PSI - Issue 84

Allan Larsen et al. / Procedia Structural Integrity 84 (2026) 1103–1110

1108

Extent of water film

Wind

Rain

Fig 6. Hanger cable model vibrating in the wind tunnel (left) and oscillating wind tunnel model (center). Overlay of two video frames captured at the extreme excursions of the model. Close-up view of the water streaks water film forming on the cable surface (right).

Unlike vortex induced vibrations, for which the wind speed and the frequency are connected through the Strouhal number, and the amplitude is governed by the Scruton number, there are no similar scientifically proven relations in case of hanger cable galloping vibrations related to the action of rain and wind. In connection with the design of dampers for mitigation of the rain-wind vibrations for the 1915 Çanakkale bridge it was decided to conduct a wind tunnel test to establish some experimental guidance for the design of mechanical dampers capable of mitigating the vibrations. Fig. 6 (center) shows the test set-up in the 2 x 2 m climatic wind tunnel at FORCE Technology. The cable model was made from a 0.11 m diameter HPDE tube suspended vertically from a flexure yielding a model natural frequency of 2.2 Hz. The model is thus a 1:1 scale model of galloping vibrations observed in the field. The rain was simulated by a vertical line of water spray nozzles located 2 m upwind of the model. The model surface was treated with polyvinyl alcohol (Ghosenol) to simulate the surface adhesion conditions of a soiled / oxidated cable surface. A close-up view of the model surface at maximum vibration amplitudes is shown in Fig. 6 (right). It is noted that rain droplets hitting the otherwise dry upwind portion of the cylinder form streaks starting from the front stagnation line and sloping downwards under the influence of gravity. The streaks add water to the downward running water film extending downwind form the separation bubble covering an angle of approximately 45 – 120 deg. from the stagnation line. Video recordings of the model reveal that the position of the water film is stationary relative to the surface of the oscillating model and does not oscillate independently as sometimes is the case for inclined models, Verweibe (1998). The recorded non-dimensional vibration amplitudes as function of non-dimensional wind speed for a Scruton number ≈ 5.8 is shown in Fig. 7 (left). The amplitude vs amplitude response for rain wind conditions is shown in