EVENTS

DIAGNOSIS

ABSTRACT
Part 1 of this article, published in Vol­ Diagnostic
measurements
ume 3 Issue 4, pages 100ff, describes
the measurements of excitation, wind­
ing resistance, turns ratio and accu­
racy as the most common diagnostic
measurements on instrument trans­

on instrument
formers (current and voltage trans­
formers) for condition and reliability
assessment. Case studies show the
failures which can be derived from the

transformers
results and underline the importance
of conducting regular diagnostic
tests. Part 2 gives more details about
the measurements of capacitance and

– Part II
dissipation/power factor, short circuit
impedance, dielectric response analy­
sis and partial discharge.

KEYWORDS
instrument transformer, electro-mag­ A classification and overview
of diagnostic measurements
netic circuit, insulation, diagnostic
tests

96 TRANSFORMERS MAGAZINE | Volume 4, Issue 1

 Florian PREDL, Dr. Michael FREIBURG, Dr. Martin ANGLHUBER

The capacitance and dissipation/power

factor measurement is a well-established
method to evaluate the insulation condi-
tion

constant greater than 1. In addition, the all parts of the insulation. The main in-
insulation material has a certain conduc- sulation, which is located between the
tivity which creates conductive losses [7]. individual turns of the primary winding,
A dissipation factor measurement meas­ cannot be accessed for measurements.
ures a combination of these losses, see However, the dissipation/power factor
Figure 9. can be measured between the primary
and secondary winding, as well as be­
A voltage tip-up test (ramping up of tween the primary winding and ground. If
the test voltage) can be used to check the transformer is equipped with a screen
whether or not there is any PD activity electrode, the measurement between the
present. An increase in the dissipation primary winding and the screen is the
factor at a certain inception voltage preferred measurement method. It de-
indicates possible PD activity. This is a pends on the type of the IVT if a screen
common diagnostic tool on generators is equipped and if it is accessible in the
and motors. However, a dissipation factor secondary terminal.
measurement does not give an exact
localization of PD. It can only give an 8. Short circuit impedance
overall representation of the insulation measurement
condition.
A CVT must have a compensation reac-
A capacitance and dissipation/power tance, often called reactance coil (Lcomp).
factor measurement on the capacitive This coil compensates the phase shift
stack of a CVT can reveal any possible caused by the capacitor stack. Hence, the
insulation degradation or even short­ reactance of the coil is tuned to the re-
ed capacitive layers. The physical con­ actance of the capacitor stack at line fre-
struction of the capacitive stack is quency. The coil is typically operated at
7. Capacitance and similar to that of condenser bushings. around 10 kV – 30 kV, depending on the
dissipa­ti­on/power factor If a capacitive layer should break down, manufacturer. In Figure 10 a simplified
measurement the overall capacitance of the stack will electrical diagram of a CVT is shown.
increase. Shorted coil turns cause the inductance to
The dissipation factor is measured by drop. The capacitor stack is therefore no
comparing the current of a test object Likewise, if the dissipation factor longer properly compensated, leading to a
to a known reference (“ideal” capacitive increases, it is an indication of an aging drift in the phase displacement.
current). The phase difference between the process taking place (moisture ingress,
reference current and the test object current partial discharge, etc.).
is determined. Calculating the tangent of δ
gives the dissipation/loss factor. Leakage currents through the insulation
of a CVT winding often lead to difficulties
The capacitance and dissipation/power in obtaining a balance of dissipation
factor measurement is a well-established factor. This means that the apparent
method to evaluate the insulation condi- dissipation factor readings are below
tion. An ideal (loss-free) insulation con- the true value, or even a negative value,
sists of a vacuum capacity also referred to although the capacitance value obtained
as the geometrical capacity C0. If insula- will be correct [8]. A change on the result
tion material other than vacuum is being of the measured capacitance results (C1
used, one or more polarization proces- in series with C2) from one routine test
ses can be observed. They represent the to another is a reason for additional
electrical behavior of the used insulation investigations.
material(s). Polarization processes cause
losses, for example due to a rotation of di- In the case of IVTs, an insulation capa-
poles. This will furthermore increase the citance and dissipation/power factor
capacitance measured due to a dielectric measurement cannot be performed on Figure 9. Insulation and its losses

w w w . t ra n sfo r m e r s - m a g a z i n e . co m 97
 DIAGNOSIS

(Figure 11). The reactive part of the com-

plex short circuit impedance should be
close to 0 Ω indicating that the capacitor
stack (C1 and C2) is properly compen­
sated.

