Beyond the Knee Point:

A Practical Guide to CT Saturation

Ariana Hargrave, Michael J. Thompson, and Brad Heilman

Schweitzer Engineering Laboratories, Inc.

© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained
for all other uses, in any current or future media, including reprinting/republishing this material
for advertising or promotional purposes, creating new collective works, for resale or
redistribution to servers or lists, or reuse of any copyrighted component of this work in other
works.

This paper was presented at the 71st Annual Conference for Protective Relay Engineers and can
be accessed at: https://doi.org/10.1109/CPRE.2018.8349779.

For the complete history of this paper, refer to the next page.
 Revised edition released April 2019

Previously presented at the

PowerTest Conference, March 2019,
54th Annual Minnesota Power Systems Conference, November 2018,
72nd Annual Georgia Tech Protective Relaying Conference, May 2018,
and 71st Annual Conference for Protective Relay Engineers, March 2018

Previous revised edition released December 2018

Originally presented at the

44th Annual Western Protective Relay Conference, October 2017
 1

Beyond the Knee Point:

A Practical Guide to CT Saturation
Ariana Hargrave, Michael J. Thompson, and Brad Heilman, Schweitzer Engineering Laboratories, Inc.

Abstract—Current transformer (CT) saturation, while a fairly understanding of CT saturation must extend beyond the
common occurrence in protection systems, is not often clearly knee point.
understood by protective relay engineers. This paper forgoes the
usual physics equations to describe how CTs saturate in a simple
and intuitive way. We explain the differences between II. CT SATURATION THEORY
symmetrical and asymmetrical saturation and how remanence To understand CT saturation, it is very important to
accumulates in the core of a CT. We then describe the CT understand the basic concept of how CTs work and what is
equivalent circuit and how it results in the familiar CT excitation actually happening when they saturate. This section describes
graph. ANSI ratings of CTs are explained, and we show how to
analyze the performance of CTs using simple equations and tools. what happens in the core of a CT during symmetrical saturation,
Finally, we explain how CT saturation can affect relay operation asymmetrical saturation, and remanence. It then explains how
and show how to detect CT saturation in protective relay event this core activity corresponds to the CT equivalent circuit,
reports. Real-world event reports are presented where correct ANSI voltage ratings, and the familiar CT excitation graph.
relay operation was compromised as a result of incorrect current
values from saturated CTs. A. How CTs Work
In its simplest form, a CT consists of two sets of wire
I. INTRODUCTION windings around an iron core, as shown in Fig. 1. The concept
Current transformer (CT) saturation is not a new topic, and is the same for a window or bushing CT, which consists of a
there have been many papers, books, application guides, and secondary winding around a core, with the primary winding
tutorials written on the subject. Sorting through this vast array being the primary conductor that passes through it.
of information to piece together a complete understanding of Transformers work based on the principle of electromagnetic
the topic is a time-consuming task and may not be realistic with induction. This principle states that an alternating magnetic flux
the schedules and demands placed on many practicing in the presence of a loop of wire induces a voltage across that
engineers. Because of this, engineers’ level of understanding is loop. Magnetic flux is simply the amount of magnetic field
often limited to the familiar CT excitation graph. The following passing through a material such as a transformer core.
is a list of common questions related to CT saturation: When alternating current IP flows in the primary winding of
• Why does a CT saturate (Section II, Subsection B)? a transformer, it generates an alternating magnetic field H,
• What is remanence, and do I have to worry about it which corresponds to an alternating magnetic flux Φ, around
(Section II, Subsection C)? the transformer core. This alternating magnetic flux passes
• What does it mean when a CT is a C800 (Section II, through the secondary winding. What happens next depends on
Subsection F)? the load connected to the secondary winding.
• I have a C800 multiratio CT tapped at 400/5. Is it still If the secondary is connected to a burden, the alternating
a C800 (Section II, Subsection F)? magnetic flux in the core induces an alternating voltage VS
across the secondary winding. This causes a corresponding
• How do I make sure my CT will not saturate for my
alternating current IS to flow in the secondary winding. The
worst-case fault current (Section III)?
alternating current in the secondary creates its own alternating
• The knee point determines the saturation voltage of a
magnetic field and alternating magnetic flux that oppose those
CT, right (Section II, Subsection E)?
created by the primary winding. These primary and secondary
• Will saturated CTs cause my relay to misoperate?
fluxes cancel, leaving a negligible amount of net flux in the
What if they just saturate a little bit (Section V)?
core. This occurs until the core becomes saturated.
• After a misoperation, how do I know if CT saturation If the secondary is open-circuited, the alternating magnetic
was a cause (Section IV)? flux in the core induces a very high alternating voltage VS
• Can modern relays prevent misoperations due to CT across the secondary winding. VS remains on the terminals with
saturation (Section VI)? no secondary current flowing, which is why it is very dangerous
The goal of this paper is to explain CT saturation to the to open-circuit an in-service CT. Because IS cannot flow, it
protective relay engineer and to answer these questions in a cannot create an opposing magnetic field, leaving a net flux in
clear and practical way. As this paper demonstrates, a proper the core equal to the flux created by the primary current.
 2

This magnetic behavior is modeled when we introduce the

CT equivalent circuit in Section II, Subsection D. For now,

Magnetic Flux Density (B)

1
know that the induction of voltage VS is a result of the magnetic
field created by IP along with the internal impedances of the
transformer and the connected load. For simplification going 0
forward, we assume a fixed linear burden is connected to the
CT. When this is the case, Ohm’s law mandates that the
secondary current in the CT and the voltage across the –1
magnetizing impedance are directly proportional.
H Φ –10 –5 0 5 10
Magnetic Field Strength (H)

IS Fig. 2. Example B-H curve

IP

How this magnetic behavior affects the ability of the CT to

VS reproduce current is simple: it is the change in flux caused by
the magnets changing direction that induces voltage VS across
the secondary winding of the CT. Voltage VS, in turn, generates
current IS through the connected circuit. When the core reaches
maximum flux density, it is fully saturated and there are no
more magnets remaining to change direction. This causes
voltage VS to drop to zero, and current IS ceases to flow.
Fig. 1. CT drawing CT saturation can occur in two forms: symmetrical
B. Why Does a CT Saturate? saturation and asymmetrical saturation. These saturation types
In an ideal world, the secondary current leaving the CT (IS) are explained in the following subsections. For a detailed
is an exact replica of the primary current (IP) divided by the explanation of the mathematics of CT saturation, see [1].
ratio of the number of turns in each winding (CT ratio). 1) Symmetrical Saturation
However, when a CT saturates, IS does not accurately replicate Symmetrical saturation is the result of a symmetrical
IP. The reason a CT saturates is related to what physically primary current being applied to the CT that is too large for the
occurs inside a CT during the electromagnetic induction CT core to handle for a given burden.
process. The iron core of a CT is made up of a fixed number of Fig. 3 shows an example of the primary current (IP) and
magnetic dipoles, which can be thought of as molecular secondary current (IS) of a CT during symmetrical saturation.
magnets. Ideally, these magnets are randomly arranged in Ideally, before the current is applied at Point a, the magnets in
polarity throughout the core, as shown in Fig. 1. the core are aligned in random directions (no remanence, which
When alternating current (IP) flows in the primary winding is discussed later in this paper). Between Points a and b, as IP
and generates the magnetic field (H), the strength of this starts to flow in the first positive half cycle, the magnets start to
magnetic field affects the magnets in the core and causes them line up in the positive direction. Because there is a change in
to start lining up (in the same direction as the magnetic field) to flux during this time, the current IS matches IP exactly
produce the magnetic flux (Φ). The more current IP that flows, (assuming a CT ratio of 1:1). Before the positive half cycle is
the stronger the magnetic field (H) becomes, and the more over, at Point b, all of the magnets available in the core are lined
magnets become lined up. The number of magnets that are lined up in the positive direction, and the core has reached maximum
up at a given time is the flux density (B). When all the magnets flux density (saturation). At this point, even though IP continues
in the core are aligned in the same direction, the maximum flux to flow, there is no more change in flux, VS drops to zero, and
density of the core is reached and the CT core is said to be IS drops to zero. IS stays at zero until IP begins to flow in the
saturated. negative direction, reversing the magnetic field. This negative
The relationship between the magnetic field strength (H) and flow, beginning at Point c, causes the magnets to begin aligning
the magnetic flux density (B) is given by the B-H curve of the in the negative direction. This changing flux allows the
core, as shown in the example in Fig. 2. Different types of core generation of voltage VS and allows IS to follow IP again until
materials have different B-H curves, which depend on the all of the magnets are aligned in the negative direction at
ability of the material to support a magnetic field. Point d. Because maximum flux density (saturation) has been
reached again, VS and IS again drop to zero.
 3

The example in Fig. 3 shows the primary current decreasing saturates. This limit is defined by a symmetrical sine wave with
in magnitude every cycle. The point of this is to show that if a fixed voltage magnitude and fixed area under the curve in
primary current magnitude decreases, the CT is saturated for both the positive and negative directions. As long as the actual
less time. The lower magnitude in the second cycle of Fig. 3 CT waveform does not exceed this positive or negative volt-
generates a weaker magnetic field, requiring less flux density time area, the CT will not saturate. Consider the dc offset of the
to replicate the current correctly. Because fewer magnets are asymmetrical current in Fig. 4. This dc offset will result in an
used, IS reliably replicates IP for a longer time until all the accumulating positive volt-time area that eventually reaches the
magnets are aligned. In the third cycle, the magnitude of IP has maximum that the CT can handle at Point d, where saturation
been lowered to the point that the CT does not saturate and occurs.
replicates current correctly the entire time.
IS
IP

Current
Time
a b c d

a a–b b b–c c–d d

Fig. 4. Primary currents, secondary currents, and magnetic dipoles in the

core during asymmetrical saturation
Fig. 3. Primary currents, secondary currents, and magnetic dipoles in the
core during symmetrical saturation The time it takes for a CT to saturate can vary depending on
current magnitude, dc offset, X over R ratio, burden, and
2) Asymmetrical Saturation
remanence. An equation to calculate the time it takes a CT to
The other form of saturation, asymmetrical saturation,
saturate is found in [2]. Symmetrical saturation normally occurs
results from high levels of dc offset in the primary sinusoidal
within the first half cycle after fault inception, whereas
current being applied to the CT. The current peaks are not
asymmetrical saturation can take several cycles to occur.
symmetrical around zero. DC offset is when there is more area
Because the magnitude of the current does not need to be as
under the curve above the zero crossing than there is below the
high for asymmetrical saturation to occur, it usually takes
zero crossing (or vice versa). Fig. 4 shows how a waveform
longer for the flux density to reach its maximum.
with high dc offset can cause a CT to quickly saturate. From
Not all saturated waveforms are as obvious to the eye as the
Points a to b, magnets are all lining up in the positive direction
waveform in Fig. 3, which has sharp edges and large chunks
and the CT has not saturated. When IP becomes negative, from
missing. With low levels of saturation, a saturated waveform
Points b to c, magnets start changing directions and lining up in
can be hard to detect. Fig. 5 shows waveforms from CTs with
the negative direction. Because there is such a small amount of
light, medium, and heavy levels of symmetrical saturation [3].
area under the curve between Points b and c, not very many of
the magnets have lined up in the negative direction when the IP a) b) c)
current goes back above the zero crossing (Point c) and they are IP IP
forced to start lining up in the positive direction again. Finally,
Current

