World Academy of Science, Engineering and Technology

International Journal of Electrical and Computer Engineering

Vol:2, No:6, 2008

Study of Tower Grounding Resistance Effected

Back Flashover to 500 kV Transmission Line in
Thailand by using ATP/EMTP
B. Marungsri, S. Boonpoke, A. Rawangpai, A. Oonsivilai, and C. Kritayakornupong

system which may occur and reach the lightning withstand

Abstract—This study describes analysis of tower grounding voltage has not yet been conducted. EGAT requires analyzing
resistance effected the back flashover voltage across insulator string the voltage level of lightning over voltage of existing 500 kV
in a transmission system. This paper studies the 500 kV transmission transmission lines.
International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

lines from Mae Moh, Lampang to Nong Chok, Bangkok, Thailand,

which is double circuit in the same steel tower with two overhead
ground wires. The factor of this study includes magnitude of Mae Moh
lightning stroke, and front time of lightning stroke. Steel tower uses
multistory tower model. The assumption of studies based on the
return stroke current ranged 1-200 kA, front time of lightning stroke
between 1 µs to 3 µs. The simulations study the effect of varying Tha Tako
tower grounding resistance that affect the lightning current.
Simulation results are analyzed lightning over voltage that causes
back flashover at insulator strings. This study helps to know causes Wang Noi
of problems of back flashover the transmission line system, and also
be as a guideline solving the problem for 500 kV transmission line
systems, as well. Sai Noi
Nong Chok
Keywords—Tower grounding resistance, back flashover,
multistory tower model, lightning stroke current.
Chom Bung Pluak Deang
I. INTRODUCTION

T HAILAND is tropical country which having a lot of

thunderstorm days and lightning discharge activities in one
year. The back bone of power system transmission is overhead
Ratchaburi

Fig. 1 Existing EGAT 500 kV Transmission System

transmission system. So, it is essential to investigate a
lightning surge for a reliable operation of a power system, EGAT have been reported the statistic of lightning occurs in
because the lightning surge over voltage is one of dominant northern region during March-August 2005[6]. The report
factors for the insulation design of the power system and the concluded that the lightning often occurs during April to May.
protection of equipments in power transmissions and But the most severe lightning takes place in June. The positive
substations. When lightning strikes the top of a transmission lightning is about 5% with magnitude between 11 kA and 171
tower, a lightning current flows down to the bottom of the kA. But the negative lightning is about 95% with magnitude
tower and causes a tower voltage rise which results in a back- between -10 kA to - 139 kA. The most lightning magnitude is
flashover across an insulator string. between -10 kA and -50 kA.
At present, Electric Generation Authority of Thailand When lightning stroke strikes the top of a tower, a traveling
(EGAT) has expanded to total about 2,600 circuits-km of 500 voltage is generated which travels back and forth along the
kV, as shown in Fig. 1[1]. Switching over voltage in EGAT tower, being reflected at the tower grounding and at the tower
system has been conducted by some researchers [1-5]. top, thus raising the voltage at the cross-arms and stressing the
However, the lightning over voltage estimation of overall insulator strings. Flashover will occurs across an insulator
string if this lightning transient voltage exceeds its lightning
withstand voltage level. Such flashover called back flashover
B. Marungsri is with Suranaree University of Technology, Nakhon
Ratchasima, 30000, Thailand (phone: +66 4422 4366; fax: +66 4422 4601; e-
Back flashover voltages are generated by multiple reflections
mail: bmshvee@ [Link]). along the struck tower and also along the shield wire for shield
S. Boonpoke, A. Rawangpai and A. Oonsivilai are with Suranaree lines at the adjacent towers. The lightning induced over
University of Technology, Nakhon Ratchasima, 30000, Thailand. voltage across an insulator string for the struck tower is not a
C. Kritayakornupong is with Electricity Generating Authority of Thailand
(EGAT).

