around 50pC say), then it is unlikely that
these PD pulses will be seen at both ends of the cable. Hence the transponded pulses
will not be related to any signal which can be seen from the measurement end of the
cable. This would be another reason to use a delayed transponded pulse, to make the
original PD pulse train visible. In such a case, it will be obvious that the transponded
pulse is not related to any signals at the measurement end.
3. If time permits, carrying out the measurements at both ends of the cable can provide
confirmation that locations have been made correctly and also provide more accurate
location results. Whilst this doubles the time taken for measurements it can sometimes
be worth it in cases where there is doubt or confusion as to the correct locations of PD on
the cable.
4. For most on-line measurements, there will be only a few sites which are responsible for
PD events. Use the Transponder so that it triggers on these events, and not on any other
pulses which may be related to noise or interference. In other words, take advantage of
the fact that larger PD events do make measurements much easier. This must be
balanced with the fact that PD sites often have a characteristic magnitude, especially
whilst they are developing. Hence try to ensure that PD events as small as possible can
be recorded, as these may be from PD sites which will grow into more active sites in the
future. Try to take advantage of good PD signals, but ensure that as good a dynamic
range of events as possible can be captured.
5. Often the PD activity can change during the course of making measurements as PD can
vary with load, voltage and time etc. It is often useful to have staff at the remote end
during on-line PD testing, to arrange for manual adjustment of the transponders during
the tests. Mobile communications are very important in this case, as is accuracy of
instructions between the two substations.
6. The cable length L can be measured using the Portable Transponder on-line if the actual
cable length is not known. Set-up the Pulse Generator (a second Transponder is ideal for
this) at the measurement end, and allow the Transponder at the remote end to be
triggered on the pulse generated at the measurement end. The time difference between
the launched pulse and the received transponded pulse, is the cable return travel time.
Check that this measured time seems sensible for the approximate length of cable being
tested. Check also that the measurement is repeatable, and not a trigger from noise at
the Transponder end.
7.
Copyright HVPD Ltd.
This is best carried out with two Transponders, as the signal levels will be large enough
to ensure noise is not a problem in triggering the remote Transponder.
8. Transponders can be put in cable circuits at Ring Main Units (RMUs) i.e. they do not
have to be put at the ends of cables. However it is important to make sure that the
transponder is well coupled to the circuit under test. Use the cable return length method
to ensure that the system responds to pulses on the cable, and that the coupling of the
transponded pulse back onto the cable is adequate. Often in such on-line cases, the
through coupling of HFCTs on earth straps is poor. In these cases, signal levels may not
be the same as at the ends. Check this if you intend to use this method.
9. The Transponders can be used off-line if required. In the off-line case, use another cable
core as the return path for the transponded signal. In this case, the transponded pulse
will never appear on the main signal path, so use a different channel on the HVPDLonghsot Tester for this application. It is generally easy to do this method off-line, as no
coupling difficulties arise. The signal level of 50V may actually be too big in this case,
and divide-by-ten probes may have to be used to detect the transponded pulse.
Please Note: The above Tips for Use are only an introduction to the application of the
technology. HVPD strongly recommend that the end user of the equipment procure a full
training package from HVPD in the use of the equipment.
1.4
Hardware Specification
Portable Transponder (PD Trigger Unit and Pulse Generator) Specification

Output Pulse magnitude
Output Pulse width
Pulse delay from Trigger Unit
Pulse delay from Pulse Generator Unit
Combined Delay
Settable output pulse delay
Trigger level range (high gain)
Trigger level range (low gain)
Trigger polarity
Input Bandwidth
Input Inhibit
Free Run mode

Free run mode intervals
Input/output protection
Input / Output connections
Power
Weight (Trigger Unit + Pulse Generator)
100V into 50Ohms

1Sec
375nSec +/- 5%
375nSec +/- 5%
750nSec +/- 5%
25Sec
5mV > 150mV
100mV -> 1000mV
Either +ve or ve pulses will trigger device
10MHz
No inputs will give triggers for 100Sec
following a trigger. This eliminates reflected
pulses re-triggering the device
Switched mode with device generating pulses
continuously at fixed interval
1mSec, 1 Sec, 1 Min
All protected by gas discharge tubes
BNC type
Battery Powered 6-8 hours continual use
4 Kg
2.0
Copyright HVPD Ltd.
PDMap Software for use with the HVPD-Longshot Unit
2.1 General
The PDMap Software gives the user the ability to record, process, display and manipulate
cable PD data maps and waveforms to provide for the on-line location of
Partial Discharge activity in HV cable feeders. The data is recorded on the HVPD-Longshot
Tektronix Waverunner Oscilloscope. The PD Map software has been developed with the
specific intention that the system can be used to draw PD maps from on-line testing, as well
as with the traditional off-line directly-coupled methods
(VLF testing etc).
The method of operation of making a PD map is as follows:
Set up the mapping equipment (HVPD-Longshot and Portable Transponder, refer to

