IR 05000369/1993024
| ML20059C039 | |
| Person / Time | |
|---|---|
| Site: | McGuire, Mcguire  |
| Issue date: | 12/16/1993 |
| From: | Lesser M, Maxwell G NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20059C000 | List: |
| References | |
| 50-369-93-24, 50-370-93-24, NUDOCS 9401040339 | |
| Download: ML20059C039 (16) | |
Text
_sj.
,
-
UNITED STATES
!
/ ato\\
NUCLEAR REGULATORY COMMISSION k
J REGloN !!
"-
"
- 01 MARIETTA STREET, N.W., SUITE 2900
j ATLANTA, GEORGIA 30323-0199
- g
/
.....
Report Nos. 50-369/93-24 and 50-370/93-24
Licensee:
Duke Power Company
'!
422 South Church Street
Charlotte, NC 28242-1007
,
!
'
Facility Name: McGuire Nuclear Station 1 and 2
,
P Docket Nos. 50-369 and 50-370 License Nos.
>
Inspection Conducted: Octobe 1993 - November 20, 1993
,
.
M// f
-.3 Inspectors:
9c.4#L,o -
.
Date Signed
[pe G. F. Maxwell f enior bnt Inspector j
J. Zeiler, Resident Inspector, Catawba
,
,
kAf4- -- -
_--
/7[4[O Approved @:
M. 'S. Lesser, Section Chief Date Signed Division of Reactor Projects
SUMMARY
,
Scope:
This routine, resident inspection was conducted in the areas of
plant operations, Engineered Safety Features System walkdown, surveillance testing, maintenance observations, meeting with local i
officials, and followup of previous inspection findings.
Backshift inspections were performed on October 20, 21, 22, 26, 29, and November 2, 3, 5, 6, 7 and 11.
.
.
Results:
In the operations area a weak response was noted during a reactor l
transient because an operator placed the Rod Control. System in
,
manual without first verifying.that plant conditions warranted
!
this action (paragraph 2.d).
However, the operator's response to i
the resulting reactor trip was acceptable.
,
In the operations area the licensee also adequately responded to a
primary to secondary: leak during startup after a refueling outage.
Management was involved in determining the magnitude of the leak, i
which did not exceed Technical Specification limits; however, it i
appeared that enough evidence was available a few hours earlier
!
for the licensee to initiate the shutdown (paragraph 2.f).
l t
[
9401040339 931216 PDR ADOCK 05000369 O
PDR:.,; *,
,
,
In the operations area, the inspectors identified lack of
,
oversight regarding the reactor coolant unidentified leak rate
surveillance.
For leak rates.below about one gallon per minute,
the calculation was not very reliable and difficult to effectively trend due to data scatter. As a result the licensee has initiated a review of the calculation to enhance its ability to accurately determine and trend leakage.
(paragraph 4.b).
.
In the maintenance area, one violation was identified involving an inadequate QA examination of a Unit 1 Steam Generator tube plug
^
weld that was faulty. A primary to secondary tube leak resulted from this faulty weld and forced the unit to shutdown on October 5 (paragraph 6).
In the maintenance area licensee management demonstrated good recognition of potentially adverse activities affecting plant safety during a reactor coolant-flow instrumentation problem.
Management maintained good oversight during the replacement of
[
this failing flow transmitter (paragraph 5.b).
-
,
In the engineering area, it was considered that poor engineering judgement was exercised when the SSF diesel generator was returned to service in a degraded condition as a result of a fuel oil leakage problem. A contributing cause appeared to be lack of management involvement by the engineering supervisor. Once he became aware that the engine had been inappropriately returned to service, the correct decision was made to continue with the maintenance. (paragraph 5.a).
'
In the plant support area, the inspectors identified declining conditions regarding plant housekeeping and material condition
'
(paragraph 2.a).
,
!
!
,
E
v,-
,
.
_
.
REPORT DETAILS
1.
Persons Contacted Licensee Employees l
'
J. Boyle, Work Control Superintendent
- D. Bumgardner, Unit 1 Operations Manager i
T. Curtis, System Engineering Manager
- E. Geddie, Station Manager
- G. Gilbert, Safety Assurance Manager
- R. Hall, Engineering Manager B. Hamilton, Superintendent of Operations
'
- T. McHeekin, Site Vice President
- T. Pederson, Safety Review Supervisor
>
- R. Sharpe, Regulatory Compliance Manager
- D. Simmons, Work Control Technical Manager l
- B. Travis, Component Engineering Manager Other licensee employees contacted included craftsmen, technicians,
operators, mechanics, security force members, and office personnel.
