GROUND POTENTIAL RISE INFLUENCE NEAR HV SUBSTATIONS IN URBAN AREAS

L Grcev, V Filiposki, V Arnautovski

St. Cyril and Methodius University, Macedonia

The ground potential rise (GPR) in high voltage 1250,1250 (m)

substations in case of ground faults may cause
dangerous voltages between telecommunication
installations and local ground. It is important to
note that not only installations serving substations,
but all other general use cables and telephone
subscriber installations in the zone of influence are
equally endangered and have to be protected. The
CCITT directives define 430 V (or 650 V) contour 0,0
as a border of the zone of GPR influence on the
telecommunication installation and all wire-line
telephone subscriber installations inside such zone -1500,-500 (m)
have to be protected.

If the high voltage substation is located out of ur-
ban area, the equipotential contours usually follow
the form of the grounding system, tending to be-
Fig. 1: Grounding system connected to cable network of
existing 110/10 kV substation in urban area.
60
Potenital (% of substation potential)

come circles at larger distances, with decaying Substation Underground cable

values in radial direction from the zone of the sub-
station, depending on the soil structure and resis- 50
tivity. Profile

However, if the high voltage substation is located 40

in the urban area, its grounding system may be
connected to a buried network of uncoated metallic
sheathed cables. Other buried metallic structures 30 Measurements
are also often located close to the substation Calculations
grounding system and are extended through out
20
the urban area, such as metal sheaths of tele-
communication and power cables, neutral wires of Distance from the origin of profile (m)
power distribution lines, water pipes, pipelines for
heating and gas, rails of traffic systems. If these Fig. 2: Normalized values of calculated and measured
buried metal structures are not taken into account values of the potential along a profile near
it may lead to underestimate of the zone of influ- 110/10 kV substation.
ence.

Although national and international regulations de-
fine the potential contour 430 V (650 V) as a bor-
der of the GPR zone of influence on telecommuni-
cation installations in case of faults to ground in-
side a high voltage substation or on a connected
power line, if the substation is in highly urbanized
area, the potentials are equalized due to large and
complex underground metallic networks. In such
cases, as more suitable to define the protection
measures are potential differences that may cause
overvoltages in the telecommunication lines.
A computer model of the substation grounding
system that include connected and near metallic
structures is developed. As an example of its ca-
pabilities, network of 10 kV buried uncoated metal-
lic sheathed cables in a 2750 × 1750 m area, il-
2
lustrated in Fig. 1, is included in the model. The
computer model is validated by comparison with
other author’s software and with direct comparison
with field measurements, Fig. 2.
This model is used to estimate the GPR zone of
influence to the consumer installations in urban
environment. The effect of the equalizing of poten-
tials by other buried metallic networks that are
typical for urban environment and are usually un-
known is included in the model as a correction
based on measured potentials in a small number
of specified points.
GROUND POTENTIAL RISE INFLUENCE NEAR HV SUBSTATIONS IN URBAN AREAS

L Grcev, V Filiposki, V Arnautovski

St. Cyril and Methodius University, Macedonia

The ground potential rise (GPR) in high voltage 1250,1250 (m)

substations in case of ground faults may cause
dangerous voltages between telecommunication
installations and local ground. It is important to
note that not only installations serving substations,
but all other general use cables and telephone
subscriber installations in the zone of influence are
equally endangered and have to be protected. The
CCITT directives define 430 V (or 650 V) contour 0,0
as a border of the zone of GPR influence on the
telecommunication installation and all wire-line
telephone subscriber installations inside such zone -1500,-500 (m)
have to be protected.

If the high voltage substation is located out of ur-
ban area, the equipotential contours usually follow
the form of the grounding system, tending to be-
Fig. 1: Grounding system connected to cable network of
existing 110/10 kV substation in urban area.
60
Potenital (% of substation potential)

come circles at larger distances, with decaying Substation Underground cable

values in radial direction from the zone of the sub-
station, depending on the soil structure and resis- 50
tivity. Profile

However, if the high voltage substation is located 40

in the urban area, its grounding system may be
connected to a buried network of uncoated metallic
sheathed cables. Other buried metallic structures 30 Measurements
are also often located close to the substation Calculations
grounding system and are extended through out
20
the urban area, such as metal sheaths of tele-
communication and power cables, neutral wires of Distance from the origin of profile (m)
power distribution lines, water pipes, pipelines for
heating and gas, rails of traffic systems. If these Fig. 2: Normalized values of calculated and measured
buried metal structures are not taken into account values of the potential along a profile near
it may lead to underestimate of the zone of influ- 110/10 kV substation.
ence.

