IR 05000483/1993010
| ML20024J019 | |
| Person / Time | |
|---|---|
| Site: | Callaway  |
| Issue date: | 08/25/1993 |
| From: | Jackiw I NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML20024J012 | List: |
| References | |
| 50-483-93-10, NUDOCS 9308310158 | |
| Download: ML20024J019 (19) | |
Text
.
.
.. _
'
,
'
.
.
.
.
U. S. NUCLEAR REGULATORY COMMISSION
REGION 111 i
'
Report No.
50-483/93010(DRP)
Docket No.
50-483 License No.
NPF-30 l
Licensee: Union Electric Company
,
Post Office Box 149 - Mail Code 400 i
St. Louis, M0 63166 l
Facility Name: Callaway Plant, Unit I
!
Inspection at: Callaway Site, Steedman, M0
l Inspection Conducted: June 1 through July 31, 1993 Inspectors:
B. L. Bartlett
>
D. R. Calhoun
'
G. A. Pick i
L. R. Wharton J. A. Gavula j
.
.
'
Approved By:
. i{.
ylu
~
,
,
,
kiw, kief, 6253
[
eactor Projects, Section 3A
' Di e
[
-
!
Inspection Summary
,
r Inspection from June 1 through July 31. 1993 (Report No. 50-483/93010(DRP))
Areas Inspected:
Routine unannounced safety inspections of onsite followup of events, review of a piping modification package, maintenance and surveillance,
'
plant operations, and followup on previous inspection findings were conducted.
,
Results: Within the areas inspected one violation and one unresolved item were identified. A summary follows:
!
Operations
!
While operations activities continued to be effectively implemented overall, some isolated weaknesses were noted.
The inspectors identified that licensed
,
operators had been aware of a longstanding minor defect in the diesel generator fuel oil day tank and had not initiated actions to repair the defect. Due to personnel error, the turbine driven auxiliary feedwater pump was returned to service without all of the necessary post maintenance tests being performed. Technical specification requirements were not violated as a result of this error.
Prompt and effective corrective actions were initiated by the licensee. As a result of fuel defects identified during this operating i
9308310158 930825 PDR ADDCK 05000483
-
G PDR
!
-
-.
,,
...,..
.,- -. -. -.. -..
..
-
- -.. -
,
.
.
.
cycle, iht licensee increased coolant sampling and drew up contingency plans
-
for any outage impacts. The fuel defects remain below the required technical specification activity limits.
Radioloaical Controls The fuel defects noted above resulted in a number of challenges to the licensee's health physics program. The licensee increased monitoring of the radiologically contaminated area and initiated corrective action to any potential contamination sources.
Effects of the fuel defects on outage radiological items was being reviewed and addressed by the licensee.
Maintenance / Surveillance Maintenance and surveillance activities were effectively performed by the licensee.
Failure to reset all appropriate relays resulted in the
'
unanticipated start of a safety-related pump.
Documentation of approval to make a minor change to a work document was delayed until prompted by the NRC inspectors.
Corrective action to the inadvertent pump start was noted to be prompt and thorough. At the present time the delay in documenting the minor work change is considered to be an isolated occurrence.
Enaineerina and Technical Support
<
The NRC inspectors' isview of the design of supports to be implemented during the next refueling outage identified weaknesses. The inspectors identified that the design of certain safety-related supports and modifications failed to consider and/or docu ent all relevant design conditions. While none of the l
errors would have relted in the installation of an unsafe design, they did result in designs that were less conservative than the licensee intended.
In addition, weaknesses in the documentation of several calculations were also noted. These weakncsses resulted in an inability for an independent reviewer l
to assess the calculations.
Safety Assessment and Quality Verification
,
l On the last day of this report period the manager of Callaway plant resigned to take a job with another utility. The previous plant manager was rotated
back to the plant manager's position. The change in plant managers is not
'
expected to negatively impact the performance of the licensee.
I
i l
. --.
.
.-.
.
. -
.__
. - _ _
._ __
_ - _ - _ _ _
,
i '
DETAILS
.
l 1.
Persons Contacted
!
D. F. Schnell, Senior Vice President, Nuclear
- G. L. Randolph, Vice President, Nuclear Operations W. R. Campbell, Manager, Callaway Plant
'
- C. D. Naslund, Manager, Nuclear Engineering J. V. Laux, Manager, Quality Assurance
- J. D. Blosser, Manager, Operations Support M. E. Taylor, Assistant Manager, Work Control D. E. Young, Superintendent, Operations M. S. Evans, Superintendent, Health Physics l
- R. C. Stevens, Engineer, Quality Assurance G. J. Czeschin, Superintendent, Planning and Scheduling
'
- H. D. Bono, Supervisor, Quality Assurance Engineering C. E. Slizewski, Supervisor, Quality Assurance Program
- D. J. Maxwell, Supervising Engineer, Design Control
,
_
l
- C. S. Petzel, Quality Assurance Engineer J. A. McGraw, Superintendent, System Engineering
- R. D. Affolter, Superintendent, Design Control
.
