Aeolian Conductor Vibration Control http://tdworld.

com/print/substations/aeolian-conductor-vibration-control

print | close

Aeolian Conductor Vibration Control

T&D World Magazine
W.F. Bückner, Transmission Line Consultant, Germany
Sun, 1998-02-01 12:00

Aeolian conductor vibration control is as important today as it was 70 years ago. After the first recorded
cases of conductor damage in 1925, reported by Stockbridge, aeolian vibrations became the object of
numerous observations; however, none of the many theoretical investigations during the past 70 years have
led to any generally applicable design guidelines.

To relate the conductor motion at a conventional suspension clamp to the likelihood of fatigue, the Institute
of Electrical & Electronic Engineers (IEEE) implemented a standardization of conductor vibration
measurements3 in 1966. It is based on the differential vertical displacement (Yb), peak to peak of the
conductor as related to a suspension clamp, measured 89 mm (3.5 inches) from the last point of contact
between the clamp and the conductor. In the U.S. a bending amplitude of about 10 mils (peak to peak) is
used as an approximation of a vibration-safe design.

In 1979, the International Conference on Large High Voltage Electric Systems (CIGRE) recommended an
improved method to determine the lifetime of conductors because of aeolian vibration1. This method is used
in the aircraft and automobile industries and is called Miner's rule4, which considers a linear damage
accumulation. The lifetime endurance capability depends on the maximum dynamic stresses on the
conductor and the related number of vibration cycles.

An improved method for vibration-damper evaluation, presented in 1991, combines the CIGRE
recommendation with complementary laboratory tests5.

Modern recorders connect directly to a PC via the RS 232 interface for readout of accumulated
measurements. This arrangement requires software for the readout and storage of data, evaluation of the
measured data, determination of the lifetime as well as easy setting of the measured parameters, such as
amplitude and frequency ranges, sampling and pause periods and start and end of measurements. The data
are stored in a matrix with at least 16 frequency plus 16 amplitude classes.

Figure 1 shows a modern vibration recorder, which simultaneously measures vibration-bending amplitudes
and frequencies as well as the wind velocity. The device determines by an internal algorithm the frequencies
and the number of cycles and, if available, wind-velocity classes. Measurements are normally taken four
times per hour with a recording period of about 10 sec. For this setting, six-month unattended operation
should be possible. The software program related to the recorders sometimes includes useful data
evaluations, such as accumulated stress curve, maximum amplitude and stress vs. frequency, danger ratio vs.
frequency, - vs. amplitude and - vs. wind-velocity class, presented in tables and graphs.

The behavior of vibrating conductors under natural wind conditions are characterized by: The accumulated
stress curve, which leads to an estimate of conductor lifetime. The occurrence of each frequency range. The
danger factor, which presents the danger share of the different frequency and amplitude/vibration angle
ranges on conductor lifetime.

Accumulated Stress Curve The measurement output is an amplitude/frequency matrix (Table 1). The matrix
shows the number of events for each amplitude/frequency pair and is the basis for the evaluation of the
measurements. The evaluation of these data will be done by software related to the recorder used.

1/4 2014.01.06. 22:06

Aeolian Conductor Vibration Control http://tdworld.com/print/substations/aeolian-conductor-vibration-control

Figure 2 shows some accumulated stress curves measured on different transmission lines. The legend
explains these curves. Measurements of OPGW cable in Germany at locations A and B only 10 km (6 miles)
apart from each other on the same line differ considerably. Another OPGW cable in Switzerland behaved
similarly to location A. The behavior of an ACSR conductor (560/50) of a triple bundle on a 380-kV line and
of a ground wire ACSR (210/35) without damper are shown too. Also shown is the SN curve for ACSR and
AAAC conductors called CIGRE Safe Border Line, a conservative limit that leads to low estimates of
conductor lifetime.

The accumulated stress curves shown in Fig. 3 belong to an AAAC conductor 228 sq mm (.353 sq inch) of a
132-kV line in Niger with damper, and to a ground wire ACSR (210/35) without damper, and to an OPGW
cable in Germany with well-selected dampers. Also, three SN curves are presented with the CIGRE Safe
Border Line, the SN curve for an ACSR conductor 210/35 as well as the SN curve for AAAC smooth
conductor 245 sq mm (.380 sq inch). The probability of breakages is 50% for the last two curves mentioned.
The SN curves are valid for conductors with suspension clamps. The fatigue limits of conductor without
fittings is about double. The static tension of the conductors was equal to 20% ultimate tensile strength
(UTS). The bending amplitude Yb multiplied by the so-called Poffenberger-Swart factor (F) leads to an
approximation of the related conductor stress6. s = Yb • F (zero to peak in N/sq mm if yb is in mm zero to
peak.)

