IR 05000528/1995023
| ML17312A543 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 02/15/1996 |
| From: | Myers C NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML17312A540 | List: |
| References | |
| 50-528-95-23, 50-529-95-23, 50-530-95-23, GL-89-10, NUDOCS 9602220169 | |
| Download: ML17312A543 (32) | |
Text
ENCLOSURE U.S.
NUCLEAR REGULATORY COMMISSION
REGION IV
Inspection Report:
50-528/95-23 50-529/95-23 50-530/95-23 Licenses:
NPF-41 NPF-51 NPF-74 Licensee:
Arizona Public Service Company P.O.
Box 53999 Phoenix.
Arizona s
,Facility Name:
Palo Verde Nuclear Generating Station, Units 1, 2, and 3 Inspection At:
Palo Verde Site, Wintersburg, Arizona Inspection Conducted:
December 11-15, 1995, and January 8-12, 1996 Inspectors:
C. Myers, Reactor Inspector, Engineering Branch Division of Reactor Satety M. Runyan, Reactor Inspector, Engineering Branch, Division of Reactor Safety Approved:
+(G'
F~
an en urg
,
ie
,
ngineering rane Division of Reactor Safety
~//5/9C Ins ection Summar Areas Ins ected Units
2 and
Special, announced inspection of the implementation of the licensee's program to meet commitments to Generic Letter 89-10.
"Safety-Related Motor-Operated Valve Testing and Surveillance,"
and followup of previously identified items.
Results Units
2 and
~E
~
With some exceptions.
the inspectors determined that the licensee had satisfactorily demonstrated the design basis capability of each of the safety-related.
motor-operated valves in its Generic Letter 89-10 program (Section 1.1).
9602220169 9602i6 PDR ADQCK 05000528 G
l I
I
l
-2-
~
The licensee had established conservative design requirements for motor-operated valve performance which often 'exceeded the safety-related design basis requirements (Section 1. 1).
~
Additional inf'ormation was requested to complete the NRC review of the following areas of the licensee's program:
Description of grouping methodology (Section 1. 1).
Justification of group valve factors (Section 1. 1),
Description of trending program (Section 1.7),
and Description and status of open findings from self-assessment (Section 1.8).'oot cause evaluations of motor-operated valve failures were consistently detailed and thorough (Section 1.7).
~
The licensee's program for trending motor-operated valve diagnostic test data was not well developed or implemented (Section 1.7).
~
The licensee's self-assessment of their motor-operated valve program was comprehensive with substantive findings (Section 1.8).
Summar of Ins ection Findin s:
~
Inspection Followup Item 528; 529;
~
Inspection Followup Item 528:
529;
~
Inspection Followup Item 528; 529;
~
Inspection Followup Item 528; 529;
~
Inspection Followup Item 528:
529:
Attachment:
530/9125-11 was closed (Section 2. 1).
530/9332-01 was closed (Section 2.2).
530/9411-01 was closed (Section 2.3).
530/9411-02 was closed (Section 2.4).
530/9411-03 was closed (Section 2.5).
~
Attachment
- Persons Contacted and Exit Meeting
l l
f
-3-DETAILS
GENERIC LETTER 89-10,
"SAFETY-RELATED MOTOR-OPERATED VALVE TESTING AND SURVEILLANCE" (2515/109)
On June 28, 1989. the NRC issued Generic Letter 89-10 requesting licensees to establish a program to ensure that switch settings for safety-related motor-operated valves were selected.
set, and maintained properly.
Subsequently, seven supplements to Generic Letter 89-10 have been issued.
NRC inspections of licensee actions implementing commitments to Generic Letter 89-10 and its supplements have been conducted based on guidance provided in Temporary Instruction 2515/109,
"Inspection Requirements for Generic Letter 89-10, Safety-Related Hotor-Operated Valve Testing and Surveillance."
Temporary Instruction 2515/109 is divided into Part 1,
"Program Review," and Part 2, "Verification of'rogram Implementation."
The Tempo ary Instruction 2515/109.
Part 1, program review at Palo Verde Nuclear Generating Station was documented in NRC Inspection Report 91-25.
Subsequent inspections, conducted under Part 2 of Temporary Instruction 2515/109.
were documented in NRC Inspection Reports 93-32 and 94-11.
