ML20212E553
| ML20212E553 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 12/17/1986 |
| From: | Ashe F, Caldwell C, Chan P, Eli M, Andrew Hon, Kellund G, Jim Melfi, Myers C, Pereira D, Richards S, Royack M, Zimmerman R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V) |
| To: | |
| Shared Package | |
| ML20212E526 | List: |
| References | |
| 50-361-86-25, 50-362-86-26, IEB-85-003, IEB-85-3, IEIN-86-034, IEIN-86-34, NUDOCS 8701050354 | |
| Download: ML20212E553 (39) | |
See also: IR 05000361/1986025
Text
_ _ _ _.
.
,
,
-
- 5 , ;-
_
'
' *
u .a.
'
,y. ID ' >
>3 + ,4 .
,
'
,
_
U. S. NUCLEAR REGULATORY COMMISSION
REGION V
Report Nos. 50-361/86-25, 50-362/86-26
Docket Nos.- 50-361 and 50-362 .
License Nos. NPF-10 and NPF-15
Licensee: Southern California Edison Company
P. O. Box 800
2244 Walnut Grove Avenue
Rosemead, Callfornia 91770
Facility Name: San Onofre Nuclear Generating Station Units 2 and 3
i
- Inspection at: ' San Onofre, San Clemente, California
. Inspection conducted: September 22 - October 3, 1986
Inspectors:-
Roy
b) 7hsf r
n Team Leader
w /f? ////Jo'
Date/ Signed
. b-* k W " / z //+ /96
S im Melfi( Fjtnetor Inspector Dat( Signed
Srch- 0roject
Dave Pereira,
- <-Inspector isIoIes
Date Signed
_',. f -
- '
pf}
Date Signed
~
Mik'e Ro' ck, Reactor Inspector
? [1,_ ) m -
/ 2 //6 N
Andrew Hon, Project Inspector Date Signed
St* M rw
Chris Myers, Resident Inspector
alolm
Date Signed
S& M rea-
Chris Caldwell, Project Inspector
111a k.
Date Signed
Sh V:OC- #2 ln f%
-Gary Kellund, Resident' Inspector
.
Date Signed
'
$ O W Y vo1 s 2 Ia Ist.
- Frank Ashe,, Electrical Systems Engineer, Date Signed
NRC Office _of Analysis and Evaluation of
Operational Data (AEOD)
8701050354 861217
PDR ADOCK 05000361
0 PDR
. _ _ - - _ _ _ _. -
- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _
.-
-2-
b')CA- Fort itl11 f 84
Paul Chan, Consultant Lawrence Livermore Date Signed
National Laboratory (LLNL)
I
bN k F-on._ ' stl0lm
Mark Eli, Consultant, (LLNL) Date Signed
Approved By: bD A- b 12.l n IEl.
Stuart Richards, Chief, Engineering Section Date Signed l
Summary:
Inspection on September 22 - October 3, 1986 (Report Nos. 50-361/86-25,
50-362/86-26)
Areas Inspected Annual, announced team inspection of the San Onofre Nuclear
Generating Station (SONGS), Units 2 and 3. The inspection focused on the
ability of the plant to safely respond to events that have actually occurred
at similar facilities. Specifically, the Rancho Seco loss of integrated
control system (ICS) power event (NUREG 1195) was reviewed in its entirety for
its applicability to SONGS. Certain findings from the Davis-Besse loss of
main and auxiliary feedwater event (NUREG 1154) and the June, 1986 Catabwa
' depressurization event were also reviewed. The licensee's review of these
events and their implementation of timely preventive corrective action was
evaluated.
<
The following areas were also reviewed:
1) Industry operational experience review program
2) General plant tour to assess equipment condition
- 3) Maintenance program (preventative and corrective)
- , 5) Motor operated valve maintenance program
j- 6)- Facility modifications (temporary & permanent)
! 7) Post trip review process
8) Calibration of pressurizer pressure & level transmitters
To the maximum extent feasible, the effectiveness of these activities were
assessed as they apply to the following plant systems. The systems were
chosen based on Frobabilistic Risk Assessment (PRA) studies, historic events
at other sites, and a review of problems at SONGS.
1) Auxiliary Feedwater System (AFWS)
-
2) Salt Water System (SWS)
'
3) Component Cooling Water (CCW)
4) Batteries and Inverters
5) Toxic Gas Isolation System (TGIS)
6) Instrument and control Air
7) Radiation Monitors (RM)
8) Shutdown Cooling System
_ - - _ _ _ _ _ _ _ _ _ _ _ _ _
i
'
,
!
_ ,,
.
-3-
This inspection was performed by eight NRC Region V inspectors, one NRC staff
engineer from AEOD, and two contractors. Inspection Procedures 30703, 37700,
38701, 40700, 41400, 42700, 61700, 61725, 62700, 62702, 62703, 62704, 62705,
72707, 72710, 72701, 90712, 92700, 92701, 92702, 92703 and 93702 were
applicable to this effort.
~
Results: Of the areas inspected, one violation of NRC requirements was
identified. (Failure to maintain fire proofing material in a cable raceway as
required by design - paragraph 3.B.)
.
d
. . . - _ . - - - .- _
. , . . - . - - .. . _ - .
&
- -=
s e
'
l
4
i
DETAILS
,
1. Persons Contacted
L
,
a. San Onofre Nuclear Generating Station (SONGS)
i
- K. Baskin, Vice President Nuclear. Engineering Safety and Licensing
- H. Ray, Vice President and Site Manager, Nuclear Generation Site
4
- M. Wharton, Deputy Station Manager
- R. Krieger, Operations Manager
- D. Shull, Maintenance Manager
- T. Mackey, Compliance Supervisor
- D. Schone, Manager, Site Quality Assurance
- D. Nunn, Manager of Nuclear Generation Services
'*H.'Horgan, Station Manager
- M. Wharton, Deputy Station Manager
- J. Curran, Nuclear Safety Manager
- A. Schramm, Supervisor Coordinator, Unit 1~
'
- P. Knapp, Health Physics Manager
- J. Patterson, Maintenance Engineering and Services Manager
-
i-
,
W. Lazear, Maintenance / Outage Quality Assurance Supervisor
- R. Joyce, Asst. Maintenance Manager, Units 2 & 3
- R. Rosenblum, Quality Assurance Manager
- W. Marsh, Unit 2's 3 Superintendent
,
- D. Pilner, Nuclear Engineering Manager-
. *J. Paterson, Manager, Maintenance Engineering and Services
-*R. Phelps, Supervisor, NSE
! *J.'Albers, Unit 2/3 Health Physics Supervisor
i *D. Peacor, Emergency Preparedness Manager
I' K. Johnson, NSSS Engineer-
- W. Zinti, Compliance Manager
- R.' Plappert, Compliance Supervisor
- C.1Gibson, Compliance Group Lead
- M. Metz,' Compliance-Engineer
~
- M. Freedman,-Compliance Engineer'~
'
- M. Zenker,' Compliance Engineer
i <
'
- Denotes those attending the final exit meeting on October 3, 1986.
! Various other craft and maintenance personnel were also contacted during
the course,of this-inspection.
!
2.
'
General Plant Tour
i
a. The inspectors toured Units 2 and 3 to gain familiarity with plant
, design and system layouts. In addition, the inspectors evaluated
plant conditions on a sampling basis with regard to the following:
o Equipment condition was observed for indications of system
leakage, improper lubrication, or other conditions that could
,
'
! prevent the equipment from fulfilling functional requirements.
1
i.
_, _ _ _ . _ _ _ _ . _ _ _ . . . _ _ _ _ _ _ _ _ . _ _ , _ , _ - . _ - . _ _ _ _ _ , -
.-~- - - - =~ ~- - - - - - - - -- -- --
'_
-
.,
, .
x -
P
.
A 4 % 3
-
.f y( ' ,
, ,
2
,
'
, ,
~
.
'
'o* Selected process instruments were observed to determine whether-
'
,
'
indicated values were consistent with expected' readings. Alarm
,
conditions were evaluated as to the,cause of alarm and whether
' appropriate corrective actions,were being instituted. _
- .
o Plant conditions and material / equipment storage were observed
tol determine the general state of. cleanliness and housekeeping.
Housekeeping.in the radiologically controlled area was
evaluated with_ respect to controlling the spread of surface and
~
,
l airborne contamination.
,
,
b. .Thefollowingbbservations',notedduringtheplanttours,
4
indicated that although adequate procedural controls appeared to
exist governing the performance of periodic tours by SONGS
r personnel, including management, greater attentiveness and
L
aggressiveness on plant rounds and tours is warranted. Findings 7
associated with a detailed walkdown of the auxiliary feedwater
'
system are documented separately in paragraph 5.
I o During a walkdown of the Unit 2 saltwater cooling system a
mechanical snubber (S2-CS-003-H-011), located about one foot
above the floor in the salt water tunnel,,was observed to be
severely corroded, bringing into question the operability. The
! anubber, which was included in the Technical Specification
'
anubber program, had apparently been in contact with residual
water in the tunnel for some time since the last visusi
.
inspection'in April, 1986. The licensee removed the snubber
! and replaced it with a qualified spare. The removed snubber
, was functionally tested and found to be frozen. Nonconformance
! Report 2-1912 was generated, and the associated engineering
i evaluation concluded that due to the minimal thermal movement
! of the 30" pipe to which it was attached, the snubber would
,
, have remained seismically operable had a design basis
i . earthquake occurred. Nevertheless, the inspector concluded
that the licensee's routine tours of the saltwater tunnel
{ should have identified the degraded snubber.
2
i
- o A chain fall was observed above the operable "B" Low Pressure
Safety injection Pump discharge check valve. Upon making the
~
licensee aware of the condition, the chain fall was immediately
l removed. The licensee was unable to determine exactly how long
the chain fall was positioned above the valve, however,
j maintenance supervision indicated it was at least two weeks.
Procedure S0123-I-1.20 Seismic Controls During Maintenance.
Testing and Inspections, addresses the need to minimize the
potential for ladders, lifting devices, and the portable
j< equipment damaging safety-related equipment in a seismic event.
The inspector noted that although the probability for damaging
4 safety equipment may have been small, normal shift rounds could
.- reasonably have been exTected to initiate the steps necessary
i to remove the unused chain fall.
<
i
t
o A minor. leak, indicated by a relatively large boric acid ;
accumulation, was noted on the Unit 2 "B" LPSI pump casing
_
'
i,
k
>
- - - - - . - - . - - . - - - - _ . - . - - - - . . , - - _ . _ - _ . . - - . -_
p
,
t
,
,
3 .
' '
.~
. .
"
+
studs. The pump was not running and no moisture was observed.
Although an indication of a minor leak does not affect the ' '
operability of.the system, it should be repaired. A maintenance' ,
work request was not submitted until addressed by the inspector
and, apparently, the licensee's tours failed to identify this
item.
o At an instrument panel, located in the outer periphery of'the
Control Room at Unit 3, an annunciator was lit indicating
component cooling water low flow to the running control element
drive mechanism (CEDM) fan (non-safety related system). The
inspector notified the licensee, who determined the lit
annunciator was due to a malfunctioning flow transmitter. A
maintenance work request was not submitted until addressed by
the inspector.
o During a tour of the Unit 2 North Cable Riser Gallery the
inspector'noted that on a portion of a horizontal running cable
tray (U2IGATA3) a fire blanketing material was laid over the
cables in only a portion of the horizontal run. The inspector
was unable to visually determine a reason which would explain
the rationale for only partially laying a fire barrier, in that
the adjacent cable trays appeared equidistant from the subject
tray for the entire length of run in question. Upon further
investigation by the inspector, with the assistance of the
licensee, it was determined that the entire length of tray
U2IGATA3 required the type 3 fire barrier (one it.ch of
Cerabianket) per the Bechtel Raceway Input Data Document -
347.00, Revision 106, dated January 6, 1986. The fire barrier
is required to meet Reg. Guide 1.75 compliance and is not part
of protection provided for. Appendix R safe shutdown compliance.
