ML20086F489
| ML20086F489 | |
| Person / Time | |
|---|---|
| Site: | Sequoyah  |
| Issue date: | 11/26/1991 |
| From: | Joshua Wilson TENNESSEE VALLEY AUTHORITY |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GL-89-10, NUDOCS 9112030247 | |
| Download: ML20086F489 (10) | |
Text
1
[G a
4 d
< Iam,
a.
a s me u, aw f. e m.,- n,, &
Jath L VMson
- ,., ma +,.masvn Novenber 26, 1991 U.S. Nuclear Ragulatory Commission ATTN: Document Control Desk Washington, D.C.
20555-4 Gentlemen:
la the-Matter of
)
Docket Nos. 50-327 Tennessee Valley Authority
)
50-328 SEQt0YAH NUCLEAR PLANT (SQN) - TVA RESPONSE TO NRC INSPECTION REPORT 18 REGARDING GENERIC LETTER (GL) 89-10 I
References:
1.
TVA letter to NRC dated October 28, 1991, " Supplemental Information for Compliance With Generic Letter (GL) 89-10" 2.
NRC letter to TVA dated September 26, 1991, "NRC Inspection Report Nos. 50-327/91-18 and 50-328/91-18" i-3.
TVA letter to NRC dated December 21,1989, " Browns Ferry Nuclear Plant (BFN), Sequoyah Nuclear Plant (SQN), and Watts Bar Nuclear Plant (WBN) - Response to Generic Letter (GL) 39 Safety-Related Motor-Operated l'alve (MOV) Testing and Surveillance" During the week of August 5-9, 1991, NRC Regicn II conducted an inspection of SQN's ongoing program for compliance with GL 89-10.
Reference 2 contains the r(sults of this inspection.
The inspection team determined that SQN had developed a basic program that adequately
. addresses most_of the GL recommendations.
Concerns and otrengtas identified by the NRC inspection team were separated into those for which a written resporet'is requested and those for which no written response is requested (see page 2 of Reference 2).
The purpose of this letter is 4:
to address two NRC concerns for which a written response was requested.
In addition, actions are being taken to address those concerns for which I
no response was requested.
3
-911dO30247 911126 hJ g
- G!
. ADOCK G5000327 FDR PDR
_.._-_~ ~...-.. -. - - -
.. ~...
c
-v i
1 P, - ~
2 l
U.S. Nuclear Regulatory Commission November 26, 1991 4
i -j.
The first concern involved SQN's schedule _for implementing a program-P within the schedule specified by GL 89-10 (i.e., five years or three i
' refueling outages). The _second concern involved a provision contained in j_
TVA's_ program that would allow differential pres.ure-_(dp) testing to be; j
omitted by grouping similar valves and testing only a' portion of the-l-
-group.-- The Reference 2 inspection report recognized that TVA intended to revise.its previous commitment (sco Reference 3) and pursue: grouping within TVA's GL 39-10 program._ By Reference 1 NRC was informed that *VA intended-to pursue grouping as technically justifiable and tha, grouping-
[
would be'a variance from TVA's previous _ commitment.
L Enclosure i provides TVA's response to the first concern regarding schedule. contains additional details and justification for-i TVA's ' philosophy _regarding grouping and selection of motor-operated valves for dp testing.--Enclosure 3 summarizes the revised TVA commitment for.SQN with respect to the proposed schedule for implementation of 4 '
{
-Please direct questions concerning this issue to D. V.'Goodin at (615) 843-7734.
i Sincerely, 4
F l
fg i'
/
JgL. Wilson i
l i
Enclosures cc'(Enclosures):
I Mr. -D. E. LaBarge Project _ Manager-8
- U.S. Nuclear Regulatory Commission
- 0ne White Flint, North
~
11555-Rockville Pike--
Rockville, Maryland 20852 1
NRC Resident Inspector Sequoyah, Nuclear: Plant J2600 Igou Ferry Road
' Soddy Daisy,' Tennessee. 37379 Mr. B.-A.
