E‐051 Overcurrent Protection Fundamentals 
 
Instructor: Velimir Lackovic 
Course ID: E‐051 
PDH Hours: 5 PDH 

PDH Star | T / F: (833) PDH‐STAR (734‐7827) | E: [email protected] 
 
   E‐051 Overcurrent Protection Fundamentals 
 

OVERCURRENT PROTECTION FUNDAMENTALS

Relay protection against high current was the earliest relay protection mechanism to
develop. From this basic method, the graded overcurrent relay protection system, a
discriminative short circuit protection, has been formulated. This should not be mixed
with ‘overload’ relay protection, which typically utilizes relays that function in a time
related in some degree to the thermal capacity of the equipment to be protected. On the
contrary, overcurrent relay protection is completely directed to the clearance of short
circuits, even though with the settings typically assumed some measure of overload
relay protection may be obtained.

CO-ORDINATION TECHNIQUE
Precise overcurrent relay usage asks for the knowledge of the short circuit current that
can flow in each section of the power network. Since large-scale measurements and
tests are typically unfeasible, system calculations have to be used. The information
needed for a relay protection setting analysis is:

- Single-line diagram of the electrical power system, presenting the type and
rating of the relay protection elements and their related current transformers

- Impedances in ohms, per cent or per unit, of all power transformers, rotating
machine and transmission lines

- Maximum and minimum figures of short circuit currents that are anticipated to
go through each protection element

- Maximum load current through protection elements

- Starting current requirements of electrical motors and the starting and locked
rotor/stalling times of induction motors

- Transformer inrush, thermal withstand and damage curves

1 
 
   E‐051 Overcurrent Protection Fundamentals 
 
- Decrement curves presenting the decay rate of the short circuit current
supplied by the generators

- Performance curves of the current transformers

The protection relay adjustments are first calculated to provide the shortest tripping
times at maximum fault currents and then verified to understand if tripping will also be
acceptable at the minimum short circuit current anticipated. It is typically suggested to
print the curves of protection relays and other protection elements, such as fuses, that
are to trip in series, on a common graph and scale. It is typically more convenient to
utilize a scale referring to the current anticipated at the lowest voltage base, or to utilize
the dominant voltage base. The options are a mutual MVA base or a different current
scale for each system voltage. The fundamental rules for proper protection relay co-
ordination can typically be presented as follows:

- Whenever feasible, utilize protection relays with the same tripping

characteristic in series with each other

- Ensure that the protection relay farthest from the source has current settings
same to or less than the protection relays behind it, that is, that the primary current
needed to trip the protection relay in front is always same to or less than the primary
current needed to trip the protection relay behind it.

RULES OF TIME/CURRENT GRADING

Among the different feasible methods utilized to accomplish precise protection relay co-
ordination are those utilizing either time or overcurrent, or a mix of both. The common
objective of all three methodologies is to provide precise discrimination. That is to say,
each one has to isolate only the faulty part of the electrical power system network,
leaving the rest of the power system untouched.

RELAY PROTECTION DISCRIMINATION BY TIME

In this system, an adequate time setting is provided to each of the protection relays
controlling the power circuit breakers in an electrical power system to make sure that

2 
 
   E‐051 Overcurrent Protection Fundamentals 
 
the circuit breaker nearest to the fault location opens first. A fundamental radial
distribution electrical system is presented in Figure 1, to demonstrate the operational
logic.

E D C B A

F 
Figure 1. Radial electrical system with time discrimination

Overcurrent relay protection is given at B, C, D and E, that is, at the infeed position of
each part of the electrical power system. Each relay protection device comprises a
definite-time delay overcurrent protection relay in which the trip of the current
sensitive element starts the time delay device. Given the setting of the current device is
below the short circuit current value, this device plays no role in the accomplishment of
discrimination. For this reason, the protection relay is sometimes known as an
‘independent definite-time delay protection relay’, since its tripping time is for practical
uses independent of the overcurrent level.

It is the time delay device, hence, which gives the means of discrimination. The
protection relay at location B is set at the shortest possible time delay to permit the fuse
to operate for a fault at location A on the secondary side of the power transformer. After
the time delay has completed, the protection relay output contact closes to operate the
power circuit breaker. The protection relay at location C has a time delay setting equal
to t1 seconds, and likewise for the protection relays at locations D and E. If a short
circuit happens at location F, the protection relay at location B will trip in t seconds and
the later tripping of the power circuit breaker at location B will clear the short circuit
before the protection relays at locations C, D and E have time to trip. The time interval t1
between each protection relay time setting must be sufficiently long to make sure that
the upstream protection relays do not trip before the power circuit breaker at the short
circuit location has operated and cleared the short circuit.

The main drawback of this discrimination procedure is that the longest short circuit
clearance time happens for short circuits in the section nearest to the power source,

3 
 
   E‐051 Overcurrent Protection Fundamentals 
 
where the short circuit level (MVA) is the greatest.

RELAY PROTECTION DISCRIMINATION BY CURRENT

Relay protection discrimination by current is based on the fact that the short circuit
current changes with the location of the fault because of the difference in impedance
figures between the source and the short circuit. Therefore, usually, the protection
relays controlling the different power circuit breakers are programmed to trip at
appropriately tapered values of current such that only the protection relay closest to the
fault operates its breaker. Figure 2 presents the method. For a fault at location F1, the
electrical system fault current is expressed as:

6350
𝐼 𝐴
𝑍 𝑍

where:

ZS - source impedance = 0.485 Ω

ZL1 = cable impedance between C and B = 0.24 Ω

Therefore 𝐼 8800 𝐴
.

Therefore, a protection relay controlling the power circuit breaker at location C and
programmed to trip at a short circuit current of 8800A would in theory save the whole
of the underground cable section between locations C and B. Nevertheless, there are
two critical practical points that impact this co-ordination procedure:

It is not efficient to differentiate between a fault at location F1 and a fault at location F2,
since the separation between these locations may be only a few metres, corresponding
to a variation in short circuit current of roughly 0.1% in practice, there would be
variations in the source short circuit level, usually from 250MVA to 130MVA. At this
lower short circuit level the short circuit current would not surpass 6800A, even for an
underground cable short circuit near to location C. A protection relay set at 8800A
would not save any part of the underground cable section concerned.

4 
 
   E‐051 Overcurrent Protection Fundamentals 
 
Relay protection discrimination by current is hence not a practical suggestion for
correct grading between the power circuit breakers at locations C and B. Nevertheless,
the issue changes appreciably when there is major impedance between the two circuit
breakers concerned. Regard the grading needed between the power circuit breakers at
locations C and A in Figure 2. Presuming a short circuit at location F4, the short-circuit
current is presented as:

6350
𝐼
𝑍 𝑍 𝑍 𝑍

Where

ZS – source impedance = 0.485 Ω

ZL1 – cable impedance between locations C and B =0.24 Ω

ZL2 – cable impedance between location B and 4 MVA transformer =0.04 Ω

Zr – transformer impedance =0.07 2.12 Ω

Therefore, I 2200 A
.

11 kV 4 MVA
200 m 200 m
250 MVA 11/3.3 kV
240mm2 240mm2
C B C
cable cable 7%

F1 F2 F3 F4

Figure 2. Radial electrical system with current discrimination

Due to this, a protection relay controlling the power circuit breaker at location B and
programmed to trip at a current of 2200A plus a safety margin would not trip for a short
circuit at F4 and would therefore discriminate with the protection relay at location A.
Presuming a safety margin of 20% to allow for protection relay errors and a further
10% for changes in the system impedance quantities, it is fair to select a protection

5 
 
   E‐051 Overcurrent Protection Fundamentals 
 
relay setting of 1.3x2200A, that is, 2860A, for the protection relay at location B. Now,
analysing a short circuit at location F3, at the end of the 11kV underground cable
supplying the 4MVA transformer, the short-circuit current is presented as:

6350
𝐼
𝑍 𝑍 𝑍

This, presuming a 250 MVA source short circuit current level:

6350
𝐼 8300 𝐴
0.485 0.24 0.04

Instead, presuming a source short circuit current level of 130 MVA:

6350
𝐼 5250 𝐴
0.93 0.214 0.04

For either value of source level, the protection relay at location B would precisely
function for short circuits anywhere on the 11kV underground cable supplying the
transformer.

