ML20045C215
| ML20045C215 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 05/24/1993 |
| From: | Wong H NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| To: | |
| Shared Package | |
| ML20045C207 | List: |
| References | |
| 50-528-93-11, 50-529-93-11, 50-530-93-11, NUDOCS 9306220220 | |
| Download: ML20045C215 (33) | |
See also: IR 05000528/1993011
Text
.- _
,
s
'
l
,
s
l
U. S. NUCLEAR REGULATORY COMMISSION
REGION V
Report Nos.
50-528/93-11, 50-529/93-11, and 50-530/93-11
Docket Nos.
50-528, 50-529, and 50-530
License Nos.
Rcensee
Arizona Public Service Company
P. O. Box 53999, Station 9082
Phoenix, AZ 85072-3999
Facility Name Palo Verde Nuclear Generating Station
Units 1, 2, and 3
Inspection
Conducted
March 25 through April 26, 1993
Inspection
Location
Wintersburg, AZ
Inspectors
J. Sloan,
Senior Resident Inspector
H. Freeman,
Resident inspector
-
D. Kirsch,
Technical Assistant
A. MacDougall, Resident Inspector
J. Mel fi ,
Resident Inspector (Trojan)
F. Ringwald,
Resident Inspector
f/3y
J
Approved By
r
W
H.1ong, CWief
V
Date Signed
Reactor ProjectsSection II
Inspection Summary:
Areas Inspected:
Routine, onsite, regular and backshift inspection by the
resident and regional inspectors. Areas inspected included:
o
review of plant activities
e
surveillance testing - Units 1 and 3
plant maintenance - Units 1, 2, and 3
e
feedwater isolation valve operability determination - Unit 1
e
i
e
steam generator U-tube mis plugging - Unit 2
e
steam generator U-tube inspections - Unit 2
e
snubber testing - Unit 2
e
leak off lines not designed to be live loaded - Unit 3
LPSI "B" breaker silent inoperability - Unit 3
e
e
turbine trip - Unit 3
inadvertent opening of multiple steam bypass control valves, plant power
e
exceeded 100% - Unit 3
evaluation of licensee self-assessment capability - Units 1, 2, and 3
e
9306220220 93052s
{DR
ADOCK 05000528
.
.
- -
. - - -
.
f
'
.
.
.
e
simulator observations - Units 1 and 3
e
maintenance record review - Units 1, 2, and 3
e
surveillance record review - Units 1, 2, and 3
e
motor operated valve inspection trending - Units 1, 2, and 3
e
class lE alarm issues - Units 1, 2, and 3
e
performance evaluation plan review - Units 1, 2, and 3
e
industry experience review program - Units 1, 2, and 3
e
followup on previously identified items - Unit 3
e
review of licensee event reports - Unit 1
During this inspection the following inspection procedures were utilized:
37700, 37828, 40500, 41500, 42001, 60710, 61726, 62700, 62703, 62705, 71707,
90700, 92700, 92701 and 93702.
,
Results:
General Conclusions and Specific Findings:
Sianificant Safety Matters:
The condition of steam generator tubes at Palo
Verde and the licensee's ability to detect defects in time to prevent
catastrophic tube failure is of significant safety concern. The licensee's
eddy current testing during the previous outage had not detected any defects
in the tube which ruptured in Unit 2 on March 14, 1993.
Summary of Violations and Deviations: Of the 21 areas inspected, two cited
violations were identified.
One violation in Unit 2 involved the failure to
plug a defective steam generator tube. The second violation in Unit 1
involved the licensee's failure to follow procedures for completing
appropriate documentation prior to work package closecut.
Two non-cited
violations were identified. One involved the failure of a work group
supervisor to verify completion of work for a lost work order in Unit 1, as
required by procedures.
The other non-cited violation involved inadequate
procedures for operations and testing in Unit 3, resulting in a minor event.
Open Items:
Ten new followup items were opened, and three followup items
were closed.
Strengths:
The licensee's self assessment capability was found to be
effective in most respects. The licensee aggressively
monitored apparent primary-to-secondary leakage indications in
Unit 1, demonstrating appropriate management oversight. The
licensee also aggressively pursued steam generator U-tube
!
inspections based on problems identified during planned eddy
current testing.
In addition, an auxiliary operator was alert
to a local panel light for a safety-related breaker which
showed that the breaker was inoperable.
Weaknesses:
Poor work verification resulted in one cycle of operation with
a defective steam-generator U-tube.
2
'
,
. - .
-
-
.
-
. -
- -
.
---
--- -
-
n
-
_ _ _ _ _ _ - - - - _ _ _ - - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ . - - _ - _ _ _ ,
l
.'
'
,
. -
.
i
'
DETAILS
i
'
1.
Persons Contacted
The below listed technical and supervisory personnel were among those
contacted:
- R. Adney,
Plant Manager, Unit 3
- J. Bailey,
Director, Eite Technical Support
- R. Bernier,
Supervisor, Nuclear Regulatory Affairs, Technical
0#*R. Bouquot,
Supervisor, Quality Audits and Monitoring
- T. Bradish,
Manager, Nuclear Regulatory Affairs
OK. Chavet,
Supervisor, Independent Operating Experience Dept.
- R. Cherba,
Manager, Quality Systems
0#P. Coffin,
Engineer, Nuclear Regulatory Affairs
- R. Flood,
Plant Manager, Unit 2
- R. Fountain,
Supervisor, Quality Audits and Monitoring
,
- R. Fullmer,
Manager, Quality Audits and Monitoring
1
1
OD. Goodlet,
Senior Planner, U-1 Work Control
- D. Gouge,
Director, Plant Support
- B. Grabo,
Supervisor, Nuclear Regulatory Affairs
- D.
Hettick,
Supervisor, Station Operating Experience Department
i
- W. Ide,
Plant Manager, Unit 1
l
- D. Leech,
Supervisor, Quality Audits and Monitoring
l
- J. Levine,
Vice President, Nuclear Production
'
- J. LoCicero,
Senior Technical Specialist, Quality Systems
- T. Matlock,
Manager, Nuclear Safety
- D. Mauldin,
Director, Site Maintenance and Modifications -
OD. Odom,
Manager, Design & Document Control
- R. Prabhakar,
Manager, Independent Safety and Quality Engineering
OG. Reeves,
Technical Manager Assistant, Central Maintenance
- R. Roehler,
Senior Engineer, Nuclear. Regulatory Affairs
- R. Rouse,
Supervisor, Station Operating Experience Department
0#*C. Russo,
Manager, Quality Control
R. Schaller,
Assistant Plant Manager, Unit 1
J. Scott,
Assistant Plant Manager, Unit 3
T. Shriver,
Assistant Plant Manager, Unit 2
Others
- J. Draper,
Site Representative, Southern California Edison
- F. Gowers,
Site Representative, El Paso Electric
- R. Henry,
Site Representative, Salt River Project
0 Denotes personnel in attendance at the Exit meeting held with J. Melfi,
NRC resident inspector, and licensee management on April-16,1993.
i
Denotes personnel in attendance at.the Exit meetings held with D. Kirsch,
'
Technical Assistant, Region V, and licensee management on April 23, 1993.
l
1
-
_ - - _ _ - _ _ - _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ . - _ _ _ _ - - - _ _ - - _ - .
'
-
.
4
- Denotes personnel in attendance at the Exit meeting held with the NRC
resident inspectors on April 27, 1993.
The inspectors also talked with other licensee and contractor personnel
>
during the course of the inspection.
2.
Review of Plant Activities - Units 1. 2. and 3 (71707)
a.
Unit 1
Unit 1 operated at essentially 100% power throughout this inspection
-
period with the exception of a reduction to 80.5% power on March 25,
1993, which was required due to a loss of the Core Operating Limit
Supervisory System (COLSS). The COLSS was restored on March 25,
,
1993; however, power was further reduced to 70% on March 26, 1993,
'
to repair a speed probe in the "A" main feedwater pump turbine. The
plant was returned to 100% power on March 26, 1993. A second
failure of the COLSS occurred on April 24, 1993, but was of
sufficiently short duration that no power reduction was necessary.
b.
Unit 2
Unit 2 began the inspection period in Mode 6 in its fourth refueling
outage.
The unit commenced core off-loading on April 1,1993, and
completed the off-load on April 4,1993. During the outage, several
degraded steam generator U-tubes were discovered by eddy current
-
testing. Additionally, as of April 26, 1993, 26 snubbers had failed
'
functional testing.
Both of these conditions are described in
,
paragraphs 7 and 8 of this report.
c.
Unit 3
The unit operated at 100% power throughout most of the inspection
period.
On April 2,1993, a malfunction of the steam bypass control
'
system (SBCS) caused five of the eight valves to modulate open.
Reactor power increased to 103% before operators could respond and
reduce power (see paragraph 12). The unit remained at 100% power
until April 20, 1993, when a turbine trip occurred due to a high
!
vibration alarm from the turbine supervisory instrument (see
paragraph 11).
Reactor power was lowered to 12% while
troubleshooting the cause of the turbine trip. Reactor power was
increased to 100% on April 23, 1993. The unit operated at 100%
power for the remainder of the inspection period.
d.
Plant Tour
,
The following plant areas at Units 1, 2, and 3 were toured by the
inspector during the inspection.:
Auxiliary Building
Control Building
Diesel Generator Building
2
i
_
._
,
i
_
,
,.
.-
.
.
e
Fuel Building
e
Main Steam Support Structure
e
Radwaste Building
e
i
e
Turbine Builoing
e
Yard Area and Perimeter
e
Containment Building
The following areas were observed during the tours:
l
(1) Operatina Loas and Records - Records were reviewed against
Technical Specifications and administrative control procedure
3
requirements.
J
(2) Monitorina Instrumentation - Process instruments were observed
for correlation between channels and for conformance with
]
Technical Specifications requirements.
(3)
Shift Staffina - Control room and shift staffing were observed
for conformance with 10 CFR Part 50.54.(k), Technical
Specifications, and administrative procedures.
(4)
Eauipment Lineups - Various valves and electrical breakers were
verified to be in the position or condition required by
Technical Specifications and administrative procedures for the
applicable plant mode.
(5)
Eauipment Taqqina - Selected equipment, for which tagging
requests had been initiated, was observed to verify that. tags
were in place and the equipment was in the condition specified.
(6)
General Plant Eauipment Conditions - Plant equipment was
observed for indications of system leakage, improper
]
lubrication, or other conditions that could prevent the systems
a
from fulfilling their functional requirements.
-
(7)
Fire Protection - Fire fighting equipment and controls were
i
observed for conformance with Technical Specifications and
)
administrative procedures.
