See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/216859148

Recommendations for grounding systems in lightning protection systems

Article in Ingenieria e Investigación · January 2010

DOI: 10.15446/ing.investig.v31n2SUP.25204

CITATIONS READS
8 2,074

1 author:

Johny Montana
Universidad Técnica Federico Santa María
51 PUBLICATIONS 265 CITATIONS

SEE PROFILE

All content following this page was uploaded by Johny Montana on 27 May 2014.

The user has requested enhancement of the downloaded file.

REVISTA INGENIERÍA E INVESTIGACIÓN Vol. 31 Suplemento No. 2 (SICEL 2011), OCTUBRE DE 2011 (5-10)

Recommendations for grounding systems in lightning protection

systems
Recomendaciones para el diseño de la puesta a tierra en los sistemas de protección
contra rayos

Johny Montaña1

out according to IEC 62305 for establishing the need for a

Abstract— This paper presents some practical recommendations for lightning protection system in a particular facility. Risk
designing grounding systems as part of an integral protection system
assessment takes aspects such as building materials, height,
against lightning strikes. These recommendations are made taking into
account the results of academic software whose development was based volume, purpose and the area’s lightning density into
on hybrid electromagnetic method and the method of moments. This consideration.
paper presents the results of impedance and transient voltage for triangle,
wye, counterpoises and mesh configurations. Some recommendation are
made concerning the use and characteristics of earth electrodes, for If an external protection system is required, the rolling
example effective counterpoise length, where to locate grounding rods, sphere method (IEEE Std-62305) can be used for determining
where to connect down conductors in a mesh and the potential difference the location of air terminations and down conductors. The
between points on the same grounding electrodes. These
recommendations guide a systems’ designer to ensure greater benefit earth terminations can be defined by means of an International
from grounding setups without wasting money. Electrotechnical Commission (IEC) standard table, taking soil
resistivity and protection level into account and giving the
Keywords: method of moments, hybrid electromagnetic method,
grounding configuration, grounding impedance, transient voltage
electrodes’ length or by means of specialised software.

Resumen— Este artículo presenta algunas recomendaciones Some technical books, standards and papers present typical
prácticas para el diseño del sistema de puesta a tierra, el cual configurations regarding earth electrodes for buildings,
hace parte del sistema integral de protección contra rayos. Estas towers, poles, houses, etc (IEEE Std-62305; Casas, 2008).
recomendaciones son el resultado de los análisis de un software Some of these configurations can be modified for obtaining
académico que se basa en el método electromagnético híbrido en
conjunto con el método de momentos. Se muestran los resultados better results concerning transient phenomenon. This paper
de la impedancia y la tensión transitoria para configuraciones presents a transient analysis of some such configurations to
como: triángulo, estrella de tres puntas, contrapesos y mallas. A ascertain how they can be modified to get better earth
partir de los resultados, se definen la aplicación y las electrode results.
características de las diferentes configuraciones, como por The results presented in this papers were obtained by means
ejemplo: longitud efectiva de los contrapesos, lugares dónde of specialised software developed using the hybrid
localizar las varillas, lugares en los cuales conectar las bajantes a
las mallas y las diferencias de potencial entre puntos de una electromagnetic method (Montaña, 2006a; Montaña, et al
misma puesta a tierra. Estas recomendaciones guían al diseñador 2006b). Impedance and voltage results are shown to provide
para obtener beneficios de las diversas configuraciones sin recommendations for geometry, injection point location, rod
desperdiciar dinero. electrode location, mesh size, etc.

Palabras claves— Método de momentos, método 2. HYBRID ELECTROMAGNETIC METHOD (HEM)

electromagnético híbrido, configuraciones de puesta a tierra,
impedancia de puesta a tierra, tensión transitoria. A methodology was used for describing structures’
electromagnetic behaviour, which may be represented by
1. INTRODUCTION cylindrical conductors (Montaña, 2006a; Visacro, 1992a;
Valencia, Moreno, 2003; Visacro, 1992b). Grounding system
A N integral lightning protection system consists of three
elements: external protection systems (air terminations,
down conductors, earth terminations), an internal protection
conductors are partitioned into a number of segments,
according to method of moments (MoM) (PCB-MoM) using
thin wire approximation. Each is considered to be an
system and personal safety guide. Risk assessment is carried electromagnetic field source produced by a transversal current
(IT) and a longitudinal current (IL) which are constant
throughout each segment (Figure 1).
1
Electrical Engineer, M.Sc. in High Voltage and Ph.D. in Electrical
Engineering from National University of Colombia. He is a Professor of
Universidad del Norte and is with the Power System Research Group-GISEL,
Barranquilla – Colombia, e-mail: [email protected]

