ML20248H791
| ML20248H791 | |
| Person / Time | |
|---|---|
| Site: | Point Beach  |
| Issue date: | 10/03/1989 |
| From: | Fay C WISCONSIN ELECTRIC POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| CON-NRC-89-119 GL-89-04, GL-89-4, VPNPD-89-519, NUDOCS 8910120082 | |
| Download: ML20248H791 (25) | |
Text
, - - _ _ - - _, - - -
l,e n
+
l.
WSCORSin Electnc ma cown 231 W. MICHIGAN, P.O. BOX 2046, M8LWAUKEE, WI 53201 (414) 221-2345 VPNPD-39-519 j -
NRC-89-119 October:3, 1989 U.
S. NUCLEAR. REGULATORY COMMISSION Document Control Desk Mail Station P1-137
-Washington, D.-C.
20555 Gentlemen:
l-DOCKETS 50-266 AND 50-301
' GUIDANCE ON-IN-SERVICE TESTING PROGRAMS GENERIC LETTER 89-0_4
-POINT BEACH NUCLEAR PLANT, UNITS 1 AND 2 i
Enclosed'is our response to Generic Letter (GL) 89-04,
" Guidance on Developing Acceptable In-Service Testing Programs."
Upon receipt of GL 89-04 on April 14, 1989, we began a preliminary comparison between our in-service testing
~
program-and implementing procedures and.the guidance provided in your letter.
On June. 13, 1989,.our representative attended the Region III public meeting on GL 89-04.
On July 5, 1989, we submitted a letter requesting an extension of the deadline for i
our'GL 89-0-4 response, along with.a request to use ASME O&M i
Standards Part 6 and 10 in preparation;of our next ten-year l
interval IST program.. These requests were.made'because our new IST program submittal is required'in December 1990.
We believed.that our efforts, as well as the NRC review efforts, would be more efficient if'the GL 89-04 response and our next program description were developed and reviewed as one package.
On August 17, 1989, during a telephone conference with our NRC Project Manager, Warren Suenson, and other NRC representatives, we'were informed that our July 5, 1989 requests were denied.
We were further instructed to respond to GL 89-04 by the original deadline (six months from date of issuance) and, for areas where full compliance would not be-achieved, to state our status, plans, and schedules for compliance.
We were informed that written correspondence regarding our July 5, 1989 request i
documenting the August 17, 1989 phone conversation would i
follow.
To date, however, such confirmation has not been received.
8910120082 691003 1R ADOCK 05000266 l
PDC g
a-E NRC-Document Control Desk october.3, 1989 j
p
_Page 2L j
l Since Point Beach Nuclear Plant is not-listed in either Table 1 Lor Table 2 of GL 89-04, we understand that GL189-04 constitutes-approval of our: presently submitted IST program and relief requests, provided we comply with the GL 89-04 Attachment 1
' positions.
-For those cases where Wisconsin Electric does not comply with the GL 89-04 Attachment lLpositions, it is our j
understanding that separate, specific relief: requests must be submitted..Further, we understand that these relief requests should not be reflected in our implementing procedures until the request is approved by the NRC.
As mentioned previously, our next ten-year inspection interval begins in December 1990.
The attachment to this letter is an item-by-item discussion of the' positions. presented in Attachment 1 to GL 89-04.
This-
' discussion includes our schedule for modifications or procedural changes, if such are necessary.
Also included is a l
one-page summary of our position on'each item.
Please contact us if you have any questions concerning this R
submittal.
Very truly~yours, 62
/
C.
'. Fay
('
i
.Vice. President Nuclear Power Enclosure Copies to NRC Regional Administrator, Region III NRC Resident Inspector l
l
.i
~i:
i i
l i
Attachment
. Response'to Generic Letter 89-04
" Guidance on Inservice Testing Program" Wisconsin Electric Power Company i
I.
1 i
^
RESPONSE TO GL 89-04 ATTACHMENT
- I Page 1-COMPLIANCE STATUS
SUMMARY
GL 89-04 Item Compliance / Status 1
No Need to complete special testing of service water system.
