210 ANNUAL REPOET SMITHSONIAN INSTITUTION, 1915. 



frequency is used, 500 per second, and that there is no laminated iron 

 used in the construction of the apparatus. 



The secret of this lies in the fact that the armature has substan- 

 tially no self-induction, and no eddy currents are generated in the 

 apparatus. This is because the copper tube forms, as will be seen, 

 the short-circuiting secondary of a transformer, of which the arma- 

 ture winding is the primary. 



This eliminates the self-induction of the armature winding. In 

 addition the upper and lower portions of the winding are wound 

 in opposite directions, and therefore there is no mutual induction 

 between the field coil circuit and the armature circuit. . With this 

 construction the amount of magnetic leakage in the armature cir- 

 cuit is very small, only a trifle more than if the armature core were 

 of wood, and as there is no alternating magnetic flux in the iron, 

 there are no eddy currents. 



As regards the capacity in kilowatts of this apparatus, it is large. 

 The armature, being wound in grooves in the armature core, so as 

 to withstand the mechanical forces acting upon it, is well cooled. 



The copper tube has no insulation to be affected, and on account 

 of its large cooling surface and high permissible temperature of 

 operation can carry very high currents without injury. 



When the oscillator is placed on a vessel or hung overboard from 

 a lightship, a large water-tight diaphragm is attached to the oscil- 

 lator. This particular type of oscillator was first tested by suspend- 

 ing it in 12 feet of water at the Boston Lightship and the signals 

 were heard plainly with a microphone lowered overboard from a 

 tug at Peaked Hill Bar Buoy, 31 miles away. Since that time tests 

 have been made with oscillators installed in the fore-peak tank of 

 the Devereux^ a collier of the Metropolitan Coal Co., and also with 

 an oscillator mounted on a diaphragm made part of the hull of the 

 vessel. The signals have been heard upward of 20 miles from the 

 Devereux running at her regular speed of 8 knots. Full power has 

 not been employed on any of the tests, and it is more than i^rob- 

 able that much longer distances can be obtained in the future. 



In addition to the tests already described the oscillator has been 

 temporarily installed on submarine boats and proved itself of im- 

 mense value, and demonstrated that a flotilla of submarines equipped 

 with oscillators will be able to make a combined attack on an enemy, 

 only one needing to show its periscope in order to direct the others, 

 or all of them can be directed by the mother ship. It therefore makes 

 possible a whole field of submarine maneuvers heretofore out of the 

 question, and perhaps, most important, it removes the principal 

 danger these boats have had to face, the risk of being run into. 



So much for the apparatus when in use as a sound generator. The 

 signals produced by the oscillator can, of course, be received by 