The licensee initiated Nonconformance Report 2-1915 to rework
the cable tray by installing the required fire blanketing
material. A fire watch was immediately posted. Fallere to
maintain the necessary fire barrier in accordance with plant
installation documents is contrary to 10 CFR 50, Appendix B,
criterion V and is a violation (361/86-25-01).
o Several instances of equipment chained to non-safety related
conduit were observed. Based on discussions with licensee
representatives the inspector learned that although the conduit
was not safety-related, the licensee has been trying to train
personnel not to tie off to any conduit. The licensee was
responsive to the specific cases identified by the NRC, and
expedited proper securing of the equipment.
c. The following plant tour observations were also made by the
inspector during the course of the inspection;
o In various places, the salt water tunnel floor was covered by
several inches of residual water. A sump pump is not located
in the tunnel, requiring the use of portable equipment to
remove water following maintenance activities which require
draining / opening of the salt water system. Pipe support
.
. 4
baseplates in the tunnel showed evidence of general corrosion.
Similarly, the inspector noted corrosion of some baseplates in
the auxiliary feedwater system.
The problem with corrosion of the pipe supports was previously-
identified by the licensee and a corrosion prevention painting
and monitoring program was implemented. Based on discussions
with licenses personnel,'the inspector concluded that the long
term corrosion concerns were appropriately considered by
licensee management. The inspector noted; however, that the
program was not formally included in a station procedure or
instruction. The licensee committed to formalise the corrosion
monitoring program for the affected portions of the saltwater
cooling and auxiliary feedwater systems for all three units by
January 1, 1987. The procedure describing the program will be
reviewed during a future inspection (361/86-25-02). The
licensee also stated that consideration was being given to
developing a comprehensive corrosion monitoring program for all
systems which showed signs of, or potential for, corrosion.
The inspector considered the licensee's efforts in this area to
date to have exhibited strong initiative and management
involvement.
o The inspector evaluated component labeling with regard to human
factors considerations to determine whether plant equipment and
plant piping runs could rapidly and accurately be identified by
, plant personnel. The quality of labeling was found to differ
significantly between systems. The licensee informed the
inspector that a major labeling upgrade was underway at all
three units, and was approximately 50% complete. The inspector
considered the enhanced labeling to be a major improvement.
The licensee stated that the label upgrading for the site was
expected to be completed by July, 1987.
o Plant instrumentation was noted to have outdated calibration
stickers affixed. The licensee stated that the use of stickers
had been abandoned, and a computer-assisted tracking method
adopted. The inspector notec that the outdated stickers could
serve as a point of confusion. The licensee stated that
inattunentation and control technicians have been removing the
,
old stickers when performing routine calibrations and that
i practice would continue. The inspector considered the
! licensee's action acceptable.
l
4
o Several instances were observed in the plant where a valve was
,
found in the open position but not locked; however, the valves
'
were shown on the controlled piping diagram as' locked open. In
j each case, the valves observed were in the correct position
(open) and agreed with that called for in the appropriate plant
l
. lineup procedures. The licensee indicated that plant personnel
were in the process of revising the piping diagrams to maintain
i consistency with plant procedures regarding normal valve
l
positions. Further, the licensee expected to complete the
'
diagram revisions by January 1, 1987.
L
, ,
,
'
- '
, .
5
. - ,
,
o During a tour of the "B" train safety injection pumps, the
inspector was informed by a licensee representative that yellow
and magenta tape on the floor represented a contaminated area,
which should be construed as an invisible vertical wall, thus
designating all equipment beyond the tape as potentially.
contaminated, not just the floor. The inspector noted that no
signs were posted to indicate that the area was contaminated,
nor was a roped barrier used. The inspector stated that the
red badge refresher retraining he attended did not specifically
address the sole use of tape to designate a contaminated area.
During subsequent discussions with health physics supervision
the inspector was informed that the use of berms had recently
been increased to minimize the spread of radioactive material
and the use of tape was also increased to designate the
contaminated area. -The inspector stated that the visual aid to
the worker notifying him/her of a contaminated area appeared to
be reduced, when a " physical" rope boundary is removed, as well
as the sign stating " contaminated area". Discussions with
about 20 plant workers indicated that several individuals who
periodically work in radiologically controlled areas were not
fully knowledgeable regarding what the floor tape signified.
Specifically, they did not realize that they shouldn't reach
across the floor tape to work on equipment. The inspector
informed licensee management that the berms appeared to be a
good feature; however, consideration should be given to using a
sign and rope to designate contaminated areas as often as
feasible. The licensee acknowledged the inspector's comments.
In addition the licensee stated that health physics refresher
training would be revised to better address the use of tape
boundaries. The licensee also stated their intention to attach
a memorandum to each worker's red badge in the security guard
house to ensure employees are familiar with the increased use
of tape boundaries,
o Portable, non-safety related fans were observed, at Units 2 and
3, providing additional cooling to the core protection
calculator cabinets. A temporary modification for the fans was
in place as a result of the licensee's. concern over the
potential for a spurious plant trip due to high temperature
effects within the cabinets. The licensee stated that the
cooling problem will be evaluated and a permanent modification
to-enhance cabinet cooling is expected to be completed during
cycle 4 reload for each unit.
o. Temporary communication cables were observed attached to the
outside of safety related cable trays in the Control Building.
The licensee previously evaluated the potential effect of the
small diameter cables, and concluded that the low electrical
carrying capacity of the cables, as well as the tray loading
effects were negligible, and did not represent an adverse
safety condition. The licensee stated that a permanent
communication system is expected to be implemented during
cycle 5 reload for each unit.
. . - . -. ..- - . - . . . . - -- - - .-
3
1 %.
'
. e.-
6
n -
4 .
% <
~
j .o (During the inspection period, the licensee-experienced some i
. , , difficulty with noble gas associated with charging pump packing :
1 '. leakoff.: The leukoff is routed to a tank.in the pump'roome
-
<
which is vented to the room atmosphere. The licensee' stated
- that a' design change is under review and is being tracked
~
-
'
through Work Request 6614. . The design change would modify the
leakoff piping arrangement to minimize the release of noble
1 gases. '
- t
- ,. l
'
It was also~noted that the noble gases were detected in the
s
. hallway outside the charging. pump room, which would indicate
1
that a portion'of the air flow was' going from an area of
3
potentially higher airborne activity to one of lower expected
s
activity. .The licensee stated that improvements were planned
, in'the near future by sealing.several open penetrations in the
i rooms. Additionally, the licensee expected to improve
, - ventilation flow balancing by the conclusion of the cycle 4
reload.
'
'
,
3. Industry operational Experience Review Program
4
The inspector reviewed the program for evaluation of industry operational
experience by the Independent Safety Engineering Group (ISEG) and
examined a sample of ISEG evaluations. ISEC is responsible for
performing technical evaluations of NRC INPO and vendor supplied
information and is also responsible for conducting surveillances of the
s licensee's Nuclear Operations activities. Following the evaluation, ISEG
'
'then makes recommendations to plant management to improve plant safety
,
.and reliability,
i
The inspector discussed the aspects of the program with the ISEG
'
supervisor and ISEG engineers and reviewed a sample of five ISEG monthly
reports'and approximately 15 evaluations. The majority of the
evaluations were of a limited scope due to the nature of the information
i evaluated; for examples NRC Information Notices, INPO Significant Event
,
Reports and Significant Operating Experience Reports. Some of the
! information evaluated; however, was'of a much broader scope, such as the
report on the loss of feedwater event at the Davis-Besse plant on-June 9,
' "
1985 (NUREC 1154). The inspector reviewed the evaluations for technical
adequacy, level of detail, and timeliness. In general, the evaluations
j were adequate with regard to these criteria. However, the inspector
'
noted that there was no formal requirement in the program that the ISEG
, verify implementation of its recommendations. The inspector observed
1 that ISEG did track and verify implementation as a matter of good
j ' practice, but there appeared to be no procedural requirement to do so. i
4 Licensee representatives stated that they would evaluate this item and
modify the program if considered necessary.
'
!.
4
} With regard to the Davis-Besse event, the inspector reviewed the details
of the licensee's evaluation and the recommendations made. The inspector
i noted that certain critical details of this event, that contributed
i significantly to it's severity, were not included in the licensee's
l evaluation or recommendations. These details included the resetting and
i relatching of the auxiliary feedwater pump turbine (AFPT) following an
!
)
i a
_ _ _ _ - _ .-,_ _ _ _ _ _ . ~ - . _ _ . . _ _ _ . _ . _ _ _ . _ . _ _ _ _
. - . .
'
e '
, ,
.
>
u,
_
. . ,
7
, -
,
4
4
t -
4 overspeed trip, as described in the NUREG. At Davis-Besse, operators had
"',
,
, considerable difficulty performing this action, delaying the recovery of. .
,
'
^ ,
the auxiliary feedwater system. NUREG 1154 states " Toledo Edison's
_
. hypothesis regarding the difficulty that the Control Room operator.
-
..
'
' experienced in regaining AFPT control was directly attributable to the
j
' '
-inability to relatch the trip-throttle valve linkage properly and the
difficulty and delay in opening the trip-throttle valve." Regarding the [
- equipment operator's difficulty in relatching the trip mechanism the
NUREG states " Toledo Edison stated that these equipment operators now
-
believe there was no mechanical problem with the mechanism, but rather '
+
that they did not know how to perform the necessary actions. It is
physically possible to pull the connecting arm sufficiently far to be
able to barely reengage the trip hook at the trip-throttle valve, but not
reset the overspeed trip back at the other end, where the overspeed
tappet and manual trip lever are located. This end of the connecting arm
is behind the governor and is not easy to see." ,
,
The licensee's review of the Davis-Besse event identified the need to
" review hands-on training for in-plant equipment." Additionally, the
licensee's evaluation of an INPO Significant Operating Experience Report
concerning reliability of auxiliary feedwater systems identified the need
- to " expand existing provisions to include demonstration that the turbine-
driven AFW pump can be started manually / locally from the tripped
condition." In neither case do these recommendations specifically
address ~part of the root cause of the AFPT problems experienced at
Davis-Besse, namely, to ensure that the overspeed tappet mechanism is
properly latched when resetting a trip. The inspector recognized that
'
the licensee's review was still ongoing, but it appeared doubtful that
this item would have been identified.
In investigating this issue to determine if San Onofre is susceptible to <
the same problems experienced at Davis-Besse with regard to resetting the
AFPT, the inspector noted the following:
'
o Operating procedures did not address the need to verify that the
'
overspeed trip tappet acchanism is properly engaged when resetting a
trip.
o Operator training lesson plans did not discuss this issue.
4
o Based on a limited sampic, operators were generally not
knowledgeable of the need to .2sure the overspeed tappet mechanism
is properly engaged.
o The overspeed tappet mechanism was covered by a shroud over the area
! that made it dark, making access and observation somewhat difficult.
Taken together, these items indicate a need for attention to reduce the
likelihood of this relatching problem occurring at San Onofre. Licensee
representatives stated that procedures and training would be modified to
address this issue. This will be verified in a future inspection
(361/86-25-03).
l
l
A
~ - e -- ,y-.-r y ~ - - -.y..-...m_.,.,-.r..-,__--,.-.,v-,.- - ,.- , r.- -, -,.e.. ~ , --,.--,..,-~,---,,.m-.#-.-~-3,,- -c.m. -, . -~-.,-e
,
Q.t [/ ,
s , ,
.f. - .
