TM Wo, 377 



The heights of the induced voltage pulses are not equal; thus, periodic gaps 

 occur in the otherwise smoothly scribed sinusoidal-like oscillations. The jagged 

 effect could easily be eliminated by careful placement of each magnet within the 

 impeller to give identical voltage amplitudes for a given speed. 



Figure V-2 is a 38-second segment of processed velocity data abstracted from 

 the record of serial 023 (BBELS-9, 0.5 m i), where the orthogonally mounted wave 

 meter (OMDUM III) was used to obtain a simultaneous u and w record - shown by solid 

 and broken lines, respectively. The data points used were interpolated at the usual 

 0.2-second intervals. The wind was from the ESE at about 3.0 - U.5 m sec"-'-, and 

 there was a well-defined swell running from the south (see table XV- 3 for other 

 information on this record). Also present were small waves building up with the 

 increasing wind from the SE. The u sensor of the wave meter was aimed directly at 

 the oncoming swell. 



The long period undulations of the u and w traces in figure V-2 vary from 7 

 to 10 seconds. The gross oscillations of the u and w components demonstrate a 

 phase shift of approximately 90° (to be expected) between the two wave motion 

 components; i.e., corresponding to sine and cosine-like motions. The increase in 

 the amplitude of the motions toward the end of the trace, similar to the increase 

 in the heights of a train of swells, is probably due to the beat of two or more 

 dissimilar frequencies of waves. The velocity patterns are quite jagged, showing 

 that higher frequency wind wave motions (which were observed) were superimposed 

 on the swell. This becomes evident when examining the auto-spectra of these 

 observations. 



Figure V-3 shows a 38-second section of the record from serial 025 (BBELS-9? 

 2.0 m i). Here again, as with the record of 023, there is a similar periodicity 

 and out-of-phase relationship between the u and w traces; however, the short 

 jagged peaks are not so prevalent. This is consistent with the fact that at 

 deeper depths the higher frequency motions become filtered. 



In general, figures V-2 and V-3 show that the gross amplitudes of these 

 orthogonal velocity components are roughly equal. This is important to note 

 because often this was not the case with the OMDUM measurements. Figure V-h, 

 for example, shows a plot of a 38-second u and w trace from serial 120 (BBELS-16, 

 0.5 m i) during a period of moderate swell from the SSE and small seas (see 

 table XV-3). The u meter of the OMDUM III system was aimed into the wind waves 

 from the west. The mean horizontal current was about 20 cm sec - -'- moving westward 

 against the wind waves. 



Figure V-U shows a strong inequality in the amplitudes of the u and w traces. 

 The vertical velocity w shows strong oscillations having amplitudes two to three 

 times those of the u component. The u component oscillates about the negative 

 d-c bias caused by the tide current and never actually changes direction (crosses 

 the zero axis). Closer eamination of the motions shown in figure V-h reveals 

 that the relatively high frequency oscillations (2-3 second period) occurring in 

 the u trace are identifiable on the w trace (A and A ! , B and B*, etc.). Note 



96 



