IR 05000461/1993019
| ML20058D019 | |
| Person / Time | |
|---|---|
| Site: | Clinton  |
| Issue date: | 11/11/1993 |
| From: | Hague R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20058D008 | List: |
| References | |
| 50-461-93-19, NUDOCS 9312030026 | |
| Download: ML20058D019 (22) | |
Text
'
-.
NOV 2 21993
.
{
'
U.S. NUCLEAR REGULATORY COMMISSION
REGION III
!
,
,
Report No.
50-461/93019(DRP)
l Docket No.
50-461 License No. NPF-62 i
Licensee:
Illinois Power Company l
500 South 27th Street
l Decatur, IL 62525 Facility Name:
Clinton Power Station Inspection At: Clinton Site, Clinton, Illinois
!
l Inspection Conducted:
September 14 - October 25, 1993 l
Inspectors:
P. G. Brochman F. L. Brush A. M. Stone Approved By:
M
//[//F3
'
M ichard L. Hague,/ Chief
'
D' ate Reactor Proje't Section 1C Inspection Summary Inspection from September 14 throuah October 25. 1993.
(Report No. 50-461/93019(DRP))
l Areas Inspected:
Routine, unannounced safety inspection by the resident inspectors of licensee actions on plant operations, maintenance, engineering, and plant support.
'
l Results: Of the four areas inspected, no violations or deviations were l
identified in one area; one violation was identified in the following area:
(failure to incorporate design requirements and perform adequate post-modification testing - paragraph 4.a).
Two non-cited violations were also identified (failure to obtain approval of overtime in excess of the l
guidelines - paragraph 2.f(l); failure to follow procedure - paragraph 3.a).
l One unresolved item was identified (operability of the high pressure core
!
spray injection valve - paragraph 4.b).
i
'
.
l
'
9312030026 931122 PDR ADDCK 05000461 G
PDRx
,
l
__
_
.
'
. _ _ - _ _ _ _ _ - _ _ _ _.
Executive Summary Plant Operations The plant was shut down September 26, 1993, for its fourth refueling
outage.
Multiple engineered safety features actuated during diesel testing due
to a blown fuse on an nuclear system protection system (NSPS) inverter.
Indications of reactor vessel level " notching" were observed by the
licensee on high speed recording instruments.
Control room instruments did not detect the anomalies.
Refueling activities were conducted in a very conservative manner.
Two
minor errors with fuel bundle orientation did occur.
Licensee efforts at minimizing shutdown risk have been excellent.
- Two instances of licensed individuals exceeding the overtime guidelines
without prior approval occurred.
(NCV)
e Maintenance Multiple examples of personnel not following the danger tag program and
one instance of operation of a danger tagged valve occurred.
(NCV)
Multiple problems with the control of foreign material over and inside
the reactor vessel occurred. Weaknesses in communications, oversight, and supervision of refueling floor activities was noted at the beginning of the outage.
The problems were addressed by the licensee.
The licensee's initiative to test safety relief valves onsite was very
effective and saved dose.
A contractor worker did not follow procedures and attempted to lift a
valve stem that was still bolted down.
The lifting device failed.
No equipment or personnel were damaged.
Engineerina Poor understanding of the design basis, weak review of a modification, e
and insufficient post-modification testing led to the
"A" fire pump being inoperable for 3 months.
(NV4 461/93019-01 (DRP))
The high pressure core spray injection valve was found with a broken
motor pinion gear key.
Evaluation of the valve's operability was underway.
Engineering evaluations of the failure of the NSPS inverter and the "B"
circulating water pump motor's bearing were very comprehensive and thorough.
Plans for testing the reactor vessel water level indicating system
'
modification (NB-31) were thorough with an appropriate test methodology.
Scheduling of testing and installation activities assured that technical specification minimum availability requirements were met.
Plant Support Degradations in housekeeping were noted during the outage.
The licensee
impleuented actions to address the issue.
_
-_
_
_
.
DETAILS 1.
Persons Contacted Illinois Power Company (IP)
'
- J. Perry, Senior Vice President
- J. Cook, Vice President and Manager of Clinton Power Station (CPS)
- J. Miller, Manager - Nuclear Station Engineering Department (NSED)
l
- R. Wyatt, Manager - Quality Assurance
- D. Thompson, Manager - Training
- J. Palchak, Manager - Nuclear Planning and Support
'
- F. Spangenberg, III, Nuclear Strategic Change leader
- R. Phares, Director - Licensing L. Everman, Director - Radiation Protection P. Yocum, Director - Plant Operations W. Clark, Director - Plant Maintenance K. Moore, Director - Plant Technical
'
- W. Bousquet, Director - Plant Support Services r
- C. Elsasser, Director - Planning & Scheduling i
R. Kerestes, Director - Nuclear Safety and Analysis
- D. Korneman, Director - Systems and Reliability, NSED
- J. Langley, Director - Design and Analysis, NSED
- J. Sipek, Supervisor - Regulatory Interface The inspectors also contacted and interviewed other licensee and contractor personnel during the course of this inspection.
,
Denotes those present during the exit interview on October 25, 1993.
,
2.
Plant Operations The unit began the report period at 90 percent power coasting down to the fourth refueling outage (RF-4), which started on September 26, 1993.
a.
Unanticipated Containment Isolation At 2:40 p.m. on September 30, 1993, while in cold shutdown, the licensee was performing the Division II, diesel generator
.
LOOP /LOCA* actuation test when several-components failed to function and an unexpected engineered safety feature (ESF)
actuation occurred.
Division II containment isolations occurred on 14 systems and RCIC (reactor core isolation cooling) received an auto start signal. Additionally, the Division 11 low pressure coolant injection (LPCI) equipment did not actuate.
Later at 2:45 p.m., the "A" reactor operator noted that level had risen in the spent fuel pool cooling (FC) system's surge tank.
