8th Mediterranean Conference on Power Generation, Transmission, Distribution and Energy Conversion

MEDPOWER 2012

1

Abstract--A safe and cost-efficient grounding system design
of a 3 MWp photovoltaic power station according to IEEE Std
80-2000 is presented. Grounding analysis is performed by
considering the metal parts of the photovoltaic panel arrays
foundations as auxiliary ground electrodes. Utilizing also
horizontal ground conductors, required solely for the
interconnections of the metal support structures of the
photovoltaic panel arrays, both safety and cost-efficiency in
grounding system design have been achieved. It is shown that in
large-scale photovoltaic power stations where the metal parts of
the panel arrays foundations are concrete encased the concrete
resistivity is not an important parameter in evaluating the
safety performance of the grounding system.

I ndex Terms--grounding system design, photovoltaic power
station, safety, touch and step voltages.
I. INTRODUCTION
he safe grounding system design of a technical
installation is based on the protection of persons against
the danger of critical electric shock. Furthermore, it allows
for the flow of normal or fault currents into the earth without
exceeding operating and equipment limits or affecting
adversely the continuity of service [1]. A safe grounding
system design requires the calculation of allowable touch
and step voltage limits which should not be exceeded, as
well as the computation of the ground resistance, the ground
potential rise and of the touch and step voltages arising in
case of the worst possible ground fault. These design
parameters depend on the soil resistivity of the installation
area. Thus, initially, accurate soil resistivity measurements at
the site of the planned installation should be performed.
In general, the design procedure of a safe grounding
system consists of five major steps:
i. Analysis of soil resistivity measurements
ii. Estimation of the allowable limits of touch and step
voltages
iii. Fault current analysis for the estimation of the
maximum grid current
iv. Grounding system design
v. Evaluation of the safety performance of the designed
grounding system.
In recent years the use of renewable energy sources, such
as wind and solar power, for the generation of green
electrical energy has increased rapidly. Numerous large-
scale photovoltaic power stations covering large areas and

Z. G. Datsios (e-mail: zdatsios@auth.gr) and P. N. Mikropoulos
(e-mail: pnm@eng.auth.gr) are with the Department of Electrical Energy,
School of Electrical and Computer Engineering, Aristotle University of
Thessaloniki, Thessaloniki, 54124 Greece.
yielding power higher than 500 kWp have already been
constructed around the globe and more are planned.
According to [2], large-scale photovoltaic power stations
installed worldwide during 2010 yield power about 3.5 GWp
and the total installed power is higher than 9 GWp. Despite
the rapid increase in the number of photovoltaic power
stations, their safe grounding system design is analyzed
sparsely in literature [3]. This paper presents the safe and
efficient grounding system design of a 3 MWp photovoltaic
power station. The grounding design is carried out according
to the IEEE Std 80-2000 [1], which is primarily concerned
with outdoor ac substations. However, as discussed in this
work, a photovoltaic power station differs from a typical
outdoor ac substation, introducing, therefore, a number of
issues related to grounding analysis.
II. PHOTOVOLTAIC POWER STATION UNDER STUDY
Fig. 1 illustrates the 3 MWp photovoltaic power station
under study which comprises 12 photovoltaic panel array
groups and covers an area of about 58000 m
2
. The generated
power is collected at three local prefabricated substations,
transferred by underground high voltage cables to the main
indoor substation and finally delivered to the 20 kV
distribution system. The metal structures of the photovoltaic

Array Group
1
Array Group
2
Array Group
7
Array Group
8
Array Group
9
Array Group
4
Array Group
12
Array Group
6
Array Group
5
2
6
5

m
8
5
m
49 m
7
8

m
219 m
Substation 1 Substation 2
Substation 3
Main Substation Auxiliary Building
Array Group
11
Array Group
10
Array Group
3

Fig. 1. Layout of the 3 MWp photovoltaic power station under study.
Safe grounding system design for a photovoltaic
power station
Zacharias G. Datsios and Pantelis N. Mikropoulos
T


