ML20202E340
| ML20202E340 | |
| Person / Time | |
|---|---|
| Site: | Arkansas Nuclear  |
| Issue date: | 02/06/1998 |
| From: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV) |
| To: | |
| Shared Package | |
| ML20202E294 | List: |
| References | |
| 50-313-97-21, 50-368-97-21, NUDOCS 9802180126 | |
| Download: ML20202E340 (50) | |
See also: IR 05000313/1997021
Text
. .- . - . - - _ - - . ~ . - - _ . - . . . . - - - . . . .-
.
..
EliCLOSURL2 i
U.S. NUCLEAR REGULATORY COMMISSION
REGION IV
-
- Docket Nos.: 50-313;50-368
Report No.: 50-313/97-21; 50 368/97-21
Licensee: Entergy Operations, Inc.
Facility: Arkansas Nuclear One, Units 1 and 2
]
- Location: Junction of Hwy,64W and Hwy.333 South
Russellville, Arkansas
Dates: October 27-31 and November 10-14,1997
, inspectors: M. Runyan, Team Leader
P. Goldberg, Reactor inspector
R. Bywater, Reactor Inspector
G. Kalman, Project Manager
'
Accompanying D. Denver, Consu' tant
Personnel
4
Approved By: T. Stetka, Acting Chief Engineering Branch, Division of Reactor Safety
'
Attachment 1: Supplemental Information
l Attachment 2: List of Documents Reviewed for Unresolved item 50-313/97201-06
!
l
l
l
l
[
,
l
'
i
9902180126 900206
- PDR ADOCK 05000313
i
G PDR
!
_ _ _ _ _ _ - _ __ - _-_-__
.
.
-2-
EXECUTIVE SUMMARY
Arkansas Nuclear One, Units 1 and 2
NRC Inspection Report 50-313/97-21; 50 368/97 21
Engineerina
-
Within a review of selected documents, the licensee's 10 CFR 50.59 safety evaluation
program appeared to be functioning satisfactorily, with the exception that some
documented word searches of the Updated Safety Analysis Report for impact were
inaccurate. However, some apparent failures to implement the program were identified
(Sections E2,1 and E8.22).
.
The failure to procedurally incorporate a vendor recommended maximum emergency
feedwater flow limit of 1500 gpm to a single steam generator into plant operating
procedures was identified as a violation of 10 CFR Part 50, Appendix B, Criterion V
(Section E8.7).
The failure to properly verify the piping configuration design inputs to a design calculation
was identified as the first example of a violation of 10 CFR Patt 50, Appendix B,
Criterion lli (Sec' ion EP.8).
Overpressure of the turbine-driven emergency feedwater pump downstream piping
required further reviews by the licensee. The existing analysis used pump performance
crit 6 ia that was less than the manufacturer's specified criteria (Section E8.9).
-
The failure to properly verify the design adequacy in four ciesign calculations was
identified as additional examples of a violation of 10 CFR Part 50, Appendix B,
Criterion 111 (Section E8.11).
The lack of a review by the licensee of a sample of existing design calcriations to
determine if the types of past problems identified by the licensee existed within other
calculations was identified as a weakness (Section E8.11).
-
The lack of over current trip testing of Unit 1 molded case circuit breakers required
further review by the NRC (Section E8.16).
-
- Although the NRC identified five discrepancies in the Final Safety Analysis
Report (FSAR), the licensee's ongoing review of the FSAR was considered sufficient to
correct these specific examples and to resolve the resulting generic concern. Therefore,
enforcement discretion was taken in accordance with the NRC Enforcement Policy
(Section E8.18).
- ___
. _ - _ _ - - _ _ _ _ _ - _ _ _
.
9
-3-
.
The failure to provide measures to ensure the correct translation of seismic design
requirements to instrument tubing installations and the maintenance of this configuration
during plant operations in Unit 1 was identified as an example of a violation of 10 CFR
Part 50, Appendix B, Criterion lli (Section E8.20), ,
.
Three apparent violations were identified associated with activities involving the removal
of the borated water storage tank vacuum relief valve for maintenance and testing on
December 4,1996. The apparent violations involved the failure to perform proper safety
evaluations as described in 10 CFR 50.59, the failure to perform the vacuum relief valve
testing under the plant conditions as prescribed in the test procedure in accordance with
10 CFR Part 50, Appendix B, Criterion V, and the failure to utilize the correct design
control processes (temporary alterations) for the temporary cover configurations placed
over the valve flange in accordance with 10 CFR Part 50, Appendix B, Criterion 111
(Section E8.22).
-
The failure to properly check the adequacy of an engineering report that failed to account
for instrument error in a calculation of vortexing in the borated v<ater storage tank was
identified as an example of a violation of 10 CFR Part 50, Appendix B, Criterion 111
(Section E8.23).
Plant Succort
'
=
During a plant walkdown, the NRC identified a fire barrier deficiency and the licensee
took prompt compensatory actions as required (Section F2).
1
_ _ - _ _ _ _ _ _ _ .
_ - - _ _ _ _ _ - _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ - -_ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _
.
.
-4
Report Details
Summarv of Plant Status
Units 1 and 2 were operated at 100 percent power during the inspection.
ll1. Enaineerina
E2 Engineering Support of Facilities and Equipment (37001)
E2.1 10 CFR 503,J Imolementation
a. Insoection Scoos (37001)
The inspectors reviewed the licensee's program to change plant design, revise
procedures, and conduct tests and experiments without prior NRC approval as described
under 10 CFR 50.59. The licensee's program was reviewed for compliance with
regulatory requirements and implementation of the program was evaluated by reviewing
three 10 CFR 50.59 evaluations. The inspectors reviewed the following documents:
ANO-1 10 CFR 50.59 Summary Report for 1996, deted May 22,1997
-
ANO 210 CFR 50.59 Summary Report for 1996, dated June 19,1997
50.59 Evaluations:
- ANO-1, Procedure Change 95-7023, " Reactor Coolant Pump Starting
Temperature"
,
- ANO-1, Condition Report 1-96-0667, " Reactor Coolant Pump Lift Pump Disabled"
- - ANO-2, Design Change Package 94-2007," Replace Charging Pump .
Downstream Check Valves CVC-22A, B, and C"
, Procedure 1000,131, "10 CFR 50.59 Review Program," Revision 3
l
b. Observations and Findings
-Within the review of the listed documents, the inspectors did not identify any
!- programmatic deficiencies in the licensee's 10 CFR 50.59 program. It was noted that
the licensee chose to evaluate the " margin of safety" of proposed changes by evaluating
the changes only against the basis section of the Techn! cal Specifications. This practice
L had been previously questioned by the NRC in Inspection Report 50-313:368/97-04 and
!
,
. - , , , - -
.
.
~5-
found to be in compliance with 10 CFR 50.59 (however, a violation for a failure to
perform a safety evaluation was issued). No deficiencies in the " margin of safety"
evaluations were identified by the inspectors during review of the 10 CFR 50.59
evaluations. Although currently in compliance, the licensee indicated that a more
encompassing design basis for margin of safety determinations will be implemented in
the future.
While review of the?e safety evaluations indicated appropriate programmatic controls,
some apparent failures to implement the 10 CFR 50.59 program were identified as
discussed in Section E8.22 of this report.
The inspectors also identified problems with the licensee's search capability for the
U) dated Final Safety Analysis Report. For example, Safety Evaluation 96-141 stated,
"The lube oil pumps are not describad in the Technical Specifications, Safety Analysis
Report, or operating license." The inspectors determined that this was in error. The lube
oil pumps were discussed in Section 4.2.2.6 of the Updated Final Safety Analysis Report.
The inspectors identified similar problems with the safety evaluatici determinations for
the procedure changes. The inspectors discussed these findings with the licens6e and
concluded that the automated search methodologies used to search for key words in
licerwag basis documents were not always effective in finding relevant reference
matenal and that the techt,ical reviewers were not ensuring that the scope of review was
adequate. The licensee agreed with these conclusions and informed the inspectors that
they had also identified this issue in the past year and had revised the safety evaluation
procedure to address the deficiencies. The inspectors review of Procedure 1000,131
indicated that improvements had been made in the Lensee's process to ensure that
adequate automated searches were performed and that they were reviewed by a second
individual.
c. Conclusions
Within the review of selected documents, the licensee's 10 CFR 50.59 safety evaluation
program, when implemented, appeared to be functioning satisfactorily, with the
exception of Updated Final Safety Analysis Report key word searches. However, some
apparent failures to implement the program were identified as discussed in
Section E8.22 of this report.
EB Miscellaneous Engineering issues (92903)
E8.1 (Ocen) Unresolved item 50-313/9623-01: Consideration of Multiole Hot Short Actuations
Backaround
This issue addressed the unplanned energizaiion of motor-operated valves (MOVs)
during a control room fire. This event, caused by a control cable shorting to an
energized conductor, is known as a hot short. Under certain configurations, a hot-
shorted MOV could stroke with the torque or limit switch removed from the control circuit,
causing the valve to stallinto its open or closed seat. Depending on the capacity of the
.
_ _ - . - _ _ _ _ _ _ . - - _ _
- - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -
'
.
.
3
motor actuator and the strength of the valve, this action could damage the valve
sufficiently such that plant operators would not be able to reposition the va!ve using the
valve's manual handwheel. The licensee's emergency operating procedures addressing
a control room fire and subsequent evacuation of the control r00m assumed that muual
control of accessible valves was available The licensee interpreted existing regvations
tu require postulation of only one hot short during a control room fire, whereas the NRC
maintained that multiple hot shorts, which is defined as more than one MOV having a hot
short et the same time, were within the scope of the regulations. Independent from the
resolution of this matter, at the request of the NRC, the licensee agreed to perform a
calculation to determine which valves would be potentially susceptible to hot short
damage. This was the anly aspect of this issue reviewed during the current inspection.
Insoector Followoo
The inspectors rev.ewed Engineering Report 97-R-0004-01, "Engineenng Information
Notice 92-18 Evaluation," Revision 1 PC-1; Calcalation 96-E-0065-01, " Unit 1 Safety
Related Valve Survivability Evaluation Under Stall Thrust / Torque, Motor Operated
Valves," Revision 1; and Calculation 96-E-0066-01, " Unit 2 Safety Related Valve
Survivability Evaluation Under Stall Thrust / Torque, Motor Operated Valves," Revision 0.
The inspectors reviewed Calculation 96-E-0065-01. The purpose of this calculation was
to determine operability with reduced margin for a select group of safe shutdown MOVs
under stall thrust / torque conditions. The evaluations were intended to reflect one-time
operation under the stated conditions and were not intended for use as design input in
other analyses or calculations. An important assumption stated that the anelvsis was
valid only for the time it was performed since actual test data was used in soine
evaluations. The results showed that 37 of the 66 valves evaluated would be able to be
manually positioned to their safe position following postulated hot short conditions.
Twenty-nino valves were susceptible to damage. The inspectors determined that the
licensee had used a valid methodology for determining the maximum torque and thrust
that an MOV would experience during a hot short event, with one exception For MOVs
that did not have previous measurements of stem friction, the licensee used a plant
averaoe stem factor correlated to a stem friction coefficient vf 0.1342. The use of a
bounding stem friction coefficient for the untested valve population, approximately 0.08,
would have resulted in calculated torques and thrusts as much as 50 percent higher than
those reported in the calculation. However, based on a qualitative judgement by the
inspectors, application of the lower stem factors would not have significantiv changed the
number of valves determined as potentially incapable of being manually repositioned.
Engineering Report 97-R-0004-01 addressed multiple hot shorts, but in situations where
two or more valves were determined to be unable to be manually repositioned, the report
stated that it wat not credible that both or all would fail at the same time in the same
direction. The inspectors noted that this reasoning reiterated the licensee's position that
multiple hot shorts were not required within the regulations.
Based on the results of the calculations and engineering report listed above and in light
of the 29 valves in Unit 1 analyzed to be susceptible to hot shorts, the inspectors
- - - - - - _ . - - - . - - . _ --- - - . - - .
':
- .
,
-7-
concluded that come safety functions could be lost if mu:tiple hot shorts occurred in a
particular manner. This was due to the fact that some of the motor operated valves may
be subjected to loads during a hot short that would render them incapable of being
manually stroked to their safe shutdown position. In some cases, as in the Unit 1 high
- pressure injection system, a total of eight valves would have to hot short in the same
direction to defeat the safety function of the system. However, the Unit 1 shutdown
cooling function could be lost with the hot short failure of only two vu.ves, CW1428 and -
CV 1429, decay heat cooler outlet valves. Both valves were catulated to be potent! ally
incapable of being manually repositioned following a hot short event.
- The above discussion was illustrative of the potential impact of multiple hot short events
and was not a complete listing of all potentially affected systems. This item will remain
open pending further reviews by the NRC program office to define the regulatory
requirements for this matter.
E8.2 (Closed) Violation 50-313/96027-03: Failure to Notify the NRC Within 1 Hour of
Declaration of Notification of Unusual Event
L Backaround
On October 17,1996, the licensee declared a Notification of Unusual Event in
accordance with its Emergency Plan for n fire in the Unit i reactor building that lasted
! more than 10 minutes. The licensee notified the NRC Operations Center of a fire in the
reactor building within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of the declaration of the Notification of Unusual Event.
'
Howover, the licensee did not inform the Operations Center that it had made a
declaration of one of the Emergency Classes (Notice of Unusual Event) specified in its
Emergency Plan in accordance with 10 CFR 50.72(a)(1)(l).
Insoector Followuo
The licensee determined that the cause of the violation was human error. The shift -
L engineer who made the initial contact with the NRC Operations Center provided details
l of the fire but neglected to inform the NRC that a Notification of Unusual Event had been
'
declared.