5
RECOMMENDATIONS FOR GROUNDING SYSTEMS IN LIGHTNING PROTECTION SYSTEMS

IT / L

IL
j j
Figure 1. Triangle (left) and wye (right) configurations

These configurations are used in simple lightning protection

Figure. 1. Current sources in one segment
systems where it is not possible to use counterpoises or mesh
grids. The down conductor is connected in one corner for
The electromagnetic coupling between each pair of
triangle configuration or is connected at the centre point for
segments is calculated by using the expressions for scalar and
wye configurations. Figure 2 shows the impedance values at
magnetic vector potentials and assuming an average potential
the injection point for both configurations. Simulations were
V for segment and a voltage drop ∆V on it. Coupling made for both soils defined in Table I. Each conductor was 5
impedance matrices ZT and ZL are thus calculated. m in length, 0.5m depth and had 0.01 m radii.
Once this has been done, circuit relationships between
voltages and currents allow the system to be represented in a 120
compact form and solved for its nodal voltages (unknown).
Once these voltages have become known, the current
100
distribution throughout the grounding system can also be Triangle (Wet)
Wye (Wet)
ascertained. Wye (Dry)
Since all the calculations involved in this methodology are 80

Impedance (Ohms)
Triangle (Dry)
carried out in the frequency domain, soil parameters, skin
effect and propagation effects’ frequency dependence are 60
easily included. Such methodology is used in this paper to
find earth electrodes’ input impedance frequency response by
40
means of the voltage-current relationship. Responses in the
time domain are computed by means of the inverse Fourier
transform (IFT) (Montaña, 2006a) from the responses in the 20

frequency domain.
0
2 3 4 5 6 7
3. PARAMETERS ANALYSIS 10 10 10 10 10 10
Frequency (Hz)
Impedance in the frequency domain and voltage in the time Figure 2. Impedance of triangle and wye configurations
domain were computed using the HEM-based software to
compare different arrangements. The impedance analysis was This simulation was used for defining which of the two
performed from 100 Hz up to 3 MHz, due to most configurations presented the lower impedance values at the
representative lightning phenomenon components occurring injection point. The results showed that the wye configuration
within that range. presented the lower impedance values for wet or dry soil.
Soil was modelled by means of permittivity, permeability Taking into account that both configurations used the same
and resistivity values. Two soil models were used to carry out conductor length, it was better to build the second one (wye
the simulation (Table I) (Grcev, 1993). configuration).
Table 1. Soil models
Characteristics Wet soil Dry soil B. Counterpoise effective length
Resistivity 100 1000
Permittivity (relative) 36 9
A simulation was developed to define the maximum length
Permeability (relative) 1 1 of counterpoises to be used in grounding systems; a 100 m,
0.01 m radii counterpoise, buried 0.5 m, was modelled. The
simulations were developed in both soil types at 100 Hz and 2
A. Triangle or wye configurations MHz frequency. The injection point was at the beginning of
the counterpoise. Figure 3 shows current distribution per unit
The first configurations being studied were triangle and
throughout the counterpoise.
wye configurations; they were named according to the
geometric figure formed by their conductors (Figure 1).

6 REVISTA INGENIERÍA E INVESTIGACIÓN Vol. 31 Suplemento No. 2 (SICEL 2011), OCTUBRE DE 2011 (5-10)
 MONTAÑA

1 40
100Hz (Wet) Counterpois
0.9 2MHz (Wet)
35 Rod beginning
100Hz (Dry) Rod end
0.8
2MHz (Dry) Two rods
0.7 30

Impedance (Ohms)
0.6
Current (p.u.)

25
0.5
20
0.4

0.3 15

0.2
10
0.1

0 5
0 10 20 30 40 50 60 70 80 90 100 2 3 4 5 6 7
10 10 10 10 10 10
Length (m) Frequency (Hz)

Figure 3. Current distribution in counterpoise Figure 4. Counterpoise impedance – with and without rods

Figure 3 shows current variation throughout the D. Counterpoises length

counterpoise for two types of soil. Distribution was uniform at
low frequency and had no dependence on soil type; however, The impedance magnitude values for different length
distribution was highly dependent at high frequency. The counterpoises was presented. The cables were modelled
current was scattered during the first meters (around 40m) for having 0.01m radii and buried 0.6 m in wet soil.
wet soil while current was almost zero for lengths greater than 50
70m for dry soil. That meant that transient analysis showed 15m
45
that using greater than 70 m counterpoise length was a waste 30m
of money because the current was going to be scattered during 40 45m
60m
the first meters nearest to the injection point. Furthermore, the 35
inductive effect in very long counterpoises could increase
Impedance (Ohms)

