Breaker Monitoring with Numeric Relays

By John Horak and Gerald Dalke

Basler Electric
There is a number of functions that a digital or numeric relay can perform with respect to what is
occurring in a power circuit breaker:

Predictive Maintenance
o Circuit breaker duty/Contact wear monitoring
o Number of operations
o Slow circuit breaker trip
Circuit Breaker Failure
o Failure to trip
o Open trip coil
o Failure to close
Recording Capabilities
o Oscillography
o Sequence of events
o metering
o Demand data and alarms

This paper will focus on the first item in the list, capabilities of numeric protective relays to
monitor breaker wear. An acknowledgement is given to the recording capabilities of relays to do
"post mortem" analysis of events and perform breaker failure relaying (50BF). Figure 1 shows
several such recording features of various events: an oscillographic record, a fault summary,
and a phasor representation of different events. Event reports such as these, and the breaker
failure function, are expected as a matter of course in modern numerical relays from any serious
player in the market, so these features are no longer the "new cool toy" in today's market. This
presentation is, instead, aimed at the question, "What are the issues associated with Breaker
Duty/Contact Wear Monitoring from the perspective of what the modern numerical protective
relay can easily monitor?"

What are the difficulties?

The difficulties can be broken down into three categories:
a) Unclear Circuit Breaker Ratings: There is a multitude of circuit breakers on the market with
varying interrupter duty ratings as stated by the manufacturer, but it can be difficult to ascertain
how these ratings correspond to actual conditions to which the breaker is subjected. While SF6
and vacuum technologies rule in new circuit breaker designs, there is a multitude of air blast,
and air-magnetic circuit breaker designs in service, although no longer manufactured, and oil
circuit breakers usually only for special applications. Within any one general design, each
manufacturer has a unique way of doing things, and breaker design evolves over time.
b) Older Designs: If starting a monitoring program of an older design breaker today, it will be
very difficult to account for the history of the breakers maintenance and past application.

Prepared by Basler Electric Company

Revised 06/10/05

c) Contact Wear Variables: There are many factors that are not easily determined which are
associated with the wear a circuit breaker may experience during an operation. Some factors
that affect performance include the following:

Various circuit breaker designs;

o Oil, Air-Magnetic, Air Blast, Vacuum, SF6; each has a different level of internal arcing
energy and arcing duration;
o Within any one manufacturers design, there may be additional sub-designs
Energy in a circuit breaker arc is hard to analyze; protective relays cannot or do not
commonly have the ability to determine the following:
o Control system voltage effects on mechanism operating speed;
o Changing Rarc of circuit breaker;
o Current magnitude throughout the operation, such as evolving fault effects and faults
with some dc offset remaining at point of interruption;
o Arcing time (measurement may require data entry from a breaker travel analyzer);
o Variations in circuit breaker speed from one operation to next;
o Where contacts part in the voltage/current wave;
o Power factor of interrupted current;
o CT saturation, particularly saturation caused by dc offset;
o Fast reclose on heated circuit breaker contacts
o Multiple breaks on HV circuit breakers and lack of coordination of multiple break;
o Additional wear that may be caused by nonstandard reclosing internals;
o For number-of-operations-based maintenance, differentiating between no-load, load
at rated current, and fault conditions;
o Mechanical condition of the circuit breaker, and its history of previous operations and
breaker maintenance.

With this extensive list of things that can confuse a relay that is monitoring circuit breaker wear,
it should be apparent that the reports available from a relay are a highly approximated number
giving general concepts and, hence, cannot be exclusively relied upon for an accurate
measurement of when a circuit breaker needs internal maintenance.

Figure 1a, 1b, 1c: Event Report Formats

Circuit Breaker Designs and Circuit Breaker Wear

Oil Circuit Breakers
Few oil circuit breakers are being manufactured today, but many are still in service. The oil
serves two purposes: It is an insulator between energized elements and the grounded portions
of the breaker, and it acts to quench the arc. When the contacts part, an arc begins and
vaporizes a small portion of the oil. This vaporization absorbs substantial amounts of energy
from the arc. The gas that is created is mostly hydrogen, which does not remain in the ionized
state easily. Further, the oil cools the arc by thermal conduction to the surrounding bulk oil. The
energy required to maintain the gas bubble is large, and the oil can cool and quench the arc at
the current zero cross.
The contacts of dead tank oil circuit breakers are enclosed in metal, grounded tanks (Fig. 2A).
In the early and lower voltage designs the arc is allowed to occur without any special arc
extinguishing scheme except the one provided by the cooling act of the oil surrounding the arc
gas bubble. In higher voltage and more advanced designs, arc extinguishing is supplemented
with specially-designed arc chutes and directed oil flow surrounding the breaker contacts (Fig.
2B). In cross blast interrupters, the arc is pulled in front of several lateral vents and forced into
the vents, which lengthens the arc and shortens the interruption time. The axial blast
3

interrupters use a similar principle as the cross blast, but the arc is not as directed, which tends
toward poorer performance.
The arc control for the minimum oil breakers is based on the same principle as the arc control
devices of the bulk oil breakers, but is designed to reduce oil requirements. This is
accomplished by a) utilizing a live tank design; therefore removing the role of the oil as
insulation between the conductors and ground, and b) utilizing an oil injection technique to
extinguish the arc.

