EXPERIMENTAL STUDY
ON LIGHTNING BREAKDOWN CHANNELS IN TIlE SOILS |
|
Song Zeqing, He Jing iiang
Wuhan University of Hydraulic & Electric Engineering
Wuhan, Hubel, P.R. China 430072 |
M.R. Ro,ghtlVeer
The University of Manitoba
Winnipeg, Manitoba, Canada R3T 5V6 |
|
|
Abstract: Presenl day models do not satisfactorily describe the
surge behavior of earth under the passage of lightning currents which
result in a single breakdown channel. This paper presents. preliminary
results of an investigation into the factors that affect lIahtning
current propagation In different soils under sinale channel formation.
The results presented In this paper were obtained uslng an indoor
experimental system and examine the influence of soil characteristics
and source capacity and voltage on lightning channel characteristics.
It is hoped that Ihese results will be useful in establishing
models which describe both single and multiple channel formation In
soils. |
|
l.introductton |
According to traditional lightning theory, when lightning
reaches the surface of earth, it spreads into the soil, which may
be represented by an equivalent hemispherical eleerrode ofrodilts
equal to that of the sparking channel. Severa! mode-Is that des.cribe
the behavior of earth under impulse currems are based on this assumption
[1][2]. But it is known, from the results of both indoor experiments
and field data conceming outdoor lightning strikes resulting in damage
to underground cables that impulse currents propagate along a single
channel during the total
propagating period or for part of the total traversed distance [3]
[4]. Present day models do not satisfactorily describe either siruation
[I]. In order to investigate the factors which influence formation
of single and multiple patins and the breakdown mechanism of soils,
we have be¡¤en carrying out cooperative research work. to date we have
addressed the following nspects:
1) When an impulse c urrent generator discharges into different kinds
of soils, what are the factors that decide whether the current propagates
along a single or multiple channels?
2) What is the influence of moisture, compaction and resistivity?
3) What is Ihe influence of generator capacitance and voltage?
4) Finally, does the lightning current exhibit a tendency to flow
along pre¡¤formed channels? This paper presents results
concerning single channel formation. Other results will be published
in future papers. |
|
2.Experimental system |
The indoor experimental system was comprised of an impulse
current generator discharging into a hemispherical pot of radius 100
mm into which different soils were packed. (Figure 1). |
|
At the bottommost point and at a height of 45mm from
the bottom, six 40mm radius round plate electrodes are located synunemcally
(one at the bottom point and five along the circle, Figure 2),
The six electrodes cover about 3/4 of the area of hemispherical surface
and their outputs are linked together to a point th at is earthed.
The current entrance electrode is an 18 mm-diameter red placed at
the center of hemisphere just above the soil suface in order to provide
a sparking connection. It is held syrometricaliy with respect to the
six
measurement electrodes and is also constrained so that it does not
move under tile impulsive force produced by discharging.
Magnetic links were used to detennine which elearode was hit by the
lightning channel. Branch currents were estimated by proof current_angle
curve of magnetic link. The signal across a shunt
was connected to TDS220 oscilloscope (Bandwidth: IOOMHZ, Sampling:
rate: lOs/S, Record length: 2500 points) to display and print the
total current wOlv-eform. The resistivity and moisture
of soi were ineasured using tile methods reponed in [5]. Soil compaction
was calculated by dividing the total weight of soH by the volume of
the hemisphere. |
|
3.Experimental results |
Although experiments were carried out in it soil model
indoors, the fonowing two observations are common to several outdoor
lightning incidents which resuhed in damage 10 communication cables
in the Guangdong province of China [4]. This commonality is very important
since it lends credibllity to the results generated indoors using
a
small model.
1) when a soil is broken down it results in holes or ditches in the
soil.
2) The conditions that cause multiple channel formation are more severe
than those which cause single channel formation. |
|
3.1 Influence of soils characteristics |
Soil resistivily is the basic parameter tbat affects
breakdown channels. Since different soils have different compaction
and moisture content, their resistivities are different, and hence
their breakdown voltages for single channel formation are different.