.
,i <
'
,, , *' p/ g
'
'
3. . ' , . i ' i . 7 _ 'c ,
~8
' '
i >
_t '
[' 'l s > 1
.
, [ . 3; *
-
, ,
- The inspector concluded that, in general, the licensee's program for '
evaluation of operational experience was acceptable.- However, for the ,
- t\
' ,
evaluation of-large, complex, multi-faceted events, such~as the
~ ~
,
>
j
Davis-Besse. event, it appeared that certain critical details of an event
~
' '
may not be evaluated in sufficient detail to identify the underlying
'
rg cause. The inspector noted that there does not appear to be a clear ,
policy regarding which' licensee group is responsible for extracting and
~
'-
evaluating very detailed,'and important information-resulting.from
evaluations of major operational events. In the case of resetting the
AFPT overspeed trip, it was not clear whether ISEG or the Operations
division had responsibility for ensuring that significant details, such
. as verifying that the overspeed tappet mechanism is properly engaged,
were considered. Licensee representatives stated they would evaluate
this concern. This item will be followed in a future inspection
'(361/86-25-04).
No violations or deviations were identified.
4. Walkdown of Rancho Seco Overcooling Event
a. Introduction
On December 26, 1985, the Rancho Seco Nuclear Generating Station
experienced a severe overcooling event which resulted from a loss of
non-vital power to the integrated control system (ICS). The NRC
conducted an in-depth investigation into the event. The results of
the Incident Investigation Team (IIT) were reported in February,
1986 in NUREG 1195. In addition to plant specific design elements
which contributed to the event, the IIT identified operational
weaknesses in the areas of human factors, procedures, and training
which appeared to have adversely affected the licensee's ability to
adequately deal with the event.
Based on an analysis of the factors identified in NUREG 1195, the
inspector developed a similar scenario of events to determine if
weaknesses similar to those affecting Rancho Seco were evident
within the licensee's program at SONGS 2 and 3. The inspector
interviewed licensed control operators and equipment operators
concerning their response to the postulated events, and walked-down
, _the sequence of actions which would be taken by each.
!~
i
.
b. Postulated SONGS 2/3 Scenario
!'
,
The inspector reviewed the main feedwater, auxiliary feedwater and
- .
main steam control systems with cognizant personnel from the
licensee's technical support and training departments. Due to
( differences between the control design of CE, at SONGS 2/3, and B&W,
'
at Rancho Seco, the inspector found that an identical sequence of
events resulting from a single failure / loss of non-1E power could
{ not ba developed at SONGS 2/3. However, to challenge the operators
! under a similar sequence of undercooling and overcooling events the
- -
inspector developed a postulated scenario involving the following
multiple failurest
L
l
!
!
'
t .
-
-
-
,
.
s ,
?
'.
& .
.
?, - La , ;
9'
1
'
s'
, . ,,
o loss of non-1E power.to the feedwater control system (FWCS)
1
o' loss of non-1E p'over to the steam bypass control system (SBCS)
.one failed' open atmospheric dump valve (ADV)
~
-o
'two failed'open auxiliary feedwater (AFW) valves
~
o
'
~
-
' The 1nspector discussed the following postulated scenario with plant -
operators, by walking through the operator responses at the
simulator. -
(1) Loss of non-1E power to the feedwater control system (FWCS)
causes main feedwater (MFW) pumps to slow to minimum speed.
(2). Low steam generator (SG) levels cause a reactor trip / turbine
trip and emergency feedwater actuation system (EFAS)
initiation. With the "B" train motor-driven AFW pump (No.
P141) out of service for maintenance, the "A" train
motor-driven AFW pump (No. P504) and the steam-driven AFW pump
(No. P140) supply emergency feedwater to both steam generators.
(3) Loss of non-1E power to the steam bypass control system (SBCS)
prevents actuation of the steam bypass valves following the
turbine trip, causing main steam safeties to lift.
(4) Control operators fully open the atmospheric dump valves (ADV)
to reseat the main steam safeties.
(5) The ADV for one steam generator (No. E088) jams open causing
low steam generator pressure which actuates the main steam
isolation system (MSIS).
(6) Two auxiliary feedwater valves (HV-4705 and HV-4730) jam open
following MSIS actuation and fail to satomatically close when
SG 1evel has recovered to 30%.
This scenario created an overcooling event requiring the operators
to deal with alternative methods of securing AFW flow to the
affected steam generator. Although this scenario describes events
due to multiple failures of equipment, which is beyond the design
basis of the plant, manual operation of the ADV and AFW valves is a
required mode of system operation in dealing with a shutdown from
outside the Control Room under the licensee's abnormal operating
instruction S023-13-2.
The inspector also walked-down an equipment operator's response to
the postuisted event as directed from the Control Room to manually
operate various valvas. The inspector discussed elements of manual
operation of systems equipment with equipment operators during these
walkdowns.
c. Observations
Based on discussions and walkdowns with licensee personnel regarding
the hypothetical sequence of events at SONGS 2/3, the inspector made
the following observations:
,
. . -
. .* 10
'
(1) The licensee's symptom related emergency procedures provided
guidance to the operators in dealing with the postulated event.
However, the inspector found the licensee's procedures to
provide limited specific direction to achieve the desired
. control. When attempting to secure AFW flow, an operators
decision to trip an operating APW pump or manually attempt to
isolate the flow.was dictated by the operator's perception of
the urgency of achieving the desired control. The licensee's
procedures did not'contain the detail in prescribing specific
operator actions in light of their priority within the event.
(2) The. operators showed no reluctance to trip an operatin8
AFW pump due to any exaggerated concern for potential problers
in restarting the pump. The operators indicated confidence in
the reliability of the AFW pumps and controls.
(3) Without the capability of remote control of the AFW valves from
the Control Room'to regulate AFW flow to the steam generators.
the initial response of the control operators was directed
toward tripping the AFW pump. However, since this action would ,
also have secured all feedwater to-the unaffected steam
generator (No. 089), the control operators readily dispatched
equipment operators to manually operate the AFW valves rather
than trip the operating AFW pump. The inspector observed that
the' control operators directed operation of manual isolation
valves to secure or throttic the AFW flow when a postulated
difficulty was encountered with the manual operation of the
motor-operated AFW pump discharge valves. The inspector found
that the operational use of the manual isolation valves was not
specifically directed in any licensee procedure, nor was the
operability of,the valves controlled under any licensee
preventative maintenance or surveillance program. This concern
for the operability of equipment in dealing with plant
transients was also emphasized in NRC Information Notice 86-61.
1 (4) To manually initiate both AFW trains in anticipation of a loss
,
of main feedwater, the control operators pushed four push
buttons to manually actuate the emergency feedwater actuation
system (EFAS) rather than individually start each pump and open
each valve to establish flow. Although the four pushbuttons
were anchored together horizontally on the control board, the
four pushbuttons were grouped with identical pushbuttons for
nanual initiation of the rain steam isolation system (MSIS) and
containment isolation system (CIS). No color coding was used
to distinguish among the pushbuttons in this cluster. The
Irbeling for each pushbutton was descriptive but of relatively
srall style print requiring close inspection to read.
At Davis-Besse, control board layout of pushbuttons was
identified as a contributory human factors design weakness. At
Rancho Seco, the lack of readily identifiable switch position
labeling prevented the operators from quickly restoring ICS
power. In NUREG 1154 the IIT commented that the labeling of
_ _ _ .__ _ .. -_ __ . _ _ _ _ _ _ _ _ _ . - _ _ _ _ _ __ -. . . _ . _-
. 11
,
. controls for manually initiating and use of the most important
systems at Davis-Besse should have been unambiguous.
(5) Equipment operator familiarity with manual operation of the ADV
and A W valves was found to vary considerably between.
individuals, suggesting that actual infrequent manual operation
of the equipment did not provide sufficient operator experience
to insure prompt, effective action in response to an event. .
At Rancho Seco, improper manual operation (use of a ' Cheater')
l.
of an AFW control valve caused damage to the valve and
,
exacerbated the overcooling event.
In NUREG 1154, the IIT concluded: "For plant events involving
'
!- conditions outside the plant design basis, operator training l
l and operator understanding of systems and equipment are key to l
> the success of mitigating actions taken by the operators. It l
' is not practical to rely on detailed step-by-step procedures i
for such events." I
t
I ;
'
(6) Hanual operation of the atmospheric dump valves required ,
overhead manipulation of the valve handwheel by an equipment
operator from a platform located approximately seven feet below
the handwheel. In addition, the isolation valve for the air
supply to the Bailey positioner was mounted on the ADV
pneumatic operator approximately 13 feet above the platform,
requiring an equipment operator to climb adjacent sumil bore
piping to reach the switch. Based on these observations of the
physical accessibility of the ADV and discussion of its manual
operation with equipment operators, the inspector found that
the human factors elements of valve design would hinder prompt
and effective manual operation of the valve.
(7) The effectiveness of radio communications throughout the plant
appeared to have been substantially improved through the
extensive use of repeater stations. Operators indicated
confidence in the usefulness of radio communications in all
areas of the plant. However, both control operators and
equipment operators complained that it was difficult for the
control operators to understand transmissions from high
background noise areas.
d. Conclusion
When presented with a postulated event based on actual industry
operating experiences involving failure beyond the design bases of
the plant, the inspector found certain common weaknesses to be
evident within the licenuee's program, indicating a need for
improvement in the licensee's readiness for manual operation of
plant equipment. This item as well as the concern associated with
the manual back seating of motor operated (paragraph 8 below) will
be verified in a future inspection report (361/86-25-05).
No violations or deviations were identified.
. _ _ _ ._______ _ - ___ _ _ ______-_ - ____ -__ _
1
, I
o 12
5. Auytliary Feedwater System (AFWS) Walkdown
The inspector toured the AFWS pump houses and tank farms for Units 2 and
3 on various occasions during the course of the inspection to assess the
physical condition of the hardware, environmental conditions and general
housekeeping in the areas. The inspector observed the following
deficiencies and identified each to licensee representatives for
resolution:
a. Two (2) station lights were apparently burned out above the
refueling water storage tanks (RWST) for Unit 2.
b. The open position marker for the yoke mounted scale of the local
valve ponition indicator (VPI) for valve No. 2HV-4712 had been
dislocated off the scale apparently due to unintended interference
with the valve stem mounted pointer.
c. No local valve stem VPI was installed on valve No. 2HV-4705.
Position indication was determined solely from the motor operator
indicating dial.
d. The yoke mounted scale of the local VPI on valve No. 2HV-4714 was
labeled only to show an open direction and did not clearly indicate
a full open or closed position.
e. No local VPI was observed on valve No. 2HV-4713. Operators
indicated that valve position could be identified based on
observation of the adjacent ilmit switches.
f. Valve No. 2HV-4730 was located below the decking level with the
valve motor operator above the steel grate deck. Local valve stem
VPI was not observable from above the grating. Valve position was
determined by observation of the motor operator dial indicator,
g. The motor operator dial position indicator for valve No. 3HV-4715
was observed to read approximately 10% open although the valve was
chut.
h. The motor operator dial position indicator for valve No. 3HV-4706
was observed to read less than 0% open when the valve was shut. No
valve stem mounted local VPI was installed. A sign was affixed to
the valve which read " Warning Valve overtravels Do Not Open More
Than 100% Per Indicator". When questioned by the inspector as to
the meaning of the warning sign, several operators were unclear as
to the reason for the posting.
The inspector further observed that the valve was located within a
posted radiation area; however, personnel access to the valve was
not restricted under the controls or any radiological entry permit.