NRC Resident Inspectors
,
G. Maxwell, SRI
,
- J. Zeiler, RI, Catawba Nuclear Station i
'
- Attended exit interview
Acronyms and abbreviations used throughout this report are listed in the last paragraph.
'
2.
Plant Operations (71707)
a.
Observations
The inspection staff reviewed plant operations during the report period to verify conformance with applicable regulatory requirements. Control room logs, shift supervisors' logs, shift
,
turnover records and equipment removal and restoration records
,
were routinely reviewed.
Interviews were conducted with plant l
operations, maintenance, chemistry, health physics, and
.;
performance personnel.
-
-
Activities within the control room were monitored'during shifts
,
and at shift changes. Actions and/or activities observed were conducted as prescribed in applicable station administrative
-;
directives.
!
Plant tours taken during the reporting period included, but were
,
not limited to, the turbine buildings, the auxiliary building, j
'
!
!
,
.
-
-
.
.
-
'
,
.
electrical equipment rooms, cable spreading rooms, and the station yard zone inside the protected area. During the plant tours, ongoing activities, housekeeping, fire protection, security, equipment status and radiation control practices were observed.
During a tour of the ECCS pump rooms the inspectors observed several minor material condition and housekeeping discrepancies.
The most notable of these were (1) missing bolts supporting the main electrical terminal box to the motor housing on the 1A and 2A Containment Spray pump motors, (2) scaffolding left in the 2B Containment Spray pump room when work was completed, and (3)
leftover hardware, paper, and a mop in the 2A Containment Spray
'
pump room. Additional material-condition items were found during
!
the ESF walkdown of the Unit 2 NI pump rooms and are discussed in paragraph 4.
The inspectors considered that these items, although minor in safety significance, indicated a lack of attention to the material condition of important plant equipment..The licensee initiated work orders to correct these discrepancies.
Based on discussions with a mechanical maintenance department general supervisor, the inspectors learned that a new maintenance program was initiated in August designating maintenance teams to
periodically inspect all areas of the plant.
b.
Unit 1 Operations Summary
,
Unit 1 began the report period in Mode 3 after being shut down on October 5 to repair a leaking steam generator tube plug. On October 18, unit startup commenced following completion of steam generator repairs. The unit reached full power operation the following day. On November 5, the reactor tripped on Overtemperature Delta-Temperature when the Turbine Governor and Intercept valves closed. These valves closed because of a
-
,
solenoid valve failure in the turbine overspeed protection system.
Details pertaining to this trip are contained in paragraph 2.d.
Reactor startup commenced November 6 and the unit was placed on-line that same day. The unit reached full power the following day and operated at full power for the remainder of the report period.
c.
Unit 2 Operations Summary Unit 2 began the report period operating at full power. 0n
November 7, reactor power was reduced to 88 percent because of a feedwater system transient that resulted from the failure of a
circuit board in the main feedwater pump speed control circuitry.
That same day the unit returned to full power following replacement of the circuit board. On November II, reactor power
'
was reduced to 15 percent to repair a leaking primary system valve 2WL-10 in the suction piping of the "A" reactor coolant drain tank l
pump.
Details pertaining to this-leak are contained in paragraph 2.e.
The unit returned to full power tM following day after the valve was repaired and remained at full power for the remainder of the report period.
f
.
e
_____ - _ _
__-______ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _
,L
d.
Unit 1 Reactor Trip on Overtemperature Delta-Temperature On November 6 Unit I was operating at full power. At approximately 2:20 a.m., the control rods began stepping into the core rapidly. The operators observed that the turbine governor and intercept valves had closed, thereby isolating steam from the turbine. Believing there was an instrument failure causing the rod movement, the operator at the controls took the rod control system to manual to scan the control board. After only a few seconds, and finding no indications of an instrument failure, the operator returned the rod control system to automatic.
Following the return to automatic control, the rods continued to step in rapidly. Simultaneously, the steam dump valves to the condenser and steam atmospheric valves began to open due to the rapid loss of steam loads.