2
panel arrays are supported by hot-dip galvanized steel piles
with buried cast-in-place concrete foundations with an outer
concrete diameter of 0.23 m; the length of the concrete
encased part of the piles is 1.1 m.
III. ANALYSIS OF SOIL RESISTIVITY MEASUREMENTS
It is well known that soil resistivity varies through the
year due to the changes in the moisture content and
temperature mostly of the upper soil layers. Generally, soil
resistivity increases with decreasing moisture content and
temperature. When temperature drops below the freezing
point, the water in the soil freezes and soil resistivity rises
abruptly [1], [4]-[7]. The highest values of soil resistivity are
obtained in prolonged dry periods of the year due to the low
moisture content of the soil or in the winter when the water
in the upper soil layers is frozen. Therefore, the time of soil
resistivity measurements should be selected properly so as to
obtain conservative results.
Several soil resistivity measurement methods exist [7]
with the Wenner four-pin method [8] being widely used.
According to this method, four electrodes are driven into the
soil to a depth b at equal distances apart in a straight line
(Fig. 2). The test current is injected into the earth via the two
outer electrodes and the arising voltage is measured between
the two inner electrodes. The measured voltage is divided by
the test current and a resistance value is obtained. For the
calculation of the apparent soil resistivity, (m), the
following approximate expression is used when the depth b
is considerably smaller than the distance between adjacent
electrodes [7], [8]
2 R to = (1)
where (m) is the spacing between adjacent electrodes and
R () is the obtained resistance. The larger the spacing
between adjacent electrodes, the deeper the test current
penetrates the soil. Thus, the measured soil resistivity is
considered as the apparent soil resistivity to a depth equal to
the electrode spacing [1]. As soil resistivity is rarely uniform
and commonly varies with depth, soil resistivity should be
measured for several electrode spacing values. Initially, a
small spacing should be used to measure the soil resistivity
of surface layers. Subsequently, the spacing should be
increased to larger values for the determination of the soil
resistivity of deeper layers.
The interpretation of the soil resistivity measurements and
the derivation of a representative soil model approximating
the actual soil conditions is a difficult task. Generally, soil
models can be divided into three categories: uniform, two-
layer and multilayer soil models. Uniform soil models can be
used only when the actual soil is practically homogeneous,
something that is not common in practice [1]. Non-uniform

V
I

b

Fig. 2. Wenner four-pin method electrode arrangement [8].
soil conditions can be represented by two-layer soil models,
which comprise an upper layer of a finite depth, h (m), and
resistivity,
1
(m), and a lower layer of different resistivity,

2
(m), and infinite thickness. Normally, these models are
sufficiently accurate in representing the actual non-uniform
soil conditions for safe grounding system design purposes
[1]. This does not apply when the apparent resistivity as a
function of electrode spacing exhibits several local extrema,
i.e. the soil is highly non-uniform; in such cases a multilayer
soil model should be employed. The relevant IEEE
Standards [1], [7] propose several methods for the
interpretation of soil resistivity measurements, including
analytical methods and graphical methods, such as Taggs
[6] or Sundes [9].
Soil resistivity measurements at the installation site of a
large-scale photovoltaic power station should be carried out
for several values of electrode spacing, including large
spacings for the estimation of the soil resistivity of larger
depths. Moreover, lateral changes of the soil resistivity
should be evaluated as well, so several positions in the vast
installation area should be selected. For the photovoltaic
installation under study, soil resistivity measurements were
performed, according to the Wenner method [8], at eight
different positions within the site of installation and with
electrode spacings of 2, 4, 8, 16 and 32 m; the average
values of the measured apparent soil resistivity are listed in
Table I. As the apparent soil resistivity generally increases
with increasing electrode spacing (Table I), a two-layer soil
model can represent satisfactorily the actual soil conditions.
Results were analyzed with the soil analysis module of the
CYMGrd software [10], which uses an expression proposed
by Tagg [6] correlating the apparent resistivity with
electrode spacing and employs computational methods to
minimize an error function in order to obtain the optimal
two-layer soil model. The variation of the apparent soil
resistivity with the electrode spacing together with the two-
layer soil model (
1
=2796 m,
2
=7250 m, h=4.45 m) are
shown in Fig 3.
TABLE I
MEASURED APPARENT SOIL RESISTIVITY
Electrode Spacing (m) Apparent Soil Resistivity (m)
2 3028
4 2944
8 4507
16 5298
32 6429

2000
3000
4000
5000
6000
7000
8000
0 5 10 15 20 25 30 35 40
Electrode spacing (m)
A
p
p
a
r
e
n
t

s
o
i
l

r
e
s
i
s
t
i
v
i
t
y

(

m
)
Calculated values
Two-layer soil model
Measured values

Fig. 3. Variation of the apparent soil resistivity with electrode spacing and
the two-layer soil model derived by using the CYMGrd software [10].


