58 BELL SYSTEM TECHNICAL JOURNAL 



Summary 



1. At times of a fault to ground on a power system with neutral 

 grounded through a Petersen reactor, the action of the latter tends to 

 extinguish the arc and to prevent its restriking. Theoretical con- 

 siderations, applied to a practical case, indicate that the transient 

 over-voltage on a sound phase at the instant of occurrence of the 

 fault is substantially the same as in a system with isolated neutral. 



2. Grounding the neutral through a Petersen earth coil instead of 

 directly or through a low resistance would largely prevent the electro- 

 magnetic inductive effects to which exposed communication circuits 

 are liable at times of faults to ground in systems grounded in the 

 latter manner. (Extensive high voltage networks are perhaps an 

 exception to this statement. But even here, the electromagnetic 

 inductive effects would in general not be greater with the reactor 

 than with isolated neutral.) However, effects due to electric induc- 

 tion similar to those from an isolated system may be expected to 

 appear. Except for long, close parallels involving open-wire com- 

 munication circuits these effects should in general be much less severe 

 than the electromagnetic inductive effects from a system with dead- 

 grounded neutral. The extent and severity of the inductive effects 

 experienced from the system grounded through the reactor would 

 further tend to be smaller than with the isolated system, because of 

 the effect of the reactor in preventing arcing grounds. 



3. Grounding the neutral through a resistance large compared to 

 other impedances involved in a short circuit to ground should have 

 an advantage over grounding through the Petersen reactor, in that 

 the former method presents fewer difficulties in respect to power 

 system protective relays, so that it would reduce the possibility of the 

 continuance of inductive disturbances over considerable periods of 

 time, which might be involved in grounding through the reactor, 

 under present relay practice. From an inductive interference stand- 

 point, a choice between the two methods would depend upon the 

 circumstances of particular cases. Advances in the art of relay 

 protection would improve the position of the reactor in such con- 

 siderations. 



4. Under normal power system operating conditions, the use of the 

 reactor may lead to excessive residual voltages of fundamental fre- 

 quency if the admittamces from phases to ground are unbalanced. 

 Such unbalance may be reduced to the extent necessary from this 

 point of view by power circuit transpositions. 



5. Under normal operating conditions, it is to be expected that the 

 residual voltages and currents of the triple harmonic series occurring 



