ML20203D474
| ML20203D474 | |
| Person / Time | |
|---|---|
| Site: | Crystal River  |
| Issue date: | 04/14/1986 |
| From: | Arildsen J, Debs B, Kleinsorge W, Mccoy F NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I), NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20203D468 | List: |
| References | |
| 50-302-86-04, 50-302-86-4, NUDOCS 8604220120 | |
| Download: ML20203D474 (11) | |
See also: IR 05000302/1986004
Text
a
UNITED STATES
/pn :t rg}o,
,
NUCLEAR REGULATORY COMMISSION
- '
R EGloN 18
i#i
j
j
to1 MARIETTA STREET.N W.
,,
ATLANTA, GEORGI A 30323
A
g
,,, . , p
.....-
Report No.
50-302/86-04
Licensee:
Florida Power Corocration
3201 34th Street, South
St. Petersburg, FL 33733
Docket No.:
50-302
License No..
OPR-72
Facility Name:
Crystal River 3
Inspection Concucted: January 3-7 1986
V[M[%
Inspectors:
F. R. McCo
'W)
Date Signed
WNb
/
W. Kleinsorge
Ddte'IS i gn ed
'
Dbra
sk
p J. A. Arildsen
Z-/
Ddte/ Signed
-
'
' '
Approved by
-
-
Brian I. Debs. Acting Section Chief
Oate Signed
Operational Programs Section
Division of Reactor Safety
SUMMARY
Scope: This soecial, unannounced inspection entailed 139 inspector-hours on site
in the area of event followup. The event under assessment was the failure of
Reactor Coolant Pumo (RCP) 1A and the suosecuent reactor trip on January 1, 1986.
Areas under review included: an inspector review of recorded values of selected
plant parameters associated with the event; an inspector review of the licensee's
Post Trip Review; an inspector review of selected plant enemistry data for
indication of fuel element failure; and an inspector review of RCP 1A's
surveillance, preventative naintenance, corrective maintenance and predictive
maintenance programs.
Results:
One violation was ic3ntified - three examples of failure to follow
written and approvec acnormal procedure (AP) 580, Reactor Protection System
Actuation during recover from a reactor trip.
(see paragraph 5.c.)
8604220120 860416
ADOCK 05000302
G
.
-
.
.
_ _
_
.-
.
.
.
REPORT DETAILS
1.
Persons Contacted
Licensee Employees
- P. F. McKee, Plant Manager
R. Wittman, Operations Superintendent
"C. F. Bennett, Operations Planning Supervis:-
- B. J. Hickle, Operations Manager
- J.
A. Frijouf, Compliance Specialist
- R. L. Thompson, Nuclear Plant Engineering Supt.
- V. R. Roppel, Plant Engineering anc Technical Services Mgr.
- D. B. Eggleston, Nuclear Safety Suoervisor
- L. W. Moffatt, Nuclear Safety Supervisor
- P. D. Breedlove, Recoras Management Supervisor
- K. F. Lancaster, Site QA Manager
- S. C. Powell, Senior Nuclear Licensing Engineer
- A. E. Friend, Nuclear Staff Engineer
G. Halnen, Nuclear Staff Engineer
Other licensee employees contacted included engineers,
technicians,
operators, mechanics, and office personnel.
Other Organizations
Precision Mechanical Analysis, Inc.
Babcock & Wilcox
NRC Resident Inspectors
- T.
K. Setka, Senior Resident
. J. E. Tedrow, Resident
- Attended exit interview
,
i
2.
Exit Interview
i
The inspection scope and findings were summarized on January 10, 1986, with
those persons indicated in paragraph 1 above. The inspectors described the
areas inspected and discussed in detail the inspection findings listed
below.
The licensee acknowledged ~the findings.
.
The licensee did identify as proprietary various materials and information
provided to the inspectors during this inspection, however, no proprietary
information is included in this report.
a
.
.
.
.
.
.
.
.
. m
.
.
.
.
.
. .
.
.
.
.
.
6-
.. .
.
2
3.
Licensee Action on Previous Enforcement Matters
This subject was not addressed in the inspection.
4.
Unresolved Items
Unresolved items are matters about which more information is required to
determine whether they are acceptable or may involve violations or devia-
tions. One unresolved item was identified in the area of reactor coolant
system pump vibration surveillance during this inspection, and is discussed
in paragraph 5.b.
5.
Failure of RCP 1A and the Subsequent Reactor Trip on Power to Flow Channels
"B" and "D".
