2002 Student Conference on Research and Development Roaxdngs, Shah ALam, Malaysia

J t O ~ m o m

SIMULATION OF LIGHTNING SURGES ON TOWER TRANSMISSION

USING PSCADEMTDC :A COMPARATIVE STUDY
Z. Zakaria., S.M. Bashi', N.F. Mailah', M.R.Othman", H. Osman"
'Department of Electrical and Eleclronic Engineering,
Faculty of Engineering,Ciniversili Pulra Malaysia
43400 Serdang. Selangor, Malaysia

..Tenaga National Research Sdn.Bhd

Bangi, Selangor. Malaysia

ABSTRACT - This paper describes the analytical and

the experimental responses of the surges strike on the
transmission line towers. A new method of calculating
transmission tower surge response has been proposed.
It has been found that the tower surge response
calculated by the proposed method has a close
agreement with the measured tower surge response
obtained from scale model and field tests. Without the
use of simulation program, this topic might be difficult
to understand. PSCADEMTDC has been selected as
the software used to generate the appropriate data
needed to graphically demonstrate this phenomenon.
1. INTRODUCTION

Tower surge impedance is one of the mosl

important factors in analyzing lightning performances
on transmission lines [I]. Jordon [2] had proposed the
first theoretical formulation for tower surge
impedance. The aim of this study is practically based
on Neumann's induction formula, which states that the
current distribution on the transmission tower was
uniform from the bottom to the top of the tower.
However the effect of return stroke current is not taken
into consideration. The tower is approximated as a
vertical cylinder having a height equal to that of the
tower, and the radius equal to the mean equivalent
radius of the tower. Propagation velocity inside of the
tower is assumed to be of the speed of light.
Beside the modeling methods, many
experimental and theoretical studies had been
performed on the surge response characteristics of
transmission towers. There are many reports on the
techniques used to estimate the tower surge
impedance. Theoretical formulation of tower surge
impedance based on the electromagnetic field theory
had been proposed by Janischewskyj [3], Wagner and
Hileman [4] and Oknmura and Kijima [5], whicb
considered the effects of the vector potential generated
by the injection current into the tower. Propagation
velocity inside the tower was assumed to be of the
speed of light.
Breuer et al [6] had canied out actual
measurements of the surge impdance of actual
transmission towers. In his work, a reflection method
was adopted, and an impedance value of 165 ohms was
obtained at the top of the tower. The measured

0-7803-7565-3/02/$17.00
02002 JEEE.

propagation velocity inside the tower was almost at the

speed of light.
Kawai [7] reported another experimental
value for actual transmission towers. He used a direct
methd to measure tower surge impedance. The value
for this impedance at the top of the tower was
measured to be 100 ohms. His experimental results
showed that the tower response to a vertical injection
current is different from the response to a horizontal
current injection. Measured propagation velocity inside
the tower was 70%-80% of the speed light.
Ishii el al[8]andNarita et a l [ 9 ]reported field
measurements of full-scale tower impedance using
direct method. These measurements were performed
using the techniques of inclined and horizontal Current
injections. Both of them proposed surge impedance of
the tower Mon the EMTF'. Propagation velocity
inside the tower was assumed to be at the speed of
light.

2. TOWER SURGE IMPEDANCE

The surge characteristics ,of transmission
have been measured using 500kV Tenaga
N a s i o d Elerhad's tower (tower height 67m) as shown
in Figure 1. This measurement has not considered the
&e&
of the ground wires and power lines.
tower

7
I
i

~~

10.szsm.

Fig. 1. Shcture drawing of 500 kV Tower.

426

From (he calculation, the surge impedances

for the tower are :-

The calculation of average value of the tower surge

impedance has been carried out by implementing the
formulas proposed by the introduction manual to
PSCADEMTDC and C E R E [IOl. In this calculation
the tower radius R has been chosen as Model (I), and
the tower was simulated by an inverted cone as shown
in Figure 2.

I. For the lower vertical cylinder conductor using

formula (1) is :- Z, = 136.61 ohm.

2. For the upper vertical cylinder conductor using

formula (2) is :- Z, = 55.883 ohm.
3. For the horizontal upper, middle and lower cross
arm using formula (3) is:- Z, = 172.86 ohm
3. SURGE RESPONSE CHARACTERISTICS OF

