Minutes
Cable Aerodynamics Meeting

Objective:Cable aerodynamic vibrations have become more of concern to both bridge and electrical engineers. This meeting aims to exchange up-to-date information on wind-induced vibration of cables and seeks to obtain suggestions for finding more reasonable and reliable vibration control.

1. The meeting was called to order at 10:30 am by Masaru Matsumoto and Nicholas P. Jones. An attendance list was circulated (attached).

2. Report of summary of responses to survey circulated by M. Matsumoto:

3. Video presentations: A number of guests presented short videos outlining either field conditions or lab simulation of stay cable vibration. Where available, short descriptions are included below.

• Cables of cable-stayed bridges in Japan (T. Yagi, Japan)

This videotape includes three examples of the wind-induced cable vibration of the cable-stayed bridges in Japan. The first example is the cable-stayed bridge with the main span length 184.2m and it has the single plane cables. The weather condition was rain and the wind speed was about less than 20m/s. The double amplitude of the vibration was 50-60cm. The dampers were installed. The second one is the cable-stayed bridge with the main span length 350m and it has the 2-plane cables. The weather condition was rain and the mean wind speed was about 10m/s. From the observation data, the maximum double amplitude of the vibration is 237cm. The dampers were also installed. The third one is the bridge with the main span length 420m and it has the 2-plane cables with twin cables. The weather condition was rain and the double amplitude was 40-100cm. The wire connection was tried and the spacers we re installed. The type of these wind-induced vibrations seems to be the rain and wind induced vibration. The details of these vibrations can be found in a article, which is "Wind-induced cable vibration of cable-stayed bridges in Japan" by M. Matsumoto, K. Yokoyama, T. Miyata, Y. Fujino and H. Yamaguchi in Proceedings of Canada-Japan Workshop on Bridge Aerodynamics, NRCC, Ottawa, Canada, 1989, pp.101-110.

• Stayed cables of Erasmus Bridge (A.J. Persoon, The Netherlands)

In November 1996 the cables of the Erasmus Bridge showed vibrations under rain and windy weather. Although the main span is of a medium size (285 m) the longest stay-cables are around 300 m because this bridge has only one pylon. Water rivulets were observed on the polyethylene casings of the cables. At medium windspeed of around 12 to 16 m/s at 30 degrees with the axes perpendicular to the bridge the cables showed vibrations of .5 to .7m (2 to 3 times the diameter) mostly in the second mode shape.

After consultation with Japanese experts the Public Works of Rotterdam (location of the bridge) proposed to install hydraulic dampers as a countermeasure. The minimum damping coefficient (percentage critical) should be .5% but to avoid any risk .8% should be the lower limit.

Full-scale tests were performed with prototype dampers to verify the required damping. A number of representative cables were excited by means of a small hydraulic actuator. Using the well-known 90 degrees phase criterion the cables were set into a number of natural frequencies up to three Hertz at sinusoidal excitation. By switching off the excitation decay's were obtained from which the damping factor was estimated.

The actual dampers were installed within one year later. Up till now no vibrations of cables (and bridge deck) were reported.

Because of traffic passing the bridge we (NLR) had some problems by adjusting the phase criterion as the cable tension changed and with that the natural frequencies.

So it is advisable to close the bridge when such type of measurements are carrying out.

• Cables of Cable-stayed bridges in U.S.A. (P. Sarkar, U.S.A.)

In April 1997, TxDOT officials recorded this footage showing large-amplitude vibrations of stay-cables at the Veterans Memorial Cable-Stayed Bridge located near Port Arthur, Texas on State Highway 87 (150 km east of Houston). This bridge was opened to traffic in 1991 (main span 191 m; 4 planes of 28 stay-cables arranged in vertical harped-configuration). The stay-cable vibrations were observed at wind speeds below 15 m/s while it was raining. At any one time, a number of adjacent stay-cables were observed to be simultaneously excited.