The purpose of this inspection was to verify completion of the licensee's commitments to Generic Letter 89-10.
The NRC has established a closure process for inspections under Generic Letter 89-10.
This was documented in a memorandum dated July 12, 1994, entitled.
"Guidance on Closure of Staff Review of Generic Letter 89-10 Programs,"
and was addressed to the NRC Regional Division of Reactor Safety Directors from Hr.
B. Sheron, Office of Nuclear Reactor Regulation.
The guidance contained in this document was used during this inspection.
The process of "closing" a licensee's Generic Letter 89-10 program can be best defined as verification that the licensee has satisfactorily applied the principles contained in Generic Letter 89-10 (or suitable alternate methods) to demonstrate the design basis capability of each motor-operated valve in the program.
The closure process does not preclude additional inspections in this area.
Additionally, there remains an expectation that the assumptions and methodologies used to develop the Generic Letter 89-10 program will be maintained for the life of the plant, a concept commonly described as a "living program."
The closure process does not convey final NRC acceptance of a licensee's approach to -the areas of periodic verification or pressure locking and thermal binding.
These areas, to be reviewed under new generic letters, were reviewed on only an interim basis for closure under Generic Letter 89-10.
Several issues necessary for NRC closure of the Generic Letter,89-10 motor-operated valve program at the Palo Verde Nuclear Generating Station were not resolved during this inspection.
Additional information was requested to complete the review process.
The inspection documented by this report was the closeout inspection at Palo Verde Nuclear Generating Station under Part 2 of Temporary Instruction 2515/109.
The inspection focused on the licensee's basis for
1f
completion of testing to demonstrate the design basis capability of the motor-operated valves in the program.
Programmatic issues were addressed during this inspection as followup from previous inspection open items and in the context of issues that developed during the course of the inspection.
As an overall assessment, the inspectors concluded that the licensee's motor-operated valves program had adequately demonstrated the operability of motor-operated valves subject to Generic Letter 89-10.
The program was thorough and was implemented in a manner consistent with the licensee's commitments to the generic letter
.
1. 1 Summar Status of Generic Letter 89-10 Motor-0 crated Valves
. The licensee had identified 336 motor -operated valves within the scope of its Generic Letter 89-10 program.
This total population for all three units included 144 gate valves, 135 globe valves.
and 57 butterfly valves.
A total of 256 valves (76 percent)
were tested under at least partial differential pressure conditions.
Some of these tests were conducted under hydrostatic conditions.
The inspectors reviewed various licensee documents and held discussions with the responsible engineers to determine if the licensee had adequately met its commitment to demonstrate the existing desi'gn basis capability of each Generic Letter 89-10 valve.
The inspectors based their determination on whether the licensee had qualified each valve using well-justified, test-based design information or in-situ design basis testing.
The inspectors found that the licensee had not adequately justified its assumptions for three critical design parameters used in its analysis of design basis capability.
These parameters were:
(1) valve factor, which correlates differential pressure to valve stem th ust requirement; (2) stem friction coefficient, which correlates the actuator output torque to valve stem thrust and, (3) rate-of-loading (alternately described as load sensitive behavior)
~ which correlates stem thrust under dynamic conditions to stem thrust under static conditions.
The licensee had grouped the total population of its Generic Letter 89-10 valves into groups based on identical valve size, pressure class, and manufacturer.
Host groups included several valves, which had been tested under differential pressure conditions.
For these tested groups, the licensee used the worst-case valve factor, stem friction coefficient, and rate-of-loading to assess the design basis capability of each motor-operated valve in the group.
This technique resulted in a conservative determination of the margin for the valves in the group. particularly those tested showing more favorable performance characteristic I
e-5-However.
for untested groups (i.e. valve groups that did not include any valves which had been tested under differential pressure conditions) the licensee had not adequately documented the justification for the assumptions for valve factor, stem friction coefficients and rate-of-loading.
For untested valve groups, the licensee generically assumed gate valves to have 0.55 or 0.6 valve factors and globe valves to have 1. 1 valve factors, stem friction coefficients to range from 0. 12 to 0.2, and rate-of-loading to be 10 percent.
