ML060690008
| ML060690008 | |
| Person / Time | |
|---|---|
| Site: | North Anna  |
| Issue date: | 03/09/2006 |
| From: | Casto C Division Reactor Projects II |
| To: | Grecheck E Virginia Electric & Power Co (VEPCO) |
| References | |
| EA-06-02 IR-05-004 | |
| Download: ML060690008 (19) | |
See also: IR 05000338/2005004
Text
March 09, 2006
EA-06-02
Virginia Electric and Power Company
ATTN: Mr. Eugene S. Grecheck
Vice President
Nuclear Support Services
Innsbrook Technical Center - 2SW
5000 Dominion Boulevard
Glen Allen, VA 23060-6711
SUBJECT: RESPONSE TO DISPUTED MINOR VIOLATION AND FOUR CROSS-
CUTTING ASPECTS CONTAINED IN NRC INTEGRATED INSPECTION
REPORT NOS. 05000338/2005004 AND 0500339/2005004 - NORTH ANNA
POWER STATION UNITS 1 & 2
Dear Mr. Grecheck:
I am writing in response to your letter dated December 9, 2005, in which you documented your
disagreement with several elements of NRC Inspection Report Nos. 05000338/2005004 and
0500339/2005004 dated October 28, 2005. In your December 9, 2005, letter, you contested
the minor violation and the existence or significance of four cross-cutting aspects described in
the inspection reports. Your letter provided reasons to support your contentions.
The NRC has carefully reviewed the documentation provided to support your positions. Based
upon our review, we have concluded, for the reasons outlined in the attached enclosure, that
the finding, which was identified during an annual inspection sample for identification and
resolution of problems, was appropriately classified as a minor violation. Three of the four
contested cross-cutting aspects occurred as stated. One non-cited violation (NCV) was
reclassified as a minor violation, therefore the NRC will modify the inspection reports
appropriately to reflect this reclassification. As a minor violation, it also does not contribute to
the assessment process used to determine if there is a substantive cross-cutting issue.
In accordance with 10 CFR 2.390 of the NRC's "Rules of Practice," a copy of this letter and its
enclosure will be available electronically for public inspection in the NRC Public Document
Room or from the Publicly Available Records (PARS) component of NRCs document system
(ADAMS). ADAMS is accessible from the NRC Web site at
http://www.nrc.gov/reading-rm/adams.html (the Public Electronic Reading Room).
2
Should you have any additional questions, please contact Kerry Landis at (404) 562-4510.
Sincerely,
/RA by Stephen J. Cahill Acting for/
Charles Casto, Director
Division of Reactor Projects
Docket Nos.: 50-338/339
License Nos.: NPF 4, NPF 7
Enclosure: RESPONSE TO DOMINIONS DISAGREEMENT WITH INSPECTION REPORT
NOS. 05000338/2005 AND 0500339/2005004
cc w/encl.: C. L. Funderburk, Manager Executive Vice President
Nuclear Licensing and Old Dominion Electric Cooperative
Operations Support Electronic Mail Distribution
Virginia Electric and Power Company
Electronic Mail Distribution County Administrator
Louisa County
Jack Davis P. O. Box 160
Site Vice President Louisa, VA 23093
North Anna Power Station
Virginia Electric and Power Company Lillian M. Cuoco, Esq.
Electronic Mail Distribution Senior Nuclear Counsel
Dominion Nuclear Connecticut, Inc.
Donald Jernigan Electronic Mail Distribution
Site Vice President
Surry Power Station Attorney General
Virginia Electric and Power Company Supreme Court Building
Electronic Mail Distribution 900 East Main Street
Richmond, VA 23219
Distribution w/encl: (See page 3)
_________________________
OFFICE RII:DRP RII:DRP RII:EICS OE NRR
SIGNATURE JTR KDL CFE LT per phone RP per email
NAME JReece KLandis CEvans LTrocine RPascarelli
DATE 02/16/2006 03/09/2006 03/09/2006 03/09/2006 03/09/2006
E-MAIL COPY? YES NO YES NO YES NO YES NO YES NO YES NO YES NO
RESPONSE TO DOMINIONS DISAGREEMENT WITH
INSPECTION REPORT 05000338, 339/2005004
I. Inspection Report Extract and Licensee Comment - Flood Protection Finding:
1R06 Flood Protection Measures
a. Inspection Scope
The inspectors reviewed internal flood protection measures for the Unit 1 and 2 air
conditioning chiller rooms (ACCRs) and adjacent air conditioning fan rooms (ACFRs).
Flooding in the ACCRs and ACFRs could impact risk-significant components in the
instrument rack rooms adjacent to the ACFRs if flood mitigation features were
degraded. ACCR and ACFR protection features were observed to verify that they were
installed and maintained consistent with the plant design basis. The inspectors
reviewed the instrumentation and associated alarms for the rooms above to verify that
the instrumentation was periodically calibrated and that the respective alarms were
appropriately integrated into plant procedures. The inspectors also reviewed licensee
instructions in the event of severe flooding and evaluated the availability of systems,
structures and components (SSCs) for safe shutdown under worst case water levels.
Documents reviewed are listed in the Attachment.
b. Findings
Inadequate Corrective Action Results in Safeguards Instrument Rack Room Flood
Problem
Introduction. The inspectors identified a self-revealing violation associated with
inadequate corrective action. Back-flow preventers were not installed in floor drains that
resulted in a flood potential for the Unit 1 and 2 Safeguards Instrument Rack Rooms.
The safety significance is under evaluation and thus the item is classified as an
unresolved item (URI).
Discussion. On July 9, 2005, back flush of control room chiller service water strainers
2-HV-S-1A and 1B as directed by engineering transmittal, ET -05-0034, Operability of
2-HV-P-22C, Service Water Pump for 2-HV-E-4C, was performed in the Unit 2 ACCR.
