1314 IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 28, NO.

3, JULY 2013

Development of a Method of Calculating the Melting

Characteristics of OPGW Strands Due to DC Arc
Simulating Lightning Strike
Mikimasa Iwata, Member, IEEE, Toshiya Ohtaka, Member, IEEE, Yasukuni Kuzuma, and
Yutaka Goda, Senior Member, IEEE

Abstract—Some strands of composite fiber-optic ground wire lightning impulse current generator cannot supply the current
(OPGW) are sometimes melted and broken by high-energy light- whose duration is so long and the electric charge is so large. In
ning strikes. DC arc tests simulating lightning strikes have been order to clarify the aforementioned characteristics of OPGW
performed to obtain the melting and breaking characteristics
of OPGW strands. The tests have to be performed under many strands, consequently, dc arc tests have been performed and
conditions concerning the arc (e.g., current, duration, polarity, gap test conditions that adequately simulate high-energy lightning
length) and the OPGW (e.g., size, type, and number of strands) to strikes have also been investigated [1], [2], [6], [7]. However,
clarify the melting and breaking characteristics of the strands. In lightning and OPGW conditions vary considerably. The light-
this paper, the calculations regarding the melting characteristics ning conditions are wide ranging as described before. On the
of strands are performed considering the transferred heat and
its area from the arc to the strands under the aforementioned other hand, there are many combinations of the types and num-
conditions. The melting characteristics of strands are calculated bers of strands of OPGW because they are selected according
with an arc current of 1–100 kA considering the measured current to the conditions of the transmission lines (e.g., transmission
of actual lightning. The calculation results of the strand melting voltages and single-line-to-ground fault currents). Figs. 1(a)
duration depending on arc current show good agreement with the and (b) shows the structure of OPGW with cross sections of
experimental values obtained in dc arc tests.
80 and 140 mm , respectively. These are one of the OPGWs
Index Terms—Arc discharges, ground wires, lightning, optical- generally used. The types of strands in Fig. 1(a) and (b) are
fiber cables, power transmission lines.
23AC and 40AC, respectively. “AC” means “Aluminum-Clad
steel”, while “23” and “40” denote that the electric conduc-
I. INTRODUCTION tivity of the AC strand is “23%” and “40%” of the electric
conductivity of copper strand with equivalent cross section to
the AC strand, respectively. If dc arc tests are performed under

G ROUND WIRES (GWs) have been used to prevent

overhead power transmission lines from being struck
by lightning. Composite fiber-optic ground wires (OPGWs)
conditions involving many combinations of the arc and OPGW
conditions, a lot of effort and cost will have to be consumed.
Since this is inefficient, it is desirable to find a way to calculate
have also been used for the aforementioned purpose as well the melting and breaking characteristics of OPGW strands. The
as transmitting high volumes of communication information. calculation way will be also a good tool for developing a new
Although GW and OPGW are designed to withstand light- OPGW strand that is hard to melt and is broken by high-energy
ning strikes, some of their strands, even if the strand is an lightning strikes.
aluminum-clad steel, are sometimes melted and broken by As a first step, this paper investigates a calculation method
high-energy lightning strikes [1]–[3]. If the GW and OPGW of melting characteristics of OPGW strands due to dc arc, and
are broken and they fall, this may cause a short-circuit fault of taking the arc and OPGW conditions into consideration. The
the power transmission lines. Besides this fault, in the event of calculation results of the strands melting duration depending on
OPGW breakage, communication interruption will also occur, the arc current show good agreement with the experimental ones
causing significant damage to the communication system of the obtained in dc arc tests for both types of OPGW shown in Fig. 1.
electric power utility and the public. Therefore, the melting and
breaking characteristics of OPGW strands must be clarified.
According to lightning measurement results, the peak current II. DC ARC TESTS
of lightning is 1–100 kA, its duration is 10 s 200 ms, and
The methods of dc arc tests [1], [2] are roughly described.
the flowing electric charge 0.1–1000 C [4], [5]. Normally, the
Fig. 2 shows an arrangement of dc arc testing in the air. A
thin copper wire (0.2 mm in diameter) is stretched between the
Manuscript received September 27, 2011; revised March 28, 2012; accepted
OPGW and a counter electrode so that current flows to vaporize
April 21, 2013. Date of publication May 15, 2013; date of current version June
20, 2013. Paper no. TPWRD-00832-2011. the copper wire to generate an arc. The counter iron bar elec-
The authors are with the Central Research Institute of Electric Power Industry trode is set above the OPGW with a gap length of about
(CRIEPI), Yokosuka 240-0196, Japan. (e-mail: [email protected]).
1500 mm. The OPGW is stretched at both ends with a tensile
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. force of 9.8–14.7 kN, a force almost equivalent to that applied
Digital Object Identifier 10.1109/TPWRD.2013.2260567 when stringing in transmission lines.