3
IV. ESTIMATION OF THE ALLOWABLE LIMITS OF TOUCH AND
STEP VOLTAGES
The estimation of the maximum allowable touch and step
voltages is required for the safe design of a grounding
system. These voltage limits are derived from the tolerable
body currents which do not cause ventricular fibrillation and
depend on the soil resistivity and the duration of the shock
current to which a person might be exposed; the latter
corresponds to the fault clearing time of primary protection
or to that of the slower backup protection [1]. The touch and
step voltages caused by ground faults should be lower than
the allowable touch and step voltage limits, respectively.
The allowable limits of touch and step voltages, E
touch
(V)
and E
step
(V), respectively, are given as [1]
( )
( )
1000 1.5
touch s s s
E C k t = + (2)
( )
( )
1000 6
step s s s
E C k t = + (3)
where t
s
(s) is the duration of the shock current and k is a
factor equal to 0.116 or 0.157 for people with body weight
of 50 kg and 70 kg, respectively [1]. C
s
is a derating factor,
given by (4) [1], introduced when a thin layer of high
resistivity surface material is spread on the earths surface
within the installation area in order to increase the contact
resistance between the ground and persons feet.

0.09 1
1
2 0.09
s
s
s
C
h

| |

|
\ .
=
+
(4)
In (4) (m) and
s
(m) are the soil and surface material
resistivities, respectively and h
s
(m) is the thickness of the
surface material. If no surface material is applied, then
s
is
equal to and C
s
is equal to 1. It is important to note that the
surface material resistivity corresponding to wet conditions
should be used in calculations in order to obtain results on
the side of safety.
According to common practice, photovoltaic power
stations are usually protected by fences and are inaccessible
to the general public. Thus, the allowable voltage limits
obtained for a person weighing 70 kg should be selected for
the installation area. On the other hand, for areas outside the
fence, the allowable limits obtained for a person weighing 50
kg may be selected as well. Furthermore, it should be
pointed out that the application of a high resistivity surface
material is not common in photovoltaic installations.
However, such surface material may be spread on certain
restricted areas of the installation, as a cost-effective method
to achieve safety.
Table II shows the calculated allowable touch and step
voltage limits of the installation under study as a function of
the surface material resistivity for three different durations of
shock current and with a surface material thickness of
0.15 m. As expected, the allowable touch and step voltage
limits are higher for a person with body weight of 70 kg than
50 kg, increase with increasing surface material resistivity
and decrease as the duration of shock current increases. In
this study, the selected limits of touch and step voltages are
1153 V and 3947 V, respectively, corresponding to a person
weighing 70 kg, a duration of shock current of 0.5 s and no
surface material application.

TABLE II
ALLOWABLE TOUCH AND STEP VOLTAGES LIMITS CALCULATED
ACCORDING TO (2)-(4); SURFACE MATERIAL THICKNESS 0.15 m
s (m)
ts (s)
0.1 0.5 1
Etouch70
(V)
Estep70
(V)
Etouch70
(V)
Estep70
(V)
Etouch70
(V)
Estep70
(V)
without 2579 8825 1153 3947 815 2791
4000 3268 11584 1462 5181 1034 3663
5000 3841 13876 1718 6205 1215 4388
6000 4414 16167 1974 7230 1396 5112
7000 4987 18459 2230 8255 1577 5837
8000 5560 20750 2486 9280 1758 6562
9000 6133 23041 2743 10304 1939 7286
10000 6706 25333 2999 11329 2120 8011