Because the rod control system was unable to keep up with the loss of steam demand, the core average temperature and delta-temperature began to increase. Approximately 11 seconds after the start of the transient, the reactor tripped when the variable Overtemperature Delta-Temperature trip setpoint was reached. As expected the turbine tripped due to the reactor trip.
All control rods fully inserted into the core.
Both motor-driven auxiliary feedwater pumps auto-started as expected when Low-Low level was reached in one steam generator. During the transient one pressurizer PORV opened, but no primary or secondary safety valve lifted.
Decay heat was dissipated via the condenser steam dumps.
The inspectors responded to the site following notification of the reactor trip and observed actions to verify the cause of the trip, discussed the trip with control room personnel, and verified appropriate reporting of the event. This review revealed that plant safety systems responded as expected to the trip and operator actions associated with the event were acceptable. A minor operator response weakness was discussed later with the licensee involving the operator initially taking the rod control system to manual without first verifying whether or not there was an actual instrument failure causing the unexpected control rod movement. However, after reviewing the transient, the inspectors determined that a reactor trip would have occurred regardless.
l On the day of the trip, the licensee's initial investigation of the cause of the unexpected closing of the turbine governor and intercept valves revealed that a turbine control system relay may have inadvertently actuated. This relay, which senses if the generator output breakers are closed, causes the following if actuated:
1) closure of the turbine governor and intercept valves, 2) opening of the generator field breaker, and 3) start of the standby condenser hotwell and booster pumps. This relay actuation was suspected to be the initiating event after two loose wires in the circuitry were discovered.
--_
.o
.
4
,
On November 8, following return of the unit to full power, the i
inspectors reviewed in detail the turbine control system logic and
reactor trip data.
Based on the sequence-of-events data, both the condenser hotwell and booster pumps started and the generator field breaker opened approximately 8 seconds after the governor and intercept valves began to close.
If the initiating event had been the actuation of the turbine control system relay, then these three items should have occurred at the same time. The inspectors discussed this with engineering personnel who initiated further investigation of the closing of the governor and intercept valves.
The licensee subsequently determined that the failure of a trip solenoid valve in the turbine emergency trip circuitry, which was discovered while they were attempting to place the turbine on-line, may have caused the governor and intercept valves to close.
Based on a detailed failure analysis the licensee determined that the solenoid could fail open because of four previously unknown mechanisms allowing the emergency hydraulic oil to be dumped to drain, resulting in the governor and intercept valve closure.
Since the solenoid valve was replaced prior to startup, the
immediate problem was corrected. At the end of the report period, the licensee was pursuing corrective actions to help preclude
-
recurrence of this failure.
e.
Unit 2 Unidentified Leakage Increase Due to Valve Leakage On November 11, at 2:40 a.m., Unit 2 was operating at 100 percent power when the operators noticed an increase in their frequency of having to pump down the containment floor and equipment sump.
Suspecting an increase in leakage inside containment, a reactor coolant leakage measurement was performed, yielding an unidentified leakage rate of 0.966 gpm. A second leakage
-
measurement indicated in 0.989 gpm. At 7:07 a.m., when a third
leakage measurement yielded 1.173 gpm, the licensee entered the TS
,
Action requirement of 3.4.6.2.b.
This TS Action required that the i
leakage be reduced below the TS limit of 1 gpm within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> or that the unit be placed in Hot Shutdown within the following 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />.
Later, a leakage inspection of lower containment was
-
,
performed and valve 2WL-10, a manual diaphragm valve in the
!
suction of the 1A NCDT, was discovered leaking. At 9:40 a.m. the l
1A NCDT was isolated and the leakage through 2WL-10 was stopped.
l At 11:05 a.m., unidentified leakage was measured to be 0.382 gom.
l The TS Action was exited at 1:05 p.m. without the need to initiate i
a plant shutdown, following a confirmatory leakage measurement
,
that was within the TS limit.
,
To repair 2WL-10, power was reduced to 15 percent.
Repair
'
activities revealed that the cause of the valve leakage was loose bolts connecting the body to the bonnet. These bolts were
,
tightened and the valve was verified to be leak tight.
Four other diaphragm valves in the NCDT system were inspected. Although no other loose bolts were identified, a torque-pass on each valve was i
.s
.
L
performed as an added precaution. The unit returned to full power at 6:25 a.m. the next morning following the repairs to 2WL-10.
.
