You are on page 1of 10

Copyright © 2011, Praise Worthy Prize S.r.l..

Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
A method for the sag-tension calculation in electrical overhead lines


I. Albizu
1
, A.J. Mazon
1
, E. Fernandez
1



Abstract – The sag and tension values of overhead conductors are influenced by the creep
developed during the line lifetime. This paper presents a method for the sag-tension calculation of
overhead conductors that is characterised by the creep sequential calculation. Thus, the creep
developed in previous stages influences the creep developed in subsequent stages. Two periods
are differenced in the creep development: the installation period and the operation period. The
relation between the creep development and the factors that influence it such as the installation
process and the operation conditions during the line lifetime is described step by step. Copyright
© 2011 Praise Worthy Prize S.r.l. - All rights reserved.

Keywords: Sag-tension, Overhead conductor, Creep


I. Nomenclature
T tension
L length
A area
E elastic modulus
α coefficient of thermal expansion
ω weight
θ temperature
s span length
σ stress
ε total strain
ε
θ
strain related to temperature
ε
T
strain related to tension
ε
mc
strain related to metallurgical creep
ε
gs
strain related to geometrical settlement
t time
g related to span geometry
c related to conductor
core related to core
a related to aluminium
o related to reference condition
II. Introduction
The aim of the sag-tension calculation is the
calculation of the installation tension as a function of the
sag and tension limits. Sag-tension calculation methods
allow the calculation of the conductor sag and tension for
different conductor temperatures, and wind and ice load
conditions, taking into account the evolution of the
conductor creep during the line lifetime [1]. The tension
is limited by the tension limit of the conductor and the
towers. The sag limit is related to the security distance to
ground and line crossings. If the crossing distance is
below the security distance, line faults could occur [2-5].
The most complete methods are those that consider an
independent core and aluminium behaviour and obtain
the creep value from experimental tests [6-8]. These
methods calculate the aluminium tension and for this
reason they calculate the knee-point temperature where
the aluminium gets slack.
The most widely used method is the graphical method
[6] that is implemented in commercial software programs
such as SAG10 or PLS-CADD. This method is based on
stress-strain and creep experimental curves. The core and
the aluminium curves are obtained separately.
The strain summation method [7] was proposed as an
alternative to the graphical method. The method is
characterized for having the conductor strain as the
dependent variable. The strain can be caused by tension,
temperature and creep. The creep is the result of the
addition of metallurgic creep and geometrical settlement.
Each of these strains is evaluated individually and they
are added to obtain the total strain.
In [8], the authors developed a method for the special
requirements of the gap-type conductors. The method
was based on the strain summation method but some
changes and improvements were carried out. For
example, independent core and aluminium reference
lengths were considered and the calculation of the creep
developed during the installation was defined in detail.
The authors have carried out the generalization of the
method. As a result, the method is valid not only for gap-
type conductors but for any type of conductor, including
the high temperature low sag (HTLS) conductors [9-12]
and the conventional conductors such as the ACSR. The
main advantage of the method is the flexibility for the
consideration of several creep stages and the ability to
take into account the influence of previous creep stages.
This paper gives a detailed description of the method
and it includes diagrams that relate all the parts of the
calculation algorithm. The relation between the outputs
and the inputs of the algorithm parts are clearly defined
from the installation of the conductor to the end of the
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
line lifetime. The creep calculation during the installation
is fully described as a function of the type of conductor
(gap-type or non gap-type) and taking into account the
time the conductor is at rest and whether there is
pretensioning or not. The calculation of creep developed
during the operation is also described.
III. Sag-Tension Calculation Algorithm
The creep developed in previous stages influences the
creep developed in subsequent stages [13,14]. For this
reason, the algorithm makes a sequential calculation of
the creep. Two periods are differenced in the creep
development: the installation period and the operation
period (Fig. 1). The creep developed during the operation
depends on the creep previously developed during the
installation. The creep developed during each operation
stage is calculated taking into account the creep
developed so far.
In each stage, the metallurgical creep ε
mc
and the creep
due to geometrical settlement ε
gs
are calculated
separately. The metallurgical creep is calculated as a
function of the conductor tension T, the conductor
temperature θ and the duration t of the stage. The strain
due to geometrical settlement is assumed to be
independent of time. It is only dependent on the
conductor construction and the historical maximum
tension T
max
experienced. This calculation process is
carried out for the aluminium and the core separately.

 
gs
inst core,
c
Installation

gs
inst a,
c
mc
inst core,
c
mc
inst a,
c
max
,inst a
T
max
,inst core
T
gs
core 1 ,
c
Operation
stage 1

gs
a 1 ,
c
mc
core 1 ,
c
mc
a 1 ,
c
max
1 , a
T
max
1 , core
T
gs
n core,
c
Operation
stage n

gs
n a,
c
mc
n core,
c
mc
n a,
c
max
,n a
T
max
,n core
T
INSTALLATION OPERATION


Fig. 1. Creep and maximum tension evolution in time

When the method characterises the conductor
installation it differentiates between the gap-type
conductors and the rest of conductors.
The conductor temperature and the wind and ice loads
are assumed to be constant during each operation stage.
Thus, the parameters that characterise each stage i are the
following:
- Conductor temperature θ
i

- Load conditions (ice and wind)
- Duration t
i

From the creep strain values calculated for the
operation stages, the tension values related to the
maximum tension conditions are calculated by the state
calculation algorithm described in section IV. The
maximum tension conditions are characterised by the
conductor temperature and the ice and wind load values.
The tension limit values are defined for these conditions.
The conditions are related to specific stages (after the
installation, 10 year operation, etc.). Thus, from the
installation tension T
inst
, the algorithm calculates the
conductor tension in the defined maximum tension
conditions. The installation tension value T
inst
is iterated
until one of the maximum tension conditions does not
allow increasing its value. The state calculation
algorithm determines the conductor tension value
calculating separately the core tension T
core
and the
aluminium tension T
a
from the conductor temperature θ
i
,
the wind and ice load and the creep
mc
i core,
c ,
mc
i a,
c ,
gs
i core,
c
and
gs
i a,
c of the corresponding stage (installation, stage 1,
stage 2, etc.).
IV. State Calculation Algorithm
The state calculation algorithm is shown in Fig. 2. The
core tension T
core
is iterated until the difference between
the span geometry length L
g
and the conductor length L
c

is below a threshold value. The aluminium tension T
a

cannot go below its minimum value. This minimum
value is zero or a negative value if aluminium
compression is considered [15]. This is taken into
account in the algorithm when aluminium tension T
a
is
evaluated.

State calculation

gs
core
c
gs
a
c
mc
core
c
mc
a
c
u
Load
Core
(2,4)
Tcore
Aluminium
(1,3)
Span
geometry

Iterate on Tcore
until Lc = Lg
Lcore
La
Lg
Ta
Lc
+
+
Tc


Fig. 2. State calculation algorithm

The state calculation algorithm is based in the
dependence of the core and aluminium lengths L
core
and
L
a
on the strain values due to tension
T
core
c ,
T
a
c ,
temperature
core
u
c ,
a
u
c , and creep
mc
core
c ,
mc
a
c ,
gs
core
c ,
gs
a
c ,
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
and the core and aluminium reference lengths
core
o
L and
a
o
L (1,2).

( )
gs
a
mc
a a
T
a
a
o a
L L c c c c
u
+ + + · = (1)
( )
gs
core
mc
core core
T
core
core
o core
L L c c c c
u
+ + + · = (2)


As described in [8], the core and aluminium reference
lengths
core
o
L and
a
o
L correspond to the reference
condition with no tension and no creep. They are
obtained from the installation condition, where the
temperature and the tension values are known and the
creep strain is estimated from the installation process
(3,4). The span geometry is characterised by the catenary
equation. The catenary equation is a function of the
conductor tension T and the weight ω. The weight value
ω depends on the conductor weight ω
c
and the wind and
ice load. The catenary length L
g,inst
is obtained from the
installation tension T
inst
and the conductor weight ω
c
(5).

