CIGRE C4 Colloquium on

Lightning and Power System,

Kuala Lumpur, 16 - 19 May 2010

LINE SURGE ARRESTERS ENERGY DUTY CONSIDERATIONS

S. Sadovic*, D. Lepley**, E. Brocard**, J.M. George**

SUMMARY
This Paper presents Line Surge Arrester [LSA] energy duty considerations. Shielded and
unshielded lines are considered. LSA current shapes are computed and presented for both line designs.
Energy duties for the first and for the subsequent strokes are presented.
There is a big difference in the energy duties for the LSA installed on the shielded and on the
unshielded lines. LSA installed on the unshielded line are much more stressed than LSA on the
shielded line.
Statistical distribution of LSA currents is computed and presented in the form of the cumulative
frequency distributions. Sigma slp software is used for the all simulations. LSA current distributions
are presented for the different tower footing resistance values and for the different line designs.
Several single stroke case studies are also performed in order to present different LSA current
shapes. It is indicated that single stroke hitting transmission line usually produce operation of more
than one LSA.

KEYWORDS
Line lightning performance, Line surge arrester, Tower footing resistance, Shielded line, Unshielded
line
(*) Sadovic Consultant, France ([email protected])

(**) Sediver S.A., France ([email protected], [email protected], [email protected])

1. INTRODUCTION
The use of line surge arresters for the quality of service improvement has increased over the last
decade [1], [2], [3], [4]. Line surge arresters (LSA) are mainly used for the transmission line lightning
performance improvement and for the reduction of double circuit outages on double circuit lines [5].
Many line surge arresters are in service today and substantial service experience has been
accumulated.
In the selection of the line surge arrester (LSA) for the line lightning performance improvement it
is very important to determine arrester energy duty. LSA energy duty depends on many factors, like:
line design, lightning data in the region where line will be operating, tower footing resistance, LSA
installation configuration, LSA design,
LSA are differently stressed on the shielded and on the unshielded line. In the case of the shielded
line a large fraction of the lightning stroke current will be diverted to the ground through the shield
wires and transmission line towers. In the case of the unshielded line, the majority of the lightning
strokes will terminate on the phase conductors. When lightning stroke hits directly phase conductors,
all energy related to the particular stroke has to be discharged by LSA.
In this paper LSA energy duties for the shielded and for the unshielded lines are considered. LSA
current shapes are presented for these two line designs.
Statistical distribution of LSA current peaks is computed by the use of sigma slp software. The
corresponding distributions are presented for the shielded and for the unshielded lines.
2. LINE SURGE ARRESTER CURRENT SHAPES
LSA energy duty is mainly related to the lightning stroke parameters (peak, shape, polarity,
multiplicity), LSA installation configuration, line design (shielded or unshielded), tower footing
resistance and stroke location.
When lightning stroke hits shield wire or tower top (in the case of the shielded line) lightning
stroke current is diverted along shield wires, tower construction and tower footing resistance. When
LSA are installed on the shielded line then only a fraction of the lightning stroke current is flowing
through LSA. In addition, LSA current tail time is shorten than original lightning stroke tail time
(duration of LSA current surge is shorter than the duration of the original lightning stroke).
Situation is completely different in the case of the unshielded lines equipped with the LSA. When
lightning stroke hits directly the phase conductor the energy related to the lightning stroke has to be
discharged through the line surge arresters installed on the stricken phase conductor. The most stressed
are the LSA, which are close to the place where lightning terminates. In general, LSA installed on the
unshielded lines are much more stressed than arresters on the shielded lines. LSA currents on the
unshielded lines have higher currents peaks and the longer durations.
When the lightning stroke hits phase conductor in the case of the shielded line (shielding failure),
duration of the LSA current is similar to the duration of the LSA currents in the case of the unshielded
lines, but current peaks are lower (shielding failures happen for the low amplitude lightning strokes).
In order to compare LSA current shape for different line design (shielded and unshielded) 123 kV
lines are considered. Line general data is given in Figure 1 (unshielded line has only three conductors no shield wire - conductor No 4).
Line insulation critical flashover voltage for both lines is 550 kV. IEC Class II polymer housed line
surge arresters are used. LSA rated voltage is 108 kV. In the case of the shielded line LSA were
installed on the bottom phase conductor, while for the unshielded line LSA were installed on the top
phase conductor.