This method can be applied at locations other than suspension clamps, but the theoretical conversion of the
bending amplitude into stress could be inaccurate because of the difficult boundary conditions of the
relevant differential equation. A modified improved conversion method that uses greater conductor stiffness
at lower bending amplitudes was recently presented7.

The calculation of the lifetime is based on the load-block diagram that delivers the number of cycles (n)i for
each measured stress level for one year as shown in Table 2. The danger rates (Di) of the different stress
levels are calculated corresponding to the Miner formula: Di = ni/Ni Ni is the relevant number of cycles of
the one-step SN curve (Wohlercurve) of the same stress range. The lifetime (LT) is according to Miner: LT =
1/ (ni/Ni)

The relevant bending amplitudes of the accumulated stress curves in Figs. 2 & 3 exceeds, in some cases, a
bending amplitude of 10 mils (peak to peak), which is often considered a safe limit in the U.S. The lifetime of
the conductor investigated by field tests determined with the Miner formula were sufficient with two
exceptions: the AAAC conductor 228 sq mm (.353 sq inch) in Niger (about a four- year lifetime) and the
ACSR 210/35 without dampers (about a six-year lifetime). The bending amplitude of 10 mils or 1/100 inch or
(0.25 mm) corresponds to a bending stress of 7.5 N/sq mm for an ACSR conductor, 6 to 1. The choice of the
optimum vibration damper by tests is a special problem. The damper has to be able to decrease the vibration
intensity, especially at the most endangered frequencies and amplitudes of the transmission line conductor
in question.

Considering energy absorption to determine the vibration intensity of conductors with and without dampers
is not reliable. No quantitative relationship exists between the energy absorption and the lifetime of the
conductor. The dynamic stresses and the number of cycles cause fatigue failure. Application of the Energy-
Balance Principle, used to determine the energy input and the free-span vibration amplitudes, is based on
the frequency calculated with the so-called Karman formula. This formula is valid only for wind tunnels with
laminar wind flow, not for transmission line conductors exposed to wind flow with variable wind velocity and
turbulence along the several miles-long sections8. In addition, the value of the self damping of the conductor
needed for this calculation fluctuates considerably as well as the determination of the maximum stress near
the hardware.

The Frequency Behavior. The frequency behavior of the conductor with and without damper will be
determined by evaluating the relevant data. Harmonic analysis of the vibration motion is recommended to
obtain the exact frequencies that occurred.

2/4 2014.01.06. 22:06

Aeolian Conductor Vibration Control http://tdworld.com/print/substations/aeolian-conductor-vibration-control

The Danger Ratios. The danger ratio of each frequency and vibration angle is given by the previously
mentioned danger factor D = ni/Ni. Figure 5 shows the danger ratios by presenting the D values of an ACSR
conductor (210/35) as a function of the frequency. The frequency with the highest occurrence in the range of
20 to 30 Hz is identical to the most endangered. The small vibration angles occur often, but most are
endangered here at an angle of 6 Min for the conductor investigated.

Accuracy of the Lifetime Calculation. How accurate the lifetime calculation needs to be depends on the
reason for the investigation. High accuracy is required in the strain conversion factor (F) and in the SN
curve. The maximum stress also needs to be considered in lifetime calculations. The recorders mainly
measure the bending amplitude near the suspension clamp, not at the conductor location where the
maximum stresses occur. In this case, complementary laboratory tests are needed. Also, the linemen and
technicians must learn how to install and handle the recorder, including how to set the readout with the
relevant software and how to evaluate and interpret the data. They must also learn to use a light and stiff
holding device to secure the recorder on the suspension clamp to avoid resonance frequencies. For accurate
lifetime calculations, the test time needed to represent one-year operation should be at least three months
depending on the climatic conditions. For simultaneous tests of conductors or hardware and dampers at the
same tower on different phases, four-week periods of measurements are sufficient. Laboratory Tests

Laboratory tests determine the relationship between maximum dynamic stress and stress at the suspension
clamp, the correct factor for converting bending amplitude into bending stress, the behavior of dampers at
the most endangered frequency-amplitude range and the SN curve for the conductor/clamp in question.

In a test-rig arrangement, the open cable was approximately 45-m (49.2 yd) long and was fastened on both
sides and suspended in the middle. Armour rods, spiral/fittings and dampers were also installed on the line.
The excitation by a shaker on one end of the conductor was carried out with test series of approximately 10
frequencies and three amplitudes for each frequency. The laboratory tests yield an easier and more accurate
determination of the maximum stresses than field tests.

