An Introduction to
Transmission Line
Lightning Performance
Kyle King
Transmission Line Design
& Operations Seminar
Lenox, Massachusetts
June 2002
Lightning
Lightning 6/02 - 2
Lightning
Lightning 6/02 - 3
Lightning
Lightning 6/02 - 4
Types of Lightning
• Negative Cloud to Ground
+++++++++++++++++++++++++
--- +++
-- -- -- -- -- -- -- -- -- -- -- -- -- --
•Ground to Cloud and Cloud -
-
-
to Cloud also exist
+++++++++++++++++++++++++
•Positive Cloud to Ground --- +++
----------------------------
Lightning 6/02 - 5
Lightning Flash as Seen
by a Streak Camera
Dart
Leader
+/- 5000 ft.
Return
Stroke
2nd
Stepped Stroke
Leader
Continuous Current
Initial Breakdown Process: 0.1 - 0.2 sec.
Process
Lightning 6/02 - 6
A Five-Stroke Lightning Flash
Time - Seconds
0 0.2 0.4 0.6 0.8 1.0
10 Time - Microseconds
0 20 40 60 80 100
kA
20
10
First Stroke in the Flash
30
20
30
Lightning 6/02 - 7
Lightning
•Current magnitudes are influenced by the
geometry and impedance of the object struck
•We assume no effect in most calculations
kA
Base of the East and
Empire State Building Hudson Rivers
Lightning 6/02 - 8
Wave Propagation Fundamentals
z How much charge in the leader as it nears the
ground
4
50 kA = 5 × 10 coulombs / s
Return stroke velocity = 1/3 velocity of light
= 108 m/s
coulombs
× 108 = 5 × 10 − 4 Coulombs/meter
m near the ground
Lightning 6/02 - 9
Wave Propagation Fundamentals
For a cylinder:
Let Q = total charges/m
Q = 2πRD = 2πRεE
At the wall, E must be roughly Q D=
D=EεE
E
25 kV/cm, or 2.5 x 106 V/m 1 Meter
D = Electric Flux Density
The air would not ionize at less E = Electric Field Intensity
than this value. Then:
Q 2R
R=
2πεE
5 × 10− 4
= = 3.6m!
1
2π × × 10− 9 × 2.5 × 106
3επ
Lightning 6/02 - 10
Lightning
What magnitudes of stroke current
will hit the line?
100 kA
kA
0
0 2 NS TIME
Lightning 6/02 - 11
Stroke Waveshape Parameters
Influencing Flashover
Lightning 6/02 - 12
Maximum Stroke Magnitudes
in a Flash
P I = 1 / [ 1 + ( I/3 1 ) 2 .6 ]
W h e re :
P I = p r o b a b ilit y t h a t a c h o s e n m a x . s t r o k e
c u r r e n t , I, in k A in a g iv e n f la s h w ill b e
exceeded.
F o r a 1 0 0 - m ile lin e o v e r 2 0 y e a r s , a t 0 .7 8
h it s /m ile /y r ., 1 5 6 0 f la s h e s w ill h it t h e lin e
( a v g .) .
Then:
7 8 0 f la s h e s w ill e x c e e d 3 1 k A
7 1 f la s h e s w ill e x c e e d 1 0 0 k A
Lightning 6/02 - 13
1 2 f la s h e s w ill e x c e e d 2 0 0 k A
Stroke Statistics
• 55% of flashes > 1 stroke
+++++++++++++++++++++++++
• 90% of flashes < 9 strokes --- +++
-- -- -- -- -- -- -- -- -- -- -- -- -- --
• Mean no. strokes/flash ≈ 3 -
-
-
• In summer, ≈ 80-90% strokes
will be negative +++++++++++++++++++++++++
• In winter, GFD << than --- +++
summers ----------------------------
Lightning 6/02 - 14
Lightning Strikes to Earth
Isokeraunic Map of the U.S
Lightning 6/02 - 15
How Many Hits to a Line?
28h 0.6 + B
N L = GFD
10
NL = # hits/100 km/yr. b
GFD = ground flash density,
flashes/km2/yr. h
B = spacing between shield wires*
*If no shield wires, B is the distance
between the outer phases and h is the
height at the phase wires.
Lightning 6/02 - 16
Lightning
Evolution of a Shielding Failure
Lightning 6/02 - 17
Lightning
Evolution of a shielding failure
Descending Leader
S = 10 I0.65
Phase A
When S = Strike distance, meters.