Series Break:
As for all breaker designs, for higher voltages the interrupters are arranged in series. For series
break devices, the design must ensure that all breaks occur simultaneously so that the
interrupted voltage is divided equally across the breaks during the interrupting process in order
to create even wear and high interrupt capacity. A natural voltage division depends on any
unreliable stray capacitances between the contacts and to the ground and, therefore, relies on
capacitances or resistors in parallel with the interrupting heads to balance the voltages between
interrupters. Series break is shown and discussed further later (Fig. 12).
Opening
Motion

Oil Level
Stationary Contact
Arc

Arc
Arcing Tip
Liner
Moving Contact

Figure 2A: Plain Break Bulk Oil Breaker

Oil
Flow

Oil
Flow

Opening
Motion

Axial Blast

Cross Blast

Figure 2B: Arc Controlled Oil Breaker

Air Break Circuit Breakers

Air-break circuit breakers (sometimes referred to as air-magnetic circuit breakers) extinguish an
arc in a normal atmospheric environment by stretching it until the cooling of the gas at the
current zero cross can rise fast enough to prevent restrikes. The longer arc has a larger cooling
surface, thus cooling increases and deionization of the gap between the contacts improves. The
increased length of the arc also increases its resistance and, therefore, decreases the current
flow and the amount of heat per unit length of the arc that is created. To increase the length of
the arc, the arc is stretched by forcing it into an arc chute with either natural convection of the
hot gas, by blowing air through the arc, or by using a magnetic blowout coil. The magnetic
blowout coil creates magnetic force on the arc that pushes or pulls the arc into the arc chute.
There are two types of arc chutes in air break circuit breakers (Fig. 3). One type is made of
insulating material, and its function is to stretch the arc. The other type is made of metal. The
metal barriers divide the arc into a series of many smaller arcs. The voltages across these
smaller arcs are much lower than the total voltage across the breaker contacts and, therefore, it
is easier to extinguish the arcs.

Barriers of insulating material

Conducting metal barriers

Arc
Arc

Arcing contacts

Arcing contacts

Figure 3: Air Break Circuit Breaker Designs

Air Blast Breakers

A blast of compressed air can be blown into the arc perpendicular to the arc (cross blast
Fig. 4A) or along its axis (axial blast Fig. 4B). The cross blast use is limited, because it is difficult
to develop the air velocity needed for high voltage breakers across the entire interrupting area.
In more recent breakers the axial blast is more common. Air blast breakers have a tendency to
chop currents when interrupting low level currents and have subsequent restrikes, adding to
contact wear. Unclean or insufficiently dry air will slow arc clearing, increasing contact wear.
Interrupting Fin

Arc

Motion
Stationary Contact

Compressed Air
Figure 4A: Cross Air Blast

Moving Contact

Interrupter
Chamber
Interrupte
Walls r

Moving
Contact

Stationary
Contact

Compressed Air Blast

Figure 4B: Axial Air Blast

Vacuum Breakers
Vacuum circuit breaker interrupters (Fig. 5) are used mostly for medium voltage applications.
Vacuum interrupting heads have been developed for up to 38 kV per break. For higher voltages,
the interrupters are connected in series, though vacuum breakers are not commonly seen at
high voltage.
When contacts separate, a portion of the metallic contact material is vaporized, so the vacuum
is partially lost, and this vaporized metal is the conducting media for the arc. At the current zero
cross, these vapors rapidly condense on the surrounding surfaces (particularly the contact
surfaces) and the vacuum is restored as well as the high dielectric properties of a vacuum.
There is a detailed science behind the contact surface material and the shape of the contact.
The shape of the contact and the distribution of contact surface materials effects how the
current in the arc is distributed as the contacts separate and as the arc progresses. For even
wear, one design factor is to attempt to keep the arc diffused over the entire surface area,
fighting the tendency of arcs to concentrate at a single point. The earlier copper- bismuth
contact material had some tendency to condense too quickly (especially at low arcing levels)
and, therefore, break down the conducting arc before a normal current zero, which creates a
high di/dt, hence, high transient voltages. Present designs reportedly use a surface material
made of sintered copper and chromium powders and do not have current chop issues.