The licensee conducted training with notification communicators about this event and
NRC notification requirements. Additionally, the licensee reviewed applicable
procedures and revised Procedure 1903.011, " Emergency Response / Notifications,"
Revision 22, to require verbatim transmission of iaformation included on
Form 1903.011Y, " Initial Notification Message," Revision 22, to the NRC Operations
Center officer followed by transmitting the form to the NRC Operations Center vi-
l. facsimile. The form explicitly identified the emergency classification for the event. The
! inspectors verified the licensee's corrective actions had been completed and considered
these actions adequate to prevent recurrence of this violation.
l
l
l
_ , - - .. - . . -_- - - - . .- . - - - - - . _- . - -- -
.
.
- 8
E8.3 (Closed) Licensee Event Report 50 313/96-009f Fire in the Reactor Buildina Durina
,
Heatuo Resulted from a Cracked Weld in an Oil Line on a Reactor Coolant Pumo Motor
i
'
Backaround
During a Unit i plant heatup on October 17,1996, a fire occurred in insulation around the
m ain feedwater nozzle ring on the Train B once through steam generator. A cracked
wekiin the discharge line of the Train B reactor coolant pump lube oil lift pump resulted
in oil spray being introduced into the insulation. The hot piping caused the oil to ignite. A
Notification of Unusual Event was declared, and fire brigade responders extinguished the
fire approximately 16 minutes after discovery The NRC conducted a specialinspection
to review the circumstances of the fire, continued safety of the plant, and corrective
actions taken by the licensee. The results of the inspection are documented in NRC
Inspection Report 50-313; -368/96-27. The !!censee initiated Condition Report CR-1-36-
0567, conducted an event evaluation, and documemed its evaluation results in a root
cause analysis report.
The licensee's event evaluation concluded that oil intrusion into the insulation caused a
wicking effect that lowered the auto-ignition point of the oil to a lower than expected
temperature.
.
The NRC conducted a special inspection documented by NRC Inspection
'
Report 50-313; -368/96-27, which identified problems with the licensee's implementation
of 10 CFR Part 50, Appendix R, fire protection requirements and that the licensee had
failed to identify and correct problems that were noted prior to the fire. These were the
subject of an NRC Escalated Enforcement Action (EA 96-512). The licensee developed
and implemented many corrective actions to correct the deficiencies in these areas.-
They are discussed in Sections E8.4 and E8.5 of this report.
Insoector Followuo
This item is closed based upon the followup and discussion in Sections E8.4 and E6.5 of
this report, which address the escalated enforcement items associated with this event.
- E8.4 (Closed) Enforcement Action 50-313:-368/96512-01013- Inadeauate Lube Oil Collection
Systems for Reactor Coolant Pumos
,
Backaround
During the NRC special inspection following the October 17,1996, Unit 1 reactor building
fire (NRC Inspection Report 50-313; -368/96-27), t1e NRC reviewed the tube oil
collection systems for all of the reactor coolant purnps in both Units 1 and 2. The
4
y n m m.e t ~-
_____-- _ - - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _ - - _ _
l .
.
.g.
inspectors determined that the lube oil collection systems for the UnP.1, Train B reactor
coolant pump and all of the Unit 2 reactor molant pumps had not been designed to
collect oilleakage from all pressurized and unpressurized locations. Also, the Unit 1
Trains A C, and D reactor coolant pumps had inadequate leakage collection from
some unpressurized locations. The deficiencies were determined to be in violation of
10 CFR Part 50, Appendix R, Section Ill, requirements.
Immediately following the fire, the licc nsee placed the unit in cold shutdewn to evaluate
the effects of the fire and water suppression on plant equipment. No significant damage
occurred other than to some insulation. The damaged insulation was repaired or
replaced as necessary. Other actions taken prior to restarting the unit included installing
a new shroud around the Train B reactor coolant pump motor and extending oil collection
drip trays for all of the reactor coolant pump motor:, to provide coveragv. af potential
leakuge points. Due to remaining concems regarding the integrity of othar portions of
the Unit 1, Train B motor lube oil piping, the licensee implemented administrative controls
to prohibit operation of the associated high pressure lift oil pumps unless a fire watch
was present. The motor manufacturer provided confirmation that the absence of tube oil
injection following an inadvertent pump trip would not adversely affect the coastdown
characteristics of the pump. This was used as supporting information in a 10 CFR 50.59
safety evaluation to conclude that operation without the lube oil!!ft pump in standby did
not constitute an unreviewed safety question.
In Unit 2, the licensee performed tube oil collection sysiem modifications following an
unplanned reactor shutdown in November 1996. The NRC performed walkdowns of both
'
units and reviewed the modifications. These activities were documented in NRC
Inspection Reports 50-313; -368/96-27, 50-313; -368/96-07, and 50-313; -368/96-08.
The NRC concluded that with the exception of the remote lube oil ti:: aystems for both
units, the tube oil collection system modifications es'.ablished compliance with 10 CFR
Part 50, Appendix R, requirements.
The Unit i remote oil fill lines (Tygon tubing) to the top of each motor were not provided
with a leakage collection system as required by Appendix R requirements. The licensee
provided administrative controls to prevent operation of the system until a modification
and any necessary regulatory exemption was applied for and obtained. The Unit 2
remote oil system was designed differently, using stainless steel tubing and stainless
steel flexible hose, but no oil collection system was provided. The licensee described the
Unit 2 system in detail and proposed compensatory administrative controls for its use in
an exemption request to the NRC dated December 23,1996. The NRC approved the
exemption request in a letter dated June 14,1937. The exemption allowed the licensee
to use the remote oil fill system without a collection system contingent upon the
licensee's implementation of several compensatory measures prior to each use.
The licensee responded to the Notice of Violation in a letter to the NRC dated May 9,
1997. Corrective actions were identified in this document, as well as the licensee event
report associated with this event, and sevaral condition reports. Some of these were
completed and reviewed previously as discussed above; others were reviewed during
this inspection.
!
l .
~
l
.
- 10-
Insoector Followuo 0
The inspectors reviewed documents, procedures, an'.8 interviewed personnel to
determine the adequacy and completeness of the Sceasee's corrective actions. Also, the
inspectors contacted the c:gnizant fire protection engineer in the NRC Office of Nuclear
Reactor Regulation regarding the adequacy of the Unit 2 lube oil collection system.
Documents reviewed included: portions of the Updated Final Safety Analysis Report;
Condition Report 196-0507, its associated Root Cause Analysis Report and corrective
action documents (including 10 CFR 50.59 safety evaluations and screenings);
Procedure 1102.002, " Plant Startup," Revision 59; Procedure 1103.006, " Reactor
Coolant Pump Operation," Revision 20; Procedure 1107.001, " Electrical System
Operations," Revision 50; Procedure 1107.004, " Battery and 125V DC Distiibution,"
Revision 8; and Procedure 1504.001, " Visual Inspection of the Unit 1 & 2 RCP's Oil .
Colloction System," Revision 4.
{
Modifications were made to the tube oil collection systems for all reactor coolant
pumps in both units. These were determined by the NRC to be acceptable for meeting
10 CFR Part 50, Appendix R, requirements. To ensure the continued operability of the
systems, the licensee imp!*mented Procedure 1504.001, for inspection of the systems
immediately prior to refueling outages and prior to startup.
The licensee placed administrative controis on operation of the Unit 1 Train B reactor
coolant pump lift oil pumps. This war, done because of uncertainty about oil system
integrity during lift pump operation on the Jeumont Industrics motor Procedure
1103.006 was revised to require that the Train B reactor coolant pump not be started or
stopped during normal operations when the reactor coolant system temperature was
greater than 350 degrees F. Operation of the lift oil system was al! owed, but only if two
fire brigade-trained operators were present in the reactor building to identify oil leakage,
inform the control room, and extinguish any resulting fire.
'
The norrr.al design of the oillift system provided an autostart of the lift pump when the
reactor coolant pump tripped. The administrative controls discussed abova prevented an
autostart from occurring. Therefore, if a reactor coolant pump trip occurred, the pump
would coast down without injection of lubrication onto the motor thrust bearing. The
inspectors reviewed the licensee's 10 CFR 50.59 Safety Evaluation 96-141, " Evaluation
of Placing RCP #32B HP Lube Oil Pump (P-808 and P-638)in Pull-to-Lock," and safety
evaluation determinations for associated procedure changes. Supporting documentation
for the licensee's evaluation included a letter from the vendor documenting that reactor
coolant pump coastdown time would not be adversely affected for at least 15 seconds
without oil lift pump operation. The relevant accident analyses for consideration in the
safety evaluation were the various sizes of loss of coolant flow accidents, and the
evaluation determined that the lack of lift pump operation had no impact. The inspectors
concluded that the safety evaluation provided an acceptable basis for concluding that
reactor coolant pump operation without availability of tube oilinjection aid not constitute
an unreviewed safety question.
)
4
1
_ _ _ _ _ _ _ _
!.
l
.
-11
The licensee initiated an evaluation of the tube oil system piping configuration to ensure
that no additional questionable welds remained. The licensee and the motor supplier
determined that the lube oil piping sysiem was constructed in accordance with applicable
commercial standards with the use of qualified welders and adequate quality controls.
No other similar weld failures were reported in more than 200 other reactor coolant pump
motors manufactured by the motor suppiier. One difference in the subject configuration
was the addition of a backup direct current powered lube oillift pump and associated
piping. To evaluate this further, the licensee developed a plan to perform inservice visual
inspections of the piping at operating pressure, collect vibration data, and perform data
analysts. The licensee considered this option for addressing piping concerns preferable
to others considered, which included piping replacement, weld replacement, and
volumetric inspections with repair of rejectable welds. After discussing these options
with the licensee, the inspectors considered the licensee's approach to evaluating the
acceptability of the lube oil piping configuration to be adequate. Administrative controls
on lift pump operation would remain in effect during future operating cycles pending
completion of vibration analyses and continued visual inspections to identify any
remaining weld failures.
The inspectors determined that the licensee had completed corrective actions to
ensure that its reactor coolant pump oil collection systems were in compliance with
10 CFR Part 50, Appendix R requirements and that proposed corrective actions were
either completed or in progress with responsive completion dates.
E8.5 { Closed) Enforcement Actions 50-313/96512-01023 and 50-313/96512-01033: Failure to
ldentify Sianificant Conditions Adverse to Quality and Take Promot Corrective Action
Relatina to Oil Accumulation in Pioe Insulation.
Backcround
'
The NRC (and the licensee) determined that there were previous opportunities to identify
that a problem existed with the reactor coolant pump oil lift system pr'ar to the fire event.
These opportunities were missed, in part, because a condition report was not initiated to
thoroughly evaluate the extent of the potential problem. Some of the opportunities to
identify the scope of the problem included: discovery of a crack in the lift oil pump
<
discharge line, the magnitude of the crack and inadequacy of the original weld, discovery
of oil on the side of the steam generator during reactor building walkdowns during plant
heatup, discovery of oil puddled under the reactor coolant pump, and identification of an
excessive amount of smoke present in the reactor building during heatup.
To correct these deficiencies in problem identification, communications, and
understanding of oil-insulation interaction consequences, the licensee revised
procedures and conducted training.
10soector Followuo
The inspectors reviewed documents, procedures, and interviewed personnel to
determine the adequacy and completeness of the licensee's corrective actions.
- _-____
0
,
-12-
Documents reviewed included: Procedure 1000.104, ' Condition Reporting and j
Corrective Actions," Revision 13; Procedure 1015.036, ' Containment Building Closeout,'
'
Revision 5; Procedure 2102.002, ' Plant Heatup," Revision 10; Procedure 1102.002,
' Plant Startup,' Revision 61; Procedure 1102.010, 'P! ant Shutoown and Cooldown,"
Revision 41; Procedure 2102.010, " Plant Cooldown,' Revision 30; Procedure 1203.034, ,
' Smoke, Fire, or Explosion," Revision _10; Procedure 2203.034, " Fire or Explosion,"
Revision 5; and training outlines and attendance records for Condition Report 196-567.
The licensee's training appeared to adequately address the circumstances and causes
of the fire and provided information on cr,uses of insulation fires; uses and limitations of
material safety data sheets; importance of proper delivery, receipt, and accuracy of
communications; and lessons learned in response to oil spills and observations of
smoke.
The condition reporting procedure _was revised to clearly address identified deficiencies
in fire protection system components including oilleaks and spills. Items of this type that
meet the definitions in the procedure now require the initiation of a condition report. The
other procedures provided requirements to perform inspections for oilleakage and
excessive sn.oke and provided actions to take in response to identified adverse
conditions. These actions could include terminating plant heatup, posting fire watch
personnel with fire fighting equipment, contacting the fire protection engineer, and othar
appropriate act.ans. The procedures also contained guidance on the effects of oil
soaked insulation on the auto-ignition temperature of the oil.
The inspectors concluded that the licensee's corrective actions were adequate to
improve staff understanding of the mnsequences of oilleaks, and the importance of
communications when adverse conditions are identified, provide actions to prevent fires
from occurring, provide additional requirements for inspections, and provide actions to
take when adverse conditions are identified.
E8.6 (Closed) Insoection Followuo item 50-313/97201-01: Licensee's Actions to Revise the
Technical Soecification Bases fo,- the Minimum Water Volume in the Condensate
Storcae Tank
3ackaround
The minimum water level in the Unit 1 safety-related condensate storage tank was based
on the capability to provide a sufficient supply of water to permit the initiation of the decay
heat removal system within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> or less of an accident initiation. However, the actual
period of time required to initiate decay heat amoval may be substantially greater than 4
hours; The licensee acknowledged that this text was ir, accurate and proposed to revise
the existing Technical Specification bases to eliminate (probably) reference to the decay
heat removalinitiation and to state that the minimum condensate storage tank level was
based on the volume that was required for 30 minutes of emergency feedwater operation
before manual switchover of the pump suction to the service water system.
__
.
13-
Insoector Followuo
The licensee stated that they were in the process of neveloping the improved Standard
Technical Specifications and that the appropriate condensate storage tank bases would
be included in the initial submittal. The licensee stated that they planned to submit the
revised Technical Specification design base along with the revised Technical
Specifications by the end of 1998. The inspectors verified that there was no safety
concern related to the minimum condensate storage tank water level and that there was
no need on the part of tne licensee to revise this level. However, a clarification of the
Technical Specification basis for the level was appropriate.
