Horn Length Estimation for Decrease of Tripout Rates on 150 kV

Transmission Lines in West Sumatra in Indonesia

Yusreni Warmi (Shizuoka University, Padang Institute of Technology)

Koji Michishita (Shizuoka University)

Abstract
The authors have carried out investigation of lightning tripout of 150 kV transmission lines in West Sumatra in Indonesia. The results of the
investigation in this paper show a positive correlation between the number of lightning tripouts on a 150 kV transmission line and the tower-footing
resistance, and also reveal that the main cause of the tripouts is back-flashover. Moreover, it turns out that the flashover at the lower arm increases
in case of high tower-footing resistance. The result of the investigation is in good agreement with the results of a simulation by the IEEE FLASH
program. As the result, it is recommended that the length of an arching horn, in the 150 kV line in West Sumatra, is 1.1 m. When the tower-footing
resistance at all towers is arranged to be 10  and the horn length is arranged 1.1 m, the critical current becomes about 80 kA. The tripout rate is
expected to be less than 15 tripout/ 100 km/ year.

Keywords : Lightning tripouts, tower-footing resistance, transmission line, IKL

studied line is a 150 kV double-circuit, balanced-insulation, and

1. Introduction
transposed transmission line. The range of the span length was 147
A 150 kV transmission line from Payakumbuh to Koto Panjang m to 434 m with an average of 333 m.
in West Sumatra passes the area whose average number of Figure 1 shows a map of the IKL of West Sumatra and location
thunderstorm days per year (IKL: Isokeraunic levels) reached up to of a 150 kV overhead transmission line between Payakumbuh -
165, and the frequency of direct lightning strokes to the Koto Panjang under study, shown as the solid line in the middle,
transmission line was high (1). with the total length of 86 km. Due to the tropical climate in
Lightning performance on a transmission line has been studied Indonesia, the maximum temperature ranges from 33° C to 36° C
for a long time. According to (2)-(4), the average ratio of lightning all year round, and thunderstorm days are as many as 165 days/year
tripouts of 110 kV to 138 kV lines in China, Mexico, and Malaysia in the area where a part of the line between No. 1 to No. 140 towers,
for five years, reached about 62%, 50%, and 70%, respectively. In 47 km in length, locates. Weather and lightning strokes in this area
Japan, the ratio of lightning tripouts of 110 – 154 kV lines was 75% are influenced by the movement of the regional wind (the monsoon
(5). Lightning tripouts mean the operation of a circuit breaker at the and the trade winds) and the local winds (sea breeze/land breeze
substation due to flashover caused by lightning (6). Furthermore, it and valley breeze/mountain wind). Thunderstorm in this area often
is shown that the tower-footing impedance has a great influence on moves from the northeast to the southwest.
the back-flashover protection performance: the lower the tower-
footing impedance, the less the back-flashover rate (6)-(12).
In this paper, lightning performance on the 150 kV transmission
line in West Sumatra is studied. To the best of our knowledge, this
is the first time that the lightning performance of the transmission
line in Indonesia is discussed. The positive correlation between the
tower-footing resistance and the number of lightning tripouts is
confirmed, and the main cause of the tripouts is inferred to be the
back-flashover. Moreover, it turns out that the flashover at phase
wires locating at the lower arm increases in case of high tower-
footing resistance. It is shown that the observed lightning tripout
rates are in good agreement with the simulation results by the IEEE
FLASH program by using average footing resistance at the towers
where the lightning tripout occurred. As the result, it is
recommended that arching horn length in the 150 kV line in West
Sumatra is 1.1 m. When the tower-footing resistance at all towers
is arranged to be less than 10 , and the horn length is arranged 1.1
m, the tripout rate is expected to decrease to half of the present rate.
2. Transmission line under study
Fig. 1. Map of Isokeraunic Level of West Sumatra and location
2.1 Thunderstorm Activity in West Sumatra The of transmission line.
 Figure 2 shows the configuration of towers. The towers consist
of four types, namely A, B, C and D types. Table 1 shows the
dimension, the component ratio and the frequency ratio of the
lightning tripouts dependent on tower types.
2.2 Tower-footing Resistance Figure 3 shows the
average tower-footing resistance dependent on the tower-number
group before and after the improvement of the tower-footing
resistance, carried out from 2010 to 2014. The average tower-
footing resistance before and after the improvement of the tower-
footing resistance was in the ranges from 24 to 50  and from 4 to
6 , respectively.