RELAY POTECTION DISCRIMINATION BY BOTH TIME AND CURRENT

Each of the two presented methodologies so far has a fundamental drawback. In the
case of discrimination only by time, the drawback is due to the fact that the more
serious short circuits are cleared in the longest tripping time. On the other side,
discrimination by current can be used only where there is considerable impedance
between the two considered power circuit breakers. It is due to the limitations
introduced by the independent usage of either time or current co-ordination that the
inverse time overcurrent protection relay characteristic has developed. With this
characteristic, the tripping time is reciprocally proportional to the short circuit current
level and the real characteristic is a function of both ‘time’ and 'current' settings. Figure
3 presents the characteristics of two protection relays given different current/time
adjustments. For a great change in short circuit current between the two feeder ends,
quicker tripping times can be accomplished by the protection relays closest to the
source, where the short circuit level is the greatest. The drawbacks of grading by time or
current alone are resolved.

6 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Figure 3. Protection relay characteristics for different settings

Relay A: Current setting=1000 A; TMS=0.1; Relay B: Current setting=300 A; TMS=0.2

STANDARD IDMT OVERCURRENT PROTECTION RELAYS

The current/time tripping characteristics of IDMT protection relays may need to be
changed according to the functioning time needed and the characteristics of other relay
protection elements used in the electrical network. For these needs, IEC 60255
determines a number of standard characteristics. These are:

- Standard Inverse characteristic (SI)

- Very Inverse characteristic (VI)

- Extremely Inverse characteristic (EI)

7 
 
   E‐051 Overcurrent Protection Fundamentals 
 
- Definite Time characteristic (DT)

The mathematical definition of the curves are presented in Table 1, and the curves
based on a common setting current and time multiplier setting of 1 second are
presented in Figure 4.

The tripping characteristics for various TMS settings using the SI curve are presented in
Figure 6.

Relay protection characteristic Formula (IEC 60255)

Standard inverse (SI) 0.14
t TMS .
I 1
Very inverse (VI) 13.5
t TMS
I 1
Extremely inverse (EI) 80
t TMS
I 1
Long time standby earth fault 120
t TMS
I 1

Table 1. Definitions of standard relay protection characteristics

Relay protection characteristic Formula

IEEE moderately inverse TD 0.0515
t 0.114
7 I . 1
IEEE very inverse TD 19.61
t 0.491
7 I 1
IEEE extremely inverse TD 28.2
t 0.1217
7 I 1
US CO8 inverse TD 5.95
t 0.18
7 I 1
US CO2 short time inverse TD 0.02394
t 0.01694
7 I . 1

Table 2. ANSI IDMT definitions of standard relay protection characteristics

𝐼
𝐼
𝐼

Where

I- Measure current

8 
 
   E‐051 Overcurrent Protection Fundamentals 
 
I - Relay setting current

TMS – Time Multiplier Setting

TD – Time Dial Setting

Figure 4. IEC 60255 IDMT protection relay Figure 5. ANSI IDMT protection relay
characteristics characteristics

9 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Figure 6. Common time/current characteristics of typical IDMT protection relay

Even though the protection curves are only presented for discrete values of TMS,
continuous settings may be feasible in an electromechanical protection relay. For other
relay protection types, the protection setting steps may be so small as to efficiently give
continuous adjustment. Also, almost all overcurrent protection relays are also equipped
with high-set instantaneous devices. In majority of situations, use of the standard SI
protection curve proves satisfactory, but if acceptable grading cannot be accomplished,
utilization of the VI or EI protection curves may assist to solve the issue. When digital or
numeric protection relays are utilized, other characteristics may be given, including the
possibility of user-definable protection curves. More information is given in the
following paragraphs. Protection relays for electrical power systems made according to
North American standards use ANSI/IEEE protection curves. Table 2 provides the
mathematical description of these protection characteristics and Figure 5 presents the
protection curves standardised to a time dial setting of 7. It is important to note that
various vendors may standardise their protection curves at various settings other than

10 
 
   E‐051 Overcurrent Protection Fundamentals 
 
TD=7. The relay protection engineer has to ensure if the factor of 7, or some other
nominal, is used.

MIXED IDMT AND HIGH SET INSTANTANEOUS OVERCURRENT PROTECTION

RELAYS
A high-set instantaneous device can be utilized where the source impedance is small in
comparison with the protected circuit impedance. This allows a decrease in the
operating time at high short circuit levels possible. It also enhances the overall electrical
system grading by allowing the 'discriminating protection curves' behind the high set
instantaneous device to be reduced. One of the benefits of the high set instantaneous
devices is to decrease the tripping time of the circuit protection. If the source impedance
stays constant, it is then feasible to accomplish high-speed relay protection over a large
part of the protected circuit. The quick short circuit clearance time helps to decrease
damage at the short circuit location. Grading with the protection relay directly behind
the protection relay that has the instantaneous devices enabled is accomplished at the
current setting of the instantaneous devices and not at the maximum short circuit level.

TRANSIENT OVERREACH
The reach of a protection relay is that portion of the protected electrical system by the
protection relay if a short circuit happens. A protection relay that trips for a short circuit
that lies beyond the intended protection zone is said to overreach. When applying
instantaneous overcurrent devices, care has to be taken in selecting the settings to stop
them tripping for short circuits beyond the protected area. The initial current due to a
D.C. offset in the current wave may be higher than the protection relay pick-up value
and cause it to trip. This may happen although the steady state R.M.S. figure of the short
circuit current for a short circuit at a location beyond the needed reach point may be
less than the protection relay setting. This process is known as transient over-reach,
and is expressed as:

% 𝑇𝑟𝑎𝑛𝑠𝑖𝑒𝑛𝑡 𝑜𝑣𝑒𝑟 𝑟𝑒𝑎𝑐ℎ 100% (1)

Where

11 
 
   E‐051 Overcurrent Protection Fundamentals 
 
I1 – RMS steady state pickup current

I2 – steady state RMS current which when completely offset just causes protection relay
pickup

When applied to power transformers, the high set instantaneous overcurrent devices
have to be set above the maximum through short circuit current than the power
transformer can supply for a short circuit across its LV terminals, to keep discrimination
with the protection relays on the transformer LV side.

VERY INVERSE (VI) OVERCURRENT PROTECTION RELAYS

Very inverse overcurrent protection relays are especially suited if there is a
considerable decrease of short circuit current as the distance from the power source
grows, i.e. there is a considerable increase in short circuit impedance. The VI tripping
characteristic is such that the tripping time is roughly doubled for decrease in current
from 7 to 4 times the protection relay current setting. This allows the use of the same
time multiplier setting for more protection relays connected in series.

Figure 8 shows the SI and VI curves for a protection relay. The VI curve is much steeper
and hence the operation increases much faster for the same decrease in current in
comparison to the SI protection curve. This allows the requisite grading margin to be
found with a lower TMS for the same setting current, and therefore the tripping time at
source can be minimised.

12 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Figure 8. Cross comparison of SI and VI protection relay characteristics

EXTREMELY INVERSE (EI) OVERCURRENT PROTECTION RELAYS

With this protection characteristic, the tripping time is roughly inversely proportional
to the square of the applied current. This makes it suited for the protection of
distribution transmission lines in which the transmission line is subjected to peak
currents on switching in, as would be the case on a power circuit supplying
refrigerators, pumps, water heaters and so on, which stay connected even after a
sustained interruption of the power. The long time tripping characteristic of the
extremely inverse protection relay at normal peak load quantities of current also makes
this protection relay especially suited for grading with fuses. Figure 9 presents common
protection curves. The EI characteristic provides a satisfactory grading margin, but the
VI or SI characteristics at the same settings do not. Another usage of this protection
relay is in conjunction with auto-reclosers in low voltage distribution lines. The
majority of short circuits are transient in nature and unneeded blowing and replacing of
the protection fuses present in final circuits of such a electrical system can be averted if

13 
 
   E‐051 Overcurrent Protection Fundamentals 
 
the auto-reclosers are made to trip before the fuse blows. If the short circuit continues,
the auto-recloser locks itself in the closed position after one opening and the fuse
operates to set apart short circuit location.

