ML20086K372
| ML20086K372 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 07/14/1995 |
| From: | Hunger G PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9507200120 | |
| Download: ML20086K372 (6) | |
Text
Cteti n Support Dep;rtment I
{
PECO Energy Cornpany Nuclear Group Headquarters 9G5 Chesterbrook Boulevard Wayne, PA 19087-5691 July 14,1995 Docket Nos. 50-277 50-278 Ucense Nos. DPR-44 DPR-56 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555
Subject:
Peach Bottom Atomic Power Station, Units 2 and 3 Submittal of Revised Relief Request No. GVAR-2, Revision 2 to the Second Ten Year Interval of the inservice Testing (IST) Program
References:
1.
Letter from G. A. Hunger, Jr. (PECO Energy Company) to U. S. Nuclear Regulatory Commission (USNRC), dated February 15,1995 2.
Letter from J. W. Shea (USNRC) to G. A. Hunger, Jr.
(PECO Energy Company), dated March 13,1995 3.
Letter from G. A. Hunger, Jr. (PECO Energy Company) to USNRC, dated April 17,1995 4.
Letter from J. F. Stolz (USNRC) to G. A. Hunger, Jr.
(PECO Energy Company), dated July 7,1995
Dear Sir:
The Inservice Testing Program (IST) for Peach Bottom Atomic Power Station (PBAPS),
Units 2 and 3 currently includes Revision 1 to Relief Request GVRR-2. In Reference 1, PECO Energy submitted a proposed Revision 2 to this relief request. A USNRC request for additional information and subsequent PECO Energy response were transmitted by References 2 and 3, respectively. In Reference 4, the USNRC denied Revision 2 to GVRR-2. In Section 2.3 of the Safety Evaluation attached to Reference 4, the NRC provided three general concerns associated with the denial. The purpose of this letter is resubmit a modified version of Revision 2 to GVRR-2, and to address the USNRC's three concerns. Each concern is paraphrased below, followed by PECO Energy's response.
9507200120 950714 DR ADOCK 0500 7
g I lI
July 14,1995 Pcge 2 USNRC concem No.1:
Ucensee has not identified any generic concern that illustrates a greater burden to perform testing other than during refueling outages.
PECO Energy Response:
Revision 1 to GVRR-2 allows excess flow check valves (EFCVs) to be exercised overy refueling outage instead of every quarter, as required by the ASME Code. This relief request was granted by the USNRC on the basis that testing the EFCVs at power introduced the potential for a plant transient or a personnel safety hazard. Because of enhancements to the on-line maintenance practices, PECO Energy is now able to perform EFCV testing, on certain valves, at power during associated system outages, without increasing the risk of a plant transient or personnel safety hazard. Additionally, PECO Energy has reviewed the EFCV test history data and has concluded that a surveillance test interval of 24 months is justified on its own safety merits, regardless of the hardships associated with any shorter surveillance test intervals. Details regarding the review of test data are provided in the attached modified version of GVRR-2.
The ability to perform EFCV testing at power during system outages coupled with the results of the review of test data prompted PECO Energy to pursue the attached relief request. This practice may not be widely used by other Boiling Water Reactors; however, we do not consider this a reason for not pursuing the initiative. As stated in the attached version of GVRR-2, the current Revision 1 of GVRR-2, which requires EFCV testing to be performed during refueling outages, results in an increase in outage duration of approximately 2 days which corresponds to approximately $900,000 per outage.
USNRC concern No. 2:
The proposed alternative does not ensure that testing will always occur on a refueling outage frequency.