During this work activity, the licensee observed water discharging from the floor drains
in the adjacent ACFR, and initiated Plant Issue N-2005-2565 to evaluate the absence of
back-flow preventers in the floor drains. The licensee initiated a flood watch, declared
the flood walls between the ACCR and adjacent ACFR on Units 1 and 2 inoperable, and
entered a Yellow 6 day maintenance rule risk condition based on the unavailability of the
flood walls to perform their function. The respective ACFR on both units are adjacent
and open to the safeguards instrument rack rooms, which contain the solid state
protection system (SSPS) and process instrumentation and are at a 2 feet lower
elevation. Each instrument rack room has a sump with two pumps rated at 40 gpm
each. On Unit 2 the sump pumps discharge line is hard-piped directly to the ACCR
sump.
Enclosure
2
However, on Unit 1 the sump pumps discharge line is routed to a drain funnel
interconnected to the floor drain system of the adjacent ACFR. The licensee
determined that this funnel did not have a back-flow preventer installed and initiated
Plant Issue N-2005-2597. A subsequent calculation, ME-0782, was performed by the
licensee to evaluate the consequences of a service water line break in either the Unit 1
or 2 ACCRs. The calculation concluded that the peak flow rate from the Units 1 and 2
ACCRs to adjacent ACFRs via the floor drain piping was 182.9 gpm and 169.4 gpm
respectively.
The inspectors reviewed the licensees corrective action database and determined that
on October 15, 2004, Plant Issue N-2004-4554 was initiated due to water discharge
from a capped floor drain outside of the ACCR. An other evaluation was assigned to
engineering to review this condition for impact on the flood protection assumed for the
ACCR and connecting areas as applicable. This evaluation did not identify and correct
the absence of back-flow preventers in the adjacent ACFR floor drains. The inspectors
also identified that Plant Issue N-1999-3405, which documented operational experience
from Three Mile Island regarding check valves missing from floor drains and the impact
on flood protection, did not result in the identification and correction of this problem.
The inspectors concluded that the inadequate corrective actions for Plant Issue
N-2004-4554 is contrary to the requirements of 10 CFR 50, Appendix B, Criterion XVI,
which requires that the establishment of measures to assure conditions adverse to
quality are promptly identified and corrected.
Analysis. The inspectors determined that the finding had a credible impact on safety
based on the potential for flooding to impact both trains of SSPS cabinets used for
engineered safeguards. The inspectors referenced IMC 0612 and determined that if left
uncorrected this finding would result in a more significant safety concern and is
consequently more than minor. Based on a review of IMC 0609 for the SDP, the
inspectors determined the finding would require a Phase III evaluation due to the loss or
degradation of equipment specifically designed to mitigate a flooding event and the
impact on two trains of a safety system. This finding is an URI pending completion of
the significance determination assessment and contains aspects relating to the
cross-cutting area of problem identification and resolution.
Enforcement. 10 CFR 50, Appendix B, Criterion XVI, Corrective Action, requires the
establishment of measures to assure conditions adverse to quality are promptly and
identified and corrected. Contrary to the above, prompt identification and correction of
deficiencies relating to Plant Issue N-2004-4554 failed to identify and correct the
absence of back-flow preventers in the Unit 1 and 2 ACFRs. This violation is
characterized as an URI pending significance determination, and is identified as URI
05000338, 339/2005004-02, Inadequate Corrective Action Results in Safeguards
Instrument Rack Room Flood Problem. This finding is in the licensee's CAP as Plant
Issue N-2005-2565.
3
Dominion Response:
As noted above, in October 15, 2004, Plant Issue (PI) N-2004-4554 was initiated due to
water leakage from a capped floor drain outside of the ACCR. This leakage was in the
ACCR sump pump discharge line in the turbine building basement. An evaluation was
performed. It was determined that due to the size of the leak and its location in the
turbine building basement it was not credible for this leaking capped floor drain to
adversely affect the operability of any equipment in the chiller room and connecting
areas.
The leak in the basement of the turbine building is isolated from the ACCR by a
floodwall and associated piping of the sump pump. There is no direct interconnected
piping that can circumvent the flood barrier. In the 2004 event, leakage would have
been required to overflow the floodwall before affecting safety-related systems. The
evaluation of the 2004 event was directed at the potential for an overflow condition and
established that the flood control design could not credibly be breeched. Therefore,
neither the leakage phenomenon nor the design features of prevention in the 2004
leakage event have any relationship with the 2005 event, which was associated with
interconnected systems and back-flow prevention to preclude flooding in the ACFR.
Based on the nature of the flooding event and the design features, it is unrealistic to
assume the 2004 evaluation should have addressed back flow preventers for areas that
would not be affected by leakage in the turbine basement.
In contrast, on June 9, 2005, during a flush of the control room chiller service water
strainers, water was noted to be discharging from the drains in the adjacent ACFR.
Actions were immediately initiated to evaluate and identify the source of the water and
subsequent corrective actions. This was captured in the Corrective Action Program
(CAP) as PI N-2005-2565. The evaluation of the condition noted in the PI resulted in
installation of back flow prevention devices on both units.
Conclusion: It is Dominion*s position that the evaluation and resultant corrective actions
for the 2004 event were necessary and sufficient because the turbine building basement
leakage was not an interconnected design issue and the leakage did not or could not
compromise the established design prevention features. Therefore, Dominion considers
that this event does not contain aspects relating to the cross-cutting area of problem
identification and resolution.
NRC Evaluation of Dominions Response:
Analysis: To satisfy the requirements of NRC Generic Letter 88-20 the licensee
performed a probabilistic risk assessment for internal flooding which included field
walkdowns to identify possible flood sources and paths. The results were published in
"Probabilistic Risk Assessment for the Individual Plant Examination Final Report North
Anna Units 1 and 2 December 1992," and subsequently validated by additional field
walkdowns in 2001. In response to the initial study the licensee performed modifications
which added flood walls and back flow preventers in various floor drains. However, in
both cases noted above, the field walkdowns failed to identify the flood path via floor
drains between the chiller room and the adjacent fan room located in the emergency
4
switchgear area. In 2004, the licensee initiated prompt identification and correction of
deficiencies relating to Plant Issue N-2004-4554. This presented an opportunity to
identify and correct the absence of back-flow preventers in the Unit 1 and 2 ACFRs.