The inspector observed a firewatch being performed in the Unit
2 control building and found it to be adequate.
.!
(8)
Plant Chemistry - Chemical analysis results were reviewed for
1
conformance with Technical Specifications and administrative
control procedures.
(9)
Security - Activities observed for conformance with regulatory
requirements, implementation of the site security plan, and
administrative procedures included vehicle and personnel
access, and protected and vital area integrity.
3
.
.
-
_
__
_.
i
'
-
i
. . .
.
.
(10) Plant Housekeepina
Nant conditions and material / equipment
]
storage were observed to determine the general state of
'
cleanliness and housekeeping.
,
(11) Radiation Protection Controls - Areas observed included control
l
point operation, records of licensee's surveys within the
-'
radiological controlled' areas, posting of radiation and high
radiation areas, compliance with radiation exposure
permits, personnel monitoring devices being properly worn, and
.
personnel frisking practices.
'
(12) Shift Turnover - Shift turnovers and special evolution
briefings were observed for effectiveness and thoroughness.-
No violations of NRC requirements or deviations were identified.
3.
Surveillance Testina - Units 1 and 3 (61726)
-,
,
Selected surveillance tests required to be performed by the Technical
Specifications were reviewed on a sangling basis to verify that:
1) The
survaillance tests were correctly included on the facility schedule; 2)' A
technically adequate procedure existed for performance of the
surveillance tests; 3) The surveillance tests had been performed at the
frequency specified in the Technical Specifications;- and 4) Test results
satisfied acceptance criteria or were properly dispositioned.
I
Specifically, portions of the following surveillances were observed by
thrt inspector during this insper' ion period:
Unit 1
Procedure
Description
Adjustable Power Signal Calibrations
41ST-!RCO2
RCS Water Inventory Balance
Unit 3
Procedure
Description
Plant "rotection System Functional Test RPS/ESFAS Logic
18 Month Surveillance Test of Reactor. Trip Breakers
Failure of Reactor Trio Breaker to Open
en March 24, 1993, the Unit 3 "C" reactor trip breaker did not open when
the local test pushbutton was depressed during the performance of
t
32ST-9SB01, "18 Month Surveillance Test of Reactor Trip Breakers." The
licensee quarantined the breaker and developed a troubleshooting. plan.
Troubleshooting' revealed that the cause of the event was the breaker
position limit switch in the cubicle. As a result, the signal for the
breaker to trip did not reacn the treaker. The licensee replaced the
4
,
_ , +
-
---
,. .
.
.
limit switch. The breaker retest was satisfactory.
The inspector
witnessed the troubleshooting and agreed with the cause determined by the
licensee.
i
.
Lost Surveillance Test Packages
The inspector reviewed Condition Report / Disposition Request (CRDR) 3-2-
0600, documenting the licensee's evaluation of the loss of two Unit 3
ASME Section XI surveillance test packages.
Surveillance tests 73ST-
3X102, "Section XI Valve Stroke Timing & Position Indication Verification
- Mode 1 through 4 - Steam Generator No. 2 Containment' Isolation Valves,"
and 73ST-3XIO7, "Section XI Valve Stroke Timing & Position Indication -
.
Verification - Mode I through 4 - GA, GR, RD, SP and WC," were performed
as required d ring November 1992, during the refueling outage,.and were
reviewed and determined to be acceptable by Operations prior to Mode 4
entry on November 18, 1992. However, the surveillance test work order
(STWO) packages (STWO 568168 and STWO 568173) were lost sometime after
.
the review by the Section XI engineer. The Mode Change Checklist for
transition to Mode 4 was signed on the basis of confirmation with the
-
Section XI engineer that all required testing was satisfactorily
completed. Original surveillance test package review sheets retained by
Operations document that acceptance reviews were completed an the
results entered in tree Station Information Management System (SIMS)
,
database, supporting the Section XI engineer's conclusions. .After
discovery of the missing STW0s on December 22, 1992, the licensee
initiated the CRDR and reconstructed the surveillance package, using
copies of the surveillance test package review sheets retained by
operations following its reviews, and data retained by the Section XI
engineer after his reviews. The inspector concluded that the
surveillances were satisfactorily completed prior to the mode change, and
that the licensee actions in response to the lost test packages were
adequate.
'
The inspector reviewed the qualifications of the Section XI engineer and
found that he met the requirements of the UFSAR, the Technical
Specifications, and the licensee's job nosition description.
The licensee acknowledged in the CRDh
o. ' ion that the lost STW0s had
i
not been submitted to the Surveillance ,a
m m Control Group or to Design
Document Control, as required by procedure 73AC-9ZZ04, " Surveillance
i
Testing." However, the regenerated STW0s, with attendant documentation,
,
were submitted, satisfying the intent of this procedural requirement.
@ violations of NRC requirements or deviations were identified.
4.
Plant Maintenance - Units 1. 2 and 3 (37828. 60710, 62703. and 62705)
,
1
During the inspection period, the inspector observed and reviewed
selected documentation associated with maintenance and problem
investigation activities listed below to verify compliance with.
j
regulatory requirements, compliance with administrative and maintenance
i
procedures, required quality assurance / quality control department
j
5
l
-
.
.
. .
.-
A
.
.
involvement, proper use of safety tags, proper equipment alignment and
use of jumpers, personnel qualifications, and proper retesting.
The
inspector verified that reportability for these activities was correct.
Specifically, the inspector witnessed portions of the following
maintenance activities:
"
Unit 1
e
Close latch reset spring inspections on 4.16 KV safety-related
Magne-Blast breakers
Replacement of the "B" EDG essential spray pond jacket water cooler
'
e
temperature element SPR-TE-196
Unit 2
Installation of essential batteries (PK)
e
e
"B" containment spray pump mechanical seal replacement
e
Number 1 steam generator hand hole installation
,
e
Spray wash 525 KV insulators
e
Reactor vessel head removal
Unit 3
e
Adjust electrolyte level in PK "C" essential battery
e
"B" LPSI breaker troubleshooting
e
Install LDCP l?-LE-HP-047 for valve HPB-UV-002
The inspector observed installation of limited design change package
(LDCP) 3LE-HP-047, " Rewire Rotork Actuators to Open Containment
Isolation Actuation Signal Circuit Upon Limit Close," to 3HPB-UV-002
on April 13, 1993.
The inspector reviewed the LDCP and work order, and observed the
radiation protection brief, operations tailboard, installation of
the modification, and the retest of the valve.
The inspector concluded that the craft were knowledgeable and
followed the appropriate procedures, and there was good quality
control involvement.
However, the inspector observed that during
the operations tailbaard there was not a discussion concerning the
possible outcomes of incorrectly installing a jumper in ESFAS
cabinet 3JSABC01-02. The inspector concluded that the operators
were not as prepared for the evolution as they could have been and
that the operations tailboardiould have been more thorough.
No violations of NRC requirements or deviations were identified.
.
5.
Feedwater Isolation Valve Operability Determination - Unit 1 (61726)
The inspector reviewed the documents associated with surveillance test
73ST-lXIl6, "Section XI Valve Stroke Timing, Partial Stroke Exercise and
6
,
i
.:*
,
,
.
Position Indication Verification - Modes 1 thru 6 - FWlV's (Economizer)"
'
of feedwater isolation valve SGB-UV-132 performed on June 9,1992 and
questioned the operability determinations associated with this test.
The valve was initially declared inoperable and tested (slow closure) at
12:24 p.m. on June 9, 1992.
The valve did not move and therefore failed
the test. Operations personnel declared the valve operable at 12:31
p.m., apparently due to pressure being greater than 5000 psig in the
'
The accumulator is the motive force for the safety function
of the valve to fast close and the hydraulic pump is used to perform slow
'
stroke surveillance testing of the valve.
Based on noises heard and a leak from the air-driven hydraulic pump, the
pump was replaced even though it was able to recharge the accumulator to
above 5000 psig. The valve was again declared inoperable and tested at
3:56 p.m., and again the valve failed to move. The valve was declared
operable at 4:16 p.m., again apparently due to pressure being greater
than 5000 psig in the accumulator.
Troublest )oting of the testing circuit was then rerformed. The valve was
again declared inoperable and tested at 9:53 p.ro.; the valve again failed
the test, but was declared operable at 9:54 p.m. with sufficient
accumulator pressure. A fourth test was performed at 10:12 p.m., which
"
also failed.
The "M" four-way valve in the hydraulic system for the
valve was then replaced and the valve tested with satisfactory results.
The valve was declared operable at 1:50 a.m. on June 10, 1992.
The licensee conducted an informal investigation of these operability
'
determinations and concluded that the determination was prudent. At the
end of the inspection period, the inspector had additional questions
regarding the operability determinations made during this testing
sequence. This item will be unresolved (Unresolved Item 50-528/93-11-01)
No violations of NRC requir:ments or deviations were identified.
6.
Steam Generator U-Tube Mis-oluaaina - Unit 2 (61726)
While performing eddy current testing (ECT) of steam generator number ??
(SG 22) during the 1993 refueling outage (2R4), the licensee identified
that one plug was installed in the wrong tube on the hot leg side during
the previous refueling outage (2R3) in 1991.
The tube which should have
been plugged (R17-C152) had been determined by ECT to have a 70% defect.
Technical Specification 4.4.4.4.b requires tubes with 40% through-wall
indications to be plugged for the SG to be operable. The cold leg side
of the tube had been properly plugged. SG 22 had been declared operable
prior to Mode 4 entry on December 31, 1991. The licensee initiated
Material Nonconformance Report (MNCR) 93-RC-2019 to document the
nonconformance, and Condition Report / Disposition Request (CRDR) 2-3-0222
to document its evaluation of how the error was made.
Ineffective corrective actions in this case are of particular safet,
significance because of the substantial risk of catastrophic tube
7
i
.
.
--
- .
-
!
'
'
.
.
.
failure. Additionally, NRC Inspection Report 50-529/90-20 documents two
previous mis-plugging events in SG 22,.further indicating the need for
increased management attention in this area,
q
l
The failure to plug tube R17-C152 prior to declaring SG 22 operable is a
'
violation of NRC requirements (Violation 50-529/93-11-02).
In response to.this condition, the licensee reperformed ECT on tube R17-
'
C152 and determined the defect to be 52% of nominal wall thickness. The
licensee also planned to plug the hot leg end of the tube, and the cold
leg of tube R17-C158 which was inadvertently plugged on the hot leg end.
The licensee developed and conducted additional tube plugging
verification training for quality control inspectors, which the inspector
attended.
i
One violation of NRC requirements was identified.