- J. M. Gloe, Supervising Engineer
!
- B. R. Newton, Engineer, Welding
!
- Deno'>s those present at one or more exit interviews.
In aduition, a number of equipment operators, reactor operators, senior reactor operators, and other members of the quality control, operations, l
maintenance, health physics, and engineering staffs were contacted.
2.
Onsite Follow Up of Events (93702_1 Fuel Assembly Defects The licensee identified that fuel defects had been causing reactor coolant system (RCS) activity to increase.
Intensive monitoring of the RCS activity was begun along with the development of contingency plans.
l l
Technical Specifications (TSs) require that the specific activity of the RCS be limited to less than or equal to one microcurie per gram Dose
!
Equivalent Iodine (DEI). DEI is that concentration of Iodine-131 which alone could produce the same thyroid dose as the quantity and isotopic mixture of I-131, I-132, I-133, I-134, and 1-135 actually present in the l
RCS. The limitations ensure that in the event of a steam generator tube
!
rupture that the resulting doses at the site boundary do not exceed a small fraction of the 10 CFR Part 100 dose guidelines.
The licensee increased the RCS letdown rate in order to maintain the RCS activity as low as possible. The licensee is continuing to monitor the DEI.
,
!
.
_ _ _ _ _ - _.
.
.
!
The licensee has formulated an action plan to ensure that long and short i
term ramifications of the fuel defects have been addressed.
Items in
!
the action plan included:
Increasing survey points and frequency to identify
.
radiological concerns within the plant.
Performing daily analysis and trending of iodine and noble
.
I gases.
Performing weekly loose parts monitoring comparisons with
.
baseline data.
Performing neutron noise analysis for indication of abnormal
.
core vibration or performance.
Limiting power level changes to a maximum rate of three
.
percent per hour.
Verifying that the existing safety analysis for normal
.
releases and accident analyses were bounded for DEI limits.
Evaluating possible refueling outage impact on fuel l
.
movement, personnel dose and surveys, and hot particle minimization.
!
l At the end of this inspection period the DEI was about 25% of the TS l
limit.
Turbine Driven Auxiliary Feedwater Pump Retest On June 21, 1993, the turbine driven auxiliary feedwater pump (TDAFP)
was restored to an operable condition without performing all the required retests for declaring the pump operable. After discovering the missed retest, the licensee had approximately 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> remaining in the action statement. This allowed sufficient time to successfully perform the retest.
Since the retest was performed within the required time limit, a TS violation did not occur.
Following maintenance activities on the TDAFP, required post maintenance tests were performed, with one exception.
Valve FC V-0312 (mechanical trip and throttle valve) was required to be fully stroked and timed to ensure operability. The valve was stroked but the procedure to time it was inadvertently not performed.
It appeared that the control room
!
operating supervisor (0S) overlooked the retest during the review of work documents.
l During an event review team (ERT) meeting on this event, several
'
contributing factors were identified.
First, the retest requirement was l
not in the preventive maintenance or the pump run work packages when the l
work documents were sent to the control room (CR).
Second, the operating crew's review of the component outage completed paperwork was
L
!
l
<
The Federal Emergency Management Agency, the State Emergency Management Agency, and the licentee were contacted to ensure that all emergency planning functions could be carried out during the flooding. The emergency plan was originally designed assuming that the Missouri River was flooding.
Using alternative routes and utilizing the Missouri National Guard as necessary, all personnel in the EPZ would be
adequately protected.
l
!
At the end of the report period the Missouri River was continuing to flood but the level was dropping with some roadways becoming again passable.
3.
Facility Modifications (37701)
As documented in NRC Inspection Report No. 50-483/92016, the essential
!
l service water (ESW) system experienced several waterhammer events in the
'
l past. The primary cause was determined to be a fundamental design l
weakness which allowed column separation at the containment coolers
during certain operational events. The licensee's proposed resolution was to quantify tue waterhammer forces and modify the piping system to accommodate these newly calculated loads. Modification, CMP 92-1034,
" Modify and Add Pipe Supports on ESW Lines to Containment Coolers," was scheduled to be implemented during the next refueling outage that begins October 1, 1993.