International Scholarly and Scientific Research & Innovation 2(6) 2008 1061 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

straightforward. The peak of lightning induced voltage will be B. Lightning Source Model
directly proportional to the peak of lightning stroke current. It is well known that the magnitude of impulse current due
This paper is organized as follows. Models of system to a lightning discharge is a probability function. Low
studied are provided in Section II. Then, the simulation results discharge levels of lightning current between 5 to 22 kA may
with various factors are discussed in section III. Finally, result in a higher tendency for the lightning strike to pass by
conclusion is presented in Section IV. any shield wires and directly hit a phase conductor. While, the
larger magnitude of lightning impulse currents may tend to
II. PARAMETERS AND MODEL OF EGAT 500 KV strike the tower top or overhead ground wire and lead to a
TRANSMISSION LINE back flashover across the insulator string[7]. Front time and
tail time of lightning waveform and lightning current
A. Structures of 500 kV Transmission Lines
amplitudes in this study are given in TABLE I. Lightning
Tower structure configuration type DQV9(9°) is used in this current from lightning stroke is modeled as an impulse current
study. Tower dimensions are illustrated in Fig. 2. The full source parallel with the impedance of a lightning path as
details of 500 kV transmission line tower are as follows: shown in Fig. 3. The resistance value is taken to be 400 Ω,
Conductor: 4 Bundle per phase (795 MCM 54/7 ACSR) which was derived by Bewley [8].
Spacing between = 45.7 cm
Diameter = 2.773 cm TABLE I
International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

DC resistance = 0.07094 ohm/km LIGHTNING AMPLITUDES AND WAVEFORMS

Ruling span = 420 m Lightning Current Amplitude Waveform tF / tT
Ultimate strength = 12,800 kg - 34 kA  1 / 30 . 2 µs
 1 . 2 / 50
Weight = 1.52 kg/m Wind span = 500 m - 50 kA  µs
Shield wire: 3/8″ EHS, Class A, Galvanized steel 
- 100 kA  2 / 77 . 5 µs
Number of wires = 2 per tower  3 / 75 µs
Shielding angle = -5° to outer phases - 139 kA
Ultimate strength = 6985 kg
Diameter = 9.1444 mm C. Tower and Transmission Line Model
Weight = 0.406 kg/m The double circuit with two overhead ground wire
Insulator: Type Suspension, ball and socket transmission tower is considered in this study is shown in Fig.
ANSI Class 52-8 and ANSI Class 52-11 1. The tower is modeled as a transmission line for surge
Based on max. conductor tension of 22% rated propagation. The surge impedance of the tower and the
Tensile strength with wind pressure of 86 kg/m2 propagation velocity down the tower are estimated and applied
Grounding system: ground resistance 10 ohms or less in a multistory tower model [9 – 11]. Each parts of tower are
A. Stub angle to rebar : used in all cases represented by distributed parameter model, with lossless high
B. Ground rods : used in all cases, one ground rod at frequency approximation. The main surge propagation path is
each tower leg modeled with frequency dependent transmission lines even if
Lightning outage rate: Less than 0.05 per 100 km per year the length is only a few meters. The tower model is shown in
Right-of-way width: 70 meters for each line Fig. 4. Multistory tower model parameters are shown in

TABLE II. Simulation of span between tower uses frequency
dependent (phase) model. The span of towers 7 spans is
h1 73 m simulated, as shown in Fig. 3, and the parameters of
h2 67 m transmission line used in this study is shown in Fig. 2.
h3 56 m
h4 45 m
d1 13.06 m D. Tower Grounding Resistance Model
d2 21 m The tower grounding resistance for fast transient surges is
d3 22.42 m not well understood. The impulse grounding resistance or high
d4 23.8 m frequency grounding resistance is less than the measured at
low frequency or calculated resistance because significant
ground currents cause voltage gradients sufficient to
breakdown the soil around the ground rod. A variable
grounding resistance approximation can be applied which is
surge current dependent as in (1) [7].