0885-8977/$31.00 © 2013 IEEE

IWATA et al.: DEVELOPMENT OF A METHOD OF CALCULATING THE MELTING CHARACTERISTICS OF OPGW STRANDS 1315

Fig. 3. Flowchart for calculation.

forming a new arc root where the arc column is brought into
contact with with the OPGW, insulating tape is wound onto both
sides of the arc-producing (exposed) part of length , about
20–40 mm.

Fig. 1. Examples of the OPGW structure. (a) 80 mm (23AC 6/(3.86)) [2]. III. CALCULATION METHODS AND CONDITIONS
(b) 140 mm (40AC 12/3.0, 40AC 8/(2.86)) [1].
Fig. 3 shows an outline of the flowchart used to calculate
the melting characteristics of OPGW due to the DC arc. By
using the arc conditions (e.g., current, duration, polarity, gap
length) and OPGW conditions (e.g., the size, type and number of
strands), the heat quantity and its area transferred to the OPGW
from the arc are calculated, whereupon the melting quantity of
OPGW due to the arc is calculated. In the calculation of this
paper, the arc current varies from 1–100 kA considering the
measured current of actual lightning, and the objects for the
calculation are the two types of OPGW shown in Fig. 1. This
chapter describes the calculation methods and conditions.

A. Heat Transferred to OPGW From the Arc and Its Area

The total heat transferred to the OPGW from the arc is
expressed by

(1)
Fig. 2. Arrangement of dc arc testing.
where is the heat due to electrons emitted from or absorbed
into the OPGW, is the convective heat transfer due to
the arc jet spouted from the counter electrode, is the heat
An arc root on the OPGW tends to be too mobile to melt the
due to the radiation from the arc, and is the joule heating in
strands, due to the electromagnetic force (emf) induced by the
the electrode. As and are relatively smaller than
current flowing through the OPGW. Accordingly, conductors
[8], this small degree of heat is not considered in this paper.
are connected to both ends of the OPGW to divert the arc current
Therefore, is estimated as the following equation:
to the earth side symmetrically so that the arc root cannot easily
move. The current is supplied by a full-wave rectifying three- (2)
phase ac using silicon diodes. The ac is generated using a short-
circuit generator (15 kV, 2500 MVA) and a transformer. The dc 1) Heat Due to Electrons : changes depending on
output can be changed between 120 kV/10 kA to 16 kV/60 kA the electrode polarity are as follows [9]
by changing the ac voltage and combination of diodes. In the dc • In the case of a cathode
arc tests, the supply dc voltage is set at about 10–17 kV so that
the supply voltage is not significantly affected by an increase
in arc voltage. Furthermore, to prevent the arc from moving by (3)
1316 IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 28, NO. 3, JULY 2013

TABLE I
VOLTAGES CONCERNING THE HEAT DUE TO ELECTRONS (SURROUNDING GAS: AIR)