BS I = Current in kA that stroke will deliver.
B = 0.6 to 1.0
Lightning 6/02 - 18
Lightning
Geometry of a shielding failure
1
2
S
Shield Wire 3
S
Phase S
BS
Uncovered Area
Lightning 6/02 - 19
Leader Progression Model
Lightning 6/02 - 20
Leader Progression Model
z 500 kV Outages Studied for the past 4 years
z 14 Total Lightning Related Outages
z 7 Caused by Stroke Currents too low to cause
outages
z Existing Model do not predict these outages
Lightning 6/02 - 21
Lightning
z What Minimum
Lightning Current is
Required for Flashover?
z 50 kA?
z 100 kA?
z 200 kA?
Lightning 6/02 - 22
Lightning
z What typical lightning
current will cause a
flashover?
Line Voltage Class Stroke Current
765 kV 40 kA
500 kV 32 kA
345 kV 20 kA
230 kV 10 kA
115 kV 7 kA
69 kV 5 kA
Lightning 6/02 - 23
Leader Progression Model
z The existing IEEE Electrogeometric model
does not correctly account for lightning channel
leader progression near the conductors
z With a zero degree or negative shielding angle,
assuming straight vertical leaders, a shielding
failure is never calculated
Lightning 6/02 - 24
Leader Progression Model
Lightning 6/02 - 25
Leader Progression Model
Lightning 6/02 - 26
Leader Progression Model
Lightning 6/02 - 27
Leader Progression Model
Lightning 6/02 - 28
Leader Progression Model
Lightning 6/02 - 29
Leader Progression Model
Lightning 6/02 - 30
Leader Progression Model
Lightning 6/02 - 31
Leader Progression Model
z The new model shows many more shielding
failures may occur than any other calculation
methods presently predict
z If the model is accurate, present lightning
mitigation measures may not be addressing
the correct problem
z Reduced ground impedance will not help
z Arresters may take substantially more energy
than anticipated
Lightning 6/02 - 32
Wave Propagation Fundamentals
z For an ideal conductor and an ideal ground plane
L∆ x L∆ x L∆ x L∆ x
V C∆ x C∆ x C∆ x C∆ x C∆ x C∆ x
Load
∆x
Lightning 6/02 - 33
Wave Propagation Fundamentals
d
−7 4h
L = 2 × 10 ln henrys / m
d
−
55.5 × 10 12
C=
4h
ln h
d
L
Surge impedance Z0 =
C
Magnetic field energy = 0.5 LI2 joules/m
Electric field energy = 0.5 CV2 joules/m
Where: V = voltage to ground, V
I = wire current, A
Lightning 6/02 - 34
Wave Propagation Fundamentals
V Z0 I = V / Z0
A. Line terminated by its surge impedance. (No reflections at
the receiving end)
Lightning 6/02 - 35
Wave Propagation Fundamentals
S
-V
V I = 2V / Z0
B. Line terminated by a short circuit (current doubles at the
receiving end. Opposite voltage is reflected back.
Lightning 6/02 - 36
Wave Propagation Fundamentals
V
V
C. Line terminated by an infinite impedance (voltage doubles
at the receiving end. Same polarity voltage is reflected back.
Lightning 6/02 - 37
Wave Propagation Fundamentals
z Each conductor has a set of forward buckets and
a set of backward wave buckets
Forward Currents
Incoming
Refract Out
Reflect Reflect
Refract Incoming
Out
Backward Currents
Lightning 6/02 - 38
Wave Propagation Fundamentals
z Reflect and Refract are functions that determine
how much current arriving at each end of the
conductor should be refracted out of the conductor
and how much should be reflected back into the
conductor.