Figure 9. Cross-comparison of protection relay and fuse characteristics

OTHER RELAY PROTECTION CHARACTERISTICS

User defined protection curves may be given on some types of digital or numerical
protection relays. The general rule is that the engineer provides a series of current/time
co-ordinates that are kept in the relay memory. Interpolation between points is used to
give a smooth trip characteristic. Such a characteristic, if present, may be utilized in
special situations if none of the standard operating characteristics is suited.
Nevertheless, grading of upstream protection may become more challenging, and it is
necessary to assure that the protection curve is adequately documented, along with the

14 
 
   E‐051 Overcurrent Protection Fundamentals 
 
application reasons. Since the typical curves cover most situations with tolerable
tripping times, and most devices is made with standard protection curves in mind, the
requirement to use this protection form is rare.

Digital and numerical protection relays may also let in pre-defined logic arrangements
using digital (relay) I/O given in the protection relay to apply typical protection
schemes such as CB failure and trip circuit supervision. This saves the provision of
separate relay protection hardware to complete these functions.

INDEPENDENT (DEFINITE) TIME OVERCURRENT PROTECTION RELAYS

Overcurrent protection relays are typically also supplied with elements having
independent or definite time protection characteristics. These protection characteristics
give means of co-ordinating more protection relays in series in arrangements in which
the system short circuit current changes very widely due to variations in source
impedance, as there is no variation in time with the change of short circuit current. The
time/current protection characteristics of this protection curve are presented in Figure
10, together with those of the standard IDMT protection characteristic, to point out that
lower tripping times are accomplished by the inverse protection relay at the higher
values of short circuit current, whereas the definite time protection relay has lower
tripping times at the lower current values.

15 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Figure 10. Cross-comparison of definite time and standard IDMT protection relays.
Vertical lines point out the reduction in tripping times accomplished by the inverse
protection relay at high short circuit levels.

PROTECTION RELAY CURRENT ADJUSTMENTS

An overcurrent protection relay has a minimum tripping current, known as the current
setting of the protection relay. The current setting has to be selected so that the
protection relay does not trip for the maximum load current in the circuit being
protected, but does trip for a current same or higher to the minimum anticipated short
circuit current. Even though by using a current setting that is only just above the
maximum load current in the circuit a degree of protection against overloads as well as
short circuits may be given, the main objective of overcurrent relay protection is to set
apart primary system short circuits and not to give overload protection. In principle, the
current setting will be chosen to be above the maximum short time rated current of the
protected circuit. Since all protection relays have hysteresis in their current settings, the
protection setting must be sufficiently high to allow the protection relay to reset when
the rated current of the protected circuit is being transfered.

The amount of hysteresis in the current setting is referred by the pick-up/drop-off ratio

16 
 
   E‐051 Overcurrent Protection Fundamentals 
 
of a protection relay – the figure for a modern protection relay is usually 0.95.
Therefore, a protection relay minimum current setting of at least 1.05 times the short-
time rated current of the protected circuit is likely to be needed.

PROTECTION RELAY TIME GRADING MARGIN

The time interval that must be granted between the tripping of two nearby protection
relays to accomplish correct discrimination between them is known as the grading
margin. If a grading margin is not given, or is not enough, more than one protection
relay will function for a short circuit, leading to troubles in finding out the position of
the short circuit and unneeded loss of supply to some devices.

The grading margin is dependent on a number of different factors:

1. The fault current interrupting time of the power circuit breaker

2. Protection relay timing errors

3. The overshoot time of the protection relay

4. CT errors

5. Final margin on tripping completion

Factors 2 and 3 are dependent on the protection relay technology. For example, an
electromechanical protection relay has a greater overshoot time than a numerical
protection relay.

Grading is initially carried out for the maximum short circuit current level at the
relaying point under consideration, but verification is also made that the needed
grading margin exists for all current levels between protection relay pick-up current
and maximum anticipated short circuit current level.

17 
 
   E‐051 Overcurrent Protection Fundamentals 
 
POWER CIRCUIT BREAKER INTERRUPTING TIME
The power circuit breaker interrupting the short circuit has to totally interrupt the
current before the discriminating protection relay ceases to be energised. The time
taken depends on the circuit breaker type and the short circuit current to be cleared.
Vendors typically give the short circuit tripping time at rated interrupting capacity and
this figure is constantly utilized in the computation of grading margin.

PROTECTION RELAY TIMING ERROR

All protection relays have errors in their timing in comparison to the ideal protection
characteristic as specified in IEC 60255. For a protection relay defined to IEC 60255, a
protection relay error index is cited that fixes the maximum timing error of the
protection relay. The timing error has to be taken into consideration when finding out
the grading margin.

PROTECTION RELAY OVERSHOOT

When the protection relay is de-energised, tripping may continue for a little longer until
any kept energy has been released. To illustrate, an induction disc protection relay will
have kept kinetic energy in the motion of the disc; static protection relay circuits may
have energy kept in capacitors. Protection relay design is aimed to minimizing and
absorbing these energies, but some adjustment is typically required. The overshoot time
is determined as the difference between the tripping time of a protection relay at a
defined value of input current and the maximum duration of input current, which when
abruptly decreased below the protection relay tripping level, is insufficient to cause
protection relay function.

CURRENT TRANSFORMER ERRORS

Current transformers have phase and ratio errors due to the exciting current needed to
magnetise their cores. The consequence is that the CT secondary current is not an
identical scaled reproduction of the primary current. This heads to protection relay
operation errors, particularly in the tripping time. CT errors are not applicable for
independent definite-time delay overcurrent protection relays.

18 
 
   E‐051 Overcurrent Protection Fundamentals 
 
FINAL MARGIN
After provisions have been made for power circuit breaker interrupting time, protection
relay timing error, overshoot and CT errors, the discriminating protection relay must
just fail to finish its function. Some additional safety margin is needed to make sure that
protection relay tripping does not happen.

SUGGESTED GRADING MARGINS

The following paragraphs provide the suggested complete grading margins for between
different protection elements.

GRADING: PROTECTION RELAY TO PROTECTION RELAY

The complete interval needed to cover power circuit breaker clearing time, protection
relay timing error, overshoot and CT errors, is dependent on the tripping speed of the
power circuit breakers and the protection relay performance. At one time 0.5s was a
typical grading margin. With quicker modern power circuit breakers and a lower
protection relay overshoot time, 0.4s is fair, while under the best circumstances even
lower intervals may be feasible. The utilization of a fixed grading margin is popular, but
it may be better to compute the needed value for each protection relay location. This
more accurate margin comprises a fixed time, covering power circuit breaker short
circuit interrupting time, protection relay overshoot time and a safety margin, plus a
varying time that allows for protection relay and CT errors. Table 3 presents common
protection relay errors according to the applied technology.

It should be kept in mind that application of a fixed grading margin is only allowed at
high short circuit levels that head to short protection relay operating times. At lower
short circuit current levels, with longer operating times, the allowed error defined in
IEC 60255 (7.5% of tripping time) may surpass the fixed grading margin, ending in the
possibility that the protection relay fails to grade precisely while keeping within
specification. This needs consideration when studying the grading margin at low short
circuit current levels.

A feasible answer for finding out the optimum grading margin is to make sure that the
protection relay closer to the short circuit location has a maximum possible timing error

19 
 
   E‐051 Overcurrent Protection Fundamentals 
 
of +2E, where E is the basic timing error. To this complete effective error for the
protection relay, an additional 10% should be added for the total current transformer
error.

Relay protection type

Electro- Static Digital Numerical
mechanical
Typical basic timing error (%) 7.5 5 5 5
Overshoot time (s) 0.05 0.03 0.02 0.02
Safety margin (s) 0.1 0.05 0.03 0.03
Common complete grading margin –
0.4 0.35 0.3 0.3
relay to relay (s)

Table 3. Common protection relay timing errors – standard IDMT protection relays

A suited minimum grading time interval, t’, may be computed as follows:

t t t t t (2)

Where:

ER – protection relay timing error (IEC60255-4)

ECT – allowance for CT ratio error (%)

t – nominal operating time of protection relay nearer to fault location (sec)

tCB – circuit breaker interrupting time (sec)

t0 – protection relay overshoot time (sec)

ts – safety margin (sec)

If, for instance t=0.5s, the time interval for an electromechanical protection relay
operating a conventional power circuit breaker would be 0.375s, while, at the lower
extreme, for a static protection relay operating a vacuum power circuit breaker, the
interval could be 0.25s.