7.
Steam Generator U-Tube Inspections - Unit 2 (61725).
Backaround
Unit 2 was forced to shut down on March 14, 1993, when a U-tube in steam
generator 22 (SG 22) failed. The unit was in Mode 1 (Power Operation)
operating at approximately 99 percent power when the U-tube ruptured._
The details of the steam generator tube rupture event were investigated
by an NRC Region V Augmented Inspection Team (AIT) and were reported in
NRC Inspection Report 50-529/93-14. The licensee subsequently discovered:
numerous axial indications in the steam generator U-tubes, mainly in
,
SG 22. Most of these indications were located in the upper regions of
the generators.
Each unit at Palo Verde has two vertical U-tube steam generators
constructed by Combustion Engineering. Each generator has 11,012 U-tubes
constructed of 0.75' inch OD NiCrfe alloy (Inconel 600) with 0.042 inch
i,
nominal wall thickness.
Ruptured Tube
A visual examination of the failed tube was performed using a video probe
'
from the inside of the tube. The examination revealed a " fish mouth"
shaped failure approximately 1.9 inches long in tube R117-C144 on the hot
!
1
leg side of the tube located approximately ten inches below the ninth
support (09H).
Eddy current testing revealed that the crack extended
,
several inches beyond the failure for ' total length of approximately
i
nine inches.
i
Eddy Current Testina
1
The licensee conducted eddy current examinations on 100 percent of the U-
j
tubes using a bobbin coil probe. Additionally, the licensee performed
eddy current testing at various locations of selected tubes using a
j
motorized rotating pancake coil (MRPC) probe, and plans to perform
8
.
!
i
,
1
Y
-e
.
.
.
additional MRPC testing. Eddy current testing using the bobbin coil
probe is the industry standard method used to satisfy . Technical
.
Specifica' tion surveillance requirements on steam generators. However,
i
the MRPC probe is used for detecting certain types of cracks and is being
used by the licensee to closely inspect susceptible tube locations.
.
Axial Indications
T
The licensee's eddy current testing revealed a significant number of
axial crack indications in SG 22. As of April 20, 1993, a total of 27
tubes in SG 22, including the ruptured tube, were discovered to have
axial indications.
Fourteen of these axial indications were larger than
the Technical Specifications plugging limit of 40% of nominal wall
thickness. Two of the tubes had indications that are greater than 80
percent of the nominal wall thickness.
Nineteen of the tubes in SG 22 had axial indications at a support
,
structure. The licensee is evaluating the indications to determine if
localized, chemistry-induced stress corrosion was a contributor to the
problem.
Six of the tubes, including the ruptured tube, had deposit
,
indications that were aligned with the axial indication. All but four of
the indications were at or above the 07H support. The two tubes with
indications greater than 80 percent are adjacent tubes with indications
-
that faced each other. Additionally, these tubes had the axial
indications aligned with indications of deposits on the tube surface.
,
Finally, most of the tubes with axial indications were located in an arc,
approximately 90 degrees long, near the outer edge of the tube bundle.
The indications were all in the hot legs, and most of them were near the
upper supports.
Local flow patterns and other characteristics were
included in the scope of the licensee's root cause of failure
i
investigation.
,
In contrast to SG 22, six axial indications were discovered in SG 21 of
which only one had indications larger than the plugging limit.
The other
five indications were not large enough to determine their depth.
An area of concern was the sensitivity of detecting flaws in the U-tubes
by eddy current testing with the bobbin coil. As of the end of this
inspection period, five axial indications had been detected by MRPC that
were not detected by bobbin coil.
,
Root Cause Investigation
l
The licensee formed a root cause investigation team.
The team is
investigating four failure modes: a) outer diameter stress corrosion
cracking (0DSCC); b) deposit induced lockup; c) cross flow; and d)
supports. ODSCC has oeen further broken down into material, stress, and
environment related failures. The licensee planned to use metallurgical
examination results from pulled U-tubes to help determine which of the
failure modes, or combinations of modes, caused the rupture and 'other
axial indications.
l
9
I
i
!
.
.
.
Tube Pulls
The licensee developed a list of eight tubes for pulling during this
outage in discussions with NRC personnel. The tubes were selected to
represent a variety of conditions in SG 22. Analysis of the pulled tubes
will be performed to help determine the root cause of the rupture and the
phenomena causing the axial indications.
The licensee considered pulling tubes from both the primary side and the
secondary side.
Because of the shape and size of the rupture, the actual
ruptured portion cannot be pulled from the primary side.
The " fish
mouth" shape would not fit through the support structures nor the tube
sheet. However, to pull the ruptured tube from the secondary side, the
.
licensee would have to " cull" approximately nine good tubes to gain
access to the failure and the effect on the stability of the remaining
tubes in the area has not been determined.
For these reasons, the
licensee plans to pull the ruptured tube, up to and including the cracks
that extend just below the rupture, from the primary side.
One of the tubes to be pulled is a tube that has an indication that was
not detected on the initial bobbin coil test.
It was detected using an
MRPC. Once the indication was noticed by MRPC, the bobbin coil data was
again reviewed and determined that there was indications by bobbin coil.
The data gained by pulling this tube is expected to help the licensee
determine the sensitivity of detecting cracks by bcbbin coil.
Conclusion
The licensee is conducting an aggressive root cause investigation into
the ruptured tew.
The axial indications may be caused by a number of
different factors or combinations. However, until the results from the
metallurgical analysis from the pulled U-tubes are known, the licensee
will be unable to determine the root cause of failure. The inspectors
and other NRC personnel will continue to monitor the licensee's root
cause of failure investigations and report the findings in a future
report. Additionally, Region V and NRR personnel initiated further
evaluation of the licensee's inspection activities, which will be
reported separately in NRC Inspection Report 50-529/93-21.
No violations of NRC requirements or deviations were identified.
8.
Snubber Testina - Unit 2 (62703)
The licensee has experienced a high failure rate of Unit 2 mechanical
snubbers during functional testing. performed during the current refueling
outage.
Functional testing of snubbers is required by Technical Specifications 4.7.9.
Unit 2 has 489 mechanical snubbers, of which 37
were to be tested during the current outage. The sample size was
increased to 235 snubbers as a result of the identified failures.
Prior
to the current testing failures, the licensee had only one snubber that
failed to pass functional testing in Unit 2.
10
_.
.
-
.
5
8
As of April 26, 1993, 26 mechanical snubbers had failed the functional
test required by Technical Specification 4.7.9.
All the snubbers that
have failed the functional testing were mechanical snubbers that were
manufactured by Pacific Scientific Applications (PSA).
Nineteen of the
26 failures were small snubbers. The licensee reported that 17 of the
small snubber failures were caused by the snubber being " locked up."
The
other seven failures were in the medium or large size snubbers, most of
which failed the test due to high drag forces.
A significant number of failed snubbers are associated with the auxiliary
feedwater (AFW) system. Eight small snubbers and one medium snubber
which failed the functional test were attached to the steam supply lines-
to the AFW pump turbine. Additionally, four snubbers passed the
functional test but exhibited signs of degradation and were replaced.
The licensee performed engineering analyses of the failed AFW system
These analyses concluded that the piping attached to the
failed snubbers did not exceed code allowable stress with those snubbers
locked and therefore did not affect operability. However, the licensee
has not yet completed the root cause of failure evaluation for all the
failed snubbers.
The inspectors noted that the due date for some of the
other engineering analyses was June 11, 1993. June 11 is after the
scheduled change in operational mode when the systems that had snubber
failures would be required to be operable. Technical Specifications
require that an engineering evaluation of the attached component be
performed within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> for systems required to be operable.
The
licensee acknowledged they were required to complete the engineering
analyses prior to the mode change.
The licensee's snubber test program, the causes of the failures and the
licensee's consideration of affect on Un'it I and 3 snubbers will be
reviewed by the inspectors and reported in a future inspection report
(Unresolved Item 50-529/93-11-03).
No violations of NRC requirements or deviations were identified.
..
9.
Leak Off Lines Not Designed to be Live loaded - Unit 3 (37700)
On April 21, 1993, during preparations to start the main turbine and
increase reactor power, shift personnel questioned whether the leak off
lines for the main steam isolation valves (MSIV) and the feedwater
isolation valves (FWIV) were properly designed.
Operations management suspended efforts to bring t% main turbine on line
'
and initiated MNCR 93-SG-305 to determine if the existing configuration
was satisfactory.
It was determined that after installation of live-
loaded packing in these valves per SMOD 03-SM-SG-024, the MSIV and FWIV
leak off lines were not modified to meet the requirements of ASME
specification 13-PN-204. The existing condition has a threaded coupling
attached to the line and a pipe plug at the end of the coupling. This
condition met the requirements for a non-pressure boundary condition, but
not a pressure boundary condition.
11
)
.
.
,
.
.
The inspector reviewed the results of the MNCR and the planned corrective
action. The engineering evaluation determined that the threaded
connection provided sufficient structural integrity to allow operation of
the valves until the next refueling outage. The evaluation also
concluded that the safety function of the valves was not affected.
The inspector concluded that the licensee's actions were appropriate and
consistent with the guidance in Generic Letter 91-18 for a nonconforming
condition. However, the failure to meet all applicable code requirements
while installing the modification was a weakness in the design control
process.
The licensee initiated CRDR 33-0202 to determine the root cause
of this error. This item will remain open pending a review of the
completed evaluation (Followup Item 50-530/93-11-04).
No violations of NRC requirements or deviations were identified.
10.
lPSI "B"
Breaker Silent Inoperabilit_y - Unit 3 (62703 and 93702)
On April 7,1993, at 12:11 p.m. a Unit 3 Auxiliary Operator (AO) noticed
that the white local panel light on the
"B" low pressure safety injection
pump breaker cubicle was not illuminated. This white light is
illuminated when control power is available and a limit switch is closed
signifying that the closing springs had reached the fully charged
position. The A0 visually inspected the breaker, noted that the closing
springs were not charged, and reported this condition to the control
room.
The licensee quarantined the breaker and developed a
troubleshooting plan. The breaker had been closed at 11:03 a.m. and
opened at 11:57 a.m.
The breaker was designed so the closing springs
would recharge following each closing operation.
The failure of the
closing springs to charge was not indicated in the control room.
The
breaker was inoperable for 14 minutes without operator awareness.
Troubleshooting revealed that the close latch spring coils were impeded
by the close latch monitoring switch mounting bracket which prevented the
switch from closing. This switch is designed to monitor the position of
the latch, and close when the latch is engaged to restrain the charging
springs wher they are charged. This problem in the breaker design
prevented the charging motor from charging the closing springs without
the close latch engaged to prevent inadvertent closing of the breaker
immediately after the springs were charged.
