TR No. 22 
is of the order of 50-70 milliseconds, In making the turbulence measure- 
ments reported herein, however, a towing velocity of approximately 400 cm-sec™ 
was superimposed on the turbulent velocity field. Therefore it was necessary 
to determine the response of the current meter to a step function change in 
velocity superimposed on a mean velocity. Wind tunnel measurements of the 
response time of the current meter are described in appendix A. It was found 
that the response time for a relatively small step function change in water 
velocity varies inversely with the mean velocity such that the product of the 
response time and the mean velocity (the response distance) is a constant with 
a value of 0.97 cm. The frequency response of the instrument is determined by 
the response time; the instrument is insensitive to variations in velocity 
occurring at frequencies greater than 
] 
Ms 
sisted << Srp re (6) 
1 
Assuming that Taylor's hypothesis is applicable, that is, 
aU 
a+ i rials eae 
(7) 
this corresponds to a wave number of 
| <¢ a K elle 
ee Peane ies = (8) 
which, from the previous measurements of response time, is 
balls mie =| 
Fe Ae Gua eas icine 
Thus the current meter hag the capability for measuring turbulence over the 
constant range of wave numbers from 0 to 0.103 om71, regardless of the mean 
velocity superimposed on the turbulent field by towing. (Actually the value 
given for Kay, is optimistic because of the size of the current meter, 15. cm 
long; a more reasonable value is of the order of 1/150 cm = 0.0068 em~,) 
Since spectral analysis of turbulence is more correctly performed with respect 
to wave number than frequency, this is an important result. 
