You are on page 1of 43

Task Force

B2.11.04



June 2005


OVERHEAD CONDUCTOR SAFE
DESI GN TENSI ON
WI TH RESPECT TO AEOLI AN
VI BRATI ONS
273
OVERHEAD CONDUCTOR SAFE
DESIGN TENSION
WITH RESPECT TO AEOLIAN
VIBRATIONS

TASK FORCE
B2.11.04


Members:
Claude Hardy (Convenor)
Hans-Jörg Krispin (Secretary)
André Leblond, Charles Rawlins, Konstantin Papailiou, Louis
Cloutier, Peter Dulhunty
Corresponding Members:
David Havard, Jean-Marie Asselin, Magnar Ervik, Tapani
Seppä, Vladimir Shkatsov




Copyright © 2005
“Ownership of a CIGRE publication, whether in paper form or on electronic support only infers right of use
for personal purposes. Are prohibited, except if explicitly agreed by CIGRE, total or partial reproduction of
the publication for use other than personal and transfer to a third party; hence circulation on any intranet or
other company network is forbidden”.
Disclaimer notice
“CIGRE gives no warranty or assurance about the contents of this publication, nor does it accept any
responsibility, as to the accuracy or exhaustiveness of the information. All implied warranties and
conditions are excluded to the maximum extent permitted by law”.

TABLE OF CONTENTS................................................................................................................ 2
EXECUTIVE SUMMARY ............................................................................................................ 3
1. INTRODUCTION.................................................................................................................. 5
2. REVIEW OF THE EVERY DAY STRESS (EDS) CONCEPT ............................................ 6
3. TURBULENCE AS A FUNCTION OF TERRAIN .............................................................. 8
4. SINGLE UNPROTECTED CONDUCTORS ...................................................................... 10
4.1 Modelling...................................................................................................................... 10
4.2 Comparison With Field Experience.............................................................................. 14
4.3 General Discussion ....................................................................................................... 16
4.4 Recommendations......................................................................................................... 18
5. DAMPED SINGLE CONDUCTORS .................................................................................. 20
5.1 Modelling...................................................................................................................... 20
5.2 Selection of Parameters................................................................................................. 21
5.3 Predicted Safe Design Tension..................................................................................... 22
5.4 Comparaison With Field Experience............................................................................ 23
5.5 Task Force Recommendations...................................................................................... 26
5.6 Limitations .................................................................................................................... 28
6. BUNDLED CONDUCTORS ............................................................................................... 29
6.1 Review of Literature ..................................................................................................... 29
6.2 Review of Field Experience.......................................................................................... 30
6.3 Determination of Safe Design Tension......................................................................... 32
6.3.1 Unspacered bundled conductors fitted or not with span-end Stockbridge dampers... 32
6.3.2 Bundled conductors fitted with non-damping spacers................................................ 32
6.3.3 Bundled conductors fitted with non-damping spacers and span-end Stockbridge
dampers................................................................................................................................. 34
6.3.4 Bundled conductors fitted with damping spacers ....................................................... 34
6.4 Comparison With Field Experience.............................................................................. 36
6.5 Task Force Recommendations...................................................................................... 38
6.6 Limitations and Warnings............................................................................................. 39
7. REFERENCES ..................................................................................................................... 40
2
EXECUTIVE SUMMARY

There may be several motives for controlling overhead conductor tension at the design stage.
One of the most obvious reasons is to ensure that maximum tension resulting from the assumed
most severe climatic loading does not exceed the conductor allowable tension. At the opposite
end, it may be required to limit minimum tension while the conductor is operating at maximum
temperature so that line clearance is not violated. A third motive, which should not be
disregarded, is to restrict conductor susceptibility to harmful conductor vibrations.

As a matter of fact, it is well known that stranded conductors get more vulnerable to Aeolian
vibrations as tension is increased. This is true for all conductor systems whether they are used in
solo or in bundles and whether they are fitted or not with damping and/or spacing devices.
Therefore, there is a need to set an upper limit to conductor unloaded tension that may prevail for
a significant period of time.

This brochure deals with this important issue. It starts in Section 2 with a critical examination of
the EDS concept which was put forward in 1960 by CIGRÉ SC6 with the intent to provide
guidance on such conductor safe design tensions with respect to Aeolian vibrations. It is noted,
for example, that the 18% EDS value which was proposed as safe for ACSR conductors did lead,
in spite of all, to fatigue failures in a significant number of cases, thus motivating the present
work further. In this conjuncture, one of the first and most important tasks of the Task Force was
to question the relevance of the EDS (% of the Ultimate Tensile Strength) as conductor tension
parameter. On account of its improved ability to rate conductor self-damping capacity, on the
one hand, and to generalise results as widely as possible, on the other hand, it was resolved to
adopt parameter H/w, the ratio between the initial horizontal tensile load H and conductor weight
w per unit length. As stated in the brochure, this tension refers to initial horizontal tension before
any significant wind and ice loading and before creep, at the average temperature of the coldest
month on the site of the line.

Another parameter of prime importance is wind turbulence since it affects to a great extent the
Aeolian power imparted to vibrating conductors. This is scrutinised in Section 3 of the brochure
insofar as it relates to roughness of the terrain crossed by the lines.

Single unarmored, unprotected single conductors of the most common types are considered in
Section 4 of the brochure. Then comes the description of analytical models from four different
organisations that were used to draw tentative recommendations about safe design tensions of
unarmored, unprotected single conductors. These recommendations were finally ratified on
account of a successful “reality check” against available field experience.

Section 5 of the brochure deals with single conductor lines protected by means of Stockbridge-
type vibration dampers set up at span extremities. Addition of dampers obviously calls for the
introduction of another parameter intended at rating the protective capacities of the damping
system. As justified in this section, the rating parameter that was selected is LD/m, the ratio of
the product of span length L and conductor diameter D to conductor mass m per unit length.

3
Similarly to what was done in the previous section, Section 5 goes on with a description of the
two analytical models that were used to predict safe design tensions, which are then compared
with all available test line and field experience, again as a “reality check”. This part of the
brochure concludes on the Task Force recommendations which are provided both graphically
and algebraically.

Finally, Section 6 of the brochure deals with bundled conductor lines, particularly, twin
horizontal bundles, triple apex-down bundles and quad horizontal bundles made up of
conventional stranded conductors fitted either with damping spacers or non-damping spacers or a
combination of non-damping spacers and span-end Stockbridge-type dampers. It starts with a
thorough review of literature about field tests carried out in the past, mainly on dedicated test
lines, on bundled conductors set up in parallel and at the same tension as identical single
conductors. It then covers a review of field experience gathered on 91 bundled conductor lines
erected in North America. This is followed by a description of the methodologies that were used
to arrive at safe design tension for each one of a number of bundled conductor systems. In this
case, these methodologies are purely empirical, relying on field experience and full scale test line
data.

The Task Force recommendations are summarised in a table at the end of Section 6 in the form
of simple algebraic expressions, each one associated to a specific conductor system and to one
out of four terrain categories. For sake of completeness, the table also incorporates the Task
Force recommendations for single conductors fitted or not with Stockbridge-type dampers set up
at the span extremities.

4
1. INTRODUCTION

There may be several motives for controlling overhead line conductor tension at the design stage.
One of the most obvious reasons is to ensure that maximum tension corresponding to the
assumed most severe climatic loading does not exceed a predefined, acceptable conductor
tension. At the opposite end, it may be required to limit tension while the conductor is operating
at maximum temperature so that line clearance is not violated. A third motive is to restrict
conductor susceptibility to harmful Aeolian vibrations.

As a matter of fact, it is well known that stranded conductors get more vulnerable to Aeolian
vibrations as tension is increased. This is true for all conductor systems whether they are used in
solo or in bundles and whether they are fitted or not with damping and/or spacing devices.
Hence, there is a need to set an upper limit to conductor unloaded tension that prevails for
significant periods of time.

Guidance to such safe design tension with respect to Aeolian vibrations was provided before,
more particularly in 1960 by the so-called EDS (Every Day Stress) Panel [1], working under
appointment of CIGRE SC6 (since disbanded). However, as discussed in the following section of
this brochure, it was felt timely to revise the Panel’s recommendations.

It was a premise for the Task Force that it would be quite difficult to work out a revised guide on
the basis of field experience only due to the rarity of well documented cases. Accordingly, the
Task Force’s terms of reference have been: To produce a practical guide for the selection of safe
design tensions in form of generic values for specific classes of single transmission line
conductors to avoid damage due to Aeolian vibrations during a typical design lifetime. This
guide should be based on existing experience and current knowledge on wind energy input,
conductor self-damping and conductor fatigue endurance. A theoretical model should be used
that translates energy balance into amplitude and number of cycles of vibration at the clamps to
assess vibration severity.

This brochure deals with this very important issue in the design and operation of overhead
transmission lines. It starts with a review of the Every Day Stress (EDS) concept. That is
followed in Section 3 by a review of the influence of terrain cover or roughness on wind
turbulence, which is known to affect to a large extent wind power imparted to conductors.
Single, unarmored, unprotected single conductors of the most common types are considered in
Section 4. The following section deals with single conductor lines protected by means of
Stockbridge-type vibration dampers set up at span extremities. Section 6 of the brochure deals in
turn with bundled conductor lines, particularly twin horizontal bundles, triple apex-down bundles
and quad horizontal bundles made up of conventional stranded conductors fitted either with
damping spacers or non-damping spacers or a combination of non-damping spacers and span-end
Stockbridge-type spacers.
5
2. REVIEW OF THE EVERY DAY STRESS (EDS) CONCEPT

Forty-five years have elapsed since the presentation, at the CIGRÉ 1960 session, of Report 223
[1] of CIGRÉ SC6, which showed the conclusions reached by the EDS Panel appointed in 1953
to investigate the fatigue damages caused to transmission line conductors by the Aeolian
vibrations.

In that report the EDS, expressed as a percent of the conductor breaking strength (or UTS,
Ultimate Tensile Strength), was defined as the maximum tensile load to which a conductor can
be subjected, at the temperature which will occur for the longest period of time, without any risk
of damage due to Aeolian vibrations. Different values of EDS were given for bare conductors
and for conductors with armour rods only, dampers only and armour rods and dampers. The
Panel’s recommendations are shown in Table 2.1.


Table 2.1 EDS panel recommendations for safe design tensions in percent UTS.
Lines equipped with Unprotected
lines
Armour rods Dampers Armour rods
and dampers
Copper conductors 26
ACSR 18 22 24 24
Aluminium conductors 17
Aldrey conductors 18 26
Steel conductors
1. Rigid clamps
2. Oscillating clamps

11
13




Since Report 223 was issued, new conductors have been put in service and lines have been built
for the first time in new areas. With the broader experience, the recommended EDS values have
appeared to be insufficient to explain the fatigue damage found on these more recent lines. In
fact, in retrospect, even the statistical conclusions drawn from EDS Panel do not seem quite
correct.

For example, Table 2.2 shows the results of an analysis of the EDS Panel data for bare ACSR
conductors. It indicates that for lines that have been in service for 10 to 20 years, 45% of the
lines with an EDS<18% and 78% of the lines with an EDS>18% revealed damages. As can be
seen, the 18% EDS value proposed as “safe” lead to fatigue failures although at a slower rate.


6
Table 2.2 Summary of damaged lines as per the EDS panel
Service life % of lines damaged
Years EDS < 18% EDS = > 18%
< = 5 5.26 25.00
> 5 < = 10 20.93 35.29
> 10 < = 20 45.00 78.00
> 20 58.93 91.67


Research and testing performed since 1960 have provided information that was not available to
the EDS Panel. Self-damping tests on conductors showed that the increased tensile load reduces
the power dissipated by a vibrating conductor. To represent such effect, it is now considered
more convenient, in order to generalise the results, to use the ratio H/w between the horizontal
tensile load and the conductor weight per unit length rather than the % of UTS. This may not
have been evident to the EDS Panel because the vast majority of the lines up to 1962 were built
with the classical 30/7, 26/7 and 54/7 stranding ACSR. With such cables, the increase of UTS
due to an increase of the conductor diameter brought also an equal increase in the conductor
weight.

The EDS Panel classified the lines on the basis of terrain, (flat, hilly, mountainous, etc.), but it is
now well known that what influences the wind power is the surface roughness of the ground
which creates turbulence.

Finally, it is well known that conductor fatigue is the result of an accumulation of dynamic
bending strains in the presence of fretting [2-3]. It follows that a parameter often ignored is the
occurrence of dangerous winds on the line. In some locales, the occurrence and direction of these
winds is only related to general meteorological conditions but in other locations, close to large
water bodies or in desert areas, the thermal differences in areas surrounding the line cause daily
air flows in one direction and in the other. Under such conditions the rate of dynamic stress
accumulation can be significantly greater than in other areas.

All this information can now better explain the dispersion of the time to failure of the lines
investigated by the EDS Panel and, in general, of present similar lines in service.

It should be pointed out that the Panel offered its recommendations with clear reservations
regarding their fundamental soundness. Much poor experience that resulted from trying to apply
the recommendations was the result of ignoring those reservations.

In view of sometimes unsatisfactory experience with the Panel’s recommendations, and of
improved current understanding of Aeolian vibration, it is felt that new proposals are
appropriate.

7
3. TURBULENCE AS A FUNCTION OF TERRAIN

Turbulence in the wind affects the amount of Aeolian vibration power that is imparted to a
vibrating conductor. Maximum power is transferred when wind causes a Strouhal frequency of
vortex shedding that is close to the frequency at which the conductor is vibrating [4]. Due to the
travelling-wave nature of conductor vibration, the conductor’s vibration frequency, or frequency
spectrum, is the same all along the span. If the wind speed is not constant all along the span,
there must be parts of the span where maximum power transfer does not occur, and total wind
power input must be less than would occur in perfectly uniform wind. Fluctuations in wind speed
with time - gustiness - also reduce wind power input, because the vibration of the span cannot
change frequency fast enough to follow short-term wind speed variations. Both the span-wise
variations and the gustiness are reflections of turbulence in the wind structure.

The turbulence that influences Aeolian vibration arises from the interaction of the mean wind
with the ground. The magnitude of the turbulence is ordinarily taken as the root-mean-square
variation of wind velocity about the mean speed, and turbulence intensity is expressed as the
ratio of that RMS variation to the mean wind velocity.

Turbulence intensity at any particular field location is strongly influenced by the local terrain,
and especially the nature of the ground cover. Obstacles to wind flow, such as trees and
buildings, and even blades of grass, shed vortices somewhat as conductors do. These vortices
comprise the turbulent component of the wind. Naturally, large obstacles shed large vortices and
result in large turbulence intensity, while small obstacles, with their small and short-lived
vortices, result in low intensity.

Very large obstacles such as hills, ridges and mountains do not cause turbulence as understood
here. Rather, they shape the flow to conform to these gross orographic features. For example,
valleys may funnel the wind, increasing its speed and actually reducing its turbulence intensity.

Field measurements have yielded tables of typical values of intensity found in various classes of
ground cover. The values in Table 3.1 pertain to elevation above ground of 10 metres [9]. Tables
with more finely-divided classifications are available. For example, Wieringa [10] provides
twice as many.


Table 3.1 Typical values of turbulence intensity.
Terrain Turbulence
Intensity
Open sea; large stretches of open water 0.11
Rural areas; open country with few, low, obstacles 0.18
Low-density built-up areas; small town; suburbs;
open woodland with small trees
0.25
Town and city centers with high density of buildings;
broken country with tall trees
0.35
8
The intensities given in these tables are typical of values measured during strong winds and, in
fact, such measurements show consistent results only during strong winds. Measurements during
light to moderate winds, say up to 8 or 10 m/s (hourly average), show a great deal of dispersion.
This dispersion results from effects of the buoyancy acquired by parcels of air that are in contact
with the surface when they are heated or cooled by the ground. Heating causes these parcels to
rise, churning the atmosphere as they do, resulting in increased turbulence. Cooling, on the other
hand, causes the atmosphere to stratify, with the cool layer sticking to the ground and blocking
the movement of surface-generated turbulence upward. Because of these effects, turbulence
intensity during the light-to-moderate winds associated with Aeolian vibration can be much
larger, or significantly smaller than the values reported in the table. For example, in rural areas,
intensity can be as high as 0.50 or as low as 0.07, depending upon whether the ground surface is
warmer or colder than the air.


9
4. SINGLE UNPROTECTED CONDUCTORS

This section aims at recommending safe design tensions for unarmored, unprotected single
conductor lines.

Modelling

All models that were employed to estimate safe tensions basically rely on the Energy Balance
Principle [11] to predict steady state vibration in terms of fy
max
(the product of vibration
frequency and maximum vibration amplitude at antinodes), that is the response of a span when
excitation by the wind is balanced by the internal damping of the span. The flow chart (Fig. 4.1)
shows parameters that, in a modelling approach, may serve to describe the properties of the
structural system (the span) and the wind excitation, and Table 4.1 summarizes which of these
parameters were considered in models of four different organizations, and how they were
quantified.

Two approaches were followed to assess computed vibration levels (the span response) with
regard to the tolerance of the conductor to vibration. In the Endurance Limit Approach, vibration
levels are considered to result in an infinite lifetime of the conductor, if they do not exceed a
definite limit value (the endurance limit in terms of fy
max
). Conductor tensions that lead to
vibration levels below the endurance limit are regarded as safe.



Effective wind recurrence
Conductor tension design guide - modelling approach
Tension
Clamping system
Conductor type
Conductor stranding
Conductor condition
Span length
Conductor size
Effective wind turbulence
Wind statistics
Terrain category
Wind power input
Energy balance
Increase or
decrease tension
Resp. : Tol.
= O.K. , limiting
design value
Response
Intensity
Recurence
Tolerance to vibrations
Damping capacity
STRUCTURAL SYSTEM
EXCITATION
< >

Figure 4.1 Parameters describing undamped structural system and wind excitation.
10
Table 4.1 Basic assumptions underlying model calculations.
Organisation RIBE/Krispin [5] IREQ/Hardy, Leblond [6] Alcoa Fujikura/Rawlins [7] Claren [8]
Approach Endurance limit Endurance limit Cumulative damage Endurance limit Cumulative damage Cumulative damage
Conductor
damping capacity
Measured (decay
method, span with
pivoted ends)
Measured
(ISWR method)
and
Calculated on the basis of similarity laws
calibrated by means of measured data
Measured
(ISWR method)
Calculated on the basis
of equations that were
derived from measured
data
Wind power
input
Laminar & reduced
for turbulence
Laminar & reduced for normally-distributed
turbulence
Laminar & reduced for turbulence Laminar & reduced for
turbulence
Wind recurrence − − Rayleigh-distributed
(4.5 km/h avrg)
− Linearly increasing
probability density
from 0 to 7 mph,
constant probability
density (0.057)
between 7 and 12 mph,
linearly decreasing
from 12 to 27 mph.
Proper winds to occur
half of the time.
Rayleigh-distributed
(2.5 m/s avrg)
Vibration mode
shape
Sinusoidal Narrow band random vibration (peak
amplitudes Rayleigh-distributed, maximum
amplitude limited to 6 times rms-value at each
frequency)
Sinusoidal Random (Flat
probability density
from 0.4 to 1 times the
EBP-amplitude)
Random (amplitudes
normally distributed
about a mean
corresponding to 50%
of the maximum
laminar amplitude at
respective frequency,
maximum amplitude
limited to 3.5 times that
mean value)
Conductor
tolerance to
vibration
Fatigue endurance as
per [13]
Fatigue endurance as
per [13]
S/N-curves based on
[13] data for several
probabilities of survival
Fatigue endurance as
per [13]
S/N-curves based on
[13] data
CIGRÉ Safe Border
Line
Clamping system Fixed metal to metal
clamps
Fixed metal to metal clamps Fixed metal to metal clamps Fixed clamps
Predicted fatigue
life
Infinite Infinite 50 years with a
probability of survival
of 95%
Infinite 100 years 30 years



In the Cumulative Damage Approach, one assigns a certain portion of fatigue damage to each
vibration cycle. These small fractions of fatigue damage are assumed to accumulate at a certain
rate during the service life of the conductor, until fatigue breakage occurs. The usual assumption
is that of linear damage accumulation (Miner's Rule). This approach requires assumptions on the
recurrence of fatigue inducing stress levels to determine the number of occurrences at different
stress levels, and data on the probability of vibration exciting wind has to be introduced into the
model. Probabilistic considerations may be expanded to the S-N-curves that define the fatigue
inducing intensity of different vibration levels, and different S-N-curves may be surmised for
different levels of probability of survival. Safe conductor tensions that are calculated on that
ground necessarily pertain to a certain predicted fatigue life of the conductor.

