1

Demagnetization of CT Cores under exposure of operating currents

Dr. Fred Steinhauser, OMICRON, Austria

Motivation
Saturated CT cores and their behavior during fault
conditions have been and still are an important
topic for protection engineers, in particular when
fault currents with transient components occur.
Many papers have been written and presented
about this. Typically, the CT cores contain some
residual magnetism when the current flow is
interrupted. Especially at reclosing, this residual
magnetism may cause undesired effects.
In this context, the question "How will the
magnetization in the CT develop when power is
restored and the CT is exposed to normal
operating currents?" is frequently raised.
Surprisingly, the answer is often based on opinion,
it is even said this may be a "matter of philosophy".
Statements like "some say the saturation level
remains there forever, some say it will go away
over time" can be heard. Much of the said is based
on assumptions, models and simulations.
But for engineers, there must be a better way than
relying "on philosophy". The recognized method is
to look for plain facts and align them with proven
theories. For this paper, systematic measurements
have been made to put some light on this matter.

Practical Implications
Measuring residual flux can be a cumbersome
procedure [6]. In the past, it was hardly feasible to
perform a large number of such measurements in
reasonable time.
Also, for establishing a relationship between a
measured residual flux and the conditions that
caused this residual flux, the full previous history
(the anamnesis) of the CT must be known. Even if
the fault records were available for some arbitrary
cases of saturated CTs, this is not a systematic
approach for reliably determining a correlation of
effects.
Another problem is that the established methods
for determining the residual flux result also in a
destruction of the residual flux [3]. It is not possible
to resume the test at the most recent point.
Thus, every single measurement has to start from
the de-magnetized state, first bringing the CT into
a defined state of magnetization (applying a
defined residual flux). Then, some other condition
which may alter the residual flux may be applied.
Then, this residual flux is to be measured.

Each of the steps involved must be performed with

adequate precision and repeatability. Modern test
equipment allows precise application of the test
quantities for each of these steps and performing
the measurements in reasonable time to make a
systematic series of tests feasible.

Establishing a
Defined Magnetization
The first step for performing measurements with
residual magnetism is to apply a defined initial
magnetization to the CT core. It is assumed that
we start with a demagnetized core. In our case,
this condition is fulfilled without further efforts,
since the measurement device for the residual flux
leaves the core demagnetized after the
measurement.

Fig. 1 Injecting current on the secondary winding of the

CT for magnetizing the core

Then, a DC current is applied as shown in Fig. 1.

The primary side of the CT is an open circuit, so
the injected current equals the excitation current
through the main inductance of the CT. As its
magnitude is increased, the magnetization will
develop along the initial magnetization curve.

Fig. 2 Magnetization along the initial magnetization

curve and the hysteresis curve

When the excitation current is lowered again, the

magnetization moves along the upper branch of
the hysteresis curve and finally reaches
when
the current is entirely removed.

OMICRON electronics GmbH 2011 International Transformer Measurement Forum

This residual flux is then measured. This can be

conveniently achieved with a test set with a
dedicated residual magnetism measurement
features as shown in Fig. 3.

Fig. 3 Setup for measuring the residual flux

Since the values scatter slightly from measurement

to measurement, some averaging has to be
applied to get reliable values. Fig. 4 shows how the
residual magnetism depends on the excitation
current applied for magnetizing the core.

Applying nominal currents at the primary side is

difficult. While it might be feasible for steady state
conditions, it is not possible to produce such
currents with the dynamic behavior as required for
the following tests, e.g. for a defined time or even
for a defined number of cycles.
The current further modifying the magnetization of
the core is again applied on the secondary side as
shown in Fig. 1 for the magnetization. As the
primary side is an open circuit, the current injected
equals the magnetization current through the main
inductance .

Estimating the Nominal Current

Fig. 5 shows the circumstances in the loaded CT at
nominal conditions. The secondary current flows
through the secondary CT impedances and the
burden, producing a voltage drop
which
occurs at the main inductance.

Fig. 5 Equivalent circuit of the CT with burden

Fig. 4 Residual flux vs. magnetization current

Using this method, it is possible to establish a well

defined magnetization of the core as a starting
condition for the tests.
To make the results more independent from the
actual ratings of the specific CTs, the quantities are
normalized. The currents are referred to the kneepoint current , which is the exciting current at the
knee point voltage. The residual flux
is
referred to the saturation flux
and expressed as
the residual magnetism. The residual magnetism is
.
defined as
The reference values
and
used here are
those obtained according to IEC 60044-1 [1].
These values are measured under special
conditions and vary depending on the standard
and method used for determining them. But the
actual standard which is used to determine these
reference values is not so crucial; some other
values would just lead to a different scaling of the
diagrams, leaving the essence of the statements
unaffected.