PECO Energy response:
During the planning of the refueling outage scope, inclusion of the EFCV testing would be considered for each system. However, if it can be determined that the EFCV testing can be performed during an upcoming system outage within the 24 month frequency utilizing appropriate plant and safety controls, the valve testing would be moved from the refueling outage scope to the system outage scope. System outages are scheduled three to four years in advance which ensures that the scope of work to be performed during the system outage is coordinated with the refueling outage scope. This coordination of work ensures that all necessary surveillance and testing are performed within the required time limits. An additional work scope check is performed 10 weeks and 5 weeks prior to the system outage. This methodical approach to planning and scheduling ensures that all valves are included in a system or refueling outage, and the scenario of a missed valve test is highly uniikely. If this unlikely scenario were to occur, PECO Energy would pursue a
July 14,1995 Pig 3 3 1
one-time change to GVRR-2 to defer the testing to the next available opportualty. PECO Energy shares the USNRC's objective of minimizing the number of licensing actions.
USNRC concern No. 3:
Ucensee has not provided a complete list of the valves that would be included in the Relief Request.
PECO Energy response:
The IST Program contains the list of valves for which this relief applies.
However, the revised Relief Request has been revised to include the specific valves. The justification for the 24 month surveillance interval applies to all the valves listed in the Relief Request. Similar to other Technical Specification and IST requirements which are on a once per operating cycle frequency, PECO Energy will ensure that the EFCV test is performed for each valve every 24 months. This testing will occur during a refueling outage or at power operation during a system outage.
If you have any questions, please contact us.
Very truly yours, 21.d.Op,%.
G. A. Hunger, Jr.
Director - Licensing Enclosure ec:
T. T. Martin, Administrator, Region I, USNRC W. L Schmidt, USNRC Senior Resident inspector, PBAPS
RELIEF REQUEST NO. GVRR-2, REVISION 2 l
Valves:
Excess Flow Check Valves Reactor & Recirculation System:
XFC-2(3)-02-007A(B) XFC-2(3)-02-008A(B) XFC-2(3)-02-011 XFC-2(3)-02-015A(B) XFC-2(3)-02-017A(B)
XFC-2(3)42419A(B) XFC-2(3)-02-021 A(B,C,D)
XFC-2(3)-02-023A(B,C,D) XFC-2(3)-02425 XFC-2(3)-02-027 XFC-2(3)42-305A(B) XFC-2(3)-02-033 XFC-2(3)-02 062A(B,C,D) XFC-2(3)-02-064A(B,C,D)
XFC-2(3)-02-037A(B) XFC-2(3)-02-073A(B,C,D,E,F,G,H)
XFC-2(3)-02-031 B(C,D,E,G.H,J,K,M,N,P,R,T,U,V,W)
RWCU System: XFC-2(3)-12-066A(B) XFC-2(3)-12-80457L(H)*
RCIC System: XFC-2(3)-13 055A(B)
Core Spray System:
XFC-2(3)-14-031A(B)
HPCI System: XFC-2(3)-23-037A(B)
Category:
A, C Testing Requirement (s):
Exercise quarterly Basis for Relief:
Excess flow check valves (EFCVs) are installed on instrument lines penetrating containment to minimize leakage In the event of an insi U,T sid line failure outside the containment in accordance with Regulatory Guide 1.11. The EFCV is a spring i
loaded ball check valve. Since the system is normally in a static condition, the valve ball is held open by the spring. Any sudden increase in flow through the valve (i.e. line break) will result in a j
differential pressure across the valve which will overcome the spring and close the valve. Functional testing of valve closure is accomplished by venting the instrument side of the valve while the process side is under pressure and verifying the absence of
{
leakage through the vent.
I The testing described above would require the removal of the associated instrument or instruments from service on a quarterly basis. Removal of any of these instruments from service outside of a scheduled refueling outage or a controlled system outage may cause a spurious signal which could result In a plant trip, an inadvertent initiation of a safety system, loss of decay heat removal and/or the defeating of safety interlocks.
Unit 3 valves will be installed during 3R10
e o
Testing of EFCVs can be performed during a scheduled system outage when appropriate plant administrative procedures and controls are utilized to ensure plant safety. System outages are performed in order to enhance system performance and maximize system availability. They are scheduled on a less than quarterly frequency (typically once an operating cycle). Taking system outages quarterly solely for the purpose of EFCV testing would result in reduced system avaRability and increased risk to the plant. NUREG 1482, Section 3.1.2 recommends minimizing equipment out of service time. The additional assurance of operational readiness afforded through surveillance testing must outweigh the impact on plant safety incurred when removing equipment from servico.