On January 1, 1986, an event was reported at Crystal River Unit 3, under 10 CFR 50.72. The event included the failure -of RCP 1A and the subsequent
reactor trip. On January 3-7, 1986, an inspection team composed of three
individuals from Region II visited the Crystal River Nuclear Plant in order
to evaluated this event.
a.
Sequence of Events
Crystal River Unit 3 was operating at approximately 93P. power with no
special
operational
evolutions
in progress.
Shift relief had
commenced, and the oncoming shift supervisor and assistant shift
supervisor had assumed the shift.
At approximately 11:37 p.m.
the
control room operators noted the sound of " clicking" relays.
The
assistant shift supervisor noted a loose parts monitor alarm and a
RCP vibration alarm.
He tnen observed an RC low flow alarm, RC low
-
pressure alarm, and low amperage indication (approximately the
no-load value) on RCP 1A.
The operators noted control rods going
in and cbserved proper indications for a reactor trip.
The operators
commenced the Immediate Actions of AP 580, Reactor Protection System
Actuation, from memory.
During the event, pressurizer level decreased to 46 inches and was
restored to greater than 50 inches by opening, more fully, charging
injection valve MUV-24.
Additionally, Reactor Protection System low
pressure channel B tripped at 47 seconds into the event. There was no
obvious indication for this occurrence.
,
Approximately 2 minutes into the event, the Shif t Supervisor, having
noted the failure of RCP 1A's first seal and a degradation of the
.
second seal, was concerned about a possible LOCA on RCP 1A's seal
r
packing failure.
He directed the securing of RCP 1A. This action was
noted to be in compliance with AP 502, Annunciator Response, which
requires securing an RCP upon verifying high vibration annunciation.
-
-
n
..-
,
.
,
3
During the transient, Emergency Feedwater actuated automatically due to
steam generator pressure perturbations associated with the turbine trip
that accompanies the reactor trip. As steam generator level increased,
Emergency Feedwater Injection valve EFV-57 did not give complete
indication of closure.
Consequently, the operator placed the valve
controller in the " Hand" position in order to manually close the valve.
When the valve controller was placed in the " Hand" position, the valve
malfunctioned and opened fully.
The operator then manually _ shut the'
valve.
Also during the transient, one secondary safety valve, MSV-41, stuck
partially open and did not reseat; and two secondary safety valves,
MSV-35 and MSV-36, chattered upon reseating.
Valve MSV-41 was
subsecuently reseated afte- lowering of steam header pressure and
manually cycling the valve, and valves MSV-35 and MSV-36 ceased
chattering upon lowering of steam header pressure.
During the conduct of this . inspection, actions were initiated by the
licensee to depressuri:e, cooldown and lower reactor water level
consistent with required plant conditions for pump repair operations.
Additionally, actions were initiated to troubleshoot pump damage
and repair the damaged pump.
The inspectors consider that the
licensee's current and planned recovery operations reflected a
controlled avolution with adequate management attention and direction.
Actions taken appeared to be properly planned and deliberately imple-
mented.
The licensee's engineering ef fort associated with recovery
from this event is considered noteworthy.
b.
Review of Transient Data, Chemistry Data, and Licensee Post Trip
Review.
A review of SPDS recall data associated with this transient did not
reflect any conditions outside of safety limits or safety analysis.
'
Additionally, there was no indication of loss of subcooling during the
A review of the licensee's Post Trio Review and Plant Review Committee
Restart Review reflected that the licensee's assessment of the event
was accurate.
The licensee's Post Trip Review and Plant Review Committee Restart
Review indicated that the following malfunctions occurred which would
require corrective action prior to restart:
RCP-1A motor and impellor separated which initiated the transient.
-
RCP-1A will be repaired and cause determined prior to restart.
EFV-57 opened when placed in manual control mode. Operators then
-
successfully shut the valve. This deficiency should be corrected
prior to restart.
!
-
.
.
.
- .
q
. *
-
.
!
4
!
!
MSIV's -35, 36, and 41 did not fully reseat following the reactor
-
trip until main steam header pressure was lowered, and MSV-41 was
cycled manually.
These valves should be reset prior to restart.
Additionally, the licensee's review indicated that the following
malfunctions occurred which do not require correction prior to restart:
-
Condensate Pump (CDP)-1A, 18 tripped following the reactor trip
(apparently on the spiking Dearator level transient that was
identified on 12/3/85 reactor trip).
One condensate pump was
-successfully restarted. This deficiency is the subject of further
engineering study and does not need to be resolved prior to
restart.
Reactor coolant pump 1A suction's narrow range temperature
-
indication, RC-5A-TT1, spiked high following the reactor trip.
t
!
Redundant instrumentation is available as wide range indication,
RC-5A-TT2.
