IEEE 99 T&D Conference

Panel session "Distribution Lightning Protection".

New Orleans, April 14, 1999

Lightning-Induced Voltages
C.A. Nucci

F. Rachidi

Dept. of Electrical Engineering

Swiss Federal Inst. of Technology

Power Systems Laboratory

University of Bologna
40136 Bologna, Italy
[email protected]

1015 Lausanne, Switzerland.

[email protected]

1. Aim of Presentation

To give some answers to the questions that are

most commonly raised by engineers and
scientific researchers dealing with the problem of
protection against lightning-induced voltages.

2. What causes induced voltages ?

Electromagnetic coupling between the field radiated
by a lightning stroke and the line conductors
TOTAL FIELD
Incident field

Scattered field

2. What causes induced voltages ?

Cont.

Essentially the return-stroke phase is responsible of

the induced voltage
However, when lightning strikes the ground nearby
the line at close distance from the line, also the
preceding leader phase can results in a significant
induced voltage

3. How to evaluate them?

Return-Stroke Current
i (0,t)

RSC

i (z,t)

Lightning ElectroMagnetic Pulse (appr. expr. for E horiz.)

i (z,t)

LEMP

E, B

ElectroMagnetic Coupling
E, B

EMC

V, I

3. How to evaluate them?

Cont.

Return-stroke current models

A review of the various return-stroke models has been

recently made by Rakov and Uman on
IEEE EMC Transactions, Special Issue on Lightning,
1998 where they have discussed, among others, the
following engineering models

Bruce-Golde (BG)
Transmission Line (TL) Uman, McLain, Krider
Traveling Current Source (TCS) Heidler
Modified Transm. Line - Linear (MTLL) Rakov and Dulzon
Modified Transm. Line - Exponential (MTLE) Nucci et al.
Diendorfer-Uman (DU)

3. How to evaluate them?

Cont.

Return-stroke current models

Experimental validation
Given a channel-base current ==>
the RSC model must reproduce the
corresponding Electromagnetic field
For Natural lightning:
PROBLEM: practically no existing data sets of
simultaneously measured current and fields
Data of this kind have been collected using
the Triggered lightning technique

3. How to evaluate them?

Return-stroke current models
n

TRIGGERED LIGHTNING:
Lightning is artificially
initiated firing small rockets
trailing grounded wires
upward a few hundred
meters under
thunderstorms.

Cont.

3. How to evaluate them?

Cont.

Return-stroke current models

Triggered lightning: A sequence of frames

3. How to evaluate them?

Cont.

V/m

Return-stroke current models

TCS
TCS

MTL

a)

b)

microseconds

microseconds

Validation by means of triggered lightning

3. How to evaluate them?

Cont.

Electromagnetic coupling

Basically, three coupling models have been used:

Rusck

[1958]

Chowdhuri [1969]
Agrawal et al. [1980]
Of the three models only the Agrawal one is
rigorous for a general external field excitation
However, for a lightning channel perpendicular to
the ground plane ===> Rusck = Agrawal

3. How to evaluate them?

Cont.

The Agrawal model: Experimental validation

Reduced scale model at the University Of So Paulo - Brazil

3. How to evaluate them?

Cont.

The
Agrawal model: Experimental validation
140
120
CALCULATED
MEASURED

100

OBSERVATION POINT

80
60

70 m
130 m

40

40 m

20
0
0

10

12

Time in s

Using reduced-scale line model

Experimental data: by A. Piantini, Univ. Of So Paulo

100 m

3. How to evaluate them?

Cont.

The Agrawal model: Experimental validation

160

Calculated dV/dt
Measured dV/dt
Measured dV/dt (filtered)

140
120
100
80
60
40
20
0
-20
-40
-60

0.5

1.5

2.5

3.5

Time in us

Using reduced-scale line model

Experimental data: by A. Piantini, Univ. Of So Paulo

3. How to evaluate them?

Cont.

The Agrawal model: Experimental validation

Using NEMP simulators

3. How to evaluate them?

Cont.

The Agrawal model: Experimental validation

Using NEMP simulators

3. How to evaluate them?

Cont.

