3H 



SCIENCE. 



[Vol. XIII. No. 325 



can be conveniently commutated from about 120 to 1,600 times per 

 minute, the lower speeds being uSed for circuits having a large time- 

 constant, and the higher for circuits with a smaller time-constant. 

 The exact range of commutation, however, is made different in the 

 various secohmmeters intended for different purposes. 



When the instrument is intended to be employed for absolute 

 measurements of the co-efficient of induction or the capacity of a 

 condenser in terms of a resistance and a time, a speed-indicator is 

 attached to the spindle seen projecting from the commutator in 



The commutators can be driven at one or other of two speeds 

 relatively to that of the driving-handle. With one arrangement 

 there are rather more than 8 reversals of both the galvanometer 

 and of the battery for one revolution of the handle ; and with the 

 other, 24 reversals of each for one revolution of the handle. The 

 secohmmeter can be conveniently driven by hand, so as to obtain 

 a steady speed of 300 to 6,000 reversals per minute of both the 

 galvanometer and the battery. 



To shift from one speed ratio to the other, press down the end of 



-AYRTON AND PERRY SECOHMMETER. 



the figure. But for comparison of the co-efficients of self or 

 mutual induction with one another, or with the capacity of a con- 

 denser, no speed-indicator is necessary. 



Another use to which the secohmmeter can be put is the meas- 

 urement of the resistance of a liquid which is liable to polarize with 

 direct currents, but which, as is well known, will not polarize with 

 rapidly alternating currents. 



This instrument consists of two rotatory commutators, each 

 with four stationary brushes. The commutators are on the 'same 



FIG. 2. — COMPARING TWO CO-EFFICIENTS OF SELF-INDUCTION. 



spindle, one at the front, and the other at the back of the secohm- 

 meter, but for convenience they are shown in the accompanying 

 symbolical figures as if they were in the same horizontal plane ; in 

 reality, however, the brushes, b^, b^, b^, b^, are at the top of the in- 

 strument. One commutator, GC, is for periodically reversing the 

 galvanometer connections ; and the other, BC, for reversing the 

 battery connections. An adjustment is provided for enabling the 

 relative positions of the two commutators to be varied, so that 

 both reversals can be made to occur simultaneously, or one a little 

 before or after the other, or one reversal midway between two suc- 

 cessive reversals of the other. 



the locking-lever at the right of the secohmmeter, and slightly push 

 in or pull out the handle, turning it slightly to assist the toothed 

 wheels engaging properly. When engaged, let go the end of the 

 locking-lever. 



I. To compare two co-efficients of self-induction, join up the ap- 

 paratus as in Fig. 2 ; ' then, if the resistances i\ and r^, of the 

 non-inductive branches of the bridge, be adjusted to give balance 

 with a steady current, balance will also be obtained on rotating 

 the secohmmeter, when 



Z;, £2, being the co-efficients of self-induction of the inductive 

 branches. The speed at which the secohmmeter is driven need 

 not be known, but the greater the speed the more sensitive the 

 test ; the rate of reversal must not, however, be too great for the 

 currents to reach their steady values between two consecutive re- 

 versals. 



2. To compare two capacities, join up the apparatus as in Fig. 

 3 ; then balance will be obtained on rotating the secohmmeter, 

 when 



/^i and F„ being the capacities of the condensers, and r^ and r^ 

 the resistances of the non-inductive branches of the bridge. As 

 before, increasing the speed at which the secohmmeter is driven, 

 merely increases the sensibility of the test without affecting the 

 ratio connecting the capacities with the resistances. 



3. In similar ways two co-efficients of mutual induction may be 

 compared with one another, or a co-efficient of mutual induction 

 with a co-efficient of self-induction, or either of these with the 

 capacity of a condenser shunted by a non-inductive resistance. 



4. To measure a co-efficient of self-induction absolutely in sec- 

 ohms by the comparative deflection method, attach a speed-indi- 

 cator to the commutator spindle, which is prolonged for this pur- 

 pose, and join up the apparatus as in Fig.4, Z being the co-efficient 

 of self-induction to be measured, and >\, r„, r^, values of the three 

 non-inductive resistances that give balance with a steady current. 

 Rotate the secohmmeter handle at some convenient speed, causing 

 the commutator spindle to make n revolutions per second, and ob- 

 serve the steady deflection, di, of the galvanometer. Next stop the 



1 The continuous lines represent the permanent connections in the secohmmeter 

 itself: the dotted lines, connections temporarily made outside the instrument. 