2 Yes Except alternate testing per page'3, of GL 89-04 was selected for some valves.
3 No Physical modifications required.
4 Yes 5
No Following the corrective actions described compliance will be achieved.
6 Yes 7
(Not applicable to PWRs) 8 Yes 9
No a.
Evaluation of containment
}
spray system needs to be done.
Both units to be completed by June 1990.
b.
Special testing of service water system scheduled to be done by January 1990.
10 Yes Except that limits apply by penetration which may be more than one valve.
11 No Scope review to be completed as part of our next 10-year IST cycle, December 1990.
L RESPONSE TO GL 89-04 ATTACHMENT' Page 2 RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 1,
" FULL FLOW TESTING OF CHECK VALVES" For normally closed check valves, when fluid flow is used to demonstrate opening, the ASME code (IWV-3522) states, "For swing or tilting disk valves, if the test is made by use of fluid flow through the valve, the pressure differential fer equivalent flow shall be no greater than that observed during the preoperational test.
For other types of check valves, it shall be shown that disk movement is sufficient to provide a flow area at least 50% of the area of the seat port, or to permit flow adequate for the function of the valve."
Further, if the check valve is also a Category A valve, IWV-3412 states:
a.
Valves shall be exercised to the position required to fulfill their function unless such operation is not practical during plant operation.
If only limited operation is practical during plant operation, the valves shall be part stroke exercised during plant operation and full stroke exercised during plant cold shutdowns.
Valves that cannot be exercised during plant operation shall be specifically identified by the Owner and shall be full-stroke exercised during cold shutdowns.
Full-stroke exercising during cold shutdowns for all valves not full-stroke exercised during plant operation shall be on a frequency determined by the intervals between shutdowns as follows: for intervals of three months or longer, exercise during each shutdown; for intervals of less than three months, full-stroke exercise is not required unless three months have passed since last shutdown exercise.
b.
The necessary valve disk movement shall be determined by exercising the valve while observing an appropriate indicator which signals the required change of disk position j
or observing indirect evidence, such as changes in system j
pressure, flow rate, level, or temperature, which reflect i
stem or disk position.
It is our understanding that the intent of GL 89-04, Item 1 is to specify the passage of the maximum required accident flow as the only acceptable flow mechanism allowed to satisfy the code requirements shown above.
l i
RESPONSE TO GL 89-04 ATTACHMENT Page 3 Our IST program presently contains 96 check valves which are required to open to perform their function.
Of this group, 82 valves are presently exercised.
These check valves fall into several distinct categories as discussed belcw.
The other 14 valves are discussed under the response to GL 89-04, Item 2.
1.
Accident required flow known and achieved during testing.
Forty-eight valves are included; we presently comply with GL 89-04 position 1 for these valves.
2.
Accident required flows known but implementing procedures need to be changed to ensure the full flow through valves is achieved.
Twenty-eight valves fall into this category.
Compliance with GL 89-04 will be achieved with the implementing procedure changes.
Procedure changes are scheduled for completion by July 1990.
3.
Six service water pump discharge check valves.
These valves are presently tested using a modified full flow system configuration.
Recently completed modifications have installed flow indicators and implementing procedure changes are being made to control their use.
It may be possible to comply with the GL 89-04 Item 1 position, provided proper system alignments are achievable; however, further plant modifications may be needed.
The possibility of using disassembly and inspection (GL 89-04, Item 2) does exist.
Special testing will be done by December 31, 1989, to determine:
(1) feasibility of full flow testing, and (2) possibility of modifications.
We will inform you of the results of our tests, including the need and schedule for modifications or disassembly and inspection following our testing.
i
)
l 1
J
p-r e -
e RESPONSE TO GL 89-04 ATTACHMENT Page 4
~
' RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 2,
" ALTERNATIVE TO FULL FLOW TESTING OF' CHECK VALVES" There are-14 check valves which-are not full-stroke exercised due to~ plant limitations.