,
The NRC inspectors reviewed portions of the following calculations and
,
support drawings for compliance with licensee commitments and NRC requirements, and had the following comments:
,
l Calculation No. P-GN01-01 CMP, " Review of supports GN01-C003, GN01-C007,
'
and EG13-C004," Revision 0, Addenda 1, January 12, 1993.
- The calculation concluded that the size of weld No. 3 should be
!
0.427 inches; however, the size was only specified as 0.375 inches.
The analyst noted that this was "close enough" based on the conservative method of modeling the structure.
While the model assumptions were conservative in some aspects, they were non-conservative in others. The bases of this engineering judgment were not well developed nor documented.
This was considered to be a weakness by the NRC inspector.
l
- The calculation did not address the adequacy of the embedded plates.
During discussions with the NRC inspector, the analyst
,
stated that they were considered acceptable based on engineering judgment. However, there were no documented bases for this judgment and no reference document regarding the capacity of the embedded plates. The lack of documented bases for engineering judgments was considered a weakness by the NRC inspector.
!
- The load on the lug weld was assumed to be pure shear; however, because of the radius on the corners of the tube steel, the actual l
load application induced significant bending moments.
The
l
!
l
.
__
_
__
_
_.-
..
._
- - -
.. _. _ _
.
_ _ _ _
_ _
.
-
calculation did not calculate nor evaluate the effects of the i
-
induced bending moment. The failure to verify the adequacy of the design was considered to be an example of a violation of 10 CFR l
50, Appendix B, Criterion III, Design Control.
(483/93010-01A)
- The support model assumed a pinned connection at one end of the 8x4 tube steel member; however, the weld configuration specified on the drawing resulted in the connection being fixed instead of pinned. This invalidated the loads used-to evaluate the welds and the weld configuration evaluated in the calculation was not the same as specified on the drawing. -The failure h corr u tly
,
translate design bases into the drawing is consW2rsd an example
'
of a violation of 10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-01B)
]
,
- The analysis utilized-the same torsional properties for both the i
horizontal I-beam and the tube steel members, even though the ' tube j
steel member was.significantly stiffer than the I-beam. This j
discrepancy compounds the pinned connection issue discussed above
_
in that the tube steel member resists more of the torsional moment i
than given in the analysis. The failure to verify the adequacy of J
the design was considered to be an example of a violation of 10 l
CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-l 010)
{
i Calculation No. P-GN01-07 CMP, " Review of support GN01-C020/231,"
i Revision 0, Addenda 1, January 12, 1993.
)
- The calculation stated, " based on the output stresses and connection loads, all welds are considered adequate by inspection." Although the calculation included a marked-up print j
of the support, the sizes shown on the marked-up print were larger than those specified on the final support drawing. _ During discussions with the NRC inspector, the analyst stated that the smaller size welds had been considered, but there was no documented basis for the difference. The lack of documentation associated with the engineering judgment was considered to be a weakness by.the NRC inspector..
- Because -the original support could not be modified to accommodate the axial loads, a new support structure was design at a location 4.5 feet from the analyzed location.
From a pipinc perspective, this change was analytically acceptable: however, because the lateral restraint did not exist at this iocation, new lateral displacements shoulo have been considered 'or the support.
Because of the initial support gap and the radius an the interfacing tube steel, the lateral displacement potentially i
shifts the pipe to one side and causes the axial load to be unequally applied by the two lugs. The failure to subject the
-
change in design to controls commensurate with the original design was considered an example of a violation.of 10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-010)
i l
---
-e v.
,
cr...
-,y
%.
., -,
,-m..-rs y
,--.
o,-p
-,
,
- -
--
...
--.
.-
~
-
-
-
.
'
- The 3/4-inch lateral gap between pipe and support structure combined with the 1/2-inch radius on the tube steel resulted in a
-
i significant bending moment being applied to the shear lug. The l
bending moment was not calculated.nor considered in the support
!
evaluation. The failure to verify the adequacy of the design was J
considered an example of a violation of 10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-ole)
l Calculation No. P-GN02-09 CMP, " Review of support GN02-R016/231,"
l Revision 0, Addenda 1, January 15, 1993.
l l
l l
- The evaluation of the lug weld only considered the shear
'
component of the applied load. This methodology neglected the bending moment ~ caused by the initial 1/2-inch gap-between the pipe and the tube steel as well as the 1-inch radius of the tube steel.