After consultation with the Office of Enforcement, this violation is better characterized as
a violation of 10 CFR 50, Appendix B, Criterion III, Design Control, in that the licensee
failed to assure that the design change requiring back-flow preventers in the Unit 1 and
2 ACFRs were appropriately specified and implemented. Final documentation of this
violation will be accomplished during the closeout of this URI.
The NRC reviewed Plant Issue N-2004-4554 and identified a comment in the evaluation
response section 3), regarding failure of a flood wall between the air conditioning chiller
room (ACCR) and turbine building area resulting in leakage to the ACCR from a
postulated turbine building flood. The NRC also identified the following comment,
NOTE: There is a flood wall in the ESGR that would prevent any water in the chiller
room from entering the ESGR. The NRC noted that the ESGR is the emergency
switchgear room which is used by the licensee to denote a general area that also
contains the solid state protection system (SSPS) and process instrumentation room
and air conditioning fan room. The NRC also noted that the ACCR is between the
turbine building and the ESGR and that a flood wall is located at each door for the
ACCR. Therefore, the above information demonstrates that the licensee considered a
flood situation starting in the turbine building and potentially involving the ESGR areas
via the ACCR. However, the licensees evaluation assumed that the flood wall between
the ACCR and ESGR was adequate and, therefore, did not include more rigorous steps
such as walkdowns and or consideration of operating experience (OE), which could
have revealed the floor drain issue. The licensees corrective action database contained
significant OE. The NRC Integrated Inspection Report Nos. 05000338/2005004 and
05000339/2005004 mentioned one, Plant Issue N-1999-3405, regarding a Three Mile
Island plant issue of check valves missing from floor drains and the impact on flood
protection. The NRC also identified another OE example: Plant Issue N-1990-0020, IN
83-44-S1: Potential damage to redundant safety equipment as a result of backflow
through the equipment and floor drain system.
NRC Conclusion:
After consultation with the Office of Enforcement, this violation is better characterized as
a violation of 10 CFR 50, Appendix B, Criterion III, Design Control. Final documentation
of this violation will be accomplished during the closeout of this URI. During the NRC
review of Plant Issue N-2004-4554, the inspectors identified a comment in the
evaluation response which addressed a flood scenario between the ACCR and ESGR
which included the SSPS instrumentation room. This evaluation missed an opportunity
to identify the absence of back-flow preventers in the adjacent ACFR floor drains.
Consequently, this design control finding contains aspects of the cross-cutting area of
problem identification and resolution.
5
II. Inspection Report Extract and Licensee Comment - Actuator Oil Leakage on Turbine
Interface Valve:
1R14 Operator Performance During Non-Routine Evolutions and Events
a. Inspection Scope
The inspectors reviewed operator logs and plant computer data for the two events listed
below to determine if plant and operator responses were in accordance with plant
design, procedures, and training. The inspectors also evaluated performance and
equipment problems to ensure that they were entered the licensees CAP.
The inspectors evaluated the response of the Unit 1 and 2 control room
operators on August 5 and 6, 2005, during an unplanned down power of Unit 1
for diaphragm replacement on 1-EH-TV-100, and,
The inspectors evaluated the response of the Unit 2 control room operators on
August 5 and 6, 2005, following an automatic reactor trip which occurred during
the Unit 1 down power event above.
.2 Unit 1 Rapid Power Reduction Due to Loss of Turbine Auto Stop Oil Pressure
Introduction: A Green, self-revealing finding was identified for not performing Unit 2
corrective actions in a timely manner on Unit 1. This resulted in the Unit 1 rapid
reduction of power from 100% to ~8% (main turbine off-line) on August 5, 2005.
Description: On August 5, 2005, the licensee rapidly reduced power on Unit 1 due to
severe oil leakage on the actuator for valve, 1-EH-TV-100 (Main Turbine Auto Stop Oil
Interface Valve). Subsequent evaluations determined that the torque specifications of
12-13 ft-lbs as specified in maintenance procedure 0-MCM-1412-01,Main Turbine
Interface Valve Diaphragm Replacement, did not provide adequate clamping force
between the diaphragm and actuator cover flange faces which resulted in diaphragm
movement and oil leakage from the actuator. The inspectors determined that an
actuator oil leak from the same valve resulted in a manual reactor trip due to low
electro-hydraulic or auto stop oil pressure on April 19, 2003. The inspectors reviewed
the root cause evaluation from that event and concluded that the licensee did not
contact the vendor for specific torque values. The inspectors also reviewed a December
2004, event involving similar leakage on the Unit 2 equivalent valve. In this case, the
resultant evaluation concluded that the interface valve diaphragm torque values should
have been 20 ft-lbs per vendor technical manual 59-264-00006, Fisher Instruction
Manual, Types 655 and 655R Actuators for Self-Operated Control. However, the
inspectors determined that associated corrective actions for Unit 1 had not been
implemented prior to the August 5, 2005, rapid down-power event.
Analysis: This finding had a credible impact on safety due to the challenge of plant
control systems from the rapid reduction of power. The inspectors referenced IMC 0612
and determined that the finding was more than minor based on the impact to the
Initiating Events cornerstone objective to limit the likelihood of those events that upset
6
plant stability and the cornerstone attribute of equipment reliability. The inspectors
referenced IMC 0609 for the SDP and determined that the finding is Green (very low
safety significance) because it did not contribute to the likelihood of a primary or
secondary system LOCA initiator or a loss of mitigation equipment functions, and did not
increase the likelihood of a fire or internal/external flood. This issue is in the licensees
CAP as Plant Issue N-2005-2984. This finding contains aspects relating to the
cross-cutting area of problem identification and resolution.