Fig. 2 Tower Configuration

International Scholarly and Scientific Research & Innovation 2(6) 2008 1062 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

Fig. 3 Span of 7 Towers in This Study

 1 E ρ
 Ig =  0 0 (2)
 2π  Rg2
 
International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

where ρ0 is soil resistivity (ohm-meter),

E0 is soil ionization gradient (about 300 kV/m).

E. Back Flashover Model

It is well-known that back flashover occurring when tower
potential is higher than lightning impulse withstand voltage or
BIL level of the insulator strings. Lightning impulse withstand
voltage level of the insulator string is not a unique number.
The insulator string may withstand a high magnitude impulse
voltage which has a short duration even it has failed to
withstand a lower magnitude impulse voltage with longer
duration. This characteristic of the insulator string is known as
the volt-time characteristic or V-t curve. A simplified
expression of withstand voltage capability for an insulator
Fig. 4 Multistory Tower Model string can be calculated as in (3) [7].
TABLE II K2
V f 0 = K1 + (3)
MULTISTORY MODEL PARAMETERS
t 0.75
 Z t1 = Z t2 = Z t3 = 200 Ω, Z t 4 = 150 Ω
where Vf0 is a flashover voltage (kV),
τ = 2 H / v µs γ = 0.8944 K1 is 400*L,
r1 = −2 × Z t1 × ln γ /(l1 + l2 + l3 ) Ω/m K2 is 710*L,
r2 = −2 × Z t 2 × ln γ /(l4 ) Ω/m L is insulator length, (m),
t is elapsed time after lightning stroke, µs.
R1 = r1 × l1 = 10.52 Ω R2 = r1 × l2 = 19.29 Ω In this study L= 4.429 m, lightning impulse withstand
R3 = r1 × l3 = 19.29 Ω R4 = r2 × l4 = 33.48 Ω voltage of the insulators insulator string can be represented by
volt-time curve as shown in Fig. 5. If the voltage across the
L1 = R1 × τ = 5.12 µH L2 = R2 × τ = 9.39 µH
insulator string exceeds its impulse voltage withstand
L3 = R3 ×τ = 9.39 µH L4 = R4 ×τ = 16.29 µH capability, the back flashover occurs.
R f = 5 − 100 Ω 20

15
Voltage, MV

Rg
Rf = 10
I (1)
1+
Ig 5

where Rf is tower grounding resistance (ohm), 0

Rg is tower grounding resistance at low current and 0 1 2 3 4 5
low frequency (ohm), Time, µs
I is surge current into ground (kA), Fig. 5 V-t Curve of the Insulator String
Ig is limiting current initiating soil ionization (kA).

International Scholarly and Scientific Research & Innovation 2(6) 2008 1063 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

III. SIMULATION RESULTS

In this section, the 500 kV transmission line as shown in
Fig. 3 has been modeled using ATP/EMTP. The tower
grounding resistance is one of factors effected the back
flashover voltage across the insulator string in transmission
system as mention earlier. In this paper, only tower grounding
resistance will be considered with various factors that affected
back flashover. The factors of this study include front time of
lightning stroke, and magnitude of lightning stroke.

A. Front Time of Lightning Stroke Current

A lightning flash generally consists of several strokes which (a) Lightning Stroke Current Waveform 1.0/30.2 µs
are lower charges, negative or positive, from the cloud to the
ground. The first stroke is most often more severe than the
subsequent strokes [7]. As shown in Fig. 5, impulse voltage
withstand capability of the insulator string depends on front
time of lightning strokes. In order to elucidate the effect of
International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

front time of lightning stroke current, lightning stroke current –

34 kA having different wave front and wave tail times, i.e.
1/30.2 µs, 1.2/50 µs, 2/77.5µs and 3/75 µs is used to simulate
lightning induced voltage across the insulator strings at upper,
middle and lower phases[12]. Simulation results are illustrated
in Fig. 6. In spite of the same magnitude of lightning stroke
current and same tower grounding resistance, obviously
differences in the magnitude of impulse voltage across the
insulator strings can be seen when comparing each phase (b) Lightning Stroke Current Waveform 1.2/50 µs
having different wave shape.

