ticity of the cable to accommodate a difference between ship -imparted 

 motion as shown in Figure 2 and depth variation of the weight. For a dead 

 weight the steady tension always equals the instantaneous tension except for 

 a resonating periodicity in the ship motion. If we now add a pair of light- 

 weight tubes which temporarily entrap a water mass of say double the dead 

 weight attached, we have the possibility of instantaneous tensions as greatas 

 3 times the steady tension and probably as low as l/3 the steady tension 

 upon release of the upward pull. Naturally the same possible resonating 

 periodicity exists, and I would imagine that these cases should be treated 

 separately. However, for nonre sonant conditions we clearly have the 

 possibility of damped vertical motion for the parallel tubes. For the sake 

 of easy recollection we can calculate casually that the ship -imparted 

 motion may be cancelled by as much as the cable strain difference due to an 

 excess as deficiency of tension equal to W sec 9 



DESCRIPTION OF DEVICE AND ITS TEST IN A MODEL BASIN 



A device which follows the foregoing principles and, as a natural con- 

 comitant, rigidly constrains the orientation of any arrangement of sensors 

 mounted upon it is shown in Figure 4. 



To fix the plane of the tubes horizontally, most of the dead weight was 

 mounted on a rigid extension about 3 ft below the suspension point providing 

 a restorative couple that keeps the tubes horizontal. The vertical vane well 

 behind the centroid constrains the tubes to align with the towing direction. 

 A small nearly horizontal vane can be adjusted so that even over a speed 

 range of several knots the ;init does not tilt downAvard or upward but keeps 

 the tube axes in the horizontal plane. 



During tests run in a model basin and in subsequent deep water sound 

 velocity surveys, two sound velocimeters are mounted at the front end of 

 the device. Since the meter designated Lockheed uses an unfolded path, it 

 could be oriented in such a way that, in addition to the sound velocity as a 

 function of path position, readings of the two meters give the rate at which 

 water sweeps by the meters. To calibrate this effect and also adjust the 

 horizontal vane for speed range up to 3 knots, the complete unit was tested 

 in January 1964 at the David Taylor Model Basin. Figure 5 shows the 

 general features of the tank used in the test. The whole tank is about 

 2600 ft long and the positions are marked every 30 ft running from west to 

 east. Carriage 5 operates on the eastern half of the southern side of the 

 tank which is divided at the 1100 -ft mark by a wave -making machine. 



For active towing we used the length between 1500 ft and 2400 ft. The 

 overhead cranes can service the carriage only at the eastern end past the 

 2600 -ft marker; therefore all runs had to start and end there. Our towed 

 vehicle is 10 ft long and was suspended at an interior part of the carriage 

 from one of its overhead tubular joists. Because of the turnaround required 

 at the unserviced end of the tank one probably cannot test a towing device 

 that is longer than 20 ft. 



59 