Figure 5: Vacuum Breaker

SF6 Breakers
In the puffer design (Fig. 6), the arc is extinguished by blowing compressed SF6 gas through
the arc via a nozzle near the arcing contacts. The mechanical energy required to compress the
gas increases very rapidly with the breaking capacity, and the size of poles and control
mechanism tends to be larger, so that 25 kA is a typical upper end design for this technology.
While higher interrupting ratings can be achieved with this design, the thermal assist design is a
more cost effective means to break high currents.
The thermal assist design (Fig. 6) uses a chamber in which SF6 gas is heated by the energy of
the arc, which causes the SF6 gas pressure in the chamber to increase. This gas, when
released, extinguishes the arc. This allows the operating mechanism to be smaller because the
extinguishing energy comes mainly from the arc. However, for low current interruption a puffer
chamber is needed to assist arc interruption due to the low energy that is available in the
breaker arc, hence, the low pressures and gas flow rates that are achieved.
A small portion of SF6 will disassociate in the high heat of the interrupting arc, which, in the
presence of any moisture and vaporized metal, can create some corrosive chemicals.
Puffer
Assist
Stationary
Conductor

Stationary
Conductor
Thermal
Assist
Contact

Blown SF6

Main
Contact

Moving
Conductor

Insulator
Moving
Conductor

Contact Movement

Thermal Pressure
Build Area

Puffer Design

Thermal Assist Design

Figure 6: Two SF6 Designs

Concept of Contact Wear: Energy in Breaker Arc

A numeric protective relay can supply a value of interrupted current to use in energy
calculations. The energy that has been interrupted can be estimated from the interrupted current
recorded, but it is not easily done. Neither the voltage across the circuit breaker (Varc-bkr) nor the
resistance of the arc (Rarc-bkr) is easily determined. Also, current and voltage are changing more
rapidly than the relay is designed to monitor. Values of Rarc-bkr vary over time through the
interruption process in a nonlinear fashion, discussed further below.

Energy arc =

( Iarc Varc bkr ) dt

Varc = Iarc Rarc bkr

Energy arc =

(I

2
arc

Eqs. 1-3

Rarc bkr dt

A common approach used by numeric relays is to use a value for Ifundamental determined just prior
to current interruption, and to assume contact wear is proportionate to I2. Other exponents are
sometimes used, ranging from 1 to 2, and some use a different exponent value for different
current levels. One reason for confusion about which exponent to use is that Rarc-bkr is nonlinear
and nonconstant during the interruption process, while the relay measures average, rms, or
fundamental frequency current levels.

Changing RARC During Fault

The diagram of Figure 7 is intended to indicate more clearly the difficulty in determining how
much energy is dissipated inside the circuit breaker during the fault. The system impedance and
the impedance of interrupted load stays relatively constant and tends to be the ruling
component that determines fault current and load current. However, the fault arc impedance
and the internal circuit breaker arc impedance may vary during the fault and tend to rise at each
current zero crossing, and then goes very high at the fault clearing point. Hence, the voltage
across the circuit breaker arc varies as the fault progresses and as the current approaches and
departs from the zero crossing point. (See Reference 1)
The humps in arc resistance shown in Figure 7 are intended to show conceptually the tendency
of the arc to break down at each zero cross. A circuit is broken when the resistance rises fast
enough and high enough to prevent the arc from continuing. In a vacuum or SF6 breaker, the
rate of rise to dielectric strength at each current zero cross is tremendous, likely leading to
blocking current flow at the first current zero cross after the contacts part. The gradual rise in the
circuit breaker arc resistance indicates the increasing arc length as the contacts part such as in
older interruption technology. This also is good for conceptually illustrating what is occurring
when current is interrupted in any circuit breaker. The final steep rise in the arc breakdown at
the final current zero cross represents the rapid rise of dielectric strength as the current flow is
blocked. The gradual decrease in fault resistance represents an evolving arc with an increasing
amount of conducting ionized gases.