1. The pointer and protective cover for the motor operator position
indicator were not installed on valve no. 3HV-4713. Crude pen
markings were visible on the valve yoke apparently nerving an a
_ _
,
~ <
. ,
t
- '
, 13 -
.,
local VPI. As in item h, the valve was also located within a posted
radiation area.- .
j. The stem rounted pointer on the rising stem manual isolation valve,
no. Sg-1305-MU-533 was observed to be midway within a clear plastic .
stem protector cap. The full open or full closed positions were not
labeled on the cap.
k. The solenoid operator conduit for valve no. 3HV-6154 was found to be
loose on the valve and not properly secured. Furthermore, the
conduit was routed in front of the chemical addition flow meter
adjacent to the A N pump which blocked the operator's view of the
flow scale. >
1. The pump bearing packing for A NS pump 2V-140 was found to be
leaking excessively.
m. The motor operator position indication'for valve no. 2HV-4706
indicated less than 0% open when the valve was shut. No other local
VP1 was apparent on the valve.
n. The bearing lubrication fittings for several manual isolation valves
were observed to be painted over exhibiting no evidence of recent
lubrication.
o. The stem mounted open travel stop on valve no. 3HV-4705 was
bent apparently due to overtravel despite a posted warning sign to
limit operation (as described in item h). A station light was
inoperative above the valve and had recently been identified and
tagged for repair by the licensee. However, an 8 hour9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> emergency
light above the valve had also been tagged by the licensee since
March 6, 1984 and was still inoperable.
Furthermore, the declutching lever on the limitorque operator was
installed backwards. When attempting to operate the valve manually,
an equipment operator must depress the declutch level to disengage
the electric motor and engage the handwheel to operate the valve.
Attempting to depress the lever as indicated with the declutch lever
installed backwards, could damage the mechanism and not allow the
handwheel to be engaged for manual operation.
p. The valve stem nut which secures the pointer for local valve
position indication for A W valves 2HV-2712 and 2HV-4705 interferes
with the valve stem packing retainer nuts. For valve 2HV-2712,'the
stem nut appeared'to have contacted the retainer nuts during valve
closure and deformed the nuts over approximately 1/2 of their
thickness. This 3,nterference was apparently not identified by
maintenance perscruel during repacking of the valve.
In general, the inspector found a lack of attention to detail in the
material condition of A W valve local position indication. The inspector
discussed this weaknes a with licensee personnel who acknowledged that
local position indication of critical motor operated valves was not
included in their current maintenance program. The inspector will review
4
~
. 14
the licensee's actions to improve the position indication of critical
motor-operated valves. This item will be followed up in a future
inspection (361/86-25-06).
No violations or deviations were identified.
6. Remote Valve Position Indication
The inspector examined the adequacy of valve position indication in the
Control Room and at the remote shutdown panel with an emphasis on the
operation of controllers and on valves that do not have direct position
feedback in the Control Room.
There are approximately ten valves that do not have direct position
indication in the Control Room except when they are fully open or fully
closed. However, in most cases the process parameter that the valve is
controlling is indicated on the controller such that direct valve
position indication is not necessary. Exceptions to this are the
atmospheric dump valves (ADVs) and the main steam isolation valve bypass
regulator. These valves only indicate a " demanded" position in the
control Room. The inspector reviewed procedure S0123-0-23.1, " Valve
Operation" and discussed operation of these valves with control
operators. The inspector verified that the procedure and operator
training directs operators to use alternate means such as a process
parameter to verify valve position. In addition, the inspector noted
that operators were generally knowledgeable about valves that displayed
demanded position and the need to use alternate means to verify this
position.
The inspector examined controllers to determine if scales were clear and
unambiguous, free of extraneous markingu; whether operator aids were
correct and properly controlled; and established differences between the
Control Room and the remote shutdown panel. No concerns were identified
regarding the scaling of controllers. The inspector noted that procedure
50123-24, " Administrative Control of Operational Aids" appeared to
adequately control the use of operator aids. With regard to differences
between controllers in the Control Room and at the remote shutdown panel,
the only controller in the sample located at the remote shutdown panel is
for the ADVs, and is identical to the Control Room controller.
The inspector interviewed QA and engineering staff to determine if human
factor reviews have been conducted to reduce ambiguities on Control Room
instrunentation and the need for operator aids. The licensee's
engineering organization has reviewed the Control Room panels for human
factors and documented this review in a detailed Control Room
supplemental report dated January 1986 (M37328). The report identifies
human factor concerns and provides solutions and a schedule for
implementation. The licensee's review of human factors for Control Room
~
panels appeared to satisfactorily in resolve areas of concern.
No violations or deviations were identified.
.
- 15
7. Licensee Action on IE Bulletins
(Closed) NRC Bulletin 85-03, Motor Operated Valve Common Mode Failures
During Plant Transients Due To Improper Switch Settings
The inspector reviewed the program established in response to NRC
Bulletin 85-03, " Motor-operated Valve Common Mode Failures During
Plant Transients Due To Improper Switch Settings," which was generated as
.
a result of the Davis-Besse event. The areas inspected included: the
licensee's program for establishing the maximum differential pressure
(dP) expected during operation of safety related valves during normal and
abnormal events; the program for establishing the baseline data used in
determiring switch settings; the program for implementing motor operated
valve testing; procedures for performance of valve testing; and the
licensee's response to the Bulletin.
NRC Bulletin 85-03 specified that a review be performed of the design
basis for the operation of valves required to be tested for operational
readiness per 10 CFR 50.55a (g). These valves are located in the high
pressure safety injection (HPSI) and auxiliary feedwater (AFW) systems.
The licensee elected to perform pressure testing of valves in these and
other systems to determine the actual dPs experienced on all safety
related valves by use of a hydrostatic testing device. This data was
compared to the system design pressure and expected dPs, and the highest
value was selected for use in the analysis of valve forces. The licensee
provided this information along with the valve and actuator
specifications to a local engineering company in order to calculate the
torque and thrust forces to be expected on these valves. With this
methodology, the required switch settings were established.
The inspector reviewed the data packages for the following valves of
various size and types
o 2HV4705, AFW Pump, P-140, discharge to Steam Generator E-088
o 2HV4715. AFW to Steam Generator E-089, isolation valve
o 2HV9303, Containment emergency sump outlet
o 2HV9434, HPSI to reactor coolant loop 2A hotleg
o 2HV9323. HPSI header No. 2 to reactor coolant loop 1A
o 2HV4716, AFW Pump, P-140, turbine stop valve
o 2HV6551, CCW Pump, P-025, minimum flow to loop A
These packages contained the engineering torque and thrust calculations;
the electrical drawing for the valve; the as-found and as-left MOV
testing data reports; and, the field change notice (FCN) that was
prepared to add torque switch and rotor settings to contact development.
The packages also included nonconformance reports (NCRs), when necessary,
if discrepancies were identified.
_ _ _ _ - _
-_-_
.
.. . .
-
- '
,. 16 <
t
The Bulletin specified that the correct switch settings be established
and maintained for the MOVs. The inspector noted that the information on
T
switch settings was included on the FCN and the electrical drawing for
the valve. The MOV testing personnel set the torque switches in
accordance with the requirements established by Engineering as a result
of the calculations and pressure testing that was performed. The
licensee set the torque bypass and Itait switches in accordance with the '
settings determined by industry practice with the concurrence of the
MOVATS company. These settings were clearly identified on the electrical
drawing for the valve and the FCN that was used in conjunction with the
maintenance order (MO) that was issued to control testing of the valve.
The inspector observed MOV testing on two valves; 3NV9331, low pressure
safety injection loop injection valve, and 3NV8150, shutdown cooling heat
exchanger E-004 outlet isolation valve. The inspector verified that the
MOV personnel changed the switch settings, when required, in accordance
with the FCN. In addition, QC inspection of the valve rewiring and
as-left conditions was noted.
At the conclusion of the MOV testing on each valve, the valves were
demonstrated operable, under zero dP conditions, to satisfy post-
maintenance testing requirements established by the MO. The Bulletin
requirements for valve testing under maximum dP conditions was satisfied
during the pressure testing phase, previously mentioned.
The inspector reviewed the following procedures used in conjunction with
MOV testings
o 50123-I-6.8, " Actuators - Limitorque Model SMB-0 through SMB-4 and
55-0 through 88-4 Disassembly, Inspection, Repair, and Assembly".
o 80123-I-6.7, "Limitorque Model SMB-000, SMB-00, and 88-00
Disassembly, Inspection, Repair, Assembly, and Adjustment".
o 50123-1-8.313. " Actuators - Motor Operated Valve Analysis and
Testing System M0 VATS".
o 5023-I-9.5, "Limitorque MOV Inspection".
o 5023-I-6.150, " Butterfly Valve Limitorque Actuators".
o Various MOs and FCNs for valves tested with the MOV test equipment.
The inspector verified that these instructions were clearly written and
followed by the MOV testing group. All instructions were reviewed and
approved by responsible licensee personnel. 50123-1-8.313 and the Mos
reviewed specified that torque switches are to be set with the valve in
mid-stroke so that the believille spring is in a relaxed condition.
50123-I-8.313 also required the as-found and as-left switch settings to
be recorded and the MO listed the QC holdpoints to be checked. The
inspector noted that none of the procedures specified whether the limit
and/or torque switch settings are based on stem or disk movement.
However, the MOV testing personnel interviewed, uniformly indicated that
the settings were based on stem movement. If the stem movement differed
from the corresponding disk movement, this would be identified on the
4
~_ .. . -_-_. _ _ . . _ _ _ . _ ._
_
, ,
t
n "
$ j.
- '
.9; 17
1< s - ., , s
, v.c , , ...
- - .sigrature t' race during the MOV . testing of the valve. This is further
' discussed sti paragraph 8.
t .
- .
'ihe inspector's review dfd'not include an assessment of the technical
adequacy of the licensae's initial response to the Bulletin. This will
be performed by 'the 'NER staff. Howevne, the inspector did review the
, ' imples.cntation of the licensed.s program detailed in the response. This
review indiestedsthat the MOV testing personnel were adequately trained
and knowledgeable'of the equipu nt usea in valve testing. In addition, ;
'
these perscenelefellowed the procedures that have been developed to check
'and/or O rify the various switch settinr,s. The inspector considered the
licuaset's is.plementation'of the progran: to be acceptable.
This item is closed for all three units; however the NRR staff's-review
and approval of the technical adequacy of the licensee's response to the
4s11etin'is still outstanding.
'
No violations or deviatioze ware identified.
6. M0_V and A0V Manual Opgatico; and Switch Settinas
The, incorrect' manual operation or inproper setting of torque and bypass
-switches may result in damage to, or inoperability of valves.
The ivspector.exantned the procedute's and training that were provided to
plant personnel to manually operate valves under emergenc.y conditions to
dinermine procedure and training aduquacy. The procedures and training
releting to switch settings were also reviewed. In the systems examined,
~
most of the valves that cou.1d be operated manually were either anual or
motor cperated valves. There was also a small number of air operated and
electrn-hydraulic valves in the systems.
The inspector reviewed the SONGG 2 and<3 training program for motor and 1
air oprated valves to assess the correlation with vendor supplied
cperating and maintotance procedures. The review included lesson plans
fc,r maintenance and operations, operating and maintenance procedures,
classroom materials including visual aids, mockups, and cutaway models.
Some of the lesson plana reviewed at 50NCS included
o 1.esson # MT-7355 "Limitorque MOV Actuators"
o N VATS 2000 Series (from Vendor)
The SONGS operations and maintenatice lesson plans, procedures and
,
'
classroom mater 161s were found to be consistent wf th vendor supplied '
materiet.L The procsdures and training material also reflected the latest
rwistona and updates from vendor and industry operating experiences.
