You are on page 1of 37

Automatic Relay Setting

By
Donald M. MacGregor, Ph.D. (Electrocon International, Inc, USA)
A. T. (Tony) Giuliante, Fellow of IEEE (ATG Exodus, Inc., USA)
Russell W. Patterson, P.E. (Tennessee Valley Authority, USA)

Presented to
56th Georgia Tech Protective Relaying Conference
Atlanta, Georgia
May 1-3, 2002

Automatic Relay Setting


Donald M. MacGregor

A. T. (Tony) Giuliante

Russell W. Patterson

1. Introduction
Modern numerical relays provide many protection functions in a single package. In order to manage
the large number of elements and settings, an automatic, systematic, setting procedure is almost
essential. Here we present setting rules for the electrical parameters, such as distance element reach
and overcurrent pickup. These rules are based largely on the standards and experience at the
Tennessee Valley Authority, and follow the principles given in [1]. They are widely applicable to both
stepped-distance schemes and directional comparison pilot schemes such as POTT, PUTT, and blocking
schemes.
A protection simulation environment, described in section 2, enables a user to encode the setting rules.
Detailed reports identify possible conflicts in the setting rules.
Section 3 shows how the user can run the algorithms and adjust the setting parameters. Each
algorithm includes a module that specifies the tap names for the relay models in use, allowing settings
to be saved in the system database and transferred to the relay electronically.
Section 4 outlines a typical case involving directional overcurrent and distance elements in a
Permissive Overreach Transfer Trip scheme. This scheme uses direct tripping from zone 1 and
permissive tripping from a forward pilot element, with two time-delayed zones for a backup steppeddistance scheme. The permissive signal is echoed to allow sequential tripping after one breaker opens.
Section 5 explains the setting rules in detail. All directional comparison pilot schemes use a
combination of underreaching, overreaching and reverse blocking zones, as follows:

Distance zone 1 (phase and ground) to cover close-in faults

Zone 2 and/or forward pilot zone to cover the entire protected line

Forward zone 3 as backup protection for downstream lines

Blinders or other load-encroachment elements

Overcurrent fault detectors

Reverse pilot zone, coordinated with the facing forward pilot zone for blocking

Power-swing (out-of-step) detectors

Ground and negative-sequence overcurrent elements with several levels of sensitivity

Directional elements

Zone timers

Neutral time-overcurrent element

An apparent impedance and an operating margin, defined in section 6, measure how close a relay
element is to its operating limit for a particular fault. For solid faults, overcurrent elements use the
multiples of pickup and distance elements use the relative distance of the apparent line impedance
from the characteristic. For resistive faults, the maximum fault resistance seen by zones 1 to 3 is
presented graphically as a function of distance, using the detailed phasor operating equations for the
particular relay model.
Section 7 demonstrates a stepped-event simulation of the permissive tripping sequence. This checks
the settings and will find any primary-backup miscoordinations.

2. Protection Simulation
The protection simulation environment allows a user to compute settings and send them to a relay, or
to read settings from a relay and test them in the modeled system. The components are:

A network model (buses, generators, lines, shunts, and transformers) and a short-circuit analysis
with high-level commands for faults and outage contingencies [2]. Currents and voltages are
treated as steady-state phasors.

A library of detailed relay models [3]. A relay model consists of instantaneous overcurrent, time
overcurrent, directional, distance, voltage, timer, and recloser elements, with auxiliary elements for
internal logic and pilot (teleprotection) schemes. Special code for each relay model interprets the
setting names and evaluates the comparators using a steady-state phasor analysis [4]. As a result,
element response is always based on the actual relay logic. Actual settings are modeled so that the
relay model is set in the same way as the physical device.

Rules for locating relays. An integrated database [5], with an interactive editor, contains the CTs
and VTs connecting the relays to the network, and specifies the protected equipment and its logical
breakers.

A macro facility. The macro language has many commands associated with a high-level language,
such as IF-THEN-ELSE, DOWHILE, and DO loops tailored to power system applications. For
example, DOXFMRS and DOLINES find the transformers and lines at a bus, and DOPATH and
DOREMOTE search through a meshed network with load-tap buses. Standard functions (e.g. SIN,
ABS, and POLAR) and special protection functions (e.g. OPERATING_CYCLES and
GET_TOC_TIME) are installed. Support for defining and looping through sets of buses, branches,
and relay elements is provided. The engineer can access all information in the database (e.g. line
impedances) and quantities developed by the programs (e.g. fault currents and source impedances).
The relay setting rules are encoded using the macro language and can be modified by a user.
Settings for a 38kV stepped- distance scheme have been obtained in this way for several years [6].

Import/export facilities to communicate with a physical relay indirectly via relay vendor databases,
or to send settings to a field engineer. Most numerical relays have their own setting software and
can store relay settings in a database. The independent simulation environment complements this
by modeling together the entire network and its protective devices from multiple vendors. Settings
can then be transferred to the relay vendors database product for subsequent electronic
transmittal to the relay, making paper setting sheets unnecessary.

3. Setting Procedure
The user chooses one relay from the system database and then selects one of the setting algorithms.
These prompt for the maximum load current, the load angle, and the minimum pickup needed to avoid
operation for unbalanced load currents. Then the primary impedance reaches and pickup currents or
voltages are computed at the network level. These settings apply to any relay model. Next, secondary
tap settings are computed for the elements that exist in the chosen relay. The user can test the
settings while they are in temporary memory or can save them in the system database. The database
can hold groups of alternative settings for trial purposes and for varied network conditions.
A user can add algorithms for other relay functions, such as loss-of-potential logic, switch-onto-fault
detection and breaker failure protection.
Actual relays use some or all of the parameters in Table 1. The values shown are typical and are easily
edited. Each algorithm includes a module that specifies the tap names for the relay models in use.
Table 1 Adjustable Relay-Setting Quantities

Stepped-Distance Mho
Parameter
USE_APPARENT_IMPEDANCE
USE_APPARENT_IMPEDANCE
DEFAULT_MAX_LOAD_CURRENT
DEFAULT_LOAD_DEG
DIST_LOAD_FACTOR
ZONE_1_FRACTION
LINE_END_FRACTION
ZONE_2_FRACTION
ZONE_3_FRACTION
ZONE_3R_FRACTION
XFMR_FRACTION12
XFMR_FRACTION3

Value
FALSE
TRUE
3000
30
1.5
0.8
1.2
0.2
1.2
0.25
0.5
0.8

Description
FALSE to set zones from line ohms only
TRUE to adjust zones using fault study
Default load amps
Default load angle
Max load ohms/Max setting
Zone 1/Line ohms
Zone 2/Line ohms at line end
Zone 2 overreach/Downstream line
Zone 3 overreach/Downstream line
Zone 3 offset/Min reverse line
Zone 2 overreach into tapped XFMR
Zone 3 overreach into tapped XFMR

Value
0.2
1.5

Description
Forward pilot overreach/ Downstream line
Reverse pilot/Max reverse line

Value
0.05 I_Rated
1.3
40
0.5
0.5
1.25

Description
Min IOC pickup in relay model
Pickup/max current for remote-bus faults
Fault resistance tested at remote bus
Pickup/Current for remote-bus resistive fault
Reverse Level 3/Remote level 2 pickup
Level 2 multiples for XFMR secondary fault

Pilot Mho
Parameter
FORWARD_PILOT_FRACTION
REVERSE_PILOT_FRACTION

Definite-Time Overcurrent
Parameter
IOC_ISEC_MIN
IOC_LEVEL_1_FRACTION
IOC_LEVEL_2_FAULT_OHMS
IOC_LEVEL_2_FRACTION
IOC_LEVEL_3_FRACTION
IOC_XFMR_FRACTION

Fault Detectors
Parameter
FD_PG_ISEC_MIN
IOC_LOAD_FACTOR
ZONE_1_PICKUP_FRACTION
ZONE_1_LINE_END_FRACTION

Value
0.1 I_Rated
1.1
0.8
0.33

Description
Default min phase-ground pickup
50PP1 pickup/Max load amps
Zone 1 pickup/Min amps for reach faults
Zone 1 pickup/Min amps with line end open

Value
0.02
0.1
0.05 I_Rated

Description
Min I2/I0 relay tap
Default min I2/I0
Current pickup in relay model

Value
10
0.1 I_Rated
30

Description
Amps/Pickup for remote-bus SLG fault
Min TOC pickup in relay model
Delay at TOC_MULT

Value
0
0
20
20
75
75
0
0
20
20
75
240
0
0
0
0
0
0
0
0

Description of time delay


Zone 1 phase
Zone 1 ground
Zone 2 phase
Zone 2 ground
Zone 3 phase
Zone 3 ground
Zone 4 phase
Zone 4 ground
Forward pilot phase
Forward pilot ground
Reverse pilot phase
Reverse pilot ground
Neutral IOC level 1 ground IOC
Neg-seq IOC level 1 ground IOC
Neutral IOC level 2 ground IOC
Neg-seq IOC level 2 ground IOC
Neutral IOC level 3 ground IOC
Neg-seq IOC level 3 ground IOC
Neutral IOC level 4 ground IOC
Neg-seq IOC level 4 ground IOC

