IR 05000445/1993010
| ML20035H809 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 05/03/1993 |
| From: | Runyan M, Westerman T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20035H802 | List: |
| References | |
| 50-445-93-10, 50-446-93-10, GL-89-10, NUDOCS 9305070163 | |
| Download: ML20035H809 (18) | |
Text
-
-.
.-
_.
-.-
.-
.
,_ -
t
.;:
.;.
.
.
!
+
APPENDIX B
-;
.
U.S. NUCLEAR REGULATORY COMMISSION l
REGION IV
Inspection Report: 50-445/93-10 i
50-446/93-10
,
Operating Licenses: NPF-87 NPF-88
_
Licensee:
TU Electric
.
Skyway Tower 400 North Olive Street, L.B. 81 Dallas, Texas 75201-l Facility Name: Comanche Peak Steam Electric Station, Units 1 and 2 iq Inspection At: Glen Rose, Texas Inspection Conducted: March 15-19 and April 7, 1993 Team Leader (Acting):
M
- L w 4-7-7-93 M. Runyan, Acting Team Leader Date Engineering Section, Division of Reactor Safety Team Members:
R. Vickrey, Reactor Inspector, Engineering Section Division of Reactor Safety.
C. Myers, Reactor Inspector, Region V T. Scarbrough, Mechanical. Engineering Branch, Office of Nuclear Reactor Regulation (NRR)
' Accompanied By:
'
M. Holbrook,. Consultant, EG&G Idaho Approved:
A
f3
-
T. Westerman, Chief, Engineerini Section Date-Division of Reactor Safety
--i -
9305070163 930503-PDR ADOCK 05000445 G-PDR j
.
_
_
-
..., ___.
.
_._
.
,
.
-...
.
_
. _.. _
_.. _
.
--
,
.
.
l SUMMARY
-
A team of NRC staff members and a consultant conducted an inspection of motor-
,
operated valve (MOV) testing and surveillance pursuant to Generic Letter (GL) 89-10 at the Comanche Peak Steam Electric Station. The inspection was conducted from March 15-19, 1993. An initial exit meeting was conducted
+
on March 19, 1993. A supplemental exit meeting was conducted on April 7,1993.
'
The NRC team utilized the guidance provided in Temporary Instruction (TI) 2515/109, " Inspection Requirements for Generic Letter 89-10, Safety-Related Motor-0perated Valve Testing and Surveillance." As delineated i
in Part 2 of TI 2515/109, this inspection was the initial review of the
'
implementation of the licensee's MOV program.
.
The team reviewed in detail the diagnostic test results of 11 MOVs including design calculations, test packages, and signature traces. The team also reviewed followup issues from the previous NRC inspection of the MOV program (TI 2515/109, Part 1).
Within the scope of the inspection, the team did not identify any issues that challenged the operability status of the MOVs selected for review or any other MOVs in the licensee's GL 89-10 program. The team concluded that the M0V program including its implementation was strong, innovative, comprehensive, and meeting the intent of GL 89-10.
The team observed several strengths. The licensee's diagnostic capabilities were very extensive and included motor dynamometer testing, actuator torque stand bench testing, and strain-gaged valve stems for measurement of thrust and torque. The licensee was refurbishing all actuators prior to baseline testing. The methodology for analyzing MOV test data was very rigorous and appeared to address all relevant points necessary to demonstrate valve operability. The licensee's HOV staff was very knowledgeable and had extensive experience in the testing and maintenance of MOVs.
The team identified several discrepancies, concerns, and weaknesses that were categorized as one potential violation, two unresolved items, and six inspection followup items.
The failure to evaluate updated values for actuator torque switch repeatability published by the actuator vendor, Limitorque, was identified as a violation.
An incorrect calculation of acceptable torque values in preparation for a MOV diagnostic test and the lack of an independent verification of the calculation was identified as an unresolved item. Based on additional information provided subsequent to the exit meeting, this unresolved item was identified as a violation with no response required.
The performance of an open-stroke dynamic diagnostic test starting from a valve position that may not have been fully closed was identified as an
- iii -
.
J
.
.
.
.
.
- unresolved item.
Based on additional information provided subsequent to the exit meeting, this unresolved item was changed to an inspection followup item.
,
A question related to the settings for torque switch bypass switches in the
-
opening direction was identified as an inspection followup item.
The use of the Kalsi test program to increase the qualified thrust ratings of Limitorque actuators beyond the range endorsed by Limitorque was identified as an inspection followup item.
The concern that the licensee's review of the pressure locking phenomenon did not consider certain normally open MOVs was identified as an inspection followup item.
The fact that the licensee's method of measuring torque may not detect abnonnal losses of torque internal to the actuator was identified as an inspection followup item.
'
A weakness in the test procedure for dynamic diagnostic tests, which lacked specific reference to evaluation of the signature trace for abnormal conditions, was identified as an inspection followup item. An MOV, which had shown abnormal behavior on a signature trace in 1991, had not been inspected internally to identify the cause of its abnormal behavior.