B. Magnitude of Lightning Stroke

Due to negative lightning stroke current in Thailand having
magnitude between -10 kA to - 139 kA and the most lightning
stroke current having magnitude between -10 kA and -50 kA,
three different magnitudes of lightning stroke current, i. e. – 50
kA, -100 kA and -139 kA are used to simulate the effect of
lightning stroke current. Fig. 7– Fig. 9 illustrated the maximum
amplitude of lightning induced voltage across phase insulator
strings. Obviously, the effect of front time on tower induced
voltage can be seen on the simulation results. From (3) and
(c) Lightning Stroke Current Waveform 2.0/77.5 µs
time at maximum value of tower induced voltage, back
flashovers across phase insulator strings due to lightning stroke
current are summarized in TABLE III – TABLE VI. No back
flashover occurs in case of lightning stroke current -50 kA
although tower grounding resistant vary from 5 – 100 Ω. In
case of -100 kA, almost back flashovers occur on the top phase
insulator string especially fast front time of lightning stroke.
However, more back flashovers also occur on the other phase
insulator strings in case of slow front time of lightning stroke.
In case of –139 kA, back flashover occurs on all phase
insulator string.


(d) Lightning Stroke Current Waveform 3.0/75 µs
Fig.6 Impulse Voltage Waveform across Phase Insulator String in
case of Lightning Stroke Current – 34 kA

International Scholarly and Scientific Research & Innovation 2(6) 2008 1064 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

(a) Waveform 1.0/30.2 µs (a) Waveform 1.0/30.2 µs

International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

(b) Waveform 1.2/50 µ (b) Waveform 1.2/50 µs

(c) Waveform 2.0/77.5 µs (c) Waveform 2.0/77.5 µs

(d) Waveform 3.0/75 µs (d) Waveform 3.0/75 µs

Fig.7 Maximum Value of Tower Induced Voltage across Phase Fig. 8 Maximum Value of Tower Induced Voltage across Phase
Insulator Strings in case of Lightning Stroke Current -50 kA Insulator Strings in case of Lightning Stroke Current -100 kA

International Scholarly and Scientific Research & Innovation 2(6) 2008 1065 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

(a) Waveform 1.0/30.2 µs (c) Waveform 2.0/77.5 µs

International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

(d) Waveform 3.0/75 µs

(b) Waveform 1.2/50 µs
Fig. 9 Maximum Value of Tower Induced Voltage across Phase Insulator Strings in case of Lightning Stroke Current -139 kA

TABLE III
BACK FLASHOVER ACROSS PHASE INSULATOR STRINGS IN CASE OF WAVEFORM 1/30.2µS
Tower grounding -50 kA -100kA -139 kA
resistance (ohm) Upper Middle Lower Upper Middle Lower Upper Middle Lower
(MV) (MV) (MV) (MV) (MV) (MV) (MV) (MV) (MV)
5 X X X O X X O O X
10 X X X O X X O O O
20 X X X O X X O O O
30 X X X O X X O O O
40 X X X O X X O O O
50 X X X O X X O O O
60 X X X O X X O O O
70 X X X O X X O O O
80 X X X O X X O O O
90 X X X O X X O O O
100 X X X O X O O O O
O : flashover X: no flashover

TABLE IV
BACK FLASHOVER ACROSS PHASE INSULATOR STRINGS IN CASE OF WAVEFORM 1.2/50 µS
Tower grounding -50 kA -100 kA -139 kA
resistance (ohm) Upper Middle Lower Upper Middle Lower Upper Middle Lower
5 X X X O X X O O X
10 X X X O X X O O X
20 X X X O X X O O X
30 X X X O X X O O O
40 X X X O X X O O O
50 X X X O X X O O O
60 X X X O X X O O O
70 X X X O X X O O O
80 X X X O X X O O O
90 X X X O X X O O O
100 X X X O X O O O O
O : flashover X: no flashover