Operations and Maintenance personnel were interviewed for their basic
' operational and nachanirs1/ electrical knowledge of motor and/or air
eterated valve,setuatore. The operations and Maintenance personnel
interviewed extfbited a knowledge of valve operation and maintenance
consistent with the traiett g material provided. However, based on
discussions with plant operttors, the inspector concluded that
- . L .- w ----- ----- -- -_-
.
f
.
..
..
O 18 '
,
-
,
.
reinstruction on valve operation procedure S0123-0-23.1 was warranted; ;
specifically, that " manually seating or backseating an MOV in other than a'x.
, 'the safety-related position may cause it to be unable to overcome the > V!
opening / closing torque of the motor and therefore shall be declared
inoperable until operated electrically". This area will be verified as
part of the rollowup action referenced in paragraph 4.d of this report.
Local valve position indication was discussed with the operators. The
operators indicated that positive local valve position indication was
used when local manual valve repositioning or valve position verification
was required. Operators were able to identify and use local valve
position indication when provided. The inspector noted in walkdowns of '}
some of the systems (e.g. auxiliary feedwater) that some of the valves ,
had local indication problems. Details are provided in paragraph 5.
Operator instructions and operarious personnel were interviewed with
respect to the use of valve wrenches and " cheaters" on valves that are
difficult to operate with standard equipnent. SONGS valve operating
procedure 50123-0-23.1 prohibite the use of " cheaters" on motor,
pneumatic and hydraulic operated valves as well as diaphragm valves. The
procedure also limits the size of the cheater to be utilized to operate
manual valves.
The operators interviewed had a working knowledge of the prohibition to ,
use cheaters on the above valves in safety related systems, and the limit
on the size allowed to be utili=ed on manually operated valves. This
prohibition is also emphasized in classroom instruction and operating
procedures.
The inspector reviewed training records and compared the records to lists
of equipment operators and maintenance personnel who could be dispatched
to operate or perform repairs on valves during emergency and/or off
normal conditions. The staffing lists indicated that sufficient trained
personnel were available on crews for operation of valves during !
emergency and/or off normal conditions. .
7
The inspector interviewed craft and supervisory personnel with respect to
the physical setting of torque and limit switches. The personnel seemed
knowledgeable on the setting of the switches and the use of the
applicable procedures. The licensee's craft personnel also used special
gloves and tools for better manual dexterity when working with the small
parts associated with setting of the switches.
When the craft personnel weta observed changing the switches, the valves
were also evaluated for material condition, environmental qualification
(EQ) and quality control (QC) checks. The actual value for changing the
limit or torque switches came from the NSSS Engineering group and
appeared reasonable. It was noted by the inspector that training was
improved to emphasize the importance of EQ and material condition checks.
The Limitorque MOV inspection procedure S023-I-9.5 now includes
additional text regarding material condition checks and EQ wiring
requirements. It was also noted that for the safety related valves that
were being analyzed, QC personnel documented the as-left condition on ,,
'
almost every valve. However, the QC personnel had only documented five r
.
5
. .. .- ~ --
- . - - - ., , -- , .-- . - - , . . . _ -. v_
., ..
y_ . _
, - -. .- - . . .
'
Q N , i> ~ 'd.', ;'
'
.
~.
'
- hi
- f . ,
~
,
..,.j
' i??:3 R f . ^
>
19 ,
'
e
[!gb [~ M
,
p p
1, , of these same safety.related valves in the as-found condition. . The.QC
iT < checks in the as-found conditions were, therefore, relying on the craft
-
personnel. The licensee committed that, in the future, the as-found
condition for analyzed safety related valves will be documented.
1-
The use and training on MOV test equipment was also inspected. The
licensee currently has nine personnel trained to use the test equipment,
eight of whom were trained in June, 1986. The training program was
=E ' analyzed for program content and compliance with vendor recommendations
, 'and appeared' adequate to provide training. The supervision of the MOV
-test personnel in the field also appeared adequate.
The inspector inte'rviewed several MOV test personnel on the signature
'
traces.. All personnel interviewed appeared knowledgeable on what the
signature trace indicated, and could refer to training material to
evaluate'the' trace. The personnel were aware of when the bypass switches
cut in os out : of possible degradations with the stem or disc, and of
possible degradations with the worm gear.
'
Several, historical signature traces were also observed by the inspector,
'
mort..of which appeared normal. An abnormal signature trace was also
observed by the inspector (Pressurizer sample valve, 2HV-0517), which did
not haveian easily recognizable trace. The licensee had developed an
hypolbeses for this signature trace, but did not want to state any cause
for the valve problems until the valve could he repaired. This valve
and similar valves were declared inoperable in both units.
' , . , g
-
The licensee keeps official records of the valve performance in its
.
Corporate DocumentNManagement department. The Maintenance department
- -
7/. ~ kept unofficial records for valve trending to observe for possible valve
degradation.,The calibration data for the MOV test equipment appeared up 1
to,date and traceable to the National Bureau of Standards (NBS) '
, requirements.
No$1olationsordeviationsvereidentified.
,
9. (Closed) NRC Information Notice 86-34, Improper Assembly, Material
!-
Selection. and Test of Valves and Their Actuators
1
) Improper assembly of valve actuators on valves can compromise the
,
- ' structural strength or operability of emergency core cooling system
](ECCS)valvestunder'.ccident. conditions.NRCInformationNoticeNo.-
y 86-34, dated May 13, 1986, describes incidents resulting from improper
material selection and inadequate assembly procedures for safety-related
-
power actuated valves. This Information Notice is applicable to valves at
the time of initial installation and during post-maintenance reassembly
- j,, *
or test.
+
This inspection examined whether SONG's valve actuator assembly program
,
included sufficient instructions to assure proper assembly of valve
'
r.ctuatora consistent with vendor instructions. The inspector reviewed
44 '
1 maintenance procedures for ECCS valves'that us. actuator assemblies to
" -
, determine if vendor file recommendations were included; specifically,
4 <
~
t
g I-
{, gt -g
'i-* ~_ _ _ , - . _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _
____m_
. . . . _
.
8
9
- * ' '
., 20 -
valve supplier and' actuator supplier data. The inspector reviewed the
folloing w procedures:
o Maintenance Procedure S023-I-6.150, " Butterfly Valve Limitorque
Actuators".
o Maintenance ProcedEre S023-1-6.31, "WKM Model "M" Gate Valve
'
Maintenance".
.
o Maintenance Procedure S023-I-6.75, "WKM Model D-2 Gate Valve
Disassembly / Assembly".
o Maintenance Procedure S023-I-6.7, "Limitorque Model SMB-000, LNE-00,
and SB-00 Disassembly, Inspection, Repair, Assembly and Adjustment".
o Maintenance Procedure S0123-I-6.8, " Actuator-Limitorque Models SMB-0
through SMB-4 and SB-0 through SB-4 Disassembly, Inspection, Repair
and Assembly". ,
o Torque Manual (M-37204).
The inspector concluded that the maintenance procedures provided for
resolution of conflicting instructions, specification of both bolting
torque and minimum required bolt thread engagement. The procedures also
specified the valve stem position during bolt torquing when specified in
the vendor manuals.- The inspector reviewed selected work orders which
involved actuator disassembly, bolt removal or replacement.- These
selected work orders specified the correct material certifications for
the replacement bolts, the correct bolt torque instructions, and
identified the QC hold point verification activities. The testing
instructions for the final check out of the valve after repair was
specified. The maintenance procedures required a maintenance data record
,
form to be completed which provided for the torquing values, restoration,
- and post-maintenance testing to be verified as complete. Any test
equipment used was also-specified with the calibration due date recorded.
The licensee's response to NRC Information Notice 86-34 was verified by
,
the Independent Safety Engineering Group (ISEG) review via ISEG Log
Number 86-ISEG-090 dated June 9, 1986. Their event analysis reported
that no further evaluation was required. Torquing requirements, if not
covered in a procedure, were addressed in SONG's Torque Manual.
It appeared that SONG's valve actuator. assembly program included
sufficient instructions to acsure proper assembly of valve actuators
consistent with vendor instructions.
I' No violations or deviations were identified in this area.
l
.
-
,
, ,- + . - . . - - . , , , , . - , , .- _- r , , . . - . - - . , , - . , -,.~.- - . - _ - - n. . . , . _ , , , , - . - - , , . - - , - - - - - . . , , , , ,
-
.
.. 21
10. Loss of Shutdown Cooling
a. Introduction
The potential significance of loss of shutdown cooling (SDC) or
decay heat removal (DHR) has long been recognized by the industry
and NRC. AEOD Case Study Report C503, dated December, 1985,
indicated that between 1976 and 1983, 130 loss of DHR events were
reported to have occurred during approximately 500 reactor years of
operation, bkny of the events which have occurred thus far may
serve as important precursors to more. serious events, though no
_
. serious damaEc has resulted from events to date. Two leading causes
identified are inadvertent automatic-closure of suction / isolation
valves and loss of reactor cooling system (RCS) inventory.
On March 26, 1986, while in cold shutdown, SONGS Unit 2 shutdown
cooling system (SDCS) experienced a total loss of flow for a period
of forty-nine minutes (LER 86-07). This occurred while the RCS
level was being reduced to repair a leaking cold leg steam generator
nozzle dam. Using the established level indication, which was later
found to be in error, the RCS was drained to a level where vortexing
occurred in the RCS/SDCS suction connection causing SDCS/LPSI pumps
to eventually become airbound. SDCS flow was reestablished in.
.
approximately.49 minutes and high pressure safety injection was then
manually initiated to further assure SDCS flow stability. Later
review by the licensee indicated that with a loss of RCS
recirculation, bulk boiling occurred in the core approximately
thirty-eight minutes into_the event. The licensee concluded the
cause of the event was erroneous level indication resulting in the
operators not recognizing the RCS low level condition prior to
complete loss of SDCS flow. A Severity Level IV violation
(50-361/86-11-02) was issued on the improper application of reactor
vessel indication which resulted in inoperability of the SDCS. The
licensee committed to a number of corrective actions with respect to
LER 86-07 and the violation.
The objectives of this inspection were: to evaluate the
susceptibility of SONGS Units 2 and 3 SDCS to a single failure which
could result in inadvertent automatic closure of SDCS purp
suction / isolation valves; and to assess the effectiveness of the
licensee's actions in response to lessons learned from the generic
loss of SDC and LER 86-07.
b. Inspection and Findings
(1) System Configuration and Surveillance Review
The licensee's logic and electrical diagrams were reviewed to
verify that redundant parallel power and control signals were ,
provided for shutdown cooling letdown valves 2/3 HV-9339, 2/3 l
HV-9337, 2/3 HV-9378 and 2/3 HV-9377. These valves were
arranged as two redundant sets of valves in series creating a
one-out-of-two-taken-twice logic for the singic letdown / suction
line. This lineup would preclude a single failure from 1)
. ,
- *
. , 22
,
~
isolating'the letdown / suction line inadvertently or 2) failing
to isolate if above the over pressure protection setpoint. The
licensee's' drawings 30678, 30565, 30677, 30562, 30555, 30564,
30554' and 30563 indicated that ' redundant parallel power and
'
'
. control signals were provided for'each of these valves. The
licensee's surveillance procedures for these valves were also-
reviewed. ~From this review, Unit 2/3 letdown / suction line did
not appear to be susceptible to single failure, resulting in a
h
subsequent loss of SDC.
(2) Licenser's Evaluation of Loss of SDC Events
A number of evaluations of both the industry and Unit 2
loss-of-SDC' events were performed by various licensee groups.
