ML20136G708
| ML20136G708 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 03/13/1997 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20136G677 | List: |
| References | |
| 50-361-97-02, 50-361-97-2, 50-362-97-02, 50-362-97-2, NUDOCS 9703180158 | |
| Download: ML20136G708 (33) | |
See also: IR 05000361/1997002
Text
. . - . _ . . _ .- . - _ _ _ . _ _ - . _ _ _ _ . . . . _- . _ _ . , _ . . . . . _ . _ _ _ . . _ . _ . _ m_
.
9
i .
i
! ENCLOSURE 2
'.
l U.S. NUCLEAR REGULATORY COMMISSION
j REGION IV
I
'
Docket Nos.: 50-361
50-362
- r
j License Nos.: NPF-10
Report No.: 50-361/97-02
50-362/97-02
Licensee: Southern California Edison Co.
Facility: San Onofre Nuclear Generating Station, Units 2 and 3
Location: 5000 S. Pacific Coast Hwy.
San Clemente, California
Dates: January 12 through February 22,1997
Inspectors: J. A. Sloan, Senior Resident inspector
J. G. Kramer, Resident inspector
J. J. Russell, Resident inspector
D. G. Acker, Senior Project Inspector
Approved By: D. F. Kirsch, Chief, Branch F
Division of Reactor Projects
i
Attachment: Supplemental Information
9703180158 970313
PDR ADOCK 05000361
G PDR
.
l
1
-2-
EXECUTIVE SUMMARY
San Onofre Nuclear Generating Station, Units 2 and 3 l
NRC Inspection Report 50-361/97-02:50-362/97-02
Operations
- A noncited minor violation was identified after the inspectors identified that access
panels in both trains of safety-related ventilation ducts for Class 1E switchgear
room cooling units were found not latched and ajar. The systems were determined
to be operable but degraded. Another example of a noncited violation was
identified after the licensee identified that the Train B control room emergency air
cleanup system (CREACUS) was degraded by an unexplained open access panel.
These conditions, although apparently having different causes, revealed an isolated
weakness in the configuration control of ventilation access panels and in the failure
to identify the discrepant conditions during operator rounds (Section 01.2).
- Plant equipment operators demonstrated excellent attention to detailin identifying
discolored oil in a Unit 2 low pressure safety injection (LPSI) pump, later determined
to have been caused by debris in the oil. Operations management acted
conservatively in declaring the pump inoperable (Section 01.3).
- One example of a violation was identified when operators inadvertently transferred
approximately 1000 gallons of water from the refueling water storage tank (RWST)
to the reactor coolant system (RCS), while Unit 2 was in Mode 5, due to a f ailure to
follow a procedure. This represents an isolated instance of lost configuration
control during the execution of the procedure (Section 01.3).
- A second example of a violation was identified when the inspectors observed that
RCS pressure was greater than allowed by the procedure in effect. Operators were
not aware of the pressure limit specified in the procedure, although there was no
safety consequence associated with the actual pressure. A contributory cause of
this situation was a weakness in the thoroughness of turnover from the previous
shift (Section 01.4).
- A control operator (CO) was not aware of a change in control element
assembly (CEA) position that resulted in the upper electrical limit (UEL) lights not
being illuminated. He was also not aware that the lights were not lit. This situation
resulted from weakness in the thoroughness of the log review and board walkdown
performed during the shif t turnover (Section 01.5).
- Performance of the draining of the RCS to midloop, near the end of the outage, was
well controlled, with operators displaying good cross-checking of component
manipulations and frequently correlating levelindications. Supervisory oversight
was continuous and effective, twice correcting inappropriate conditions.
Communications were effective but were not always as formal as expected by
licensee management and the tailboard briefing prior to the evolution was weak in
l
'l
l
1
. i
l
i
l
l
-3 1
that it f ailed to address several of the topics recommended in the procedure for
draining the RCS (Section 01.6).
Operations personnel demonstrated good overall performance during an inservice
test af a high pressure safety injection (HPSI) pump (Section M1.3).
Maintenance
- The prejob briefing prior to the inservice test of a HPSI pump was excellent
(Section M1.3).
- Maintenance demonstrated careful control of the installation of the upper guide l
structure (UGS) in the Unit 2 reactor vessel (Section M1.4).
- Instrumentation and Control (l&C) technicians did not raise inspector-identified
procedural deficiencies to the attention of l&C supervision until prompted by the
inspectors during the seismic trigger calibration execution (Section M1.5).
l
- The licensee's plant preservation efforts have resulted in significantly improved i
appearance of many plant areas, and substantially contributed to improved j
identification and correction of minor material deficiencies (Section M2.1). '
- A noncited violation was identified after the licensee determined that an air-operated
containment isolation valve would not close under design basis conditions. The
vendor had used incorrect information in determination of the required actuator 1
setpoint, and the licensee valve reassembly procedure was deficient (Section M8.1).
- Enaineerina
- System engineer support of the preparations for, and inservice test of, the HPSI
pump was effective (Section M1.3).
- The licensee implemented an air-operated valve (AOV) program, and was proactive
in identifying that a containment isolation valve would not close under design basis
conditions (Section M8.1)
- The root cause assessment of a failed RCS thermowell was extremely thorough and
represented substantial technical depth (Section E1.1).
- Engineering's evaluation and support of the steam generator (SG) chemical cleaning
process were thorough and excellent. Actions in response to the in-process
corrosion indications were conservative. Engineering predictions for the amount of
internal SG corrosion products that would be removed by chemical cleaning were
accurate (Section E1.2).
- The inspectors identified that periodic independent inventories of the components in
the reactor core and spent fuel pool were not being performed during the core
1
l
l
. !
-4- 1
l
reload, as described in the Updated Final Safety Analysis Report (UFSAR)
(Section E1.3). !
1
- The licensee's Independent Safety Engineering Group (ISEG), in conjunction with
'
Nuclear Engineering Design and Station Technical, demonstrated aggressive and
)
i
prompt response to two notifications of generic problems (Section E2.2).
- A safety evaluation performed to determine the acceptability of leaving some debris
in the reactor vessel was rigorous (Section E2.3).
- The licensee's evaluation of SG eddy current testing (ECT) results, and associated
corrective action, was thorough (Section E2.4).
l
- The inspectors identified a violation when contract personnel did not follow the
procedure specified hold time for in-situ pressure testing of Unit 2 SG tubes. The
- prejob briefing was inadequate in that the pressure hold times were not discussed,
, contributing to the violation. However, the test personnel used good j
communications and tube verification techniques (Section E4.1). '
- The licensee's engineering analysis, and testing done at a contracted laboratory, l'
were aggressive in identifying and minimizing irregularities with the containment
high range radiation monitors (HRRMs) (Section E8.2).
1
- A noncited violation was identified by the licensee related to deficiencies in the
completeness of the reactor coolant pump (RCP) motor lube oilleakage collection
system (Section E8.3).
Plant Sucoort
- Health Physics coverage of the installation of the UGS in the Unit 2 reactor vessel
was excellent, and technicians provided routine updates of the radiation levels to
the maintenance personnel (Section M1.4).
1
- The inspectors identified loose surf ace contamination, due to built up boric acid
resulting from a leak, exceeding the licensee's posting limits, that was not posted
as a contamination area. The time duration required to accomplish the boric acid
buildup indicated that Health Physics was not sufficiently timely or thorough in
identifying and posting such conditions (Section R1.1).
.
4
4
._
,
l - j
l
.
I
- Report Details
l
i 1
'
Summary of Plant Status l
l
Unit 2 began this inspection period in Mode 5 during the Cycle 9 refueling outage. On
February 19,1997, the unit entered Mode 4 in preparation for startup. However, a SG l
j
tube leak of approximately 3 gpm necessitated returning to Mode 5 later the same day for i
'
repairs. The unit operated in Mode 5 for the remainder of this inspection period.
Unit 3 operated at essentially full power throughout this inspection period.
l
1. Operations l
01 Conduct of Operations
01.1 General Comments (71707)
Operations during this inspection period were generally characterized by
conservative management oversight. Management response to low-safety
consequence surveillance testing discrepancies and various outage issues was
effective and conservative.
01.2 Safetv-Related Ventilation Ductina Access Panels Found Ooen - Unit 3
a. Inspection Scope (71707)
On January 15,1997, during a tour of the Unit 3 Class 1E switchgear rooms, the
inspectors observed access panels for safety-related heating, ventilation and air
conditioning (HVAC) ducting on the suction of Class 1E switchgear room cooling
units uniatched and ajar. On January 28,1997, Operations personnel observed an
access panel for safety-related Train B CREACUS Emergency Cooling
Unit (ECU) E419 untatched and ajar. The inspectors inspected the circumstances
surrounding these observations. The inspectors reviewed HVAC Drawing 40088,
Revision 3, Procedure SO23-1-5.1, Revision 2, " Auxiliary Building Emergency HVAC
Cooling Unit Operation," and interviewed cognizant Station Technical and
Operations personnel. The inspectors also reviewed Unit 3 Technical
Specification (TS) Surveillance Requirement (SR) 3.7.10.2 and portions of
Procedure SO23-3-3.12, Issue 2, Revision 12, " Integrated Engineered Safety
Features (ESF) System Refueling Tests," which the licensee used to meet the SR for
the ducting found open by the inspector. The inspectors also reviewed Action
Request (AR) 970101682, written when the licensee discovered the CREACUS
panel open, and CREACUS Air Flow Diagram 40096-13.
b. Observations and Findinas
On January 15,1997, the inspectors observed two hinged access doors on the
ventilation intake, adjacent to Train A ECU 255, unlatched and ajar. The doors
measured approximately 2 feet by 4 feet. Tne inspectors also observed one similar
access door unlatched and ajar on Train B ECU 257. The access doors were
- --- ~ _ _ . . - . _ - ~ . . .. - - . .- - ---
.
l
.
