ML20134P507
| ML20134P507 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 02/06/1997 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20134P501 | List: |
| References | |
| 50-341-96-13, NUDOCS 9702260006 | |
| Download: ML20134P507 (23) | |
See also: IR 05000341/1996013
Text
, ,
. . ;
I
U.S. NUCLEAR REGULATORY COMMISSION
l
'
REGION 3
Docket No: 50-341
License No: NPF-43
Report No: 50-341/96013
Licensee: Detroit Edison Company (DECO) j
Facility: Enrico Fermi, Unit 2
Location: 6400 N. Dixie Hwy.
Newport, MI 48166 '
!
Dates: October 26 through December 16,1996
Inspectors: A. Vegel, Senior Resident inspector
C. O'Keefe, Resident inspector
S. Stasek, Senior Resident inspector, Davis-
Bessie
A. Kugler, Fermi 2 Project Manager, NRR
Approved by: Mike Jordan, Chief, Branch 5
Division of Reactor Projects
_
9702260006 970206
PDR ADOCK 05000341
0 PDR
i
,
,
1
EXECUTIVE SUMMARY
Enrico Fermi, Unit 2
NRC inspection Report 50-341/96013 ;
l
This inspection included aspects of licensee operations, engineering, maintenance, and l
plant support. The report covers a 7-week period of resident inspection, in addition, it
includes the results of an announced 10 CFR 50.59 inspection by the Fermi 2 Project !
Manager.
1
Ooerations
e Performance of operations activities during plant startup, shutdown, and refueling
activities were well-controlled. Communications and coordination were good (01.2,
01.4). In contrast, a number of routine evolutions were not properly controlled and
errors were made which resulted in one scram and overflowing the spent fuel pool !
(O4.1) A non-cited violation was issued for the performance related to the scram.
'
e Shutdown cooling was lost for 24 minutes when the suction valve shut and the
running pump tripped. The exact cause was not determined, but was believed to
be an invalid spurious trip (O2.2). ;
l
l
Maintenance
l
e The licensee identified two cases of lifted leads that were not relanded following
maintenance. Both cases were found by workers but not reported or corrected
until post maintenance checks. In one case, the investigation did not identify the
cause. The inspectors determined the surveillance procedure was inadequate to I
verify proper system restoration (M1.2). A Notice of Violation was issued.
e inspectors noted several material condition issues which were not corrected during I
the outage due to improper planning. Also, a number of instances of inadequate
preventive maintenance were discussed (M1.3). i
l
e The control center emergency makeup air filter overheated due to
improper / inadequate maintenance (M2.1). The investigation was unable to
determine how the heater came to be set 100F too high. Response to the event
was proper, but the filter had to be replaced. Valves needed to deluge the filter
were found incorrectly labelled.
e Two surveillances were started without meeting the plant conditions to complete
the tests. Reviews by schedulers and operators were inadequate (M3.1).
2
j
- .
Enaineerina
l
1
e The control rod position indication system upgrade project and the low pressure l
turbine replacement and steam path upgrade project were carefully planned and
executed with few problems. Coordination for both projects was good. Both have 4
yet to be tested completely, however (E2.1, E2.2).
e Safety Relief Valve setpoint testing indicated significant drift. Evaluations showed
the plant would not have met its design basis and could have exceeded a safety
limit. A modification to reduce the susceptibility to set point drift was implemented i
during the outage (E2.3).
l
l
e The inspectors identified informal and untimely corrective actions for a problem
with the EDG 12 muffler. The issue was improperly assigned a low significance
and closed. Corrective actions were rescheduled without reassessing the I
operability determination (E2.4). A Notice of Violation was issued. I
e The inspectors identified that a technical specification (TS) change was required as
a result of the core reload analysis. Existing rod block monitor (RBM) operability ;
requirements were not adequate to prevent exceeding fuel mechanical overpower l
limits during a rod withdrawal error event, but the 50.59 evaluation did not
determine a TS change was necessary (E3.1). A Notice of Violation was issued.
e The 10 CFR 50.59 program was generally adequate. However, a safety evaluation
for the impact of using the Emergency Equipment Cooling Water (EECW) System to
supplement drywell cooling during extended hot weather did not consider all
applicable scenarios (E3.2).
4
l
l
j
._ - -. _-- __ _ - _ _ - _ . _ . -- . . . . - . .-..
. .
,
,
,
Report Details i
i l
Summarv of Plant Status
.
Unit 2 began this inspection period shutdown for its fifth refueling outage (RFO5). The
. plant was started up on November 29, but shutdown to investigate an indication problem j
with Safety Relief Valve (SRV) "A" on December 2. The plant was started up again on 1
December 7, but shutdown the following day bacause the SRV "A" indication problem
recurred and remained shutdown through the close of the inspection penod. ;
i l
!
1. Operations
01 Conduct of Operations i
!
, O 1.1 General Comments (71707)
- I
! J
'
Using Inspection Procedure 71707, the inspectors conducted frequent reviews of.
? ongoing plant operations, inspectors noted a continued contrast in operator
teamwork and performance between high-visibility operations and routine
l
operations. Several self-revealing events which were caused by operator errors
'
3
occurred during routine operations. This disparity has been documented in prior
inspection reports and the most recent SALP report. Specific events and notewor-
thy observations are detailed in the sections below.
l
.
01.2 Startuo and Shutdown Observations
,
inspectors observed the first partial startup and both shutdowns during this period.
The inspectors observed good teamwork and mostly formal communications.
Evolutions were performed in accordance with procedures. Briefings and the use of
past experience were a particular noted strength. The startup schedule was
detailed, enabling timely support of post maintenance testing (PMT) and evolutions
l by the entire site organization; this was an improvement for the last forced outage
startup.
