ML13330B224
| ML13330B224 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 07/02/1987 |
| From: | Medford M Southern California Edison Co |
| To: | NRC Office of Administration & Resources Management (ARM) |
| References | |
| TAC-65229 NUDOCS 8707070279 | |
| Download: ML13330B224 (4) | |
Text
Southem California Edison Company P. 0. BOX 800 2244 WALNUT GROVE AVENUE ROSEMEAD, CALIFORNIA 91770 M.O.MEDFORD TELEPHONE MANAGER OF NUCLEAR ENGINEERING (818) 302-1749 AND LICENSING July 2, 1987 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D.C. 20555 Gentlemen:
Subject:
Docket No. 50-206 Pressure Transmitter PT-459 Failure San Onofre Nuclear Generating Station Unit 1 By letter dated May 19, 1987, the NRC staff was provided an analysis of the San Onofre Unit 1 containment pressure response for a main Feedline
- Break (FLB) downstream of the in-containment check valves.
The analysis demonstrated that a high containment pressure reactor trip occurs earlier than the 14 second trip used in the previously submitted analysis. This submittal also provided the plant response for a FLB upstream of the in-containment check valves which includes breaks outside containment. For this transient the reactor trip occurs at 84 seconds on high pressurizer level at 50%. During subsequent telephone discussions, the NRC staff requested information regarding the impact which consideration of a small FLB would have on the transient results for the downstream FLB and an evaluation of the radiological consequences of the FLB outside containment. The purpose of this letter is to provide the requested information.
Though small FLBs have not been specifically reviewed as a design basis event at San Onofre Unit 1, the largest FLB has always been considered to be bounding. When credit was taken for the high containment pressure safety injection reactor trip, the impact of a smaller break was evaluated to determine if the large break continued to be bounding. As indicated by letters dated April 30, 1987 and May 19, 1987, the large FLB would result in acceptable consequences even if the reactor trip was arbitrarily delayed up to 44 seconds.
The containment response calculation determines the mass energy release rate needed to obtain the trip in 44 seconds and therefore the minimum 8707070279 870702 PDR ADOCK 05000206 P
Document Control Desk July 2, 1987 break size for which the trip would be conservatively estimated to respond within the available time. This was calculated to be 42,500 lbs. in 44 seconds or approximately 1000 lb/sec. This break size was estimated to be 0.09 ft2 which is considerably smaller than the large break of 0.785 ft2.
This break size (.09 ft2) represents the largest break for which a reactor trip other than containment pressure safety injection would have to be relied upon.
An additional consideration in this evaluation was the behavior of the main feedwater system. For the small break all of the feedwater cannot be assumed to be going into containment since this would be non-conservative for the containment response since it would increase the discharge rate and result in an earlier trip on high containment pressure. The main feedwater system would therefore continue to deliver some flow to the two steam generators in parallel with the broken loop. The main feedwater system normally delivers approximately 1500 lb/sec to all three steam generators. If the break flow identified above of 1000 lb/sec is assumed to be going to containment for the largest small break under consideration then approximately 500 lbs/sec would be delivered to the other two steam generators or approximately 33% of full flow.
The primary system response for this smaller break would of course be different than the large break with trip required at 44 seconds and in fact more time would be available before reactor trip was needed. In order to establish the transient response of the system under these circumstances a transient analysis for this case was run using the same codes and methods used for the large breaks. The break assumed was.09 ft2, downstream of the in-containment check valve with 15% of main feedwater flow to each of the intact steam generators for a total of 30% flow. Reactor trip occurred on high pressurizer level (50% setpoint) at 120 seconds. Auxiliary feedwater flow of 250 gpm was provided at 15 minutes. The results for this case are acceptable since the acceptance criteria are met. This analysis demonstrates that small breaks are bounded by the large breaks in the main feedwater system.
The NRC staff also requested an evaluation of the radiological consequences of a FLB outside containment. The FLB outside containment was assumed to be limiting since this was the case for the previously evaluated steamline breaks. As will be discussed below, because of the conservative modelling assumptions, the FLB outside containment is indeed the bounding case.
In response to the NRC staff request for an evaluation of the radiological consequences of a FLB outside containment, we requested that Westinghouse determine specifically if the DNB ratio falls below acceptable limits (1.30).
