It was concluded from a study of the static and dynamic stress levels within the Cognac piles dur- 

 ing driving that the large cross flow displacement amplitudes (of the level shown in Fig. 2.24) would 

 triple the stresses from a corresponding stationary 130 m (426 ft) long pile. The apparent steady drag 

 coefficient on the oscillating pile was Cq = 2.12; this is an amplification of 230 percent from the drag 

 coefficient (C^o = 0.93) when the pile was restrained. A fatigue life of four days was predicted for a 

 stabbed pile (without a hammer attached) when it was exposed to a current of magnitude 0.46 m/sec 

 (0.9 kt). Additional details and assumptions pertaining to the study are discussed in reference 40. 



A flag-type or flexible tail fairing type of wake interference device was developed to suppress the 

 cross flow oscillations. Such a device was tested successfully on the model piles, but the particular 

 configuration was chosen because of the nearly unidirectional currents at the Cognac site (40). Few 

 actual problems were encountered during the field installation, but in the case of one pile typical peak- 

 to-peak displacement amplitudes of 3 m (9.8 ft) were measured. The cause of these cross flow oscilla- 

 tions was attributed to alternate vortex shedding. 



A program of experiments recently was conducted to assess the effects of shear on vortex shed- 

 ding from smooth and rough cylinders at large Reynolds numbers (31). The experiments were con- 

 ducted with a cylinder of aspect ratio L/D = 16 at Reynolds numbers in the range of 1.5 (10^) to 3 

 (10^) in order to assess the minimum shear (as denoted by the shear parameter p given above) at 

 which the characteristic lengthwise cellular vortex shedding pattern was initiated. An incipient cellular 

 pattern of vortex shedding was observed at the weakest shear gradient, /3 = 0.007, and persisted in 

 stronger form over the test range to shear levels given by ^ = 0.04. Most of the test runs, however, 

 were carried out at values of the shear parameter, p = 0.007 to 0.02, which are representative of ocean 

 site conditions. The results obtained by Peltzer and Rooney provide a reasonably extensive data base of 

 circumferential mean pressure and vortex shedding frequencies for smooth and rough circular cylinders 

 at subcritical, critical and supercritical Reynolds numbers. 



Peterka, Cermak and Woo (41) are conducting experiments to study the vortex shedding from 



large aspect-ratio (L/D = 12 to 128), stationary and vibrating circular cylinders and cables in a linear 



25 