01.4 Refuelina Activities (60710)
!
Throughout the outage, inspectors observed refuel floor activities and fuel moves.
Procedure adherence, communications, and formality were assessed. Radiation
, worker practices, safety practices, and foreign material exclusion controls were also
observed.
"
Refueling bridge operations were formal and performed according to procedure,
- with good communications on the bridge and with the control room. Over 1100
i core alteration steps were performed without error. Fuel sipping was performed to
identify the two fuel bundles with smallleaks which had been suspected, based on
testing during the operating cycle.
4
4
i
.
. .
In contrast to a nearly flawless performance during core alterations, production
wwk on the refueling floor had several miscues, including removing the wrong
reactor pressure vessel thermocouples, and the previous'y discussed overflowing
the SFP, and tripping both Fuel Pool Cooling Pumps while filling MSLs
simultaneously. Radworker practices were performed in accordance with licensee
requirements, and Radiation Protection personnel provided good support.
O2 Operational Status of Facilities and Equipment
O2.1 Enaineered Safety Feature System Walkdowns (71707)
The inspectors used inspection Procedure 71707 to walk down accessible portions
of the following ESF systems:
- Divisions 1 and 2 Residual Heat Removal
e Emergency Diesel Generators 11,12,13 and 14
e Divisions 1 and 2 Core Spray Systems
Equipment operability, material condition, and housekeeping were acceptable in all
cases. Several minor discrepancies were brought to the licensee's attention and
were corrected. The inspectors identified no substantive concerns as a result of
these walkdowns.
While inspecting the top of the torus, inspectors identified that several test plugs
used to perform leak rate testing of the downcomer seal bellows were missing.
Inservice Testing personnel were able to produce a memorandum which identified
these caps as not being a primary containment boundary, but ensuring they were
installed was identified as a good practice. As a result of this finding, the licensee
was planning to change the surveillance for checking test connections at the end of
an outage to add these connections.
During the same inspection, inspectors identified an example of insufficient space
between the Scram Discharge Instrument Volume Drain Line and a fire header,
severalloose electrical junction box cover screws, and severalloose pipe hangers
for drywell pneumatics. These deficiencies were reported to the licensee. The first
was assessed to be acceptable as is, the others were corrected.
02.2 ESF Actuation Renort on Loss of Shutdown Coolina
On December 3, with the plant in cold shutdown at 157F, shutdown cooling (SDC)
was lost when the shutdown cooling suction valve (E11-F009) closed
unexpectedly, tripping the running Residual Heat Removal (RHR) Pump C. The
licensee believed the trip was due to an invalid reactor high pressure isolation signal
though a specific cause was not determined. Shutdown cooling flow was restored
after 24 minutes with no temperature rise in the reactor. This event was reported
per 10 CFR 50.72.b(2)(ii), as en ESF Actuation to the NRC Operations Center as a
four hour report. The inspectors identified no violations of NRC requirements.
5
.
. ,
Corrective actions and the root cause determination results for this event will be
reviewed under Licensee Event Report (LER)96-020.
04 Operator Knowledge and Performance
04.1 Ooerator Errors
04.1.1 ESF Actuation Reoort for Unintentional Scram While Shutdown
On December 4, following completion of a control rod interlock surveillance, a
licensed operator moved the Reactor Mode Switch from " Refuel * to " Shutdown."
This initiated an expected manual scram signal. The operator reset the scram logic,
then went to get the key to bypass the High Scram Discharge Instrument Volume
(SDIV) Level scram logic. Before the logic was bypassed, however, the SDIV filled
and initiated an unexpected scram on high SDIV water level. This event was
reponed per 10 CFR 50.72.b(2)(ii), as an ESF Actuation to the NRC Operations
Center as a four hour report.
The inspectors reviewed applicable procedures and determined that Abnormal
Operating Procedure 20.000.21, " Reactor Scram," Revision 38, required that the
operator verify that a Scram Discharge Volume Level High alarm was received, then
the Scram Discharge Volume High Water Level Scram, and then reset the scram.
By performing this out of sequence, a second scram resulted. Failuro to follow
procedures was a violation. (50-341/96013-01). Corrective actions for this event
will be followed up under LER 96-021.
04.1.2 Miscositioned Valve Results in Overflowina the Soent Fuel Pool
a. insoection Scone (92901)
The inspectors reviewed the events surrounding overflowing the Spent Fuel Pool
(SFP). Administrative controls .9nd training for valve operations by refueling floor
workers were discussed with Refueling Floor supervision and Operations
management. Surveys of the spill and other documentation were reviewed.
b. Observations and Findinas
On October 30, the licensee overflowed the SFP. The normal method of filling the
SFP, a manual valve located in a covered recess on the refueling floor, was found
two turns open. Connecting to the same supply line was a hose connection used
by the refueling crew for decontamination.
Because the valve pit was iriside the contaminated area, Operations had permitted
deconners to operate the valve used for the hose after obtaining control room
permission. All refueling crew personnel were trained on this arrangement, and
were instructed that the larger (manual SFP fill) valve was not required to be open
to get water from the hose.
6
-- . _., - - .. - -. .- - - - -
' '
! i
.
l
!
. A short time before the SFP overflowed, deconners were unable to obtain water
from the hose and asked permission to open the P11-F015 rnanual fill valve. l
Permission vras denied. Operators then realized that no water was available to the '
Reactor Building fifth floor due to heavy plant usage of the water supply, so a
- second pump was started. Approximately a half hour later, the Fuel Pool Cooling
- Trouble Alarm was received. This alarm had five inputs which could cause the
1
alarm, one of which was SFP Skimmer Surge Tank High Level. Because no SFP
Skimmer Surge Tank water level indication was available in the control room, an l
operator was dispatched to the Reactor Building. He identified that water had filled I
- the surge tanks, then filled the SFP until it overflowed into the ventilation ducts, j
. just above the normal SFP water level. This resulted in spills on floors below as 1
water dripped out of the ducts, so access to the Reactor Building was secured.