Licensee investigation has revealed an inadequate
connection at the RTD which is under a work request for repair.
Although the licensee's initial evaluation determined that
correction of this malfunction was not equired prior to restart,
subsecuent investigation has led the licensee to list this item
for repair prior to restart.
The licensee also indicated t m the Qaactor Protection System a flux to
flux to flow set point needs to be verified with detailed calculations
based on data derived from this event, and that Surveillance Procedure
i
(Sp) 110, Reactor Protection System Functional Testing, be performed
for this trip function prior to restart.
-
The insoectors noted that Reactor Protection System low pressure
channel B tripped at approximately 45 seconds into the event. The Post
Trip Review recognized the trip; yet, did not address the lack of
obvious indication that the plant pressure attained a point which would
have perpetuated the low pressure trip.
The inspectors were unable
to find any licensee evaluation to determine the specific cause of this
occurrence, and their review of selected documentation including the
Reactor Trip Report and Plant Review Committee Restart Review failed
l
to indicate that such an evaluation is to be performed. The licensee
stated that Reactor Protection System calibration of the low pressure
trip instrumentation and other trip functions is scheduled for
'
accomplishment during this outage.
The inspectors consider such
action to be warranted as a result this occurrence, and also consider
that a detailed licensee evaluation should be conducted to determine the
specific cause of this occurrence.
Resolution of this concern is
identified as an inspector followup item (302-86-04-01).
The inspectors also noted that the cause of failure of RCP-1A may
necessitate the performance of further evaluation of the other RCP's.
During interviews with licensee personnel, it was apparent that this
issue has been recognized; but it was not specifically addressed in the
documentation reviewed by the inspectors.
Resolution of this ' concern
is identified as an inspector followup item (302-86-04-02).
. -
-
- - -
-
-
-
-
-
-
-
--
--
^
,
i
-
.
5
A review of selected chemistry data associated with this event
indicates an increase in Dose Equivalent I 281 Activity of approximately
346% over pretrip levels and an increase in the I */I
5
Ratio by a
factor of 38.5 to a level of 5.89 (as of last sample January 4,1986).
This ratio is in excess of the criteria 0-0.6 specified in Enclosure 2
of the Chemrad Report, CH-409, Rev. 05. Interviews with the licensee's
chemistry staff indicated that a licensee evaluation of this and other
chemistry data for the event had been conducted, and that fuel failure,
as a result of the event, is not evident from this data. A review of
selected past trip data confirmed that, though higher, this data is not
significantly out of proportion with that from past react Jr trips.
Further review of this data by the NRC Region II Radiologic,1 Effluents
and Chemistry Section confirms the licensee's conclusions that fuel
failure is not evident from this data.
c.
Failure to Follow Written Approved Procedures
The inspectors interviewed onshif t operaters in evaluation of the
event.
In the performance of AP-530, Reactor Protection System
Actuation. Immediate Actions and some Followup Actions were performed
from memory.
However, during the course of this performance, three
_
examples of failure to follow written and approved procedures were
noted:
(1)
Immediate Action 1 of AP-580 requires the operators to ensure
group 1-7 rods are inserted by depressing the reactor trip push
button and by observing that the " TRIP CONF" light is lit on the
diamond panel. The operators failed to depress the reactor trip
push button.
(2)
Immediate Action 6 of AP-580 requires the operators to ensure
pressurizer level greater than 50 inches and specifies four
remedial actions to be taken in the event level is less than 50
incnes.
These actions are in order of 1 to 4 as follows:
1.
open suction from BWST
2.
start 2nd MVP
3.
open MOV-24
4.
close MUV-51
The operators only performed remedial action 3, open MUV-24. This
action was sufficient to recover pressurizer level, however, it
was not fully in accordance with the procedure since the other
three actions were not performed.
(3)
Followup Action 2 of AP-580 recuires that makeup tank level be
maintained between 55 inches and 86 inches with a detailed action
to be performed if makeup tank level is less than or equal to 55
inches. This detailed action is to open suction from the Borated
Water Storage Tank (BWST) to the running makeup pump (s). When
makeup tank level decreased to less than 55 inches, the operators
--
.
--
- -
r1
~~
' s .
i
A
.
!
j
6
did not perform this detailed action, but instead, restored makeup
tank level by filling the makeup tank from the reactor coolant
<
bleed tank.
The inspectors questioned the operators about the
boron concentration of water in the RC bleed tank and were
informed by the operators that the concentration was approximately
200 parts per million greater than the concentration which was in
the reactor coolant system at that time.
!