>
The team concluded that the licensee will need to establish additional
.
justification for-its occasional use of motor running efficiency for MOV open
strokes, for the use of a linear extrapolation technique for predicting
'
required thrusts and torque, and for the use of increased thrust ratings established by Westinghouse for SB-00 actuators.
The team observed that dynamic test results for two identical valves, one in Unit I and the other in Unit 2, were very dissimilar, demonstrating limitations in qualifying MOVs by comparison to similar valves.
The team discussed several issues related to butterfly valves and was satisfied that the licensee would address all relevant concerns.
At the exit meeting, the licensee committed to provide the NRC additional information related to the two unresolved items by April 2, 1993.
Overall, the inspection team concluded that the licensee had satisfactorily demonstrated the operability of the MOVs selected for review and had established a strong, comprehensive program that met the intent of and its commitments to Generic Letter 89-10. The weaknesses and discrepancies discovered during the inspection were not considered excessive considering the complexity of the program and the extent of the material that was reviewed.
- iv -
,
.
-
- - - -
- - - - - - -
- -
- - - - - - -
- - - - - - - - - - - - - - - - - - - - -
_
-
.
.
.
.
DETAILS
GENERIC LETTER (GL) 89-10 " SAFETY-RELATED MOTOR-OPERATED VALVE TESTING AND SURVEILLANCE" (2515/109)
On June 28, 1989, the NRC issued GL 89-10, which requested licensees and construction permit holders to establish a program to ensure that switch settings for safety-related motor-operated valves (MOVs) were selected, set, and maintained properly. Subsequently, four supplements to the GL have been issued and one issued for comment. NRC inspections of licensee actions implementing commitments to GL 89-10 and its supplements have been conducted based on guidance provided in Temporary Instruction (TI) 2515/109, " Inspection Requirements for GL 89-10, Safety-Related Motor-0perated Valve Testing and Surveillance." TI 2515/109 is divided into Part 1, " Program Review," and Part 2, " Verification of Program Implementation." The TI 2515/109 Part I program review was conducted at Comanche Peak from September 30 to October 4, 1991, and is documented in NRC Inspection Report 50-445/91-51; 50-446/91-51.
The inspection documented by this Report was the initial inspection at Comanche Peak under Part 2 of TI 2515/109, and thus was focused on verification of program implementation. Nevertheless, programmatic issues were addressed during this inspection in response to followup of findings in
-
the Part 1 inspection and in the context of issues that developed in the course of the inspection.
The team reviewed the diagnostic test results of 11 motor-operated valves.
Each of the selected valves had been tested under differential pressure conditions, some at greater than and some at less than the maximum expected differential pressure (MEDP). The review consisted of examining the test data packages, the diagnostic signature traces (on a computer), and, in some cases, the design calculations of MEDP and thrust requirements. The following valves
.
were reviewed:
1-8716A RHR Pump 1 Crosstie 1-8804B RHR Pump 2 to CCP Suction 1-8821B SI Pump 2 Crosstie 1-8835 SI to Cold leg 1 and 4 Injection Isolation 1-HV-24928 Motor-drivan AFW Pump Discharge to Steam Generator Loop 2 Isolation 2-8000A PORY Block Valve 2-8105 Charging Pump B to RCS Isolation 2-8511A CCP 1 Alternate Miniflow Isolation 2-8716A RHR Pump 1 Crosstie-1-
.
.
.
.
- 2-8809A RHR to Cold Leg 1&2 Injection Isolation 2-HV-4776 Containment Spray Heat Exchanger 1 Outlet In the scope of this review, no issues were identified that immediately challenged the operability status of the 11 MOVs. However, some concerns were identified that will require review, evaluation, and, in some cases, corrective action on the part of the licensee to ensure that the selected valves as well as other valves in the GL 89-10 program are capable of performing their design safety function. The concerns and observations that were identified during the inspection are discussed in the subsections that follow. Notwithstanding the noted concerns and weaknesses, the team concluded that the licensee's GL 89-10 program was strong, innovative, comprehensive, and meeting the intent of the GL.
1.1 MOV Sizino and Switch Settina The inspectors reviewed the licensee's methodology for determination of thrust and torque requirements for the valves listed above.
The licensee's gate
valve thrust equation provided by Westinghouse incorporated valve disk friction coefficients ranging from 0.45 to 0.55.
MOV load sensitive behavior (also known as " rate of loading" or "ROL") was addressed by adding a margin i
based on the valve's stem factor using a 0.15 stem friction coefficient divided by the stem factor using a 0.10 stem friction coefficient. Thrust requirements were adjusted to account for diagnostic equipment inaccuracy and torque switch repeatability. Once dynamic test results were available for a given valve, the thrust requirements were revised to reflect the results of the dynamic test, including consideration of the observed load sensitive behavior and extrapolation to design-basis conditions (if required).
The licensee utilized the close limit switch to control seating of several SB actuators by adjusting the close limit switch until a minimum thrust level was exceeded. When a differential pressure test was performed, the torque output generated during the dynamic test was compared to the torque generated during
the static test and the ratio of the two torques' values was computed. This ratio was used to reduce the torque limits utilized during subsequent static tests to ensure that actual torque limits would not be exceeded under design-basis conditions.