IP
Current

Current

Time Time Time

at Point d, all of the available magnets are lined up in the IS IS IS
positive direction and the core saturates. The dc offset—not the
magnitude of the fault current itself—is what causes the
saturation. The fact that IP is not below the zero crossing long Fig. 5. Varying degrees of CT saturation: light saturation (a), medium
enough to reset the magnets in the opposite direction is what saturation (b), and heavy saturation (c)
causes saturation to occur. Notice that the amount of time the
The waveshape of secondary current IS during saturation
CT is saturated is less with each cycle as the dc component
also depends on the type of load connected. Fig. 6 shows the
decays.
waveshape with a purely resistive burden compared with a
The magnetic behavior described previously is referred to in
burden that has both resistive and reactive components of
[1] as the integration of the volt-time area. Recall that with a
similar magnitudes. The difference in waveshape is because of
fixed burden, current is directly proportional to voltage. The
the fact that current through an inductive load cannot change
amount of flux that accumulates in the core of a CT is directly
instantaneously, so it takes some time for the current to decay.
proportional to the area under the voltage (or current) curve. A
given CT can handle some maximum amount of flux before it
 4

(a) The only way to remove this remanence is by degaussing the

IP
IS CT. Degaussing can be done by applying primary rated current
Current
and a variable load to the CT secondary terminals. Start the load
Time at a high resistance to cause the CT to saturate in both the
positive and negative directions. Then bring the CT out of
saturation by slowly reducing the load (and thus the secondary
voltage) to zero. The only convenient time to do this is during
(b) system maintenance, but in practice it is almost never done.
IP Annex C of [2] shows that in a survey of 141 CTs on a 230 kV
IS system, 43 percent had more than 40 percent remanence when
Current

measured.
Time
The effect of remanence on CT saturation is shown in Fig. 7.
In this example, the CT is sized to perfectly handle an ac current
signal of a certain magnitude without saturating. In addition,
there is some remanence left over in the core of the CT. Before
Fig. 6. Saturated waveshapes for resistive (a) and resistive-inductive (b)
loads the current is applied, between Points a and b, some of the
magnets in the core are already aligned in the positive direction
C. Remanence because of remanence. Between Points b and c, as IP starts to
If a CT has reached saturation and a switch is opened to flow in the first positive half cycle, the remaining magnets also
remove the primary current, we would expect the magnetic field line up in the positive direction. Because there is a change in
(H) to disappear and the flux density (B) to reduce to zero. flux during this time, the current IS matches IP divided by the
However, flux density does not go to zero when the primary turns ratio. Before the positive half cycle is over, at Point c, all
current stops flowing. When primary current is removed, the the magnets available in the core are lined up in the positive
magnetic field that causes the magnets to change orientation direction and the core has reached saturation. At this point, even
disappears, and the magnets in the core remain in their present though IP continues to flow, there is no more change in flux and
orientation. The magnets will not move again until exposed to VS and IS drop to zero. IS stays at zero until IP begins to flow in
another magnetic field. The amount of flux density remaining the negative direction, reversing the magnetic field and
in the core is called remanence. The fact that the magnets still allowing flux to accumulate in the opposite direction. This
point in the direction they were in when the magnetic field was occurs at Point d, where magnets begin to align in the negative
removed gives the core “memory” (like a permanent magnet). direction and the changing flux allows IS to follow IP again. By
This remanence remains in the core until primary current is the time IP reaches its negative peak at Point e, only half of the
reapplied. If the reapplied current is opposite in polarity from maximum core flux is aligned in the negative direction. This
the original, flux density is created in the opposite direction of allows IS to continue to follow IP until all the magnets are
the prior remanence. aligned in the negative direction, at Point f. Although we have
The example of a switch being opened to remove primary reached maximum flux density again, this is where IP changes
current is exactly what happens when a relay trips a circuit direction and we start accumulating flux in the positive
breaker during a fault. Recall that when a breaker operates, direction.
current is interrupted at a zero crossing. In Section II, IP
Subsection B, we show that for both symmetrical and
IS
asymmetrical currents, there is a positive or negative flux
density in the core when the current zero crossing occurs. This a b c d e f g
flux density can be significant during high-magnitude
asymmetrical current (when a dc transient is present). This
remanence remains in the CTs after the breaker opens and
affects their behavior the next time they are energized.
Remanence can either help or hinder a CT’s performance,
depending on whether the remanence is of the same polarity or
a–b b–c c c–d d–e e–f g
opposite polarity of the next current that the CT measures. It
takes more time for the CT to saturate if the remanence is the Fig. 7. Primary currents, secondary currents, and magnetic dipoles in the
opposite polarity of the current and less time if it is the same core with remanence
polarity. As we can see from this example, saturation as a result of
Examination of the example B-H curve in Fig. 2 shows remanence is short-lived, lasting about half a cycle. Because of
another factor that causes the magnetic core to have remanence. this short saturation time, remanence has little effect on
Note that the curve follows a loop trajectory. The flux density standard protection algorithms and is normally neglected in CT
(B) lags the field intensity (H) as it goes through a power system saturation calculations [4] [5]. Any relaying algorithm that
cycle. This phenomenon is called hysteresis. performs faster than this should include some level of built-in
protection against CT saturation.
 5

D. CT Equivalent Circuit when the ZE impedance is large (during normal conditions) and
We can represent the behavior of the C-class CT shown in large when the ZE impedance is small (during saturation). The
Fig. 8a with the simplified equivalent circuit shown in Fig. 8b. current measured by the load (IS) is what is left of the ratio
current (IST) after the error current (IE) is lost through the
(a) Secondary
magnetizing impedance (ZE). Therefore, the current the relay
sees is accurate during times of high ZE impedance (low error
current), and it is much lower than expected during times of low
ZE impedance (high error current). This behavior gives us the
Primary
classic saturation waveform shown in Fig. 3.
Fig. 9, adapted from [6], explains the saturation process in
the time domain. The graph in Quadrant III shows the primary
Fully Distributed Windings current IP being applied to the CT. The graph in Quadrant II
IST RS IS
shows the flux density over time. At time t0, when the CT is
(b)
first energized, the flux density starts to increase from whatever
IE
remanent value was left in the CT at the last de-energization.
Flux density continues to rise in the CT during the positive half
IP VE VS ZB
ZE cycle and decreases during the negative half cycle. Time t3 is
the point of maximum flux density, when the CT is fully
saturated. The graph in Quadrant IV shows the excitation
N1 N2
(error) current over time. We can see that the highest excitation
Fig. 8. Class C CT (a) and Simplified CT equivalent circuit (b) current occurs at the point of maximum flux density. The graph
IP (the primary current), VS (the secondary CT terminal in Quadrant I shows flux density versus excitation current and
voltage), and IS (the secondary load current) are is made up of the intersection of points between the graph in
self-explanatory. N1 and N2 represent the turns ratio of the CT, Quadrant II and the graph in Quadrant IV. The graph in
and the ratio current IST is the primary current divided by this Quadrant I is not time-dependent, but simply shows the amount
ratio. The circuit connected to the secondary of the CT is the of excitation current that is measured for a given flux density.
burden represented by ZB and includes the impedance of the This graph shows that when the flux density is low, the
relay and CT leads. The resistance RS represents the secondary excitation (error) current is low (because ZE is a high value). As
winding resistance of the CT. The magnetizing behavior of the flux density increases and the CT approaches saturation, the
core of the CT is represented by a varying reactance, ZE. This excitation current increases substantially (because ZE is a lower
impedance changes as the CT saturates, resulting in varying value), meaning less current will be available to the burden.
secondary excitation voltage VE and varying exciting branch Notice that the graph in Quadrant I corresponds directly to the
current IE. (Although ZE technically has a real and reactive B-H curve in Fig. 2, where the excitation current IE is
component, most practical CT analysis is close enough when proportional to the magnetic field strength H.
neglecting the angles of the impedances.) The flux density in Fig. 9 is directly proportional to both the
In the CT equivalent circuit in Fig. 8, the constant current flux and the excitation voltage VE. We can therefore interpret
source IP drives the total secondary current IST as determined by the graph in Quadrant I as the instantaneous relationship
the turns ratio. IST is the current that is measured by the load if between VE and IE, and use it to understand the changing
the CT is ideal. However, we must take into account the impedance of the CT excitation branch (ZE). The slope of the
magnetizing behavior of the CT core that is represented by the curve is the ratio of VE/IE, which Ohm’s law allows us to
varying impedance ZE. Because magnetization is nonlinear, characterize as ZE. We can easily visualize that the curve is
different values of impedance (ZE) must be used for different made up of two linear sections and that the knee point is where
states of operation (normal conditions vs. saturated conditions). the two linear sections transition. The steep part of the curve,
We can think of ZE as a varying impedance that changes with where the ratio of VE/IE is high (ZE is high), represents where
flux density. It has a very high value under normal conditions the core is not saturated. This is called the iron-core reactance
and a very low value (basically a short-circuit) when the CT is region of the curve. The flat part of the curve, where the ratio
saturated. The excitation current (IE) is the current that is lost of VE/IE is low (ZE is low), represents where the core is
through the magnetizing branch and corresponds to the CT saturated. This is called the air-core reactance region of the
error. We can also refer to this as “error current.” IE is small curve because the saturated iron core has no better permeability
than air.
 6

II Flux Density I Excitation graphs are developed from measurements at the

factory. Graphs for multiratio CTs show multiple lines, one for
each tap. Fig. 10 shows the highest and lowest taps of a C400,
2000/5 multiratio CT. The magnetizing impedance ZE can be
calculated at any point on the graph by dividing VE by the
Remanence Excitation
Time Current corresponding value for IE. This results in values for ZE that
t0 t1 t2 t3 vary nonlinearly. Recall again that the values on the excitation
graph are all rms values. In reality, ZE, VE, and IE change
continually with every instantaneous value during a power
system cycle.
A simple test can be performed to obtain the excitation graph
Primary
Current IP of a CT. With the primary of the CT open, a user can apply a
III
Time
IV
known voltage across the secondary terminals and measure the
current flowing from the source into the CT. With voltage
Fig. 9. Relationship between primary current, core flux density, and error
current in a CT applied on the secondary side, the equivalent circuit of the CT
changes. In this test, the voltage applied is VE and the current
E. Excitation Graph measured is IE. The user can then plot VE versus IE for various
Being able to interpret the classic knee-point (excitation) values of VE.
graph of a CT is an integral part of understanding CT saturation. Using the graph in Fig. 10 for the 2000/5 CT as well as the
The previous sections explain what happens as a CT saturates CT equivalent circuit in Fig. 8, we can analyze how the CT
in the time domain and why CT saturation results in secondary behaves during different operating conditions (normal vs.
waveforms like those in Fig. 5. However, it is sometimes saturated). For simplicity, these examples neglect the CT
helpful to think in terms of rms quantities instead of samples on secondary resistance RS.
a waveform over time. The excitation graph in Fig. 10 shows
the relationship between the excitation voltage VE and 1) Normal Operation
The area of normal operation is along the high linear slope
excitation (error) current IE of a CT in rms quantities. In other
portion of the excitation graph, where IE and VE increase
words, the graph shows what we would read if we varied the
proportionally to each other. We select a random point on the
magnitude of the ac current on the primary of the CT and were
line in this region as (IE = 0.01 A, VE = 60 V) and apply these
somehow able to measure the VE voltage and IE current with an
values to our equivalent circuit, as shown in Fig. 11. We then
rms meter. This graph shows the rms value of the entire
solve for ZE = 6,000 Ω, which is a very high impedance. If we
waveform, not just a specific point in time when the CT is not
connect a burden of 4 ohms, we can solve for IS = 15 A and IST
saturated or fully saturated. The graph in Fig. 10 looks similar
= 15.01 A. We see that because the error current is so small, the
in shape to the graph in Quadrant I of Fig. 9, but Fig. 9 consists
ratio current (IST) is very close to the current the load actually
of instantaneous quantities and Fig. 10 consists of rms
sees (IS), meaning the CT is behaving properly.
quantities. Another difference between Fig. 9 and Fig. 10 is a
result of the equation Φ = B • A. Basically, the flux density (B) IST = 15.01 A IS = 15 A
in Fig. 9 is multiplied by the area of the core (A) to get the
IE = 0.01 A
magnetic flux (Φ), which is proportional to the excitation
voltage (VE). VE = 60 V VS = 60 V
IP ZB = 4 Ω
ZE = 6,000 Ω