The failure to verify the adequacy of the design was considered an
{'
example of a violation of 10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-01F)
- The initial support design specified a wall thickness of 1/4 inch for the 3x5 tube steel; however, in an attempt to minimize the variety of material being ordered, the thickness was I
changed to 1/2-inch wall tube steel. Although this was a stronger
'i member, the material was changed without considering the effect on
,
the construction of the support. The thicker wall of the tube steel increased the corner radius from 1/2 inch to 1 inch. This no longer allowed the specified fillet weld to be_ made,'but instead required a flare bevel grove weld. Also the larger corner radius effectively increased the bending moment being applied to the shear lug..The failure to subject the change in design to controls commensurate with the original design was considered an j
example of a violation of 10 CFR 50, Appendix B, Criterion III, i
Design Control.
(483/93010-OlG)
Hanger Drawing Nos. 2-GN01-C009/252 Revision 1 and 2-GN01-C019/252 Revision 3.
- These two drawings were to be worked together since they were structurally attached to each other. None of the horizontal members were dimensionally. located with' respect to the piping centerline.
Instead, gaps between.the pipe and structural
!
elements were specified with shims shown "as required." Based on j
the centerline distance between the two pipes, the support could
'
not be constructed because-the lengths of the vertical members was insufficient to span the given distance. Depending on which vertical member was eventually changed, the shims on the upper support would affect the moment induced into the lug weld. The failure to correctly translate the design basis into the drawing is considered an example of a violation of 10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-01H)
I t
!
!
- The size of the fillet weld attaching the lugs to the pipe was specified as 1/4 inch; however, the original code of construction, American Society of Mechanical Engineers (ASME) Boiler and
>
Pressure Vessel Code (Code),Section III, Appendix XVII, Table XVII-2452.1-1, " Minimum Size of Fillet Welds," gives a minimum size of 5/16 inch for this configuration. When questioned by the NRC inspector, the licensee stated that subsequent Code Cases and later editions of the Code allow smaller sizes to be specified.
These documents, however, were not referenced in any of the calculations or the design specification. The lack of a documented design basis was considered a weakness by the NRC inspector.
Hanger Drawing No. 2-GN01-R009/252, Revision 2.
!
l
- The drawing specified a 1/16 in gap between the bottom of the pipe and the support structure.
In accordance with UE's design practice, this results in the support being active in the vertical direction. However, no vertical loads were specified for this support and it was not analyzed to resist any loads in the vertical direction. The failure to correctly translate the design basis into the drawing is considered an example of a violation of l
10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-l 011)
Hanger Drawing No. 2-GN01-R014/252, Revision 6.
- The location of item No. 4 was specified with a vertical l
dimension without giving an angle or a horizontal dimension.
The licensee stated that the length of this 6-inch structural brace, given in the Bill of Material, combined with the vertical dimension was sufficient to locate the member.
Based on the NRC inspector's calculation, this method resulted in an angular dimension of approximately 55* and the brace could not be installed as shown on the drawing. According to the licensee, the support calculation utilized an angle of 45 in the analysis of the structure. The failure to correctly translate the design basis into the drawing is considered an example of a violation of 10 CFR 50, Appendix B, Criterion III, Design Control.
(483/93010-01J)
The NRC inspectors reviewed a sample of quality assurance audits and surveillances in an effort to determine what assessment QA had performed in the engineering area. While none of the audits were specifically on piping, the audits reviewed did verify assumptions, check calculations, and did perform independent calculations. The lack of QA assessment in the piping area appeared to be an isolated occurrence of missing a narrow focus, highly complex area.
I
. _.
.
}
.
Conclusions One violation with multiple examples was identified.
The quality of the engineering evaluations for the reviewed modification was poor.
Documentation was weak, assumptions were poorly stated or sometimes not stated at all, engineering judgment was used where inappropriate, and when the design was changed during the design phase, additional errors were introduced. While none of these weaknesses resulted in a failure to meet design codes they did show significant deficiencies.
,
'
t l
4.
Maintenance / Surveillance (627C5) (61726)
Selected portions of the plant surveillance, _ test, and maintenance
'
activities on safety-related systems and components were observed or i
reviewed to ascertain that the activities were performed in accordance with approved procedures, regulatory guides, industry codes and standards, and the Technical Specifications. The following items were considered during these inspections:
the limiting conditions for operation were met while components or systems were removed from service; approvals were obtained prior to initiating the work; activities were accomplished using approved procedures and were inspected as applicable; functional testing and/or calibration was performed prior to returning the components or systems to service; parts and materials that were used were properly certified; and appropriate fire prevention, radiological, and housekeeping conditions were maintained.
a.
Maintenance The reviewed maintenance activities included-
-
Work Reouest No.
Activity C534012 Re-route piping and relief valve to diesel
generator "A" auxiliary lube oil pump.
C469351 Replace _ obsolete time delay relay.