• In the case of an anode (933 K for aluminum and 1809 K for iron [17]) during testing.
for in the room temperature is larger by about
10%–20% than that for the melting point. In the calculation,
(4) is assumed to be 300 K in order to determine the max-
where and are the cathode-fall and the anode-fall volt- imum value of .
ages, respectively; is the work function of the electrode; and The heat-transfer coefficient is estimated using the fol-
is the voltage estimated by dividing the enthalpy due to the lowing equation [16]:
electron drift by the arc current . is given by the following
(8)
equation, in which Bolzmann’s constant and the value of elec-
tronic charge are considered [9]
where is the thermal conductivity of the flowing gas, the air
(5) arc jet is contaminated iron vapor in the calculation, is the
characteristic dimension, the outer diameter of the OPGW in the
where is the arc temperature near the electrode. calculation. Nu is the Nusselt number, which varies depending
These voltages , and vary depending on the elec- on the heat-transfer conditions and is estimated using equations
trode material and the surrounding gas. In this paper, these volt- obtained by experiments. In the calculation, the heat transfer
ages are the ones for the electrode material aluminum and iron due to the flow collision is considered, because the arc jet flow
and the surrounding gas air, because the type of OPGW strand spouting from the counter electrode collides with the OPGW. In
used, as shown in Fig. 1, is AC steel, and dc arc tests are per- this case, Nu varies depending on gap length and the diam-
formed in the air as described in Section II. Although the cen- eter of the counter electrode, and is expressed as the following
tral part material used in the actual strand is steel, the material equation [16] using the Prandtl number (Pr) and the Reynolds
is iron in the calculation because the main component of steel is number (Re). In the case of
iron and its thermal properties are easily obtained from the ref-
erences. Table I shows the voltages used for calculations. is (9)
the voltage estimated by dividing of (3) and (4) by .
is estimated using (5), where the dependence of the arc temper- where C is as follows.
ature on arc current and the electrode material [13], [14] is In the case of
considered. The transferred heat area for is estimated as
an area touched by the arc using the following equation: (10)

(6)
in the case of
where means a function where the smallest value in the
(11)
bracket is picked, is the current density on the OPGW,
is the exposed area of the OPGW mentioned in Section II. ,
in the case of
for the aluminum and iron is 60.0 and 33.8 A/mm , respectively,
and independent of its polarity, which is estimated using the arc
root diameter in the electrode measured in [15].
2) Convective Heat Transfer : As shown in Fig. 2 re-
garding dc arc testing, an arc is ignited between the OPGW and
(12)
the iron counter electrode set above it. A high-current arc can
produce a high-temperature arc jet containing iron vapor from
where Pr and Re are obtained using the following equations:
the counter electrode at high speed from the arc root on the elec-
trode. The convective heat transfer to the OPGW due to the arc (13)
jet flow is calculated using the following [16]:

(7) (14)

where is the heat transfer coefficient, and is the tem- where , and are the specific heat at constant pressure,
perature of the OPGW surface. is estimated assuming the the viscosity, and density and velocity of the flowing gas, which
same value as for . is considered to be room tem- is the air arc jet contaminated with iron vapor in the calculation.
perature (300 K) before dc arc testing, and near melting point These values and are referred to in [13], [14], and [18]–[20].
IWATA et al.: DEVELOPMENT OF A METHOD OF CALCULATING THE MELTING CHARACTERISTICS OF OPGW STRANDS 1317

Fig. 5. Dependence of heat transferred to OPGW (80 mm ) on arc current.

Fig. 4. Simplification of the cross-sectional structure of OPGW strands.
(a) Actual structure. (b) Simplified structure.