The breaker was a General Electric (GE) 4.16 KV model AM-4.16-250-91200
ampere Magne-Blast breaker with the model ML-13 mechanism. This
condition is generically applicable to any GE Magne-Blast breakers with
the model ML-13 mechanism with the older tension-style close latch
spring. Two different designs exist for the close latch spring. The
design which failed was the older tension-style spring which was
positioned beside the close latch monitoring switch.
A newer design has
the torsion spring wrapped around the close latch shaft. The tension
spring connected to a stationary member on the breaker frame, and to a
plate attached to the close latch shaft.
Both style springs pulled the
,
'
plate so it would depress the plunger of the close latch monitoring
12
i
'
,
.
.
switch to close the switch. This action rotated the clt
latch shaft
and engaged the close latch. The plate was attached to the close latch
shaft with a single screw. This single screw permitted the plate to
rotate, and this rotation affected the alignment of the tension spring
either away from or close to the switch mounting bracket.
In the Unit 3
"B" LPSI breaker, the spring was so close to the switch mounting bracket
that the coils of the spring caught on the switch mounting bracket which
prevented the plate from fully depressing the switch plunger. This
prevented the switch from closing, the close latch from fully engaging,
and rendered the breaker inoperable.
The licensee questioned the vendor regarding the spring design change.
The vendor answered that this was an enhancement and not a response to
any particular concern.
the endor also stated that there has not been a
history of problems of thu type with this breaker.
This as the fourth silent inoperability of a safety-related 4.16 KV
Magne-Blast breaker since the beginning of 1992. The three other events
are described in NRC Inspection Reports 50-529/92-05, 50-530/92-31, and
50-529/92-35. The licensee immediately issued night orders to direct
operators to visually observe the white local panel lights immediately
following any safety-related Magne-Blast breaker closing operation.
In
addition, the licensee visually inspected the close latch spring on all
safety-related 4.16 KV Magne-Blast breakers in all Units within a few
days of the event.
The licensee initiated Condition Report / Disposition
Request 3-3-0152 and is evaluating additional corrective actions.
The
inspector concluded that the A0 observation of the inoperable breaker was
appropriate and timely. The inspector further considered the
compensatory action issued via the Night Orders to have A0s verify
closing springs charged foilowing breaker closing operations to be
prudent. This item will remain open pending a review of the CRDR
3-3-0152 (Followup Item 50-530/93-11-05).
No violations of NRC requirements or deviations were identified.
11. Turbine Trio - Unit 3 (71707 and 93702)
On April 20,1993, Unit 3 was operating at 100% power with minor
maintenance being performed on the Turbine Supervisory Instrument (TSI).
The TSI trips had been disabled for the reinstallation of vibration cards-
after solder repairs to the test jacks. At 13:28, approximately 2
minutes after the vibration cards were reinstalled and the TSI trips were
enabled, a main turbine trip was received. A reactor power cutback
(RPCB) on a large load rejection occurred. Reactor power stabilized at
approximately 47% as a result of the RPCB and operation of the steam
bypass control system, and a downpower to 10% was commenced.
The cause of the turbine trip was determined to be a high vibration alarm
from the TSI #1 amplifier card. The system engineer tested the amplifier
card and could not determine the exact cause of the high output (greater
than 10 mils).
The card was replaced with a new version and the turbine
was successfully brought on line on April 21. The exact cause'of the
13
,
- - -
-
'
.
.
.
card failure was still beint
luated at the end of the inspection
period.
Operator response to the event was adequate, however, reactor trip
"
breaker "C" was manually tripped by the reactor operator (RO). The
inspector interviewed operations and training department personnel, and
reviewed procedures to determine if the R0's response to the event was
'
appropriate. The inspector noted that the unit operated without the RPCB
feature from late 1991 until February 4,1993. During this time,
operators were trained in the plant simulator to expect a plant trip on
high pressurizer pressure when a large load rejection occurred without a
RPCB .
The simulator software was subsequently changed to incorporate
the RPCB on April 20.
The inspector interviewed the crew on shift to determine why the R0 was
poised to manually trip the reactor if a RPCB was occurring and
pressurizer pressure was normal. At the end of the inspection period,
the inspector was unable to interview the R0 who was actually involved in
the event at the close of the inspection period. The inspector's
determination of whether the RO's action was consistent with procedures
isanunresolveditem(UnresolvedItem 50-530/93-11-06).
No violations of NRC requirements or deviations were identified.
12.
Inadvertent Opening of Multiole Steam Bypass Control Valves. Plant Power
Exceeded 100% - Unit 3 (93702)
On April 2,1993, five steam bypass control valves (SBCV) opened during a
retest of a single valve, causing instantaneous plant power to rise to
approximately 103%. Average plant power remained within license limits.
Plant operators were performing a retest of SBCV-1005, following vendor
recommended instructions, when the event occurred. The retest required
operators to quick open SBCV-1005 using the steam bypass control system
(SBCS) master controller test panel mode selector switch.
Operators took
the master controller to manual with zero output, then took the test
panel mode selector switch to the valve test position. This caused SBCVs
1001,1002,1003,1004, and 1006 to open inadvertently. A control board
operator took the SBCS to emergency off which shut the valves. The shift
supervisor directed operators to stop testing, returiNhe test panel mode
selector switch to the operate position, and restore the SBCS to
operation.
Subsequent troubleshooting revealed that Design Change Package (DCP)
13XJ-SF-31, which was recently implemented in all three units to add
modulation tracking to the SBCS, appeared to also produce this potential
for inadvertent multiple valve opening. According to Site Technical
Support personnel, the modification was developed by the SBCS vendor,
Combustion Engineering. While the test panel mode selector switch was
shifted from the operate to the valve test position, the modulation
tracking circuitry generated open signals to six SBCVs, SBCVs 1001-1006.
SBCV 1005 did not open because it was isolated as required by the retest
procedure.
14
_
,
.
.
.
'
The inspector noted four previous occurrences of inadvertent opening of
multiple SBCVs at Palo Verde. The first occurred on January 9,1986.
The second occurred on September 11, 1986, and was reported in Licensee
EventReport(LER) 50-528/86-53. The third occurred on March 3, 1989.
The fourth occurred on October 20, 1990, and was reported in NRC
Inspection Report 50-530/90-45.
Following the event on October 20, 1990,
the licensee issued a Justification for Continued Operation (JCO),
performed additional analysis, and conducted a failure modes and effects
analysis (FMEA) to identify all single components which, upon failure,
,
could cause inadvertent opening of more than one SBCV. A modification to
,
separate power supplies was implemented by DCP 13XJ-SF-30 and 13XJ-SF-32
in all 3 units as a result of this FMEA.
The inspector concluded that the immediate operator action to terminate
the event by placing the SBCS in emergency off was prompt and
'
appropriate. The inspector further concluded that DCP 13XJ-SF-31 was not
thoroughly reviewed for all potential effects, and the licensee review
prior to implementation of the DCP did not provide a backstop to prevent
,
this event. The licensee initiated condition report / disposition request
(CRDR) 3-3-0143 which is scheduled to be completed on May 28, 1993.
This
item will remain unresolved pending a review of CRDR 3-3-0143 and a-
review of how this event occurred despite the DCP was initiated as a
result of the FMEA performed followin the previous event on October 20,
1990 (Unresolved Item 50-530/93-11-07 .
'
No violations of NRC requirements or deviations were identified.
13.
Evaluation of Licensee Self-Assessment Capability - Units 1. 2 and 3
(405001
The inspector examined the licensee's self-assessment capability by
examining the activities of the Quality Assurance and Monitoring (QA&M),
Quality Systems, Quality Control (QC), and Independent Safety
Engineering / Quality Engineering organizations.
In that regard, the inspector reviewed the following procedures:
QA Organization and Responsibility Policy (610G-02201)
e
Quality Auditing (62DP-0QQ01)
e
Conduct of Monitoring (62DP-0QQll)
e
Indoctrination and Qualification of Independent Safety and
Quality Engineering (ISEG) Personnel (64DP-0TR01)
'
e
ISE Assessment Performance and Reporting (64DP-0QQ17)
ISE Assessment Administration (64DP-0QQ18)
e
e
ISE Assessments (64DP-0QQ02)
Condition. Reporting (90AC-OlP04)
e
The inspector concluded that the above procedures provide adequate
program definition and administrativo controls for the subject oversight
activities.
15
.
-
-
-
l
.
4
a.
Quality Auditing and Monitoring
The inspector examined this oversight activity by reviewing eight
audits conducted by the QA&M organization, interviewing the manager,
and reviewing documents providing the manager's assessments and
impressions to the Executive Vice President.
The inspector found the audits to be substantial examinations of the
audited areas which resulted in significant findings.
The findings
were referred to appropriate organizations for resolution and
tracked to and verified at completion. Feedback was candidly
provided to the Executive Vice President regarding the problems
found by QA&M and the candid assessments of the QA&M manager.
b.
Performance Assessment Process
Palo Verde had recently (April 1993) instituted a process for
performing Performance Assessments of PVNGS activities.
The process
periodically analyzes conditions adverse to quality for trends and
reports the results to management. The process is controlled by
procedure 60AC-0QQ20, "PVNGS Performance Assessment Process," dated
March 31, 1993.
The process provides for a team of five representatives from
different PVNGS organizations, at the manager level, or above, to
analyze deficiency data from Nuclear Production, Nuclear Engineering
and Projects, Nuclear Safety, QA, and Regulatory and Industry
Affairs organizations. The deficiency data is reviewed by the team
and a consensus reached regarding whether the activity being
asseised demonstrates strengths, no significant deficiencies, or
significant weaknesses which deserve awareness by management,
management attention, or prompt management action.
To date, two quarterly reports have been issued (4th QTR 92, and 1st
QTR 93). These reports, in the form of annunciator window
representations, are color coded for quick visual determination of
activity assessment conclusions and trends.
The inspector considers this to be a useful and substantial program.
The results of the first two assessment reports were in agreement
with NRC perceptions of activity effectiveness in several areas.
Thi.; process represents a significant initiative by the licensee to
provide oversight of wide range of activities and provide the
results in an easily understood manner.
c.
Qualit_y Control
The inspector examined reports of Quality Control (QC) activities to
senior management and several QC organization initiatives to improve
performance.
16
-
. - .
_
-.-.
-.-
--
.
-
-
-
--
,
,
.
.
The inspector found the monthly reports of QC Inspection and
'
Observation results identified significant findings and actions
taken regarding the findings.