Inspection and decontamination efforts permitted restoring access over the next
i seven hours. Contamination levels up to 110K DPM were detected, and the water
,
spilled was estimated to be 100-200 gallons.
A licensee investigation was unable to determine how the manual fill valve was
- repositioned. The licensee suspected that the contract deconner who was
attempting to use the hose and had requested permission to open the manual fill
valve, had actually opened it before requesting permission. However, the individual
'
denied doing no. Security performed an investigation to determine if wrongdoing
a was involved, but no evidence of malicious intent was found.
Also, the licensee determined that the SFP High Level Alarm was never received as
it should have been. The level switches had not been calibrated in about two
5
years. The switches were calibrated, and the periodicity of the preventive
maintenance event was reduced.
Corrective ac: ions included stopping all refueling floor work and reemphasized j
administrative controls on valve operations to all workers assigned to that area. ,
The licensee also initiated an engineering request to install water level indication for
the SFP in the control room, and were considering additional measures to avoid
accidental filling of the SFP.
There were no personnel contaminations during this event.
c. Conclusions
immediate cc.rrective actions for this event were adequate. However, because a
responsible individual could not be identified and a reason for the valve being
mispositioned determined, more substantial corrective actions were delayed.
Following discussions with the inspectors, Operations recognized that SFP system
design contributed to this event, because operators must respond to the refueling
floor to determine the cause of the alarm. These switches were not TS equipment
and no NRC requirements were violated. However, the failure to obtain permission
to operate the valves is a failure to follow procedures. (VIO) (50-341/96013-02)
7
. .
. . j
08 Miscellaneous Operations issues (92700) ,
l
,
08.1 Institute Of Nuclear Power Ooerations (INPO) Evaluation Revien
The inspectors reviewed the June 1996 INPO evaluation of Fermi 2. The inspectors
determined that no safety issues requiring NRC followup were identified.
II. Maintenance
M1 Conduct of Maintenance i
i
M 1.1 General Comments l
l
Attention to detail was an area of concern in maintenance work, improperly landed i
leads and inconclusive surveillance documentation are discussed in M1.2. An
improperly set charcoal heater following maintenance on a safety system is
discussed in M2.1. Specific failures which led to these problems could not be
determined by the licensee, which limited corrective actions.
a. Insoection Scope (62703)
The inspectors observed all or portions of the following work activities:
- EDG 11 Loss of Offsite Power / Loss of Coolant Accident Surveillance I
- EDG 11 Voltage Regulator Troubleshooting
- EDG 14 Loss of Offsite Power / Loss of Coolant Accident Surveillance
- Rod Worth Minimizer Operability Surveillance
- Position Indication Probe (PIP)/ Cable Changeout Work
- PIP Post-Modification Testing
- SFP High Level Alarm Calibration
- Reactor Moisture Separator Installation
- Emergency Equipment Cooling Water / Service Water (EECW/SW) Cross-tie
Check Valve Leak Rate Testing
- Work Activities Associated with Installation of Freeze Protection for both
Divisions of the Ultimate Heat Sink
- Division 1 Control Air Compressor Capacity Test
- Hydraulic Control Unit Pressure and Level Switch Calibrations
- Scram Solenoid Pilot Valve Replacements
- Ultimate Heat Sink Cross-tie Valve (E1150-F601 A) Repairs
b. Observations and Findinas
Maintenance work observed was performed in a professional manner. In most
cases, procedures were available and followed. However, the following issues
were identified by the inspectors and discussed with senior licensee management:
- Inspectors noted that the source of hot water input to the northwest
Emergency Core Cooling System (ECCS) Sump (D073), which was
8
i
I
.
I
.
,
discussed in Inspection Report (IR) 96007, was not specifically identified
'
prior to plant shutdown. As a result, the wrong Reactor Water Cleanup
(RWCU) system valves were worked during the outage. When the plant was l
started up, leakage out of the RWCU was about 15 gpm by control room I
indication. A licensee inspection of the RWCU system was then performed, !
and the source of the leakage was identified. The correct valves were i
repaired following plant shutdown. l
l
e A leaking Reactor Coolant Isolation Cooling (RCIC) system gland exhaust l
hose was not replaced during the outage because the wrong part was i
ordered for the second time. The leak was first identified in January 1996.
Following shutdown from the aborted startup, the issue was again raised by
Operations, and priority was placed on obtaining the correct hose. The hose
was ordered and installed in three days.
- Inspectors identified three instances where radiologically controlled vacuum
cleaners located adjacent to work areas had hoses crossing contamination
area boundaries which were not properly secured to prevent the spread of
contamination. In each case, Radiation Protection initiated appropriate
corrective actions.
M1.2 Lifted Lead issues
a. Insoection Scoce (92902)
The inspectors investigated an issue that involved lifted electricalleads on safety
related equipment which were not relanded following work. The inspectors
interviewed maintenance engineers and supervisors from the electrical and l&C
maintenance groups and reviewed work documentation and licensee investigation
results. The findings were discussed with senior licensee management.
b. Observations and Findinos
On November 10, a non-licensed operator on rounds it'entified that the "A" Reactor
Recirculation Motor Generator (RRMG) field breaker was shut while the RRMG was
shutdown. This breaker should have tripped during RRMG shutdown. Deviation
Event Report (DER) 96-1616 was written to document the event and track
corrective actions. Troubleshooting determined that there was a lifted lead (KK18)
which affected the normal trv coil, which also performed an Anticipated Transient
,
Without Scram safety function. However, the licensee was unai;le to determine
any work activity which could have lifted the lead and not relanded it. Work
documents showed that the lead was lifted during three surveillances, but
surveillance documentation indicated that the lead was properly relanded. No other
documented work in the vicinity of the lifted lead required or documented lifting it.