Operators then recognized that containment isolations had occurred and performed
A loss of offsite power accident coincident with a loss of coolant
accident.
,
-
.
.
-
..
1 ihr auto. isolation checklist for reactor vessel low level-2.
No c.ner problems were found and the isolated systems were restored to operation. There was no impact on the reactor core or spent fuel pool. Only the gland seal compressor for RCIC turbine
started as there was no main steam to start the RCIC turbine
itsel f.
,
The cause of the equipment failure and the unplanned actuations
,
l
'
was a loss of power to the Division II NSPS inverter. This was
!
caused by the blowing of the dc input fuse F-1.
.This resulted in
.
l
'
the inverter having neither an ac nor de power supply during the l
test. This led to a loss of power on the NSPS bus and caused the
.
!
fa lures and actuations. The blown fuse was replaced and the test reperformed on October 1, 1993.
This time fuse F-1 blew after the i
diesel started and restored power to the 4.16 kV bus.
Therefore, l
the NSPS bus remained energized from its ac source and the LPCI l
equipment actuated as designed.
{
l The licensee's evaluation of this event was ongoing and the
!
inspectors will perform further. reviews after the licensee event report (LER) is issued. The operators response to this event was good. Review of the engineering departments evaluation and root l
cause analysis for this problem is discussed in paragraph 4.c.
l b.
Reactor Water Level "Notchina" f
During plant cooldown in preparation for the fourth refueling
,
outage, the licensee monitored reactor water level indications to
!
l determine if the " notching" phenomenon would occur.
" Notching" is
,
!
a change in indicated reactor water level caused by non-i
!
condensable gases in the water level indication reference legs
coming out of solution. Data was taken on four of the water
reference leg channels from 500 psig to near 0 psig using a plant computer system. Notching was observed on all four channels with
the largest eifect being less than 4 inches for under 35 seconds.
However, the level anomalies were not large enough to be observed l
'
l on control room instrumentation.
l The licensee is installing modifications.to the reactor water level instrumentation reference legs which will preclude the i
buildup of non-condensable gases, as required by NRC Bulletin 93-03.
The inspectors are following the licensee's efforts in
,
I this area.
c.
Refuelina Activities (60710)
The inspectors observed refueling activities in the control room,
.
the fuel building, and on the refueling floor. There were good
!
communications between personnel in the various. locations, procedures were strictly followed, and licensee management was in attendance during all fuel moves. Overall, core alterations were performed in a conservative and methodical manner. However, there l
.
!
L
.
-
-
-
-
-
.
were two instances where fuel bundles were found in an incorrect orientation.
On October 12, 1993, the oncoming senior reactor operator
noticed that a fuel bundle in the upper storage pool was in the wrong orientation. The special nuclear material (SNM)
transfer list was updated and station nuclear engineer informed. The bundle was reoriented and a verification was made of the other fuel bundles.
Three individuals had verified that the bundle was in its proper orientation.
On October 23, 1993, a fuel bundle on the periphery of the
core was found in the wrong orientation during the post-core video verification. The bundle was reoriented and the verification was continued. The licensee was still evaluating this event and its causes, d.
Assurina Shutdown Safety The licensee has aggressively implemented actions to ensure shutdown safety.
These actions include:
A detailed review of the overall outage schedule was
performed before the outage started by the ISEG' to ensure that at least one train of onsite and offsite ac power sources, emergency core cooling systems (ECCS), and decay heat removal (DHR) systems remained available.
A senior reactor operator reviewed all activities
(maintenance, testing, or changes to plant conditions) on a daily basis for any shutdown safety impacts.
Status of ac power, ECCS, and DHR systems was discussed at
all daily outage management and scheduling meetings.
Notice boards, throughout the plant, informed the average
worker what the status of these important systems was on a daily basis.
Overall, the inspectors concluded that these actions have had a very positive effect on reducing shutdown risk.
!
e.
Latham Line Repaired On October 2,1993, the licensee completed final repairs on the Latham line and released it for unrestricted use. The line is one of three 345 Kv transmission circuits connecting Clinton to the grid.
It was damaged on June 8, 1993, when a storm took out 17 miles of transmission towers. The work was completed ahead of schedule and repair activities were scheduled to minimize shutdown risk impacts.
' Independent Safety Engineering Group
.
_ - _ _.
..
.
.
-
,
.
f.
Operational Safety (71707)
i The inspectors observed control room operation, reviewed applicable logs, and conducted discussions with control room operators.
During these discussions and observations, the operators were alert, cognizant of plant conditions, attentive to
,
changes in those conditions, and took prompt action when appropriate. The inspectors verified the operability of selected emergency systems, reviewed tagout records, and verified the proper return to service of affected components.
Tours of the circulating water screen house and auxiliary, containment, control, diesel, fuel handling, rad-waste, and
turbine buildings were conducted to observe plant equipment conditions, including potential fire hazards, fluid leaks,
,
excessive vibrations, and to verify that maintenance requests had
,
been initiated for equipment in need of maintenance. The
!
inspectors verified implementation of radiation protection.
controls and physical security plan.
l (1)
Exceedina Overtime Limits Without Prior Authorization j
On October 12, 1993, the licensee identified that two
!
operations personnel had exceeded the overtime guidelines for working more than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> in a 48 hour5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> period and working more than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> in a 7 day period.
Both of these individuals were involved in critical activities and these i
problems were documented in condition reports. Management re-reviewed procedure CPS 1001.10, " Control of Working
'
Hours," with each individual and reiterated the need for each individual to closely monitor their hours during the
,
busy outage and obtain prior approval when the limits needed to be exceeded.