Wilson, Project Chief
'U.S. Nuclear Regulatory Commission l
_ Region-Il
}
-101 Marietta Street, NW ' Suite 2900 l~
-Atlanta,_ Georgia 10323 i
4
ENCLOSURE 1 JUSTIFICATION FOR GENERIC LETTER (GL) 89-10 TVA recognizes that the present schedule for the implementation of GL 89-10 at Sequoyah Nuclear Plant (SQN) (the Cycle 6 refueling outages for both units) may not be achievable.
Because of the number of unresolved industry issues relsted to the implementation of GL 89-10. TVA's approach has been to address the most demanding applications first (priority given to gate and globe valves that operate against high differential pressures) in order to gain data and work toward the resolution of these unresolved issues while progressing toward the implementation of the GL.
Additionally, it has been anticipated that the technical justificatica would be developed through ongoing industry initiatives to afford the application of grouping to some subset of SQN's GL 89-10 motor-aperated valves (MOVs).
IVA recognizes that the extent of the grouping application is not yet fully defined; the inability to apply grouping would impact the cur ently defined scnedule for implementatien.
SQN began testing MOVs for GL 89-10.iuring the Unit 1 Cycle 5 refueling outage i
currently in progress.
Design reviews and thrust calculations have been completed for 30 MOVs, and differential pressure (dp) testing has been perf ormed on 12 of the 17 MOVs scheduled for testin6 this outage.
The five remair.ing MOVs to be dp tested must be tested during start-up (i.e., Mode 3) when steam is available.
Datc gathered and experience gained during dp testing during this outage further lead to the conclusion that the original schedule was overly optimistic.
As a result of the above-described considerations, TVA is extending SQN's schedule for implementing its GL 39-10 program to the end of the Cycle 7 refueling outages for each unit (November 1994 for Unit 1 and May l995 for Unit 2).
c 1
-_~._
ENCLOSURE 2 JUSTIFICATION AND PHILOSOPHY REGARDING GROUPING ll.0' INTRODUCTION The primary objective of Generic Letter (GL) 89-10 is to ensure that
. motor-operated valves (MOVs) are functional at design basis conditions.
Because of uncertainties in the analytical prediction of MOV behavior and
-in the application of test results from one MOV to another, testing of ati MOVs at design basis conditions has been recommended, where practicable.
TVA believes that advancements in the_MOV thrust calculation methodology and a better understanding of design characteristics that affect MOV performance will enable a :ogram to be developed that places greater emphasis onttbe engineering analysis with a reduced need for the testing 1-of MOVs at maximum expected differential pressore (dp) conditions.
TVA's approach to GL 89--10 implementation will ensure that the following objectives are met:-
a.
MOVs are capable of performing their desi n basis function.
6 b.
The setpe of dp testing performed to valicate MOV functional capability is technically warranted,
-Testing is performed periodically "o ensure MOV functional capability.
c.
An engineering basis is documented to address MOVs that are not
^
d.
tested at full dp and flow conditions.
In order to accomplish the above objectives, a grouping methodolcgy is i
being developed that will serve two purposes.
The primary purpose is to allow the evaluation of MOVs that cannot be tested at design basis l -
conditions._ Tnis-process is consistent-with the GL 89-10_two-phase approach andLallows the program implementation to progress, with the results of_ industry testing _ factored in as thiu information becomes available.
The second purpose for grouplag is to reduce the amount of dp testing when it can be shown that test results from identical and/or similar h
-valves demonstrate the ability-of other MOVs within that grouping to l
function at design basis conditions. This would help to reduce wear and L
teac on equiprent, minimize program implementation costs, reduce l
personnel radiation _ exposure, and reduce 1 system outage time.
This philoscphy complies with the goals of GL 89-10 recommendations to achieve
[
higher component reliability and plant safety, p
The grouping process is still under development by TVA and will be refined as additional industry information and test results become available. The following provides TVA's preliminary apprcach to the
. grouping process.
Sequoyah Nuclear Plant (SQN) will maintain a description of the grouping process with supporting' technical L
justification onsite in accordance with GL 39-10.