Eight of these valves are disassembled and inspected.every refueling outage.
Four valves are disassembled and inspected at least once every 10 years.
The remaining two valves, 1SI-842B and 2SI-842B (accumulator discharge check valves),-
are not instected due to plant limitations on isolating these valves.
All'of these valves are partially exercised.
The eight valves which are inspected every refueling are opened, manually stroked, and inspected.
The inspection procedures for these valves are presently general in nature.
A procedure encompassing the recommendations of GL 89-04 is-being developed
.which will contain' specific directions on inspection and I
disassembly of check valves.
This procedure is scheduled for completion by September 1990.
Until this procedure is completed, the existing instructions are considered adequate.
Generic' Letter 89-04 states that all valves should be disassembled and inspected on a frequency of no greater than six years.
It does allow for an extension of this interval to longer than every six years.
The letter. states that for this extension, the licensee should develop.the following information:
1.
Disassemble and inspect each valve in the group.
2.
Review industry experience regarding the same type of valve j
used in similar service.
3.
Review the installation of each valve addressing the "EPRI
' Application Guidelines for Check Valves in Nuclear Power Plant" for problematic locations.
We have reviewed these points in regard to the four valves inspected at least once per 10 years and the two valves which are not inspected.
These six valves comprise a single group.
f l
RESPONSE TO GL 89-04 ATTACHMENT I
Page 5 I
Disassembly and Inspection 1
Four valves, two per unit (SI-842A and SI-867A), have been j
opened and inspected.
The results of these inspections are documented and have been reviewed.
No problems were noted which could ~ affect valve operability.
Two other valves (lSI-842B and 2SI-842B), will be disassembled and inspected during the upcoming refueling outages.
These results.will be reviewed when available.
Review of Industry Experience A listing of the reported failures for valves of this type was requested from the INPO NPRDS data.
There were 1
22 failures reported, including three failures at Point l
Beach.
All of these failures involved leakage past the j
seat.
There were no reported cases of failure of these valves to open, or operate freely.
Review of EPRI Installation Guidelines These six valves have been reviewed against the EPRI Guidelines.
It was noted that all but one valve (lSI-842A) is located close to a source of turbulence.
The past maintenance history has identified no problems.
Recommendations as a result of this review was to inspect and replace the retaining block hold down studs.
This is being planned for future refuelings.
No other action is suggested.
Additional Discussion These six valves were discussed at length during our response of April 2, 1987, to the draft safety evaluation report and technical evaluation report addressing our Inservice Testing Program.
The discussion is contained in Item 4 of that letter.
A copy of that discussion is also contained in Appendix A of this response.
N RESPONSE TO GL 89-04 o
' ATTACHMENT Page 6 We believe_our existing program of disassembly and inspection is adequate based on the above information.
The results of the inspection of the additional valves will be reviewed, when completed.
If any problems are identified during these inspections, we will review this position again.
The past valve maintenance history and test performance leaves us confident that no problems will be found.
We are not in full compliance with GL 89-04 Item 2 as it is presented, but we have addressed the issues discussed on page 3 of GL 89-04 regarding alternative testing due to design considerations.
Documentation of'the justification for the above will be included in our IST program.
l l
_m_.-_______-_m.
__.________.--___.m
RESPONSE TO GL 89-04 ATTACEMENT Page 7 RESPONSE TO'GL 89-04, ATTACHMENT 1, ITEM 3, i
" BACKFLOW TESTING OF CHECK VALVES" Our review was' conducted on check valves presently contained in our.IST program with respect to proper classification and testing.
Valves currently not in our program which may fall under the description of this item will be addressed during our scope review (see Item 11).
The program review identified 30 check valves that should be Category A/C and are presently neither tested for backleakage nor have a backleakage limit specified.
The physical configuration of the systems containing these valves make it impossible to test them in accordance with IWV-3434a or b.
The program review also identified 66 check valves that do not have their position verified in accordance with IWV-3522.
One group of eight check valves, the main feedwater line check valves are not tested. individually as required by IWV-3424.