Enforcement: Since this finding is associated with nonsafety-related secondary plant
equipment, no violation of regulatory requirements occurred. Therefore, this finding is
identified as a Green finding FIN 05000338/2005004-04, Untimely Corrective Actions for
Actuator Oil Leakage on Turbine Interface Valve Results in Rapid Down Power.
Dominion Response:
On December 30, 2004, it was identified that oil drops were hanging from each bolt
around the diaphragm of the Unit 2 Autostop Oil Interface Valve, 2-EH-TV-200. The oil
was removed and the valve monitored. On December 31, 2004, oil was again identified
in the threads of the diaphragm bolts, but no drops had formed. Subsequently, an
engineering evaluation concluded that the interface valve diaphragm torque values
should have been 20 ft-lbs. when the diaphragm was replaced. The maintenance
procedure for diaphragm replacement was revised on July 14, 2005, to include the
20 ft-lbs. value.
As a result of the December observations, both units* Main Turbine Auto Stop Oil
Interface Valves were being routinely monitored by Operations during normal rounds
with periodic monitoring by System Engineering during their walkdowns. Since there
was no observed active leakage at that time, checking the torque of both units* interface
valves was not immediately performed due to the potential threat of tripping the units
while online. The only means to completely address (i.e., valve disassembly) this issue
immediately was to initiate a two-unit shutdown with the attendant risk associated in
such an evolution. Based on the limited presence of oil leakage observed up to that
point and the associated risk of immediate action, a decision was made to continue
monitoring and await the first available and more appropriate time to check the torque of
the diaphragm bolts.
Due to a minor leak on one bolt, on August 1, 2005, the Unit 2 Main Turbine Auto Stop
Oil Interface Valve (2-EH-TV-200) was torqued to 12 ft-lbs. Based on the lack of
leakage and past history indicating satisfactory performance of 1-EH-TV-100, checking
the torque on 1-EH-TV-100 was not immediately attempted. Work requests to check
the torque on 1-EH-TV-100 were written with plans to implement during the week of
August 8, 2005. Again, at the time there were no immediate operability concerns. On
August 5, 2005, Dominion rapidly reduced power on Unit 1 due to severe oil leakage on
the actuator for 1-EH-TV-1 00. The diaphragm was replaced and torqued to 20 ft-lbs. in
accordance with the revised maintenance procedure.
Conclusion: It is Dominion*s position that problem identification was appropriately
documented and the resolution had been purposefully scheduled to minimize the risk of
tripping the units. As a consequence, we conclude that this should not be considered a
7
cross-cutting concern in the area of problem identification and resolution, as the
identification and resolution process were purposefully and reasonably exercised.
NRC Evaluation of Dominions Response
Analysis: Dominion stated that a subsequent engineering evaluation concluded that the
diaphragm torque value should have been 20 ft-lbs when the valve diaphragm was
previously replaced on 2-EH-TV-200, and that a procedure change to include the 20 ft-lb
value was completed on July 14, 2005. The NRC determined that the licensee
documented their awareness of the required torque value on March 22, 2005, in Plant
Issue N-2004-5408 which also included statements about the vendor technical support
staff acknowledging that the required torque setting is different from the licensees
current torque valve specified by procedure and, He was more concerned about the low
actuator diaphragm casing torque value, since this is directly responsible for retaining
the diaphragm. The plant issue also stated that eminent failure of the diaphragm is not
expected at the current torque valve, since it has been used successfully for the past
25+ years. However, the NRC determined that this is an incorrect statement based on
the similar event that resulted in a manual reactor trip on April 19, 2003. The NRC
identified that operating experience, OE 17170: Turbine Lube Oil Interface Valve Oil
Leak, as listed in the licensees corrective action program as simply Plant Issue, was
received on October 30, 2003, and reviewed on November 5, 2003, and is, therefore,
within the 2 year window of the August 5, 2005 down-power event. The licensees
review of this OE stated, Close: No definitive root cause information provided; RCE
N-2003-1761 addresses the torque p(r)ocess under section 4.2; FYI for technical
information. The NRC review of section 4.2 found that the only statement addressing
torque information was, Turbine Generator Coordinator (TGC) oversight of the
diaphragm cover installation torque process will also be added. The NRC concluded
that this failure to determine the correct torque requirements from the Plant Issue
addressing OE 17170 is also inadequate corrective action.
Dominion stated that the only means to completely address this issue immediately was
to initiate a two-unit shutdown with the attendant risk associated in such an evolution.
Yet, the NRC noted that Dominion also performed maintenance with Unit 2 in service
based on their statement that due to a minor leak on one bolt, on August 1, 2005, the
Unit 2 Main Turbine Auto Stop Oil Interface valve (2-EH-TV-200) was torqued to 12
ft-lbs. The NRC reviewed the associated work order, 00605404-01, Torque Diaphragm
casing bolts, and noted that the actions taken were, Set torque wrench to 12 ft.lbs.
Tightened bolting outside leak working inwards to center leak. Did the same on the
opposite side. All bolts except the one at the south side leak have less than 12 ft.lbs.
The bolt at southside leak is torqued to 12 ft.lbs. The NRC also reviewed engineering
logs and noted that the entry dated August 5, 2005, at 1043 hours0.0121 days <br />0.29 hours <br />0.00172 weeks <br />3.968615e-4 months <br />, stated that an oil
drop had accumulated on the bottom of the North bolt location for the Unit 1 valve,
1-EH-TV-100. This is different from previous entries which denote only slight seepage,
no oil drips.
The potential for the valve diaphragm to rapidly degrade causing a plant transient was
demonstrated by the fact that an oil drop had accumulated on Unit 1 valve,
1-EH-TV-100 on August 5, 2005, at 1043 hours0.0121 days <br />0.29 hours <br />0.00172 weeks <br />3.968615e-4 months <br /> and subsequently that day Dominion
rapidly reduced power on Unit 1 due to severe oil leakage on the actuator for
8
1-EH-TV-100.