Safe design tensions as predicted by the different organizations are depicted in Figs. 4.2 and 4.3
as a function of wind intensity of turbulence. Figures 4.2a & b were worked out on the basis of
the Endurance Limit Approach while Figs. 4.3a & b were prepared on the basis of the
Cumulative Damage Approach for multi-layer A1/Syz (ACSR) and A1 (AAC) conductor classes
[12] as well as for multi-layer A3 (AAAC) and A1/A3 (ACAR) conductor classes respectively.
Also shown in the figures are the tensions recommended by the EDS Panel, translated in terms of
H/w, for different conductor types. As the Panel did not account for the effect of turbulence, their
recommended tensions appear as flat horizontal lines in the figures.
11

Figure 4.2 Predicted conductor safe design tension according to endurance limit approach. (a)
Multi-layer A1 (AAC) and A1/Syz (ACSR) conductors; (b) Multi-layer A3
(AAAC-6201) and A1/A3 (ACAR) conductors.
12

Figure 4.3 Predicted conductor safe design tension according to cumulative damage approach.
(a) Multi-layer A1 (AAC) and A1/Syz (ACSR) conductors; (b) Multi-layer A3
(AAAC-6201) and A1/A3 (ACAR) conductors.
13
On the contrary, calculated safe design tensions are seen to increase with intensity of turbulence
pointing out that conductors are less susceptible to Aeolian vibration excitation by turbulent
winds.

It will be noticed that despite a considerable scatter in predicted safe tensions according to
organizations, they generally stand well below the corresponding EDS values, particularly at the
lowest intensities of turbulence. As for the scatter, it appears to result mainly from differences in
predicted conductor self-damping.

Three organizations computed admissible H/w for A3 (AAAC-6201) all aluminum alloy
conductors that are from 30 to 50% lower than those for A1 (AAC) all aluminum conductors or
A1/Syz (ACSR) steel reinforced aluminum conductors. This is the result of their calculations
being performed on the basis of lower fatigue endurance characteristics for the A3 conductors as
per reference [13].

Lastly, it may be observed that recourse to the cumulative damage approach leads to more
permissive H/w than the alternative approach. That stands to reason on account of the former
approach allowing for a certain number of vibration cycles above the conductor endurance limit
while the latter does not.

Comparison With Field Experience

Uncertainties in the data and assumptions that are required to determine maximum safe tensions
on the basis of the Energy Balance Principle have made it necessary to consider known
experience with existing lines when specifying protection for new lines. For that purpose,
collections have been made of historical data on existing lines, relative to their design tensions
and the occurrence or not of fatigued strands. Analyses of such collections have been published
by Zetterholm [1], Rawlins et al [14], Dulhunty et al [15] and Doocey et al [16].

These analyses rely upon evidence that there exist parameters that can rank spans according to
their risk of fatigue damage, and that well-chosen ranking parameters can properly rank
collections of spans of different design, as long as each collection is confined to a certain limited
class. For example, ACSR spans with dampers would belong to a different class from ACSR
spans with armor rods, and copper conductor spans, with dampers or with armor rods, would
belong to different classes from the ACSR spans. The power of such analyses hinges upon
choosing the ranking parameters well, defining the various classes narrowly enough, and yet
retaining enough cases in each class to define a pattern that distinguishes spans that are likely to
experience fatigue from the general population.

As noted earlier, the method adopted by the Task Force, to estimate maximum safe tensions for
Aeolian vibration, was Energy Balance Analysis. However, it also made use of the field
experience approach, as a “reality check.” In doing that, it was the view of the Task Force that,
since it was first concerned only with undamped, unarmored spans, the most significant design
parameter influencing the probability of fatigue is conductor tension, because of the impact of
tension upon conductor self-damping. However, tension can be expressed in various forms, such
as force, stress, percent of rated strength and others. The Task Force was faced with the need to
14
choose a form, for use as a ranking parameter that would be indifferent to such things as
conductor diameter, in order gather as many field cases into each class as possible. The
parameter that was selected was H/w.

Since tension H for any span is not constant, but varies with temperature, ice or wind loading
history and creep, it was necessary to define the reference condition for determining H. The Task
Force decided to use the average temperature for the coldest month as reference temperature, and
to determine tension for initial conditions, i.e., before wind and ice loading and creep.

The Task Force drew from several sources to form a collection of field experience cases for
undamped spans, i.e., spans equipped with neither dampers nor armor of any type. This
collection was used as a “reality check” on the predictions of maximum safe design tensions
based on the Energy Balance Principle, developed by Task Force members. The collection
contained enough cases to be useful only for conventional round strand ACSR. There were not
enough cases to indicate safe tension limits for aluminum or aluminum alloy conductors, ground
wires or OPGW. The ACSR collection is shown in Table 4.2.

In Table 4.2, the individual rows pertain to existing undamped ACSR lines for which data is
available. They are arranged in order of increasing H/w, and for each line it is indicated whether
conductor fatigue is known to have occurred.

The tension-ranking parameter H/w is expected to correlate with failure experience, and a
limiting value of H/w below which failures do not occur is expected to exist. Spans having H/w
values greater than this limit may survive without experiencing fatigue failure due to the
protective effects of terrain features such as trees and buildings, or where the line runs parallel to
a valley that channels the wind. However, no failures should occur where H/w is below the limit.

The table suggests that the limit may be about H/w = 1000 meters. However, there are not
enough experience cases in the vicinity of that value to support more than a tentative conclusion.
The lowest failure case, with H/w = 1030 meters, involved pin-type insulators with hand-formed
ties, whereas almost all the other cases involved spans supported by suspension clamps. Thus,
the departure of this case from the main pattern may indicate that pin insulator supports cannot
be included in the same class as suspension clamp supports. In addition, the span was short,
making the determination of tension particularly sensitive to the precision of “sagging in” and to
the choice of reference temperature. If this case is ignored, the safe limit may be as much as
H/w = 1400 meters. However, there are not enough experience cases in the interval between
1000 and 1400 m to determine where in this range the limit should fall. One can only conclude it
is somewhere in this interval.
15
Table 4.2 Field Experience Cases for Undamped ACSR.
Conductor Diam Al / St Average H/w Fatigue
(mm) Strands Span (m) (m) failure
21.9 36/12 200 707
21.9 36/12 395 844
24.2 54/7 137 934
8.0 6/1 61 1029 Yes
21.8 26/7 183 1358
26.6 26/19 362 1397 Yes
16.5 310 1405 Yes
18.8 30/7 396 1511 Yes
18.8 30/7 350 1554
10.7 12/7 300 1607 Yes
21.8 26/7 274 1638
25.9 30/7 396 1655 Yes
21.8 26/7 326 1723
20.5 26/7 300 1731 Yes
25.4 54/7 346 1735
19.9 26/7 170 1738
22.4 30/7 333 1747
21.0 30/7 390 1761 Yes
12.7 6/1 107 1772
22.4 30/7 340 1865 Yes
21.7 48/7 295 1881 Yes
18.8 30/7 270 1908
18.8 30/7 360 1959
11.7 12/7 264 1996 Yes
19.6 30/7 350 2001
19.6 30/7 350 2001
26.4 32/19 655 2095 Yes
26.4 32/19 445 2108 Yes
26.4 32/19 520 2116 Yes
22.4 34/19 475 2154
25.1 26/7 580 2162
26.4 32/19 475 2176
21.9 36/12 315 2243
18.3 26/7 440 2279 Yes
31.7 114/37 425 2297 Yes
31.7 114/37 415 2300 Yes
31.7 114/37 374 2382 Yes
14.4 12/7 194 2458 Yes
22.3 30/7 350 2861 Yes


General Discussion

Attempts to determine maximum safe tensions for Aeolian vibration on the basis of field
experience only have been difficult and sometimes unsuccessful. This is due to the fact that
actual conductor condition, tension history and even, on occasions, design tensions are not
known. One way round is to have recourse to modelling which is the method adopted by the
Task Force. One substantial advantage of this approach is the possibility to consider all relevant
parameters and variables in a systematic manner.

However, caution is compulsory despite the improved scientific appearance of the modelling
approach. Vibration-induced fatigue degradation of conductors is a problem of a highly complex
16
and highly random nature that is not easily amenable to mathematical analysis. From the very
first, the wind, the sole cause to Aeolian vibrations, is itself strongly random, with the wind
speed and direction fluctuating continuously both time-wise and space-wise, as noted above. The
conductors respond to variations in weather conditions associated with the passage of major
weather systems by displaying markedly different amplitudes and frequencies on different days,
and even different hours of the same day. They respond to differences in wind structure by
vibrating less when and where the wind is turbulent or gusty. Thus, the rate of accumulation of
fatigue cycles is random over time and varies markedly with location.

What is more, conductor self-damping, which determines the vibration intensity, is a strong
function of tension which itself is far from being a constant depending on non-deterministic
variables such as conductor temperature and also creep which in turn depends on wind and ice
loading history. For ACSR conductors, self-damping is even dependent on the share of tension
between the aluminium strands and the steel strands and that is again a function of temperature
and creep.

Conductor fatigue endurance, the conductor faculty to resist vibratory motion without strand
failure, is another random variable that comes into play. It has been determined that lifetime
before first strand failure of several samples of the same conductor, tested according to the same
ideal laboratory protocol, may be scattered over a range as large as 1-50. Dispersion in the cycles
to failure is indeed inherent to metal fatigue, and is increased by effects of variability in mill
processing and, particularly, effects of fretting.

Moreover, there is already a large scatter in the available experimental data yet, pertaining to the
same simplified vibratory schemes. Hence, there is a 2 to 1 factor between the highest and the
lowest value of maximum wind power input into conductors according to investigators [17].
What is even more consequential, there may be one order of magnitude difference in conductor
self-damping estimate at the same tension according to testing methods [17]. Likewise, as regard
conductor fatigue, there is a 2 to 1 ratio in the endurance limit of multilayer ACSR’s whether it is
extracted from EPRI’s Orange Book [13] or from the CIGRÉ Safe Border Line [18]. That entire
scatter would probably intensify had the true nature of the beat-pattern-like vibrations been duly
considered.

All that without saying anything about the great diversity in conductor types, sizes and makes as
well as in suspension clamps or other supporting devices.

All those complexities and uncertainties just enhanced the necessity of resorting to sound
analysis and judgement. Obviously, simplification is required to overcome complexities. Some
conservatism is needed to counterbalance uncertainties. And comparison against field experience
and test line results is a necessity to gain confidence in the recommended safe tensions. That has
been the philosophy that has guided the Task Force throughout.

Hence, initial tension at the average temperature of the coldest month was chosen as a base to the
recommendations knowing that any temperature increase, any wind and ice loading as well as
the inevitable creep would soon reduce tension permanently, thus alleviating the vibration
17
severity. Furthermore, the recommended safe tensions stand well in the lower range of the
predicted values.

The recommended safe H/w may appear over-conservative. Nevertheless, it should be noticed
that they generally stand above the 17-18% of UTS recommended by the EDS Panel for all
aluminum A1 (AAC) conductors and low steel-content aluminum conductors A1/Syz (ACSR).

Recommendations

The maximum safe design tensions with respect to Aeolian vibrations of undamped and
unarmored conductors, as recommended by the Task Force, are shown in Table 4.3 as a function
of terrain category. The table uses H/w, the ratio of horizontal tension in the span to conductor
weight per unit length, as the tension parameter. It is important to note that this horizontal
tension refers to initial horizontal tension, before wind and ice loading and before creep, at the
average temperature of the coldest month on the site of the line.

Recommended safe tensions apply to the following round strand conductors: all aluminium A1
(AAC) conductors; all aluminium alloy A2 or A3 (AAAC) conductors; aluminium/aluminium
alloy A1/A2 or A1/A3 (ACAR) conductors and steel-reinforced aluminium A1/Syz (ACSR)
conductors. It was resolved to give a uniform recommendation for all types of conventional
conductors using aluminium and/or aluminium alloy. Although a lower fatigue endurance of A2
(AAAC) conductors may be surmised from reference [13], there seems to be no well
documented field evidence to support a more pessimistic tension recommendation for these
conductors.

Terrains have been divided into four categories according to general characteristics. Should there
be any doubt about real terrain category, the lowest class should be selected.



Table 4.3 Recommended conductor safe design tension at the average temperature of the
coldest month as a function of terrain category. Valid for homogeneous aluminium
and aluminium alloy conductors Ax (AAC and AAAC), bi-metallic aluminium
conductors Ax/Ay (ACAR) and steel reinforced aluminium conductors A1/Syz
(ACSR).
Terrain
category
Terrain characteristics
( )
H
w
adm

(m)
1 Open, flat, no trees, no obstruction, with snow cover, or
near/across large bodies of water; flat desert.
1000
2 Open, flat, no obstruction, no snow; e.g. farmland without any
obstruction, summer time.
1125
3 Open, flat, or undulating with very few obstacles, e.g. open
grass or farmland with few trees, hedgerows and other
barriers; prairie, tundra.
1225
4 Built-up with some trees and buildings, e.g. residential
suburbs; small towns; woodlands and shrubs. Small fields with
bushes, trees and hedges.
1425
18
The maximum safe design tensions recommended herein should be suitable most of the time.
However, special situations require specific attention. Such is the case for extra long spans, or
spans exposed to pollutants that may decrease the self-damping or the fatigue endurance of the
conductor, or spans often covered with ice, rime or hoarfrost, or spans operated at high
temperature.

Generally speaking, damping spans is inexpensive and that is certainly preferable to hazarding
conductor fatigue breaks. Moreover, use of damping may allow higher tensions resulting in
significant cost savings in line construction.

Use of armour rods or special supporting devices such as cushioned clamps and helical
elastomer-bushed suspensions may justify higher design tensions on otherwise unprotected
conductors. Information on safe tension, when these devices are employed, should be obtained
from their suppliers.

Existing lines using single unprotected conductors strung at a tension exceeding the
recommended value for the terrain may require inspection and field measurement. Techniques to
perform such vibration measurements have been described previously [19].

Incidentally, in some countries, the maximum safe design tension may be governed by the
maximum climatic loading rather than by Aeolian vibrations.

19
5. DAMPED SINGLE CONDUCTORS

This section aims at recommending safe design tensions for single conductor lines protected by
means of Stockbridge-type vibration dampers set up at the span extremities.

Modelling

The two models that were employed to estimate safe tensions basically rely on the Energy
Balance Principle [11] to predict steady state vibration in terms of f y
max
(the product of vibration
frequency and maximum vibration amplitudes at antinodes), that is the response of a span when
excitation by the wind is balanced by the self-damping in the conductor and by damping devices.

The flow chart in Fig. 4.1 shows parameters that, in a modelling approach, may serve to describe
the properties of the structural system (the span and span-end damping arrangements) and the
wind excitation, and Table 5.1 summarizes which of these parameters were considered in models
of different organisations, and how they were quantified.


Table 5.1 Basic assumptions underlying model calculations.
Organisation IREQ / Leblond & Hardy [25] Alcoa Fujikura / Rawlins [26]
Approach Endurance Limit Endurance Limit
Conductor damping
capacity
Calculated on the basis of similarity
laws calibrated by means of
measured data
(ISWR method).
Measured
(ISWR method)

Span-end damping Travelling wave approach using
complex damper stiffness measured
on shaker
Measured efficiency of span-
end damping arrangement
Wind power input Laminar or reduced for normally-
distributed turbulence
Laminar or reduced for
turbulence
Vibration mode
shape
Narrow band random vibration
(peak amplitudes Rayleigh-
distributed, maximum amplitude
limited to 3.5 times RMS-value at
each frequency)
Sinusoidal
Conductor tolerance
to vibration
Fatigue endurance as per [13] Fatigue endurance as per [13]
Clamping system Fixed metal to metal clamps Fixed metal to metal clamps



20
As the Task Force focused on Stockbridge-type dampers applied at span ends, information on the
span-end damping was needed to be employed in the Energy Balance analysis. Data was
available that was obtained by indoor testing techniques described in IEEE Std 664-1993, “IEEE
Guide for Laboratory Measurement of the Power Dissipation of Aeolian vibration Damper for
Single Conductors” (see also ref. [20]). One technique uses a conductor span to determine the so-
called efficiency of a span-end damping arrangement which can directly be used in the Energy
Balance [21]. Another technique consists in measuring the dynamic characteristics of a
Stockbridge-type vibration damper on a shaker. Travelling wave analysis was then used to
predict the span-end dissipation characteristics from the measured complex characteristics of the
damper [21].

The Endurance Limit Approach, as defined in Section 4, was generally followed to assess
computed vibration levels (the span response) with regard to the tolerance of the conductor to
vibration. In this approach, vibration levels are considered to result in an infinite lifetime of the
conductor, if they do not exceed a definite limit value (the endurance limit in terms of f y
max
).
Conductor tensions that lead to vibration levels below the endurance limit are regarded as safe.

Selection of Parameters

The Task Force has focused on Stockbridge-type dampers, since this is at present the type most
widely used on conductors. It is noted that there are differences in design and application rules
among Stockbridge dampers of different sources. The Task Force is not in a position to
distinguish among these, and does not feel that it should. Thus, it has been forced to seek a single
safe tension criterion for a population that is not homogeneous, even though limited to
Stockbridge-type dampers.

The Task Force was aided, in dealing with this difficulty, by the observation that tests on
laboratory spans as well as on an outdoor test span [22] have shown somewhat similar level of
damping efficiency
1
using dampers of various sources. This observation was found to be true,
even for widely different conductor sizes, when applied to well-designed and properly applied
dampers of sizes appropriate to the conductors in question. In fact, certain utility specifications
specify the same minimum level of performance in terms of damping efficiency for a significant
range of conductor sizes [23].

The similarity in damping efficiency levels pointed toward a particular parameter to use in rating
the protective capabilities of dampers. This rating parameter, m H D L , (where L is actual span
length, D is conductor diameter, H is horizontal tension in the conductor and m is mass of the
conductor per unit length) has been used in analyses of collections of field experience data on
fatigue of overhead conductors [24] to rank spans according to the difficulty in damping them,
based on the line design variables, span length, conductor size and tension. The parameter is
proportional to the damping efficiency required to control vibration amplitude to a given level.
Since the Task Force had adopted the parameter H/w (where w is weight of the conductor per

1
Damping efficiency is defined [21] as the ratio of power actually dissipated by the damper to the power that would be dissipated by
a perfect damper - one that absorbed incident waves without reflection - at the same frequency and amplitude. In the laboratory,
damping is measured through the inverse standing wave ratio [20].
21
unit length) to rate the effect of tension on conductor self damping, it was able to simplify the set
of rating parameters m H D L and H/m to LD/m and H/w respectively.

