ML20248D466
| ML20248D466 | |
| Person / Time | |
|---|---|
| Site: | Pilgrim |
| Issue date: | 09/27/1989 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20248D458 | List: |
| References | |
| NUDOCS 8910040275 | |
| Download: ML20248D466 (7) | |
Text
- --
nacuq'o
~
' 4.-
UNITED STATES 8"
N l
NUCLEAR REGULATORY COMMISSION k
7, j
WASHINGTON, D. C. 20555 L
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION FACILITY OPERATING LICENSE NO. DPR-35 BOSTON EDISON COMPANY PILGRIM NUCLEAR POWER STATION DOCKET N0. 50-293 INTRODUCTION By letter dated July ~ 7,1989 and supplemented on September 1,1989, Boston Edison Company (BEC0/ Licensee) requested an exemption from the requirements of 10 CFR 50, Appendix J, Sections III.A.6.(b), "Additfor,al Requirements-Type A Test," III.D.2(a), " Type B test" and III.D.3., " Type C test." BEC0 has proposed to extend the test period for the Type A (Primary Containment Integrated Leakage Rate Test) test and certain Type B and C tests (Local Leakage Rate Tests)'until the next refueling outage which will begin in the March 1991 timeframe. This would be approximately a 22-month extension for the Type A test and approximately a six-month extension for the Type B and C tests. These exemptions are necessary to prevent a midcycle shutdown to perform the Type A, B and C tests. BECO has also requested, as part of the to be allowed to resume the Type A~ test schedule of Section
. exemption} of Appendix J; the staff will grant this request only oli the III.D.I(a
/
condition that the next Type A test is successful, as discussed below.
DISCUSSION 1.
Section III.A.6.(b) of 10 CFR 50, Appendix J, requires that:
"If two consecutive periodic Type A tests fail to meet the applicable acceptance criteria in III.A.S.(b), notwithstanding the periodic retest schedule of III.D, a Type A test shall be performed at each plant shutdown for refueling or approximately every 18 months, whichever occurs first, until two consecutive Type A tests meet the acceptance criteria in i
III.A.5(b), after which time the retest schedule specified in III.D may be resumed."
The Type A tests performed at Pilgrim during the last three refueling outages (1982, 1984, and 1987) failed the "as-found" acceptance criteria.
l BECO contemplated performing the next test during a Maintenance Outage originally scheduled for October 1989 (since rescheduled to Spring 1990).
Pilgrim's next refueling outage will begin in the March 1991 timeframe.
The refueling outage schedule is due to the length of the Pilgrim Restart Program. The 1991 Type A retest date would exceed Appendix J's 18-month requirement by approximately 22 months.
8910040275 890o27 PDR ADOCK 03000293 P
PDC 3
e 4'
y 2'
F A review of the 1982,1984, and 1987 Type A test results indicates that the "as-left" condition of the containment satisfactorily met the require-ments' of' Appendix J.. ~ However, "as-found" Type A tests did not meet the acceptance requirements and were deerned failures. The cause for.the "as-found" failures was ~ due to excessive Type B and C test leakages.
2.
Section III.D.2.(a) of 10 CFR 50, Appendix J, requires that:
p:
" Type B tests, except tests for air locks, shall be performed during reactor shutdown for refueling or other convenient intervals but in no case at intervals greater than 2 years."
. An exemption is requested only for the Drywell Head and Drywell Head Access Hatch. These Type B local leak rate tests (LLRT) leakages have his-torically not been a source of significant "as-found" leakage.
The exemption is requested to prevent the necessity of testing during a midcycle shutdown and allow the tests to be performed during the next refueling. outage (RFO-8).
3.
Section III.D.3 of 10 CFR 50, Appendix J,- requires that:
" Type C tests shall be performed during each reactor shutdown for refueling but in no case at intervals greater than two years."
Exemption is requested for four valves in two systems because testing at.
other than refueling outages presents significant difficulties for'these systems. BEC0 requests a schedular exemption be granted because the next scheduled refueling outage will exceed the provision of this requirement by approximately six months. Currently this testing would be required by November or December 1990.
The subject valves have good test histories, having passed at least the last three Type C tests.
EVALUATION Type A Test Pilgrim has failed three consecutive "as-found" Type A primary containment integrated leak rate tests (PCILRTs). BEC0 was required by Appendix J to increase the testing frequency for the ILRT from three times in ten years to once per outage until two consecutive "as-found" tests are acceptable.