(An exception to this generic assumption is discussed later in this section.)
The licensee had not documented a basis for selecting these particular values for the untested groups of valves.
The inspectors reviewed the range of values obtained during licensee testing and found that these three assumptions did not conservatively bound all the test results.
Some tested valve factors exceeded 0.6.
some stem friction coefficients were less than 0. 12, and many rate-of-loading values were greater than 10 percent.
The inspectors concluded that, because of the use of assumptions that were not bounded by the licensee's test data and lacking other documented engineering basis for its assumptions, the licensee had not adequately justified the design basis capability for untested valve groups under the "two-stage approach" discussed in Generic Letter 89-10.
The generic letter stated that for untestable valves an original best estimates assumption (Stage 1) should be followed up by some comparison to similar valves by grouping with other valves in the plant, prototype testing, tests in other nuclear plants, or other industry testing (which would constitute Stage 2 qualification).
The inspectors considered that the licensee had completed only the first stage of this process.
The inspectors noted that
~ for five marginal motor-operated valves in untested groups.
the licensee departed from its generic assumption for valve factor and instead used less conservative test values reported by the Electric Power Research Institute (EPRI) for similar valves.
This testing had been performed oy EPRI in development of their performance prediction methodology for motor-operated valves.
Howevers the licensee did not intend to use the EPRI performance prediction methodology to establish the design basis capability for untested valves.
The inspectors were concerned that the licensee had used EPRI test results to demonstrate the design basis capability of its five marginal valves without further justification.
The inspectors considered that valve factors reported by EPRI were appropriate for licensee use as the best
,available data "for justifying the design basis capability of individual valves
'under Stage 1 of a two-stage approach, as recommended by Generic Letter 89-10.
However, without further justification, the inspectors did not consider the licensee's use of the limited EPRI data to be appropr iate for program closure as a Stage-2.qualification.
The inspectors discussed with the licensee that the EPRI test data was not intended to be used as bounding values for untested valve populations because the EPRI testing generally included tests of only one valve representing a given manufacturer, size.
and pressure clas I l)
I'
-6-For Valves 1SGAUV0134, 1SGAUV0138, 2JSGAUV0134, and 2JSGAUV0138, the licensee used a valve factor of 0.453, measured during EPRI testing of a 6-inch, 900-pound pressure class, Anchor-Darling gate valve.
The four Palo Verde valves were identical in specification to the EPRI test valve.
For Valve 2SIAUV0672, the licensee used a valve factor of 0.56, reported from EPRI testing of a 12-inch.
300-pound pressure class, Borg-Warner flex-wedge gate valve.
Valve 2SIAUV0672 was an 8-inch valve, as compared to the 12-inch EPRI valve.
The licensee had not justified the use of the EPRI data for the different valve size.
The inspectors concluded that the licensee had not sufficiently justified use of the EPRI valve factors for the valves in question.
The inspectors concluded that an immediate operability concern did not exist with any of the Generic Letter 89-10 valves because the assumed parameters appeared reasonable and, combined with other conservatisms in the licensee's program, provided reasonable assur ance that these valves could perform their design basis safety functions.
However. for closure under Generic Letter 89-10. additional description of the licensee's grouping methodology and justification of the valve factors used in each group is necessary.
The inspectors identified this issue as a response item to be completed by the licensee within 60 days of this report.
Dur,ing the inspection.
the licensee provided draft information to address the inspectors'oncerns regarding validation of performance parameters.
Some of this information provided additional assurance, to support the inspectors'onclusion that no immediate operability concerns existed.
The inspectors found that the licensee had not included a specific margin for valve factor degradation to account for changes in valve factor performance over ti'me or service conditions.
Industry testing had indicated that valve factors may increase in service.
The inspectors requested additional information to assess the effective performance margin in both the closing and open di rection for each motor -operated valve in the Generic Letter 89-10 rogram.
The inspectors reviewed this information to determine if the icensee could demonstrate an excess margin avai lable for each valve.
This review was limited to the gate valves in the program, and, with only a few exceptions, all of the gate valves were shown to have at least a
10 percent margin.
However. in many cases.
the licensee credited valve inertia to achieve a
10 percent margin.
