FISHERY BULLETIN: VOL. 72, NO. 3 



As can be seen, the value for the load resistance 

 of a 5-square meter by 10-m array compares 

 favorably when determined by the two different 

 equations. The higher value of 0.01558 ohm was 

 used in making power calculations since any 

 electrode array will have some additional resis- 

 tance due to connection losses. 



Results from our field study thus provided the 

 following set of basic design specifics for our proto- 

 type pulse generator for use with attracting lights 

 in a netless fish harvesting application: 



1. Minimum field strength- 15 V/m. 



2. Pulse rate - 20-35 pulses/s. 



3. Pulse width- >0.5 ms. 



4. Array size - 5 x 5 x 10 m. 



5. Load resistance of array- 0.01558 ohm. 



Using these specifications, we determined the 

 output capability of the pulse generator which 

 would satisfy our requirements by the following 

 equation: 



P = VI X fl (6) 



where P = power, watts 



V = output voltage, volts 



/ = current, amperes 



f = pulse rate, pulses per second 



I = pulse length or width, seconds. 



To insure an adequate field strength throughout 

 our electrode array, we chose a value of 20 V/m 

 for the power calculations. We also selected a 

 maximum pulse rate of 50/s and pulse widths of 

 0.5, 0.75, and 1.0 ms to give the pulse generator 

 more versatility. Using Equation (6), the power 

 requirement is: 



V = 20 X 10 = 200 V for 10-m array 



V 



I =- 



200 



Rt 0.01558 



= 12,837 A, 



and at 50 pulses/s and 0.75 ms pulse width 



P =(200)(12,837)(50)(0.75 X 10-=^) 

 P = 96,278 W. 



In an applied system, a cable and connection 

 loss will be experienced. Because of the very low 

 load resistances of seawater, a 25*^ cable loss can 



easily be expected. Rounding off our requirement 

 to 90 kVA and after allowing for a 25% loss, we 

 need a pulse generator of 120-kVA output to 

 satisfy the system requirements we established. 

 As a crosscheck of the above designed system, 

 the following formula (Kreutzer, 1964) is used to 

 calculate the effective fish control range of one 

 electrode: 



R 



I X L X P 

 G X 2 xn 



where R 

 I 

 L 



9 

 G 



= effective range, meters 



= current into the water, amperes 



= length of fish, meters 



= water resistivity, ohm-meter 



= body voltage of fish. 



To determine the effective range of 20 V/m, a 

 value of 1 m is used for the fish length, fish body 

 voltage is 20 V, and the resistivity is again 0.189 

 ohm-m. 



Allowing a 25% cable loss requires a total input 

 voltage of 267 V at a total load resistance of 

 0.0208 ohm, and the current in the water is found 

 to be: 



/ = 



V 



267 



12,837 A. 



R 0.0208 

 Using these values, range (R) is found to be: 



R = 



R 



(12,837 X 1 X 0.189 



20 X 2 X 3.14 

 4.40 m. 



Since this value is computed for one electrode, 

 the 20 V/m range of two electrodes will be 8.8 

 m. In actual practice, however, the range of two 

 electrodes paired together is greater than twice 

 the reach of one, and we can supply a 5 x 5 x 10 

 m array with 20 V/m. At our minimum specifica- 

 tion of 15 V/m, the calculated reach of one 

 electrode is 5.08 m. 



Since the configuration of the electrode array 

 determines array resistance, various combina- 

 tions of electrode size and separation distance can 

 change the pulse voltage and current require- 

 ments. For this reason, a certain degree of flexi- 

 bility was designed into the netless fish harvest- 

 ing mode of the pulse generator. The system is 



666 



