ABSTRACT 



A surface ship model was constrained to perform forced heaving 

 oscillations in still water and the resulting lift forces and pitching mo- 

 ments obtained. The influence of forward speed, frequency and ampli- 

 tude of oscillation were investigated. 



The added mass and damping of the ship motion were determined 

 as well as the coupling moments (the pitching moments due to heaving 

 velocity and acceleration). Comparisons are made between the various 

 theoretical prediction procedures and these experimental results. The 

 significance of the observed nonlinear damping forces is discussed. 



INTRODUCTION 



In the past five years the Transactions of the Society of Naval Architects and Marine 

 Engineers has included a succession of papers dealing with the motion of ships in waves. 

 These papers, by Weinblum, St. Denis, Pierson, and Korvin-Kroukovsky, present the tools 

 necessary for the analytical determination of these motions. 



Uniformly, the assumption is made that the forces on the body are of two types. The 

 first is due to wave action and the remaining forces are proportional to the instantaneous posi- 

 tion, velocity, and acceleration of the ship. This yields two linear, second order, differential 

 equations with constant coefficients, which are simple to solve. 



Unfortunately, these methods do not represent exact solutions to the boundary value 

 problem. The analytical procedures require strong assumptions to be made in order to gain 

 simplicity in the solution. Among these are the use of two-dimensional solutions, and the 

 neglect of the free-surface influence, of nonlinear terms and of the speed dependence of the 

 coefficients. Such an approximate procedure may be justified in practical ship problems, but 

 the range of validity must be firmly established by adequate experimental investigations. 



There have been few experimental studies wherein measured model motions have been 

 compared with predicted motions using these analytical procedures. The little data available 

 indicates that discrepancies do exist. No doubt, part of the problem involves the inherent 

 experimental difficulties which make for poor accuracy in the measurements. 



When discrepancies are revealed it is then difficult to decide whether to ascribe the 

 fault to the form of the equations or to the particular coefficients used in the computation. 

 This difficulty can be avoided by experimentally determining the coefficients. The body can 

 be oscillated and either the forces or motions measured. The dependence of these coeffici- 

 ents on speed of advance, frequency, and amplitude of oscillation can be determined. This 

 experimental procedure is quite analogous to that used in aerodynamic stability investiga- 

 tions wherein the wind tunnel balance, whirling arm and oscillator are used to measure the 



