6 Prelocation Transient Methods En

7/25/2019 6 Prelocation Transient Methods En

1/13

Faultlocationinpowercables

PrelocationTransient Methods


2/13

2

Contents:

1. Introduction2. DECAY (voltage decoupl ing)3. ICE (current decoupl ing)

1. Introduction

In the case of transient methods a breakdown is generated at the fault location. Itburns for a few milliseconds and is a very low-impedance short circuit during this

period. Due to this abrupt breakdown a travelling wave spreading within the cable isgenerated during ignition. It spreads from the fault location towards both ends and isalso reflected to the fault there, while the still-incoming burning arc prevents thetravelling wave from passing the fault location. Just like when using a standardreflectometer pulse, it is again reflected at this spot by the still burning short circuitwith polarity reversion. From on particular end only the travelling wave of thatparticular section of the cable up to the fault is visible. There are different ways totrigger these transients and there are also different methods to decouple andevaluate them later on.

The transient waveA transient wave always contains a voltage component and a current component.The voltage and the current components behave differently in closed cable ends(short circuit, open).The current pulse method only evaluates the current component of the transientwave (current wave).

The open ended cableWhen a transient wave arrives at the open end of the cable, due to the infinitely highimpedance mismatch a total reflection of the current and voltage waves is generatedhere. Figure 1 shows how the incoming voltage wave is reflected with the same

amplitude and polarity at the open cable end. The resulting voltage amplitude of theincoming and returning wave can reach up to twice the value of the applied voltage.


3/13

3

Figure 1: The transient voltage wave

The current wave which initially arrives at the cable end in phase with the voltagewave is also reflected with the same amplitude, but with opposite polarity, so that

both current wave parts cancel each other at the cable end.

Figure 2: The transient current wave

The cable with a short circui t at the endIn this case the current wave is reflected with the initially same amplitude and polarityso that the current amplitude increases up to twice the value of the incoming wave.However, here there is a polarity reversal in the voltage wave which causes thevoltage at the short-circuited cable end to be zero.

2. DECAY (voltage decoupl ing)

The simplest and probably the oldest method is the travelling wave pre-localisation,in short the DECAY method. For this purpose the cable is charged with direct voltageuntil the voltage exceeds the breakdown voltage of the fault. The energy stored in thecable capacitance discharges itself via the fault and creates a travelling wave thatcan be measured as damped (decaying) oscillation. The length of this oscillationcontains the distance to the fault; but it also includes test leads. During this processthe signal is decoupled from the hot part of the high voltage via a capacitive coupler.

+Uinc.

+Ureturn

+U

+2U

cable start cable end

i-inc.

i-return

+i +i

cable start cable end

-i


4/13

4

Description of the processTotal length of an oscillationAt the moment the breakdown takes place, the cable is negative charged with thetest voltage. Therefore, the discharge created by the fault creates a travelling wave

with positive rising slope. The front runs with a velocity of approx. 160 m/s from thefault to the cable start where it meets the very high output impedance of the HV testdevice. Consequently, a reflection with unchanged polarity takes place. Now thetravelling wave returns from the cable start to the fault where the breakdown ark isstill burning. This short circuit reflects the travelling wave. This time the polarity isreversed, i.e. the travelling wave now has a negative front.This process is repeated until the energy of the travelling wave is attenuated or theburning breakdown arc is extinguished.Due to the spreading behaviour of the travelling wave the fault distance is includedtwice in every oscillation. Therefore, the length of an oscillation must be divided by 2and subsequently, the length of the test leads must be subtracted from the

calculation.

Fault distance = leadstest2

noscillatioanoflenghtTotal

For preparing the measurement it is very helpful to adjust the distance range to 5 to10-times the known cable length and to reduce the amplification at least by factor 2 inrespect to a normal reflection measurement before starting the measurement. Thisensures that the first shot is a hit providing a result which only requires a minimumamount of correction.

The advantage of this technology is, that the voltage is nearly unlimited. Pre-localisations of up to 400 kV and even higher have already been achieved with thismethod.A disadvantage to be mentioned is, that due to the high-frequency of the travellingwave, in combination with the very high voltage, additional damage to cables that arealready aged by operation, can happen. In such cases it is recommended to proceedvery carefully to avoid possible consequential damage.


5/13

5

Figure 3: Basic block diagram of the DECAY methodThe transient voltage wave is reflected with unchanged polarity in the HV test device and movesback to the arcing fault. This wave is then reflected at the arc with reversed polarity.

Figure 4: Decay travelling wave decoupling on an 8 km cable with marker and cursor

=

G

50m

TDR


6/13

6

Figure 5: Reflectogram of a fault location measurement with the DECAY methodThe software of some reflectometers already performs the division by 2 automatically andsubtracts the test leads.

Reflections are caused each impedance change, e.g. joints, connection points of the test leads tothe test object.