30
impedance magnitude. On the other hand, there is not limit to
counterpoise length in AC analysis because the current is 25
uniformly distributed. However, grounding electrodes are 20
used nowadays in AC and transient at the same time so
maximum counterpoise length should be close to 70 m in high 15

resistivity soils and 40 m in low resistivity soils. 10

5
C. Counterpoise with or without rods 0
2 3 4 5 6 7
The impedance magnitude values of a counterpoise with 10 10 10 10 10 10
Frequency (Hz)
and without rods is presented. The rods were located at the
Figure 5. Impedance of counterpoises
beginning and in the open end. The counterpoises were
modelled with 0.01m radii, 15 m length and buried 0.6 m; the The counterpoise impedance values had no important
rods measured 2.4 m in length and had 0.01 m radii. They variations regarding counterpoise length at high frequencies,
were modelled in wet soil. meaning that the transient response of different counterpoise
lengths was the same. The voltage of four counterpoises was
Figure 4 shows the difference between impedance at the modelled in the time domain to complement this analysis;
injection point when the rods were not included, when the rod current was 1 kA and 1/20 µs. The results are shown in Figure
was included at the beginning, at the open end and at both the 6.
beginning and open ends. The results showed that better
performance was achieved when the rod was included at the Figure 6 shows injection point voltage for each
beginning of counterpoise, because including the rod at the counterpoise. The peak values for four counterpoises were the
open end led to no significant differences at high frequencies. same; the differences shown in the tails of the waveforms
explained because impedance magnitude had variations at low
frequencies but not at high frequencies. It may thus be
concluded that increased counterpoise length modified
transient voltage tail but not the peak.

REVISTA INGENIERÍA E INVESTIGACIÓN Vol. 31 Suplemento No.2 (SICEL 2011), OCTUBRE DE 2011 (5-10) 7
RECOMMENDATIONS FOR GROUNDING SYSTEMS IN LIGHTNING PROTECTION SYSTEMS

between conductors in both directions was 6 m (constant). For

14000
15m
example, 1x1 mesh had one grid (6x6 m), 4x4 had sixteen
12000
30m grids (24x24 m) and 8x8 had sixty-four grids (48x48 m). All
45m meshes were injected at one corner. Figure 8 shows the
60m
10000
results.
25
Voltage (V)

8000
1x1
2x2
6000 4x4
20
8x8

4000

Impedance (Ohms)
15

2000

10
0
0 1 2 3 4 5 6 7 8 9
Time (s) -5
x 10
Figure 6. Counterpoise voltage at injection point 5

E. Voltage throughout the counterpoise 0

2 3 4 5 6 7
10 10 10 10 10 10
Transient voltage is presented at three different points on a Frequency (Hz)
counterpoise measuring 60 m length, 0.01 m radii, buried 0.5 Figure 8. Mesh impedance
m. The simulation was carried out with wet soil parameters in
the time domain when the current was 1 kA and 1/20 µs. From Figure 8, it was concluded that the high frequency
Figure 7 shows transient voltage at the injection point, centre impedance value had no large variation for the four different
point and open end of the counterpoise. meshes being studied; variation took place at low and medium
frequency (up to 100 kHz), meaning that differences were
14000
0m
again presented in the time domain in the tail of the transient
12000 30m voltage.
60m

10000
G. Injection point dependence in a mesh (impedance)
Impedance was simulated at four injection points in a
8000
24x24 m mesh having 16 inner grids to determine injection
Voltage (V)

6000
point dependence in the impedance value (see Figure 9). The
mesh was modelled in wet soil, built with conductors
4000 measuring 0.01 m radii, buried 0.5 m. The impedance for four
injection points are shown in Figure 10.
2000

-2000
0 1 2 3 4 5 6 7 8 9
Time (s) -5
x 10
Figure 7. Voltage at three different points on a 60m counterpoise
Figure 9. Mesh 24x24 m with four injection points.
Based on these results, it is shown that the concept of 25
“equipotentiality” has a different meaning in transient Corner
Inside (6,6)
analysis, since (as shown in Figure 7) there was a voltage Center (12,12)
20
difference between points on the same conductor and the Edge (0,12)

peaks happened at different times. It would thus be advisable

Impedance (Ohms)

to connect different devices at the same grounding system 15

point to avoid large voltage differences which could damage a

device’s insulation. 10

F. Mesh for different areas 5

Different sized meshes’ impedance values are now

presented. The meshes were modelled in wet soil, built with a 0
0.01 m radii conductor, buried 0.5 m. The meshes were 10
2
10
3
10
4
10
5 6
10
7
10
Frequency (Hz)
defined by means of the number of inner grids; the distance
Figure 10. Impedance for 24x24 m mesh with four injection points