E8.7 (Closed) Unresolved item 50-313/97201-02: Licensee's Failure to incoroorste Maximum
Emergency Feedwater Flow limits into Plant Procedures to Monitor and Preclude
Exceedina Recommended Limits
Backaround
A 1991 Babcock and Wilcox Company engineering report (92-R-1019-01) identified that
the maximum emergency feedwater flow rate to preclude damage to the Unit 1 steam
generator tubes was 1500 gpm. The NRC determined that the licensee had not revised
plant procedures to incorporate this upper flow limit. Specifically, Procedure 1010.002,
"hansient History / Transient Cycle Logging,"in effect at the time was not revised. The
NRC reviewed operator logs and identified two occurrences where the emergency
feedwater flow exceeded the 1500 gpm limit.
Insoector Followuo
The inspectors reviewed Condition Report CR-1-97-0081, dated March 27,1997. which
the licensee generated to resolve the maximum emergency feedwater flow issue. A
licensee contractor prepared an analysis which determined that the maximum feedwater
flow rate to prevent steam generator tube damage was 2214 gpm. The inspectors
reviewed the calculation and determined that the higher flow limit was adequately
justified.
The licensee stated that the upper flow limit was not included in operating procedures
because high emergency feedwater flows would cause a cooldown, during which
operators would take procedural actions to correct the high flow rate problem. However,
the inspectors determined that the licensee had not previously evaluated the effects of
the high flow on the steam generator tubes for the periods of time before operator actions
would restore an acceptable flow rate.
The inspectors reviewed Procedure 1010.010, " Unit One Transient Cycle Logging and
Reporting," Revision 1, which was issued to provide guidance for determining excessive
emergency feedwater flow and provide specifications for initiation of a condition report.
This procedure superseded Procedure 1010.002. The inspectors determined that the
Procedure 1010.010 required an engineering evaluation if the upper limit of the '
emergency feedwater flow was exceeded.
.
.. .
_ _ _ _ _ - _ _
- - _ _ _ - _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ ___
.
-14-
During the inspection, the inspectors reviewed in detail Framatome Technologies final
draft calculation 321229968-00 and Framatome Technologies Calculation 32-5000450-
00,"ANO Auxiliary Feedwater Capacity - Phase 2." The purpose of these calculations
was to determine the limit for emergency feedwater flow to preclude unacceptable steam
generator tube damage or wear. The results showed that a flow rate of 2214 gpm
resulted in less turbulence induced vibration than at full power conditions. As part of this
review, a conference call was held with Framatome Technologies and licensee
personnel during which they responded to questions from tM inspectors. Based on their
review and the information presented during the conference call, the inspectors
determined that the analysis provided in these calculations demonstrated that
emergency feedwater .peration at flow rates up to 2214 gpm for limited periods of time
would have no significant effect on steam generator performance. The inspectors
verified that design of the emergency feedwater system would not resun la flow rates in
excess of this amount.
The inspectors determined that the licensee failed to procedurally incorporate a vendor
recommended maximum emergency feedwater flow limit of 1500 gpm to a single steam
generator. As a result, flow rates in excess of this amount were permitted to occur for
short periods of time. Later, in response to the NRC's questions, these excess flow rates
were shown to be acceptable.
10 CFR Part 50, Appendix B, Criterion V, requires that activities affecting quality be
prescribed by and performed in accorJance with instructions, procedures, or drawings.
The licensee's failure to incorporate vendor recommended flow rate limits into
procedures was identified as a violation of 10 CFR Part 50, Appendix B, Criterion V
(50-313/9721-01),
Conclusions
The licensee failed to incorporate a vendor specified emergency feedwater flow limit into
plant procedures, resulting in the failure to evalus'e the effects of exceeding the vendor
recommended flow limit during two previous periw of plant operation. Concerns
related to possible damage to the steam generator tunes were adequately resolved by a
subsequent arialysis demonstrating the acceptability of higher flow limits.
E8.8 (Closed) Unresolved item 50-313/97201-03: EFW Pioina Confiauration aifferences
Backaround
in review of Calculation 82-D-2086-02, which evaluated net positive suction head
requirements for the motor-driven and turbine-driven emergency feedwater pumps, the
NRC identified that the suction piping configuration used in the calculation to determine
pressure drops was not consistent with the piping configuration identified in the piping
isometric drawings. The pressure drops in the piping were negligible and the effect on
the results of the calculation due to the differences was small. This item was opened
pending action to revise the calculation and take any appropriate generic actions.
_ _ _ _ _ _ - _ _ _ _ _ _ _ _ _
'<
.
I
..
15 -
Inspector Followup
The licensee initiated Action item Requests L97-0198 and L97-0285 to revise
Calculation 82-D-2086-02 to address the as-built configuration. The inspectors verified
that these action items had been assigned to the appropriate engineering organization
for completion. The licensee established a Uue date for completion of the calculation
revision of March 1,1998. Based on the minhal impact on the results of the calculation -
expected by incorporating the as-built piping configuration differences.in the calculation,
the inspectors considered the licensee's schedule for completion adequate. The broader
3
issue of identified deficiencies in the licensee's maintenance of calculations is addressed
in the inspectors' followup of Unresolved item 50 313/97201-06 (Section E8.11).-
10 CFR Part 50, Appendix B, Criterion Ill, " Design Control," states, in part, that design
control measurss, such as design reviews, shall provide for verifying or checking the
adequacy of design. The licensee's failure to properly verify that the design inputs for
Calculation 82 D-2086-02 were consistent with the installed piping configuration was
considered as the first example of a violation of 10 CFR Part 50, Criterion lli (50-
313/9721-02).
Conclusion
The licensee's failure to properly verify the design inputs to Calcu;ation 82-D-2086-02 ^i
- was identified as a violation of 10 CFR Part 50, Appendix B, Criterion Ill.
E8.g (Ocen) Unresolved item 50-313/97201-04f 1: *n=*a Pinino Pr===nre.ggi .
Temperature Specifications
,
Background
During review of Calculation 88-E-0100-16, which evaluated pressure and temperature
- limitations of emergency feedwater system piping, the inspectors identified several
discrepancies. For example, in the determination of maximum emergency feedwater
discharge pressure limitations, the calculation used a pump suction alignment from the
condensate storage tank instead of the service water system as input. The use of the
service water system as a source would result in a higher emergency feedwater system
discharge pressure than if suction were taken from the condensate storage tank.-
The licensee initiated Engineering Request 973848 for the design engineering staff to
review and revise Calculation 88-E-0100-16. The rrque'st was assioned a completion
date of December 31,1997.'
- _ _ _ _ - _ _ _. l
___
.
.
i
13
Insoector Followup
The inspectors reviewed the licensee's followup action item response to the original
inspection concern. The response identified that the design pressure of the emergency
feedwater system discharge piping was 1850 psig. At rated speed of the turbine-driven
emergency feedwater pump (3795 rpm) and with the suction aligned to the condensate
storage tank, the calculated discharge pressure at shuto# head was 1732 psig. When
the system was aligned to the service water system, the calculated discharge pressure at
shutoff head was 1842 psig, resulting in a margin from the design pressure of only 8 psi.
However, the ca!culation did not account for speed controller tolerances, nor did the
calculation assume a failure of the speed controller. If a failure of the speed controller
occurred, procedural guidance was provided for an operator to take manual control of the
turbine-driven emergency feedwater pump. This procedure instructed the oper; tor to not
exceed a pump speed of 3900 rpm to avoid overpressurization of the discharge piping.
The licensee determined that at a pump speed of 3900 rpm, the resulting maximum
discharge pressure was 1844 psig when suction was aligned to the condensate storage
tank and was 1932 psig when suction was aligned to the service water system.
Therefore, manual operation of the turbine-driven emergency feedwater pump at the
maximum procedurally-allowed speed would result in a maximum discharge pressure >
that exceeded the design pressure of the piping.
The licensee's response stated that the utilization of the 1932 psig maximum ressure as
a basis for discharge piping design pressure was not considered because manual pumr>
control while aligned to the service water source would require an equipment failure in an
already infrequent emergency plant condition.
The inspectors had a concern with this position. The condensate storage tank was
designed to provide a tomado-protected 30-minute supply of water to the emergency
feedwater system. The 30 minutes allowed the operations staff time to transfer the
emergency feedwater supply to the service water system. However, if the turbine-driven
speed control was out of tolerance or failed and the pump was operated manually per
procedure, the maximum discharge pressure could exceed the design pressure of the
turbine-driven pump discharge piping.
The licensee's response also addressed historical mp performance during testing at a
pump speed of 3920 rpm. Analysis of the test results in Calculation 80-D-10838-102A
concluded that the discharge pressure would not have exceeded the 1850 psig design
pressure, even if suction were aligned to the service water system at maximum pressure,
because the pump was not operating at full potential. However, the pump had sufficient
capacity to meet minimum specifications and, therefore, was operable.
The inspectors discussed these issues further with the licensee and were informed that
the original calculation would be reviewed to determine the extent that higher-than
nominal pump speeds need to be considered, both from a speed controller tolerance
perspective and a manual operation perspective. Additionally, since the pump appeared
to be under performing with respect to its certified pump curve, using the pump curve for
, _ _ _ _ ___
_ _ _
.
f
.
-17
piping design pressure determination may be overly conservative. The inspectors were
unable during this inspection to evaluate this issue further to determine whether the
pump was performing in an acceptable manner with respect to the pump curve.
However, the inspectors noted that the pump was meeting its Technical Specification
limits. The licensee'e due date for completion of this review and revision to the
calculation was December 31,1997. During the exit meeting conducted January 6,
1998, the licensee indicated that the due date for this effort had been extended to
January 15,1998.
The inspectors considered this item open pending further NRC review of the licensee's
ongoing evaluation and review of the performance of the turbine-driven emergency
feedwater pump.
E8.10 (C10MLUDIggived item 50-313/97201-05: Evaluation of EFW Pumo Room
Environmerd
Backaround
The licensee identified an environmental qualification problem in the Unit 1 EFW pump
room, which houses the turbine-driven and motor-driven EFW pumps, Steam traps
located in the steam supply lines to the turoine-driven pump, which were not designed for
seismic loads, were vented to the atmosphere of the room. A failure of one these traps
(two in the high pressure piping and two in the low pressure piping) would cause a
continuous b!owdown of hot steam in the room, raising the ambient temperature and
humidity, and possibly challenging the operability of environmentally-sensttive equipment
in the room. The licensee took interim measures to leave one steam trap isolated and to
permanently open a door to the room. The licensee then calculated the impact of a
steam trap failure under this configuration assuming that operators could isolat; the
steam blowdown within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />. The calculation (87-E-0026-09, Revision 0) co%ded
that temperature in the room wouid peak at 149.5 degrees F for a brief time, but that the
c steady state temperature would be 134 degrees F. Since the limiting long-term
temperature for the room was 148 degrees F, the licensee concluded that the EFW
system remained operable. The inspectors agreed with the licensee that humidity
resulting from the event would not adversely affect equipment critical to the function of
the EFW system.
This item was opened to track licensee efforts to make modifications to resolve this issue
permanently and to make necessary revisions to related design basis documents. The
licensee also indicated that an assessment of past operability of the EFW room would be
performed.
--
.
! *
-18-
Insoector Followuo
in review of this finding, the licensee determined that the original design of the EFW
steam traps was inadequate for both Units 1 and 2. The safety implications for Unit 2
were mitigated because the steam traps exhausted outside the EFW room. The licensee
decided to modify EFW steam traps on both units: Unit 1 during Refueling Outage 1R14
(Spring 1998) and Unit 2 during Refueling Outage 2R13 (first quarter 1999).
The licensee performed a past operabmty assessment and concluded that the EFW
system was operable in the past considering the room door closed, all traps in operation,
and a 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> operator response to isolate the failed trap.
,
The inspectors reviewed Root Cause Aralysis Report CR C-97-0048, " Unqualified
Steam Trap Lines in the EFW Turbines' Steam Supply Lines." In this document, the
licensee determined that the architect / engineer (A/E) for the design of the plant assumed
that failure of the steam trap lines would not reduce the functional capability of the EFW
systems. However, the A/E had evidently focused on sufficiency of the resulting supply
of steam to the turbine and not on the environmental consequences of the failure.
The inspectors questioned the assumption that operator action to isolate steam would
occur within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of a steam trap failure, an assumption essential to the past operability
status of the system. The licensee stated that an operator surveillance of the room
(checking lube oil flows) occurs by procedure soon after each automatic start of the EFW
system, which would presumably result in identification of the steam trap failure.
However, since no time parameter was specified for this surveillance, it could be '
indefinitely delayed if other complications were to occur. The inspectors asked about the
effect of a response if the failure occurred after this surveillance effort. In this case, since
no remote indication of temperature in the room is available, it did not seem reasonable
to assume that steam isolation within one hour would occur. The licensee stated that the
fire detection system would probably alarm given the amount of steam that would exist in
the room. On one previous occasion, in 1992 (CR 1-92 0361), the deluge valve tripped
open due to an excess amount of steam from the steam trap drain lines. However, on
further discussion with the fire control engineer, the inspectors learned that the fire
detection device in the EFW room is not sensitive to steam and would only possibly
alarm in the event of a condensation buildup on the device. The inspectors determined
e
that the fire detectors were not a reliable indication of a steam leak.
r
The evaluation cf past operability indicated that, with the room door closed, the room
temperature would reach 164 degrees F 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> after the steam trap failed, it seemed
questionable whether an ooerator could enter an anwonment this harsh to pen ~orm the
required valve isolation. In response to the inspectors' concern, the licensee generated
a new action item (Item Number 9) for CR C-97-0048 to address the issue of accessibility
as it pertained to the past operability determination. The licensee recalculated the room
temperature profile using more realistic assumptions, including condensing heat transfer,
and found that the room condition at 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> would be 133 degrees F at a relative
humidity of 57 percent. This corresponded to a wet bulb globe temperature (WBGT) of
121.5 degrees F. At this WBGT, the licensee, based on US Navy
i
1
~ . _ _ _ _ - _ _ _ _ - _ _ _ _ .