-2-
located on either side of suction filters,immediately adjacent to the suction of the
cooling units. The ducting itself measured approximately 4 feet by 4 feet. The
units provided emergency cooling to Class 1E switchgear and distribution Rooms
308A,310A, and 310A (Train A) and Rooms 302A,310A,310C, and Room 311
(Train B). These ECUS were used to cool the switchgear roorns under accident
conditions, and were cooled themselves by emergency chilled water. On
January 15,1997, the common chilled water system was also powered from Unit 3
switchgear, which would have had space cooling from these ECUS. These ECUS
l
were required to be tested for proper actuation by Unit 3 TS SR 3.7.10.2 every
24 months.
l The system alignment for the auxiliary building emergency HVAC system, of which
I
these coolers were a part, contained electrical and switch lineups, but required no
check of the configuration of access doors in the ducting. The surveillance
performed also did not require a check of access door configuration, but did check
that the control room indication of ECU start was satisfactory.
In order to investigate the reason the panels were not latched, the licensee started
the ECU units on January 15,1997. Operations and Station Technical personnel
noticed that these access doors were closed, due to the high differential pressure
between the room and suction during unit operation, and that vibration and loose
latches caused the latching dogs to disengage. Consequently, the doors became
ajar when the unit was shut down. Prior to January 15,1997, the ECUS were last
operated on January 9,1997 (Train A), and December 26,1996 (Train B). Based
on the above, the inspectors concluded that the ECUS remained operable with the
access panels ajar as they were found. However, the inspectors also found that the
ECUS were degraded, because consistent closure during unit operation could not be
assured. If the doors did not close, there would be significant air flow from the
space the unit was located in, and less than the design air flow from the spaces
'that were to be cooled. Even though the inspectors could determine the last time
the ECUS were operated, the inspectors could not determine how long the doors
had been ajar, because the licensee did not verify the position of these doors before
or after unit operation, and did not appear to be sensitive to the position of the
doors during operator rounds or Station Technical walkdowns.
In response to the inspectors' observations, the licensee generated a maintenance
order to tighten the latching lugs on the access panels and was evaluating
modifying the design of the latches at the end of this inspection period.
10 CFR Part 50, Appendix B, Criterion V, states, in part, that activities affecting
quality shall be accomplished in accordance with applicable drawings. Licensee
HVAC Drawing 40088-3 shows the suction flow of air to ECU 255 from the spaces
described above, and from a louver adjacent to the cooling unit, through intact
ducting. The same drawing shows the suction flow of air to ECU 257 from the
spaces described above, and from a louver adjacent to the cooling unit, through
intact ducting. Contrary to this, the inspectors identified the access panels
- - ._ - - = . -
.
.
A
, .
i
-3-
2
described above not secured closed (open) on the suction ducting of ECU 255 and
ECU 257, The inspectors considered the guidance of Section 6.3.1.1 of the NRC
Enforcement Manualin assessing the significance of this issue. This failure
constitutes a violation of minor significance and is being treated as a noncited
violation, consistent with Section IV of the NRC Enforcement Policy (Violation
362/97002-01).
On January 28,1997, Operations personnel found another hinged access panel 6 to
j 8 inches open. The panel was for Train B CREACUS ECU E419, and was similar
'
but had larger dogging devices. No maintenance activities were in progress, and
the licensee closed the panel. This panel was one of four access panels on the
charcoal unit and measured approximately 6 feet by 2 feet. The CREACUS
ventilation units provided filtration and cooling for the control room habitat in the
event of detection of toxic gas or radioactivity in the control room environment.
The licensee conducted extensive investigations to determine the circumstances
which resulted in this situation, but was not able to establish the reasons.
On February 21,1997, Station Technical and Operations personnel started the
Train B CREACUS unit with the door, previously found open 8 inches, open to test
system response. The door was forced shut by negative system pressure when the
unit was started, with a small amount of air leakage past the seal. Based on this,
the inspector concluded the unit remained operable with the door ajar, but that it
was degraded because door closure could not be consistently assured. Licensee
HVAC Drawing 40096-13 shows the access panels for the charcoal unit all shut.
The NRC considers that the design drawing clearly demonstrates the designer's
intent to provide for a closed control room emergency ventilation system with
minimum in-leakage from an uncontrolled atmosphere.
10 CFR Part 50, Appendix B, Criterion V, states, in part, that activities affecting
quality shall be accomplished in accordance with applicable drawings. The failure to
maintain the door closed on CREACUS Train B is another example of a noncited
violation of 10 CFR Part 50, Appendix B, Criterion V
(Violation 361;362/97002-01).
c. Conclusions
The inspectors identified a minor noncited violation in that Class 1E switchgear
ECUS E255 and E257 were degraded, but operable, due to poor material condition
- which was not identified by licensee personnel. A similar licensee-identified
condition existed for a CREACUS cooling unit and is considered a violation of design
drawing requirements.
01.3 Inadvertent Transfer of Water to the RCS in Mode 5 - Unit 2
a. Jntspection
r Scope (71707)
4
On January 29,1997, the inspectors were informed by the Operations manager
that, due to operating a valve out of sequence, while making containment
.
.
4
spray (CS) Pump 2P012 available for shutdown cooling, about 1000 gallons of
water from the RWST had been inadvertently transferred to the Unit 2 RCS. The
inspectors interviewed Operations supervisory personnel, the two equipment
operators, and the unit CO involved. The inspectors reviewed
Procedure SO23-3-2.6, Revision 10, Attachment 15, " Aligning CS Pumps as
Standby Shutdown Cooling (SDC) Pumps;" Procedure SO123-0-23, Revision 3,
" Control of System Alignments;" and control room logs and records of RCS and 3
RWST parameters for the time of the occurrence. I
b. Observations and Findinas
On January 28,1997, while on normal rounds, Unit 2 equipment operators
observed that the oil in LPSI Pump 2P015 appeared to be an improper color. The
cognizant engineer later confirmed this finding and observed some debris in the oil.
LPSI Pump 2P015 was operating as the SDC pump, with Unit 2 in Mode 5 at about
91 * F and about 35 percent of pressurizer level. The RCS was vented to ,
containment atmosphere through an open pressurizer manway. SG primary side j
nozzle dams were in place. Operations management decided to conservatively j
declare Pump 2P015 (the Train A pump) inoperable, even though the pump had {
been operating for SDC with no abnormalities in bearing temperatures or motor i
amperage. Operations personnel then recognized the need to align a CS pump as l
the standby pump for SDC, and then to transfer SDC to Train B LPSI Pump 2P016,
in order to comply with TS 3.4.7, which required two operational (including one
operating) SDC trains.
Night shift operators tailboarded the evolution, and knew that CS Pump 2P012
needed to be vented. During the venting evolution, Suction Valve 2MUO62 for the
pump, from the RWST, was to be opened, and the pump casing vented. The j
suction valve was then to be closed and discharge Valve 2MU994, from the pump
to the suction of the LPSI pumps, was to be opened. During the tailboard,
Operations personnel noted that having both these valves opened simultaneously
would provide a flowpath from the RWST to the suction of the operating LPSI pump
(2P015), causing an addition of water to the RCS. Both these valves were operated
locally with no control room indication for valve position available.
i
One equipment operator opened Valve 2MUO62 and vented the CS pump. He then
closed Valve 2MUO62 about three-quarters of the way, and went to another room
where a second equipment operator was standing by. He reported to the second
equipment operator that a satisfactory vent had been performed. The second
operator then opened Valse 2MU994. When Valve 2MU994 was opened, control
room operators received an alarm for pressurizer level deviating from setpoint.
Control room operators paged the local operators and directed them to fully shut i
Valve 2MUO62, which stopped the water flow. Pressurizer level rose approximately
10 percent, to about 45 percent, indicating that about 1000 gallons of water had
been transferred in about 3 minutes. RCS temperature lowered about one degree,
to approximately 90* F, and RCS pressure remained stable.
-- - - - - -- . - - ... - - - . - - - - - - - - _ - - - -
I
.
j
'
.
!
-5-
3
The inspectors determined that the primary root cause of the inadvertent water
, addition was failure to follow Procedure SO23-3-2.6. Step 2.1.8. of Attachment 15 I
to this procedure was to lock closed Valve 2MUO62. Step 2.3.2 was to lock open
- - Valve 2MU994. These steps were to have been performed in sequence. ;
3 Equipment operators failed to close Valve 2MUO62 prior to opening Valve 2MU994,
- contrary to procedure, causing the inadvertent water addition.
