Author: Yujie I. Lu and Jeffrey Pond, National Grid, USA
On August 14th, 2005, the National Grid System Control Center reported that at 16:01:30 a lightning stroke caused a fault on the 115kV O-141S line near the company's Millbury #302 substation.
The substation and faulted line configuration are shown in Figure 2. The circuit breakers O141 and 29-41 at Millbury #302 and O141S at Nashua St tripped correctly by the line protection relays. As per the report, at the time of the O141S operation, the 115kV P142S line, one of the parallel lines with the faulted O141S line, tripped by their line protection relays at Millbury #302 and Rolfe Avenue substations for the O141S fault. Targets reported for these two operations are shown in table1.
Why did the P142S line relays respond to the O141S fault? Was it a single circuit fault only? First, the investigation team verified that both lines share double-circuit towers with a vertical design between Millbury #302 and Rolfe Avenue, where the P142S line is more highly insulated than the O141S line. The conductor phasing arrangements, top to bottom, are B-C-A on the O141S and A-B-C on the P142S.
Knowing the construction of these lines, it was believed that it would have had to have been a high current stroke to trip both lines. Next, the sequence-of-event and digital fault recorder (DFR) records captured at Millbury #302 were collected, which indicated that the fault occurred on both O141S and P142S lines at 3.06 miles from Millbury #302. Based on the records, the fault involved A-phase-, B-phase- and C-phase-to-ground on the O141S line and B-phase-to-ground on the P142S line with fault impedance (Figure 1). It was then further noticed that:
Note: Current is shown in secondary amperes in the record. The CTR for the O141S and P142S relays at Millbury #302 is 3000:5.
In order to verify that it was a double-circuit non-bolted fault and all line relays of O141S and P142S responded to the event correctly, a simultaneous fault was simulated by using the short circuit program simultaneous fault module. The simulation results proved that:
Furthermore, to quantitatively match the fault records captured by the DFR at Millbury 302, more simulations were conducted with a different fault resistance on each phase.
Finally, the results concluded that the fault resistance for the O141S three-phase-to-ground fault was approximately as follows: Ra = 5.0 primary-ohms, Rb = 11.3 primary-ohms and Rc = 7.0 primary-ohms, and the fault resistance for the P142S B-phase-to-ground fault, Rg, was around 6.5 primary-ohms. (Figure 3 &Table 2)