Which component of the LEMP does affect

most the induced voltages?
Vertical E component?
Horizontal E component?
Other components?
Let us assume, for simplicity, a lossless line

3. How to evaluate them?

i(x)

L'dx

E ix dx
-

-u i (0)

u (x)

Cont.

i(x+dx)

C'dx

u s (x+dx)

RL

R0
0

-u i (L)

u (x)

Agrawal et al.

x+dx

3. How to evaluate them?

i(x)

L'dx

E ix dx
-

-u i (0)

u (x)

Cont.

i(x+dx)

C'dx

u s (x+dx)

RL

R0
0

-u i (L)

u (x)

Agrawal et al.

x+dx

3. How to evaluate them?

Cont.

80

60

Total
E i dx
i(x) L'dx x
+
- +
-ui(0,t) s
C'dx
u (x)
R0

40
E X contribution

20

0
E contribution
-20
0

8
Time (s)

ui(x)

i(x+dx)
+
us(x+dx) -u i(L)
RL
x+dx

3. How to evaluate them?

Cont.

E i, B i

y
h

i
i
B y ( x, z, t )dz = E x ( x, h, t ) +
E zi ( x, z, t )dz
t
x

Nucci and Rachidi, IEEE Trans. on EMC,

Vol. 37, No. 4, November 1995.

3. How to evaluate them?

i(x)

L'dx

d i
dt (B y (x,z) dz)dx
-

u (x)
R0
0

Cont.

i(x+dx)

d
-C (E i (x,z) dz)dx
dt z

C'dx

u (x+dx)
RL

x+dx

Taylor et al.

3. How to evaluate them?

Cont.

80

60

Total
Ez contribution

d (B i (x,z) dz)dx
dt y
i(x) L'dx
i(x+dx)
+
C'dx
u (x)
u (x+dx)
R0
d i
-C (E (x,z) dz)dx
dt z

40

20

-20
By contribution
-40

-60

x+dx

RL
L

3. How to evaluate them?

Cont.

1 B i (x,0) dz
y
L
0

u (x)
0

i(x)

L'dx

1 B i (x,L) dz
y
L
0

h i
1 B x(x,z) dz ]
[- y
L
0

Rachidi

C'dx
RL

3. How to evaluate them?

Cont

80

60

i
dz
-1
B y(x,0)
L
i(x+dx)

B iy(x,0) dz

-1
L0

Total

i(x) L'dx

u (x)

40

C'dx u (x+dx)
RL

R0

BX contribution

Bx (x,z)
[-1
dz ] dx

y
0

20

-20
0

By contribution
2

Time (s)

x+dx L

3.

Cont.

The contribution of a given electromagnetic field

component in the coupling mechanism depends
strongly on the used model.
Thus, when speaking about the contribution of a
given electromagnetic field component to the
induced voltages, one has to specify the coupling
.

4.

are
for lightning induced
Induced voltage magnitude and shape significantly
depend on
lightning return stroke parameters (channel-base
current parameters, return stroke velocity),
distance and relative position with respect to the
transmission line,
line configuration and terminations.

Induced overvoltages can reach magnitudes up to few

hundreds of kV and can therefore cause line flashover.

4. What magnitudes and shape are typical

for lightning-induced overvoltages ? Cont.
I in kA

12
10
8
6
4

1 km

2
0

4
Time in us

Shape

60
U in kV 50
A
40
30
20
10
-10
-20
0

3
4
Time in us

However ...

4. What magnitudes and shape are typical

for lightning-induced overvoltages ? Cont.
60
U in kV 50
A
40
30
20
10

3
4
Time in us

4
3
Time in us

-10
-20
0

0
2
4
6
8
12
I in kA
10

4. What magnitudes and shape are typical

for lightning-induced overvoltages ? Cont.

1 km

Magnitude

However ...

4. What magnitudes and shape are typical

for lightning-induced overvoltages ? Cont.
For perfectly conducting grounds and for an
infinitely long wire

U max

I max h
= Z0
d

where Z 0 = 1 / 4 0 / o = 30

the simplified Rusck formula allows for a

satisfactory estimation
HOWEVER

5. How far away can lightning strokes be that

cause an induced voltage flashover ?
Generally within 200 m
However it depends on many parameters
(=> computer code)
Lightning strokes occurring beyond a few
hundred meters from the line can cause a line
flashover for poor conducting soils (De La
Rosa et al., IEEE Trans. on PWDR, 1988)

6. How does the induced voltage drop as a

function of distance from the line ?
U A in kV
90
80
70

30m

60
60m

50
40

90m

30
20
10

1 km
-10
-20
0

Time in us

Perfectly conducting ground and stroke location equidistant to the

line termination => nearly proportionally to 1/d
HOWEVER ...