Section 1.2 of this Manual), VLF high voltage supplies (if required i.e. for off-line testing)
and HFCT sensors (for measurement and pulse-injection, as per Figure 8 in section 1.4).
Run the PD Map Software (double-click on PD Map shortcut icon on Longshot Windows
Page) and fill in the data input sheet (this should include the positions of the joints and
terminations in the circuit, if available)
Apply the test voltage (or run on-line if required)
Record the PD waveforms for each phase of the circuit tested as required (use the start,
stop, and re-start buttons for each phase of the cable).
Process the waveforms after they have all been captured.
Draw the PD maps and display them as required
The software provides all the facilities to set up the HVPD-Longshot Oscilloscope into a
standard state, which can be used directly for mapping measurements. Full manual control is
available for more experienced users.
The manual is organised around the four operating windows, which comprise the PD Map
software. The tasks are split between these windows, and are as follows:1)
2)
3)
4)
Data Input Window

Data Capture Window (this is the main window)
Data Processing Window
PD Map Visualisation Window
There is also a Print Preview Page available, which is used to print a single page output of the
chosen PD map, the associated data, and any words which are recorded in the Report part
of the data input page. This allows for a simple report to be output by the user which contains
data, maps and recommendations, in a single printed page suitable for presentations &
reports.
2.2
Copyright HVPD Ltd.
PD Map Software - Installation
Installation Instructions
The HVPD-Longshot unit is supplied with the PD Map software pre-programmed into it. A
back-up copy of the software is also provided on a CD-ROM and the software is a single
executable file. The software is designed like this so that no further libraries, DLLs, or other
system files are needed to run the software.
The software can be run from any part of the HVPD-Longshot units hard disc drive (HDD)
and thus can be installed in any user selectable folder on the HDD. Whilst the software can
be run from the CD-ROM although for best speed of operation, it is advised that it is operated
from the Hard Drive. The set-up instructions in the likelihood that the software does need to
be re-installed are as follows:1) Create a new folder on the HDD to hold the PD Map Software. If an existing folder is
regarded as suitable, this is fine.
2) Copy the executable file PDMap.exe to the new folder.
3) Run the program by double clicking on the filename in Windows Explorer.
4) A shortcut can be placed on the desktop using the Windows Explorer wizard. The wizard
can be run from windows explorer by clicking the filename (i.e.PDMap.exe) and, from the
file menu, click on the new shortcut item. Then follow the wizard instructions to create the
shortcut.
On initialisation, the program checks to confirm that the oscilloscope is attached to the
internal GPIB card. If there are, then the addresses are used and noted on the capture page.
If no oscilloscopes are present, the program puts No scope in the address slot, and only offline tasks can be performed.
The software is designed to run on PCs with a display resolution of at least 1024x768 (i.e.
XGA).
2.3
PD Map Software - Main Capture Window
2.3.1
Operation
Copyright HVPD Ltd.
The PD Map software opens with the Main Capture page shown in Figure 9.
Figure 9: PD Map Software - Main Capture Window

2.3.2
Selection of the Other Windows in PD Map
The PD Map software system can be navigated by choosing another page from one of the
menu items on the capture page. The menus are arranged as follows:Data Input Window
Data Processing Window
Mapping Window
File menu
Processing menu
Mapping Menu
Before operation of the Data Capture window, the Data Input page should be completed with
details of the test set-up and the cable to be tested (refer to the section on Data Input).
Operators are urged to carry out the completion of the data input page with as much care as
possible, because information about the tests are always difficult to produce and/or remember
after the tests have been completed.
2.3.3
Copyright HVPD Ltd.
Setting up the HVPD-Longshot Oscilloscope
The HVPD-Longshot Units in-built Lecroy Waverunner Oscilloscope can be set up to a

standard, pre-set state, in readiness for capturing the PD Mapping waveforms. To achieve
this simply click onto the Scope Set up button. The button sets up the scope in the following
way
Sets up communications protocol