These valves are tested in series, in accordance with the request for relief, VRR-21, included with our present IST program.
We consider the existing request for relief still valid and do not intend to submit an additional request.
l 1
-RESPONSE TO GL 89-04 ATTACHMENT Page 8 RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 4,
" PRESSURE ISOLATION VALVES" a.
All pressure isolation valves listed in Technical Specifications are identified in the IST program.
These are only the Event V valves.
These valves are tested using Technical Specification implementing procedures rather than IST program implementing procedures.
b.
The Technical Specification implementing procedures used to test Event V PIVs were reviewed.
All Event V check valves are individually tested.
Measured leak rates for individual valves are compared with Technical Specification acceptability limits.
l
L l
I RESPONSE TO GL 89-04 ATTACHMENT Page 9 RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 5,
" LIMITING VALVES OF FULL-STROKE TIMES FOR POWER OPERATED VALVES" Stroke time limits prasently used at Point Beach were established over nine years ago using historical performance data and component design information.
In accordance with GL 89-04, stroke time limits were reviewed relative to safety function.
The results of this review found 33 time limits to be inconsistent with safety analysis requirements.
This group of valves can be divided into four distinct categories.
The information provided demonstrates that we will comply with the Item 5 position.
1.
Performance potentially beyond safety analysis requirements.
Twelve valves had operating times which apparently exceed the safety analysis requirements.
The validity of these stroke time requirements is being investigated.
If stroke times are in excess of those actually required for the system to perform its function, these valves will be adjusted, if possible, to limit their stroke times to within the safety analysis.
If adjustment is not possible, equipment modification or replacement will be done.
A schedule cannot be determined until the appropriate corrective mechanism can be identified.
Additionally, procedure changes will_be submitted, if needed, to make the IST program limits fall within the safety analysis requirements.
This action will be completed by i
December 1990.
I l
2.
Actual performance within safety analysis requirements.
Nine valves have IST limits outside the safety analysis requirements but their actual stroke times have not exceeded the safety analysis requirements.
Procedure changes have been submitted to make the IST program limits fall within the safety analysis requirements.
These procedure changes will be completed by February 1990.
3.
Pressurizer PORVs Approximately six months ago it was recognized that these four valves did not have an appropriate limit in the IST program.
Changes have been submitted to make the 1
IST program limits appropriate.
These changes will be
(
implemented by February 1990.
l 1
J i
'0
'- i RESPONSE TOLGL'89-04' ATTACHMENT
'Page 10l It was.also identified that'the PORVs were not stroking
'withincthe safety' analysis. requirements. 1 Temporary
. modifications,are presently completed or scheduled.to meet the safety' analysis' requirements.
Permanent modifications are planned.... Unit 2 is scheduled for completion during its refueling outage which: began September 22, 1989, and Unit 1-
.isLscheduled for the spring 1990 refueling outage.
4.
Purge' Supply.&. Exhaust Valves The safety: analysis limit for these eight valves-is based'on their containment isolation function, shutting from.the open position._
Initial' plant design anticipated that the purge.
supply and exhaust valves would be open during reactor power operation.
It has been over a decade since the valves have been open l
during-power; operations.
Technical Specifications require
-that;they are normally shut,--with electrical power removed.
The valves.are only opened, except for repair, when the unit is in.the cold;and refueling shutdown conditions thus changing theLscope of their safety analysis function.
We.will review the function of the purge supply and exhaust valves to determine appropriate' stroke time limits.
We1will update the IST program limits as needed.
The review-and-i program update will be be included iniour 1990 IST program
.(see. Item 11).
9 5
RESPONSE TO.GL 89 ATTACHMENT
- Page-11
^!
l I
i RESPONSE TO GL 89-04, ATTACHMENT 1,. ITEM 6
" STROKE TIME MEASUREMENTS FOR RAPID-ACTING VALVES" The IST program implementing procedure changes and analysis procedure changes'to reflect Item 6.are presently being reviewed and will be issued by February 1990.
