TRANSACTIONS OF SECTION A. 779 



knife edge, counterpoised for a suitable position. For exhibition to the section 

 these detectors rephiced the more sensitive rotating indicators which are adapted 

 only for inspection from quite short distances. 



The application of the experiment to the illustration of cases of air motion on 

 a larger scale in certain examples of ventilation was pointed out. 



FRIDAY, SEPTEMBER 9. 



The following Papers were read : — - 

 1 . Comparison bet ween Charging a Secondary Cell at Constant Potential and 

 at Constant Current, more especially/ as regards Efficiency. By A. A. 

 Cahen and J. M. Donaldson. 



The method of charging secondary cells at constant potential has only com- 

 paratively recently come into vogue, and the probable reason for its adoption now 

 is the saving of time thereby effected. 



Little, however, is known and nothing, as far as we are aware, has been published 

 with regard to the efficiency and phj'sical characteristics of the method. 



The following tests were carried out to investigate these matters, and in order 

 to make a fair comparison between charging at constant potential and charging 

 at constant current a complete set of experiments by each method was tried on 

 one cell. 



This cell was of the Tudor type and had two positive plates (pasted) and three- 

 negative plates (unpasted). The size was that called 11 L.A., and its listed 

 capacity, charging at 20 amperes and discharging at 36, was 108 ampere hours. 



In order to get the true or working efficiency of the cell, it was charged and 

 discharged many times without intermission, until charge and discharge curves 

 obtained in consecutive experiments did not appreciably differ. 



On February 24, 1898, charging at constant potential was begun, and, in all, 

 50 charges and discharges were carried out, the intervals between charge and 

 discharge, discharge and charge being, on an average, one minute. 



During the charge the potential difference between the terminals of the cell 

 w-as kept constant at 2-51 volts, and charging was continued until the current had 

 fallen to 10 amperes. 



During the discharge the current was kept constant at 36 amperes, and the 

 discharge was stopped when the P.D. had fallen to 1"82 volts. 



The kst charge was given on March 4, and on March 7, after a preliminary 

 discharge, charging at constant current was started. 



The cell was charged at a constant current of 20 amperes, and charging was 

 continued in the first 22 tests until the P.D. had risen to 2-51 volts. During the 

 twenty-third and following charges this limit was increased to 2*58 volts, in order to 

 increase, it' possible, the capacity, which was low in comparison with that obtained 

 after a charge at constant P.D. 



Discharge took place at a constant current of 36 amperes and was stopped 

 when the P.D. had fallen to 1-82 volt. 



Forty-six charges and discharges were carried out. 



As, however, the capacity of the cell when charged at constant current was 

 considerably less than that after a constant P.D. charge, we suspected that the 

 heavy currents passed through the cell during the charges at constant P.D. had 

 injured it, and so this method was again started to see if the capacity was still the 

 same. 



This was on March 17, after an interval of only half an hour. 



Seven charges and discharges were sufficient to show that the capacity had not 

 diminished, and these experiments brought the tests to an end. 



Results obtained by Charging at Constant P.D. 

 Capacity and Efficiency. — The curves of charge and discharge drawn from the 

 results of experiments 37-46 inclusive lie very near together, and the mean of 

 the results was used in calculating the working efficiency. 