The inspectors monitored the licensee's activities in response to the increased unidentified leakage and determined that their actions were proper. The operators demonstrated that they were adequately monitoring and capable of detecting NC system leakage using indications other than the results of leakage measurements.
In addition, in response to the loose bolts found on 2WL-10, the
.
licensee plans to inspect similar valves on the other unit as well as review the possibility that diaphragm valves in other plant systems may have loose bolts.
f.
Review of Unit 1 Primary to Secondary Steam Generator Tube Leakage During Startup During this report period the inspectors reviewed the events l
leading to the licensee's discovery of the Unit 1 primary to secondary steam generator tube leak on October 5.
The following
!
sequence of events occurred according to plant logs, chemistry
.
data, and discussions with licensee personnel.
On October 4, prior to entering Mode 4, sampling for specific activity of the secondary coolant system was conducted. The
,
sample results indicated normal activity in the coolant. The y
following day, startup commenced and criticality was achieved at
!
4:47 a.m.
Around 10:00 a.m., with the unit at 3 percent power, a technician in the secondary chemistry lab noticed a slight increasing count rate in the CSAE exhaust radiation monitor (EMF-33).
EMF-33 monitors the gaseous activity in the condensate i
system. At that time, the count rate had not reached the alarm
^
setpoints. The technician discussed these readings with IAE and
,
operations personnel to determine if any ongoing maintenance could i
be causing the increase. No activity affecting the monitor was
,
'
identified. At 12:00 p.m., the chemistry manager arrived on-site and, after being briefed on the EMF-33 trends, directed further sampling of the steam generator and NC system.
,
i At 3:45 p.m., with power at 5 percent and increasing, analysis of
+
'
a gas sample from the CSAE was completed. The results of this analysis revealed that there were radioisotopes of Xenon present, i
explaining the increase in EMF-33 readings and confirming the existence of some primary-to-secondary tube leakage. An NC sample
)
was obtained for activity analysis at this time so that a leak i
rate could be calculated.
At 4:30 p.m., during the regularly scheduled outage meeting, the chemistry manager discussed with other management personnel the possibility of primary-to-secondary leakage. The licensee decided to continue power escalation until chemistry evaluations were completed.
i
!
.)
.
.
At 5:00 p.m., with the unit at 15 percent power, analyses of blowdown samples from each steam generator were completed. The results did not indicate the presence of the isotope Iodine 133.
Iodine 133 activity is used to calculate primary-to-secondary
'
leakage. At about the same time, the results of an NC system activity sample was completed.
By using the gas sample results and measuring the offgas flow, a chemistry technician calculated a primary to secondary leakrate of 468 GPD. The technician later
'
explained that the confidence in this leakrate was not good due to a problem in reading the offgas flowmeter. The flowmeter was pegged high when the reading was taken, indicating a higher than actual leakrate.
,
l At 5:30 p.m., with the unit at 16 percent power, readings from the
-
N-16 monitors, which detect Nitrogen 16 in the main steam lines, began to increase, and EMF-33 counts had increased to the alarm setpoint.
It was also reported that the problem with the pegged offgas flowmeter was corrected.
i At 7:00 p.m.,
a decision was made to hold reactor power at 22 percent until another set of chemistry samples could be collected and analyzed. The results of these samples were to be used to calculate a leakrate based on EHF-33 counts, NC system Xenon
!
activity, and the CSAE offgas flowrate.
Based on the results of
'
this sampling, a 410 GPD leakrate was calculated; the calculation did not exceed Technical Specification limits. The results of blowdown samples indicated that the leakage was from the "A" steam generator.
Licensee management concluded that, because of the
_
r transient conditions and low power level, the leakrate calculations were not reliable.
However, the actual leakrate clearly exceeded the administrative limit of 50 gpd. At 10:10 p.m., the unit commenced a shutdown to Mode 5 for steam generator repairs.
After reviewing the above events, the inspectors determined that the licensee had no evidence of leakage prior to commencing unit r
startup. However, the inspectors considered that there was sufficient data for the licensee to have concluded earlier that a
-
primary-to-secondary leak had occurred and that the leak exceeded
their 50 gpd administrative limit.
For instance, at around 5:30 p.m., a leakrate determination could have been performed since an NC system activity sample had recently been completed and the CSAE flowmeter problem had been corrected.
Further steam generator blowdown sampling around this time may have revealed
,
which steam generator had the leak.