The reports trended the findings by
activity and organization responsible and provided the trends to
management. The reports also highlighted generic issues, where
findings would indicate more broad based concerns in particular
areas, for management review and information.
Individual QC reports
-
Issued for particular inspections were found to be sufficiently
'
detailed and issued in a timely manner, with problem reports sent to
'
the offending work group and cognizant manager for correction.
QC has developed a detailed listing of 18 attributes for use by
inspectors during their inspection, from initial tailboard to
-
restoration after completion of the work activity, including
examination criteria for each attribute. The inspector reviewed
this attribute listing and found it to be comprehensive and
substantial. The QC reports document the use of the
attribute / criteria during the individual inspections.
Supervisory oversight of QC activities occurs in two ways.
First,
the level III QC personnel periodically conduct oversight
observations of QC inspectors performing inspection work to assess
the effectiveness of QC training in the area being inspected and the
effectiveness of the QC inspector / inspection. Secondly, supervisors
are tasked with providing personal oversight of QC inspectors in the
field on a frequency of about 6 field obser vations per month.
The
results of the oversight observations are documented and, based upon
the inspector's review, appear to be substantial in content.
'
QC had developed a listing of jobs which have exhibited a history of
problems and conducts pre-job briefings of the work groups for those
of jobs. The pre-job briefing content is pre-planned, approved by
the QC Manager and discusses the problems experienced, job
expectations, and techniques for problem avoidance.
QC had also developed a Briefing Log wherein, for each major problem
identified in the field, a briefing form describing the problem and
lessons learned is generated e.nd put into the log. The log is a
required QC reading file to assure the staff is aware of major
problems and issues.
d.
Assessment of Engineerina Performance
The inspector observed the Quality Assurance audits identified that
improvement was needed in the areas of modification impact reviews
and post-modification testing. The inspector interviewed licensee
engineering and assessment staff personnel regarding activities to
assess engineering performance improvement.
The licensee's incident investigation staff had recently completed
(January 1993) an assessment of the Nuclear Engineering and Projects
Organization, as requested by the Vice President, Nuclear
17
.
..
-
.
~. -
. - ~ . . - .
. - - -
.
---
.
.
i
.
..
a
.
.
'
Engineering (NE). The assessment was an organizational and
programmatic assessment, and resulted in the identification of
improvements needed in the NE self-improvement and root cause
culture, interface and communications with other organizations and
within NE, and the need for more managers in several areas. The
,
licensee is engaged in effecting corrective actions.
The licensee has implemented other actions to improve engineering
,
performance.
These are: (1) a rotation pro
between the site and corporate office; (2) gram for design engineers
separating system
performance engineers from the system engineers to improve
i
performance monitoring, tracking, and trending; and (3) instituting
the practice of formal system performance assessments.
i
i
System performance assessments are in place for certain critical
systems and the licensee plans to expand the program to other
systems. System Performance monitoring, tracking, and trending
programs are in place for critical systems and the licensee plans to
!
'
include the o'ther systems by June 1993.
The licensee's staff was in the process of specifying corrective
actions for the recent audit findings regarding the need to improve
modification impact reviews and post-modification in testing.
,
!
The Nuclear Engineering organization had recently instituted the
practice of documenting and correcting all human performance
i
deficiencies using the CRDR process. This has resulted in an
increase in the number of CRDRs; however, NE anticipates that this
will facilitate an objective assessment of error sources and
corrective action evaluation and assignment.
The inspector concluded that the licensee was proactively engaged in
,
the assessment of engineering performance, identifying deficiencies
l
'
and providing needed corrective actions.
e.
Independent Safety Enaineerina Group (ISEG) Activities
,
'
The inspector examined ISEG activity reports for the first 3 months
of 1993 and the 1992 annual activity report. The inspector
concluded that ISEG reviews were thorough, in-depth reviews covering
several functional areas.
Recommendations are tracked to closure
and verified at closure.
ISEG reports are submitted to senior
management for analysis and action.
The inspector exam'ined the experience and training of the ISEG staff
,
and concluded that the staff consisted of well experienced, degreed
engineers with several years nuclear experience in a variety of
areas.
Each of the staff had attended a root cause analysis
'
training course and technical staff training consisting of a wide
range of Palo Verde specific topics. One individual, however, was
new and in the training process.
18
)
--pi
- , - -
-
1m-y-
-.-,-m
--
gy
y-m--
e-
n
- -
e
-
-
.-.
.
--
.-.
-
.
. - -
.
-
-
.- -
--.
.
l
.
A
The ISEG staff consists of S engineers and I supervisor, a level
sufficient to comply with regulatory requirements. The inspector
noted that the staffing level was lower than other ISEG
organizations in Region V.
However, all the functions generally
performed by an ISEG staff were being done at Palo Verde, in some
cases by other organizations.
The ISEG organizatit,a had an assessment schedule with three
assessments scheduled for 1993. However, the organization appears
to be highly reactive in taking on emergent work due to problems or
management requests and the schedule appears to be designed to
accommodate reactive work. For example, ISEG was currently engaged
in an assessment of procedure adherence at Palo Verde.
An ISEG engineer was assigned to each unit, to attend morning
'
,
planning meetings and evaluate, for example, the impact of removal
of safety system from service and the most significant activities
occurring in the units so that monitoring activities could be
focused to the highest safety significant activities.
The ISE manager had engaged the services of an independent
consultaat to assess ISE work practices and precedures. The
consultant's report was recently completed and was being provided
for senior management review.
The inspector observed that the
consultant's findings and conclusions closely paralleled the
inspector's.
ISEG management acknowledged the need to better
uti.lize trending information to focus their limited staff on those
areas exhibiting the most serious and safety significant problem
history.
f.
Condition Report / Disposition Reauest Proaram (CRDR)
During examination of the auditing program, the inspector became
concerned regarding the high number of CRDRs which apparently were
'
not evaluated properly. The licensee's audits indicated that about
20% of the CRDRs were not adequately evaluated and that about 10% of
the closed CRDRs had been closed without all of the specified
corrective actions having been completed.
The inspector learned that a comprehensive assessment of the program
had not been done, nor was one planned. The inspector questioned
licensee representatives regarding the action taken to improve the
'
evaluations of the CRDRs and learned that CRDR program
representatives had discussed the expectations regarding evaluations
with the people generating the CRDRs and their supervision; however,
they had not discussed the organizational problem sources with
senior management to obtain support and make management aware of the
situation.
Further, the inspector's discussions with licensee
personnel identified that there were pockets of resistance to the
licensee's problem identification ' philosophy in the field; a
circumstance which apparently had not been brought to the attention
of senior management. These findings were of concern to the
19
.
w
w
.i.
y
a.-
-
-p
_.
. _ _ _
..
_ ._ _
_
_ _ .
_
._ -
_.
__
_
_ _.
l
.
.
.
,
'
inspector because senior licensee management apparently had not been
made aware of this problem and was not able to take effective
corrective actions to assure that their expectations were clearly
understood and effectively implemented by the Palo Verde staff. The
manager of Quality Auditing and Monitoring was, however, quite
candid in sharing impressions of organizational effectiveness with
>
the Executive VP in periodic communications.
The inspector
expressed these concerns to senior utility management who indicated
that actions would be taken to assure that their expectations were
clear and implemented in the field. This item will remain open
pending review of licensee actions (Followup Item 50-528/93-11-12).
g.
Trendino Program
'
The inspector discussed the licensee's trending programs and
,
practice with each oversight department manager.
It was apparent
that the program was being used to make senior management aware of
general problem sources. The inspector found that the trending
e
program could be made more useful by providing the ability to
identify which organizations had the largest numbers of problems
(identified by QA, CRDRs, and QC inspections, etc.) identified in
their work activity and identifying which work activities resulted
in an inordinate number of problems. However, the inspector
observed that, with little modification, the program would be able
to focus upon organizations experiencing the most problems and upon
the activities responsible for a large number of problems. This
knowledge would allow the licensee the opportunity to use their
overr,ight resources more effectively by focussing upon those
organizations and activities in need of the greatest improvement.
The licensee agreed and will evaluate improvements to the trending
program.
No violations of NRC requirements or deviations were identified.
14.
Simulator Observations - Units 1 and 3 (41500 and 42001)
The inspector observed two simulator examination scenarios conducted by
licensee instructors.
Unit 1 - March 30. 1993
This scenario was an excess steam demand event.
The Operations Manager
observed the scenario and participated in the critique.
The control room
>
supervisor (CRS) exhibited excellent command and control.
Communications
were generally good, with a few instances of incomplete two-way
communication. The status and direction announcements to the operating
crew in the control room were frequent and informative,
No similar
announcements were made to the auxiliary operators.
Two operator weaknesses were observed by the inspector which were not
discussed by the licensee during the critique. The first involved the
secondary operator (50) reporting to the CRS that steam generator levels
l
.
20
..
,
,
,
,
were unexplainably diverging even though they were being fed and steamed
at the same rate.
The CRS appropriately pointed out that this was not
true, with one main steam isolation valve and one steam bypass control
valve stuck open, one steam generator was indeed steaming at a
substantially higher rate than the other.
Both of these-events had
s
occurred and were announced to the crew prior to this discussion.
The
second involved the 50 demonstrating cpparent unfamiliarity with the
safety equipment status system (SESS) in that one 50 action caused SESS
.
alarms without any announcement by the 50, while later in the scenario
the 50 announced that another action would cause SESS alarms when indeed
-
'
it did not and should not in that circumstance.
In addition to acknowledging the fir:t weakness, the Operations Manager
responded to the first weakness by noting that the crew concept
functioned in this instance in that a deficiency by one operator was
raised, corrected, and the operators continued to work together. The
Operator Training Manager responded to both weaknesses by stating that
-
during this exam cycle, he and his staff identified instructor weaknesses
,
in observing SESS and secondary plant issues.
To address these
weaknesses, the Operator Training Manager has scheduled additional
training for instructors to address secondary plant operation and the
SESS.
'
Unit 3 - April 2.1993
-This scenario was a steam generator tube rupture event. The Operations
Supervisor was assigned as shift supervisor. The Operations Manager
.
observed the scenario, but did not attend.the critique.
Communication
was adequate with a few instances of incomplete two-way communication.
-
The inspector observed that there were no status and direction
announcements to the control room operators nor the auxiliary operators.
,
After the control room supervisor entered the functional reccvery
procedure, the shift supervisor and control room supervisor focused on
the procedure for approximately 20 minutes. The inspector concluded that
'
either the functional recovery procedure was too detailed, or operators
were not sufficiently familiar with it to use it effectively.