As a result, there was no corrective action initiated.
The licensee investigation discovered that two l&C technicians found the KK18 lead
lifted during unrelated work in the same panel. They taped the end, but never
9
'
l
- ,
1
l
l
investigated or reported this discovery.
The inspectors determined that the safety function (ATWS Recirculation Pump Trip)
was affected. However, the function remained operable because any of the
redundant relays could have succeeded in tripping the pump during an ATWS.
The inspectors reviewed the completed surveillances and found that the
documentation for relanding the KK18 lead was not conclusive; the l&C policy on
how procedure steps were signed resulted in one person signing for all steps being
performed, including verifications, based on reports from workers. The inspectors
also determined that neither of the last two surveillances which were known to
have lifted the KK18 lead,44.030.251 and 44.030.253, " Division 1 Reactor Vessel
Water Level Channel Functional Tests (Channels A and C)," adequately verified that l
it was relanded. The normal methodology used for l&C procedures for relanding I
leads was to either independently verify the lead to have been returned to normal, i
or else functionally test that part of the circuit. Neither was done in the case of
these procedures.
c. Conclusions
The licensee's investigation of the RRMG field breaker problem did not identify a
probable cause, did not recommend any corrective action, and did not identify the
lack of procedural rigor. The inspectors were concerned particularly with the
importance of the safety function affected. The inspectors believed that
surveillance documentation was inconclusive and the procedure was inadequate to
prove that the lead was landed and the RRMG Field Breaker functioned properly at
the end of the work. The l&C policy of signing for step completion contributed to
the inconclusiveness of the documentation. The inadequate surveillance procedures
were considered a violation (VIO) (50-341/96013-03).
M1.3 Conclusions on Conduct of Maintenance
Maintenance activities were generally completed professionally. The licensee
employed a larger number of contract workers during this outage than during past
outages, but contractor control issues were limited. For the most part, adequate
supervision of contract workers by licensee personnel was evident.
As discussed in this report, attention to detail during maintenance work, particularly
in documentation and closecut reviews, was not always adequate. Several
examples of inadequate scheduled preventive maintenance activities were evident
in investigating equipment performance problems. Also, though not specifically
determined, the inspectors concluded that the RRMG lifted lead and imprnperly set
CCHVAC heater controller (M2.1) were due to improper restoration from
maintenance activities.
10
.
.
-
_ _ _
- *
. l
. \
M2 Maintenance and Material Condition of Facilities and Equipment
M2.1 Control Center Heatina. Ventilation and Air Conditionina (CCHVAC) Filter
Overheated - Preventive Maintenance Inadeauscles Hiahliahtgd
.
a. Insoection Scope (93702,92903)
The inspectors observed licensee response to alarms indicating a high temperature
condition in the CCHVAC Emergency Makeup Filter (EMUF). The investigation
results and maintenance issues identified by the licensee were discussed with the !
investigation team, system engineer, and senior management.
b. Observations and Findinas
On October 25, the CCHVAC EMUF overheated following charcoal replacement.
This resulted in a fire alarm and auto oxidation of the filter, an exothermic reaction.
The fire brigade responded. Operators found that the heater was set at 250
degrees fahrenheit vice the correct setting of 150 degrees fahrenheit. The Alarm
Response Procedure discussed lining up the fire protection system to deluge the
-
filter if necessary, but the operators found that valve labels did not agree with the
procedure or the system drawings. The licensee conservatively decided to deluge
the filter and replace the charcoal when temperatures inside the filter did not
consistently decline with the heater off, even though temperatures were well below
4 the flashover point.
i The heater cutout switches were found to be out of calibration 200 degrees
- fahrenheit high and not covered by scheduled preventive maintenance. Heater
controllers were also found to be out of calibration. System records indicated that
- the heater cutout switches were supposed to be set 300 degrees fahrenheit higher
that the fire alarm setpoint, rather than below the fire alarm setpoint. The design
basis of this setpoint was being reviewed by the licensee at the end of this report
period, as well as possible deactivation of the heater.
c. Conclusions i
The inspectors considered the licensee's response to the event was appropriate and
conservative. The inspectors considered that the licensee's investigation was
sufficiently broad and detailed, and identified a number of existing equipment
problems, most significantly a lack of scheduled preventive maintenance for some
components. The four mislabelled water deluge valves were identified before they
hampered fire brigade response. The lack of adequate preventive maintenance is a
violation. Since the requirements of Section Vil of NUREG-1600, " General
Statement of Policy and Procedures for NRC Enforcement Actions" were met, this
is considered to be a non-cited violation (NCV)(50-341/96013-04).
11
. _ _ . . , .-_ . _-. .- . ._ ._ _ _ - - --
. .
,
. . ,
t
1
M3 Maintenance Procedures and Documentation
!
M3.1 Two Surveillances Started Without Meetino Reouired Plant Conditions - Pre-lob
Reviews inadeouate
a. Inspection Scoce (92903)
~
I
Following findings in the last Resident inspection Report (96010), that scheduled i
,
equipment lineup changes that affected shutdown cooling (SDC) were not
adequately reviewed for plant impact. The inspectors reviewed scheduling of
surveillances during the outage. Members of the Independent Safety Engineering
Group (ISEG) were interviewed to determine what reviews were performed by
ISEG. Operator logs and applicable DERs were reviewed for problems encountered.
Findings were discussed with senior plant management.