Technical Specification 6.2.2.f requires, in part, that deviations from the overtime guidelines shall be approved by the Manager - Clinton Power Station. The overtime guidelines include working no more than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> in a 48 hour5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> period and working no more than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> in a.7 day period. The failure to obtain prior approval was a violation of TS 6.2.2.f.
However, the violation is not being cited because the criteria specified in Section VII.B.1 of the " General Statement of Policy and Procedures for NRC Enforcement Actions," (Enforcement Policy, 10 CFR Part 2, Appendix C) were met.
(2)
Water Rainina into Main Control Room On October 9, 1993, the inspector observed water raining from a main control room ventilation duct onto the back of a main control room panel (engineered safety features
controls). The control operators had responded promptly to this event and covered the panel with plastic.
The-
,
.
humidification boiler for the main control room ventilation system had just been placed in service.
The boiler was secured and a work request initiated.
The problem was believed to be a malfunction (plugging) of the steam trap which allowed condensate to back up into the ventilation duct and rain into the control room.
The inspectors discussed this issue with operations management and have no further concerns.
(3)
Communications Problems Contribute to Floodina of Fuel Handlina Buildina Ventilation System On October 3, 1993, operations personnel were draining the refueling cavity due to indications of a leaking flange IRE 68A' which had developed during refueling cavity flood up.
Due to incorrect coordination and communication by operations personnel, cavity drain valves IFC003A & B were opened before skimmer surge tank valves IFC012A & B were closed. This allowed the surge tank and the spent fuel pool to overflow and sent contaminated water into the fuel l
handling building ventilation (FHB) ducts. The water then leaked from the ventilation ducts onto the FHB floor.
'
Valves IFC012A & B were inside contamination zones. This delay in accessing the valves was not anticipated and as the area was also a radio dead zone personnel could not contact
!
the control room to tell it that the valves had not yet been closed. The inspectors concluded that poor operations
,
!
communication and coordination was a contributing cause to the spill of water. The principal cause was a maintenance error in assembling the IRE 68A flange, which is discussed
further in paragraph 3.c.
!
l No deviations were identified. One non-cited violation was identified.
i 3.
Maintenance (61726 & 62703)
a.
Problems with Danaer Taas l
'
During the course of the outage there have been five instances in which personnel failed to adhere to licensee program requirements l
on the use of danger tags--four by contractors and one by control and instrument personnel. No personnel were injured or equipment l
damaged by these acts.
Never-the-less, these events were significant and were of serious concern to the NRC.
On October 1,1993, motor operated valve (MOV) IB21F021, in
the main steam system, was tagged out for troubleshooting of the limit switches, as the valve would not stroke fully shut.
There was a note on the tagout (93-0917) to
'
Reactor Core Isolation Cooling Head spray leakoff line
!
I
~
reevaluate the tagout boundaries if valve operator removal was required.
This note was not clearly understood by the workers and the motor operator was removed without expanding the tagout boundaries. This event was documented in condition report (CR) 1-93-10-015. The work was suspended and new tagout boundaries established.
On October 7,1993, a tag located on the manual handwheel
for MOV IE12F027A was removed when the handwheel was removed by contractor workers to support a modification of the motor operator. The tag should have been relocated by operations personnel if the handwheel needed to be removed. This event was documented in CR l-93-10-022. The handwheel was reinstalled and the valve verified shut.
On October 11, 1993, control and instrument (C&I) personnel
were working on a turbine lube oil pressure sensor, which was inside of the boundaries of a tagout on the lube oil system.
However, the C&I technicians had not signed onto the tagout before commencing their work.
Subsequently, operations personnel cleared the tagout, while the system was partially disassembled. This resulted in a spill of 20 gallons of lube oil onto the main steam lines. This event was documented in CR 1-93-10-031. The lube oil system was shutdown and the hazmat team responded and cleaned up the spilled oil. The pressure switch was reinstalled.
On October 15, 1993, an auxiliary operator (AO) found valve
ISOH096, in the turbine seal oil system, shut but tagged open; and a pipe cap was installed on the drain line downstream of the valve. While the A0 was there, a contractor worker came up and removed the pipe cap and opened the valve. The worker stated that the valve was now back in its initial position. The A0 told the worker to never operate a danger tagged component and immediately i
reported the event to the main control room. This event was documented in CR l-93-10-051.
The tagout was reverified and no other discrepancies were found. Disciplinary action was j
taken against the worker and briefings on this event were given to all craft supervisors and workers.
On October 24,1993, valve 1G33F041 had its actuator removed
,
without a tagout or approval to do work. Valve 1G33F041 was I
in fact a tagged boundary valve for work on the main
'
condenser. However, removal of the actuator did not change the position of the valve disk-it remained shut.
Upon discovery, the valve actuator was immediately reinstalled.
The number of these events occurring during RF-4 was greater than the total number which occurred in 1991.
None of these events resulted in personnel injury or equipment damage. All of these events were identified by the licensee and documented in its corrective action program.
Taken together, they indicated a
I
~
\\
declining trend which is a serious concern to the NRC. The event on October 15 was the most egregious and indicated at best a careless disregard for personnel and equipment safety by the contractor worker.
The licensee's corrective actions included specific retraining for all contractor supervisors, contractor workers, and maintenance personnel.
The importance of safety tags was reiterated in a message from plant management to all employees.
The quality assurance organization has initiated an investigation into the causes for this problem. Disciplinary action was taken against the individual involved in the October 15 event.
Licensee procedure CPS 1014.01, " Safety Tagging," section 4.2 requires that a component with a red tag attached shall not be altered or removed from its physical location....
Section 2.2.3 defines a danger (red) tag as a tag that prohibits operation or status change of a piece of equipment.
Part 50 of Title 10 of the Code of Federal Regulations, Appendix B, Criterion V, requires that activities affecting quality shall be prescribed by documented instructions or procedures of a type appropriate to the circumstances and shall be accomplished in accordance with these instructions. The failure to adhere to the requirements of CPS 1014.01 is a violation of Criterion V.