Fig. 3. Average value of tower-footing resistance before and after

Fig. 2. Configuration of towers. improvement.

The line had two galvanized overhead ground wires (OHGW),

9.6 mm in diameter, and the sag was 1.5% times the span length of
the towers. Each phase conductor was the ACSR (Aluminum
Conductor Steel Reinforced), 25.5 mm in diameter, and the sag was
2% times the span length of the towers. The string of insulators
made of porcelain had the BIL (Basic Insulation Level) of 1.21 MV.
In addition to the ground wires, transmission towers had other
protection systems such as arching horns, TLAs (Transmission Line
Arrester), down conductors and the grounding system. Each tower
had arching horns with the horn length of 0.9 - 1.6 m. The 4 to 6
pieces of the TLAs were installed at 10 towers among No.1 to
No.140 towers dependent on the frequency of the lightning strokes.
There were down conductors and grounding systems at 36 towers
(13).
3. Lightning Tripout
3.1 Cause of Tripouts Based on the data for 5 years from
2010 to 2014 provided by P3B Sumatra and Padang UPT, the causes
of tripouts on the 150 kV transmission line are shown in Figure 4
(13). The most frequent cause of tripouts on the line is lightning,
amounting to 66% of all tripouts (13). The ratio of lightning tripouts
are almost the same as in (2)-(5). The high number of the lightning
tripouts occurred in April 2011 and November 2012, and these
months had many thunderstorm days (14).

Table 1. Dimension, component ratio and frequency of

lightning tripout dependent on tower types.
Tower Types
A B C D
d OHGW 7.0 6.8 6.8 7.0
d Upper 7.6 7.0 7.0 7.6
d Middle 8.0 7.4 7.4 7.6
d Lower 8.45 7.8 7.8 7.6
Fig. 4. Tripout causes.
d Tower 5.0 5.4 5.4 5.6
h OHGW 32.2 31.7 31.7 31.7 3.2 Tripout Rate The tripout rates of the line between No.
h Upper 28.1 27.7 27.7 27.7 1 to No. 140 towers before and after the improvement of the tower-
h Middle 23.8 23.6 23.6 23.6 footing resistance, and those were 114 and 22 tripouts /100 km/year,
respectively.
h Lower 19.5 19.5 19.5 19.5
According to (15), the average lightning tripout rates of 110 kV
Component ratio
53.6 28.6 10.7 7.1 to 154 kV lines from 1980 to 2000 was 3.25 tripouts/100 km/year
(%)
in Japan where the average IKL was from 20 to 30. The high tripout
Frequency ratio of rate in Indonesia even after the improvement of the tower-footing
60.5 30.2 8.1 1.2
tripout (%) resistance might be due to the high IKL and the short arching horn
length.

