ML20207B709

SER Accepting one-time Exemption from 2-yr Type B & Type C Test Interval Requirements as Prescribed in App J,Until 880326,per Util Request

ML20207B709
Person / Time
Site:	Dresden
Issue date:	07/21/1988
From:	NRC
To:
Shared Package
ML20207B712	List: ML20207B709
References
NUDOCS 8808040079
Download: ML20207B709 (3)

Text

.

~

Enclosure S

SAFETY EVALUATION REPORT APPENDIX J TEST INTERVAL EXEPPTION DRESDEN STATION, UNIT 3 DOCKET NO. 50-249

1.0 INTRODUCTION

On January 7,1.088, the staff was.infonned by NRC Region III that the licensee (Cowonwealth Edison Company) had logged a single leak rate testing date for Dresden-3 local leak rate tests (LLRTs) based on the date of completing all the tests rather than the date of each individual component being tested. Thfs recordina system violates the test interval requirement of Appendix J to 10 CFR 50 and causes certain components to exceed their two-year test limit.

By letter dated January 10, 1988, the ifcensee requested a one-time exemption from the test interval requirement of Appendix J for these components. (bellows, manway psket seal, flanges, and isolation valves) beyond the two-year Type R or Type C test interval. The licensee stated that these components either can not be tested while the reactor is at power or would require entry into a limiting condition of operation (LCO) on primary containment integrity.

For those components which are required by Appendix J to be, tested prior the refueling outage, a 90 day extension is requested by the licensee to allow a delay in the Type B and Type C testing for these components until the scheduled refueling outage. This is the second time the licensee recuested Appendix J exemption for test deferral. The licensee made a similar recuest during l

l its last shutdown for refueling in 1985.

Paragraphs III.D.?(a) and III.D.3 of Appendix J specifies that both Type B and Type C tests shall be perfonned during reactor shutdown for refueling, but in no case at an interval greater than two years. Dresden 3 last shutdown for refueling at the end of cycle 9 (E0C 9) which started on October 28, 1985. Due to the extensive recirculation pipe replacement program and other outage l

related problems, the outage was unusually long with startup not occuring until Septerber 1986. The required Appendix J 1eak rate testing comenced on i

September 27, 1985 and continued through August 1986. As a result of the extensive E0C 9 outage, the LLPT postponement became recessary to coincide with the delayed EOC 10 refueling outage which is scheduled to begin on March 26, 1088. The licensee proposes to temporarily postpone testing for about 53 components, identified in Attachment 1 of the licensee's submittal dated January 10, 1988, l

to avoid the potential for an earlier reactor shutdown. The proposed test postponement for these components to March 26, 1988 will exceed the required 24-month Appendix J test interval by 49 to 147 days. The licensee has comitted to perfonn Apperdix J testing as soon as possible following the refueling outage or during any earlier outage of suitable duration should one occur prior to March 26, 1988.

l 8808040079 880724 PDR ADOCK 05000249 P

PDC

c 2.0 EVALUATION In its letter dated January 10, 1988, the licensee provided a list of components subject to an Appendix J exemption. The licensee indicated that these components are not able to be tested due to Technical Specification (TS) limitations while at power or due to the containment entry hazards involved in exposing personnel to high radiation and/or temperature levels. These components are the containment isolation barriers for the following system lines: drywell pneumatic supply, feedwater line, standby liquid control injection, LPCI loops, core spray iniection lines, condenser steam supply and vents, recirculation loop sample line, reactor head cooling, RPCCW to drywell coolers, drywell head manway double gasket seal, main steamline drain HPCI steam supply and pump suction, and scram discharge lines. The staff has reviewed these system line isolation barriers and finds the licensee's bases for Appendix J test deferral acceptable. These bases are sunrarized below.