However, the violation is not being cited because the criteria specified in Section VII.B.2 of the " General Statement of Policy and Procedures for NRC Enforcement Actions," (Enforcement Policy, 10 CFR Part 2, Appendix C) were met.
b.
Control of Foreian Material over the Reactor Vessel The licensee has experienced several instances where foreign material was lost or fell into the reactor vessel. These events were caused by contractor personnel and indicated weaknesses in communications, management oversight, and in evaluatirg the impact of changes to equipment and tools.
On October 1,1993, a plug being installed in a reactor
vessel stud hole failed and a piece flew into the reactor vessel.
It was subsequently retrieved.
On October 5,1993, the nozzle tip on a flushing wand came
loose and was stuck in a reactor vessel penetration.
This problem occurred after contractor technicians (GE) modified a nozzle tip by attempting to join two pieces of stainless steel with a brass Swagelock fitting.
Contributing causes were:
Licensee refueling floor supervisors were not informed
of the problem or the proposed corrective actions.
The GE supervisors oid not understand the technical
implications of the change or why locking tabs had been added to the tool.
The GE technicians had not received formal training on i
assembly of swagelock fittings.
i
GE personnel did not understand that brass swagelock
fittings cannot be used on stainless steel tubing.
Licensee practices on material control and design
changes of equipment were not followed by GE personnel.
Communications between GE technicians, supervisors,
and licensee supervisors was weak.
Work was suspended and the licensee developed a plan to retrieve the nozzle tip from the vessel penetration. The licensee conducted detailed fact finding. The nozzle tip and a plastic tube were subsequently flushed from the penetration to the annulus space. The nozzle tip was retrieved.
On October 8, 1993, an underwater pole, being used to help
in the retrieval of the plastic tube, was suspended from the auxiliary bridge railing with t m. This 70 foot pole came loose and dropped approximately feet onto the reactor core. The GE technicians on the.5.Jge did not report this
,
immediately. They did contact radiation protection and retrieved the dropped tool and secured it to the bridge railing with rope. One of the GE technicians mentioned this to the GE refueling supervisor, several hours after the problem occurred. GE management immediately suspended all refueling floor activities and notified licensee senior management. The licensee conducted a critique on the issue and evaluated any potential damage to the fuel as well as revising methods of operating and licensee supervision of refueling floor activities. The licensee concluded that the dropped pole did not have any impact on the fuel.
The plastic tube was retrieved later.
On October 9,1993, a magnet being used to retrieve objects
in the reactor annulus broke. Management was immediately notified and the pieces were subsequently recovered.
Additionally, on October 1,1993, GE personnel discovered
metal filings on both the refueling and fuel handling building bridges. These metal filings were the result of drilling holes in the bridges to mount new communications equipment. The filings were on top of I-beams where the workers had drilled up from the bottom.
The maintenance work request directed workers to collect filings from underneath the I-beam, but they never looked on top. The
filings were cleaned up and both bridges inspected.
Overall, the inspectors consider this poor performance in maintaining foreign material exclusion (FME) and oversight of contractors.
The inspectors were especially surprised by the
'
events on October 5 and 8, as they indicated a return to the ways and problems of past refueling outages.
The inspectors discussed their concerns with senior licensee management.
Subsequent observations of refueling floor activities indicated significant improvements in FME control and management oversight.
.--
j
-.
I c.
Floodina in Fuel Handlina Buildina Ventilation Ducts On October 3,1993, the surge tank and spent fuel pool overflowed
and sent contaminated water into the FHB ducts. While the cause i
of this event was incorrect coordination of the valve manipulations by plant operations personnel, the reason the refueling cavity needed to be drained was that maintenance workers
had inste %d the wrong gasket on flange IRE 68A. The correct i
gasket was specified in the maintenance package and procedure; however, the wrong type of gasket was installed. Since this wrong
,
gasket did not provide a proper seal on the flange, leakage
occurred. When the leakage was observed, operations personnel initiated efforts to lower the refueling cavity level to repair
i the flange.
Operations aspects of this event are discussed in paragraph 2. f. (3).
In evaluating this event, the inspectors concluded that without I
the maintenance error the operations error would not have
occurred. Consequently, the maintenance error was the root cause i
of this event and this showed poor maintenance performance.
[
'
d.
Safety Relief Valve Testina
!
The licensee developed a test stand with the capability of lift l
testing a main steam safety relief (SRV) valve using nitrogen
!
(N,).
This has reduced dose and schedule impact.
In previous j
outages it was necessary to remove four SRVs and ship them to an
!
offsite vendor for testing and resetting.
If any of the initial 4
!
-
valves failed, the.imaining 12 valves would be removed and
!
shipped offsite for testing; even if only.1 valve in the initial.
!
group of 4 had f ailed. The ASME Code * only required that if one
,
valve in the first group of four failed, then four additional
.
valves be tested. However, because of the impact of equipment removal, reinstallation, shipping, and test facility availability,
!
the licensee did not want to risk the potential delay. To_ avoid j
outage delay, the licensee would immediately remove and test the
remaining 12 valves. Consequences of this policy were increased
,
personnel exposure and potential damage to associated components such as acoustic monitors and flexible conduits during the removal
!
and reinstallation process. The licensee has had problems with
!
!
incidental damage to acoustic monitors and electrical conduit during previous refuelings.
The new onsite testing capability allowed the licensee the time
.
l and flexibility to test only the required number of SRVs. The
.,
idea to develop this capability occurred during a maintenance l
management observation of a refueling outage in a BWR-6 reactor in l
Switzerland.
l i
!
I American Society of Mechanical Engineers, Boiler and Pressure Vessel
)
'
Code,Section XI
,
l i
!
.
__
.
