THE BALLISTIC GALVANOMETER 103 



field and extreme care in the preparation of the materials used, 

 and in the winding of the coil. The "set" may be minimized by 

 the proper choice of suspension strip and by care in mounting it. 



When readings are made they should all be taken toward the 

 same end of the scale and the coil should not be allowed to swing 

 very much beyond zero on its return; proper damping or use of 

 the checking device will insure this. Before taking any readings, 

 a deflection in the proper direction and as large as any which are 

 to be used should be given the system ; after this there will be no 

 appreciable change of zero. This precaution should be taken 

 each time the instrument is used. 



If it is necessary to take a reading when the coil is not abso- 

 lutely at rest, but swinging so that the amplitude as read on the 

 scale is only a small fraction of a centimeter, the impulse shoud 

 be given to the system when the swing is at its maximum 

 and the elongation 6 should be calculated from the true mechan- 

 ical zero, not from the scale reading when the discharge was 

 passed. This applies when the swinging on either side of the 

 mechanical zero is not more than about 3 per cent, of the first 

 elongation, B\. 



Thermo-elcctromotive forces in any part of the circuit are 

 troublesome; those arising in the galvanometer itself should be 

 minimized by shielding from draughts or anything which could 

 cause irregularities of temperature. In the best instruments the 

 binding posts, connections to the movable coil and the coil itself 

 are all of copper. In specially designed instruments the current 

 is not taken in through the suspension, which may be of steel, 

 but through spiral connections made of very thin copper strip. 

 This strip may be made by rolling out a fine wire, about No. 40. 

 The spirals may be made so delicate that they contribute prac- 

 tically nothing to the restoring moment. 



The Calibration of a Ballistic Galvanometer. It will be shown 

 that the quantity of electricity which is instantaneously dis- 

 charged through a ballistic galvanometer is given by 



where 



T = time of a complete swing. 