specific heat in the calculation is the average value of one for

B. Simplification of the Cross-Sectional Structure of OPGW room temperature and one for the melting point to facilitate the
Strands calculation. The thermal property values mentioned before are
those in [17] and [21].
As shown in Fig. 1, the OPGW consists of single-layer or
multilayer AC steel strands. In the calculation, the melting phe- IV. RESULTS
nomena of the only outer-layer strands are investigated consid-
ering the melting and breaking of actual OPGW strands by light- A. Total Heat Transferred to OPGW From the Arc
ning strikes reported in [1]. In the case of the actual strand, as As expressed in (2), the total heat transferred to the
shown in Fig. 4(a), there is iron in the central part, which is OPGW from the arc is the sum of and . In the
covered with aluminum. In the calculation, the cross-sectional calculation, the gap length varies from 50–1500 mm,
structure of the strand is simplified, as shown in Fig. 4(b), and although is about 1500 mm in the dc arc tests as described
there is an aluminum layer on the iron layer. The outer diam- in Section II. Consequently, the percentage of in
eter , the cross-sectional area of the aluminum layer, and the increases as rises. For example, in the case of 80 mm
iron layer of Fig. 4(b) are the same as those of Fig. 4(a). It is OPGW shown in Fig. 1(a), the percentages are around 80%,
assumed that the heat transferred from the arc is injected into 95%, and 99% for of 50, 500, and 1500 mm, respectively.
the aluminum layer, which is subsequently melted and removed, The percentages in the other cases vary somewhat depending
whereupon the heat from the arc is injected into the iron layer. on the arc current , the OPGW type, and its polarity.
The aluminum and iron layer can be evaporated by a tempera- Fig. 5 shows the dependence of on for 80 mm
ture rise due to the heat from the arc. In the calculation, how- OPGW and 1500 mm. The calculation parameters are the
ever, the evaporation is not taken into account, because it is con- OPGW strand material (aluminum, iron) and its polarity. In
sidered that the melting part of the layer will be blown off by every case, increases almost proportionally with growing
a tensile force applied to the OPGW strands and by an arc jet . for the iron and cathode exceeds that for the alu-
spouting from an arc root on the layer. minum and anode. This is because the voltage regarding
In the calculation, after a part, which is estimated by multi- the electrons transfers for iron and the cathode exceeds that for
plying the transferred heat area by the thickness of the alu- aluminum and anode, as shown in Table I. for 140 mm
minum or iron layer, is melted completely, an arc moves to an- OPGW is almost the same as the results in Fig. 5. This is due
other part and similarly causes it to melt. For a high arc current to the fact that is nearly equal to for 1500 mm
and/or long arcing duration, all strands can be melted. In the cal- as mentioned before, although varies depending on the
culation, however, the melting of the upper strands, which are OPGW size.
easily touched by the arc (e.g., the upper three strands in the case
of Fig. 4) is investigated. When a part in the entire cross-section B. Transferred Heat Area to OPGW From the Arc
region is melted, even if its length along the OPGW axis-direc- Fig. 6 shows the dependence of the transferred heat area
tion is a bit, the melting quantity is considered 100%. on . Fig. 6(a) and (b) shows the calculation results for 80
The energy required to melt the strands is the sum of “the and 140 mm OPGW, respectively. In the calculation, the ex-
energy required for a temperature rise from a room tempera- posed part length of OPGW is the same as that in the dc arc
ture to melting point” and “the energy required for complete tests (i.e., 40 mm [2] for Fig. 6(a) and 20 mm [1] for Fig. 6(b),
melting after reaching the melting point.” The former energy is respectively).
estimated using the specific heat and density of the strand ma- In every case, for iron exceeds that for aluminum when
terial (aluminum, iron). The latter energy is estimated using the is below 20–30 kA. This tendency results from a calculation
latent heat of melting and the density of the strand material. Al- condition where the current density for iron is smaller than
though specific heat varies depending on the temperature, the that for aluminum, as described in Section III-A. Conversely,
1318 IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 28, NO. 3, JULY 2013

Fig. 6. Dependence of transferred heat area to OPGW on arc current.

