Cable fault location in power cables

Fault classification:
Insulation and resistance measurement
Contents:

1. Introduction
2. Insulation test
3. Measuring the capacitance of the test object
4. Measuring the resistance of a fault
5. Time-dependent resistance test
6. Overview measurement with TDR

1. Introduction

The first step in locating a cable fault – fault classification – should be very thorough. Exact
cable plans, knowledge of the cable network as well as any civil engineering work can
provide the first clues regarding the location of the fault and prevent any misinterpretation.
The following insulation testing and resistance measurement can supply information about
the characteristics of the fault.
After these measurements are performed a reflection measurement device (TDR) is
connected in order to detect the cable length, sleeves or other changes in impedance. One
should always compare conductors which are faulty with conductors which are not faulty.
The more information gathered this way, the easier and more reliable the overall fault
location process will be.

2. Insulation test

The insulation test shows the type of fault. To do this, the ohmic insulation resistance
between the conductors and shield (phase – phase or phase – shield) is measured.
The results of the insulation test are important in deciding how to proceed further with the
fault location process and can be classified as follows:

No fault (no deviation between the resistance values)

High resistive cable fault (kOhm, MOhm)

Low resistive cable fault (contact between two conductors/screen)

2
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06
Normally, the insulation test can be performed directly from the cable test van. The
maximum voltage is 1000 V. If higher voltages are required, individual apparatus must be
used on site. Equipment with an analogue display has proven reliable in practice. For
example, anomalies during the charging process can show up faults due to the presence
of moisture. These tendencies can only be detected using pointer instruments.
With faults of very high resistance, direct current must be used to establish the level of
breakdown voltage in the cable fault.

Equipment: Insulation testers 500 V, 1 kV, 2.5 kV, 5 kV,

integrated insulation testers up to 1 kV

3. Measuring the capacitance of the test object

For the measurement of the insulation resistance and test object capacitance, the
following settings can be adjusted in the system Centrix:
In automatic mode, both, resistance and capacitance are measured. A selection must
only be made between the upper test voltages of 500 V and 1000 V. The system
automatically switches to low voltage when determining a low-ohm resistance, since the
resistance values in this measuring range can only be measured with low voltage.
In manual mode, you may only determine either the resistance or the capacitance during
a test. The three test voltages of 500 V, 1000 V and low voltage (<6 V) are available for
selection. No automatic switching between the voltages occurs during the measurement.

The measured parameter, date and time of measurement as well as the test voltage used
can be read from the table header for each measurement (column).

3
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06
4. Measuring the resistance of a fault

When choosing the pre-location method to be used, it is very important to have precise
knowledge about the resistance of the fault and the position of the fault. The results of both
measurements should be saved or noted. With multiple faults, the faults are often parallel.
A drawing can help during evaluation.

2 Ω

10 k Ω

E E
10 kΩ 10 kΩ ∞ 2Ω RRis ∞
o-

In solely plastic low-voltage cable networks without shielding or armouring, it is

recommended that the PEN be disconnected on all sides and a check then performed to
see if there are any faults due to contact with the earth.

Equipment: Low-resistance ohmmeter, up to 10 ohm

5. Time-dependent resistance test

Through a resistance measurement over time, the chronological change of the absorption
characteristics and thus the degree of moisture and dirt in an insulation can be tested.
A continual rise of the recorded resistance indicates an intact insulation. A flat or
downward sloping curve on the other hand can indicate a dirty, moist or damaged
insulation.
To obtain comparable measurement results, the chronological progress can be used to
calculate the coefficients PI (Polarization Index) and DAR (Dielectric Absorption Ratio).
For the DAR coefficient, the value measured after one minute is divided by the measured
value after 30 seconds. This coefficient should therefore be used primarily for evaluation of
newer insulating materials, which exhibit a faster decline of dielectric absorption currents.
For other insulation materials, with absorption characteristics that normalise more slowly,
the PI coefficients should be determined. For these, the value measured after 10 minutes
is divided by the measured value after one minute.
The determined resistance values are shown on the screen during the course of the
measurement as a curve over time.
With the completion of the measurement, the coefficients that have been determined (DAR
and / or PI) are shown in a dialog box.

4
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06
The course of the curve itself as well as the determined coefficients can provide
information on the state of the insulation.