The inspector reviewed the following evaluation reports:
o " Nuclear Safety Organization Independent Evaluation of
Unit 2 Loss of Shutdown Cooling Event" May, 1986.
o " Loss of Shutdown Cooling" San Onofre Nuclear Generation
Station, Station Incident Report No. SO2-86-004, July,
1986.
o ISEG Operating Experience Evaluation: SOER 85-04, " Loss
or Degradation of Residual Heat Removal Capability in
PWRs", July, 1986.
o " Shutdown Cooling Incident Experience Review", SCE Nuclear
Training Division, May 1986.
The conclusions and recommendations in these studies appeared
to be consistent and adequate. The inspector reviewed the
implementation of the following specific recommendations and
commitments.
(3) Reactor Vessel Level Indication
The licensee's post event investigation revealed that during
the March 26, 1986 event the Unit 2 vide range refueling water
level indicator (RWLI) was out of calibration. The detector's
performance was also adversely affected by operation of the RCS
educator. Furthermore, a 2 1/2 inch error in the tygon
manometer reference scale combined with a 10 1/2 inch error
caused by an air bubble that existed in the tygon tubing
resulted in an indicated level of 13 inches higher than actual
level. This resulted in non-conservative level information
which led the operator to believe that the reactor vessel level
was higher than actual, and resulted in reducing the level to
the point of vortexing the SDCS/LPSI pumps.
i
The centrib+1ng causes to the erroneous level indication were
determined to be:
.
__ m.. _
_. -_- . . _ . ._ .
,-
- >- '23
s !
-
o RCS refueling level detectors were classified as
non-safety related, thus, their control was not-as-
formalized as necessary.
o There was no formal control over the installation of the
tygon tubing.
To correct these deficiencies, the licensee-indicated the
following corrective actions were taken:
o The RWLI reference legs were relocated to a different
, pressurizer instrument tap to preclude adverse effects
from educator operation.
o Procedure S023-I-3.43 " Refueling Water Level. Indicator
(RWLI) Tubing Installation and Removal" was issued to
formalize the tygon tubing installation. It included QC
check points to independently verify proper installation.
o Proced'ure S023-3-2.8.1 " Refueling Cavity Draining
_ Operations" was revised to caution operators not to change
RCS level until the RWLI is in service.
o Procedure 5023-3-1.8 " Draining the Reactor Coolant System"
'
maintain: Heated Junction;Thermocouples
b , was revisedservice
'(HJTC),in' to: until the RWLI and RCS level tygon
tubing:are in service'and' satisfactorily correlated;
include' level correlation data for HJTC; and include a RCS
-level correlation chart;for RWLI, tygon tubing,
pressurizer. level, and HJTC for various plant elevations
such.as RCS mid-loop. Furthermore, the procedure directs
the operator not to change RCS level or SDC flow prior to
assuring level is six inches above mid-loop, if the RWLI
and tygen tube differ by more than one inch. The
inspector: reviewed several of these changes and found them
to be satisfactory.
(4) Operating Instructions
Abnormal operating instructions lacked sufficient instructions
for mid-loop operations, such as data on the potential for
vortexing at lower RCS levels. The licensee indicated the
following corrective actions were taken:
o Procedure S023-3-1.8 " Draining the Reactor Coolant System"
was revised to include a new SDC flow vortex curve and
- operating window for SDC flow rate for various hot leg
l- levels. The maximum SDC flow was defined to prevent
vortexing and the minimum flow rate was defined to assure
adequate core cooling.
o S023-13-15 " Loss of Shutdown Cooling" was revised to
include steps to check for adequate sub-cooling and
venting the SDC suction.
4
y -- .--:,,--,p- - -
,-w s -,. . . , - , , - , ~ - - - + , - , - , , - - , - - - - - - - , - e , - ~,n w,
_ . - . _._ ~- __ . _ . _. __ _
- *
24
-
- .1
4
The inspector reviewed these corrective actions and found them
to be adequate.
,
'
(5). Operator Training
s,: ,-
, , Operator training on SDC was_found to be weak. For example,
.
the significance of vortexing experienced on the previous shift
! and the method for correcting the abnormal condition was not'
!
.
recognized and communicated to the shift involved in the event. ,
, _ [. >
-The operator also did not recognize the onset of core boiling -
1
/ in 38 minutes after loss of SDC. The operator did not expect < ,'
-
,
,
'
core boiling to take place in less than an hour because-
Technical Specifications allow the shutdown cooling pump.to be.
.' ,
,
'
de-energized for'up to one hour provided 1)'no operations are
., , permitted _that would cause dilution of the reactor' coolant-
'
-
system boron concentration, and 2) core outlet temperature is
f, '
' - -
'
maintained at least 10 degrees F below saturation temp'e'rature. 7
it
, .
However, core outlet temperature was not available~at,the time, 7-
!
"
,
because core exit thermocouples were removed.from service .
'
!
'
and the hot leg temperature reading was not. representative of
the core outlet temperature, once the flow was interrupted.
i~ - The operator's training prior to the event did not focus .
sufficiently on conditions that could lead to a loss of SDC.
!, . '
s For corrective action with regard to training, the licensee
i implemented a major training upgrade to focus on the loss of
. shutdown cooling experience review. This training covered-17
elements and included the factors leading to the event,
, industry statistics for the potential'for a loss of shutdown
cooling, potential consequences of this event, procedures which
address the operation of the shu down e cooling system,
indications available and actions necessary to mitigate the
consequences of this type'of event. _Upon completion of
'
training for a loss of shutdown cooling event, which was held
during requalification training, the operators were required to
take an examination covering the topics discussed. Other
,
corrective actions taken by the licensee included pre-shift
i
briefings and a priority 1 (prior to assuming the shift)
required reading to provide information important to safe
,
operation of the plant.
The inspector reviewed the supporting documentation to verify
'
l that the corrective actions had been implemented. These
L documents included training logs, required reading logs,
j experience review reports, lesson plans, and exam results. All
were found to be complete and appeared to be satisfactory to
5
preclude the occurrence of similar events in the future. In
'
addition, the inspector interviewed several licensed operators '
whose records were on file, and determined that they were
trained on these events as the records indicated.
I
l
'
Based on the above review, the inspector found the licensee's
corrective action to the March 26, 1986 event, as well as the
g2neric loss of SDC issue, to be responsive and adequate. The
[
- , ._, - _. _ . _ _ _ _ - . _ _ - _ . _ - - _ . - . _ . . . _ . . _ _ _ _ . . . _ . _ . _ _ _ _ . . _ , , . _
_
.
' '
- . 25 *
'
r a'.
d : #
previous open items'related to the March 26, 1986 event
'
'50-361/86-11-02, 86-11-03, 86-07-LO are considered closed..
- # '
.. J 'No violations or deviations were identified. ,
,
, s. ,
- - ,
+ l11. Review'of Salt Water Heat Exchanger (HX) Backflush Modification ,
'
. .
.
,
' -*-
,
.As a result of periodic fouling of the salt water.HX, thi licensee
'
.
instituted a permanent design modification to allow backflushing of the ^
1 l -
salt, water heat exchangers.. The inspector reviewed the heat. exchanger-
backflush modification to ensure minimum flow requirements are met-both
'
. .
'+' ,, in the normal flow path and in the backflush mode. ;
' '
i ,
'
. .
.
- The review of the design change package (DCP) 6204.15K. indicated no
' t abnormal design problems in installatica, testing, and final operation of-
'.
.
e the backflush valves and piping. Review-of engineering calculation no.
.
M-0027-10, "SWCS Flow Requirements", indicated that thei system flow
requirements are adequately met in the backflush mode. The flow curves: i
<- appeared to be consistent between-Units 2 and 3 and their calculated
, points were correctly determined from the calculation. The flow
' "
' indicators (2 FI-6399 and 3 FI-6399) and the local temperature indicators
.(TI-6205 and TI-6235) were in SONG's calibration program and were
currently calibrated.
Review of operating instruction S023-2-8, entitled " Saltwater Cooling
System Operation", dated September 20, 1986, determined that the
backflush mode of operation was explained in detail, describing system
initiation, operation, and securing. The procedure provided the
. prerequisites, precautions, and alarm conditions necessary to perform
backflush of the SWC heat exchangers. The procedure verified'that the
. CCW HK heat removal. criteria was achieved by monitoring SWC flow and CCW
outlet temperature-after the backflush was completed.
.
The inspector's review of DCP-6204.1SM indicated satisfactory system
operation in the backflush mode. The procedure was^ concise, clear and
! the curves for determining operability were easy to read and use.
,
i, No violations or deviations were identified.
i 12. . Spurious Engineered Safety Features Actuation Due To The Radiation
Monitoring (RMS) and Toxic Gas Isolation Systems (TGIS)
a. Introduction
i.
Since the operating license was issued:to Unit 2, there had been
a significant number of engineered safety features actuation due to
spurious signals from the RMS and TGIS. These actuations resulted
in the operation of the containment purge isolation system (CPIS),
l control room isolation system (CRIS), control room emergency air
I
cleanup system (CREACUS) and the fuel handling isolation system
(FHIS). Although these spurious actuations did not directly impact
I plant safety, they tended to distract operators from performing
i
their safety functions and could hinder their ability to recognize a
j genuine hazard.- The objective of this inspection was to evaluate
t
jr
i
_ - ,
'
,
~
.ci__ _
, , -- . - . . . _ - . .- - . . . -
l
-
. .
, 26
the' licensee's actions to identify the causes and correct the.
'
deficiencies.
, l*
b. Inspection and Findings-
(1) ' Radiation Monitor System (RMS)
SONGS Units 2 and 3 RMS_ detector inputs to the ESF systems
listed above are single channel, i.e. without coincidence logic
to minimize spurious actuation caused by noise. The licensee
initiated an extensive effort to reduce the electrical noise
- problem-in the RMS. Outside consultants in electromagnetic
,
shielding; technology were contracted to study the problem and
recommend improvements. DCP 6017 was being implemented for the
following design changes:
'
o' The mounting assembly of the detector was redesigned to
assure good ground connections.
o The photo multiplier tubes.were being upgraded.
_
, . .
o The< pre-amp sub-assembly and the high voltage card were
being replaced to enhance the signal.
The licensee is implementing these changes according to the
refueling outage schedule. Eighteen out of the forty four
detectors had been upgraded. The-rest were expected to be
completed within one year. The licensee indicated that no
,
'
spurious actuations had occurred from those upgraded detectors.
However, the operational experience of these upgraded detectors
is still too limited to conclude the overall effectivaness. The
inspector reviewed the licensee's study on the spurious noise
spikes on the RMS, and on DCP 6017,.and found the licensee's
actions to be reasonable and adequate.
The inspector also reviewed samples of the surveillance
procedures and found them to be consistent with the current
'
Technical Specification requirements. The licensee indicated
that the current RMS setpoints were overly conservative low '
and, thus, too close to the background level. The licensee is
- in the process of requesting Technical Specification changes
l^
from NRR to raise the setpoints to more realistic levels.
The licensee's RMS instrumentation group was recently
reorganized to be separated from the Instrumentation and
i Control group as a dedicated group on RMS. The technicians had
on the average eight to nine years of experience. The licensee
indicated these technicians were in the process of meeting the
INPO accreditation requirement for technicians. I
Based on these reviews, the inspector found the licensee to be
I
responsive in resolving the problem of spurious actuations of
i ESF equipment caused by noise in the RMS. The hardware
upgrades appeared to be prudent, though limited operating
,
r ,- . , ---_n ,. - , , , , , - , , , - ~ , . , , , - - - . , ~ , . . - - .
-
< - - . . - - -- - . .
. _
- *
. 27
i
i
history was not sufficiently long to indicate'the
effectiveness.