_ _ _ _ _ _ _ _ - _ _ _ _ _ _
,
.
19-
Document OPNAVINST 5100.19C, Changa 1, dated May 15,1990, determined that
strenuous work at a maximum duration of 30 minutes could occur in this environment.
The amount of time needed to isolate a failed steam trap would be only several minutes.
The inspectors noted that the shapa of the temperature profile suggested that the
temperature would peak out at something less than 140 degrees F at a relative humidity
of less than 60 percent. This suggested that the room would be fit for the necessary
operntor's actions to take olace and that equipment operability would not be threatened.
Dunng the inspection, the inspectors reviewed Calculation 87 E-0026-09, EFW Pump
Ruom Tempaniture Profiles." This calculation provided an analysis of the heat up of the
EFW puu : u sum!ng various failed op9n steam traps in addition to the normal thermal
load proviued by the EFW pumps and other equipment. The analysis was carried out
with an ANO developad and maintained software package,"RHUR," which calculates
roora heat up from art >ltrary sources. RHUR was p ogrammed in a ' macro' language
executed in a spreadsheet program, SYMPHONY. It was documented as a software
package and was controlled and maintained under the software control program. The
RHUR model devebped for thlS analysis was constructed by substantially modifying the
" macro" statements that make up the RHUR software package.
'
The inspector reviewed the analysis provided in the calculation and determ!ned that the
results of the four cases presented seemed reasonable and conservative for the given
configurations, in response to the NRC concern, the licensee provided additional
supporting analysis in the form of an alternate calculation using the GOTHIC software
program that substantially verified the results presented in Calculation 87-E 0026-09. (
The inspectors considered that the licensee had adequately established thJ past
operability of the EFW room equipment. Based on the less severe conditions of having
the room door removed, the insoectors also considered this in9rmation to support the
licensee'e assessment of current operability However, the inspectors identified a
weakness in the licensee's original response to this issue in that it frJted to document
consideration of the ability of an operator to perform an assumed operation in a harsh
environment. A tecond weakness was the lack of verification of the software model
used to calculate the postulated EFW room thermal transient. Although considered good
enginee% practice, licensee procedt,res did not specifically address this validation.
The licensee was in the process of developing a modification for both units to mitigate
the effects of pressure boundary failures and the blowdown of live steam from steam
traps in the EFW rooms, while preserving adequate condensato removal. Through
discussions, the inspectors learned that the modification w jd reroute steam trap piping
to exhaust outside the EFW room and upgrade the re a be safety related and
seismically qualified. The inspectors determinM 'N modification would resolvo the
hardware concerns associated with this issue.
4
.__ _ _ _ _ _ _ - _ _ _ _ _
0
.
20-
E8.11 (Cloand) Unresolved item 50-313/9720106: Desion Control Weakness
Bsckaround
This issue involved several calculation deficiencies. The examples included calculations
containing plant configuration discrepancies as well as calcult.tions that required
enhancement, updating, or to be superseded. The specific examples were as follows:
- Calculation 88 E-0086 01,"lE Bulletin 88 04 Review for P7A and P7B Minimum
Flow Evaluation," did not adequately document the effect of two parallel EFW
pumps operating on minimum recirculation.
- Calculation 92 E 0077-04, " Unit 1 EFW System Pump Performance
Requirements," used an incorrect level in tank T-418 for EFW actuation due to a
loss of offsite power following a tornado.
Calculation 92 E 0021-01," Emergency Nty Cycle and Battery Sizing
Calculation," was not revised to reflect plant modifications that resulted in
changes in the continuous and locked rotor current for EFW Valves CV 2620 and
CV 2627, and a modification that resulted in an additional load due to the reactor
building spray puinps close and trip circuits.
Voltage Condition," was not superseded as required.
The actions implemented or planned by the licensee included those to correct the
specific examples and to improve design control programs to prevent recurrence.
Insoector Followuo
The inspectors reviewed Licensee information Requests (LIRs) L97 0289 and 0290 as
well as Condition Reports CR C-97-0058 and CR-C 97-0059.
'
LIR L97 289 addressed the failure to updato Battery Loading Calculation 92 E-0021-01,
following iniplementation of Design Change Packages (DCPs) 92-1003 and 93-008. The
licensees's response to the inspection report (letter 1CAN099703, dated September 22,
1997), indicated that this calculation would be superseded before January 15,1998.
F sever, at the time of this inspection, this calculation had already been updated and
(as reviewed by the inspectors and found acceptable, in discussions regarding this
inconsistency, the licensee explained that the oversight resulted from the existence of a
large backlog of dats that had not been entered into the Design Configuration
Information Management System (DCIMS). This condition resulted in a failure to identify
affected calculations when engineers accessed the DCIMS system to perform searches
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ - _ - _ - _ _ _ _ - _ _ _ _ _ __
.
.
21-
based on key words related to a particular modification or calculation revision. The
licensee indicated that this backlog had been eliminated and that current practice would
preclude new backlog formation. The inspectors confirmed that no backlog existed and
that the DCIMS was up to-date at the time of this inspection. In additia, the inspectors
reviewed Memorandum ANO 97 00225 issued to the engineering staff that reinforced the
existing procedures and guidance related to calculation updates for load changes.
The second issue within LIR L97 289 dealt with Calculation 80-D 1083A 02, *EFIC DC
Valve Torque Calculation Under Reduced Voltage Condition" which should have been
identified as superseded. The licensees's response to the inspection report indicated
that this calculation will be superseded before January 15,1998. Discussions with
licensee personnel and review of the applicable procedure, OP 5010-015, " Engineering
Calculations," indicated that clear direction existed to ensure that superseded
calculations are identified, The cause of the specific instance was attributed to a backlog
of data to be entered in the primary tracking tool, DCIMS, that had previously existed. As
discussed above, the inspectors confirmed that no backlog existed at this time of this
inspection.
LIR L9/ 290 addressed Calculations 88 E-0086-01 and 92 E-0077-04, both of which
needed enhancement for various reasons. The licensees's response to the inspection
ecport indicated that these calculations would be revised before March 1,1998.
10 CFR Part 50, Criterion lil, * Design Control," states that measures shall be established
to ensure that applicable requirements and the design basis are correctly translated into
specifications, drawings, procedures, and instructions.
The licensee's failure to properly account for design information in the
Calculations 88 E-0086 01,92 E 0077-04,92 E-002101, and 80-D-1083A-02 was
considered as additional examples of a violation of 10 CFR Part 50, Appendix B,
Criterion 111 (50-313/9721.~1).
The inspectors review of Condition Report CR C 97-0058 indicated that the condition
report dealt with prioritization and completion of the Upper Level Documents (ULD) and
ULD-related discrepancies. The actions taken as a result of the condition report included
- determining completion priority and establishing and communicating to the engineering
' organization the philosophy for use of the ULDs. The inspector reviewed the actions
taken on these issues and confirmed that they were appropriate.
The inspectors review of Condition Report CR-C 97 0059 !ndicated that this condition
report dealt with completion of the Design Configuration Documentation (DCD) Project
I and resolution of discrepancies identified during this project. The project was completed
l in December 1994; however, approximately 100 discrepancies remained unresolved.
The discrepancies were originally screened for significance and operability, but this CR
recognized that recently there was greater interest in and higher standards being applied
to design configuration issues, in addition, several other recent CRs have been initiated
to reexamine previously closed discrepancies that were found to have inadequate
resolution by current expectations. The CR-C 97-0059 evaluation established an action
!
_ _ _____,
.
22
to priontize review of the remaining open discrepancies. This action was completed. In
addition, previously closed discrepancies have been reviewed to ensure that they are
adequate. This completed action led to further actions to evaluate 44 previously closed l
items. These subsequent actions were not available for review at the time of this )
inspection. '
In addition to the specific observations and related corrective actions reviewed above,
the inspectors reviewed other condition reports, procedures, guidelines, and
programmatic evaluations that impact the control of calculations. These are listed as
Attachment 2 to this report. Several recent calculations and design change packages
were reviewed to assess the results of ongoing improvement efforts on current
calculation practices. These actions addressed broader issues associated with
calculation processes and control.
The licensee initially identified a concem regarding the adequacy of their calculation
processes in Condition Report CR C 96-0060 initiated in March 1996. The most
significant outgrowth of efforts to resolve this concern led to the formation of a Quality
Action Team (QAT) to study calculation process issues. The team's final report,
published in September 199d, identified a large number of concerns (56) organized into ,
the following broad categories: l
- Calculation in processing time was too long creating backlogs and delays for
engineering personnel.
- The DCIMS data base contained inaccurate, incomplete, not up-to-date or
erroneous information.
- The calculation preparation, revision, and approval process requirements were
not clearly and consistently defined, documented, or understood by engineering
personnel.
t
- Retrieval of existing calculations was sometimes difficult.
! + Inconsistent interpretation of standards for legibility of calculation documents
l resulted in delays for processing calculations through quality assurance records
l storage.
l
- Data may be buried within a calculation with its relationship to other processes
not defined,
- Program engineers were not alwa ys aware when other engineers were revising
,
l calculations that impact their program.
- -
In reviewing the QAT action plan, the intpectors determined that the proposed corrective
actions were appropriate and that they r hould provide significant calculation process
improvements when fully implemented in addition, the interim compensatory actions
. .- - - - _ - . - . - - . - - - . - , - - . .
. _ _ -
_ _ . _ . . _ _ _ _ _ _ _
.
.
23-
that were implemented in the short term appeared to be of sufficient quality to reduce the
occurrence of calculation deficiencies.
In addition to reviewing the licensee's action plan, the inspectors reviewed a sample of
approximately 30 calculations generally chosen at random. These are listed in
Attachment 2. The inspectors found that the calculations were thorough, generally clear,
and well documented. Assumptions and other data sources were clearly identified e7d
referenced. The few exceptions were found in some older calculations performed during
original plant design and construction, approximately 20 years ago.
The inspectors focused on the licensee's process for assigning status to calculations.
This included the use of new, pending, as installed, amending, and superseded status
levels. Although the inspectors found the use of the amending status somewhat
confusing and cumbersome, no erroneous cross references were found in the sample
reviewed. For example, there are m limits to the number of calculations that can amend
a given calculation. Nor is there a time limit beyond which amending calculations would
require incorporation. In practice, however, the only calculations found with a large
number of amending calculations were the electricalloading calculations. In this case,
the licensee's practice was to assign a coordinator or owner for these calculations. The
coordinator was responsible for incorporating all amending calculations once each cycle.
The incorporated calculations were then assigned a status of superseded. The
inspectors reviewed and found this to be correctly performed for a recently issued
revision to Calculation 92 E 0021-01. " Emergency Duty Cycle and Battery Sizing
Calculations." In addition, as mentioned above, the licensee instituted a Calculation
Revision Notice (CRN) process that allows minor changes to calculations to be issued
and more easily tracked. This process was designed to reduce the number of amending
calculations to no more than nine in practice, based on a sample review, the inspectors
found no examples with more than two CRNs.
In spite of an apparently comprehensive effort to improve the calculation process, the
inspectors recognized a potential inconsistency in the way the licensee handled the
issue. The licensee's quality assurance team identified a concern related to the
preexistent backt0g in the DCIMS data base that had apparently been the cause of
several NRC-identified discrepancies. In addition, other persistent problems within the
calculation process were identified by the team. However, the licensee did not perform a
sample review of existing calculations to determine whether the DCIMS problem, in
addition to the other identified problems listed above, had created a situation where a
large number of calculations currently in effect may contain technical errors as a result of
these problems. The inspectors considered the lack of review of a sample of
calculations to represent a weakness in the licensee's overall response to the issue.
The inspectors reviewed the capability and use of the DCIMS database. It was a very
useful and powerful tool for searching for needed information concerning calculations,
evaluations, and reports.
_ _
,. . _ _ ._ .__
,
L
.
24
Conclusips
A violation of 10 CFR Part 50, Appendix B, Criteilon lil, was identified conceming
discrepancies identified with four licensee calculations. The lack of a review of existing
calculations to determine if past problems identified by the licensee affected other
calculations of record was identified as a weakness.
E8.12 (CJosed) Insoection Followeo item 50-382/9720107: Licensee's Actions for Revising
the Modification Procedure to Address Field Routed Installations
Background
The NRC identified that the layout configuration of certain installed conduits was different
from that shown on the conduit and cable tray layout drawings. The NRC also identified
that there were no drawings or records of the seismic support details for the installed
conduit and cable tray supports. The licensee stated that this was due to conduits being
field routed.
The NRC noted that conduits attached to the wall of the emergency feedwater pump
room contained redundant cables supported on the same unistrut support. The licensee
stated that the supports were designed to withstand seismic design loads and were
passive. In addition, the licensee stated that redundant trains could be supported from a
common support from a seismic standpoint provided Appendix R and other criteria were
addressed.
The licensee's modification procedure discussed a constructability walkdown involving
the modification engineer, modification supervisor, craft supervisor, and quality engineer,
During discussions, the licensee stated that they would document the decisions arrived
at during the walkdown by including these in the form of work steps in the controlled work
package. The licensee stated that this would include conduit routing and seismic support
details that would be documented as sign-off steps. The need to revise the modification
procedure to address field routed installations was identified as Inspection Followup
Item 50 313/9720107.
Insoector Followuo
The inspectors reviewed the licensee's response to the inspection followup item
documented in Letter 1CAN099703, dated September 22,1997 (response letter to NRC
Inspection Report 50-313/97-201). The inspectors also interviewed licensee personnel.