The team noted that the S8-00 actuator thrust limits for the closing direction had been extended to 16,000 pounds based on testing conducted for Westinghouse. Peak closing thrust values were allowed up to 17,000 pounds.
The maximum allowed opening thrust was 15,500 pounds. The staff determined that these limits were presently acceptable. However, for the long term, the licensee will be expected to address Section 3.3.3 of NRC Inspection Report 99900404/92-01 (dated August 14,1992) to resolve concerns regarding the Westinghouse actuator testing.
I t-2-
.
-
.
l 1.2 Desian-basis Capability
'
The inspectors reviewed Procedure No. MSE-P0-8004, " Safety-Related Rising Stem MOV Testing," Revision 4, static test packages, and dynamic test packages for
'
the selected valves.
Where testing could not be practically conducted at design-basis conditions, the licensee utilized a straight line extrapolation of the thrust to overcome differential pressure using the ratio of design-basis differential pressure to the test differs tial pressure. However, the licensee had not performed multi-point testirg or developed other methods to justify the long-term acceptability of ts extrapolation method. Therefore, the team considered these tests to be the first stage of a two-stage approach where the valves have been set up initially using the best available data, as discussed in GL 89-10. The licensee would be expected to justify their method of
'
extrapolation by the schedule commitment date for the completion of their GL 89-10 program.
,
During review of the static testing, it was noted that a test conducted on October 28, 1992, for Valve 1-8821B, indicated that closing thrust at control
'
switch trip (CST) minus the running thrust load did not achieve the thrust i
estimated by the licensee to be required to close the valve under design-basis
"
conditions (based on previous dynamic test results). Further, the torque
!
generated at CST was greater than the actuator's torque limit (based on output capacity at 80 percent degraded voltage conditions). The licensee's
'
resolution of these issues was documented in ONE Form 92-1053, dated l
October 28, 1992. The original thrust requirement was based on dynamic
.
testing that was conducted at 118 percent of design-basis conditions.
!
However, the licensee originally took a conservative position and did not
extrapolate these results downward to arrive at the test-based thrust
requirements. After removing this conservatism, the valve passed the revised
!
thrust acceptance criteria. The original torque limit was based on calculated t
output capacity under 80 percent degraded voltage conditions. The licensee
!
revised this limit upward based on dynamometer test results conducted at 80
percent degraded voltage conditions. This provided enough margin to allow the
,
valve to pass the torque acceptance criteria. However, this left little i
margin to account for degradation. The licensee intends to change the gear
ratio for this valve (and its companion valves) in order to improve setup
margin. The team agreed with the licensee's evaluation and the plan to modify
Valve 1-88218.
During review of Procedure MSE-P0-8004 and its attachments, the inspectors l
noted that while static test sections contained acceptance criteria that required inspection of the data traces to ensure that they were free of i
anomalies (Criterion 2.1.8), the dynamic test section did not have this
!
criterion. This was of concern because dynamic testing is the most likely place where anomalies would be present.
For example, a review of the first
differential pressure test of Valve 1-88218 indicated that an anomalous behavior occurred during the seating portion of the trace. This was indicated by a sharp upward spike that indicated possible difficulty in transitioning-3-
.
t
. _
.
-.
.
.
.
.
.
I
-
from the valve guides to the valve seating surface. The licensee noted the
,
anomaly during the testing and ge-ated a technical evaluation
,
(TE No. 92-2404, dated November 18,1992), which indicated that the valve
,
internals were potentially degraded. However, no test acceptance criteria
were violated and a ONE Form was not generated. The team considered this to
'
be a weakness in the licensee's acceptance criteria and identified this as an
inspection followup item (445/9310-01; 446/9310-01). The licensee stated that it intends to inspect the valve internals of Valve 1-8821B at the next
opportunity, and to revise the dynamic test acceptance criteria to include a qualitative evaluation of the traces.
The dynamic test package for Valve 2-8000A (Unit 2 pilot-operated relief valve block valve), identified an over-torque condition where the closing torque at CST exceeded the maximum allowed torque (based on actuator capacity at
,
80 percent degraded voltage conditions).
Further, the testing indicated that
'
approximately 23 percent more torque was required to close the valve under dynamic conditions versus closing the valve under static conditions. Thus, this " rate of loading" value would be used to reduce the torque ratings imposed during static testing. TU Evaluation TUE No. 92-5986, dated August 5, 1992, documented the over-torque condition and the resolution of this issue which included a change in overall gear ratios from 34.1:1 to 52.2:1.
Subsequently, it was decided to make the same modification to the remaining pilot-operated relief valve block valves in Unit 2 and Unit 1.
Evaluation TUE No. 92-5986 determined that the new gear set would result in an actuator output of 248.05 foot-pounds at 80 percent degraded voltage conditions.