NRC Conclusion:
The NRC determined that Dominion had specific knowledge that:
1. The licensee was aware of the fact that the licensees Turbine Interface Valve
bolt torque requirement of 12 ft.lbs was less than the vendor specified torque
requirement of 20 ft.lbs. Additionally the licensee knew that adequate torque
was critical to retaining the diaphragm as documented in the March, 22, 2005,
Plant Issue response following the December 2004 event involving similar
leakage on the Unit 2 Main Turbine Auto Stop Oil Interface Valve, 2-EH-TV-200;
2 The 2-EH-TV-200 bolting was found with less than 12 ft.lbs torque values after
valve maintenance was performed on August 1, 2005, with the unit remaining in
service; and extent of condition on Unit 1 was not evaluated;
3. An increase in the oil leakage was noticed on 1-EH-TV-100 actuator on the
morning of August 5, 2005, and later during that day increasing oil leakage
resulted in a rapid down-power event.
Dominion failed to take adequate and timely corrective action for the actuator oil leakage
on the Unit 2 Turbine Interface Valve following the December 2004 event to preclude
the Unit 1 rapid down-power event on August 5, 2005.
III.1. Inspection Report Extract and Licensee Comment - Quench Spray Pump Safety
Related Breaker - Finding relating to SSPS Testing:
1R22 Surveillance Testing
a. Inspection Scope
For the nine surveillance tests listed below, the inspectors examined the test procedure,
witnessed testing, and reviewed test records and data packages, to determine whether
the scope of testing adequately demonstrated that the affected equipment was
functional and operable, and that the surveillance requirements of the TS were met:
- 1-PT-63.1A, Quench Spray System A Subsystem (1-QS-P-1A), an inservice
test,
- 2-PT-71.2Q, Unit 2 Motor Driven Auxiliary Feedwater (2-FW-P-3A) Pump Test;
- 1-PT-52.2, Reactor Coolant System Leak Rate (Hand Calculation) VPAP-0502 -
Procedure Process Control;
- 2-PT-82J, 2J Diesel Generator Test Slow Start Test;
- 2-PT-63.1B, Quench Spray System - B Subsystem;
- 2-PT-213.8B, Valve Inservice Inspection (B Train of Safety Injection System);
- 2-PT-31.7, Pressurizer Level Channel (2-RC-L-2459) Channel Operational
Test;
- 1-PT-75.2B, Unit 1 Service Water Pump (1-SW-P-1B); and,
- 2-PT-57.1B, Emergency Core Cooling Subsystem - Low Head Safety Injection
Pump (2-SI-P-1B).
9
b. Findings
.1 Failure to Follow Procedures During SSPS Testing
Introduction. A Green, self-revealing NCV of TS 5.4.1.a was identified for failure to
implement a surveillance procedure which resulted in placing an incorrect bistable in a
trip condition.
Description. On July 22, 2005, during the performance of SSPS testing on Unit 2 in
accordance with procedure 2-PT-31.7, Pressurizer Level Channel I (2-RC-L-2459)
Channel Operational Test, of which step 6.1.5 requires placement of trip switches BS1
and BS2 on card C1-442 in the trip position, instrument technicians incorrectly placed
switches BS1 and BS2 on card C1-422 (same switch designation but a different card) in
the test position, which initiated an unexpected alarm (LO LO Tave Interlock Loop 1
A-B-C) in the control room. This caused Unit 2, Loop 1 T cold inputs to the SSPS
Relays K148 (Lo Lo Tave)(BS1) and K140 (Lo Tave)(BS2) to fail safe and show a trip
condition. A subsequent review by the inspectors of I/C drawings revealed that these
relays were Channel I inputs for P-12 (Lo Lo Tave Steam Dump Interlock) and
feedwater isolation permissives. The inspectors concluded that since loops two and
three were not in a trip condition, the two out of three logic was not satisfied, and the
plant was not affected.
Analysis. The inspectors reviewed IMC 0612 and determined that the finding was more
than minor because it could reasonably be viewed as a precursor to a more significant
event. If another channel in the logic had already been tripped, the plant would have
been adversely affected by the performance deficiency. The inspectors consulted IMC 0609 for the SDP and determined that the finding is Green (very low safety significance)
because it did not involve any LOCA initiators, did not contribute to both a reactor trip or
mitigating system unavailability, nor increase the likelihood of a fire. This finding
contains aspects relating to the cross-cutting area of human performance.
Enforcement. TS 5.4.1.a, requires that written procedures shall be established,
implemented, and maintained per RG 1.33, Appendix A, of which Part 8 stipulates
procedures for surveillance tests. Procedure, 2-PT-31.7.1, step 6.1.5. states, Place the
following comparator trip switches in TEST: On card C1-442, BS1 and BS2. Contrary
to the above on July 22, 2005, step 6.1.5 was improperly implemented in that
comparator switches, BS1 and BS2, on card C1-422 were placed in trip as opposed to
the switches on the correct card, C1-442. This finding is of very low safety significance
or Green, is in the licensees CAP as Plant Issue N-2005-2755, and thus is
characterized as an NCV, consistent with Section VI.A of the NRC's Enforcement Policy:
NCV 05000339/2005004-04, Failure to Follow Procedure During Solid State Protection
System Testing.
Dominion Response:
The event described above did include a human performance error. However, this error
was immediately identified and the channel returned to service. Furthermore, the
testing was stopped until the issue could be understood and resolved.
10
The plant was not adversely affected because loops two and three were not in a trip
condition and the human performance issue was immediately resolved. Without
consideration of potential additional failures, this issue did not significantly impact the
Initiating Events cornerstone and should not be considered as meeting the criteria for a
cross-cutting issue. To escalate human performance deficiencies with the burden of
additional single failures would render all procedural errors as substantive cross cutting
issues.
Conclusion: Dominion does not agree that the identified human performance deficiency
should be considered as relating to the cross-cutting area of human performance since
the Initiating Events cornerstone was not significantly and therefore not substantively
affected by the immediately corrected error.