.
Of the initial group of four SRVs removed in RF-4, three passed.
One failed 3.4% low.
Acceptance criteria is i 3% of lift pressure.
All four of next group of SRVs passed. The inspector observed one of the lift tests and was favorably impressed with the testing methodology and equipment.
e.
Riaaina Eauipment Damaaed Durino Turbine Ston Valve Maintenance On October 10, 1993, contractor workers were disassembling and removing the internals of main turbine stop valve ITGMSV4. The
.
valve stem and pressure seal head were suspended from a 3 ton
chainfall, which was connected to the tarnine building crane.
The
worker had been directed to blue check the valve and then i
disassemble the lower end of the valve. However, he forgot to remove the pressure seal head bolts. This would prevent the valve i
stem from being lifted out more than 8.5 inches.
When the
chainfall bound up the worker concluded that the pressure seal
'
head was stuck and directed the crane operator to raise the hook.
When this happened, the. lead chain from the 3 ton chainfall snapped and the valve dropped back onto its seat.
No permanent i
equipment was damaged or personnel injured.
I The licensee's corrective actions included: suspension of all l
turbine floor rigging activities, additional review for craft supervisors on rigging policy and what to'do with stuck
.
components, review of CPS 8273.10 "High Pressure Turbine Stop-Valve Maintenance," to make it simpler and easier to work the steps in sequence, and disciplinary action against the worker.
The inspectors discussed these actions with maintenance management.
The inspector reviewed procedure CPS 8273.10, sections 8.1.10 s
through 8.1.14, and noted it was very clear that the pressure seal head bolts were to be removed before a chainfall was attached.
The failure to remove the pressure seal head bolts before removing
.
l the valve stem showed inattention to procedural requirements by j
the contractor workers.
However, based on licensee corrective
I actions and inspector discussions with licensee management, no further actions are warranted.
f.
Emeraency Siren Transmitter not Restored to Service After Maintenance on the Meteoroloaical Tower On September 7, 1993, the primary emergency siren transmitter failed its monthly test. The transmitter had been deenergized on August 28, 1993, in support of maintenance activities on the meteorological tower.
When the maintenance activities were completed on August 30, 1993, the DeWitt County ESDA' coordinator.
was notified that the transmitter had been restored to service.
'
Illinois Emergency Services and Disaster Agency (Since Retitled as 111inois Emergency Management Agency (IEMA))
]
,
i
.
_
. _..
_
.
. _ _,.__
-
.
However, this was incorrect since the transmitter remained deenergized.
The emergency notification sirens for the emergency planning zone surrounding Clinton station are controlled by two redundant transmitters. The primary is located on the meteorological tower at Clinton station and a backup is located at the fire department in the city of Clinton. When work was performed on the sensors on the meteorological tower, the transmitter was deenergized to prevent a possible radio frequency (RF) hazard to the worker.
The guidance contained in procedure CPS 9437.14, " Metrology System Loop Calibration," revision 32, was to deenergize the transmitter; however, no mention was made of tagging out the transmitter nor was a specific step included to reenergize the transmitter.
The procedure only required that emergency planning be notified to pass a message to ESDA when the transmitter was returned to service. Additionally, no requirement existed to perform a post-maintenance (or return to power) operability test on the transmitter.
After this event the licensee revised CPS 9437.14, to require:
that the transmitter be tagged out in accordance with the
owner control area tagout procedure (NP&S 4.27) if work was to be done on the tower, that the tagout be cleared after the work was done, and
that ESDA perform a " Cancel Test" to verify proper operation a
of the transmitter.
The inspectors have reviewed these corrective actions and agree they are appropriate.
The inspectors also believe that this was an example of poor work control outside of the protected area.
It was apparent from reviewing the procedure and discussions with personnel that normal practices of safety tagging for personnel hazards and verification of equipment operability after maintenance or tag out were not followed for this work.
Post-maintenance testing was performed on the meteorological sensors which had been worked on.
While this was poor performance, it did not have any safety impact since the backup transmitter was functional.
Based on the safety impact, licensee corrective actions, and inspector discussions with licensee management, no enforcement actions are warranted.
g.
Poor Maintenance Practices Contributes to Damaae to the "B" Circulatina Water (CW) Pumo Motor's Lower Bearina The engineering department's evaluation of the failure of the "B" CW pump motor's lower bearing is discussed in paragraph 4.d.
The poor performance by maintenance personnel contributed to the conditions necessary to cause the failure. Three specific contributors were:
l
'
l
.
i
!
Four one pint oil samples were drawn from the lower bearing,
over a 1 year period, and no procedural guidance existed to i
'
replace the sampled oil.
i The oil sight glass on the lower bearing was replaced and
the mark indicating the required level was not correctly
.
positioned.
The required depth of oil in the lower bearing was not
clearly understood from the vendor's manual.
I h.
Observations Of Work Activities The inspectors observed maintenance and surveillance activities of both safety-related and nonsafety-related systems and components listed below. These activities were reviewed to ascertain that i
they were conducted in accordance with approved procedures, l
regulatory guides, industry codes or standards, and in conformance with technical specifications.
l Document Activity D25237 IE22F994 Rework D35525 1E22F004 Cablerun 026415 IB21F032A Actuator D09911 IB21F0228 Rework l
D33044 Main Generator Inspection I
D26416 1821F0328 Rework 9080.22 Div II LOOP /LOCA Test No deviations were identified. One non-cited violation was identified.
4.
Enaineerina i
i a.
Incorrect Modification to the "A" Fire Pump Renders it inoperable
)
Diesel driven fire pump "A" (0FP0lPA) was inoperable for over 3 months due to an improperly designed modification and insufficient post-modification testing. A modification to 0FP0lPA was completed on June 2,1993, which replaced the mechanical overspeed device with an electronic overspeed device.