N
1 2.0 JUSTIFICATION FOR UTILIZING GROUPING TO ICENTIFY MOVs TO BE DP TESTED 2.1 ENHANCED METHODOLOGY FOR MOV THR'JST PREDICTION TVA has taken an aggressive approach to ensure that calculations used to precict MOV thrust requirements are conservative and will bound test results.
Dirferences in test results between M0'!s, even those that are identical in design and have the same functional requirements, are expt ted because of issues such as the following:
a.
Diagnostic equipment accuracy b.
Test boundary conditions c.
Differences in design features between valves (e.g. internal clearances) d.
Condition of valve and actuator internals (aging process) e.
Maintenance or preventive maintenance that has been performed Rate of loading g.
Material strength properties TVA's thrust calculation methodology accounts for these and other issues that affect thrust requirements, and the amount of testing required to validste these calculacions would be less than if typical vendor thrust equations and valve factors were used.
Preliminary results from dp testing of 12 medium-to high-pressure gate valves during the current Unit 1 Cycle 5 outage have shown that l
switch settings implemented based upon TVA calculations resulted in valve functional apabt.'.ty at dp test conditions.
TVA's calculation resulta have been compared with the results of other equations such as those developed by the Eh ctric Power Research Institute (EPRI) and the idato National Engineering Laboratorys and it has been shown that TVA's calculation methodclogy produces results that bound these equations.
It should clso be noted that TVA's MOV calculation methodology was identified as a strength during the NRC audit at SQN.
1 2.2 EPRI AND/OR INDUSTRY SUPPORT OF GROUPING CONCEPT T*JA's approach to grouping parallels the objective of tt.c EPRI MOV Performance Prediction Program.
EPRI expects to provid-guidelines, validated by test results, that could help utilities predict MOV r
perfornance so that dp testing at design basis condftlons is not I
required for all MOVs.
TVA is also interf acing witt. other utilities in developing a validated grouping philosophy.
i I
.. _ _. _. ~.. _ _. _.. _ _ _ _
?
- i 2.3 globe AND QUARTER-TURN MOV CONSIDERATIONS j
Globe and quarter-turn MOVs are crpected to be appropriate for the limited dp test program because of internal-design characteristics that produce better consistency when-predicting thrust and/or torque requirements.
In addition, industry information to date does not indicate that these types _of valves are subject to many of the concerns that apply to gate valven.- This philosophy will be evaluated against the results of further indestry tecting-and t.odified if necessary.
2.4 TVA MOV SWITCH SETTING METl!0DOLOGY As'an added measure to ensure that MOVs are functional at design basis conditions, the control circuitry for most_ safety-related MOVs at SQN is set up to provide. actuator stall torque capacity (if called upon) for opening the valve and for 95 to 98 percent of the closing stroke. This methodology for setting control swicches ensures that.the valve performs its safety function,.regardless of j
the torque switch setti.ng._
3.0
SUMMARY
OF THE GROUPING PROCESS 3.1 - DEFINITION OF IDENTICAL AND SIMILAR MOV ' GROUPS-One of the first tasks to be accomplished in developing a successful-approach to the grouping ofLMOVs is to aort MOVs into identical and similar_ groups that will facilitate a meanirgful comparison of test data and other issues. relative' to the GL 39-10 program.
\\
j An identical group'af MOVs is considered to be one sharing the' l-following characteristics:
a.
Same~ manufacturer
._b.
Ssme type, design, and materials _of construction
-c.
Same size d.
Same or similar application ar.d pressure conditions e.
Same pressure-class f.
Same valve identification number A similar group of MOVs is normally one consisting of valves of the same manufacturer, but having different-attributes such as size, pressure class, and model number, r:
I L
._._.___._____m
^
t If it is necessary to include valves-of different manufacturers in a similar group because of the inability to dp test valves of a parti ular manufacturer, a study will be performed to document the basis for this grouping. This study may include a finito element analysis. The grouping of valves.from a different manufacturer would be used as a " Phase 1" approach, as defined _by GL 89-10, until additional industry information is available.
The following valve characteristics must be comparable in order to define valve ~ similarity:
Similar disc desi n (e.g., solid wedge, flex wedge, double disc, a.
6 and parallel disc) b.