Negative-Sequence Directional
Parameter
MIN_A2
DEFAULT_DES_A2
MIN_50Q

Inverse-Time Overcurrent
Parameter
TOC_MULT
TOC_ISEC_MIN
TOC_CYCLES

Timers for Tripping


Parameter
Z1_PH_CYC
Z1_GND_CYC
Z2_PH_CYC
Z2_GND_CYC
Z3_PH_CYC
Z3_GND_CYC
Z4_PH_CYC
Z4_GND_CYC
FWP_PH_CYC
FWP_GND_CYC
RVP_PH_CYC
RVP_GND_CYC
D67N1_CYC
D67Q1_CYC
D67N2_CYC
D67Q2_CYC
D67N3_CYC
D67Q3_CYC
D67N4_CYC
D67Q4_CYC

Power-Swing Detector
Parameter
BLINDER_INNER_R_DIV_ZT
BLINDER_RATIO

Value
0.288
2.5

Description
Blinder inner resistance/System impedance
Min (outer R / inner R )

Min (outer R - inner R) (sec. ohm)


Inner reactance/Blocked-zone forward reach
Inner reactance/Blocked-zone reverse reach
Min reverse reach/line ohms
Min (outer X - inner X) (sec. ohm reactance)
Max swing cycles treated as a fault
Min swing cycles on way in for tripping
Min swing cycles on way out for tripping
Min swing cycles for blocking
Swing cycles before overriding the out-of-step
elements

2.0
1.1
1.1
0.1
0.1
0.02 * Frequency
0.02 * Frequency
0.02 * Frequency
0.05 * Frequency
0.4 * Frequency

BLINDER_DELR
OS_X_RATIO_FWD
OS_X_RATIO_REV
OS_MIN_X_REV
OS_DELX
OS_OPER_CYC
OS_WAY_IN_TRIP_CYC
OS_WAY_OUT_TRIP_CYC
OS_BLOCK_CYC
OS_OVERRIDE_CYC

4. Parallel 161kV Lines at TVA


To illustrate the setting rules, we use an actual case (Figure 1) at the Tennessee Valley Authority,
involving two coupled 161kV lines with equal lengths; line #1 is tapped at Ackerman station.
232 Sturgis
161.0 kV

109 Red Hill 161


161.0 kV

(2)

248.000

239 Louisvi SS 5
161.0 kV
147.200

21.300 No_Op
51N LFZP_TZ3

Z2GD,51N

MDAR_T2G,MDAR_TOC

24026 Red Hill U1S


20.0 kV

1151 Adaton
161.0 kV

1159 Starkvile T2
161.0 kV

94.200

LFZP_TZ3

161_BU_GND

(1)

4.200

376 Ackerman 161


161.0 kV

1348 Maben Tap 5

231 Eupora, MS 5
161.0 kV

161.0 kV

(1)

4.000
Z1G

67N2,RECEIVER

CE

No_Op
LFZP_TZ3

LFZP_TOC
3

111.400
161_BU_GND

69.0 kV

595 Ackrmn 69-1


69.0 kV

Figure 1 One-line diagram of protected region.


The impedances in primary ohms are:
Positive Sequence

Zero Sequence

Red Hills - Ackerman #1

3.77 @ 85.3 deg

9.37 @ 77.5 deg

Ackerman - Sturgis #1

3.77 @ 85.3 deg

9.37 @ 77.5 deg

Red Hills - Sturgis #2

7.57 @ 85.5 deg

19.03 @ 77.9 deg

34.5 @ 90.0 deg at


16lkV

28.2 @ 85.2 deg at


161kV

Each tapped transformer at bus 376

161.0 kV
484.500

2446 Sturgis
69.0 kV

9409 Ackrmn 69-2

229 Calhn City 5

The computed positive-sequence source-impedance ratio for this case ranges from 1.3 to 4.4 depending
on fault location; this is a medium line [1].
Here, a Permissive Overreach Transfer Trip (POTT) scheme uses phase and ground distance relay
elements with backup ground and negative-sequence overcurrent elements. Figure 1 shows the
operation for a close-in phase-A-ground fault at Sturgis. Zone 1 opens the local breaker at Sturgis; the
directional overcurrent element operates at Red Hills with the permissive signal.
The elements in a single zone are shown in Figure 2.

50ABC
32QF

21P

50PP

PHASE
DISTANCE

50L
50G
32Q

21G
GROUND
DISTANCE

50N

TRIPPING LOGIC
OR
PILOT SIGNAL

67N
32Q

INSTANTANEOUS
OVERCURRENT

50Q
67Q
32Q

INSTANTANEOUS
OVERCURRENT

51NP
51N
32Q

GROUND TIME
OVERCURRENT

Figure 2 Supervised Distance and Overcurrent Relay Elements.


At each line terminal, zone 1 elements (21P1, 21G1, 67N1 or 67Q1) trip instantaneously for internal
faults within their set points, and forward directional pilot elements (21P2, 21G2, 67N2 or 67Q2) cover
the entire line with some overreach. For internal faults, each forward pilot element will transmit a
permissive signal to the other terminal. When this signal is received, the local forward pilot elements
that have operated trip the corresponding breakers, at buses 109 and 232 in Figure 1.
This pilot scheme includes echo keying logic and is described in [1 ] as a directional comparison
hybrid scheme. A second pilot zone reaches in the reverse direction. External faults at a terminal will
suppress the transmitter functions and inhibit pilot tripping. A permissive signal received at a
terminal, e.g. Sturgis, is echoed (after a precautionary delay), unless any of the reverse elements at
Sturgis (21P3, 21G3, 67N3, or 67Q3) have detected an external fault. If the breaker at Sturgis is
already open (for maintenance or other reasons) and a close-in fault occurs there, the relay at the

opposite terminal (Red Hills) will send a permissive signal to Sturgis, and this signal will be echoed
back to Red Hills after a set delay (typically 2 cycles). Receipt of the echo signal will allow the breaker
at Red Hills to open and clear the fault.
For extra reliability, TVA is using in parallel two independent relay sets from different manufacturers.
In one manufacturers relay, the pilot zones also serve as zones 2 and 3 and provide time-delayed
backup for the adjacent lines. The other relay uses zone 1 for instantaneous tripping, uses dedicated
forward and reverse pilot zones, and uses separate zones 2 and 3 for time-delayed backup in a steppeddistance scheme.
In one relay model, the echo signal starts only when the remote permissive signal is received, the local
reverse zone 3 has not operated, and the local forward pilot element has not operated within a set time
(the echo block time delay). In the other relay, the echo signal also requires a 52b switch to assert
with the local breaker open.

5. Rules for Relay Setting


Here we explain the detailed setting rules for:

Distance zones and their fault detectors


Load-encroachment elements
Power-swing detectors
Definite-time overcurrent elements
Neutral time overcurrent element
Negative-sequence directional element

The algorithms set one element at a time and warn where setting rules conflict. They do not
automatically coordinate relays at different locations, but the user can investigate coordination
problems graphically (section 6).

5.1 MHO Distance Elements


The R-X diagram in Figure 3 shows the static mho characteristics of zone 1, the forward pilot element,
and a facing reverse pilot element that blocks tripping for external faults.

X
P.Ohms

40

30

20

10
232

109

-20

-10

R
10

20

P.Ohms

-10

Figure 3 Local Forward Mho Distance Characteristics and Remote Reverse Characteristic.
The algorithms use solid faults for initial mho settings. Then all phase zones are checked for load
encroachment. The user supplies the maximum forward and reverse load currents and worst load
angle, from separate load flow computations.
The actual memory-polarized characteristics are expanded circles and allow the zone to cover resistive
faults [4 ,7, 8]. However, for solid faults, the angle of the apparent impedance and the MTA are both
within a few degrees of the line angle. For settings based on solid faults, therefore, it is a useful
approximation that a distance element will operate whenever the apparent impedance magnitude is
less than the reach setting. Subsequently we check the operation for resistive faults using the complete
comparator equations.