Sets up the required active channels on the screen
Sets up the largest number of points in the waveforms
Sets up the channel offset (to zero), removes any bandwidth limit, sets the coupling to
AC, sets the trigger position to 10% and sets the attenuator multipliers to unity
Sets the channel volts/div (to 20mV/Div)
Sets a typical time per div ( this is set at 2uSec per division i.e. 20uSec across the screen
which corresponds to a cable return length of around 1600m )
Selects an edge/window trigger, with DC coupling, with no offset, and a level of 10mV.
Following the addressing of the Oscilloscope, the Lecroy Waverunner Scope is then put in a
state of remote control. This means that the front panel buttons will not respond, other than
the one which states Return to Local (if this menu button is pushed on the scope, then the
front panel controls again become active).
2.3.4
Capturing the Set-up Waveform (to calculate Cable Return Time)
When setting up mapping measurements, it is essential that the user measures the PD Pulse
Return Time of the cable under test. This is the time it takes for a PD Pulse (or PD calibration
pulse) to travel along and back the entire length of the cable. In the software this is called the
Set-Up Waveform and is very useful later when the user is processing the PD waveforms
received from the test. Measurement of the PD Pulse Return Time of the cable is achieved in
practice by measuring the return time of an injected pulse from the Portable Pulse Generator
and an HFCT to inject a pulse onto the earth screen of the cable at either end of the cable. A
second Measurement HFCT is used with the unit to measure the time difference between
the injected pulse and the reflected pulse which arrives back from the far end (refer to Section
1 of this Manual for details).
The cable return time will be equal to:
Cable Return Time
Length of Cable/Pulse Propagation Speed
It is known that the average PD Pulse Propagation Speeds for various types of cable can be
estimated as follows:
Average PD Pulse Propagation Speed for XLPE Cables = 76m/Sec
Average PD Pulse Propagation Speed for PILC Cables = 80m/Sec
Average PD Pulse Propagation Speed for PILC/XLPE Mixed Cables = 78m/Sec
Copyright HVPD Ltd.
So for example, with a 3km long, wholly-XLPE cable the PD Pulse Return Time would be:
PD Pulse Return Time = 3000m/ 76 m/Sec = 39.47 Secs.
During the test the user should ensure that they have the right timebase on the oscilloscope
for the length of cable under test so that they can see both the injected pulse (from the pulse
generator) and the second pulse (reflected pulse). This can be achieved by simply dividing
the length of the cable under test (in m) by 80 to provide a rough estimate of the total
timebase which should be displayed on the oscilloscope.
To store the Set-up waveform in the software click on the Store Setup Waveform button.
This always stores the present waveform displayed on the oscilloscope screen into record No.
0 of the internal arrays. This can be done at any time during the testing.
2.3.5
Start and Re-start Waveform Capture
The PD Waveform capture process is started by clicking the Start Capture button in the
window. A file dialog box is then used to specify the filename in which all the data of the
current measurement can be stored. The measurements can be re-started as many times as
required, with different phases of the cable, so all the data is kept in the same file. To re-start
the capture process, simply click on the Re-Start Capture button.
The phase of the cable under test should be selected, as later this will be used to display the
phase-related PD data from the cable. If no phase is selected by the user (as may be the
case where it is not possible to access the individual phase of the cable with the HFCT
sensor) the software treats this as combined, three-phase test, called no phase. The phase
names should be typed into the edit boxes by the User if the default names (red, yellow, blue)
are not suitable.
The PD waveforms can be seen in real time on the PC screen, if the Radio buttons in the
Display Control are highlighted. The data transfer is slightly faster with no display of
waveforms, but this is probably best sacrificed for clarity, as the process is better seen with
the waveforms displayed. This gives some control over the quality of the measurements
obtained.
2.4
PD Map Software - Data Input Window
2.4.1
Entering and Setting channels
Copyright HVPD Ltd.
The measurement system can use up to 16 channels. These are selected using the data
input page. The data input window is accessed through the Data Input Form menu item in
the file menu on the main window.
Figure 10 below shows an example of the Data Input window. The essential box to fill in is
the measured Calculated Return Time, measured in Secs (refer to section 2.2.4 of this
Manual for details) as the program will not proceed unless this is available. Furthermore, it
must be urged that for the best results, it is essential to fill in all the data boxes below the
Joint Sections line to include details of cable sections and joints along the entire length of the
cable feeder. All the edit boxes are simple and intuitive to fill in. When all of the boxes have
been filled in and have been checked by the user then the Save Settings button should be
clicked.
Figure 10: PD Map - Data Input Window
2.4.2
Copyright HVPD Ltd.
Entering data about Joint Positions on the cable.
Up to 50x cable joints can be entered in the Joint Position boxes in the software. To enter a
cable section into the circuit database, click on the Insert after current Line button, and enter
the length of the section in the length edit box. After the correct cable section length has
been entered, click on the Save Line button to enter this section into the software. All the
start and end positions are automatically calculated, and are not required to be entered.
2.4.3
Entering Default Data about joint sections on the cable.
The details of cable type, manufacturer, propagation speed, and attenuation can be entered
at the same time as the cable section length is inputted. When the Save Settings button is
clicked, all the data will be entered into the section length. However, to save effort, all the
cable sections can be filled with a user chosen set of defaults to save time. To achieve this fill
in the default propagation speed, attenuation, cable type and manufacturer edit boxes, and
click on the fill with default button. This fills all the data with the default type. Hence if the
circuit was made up of sections of identical cable type, etc, then using the default entry will
save much detailed entering of cable data.
2.4.4
Deleting a Cable Section from the Inputted Data
If data has been entered incorrectly, or subsequently found to be wrong, the data in the cable
sections can be simply changed by the user. Use the scroll bars to access the cable section,
and simply enter the correct section length and other details. A click on the Save Line
button re-saves the new line. If the section is to be deleted, then click on the delete line
button and the section is deleted from the data.
2.4.5
Entering Calibration Data for Direct and HFCT type measurements.
PD measurement voltage calibration is required to ensure that the PD Map which is plotted
can be made using pico-Coulombs (pCs) as the PD measurement units. The magnitudes of
the PD pulses are calculated for direct voltage measurement using the formula:Where Const is the actual/measured voltage calibration constant and Z is the impedance of
endPulse
Magnitude( pC )
I dt
Startpulse
endPulse
Const
V dt
Z
Startpulse
the cable. Note that for 50 Ohm systems with no other attenuation or amplification, then
Const=2. It has been the custom to use Z = 20Ohms in the past, and this has been entered
as a default in the software. However, if specific Impedance data about the cable is known,
then this can be entered. The software calculates the
Copyright HVPD Ltd.
Integral of the PD pulse, measured in pCs, and the only user requirement is to ensure that
the cursors are correctly set by the software in the processing page.
If high frequency current transformer (HFCT) measurements are made on-line, then the
calculations for the PD magnitude are slightly different. The current is actually measured by
the CT, and the transfer impedance of the CT can be used to calculate the current, and hence
the charge in pC. This assumes that the measurement impedance is roughly the same
impedance as the cable which is the case.
The PD magnitude in pC for HFCT measurements can be calculated as follows:endPulse
endPulse
1
Magnitude( pC ) I dt ( ) V dt
Tr Startpulse
Startpulse
Where Tr is the Transfer Impedance of the HFCT used for the data capture. The IPEC HV
HFCTs supplied with the mapping system have a transfer impedance of Tr = 5.0, so the
constant of calibration is (1/Tr) = 0.20, and this number is entered in the CT calibration edit
box. If other HFCTs are used with different Trs, then the specific HFCT Transfer Impedance
should be changed here.
During the calculation, no user intervention is required, except that the recording type radio
buttons (on the processing page) must be correctly set to choose between on-line CT
measurements, and the off-line direct voltage type measurements.
2.5
PD Map Software - Data Processing Window
2.5.1
Display and Data Processing of the collected data
Following the data capture, the PD waveforms obtained need then to be processed by the
positioning of the time cursors in the software to enable the pulse magnitudes and locations of
the PD waveforms to be calculated. Figure 11 overleaf shows an example of the Data
Processing window. The main task in this window is to allocate the correct cursor positions
for each waveform collected. The first two cursors (green and yellow) must be placed at the
start (green cursor line) and end (yellow cursor line) of the initial pulse to enable the pulse
magnitude (in pC) of the pulse to be calculated by the software. The third (blue) cursor must
then be placed at the start of the reflected pulse, to enable the PD location to be calculated.
The PD location is calculated from the cursor positions as follows:-
Location(%) (1
T
) 100
Tl
Where Tl is the PD Pulse return time, and T is the time between the start of the first pulse,
and the start of the pulse which has been reflected from the far end of the circuit (this is the
third, blue cursor). Once the cursors have been set, the program automatically calculates the
location of the PD site which produced the waveform. A fourth (white) cursor is also
automatically placed at exactly Tl following the first cursor, to help in identifying the peaks in
the waveform.
Copyright HVPD Ltd.
The cursors shown in Figure 11 are set automatically by the PD Map program, and if these
are correct, then clicking the Approve & Next button will record the users approval of each
and then move onto the next waveform collected. If the cursors are not in the correct
position, then the cursor scroll bars can be moved manually by the user until they are in a
satisfactory position. To aid this procedure, the position of the third cursor (i.e. the start of the
pulse reflected from the far end) can be re-calculated by the program. The cursors are set by
a simple algorithm. The largest excursion of the waveform is traced to the average line, and
the intersection used for the cursor positions. Hence the first pulse is calculated from the
largest excursion of the waveform. By asking for a re-calculation of the second pulse, the
program uses the next highest pulse (either +ve or ve) as the pulse starting point. This can
be done as often as required. Often the noise and other factors may conspire to make the
reflected pulse not the highest in the waveform. Hence this method provides a relatively
quick way to try some new possibilities.
Figure 11: PD Map - Data Processing Window