Fig. 11. Equivalent circuit example of normal operation

2) Saturation
The area of saturation is along the low linear slope portion
of the excitation graph, where a small change in VE results in a
very large change in IE. We select a random point on the line in
this region as (IE = 10 A, VE = 496 V) and apply these values to
the equivalent circuit, as shown in Fig. 12. We then solve for
ZE = 49.6 Ω, which is a very small impedance. If we connect a
burden of 4 Ω, we can solve for IS = 124 A and IST = 134 A. We
Fig. 10. Example excitation graph for a C400 CT [7] see that because the error current is so large, the ratio current
(IST) is not very close to the current the load actually sees (IS),
meaning the CT is behaving poorly.
 7

IST = 134 A IS = 124 A 1) Ratio Error

The accuracy of a CT is defined as the extent to which the
IE = 10 A
current in the secondary circuit reproduces the current in the
IP VE = 496 V VS = 496 V ZB = 4 Ω
primary circuit in the proportion stated by the marked ratio [2].
ZE = 49.6 Ω
There are two ways to measure the accuracy of a CT—a direct
test or an indirect test. A direct test is done by injecting a known
primary current into the CT and calculating the difference
between the injected primary current and the measured
Fig. 12. Equivalent circuit example of saturation
secondary current at the terminals (after taking into account the
The shape of this area of the excitation graph confirms what marked ratio). Although this is the simplest and most accurate
is happening magnetically when a CT saturates. Recall that VE way to test CT accuracy, it is not practical for relaying CTs
is generated by the magnetic field from the current flowing in because the extremely high current magnitudes that are required
the primary winding. In addition, recall from our time-domain to be injected into the primary to achieve 20 times nominal are
analysis that during the parts of the waveform when the CT is difficult to generate.
saturated (high levels of VE), the current is diverted away from An indirect test is described in [9] as follows, which is
the load and into the magnetizing branch. If this happens for a similar to the excitation test described in Section II,
large percentage of the waveform, we can expect a higher rms Subsection E. First, calculate the equivalent VE for a desired
current for IE. secondary current IS and total burden (ZB + RS). Next, with the
The 2000/5 CT in Fig. 10 has a maximum VE of 496 V. This primary open-circuited, inject this voltage between the
means that no matter how much primary current is coming into secondary terminals of the CT and measure the excitation
the CT, the resulting magnetic field can only produce a voltage current IE. Then use (1) to calculate the percent ratio error. We
as high as 496 V on the secondary winding. This voltage is define saturation as the point when the CT ratio error exceeds
limited by the amount of core material, and thus magnetic 10 percent.
dipoles, available in the CT. A CT with a larger core is able to IE
support a higher voltage and has a higher rating (discussed later Ratio error (%)  •100 (1)
IS
in the paper). The actual voltage on the secondary terminals is
496 V minus the voltage drop across the secondary winding At first glance, (1) might seem like an incorrect equation to
resistance RS. calculate error. When we try to define how well a CT
reproduces current, a standard error calculation would compare
3) Definition of Saturation and Knee-Point Voltage
the actual secondary current coming out of the CT terminals (IS)
In previous sections of the paper that explain saturation in
with the expected or ideal secondary current (IST). Instead, (1)
the time domain, we said that the CT was saturated during the
compares the error current (IE) with the actual secondary
sections of the waveform where the flux was not changing and
current (IS). Appendix A explains where (1) comes from and
the secondary current dropped to zero. This is an accurate
relates it to the more intuitive definition of error.
definition of saturation in the time domain.
It is also important to define saturation in terms of rms 2) Terminal Voltage Rating
quantities so we can work with the CT equivalent circuit and Most protection CTs in the United States are ANSI Class C.
the CT excitation graph. IEEE does not define a specific To be classified as a Class C CT, the CT must be constructed
saturation point, but does define the knee point of the excitation so that it has negligible leakage flux. If the leakage flux can be
curve as the voltage at the point where a 45-degree line hits neglected, the performance of the CT can be determined by
tangent to the curve. This is shown in Fig. 10 and results in a calculation (Class C). That generally requires that the CT be
knee-point voltage of 200 V for the 2000/5 CT. The knee point constructed with a toroidal core with fully distributed windings
of the graph defines where the CT starts to exhibit nonlinear and a single primary winding, as shown in Fig. 8. The generic
behavior resulting from the beginning of core saturation. CT shown in Fig. 1 would be considered a Class T because the
However, for rms analysis, it is not practical to define saturation leakage flux in a CT constructed like this cannot be neglected.
at the point where it first begins. Instead, saturation typically Class T means that the performance of the CT can only be
refers to the point where the CT error starts to exceed accurately determined by test.
10 percent. This aligns with definitions of the ratio error and the The number following the C is the secondary terminal
terminal voltage rating in Section II, Subsection F. For now, voltage rating. IEEE defines the secondary terminal voltage
understand that the point of saturation is not the same as the rating as “the CT secondary voltage that the CT will deliver
knee-point voltage (a very common misconception). The knee- when it is connected to a standard secondary burden, at
point voltage is typically 46 percent of the saturation voltage 20 times rated secondary current, without exceeding a 10%
[8]. ratio error [2].”
This is the highest VS voltage the core will support without
F. CT Accuracy Class
going into significant saturation and assumes steady-state,
CT accuracy is often discussed using the CT ratio error and symmetrical (no dc offset) current. An increase in primary
terminal voltage rating. current that tries to force a voltage beyond this point simply
results in more error current and a very small, disproportionate
 8

increase in current delivered to the load. The definition also of the CT has a much lower terminal voltage at 10 A of error
states that assuming a 5 A nominal CT, 100 A is delivered to a current than the full 2000/5 tap. For multiratio CTs, the voltage
standard burden when VS matches the voltage rating of the CT. rating of a tap is directly proportional to the CT ratio
IEEE defines standard burdens for relaying CTs as 1, 2, 4, and corresponding to the tap divided by the full CT ratio (assuming
8 Ω. Consequently, the standard voltage ratings for 5 A nominal the windings are fully distributed around the core). For
relaying CTs are C100, C200, C400, and C800 (20 times the example, if a C400, 1200/5 CT is operating on a 600/5 tap, the
5 A rated secondary current) [9]. For example, a C400 accuracy voltage rating at 600/5 is calculated as shown in (2).
class on a 5 A CT means that the ratio error will not exceed  600 / 5 
VTAP  C400 •   C200 (2)
10 percent for any current up to 100 A secondary (20 times the 1200 / 5 
rated current) as long as the burden does not exceed the 4 Ω
standard burden. This means that the CT supports a voltage of 200 V at 20
From this definition, the voltage rating of a CT defines times rated secondary current of 100 A, which results in a
voltage VS, burden ZB, and current IS in the example shown in standard burden of 2 Ω.
Fig. 13 for a C400, 5 A nominal CT. The definition also tells us A final note on CT ratings and the knee point: notice that the
that at this voltage, the ratio error (IE / IS) is a maximum of knee point of the graph in Fig. 10 is VE = 200, which is half of
10 percent. If we assume this worst case, then IE must equal the ANSI voltage rating of the CT (C400). In fact, 400 V is not
10 A. We can then back-calculate to find IST = 110 A, which is even on the high linear slope portion of the graph. Recall again
the primary current divided by the turns ratio. If we go above that the knee point only shows where the CT behavior starts to
this, we will start getting more than 10 percent error on the become nonlinear, while the ANSI rating defines the point of
output, and the current the connected load reads will not be 10 percent error. The ANSI rating, not the knee point, defines
reliable. Similarly, increasing the burden ZB to a value greater the practical threshold for saturation. Because the knee-point
than the standard burden will also result in an increased error. voltage is normally 46 percent of the saturation voltage, a
When applying CTs, a simple rule of thumb to avoid a popular rule of thumb to avoid saturation is to ensure that the
symmetrical saturation error over 10 percent is to use a burden ANSI rating is twice the terminal voltage developed by the
equal to or less than the standard burden and ensure that the maximum fault current [8]. This ensures operation near the
worst-case secondary fault current is less than 100 A. knee point for the maximum symmetrical fault current.
IST = 110 A RS IS = 100 A
III. ANALYZING CT PERFORMANCE
IE = 10 A
When a CT saturates, it provides distorted information to the
IP VS = 400 V ZB = 4 Ω
connected relay, which can cause the relay to behave
unexpectedly. When selecting a CT for a protective relay
application, it is important to check if the CT saturates for the
levels of fault current that will occur on the system. This section
Fig. 13. Example for a C400, 5 A nominal CT explains an equation that can be used to determine if a CT will
saturate for given fault conditions.
IEEE defines common CT voltage ratings as C100, C200,
C400, and C800. Because the ratings are defined at specific A. CT Steady-State Performance Analysis
voltage levels, not all CTs with the same voltage rating are Equation (3) is the criterion to avoid CT saturation for
created equal. For example, a CT with a terminal voltage of symmetrical fault currents. Real-world faults are rarely
810 V and a CT with a terminal voltage of 1,000 V would both symmetrical, so this equation should never be used practically
be classified as C800. The CT with the excitation graph shown to analyze CT performance. It is explained here and derived in
in Fig. 10 has an excitation voltage of 496 V at an error current Appendix B as a basis for the more practical equation in
of 10 A. To get the terminal voltage VS, we subtract the voltage Section III, Subsection B. Equation (3) can be used to
drop across the secondary winding resistance (RS • 100 A) from determine the maximum allowable fault current for a given
the excitation voltage of 496 V. If the resulting terminal voltage burden or maximum allowable burden for a given fault current.
is over 400 V, we classify the CT as a C400 CT [2]. I FAULT ZB  R S
It is important to note that these voltage ratings only apply •  20 (3)
I PRI ZB STD  R S
to the full winding ratio of the CT, and tapping down a CT
reduces its accuracy. Because both windings are distributed The terms in (3) are defined as follows:
around the same core, they are each subject to the same flux and IFAULT is the maximum fault current in primary amperes
the voltage induced across every turn of any winding is the for a given fault.
same. That is, the maximum volts per turn the CT can support IPRI is the primary current rating of the CT (e.g., for a
without symmetrical saturation is fixed by the cross-sectional 2000/5 CT, IPRI is 2,000.)
area of the core. If we choose to tap a CT at anything other than ZB is the actual burden of the CT’s secondary circuit.
its full winding, there will be less voltage available at the RS is the internal resistance of the CT secondary winding.
terminals because not all of the available turns are connected to ZB STD is the standard burden of the CT (e.g., for a C800
the load. In Fig. 10, for example, we can see that the 300/5 tap CT, the ZB STD is 8 Ω).
 9