8.1 Case study III – CVT accuracy

measurements

Two CVTs were investigated after one of

the two units revealed high gas levels after
oil sampling. The Dissolved Gas Analysis
(DGA) result indicated PD and arcing.

Both devices were measured to check

the integrity of the electrical circuit. The
nameplate information is shown in Table 2.

Figure 10. Capacitive voltage transformer – simplified electrical diagram The CVT with the elevated dissolved gas
results during oil sampling also showed
a much higher ratio error and phase
displacement. A closer look at the short
A short circuit impedance test at line frequen- circuit impedance test result confirmed
cy can be used to check the integrity of a CVT’s that the reactive part of the “faulty” CVT
showed capacitive behavior.
reactance coil
This confirmed that the reactance coil had
shorted turns. The capacitor stack was no
A short circuit impedance test at line fre- drop and the phase angle between voltage longer compensated at line frequency.
quency can be used to check the integrity and current is measured across the second­
of the coil. An AC current is injected into ary winding. The primary side (capacitor The ratio error and phase displacement of
the secondary winding while the voltage stack) must be short circuited to ground the faulty CVT are indicated in Figures 12
and 13.

9. Dielectric response
analysis
A high water content in the oil-paper in-
sulation of ITs can lead to a failure of the
insulation and, as a consequence, can
even result in the complete destruction of
the asset. Therefore, it is important to be
able to assess the ITs’ water content. This
proves to be quite challenging as, in con-
trast to power transformers, its measure-
ment techniques such as oil sampling are
quite difficult to perform due to the rather
small oil volume and often a lack of simple
Figure 11. Short circuit impedance measurement on a CVT and easy access to it.

Table 2. Nameplate information of the CVTs under test

Rated primary voltage 220/√3 kV Line to ground

Low-voltage terminals Secondary voltage (V) Ratio (to 1) Accuracy class Rated output (VA)
1a2 – 1a1 110/√3 2000 0.2/3P 0 - 100
2a2 – 2a1 110/√3 2000 0.2/3P 0 - 100
Total nominal capacitance 6200 pF C1 7106 pF C2 76393 pF

98 TRANSFORMERS MAGAZINE | Volume 4, Issue 1

Over the last few years the dielectric res­
ponse analysis has become well estab-
lished to assess the moisture in the solid
paper insulation. It is done by measuring
the power factor/dissipation factor over
frequency.

The measurement of the dielectric re­

sponse over a wide frequency range (for
example, 100 μHz up to 5 kHz) provides
information about the insulation condi-
tion and, especially for oil-paper insula­
tions, about the water content in the solid
insulation.

For calculating the water content, the

Figure 12. Ratio error of the faulty CVT
measured dielectric response curve is
compared to a modeled curve (Figure 14).
The curve modeling is done with help of
a database including material properties
of cellulosic material with different water
contents and temperatures. Using the so-
called XY model [9] a dielectric response
is calculated under consideration of the
insulation geometry, temperature, oil and
moisture content. A matching algorithm
aligns the modelled response of the data­
base to the measured curve of the real
insulation and automatically delivers the
water content of the cellulosic material as
well as the water saturation or the oil con-
ductivity.

The appropriate test setup on current Figure 13. Phase displacement of the faulty CVT
transformers depends on whether or not
the CT has a screen electrode and whether
or not the screen electrode is accessible.
Sometimes this is documented in the A dielectric response analysis assesses the
datasheet of the CT. In case there is no moisture in the solid paper insulation
information about the screen, contact
the manufacturer. Figures 15 to 17 show
proposed setups for these cases [10].