'
Procedure SO23-3-2.6is applicable to Regulatory Guide 1.33, Revision 2. Unit 2
>
TS 5.5.1.1 requires that procedures applicable to Regulatory Guide 1.33 be l
i implemented. The failure to follow Procedure SO23-3-2.6is the first example of a
violation of TS 5.5.1.1 (Violation 361/97002-03). I
i l
l The inspectors observed that the preevolution tailboard was thorough, and that all i
'
operators seemed to understand their duties, their effect on the plant, and possible l
misoperations and their effects. Operator workload, shift manning, training, end
> physical condition did not appear to be factors in the occurrence. I
! Control room operators gave Attachment 15 to the equipment operator and directed
i him to call before he operated the first component and then to perform the
! attachment up to, and including, breaker alignments. No hold points or other
j communication controls, until the attachment was completed, were in place. This j
- was routine guidance for equipment operators performing valve or breaker !
i alignments, even if the alignments could affect RCS status. The licensee relied )
l heavily on thorough tailboards to ensure no errors were made, and generally did not 1
i control evolutions outside the control room in a manner other than was done in this
- case. The control room operators generally did not issue orders in more of a
valve-by-valve sequence.
'
Procedure SO123-0-23 provided for, although did not require, the use of
,
independent verification as a hold point for more critical valve alignments. This
form of second check was not used in this case, and was generally used only when
>
verifying operability of a component prior to taking the opposite train component
, out of service. Normally, independent checks were not required prior to performing
! subsequent steps in a procedure. Consequently, the use of independent verification
- to ensure steps were performed in a certain sequence, although provided for
programmatically, was seldom used.
- Communications between the two equipment operators were incomplete. Both
knew that the closing of Valve 2MUO62 was important prior to opening
- Valve 2MU994. However, no common understanding was verbally established
between the two operators as to the status of this valve. The operator assigned to
open Valve 2MU994 assumed that Valve 2MUO62 was closed, based on the report
of a satisf actory pump vent. Therefore, he assumed that it was acceptable to open
,
Valve 2MU994. Although this assumption would have been valid based on
i discussions of the proper sequence of steps during tanooards, more complete
- communications would have confirmed the status of Valve 2MUO62.
j The inspectors concluded that the safety consequence and significance of this
event was minimal. SG nozzle dams were rated at 20 psig nominal pressure, from
l
i
i
l
. - -- . _ . - - ~ . . - . . . . . - . . - . - - . - - - - - . - - - . - - - - .
j' .
t
'
l-
,
1 -6-
i
'
the RCS loop, and for a 50 psig pressure excursion. The dams could withstand the
50 psig without leakage, but may have incurred some damage. The open
- pressurizer manway was at the top of the pressurizer, and, if the pressurizer had
l completely filled, there would have been about 50 feet of static head of water at
the nozzle dam level. This condition would have resulted in about 23 psig against
'
- - the nozzle dams, well within their one-time rating. Consequently, the inspectors
- found that there was no credible danger of causing excessive leakage from the
i nozzle dams. Licensee inspection of the nozzle dams after the incident did not
reveal any damage. Reactor coolant could have flowed out the top of the
! pressurizer, via the manway,if the operators had delayed any action for
! approximately another 17 minutes.
c. Conclusions
A plant equipment operator demonstrated excellent attention to detailin observing
discolored oilin LPSI Pump 2P015, and Operations management acted
conservatively in declaring the pump inoperable.
A violation was identified for f ailing to follow Procedure SO23-3-2.6, resulting in
inadvertently transferring about 1000 gallons of water from the RWST to the RCS
during Mode 5 operations. Weaknesses in control of the evolution from the control
room and communications between equipment operators were identified. Safety
consequences were minimal.
01.4 RCS Pressure Not Maintained Within Procedural Limits - Unit 2
a. Insoection Scope (71707)
On February 13,1997, the inspectors identified that Unit 2 (in Mode 5) control
room operators were not maintaining RCS pressure within the limits established in
the procedure in effect. The inspectors reviewed the procedure in effect,
interviewed operators, and reviewed plant data from the plant monitoring system
and the critical functions monitoring system along with control board indications.
b. Observations and Findinas
On February 13,1997, Unit 2 operators were filling safety injection tanks (SITS)
using Procedure SO23-3-2.7.1, Revision 4, Attachment 10, Procedure Modification
Permit 1, " SIT Operation." The procedure had been modified using the procedure
modification permit process to use the operating LPSI pump and a HPSI pump to fill
the SITS one at a time. Step 1.2 directed the operators to maintain RCS pressure
less than 350 psia.
At approximately 9:50 a.m., the inspectors noted control room indications of
pressure were 350 psia. The pressure control band being used by the operators
was 340 to 360 psia. The inspectors determined, as a result of discussions with
the operators, that the operators were not aware of the procedurallimit. Operators
.
.
-7-
initially reset the pressure band to 340 to 349 psia, and later changed the procedure
to increase the allowable pressure.
A review of RCS Pressure Instruments P104 and P103 (P104 was being monitored
by cperators using the plant monitoring system), HPSI flow, and pressurizer level
indications showed that the evolution had commenced at around 5:30 a.m. that
morning. The RCS pressure, since the day shift assumed the watch, was between
349 and 353 psia, with spikes to 355 psia when operators secured from filling
SIT 2T008.
The safety consequence was minimal because the basis of the upper pressure limit
was to avoid lifting a low temperature / overpressure relief valve which lifted at
392 i10 psia. Therefore, the more restrictive upper limit had no operational or
safety basis. However, the operating crew demonstrated a poor familiarity with the
operating limits in effect, in that the operators did not know the pressure limitation
specified in the procedure. Because pressure was allowed to rise af ter the day shift
assumed the watch, the inspectors were concerned that the turnover, specifically
the review of evolutions in progress, may not have been sufficiently complete and
thorough. The licensee's evaluation of this situation determined that the shift
turnover was not sufficiently complete and failed to adequately cover the rnore
restrictive pressure limit. The failure to maintain pressure in accordance with
procedure requirements constitutes the second example of a violation of TS 5.5.1.1
(Violation 361/97002-03).
Licensee corrective actions were to change the SIT operation procedure to raise
allowable pressure to 364 psia during future SIT fill evolutions, and to brief all
operating crews on the incident. Operations management also discussed the
incident with the crew directly involved.
c. Conclusions
Unit 2 operators demonstrated poor familiarity with an operating limit in effect, and
violated the limit. Incomplete shift turnover contributed to this error. The safety
consequences were minimal.
01.5 Ooerator Knowledae of Indications - Unit 3
a. Insoection Scope (71707)
On February 3,1997, the inspectors performed a routine control board walkdown,
observed CEA UEL light variations, and discussed the variations with the control
room staff.
b. Observations and Findinas
The inspectors observed that the UEL lights for CEAs 23,60, and 79, were not
illuminated. The inspectors informed the CO of the discrepancy. The CO was not
aware that the CEA UEL lights were not illuminated. The CO mechanically agitated
-
'
1
.
-8- I
l
the lights for the CEAs, and the UEL light for CEA 60 illuminated. The CO then i
performed a lamp test of the panel and all lights illuminated. The CO eliminated a l
'
blown bulb as the cause of the light not being illuminated. The CO indicated that
the light discrepancy would be discussed with the control room supervisor. The
,
control room supervisor subsequently informed the inspectors that CEAs 23 and 79 l
were inserted one step as part of the troubleshooting activities associated with i
'
Control Element Assembly Calculator 2 and, therefore, the UEL light would not be
illuminated, i
l
The inspectors discussed the CO's lack of knowledge of the CEA positions with the
4 shift superintendent (SS). The SS noted that the position of the CEAs was not
documented in the shift relief status sheet; however, the reason for the CEA
position deviation was documented in the long term condition section of the station
log book reviewed by the CO as part of shift turnover. The SS indicated that the
status of the CEAs would become part of the CO turnover sheet to further increase
the operators' awareness of the abnormal condition. The SS discussed the situation
with the CO. The inspectors reviewed the subsequent turnover sheets and noted a ;
reference to the CEA positioning. l
c. Conclusions l
1
i
The CO responsible for Unit 3 demonstrated weakness in attention to detail during ,
his control room board walkdown and failed to recognize an abnormal CEA l
alignment. Further, the CO did not adequately review the CO log book that
4
documented the CEA deviations. l
01.6 RCS Drain to Midlooo - Unit 2
.
a. Insoection Scope (71707)
The inspectors observed the operators perform Procedure SO23-3-1.8, Revision 10,
" Draining the RCS," to a midloop condition,
b. Observations and Findinas
On February 7,1997, operators drained the RCS to a midloop condition. The
inspectors identified that the prejob briefing did not include several aspects of the
performance of the evolution, as recommended in the Procedure SO23-3-1.8 In
, addition, operators at times did not employ 2-way or 3-way communications, but
they were effective in conveying the required messages.