The inspectors did not consider the licensee's use of inertia to be adequately justified.
The licensee had not performed tests necessary to demonstrate the validity of the assumptions.
The inspectors concluded that the licensee needed to remain aware of ongoing industry testing to determine the rate at which valve factors increase, if at all, with time and valve strokes to ensure the continued capability of its valves to perform their intended safety function ~
l
-7-The licensee had established conservative guidelines for calculating the design basis conditions under which the Generic Letter 89-10 valves were assumed to operate.
In its review of design basis requirements.
the licensee had considered certain aspects that were not generally considered by other
.licensees.
These items included increased pump differential pressures resulting from fluctuations in diesel generator frequency, and valve operation under all normal.
abnormal, and emergency conditions.
Also. the licensee assigned uncertainty to visual gage readings and performed extensive error analysis for pump curves.
The licensee credited this general conservatism to mitigate the inspectors'oncerns regarding justification of assumptions of valve performance parameters.
The inspectors agreed that a conservative design basis was a strength in the licensee's program.
However, the inspectors were unable to assess the adequacy of individual design assumptions for untested valves based solely on this justification.
1.2 M~iiti Generic Letter 89-10 had originally recommended that the scope of licensee programs include valves which could be inadvertently mispositioned from the control room.
Supplement 7 to Generic Letter 89-10, dated January 24, 1996, removed this recommendation for pressurized water reactors.
At the time of the inspection, the inspectors found that the licensee continued to include valves which could be mispositioned within the scope of'ts motor-operated valves program.
According to the licensee, the scope of its Generic Letter 89-10 program also included valves that were considered to be capable of repositioning following a fire in the control room.
The licensee included these valves because the capability to recover from potential mispositioning induced by hot shorting was credited in its fire safety hazards analysis under
CFR 50. Appendix R.
1.3 Pressure Lockin and Thermal Bindin At the time of the inspection, the licensee was in the process of developing a
submittal in response to Generic Letter 95-07,
"Pressure Locking and Thermal Binding of Safety-Related Power-Operated Gate Valves."
The inspectors reviewed Engineering Study 13-HS-A96,
"Gate Valve Pressure Locking and Thermal Binding Evaluations,"
Revision 0, which documented the licensee's engineering justifications supporting the current operability of susceptible valves.
The licensee had identified 10 susceptible valves per unit'(30 valves total) where predicted pressure locking forces were in excess of the nominal pullout thrust capability of the valves.
The licensee had used a double-disk-area methodology to estimate the thrust requirements under pressure locked conditions.
These 10 valves were SI-653/654, SI-671/672, SI-686/696, SI-685/694, and SI-688/693.
Six of the valves exhibited only minor thrust capability deficits.
Four of the valves had pressure locking thrust requirements well in. excess of their nominal thrust capability.
The
I
inspectors reviewed the licensee's operability determination for,the following four valves:
Pressure Locking Thrust Available Thrust SI-653/654 SI-686/696 143,571 lbf 152.090 lbf 69,044 lbf 97,310 lbf The licensee intended to install relief valves on the bonnets of these valves to eliminate the potential for pressure locking at the next available train outage.
For the interim period, the licensee had justified the operability using a defense-in-depth approach that included taking credit for an unventable air volume at the top of the bonnet and inservice leak testing data.
The inspectors found that due to the orientation of the valves.
an unventable air volume was credible but had not been positively verified by testing.
However, the inspectors did not consider the inservice leak test data to be a valid basis for mitigating bonnet pressurization.
The leak test was performed using 50 psig air, and the licensee did not have any correlation to convert measured air leakage to a corresponding water leakage rate.
Additionally, the inservice leak test system involved multiple test boundary valves, such that the source of the leakage could not positively be ascribed to the valve seat.
The inspectors requested additional justification for the basis for operability.
Coincident with the inspectors'equest for additional information, the licensee was in the process of incorporating pressure locking test data attained at the Braidwood nuclear facility.
Palo Verde participated in the testing at this site.
In response to the inspectors'oncern, the licensee recalculated the pressure locking thrust requi rements, using the Braidwood test data, for the 30 susceptible valves and found that in each case, the thrust requirement decreased.
The licensee concluded that this provided evidence that its original methodology was conservative.