8 REVISTA INGENIERÍA E INVESTIGACIÓN Vol. 31 Suplemento No. 2 (SICEL 2011), OCTUBRE DE 2011 (5-10)
 MONTAÑA

Injection point dependence is presented in Figure 10. The

differences were noticeable at frequencies over 10 kHz,
becoming lower when the injection point was located in the
centre of the mesh and higher at the corners. The differences
were thus shown at peak transient voltage not in the tail, as
will be shown in the next section. Notice that the same mesh
may depict diverse performance depending on the injection
point. Figure 13. 24x24 m mesh with one injection point to compute transient
voltage at three different points
H. Injection point dependence in a mesh (voltage)
To support the above conclusion, the same mesh was fed
with 1 kA and 1/20 µ current . The current was injected at the
4000
Center
centre point and in the corner (Figure 11). 3500 Edge
Corner

3000

2500

Voltage (V)
2000

1500

1000

500
Figure 11. 24x24 m mesh having two injection points to compute transient
voltage 0
0 1 2 3 4 5 6 7 8 9
Time (s) -5
x 10
Figure 12 shows the differences between transient voltage Figure 14. Transient voltage in three different points in a mesh 24x24 m.
when the injection point was in the centre and in a corner. The
differences mainly occurred in peak waveform not in the tail; Figure 15. Transient voltage at three different points in a 24x24 m mesh
the differences were almost twice higher when the injection
point was located in a corner. As was concluded in Figure 7, when a grounding electrode
8000 was analysed in the transient domain, it had voltage
Injection:corner differences between points within itself. Figure 14 shows that
7000 Injection:center
the differences were obvious when the observation point was
6000
moved. In this case, the difference between the centre point
and the corner was 1.5 kV when the transient voltage in the
5000 injection point was 3 kV peak.
Voltage (V)

4000

3000 4. CONCLUSIONS
2000
Based on the transient analysis of grounding configurations
it may be concluded that:
1000

0
• When there is not enough area to build grounding
0 1 2 3 4 5 6 7 8 9
Time (s) -5
electrodes, it is better to use a wye configuration instead of
x 10
the triangle configuration;
Figure 12. Transient voltage in two injection points; mesh 24x24 m.
• Effective counterpoise length was close to 70 m in high
resistivity soils and 40 m in low resistivity soils;
I. Voltage difference in a mesh
• The best performance was achieved when a rod was
Continuing mesh analysis, voltages were then found at included at the beginning of a counterpoise, not at the
different points in the same mesh (24x24m) when current was open end;
fed at the centre point to identify voltage difference in the • Increasing counterpoise length modified transient voltage
same system due to transient performance. The transient
tail, not its peak;
voltage was computed at the injection point (centre), in a
• The down conductors had to be connected in the centre of
corner and at the edge of the mesh (Figure 13).
the meshes, not in the corner or at the edges, to reduce
impedance; and

REVISTA INGENIERÍA E INVESTIGACIÓN Vol. 31 Suplemento No.2 (SICEL 2011), OCTUBRE DE 2011 (5-10) 9
RECOMMENDATIONS FOR GROUNDING SYSTEMS IN LIGHTNING PROTECTION SYSTEMS

• When a grounding electrode was analyzed in the transient Montaña, J., Grounding system, Soil electric parameters variation with
frequency and software for computing transient voltages, Thesys, National
domain, it had voltage differences between its points so University of Colombia, Doctor of Philosophy, 2006a.
that it should be mandatory to connect different devices at
the same grounding system point to avoid large voltage Montaña, J., Montanyá J., et al, Quasi-static approximation of concentrated
ground electrodes: experimental results, 28th International Conference on
differences. lightning protection, ICLP, Kanasawa – Japan, 2006b.

5. ACKNOWLEDGMENT PCB-MoM Software Manuals, A method of Moment Program for Radiated

Emission and Susceptibility Analysis of Printed Circuit Board.
The author wishes to express his gratitude to
COLCIENCIAS and Universidad del Norte for financing this Valencia, J., Moreno, G., Modelación de puestas a ierra para evolución de
work. sobretensiones transitorias, Congreso Iberoamericano de Alta Tensión y
Asilamiento Eléctrico, ALTAE, 2003.7

6. REFERENCES Visacro, S.F., Modeling of earthing systems for lightning protection

Casas, F. (ed), Tierras, soporte de la Seguridad Eléctrica, 4a ed., 2008. application, including propagation effects, 21st International Conference
on lightning protection, ICLP, Berlin – Germany, 1992a.
Grcev, L., More Accurate Modeling of Earthing Systems Transient Behavior,
15th Telecommunication Energy Conference, INTELEC, vol. 2, pp. 167- Visacro, S.F., Modelagem de Aterramentos Eléctricos, Thesys, Federal
173, 1993. University of Rio de Janeiro – Brazil, Doctor of Philosophy, 1992b.

IEEE Std. 62305, Protection against Lightning Part I, II, III.

10 REVISTA INGENIERÍA E INVESTIGACIÓN Vol. 31 Suplemento No. 2 (SICEL 2011), OCTUBRE DE 2011 (5-10)

View publication stats