Predicted Safe Design Tension

Safe design tension as predicted by Leblond and Hardy [25] for one conductor type over a full
range of span lengths and by Rawlins [26] for three sets of conductor types and span lengths is
depicted in Fig. 5.1 in terms of the span parameter LD/m and the tension parameter H/w. As a
rule, the calculations were carried out using the endurance limit approach for the A1/Syz
2

(ACSR) aluminium-conductors-steel-reinforced indicated in the legend. The results shown here
were determined on the basis of an assumed constant level of wind turbulence ranging from 5%
to 30%. It will be noticed that despite a large scatter in the calculated safe design tension
according to the source, there is a fair agreement in the minimum, more conservative, values.



3
1000 1500 2000 2500 3000 3500
H/w, (m)
0
2
4
6
8
10
12
14
16
18
20
22
L

D

/

m
,

(
m

/
k
g
)
E
D
S

-

A
C
S
R

4
2
/
7
E
D
S

-

A
C
S
R

4
8
/
7
E
D
S

-

A
C
S
R

2
6
/
7
E
D
S

-

A
C
S
R

3
0
/
7
E
D
S

-

A
l
d
r
e
y
Leblond and Hardy [14] : Bersfort : , 8% turb.; _, 15% turb.; l, 22% turb.; ¯, 30% turb.
Rawlins [15] : Chukar : ., 5% turb.; >, 10% turb.; Condor : , 5% turb.; ~, 10% turb.;
Drake, damper A : W, 5% turb.; ³, 10% turb.; Drake, damper B : ª , 5% turb.; ¦, 10% turb.

Figure 5.1 Predicted safe boundaries according to endurance limit approach.



Also shown in Fig. 5.1 are the safe tensions recommended by the EDS Panel, translated in terms
of H/w, for different damped A1/Syz (ACSR) and Aldrey conductors. As the Panel did not

2
International Electrotechnical Commission (IEC) conductor designation [12] : see Table 5.3.
22
account for the span length, their recommended tensions appear as straight vertical lines in the
graphics. For the common range of span parameters, 5 < LD/m < 15 (m3/kg), it may be observed
that the safe design tension as calculated by the endurance limit approach for 8% wind
turbulence is respectively more permissive, about equally permissive or less permissive than the
corresponding EDS values for low steel content, medium steel content or high steel content
A1/Syz (ACSR) conductors. However, for Aldrey conductors, the calculated safe design tension
is definitely much more conservative over all the span range than the value recommended by the
EDS Panel, i.e. 26% of UTS or about 3000 m.

Comparison With Field Experience

In the previous section dealing with Unprotected Single Conductors, it was remarked,
“Uncertainties in the data and assumptions that are required to determine maximum safe tensions
on the basis of the Energy Balance Principle have made it necessary to consider known
experience with existing lines when specifying protection for new lines.” These uncertainties are
even more significant in the present case where dampers are involved. The analytical basis for
applying the Energy Balance Principle is still open to question and is, in fact, currently under
scrutiny by CIGRÉ TF B2.11.01 “Vibration Principles”. Thus, the need to test the present
recommendations against past experience is even greater than before.

It is worthwhile to examine the range of the ranking parameters LD/m and H/w actually
represented in overhead lines. Fig. 5.2 shows a collection of points representing a number of
actual lines, all of which are protected by Stockbridge-type dampers. It is not certain that all of
these lines were free of fatigue damage, but most probably were. The collection was drawn from
files of Task Force members.

Figure 5.2 includes the estimated safe boundaries depicted in Fig. 5.1. It is evident that many
existing lines fall on the safe side of the estimated safe boundaries, but a significant number fall
on the “unsafe” side. It is likely that a few of these “unsafe” lines did experience damage, and
that others were protected from severe vibration by very rough terrain, beyond Category 4.
However, the Task Force interprets the presence of so many cases on the unsafe side as
indicating a conservative bias in the estimated safe boundaries.

The TF found only meagre information on field experience cases where conductor fatigue
occurred in lines protected by Stockbridge type dampers. Table 5.2 summarises cases from
questionnaires collected by CSC6 [2] and SC22 WG04 for all damper types. Only three cases,
Items 1, 5 and 6, clearly pertain to Stockbridge dampers. Items 7 and 8 may or may not pertain
to Stockbridge dampers, but it is considered likely that the fatigue occurred before any dampers
were installed. For Items 1 and 6, the damage may have occurred in spans that were not fitted
with dampers. For Item 1, the figure for H/w applies to the design value of H. Actual tension
was higher due to contractor error, but an actual value was not reported. Thus, Item 5 represents
the only case that is clearly valid for testing the estimated safe boundaries against field
experience. The case led the utility to increase the number of dampers in designing its lines,
implying that the damaged line had originally been fitted with less damping than might have
been required.

23
3
Terrain
Category : 1 2 3 4
1000 1500 2000 2500 3000 3500
H/w, (m)
0
2
4
6
8
10
12
14
16
18
L

D

/

m
,

(
m

/
k
g
)
Ruling spans
Maximum spans

Figure 5.2 Ranking parameters of damped ACSR lines in North America (no water crossings) in
relation to estimated safe boundaries.


Table 5.2 Field experience conductor fatigue cases – ACSR.
Item
Diam.
mm
Stranding
Span
metres
H/w
metres
LD/m
m
3
/kg
Damper type
Note
1 16.28 26/7 167 1542 4.983 Stockbridge 1
2 31.59 54/7 360 2017 5.755 Torsional
3 27.00 30/7 290 1851 5.861 Dumbell
4 27.00 30/7 305 1851 6.164 Dumbell
5 26.60 26/19 380 1994 6.971 Stockbridge
6 31.77 54/7 320 2031 5.086 Stockbridge 2
7 27.72 54/7 305 1677 5.691 “yes” 3
8 24.20 54/7 268 1406 5.592 “yes” 3
9 26.40 32/19 510 2227 8.496 Bretelles
10 27.72 54/7 330 1734 6.140 Elgra
1. Dampers may have been installed in dead-end spans and at angle towers only. H/w based on design tension.
Actual tension was higher due to contractor error.
2. Not all spans were damped. Damage may have only occurred in spans that were protected by armour rods only.
3. Lines were built in 1930 and 1927, respectively, before dampers were commercially available. Evidently, dampers
were added later. Damage likely occurred before they were installed.
24
Figure 5.3 shows all of the cases from Table 5.2, for comparison with the estimated safe
boundaries. All but Item 9 fall on the “safe” side of all four boundaries. The one valid case for
Stockbridge dampers, Item 5, falls slightly on the safe side of the Terrain Category 1 boundary.
Item 6, which may or may not be valid, falls similarly. Both cases came from Category 1 terrain.
Their position in the plot suggests that the proposed safe boundaries are not excessively
conservative.

Figure 5.3 also includes data from a number of operating lines where there is information that
fatigue damage to conductors has not occurred. The cases drawn from the test line are
particularly well documented because of the close control of conductor tensions, careful
vibration measurements and close inspection for damage. All these cases are in harmony with
the proposed safe boundaries.



3
Conductor damage reported : ¯, valid case;
³, other cases;
e, other damage reported.
No damage reported : W, individual spans;
, one damper/span test line safe cases;
, Zebra conductor over 2800 route km;
, Lynx conductor over 4000 route km.
1000 1500 2000 2500 3000
H/w, (m)
0
2
4
6
8
10
12
L

D

/

m
,

(
m

/
k
g
)
Terrain category :
1 2 3 4

Figure 5.3 Available field experience regarding damped single conductor lines in relation to
estimated safe boundaries.
25
Task Force Recommendations

As explained above, the Task Force has resolved to provide guidance to conductor safe design
tension with respect to Aeolian vibrations in terms of two parameters: the tension parameter H/w,
the ratio of horizontal tension H in the span to conductor weight w per unit length and the span
parameter, LD/m, the ratio of actual span length L times conductor diameter D to conductor mass
m per unit length. The tension H refers to initial horizontal tension before any significant wind
and ice loading and before creep, at the average temperature of the coldest month on the site of
the line.

The Task Force recommendations are depicted in Fig. 5.4 in the form of four sets of curves, each
one set associated to a particular terrain category described in the legend. The corresponding
information is provided in Table 6.3 in algebraic form. Terrains have been divided into four
categories according to their general characteristics. Should there be any doubt about real terrain
category, the lowest category should be selected.

The basic Safe Design Zone - No Damping applies to undamped and unarmoured single
conductors, as already shown in Section 4. This zone is defined in terms of the H/w parameter
only and it is unlimited in the LD/m parameter.

The Safe Design Zone - Span End Damping constitutes a zone where full protection of single
conductors against Aeolian vibrations is assuredly feasible by means of one or more
Stockbridge-type damper(s) set up at span extremities. Hence, within the limits of this zone,
Aeolian vibrations should not be a constraint on design tension.

For line parameters falling in the Special Application Zone, Aeolian vibrations may or may not
be a constraint and it is recommended that line designers determine the availability of adequate
protection before finalising the design.

This guide applies to all round wire, concentric lay, overhead electrical conductors shown in
Table 5.3.



Table 5.3 Conductor types to which recommendations apply.
Metal Combination Common
Designation
IEC
Designation
All 1350-H19 ASC or AAC A1
All 6101-T81 AASC or AAAC A2
All 6201-T81 AASC or AAAC A3
1350-H19 / Steel ACSR A1/S1A
1350-H19 / 6101-T81 ACAR A1/A2
1350-H19 / 6201-T81 ACAR A1/A3
26
3
0 500 1000 1500 2000 2500 3000
H/w, (m)
0
2
4
6
8
10
12
14
16
18
20
L

D

/

m
,

(
m

/
k
g
)
Terrain Category
#1 : Open, flat, no trees, no obstruction, with snow cover,
or near/across large bodies of water or flat desert.
#2 : Open, flat, no obstruction, no snow; e.g. farmland
without any obstruction, summer time.
#3 : Open, flat, or undulating with very few obstacles, e.g.
open grass of farmland with few trees, hedgerows
and other barriers; prairie, tundra.
#4 : Built-up with some trees and buildings, e.g. residential
suburbs; small towns; woodlands and shrubs. Small
fields with bushes, trees and hedges.
Safe Design
Zone
No Damping
Special Application
Zone
Safe
Design
Zone
Span-
End
Damping

Figure 5.4 Recommended safe design tension for single conductor lines. H: initial horizontal
tension; w: conductor weight per unit length; L: actual span length; D: conductor
diameter and m: conductor mass per unit length.



It was resolved to give a uniform recommendation for all types of conventional conductors using
aluminium and/or aluminium alloy. Although a lower fatigue endurance of A3 (AAAC)
conductors may be surmised from reference [13], there seems to be no well documented field
evidence to support a more pessimistic tension recommendation for these conductors.

27
Existing lines using single conductors strung at a tension exceeding the recommended value for
the terrain and span length may require inspection and field measurement. Techniques to perform
such vibration measurements have been described previously [19].

It may be noted here that conductors do not get more susceptible to other forms of wind-induced
vibrations when tension is increased. As a matter of fact, their propensity to both galloping [27]
and wake-induced oscillations [28] (of bundled conductors) has been shown to decrease with
increased tension.

Limitations

As mentioned above, the Task Force has focused on Stockbridge-type dampers. It is noted that
the “Bretelle” is also used in some countries, helical impact dampers are widely used on earth
wires and special designs such as “Festoon” dampers are employed in certain applications such
as fjord crossings. However, it was judged not feasible at this juncture to arrive at safe tension
recommendations for these types.

The Task Force has used the parameters H/w and LD/m to construct boundaries within which
satisfactory performance should be commercially available, often from multiple sources. This
statement should in no way be taken to mean that all commercially available dampers will
provide such protection up to the limits of those boundaries or that they could provide such
protection indefinitely without themselves failing from fatigue. Rather, appropriate field testing
or evaluation of available dampers is likely to reveal at least some that do. It falls to the line
designer to identify them.

The guidance provided herein should be suitable most of the time. However, special situations
require specific attention. Such is the case for extra long spans; for spans often covered with ice,
rime or hoarfrost in which case dampers may break up prematurely as a result of galloping
and/or excessive Aeolian vibrations; for spans exposed to pollutants that may decrease the
fatigue endurance of the conductors; for spans equipped with warning devices and for spans
using non-conventional conductors such as compact or high temperature conductors.

Use of armour rods or special supporting devices such as cushioned clamps and helical
elastomer-bushed suspensions may justify higher design tensions. Information on safe tension,
when these devices are employed should be obtained from their suppliers.
28
6. BUNDLED CONDUCTORS

The present section aims at recommending safe design tensions for bundled conductor lines. The
recommendations cover twin horizontal bundles, triple apex-down bundles and quad horizontal
bundles made up of conventional stranded conductors fitted either with damping spacers or non-
damping spacers or a combination of non-damping spacers and span-end Stockbridge-type
dampers.

Review of Literature

Numerous field tests have demonstrated that bundled conductors respond less to Aeolian
excitation than single conductors of the same size and at the same tension as those of the bundle.
For example, Fig. 6.1 shows results of simultaneous recordings at Alcoa’s outdoor laboratory
in1960 [29]. Bundling reduced vibration amplitudes by about half, with and without dampers on
the span.





Figure 6.1 Comparison of vibration in single versus bundled conductors
Drake ACSR in 1200-foot span [29]. (MILS ≡ in x 10
-3
; CPS ≡ Hertz)
1 – Single conductor 2,3 – Subconductors in horizontal two-conductor bundle
A – No dampers B – One Stockbridge damper at end opposite recorder.



Early investigations showed that the reduction in amplitude becomes greater as the number of
subconductors in the bundle increases. Leibfried and Mors reported that spans in the
Hornisgrinde Test Station displayed, on average, amplitudes in the ratios 4:2:1 in spans with
single, twin- and quad-bundles respectively [30]. Liberman and Krukov reported that amplitudes
29
were reduced by factors of 1.5 to 2.5 for horizontal two-bundles, and of 5 to 10 for triple and
quadruple bundles [31]. There was some influence by spacer design, but in all of these tests the
spacers did not include intentional damping. Thus, the benefits of bundling were attributed to the
effects of mechanical coupling among the subconductors interfering with the vortex excitation
mechanism.

The effect of intentional damping in spacers was investigated initially by Edwards and Boyd at
Ontario Hydro [32]. They found that damping of a certain type reduced amplitudes in a twin
bundle by a factor of 5, relative to a comparable single conductor, and by a factor of 20 in a quad
bundle. The benefit of damping was confirmed by Diana et al. at the Porto Tolle Test Station in
Italy [33]. The effect of bundle configuration upon this benefit was investigated by Hardy et al.
at the Magdalen Islands Test Station of IREQ [34]. Damping in spacers is beneficial in all bundle
configurations, but seems greatest in those that have size in the vertical direction, such as triple
and quad bundles.

Table 6.1 collects the test conditions and salient data from a number of these field tests.

Review of Field Experience

Table 6.2 summarises some line design variables of 91 bundled conductor lines in North
America which have most likely operated for many decades without any Aeolian vibration
problem. It comprises 70 twin horizontal bundled lines (2H), out of which 19 lines have been
fitted with non-damping spacers (NDS) alone, 48 lines fitted with a combination of such spacers
and span-end Stockbridge dampers (NDS+Stk) and 3 lines fitted with damping spacers (DS)
alone. The tension parameter H/w ranged up to 2088 m in the first case, 2959 m in the second
and 1937 m in the third.

Table 6.2 also gives details of 14 triple apex-down bundled lines (3AD), out of which one line
only has been equipped with NDS, 4 lines with NDS+Stk and 9 lines with DS. This time, the
tension parameter H/w reached 1627 m in the first case, 2056 m in the second and 2096 m in the
third.

Finally, Table 6.2 includes 7 quad horizontal bundled lines (4H), out of which 3 are protected by
means of NDS+Stk and 4 others by means of DS. In this case, the tension parameter H/w was set
to a maximum of 1488 m and 1937 m respectively.

30
Table 6.1 Synthesis of field test experience about comparative vibration behaviour of bundled
conductors.
Ref. Bundle Nominal Test span Spacer End Amplitude ratio
type conductor H/w LD/m type damper single/bundle
(m) (m
3
/kg)
[29] Hor. twin 1755 6.3 Articulated No >2,1
[29] Hor. twin 1755 6.3 Articulated Yes >1,8
[29] Hor. twin 1755 6.3 Ball-&-socket No >2,7
[29] Hor. twin 1755 6.3 Ball-&-socket Yes >1,9
[30] Hor. twin 1295 6.5 Various No ~2
[30] Hor. quad 1295 6.5 Various No ~4
[31] Hor. twin >1454 6.5 Articulated No >1,5
[31] Hor. twin >1437 7.5 Articulated No >1,5
[31] Hor. twin >1730 6.3 Articulated No >1,7
[31] Vert. twin >1454 6.5 Articulated No >1,7
[31] Vert. twin >1730 6.3 Articulated No >5
[31] Triple >1437 7.5 Grouped twins No >5
[31] Hor. quad >1730 6.3 Grouped twins No >5
[33] Hor. twin 1743 7.2 Rigid No >1,3
[33] Triple 1743 7.2 Damping spacers No >5
[34] Hor. twin 1550 6.8 Damping spacers No ~2,4
[34] Triple 1550 6.8 Damping spacers No ~6,5
[34] Hor. quad 1550 6.8 Damping spacers No ~7,7
[34] Hor. twin 2325 6.8 Damping spacers No ~3,6
[34] Triple 2325 6.8 Damping spacers No ~8,6
[34] Hor. quad 2325 6.8 Damping spacers No ~13,0




Table 6.2 Outlook of field experience with bundled conductor lines in North America.
No. of lines Bundle Protection Range of Range of
type mean LD/m initial H/w
(m
3
/kg) (m)
19 2H NDS 2.19 - 4.63 802 - 2088
48 2H NDS+Stk 3.14 - 7.27 910 - 2959
3 2H DS 5.03 - 6.60 1636 - 1937
1 3AD NDS 5.62 1627
4 3AD NDS+Stk 6.20 - 6.93 1166- 2056
9 3AD DS 3.93 - 7.81 1401 - 2096
3 4H NDS+Stk 6.63 - 7.89 1452 - 1488
4 4H DS 7.33 - 8.38 1633 - 1937
31
Determination of Safe Design Tension

As an accommodating reference and benchmark, the safe design tensions that were
recommended previously for undamped single conductors (conductor system #1) and single
conductors fitted with Stockbridge dampers (Stk) at the span extremities (system #2) are shown
in Table 6.3 for each one of the four terrain categories that were then defined.

6.3.1 Unspacered bundled conductors fitted or not with span-end Stockbridge dampers
Unspacered bundled conductors have sometimes been used with the object of reducing their
susceptibility to galloping. In such cases, it appears reasonable to use the same safe design
tension limits as for equivalent single conductors whether they are undamped (system #3) or
damped by means of Stk (system #4), as indicated in Table 6.3

6.3.2 Bundled conductors fitted with non-damping spacers
The Task Force could not find any well-documented, instrumented field test data related to twin
horizontal bundles fitted with NDS alone. But field experience suggests that a tension of
H/w = 2100 m should be safe for terrain category #2, as it could be appreciated for one of the
lines in Table 6.2. However, to keep on the prudent side, it was resolved to associate that safe
design tension to terrain category #3. The safe design tensions for terrain categories #1, 2 and 4,
as shown in Table 6.3 (system #5), were then determined on the basis of terrain category #3
using the same terrain-to-terrain ratio as for the undamped single conductor case.

For triple apex-down bundled lines fitted with NDS, reference was made to comparative field
tests [35] carried out at the IREQ test line in Varennes clearly showing such a tension
H/w = 2100 m to be safe for terrain category #2, applicable to the test station. The safe design
tensions for the other terrain categories were determined using the same transposition principle
as for the twin bundles. However, as indicated in Table 6.3 (system #8), it was resolved to limit
the safe design tension for terrain category #4 to H/w = 2500 m as a matter of prudence.