The inspectors did not review the new pressure locking calculations for technical acceptability.
However, the inspectors noted that there remained two valves per unit (SI-653/654) that had available thrusts that were less than the new estimated pressure locking forces.
The inspectors requested the licensee to recalculate the thrust requi rement using only the unventable air pocket as a means to mitigate the pressure rise in the bonnet.
The licensee prepared a calculation entitled, "Alternative Pressure Locking Evaluation of 1,2-SI653/654 for a 110'F to 220'F Temperature Increase."
The calculation showed that the unventable air volume was capable of limiting the bonnet pressure increase within the capability of the actuator to open the valve.
The inspectors found that the licensee's evaluation was adequate to resolve their concerns.
Additional NRC review of the licensee's pressure locking and thermal binding program is expected to be conducted pursuant to Generic Letter 95-0 l
-9-The inspectors reviewed the licensee's treatment of measurement error applied to the analysis of diagnostic data from testing of motor-operated valves.
The inspectors found the licensee's analysis to be consistent with its response to Supplement 5 of Generic Letter 89-10.
The licensee had incorporated appropriate vendor accuracy considerations in its evaluation of valve stem thrust and torque measured during dynamic and static testing.
1,5
~Gnou in The licensee categorized all Generic Letter 89-10 valves by groups containing valves of identical size.
valve manufacturer, valve model, actuator size, pitch/lead of the stem, process medium, differential pressure, flow, and temperature.
The worst-case performance characteristics were applied to each valve in any individual group.
The inspectors determined that the licensee's general grouping methodology was consistent with the guidelines. in Supplement 6 to Generic Letter 89-10.
However, in cases where differential pressure testing was not conducted within a given group, the licensee stated that test results of'alves in other groups would be considered in an effort to justify the assumed valve factor.
This effort was not complete at the time of this inspection, but was included as part of a detailed data reconciliation task assigned with a due date of June 30.
1996.
The inspectors considered that thi s effort would result in a revision of the originally-defined groups.
The inspectors requested additional information identifying and justifying the resulting grouping criteria to be used as the basis for demonstrating the design basis capability of untested valves.
This request for additional information was included as part of a response item to be addressed by the licensee within 60 days of the date of this report.
1.6 Periodic Verification and Post-Maintenance Testin The inspectors reviewed licensee Procedure 39AC-92202,
"Valve Services Maintenance,"
Revision 0.
The inspectors found that the licensee had established both static and dynamic test requirements following motor-operated valve maintenance and modification.
The inspectors found that the licensee conducts static diagnostic testing for each motor-operated valve in their program as periodic verification of design basis capability.
The licensee performs this static testing of each valve on a frequency of two refueling outages.
Each refueling outage, the licensee performs preventive maintenance including stem lubrication and grease inspection.
Every fourth refueling outage, the licensee refurbishes each actuator and performs a static diagnostic test of the valve.
The inspectors found that the licensee had implemented preventive maintenance tasks to perform the periodic static testin I
-10-The licensee also planned to conduct selected periodic differential pressure
.testing as part of its periodic verification activities.
The inspectors found that the licensee had not identified the periodic dynamic testing to be performed.
The inspectors found that the licensee was considering the risk of accelerated wear or damage in weighing the benefits of ongoing dynamic testing of motor-operated valves.
The NRC staff is preparing a generic letter on periodic verification of motor-operated valve design basis capability.
It is anticipated that the licensee's program for periodic verification will be reviewed in more detai l following issuance of the new generic letter,.
1.7 Failure Anal sis and Trendin The inspectors reviewed selected condition reports/disposition requests for the last 2 years to evaluate the licensee's analysis of motor-operated valve problems.
The inspectors found that the licensee's root-cause evaluations of motor-operated valve fai lures and test anomalies were consistently thorough and technically detailed.
The licensee had identifi'ed significant problems with torque switch contact bar flipping, torque switch roll pin failures, and excessive bearing friction in butterfly valves.
The inspectors considered that the direct involvement of engineering in the identification and resolution of motor -operated valve problems was instrumental in achieving the strong performance in this area.
The inspectors reviewed licensee Procedure 39AC-9ZZ01, "Valve Services Engineering,"
Revision 0.