5.1.1 Zone 1
Zone 1 reaches 80% along the protected line. Specifically, the total positive-sequence and zero-sequence
line impedances Z1 and Z0 are found from the database. The maximum torque angle (MTA) equals the
zone 1 line angle arg(Z1), and the set reach is 80% of the magnitude of Z1. Then the phase zone is

further limited to 66 percent of the apparent impedance at maximum forward load current and a
specified power angle, typically 30 degrees.
The zero-sequence compensation factor [9] multiplying the neutral current is set as:

k 0 = ( Z0 /Z1 -1) /3
To avoid overreach due to mutual coupling [7], the reach of the zone 1 ground distance element is also
limited to 80% of the least apparent impedance for a solid single-line-ground fault on the remote bus.
The calculation is run first with all lines in service and then with coupled lines grounded one at a time,
and with intermediate infeed removed.
The supervising phase fault detector is useful to prevent instantaneous operation on loss of potential.
It is desirable to set it above expected load current while maintaining a margin below expected fault
current to allow the distance elements to operate reliably. In relays with a separate fault detector
(50PP1) for zone 1, the recommended setting (based on experience) is the lower of:
(a) 0.8 times the least fault current for solid faults 80% along the line, with sources out one at a time
behind the relay bus, and
(b) 0.33 times the fault current for the same faults with the remote breaker open, for sequential
tripping.
To prevent zone 1 from tripping with a loss of potential under load, this pickup must also be set at least
10 percent above the maximum load current from temporary overloads and heavy loads under reduced
system voltage.
The single-phase and ground-fault detectors (50L and 50G level 1 or 2) are allowed to operate with load
currents. This is because both 50L and 50G must operate to trip a ground distance zone, and the 50G
residual elements do not operate with load except under abnormal conditions such as a breaker with
one phase open.
In relays with only one fault detector for all zones, current sensitivity for remote faults takes priority,
including those in the reverse pilot zone or zone 3. The setting may necessarily be below the maximum
load current. These relays have separate elements to block the distance elements under loss of
potential.

5.1.2 Forward Pilot Zone and Zone 2


These are set from line ohms and from zone 1 apparent impedance and are subsequently checked for
infeed and mutual coupling. Where zone 2 provides time-delayed direct tripping, it must not overreach
a downstream zone.
Depending on the status of coupled lines, infeed or mutual coupling along the protected line may limit
the zone 2 reach, especially for resistive faults. Therefore, these overreaching zones are set to cover the
larger of the actual and apparent protected-line impedances, plus a chosen portion (e.g. 20 percent) of
the shortest adjacent line. The largest apparent impedance is found for faults on a line-end bus, first
with all lines in service and then with coupled lines outaged one at a time.
The phase zone is further limited to 66 percent of the apparent positive-sequence impedance at
maximum forward load current and a specified power angle.

To avoid tripping for faults on the secondary winding of a tapped transformer, zone 2 should not
overreach the primary bus by more than a given percentage (20% to 50%) of the transformer reactance.
Where separate fault detectors 50PP2, 50L2 and 50G2 for zone 2 are available, they are set at the least
relay current (with one source removed) for remote-bus faults with fault resistance of 40 primary ohms.
The zone 2 timer is set at 20 cycles. Downstream zone 1 elements and breaker-failure protection at the
line-end bus (about 15 cycles delay) are allowed to operate first.

5.1.3 Reverse Pilot Zone


A reverse pilot zone blocks echoing of a permissive signal for external faults. Both phase and ground
elements should cover 150 percent of the largest apparent impedance calculated for line-end faults (on
the line-side of an open remote breaker) behind the relay. This setting allows the reverse zone to
operate for all external faults seen by the forward pilot zone at the other end of the protected line, as in
Figure 3. A detailed coordination check is shown in section 6.
The phase-distance element must not operate under load alone. This limit is set as 66 percent of the
apparent impedance due to maximum load current from the protected line into the relay bus, with a
specified power angle.
For the fault detectors (such as 50PP3, 50L3 and 50G3), minimum settings are allowed because: (a)
zone 3 direct tripping is delayed by 75 cycles, giving ample time for Loss of Potential Logic to assert and
block tripping of distance elements, and (b) the zone 3 distance elements are set to avoid operation
under maximum load conditions.

5.1.4 Zone 3 of Stepped Distance Scheme


A forward zone 3 should see 1.2 times the impedance to the most distant bus at a depth of 2, but the
phase element is also limited by forward load as above. Both phase and ground elements are limited to
cover 80 percent of the reactance of tapped transformers. The timer is set at 75 cycles.
Certain relay models allow zone 3 to have a reverse offset. Where this is used, it is set at 25% of the
shortest adjacent line behind the relay, to detect local bus faults.
Example of Distance Zone Settings (excerpts):
******************************************************************
Setting MHO DISTANCE elements for Permissive Overreach
******************************************************************
Substation: RED HILLS STEAM PLANT
Line:
Sturgis 161-kV line No. 1
Maximum forward load current 3000 primary amps
Highest power factor angle (deg) for forward load = 30
Worst forward load: 30.9854 Primary Ohms at 30 Degrees
836.556 MVA
724.479 MW
418.278 MVAR
Worst reverse load: 30.9854 Primary Ohms at
836.556 MVA
724.479 MW
418.278 MVAR

10

30 Degrees

************************************************************
Relay on 109 376 Ckt 1 RED HILLS STEAM PLANT - ACKERMAN 161-kV SUB 161 kV
Base kV 161
Base ohms 259.210
************************************************************
*** Zone 1 = 0.80000 * min apparent impedance for remote-bus faults
Zone 1 path
109 376 1 to 376
Total line ohms 7.54279
Phase Setting
0.80000 * 7.54279 =
6.03 Ohm
Ground Setting 0.80000 * 7.15863 =
5.73 Ohm
MTA
85.3 deg
*** Zone 2 = Longest line or apparent impedance + 0.20000 * shortest adjacent line
Zone 2 path 109 376 Ckt 1 to 232 109 Ckt 2
Path 109 376 Ckt 1 to 232 to 232 109 Ckt 2
Setting Phs
7.543 @ 85 deg + 0.20000 *

7.567
9.056
7.567
13.033

=
Setting Gnd
Phase Zone 1
Ground Zone 1
Phase Zone 2
Ground Zone 2

11.521 @
/
/
/
/

Max
Max
Max
Max

82 deg +

impedance
impedance
impedance
impedance

to
to
to
to

0.20000 *

depth
depth
depth
depth

=
80.0
75.9
120.1
172.8

1
1
1
1

%
%
%
%

@
@
@
@

85
85
85
83

deg
deg Ohm
deg
deg Ohm

(note 1)

*** Zone 3 = 1.2 * longest path to depth 2


Zone 3 path 109 376 1 to 229
Using load to limit all phase DIST zones
Worst (least) load impedance / 1.50000 = 20.6569 Primary Ohms at 30 Degrees
Desired reach to avoid load = 36.2578 Primary Ohms at 85.2691 deg
Max allowed load at
Zone Pri.Ohm Load
1
5.16
2
7.74
3
25.93
FP
7.74
RP
23.29

1.50000 * reach at given load angle


deg
Amps
MVA
MW
30.
18026.
5027.
4353.
30.
12011.
3349.
2901.
30.
3585.
1000.
866.
30.
12011.
3349.
2901.
30.
3992.
1113.
964.

Checking Transformer Overreach for ground zone 2


4 remote XFMR buses at bus 376 161 kV
Zone reach (p. ohms)
13.033 @ 85 deg
Zone reach into XFMR (p. ohms) 9.24786 @90 deg
Zone reach into XFMR
27.9 % of XFMR
*** Warning: reducing zone ohms from 13.0325 to 10.4028

New reach 20.0 % (note 1)

Phase and Ground MHO elements: desired primary ohms


Without XFMRS With XFMRS With XFMRS and LOAD
Zone 1 Phs
forward
6.03
6.03
6.03
Zone 1 Gnd
forward
5.73
5.73
5.73
Zone 2 Phs
forward
9.06
9.06
9.06
Zone 2 Gnd
forward
13.03
10.40
10.40
Zone 3 Phs
forward
47.85
30.34
30.34
Zone 3 Gnd
forward
47.85
30.34
30.34
Pilot Phs
forward
9.06
9.06
9.06
Pilot Gnd
forward
13.03
10.40
10.40
Pilot Phs
reverse
27.25
27.25
27.25
Pilot Gnd
reverse
13.03
27.25
27.25
MTA (deg) for all zones 85.3

11

MVAR
2513.
1675.
500. (note 2)
1675.
557.

Notes:
1. The ground zone 2 and the forward pilot zone are computed as 1.728 times the protected line ohms
to overcome infeed and mutual coupling, but are then reduced from 13.0325 to 10.4028 ohms to
avoid the tapped transformers at bus 376.
2. With these settings, a load up to 1000 MVA at 30 degrees will appear outside zone 3 with a margin
of 50 percent. This is the relay load limit [10], above which the element may misoperate.
Example of Fault Detector Settings (excerpts):
******************************************************************
Setting Fault Detectors for Permissive Overreach
******************************************************************
Minimum allowed fault-detector pickup = 200 Primary A
Zone 1 fault detectors: no load;
0.80000 * min current from solid faults 0.8 along line
50PP1 Phase-phase Primary A 4282.21
50L1
Single phase Primary A 2141.52
50G1
3*Izero
Primary A 2345.51
Zone 1
0.33 *
50PP1
50L1
50G1

fault detectors: no load;


min current from solid faults 0.8 along line with remote breaker open
Phase-phase Primary A 2263.65
Single phase Primary A 1131.83
3*Izero
Primary A 1048.90

Zone 1
50PP1
50L1
50G1

fault detectors: no load; maximum recommended settings:


Phase-phase Primary A 2263.65
Single phase Primary A 1131.83
3*Izero
Primary A 1048.90