The same absolute limit (H/w = 2500 m) was indeed applied to all systems even if they are
strung over the roughest terrain.
32
Table 6.3 Recommended conductor safe design tension with respect to Aeolian vibrations.
Conductor system
H/w (m) LD/m (m
3
/kg) H/w (m) LD/m (m
3
/kg) H/w (m) LD/m (m
3
/kg) H/w (m) LD/m (m
3
/kg)
1. Undamped single conductor < 1000 < 1125 < 1225 < 1425
2. Single conductor with span-end Stockbridge dampers < 2615/(LD/m)
0.12
< 15 < 2780/(LD/m)
0.12
< 15 < 2860/(LD/m)
0.12
< 15 < 3030/(LD/m)
0.12
< 15
3. Undamped, unspacered twin, triple & quad bundled < 1000 < 1125 < 1225 < 1425
conductors
4. Unspacered twin, triple & quad bundled conductors < 2615/(LD/m)
0.12
< 15 < 2780/(LD/m)
0.12
< 15 < 2860/(LD/m)
0.12
< 15 < 3030/(LD/m)
0.12
< 15
with span-end Stockbridge dampers
5. Twin horizontal bundled conductors with non-damping < 1725 < 15 < 1925 < 15 < 2100 < 15 < 2450 < 15
spacers
6. Twin horizontal bundled conductors with non-damping < 2615/(LD/m)
0.12
< 15 < 2780/(LD/m)
0.12
< 15 < 2860/(LD/m)
0.12
< 13 < 3030/(LD/m)
0.12
< 6
spacers and span-end Stockbridge dampers < 2100 > 13 ; < 15 < 2450 > 6 ; < 15
7. Twin horizontal bundled conductors with damping < 1900 < 2200 < 2500 < 2500
spacers
8. Triple apex-down bundled conductors with non-damping < 1850 < 15 < 2100 < 15 < 2275 < 15 < 2500 < 15
spacers
9. Triple apex-down bundled conductors with non-damping < 2615/(LD/m)
0.12
< 15 < 2780/(LD/m)
0.12
< 10 < 2860/(LD/m)
0.12
< 7 < 3030/(LD/m)
0.12
< 5
spacers and span-end Stockbridge dampers < 2100 > 10 ; < 15 < 2275 > 7 ; < 15 < 2500 > 5 ; < 15
10. Triple apex-down bundled conductors with damping < 2500 < 2500 < 2500 < 2500
spacers
11. Quad horizontal bundled conductors with non-damping < 1850 < 15 < 2100 < 15 < 2275 < 15 < 2500 < 15
spacers
12. Quad horizontal bundled conductors with non-damping < 2615/(LD/m)
0.12
< 15 < 2780/(LD/m)
0.12
< 10 < 2860/(LD/m)
0.12
< 7 < 3030/(LD/m)
0.12
< 5
spacers and span-end Stockbridge dampers < 2100 > 10 ; < 15 < 2275 > 7 ; < 15 < 2500 > 5 ; < 15
13. Quad horizontal bundled conductors with damping < 2500 < 2500 < 2500 < 2500
spacers
Terrain category # 1 : Open, flat, no trees, no obstruction, with snow cover, or near/across large bodies of water or flat desert.
Terrain category # 2 : Open, flat, no obstruction, no snow; e.g. farmland without any obstruction, summer time.
Terrain category # 3 : Open, flat or undulating with very few obstacles, e.g. open grass of farmland with few trees, hedgerows and other barriers; prairie, tundra.
Terrain category # 4 : Built-up with some trees and buildings, e.g. residential suburbs; small towns; woodlands and shrubs. Small fields with bushes, trees and hedges.
H: initial horizontal tension; w: conductor weight per unit length; L: actual span length; D: conductor diameter and m: conductor mass per unit length.
Terrain Cat. #1 Terrain Cat. #2 Terrain Cat. #3 Terrain Cat. #4



33
Quad bundles (system #11) fitted with NDS should be somewhat less prone to Aeolian vibrations
than their triple equivalents. However, due to lack of accurate information from the field, quad
bundles were assigned the same safe design tensions as the triple bundles (system #8).

For all systems using NDS or NDS+Stk (systems #5, 6, 8, 9, 11 and 12), it was resolved, as a
matter of prudence, to limit the range of application of the given safe design tensions H/w to
spans for which LD/m < 15 m
3
/kg.

6.3.3 Bundled conductors fitted with non-damping spacers and span-end Stockbridge dampers
Again, the Task Force could not find any well-documented field test data related to bundles fitted
with NDS+Stk which could have been used to determine safe design tensions rigorously.
However, it appears reasonable and at one and the same time prudent to recommend the same
safe design tensions as for either single conductors fitted with Stk (system #2) or bundles fitted
with NDS alone (systems #5, 8 and 11), whichever are more permissive. That applies to twin
horizontal bundles (system #6), to triple apex-down bundles (system #9) and quad horizontal
bundles (system #12) as shown in Table 6.3.

6.3.4 Bundled conductors fitted with damping spacers
To determine safe design tensions of bundles fitted with DS, the Task Force could rely on the
results of an extensive program of tests [34] carried out at the IREQ test line in the Magdalen
Islands. The test program covered a horizontal twin, an apex-down triple and a horizontal quad
bundle of ACSR Bersfort conductors fitted with the same number of DS, staggered in a similar
manner, as well as an undamped single conductor of the same type. Each conductor system was
tested successively at a nominal tension H/w of 1550 m, 2325 m and 2870 m respectively.

The most relevant results are shown synthetically in Fig. 6.2 in terms of a relative effective
amplitude as a function of the tension parameter H/w. The effective amplitudes relate to a single,
integrated figure expressing the overall Aeolian vibration response of each system according to a
severity or fatigue-damage point of view. The effective amplitudes are then normalized by means
of the effective amplitude of the single conductor strung at H/w = 2325 m to yield the relative
effective amplitudes.

The resultant relative effective amplitudes are plotted as data points in Fig.6.2 which are then
related by straight line segments. There are two and three such points for the single conductor on
the one hand and each one of the twin, triple and quad bundles on the other hand. It may be
observed first that the relative effective amplitude of vibration is from 2 to 4, 6 to 9 and 8 to 13
times less severe on the twin, triple and quad bundles respectively, at the same tension, in
comparison to the undamped single conductor.
34
0 500 1000 1500 2000 2500 3000
Tension parameter H/w, (m)
0.0
0.2
0.4
0.6
0.8
1.0
R
e
l
a
t
i
v
e

e
f
f
e
c
t
i
v
e

a
m
p
l
i
t
u
d
e
B
C
D
A
U
n
d
a
m
p
e
d

S
i
n
g
l
e
Damped Quad
D
a
m
p
e
d
T
rip
le
±
±
±
±
±
±
±
±
±
± ±
±
D
a
m
p
e
d

T
w
i
n
1 2 3 4
1
2
3
4
1
2
3
4

Figure 6.2 Determination of safe design tension for twin horizontal bundled conductor lines
fitted with damping spacers on the basis of comparative field test results [34].



In order to proceed further, one would need at least one data point for the undamped single
conductor that would stand on the low side of the range of safe design tensions for such a
conductor, i.e. 1000 m < H/w < 1425 m. For lack of such a point from the test program, the
IREQ’s calculation model [36] was called upon to determine the expected response of the
undamped single conductor below data point B in Fig. 6.2, assuming a turbulence intensity
corresponding to terrain category 2, as the test site was classified.

Now, if line segment AB is accepted as a reasonable approximation of the actual single
conductor response in the range 1000 m < H/w < 1550 m, one can proceed as follows: for
instance, going along arrowed line 2 in Fig.6.2, it may be seen that the relative effective
amplitude of the undamped single conductor is expected to stand at about 0.27 when the tension
is set at H/w = 1125 m while for the twin bundled conductors, this amplitude is reached when the
tension is set at H/w = 2200 m. As a result, it looks reasonable to conclude that H/w = 2200 m
may be used as the safe design tension for the twin bundle fitted with DS where H/w = 1125 m is
considered safe for the undamped single conductor.

35
Strictly speaking, the above procedure for transposing the safe design tension of the undamped
single conductor to the twin bundle fitted with DS would be valid only for terrains category 2 as
for the test site. However, it was argued that, had the same tests been repeated in other terrains,
the effective amplitudes would have certainly moved up or down in absolute terms, but it seems
reasonable to assume that they would have moved everywhere in even proportion. As a result,
the respective locus for the undamped single conductor and each one of the damped conductor
bundles when shown in relative terms would be expected to be approximately unchanged.

To check that assumption to a certain extent, the IREQ computer model for undamped single
conductors was run at H/w = 1500 m, 2000 m and 2500 m respectively for each of the four
terrain categories. It could be confirmed that the locus of the predicted, effective amplitude for
each terrain category almost coincides when normalized appropriately. Besides, this result makes
sense since it is expected that the slope of the line expressing the effective response of any
conductor system as a function of the tension parameter H/w should be determined entirely on
structural grounds and not on aerodynamics grounds.

Thus, it appears that Fig. 6.2 could also be used to determine the safe design tensions of twin
bundles for the other terrain categories, according to the same transposition procedure. Doing so,
it came out that the safe design tensions H/w = 1000 m, 1225 m and 1425 m for the undamped
single conductor in terrain category #1, 3 and 4 respectively translate into safe design tensions
H/w = 1900 m, 2500 m and 3050 m respectively for the horizontal twin bundled conductors
(system #7 in Table 6.3) in the same terrain categories. However, in the latter case, the Task
Force resolved again to limit the safe H/w to 2500 m as a matter of prudence.

The above safe H/w figures are well supported by the results of the test on the twin bundle at
H/w = 2870 m at the test site, in terrain category 2, which indicated a safe condition.

As for the triple and the quad bundles (system #10 and #13), it is clear from Fig. 6.2 that their
response is much lower than that of the twin bundle. Using the same transposition procedure as
above would lead to quite high safe design tensions even for terrains category 1. Again, the Task
Force decided to limit their respective safe design tension to H/w = 2500 m for any terrain
category.

It should be noted that safe design tensions H/w for bundled conductors fitted with DS are
provided in a way independent of the span parameter LD/m as it was considered that the benefits
of damping are usually well distributed over the span length.

Comparison With Field Experience

Figures 6.3 and 6.4 depict the proposed safe design tension limits together with the
corresponding field cases in Table 6.2 for the twin horizontal bundles fitted with NDS or
NDS+Stk respectively. It will be seen that the proposed limits reflect well field experience.

Such is also the case for both triple apex-down and quad horizontal bundles fitted with either
NDS+Stk or DS, as shown in Fig. 6.5.

36
3
0 500 1000 1500 2000 2500 3000 3500
H/w, (m)
0
5
10
15
L

D

/

m
,

(
m

/
k
g
)
Field cases
Terrain #1
Terrain #2
Terrain #3
Terrain #4
Safe Design Zone
Special
Application
Zone

Figure 6.3 Ranking parameters of twin horizontal bundled lines in North America fitted with
non-damping spacers in relation to estimated safe boundaries.


Safe Design Zone
3
0 500 1000 1500 2000 2500 3000 3500
H/w, (m)
0
5
10
15
L

D

/

m
,

(
m

/
k
g
)
Field cases
Terrain #1
Terrain #2
Terrain #3
Terrain #4
Special
Application
Zone

Figure 6.4 Ranking parameters of twin horizontal bundled lines in North America fitted with
non-damping spacers and span-end Stockbridge dampers in relation to estimated safe boundaries.

37
3
0 500 1000 1500 2000 2500 3000 3500
H/w, (m)
0
5
10
15
L

D

/

m
,

(
m

/
k
g
)
Field cases
Terrain #1, NDS+Stk
Terrain #2, NDS+Stk
Terrain #3, NDS+Stk
Terrain #4, NDS+Stk
Any terrain, DS
Safe Design Zone
Special
Application
Zone

Figure 6.5 Ranking parameters of triple apex-down and quad horizontal bundled lines in North
America fitted either with non-damping spacers (NDS) plus span-end Stockbridge dampers (Stk)
or damping spacers (DS) in relation to estimated safe boundaries.



Task Force Recommendations

As explained above, the Task Force has resolved to provide guidance to conductor safe design
tension in terms of two parameters: the tension parameter H/w, the ratio of tension H to
conductor weight w per unit length and, as the case may be, the span parameter LD/m, the ratio
of actual span length L times conductor diameter D to conductor mass m per unit length. The
tension H refers to initial horizontal tension before any significant wind and ice load and before
creep, at the average temperature of the coldest month.

The Task Force recommendations regarding each one of several bundled conductor systems, in
addition to two single conductor systems, are summarized in Table 6.3 in the form of simple
algebraic expressions, each one associated to a particular terrain category described in the
legend. Terrains have been divided in four categories according to their general characteristics.
Should there be any doubt about real terrain category, the lowest should be selected.

For some systems, the safe design zones are defined in terms of the H/w parameter only in which
case they are unlimited in the LD/m parameter. However in all cases, they constitute zones where
full protection of conductors against Aeolian vibrations is certainly feasible by means of a
reliable system of non-damping spacers, combined or not with span-end Stockbridge dampers, or
38
alternatively, a reliable system of damping spacers. Hence, within the limits of these zones,
Aeolian vibrations should not be a constraint on design tension.

For line parameters falling outside of these zones, Aeolian vibrations may or may not be a
constraint and it is recommended that line designers determine the availability of adequate
protection before finalising the design.

In the context of this paper, non-damping spacers should be understood as spacers allowing a
certain relative mobility of the attachment points with the subconductors in the vertical direction.
Lack of such relative mobility may lead to harmful vibration trapping.

This guide applies to all round wire, concentric lay, overhead electrical conductors listed in
Table 5.3 supported in conventional metal-to metal clamps.

Use of armour rods or special supporting devices such as helical elastomer-bushed suspensions
may justify higher design tensions. Information on safe design tension, when these devices are
employed should be obtained from their suppliers.

Limitations and Warnings

The Task Force has recommended safe design boundaries within which satisfactory performance
should be available, often from multiple commercial sources. This statement should in no way be
taken that all commercially available spacing and/or damping systems will provide such
protection up to the limits of those boundaries or that they could provide such protection
indefinitely without themselves failing from loosening, fatigue or wear. Rather, appropriate field
testing or evaluation of available systems is likely to reveal at least some that do. It falls to the
line engineer to identify them.

The guidance provided herein should be suitable most of the time. However, special situations
require specific attention. Such is the case for extra long spans; for spans often covered with ice,
rime, or hoarfrost; for spans equipped with aircraft warning devices and for spans using non-
conventional conductors.

39
7. REFERENCES

[1] Zetterholm. O.D., “Bare Conductors and Mechanical Calculation of Overhead
Conductors”, CIGRÉ Session 1960, Report No. 223.
[2] Fricke, Jr., W.G. and Rawlins, C.B., “The Importance of Fretting in Vibration Failure of
Stranded Conductors”, IEEE Trans. on Power Apparatus & Systems, Vol. PAS-87, No. 6,
June 1968, pp. 1381-1384.
[3] CIGRÉ Study Committee 22, Working Group 04, “Recommendations for the Evaluation
of the Lifetime of Transmission Line Conductors”, Electra No. 63, March 1979, pp. 103-
145.
[4] Rawlins, C.B., “Wind Tunnel Measurement of the Power Imparted to a Model of a
Vibrating Conductor”, IEEE Trans. on Power Apparatus & Systems, Vol. PAS-102, No.
4, April 1983, pp. 963-971.
[5] Krispin, H.-J., "Parametric Studies: Vibration Intensity of Various Conductors as a
Function of Tension and Wind Turbulence", Report CIGRÉ SC22-WG11-TF4-95-6,
April 1995.
[6] Hardy, C. and Leblond A., "Estimated Maximum Safe H/w for Undamped Conductor
Spans", Report CIGRÉ SC22-WG11-TF4-95-13, September 1995.
[7] Rawlins, C.B., "Exploratory Calculations of the Predicted Fatigue Life of Two ACSR
and One AAAC", Report CIGRÉ SC22-WG11-TF4-96-5, April 1996.
[8] Claren, R., "A Contribution to Safe Design T/m Values", Report CIGRÉ SC22-WG11-
TF4-94-10, 1994.
[9] ESDU, “Across-Flow Response Due to Vortex Shedding”, Eng. Science Data Item
Number 78006, Oct. 1978, ESDU, London.
[10] Wieringa, “Updating the Davenport Roughness Classification”, J. Wind Engineering &
Industrial Aerodynamics, 41-44, 1992, pp. 357-368.
[11] CIGRÉ Study Committee 22, Working Group 01, “Report on Aeolian Vibration”, Electra
No. 124, May 1989, pp. 40-77.
[12] International Electrotechnical Commission, IEC 61089 Publication, “Round Wire
Concentric Lay Overhead Electrical Stranded Conductors”, June 1991.
[13] Rawlins, C.B., “Fatigue of Overhead Conductors”, Chapter 2 of “Wind-Induced
Conductor Motion”, EPRI Transmission Line Reference Book, Electric Power Research
Institute, Palo Alto, California, 1979.
[14] Rawlins, C.B., Greathouse, K.R. and Larson, R.E. "Conductor Vibration - A Study of
Field Experience", AIEE Paper No. CP61-1090, presented at the AIEE Fall General
Meeting, Detroit, MI, Oct. 15-20, 1961.
[15] Dulhunty, P.W., Lamprecht, A. and Roughan, J., "The Fatigue Life of Overhead Line
Conductors", CIGRÉ SC22-WG04 Task Force document, 1982.
[16] Doocey et al., "Wind-Induced Conductor Motion", EPRI Transmission Line Reference
Book, Electric Power Research Institute, Palo Alto, California, 1979, Chapter 3.
40
[17] CIGRÉ Study Committee 22, WG11, Task Force 1, "Modelling of Aeolian Vibrations of
Single Conductors : Assessment of The Technology". Electra No. 198, December 1998,
pp. 53-69.
[18] CIGRÉ Study Committee 22, Working Group 04, "Endurance Capability of Conductors",
Final Report, July 1988.
[19] CIGRÉ Study Committee 22, Working Group 11, Task Force 2, “Guide to Vibration
Measurements on Overhead Lines”, Electra No. 163, Dec. 1995.
[20] CIGRÉ Study Committee 22, “Guide on the Measurement of the Performance of Aeolian
Vibration Dampers for Single Conductors”, Électra No. 76, May 1981, pp.21-28.
[21] Tompkins, J.S., Merrill, L.L. and Jones, B. L., “Quantitative Relationship in Conductor
Vibration Damping”, AIEE Transactions on Power Apparatus and Systems, Vol. 75, Pt.
III, 1956, pp. 879-894.
[22] Van Dyke, P., Hardy, C., St-Louis, M. and Gardes, J.L., “Comparative Field Tests on
Various Practices for the Control of Wind-Induced Conductor Motion”, IEEE
Transactions on Power Delivery, Vol. 12, No. 2, April 1997.
[23] Bonneville Power Administration, Specification ETF 60-25.2G, October 1, 1987.
[24] Rawlins, C.B., Greathouse, K.R. and Larson, R.E., “Conductor Vibration – A Study of
Field Experience”, AIEE Paper No. CP 61-1090, presented at the AIEE Fall Meeting,
Detroit, MI, October 15-20, 1961.
[25] Leblond A. and Hardy, C., “IREQ Predicted Safe Design Tension”, Report CIGRÉ
SC22-WG11-TF4-99-6, April 1999.
[26] Rawlins C.B., “Safe Tensions with Dampers”, Report CIGRÉ SC22-WG11-TF4-99-5,
April 1999.
[27] Rawlins, C.B., “Analysis of Conductor Galloping Field Observations – Single
Conductors”, IEEE Trans. On Power Apparatus & Systems, Vol. PAS-100, No. 8,
August 1981, pp. 3744-3753.
[28] Hardy, C. and Van Dyke, P., “Field Observations on Wind-Induced Conductor Motions”,
Journal of Fluids and Structures, Vol. 9, 1995, pp. 43-60.
[29] Rawlins & J.R. Harvey, “Improved Systems for Recording Conductor Vibration” Trans.
of the AIEE, Transmission and Distribution Committee, Feb. 1960, pp. 1494-1501.
[30] W. Leibfried & H. Mors, “Die Bündelleiter-Vershuchsanlage Hornisgrinde der
Badenwerk AG, Karlsruhe”, 1964.
[31] A.J. Liberman & K.P. Krukov, Vibration of Overhead Line Conductors and Protection
Against it in the U.S.S.R.”, CIGRÉ General Session, 10-20 June 1968.
[32] A. T. Edwards & J. M. Boyd, “Bundle Conductor Design Requirements and
Development of ‘Spacer-Vibration Damper’”, IEEE Transactions on Power Apparatus &
Systems, Vol. PAS-84, No. 10, October 1965, pp. 924-932.
[33] G. Diana, M. Gasparetto, F. Tavano & U. Cosmai, “Field Measurements and Field Data
Processing on Conductor Vibration (Comparison Between Experimental and Analytical
Results)”, CIGRÉ General Session, 1-9 September 1982.
41
[34] C. Hardy, P. Jahier & S. Houle, “Field Evaluation of Aeolian Vibration of Bundled
Conductors”, Canadian Electrical Association Report No. 220T553, December 1990.
[35] P. Van Dyke, C. Hardy, M. St-Louis & J.L. Gardes, “Comparative Practices for the
Control of Wind-Induced Conductor Motion”, IEEE Trans. On Power Delivery, Vol. 12,
No. 2, April 1997.
[36] A. Leblond & C. Hardy, “Assessment of Safe Design Tension with Regard to Aeolian
Vibrations of Single Overhead Conductors”, Proc. of the ESMO 2000 Conference,
Montreal, 8-12 October 2000, pp. 202-208.
42

OVERHEAD CONDUCTOR SAFE DESIGN TENSION WITH RESPECT TO AEOLIAN VIBRATIONS
TASK FORCE B2.11.04

Members: Claude Hardy (Convenor) Hans-Jörg Krispin (Secretary) André Leblond, Charles Rawlins, Konstantin Papailiou, Louis Cloutier, Peter Dulhunty Corresponding Members: David Havard, Jean-Marie Asselin, Magnar Ervik, Tapani Seppä, Vladimir Shkatsov

Copyright © 2005 “Ownership of a CIGRE publication, whether in paper form or on electronic support only infers right of use for personal purposes. Are prohibited, except if explicitly agreed by CIGRE, total or partial reproduction of the publication for use other than personal and transfer to a third party; hence circulation on any intranet or other company network is forbidden”. Disclaimer notice “CIGRE gives no warranty or assurance about the contents of this publication, nor does it accept any responsibility, as to the accuracy or exhaustiveness of the information. All implied warranties and conditions are excluded to the maximum extent permitted by law”.