As a result of the event, the licensee initiated several actions to improve their ability to identify primary to secondary steam generator tube leakage during unit startup and power escalation.
These actions revolved around closer monitoring of EMF-33 count
.
rate since this monitor gives a real-time indication of change in
,
primary to secondary leakage. The actions include (1) resetting
!
,
.
.
,
,
the EMF-33 alarm setpoints prior to unit restart, and (2)
suspending startup or power escalation if the EMF-33 count rate increases above 3 times the background.
In addition, e.hemistry personnel are investigating other techniques used throughout the industry to effectively detect leakage during startup.
'
g.
Review of Sequoyah Pod Control Step Demand Counter Problem During this report period the inspectors learned that'one of TVA's
.
Sequoyah units had experienced a problem with their control rod l
group step demand counters. While moving the control rods manually, a step counter began changing by factors of ten as opposed to single step changes.
Concerns were heightened when
several new replacement step counters experienced similar failures.
In accordance with their TSs, the reactor trip breakers were opened (the unit was in Mode 3) as a result of the discrepancy between the step counter and the actual rod positions.
l The inspectors reviewed the similarity between the Sequoyah and
,
McGuire step counters and TSs. The step counters used are
.
'
identical.
The counters are of a mechanical analog design and are manufactured by Whittaker Electronic Resources Division. The counters were bought and supplied by Westinghouse. The inspectors j
discussed the counter design and failure history with licensee
,
engineering personnel.
Step counter testing is conducted during l
each refueling outage. Although some counter failures, attributed t
to aging, wear, and dirt accumulation, have occurred, an excessive
,
number of failures have not been identified. Also, there have e
been no reported failures of new counters obtained directly from stock.
In addition, prior to replacement of any step counter, a
!
bench check is performed to ensure that they work properly. The
!
licensee indicated that they anticipate replacing their counters
-
with a digitel design during the next outage on each unit.
a The inspectors also reviewed the McGuire TS to determine what action must be taken in the event that a step counter fails with
!
the unit in Mode 3.
There are no TSs similar to Sequoyah that _
l require a plant shutdown in the event of the inoperability of. the l
a step counter in Mode 3.
i No violations or deviations were identified.
!
3.
Engineered Safety Feature System Walkdown (71710)
i
!
During this report period, the inspectors completed a detailed walkdown
,
of accessible portions of both trains of the Unit 2 NI System. Using the licensee's NI system lineup procedure, OP/2/A/6200/06, and the Unit 2 startup procedure, OP/2/A/6100/01, the inspectors verified that
-
outside containment main system flowpath valves and assorted system drain and vent valves were in their proper positions. Valve positions
,
were verified using both the control room board indication and local
!
valve position where possible.
Based on this review, the inspector i
,
.
,
,
determined that major system flowpath valves were in their proper position. A minor procedural discrepancy was identified with
,
OP/2/A/6200/06 involving an incorrect valve position. The valve lineup checklist specified valve 2NI-115B to be positioned closed.
2NI-115B is
[
the "A" NI pump miniflow valve and should be in the open position for proper operation of the pump. The inspector verified that the actual
valve position was open. The inspectors reviewed the last time this procedure was performed to determine why this procedure discrepancy was
not identified. When last performed, an earlier revision of the procedure was used and this revision accurately required the valve to be
'
in the open position.
Operations personnel initiated a procedural change to properly reflect the correct valve position. The inspectors also noted that the NI system lineup procedure (as well as the other ECCS lineup procedures) did not require the verification that pipe caps are properly installed.
Pipe caps are typically used downstream of all vent and drain valves located in piping arrangements and provide a second isolation to the vent and drain valves. Operations management
'
personnel indicated that there was an expectation that personnel performing the valve lineup procedures would ensure that pipe caps are installed.
During the system walkdown the inspectors examined manual and motor-operated valves to ensure that they were installed correctly and with no
bent stems, missing handwheels, or improper labeling. Open work requests on components in the system were examined to ensure that no
,
major maintenance, which could possibly affect the system's performance, had been performed.
Selected process instrumentation was examined to ensure proper installation and function. Where possible, local instrument readings were compared with control room indications. The i
inspectors also compared portions of the as-built system configurations against plant drawings to ensure that the as-built system reflected the
!
current system design.
No major discrepancies were noted from the above reviews; however, the inspectors observed several minor material condition problems.