W149385 Replace gaskets at safety injection pump
"B" strainer.
G536027 Clean and unplug "B" ESW pre-lube tank.
P507167 Inspect pre-lube tank strainer
"B".
W157254 Replace relief valve, fuel pool heat exchanger "B" tube side.
W145111 Replace flange gasket on valve EC V-033, fuel pool heat exchanger "B" to fuel pool cleanup.
_ _ _ _ _ _ _ _ _ _ _ _ -
-
.
W158789 Flush out instrumentation tubing of i '
differential pressure transmitter.
!
C533562 4GKYO2BF cable pull from NN04 to NG04C.
Time Delay Relay Replacement During the performance of C469351 the workers were required to add some minor steps to their procedure. They received verbal permission to add the steps but did not document the changes in the procedure until l
prompted by the inspectors.
C469351 replaced an obsolete time delay relay in the ultimate heat sink
,
cooling tower fan motor control circuit, The existing relays could no longer be purchased, so on a periodic basis they were being changed out
'
with a new style of relay.
The work had been performed previously on the "B" train without any problems being encountered. However, due to the arrangement of items in the breaker cubical being slightly different on
"A" train than on
"B" train there was not enough room to properly drill the new mounting
,
holes. A small 480 v to 120 v ac transformer and a relay would have to i
[
be moved to allow room to perform drilling and tapping activities in the sheet metal for mounting the new style relays.
i The workers and the quality control inspector had discussed the additional work with the responsible construction engineer as required by procedure. The responsible engineer informed them to go ahead and temporarily remove the transformer and relay but to document all work performed in the comments section of the work request.
i The NRC inspectors began observing work activities immediately prior to l
any equipment being removed. When the inspectors noted that the procedure did not address removing other equipment they questioned the
'
,
i workers. The workers informed the inspectors that engineering had approved the minor scope change verbally. Additional discussions with the workers revealed that the work document should be changed before
work was performed to match the new scope. The engineer and work
!
supervisor added a note to the work request authorizing the removal and l
reinstallation of the additional components and then the work continued.
l b.
Surveillance The reviewed surveillances included:
i
'
Procedure No.
Activity ETP-GN-0001B Train "B" containment cooler performance test,
i i
l l
I
_ - _ _ _ _ _ _ - _ _ -
ITG-ZZ-GPS 01 Calibration of diesel generator
"A" cylinders one through seven turbocharger inlet pressure switch.
MSE-NK-QB001 Weekly battery inspection on large station batteries, NKll and NK13.
OSP-AL-P0002 Section XI turbine driven auxiliary feedwater pump operability.
Unplanned Start of A S:fety-Related Pump On July 12, 1993, while performing an operations surveillance procedure, the "A" essential service water (system designator EF) pump auto-started. The reactor operator (RO) immediately secured the pump and began investigating the unintended pump start. The pump started when a nonsafety-related circuit was made up.
During the surveillance procedure EF flow was throttled causing a temporary low flow condition.
Previous work activities had caused a relay which indicates an undervoltage condition on the bus to be energized.
The low flow coincident with an indicated low voltage condition resulted in the automatic pump start.
Initially, the licensee reported the engineered safety features component actuation pursuant to 10 CFR 50.'72 requirements. After further review of the event, the licensee rescinded the reportability call on July 22, 1993.
Instrumentation and control (I&C) technicians along with the system engineer had been requested on July 2,1993, to troubleshoot the load shedding and emergency load sequencing system (system designator NF),
NF039B, due to system problems.
The I&C personnel identified a failed power supply card on channel III. As part of the repair efforts, both undervoltage channels I and III were downpowered, for personnel safety concerns in accordance with workman's protection assurance (WPA). After the defective card was replaced, the I&C personnel proceeded to perform retesting activities on channel III as directed by the work instructions.
There was no requirement to perform testing of channel I since no work had actually been conducted on the channel.
However, when channel I had been downpowered, this had caused the energization and seal-in of relay
,
'ARE'.
This relay is normally de-energized and was associated with a
'
nonsafety-related circuit that initiates an EF pump start when certain conditions are met.
.
A note which directed the resetting of relay 'ARE' to prevent an l
inadvertent pump start was only included on the EF drawings. Thus, the l
energized condition of relay 'ARE' went undetected because the I&C
'
personnel relied upon the NF drawings during restoration.
Relay ' ARE'
had been left in an energized state from July 2 until July 12, 1993.
!
l
l l
l
!
i
-
.
<
,
-
J The corrective actions included.
.
-
,
i Adding this event. to the training cycle for operators, I&C
.
,
planners, and I&C engineers.
