THE MEASUREMENT OF RESISTANCE 165 



of the current can be sent through the galvanometer. During 

 the final adjustment, when full sensitiveness is desired, the 

 movable arm should be turned so far to one side that it breaks 

 the shunt circuit and the entire current flows through the 

 galvanometer. The various positions of the movable arm are 

 usually so arranged that the fractional parts of the full current 

 which can be sent through the galvanometer are 0.001, 0.01, 0.1 

 and 1. When using a Wheatstone bridge one should always begin 

 measurements with the galvanometer heavily shunted. Violent 

 deflections of the instrument are thus avoided. 



The Null Method of Making a Measurement. The coil of 

 unknown resistance is inserted, as indicated in Fig. 90. All 

 connections must be electrically perfect ; all binding posts 

 should be screwed up tightly, but without using undue force; 

 all plugs should be firmly inserted and the galvanometer heavily 

 shunted. A rough idea of the magnitude of X is obtained as 

 follows: Make M = N; draw the 1-ohm plug in P; depress the 

 key with care, being ready to release it immediately should the 

 deflection of the galvanometer be violent. The deflection will be 

 assumed to be toward the left. Note this deflection, plug up the 

 1-ohm coil, and draw the 5,000-ohm or other high resistance plug; 

 proceed as before. The deflection may be toward the right; 

 it is then known that P and consequently X is between 1 ohm 

 and 5,000 ohms. If one deflection is greater than the other, 

 it shows that the proper value of P is nearer the resistance which 

 gives the smaller deflection. Next try 10 ohms in P. Suppose 

 the deflection to be still toward the left; the proper value of P 

 is between 10 ohms and 5,000 ohms. Proceed in this manner, 

 always narrowing the limits between which the right value of 

 P must be located. Having obtained an apparent balance, the 

 shunt resistance is increased and a better approximation obtained. 

 Suppose that the bridge finally balances with P = 25 + ohms, 

 25 ohms being too small and 26 ohms too large ; then X is between 

 25 and 26 ohms. It is obvious that the determination of X is 

 good only to about 2 or 3 per cent. Suppose that X is desired 

 to 0.1 per cent.; then as the smallest coil in P is 1 ohm, P must be 

 between 2,500 and 2,600 ohms in order that the smallest step 

 may represent most nearly the desired precision. Accordingly 