The most notable items were (1) a section of the Auxiliary Building Ventilation duct in the 2B NI pump room was not properly supported in that a metal support that connected the duct to the wall had been removed, and (2) an excessive number of instrument piping connections and valves in the 2A NI pump room had boron buildup. Although no visible liquid leakage was observed at the time of the inspection, the boron buildup both on these components and on the floor indicated that there had been leakage previously.
The inspectors found no catchments installed to reduce the potential spread of contamination or evidence that Work Requests had
'
been initiated to repair the leaks. These items were discussed with licensee personnel, who indicated that work requests would be initiated to correct them.
No violations or deviations were identified.
i
]
(.
+
4.
Surveillance Testing (61726)
The resident inspectors reviewed and/or witnessed selected surveillance tests to assess the adequacy of procedures and performance as well as conformance with the applicable TS.
Selected tests were witnessed to verify that (1) approved procedures were available and in use, (2) test equipment in use was calibratW, (3)
test prerequisites were met, (4) system restoration was completed, and (5) acceptance criteria were met.
The selected tests listed below were reviewed or witnessed in detail:
a.
Accumulator Pressure and Level ACOTs On October 27 the inspectors witnessed surveillance testing performed in accordarme with Work Orders 93073718-01 and 93073717-
01. The procedures used were PT/2/8/4600/07, NI Cold Leg Accumulator Pressure, and PT/2/B/4600/06, NI Cold Leg Accumulator Level. These procedures were implemented to satisfy the monthly ACOTs in accordance with TS surveillance requirement 4.5.1.1.2.
The ACOTs verified that each cold leg injection accumulator water level and pressure channel was operable. The inspectors verified that (1) approved procedures of the correct revision were used and properly followed, and (2) test equipment was calibrated. The irspectors observed good communication between test and operations personnel. Test personnel discussed the test scope and expected unit response with operations personnel prior to the start of testing. Based on discussions with the technicians, t'
were knowledgeable of the procedure and equipment.
Testing a s accomplished without error and both the level and pressure channels were found to be within established test parameters.
b.
Review of Reactor Coolant Leakage Measurements During the report period, the inspectors began close monitoring of the licensee's Unit 2 unidentified reactor coolant system leakage results. This monitoring was considered necessary due to the degraded condition of valve 2NC-14, the manual isolation for letdown between the reactor coolant system and the CVCS.
_
'
External leakage was first discovered from 2NC-14 at the start of the July refueling outage.
During the outage and preparation for startup, leakage repairs were performed by injecting the valve with a sealant.
Following startup, external leakage in excess of
-
the 1 gpm TS limit developed from the valve forcing the unit to shutdown on September 27. While shutdown, the valve was injected with sealant and seal welds were installed in three locations
.
where leakage could occur. However, two pinhole leaks that developed in the weld area were discovered during preparations for restart of the unit. The valve was injected with sealant a final time at full system temperature and pressure.
When unit startup
s a
w
,
commenced on October 15, there was no evidence of external leakage. Since this valve was located inside containment in a
high radiation area, the licensee had no plans following startup to visually verify that leakage had not started.
Based on the problems experienced with 2NC-14,- the inspectors considered that
,
the potential existed for external leakage to develop during the operating cycle warranting increased management and operator awareness for NC leakage.
'
The licensee issued an Operation's Special Order (No. 93-17) on October 18 requiring that NC leakage measurements be conducted daily (as opposed to every 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />) as well as requiring
.
increased operator monitoring for signs of NC leakage. On
'
October 22, the inspectors noticed that the unidentified leakage was approximately 0.5 gpm, increasing almost 0.4 gpm over the
>
previous 3 days. The inspectors became concerned that neither operations nor engineering personnel were trending the leakage results from day to day and were not cognizant of or concerned
>
with the adverse trend. The operators indicated that it was not uncommon to see as much as a 0.2-to 0.3-gpm change from one
,
leakage measurement to the next.
'
Based on concerns over the amount of leakage measurement data scatter indicated by the operators, the inspectors reviewed the licensee's leakage measurement procedure PT/A/4150/01B, Reactor Coolant Leakage Calculation. The test method involved uses a water inventory balance around the NC and associated systems.
This is typical of such leakage calculations used in the industry.
However, the inspectors noted that the minimum leakage run time was only 30 minutes. Since the leakage measurement is based on
.
the difference in volumes at the beginning and endpoint states,
the accuracy of the program is highly dependent on the test
,
duration.