Adding a note to~the computerized equipment list discussing the
.
need to reset the relay. This would ensure that when the i
component was included in work activities that-the note would be automatically listed.
>
>
Adding a note to the WPA file for this activity. Whenever a WPA-
.
,
was brought up for use'on this equipment the note would be
,
automatically included.
l j
Conclusions i
Whenever the workers encountered the need' to modify their work l
instructions they properly contacted their supervision and engineering.
.'
However, the workers need to ensure that the work instructions properly reflect any additional guidance supplied during these discussions.
The root cause analysis and corrective' actions to the inadvertent EF
pump start were prompt, thorough, and detailed. -The necessary procedure i
L and process changes were. initiated to prevent reoccurrence.
'
No violations or deviations were identified.
l 5.
Plant Operations (71707)
J l
The objectives of this inspection were to ensure that the facility was j
l being operated safely and in conformance with license. and regulatory i
l requirements and that the licensee's management control systems were
effectively discharging the licensee's responsibilities for continued
!
safe operation. The methods used to perform this inspection included i
direct observation of activities and equipment, tours of.the facility, interviews and discussions with licensee personnel,-independent verification of safety system status and limiting conditions for operation, corrective actions, and review of facility records.
t
'
Areas reviewed during this inspection included, but were not limited to, control room activities, routine surveillances, engineered safety feature operability, radiation protection controls, fire protection, I
security, plant cleanliness, instrumentation and alarms, deficiency reports, and corrective actions.
Section XI Testing L
The NRC inspectors identified that a number of components were being
)
'
unnecessarily tested. The components had been placed on an increased testing frequency but had na been evaluated for return to a normal testing frequency. This resulted in unnecessary operator distractions,
l
.
- __
-_ _,.
_
..
_, ~.,. _ -.... - -
-
.
i
.
unnecessary pump starts and valve manipulations, and resulted in
,
additional out-of-service time on the emergency diesel generators.
{
-
During a routine tour of the control room the inspectors had noted a work request (WR) on the discharge pressure gauge of the "B" safety injection (SI) pump. The WR stated that the pressure gauge was reading about two percent low. The reactor operators informed the inspectors l
that the SI pump was in an increased testing frequency due to low differential pressure.
TSs require that the licensee periodically test certain components in
!
accordance with the ASME Boiler and Pressure Vessel Code,Section XI.
The testing for most components occurs on a quarterly basis.
If during
.
this testing, the component has been found to be degraded, then the
!
testing interval must be shortened (usually to a monthly interval) or l
the component declared inoperable if appropriate. Due to the discharge pressure gauge reading low during the performance of the quarterly test of the SI pump, it appeared that the pump was developing a low i
differential pressure. This caused the pump testing frequency to be
i changed from quarterly to mor.thly.
i l
The inspectors were concerned that even though the pressure gauge had
,
l been identified as reading low before the quarterly test, the licensee
!
had performed the test before recalibrating the gauge. This caused the
,
SI pump to unnecessarily be declared degraded and placed on an increased
!
testing frequency.
It also resulted in additional unnecessary pump l
starts and unnecessary operator work load. At the time of discovery, the pump had been on increased testing for slightly over one month. The licensee stated that when the gauge had been identified as reading low that it had been scheduled to be recalibrated before the next performance of the quarterly test.
However, the schedule had been mis-read and the test was performed before the gauge was recalibrated.
In following up on this issue the inspectors requested that the licensee supply them with a list of all pumps and valves that were on an
,
increased testing frequency. The data showed that there were 28 items on increased testing frequency with the oldest item dated January 1987.
Four of the components were safety-related pumps and the remaining items were safety-related valves. The oldest pumps on the list were the two l
emergency diesel generator fuel oil transfer pumps, which had been on an increased testing frequency since May 1988. With the exception of the fuel oil transfer pumps which are discussed below, the unnecessary testing did not result in any significant problems.
It did result, l
however, in some unnecessary wear and tear on components and unnecessary work activities in the control room.
Fuel Oil Transfer Pumos Surveillance Testina The fuel oil transfer pumps take a suction on the main fuel oil tanks and supply fuel to the diesel generator (DG) day tank. The fuel oil transfer pumps operate to ensure that the DG day tank maintains its fuel
oil supply.Section XI required that the pumps be periodically tested l
i
!
l
- ____
,
to ensure that pump flow and discharge pressure were not degraded.
There was no flow indicating device in the fuel oil transfer pump discharge line to the day tank.
The licensee tested the pumps by
-
draining the day tank and timing how long it took the transfer pump to
'
refill the tank. During the time that the day tank was partially drained the DG is without a source of fuel supply and is inoperable.