Figure 6: Decay travelling wave decoupling on 8 km cable with shift technology


7/13

7

3. ICE sing le-phase current decoupl ing

This method is comparable to the Decay, but does not work capacitive but inductiveby decoupling the current. Instead of using a capacitor for decoupling, a coil or

transformer is used. Typically Rogowski coils are used. The standard ICE method isdone with a surge generator. A capacitive discharge of the surge generator via thesurge switch triggers and ignites the fault and causes it to a flash over. This flashoverresults in a travelling wave shown on the Reflectometer.

For evaluation, the second largest reflection is taken as reference. Any reflectionsvisible before this reflection are delayed by the ionisation time and represent theprocess during which the fault begins to ignite. Furthermore, an attenuated transientwave subsequently travels back and forth between fault and surge generator. Herethe capacitor of the surge generator as well as the arc at the fault represent a shortcircuit for the high-frequency wave. The result is an oscillation where the period

length corresponds directly to the fault distance. The test leads must be subtracted todetermine the fault distance.

Fault distance = length of an oscillation test leads

The most accurate measurement results are achieved when the oscillationmeasurement is done at the zero crossings. Another alternative is the shifttechnology, where an identical copy of the trace is side shifted until the nextoscillation is completely overimposed with the original trace. For preparing themeasurement it is very helpful to set the distance range to 5 to 10-times the cable

length before starting the measurement. However, in this case the amplificationshould be increased in respect to the normal reflection measurement since thesignals received from the decoupling coils are significantly weaker. An advantage ofthe ICE technology is that the decoupling coil is in the earth path of the cable. It is notexposed to any high voltage. Therefore, no complex insulation design is necessary.The coil itself is very small and can be integrated in almost any device.

As for all transient methods, the disadvantage is that there is only information aboutthe fault distance without any other details on the cable or the fault itself. Theprecision of transient methods is partly limited and cannot be compared with genuinereflection methods.

Current measuring method

Another possibility is charging the cable up to the breakdown voltage with the surgegenerator with closed surge switch and then using the cable capacitance itself as acapacitor. With this method the available surge capacitance can be significantlyincreased and this is particularly helpful with very long cables. The surge energydoes not have to run from the surge generator to the fault as it is already provided bythe cable and its charged capacitance. There is no ionisation time.. The remainingmeasuring process is completely identical to the normal ICE current decouplingprocess.


8/13

8

Figure 7: Basic diagram for the ICE process single phaseThe fault does not ignite transient current wave is reflected at the open cable endwith the same amplitude, however with opposite polarity.

Figure 8: Reflectogram of a fault location measurement with the ICE method single phaseThe fault does not ignite reflection at open cable end

tx tx

parasiticreflections

TDR

tx


9/13

9

Figure 9: Basic schematic for the ICE process short circuit

Figure 10: Reflectogram of a fault location measurement with the ICE method single phaseshort circuit reflection of the transient current wave at the fault location without ignition delay,amplitude and polarity remain the same

tx tx tx parasiticreflections

TDR

tx


10/13

10

Figure 11: Basic schematic for the ICE process single phase flashing fault

Figure 12: Reflectogram of a fault location measurement with the ICE method single phaseflashover reflection at the fault location with ignition delay

Figure 13: Reflectogram of a fault location measurement with the ICE method single phaseflashover ignition delay is so extensive that the reflection first takes placeat the cable end

tx+ tx tx


t

tx tx tx


t

tparasiticreflections

TDR

tx


11/13

11

The ICE current decoupling processes have proven their suitability in particular withfaults in the lower kOhm range and very large fault distances where it is not possibleto get results with the ARM process. However, it is always a requirement that the arc

lasts for a few milliseconds. The capacity of the surge capacitor and the impedanceof the cable determine the pulse widths of the travelling wave in the s range. Theseflat slopes caused by the wide pulses as well as by the attenuation of the cablereduce the accuracy of the results. Usually, the fault distances determined that wayare up to 5 to 10 % longer. For exact and fast fault location process, themeasurement technician should walk from the pre-localised distance into thedirection of the fault location system. The test leads must be subtracted from theresult as for all transient measurement methods.

Figure 14: ICE automatic measurement

large measurementrange


12/13

12

Figure 15: ICE automatic measurement and manual correctionReflections are additionally created at each impedance change, e.g. joints, connection points ofthe test leads to the test object.

Figure 16: ICE current decoupling on an 8 km cableThe ignition delay time can be seen very clearly. The measurement should always take placebetween the 2

ndand 3

rdpeak/wave or their multiples.

fault distance


13/13

13

4. Three-phase current decoupl ing

These fault location methods for measurements in branched (T-eed) networks, inparticular in branched medium-voltage networks, will be part of one of the nextreports on fault location in power cables.

Additional focus in the next applications will be on testing and diagnosis as well asthe application of the tools.

6 Prelocation Transient Methods En

Documents