BECO has stated that the cause for the "as-found" PCILRT failures has been due exclusively to Type B and C leakages. As such, BEC0 has increased its efforts to minimize Type B and C leakage, j
NRC Information Notice 85-71, " Containment Integrated Leak Rate Tests," dated August 22, 1985, states in part:
[m l
+
(
l.
Es 3
"If Type B and C leakage rates constitute an identified contributor to this failure of the "as-found" condition for the PCILRT, the general purpose of maintaining a high degree of containment integrity might be better served through an improved maintenance and testing i
program for containment penetration boundaries and isolation valves.
l In this' situation, the licensee may submit a Corrective Action Plan with an alternative leakage test program proposal as an exemption request for NRC staff review.
If this submittal is' approved by the NRC staff, the licensee may implement the corrective action and alternative leakage test program in lieu of the required. increase in Type A test frequency incurred after the failure of two successive Type A tests."
As an alternative to performing a Type A test in the near term, BECO proposes that'a more effective approach to maintaining a high degree of containment integrity is through implementing an improved maintenance and testing program for containment isolation valves and penetrations. The Local Leak Rate Testing Corrective Action Plan described in Attachment B to the subject letter 'provides details of the program for implementation at Pilgrim. This Corrective Action i
-Plan was developed using the guidance of NRC Information Notice'85-71, and is designed to eliminate the root cause of the successive Type A failures at Pilgrim by aggressively addressing leakage observed from the pathways of Types B and C LLRTs.
Prior to the development of the Corrective Action Plan, BECO established a LLRT Failure Analysis Team in October 1986 as a " standing entity" to conduct root cause analyses and to recommend and follow up on corrective actions for all local leak rate test failures that occurred during the last refueling outage.
By intensifying the level of attention at resolving the actual root causes for Type B and C test failures, the results of the "as-left" PCILRT conducted at-Pilgrim in December 1987. were the lowest leakage in the plant's history.
Enclosure 1 of the subject letter summarizes the root cause analyses and corrective actions taken for the LLRT failures at the last refueling outage. Also, BECO has already j
initiated more frequent LLRTs. Two additional LLRTs have been performed since the December 1987 ILRT "as-found" failure, and more testing is planned for Spring 1990.
The staff has reviewed BECO's submittal and concluded that it would be acceptable to extend the Type A test period until the next refueling outage.
The staff's conclusion is based on the following.
The "as-found" failure of the ILRT was due to excessive Type B and C leakage.
As such BEC0 has taken and will take actions to improve the Type B and C leakages. An important example of these actions follows:
The majority (83%) of the net equivalent LLRT 1eakage that caused the "as-found" failure of the Type A test in December 1987 was due to excessive leakage l
past the feedwater check valves. These feedwater check valves had a history of LLRT failures due to worn soft seats, leaking hinge pins, and cracked welds on bushings.
l
-j 1
l a
4
)
The failure root cause analysis performed by the LLRT Failure Analysis Team determined that the failure was caused by degraded soft seat material (Sargent SR740-70), and bushings and hinge pins from an older design. As short-term corrective action, the old hinge pins and bushings were replaced with new designs and the soft seat material was replaced with a new material (Parker I
E692) which has passed thermal aging tests simulating six years of accelerated thermal aging.
The following long-term corrective actions were identified:
o Continue to closely trend the LLRT performance of the refurbished valves.
BECO will open and inspect at least one feedwater check valve during o
each refueling outage to evaluate the new soft seat material's performance.
The minimal leakage experience during tests of these feedwater check valves thus far indicates the repairs have been successful.
The staff has reviewed the actions described by BECO and agrees these actions should improve leakage through historically poor penetrations and provide BECO a method to detect and focus its attention on future bad performers. Therefore, the staff would expect that the containment integrity will be maintained during the extension of the test period.
However, the staff is not prepared to grant the request to return to the normal Type A test schedule of Section III.D.1(a) of Appendix J until Pilgrim has passed at least one "as-found" Type A test. Such a success would demonstrate the effectiveness of the Corrective Action Plan and justify the full exemption from the increased test schedule that BECO seeks. Therefore, the staff has determined that Pilgrim may return to the normal test schedule of Appendix J,Section III.D.1(a), on the condition that the next Type A test passes on an "as-found" basis. Also, the staff has determined that the next Type A test may be performed during RFO-8.