(a) 80 mm . (b) 140 mm . Fig. 7. Dependence of melting time of OPGW (80 mm ) on arc current.
(a) Cathode. (b) Anode.

is constant when exceeds 30–40 kA, and the constant

value of differs in Fig. 6(a) and (b). This is because the con- time on the arc current for the melting quantity 40%,
stant value of is the exposed area of the OPGW, and the area 70%, and 100%. For reference, the relation between time and
determined by the OPGW size and the exposed part length current for 100, 500 and 1000 C is shown as a dotted line in
differs in the 80 and 140 mm OPGW. The constant value of these figures. Figs. 7 and 8 are for the OPGW polarity negative
for iron is smaller than that for aluminum because of the (cathode) and positive (anode), respectively. The plotted points
simplification of the cross-sectional structure of OPGW strands in these figures show the experimental results (i.e., the arc cur-
that is mentioned in SectionIII-B (i.e., the outer diameter of the rent and arcing duration) in which 3 of 6 strands [2] (Fig. 7)
iron layer is smaller than that of the aluminum layer shown in and 6 of 12 strands [1] (Fig. 8) are broken, respectively. The
Fig. 4(b) as an example). arc current in dc arc tests, meanwhile, shows a waveform in
which a ripple component is superimposed on a constant value
C. Melting Characteristics of OPGW Strands because the arc current is obtained by full-wave rectification. In
In the dc arc tests, the melting and breaking characteristics of the calculation, the arc current is assumed to be constant to save
OPGW strands have been evaluated by the value estimated by calculation efforts. Therefore, the arc current of the plotted ex-
integrating the current by time (i.e., the flowing electric charge perimental results is a time-averaged value estimated using the
) [1], [2], [6], [7]. In general, the number of melting and arcing duration and the flowing electric charge obtained in the
breaking strands rises with increasing the charge . In [1] and dc arc tests.
[2], similar to this tendency, increases with growing , but Figs. 7 and 8 suggest the following:
varies considerably depending on the arc current even 1) In both OPGWs, the melting time for the cathode is
if is constant. This result means that cannot be evaluated shorter than that for the anode, because the total heat
by but by . In this paper, the melting characteristics of the transferred to the OPGW for the cathode exceeds that for
strands are evaluated by the arc current . the anode as mentioned in Fig. 5.
Figs. 7 and 8 show the calculation results for 80 and 140 mm 2) In every case, when the arc current exceeds 20–30
OPGW, respectively. These are the dependence of the melting kA, the gradient of the calculation result approaches that
IWATA et al.: DEVELOPMENT OF A METHOD OF CALCULATING THE MELTING CHARACTERISTICS OF OPGW STRANDS 1319

exposed area , as estimated using (6). Since

is constant, is independent of the arc current .
The total transferred heat increases almost pro-
portionally with growing , as described in Fig. 5.
In this case, therefore, the melting time is almost in-
versely proportional to .
When the area of arc root is smaller than the exposed
area of the OPGW, the transferred heat area is the
arc root area, as estimated using (6). Since the arc root
area increases proportionally with growing
increases similarly, while increases almost pro-
portionally with growing . In this case, therefore,
the melting time is almost independent of .
b) Where the arc current is low and/or the OPGW
size (the outer diameter ) is large, the arc diameter
is shorter than half the length of the OPGW circum-
ference. Since the current density is constant, as de-
scribed in Section III-A, the diameter of the arc root
area is proportional to the square root of , and
increases almost proportionally with growing
. In this case, therefore, the melting time is al-
most inversely proportional to the square root of .
3) The melting time in Fig. 7 is longer than that in Fig. 8
under the same conditions of arc current and melting
quantity . This is because the energy required for iron
to melt, including the latent heat of melting, is about three
times as high as for aluminum, and the cross section of the
iron part for Fig. 7 exceeds that for Fig. 8.
4) In the dc arc tests the OPGW strands are broken, when the
melting quantity , which is estimated using the calcu-
lation results and the plotted points, is about 70%–100%.
This result is considered attributable to the fact that the
Fig. 8. Dependence of melting time of OPGW (140 mm ) on arc current.
(a) Cathode. (b) Anode. strands can be broken even when is below 100% be-
cause the melting part of the strands can be blown off by a
tensile force applied and by an arc jet spouted from an arc
of the dotted line where is constant, which means the root on the strands.
melting time is inversely proportional to . Con-
versely, in the region below the current, the variation of V. CONCLUSION
with changing diminishes, which means dimin- Although OPGWs are designed to withstand lightning
ishes more than when is inversely proportional to . strikes, some OPGW strands are sometimes melted and broken
This calculation result shows good agreement with the by the latter. To clarify the melting and breaking characteristics
tendency of the experimental results shown in the plotted of OPGW strands, dc arc tests have been performed and test
points in Fig. 7(a). As described before, in the dc arc tests, conditions that adequately simulate high-energy lightning
the melting and breaking characteristics of OPGW strands strikes have also been investigated. However, the lightning and
have been evaluated by based on experimental data in OPGW conditions are wide-ranging. Since performing dc arc
which the melting quantity of the electrode is proportional tests under many conditions are not efficient, the development
to . If the melting characteristics of OPGW strands are of a calculation method to determine the melting and breaking
calculated based on the thought mentioned before, the cal- characteristics of OPGW strands is desired. In this paper, a
culation results must be parallel to the dotted lines, where method of calculating the melting characteristics of OPGW
is constant. However, the calculation and experimental strands due to dc arc is investigated considering the arc con-
results in Figs. 7 and 8 do not show the tendency. These ditions (e.g., current, duration, polarity, gap length) and the
results are considered due to the following: OPGW conditions (e.g., the size, type, and number of strands).
a) Where the arc current is high and/or the OPGW By using these conditions, the heat quantity and its
size (the outer diameter ) is small, the arc diam- area transferred to the OPGW from the arc are calculated,
eter is longer than half the length of the OPGW whereupon the melting quantity of OPGW due to the arc is
circumference. calculated.
When the area of the arc root exceeds the exposed The OPGW consists of single-layer or multilayer AC steel
area of the OPGW, the transferred heat area is the strands. In this paper, the melting phenomena of the only outer-
1320 IEEE TRANSACTIONS ON POWER DELIVERY, VOL. 28, NO. 3, JULY 2013