Although national and international regulations de-
fine the potential contour 430 V (650 V) as a bor-
der of the GPR zone of influence on telecommuni-
cation installations in case of faults to ground in-
side a high voltage substation or on a connected
power line, if the substation is in highly urbanized
area, the potentials are equalized due to large and
complex underground metallic networks. In such
cases, as more suitable to define the protection
measures are potential differences that may cause
overvoltages in the telecommunication lines.
A computer model of the substation grounding
system that include connected and near metallic
structures is developed. As an example of its ca-
pabilities, network of 10 kV buried uncoated metal-
lic sheathed cables in a 2750 × 1750 m area, il-
2
lustrated in Fig. 1, is included in the model. The
computer model is validated by comparison with
other author’s software and with direct comparison
with field measurements, Fig. 2.
This model is used to estimate the GPR zone of
influence to the consumer installations in urban
environment. The effect of the equalizing of poten-
tials by other buried metallic networks that are
typical for urban environment and are usually un-
known is included in the model as a correction
based on measured potentials in a small number
of specified points.
GROUND POTENTIAL RISE INFLUENCE NEAR HV SUBSTATIONS IN URBAN AREAS
L Grcev, V Filiposki, V Arnautovski
St. Cyril and Methodius University, Macedonia
INTRODUCTION
The ground potential rise (GPR) in high voltage
substations in case of ground faults [1] may cause
dangerous voltages between telecommunication
installations and local ground. It is important to
note that not only installations serving substations,
but all other general use cables and telephone
subscriber installations in the zone of influence are
equally endangered and have to be protected
[2,3,4]. The CCITT directives [5] define 430 V
contour as a border of the zone of GPR influence
on the telecommunication installation and all wire-
line telephone subscriber installations inside such
zone have to be protected [6].
If the high voltage substation is located out of ur-
ban area, the equipotential contours usually follow
the form of the grounding system, tending to be-
come circles at larger distances. Typically the
ground potential value decays in radial direction
from the zone of the substation, depending on the
soil structure and resistivity [7].
However, if the high voltage substation is located
in the urban area, its grounding system is often
connected to a buried network of uncoated metallic
sheathed cables [8]. Although such cables are no
longer manufactured in many countries, many of
them are still in operation. Other buried metallic
structures are also often located close to the sub-
station grounding system and are extended
through out the urban area, such as metal sheaths
of telecommunication and power cables, neutral
wires of power distribution lines, water pipes, pipe-
lines for heating and gas, rails of traffic systems. If
these buried metal structures are not taken into
account [9], this may lead to underestimation of the
zone of influence, as it is shown in [10]. Previous
papers [11,12,13] have shown that metallic net-
works directly connected to the substation
grounding system substantially affect the GPR
zone of influence. However, all elements of the
urban environment that affect potential distribution
cannot be included in any model, firstly, because
of the complexity of the problem, and, secondly,
because there are numerous unknown elements of
the urban environment.
Fault to Ground
Underground Cable Network
110/10 kV
Substation 10/0.4 kV
Substation
Figure 1: Grounding system of high voltage
substation connected to underground
cable network.
This paper presents results of measurements that
confirm this conclusion in a zone near the substa-
tion, while at a distance, especially in the highly
urbanized zone, the additional buried networks of
conductors tend to equalize potentials. The level of
such equalizing of the potentials has to be deter-
mined experimentally [6,14]. This effect is included
in the computer model by forcing the model to ac-
cept equalizing of potentials in extent determined
by measurements in a small number of points.
MODEL OF THE GROUNDING SYSTEM
CONNECTED TO UNDERGROUND CABLE
NETWORK
The model of a substation grounding system con-
nected to underground cable network is illustrated
in Fig. 1. It is based on the following assumptions
[8,11,17,]. The grounding electrodes are modeled
as a set of connected conductors buried in homo-
geneous or two-layer soil. The uncoated metallic
cable sheaths are directly connected to the
grounding conductors and are connected to 10/0.4
kV local substations or to other cables.
The grounding electrodes are divided into ng
segments, and the cables are divided in nc
segments. Cables are considered to have
 z1 I1 u1 z2 I2 u2
ug
i1 i2
Figure 2: Equivalent circuit of the uncoated
metallic sheathed cable.
longitudinal impedance, with equivalent circuit
illustrated in Fig. 2 [8,17]. The parameters yk
denotes the longitudinal admittance of cable k
sheat segment, ik denotes the current emanating to
the soil from sheat segment k, while Ik denotes
current flowing through sheat segment k. The
potential of the source substation is ug, while uk is
the potential of the sheat segment k. The
consumer substation grounding system is denoted
by the admittance ykt. The parameters needed for
the circiut in Fig. 2 are obrained using measured
data for given types of cables [11].
For this model of a cable, the following relation can
be written
- i k = Ygk ×ug + Yk ×uk ,
where
t
Ygk º [- y k 0...0 ] , and
zk Ik uk
The previous equations imply that there is no mu-
tual magnetic coupling between cable segments.
Such an assumption holds practically for the un-
ik coated, steel armored lead sheathed three-phase
zkt cables that are commonly used in distribution sys-
tems. For unarmored cable constructions with un-
coated metal sheaths the expressions (6) are ap-
proximate.
Voltages uc depend on the currents flowing from
cable sheath segments and from the source sub-
station ground electrode into the soil. This relation-
ship can be expressed as
uc = Rgc ×i g + Rcc ×ic (7)
The analogous equation for the source substation
ground electrode will be
t
1g ×ug = Rgg ×i g + Rgc ×ic (8)
From the current continuity law it follows
1tg ×ug + 1ct ×ic = J (9)
If we substitute uc from (7) into (4), the following
relationship is obtained
(1)
Yc ×Rgc ×i g + (Yc ×Rcc + E )×i c + Yg ×ug = 0 (10)
Relations (8)-(10) build a closed system of linear
(2) equations that can be solved for ig, ic and ug.