Etouch50
(V)
Estep50
(V)
Etouch50
(V)
Estep50
(V)
Etouch50
(V)
Estep50
(V)
without 1905 6521 852 2916 603 2062
4000 2415 8559 1080 3828 764 2707
5000 2838 10252 1269 4585 898 3242
6000 3261 11945 1459 5342 1031 3777
7000 3685 13638 1648 6099 1165 4313
8000 4108 15331 1837 6856 1299 4848
9000 4531 17024 2026 7613 1433 5384
10000 4954 18717 2216 8371 1567 5919
V. ESTIMATION OF THE MAXIMUM GRID CURRENT
For the safe grounding system design, fault current
calculations should be performed so as to determine the most
dangerous ground fault, which is usually associated with the
highest value of grid current. The latter is defined in [1] as
the portion of the symmetrical ground fault current that
flows between the grounding grid and the surrounding earth.
The rms value of the symmetrical grid current, I
g
(A), is
given as [1]

g f f
I S I = (5)
where I
f
(A) is the rms value of the symmetrical ground fault
current and S
f
is the fault current division factor. S
f
is lower
or equal to unity and takes into account the portion of the
symmetrical ground fault current flowing away from the
grounding grid through overhead ground wires, neutral
conductors, counterpoise conductors, buried metal pipes and
cables with sheaths or armor in contact with the soil.
However, the actual fault current is usually asymmetrical.
Thus, an rms equivalent of the actual grid current shall be
estimated in order to take into account the asymmetry of the
fault current in the grounding system design. This rms
equivalent current, called maximum grid current, I
G
(A), is
given as [1]

G f g
I D I = (6)
where D
f
is a decrement factor introduced so as to consider
the dc offset of the asymmetrical fault current. This factor
depends on the duration of the fault, t
f
(s), and the X/R ratio,
which is the ratio of inductive reactance to resistance of the
system at the fault location, indicating the rate of the dc
offset decay. The decrement factor, D
f
, is given as [1]

( ) ( )
2
1 1
f a
t T
f a f
D T t e

= + (7)
where the dc offset time constant T

(s) is equal to X/(R).

4
The most dangerous ground fault for a photovoltaic
power station typically corresponds to a ground fault at the
high voltage side of the step-up transformers. However,
exceptions may occur, as in [3], where the worst ground
fault occurs at the switching station of a power plant
adjacent to the photovoltaic power station. In the present
photovoltaic power station the worst ground fault generating
the highest grid current corresponds to a 20 kV single phase
ground fault at the main substation. The value of 1 kA for
the symmetrical ground fault current, I
f
, was considered as a
worst case scenario; the ground fault currents at the 20 kV
side of the step-up transformers are limited to values lower
than 1 kA by the 12 current limiting resistance utilized in
the particular medium voltage distribution system. The fault
current division factor, S
f
, was taken equal to 1, since no
overhead ground wires or neutral conductors exist; therefore
the symmetrical grid current I
g
equals to 1 kA according to
(5). The decrement factor, D
f
, was found 1.0127, calculated
according to (7) for a X/R ratio equal to 4 and a fault
duration of 0.5 s corresponding to the fault clearing time of
the primary protective devices. Hence, the maximum grid
current, I
G
, used for the grounding system design was found
1012.7 A according to (6).
VI. GROUNDING SYSTEM DESIGN
In order to ensure a safe, yet cost-efficient, grounding
system design for a photovoltaic power station, the metal
parts of the foundations supporting the metal structures of
the photovoltaic panel arrays should be considered as
auxiliary ground electrodes. This is essential as it reduces
considerably the photovoltaic power stations ground
resistance and, consequently, the ground potential rise and
the arising touch and step voltages within the installation in
case of a ground fault. Hence, both safety and cost-
efficiency can be achieved without requiring excessively
large lengths of ground conductors or a high resistivity
surface material spread to the whole installation area of the
photovoltaic facility.
A large number of different foundation types can be
utilized to support the metal structures of the photovoltaic
panel arrays. From the grounding system design point of
view, they can be divided into those with metal parts in
direct contact with the soil, e.g. driven piles and ground
screws, and those with concrete encased metal parts, e.g.
concrete encased piles and concrete foundations with steel
reinforcing bars. The cross section of the metal parts of these
foundation types is normally large enough to carry the
ground fault currents into the earth. This also applies for the
metal structures which support the photovoltaic panels and
interconnect the metal parts of the foundations.
Nevertheless, attention should be given to the
interconnections of the metal structures because paint films,
protective coatings and metal surface treatments could
introduce high resistances, resulting in unreliable electrical
connections.
Concrete encased electrodes were first used as ground
electrodes for lightning protection systems of military
installations in the 1940s [11]. Concrete is a hygroscopic
material that means it attracts moisture from the surrounding
soil [1], [12] and therefore it retains a high water content and
low resistivity even under adverse conditions [12]. The
IEEE Std 80-2000 [1] proposes for the resistivity of concrete
values ranging between 30 m and 90 m. However,
several authors, based on literature review, have proposed
higher values for concrete resistivity, such as 100-300 m
[13] and 100-400 m [14]. It is noteworthy that the ground
resistance of concrete encased electrodes is lower than the
ground resistance of the same electrodes embedded directly
in soil when the resistivity of the latter is relatively high [1].
For the photovoltaic power station under study the
concrete encased parts of the hot-dip galvanized steel piles
supporting the photovoltaic panel arrays were considered as
auxiliary ground electrodes. The 28 piles belonging to each
photovoltaic panel array (Fig. 4) are all interconnected
above ground by the metal structures supporting the
photovoltaic panels. Also, horizontal ground conductors,
buried next to the array groups at a depth of 0.5 m, were
employed to interconnect the metal support structures of all
photovoltaic panel arrays, as shown for the array groups 1
and 2 in Fig. 4.
The grounding systems of the substations and of the
auxiliary building consist of a foundation earth electrode and
a ring ground electrode being buried at a depth of 0.5 m and
surrounding the buildings at a distance of 1 m from their
boundaries. These grounding systems are connected to the
photovoltaic power stations grounding network with two
horizontal ground conductors buried at 0.5 m.
All the ground conductors are made of copper-clad steel
and have the commercially available diameter of 10 mm. It
must be noted that the minimum conductor size based on
fusing was calculated 5.36 mm
2
according to [1] for the
maximum grid current of 1012.7 A and for the fault clearing
time of 1 s corresponding to the backup protection.
Array Group
1
Array Group
2
9 m 1 m 3 m