( )
gs
inst a
mc
inst a inst a
T
inst a
inst g
a
o
L
L
, , , ,
,
1 c c c c
u
+ + + +
=
(3)
( )
gs
inst core
mc
inst core inst core
T
inst core
inst g
core
o
L
L
, , , ,
,
1 c c c c
u
+ + + +
=
(4)
|
|
.
|

\
|
·
·
· · =
inst
c
c
inst
inst g
T
s T
L
2
sinh 2
,
e
e
(5)
V. Creep Developed during the
Installation
V.1. Gap-type Conductors
The gap-type conductors allow a relative displacement
between the core and the aluminium during the
installation of the conductor. In this way, the aluminium
gets slack during the installation and the knee-point
temperature is forced to be the temperature of
installation. Thus, the installation process of the gap-type
conductors is special and comprises several steps. For a
few minutes, during the pre-sagging step, around 70 %
of the sagging tension is applied only to the aluminium.
In the final installation step, the whole sagging tension is
applied only to the steel for a few hours, between 2 and
24 hours depending on the span. Due to this special
installation process, the total strains of the aluminium ε
a

and the core ε
core
have different values.
Fig. 3 shows a diagram of the algorithm that
calculates the creep developed by the aluminium during
the installation of the gap-type conductors. The input
values of the algorithm are the installation temperature
θ
inst
, the installation sagging tension T
inst
, the percentage
of the sagging tension T
a,inst
(%) the aluminium is
expected to support during the pre-sagging step (usually
70 %) and the duration t
pre-sagging
of the pre-sagging step.
The outputs of the algorithm are the maximum tension
max
, a inst
T experienced and the deformation due to
metallurgical creep
,
mc
a inst
c and geometrical settlement
,
gs
a inst
c .
From the installation sagging tension T
inst
and the
percentage of the sagging tension T
a,inst
(%), the
aluminium tension T
a,inst
is calculated.
The metallurgical creep
,
mc
a inst
c is calculated as a
function of the aluminium tension T
a,inst
the installation
temperature θ
inst
and the duration t
pre-sagging
of the pre-
sagging step. The metallurgical creep in the aluminium
follows the law given in (6), where K, Φ, β and μ are
constant coefficients that represent the behaviour of the
aluminium [14].

µ | u
o c t e K
mc
· · · =
u
(6)

The deformation due to geometrical settlement
,
gs
a inst
c
is obtained from the modified stress-strain curves of the
aluminium. The strain in the stress-strain curve [16] is
composed of three components related to the stress, the
geometrical settlement and the metallurgical creep
developed during one hour. To obtain directly the value
of the geometrical settlement from the curve, this is
modified removing the strain related to the metallurgical
creep. For this purpose, the metallurgical creep at the
stress-strain test temperature θ
s-s
and different stress
values σ is calculated by (6). Thus, a modified curve is
obtained as it is shown in Fig. 4.
The total strain ε
a
corresponding to the tension T
a
at
the installation temperature θ
inst
is obtained from the
modified aluminium stress-strain curve. The modified
stress-strain curve corresponding to the stress-strain test
carried out at the temperature θ
s-s
is displaced in the
strain axis in order to model the deformation
, a inst
u
c due
to thermal expansion (Fig. 5). This deformation
, a inst
u
c is
a function of the installation temperature θ
inst
, the
temperature θ
s-s
of the stress-strain test and the
coefficient of thermal expansion α
a
of the aluminium (7).
From the tension T
a,inst
and the area of the aluminium A
a
,
the aluminium stress σ
a,inst
is calculated and the strain ε
a

is obtained from the aluminium stress-strain curve (Fig.
5). The strain due to tension
T
inst a,
c is obtained from the
aluminium stress σ
a,inst
and the elastic modulus of the
aluminium E
a
(8). The deformation due to geometrical
settlement
,
gs
a inst
c is then calculated subtracting from the
total strain ε
a
the strain due to tension
T
inst a,
c and the
strain due to thermal expansion
, a inst
u
c (9).

Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
 
θ
inst

T
inst

mc
inst a,
c
gs
inst a,
c
Modified aluminium
stress-strain curve
(Fig. 5)
inst a,
c
T
inst a,
c
Thermal
expansion (7)
u
c
inst a,
T
a,inst

Elastic
deformation (8)
T
a,inst
(%)
Aluminium
pre-sagging
tension
σ
ε
Geometrical
settlement (9)

Metallurgical
creep (6)
t
pre-sagging

max
,inst a
T
o
c


Fig. 3. Creep developed in the aluminium of a gap-type conductor during the installation

 
σ
ε
modified
original


Fig. 4. Modified (1 h metallurgical creep removed) stress-strain curve

 
σ
ε
ε
θ
a,inst

σ
a,inst
ε
a,inst



Fig. 5. Modified (1 h metallurgical creep removed) aluminium
stress-strain curve

( )
s s inst a inst a ÷
÷ · = u u o c
u
,
(7)
a inst a
T
inst a
E
, ,
o c = (8)
u
c c c c
inst a
T
inst a inst a
gs
inst a , , , ,
÷ ÷ = (9)

Fig. 6 shows a diagram of the algorithm that
calculates the creep developed by the core during the
installation of the gap-type conductors. The input values
of the algorithm are the installation temperature θ
inst
, the
installation sagging tension T
inst
and the time the core is
at rest under installation sagging tension t
rest
. The outputs
of the algorithm are the maximum tension
max
, core inst
T
experienced and the deformation due to metallurgical
creep
,
mc
core inst
c and geometrical settlement
,
gs
core inst
c .
The metallurgical creep
,
mc
core inst
c is calculated as a
function of the core tension T
inst
, the installation
temperature θ
inst
and the time the core is at rest t
rest
. The
metallurgical creep in the core follows the law given in
(6), where K, Φ, β and μ are constant coefficients that
represent the behaviour of the core steel. These
coefficients are different from those previously given for
the aluminium.
The deformation due to geometrical settlement
,
gs
core inst
c is obtained from the stress-strain curves in a
similar way to the aluminium (10-12). This process has
been described above.

( )
s s inst core inst core ÷
÷ · = u u o c
u
,
(10)
core inst core
T
inst core
E
, ,
o c = (11)
u
c c c c
inst core
T
inst core core
gs
inst core , , ,
÷ ÷ = (12)
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
 
θ
inst

T
inst

mc
inst core,
c
gs
inst core,
c
Modified core
stress-strain curve
core
c
T
inst core,
c
Thermal
expansion (10)
u
c
inst core,
Elastic
deformation (11)
t
rest

σ
ε
Geometrical
settlement (12)

Metallurgical
creep (6)
max
,inst core
T
o
c


Fig. 6. Creep developed in the core of a gap-type conductor during the installation

V.2. Non gap-type Conductors
In the case of non gap-type conductors there is no
relative displacement between the core and the
aluminium during the installation of the conductor.
Hence, the total strains of the conductor ε
c
, the
aluminium ε
a
and the core ε
core
have the same values. For
this reason, the reference lengths of the core
core
o
L and the
aluminium
a
o
L have the same values.
It is recommended to maintain the conductors at rest
under the sagging tension T
inst
for some hours in order to
reduce the metallurgical creep during line operation. 12
hours is recommended and 48 hours is desirable.
The deformation due to geometrical settlement is
independent from the duration of the period at rest. It
depends on the sagging tension T
inst
and the installation
temperature θ
inst
. Fig. 7 shows a diagram of the algorithm
that calculates the geometrical settlement creep
developed in the core
,
gs
core inst
c and the aluminium
,
gs
a inst
c
during the installation. The deformation due to
geometrical settlement is obtained from the stress-strain
curves of the conductor. In this case, the total strains of
the aluminium ε
a
and the core ε
core
have the same values.
Fig. 8 shows the way these strains are obtained. Firstly,
the core and aluminium curves are displaced in the strain
axis to take into account the strain due to temperature
, core inst
u
c and
, a inst
u
c . Then, from the tension T
inst
and the
area of the conductor A, the conductor stress σ
inst
is
calculated and the total strain ε
c
is obtained from the
conductor stress-strain curve. This strain value is the
same as the aluminium ε
a
and the core ε
core
strains.
Besides, the stress values of the aluminium σ
a
and the
core σ
core
are calculated (13,14) from the virtual stress
values
virtual
a
o and
virtual
core
o obtained from the virtual
stress-strain curves of the aluminium and the core as it is
shown in Fig. 8. From the stress values, the tension
values are obtained (15,16).
a
virtual
inst a inst a
A
A
· =
, ,
o o
(13)
core
virtual
inst core inst core
A
A
· =
, ,
o o
(14)
a inst a inst a
A T · =
,
max
,
o (15)
core inst core inst core
A T · =
,
max
,
o (16)