é2 y k - yk 0 ... 0 ù
ê ú VALIDATION OF THE COMPUTER MODEL
ê- y k 2y k - yk ... 0 úú
Yk º êê . (3)
ê ... ... ... ... ... úú
ê0 0 ... - y k y k + y kt úúû The validation of the results of computer analysis
ëê is based on the comparison with field measure-
ments performed by the Electricite de France,
Equation (1) is valid for all cables emanating from Paris, France, in the period 1976-85. An extensive
the source substation if the specific values for pa- set of experiments had been performed for differ-
rameter yk and ykt are used. Consequently, based ent grounding electrodes arrangements [15]. Some
upon (1), the following general equation, including comparisons for low and high frequencies are
all radial positioned cables, can be written documented in [16].
- i c = Yg ×ug + Yc ×uc (4) Figure 3 shows comparison with measurements in
case of irregular and complex arrangement of
where ground electrodes [18], and Fig. 4 direct compari-
t t
ic º éêëi1t ... i mt ùúû , uc º éêëu1t ... umt ùúû (5) son in urban area is presented for the practical
case given in in the next part.

t ùt The computer model was also compared for power

Yg º éêëYgt1 ... Ygm úû , Yc º diag {Yk }
In case of complex cable configuration, the matrix
Yc is not diagonal, and it takes the form, which cor-
responds to mutual cable connections.
(6)
frequencies with other authors’ software [19], with
excellent agreement of the results, Fig. 5. Results
in Fig. 5 are magnitudes of the electric field com-
ponent along profile A-A at earth’s surface for the
illustrated ground grid. Conductors are constructed
of copper with radius 0.5 cm, and grid is buried in POTENTIAL DISTRIBUTION AROUND
soil with ρ = 100 Ω⋅m at 0.5 m depth. Current of 1 110/10 KV SUBSTATION IN URBAN AREAS
kA at power frequency is injected at the central
point.
The computer model is used for estimation of the
Earth potential GPR zone of influence around an existing 110/10
kV substation in case of a ground fault. The net-
100 Measurement work of 10 kV buried uncoated metallic sheathed
cables in a 2750 × 1750 m area, illustrated in Fig.
Simulation 2
%
6, is included in the model. The substation’s sur-
60 rounding is a highly urbanized area, on the right
side in Fig. 5, and partially urbanized area, on the
40 left side in Fig. 5. For the same case detailed
20 measurements were made. The measurements
were done using current injection method, follow-
0 ing procedures in [14].
0 50 100 m 200
Distance along profile

Figure 3: Comparison with field measurements. 1250,1250 (m)

60
Potenital (% of substation potential)

Substation Underground cable

50
Profile
0,0
40

-1500,-500 (m)
30 Measurements
Calculations
Figure 6: Grounding system connected to cable
20 network of existing 110/10 kV
substation in urban area.
Distance from the origin of profile (m)
0
Figure 4: Normalized values of calculated and Average value of the
Voltage between substation grounding

measured values of the potential along measured voltage

1
a profile near 110/10 kV substation.
system and local ground (V)

Electric field (kV/m)

1 3

0.8 4 Level of the potential equalising

A A
10 m
0.6 5 Substation zone
Partially urbanized Highly urbanized area
10 m area
0.4 6
-400 -200 0 200 400 600 800
Radial distance from the substation center (m)
0.2
Figure 7: Measured values of the voltage be-
0
0 5 10 15 20 tween grounding system and local
Distance along profile (m) ground using current of 100 A.

Figure 5: Comparison with other authors’ model.