Demagnetization
The goal of this work is to investigate how the
magnetization of the saturated core is affected by
normal operating currents. Therefore, conditions as
under nominal currents shall be applied.

With the total secondary impedance

the magnitude of the magnetization current at a
given burden
can be easily calculated:

As it has to be expected for a reasonable CT, this

is a fairly small current, only a few percent of the
knee-point current. In reality, this means only a few
Milliamperes.

Demagnetization
With Injected Current
The current source injects a sinusoidal current into
the secondary winding. As long the compliance
voltage of the current source is not exceeded, the
voltage adapts to the load impedance and the
current is kept constant. As the secondary
impedances are almost negligible compared to the
impedance of the main inductance, the voltage
measured at the terminals is almost identical to the
voltage at the main inductance and according to
hysteresis curve.
It must be noted that this injection of a defined
current is different from the procedures described
in most standards as IEC 60044-1, where a
sinusoidal voltage is applied and a current with
potentially non-sinusoidal waveform is obtained.

OMICRON electronics GmbH 2011 International Transformer Measurement Forum

This current
was now applied to the CT for a
few seconds and the remaining residual
magnetism was measured. It was significantly
reduced from the value established through the
magnetization. This was a first finding, but further
measurements with longer exposure time of the
nominal current showed the same result, the
residual magnetism did not drop further.
Consequently, the exposure time was reduced until
a change of the measured residual magnetism
could be observed. It turned out that the time
interval to be considered is less than a second or
only a few cycles. The result is shown in the
following figure.

reduction would have occurred most likely already

after 24 cycles and not only after 24 hours.
Reference [3] displays some oscillographic
recordings which indicate that most of the change
of the residual magnetism takes place in a few
dozen of cycles.
Reference [4] states that under the flow of rated
current the residual flux in CTs is reduced to 60 %
to 70 % of the remanent flux within a few seconds.
So far, the measurements and results described
are just in line with the state of the art.

Frequency Dependency
of the Effect
For the measurements described above it was
implicitly assumed and therefore not explicitly
mentioned that the current applied for changing the
magnetization of the core had nominal frequency,
in this case 50 Hz.
But it turns out that the reduction of the
magnetization is essentially more significant if the
frequency of this current is lowered. Fig. 7 shows
how the residual magnetism develops for different
frequencies of the applied current.

Fig. 6 Reduction of the residual magnetism when

applying nominal current

Fig. 6 shows how the residual magnetism develops

under the application of a current of about nominal
magnitude. The process is more or less completed
after about 20 cycles and the residual magnetism
is roughly reduced to half of the initial value.
This result could be used to get to an answer to
the question raised in the motivation. And as in
most cases, the extremes (stays forever / goes
away) do not apply and the actual circumstances
might depend on many other parameters which
were not considered in the measurement.
The insight summarized in Fig. 6 is not unique and
not new. There are several publications [2],[3],[4],
one of them dating back almost 70 years that
describe similar findings. Some of these
statements shall be briefly reviewed here.
Reference [2] contains a statement, called a
hypothesis, which nicely resembles what was
concluded above: "This residual flux remains in the
current transformer core practically undiminished
unless the circuit is loaded, in which case it is
decreased gradually but not usually removed
entirely." It describes one specific case where a CT
was magnetized to a certain value and then
exposed to a current of 5 A for 24 hours. After this,
the residual flux had been reduced to about 31 %
of the initial magnetization. At the time when this
experiment was made, it would have been
probably very difficult to find out that the same

Fig. 7 Reduction of the residual magnetism at different

frequencies

The initial magnetization was established with

, so the yellow curve (for 50 Hz) is
identical to the lower curve in Fig. 6. The residual
magnetism is already heavily reduced at 20 Hz
(red curve), and it is essentially removed at 10 Hz
after only 5 to 10 cycles (blue curve).
The observed differences in the levels of the
residual flux are remarkable. Obviously, the
magnetic properties of the core material are very
differently involved at the different frequencies.
A further attempt to visualize this is to look at the
voltages that occur at the CT when the current is
applied. The current is injected from a current
source, imposing the sinusoidal waveform of the
current. Any influence from the core material will
become visible in the waveform of the voltage.

OMICRON electronics GmbH 2011 International Transformer Measurement Forum

Recorded waveforms from two cases are shown in

Fig. 8. The lower (yellow) trace shows the injected
current. As explained above, it is sinusoidal and
the burst contains exactly five cycles.