9.1 Case study IV – IT insulation test

Two combined ITs were investigated as

one unit showed a high concentration of
hydrogen in an oil sample. The concen­
tration was 699 ppm. There was no meth­
ane involved. Therefore, it was not very
likely that the high H2 concentration was
caused by PD activity.

The test results in Table 3 show a sum­

mary of the dielectric response values
measured. CT2 is the IT with a high
concentration of H2.

The ratio of the complex capacitance

measur­ed at 10  mHz and 50  Hz provides
further information on the insulation Figure 14. Calculation of the water content based on the XY Model

w w w . t ra n sfo r m e r s - m a g a z i n e . co m 99
DIAGNOSIS

conditions of ITs. Field studies have

Output shown that the capacitive ratio should be
CHL below 1.05 for a healthy and dry insula-
tion [11]. The advantage of the capacitive
V CH1
ratio is that this parameter is not depend­
A ent on the geometry of the insulation.
CH2
A The dielectric parameters obtained on
both ITs did not indicate any aged insu-
lation. The very high concentration of H2
was most likely caused by stray gas. The
transformer manufacturer was asked for
information related to the stray gas and
Figure 15. Test setup for a CT with an accessible screen electrode; measured insulation
HV to screen; guard applied to ground they revealed that a chemical reaction
between a certain detergent and the in-
side materials of the transformer could
Output
have produced the high H2 values. There-
fore, the tested transformer is still in ope-
CS
ration.
V CH1
10. Partial discharge
A
measurements
CH2
A PD is a localized dielectric breakdown of a
small portion of a solid or liquid elec­trical
insulation system under high-voltage
stress. PD only partially bridges the insu-
lation between conductors [12]. PD ac­
Figure 16. Test setup for a CT without accessible screen electrode; measured insulation t­iv­ity deteriorates the insulation material
HV to ground; no guard applied over time, which can eventually lead to a
total breakdown of the insulation.
Output
PD releases parts of the energy as an elec-
CHL tromagnetic wave. For PD measurements
a test circuit is installed so the shorted ca-
V CH1 pacitance is reloaded from the coupling
A capacitor. The current during reloading
CH2 can be measured and correlated to the dis­
A charge level. PD is measured in pC either
according to IEEE Std C57.13TM-2016
[13]) or according to the IEC  60270
stand­ard.

Figure 17. Test setup for a CT without screen electrode; measured insulation HV to se­ Figure 18 shows a PD measurement setup
condary winding; guard applied to ground according to IEC 60270 [12]. It involves a
blocking impedance, a coupling capacitor,

Blocking
impedance Coupling Table 3. Dielectric response results
z capacitor
CT1 CT2 
Ck
Tan(δ) @ 50 Hz 0.28 % 0.29 %
ut(t)

Ca
CD Oil conductivity 23 fs/m 22 fs/m

Moisture content 1.6 % 1.8 %

Test Coupling
Object device
C_10 mHz/C_50 Hz 1.02 1.03
Figure 18. PD Measurement setup according to IEC 60270

100 TRANSFORMERS MAGAZINE | Volume 4, Issue 1

Partial discharge meas­
ure­­ments reveal weak A

points in the insulation

before a total break- 2CS

down of the insulation Cp/2 CF ε0

εr
Cp/2
can occur
2CS

and a coupling device which is attached to

the PD measurement instrument. B

ITs for medium-voltage (1 kV up to 75

kV) applications typically have a cast resin A
insulation. Voids or cavities in this insula- IS(t)
tion can be a result of shock and vibration
or manufacturing faults. If the electrical
field strength in the insulation becomes I2(t)
higher than the dielectric strength of the I3(t) CS
gas inside the void, a total breakdown
R1
will appear inside the void. At this very
Ut
moment the electrical field in the void ex- CP
tinguishes. The dissipated energy will be U1
recharged by the coupling capacitor. The I1(t) S
CF
coupling device connected to the coupling
capacitor is able to measure the recharge
current. The recharge process depends on
the voltage gradient of the applied voltage. B
The process is fastest at the steepest part of
the voltage gradient. Therefore, PD often
occurs close to the zero crossing of the ap-
plied voltage (Figures 19 and 20). Figure 19. Recharge process explained on the principle of a void discharge

Figure 21 shows a typical phase-resolved

PD pattern (PRPD pattern) for a void
discharge happening inside a solid insu-
lation of a medium-voltage transformer.
The cluster represents a histogram of all Ut(t)
discharges recorded over 1 min 36 sec. In U1(t)
accordance with the IEC 61869-1 stand­ Uz
ard [14] and depending on the test volt­ UL U’t(t)
age, the discharge level should not exceed -UL t
50 pC, where in this case discharges up to
-Uz
several nC have been measured.