ZB includes both the impedance of the connected relay and in less saturation. Therefore, a value over 20 in this equation
the impedance of the leads from the CT to the relay. does not mean that the CT will saturate for that fault current—
Microprocessor-based relays have a negligible burden, but just that it could, assuming the fault current had the worst-case
electromechanical relays do not. The impedance of the leads is dc offset.
determined by the wire gauge, the length of the leads, how the
C. What About Remanence?
CTs are connected, and the fault type. #10 AWG copper wire
has a resistance of 0.9989 Ω per 1,000 feet. Reference [10] There is no way to predict the value of remanence that may
shows how to calculate the resistance of other wire gauges. exist in a CT at a given instant in time. As described in
When using wye-connected CTs and performing the calculation Section II, remanence can either impair or improve the CT
for a single-phase-to-ground fault, a two-way lead burden behavior for a given fault. The worst case is when a fault
should be used because the fault current has to make a full loop produces flux of the same sign as the remanence. In this case,
through the CT circuit (phase and neutral) and back to the relay. the flux change required to saturate the CT equals the difference
When using wye-connected CTs and performing the calculation between the core saturation flux and the remanence.
for a three-phase fault, a one-way lead burden should be used Regardless of how much remanence exists in a CT, recall
because the fault current for a single phase sums with the other from Section II that the effect remanence has on saturation only
two phases at the neutral point of the wye connection and no lasts about half a cycle. Modifications to (5) have been made to
current will flow through the neutral. This is derived in [5], account for the effects of remanence by derating the CT, but
which also shows how to calculate the burden when using delta- they are only valid for the first half cycle of the fault [2]. Using
connected CTs. these modifications to size CTs is unnecessarily conservative
RS, the secondary winding resistance of the CT, is specified and often yields impractically large CTs. Be aware that because
on the CT data sheet. The RS value is directly related to the of the random nature of remanence, even a CT that was properly
resistance of the wire and the number of turns and is therefore sized using (5) may still saturate momentarily. This saturation
dependent on the tap for multiratio CTs. will be short-lived and should have little effect on the
ZB STD is the standard burden of the CT and must be adjusted performance of standard relay algorithms.
if the CT is not tapped at its full ratio. In these instances, the Despite the best intentions, there are certain times when the
standard burden must first be multiplied by the ratio of the tap use of (5) to select CTs that will never saturate becomes an
rating to the full rating. For example, for a C800, 2000/5 CT exercise in futility. One such example is on applications near a
tapped at 1200/5, the standard burden is shown in (4). The generator bus. Here, the X/R ratios and fault current magnitudes
resulting C-rating for the tap would be 4.8 • 100 A = C480. are extremely high, and it is usually impossible to avoid
asymmetrical saturation even with the largest CTs.
800 1, 200 Reference [2] recommends selecting the highest practical rating
ZB STD  •  4.8 (4)
20 • 5 2, 000 for the CTs and always matching the CTs on the terminal and
B. CT Transient Performance Analysis neutral sides of the generator to each other (same ANSI voltage
rating, CT ratio, and connected burden). This way, even though
Section II shows that symmetrical fault currents are not the
we know the CTs will saturate, they will saturate in the same
only risk for CT saturation. Fault currents with an exponentially
way and the saturated waveforms will cancel each other out in
decaying dc offset, caused by the X/R ratio of the system, can
the differential calculation. Because not all CTs with the same
produce significant CT saturation. Selecting CTs based only on
voltage rating are created equal, it is important to make sure that
symmetrical fault current is not advised because it ignores the
both CTs have the same excitation curve, knee-point voltage,
risk of heavy CT saturation when the fault current includes dc
and terminal voltage at 10 A of excitation current. See
offset.
Section III, Subsection D for an example illustrating this point.
To account for dc offset, we can improve (3) by including
Another example of not being able to avoid saturation is in
an extra X/R term, resulting in (5) [5]. This extra term is the
low- and medium-voltage switchgear applications, where fault
X/R ratio of the entire Thevenin equivalent of the system, from
currents are high and space requirements force the use of small
the source to the fault, through the particular CT. Equation (5)
and poorly-rated CTs. Refer to [11] and [12] for selection
only considers dc offset from sinusoidal fault current, not offset
criteria for low-voltage applications. For more information on
from nonsinusoidal sources like transformer inrush.
selecting CTs for various protection applications, see [1], [2],
Appendix C gives an example of using (5) to analyze
[7], [10], [13], and [14].
asymmetrical saturation in a CT.
I FAULT ZB  R S  X  D. Transient Performance of Different CTs With the
•   1  20 (5) Same Rating
I PRI ZB STD  R S  R 
To illustrate the point discussed in the previous section about
The initial magnitude of the dc offset that occurs is trying to make sure that the terminal and neutral-end CTs on a
determined by the fault incidence angle, and the time it takes generator differential have the same excitation curve, we can
for the dc offset to decay is determined by the X/R ratio of the look at two different C800 CTs manufactured to two different
system. Equation (5) is quite conservative and assumes a worst- designs. Fig. 14 shows two CTs that are both classified as C800
case, completely offset waveform, which is not often the case because they each deliver 800 V to a standard burden of 8 Ω at
in the real world. Any fault other than the worst case will result 100 A secondary current. The construction of the two CTs,
 10

though, is different. The CT in Fig. 14a (CT A) is constructed

of more copper (RS is smaller) and less iron (VE is lower), and
the CT in Fig. 14b (CT B) is constructed of more iron (VE is
higher) and less copper (RS is larger).
If we run each CT through (5) for the same fault condition,
we find that the CT in Fig. 14b has a higher chance of
saturation—even if the burden loops are exactly the same.
Assuming a 2000:5, C800 CT being subjected to a single-line-
to-ground fault of 18,000A with an X/R ratio of 25 and an
external burden loop of 1.2 Ω, we get the following for the CT
in Fig. 14a (smaller RS): Fig. 15. Transient model of the CTs in Fig. 14

18, 000 A 1.2 Ω + 1 Ω Fig. 16 shows the calculated operate and restraint values for
• • ( 25 + 1) =
57.2 (6)
2, 000 A 8 Ω + 1 Ω a differential relay connected to these two CTs. We can see that
the false differential current reaches over 25 A secondary for
We get the following for the CT in Fig. 14b (larger RS): almost 1.5 cycles. This relay would have to have a fairly high
18, 000 A 1.2 Ω + 2 Ω slope ratio of 40 percent to ride through the transient differential
• • ( 25 + 1) =
74.9 (7) current from these two CTs with the same rating.
2, 000 A 8 Ω + 2 Ω
100
Even though the CT in Fig. 14b has more iron (VE is higher),
90
transiently, (5) predicts that it will saturate more severely.

Current (secondary A)
80
70
(a) IST = 110 A RS = 1 Ω IS = 100 A Restraint
60
IE = 10 A 50
40
30
IP VE = 900 V VS = 800 V ZB = 8 Ω
20
10 Operate
0
0.000 0.017 0.033 0.050 0.067 0.083 0.010
Time (seconds)
(b) IST = 110 A RS = 2 Ω IS = 100 A
Fig. 16. Operate and restraint values from a differential relay connected to
IE = 10 A the CTs in Fig. 14

IP VE = 1000 V VS = 800 V ZB = 8 Ω
E. CT Analysis Tools
Equation (5) shows how to verify if a CT will saturate for a
given fault current and burden. In addition to this equation,
other tools exist to analyze the behavior of a CT or sets of CTs.
Fig. 14. Two C800 CTs, one with more copper and less iron (a) and one
with more iron and less copper (b)
1) CT Saturation Theory and Calculator
Perhaps the most common is the “CT Saturation Theory and
Using a modeling tool like those described in Section III, Calculator,” an Excel® spreadsheet created by the IEEE Power
Subsection E, we can plot the response of both of these CTs and System Relaying Committee (PSRC) [15]. This spreadsheet is
compare the results. Fig. 15 shows the two CTs modeled in a shown in Fig. 17 and is well documented. Using the available
differential circuit, with the differential current being the tutorials, a level of proficiency can be gained in a short period
difference between the two CTs. We can see that the two CTs of time. The necessary inputs are a description of the CT
do indeed perform very similarly except in the second cycle, (e.g., ratio, accuracy class, and so on) and data on the available
where CT B saturates sooner than CT A as the evaluation of (5) fault current and X/R ratio. The calculator allows the settings
predicted. After the second cycle, the CTs have nearly the same and specification engineer to play “what if” scenarios with the
error and the errors mostly cancel (resulting in little differential CT. Being able to adequately predict the CT’s behavior during
current). fault conditions can help form well-reasoned decisions when
creating relay settings.
 11

Fig. 17. Simulation in the “CT Saturation Theory and Calculator” spreadsheet for a C400 CT [15]

Fig. 18. Simulation in the “CT Saturation Theory and Calculator” spreadsheet for a C800 CT [15]

The simulations in Fig. 17 and Fig. 18 show the difference 2) Mathcad®

between a C400 (Fig. 17) and C800 (Fig. 18) CT selection for Mathcad is another valuable tool that an engineer can use to
the same system and burden conditions. If a user runs this model the output of a CT. A simulation in Mathcad for the C400
simulation for a C400 CT at a certain location, they can easily and C800 CTs in the previous example is shown in Fig. 19. This
see that the CT will saturate for the given fault current. At this simulation uses the Mathcad code from [16]. In both the IEEE
point, the user has some choices. Because the system fault “CT Saturation Theory and Calculator” spreadsheet and
current and X/R ratio cannot be changed, they can try lowering Mathcad, it is possible to create an output file
the burden or using a CT with a higher voltage rating or a higher (e.g., COMTRADE) for simulation with a test set, but it is
turns ratio. Fig. 18 shows that a C800 CT will still saturate but arguably easier to implement in Mathcad. Simulations in
less quickly and severely. Mathcad allow for more user customization, such as changing
the relative permeability of the CT core material.
 12

Fig. 19. Mathcad simulation of C400 and C800 CTs [16]

3) Two CTs In addition to these tools, other software exists to model the
The authors of [8] developed a program that simulates the transient behavior of CTs—many of which are described in
transient behavior of two CTs in a differential circuit. The [17].
program was originally written in BASIC programming
F. Mitigating Saturation
language, but has since been converted to an executable
program that can be run on a Windows® operating system If any of the analysis in the previous subsections proves that
(shown in Fig. 20). In addition to plotting the output of the CTs, CT saturation is a concern, there are several practical solutions
the program produces ASCII and COMTRADE files that can that can be used to mitigate the situation. The most obvious
be used to play the signals into a relay for testing. solution is to use a CT with a higher accuracy class, but
sometimes these CTs are simply too large physically and do not
fit in the given space requirements. Another option is to reduce
the CT burden by adding another set of CT cables in parallel.
Halving the burden has the same impact on saturation as
doubling the accuracy class. Another option is to use CT taps
to double the turns ratio. This has an even larger impact—the
same as quadrupling the accuracy class. This is because
doubling the turns ratio results in halving the secondary current
as well as doubling the available voltage at the CT terminals to
drive that current through the connected burden. Whenever the
turns ratio is increased, ensure that the currents the relay
measures during normal operation are still high enough that
sensitivity is not lost.
Fig. 20. Two CTs simulation program
 13