Z
Zload

Zsys

R-fault

R-arc,
b re a k e r

Time

Bkr Trip
Starts

Fault Starts

Contacts
Part

Current Zero
Crosses

Fault Duration

Fault Cleared

Figure 7: Changing RARC During Fault

Arcing Time vs. Circuit Breaker Mechanism Travel and 52a/b Contact
The relay may or may not know when a breaker trip command is issued to the breaker.
However, for accurate measurement of breaker arc time, the relay needs to know some
information that correlates breaker contact parting to some other input to the relay, most
commonly a 52a or 52b breaker position indicator. The value Tc in Fig. 8 provides this
information. The difficulty is that such data requires a breaker travel analysis that compares
main contact travel information as compared to the 52a/b travel. Such data may not be readily
available. However, even if a travel analysis were done, there may be some area where grading
resistors may be in place that will affect the arc pattern and the contact wear and is yet another
confusing point on the energy dissipation in the breaker during the opening process. Relays will
also know when current reaches a 0 level. Once the point where the main contacts part is
known and the current zero is known, the relay can give an estimate of actual I2Rt (although R
must be an estimate, so only I2t is all that actually can be collected).

10

Figure 8: 52a/b vs. Breaker Arcing Time

The point at which the 52a and 52b contacts change state relative to the status of the main
contacts varies among circuit breakers. The only common thing is that the 52a and 52b auxiliary
contacts will not be closed simultaneously and that the 52a and 52b, because they do not
change state simultaneously, are not true inverses of one another (Fig. 9). There are also other
types of 52 contacts, variously named 52aa, 52a', 52a'', 52bb, 52b', etc. used in anti-pump
circuits and sometimes available for external applications, that have slightly different timing.
Arc Extinguished

Open

52aa

52b

52a

Contact Arcing

52bb

Main Contacts
Part
Closed

Figure 9: 52a/b vs. Breaker Travel

Circuit Breaker Mechanism Speed Variations

Circuit breaker mechanism operate time is fairly consistent, but there are no guarantees that
every mechanical operation will occur at the same speed for each fault cleared over the same
time period. Some items that can affect circuit breaker mechanics or clearing time are control
system operating voltage, time since last operation, temperature, age, mechanical wear,
pneumatic pressure, where in the sine wave the trip command is issued (which affects the time

11

to the next current zero cross), and in some circuit breaker designs (particularly air magnetic)
the current in the circuit breaker (low currents are interrupted more slowly then high currents).

Power Factor Effects on Contact Wear

The power factor of the load also affects contact wear. At the moment the arc is broken and the
current is driven to 0, the system voltage in the circuit must appear entirely across the contacts.
In low power factor loads (capacitor banks or inductors) or faults, when a current zero occurs,
there is high voltage across the load or system power delivery equipment. To successfully break
the arc, this load or equipment voltage must be transferred from the load or equipment to the
interrupter very rapidly, hence a high dR/dt is required of the interrupter relative to what would
be required to break high power factor loads.
Note that the energy developed in the interrupter is P = V * I. At the moment that current is
being interrupted (Fig. 10), V is rising higher in low power factor loads relative to the same
current in high power factor loads, so there is more energy dissipation in the breaker when
interrupting low power factor loads compared to high power factor loads. High power factor
loads cause much lower interrupter voltage at the point current is interrupted, so less energy is
dissipated in the breaker at the moment of interruption. Also, at high power factor, the breaker is
less prone to restrikes, and current is possibly cleared earlier in the interrupter opening process.
V-system
I-capacitive

I-inductive
Low V-breaker
at current zero

I-high PF load

High V-breaker
required at current
zero for low PF
conditions

Figure 10: Power Factor vs. Breaker Interrupting Voltage

DC Offset Effects on Breaker Wear

DC Offset increases the current that a breaker carries relative to the symmetrical waveform.
Relays that report fundamental current and filter out DC (most relays) give a low report of
interrupted current. The first half cycle of a highly offset wave (Fig. 11) will have an RMS
equivalent that is about 1.6 times the RMS value of a non-offset wave. By the time a breaker
begins its opening process, much of the DC offset has fallen off or decayed, but this may not
always be the case.

12

A couple of slightly related issues that may be good to be aware of are: a) Most short circuit
software will give the symmetrical fault level that will be seen in a fault, and some give the RMS
symmetrical equivalent of the asymmetrical waveform. Be aware of what the fault study is
reporting, and b) Beginning in 1964, ANSI Standards began the change to rating breakers on a
symmetrical current basis [2]. This allowed the end user to avoid the confusing aspects of trying
to determine the DC offsets equivalent symmetrical current value at the breakers location or to
apply multiplying factors when the breaker interrupting time was faster than a specified
interrupting time. Breakers made before the ANSI Total Current Breaker Rating Standards of
1953 did not consider the effects of asymmetrical current level.
Current
~2.7