Besides the frequencies, the amplitudes at the antinode and node positions of three loops and the locations
of the nodes were established, as well as the power input at the shaker and its force, velocity, phase angle and
impedance plus free-span vibration angle. Laboratory test results will improve the accuracy of the lifetime
determination and will help to overcome some doubts about the reliability of the field tests. When
conductors fail because of fatigue, the failure locations usually are confined to places where hardware is
fastened to the conductors. The dynamic stresses (smax) at the hardware, apart from the suspension clamp,
are mostly higher than the stresses (ss) at the edges of the suspension clamp. The relationship F = smax/ss
varies considerably with frequencies and amplitudes and also how they affect the lifetime. Figure 6 shows the
data for the four Stockbridge dampers, which four manufacturers supplied for an OPGW cable in Germany.
The variation of (F) lies between 2 and 25, and the values of cable plus damper No. 4 exceeds the values for
the conductor without damper, especially for frequencies above 15 Hz. For the evaluation of the different
dampers, the consideration of the maximum relationship (stress power), called Stress-Power-factors Si, for
each frequency vibration angle pair, are recommended5. Lower values for Si are better; s is stress per watt.

Figure 7 represents these data for the four dampers. The lump sum of the values (SSi) for all frequency-
amplitude pairs leads to a first evaluation of each damper plus cable system. The figure shows that the
stress-power factor of damper No. 4 is worse than the cable without damper. Evaluation one considers all
frequency-amplitude pairs equally without considering their occurrence on the vibrating conductor (Table 1).

Field and Laboratory Tests The final evaluation has to consider the unequal occurrence of the frequency-
amplitude pairs and especially its unequal influence on the lifetime, represented by the unequal danger
factor (D) based on outdoor vibration tests. These tests, involving the strain-power factor (S) for all
frequency-amplitude pairs determined by the laboratory tests, will be readjusted by the danger factors (Di),
determined by field tests. In practice, the calculation will be done by multiplication of Di and Si and leads to
an evaluation factor (Vi) for each frequency-amplitude pair. The lump sum of these factors (SVi) is the
measure of the final evaluation 2 of the conductor-damper in question. Low values of (Vi) are desired. Table

3/4 2014.01.06. 22:06

Aeolian Conductor Vibration Control http://tdworld.com/print/substations/aeolian-conductor-vibration-control

1 shows the evaluations of the four dampers investigated for the OPGW cable. Damper No. 3 was the
apparent winner in the evaluation 1, but the final evaluation 2 shows that damper No. 2 is the actual winner.

Finally, the behavior of damper Nos. 2 and 3 on the OPGW cable was field tested at the same tower locations
as the cable without dampers. The results of this test confirmed the final evaluation method based on
outdoor tests, characterized by the danger rates, and in combination with the laboratory tests characterized
by stress-power factors.

Conclusions As illustrated by some examples, comprehensive investigations, including field and laboratory
tests, can lead to more reliable and cost-effective vibration control. Application of these lifetime calculation
methods for conductors with or without dampers can lead to further improvements in estimating the lifetime
of old lines and evaluating the design of new lines.

Note: The author thanks Jim Poffenberger, a U.S. consultant, for his help in editing this article. References 1.
CIGRE 22 WG 04, Recommendations for Evaluation of Lifetime, Conductors 1979, ELECTRA No. 63 1
S.103-145. 2. CIGRE 22 WG 04, Guide for Endurance Tests of Conductors + Clamp 1985, ELECTRA No. 100
S.77. 3. IEEE, Standardization of Conductor Vibration Measurements 1966, IEEE Tr.Action, 31TP65-156. 4.
Miner, M.A., Estimation of Fatigue Life with Particular Emphasis on Cumulative Damage 1959, Fatigue Life:
McGraw Hill Co., Inc. 5. Buckner, W.F., Improved Diagnostic Techniques for Vibration Dampers 1991,
Transmission Distribution International. 6. Poffenberger, J.C., Swart, R.L., Differential Displacement and
Dynamic Conduct, Strain 1965, EIII Paper 31 TP 65-158. 7. Papailiou, K.O., Die Seilbiegung mit einer durch
die innere Reibung, die Zugkraft und die Seilkruemmung veraenderliche Biegesteifigkeit 1995, Doctoral
Thesis Nr. 11057, ETH Zurich. 8. Wolfs M., Mouchard, P., Bückner, W.F., Improved Reliability of Overhead
Lines by Controlling the Aeolian Vibration of Conductors 1995, International Symposium on Cable Dynamics
in Liege. TDW Dr. Walter F. Bückner began his professional career in airplane engineering and cooperated in
the design of the Heinkel 164 jet plane. He joined Siemens in 1948 and became head of the mechanical
calculation and line design in 1950. Dr. Bückner received the Doktor Ingenieur degree from the University of
Munich in 1956. He became chief manager for design, sales and construction in 1970 and joined GEA
Gesellschaft in October 1970 as CEO. Dr. Bückner established a transmission line consulting firm in 1985. He
serves as expert advisor to CIGRE Committee 22-Overhead Lines and was the international convener of
Working Group 04-Endurance Capability of Overhead Line Conductors from 1965 until 1986.

Source URL: http://tdworld.com/substations/aeolian-conductor-vibration-control

4/4 2014.01.06. 22:06