the voltage induced by the water velocity; (2) the alternating field 

 would cause an induced voltage in the water even though the water 

 velocity were zero; (3) the alternating induced voltage would have 

 to be rectified before it could be recorded in a direct current 

 instrument; and (4.) an alternating current power supply cannot be pro- . 

 vided for a field installation as simply as a direct current supply 

 such as portable storage batteries, FoT these reasons it -..r.n iscided 

 to employ direct current in the investigations conducted at tho 3sach 

 jErosion Board's laboratory, with the expectation that the one disadvan- 

 tage associated with its use, that of polarization, could be satisfactorily 

 over come . 



Calculations indicated that an instrument using a toroidal exciting 

 coil, energized by direct cu •'.-a;it, with two pairs of electrodes aligned 

 on perpendicular axes to record simultaneously induced voltages on two 

 General Electric Photoelectric Potentiometer Recorders, would be 

 practicable. However, before assembling an instrument which would be 

 practical for field tests, preliminary tests were made with a small 

 laboratory model utilizing the same basic principles. For this purpose 

 a small field coil and a storage battery were used. Copper wire 

 electrodes coated with colloidal graphite were introduced into a 

 specially built flume which provided a known vrater velocity, and the 

 induced voltage was recorded on a portable d'Arsonval galvanometer. On 

 the basis of the known water velocity and field strength, the anticipated 

 induced voltage was computed to be approximately 0.17 millivol'^s. Dur- 

 ing these tests, deflections of approximately one unit were observed on 

 the galvanometer whose sensitivity was estimated at 0.15 millivolts per 

 vinit. However, throughout the tests varying deflections of the galvano- 

 meter were noted which apparently were caused by a varying potential 

 induced by some other source than the magnetic field and water velocit;/ . 

 This externally ind -cad niasking potential made observation of the sraallei' 

 deflection caused oj the water velocity induced potential very difficult, 

 •^u'>litatively though, these preliminary tests indicated that the induced 

 voltage was directly proportional to the velocity of the water and the 

 separation of the electrodes. 



Another series of tests was then initiated utilizing the same equip- 

 ment described above with the exception that various types of electrodes 

 were used, and the electrodes were connected to a General Electric Photo- 

 electric Potentiometer Recorder, model fiCE5 DM5Y-I, The results of this 

 series of tests were unsatisfactory in that the masking potential was 

 still present. No noticeable change in reading was obtained from the 

 recorder upon applying a voltage to the field coil, but a potential 

 of greatly vaiying amplitude and varying polarity was present at all 

 times. The induction of this troublesome masking potential into the 

 measuring circuit was attributed to the chemical electrolysis between 

 the water and the electrodes. To minimize the effect of this undesirable 

 chemically induced voltage, another series of tests was made with a 

 different coil providing a magnetic field of considerably greater 

 strength. Calculations of induced voltage for different velocities 



20 