Similar guide design (considering length, machined er cast, hard surfaced, clearances, and'other attributes)
.c.
Design of stem _to T-head connection (proper orientation to flow and proper clearances) d.
Seats (materials of construction, hard surfaced, and angle) 3.2 USE OF AN ENGINEERED APPROACH FOR GROUPING
'l Grouping would be supported by a detailed. engineering analysis that would demonstrate the functional capability-of MOVs. This analysis would address.the following as a minimum a.
Methodology and justification for establishing identical and similar MOV groupings b.
Methodology for performing validation testing, including selection-criteria for MOVs to be tested and'deta ecliection requireuents c.
Methodoiogy for extrapolating dp tests to design basis conditions, where required d.
Initial assessment of the ability of each MOV to perform its
-design basis function (s)
=e.
Resolution et any defielencies identified by the initial assessment f.
Final essessment of the ability of each MOV to perform _lts intended safety function, based upon test results, industry
~,
feedback, and lessons learned from program implementation
.TVA is currently negotiating with Siemens Nuclear Power Services to further develop in engineering approach for grouping MOVs.
i.
-. - - - - + - -...
4 1
4
=$.
3.3 - VALIDATION OF THE GROUPING PROCESS i
TVA's grouping philosophy will utilize in-plant-testing and other available _ test data for validation. The following general acceptance-criteria will be applied for MOVs that are dp tested!
- a. _Each-tested MOV shall perform its safety function to fully open and/or close as-defined by the design evaluation.
Full closure shall be indicated by the ability to seal off flow (with~some l
1eakage allowed), unlest a more stringent closure criterion is
-specified by the design evaluation.-
b.
MOVs tested at less than design basis conditions shall be evaluated for functional capability at_the design basis condition.
If sample testing from a valve group produces unacceptable test results, a thorough engineering review will be performed to determine the root cause.
This review will address factors such as.
validity of_ test data'; valve design characteristics (e.g.,
materials and clearances); previous maintenance history; and other factora that could !nfluance valve performance. A determination will also be nede as to..mther the condition causing the -
unacceptable results is an isolated occurrence. Additional valves a
from the group will be tested as det"rmined appropriate f om the
^
engineering review.
Where additic:ri testing is not practical, and-appropriate alternative corrective action will be identified and implemented.
The engineering evaluation and test-results_will also be reviewed
[
to. ensure special attention in_ areas of marginal performance.
The margin is defined as the difference between the calculated actuator capability and the thrust. required to perform the safety function.
Ear measured by the dp test. -Special attention will-be'given to marginal MOVs-'to ensu e they will centinue to perform their safety' function (e.g., no indication of mechanical problems based on the -
signature analysis). The corrective action will be identified where appropriate f.: marginal performance. The criteria for marginal performanc have not been developed, but this will be established and docunented when ~ applicable. Where margin is a concern, special emphasis.will be given to dp testing and identical and'similar MOVs, where practical.
u Dif f erences.in thrust requirements obtained during dp testing of identical valves will be evaluated and the differences documented.
4.0 DOCUMENTATION FOR GROUPING AND/0R TEST SELECTION PROCESS L
l-Jur*.ification will be documented f or MOVs that are either not dp tested or tested at less than design basis conditions based upon established grouping _ methodology.
This information will be available onsite for i
review by NRC.
5.0 SHORT-TERM ACTION BY SQN I
Because of concerns related to gate valves, especially in high-and I
medium-pressure systems, and in order to obtain as much short-term test information as possible, Sequoyah's schedule for testing requires that i
these valves be tested at or near design basis conditions where practical. This will further support the grouping practice as test results are collected and evaluated.
These test results may be used to justify a method (basis) for the limited testing of remaining MOVs.
Conversely, the results of testing may indicate that the testing of all MOVs should continue
- .o be cerformed where practical.
Sequoyah is connu1*.ted to perfotming as much testing as is necessary to adequately demonstrate the functional capability of MOVs at design basis conditions.
P.NCLOSURE 3 Commitment The SQN Generic Letter 09-10 program implementation will be completed by the end of the Cycle 7 refueling outages f or each uni t.
Y e~
4 9
l