If a return to the original state is required, then click on re-start cursors to return to a original
position of the cursors.
If the waveform is Un-interpretable as a PD Waveform, or simply noise or some other
interference, then the user should click the Reject & Next button. This will ensure that the
waveform will not be used to draw maps in the final page (see PD Mapping Window).
Copyright HVPD Ltd.
Note that only those waveforms which have been approved by this method, will be used in
the final PD Map. This avoids any mistakes which could be made in an automatic system and
allows the user to filter out any spurious readings.
2.5.2
Displaying a Preview of the Cable Map
The software provides a facility for a preview of the map on this page. In the file menu, the
Process File menu item will give an instant view of the map of the recorded data, according
to the simple algorithm of setting the cursors described above. Even though this may be an
accurate eventual map, the waveforms still have to be individually approved, to avoid any
errors in the locations of the waveforms.
2.5.3
Scrolling through the records
The records (i.e. waveforms) can be viewed simply by scrolling the record number scroll bars.
For each record, the waveform will be displayed, and the calculated parameters for this
waveform will also be displayed in the edit boxes.
2.6
PD Map Software - PD Map Window
2.6.1
Display and Analysis of the PD Map
Figure 12 overleaf shows the main PD Map window from the software. The operation of this
window is designed to be intuitive. This Main Map is displayed whenever the user has
confirmed any approved data in the current file. The joint positions can be displayed on the
Map using the Joint Display button (assuming that this data has been inputted in the Data
Input Page). This button toggles the joint positions on or off the map as required.
The PD Map data can be displayed in four modes, as follows:1. As vertical lines so that each waveform is represented by a vertical line at the location of
the PD event, and with a height equal to the calculated magnitude (as shown in Figure
12.
2. As dots so that each waveform which has been approved is represented by a dot at the
location of the PD event, and with a height equal to the calculated magnitude
3. As the sum of the number of events in a fixed width slot along the cable (each slot being
1% of the cable length)
4. As the sum of the product of the number of events*calculated magnitude. Again these
slots are each of width of 1% of the cable length.
These display types are toggled using the Map Display Points radio buttons.
The data can be displayed all on one graph, or on three separate graphs, one for each phase
to allow comparison between phases. The User can click on the Map control radio buttons
to change between the single and three phase display.
Copyright HVPD Ltd.
The map can be copied to the windows clipboard using the menu item Copy Map To
Clipboard in the Report menu of this window.
Figure 12: PD Map Software - Main PD Map Window