, b. Observations and Findinos l
l
Independent Safety Engineering Group (ISEG) reviews of the outage schedule were
performed in two phases. The pre-outage review was performed several months in
advance of the outage, and resulted in a detailed report which included safety
- system availability and required power sources, and a number of other useful data. )
However, the inspectors determined through discussions with ISEG that this report
did not reflect the actual outage schedule by the time it was issued.
The second phase of review involved ISEG maintaining a running review of the
impact of outage schedule changes. The Independent Safety Engineering Group
acknowledged that this running review may not have been comprehensive because '
'
of the large number of schedule and scope changes made in the month before the
outage. It became comprehensive again after the outage started and the number of
changes became smaller. During the outage, ISEG reviewed the entire list of l
approved activities daily, and attended Fermi's integrated Resource Support Team i
,~
screening meetings to review non-scheduled work. However, the depth of reviews ;
varied. .
!
'
Two surveillances were begun without realizing that plant conditions were not i
appropriate. On October 3, Surveillance 42.302.04, " Division 2 Bus 65E/13EC UV I
Logic Functional Test," conflicted with Non-interruptible Air System being cross-tied
. for Reactor Building Closed Cooling Water (RBCCW) system outage. Deviation l
i Event Report (DER) 96-1291 was written to document the event and track l
corrective actions. On October 28, Surveillance 44.030.052, "ECCS - RHR Division
2 Logic Functional Test" required securing Division 1 SDC at a time when it was
required by TS, and an alternate method of removing decay heat was not available.
Deviation Event Report (DER) 96-1511 was written to document the event and
track corrective actions. Operators identified these conflicts after starting the j
surveillances, and responded properly by reporting the problem and backing out of
'
the procedure. The inspectors reviewed the two surveillance procedures. The !
impact statement and prerequisites for 44.030.052 did not address the need to
secure both divisions of shutdown cooling.
12
I
1
.
- ,
c. Conclusions
The inspectors determined that no technical specification (TS) requirements were
violated. The surveillances were stopped before violations of TS occurred.
However,.the inspectors were concerned that licensee reviews of the impact of
these surveillances, both in determining when to schedule performance and during
operator preparation to run the tests, failed to identify that plant conditions were l
not appropriate.
M8 Miscellaneous Maintenance issues (92902)
M8.1 (Closed) Follow-Uo item 50-341/96010-01: Ultimate Heat Sink Cross-tie Valve
(E1150-F601 A) failed to operate. The licensee determined that the valve failed to
stroke because a set screw had come loose in the gear connecting the motor
operated actuator and the valve shaft. Repairs were performed, and the other three
cross-tie valves inspected satisfactorily. corrective actions for this event will be
reviewed under LER 96-14. This item is closed.
M8.2 (Closed) Unresolved item 50-341/94016-04: Poor cleanliness in the torus and
drywell. The inspectors conducted closecut inspections of the primary containment
during the forced outage in April 1996 and at the conclusion of RFO5, and found
cleanliness control much improved. Emphasis on housekeeping during work
activities insido containment and detailed walkdowns during closeout preparations
were evident. The suppression pool was observed by inspectors and found to be
free of fibrous material and debris, with the interior coating in good condition. The
licensee vacuumed the suppression pool during this outage to remove minor
accumulations of rust chips. This item is closed.
Ill. Enaineerina
E2 Engineering Support of Facilities and Equipment
E2.1 Control Rod Position Indication Reolacement Proiect
a. Insoection Scone (92902. 92903. 37551._40500)
Inspectors discussed Position Indication Probe (PIP) replacement plans with the
project staf f. Work in progress and testing performance were observed. DERs
identifying problems were reviewed, and results discussed with members of the
project team. Operator logs were reviewed for PIP problems following replacement.
b. Observations and Findinas
The work for this project was carefully planned in great detail. Close cooperation
among engineering and maintenance groups and vendors resulted in several
innovations which simplified installation while improving the cable design and
reliability. Pre-installation testing was maximized, so the majority of problems
identified were found, documented, and corrected before the drywell work began.
13
.
% ~ m s
, . 1
. .
l
l
Dose controls and shielding were well-planned, and constantly adjusted based on
worker feedback. Actual dose was just under the original estimate of 25 Rem.
Lab testing of the new assemblies was performed under extreme laboratory
conditions at elevated temperatures before the outage, which confirmed that past
problems with loss of electrical continuity were not experienced by the new design.
Post modification test;ng identified a minor problem which caused position
indication flickering ouring rod movement. Troubleshooting and vendor bench
testing identified that this phenomenon was probably caused by slightly strong
magnet or overly seasitive reed switches in some PIP probes. A number of PIP l
probes were replaced to minimize the problems, even though they were not
expected to impact plant epcGion.
c. Conclusions
l
The PIP replacement project was planned and implemented well. Initial results
indicated that past problems appeared to have been corrected. However, past
problems were most significant at power with high core flow, which have not been
experienced since replacements.
E2.2 Turbine Reolacement
inspectors discussed the low pressure turbine replacement and steam path upgrade l
plans with the project staff. Work in progress was observed and DERs identifying
problems were reviewed. The work was carefully planned in great detail and
performed smoothly. Few problems were encountered, although the turbine had
not been rolled with steam at the conclusion of this report. The inspectors
identified some early issues with housekeeping and radworker practices, which
were discussed with the licensee staff, and improvement was later seen. Radiation
Protection support was noted as particularly good, and was instrumental in the
radworker practice improvements. Damper installation on the steam lines and high
pressure turbine control valve problems corrected during valve inspections were l
expected to result in a reduction in steam line vibration and improved steam flow
characteristics compared to last cycle.
