A SHIELDED BRIDGE FOR INDUCTIVE IMPEDANCE 



169 



Consideration of these figures led to the following conclusions: 



1. The change in inductance balance due to reversal of the ratio 

 arms is not more than eight parts in 200,000 or 0.004 per cent. It has 

 been shown previously that the phase-angle balance of the ratio arms 

 is not critical with respect to inductance readings. Hence, the change 

 in inductance may be considered to be closely indicative of the re- 

 sistance unbalance of the ratio arms. From the above data it is 

 seen that this does not exceed 0.01 per cent. 



2. The change in resistance due to reversal of the ratio arms being 

 within the limits of observational error, the phase angles of the coils 

 themselves are, as nearly as can be determined by this test, exactly 

 balanced. 



3. Since the change of inductance balance due to reversal of the 

 impedance arms is no more than that due to the ratio arm reversal, 

 the capacitance balance of the impedance arms is apparently satis- 

 factory. It should be noted, however, that this balance is not critical 

 under these test conditions. (See eq. 9.) 



4. Since the resistance balance was appreciably affected by reversal 

 of the impedance arms, it appeared that there was an unbalancing 

 factor present which, if afifecting the ratio arms, was not reversed with 

 them but which if present in the impedance arms was reversed. The 

 latter might have been an unbalance of the impedance arm shunt 

 conductances but it was assumed that this unbalance was quite small. 

 On the other hand, as was mentioned earlier in the paper, there are 

 two small inter-shield capacitances shunting the ratio arms which are 

 not reversed by the ratio arm switch and it seemed likely that an un- 

 balance of these capacitances was causing the change in resistance 

 reading. This proved to be the case, as adjustment of the balance of 

 these capacitances for which, as previously noted, provision had been 

 made, resulted in identical resistance readings being obtained for both 

 positions of the impedance arm switch. After this adjustment had 

 been made the previous tests were repeated, resulting in readings as 

 given in Table V. 



TABLE V 