20 
 
   E‐051 Overcurrent Protection Fundamentals 
 
When the overcurrent protection relays have independent definite time delay
protection characteristics, it is not required to include the provision for CT error.
Therefore:

𝑡 𝑡 𝑡 𝑡 𝑡 (3)

Computation of particular grading times for each protection relay can often be
demanding when completing a relay protection grading computation on an electrical
power system. Table 3 also provides feasible grading times at high short circuit current
levels between overcurrent protection relays for different technologies. Where
protection relays of different technologies are utilized, the time appropriate to the
technology of the downstream protection relay should be utilized.

GRADING: FUSE TO FUSE

The tripping time of a fuse is a function of pre-arcing and arcing time of the fusing
component, which adheres to I2t law. So, to accomplish precise co-ordination between
two fuses in connected series, it is mandatory to assure that the complete I2t taken by
the smaller fuse is not higher than the pre-arcing I2t value of the bigger fuse. It has been
founded by tests that acceptable grading between the two fuses will typically be
accomplished if the current rating ratio between them is higher than two.

GRADING: FUSE TO PROTECTION RELAY

For grading inverse time protection relays with fuses, the basic strategy is to make sure
whenever feasible that the protection relay backs up the fuse and not vice versa. If the
fuse is upstream of the protection relay, it is very challenging to keep precise
discrimination at great values of short circuit current because of the fast tripping of the
fuse. The protection relay characteristic best fitted for this co-ordination with fuses is
typically the extremely inverse (EI) protection characteristic as it uses a similar I2t
protection characteristic. To make sure acceptable coordination between protection
relay and fuse, the primary current setting of the protection relay should be roughly
three times the current rating of the fuse. The grading margin for precise coordination,
when conveyed as a fixed figure, should not be less than 0.4s or, when conveyed as a
varying quantity, should have a minimum figure of:

21 
 
   E‐051 Overcurrent Protection Fundamentals 
 
𝑡 0.4𝑡 0.15 (4)

where t is the nominal tripping time of the fuse.

COMPUTATION OF PHASE FAULT OVERCURRENT PROTECTION RELAY

SETTINGS
The precise co-ordination of overcurrent protection relays in an electrical power system
asks for the computation of the approximated protection relay settings in terms of both
current and time. The final settings are then typically printed in appropriate log/log
format to present pictorially that an appropriate grading margin exists between the
protection relays at adjacent substations. Printing is typically completed using suitable
software even though it can be completed by hand. The main protection relay
information can be put in a table such as that presented in Table 4, populating the first
five columns.

Relay current
Fault current (A)
Maximum setting Relay time
CT
Location load Primary multiplier
Ratio Per
Maximum Minimum current (A) Current setting
Cent
(A)

Table 4. Common protection relay data table

It is common to print all time/current characteristics to a common voltage/MVA base

on log/log scales. The graph considers all protection relays in a single path, starting
with the protection relay closest to the load and finishing with the protection relay
closest the source of supply. A different graph is needed for each independent path.
Settings of any protection relays that lie on multiple paths have to be cautiously
considered to make sure that the final adjustments are appropriate for all operational
scenarios. Ground short circuits are treated separately from line short circuits and need
different graph. After protection relay settings have been completed they are out into a
table such as that presented in Table 4, entering information in the last three columns.
This also helps in record keeping during commissioning of the protection relays at site.

22 
 
   E‐051 Overcurrent Protection Fundamentals 
 
INDEPENDENT (DEFINITE) TIME PROTECTION RELAYS
The choice of protection settings for independent (definite) time protection relays
presents little trouble. The overcurrent protection elements have to be given settings
that are lower, by a sensible margin, than the short circuit current that is likely to go to a
fault at the remote end of the electrical system up to which back-up relay protection is
needed, with the minimum elements in operation. The protection settings have to be
sufficiently high to avert protection relay tripping with the maximum probable load, a
suited margin being allowed for large motor starting currents or transformer inrush
transients. Time settings will be selected to provide suited grading margins.

INVERSE TIME PROTECTION RELAYS

When the electrical power system consists of a series of short sections of underground
cable, so that the complete transmission line impedance is low, the value of short circuit
current will be checked mainly by the impedance of transformers or other fixed devices
and will not change greatly with the position of the short circuit. In such situations, it
may be feasible to grade the inverse time protection relays in the same way as definite
time protection relays. Nevertheless, when the prospective short circuit current
changes considerably with the position of the short circuit, it is feasible to make use of
this fact by utilizing both current and time grading to enhance the total tripping of the
protection relay.

The process starts by selection of the adequate protection relay characteristics. Current
settings are then selected, with eventually the time multiplier settings to provide
adequate grading margins between protection relays. Otherwise, the process is similar
to that for definite time delay protection relays.

DIRECTIONAL PHASE SHORT CIRCUIT OVERCURRENT PROTECTION RELAYS

When short circuit current can go in both directions through the protection relay
location, it may be required to make the response of the protection relay directional by
the initiation of a directional control device. The device is provided by use of extra
voltage inputs to the protection relay.

23 
 
   E‐051 Overcurrent Protection Fundamentals 
 
PROTECTION RELAY CONNECTIONS
There are many ways for an appropriate connection of voltage and current signals. The
different connections depend on the phase angle, at unity system power factor, by
which the current and voltage used to the protection relay are displaced. Nevertheless,
only very few connections are utilized in current practice and these are presented
below. In a digital or numerical protection relay, the phase displacements are
determined by software, while electromechanical and static protection relays typically
get the needed phase displacements by connecting the input signals to the protection
relay. The history of the topic ends in the protection relay connections being specified
as if they were received by appropriate connection of the input signals, irrespective of
the actual process used.

90° PROTECTION RELAY QUADRATURE CONNECTION

This is the typical arrangement for static, digital or numerical protection relays.
Depending on the angle by which the used voltage is shifted to generate maximum
protection relay sensitivity (the protection Relay Characteristic Angle, or RCA), two
types are usable.

90°-30° CHARACTERISTIC (30° RCA)

The A phase protection relay element is furnished with Ia current and Vbc voltage
displaced by 30° in an anti-clockwise direction. In this situation, the protection relay
maximum sensitivity is generated when the current lags the system phase to neutral
voltage by 60°. This relay connection provides precise directional tripping zone over the
current range of 30° leading to 150° lagging as shown in Figure 11. The protection relay
sensitiveness at unity power factor is 50% of the relay maximum sensitivity and 86.6%
at zero power factor lagging. This characteristic is suggested when the protection relay
is used for the protection of plain lines with the zero sequence source behind the
relaying location.

   

24 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Ia
Zero torque line

Va

30°
Vbc’

30°

Vbc

Vc Vb

A phase device connected Ia Vbc

B phase device connected Ib Vca
C phase device connected Ic Vab

Figure 11. Vector graph for the 90°-30° arrangement (phase A device)

90°-45° CHARACTERISTIC (45° RCA)

The A phase protection relay device is furnished with current Ia and voltage Vbc shifted
by 45° in an anti-clockwise direction. The protection relay maximum sensitiveness is
generated when the current lags the system phase to neutral voltage by 45°. This
arrangement provides a precise directional tripping zone over the current range of 45°
leading to 135° lagging. The protection relay sensitivity at unity power factor is 70.7%
of the maximum torque and the same at zero power factor lagging as shown in Figure
12. This arrangement is suggested for the protection of transformer feeders or lines that
have a zero sequence source in front of the protection relay. It is requirement in the
situation of parallel transformers or transformer lines, to assure precise protection

25 
 
   E‐051 Overcurrent Protection Fundamentals 
 
relay tripping for short circutis beyond the star/delta transformer.

Ia

Zero torque
Va
line
Vbc’

45° 45°

Vbc

Vc Vb

A phase device connected Ia Vbc

B phase device connected Ib Vca
C phase device connected Ic Vab

Figure 12. Vector graph for the 90°-45° arrangement (phase A device)

For a digital or numerical protection relay, it is typical to grant user selection of the RCA
within a wide range.