TxDOT officials recorded this footage at the Fred Hartman Cable-Stayed Bridge located near La Porte, Texas on US146 (25 km south of Houston). This twin-deck bridge was opened to traffic in 1995 (main span 381 m; 4 planes of cable per deck arranged in fanned configuration). The cable that is observed to vibrate here is the longest among the 24 cables that comprise one plane of stay cables, and hence it has the lowest natural frequency among all the cables (first natural frequency: 0.67 Hz). It was observed to vibrate in the third mode. The winds were gusting up to 22 m/s without the presence of rain. This observation showed that cables can vibrate even without the presence of rain at wind speeds above the range of wind speeds where rain-wind induced vibrations usually occur. The next footage shows rain-wind induced vibration of stay-cables at the Fred Hartman Bridge at wind speeds below 15 m/s in the presence of rain. A set of adjacent cables was observed to excite simultaneously.

• Rain-wind induced vibration of cables (H. Ruscheweyh, Germany)

Video about the oscillation of the water rivulets.

The oscillation of the water rivulets at a cantilevered aeroelastic model (diameter 100 mm, length 2 m) has been observed and documented by a video film (C.Verwiebe). At lower wind speed the model is excited to cross-wind vibrations and the rivulets oscillate asymmetrically in front of the meridian.

By increasing the wind speed the two rivulets are shifted to a position behind the meridian and the model vibrates in a in-wind mode. Both rivulets oscillate symmetrically to and fro.

• Hanger ropes cable of Akashi Kaikyo Bridge (H. Katsuchi, Japan)

The video showed an example of wind-induced vibration observed at the hanger cables of the Akashi-Kaikyo Bridge.

In the Akashi-Kaikyo Bridge, polyethylene-covered round hanger cables were used instead of conventional CFRC spiral cables, except of some short-length cables. The length of hanger cables ranges from 26 m to 200 m with a diameter of 87 mm. Since two hanger cables are aligned at one fixing point, wake-galloping was feared to occur at the designing stage. However, it was reported that wake galloping would occur if the distance of two cables ranged from 1.5 to 6 times diameter, and the distance of two hangers of the Akashi-Kaikyo Bridge was set to 9 times diameter. Therefore, no-occurrence of wake-galloping was judged in the Akashi-Kaikyo Bridge.

During construction stage, two kinds of vibration were observed. One is vortex-induced vibration at a low wind speed. The other is unknown large-amplitude vibration. The vibration typically occurred at a wind speed of more than 12 m/s, on 100 - 200m long leeward cables, with in-line vibration with an elliptical orbit. 1st - 7th modes were observed and their amplitude reached 8 times diameter at the maximum.

In order to investigate the mechanisms and countermeasures, wind-tunnel investigation was carried out. A pair of hanger cables of 140m long were modeled as a section model with a natural frequency of 0.71 Hz. A windward cable was fixed and leeward one was supported by springs. The result showed that vertical movement on the leeward cable dominated at a low wind speed, but lateral movement gradually dominated as wind speed increased. Based on these results, the unknown vibration was identified wake-induced flutter which was often observed on power cables.

Wake-induced flutter occurred at restricted angles of attack in both the full scale and wind-tunnel testing. Finally, winding 2 helical ropes, which effectively reduced the amplitude in the wind-tunnel testing, was adopted as a countermeasure for wake-induced flutter. Countermeasure work is now in progress.

• Back-stay bundle of a suspension bridge (E. Hjorth-Hansen, Norway)

Violent response of suspension bridge backstays; Askoy Bridge, Bergen, Norway

During construction when only main cables and catwalks were in place, the structure was hit by the storm on new year’s day, 1992. The temporary spacers fixing the patent lock-coil main cable members into an open, rectangular bundle disintegrated in the backstay part. So, the individual cable members got free and hit one another with a sound resembling that of advancing cavalry.

After the storm the "wounds" were healed in place by paint etc.

The film taken by Norwegian Public Roads Administration shows a case where any prediction method (wind-tunnel modeling or computational fluid-structure interaction) are likely to fail.

• Transmission lines (K. Kimura, Japan)

Galloping of ice-accreted twin bundled conductors overhead line observed at the Tsuruga Test Line of Kansai Electric Power Company in Japan was shown on the video. The span of the line is 234m and the peak to peak amplitude seems to be approximately 1m. The wind speed was probably around 15m/s. The accreted ice was soft rime on the windward side of the conductors and its length was probably around 10cm.