The licensee did not fully drain the day tank to perform this test.
'
This meant that the inaccuracies in the test method combined with the small test volume resulted in inconsistent flowrate data.
The inconsistent data caused the fuel oil transfer pumps to be placed on an increased testing frequency.
The licensee had attempted to correct the inconsistent data by draining the day tank further thus resulting in a larger test volume. This resulted in more consistent data but it also resulted in longer out-of-service times for the DG in order to allow more time to drain the tank. The drain systems of the two tanks were different so that on average it took 4h hours to perform the Section XI test on the "A" DG and 7 hours8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> to perform the test on the "B" DG.
With the pumps being on an increased testing frequency, they were being tested once per month instead of once per quarter.
By not promptly initiating corrective action to improve the testing methodology and then failing to remove the pumps from the increased frequency, up to two thirds of all at-power out-of-service time was unnecessary.
Using test records, the inspectors determined that between June 1990 and May 1992 the "A" DG had approximately 52 hours6.018519e-4 days <br />0.0144 hours <br />8.597884e-5 weeks <br />1.9786e-5 months <br /> of unnecessary out-of-service time and the "B" DG had approximately 66 hours7.638889e-4 days <br />0.0183 hours <br />1.09127e-4 weeks <br />2.5113e-5 months <br /> of unnecessary out-of-service time while the unit was at power.
Section XI Testino Corrective Actions The licensee determined that there was no formal internal requirement to evaluate items that were on an increased testing frequency and to have them removed as soon as possible. Through analysis and review of testing data, the licensee determined that all 4 pumps and 12 of the
,
valves could be removed from increased testing and placed back on quarterly testing.
In addition, those items on an increased testing frequency will be regularly reviewed by management to ensure removal as soon as possible.
Diesel Generator Day Tank Internal Leakage During a routine tour of the control room the inspectors noted that there was an apparent discrepancy in the indicated level of the two DG day tanks. Discussions with the reactor operator revealed that she knew of the discrepancy and that it was due to internal leakage of the two t
tanks. Apparently one of two valves in the lines to and from the main fuel oil tank for each day tank was leaking slightly. This caused the day tank level to slowly drop until it reached the low set point, the transfer pump would then start and refill the tank.
An interview of the other reactor operator revealed that she too was aware of the internal leakage.
Interviews and analysis of limited computer data showed that
l
!
i i
.
l
i
.
the "A" day tank transfer pump was refilling the day tank about once j
every 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> and that the
"B" day tank was being refilled about once
'
every day.
Interviews with other reactor operators revealed that most of them were aware of the problem and that it had been in existence for several years. The system engineer was also knowledgeable of the problem. The senior reactor operators and plant management were unaware of the internal leakage.
'
While the internal leakage did not affect the operability of the DGs, it was an example of a minor maintenance problem that was not being repaired. Work requests to identify and repair the source of the leakage should have been written.
Instead, the system engineer and the l
R0s assumed it was not worth bringing to management's attention. The R0s and/or the system engineer should have questioned the leakage and
'
initiated repair efforts instead of accepting a known deficiency.
Temporary Repair of a Service Water Leak A 4-inch diameter pipe leading to one of the safety-related control building air conditioners developed a small leak. The leak was promptly repaired and the piping returned to service. The NRC inspectors developed concerns that while the repair was acceptable for temporary repair, it may not have met ASME code requirements.
On July 17, 1993, a non-licensed equipment operator discovered a small leak during a routine tour of the control building. The leak was about 2 drops per minute. The leaking component was a 4-inch, ASME Class III (low pressure) pipe leading to a safety-related air conditioner. The licensee performed ultrasonic testing of the pipe to identify the extent of the defect and initiated actions to repair the leak.
The leak was found to be localized pitting most probably due to microbiologically induced corrosion (MIC).
It was decided to grind out the thin wall areas and fill the hole with weld material. The system was then pressure tested and returned to service.
Due to previous problems with MIC, the licensee had already scheduled the pipe to be replaced during the upcoming refueling outage (scheduled to start October 1, 1993).
The NRC inspectors became concerned that while the repair was acceptable for temporary repair, it may not meet all pertinent sections of the ASME code for repair of a through wall leak. Discussions were held with the licensee and members of the NRC technical staff.
At the end of this inspection period the technical staff had yet determined whether the repair met code requirements. Consequently, pending the decision of the NRC technical staff this issue will remain as an unresolved item (483/93010-02).
,
i
~
.
l l
.
-
!
Foaming of Containment Penetrations During Refueling Outages
.