Procedure MSE-P0-8004 was used to conduct a static test performed October 17, 1992, after the overall gear ratio was changed. As required by Step 6.8.8, licensee personnel obtained the maximum allowable torque value, 248 ft-lb i
(TQRUN80) identified in Evaluation TUE No. 92-5986, in order to fill in the torque limits in the verification data package. However, an additional
!
requirement of Step 6.8.8 was not performed, in that, the torque limit was not adjusted downward by the amount of " rate of loading" identified in the dynamic test package prior to entering this value in the data sheet.
Further, l
licensee personnel failed to adjust in a similar manner, the total torque limit used to account for peaks due to inertia. This was considered contrary
-
to Criterion V, 10 CFR 50, Appendix B, " Procedures," which combined with a consideration of Criterion III below, became the subject of a violation with
,
no response required (445/9310-02; 446/9310-02). These actions resulted in
,
non-conservative torque limits being used in the test acceptance criteria. A review of the test results using the corrected torque values indicated that no
,
actual limits were exceeded.
,
Upon further review of the test data for Valve 2-8000A, it was noted that the
'
same individual who failed to include " rate of loading" in the torque limit
,
calculation later signed for review of the data package. This review was conducted 4 days after the test was conducted and included review of the diagnostic data acquisition section, which was recorded by a different
individual. However, this apparent lack of independent review of the torque limits, which formed the basis of the acceptance criteria, was considered to
-
[
_4_
.
.
b be a violation (445/9310-02; 446/9310-02) of Criterion III,10 CFR 50, Appendix B, " Design Control," in that the checking process was not performed by an individual other than the one who entered the original design values.
In a telephone conversation on April 5,1993, the licensee informed the NRC that a procedure change notice (PCN) would be issued by April 13, 1993, to the MOV test procedure, MSE-PO-8004, to stipulate that the individual who performs the prerequisite calculations for a given test package cannot sign as the reviewing authority.
Based on information provided by the licensee on March 30, 1993, the licensee indicated that a generic review of test packages
.
'
had revealed one other instance where input torque numbers had been incorrectly delineated (Valve 2-LCV-Il2D).
For both valves affected by these errors, the licensee determined that test results were in specification with the corrected torque limits (and, hence, no operability issue existed).
In i
consideration of the prompt and comprehensive corrective actions, the licensee will not be required to respond to this violation.
The team reviewed the licensee's demonstration of design-basis capability for MOV l-8835. MOV 1-8835 is a normally open 4-inch Westinghouse gate valve, which the licensee had determined to have an active safety function to close under 245 pounds per square inch differential (psid) and open against 1001 psid. During the test, the M0V had been closed and opened twice against differential pressure (dp) conditions which were in excess of design-basis conditions.
Prior to the demonstration of opening capability, the MOV had been stroked closed under 1750 psid (710 percent of the expected design-basis dp). Not unexpectedly, the valve had failed to completely close during that closing stroke since the torque switch setting was established based on a
lower design-basis dp. However, the diagnostic data from the subsequent opening stroke was recorded as the open-thrust verification signature. The inspector found that the initial condition of the valve prior to the opening demonstration did not appear to be representative of worst case opening conditions since the disc had not wedged into the seat during the previous closing stroke. The thrust verification signature did not appear consistent with the test procedure. Test procedure MSE-P0-8804, Revision 2, indicated that two closure attempts had been performed prior to the opening stroke. The inspector noted that the recorded thrust signature indicated that only one closure attempt had been made prior to opening. The team observed that the
- >
thrust verification signature showed that hard seat contact occurred during
,
closing. This indicated that full closure had been achieved. However, no i
'
close light indication was indicated in the signature.
In addition, flow-i noise was noted following closure indicating to the licensee test personnel that full closure had not been achieved. The inspector noted that a failure to fully close had been identified in the test log. The data recorded for the open torque switch bypass was not taken from the same opening stroke verification signature as the other data.
It was obtained from a second opening stroke which was performed after the valve had been closed under less
'
severe dp conditions. The team was concerned that the documentation of the testing did not resolve the testing anomalies to assure an adequate
demonstration of design-basis capability.
l
!
-5-
.
.
.
.
.
.
.
.
.
,
In response to the team's concerns, the licensee initiated a ONE Form to
review the adequacy of the testing performed for the verification of design-basis capability.
Subsequent to the exit meeting, the licensee provided additional information on this item.
In their review, the licensee identified two other valves (1-8802A and 2-8802A) for which data used to evaluate the open acceptance criteria were not the most conservative available (i.e., these valves were not fully closed prior to the open stroke used to demonstrate their operability). However, the licensee determined that Valve 1-8835 had been fully closed prior to its official open stroke, even though some flow may
'
still have been present, and that the original evaluation of this valve's capability was correct. The licensee committed to conduct retraining of all
'
Level 2 qualified MOV test personnel to ensure that future test data evaluations are suitably conservative. This item has been identified as an inspection followup item (445/9310-03; 446/9310-03).
.