While the conductor is at rest during the period t
rest
,
the deformation due to creep develops, the conductor
stress value decreases and as a consequence the creep
deformation developed afterwards decreases too. In other
words, the creep that results from a constant stress value
equal to the initial stress of the period is higher than the
creep that actually develops. To take into account this
fact, the calculation method follows the following steps.
Firstly, the initial reference length L
o,ini
is calculated
taking into account the catenary conductor length L
g

related to the installation tension T
inst
, the creep strain
due to geometrical settlement
,
gs
a inst
c calculated
previously, the strain due to tension
T
inst a,
c
and the strain
due to temperature
u
c
inst a,
(17). The metallurgical creep is
not taken into account because it has not started
developing yet.

( )
gs
inst a inst a
T
inst a
g a
ini o
core
ini o ini o
L
L L L
, , ,
, , ,
1 c c c
u
+ + +
= = =
(17)

Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
 
θ
inst

T
inst

Modified conductor
stress-strain curves
(Fig. 8)
gs
inst core,
c
core
c
T
inst core,
c
Thermal
expansion
(7,10)
u
c
inst core,
Elastic
deformation
(8,11)
Geometrical
settlement (9,12)

a
c
gs
inst a,
c
u
c
inst a,
inst a,
o
inst core,
o
Tension
(15,16)
max
,inst a
T
max
,inst core
T
T
inst a,
c
σ
ε


Fig. 7. Geometrical settlement creep developed in the core and the aluminium during the installation

 
core
σ
ε
aluminium
conductor
virtual
inst a,
o
virtual
inst core,
o
inst
o
u
c
inst a,
u
c
inst core,
a
c
core
c
c
c

Fig. 8. Modified conductor stress-strain curve

Then, to calculate the metallurgical creep developed
during the rest period t
rest
, the method proposed by
CIGRE [14] is used. This method divides the period of
time t
rest
in short sub-periods in which the change in
stress and creep strain is small enough. Sub-periods in
which the change in strain is around 20 μm/m are
considered. The initial tension of the first sub-period is
the installation tension T
inst
. At the end of each sub-
period, with the new creep values, the core and
aluminium tension values are updated by the state
calculation algorithm and these are used to evaluate the
creep in the following sub-period. This process is carried
out until the period is completed. As a result, the
metallurgical creep developed by the core
mc
inst core,
c and
the aluminium
mc
inst a,
c , and the tension value T’ at the end
of the period are obtained.
At the end of the period, the conductor catenary length
L
g
’ related to the final tension T’ is higher than the initial
conductor catenary length L
g
related to T
inst
, and for this
reason the conductor is retensioned to the initial value.
Hence, a portion of the conductor is removed. Thus, the
new reference value L
o
is lower than the initial reference
value L
o,ini
and it is given by (18) taking into account the
metallurgical creep
mc
inst a,
c
developed during this period.

( )
mc
inst a
gs
inst a inst a
T
inst a
g a
o
core
o o
L
L L L
, , , ,
1 c c c c
u
+ + + +
= = =
(18)

The conductor can be pretensioned during the
installation process causing the geometrical settlement of
the conductor and decreasing the deformation developed
during the operation. During the pretensioning period,
the conductor is under a tension T
pret
that is higher than
the installation tension T
inst
.
The calculation of the deformation due to geometrical
settlement is carried out in a similar way to the
calculation when there is no pretensioning. The only
difference is the value of the tension. Thus, the
calculation algorithm is that given in Fig. 7 but with T
pret

instead of T
inst
.
The calculation of the metallurgical creep is also
carried out in a similar way to the calculation when there
is no pretensioning. The only difference is that when the
initial reference length L
o,ini
corresponding to the
beginning of the period at rest is calculated, the creep
developed during the pretensioning period is taken into
account (19).

( )
mc
pret a
gs
pret a inst a
T
inst a
g
ini o
L
L
, , , ,
,
1 c c c c
u
+ + + +
=
(19)
VI. Creep Developed during the Operation
To calculate the creep developed during the operation,
the line lifetime is divided in operation stages where the
conductor temperature and the wind and ice loads are
assumed to be constant. The method calculates
sequentially each of the operation stages.
Fig. 9 shows the calculation of the creep in the
operation stage i. The input values are the creep strain
values
mc
i core 1 , ÷
c ,
mc
i a 1 , ÷
c
,
gs
i core 1 , ÷
c and
gs
i a 1 , ÷
c
and the historical
maximum tension values
max
1 , ÷ i core
T
and
max
1 , ÷ i a
T
at the end of
the stage (i-1). The output values are the same but at the
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
end of the stage i.
 
gs
i core 1 , ÷
c
gs
i a 1 , ÷
c
mc
i core 1 , ÷
c
mc
i a 1 , ÷
c
max
1 , ÷ i a
T
max
1 , ÷ i core
T
Operation stage i

gs
i core,
c
gs
i a,
c
mc
i core,
c
mc
i a,
c
max
,i a
T
max
,i core
T
i
u
Load
i

State
Calculation
(Fig. 2)
T
o
core,i

T
o
a,i

Metallurgical
creep
calculation
Geometrical
settlement
calculation
t
i



Fig. 9. Creep calculation in the operation stage i

In a first step, the initial tension values
,
o
a i
T and
,
o
core i
T
corresponding to the creep at the end of the stage (i-1)
and the temperature and the load of the stage i are
calculated by the state calculation algorithm. From these
tension values the geometrical settlement is calculated
first and the metallurgical creep is calculated afterwards.
The initial tension values
,
o
a i
T and
,
o
core i
T of the stage i
are compared with the historical maximum tension
values
max
1 , ÷ i core
T and
max
1 , ÷ i a
T . If they are lower, the creep
strain due to geometrical settlement and the maximum
historical tension values do not change. If the initial
tension value is higher, the creep strain due to
geometrical settlement and the maximum historical
tension values are recalculated. Fig. 10 shows the
calculation of the geometrical settlement of the
aluminium
gs
i a,
c . The total strain
i a,
c is given by (20) and
the geometrical settlement of the aluminium
gs
i a,
c is
calculated by the equation (21). The calculation for the
core is carried out in a similar way.

a
o
a
i a
L
L
=
,
c
(20)
mc
i a i a
T
i a i a
gs
i a 1 , , , , , ÷
÷ ÷ ÷ = c c c c c
u
(21)
The load due to wind or ice affects the span geometry
whereas the conductor behaviour is affected by the
conductor temperature θ
i
. The algorithm iterates the
tension value T
c
until the difference between the span
geometry length L
g
and the conductor length L
c
is below
a threshold value. The conductor is characterised from
the stress-strain curves of the core and the aluminium.
These curves have been modified as it has been
described above (Fig. 4). Besides, below the maximum
historical tension value, the behaviour of the aluminium
and the core is linear and is a function of the elastic
modulus (Fig. 11).
To make the calculation of the metallurgical creep
developed during the stage i, the period t
i
is divided in
several sub-periods where stress and temperature values
are considered to be constant. As it has been mentioned
before, sub-periods in which the change in strain is
around 20 μm/m are considered. To evaluate the creep
during the first sup-period, the core and aluminium stress
values are calculated by the state calculation algorithm
from the geometrical settlement calculated for the stage i,
the metallurgical creep at the end of the stage (i-1) and
the temperature
i
u
. At the end of each sub-period, the
core and aluminium tension values are updated. As a
result, the metallurgical creep developed by the core
mc
i core,
c and the aluminium
mc
i a,
c are obtained.
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
 