Zone 2 fault detectors: no load; 40-ohm faults


50PP2 Phase-phase Primary A 827.912
50L2
Single phase Primary A 54.3842
50G2
3*Izero
Primary A 163.153
*** 50L2 below minimum 200; using minimum
*** 50G2 below minimum 200; using minimum
Zone 3
50PP3
50L3
50G3

fault detectors are set at the minimum allowed:


Phase-phase Primary A 200
Single phase Primary A 200
3*Izero
Primary A 200

Checking load current for IOC 50PP1


Maximum load current through the relay, in amps
= 3000
Maximum load current through the relay, in percent= 836.556
Maximum phase-phase load current in amps
= 5196
Maximum PP load current * 1.10000
= 5715.60
50PP1 Primary A increased to 5715.60
50PP2 and 50PP3 do not require adjustment above load

12

5.2 Quadrilateral Elements


In the TVA application, Zones 1 to 3 use polarized mho phase and ground elements; the directional
overcurrent elements are adequate to extend the protection for resistive faults.
Some relays provide optional quadrilateral (quad) ground elements. These are particularly useful to
cover resistive faults on short lines with strong sources, where the mho characteristic may not expand
sufficiently [4, 11]. The mho and quad elements have equal reaches in the line-angle direction.
Typically, the quad element resistive reaches are set at 20 primary ohms. For the quad characteristic,
the constant-reactance line is tilted automatically in the line impedance plane. The tilt eliminates
overreach or underreach for resistive faults with outward or inward load current. One type of relay
[11,12] uses negative-sequence current polarization without additional settings. Another type [9] uses
zero-sequence polarization and an extra tap setting T which is defined as the phase difference

T = arg (total fault I 0 ) - arg (local relay I 0 )


Here I0 is the zero-sequence current due to a single-line-ground fault at the zone 1 reach point. T is a
function of fault location and the network impedances and is typically between zero and - 10 degrees.
The angle T is zero for a homogeneous system (where the zero-sequence source impedance angles at the
line ends are both equal to the line angle). If T is set exactly, the reach is independent of load. If T
varies along the line, T should be set at the largest negative value from a fault study, tilting the
reactance line down to the right in the system impedance plane. Then any error in T causes
underreach rather than overreach, increasing security.

5.3 Load Encroachment


Many relays provide shaped load encroachment elements [8] or blinder elements [1,13 ] in case the load
severely restricts the reach along the line. These are set directly to exclude the expected load
impedances with a settable safety margin. In both cases, the resistive impedance from the origin to the
load encroachment element, measured along the R axis, must be inside the load impedance region. In
the following example, the least load resistance is reduced by a safety factor of 1.5 to give the blinder
setting.
Example of Load Encroachment Settings
******************************************************************
Setting Blinder and Impedance Elements for Load Restriction
******************************************************************

Substation: RED HILLS STEAM PLANT


LZOP:
Sturgis 161-kV line No. 1
Starting branch is found from element 6472 DIST BLNDR_INNER 6
Maximum forward load current 3000 primary amps
Highest power factor angle (deg) for forward load = 30
Worst forward Load: 30.9854 Primary Ohms at 30 Degrees
Forward power = Sqrt(3) * Base kV/1000* Max load current /_ power factor angle
Forward power = 0.27885 * Max load current /_ power factor angle
836.556 MVA
724.479 MW
418.278 MVAR

13

Maximum reverse load current 3000 primary amps


Highest power factor angle (deg) for reverse load = 30
Worst reverse Load: 30.9854 Primary Ohms at 30 Degrees
Reverse power = Sqrt(3) * Base kV/1000* Max load current /_ power factor angle
Reverse power = 0.27885 * Max load current /_ power factor angle
836.556 MVA
724.479 MW
418.278 MVAR
Settings for forward load
Worst (least) load impedance / 1.50000 = 20.6569 Primary Ohms at 30 Degrees
Forward load encroachment setting 20.6569
Forward load encroachment angle
30
Blinder angle to R axis 85.2691 deg
Max Blinder R along axis 17.0346 primary ohms for forward-load restriction
Settings for reverse load
Worst (least) load impedance / 1.50000 = 20.6569 Primary Ohms at 30 Degrees
Reverse load encroachment setting 20.6569
Reverse load encroachment angle
30
Blinder angle to R axis 85.2691 deg
Max Blinder R along axis 17.0346 primary ohms for reverse-load restriction

5.4 Power Swings


Double blinders with timers are used to detect power swings [1]. In case a detailed stability study is
not available, the following simple rules set the blinder positions in all relays.
A typical inner blinder setting [13 ] is 0.288 ZT, where ZT is the magnitude of the positive-sequence
system impedance: the total of line impedance and the lowest positive-sequence source impedances at
the ends. The lowest source impedance at each end of the line is found as
(-1)*(Change in positive-sequence voltage)/ (Change in positive-sequence current).
This is evaluated for sliding faults and close-in faults with up to one line out at a time behind the end
bus. The factor 0.288 is 0.5/tan(60 deg) and corresponds to a phase difference of 120 degrees or larger
between the source EMFs feeding the line, for which a power swing is assumed to be unstable.
The outer blinder and the timers distinguish stable or unstable power swings from three-phase faults.
The outer blinder is set at
Max of (inner blinder resistance + 2 ohms, inner blinder resistance * 2.5)
For a rectangular zone, the inner reactance should be 10 per cent larger than the blocked zones, and
the outer reactance is an additional 0.1 relay ohm [14].
Power-swing blinders must not operate for load alone. If the outer blinder setting exceeds 0.66 times
the resistance to the region of highest load, the user is warned that the setting requirements conflict.
If the balanced apparent impedance moves from the outer blinder to the inner blinder over a time
interval longer than the set delay, a power swing is assumed; faster impedance changes are treated as
three-phase faults. One manufacturer [14] provides one timer setting for unstable swings (allowing
tripping) and a longer timer setting for stable swings (to block selected zones). Another relay [13 ] uses
a fixed 50 ms delay. For some relays, the user may specify whether unstable power swings should trip
the relay (if system separation is required) or should block tripping in specified zones (if system
separation is desired on another line).

14

5.5 Directional Instantaneous Overcurrent (IOC) Elements


These elements (67N and 67Q) detect the direction and magnitude of residual and negative- sequence
current. They provide sensitive protection for resistive faults that are missed by the mho elements.
They contain IOC pickup settings (tap 50N with 3I0 for 67N, or tap 50Q with 3I2 for 67Q) and the
negative-sequence directional elements 32Q.
The following setting rules automatically monitor all three unbalanced fault types: single-line-ground
(SLG), line-line (LTL) and double-line-ground (DLG).

5.5.1 Level 1 IOC for Instantaneous Direct Tripping


Level 1 (High Set) elements 67N1 and 67Q1 are set with a safety margin of 1.3 times the maximum
current for a fault at the remote bus, with infeed branches outaged and with coupled lines grounded
one at a time. Monitors record the maximum zero- and negative-sequence currents at the relay over all
faults with the outage contingencies chosen.

5.5.2 Level 2 IOC for Pilot Signaling and Time-Delayed Backup


Level 2 elements (67N2 and 67Q2) are backup controls for the pilot signal or time-delayed zone 2
tripping. They are set as 0.5 times the minimum relay current for a 40-ohm ground fault at the remote
bus. The following report presents the 3I0 and 3I2 fault currents for the four types of faults. SLGR
refers to a single-line-to-ground fault with 40-ohm fault resistance (primary ohms).

Example of Overcurrent Element Settings


*******************************************************************
Ground and Neg-Seq Overcurrent Elements for Permissive Overreach
*******************************************************************
Minimum allowed IOC pickup for all levels = 100 Primary A

Level 1 IOC
Max
50N1 Primary A 4881.84
50Q1 Primary A 5582.18
Level 2 IOC:

1.30000* Max
6346.40
7256.83

40-ohm faults at remote bus 232


Fault

3I0

3I2

TPHR
LTLR
DLGR
SLGR

0.0
0.0
403.4
405.5

0.0
1275.9
464.6
458.3

Level 2 IOC at RED HILLS STEAM PLANT to ACKERMAN 161-kV SUB


Desired setting = 0.5 * min relay amps for 40-ohm faults at remote bus 232
Level 2 IOC
Min
0.50000 * Min
50N2 Primary A
403.387
201.694
50Q2 Primary A
458.263
229.132

15

5.5.3 Level 2 IOC Settings with Tapped Transformers


If the protected line has tapped transformers, level 2 must be increased to 1.25 times the highest
current due to faults on the transformer secondary. The fault study applies three-phase, phase-phase
and phase-ground faults on the secondary side; one does not need to know the transformer connection.
Example of limits due to transformers:
4 tapped transformers at bus 376; adjusting Level 2
Faults on tapped XFMRs; remote breaker closed
Faults on tapped XFMRs; remote breaker open at 232
Solid faults on tapped XFMR secondaries
Level 2 IOC
MAX
1.25000 * MAX
50N2 Primary A
1454.83
1818.53
50Q2 Primary A
2976.65
3720.81

The following table shows how the tapped transformers limit the sensitivity allowed:
Level 2 IOC elements (Primary A)
Without XFMRS
201.69
229.13

50N2
50Q2

With XFMRS
1818.53
3720.81

5.5.4 Level 3 IOC Settings


The local level-3 reverse IOC settings (taps 50N3 and 50Q3) should have at most half of the remote
level-2 values, to block echo signals for external faults (Figure 4). These settings are secure, since zone
3 sees at least as much line current as its remote level 2 element.