International Scholarly and Scientific Research & Innovation 2(6) 2008 1066 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

TABLE V
BACK FLASHOVER ACROSS PHASE INSULATOR STRINGS IN CASE OF WAVEFORM 2/77.5µS
Tower grounding -50 kA -100 kA -139 kA
resistance (ohm) Upper Middle Lower Upper Middle Lower Upper Middle Lower
(MV) (MV) (MV) (MV) (MV) (MV) (MV) (MV) (MV)
5 X X X X X X O O X
10 X X X X X X O O X
20 X X X X X X O O X
30 X X X X X X O O X
40 X X X X X X O O X
50 X X X X X X O O X
60 X X X X X X O O X
70 X X X O X X O O O
80 X X X O X O O O O
90 X X X O O O O O O
100 X X X O O O O O O
O: flashover X: no flashover

TABLE VI
International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

BACK FLASHOVER ACROSS PHASE INSULATOR STRINGS IN CASE OF WAVEFORM 3/75µS

Tower grounding -50 kA -100 kA -139 kA
resistance (ohm) Upper Middle Lower Upper Middle Lower Upper Middle Lower
(MV) (MV) (MV) (MV) (MV) (MV) (MV) (MV) (MV)
5 X X X X X X O X X
10 X X X X X X O X X
20 X X X X X X O X X
30 X X X X X X O O X
40 X X X X X X O O O
50 X X X O X X O O O
60 X X X O O O O O O
70 X X X O O O O O O
80 X X X O O O O O O
90 X X X O O O O O O
100 X X X O O O O O O
O: flashover X: no flashover

[3] C. Jaipradidtham, “ Electromagnetic Field Analysis on Surge Response

IV. CONCLUSION
of 500 kV EHV Single Circuit Transmission Tower in Lightning
This paper has described an analysis of Tower grounding Protection System using Neural Networks”, Proc. of Int. Conf. on
resistance effected the back flashover voltage across phase Control, Automation and System, KINTEX, Gyeonggi-Do, Korea, 2-5
June 2005.
insulator string in EGAT 500 kV transmission system. As seen [4] N. Petcharaks, C. Y. Yu and C. Panprommin, “ A Study of Ferranti and
from simulation results, the shorter front time of lightning Energization Overvoltages case of 500 kV Line in Thailand”, 11th Int.
stroke current will increase the lightning induced over voltage Sym. on High Voltage Engineering, Vol. 1, Conf. Publ. No. 467, 1999,
across phase insulator string. No back flashover occurs in case pp. 291-294.
[5] P. Yadee and S. Premrudeepreechacharn, “Analysis of Tower Footing
of the most lightning stroke current having magnitude between Resistance Effected Back Flashover Across Insulator in a Transmission
-10 kA and -50 kA. When the magnitude of lightning stroke System”, Proc. of Int. Conf. on Power Systems Transients (IPST’07) ,
current is up to -100kA, the back flashover always occurs with Lyon, France, June 4-7, 2007.
any tower footing resistances. The simulation results have [6] Report of flashover voltage in Northern Region Operation Division
1996-2003, EGAT, Thailand. (in Thai)
shown that the higher tower grounding resistance has potential [7] A. R. Hileman, Insulation Coordination for Power Systems, Marcel
to damage the transmission line by back flashover across the Dekker, New York, USA, 1989.
phase insulator strings. However, it still has other factors and [8] L. V. Bewley, Traveling Waves on Transmission Systems. Dover
counterpoint measures to consider reducing the back flashover Publications, New York, N.Y., USA,1951.
[9] M. Ishii, T. Kawamura, T. Kouno, E. Ohsaki, K. Shiokawa, K. Murotani
for transmission line. and T. Higuchi, “Multistory Transmission Tower Model for Lightning
Surge Analysis”, IEEE Trans. on Power Delivery, Vol. 6, No. 3, July
REFERENCES 1991, pp. 1327 – 1335.
[10] T. Yamada, A. Mochizuki, J. Sawada, E. Zaima, T. Kawamura, A.
[1] K. Ngamsanroaj and W. Tayati, “An Analysis Of Switching
Ametani, M. Ishii and S. Kato, “Experimental evaluation of a UHV
Overvoltages in the EGAT 500 kV Transmission System”, Large
tower model for lightning surge analysis”, IEEE Trans. on Power
Engineering Systems Conference on Power Engineering, 7-9 May 2003,
Delivery, Vol. 10, No. 1, January 1995, pp. 393 – 402.
pp. 149 – 153.
[11] A. Ametani, N. Nagaoka, T. Funabashi and N. Inoue, “Tower
[2] C. Y. Yu, N. Petcharaks and C. Panprommin, “The Statistical
Structure Effect on a Back-Flashover Phase”, Proc. of Int. Conf. on
Calculation of Energization Overvoltages, Case of EGAT 500 kV
Power Systems Transients (IPST’05), Paper No. IPST05 – 190,
Lines”, IEEE Power Engineering Society Winter Meeting , Vol. 4, 23-
Montreal, Canada, June 19-23, 2005.
27 Jan. 2000, pp. 2705 - 2709.