Lightning 6/02 - 39
Wave Propagation Fundamentals
z How voltages couple between conductors
(2-wire case)
C10∆ x C10∆ x C10∆ x C10∆ x C10∆ x
I1 L11∆ x L11∆ x L11∆ x L11∆ x
1
I1
L12∆ x L12∆ x L12∆ x L12∆ x
C12 C12 C12 C12 C12
I2
2
L22∆ x L22∆ x L22∆ x L22∆ x I2
C20∆ x C20∆ x C20∆ x C20∆ x C20∆ x
Lightning 6/02 - 40
Wave Propagation Fundamentals
z Earth skin effect: For a 2 µs crest current wave
z A half period of 4 µs corresponds to a frequency of
125 kHz
0 2µs
Time
4µs
Lightning 6/02 - 41
Wave Propagation Fundamentals
z Earth skin effect: For a 2 µs crest current wave
z Skin depth of δ of the earth for an earth resistivity of
100 ohm-meters
1 Where:
δ=
πfµ f = frequency, Hz
µ = earth permeability = 4π x 10-7 henrys/m
ρ
ρ = earth resistivity, ohm-meters
d = 14 m for 125 kHz; d = 650 m for 60 Hz
As a fair approximation for lightning frequencies, we can ignore
the earth skin effect because impedances vary by the log of the
conductor height.
Lightning 6/02 - 42
Wave Propagation Fundamentals
z Conductors and the ground plane are not ideal
in the real world
z Conductors are surrounded by a corona envelop that
increases their capacitance and distorts the
waveshape
z The earth has a “skin effect” that moves the effective
ground plane down into the earth. This changes the
inductance and increases resistance.
Lightning 6/02 - 43
Wave Propagation Fundamentals
z Tower surge impedance
z The tower as a short, vertical transmission line
z The principle behind nanosecond geometrical
models:
If a 50:1 scale model of a tower is impulsed with a
current wave that is 50 times faster than a current
wave injected into the full-scale model, all impedances
in the model will equal the same impedances in the
full-scale tower.
Impedances as ratios of voltage to current are
relatively easy to measure.
z Tower surge impedances are influenced by injected
current waveshapes.
Lightning 6/02 - 44
Tower Surge Impedance
Zs
2 V ZT
I
Tower
Surge
Impedance
Zs Zs •Stroke current injected into
the tower top creates a
traveling wave of voltage on
V
the shield wire.
R ZS V
ZT = Where R=
ZS − 2R I
Lightning 6/02 - 45
Nanosecond Geometric Models
Downward Leader
( A Helix with Dipole )
Charging Resistor
100 V DC Supply Mercury Switch
Probe
Scope
50:1
Scale
Model
Tower
Footing Resistance
Metal Ground Plane
Lightning 6/02 - 46
Wave Propagation Fundamentals
z The Chisholm tower surge impedance equations:
z Average tower radius
Ravg = R1h2+ R2 (h1+h2) + R3h1
h1 + h2
Where: Ravg = average tower radius
R1 = radius at the tower top
R2 = radius at the tower midsection
R3 = radius at tower base
h1 = height from base to midsection
h2 = height from midsection to top
z Tower surge impedance Zavg
1 −1 R avg
Zavg = 60 ln cot tan
2 h1 + h 2
Lightning 6/02 - 47
Surge Impedance of Conductor
Stroke Current
IF
IS IS
Shield Wire
VP
P Phase
IT = IF - 2 I S
ZP
VP = (IT = IF - 2 I S)
R
Resistive component of insulator voltage.
Lightning 6/02 - 48
Surge Impedance of Conductor
z How a transformer works.
V
Voltmeter
V = Total rate of change of magnetic flux in loop A.
Lightning 6/02 - 49
Surge Impedance of Conductor
Stroke Current
IF
IS IS
Shield Wire
VP
P Phase
IT = IF - 2 I S
do A ZP
dt
VP = A d o
dt
Inductive component of insulator voltage.
Lightning 6/02 - 50
Surge Impedance of Conductor
z How coupling Cx modifies insulator volts.
Stroke Current
IS IF I S
Shield Wire
C1
C2
V1
V2
P1 Phase
P2
Z P1
Z P2 IT = IF - 2 I S
R
The resistive component of V2 will be larger
than V1 because C2 is smaller than C1.
Lightning 6/02 - 51
Simple Equivalent Circuit
for a Hit to a Tower
ZP = Phase wire surge imp.; Zs = Static
Lightning wire surge imp.