2.6.2
Print and Preview of Main PD Map
The PD Map can be printed using the Print Preview menu item in the report menu.
Whatever type of graph has been selected is used for the preview (and hence printing). The
PD report shows all the PD data, the PD Map, and the words of the section at the foot of the
data input page. This allows for all the data, and some relevant comments to be made and
printed on a single sheet. Whilst the Printer to be used should be capable of printing graphics
this is not normally now a restriction with most modern printers.
Copyright HVPD Ltd.
Appendix 1: Case Study of Application of PD Map Software and Transponder
OSM Case Study
On-line Cable Monitoring and Pre-Fault Location using

OSM On-Line Cable Mapping System
Executive Summary
This report describes the identification, measurement
and location of severe partial discharge activity on a
2677m length of 33kV mixed cable (PILC, XLPE & EPR)
between two substations within a UK Electricity
Companys Distribution Network in May and June 2003.
The PD activity in the cable was initially detected by
HVPDs HVPD-F64 Permanent OSM system which was
installed at the customers substation. An increasing
level of discharge activity activated a threshold alarm on
the OSM and the customer requested a series of spot
tests to identify the source(s) of the partial discharge
HVPD utilised our HVPD-ST1 Spot Tester with
integrated On-Line Cable Mapping software (Cable
TM
Map ) and Portable Transponder to find the location of
two discharging sites on the cable (at Trifurgating
joints). Location was confirmed with further tests
following excavation and the joints where the excessive
PD activity were replaced.
Final tests show that the discharge activity on the cable
had been successfully removed through replacement of
the discharging TRIF joints. The total time from
detection of excessive levels of discharge to
replacement of the joints was 8 weeks.
Case Study prepared by:

Dr Lee Renforth, Director
HIGH VOLTAGE PARTIAL DISCHARGE LTD.
October 2004
Copyright HVPD Ltd.
Introduction
The UK Electricity Distribution Companys Grid/Primary (132kV/33kV) substation had an
HVPD-F64 Partial Discharge Permanent Monitor installed in July 2002. Fourteen switchgear
panels and their feeder cables were being monitored due to their critical location in the
system, supplying power for the Commonwealth Games complex.
th
On 29 April 2003 the OSM monitor sent out a warning alarm to the Customer and IPEC that
there was excessive PD activity levels in the Queens Park/Eastlands circuit. A site visit was
made by Dr Lee Renforth of IPEC and Dr Ross Mackinlay of HVSL (prior to the joining of the
two companies to form IPEC High Voltage) who used the OSM PD Spot Tester. High levels of
discharge in the circuit in question were confirmed with this testing and the customer was
alerted. Further to discussions between the customer and IPEC engineers from IPEC and
rd
HVSL carried out tests with the OSM PD Tester and Portable Transponder on 23 May 2003
in order to locate the source of the discharge activity.
From the panel where the discharges were detected, a cable tee led to two transformers at
Queens Park and Eastlands substations. Using the OSM On-Line Cable Mapping technology,
two discharge sites were detected in the Queens Park cable at approximately 1564m and
1681m from Stuart St substation. The discharges were approximately 6,00pC and 9,000pC in
magnitude respectively and both were on the red phase.
Figure A1: Discharge sites on Queens Park cable.