International Scholarly and Scientific Research & Innovation 2(6) 2008 1067 [Link]/1307-6892/3915
 World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:2, No:6, 2008

[12] R. B. Anderson and A. J. Erikson, “Lightning Parameters for

Engineering Applications”, ELECTRA 69, March 1980, pp. 65 – 102.

Boonruang Marungsri was born in Nakhon

Ratchasima Province, Thailand, in 1973. He
received his B. Eng. and M. Eng. from
Chulalongkorn University, Thailand in 1996 and
1999 and D. Eng. from Chubu University, Kasugai,
Aichi, Japan in 2006, all in electrical engineering,
respectively. Dr. Marungsri is currently a lecturer in
School of Electrical Engineering, Suranaree
University of Technology, Thailand. His areas of interest are electrical power
system and high voltage insulation technologies.

Suphachai Boonpoke was born in Surin Province,

Thailand, in 1984. He received his B. Eng. in Electrical
Engineering from Suranaree University of Technology,
Nakhon Ratchasima, Thailand, in 2005. Recently, he is a
graduate student in School of Electrical Engineering,
International Science Index, Electrical and Computer Engineering Vol:2, No:6, 2008 [Link]/Publication/3915

Suranaree University of Technology. His research

interesting area is high voltage insulation technology.

Anucha Rawangpai was born in Ratchaburee Province,

Thailand, in 1985. He received his B. Eng. in Electrical
Engineering from Suranaree University of Technology,
Nakhon Ratchasima, Thailand, in 2005. Recently, he is a
graduate student in School of Electrical Engineering,
Suranaree University of Technology. His research
interesting area is high voltage insulation technology.

Anant Oonsivilai was born in Khon Kaen Province,

Thailand, in 1963. He received his B. Eng. fron Khon
Kaen University, and M. Eng. from King Mongkut
Institute of Technology North Bangkok, Thailand and
PhD. from Dalhousie University, Canada, all in
electrical Engineering, in 1986, 1992 and 2000,
respectively. Dr. Oonsivilai is currently an Assistant
Professor in School of Electrical Engineering, Suranaree
University of Technology, Thailand. His areas of interest are electrical power
system, stability, control technology, advance alternative, and sustainable
energy.

Chanin Kritayakornupong was born in Bangkok,

Thailand, in 1974. He received his [Link] and [Link]
both in electrical engineering from Chulalongkorn
University, Bangkok, Thailand, in 1996 and 1998,
respectively. In 1999, he joined Electricity Generating
Authority of Thailand (EGAT). He is now an Engineer
Level 6 of transmission line maintenance technology
section, transmission line technology and aviation
department, transmission system maintenance division.

International Scholarly and Scientific Research & Innovation 2(6) 2008 1068 [Link]/1307-6892/3915