Current
Generator Static Z5P= Mutual impedance between shield
Tower Top Wires
wire & phase
I5
VT= Tower top voltage; LT = Tower
Zs Crossarm inductance
Z5P RT = Footing resistance; I = Phase
I VT
LT insulator
ID Zs
VI = voltage across phase ins.; ID =
VT
Phase reflected current from adj. towers
ZP
IT
Lightning 6/02 - 52
Volt-time Curves
of Insulators and Air Gaps
3000 kV Negative Polarity
Lightning Surge
Voltage
Gap kV
2000 kV
1000 kV
Applied
Voltage
Breakdown
Time
0 1 2 3 4 5 6
Lightning 6/02 - 53 Time to Breakdown ( µ s)
Impulse Impedance of Grounds
•Impulse breakdown of soils
•Low-current, Low-frequency resistance:
Ro = Po [log (4L/ro)-1] / 2π L
I
Rod with radius Ro and length L
Earth
P = Po = Earth resistivity
( ohm-meters )
E = Earth Dielectric Constant
Arc
Streamers
Electrolytic Conduction
Lightning 6/02 - 54
Impulse Impedance of Grounds
The simple model
I
Soil breaks down set to a
Eo r
distance r where the electric
field gradient is equal to the
critical breakdown gradient
Eo of the soil.
Thus, at high currents the rod can be modeled
as a hemispherical electrode.
Lightning 6/02 - 55
Soil Ionization Model
z The ground resistance value reported by the line
construction and maintenance crews is not
representative of the dynamic impedance during a
lightning event
z Existing nonlinear Weck model is better than the
meggered value but it does not represent the true
dynamic impedance
Ro
R= ρEs
I IC = Es = Soil Ionization Gradient
1+ 2πRo2
IC
Lightning 6/02 - 56
Soil Ionization Model
Lightning 6/02 - 57
Soil Ionization Model
z The impedance reduction depends on
z Electrode configuration
z Soil type
z Lightning current
Lightning 6/02 - 58
Lightning
z The soil ionization impedance reduction effects
which phase backflashes
z Lower impedance causes higher phases to
flashover
z Higher impedance causes lower phases to
flashover
z Backflash location may determine arrester
placement
Lightning 6/02 - 59
Soil Ionization
Lightning 6/02 - 60
Soil Ionization
Lightning 6/02 - 61
Soil Ionization
Lightning 6/02 - 62
High Current Response
of Ground Rods
I
•Inside streamers are shorter
than outside streamers
Earth
Breadown
Streamers
Lightning 6/02 - 63
High Current Response
of Ground Rods
Wood Pole
Ground Rod Ground Rod
•Earth breakdown
streamers
can only form in the
direction
away from the pole
Earth
Breakdown
Lightning 6/02 - 64 Streamers
Impulse Impedance of Grounds
I
•Installation of Ground
Electrodes
-Rods
Concrete -Concrete Pipes
Lightning 6/02 - 65 20 cm
Impulse Impedance of Grounds
Puncture of concrete foundations
Puncture from
reinforcing
to water pocket
Water
Pocket
Drive ground rod near concrete foundation to prevent
puncturing of the concrete.
Lightning 6/02 - 66
Impulse Impedance of Grounds
Counterpoise
–One or more long
or continuous ground rods laid horizontal
–Continuous counterpoise
1-2m
Lightning 6/02 -Continuous
67 Counterpoise
Impulse Impedance of Grounds
Radial counterpoise
Usually much better than
continuous counterpoise
5 - 50 meters
Lightning 6/02 - 68
Induced Voltage Flashovers
z Surge impedance equivalent circuit
I F
IB
A
+
IC
V1 Z0 V2 h
Z0
IC
∆x
z Inducing magnetic field H induces circulating current
Ic in the loop: µh dH
IC = ∆X
2 Z0 dT
Lightning 6/02 - 69
Induced Voltage Flashovers
z Surge impedance equivalent circuit
IF
IB
A
+
IC
V1 Z0 V2 h
Z0
IC
∆x
z Ic
can be separated into a forward current, IF, and a
backward current, IB, which are equal.