(2) Toxic Cas Isolation System (TGIS)
In response to the frequent spurious actuation of the TGIS, the
. licensee chartered a task force to determine-the cause and
correct the problem. The following modifications were
implemented as a result of the task force recommendation:
.o Protected the front instrument panel against tampering and
electromagnetic interference.
o Rerouted personnel and equipment traffic in the vicinity
of the unit.
o Added a time delay relay to make the system less
susceptible to noise spikes.
o Relocated the air pumps outside the instrument cabinet to
reduce vibration and interference.
o Added a second gas supply bottle which is checked each
shift.
o Added another cooling fan for the instrument cabinet.
o- .Added diagnostic display panel to show the cause of
actuation.
o Added system trouble alarm in the Control Room.
'
- The inspector reviewed samples of the change packages to verify
proper design change reviews were made. The inspector also
verified these changes had been implemented by a physical
walkdown of the system. In addition, samples of the
surveillance records for' TGIS were reviewed for Technical
Specification setpoint and response time compliance. They were
found to be satisfactory. From the LERs submitted by the
licensee, it appeared that the number of spurious TGIS
actuations were reduced significantly since the implementation
of the modifications'.
Based on the above review, the inspector found the licensee's
corrective action to be responsive and effective.
There were no violations or deviations identified.
13. Static Inverters
a. Background Information
The San Onofre Units 2 and 3 electrical design was implemented with
nine static inverters per unit. In addition to these, one static
._
. _ _ _ _ . _.
. . . . . .- - . _ .
,-
' * *
. 28
inverter is shared between the two units. Six inverters per unit
are classified as Class 1E and the remaining ones are classified as
. non-Class 1E. Four of the six Class 1E were manufactured by.Cyberex
and the remaining two were manufactured by Elgar. The non-Class 1E
4 inverters were manufactured by Solid State Controls. The rated ,
capacity of the inverters range from 5 KVa to 150 KVa with some
providing single phase outputs and others providing three phase
outputs. Among the electrical loads the inverters power are safety.
j< related electrical equipment and indicators; critical functions
monitoring system; piant monitoring system computer; health physics
computer; backup core operating limit supervisory system and-
security computers; and non-Class IE instrumentation.
I The specific procedures reviewed by the inspector for inverters -
,
, were:
,
L ,
- S0123-II-11.163, " Inverter Inspection and Cleaning"
,
- S023-6-17, " Class IE 120 Vac Vital Bus Power Supply System
Operation" ,
'
c ..
.
.
- S023-II-11.177, "YO12 Inverter and Static Switch
Test / Calibration"
5
- 'S023-II-11.185, " Vital Bus Inverter Test / Calibration"
,
. The inspection of the inverters consisted of reviewing system-
descriptions and their electrical diagrams, reviewing instrument and -
test procedures and examining operating instructions and maintenance
records. The physical installation of the inverters and the loading
- > increases which occurred was also inspected. Several of the station ,
engineers, test technicians and Operations personnel cognizant of
1 inverters were also interviewed. ,
'
b. Inspection Findings ,
l As a result of this inspection, the following items were noted by
_
,
the inspector:
o Manual Transfer Switch
The four Class 1E Cyberex inverters are equipped with a manual
! transfer switch. One. function of this switch is to transfer
, the power source for a vital bus from the inverter output to an
alternate source.- It was noted that the switch handle provided
to operate the switch is about the size of a human hand and is
,
physically located near the top of the inverter cabinet in
- excess of six feet from the floor. The inspector was informed
- by several licensee personnel who had operated the switch that
it was difficult to operate and that hang-ups in the
mid-position had occurred. This apparently is due to the
physical location of the switch relative to the floor and that
a considerable amount of force is required to operate the
! switch smoothly. The inspector expressed concern about smooth
. - _. -_. _ . ._ - - - ~ . . . . . _ _ _ _ _ _ _ - . . . _ . . _ _ _ _ _ _ _ _ _ _ - - - . - _ _
-. . .
. >
} . . l. ~ 39
,-
I
'
operation of the switch, particularly by physically smaller l
operators following an inverter loss at a. time when power to
the vital bus is needed. The licensee's. engineering staff.
stated that the switch was hard to. rotate and is operated
infrequently, however modification of the switch is not
justifiable. The inspector also noted that depending on the
status of the inverter output circuitry.in conjunction with the
make-before-break feature of the switch, a switch hangup in the -;
,
- ' mid-position may result in the loss of the alternate power. ,
'.
source. 'The inspector noted that the licensee did not consider
- '
the difficulty in switch operation to be a safety concern.,
However, to offset inherent design weaknesses and infrequent
operation of this switch, continued training (to include. i
' '
operator aids) which demonarrr.ces operator proficiency of this ,
'
< '
switch was suggested by the inspector.
>
<
o Monitoring of Temperatures Within Inverter Cabinets
- +
. :
'
Each of the four Class IE Cyberex inverters attendant to Unit 3
"," are located in a separate room. The four rooms are adjacent to
,
,i , each other on one side of a hallway. A battery charger and a 3
L ,
. power distribution panel'are also located in each of the four ,
s !
rooms. Ventilation to each of these rooms is provided.by, ,
'
overhead duct work. The thermostat which controls the air.
' ~
< temperature within these rooms is located within a switchgear
(
room. This switchgear room is. located across from the vital ,
'.
.s!' inverter rooms on the other side of the hallway. Access to the
switchgear room and the vital inverter rooms are provided by 3
,
doors which-are normally cloted. The inspector observed a ,
noticeable increase in the air temperature between the hallway
'
and inside the vital inverter rooms. In view of this
observation and the thermostat arrangement for the ventilation
system, the inspector expressed concern regarding verification ,
that temperatures within each of the four inverter cabinets
were within the specifications provided by.the manufacturer.
The licensee informed the inspector that design calculations
showed that air temperatures within the switchgear room should
be greater than air temperatures within the vital inverter
rooms. In addition, the licensee informed the inspector that
air temperatures over.the top of one of the inverter cabinets
had been measured; however, a record of this measurement was
not avaitcble at the site. The inspector expressed concern
'
ebout verifying that internal cabinet temperatures for each
Class 1E inverter was within the specifications provided by the
manufacturer. This item will remain open pending additional
inspector review (361/86-25-07). *
o Loading Increases
The inspector noted that Class IE inverter 3Y002 had a total
calculated connected load of 18.01 KVA in the 1977 design. The
present calculated connected load has increased to 19.8 KVA.
,
, - .. - . . - - _ .
- a '
.; .
- 30
.
- A review oflthe loads on other vital buses showed similar
increases in load. While it is recognized that calculated
connected load and actual running load are different, the
inspector expressed concern about additional loading increases.
The inspector also expressed concern about the lack of
definition of the inductive portion of the load and its
frequency of operation. The licensee indicated that future ~
load additions would be monitored.
o Design Change
,
The inspector noted that a design modification resulting in a
' load increase involved the use of a 15 ampere fused circuit
protective device assembly to feed two branch circuits. This
.was noted in inverter supplie'd' pan'ls e 2Y002 circuit 10, and
.3Y002 circuit 10, which power the Engineered Safety Features
Actuation Signal (ESFAS) Auxiliary Relay cabinets. The
'
, inspector' noted that this is not in accordance with the
,
ENational Electric Code (NIC) and Underwriter Laboratory (UL)
- Standard. It~was also noted that as presently installed, that
the= termination connector which supplies the two branch'
,
circuits is not in accordance with the UL standard on the use
,
of electrical materials. Further it was noted that Section
- 210-22(c) and '384-16(c) of the NEC ruled that the total load on
s .any overcurrent device located in a panelboard shall not exceed
1 3 u80 percent of its rating. UL has a rule that any fuse or
,
' breaker not marked for continuous load must have its load
limited to 80' percent of its rating. The inspector noted that
each 'of the subject ' circuits have two loads with only a 1200 VA
load c'onnected in parallel with a 480 VA load. This adds up to
a total load of 14 ampere,'and thus exceed the 12 ampere
. maximum load that can be placed on.the circuits in question.
The licensee conferred with its consultant from Bechtel and
i informed the' inspector that the Bechte1~ design was based on an ,
I analysis that the actual load would be less than 1200 and 480 -
VA respectively. The inspector found that these 1200 and 480
VA load ratings were in fact obtained and factored in the
original design and concluded that these load ratings should
have been adhered to in subsequent design modification. This
item will.be further reviewed during a future inspection
,
-(361/86-25-08).
o. Formal Training for Test Technicians
The inspector noted that in the past little or no formal
training from inverter manufacturers has been provided to test
technicians who service inverters. This has been recognized by
the licensee and some training of this type is planned. The
inspector emphasized the need for a well structured and well
,
defined formal training program which included inverter
,
manufacturers. The licensee appeared receptive to this
Concern.
,
o Cleanliness of Class IE Inverter Rooms
t
, , , ,. , -- - - ,-.---- ,.-- n,- . , , , - - - , .-. ... . ---,,--- -
- - , _ . . - - -
. - , - . . . - - - . - . .- ,,,
. . . ,
' ,
. - - - -
_m._ r. -_ ._ . . _ . _ _ _ _ - _
'
- ,
-
,
. ;
, <
,
> . .
,x .
. . *-- * > '
31
-
- ?
.The inspector noted that all class 1E inverter rooms and
attendant equipment were clean and neatly arranged. In' .
addition, the inspector visually inspected the dirt and dust
filters mounted on the front of.the Class 2E Cyberex inverters.
These filters were also clean.
-
No violations or deviations were identified.
,
14'. Station Batteries
s
-
's. Inspection Activities ,
The inspector reviewed procedures which. implemented Technical
Specification required surveillances for the Class.1E 125 volt
- safety related batteries. The inspector examined the maintenance
procedures provided for Class 1E and Non-Class 1E batteries. The
- . inspection activities also included discussions with licensee
personnel concerning initial and current problems with batteries and
battery chargers. The physical installations of Class 1E and
'
Non-Class IE batteries were inspected. -The inspector examined
'
surveillance records for. required performance and service testing of
, Class IE batteries..
h . ~
The inspector reviewed the following procedures used for station
- - batteries
- ;
i
'
-. 5023-I-2.12 " Weekly Inspection of IE Batteries".
- S023-I-2.13, " Quarterly Inspection of Batteries". !
,
' - S023-I-2.14,- " Refueling Interval Physical Inspection of
4 Batteries".'
- S023-I-2.15, " Refueling Interval Battery Service Test".
- S023-I-2.16, " Battery Performance".
-
.S023-I-4.78, " Spare Battery Charger Installation'and
Testing". i
' -
S023-I-9.14, " Battery Charger Inspection, Cleaning and
. Testing". ? '
'
r-
.S023-I-9.47, " Battery Weekly Inspection of Non-1E Station
-Batteries". .
-
S023-I-9.48, " Quarterly Inspection of Non-1E Station
Batteries".> !
-
SO23-I-9.49, " Annual Inspection of Non-1E Station' Batteries".
'
The inspector.also reviewed surveillance records for performance and
t service,te' sting. The performance tests in February 198? (batteries
38008 and 38010) and:the tests from March'1983 (batteries 38007 and
- 38009) ~ were reviewed.: All had battery capacities in excess of 100%
1
.- (minimum' acceptance is;90%). The service tests on the same
batteries done about the same time also had acceptable results. ;
b. - Battery Cell Problems During Startup'
Theinspectorwas[informedbylicenseepersonnelthatpriorto
initial startup a problem was experienced with a number of
_
,
b
_s-
, . _ _ _ . _ . . ._ _ _ _ - . _ _. _
v
'- *
,< 32
.