1
6

Protected Line
Local Relay

IOC 67N3

2
Remote Relay
IOC 67N2

F = Reach point of remote level-2


overcurrent element
Level 3 overcurrent element 67N3 must
see all reverse faults at points F.

Figure 4 Coordinating definite-time overcurrent elements.

16

5.6 Backup Directional Ground Time-Overcurrent (TOC) Element


Element 51N provides current-dependent time-delayed clearance of high-resistance faults along the
protected line. It provides backup protection for remote elements, and supplements zone 2 when the
pilot scheme is out of service. A suitable time delay is 30 cycles for a solid fault at the remote bus and a
pickup setting of 0.1 times the relay current for this fault. The curve shape is chosen separately for
coordination with neighboring TOC elements. A Very Inverse curve is typical for a looped system
without fuses.
Figure 5 shows the operating times for sliding 40-ohm faults along the protected line from Red Hills to
Sturgis.

Figure 5 Time-overcurrent characteristic with sliding 40-ohm faults. This element trips in 30 cycles
for a close-in 40-ohm fault and 300 cycles for a 40-ohm fault on the remote bus.

5.7 Current Direction


A fault on a coupled external line can reverse the zero-sequence voltage and make the relays at both
ends of the line see the fault as forward [1 , 15]. In other cases, the current direction may reverse. The
setting algorithm reports the direction seen by the relay for:
(a) Sliding faults on the relay line with all coupled lines in service,
(b) Sliding faults on the relay line with coupled lines grounded one at a time, and

17

(c) Sliding faults on each coupled line with all lines in service.
Generic zero-sequence and negative-sequence directional elements are used to measure the direction as
in reference [15]. The user is shown where the current may reverse and is warned where an external
fault appears as internal.
The following example verifies that internal faults remain as forward (F) at both ends of the line when
a coupled line is grounded. The phase differences between the voltages (V) and currents (I) at the ends
of the line are labeled Arg(Right/Left).
Direction at end buses 109 and 232 for single-line-ground faults on line 109 376 1 to 232
Fault: SLG
Relay branch 109 376 1
Fault branch 109 376 1 to 232 Internal
0 seq
-seq
0 seq
-seq
Fault direction
Fault direction
Arg(Right/Left)
Dist Left end Right end
Left end Right end
V
I
V
I
1 0.25
F
F
F
F
-3.8
2.8
-0.7
3.4
2 0.50
F
F
F
F
-7.5
0.9
-0.7
3.1
3 0.75
F
F
F
F
-10.0
-0.3
-0.7
3.4

Fault: SLG
Line Grounding 109 Red Hill 161 to 232 Sturgis
Relay branch 109 376 1
Fault branch 109 376 1 to 232 Internal
0 seq
-seq
Fault direction
Fault direction
Dist Left end Right end
Left end Right end
1 0.25
F
F
F
F
2 0.50
F
F
F
F
3 0.75
F
F
F
F

5 Ckt 2

0 seq
-seq
Arg(Right/Left)
V
I
V
I
-3.8
6.1
-1.2
5.7
-4.7
5.1
-1.3
5.7
-4.9
5.0
-1.2
5.7

The following example checks the direction measured for external faults on a coupled branch (109 232 2
to 232). At least one end must see the external fault as reverse (R); otherwise a warning is shown.
Normal current reversal occurs when the fault location changes on the parallel line.
External faults on coupled branches; all lines in service
Fault: SLG
Relay branch 109 376 1
Fault branch 109 232 2 to 232 External
0 seq
-seq
0 seq
-seq
Fault direction
Fault direction
Arg(Right/Left)
Dist Left end Right end
Left end Right end
V
I
V
I
1 0.25
R
R
R
F
-3.8 -15.6
-0.7 180.0
2 0.50
F
R
R
F
-7.5 172.9
-0.7 180.0
3 0.75
F
R
F
R
-10.0 178.3
-0.7 -180.0

18

5.8 Negative-Sequence Directional Element 32Q


This element supervises all zones for the ground distance elements (forward or reverse) and also
supervises the phase mho elements except when a three-phase fault is detected. One type of relay [15]
has a fixed pickup for the torque product

Torque = Re ( V2 conjg (I 2 ) exp(-jMTA) )


from relay negative-sequence voltage V2 and current I2. When the product is negative (for a forward
fault), the relay bit 32QF is asserted. For a reverse fault, the relay bit 32QR is asserted.
Another type [8] increases the torque limit for high fault current by measuring an impedance
component instead:

Z2 = Re ( ( V2 /I 2 ) exp ( -jMTA ) )
The angle MTA is a tap setting, usually set equal to the +/- sequence line angle arg(ZL1). Z2 changes
abruptly from a negative value for close-in forward faults to a positive value for reverse faults.
Directionality is assured by computing the least negative Z2 for forward faults, and the least positive Z2
for reverse faults. Then the forward and reverse impedance pickups Z2F and Z2R are set between
these limits, with

Z2R Z2F + 0.5 / (rated current)


in secondary ohms.
The 50QF and 50QR pickups (3I2) must allow the most sensitive supervised IOC, TOC or distance
element to operate [16]. The required minimum 3I2 values are computed from line-line, single-line, and
double-line faults at the operating limits of the supervised elements.
The tap setting (a2) equals the least allowed magnitude of (I2/I1) for operation, to avoid operation due to
untransposed lines with load current. If an unbalance factor a2 is supplied, the algorithm must warn
the user where the desired 50QF or 50QR are less than a2 * (max load current). Alternatively, 50QF
and 50QR are set from the fault studies and the unbalance factor is set as:

a2 = min (50QF, 50QR) / (max load current)


This value is chosen because any larger a2 value would cause the maximum load current to raise the
threshold 3I2 above the required 50QF setting.
Example:
************************************************************************
Setting Negative-Sequence Directional Element for Permissive Overreach
************************************************************************
Substation RED HILLS STEAM PLANT
Maximum load current 3000 primary amps

19

Branch impedance (pu)


0.015 @ 85 deg
Total line ohms (primary) 7.54279
CTR 400 VTR 1399.89
Primary (network) quantities for forward faults
|Zsourcen| Re(V2/I2/_-MTA)
|I2| Contingency
12.4908
-12.485
3106.60 All lines in service
13.4036
-13.398
2565.09 Midline fault at 0.25000
14.5186
-14.512
2166.46 Midline fault at 0.50000
15.9117
-15.904
1857.16 Midline fault at 0.75000
17.6934
-17.683
1607.01 Midline fault at 1
18.3737
-18.336
2140.20 Line out 109 232 2
38.8598
-38.823
750.487 Line out 109 24026 1
Negative V2/I2 component of least magnitude for forward faults -12.485 primary ohms
(-3.567 secondary ohms).
Primary (network) quantities for reverse faults
|Zsourcen| Re(V2/I2/_-MTA)
|I2| Contingency
38.7379
38.6949
1001.71 All lines in
23.1619
23.1450
1697.76 Line out 109
38.7379
38.6949
752.850 Line out 109
Positive V2/I2 component of least magnitude for
(6.613 secondary ohms).

service
232 2
24026 1
reverse faults 23.1450 primary ohms

Default settings
Rated amps
5
Total line ohms
7.54279
Total secondary ohms 2.15524
Default Z2F = 0.5 * total secondary ohms= 1.07762
Default Z2R = Z2F + 0.5/(rated amps)
= 1.17762
Default values lie within required range (-3.567, 6.613) secondary ohms
Primary 3*I2 (A)
TOC_FACTOR chosen as
Primary pickup (3*I2)
CTR
Secondary pickup (3*I2)

2133.70
0.10000
213.370
400
0.53343 relay A

Settings chosen with fixed a2 and max load 3000 A


a2 (min I2/I1)
0.10000
50QF & 50QR 3I2 pickup
2.25000
Alternative settings for IOC pickup and max load 3000 A
a2 (min I2/I1)
0.02371
50QF & 50QR 3I2 pickup
0.53343

20

5.9 Summary of Setting Rules


Table 2 is a summary of the setting rules.
Table 2 Summary of Rules for Distance and Overcurrent Settings

Phase DIST zone 1

80% along the protected line; all phase zones are checked for max
load.

Ground DIST zone 1

Minimum of 80% along the protected line and 80% of apparent


impedance due to mutual coupling for a remote A-G bus fault, with
infeed removed and a coupled branch grounded.

Fault detector for zone 1

0.8 times the least fault current for solid faults 80% along the line.
0.33 times the fault current for the same faults with the remote breaker
open.
Phase pickup 10% above maximum load current.

Forward pilot zone

Maximum of (line ohms and largest apparent impedance for remotebus fault) + 20% of the shortest downstream line.
Phase element limited to 66% of the apparent impedance at maximum
load current and a power angle of 30 degrees.