In theory, three short circuit conditions can start maloperation of the directional device:

- a line-line-earth short circuit on a plain line

- a line-earth short circuit on a transformer line with the zero sequence source in
front of the protection relay

- a line-line short circuit on a power transformer with the protection relay

looking into the transformer delta winding. These circumstances are presumed to
establish the maximum angular shift between the current and voltage quantities at the
protection relay. The magnitude of the current input to the protection relay is

26 
 
   E‐051 Overcurrent Protection Fundamentals 
 
insufficient to start the overcurrent device to trip. The possibility of maloperation with
the 90°-45° arrangement is non-existent.

USAGE OF DIRECTIONAL PROTECTION RELAYS

If non-unit, non-directional protection relays are used to parallel lines having a common
generator, any short circuits that might happen on any one transmission line will,
irrespective of the protection relay settings utilized, set apart both feeders and totally
disconnect the power supply. With this system arrangement, it is essential to use
directional protection relays at the receiving end and to grade them with the non-
directional protection relays at the sending end, to assure precise discriminative
tripping of the protection relays during transmission line faults. This is accomplished by
setting the directional protection relays R’1 and R’2 in Figure 13 with their directional
components looking into the protected feeder, and providing them with lower time and
current settings than protection relays R1 and R2. The common practice is to program
protection relays R’1 and R’2 to 50% of the normal full load of the protected feeder and
0.1 TMS, but care has to be taken to assure that the uninterrupted thermal rating of the
protection relays of twice rated current is not surpassed.

R1 R1’

I>  I> 
Fault
Load

R2 R2’

I>  I> 

Figure 14. Directional protection relays utilized to parallel lines

RING MAINS
Typical scheme within electrical distribution networks is the Ring Main. The main
reason for its application is to keep supplies to consumers in case of short circuit

27 
 
   E‐051 Overcurrent Protection Fundamentals 
 
conditions happening on the interconnecting lines. Current may go in either direction
through the different protection relay locations, and hence directional overcurrent
protection relays are used.

In the situation of a ring main fed at the single point, the settings of the protection relays
at the supply side and at the mid-point substation are same. Hence, they can be made
non-directional, if, in the latter situation, the protection relays are placed on the same
line, that is, one at each end of the line. It is interesting to note that when the number of
lines round the ring is an even number, the two protection relays with the same tripping
time are at the same substation. Hence, they will have to be directional. When the
number of lines is an odd number, the two protection relays with the same tripping time
are at different substations and hence do not need to be directional. It may also be
remembered that, at intermediate substations, whenever the tripping time of the
protection relays at each substation are different, the difference between their tripping
times is never lower than the grading margin, so the protection relay with the longer
tripping time can be non-directional. With modern numerical protection relays, a
directional facility is often applicable for little or no additional cost, so that it may be
easier in practice to use directional protection relays at all locations. Also, in the case of
an extra line being installed subsequently, the protection relays that can be non-
directional have to be re-determined and will not inevitably be the same – giving rise to
issues of changing a non-directional protection relay for a directional one. If a VT was
not initially given, this may be very challenging to provide at a later date.

RING MAINS GRADING

The typical grading process for protection relays in a ring main circuit is to trip the ring
at the supply point and to grade the protection relays first clockwise and then anti-
clockwise. That is, the protection relays looking in a clockwise direction around the ring
are made to trip in the sequence 1-2-3-4-5-6 and the protection relays looking in the
anti-clockwise direction are made to trip in the sequence 1’-2’-3’-4’-5’-6’.

The directional protection relays are set in accordance with the invariable standard,
relevant to all forms of directional relay protection that the current in the system has to
go from the substation bus into the protected feeder so the protection relays may trip.

28 
 
   E‐051 Overcurrent Protection Fundamentals 
 
Tripping of the faulted feeder is completed according to time and short circuit current
direction. As in any parallel electrical system, the short circuit current has two parallel
paths and separates itself in the inverse ratio of their impedances. Therefore, at each
substation in the ring, one set of protection relays will be made defunct because of the
direction of current flow, and the other set operative. It will also be noted that the
tripping times of the protection relays that are defunct are quicker than those of the
operative protection relays, with the exclusion of the mid-point substation, where the
tripping times of protection relays are equal.

The protection relays that are operative are graded downwards towards the short
circuit location and the last to be impacted by the short circuit trips first. This is
applicable to both paths to the short circuit. Accordingly, the affected feeder is the only
one to be switched off from the ring and the power supply is kept to all the substations.

When two or more power sources supply the ring main, time graded overcurrent
protection is challenging to use and complete discrimination may not be achievable.
With two power sources of supply there are two possible solutions. The first is to trip
the ring at one of the supply locations, whichever is more practical, by means of a suited
high set instantaneous overcurrent protection relay. The ring is then graded as in the
situation of a single supply. The second technique is to treat the portion of the ring
between the two supply locations as a continuous bus separate from the ring and to
save it with a unit protection arrangement, and then go forward to grade the ring as in
the case of a single supply.

GROUND FAULT PROTECTION

In the foregoing paragraph, care has been primarily orientated towards line fault
overcurrent protection. More sensitive protection against ground short circuit currents
can be accomplished by utilizing a protection relay that acts only to the residual system
current, since a residual component is available only when short circuit current goes to
ground. The ground fault protection relay is hence totally untouched by load currents,
whether balanced or not, and can be provided a setting which is determined only by the
equipment arrangement and the presence of unbalanced leakage or capacitance

29 
 
   E‐051 Overcurrent Protection Fundamentals 
 
currents to ground. This is crucial consideration if settings of only a several percent of
system rating are looked at, since leakage currents may generate a residual quantity of
this magnitude.

On the whole, the low settings allowable for ground fault protection relays are very
practical, as ground short circuits are not only by far the most frequent of all short
circuits, but may be determined in magnitude by the neutral grounding impedance, or
by ground contact resistance. The residual element is extracted by linking the phase
current transformers in parallel as presented in Figure 15. The simple arrangement
presented in Figure 15(a) can be extended by linking overcurrent devices in the
individual phase leads, as presented in Figure 15(b), and placing the ground fault
protection relay between the star points of the protection relay group and the current
transformers.

Line short circuit overcurrent protection relays are typically given on only two lines
since these will sense any interphase short circuit; the arrangements to the ground
short circuit protection relay are unaffected by this condition. The principal scheme is
presented in Figure 15(c).

   

30 
 
   E‐051 Overcurrent Protection Fundamentals 
 

I >

I> I> I> I >

I> I> I >

Figure 15. Residual arrangement of current transformers to ground fault protection

relays

The common settings for ground fault protection relays are 30%-40% of the total-load
current or minimum ground short circuit current on the portion of the system being
protected.

31 
 
   E‐051 Overcurrent Protection Fundamentals 
 
EFFECTIVE SETTING OF GROUND FAULT PROTECTION RELAYS
The primary setting of an overcurrent protection relay can typically be taken as the
protection relay setting multiplied by the CT ratio. The CT can be presumed to keep a
sufficiently precise ratio so that, conveyed as a percentage of rated current, the primary
setting is directly relative to the protection relay setting. Nevertheless, this may not be
correct for ground fault protection relay. The operation changes according to the
applied protection relay technology.

STATIC, DIGITAL AND NUMERICAL PROTECTION RELAYS

When static, digital or numerical protection relays are applied the relatively low value
and fixed variation of the protection relay burden over the protection relay setting
range ends in the above statement being true. The variation of input burden with
current should be verified to assure that the change is sufficiently small. If not,
substantial errors may happen, and the setting process will have to follow that for
electromechanical protection relays.

ELECTROMECHANICAL PROTECTION RELAYS

When using an electromechanical protection relay, the ground fault device typically will
be similar to the phase devices. It will have a similar VA usage at setting, but will enforce
a far greater burden at nominal or rated current, because of its lower setting. For
instance, a protection relay with a setting of 20% will have an impedance of 25 times
that of a similar device with a setting of 100%. Very often, this burden will surpass the
rated burden of the current transformers. It might be believed that correspondingly
higher current transformers should be applied, but this is conceived to be unneeded.
The current transformers that handle the line burdens can trip the ground fault
protection relay and the greater errors can be allowed for.