IV. DETECTING CT SATURATION IN RELAY EVENT REPORTS 2) Filtered Event Reports

Analyzing relay event reports after an operation can be a Filtered event reports are captured after the cosine filter in
challenging task, especially when there is doubt about the Fig. 21. Because relay algorithms also use these quantities, this
validity of the operation. Protective relays depend on accurate makes filtered event reports useful in analyzing protective
current and voltage measurements to do their jobs properly. element behavior after a fault. However, because they contain
When a CT saturates, the current the relay measures does not information after the cosine filter, filtered event reports are
represent what is truly on the power system and can cause the missing any higher-level harmonics and dc offset, leaving only
relay to behave unexpectedly. The ability to detect when CTs a 60 Hz waveform that is symmetrical around the zero-axis.
have saturated using relay event reports is necessary to properly This makes it especially difficult to see CT saturation.
analyze relay operation. An example of the difference between the signals in raw and
filtered event reports is shown in Fig. 22. Fig. 22a shows
A. Raw vs. Filtered Event Reports signals from both an unsaturated and a saturated CT in a raw
Most modern microprocessor-based relays store at least two event report. The waveform from the saturated CT (gray line)
types of event reports: raw (unfiltered) and filtered. Both event is very similar to the waveforms in Fig. 5. Fig. 22b shows what
report types are important and are used for different purposes. this same event looks like in a filtered event report. Notice that
To understand how to detect CT saturation, we first need to the characteristics that make the saturated CT waveform unique
understand the differences between these two report types. are eliminated by the cosine filter, and the resulting signal just
Fig. 21 is a simplified diagram showing how currents are looks like normal fault current. As shown in Fig. 22a, most of
brought into and processed by a microprocessor-based relay. the signal in the saturated waveform exists at the beginning of
The process for voltages is similar. Current measured by a CT each cycle and drops to zero when the CT saturates. Because of
(typically 5 A nominal carried by the CT secondary leads) is this, the resulting phasor (after the cosine filter) will have a
brought to the relay inputs and immediately stepped down by reduced magnitude and leading angle compared with what it
an internal transformer to a milliampere-level signal. This would be if the CT did not saturate [20].
current is dropped across a small resistor, resulting in a
(a)
millivolt-level signal that the relay circuit boards can handle. 100
Current (A secondary)

Next, the voltage is read by an analog-to-digital converter that Iunsat

converts the analog data to digital samples. These samples are 50
then passed through a cosine filter that calculates an equivalent Isat
0
60 Hz phasor. The goal of this filter is to remove all harmonics
outside of the fundamental frequency (60 Hz), as well as any dc –50
offset, to create a 60 Hz signal that can be represented as a
0 2 4 6 8
phasor with a magnitude and an angle. Most relay algorithms Cycles
use these phasors (instead of raw samples) to perform their
protection functions. For more information on relay filtering, (b)
50
Current (A secondary)

see [18] and [19].

Relay Iunsat
0
Isat
Cosine Protection
A/D
Filter Algorithms

Internal Raw Event Filtered –50

Relay Input
Transformer Report Event Report 0 2 4 6 8
Cycles
Fig. 21. Simplified signal processing in a microprocessor-based relay
Fig. 22. CT saturation in a raw (a) and filtered (b) event report [20]
1) Raw Event Reports
3) Which Type of Event Should I Download?
Raw event reports are captured before the cosine filter in
Both raw and filtered event types are useful and necessary
Fig. 21. Because of this, the data in raw event reports most
in event analysis. The type of event report that should be used
closely match the actual signals that occur on the power system.
for analysis depends on whether the protection algorithm being
The waveforms in these event reports include dc offset and
analyzed uses raw or filtered data. The vast majority of
harmonics outside of the fundamental frequency. Current
protection algorithms use filtered data, and thus the filtered
signals in raw event reports are often not as smooth as filtered
event report is the most often analyzed. However, as shown in
event reports because the cosine filter has yet to be applied.
Fig. 22, some power system conditions (CT saturation, inrush,
Event reports downloaded in the COMTRADE file format are
and so on) can only be detected in raw event reports because
always raw data and can be replayed via a test set to a connected
the relay filters remove the characteristics that make them
relay. Users can change settings and replay an event into their
unique. Even the most seasoned event analysis expert will not
relays to verify correct operation.
have much luck looking for CT saturation in filtered event
reports. Be sure to understand whether the data is raw or filtered
 14

when analyzing a relay operation, and always download both the B-phase CT is saturated slightly, and in Fig. 25c, there is
raw and filtered event reports after a fault. severe CT saturation on all three phases.
In addition to whether the event report is raw or filtered, it is
a) 20
important to be aware of the sampling rate of the data in the 15
10 2:IA69
event report. Fig. 23 shows CT saturation error between two

Current
5 2:IB69
CTs that only lasts for a quarter cycle but manages to cause a 0 2:IC69
–5
misoperation [21]. This could easily have been missed if the –10
user was looking at a 4 sample-per-cycle event report. It is –15
important to always download the highest resolution data b) 600
400
possible after a fault. 200 1:IA_A

Current
0 1:IB_A
1:IBW –200 1:IC_A
8,000 –400
1:IBX_NEG
–600
Current (A)

4,000 –800
0 c) 600
400
–4,000 200 1:IA
1:IB

Current
0
–8,000 1:IC
25.030581 25.040581 25.050581 –200
–400
1,400 1:IB_SUM –600
Current (A)

1,000 –800
Time
600
200 Fig. 25. Inrush currents with no CT saturation (a), slight CT saturation (b),
–200 and severe CT saturation (c)
25.030581 25.040581 25.050581
Time (s) 2) Look for Frequency Changes
Depending on the level of CT saturation and how long it
Fig. 23. Difference error from CTs lasting for a quarter cycle persists, it may not always be as obvious as what we have seen
B. How to Detect Saturation in an Event Report so far. Light saturation, while capable of still causing relay
To determine if CTs have saturated, relay event reports can misoperations, does not always give us the typical sawtooth
be inspected for several known characteristics. waveforms shown previously. Fig. 26 shows a current
waveform from a saturated CT that might not be immediately
1) Look for the Characteristic Waveform obvious. In cases like this, we can use the fact that saturated
The simplest way to detect CT saturation is to look in the CTs output currents rich in harmonics to our advantage. One
raw event data for the characteristic “sawtooth” current way to detect higher-level harmonics is to look for frequencies
waveforms shown in Fig. 5. Fig. 24 shows a CT that saturated above 60 Hz.
and then recovered during a phase-to-phase fault [22]. This We can detect a change in frequency by using event analysis
waveform is very similar to the textbook cases, and CT software to measure and compare the periods of different parts
saturation should be immediately suspected. of a signal in a raw event report. In Fig. 26, we placed two
100
vertical lines to create a window that starts at one zero crossing
and ends at another zero crossing approximately 1 cycle later.
60
1:ICW2 The software shows that the time period between these two
Current (A)

20
lines is 16.147 ms, which is very close to 16.666 ms (1 cycle,
–20
based on a 60 Hz nominal frequency). This means that the
–60
current signal between these two points in time is almost a true
–100 60 Hz signal and the CT is not experiencing much saturation.
3 4 5 6 7 8 9 10
Cycles
20,000
Fig. 24. C-phase CT saturates and recovers during a phase-to-phase fault 10,000

Realize that the currents the relay sees will vary depending 0
Curren t

on what system condition is causing the saturation (fault, –10,000

inrush, and so on). In general, look for currents to not accurately –20,000
–30,000
follow the waveform expected for the given system conditions.
–40,000
Take the system condition of inrush on a power transformer, for 16.147 ms
example. A relay sees inrush currents when a downstream 12:26:32.97 12:26:32.99
transformer is first energized, and these inrush currents have an Time

expected signature. Fig. 25 shows three examples of event Fig. 26. No frequency distortion exists when the period of a signal is
reports during an inrush condition, each with varying levels of 16.666 ms
CT saturation. Fig. 25a shows an inrush event with no CT Now, we move the time cursors further along the waveform
saturation—this is what inrush is supposed to look like [22]. and compare the results. In Fig. 27, the period between the two
Fig. 25b and Fig. 25c show a different inrush event. In Fig. 25b,
 15

zero crossings has decreased to 14.616 ms. This shows that the Just like with frequency changes, CT saturation is not the
signal has a frequency higher than 60 Hz, and it could be a sign only cause of harmonic content in current waveforms. The
that the CT is saturating. inrush condition shown in the previous section also generates
large amounts of harmonic content that the relay will measure.
20,000 Measuring harmonic content is just one method that can be used
10,000 when trying to determine if a CT has saturated.
0
4) Look for a Sudden Loss of DC Offset
Curren t

–10,000
In cases where a fault current contains dc offset, the sudden
–20,000 dissipation of this offset can be a sign that CTs have saturated.
–30,000 Theory and testing that shows why dc disappears from the
–40,000 secondary current when saturation occurs is given in [24].
14.616 ms
Fig. 29 gives an illustration of what a sudden loss in dc offset
12:26:32.99 12:26:33.00
Time
looks like. Normally, we would expect a current with dc offset
to slowly decay along an exponential curve (shown in red in
Fig. 27. Frequency is greater than 60 Hz when the period of a signal is less
Fig. 28). We can see that the waveform follows the exponential
than 16.666 ms
curve for the first 2–3 cycles, at which point the dc offset
This same technique can be applied to filtered event reports completely goes away and the waveform is symmetrical around
to a certain extent. When a saturated waveform gets passed the 0 axis. If there is rapid dissipation of dc offset in a raw event
through the cosine filter, there is some distortion in the output report, suspect CT saturation.
compared with when a nonsaturated waveform gets passed
4,000
through. For example, the angle of the filtered current phasor
will start to go in the leading direction. It is important to note 2,000
that distortion in a filtered waveform may not be the result of
0
CT saturation—that is just one possible cause. Step changes in
Curren t

magnitude or current reversals from sequential breaker opening –2,000

(that result in a step change in angle), and evolving faults can
–4,000
cause similar results. However, this distortion is a good hint to
download and analyze the raw event report and check for CT –6,000
saturation. –8,000
40.33.970 40.34.000 40.34.030 40.34.060 40.34.090 40.34.121
3) Look for Harmonic Content
Time
Another way to detect if a CT has saturated is to look at the
individual harmonics. Asymmetrical saturation produces a Fig. 29. Example of a sudden loss of dc offset
large amount of even harmonics and symmetrical saturation 5) Look for False Residual
produces mainly odd harmonics [23]. Event analysis software False residual current is one sign of CT saturation that can
can be used to show the amount of harmonic content in the two be found in filtered event reports. CT saturation in any of the
one-cycle windows of the previous event shown in Fig. 26 and three phase CTs throws off the balance between the phases and
Fig. 27. Fig. 28a shows the second-harmonic content during the creates false residual ground current. Fig. 29 shows a filtered
first one-cycle window where the CT is not saturating (Fig. 26). event report from a delta-wye transformer after an external
Fig. 28b shows the second-harmonic content during the second phase-to-phase fault on a downstream feeder [22]. The top
one-cycle window where the CT is saturating (Fig. 27). Notice traces (W1) show the phase and residual currents on the wye
that there is significantly more second-harmonic content in the side of the transformer. CT saturation is suspected because we
waveform when the CT has saturated. would not expect zero-sequence current to exist for a phase-to-
a) b) phase fault. Also notice how the residual current slowly decays
100
1:IB
with time, which corresponds to the CTs coming out of
80
1:IB
37.58% saturation. The bottom traces (W2) show the phase and residual
Percent