7. What BIL is needed to prevent induced

flashovers ?
No. of events having amplitudes exceeding
the BIL / (100 km year)

1000.00

100.00

10.00

1.00

perfect ground

0.10

ground conductivity = 0.01 S/m

ground conductivity = 0.001 S/m
0.01
50

100

150

200

250

Basic Insulation Level [kV]

300

8. What arrester spacing is needed to prevent

flashover?
In order to prevent direct-stroke flashover, arrester
spacings of 300-400 m is generally recommended.
For the case of induced flashovers, a given
configuration of line arresters can result in different
performances depending on the location of lightning
strike [29].
Further studies are needed in this respect.

8. What arrester spacing is needed to prevent

flashover?

120

Stroke location: B1
100
80
60
40
20
0
X0

a)

Transf.
Transf. + Surge arr.
Matched

X1

X2

X3

X4

X5

8. What arrester spacing is needed to prevent

flashover?

400

Stroke location: E1

350
300
250
200
150
100

Transf.
Transf. + Surge arr.
Matched

50
0
X0

b)

X1

X2

X3

X4

X5

8. What arrester spacing is needed to prevent

flashover?

400

Stroke location: H1

350
300
250
200
150
100

Transf.
Transf. + Surge arr.
Matched

50
0
X0

c)

X1

X2

X3

X4

Observation Point Along the Line

X5

9. Will a shield wire help ?

4
3.7 m
3.7 m

3
3.7 m
5

2
3.7 m

3.7 m
1

3.7 m 3.7 m
10 m

10 m

b)

a)

9. Will a shield wire help ?

Cont.

Shield wires help in reducing the magnitude of

induced voltages by a factor of about 20 to 40 %.
wire
1
2
3

Protective
Ratio
PR
PR
PR

Vertical
Config.

Horizontal
Config.

0.81

0.67

0.78

0.60

0.72

0.67

This implies about the same reduction of the fault

frequency.

10. Is horizontal or vertical construction

best ?
4
3.7 m
3.7 m

3
3.7 m
5

2
3.7 m

3.7 m
1

3.7 m 3.7 m
10 m

10 m

b)

a)

10. Is horizontal or vertical construction

best ?
Cont.
Ratio between peak values of the induced voltages on a line
conductor Vi and those corresponding
to a single-conductor line of the same height V(hi).

Voltage
Ratio

Vertical
Configuration

Horizontal
Configuration

V1/V(h1)

0.75

0.85

V2/V(h2)

0.79

0.81

V3/V(h3)

0.89

0.85
a)

10. Is horizontal or vertical construction

best ?
Cont.
The induced voltage magnitude for typical distribution
lines is virtually proportional to the line height.
As a consequence, an important factor determining the
magnitude of lightning-induced voltage is the line
height above ground, rather than the type of
construction.
In general, a construction allowing a shorter height for
the conductors is epected to experience lower induced
overvoltages.

11. What effect does pole grounding and

ground resistivity have ?
Influence of pole grounding

Pole grounding affects the performance of the ground

wire in reducing the induced overvoltages.
In general, lower the pole ground impedance, better
the performance of the ground wire.

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

The ground resistivity affects:

1. electromagnetic field
2. propagation of the surges

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

Cont.

Perfect ground

Lossy ground:
Lossy ground

microseconds

= 0.001 S/m

0.5 km

0.5 km

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

Cont.

Lossy ground
( = 0.001 S/m)
Ideal field

Ideal field:
Lossy ground
Ideal line

ground resisitivity only in

the expression of the
ground impedance

Ideal line:
b)

ground resistivity only in

the expression of the
incident field

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

Cont.

Lossy ground
( = 0.001 S/m)

Ideal field

Ideal field:
ground resisitivity only in
the expression of the
ground impedance

Ideal line:

Lossy ground

ground resistivity only in

the expression of the
incident field

Ideal line

microseconds

2.5 km

2.5 km

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

How does the ground resistivity affect

magnitude and shape of the induced
voltages?

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

Cont.

110

=0.01 S/m
90

infinite

70

50

0.5 km

0.5 km

30
0

200

400

600

800

Obs. point along the line in m

1,000

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

Cont.

40

infinite

20

= 0.01 S/m
-20
1 km

-40

200

400

600

800

Obs. point along the line in m

1000

11.What effect does pole grounding and

ground resistivity have ?
Cont.
Influence of ground resistivity

Cont.