Three joints were dug up for direct inspection with IPECs RF discharge detecting equipment.
With reference to the cable route map shown in Appendix A, they were at 1,575m, 1,700m
and 1,703m from Stuart St. After excavation Radio Frequency (RF) PD Tests were performed
on the joints with a portable detector and local discharge activity was confirmed.
th
The joints and the cable between them were replaced on 26 June 2003. Further PD tests
th
were carried out at Stuart St. on 4 July, showing that the discharge activity had been
eliminated. This Case Study briefly presents the techniques used to perform these tests and
the results they gave.
OSM installed system

The HVPD-F64 monitor permanently installed in Stuart St has IPEC 100/50 HFCTs (High
Frequency Current Transformers) connected to the cable earth straps from all 14 33kV
switchgear feeder panels. PD signals detected by the sensors are monitored constantly and
raise alarms when pre-set thresholds are exceeded. Data is downloaded automatically to
IPECs server every night and stored in a database for future data-mining. The PD signals
Copyright HVPD Ltd.
recorded are resolved with respect to phase. Figure A2 below shows the phase-resolved
signals picked up from the Queens Park / Eastlands Tee over a one hour
Copyright HVPD Ltd.
th
period on 19 June before the joints were replaced. Figure A3 show signals taken over a
similar period on the same circuit after the joints were replaced.
Figure A2: Activity on Queens Park cable

before joint replacement
Figure A3: Activity on Queens Park cable

after joint replacement.
The OSM monitor can be accessed directly over the phone line by any computer and the data
viewed using a standard Internet browser. Figure A4 below shows the discharge activity on
the Queens Park cable as seen directly from the system.
Figure A4: Activity as presented by the monitors web server.

HVPD Longshot PD Spot Tester and PD Gold
On-line tests were carried out from Stuart St using the OSM PD Spot Tester which
incorporates the PD Gold discharge analysis software. The OSM Spot Tester captures
signals from the HFCTs and allows accurate PD level measurement and phase resolution.
A report was produced from measurements taken before (Report 1) and after (Report 2) the
discharging joints were removed. The results of this testing are shown below in Figures A5
and A6 (using the standard PD Gold Report Format).
The actual signals were analysed and confirmed as being from a discharge source. As the
discharge levels within the cable were relatively high (50mV for a 33kV cable) the customer
decided to carry out some Off-Line Cable Mapping testing which identified the PD problem to
be at 2x transition joints. A
Copyright HVPD Ltd.
decision was made to replace these cable joints. Further PD testing and monitoring was
carried out after replacement of the discharging joint. The result of this testing are shown in
Figures A5 and A6. It can be noted in OSM PD Reports show that the signal levels after the
joints were replaced were an order of magnitude lower (approx. 5mV c.f 50mV) than before
replacement. This result was further confirmed from the OSM F64 PD Monitoring results
shown in Figure A6.
Filename of measurement
Date of Measurement Time at Start
C:\Mydelphi\PDgold\TekMeasure\Results\IPEC\Stuart St 1
15:18:44
29/04/02
PD parameters measured between cursors

Largest Peak Value......................... 50.20 mV
Largest Average Value...................... 5.49 mV
Largest 1% Level Value..................... 41.57 mV
Largest Short Term Severity................ 4078.43 mV.Counts
Largest Counts per Cycle................... 106.00 Counts/Cycle
Comments
Measurement Comments
Transponder HFCT measurement. Red Phase measured. Double cables into 33kV
Switchgear
Results Comments
This cable is active at a single site. It is only a single phase (most probably the red phase)
and a single position which is responsible for the PD activity above. Previous experience
shows that this cable is at risk of failure, but may still have some months or even several
years before failure. However, proper locations should be carried out, and an action plan
made. Locations could be done on-line in this case, as the signals are clear and relatively
noise free. Monitoring should be continued to track the progress of the PD site.
Figure A4:
Copyright HVPD Ltd.
HVPD-ST1 PD Report from Discharging 33kV Cable
Filename of measurement
Date of Measurement Time at Start
C:\Mydelphi\PDgold\TekMeasure\Results\IPEC\Stuart St 3
20:15:02
04/07/2002
PD parameters measured between cursors

Largest Peak Value......................... 4.71 mV
Largest Average Value...................... 1.10 mV
Largest 1% Level Value.................... 2.12 mV
Largest Short Term Severity................ 233.41 mV.Counts
Largest Counts per Cycle................... 125.00 Counts/Cycle
Comments
Measurement Comments
Transponder CT measurement. Red Phase measured. Double cables into 33kV Switchgear
Results Comments
These measurements were made following repair of the cable to Queens Park, where two
discharging joints were replaced. The large PD events from before the replacement are now
absent. The occasional noise interference is still small compared to the original PD measured
before replacement. This clearly shows the joints were responsible for the PD activity.
Figure A5:
HVPD-ST1 PD Report on 33kV Cable after Replacement of