During refueling outages TS 3.9.4 requires that each reactor building penetration providing direct access from the reactor building atmosphere to the outside atmosphere be capable of being closed by an isolation
valve, blind flange, or manual valve. The licensee had determined that the use of temporary penetration barrier material in lieu of a blind l
flange was an acceptable method for maintaining containment closure.
This is due to the need to utilize spare containment penetrations to run cables, piping, and hoses to support steam generator inspections.
While reviewing industry information the licensee observed that another facility had no configuration control of the temporary barriers.
The licensee evaluated the control and use of temporary barriers and determined that additional controls should be implemented.
The licensee'c temporary barriers consisted of silicone foam seals that
,
,
!
were installed after the penetrating items were installed through the l
!
The licensee had previously used a procedure to install the barriers but this procedure was utilized for all types of penetrations. The licensee decided that the containment penetrations should be performed using a separate procedure with Quality Control t
inspections. At the close of the inspection period the procedure was l
still in draft but it was scheduled to be finalized prior to the start l
of the next refueling outage.
t i
As part of the analysis to determine that foamed penetrations were l
acceptable in lieu of flanges, the licensee compared the pressure rating l
of a properly installed foam penetration to expected containment l
pressure following postulated refueling outage accident scenarios.
Through testing at a separate facility the licensee had previously determined that the foamed penetrations would remain functional with up to a 6 pounds per square inch pressure differential (psid).
UE calculation M-YY-44 performed April 28, 1993, determined the peak
,
containment pressure during a refueling accident. Some conservative assumptions used in the calculation were:
j The fuel decay heat value was at maximum in that refueling
.
was started with the minimum decay time allowed by TS.
The temperature reducing effects of the concrete and steel
.
inside containment were ignored. During a refueling accident as steam and water vapor would be given off, the concrete and steel of the containment structure would be expected to act as a heat sink helping to lower peak containment temperatures and pressures.
l Containment leakage was assumed to be zero. This would
.
maximize the amount of air inside containment and increase the expected peak pressure conservatively.
_
-
.
.
.
The release of steam causing containment pressure to l
-
.
l increase was not limited to only the existing reactor
'
coolant system inventory.
The peak pressure calculated by the licensee was 5.6 psid which was less
-
than the 6.0 psid that the foamed penetrations could be expected to handle. However, the licensee also assumed that two of four containment coolers would be available to help cool the containment atmosphere and limit the peak pressure. The licensee intends to keep the coolers
available during refueling outages but does not intend to maintain them
in an operable condition.
l No violations or deviations were identified.
6.
Followup On Previous NRC Inspection Findinas (92701)
(CLOSED) VIOLATION 483/92012-01:
Failure to have a locking device installed on damper GK D-0325 and the upper cable spreading room exhaust register.
l
'
During a walkdown of the control room ventilation isolation system (system designator GK), the inspectors identified that damper GK D-0325 failed to have a locking device installed. Upon notifying plant management, a subsequent system walkdown, by utility personnel, of all GK dampers was undertaken. This audit discovered one additional register without a locking device.
Both damper and register were verified to be in the correct position and locking devices were installed on each component.
- The licensee's corrective action to prevent future occurrences included revising the locked component control procedure, ODP-ZZ-00004.
It was believed that the locks were removed during flow verification activities by either the system engineer or maintenance personnel. The procedural revision limited the removal and restoration of locks to operations, l
radwaste, and chemistry personnel.
Procedures MPE-GK-QC001 and HPE-GK-QC002, " Control Room Emergency Ventilation System Trains A and B Flow
'
Verification" also were modified to direct operators to remove and install all locking devices as well as maintain the locked component deviation log in accordance with the procedure.
These procedures required notifying the CR upon completion of testing so that locks can be installed.
In addition, retests GK PM-01 and GK PM-02
,
were generated to have the operations department perform a walkdown to ensure all locks had been properly re-installed. The inspectors verified all applicable procedural changes had been made and that
)
retests were scheduled under the work request program.
This violation is considered closed.
!
.
4.m
_A-
-
.
,
l l
e l
!
!
-
7.
Unresolved Item i
I Unresolved items are matters about which more information is required in order to ascertain whether it is an acceptable item, a violation, a failure to meet a licensee commitment, or a deviation. An unresolved item is discussed in paragraph 5.
.
!
8.
Exit Meetino (71707)
The inspectors met with licensee representatives (denoted under Persons Contacted) at intervals during the inspection period. The inspectors summarized the scope and findings of the inspection. The licensee representatives acknowledged the findings as reported herein. The t
l 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 / processes as proprietary.
l
,
I i
.
i
- -
-
-.