~1.6
Time

Figure 11: DC Offset Example

Effects of Multiple Series Breaks

Extra High Voltage breakers frequently have multiple breaks, with a grading resistor or capacitor
that is held in place until late in the opening process (Fig. 12). Typical relay monitoring of
interrupted current can only give one measure of overall interrupter wear. If one contact is worn
more than others due to lack of contact opening coordination, breaker wear measurements
based on interrupted current will miss the problem. Measuring for uneven wear on multiple
break breakers requires high voltage breaker analysis at routine maintenance testing.
Grading R or C

Figure 12: Multiple Series Breaks

Fast Reclosing Effects on Breaker Wear

Since modern circuit breakers have much better heat dissipation, reclosing may not seriously
affect contact wear. However, if following IEEE Standard C37.04 guidelines, some derating of

13

the breaker may be called for. A fault heats up interrupter contacts, so if subsequent rapid
operations occur with heated contacts, they are more likely to be damaged. The standard CO
duty cycle is, as defined by ANSI/IEEE C37.04-1999, two operations with a time interval of 15 s
between operations (CO + 15 s + CO). This means that the circuit breaker should only operate
twice at its rated symmetrical interrupting capability current with 15 s of dead time between
operations. If reclosing is set to occur faster, then the standards provide a method of calculating
derating factors to reduce the rated interrupting current. The effects of unscheduled faster
reclosing are not compensated for in the programming of numeric relay duty monitoring.

Close Operation Effects on Breaker Wear

C37 Standards for breaker ratings are based upon the number of Close/Open operations. It is
typically assumed that the open operation causes most contact wear, but closing operations
also create wear. Protective relays typically are only monitoring the trip operations. If a breaker
is always closed into a high load, it will have more wear than the breaker that is always closed
into no load. Ignoring the effects of close operations in the relay will affect the accuracy of
breaker contact wear measurements. However, since the contact wear calculation in relays is a
highly approximated wear measurement, the error in not accounting for the close operation may
be considered a negligible increase in the overall inaccuracy issue.

No-Load Operation Considerations

The typical circuit breaker is very highly rated mechanically, and the number of no-load
operations is very high. The IEEE Standard C37.06 Operations Endurance Tables vary from
500 to 10,000 no-load operations depending on interrupting current, voltage, and full-load
current ratings. Circuit breakers also have a "tighten, adjust, and inspect" requirement after 2502000 operations. Due to the high number of no-load operations permitted, it is justifiable to
ignore no-load operations in the relay duty information; however, the circuit breaker operations
counter should be monitored to see when the number-of-operations-based maintenance is
required.

Relay Testing Effects on Relay Wear Records

Relay testing will make a change in the accumulated circuit breaker duty that the relay reports,
so accumulation of duty current needs to be defeated when applying current to the relay during
testing. Numeric relays sometimes have a "Block Circuit Breaker Duty Accumulation" logic input
feature that allows a relay to stop circuit breaker duty summation when relay testing is
proceeding. If the relay lacks this feature, one must note circuit breaker duty data before starting
tests and must reset data at the completion of the testing.

Circuit Breaker Duty Monitoring Capabilities of Numeric

Relays
Numeric relays typically only record an rms or fundamental frequency value of the current at the
approximate time when the circuit breaker interrupter opens. Thus, the relay can determine:
Monitor the number of operations;
Representative value of the total interrupted current;
Monitor breaker opening time
14

The variations of total interrupted current found in relays on the market can be I1, I2, IN, and
other variations that vary the exponent N for current-level variations.
The next issue is to determine breaker duty settings for the relay. What ratings for the circuit
breaker should be used? In order for a company to standardize on a program of planned circuit
breaker duty monitoring using relay data, some simplified approach to interpreting circuit
breaker duty ratings needs to be developed based upon the circuit breaker rating data that is
commonly available, the limited ability of a relay to measure breaker wear, and the limited value
of carrying the analysis to a very high degree of accuracy. Maximum circuit breaker short-circuit
ratings and continuous current ratings are available from the manufacturer, but the number of
allowed interruptions at some intermediate current level may not be available. An alternate
position is to use standard values from IEEEC37.04 and ANSI C37.06 circuit breaker standards
in lieu of manufacturers data and then compare this to the current interrupted available from the
relay.