2
 Figure 5 shows the rate of lightning tripouts for tower-number
groups before and after the improvement of tower-footing
resistance. The location of the flashover resulting in tripouts was
estimated by the fault locator. The highest rate of lightning tripouts
before and after the improvement of tower-footing resistance, being
found at No. 1 to No. 17 towers, were 234 and 57 tripouts/100
km/year, respectively. This is due to the geographical condition that
these towers stand on the hill about 225 m above the sea level in the
north where thunderstorm activity is high as shown by the IKL in
Figure 1. Before the improvement of tower-footing resistance, the
highest tripout rate was found at No. 16 tower with the tower-
footing resistance of 35  and the average tower-footing resistance
at the towers experiencing lightning tripouts was 48 . After the
improvement of the tower-footing resistance, the high tripout rate
occurred at No. 47 tower with the tower-footing resistance of 15 
and the average tower-footing resistance of the towers experiencing
lightning tripouts, was 11 . The tower-footing resistance at the
towers experiencing tripout is more than 1.5 times as high as the
average tower-footing resistance.
Fig. 5. Rate of tripouts for tower-number groups before and after
improvement of tower-footing resistance
Figure 6 shows the tripout rate of the lines extending from the
east to the west, namely the line at No. 1 to No. 37 towers, and the
line from north to the south, namely the line from No. 38 to No. 140
towers. The high rate of lightning tripouts occurs on the line from
No. 1 to No. 37 towers, extending from the east to the west. This is
due to the fact that the lightning is active at the north area as shown
in Figure 1.
Fig. 6. Rate of tripouts for tower-number group.
dependent on line location.
3
Figure 7 shows the tripout rate dependent on the arm location.
The high rate of lightning tripouts before the improvement of the
tower-footing resistance arises at the lower arm. After the
improvement of tower-footing resistance, the high tripout rate
occurs at the upper arm. Since the location of the tripout is not
influenced by the tower-footing resistance and depends on the
struck phase in case of shielding failure, the main cause of the
flashover is inferred to be the back-flashover.
Fig. 7. Rate of tripouts depend on arm location.
By numerical simulation, it was pointed out that the flashover
frequently occurred at the upper and lower arms for the tower-
footing resistance of 10 and 100 , respectively (16). The result of
the investigation in this paper qualitatively agrees with the results
in (16). However, the average tower-footing resistance measured at
the frequency of 151 Hz before improvement of the tower-footing
resistance in Indonesia is less than 100 . In the case of inductive
response, the tower-footing impedance increases in the frequency
range of lightning compared with the resistance at the low
frequency (9), (17)-(18). This might be the possible cause of the above-
mentioned difference.
Figures 8 shows tower-footing resistance and the ratio of
lightning tripouts at the lower arms before and after the
improvement of tower-footing resistance.
Fig. 8. Tower-footing resistance and ratio of tripouts at lower arm
before and after improvement of tower-footing resistance.
The high correlation between the rate of lightning tripouts at the
lower arm and the tower-footing resistance was found at the towers
with the tower-footing resistance of more than 30  before the
improvement of tower-footing resistance. Such a trend cannot be Figure 1. In this way, the ground flash density is calculated to be
found after the improvement of tower-footing resistance, in the 12.7 flashes/km2 /year.
resistance range from 4 to 15 .
Figure 9 shows tower-footing resistance and the arching horn
𝑁𝑔 = 0.04 𝑇𝐷1.25 ............................................................ (2)
length in case of damage of insulators due to lightning after the
improvement of tower-footing resistance. The insulator damages
occur at the towers with the horn longer than 1.4 m regardless of
where 𝑵𝒈 is the ground flash density (flashes/𝐤𝐦𝟐 /𝐲𝐞𝐚𝐫), 𝑻𝑫 is
the tower-footing resistance.
the thunderstorm days (days/year).
Lightning incidence to overhead lines was calculated to be 293
flashes/100 km/year through (3) (21)–(23).

𝑁𝑔
N𝑠 = (28ℎ𝑡0.6 + 𝑑𝑂𝐻𝐺𝑊 ) ............................................. (3)
10

where N𝑠 is the number of stroke incidence to a line (flashes/100

km/year), ℎ𝑡 is tower heights (m), 𝑑𝑂𝐻𝐺𝑊 is overhead ground
wire separation distance (m) in Figure 2.
The cumulative frequency distribution of the lightning current is
assumed to be given by (4) (20)-(24).