TABLE OF CONTENTS................................................................................................................ 2 EXECUTIVE SUMMARY ............................................................................................................ 3 1. 2. 3. 4. INTRODUCTION .................................................................................................................. 5 REVIEW OF THE EVERY DAY STRESS (EDS) CONCEPT ............................................ 6 TURBULENCE AS A FUNCTION OF TERRAIN .............................................................. 8 SINGLE UNPROTECTED CONDUCTORS ...................................................................... 10 4.1 4.2 4.3 4.4 5. 5.1 5.2 5.3 5.4 5.5 5.6 6. 6.1 6.2 6.3 Modelling...................................................................................................................... 10 Comparison With Field Experience.............................................................................. 14 General Discussion ....................................................................................................... 16 Recommendations......................................................................................................... 18 Modelling...................................................................................................................... 20 Selection of Parameters................................................................................................. 21 Predicted Safe Design Tension ..................................................................................... 22 Comparaison With Field Experience ............................................................................ 23 Task Force Recommendations...................................................................................... 26 Limitations .................................................................................................................... 28 Review of Literature ..................................................................................................... 29 Review of Field Experience.......................................................................................... 30 Determination of Safe Design Tension......................................................................... 32

DAMPED SINGLE CONDUCTORS .................................................................................. 20

BUNDLED CONDUCTORS ............................................................................................... 29

6.3.1 Unspacered bundled conductors fitted or not with span-end Stockbridge dampers... 32 6.3.2 Bundled conductors fitted with non-damping spacers................................................ 32 6.3.3 Bundled conductors fitted with non-damping spacers and span-end Stockbridge dampers ................................................................................................................................. 34 6.3.4 Bundled conductors fitted with damping spacers ....................................................... 34 6.4 Comparison With Field Experience.............................................................................. 36 6.5 6.6 7. Task Force Recommendations...................................................................................... 38 Limitations and Warnings............................................................................................. 39

REFERENCES ..................................................................................................................... 40

2

EXECUTIVE SUMMARY
There may be several motives for controlling overhead conductor tension at the design stage. One of the most obvious reasons is to ensure that maximum tension resulting from the assumed most severe climatic loading does not exceed the conductor allowable tension. At the opposite end, it may be required to limit minimum tension while the conductor is operating at maximum temperature so that line clearance is not violated. A third motive, which should not be disregarded, is to restrict conductor susceptibility to harmful conductor vibrations. As a matter of fact, it is well known that stranded conductors get more vulnerable to Aeolian vibrations as tension is increased. This is true for all conductor systems whether they are used in solo or in bundles and whether they are fitted or not with damping and/or spacing devices. Therefore, there is a need to set an upper limit to conductor unloaded tension that may prevail for a significant period of time. This brochure deals with this important issue. It starts in Section 2 with a critical examination of the EDS concept which was put forward in 1960 by CIGRÉ SC6 with the intent to provide guidance on such conductor safe design tensions with respect to Aeolian vibrations. It is noted, for example, that the 18% EDS value which was proposed as safe for ACSR conductors did lead, in spite of all, to fatigue failures in a significant number of cases, thus motivating the present work further. In this conjuncture, one of the first and most important tasks of the Task Force was to question the relevance of the EDS (% of the Ultimate Tensile Strength) as conductor tension parameter. On account of its improved ability to rate conductor self-damping capacity, on the one hand, and to generalise results as widely as possible, on the other hand, it was resolved to adopt parameter H/w, the ratio between the initial horizontal tensile load H and conductor weight w per unit length. As stated in the brochure, this tension refers to initial horizontal tension before any significant wind and ice loading and before creep, at the average temperature of the coldest month on the site of the line. Another parameter of prime importance is wind turbulence since it affects to a great extent the Aeolian power imparted to vibrating conductors. This is scrutinised in Section 3 of the brochure insofar as it relates to roughness of the terrain crossed by the lines. Single unarmored, unprotected single conductors of the most common types are considered in Section 4 of the brochure. Then comes the description of analytical models from four different organisations that were used to draw tentative recommendations about safe design tensions of unarmored, unprotected single conductors. These recommendations were finally ratified on account of a successful “reality check” against available field experience. Section 5 of the brochure deals with single conductor lines protected by means of Stockbridgetype vibration dampers set up at span extremities. Addition of dampers obviously calls for the introduction of another parameter intended at rating the protective capacities of the damping system. As justified in this section, the rating parameter that was selected is LD/m, the ratio of the product of span length L and conductor diameter D to conductor mass m per unit length.

3

The Task Force recommendations are summarised in a table at the end of Section 6 in the form of simple algebraic expressions. on bundled conductors set up in parallel and at the same tension as identical single conductors. the table also incorporates the Task Force recommendations for single conductors fitted or not with Stockbridge-type dampers set up at the span extremities. which are then compared with all available test line and field experience. relying on field experience and full scale test line data. triple apex-down bundles and quad horizontal bundles made up of conventional stranded conductors fitted either with damping spacers or non-damping spacers or a combination of non-damping spacers and span-end Stockbridge-type dampers. In this case. 4 . Finally. It then covers a review of field experience gathered on 91 bundled conductor lines erected in North America. This part of the brochure concludes on the Task Force recommendations which are provided both graphically and algebraically. For sake of completeness. particularly. This is followed by a description of the methodologies that were used to arrive at safe design tension for each one of a number of bundled conductor systems. again as a “reality check”. mainly on dedicated test lines. twin horizontal bundles. It starts with a thorough review of literature about field tests carried out in the past. Section 5 goes on with a description of the two analytical models that were used to predict safe design tensions.Similarly to what was done in the previous section. each one associated to a specific conductor system and to one out of four terrain categories. these methodologies are purely empirical. Section 6 of the brochure deals with bundled conductor lines.

Accordingly. A theoretical model should be used that translates energy balance into amplitude and number of cycles of vibration at the clamps to assess vibration severity. Section 6 of the brochure deals in turn with bundled conductor lines. particularly twin horizontal bundles. It starts with a review of the Every Day Stress (EDS) concept. as discussed in the following section of this brochure. unarmored. Hence. working under appointment of CIGRE SC6 (since disbanded). it may be required to limit tension while the conductor is operating at maximum temperature so that line clearance is not violated. 5 . As a matter of fact. triple apex-down bundles and quad horizontal bundles made up of conventional stranded conductors fitted either with damping spacers or non-damping spacers or a combination of non-damping spacers and span-end Stockbridge-type spacers. One of the most obvious reasons is to ensure that maximum tension corresponding to the assumed most severe climatic loading does not exceed a predefined. INTRODUCTION There may be several motives for controlling overhead line conductor tension at the design stage. This brochure deals with this very important issue in the design and operation of overhead transmission lines. It was a premise for the Task Force that it would be quite difficult to work out a revised guide on the basis of field experience only due to the rarity of well documented cases. At the opposite end. acceptable conductor tension. which is known to affect to a large extent wind power imparted to conductors. This guide should be based on existing experience and current knowledge on wind energy input. This is true for all conductor systems whether they are used in solo or in bundles and whether they are fitted or not with damping and/or spacing devices.1. A third motive is to restrict conductor susceptibility to harmful Aeolian vibrations. it was felt timely to revise the Panel’s recommendations. However. there is a need to set an upper limit to conductor unloaded tension that prevails for significant periods of time. Single. more particularly in 1960 by the so-called EDS (Every Day Stress) Panel [1]. the Task Force’s terms of reference have been: To produce a practical guide for the selection of safe design tensions in form of generic values for specific classes of single transmission line conductors to avoid damage due to Aeolian vibrations during a typical design lifetime. Guidance to such safe design tension with respect to Aeolian vibrations was provided before. it is well known that stranded conductors get more vulnerable to Aeolian vibrations as tension is increased. The following section deals with single conductor lines protected by means of Stockbridge-type vibration dampers set up at span extremities. That is followed in Section 3 by a review of the influence of terrain cover or roughness on wind turbulence. conductor self-damping and conductor fatigue endurance. unprotected single conductors of the most common types are considered in Section 4.

With the broader experience. 6 . at the temperature which will occur for the longest period of time.2. In that report the EDS. Table 2. Lines equipped with Unprotected lines Armour rods Dampers Armour rods and dampers 26 18 22 24 24 17 18 26 11 13 Copper conductors ACSR Aluminium conductors Aldrey conductors Steel conductors 1. For example. Different values of EDS were given for bare conductors and for conductors with armour rods only. at the CIGRÉ 1960 session. Table 2. the 18% EDS value proposed as “safe” lead to fatigue failures although at a slower rate. even the statistical conclusions drawn from EDS Panel do not seem quite correct. new conductors have been put in service and lines have been built for the first time in new areas. the recommended EDS values have appeared to be insufficient to explain the fatigue damage found on these more recent lines. of Report 223 [1] of CIGRÉ SC6. Oscillating clamps Since Report 223 was issued. Rigid clamps 2. dampers only and armour rods and dampers. REVIEW OF THE EVERY DAY STRESS (EDS) CONCEPT Forty-five years have elapsed since the presentation. As can be seen. It indicates that for lines that have been in service for 10 to 20 years.1. In fact. 45% of the lines with an EDS<18% and 78% of the lines with an EDS>18% revealed damages. expressed as a percent of the conductor breaking strength (or UTS. The Panel’s recommendations are shown in Table 2. without any risk of damage due to Aeolian vibrations. in retrospect. was defined as the maximum tensile load to which a conductor can be subjected.1 EDS panel recommendations for safe design tensions in percent UTS. Ultimate Tensile Strength). which showed the conclusions reached by the EDS Panel appointed in 1953 to investigate the fatigue damages caused to transmission line conductors by the Aeolian vibrations.2 shows the results of an analysis of the EDS Panel data for bare ACSR conductors.

the occurrence and direction of these winds is only related to general meteorological conditions but in other locations. The EDS Panel classified the lines on the basis of terrain.67 Research and testing performed since 1960 have provided information that was not available to the EDS Panel. hilly.00 78. it is well known that conductor fatigue is the result of an accumulation of dynamic bending strains in the presence of fretting [2-3].93 35. It follows that a parameter often ignored is the occurrence of dangerous winds on the line.). Finally. but it is now well known that what influences the wind power is the surface roughness of the ground which creates turbulence. of present similar lines in service. With such cables. it is now considered more convenient.00 58. mountainous. To represent such effect. and of improved current understanding of Aeolian vibration. close to large water bodies or in desert areas. In some locales. This may not have been evident to the EDS Panel because the vast majority of the lines up to 1962 were built with the classical 30/7. the thermal differences in areas surrounding the line cause daily air flows in one direction and in the other.29 45. It should be pointed out that the Panel offered its recommendations with clear reservations regarding their fundamental soundness.00 20. to use the ratio H/w between the horizontal tensile load and the conductor weight per unit length rather than the % of UTS. etc. 7 .Table 2. All this information can now better explain the dispersion of the time to failure of the lines investigated by the EDS Panel and. the increase of UTS due to an increase of the conductor diameter brought also an equal increase in the conductor weight. Self-damping tests on conductors showed that the increased tensile load reduces the power dissipated by a vibrating conductor. Under such conditions the rate of dynamic stress accumulation can be significantly greater than in other areas.26 25. in general.2 Summary of damaged lines as per the EDS panel Service life Years <=5 > 5 < = 10 > 10 < = 20 > 20 % of lines damaged EDS < 18% EDS = > 18% 5. (flat. in order to generalise the results. Much poor experience that resulted from trying to apply the recommendations was the result of ignoring those reservations. 26/7 and 54/7 stranding ACSR.93 91. In view of sometimes unsatisfactory experience with the Panel’s recommendations. it is felt that new proposals are appropriate.

and even blades of grass.also reduce wind power input. obstacles Low-density built-up areas. Turbulence intensity at any particular field location is strongly influenced by the local terrain. The turbulence that influences Aeolian vibration arises from the interaction of the mean wind with the ground.3. open country with few. the conductor’s vibration frequency. Table 3. For example. Obstacles to wind flow.1 Typical values of turbulence intensity. there must be parts of the span where maximum power transfer does not occur.18 0. they shape the flow to conform to these gross orographic features. such as trees and buildings. Fluctuations in wind speed with time . small town. For example. Naturally. Both the span-wise variations and the gustiness are reflections of turbulence in the wind structure. result in low intensity. Tables with more finely-divided classifications are available. large obstacles shed large vortices and result in large turbulence intensity. open woodland with small trees Town and city centers with high density of buildings. increasing its speed and actually reducing its turbulence intensity. or frequency spectrum.35 8 . Very large obstacles such as hills. Wieringa [10] provides twice as many. and especially the nature of the ground cover. because the vibration of the span cannot change frequency fast enough to follow short-term wind speed variations. while small obstacles. ridges and mountains do not cause turbulence as understood here.gustiness . The values in Table 3. The magnitude of the turbulence is ordinarily taken as the root-mean-square variation of wind velocity about the mean speed. broken country with tall trees Turbulence Intensity 0.1 pertain to elevation above ground of 10 metres [9]. Due to the travelling-wave nature of conductor vibration. large stretches of open water Rural areas. shed vortices somewhat as conductors do. These vortices comprise the turbulent component of the wind. Rather. and total wind power input must be less than would occur in perfectly uniform wind. is the same all along the span. If the wind speed is not constant all along the span. low. and turbulence intensity is expressed as the ratio of that RMS variation to the mean wind velocity. with their small and short-lived vortices. TURBULENCE AS A FUNCTION OF TERRAIN Turbulence in the wind affects the amount of Aeolian vibration power that is imparted to a vibrating conductor. Terrain Open sea. valleys may funnel the wind. Maximum power is transferred when wind causes a Strouhal frequency of vortex shedding that is close to the frequency at which the conductor is vibrating [4].11 0.25 0. suburbs. Field measurements have yielded tables of typical values of intensity found in various classes of ground cover.

This dispersion results from effects of the buoyancy acquired by parcels of air that are in contact with the surface when they are heated or cooled by the ground. Because of these effects. Heating causes these parcels to rise. show a great deal of dispersion. such measurements show consistent results only during strong winds. in fact.The intensities given in these tables are typical of values measured during strong winds and. depending upon whether the ground surface is warmer or colder than the air. For example. 9 . intensity can be as high as 0. or significantly smaller than the values reported in the table. resulting in increased turbulence. say up to 8 or 10 m/s (hourly average). Measurements during light to moderate winds. on the other hand. churning the atmosphere as they do.50 or as low as 0. Cooling. causes the atmosphere to stratify. in rural areas.07. turbulence intensity during the light-to-moderate winds associated with Aeolian vibration can be much larger. with the cool layer sticking to the ground and blocking the movement of surface-generated turbulence upward.

may serve to describe the properties of the structural system (the span) and the wind excitation.1) shows parameters that. The flow chart (Fig. Conductor tension design guide . Conductor tensions that lead to vibration levels below the endurance limit are regarded as safe. . 4. : Tol. and Table 4. Modelling All models that were employed to estimate safe tensions basically rely on the Energy Balance Principle [11] to predict steady state vibration in terms of fymax (the product of vibration frequency and maximum vibration amplitude at antinodes).1 Parameters describing undamped structural system and wind excitation. if they do not exceed a definite limit value (the endurance limit in terms of fymax). and how they were quantified. in a modelling approach. Two approaches were followed to assess computed vibration levels (the span response) with regard to the tolerance of the conductor to vibration.4. unprotected single conductor lines. 10 . In the Endurance Limit Approach. limiting design value Resp.K.1 summarizes which of these parameters were considered in models of four different organizations. Tension Energy balance Wind statistics Terrain category Effective wind turbulence Effective wind recurrence EXCITATION Wind power input Figure 4. SINGLE UNPROTECTED CONDUCTORS This section aims at recommending safe design tensions for unarmored. vibration levels are considered to result in an infinite lifetime of the conductor.modelling approach Increase or decrease tension STRUCTURAL SYSTEM Clamping system Conductor type Conductor stranding Conductor condition Span length Conductor size < Tolerance to vibrations Damping capacity Response Intensity Recurence > = O. that is the response of a span when excitation by the wind is balanced by the internal damping of the span.

