Table 2.5. The degree of correlation (denoted by the maximum 



value of the correlation coefficient p^g) in the wake of a 



vibrating cable; from reference 22. 





Reynolds number 



= 1300 





Frequency ratio, ///j 



Displacement amplitude, 



2YID 



Degree of correlation, Pab.max 



0.9 



0.1 





0.89 





0.2 





0.94 





0.3 





0.96 





0.4 





0.96 





0.5 





0.92 





0.6 





0.92 



1.0 



0.1 





0.88 





0.2 





0.96 





0.3 





0.96 





0.4 





0.97 





0.5 





0.97 





0.6 





0.96 



A further study of the coherence of the vortex shedding in the wake of a flexible cable was con- 

 ducted by Ramberg and Griffin (23). It was found from spectral analysis of the vortex shedding that 

 the component of the fluctuating pressure or velocity at the cable frequency was 15 to 20 times the 

 component at the Strouhal frequency. The spanwise correlation coefficient p ^g attained values compar- 

 able to those shown in Table 2.5, as shown in Fig. 2.12. Note that magnitude of the spectral com- 

 ponent C„ in Fig. 2.12 follows the distribution in displacement amplitude along the cable; this and other 

 similar observations suggest that predictive models for vortex-excited oscillations can be correctly based 

 upon local conditions, i.e. lift and drag forces, so long as the vortex and vibration frequencies are 

 locked-on. As a practical matter it is reasonable to assume that at large displacement amplitudes the vor- 

 tex shedding is coherent in degree and extent between nodal points on a vibratmg flexible structure. As 

 a nodal region is traversed lengthwise there is a 180 degree phase shift; the vortex shedding is again 

 coherent in degree and extent but is shifted in phase from neighboring sections of the cable. The 

 steady and unsteady hydrodynamic forces (drag, excitation, damping, etc.) vary with local displacement 

 amplitude and are not constant over the half wave length of the cable. 



Similar findings were obtained by Novak and Tanaka (19) and Howell and Novak (20) for experi- 

 ments conducted with cylinders vibrating in smooth flow. Results similar to those in Fig. 2.11 were 



16 



