8 



OCEAN ATMOSPHERIC-ELECTRIC RESULTS 



3. Remove recorder from wooden box on stern rail, take 

 to darkroom and remove the week's roll of photo- 

 graphic paper and install new roll. 



4. Wind driving clock and set to correct Greenwich Time. 



5. Replace recorder in box on stern rail. 



6. MaJce notes of above operations on daily record form. 



Conductivitv Apparatus 8A. --The method employed 

 on cruise Vn for the visual measurement of the conduc- 

 tivity of the air was the same in principle as that used 

 on previous cruises although the arrangement of appara- 

 tus was somewhat different. The method is that devised 

 by Gerdien (5). In this method air is drawn by a fan 

 through the space between two concentric cylinders, the 

 central member of which is charged and connected to 

 the insulated fiber system of an electrometer. The the- 

 ory of the instrument shows that as long as the velocity 

 of the air stream is large enough to insure that the cen- 

 tral cylinder cannot collect from the air all the ions at- 

 tracted toward it as the air passes through, the rate of 

 loss of charge by that cylinder is independent of the air 

 velocity and dependent only on the potential applied and 

 on the conductivity contributed by the ions of sign oppo- 

 site to the charge on the cylinder. 



The outer tube of the air -flow system has been shown 

 in figures 7 and 8. Air was drawn into the intake of the 

 air -flow tube on the roof of the observing cabin. It 

 passed around a small cylinder which was supported in an 

 amber insulator concentrically within the air -flow tube, 

 and at a height convenient for attachment of an eye- 

 reading electrometer or a recording apparatus located 

 on the shelf in the cabin. A difference of potential was 

 maintained between the small cylinder and the air -flow 

 tube by means of a battery, the battery being carefully 

 protected against the sea air by being kept in a suitable 

 cabinet or compartment. In the upper partof the air -flow 

 tube, above the small central cylinder, two concentric 

 cylinders were installed, the innermost being connected 

 directlyto the air -flow tube and the other one insulated 

 from both. A high potential could be applied across the 

 cylinders whenever desired, to "sweep" out of the air 

 stream all the ions entering the tube, a procedure which 

 was used both with eye-reading and recording apparatus. 

 With eye-reading apparatus, the sweeping potential was 

 applied when tests were made of insulation leak, and with 

 recording apparatus it was applied for a few minutes once 

 each hour to establish a base line from which the deflec- 

 tions during regular recording could be measured. 



The procedure on cruise VII, as on previous cruis- 

 es, for eye-reading measurements of conductivity con- 

 sisted essentially in observing the rate of discharge of 

 the central cylinder and the attached electrometer when 

 charged to an average potential, V. In determining the 

 rate of discharge it was convenient to observe the time, 

 r, required for the fiber (only one fiber is observed in 

 this work) to move over a specified number of scale di- 

 visions, d, corresponding to a change in potential, 5 V. 

 The symbols used here are those used on a form sup- 

 plied for recording the observations (Department form 

 101), and the potential is expressed in volts. The capac- 

 itance, Ci, of the central cylinder, the electrometer and 

 the connections of electrometer to central cylinder, and 

 also the capacitance, C2, of that part of the insulated 

 system exposed to the air flow, enter the determination 

 of conductivity. Both values of capacitance must be 

 measured rather than computed; otherwise inaccuracy 

 is introduced in the conductivity determination (6). The 

 formula for the conductivity is as follows: 



V 4irC2 



where tnJ is obtained from leak tests as explained here- 

 after. 



Procedure for Eve-Reading Conductivity Measurements 

 on Cruise VII 



1. Preliminary 



(a) Ascertain that the bifilar electrometer and bat- 

 teries are in proper working condition. Desirable 

 electrometer sensitivity is about two volts per di- 

 vision. 



(b) Start fan motor and apply the sweeping potential. 



(c) Calibrate the electrometer over the range to be 

 covered in the observations, using for this pur- 

 pose a Weston voltmeter connected in parallel 

 with it. Two points near the limits of the range 

 will be adequate. 



2. Initial leak test 



(a) Apply to the central cylinder and electrometer 

 system a potential slightly higher than that desired 

 at the beginning of the conductivity observations 

 proper. 



(b) Observe leak in scale divisions for a definite in- 

 terval of time, say 180 seconds. Record time- 

 duration of leak test (on form 101 opposite Ati). 

 Observe and record position of fiber at the begin- 

 ning and at the end of the leak test and enter on 

 form after 9j. 



3. Conductivity observations 



(a) Remove sweeping potential one or two minutes be- 

 fore starting observations. 



(b) Select the number of scale divisions, 6, over which 

 the electrometer shall discharge, say four divisions, 

 and charge the apparatus as in 2(a). 



(c) Observe the time at which the fiber reaches the be- 

 ginning and the end of the four-division range and 

 record the times on form 101; the time interval in 

 seconds is r. Recharge the fiber and repeat this 

 operation three times or more to obtain several 

 values of r. 



4. Final leak test 



(a) Apply the sweeping potential and begin the leak 

 test when the fiber reaches a position approximate- 

 ly as far below the scale range used in 3(c) as it 

 was above the scale range at the end of the initial 

 leak test. Complete the test as in 2(b), recording 

 At2 and 62- 



5. Computation of conductivity 



(a) Compute the several values of r, take the recip- 

 rocal of each, and obtain the average, which is 

 ''"m"^) i" 10"^ units. 



(b) Compute ti'lfrom the formula ti"! = S\/e(M\) 

 and t2'^from a similar formula, expressing the 

 values in 10"^ units. Average the two values for 



tm-1- 



(c) The value of capacitance of Ci is 14.9 cm and that 



of C2 is 6.14 cm; consequently Ci/4TrC2 = 0.193. 



(d) From the calibration made in 1(c), determine the 

 voltages Vj and V2 for the scale readings at the 

 beginning and end of the four-division range used 

 in observing conductivity. Fromthese, 5V = Vi-V2 

 and V = (Vi + V2)/2. 



(e) Compute conductivity from the working formula 

 A =0.193(5V/V')(rm-1 - tm"^)- 