Shielded line conductor data:

Cond No
1
2
3
4

x (m)
2,5
-3
3,5
0

y(m)
22,7
20,5
18,3
28,9

r (mm)
8,54
8,54
8,54
4,5

Sag(m)
8,6
8,6
8,6
7,6

LSA

LSA
Tower height:
Tower surge impedance:
Line span:
Tower footing resistance:
Soil resistivity:

28,9 m
184,7
305 m
40
1200 m

a) Shielded line

b)Unshielded line

Figure 1 - Shielded and unshielded 123 kV lines

In order to compare LSA current shapes, two lightning strokes are considered in the study:
a) Lightning stroke, which has median parameters from the first stroke distributions for negative
downwards flashes [7]
b) Lightning stroke, which has median parameters from the subsequent stroke distributions for
negative downwards flashes [7]
Stroke parameters are presented in Table 1.
Table 1 - Parameters of the strokes used in the study

First Stroke
Subsequent stroke

I0 (kA)
31,1
12,3

tf (s)
3,83
0,67

tt (s)
77,5
30,2

I0 - Stroke peak (kA)

tf - Stroke front time (s)
tt - Stroke tail time (s)
In the case of the shielded lines lightning strokes are applied to the tower top, while for the
unshielded line strokes are injected to the middle of the span (top phase conductor) - Figure 1.
Line surge arrester current shapes for the shielded line are presented in Figure 2. From this Figure
we can see that the arrester current peaks are not so high, compared to the peaks of the original
lightning strokes (more that ten times lower compared to the original lightning strokes).
It is also important to note that the durations of the arrester currents are much shorter than the
durations of the original lightning strokes (almost four times shorter than the original lightning stroke).
The associated energy duties of LSA for this studied case are given in Table 2 (7,7 kJ and 0,6 kJ
respectively). The energy capability of the considered LSA is 450 kJ.

If (kA)

Is (kA)

2,5

Current (kA)

First stroke
1,5

Subsequent stroke

0,5

0
0

10

15

20

25

30

35

Time (microsec)

Figure 2 - Shielded line: Arrester current for the first and for the subsequent stroke

Line surge arresters current shapes for the unshielded line are presented in Figure 3. From the
presented results we can see that surge arrester current peaks are much higher than in the case of the
shielded line (for the same original lightning strokes).

Table 2 - Arrester energies W (kJ)

First Stroke
Subsequent stroke

Shielded line
W (kJ)
7,5
0,6

Unshielded line
W (kJ)
83,2
15,8

The duration of the arrester currents is also much longer than for the shielded line case (being very
close to the durations of the original lightning strokes).
The corresponding arrester energy duties are 83,2 kJ for the first stroke and 15,8 kJ for the
subsequent stroke (Table 2). These values are much higher than in the case of the shielded line.

40

If (kA)

Is (kA)

16
14
12
Current (kA)

First stroke

10
8
6
Subsequent stroke

4
2
0
0

10

20

30

40

50

60

70

80

90

100

110

Time (microsec)
Figure 3 - Unshielded line: Arrester current for the first and for the subsequent stroke

3. STATISTICAL STUDY OF LINE SURGE ARRESTER CURRENTS: SHIELDED LINE

Statistical distributions of LSA current peaks are computed using sigma slp software. Shielded line
design corresponds to the previously presented line. Tower footing resistance was varied from 10 to
100 . For each studied case the total number of 1000 statistical simulations is performed. Two lines
CIGRE stroke distribution, modified for the flat ground, is used.
For each statistical case the highest LSA current peak is extracted. The cumulative frequency
distributions of the highest LSA current peaks are presented in the following tables:
Table 3: All three phase conductors LSA installation configurations [LSA 1,2,3]
Table 4: Middle and bottom phase conductors LSA installation configurations [LSA 2,3]
Table 5: Bottom phase conductor LSA installation configurations [LSA 3]

120

Table 3 - Shielded line: Cumulative frequency distributions of LSA Currents

Arrester configuration LSA 1,2,3 [All three phase conductors]

RT ()
10
20
30
40
50
60
70
80
90
100

I1% (kA)
3,36
6,65
8,91
10,93
11,52
12,33
12,97
13,49
13,92
14,29

LSA Current cumulative frequency distributions (%)

I2% (kA)
I5% (kA)
I10% (kA)
I25% (kA)
2,75
1,62
1,10
0,61
4,88
2,78
2,00
0,95
6,86
3,83
2,84
1,33
8,21
4,66
3,45
1,71
9,22
5,38
3,99
2,05
9,94
6,06
4,43
2,35
10,51
6,53
4,80
2,57
10,97
6,89
5,08
2,74
11,34
7,17
5,29
2,92
11,68
7,39
5,51
3,07

I50% (kA)
0,29
0,49
0,65
0,80
0,92
1,05
1,65
1,26
1,34
1,45

Table 4 - Shielded line: Cumulative frequency distributions of LSA Currents

Arrester configuration LSA 2,3 [Middle and bottom conductor LSA installation]

RT ()
10
20
30
40
50
60
70
80
90
100

I1% (kA)
3,89
7,75
9,61
10,91
11,64
12,17
12,66
13,13
13,50
13,86

LSA Current cumulative frequency distributions (%)