The inspectors discussed the observation with the plant superintendent. The plant
superintendent agreed with the inspectors' observations that the prejob briefing was
less effective than previously observed, and that the operators' communications did
not fully meet management expectations.
The inspectors observed two instances in which the team concept improved the
performance of the evolution. The operations superintendent noted that the display
~ .
. --
,
-
l
t
.
i .g.
- on refueling water Level Indication 2Ll1520W was indicating the alarm setpoint
- instead of the actuallevel. The CO directing the evolution had not recognized the ;
discrepancy in the display, and corrected the indication. The CO directed an extra l
reactor operator assigned to the shift to enable the early warning heated-junction
thermocouple alarm before the procedure directed that the evolution should be
'
performed. The SS recognized the step should not be performed and stopped the
task from being performed.
The inspectors observed several strengths during the evolution. The control room
operators used good cross-checking during component manipulations. Operators
performed frequent level correlations and volumetric calculations of inventory
drained. Shift supervision and operations management provided continuous
coverage of the evolution.
I
c. Conclusions l
Operators successfully drained the RCS to a midloop condition with consistent shift
supervision present, good cross-checking of component manipulations, and frequent
level correlations. Weaknesses were identified in the prejob briefing; however,
communications, although adequate, did not meet management expectations.
1
08 Miscellaneous Operations issues (92700,92901)
08.1 LOpen) Unresolved item 361:362/96018-02: Technical Specification Improvement
Program (TSIP) SR implementation / emergency diesel generator (EDG) surveillance
tests not performed.
a. Inspection Scope
The inspector reviewed available information related to surveillance tests performed
to meet SR 3.8.1.9.c (EDG single load reject). During the course of this inspection,
other surveillance deficiencies were identified by the licensee. The inspector
incorporated those additional issues in the review of this unresolved item.
b. Observations and Findinas
The licensee determined that there was no surveillance test of record demonstrating
that the requirements of SR 3.8.1.9.c were satisfied for EDG 2G002, from the time
of TSIP implementation (August 5,1996) until January 6,1997. The SR required
that the EDG frequency be less than or equal to 61.2 Hz within 4 seconds after
rejecting a single load of at least 682 kW. The previous version of the TS did not
include that requirement, and the data from the previous performance of the single
load reject test showed that the frequency after 4 seconds was approximately
61.75 Hz. The cause of the high frequency was that the electronic governor speed
setting was high in order to load the EDG onto the grid, and that when the EDG bus
was separated from the grid, the EDG frequency increased to the high speed
setting. The licensee determined that, whenever the EDG was initially loaded to
greater than 1600 kW on the grid, the speed setting would be greater than 61.2 Hz.
- - - . _ - - - . . - . _ . - - . - . - _ _ . - . - _ . - _ - - . - -
-
i
i
.
! .
4
- 10-
l
The licensee performed a test on January 6,1997, with the initial load less than
} 1600 kW, and satisfied the frequency criteria of SR 3.8.1.9.c.
j The licensee reviewed other EDG SRs contained in the TSIP and determined that
other requirements were also not met by the most recent tests. SR 3.8.1.14
j (24-hour run) and SR 3.8.1.15 (hot restart) were not met by previous tests in that
l the loading in the pre-TSIP tests was greater than allowed in the TSIP SRs. While
the previous tests were determined to be a more rigorous test of the EDGs, the
-
tests did not comply with the restrictions of the current SRs. On January 12, ,
i 1997, NRR granted the licensee a Notice of Enforcement Discretion (NOED) for l
. SR 3.0.3 for these two SRs, extending the 24-hour time for completing missed )
I surveillances until NRC approved an exigent TS amendment deferring the SRs until
the next Unit 3 refueling outage. The licensee submitted the TS amendment i
4 request on January 14,1997. l
i
j The licensee also determined that compliance with SR 3.8.1.10 (full load rejection)
! was not demonstrated by existing test records, also because the tests had been
! performed at an initial load greater than now allowed by the TS. The licensee
l entered SR 3.0.3 and satisf actorily completed tests in accordance with the revised
j requirements for the Unit 3 EDGs.
1
'
At the request of the NRC, the licensee submitted a letter to the NRC on
l January 12,1997, documenting the licensee's interpretation of the initial loading
! requirements of SR 3.8.1.9 and how the current EDG testing satisfies this SR.
i
l The licensee determined that SR 3.8.1.8 (verification of automatic and manual
l transfer of AC power sources from the normal offsite circuit to each alternate ;
[ required offsite circuit) had not been performed for the unit auxiliary transformer
'
{ breakers. The licensee determined that the safety busses were operable as long as
] they were connected to the reserve auxiliary transformers (normal configuration) l
! and not connected to the unit auxiliary transformers. This position was confirmed
'
during a conference call between the licensee and NRC personnel on January 16, i
'
j 1997.
As a result of the discrepancies identified in the licensee's limited review of
l surveillances, the licensee initiated a review of all SRs to confirm that current SRs
{ were properly incorporated into surveillance procedures and that records of tests
i demonstrated strict compliance with the requirements.
'
The inspector interviewed licensee personnel involved in the preimplementation
j review of the TSIP SRs for the EDG. The licensee's TSIP implementation
coordinator assigned the divisions responsible for accomplishing the surveillances
.
the responsibility of reviewing the applicable SRs. For the SRs discussed above,
! Operations was principally responsible. Operations reviewed the test of record for
the single load reject test and incorrectly determined that the 4-second frequency
4 criterion was satisfied for EDG 2G002. Operations personnel met periodically with
l engineering and licensing personnel to discuss discrepancies and other questions
- associated with TSIP implementation. However, the apparent consensus was that
l
.
l
- _- - ,- , ,
1
.
.
-11-
the new requirements were bound by the old requirements, as far as the several
EDG loading initial conditions were concerned. This logic was sound from a safety
and function perspective, but was flawed from a compliance perspective, since the
old and new criteria were mutually exclusive.
On February 4,1997, the licensee determined that existing Unit 3 surveillances did
not fully satisfy SR 3.1.5.4 (CEA reed switch position transmitter surveillance), and
that previous tests in Unit 2 did not satisfy SR 3.1.5.4. On February 5,1997, a
NOED was verbally granted during a conference call between the licensee and NRC
personnel. The licensee submitted a TS amendment request on February 7,1997,
addressing this issue.
1
On February 14,1997, the licensee identified that the actuation of the "K" relays ;
had not been properly timed and included in the overall response time for the ESF, '
as required by SR 3.3.5.6. Conference calls were conducted between the licensee
and NRC personnel on February 14 and February 15,1997, during which the
licensee requested approval of a NOED for the relays in Unit 3. In 1993, the
licensee had determined that similar relays actuated in an average of about 150
msec. The licensee calculated that the three standard deviation value of 300 msec
was reasonable and encompassed the expected performance of all similar relays, l
and subsequently added the 300 msec value to the time response of the rest of the
ESF system actuation for each ESF component (measured at the specified
surveillance interval) to calculate the total ESF response time. However, the
licensee never actually timed the installed relays at the specified surveillance i
interval. Additionally, all the relays had been replaced with a new design between l
1993 and 1995, and the actuation of the new relays had not been timed.
The licensee observed that a phrase that had been in the original TS basis had been
relocated to the definition of ESF response time in the TSIP, and that the change
had led the licensee to a different interpretation of the TS requirement, which was
otherwise unchanged between the original TS and the TSIP. The Vice President,
Nuclear Generation, stated during the February 15,1997, conference call that the
licensee considered that the method that had been used for the SR satisfied the
original TS requirements, but did not satisfy the requirements of the TSIP. The NRC
considered that the licensee had made an error in the review of the TSIP. On
February 15,1997, the NRC verbally granted the NOED for the ESF response time
testing. The licensee submitted a written request for the NOED on February 16,
and a TS amendment request on February 18. The NRC's approval letter for the
NOED, dated February 19,1997, states that the bounding response time evaluation l
developed in 1983 may not have been a valid method of meeting the original TS l
requirements, and that the relay subgroups may have required specific response i
time testing to satisfy the origir:al TS requirements.
Additional examples of SRs not having been properly performed were reported in
licensee event reports (LERs) and are discussed in Sections 08.2 and E8.1. The
issues reported in those LERs will be incorporated in to the ongoing inspection of
this unresolved item.
.
.
-12-
c. Conclusions
The licensee had not performed tests demonstrating that SRs 3.1.5.4, 3.3. 5.6,
3.8.1.8., 3.8.1.9.c, 3.8.1.10,3.8.1.14, or 3.8.1.15 were satisfied, from August 5,
1996, until testing was completed during the Unit 2 refueling outage, which began
on November 30,1996. The licensee was continuing to review the implementation
of the TSIP surveillance requirements; therefore, this unresolved item will remain
open pending the completion of that review and NRC evaluation of the results.
08.2 (Closed) LER 50-361:362/97-001-00: surveillances not current upon improved TS
implementation. This LER describes the issues already discussed in Section 08.1,
above, and includes additional sc.veillances found not to be current, that were
identified by the licensee as put of an exhaustive review of the TSIP surveillances.
