Diagnostic Insulation Testing

Introduction

A major cause of equipment failure is breakdown of the insulation. A range of diagnostic techniques can be applied
to monitor insulation condition in order to take timely maintenance action. Each test gives one perspective on the
overall condition of the equipment and their combination can form a more complete picture than is provided by a
single test.

In the practical operational world there is not the time to completely analyse a piece of equipment and choices have
to be made on the appropriate test strategy for different types of equipment.

Insulation Degradation

There are 5 basic initiators for degradation of the insulation; electrical and mechanical stress, chemical attack,
thermal stress and environmental contamination. Normal cycles of operation will lead to 'aging' through these
mechanisms. The aging of insulation is a slow process of degradation as these factors interact with each other in a
gradual spiral of decline.

The rate of this insulation deterioration is of increasing interest to engineers dealing with electrical supply networks
such as those in Europe and the USA, where large parts of the network were installed 30 or 40 years ago in a burst
of investment in the infrastructure.

Testing

As insulation deteriorates leakage current may increase, the dielectric loss characteristic will change and the degree
of polarisation will alter. Insulation tests look for a change in one of these responses to indicate the deterioration.

Go/NoGo Diagnostic Insulation Tests

The traditional insulation resistance test is the simplest way to gain an overall indication of the condition of the
insulation. Although the Insulation Resistance test can be applied as a simple Go/NoGo test, it can also be used to
give more extensive diagnostic information. The most common voltages applied for non-destructive d.c. insulation
tests are 2.5 and 5 kV (defined in BS6266 and IEEE43).

'Spot' Test

This is the simplest insulation test, giving a reading of Insulation Resistance in MW. The test is applied for a short,
but specific period of time, after which a reading is taken. The time is typically 60 seconds (allowing the capacitive
charging current to reduce).

On installation of the equipment, these readings will be compared to the required minimum specification. Insulation
resistance is significantly temperature dependent and correction factors can be applied to show a trend more clearly
from a history of insulation tests.

The factor to correct the temperature to 40C is given by; Rcorrected = Kt x Rtest where Kt, the compensation factor,
doubles for each 10C rise

d.c. Insulation Test Currents

The test current in the body of the insulation can be split into three components; the capacitance charging current,
the polarisation (or absorption) current and the conduction or leakage current. In addition there may exist a surface
leakage current, which it may be advantageous to remove from the test result using the guard terminal of the
Insulation Tester.

Capacitive Current
This is initially large, but goes to zero as the test piece is charged.

Polarisation Current

Caused by charges in the insulation material moving under the effect of the electric field or by molecular di-poles
lining themselves up with the applied field (orientational polarisation). It is greatly affected by moisture or
contamination in the insulation, as the water molecule has additional orientational polarisation. This process takes
much longer than the capacitive charging.

Leakage current

Steady (resistive) current through the insulation, which is usually represented by a very high resistor in parallel with
the capacitance of the insulation.

Figure 1. Insulation Test Currents

It is immediately apparent that the Insulation Resistance reading is time dependent. In general, it is mainly the
capacitive current that is seen in the first seconds after the test

is started, as the needle on your MEGGER® Insulation Tester climbs. Then, at the one minute period most
commonly used for a 'spot' test, the current is a combination of polarisation & leakage current. By 10 minutes you
are reading mainly leakage current, though it can take up to 30 minutes for polarisation effects to be complete.

Discharge Currents

During the discharge phase the reverse occurs with the exception of the leakage current. There is no test voltage so
the leakage current is insignificant. The capacitive current decays quickly and the re-absorption current takes several
minutes, and perhaps hours, to reduce to zero.

Time-Resistance Tests (PI & DAR)

Time-Resistance tests take successive readings at specified times and have the great advantage of being independent
of temperature. They also help in the situation where past test records are sketchy, as, being a ratio, they are also
independent of the size of the equipment, although it is always more valuable if a trend can be established. This is a
lot easier with the time-resistance (Polarisation Index and Dielectric Absorption Ratio) tests than with Spot Tests, as
temperature correction is unnecessary.

Good insulation generally shows an increase in resistance over the 10 minute period, but with contaminated
insulation the polarisation effects are masked by high leakage currents and the readings are flat.

Table 1. Test Result Analysis

Insulation PI DAR TC DD
Condition

10 minute 60 seconds Discharge current

Definition Resistance x Capacitance
1 minute 30 seconds (at 60 s)/(V x C)

Poor <1 <1 <100 >4

Questionable 1 to 2 1 to 1.4 100 to 800 2 to 4

OK 2 to 4 1.4 to 1.6 800 to 2000 <2

Good >4 > 1.6 >2000

These values can give a guide to condition of insulation, although the figures are better interpreted in the context of
the equipment history. If a PI falls by 30% or more, then remedial action such as cleaning, or further investigation,
should be considered.