Programs that run longer tend to reduce the amount of
~
data scatter. This was discussed with management, and on October 29 the test duration was extended to a minimum of I hour.
On October 29, using PT/A/4150/01B, the licensee calculated a total Unit 2 leakage of 1.192 gpm and an unidentified leakage of 0.294 gpm. To verify the acceptability of the procedure and the constants used in it, the inspectors analyzed the same leakage data recorded. This analysis was conducted using the NRC's RCSLK9 computer program which is described in NUREG-1107, "RCSLK9; Reactor Coolant System Leak Rate Determination for PWRs." The i
results of this analysis were in good agreement with the
'
licensee's leakage results.
By the end of the report period, the licensee had initiated
-
several other good initiatives in addition to the increased i
leakage duration. The following were initiated: (1) efforts were
,
made to install a camera to monitor leakage from 2NC-14, (2)
enhancements to the leakage program statistical analysis were made, and (3) an Operation's Special Order implementing
>
,
.
.
.
_
(,'.
,
,
administrative limits to shutdown prior to unidentified leakage reaching the TS limit was issued.
,
The inspectors considered these actions to be useful in improving the licensee's ability to reliably monitor NC leakage.
No violations or deviations were identified.
5.
Maintenance Observations (62703)
l
Routine maintenance activities were reviewed and/or witnessed to assess procedural and performance adequacy and conformance with the applicable TS.
The selected activities witnessed were examined to verify that,
.'
where applicable, approved procedures were available and in use, prerequisites were met, equipment restoration was completed, and j
maintenance results were adequate.
-
The following maintenance activities were reviewed or witnessed in detail-I
a.
SSF Diesel Generator Lube Oil Problem On November 11 the inspectors began monitoring a maintenance problem with the SSF diesel generator involving diesel fuel leaking into the diesel's lube oil system.
Leakage was first discovered on November 9 during routine preventative maintenance.
Part of the leakage was from a lube oil check valve that was operating improperly. This valve was.
i replaced, but subsequent lube oil sample results indicated that fuel was still leaking. Since the sample results did not exceed the vendor's recommuided limit, an engineering representative responsible for the diesel recommended that it be returned to
.
service and further inspections and repairs be conducted later.
l The SSF was returned to service on November 10 at 6:05 p.m.
'
The following morning, when the engineering supervisor learned that the SSF had been returned to service with the fuel oil leak not corrected, a decision was made to remove the diesel.from service again and repair the leak. The repair work was conducted using Work Requests 93081228-01 and 93083063-01. The cause of the leakage was a degraded o-ring in a fuel injector.
Following i
replacement of all fuel oil. injectors and the completion of i
functional testing, the SSF was returned to service on November.14 l
at 4:01 a.m.
The inspectors witnessed portions.of the repair activity, noting that approved procedures were being used and work was being j
performed in accordance with these procedures. The inspectors
also discussed with engineering personnel the decision to return
'
the diesel to service initially without correcting the problem.
Based on these discussions, the inspectors considered the decision i
.
-
-
.
to return the diesel to service in the degraded condition to-reflect non-conservative judgment.
It was also apparent that the engineering supervisor was not involved with the initial decision, which under the circumstances warranted additional oversight.
b.
Unit 1 NC Flow Transmitter Replacement On November 17 operations personnel noticed that the "C" NC Loop flow Channel II was reading lower than the other two flow channels on that loop. Since this channel was still reading within 5 percent of the other two it was considered operable. Trouble-shooting conducted by IAE personnel revealed that the flow transmitter was failing. On November 18 a management meeting was held to discuss the problem which resulted in a decision to replace the transmitter on-line. Since all three NC flow channels for each loop share a common high side pressure tap, the licensee was concerned that a pressure perturbation in the instrument line while valving in the replacement transmitter could cause a reactor trip. A reactor trip world result from 2 out of 3 low NC flow signals.
The replacement activity was conducted using Work Request 93084103-01. The inspectors reviewed this work package and monitored the activity from the Control Room. Management demonstrated good recognition of the potential safety impact to the plant in that industry guidelines for coordinating an infrequent or high risk evolution were consulted. A detailed briefing was held to discuss the replacement'and the potential safety precautions. The inspectors observed good communication and coordination of the activity, which was accomplished without incident.
No violations or deviations were identified.
6.