Circuit Breaker Duty Rating Approximation

For a given circuit breaker, if one assumes a relatively constant arcing time t and a relatively
constant Rarc in the circuit breaker, then one can assume the energy absorbed by the circuit
breaker contacts will be proportional to the current interrupted. If two points of "number of
operations at a given current level" can be obtained, then we can find an approximate means to
determine when a circuit breaker has absorbed an approximation of maximum wear.
Assume that:
Average arc time tarc is relatively the same from one operation to the next.
Averaged over the entire opening process, circuit breaker arc resistance Rarc-bkr is
relatively the same from one circuit breaker operation to the next.
The result is energy dissipated in the circuit breaker proportionate to a summation of
interrupted current, so we can assume that circuit breaker wear can be estimated from:
Bkr Wear Energyarc bkr

k IN

Eq. 4

The earliest relays and breaker wear concepts were based on energy being proportionate to I2
but, in some cases, the concept of monitoring only I1 was available; hence, N=2 or N=1 was
used. Therefore, the relay would set k=1 and sum I2 or I until a user-assigned maximum was
reached. The problem is in determining the appropriate maximum. If one can determine a value
for "number of operations at a given current level" data for the circuit breaker, then one can
determine when a circuit breaker is approaching wear limits and needs internal maintenance. A
position that will be taken in the balance of this paper to find these wear ratings is to use ratings
from IEEE Standards C37.04-1999 and C37.06-1997 as shown in Table 1.

15

Rated Continuous Current

2,000A

Voltage Rating

15kV

Isc,max Max Symmetrical Interruption Rating @ max KV

31.5kA

Operations between servicing (lubricating, tightening)

2,000

Rated No-Load Operations

10,000

Full Load C/O at >0.8 pf (Note: This rating will not apply to cap
switching, motor starts, highly inductive or capacitive loads, low level
faults.)

1,000

Sum of I interrupted that the circuit breaker should be able to

withstand as long as each fault is less than 85% of Isc,max.
Note 1, 2, 3, 4

800% of Isc,max

Table 1: Sample Breaker Ratings

Note 1: Note the last line in Table 1 is a rating based on I1, not I2. Note 800% @ 85% rule from C37.04,1999, section
5.8.2.5 effectively means the circuit breaker shall be able to withstand a calculated value of 9.41 faults at 85% of
2
Isc,max (800% *31.5 kA/85% * 31.5 kA=9.41 operations). This will be important to note when changing to I
interruption summations.
Note 2: It should also be noted that in older copies of C37.04, a 400% factor was specified so a 4.7 factor would be
used for 85% faults in circuit breakers rated per the older standard.
Note 3: For simplicity, the asymmetrical adjustment factor for high X/R systems called for in section C37.04 section
5.8.2.2 is not used in the following examples.
Note 4: For simplicity, using a K range factor of 1. Higher K range factors are described in older versions of C37.04
and C37.06.

Table 1 effectively provides two measuring points available for wear calculations:
a) Number of operations at rated current =1000;
b) Isc,max rating = 31.5 kA, which implies 26,775 A (85%) can be interrupted 9.41 times.
See Table 1, Note 1, for an explanation of the 9.41 factor.
Now calculate the sum of I1 and I2 with these two wear points:
1,000 ops at IFL, PF>0.8

800% of Isc,max, at 85% fault

levels:

Sum I Interrupted

1,000 * 2000A = 2,000kA

9.41 * (0.85 * 31.5kA) = 252kA

Sum I2 Interrupted

1,000 * (2000)2 = 4,000MA

9.41 * (0.85 * 31.5kA)2 =

6,747MA
MA = Million Amps; kA = Thousand Amps

Table 2: Sample I and I2 Interrupted Breaker Ratings

The problem immediately seen is that we have four different wear values: two that might be
appropriate if I1 is accumulated, and two that might be appropriate if I2 is accumulated. There is
no clear answer on how to approach the issue. However, if the relay would allow N to vary from
1 and 2, a resolution to the issue can be found by adjusting the exponent N. If we apply our two
wear rating points to equation 4, we can derive a value for N that gives consistent ratings. This
gives us the following equality:

16

k 1000 2000N = k 9.41 ( 31500 0.85 )

Eq. 5

We can solve this for N. For this example:

1, 0 0 0
9 .4 1
3 1, 5 0 0 0 . 8 5
2, 0 0 0

lo g
N =
lo g

= 1 .7 9 8 5 4

Eq. 6

Thus, the exponent 1.8 may be more appropriate for wear calculations. The effect will be seen
further in the paper, especially in Table 4 discussions.

Breaker Duty Monitoring in Basler Electric Relays

The balance of the paper will explain the different ways Basler Electric implements circuit
breaker duty monitoring in its relays, identified as Circuit Breaker Failure Relays BFR1, BFR2
and BFR3. This may or may not apply to how circuit breaker duty monitoring is calculated in
other manufacturers relays but is used for illustration of some of the different methods of
estimating breaker duty.