The inspectors observed'a demonstration of the computer-based data collection performed by the licensee and discussed trending activities with licensee cognizant personnel.
The inspectors reviewed the licensee's program for the trending of diagnostic test data and found it to be poorly developed and weakly implemented.
Detailed procedures had not been developed and general guidelines were not being followed.
The inspectors found that early licensee attempts at accumulating and organizing the test data 'for meaningful evaluation had been abandoned.
No clearly defined program had been established to define meaningful performance parameters to maintain design basis capability through the periodic evaluation of test data.
The only parameter being trended at the time of the inspection was the measured thrust at torque-switch-trip.
The inspectors found that the licensee treated this measur ed parameter as'
process variable with an expected normal distribution.
The licensee did not historically trend individual valve data, but rather applied a three sigma statistical confidence interval as the
'xpected variability of individual valve data before an outlier would be identified and trended.
Based on the licensee's data.
the inspectors found that measured thrust would have to change at least 50 percent before it would be considered by the licensee to be an outlier.
Compared to.the measurement
, repeatabi lity of less than 10 percent, the inspectors considered the licensee's methodology for trending to be relatively insensitive.
Furthermore, the inspectors found that the licensee had not integrated an
(
I
allowance for this expected variability into the margin assessment for individual valves.
The inspectors noted that the trending program had also been identified as a
area of concern in a recent self-assessment of the licensee's Generic Letter 89-10 program.
The licensee had committed to implement recommended action (h) of Generic Letter 89-10 related to trending.
Action (h) recommended that each motor-operated valve failure and the corrective action taken, including repai rs, alterations, analyses, tests, and survei llances, should be analyzed or justified and documented.
The documentation should include the results and history of each as-found deteriorated condition, malfunction, test, inspection.
analysis.
repai r, and alteration.
The recommended action further suggested that the data be periodically examined as part of a monitoring and feedback effort to establish trends of motor-operated valve operability.
The inspectors considered that meaningful performance based trending of motor-operated valve test data continues to develop as an industry issue and a long term aspect of Generic Letter 89-10.
However, the inspectors concluded the licensee's trending program for motor-operated valves did not appear adequately developed or implemented for closure under Generic Letter 89-10.
The inspectors identified this issue as a response item to be completed by the licensee within 60 days of this report.
1.8 ualit Assurance Involvement/Self Assessment The inspectors reviewed the following Nuclear Assurance assessments of the Generic Letter 89-10 program:
Independent Safety and Quaiity Engineering Department Assessment 93-02,
"Generic Letter 89-10 Motor-Operated Valve Programmatic Assessment,"
(9/21/93);
~
Independent Safety and Quality Engineering Department Assessment 94-A-M0-005,
"PVNGS Motor-Operated Valve Program Operability Review," (8/9/94):
and,
~
Nuclear Assurance Division Evaluation Report 95-1015,
"Closure of Generic Letter 89-10," (12/9/95).
The inspectors found each of the above self assessments to be comprehensive and technically detailed.
Several significant findings were identified in these reports including. in the 1994 and 1995 reports, a significant concern in the motor-operated valve trending program.
The inspectors considered these self assessments to be an area of strong licensee performance in closure of the licensee's Generic Letter 89-10 program.
Several of the findings from the recent self assessment had not been dispositioned by the licensee at the time of the inspection.
The inspectors
]
t
-12-considered that licensee resolution of the concerns identified in the self assessment was instrumental for closure under Generic Letter 89-10.
The inspectors identified this issue as a response item to be completed by the licensee within 60 days of this report.
FOLLOWUP (92903)
2. 1 Closed Ins ection Followu Item 528 529 530/9125-11:
Evaluation of DC Motor Stroke Time Back round Licensee testing had shown that several dc motor-operated valves had stroke times that were close to the acceptable limits specified in the Final Safety Analysis Report.
The tests had been conducted under static conditions at rated voltage.
The stroke time of dc motor-operated valves is dependent on dc motor speed which varies with load and voltage.
The inspectors had been concerned that under worst-case design basis conditions of maximum differential pressure and degraded voltages some of these valves may be unable to meet the required stroke time.