134.4199 (rms)
60 4.61%
120 Hz
currents on the delta side of the transformer. We also suspect
40 4.1226 (rms)
Second Harmonic this to be a false residual current because we would not expect
120 Hz
20 Second Harmonic zero-sequence current to exist on the high side of the delta
0 connection for this fault. Note that only unexpected residual
0 120 240 360 0 120 240 360 current (calculated by the relay as the sum of all three phase
Hz Hz
CTs or wired into a relay input in a residual connection) is a
Fig. 28. Second-harmonic content of current waveform when CT is not sign of CT saturation—unexpected neutral current through a
saturated (a) and is saturated (b) core-balanced CT is not. Whenever CT saturation is suspected
from filtered event reports, the raw data should be downloaded
to verify.
 16

60 Based on the principle of differential protection, it is

20
IAW1 expected that for this external fault, IBW1 will be equal to and
IBW1
5ICW1 180 degrees out of phase with IBW2. The same is true for ICW1
–20
10:01:17.615000 AM and ICW2. Fig. 31 shows the B-phase and C-phase currents
–60
14
compared across the differential. The Winding 2 data points are
10 IRW1.Mag negated to make the difference between waveforms easier to
6 see. If there had been no CT saturation, the curves would be on
2
10:01:17.615000 AM
top of each other. However, we can see that the Winding 1
Current (A)

–2
currents experienced saturation around the 670 ms point in the
80
IAW2
fault data, causing distortion and a difference from Winding 2.
40
0 IBW2
5ICW2 The magnitudes of the differential signals resulting from the
–40 saturation on Winding 1 can be found by adding the Winding 1
10:01:17.615000 AM
–80 and Winding 2 data points, as shown by the black traces
12 in Fig. 31.
8 IRW2.Mag
4
40 IBW1
0 10:01:17.615000 AM
20 NEG_IBW2
–2 –1 0 1 2 3 4 5 6 5
PARTIAL_IBT
Cycles 0

Fig. 30. False residual current appears when CTs saturate during an external –20
fault [22] –40

6) Differential Applications: Look for –60

TREF = 9:53:45 PM

Current (A)
Unmatched Currents
Most misoperations of differential applications occur when
CTs saturate during external faults. To detect saturation in 60 ICW1
differential events, we must remember that the differential NEG_ICW2
5
40 PARTIAL_ICT
principle is based on the expectation of the current entering a
20
zone of protection being equal to the current leaving the zone
of protection during normal conditions. Fig. 30 shows a raw 0
event report from an external B-phase-to-C-phase fault on a –20
two-terminal differential application. Note the subtle change in –40 TREF = 9:53:45 PM
IBW1 and ICW1 currents near the 670 ms vertical (orange) 640 650 660 670 680 690 700
time marker. Although the change looks minor, it was enough Time (ms)
for this relay to misoperate.
Fig. 32. Comparison of phase CTs with Winding 2 negated
80
IAW1
Whenever phase currents across a differential are not equal
40 IBW1 and 180 degrees out of phase during an external fault condition,
5
ICW1 CT saturation is suspected. Remembering to account for CT
0 ratio differences and transformer and CT connections is
important, as these things may also result in the phase currents
–40
not being equal and 180 degrees out of phase during normal
TREF = 9:53:45 PM conditions.
Current (A)

–80

80
V. COMMON MISOPERATIONS BECAUSE OF CT SATURATION
IAW2
40 IBW2 Section III shows guidelines to follow to reduce the
5
ICW2 likelihood of saturation occurring in CTs. Despite our best
0 efforts to follow those guidelines, it is inevitable that sometimes
CTs saturate and feed unreliable signals to the connected relay.
–40
Because relays require undistorted CT secondary current to
TREF = 9:53:45 PM
–80 perform accurate phasor measurements, saturation can lead to
640 650 660 670 680 690 700
misoperations. In other words, “garbage in = garbage out.” This
Time (ms)
section shows two different field cases of CT saturation leading
Fig. 31. External B-phase-to-C-phase fault on a two-terminal differential to misoperations in differential installations. In addition to
these, there are many other references that document the
performance of relay elements during CT saturation, including
[1], [5], [7], [10], [20], and [21].
 17

A. Generator Differential Trips During Inrush generator and the differential to be balanced for this external
A utility feed and a bank of backup generators both feed a inrush event. The fact that this is not the case caused the
switchgear bus. Seven 2,500 kW backup generators are differential element to measure false operate current and trip.
connected to the 13.8 kV generator bus along with a load bank 800
breaker, per Fig. 32. During testing, a 3,500 kVA transformer 400
IA87
IB87
is connected to the load bank breaker, the breaker is closed into 0
5IC87
the dead generator bus, and the generators are started. The relay
–400 7:39:07 PM

Current (A)
protecting the first generator to close in trips on its differential
element. 600
IA
200
IB
5IC
–200
–600 7:39:07 PM

170 190 210 230 250 270

Time (ms)

Generator Fig. 35. Raw event report recorded when generator relay trips during startup
Switchgear
Recall from Section II that the presence of asymmetrical
fault current can result in CT saturation. This is because of the
dc offset in the primary current causing flux to accumulate in
A1 one direction more than the other, resulting in the CT eventually
Switchgear hitting its positive or negative volt-time area limit. An inrush
waveform (Fig. 34) is harder on CTs than asymmetrical fault
current because all of the currents are monopolar. The C-phase
current, for instance, only allows accumulation of flux in the
Fig. 33. One-line diagram of utility feed and backup generators positive direction, resulting in hitting the maximum volt-time
area much faster than if the current waveform dropped below
The differential relays for each backup generator are zero for some period of time to allow the CT to partially
connected as shown in Fig. 33. recover.
Generator Fig. 35 shows each phase current from the bus-side CTs
A
52 Switchgear compared with its corresponding phase on the neutral side. The
B
bus-side CT signals have been flipped 180 degrees to account
G
C
for the difference in CT polarity. We can see that the signals
start out completely equal to each other (which is expected) and
then fall away as the neutral CTs saturate. C-phase is the worst
IC IC87 offender, and it is the phase that tripped the relay.
IB IB87 350
IA IA87 250 IA
IA87_NEG
150
Fig. 34. Connections of differential relays on backup generators 50
Fig. 34 shows the raw event report that the differential relay –50
7:39:07 PM
recorded. Notice the difference between the bus-side currents –150

and the neutral-side currents. The bus-side currents show a 600

Current (A)

IB
classic, although saturated, inrush waveform (as described 400
IB87_NEG
previously in Section IV, Subsection B), which makes sense 200
0
because of the downstream transformer being energized. 7:39:07 PM
–200
Section III mentions that we should always match the phase and
neutral-side CTs in generator applications to account for the 200
0 IC
saturation that will inevitably occur. We know the CTs will –200 IC87_NEG
saturate, but matching the CTs will make sure they saturate in –400
the same way, and the signals will cancel out in the differential. –600 7:39:07 PM

In this example, because both sets of CTs see the same current 170 190 210 230 250 270
Time (ms)
with a 180-degree polarity difference, we would expect the
same inrush waveform to be mirrored on the neutral side of the Fig. 36. Currents from bus-side CTs vs. neutral-side CTs
 18

Upon investigation, it was found that the CTs in fact were currents remain steady throughout a fault, is an indication of CT
sized the same on both sides of the generator—both 200/5, C20, saturation [22].
intentionally sized to the generator’s full-load ampere rating of
2
150 A. With CTs rated similarly, it is unlikely that one set IOP1
would perform so differently than the other. It was then found IRT1
1
that the length of the CT secondary leads between the CTs and
the relay were considerably different. The one-way lead length 0

of the bus-side CTs was 15 feet, while the neutral-side CTs

were 20 times longer—300 feet! This greatly increased the

Current (A)
0.8 IOP2
IRT2
burden of the neutral-side CTs (2.44 Ω for the neutral side vs. 0.4
0.17 Ω for the bus side) and resulted in the difference in 0
saturation behavior.
The ideal solution to this problem is to match the lead length 1.2
IOP3
burden on both sets of CTs to where they saturate similarly and 0.8 IRT3
balance each other out. Until that could be accomplished, a 0.4

temporary measure was taken by forcing the relay into high 0

security mode for a short time during startup, when the –4 –2 0 2 4 6

Cycles
generator breaker first closed. High security mode was then
programmed to have a higher minimum operate current and a Fig. 38. Filtered differential event shows decreasing operate current over
time
more secure slope setting than the standard settings.
Fig. 38 shows the raw event report. Here we can see that the
B. Transformer Relay Operates During DC Offset
A-phase currents on Winding 1 and Winding 2 have a
In March of 2017, a transformer differential relay significant amount of dc offset at the beginning of the fault.
misoperated on an external A-phase-to-ground fault on the low After about seven cycles, the dc offset goes away and the
side of a 138/69 kV wye-wye autotransformer. The filtered waveforms balance out. Note that this dc offset cannot be seen
event report from this operation is shown in Fig. 36. The relay in the filtered event report because it is removed by the cosine
and CTs are configured such that the taps on both sides are filter. The dc offset itself does not cause the misoperation, but
equal, and thus we would expect the A-phase currents on it does cause the CTs to saturate. The bottom graph in Fig. 38
Winding 1 and Winding 2 to be equal and 180 degrees out of shows the A-phase current on Winding 1 compared with the
phase with each other. We can tell from the waveforms on the inverse of the A-phase current on Winding 2, and it proves that
bottom of Fig. 36 that this is not the case. We would also expect these signals are not equal and 180 degrees out of phase because
the nonfaulted phases to remain balanced, with 120-degree the CTs have saturated. Notice that the period of the event that
separation—but, Fig. 36 shows that this is also not the case. We has the most difference between the current signals in Fig. 38
can suspect CT saturation, but we will not know for sure until corresponds to the period in the event with the highest operate
we look at the raw signals. current in Fig. 37. It is this difference in CT behavior that
10 causes false operate current and results in the relay tripping on
6 IAW1 its differential element. A very similar event is shown in [22]—
2 IBW1
5ICW1 an external phase-to-ground fault on the wye side of a delta-
–2
–6 wye transformer with significant dc offset.
6:19:22.971000 AM
–10
15
15 5 IAW1
Current (A)

IAW2 IBW1
5 5ICW1
IBW2 –5
5ICW2
–5 –15
6:19:22.971000 AM 6:19:22.971000 AM
–15 –25
15 25
IAW1 15 IAW2
IBW2
Current (A)

5 IAW2 5ICW2
5
–5
–5
6:19:22.971000 AM 6:19:22.971000 AM
–15 –15
–3 –1 1 3 5
Cycles 15
IAW1
Fig. 37. Filtered event report for external A-phase-to-ground fault 5 IAW2_NEG
–5
Fig. 37 shows the filtered differential event report. Here, we
–15
see the restraint currents steadily decreasing throughout the
6:19:22.971000 AM
event, while operate currents increase and then decrease. –25
–5 –3 –1 1 3 5
Decreasing operate current over time, while filtered phase Cycles