NRC Evaluation of Dominions Response:
Analysis: The Dominion response agrees that the event did include human error,
however, the licensee makes three other points: (1) the human performance issue was
immediately resolved, (2) Without consideration of potential additional failures, this issue
did not significantly impact the Initiating Events cornerstone and should not be
considered as meeting the criteria for a cross-cutting issue, and (3) To escalate human
performance deficiencies with the burden of additional single failures would render all
procedural errors as substantive cross cutting issues.
Point 1: Whether the event was immediately corrected by returning the channel to
service has no bearing on the fact that human performance was a contributing cause of
the event or the identification of a human performance as a aspect of the event.
Point 2: The description and analysis In the inspection report referenced the other
channels of SSPS logic to clearly state that the plant was not effected. The licensee
appears to be taking issue with this violation being more than minor rather than meeting
the criteria for a cross-cutting issue.
Point 3: The case is being made that since there was no additional single failure that
any violation in this circumstance should not be characterized as more than minor
because the condition did not involve actual consequences affecting safety and/or
operability. After further review of IMC 0612, Appendix E.4. Example b, this violation
closely matched this example and is being reclassified as a minor violation.
NRC Conclusion:
The SSPS testing event was a procedure violation involving a human performance error,
however it is a minor violation based upon matching closely example b described in IMC 0612, Appendix E.4. Consequently, this minor violation and its cross-cutting aspect will
not be part of the evaluation to determine if there is a substantive cross-cutting issue.
11
III.2. Inspection Report Extract and Licensee Comment - Quench Spray Pump Safety
Related Breaker - Finding relating to the breaker overload device instantaneous pickup:
1R22 Surveillance Testing
a. Inspection Scope
For the nine surveillance tests listed below, the inspectors examined the test procedure,
witnessed testing, and reviewed test records and data packages, to determine whether
the scope of testing adequately demonstrated that the affected equipment was
functional and operable, and that the surveillance requirements of the TS were met:
- 1-PT-63.1A, Quench Spray System A Subsystem (1-QS-P-1A), an inservice
test,
- 2-PT-71.2Q, Unit 2 Motor Driven Auxiliary Feedwater (2-FW-P-3A) Pump Test;
- 1-PT-52.2, Reactor Coolant System Leak Rate (Hand Calculation) VPAP-0502 -
Procedure Process Control;
- 2-PT-82J, 2J Diesel Generator Test Slow Start Test;
- 2-PT-63.1B, Quench Spray System - B Subsystem;
- 2-PT-213.8B, Valve Inservice Inspection (B Train of Safety Injection System);
- 2-PT-31.7, Pressurizer Level Channel (2-RC-L-2459) Channel Operational
Test;
- 1-PT-75.2B, Unit 1 Service Water Pump (1-SW-P-1B); and,
- 2-PT-57.1B, Emergency Core Cooling Subsystem - Low Head Safety Injection
Pump (2-SI-P-1B).
b. Findings
.2 Failure to Follow Procedures Affecting Safety-Related Breakers
Introduction. A Green, self-revealing NCV of TS 5.4.1.a was identified for a failure to
follow procedures resulting in a trip of the Unit 2 Quench Spray Pump, 2-QS-P-1B.
Description. On August 19, 2005, during performance testing of 2-QS-P-1B per
2-PT-63.1B, Quench Spray System - B Subsystem, the respective motor breaker,
2-EE-BKR-24J1-4, closed and then immediately tripped open. The licensee
subsequently determined that two of the three as-found phase values of the breaker
overload device instantaneous pickup were low when compared to the North Anna
Setpoint Document (NASD) procedure which contains the setpoints, trip times and test
currents for all overload trip devices for 480-volt BBC/ITE K-Line Breakers. Therefore,
the motor starting current of approximately 3028 amps compared to the overload
instantaneous setpoints of 2268 amps and 2912 amps for B and C phases
respectively resulted in a premature trip of the breaker. The licensee previously
performed maintenance on this breaker on February 19, 2005, when the overload
devices were set and tested in accordance with electrical maintenance procedure,
0-EPM-302-02, BBC/ITE 480-volt K-Line Breaker & Associated Switchgear Cubicle
Maintenance, which references the NASD. Procedure 0-EPM-302-02, step 6.19.4.a.2
states, If the trip setpoint is within tolerance (80-120 percent) that was recorded in step
12
6.19.1, then go to substep 6.19.4.b, and if not, then make adjustments using
Attachment 5, Instantaneous And Short-Time Pickup Adjustment, and repeat steps
6.19.4.a.1 and 6.19.4.a.2. Contrary to the above, the technician performing the
maintenance left the B and C phase instantaneous overload setpoints low outside of
the allowable procedural tolerance at 3030 & 3002 amps respectively instead of within
the allowable procedural tolerance of 3080 to 4620 amps. The licensee determined that
a contributing cause was setpoint drift on the associated overload device. However, the
inspectors determined that given the worst case drift, B phase at 812 amps, and an
initial setpoint of 3850 amps (middle of the established ban), the resulting drift would
have resulted in a value above the motor starting current.
Analysis. The inspectors referenced IMC 0612 and determined that the finding was
more than minor because it affected the Barrier Integrity cornerstone objective to
provide reasonable assurance that the containment physical design barriers protect the
public from radio nuclide releases caused by accidents or events and the cornerstone
attribute of human performance. The inspectors referenced IMC 0609 for the SDP and
determined that the finding is Green (very low safety significance) because it did not
impact design deficiencies, result in a loss of system safety functions, exceed related TS
outage times, nor involve a seismic, flooding, or severe weather initiating event. This
finding contains aspects relating to the cross-cutting area of human performance.