I2% (kA)
I5% (kA)
I10% (kA)
I25% (kA)
3,04
1,82
1,23
0,66
5,43
3,41
2,37
1,08
7,55
4,51
3,32
1,57
8,61
5,35
4,05
2,01
9,48
6,02
4,58
2,44
10,33
6,63
4,97
2,74
10,86
7,10
5,35
3,00
11,08
7,53
5,66
3,15
11,28
7,86
5,75
3,31
11,48
7,96
6,00
3,47

I50% (kA)
0,31
0,55
0,75
0,93
1,12
1,28
1,41
1,54
1,63
1,73

Table 5 - Shielded line: Cumulative frequency distributions of LSA Currents

Arrester configuration LSA 3 [Bottom conductor LSA installation]

RT ()
10
20
30
40
50
60
70
80
90
100

I1% (kA)
4,53
8,16
10,12
10,81
11,59
11,91
12,61
13,14
13,80
13,75

LSA Current cumulative frequency distributions (%)

I2% (kA)
I5% (kA)
I10% (kA)
I25% (kA)
3,65
2,10
1,36
0,70
6,69
4,00
2,80
1,26
8,20
5,21
3,92
1,95
9,51
6,40
4,65
2,56
10,04
6,90
5,14
2,98
10,08
7,56
5,77
3,25
10,16
7,95
6,08
3,48
10,37
8,23
6,59
3,71
10,82
8,51
6,69
3,85
11,22
8,68
6,89
4,04

I50% (kA)
0,34
0,62
0,88
1,15
1,43
1,63
1,77
1,93
2,05
2,21

It is important to note that for the less than all phase conductors LSA installation configurations
[for LSA 2,3 and LSA 3] and for the high values of the tower footing resistance insulation flashovers
happen on the conductors without LSA.
If we take 2% values for the comparison purposes then we can conclude:
The maximum 2 % LSA current peaks are lower than 12 kA [for all studied LSA installation
configurations and for the different tower footing resistances].
2 % LSA current peaks increase with the increase of the tower footing resistance.
For the lower values of the tower footing resistance, there are no insulation flashovers and LSA
currents are lower for more LSA installed [stroke current is better shared through more LSA installed
on the same towers].
Each lightning stroke hitting the line usually produces the operation of several LSA. As indicated
before for each lightning stroke hitting the line the maximum LSA current for all installed LSA is
extracted. In Figures 4 and 5 LSA current shapes are presented for the stroke hitting tower top and the
stroke hitting the phase conductor [shielding failure].

75 kA, tf=5 s

Figure 4 - LSA current shapes: Shielded line > Stroke to the tower top
Stroke current peak 75 kA and stroke front time of 5 s
Tower footing resistance 30
In Figure 4 lightning stroke having current peak of 75 kA, front time 5 s and tail time of 75 s hits
tower top of the shielded line. Tower footing resistance is 30 [soil resisitivity 900 m].
This stroke produces operation of the LSA on the struck tower but also operation of the LSA on the
neighbouring towers.
From the presented shapes we can see that LSA currents tail times is around 25 s.

15 kA, tf=2,5 s

Figure 5 - LSA current shapes: Shielded line > Stroke to the top phase conductor
Stroke to the 1/4 of the span length
Stroke current peak 15 kA and stroke front time of 2,5 s
Tower footing resistance 30

4. STATISTICAL STUDY OF LINE SURGE ARRESTER CURRENTS: UNSHIELDED LINE

The unshielded line used for the simulations is the same as that one used before but with the
removed shield wire. LSA surge arresters are installed on the top and on the middle phase conductors.
Tower footing resistance was varied from 10 to 100 . For each studied case the total number of
1000 statistical simulations is performed. Two lines CIGRE stroke distribution, modified for the flat
ground, is used.
For each statistical case the highest LSA current peak is extracted. The cumulative frequency
distributions of the highest LSA current peaks are presented in Table 6.

Table 6 - Unshielded line: Cumulative frequency distributions of LSA Currents

Arrester configuration LSA 2, 3 [Top and middle conductor LSA installation]

RT()
10
20
30
40
50
60
70
80
90
100

I1% (kA)
48,94
47,36
46,23
45,41
45,86
45,38
45,02
44,73
44,49
43,98

LSA Current cumulative frequency distributions (%)

I2% (kA)
I5% (kA)
I10% (kA)
I25% (kA)
39,08
31,84
24,11
13,93
38,15
31,16
23,61
13,86
37,29
30,07
23,23
13,92
36,66
30,02
22,95
13,90
36,15
29,44
22,66
13,83
36,02
29,10
22,41
13,79
36,32
29,45
22,19
13,79
36,05
28,95
21,99
13,72
36,08
29,14
21,88
13,67
35,76
29,57
21,87
13,60

I50% (kA)
3,83
5,46
6,34
6,93
7,20
7,44
7,56
7,66
7,75
7,80

From the presented results we can see that 2 % LSA maximum currents might be close to 40 kA.
LSA currents are slightly reduced with the increase of the tower footing increase because:
-

With the increase of the tower footing resistance less current is flowing through the tower
footing resistance and more current has to be diverted through LSA.
For the high values of the tower footing resistance insulation flashovers may happen on the
bottom conductor [reducing current through the LSA].