The inspector considered these additional surveillances, SR 3.7.8.4 (saltwater
cooling pump auto start) and SR 3.3.7.1 (channel checks on EDG undervoltage) to
be more examples of the condition discussed in Section 08.1. This will be
reviewed as part of Unresolved item 361;362/96018-02.
II. Maintenance
M1 Conduct of Maintenance
M 1.1 General Comments
a. Insoection Scope (62707)
The inspectors observed all or portions of the following work activities:
- Motor-operated valvo inspection of Valve 2HV4716, steam supply to
auxiliary feedwater pump Turbine K007 (Unit 2)
system (Unit 2)
- Repair of feedwater check Valve 2MUO129 (Unit 2)
- Repair of feedwater check Valve 2MUO36 (Unit 2)
i
b. Observations and Findinas
The inspectors found the work performed under these activities to be thorough. All
work observed was performed with the work package present and in active use. l
Technicians were knowledgeable and professional. The inspectors frequently ;
observed supervisors and system engineers monitoring job progress, and quality ;
control personnel were present whenever required by procedure. When applicable,
appropriate radiation controls were in place.
l
.
.
-13-
In addition, see the specific discussions of maintenance observed under
Section M1.4, below.
M1.2 General Comments on Surveillance Activities
i
a. insoection Scope (61726) ?
l
, The inspectors observed all or portions of the following surveillance activities:
- Containment Penetration C203, Containment Emergency Airlock Door,
Interlock Test (Unit 3)
!
- EDG 2 GOO 2 Monthly Surveillance (Unit 2)
4 * Integrated ESF System Refueling Test (Unit 2) l
b. Observations and Findinas
The inspectors found all surveillancas performed under these activities to De
thorough. All surveillances observed were performed with the work package
present and in active use. Technicians were knowledgeable and professional. The 1
inspectors frequently observed supervisors and system engineers monitoring job ,
- progress, and quality control personnel were present whenever required by (
procedure. When applicable, appropriate radiation controls were in place. ,
i
in addition, see the specific discussions of surveillances observed under
Sections M1.3 and M1.5, below.
j M 1.3 HPSI Pumo inservice Testina - Unit 3
a. Inspection Scope (61726) !
On January 14,1997, the inspectors observed inservice test technicians and
operators perform Procedure SO23-3-3.60.1,"HPSI Pump 2(3}MP-018 and Valve
Testing for Train A."
b. Observations and Findinas
The prejob brief for the surveillance test included a description of the evolution, the
necessary communications between the control room and personnel at the pump,
expected pump parameters during the test, pump operational limitations, "as low as
reasonably achievable" radiation dose reduction considerations, and industrial
safety. The crew assigned an additional reactor operator to monitor the control
boards so the control room crew involved in the evolution could focus on the
briefing.
The operators correctly performed the procedure and accurately obtained the pump
flow and vibration readings. In addition, the operators displayed good independent
. .- -. - _- ~ ~- -. -- - - - _- - - - _ .
4 .
4 .
<
14-
i verification techniques. The system engineer monitored pump performance and
~
i provided technical guidance during the test.
c. Conclusions
Operations personnel demonstrated an overall strong performance during the prejob
briefing and the HPSI surveillance test. On-scene support by the system engineer
was effective.
M 1.4 UGS Installation - Unit 2
a. Inspection Scoce (62707)
On January 15,1997, the inspectors observed the installation of the UGS into the
reactor vessel in accordance with Procedure SO23-I-3.14,"in-core Instrumentation
Assembly Installation and Restoration of the UGS."
b. Observations and Findings
Maintenance personnel used appropriate controls and safety precautions when
lifting the UGS from the storage stand and placing it in the reactor vessel. The
technicians properly used the procedure and followed the quality control hold
points. Health physics technicians maintained continuous coverage of the evolution
and provided routine updates of the radiation levels to the maintenance personnel.
c. Conclusions
Licensee personnel exhibited strong performance during the UGS lift and placement
in the reactor vessel. >
M1.5 Seismic Monitor Surveillance - Unit 2
a. insoection Scope (61726)
The inspectors observed the calibration of the seismic triggers in accordance with
Procedure SO23-II-2.5, Revision 8, "SR Seismic Time History Accelerograph
Models SMA-2/ Trigger Package and SSA-1 Channel Calibration."
b. Observations and Findinas
On January 28,1997, the inspectors observed I&C technicians perform a
calibration on Seisrnic Trigger XS-8020F. The seismic triggers actuate
accelerographs which are used to determine if the design limitation of the ,
equipment is exceeded during a seismic event. The calibration involved checking
the positive and negative acceleration setpoints in the vertical, longitudinal, and
transverse channels. Each of the channels had a sensitivity potentiometer to adjust
the trigger setpoint that was common for both the positive and negative
a
_. _ _ __. - - -
-
!
l
_
l
-15-
accelerations; therefore, any sensitivity potentiometer adjustment would affect both
the positive and negative setpoint values. !
During the performance of Procedure SO23-II-2.5, the I&C technicians performed a
caQration of the vertical channel positive acceleration and observed that the
as-found setpoint was within the range of the acceptance criteria. The technicians
observed that the negative acceleration setpoint was outside the acceptance criteria l
'
range and required adjustment. The technicians followed the procedure, adjusted
the potentiometer and obtained satisfactory as-left results for the negative ]
acceleration. The inspectors observed that the procedure did not direct the i
technicians to check the as-left positive acceleration setpoint that was also adjusted
by the common potentiometer. The inspectors raised the concern to the
technicians, who agreed with the inspectors' observations that the positive
acceleration setpoint was affected, but continued with the procedure to check the j
remaining two channels. i
Upon completion of the remaining two channel calibrations, the inspectors again
questioned the technicians about the as-left value for the positive acceleration of {
the vertical channel. The technicians noted that the procedure did not provide i
guidance on rechecking the positive acceleration and decided to discuss the
inspectors' concerns with their supervisor,
l&C supervision acknowledged that the procedure did not provide direction to verify
the posiuve acceleration as-left setpoint, and discussed the procedure inadequacies
with the document author, who agreed to initiate a temporary change notice to
correct the procedure deficiencies, in addition, an I&C technician identified that,
should the positive acceleration setpoint be outside the acceptance criteria, the ;
procedure would direct the technicians to adjust the setpoint and then the as-found j
'
data for negative acceleration would be lost. The instructions for adjustment of the
seismic trigger were not adequate, in that they assumed that the as-left conditions
were still within the acceptance criteria. However, the instruments were classified ,
as Quality Class IV and, therefore, not subject to the requirements of 10 CFR j
Part 50, Appendix B. j
!
I
l&C supervision used guidance in Procedure SO123-1-1.3," Work Activity i
Guidelines," to allow the technicians to reperform steps of the procedure to obtain '
the as-left positive acceleration setpoint values. The l&C technicians reperformed i
the section of the procedure and obtained an acceptable as-left value for the l
positive acceleration. !
!
The licensee initiated AR 970200065to document and evaluate the procedure and l
personnel weaknesses. The l&C superintendent of plant maintenance indicated that l
the technicians did not meet management expectations of stopping the evolution j
when they initially became aware of potential procedure deficiencies. The licensee !
reviewed the vendor recommended guidance for the performance of the test and i
noted that the procedure reflected the vendor guidance. The inspectors agreed that .i
the procedure reflected the vendor guidance on how to check the acceleration l
setpoint, but observed that the vendor technical manual did not include aspects on j
!
l
,
.
.
16-
recording the as-found and as-left setpoints. The licensee planned to review other
seismic instrument procedures for similar deficiencies,
c, Conclusions
I&C technicians did not meet management expectations in that they did not
promptly raise potential procedure deficiencies to supervision for resolution. The ,
inspectors identified that the procedure for adjustment of the seismic trigger did not j
adequately assure that the as-left conditions were within the acceptance criteria:
however, this was not a violation because the instruments were not subject to the
10 CFR Part 50, Appendix B, regulations.
M2 Maintenance and Material Condition of Facilities and Equipment
M 2.1 Facility Preservation (71707) j
During routine plant walkdowns, the inspectors observed continued improvement in
the material condition of general plant areas, including the saltwater cooling pump
rooms, the EDG rooms, and the auxiliary building roof. The licensee recently !
completed a significant project to preserve the ventilation housing on the side of the j
containment building, and had continued to expend significant resources to repaint '
the turbine building and other areas susceptible to the corrosive salt air
environment. The inspector observed that these efforts resulted in better lighting l
condit i ons in most areas, and that more minor material deficiencies were being
identified (and enrrected) as the ability to see them improved. The inspector I
concluded that the licensee's preservation efforts contributed substantially to the I
material condition of the facility.