Enforcement. TS 5.4.1.a, requires that written procedures shall be established,
implemented, and maintained as documented in RG 1.33, Appendix A, of which Part 9
stipulates procedures for maintenance. Procedure 0-EPM-302-02, step 6.19.4.a.2
stated, If the trip setpoint is within tolerance (80-120 percent) that was recorded in step
6.19.1, then go to substep 6.19.4.b, and if not, then make adjustments using
Attachment 5, Instantaneous And Short-Time Pickup Adjustment, and repeat steps
6.19.4.a.1 and 6.19.4.a.2. Contrary to the above, on February 19, 2005, this step was
not properly implemented or followed resulting in improper instantaneous overload
setpoints on B and C phases and a subsequent trip of 2-QS-P-1B. This finding is of
very low safety significance or Green, is in the licensees CAP as Plant Issue
N-2005-3225, and thus is characterized as an NCV, consistent with Section VI.A of the
NRC's Enforcement Policy: NCV 05000339/2005004-05, Failure to Follow Procedures
Affecting Safety-Related Breakers.
Dominion Response:
The procedure implementation issue is correct as stated. However, contrary to the
conclusions stated above, the root cause evaluation (RCE) determined that instrument
drift was the direct cause of the pump trip. The instantaneous overload device on
Breaker 2-EE-BKR-24J1-4 had drifted 27.3% from the previous as-left setpoint. This is
outside the acceptance criteria of +/-20% outlined in BBC Bulletin lB-8203 (Procedure
for Field Testing/Calibration of ITE K-Line Overcurrent Trip Devices). It should be noted
that the identified drift, by itself, was sufficient to cause the breaker failure.
The calibration was performed on February 19, 2005, and the pump was successfully
started twice prior to the failure on August 19, 2005. A human performance error did
occur when the trip setpoints were being installed on the breaker. However, this error
was not the cause of the subsequent failure. Had incorrect setpoints been the cause of
13
this event the pump would not have passed the post maintenance test.
As clarification, 3002 amps noted in the description section in the IR above is typed
incorrectly, it should have read 3020 amps.
Conclusion: Dominion does not agree that the event contained aspects relating to the
cross-cutting area of human performance since the root cause was determined to be
instrument drift. Specifically, the human performance deficiency was not substantive
and did not cause the pump trip nor impact a ROP cornerstone directly.
NRC Evaluation of Dominions Response:
Analysis: Subsequent to issuance of NRC Inspection Report 05000339/2005004 the
root cause evaluation (RCE) determined that instrument drift was the direct cause of the
pump trip. The NRC agrees that the A phase overcurrent instrument setpoint drift was
sufficient to trip the breaker. However, during review of the RCE, the inspectors
observed that even if the A phase setpoint had not drifted, the C phase as-left setpoint
could reasonably be expected to trip the breaker during some starts. The starting
current for the motor varies from the locked rotor current of 1750 amps to the maximum
calculated inrush current of 3028 amps depending upon the AC phase angle when the
breaker closes. Thus, the human error associated with leaving the C phase overcurrent
setpoint at 3020 amps, which was outside the allowable band, could randomly cause a
breaker trip. Furthermore, the NRC inspectors observed in the RCE that the allowable
band of 3080 to 4620 amps was unacceptable. The lower value of 3080 is only 52
amps above the maximum calculated inrush current of 3028 amps. This does not
provide sufficient margin to allow for possible instrument setpoint drift during the five
year calibration interval to prevent tripping of the breaker during a demand start. The
RCE also identified a contributing human performance error. The licensee identified
that the overcurrent instrument setpoint was approximately 20% lower than the
requirements in the North Anna Setpoints Document. Thus, the NRC considers that the
event revealed two human performance issues which directed impacted the reliability of
the Unit 2 B Quench Spray Pump.
This finding has a cross-cutting aspect of human performance and by itself does not
constitute a substantive cross cutting issue relative to human performance and
procedure errors.1
NRC Conclusion:
Although the NRC acknowledges that the A phase setpoint drift tripped the breaker,
there is no evidence that only the A phase tripped the breaker. As noted above, the
1
From MC 0305, a substantive cross-cutting issue should be corroborated by the
existence of a significant number (more than three (3)) of findings. The findings should share a
common performance characteristic more specific than the three cross-cutting areas and
should be from more than one cornerstone. However, it is recognized that given the significant
inspection effort applied to the mitigating systems cornerstone, that a substantive cross-cutting
issue may be observed through inspection findings associated with only this one cornerstone.
14
human errors associated with the as-left C phase setpoint and allowable band also
provided a mechanism for the breaker to trip at some random interval and thus impacted
the pumps reliability. The NRC considers the identification of a human performance
cross-cutting aspect as appropriate and will be part of the evaluation to determine if
there exists a substantive cross-cutting issue.
IV. Inspection Report With Contested Comment - TDAFW Outboard Bearing Leak:
4OA2 Identification and Resolution of Problems
a. Inspection Scope
The inspectors reviewed the licensees assessments and corrective actions for Plant
Issue N-2005-2320, during the performance of 1-PT-71.1Q (1-FW-P-2, Turbine Driven
Auxilliary Feedwater (TDAFW) pump), noted the outboard bearing slinger ring leaking oil
at approximately 3-4 drops per second. The Plant Issue was reviewed to ensure that
the full extent of the issue was identified, an appropriate evaluation was performed, and
appropriate corrective actions were specified and prioritized. The inspectors also
evaluated the Plant Issue against the requirements of the licensees CAP as specified in
VPAP-1601, Corrective Action Program, VPAP-1501, Deviations and 10 CFR 50,
Appendix B. Additional documents reviewed are listed in the Attachment.
b. Findings and Observations
No findings of significance were identified. On June 21, 2005, the licensee initiated
Plant Issue N-2005-2320 in response to an oil leak on the Unit 1 TDAFW pump
outboard bearing identified during the quarterly surveillance test. The licensee
completed a functional evaluation and declared a GL 91-18 condition (operable but
degraded) for the component. During subsequent testing, the licensee better quantified
the leak at 1.58 gallons per day as opposed to the original estimate of 8.5 gallons per
day. The inspectors verified the licensee functional evaluation which considered the
following facts that the design basis accident mission time for TDAFW operation is 8
hours and that the pump oil reservoir is maintained at 12 - 18 gallons of which 8 gallons
are below pump suction. This would result in a leakage of .53 gallons during the 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />
mission time resulting in the maintenance of pump operability. The inspectors reviewed
the history of bearing oil leaks for the Unit 1 and 2 TDAFW pumps which included work
order, 00505761-01, for an oil leak on the Unit 1 TDAFW pump outboard bearing which
was completed on September 18, 2004. The licensee subsequently identified this
corrective action as rework. The inspectors also found for the Unit 2 TDAFW pump an
Item Equivalency Evaluation Review (IEER) report, N95-5022-000, which installed new
seals of a different design due to similar problems of oil leakage. The licensee could not
explain why this same design had not been considered for the Unit 1 TDAFW pump.