Organisation Approach Conductor damping capacity RIBE/Krispin [5] Endurance limit Measured (decay method. translated in terms of H/w. Safe conductor tensions that are calculated on that ground necessarily pertain to a certain predicted fatigue life of the conductor. Safe design tensions as predicted by the different organizations are depicted in Figs. Also shown in the figures are the tensions recommended by the EDS Panel. The usual assumption is that of linear damage accumulation (Miner's Rule). and different S-N-curves may be surmised for different levels of probability of survival. span with pivoted ends) IREQ/Hardy. 4. and data on the probability of vibration exciting wind has to be introduced into the model. These small fractions of fatigue damage are assumed to accumulate at a certain rate during the service life of the conductor.057) between 7 and 12 mph. for different conductor types. Proper winds to occur half of the time. Probabilistic considerations may be expanded to the S-N-curves that define the fatigue inducing intensity of different vibration levels.Table 4. maximum amplitude limited to 3. their recommended tensions appear as flat horizontal lines in the figures. one assigns a certain portion of fatigue damage to each vibration cycle. Figures 4.3 as a function of wind intensity of turbulence. linearly decreasing from 12 to 27 mph. until fatigue breakage occurs. constant probability density (0. As the Panel did not account for the effect of turbulence.1 Basic assumptions underlying model calculations.5 km/h avrg) Alcoa Fujikura/Rawlins [7] Endurance limit Cumulative damage Measured (ISWR method) Claren [8] Cumulative damage Calculated on the basis of equations that were derived from measured data Laminar & reduced for turbulence Rayleigh-distributed (2.3a & b were prepared on the basis of the Cumulative Damage Approach for multi-layer A1/Syz (ACSR) and A1 (AAC) conductor classes [12] as well as for multi-layer A3 (AAAC) and A1/A3 (ACAR) conductor classes respectively. 4.5 m/s avrg) Wind power input Wind recurrence Laminar & reduced for turbulence − Laminar & reduced for turbulence − Linearly increasing probability density from 0 to 7 mph. This approach requires assumptions on the recurrence of fatigue inducing stress levels to determine the number of occurrences at different stress levels. Leblond [6] Endurance limit Cumulative damage Measured (ISWR method) and Calculated on the basis of similarity laws calibrated by means of measured data Laminar & reduced for normally-distributed turbulence Rayleigh-distributed − (4. 11 .5 times that mean value) CIGRÉ Safe Border Line Fixed clamps 30 years Fixed metal to metal clamps Infinite 100 years Infinite In the Cumulative Damage Approach.4 to 1 times the EBP-amplitude) Vibration mode shape Sinusoidal Narrow band random vibration (peak amplitudes Rayleigh-distributed.2a & b were worked out on the basis of the Endurance Limit Approach while Figs. maximum amplitude limited to 6 times rms-value at each frequency) Sinusoidal Conductor tolerance to vibration Clamping system Predicted fatigue life Fatigue endurance as per [13] Fixed metal to metal clamps Infinite Fatigue endurance as per [13] S/N-curves based on [13] data for several probabilities of survival Fixed metal to metal clamps 50 years with a probability of survival of 95% Fatigue endurance as per [13] S/N-curves based on [13] data Random (amplitudes normally distributed about a mean corresponding to 50% of the maximum laminar amplitude at respective frequency.2 and 4. Random (Flat probability density from 0.

(a) Multi-layer A1 (AAC) and A1/Syz (ACSR) conductors. 12 .2 Predicted conductor safe design tension according to endurance limit approach.Figure 4. (b) Multi-layer A3 (AAAC-6201) and A1/A3 (ACAR) conductors.

(a) Multi-layer A1 (AAC) and A1/Syz (ACSR) conductors. 13 .3 Predicted conductor safe design tension according to cumulative damage approach. (b) Multi-layer A3 (AAAC-6201) and A1/A3 (ACAR) conductors.Figure 4.

For example. This is the result of their calculations being performed on the basis of lower fatigue endurance characteristics for the A3 conductors as per reference [13]. as long as each collection is confined to a certain limited class. These analyses rely upon evidence that there exist parameters that can rank spans according to their risk of fatigue damage. since it was first concerned only with undamped. relative to their design tensions and the occurrence or not of fatigued strands. The Task Force was faced with the need to 14 . However. such as force. was Energy Balance Analysis.” In doing that. the most significant design parameter influencing the probability of fatigue is conductor tension. Dulhunty et al [15] and Doocey et al [16]. particularly at the lowest intensities of turbulence. and that well-chosen ranking parameters can properly rank collections of spans of different design. with dampers or with armor rods. However. It will be noticed that despite a considerable scatter in predicted safe tensions according to organizations. the method adopted by the Task Force. percent of rated strength and others. ACSR spans with dampers would belong to a different class from ACSR spans with armor rods. it may be observed that recourse to the cumulative damage approach leads to more permissive H/w than the alternative approach. tension can be expressed in various forms. calculated safe design tensions are seen to increase with intensity of turbulence pointing out that conductors are less susceptible to Aeolian vibration excitation by turbulent winds. defining the various classes narrowly enough. The power of such analyses hinges upon choosing the ranking parameters well. it appears to result mainly from differences in predicted conductor self-damping. collections have been made of historical data on existing lines. and copper conductor spans. as a “reality check. Analyses of such collections have been published by Zetterholm [1]. For that purpose. because of the impact of tension upon conductor self-damping. it also made use of the field experience approach.On the contrary. As for the scatter. stress. would belong to different classes from the ACSR spans. That stands to reason on account of the former approach allowing for a certain number of vibration cycles above the conductor endurance limit while the latter does not. Three organizations computed admissible H/w for A3 (AAAC-6201) all aluminum alloy conductors that are from 30 to 50% lower than those for A1 (AAC) all aluminum conductors or A1/Syz (ACSR) steel reinforced aluminum conductors. As noted earlier. Lastly. Rawlins et al [14]. Comparison With Field Experience Uncertainties in the data and assumptions that are required to determine maximum safe tensions on the basis of the Energy Balance Principle have made it necessary to consider known experience with existing lines when specifying protection for new lines. unarmored spans. to estimate maximum safe tensions for Aeolian vibration. they generally stand well below the corresponding EDS values. and yet retaining enough cases in each class to define a pattern that distinguishes spans that are likely to experience fatigue from the general population. it was the view of the Task Force that.

i. and a limiting value of H/w below which failures do not occur is expected to exist. The tension-ranking parameter H/w is expected to correlate with failure experience. the departure of this case from the main pattern may indicate that pin insulator supports cannot be included in the same class as suspension clamp supports. and for each line it is indicated whether conductor fatigue is known to have occurred. spans equipped with neither dampers nor armor of any type. the safe limit may be as much as H/w = 1400 meters. Spans having H/w values greater than this limit may survive without experiencing fatigue failure due to the protective effects of terrain features such as trees and buildings. no failures should occur where H/w is below the limit. The Task Force drew from several sources to form a collection of field experience cases for undamped spans. for use as a ranking parameter that would be indifferent to such things as conductor diameter. there are not enough experience cases in the vicinity of that value to support more than a tentative conclusion. i. it was necessary to define the reference condition for determining H.2. The ACSR collection is shown in Table 4. The lowest failure case. Thus. involved pin-type insulators with hand-formed ties. or where the line runs parallel to a valley that channels the wind.. They are arranged in order of increasing H/w.e. However. One can only conclude it is somewhere in this interval. developed by Task Force members. However. but varies with temperature. The table suggests that the limit may be about H/w = 1000 meters. The Task Force decided to use the average temperature for the coldest month as reference temperature. ice or wind loading history and creep. The parameter that was selected was H/w. before wind and ice loading and creep. There were not enough cases to indicate safe tension limits for aluminum or aluminum alloy conductors. This collection was used as a “reality check” on the predictions of maximum safe design tensions based on the Energy Balance Principle.. Since tension H for any span is not constant.e. 15 . The collection contained enough cases to be useful only for conventional round strand ACSR. In addition. the individual rows pertain to existing undamped ACSR lines for which data is available. in order gather as many field cases into each class as possible. In Table 4. there are not enough experience cases in the interval between 1000 and 1400 m to determine where in this range the limit should fall. with H/w = 1030 meters.choose a form. If this case is ignored.2. whereas almost all the other cases involved spans supported by suspension clamps. making the determination of tension particularly sensitive to the precision of “sagging in” and to the choice of reference temperature. the span was short. ground wires or OPGW. However. and to determine tension for initial conditions.

0 21.6 26.2 Field Experience Cases for Undamped ACSR. tension history and even.3 31.3 Al / St Strands 36/12 36/12 54/7 6/1 26/7 26/19 30/7 30/7 12/7 26/7 30/7 26/7 26/7 54/7 26/7 30/7 30/7 6/1 30/7 48/7 30/7 30/7 12/7 30/7 30/7 32/19 32/19 32/19 34/19 26/7 32/19 36/12 26/7 114/37 114/37 114/37 12/7 30/7 Average Span (m) 200 395 137 61 183 362 310 396 350 300 274 396 326 300 346 170 333 390 107 340 295 270 360 264 350 350 655 445 520 475 580 475 315 440 425 415 374 194 350 H/w (m) 707 844 934 1029 1358 1397 1405 1511 1554 1607 1638 1655 1723 1731 1735 1738 1747 1761 1772 1865 1881 1908 1959 1996 2001 2001 2095 2108 2116 2154 2162 2176 2243 2279 2297 2300 2382 2458 2861 Fatigue failure Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes General Discussion Attempts to determine maximum safe tensions for Aeolian vibration on the basis of field experience only have been difficult and sometimes unsuccessful.4 21.8 20.8 26.Table 4.7 22.7 31.7 31.4 25.4 22. This is due to the fact that actual conductor condition. However.4 21.9 18.7 21.7 19.8 10.5 18.7 14. design tensions are not known.8 18.2 8.9 21.1 26.7 18.6 16.9 22. One substantial advantage of this approach is the possibility to consider all relevant parameters and variables in a systematic manner.4 22. One way round is to have recourse to modelling which is the method adopted by the Task Force.8 25.0 12.4 26.4 26.8 18.4 19.9 24. on occasions.9 21.6 19.4 21.8 11. Vibration-induced fatigue degradation of conductors is a problem of a highly complex 16 .5 25. caution is compulsory despite the improved scientific appearance of the modelling approach. Conductor Diam (mm) 21.

the wind. and even different hours of the same day. initial tension at the average temperature of the coldest month was chosen as a base to the recommendations knowing that any temperature increase. may be scattered over a range as large as 1-50. the sole cause to Aeolian vibrations. thus alleviating the vibration 17 . which determines the vibration intensity. All those complexities and uncertainties just enhanced the necessity of resorting to sound analysis and judgement. It has been determined that lifetime before first strand failure of several samples of the same conductor. any wind and ice loading as well as the inevitable creep would soon reduce tension permanently. They respond to differences in wind structure by vibrating less when and where the wind is turbulent or gusty. Moreover. What is more. is another random variable that comes into play. The conductors respond to variations in weather conditions associated with the passage of major weather systems by displaying markedly different amplitudes and frequencies on different days. And comparison against field experience and test line results is a necessity to gain confidence in the recommended safe tensions. From the very first. the conductor faculty to resist vibratory motion without strand failure.and highly random nature that is not easily amenable to mathematical analysis. there is a 2 to 1 ratio in the endurance limit of multilayer ACSR’s whether it is extracted from EPRI’s Orange Book [13] or from the CIGRÉ Safe Border Line [18]. Thus. effects of fretting. and is increased by effects of variability in mill processing and. Some conservatism is needed to counterbalance uncertainties. simplification is required to overcome complexities. self-damping is even dependent on the share of tension between the aluminium strands and the steel strands and that is again a function of temperature and creep. Obviously. Likewise. That entire scatter would probably intensify had the true nature of the beat-pattern-like vibrations been duly considered. with the wind speed and direction fluctuating continuously both time-wise and space-wise. is itself strongly random. the rate of accumulation of fatigue cycles is random over time and varies markedly with location. is a strong function of tension which itself is far from being a constant depending on non-deterministic variables such as conductor temperature and also creep which in turn depends on wind and ice loading history. pertaining to the same simplified vibratory schemes. there is a 2 to 1 factor between the highest and the lowest value of maximum wind power input into conductors according to investigators [17]. Conductor fatigue endurance. Hence. sizes and makes as well as in suspension clamps or other supporting devices. particularly. For ACSR conductors. there is already a large scatter in the available experimental data yet. That has been the philosophy that has guided the Task Force throughout. Dispersion in the cycles to failure is indeed inherent to metal fatigue. as noted above. there may be one order of magnitude difference in conductor self-damping estimate at the same tension according to testing methods [17]. tested according to the same ideal laboratory protocol. All that without saying anything about the great diversity in conductor types. conductor self-damping. Hence. What is even more consequential. as regard conductor fatigue.

Should there be any doubt about real terrain category. Terrains have been divided into four categories according to general characteristics. all aluminium alloy A2 or A3 (AAAC) conductors.g.3 as a function of terrain category. e. Although a lower fatigue endurance of A2 (AAAC) conductors may be surmised from reference [13]. ( H w) adm (m) 1000 1125 1225 4 1425 18 . summer time. Open. Recommendations The maximum safe design tensions with respect to Aeolian vibrations of undamped and unarmored conductors. The table uses H/w. or undulating with very few obstacles. Recommended safe tensions apply to the following round strand conductors: all aluminium A1 (AAC) conductors. tundra. with snow cover. no trees. as the tension parameter. Small fields with bushes. small towns. or near/across large bodies of water. residential suburbs. flat desert. trees and hedges. the lowest class should be selected. farmland without any obstruction. Terrain category 1 2 3 Terrain characteristics Open.3 Recommended conductor safe design tension at the average temperature of the coldest month as a function of terrain category. no obstruction. bi-metallic aluminium conductors Ax/Ay (ACAR) and steel reinforced aluminium conductors A1/Syz (ACSR). aluminium/aluminium alloy A1/A2 or A1/A3 (ACAR) conductors and steel-reinforced aluminium A1/Syz (ACSR) conductors. no snow. Valid for homogeneous aluminium and aluminium alloy conductors Ax (AAC and AAAC). woodlands and shrubs. flat. e. prairie. open grass or farmland with few trees.g. no obstruction. hedgerows and other barriers. the ratio of horizontal tension in the span to conductor weight per unit length. Built-up with some trees and buildings. the recommended safe tensions stand well in the lower range of the predicted values. Furthermore. as recommended by the Task Force.g. The recommended safe H/w may appear over-conservative.severity. e. flat. Nevertheless. before wind and ice loading and before creep. It was resolved to give a uniform recommendation for all types of conventional conductors using aluminium and/or aluminium alloy. It is important to note that this horizontal tension refers to initial horizontal tension. Open. Table 4. there seems to be no well documented field evidence to support a more pessimistic tension recommendation for these conductors. are shown in Table 4. flat. at the average temperature of the coldest month on the site of the line. it should be noticed that they generally stand above the 17-18% of UTS recommended by the EDS Panel for all aluminum A1 (AAC) conductors and low steel-content aluminum conductors A1/Syz (ACSR).

19 . Techniques to perform such vibration measurements have been described previously [19]. use of damping may allow higher tensions resulting in significant cost savings in line construction. or spans often covered with ice. when these devices are employed.The maximum safe design tensions recommended herein should be suitable most of the time. Use of armour rods or special supporting devices such as cushioned clamps and helical elastomer-bushed suspensions may justify higher design tensions on otherwise unprotected conductors. or spans exposed to pollutants that may decrease the self-damping or the fatigue endurance of the conductor. Incidentally. Generally speaking. Such is the case for extra long spans. damping spans is inexpensive and that is certainly preferable to hazarding conductor fatigue breaks. in some countries. the maximum safe design tension may be governed by the maximum climatic loading rather than by Aeolian vibrations. Information on safe tension. or spans operated at high temperature. should be obtained from their suppliers. Moreover. Existing lines using single unprotected conductors strung at a tension exceeding the recommended value for the terrain may require inspection and field measurement. However. special situations require specific attention. rime or hoarfrost.

and Table 5. Alcoa Fujikura / Rawlins [26] Endurance Limit Measured (ISWR method) Span-end damping Travelling wave approach using Measured efficiency of spancomplex damper stiffness measured end damping arrangement on shaker Laminar or reduced for normallydistributed turbulence Narrow band random vibration (peak amplitudes Rayleighdistributed.1 summarizes which of these parameters were considered in models of different organisations.5 times RMS-value at each frequency) Fatigue endurance as per [13] Fixed metal to metal clamps Laminar or reduced for turbulence Sinusoidal Wind power input Vibration mode shape Conductor tolerance to vibration Clamping system Fatigue endurance as per [13] Fixed metal to metal clamps 20 . that is the response of a span when excitation by the wind is balanced by the self-damping in the conductor and by damping devices.5.1 Basic assumptions underlying model calculations. Organisation Approach Conductor damping capacity IREQ / Leblond & Hardy [25] Endurance Limit Calculated on the basis of similarity laws calibrated by means of measured data (ISWR method). Table 5. in a modelling approach. maximum amplitude limited to 3. The flow chart in Fig. Modelling The two models that were employed to estimate safe tensions basically rely on the Energy Balance Principle [11] to predict steady state vibration in terms of f ymax (the product of vibration frequency and maximum vibration amplitudes at antinodes).1 shows parameters that. and how they were quantified. may serve to describe the properties of the structural system (the span and span-end damping arrangements) and the wind excitation. 4. DAMPED SINGLE CONDUCTORS This section aims at recommending safe design tensions for single conductor lines protected by means of Stockbridge-type vibration dampers set up at the span extremities.

The parameter is proportional to the damping efficiency required to control vibration amplitude to a given level. The Task Force was aided. if they do not exceed a definite limit value (the endurance limit in terms of f ymax). Conductor tensions that lead to vibration levels below the endurance limit are regarded as safe. in dealing with this difficulty.one that absorbed incident waves without reflection . In the laboratory. since this is at present the type most widely used on conductors. The Endurance Limit Approach. In this approach. based on the line design variables. (where L is actual span length. D is conductor diameter.As the Task Force focused on Stockbridge-type dampers applied at span ends. span length. H is horizontal tension in the conductor and m is mass of the conductor per unit length) has been used in analyses of collections of field experience data on fatigue of overhead conductors [24] to rank spans according to the difficulty in damping them. This rating parameter. L D H m . even for widely different conductor sizes. One technique uses a conductor span to determine the socalled efficiency of a span-end damping arrangement which can directly be used in the Energy Balance [21]. Data was available that was obtained by indoor testing techniques described in IEEE Std 664-1993. by the observation that tests on laboratory spans as well as on an outdoor test span [22] have shown somewhat similar level of damping efficiency1 using dampers of various sources. as defined in Section 4. certain utility specifications specify the same minimum level of performance in terms of damping efficiency for a significant range of conductor sizes [23]. 1 21 . Another technique consists in measuring the dynamic characteristics of a Stockbridge-type vibration damper on a shaker. The similarity in damping efficiency levels pointed toward a particular parameter to use in rating the protective capabilities of dampers. Thus. vibration levels are considered to result in an infinite lifetime of the conductor. [20]). It is noted that there are differences in design and application rules among Stockbridge dampers of different sources. In fact. damping is measured through the inverse standing wave ratio [20]. information on the span-end damping was needed to be employed in the Energy Balance analysis. even though limited to Stockbridge-type dampers. This observation was found to be true. when applied to well-designed and properly applied dampers of sizes appropriate to the conductors in question. Since the Task Force had adopted the parameter H/w (where w is weight of the conductor per Damping efficiency is defined [21] as the ratio of power actually dissipated by the damper to the power that would be dissipated by a perfect damper . and does not feel that it should. it has been forced to seek a single safe tension criterion for a population that is not homogeneous. “IEEE Guide for Laboratory Measurement of the Power Dissipation of Aeolian vibration Damper for Single Conductors” (see also ref. Travelling wave analysis was then used to predict the span-end dissipation characteristics from the measured complex characteristics of the damper [21].at the same frequency and amplitude. Selection of Parameters The Task Force has focused on Stockbridge-type dampers. The Task Force is not in a position to distinguish among these. was generally followed to assess computed vibration levels (the span response) with regard to the tolerance of the conductor to vibration. conductor size and tension.