The IST program will identify the maximum stroke time for rapid' acting valves at 2.0 seconds..
The analysis procedures will associate " operability" at the time limit of 2.0 seconds.
The IST program identifies " rapid acting" valves by referencing
-VRR-0.
Our PORVs were recently defined as being in the " rapid acting" category.
A change to VRR-O has been developed for addition to the IST' program to document the status of the FORVs.
We~have numerous valves which typically operate in 2.0 seconds or less but we have not-encountered the problem described in GL 89-04.
We believe this occurs because we do not round off to the nearest second, rather we record stroke time to the nearest one-tenth second.
Since'we have not experienced difficulty, we will continue to perform the analysis as we have, not identifying valves as rapid acting.
L_ _ _ _ ______ _ _____ __ _ _ _ _______
a l
l RESPONSE TO'GL 89-04' ATTACHMENT.
1Page112
]
RESPONSE'TO GL 89-04, ATTACHMENT 1, ITEM 8,
'l
" STARTING POINT FOR' TIME PERIOD IN TECHNICAL SPECIFICATION ACTION STATEMENTS" i
j For.approximately three years, Point Beach Nuclear Plant has been evaluating IST data against operability limits-immediately upon test completion and making subsequent inoperability j
declarations when limits were exceeded.
-We'will continue this practice.
It is presently described in plant procedures.
We are making a procedure' change to include
~a statement of this policy in our IST program and will add a l
. statement to the analysis section of our implementing procedures.
This' change should be completed and issued by February 1990.
4 l
j vu
RESPONSE TO GL 89-04 ATTACHMENT Page '13 RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 9,
" PUMP TESTING USING MINIMUM FLOW RETURN LINE WITH OR WITHOUT FLOW MEASURING DEVICES" l
Presently we test safety injection, residual heat removal, motor-driven auxiliary feedwater, turbine-driven auxiliary feedwater in compliance with the positions stated and allowed by i
Item 9 of GL 89-04.
The containment spray pumps do not have full flow test capabilities.
The testing of the containment spray pumps is performed using mini-flow recirc only.
A modification is required to install suction pressure gauges for these pumps.
Completion of these modifications is scheduled during the 1990 refueling outages.
Presently the recirc path does not contain a flow indicator.
An existing alternate path may be available for use during periodic testing that does contain a flow indicator.
This configuration maintains only single valve isolation between the spray pump discharge and the spray header in containment.
Special testing will be conducted to determine if this flow path is viable for routine testing.
Unit 2 testing should be complete by December 31, 1989, and Unit 1 by June 1, 1990.
If the alternate i
path is not viable, recirc line flow indicators will need to be installed.
If modifications are needed, we will inform you by September 1990, with a proposed schedule for completing the modifications.
The service water system does not have mini-f]ow recirc capabilities.
The pumps are presently tested using a modified full flow configuration and are not in full compliance with Item 9 criteria.
Recently modifications were completed to install flow indicators and implementing procedure changes are being made to j
control their use.
It may be possible to comply with Item 9 j
criteria provided proper system alignments are achievable; j
additional plant modifications ray be needed.
Special testing will be done by December 31, 1989, to determine the:
(1) feasibility of single-pump full flow testing and (2) necessity for modifications.
If modifications are needed, we will inform you by March 1990 of the schedule for installation.
We presently include the spent fuel pool recirculation cooling I
pumps in our IST program.
These pumps are full-flow tested and j
have no recirc flow test capabilities.
1 I
1 l
u
' ' RESPONSE TO GL 89-04 ATTACHMENT 1
Page 14 l
i RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 10,
" CONTAINMENT ISOLATION VALVE TESTING" i
j We do not strictly comply with IWV-3426 in that there are cases
]
in which specific limits are not applied to individual valves.
In such cases, a limit is applied to a group of valves which make o
up a given penetration.
By limiting the penetration leakage to some conservative value, we ensure that no single valve can have leakage greater than the allowed penetration leakage.
We believe this fulfills the intent of Item 10.