Fig. 8 Voltages (top and middle trace) and current

(bottom trace) when applying nominal current to
the magnetized CT

The special scaling axes in Fig. 8 must be kept in

mind. The horizontal axis shows the angle ( ),
giving the same distance for a cycle independent
of frequency. As the frequencies are different by a
factor of 5, the voltage ranges are also different
and not to scale.
The upper (red) trace is the voltage at 50 Hz. After
the first cycle, the waveform is close to sinusoidal,
not showing much of a visible effect from the
hysteresis of the core material. The middle (blue)
trace is the voltage at 10 Hz. Its waveform is visibly
affected from the hysteresis.
The first cycle at 10 Hz is where the largest change
in the residual magnetism occurs. This is illustrated
in Fig. 9 in detail.

Fig. 9 Magnetization changing during the first cycle at a

frequency of 10 Hz

By performing an integration of the voltage and

appropriate scaling, the approximate course of the
residual magnetism over the applied current can
be constructed.
The current value on the horizontal axis is referred
to
, which is the peak value of the sinusoidal
current applied.
During the positive half wave of the current, the
magnetization curve describes a small hysteresis
loop superimposed to the remanence point
established by the initial magnetization. At the end
of the positive half wave, the magnetization is more
or less back where it started off when the current
was applied.
During the negative half wave, the magnetization is
heavily eroded as long the magnitude of the
current is increasing. But even when the current
moves back toward zero after reaching its
maximum value, the magnetization becomes
further decreased until the positive zero crossing of
the voltage takes place.

An Attempt for an Explanation

So, why can such different behavior be observed
at frequencies not higher than the rated
frequency? Or is the rated frequency already a
"considerably high" frequency when it comes to the
demagnetizing effect of a magnetic field caused by
a rated current? This thought shall be further
pursued and some reasoning shall be given.
One approach is reference [7] that does not aim at
the magnetic polarization issue, but focuses on the
frequency dependency of the permeability instead.
The key point is that this paper gives insight how
the magnetic field penetrates magnetic sheet
metals as used for transformer cores.
It is well known that eddy-currents occur when a
magnetic material is exposed to a changing outer
magnetic field. These eddy-currents build by
themselves a magnetic field that is opposed to the
outer field. By doing so, the eddy-currents shield
the inner portions of the magnetic material from the
outer field. The effects of the eddy-currents
become essential when the frequency of the outer
field is approaching the Wolman frequency

The above formula applies for sheet metals, where

is the specific resistance of the material, is the
thickness of the sheet, and
is the "overall initial
permeability" which is used throughout the
calculations in the mentioned paper.
This Wolman frequency can assume surprisingly
low values. For typical materials and thicknesses
of the laminated core material, it can be as low as
only a few hundred Hertz or even below.

OMICRON electronics GmbH 2011 International Transformer Measurement Forum

Fig. 10 from [7] shows the locus of the complex

permeability [8],[9] as a function of frequency for a
Chrome-Permalloy sheet metal. The frequency is
referred to the Wolman frequency.

In the case of an outer field resulting from an

operating current in a CT, this inner portion of the
core
material
is not
involved
in
the
demagnetization process and remains magnetized
at its former level.
The value of is inversely proportional to . So
for lower values of , increases and more of the
material becomes involved in the demagnetization
process, which would explain the effects shown in
Fig. 7 and Fig. 8.
Thus, even when the sheets in a laminated
transformer core are obviously thin enough for
reducing the eddy-current losses to an acceptable
level, the sheets may still appear to be thick when
it comes to the penetration by an outer field of
rated frequency.

Conclusions

Fig. 10 Locus of the complex permeability (solid:

measurement; dashed: calculation) [7]

The vertical axis represents the inductive

component, while the horizontal axis represents
the resistive component. The resistive component
is connected to the eddy-current losses.
Reference [7] is very challenging, containing more
than 50 numbered equations for mathematically
encompassing the effects. Further down, the paper
displays Fig. 11 for the local permeability over the
cross section of the sheet metal.

Fig. 11 Local permeability over the cross section of a

36 % Nickel-Iron sheet of thickness
[7]

On the surface, the local permeability is zero and

increases when going deeper into the material.
The local permeability is essentially affected by the
outer field only within a certain depth of about the
value , which characterizes where the local
permeability reaches the value of the overall initial
permeability. The permeability remains at high
levels in a large inner portion of the sheet, because
it is effectively shielded by the eddy-currents and
not exposed to the outer field.