Fig. 39. Raw event report shows significant dc offset

 19

Several solutions could be implemented to prevent this type B. External Fault Detector Enables High Security Mode
of misoperation. In this case, the utility chose to temporarily When an external fault occurs and causes the CTs to
raise the minimum operate current threshold and slope setting saturate, there is a small window of time when the fault first
in the differential relay. At a later date, they plan to increase the occurs that the CTs have not yet saturated. During this time
taps of the CTs to allow for a higher voltage before saturation period, IRT increases because of the elevated fault current, but
(from 300/5 to 800/5 on the 138 kV side, and from 600/5 to IOP does not. We can use this to our advantage to develop logic
1200/5 on the 69 kV side) and return the slope to its previous that detects an external fault before the CTs have had a chance
setting. The external fault detection algorithm described in the to saturate. When the relay detects a large change in IRT with no
next section would also have prevented this misoperation (or very small) change in IOP, it declares an external fault. This
because restraint current increased in all three phases is shown in Fig. 39. The success of this logic depends on CTs
significantly before the operate current increased. This would providing valid output for some amount of time (2 ms to a half
have put the relay into a more secure mode with a higher slope cycle) [14] [21].
setting and kept it there until the dc offset decayed.
DIO PR
+
∆IOPR DIO P
VI. RELAY ALGORITHMS ACCOUNT FOR SATURATION DIO PP –

Even with comprehensive analysis and our best intentions, DIRTR

+
DIRT
it is often difficult or impossible to avoid saturation for a given ∆IRTR
DIRTP –
system. In these instances, relay algorithms may be able to
assist by either desensitizing or delaying operation during Fig. 40. External fault detection logic
CT saturation.
Using this logic, a relay can automatically desensitize itself
A. Percentage-Restrained Differential to an external fault (when CT saturation is likely to occur).
A percentage-restrained differential relay works off the Although not every external fault will result in saturated CTs,
principle that the current going into a zone of protection must it is best to default to a higher security (less sensitive) protective
equal the current leaving the zone under balanced conditions. mode for a short period of time. This higher security mode can
This can be implemented by calculating an operate and restraint include a higher slope value and/or a higher minimum operate
quantity using (8). current required to operate the differential element. Some relays
  use the second-harmonic content present in the waveforms of
IOP  I1  I2 saturated CTs to extend the time the differential element is in
  (8) high security mode [14].
I RT  I1  I2
C. Cosine-Peak Adaptive Filtering
Because the CTs around the zone are assumed to have In switchgear applications, CTs are notoriously undersized,
opposing polarities, the operate quantity (IOP) calculates as the primarily because of space restrictions. Higher accuracy classes
difference between the current into and out of the zone. Under require more turns or more iron and result in physically larger
normal load or through-fault conditions, IOP should equal 0, and CTs. Even when it is known that a higher accuracy class is
under internal fault conditions, IOP calculates to a value. needed, there is simply not room for these larger CTs in many
However, if one of the CTs enclosing the zone saturates during switchgear lineups. In addition, switchgear is serving loads in
an external fault, the currents from the CTs no longer cancel some of the highest magnitude fault current locations in the
each other out exactly, and IOP calculates to some value. We do power system. This combination yields fault currents that can
not want to trip for this condition, so we must add a margin to exceed 200 times the CT primary current rating and all but
provide security for CT errors. This is normally done by assures that saturation will regularly occur. Overcurrent
comparing IOP with IRT times a slope (see (9))—instead of elements are used very frequently in these applications, so it is
comparing it with a fixed value—which can be set to an important that they function correctly, even during saturation.
appropriate margin to account for CT saturation errors. This Section IV shows that current magnitudes and angles are
works well because when currents are high (high values of IRT) calculated from analog data using a cosine filter. This cosine
and the chance for saturation is greater, IOP needs to overcome filter is excellent at removing dc offset and harmonics, but it is
a larger threshold before the relay trips. Similar to the slope in slow to provide the correct magnitude when CTs are saturated.
percentage-restrained differential relays, the angle of the A slow filter output results in overcurrent elements taking
characteristic of an alpha-plane differential relay also helps longer to operate for a fault because the reported current
account for CT saturation differences between the two sets of magnitude stays below the pickup threshold for a longer time.
CTs. Another type of filter, the bipolar peak detector, provides a
IOP  SLP • I RT (9) better estimation of the current magnitude during CT saturation.
The response of these two filters to a saturated signal from a CT
is shown in Fig. 40, along with an instantaneous overcurrent
pickup setting of 80 A secondary.
 20

200
80 A • There are two types of CT saturation: symmetrical
A/D Output Cosine Filter Peak Filter
150
INST saturation and asymmetrical saturation. Symmetrical
saturation is caused by symmetrical fault currents high
100 in magnitude, while asymmetrical saturation is caused
by fault currents with dc offset.
50
• When a breaker trips during asymmetrical current
Current (A)

0 (before the dc component dissipates), remanence can

remain in the CT core and affect its behavior when it
–50
is next energized.
–100 • A simple CT equivalent circuit can be used to model
CT behavior using rms quantities.
–150
• The knee point of a CT excitation graph is not the
–200 same thing as the saturation voltage. The accuracy
0 1 2 3 4 5 6
Cycles class voltage rating of the CT is a better estimate of
the saturation voltage.
Fig. 41. Response of cosine filter and peak detector filter to a saturated
waveform [24] • IEEE defines the voltage rating as the CT secondary
voltage that the CT delivers when it is connected to a
When an instantaneous overcurrent element is set with a standard secondary burden at 20 times rated secondary
pickup of 40 A secondary or higher, modern relay algorithms current without exceeding a 10 percent ratio error [2].
can use harmonic content to detect when CTs are saturating and • The voltage rating for a CT is only valid for its full
switch between the cosine and bipolar peak detector filters ratio and changes if the CT is tapped down.
accordingly. This results in secure performance at lower fault
• Equation (5) can be used to determine the maximum
currents and faster performance during CT saturation. For more
allowable fault current for a given burden, or
information on the bipolar peak detector filter, see [25], [26],
maximum allowable burden for a given fault current,
and [27].
while avoiding saturation.
Although modern relay algorithms can help accommodate
• It is important to understand the difference between
some level of CT saturation, these algorithms are not an excuse
raw and filtered event reports and to download both
to bypass the calculations in Section III or use underrated CTs.
types of data (at the highest sampling rate available)
If saturation is expected and the relay algorithms are depended
from relays after a fault.
upon to account for it, it is always best to simulate the fault
• CT saturation can be detected in raw event reports by
using a transient program to validate how the relay responds.
looking for the characteristic waveform, frequency
changes, or harmonic content. Filtered event reports
VII. CONCLUSION
can be inspected for false residual current.
The following are important takeaways from this paper: • Saturated CTs feed incorrect signals to relays and can
• A CT works by the principle of electromagnetic result in relay misoperations, mainly in differential
induction. Current flowing in the primary winding of a applications.
CT generates an alternating magnetic flux, which in • Although many existing protection algorithms account
turn induces an alternating voltage across the for some degree of CT saturation, this does not excuse
secondary winding. the engineer from performing due diligence in
• The iron core of a CT is made up of a fixed number of selecting appropriate CTs for the application.
molecular magnets, which line up dynamically with
the alternating magnetic field. When all the magnets
are aligned in the same direction, the maximum flux
density is reached and the CT core is said to be
saturated.
 21

VIII. APPENDIX A Therefore, the actual secondary current coming out of the
To meet the IEEE C57.13-1993 standard, a CT must not CT must be greater than or equal to 90 percent of what we
exceed 10 percent ratio correction (PRC) [28]. The standard expect it to be. This means that only 10 percent of the expected
defines the following: signal can be lost as error.
• The ratio correction factor (RCF) is the ratio of the IEEE C57.13-2016 states that CTs must have ≤10% ratio
true ratio to the marked ratio. The primary current is error to meet the standard [9]. Ratio error is defined similarly
equal to the secondary current multiplied by the to PRC in IEEE C57.13-1993, but with an added sine term to
marked ratio times the RCF. account for the fact that IE and IS are not exactly in phase for
• The percent ratio correction (PRC) is the difference cases when the connected burden has a resistive component.
between the RCF and unity, expressed in percent. When the connected burden is not purely inductive, IE and IS
We can write the definition of RCF as shown in (10), where cannot simply be added algebraically, and this extra term is
CTR is the marked ratio of the CT. needed for accuracy.
IEEE C57.13-2016 also defines the composite error the
I P / IS same way IEEE C57.13-1993 defines PRC, and it says that this
RCF =
CTR can be used in place of ratio error for ring-type cores of low
(10)
IST reactance. For more on CT measurement errors, see [29].
RCF =
IS For simplicity in this paper, we define ratio error the same
as PRC, as shown in (16).
We can write the definition of PRC as shown in (11).
IE
PRC
= ( RCF − 1) •100 (11) Ratio Error (%) = •100 (16)
IS
Substituting (10) into (11), we get (12).
I  IX. APPENDIX B
=  ST − 1 •100
PRC
 IS  Recall from Section II that the ANSI voltage rating of a CT
defines the minimum secondary voltage (VS) that the CT must
I I 
=  ST − S  •100
PRC deliver to a standard burden at 20 times rated secondary current
 IS IS  (12) without exceeding a 10 percent ratio error [2]. If the fault
I −I  current through a CT is higher than 20 times the rated current,
PRC =  ST S  •100 or the connected burden is higher than the standard burden, we
 IS  may risk going into saturation (over 10 percent error). We can
IE use the definition for the ANSI standard voltage rating to
PRC = •100
IS develop an equation for the level of fault current (or connected
burden) that, if we go above, will result in more than 10 percent
IEEE C57.13-1993 states that (12) must not exceed
error. We can call this the saturation point for symmetrical
10 percent [28]. We want to know what this limit means in
current (symmetrical saturation).
terms of how the actual secondary current corresponds to the
We can write the definition of the ANSI standard voltage
expected, or ideal, secondary current. A more intuitive form of
rating as follows:
I I
calculating CT error would be S_actual , or S . We can write VSTD  20 • IS RATED • ZB STD (17)
IS_expected IST
this as shown in (13). where:
VSTD is the secondary terminal voltage rating
IS IS
= (13) IS RATED is the rated secondary current
IST IS + I E
ZB STD is the standard burden
If we divide all terms on the right side by IS, we get (14). Fig. 41 is a reproduction of Fig. 8, for convenience.
IS 1 IST RS IS
= (14)
IST I IE
1+ E
IS
IP VE VS ZB
IE ZE
Since ≤ 10 percent, we get (15).
IS
N1 N2
IS
≥ 0.909 (15)
IST Fig. 42. Simplified CT equivalent circuit
 22