In Figures 6 and 7 LSA current shapes are presented for the stroke hitting the phase conductor and
tower top.

75 kA, tf=5 s

Figure 6 - LSA current shapes: Unshielded line > Stroke to the top phase conductor
Stroke to the 1/4 of the span length
Stroke current peak 75 kA and stroke front time of 5 s
Tower footing resistance 30
In Figure 6 lightning stroke having current peak of 75 kA, front time 5 s and tail time of 75 s hits
top phase conductor. Tower footing resistance is 30 [soil resisitivity 900 m]. Stroke distance is
one quarter of the span length. LSA current shapes for the top LSA are presented.

75 kA, tf=5 s

Flashover

Flashover

Figure 7 - LSA current shapes: Unshielded line > Stroke to the tower top
Flashover on the bottom conductor insulator
Stroke current peak 75 kA and stroke front time of 5 s
Tower footing resistance 30
In Figure 6 lightning stroke having current peak of 75 kA, front time 5 s and tail time of 75 s hits
tower top of the unshielded line. Tower footing resistance is 30 [soil resisitivity 900 m].
This lightning stroke produces also flashover on the bottom conductor of the struck tower and also
flashover on the neighbouring tower.

LSA current shapes on the struck and on the neighbouring tower are presented.

5. CONCLUSIONS
1.
2.

3.
4.
5.

In the design and the selection of the Line Surge Arresters, it is very important to know the
shapes of the current flowing through the arresters.
Line Surge Arresters installed on the shielded and on the unshielded line are differently stressed.
LSA on the unshielded line are more stressed than arrester on the shielded line. The main reason
for this is in the difference in the arresters current shapes.
The shapes of the surge arrester current can be determined by the software simulations and by the
field measurement.
Lightning stroke hitting transmission line usually produces operation of several line surge
arresters.
In addition to the line design [shielded or unshielded] line surge arrester installation configuration
and tower footing resistance influence arrester energy duty.

6. REFERENCES
[1] S. Sadovic, R. Joulie, S. Tartier, Transmission Lines Lightning Performance Improvement by
the Installation of Line Surge Arresters , Ninth International Symposium on High Voltage
Engineering, Graz Austria 1995, paper 6731
[2] S. Sadovic, R. Joulie, S. Tartier, E. Brocard Use of Line Surge Arresters for the Improvement of
the Lightning Performance of 63 kV and 90 kV Shielded and Unshielded Transmission Lines, ,
IEEE Transactions on Power Delivery, vol. 12, no. 3, July 1997, pp. 1232 - 1240
[3] S. Sadovic, R. Joulie, S. Tartier, E. Brocard Line Surge Arresters and Unbalanced Insulation in
the Reduction of Double Circuit Outages on a 225 kV Transmission Line, X International
Symposium on High Voltage Engineering, August 25-29, 1997, Montreal, Canada
[4] M. Babuder, M. Kenda, P. Kotar, E. Brocard, S. Tartier, R. Joulie, S. Sadovic, Lightning
Performance Improvement of 123 kV Transmission Line by use of Line Surge Arresters, XI
International Symposium on High Voltage Engineering, August, 1999, London, U.K.
[5] Y. Wahab, Z. Abidin, S. Sadovic Line Surge Arrester Application on the Quadruple Circuit Transmission Line, Bologna PowerTech 2003 International Conference, June 2003, Bologna, Italy
[6] sigma slp: Software for the determination of the transmission on the distribution line lighnting
performance, with special reference to the application of line surge arresters, www.sadovic.com
[7] CIGRE WG 33.01, Guide to Procedures for Estimating the Lightning Performance of
Transmission Lines, CIGRE Technical brochure no 63, October 1991
[8] S. Sadovic, G. Couret, Z. Abidin, M. Puharic, L. Peter Quality of the Service Improvement of
the Compact Lines by the Use of Line Surge Arresters, Bologna PowerTech 2003 International
Conference, October 2005, Cape Town, South Africa
[9] S. Sadovic, M. Muhr, T. Sadovic Line Surge Arrester Energy Duty Considerations on the
Compact Unshielded Transmission Lines - 15th International Symposium on High-Voltage
Engineering, ISH 2007, Ljubljana, Slovenia, August 27-31, 2007.