M8 Miscellaneous Maintenance issues (92700)
M8.1 (Closed) LER 50-361/96011-00; air-operated containment isolation valve
inoperable. The licensee identified that pressurizer surge line sample
Valve 2HV0513 would not close under design basis conditions. The other series
valve in the line remained operable. The licensee performed an evaluation and
determined that two separate errors could have caused the insufficient closing
force. The vendor used incorrect information in the determination of the required
actuator setpoint and the licensee's valve reassembly procedure was deficient. The
licensee reset and retested Valve 2HV0513 and similar Unit 2 AOVs during the
recent outage. In addition, the licensee reviewed valve stroke tests and confirmed
that similar Unit 3 AOV actuator closir.g force settings were acceptable.
The licensee concluded that Valve 2HV0513 had been inoperable when Unit 2 was
in Mode 1 and the required action of TS limiting condition for operation 3.6.3 was
-not performed. The inspector concluded that the licensee's AOV program was
proactive in identification of the degraded valve and corrective actions were
appropriate. This licensee-identified and corrected violation is being treated as a
noncited violation, consistent with Section Vll.B.1 of the NRC Enforcement Policy
(NCV 361/97002-04).
_ _ , _
.
.
-17-
lli. Enaineerina
E1 Conduct of Engineering
E1.1 Root Cause Analysis - RCS Thermowell Failure (37551)
The licensee completed an evaluation of the root cause of the thermowell failure
which had resulted in pressure boundary leakage from the RCS in Unit 3 (see NRC
Inspection Report 50-361/96-11:50-362/96-11). The final evaluation included
detailed metallurgical analysis, including micrographic photographs, which confirmed
that the thermowell f ailed due to high cycle fatigue. The licensee further
determined that the thermowell had been incorrectly installed during initial plant
construction. The licensee removed and inspected three similar thermowells during
the Unit 2 refueling outage and found no wear indications like those observed in the
failed thermowell. The licensee also considered the potential for the threaded plug
to have failed, and determined that such a failure would have been well bounded by
the licensee's small break loss of coolant accident analysis. Based on a review of
the evaluation and discussion with the licensee's metallurgist, the inspectors i
concluded that the root cause assessment was extremely thorough and represented
substantial technical depth.
E1.2 SG Chemical Cleanina Results - Unit 2
a. Insoection Scooe (37551)
The inspectors reviewed portions of the licensee's Engineering support activities
associated with the performance of SG chemical cleaning in Unit 2, including the
licensee's assessment of the effects of the chemical cleaning on SG integrity. The
inspectors discussed the chemical cleaning process with the licensee before, during, l
and after the process was conducted. l
l
b. Observations and Findinos l
l
The licensee performed chemical cleaning of the Unit 2 SGs. Prior to the outage,
the licensee evaluated the various aspects of the process per 10 CFR 50.59, and
concluded that the process did not represent an unreviewed safety question and l
was acceptable. The licensee also obtained a site-specific qualification report from l
the vendor for the process.
1
The licensee determined that the chemical cleaning process would result in some
corrosion of SG internal components, and that the tube support egg crates were the
most limiting components subject to this anticipated corrosion. The licensee ,
I
installed a corrosion monitoring system to provide real-time indication of the
corrosion rates within the SG during the process, and installed corrosion coupons to
allow verification of the indicated corrosion rates. After the first iron removal step,
the licensee determined that a coupon showed significantly greater corrosion than I
anticipated or indicated by the corrosion monitoring system. The licensee
subsequently determined, and documented in AR 961201816,that the placement
i
l
.
- 18 -
of the coupon had resulted in it becoming a sacrificial anode in a galvanic field
greater than that which would be representative of the egg crate supports. The
licensee then redesigned and relocated the coupon to be more representative of the
egg crate supports. The licensee also reduced the level of the chemicals in the SG
to be below the eighth support, as the upper supports had the least margin. One
high-temperature iron removal step was also canceled to minimize corrosion of SG
internals. The licensee's inspection of the coupons, portions of the vertical straps,
and portions of the lowest egg crate support, after the completion of the chemical
cleaning process, confirmed that the corrosion monitoring system was reasonably
accurate and that the extent of the corrorion of the egg crates was within the
predicted and allowed tolerances.
Approximately 18,000 pounds of corrosion products were removed by the chemical
cleaning and sludge lancing processes from each SG. The licensee had estimated
the SG deposit loading to be a nominal 15,695 pounds, and a maximum of
19,214 pounds. A visualinspection of the tubes performed by the licensee showed
that the tubes were clean, although a small amount of deposits remained in the
crevices between the tubes and some of the egg crate supports.
'
c. Conclusions
The licensee's engineering evaluation and support of the chemical cleaning process ,
were thorough. The licensee's actions in response to the in-process corrosion I
indications were conservative. Engineering predictions for the amount of internal
SG corrosion products that would be removed by chemical cleaning were accurate. l
Engineering support for the chemical cleaning process was excellent.
E1.3 Indeoendent Checks While Reloadina the Core (37551)
While reviewing a portion of Updated UFSAR Chapter 15, the inspectors observed
that Section 15.4.3.1.1.2," Erroneous Placement or Orientation of Fuel
Assemblies," stated that periodic independent inventories of components in the
reactor core and spent fuel pool would be made to ensure the refueling tag board in
the control room was accurate. These inventories were to be independent of the
operators in the control room and containment controlling core reload, and were to
be incorporated in procedure. The inspectors found, based on previous observation
of core reload, that independent checks of this nature were not being performed and
were not proceduralized. On January 30,1997, the refueling manager
acknowledged this finding, and planned on resolving it prior to the upcoming Unit 3
refueling outage. This UFSAR discrepancy is similar to issues described in
Violation 361;362/9526-02. The corrective actions being performed by the
licensee for that violation have not been completed. The licensee's corrective
actions will be reviewed as a followup item in conjunction with the review of that
violation (Inspector Followup Item 361;362/97002-05).
.
.
-19-
E2 Engineering Support of Facilities and Equipment
E2.1 Review of Facility and Eauioment Conformance to UFSAR Description (37551)
A recent discovery of a licensee operrting its facility in a manner contrary to the
UFSAR description highlighted the need for a special focused review that compares
plant practices, procedures and/or parameters to the UFSAR description. While
performing the inspections discussed in this report, the inspectors reviewed the
applicable sections of the UFSAR that related to the inspection areas inspected.
The following inconsistency was noted between the wording of the UFSAR and the
plant practices, procedures and/or parameters observed by the inspectors.
Periodic independent inventories of the reactor core and spent fuel pool were not
being performed during core reloading, as described in UFSAR Section 15.4.3.1.1.2
(Section E1.3).
E2.2 .Resoonse to Industrv Issues
a. Inspection Scoce (37551)
The inspectors reviewed the licensee's response to notifications of two generic
problems received by the licensee during this inspection period.
b. Observatics _and Findinas
On December 29,1996, the licensee received a notification from the Vogtle facility
licensee that motor coolers for fully-enclosed Westinghouse motors for emergency I
core cooling system pumps could be installed in an incorrect orientation, resulting in i
the coolers having one-pass instead of three-pass flow paths, and rendering the I
coolers incapable of removing design heat loads. The licensee's ISEG initiated
contact with both Vogtle and the motor cooler vendor. Within one week, the
licensee visually inspected all susceptible motors, identifying three motors that '
required further inspection. The additionalinspections were accomplished within
the next 2 days. No incorrectly oriented coolers were identified.
In January 1997 the licensee's Combustion Engineering Owners Group
representative received an electronic notification (via the Internet) from i
Asea Brown-Boveri(ABB) that calculations for main steam safety valve lift setpoints i
may not have considered the pressure drop from the SGs to the valves. The
licensee promptly confirmed with ABB that the ABB calculational error did not affect l
valves installed at San Onofre. This effort principally involved both ISEG and
Nuclear Engineering Design. Nuclear Engineering Design had recognized and
dispositioned the issue with ABB several months earlier as it related to San Onofre.
c. Conclusions
The licensee's ISEG and Engineering groups demonstrated aggressive and prompt
response to two notifications of generic problems. The value of electronic means of
_ - .- - . - . . - . . . _ ---
,
o !
.
'
20-
communication was also evident in promulgating the notifications to the licensee
'
- and among licensee personnel.
,
E2.3 Debris Left in Reactor Pressure Vessel- Unit 2
j a. Insoection Scoce (37551)
i
The inspectors reviewed a licensee evaluation done for foreign material being left in ,
'
the Unit 2 RCS. This evaluation was contained in Nonconformance
l
. Report 970100329, Revision O. The inspectors also reviewed ARs 970100329,
970100205,and 970100276, written to document both foreign material that had
been removed from the reactor vessel, and foreign material that had not been
removed.
,
j b. Observations and Findinas
1
l Licensee refueling personnel had conducted remote camera inspections of the
reactor vessel, prior to core reload with the core support plate in place, in order to
! identify and retrieve any loose debris. The licensee identified approximately ten I
items of debris. Six of these items were removed from the vessel. Licensee
i
refueling personnel determined four could not be removed: an object resembling a
j cap screw, a smalllight bulb, a small piece of paint or metal, and an object
'
resembling a washer. These objects were evaluated for the acceptability of being
- left in the vessel. The cap screw and washer were identified as being below the
i
core support plate at the bottom of the vessel. The light bulb was noted to have
! become detached from a remote camera, but was not found by further camera
i inspections. The paint or metal chip was seen on top of the core support plate, and l
- was thought to have been vacuumed; however,it was not found in the vacuum
- filter. Licensee refueling personnel attempted to retrieve the washer and cap screw
by vacuuming for approximately 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. Due to the location of the objects, below
i flow distributinn holes, which made positioning the vacuum hose difficult, they
were not recovered.