~Fol 1 owu The inspectors reviewed licensee Calculation 13-JC-ZZ-221,
"AC and DC Powered Motor-operated Valve Stroke Times."
The licensee had developed this calculation to estimate motor -operated valve stroke times under dynamic conditions and under design voltage conditions.
In its calculation, the licensee assumed that the running-load motor speed (i.e., the motor speed resulting under running load conditions only during static testing) would decrease by 50 percent over 20 percent of the valve stroke.
The licensee used the motor performance curves published by Limitorque to support this assumption.
Correction factors were then applied to the static stroke times of each motor-operated valve.
The licensee estimated that 19 motor-operated valves may exceed the analyzed stroke time under dynamic conditions at degraded voltage.
In each of these cases, the licensee evaluated in more detail the basis for the stroke limits and determined that the estimated stroke time was acceptable.
The inspectors reviewed Calculation 13-JC-ZZ-221 and concluded that the licensee had satisfactorily resolved this issue.
2.2 Closed Ins ection Followu Item 528 529 530/9332-01:
Pressure Lockin and Thermal Bindin Back round During a previous inspection, the licensee had not completed evaluating the susceptibility of its safety-related motor-operated valves to the phenomenon
l f
I I
l
-13-of pressure locking and thermal binding.
This item was opened to ensure an NRC review of this effort upon completion.
~Foll owu During this inspection, the inspectors reviewed the status of the licensee's evaluation of pressure locking and thermal binding.
The inspectors found that the licensee had completed its evaluation and was taking corrective actions for susceptible valves.
This review is documented in Section 1.3 of this report.
Further NRC review of this subject is anticipated in the in review of the licensee's response to Generic Letter 95-07,
"Pressure Locking and Thermal Binding of Safety-Related Power-Operated Gate Valves."
2.3 Closed Ins ection Followu Item 528 529 530/9411-01:
DC Motor-0 crated Valves Desi n Basis Ca abi lit Under De raded Volta e
~Back round In its evaluation of the design basis capability for certain dc motor-operated valves (1.2,3-AF-32/33, 1,2,3-AF-36/37, 1,2,3-AF-54, and 1,2,3-SG-134/138),
the licensee had used Limitorque's published pullout efficiency of 0.40 as a
conser vative estimate of actuator efficiency when closing under degraded voltage.
Limitorque has stated that their published dynamic efficiencies were appropriate with motors operating at nominal rated speed of 1700 rpm.
In se'lecting an appropriate value for dynamic actuator efficiency, the licensee had not justified the use of pullout efficiency under slow motor speed.
The inspectors had observed that the motor speed for certain dc motor-operated valves was expected to slow to approximately 200 rpm under design basis degraded voltage conditions.
For these marginally-sized motor-operated valves'he inspectors were concerned that the use of pullout efficiency would not be conservative and had not been adequately justified.
~Fol 1 owu In response to the inspectors'oncern, the licensee conducted a test of a dc motor actuator setup in the shop.
The motor was identical to the installed dc motors. of concern, although the actuator was of a different size.
The test results, which were documented in Attachment A to Condition Report/Disposition Request CRDR-9-4-0449.
indicated that dc actuator efficiency at 100 rpm was 41 percent.
This value was greater than the assumed pullout efficiency of 40 percent.
The inspectors considered that the licensee's testing had demonstrated that the use of published pullout efficiency was a realistic estimate of the actuator efficiency resulting under slow speed operation.
The licensee planned to modify the marginal valves by increasing the actuator gear ratios to increase the design margin.
For the interim period. the licensee relied on the use ot stall torque from generic motor curves rather than the rated motor starting torque (40 foot-pounds) in order to demonstrate
I
-14-the operability of these valves.
To account for uncertainties in the use of the generic curves, the licensee had applied a
10 percent correction factor to lower the indicated stall torque capability from approximately 62 foot-pounds to 56 foot-pounds.
During its test of the shop dc valve, the licensee measured a stall capability of 58 foot-pounds.
The licensee considered that this test result was conservative for use in the analysis of the installed dc motors because the test motor was not new and had been repeatedly stalled.
The inspectors considered that the licensee had adequately justified its use of the generic motor curves.