IF - IB = IC
Lightning 6/02 - 70
Induced Voltage Flashovers
z Surge impedance equivalent circuit
IF
IB
A
+
IC
V1 Z0 V2 h
Z0
IC
∆x
z ICand IF flow into other wire sections to the left and
right, respectively
z V1 and V2 are equal and opposite
Lightning 6/02 - 71
Induced Voltage Flashovers
z Surge impedance equivalent circuit
IF
IB
A
+
V1 Z0 V2 h
Z0 V0
∆E
∆x
z Inducing vertical electric field ∆E for a time ∆t
z Vo = (∆E)h until charges flow upward through the
surge impedance to move the voltage back toward 0
Lightning 6/02 - 72
Induced Voltage Flashovers
IF
IB
A
+
V1 Z0 V2 h
Z0 V0
∆E
∆x
z But V0 = (IF - IB) Z0
z Since IF must equal IB in magnitude and V0 is not 0,
IB = - IF and IF = (∆E)h/2Z0 = - IB
Lightning 6/02 - 73
Induced Voltage Flashovers
z Line with BIL>300 kV are unlikely to experience induced
flashovers
z Simplified form of the Rusck equation:
38.8I max h
v=
Where:
y
v = voltage induced on a conductor, kV
Imax = crest kA of the stroke current
h = conductor height, m
Y = horizontal dist. from the conductor to the inducing stroke, m
Lightning 6/02 - 74
Induced Voltage Flashovers
Example:
For a 50 kA stroke 130 m from a 10 m high conductor:
v=
(38.8)(50 )(10 )
= 149 kV
130
For a 200 kA stroke 315 m from a 10 m high conductor:
v=
(38.8)(200 )(10 )
= 246 kV
315
Lightning 6/02 - 75
Using Masts
to Divert Lightning
The Old Shadow Method
Hit to
Mast Hit to •For a high amplitude
Ground
flash, 2h was used
• For a weak flash, it
might be closer to h,
h instead of 2h
Zone of
Protection
Lightning 6/02 - 76
2h 2h
Using Masts
to Divert Lightning
Eriksson’s Equation
I
S S = 0.84h0.6 I0.74
Where: S = strike distance, m
H = mast height, m
I = stroke current, kA
h
Lightning 6/02 - 77
Lightning Flashover Types
z Backflash – caused by stroke to tower or shield
wire
z Shielding Failure – caused by stroke to phase
conductor
z Induced Flashover – caused by nearby stroke
to another object or ground
Lightning 6/02 - 78
Lightning Mitigation Options
z Mitigation Options
z Shielding
z Grounding
z Insulation
z Surge Arresters
z Double Circuit Strategies
Lightning 6/02 - 79
Lightning Calculation Parameters
•Ground flash densities
•Stroke current waveshapes
•Lightning stroke multiplicity
•Magnitude probabilities
•Stroke polarities
•Multiple stroke models
•Charge delivered
Lightning 6/02 - 80
Lightning Calculation Parameters
•Leader model and channel geometry
•Return stroke model
–Return stroke velocities
–Electromagnetic field
simulation
–Ground effects
Lightning 6/02 - 81
Lightning Calculation Parameters
•Lightning attraction to lines
–Electrogeometric models
–Effects of line height
–shield angles
–Midspan, quarterspan, and tower hits
•Exposure and natural shielding
–Hills
–Valleys
–Sloping hillsides
–Trees and Natural Shielding
Lightning 6/02 - 82
Lightning Calculation Parameters
•Development of insulator voltages
–Footing resistance contribution
•Linear
•Non-linear
•Electrode geometries
–Electromagnetic field contribution
•Tower contribution
•Shield wire non-linear coupling
•Induced voltages
•Stroke channel impedances
Lightning 6/02 - 83
Lightning Calculation Parameters
•Development of insulator voltages
–Insulator/air gap flashover
•Power frequency effects
•Non-standard waveshapes
•Complex gap geometries
•Midspan flashover/predischarge effects
–Effects of adjacent towers
–Multi-phase/multi-circuit flashover
–Arrester applications
Lightning 6/02 - 84
Lightning Eliminators:
Do They Work?
Two claims have
been made:
•They neutralize the
charged clouds
overhead
One of
many
•They create an ion path versions
above the eliminator
that diverts lightning
that would otherwise hit
unprotected objects.
Lightning 6/02 - 85
Lightning Eliminators:
Do They Work?
Abdul M. Mousa, “The Applicability of Lightning Elimination Devices
to Substations and Power Lines,” IEEE Transactions on Power
Delivery, Vol. 13, No. 4, Oct. 1998m pp. 1120-1127. Available from:
[email protected])
•Paper was peer-reviewed by six reviewers under threat of
lawsuits. Paper states that these devices do not work as the
manufacturers claim.
1997 Report on Dissipation Arrays, funded by FAA, Naval Research
Labs, NASA, and USAF
•The report, 274 pp., compiled by 17 scientists and engineers
from around the world, provides no definitive physical or
theoretical evidence that lightning dissipation arrays prevent
lightning. The USAF presented photos showing the arrays being
hit by lightning. Contact Mousa for more information.
Lightning 6/02 - 86