Basler BE1-BPR Relay

In the BE1-BPR circuit breaker failure relay, circuit breaker duty measurement includes the time
of the arc. The relay only uses the I2 exponent; it does not allow other exponents to be used in
the equation. It uses a one-cycle rms equivalent of the current at the approximate time of
interruption.
The BE1-BPR relay sums I2t where:
t is taken from monitoring the time between when a 52b changes state, +/- a user offset
time ("Tc"), until the relay senses current has been interrupted. See Fig. 8.
I is a quasi-rms (low sampling rate) value of one cycle of current at the time the circuit
breaker contacts began parting (Fig. 8).
To set Tc, one must compare a 52a or 52b to main contact travel, using a circuit breaker travel
analyzer. An alarm is set when I2t sums to greater than a user-defined setting.
The BPR's max I2t setting [DMAX (short for Duty-Max)] is defined by:

DMAX = ( Max # Ops @ ISC Max ) ISC Max 2 ( Typical Bkr Arc Time )

Eq. 7

Typical arc time is hard to define; it is the time that the circuit breaker internal arc is ongoing,
from the time of initial contact parting to fault interruption. It should not be confused with
published opening time. Not many circuit breaker owners know the internal arc time of their
circuit breaker, so this is a notable source of error and makes the calculation of the DMAX
setting only a guiding approximation. Assuming the circuit reaker ratings from the earlier
example, and assuming a typical arc time of 3/4 cycle (0.0125s):

D M A X = ( 9.41 ) ( 0.85 31, 500 ) ( 0.0125s ) 84.3 E 6

2

This is the calculated maximum duty setting.

17

Basler BE1-851, -951, -IPS, -GPS, and -CDS220 Numerical Relays

Compared to the BE1-BPR, the BE1-851, -951, IPS, -GPS, and -CDS relays discard the
inclusion of arcing time in the equation but add the ability to monitor I or I2 where I is the value
of current the relay measured at the time just preceding the circuit breaker opening. While the
relay sums both I and I2, only one of these values can be selected for alarming.
Similar to the BE1-BPR, the Max Circuit Breaker Duty is set in terms of DMAX. If the relay is set
to monitor the summation of I1, DMAX is set by the equation:

DMAX = Iinterrupt # of operations

Eq. 8

If the relay is set to monitor I2, then the relay internally uses (DMAX)2. This means you need to
calculate and enter a DMAX setting using the equation:

DMAX = I2interrupt # of operations

0.5

Eq. 9

Using our earlier example and Equations 8 and 9, there are four possible settings for DMAX,
each valid and suitable depending on the assumptions one makes and the application of the
circuit breaker (see Table III).
1,000 ops IFL, PF>0.8

800% of Isc,max, at 85% fault

levels:

Sum I Interrupted

1,000 * 2000A = 2,000kA

(9.41 * (0.85 * 31.5kA) =252kA

Sum I2 Interrupted

(1,000 * (2000)2)0.5 = 63.2kA

(9.41 * (0.85 * 31.5kA)2)0.5 =

82.1kA

Table 3: Possible DMAX Settings (bold numbers) for BE1-851, 951, IPS, GPS, CDS220

The variety of values that could be used for maximum I or I2 means that Basler basically leaves
settings recommendations to the good judgment of the end user.

BE1-1051 and BE1-CDS240

These two relays allow the user to set the duty exponent N in Equation 4 between 1.0-3.0. By
allowing the user to set the exponent N, the relay offers a better ability for the relay duty
measurement to track the two wear data points previously discussed.
These relays sum interrupted IN, where I is the current value that the relay measured at the time
just preceding the circuit breaker opening. Recall earlier calculations using Equation 5 to solve
for N, where we determined that N=1.8 would be appropriate.
Similar to the condition for BE1-851, -951, et al when I2 is used, DMAX needs to be calculated
from an equation that backs out the N from the entered value for DMAX. The equation is:

DMAX = Iinterrupt N # of operations

18

1/N

The need to include the Nth root function when using the IN version of circuit breaker duty is
caused by Baslers requirement for DMAX to be entered in terms of I, not IN and because Basler
internally uses the Nth power of DMAX for whatever data is entered for DMAX.
Allowing the user to set N gives a circuit breaker duty measurement and DMAX that is
consistent across the two duty points (see Table IV):

Sum IN Interrupted

1,000 ops IFL, PF>0.8

800% of Isc,maxx, at 85% fault

levels:

(1,000 * (2000)1.8)1/1.8 =
93.1kA

(9.41 * (0.85 * 31.5kA)1.8)1/1.8 =

93,1kA

Table 4: DMAX Settings(bold numbers) for BE1-1051, -CDS240

Hence, set the relay to N=1.80 and DMAX=93,100.

Other Breaker Monitoring Functions in Numeric Relays

There are other useful features besides circuit breaker wear measuring functions available in
numeric relays such as the following.