Fig. 4. Schematic of the grounding system of array groups 1 and 2. Circles
represent the piles supporting the metal structures of the photovoltaic panel
arrays. Yellow lines depict horizontal ground conductors buried at 0.5 m.
VII. SAFETY PERFORMANCE EVALUATION
The safety performance of the grounding system was
evaluated with the aid of the CYMGrd software [10]. Fig. 5
shows the computer model of the photovoltaic power
stations grounding system. The model includes 3766
concrete encased steel piles, arranged according to the exact


5

Fig. 5. Computer model of the photovoltaic power stations grounding system. The graph is not according to scale.
layout of the photovoltaic panel arrays, and approximately
1.7 km of buried horizontal ground conductors. The
connections of the horizontal ground conductors with the
adjacent piles were not modeled except for those at the outer
corners of the photovoltaic facility, where the highest touch
and step voltages are typically expected; this would result in
conservative results. The foundation earth electrodes of the
substations and the auxiliary building were modeled as
sparse grounding grids. The concrete resistivity was assumed
equal to 90 m, which the highest value proposed by the
IEEE Std 80-2000 [1]. It must be mentioned that the
CYMGrd software [10] assumes that all the electrodes are
electrically connected together and are elevated to a single
potential, the ground potential rise.
It was not possible to analyze the complete computer
model of the photovoltaic power stations grounding system
(Fig. 5) due to the large number of piles employed; this is
common for large-scale photovoltaic power stations. To
overcome this limitation, the number of the modeled piles
could be reduced appropriately. However, the exact layout
of the piles should always be maintained in the areas where
the safety performance of the grounding system is evaluated.
Nevertheless, the modified computer model would yield
results on the side of safety due to the reduced number piles
employed in analysis. As an alternative approach [3], the
modeled piles, with the exception again of those positioned
at the areas where the safety performance is evaluated, could
be replaced by sparse horizontal ground conductors,
obtaining, thus, conservative results.
In the present case, the computer model of the
photovoltaic power stations grounding system (Fig. 5) was
divided in two sections A and B comprising array groups 1-6
and 7-12, respectively (Fig. 1). Subsequently, the number of
the modeled piles in section A was reduced appropriately
and safety performance was evaluated in section B and vice
versa. For both cases, the ground resistance and ground
potential rise were found approximately 11.9 and 12 kV,
respectively; according to above, these are conservative
values, resulting, therefore, in a safer grounding design.
Fig. 6 shows touch voltage contours for the section A of
the computer model of the photovoltaic power stations
grounding system; the number of the modeled piles was
reduced appropriately in section B. From Fig. 6 it can be
seen that, with the exception of the two outer corner areas of
the photovoltaic facility, the touch voltages computed for
distances up to 1 m from the photovoltaic panel arrays and
substations metal structures are well below the
corresponding 1153 V touch voltage limit. This is also true
for the section B of the computer model, where the highest
touch voltage (1486 V) within the installation area is found
at the outer corner of the array group 12 (Fig. 1). In order to
ensure safety in the four outer corner areas of the
photovoltaic installation, a surface material with resistivity
equal to or higher than 5000 m (Table II) can be used
locally to increase the allowable touch voltage limit (1718
V) above the computed touch voltages. Otherwise, a short
horizontal ground conductor with length about 9 m can be
installed at each outer corner area, parallel to the last
photovoltaic panel array at a distance of 1 m from its
boundary, as depicted in Fig. 4 for array group 1. Fig. 7
shows the touch voltages computed along a profile parallel
to the outermost array of the facility in array group 12 (Fig.
1) at a distance of 1 m from its metal structure for the cases