- Class 1E battery cells. The problem as described-involved
stratification.of the electrolyte contained in the cells. This
+
problem was attributed to the manner in which these cells were
i stored at the site prior to initial startup. For Unit 2 Class 1E
- batteries 28007 and 28008, all of the initial supplied Exide GN-13
i
'
cells were replaced with Exide CN-15. cells. Licensee personnel
indicated that following initial startup, recurrence of this problem
had not occurred.
1
I- c. Battery Performance Testing In Lieu of Service Testing
} Once per sixty months, the Technical Specification permits battery
'
performance testing in lieu of service testing. Licensee personnel
. indicated that this provision was used for battery testing in May of
4
this year. This being the case, the inspector expressed concern
. that the more demanding ninety minute service test loading profile
may indicate weaknesses in the battery installation which
,
performance testing would not. For batteries which this provision
'-
had been used, the inspector suggested that close attention be given
to the results of the next service test. The licensee was receptive
to this concern and suggestion.
, d. Battery Service Test Loading Profiles and Surveillance Records ,
<.
{ k
-
Design documents providing information on battery service test
.
2
loading profiles were examined. These documents revised.the 90
! minute service test loading profiles for Class 1E. batteries 28007,
'
28008, 28009 and 28010. These documents showed that service testing
t- loading profiles for San Onofre 2 and 3 were established by
, e calculations. No concerns were identified associated with these
documents.
' * '
-
The inspector reviewed surveillance records for required battery "
i service and performance testing. These records showe'd
i
'
i ' that the 1982 service and performance test rasults for Class-1E :
'
batteries 38007,.38008, 38009 and 38010 were satisfactory. However, +
.', in view of time constraints and the manner in which the records are r
1 arranged at the site, surveillance records providing recent test s
'
i '
_
results were not reviewed. This being the case, recent test results
"
'.
for battery service and performance testing and record availability
,
will be examined in a future inspection (361/86-25-09). .
.
e. Battery Maintenance and Surveillance Program ,
'
The inspector reviewed the battery maintenance and surveillance -
'
procedures. These procedures satisfied the requirements of the
l Technical Specifications, Regulatory Guide 1.129 "Maintenance,
! Testing and Replacement of Large Load Storage Batteries for Nuclear
Power Plants", IEEE Standard 43C-1980 "IEEE Recommended Practice for
Maintenance, Testing and Replacement of Large Load Storage Batteries !
s for Generating Stations and Substations" and manufacturer's
technical instructions. In view of this, the inspector concluded
- that an adequate surveillance program exists for the San Onofre
- Units 2 and 3 batteries.
1
, , , y- m, ,,_ - - _ , . _ _ _ , . . . . , . . . ~ , _ . _ , - - - ~ , - . , . - - - - . ~ - , , , m . . . , .- . . . _ . . . . - _ , _ - . .
,
,
- 3'
33 ,
f. Inspection of Battery Installations
The inspector noted that Class 1E and Non-Class IE battery cells
were clean and intercell connectors were free of particles and
- corrosion. The interceli connectors to battery cell past
connections were well lubricated. The battery rooms were well
. lighted and clear of extraneous equipment. .However, the inspector
'noted that cleanliness of battery room floors and battery. cell
, retaining racks could be improved. The licensee was receptive to
this observation.
No violations or deviations were identified.
'
15. Control Air System Moisture Buildup
-Industry experience has shown that moisture buildup in air lines can
cause failure of air operated valves, particularly during periods of high
air demand. In July 15, 1981 the Palisades power plant was operating in
the shutdown cooling mode with the primary system level drained-near the
hot-leg centerline for replacement of a reactor coolant pump (RCP) seal
package, when a loss of shutdown cooling capability occurred because of
isolation of the single shutdown heat exchanger outlet control" valve.
'
The outlet valve malfunctioned because of water accumulation in~its
,
control air system.
t
The purpose of this inspection was to determine whether SONGS has'
implemented a program that effectively prevents and/or identifies and
' 4
'
i
' corrects moisture accumulation in control air lines. The inspector
m reviewed operating instruction S023-1-1, " Instrument and Service Air
.N System", dated October 2, 1983. This operating instruction described the
t
normal and abnormal operation of the instrument and service air systems,
+
provided procedures for removing various system components from service,
1- . and provided a detailed procedure for operating the instrument air-
dryers. This procedure, in paragraph 5.5 " Blowdown of Air System
'
Drains", provided once-a-shift blowdown of the three air receivers-(E090,
' '
E091 and_E092) and the control air supply line filters (C-001, C-002 and
C-003) by cracking open the drain valve and then reclosing the drain
"
valve when moisture free air issues from the drain. The inspector ,
witnessed this once-a-shift blowdown. In addition, every refueling, a
'
,
check-off list is used to verify that no water has accumulated in the
instrument air lines.
The inspector interviewed air and emergency core cooling system (ECCS)
engineers, and craft, including supervision, to determine if moisture
control problems have been experienced and any corrective actions taken.
The results of the interviews indicated that only in the startup phase of
.the control air system was there a moisture buildup problem. No
equipment history records indicated a problem with moisture accumulation
in the control air. The inspector reviewed the system design of the
control air system, which has two refrigerated type air heaters, and
once-a-shift. blowdown of the air receivers and the control air supply
line filters. In addition, system design features include the use of
copper lines, air compressors that are non-lubricated, air filters that
. - - . - __
. . _ _ _ _ _ . _ _ -. ..
'
<
../t
- *
c .._ 34
-
are designed to remove particles greater than 5 microns, and air operated
'
' valves upstream that use a filter.
LThe inspection of the control air system revealed no present day problems.
j with moisture accumulation. The system is blown down daily'and complete
f
.
' low point drain blowdown is performed each refueling.
L No violations or deviations were identified in this area.
16. Post Trip Reviews
<
The. inspector'_ reviewed a total of six post trip review reports _
The
'
associated with the most recent reactor trips at_ Units 2 and 3.
purpose of the review was to ensure that: 1) anomalous plant equipment
responses were identified, and corrective actions implemented as .
' '
required. 2) the licensee's evaluation was sufficiently thorough to
, identify the cause of the reactor trip, and potentially generic concerns
were addressed at the other two operating units.- The post trip review
packages evaluated were for the following reactor trip dates:
'
UNIT 2 UNIT 3
m July 14,1986 April 13, 1986
-
August 12, 1986 July 26, 1986
September 13, 1986 September 4, 1986
The inspector found the licensee's reviews to be adequate. The inspector
{ suggested the licensee consider cont' acting the Palo Verde Nuclear
Generating Station (PVNGS) to take advantage of recent program
'
improvements which had been implemented at that. facility, regarding post
- trip review evaluations. ;The licensee: stated that a representative would
'
contact.PVNGS and request that a sample trip report be sent to SONGS for
j_ their review.
No violations;or. deviations were identified.
17. Temperary Facility ~ Modifications- ',
- When~a plant structure, system, or component used to perform its design
.
'-
function, is in an-other-than-designed condition, a Temporary Facility
i Modification (TFM) would normally be expected to be initiated. The TFMs
- were inspected for several of the systems reviewed during this
- inspection.
The use of TFMs was governed by procedure S0123-V-5.10. " Temporary
Facility Modification", and was supported by other procedures including
desk Procedure 3.1-02 for " Configuration control Document Maintenance"
i
and S023-0-16, " Temporary. Facility Modification Control". The number of
- open TFMs at SONGS 2 and 3 was 46, 2 from 1983, 1 from 1984, 9 from 1985
i and 34 in 1986. For Units 2 and 3, their were no current TFMs for the
CCW, SWC, and AFW systems. The timeliness of the TEHs was assessed, and
l the average age of the temporary modifications was determined to be
'
approximately 6 months. The median age of the TFMs was found to be
approximately 3 months. Safety evaluations (50.59) associated with a
!
i
f
a
..w -- - ., .- -,-,,., _. -_.__, y- ..-..-.,-,.w-,,,,.- . , , . ,,,..w._,.-,_,.._ . - - , _ . _ _ - . - - - _ . - , - - . , _ . , _ - - - -
-
,-,,-,-,---.,-_m.
. . . - ,
,
.
'" *
. 35 +
sampling'of.TFMs were' considered adequate. The implementing procedure
required reevaluation'of existing'TFMs every three months.
No violations' or deviations were identified.
18. Calibration of Pressurizer Pressure and Level Transmitters
Actuation of the Engineered Safety Feature Actuation System (ESFAS)
instrumentation channels depends upon accurate and correctly calibrated
instruments. For this reason, the pressurizer level and pressure
transmitters need special attention in their calibration. The
pressurizer level transmitters need to be calibrated from the lower tap
to the upper tap for the calibration range per the Combustion Engineering
(CE) data sheet. The pressurizer pressure transmitters need to have a
height of water (head) correction removed from its input bistable,
otherwise incorrect ESFAS setpoint actuation occurs.
The inspector reviewed SONG's engineering calculation no. M107-12-3 for
the suppression and span adjustment for pressurizer level transmitters.
This Bechtel calculation had a six level review. The pressure
transmitter head correction calibration was determined by the Instrument
and Control (I&C) group. Both calculations were considered concise,
clear and correct based upon the inspector's review.
No violations or deviations were identified.
'
19. Testing Time Delay Relays
Time delay relays provide the proper sequencing of emergency core cooling
systems (ECCS) and safety injection (SI) activation systems under
accident conditions. Since these relays provide for the starting of the
ECCS and SI systems, improper or non-existent testing of these relays can
compromise the operability of the ECCS equipment under accident
conditions.
The Trojan nuclear power plant in September, 1984 experienced a safety
injection actuation signal in which the diesel driven auxiliary feedwater
(AFW) pump failed to start. Failure of the diesel driven AFW pump to
start was attributed to the absence of procedures for the periodic
calibration of time delay relays associated with permissive and
protective functions of the diesel driven AFW pump.
The above incident revealed that failure of testing time delays relays
can compromise the operability of the ECCS equipment. The purpose of
this inspection was to determine if SONG's program of testing time delay
relays ensures ECCS equipment operability.
The inspector reviewed the following procedures: j
o Operating Instruction S02-II-11.1 - Unit 2 Loss of Voltage (LOV) and
Sequencing Relay and Circuit Test,. dated August 1, 1984.
,
o Operating Instruction S03-II-11.1 - Unit 3 Loss of Voltage (LOV) and !
Sequencing Relay and Circuit Test, dated August 1, 1984.
,
I
i
l
1
'
, _ . , - - - . _ , - - . _ _
_____. _ __- _ _ - . - .
-
,
. .
v. * * 36
o Operating Instruction S023-3-3.12 - Integrated ESF System Refueling
Test, dated May 9, 1986.
The above procedures provided the mechanism to test the Unit 2 and
Unit 3 individual time delay relays and conduct an integrated engineered
safety feature (ESP) test every 18 months. The inspector's review
determined that SONG's program of testing time delay relays was
acceptable. Tests were included on the surveillance' schedule routinely,
and the recorder wires were permanently run throughout Units 2 and 3 for
easy installation into the various time delay relays.
No violations or deviations were identified.
20. Plant Personnel Interviews
In addition to routine discussion's with licensee individuals to gain
information as part of the normal inspection process, informal interview
were held between the inspector and 28 plant personnel working within
various disciplines having responsibility for safety-related matters.
Discussions were held with both craft and engineering personnel. The
' purpose of the interviews were to determine whether; 1) workers felt they
could bring safety concerns to their supervision, 2) whether schedular
pressures have compromised plant safety and 3) whether the Quality
Assurance Department was effective in carrying out its responsibilities.
No concerns or responses, which could indicate a potential problem, were
brought to the inspector's attention.
No violations or deviations were identified.
21. Exit Meeting
The results of this inspection were discussed with the licensee on
October 3, 1986. The inspector summarized the scope of the inspection
and findings as described in this report.
,
t