10% turb. 5% turb.. The results shown here were determined on the basis of an assumed constant level of wind turbulence ranging from 5% to 30%. Rawlins [15] : Chukar : . for different damped A1/Syz (ACSR) and Aldrey conductors. more conservative.ACSR 42/7 EDS . . 15% turb. values. As the Panel did not 2 International Electrotechnical Commission (IEC) conductor designation [12] : see Table 5. It will be noticed that despite a large scatter in the calculated safe design tension according to the source. Condor : . (m) Leblond and Hardy [14] : Bersfort : . Also shown in Fig. 8% turb. As a rule.Aldrey 22 22 .1 in terms of the span parameter LD/m and the tension parameter H/w. Drake. Figure 5. . 5% turb. 5. 10% turb.ACSR 26/7 EDS . it was able to simplify the set of rating parameters L D H m and H/m to LD/m and H/w respectively. damper B : .. damper A : . .1 are the safe tensions recommended by the EDS Panel. (m /kg) 14 12 10 8 6 4 2 0 1000 1500 2000 2500 3000 3500 3 H/w. 5.. Predicted Safe Design Tension Safe design tension as predicted by Leblond and Hardy [25] for one conductor type over a full range of span lengths and by Rawlins [26] for three sets of conductor types and span lengths is depicted in Fig....3. . there is a fair agreement in the minimum. 10% turb.ACSR 48/7 EDS . EDS . . 22% turb. .unit length) to rate the effect of tension on conductor self damping. Drake.1 Predicted safe boundaries according to endurance limit approach. 10% turb. 5% turb. EDS ..ACSR 30/7 20 18 16 L D / m.. 30% turb. translated in terms of H/w.. the calculations were carried out using the endurance limit approach for the A1/Syz2 (ACSR) aluminium-conductors-steel-reinforced indicated in the legend.. 5% turb. .

1. but most probably were. their recommended tensions appear as straight vertical lines in the graphics. However. it was remarked. the figure for H/w applies to the design value of H. the damage may have occurred in spans that were not fitted with dampers. but it is considered likely that the fatigue occurred before any dampers were installed. for Aldrey conductors. Items 1. However. Fig.e. it may be observed that the safe design tension as calculated by the endurance limit approach for 8% wind turbulence is respectively more permissive. the Task Force interprets the presence of so many cases on the unsafe side as indicating a conservative bias in the estimated safe boundaries. beyond Category 4. It is not certain that all of these lines were free of fatigue damage.” These uncertainties are even more significant in the present case where dampers are involved. but a significant number fall on the “unsafe” side. It is evident that many existing lines fall on the safe side of the estimated safe boundaries. For Items 1 and 6. 26% of UTS or about 3000 m. Comparison With Field Experience In the previous section dealing with Unprotected Single Conductors. but an actual value was not reported. For Item 1.account for the span length. 5. the need to test the present recommendations against past experience is even greater than before. clearly pertain to Stockbridge dampers. Actual tension was higher due to contractor error. in fact. The analytical basis for applying the Energy Balance Principle is still open to question and is. i. The TF found only meagre information on field experience cases where conductor fatigue occurred in lines protected by Stockbridge type dampers.2 includes the estimated safe boundaries depicted in Fig. and that others were protected from severe vibration by very rough terrain. 5. Table 5. For the common range of span parameters. 5 < LD/m < 15 (m3/kg).11. 23 . all of which are protected by Stockbridge-type dampers.01 “Vibration Principles”. 5 and 6. medium steel content or high steel content A1/Syz (ACSR) conductors. Only three cases. The collection was drawn from files of Task Force members. Items 7 and 8 may or may not pertain to Stockbridge dampers.2 summarises cases from questionnaires collected by CSC6 [2] and SC22 WG04 for all damper types. It is likely that a few of these “unsafe” lines did experience damage. It is worthwhile to examine the range of the ranking parameters LD/m and H/w actually represented in overhead lines. about equally permissive or less permissive than the corresponding EDS values for low steel content. Figure 5. Item 5 represents the only case that is clearly valid for testing the estimated safe boundaries against field experience. “Uncertainties in the data and assumptions that are required to determine maximum safe tensions on the basis of the Energy Balance Principle have made it necessary to consider known experience with existing lines when specifying protection for new lines. the calculated safe design tension is definitely much more conservative over all the span range than the value recommended by the EDS Panel. Thus.2 shows a collection of points representing a number of actual lines. Thus. implying that the damaged line had originally been fitted with less damping than might have been required. The case led the utility to increase the number of dampers in designing its lines. currently under scrutiny by CIGRÉ TF B2.

respectively.983 5.20 26.28 31.755 5. (m /kg) 12 10 8 6 4 2 0 1000 1500 2000 2500 3000 3500 Ruling spans Maximum spans 3 H/w.00 26.59 27.77 27. Table 5.164 6.72 Stranding 26/7 54/7 30/7 30/7 26/19 54/7 54/7 54/7 32/19 54/7 Span metres 167 360 290 305 380 320 305 268 510 330 H/w metres 1542 2017 1851 1851 1994 2031 1677 1406 2227 1734 LD/m m3/kg 4.086 5.00 27. Damage likely occurred before they were installed. Evidently. mm 16.861 6. Item 1 2 3 4 5 6 7 8 9 10 Diam. Lines were built in 1930 and 1927.72 24. dampers were added later. Dampers may have been installed in dead-end spans and at angle towers only.18 16 14 Terrain Category : 1 2 3 4 L D / m.2 Field experience conductor fatigue cases – ACSR.2 Ranking parameters of damped ACSR lines in North America (no water crossings) in relation to estimated safe boundaries.691 5. Actual tension was higher due to contractor error.60 31. 24 . (m) Figure 5. 3.592 8. H/w based on design tension. Damage may have only occurred in spans that were protected by armour rods only.496 6. before dampers were commercially available.971 5. 2.40 27.140 Damper type Note Stockbridge Torsional Dumbell Dumbell Stockbridge Stockbridge “yes” “yes” Bretelles Elgra 2 3 3 1 1. Not all spans were damped.

. .Figure 5. falls slightly on the safe side of the Terrain Category 1 boundary. (m) Conductor damage reported : No damage reported : . . All but Item 9 fall on the “safe” side of all four boundaries. 25 .3 shows all of the cases from Table 5. All these cases are in harmony with the proposed safe boundaries. Lynx conductor over 4000 route km. Figure 5. Zebra conductor over 2800 route km. one damper/span test line safe cases. Item 6. . valid case. Their position in the plot suggests that the proposed safe boundaries are not excessively conservative. 12 10 Terrain category : 1 2 3 4 L D / m. other damage reported. Figure 5. careful vibration measurements and close inspection for damage. which may or may not be valid. The one valid case for Stockbridge dampers. (m /kg) 8 3 6 4 2 0 1000 1500 2000 2500 3000 H/w. . individual spans. . The cases drawn from the test line are particularly well documented because of the close control of conductor tensions. falls similarly. for comparison with the estimated safe boundaries. Item 5.3 also includes data from a number of operating lines where there is information that fatigue damage to conductors has not occurred.3 Available field experience regarding damped single conductor lines in relation to estimated safe boundaries. Both cases came from Category 1 terrain.2. other cases.

The Task Force recommendations are depicted in Fig. Metal Combination All 1350-H19 All 6101-T81 All 6201-T81 1350-H19 / Steel 1350-H19 / 6101-T81 1350-H19 / 6201-T81 Common Designation ASC or AAC AASC or AAAC AASC or AAAC ACSR ACAR ACAR IEC Designation A1 A2 A3 A1/S1A A1/A2 A1/A3 26 . For line parameters falling in the Special Application Zone. Aeolian vibrations should not be a constraint on design tension. at the average temperature of the coldest month on the site of the line.3 Conductor types to which recommendations apply.No Damping applies to undamped and unarmoured single conductors. The corresponding information is provided in Table 6.3 in algebraic form. Hence. the ratio of actual span length L times conductor diameter D to conductor mass m per unit length. the Task Force has resolved to provide guidance to conductor safe design tension with respect to Aeolian vibrations in terms of two parameters: the tension parameter H/w. LD/m. Table 5.3. each one set associated to a particular terrain category described in the legend.Task Force Recommendations As explained above.Span End Damping constitutes a zone where full protection of single conductors against Aeolian vibrations is assuredly feasible by means of one or more Stockbridge-type damper(s) set up at span extremities. The Safe Design Zone . Should there be any doubt about real terrain category. The basic Safe Design Zone . as already shown in Section 4. Terrains have been divided into four categories according to their general characteristics. This guide applies to all round wire. within the limits of this zone.4 in the form of four sets of curves. concentric lay. This zone is defined in terms of the H/w parameter only and it is unlimited in the LD/m parameter. the ratio of horizontal tension H in the span to conductor weight w per unit length and the span parameter. The tension H refers to initial horizontal tension before any significant wind and ice loading and before creep. overhead electrical conductors shown in Table 5. 5. Aeolian vibrations may or may not be a constraint and it is recommended that line designers determine the availability of adequate protection before finalising the design. the lowest category should be selected.

summer time. flat. prairie. Figure 5. woodlands and shrubs. tundra.4 Recommended safe design tension for single conductor lines. #2 : Open. or near/across large bodies of water or flat desert. (m) Terrain Category #1 : Open. D: conductor diameter and m: conductor mass per unit length. Although a lower fatigue endurance of A3 (AAAC) conductors may be surmised from reference [13]. L: actual span length. #4 : Built-up with some trees and buildings. no snow. there seems to be no well documented field evidence to support a more pessimistic tension recommendation for these conductors. farmland without any obstruction. It was resolved to give a uniform recommendation for all types of conventional conductors using aluminium and/or aluminium alloy. Small fields with bushes. 27 . small towns.g. open grass of farmland with few trees. residential suburbs. or undulating with very few obstacles. trees and hedges. flat. w: conductor weight per unit length. e. no obstruction.g.g. H: initial horizontal tension. no obstruction. (m /kg) 3 12 10 8 6 4 2 0 0 500 1000 1500 2000 2500 3000 Safe Design Zone No Damping Safe Design Zone SpanEnd Damping H/w. e. hedgerows and other barriers.20 18 16 14 Special Application Zone L D / m. e. no trees. #3 : Open. with snow cover. flat.

This statement should in no way be taken to mean that all commercially available dampers will provide such protection up to the limits of those boundaries or that they could provide such protection indefinitely without themselves failing from fatigue. The Task Force has used the parameters H/w and LD/m to construct boundaries within which satisfactory performance should be commercially available. rime or hoarfrost in which case dampers may break up prematurely as a result of galloping and/or excessive Aeolian vibrations. Use of armour rods or special supporting devices such as cushioned clamps and helical elastomer-bushed suspensions may justify higher design tensions. it was judged not feasible at this juncture to arrive at safe tension recommendations for these types. However. As a matter of fact. It may be noted here that conductors do not get more susceptible to other forms of wind-induced vibrations when tension is increased. for spans exposed to pollutants that may decrease the fatigue endurance of the conductors. Rather. often from multiple sources. special situations require specific attention. 28 .Existing lines using single conductors strung at a tension exceeding the recommended value for the terrain and span length may require inspection and field measurement. the Task Force has focused on Stockbridge-type dampers. their propensity to both galloping [27] and wake-induced oscillations [28] (of bundled conductors) has been shown to decrease with increased tension. The guidance provided herein should be suitable most of the time. for spans equipped with warning devices and for spans using non-conventional conductors such as compact or high temperature conductors. Information on safe tension. However. appropriate field testing or evaluation of available dampers is likely to reveal at least some that do. Techniques to perform such vibration measurements have been described previously [19]. when these devices are employed should be obtained from their suppliers. for spans often covered with ice. It falls to the line designer to identify them. It is noted that the “Bretelle” is also used in some countries. Such is the case for extra long spans. helical impact dampers are widely used on earth wires and special designs such as “Festoon” dampers are employed in certain applications such as fjord crossings. Limitations As mentioned above.

CPS ≡ Hertz) 1 – Single conductor 2. BUNDLED CONDUCTORS The present section aims at recommending safe design tensions for bundled conductor lines. Fig. Liberman and Krukov reported that amplitudes 29 . twin. triple apex-down bundles and quad horizontal bundles made up of conventional stranded conductors fitted either with damping spacers or nondamping spacers or a combination of non-damping spacers and span-end Stockbridge-type dampers.1 Comparison of vibration in single versus bundled conductors Drake ACSR in 1200-foot span [29].1 shows results of simultaneous recordings at Alcoa’s outdoor laboratory in1960 [29]. Figure 6.and quad-bundles respectively [30]. Early investigations showed that the reduction in amplitude becomes greater as the number of subconductors in the bundle increases. 6. Leibfried and Mors reported that spans in the Hornisgrinde Test Station displayed.6. with and without dampers on the span.3 – Subconductors in horizontal two-conductor bundle A – No dampers B – One Stockbridge damper at end opposite recorder. (MILS ≡ in x 10-3. amplitudes in the ratios 4:2:1 in spans with single. Review of Literature Numerous field tests have demonstrated that bundled conductors respond less to Aeolian excitation than single conductors of the same size and at the same tension as those of the bundle. The recommendations cover twin horizontal bundles. on average. Bundling reduced vibration amplitudes by about half. For example.

There was some influence by spacer design. Finally. 2056 m in the second and 2096 m in the third. and by a factor of 20 in a quad bundle. relative to a comparable single conductor. but in all of these tests the spacers did not include intentional damping. Table 6.2 also gives details of 14 triple apex-down bundled lines (3AD). but seems greatest in those that have size in the vertical direction. such as triple and quad bundles. They found that damping of a certain type reduced amplitudes in a twin bundle by a factor of 5. out of which 19 lines have been fitted with non-damping spacers (NDS) alone. The benefit of damping was confirmed by Diana et al. Table 6. This time.2 includes 7 quad horizontal bundled lines (4H). It comprises 70 twin horizontal bundled lines (2H). The effect of intentional damping in spacers was investigated initially by Edwards and Boyd at Ontario Hydro [32]. The effect of bundle configuration upon this benefit was investigated by Hardy et al. out of which one line only has been equipped with NDS. 48 lines fitted with a combination of such spacers and span-end Stockbridge dampers (NDS+Stk) and 3 lines fitted with damping spacers (DS) alone. at the Magdalen Islands Test Station of IREQ [34]. and of 5 to 10 for triple and quadruple bundles [31]. In this case. Table 6. the benefits of bundling were attributed to the effects of mechanical coupling among the subconductors interfering with the vortex excitation mechanism. Damping in spacers is beneficial in all bundle configurations.1 collects the test conditions and salient data from a number of these field tests. Thus. the tension parameter H/w reached 1627 m in the first case. out of which 3 are protected by means of NDS+Stk and 4 others by means of DS.2 summarises some line design variables of 91 bundled conductor lines in North America which have most likely operated for many decades without any Aeolian vibration problem. at the Porto Tolle Test Station in Italy [33]. the tension parameter H/w was set to a maximum of 1488 m and 1937 m respectively. Review of Field Experience Table 6.were reduced by factors of 1.5 to 2. 30 . 4 lines with NDS+Stk and 9 lines with DS.5 for horizontal two-bundles. 2959 m in the second and 1937 m in the third. The tension parameter H/w ranged up to 2088 m in the first case.

0 Spacer type End damper Amplitude ratio single/bundle Table 6.3 6. twin Hor.6.63 .5 6.8 6.7 >1.27 5.14 . quad Hor.7 ~3.20 .Table 6.33 .7 >1.8 6.4.2959 1636 .7. twin Vert.8 6.6.89 7. twin Hor.1488 1633 .38 Range of initial H/w (m) 802 .60 5.03 .63 3.2056 1401 .7.8.81 6.5 6.2096 1452 .1 Synthesis of field test experience about comparative vibration behaviour of bundled conductors.8 Articulated Articulated Ball-&-socket Ball-&-socket Various Various Articulated Articulated Articulated Articulated Articulated Grouped twins Grouped twins Rigid Damping spacers Damping spacers Damping spacers Damping spacers Damping spacers Damping spacers Damping spacers No Yes No Yes No No No No No No No No No No No No No No No No No >2. twin Triple Hor.5 6.2 Outlook of field experience with bundled conductor lines in North America.1 >1.3 6.9 ~2 ~4 >1. twin Triple Hor.3 6.5 ~7.3 7. twin Hor.3 6.6 ~8.3 7. twin Triple Hor. Ref.5 >1.4 ~6.8 >2.5 7.1937 1627 1166.5 6. of lines Bundle type 2H 2H 2H 3AD 3AD 3AD 4H 4H Protection Range of mean LD/m (m3/kg) 2.5 6. Bundle type [29] [29] [29] [29] [30] [30] [31] [31] [31] [31] [31] [31] [31] [33] [33] [34] [34] [34] [34] [34] [34] Hor. twin Vert.93 3.8 6.93 . twin Hor.3 6. quad Nominal conductor H/w (m) 1755 1755 1755 1755 1295 1295 >1454 >1437 >1730 >1454 >1730 >1437 >1730 1743 1743 1550 1550 1550 2325 2325 2325 Test span LD/m (m3/kg) 6. No.7 >5 >5 >5 >1.7.6 ~13.62 6.8 6. twin Triple Hor.2088 910 .2 7.1937 19 48 3 1 4 9 3 4 NDS NDS+Stk DS NDS NDS+Stk DS NDS+Stk DS 31 . twin Hor.3 >5 ~2.5 >1. twin Hor.2 6.19 . twin Hor. quad Hor. quad Hor.

2.3.3 6. The same absolute limit (H/w = 2500 m) was indeed applied to all systems even if they are strung over the roughest terrain. In such cases. instrumented field test data related to twin horizontal bundles fitted with NDS alone. 2 and 4. applicable to the test station. For triple apex-down bundled lines fitted with NDS. reference was made to comparative field tests [35] carried out at the IREQ test line in Varennes clearly showing such a tension H/w = 2100 m to be safe for terrain category #2. were then determined on the basis of terrain category #3 using the same terrain-to-terrain ratio as for the undamped single conductor case.Determination of Safe Design Tension As an accommodating reference and benchmark.3 for each one of the four terrain categories that were then defined.2 Bundled conductors fitted with non-damping spacers The Task Force could not find any well-documented. to keep on the prudent side. as it could be appreciated for one of the lines in Table 6. it was resolved to associate that safe design tension to terrain category #3. as shown in Table 6. 6. But field experience suggests that a tension of H/w = 2100 m should be safe for terrain category #2. it appears reasonable to use the same safe design tension limits as for equivalent single conductors whether they are undamped (system #3) or damped by means of Stk (system #4). as indicated in Table 6. the safe design tensions that were recommended previously for undamped single conductors (conductor system #1) and single conductors fitted with Stockbridge dampers (Stk) at the span extremities (system #2) are shown in Table 6.3 (system #8). However.3 (system #5). as indicated in Table 6.1 Unspacered bundled conductors fitted or not with span-end Stockbridge dampers Unspacered bundled conductors have sometimes been used with the object of reducing their susceptibility to galloping. The safe design tensions for the other terrain categories were determined using the same transposition principle as for the twin bundles.3. The safe design tensions for terrain categories #1. However. it was resolved to limit the safe design tension for terrain category #4 to H/w = 2500 m as a matter of prudence. 32 .