1
P(I) = ................................................................. (4)
𝐼 2.6
1+( )
31
Fig. 9. Tower-footing resistance and arching horn length at
towers with insulator damage after improvement of tower-
where I is lightning current amplitude in units of kA, P is
footing resistance.
probability of occurrence of lightning current amplitude higher than
4. Analysis I.
.
The back-flashover rate BFR is given as the product of this
4.1 Method of Analysis The authors estimated the
probability and the number of strokes that terminate on the towers.
lightning tripout rate by the IEEE FLASH program (19) and by the
For simplicity, the number of strokes to the towers is assumed to be
analytical method (20) to study the tripout rate dependent on the
60% of all strokes to the overhead wires (23), (25).
location of phase conductors.
4.2 Line Model and Lightning Incidence to
BFR = 0.6 𝑁𝑠 𝑃(𝐼𝐶 ) ....................................................... (5)
Transmission Line A 47 km – long double-circuit line with the
span length of 333 m, simulating the line from No. 1 to No. 140
where 𝐼𝐶 is the critical current resulting in back-flashover (23).
towers, was selected to estimate the lightning performance of the
transmission line under study. The surge impedance of the towers 5. Results and Discussion
calculated by using (1) (20) is shown in Table 2. Regardless of the
5.1 Tripout Rates Figure 10 shows the comparison of
tower types, values of surge impedance are almost the same.
investigation results with the calculation by the analytical method
Table 2. Surge impedance dependent on tower types. in (20) and by the IEEE FLASH program version 1.81 (19) with the
Surge average tower-footing resistance before and after the improvement
Type of Tower Height Tower Base
Impedance of the tower-footing resistance at No. 1 to No. 140 towers of 33 and
Tower (ℎ𝑡 ) (m) (𝑟𝑇𝑜𝑤𝑒𝑟 ) (m)
(𝑍𝑡 ) () 5.6 , respectively. The tripout rates by the analytical method in
A 32.2 2.5 174
(20) or the IEEE Flash program is smaller than the investigated
result, and such a trend is significant after the improvement of the
B 31.7 2.7 169
tower-footing resistance. The calculated lightning tripout rate
C 31.7 2.7 169 dependent on flashover location by the analytical method is also
D 31.7 2.8 167 less than the investigated results before and after the improvement
of the tower-footing resistance.
Figure 11 shows the comparison of investigation results with
2(ℎ𝑡2 +(𝑟𝑇𝑜𝑤𝑒𝑟 )2 )
𝑍𝑡 = 30 𝑙𝑛 [ ] ............................................. (1) the calculation by the analytical method in (20) and by the IEEE
(𝑟𝑇𝑜𝑤𝑒𝑟)2
FLASH program with the average tower-footing resistance of
towers experiencing lightning resulting in tripouts.
where 𝑍𝑡 is the surge impedance of the tower (), ℎ𝑡 and
The average tower-footing resistance before and after the
𝑟𝑡𝑜𝑤𝑒𝑟 is the tower height and the tower base radius (m) in Figure 2,
respectively. improvement were 48 and 11 , respectively. The results of the
investigation are almost the same with the simulation results. These
The ground flash density is calculated from the IKL by (2) (21)-
(23). The No. 1 to No. 140 towers located in a high lightning activity results show that the overall performance of an entire transmission
line was influenced by the individual performance of the tower (22).
area. Therefore, the thunderstorm day is assumed 100 days/year
The calculated flashover rate dependent on the location reasonably
almost equal to the average values of the IKL at two sections in