Time-delayed zone 2

Maximum of (line ohms and largest apparent impedance for remotebus fault) + 20% of the shortest downstream line.
Phase element limited to 66% of the apparent impedance at maximum
load current and a power angle of 30 degrees.
Must not overreach a downstream zone 1.
Must not overreach the primary bus of a tapped XFMR by more than
20 to 50% of the transformer reactance.

Fault detector for zone 2

Least relay current (with one source removed) for remote-bus faults
with fault resistance 40 primary ohms.

Zone 2 timer

20 cycles.

Reverse pilot zone

150% of the largest apparent impedance calculated for line-end faults


(on the line-side of an open remote breaker) behind the relay.
Phase element limited to 66% of the apparent impedance due to
maximum load current from the protected line into the relay bus, with a
power angle of 30 degrees.

Fault detector for reverse


pilot zone

Minimum setting.

21

Time-delayed forward
zone 3

1.2 times the impedance to the most distant depth-2 bus.


Phase element limited to 66% of the apparent impedance at maximum
load current and a power angle of 30 degrees.
Must not overreach the primary bus of a tapped XFMR by more than
80% of the transformer reactance.

Zone 3 timer

75 cycles.

Fault detector for zone 3

Minimum setting.

Zone 3 reverse offset

25% of zone 1.

Quad element resistive


reach

20 primary ohms.

Load encroachment
blocking

Resistive reach of 0.66 * apparent resistance at maximum load


current.

Power swing detector:


Inner blinder

Resistance of 0.5/tan(60) * Abs (protected line + total source


impedance).

Outer blinder

Max (Inner blinder * 2.5, Inner blinder + 2 secondary ohms).

Reactance reach

1.1 * largest controlled zone; outer reach 0.1 secondary ohms larger.

Level 1 IOC

1.3 times the maximum current for a fault at the remote bus with infeed
branches outaged and one coupled line grounded.

Level 2 IOC

0.5 times the relay current for a 40-ohm ground fault at the remote
bus.
At most 1.25 times the highest current due to faults on the transformer
secondary.

Level 3 IOC

Half of the remote level-2 pickup.

Ground TOC with Very


Inverse curve

30 cycles delay for a solid fault at the remote bus and a pickup setting
of 0.1 times the relay current for this fault.

Negative-Sequence
Directional Element

Z2F = 0.5 * (secondary line ohms); Z2R = Z2F + 0.5/(rated current).


Forward and reverse pickups (3I2) allow the most sensitive supervised
element to operate.
Min (Ineg/Ipos) tap = pickup / (max load current).

22

6. Checking the Settings


Settings must allow for fault resistance, imprecise network data, and instrument-transformer error.
The operating margin defined below shows how close an operating relay is to its limit, or how close a
non-operating relay is to an incorrect operation. Where the calculations involve only a single element,
the fault studies are performed by the setting algorithm to warn the user of setting conflicts.
Overcurrent elements use the multiples of pickup to measure the safety margin, and the pickup taps
are computed directly from the desired margin in the fault studies above.
Distance elements are set directly from the line impedances and need additional fault studies
(performed by the setting algorithm) to test for infeed and mutual coupling.

6.1 Reach Margin for Distance Elements


Here we define the reach margin in the positive-sequence line-impedance plane (Figure 6) for a solid
fault on one of the protected lines as

Margin = Set reach / (Set reach + distance from boundary)

[Fault beyond characteristic]

Margin = Set reach / (Set reach - distance from boundary)

[Fault within characteristic]

using the shortest distance from the apparent impedance point to the boundary of the operating region.

F1
A

F2

Mho circle

MTA
R
Margin =

Fault beyond
Diameter
Diameter + dist A to F1 characteristic

Margin =

Fault within
Diameter
Diameter dist A to F2 characteristic

Figure 6 Operating margins of mho distance characteristic in line impedance plane.

23

The set reach depends on the tap settings. For a quadrilateral, the reach is measured from the origin
to the reactance line. For a mho circle (Figure 6), the reach is the diameter. By design, this circle
passes through the set reach point at the MTA. Although the actual characteristic expands away from
this angle, solid faults on the protected lines have apparent impedance angles within a few degrees of
the MTA setting, so useful margin estimates for solid faults can be based on a fixed set reach. This
simplification makes the margin calculation independent of the particular relay comparator.
The phase elements are tested for three-phase faults and the apparent impedance is computed as:

Impedance = ( Vb -Vc ) / ( I b -I c )
where the relay phase voltages relative to local ground are (Va, Vb, Vc) and the line currents at the relay
are (Ia, Ib, Ic).
For the ground elements, the apparent impedance is computed. For phase-A-ground element, the
apparent impedance is:

Impedance = Va / ( I a + 3I 0 k 0 )
where the zero-sequence compensation factor k0 is approximated by the relay tap settings.
These impedances equal the apparent positive-sequence ohms between the relay and its local zerovoltage point and therefore give the apparent fault location on the line.
Applications of the reach margins are described in the following sections.

6.1.1 Zone 1 and 2 on Protected Line


With solid faults at the end bus of the protected line, the zone 1 reach margin should be 0.8 or less. For
the forward pilot zone or zone 2, the reach margin should exceed 1.2 for the same faults.
This check is part of the setting algorithm. The report warns that the forward pilot zone and zone 2 set
as above both underreach for ground faults:
Margins for zone 1 phase (underreaching)
Line
109 232

Fault
TPH

Reach
6.03

App P.Ohm
7.54

MHO margin
0.80 Underreach

Margins for zone 1 ground (underreaching)


Line
109 232

Fault
SLG

Reach
5.73

App P.Ohm
11.52

MHO margin
0.50 Underreach

Margins for zone 2 (overreaching)


Line
109 232
109 232

1
1

Fault
TPH
SLG

Reach
9.06
10.40

App P.Ohm
7.54
11.52

24

MHO margin
1.20 Overreach
0.90 Underreach

** Warning **

Margins for forward pilot zone (overreaching)


Line
109 232
109 232

1
1

Fault
TPH
SLG

Reach
9.06
10.40

App P.Ohm
7.54
11.52

MHO margin
1.20 Overreach
0.90 Underreach

** Warning **

Here the automatic procedure cannot set an optimum value, and the engineer must choose compromise
settings for forward pilot operation and for time-delayed tripping. Zone 1 (phase or ground) is critical
and should already have been set to prevent overreach. To increase zone 2 to cover the line, or to relax
the load restrictions or the tapped transformer limit, the user will change the setting factors shown in
Table 1 and repeat the calculation.

6.1.2 Zone 2 Coverage of Downstream Line


For zone 2 stepped distance elements, the algorithm applies a fault at each of the downstream zone 1
limits, with infeed branches removed to maximize the zone 2 reach. For a practical approximation, the
downstream zone 1 is assumed to reach 0.8 times the least apparent impedance for a solid single-lineground fault at the end of its line. This is the same zone-1 rule as above. The parallel line in this
network is treated as any other downstream line.
If the zone 2 reach margin exceeds 0.8, the fault is close to the downstream zone 1 characteristic, so
there is a risk that the local zone 2 will misoperate for faults beyond zone 1 of the downstream relay.
The coordination can be maintained by increasing the local zone 2 time delay [1].
Margins for GROUND zone 2 overreach in downstream zone 1
Overreaching zone of relay on 109 376 1 must not reach ends of downstream zone 1
Faults at zone 1 reach point on lines from remote bus 232
All infeed branches in service
Remote line
0.75
0.79
0.80
0.79
0.79

along
along
along
along
along

232
232
232
232
232

109 2 to 109
239 1 to 239
1151 1 to 36
1348 1 to 229
1348 1 to 590

Fault

Reach

SLG
SLG
SLG
SLG
SLG

13.03
13.03
13.03
13.03
13.03

Fault

Reach

SLG
SLG
SLG
SLG
SLG

13.03
13.03
13.03
13.03
13.03

App P.Ohm
15.251
69.296
89.135
142.599
83.623

@-105
@ 82
@ 81
@ 82
@ 84

MHO margin
deg
deg
deg
deg
deg

Rev Flt
0.19
0.15
0.09
0.16

Underreach
Underreach
Underreach
Underreach
Underreach

OK
OK
OK
OK
OK

All infeed branches outaged


Remote line
0.75
0.79
0.80
0.79
0.79

along
along
along
along
along

232
232
232
232
232

109 2 to 109
239 1 to 239
1151 1 to 36
1348 1 to 229
1348 1 to 590

App P.Ohm
13.700
20.426
29.017
36.024
22.622

@
@
@
@
@

85
83
83
83
83

MHO margin
deg
deg
deg
deg
deg

0.95
0.64
0.45
0.36
0.58

Underreach
Underreach
Underreach
Underreach
Underreach

*Warning*
OK
OK
OK
OK

All infeed branches outaged; end breaker open


Remote line
0.75
0.79
0.80
0.79
0.79

along
along
along
along
along

232
232
232
232
232

109 2 to 109
239 1 to 239
1151 1 to 36
1348 1 to 229
1348 1 to 590

Fault

Reach

SLG
SLG
SLG
SLG
SLG

13.03
13.03
13.03
13.03
13.03

App P.Ohm
10.269
20.429
28.844
36.039
23.142

25

@
@
@
@
@

89
83
83
83
84

MHO margin
deg
deg
deg
deg
deg

1.26
0.64
0.45
0.36
0.56

Overreach **Miscoord.**
Underreach
OK
Underreach
OK
Underreach
OK
Underreach
OK

Here we have allowed zone 2 to cover up to 50 percent of tapped-transformer reactance. In this


application, zone 2 is short enough to coordinate with every downstream zone 1 when all lines are in
service, but it shows a miscoordination with zone 1 along the parallel line (232-109 circuit 2) when
intermediate sources are removed and the breaker is open on the parallel line at the relay bus 109.
This is acceptable when zone 2 is used for a pilot scheme.