0
(0%)
288 V
(25%)
577 V
(50%)
865 V
(75%)
1153 V
(100%)

Fig. 6. Touch voltage contours of computer model section A. The graph is
not according to scale; the highest touch voltage is 1485 V.

500
600
700
800
900
1000
1100
1200
1300
1400
1500
0 5 10 15 20 25 30 35 40 45 50
Length (m)
T
o
u
c
h

v
o
l
t
a
g
e

(
V
)
Without ground conductor
With ground conductor
Touch voltage limit

Fig. 7. Touch voltages computed along a profile parallel to the outermost
array of the photovoltaic facility in array group 12 at a distance of 1 m
from its metal structure; cases of with and without the 9 m horizontal
ground conductor, red line depicts the touch voltage limit of 1153 V.


6
of with and without the horizontal ground conductor. It is
obvious that the addition of the short ground conductor
reduces the computed touch voltages to values lower than
the touch voltage limit.
The maximum computed surface potential difference
within the installation area is 3655 V. Thus, the step voltages
being lower than the 3947 V step voltage limit (Table II),
are not a threat for persons in the installation area.
Summarizing, as the computed touch and step voltages
arising in the photovoltaic facility in case of the worst
possible ground fault are lower than the allowable limits of
touch and step voltages, the designed grounding system
ensures the safety of persons in the installation area.
Finally, analysis of the photovoltaic facilitys grounding
system was also performed for different values of concrete
resistivity, ranging from 30 m to 400 m. The limits of
this range correspond to the lowest and highest values of
concrete resistivity proposed in [1], [12] and [14]. The effect
of concrete resistivity on ground resistance, thus also ground
potential rise, was found to be negligible, as the ground
resistance varied less than 1.5% over the range of concrete
resistivity studied. Actually, the ground resistance increased
only slightly (12.05 ) when the pile concrete encasement
was disregarded in analysis, in spite of the relatively high
resistivity of the upper soil layer (2796 m). This is because
of the large number of ground electrodes in the installation
area; in such cases, the resistivity of the concrete encasing
ground electrodes is not an important parameter in
evaluating the safety performance of the grounding system.
VIII. CONCLUSIONS
A safe and cost-efficient grounding system of a 3 MWp
photovoltaic power station has been designed according to
IEEE Std 80-2000 with the aid of grounding analysis
software. The metal parts of the photovoltaic panel arrays
foundations should be considered as auxiliary ground
electrodes. In this case, by utilizing also horizontal ground
conductors required solely for the interconnections of the
metal support structures of the photovoltaic panel arrays
rather than forming a mesh, both safety and cost-efficiency
in grounding system design are achieved. In restricted areas,
mainly at the outer boundaries of the photovoltaic panel
array groups, where touch voltages exceed the allowable
limit, a high resistivity surface material can be spread or
horizontal ground conductors of relatively short lengths can
be added.
Grounding analysis of large-scale photovoltaic power
stations, involving a large number of ground electrodes, can
be performed by reducing the number of the ground
electrodes in the computer model, provided that their exact
layout is always maintained in the areas where the safety
performance of the grounding system is evaluated. Such a
modified computer model yields a higher ground resistance,