The licensee has provided leakage test results and maintenance infonnation on these components for the past three testing programs conducted in 198'2, 1984 and 1985. The total "as found" leakage rates for these components, in standard cubic feet per hour (SCFH), are sumarized as follows: 70.44 for value, 328.75 for 1984, and 133.64 for 1985. The averace or "expected" leakage 1982 for the three tests is 200.29. The leakage rate for the remaining unexpired and recently tested components is 299 SCFH. 'The licensee also sumed the highest individiual component leakage for each component selected from the three tests. This total leakage rate, which is considered as a maximum credible limit, is 437 SCFH. Adding this result to the remaining l

Type B and Type C leakage total, yields a value of 736 SCFH or about 0.9La.

Adding the laroest "as found" LLRT leakaoe (which was the 1984 testi to the t

remaining Type B and Type C leakage totai yields a value of 6?7 SCFF (about 0.76La).

If the average value for the past three tests is used, the total leakage is A99 SCFH or 0.6La. This value can be considered as the expected leakage. Other than the mentioned worst cases, the combined Type B and Type C leakage rates for most of the tests are all below the 0.6La limit.

Paragraphs III.B.3 and III.C.3 of Appendix J states that the combined leakage rate for all penetrations and valves subfect to LLRTs shall be less than 0.6La. As seen from the licensee's test results, the "as found" leakage for 1984 test procram had exceeded the acceptance criterien of 0.6La but still remained within the La (821 SCFH) limit.

It should be noted that la is the limit used to compute dose consequences. Technically, if the worst case "as found" leakage is used, the leakage would still be below that used within the accident analysis. The acceptance criterion of 0.6La is applied to the "as left" leakage rates measured before the plant start-up to assure an adeounte safety margin. Furthennore, the "as found" leakage rate in the 1985 test was i

found to be decreased compared to the leakage found in 1984 after repairing i

and adiusting a few severely leaking valves. These test results showed that the condition of these problem components should not be expected to chance significantly dur ig the short extended test period.

l

i.

. The maintenance information provided by the licensee showed an average maintenance interval of 7.14 years per penetration (service interval ranging from 7.74 years to 1.53 years per penetration). The maintenance interval start date corresponds to the start-up date following the 1980 refueling outage (4-24-80) and ends with actual repair date unless the valve has required no service since the 1980 outage. The average "potential maintenance interval" is 7.01 years franging from 7.93 years to 2.31 years for each specified penetration). The potential maintenance interval is the time period from last valve refurbishment to March 26, 1988. These figures showed that the averace service required on each penetration is low. Therefore, these components are not expected to see a significant service interval during the short period of the test interval extension.

The staff has reviewed the licensee's submittal regarding the Appendix J test interval exempti.on request. Rased on the above discussion, the staff finds that for the 53 components, an exemption from the LLRT test frequency specified in Appendix J should be granted based on the following:

1.

The plant was in an extended refueling outage (Cycle 9) during which these components were not sub,iect to an operating environment. Because of the delayed restart for Cycle 10, these components would be sub,iect to much less than a 24 month service condition between tests. This gives a safety margin to reduce potential degradation of these components during the extended l

test interval.

2.

Historically, testing has shown low "as found" leakage with the exception of the 1984 test. The ample margin between the measured leakaoe and the allowable leakage should accommodate any degradation likely to be experienced for these components during the extended period. The higher leakace record in the 1984 test was improved after repairino and ad.iusting a few severely leaked valves.

3.

The intent of Appendix J was that Type B and C testing be performed during a refueling outage. The exemption would provide one-time relief from the requirement of Appendix J to allow a test interval extension for these components.

3.0 CONCLtlSION 4

Based on the above, the staff concludes that the licensee proposed extension of the test intervals for these components identified in its submittal are acceptable. This is a one-time exemption from the two-year Type B and Type C test interval requirements as prescribed in Appendix J and is intended to be in effect until March 26, 1988. This approval is based on the assumption that all other tests will be conducted in accordance with the requirements of Appendix J.

n.-,--

,..------,.,v.-,

..,,,._ _.--,-..-,_,_-...-__,-. - _-.,-.----._,_ -_ -..,-,_ -.