E2.3 Safety Relief Valve (SRV) Failures and Resolution to LER 96-017
a. Insoection Scone (92902)
On December 1, during surveillance testing, safety relief valve (SRV) B21-F013A
did not properly indicate when the valve was open, although secondary indications
(steam flow and tailpipe temperature) did confirm that the valve opened and closed
properly, inspectors observed control room operators response to the failure, and
reviewed applicable documentation including Surveillance Procedure 24.137.11,
" Safety Relief Valve Operability Test," and Alarm Response Procedure 1D61, "SRV
Open."
14
. .
. ,
,
b. Observations and Findinag
While attempting to cycle SRVs during plant startup, control room operators
observed that when SRV B21F013A was commanded open, it failed to indicate
open. The operators did note that main steam line flow and bypass valve position
responded as expected for the valve actually opening. On a second attempt, the
"open" light came on briefly. The SRV open indication, which was produced by a
pressure switch connected to the SRV tailpipe, was declared inoperable and
appropriate Technical Specification (TS) Limiting Conditions for Operation action
requirements were initiated. Initial troubleshooting indicated possible instrument
line blockage. The licensee decided to shutdown to correct this and several other
equipment problems on December 2 (see section M1.3).
Troubleshooting did not identify a specific problem, but the licensee believed that
some blockage was blown from the instrument line. The plant was again started
up, but on December 7 Surveillance Procedure 24.137,11 was again unsatisfactory
for SRV "A." The open indication for the SRV failed to remain energized when the
valve was open and spuriously indicated open when the valve was being closed.
Based on steam flow and bypass valve position changes, the SRV was verified to
have opened and closed as required. The reactor was subsequently shutdown and
additional troubleshooting was initiated.
The licensee decided the problem was likely a design problem, and did a deteiled
historical performance review of all SRVs. Vendor assistance was requested.
Advanced flow modeling was performed which showed that the instrument tap
existed near an area of low pressure due to sonic flow. A modification was
implemented which moved the tap several feet downstream. All the SRV tailpipes
were evaluated, but the remaining 14 were determined to be free from this
problem. Also, historical data supported the conclusion that only SRV "A" had
intermittent indication problems.
The investigation included a design and licensing basis review, which identified and
corrected some documentation inconsistencies of minor consequence. The licensee
was also evaluating this for a potential 10 CFR 21 report.
The modification was planned to be installed and tested after the conclusion of this
inspection period. This will be tracked as an inspection Followup Item to verify the
modification was implemented and tested satisfactorily. (IFI) (50-341/96013-05)
c. Conclusion _1
During the first shutdown, the inspectors concluded that the licensee took
reasonable actions to determine the pressure switch was functioning, then
maintained a questioning attitude and planned for contingencies during the
subsequent startup. When the problem recurred, pertinent data was collected and
the reactor was appropriately shutdown in accordance with TS requirements. The
subsequent investigation of the problem was broadened appropriately, and industry
assistance was sought. Troubleshooting efforts displayed considerable coordination
15
. .
!
. .
. ,
I
among the various work groups involved. No violations of NRC requirements were l
identified. '
E2.4 Emeraency Diesel Generator 12 Muffler - Corrective Actions for EDG Muffler l
Problem Informal and Ooerability Vaaue
j
a. Insoection Scooe (92902,40500)
The inspectors condccted a followup on an NRC-identified rattle in the EDG 12
muffler, as discussed in IR 96002. The disposition of the original DER and work
request were reviewed. Findings were discussed with the system engineer, safety
engineering, licensing, and senior management.
b. Observations and Findinas
in January,1996, inspectors identified a metallic rattle in the exhaust muffler for
EDG 12 while the engine was running. The condition was documented in DER 96- l
0026. Based on a discussion with the vendor, a licensee determination was made
that EDG 12 operability would not be impaired provided engine exhaust
temperatures did not rise, the rattle did not get louder, and the rattle did not change i
locations in the muffler, The probable cause of the noise was determined to be i
"some sort of internal baffle failure," and the DER dispositioned the problem by '
scheduling the muffler to be replaced during Refueling Outage (RF) 05, to be
completed by November 30.
The inspectors reviewed the issue and found that the muffler was no longer
scheduled to be replaced during the outage. However, the DER was closed under a
recent program, called " Closed to Process," (CTP). This process allowed closing i
DERs of low importance provided the disposition was traceable and handled under
another process, such as a maintenance work request. In this case, the EDG 12
muffler was tracked under Work Request 000Z961758 and DER 96-0026 was
closed.
System engineering had determined that the significance of the rattle was low
because the vendor stated the likelihood of an intema! muffler failure leading to
blocked exhaust line was low. However, the inspectors noted that the
consequences of a partially or fully blocked exhaust line would be reduced load
capability for the EDG, up to complete loss of safety function. The inspectors
determined that this was not evaluated by system engineering or safety
engineering, despite the high safety significance of the EDG.
Subsequent to closing DER 96-0026, maintenance / planning decided to schedule the
work for the next system outage, targeted for May,1997. This was a considerable
extension of the intended time to replcce the muffler over what was assumed at
the time the operability determination was made, yet the extension was not
discussed with those responsible for making the original decisions because the DER
was closed.
16
-
. .
I r g
4
The inspectors found that the continued operability determination for EDG 12 was j
not formally documented or tracked. The EDG System Engineer was monitoring
surveillance runs, listening to the muffler, and reviewing logs for exhaust
temperature. Licensee management was unaware that there was an ongoing
operability assessment or that a delay in replacing the muffler was planned until
brought to their attention by the inspectors.
Following the inspectors' raising the above issues, the conditions being monitored
to ensure continued EDG 12 operability in light of the muffler degradation were
added to the monthly surveillance procedure. Licensee management decided to
review the CTP program from the standpoint of lessons learned from this issue.