Time Constant (TC)

The time constant of the insulation is related to the basic characteristics of dielectric constant and resistivity, as well
as having some influence from polarisation. Trending this simple parameter allows degradation to be monitored,
with a reduction in TC showing problems in the insulation.

Discharge based tests

There are a range of techniques looking at the response of the insulation during its discharge. These tests have all
targeted the polarisation behaviour of the insulation, which is very sensitive to moisture. As all three components of
current (charging, polarisation and leakage currents), are present during the charging phase of an insulation test, the
determination of polarisation or absorption current is potentially confused by the presence of the capacitive and
leakage currents. The discharge phase of the test can more rapidly remove these effects, giving the possibility of
interpreting the degree of polarisation of the insulation and relating this to moisture and other polarisation effects.

The Dielectric Discharge (DD) Test

This test operates during the discharge of the dielectric, but is set up to be a simple and practical test. The DD test
was developed for generators, by EdF in France.

The insulator is charged for a sufficient time to be 'stable' (usually 30 minutes) This means that charging and
polarisation are complete and the only remaining component of the current is leakage current. The insulator is then
discharged and the resulting current is measured. This current constitutes the capacitive discharges and the
'reabsorption currents', combining to give the total 'dielectric discharge'.

In order to make comparisons between equipment, the current is measured after a standard time of 1 minute, which
is much greater than the primary time constant of the capacitive discharge. The result is not affected by surface
leakage, which is effectively short-circuited.

The resultant current is dependent on the overall capacitance (C), the final test voltage (V) and the degree of
polarisation of the dielectric. During the discharge, the voltage and capacitance of the equipment are measured so
that the 'Dielectric Discharge' can be quoted as a simple number;

DD = I1 min / V x C (mA/V/F)

The Dielectric Discharge can identify absorbed moisture in an insulation as this affects the absorption behaviour of
the dielectric and is masked by leakage effects if we try to measure it on the charging cycle.

a.c. based tests

In a.c. testing, the polarisation current never has time to die away and, together with capacitive current, predominates
in the total dielectric loss. It is therefore also sensitive to moisture and degradation of the insulation material.

Some a.c. tests do not require a high voltage test signal and so do not stress the insulation. This means there is very
little chance of damaging the dielectric.

Tan-Delta (Power Factor) Testing

The loss angle / tan delta / power factor of the insulation is more sensitive to small changes in the condition of the
insulation than the raw d.c. insulation resistance and in some circumstances this may be important. It is therefore a
very useful test for monitoring insulation condition from initial installation. Comparing previous results is essential
to be able to spot deterioration; a single result cannot give much information.

The use of an a.c. source (often 2.5 or 10 kV) makes the Power Factor test set heavier and more expensive than a
d.c. tester. Also there is a limit to the capacitance that can be tested, but the early warning given by monitoring this
parameter makes it invaluable.

Summary

Diagnostic insulation tests allow more information to be gained from an insulation test which may prove invaluable
in a maintenance program. An a.c. test also provides early information that can point to future insulation problems.
Each type of test helps to give a more complete picture of the condition of the insulation.

Table 2. Summary of Diagnostic Insulation Tests

 Test Type Definition Application

Insulation Resistance Single insulation value, often after General condition, need
(Spot test) 60 s temperature compensation

Ratio of 10 min.:1 min. insulation Level of dirt/moisture, less

Polarisation Index (PI) values temperature dependent
d.c.
charging Dielectric Absorption Ratio of 60 s:30 s insulation Quicker version of PI
tests Ratio (DAR) values

Time Constant Resistance x Capacitance Figure of merit

Comparison of insulation at
Step Voltage (SV) Presence of cracks & holes
different test voltages

Dielectric Discharge Discharge current after 60 s/(Volts Level of absorbed

(DD) x Capacitance) dirt/moisture

Recovery Voltage Discharge for 50% of charge time. State of transformer insulation
/Polarisation Spectrum Measure peak, initial slope and including moisture in oil and
d.c. Analysis recovery voltage paper
discharge
tests
Isothermal Relaxation Discharge characteristic after 60 s XLPE cable insulation
Current Analysis charge at 1 kV condition

Current & voltage characteristics Motor & generator insulation

EDA Test + capacitance condition

Type tests with go/no-go trip

Flash (Hi-Pot) Leakage current (a.c.) levels

Measures small charges caused by Usually used for component

Partial Discharge voltage stress testing
a.c. tests
Power Factor/'Tan Difference in the in-phase and Very sensitive to insulation
Delta' out-of-phase current moisture and degradation

VLF (Very Low Insulation value at 0.1 Hz Non-destructive cable testing

Frequency)