One simulator fidelity issue was noted during each observed scenario.
The first issue was that all charging pumps tripped off simultaneously on
low suction pressure yet in the plant, they trip off at different times
and this provides some warning which can enable operators to take action
to prevent the loss of all charging pumps. The second issue was that the
simulator did not implement a recent change to operate with one steam
bypass control valve in off. The licensee responded by evaluating the
first fidelity issue, and immediately implementing the second. The
inspector concluded that the training conducted was effective. The
inspector further concluded that except as noted, the instructors were
-j
reasonably thorough in their evaluation, and appropriately critical of
operator weaknesses. During the critique the instructors demonstrated a
good understanding of NUREG 1021, " Operator Licensing Examiner
_
21
,
1
-
.;-
,-
. -
.
~.
._
_
-.
-
.
..
,
.
.
1
Standards," Revision 7.
In addition to the specific responses noted, the
licensee acknowledged the inspector's comments.
No violations of f4RC requirements or deviations were identified.
i
15. Maintenance Record Review - Units 1. 2. and 3 (62700)
The inspector reviewed recent, completed maintenance packages in safety-
related systems that the licensee archived between January and April,
1993.
The inspector reviewed the maintenance packages to assess
e
The level of detail in the work orders,
e
The attention to detail in tha work orders.
If there was a work order change of intent without a quality review.
e
If the work order followed the licensee's administrative processes.
If post-maintenance tests were adequate.
,
The inspector reviewed 21 packages and found the following discrepancies:
Theinspectornotedthatworkorder(WO) 00328044 for Unit I was lost and
j
the work reverified. The inspector noted that this work order was not
completed in accordance with procedure 30DP-9WP02, " Work Order
'
Development."
Step 3.9 of this procedure states that lost W0s shall be
verified by a Work Group Supervisor (WGS). This procedure was not
,
follcwed since the planner for this work order verified completion of
r
this work. The licensee committed to revise this procedure by June 30,
1993, to allow planners to verify work on lost work orders.
Based on the
limited safety significance, an apparent isolated case and the licensee's
proposed corrective actions, this violation is not being cited since the
criteria specified in Section VII.B of the Enforcement Policy were
satisfied (f4CV 50-528/93-11-08) .
The inspector found a completed WO (00365041) which had a blank
restoration step.
The W0 had been signed off as complete by the Work
'
Group Supervisor. This WO replaced a Hancock valve with a Kerotest
valve.
Step 6.4 of this work order, which specified updating the valve
designation lists, was left blank. Further licensee review indicated
that the list was not updated. This is contrary to licensee procedure
30DP-9MP01, " Conduct of Maintenance." This procedure, step 3.10.2 states
"The WGS indicates, by signature, proper completion of the work and work
package documentation." This package was completed.
This is a violation
of NRC requirements (Violation 50-528/93-11-09).
Of the records reviewed the inspector noted that the timeliness from
completion of work to archiving of records was generally less than 4
months. However, for five of the twenty-one work orders reviewed, the
time between completion of work and archiving was between 10 and 36
months.
The inspector also reviewed documents associated with maintenance in
.
Unit 2 during the previous refueling outage.
These records were
associated with reactor cavity sump level indicators and rigging in the
22
,
_
.
.
_
_ . _ - -
_
;
,
.
.
,
vicinity of the reactor vessel head. The inspector identified an issue
involving rigging reactor vessel head insulation which warranted
additional discussion. The load path for these insulation pieces, as
described in maintenance procedure 31MT-9RC01, was directly over the top
of the reactor vessel head. Since another load path not directly over
the vessel was available, it would have been a more conservative load
path.
This is, nonetheless, not a requirement since the regulations
address heavy loads as defined in NUREG 0612 and the insulation pieces
weigh less than a fuel assembly and are, therefore, not heavy loads. The
,
licensee acknowledged this comment.
One cited violation and one non-cited violation of NRC requirements were
identified.
16.
Surveillance Record Review - Units 1. 2. and 3 (62700)
-
The inspector reviewed completed surveillance test (ST) rccords required
for various Technical Specifications. The inspector reviewed 12 recently
completed STs.
The inspector chose his sample based on Technical
Specification frequency requirement that had long surveillance intervals
(greater than or equal to 18 months) and the inspector reviewed these STs
for completeness and adequacy in meeting the Technical Specifications.
i
Overall, the ST documents were completed appropriately. The inspector
only noted one item during this review involving the Technical
Specification requirement to test the pressurizer relief valves for the
inservice test (IST) program.
The last completed surveillance procedure
referenced in the Technical Specifications / procedure cross reference was
done four years ago. This appeared inconsistent with the IST code
frequency requirement. The licensee verified that this was the last test
-
performed; but they were sending all the relief valves offsite for
testing every refueling. The licensee provided the inspector with a
table showing when they tested the relief valves for all units. The
inspector determined that'they met the frequency requirements of the IST
code.
The inspector noted that the Technical Specifications / procedures
cross reference was not up-to-date, and the licensee acknowledged the
inspector's comment.
The inspector also reviewed documents associated with surveillances in
Unit 2 during the previous refueling outage. These records were
associated with the loose parts monitoring system, reactor coolant pump
speed probes, and atmospheric dump valves. The inspector identified an
issue in Work Order 502668 involving the reactor sump level indicators
not reading correctly in the control room.
Engineering Evaluation
'
Request 91-RD-009 evaluated this issue and determined that there was a
fixed offset between the actual sump level and the indicated sump level,
and recommended a design change to correct the discrepancy. This design
change was ultimately canceled. This condition did not represent a
safety concern because the actual level was not needed for any safety-
related plant activity.
The calibration required the indicator to track
as level was changed in the sump, but did not compare actual sump level
versus indicated level. The sump levels were only used in the emergency
23
i
-
-
-
,
~.
-
...
.
-
.
.
. . .
- -
.
' ' '
,
.
,
operating procedures to evaluate the trend of the sump level. Actual
sump level was never needed.
This-condition appeared to meet the minimum
regulatory requirements. However, the inspector noted that when the
design change to correct the problem was proposed, operators supported it
with Unit 2 operators particularly emphatic.
In addition, this condition
did not appear to be consistent with NUREG 0700 guidelines. The
inspector concluded based on a review of the documentation, that all the
work reviewed appeared to have been appropriately performed. The
licensee acknowledged the inspector's comments.
No violations or deviations were identified.
17. Motor Operated Valve Inspection Trendino - Units 1. 2. and 3 (62703)
.
The inspector reviewed the licensee's procedure for trending the
condition of grease in Motor-Operated Valve (MOV) operators.
The inspector noted that condition of grease in MOV operators was checked-
on an 18 month frequency using procedure 39MT-9ZZ02, "PM/EQPM Inspection
of GL 89-10 Limitorque SMB/SB Valve Motor Operators." This procedure
requires inspection of the quantity, quality, and consistency of the
- '
grease in the main housing cavity.
Appendix "D" of the procedure
contains detailed guidance on how to inspect the condition of the grease.
The inspector verified that the condition of grease in a M0V was not'
'
reportable in the Nuclear Plant Reliability Data System (NPRDS) unless
'
the condition of the grease caused the valve operator to fail. -However,
the inspector r.oted that the condition of the grease was not currently
trended to evaluate the adequacy of. the inspection frequency. The M0V
Phase One inspection also identified the trending of MOV inspection
results as a weakness in NRC Inspection Report 50-528/91-25. The-
licensee was aware of this shortfall and recently revised-the MOV program.
.
instruction, 39PR-9ZZ01 dated March 6,1993, to include trending of
various MOV parameters including the condition of the gear box grease.
The performance monitoring group was responsible for the trending of
various parameters and making appropriate recommendations based on the
results.
-
The inspector concluded that the licensee was appropriately inspecting
,
and documenting the condition of grease in MOVs. The inspector
considered the formation of a dedicated engineering group to conduct
periodic trending of MOV inspection data a positive initiative deserving-
of strong management support in order to meet-the requirements of NRC Generic Letter 89-10.
No violations of NRC requirements or deviations were identified.
18. Class IE Alarm Issues - Units 1. 2. and 3 (71707)
On April 19, 1993, the inspector noted that the Unit 2 Class 1E control
- room alarms could be silenced using the flasher reset pushbutton.
Further inspection revealed that this was also true in Unit 3, but not in .
24
.
m
, . - ~
g
,
>w-
m-
,
. . _ _ _ - .
. _ . _ _ _ .
__ _ _ _ _
_ . .
_
_
_
..
.
.> .
,
.
,
Unit 1.
Updated Final Safety Analysis Report (UFSAR), section 7.6.2.1.3,
states that for the Class lE alarm system, "The audible alarms for each
'
channel must be silenced with the use of a key which is under
administrative control."
The reason for this problem was traced to Engineering Evaluation Request
(EER) 90-RK-005 which addressed another problem with the Class IE alarm
system.
On March 28, 1990, Unit 3 operators noted that a particular
sequence of key presses could set up the Class IE alarm system so that
the next annunciation condition would illuminate the alarm window, but
would not sound the audible alarm. The EER evaluated this and determined
that a c;rcuit change would eliminate the potential for this condition.
The EER was converted to a design equivalent change and implemented in
Unit 2 and 3.
This change also altered the system such that the audible
alarm could be silenced without the key lock switch.
Since this change was not implemented in Unit 1, it was still susceptible
to an alarm condition without audible indication. The licensee initiated
Condition Report / Disposition Request (CRDR) 2-3-0233 as a result of the
apparent difference between the UFSAR description and alarm function.
In
addition, the licensee is evaluating the situation in Unit 1.
This issue
will remain unresolved pending a review of CRDR 2-3-0233 ind licensee
actions regarding Unit 1 (Unresolved item 50-528/93-11-10).
No violations of NRC requirements or deviations were identified.
19.
Performance Evaluation Plan Review - Units 1. 2. and 3 (71707)
The inspector reviewed the licensee's perforrrance evaluation plan (PEP)
to ensure that performance targets do not adversely impact the use of
,
self-identification programs such as quality deficiency reports (QDRs)
i
and condition reports / disposition requests (CRDRs).
'
,
The inspector noted that in the spring of 1992, the licensee developed
and implemented a new training course on target setting. Targets are
used in the performance evaluation plan to measure an individual's
performance in various key areas. As part of the training program,
targets which referred to the number of QDRs written against the
individual were used as examples in the training class. This was
'
identified by the licensee as a problem because using the number of QDRs
as a performance target may be viewed by personnel as an incentive not to
,
identify problems.