The inspectors requested that the licensee supply details concerning how the Appendix
R requirements were met for the redundant cables supported on the same unistrut in the
emergency feedwater pump room.
The inspectors rewewed draft Calculation 85-E-0086-1, " Safe Shutdown Capability
Assessment," Revision 3, and found that the turbine driven emergency feedwater pump
cable and the motor driven emergency feedwater pump cable had a 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> fire barrier
installed. In addition, the inspectors found that an NRC exemption was granted for
.- - . - - .
.
.
25-
configurations having less than 20 feet of separation. The inspectors reviewed licensee
Letters 1CAN068706, dated June 24,1987, and 1CAN048708, dated April 22,1987,
which requested the exemption for the cables. The exemption, granted in an NRC letter
dated October 26,1988, contained a stipulation that a 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> fire wrap was necessary in
addition to a fire suppression system. During a walkdown, the inspectors determine)
that there was a fire suppression system in the pump room and that the cable was
wrapped with fire wrap.
The inspectors reviewed the revised Procedure 6030.005, " Control of Modification
Work," Revision 3, Change PC-1. The inspectors noted that the procedure was revised
to include instructions for the development of a single line drawing depleting the types of
supports used, the support spacing, and the conduit layout. The procedure required that
the draw!ag be included with the controlled work package.
E8.13 Ogied) Insoection Followuo item 50 313/97201-08: Licentee's Actions to Perform a
Revised Cable Pullina Calculation Usino Additional Desian Information Obtained from
the Cable Vendor
flitCMIQMDd
The NRC identified that Procedure / Work Plan 6030.112, " Installation of Raceway
Systems," used for the installation of raceways, included a table of conduit bend radil for
manufacturers' standard bends but not for long radius bends. The NRC asked the
licensee which cables required long radius bends as a result of the minimum bending
radius and maximum sidewall pressure restrictions during pulling. The licensee provided
a calculation that compared the cable minimum bending radil to the bend radil of the
conduit that was appropriate for a single cable. The NRC questioned the values for the
redil on the inside surface of the conduit bends as well as the accuracy of the cable
diameters and maximum cable bend radii. The licensee stated that they would obtain
updated cable vendor data to verify the conduit bend dimensions and would revise the
calculation accordingly. The NRC reviewed the preliminary calculation and noted that a
few largo diameter cables would require long radius bends. This item was opened
pending licensee actions to obtain additional design information from the cable vendor in
order to perform a revised cable pulling calculation,
insoector Followuo
The inspectors reviewed the licensee's response to the inspection followup item in
Letter 1CAN099703, dated September 22,1997, and interviewed licensee personnel.
The inspectors also requested a listing of condition reports involving damage to cables
that resulted from cable pulling; however, no such conditions were found.
- -
. - . - - - . - . - .- .- - - - - . - _ -.- --. - . - - - - - -
.
!
'
- .
26
The licensee stated that updated cable information would be obtained from cable '
vendors for approximately 10 percent of the cable codes listed in the cable minimum
bend radius review table. The licensee expected to receive this information by mid.
January,1998. The licensee also stated that the updated cable information obtained
would be used to revise the calculation for each of the applicable cable codes. The e
licens*a stated that the cable calculations would be revised by March 1998.
The inspectors determined that the licensee's corrective actions were adequate for the
current closure of this item based on the fact that the condition report search had not _
found any cables damaged by cable pulling and that the actions necessary to revise the
cable code calculations were being tracked and had acceptable completion dates.
E8.14 (Closed) Insoection Followun item 50-313/97201 11: Verification of Calculation Changes
Associated with DC Batteries
Backaround
The NRC identified that Calculation 92 E 0021-04," Battery DOS and 007 Recharge
Time," Revision 0, was not revised to show the reduced charging time resulting from a
change out of the Unit 1200 amp battery chargers with new 400 amp chargers. An
amending calculation,93 D 1010-04, " Recharge Time for D06 and D07 with Changes
per DCP 931010," Revision 0, calculated the revised charging time. This item was
opened pending incorporation of the amending calculation information into the parent
calculation,
insoector Followuo
The inspectors determined that this condition did not represent an actual prob'sm
because the battery recharge times were not required by license conditions and did not
affect battery capacity. In addition, as discussed in Section E8.11 of this report, thers
were no licensee procedures requiring incorporation of amending calculations into the
parent calculation within a specific time period. Based on discussions with the licensee,
the inspectors determined that information provided in amending calculations, but
missing from the parent calculation, should not result in a quality concern because
annotations provided in the parent calculation positively identified the existence of the
amending calculations.
The inspectors verified that amending Calculation 93 D 1010-04 had been incorporated
into Calculation 92 E 002104. The inspectors observed that the revised recharge times
were based on the updated 400 amp recharge capacity.
I '
d
m_ _ - - _ _ . _ . . , ..,,_,_--m =--. - . , _ , _ , - ---,--r-- -- - m. -
-__. .__ . .. . ._ _
.
.
27
E8.15 (Closed) Insoection Followuo item 50 313/97201 12: Review of Current Desion BM!s for
Bvoassina Emergency Diesel Generator (EDG) Protective Trio
Background
For emergency starts of the Unit 1 EDGs, the licensee did not bypass those protective
functions delineated by Regulatory Guide 1.9, " Selection, Design, and Qualification of
Diesel-Generator Units Used as Standby Onsite Electrical Power Systems at Nuclear
Power Plants," to be bypassed under accident conditions. This was because the
licensee was committed to Safety Guide 9, " Selection of Diesel Generator Set Capacity
for Standby Power Supplies," which does not specify protective functions to be
bypassed. The concern was that the licensee's setup could result in a higher probability
of EDG loss during accident conditions. This item was left open pending review by the
NRC program office,
insoector Fougsp
Based on discussions with the NRC program office, the inspectors determined that the
licensee's adherence to Regulatory Guide 1.9 was satisfactory.
E8.16 (Ooen) U.Q[esolved item 50-313/97201 14: Lack of Testina Unit 1 Molded Case Circuit
Breakers
Background
The Updated Final Safety Analysis Report identified in Section 8.3 that the Class 1E
electrical distribution system was designed to meet the requirements of IEEE Standard
3081971, *lEEE Standard Criteria for Class 1E Electric Systems for Nuclear Power
Generating Stations ' Section 6.3 of this standard required periodic testing at scheduled
intervals to demonstrate that components that were not exercised during normal
operation of the station were operable. The standard further stated,'The specific tests
and the frequency at which they are performed depend upon the specific components
installed, their function, their environment, and the fact that they are in the maintenance
program of the unit (s). Illustrative examples of tests are given in Table 2." Table 2 of the
standard identified switchgear and one of the tests identified for illustrative purposes for
that type of component was an overcurrent trip test.
In contrast to these requirements, the licensee did not have a program to perform
periodic overcurrent trip testing of its Unit 1, Class 1E, molded-case circuit breakers.
The licensee developed a position paper that concluded that IEEE 308-1971 was not
applicable to molded-case circuit breakers. This conclusion was based on an
interpretation that Table 2 of the standard identified four tests for 600 V and below
switchgear: operation test, mechanical inspection, overhaul, and overcurrent trip test.
_ __ _
_ - . - - .
- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ______________ _ ___
i
. :
.
28-
Due to the physical construction of molded caso circuit breakers, mechanicalinspection
and circuit breaker overhaul were not possible. Therefore, the licensee concluded that
the perindic tests in Table 2 were not applicoble to molded-case circuit breakers but were
rather intended for load center switchgear.
The inspectors considered the licensee's interpretation of Table 2 of IEEE 3081971 to
be in error, based on the following points:
- Table 2 identified illustrative periodic tests for Class 1E components.
.
The identified tests were not an exhaustive or restrictive lists nor was Table 2 i
intended to exclude components for which it was not physically possible to l
perform all of the identified tests.
)
- The user of the standard was expected to evaluate which of the identified tests
(or others) were appropriate for each component.
As a result of its operating experience assessments for NRC's Information Notices 92 51,
- Misapplication and inadequate Testing of Molded-Case Circuit Breakers" and 93-64,
- Periodic Testing and Preventive Maintenance of Molded-Case Circuit Breakers," the
licent,ee initiated a practice of performing electrical testing of molded case circuit
breakers in the receipt inspection process and performing mechanical exercising of the
breakers on a periodic basis. However, they did not implement the practice of periodic ;
electrical testing. Information Notice 93-64 referred to Electrical Power Research
Institute Report NP 7410," Breaker Maintenance." Volume 3 of this document," Molded
Case Circuit Breakers," recommended electrical testing of molded case circuit breakers
on a 4 to 6 year periodicity.
Prior to the inspection documented in NRC Inspection Report 50-313/97 201, the
licensee performed a self assessment that identified that periodic electrical testing of
molded-case circuit breakers was not performed. The licensee established a task group
to determine if a testing program, replacement program, or a combination of both should
be adopted.
Insoector Followuo
The inspectors conducted interviews with personnel and reviewed the licensee's position
paper on molded-case circuit breaker testing and plans for developing a periodic testing
program for molded-case circuit breakers.
The licensee identified that they performed several activities related to safety related
molded-case circuit breakers to venfy that they will function under design basis
conditions and that the breakers are maintained accordhg to vendor requirements.
,
- F
>
2g.
These activities included: testing Unit 2 electrical penetration breakers in accordance
with the Technical Specifications, tet, ting all new breakers prior to installation, preventive
maintenance inspection and cleaning of breakers and panels, manual cycling of some
breakers, and performing periodic thermography inspections of 480 V molded case
circuit breakers. The licensee stated that these testing activities have historically
resulted in very low failure rates.
The licensee informed the inspectors that they had undertaken a project to develop a
periodic electrical testing program for molded case circuit breakers. Specifically, the
licensee stated that a program would be developed to test the Unit 1 containment
electrical penetration breakers and the program would be implemented in the refueling
outage scheduled for the spring of 1998. At the time of the inspection, the licensee had
identified a sample population of 64 molded-case circuit breakers for inclusion in the
program, of which, approximately 12 wcre scheduled for testing in the next refueling
outage. Additionally, the licensee stated that it would categorize the remaining
safety related Unit i breakers by risk significance and develop and begin implementation
of a testing program by the end of 1998. With respect to Unit 2, the licensee identified
that safety related molded-case circuit breakers that were not included in the Technical-
Specification-required testing program would be evaluated for risk significance and a
testing program would be developed and implemented by the Unit 2 refueling outage
scheduled in 1999.
Based upon the previous record of satisfactory Technical Specification-required electrical
testing experience of Unit 2 molded-case circuit breakers and other maintenance and
testing activities the licensee has implemented, the inspectors did not consider the
licensee's lack of a periodic testing program for Unit 1 molded-case circuit breakers an
immediate safety concern. However, the licensee's practices will be reviewed further
with the NRC's program office to determine if the licensee's reference to IEEE 308-1971
in its Updated Final Safety Analysis Report imposed a regulatory requirement to
implement an electrical testing program for molded-case circuit breakers. This
unresolved item was left open pending review of this issue by the NRC program office.
E8.17 (Closed) Unresolved item 50-313/97202-15: Lack of includino all Safetv-Related Fuses
in Proce&dC
BacharWad
The NRC identified that the licensee had not adequately resolved previous concerns
re;ated to incorrect electrical fuse installations. The licensee had established a like for-
like replacement policy for installed fuses but had physically verified only 4 percent of the
fuses in Unit 1 to be correct in all aspects. The fuse control procedure required an
engineering evaluation only for blown fuses or when replacement fuses were different
from those installed. All other 3pplications involved a like for like replacement without an
engineering review. Under these conditions, an existing incorrect fuse would be
replaced with a like fuse. The NRC considered the like-for like fuse replacement policy
to be inappropriate considering the small percentage of fuses validated.
_ _
- . _ _ _. _ _ _ _ _._ _ _ _ _ _ _ _ _ . ._ _ _ _ _ _ _ ._ _
.
- ,
- 30-
Insoector Followup
In response to the NRC finding, the licensee proposed to add the following DC
distribution equipment to the fuse control program by June 30,1998: D01, D02, D15,
D16, and D17 for Unit i and 2001,2002,2D26,2D27,2D41,2D42,2D51, and 2D52 for
Unit 2. Also, as a long term corrective action, the licensee stated that whenever a design
change was r ade to a safety related circuit protected by a fuse, a review of tha fuse list
would be performed by engineering to verify, correct, or add fuses to the procedure as
- necessary.
>
The licensee reviewed their condition report database for fuse-related discrepancies
occurring since January 1,1993. Seven condition reports during this period addressed
wrong size or type of fuse. Only one of these condition reports affected Unit 1, and this
, fuse was installed in a nonsafety related circuit. One was in a nonsafety circuit common
to both units and five affected Unit 2. Of these five, only three affected safety related
circuits. None affected circuit operability. Based on these facts, the licensee concluded
that the fuse control program was being effectively managed. The inspectors determined
that it was unlikely that a pervasive fuse control problem existed and, therefore, did not
consider the matter to represent a safety concern.
E8.18 (Closed) Unresolved item 50-313/97201-16: FSAR Disneoancies
Backaround
The NRC identified five discrepancies in the license's FSAR, mostly involving
, Table 7-11 A, "R.G.1.97 Post Accident Monitoring Variables." The discrepancies ~
involved inconsistencies between the table and design documents concerning instrument
ranges and display configurations. One additional discrepancy concerned a note in
FSAR Section 14.2.2.1.3.1 stating that both steam generators would be isolated in the
, event of a steamline break; whereas, in actuality, only one steam generator would be
isolated. i
Insoector Followun
! Prior to the identification of these discrepancies the licensee was in the midst of a
comprehensive FSAR review for both units. This program was described to the NRC in
Letter CNRO 97/00010 "EOl Licensing Basis Assessment and UFSAR Review
Initiatives," dated May 7,1997. The licensee stated that this effort would resolve the
, generic concerns iaised by this item since this review effort would identify the same
types of errors discovered by the NRC. The specific errors were to be reviewed and
_
corrected using the ticensing document change process. These corrections were to be ,
made effective in the 1998 submittal of the ANO Unit 1 FSAR.