layer strands of 80 and 140 mm OPGW are investigated con- [8] M. Shibuya, T. Inaba, and Y. Goda, “Molten losses on cut end of rod
sidering the melting and breaking of actual OPGW strands by electrodes in AC high current arc discharge,” Inst. Elect. Eng. Jpn.
Trans., vol. 113-B, pp. 327–334, 1993.
lightning strikes, and the arc current varies from 1–100 kA [9] Y. Kito and T. Mizuno, “An experimental approach to melting of alu-
considering the measured current of actual lightning. In the case minium alloy string at an arc spot formed by surging current,” Inst.
of the actual strand, in the central part, there is steel whose main Elect. Eng. Jpn. Trans., vol. 108-B, pp. 577–584, 1988.
[10] Y. Yokomizu, T. Matsumura, R. Henmi, and Y. Kito, “Total voltage
component is iron, which is covered with aluminum. In the cal- drops in electrode fall regions of , argon and air arcs in current
culation, therefore, the cross-sectional structure of the strand is range from 10 to 20000A,” J. Phys. D, Appl. Phys., vol. 29, pp.
simplified to one which has an aluminum layer on an iron layer. 1260–1267, 1996.
[11] Electrical Discharges Handbook. Chiyoda, Tokyo, Japan: IEEJ,
In the calculation, after a part, which is estimated by mul- 1974, p. 154.
tiplying by the thickness of the aluminum or iron layer, [12] H. B. Michaelson, “The work function of the elements and its period-
is melted, an arc moves to elsewhere and similarly causes the icity,” J. Appl. Phys., vol. 48, no. 11, pp. 4729–4733, 1977.
[13] M. Iwata, S. Tanaka, K. Ikeda, and Y. Goda, “Temperature of arc jet
part to melt. The melting of the upper strands, which are easily of 50 kA class AC Arc in long gap (Part 2)—Temperature and energy
touched by the arc (e.g., the upper three strands in the case of density of arc jet spouted from iron electrode,” Yokosuka, Kanagawa,
80 mm ) is investigated. When parts of the entire cross section Japan, CRIEPI Rep. W00026, 2001.
[14] M. Iwata, K. Ikeda, S. Tanaka, and Y. Goda, “Temperature of arc jet of
region are melted, even if its length along the OPGW axis-di- 50 kA Class AC arc in long gap (Part 3)—Axial distribution of temper-
rection is slight, the melting quantity is considered 100%. ature and energy density,” Yokosuka, Kanagawa, Japan, CRIEPI Rep.
Consequently, the melting time for the cathode is shorter than W01010, 2002.
[15] Y. Goda, M. Iwata, K. Ikeda, and S. Tanaka, “Behaviour of arc jet of 50
that for the anode, because the total heat transferred to kA class high current arc in a long gap,” Yokosuka, Kanagawa, Japan,
the OPGW for the cathode exceeds that for the anode. When CRIEPI Rep. W00034, 2001.
the arc current exceeds 20–30 kA, the gradient of the cal- [16] JSME Data Book: Heat Transfer, 4th ed. Shinjuku, Tokyo, Japan:
JSME, 1986, p. 42, 65.
culation result approaches that of the property where the flowing [17] Tables of Physical Constants, S. Iida, K. Ohno, S. Kamimae, H. Ku-
electric charge is constant, meaning the melting time is in- magaya, and S. Sawada, Eds. Shinjuku, Tokyo, Japan: Asakura Pub.
versely proportional to . Conversely, in the region below the Co. Ltd., 1978, p. 20, 108, 141, and 143.
[18] S. Tanaka, K. Ikeda, M. Iwata, and Y. Goda, “Flow velocity character-
current, the variation of with changing diminishes, which istics of high current arc jet in atmospheric air (Part 2)—Flow velocity
means diminishes more than in the case where is inversely in 50 kA class arc jet spouted from iron and aluminium electrode,”
proportional to . This calculation result shows good agree- Yokosuka, Kanagawa, Japan, CRIEPI Rep. W01021, 2002.
[19] T. Amakawa, “Transport properties of high temperature air with metal
ment with the tendency of the experimental results. These re- vapor,” Yokosuka, Kanagawa, Japan, CRIEPI Rep. W86009, 1986.
sults suggest that the developed calculation method for melting [20] M. Boulos, P. Fauchais, and E. Pfender, Thermal Plasmas—Funda-