The ground resistivity can increase or

decrease the magnitude of the induced
voltages depending on the stroke location
and the observation point along the line
=> the calculation is not trivial

12.Some of the current researches which are

being done on induced voltages
- Lightning electromagnetic field characterization
using natural and artificially-initiated lightning
- Experimental validation of field-to-transmission line
coupling models
- Development of engineering tools for the protection
of power networks against lightning-induced
overvoltages
- Leader-induction effect
- Effect of ground conductivity on lightning-induced
overvoltages

The LIOV-EMTP code

LIOV-line

0
Overhead
Distribution Line
n-port

The LIOV code calculates:

LEMP (using the MTL model and Cooray-Rubinstein expr.)
Coupling using the Agrawal model.
The EMTP :
calculates the boundary conditions
makes available a large library of power components

The LIOV-EMTP code

Contd

LIOV has been developped within the framework

of an international collaboration involving
University of Bologna
Swiss federal Institute of Technology (EPFL, Lausanne)
University of Roma La Sapienza
Its link with EMTP has been realized in collaboration with
ENEL-CESI (Univ. Bologna)
Other methods have been proposed
EdF (EPFL )

The LIOV-EMTP code

i0 u 0

i(x,t)

Contd
E ix (x,h,t)dx

L'dx

u s (x,t)

i(x+dx,t)

u s (x+dx,t)

C'dx

+
-u i (0,t)
-

u =u +u =u
t

Bergeron
linea Bergeron
u1
u 1'
line
i0'

i0

LIOV
linealine
LIOV

u2
i1

i
i2

-u

E zi ( x , t )dz
u1

u1'
Zc

-u (0,t)
-

i0

Zc

u2

0'

+
i (0,t)

x+dx

u i (x,t)

+
V1'

Link between LIOV and EMTP

i1

i2

The LIOV-EMTP code

Contd

u s ( x , t )
i( x , t )
+ L
= E xi ( x , h, t )
x
t
i ( x , t )
u s ( x , t )
+ C
=0
x
t
h

u s ( 0, t ) = u1 ( t ) = o (i( 0, t )) + E zi ( 0, z , t )dz
0

u1 ( t ) = Z c i0 ( t ) + [u1' ( t t ) + Zc i0' ( t t )]
= Z c i0 ( t ) + V1 ( t t )

u1' ( t ) = Zc i0' ( t ) + [u1 ( t t ) Zc i0 ( t t )]

= Zc i0' ( t ) + V1' ( t t )

12.Some of the current researches which are

being done on induced voltages
Cont.

Validation using data from a more complex system

Data: courtesy of Dr. A. Piantini, Univ. Sa Paulo

12.Some of the current researches which are

being done on induced voltages
Cont.
200

Calculated voltage
Measured voltage

150
100
50
0
-50
-100
0

Time in us

Validation using data from a more complex system

Data: courtesy of Dr. A. Piantini, Univ. Of Sa Paulo

12.Some of the current researches which are

being done on induced voltages
Cont.
calculated voltage
Measured voltage

60
50
40
30
20
10
0
0

Time in us

Validation using data from a more complex system

Data: courtesy of Dr. A. Piantini, Univ. Of Sa Paulo

12.Some of the current researches which are

being done on induced voltages
Cont.

- Return stroke modeling and influence of elevated

strike objects on lightning current and radiated fields
- Adequacy of the available lightning return stroke
current statistical data
- Effect of corona on lightning-induced voltages
- Lightning detection and location systems
- Lightning channel tortuosity and inclination

13.What is the CIGRE working group doing on

induced voltages ?
Within the framework of
CIGRE working group WG 33.01 "Lightning",
Task Force 33.01.01 "Lightning induced voltages"
established some years ago.
C.A. Nucci (responsible member), P. Chowdhuri, G. V.
Cooray, M.T. Correia de Barros, M. Darveniza, F. De la Rosa,
G. Diendorfer, F. Heidler, M. Ishii, W. Janischewskyj, T.
Kawamura, C. Mazzetti, P. Pettersson, F. Rachidi, V. Rakov,
M. Rubinstein, T. Short, J.V. Shostak, M.A. Uman, S.
Yokoyama

13.What is the CIGRE working group doing on

induced voltages ?
TF 33.01.01 has already produced two papers
published in Electra dealing respectively with
lightning return stroke models (August 95) and
lightning electromagnetic field-to-transmission line
coupling models (October 95).
A third paper, dealing with a sensitivity analysis and
aimed at providing ranges of overvoltage values to
be expected in the different typical line
configurations, is in preparation.