Not only is the exciting current of the energising current transformer relatively high due
to the great burden of the ground fault protection relay, but the voltage drop on this
protection relay is impressed on the other current transformers of the paralleled group,
whether they are transferring primary current or not. The overall exciting current is
hence the product of the magnetizing loss in one CT and the other current transformers
operating in parallel. The overall magnetising loss can be considerable in comparison

32 
 
   E‐051 Overcurrent Protection Fundamentals 
 
with the working current of the protection relay, and in utmost situations where the
setting current is small or the current transformers are of low performance, may even
surpass the output to the protection relay. The ‘effective setting current’ in secondary
terms is the sum of the protection relay setting current and the overall excitation loss.
Strictly speaking, applied setting is the vector sum of the protection relay setting
current and the overall exciting current, but the arithmetic sum is near sufficient,
because of the similarity of power factors. It is informative to compute the applied
setting for a range of setting values of a protection relay, a process that is shown in
Table 5, with the results presented in Figure 16.

The effect of the comparatively high protection relay impedance and the summation of
CT excitation losses in the residual circuit is increased still further by the fact that, at
setting, the flux density in the current transformers matches to the bottom bend of the
excitation characteristic. The exciting impedance under these circumstances is
comparatively low, causing the ratio error to be big. The current transformer really
enhances in operation with raised primary current, while the protection relay
impedance reduces until, with an input current few times higher than the primary
setting, the multiple of setting current in the protection relay is appreciably greater than
the multiple of primary current setting which is implemented on the primary circuit.
This causes the protection relay tripping time to be shorter than might be anticipated.

At still greater input currents, the CT operation falls off until eventually the output
current ceases to grow considerably. Beyond this value of input current, function is
additionally complicated by distortion of the output current waveform.

   

33 
 
   E‐051 Overcurrent Protection Fundamentals 
 
Table 5. Computation of actual relay settings

Protection relay
Coil voltage Exciting Applied setting
plug setting
at setting (V) current Ie
% Current (A) Current (A) %
5 0.25 12 0.583 2 40
10 0.5 6 0.405 1.715 34.3
15 0.75 4 0.3 1.65 33
20 1 3 0.27 1.81 36
40 2 1.5 0.17 2.51 50
60 3 1 0.12 3.36 67
80 4 0.75 0.1 4.3 86
100 5 0.6 0.08 5.24 105

35

30
Secondary Voltage

25

20

15

10 Current transformer
excitation characteristic
5

0
0 0.5 1 1.5
Exciting current (A)

100
90
80
Effective setting (%)

70
60
50
40
30
20
10
0
0 20 40 60 80 100
Relay setting (%)

Figure 16. Applied setting of ground fault protection relay

34 
 
   E‐051 Overcurrent Protection Fundamentals 
 

TIME GRADING OF ELECTROMECHANICAL GROUND FAULT PROTECTION

RELAYS
The time grading of ground fault protection relays can be organized in the same style as
for phase short circuit protection relays. The time/primary current characteristic for
electromechanical protection relays cannot be kept proportionate to the protection
relay characteristic with anything like the precision that is feasible for phase short
circuit protection relays. As mentioned above, the ratio error of the current
transformers at protection relay setting current may be very big. It is evident that time
grading of electromechanical ground fault protection relays is not simple issue as the
process used for phase protection relays in Table 4. Either the above considerations
must be considered with the errors computed for each current level, making the overall
procedure much more challenging, or longer grading margins must be applied.
Nevertheless, for other relay protection types, the process used for line fault protection
relays can be applied.

SENSITIVE GRONUD FAULT RELAY PROTECTION

LV electrical systems are not typically grounded through impedance, due to the over-
voltages that may take place and eventful safety considerations. HV electrical systems
may be organized to suit such over-voltages, but not the majority of LV systems.
Nevertheless, it is quite typical to ground HV electrical systems through impedance that
fixes the ground fault current. Next, in some countries, the resistivity of the ground path
may be great due to the nature of the earth itself (e.g. desert). A short circuit to ground
not involving ground conductors may end in the transfer of only a small current,
insufficient to trip a typical relay protection mechanism. A similar trouble also occurs in
the case of broken line conductors, which, after coming down on to hedges or dry
metalled roads, stay energised because of the low leakage current, and hence present a
risk to life. To resolve the issue, it is mandatory to supply ground fault relay protection
mechanism with a setting that is substantially lower than the normal line relay
protection. This is not an issue for a modern digital or numerical protection relay.
Nevertheless, older electromechanical or static protection relays may introduce
problems due to the great effective burden they may present to the CT.

35 
 
   E‐051 Overcurrent Protection Fundamentals 
 
The needed sensitivity cannot typically be given by means of conventional CTs. A core
balance current transformer (CBCT) will typically be utilized. The CBCT is a current
transformer placed around all three line (and neutral if available) conductors so that the
CT secondary current is relative to the residual (i.e. ground) current. Such a CT can be
designed to have any convenient ratio suited for tripping a sensitive ground fault relay
protection device. By use of such arrangements, ground fault settings down to 10% of
the current rating of the line to be protected can be achieved.

Attention has to be taken to place a CBCT correctly in an underground cable circuit. If

the underground cable sheath is grounded, the ground link from the underground cable
gland/sheath junction has to be taken through the CBCT primary to assure that line-
sheath short circuits are sensed. Figure 17 presents the correct and wrong
arrangements. With the wrong arrangement, the short circuit current in the sheath is
not detected as an unbalance current and therefore protection relay tripping does not
happen.

The normal residual current that may go during normal operation limits the usage of
non-directional sensitive ground fault relay protection. Such residual effects can happen
due to unbalanced leakage or capacitance in the electrical system.

   

36 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Cable gland

Cable box

Cable gland / sheath

ground connection I >

No operation

I >

Operation

I >

Figure 17. Installation of core balance current transformers (a) Physical arrangement
(b) Wrong placement (c) Correct placement

37 
 
   E‐051 Overcurrent Protection Fundamentals 
 
DIRECTIONAL GROUND FAULT OVERCURRENT RELAY PROTECTION
Directional ground fault overcurrent has to be used in the following cases:

- for ground fault protection where the overcurrent relay protection is by

directional protection relays

- in insulated-ground power networks

- in Petersen coil grounded power networks

- where the sensitivity of sensitive ground fault relay protection is deficient–

application of a directional ground fault protection relay may give higher
sensitivity

The protection relay device previously mentioned as phase short circuit elements react
to the flow of ground fault current, and it is crucial that their directional reaction is
adequate for this situation.

PROTECTION RELAY CONNECTIONS

The residual current is extracted as presented in Figure 15. Since this current may be
gained from any line, to get a directional response it is mandatory to get an adequate
quantity to polarise the protection relay. In digital or numerical protection relays there
are typically two given options.

RESIDUAL VOLTAGE
A suited amount is the residual voltage of the system. This residual voltage is the vector
sum of the individual line voltages. If the secondary windings of a three-line, five limb
voltage transformer or three single-line units are linked in broken delta, the voltage
built up across its terminals will be the vector sum of the line to earth voltages and
therefore the residual voltage of the electrical system, as presented in Figure 18.

38 
 
   E‐051 Overcurrent Protection Fundamentals 
 

3I0

Va Va
Va2

3V0
Vc Vb Vc Vb

Figure 18. Voltage polarized directional ground fault protection relay (a) Protection
relay links(b) Balanced system (zero residual volts) (c) Unbalanced system phase A to
earth fault (3V0 residual volts)

The primary star point of the VT has to be grounded. Nevertheless, a three-line, three
limb VT is not suitable, as there is no path for the residual magnetic flux.

When the main voltage transformer related with the high voltage system is not supplied
with a broken delta secondary winding to polarise the directional ground short circuit

39 
 
   E‐051 Overcurrent Protection Fundamentals 
 
protection relay, it is allowable to utilize three single-line interposing voltage
transformers. Their primary windings are arranged in star and their secondary
windings are arranged in broken delta. For satisfactory functioning, nevertheless, it is
required to assure that the main voltage transformers are of a suitable design to
replicate the residual voltage and that the star point of the primary winding is directly
grounded. Also, the star point of the primary windings of the interposing voltage
transformers has to be linked to the star point of the secondary windings of the main
voltage transformers.