Aluminium geometrical settlement


L
c

gs
i a,
c
Span
geometry
(catenary
equation)
i
u
Load
i

L
g

Iterate on T
c

until L
g
= L
c

Conductor
stress-strain
curve at stage
(i-1)
T
c

T
a

max
,i a
T
Deformation
calculation
(20)
Elastic
deformation
L
a

Thermal
expansión
Geometrical
settlement (21)

i a,
c
T
i a,
c
u
c
i a,
mc
i a 1 , ÷
c


Fig. 10. Aluminium geometrical settlement calculation in the operation stage i

 
aluminium
max
,i a
o
max
1 , ÷ i a
o
o
c
u
c
i a,
mc
i a 1 , ÷
c
gs
i a 1 , ÷
c
T
i a,
c
i a,
c
gs
i a,
c
stage (i-1)
stage i


Fig. 11. Stress-strain curve of the aluminium at the end of the stages
(i-1) and i
VII. Application Example
The described method is applied in an application
example. The span length is 350 m and the conductor is
the ZTACIR Hen. The installation tension is 1681 kg (15
% RTS) and it has been carried out at 15 ºC.
The maximum tension conditions evaluated are those
established in the Spanish regulation. The maximum
tension condition in Spanish lines considers ice load at
-20 ºC. Besides, a high temperature operation of the line
is expected. In order to model the effect of different
operation temperatures, the conductor temperature is
assumed to be 15 ºC for 6 months, 30 ºC for 3 months,
60 ºC for 2 months and 120 ºC for one month every year.
VII.1. Creep Developed during the Installation
The ZTACIR conductor is a non gap-type conductor
and for this reason the algorithm described in sub-section
V.2 is applied. Thus, the geometrical settlement is
calculated with the algorithm given in Fig. 7 The inputs
are the installation temperature θ
inst
(15 ºC) and tension
T
inst
(1681 kg). The obtained results are the geometrical
settlement of the core
,
gs
core inst
c (6.54·10
-6
) and the
aluminium
,
gs
a inst
c (1.7·10
-4
), and the historical maximum
tension of the core
max
,inst core
T (812 kg) and the aluminium
max
, a inst
T (869 kg). For the metallurgical creep calculation,
the period the conductor is at rest t
rest
is assumed to be
one hour. The obtained results are the metallurgical creep
developed by the core
mc
inst core,
c (3.55·10
-6
) and the
aluminium
mc
inst a,
c (8.5·10
-6
). As the ZTACIR is a non
gap-type conductor, the reference lengths
core
o
L and
a
o
L
given by the equation (18) have the same value (350.65
m). They are obtained from the catenary length of the
installed conductor L
g
(350.89 m).
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
VII.2. Creep Developed during the Operation
The creep developed for 10 years is calculated. The
first step is the definition of the operation stages. For this
purpose, in addition to the duration of the stage, the
conductor temperature and the wind and ice loads have
been defined. Each year is divided in 4 stages of
different length (6, 3, 2 and 1 month) where different
conductor temperatures are assumed (15 ºC, 30 ºC, 60
ºC, 120 ºC). Besides, after 5 years in operation an ice
load condition is assumed at -20 ºC. Hence, 41 stages are
calculated. The algorithm described in Fig. 9 is applied
to each of the stages.
Fig. 12 shows the evolution of the creep during the 10
year period. The first 5 years there is an increase of the
metallurgical creep that decreases the tension value. As a
consequence, the tension values are below the historical
tension values and no geometrical settlement is
developed. When the ice load occurs new historical
tension values are obtained for the core
max
core
T
(1494 kg)
and the aluminium
max
, a inst
T
(2109 kg). Hence, geometrical
settlement is developed and it mainly affects the
aluminium. During the last 5 years the metallurgical
creep develops but much slower than at the beginning.

0,E+00
5,E‐05
1,E‐04
2,E‐04
2,E‐04
3,E‐04
0 1 2 3 4 5 6 7 8 9 10
M
e
t
a
l
l
u
r
g
i
c
a
l
 
c
r
e
e
p
 
s
t
r
a
i
n
Time, year
Core Aluminium

0,E+00
1,E‐04
2,E‐04
3,E‐04
4,E‐04
5,E‐04
0 1 2 3 4 5 6 7 8 9 10
G
e
o
m
e
t
r
i
c
a
l
 
s
e
t
t
l
e
m
e
n
t
 s
t
r
a
i
n
Time, year
Core Aluminium

Fig. 12. Creep developed during the operation
VIII. Conclusion
A complete description of a sag-tension calculation
algorithm developed by the authors has been presented
including a calculation example. The relation between
the creep development and the factors that influence it
such as the installation process and the operation
conditions during the line lifetime is described step by
step. The algorithm is characterised by the creep
sequential calculation. Thus, the creep developed in
previous stages influences the creep developed in
subsequent stages. Two periods are differenced in the
creep development: the installation period and the
operation period.
The method is suitable for modelling the conductor
behaviour including the multiple stages during the line
lifetime. Besides, it allows a detailed modelling of the
installation process. The described algorithm takes into
account the interaction between the metallurgical creep
and the geometrical settlement. Thus, the method
calculates the installation tension for new lines taking
into account the expected conditions during the line
lifetime. Furthermore, the method is also useful for the
calculation of the current state of lines in operation
whose historical operation conditions are known.
The method has several advantages over other
methods proposed in literature. Some advantages of the
developed method over the graphical method are related
to the creep stages (several stages are calculated
sequentially and there is interaction between
metallurgical creep and creep due to wind or ice loads),
the gap-type conductor installation (the aluminium creep
is modelled and the steel is assumed to be at rest during a
configurable duration), the high temperature
metallurgical creep (independent core and aluminium
creep calculation and coefficients as a function of the
conductor type) and the pretensioning during the
installation (it is included in the calculation method).
Some advantages over the strain summation method are
the independent core and aluminium reference lengths
for gap-type conductors and the detailed calculation of
the creep developed during the installation.
Acknowledgements
This work is financially supported by the University
of the Basque Country UPV/EHU under project
EHU09/18, the Ministerio of Ciencia e Innovación under
project DPI2009-08454 and the Eusko Jaurlaritzako
Hezkuntza, Unibertsitate eta Ikerketa Saila (Euskal
unibertsitate-sistemako ikerketa-taldeak Ref. IT532-10).
References
[1] CIGRÉ B2-12 Brochure (Ref. No. 324), Sag-tension calculation
methods for overhead lines, 2007.
[2] H. Mokhlis, H.Y. Li, A.R. Khalid, The application of voltage sags
pattern to locate a faulted section in distribution network,
International Review of Electrical Engineering, vol. 5, n. 1, 2010,
pp. 173-179.
[3] H. Mokhlis, A.H.A. Bakar, D.N.A. Talib, H. Mohamad, The
improvement of voltage sags pattern approach to locate a fault in
distribution network, International Review of Electrical
Engineering, vol. 5, n. 3, 2010, pp. 1159-1164.
[4] H. Mokhlis, H.Y. Li, H. Mohamad, A.H.A. Bakar, A
comprehensive fault location estimation using voltage sag profile
for non-homogenous distribution networks, International Review
of Electrical Engineering, vol. 5, n. 5, 2010, pp. 2310-2316.
Copyright © 2011, Praise Worthy Prize S.r.l..
Reprinted, with permission of Praise Worthy Prize S.r.l.. from the International Review of Electrical Engineering, IREE, Vol. 6. n. 3
[5] N. Kang, Y. Liao, New fault location technique for series
compensated transmission lines, International Review of
Electrical Engineering, vol. 4, n. 6, 2009, pp. 1385-1390.
[6] T. Varney, Graphic method for sag-tension calculations for ACSR
and other conductors, Alcoa, 1926.
[7] J.S. Barrett, S. Dutta, O. Nigol, A new computer model of ACSR
conductors, IEEE Trans. Power App. Syst., vol. PAS-102 n. 3,
1983, pp. 614-621.
[8] I. Albizu, A.J. Mazon, I. Zamora, Flexible strain-tension
calculation method for gap-type overhead conductors, IEEE
Trans. on Power Del., vol. 24 n. 3, 2009, pp. 1529-1537.
[9] CIGRÉ B2-12 Brochure (Ref. No. 244), Conductors for the
uprating of overhead lines, 2004.
[10] I. Albizu, A.J. Mazon, V. Valverde, G. Buigues, Aspects to take
into account in the application of mechanical calculation to high
temperature low sag conductors, IET Gener. Transm. Dis., vol. 4
n. 5, 2010, pp. 631-640.
[11] Technical Specifications for the Installation of Gap Type
Conductors. SAPREM® Method, SAPREM, 2006.
[12] M. Landeira, P. Morentin, A.J. Mazon, I. Albizu, The high
temperature cable solution for electrical overhead distribution
lines, DYNA, vol. 82 n. 5, 2007, pp. 226-230.
[13] CIGRÉ 22-05, A practical method of conductor creep
determination, ELECTRA, vol. 24, 1972, pp. 105-137.
[14] CIGRÉ 22-05, Permanent elongation of conductors. Predictor
equations and evaluation methods, ELECTRA, vol. 75, 1981, pp.
63-98.
[15] O. Nigol, J.S. Barrett, Characteristics of ACSR conductors at high
temperatures and stresses, IEEE Trans. Power App. Syst., vol.
PAS-100 n. 2, 1981, pp. 485-493.
[16] A method of stress-strain testing of aluminum conductor and a test
for determining the long time tensile creep of aluminum
conductors in overhead lines, Electrical Technical Committee of
The Aluminum Association, 1999
Authors’ information
1
Electrical Engineering Department
Faculty of Engineering of Bilbao
UPV/EHU University of the Basque Country
Alda. Urquijo s/n
48013-Bilbao
Spain
e-mails igor.albizu@ehu.es
javier.mazon@ehu.es
elvira.fernandezh@ehu.es