characteristics of OPGW strands due to dc arc is a good tool for mentals and Applications. New York: Plenum, 1994, vol. 1, pp.
413–417.
performing the dc arc tests efficiently and clarifying the melting [21] Chronological Scientific Tables, Tokyo Astronomical Observatory
and breaking characteristics of OPGW strands analytically. Fur- ed. Chuo, Tokyo, Japan: Maruzen Co. Ltd., 1988, p. 472.
thermore, the obtained results in this paper also suggest that
Mikimasa Iwata (M’12) was born in Aichi Prefec-
the developed calculation method will be able to estimate the ture, Japan, on May 14, 1964. He received the B.E.,
melting and breaking characteristics of conventional and newly M.E., and Ph.D. degrees in electrical engineering
developed lightning-resistant OPGW strands due to high-en- from Nagoya University, Aichi Prefecture, Japan, in
1988, 1990, and 2000, respectively.
ergy lightning strikes, assuming that the dc arc tests adequately In 1990, he joined the Central Research Institute of
simulate the actual lightning strike. Electric Power Industry (CRIEPI), Yokosuka, Japan.
From 2006 to 2007, he was a Visiting Research As-
sociate of RWTH Aachen University, Aachen, Ger-
many. Currently, he is a Senior Research Scientist of
REFERENCES the Applied High Energy Physics Sector and the High
[1] Y. Goda, S. Yokoyama, S. Watanabe, T. Kawano, and S. Kanda, Power Testing Laboratory, CRIEPI. His research interests are mainly electrode
“Melting and breaking characteristics of OPGW strands by lightning,” erosion and stability of AC arc plasma, high-current arc phenomena on power
IEEE Trans. Power Del., vol. 19, no. 4, pp. 1734–1739, Oct. 2004. equipment, and synthesis of ceramics nanopowder by means of arc plasma.
[2] Y. Goda, M. Shimizu, and A. Matsumoto, “DC arc tests of OPGW Dr. Iwata is a member of the Institute of Electrical Engineers of Japan (IEEJ).
simulating high energy lightning strike,” in Proc. 28th Int. Conf. Light. In 2006, he was awarded the Technical Development Award by the IEEJ.
Protect., 2006, pp. 959–964.
[3] J. Huang, X. Chen, and J. Xu, “Investigation and analysis of an OPGW
lightning strike incident,” in Proc. 58th Int. Wire Cable Symp., 2009,
pp. 389–392. Toshiya Ohtaka (M’03) was born in Kanagawa
[4] Working Group on Lightning Protection for Transmission Systems, Prefecture, Japan, on March 9, 1976. He received
“Lightning observation on Japan sea coast in winter,” Komae, Tokyo, the B.E., M.E., and Ph.D. degrees in electrical
Japan, CRIEPI Rep. T10, 1989. engineering from Waseda University, Tokyo, Japan,
[5] T. Shindo, S. Yokoyama, T. Sunaga, A. Asakawa, H. Motoyama, A. in 1998, 2000, and 2003, respectively.
Wada, and H. Goshima, “Lightning characteristics in winter season to From 2001 to 2003, he was a Research Associate
a high stack—Ten-year observation results from 1989 to 1998 at Fukui at Waseda University, and from 2003 to 2004, he was
Observation Site” Komae, Tokyo, Japan, CRIEPI Rep. T58, 1999. a Visiting Lecturer. In 2004, he joined the Central Re-
[6] M. Böhme and K. Möller, “Arc motion during lightning test of power search Institute of Electric Power Industry (CRIEPI),
ground wires with optical fibers (OPGW),” in Proc. 24th Int. Conf. Yokosuka, Japan, and is currently a Research Scien-
Light. Protect., 1998, pp. 571–575. tist of the Applied High Energy Physics Sector and
[7] S.-H. Xie and R.-S. Yang, “Study of OPGW performance against light- the High Power Testing Laboratory, CRIEPI. His research interests are mainly
ning strikes in UHV transmission lines,” in Proc. 58th Int. Wire Cable high-current arc phenomena and power system analysis.
Symp., 2009, pp. 393–398. Dr. Ohtaka is a member of the Institute of Electrical Engineers of Japan.
IWATA et al.: DEVELOPMENT OF A METHOD OF CALCULATING THE MELTING CHARACTERISTICS OF OPGW STRANDS 1321