Results of the measurements using simulated
ground fault current of 100 A are given in Fig. 7.
Two areas could be separated in radial distance
around the substation. On the right side in Fig. 7,
the highly urbanized area that include complex
underground network of water pipes, pipelines for
heating, metallic sheathed cables for telecommu-
nication and power, local grounding and other
buried metallic structures, produces equalizing of
the potential at a distance of the substation. Such
effect is not visible on the left side in Fig. 7 for par-
tially urban area. This result is in accordance with
similar already reported in [6,14,20].
It may be concluded that the protection of the sub-
scriber telecommunication installation based on
the concept of the GPR zone of influence is not
always applicable in the highly urbanized areas,
where potentials may be equalized due to large
underground networks of conductors. In such
cases protection should be based on the potential
difference that may cause overvoltages in the tele-
communication lines.
PROTECTION OF SUBSCRIBERS’ AND
EXCHANGE INSTALLATIONS
In case when there is a potential difference be-
tween the subscriber local ground near the substa-
tion and the exchange local ground larger than 430
V, gas discharge tubes (GDTs) were used for pro-
tection, Fig. 8. To protect subscribers’ terminal
equipment against overvoltages coming in by the
telecommunication lines caused by GPR at high
voltage substation, GDTs are connected between
the line conductors and local earth [5]. The same is
done on the side of the exchange.
CONCLUSION
When the grounding system of a high voltage sub-
station is connected to uncoated metallic shielded
cables, they act as a part of the grounding system,
and have large influence on the potential distribu-
tion around the substation. National and interna-
tional regulations define the potential contour 430
V as a border of the GPR zone of influence on
telecommunication installations in case of faults to
ground inside a high voltage substation or on a
connected power line. However, if the substation is
in highly urbanized area, the potentials are equal-
ized due to large and complex underground metal-
lic networks. In such cases, as more suitable to
define the protection measures are potential differ-
ences that may cause overvoltages in the tele-
communication lines.
A computer model of the substation grounding
system and all connected uncoated metallic
sheathed cables is developed. This model is used
to estimate the GPR zone of influence to the con-
Subscriber
> 430 V Exchange
Figure 8: Protection on the side of the subscriber
and the exchange by gas discharge
tubes.
sumer installations in urban environment. The ef-
fect of the equalizing of potentials by other buried
metallic networks, typical for urban environment, is
included in the model based on measured poten-
tials in a small number of specified points.
ACKNOWLEDGMENT
The work was partially supported by the Ministry of
Science of Republic of Macedonia.
REFERENCES
