where f (<j>) is the normal hydrodynamic loading function, 



f (<j>) is the tangential hydrodynamic loading function, 



R is the cable drag per unit length when the axis is 90 to the 



flow (R = *2pC_,V 2 d) , 



p is the density of seawater, 



C is the normal drag coefficient based on cable diameter, 

 R 



V is velocity, and 

 d is cable diameter. 



As seen in Equation (2), F and G are expressed as the products of two terms. 



Essentially the process of determining the hydrodynamic loading functions is one 



of assuming various forms of f (<J>) and f (<f>) until, through the regression analysis, 



a value of C (as a function only of Reynolds number) is obtained for which the 



computed configurations match those measured. Since the fitting process is based 



on the products f (<j>)* R and f (<(>)• R, it appears that there could be a family of 



solutions rather than a unique solution. However a constraint on the range of 



solutions is that the C values be plausible. Nonetheless it must be recognized 

 R 



that the derived value of C may be different from that which would be obtained 



K 



by physical measurement and, therefore, it is not valid to imply that this C is 



K 



a characteristic of the cable independent of the hydrodynamic loading functions. 



The above caveats notwithstanding, there is confidence in this technique for 

 developing the hydrodynamic loading functions and in applying these functions to 

 the configuration predictions of ribbon towcable provided: 



1. the ribbon towcable design is similar, 



2. cable diameter is not greatly different from that measured, and 



4 5 



3. the Reynolds number is within the range from 5.2 x 10 to 1.28 x 10 . 



4 ' 

 In conformance with established methodology, f (<|>) and f (<t>) are represented 



in this analysis by selected terms from the following trigonometric series: 

 f (<f>) = A + A cos<J> + A„ sin<J> + A cos2<J> + A, sin2c|). 



