ML13330B232
| ML13330B232 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 07/16/1987 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML13330B231 | List: |
| References | |
| TAC-65229, NUDOCS 8707230206 | |
| Download: ML13330B232 (4) | |
Text
- 'PREGU(
SO RE0 4UNITED STATES C,
NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATING TO MAIN FEEDWATER LINE BREAK REANALYSIS SOUTHERN CALIFORNIA EDISON COMPANY SAN ONOFRE NUCLEAR GENERATING STATION UNIT 1 DOCKET NO.: 50-206
1.0 INTRODUCTION
By a letter dated April 30, 1987 (Ref. 1), Southern California Edison Company (SCE) informed the NRC that a modeling error was discovered in their main feedwater line break analysis submitted with the Amendment Application No. 138 in a November 12, 1986 letter (Ref. 2). In that analysis it was assumed that in the event of a main feed line break, the reactor would trip on the variable low pressure (VLP) trip if the steam flow/feed water flow mismatch trip was not available. The modeling error was an assumption that a reactor trip would occur when one loop reached the VLP trip setpoint, whereas, the two-out of-three logic would actually prevent a reactor trip by just one signal.
Failure of the VLP trip may delay the reactor trip and result in failure to meet the acceptance criteria. The licensee, therefore, re-analyzed the main feedwater line break by neglecting the VLP trip. The reanalysis results were submitted in a letter of May 19, 1987 (Ref. 3).
Our evaluation is discussed in the next section.
2.0 STAFF EVALUATION In the process of reanalysis, the licensee also found that the previous analysis did not account for the check valves installed during the last refueling outage which ended in July 1986. Each of these check valves was installed inside the containment on a feedwater line in addition to the existing feedline check valve outside the containment. Therefore, the reanalysis considered two scenarios: (1) a feedline break at a location between the two check valves, and (2) a feedline break downstream of the newly installed valve before a steam generator.
The reanalysis of the first scenario, i.e., a feedline break at a location upstream of the newly installed check valve, was performed using the approved Westinghouse LOFTRAN computer code to simulate the transient. The initial conditions and assumptions in the analysis were essentially the same as those in the previous analysis except for the effect of the new check valves and the bypass of the variable low pressure trip. Since the break occurs upstream of the new check valve, the steam generators remain pressurized following the break. The decrease in the primary to secondary heat transfer causes the 8707230206 870716, 7
PDR ADOCK 05000206 P
I0
-2 primary system pressure, temperature and the pressurizer water level to increase. About 14 seconds after the break, the pressurizer power operated relief valves (PORVs) open; and the reactor trip occurs on high pressurizer level of 50% at 74 seconds. The pressurizer safety valves open at the setpoint of 2500 psia at about 700 seconds into the transient, and the pressure reaches the peak value of 2558 psia at about 900 seconds when the auxiliary feed water is started. This peak pressure is less than 110% of the design pressure (2750 psia) and, thus, satisfies the over pressure criterion specified in the standard review plan.
The pressurizer is filled at about 320 seconds. Water is released through the PORVs and safety valves when they are open. However, the amount of water release is not large enough to cause core uncovery. After the AFW pumps are started, the auxiliary feedwater flow is adequate in removing the decay heat. The safety valves and PORVs are subsequently reseated and the core remains covered and coolable during the entire transient. This satisfies the core coolability criterion specified in the SRP.
For the second scenario, the break can only occur inside the containment since the break occurs downstream of the newly installed check valves. After the break, the steam generator depressurizes and the containment pressure increases due to blowdown of mass and energy from the broken loop steam generator.
Reactor trip will occur on the safety injection signal when the containment pressure exceeds the setpoint of 2 psig.
In the previous analysis (Ref. 2) of the main feedline break, which resulted in SG depressurization, and the reactor tripped at 14 seconds on the VLP signal, it was found that the acceptance criteria had been met. Indeed, the licensee has determined that even if the reactor trip was arbitrarily delayed to 44 seconds, the acceptance criteria would also be met (Ref. 1).
Therefore, if the safety injection signal on the high containment pressure occurs soon enough to initiate a reactor trip, this scenario may be bounded by the previous analysis. The licensee, therefore, performed a quasi-steady state containment analysis to determine the time required to trip on the high containment pressure (2 psig) safety injection signal. The analysis was based on the conservation of mass and energy, and assumed thermal equilibrium of steam, water and air in the containment. Energy loss from passive heat sinks and through steam venting were included. The blowdown flow was assumed from the broken loop steam generator with an enthalpy conservatively assumed the same as feedwater enthalpy. The result of calculation showed that the containment pressure would reach 2 psig within 6 seconds after a large break. This is bounded by the 14 seconds of the previous analysis and well within the 44 seconds acceptable time. The licensee also performed a sensitivity study by using a more realistic enthalpy assumption along with a transient analysis. The result showed that a containment pressure would reach 2 psig within 5.1 seconds compared to the conservative steady state calculation of 6 seconds.
Therefore, we are confident that the reactor would be tripped on high containment pressure SI signal before 14 seconds and the previous analysis with VLP trip at 14 seconds is bounding.
-3 The licensee did not initially provide information regarding the amount of radioactive dose release during this transient. At the request of the staff, the licensee in a subsequent letter (Ref. 4) indicated that analyses performed by Westinghouse have shown that the core did not experience departure from nucleate boiling during the entire feedline break transient and therefore, no additional fuel failure has resulted. A conservative radioactive dose release calculation for the accident-initiated iodine spike and pre-accident iodine spike cases shows the two-hours site boundary thyroid doses of 29.2 rem and 50.6 rem, respectively, which are within the limits set forth in 10 CFR 100.
3.0
SUMMARY
The staff has reviewed the SCE submittal regarding a reanalysis of the main feedwater line break without a VLP trip. In both scenarios assumed in the analysis, the results indicated that the acceptance criteria with regard to the system pressure, core coolability and radiological consequence are met.
We conclude that the reanalysis has properly addressed the modeling error of the previous analysis and is acceptable.
4.0 ACKNOWLEDGEMENTS This evaluation was prepared by G. Hsii and C. Li.
- e S
-4 REFERENCES
- 1. Letter from M. 0. Medford (SCE) to US Nuclear Regulatory Commission, "Docket No. 50-206, Amendment No. 97, San Onofre Nuclear Generating Station, Unit 1," April 30, 1987.
50-206, Amendment Application No. 138, San Onofre Nuclear Generating Station, Unit 1," November 12, 1986.
- 3. Letter from M. 0. Medford (SCE) to US Nuclear Regulatory Commission, "Docket No. 50-206, Pressure Transmitter PT-459 Failure, San Onofre Nuclear Generating Station, Unit 1," May 19, 1987.
- 4. Letter from M. 0. Medford (SCE) to U.S. Nuclear Regulatory Commission, "Docket No. 50-206, Pressure Transmitter PT-459 Failure, San Onofre Nuclear Generating Station, Unit 1," July 2, 1987.