I. Albizu was born in Zumaia, Spain, in 1975.
He received the M.Sc. degree in electronic
instrumentations systems from the University
of Manchester Institute of Science and
Technology, Manchester, U.K., in 1999, and
the Ph.D. degree from the University of the
Basque Country, Bilbao, Spain, in 2008.
Currently, he is a Lecturer with the Electrical
Engineering Department, Faculty of Engineering of Bilbao, University
of the Basque Country. His research activities are concentrated in the
area of transmission-line thermal rating..

A.J. Mazon was born in Bilbao, Spain, in
1965. He received the Ph.D. degree from the
University of the Basque Country, Bilbao,
Spain, in 1994. In 1992, he was with Labein
Research Laboratories.
Currently, he is a Full-Time Professor in the
Electrical Engineering Department, Faculty of
Engineering of Bilbao. His research activities
are concentrated in the area of electric power systems, transients
simulation, fault analysis, and transmission-line thermal rating.


E. Fernandez was born in Bilbao, Spain, in
1973. She received her Ph.D. degree from the
University of the Basque Country, Bilbao,
Spain, in 2008.
Currently, she is a Lecturer with the Electrical
Engineering Department, Faculty of
Engineering of Bilbao, University of the
Basque Country. Her research activities are
concentrated in the area of transmission-line thermal rating and low-
voltage arc chutes studies.

  INSTALLATION Installation mc  core. This minimum value is zero or a negative value if aluminium compression is considered [15]. 2. 10 year operation.1  La mc a Tamax .i gs and  a. etc. This calculation process is carried out for the aluminium and the core separately. Tamax . with permission of Praise Worthy Prize S. The strain due to geometrical settlement is assumed to be independent of time. In each stage.r. .n gs  a. III. 3 line lifetime. The core tension Tcore is iterated until the difference between the span geometry length Lg and the conductor length Lc is below a threshold value. 1.i ..l. and creep  core . 2.Copyright © 2011. The creep developed during each operation stage is calculated taking into account the creep developed so far. IV.inst gs  a. Vol. from the installation tension Tinst.  core .1 mc  a. Two periods are differenced in the creep development: the installation period and the operation period (Fig. It is only dependent on the conductor construction and the historical maximum tension Tmax experienced.  a .inst max Tcore .inst Tamax . State Calculation Algorithm The state calculation algorithm is shown in Fig. The installation tension value Tinst is iterated until one of the maximum tension conditions does not allow increasing its value.1 gs  core . etc.). Thus. stage 1.n gs  core. The aluminium tension Ta cannot go below its minimum value.   mc mc gs gs temperature  core . the parameters that characterise each stage i are the following:  Conductor temperature θi  Load conditions (ice and wind)  Duration ti From the creep strain values calculated for the operation stages.14].1 max Tcore .n max Tcore . The creep calculation during the installation is fully described as a function of the type of conductor (gap-type or non gap-type) and taking into account the time the conductor is at rest and whether there is pretensioning or not. The Load Tc Span geometry Lg + + Fig. 6. the tension values related to the maximum tension conditions are calculated by the state calculation algorithm described in section IV. the metallurgical creep εmc and the creep due to geometrical settlement εgs are calculated separately.n Ta Fig. Sag-Tension Calculation Algorithm The creep developed in previous stages influences the creep developed in subsequent stages [13. the algorithm makes a sequential calculation of the creep. 1).i . Creep and maximum tension evolution in time  ags When the method characterises the conductor installation it differentiates between the gap-type conductors and the rest of conductors. mc mc gs the wind and ice load and the creep  core.  a . For this reason. This is taken into account in the algorithm when aluminium tension Ta is evaluated.n Core (2. The conditions are related to specific stages (after the installation.inst maximum tension conditions are characterised by the conductor temperature and the ice and wind load values.  a. The metallurgical creep is calculated as a function of the conductor tension T.  a . the algorithm calculates the conductor tension in the defined maximum tension conditions.i of the corresponding stage (installation. State calculation algorithm The state calculation algorithm is based in the dependence of the core and aluminium lengths Lcore and T T La on the strain values due to tension  core .inst mc  a. stage 2. the conductor temperature θ and the duration t of the stage.. n.1 Operation stage n  mc core.4) Lcore Lc Iterate on Tcore until Lc = Lg Aluminium (1. The conductor temperature and the wind and ice loads are assumed to be constant during each operation stage. Reprinted.  a . The state calculation algorithm determines the conductor tension value calculating separately the core tension Tcore and the aluminium tension Ta from the conductor temperature θi. The tension limit values are defined for these conditions. from the International Review of Electrical Engineering.). n mc  core gs  core mc  a . Praise Worthy Prize S.r. State calculation OPERATION Operation stage 1 mc  core . IREE. The calculation of creep developed during the operation is also described. The creep developed during the operation depends on the creep previously developed during the installation.inst  ags1 . Thus.3) Tcore gs  core.  core.l.