Followup on Previous Inspection Findings (92701, 92702, and 71707)
(Closed) UNR 50-369/93-22-01:
Incomplete Weld for Steam Generator Tube Plug.
On October 5 Unit I was forced to shutdown from approximately 20 percent power due to a primary to secondary steam generator tube leak.
Subsequent inspection of the "A" steam generator revealed that the hot leg plug located in tubesheet position 39-72 was leaking. This plug was installed on September 9 after portions of this and one other tube were
!
re w ed for analysis.
Upon examination of the plug weld, the licensee
'
determined that it was faulty because there was insufficient weld overlap at the weld start and stop points. The poor quality of the weld was not identified during the post-weld remote visual QA examination.
'
In the previous inspection, the inspectors began reviewing the circumstances that led to the acceptance of this faulty weld. The two steam generator tube plugs were welded and visually examined in i
,
_ _ -
-
,
.~s
.
accordance with Work Request 93064384-01. The inspectors reviewed this maintenance work package, noting that the activities were properly controlled and performed by both a qualified welder and QA examiner.
The weld examination was performed remotely.using a. video recorder.
Remote visual inspection is allowed by ASME weld examination and acceptance standards; however, the remote inspection must be at least equivalent to that which is attainable by direct visual examination.
The resident inspectors reviewed the video recording made of the examination. The clarity and resolution of the recording was good and there was adequate illumination to conduct the examination'. The welder.
had made two weld passes around the plug and there was an obvious discontinuity in a section of the outermost weld pass.
The QA examination was inadequate in that the faulty weld was not identified and rejected.
10 CFR 50 Appendix B, Criterion X, Inspection, requires in part that a program for inspection of activities affecting quality be established and executed by or for the organization performing the activity to verify conformance with the documented instructions, procedures, and drawings for accomplishing the activity.
Performance of examinations,
.
measurements, or tests of material or products processed is required for each work operation as necessary to assure qushty.
Implicit in those
,
requirements is the requisite that the inspections be adequate to t.nsure quality.
This issue is considered a violation of 10 CFR 50 Appendix B, Criterion X, for failing to perform an adequate examination of the weld for tube plug 39-72.in the Unit 1
"A" steam generator on September 9, 1993. The Jefective weld led to a primary to secondary leak that was detected following unit startup on October 5, 1993. An Unresolved Item 50-369/93-22-01 is closed and will be identified as Violation 369/93-24-01:
Inadequate QA Examination of SG Tube Plug Weld.
,
One violation was identified.
,
7.
Information Meeting With Local Officials (94600)
During the weeks of November 1 and 8, the Senior Resident Inspector met with local elected officials from Mecklenburg and Lincoln counties. The inspector discussed the role and location of the Resident Inspector's Office. On November 12 the inspector toured the Lincoln County Emergency Preparedness Center. The inspector met the Lincoln County Emergency Preparedness personnel and discussed the role of the Site Resident Inspector as it relates to the actuation of the County
Emergency Preparedness Center.
_
_
'
I
_( 4 a:
8.
Exit Interview The inspection scope and findings identified below were summarized on November 22, 1993, with those persons indicated in paragraph 1.
The following items were discussed in detail:
Item Number Description and Reference VIO 369/93-24-01 Inadequate QA Examination of SG Tube Plug Weld (paragraph 6).
The licensee representatives present offered no dissenting comments, nor did they identify as proprietary any of the information reviewed by the inspectors during the course of their inspection.
9.
Acronyms and Abbreviations
,
,
ACOT Analog Channel Operational Test
-
ASME -
American Society of Mechanical Engineers CSAE -
Condensate Steam Air Ejector
!
CVCS -
Chemical and Volume Control System ECCS -
Emergency Core Cooling System ESF
-
Engineered Safety Features GPD
-
Gallons Per Day
GPM
-
Gallons Per Minute i
IAE
-
Instrumentation and Electrical i
LER
-
Licensee Event Report NC
-
NCDT -
Reactor Coolant Drain Tank NI
-
Safety Injection System Nuclear Regulatory Commission NRC
-
PORY -
Power-0perated Relief Valve QA
-
Quality Assurance
RI
-
Resident Inspector SG
-
-
Senior Resident Inspector
SSF Standby Shutdown Facility
-
TS
-
Technical Specification
,
-
Tennessee Valley Authority i
-
Unresolved item VIO
-
Violation l
!
.
O