The inspectors determined that the safety evaluation, which included an evaluation
j as to whether leaving the debris in the RCS represented an unreviewed safety
- question as required by 10 CFR 50.59, was rigorous and comprehensive. Based on
- the size, location, and composition of the material, and on review of the evaluation,
the licensee determined that leaving the debris in the vessel was not an unreviewed
safety question and was acceptable.
l The unrecovered debris was found on the top of the core support plate, prior to
- core reload. The nonconformance report indicated that this location, and the
.
radioactivity of recovered items, was used to conclude that the unrecovered items,
'
with the exception of the light bulb, had been present throughout the previous fuel
- cycle. The inspector determined that another possibility was that the objects were
dropped into the vessel during this refueling outage and that they would have fallen
j on top of the core support plate and could have drifted beneath the plate, where
- they were found. In addition, whether or not the unrecovered items were
.
A ..
I
s
1
-- ._ _ - . ,
- .-
_
- . --_-.
l o
l
.
l
l -21-
radioactively activated was not known. Because the evaluation logic t'!d not
depend on the argument that the debris had already been in the RCS for the
I
previous cycle, the inspector still considered it acceptable to leave the items in the
l RCS.
c. Conclusions
A safety evaluation performed to determine the acceptability of leaving some debris
in the RCS was thorough and rigorous.
E2.4 ECT Results - Unit 2
a. Insoection Scope (37551)
The inspectors reviewed the results of the licensee's ECT of the Unit 2 SG tubes,
and discussed the results with licensee and NRC personnel,
b. Observations and Findinas
The licensee performed an ECT inspection of the SG tubes (approximately
9000 tubes in each SG) and identified significant unanticipated tube degradation in
both SGs. The inspection program and results were described in "Special Report:
Inservice inspection of SG Tubes," submitted by the licensee on February 6,1997.
The inspectors monitored ECT progress and reviewed the special report. The ECT
results indicated three notable conditions. Top-of-tubesheet transition area defects,
both axial and circumferential, were identified in approximately 254 tubes. Axial
cracking, primarily in the lower bundle region, was identified in approximately
29 tubes in the freespan and at egg crate tube support locations in another 17
tubes. Dents were identified in approximately 1900 tubes. As a result of the ECT
inspections, the licensee removed 332 tubes from service.
The licensee performed in-situ pressure testing of 21 tubes, of which one tube
f ailed, and removed three tubes for laboratory analysis. The licensee agreed to
perform a midcycle inspection of the SG tubes, the timing of which would be
determined on the basis of the results of the laboratory analysis of the removed
tubes, pressure testing, and ECT data analysis.
Ecitewing the recommendations of an expert panel that reviewed the status of SG
corrosion issues at San Onofre, the licensee's Onsite Review Committee approved
plans to begin titanium dioxide injection in the feedwater. Titanium dioxide is
intended to condition the tubes to inhibit intergranular attack / stress corrosion
cracking propagation. The inspectors reviewed the licensee's 10 CFR 50.59 safety
evaluation and concluded that it thoroughly addressed the safety implications of the
use of titanium dioxide.
l
..---.- _ _ . _ --. - - - . . _ _ _ _ _ - - _ . _ _ _ _ . . __ _
'
.
1
,
..
! -22-
c. Conclusions
The inspectors concluded that the licensee's evaluation and corrective actions to
i date'were thorough for SG tube inspection results.
,
E4 Engineering Staff Knowledge and Performance
t
'
E4.1 SG In-situ Pressure Testina - Unit 2
!
! a. inspection Scope (37551)
i The inspectors observed the licensee performance of Procedure SO23-XXVil-4.34,
i "In-situ Hydro Test," in SG 20089.
I
T
b. Observations and Findinas
On January 25,1997, the inspectors observed the licensee perform in-situ pressure
testing of the SG tubes. The inspectors performed a detailed review of the
procedure and noted that the licensee pressurized the tubes to the correct pressure,
but did not maintain the pressure for a minimum of 5 minutes as required by
procedure. The licensee held the pressure for 1 minute. The inspectors notified !
the licensee of the discrepancy. The licensee had previously suspended the testing
for the day shift and indicated that an evaluation of the results and procedure
compliance would be performed. The failure of the licensee to follow
Procedure SO23-XXVI-4.34 is a violation of 10 CFR Part 50, Appendix B,
Criterion V (Violation 361/97002-02).
The inspectors determined that several personnel missed opportunities to identify
the required pressure hold time. A contractor represcatative supervising the
activities inside containment failed to identify the correct hold time. In addition,
both a contractor quality control personnel, and a licensee engineer providing
process oversight failed to recognize the hold time requirement.
The inspectors discussed the event with the Site Technical Support SG engineer. .;
The engineer indicated that, prior to recommencing the pressure testing, an
extensive procedural review was conducted. The licensee conducted an additional
prejob brief and added a blank on the data sheets to document the required pressure
hold time. The licensee found that the contractors involved in the in-situ pressure
testing had significant industry experience performing this evolution, and may have
used the 1-minute pressure hold time used by another utility. Management
expectations for following procedures were discussed with applicable personnel.
The inspectors questioned the system engineer about the prejob briefing. The
engineer indicated that the briefing focused on health physics issues and the
interf ace with the contractor. In addition, the briefing included discussions on what
to do if a tube leaked or burst, and the required test pressures. The brief did not
include pressure hold times.
.- - .-.-. . _. - _ - - - - . -. - .- -.. ~. .. -
.
..
i
23-
The inspectors observed positive verification of the tube to be tested, by both the
contractor quality control representative and the licensee engineer providing
oversight, using known locations in the SG map. In addition, the inspectors ;
I observed good communications between personnelinside containment and at the '
,
rcmote control station.
- c. Conclusions l
'
Personnel performing the in-situ pressure test of the SG tubes failed to follow
procedure and maintain the test pressure for the required time interval. In addition,
i the prejob briefing was inadequate in that pressure hold times were not discussed,
i
However, the test personnel used good communications and tube verification
techniques. The licensee repeated, successfully, the pressure tests, using the
required five minute pressure hold time, on the tubes which had been tested at the
'
one minute hold time.
4
I
.
E8 Miscellaneous Engineering issues (92700,92903)
i
i E8.1 (Closed) LER 50-361: 362/96-009-00and 01: failure to complete surveillance
testing of EDG noncritical trips.
i. a. Backaround i
l
! Each EDG has a number of noncritical trips which will trip the EDG under normal l
operation, but are bypassed with a safety injection actuation signal (SIAS) present.
Prior to August 1996, SR 4.8.1.1.2.d.7 required that the licensee verify that these
trips were bypassed when, " Simulating a loss of offsite power in conjunction with
an ESF [SIAS) test signal . . ."
,
in August 1996 the licensee implemented standard TS resulting from the TSIP.
SR 3.8.1.13 required that the noncritical trips be verified as bypassed "on actual or
simulated loss of voltage signal on the emergency but concurrent with an actual or
simulated ESF [SIAS) actuation signal . . ."
,
'
in December 1996 the licensee identified that the procedure that had been used
since 1983 to perform the test of the EDG-bypassed trips did not simulate a loss of
-
offsite power. The licensee then completed SR 3.8.1.13, simulating both a loss of
voltage and a SIAS within the time allowed by SR 3.0.3.
The licensee identified a number of root causes for this error, including failure to
ensure that their procedure adequately implemented the TS requirements during
'
initial procedure preparation in 1983 and during reviews associated with the TSIP in
1996. The licensee stated that a simulated loss of voltage /offsite power had no
effect on the EDG trips, and considered that the bypass of noncritical EDG trips had
always been adequately verified.
4
A
4
_ _ , - , . _ _ ..
-
,
d
-24-
The licensee took a number of corrective actions, including reviewing their
procedures for completing the TSIP for other potential omissions or errors,
b. Insoection Scope l
s
The inspectors reviewed the LER, circuit diagrams associated with EDG trip circuits,
and the licensee's root cause invastigation results.
l
c. Observations and Findinas l
The inspectors determined that the licensee's circuit drawings indicated that a loss
of voltage /offsite power signal had no affect on EDG trip signals. l
l
The inspectors found that, in May 1996, operators, reviewing existing procedure i
requirements to ensure compliance with the TSIP SRs, identified that Surveillance
Operating instruction SO23-3-3.23.1, Revision 8, " Diesel Generator Refueling ,
Interval Tests," credited by the licensee for completing TSIP SR 3.8.1.13, was j
being accomplished by initiating a SIAS but not a loss of voltage /offsite power. The l
operators identified the problem in writing to a licensing supervisor who signed an l
evaluation that the existing procedure was adequate to meet TSIP SR until the SR
could be changed. Approximately 6 weeks later, this same supervisor signed
approval of his own evaluation. The licensee implemented the TSIP on August 5,
1996.