In Section 4.1.4 of NUREG 1482, the USNRC approves the deferral of backflow testing of check valves to refueling outages when the testing requires the installation of test equipment. The intent of this request is the same in that EFCV testing requires a plant evolution which should be avoided unless appropriate plant administrative controls are in place.
As discussed in NUREG 1482, the staff recommends that the basis for relief address whether: (1) tha proposed attemative gives an acceptable level of quality and safety, (2) compliance would result in a hardship without a compensating increase in the level of safety, or (3) complying with Code requirements is Impractical. Two of these criteria, and the basis for meeting the criteria are provided below:
The orooosed afternative alves an acceotable level of oualltv and safety.
A review of NPRDS industry failure data for the Dragon excess flow check valves, which is the manufacturer of the valves used at PBAPS, reveals only 7 failures. The 7 faRures break down as follows; 2 were a result of an IST surveillance which failed to met the acceptance criteria for leakage,1 was an Indication (limit switch) problem, and 4 were leakage caused by a bad gasket. Both IST failures occurred at Peach Bottom. A thorough review of Peach Bottom excess flow check valve test history has shown that the 2 NPRDS failures above are the only 2 out of 888 valve tests since 1980. One of the faBures,1.1 gpm measured leakage, was due to dirt on the seating surface of the valve; the other,4.0 gpm measured leakage, was due to a defective seating surface in the manual bypass portion of the EFCV. These are 2 independent f tHures which have not been repeated. Only 2 failures out of 808 valve tests indicates that the valves are highly reliable. Further, this review of surveillance test history shows evidence of no time based faDure mechanisms or chronic failures associated with the excess flow check valves. Although the testing of the EFCVs was performed on the previous refueling cycle frequency (approximately 18 months), testing of the EFCVs has been performed since 1993 on a 24 month refueling outage frequency. Therefore,
- c. 3 l
4 performing the testing of the EFCVs on a 24 month frequency is l
~
appropriate.
i Comoliance would result in a hardshio without a compensating j
increase in safety.
Personnel safety would decrease if EFCV testing would be
' performed on a quarterly basis during plant operation without an appropriate system outage During power operation, the i
process side of the EFCVs is normally high pressure i
(>500 psig) and/or high temperature (>2000F) and highly contaminated reactor coolant. Testing EFCVs during system f
outage windows with the appropriate administrative procedures and controls applied will ensure personnel safety. Additionally, testing at a frequency greater than once per operating cycle would also result in increased radiation dosage and reduced i
system availabNity without any compensating increase in safety.
Improvements in work planning and scheduling have resulted in i
a significant reduction in outage duration at Peach Bottom. As i
a result of these improvements, EFCV testing has become an j
outage critical path activity. Due to the large number of EFCVs and the plant conditions required to perform the testing (reactor l
pressure > 500 poig), testing all the valves during refueling j
results in an outage duration increase of approximately 2 days.
Based on current replacement power costs, this equates to an expenditure of $900,000 por year for the life of the plant.
In order to reduce this level of burden, extensive programmatic and procedural controis are used during system outages to ensure that the impact on plant safety is understood prior to l
removing equipment from service This process is consistent with industry practice and NRC guidance, and has been I
recognized as an effective method of controlling the impact of j
plant activities on safety. During a refueling outage, the l
constraints on resources are at a premium, and the elimination l
of work which can be safely performed independent of those constraints is both practical and prudent.
j In summary, considering the extremely low faHure rate, personnel and plant safety concerns, and the high monetary I
cost of testing during refueling outages EFCV testing at a frequency greater than once per operating cycle and exclusively during refueling outages is impracticable and results in a i
hardship without a compensating increase in the level of safety.
Alternate Testing:
Functional testing wNl be performed once per operating cycle during a refueling outage or system outages when appropriate plant administrative controls are in place.
ll 4
i,71%9 m
e
_~a
-+--- -
, _-.-