ML20135H300
| ML20135H300 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 05/14/2020 |
| From: | Christopher Hunter NRC/RES/DRA/PRB |
| To: | |
| Littlejohn J (301) 415-0428 | |
| References | |
| LER 219-1992-005 | |
| Download: ML20135H300 (5) | |
Text
B-6 B.4 LER Number 219/92-005 Event
Description:
Loss of Offsite Power Due to Forest Fire Date of Event: May 3, 1992 Plant: Oyster Creek B.4.1 Summary Oyster Creek lost offsite power for 5 min when a forest fire near the plant caused the offsite transmission lines to fault. The two emergency diesel generators (EDGs) operated as designed. Although offsite power was restored in 5 min, the emergency buses were supplied from the EDGs for 17 h until reliability of the offsite power supply could be assured. The conditional core damage probability estimated for this event is 7.1 x 10-1. The relative significance of this event compared to other postulated events at Oyster Creek is shown in Fig. B. 1.
LER 219/92-005 1E-7 1E-6 1E-5 ME4 1&-3 ME2 p roL cL LOFW+IC LOOP Fig. B.1. Relative significance of LER 219/92-005 compared with other potential events at Oyster Creek.
B.4.2 Event Description On May 3, 1992, at 1310 hours0.0152 days <br />0.364 hours <br />0.00217 weeks <br />4.98455e-4 months <br />, the control room at Oyster Creek was informed that a forest fire was burning to the west of the plant near the 230-kV offsite distribution lines. At 1326 hours0.0153 days <br />0.368 hours <br />0.00219 weeks <br />5.04543e-4 months <br />, a full reactor scram occurred following the loss of the 230-kV lines. It is believed that the heavy smoke and heat from the fire ionized the air near the lines and caused the line to fault. The 34.5-kV supply was also lost and the result was a complete loss of offsite power (LOOP). The two EDGs started and loaded onto the two emergency buses (IC and ID). However, control rod drive (CRD) pump A failed to start during the loading sequence because of high-resistance contacts in its time-delay relay. Offsite power was restored LER NO: 219/92-005
B-7 from the 34.5-kV system through the two startup transformers at 1331 hours0.0154 days <br />0.37 hours <br />0.0022 weeks <br />5.064455e-4 months <br />, and the two nonemergency buses were reenergized. The plant staff questioned the reliability of the offsite supply due to the proximity of the fire to the station and the reduced number of offsite supply lines that were available.
In addition, difficulties were encountered in transferring the emergency buses to offsite power. As a result, the emergency buses continued to be supplied from the two EDGs for another 17 h. By 0631 hours0.0073 days <br />0.175 hours <br />0.00104 weeks <br />2.400955e-4 months <br /> on May 4, 1992, the emergency buses were restored to their normal offsite supplies.
B.4.3 Additional Event-Related Information Oyster Creek has three 230-kV supply lines and five 34.5-kV offsite lines. Two of the three 230-kV lines share double-circuit transmission towers. Normal operation is with two or three of the 230-kV lines and at least three of the 34.5-kV lines in service.
During startups and shutdowns, station power is supplied from the 34.5-kV system to the two startup transformers. During normal operation station power is supplied from the main generator through an auxiliary transformer and no loads are carried by the startup transformers. The two 4160-V emergency buses (IC and ID) are normally supplied by the auxiliary transformer via the two nonemergency buses (IA and 1B). The EDGs associated with each emergency bus can supply power in case of a LOOP.
B.4.4 Modeling Assumptions This event was modeled as a recoverable LOOP. To reflect the impact of the fire on the 230-kV lines and the extended time on the EDGs, nonrecovery probabilities for short-term and long-term ac power were developed by averaging the probabilities normally used for plant-centered and grid-related LOOPs.
(See ORNL/NRC/LTR-89/1 1, Revised LOOP Recovery and PWR Seal LOCA Models, August 1989).
This calculation results in somewhat higher short-term and long-term nonrecovery probabilities when compared to the plant-centered LOOP model and gives credit for the startup transformers as a source of supply for the safeguards buses that was available but not utilized. The nominal LOOP includes the effects of extreme severe weather and severe weather induced LOOPs in addition to the plant-centered and grid-related LOOPs, with correspondingly higher nonrecovery probabilities. Therefore the core damage probability for this event is less than that for the nominal case.
The failure of the CRD pump to start during EDG loading was not addressed in the event model. This pump would have been manually started if required (operator action to start and align the CRD system is included in the branch model).
B.4.5 Analysis Results The conditional probability of core damage estimated for this event is 7.1 x 10-1. The dominant core damage sequence, highlighted on the following event tree in Fig. B.2, involves a LOOP with a postulated failure of emergency power and failure to restore ac power prior to battery depletion.
LER NO: 219/92-005
B-8 EI L~LOO
~OOP E I ftT (LONGC
~L SV ONO:CHA RI II /U WC lCO I ADS I 1 R* LCLPC SDC CC OH~
91! SEQ NO END STATE OK OK OK 41 CD OK OK 42 CD OK OK 43 CD OK 44 CD OK 45 CD 46 CD 47 CD OK OK 48 CD OK OK 49 CD OK 50 CD OK 51 CD 52 CD 53 CD OK OK OK 54 CD OK OK 55 CD OK OK 56 CD OK 57 CD OK 58 CD 59 CD 60 CD 98 ATWS OK 61 CD 62 CD OK 63 CD 64 CD 97 ATWS Fig. B.2. Dominant core damage sequences for LER 219/92-005.
LER NO: 219/92-005
B-9 LER NO: 219/92-005
B-1O LER NO: 219/92-005