Currently, the leakage of any penetration is specifically noted when it exceeds 2000 SCCM.
This limit is approximately the total allowed Technical Specification containment (Type C) leakage limit divided by the total number of leak paths.
However, we have not required repair or replacement of valves which comprise a penetration which exceeds 2000 SCCM leakage; we also review total Type C leakage to determine when repairs are needed.
In order to comply with Item 10, we will require repair or replacement at 10000 SCCM regardless of total containment leakage.
We will continue to use our 2000 SCCM administrative limit as we have in the past.
This will be incorporated in our program by June 1990.
Please note, however, the 10000 SCCM leakage limit criteria wi]l be administered by engineering personnel and will not be reflected in individual implementing procedures.
l
_______.__-__________m_
1 RESPONSE TO GL 89-04 ATTACHMENT j
Page'15-I RESPONSE TO GL 89-04, ATTACHMENT 1, ITEM 11, "IST PROGRAM SCOPE" Ongoing scope review has been conducted for several years.
{
Approximately two years ago the spent fuel pool cooling system was included in the IST program.
Recently it was decided to add portions of the fuel oil transfer system, portions of the instrument air system and portions of the component cooling water system in our IST program.
Inclusion of these systems in the IST program as well as preparation of implementing procedures are currently in progress.
We propose to complete a comprehensive scope review, relative to plant design, system function, GL 89-04 and code scope statements in order to prepare our next 10-year IST program due by December 1990.
Appropriate implementing procedures and other support documentation will also be in place by December 1990.
Any relief requests submitted with the 1990 plan will not be implemented until NRC approval is received.
Until such approval is received, we will continue to test in accordance with the present program or, for newly added components, we will not conduct any testing until relief is approved.
l I
l
w7 -
.t
' -RESPONSE TO GL 89-04
- ATTACHMENT.
APPENDIX A SIX SELECTED PAGES PERTAINING TO ITEM 4 FROM APRIL'2,.1987 LETTER FROM WISCONSIN ELECTRIC POWER COMPANY TO THE NUCLEAR REGULATORY COMMISSION 4
I I
'I J
i Document Control Desk April 2,.1987 Page 6 1
.ent than'the other four systems for which flow rate measure-ment relief requests were submitted in that generally two of the six pumps are always operating at or near design condi-
.tions to provide the required system heat removal capability.
Ilowever, the existing instrumentation configuration does not measure or allow derivation of the total system flow rate.
Therefore, if'the service water system can be reasonably i
modified to provide the measurement of the system flow rate of each of the six service water pumps, then the data would enhance our testing program and capability to monitor the performance of the service water pumps.
An engineering review of the service water system will be conducted within the next nine months to determine the feasibility of installing flow instrumentation on the ser-vice water system.
By January 15, 1988, we will provide-the results of this evaluation.
4.
PERIODICITY TO DISASSEMBLE, INSPECT, AND MANUALLY FULL STROKE EXERCISE VALVES 842A, 842B, 867A, AND 867B, VALVE RELIEF REQUEST NO. 4 We had proposed to disassemble, inspect, and manually full-stroke exercise the accumulator discharge (842A and 842B) and combined accumulator / safety injection discharge (867A) check valves once every ten years and to partial-stroke exercise the valves on a cold shutdown frequency in lieu of the requirements of Section XI, IWV-3400 and 3500.
NRC Exception and Comment No. 7 The Draft SER maintains that valve disassembly, inspection and manual stroking is an acceptable alternate testing method to full-stroke exercising check valves that cannot be full-stroke exercised with system flow.
However, the Draft SER also stated that our proposed disassembly frequency wat not in accordance with the staff guidelines which prescribe a refueling outage frequency.
Accordingly, it was suggested that we test these valves on a refueling outage frequency either individually or on a group sampling basis.
Response
Figure VRR 4-1 (Attachment 2) is provided to help clarify the following discussion about the function, maintenance history and proposed relief requested in regards to the safety injec-tion accumulator check valves.