The licensee stated that the evaluation under Condition Report/Disposition Request 9-4-0449 was an interim justification until the next scheduled outage for each unit. at which time the subject valves were planned to be modified.
Following modification, the dc motor-operated valves were to be evaluated by the conventional Limitorque method of using the nominal rated starting torque in combination with the pullout efficiency.
The inspectors reviewed the licensee's test results and discussed this issue with the responsible engineers.
The inspectors concluded that the licensee had established an acceptable basis for current operability of the dc motor-operated valves.
2.4 Closed Ins ection Followu Item 528 529 530/9411-02:
Lack of Mar in for S rin ack Relaxation Backcnround The licensee had not provided a margin in its evaluation process to account for long-term relaxation of actuator springpacks.
This degradation causes a
reduction in the actuator torque delivered at torque switch trip and could result in a fai lure of the valve to fully close under design basis conditions.
Limitorque Maintenance Update 93-02 addressed this concern.
Normally-closed motor-operated valves were identified as most susceptible because the springpacks for these valves are compressed for long periods of time.
~Fol 1 owu The inspectors reviewed the licensee's motor-operated valve setpoint control document.
The inspectors found that the licensee had revised its methodology to incorporate a margin for springpack relaxation.
consistent with the information presented in Limitorque Maintenance Update 93-02.
The inspectors found that the new margins had been applied for applicable Generic Letter 89-10 motor-operated valve,
-15-2.5 Closed Ins ection Followu Item 528 5Z9 530/9411-03:
Weaknesses in Test Data Evaluation Procedure Back round Two concerns had been identified related to the licensee's process for evaluating motor-operated valve dynamic test data.
The first concern involved the adequacy of the preliminary operability review performed prior to returning-a tested valve to service.
The licensee had not included consideration of uncertainties for torque switch repeatabi lity and diagnostic system error in its evaluation.
Furthermore, the evaluation had used the torque switch trip thrust from the previous static stroke rather than the measured value from the dynamic test to satisfy the setpoint acceptance criteria.
The second concern involved the detailed review of the test data, which the licensee requi red to be completed within 7 days following return to service.
The licensee's review had been considered to be potentially untimely because any deviations from expected performance were transferred to a condition report/disposition request for further evaluation without a specified time for completion.
Because of this, an inoperable valve may not have been identified for an extended period of time.
~Fo1 1 owu The licensee revised Procedure 39DP-92201 (Revision 4) to include in the preliminary operability review a consideration of diagnostic system error, dynamic and differential pressure effects, degradation margin, and extrapolation of required thrust up to the design basis conditions.
The inspectors reviewed this procedure change and concluded that both concerns were resolved in that the new assessment performed a complete operability evaluation of the valve prior to returning it to service.
Any inoperable conditions would therefore be identified before the valve was relied upon for safety-related servic,
PERSONS CONTACTED 1. 1 Licensee Personnel ATTACHMENT S.
Bauer.
Licensing Section Leader.
Nuclear Regulatory Affairs S.
Burns, Department Leader, Nuclear Design and Plant Engineering M. Hooshmand, Section Leader, Valve Services Engineering A. Krainik, Department Leader, Nuclear Regulatory Affairs J. Kriner, Engineer, Valve Services Engineering D. Maudlin, Director. Maintenance J. Minnicks. Department Leader, Valve Services M. Renfroe, Section Leader, Valve Design Engineering R. Rogalski, Engineer, Nuclear Regulatory Affairs C.
Russo, Department Leader, Nuclear Assurance
- Maintenance M. Salazar, Section Leader, Valve Services S.
Scow, Engineer, Nuclear Assurance
- Maintenance E. Smith Jr., Consulting Engineer, Valve Services Engineering 1.2 Other Or ani zations R. Henry, Site Representative, Salt River Project 1.3 NRC Personnel K. Johnston, Senior Resident Inspector The personnel listed above attended the exit meeting.
In addition to the personnel listed above, the inspectors contacted other personnel during this inspection period.
EXIT MEETING An exit meeting was conducted on January 12, 1996.
During= this meeting, the inspectors reviewed the scope and findings of the report.
The licensee acknowledged the inspection findings documented in this report.
The licensee did not identify as proprietary any information provided to, or reviewed by, the inspector l
~
J
~
,
l