In December 1996 Engineering personnel, reviewing documentation associated with
SR 3.8.1.13, again concluded that Procedure SO23-3-3.23.1 was not in compliance
with TSIP SR 3.8.1.13, and brought this issue to the attention of senior personnel, 4
who directed that the procedure be completed as required by the TSIP, l
The licensee's investigation determined that the licensing supervisor based his l
decision on an engineering evaluation, which indicated that the loss of
voltage /offsite power signal did not affect the test, and concluded that this decision
was inadequate. The licensee also concluded that failure to have an independent
review of the decision contributed to the error. The inspector considered that the
licensee also f ailed to identify that Procedure SO23-3-3.23.1 did not appear to meet
the old TS requirements,in effect at the time of the review.
The inspectors determined that the licensee failed to ensure that the procedure of
record met the requirements of TSIP SR 3.8.1.13. This issue will be reviewed as
part of Unresolved item 361: 362/96018-02,in conjunction with other TSIP SR
deficiencies recently identified by the licensee (see Section 08.1).
d. Conclusions
Licensee personnel used poor judgment in May 1996 in concluding that
Procedure SO23-3-3.23.1 met the requirements of TSIP 3.8.1.13.
Procedure SO23-3-3.23.1 was technically adequate; therefore, there was little
safety consequence to the licensee's failure to meet the surveillance requirement.
a
.
.
-25- l
This item will be reviewed as an unresolved item along with other deficiencies I
recently identified with implementation of TSIP SRs. I
E8.2 (Closed) Insoection Followuo item 50-361:362/96005-04.LER 50-361:
362/96-005-00and 01: anomalous performance of containment HRRMs. This item
involved a licensee evaluation of two effects scen when containment HRRM signal
cabling was subjected to changes in temperature and/or a high steam environment.
The temperature change induced currents in the signal cable that caused erroneous
readings of the instrument. The steam environment caused moisture to penetrate
the cable insulation, migrate through the cable around the inner shield, and caused
amphenol connections on a pigtail to the penetration to fill with water and cause the
instrument to fail.
The licensee replaced the coaxial cable and the pigtail arrangement for the Unit 2
HRRMs. The licensee replaced the Rockbestos RSS-6-104LEcable with ABB/CE
coaxial cable, which was less sensitive to water intrusion and temperature induced
currents. The licensee also changed the penetration location for the signal cable to
a location that facilitated a penetration without a pigtail; one in which the signal
cable attached directly to the penetration plate.
The licensee planned on submitting LER 96-005-01 to clarify statements made in
Revision 0 regarding the penetration, and to enclose portions of the testing results.
The inspector reviewed Environmental Qualification Report M85114, Revision 0, '
dated November 15,1996. Based on this review, the inspectors found the
licensee's engineering analysis and testing done at Wyle Laboratory comprehensive
and aggressive in identifying and minimizing the irregularities.
E8.3 (Closed) URI 361: 362/06011-01: RCP oil collection system deficiencies. In 1996
the licensee identified several potential leak points on the RCPs in both units that
were not protected by the RCP oil collection system, as prescribed in 10 CFR
Part 50, Appendix R, Section Ill.O. The NRC completed a review of the deficiencies
and applicable requirements, and concluded that the RCP oil collection systems in
Units 2 and 3 did not comply with Section Ill.0, which required that the collection
systems be capable of collecting lube oil from all potential pressurized and ]
unpressurized leakage sites in the RCP lube oil systems. This licensee-identified and !
corrected violation is being treated as a noncited violation, consistent with
Section Vll.B.1 of the NRC Enforcement Policy (NCV 361:362/97002-06).
The licensee completed the extension of the tube oil collection system under the l
potential leakage sites in Unit 2 in January 1997,in accordance with Field Change l
Notice F-1716C. On February 6,1997, the inspectors inspected the installed
modification to the oil collection system in Unit 2. The inspectors observed that the
areas of previous concern for oil collection were properly protected for the three
Siemens-Allis motors. The licensee replaced the fourth motor with an ABB motor
end the oil collection piping was in the process of being installed. The licensee ;
plans to complete the corrective actions in Unit 3 during the next refueling outage.
i
l
i
.
,
,
,
O
-26- l
IV. Plant Support
R1 Radiological Protection and Chemistry Controls
,
l
R 1.1 Unoosted Loose Surface Contamination Area - Unit 2(71750) i
On February 10,1997, the inspectors observed that wet boric acid was built up on
a sample point for the common miscellaneous waste evaporator, downstream of
Valve 2/3AP7298. The sample point was located in Room 505c, a small room off ;
the radwaste operator's station in a radiologically controlled area. The inspectors
surveyed a small portion of the buildup. A licensee Health Physics technician
confirmed the survey results, and found 8000 disintegrations per minute (dpm) of )
activity. However, this was not a violation of requirements because there was no 1
specific licensee programmatic requirement, or NRC requirement, to post loose
surf ace contamination, as long as radiation exposure was maintained as low as
reasonably achievable. Licensee management expectation was that any activity
over 1000 dpm/100 centimeter squared would be posted as a contaminated area.
In response, the licensee posted the area as contaminated. The inspectors found
. that the response to the their concerns was good, but that, in this instance, licensee
Health Physics personnel were not sufficiently timely in identifying and posting
loose surface contamination in excess of limits.
V. Manaaement Meetinas
X1 Exit Meeting Summary
The inspectors presented the inspection results to members of licensee management
at the exit meeting on February 25,1997. The licensee acknowledged the findings
presented.
. The inspectors asked the licensee whether any materials examined during the
inspection should be considered proprietary. No proprietary information was
identified.
I
1
I
i
-r
.
-2 7-
ATTACHMENT
SUPPLEMENTAL INFORMATION
PARTIAL LIST OF PERSONS CONTACTED
Licensee
D. Brieg, Manager, Station Technical
J. Fee, Manager, Maintenance
G. Gibson, Manager, Compliance
R. Krieger, Vice President, Nuclear Generation
D. NunniVice President, Engineering and Technical Services
T. Vogt, Plant Superintendent, Units 2 and 3
R. Waldo, Manager, Operations
INSPECTION PROCEDURES USED
IP 37551: Onsite Engineering
IP 61726: Surveillance Observations
IP 62707: Maintenance Observations
IP 71707: Plant Operations
IP 71750: Plant Support Activities
IP 92700: On Site LER Review
IP 92901: Followup - Operations
IP 92903: Followup - Engineering
.
o
_ __ _ .. _ _. _. - _ - _ . .
f
.
.
.
4
i 28-
ITEMS OPENED AND CLOSED
Onened l
50-361/97002-03 VIO Inadvertent transfer of water to RCS, and
failure to maintain RCS pressure within limits
$ 50-361/97002-05 IFl UFSAR discrepancy: Core reload checks i
50-362/97002-05
50-361/97002-02 VIO In-situ pressure testing of SG tubes
Ooened and Closed
50-361/97002-01 NCV ECU access panels found unlatched and ajar and CREACUS
50-362/97002-01 access panel found open
50-361/97002-04 NCV Air-operated containment isolation valve inoperable
50-361/97002-06 NCV RCP oil collection system deficiencies
50-362/97002-06
Closed
50-361/97001-00 LER Surveillance not current upon improved TS implementation
50-362/97001-00
50-361/96011-00 LER Air-operated containment isolation valve inoperable
50-361/96009-00 LER Failure to complete surveillance test of EDG noncritical trips
50-362/96009 00
50-361/96009-01
50-362/96009-01
50-361/96005-04 IFl Anomalous performance of containment HRRMs
50-362/96005-04
50-361/96005-00 LER Anomalous performance of containment HRRMs
50-362/96005-00
50-361/96005-01
50-361/96005-01
50-361/96011-01 URI RCP oil collection system deficiencies
50-362/96011-01
_ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ - _ . . _ - _ _ - . _ . _ _ _ . _ - _ _
-
_ . .._ _ _ _ _ - . _ _ _
i
4
.
.
-2 9-
Discussed
50-361/96018-02 URI surveillance tests not performed in required manner
50-362/96018-02
LIST OF ACRONYMS USED -
ABB Asea Brown-Boveri
AOV air-operated valve
AR action request
CEA control element assembly
CO control operator
CREACUS control room emergency air clean up system !
ECU emergency cooling unit
EDG emergency diesel generator
ESF engineered safety features
HPSI high pressure safety injection
HRRM high range radiation monitors
HVAC heating, ventilation and air conditioning
I&C instrumentation and control
ISEG Independent Safety Engineering Group.
LER licensee event report
LPSI low pressure safety injection
NOED notice of enforcement discretion
PDR Public Document Room
RCP reactor coolant pump
RWST refueling water storage tank
SIAS safety injection actuation signal
SIT safety injection tank
SR surveillance requirement ,
SS- shift superintendent i
TS Technical Specification l
TSIP Technical Specification Improvement Program
UEL upper electrical limit
UFSAR Updated Final Safety Analysis Report
UGS upper guide structure