Valves 842A, 842B, 867A, and
k 2
m 3
4 Doc'ument' Control Desk i
' April 2, 1987 L
page'7.
867B will stroke open if the SI Accumulators were to dump their contents to the reactor coolant system.
Valves 867A and 867B will. stroke open when the high head safety inspec-y tion pumps are supplying water to the reactor coclant system.
Additionally, valve 867B will' stroke open anytime RIIR cool-down is initiated.
Check valves 842A, 842B, 867A, and 867B are. identical 10" stainless steel, Darling swing check valves.
The eight identical valves of Unit 1 and Unit 2 p
operate under similar service conditions.
'Since 1977, valves 842A, 842B, 867A and 867B have been par-tially stroked open at least annually during the transition from hot. shutdown to cold shutdown.
By reducing reactor coolant system pressure in a controlled manner to a pressure which is less than that of the SI accumulators, the differ-ential pressure across the check valves is sufficient to open the valves and initiate flow into the reactor coolant system from the SI accumulators.
Although the flow rate through the check valves is not the design flow rate, it does verify freedom of movement of the valve disc to a partially open position.
Since 1974, valvas 867A and 867B have been stroked open each annual refueling outage during the full-flow testing of the high head' safety injection pumps.
The flow of one SI pump, at approximately 80% of design capacity (700 gpm), is direc-ted through each of these valves separately. Although full design flow through the check valves is not achieved during this test, sufficient flow is passed to stroke the valve substantially and verify freedom of disc movement.
Since each unit was placed into commercial operation, i
approximately 16 years ago, valve 867D has also been stroked open at least annually during initiation of cooldown on RilR.
During conditions requiring RilR, approximately 2,200 gpm are passed through the valve.
This flow rate is sufficient to fully stroke the valve open.
Over the last 16 years we have observed nothing which would be indicative of a problem that would inhibit any of the check valves' ability to stroke fully open.
The 867A check valves on both units were opened and inspected after approximately six years of service due to suspected seat leakage.
In both cases, seating surface wear was observed but no problems were noted with either valve's ability to stroke open freely.
The successful operation of valves 867A and 867B for more than 32 reactor years indicates that the valves are reliable. The occurrence of multiple generic failures within the group of eight identical valves is unlikely.
L Document Control Desk April 2,
1987 Page 8 Because of their elevation and their proximity to the resi-dual heat removal connection to the reactor coolant system, Valves 867B and 842B cannot ba opened and inspected unless the entire core is unloaded and the reactor coolant system is drained to the elevation of the reactor vessel nozzles.
The need to achieve this plant condition is rare.
It has only occurred once for Unit 1 and once for Unit 2.
To achieve the required plant condition and to disassemble and inspect Valve 842B would require approximately five addi-tional critical path days.
valves 867A and 842A in both units are at an elevation and location such that they can be opened for inspection during normal refueling outages without impacting a normal refueling outage schedule.
Because of the additional outage time which would be required to disassemble and inspect Valve 842B, our alter-nate t? sting program for the eight valves does not include full-stroking or disassembly and inspection of Valve 842B.
We believe that based on the maintenance history to date of all eight identical valves, the partial stroke testing of Valve 842B, and the testing progr.am for the other six valves provides sufficient assurance that Valve 842B will perform its design function.
Based on the above discussion, we are proposing the alter-nate testing for the safety injection accumulator check valves outlined in our Valve Relief Request No.
4.
5.
EXERCISING Tile CONTAINMENT SPRAY CHEMICAL ADDITIVE TANK VACUUM BREAKERS (VALVES 840A AND 8408), VALVE RELIEF REQUEST NO. 5 Since in-place testing of Valves 840A and 840B is undesirable due to the nature of the spray additive solution and the system arrangement, to fulfill the testing requirements of IWV-3520 would require valve removal.
However, to maximize system availability we had proposed not testing the valves at operation or during cold shudowns.
Alternatively, we had proposed exercising these valves during reactor refueling outages.
NRC Exception and Comment No. 8 The Draft SER found the relief request acceptable provided that the additional information which was provided during the November 1 and 2, 1983, working meeting is formally documented.
lC
~
' (* i.