Current operability was reviewed by the licensee and determined to be unchanged
since the time of initialidentification of the rattle.
c. Conclusions ,
The inspectors agreed based on their observations that the EDG 12 operability was
unchanged. However, the inspectors were concerned that the CTP method of
handling this issue did not handle ultimate resolution with the proper attention due
a safety system. Low significance was assigned based on a low probability of
catastrophic failure, rather than evaluating the consequences of further degradation.
The CTP procedure did not require that changes to the intended corrective action be
reviewed with those responsible for the original action. Also, the informal method
by which the ongoing assessment of EDG 12 operability has handled was
considered weak, particularly because it resided with a single individual. Failure to
take corrective actions in a timely manner commensurate with the safety
significance of the system was considered a violation of 10 CFR 50, Appendix B,
Criterion XVI (VIO) (50-341/96013-06).
E2.5 UFSAR Reouirement Review
A recent discovery of a licensee operating their facility in a manner contrary to the
Updated Final Safety Analysis Report (UFSAR) description highlighted the need for
a special focused review that compares plant practices, procedures, and parameters
to the UFSAR descriptions. While performing the inspections discussed in this
report, the inspectors reviewed the applicable portions of the UFSAR that related to
the areas inspected. The inspectors verified that the UFSAR wording was
consistent with the observed plant practices, procedures, and parameters.
As discussed in E2.3, the licensee identified a discrepancy in UFSAR description of
the licensing basis of SRVs and associated accident monitoring indication. This
issue was discussed with NRR, and licensee actions to verify the design by testing
and resolve UFSAR inconsistencies were determined to be adequate.
E3 Engineering Procedures and Documentation
E3.1 Insoectors identifv Need for Technical Soecification Chanae
17
1
I
e ,
,
a. Insoection Scoce (92902,71707)
1
The inspectors reviewed the Core Operating Limits Report (COLR) for Cycle 6 and l
the associated Technical Specification Clarification (TSC)96-004. When it l
appeared that a TS change was required based on this review, the issue was
discussed with NRR Technical Specifications Branch, as well as members of the
licensee's Reactor Engineering and Licensing staffs. The associated preliminary
evaluation and a safety evaluation were also reviewed,
b. Observations and Findinas
,
Technical Specifications 3.1.4.3 and 3.3.6 described the conditions under which I
the Rod Block Monitor (RBM) must be operable. The Fermi Core Operating Limits
Report for Cycle 6, Revision 0, stated that "In addition to these requirements, at
least one RBM channel must be operable when moving control rods with thermal
power greater than or equal to 30 percent of rated thermal power in order to
protect for mechanical overpower limits."
The inspectors discussed this issue with Reactor Engineering and Licensing, and
concluded that this statement inferred that the existing TS requirements were ;
inadequate to protect the core during a rod withdrawal error event. NRR Technical !
, Specifications Branch reviewed the issue, and agreed that a TS change was
required. The licensee agreed and submitted a license amendment request on
December 2. Based upon the licensee's commitment to make a license change
submittal and administrative controls in place, NRR determined that Fermi could
start up while the amendment was being reviewed.
The inspectors determined that Safety Evaluation 96-0128, Revision 0, and the
associated Preliminary Evaluation, evaluating the COLR for the Cycle 6 core, were
inadequate in that they improperly concluded that a TS change was not required to
ensure that mechanical overpower limits would be met during a rod withdrawal
error event. Instead, the licensee issued TSC 96-004 to indicate that the RBM
operability requirements of the COLR would be followed, beyond the TS
requirements.
c. Conclusions
The NRC concluded that a change was required to the existing TS in order to
ensure that mechanical overpower limits would be met during a rod withdrawal
error event. Failure to identify that a TS change was required during the review of
the proposed modification to the facility was a violation of 10 CFR 50.59 (VIO)
(50-341/96013-07).
E3.2 Proaram to Evaluate Chanaes. Tests and Exoeriments Pursuant to 10 CFR 50.59
a. Insoection Scoce (37001)
The inspectors reviewed selected preliminary evaluations (PE) and safety
18
- - . . _ - . - . --- . - -- .
. .
!
i . , ,
l
evaluations (SE) performed by the licensee to satisfy the requirements of
b. Observations and Findinos
The inspectors reviewed a total of 41 PE and SE reports completed by the licensee
within the last two years. Roughly half of the evaluations evaluated changes to the
facility while the balance was split between changes to procedures and tests. Most
of the evaluations had been performed in conformance with the regulations and the
licensee's procedures. However, the inspectors did find problems in some of the
evaluations.
The licensee prepared a preliminary evaluation and a safety evaluation (SE 96-0128)
for the Core Operating Limits Report (COLR) for the Cycle 6 core. As discussed in
detail in E3.1 above, the evaluations failed to identify the need to amend the TS to I
'
ensure fuel mechanical overpower limits would be met during a rod withdrawal
error event.
The licensee prepared SE 95-0036 to evaluate operating the Emergency Equipment
Cooling Water (EECW) system to supplement the Reactor Building Closed Cooling I
Water system. The SE did not fully evaluate the consequences of this change in I
the operation of EECW. In particular, the SE didn't evaluate the possibility that .)
operating in this mode could cause reduced flows to essential loads under certain
accident conditions. The licensee wrote DER 96-1836 to document the deficiency
and track corrective actions.
In addition to these problems, the inspectors noted some weaknesses in the
licensee program for evaluating changes, tests and experiments. The licensee
procedures did not specify reviewing for any potential unreviewed safety question
which might exist during the implementation of a change. Also, some reviewers
appeared to interpret the scope of review under 10 CFR 50.59 for changes to
procedures to only include those items specifically identified as procedures in the
UFSAR. The inspectors determined that Revision 3 to MLS02, " Preliminary
Evaluations and 10 CFR 50.59 Safety Evaluations," step 4.1.9, indicated that the
scope of this review should be broader. Finally, some preliminary evaluations didn't
contain sufficient information to allow a subsequent reviewer to fully understand
the safety significance of the change. The licensee was mada aware of these
deficiencies and indicated that corrective actions were being evaluated.
c. Conclusions
The inspectors determined the licensee's 10 CFR 50.59 process to be generally
adequate for the review of proposed changes. However, failure to perform an
adequate evaluation for the COLR was cited as a violation in E3.1, and the EECW
safety evaluation will be tracked as an Unresolved item pending NRC review of the
licensee re-evaluation (URI) (50-341/96013-08).