Because the unsteady aerodynamic force characteristics acting on the conductors while they are vibrating with large amplitude are not known, a series of experiments is now carried out by using a newly designed apparatus that can measure the aerodynamic forces under forced harmonic vibration with amplitude of +-500mm in vertical and horizontal and +-55deg. in rotational direction.

4. Panel and open discussion:

Panelists: G. Diana (Italy), Y. Fujino (Japan), E. Hjorth-Hansen (Norway), P. King (Canada), H. Niemann (Germany), G. Piccardo (Italy), H. Ruscheweyh (Germany), P. Sarkar, (U.S.A.), J. Xie (Canada)

Potential topics suggested included:

• Practical experiences of cable vibrations and damages
• Research works
• Site measurements
• Wind tunnel tests
• Analytical approaches
• Vibration control

A number of the panelists gave short presentations/remarks:

• Y. Fujino described recent efforts using active, semi-active, and magnetic damping in a number of applications.
• G. Diana noted that considerable investigation into cable dynamics has occurred in the transmission line industry, and that we should ensure links with relevant CIGRE and IEEE working groups. He also noted that damping may not work for wake galloping problems, whereas vibration absorbers may be appropriate for ice-related galloping.
• H. Ruscheweyh described a case where vibration absorbers/dampers worked well for rain-induced vibration in near-vertical hangers. Tunable fluid dampers produced considerable reductions in amplitude. (Later discussion generally supported this type of solution for this application.)
• P. Sarkar showed and discussed some wind tunnel testing of stay cable sections with simulated rain effects, including a discussion of several aerodynamic mitigation methods.
• G. Piccardo described an analytical technique for the nonlinear analysis of galloping cables.
• M.Matsumoto asked the questions: 1. Is rain necessary for large-amplitude vibration? 2. Is a log-dec damping of 0.02 as frequently proposed sufficient? Data were showed indicating the presence of beating oscillation (multiple modes) as well as single mode.

General discussion ensued, with the following points made or issues raised:

• Log-dec damping in stays as low as 0.0005 have been measured. (Verwiebe)
• It was suggested that we focus on three areas: 1. Understanding the mechanics of the phenomenon; 2. Modeling (including field and laboratory experiments); and 3. Mitigation techniques. (Fujino)
• The group might provide a forum to share and/or publish information on, for example, the variability of responses and response types, and to share information about cable parameters and characteristics (e.g., frequencies, dampings, mode shapes, wind directions, etc.) (Xie, Cooper, Ruscheweyh, Persoon). Even dissemination of (initially) worst-case data and associated parameters was encouraged (Ruscheweyh). The use of a web site was suggested, with Soren Esdahl volunteering to help in this effort. (Larsen, Jones, Sarkar)
• The fact that the cables are part of a large, complex, dynamic system should not be overlooked. (Larose)
• What are the critical parameters? (Larsen)
• Addition of damping (e.g., by using a Scruton number criterion) was pointed out as a potential solution (Diana, Sarkar)
• Performance of damper systems on Erasmus (Persoon) and Huntington (Jones) was reported as successful. Failure of a restrainer system in Texas was reported (Jones). Aerodynamic solutions were raised and discussed (Sarkar, Matsumoto). It was noted that aerodynamic solutions require little maintenance vs. mechanical. It was also noted that on the second Severn bridge, addition of cross ties to mitigate cable vibration led to other (broader) vibration problems. (Larose)
• The formation of specific working subgroups was suggested, and need to coordinate with the power line community reiterated. (Diana)
• Attention was brought to the upcoming International Conference on Cable Dynamics in mid-August in Trondheim.

5. Closure (N. Jones, M. Matsumoto)

The chairs thanked all the attendees for their participation, and reiterated that active participation would be needed in the future to assure the success of the working group. Minutes will be distributed in PDF format, and efforts will be made to get the web site up and running as soon as possible.

The meeting was adjourned at 12:30 pm.

List of Attendees:

Name, First Name, Second Affiliation e-mail