The inspectors reviewed Letter CNRO-97/00010 and noted that it referred to a generic
review of the Unit 1 and Unit 2 FSAR for accuracy. The inspectors reviewed a document
entitled, " Safety Analysis Report Upgrade Project," that explained the scope of this effort,
currently scheduled for completion in July 1998. Based on review of this material and
,
,, ,-,s.v ., ,_r.- --
..,y, ,-- -y , . w, - - - ,, -, -- - - , - __. , , ,,--- ---,-
.
.
-31-
discussions with licensee personnel, the inspectors concluded that the project as
pirnned would resolve the generic concerns related to this issue. At the time of this
inswa ine licensee was approximately 50 percent complete with this project and had
identified 895 potential discrepancies. Of these,391 vea strictly editorial, leaving 504
items having potential significance. However, the licensee stated that many items may
subsequently be deterrnined not to be FSAR errors since the reviewers have been
instructed to docuinent everything that might be a discrepancy.
The inspectors reviewed licensing Information Requests L97-0213, L97 0326, and
L97 0310 and confirmed that the specific errors identified by the NRC were scheduled to
be corrected in the next FSAR update.
The inspectors determined that the licensee's efforts were sufficient to correct the
specific and generic concerns associated with this inspection finding.
10 CFR 50.71(e) states that the updated FSAR shall be revised to include the effects of
all changes made to the facility or procedures as described in the FSAR. The licensee's
failure to properly update the FSAR regarding the five discrepancies described above
was considered a violation of 10 CFR 50.71(e). However, because of the favorable
progress being made by the licensee's FSAR review team and the fact that a description
of this project has been placed on the docket, and the NRC's view that the discrepancies
identified by the NRC likely woule. have been identified by the licensee's program, the
NRC is exercising discretion in accordance with Section Vll.B.3 of the Enforcement
Policy and is not taking formal enforcement action for this finding.
E8.19 (Closed) Unresolvnd item 50-313/97201-17: Ineffective Resolution of Previous
Corrective Actions
Backaround
During a plant walkdown, the NRC noted that two in-parallel check valves located in the
steam supply piping to the turbine-driven emergency feedwater pump (P7A) were
continually chattering because of steam leakage through downstream solenoid valves.
The NRC's concern was that wear from this mechanical action could eventually result in
failure of the check valves to perform their reverse seating function,
insoector FollQEuD
The Wspectors were informed that leakage through the solenoid valves had been a
problem at ANO since at least 1988 when CR 1-88 0015 identified this problem. The
licensee had performed maint3 nance on the valves on several occasions, but the
leakage problem persisted. At some point, the licensee determined that the small
amount of leakage was not detrimental to the system, since the check valves were not
experiencing abnormal wear and the heating of the downstream piping was considered
beneficial in reducing the potential for condensate-induced turbine overspeed events.
1 Then, the licensee deferred any plan for modification. However, in October 1996,
the licensee identified accelerated wearing of elastomeric surfaces on one of the
.
.
32-
two solenoid valves in the EFW steam lines. As documented in Condition
Report CR 196 0569, electricians found the O-ring on the operator of Solenoid
Valve SV 2663 to be hard and brittle following its installation on July 14,1992. The
licensee's equipment qualification program stated that this 0 ring was qualified for
60 years of service but did not consider seat leakage to be a normal condition for the
establishment of qualified life. The licensee determined that the solenoid valve remained
operable with the damaged 0-ring. Based on the O ring failure, the qualified life of the
O rings was being revised to be from 2 to 8 years depending on the service condition.
While the licensee considered a modification to the solenoid valves, they were also
considering installation of motor-operated globe valves and relecating the valveb closer
to the turbine.
The check valves performed a safety related function by isolating the piping from a
faulted steam generator. The licensee stated that one of these valves would be
disassembled and inspected during the next refueling outage in the Spring of 1998. In
February 1995, the other check valve was inspected and found structurally sound with ,
slight indications of scoring. During past tests of the check valve's reverse seating
capability, the results have been acceptable.
The inspectors were informed that the check valves had chattered since their installation
in 1986, even before the solenoid valves began to leak, because of the presence of a
drain / strainer orifice that creates a constant steam demand of 1600 pounds-mass per
hour. Since the check valves require iOOO pounds-mass per hour to remain constantly
open, the solenoid valve leakage couid actually reduce the chattering if it were of
sufficient magnitude. However, the current chatter rate of the check valves was 10 taps
per second, indicating that the solenoid leakage was well less than 3400 pounds mass
per hour.
The inspectors determined that the licensee had adequately ensured that EFW system
operabililty had not been compromised by the solenoid leakage problem. The inspectors
recognized that the check valve chattering was partNily independent of the leakage
problem and that the leakage had been observed tc to beneficial for preventing
condensation in the EFW turbine steam supply line Gese facts mitigated any safety
concerns re' < ed to this condition.
E8.20 (Closed) Unresolved item 50-313/97201-18: Lack of Seismic Sucoort of OTSG Pressure
Transmitters
(Closed) Unresolved item 50-313/97201-19: Lack of Desian Basis for Sucoort of Steam
Generator Instrument Sensina Lines
Backaround
The NRC observed that a clip designed to restrain the isolation valve and sensing line for
Steam Generator Pressure Transmitter PT 2667B was missing. The concern was that
the line could fail during a seismic event, potentially affecting the operability of the steam
_ _ _ _ _ _ _ _ _ .
.
.
33-
generator pressure controlloop. The NRC also observed that the corresponding clips on
the three other redundant pressure instrument channels in the same vicinity were loose.
The licensee initiated Condition Report CR 197-0058 to determine the operability status
and potential generic concerns. Before the end of the inspection, the licensee reported
to the NRC that the as found configuration was operable and that a walkdown of other
instrument tubing restraints indicated that the mounting clip problem was not widespread.
In the Unit 1 EFW pump room, the NRC found a 10-foot unsupported span of instrument
tubing for a pressure tranomitter (PT/PI 2811). This is a safety related sensing line that
monitors the discharge pressure of the turbine driven emergency feedwater pump. A
m. mum unsupported span of 2.5 feet was generally specified in the seismic drawing
details. The licensee issued Condition Report CR 107-0074 and, subsequently,
determined that the as-found configuration was operable. The licensee determined that
a generic issue existed and elevated CR 197-0074 to a significant condition, requiring a
root cause determination. The licensee performed a detailed walkdown of the EFW,
Decay Heat / Low Pressure injection, and High Pressure injection systems and identified
the following conditions;
Several installations were undocumented or unanalyzed, such as tubing supports
exceeding the 30-inch unsupported span and use of nonstandard instrument
supports.
Lack of design bases or criteria for determining instrument line slopes.
Drawing discrepancies between drawings and the as built condition.
Missing or incorrect tags on instrument valves.
Based on these findings, the licensee expanded the scope to include other nystems. For
example, Condition Report CR-197 0087 was later opened to document insidequately
supported emergency diesel generator instrument lines. In all,37 instrument
installations were found in Unit 1 that did not conform to either seismic or instrument
installation requirements as defined in the standard drawing details. Design engineering
deiermined that none of the 37 discrepancies created an inoperable condition.
Insoector Followuo
The inspectors reviewed Root Cause Analysis Report CR 197-0074," Instrument Tubing
not Mounted in Accordance with Seismic Design Requirements," dated March 24,1997.
The licensee identified the following root causes for this event:
(1) Inadeouate_trainina: Personnel involved in the installation and maintenance of
instrument tubing were not adequately trained or sensitized to seismic and
configurations requirements.
_ _ _ _ - _ _ _ _
. _ . . . ._. - - -- .. . .. -
.
.
34-
(2) Work oractics Personnelinstalling and maintaining instrument tubing either did
not use required configuration documentation during work activities or did not
follow these documents correctly.
(3) Desian confiaura@n and analygls: Design documentation and prints were
inadequate or not easily accessible.
The licensee walked down 40 instrument installations in Unit 2 and found that the quality
of the installations was much better than in Unit 1. Only four of the installations
inspected had notable deficiencies involving damaged mounting clips or brackets
resulting from removal and reinstallation to support work on nearby equipment.
The licensee performed the following corrective actions: (1) the individual discrepancies
were repaired as necessary, (2) a letter was issued on April 4,1997, to site groups
reiterating instrument tubing requirements, and (3) design engineering reconfirmed the
initial operability assessments fiom the tubing walkdowns. Additionallong term actions
included the following: (1) develop a scope for additional walkdowns, (2) determine
whether existing seismic tubing mounting requirements can be relaxed, (3) evaluate
enhancements to design drawings associated with tubing, (4) evaluate the work planning
processes that involve instrument tubing or the removal of tubing for interference
purposes, and (5) provide training on seismic and configuration requirements for
instrument tubing.
The inspectors discussed the status of the proposed long term actions discussed above
with the licensee. A calculation was in the final stages of review that redefined the
acceptable support span lengths for various configurations. Once this calculation was
complete, the licensee was planning to perform additional walkdowns of those safety-
related components in Unit 1 that were potentially susceptible to instrument tubing
discrepancies. This effort was to be completed by May 1998. The licensee was also
planning to provide personnel training and to clarify drawings that contained confusing
tubing support information.
l The inspectors determined that the licensee's completed and proposed corrective
actions (each of which were assigned completion due dates) were sufficient to address
the issues raised by this item and considered the overall response to be strong,
particularly considering that none of the identified discrepancies had been evaluated to
affect operability.
10 CFR Part 50, Appendix B, Criterion 111, " Design Control," states that the design control
measures shall provide for verifying or checking the adequacy of the design.
The licensee's failure to provide measures to ensure the correct translation of seismic
design requirements to instrument tubing installations and the maintenance of this
configuration during plant operations in Unit 1 was identified as a third example of a
violation of 10 CFR Part 50, Appendix B, Criterion ill (50-313/9721-02).
!
1 -. _-. _ _
.
i
35-
Conclusions
The licensee's failure to provide measures to ensure the correct translation of seismic
design requirements to instrument tubing installations and the maintenance of this
configuration during plant operations in Unit 1 was identified as a violation.
E8.21 (Gosed) Unresolygd item 50-313/9720120: Inadeauate Work Plan for the Controlof
Post-maintenance Testina
Backomund
The NRC identified a misaligned limit switch on Valve CV-4804, rector building vent.
The misalignment caused the actuator cam to drag on the limit switch arm toward the
end of the valve stroke instead of riding on the roller bearing. Repeated operation in this
configuration could have resulted in a faulty valve position indication. Although this
particular application was not safety related, there existed a concern that other safety-
related applications may have a similar misalignment problem.
The licensee recognized that the post modification plan and job order for this valve
(00954697) did not include a post maintenance test to verify proper operation of the limit
switch cam assembly. The licensee reopened this job order and accompanying work
plan with the intent to add steps to reconfigure and test the assembly. One other valve in
Unit 1 (CV-1845, quench tank sample valve) had a similar mounting design, and a similar
revision was scheduled for its work plan. The licensee performed a review for other
similar limit switch configurations but did not identify any other of these configurations in
either Units 1 or 2.
Insoector Followup
The inspectors reviewed the work plan revisions discussed above and observed that
steps were added to verify that the limit switch rollers were in proper contact with the
cams. The licensee considered that this verification in the work plan coupled with the
existing post maintenance test that ver;fies smooth motion and proper valve position light
sequencing would be sufficient to ensure that the condition would not recur (i.e., no
revision to the post modification plan was considered necessary). The inspectors
observed that the post maintenance testing problem,in this instance, did not affect
safety related equipment and determined that the licensee had adequately addressed
the specific and generic implications of this finding.
E8.22 (Closed) Unresolved item 50-313/972121: Inadeauate 50.59 Review Associated v@
the Removal of the Unit 1 BWST Vacuum Breaker and Followuo Corrective Action
,
Backaround
This item hvolved an inadequate 10 CFR 50.59 safety evaluation performed by the
licensee for a temporary covering placed over the borated water storage tank vacuum
relief valve flange. The evaluation was inadequate, because it did not consider the air
_
'
.. .. . .. ..
.
. . .
______ ____________
.
.
36-
flow restricting effects of a mesh screen that was an integral part of the temporary
covering. The following is a summary of key events:
- On December 4,1996, while Unit 1 was at full power, the iicensee removed the
8-inch diameter vacuum relief valve from the borated water storage tank for a
surveillance test. When the valve was removed, the licensee installed a white
plastic bag over the 8 inch diameter open tank flange to act as a foreign material
exclusion (FME) cover. Two slits, each approximately 8 inches long, were cut in
the plastic cover to allow air flow. This work was done using Job Order
00952943. The job order stated that if suction for the tank was necessary, a
supplemental vent path was provided by a 4 inch tank overflow line that was
always open during plant operations.
- On December 5,1996, the system engineer observed that the tank flange was
covered with what he believed to be a solid FME cover and initiated Condition
Report 106 0663 to document the potentialinoperability of the borated water
storage tank. Subsequently, the system engineer determined that the tank was
operable since the FME cover contained two slits that would permit tank pressure
and vacuum relief. The plastic bag was installed for approximately 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />.
The licensee subsequently ter*ved the plastic bag and installed a blank flange
on the tank flange with a 1 inch gap between the flange and tank flange to permit
pressure and vacuum relief. Around the 1 inch opening, a mesh screen was
placed to prevent foreign material introduction into the tank.
- Condition Report 196-0663 also documented the discovery of a through wall
crack on the inlet flange of the removed vacuum relief valve and assessed the
operability of the tank. The operability assessment did not address tank
operability while the plastic bag was installed as an FME cover. The assessment
only addressed the flange and 1 inch gap and stated that the blank flange with
the 1 inch air gap installed over the tank flange would provide sufficient air
movement to vent the tank during rapid drain downs. At this time, any air flow
restriction from the installed mesh screen was not considered,
.