Discharging Joints
Copyright HVPD Ltd.
PD Site Location: HVPD PD Spot Tester, Portable Transponder and PD Map Software
In order to locate the source of the discharges the PD Map On-Line Mapping System was
used. A Potable Transponder with HFCTs was fitted to the earth of the Queens Park cable, at
the Stuart St site. The mapping system was set up to inject a large (100V) high frequency
impulse onto the earth when a discharge signal was detected. By measuring the time
difference between the transponder signal and its reflection from the end of the cable, the
cable length was verified and cross-checked with the customers cable route map.
Measurements were then made at the remote transformers, Queens Park and Eastlands,
again using an HFCT sensor to measure the time interval between the discharge and
transponder signals in the earth.
Results
The mapping results are shown in Figure A6. The measured magnitudes are large at around
6,000pC and 9,000pC peak value. It should be noted that the discharges are attenuated by
their travel along the cable due to the cables surge impedance.
PD Map of Circuit
Stuart St - Queens Park (tee Eastlands)
9,000
All
8,000
All Phases PD
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105
Location (% along cable)
Figure A6: PD Mapping at Stuart St

The results showed two sites of PD activity on the feeder. The PD Map waveforms for these
sites are shown in Figures A7 and A* overleaf.
The map shows locations for these positions are as follows:First Site at 58.4% from Stuart St (1564m from Stuart St)
Second Site at 62.8% from Stuart St (1681m from Stuart St)
Copyright HVPD Ltd.
Waveform data in Time

Chan 1 (mV)
Cur 1
Cur 2
Cur 3
Cur 5
30
Voltage (mV)
20
10
0
-10
-20
-30
10
15
20
25
Time (uSec)
30
35
40
45
50
Figure A7: Waveform of on-line PD (location 58.4% from Stuart St End)
Waveform data in Time

Chan 1 (mV)
Cur 1
Cur 2
Cur 3
Cur 5
30
Voltage (mV)
20
10
0
-10
-20
-30
10
15
20
25
Time (uSec)
30
35
40
45
50
Figure A8: Waveform of on-line PD (location 62.8% from Stuart St End)

These results were compared to the cable route map and locations corresponded to
trifurcating joints at 1575m (SJ8699/27) from Stuart St and a further trifurcating joint at 1695m
from Stuart St (SJ8699/18).
The locations using this technique will result in measurements that are about 1% accurate
over the cable length. The measurements in this case are better than 1%, as data was
obtained from both ends of the circuit.
Although the measurements were made on the earth straps of the cables at Stuart St, the
largest signals were obtained on the red phase.
Copyright HVPD Ltd.
Discharge tests
Three joints; SJ8699/18 SJ at 1700m, TRIF JT 8SJ699/18 at 1695m and TRIF JT SJ8699/27
at 1575m, were dug up and tested using Transient Earth Voltage (TEV) transducers. These
pick up, through capacitive coupling, very fast transients that are induced in the cable earth
every time a partial discharge occurs. As the earth acts as a screen around cables and joints,
TEV voltages only emanate from points where there is a break in this screen. From these
points they travel along the outside surface of the earth. They are very high frequency and
therefore attenuate rapidly with distance from the source.
Using filters to eliminate interference from other sources, the TEV signals were monitored and
measured on a fast digital oscilloscope.
Tests were also carried out using IPECs Portable RF interference detector. A 0.4m whip
aerial is used to pick up airborne RF activity and a multi-turn tuner allows continuous
frequency selection in the in the range 48 to 860MHz. Detected RF levels are displayed on a
linear analogue meter and the demodulated RF signals can be listened to using earphones
with an adjustable volume. The detector is shown in figure A9 below.
Figure A9: HIGH VOLTAGE PARTIAL DISCHARGE LTD.RF Detector.
Copyright HVPD Ltd.
Results
High levels of activity were recorded around joints SJ8699/18 SJ and TRIF JT 8SJ699/18.
The highest signals detected appeared to emanate from the Stuart St end of SJ8699/18 SJ
and were up to 500mV in peak to peak magnitude. An example signal is shown in figure A10.
Figure A10: TEV signal from joint SJ8699/18 SJ.

The probe was placed at various points on and between the joints. The activity was measured
at each location and the distribution of activity is shown in figure A 11.
Figure A11: Relative levels detected around J1 and J2.

Similar tests were carried out on TRIF JT SJ8699/27. Activity was again recorded with the
TEV but levels were approximately one tenth of those found at SJ8699/18 SJ. This difference
in magnitude is due to differences in screening.
Copyright HVPD Ltd.
Summary and Conclusions

The installed OSM (On-line Substation Monitor) at Stuart St detected partial discharge on the
Queens Park circuit through remote analysis of data acquired. This pd exceeded activity level
thresholds and an alarm was raised. High resolution data from the circuit was downloaded to
the IPEC server and analysis of the data confirmed the presence of partial discharge activity.
The Customer were alerted and a series of tests on the circuit were set up.
Tests were carried out at Stuart St and Queens Park substations. Using the HVPD-Spot
TM
TM
Tester Unit with PD Gold and PD Map Software the pd activity was established to be from
two sources of similar magnitude on the red phase of the cable. They were calculated to be
respectively 1564m and 1681m from Stuart St.
Three trifurcating joints were excavated at the location sites and using TEV signal
measurement were successfully identified as the discharging joints. The accuracy of the
location was approximately 0.4% and 0.7% of the cable length for the respective joints.
As the PD activity was sufficiently large to put the security of the circuit at risk the joints and
the cable between the joints were replaced.
Further tests were carried out at Stuart St after the joint and cable replacement and these
showed that all discharge activity had completely disappeared and that the identification and
replacement of the active joints had been successful.
Copyright HVPD Ltd.
Figure A12: Cable Feeder Route Length Diagram
Copyright HVPD Ltd.
Appendix 2: Velocity of Cable PD Pulses for different types of HV Cables