6.1.3 Zone 3 Coverage of Depth-2 Buses


For a forward zone 3, the setting algorithm applies a fault at each bus at a depth of 2, with all branches
in use and with branches out one at a time behind the relay, to check minimum source conditions.
Warnings appear in the following example because zone 3 has been limited by tapped transformers and
cannot reach the depth-2 buses.
Margins for zone 3 at depth 2: all lines in service
Line

109
109
109
109
109
109
109
109

376
376
376
376
376
376
376
376

1
1
1
1
1
1
1
1

to
to
to
to
to
to
to
to

239
36
229
590
239
36
229
590

Fault

Reach

TPH
TPH
TPH
TPH
SLG
SLG
SLG
SLG

30.34
30.34
30.34
30.34
30.34
30.34
30.34
30.34

App P.Ohm

69.58
54.37
120.92
88.00
85.82
84.83
169.84
105.63

MHO margin

0.44
0.56
0.25
0.34
0.35
0.36
0.18
0.29

Underreach
Underreach
Underreach
Underreach
Underreach
Underreach
Underreach
Underreach

**
**
**
**
**
**
**
**

Warning
Warning
Warning
Warning
Warning
Warning
Warning
Warning

**
**
**
**
**
**
**
**

6.2 Sensitivity to Ground-Fault Resistance


A plot of the largest detectable fault resistance against fault location is a convenient measure of ground
element sensitivity [16]. We fix the fault type, open chosen breakers, and compute the largest fault
resistance that will trip the element at each location, again using the actual relay comparator
equations.
Figure 7 plots the maximum fault resistance that the distance and directional overcurrent elements
can detect for single-line-ground (1Ph) faults in circuit 1.

26

Gnd IOC (1Ph)

NSeq IOC (1Ph)

Max R

Gnd DIST (1Ph)

Open at ST

Open at ST

Distance/Line

Distance/Line

Max R

Open at ST

Distance/Line

Figure 7 Threshold fault resistance (primary ohms) for zone 1 (solid lower curves) and zone 2
(dashed upper). Line-end breaker open for lower plots.
Red Hills (RH) substation is at the left (Distance/Line = 0) and Sturgis (ST) is at the right. The largest
resistance seen by zones 1 and 2 at Red Hills decreases with increasing distance along the line. The
region below the zone 1 curves shows the faults in zone 1 that trip directly; Zone 1 reaches no more
than 80 percent along the line as required. The region under both zone 2 curves shows the higher fault
resistances detectable by the pilot scheme. This region must cover the entire line. The NSeq IOC
elements have been made less sensitive than the Gnd IOC elements, to avoid the tapped
transformers; hence the zone 1 NSeq IOC element at Sturgis does not operate for 1Ph faults.
A 15-ohm close-in fault at Sturgis is not seen from Red Hills because of neutral infeed from the two
tapped autotransformers at Ackerman. Such faults will be cleared with zone-2 time delay (20 cycles),
by the TOC element, or sequentially if the current rises enough after the breaker opens at Sturgis. The
lower plots in Figure 7 (open at ST) show the increased coverage in this case. The Gnd IOC curve
flattens out for faults beyond the transformers; this is a result of the neutral infeed, which does not
affect the NSeq IOC element.
We exploit the automatic process by including more thorough fault studies than would be practicable
manually. For example, to set the largest allowed fault detector pickup requires about 50 fault
calculations, and setting the overcurrent elements uses about 85 fault calculations, with various lines
temporarily outaged. Figure 7 required over 6000 fault calculations (under 15 minutes on a 450 Mhz

PC).

27

6.3 Reverse Pilot Reach


The reverse pilot zone must cover the overreach region of the forward pilot zone at the opposite end of
the line, in order to block all echoed signals. This coordination check is made graphically from the
settings of both relays. For the worst case (maximum overreach of line 109-376-232), the tapped
transformers at bus 376 are removed.
Figure 8 shows the maximum resistance seen by the forward zones (upper half of diagram) and the
reverse pilot zone of the facing relay (lower half), for single-line-ground faults around the loop of
parallel lines. Zone 1 sees only part of its protected line, as required. The overreach area of the forward
pilot zone is covered by the reverse pilot zone plotted below it.

28

Figure 8 Threshold fault resistance (primary ohms) for SLG faults around loop of parallel lines.
The upper curves show two forward zones on Red Hills Sturgis line 1. The lower curve shows the
reverse pilot element of the facing relay on Sturgis-Red Hills line 1. The reverse element covers the
forward pilot overreach region right of Sturgis.
Coverage of fault impedance is also shown in the circles in Figure 9. These circles are similar in
appearance to traditional mho circles but are not to be confused with them. Each plots the largest fault
impedance seen for a single-line-ground fault at bus 232, using the full relay operating equations. This
impedance may be inductive, resistive, or capacitive: the algorithm searches for the limit at each faultimpedance angle in turn. The limit depends on the equivalent source impedance at the relay, and
hence also varies with fault location. The forward pilot limit is well within the facing reverse
characteristic, which must enclose all external faults that the forward pilot sees.

29

X
P.Ohms

30

20

10
232

12 Ohms

109

-20

-10

R
10

20

P.Ohms

-10

Figure 9 Limits of fault impedance for single-line-ground faults at remote bus 232: for forward pilot
relay at 109 facing the reverse pilot relay at 232. The forward element (smaller circle) barely
operates for a solid fault at bus 232. The reverse element (larger circle) will operate for a 12-ohm
resistive bus fault at bus 232. Both elements see capacitive fault impedance of about j10 ohms at
bus 232.

30

Table 3 summarizes the extra checking rules. Again, these apply to any pilot scheme using phasor
distance relays. If these tests produce warnings, the user must use judgment in finding a compromise
setting.
Table 3
Summary of Extra Checking Rules

Distance zone 1

Reach margin < 0.8 for line-end solid faults on the


protected line or on a coupled line.

Forward pilot zone and time-delayed


distance zone 2

Reach margin > 1.2 for line-end solid faults.


Reach margin < 0.8 for faults at downstream zone 1 limits
with infeed branches outaged.

Time-delayed forward distance zone 3

Reach margin > 1.2 for solid faults at a depth of 2 buses.

Reverse distance pilot zone

Cover the overreach region of the forward pilot zone at the


opposite end of the line. Plot the fault impedance limits in
the line-impedance plane.

Distance and overcurrent zones 1


and 2

Plot the largest detectable fault resistance for sliding faults


on the protected line.

Directional overcurrent

Confirm that internal faults on the protected line are


measured as forward at both ends, with coupled branches
outaged or grounded.
Confirm that external faults on coupled lines are measured
as reverse at one or both ends of the protected line.

7. System Simulation
After setting the relay, the engineer can test its operation in the network. The stepped event
simulation uses detailed phasor models of the relays, including the TOC curves and distance- element
comparators, and accounts for the logic of multiple relays in the scheme. It verifies that the primary
protection can successfully clear faults on the protected line, and that other relays will not operate
unintentionally.
Figure 1 shows a close-in solid single-line fault at Sturgis for which the zone 2 ground element at Red
Hills does not operate. The 67N2 pilot element at Red Hills trips on receiving the permissive signal.
The following events are reproduced:
Event

Cycles from start

Close-in fault at Sturgis


Zones 1 and 2 assert at Sturgis
161 kV breaker starts to open at Sturgis

0.0
1.0
1.0

31

Transmission to Red Hills


Signal received at Red Hills
67N2 asserts at Red Hills

1.0
1.2
1.2 (includes 0.2 cycles for
torque control element)
1.2
4.0
4.2

161 kV breaker starts to open at Red Hills


Breaker opens at Sturgis
Breaker opens at Red Hills

The total time has about 0.5 cycle of random error, since the prefault voltage angle at the instant of the
fault is unknown in a phasor model.
Figure 10 simulates a single-line fault with fault resistance of 6 ohms.
232 Sturgis
161.0 kV

109 Red Hill 161


161.0 kV

(2)