The general townendency exhibited is that the lower the resistivity,
the smaner the minimum breakdown voltage for a single channel. --See
Table 1 From Tables 1 and 2 it is also seen that when
moisture or compaction increased. the soils resistivity decreases
and its minlmum breakdown voltage for single channel formation decreases
also. A dry soil has a very high res istivity, and
its breakdown voltage is very high, but only a little water sprinkled
on the surface of soil (simulating natural r ain) causes a sig:tificant
decrease in the breakdown voltage of a single channel. Table 1 shows
that send wjth zero moisture content could not be broken down even
when the generator
yoltage was increased to 46 kV. From data concerning dielecnic strength
published in [3], it can be calculated that the break down voltage
for this case is approximately 70kV. However, after only about 15
to 20 dtops of tap water were evenly sprinkled on the surface of the
soil the breakdown voltage decreased to 35 kV with resulting peak
value of current of 500A |
Table I:Minimum breakdown voltage for single
channel in dlfferent soils |
|
Table 2:Influence of compaction |
|
|
3.2 Influence o f impulse generator capaCitance
and voltage |
When the generator capccitors DC Voltage is at the minunum
value for a particular soils breakdown, it results in the formation
of a single channel. If the voltage is increased, only one channel
forms but the value of current
increases. If the voltage is increased beyond a certain value, the
current path has a tendency to split into two
branches. |
TABLE 3: Influence of gene rotor capacitance
Cultivat-cd soil , 13.7%moislUre , 1 27g/§¨ 1900§Ùm |
|
When the generator capacitance is decreased, the breakdown voltage
for formation of a single channel increases and the total current
decreases. On the other hand, the voltage value for multiple path
formation does not change appreciably despite the fact that both total
and branch currents decrease. - See Table 3. |
|
3.3 Location of channels: |
If the soil condition is restored to its original condition
after each discharge of the generator, each plate electrode is equally
likely to be the recipient of a chaMel which forms randomly. The
reason is that the six electrodes are situated equidistantly from
the top entrance electrode from an electromagnetic field point of
view. This observation was confmned for various values of generator
capacitance. 1f the soil is left undisturbed after
each discharge, the next discharge (for same value of generator voltage)
propagates in the soil along or close to the channel formed upon previous
application of voltage. Under such circumstances all channels form
close to each other. Table 4 shows that the breakdown channels are
confined to a small volume formed by the entrance electrode and plate
electrodes #2 to #4. |
|
Table 4:Location of channe |
|
3.3 Location of channels: |
Figure 3 shows a typical waveform of the total current.
Because breakdown resistance varies under different condition, it
affects the total current waveform. The Time lag Tl
is the time difference between the application of a voltage sufficient
to cause soil breakdown and occurrence of breakdown itself. For
any type of soil, Time lag T1 decreases as generator voltage is increased.
When the voltage applied corresponds to the minimum voltage for single
breakdown channel fonnation, Tl reaches a maximum that may be several
hundred microseconds. When the capacitor voltage exceeds a certain
value, TI becomes shorter than 3
microseconds and does not change significantly thereafter. If other
conditions are not changed the lower the resistivity of soil, the
sholter the Tl. The duration of impulse current Td
is also affected by applied voltage. As the generator capacitor voltage
is increased beginning from the minimum value necessary for single
channel fonnation, the duration Td shows a tendency to decrease. After
the voltage exceeds a certain value, the duration changes very little.
-See Table 5 |
|
Table 5: Tl and Td |
|
With respect to some soils, when the generator voltage
is increased, the wavefonn of total current exhibits a negative overshoot.
--See Figure 4. |
|
4.Conculsions |
--Resistivity is the basic parameter that affects formation of breakdoVl1
channels. With regard to homogeneous soils the influence of moisture
and compaction is reflected by resistivity changes.
--The mechanism of breakdown is different for dry soils, soils that
are wetted only at the top and homogeneously wetted soils.
--In model studies breakdoVl1 is more likely to occur along or close
to preformed channels if the soil is left undisturbed between discharges
of the impulse current generator.
--Breakdown channel has different equivalent resistance under different
condition, so it affects the total current waveform. |
|
5. References |
[1] AC.Liew, "Dynamic model of impulse characteristics of concentrated
earths", Proc. lEE, VoL 121, No.2, Feb. 1974
[2] M.E. Almedia, "Modeling the hysteresis behavior of the transmission
tower footing", 9th ISH in Graz., Australia, 1995
[3] l.A.l Pettinga, "Fulgurites useful to deternine the charge
of a lightning strokes", 10th ISH in Montreal, Canada, ] 997
[4 J Xue Lei, "Lightning protection for underground communication
cables", Lightning and Electrostatics, Vo1.1 No.1, 1985 |