1.3 Limitoraue Maintenance Update 92-02 During discussions of general topics, the team discovered that the MOV engineers were unaware of changes made by Limitorque to its published values for torque switch repeatability. The licensee's program assigned a torque switch repeatability of +/- 10 percent for all of its actuators. Of specific concern to the team, was Section 4 (Case 4) in Limitorque Maintenance Update 92-02 that assigned a torque switch repeatability of +/- 20 percent for actuators with the torque switch set at the No. I setting and having less than 50 foot-pounds of actuator output torque. The team briefly reviewed the MOV information matrix provided by the licensee along with the master setpoint list (MSL) and identified three MOVs (1-FCV-0610, 1-FCV-0611, AND l-LCV-112B)
that appeared likely to be affected by the new repeatability limits. The team reviewed the most recent test packages for these three MOVs and concluded that sufficient margin existed to accommodate the higher torque switch repeatability range. However, other MOVs may have also been affected by the
,
new limits.
The licensee received Limitorque Maintenance Update 92-02 on or before January 4,1993, and routed the document for inclusion in maintenance department manuals and the lubrication manual. The update was not sent to the engineering department, resulting in the lack of an applicability review for impact on the MOV program. The provisions of Procedure STA-206, " Review of Vendor Documents and Vendor Technical Manuals," Revision 17, were not
,
appropriately implemented. The licensee indicated that the M0V program would not have evaluated the new limitorque information for applicability without
>
the team identifying the problem. The failure to properly follow Procedure STA-206 in response to the receipt of Limitorque Maintenance Update 92-02 is considered a violation of 10 CFR 50, Appendix B, Criterion V (445/9310-04; 446/93-10-04).
In its response to this violation, the licensee is expected to evaluate the Limitorque maintenance update for applicability and to complete a review of the MOV data base to determine that there are proper margins.
-6-
.
-
- -
-
-
.
.
.
,
.
$
1.4 Toroue Switch Bypass in Ooen Direction The MOV program was being revised to establish the minimum bypass of the open l
torque switch as 20 percent of full stoke or the point where flow effects
!
terminate during an open dynamic diagnostic stroke, whichever is greater. The team questioned whether the point at which flow effects cease on a partial differential pressure test (50 percent MEDP, for example) would be the same as at 100 percent MEDP. This was a concern because open torque switches were being set at levels only slightly in excess of running loads to provide maximum protection against backseating loads.
If the open torque switch were set at a point that did not envelop the duration of 100 percent MEDP flow
-
effects in the open direction, valve motion could cease at mid-cycle. The licensee stated that this concern would be evaluated to determine if a change to the acceptable torque switch bypass range is needed. This issue will be tracked as an inspection followup item (445/9310-05; 446/9310-05).
1.5 Actuator Performance Predictability
.
15e team noted that the differential pressure test results for MOVs 1-8716A and 2-8716A, identical valves in each unit (including actuator type, torque switch setting, pitch, and lead), were very dissimilar. The extrapolated thrust needed to close the valve at 100 percent MEDP was 9070 pounds for Valve 2-8716A compared to 14361 pounds for Valve 1-8716A. The magnitude of this disparity suggests a limitation in the use of test results of similar valves to demonstrate capability under the two-stage approach of GL 89-10.
In
,
this case, had Valve 1-8716A been set based on the test results of
-
Valve 2-8716A, it may have been unable to close against the MEDP. The
,
licensee was aware of this phenomenon from a generic perspective and stated i
that it would be appropriately considered when analytical estimates are made of the capability of valves that can be tested only under static conditions.
i 1.6 Reliance on Kalsi Study of limitoroue Actuator Over-Thrust Capability The licensee stated that it was relying on a study conducted by Kalsi Engineering of the structural thrust capability of certain Limitorque actuators.
In light of that Kalsi study, Limitorque, in its Technical
Update 92-01, raised the allowable structural limit of the actuators within the scope of the study to 140 percent of their thrust ratings for 2000 cycles.
Further, Limitorque has informed those licensees with access to the
,
proprietary Kalsi engineering report that the applicable actuators can
!
withstand 162 percent of their structural thrust ratings for 2000 cycles if
'
the provisions of the Kalsi. engineering report are met. The inspectors found that the licensee had identified 12 MOVs to be overthrusted up to 40 percent above their rating and 2 MOVs to be overthrusted up to 67 percent above their rating in each reactor unit.
For the two MOVs in each unit being overthrusted
up to 67 percent (HV-4696 and 4709), the inspectors reviewed ONE form 92-1049
-
in which the licensee evaluated the number of cycles that these MOVs had
'
experienced. The inspectors considered the licensee to have provided adequate justification for the structural thrust capability of these overthrusted MOVs.
.
-7-
.
i
.
.
.
.
.
s The NRC staff position'concerning the use of the Kalsi study to increase actuator thrust ratings will be provided separately by the staff. This issue was identified as an inspection followup item (445/9310-06; 446/9310-06).
1.7 Use of Run Efficiency in Evaluatina Motor Pullout Caoability The licensee typically assumed nominal motor-start torque, an application
,
factor of 0.9, pullout efficiency, and 80 percent degraded voltage in determining the capability of each motor actuator to initially open its valve.
