Testing Ship Models in Transient Waves 



100 



.2 .3 4 .5 .6 



Frequency in eyries per second 



,7 .6 .9 1.0 



Fig. 31 - Comparison of transient wave height 

 spectra measured with and without model in 

 water at zero speed 



To assist in these investigations and to speed up data analysis, digital com- 

 puter programs are being written for Fourier transform computations. An im- 

 portant part of these programs will be the "smoothing" of the transient records 

 prior to transformation, or the multiplication of all time histories by a quantity 

 which is unity over the duration of the test and eases to zero at the beginning 

 and end of the test. This smoothing, standard in spectral analysis, will consid- 

 erably reduce the effect of residual noise in the water near the end of a run. 



After more experience has been gained in forward speed runs, a critical 

 analysis must be made to determine the effect of surge variations in transient 

 analysis. This very knotty theoretical problem might require conducting wave 

 transient tests at reduced wave height levels in order to avoid the time distor- 

 tion of motion records. 



The particular problems associated with tests in astern waves must be re- 

 solved. Unfortunately, a given frequency measured in the water with respect to 

 the moving model can originate from astern waves at three different wave 

 lengths. This ambiguity may force the use of transient waves with energy in 

 more limited frequency bands, the use of multiple wave probes to make use of 

 phase information, or both. 



Special Use of the Transient Testing Technique 



In one interesting use of a transient test, the variance of all motions in a 

 given unidirectional random seaway can be found directly without need for 



535 