- On December 12,1996, the gap dimension was changed from 1 to 3-inches
under Job Order 00952943. This was done to increase the venting capacity
based on engineering judgement that the 1 inch gap may have been marginal.
- On January 22,1997, Significant Condition Report 197-0019 was initiated and
identified that a safety evaluation was not performed for se plastic bag installed
over the tank flange for approximately 30 hours3.472222e-4 days <br />0.00833 hours <br />4.960317e-5 weeks <br />1.1415e-5 months <br />. In addition, the condition report
identified that the procedure for testing the valve (Procedure 1306.034, " Testing
of Unit 1 Pressure Vacuum Relief Valve PSV 1617, PSV 2423, and PSV 1412,"
Revision 2) did not allow removal of the vacuum relief valve during power
._ _ __ __ _ __- _ _ _ _.. _ __ _ _ _ _ _ _ _ . _ . . _ . _ _ _ _
- -
.
.
37 -
. operations and that the safety evaluallon for the procedure did not include
l consideration for valve removal at powcr. The condition report included an _
>
operability assessment for the tank when the plastic t sg was installed on the tank
i flange. The licensee sta;ed that the tank was operable based on engineering
judgerrent that the plastic bag would break before an internal vacuum would
'
,
cause damage to the tank.
'
+ On February 3,1997, the licensee issued Condition Report CR 197-0031 to
address the previously made assumption that the 4-inch overflow line would, by
itself, provide sufficient vacuum relief. The conclusion of this effort was that,
< contrary to prior judgement, the 4 inch line was not large enough to prevent the
formation of an internal vacuum in the tank during rapid draindown. .
- On February 11,1997, the licensee prepared Temporary Alteration 971-001 that
documented the removal of the vacuum relief valve and the installation of the
i blank flange with a 3 inch gap over the tank flange. The temporary alteration was
prepared approximately 2 months after the actualinstallation. Temporary
alterations were not issued for the previous two temporary coverings (plastic bag
and 1 inch gap flange connection). Temporary Alteration 971001, which >
contained the first 10 CFR 50.59 evaluation performed in association with these .
events, concluded that the borated water storage tank was operable with the ;
3 inch gap blank flange installation. j
'
- On March 4,1997, in response to concerns from the NRC, Condition Report 1
,
96-0663 was amended to incluc's a discussion of the air flow blockage caused by
the mesh screen over the 1 inch gap. After considering the screen restriction, the -
licensee concluded that the borated water storage tank was still operable under
this configuration.
,
insoector FolloWMD
The inspectors reviewed the documents referenced above and had discussions with
licensee personnel, in review of the facts, the inspectors identified several concerns in
addition to those addressed in the previous inspection, as discussed below. <
Procedure issu_es
'
The inspectors found, in agreement with Significant Condition Report CR-197 0019, that
'
the licensee's use of Procedure 1306.034 during power operations was improper. This
procedure was originally written for use during refueling outage conditions as indicated ;
by paragraph 4.3.1 of the procedure, which stated that inspection of the valve would be
performed each refueling outage. The 10 CFR 50.59 evaluation for the procedure did
4
not address performing maintenance or testing of the vacuum relief valve while the plant
was at power. Also, the procedure did not address temporary provWons for vacuum
relief, which would be an essential element for online maintenance of the vacuum relief
valve. Although one of the corrective actions from the condition report was for the
,
,--v--e - - ,,,,, y ,-e , , , . ,-, ~ . ~ , - , ., , - ,,-r , - , , - , , - - , ,yw --e , ,e~ n.- -,--a, n- -rm.,.
.
.
38-
licensee to revise Procedure 1300.034 to allow valve removal and maintenance of the
VMye while at power, this revision had not been corr,pleted at the time of this inspection.
10 CFR Pr.rt 50, Appendix B, C6terion V, require', thst activities affecting quality.shall be
oresenbod by docum6ntsd procedures of a type appropriate to the circumstances and
shall be accomplished !,1 accordance with these procedures. The licensee's failure to
comply with the plant condition prerequisites of Procedure 1306.034 was identified as an
apparchtTiolent.,n (50 341iW21-03).
Desian Controlinna
The inspectors determined that the licensee had performed temporary changes to the
borated water storage tank without the use of proper design control measures. The
licensee removed the tank vacuum relief valve and sequentially installed three temporary
covering configurations, each of which constituted a change to a safety related
con'oonent, without use of the temporary alteration procedure (a temporary alteration
wat eventually prepared for t,1e third configuration, but this was issut j approximately 2
months after installation).
The inspectors reviewed Procedure 1000.028," Control of Temporary Alterations,"
Revision 20. Attachment i to this procedure listed the criteria for requiring a temporary
alteration, which included a situation where a design change was made to safety-related
equipment without the removal of the equipment from service. Using this criteria, the -
inspectors concluded that, because the borated water storage tank was not taken out of
service for these evolutions, a temporary alteration was required. The use of a
temporary alteration was intended to ensure that plant changes were subjected to design
control measures commensurate with the original design.
10 CFR Part 50, Appendix B, Criterion til, requires those design changes, including field
changes, shall be subject to design control measures commensurate with those applied
to the original design. The licensee's failure to provide design control measures
<
commensurate with initial design for three temporary changes made to the borated water
storage tank vacuum relief valve was identified as an apparent violation
(50-313/9721 04).
Operability issues
The plastic bag was attached to the tank flange in a manner such that the 8-inch
'
ventilation slits may not have been pulled into the opening during an event (e.g., a safety
injection event due to a large nreak loss of coolant accident that would cause a rapid
tank draindow). Therefore, the venting function of the line could have been defeated.
This would have left the 4-inch overflow line as the only venting source for the tank,
which, according to the licensee's calculations, was not adequately sized to prevent a
vacuum from forming in the tank during a rapid tank draindown. Consequently, the
inspectors questioned the operability of the borated water storage tank during the 30
-__-
.
.
t 39
hours that the plastic bag was attached to the vent flange. Because of a lack of
redundancy, an inoperable borated water storage tank would render the entire
emergency core cooling system inoperable. The licensee performed several tests and
calculations in an attempt to support the original engineering judgement that the tank
was operable in this configuration. Following the onsite inspection period; the licensee
provided the following evaluations and calculations for NRC review;
- *BWST (borated water storage tank) Past Operability," internal memorandum
from Don Phillips to Charles Tyrone, dated November 25,1997.
- Calculation 97 E-02100 01, "BWST Venting Capability Through 4 inch Overflow,"
Revision 0, December 3,1997.
- Calculation 97 E 021101, "BWST Level Analysis," Revision 1, December 4,
1997.
.
Calculation 97 E-0209-01," Evaluation of BWST Tank T 3 for Potential Vacuum
Load in Support of Operability Determination associated with CR-196-0663,"
Revision 0, December 8,1997.
The licensee concluded that borated water storage tank was operable during the period
of time the plastic bag was attached to the vacuum relief valve flanga. However, the
analyses clearly demonstrated that the borated water storage tank was in a degraded I
condition and could have plastically deformed above the water level in the tank in the
event of a large or medium break loss of coolant accident. This deformation, according
to the licensee's evaluation, would not have prevented the operation of the emergency
core cooling system as designed. Also, the effect of the internal vacuum on the tank
level instrumentation could have caused an early transfer of emergency core cooling
water system suction from the borated water storage tank to the reactor building sump.
This early transfer would have reduced the net positive suction head margins for the
emergency core cooling system pumps. The licensee determined that this would not
have rendered any of these pumps inoperable.
'
10_CFR 50 59 Issues
The inspectors determined that the licer,see had failed to evaluate whether temporary
changes made to the borated water storage tank had created an unreviewed safety
- question.
10 CFR 60.59, " Changes, Tests and Experiments," permits the licensee to make
changes to the facility and to procedures as described in the Safety Analysis Report
without prior Commission approval provided the change does not involve a change to the
Technical Specification and does not involve an unreviewed safety question. A
!
.
.
-4 0-
proposed change, test, or experiment shall be deemed to involve an unreviewed safety
question if a possibility for an accident or malfunction of a different type than any
evaluated previously in the safety analysis report may be created. The licensee shall
maintain records of changes in the facility, and these records must include a written
safety evaluation that provid6s the bases for the determination that the change does not
I involve an unreviewed safety question.
The inspectors identified the following two examples where the licensee made changes
to the borated water atorage tarik vacuum relief line without adequately eva!uating
whether an unreview6d safety question existed:
- The licensee reinoved the existing vacuum relief valve and covered the tank
flange with a plaatic bag taped to the tank. In the absence of a safety evaluation,
the licensee used engineering judgement to consider the tank operable under this
configuration. However, when questioned, the licensee had to perform extensive
testing and analysis beicte concluding that the original engineering assumption
was correct and that the tank was operable. The installation of the plastic bag
appeared to create the possibility for an accident or a malfunction of a different
type than any evaluated previously in the Safety Analysis Report and may have
created an unreviewed safety question.
- The licensee installed a blind flange separated by a 1 inch gap from the tank
flange and installed a foreign material exclusion screen over the gap. The
unreviewed safety question evaluation of this configuration failed to consider the
air flow restriction created by the foreign material exclusion screen.
The failure to properly determine whether an unreviewed safety question existed as
t
required by 10 CFR 50.59 was identified as an apparent violation (50-313/9721-05).
Conclusion
Three apparent violations were identified associated with temporary coverings placed
over the borated water storage tank vacuum relief valve flange, when this valve was
removed on December 4,1996, for testing and maintenance. The apparent violations
involved the failure to perform proper safety evaluations as described in 10 CFR 50.b9,
the failure to perform the vacuum relief valve testing under the plant conditions as
prescribed in the test procedure in accordance with the requirements of 10 CFR Part 50,
Appendix B, Criterion V, and the failure to utilize the correct design control processes
(temporary alterations) for the temporary cover configurations placed over the valve
flange in accordance with 10 CFR Part 50, Appendix B, Criterion Ill.
,. .. . . . .. .
.
..
.
-o
-_. . _. - . . _ - . ---- - - -
0
0
41-
E8.23 (Cloigd) Unresglyed item 50-313/9720122: Revised Calculations Associated with
BWST Vonexing and Pumo NPSH
Background
The licensee prepared an engineering report,93 R 1002-01, "ANO-1 BWST Outlet
Vortex Suppressor," dated February 5,1993, to examine the potential for vortexing in the
BWST resulting in a loss of net positive suction head (NPSH) for the emergency core
cooling pumps. The conclusion of the report was that the ECCS pumps would have
sufficient NPSH. Within this report, the licensee identified that it had failed to consider
the effect of instrument error in the vortexing calculation. This meant that the actual level
in the tank at the time of switchover from the BWST injection mode to the recirculation
mode may have been less than the nominal value, bringing the hydrodynamic system
closer to a state of cavitation and air entrainment than had been originally assumed. The
licensee issued Condition Report CR 197-0039 to investigate this condition. Within this
report, the licensee concluded that the ECCS system would have remained operable.
The licensee concluded that only the high pressure injection pumps would be susceptible
to an NPSH problem but only at low tank levels near the switchover point to the
recirculation mode. At this time in the accident, the high pressure injection (HPI) pumps
could be shut down because the low pressure injection pumps alone could provide
adequate core cooling. Within CR 197-0039, the licensee also was to complete
Revision 1 to Engineering Report 93 R 1002 01 and to determine whether other
calculations may have been affected by a similar failure to consider instrument error.
Insoector Followup
The inspectors reviewed the documents listed above and discussed the issue with
licensee engineers. The inspectors determin. d that this instrument error was only
reflected in the vortexing calculation and did not affect actualinstrument readings.
10 CFR Part 50, Criterion lil, " Design Control," states that the design control measures
, shall provide for verifying or checking the adequacy of design, such as by the
l performance of design reviews.
The licensee's failure to check properly the adequacy of Engineering
Report 93 R 1002-01 was identified as a fourth example of a violation of 10 CFR Part 50,
Appendix B, Criterion lli (50 313/972102).
Conclusioni
The licensee's failure to check the adequacy of an engineering report properly that failed
to account for instrument error in a calculation of vortexing in the borated water storage
tank was identified as a violation.
1
_ _ -.
.
.
l -4 2-
E8.24 [Gosed) Unresolved item 50 313/97202 23: InadtSuate 10 CFR 50.59 Evaluation
6110 dated With BWST Releastts
Background
This issue involved a problem witi 10 CFR 50.59 safety evaluation performed for
Temporary Alteration 97-1001, whi n is discussed in Section E8.23 of this report. In
assessing the potential post loss of coolant accident (LOCA) radioactive release through
the 3 inch gap foreign material exclusion (FME) covedag on the flange of the borated
water storage tank (BWST) vent, the licensee addressed only offsite exposure and did
not address control room exposure, which could ba considerably more significant. Also,
the safety evaluation did not refer to the licensee's response to information Notice 9156,
- Potential Radioactivity Leakage to Tank Vented to Atmosphere "
Insoector Fe20Wup
The licensee revised the 10 CFR 50.59 evaluation for Temporary Alteration 971-001 to
include a discussion of control room dose resulting from potential backleakage to the
BWST. In the evaluation, the licensee stated that no accident consequence analysis
takes credit for holdup or filtering of radioactive releases to the BWST, and, therefore,
back leakage to the BWST is beyond the licensing basis. Based on recent test results of
the BWST supply and return valves, the licensee stated (in the revised safety evaluation)
that little or no backleakage would be expected to occur post LOCA. Consequently, the
licensee determined that there would be no change to the estimated control room
dosage as a result of the temporary alteration.
The licensee reviewed its response to information Notice 9156 and determined that the
conclusions reached as a result of Temporary Alteration 97-1001 were not affected by
this iirmation notice.