The fire pump's
design required that it have two batteries, each capable of starting the diesel engine; and that in the event that one battery i
is inoperable, the control system shall lock-in on the remaining battery (NFPA-20, Section 9-5.3.2.c').
However, due to errors in I
the modification's design, this would not occur.
If battery number 2 was inoperable, the loss of voltage to the overspeed device would cause it to lock-out and would prevent the diesel
'
National Fire Code NFPA-20, " Standard for The Installation of Centrifugal Fire Pumps"
l l
.
from starting manually or automatically. Additionally, this design error was not identified in the post modification testing.
It was identified on September 2, 1993, during performance of a routine surveillance test.
The design of this device was not the same as the one installed on fire pump "B".
That device had an auctioneered power supply card.
When the design engineer contacted the manufacture, he was told
<
that this card was not necessary. The engineer interpreted this as meaning only that a " reliable" power supply was needed.
He had the choice of powering the overspeed device from either a bus powered by battery I, battery 2, or a common bus from both batteries. He chose to power the device from bus 2, rather than the common bus.
Consequently, if battery 2 were inoperable---
voltage degraded to less than 9 volts---the overspeed device would trip and the fire pump would be prevented from starting.
Engineering department management performed a thorough review of this event and identified the following causes:
The design basis requirements were clearly specified in the
applicable code, but they were not understood by the design engineer.
The design engineer did not evaluate the potential of this
modification to create a new failure, as it was duplicating a device already installed in fire pump
"B".
Personnel performing the design review and verification
!
function did not have a clear appreciation of the design basis and did not challenge the design.
The post-maintenance test performed a routine surveillance
and did not test to see if a new failure mode had been
'
introduced or prove the design basis.
Plant test engineers were not involved in either the initial
design activities or in developing the test acceptance criteria or methodology.
.
'
Part 50 of Title 10 of the Code of Federal Regulations, Appendix B, Criterion III, requires, in part, that measures shall i
be established to assure that applicable regulatory requirements and the design basis for those structures, systems, and components to which this appendix applies are correctly translated into specifications, procedures, drawings, and instructions.
Part 50 of Title 10 of the Code of Federal Regulations, Appendix B, Criterion XI, requires, in part that a test program shall be j
established to assure that all testing required to demonstrate that structures, systems, and components will perform
satisfactorily in service is identified and performed in accordance with written test procedures which incorporate the requirements and acceptance limits contained in applicable design documents. The failure to design the overspeed circuit to meet the requirements of NFPA-20 is a violation of Criterion III. The i
15 i
'
l
__ _
_
,
_
_
.
.
failure to adequately test the modification to ensure that the overspeed device functioned as designed and that no new failure
modes were introduced is a violation of Criterion XI (461/93019-01(DRP)).
The licensee's corrective actions included reviewing this event with all design, design verification, and system engineers.
Aspects of this included:
Understand the design basis and verify that design input
requirements have been adequately addressed.
Use an independent approach to designing. Don't rely on
previous designs.
Assume that errors exist.
Challenge the design to prove
otherwise.
Perform an intrusive design review process, to determine
what the assumptions were and where the weaknesses are.
Integrate the test engineer into the design process _early to
help define the modification's acceptance criteria.
Test components and systems to verify that they can still
perform their design basis requirements and that no new failure modes have been introduced.
This training was scheduled to be completed by the end of 1993.
Based on the licensee's actions, the inspectors have no further concerns.
This violation is considered closed.
b.
Potentially Inonerable Hiah Pressure Cor'e Spray (HPCS) In.iection Valve During rework of the HPCS injection valve IE22F004, as part of the licensee's NRC Generic Letter 89-10 program, two significant problems were noted.
The first was that the actuator's motor pinion gear key was sheared and the gear had rotated on the shaft.
When the gear was removed from the shaft, the remains of the key were aligned and there was no heat tempering on the shaft or gear.
This indicated that the gear had slipped only one or two times.
The licensee believed this indicated that the gear would operate under normal loads.
However, under design basis loads the gear might rotate on the shaft.
If this happened, the valve would fail
'
to operate.
The key was tested and found to be of the correct material. The motor did not produce sufficient torque to exceed the shear stress limit on the key, so the force must have come from the valve.
Secondly, the valve body guide ribs were out of alignment.
The licensee performed a weld repair and machined the guide ribs to realign them.
The licensee concluded that there were five MOVs with 3600 rpm high inertia motors that might also be susceptible to key failures.
Two have already been inspected; with no
,
'
.
,
t problems found. The remaining three will be inspected during the outage._ Review of the licensee's evaluation will be tracked as
-
unresolved item 461/93019-02(DRP).
,
c.
Failure of NSPS Inverter Fuse F-1
?
Engineering personnel-evaluated the equipment failures and
unplanned actuations that occurred during the Division 11 DG's i
LOOP /LOCA ' test. Based on several of the annunciators which.
came in during the test, an operator was dispatched to the Divisions II NSPS inverter.
He found the dc input fuse (F-1)
blown and the inverter transferred to the alternate ac supply.
'
,
The licensee has theorized that fuse F-1 was blown before the test
!
or in the initial seconds before.the diesel reenergized the ac loads. The ac source of power is fed from the 4.16 Kv bus powered by the Division II diesel.
In either case there would be no power
,
'
to the inverter for a short period of time.
If the manual low pressure coolant injection (LPCI) pushbutton were released before the inverter was reenergized, the LOCA signal would not seal in
,
and the LPCI components would fail to actuate. Additionally, when the NSPS bus was reenergized either a relay race or the power-on-initialization feature caused the ESF logic system to think that a reactor vessel low water level 2 signal existed; thereby, causing the containment isolations and RCIC actuation.
The blown fuse was replaced and the test reperformed on October 1, 1993.