.inst the installation temperature θinst and the duration tpre-sagging of the presagging step.inst mc metallurgical creep  a ..inst(%). the total strains of the aluminium εa and the core εcore have different values. the installation sagging tension Tinst.inst T core . For this purpose.. The span geometry is characterised by the catenary equation. Praise Worthy Prize S. the metallurgical creep at the stress-strain test temperature θs-s and different stress values σ is calculated by (6). They are obtained from the installation condition. with permission of Praise Worthy Prize S. From the installation sagging tension Tinst and the percentage of the sagging tension Ta. the core and aluminium reference lengths Lcore and La correspond to the reference o o condition with no tension and no creep. IREE. the aluminium tension Ta.   (3) (4) (5) 1   L g . Reprinted. Fig. 3 shows a diagram of the algorithm that calculates the creep developed by the aluminium during the installation of the gap-type conductors. the aluminium stress σa.inst (%) the aluminium is is obtained from the modified stress-strain curves of the aluminium.inst   a . The outputs of the algorithm are the maximum tension Tamax experienced and the deformation due to . inst mc gs   core. This deformation  a .inst is a function of the installation temperature θinst.4).inst  2  Tinst  s c  sinh  2 T c inst      V. this is modified removing the strain related to the metallurgical creep.inst   core. The total strain εa corresponding to the tension Ta at the installation temperature θinst is obtained from the modified aluminium stress-strain curve. between 2 and 24 hours depending on the span. As described in [8]. 6. To obtain directly the value of the geometrical settlement from the curve. the percentage of the sagging tension Ta.inst   core. 5). the whole sagging tension is applied only to the steel for a few hours.inst is obtained from the aluminium stress σa. Due to this special installation process. β and μ are constant coefficients that represent the behaviour of the aluminium [14]. The modified stress-strain curve corresponding to the stress-strain test carried out at the temperature θs-s is displaced in the  strain axis in order to model the deformation  a . For a few minutes. The weight value ω depends on the conductor weight ωc and the wind and ice load. the geometrical settlement and the metallurgical creep developed during one hour. from the International Review of Electrical Engineering.  mc  K  e      t  (6) La  o Lcore  o 1   L g . o T  mc La  La   a   a   a   ags o Lcore  L core o     gs core T core   core    mc core   (1) (2) expected to support during the pre-sagging step (usually 70 %) and the duration tpre-sagging of the pre-sagging step. The strain due to tension  a. Vol.l. Thus. where the temperature and the tension values are known and the creep strain is estimated from the installation process (3. T 5).inst and the  strain due to thermal expansion  a . The catenary length Lg. The strain in the stress-strain curve [16] is composed of three components related to the stress. Creep Developed during the Installation V.inst is calculated. The deformation due to geometrical settlement  agsinst . In this way. 4.inst and the elastic modulus of the aluminium Ea (8). a modified curve is obtained as it is shown in Fig. 3 and the core and aluminium reference lengths Lcore and o La (1.inst and the area of the aluminium Aa.1.r. during the pre-sagging step. From the tension Ta. mc The metallurgical creep  a .inst is obtained from the installation tension Tinst and the conductor weight ωc (5).inst due  to thermal expansion (Fig. the installation process of the gap-type conductors is special and comprises several steps. n. The deformation due to geometrical settlement  agsinst is then calculated subtracting from the . where K.inst (9).inst mc   a .inst   agsinst .r. The input values of the algorithm are the installation temperature θinst. Thus.inst is calculated and the strain εa is obtained from the aluminium stress-strain curve (Fig.inst   L g . T total strain εa the strain due to tension  a.Copyright © 2011. Φ. The metallurgical creep in the aluminium follows the law given in (6). the temperature θs-s of the stress-strain test and the coefficient of thermal expansion αa of the aluminium (7). In the final installation step. The catenary equation is a function of the conductor tension T and the weight ω.inst is calculated as a function of the aluminium tension Ta. around 70 % of the sagging tension is applied only to the aluminium. the aluminium gets slack during the installation and the knee-point temperature is forced to be the temperature of installation.l. .2).inst T a . Gap-type Conductors The gap-type conductors allow a relative displacement between the core and the aluminium during the installation of the conductor.inst and geometrical settlement  agsinst .

inst   a   inst   s  s  T  a .inst   core   inst   s  s  Fig. The outputs max of the algorithm are the maximum tension Tcore.inst Ecore  T  agsinst   a .inst   core.inst   a . n..inst and geometrical settlement  core .inst (%) Aluminium pre-sagging tension Thermal expansion (7)  a. Reprinted. β and μ are constant coefficients that represent the behaviour of the core steel.inst .  gs T  core.inst  Geometrical settlement (9) Modified aluminium stress-strain curve σ (Fig.inst   core   core. This process has been described above.inst   core.inst is obtained from the stress-strain curves in a similar way to the aluminium (10-12). IREE. (7) (8) (9)   core.inst Fig. These coefficients are different from those previously given for the aluminium. with permission of Praise Worthy Prize S. 6 shows a diagram of the algorithm that . 6.l.inst ε  Elastic deformation (8) T  a. from the International Review of Electrical Engineering.r.. where K. Creep developed in the aluminium of a gap-type conductor during the installation   σ modified original calculates the creep developed by the core during the installation of the gap-type conductors. the installation sagging tension Tinst and the time the core is at rest under installation sagging tension trest.inst ε function of the core tension Tinst. The input values of the algorithm are the installation temperature θinst.inst   a . The deformation due to geometrical settlement gs  core.l. the installation temperature θinst and the time the core is at rest trest.inst εa.inst Tamax . 5) gs  a.inst Ta.r.inst is calculated as a   σ σa. 3   θinst Tinst Ta. 4. 5.inst   a . Φ.inst Metallurgical creep (6) mc  a.inst E a (10) (11) (12)  T core. 3.inst  a.inst ε Fig. Modified (1 h metallurgical creep removed) stress-strain curve experienced and the deformation due to metallurgical mc gs creep  core . Modified (1 h metallurgical creep removed) aluminium stress-strain curve   a . mc The metallurgical creep  core .inst ε θ a.inst .inst tpre-sagging Fig.Copyright © 2011. The metallurgical creep in the core follows the law given in (6). Praise Worthy Prize S. Vol.

IREE. the creep strain due to geometrical settlement  agsinst calculated . It depends on the sagging tension Tinst and the installation temperature θinst.ini  o .  (17) . 3   θinst Tinst Thermal expansion (10)   core. the total strains of the conductor εc. Firstly. Praise Worthy Prize S.inst  Acore While the conductor is at rest during the period trest. 8 shows the way these strains are obtained. the aluminium εa and the core εcore have the same values. 6.inst and the strain  due to temperature  a. with permission of Praise Worthy Prize S. from the International Review of Electrical Engineering.inst  Aa .inst and the aluminium  a .r.inst   core.16).Copyright © 2011..ini is calculated taking into account the catenary conductor length Lg related to the installation tension Tinst.r. the initial reference length Lo.inst during the installation. T previously.inst   agsinst .2.inst .inst   a . the deformation due to creep develops. Then. Non gap-type Conductors In the case of non gap-type conductors there is no relative displacement between the core and the aluminium during the installation of the conductor. the core and aluminium curves are displaced in the strain axis to take into account the strain due to temperature    core. the conductor stress σinst is calculated and the total strain εc is obtained from the conductor stress-strain curve. from the tension Tinst and the area of the conductor A.inst  Tamax   a . the creep that results from a constant stress value equal to the initial stress of the period is higher than the creep that actually develops. the conductor stress value decreases and as a consequence the creep deformation developed afterwards decreases too.inst  A Aa A Acore (13) (14) (15) (16) virtual  core. the total strains of the aluminium εa and the core εcore have the same values. In this case. Creep developed in the core of a gap-type conductor during the installation virtual virtual values  a and  core obtained from the virtual stress-strain curves of the aluminium and the core as it is shown in Fig.inst Geometrical settlement (12) σ Modified core stress-strain curve gs  core. the strain due to tension  a. From the stress values. In other words. This strain value is the same as the aluminium εa and the core εcore strains. To take into account this fact.ini  Lcore  La . Reprinted. Firstly.inst T max core . The metallurgical creep is not taken into account because it has not started developing yet. the stress values of the aluminium σa and the core σcore are calculated (13.inst (17). 8.inst Metallurgical creep (6) mc  core. Besides. the tension values are obtained (15. The deformation due to geometrical settlement is obtained from the stress-strain curves of the conductor.inst and  a .14) from the virtual stress virtual  a .ini o 1   Lg T a .inst   core. the calculation method follows the following steps. The deformation due to geometrical settlement is independent from the duration of the period at rest. n. the reference lengths of the core Lcore and the o aluminium La have the same values.l.l.inst    a . 7 shows a diagram of the algorithm that calculates the geometrical settlement creep gs gs developed in the core  core . 12 hours is recommended and 48 hours is desirable. Vol.inst max Tcore . For this reason. Lo .inst  core ε  Elastic deformation (11) T  core.inst trest Fig. Fig.. o It is recommended to maintain the conductors at rest under the sagging tension Tinst for some hours in order to reduce the metallurgical creep during line operation. Fig. Hence. 6. V.