Circuit Breaker Clearing Time Alarms

A circuit breaker may become just slow enough to create an alarm but still clear a fault. When
this occurs, it is an indication that the circuit breaker mechanism needs lubrication or adjustment
[3]. The slow breaker clearing time alarm is primarily looking for a clearing time greater than the
users input value, which is based on the manufacturers stated opening time of the circuit
breaker, plus an experience-based margin, indicating a need for maintenance. Another option is
to use historical or seasonal operation records to determine typical time to interrupt current. A
third approach is to use an anticipated maximum clearing time plus some margin, such as two
cycles less than circuit breaker failure relay operate time.

Number of Operations Counter

The operations counter in the relay serves as a means to create an operations-based
maintenance interval. It serves the same purpose as a station log that tracks circuit breaker
operations but is a value that can more easily be automatically or remotely monitored. The
function does not provide any weighting for interrupted current, so it is most useful in indicating
a mechanical operations maintenance point has been reached.

Trip Coil Continuity Monitoring

Simple logic is required on the part of the relay to monitor trip coil (TC) continuity, though it does
use up some Input/Output capability of the relay. In some unusual circumstances the Trip Circuit
Monitor (TCM) has introduced stray unwanted signals in the trip circuit.
In the circuit of Fig. 13, when the circuit breaker is closed, the 52a in the figure will be closed, so
voltage across the trip contact should be high, approximately equal to dc power supply. The

19

"High R" device in the relay allows a trickle current around the trip contact. Normally, this is not
enough current to build any substantive voltage across the TC, and the voltage across the trip
contact remains nearly equal to the dc supply voltage. However, if either the TC or 52a in series
with it fails in an open-circuit mode, the voltage across the trip contact falls to 0. The relay
monitors trip contact voltage via device "V" in the diagram, and it alarms if the voltage is low
when the circuit breaker is closed.
The TC monitor system must be turned off when the circuit breaker opens, since a 52a in the
circuit breaker opens up the trip path. When the circuit breaker opens, the TC is open circuited
by the 52a contact. For this purpose, the relay needs to know the status of the circuit breaker,
which is supplied from another auxiliary contact not shown in Fig. 13. It should be noted that this
would be used in lieu of the common red light in the trip circuit used to show trip circuit
continuity.
In occasional cases, the leakage current can be a problem. If the TC itself was actually a high
impedance monitoring device that provides trip information rather than being the actual trip coil.
The trip coil monitor will see an open circuit and alarm, or the leakage current in the resistor
may be enough to tell the high impedance monitor that the trip contact has closed.
+DC

High
R

Trip
Contact

RELAY
52a

TC

-DC
Figure 13: Trip Coil Continuity Monitor Circuit

Relay Recording Capabilities

Some recording features available in most numeric relays are: oscillography records, fault
summary records, sequence of events and phasor representation, for each abnormal condition
or fault event. These report making functions and the circuit breaker failure function are
expected as a matter of course in modern numeric relays.

20

Conclusion
Numeric relays are useful for monitoring power circuit breakers to obtain predictive maintenance
information. However, there is not a universal approach among relay or circuit breaker
manufacturers for using interrupted current to measure breaker wear. The instruction manual for
each individual relay must be studied to determine the methodology used in applying and
setting that relay, since relay manufacturers may not be consistent across their product lines
about how duty factor is calculated and set in their relays. When applied properly, the
information obtained from the circuit breaker duty monitoring feature of numeric relays is a tool
that is waiting to be used for condition-based maintenance scheduling of power circuit breakers.

References
1) D. L. Hickery, E. J. Bartlett, and P. J. Moore. January 2003. Investigation into Physical
and Electrical Process of Power System Fault Arcs. University of Bath, Bath,
U.K.;[Online] Available: http://staff.bath.ac.uk/eesip/other_publications/FaultArc.pdf
2) W. R. Speed, Circuit breaker operator signature analysis, TXU Electric, Ft. Worth, TX,
May 2002.

21

If you have any questions or need

additional information, please contact
Basler Electric Company.
Our web site is located at:
http://www.basler.com
e-mail: info@basler.com

Route 143, Box 269, Highland, Illinois U.S.A. 62249

Tel +1 618.654.2341 Fax +1 618.654.2351

e-mail: info@basler.com
P.A.E. Les Pins, 67319 Wasselonne Cedex FRANCE
Tel +33 3.88.87.1010 Fax +33 3.88.87.0808

e-mail: beifrance@basler.com

1300 North Zhongshan Road,

Wujiang Economic Development Zone
Suzhou, Jiangsu Province, PRC 215200
Tel +86(0)512 6346 1730
Fax +86(0)512 6346 1760
e-mail: beichina@basler.com