This time fuse F-1 blew after the diesel started and restored power to the 4.16 Kv bus.
Consequently, the NSPS bus
!
remained energized and the LPCI equipment actuated as designed. A review of maintenance records indicated that fuse F-1 has blown several times, including previous performances of this test.
In 1989, the licensee replaced some components on the inverter's
,
control circuit.
This appeared to address the problem and there
'
were no instances of blow fuses since then. The licensee was also evaluating the possibility of induced voltages in either the dc
supply cable or the ground plane affecting the control circuit.
The licensee believes that the ability of the control circuit to gate on and off the silicon controlled rectifiers (SCR)'may be affected. The SCRs are gated on and off in an alternating fashion
-
to produce the 60 Hz alternating waveform.
If the control circuit i
misfires and both SCRs gate at the same time, a direct short across the battery is created.
That short will draw more than enough current to blow the 300 amp fuse F-1.
!
,
7 A loss of offsite power accident coincident with a loss of coolant
'
accident.
!
!
I
i i
.
.
. - - -
-
i
'
.
The licensee was still evaluating possible corrective actions at the end of the report period.
The inspectors will perform further reviews after the LER is issued.
d.
Cause of "B" Circulatina Water Pumo's Bearina Failure Engineering personnel completed a failure analysis of the "B" l
circulating water (CW) pump's motor, which was damaged on April 12, 1993. The cause of the bearing failure was a lack of lubrication caused by low oil level.
The motor was shipped to a vendor and the lower bearing and integral oil pump were examined.
Major damage had occurred to the babbitt and over 50% of it was missing.
The licensee's evaluation concluded the following factors resulted in the low oil level:
The oil level was maintained at the midpoint of the sight
glass, which was 0.5 inches below the vendor required level.
Four one pint oil samples were drawn over a 1 year time
period and no replacement oil was added.
A temperature reduction of 50 F occurred when the temporary
enclosure was removed. This caused a volumetric contraction in the oil level of approximately 3/16 inches.
This combination allowed the oil pump to loose suction and cause the failure.
Contributing factors included:
No procedural guidance to replace the sampled oil.
- Failure to mark the sight glass to the required level when
it was replaced.
Conflicting information between the motor nameplate and
vendor manual on the oil capacity of the lower bearing.
A common misconception that the middle of the sight glass
was an acceptable oil level.
- As corrective action, the licensee checked the oil levels in the
"A" and "C" CW pumps. Both required additional oil and the worst case oil level was 0.75 inches low. Other motors with oil sight glasses will be evaluated to ensure the proper level is indicated.
The inspectors concluded that the licensee had done a good job in evaluating the cause of the bearing failure.
Based on this evaluation, the inspectors have no further concerns on this issue.
e.
Reactor Recirculation Pumo Vibration Monitorina The licensee informed the NRC that it was revising a commitment made in response to a confirmatory action letter issued on June 1, 1989. Confirmatory Action Letter (CAL) RIII-89-016 was issued in June 1989, due to the failure of both seals at the
"B" reactor recirculation (RR) pump. At that time, the licensee i
,
f
.
.
.
.
believed that monitoring of pump vibration would aid in the detection of imminent seal failure. The licensee installed sensors and cabling from the RR pumps to a location outside the dry well (and inside of containment) under a temporary
,
modification. A permanent modification was designed, which would run cabling up to the main control room.
After a year of monitoring the pumps' vibrations and having a consultant review the data, the licensee concluded that. vibration l
was not a useful predictor of imminent seal failure.
Seal
pressures, temperatures, and leakoff flow rates were much better indications of imminent seal failure. Since the seal failure in
1989, the licensee has significantly improved the training procedures, and techniques for installing the seals.
Performance of the seals during the last two operating cycles has been
,
excellent.
Based on this information, the licensee informed the NRC via a letter from J.G. Cook to J. B. Martin, dated September 1, 1993, that it was modifying the commitment made in response to CAL RIII-89-016. The licensee will not install a permanent
modification in RF-4 to monitor RR pump vibration. The licensee intends to continue collecting vibration data for equipment reliability purposes and will evaluate how this function should be made permanent.
!
f.
Reactor Vessel Water Level Modification Testina The licensee discussed its comprehensive plan for testing the i
modification to the reactor vessel water level indication system that was required by NRC Bulletin 93-03.
The testing will verify
,
that hydraulic transients induced in the keep fill system installed to Divisions I and II will not cause spurious transients or equipment actuations in the reactor protection or engineered safety feature actuation systems.
The testing will consist of three parts.
One in cold shutdown with no control rod movement; one in cold shutdown with control
. I rod movement; and one with the reactor critical at low power with pressure greater than 750 psig. The form of the testing consists of inducing hydraulic transients and then monitoring temporary Rosemount transmitters with high speed recorders for evidence of signals large enough to cause actuations.
The licensee also discussed the impact on system operability and how the testing would be scheduled to ensure that technical specification limits are complied with.
Until all of the testing is completed, only one division at a time, will be placed in service and tested. The inspectors.believe the licensee has done
,
a thorough job in developing the test procedure and evaluating the impact on reactor safety and technical specification compliance.
'
.
-
-
-
-
. --
_.
-
.
.
.
-
_
However, the inspectors recommended that the licensee evaluate taking data during single control rod scram time testing performed when the reactor is critical at over 900 psig.
Subsequent to the end of the inspection period, the inspectors were informed that the licensee will collect data during single rod testing at 401
!
power. This will be supplemental data after the modification has been released for operation.
Based cn this additional information, the inspectors have no further concerns on the testing methodology. The inspectors will review the
.;
post-modification test results in a subsequent inspection.
'l
g=
Failure of the Shutdown Service Water (SX) Pumo l A to Start Durino l
A Surveillance Test
,
On October 4, 1993, SX pump 1A failed to start during performance i
of the Division I DG LOOP /LOCA test. When the licensee inspected the breaker, the anti-pumping relay (Y relay) was found energized.