Type B and C Tests BEC0 has requested that the Type B and C test intervals be extended for four valves in two systems and for the Drywell Head and Drywell Head Access Hatch, as follows:
1.
Exemption from the Type C LLRT Retest Requirement for Shutdown Cooling Suction Isolation Valves MO-1001-47 and MO-1001-50 Section III.D.3 of 10 CFR 50, Appendix J requires a Type C LLRT to be performed on all primary containment isolation valves during each refueling outage, but in no case at intervals greater than two years.
LLRTs on Shutdown Cooling Suction Isolation Valves M0-1001-47 and M0-1001-50 were last performed on November 28, 1988, near the end of RF0 7.
Accordingly, the next LLRT on these in-series isolation valves
~
- g. :
'k.
5 L
for primary containment Penetration X-12.would be required on or before.
p November 28, 1990. However, because RF0-8 will not occur until at least flarch 1,1991, the two-year intervil would be exceeded.
~
These valves are used to isolate the single suction line from the reactor waterfrecirculation system to the shutdown cooling system pumps. LLRTs on L
these valves during outages other than refueling outages would have an adverse impact on the control of reactor coolant temperature because the shutdown cooling system is the only normal means of. removing ' decay heat from the reactor vessel during short outages. A more appropriate-alternative would.be to perform the LLRTs on these two valves during RF0-8 while'the reactor is refueled and shutdown cooling is not required. Accordingly, BEC0 requests a one-time exemption from the maximum _ two-year LLRT retest interval for Shutdown Cooling Suction Isolation Valves M0-1001-47 and MD-1001-50. - These valves will be tested during RF0-8 and the. retest interval will exceed two years by approximately six months.
Shutdown Cooling Suction Isolaton Valves M0-1001-47 and M0-1001-50 have been LLRT tested four times each since May 1987. The only LLRT failure on these valves occurred in May 1987, when leakage of 13.5 sim was observed past Isolation Valve MD-1001-50. The valve was disassembled, reworked, and components were replacsd. Since the failure, Isolation Valve M0-1001-50 has successfully passed three LLRTs. The last LLRTs were performed on these valves on November 28, 1988 in anticipation of startup for Cycle 8.
The maximum two-year retest was required. However, because
. of the length of the Pilgrim Restart Program,-Cycle 8 will exceed this maximum two-year retest interval.
I The requested exemption would provide only temporary relief from the requirements of Section III.D.3 of 10 CFR 50, Appendix J to the extent I
that the Type C LLRTs for Shutdown Cooling Suction' Isolation Valves M0-1001-47 and M0-1001-50 would be permitted to exceed the maximum two-year retest interval by approximately six months. BEC0 has made good faith efforts to comply with the regulation based on the increased LLRT retest frequency on these valves since May 1987.
2.
Exemption from the Type C LLRT Retest Requirement for Reactor Building Closed Cooling Water (RBCCW) Isolation Valve MO-4002 and Check Valve 30-CK-432 LLRTs on RBCCW Isolation Valve MO-4002 and Check Valve 30-CK-432 were last performed on December 19, 1988, near the end of RF0-7. Accordingly, the next LLRT on these valves would be required by December 19, 1990. However, because RF0-8 is not scheduled to begin until at least March 1,1991, the two-year interval would be exceeded.
Train B of the RBCCW is the only closed-water system that penetrates the Pilgrim drywell.
It is used to provide component cooling inside the drywell for the drywell coolers, the oil coolers to both recirculation pumps, and the drywell fans. Outside the drywell, Train 8 of the RBCCW is used to provide component cooling for both control rod drive (CRD) pumps, 6
i
.D-l 6-both. reactor water cleanup (RWCU) pumps, and the heat exchanger on Train B.
of the fuel pool cooling system. Containment isolation.for Train B of the RBCCW is provided by Isolation Valve.M0-4002 on the line from the drywell (Penetration X-24) and by Check Valve 30-CK-432.on the line to the drywell-(Penetration-23). Only one isolation valve is provided on each -
penetration because Train 8 of the RBCCW. is a closed-water system and it does not provide a direct leakage flowpath from the drywell to secondary containment.
LLRT testing of RBCCW Isolation Valve M0-4002 and Check Valve 30-CK-432 requires plant operators to close the valves to each component in the drywell served by Train 8 of RBCCW. This is done to isolate any leakage that would remain within the drywell and need not be measured.
