28 BELL SYSTEM TECHNICAL JOURNAL 



meter and grounded battery, and noting the amount of momentary 

 deflection on the good and bad conductors, respectively, when the 

 polarity of the battery was reversed. The ratio of these deflections 

 gave a general indication of the location of the fault. This method, 

 while giving more accurate results than the continuity test, was still 

 only an approximation and as such was materially affected by line con- 

 ditions. An appreciable error was produced by leakage due to trees 

 and other causes, and much depended on the judgment of the tester. 



Later, the Wheatstone bridge ^ largely displaced the voltmeter. 

 A standard capacitance was compared with the impedance between 

 the open conductor and ground, by varying the ratio arms of the 

 impedance bridge. In this comparison it was necessary to employ 

 some form of alternating testing potential. At first, the simple 

 expedient prevailed of reversing the bridge battery, as in the voltmeter 

 test, the bridge being balanced until no transient unbalance current, 

 or "kick," was indicated by the galvanometer when the battery was 

 reversed. However, when the battery was reversed rapidly, the 

 galvanometer displayed a tendency to stand still at all times. A 

 considerable improvement in the method was effected by providing 

 or the reversal of the galvanometer connections at the same time the 

 battery was reversed. With this arrangement, the galvanometer 

 always read in the same direction, and a balance could be more 

 easily obtained. 



Later a relay system was arranged for automatically reversing the 

 connections to the galvanometer as well as for reversing the testing 

 battery. This arrangement relieved the operator of the necessity of 

 doing the reversing manually. Following this, a source of 20-cycle 

 ringing voltage was used for the bridge and also for operating the 

 galvanometer reversing relay. With this arrangement open locations, 

 made on aerial wire lines and short lengths of cable, were fairly 

 satisfactory. However, with the extensive installation of the long 

 toll telephone cables, it was found that open locations made on this 

 type of conductors, did not give a consistently accurate indication 

 of the location of the fault. 



As a preliminary step in the development of a suitable open location 

 method and associated apparatus, an analysis was made of all errors 

 which, in general, might enter into a open location. These errors can 

 be classed in several groups for treatment or correction. One group 



^References: Frank A. Laws, "Electrical Measurements," 1917, McGraw-Hill 

 Book Co., Inc., N. Y.; page 381, "Bridge Measurements of Capacity and Induc- 

 tance." 



"Bridge Methods for Alternating-Current Measurements," D. I. Cone, Trans- 

 actions of A. L E. E., July, 1920. 