The licensee retested the pump and was able to get this condition to recur. The Y relay was energized because the 52b contacts 7
'
'
and 8 makeup faster than time delay relay 2-SX1PA contacts 3 and t
4.
Contacts 3 and 4 were to open on an undervoltage trip due to j
the bus undervoltage signal.
This resulted in inadvertently i
sealing in the anti-pump coil through relay FC-SX1PA contacts 5 i
i l
and 7.
Consequently, the anti-pump coil would remain energized i
until the dc control power fuses were pulled. This condition l
would not allow the pump to start automatically or manually.
This condition would exist if the pump were running due to a maintained LOCA signal or low pressure from the plant service
water (WS) system signal and a bus undervoltage signal was to occur. The timing of the 52b contacts could be adjusted as a
,
l method of correcting this problem. Another corrective action was
,
eliminating a seal-in feature of the FC-SX1PA relay. The licensee chose the second action to correct the problem. This was
completed on Division I and will.be performed on the Division II
breaker. The inspectors will perform further reviews of this issue after the LER is issued.
No deviations were identified. One violation was identified. One
.
unresolved item was identified.
!
5.
Plant Sucoort
,
t a.
Radiation Protection
.
ALARA Imolications of Testino Safety Relief Valves Onsite
~
!
!
j Paragraph 3.d discussed a licensee initiative in the maintenance
'
l area to perform onsite testing of main steam safety relief valves I
(SRV).
In addition to reducing outage schedule impact and cost, the development of this capability has reduced the total dose that
would have been accumulated. Dose savings will be achieved not j
only by removing and reinstalling fewer SRVs, but in less
>
l l
,
-
-
_,
.
--..
.
,
.
i
-
.
,
.
associated work such as removing, reinstalling, and testing acoustic monitors and removing and reinstalling electrical conduits.
Overall the inspectors concluded that in addition to improving maintenance's ability to test SRVs, this initiative resulted in dose savings. The licensee's initial estimate was that approximately 4 person Rem would be saved.
b.
Security
'
On October 1,1993, the wrong key card was issued to a security supervisor who was responding to an apparent medical emergency.
The supervisor did not identify that he had the wrong key card until he was inside the vital area.
He immediately contacted the central alarm station and obtained an escort to the scene of the emergency. The injured worker was treated and the supervisor was escorted out of the protected area. Security personnel verified that the correct key card was still in the badge rack and initiated an investigation.
The results of the licensee's investigation will be reviewed by a regional inspector during a
,
subsequent security inspection.
c.
Housekeepino Controls (71707)
Tours of the circulating water screen house and auxiliary,
!
containment, control, diesel, fuel handling, rad-waste, and turbine buildings were conducted to observe plant equipment conditions, including potential fire hazards, fluid leaks, i
housekeeping, and cleanliness conditions.
'
During the course of RF-4, declines were noted in housekeeping in
,
some areas, such as the drywell. The licensee's expectations for
work crews cleaning up after their job were not fully met and
!
remediation efforts were implemented.
By the end of the report
'
period the licensee had an action plan in place to address this
!
problem.
i Leaks in the Fuel Handlina Buildina Roof l
On September 15, 1993, a tour of the basement of the auxiliary and
!
fuel handling buildings discovered quite a bit of standing water
,
on the floors of the high pressure core spray, and reactor core
isolation cooling rooms.
Licensee management was aware of this water and attributed it to leaking roofs. The NRC expressed its concerns to the licensee on this issue. The licensee has accelerated repair efforts on the roofs of the fuel and auxiliary
,
'
buildings.
d.
Initial Drywell Inspection (71707)
e The inspectors performed a walkdown of the drywell prior to the start of maintenance activities. There was some debris, such as
]
duct tape and rope in the suppression pool inside the weir wall as-well as in other areas. The drywell cooling system had numerous
i l
.
.
o
,
.
.
leaks.
The licensee surveyed the bottom of the suppression pool inside the weir wall and located a number of tools and pieces of equipment. They removed them and placed netting over the pool inside the drywell to catch any dropped objects. The inspectors relayed their observations to the licensee and will perform a drywell closeout inspection at the end of RF-4.
e.
Licensee Event Report Follow-up (90712 & 92700)
Through direct observation, discussions with licensee personnel, and review of records, the following licensee event reports (LER)
were reviewed to determine that the reportability requirements were fulfilled, immediate corrective action was accomplished, and corrective action to prevent recurrence had been accomplished in accordance with technical specifications.
LG Title 461/92007 Inadequate Logic System Functional Testing-
<
No violations or deviations were identified.
,
6.
Unresolved Items Unresolved items are matters about which more information is required in i
order to ascertain whether they are acceptable items, violations, or deviations. An unresolved item disclosed during the inspection is
'
discussed in paragraph 4.b.
7.
Non-Cited Violation The NRC uses the Notice of Violation to formally document failure to meet a legally binding requirement. However, because the NRC wants to encourage and support licensee's initiatives for self-identification and correction of problems, the NRC will not issue a Notice of Violation if the requirements set forth in 10 CFR Part 2, Appendix C, are met.
Violations of regulatory requirements identified during the inspection, i
for which a Notice of Violation will not be issued, are discussed in paragraphs 2.f(l) and 3.a.
8.
Exit Interview i
The inspectors met with the licensee representatives. denoted in
!
paragraph I at the conclusion of the inspection on October 25, 1993.
The inspectors summarized the purpose and scope of the inspection and the findings. The inspectors also discussed the likely informational content of the inspection report, with regard to documents or processes
,
reviewed by the inspectors during the inspection.
The licensee did not
'
identify any such documents or processes as proprietary.
!
l
l
,
!