4
agrees with investigated results in Figure 11.
Fig. 10. Rate of lightning tripouts calculated by using
average tower-footing resistance at No. 1 to No. 140
towers.
Fig. 11. Rate of lightning tripouts estimated by using average
tower-footing resistance at the towers experiencing lightning
tripouts.
5.2 Proposal for Decrease of Tripout Rates Figure 12
shows the tripout rate calculated by using the IEEE FLASH
program dependent on the tower-footing resistance from 1 to 15 
with the horn length from 0.9 – 1.3 m.
Fig. 12. Rate of tripouts depend on tower-footing
resistance and horn length.
5
The tripout rate increases with the decrease of the horn length
and the increase of tower-footing resistance. When the tower-
footing resistance at all towers is reduced to 10 , as is regulated in
Japan, the tripout rate is expected to decrease from 29 tripouts/ 100
km/ year to 26 tripouts/ 100 km/ year.
Figure 13 shows the critical flashover voltage (CFO)
calculated by using the IEEE FLASH program for the horn length
from 0.9 – 1.3 m resulting in the flashover at the arching horn. The
flashover at the surface of the insulators occurs when the CFO
becomes equal to or higher than the insulation strength under
standard lightning impulse voltages (BIL).
From the result in Figure 13, it is recommended aching horn
length in the 150 kV Payakumbuh – Koto Panjang line is 1.1 m to
avoid the insulator damages. This result is supported by the result
in Figure 9 where insulator damages occurred at the arching horn
longer than 1.4 m regardless of tower-footing resistance. When the
tower-footing resistance at all towers is arranged to be less than 10
 and the horn length is arranged 1.1 m, the tripouts rate is expected
to be less than 15 tripout/ 100 km/ year, about half of the present
tripout rate.
Fig. 13. Critical flashover voltage (CFO) depend on tower-footing
resistance and horn length.
Fig. 14. Critical current and tripout rate depend on tower-footing
resistance for horn length of 1.1 m.
Figure 14 shows that the critical current and the tripout rate
depend on tower-footing resistance calculated by using the
analytical method (21) for the horn length of 1.1 m. The result shows
that when the tower-footing resistance at all towers is arranged to
be less than 10 , the critical current is about 80 kA, and the tripout
rate is expected to be less than 15 tripout/ 100 km/ year. (14) Y. Warmi and K. Michishita, “A Study on Lightning Outages on 150 kV
Transmission Line of Payakumbuh – Koto Panjang in West Sumatra in
6. Conclusion Indonesia,” in Proc. 19th ISH August 2015, number. 170.
(15) Subcommittee for transmission lines, Lightning protection design committee,
Based on the investigation of the 150 kV transmission line in “Guide to Lightning Protection Design for Transmission Line, ”CRIEPI
Report T-72 2002 (in Japanese).
West Sumatra in Indonesia, the following insights are obtained:
(16) Shozo Sekioka,“A Study on Lightning Surge with Special Reference to
1. The rate of lightning tripouts before and after the Multiphase Flashover,” Doctoral Thesis, Faculty of Engineering Doshisha
improvement of tower-footing resistance are 114 and 22 University, Japan, March, 1997.
tripouts/100 km/year, respectively. (17) K. Michishita, M. Ishii, A. Asakawa, S. Yokoyama, and K. Kami, “Voltage
induced on a test distribution line by negative winter lightning strokes to a
2. The rate of lightning tripouts at the lower arm increases tall structure,” IEEE Trans. Electromagn. Compat., vol. 45, no. 1, pp.135 –
with the increase of the tower-footing resistance. The high 140, February 2003.
rate of lightning tripouts before the improvement of the (18) W. A. Chisholm and W. Janischewskyj, “Lightning surge response of ground
electrodes,” IEEE Trans. Power Delivery, vol. 4, no. 2, pp.1329 – 1337, April
tower-footing resistance arises at the lower arm, with the
1989.
tower-footing resistance being more than 30 . After the (19) T. E. Mc Dermott, “Using IEEE Flash to estimate transmission and
improvement of tower-footing resistance, the high tripout distribution line lightning performance,” Proc. IEEE Power Eng. Soc.
rate occurs at the upper arm. Transm. Distrib. Conf., vol. 15236, pp. 8–9, May 2012.
(20) Anderson, J.G., ”Lightning Performance of Transmission Line,”