Example of a resistance curve of a faultless insulation (DAR value of approx.: 1.5)

Example of a resistance curve of a faulty insulation (moisture, dirt) (DAR value <1.2)

The following table provides generally accepted guideline values which may be used for
evaluating the measurement results:

PI value DAR value Condition of

insulation
<1 <1 poor
1…2 1 … 1.3 questionable
2…4 1.3 … 1.6 good
>4 >1.6 excellent

A comparison with intact cables of identical construction or with previous measurements

should also be taken into account.
To do so, previous measurements can be called up from the history database and the
resistance curves can be compared. By selecting the menu item, the determined DAR
and/or PI coefficients can also be displayed for comparison.

5
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06
Technical dates Centrix System

Voltage Uiso=500V, Measuring range: 500k…2GΩ (gem. IEC 61557-2)

Range [Ω] Resolution [Ω] Accuracy

1,00k…9,99k 0,01k
10,0k…99,9k 0,1k
100k…249k
1k
250k…999k
1,0M…9,99M 0,1M ± (3% of display)
10M…2G 1M

Voltage Uiso=1000V, Measuring range: 1M…2GΩ (acc. IEC 61557-2)

Range [Ω] Resolution [Ω] Accuracy

1,00k…9,99k 0,01k
10,0k…99,9k 0,1k
100k…499k
1k
500k…999k
1,0M…9,99M 0,1M ± (3% of display)
10M…2G 1M

Resistance

Range [Ω] Resolution [Ω] Accuracy

0,0…9,9 0,1 ± (3% of Display)

10…999 1 ± (3% of Display)

Capacity

Range [µF] Resolution [µF] Accuracy

0,0µ…19,9µ 0,1µ ± (5% of Display)

6
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06
6. Overview measurement with TDR

The following overview measurements should be carried out with a reflectometer (TDR):
- Comparative measurement – length measured according to length on cable plan,
if necessary, correction of diffusion speed V/2
- Sleeve calibration
- Comparison of faulty and non-faulty wires
- Saving reflectograms for later comparison

Equipment: Teleflex MX (Centrix)

Teleflex T 30-E
Digiflex Com, Easyflex Com

Results of fault classification

Short circuit, 0 ohm

A short circuit is a direct, metal connection between conductors, i.e. these conductors
have fused together or are touching one another. This results in all the acoustic pinpointing
methods failing; due to the direct metal contact no breakdown noises can be produced. On
the other hand, a short circuit can be very easily seen using classic reflection measure-
ments. Nevertheless, an attempt should be made to send a shock of a higher resistance to
this fault – with a high-power shock discharge generator – so that acoustic pinpointing can
be undertaken.

Resistance fault greater than 0 ohm

Resistance faults do not allow the cable to be charged. However, these faults are visible
when almost any of the pre-location or pinpointing methods based on high voltage are
used.

Very high resistive faults

The resistance of these faults is so high that, in many cases, ignition using the normal
voltage of a shock discharge generator can no longer take place. These faults are charged
up to the flash-over voltage. The entire energy stored in the cable capacitance is
discharged via the fault. DECAY and DECAY-Plus pre-location methods as well as
acoustic pinpointing are possible.

Faults due to contact with earth

Pre-location using the bridge method and the voltage drop method

7
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06
Which faults can be seen with the reflectometer (TDR)?

The following are visible The following are not visible

-All impedance changes below cable -Faults of which the resistance is in a
impedance range many times the impedance of the
e.g.: parallel and series ohmic faults cable. In theory, these faults are visible
-Sleeves but the visible change is so small that it
-Strong deflections gets lost in the normal interference level
-Damaged areas and pressure points or disappears due to the insulation.
-Incoming water With modern cables and correctly fitted
-Changes in cross-section sleeves, these changes in impedance
-Contact problems caused by corrosion can be so small that they can no longer
be detected.
-Faults that normally do not exhibit any
resistance (R=infinite). These faults
(spark gaps) are ignited by applying an
DC or VLF voltage. The voltage level
depends on the defective insulation.
-Where a cable of unknown length has
severed – a mix-up between cable end
and severed point is possible. When
there is doubt, short-circuiting at the far
end can at once clearly confirm that
there is such a fault present.

Depending on the type of fault, the visible reflections can be so small that they are
inconspicuous and are therefore undetectable.

Remark: Further information can be found in the following article on reflection

measurements

8
Seba Dynatronic Mess- und Ortungstechnik GmbH Registergericht, D-96045 Bamberg, HRB-Nr. 195
[email protected] – www.sebakmt.com All rights reserved 2010/06