To prevent this perception, the licensee corrected
the example, sent letters to all the employees who had received the
training with the inappropriate target as an example, and sent out PEP
update #5 on January 14, 1993. The PEP update expanded the subject to
include all self-identification programs, including CRDRs.
The inspector reviewed a sample of PEPS to verify that.the number of
CRDRs was not being used as a performance target. The inspector
concluded that the PEPS reviewed did not inappropriately use the number
1
of CRDRs as a performance target. However, the inspector noted that the
quality of evaluating CRDRs. and the timeliness of resolving issues was
25
P
,..
,
_ _ . , _ . .
m
_.. ,
.
,
.
. . .
,
,
.
.
sometimes used as a performance target. The inspector concluded that
these targets were appropriate because they stressed the importance of
excellence in these areas.
The inspector concluded that senior management is effectively
communicating their expectations that PEP targets not discourage the
generation of QDRs and CRDRs to correct self-identified problems.
No violations of NRC requirements or deviations were identified.
'
20.
Industry Experience Review Proaram-- Units 1. 2. and 3 (90700)
Three Mile Island (TMI) action plan item I.C.5 required nuclear utilities
to establish programs that would review and apply industry experience to
their power plant.
Industry experience includes data from the following
NRC and Institute of Nuclear Power Operations (INPO) documents:
INPO Significant Operating Experience Reports (50ERs)
e
e
INP0 Significant Event Notification (SENs)
INP0 Significant Event Reports (SERs)
e
NRC Information Notices (ins)
'
NRC Generic Letters (GLs)
e
NRC Bulletins (NRCBs)
The licensee's program also included vendor bulletins.
The inspector
assessed the licensee's program, its effectiveness in reviewing industry
experience and disseminating this experience to plant staff.
Program
The licensee's industry operating experience review program was described
'
in Updated Final Safety Analysis Report (UFSAR), section 18.I.C.5, and in
procedure 95DP-0NS01, " Industry Operating Experience Review Program
Department Instruction." The licensee's Industry Affairs Group reviews
the industry information.
The licensee's program requires Industry
Affairs to screen industry _ information (within 10 days), determine
'
applicability and significance, assign a priority and a response due
'
date, and assign the information to the affected groups. UFSAR, section
18.1.C.5, further notes that the licensee participates in INP0
Significant Event Evaluation and Information Network (SEE-IN). The
licensee's program also requires tracking of the notices and evaluation
of the response.
The inspector found that the licensee screened information notices within
ten working days of receipt of the notice.
Since the SEE-IN database is
used, significant items are screened before there is an NRC or INP0
document issued.
Sta f fing
The industry affairs group staff had four engineers, one engineering
assistant, one supervisor and a manager. The inspector considered that
26
.
-
-
. -
_ .
_
. ..
~.
- ,
[
. . .
.
.
this group had a diverse background, adequate to assess industry
information.
Effectiveness Reviews
Licensee procedure 95DP-0NS01 stated that the Industry Affairs Group will
have periodic (every 12 months) effectiveness reviews of their program.
,
These effectiveness reviews are performed by different organizations
about every 12 months.
The last QA audit (Audit Report 92-006, July 23,1992) found strengths
and weaknesses in the program. These weaknesses were generally
administrative in nature or observations that the procedure was not
,
consistently implemented.
Information Notice Review
The inspector assessed the licensee's evaluation of 24 NRC Information
Notices (Ins 85-20, 85-20 Supplement 1, 91-41, 91-46, 91-50, 91-51,
91-54, 91-55, 91-56, 91-73, 91-75, 91-82, 91-85, 92-04, 92-16, 92-16
Supplement 1, 92-27, 92-30, 92-42, 92-45, 92-48, 92-49, 92-64, 92-67,
92-68, and 92-86), which were generally between 5 and 22 months old. The
majority of the Information Notices were either acceptably evaluated or
were not applicable. The inspector had comments on six of the licensee
'
evaluations:
a.
IN 85-20 and IN 85-20. Supplement 1. " Motor-0perated Valve Failures
Due to Hammerina Effect" These ins documented valve failures due to
the torque switch resetting in the Motor-0perated Valves (MOV)
actuator when the motor circuitry incorporated a seal-in feature.
In the example discussed, the motor energized and the valve closed.
The motor stopped when the torque switch opened. The torque on the-
valve would relax and the torque switch would reset. With the seal-
.
in feature present, the motor would reenergize, driving the' valve
further closed.
While this evaluation discussed Limitorque manufactured valves, this
situation would probably apply to other valve manufacturers (e.g.,
Rotork). The licensee's evaluation did not include other valve
manufacturers.
In response, the licensee stated that it is addressing this effect
for Rotork and other manufacturers as part of its response to NRC
Generic letter 89-10. However, this IN was issued four years
earlier than this generic letter.
b.
IN 91-54. " Foreign Experience Regardina Boron Dilution" This IN
involved a foreign safety analysis scenario that could lead to a
boron dilution. The specific accident sequence begins with a
partial loss of offsite power during dilution of the reactor coolant
system (RCS). The reactor coolant pumps (RCPs) are lost when
offsite power is lost, but dilution water continues to inject. .The
i
27
,
_ _ - _ _ _ _ _ .
-
.
'.
.
/ ,O
ls
s
s
,
-
'
esp
///
,, %
,,
' ',
,jjpg/
O
IMAGE EVALUATION
}////
.
'~
/
' [.+
/g
~
TEST TARGET (MT-3)
g3/ gp'
k?
i[
1.0
F"
' ;. 2 2
,-
l,l
'
w . _s =<
$! 1.8
uxa
i 1.25
1.4
l
1.6
l
---
i m=
m _s
4-
- ._.. . _ -. -- -- - - 1 5 0 m m
>
<
--
- . - - _
. . _.
6"
---
- - - -
>
A
8>)y
/+ por
<go
c
c7
- ;;
-
(. 4x
_,
.
@fz
- c
,*
O ,-
pg
.
q.
s
'
'
- 't;
~, '~
--
-
- ^
-
_ , , _ ,
,__,._,_y
_ _ _ ,
p't>
&
4
/O
c
>
w-
,, &.
s.
IMAGE EVALUATION
. '+h
- ,
,
Gh?'
\\/,O
TEST TARGET (MT-3)
/2F4;^g,
/
/'
'5
'
gy ;p\\>,s
lh//g,/,C:*j*ge
Afb
.
'
s
y
g
-
Tk
,
(4
1.0
~
"
"
~
~
w
t
n=ev
m
V--
)
t
l,l
_._=
<
lbl=1.8
!
=
l.25
1.4
11'
,
-
yl_i.6_
4-
-
150mm
>
4_..__.
_ . . _ _ . _ . . . . _ . . - - - -
-
-
6"
>
3
%,4
+ + ,Ao
.
A
eqy
,, 7
,
p, 3., ,,//s
_
.,
o
,
f;n?
c
,
.
"
p
. fy,
r
- .a
,
O
,f h "#De
1
,,
,
l
.j
J
be
euasaa:els!
_ , _ , i_ t -
j
s
_
-
- - - - _ ,
-
- Na
.A
.6
/.o
-
a' , . ,"
,;, e
fe
i.,
7"
IMAGE EVALUATION
- b,
O
O
g.[
4,
'
[
ff[y(g*
////
-
4)h*
TEST TARGET (MT-3)
.
s-
~ , '
'
N
.
q ,,
1.0
.: "
-
.
,
l
? rh_t
u
L
r 2
lA
I,l
!=
m
'! l.8
\\ ===
1.25
1.4
1.6
__.
4
. _ . _ _ _ _ _ _ . . - -
150mm
>
4_.
._ _ __
-
6"
>
zh
'h
y
46 %, p@p
spl*/,,,;O
hiY
ty
4
d* x
t#p
3,47
-
- , 4. y d f /4
.
w g / ,, g-
w
,
Oy j
qb; <ge
R&
4f
.
,
,
L
d
!
w.
._
..
-,
l
,e
y, " g
0
e
- v
4,
...
/I,
. . -
/
[ ,jp%* 44,
IMAGE EVALUATION
g [Of/
'
' h'
TEST TARGET (MT-3)
.
q q;;*
%q ;j;f
'"
"
1.0
.
L l
ll6 2.2
- um;
,u
Y
^
l,l
ne_ e.
ill I.8
!!!as
1.25
l
1.4
!jl 1.6
i
a__
w --__
_
150mm
>
4.._.-
_
. _ . _ _ _ _ _ . - - -
6"
>
es
b'
4 ,-
.
,, n. g.
s;g
,,
I ,
,
, et.,
' #[i
'C$
Nh'
l
'
t
/
cp
}
U
-: ,
lh0c,,a&::, N Nbt* Y
?
_ _ . ,
_
,
-
~ ^ ~ - ~
w
g
.
__
.
_ _ _ _ .
_ _ _ -
_ _ _ - _ _ . . .
!
3
,
- Y
y
.m.
'
3
,-
E
reactor' coolant does not adequately mix due to a lack of convective
~
g
. cooling. A volume of undiluted water builds in the RCS cold legs.
Subsequent startup of the RCPs pushes this. undiluted water through
the core, possibly leading to an inadvertent criticality or core
damage.
The licensee' canceled its review of this IN based on'INP0 SER 90-13,
which discussed actual dilution events at foreign reactors. The
inspoctor's- review of INPO SER 90-13 did not identify where this
sequence was' discussed. The licensee did not evaluate the sequence-
mentioned in this IN.
After questioned by the inspector, the licensee found that this item
r
-
was being addressed in their review of shutdown. risk. The. licensee
subsequently found engineering evaluation request (EER) 91-RX-018,.
which included a safety analysis evaluating the condition of the RCS
boron concentration less than cold shutdown concentration during
cooldown following a loss of coolant accident.
CE analysis stated
that the reactor would not go critical even if all the water in the
pressurizer went through the core. Based on this, the licensee does
,
not consider 'that this problem sequence is applicable to PVNGS.
c.
IN 91-73. " Loss of Shutdown Coolina (SDC) Durino Disassemb1v of Hiah
Pressure Safety Iniection (SI) System Check Valve" This IN involved
the loss of SDC during RCS drain down due to improper coordination.
'
between operations and maintenance. .The work control process
-
to be changed without the appropriate
allowed this maintenance' evaluation took credit for INP0 SER 91-19
reviews. The~ licensee's
and' closed this IN.
Generally, the licensee off-loads fuel before
going to a midloJp condition. There is no requirement to assure
that the licensee would always do this,
i
The administrative barriers in place to prevent poor coordination
between groups was not easily discerned in the licensee's
'
.
evaluation. 'The inspector discussed this issue with the fall outage
manager and found that the licensee has administrative barriers in
l
place to prevent-a similar occurrence.