[1] K. H. Feist, 1978, “Zone of Influence of Earth
Potential Rise”, Electra, 60, 57-68.
[2] IEEE Std. 487, 1980, "IEEE Guide for the Pro-
tection of Wire-Line Communication Facilities
Serving Electric Power Stations".
[3] G. Grand, A. J. Pesonen, 1976, “Protection of
the Telecommunication Circuits Entering
Power Stations and Substations”, Electra, 48,
51-72.
[4] A. Pesonen, J. Tattelus, P. Alatalo, 1970, “Earth
Potential Rise and Telecommunication Lines”,
CIGRE Session, Paper 36-04.
[5] CCITT, 1989, " CCITT Directives Concerning
the Protection of Telecommunication Lines
Against the Harmful Effects from Electric
Power and Electrified Railway Lines", Vol. I-IX,
ITU, Geneva, Switzerland.
[6] S. Pispiris, 1994, “Experimental Determination
of the Urban and Industrial Zones Equipoten-
tialising when High Voltage Energy Facilities
are Under Transitory Regime”, CIGRE Ses-
sion, Paper 36-201.
[7] F. Dawalibi, D. Mukhedkar, 1978 ,“Transferred
Earth Potentials in Power Systems”, IEEE
Trans. PAS, PAS-97, 90-101.
[8] J. Nahman, 1996, “Earthing Effects of Uncoated
Underground Cables and Transferred Poten-
tials”, Electr. Eng., 79, 55-60.
[9] ANSI/IEEE Std. 367, 1979, "IEEE Guide for
Determining the Maximum Electric Power Sta-
tion Ground Potential Rise and Induced Volt-
age from a Power Fault"
[10] M. Kuussaari, A. J. Pesonen, 1978 , “Earthing
Impedance Measurements of Substations”,
CIGRE Session, Paper 36-02.
[11] L Grcev, V. Filiposki, 1996, “Estimation of the
Zone of Influence of Earth Potential Rise on
Telecommunication Installations in Urban Envi-
ronment”, EMC’96 ROMA, 76-81.
[12] L. Grcev, V. Filiposki, 1997, “Earth potential
distribution around high voltage substations in
rural and urban areas”, EMC Zurich ’97, Paper
92N3, pp. 483-488.
[13] L. Grcev, V. Filiposki, 1997, “Zone of Influence
of Ground Potential Rise on Wire-Line Com-
munication Installations in Urban Areas”, IEEE
Int. Symp. EMC, pp. 580-585.
[14] J. Valjus, R. Sarmanto, 1984, “Practical
Earthing Measurements of Large Rural and
Urban Substations,” CIGRE Session, Paper
36-04.
[15] H. Rochereau, “Response of Earth Electrodes
when Fast Fronted Currents are Flowing Out”,
EDF Bulletin de la Direction des Etudes et Re-
cherches, serie B, no. 2, 1988, pp. 13-22.
[16] L. Grcev, 1996, “Computer Analysis of Tran-
sient Voltages in Large Grounding Systems,”
IEEE Trans. Power Delivery, 11, 815-823.
[17] J. Nahman, D. Salamon, 1992, “Effects of the
Metal Sheathed Cables upon the Perform-
ances of the Distribution Substations Ground-
ing Systems,” IEEE Trans. Power Delivery, 7,
1179-1187.
[18] Measurements and test of grounding system
and earth potential distribution at power station
Tikves, Report for Power Corporation of Ma-
cedonia no. 1102, Rade Koncar, Zagreb, 1971.
[19] F. Dawalibi, F. Donoso, 1993, “Integrated
Analysis Software for grounding, EMF, and
EMI”, IEEE Computer Applications in Power, 6,
19-24.
[20] G. Buse, ”Betrachtungen zur Structur des Er-
dreiches im Bereich einer Grosstadt”, Elek-
trizitatswirtschaft, 66, 646-650.