The residual voltage will be zero for balanced line voltages. For simple ground short
circuit circumstances, it will be same to the depression of the faulted line voltage. In all
situations the residual voltage is same to three times the zero sequence voltage drop on
the source impedance and is hence shifted from the residual current by the
characteristic angle of the source impedance. The residual quantities are applicable to
the directional device of the ground short circuit protection relay. The residual current
is phase offset from the residual voltage and therefore angle adjustment is needed.
Generally, the current will lag the polarising voltage. The grounding system
arrangement also impacts the Relay Characteristic Angle (RCA), and the next settings
are typical:

- Resistance-grounded system: 0° RCA

- Electrical distribution system, directly-grounded: -45° RCA

- Electrical transmission system, directly-grounded: -60° RCA

The different adjustments for distribution and transmission electrical systems come up
from the various X/R ratios found in these systems.

NEGATIVE SEQUENCE CURRENT

The residual voltage at any system point may not be sufficient to polarise a directional
protection relay, or the voltage transformers available may not meet the conditions for
providing residual voltage. In these situations, negative sequence current can be applied
as the polarising quantity. The short circuit current direction is found out by cross

40 
 
   E‐051 Overcurrent Protection Fundamentals 
 
comparison of the negative sequence voltage with the negative sequence current. The
RCA has to be set based on the angle of the negative phase sequence source voltage.

GROUND FAULT RELAY PROTECTION ON INSULATED ELECTRICAL NETWORKS

Periodically, a power system is operated totally insulated from the ground. The benefit
of this arrangement is that a single line-ground short circuit on the system does not
cause any ground short circuit current to flow, and so the whole electrical system stays
functional. The electrical system has to be made to resist high transient and steady-state
over-voltages nevertheless, so its application is typically limited to low and medium
voltage systems. It is important that sensing of a single line-ground short circuit is
accomplished, so that the short circuit can be traced and corrected. While electrical
system performance is unaffected for this situation, the occurrence of a second ground
short circuit allows significant currents to flow.

The absence of ground short circuit current for a single line-ground short circuit
evidently presents some challenges in short circuit current detection. Two techniques
are available using modern protection relays.

RESIDUAL VOLTAGE
When a single line-ground short circuit happens, the healthy line voltages increase by a
factor of 3 and the three phase voltages no longer have a vector sum of zero. Therefore,
a residual voltage device can be utilized to sense the short circuit current. Nevertheless,
the technique does not allow any discrimination, as the unbalanced voltage happens on
the complete of the impacted portion of the electrical system. One benefit of this
technique is that no CTs are needed, as voltage is being measured. Grading is a problem
with this technique, since all protection relays in the impacted section will see the short
circuit. It may be feasible to utilize definite-time grading, but in principle, it is not
feasible to give completely discriminative protection using this method.

41 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Ia1
Ib1

IR1
-jXc1

Ia2
Ib2
IH1
IR2
-jXc2

Ia3
IH2
Ib3
IH1+ IH2+ IH3

IR3 -jXc3

IR3= IH1+ IH2+ IH3- IH3

IH3
= IH1+ IH2

Figure 19. Current distribution in an insulated electrical system with a C phase-ground

short circuit

SENSITIVE GROUND FAULT

This technique is generally used to MV electrical systems, as it relies on detection of the
imbalance in the per-line charging currents that exist. Figure 19 presents the case that
happens when a single line-ground short circuit is present. The protection relays on the
healthy lines see the unbalance in charging currents for their own lines. The protection
relay in the faulted line detects the charging currents in the rest of the electrical system,
with the current of its’ own lines cancelled out. Figure 20 presents the phasor graph.

   

42 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Va
Restrain
Vap
IR1
Ib1

Operate
Ia1

Vbf
Vcpf
Vbpf
An RCA
setting of+90°
shifts the
“center of the
characteristic”
Vres=-3V0
IR3=-(IH1+IH2)

Figure 20. Phasor graph for insulated electrical system with C line-ground short circuit

Use of Core Balance CTs is mandatory. With reference to Figure 20, the unbalance
current on the healthy lines lags the residual voltage by 90º. The charging currents on
these lines will be √3 times the normal value, as the line-ground voltages have
increased by this amount. The magnitude of the residual current is hence three times
the steady-state charging current per line. As the residual currents on the live and
faulted lines are in anti-phase, application of a directional ground fault protection relay
can give the needed discrimination.

The polarising quantity applied is the residual voltage. By shifting it by 90°, the residual
current detected by the protection relay on the faulted line lies within the ‘operate’ area
of the directional characteristic, while the residual currents on the healthy lines lie
within the ‘restrain’ region. Therefore, the RCA needed is 90°. The protection relay
setting has to lie between one and three times the per-line charging current.

This may be computed at the design stage, but check by means of tests on-site is
common. A single line-ground short circuit is deliberately applied and the resulting

43 
 
   E‐051 Overcurrent Protection Fundamentals 
 
currents noted, a procedure made simpler in a modern digital or numeric protection
relay by the measurement facilities given. As previously noted, usage of such a fault for
a short period does not call for any interruption to the electrical network, or short
circuit currents, but the duration needs to be as short as possible to save against a
second such short circuit happening.

It is also feasible to dispense with the directional device if the protection relay can be
programmed at a current value that lies between the charging current on the line to be
protected and the charging current of the rest of the electrical system.

GROUND FAULT RELAY PROTECTION ON PETERSEN COIL GROUNDED

ELECTRICAL NETWORKS
Petersen Coil grounding is a special arrangement of high impedance grounding. The
electrical system is grounded via a reactor, whose reactance is designed to be same to
the complete electrical system capacitance to ground. Under this condition, a single line-
ground short circuit does not end in any ground fault current in steady state operation.
Hence, the impact is similar to having an insulated electrical system. The effectiveness
of the arrangement depends on the precision of tuning of the reactance value –
modifications in system capacitance (for example, due to electrical system configuration
variations) demand modifications to the coil reactance. In reality, absolute matching of
the coil reactance to the electrical system capacitance is hard to accomplish, so that a
small ground short circuit current will flow. Petersen Coil grounded electrical systems
are typically found in locations where the electrical system consists dominantly of rural
overhead transmission lines, and are especially useful in locations subject to a high
incidence of transient short circuits.

To understand how to exactly use ground short circuit protection to such electrical
systems, system behaviour under ground fault conditions has to be understood. Figure
21 presents a basic electrical network grounded through a Petersen Coil. The formulas
clearly present that, if the reactor is precisely tuned, no ground fault current will flow.

Figure 22 presents a radial distribution system grounded using a Petersen Coil. One
distribution feeder has a line-ground short circuit on phase C. Figure 23 presents the

44 
 
   E‐051 Overcurrent Protection Fundamentals 
 
resulting phasor graphs, presuming that no resistance is present. In Figure 23(a), it can
be noted that the short circuit causes the healthy line voltages to increase by a factor of
√3 and the charging currents lead the voltages by 90°.

-IB

-IC

-IB -(Vac/jXc)=-Ic
Van/jXL
If -jXc -jXc -jXc
Petersen
jXL
coil If=-IB-IC+Van/jXL=0
If Van/jXL=IB+IC

‐IC 
A 
IL 

‐IB 
Vac Vab

N 

C B

Figure 21. Ground short circuit in Petersen coil grounded electrical system

45 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Ia1
Ib1

IR1
-jXc1

Ia2
IL Ib2
IH1
jXL IR2
-jXc2

Ia3
IH2
Ib3
Ic3=IF

IR3 -jXc3
IF

IL=IF+ IH1+ IH2- IH3 IH3

Figure 22. Distribution of currents during a C line-ground fault-radial distribution

electrical system

46 
 
   E‐051 Overcurrent Protection Fundamentals 
 

IH A 3V
I
IH2

IH
Ibf 
Ia1
N 

C B

IL
IR1=IH1 -IH1
Ibf
-IH2
IR3=IF+IH3=IL-IH1-IH2
Ia1
IR3

Vres=-3V0 Vres=-3V0

Figure 23. C line-ground short circuit in Peterson Coil grounded electrical system:
theoretical situation-no resistance present in XL or XC (a) Capacitive and inductive
currents (b) Unfaulted feeder (c) Faulted feeder

Utilizing a CBCT, the unbalance currents detected on the healthy transmission lines can
be seen to be a simple vector addition of Ia1 and Ib1 and this lies at precisely 90o lagging
to the residual voltage (as shown in Figure 23(b)). The magnitude of the residual
current IR1 is same to three times the steady-state charging current per line. On the
faulted transmission line, the residual current is same to IL - IH1 - IH2, as presented in
Figure 23(c) and more clearly by the zero sequence network of Figure 24.