11. Conclusion
Different diagnostic measurements on
instrument transformers help to assess
their condition. Their results give valu­ I1(t)
able information about possible faults re-
lated to specific parts of an IT (Table 1). By
combining the test results of the various
diagnostic tests, an overall picture of the
t
health condition of the IT can be derived.
As a consequence, failures can be detected
before they turn into severe failures which
endanger people or result in costly dam­
age to connected equipment. Figure 20. Recharge process explained on the principle of a void discharge

w w w . t ra n sfo r m e r s - m a g a z i n e . co m 101
DIAGNOSIS

Figure 21. PRPD pattern of void discharges

Bibliography Authors
Florian Predl started with OMICRON Austria in 2007 as an
[7] A. Küchler, Hochspannungstechnik, application engineer within the Engineering Services team
Heidelberg 2009 with special focus on advanced instrument transformer
[8] Trench Instruction Manual Coupling diagnostics. He also provided technical support to world-wide
Capacitor Voltage Transformers, Bullet­ users of OMICRON products. In 2013 Florian joined the
ing 20 95 05, Revision 03, 02/2012 OMICRON team in Australia where he is currently employed
[9] M. Koch, Reliable Moisture Deter- as a Field Application Engineer. Before starting at OMICRON
mination in Power Transformers, PhD he attended the Federal Higher Technical Institute in Rankweil, Austria, where
thesis, Institute of Energy Transmission he graduated in 2007 with a focus on high-frequency technology. His final thesis
and High Voltage Engineering, University focused on range extension of RFID systems for business applications by using high-
of Stuttgart, Sierke Verlag Göttingen, Ger- frequency amplifiers.
many, 2008
[10] DIRANA application note: Measur­ Dr. Michael Freiburg is responsible for instrument transformer
ing and analyzing the dielectric response tests and diagnostic equipment and is currently working as a
of current transformers product manager at OMICRON electronics in Austria. Prior
to that, he worked as a research and teaching assistant at the
[11] M. Anglhuber, Isolationsdiag­nose
Technical University in Dortmund, Germany. His research
an Messwandlern mit dielektrischer
interests include the diagnostics of high voltage equipment and
Antwortmessung, Diagnosewoche 2015
material science. In his undergraduate studies he focused on
in Austria
automation and control engineering before studying power engineering in his post-
[12] IEC 60270 Third Edition / 2000-12, graduate courses. He received an engineering degree in 2010 and his PhD degree in
High-voltage test technique – Partial high voltage engineering in 2014.
discharge measurement, Reference num-
ber CEI/IEC 60270:2000 Dr. Martin Anglhuber received his degree in electrical
[13] IEEE Std C57.13TM-2016, IEEE engineering from the TU München in 2007. From 2007 to
Stand­ ard Requirements for Instrument 2011 he worked as a scientific assistant at the Institute for
Transformers, New York High Voltage Technology and Power Transmission of the TU
[14] IEC 61869-1 Edition 1.0 / 2007-10, München, Germany and performed research on polymer
Instrument transformers – Part 1: Gen­ nanocomposites as insulating material in high-voltage
eral requirements apparatus. He received his Dr.-Ing. (Ph.D.E.E.) degree in 2012.
[15] IEEE, Standard Requirements for He joined OMICRON in 2012 as an Application Engineer and currently holds the
Instrument Transformers, IEEE Std position of a Product Manager in the area of dielectric transformer diagnostics. He
C57.13TM-2008 is member of VDE and IEEE.

102 TRANSFORMERS MAGAZINE | Volume 4, Issue 1