The inspectors found that, for six butterfly MOVs, the licensee had replaced the pullout efficiency with run efficiency in its prediction of motor actuator
'
capability. This results in an increase in predicted motor actuator capability. The inspectors did not consider the licensee to have provided sufficient test data to justify generic use of run efficiency rather than
'
pullout efficiency in predicting motor actuator capability when opening valves. The inspectors reviewed the licensee's evaluation of the six butterfly MOVs: ONE Form 92-1469 for M0V 1-HV-4286, ONE Form 92-1407 for
,
MOV l-HV-4287, Evaluation TUE Form 92-6598 for MOV 2-HV-4525, ONE Form 93-263 for M0V 2-HV-4575, ONE Form 93-63 for M0V 2-HV-4286, and Evaluation TUE Form 92-6349 for MOV 2-HV-4287.
In each case, the inspectors found that the overall conservatism of the licensee's evaluation was sufficient to demonstrate the capability of the motor actuator.
For example, the licensee's evaluations included such factors as nominal motor torque, 80 percent degraded
,
voltage although the minimum voltage was calculated to be higher, and
,
partially open valves. The licensee will need to justify any long-term use of run efficiency in its evaluation of motor actuator capability for opening
valves. This issue is identified as an inspection followup item (445/9310-07;
,
446/9310-07).
1.8 Linear Extrapolation In evaluating its MOV test data, the licensee performed a linear extrapolation of the thrust required to overcome differential pressure from test conditions
<
to design-basis conditions. The licensee had not justified its linear
extrapolation method (such as by performing multiple differential pressure
,
tests of individual MOVs). Therefore, the inspectors considered the capability of the MOVs based on the licensee's extrapolated test data to only satisfy the first stage of the two-stage approach described in GL 89-10 and
its supplements. The licensee will need to complete the justification for its extrapolation method within its GL 89-10 schedule commitment by in-plant testing or industry test information.
1.9 Butterfly MOV Testino
The licensee has been testing butterfly MOVs as part of its GL 89-10 program.
In reviewing the' test results, the inspectors noted that, in Technical-Evaluation TE No.93-581, the licensee reported that MOV 2-HV-5543 had demonstrated a lower torque output at torque switch trip during loaded testing
-
on a torque test stand than that delivered at torque switch trip during static in-situ testing. The licensee demonstrated that no sperability concern-8-
,
.
.
.
existed for this M0V. However, this apparent reduction in torque output at torque switch trip under loaded conditions for a butterfly MOV could have implications for other butterfly MOVs at Comanche Peak and other facilities.
The inspectors also noted that, in Evaluation TUE No.91-3065, the licensee reported that MOV l-HV-4393 had its open and close torque switches wired backward. The licensee determined that the cause of the problem was that Limitorque's labelling of the open and close torque switches was intended for HOVs designed with the handwheel turned clockwise to close the valve (direct-acting MOV), whereas, M0V l-HV-4393 is designed with the handwheel turned counterclockwise to close its valve (reverse-acting MOV). The licensee has revised Section 8.2 of Procedure MSE-P0-8005, " Comanche Peak Steam Electric
!
Station Safety-Related Quarter Turn Motor-0perated Valve Testing Procedure,"
Revision 2 (effective May 20,1992), to include verification of correct torque switch wiring.
1.10 Periodic Verification Procedure STA-754, " Comanche Peak Steam Electric Station Motor-operated Valve Program," Revision 0 (effective May 1, 1993) states in Section 6.7, that,
"after initial baseline testing, MOVs shall be assessed by performing periodic testing as appropriate to confirm continued readiness." During the inspection, the licensee stated that it planned to perform static tests in the effort to satisfy its commitment to GL 89-10 regarding periodic verification of design-basis capability. The licensee had not developed specific justification for its plan for periodic verification.
1.11 Schedule In a letter dated December 21, 1989, the licensee committed to meet the schedule of GL 89-10.
During the inspection, the licensee reported that 33 MOVs remained to be dynamically tested in Unit 1, which will be performed
'
before the end of 1993. The licensee reported that all of the M0V dynamic testing had been completed in Unit 2.
When dynamic testing is complete, the
licensee estimated that approximately half of the 123 GL 89-10 MOVs in Unit I and almost all of the 119 GL 89-10 MOVs in Unit 2 will have been dynamically tested under its GL 89-10 program. The licensee plans to set up the MOVs, which are not dynamically tested, using the best data available (primarily MOV test data from Unit 2). The licensee intends to complete initial setup and testing to demonstrate the design-basis capability of_ all MOVs within the
,
scope of GL 89-10 by June 1994. The licensee intends to continue the long-term aspects of GL 89-10 following the completion of initial MOV setup and e
testing.
1.12 Pressure Lockina
,
The Office for Analysis and Evaluation of Operational Data (AE00) has completed a study of pressure locking and thermal binding of gate valves.
AE00 concluded in its Report that licensees have not taken sufficient action to provide assurance that pressure locking or thermal binding will not prevent a gate valve from performing its safety function. The NRC regulations require-9-
.
-
. -
-
-
.
.
.
.
.
,
that licensees design safety-related systems to provide assurance that those systems can perform their safety functions.