The inspectors considered the lack of reference to control room dose in the original
10 CFR 50.59 evaluation to be an oversight, but observed that two mitigating factors
existed. First, the original evaluation did not quantify a release rate based on the
assumption that the release would be negligible. Had the release been quantifiable, an
assessment of control room dose would have been necessary. Also, the fact ' hat no
credit was taken in the licensee's accident analysis for containment of BWST radioactive
leakage relegated the issue to a beyond-licensing basis status, The inspectors
considered the licensee's response to this issue to be adequato.
.
.
.
-43
IV. PlanLSunpart
1
F2 Status of Fire Protection Facilities and Equipment
(
a. Insoection Scooe (92903)
During a walkdown of the Unit i emergency feedwater system pump room, which was
performed in support of the review of various items discussed in Section E, the
inspectors inspected the condition of the electrical cable fire barrier material,
b. Qhstryations and Findings
The inspectors identified a deficiency in the fire wrap for Conduit EJ2029, which provided
protection for the control circuitry for the turbine-driven emergency feedwater pump. On
a vertical run of the conduit, a collar joining two sections of fire wrap had loosened and
slid down. This resulted in about a 2 inch portion of the conduit being exposed.
The inspectors reported the condition to the licensee and showed the deficiency to a fire
protection engineer and an operator. The licensee agreed that the as-found condition
was unacceptable and immediately declared the fire barrier inoperable. In accordance
with the requirements of its fire protection program, the licensee verified the operability of
smoke detection equipment and control room alarm capability and established an hourly
fire watch. Condition Report 197-0313 was initiated to document the item and Job
Request 928623 was initiated to repair the fire wrap.
The inspectors asked the licensee whether there was a program for inspecting fire
barriers and whether the fire wrap in the emergency feedwater pump room was included
within the scope of the program. The licendee provided the inspectors with
Procedure 1307.062, * Unit 1,1 Hour Cable Fire Wrap Inspection," Revision 0. The
inspectors reviewed the procedure, discussed fire wrap inspection techniques and the
process of dispositioning inspection observations with licensee personnel, and verified
that inspection of the pump room was included within the scope of the procedure. The
inspection of the sub; set ' ire wrap was last performed with satisfactory results in July
1997. The procedure was required to be performed every 6 months. At the time of the
inspection, no explanation was available for the cause of the deficiency. The licensee
determined that a noncompliance did not exist, because the surveillance interval had not
been exceeded and the damage could reasonably be assumed to have occurred
following the most recent inspection.
The inspectors considered the licensee's actons on this matter to have been
appropriate,
I
O
e i
+44 * r
'
c. Conclusions
The inspectors identified a fire barrier deficiency and the licensee took prompt
compensatory actions as required.
V. ManSment Meetings
XI Exit Meeting Summary
The inspectors presented the inspection resu'ts to members of licensee management by
telephone on January 6,1998. -The licensee acknowledged the findings presented.
During this meeting, the licensee stated that it will expand its definition of margin of
safety as it pertains to the performance of safety evaluations performed under
10 CFR 50.59. This issue is further discussed in Section E2.1 of this report. i
The inspectors asked the licensee whether any materials examined during the inspection !
should be considered proprietary. The licensee sta..J that one document reviewed by
the inspectors contained proprietary information. The proprietary information from this
- document was not discussed within this inspection report.
,
4
1
. -.. , - , . . , . ._ ._ . . - ,m.._,
.
i
l
.
ATTACHMEliT 1
SUPPLEMENTAL INFORMATION
PARTIAL LIST OF PERSONS CONTACTED
Licemen
G. Ashley, Licensing SupeMsor
E. Blackard, Design Engineer
M. Cooper, Licensing Specialist
G. Dobbs, Electrical and instrumentation and Control Supervisor
J. Howell, Design Engineer
R. Hutchinson, Vice President, Arkanshs Nuclear One
R. Lane, Director, Design Engineering
D. Mims, Licensing Director
D. Phillips, Nuclear Steam Supply Syatem Supervisor
J. Richardson, Design Engineer
C. Tyrone, Director of Design Engineering
T. Waldo, Supervisor, Modifications
C. Zimmerman, Plant Manager, Unit i
Nf1C
J. Melfi, Resident inspector
INSPECTIQfiERQCIQUEES USED
37001 10 CFR 50.59 Safety Evaluations
92903 Followup of Engineering issues
ITEMS OPENED. CLOSED. AND DISCUSSED
Ooened
50-313/9721-01 VIO Failure to incorporate EFW Flow Limits into Procedures
(Section E8.7)
50-313/9721-02 VIO Design Calculation Deficiencies (Sections E8.8, E8.11,
E8.20, and E8.23)
50-313/9721-03 APV Failure to Perform BWST Vacuum Relief Valve Testing
Procedure Under Proper Plant Conditions (Section EB.22)
50-313/9721 04 APV Failure to implement Temporary Alterations for BWST
Vacuum Relief Valve Removal (Section E8.22)
l
'
_ _ _ _
.
.4
2-
50-313/9721 05 APV Failure to Perform 10 CFR 50.59 Evaluations for BWST
Vacuum Relief Valve Removal (Section E8.22)
Closed
50-313/9627-03 VIO Failure to Notify NRC Within 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> of Declaration of NUE
50 313/96-009 LER Fire in Reactor Building During Heatup Resulted from
Cracked Weld in Oil Line
50-313; 368/96512-01013 EA inadequate Lube Oil Collection Systems for Reactor
Coolant Pumps
50 313/96512-01023, and EA Failure to identify and Take Prompt Action to Lube Oil
01033 Leakage
50-313/97201 01 IFl Licensee's Actions to Revise the Technical Specification
Bases for the Minimum Water Vo!ume in the Condensate
Storage Tank
50 313/97201 02 URI EFW Flow Rates Exceeding B&W Flow Rates
50 313/97201-03 URI EFW Piping Configuration Differences
50 313/97201-05 URI Evaluation of EFWP Pump Room Environment
50 313/97201 06 URI Drawing and Calculation Revisions
50-313/97201-07 IFl Modification Work Procedure Revisions
50 313/97201 08 IFl Add;tional Vendor infonnation Necessary for Revised
Caole Pulling Calculation
50-313/97201 11 IFl Verification of Calculation Changes Associated With DC -
Batteries
50 313/97201 12 IFl Review of Current Design Basis for Bypassing EDG
Protective Trip
50-313/97201 15 URI Lack of including all Safety Related Fuses in Procedure-
50 313/97201 16 URI FSAR Discrepancies
50-313/97201 17 URI Ineffective Resolution of Previous Corrective Actions
50 313/97201-18 URI . Lack of Seismic Support of OTSG Pressure Transmitters
50-313/97201 19 URI Lack of Design Basis for Support of OTSG Instrument
Sensing Lines
50 313/97201 20 URI Inadequate Work Plan for the Control of Post-
Maintenance Testing
.50 313/97201 21 URI Inadequate 50.59 Review Associated with BWST Breaker
.
/
.
. . .
.
_ _ _ _ _ _ _ _ _ _
,
o
D
3-
50-313/97201 22 URI Revised Calculations Associated with Vortexing and
Pump NPSH
50 313/97201-23 URI Inadoquate 50.59 Evaluation Associated with BWST
Releases
Discussed
50-313/97201-04 URI Inadequate Piping Pressure and Temperature
Specifications
50-313; URI Consideration of Multiple Hot Short Actuations
-368/9623-01
1
50-313/97201 14 URI Lack of Testing of Unit 1 Molded Case Circuit Breakers
LIST OF ACRONYMS USED
A/E Architect / Engineer
BWST Borated Water Storage Tank
CR Condition Report
CRN Calculation Revision Notice
DC Direct Current
DCD Design Configuration Documentation
DCIMS Design Configuration Information Management System
EDG Emergency Diesel Generator
FME Fore;gn Material Exclusion
FSAR Final Sa'ety Analysis Report
MOV Motor-operated valve
NPSH Net Positive Suction Head
QAT Quality Assurance Team
SIMS Safety Information Management System
ULD Upper Level Document
WBGT Wet Bulb Globe Temperature
._ _ _ _ _ _ _ . -
.-__ - __ -__-___ -_-_ _ _ _ _ _ - _
e
o
AIIACHMENT 2
LIST OF DOCUMENTS REVIEWED FOR ITEM 50 313/97-201-06
CALCULATIONS
Lalculation A-86," Items Discussed with Daleas R.S. of CE MSLB, ECCS and LOCA,"
Revision O.
Calculation Quality Action Team Final Reput, September,1996.
Calculation 32-1229968-00, " Final Draft, ANO AFW Capacity - Phase 1".
Calculation 32 5000450-G0, "ANO AFW Capacity, Phase 2," Revision 0.
Calculation 80-D-1083A-02, "EFIC DC Valve Torque Calculation Under Reduced Voltage
Condition," Revision 1.
Calculation 80-D-2085-01, " Correlation of Teletector to Activity Release," Revision 0.
Calculation 80-D-2085-02, " Provide Correlation Between MR/hr Reading Detector Activity Steam
Line Result Table Method," Revision 0
Calculation 85-D-1042-03, "AMSAC Instr Loop Errors & Setpoints," Revish15.
Calculation 85-E-0053-20, " Combustible Loading Calc for Fire Area G," Revision 29.
'
Calculation 86-E-0002-01, "EDG 1 & 2 Load Study," Revision 8.
- Calculation 86-E-0020-01, "lE Station Battery 2D11 Duty Cycle'& Sizing Calc," Revision 7.
Calculation 87 E-0026-09, "ErM Pump Room Temperature Profile," Revision 0.
Calculation 88 E-0086-01, "lE Bulletin 88-04 Review for P7A and P78 Minimum Flow
Evaluation," Revision 0.
Calculation 89-E-0029-02, " Cal Factor Calc ANO-2 MS Line Rad Monitor," Revision 0.
Calculation 89-E-OiO2-01, " Terminal Voltage Calc for DC MOV," Revision 8.
Calculation 89-E-0144-01, "EDG Loading for 2A3 & 2A4 Bus," Revision 3.
Calculation 90-E-0116-01, "ANO-2 EOP Setpoint Document," Re.wision 4.
Calculation 91 E-0090-06, " Heat Load Determination for Post Accident for Rooms 95 98 99100
104109110 & 149 for Post Accident Cooling," Revision 2.
Calculation 92-E-0021-01, " Emergency Duty Cycle and Battery Sizing Calculations," Revision 4
.mu _ __ ___-___
_ _ _ _ _ _ _ _
!
,,
9
2-
Calculation 92-E-0077-04, " Unit 1 EFW System Pump Performance Requirements,' Revision 0,
- Calculation 93-D-1008-05, 'MOV Sizing Calc for CV-2620,* Revision O.
Calculation 93-R-1029-30,'ANO-1 SRR for Auxiliary Building HVAC System ABH V,'
Revision O.
Calculation 94-D-5033-01, * Removal of Local Fire Trouble Horns from D21 Breaker 34,"
Revision 0.
Calculation 94-E-006101," Tech Spec Allowable Outage Time (AOT) Utilizing Attemate AC ,
EDG," Revision O.
Calculation 95-D-1004-03, 'DCP 95-1004 Load Addition to RC1," Revision O.
Calculation 95-D-1011-01,"EDG Load Add for DCP 95-1011," Revision 1.
Calculation 95-D 7064-01, " Vital AC Panels & Instrument AC Panels," Revision 0.
Calculation 95-D-7072-01,"D01 Battery Loading Amendment," Revision O.
Calculation 96-E-0065-01, " Unit i Safety Survivability Eval Under Stall Thrust / Torque,"
Revision 1.
,
Calculation 96-R 0002-02, " Generic Letter 89-10 Program Valves Limit Switch Setpoint,"
Revision 1
j
CONDITION REPORTS
CR-C-96-0060, re Calculation Room Control and Discrepancies, March 22,1997.
- CR-C-97-0053, re Calculation 92-D-2001-49 Approval, February 11,1997.
CR-C-97-0058, re Upper Level Document Discrepancies, February 13,1997.
CR-C-97-0059, re Design Configuration Documentation Discrepancies, February 13,1997.
CR-1-96-0184, re Seismic Qualification Calculations, May 23,1996.
CR-1-97-0017, re Accuracy Analysis Calculations for EFW Flow Transmitters, January 21,1997.
CR 2-97-0304, re Discrepancy in Instrument Setpoint Calculations 2LT-5659-1 and 2LT-5668-2,
May 28,1997
..
.. . . . . .
- _--_ ___-_-__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -
%l
'
o
3-
DESIGN CHANGES
Design Change Package 94-1003, "Feedwater (MFW Venturi Replacement)," Revision 2
Design Change Package 95-2001, " Control Room Emergency Chiller Installation - Air
Conditioning Modification," Revision 1
ENGINEERING STANDARDS
'
Engineering Standard GES-39, " Calculation Numbering Schemes and Codes," Revision 2
LICENSING INFORMATION REQUESTS
LIR L97-0035, ' Action items from ANO 50.54(f) Design Bases Assessment: Maintenance of
Calculations [1.1,1.2),",
LIR L97-0037, " Action items from ANO 50.54(f) Design Bases Assessment: DCIMS
Improvements [1.6],",
LIR L97-0201,"NRC AE Design Basis inspection," July 31,1997
LIR L97 0202,'NRC AE Design Basis inspection," July 31,1997.
LIR L97-0289, " Response to IR 97-202 AE Design Basis inspection," September 28,1997
LIR L97-0290, " Response to IR 97-201 AE Design Basis Inspection," September 26,1997
PROCEDURES-
Procedure OP-5010-015," Engineering Calculations," Revision 1 PC1,
Procedure OP-5010-019, " Computer Software Control," Revision 1 PC1
MISCELLANEOUS
Memorandum ANO-97-00225, M. Stroud to ANO Engineering Personnel," Reminder of Design
Considerations for Possible Impact to Battery and EDG Lording Calculations," March 3,1997
Design Configuration Documentation Discrepancies 2M7-07-01,02,03,07,08,09,12,14.
Design Configuration Documentation Discrepancies RBS 02-05,06
d
_ --__-_-_ _ a