.j
'
d.
IN 91-85. " Potential failures of Thermostatic Control Valves for
Diesel Generator Jacket Coolina Water" This IN involved.the failure
-
of a 3-way the mostatic control valve due to intergranular stress
corrosion cracking (IGSCC). This IGSCC failed two out of the~four
" power elements" in this valve, allowing the internal. fluid to leak
out and the valve to fail partially open. This failure occurred 12
?-
,
years into a nominal 15-year design life.
The licensee determined it had a different model number, bet a
F
similar design to the one described in IN 91-85. The licensee noted
that it did not have any failures, but .that this. element was not on
a Preventive Maintenance (PM) schedule. The licensee established a
-,
PM with a-replacement interval of 10 years, the maximum allowed by
-
,
28
j
.;
i
l
2
.-
.
..
.
.-
--
_
-..
' ,
.
the vendor.
The licensee further noted that this failure could lead
to a valid engine failure on an emergency start.
The inspector questioned why placing the replacement interval at the
vendor's maximum was acceptable, since the similar design failed
before its design life was over. Subsequent discussions with the
system engineer revealed that this issue was considered by the
Cooper-Bessemer diesel owners group. This industry group now has
this temperature control valve replaced every five years, and the
licensee has implemented this recommendation. The system engineer
did not recall any failures associated with this component.
e.
IN 92-27. " Thermally Induced Accelerated Agina and Failure of
ITE/G0ULD A.C. Relavs used in Safetv-Related Applications" This IN
involved specific relay failures due to heat. These relays
'
generated excessive heat since they were normally energized and
mounted in a horizontal " ganged" arrangement. The relays that
failed were in the middle of this " ganged" arrangement and failed
within seven years.
The licensee previously evaluated an INP0 document (SEN 86) on
February 7,1992, on the same issue and found that this relay was
not used in safety-related applications. The licensee's
documentation for IN 92-27 noted that this relay was within the
controller for the motor-driven fire pump. The motor-driven fire
pump was not safety-related, but was considered by the licensee to
be quality-related (or augmented quality). The licensee's
documentation did not show why the motor-driven fire pump was
The inspector found that the licensee had previously determined that
the relays are normally deenergized and are not in the " ganged"
arrangement mentioned in the IN. The licensee also informed the
inspector that these relays are tested when the pump is tested and
were on a preventive maintenance interval to be replaced.
Based on this new information, the inspector concluded that the
licensee conclusions were appropriate and that the motor-driven fire
pump was operable.
f.
IN 92-42. " Fraudulent Bolts in Seismically Designed Walls" This IN
involved fraudulent anchor bolts and through-wall bolts in seismic
structures. These bolts did not go through the walls or through
angle iron that provided lateral support to seismic structures.
These walls were built in 1973 by a sub-contractor to the architect-
engineering design firm.
That licensee eventually concluded that
some of the walls were not capable of performing their design
function, and shutdown to repair these walls.
The licensee's evaluation concluded that Palo Verde did not have the
problem since there were no corrective action documents similar to
this problem, and the licensee oversaw construction activities of
29
-.
5
a
>
J a3ia
-
an
=
-44-
- - - - >
i._V
.n4~
m.en.A4
_2_.g.
.
p.
-4t._a+4
a
-
a
yA
2As
-
t
c
.o ,
,
,
,
.
!
their architect-engineering firm. Their architect engineering firm-
did not sub-contract this work.
!
The inspector evaluated the thoroughness of the licensee's
evaluation. The licensee performed additional records review and
1
identified that Deficiency Evaluation Report (DER) 83-53 addressed a
similar concern-at PVNGS. At that time, the licensee inspected
approximately 38,000 plates and bolts in the units. The licensee
found only four plates that were defective, which indicates that the
installation of anchor bolts and plates was adequately controlled.
The industry affairs group added DER 83-53 to their evaluation
package.
l
Conclusions
The inspector found the licensee's operating experience review program
met TMI action plan item I.C.5 and the UFSAR. Tne licensee processed and
screened these notices within ten days of receipt of the information
notice. The licensee's program was evaluated about every 12 months. The
,
'
licensee also actively screened new industry events to assure a rapid
,
evaluation on more significant items.
]
With respect to the quality of the licensee's evaluation of the
information notices, the majority were satisfactory. The inspector found
several information notice evaluations that did not initially appear
,
adequate, though subsequent information provided by the licensee
H
established that they were acceptable.
For some of these information
notices, the licensee closed out the information notice early, based on
earlier, similar INP0 documents. This led the licensee to overlook some
relevant points that were not in the INP0' documents.
No violations of NRC requirements or deviations were identified..
21. Followup on Previously Identified Items - Unit 3 (92701)
(Closed) Followup Item 50-530/92-41-0L "Refuelina Water Level Indication
Error" - Unit 3 (92701)
This item resulted from a November 2,1992, event during which refueling
'
water level increased to about one inch below the refueling pool skimmers
during performance of safety injection system Section XI valve testing.
Performance of the test normally increases refueling. pool level
approximately three feet. The inspector reviewed Condition
Report / Disposition Request (CRDR) 3-2-0480, which documents the
licensee's evaluation.
,
1
The licensee determined that a mismatch developed between actual
refueling pool level and the level indicated by the refueling water level
i
indication system (RWLIS) in the control room. This occurred because the
'
pressurizer safety valves- had been reinstalled, so that the pressurizer
was no longer vented. With the filling of the pressurizer and
compression of the air space in the top of the pressurizer, increased
30
)
.
.
.
-.
.
- -
-
. __
_.
.- .
.
-
.
.. .
'
.
,
pressure was sensed by part of the RWLIS, resulting in a falsely low
indication. When the pressurizer was vented, the indication reflected
actual level conditions.
During this event, an auxiliary operator called the control room three
times to report the unexpectedly high refueling pool level.
Additionally, the containment coordinator and an outage services
representative also called from containment to report the conditions.
Upon receipt of these reports, control room operators verified the RWLIS
indications, noting that they were consistent with each other (both
channels vent to the pressurizer). Additionally, pressurizer level
indicated as the operators expected. The operators did not believe the
,
accuracy of the visual reports, did not question their indications, and
did not stop the testing evolution until the level was one inch below the
skimmers. Once the problem with the level indication was accepted, the
appropriate actions were taken by the control room staff.
The CRDR documents that the most significant aspect of this event was
that the control room operators rejected multiple reports which
contradicted their instrumentation indications, and did not resolve the
disparity before proceeding with the test. However, the inspector
interviewed the shift supervisor involved in the event and learned that
the control room operators thought the auxiliary operator who made the
reports was confused about the reference points within the refueling
cavity.
The auxiliary operator was relatively inexperienced, and the
levels he reported correlated with control room instrumentation if he was
confusing the skimmers with the pool cooling return line. The control
room operators contacted the containment coordinator to get another
confirmation, and the evolution was stopped when the report was received,
confirming the auxiliary operator's reports.
The licensee determined that several procedures were deficient in not
requiring the pressurizer to be vented. These included 43ST-35116,
"Section XI HPSI Pump Test," 400P-9PC02, " Filling and Draining the
Refueling Pool using the CS, LPSI, and HPSI Pumps," and 400P-9PC06, " Fuel
Pool Cleanup and Transfer."
Despite the operator actions not being addressed in the CRDR evaluation,
the inspector concurred with the CRDR conclusions, as the operators
should have been more aggressive in resolving the reported discrepancy,
and the procedures were not appropriate for the conditions, as they did
not require the pressurizer to be vented. The inspector further noted
that the error would have been much more significant if it involved a
draining evolution, as level would have been lower than indicated,
possibly resulting in a loss of shutdown cooling. The inspector
determined that the CRDR evaluation was deficient in not addressing
actions operators took to verify actual refueling pool level.
Additionally, the CRDR referenced procedure 40AC-9PC02, which is non-
existent.
As corrective actions, the licensee modified their procedures to require
that the pressurizer be vented. The inspector reviewed these procedural
0
31
--
.
__ -- .
~
-
_.
..
.
.
.
.-
-
-
-
- *K
e
.
.
changes and found them adequate. This event was discussed in " Industry
Events" training for operators and technical staff, and the operating
crew involved conducted a self-critique of their effectiveness as a team
during the event. The I' mnsee also stated that the CRDR would be
revised, as appropriate.
Procedure 40AC-90P02, " Conduct of Shift Operations," required that when
i
the desired or expected results are not achieved after the completion of
a step, or when there is reason to believe that performance of a
procedure step will damage equipment, result in an unsafe condition, or
!
is not correct, operators are to stop the evolution or step and resolve
the problem. The inspector observed that the operator's response to the
reports from containment were not fully consistent with their guidance,
in that the evolution was not stopped to resolve the discrepancy until
confirmation was received that a discrepancy existed.
Until confirmation
was received, however, the operations were other than expected. The
inspector considered that the operator's decision to continue the
-
evolution despite the uncertainty was poor, particularly as immediate
confirmation was not obtained.
10 CFR Part 50, Appendix B, Criterion V, requires that activities
.
!
affecting quality be described by procedures appropriate to the
circumstances. The failure of the licensee to have procedures
.
'
appropriate to the circumstances is a violation of NRC requirements. The
licensee-identified violation is not being cited because the criteria
i
specified in Section VII.B. of the Enforcement Policy were satisfied (NCV
50-530/93-11-11).
!
The licensee acknowledged the inspector's comments.
One non-cited violation of NRC requirements was identified.
22. Review of Licensee Event Reports (LER) - Unit 1 (92700)
t
Through direct observations, discussion with licensee personnel, or
review of the records, the following LERs were closed.
Unit 1
(1)
93-01,
Revision LO
" Turbine Trip / Reactor Trip on Moisture
Separator Reheater High Level"
(2)
93-03,
Revision LO
" Loss of Power to Train "A" Class 1E
,
4.16 Kv Bus"
!
No violations of NRC requirements or deviations were identified.
F
32
.
_
_
.
-
. . .
..
-
23.
Exit Meetina_171707)
An exit meeting was held on April 16, 1993, with licensee management and
J. Melfi during which the observations and conclusions of his inspection
were discussed.
An exit meetings was held on April 23, 1993, with licensee management
oversight groups and D. Kirsch during which the observations and
conclusions of his inspection were discussed.
>
An exit meeting was held on April 27, 1993, with licensee management and
resident inspectors during which the observations and conclusions in this
report were discussed.
The licensee did not identify as proprietary any materials provided to or
reviewed by the inspectors during the inspection.
33
.