Yasukuni Kuzuma was born in Kanagawa Prefec- Yutaka Goda (SM’94) was born in Ehime Prefec-
ture, Japan, on March 29, 1945. He received the B.E. ture, Japan, on April 1, 1956. He received the B.E.
degree in electrical engineering from Kanto Gakuin and Ph.D. degrees in electrical engineering from
University, Kanagawa Prefecture, Japan, in 1969. Nagoya University, Aichi Prefecture, Japan, in 1979
In 1963, he joined the High Voltage Power and 1998, respectively.
Laboratory (HVPL), Yokosuka, Japan. Since 1977, In 1979, he joined the Central Research Institute of
he has been with the Central Research Institute of Electric Power Industry (CRIEPI), Yokosuka, Japan.
Electric Power Industry (CRIEPI), Yokosuka, Japan. From 1991 to 1992, he was a Visiting Research As-
Currently, he is on the staff of the High Power sociate at the University of Connecticut, Storrs. Cur-
Testing Laboratory, CRIEPI. His research interests rently, he is a Senior Research Scientist and Director
are mainly arc phenomena on power equipment. of the High Power Testing Laboratory, CRIEPI. His
Mr. Kuzuma is a member of the Institute of Electrical Engineers of Japan. research interests are mainly arc phenomena on power equipment.
Dr. Goda is a member of the Institute of Electrical Engineers of Japan
(IEEJ) and Working Group 12 of IEC-TC42. He was awarded the Technical
Development Award and the Best Article Award by the IEEJ in 1997 and 1998,
respectively.