There can be a wide range of PD Pulse Propagation speeds along the conductor or earth
sheath of HV cables.
The speed of propagation is dependant upon the Relative
Permittivity/Dielectric Constant ( of the cables insulation.
The Absolute Velocity of Propagation, Vf is given by
Where:
s is the velocity of light in free space (300 m/s) and is the relative permittivity of the
dielectric.
The term velocity factor is sometimes used instead of the velocity of propagation.
velocity factor is the ratio of the velocity of propagation to the velocity of light.
The
For cable PD mapping and/or fault location testing it is common to use the Return Speed of
the pulses as opposed to the Absolute Velocity. The Return Speed is the Absolute Velocity
divided by 2 and is used in practice because when measuring the time between received
pulses the test engineer is measuring the time for the pulse to go there and back to the far
end of the cable.
If both the Time of Flight of the PD signals up and down the cable and the cable length is
known than the Absolute Velocity of propagation and Return Speed can be calculated.
Where the cable length is not known the distance to any feature can be calculated from:
tf
D Vf
2
Where:
Vf
tf
=
=
absolute velocity of propagation for the cable

time to the feature
The measured time is divided by 2 because the measured time is the there and back time for
the pulse.
Insulation
Type
Typical
Dielectric
Constant
Loss
Tangent
(Tan )
For Good
Cable
(radians)
PVC
(type T1)
XLPE
EPR
Paper
(Oil Rosin)
Paper
(MIND)
Vf - Absolute
Velocity of
propagation of PD
Pulses in cable
s
Return Speed
(Vf / 2)
(m/s)
(m/s)
6.0 - 7.0
0.08
113 122
57 - 61
2.3 - 5.2
3.2 3.8
2.5-3.6
0.0003
0.01
0.003
132 198
154 168
158 - 190
66 99
77 - 84
79 - 95
2.4-3.5
0.002
160 - 194
80 - 97

HVPD Longshot PD Mapping System User Manual Sep 2009

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

HVPD Longshot PD Mapping System User Manual Sep 2009

Загружено:

Авторское право:

Доступные форматы

High Voltage Partial Discharge Ltd

Empress Business Centre

HVPD On-line Cable PD Mapping

Stuart St - Queens Park (tee Eastlands)

HV Cable PD Mapping (location) using the HVPD-Longshot PD Spot

PD Map Software Manual Version 4 - September 2009

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Introduction to PD Mapping Methodology and Hardware... 3

Requirements for Portable Transponder (Trigger and Pulse Booster).. 4

Portable Transponder Hardware (PD Trigger and Pulse Booster Units)... 6

Tips for using the Portable Transponders for PD location...12

PD Map Software for use with the HVPD-Longshot..........14

PD Map Software - Main Capture Window..16

PD Map Software - Data Input Window.19

PD Map Software - Data Processing Window.21

Display and Data Processing of the collected data21

PD Map Software - PD Map Window23

Display and Analysis of the PD Map..23

Appendix 1: Case Study of use of PD Map Software and Transponder..25

Appendix 2: Velocity of Cable PD Pulses for different types of HV Cables..36

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Introduction to PD Mapping Methodology and Hardware

Figure 1: Single-Ended PD Site Location Method

Figure 2: PD Pulse Trains as seen from the Measurement End

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Requirements for Portable Transponder (Trigger and Pulse Booster)

Figure 3: Portable Transponder Unit Type PTT 2000-CT

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Reflected, Transponded pulse

Figure 4: Example of waveform with Transponded Pulse

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Figure 5: Using the Manual Transponder with Transponder Delay

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

PTT 2000-CT Discharge Trigger Unit

PTT 2000-CT -Discharge Trigger Unit

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

BNC Output (PULSE OUT)

Trigger Unit Set-Up - General

PTT 2000-CT Portable Pulse Generator Unit

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Figure 7: PTT 2000-CT PORTABLE PULSE GENERATOR and DISCHARGE TRIGGER

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Transponder & HFCT Connections For On-Line Cable Mapping

Standard Test Set-Up for Double-Ended On-Line PD Cable Mapping

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Tips for using the Portable Transponders for PD location on HV Cables

HVPD-Longshot - PD Map Software Manual v3 May 2009

Copyright HVPD Ltd.

Portable Transponder (PD Trigger Unit and Pulse Generator) Specification

Free Run mode

100V into 50Ohms

HVPD-Longshot - PD Map Software Manual v3 May 2009