due to the reduced number of ground electrodes employed in
analysis, however results are on the side of safety.
In large-scale photovoltaic power stations where the
metal parts of the panel arrays foundations are concrete
encased the concrete resistivity does not affect appreciably
the ground resistance, thus it is not an important parameter
in evaluating the safety performance of the grounding
system.
IX. REFERENCES
[1] IEEE Guide for Safety in AC Substation Grounding, IEEE Std 80-
2000, Jan. 2000.
[2] PV Power Plants 2011-Industry Guide, Renewables Insight (RENI),
2011.
[3] J. Ma and F. P. Dawalibi, Grounding analysis of a solar power
generation facility, in Proc. 2010 Asia-Pacific Power and Energy
Engineering Conference (APPEEC), Chengdu, China, Mar. 2010.
[4] P. J. Higgs, An investigation of earthing resistances, Journal of the
IEE, vol. 68, no. 402, pp. 736-750, Jun. 1930.
[5] R. Rdenberg, Grounding principles and practice I-Fundamental
considerations on ground currents, Electrical Engineering, vol. 64,
no. 1, pp. 1-13, Jan. 1945.
[6] G. F. Tagg, Earth resistances, Pitman Publishing Corporation,
London, 1964.
[7] IEEE Guide for Measuring Earth Resistivity, Ground Impedance,
and Earth Surface Potentials of a Ground System Part I: Normal
Measurements, ANSI/IEEE Std 81-1983, 1983.
[8] F. Wenner, A method of measuring earth resistivity, Bulletin of the
Bureau of Standards, Scientific Paper 258, vol. 12, pp. 469-478, Jul.
1915.
[9] E. D. Sunde, Earth conduction effects in transmission systems,
Dover Publications Inc, New York, 1968.
[10] CYMGRD 6.3 for Windows Users guide and reference manual,
CYME International T&D Inc., Oct. 2006.
[11] H. G. Ufer, Investigation and testing of footing-type grounding
electrodes for electrical installations, IEEE Trans. Power App. Syst.,
vol. PAS-83, no. 10, pp. 1042-1048, Oct. 1964.
[12] E. J. Fagan and R. H. Lee, The use of concrete-enclosed reinforcing
rods as grounding electrodes, IEEE Trans. Ind. Gen. Appl., vol.
IGA-6, no. 4, pp. 337-348, Jul./Aug. 1970.
[13] J. Preminger, Evaluation of concrete-encased electrodes, IEEE
Trans. Ind. Appl., vol. IA-11, no. 6, pp. 664-668, Nov./Dec. 1975.
[14] A. M. Mousa, Discussion of Fault induced voltages on metallic
fencing located in the vicinity of a high voltage substation, IEEE
Trans. Power App. Syst., vol. PAS-101, no. 3, pp. 746-750, Mar.
1982.
X. BIOGRAPHIES
Zacharias G. Datsios was born in Kozani, Greece, in 1986. He received
the M.Eng. degree in electrical and computer engineering from Aristotle
University of Thessaloniki (AUTh), Thessaloniki, Greece, in 2010, where
he is currently pursuing the Ph.D. degree at the High Voltage Laboratory.
His research is related to grounding systems, especially dedicated to the
soil ionization phenomenon.

Pantelis N. Mikropoulos was born in Kavala, Greece, in 1967. He
received the M.Eng. and Ph.D. degrees in electrical and computer
engineering from Aristotle University of Thessaloniki (AUTh),
Thessaloniki, Greece, in 1991 and 1995, respectively.
He held postdoctoral positions at AUTh and the University of
Manchester, Manchester, UK. He was Senior Engineer with Public Power
Corporation SA, Athens, Greece. He was elected Assistant Professor and
Associate Professor in High Voltage Engineering at AUTh in 2003 and
2010, respectively. Since 2005 he has been the Director of the High
Voltage Laboratory at AUTh. His research interests include the broad area
of high-voltage engineering with emphasis given to air and surface
discharges, electric breakdown, lightning protection and insulation
coordination for power systems.