Using the equivalent circuit in Fig. 41 and the definition of Because we are concerned with primary fault current values,
the ANSI voltage rating, we know that the ratio error will not we can incorporate the CT ratio and convert the first term in
exceed 10 percent as long as the secondary terminal voltage VS (27) to primary values. This results in (28).
is less than the secondary terminal voltage rating VSTD, as I FAULT ZB  R S
shown in (18). •  20 (28)
I PRI ZB STD  R S
VS  VSTD (18)
where:
The voltages VS and VSTD are driven by the core excitation IFAULT is the maximum fault current in primary amperes
voltage, VE. Therefore, we can also write the relationship in for a given fault.
(18) as (19), which considers the voltage drop across the IPRI is the primary current rating of the CT (e.g., for a
secondary winding resistance RS. 2000/5 CT, IPRI is 2000.)
VE  VE STD (19) ZB is the actual burden of the CT’s secondary circuit,
which includes both the impedance of the connected relay
We can write the definition of VE STD as: and the impedance of the leads from the CT to the relay.
VE STD  20 • IS RATED •  ZB STD  R S  (20) RS is the internal resistance of the CT secondary winding.
ZB STD is the standard burden of the CT (e.g., for a C800
We can write the actual voltage VE as:
CT, the ZB STD is 8 Ω.)
VE  IS •  ZB  R S  (21) Equation (28) can be used to evaluate CT performance for
Therefore, based on the relationship in (19): symmetrical fault currents but should never be used in practice.
This derivation of (28) is shown simply as a basis for (5).
IS •  ZB  R S   20 • IS RATED •  ZB STD  R S  (22)
Equation (28) is not adequate for analyzing how a CT will
Rearranging terms, we obtain: behave in the presence of (typically asymmetrical) fault current,
IS ZB  R S and (5) should always be used in practice. For even better
•  20 (23) accuracy, use a transient modeling program to model CT
IS RATED ZB STD  R S
behavior in the time domain.
To make this equation useful, we must write it in terms of Note that the basis for (28) was first introduced in [1]. Later,
the secondary fault current, which is IST in Fig. 41. IST is the in [5], the original equation was adjusted to consider the CT
sum of IS and IE. For the worst case right at the saturation point, secondary resistance RS, resulting in the more accurate analysis
[9] defines the error as 10 percent (IE/IS ≤ 0.1). Using this reflected here.
relationship, we can define:
X. APPENDIX C
IS  0.909 • IST (24)
Equation (5), repeated in (29), is used to analyze CT
We can then write (23) as: saturation resulting from fault current with dc offset.
0.909 • IST ZB  R S A C800, 2000/5 CT with RS = 0.5 Ω is connected to a 1 Ω
•  20 (25)
IS RATED ZB STD  R S burden (including relay and lead resistance). The system X/R
ratio is 12. What is the maximum primary three-phase fault
If we define IST as the secondary fault current If sec, (25) current that can be applied to this CT without exceeding the
becomes: maximum flux density (which is proportional to the volt-time
If sec ZB  R S area)?
•  22 (26)
IS RATED ZB STD  R S To determine the maximum fault current, set (5) to the
maximum of 20 and solve for IFAULT, as shown in (29) and
We can use (26) to solve for the maximum fault current (30).
(If sec) that the CT can handle before it goes into saturation. We
can also use it to solve for the maximum burden for a given I FAULT ZB  R S  X 
•   1  20 (29)
fault current. I PRI ZB STD  R S  R 
In development of (26), we assumed an inductive burden. I FAULT 1   0.5 
Therefore, IS and IE are in phase and can be added algebraically • 12  1  20
2000 8   0.5  (30)
to get IST. This assumption also allows us to use the simplified
error equation for ratio error in Appendix A. Using modern I FAULT  17.4 kA
resistive burdens with these simplified equations produces
errors. To account for these errors, a more conservative form of XI. REFERENCES
(19) is used: [1] S. E. Zocholl, Analyzing and Applying Current Transformers.
Schweitzer Engineering Laboratories, Inc., Pullman, WA, 2004.
If sec ZB  R S
•  20 (27) [2] IEEE Standard C37.110-2007, IEEE Guide for the Application of
IS RATED ZB STD  R S Current Transformers Used for Protective Relaying Purposes.
 23

[3] D. Subedi and S. Pradhan, “Analyzing Current Transformers Saturation proceedings of the 68th Annual Conference for Protective Relay
Characteristics for Different Connected Burden Using LabVIEW Data Engineers, College Station, TX, March–April 2015.
Acquisition Tool,” International Journal of Electrical, Computer, [22] D. Costello, “Lessons Learned Through Commissioning and Analyzing
Energetic, Electronic and Communication Engineering, Vol. 9, Data From Transformer Differential Installations,” proceedings of the
Issue 10, 2015. 60th Annual Conference for Protective Relay Engineers, College
[4] B. Kasztenny, N. Fischer, D. Taylor, T. Prakash, and J. Jalli, “Do CTs Station, TX, March 2007.
like DC? Performance of Current Transformers With Geomagnetically [23] L. F. Kennedy and C. D. Hayward, “Harmonic-Current-Restrained
Induced Currents,” proceedings of the 69th Annual Conference for Relays for Differential Protection,” Electrical Engineering, Vol. 57,
Protective Relay Engineers, College Station, TX, April 2016. Issue 5, May 1938, pp. 262–271.
[5] M. Thompson and R. Folkers, “Secure Application of Transformer [24] B. Kasztenny, N. Fischer, D. Taylor, T. Prakash, and J. Jalli, “Do CTs
Differential Relays for Bus Protection,” proceedings of the 58th Annual like DC? Performance of Current Transformers With Geomagnetically
Conference for Protective Relay Engineers, College Station, TX, Induced Currents,” proceedings of the 66th Annual Conference for
April 2005. Protective Relay Engineers, College Station, TX, April 2016.
[6] S. Hodder, B. Kasztenny, N. Fischer, and Y. Xia, “Low Second- [25] L. Underwood, “Current Transformer Selection Criteria for Relays with
Harmonic Content in Transformer Inrush Currents – Analysis and Adaptive Overcurrent Elements,” April 2005. Available: selinc.com.
Practical Solutions for Protection Security,” proceedings of the
[26] S. E. Zocholl and J. Mooney, “Primary High-Current Testing of Relays
67th Annual Conference for Protective Relay Engineers, College With Low Ratio Current Transformers,” proceedings of the 57th Annual
Station, TX, March 2014. Conference for Protective Relay Engineers, College Station, TX,
[7] H. J. Altuve, N. Fischer, G. Benmouyal, and D. Finney, “Sizing Current April 2004.
Transformers for Line Protection Applications,” proceedings of the [27] G. Benmouyal and S. E. Zocholl, “The Impact of High Fault Current and
66th Annual Conference for Protective Relay Engineers, College CT Rating Limits on Overcurrent Protection,” proceedings of the
Station, TX, April 2013. 56th Annual Conference for Protective Relay Engineers, College
[8] S. E. Zocholl and D. W. Smaha, “Current Transformer Concepts,” Station, TX, April 2003.
proceedings of the 46th Annual Georgia Tech Protective Relaying [28] IEEE Standard C57.13-1993, IEEE Standard Requirements for
Conference, Atlanta, GA, April 1992.
Instrument Transformers.
[9] IEEE Standard C57.13-2016, IEEE Standard Requirements for
[29] Forest K. Harris, Electrical Measurements. John Wiley & Sons,
Instrument Transformers. Hoboken, NJ, 1952.
[10] G. Benmouyal, J. Roberts, and S. E. Zocholl, “Selecting CTs to
Optimize Relay Performance," proceedings of the Pennsylvania Electric
Association Relay Committee Fall Meeting,” September 1996. XII. BIOGRAPHIES
[11] S. Manson and A. Upreti, “Current Transformer Selection Techniques Ariana Hargrave earned her B.S.E.E., magna cum laude, from St. Mary’s
for Low-Voltage Motor Control Centers,” proceedings of the University in San Antonio, Texas, in 2007. She graduated with a Master’s of
63rd Annual Petroleum and Chemical Industry Technical Conference, Engineering in electrical engineering from Texas A&M University in 2009,
Philadelphia, PA, September 2016. specializing in power systems. Ariana joined Schweitzer Engineering
Laboratories, Inc. in 2009 and works as a protection application engineer in Fair
[12] J. R. Linders, C. W. Barnett, J. W. Chadwick, P. R. Drum, Oaks Ranch, Texas. She is an IEEE member and a registered professional
K. J. Khunkhun, W. C. Kotheimer, P. A. Kotos, D. W. Smaha, J. W. engineer in the state of Texas.
Walton, P. B. Winston, and S. E. Zocholl, “Relay Performance
Considerations With Low-Ratio CTs and High-Fault Currents,”
Michael J. Thompson received his B.S., magna cum laude, from Bradley
IEEE Transactions on Industry Applications, Vol. 31, Issue 2,
University in 1981 and an M.B.A. from Eastern Illinois University in 1991.
March–April 1995.
Upon graduating, he served nearly 15 years at Central Illinois Public Service
[13] S. E. Zocholl, “Rating CTs for Low Impedance Bus and Machine (now AMEREN). Prior to joining Schweitzer Engineering Laboratories, Inc.
Differential Applications,” proceedings of the 27th Annual Western (SEL) in 2001, he was involved in the development of several numerical
Protective Relay Conference,” Spokane, WA, October 2000. protective relays while working at Basler Electric. He is presently a Fellow
[14] M. Donolo, A. Guzmán, M. V. Mynam, R. Jain, and D. Finney, Engineer at SEL Engineering Services, Inc. He is a senior member of the IEEE,
“Generator Protection Overcomes Current Transformer Limitations,” member of the IEEE PES Power System Relaying and Control Committee, past
proceedings of the 41st Annual Western Protective Relay Conference, chairman of the Substation Protection Subcommittee of the PSRCC, and
Spokane, WA, October 2014. received the Standards Medallion from the IEEE Standards Association in
2016. Michael is a registered professional engineer in six jurisdictions, was a
[15] IEEE Power System Relaying and Control Committee, “CT Saturation contributor to the reference book “Modern Solutions for the Protection Control
Theory and Calculator.” Available: http://www.pes-psrc.org. and Monitoring of Electric Power Systems,” has published numerous technical
[16] ECE 525: Power Systems Protection and Relaying (Lecture, University papers and magazine articles, and holds three patents associated with power
of Idaho, Moscow, ID, February 2016). Available: www.ece.uidaho. system protection and control.
edu/ee/power/ECE525/Lectures/L8/CT_Model.pdf.
[17] R. Folkers, “Determine Current Transformer Suitability Using EMTP Brad Heilman, P.E., has a B.S.E.E. degree from the South Dakota School of
Models,” proceedings of the 26th Annual Western Protective Relay Mines and Technology. From 1990 to 1997, he was employed as a system
Conference, Spokane, WA, October 1999. protection engineer for Black Hills Power & Light. His utility work included
[18] E. O. Schweitzer, III and D. Hou, “Filtering for Protective Relays,” transmission line and generator protection, power plant and substation
proceedings of the 19th Annual Western Protective Relay Conference, commissioning, and SCADA system upgrades. In 1997, he joined Schweitzer
Engineering Laboratories, Inc. as a field application engineer supporting
Spokane, WA, October 1992.
customers in Arizona, California, Montana, Wyoming, Utah, North Dakota,
[19] S. E. Zocholl and G. Benmouyal, “How Microprocessor Relays Respond Nebraska, Nevada, Idaho, and South Dakota. He is an IEEE member and
to Harmonics, Saturation, and Other Wave Distortions,” proceedings of licensed professional engineer in the state of South Dakota.
the 24th Annual Western Protective Relay Conference, Spokane, WA,
October 1997.
[20] J. Mooney, “Distance Element Performance Under Conditions of CT
Saturation,” proceedings of the 61st Annual Conference for Protective Previously presented at the 2018 Texas A&M
Relay Engineers, College Station, TX, April 2008. Conference for Protective Relay Engineers.
[21] D. Costello, J. Young, and J. Traphoner, “Paralleling CTs for Line © 2018, 2019 IEEE – All rights reserved.
Current Differential Applications: Problems and Solutions,” 20190418 • TP6811-01