3. The result of the investigation in this paper reasonably Transmission Line References Book 345 and Above, California, 1982, pp.
agrees with the calculated results by using average tower- 545–597.
footing resistance of the towers experiencing lightning (21) IEEE Working group report Lightning Performance of Transmission Lines,
“IEEE Working Group Report – Estimating Lightning Performance of
tripouts. Transmission Line II – Updates to Analytical Models,” IEEE Trans. Power
4. The recommended arching horn length, in the 150 kV line Deliv., vol. PWRD-8, no. 3, pp. 1254 – 1267, July 1993.
in West Sumatra, is 1.1 m. When the tower-footing (22) IEEE Working group Estimating Lightning Performance of Transmission
Lines, “A simplified Method for Estimating Lightning Performance of
resistance at all towers is arranged to be less than 10 , and
Transmission Lines,” IEEE Trans. Power Appar. Syst., vol. PAS-104, pp.
the horn length is arranged 1.1 m, the critical current is less 919 – 932, July 1985.
than 80 kA, and the tripout rate is expected to decrease to (23) Hileman, A.R.,” Insulation co-ordination for Power System,”, Marcel
half of the present rate. Dekker Inc., New York, 1999.
(24) T. Udo, “Estimating of Shielding Failure and Mid-Span Back-flashover
Based on the Performance of EHV Double Circuit Transmission Lines,”
IEEE Trans. Power Deliv., vol. 12, no. 2, pp.832 – 836, 45, April 1997.
References (25) A. C. Report, “A Method of Estimating Lightning Performance of
Transmission Lines,” Trans. Am. Inst. Electr. Eng., vol. 69, no. 2, pp. 1187–
(1) Meteorological and Geophysical of Padang Panjang, “Annual report,”
1196, Jan 1950.
BMKG Padang Panjang, Indonesia, Dec. 2010.
(26) Z. G. Datsios, P. N. Mikropoulos, T. E. TSovilis, and S. I. Angelakidou,
(2) J. He, X. Wang, Z. Yu, and R. Zeng, “Statistical Analysis on Lightning
“Estimation of the Minimum Backflashover Current of Overhead Lines of
Performance of Transmission Lines in Several Regions of China,” IEEE
the Hellenic Transmission System through ATP – EMTP Simulations,”
Trans. Power Deliv., vol. 30, no. 3, pp. 1543–1551, 2014.
CRIEPI Report F-75008 2016.
(3) R. de la Rosa, G. Enriquez, and J. L. Bonilla, “Contributions to lightning
research for transmission line compaction,” IEEE Trans. Power Deliv., vol.
3, no. 2, pp. 716 – 723, Apr. 1988.
(4) A. H. A. Bakar, D. N. A. Talib, H. Mokhlis, and H. A. Illias, “Lightning back
flashover double circuit tripping pattern of 132 kV lines in Malaysia,” Int. J.
Electr. Power Energy Syst., vol. 45, no. 1, pp. 235–241, 2013.
(5) Subcommittee for transmission lines, study committee an lightning risk,
“Application Guide for Transmission Line Surge Arrester,” CRIEPI Report
H-07 2012 (in Japanese).
(6) IEEE Guide for Improving the Lightning Performance of Transmission
Lines, IEEE Std 1243-1997, pp. 1 – 44, Dec. 1997.
(7) F. M. Gatta, A. Geri, S. Lauria, M. Maccioni, and F. Palone, “Tower
Grounding Improvement vs. Line Surge Arresters: Comparison of Remedial
Measures for High-BFOR Subtransmission Lines,” IEEE Trans. Ind. Apl.,
vol. 51, no. 6, pp. 4952 – 4960, June 2015.
(8) S. Wu and W. Sun, “Back flashover protection performance analysis of
220kV double circuit transmission line,” in Proc. Asia-Pacific Conference
Power and Energy Engineering, (APPEEC), Wuhan, Cina, March 2011, pp.
1 – 4.
(9) A. Ametani and T. Kawamura, “A method of a lightning surge analysis
recommended in Japan using EMTP,” IEEE Trans. Power Deliv., vol. 20, no.
2, pp. 867–875, Apr. 2005.
(10) J. Sardi and M. Z. A. Ab Kadir, “Investigation on the effects of line
parameters to the lightning performance of 132 kV Kuala Krai-Gua Musang
transmission line,” in Proc. 7th International Symposium on Power
Engineering and Optimization, (PEOCO), Langkawi, Malaysia, June 2013,
pp. 594 – 599.
(11) E. F. KONCEL, “Potential of a Transmission-Line Tower Top When Struck
by Lightning,” Trans. Am. Inst. Electr. Eng., vol. 75, no. 3, pp. 457 – 462,
Jan. 1956.
(12) A. Holdyk and B. Gustavsen, “Inclusion of Field Solver-Based Tower
Footing Grounding Models in Electromagnetic Transients Programs,” IEEE
Trans. Ind. Apl., vol. 51, no. 6, pp. 5101 – 5106, 2015.
(13) The Distribution and Load Control Center Sumatra (Sumatra P3B), “monthly
report,” Sumatra P3B, Padang UPT, Indonesia, July 2013.