This parameter is not normally a subject of concern for FLB transients and is therefore not automatically calculated by Westinghouse. The results of the Westinghouse calculations indicate that DNB is not approached during the FLB transient outside containment. This includes the time period prior to reactor trip as well as the time at about 900 seconds just before the auxiliary feedwater system is initiated when the primary system reaches the highest temperature. No additional fuel failures will therefore occur.
Document Control Desk July 2, 1987 For the steam side paths to the environment, the FLB dose consequences would be the same as those calculated for the Main Steamline Break (MSLB).
This calculation is documented in SEP TOPIC XV-18, "Radiological Consequences of Main Steam Line Failure Outside Containment", which was transmitted by NRC letter dated December 31, 1979.
The results of this SEP topic review for the 2 hr. site boundary thyroid doses are as follows:
MSLB (with accident initiated iodine spike) 13 rem MSLB (with previous iodine spike) 22.5 rem The FLB transient results for the Cycle 9 event result in an additional contribution to the resultant dose since the PORV's discharge primary fluid into containment. This contribution has been calculated using extremely conservative assumptions to obtain a bounding dose contribution from the containment path. The methodology was the same as that used in the SEP topic referenced above except for the very conservative nature of the assumptions described below.
No credit was taken for reactor coolant flashing. The iodine release to the containment atmosphere is equal to the iodine carried by the flashed and vaporized coolant. It is expected that 40% to 50% of the relieved fluid would flash to steam. The non-flashed (liquid) portion of the release will atomize. A significant portion of these atomized coolant droplets (perhaps 10%) are expected to vaporize. If credit were taken for flashing, the 40% of the reactor coolant released to containment which would remain as liquid retaining approximately 40% of the iodine, would not contribute to the dose.
Thus, consideration of flashing would reduce the containment portion of the dose by approximately 40%.
No credit was taken for elemental iodine deposition on internal containment surfaces. Instantaneous deposition of half the released iodine as recommended in Regulatory Guide 1.4, would reduce the containment dose by 50%.
A time dependent deposition removal term, as suggested in the Standard Review Plan Section 6.5.2, would have a smaller dose reduction effect of 10% to 20%.
The containment vent valves were assumed to be closed manually at 15 minutes rather than automatically at an earlier time by either the radiation monitoring equipment or a high pressure containment isolation signal.
This assumption assures that the FLB break outside containment is the limiting case.
A break inside containment would add the secondary side discharge to the primary coolant discharge through the PORV's and result in a high pressure containment isolation in less than 44 seconds for the worst case. The path through the vent valves is the major contributor to the dose.
The resulting 2 hr. thyroid dose at the site boundary for the containment path with pre-accident (PA) iodine spike is 28.1 rem. In addition, the dose for the accident initiated (AI) iodine spike was estimated using the ratio of AI/PI doses from the SEP Topic review. Based on this ratio (13/22.5) the containment dose contribution with an AI iodine spike is estimated to be 16.2 rem.
Document Control Desk July 2, 1987 The total thyroid dose at the site boundary for the FLB is the sum of the doses from the two paths considered above. These bounding doses are summarized as follows:
FLB (AI iodine spike) 29.2 rem FLB (PA iodine spike) 50.6 rem The doses calculated for the feedline break, while greater than those for the steam line failure currently reported in the SEP, do not constitute a significant increase in the consequence of the accident. This judgment is based on the fact that the doses are within their respective fractions of 10 CFR 100. Specifically, the dose guideline for similar events incorporating an accident initiated iodine spike (SGTR and MSLB) is defined by the NRC in NUREG-0800 as a "small fraction" of 10 CFR 100, that is, 30 rem thyroid. The dose guideline for the same events with a pre-accident spike, in place of the accident initiated iodine spike, is the full 10 CFR 100 limit of 300 rem. In addition, the doses reflect a large margin of conservatism as described above.
Consideration of these factors in a rigorous dose calculation would significantly reduce the containment dose contribution.
In the submittal of May 19, 1987 it was indicated that the ESF Single Failure Analysis as well as the design information and safety analyses for the Cycle 10 modifications to the Feedwater System and the Reactor Protection System to eliminate single failure concerns would be provided by June 19, 1987.
Due to the additional efforts which have recently been required related to this project as well as the current outage extension, the revised date for submittal is July 31, 1987.
If you have any questions or desire additional information regarding this subject please contact me.
Very truly yours, cc:
R. F. Dudley, NRR Project Manager, San Onofre Unit 1 J. B. Martin, Regional Administrator, NRC Region V F. R. Huey, NRC Senior Resident Inspector, San Onofre Units 1, 2 and 3