9 shows the calculation of the creep in the operation stage i. 7.inst   a .inst max Tcore .inst ε Elastic deformation (8. Fig. The initial tension of the first sub-period is the installation tension Tinst.r. the core and aluminium tension values are updated by the state calculation algorithm and these are used to evaluate the creep in the following sub-period. the new reference value Lo is lower than the initial reference value Lo. with the new creep values. 8. The only difference is that when the initial reference length Lo. Lo.inst and mc the aluminium  a. During the pretensioning period. 8) σ  core gs  core.inst mc   a . a portion of the conductor is removed. Modified conductor stress-strain curve Then. 3   Geometrical settlement (9. This method divides the period of time trest in short sub-periods in which the change in stress and creep strain is small enough. The calculation of the deformation due to geometrical settlement is carried out in a similar way to the calculation when there is no pretensioning. the calculation algorithm is that given in Fig. Sub-periods in which the change in strain is around 20 μm/m are considered. Thus. Reprinted.inst   agspret   a .ini and it is given by (18) taking into account the mc metallurgical creep  a. n. The output values are the same but at the . max maximum tension values Tcore .inst a  core c ε Fig.  a .inst a  a . The input values are the creep strain mc gs mc values  core .ini  1   Lg T a . The only difference is the value of the tension.inst developed during this period. Tension (15.inst .l.. the method proposed by CIGRE [14] is used.i 1 .i 1 the stage (i-1).i 1 and  agsi 1 and the historical . Praise Worthy Prize S. 6. from the International Review of Electrical Engineering.i 1 and Tamax at the end of .Copyright © 2011. Geometrical settlement creep developed in the core and the aluminium during the installation   σ  inst virtual core. Creep Developed during the Operation To calculate the creep developed during the operation.11) T  a.inst .r. At the end of each subperiod. VI.  core .   (19) of the period are obtained. As a result.inst   agsinst   a .inst Lo  Lcore  La  o o 1   Lg T a . the creep developed during the pretensioning period is taken into account (19).16) Tamax . with permission of Praise Worthy Prize S. Hence. 7 but with Tpret instead of Tinst. the conductor is under a tension Tpret that is higher than the installation tension Tinst.   (18)  virtual  a. Vol. the conductor catenary length Lg’ related to the final tension T’ is higher than the initial conductor catenary length Lg related to Tinst.i 1 . IREE. The method calculates sequentially each of the operation stages. pret .inst Fig. and for this reason the conductor is retensioned to the initial value.10)   core . At the end of the period. the mc metallurgical creep developed by the core  core. to calculate the metallurgical creep developed during the rest period trest.inst  agsinst . and the tension value T’ at the end The conductor can be pretensioned during the installation process causing the geometrical settlement of the conductor and decreasing the deformation developed during the operation..inst conductor core aluminium   core.l. Thus.inst   a . This process is carried out until the period is completed. the line lifetime is divided in operation stages where the conductor temperature and the wind and ice loads are assumed to be constant.inst  core.inst θinst Tinst   Modified conductor stress-strain curves (Fig. The calculation of the metallurgical creep is also carried out in a similar way to the calculation when there is no pretensioning.inst T  core .inst mc   a .ini corresponding to the beginning of the period at rest is calculated.12) Thermal expansion (7.

the behaviour of the aluminium and the core is linear and is a function of the elastic modulus (Fig. o The initial tension values Tao. from the International Review of Electrical Engineering. The calculation for the core is carried out in a similar way.i 1 .i 1 and Ta . i Loadi gs  core. The conductor is characterised from the stress-strain curves of the core and the aluminium. 6. 4). If the initial tension value is higher.i Fig.i is given by (20) and gs the geometrical settlement of the aluminium  a. Vol.i corresponding to the creep at the end of the stage (i-1) and the temperature and the load of the stage i are calculated by the state calculation algorithm. If they are lower.i of the stage i are compared with the historical maximum tension values Tcore . the creep strain due to geometrical settlement and the maximum historical tension values are recalculated. The load due to wind or ice affects the span geometry . the creep strain due to geometrical settlement and the maximum historical tension values do not change.i 1 State Calculation (Fig.i . From these tension values the geometrical settlement is calculated first and the metallurgical creep is calculated afterwards.i ti mc  a. the metallurgical creep at the end of the stage (i-1) and the temperature  i .i and Tcore.Copyright © 2011. sub-periods in which the change in strain is around 20 μm/m are considered. 9.i and the aluminium  a. with permission of Praise Worthy Prize S.i gs  a. whereas the conductor behaviour is affected by the conductor temperature θi. the core and aluminium stress values are calculated by the state calculation algorithm from the geometrical settlement calculated for the stage i.i 1  mc a . the metallurgical creep developed by the core mc mc  core. 11). As a result. The algorithm iterates the tension value Tc until the difference between the span geometry length Lg and the conductor length Lc is below a threshold value.r. below the maximum historical tension value.i 1 Tamax .i mc  core . the core and aluminium tension values are updated.l. Creep calculation in the operation stage i o In a first step.i Tamax . IREE.i is max max calculated by the equation (21). At the end of each sub-period. Praise Worthy Prize S.i Tocore. the period ti is divided in several sub-periods where stress and temperature values are considered to be constant. Fig.i gs  core .i 1 max Tcore . To evaluate the creep during the first sup-period.  a .i  La La o (20) (21) T  mc  agsi   a .i max Tcore . 3 end of the stage i. As it has been mentioned before.i 1  agsi 1 .i   a .i 1 . 10 shows the calculation of the geometrical settlement of the gs aluminium  a.. Reprinted.   Operation stage i Metallurgical creep calculation mc  core.i and Tcore. These curves have been modified as it has been described above (Fig..i   a .l. n. 2) Geometrical settlement calculation Toa.i are obtained.r. The total strain  a. the initial tension values Tao.i   a . Besides. To make the calculation of the metallurgical creep developed during the stage i.

inst (8. n.7·10-4).1. a high temperature operation of the line is expected. The maximum tension condition in Spanish lines considers ice load at -20 ºC. .i La Deformation calculation (20) Elastic deformation Thermal expansión  a .inst (6.i  a . Application Example The described method is applied in an application example. . Stress-strain curve of the aluminium at the end of the stages (i-1) and i VII.i max a .   T a.i 1 aluminium stage (i-1) stage i assumed to be 15 ºC for 6 months.i  a. Creep Developed during the Installation The ZTACIR conductor is a non gap-type conductor and for this reason the algorithm described in sub-section V..i  amc1 . 30 ºC for 3 months.Copyright © 2011. and the historical maximum . 6.i Fig. the geometrical settlement is calculated with the algorithm given in Fig.i   a .2 is applied. Reprinted. 3   Aluminium geometrical settlement Loadi Span geometry (catenary equation) Lg Iterate on Tc until Lg = Lc Tc i Conductor Lc stress-strain Ta curve at stage (i-1) Tamax .i Fig. Aluminium geometrical settlement calculation in the operation stage i      max a . from the International Review of Electrical Engineering. 7 The inputs are the installation temperature θinst (15 ºC) and tension Tinst (1681 kg). 60 ºC for 2 months and 120 ºC for one month every year. For the metallurgical creep calculation.inst (3. Vol.i1  agsi1 . max tension of the core Tcore. the reference lengths Lcore and La o o given by the equation (18) have the same value (350. Besides. the conductor temperature is Tamax (869 kg).5·10-6). They are obtained from the catenary length of the installed conductor Lg (350. The installation tension is 1681 kg (15 % RTS) and it has been carried out at 15 ºC.inst the period the conductor is at rest trest is assumed to be one hour.i  gs a. IREE.89 m).r. Praise Worthy Prize S.i Geometrical settlement (21) gs  a. The span length is 350 m and the conductor is the ZTACIR Hen.54·10-6) and the aluminium  agsinst (1. As the ZTACIR is a non gap-type conductor. The obtained results are the metallurgical creep mc developed by the core  core.l.r.i T  a .65 m). VII. 11.l.55·10-6) and the aluminium mc  a. Thus. The obtained results are the geometrical gs settlement of the core  core . 10.inst (812 kg) and the aluminium mc   a.. The maximum tension conditions evaluated are those established in the Spanish regulation. In order to model the effect of different operation temperatures. with permission of Praise Worthy Prize S.