Referring to the question raised in the motivation

section, a rough answer can be given: The
magnetization of a saturated CT core exposed to
normal operating currents will neither "remain there
forever", nor will it completely "go away over time".
The actual degree of the demagnetization due to
operating currents depends on several parameters,
which were not all investigated in detail.
Given the observed speed of the demagnetization
(only a few cycles in some cases), it might be
realistic that saturation effects will be quickly
diminished in a CT core at reclosing after
successful fault clearing.
The explanations and conclusions given above
could be verified by systematic repetition of the
measurements with other CTs, preferably with the
exact knowledge of the construction.
There is the question if the use of thinner sheet
metals in the core yields in more effective
demagnetization due to operating currents and
better recovery after saturation. This could be
investigated by comparing CTs with cores with
different lamination thickness at otherwise identical
data. This approach may collide with a statement
from reference [5] that points out that "the eddycurrent loss in a laminated core may increase if the
lamination thickness is reduced". The overall
optimum for the technical compromise may depend
on how important an improvement of the recovery
behavior is regarded compared to other properties
of a CT.

OMICRON electronics GmbH 2011 International Transformer Measurement Forum

Literature

Dedication

[1]

IEC 60044-1, Instrument transformers

Part1: Current transformers. Ed. 1.2, IEC,
Geneva 2003

[2]

Seeley, H. T.: The Effect of CurrentTransformer

Residual
Magnetism
on
Balanced-Current or Differential Relays. AIEE
Trans., Vol. 62, April 1943

[3]

Bruce, R. G., Wright, A.: Remanent Flux in

Current-Transformer Cores. Proc. IEE, Vol.
113, No. 5, May 1966

[4]

Dickert, J., et al.: Investigation on the

Behavior of the Remanence Level of
Protective Current Transformers. MEPS'06,
Wroclaw 2006

[5]

Mukerji, S. K., et al.: Eddy Currents in

Laminated Rectangular Cores. Progress In
Electromagnetics Research, PIER 83, 2008

[6]

Pfannenstiel, M.: Residual magnetism of

current transformers, OMICRON ITMF, 2010

[7]

Feldtkeller, R.: Permeabilitt und Wirbelstrme in Blechkernen bei sehr hohen

Frequenzen. Frequenz, Band 3/1949 Nr.4

[8]

Boll, R.: Weichmagnetische Werkstoffe.

Vacuumschmelze GmbH. 4th Edition, 1980

[9]

Michalowsky, L., Schneider, J.: Magnettechnik. Vulkan-Verlag GmbH. 3rd Edition,

2006

This work is dedicated to our

deceased colleague Franz
Sss,
who
unexpectedly
passed away on 3rd October
2008. Franz was the creator
and father of the OMICRON
CT Analyzer which revolutionized the way CTs are tested.
His legacy, now enhanced with the RemAlyzer
function, made it feasible to perform the extensive
measurements involved in investigations as
required for this paper in reasonable time.

About the Author

Fred Steinhauser was born
in Austria in 1960. He studied
Electrical Engineering at the
Vienna
University
of
Technology,
where
he
obtained his diploma in 1986
and received a Dr. of
Technical Sciences in 1991.
In 1998 he joined OMICRON,
where he worked on several aspects of testing
power system protection. Since 2000 he works as
a product manager with a focus on substation
communication issues.
Fred Steinhauser is a representative of OMICRON
in the UCA International Users Group. As a
member of WG10 and WG17 in the TC57 of the
IEC he contributes the standard IEC 61850. He is
also a member of SC B5 of CIGR.

OMICRON electronics GmbH 2011 International Transformer Measurement Forum

OMICRON is an international company serving the electrical power industry with

innovative testing and diagnostic solutions. The application of OMICRON products
allows users to assess the condition of the primary and secondary equipment on
their systems with complete confidence. Services offered in the area of consulting,
commissioning, testing, diagnosis, and training make the product range complete.
Customers in more than 140 countries rely on the companys ability to supply
leading edge technology of excellent quality. Broad application knowledge and
extraordinary customer support provided by offices in North America, Europe,
South and East Asia, Australia, and the Middle East, together with a worldwide
network of distributors and representatives, make the company a market leader in
its sector.

Americas
OMICRON electronics Corp. USA
12 Greenway Plaza, Suite 1510
Houston, TX 77046, USA
Phone: +1 713 830-4660

+1 800-OMICRON
Fax:
+1 713 830-4661
info@omicronusa.com

OMICRON

Asia-Pacific
OMICRON electronics Asia Limited
Suite 2006, 20/F, Tower 2
The Gateway, Harbour City
Kowloon, Hong Kong S.A.R.
Phone: +852 3767 5500
Fax:
+852 3767 5400
info@asia.omicron.at

Europe, Middle East, Africa

OMICRON electronics GmbH
Oberes Ried 1
6833 Klaus, Austria
Phone: +43 5523 507-0
Fax:
+43 5523 507-999
info@omicron.at

www.omicron.at www.omicronusa.com