In GL 89-10, the staff requested licensees to review the design bases of their safety-related MOVs.
The team reviewed the licensee's evaluations of the potential for pressure locking and thermal binding of gate valves as part of their design-basis f
review. Where such a potential was identified, the team reviewed the
,
licensee's actions taken to prevent pressure locking and thermal binding. The licensee's initial search identified 23 valves that were potentially susceptible to pressure locking or thermal binding. The licensee committed to use the results of a design-basis review done in accordance with GL 89-10 and l
expanded the scope of the review by adding 11 valves, which had been excluded
'
based on low-system pressure. A total of 8 valves were recommended to be modified.
Further reviews determined that because of plant operating conditions, only 4 valves in each unit required modification. These valve design changes were completed (DCA-95344 and DM-89-303).
,
!
Although the licensee had addressed normally closed valves, they did not
[
consider pressure locking to be credible for certain normally open valves.
'
The justification was that inadvertent closure, which could lead to a safety-related recovery to the open position, would be accompanied by light indications or alarms.
Plant operators would be expected to restore the valves to their open position prior to substantial heating of water, which may be trapped in the bonnet cavity. The inspectors questioned the licensee as to whether these administrative controls were sufficient in the event of
'
maintenance, surveillance, or testing activities which might preclude the restoration of these valves for a longer duration of time. The licensee informed the inspectors that further evaluation of these normally open valves would be necessary to establish whether the administrative controls were sufficient. The licensee's further evaluation of normally open valves will be reviewed during a future inspection and has been identified as an inspection followup item (445/9310-08; 446/9310-08).
1.13 MOV Failures. Corrective Actions. and Trendina In GL 89-10, the NRC requested that the licensees analyze or justify each MOV failure and corrective action. The documentation should include the results and history of each as-found deteriorated condition, malfunction, test,
,
inspection, analysis, repair, or alteration. All documentation should be
retained and reported in accordance with plant requirements.
It was also
'
suggested that the material be periodically examined as part of the monitoring i
and feedback effort to establish trends of MOV operability. 'These trends
!
could provide the basis for a licensee revision of the testing frequency
.
established to verify MOV capability on a periodic basis. The GL indicated
that a well-structured and component-oriented system would be necessary to i
track, capture, and share equipment history data.
[
At the time of the GL 89-10, Part 1 inspection, the licensee was using established programs for trending; however, they planned to develop a trending
,
system specifically for the GL 89-10 MOVs. During this inspection, the
.
- 10 -
.
.
.
.
i inspectors reviewed the licensee's progress in developing a trending system for the GL 89-10 MOVs. The licensee had issued Procedure STA-754, " Motor-Operated Valve Program," Revision 0, on March 12,1993.
This procedure established responsibilities for developing and maintaining the MOV performance monitoring and trending program and for performing trending analysis.
It further required that procedures be developed to specify MOV performance monitoring and trending activities and techniques for assessing performance changes. The licensee had two Technical Support Procedures (TSPs)
,
in final draft format that would establish the administrative requirements for MOV performance monitoring and trending (TSP-711) and MOV failure analysis and
-
trending (TSP-710). The team considered the existing and proposed procedural controls to be sufficient to establish a good trending program.
1.14 Toroue Measurement Technioue The team noted that the licensee utilized stem-mounted strain gages to directly measure torque as part of their diagnostic instrumentation.
Further, the licensee calibrated the strain gages in-situ by manually applying torque directly to the stem using a calibrated electronic torque wrench. The licensee subsequently compared the measured stem torque to the allowable torque rating for the actuator and the motor to determine adequate component
-
sizing and proper torque switch setpoints.
While the licensee's technique appeared to be an accurate measurement of reaction torque of the stem, the team questioned the licensee's use of stem reaction torque as a direct measurement of actuator output torque and an indicator of applied motor torque. The team was concerned that internal actuator losses because of improper maintenance would not be identified using i
the licensee's method, especially for valves that were set to close on a limit switch.
For example, improper shimming of the upper thrust bearing of the actuator has been found to reduce available motor torque because of increased i
internal friction losses. These losses would not be identified nor accounted for using the licensee's technique.
,
The licensee acknowledged the team's concern and stated that a quantitative check of springpack displacement would be included in the test acceptance criteria to identify any significant losses of torque internal to the actuator. This issue will be tracked as an inspection followup item (445/9310-09; 446/9310-09).
1.15' followuo of Weaknesses and Observations Enumerated in the Part 1 Report
!
At the time of the Part 1 inspection, four H0Vs in the component cooling water (CCW) system were in the process of being excluded from the GL 89-10 scope. Based on a review of Technical Evaluation 92-1130, the NRC concluded
that the exclusion of these valves was acceptable as documented in NRC Inspection Report 50-445/92-34; 50-446/92-34. Another valve, 1-8109, the positive displacement charging pump minimum flow recirculation valve, was also identified as being excluded. The licensee has subsequently reinstated Valve 1-8109 into the GL 89-10 program. However, the exclusion of this valve
- 11 -
,
.
.
.
.