ACTUAL TOWER

3.1 Measurement Conditions

Figure 3 shows the measurement conditions
and the arrangement of measuring equipment. The
tower under measurement is the 500kV steel tower
with an overall height of 62.8m and a base width of
10.8m. The ground wires were insulated from no. 1
tower by insulators to avoid reflection at this position.
since the span between no. 2 and no. 3 tower is short.
Fig. 2.
All the phase conductors were directly connected lo
no. 1 and no. 4 tower.to reduce the noise, which may
2.1 For A Vertical Cylindrical Conductor
affect the measurement.
The theoretical value of the vertical cylinder
A 50 ohm coaxial cable of 300m in length
impedance has been given in the literature [IO] in a
was used to inject current into no. 3 tower from a pulse
different form: generator (PG) on the ground as Seen in Figure 3. The
tower surge impedance is chosen at 115 ohms. The
i) CIGRE formula of lower vertical cylinder is given
sheath of the cable was left open at the tower end,
as: while it is connected to the earth at other end, where
the configuration is equivalent to as to set the PG on
Z, = 60 In [cot 0.5 {tan (R ,@)}
(1)
the ground wire across the arm, as no external field
exists before the traveling wave within the coaxial
Where (see figure 2) :
cable from the PG reaches the tower top no. 4. Thus
h = tower height (m)
current injection is similar to Kawai 171 and is
KY8
= average tower radius (m)
classified a direct method (no other lines for
= (rl hz +rz h + r3 h i ) h
measurement were installed).
rl = radius at mid point hl tower (m)
The injected current was measured by the
rz = radius at h2 from the mid point (m)
current shunt and the voltage across the current strings
r3 = tower's base radius (m)
was measured by a 10K ohm resistive voltage divider.
These signals were converted into light signals and
ii) PSCADEMRDC formula for upper vertical
were transmitted to waveform recording devices on the
cylinder is given as :ground through optical wave-guides. Bundle fibers
were used for the optical wave-guides, which resulted
in the change of sensitivity after the wave-guide was
.
disconnected and reconnected to the transducer.
$,.;si
Where:
Therefore the sensitivity was calibrated before and
h = height of the tower (m)
after each measnrement by a calibrator incorporated in
d = diameter of the conductor (m)
the optical signal transmission system and the accuracy
was maintained within 2% of error. In a measurement
2.2 For A Horizontal Cylindrical Conductor
without connecting ground wires to the tower, the
zO for the
The theoretical Surge
voltage between an isolated ground wire and the cross
horizontal upper, middle and h e r cylinders can be
arm was also measured. The frequency range of the
approximated to be :measuring system was from DC up to 12 MHz, which
was limited by the electro-optical signal converter.
Z, = (377/2n) * In (2ldr)
(3)
3.2 Recorded Waveforms
Where:
Figure 4 shows the resulted waveform when
h = height of conductor above the ground (m)
the current was injected at the top point of the tower.
r = radius of conductor (m)
In the case of steep front wave, a PG was sent out at a
neat rectangular waveform into a 50-ohm connecting
'

.~(.~A'.

427

cable. The resulted waveform, which was measured at

the top point of the tower, was found to be deformed
which can be seen in Figure 4(a). Figure 4@) shows
various waveforms calculated for conventional and
new tower models shown for comparison purpose. It is
clear that the calculated results agree accurately with
the measured results and the proposed model can
reproduce the effect of the direction of the injected

current line. This shows that from the experiment and

theory, the setup current injection line has a great
effect on the surge response. This measured value of
the propagation velocity along the towers is almost
80?6-90% speed of the light. The fact that the
propagation velocity depends on the strnctures such as
the cross arms, horizontal elements and inclined
elements are clearly shown here.

Figure 3. Setup for lightning surge response measurements

Measured waveforms

Calculated waveforms

Figure 4. Calculated and measured waveform

(a) Upper phase insulator voltage
(b) Middle phase insulator voltage
(c) Lower phase insulator voltage

428

4. CONCLUSIONS

This paper reports on the results of

measurements and calculations of the lightning surge
response characteristics of a 500 kV transmission
tower. Based on the above calculation, a new
multistory tower model will be developed to be used in
the multistoly conductor lightning surge analysis by
EMTDC. This analysis enables us to predict the
waveform of each insulator voltage, which is useful in
determining the phase. Measurements of the tower
surge response were carried out. It is known from the
measurements .and calculations that the actual surge
velocity in the tower is closed to the speed of light.
From the comparisons between the calculated results
obtained using the proposed model and the measured
results, it is clearly shows the direction of the current.
Injection line significantly affects the surge
characteristics of the transmission tower. This fact is
important in understanding the surge response of the
tower.

[4] C.F Wagner and A.R. Hileman, A New Approach

to the Calculation of the Lightning Performance of
Transmission Lines , 111-A Simplified Method Stroke
to Tower, AIEE Trans. Vol 79, 1980, pp 589-603.

5. REFERENCES

[8] M. Ishii, T. Kawamura, T. Kouno, E. 0hsaki.K.

Murotani and T . Higuchi, Multisory Transmission
Tower Model for Lightning Surge Analysis, IEEE
Trans. Power Delivery, Vol. 6 no. 3 pp 1327-1335,

[l] Hideki Motoyama and Hiroji Matsnbara,

Analytical and Experimental Study on Surge
Response Transmission Tower, IEEE Trans., PWRD,
vol 15, no 2, April 2000.

[5] K.Okumura and A. Kijima. A method for

Computing Surge Impedance of Transmission Line
Tower by
Electromagnetic Field Theory, IEE of Japan. Trans B,
Vol 105, pp 733-740. 1985.

[6] G Breuer, A.J, Schultz, R.H. Schlomann and W.S

Price, Field Studies of the Surge Response of a
345 kV Transmission Tower And Ground Wires,
ALEE Trans., pt I11 vol 77, 1958, pp 1392-1396
171 M. Kawai, Studies of the Surge Response on a
Transmission Line Tower, IEEE Trans, 1964 . PAS88. pp30-34.

1991.

[9] ita.
Tomomi,Yamada, Takeshi, Mochizki,
Azuma, Zaima, Ishii, and Masaru , Observation of
Current Waveshapes of Lightning Strokes on
Transmission Towers, IEEE Transactions on Power
Delivery, Vol. 15, Jan 2000, pp 429-435

[2] C.A. Jordon, Lightning Computations for

Transmission Lines with Overhead Ground Wires
Part 11, General Electric Review. Vol 34, pp180-185,
1994.

[IO] CIGRE Working Group 33.01, Guide to

Procedures for Estimating the Lightning Performance
of Transmission Lines, CIGRE Technical, Brochure
63, Oct. 1991.

[3] Janischewskyj, Lightning Electric Characteristics

of first and subsequent return Strokes to a Tall Tower,
Published in the Proceeding of 11 International
Symposium on high voltage Engineering ,Vol. 1, 467,
London, Aug 23-Aug 27 1999.

429