248.000

239 Louisvi SS 5
161.0 kV

22.500 No_Op

190.800

51N LFZP_TZ3

Z3GD,51N

MDAR_TOC

24026 Red Hill U1S


20.0 kV

1151 Adaton
161.0 kV

1159 Starkvile T2
161.0 kV

No_Op
LFZP_TZ3

161_BU_GND
376 Ackerman 161
E 10.400

161.0 kV

(1)

1348 Maben Tap 5

231 Eupora, MS 5
161.0 kV

161.0 kV

(1)

4.000
Z1G

Z2G,ECHO_RECEIVER,6

CE

No_Op

229 Calhn City 5


161.0 kV
651.600

Not tripped

LFZP_TOC
2446 Sturgis

69.0 kV
118.900
161_BU_GND

9409 Ackrmn 69-2

595 Ackrmn 69-1

69.0 kV

69.0 kV

Figure 10 Delayed tripping at Red Hills for 6-ohm ground fault. This fault is outside the reach of
zone 2 at Red Hills until the breaker opens at Sturgis.
It has already been shown that the zone 2 mho and IOC elements may both underreach in this case, so
the pilot zone at Red Hills sees the fault only after the breaker has opened at Sturgis. The following
events are reproduced:
Event

Cycles from start

Close-in fault at Sturgis


Zone 1 tripping at Sturgis
161 kV breaker opens at Sturgis
Zone 2 sequential tripping at Red Hills
Transmission to Sturgis
Signal received at Sturgis
Echo transmission to Red Hills
Echo received at Red Hills
161 kV breaker opens at Red Hills

0.0
1.0
4.0
5.0
5.0
5.2
7.2
7.4
10.4

In a systematic search for miscoordinations, the engineer can run many different faults. A typical
series of stepped-event simulations varies the fault resistance for close-in and midline single-lineground faults on the protected line:

32

Protected line: local branch 232 376 circuit 1 to remote bus 109
Fault type

Location

Single-line-ground (SLG)
Single-line-ground
Single-line-ground

Close-in at 232
Close-in at 109
0.5 from 232 376 1 to 109

SLG - 2 Ohms
SLG - 2 Ohms
SLG - 2 Ohms

Close-in at 232
Close-in at 109
0.5 from 232 376 1 to 109

SLG - 6 Ohms
SLG - 6 Ohms
SLG - 6 Ohms

Close-in at 232
Close-in at 109
0.5 from 232 376 1 to 109

SLG - 10 Ohms
SLG - 10 Ohms
SLG - 10 Ohms

Close-in at 232
Close-in at 109
0.5 from 232 376 1 to 109

SLG - 20 Ohms
SLG - 20 Ohms
SLG - 20 Ohms

Close-in at 232
Close-in at 109
0.5 from 232 376 1 to 109

SLG - 40 Ohms
SLG - 40 Ohms
SLG - 40 Ohms

Close-in at 232
Close-in at 109
0.5 from 232 376 1 to 109

The simulation evaluates the time delay between the fastest primary and fastest backup local zones of
protection, reports any miscoordinations or time intervals below the chosen minimum, and continues
with the next fault. On line 2 it is found that faults up to 60 ohms are cleared in 4.4 cycles when the
breakers open simultaneously at both ends. However, on line 1 the instantaneous overcurrent settings
have been limited by the tapped transformers, and the result is that tripping for 20-ohm faults on line 1
requires 25 cycles using the time-overcurrent elements.

8. Summary
Automated setting of complex modern relays improves productivity by applying utility rules
consistently, simplifying routine data-handling and avoiding human error. Fast computers can run
more thorough fault studies than those previously conducted manually. The algorithms presented here
calculate the electrical settings for multiple overcurrent and distance elements and report the margins
of secure operation. They apply to any directional comparison pilot scheme and to any relay model.
The settings are verified graphically and in a system simulation that includes the pilot logic.

References
[1] IEEE Guide for Protective Relay Applications to Transmission Lines, IEEE Standard C37.1131999, Institute of Electrical and Electronics Engineers, New York, NY, February 2000.

33

[2] Fernando L. Alvarado, Sao Khai Mong, and Mark K. Enns, A Fault Program with Macros,
Monitors, and Direct Compensation in Mutual Groups, IEEE Transactions on Power Apparatus and
Systems, vol. PAS-104, No. 5, pp. 1109-1120; May 1985.
[3]. Paul F. McGuire, Donald M. MacGregor, John J. Quada, and Daryl B. Coleman, A Stepped-Event
Technique for Simulating Protection System Response, presented at 6th Technical Seminar on
Protection and Control, Natal, Brazil; September 27 - October 2, 1998.
[4] A. T. Giuliante, S. P. Turner, and J. E. McConnell, Considerations for the Design and Application
of Ground Distance Relays, 22nd Annual Western Protective Relay Conference, Spokane, Washington;
October 1995.
[5] Mark K. Enns and Paul F. McGuire, Data Base Organization for Protection Engineering, CIGRE
Study Committee 34 Colloquium, Johannesburg, South Africa, October 1-3, 1997.
[6]. Donald M. MacGregor and Hugh Borland, Computer-Aided Setting and Coordination of Distance
Relays in 38 kV Distribution Networks, 13th International Conference on Electricity Distribution
(CIRED 1995), Brussels, Belgium; May 1995.
[7]. George E. Alexander and Joe G. Andrichak, Ground Distance Relaying: Problems and Principles,
47th Annual Georgia Tech Protective Relaying Conference, Atlanta, Georgia; April 28-30, 1993.
Protective Relay Conference, Spokane, Washington; October 24-26, 2000.
[8]. E. O. Schweitzer III and Jeff Roberts, Distance Relay Element Design, 46th Annual Conference
for Protective Relay Engineers, Texas A&M University, College Station, Texas; April 12-14, 1993.
[9] S. E. Zocholl, Three-Phase Circuit Analysis and the Mysterious k0 Factor, 22nd Annual Western
Protective Relay Conference, Spokane, Washington; October 1995.
[10] Transmission Line Protective Systems Loadability, report by the IEEE Power System Relaying
Committee Working Group D6, presented at 28th Annual Western Protective Relay Conference,
Spokane, Washington; October 23-35, 2001.
[11] Solveig Ward, Comparison of Quadrilateral and Mho Distance Characteristic, 26th Annual
Western Protective Relay Conference, Spokane, Washington; October 1999.
[12] Walter A. Elmore, Fernando Calero and Lifeng Yang, Evolution of Distance Relaying Principles,
48th Annual Conference for Protective Relay Engineers, Texas A&M University, College Station,
Texas; April 3-5, 1995.
[13 ] REL 512 Line Protection and Breaker Control Terminal, manual I.L.40-512, ABB Power
Automation and Protection Division, Coral Springs, FL; July 2001.
[14] Daqing Hou, Shaojun Chen and Steve Turner, SEL-321-5 Relay Out-of-Step Logic, Application
Guide AG97-13, Schweitzer Engineering Laboratories, Inc.; 1997.
[15] Walter A. Elmore and Elmo Price, Polarization Fundamentals, 27th Annual Western Protective
Relay Conference, Spokane, Washington; October 24-26, 2000.
[16] Jeff Roberts, E. O. Schweitzer III, Renu Arora, and Ernie Poggi, Limits to the Sensitivity of
Ground Directional & Distance Protection, 22nd Annual Western Protective Relay Conference,
Spokane, Washington; October 24-26, 1995.

34

A.T. Giuliante is president and founder of ATG Exodus. Prior to forming his company in 1995, Tony
was Executive Vice President of GEC ALSTHOM T&D Inc.- Protection and Control Division, which he
started in 1983. From 1967 to 1983, he was employed by General Electric and ASEA. In 1994, Tony
was elected a Fellow of IEEE for contributions to protective relaying education and their analysis in
operational environments. He has authored over 40 technical papers and is a frequent lecturer on all
aspects of protective relaying, including electromechanical, solid state and digital based equipment.
Tony is a past Chairman of the IEEE Power System Relaying Committee 1993-1994, and past
Chairman of the Relay Practices Subcommittee. He has degrees of BSEE and MSEE from Drexel
University 1967 and 1969.
Donald M. MacGregor is a Lead Engineer at Electrocon International, Inc. He received his B.A. degree
with Honors in mathematics in 1970, from St. Catharines College, Cambridge, England. He next
attended University College of North Wales in Bangor, where he earned his Ph.D. in Electronic
Engineering in 1973. He joined Electrocon in 1973 and has made significant contributions to software
for fault analysis, the modeling of power transformers, and power system protection, including detailed
models of multifunction relays.
Russell W. Patterson is a Project Specialist, System Protection, for the Tennessee Valley Authority
(TVA) in Chattanooga, Tennessee. He is responsible for reviewing and making protective relaying
recommendations on new construction and retrofit projects for the generation and transmission system.
He also has responsibility for protective relaying and control systems and field support. Prior to his
position as Project Specialist, he was TVAs Power Quality Manager responsible for field and customer
support on PQ related issues and disturbances. Mr. Patterson earned the BSEE degree from
Mississippi State University in 1991 and has completed all coursework toward the MSEE at
Mississippi State University. He is a registered professional engineer in the State of Tennessee.

35