Nevertheless, in real situations, resistance is there and Figure 25 presents the resulting
phasor graphs. If the residual voltage Vres is utilized as the polarising voltage, the

47 
 
   E‐051 Overcurrent Protection Fundamentals 
 
residual current is phase shifted by an angle lower than 90° on the faulted transmission
line and higher than 90° on the healthy transmission lines. Therefore, a directional
protection relay can be applied, and with an RCA of 0°, the live transmission line
residual current will fall in the ‘restrain’ section of the protection relay characteristic
while the faulted circuit residual current falls in the ‘operate’ section.

IOF
IROF Faulted
line
IROF
Healthy
IROH
lines

IL IH3 IH2 IH1

3XL -V0 Xco

Where:

IROF – Residual current on faulted line

IROH – Residual current on healthy line

Hence, it can be seen that:

IOF=IL-IH1-IH2-IH3
IROF=IH3+IOF

So:
IROF=IL-IH1-IH2

Figure 24. Zero sequence electrical network presenting residual currents

   

48 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Resistive component in line

Resistive component (IH1+ IH2+ IH3)’

A 3V0
in earthing coil
IL’

N 

C B

IL
Operate Restrain Operate Restrain
IR1=IH1
-IH1-IH2

IR3 IR3=IF+IH3=IL-IH1-IH2

Vres=-3V0 Vres=-3V0

Zero torque
line for 0° RCA

Figure 25. C line-ground short circuit in Petersen Coil grounded network: realistic
situation with resistance present in XL or XC (a) Capacitive and inductive currents with
resistive components (b) Unfaulted circuit (c) Faulted circuit

Usually, a resistance is intentionally placed in parallel with the Petersen Coil to make
sure a measurable ground short circuit current and increase the angular difference
between the residual signals to help protection relay application. Having demonstrated
that a directional protection relay can be applied, two options are available for the
method of protection element that can be used – sensitive ground short circuit and zero
sequence watt-metric.

49 
 
   E‐051 Overcurrent Protection Fundamentals 
 
SENSITIVE GROUND SHORT CIRCUIT PROTECTION
To use this form of relay protection, the protection relay has to meet two requirements:

- current measurement setting capable of being set to very low figures

- an RCA of 0°, and capable of fine adjustment around this figure

The sensitive current component is needed because of the very low current that may
flow – so adjustments of less than 0.5% of rated current may be needed. Nevertheless,
as compensation by the Petersen Coil may not be complete, low levels of steady-state
ground short circuit current will go and raise the residual current detected by the
protection relay. Typically applied setting value is the per line charging current of the
circuit being protected. Fine adjustment of the RCA is also needed around the 0° setting,
to compensate for coil and circuit resistances and the performance of the CT applied. In
reality, these settings are best implemented on site through careful application of short
circuits and keeping record of the resulting currents.

SENSITIVE WATTMETRIC PROTECTION

It can be noted in Figure 25 that a small angular difference exists between the spill
current on the live and faulted transmission lines. Figure 26 presents how this angular
difference provides increase to active components of current which are in anti-phase to
each other.

   

50 
 
   E‐051 Overcurrent Protection Fundamentals 
 

Vres=-3V0
Active
component of IR3 -IH1-IH2 Operate
residual current IL

faulted line

Active
component of
IR1 Zero torque line
residual current Restrain
for 0° RCA
live line

Figure 26. Resistive components of spill current

Accordingly, the active portions of zero sequence power will also lie in similar planes
and a protection relay able to sense active power can make a discriminatory call. If the
watt-metric portion of zero sequence power is sensed in the forward direction, it
suggests a short circuit on that transmission line, while a power in the reverse direction
shows a short circuit elsewhere on the electrical system. This protection arrangement is
more popular than the sensitive ground short circuit arrangement, and can give higher
security against false operation due to spurious CBCT output under non-ground short
circuit conditions. Watt-metric power is computed using residual measures instead of
zero sequence ones. The final figures are hence nine times the zero sequence quantities
as the residual figures of current and voltage are each three times the corresponding
zero sequence figures. The formula applied is:

𝑉 𝐼 cos 𝜑 𝜑 9 𝑉 𝐼 cos 𝜑 𝜑 (5)

Where

V – residual voltage

I – residual current

V – zero sequence voltage

I – zero sequence current

51 
 
   E‐051 Overcurrent Protection Fundamentals 
 
φ – angle between V and I

φ – protection relay characteristic angle setting

The current and RCA adjustments are as for a sensitive ground short circuit protection
relay

HI-Z – HIGH IMPEDANCE DOWNED CONDUCTOR RELAY PROTECTION

High impedance (“Hi-Z”) short circuits are typically determined as the unwanted
contact of an electrical circuit with a nonconductive surface like asphalt road, sand,
wooden fences or some other area or object which prevents the short circuit current to
a level below that which can be sensed by conventional overcurrent and ground fault
protection relays. In some situations even sensitive ground short circuit current relay
protection cannot reliably sense such low levels of short circuit current transfer.
Undetected high impedance short circuits such as downed conductors are unsafe for
nearby staff, the public, and livestock. The main aim of clearing such short circuits is
hence towards protection of human life and not towards the integrity and selectivity of
the electrical power system. Hence, high-impedance short circuit detection is becoming
more and more important for utilities and relay protection engineers, as moral and legal
challenges push them to take a greater care for all that may be in the proximity of
electrical power assets.

The common short circuit scenario is where an overhead distribution conductor has
fallen. As mentioned, non-conductive surfaces will try to stop the short circuit current
which flows. This is due to their high resistivity, and the requirement for the ground
short circuit current to go back to the generator, and a legitimate zero sequence current
source. Generally, this will demand the current going back to the nearest adjacent
grounded tower, for the return current then to go in any aerial ground wire. If the
circuit has no ground wire, the short circuit current will need to return to the ground
star-point of the upstream distribution transformer. In the situation of a line falling onto
rock or sand, the challenge is made all the harder in that the initial contact surface, and
many metres of short circuit current flow in the same material composition which can
drastically fix the prospective short circuit current. In the situation of a conductor
falling onto a fence, if the wood is dry this may have a great resistivity, but that great

52 
 
   E‐051 Overcurrent Protection Fundamentals 
 
resistance may only apply for a several metres, until the current can go in moist soil
underground.

Sand is a special problem, because once a feeder falls onto it and arc current strikes, the
heat in the arc can induce clumps of the surrounding sand to turn to glass, which partly
insulates the line from the ground. Asphalt roads too can pose short circuit detection
issues due to the natural good insulating properties, particularly if the road is dry.

It has been pointed out that such a downed feeder will tend to strike a fault arc. This
provides a real benefit in terms of sensing of such short circuits, because arcs have
special characteristics:

- Arcs are rich in harmonics, with a persistence and randomness of the harmonic
profile which is not commonly detected in normal load current.

- The heat and energy in the arc, plus remaining tension in the conductor, tend to
induce movement of the line – this leads to randomness in the flow of
fundamental current too.

- On several surfaces, the heat from the arc will impact the insulating
characteristics, and any moisture in the contact location – this will cause
randomness in current harmonics, fundamental current, and the continuity of
any short circuit current flow.

Modern line management protection relays provide numerical algorithms which act to
(1) sustained intermittence in current flow, and (2) strange levels or preponderance of
harmonics, to be used independently, or in combination, as a reliable arrangement to
sense high impedance short circuits such as downed feeders. These arrangements can
be directionalised using power-based methods.

53