In addition, staging must be built above the torus to provide access to the valves for the LLRTs. After the tests, the staging must be removed. All of this work to manipulate the valves of affected drywell components, build and remove staging, and perform the LLRTs would result in radiation exposure (approximately 1.41 person rem) of plant workers.
RBCCW water contains nitrites for corrosion control and approximately 2,500 gallons of this water must be drained and treated by the radwaste demineralized during the LLRT tests on these valves. Because nitrites act to break down the demineralized resins, this water must be carefully collected and treated by the radwaste demineralized'in measured amounts.
In any case, the treatment of this high-nitrite water would result in more frequently changed resin and ultimately result in greater production of radwaste.
In addition, performance of the LLRT tests on RBCCW Isolation Valve M0-4002 and Check Valve 30-CK-432 would have a large impact on plant operation because all components cooled by Train B of the RBCCW would be inoperable during the tests.. The affected systems include the drywell coolers, both recirculation pumps, both CRD pumps, both RWCU pumps, and Train B of the fuel pool cooling system.
LLRTs on RBCCW Isolation Valve M0-4002 and Check Valve 30-CK-432 performed since December 1981 have been successful, with indicated leakage on each valve measuring 0.1 sim. A leakage of 0.1 slm corresponds to the minimum sensitivity of the leakage detection instrumentation.
In addition, LLRTs on these valves have been performed successfully three times each since July 1987 The requested exemption would provide only temporary relief from the requirements of Section III.D.3 of 10 CFR 50, Appendix J to the extent that the Type C LLRTs for RBCCW Isolation Valve M0-4002 and Check Valve 30-CK-432 would be permitted to exceed the maximum two-year retest interval.by approximately six months. BEC0 has made good faith efforts to comply with the regulation based on the increased LLRT retest frequency on these valves since July 1987.
n n,..
L.
3.
Exemption from the Type B LLRT Retest Requirement for the' Drywell Head and Drywell Head Access Hatch Section III.D.2(a)-of 10 CFR 50, Appendix J' requires a Type B LLRT to be performed on all primary containment penetrations, whose design incorporates components such as resilient seals and gaskets, during each refueling outage, but in no case at intervals greater than two years. The LLRTs on the Drywell Head and Drywell Head Access Hatch were last performed on
. November 29, 1988, near the end of RFO-7. Accordingly, the next LLRT on.
these penetrations would be required by November 29, 1990. However, because RFO-8 is currently not scheduled to begin until at least March 1,1991, the two-year interval would be exceeded.
To' perform the LLRT tests on the Drywell Head and Drywell Head Access Hatch, seven of the nine shield blocks that rest above the drywell must be removed. The shield blocks are normally removed only during refueling outages when the Drywell Head must be removed for fuel loading and unloading.
In this circumstance, the shield blocks would be moved only to allow the LLRTs. Also, the work to perform these LLRTS would be in a high radiation area and would result in radiation exposure (approximately 0.6 person rem) of plant personnel.
The Drywell Head and Drywell Head Access Hatch have lifetime histories of successful LLRTs at Pilgrim, with indicated leakage on each penetration measuring 0.1 sim. A leakage of 0.1 slm corresponds to the minimum sensitivity of the leakage instrumentation.
The requested exemption would provide only temporary relief from the requirements of Section III.D.2(a) of 10 CFR 50, Appendix J to the extent that the Type B LLRTs for the Drywell Head and Drywell Head Access Hatch would be permitted to exceed the maximum two-year retest interval by approximately six months.
Based on the above review, the staff has determined that extending the surveillance intervals as described is acceptable and the requested temporary exemption for Type B and C test schedules should be granted.
CONCLUSION The staff has concluded, based on the above, that the next Type A test may be performed during RFO-8 and the licensee may return to the normal test schedule of Section III.D.1(a) of Appendix J to 10 CFR Part 50. This conclusion is based on the condition that the next Type A test meets the NRC acceptance criteria on an "as-found" basis. The staff further concludes that a one-time schedule extension from the required test intervals of Section III.D.2(a) and III.D.3 of Appendix J to 10 CFR Part 50 should be granted for the containment penetrations discussed above. The required tests will be performed during RFO-8.
~ ~
Principal Contributor:
J. Pulsipher Dated: September 27, 1989
_ _ -