19
. - _
- . _ - . . _ . . - . - .
. .
t i e a
1
IV. Plant Suocort
R1 Radioiogical Protection and Chemistry Controls
i
All Radiation Protection topics were discussed in the sections above. No separate
, writeup will be provided.
'
V. Manaaement Meetinos
X1 Exit Meeting Summany
The inspectors presented the inspection results to members of licensee management at the
conclusion of the inspection on December 17,1996. The licensee acknowledged the
findings presented.
I
The inspectors asked the licensee whether any materials examined during the inspection
- should be considered proprietary. No proprietary information was identified.
!
X3 Management Meeting Summary
l
1
On December 17, J. Hannon, Director, Project Directorate ill-3, NFiR met with D. Gipson,
Senior Vice President, Nuclear and members of his staff on site ta discuss the environment
for reporting safety concerns at Fermi.
20
. .
. . . .
PARTIAL LIST OF PERSONS CONTACTED
,
Licensee
,
S. Booker, General Supervisor, Electrical Maintenance
- C. Cassise, General Supervisor, Mechanical Maintenance
I
W. Colonnello, Director, Safety Engineering
i R. Delong, Superintendent, System Engineering
- T. Dong, NSSS, Technical Engineering
P. Fessler, Plant Manager, Operations
E. Kokosky, Superintendent, RP and Chemistry
,
R. McKeon, Assistant Vice President, Operations
, W. Miller, Technical Support
J. Nolloth, Superintendent, Maintenance
i J. Nyquist, Acting General Supervisor, ISEG
J. Plona, Technical Director
i W. Romberg, Assistant Vice President and Manager, Technical
P. Smith, Director, Nuclear Licensing
G. Trahey, General Supervisor, ISEG
E. Vinsko, General Supervisor, l&C
NRC
i
M. Weston, NRR
- G. Hammer, NRR
,
t
0
4
i
e
a
i
1
!
G
21
- - . ._- - ._ . . . - . - _ _ - = - ..- - - . . - . _ . -
F
,
, .. ,
INSPECTION PROCEDURES USED
!
IP 37001: 10 CFR 50,59 Safety Evaluation Program {
IP 37551: Onsite Engineering ;
IP 40500: Effectiveness of Licensee Controls in Identifying, Resolving, and Preventing !
Problems
IP 60710: Refueling Activities
IP 62703: Maintenance Observation
IP 71707: Plant Operations l
lP 92901: Followup - Operations
IP 92902: Followup - Engineering
IP 92903: Followup - Maintenance l
IP 93702: Prompt Onsite Response to Events at Operating Power Reactors !
!
<
ITEMS OPENED, CLOSED, AND DISCUSSED
Onened
50-341/96013-01 VIO Unexpected Scram When Manual Scram Roset Prematurely ,
50-341/96013-02 VIO Inadequate Procedure failed to prevent overfill of SFP !
50-341/96013-03 VIO Inadequate Procedure Failed to Verify Electrical Lead Properly
Connected After Removal ;
50-341/96013-04 NCV inadequate preventive maintenance on CCHVAC i
50-341/96013-05 IFl SRV "A" Indication Modification and Testing
50-341/96013-06 VIO Loose Baffle Plate in EDG 12 Not Repaired in a Timely Manner
50-341/96013-07 VIO Inadequate Safety Evaluation Failed to Identify Required TS
Change
50-341/96013-08 URI EECW SE Failure to Fully Evaluate Consequences of Change to
Supplement RBCCW System
Cipsed
50-341/96010-02 IFl Ultimate Heat Sink Cross-Tie Valve Failed to Operate
50-341/94016-04 URI Poor Cleanliness in Torus and Drywell
l
l
1 ;
l
l
l 22 i
l
l. .
'
. . , ,
i
LIST OF ACRONYMS USED
CCHVAC Control Center Heating Ventilation Air Conditioning l
CFR Code of Federal Regulations
COLR Core Operating Limits Report
l
CTP Close to Process
DER Deviation Event Report
DPM Disintegrations per minute
EDG Emergency Diesel Generator
ECCS Emergency Core Cooling System
EECW Emergency Equipment Cooling Water
EMUF Emergency Makeup Filter
ESF Engineered Safety Feature
GPM Gallons per Minute
I&C Instrumentation and Control
INPO Institute of Power Operations
IR Inspection Report
ISEG Independent Safety Engineering Group
LER Licensee Event Report
NCV Non-Cited Violation
NRC Nuclear Regulatory Commission
NRR Office of Nuclear Reactor Regulation ,
PIP Position Indication Probe l
PMT Post Maintenance Testing
RBM Rod Block Monitor
RBCCW Reactor Building Closed Cooling Water
RCIC Reactor Coolant isolation System
RF Refueling Outage
RFC Refueling Floor Coordinator
RRMG Reactor Recirculation Motor Generator
RWCU Reactor Water Clean-Up
SALP Systematic Assessment of Licensee Performance
SDIV High Scram Discharge Instrument Volume
SE Safety Evaluation
SFP Spent Fuel Pool
TS Technical Specification
TSC Technical Specification Clarification
UFSAR Updated Final Safety Analysis Report
URI Unresolved item
VIO Violation
l
{
,
4
23
,
l