4 VALVE RELIEF REQUEST NO. 4-
}:>
System:
K Safety injection, Units 1 and 2.
Component:
867A, 842A, 842B Category A/C Class:
1 867A 2'
842A&B Functions-Valves 867A, 842A, as.
842B open with differential pressure to provide flow from the SI accumulators and/or the SI pumps to the reactor coolant system during an accident.
These valves are normally shut.
In the shut position, these valves also serve as reactor coolant system pressure isolation valves.
b Basis For Relief:
During normal operation,' safety injection pump discharge pressure of 1500 psig or accumulator pressure of 760 psig is not sufficient to overcome reactor coolant system pressure.
Full or partial stroke testing is, therefore, not possible.
7 During cold shutdowns, partial or full stroke testing via-the use of the accumulators or safety injection pumps is not allowed so as to prevent the possibility of a low temperature g
overpressurization event.
A full stroke test by dumping the accumulator to the reactor coolant system could be possible during refueling, when the reactor vessel head is removed, but the volume and flow rate required for the test could damage core internals.
There would also be the possibility of forcing a nitrogen bubble through the reactor coolant system and refueling cavity resulting in possible safety implications which makes this testing concept inadvisable.
Alternate Testing:
The following alternate testing will be conducted on the SI accumulator check valves.
1.
A partial stroke test of 867A, 8678, 842A, and 842B wi.11 be conducted during the transition from hot shutdown to cold shutdown.
This will be considered a cold shutdown test. This test will not be performed if it will disturb an " Event V" valve which is not required to be tested within the associated cold shutdown. At a minimum, however, this test will be performed once every refueling outaoe.
f
- +
Page 2 i
VALVE RELIEF REQUEST NO. 4 (Cont.)
2.
A seat leakage test _ of 667A and 867B will be conducted in accordance with Technical Specification 15.3.16,
" Reactor Coolant System Pressure Isolation valve Leakage Tests."
3.
A seat leakage test will be performed quarterly coincident with the SI pump tests on 842A and 8428.
A seat leakage rate of 5 gpm or less will be considered acceptable.
4.
A full stroke test of 867B will be conducted once each cold shutdown while~on RilR cooling.
5.
Once within each 120-month inspection interval, f
valves 1-867A, 1-842A, 2-867A, and 2-842A will be opened
!~
and.their discs will be checked to verify freedom of movement. The inspection will be staggered such that one valve from the gtoup of four (includes both 'Jnit I and Unit 2 valves) will be opened and inspected approximately every two to three years.
If a condition is discovered during the inspection of a given valve that would have prevented it from stroking fully open, the inspection sample will be expanded.
A second identical check valve in the same unit will be opened and inspected.
Also, during the next refueling outage on the opposite unit, the sister valve to the inoperable valve will also be inspected.
If a second valve is found inoperable in the expanded sample, all five remaining check valves from the group of eight will i
be inspected.
The group of eight check valves includes 1-842A, 1-842B, 1-867A, 1-867B, 2-867A, 2-8678, 2-842A, 2-842B.
If the inspection must. be expanded to include all eight valves, the inspection of those valves in the unit which is not in a refueling shutdown condition shall be performed during the next regularly scheduled refueling shutdown.
1 j
Status:
Submitted for review Document:
1
_________.a_______-
- - - - - - - - - - - - - ^ - - - - - - - - - - - ' - - - -
-cn as/
fgyg>[
ygg
_/
SI ACCunutArorg ;cygtx g4ty35 1
.SI ACCUM l
M
FROM sZ PUMPS
(
l k
n a
y d ddP A l
d>
/j
'l g
8474 8424 8 +'4 h
sr.
ACCT /M j.
s.
PHR suPPLV
- FROM SZ* Pt/MPs l
Loop g G
G c o s.D t r y t/1
/
GQ 1-8678 9928 ggg
-___ ___ __ ____ ___ __ _ ____ _ _ _ _ _ _