.
f from the program will be further evaluated by the licensee when additional clarification on " position changamble" valves is received from the NRC.
The licensee was using various contractors to aid them in developing their program. The licensee agreed to review all documents for applicability to Comanche Peak Steam Electric Station prior to being included in the Comanche Peak Steam Electric Station controlling documents.. The design-basis review calculations performed by contractors were reviewed by the licensee for plant-specific information. Applicable information was consolidated in the Comanche Peak Steam Electric Station controlling document, Calculation ME-CA-0000-1093, Revision 2, for Unit I and Calculation 2-ME-0241, Revision 0, for Unit 2.
A study reviewed during the Part 1 inspection, which allowed the use of 2.5 times the manufacturer's rated thrust capacity and 2.0 times the manufacturer's rated torque capacity for Limitorque actuators, had not been adequately justified for use at Comanche Peak Steam Electric Station and, therefore, was considered inconsistent with the recommendations of GL 89-10.
A one-time allowable load of 2.0 times the manufacturer's rated torque capacity and 2.5 times the manufacturer's rated thrust had been established by Limitorque Corporation and was observed during this inspection to have been documented in Calculation ME-CA-0000-1093, Revision 2, Attachment B.
The inspectors on the Part I inspection found that the MOV setpoints appeared to be in at least three documents for each unit. The licensee stated that once the calculations were approved, all MOV setpoints would be placed in the 2401 drawings and controlled in accordance with their design control program.
Current design documents, station procedures, and testing procedures provide proper M0V setpoint control at Comanche Peak Steam Electric Station.
The licensee planned to use equipment developed by several vendors and combine the abilities of this equipment to monitor the capability of their MOVs. The inspectors viewed this proactive approach as a strength in the licensee's GL 89-10 program. However, as with any new technique or methodology, the burden of proof to ensure the validity of this approach is upon the licensee.
Further, at the time of the previous inspection, the accuracies for the licensee's equipment had not been documented. VITALS (M0 VATS 3500) and transducer accuracies have now been accounted for in the licensee's test acceptance criteria.
- 12 -
j
.
.
.
.
'
,
ATTACHMENT 1 1 PERSONS CONTACTED 1.1 Licensee Personnel
- 0. Bhatty, Licensing
- B. Black, Senior Engineer
.
- M. Blevins, Director of Nuclear Overviews
- W. Cahill, Group Vice President
- R. Calder, Manager, Reactor Engineering
- C. Catino, Test Engineer
- P. Chiu, Senior Engineer
- J. Donahue, Manager, Operations
- S. Ellis, Manager, Power Ascension
- W. Guldemond, Manager, Independent Safety Engineering Group
- C. Harrington, Supervisor, Mechanical Engineering
- T. Heatherly, Licensing Engineer
- T. Hope, Licensino Manager, Unit 2
- F. Madden, Manager, Mechanical Engineering
- D. Manning, Diagnostic Test Lead
- D. McAfee, Manager, Quality Assurance
- P. Mills, Senior Quality Assurance Specialist
- J. Muffett, Manager, Technical Support and Design Engineering
- M. Sunseri, Manager, Maintenance Engineering
- C. Terry, Vice President, Nuclear Engineering and Support
- R. Withrow, Supervisor, Component Test 1.2 NRC Personnel
,
- B. Jones, Senior Resident Inspector
- G. Werner, Resident Inspector In addition to the personnel listed above, the inspectors contacted other personnel during this inspection period.
-
,
- Denotes personnel that attended the exit meeting on March 19, 1993.
- Denotes personnel that participated in the supplemental exit meeting on April 7, 1993.
>
2 EXIT MEETING An exit meeting was conducted on March 19, 1993. During this meeting, the inspector reviewed the scope and findings of the report. The licensee-identified one document reviewed by the team as proprietary. This document
,
contained design-basis equations prepared generically for Westinghouse nuclear plants. No proprietary information is included in this report.
-t A supplemental exit meeting was conducted on April 7,1993, during which the -
NRC informed the licensee of the disposition of two unresolved items i
identified during the inspection.
,
.
.
P~
,
.
.
-
!
.
ATTACHMENT 2
)
i INSPECTION FINDINGS INDEX Inspection Followup Item 445/9310-01; 446/9310-01 was opened
-
(Section 1.2).
,
Violation 445/9310-02; 446/9310-02 was opened (Section 1.2).
Inspection Followup Item 445/9310-03; 446/9310-03 was opened
(Section 1.2).
Violation 445/9310-04; 446/9310-04 was opened (Section 1.3).
.
Inspection Followup Item 445/9310-05; 446/9310-05 was opened
(Section 1.4).
Inspection Followup Item 445/9310-06; 446/9310-06 was opened
!
(Section 1.6).
i Inspection Followup Item 445/9310-07; 446/9310-07 was opened
-
(Section 1.7).
,
r Inspection Followup Item 445/9310-08; 446/9310-08 was opened
(Section 1.12).
Inspection Followup Item 445/9310-09; 446/9310-09 was opened
-
(Section 1.14).
i
?
.[
P
.
.
-
--.
J