Table 6.12 < 15 < 1725 < 2615/(LD/m) 0. Twin horizontal bundled conductors with non-damping spacers and span-end Stockbridge dampers 7.12 < 15 < 10 > 10 . Triple apex-down bundled conductors with non-damping spacers and span-end Stockbridge dampers 10. farmland without any obstruction. D: conductor diameter and m: conductor mass per unit length. w: conductor weight per unit length. < 15 < 2500 < 1850 < 2615/(LD/m) < 2100 < 2780/(LD/m) < 2100 < 2500 0. hedgerows and other barriers.12 < 2200 < 15 < 15 < 2100 < 2780/(LD/m) < 2100 < 2500 < 15 0. unspacered twin. summer time.12 < 15 < 2860/(LD/m)0. Twin horizontal bundled conductors with non-damping spacers 6.12 < 15 < 15 < 1925 < 2780/(LD/m) 0. no obstruction.3 Recommended conductor safe design tension with respect to Aeolian vibrations. tundra. < 15 < 2500 < 3030/(LD/m) < 2500 < 2500 0. woodlands and shrubs. L: actual span length. < 15 < 15 < 2500 Terrain category # 1 : Open. prairie. < 15 < 2275 < 2860/(LD/m) < 2275 < 2500 0. small towns. no obstruction. Terrain category # 2 : Open. Conductor system Terrain Cat.12 < 15 <6 > 6 . Undamped.12 < 15 <5 > 5 . #4 LD/m (m3/kg) H/w (m) < 1425 < 3030/(LD/m)0. e.12 < 15 < 2780/(LD/m)0. with snow cover. < 15 < 1900 < 1850 < 2615/(LD/m) 0. < 15 < 2500 < 3030/(LD/m) < 2500 < 2500 0. e.12 < 15 < 3030/(LD/m)0.g. flat. #1 LD/m (m3/kg) H/w (m) < 1000 < 2615/(LD/m)0. Quad horizontal bundled conductors with damping spacers < 2615/(LD/m)0. e. Quad horizontal bundled conductors with non-damping spacers and span-end Stockbridge dampers 13. or near/across large bodies of water or flat desert.12 < 15 <7 > 7 . triple & quad bundled conductors with span-end Stockbridge dampers 5. Undamped single conductor 2. flat or undulating with very few obstacles. 33 .12 < 1000 < 15 Terrain Cat. flat. Unspacered twin. Small fields with bushes.g.12 < 15 < 10 > 10 . Single conductor with span-end Stockbridge dampers 3. no snow.12 < 15 < 13 > 13 . no trees. < 15 < 2450 < 3030/(LD/m) < 2450 < 2500 0. #3 LD/m (m3/kg) H/w (m) < 1225 < 2860/(LD/m)0. residential suburbs. Terrain category # 4 : Built-up with some trees and buildings.g.12 < 15 < 15 < 2100 < 2860/(LD/m) < 2100 < 2500 0.12 < 1425 < 15 1. Quad horizontal bundled conductors with non-damping spacers 12. < 15 < 2275 < 2860/(LD/m) < 2275 < 2500 0. Twin horizontal bundled conductors with damping spacers 8.12 < 15 <7 > 7 . Triple apex-down bundled conductors with damping spacers 11. Triple apex-down bundled conductors with non-damping spacers 9.12 < 15 <5 > 5 . triple & quad bundled conductors 4. H: initial horizontal tension.12 < 1125 < 15 Terrain Cat. #2 LD/m (m3/kg) H/w (m) < 1125 < 2780/(LD/m)0.12 0. Terrain category # 3 : Open. open grass of farmland with few trees. trees and hedges.12 < 1225 < 15 Terrain Cat.

However. as a matter of prudence. 8 and 11). as well as an undamped single conductor of the same type.4 Bundled conductors fitted with damping spacers To determine safe design tensions of bundles fitted with DS. 6. to triple apex-down bundles (system #9) and quad horizontal bundles (system #12) as shown in Table 6. For all systems using NDS or NDS+Stk (systems #5. 6 to 9 and 8 to 13 times less severe on the twin. the Task Force could not find any well-documented field test data related to bundles fitted with NDS+Stk which could have been used to determine safe design tensions rigorously.2 in terms of a relative effective amplitude as a function of the tension parameter H/w. 6. due to lack of accurate information from the field.Quad bundles (system #11) fitted with NDS should be somewhat less prone to Aeolian vibrations than their triple equivalents. The effective amplitudes are then normalized by means of the effective amplitude of the single conductor strung at H/w = 2325 m to yield the relative effective amplitudes. 8. Each conductor system was tested successively at a nominal tension H/w of 1550 m. staggered in a similar manner. an apex-down triple and a horizontal quad bundle of ACSR Bersfort conductors fitted with the same number of DS. the Task Force could rely on the results of an extensive program of tests [34] carried out at the IREQ test line in the Magdalen Islands.3 Bundled conductors fitted with non-damping spacers and span-end Stockbridge dampers Again. 6. 2325 m and 2870 m respectively. quad bundles were assigned the same safe design tensions as the triple bundles (system #8). whichever are more permissive. triple and quad bundles on the other hand. The effective amplitudes relate to a single. 11 and 12). to limit the range of application of the given safe design tensions H/w to spans for which LD/m < 15 m3/kg. There are two and three such points for the single conductor on the one hand and each one of the twin.3. triple and quad bundles respectively. 9. it appears reasonable and at one and the same time prudent to recommend the same safe design tensions as for either single conductors fitted with Stk (system #2) or bundles fitted with NDS alone (systems #5.3. in comparison to the undamped single conductor. 6. The most relevant results are shown synthetically in Fig.2 which are then related by straight line segments.6. 34 . The test program covered a horizontal twin. integrated figure expressing the overall Aeolian vibration response of each system according to a severity or fatigue-damage point of view. It may be observed first that the relative effective amplitude of vibration is from 2 to 4. That applies to twin horizontal bundles (system #6). it was resolved. The resultant relative effective amplitudes are plotted as data points in Fig. at the same tension. However.3.

one would need at least one data point for the undamped single conductor that would stand on the low side of the range of safe design tensions for such a conductor. this amplitude is reached when the tension is set at H/w = 2200 m.2. 1000 m < H/w < 1425 m.0 0 500 1000 ple e d Tr i Damp d Damped Qua 2000 2500 3000 1 2 T ed mp Da 3 win 4 1500 Tension parameter H/w.4 3 2 1 4 Un d am ped D 0. as the test site was classified. For lack of such a point from the test program.6 B 0. if line segment AB is accepted as a reasonable approximation of the actual single conductor response in the range 1000 m < H/w < 1550 m. one can proceed as follows: for instance.6.e. (m) Figure 6. the IREQ’s calculation model [36] was called upon to determine the expected response of the undamped single conductor below data point B in Fig.2 A 1 2 3 4 0.2. i. it may be seen that the relative effective amplitude of the undamped single conductor is expected to stand at about 0.27 when the tension is set at H/w = 1125 m while for the twin bundled conductors. 6.0 Relative effective amplitude 0. assuming a turbulence intensity corresponding to terrain category 2. In order to proceed further. 35 . As a result.2 Determination of safe design tension for twin horizontal bundled conductor lines fitted with damping spacers on the basis of comparative field test results [34].C 1. going along arrowed line 2 in Fig.8 Sin gle 0. it looks reasonable to conclude that H/w = 2200 m may be used as the safe design tension for the twin bundle fitted with DS where H/w = 1125 m is considered safe for the undamped single conductor. Now.

As for the triple and the quad bundles (system #10 and #13). according to the same transposition procedure. Using the same transposition procedure as above would lead to quite high safe design tensions even for terrains category 1. this result makes sense since it is expected that the slope of the line expressing the effective response of any conductor system as a function of the tension parameter H/w should be determined entirely on structural grounds and not on aerodynamics grounds. 2000 m and 2500 m respectively for each of the four terrain categories. 6.2 that their response is much lower than that of the twin bundle. It could be confirmed that the locus of the predicted. it was argued that. the effective amplitudes would have certainly moved up or down in absolute terms. Comparison With Field Experience Figures 6. in the latter case. 36 . Doing so. the IREQ computer model for undamped single conductors was run at H/w = 1500 m. which indicated a safe condition. 6. However.5. in terrain category 2.2 could also be used to determine the safe design tensions of twin bundles for the other terrain categories. the respective locus for the undamped single conductor and each one of the damped conductor bundles when shown in relative terms would be expected to be approximately unchanged. had the same tests been repeated in other terrains. 2500 m and 3050 m respectively for the horizontal twin bundled conductors (system #7 in Table 6. the Task Force resolved again to limit the safe H/w to 2500 m as a matter of prudence.Strictly speaking. To check that assumption to a certain extent. Such is also the case for both triple apex-down and quad horizontal bundles fitted with either NDS+Stk or DS. the above procedure for transposing the safe design tension of the undamped single conductor to the twin bundle fitted with DS would be valid only for terrains category 2 as for the test site. Besides. 6. As a result.4 depict the proposed safe design tension limits together with the corresponding field cases in Table 6. 3 and 4 respectively translate into safe design tensions H/w = 1900 m.3) in the same terrain categories. Again. effective amplitude for each terrain category almost coincides when normalized appropriately. it is clear from Fig. as shown in Fig. the Task Force decided to limit their respective safe design tension to H/w = 2500 m for any terrain category. 1225 m and 1425 m for the undamped single conductor in terrain category #1. However. but it seems reasonable to assume that they would have moved everywhere in even proportion. The above safe H/w figures are well supported by the results of the test on the twin bundle at H/w = 2870 m at the test site.3 and 6. it came out that the safe design tensions H/w = 1000 m. it appears that Fig. It will be seen that the proposed limits reflect well field experience. It should be noted that safe design tensions H/w for bundled conductors fitted with DS are provided in a way independent of the span parameter LD/m as it was considered that the benefits of damping are usually well distributed over the span length. Thus.2 for the twin horizontal bundles fitted with NDS or NDS+Stk respectively.

15 Field cases Terrain #1 Terrain #2 Terrain #3 L D / m. (m) Figure 6.4 Ranking parameters of twin horizontal bundled lines in North America fitted with non-damping spacers and span-end Stockbridge dampers in relation to estimated safe boundaries.15 Field cases Terrain #1 Terrain #2 L D / m. (m /kg) 10 Terrain #3 Terrain #4 3 Safe Design Zone 5 Special Application Zone 0 0 500 1000 1500 2000 2500 3000 3500 H/w. (m) Figure 6. 37 . (m /kg) 10 Terrain #4 3 Safe Design Zone Special Application Zone 5 0 0 500 1000 1500 2000 2500 3000 3500 H/w.3 Ranking parameters of twin horizontal bundled lines in North America fitted with non-damping spacers in relation to estimated safe boundaries.

combined or not with span-end Stockbridge dampers. Terrains have been divided in four categories according to their general characteristics. NDS+Stk L D / m. DS 3 Safe Design Zone Special Application Zone 5 0 0 500 1000 1500 2000 2500 3000 3500 H/w. NDS+Stk Terrain #2. The Task Force recommendations regarding each one of several bundled conductor systems. the safe design zones are defined in terms of the H/w parameter only in which case they are unlimited in the LD/m parameter. For some systems. (m /kg) 10 Terrain #4. However in all cases. the Task Force has resolved to provide guidance to conductor safe design tension in terms of two parameters: the tension parameter H/w. at the average temperature of the coldest month. are summarized in Table 6. as the case may be.3 in the form of simple algebraic expressions. the span parameter LD/m. (m) Figure 6. NDS+Stk Terrain #3. Task Force Recommendations As explained above. or 38 . the ratio of tension H to conductor weight w per unit length and. the lowest should be selected. they constitute zones where full protection of conductors against Aeolian vibrations is certainly feasible by means of a reliable system of non-damping spacers. each one associated to a particular terrain category described in the legend.15 Field cases Terrain #1. in addition to two single conductor systems.5 Ranking parameters of triple apex-down and quad horizontal bundled lines in North America fitted either with non-damping spacers (NDS) plus span-end Stockbridge dampers (Stk) or damping spacers (DS) in relation to estimated safe boundaries. NDS+Stk Any terrain. Should there be any doubt about real terrain category. The tension H refers to initial horizontal tension before any significant wind and ice load and before creep. the ratio of actual span length L times conductor diameter D to conductor mass m per unit length.

Lack of such relative mobility may lead to harmful vibration trapping. 39 . Hence. However. fatigue or wear. Use of armour rods or special supporting devices such as helical elastomer-bushed suspensions may justify higher design tensions. The guidance provided herein should be suitable most of the time. It falls to the line engineer to identify them. Aeolian vibrations may or may not be a constraint and it is recommended that line designers determine the availability of adequate protection before finalising the design. concentric lay. For line parameters falling outside of these zones. non-damping spacers should be understood as spacers allowing a certain relative mobility of the attachment points with the subconductors in the vertical direction. overhead electrical conductors listed in Table 5. Rather. This guide applies to all round wire. Limitations and Warnings The Task Force has recommended safe design boundaries within which satisfactory performance should be available. within the limits of these zones. This statement should in no way be taken that all commercially available spacing and/or damping systems will provide such protection up to the limits of those boundaries or that they could provide such protection indefinitely without themselves failing from loosening. rime. when these devices are employed should be obtained from their suppliers. Aeolian vibrations should not be a constraint on design tension. Information on safe design tension. often from multiple commercial sources. for spans often covered with ice. In the context of this paper. or hoarfrost. for spans equipped with aircraft warning devices and for spans using nonconventional conductors. appropriate field testing or evaluation of available systems is likely to reveal at least some that do.3 supported in conventional metal-to metal clamps.alternatively. special situations require specific attention. Such is the case for extra long spans. a reliable system of damping spacers.

1979.. Detroit. Rawlins. Eng. 963-971. A. "A Contribution to Safe Design T/m Values"..B. 124. and Larson. Electric Power Research Institute. “Across-Flow Response Due to Vortex Shedding”. Vol. California. ESDU. CIGRÉ Study Committee 22.7. April 1983. “The Importance of Fretting in Vibration Failure of Stranded Conductors”. PAS-87. C. Dulhunty. C.. April 1995. Greathouse. “Bare Conductors and Mechanical Calculation of Overhead Conductors”. Working Group 04. Rawlins.. R. "Conductor Vibration . Science Data Item Number 78006. C.B.. Rawlins. and Roughan. Electric Power Research Institute.. C. Krispin. C. IEEE Trans. pp. CIGRÉ SC22-WG04 Task Force document. Chapter 3. 1961. CIGRÉ Session 1960. London. International Electrotechnical Commission. Report CIGRÉ SC22-WG11-TF4-95-13. "Wind-Induced Conductor Motion".B. Electra No. K. and Leblond A. Oct. Lamprecht. Fricke.R. Report CIGRÉ SC22-WG11TF4-94-10. “Recommendations for the Evaluation of the Lifetime of Transmission Line Conductors”. Vol. Hardy.D. 1979. presented at the AIEE Fall General Meeting. No. 1992. 223. CIGRÉ Study Committee 22. 4. Jr.. on Power Apparatus & Systems. EPRI Transmission Line Reference Book. pp. IEEE Trans.B. April 1996. 1982. March 1979. J. “Updating the Davenport Roughness Classification”. "Exploratory Calculations of the Predicted Fatigue Life of Two ACSR and One AAAC". Working Group 01..A Study of Field Experience". Wieringa. MI. Report CIGRÉ SC22-WG11-TF4-95-6.. 103145.W.-J. Electra No. ESDU.. "Estimated Maximum Safe H/w for Undamped Conductor Spans".E. 40 [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] . W. 1994. California. June 1991. P. “Wind Tunnel Measurement of the Power Imparted to a Model of a Vibrating Conductor”.. PAS-102. "Parametric Studies: Vibration Intensity of Various Conductors as a Function of Tension and Wind Turbulence". 63. 15-20. Chapter 2 of “Wind-Induced Conductor Motion”. Wind Engineering & Industrial Aerodynamics. C. on Power Apparatus & Systems. Palo Alto. "The Fatigue Life of Overhead Line Conductors". “Round Wire Concentric Lay Overhead Electrical Stranded Conductors”. Report No. Oct. 1978. pp. 40-77. May 1989. R.B. Rawlins. pp.G. J. Palo Alto. “Fatigue of Overhead Conductors”. pp.. REFERENCES [1] [2] Zetterholm. EPRI Transmission Line Reference Book. Doocey et al. H. CP61-1090. “Report on Aeolian Vibration”. and Rawlins. 6. Claren. Report CIGRÉ SC22-WG11-TF4-96-5. AIEE Paper No. No. IEC 61089 Publication. 1381-1384. O.. 357-368. June 1968. September 1995. 41-44.

53-69. 1995. Vol. “Comparative Field Tests on Various Practices for the Control of Wind-Induced Conductor Motion”. CIGRÉ General Session. Van Dyke. Harvey. C. IEEE Transactions on Power Delivery. Leblond A. Vol. “Improved Systems for Recording Conductor Vibration” Trans. and Hardy. “Safe Tensions with Dampers”. pp. July 1988.”.L. AIEE Paper No. pp. F. 1961. C. pp. Tavano & U. pp. 43-60. P. August 1981. 924-932. "Endurance Capability of Conductors". 2. C. Électra No. 1964. April 1999. of the AIEE. 1960. 10-20 June 1968. No. WG11. 1494-1501. W. Rawlins & J. 1-9 September 1982. “Quantitative Relationship in Conductor Vibration Damping”. 41 [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] ..B. J. M... pp. “IREQ Predicted Safe Design Tension”. 10.R. Karlsruhe”. AIEE Transactions on Power Apparatus and Systems. April 1999. Vol. No. 879-894. Electra No. October 1. Working Group 04. III. 76. “Guide on the Measurement of the Performance of Aeolian Vibration Dampers for Single Conductors”. Electra No. 3744-3753.S. 75. 8. A. L. T. Bonneville Power Administration. Gasparetto. No. October 1965. “Field Observations on Wind-Induced Conductor Motions”. and Jones.2G.. M. 9. IEEE Trans. Rawlins. Leibfried & H. CIGRÉ Study Committee 22.S. presented at the AIEE Fall Meeting. Working Group 11. L. B. and Larson. Edwards & J. 1956. M. April 1997.B. 1995. Merrill. CIGRÉ Study Committee 22. PAS-84.[17] CIGRÉ Study Committee 22. pp. Cosmai. 12. Report CIGRÉ SC22-WG11-TF4-99-5. Detroit. Hardy. Dec. IEEE Transactions on Power Apparatus & Systems... CIGRÉ Study Committee 22.. J. “Analysis of Conductor Galloping Field Observations – Single Conductors”. Feb.. Tompkins. Boyd. CIGRÉ General Session. Mors. Vol.. Transmission and Distribution Committee. 163. R. “Field Measurements and Field Data Processing on Conductor Vibration (Comparison Between Experimental and Analytical Results)”. K. CP 61-1090. Final Report. C.R. and Gardes. Task Force 1. C. May 1981. “Guide to Vibration Measurements on Overhead Lines”. Hardy. 1987. Diana. December 1998. Liberman & K. Vibration of Overhead Line Conductors and Protection Against it in the U. Task Force 2. “Die Bündelleiter-Vershuchsanlage Hornisgrinde der Badenwerk AG. St-Louis.R. On Power Apparatus & Systems. PAS-100. Rawlins C. “Conductor Vibration – A Study of Field Experience”. 198. pp.S.B.J. G. MI.21-28. Vol. A. October 15-20. Rawlins. P. Journal of Fluids and Structures.. “Bundle Conductor Design Requirements and Development of ‘Spacer-Vibration Damper’”. and Van Dyke. Pt. Report CIGRÉ SC22-WG11-TF4-99-6..E. Krukov. Greathouse. "Modelling of Aeolian Vibrations of Single Conductors : Assessment of The Technology".L.P. Specification ETF 60-25.

Hardy. P. St-Louis & J. 220T553. A. “Field Evaluation of Aeolian Vibration of Bundled Conductors”. “Comparative Practices for the Control of Wind-Induced Conductor Motion”. 12. April 1997. Gardes. Hardy. On Power Delivery. Vol. Hardy. Proc. C.[34] [35] C. M. P. pp.L. Canadian Electrical Association Report No. 2. [36] 42 . IEEE Trans. Jahier & S. December 1990. “Assessment of Safe Design Tension with Regard to Aeolian Vibrations of Single Overhead Conductors”. Leblond & C. of the ESMO 2000 Conference. Houle. Van Dyke. 202-208. No. 8-12 October 2000. Montreal.