The inspectors reviewed the IEER process as implemented by VPAP-0708, Item
Equivalency Evaluation, and the corrective action process as implemented by
VPAP-1601 and VPAP-1501. The inspectors determined that VPAP-0708 did not
perform an extent of condition review nor reference, consider or require a plant issue.
The inspectors also determined that neither VPAP-1601 or VPAP-1501 discussed the
IEER process as part of the CAP. The inspectors concluded the failure to implement
15
adequate corrective action for the Unit 1 TDAFW pump constituted a minor violation.
This finding is not yet captured in the licensees corrective action program.
Dominion Response:
The minor violation as stated is incorrect. VPAP-0708, Item Equivalency Evaluation,
establishes the requirements and methodology to ensure that alternate replacement
parts are evaluated for their interfaces, interchangeability, form, fit, and function for
parts installed in safety and non-safety related systems/components. This process
(lEER report, N95-5022) was used in 1995 to justify the installation of seals of a different
design on the Unit 2 TDAFW pump due to a problem with oil leakage. The new seal
design supported by the IEER resolved the oil leakage issue on the Unit 2 TDAFW
pumps. However, due to the continued and extended satisfactory performance of the
Unit 1 seals, it was not considered necessary or desirable to take immediate actions to
replace the existing seals with a design with no previous operational experience at North
Anna. The Unit 1 TDAFW pump*s oil seals continued to operate satisfactorily for the
next nine years. Once minor oil leakage was identified, as documented in PI -2005-
2320, corrective actions were initiated. The Operational Decision Making Report, written
in response to PI -2005-2320, determined the oil leak will not affect the TDAFW pump*s
ability to perform its design function for its established mission time. Therefore,
installation of new pump seals was scheduled for the Spring 2006 Unit 1 refueling
outage.
Conclusion: Dominion does not agree that this issue constitutes a minor violation
relating to the identification and resolution of problems. Replacement of the Unit 2
TDAFW pump seals used a process to ensure equivalency for the replacement seal.
The original design seal was still acceptable to perform its design function, and it did just
that. The initial Unit 2 seal leaks were corrected by a change in seal design and the Unit
1 seals were monitored. The Unit 1 TDAFW pump operated satisfactorily for 9 years
before a minor leak was identified. Once leakage was identified on the Unit 1 TDAFW
pump, corrective actions were scheduled commensurate with the safety significance.
NRC Evaluation of Dominions Response:
Analysis: The inspection report states that: "the failure to implement adequate
corrective action for the Unit 1 TDAFW pump constituted a minor violation." The minor
violation involves failure to take adequate corrective action for the September 18, 2004,
oil leak on the Unit 1 TDAFW pump outboard bearing. As noted in the inspection report,
work order 00505761-01 was initiated to repair the oil leak. The work order documented
as found and required clearances associated with the oil deflector. The required oil
deflector running clearance on the outboard bearing is .050 to .060 inches. The as
found clearance with the rotating element thrusted inboard was .052 inches. However,
when the rotating element was thrusted outboard, the as found clearance was
documented as .028 inches. The inspectors noted that there was no engineering review
signature on the work order. Additionally, the work order stated that the oil film may
have come from oil spillage during the pre-lube of the outboard bearing prior to a
surveillance test. This assumption was not verified as being accurate by an additional
run of the Unit 1 TDAFW pump. The inspectors could find no other corrective action
16
document which evaluated the cause of the leak or specified corrective actions to be
implemented such as continued monitoring during future pump tests. Subsequently, the
oil leakage on the Unit 1 TDAFW pump outboard bearing resulted in initiation of a Plant
Issue on June 21, 2005. The licensee performed an operability evaluation on July 8,
2005, and determined the Unit 1 TDAFW pump was operable, but degraded. The root
cause evaluation has been delayed until the refueling outage planned for March 12,
2006, at which time the licensee has planned to implement a modification to replace the
oil deflector seal with a labyrinth seal similar to that previously installed on the Unit 2
TDAFW pump. A seal oil leak is a condition adverse to quality, and thus actions are
required to identify and correct the problem as specified in 10 CFR 50 Appendix B,
Criterion XVI. This issue was inspected as an annual inspection sample for
identification and resolution of problems and documented as a sample required by the
inspection program. Using NRC guidance, this issue was also evaluated and identified
as being a minor violation.
The discussion concerning the IEER process involved the inspectors observations
concerning historical information about previous corrective actions for earlier leaks.
While the inspectors were reviewing available documentation to see if proper corrective
action was documented for the Unit 2 TDAFW, the IEER documentation was found and
reviewed. It was not considered as part of the minor violation.
NRC Conclusion:
The NRC considers failure to take adequate corrective action for the September 18,
2004, Unit 1 TDAFW pump outboard bearing seal oil leak as a violation, in that the
inspectors could find no corrective action documentation which adequately evaluated
the cause of the leak and specified corrective actions to be implemented to correct this
condition adverse to quality. Subsequently, the Unit 1 TDAFW pump outboard bearing
oil leak resulted in degraded operation of the Unit 1 TDAFW pump as documented in a
Plant Issue on June 21, 2005. The IEER process was reviewed but not considered part
of the minor violation.