A. 2010. The method has several advantages over other methods proposed in literature.E‐05 0. 12 shows the evolution of the creep during the 10 year period. after 5 years in operation an ice load condition is assumed at -20 ºC.inst settlement is developed and it mainly affects the aluminium.E‐04 0. 120 ºC). Besides. vol.E+00 0 1 2 3 4 5 Time. International Review of Electrical Engineering. Mokhlis. Hence.E‐04 2. 3.E‐04 1.r. IREE. International Review of Electrical Engineering. A. The first 5 years there is an increase of the metallurgical creep that decreases the tension value. Reprinted. Hence. Furthermore. The first step is the definition of the operation stages. Unibertsitate eta Ikerketa Saila (Euskal unibertsitate-sistemako ikerketa-taldeak Ref. 2 and 1 month) where different conductor temperatures are assumed (15 ºC. References [1] [2] CIGRÉ B2-12 Brochure (Ref.l. Li. H.E‐04 3.H. 3. vol. When the ice load occurs new historical max tension values are obtained for the core Tcore (1494 kg) and the aluminium Tamax (2109 kg). Talib. During the last 5 years the metallurgical creep develops but much slower than at the beginning. The described algorithm takes into account the interaction between the metallurgical creep and the geometrical settlement. H. Sag-tension calculation methods for overhead lines. A. 2010. 30 ºC.N. Metallurgical  creep strain Geometrical settlement strain 5.E‐04 5. H.A. No. As a consequence.E+00 0 1 2 3 4 5 Time.E‐04 2. D. Thus. The algorithm is characterised by the creep sequential calculation. pp. geometrical . Khalid. Praise Worthy Prize S. Vol. 5. 12. 2007. Fig. the method is also useful for the calculation of the current state of lines in operation whose historical operation conditions are known. the conductor temperature and the wind and ice loads have been defined.r. Mokhlis. the creep developed in previous stages influences the creep developed in subsequent stages. Mohamad. Some advantages of the developed method over the graphical method are related to the creep stages (several stages are calculated sequentially and there is interaction between metallurgical creep and creep due to wind or ice loads). pp. 324). IT532-10). it allows a detailed modelling of the installation process. 1159-1164. Creep developed during the operation VIII.Y. 173-179. the method calculates the installation tension for new lines taking into account the expected conditions during the line lifetime. 5. Creep Developed during the Operation The creep developed for 10 years is calculated. vol.H. the high temperature metallurgical creep (independent core and aluminium creep calculation and coefficients as a function of the conductor type) and the pretensioning during the installation (it is included in the calculation method).A. Two periods are differenced in the creep development: the installation period and the operation period. 1. from the International Review of Electrical Engineering. year Core Aluminium 6 7 8 9 10 Acknowledgements This work is financially supported by the University of the Basque Country UPV/EHU under project EHU09/18. The improvement of voltage sags pattern approach to locate a fault in distribution network. 3..E‐04 4. Bakar. Thus.R. 2010.Y. n. the Ministerio of Ciencia e Innovación under project DPI2009-08454 and the Eusko Jaurlaritzako Hezkuntza. 41 stages are calculated. 9 is applied to each of the stages. A. For this purpose. The relation between the creep development and the factors that influence it such as the installation process and the operation [3] [4] . Fig. Li. The algorithm described in Fig. A comprehensive fault location estimation using voltage sag profile for non-homogenous distribution networks. Some advantages over the strain summation method are the independent core and aluminium reference lengths for gap-type conductors and the detailed calculation of the creep developed during the installation. Mokhlis.E‐04 1. n. 2310-2316. Besides.Copyright © 2011.. the gap-type conductor installation (the aluminium creep is modelled and the steel is assumed to be at rest during a configurable duration). 3 VII. 6. Each year is divided in 4 stages of different length (6.l. Conclusion A complete description of a sag-tension calculation algorithm developed by the authors has been presented including a calculation example. H. H. 5. H. year Core Aluminium 6 7 8 9 10 conditions during the line lifetime is described step by step. The application of voltage sags pattern to locate a faulted section in distribution network. pp. 5.E‐04 2. Bakar. Mohamad. the tension values are below the historical tension values and no geometrical settlement is developed.2. n. International Review of Electrical Engineering. 60 ºC. with permission of Praise Worthy Prize S. H. The method is suitable for modelling the conductor behaviour including the multiple stages during the line lifetime. n. in addition to the duration of the stage.

A. Y. 5. Currently. 2007.D. Aspects to take into account in the application of mechanical calculation to high temperature low sag conductors. 614-621. Faculty of Engineering of Bilbao. vol. Manchester. M. on Power Del.J. 1983. He received the Ph. 4 n. 2004. P. pp. Vol. Permanent elongation of conductors. Predictor equations and evaluation methods. Spain. Dis. He received the M. Morentin. Liao. vol. Currently. 3. New fault location technique for series compensated transmission lines. 105-137. degree from the University of the Basque Country. 4. T. 2006. Nigol. I.l. he is a Full-Time Professor in the Electrical Engineering Department. and transmission-line thermal rating. Reprinted. 82 n. Currently. Power App. SAPREM® Method. she is a Lecturer with the Electrical Engineering Department.r. pp. University of the Basque Country.J.S. J. transients simulation.mazon@ehu. Albizu. IEEE Trans. ELECTRA.albizu@ehu. Faculty of Engineering of Bilbao. IEEE Trans. Electrical Technical Committee of The Aluminum Association. IREE. A. O. vol. Graphic method for sag-tension calculations for ACSR and other conductors. No. Nigol. A. from the International Review of Electrical Engineering. Dutta. Syst. 1926. in 2008. pp. vol. 3. Power App. Spain. 24 n. fault analysis. Barrett. Bilbao. PAS-100 n. Mazon. Kang. Buigues. Mazon was born in Bilbao. 2009. A new computer model of ACSR conductors. I. Praise Worthy Prize S. 6. pp. vol. Albizu. S.es elvira. Spain. Albizu.r. Spain.Copyright © 2011. E. pp.. The high temperature cable solution for electrical overhead distribution lines. and the Ph.fernandezh@ehu. 6. Spain. degree from the University of the Basque Country. I. ELECTRA. Zamora. Albizu was born in Zumaia. O.. 485-493.J.es javier. Fernandez was born in Bilbao.D..l. A method of stress-strain testing of aluminum conductor and a test for determining the long time tensile creep of aluminum conductors in overhead lines. A practical method of conductor creep determination. I. Valverde. A.. 3 [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] N. in 1975..Sc. 24. Barrett. Authors’ information Electrical Engineering Department Faculty of Engineering of Bilbao UPV/EHU University of the Basque Country Alda. G. he was with Labein Research Laboratories. University of the Basque Country. She received her Ph. IEEE Trans.S. Urquijo s/n 48013-Bilbao Spain e-mails igor. 63-98. degree in electronic instrumentations systems from the University of Manchester Institute of Science and Technology.. in 1999. 226-230. Mazon. in 2008. Landeira.. Flexible strain-tension calculation method for gap-type overhead conductors. pp. n. Spain. pp. Bilbao. 1385-1390. he is a Lecturer with the Electrical Engineering Department. CIGRÉ 22-05. DYNA. 244).K. His research activities are concentrated in the area of electric power systems. 1999 I. 2010. SAPREM. Conductors for the uprating of overhead lines. n. with permission of Praise Worthy Prize S.. PAS-102 n. vol. vol. degree from the University of the Basque Country. U. CIGRÉ B2-12 Brochure (Ref. In 1992. Bilbao. 631-640. IET Gener. His research activities are concentrated in the area of transmission-line thermal rating. 1981. CIGRÉ 22-05. Syst. 1972.es 1 . 2009. Her research activities are concentrated in the area of transmission-line thermal rating and lowvoltage arc chutes studies. 1981. Mazon. 1529-1537. in 1994. 2. pp. J.J. Faculty of Engineering of Bilbao. 5. V. Alcoa. Characteristics of ACSR conductors at high temperatures and stresses. 75. Technical Specifications for the Installation of Gap Type Conductors. Transm.D. International Review of Electrical Engineering. in 1965. in 1973. vol. Varney.