Observation of Karman Vortex Street on Flowing Soap Film behind Cylinders at Low Reynolds Numbers

Introduction. The Karman-vortex street formed behind two-dimensional bluff bodies has attracted interest because of its beautiful flow pattern. It goes without saying that this beautifulness results from the alternative vortices regularly shed from the body. Vortex shedding induces aerodynamic force perpendicular to the mean flow. This force causing mechanical vibration of the model is accompanied with aeroacoustic emission. The Karman vortices in the wake of twodimensional obstacles appears at wide Reynolds numbers in laboratory and also in nature. Thus, numerous studies associated with the Karman vortices have been made from an engineering point of view. Velocity deficit is formed behind the body, but its deficit rapidly recovers due to active vortex motions, and results in widening the wake in the downstream direction. As a result, the formation of Karman vortices induced in the near wake of the model is mismatched to the far wake. Taneda [1] visualized the wakes of two-dimensional cylinders and flat plates in a water towing tank at Reynolds range between Re=50 and 150 based on the towing speed and the representative length of the model. He observed the secondary vortex street in the far wake. Prior to secondary-vortex formation, the primary Karman-vortex street was decayed. It is, therefore, suggested that there is no relation between the prominant frequency of the secondary vortex street and the primary frequency. Using smoke-wire and hot-wire techniques, Matsui's group [2-5] visualized the Karman-vortex street and made quantitative measuremnts behind a cylinder at Re=100-160 in a low speed wind tunnel. The secondary-vortex street is formed resulting from pairing that is the mechanism for the change in scale. The passing frequency of the secondary vortices is not always exactly one-half because not all of the vortices pair up: a few are left stranded. Also, by means of acoustic forcing of one-half or one third of the Karman vortex frequency, they observed pairing or tripling of the primary vortices, showing direct evidence that the primary vortex street was rearranged in the far downstream wake. Cimbala et al.[6] observed the secondary vortex street in the far wake behind cylinders at Re=140, following exponetial decay and resultant annihilation of the primary vortex street. This observation is similar to that by Taneda [1]. They also reported that the passing frequencies of the secondary vortices are in good agreement with those predicted by two-dimensional locally parallel inviscid linear stability theory taking account for spatial growth of wake width. Plausable reason why the primary Karman-vortex street exponentially decays still remains. Visualizing the flow past a low aspect-ratio model at around Re=100 with the smoke-wire method, Mizushima et al. [7] observed annihilation of the primary Karman vortex street and then secondary streets downstream. It is reported that the flow in the spanwise direction along the model axis is three-dimensional. They also analyzed characteristics of the basic wake flow in combination with linear stability analyses and numerical simulations. Observable region of the primary and secondary vortex streets in the streamwise direcion was mapped out. In the present paper, we made comparative experiments using a water towing tank and a soapfilm tunnel to visualize the formation of vortices shedding from a triangle cylinder and a circular cylinder for comparison at low Reynolds number.