It was noted that, programmatically, the licensee's operators and
operations management consider that Remedial Actions associated
with fulfilling Immediate Actions and Details associated with
accomplishing Followup Actions are not required to be followed
vercatim as prescribed in the Abnormal Procedure provided that
actions taken ca- restore the requisite parameters.
Inspector
review of administrative procedures in this regard indicate that
authority for operators to deviate from prescribed Remedial or
Detail Actions does not exist.
'
These examples of failure to follow procedure constitute a viola-
tion (302-86-04-03).
.
,.
l
Inspector evaluation of ooerator actions during the transient of
this event noted that in responding to acnormal events, operators
!
take all tne Immediate Actions and some of the Followup Actions
l
based on operator training and experience.
The procedure is
referred to af ter accomplishment of these actions to verify that
all requisite actions were properly taken.
This practice, in
j
general, is considered acceptable if the operators are properly
-
trained to respond in this fashion.
If it is intended that all
Immediate and some Followup Actions taken in the early phases of
j
l
an actual event be performed without use of open procedure to
facilitate a timely response, and if it is intended that the
orocedure is to be subsecuently used to verify all requisite
actions were properly taken, the operators should respond in a
similar manner during simulator examinations conducted by either
i
the licensee or the NRC. Resolution of this concern is identified
as an inspector followup item (302-86-04-04).
d.
Review of RCP Operational and Maintenance Data for Prior Indication of
Impending Pump Failure
An inspector review of selected portions of machinery history, opera-
l
tional data, and pump vibration data associated with RCP 1A; and
!
interviews with staff from Precision Mechanical Analysis, Inc. (PMA)
l
indicated that, with regard to the data reviewed, there was no prior
indication of RCP 1A's degradation.
. - -
-
-
- -
-
- - -
. -
. .
.
. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _
___-
.
. .
. . . . . . . .
h
,
.
.
7
The below listed documents were reviewed to evaluate surveillance,
corrective maintenance,
preventative maintenance,
and predictive
maintenance for RCP No. lA.
FPC-SP-300, Rev. 82
" Operating Daily Surveillance Log"
FPC-PM-101, Rev. 11
"4.16kV and 6.9 kV Switchgear"
FPC-FM-103, Rev. 10
" Maintenance of the Reactor Coolant Pump
Motor Oil Reservoir"
FPC-PM-105, Rev. 9
" Insulation Resistance, High Potential
Testing, Inspection, Cleaning and Drying
Windings of Electric Motors"
FPC-PM-133, Rev. 22
" Equipment Lubrication Procedure"
Prever*.ative Maintenance Control Sheet (PMCS) No. 139
PMCS No. 846
PMCS No. 1102
PMCS No. 1468
" Machine History Trending Program" Printout for Tag No RCP-001A
Westinghouse Oil Analysis Report dated 6/14/85.
The licensee's documented preventative maintenance program for Reactor
l
Coolant Pumps consists cf the following:
PM-133 (Section 7 and 10)
Frequency 180 days
PM-102 (Calibrate 1elays)
Frequency 547 days
PM-101 (Clean and nspect)
Frequency 547 days
PM-105 (sections 8 nd 9)
Frequency 365 days
The licensee replacts pump seals at each refueling outage, not as part
of a documented pro ram, but at the direction of the reactor pump
engineering section based on the engineer's experience and engineering
judgement.
Under contract to the licensee, PMA performs real time
vibration spectrum analysis monthly on the reactor coolant pumps. This
analysis measures and trends the ten frequencies of largest amplitude
in the vibration spectrum from data taken in the control room.
The
PMA . reports, since the last seal change (summer 1985 outage), and
discussions with PMA personnel indicate that the 1A pump vibration was
normal, lower than the other pumps, and the trend was flat.
During
the summer 1985 outage, the oil in all four pumps was filtered and
analyzed by Westinghouse for: insoluble metals, as additives, contami-
nants and wear metals; solid content by volume; water content by
volume; viscosity at 40 and 100 degrees centigrade; neutralization
number, acid or base indexes; and particle size by scanning electron
microscopy. The licensee determined that the oil was acceptable. The
licensee used the acceptable analysis to satisfy the requirements of
preventative maintenance control sheet (PMCS) 139 (sections 7.0 and
10.0 of PM-133).
-
.
... -.
.
. . . .
.
. ,
-
>
a
-
.
8
During this area of review, several problems with the accumulation and
evaluation of RCP vibration data were noted.
Specifically, SP-300
requires personnel, reporting to the shift supervisor, to record
Reactor Coolant Pump data including but not limited to the following:
pump seal pressures; seal leak flow data; seal water flow; control
bleed-off flow; component cooling water temperature; control bleed off
temperature; and radial vibration daily.
SP-300 paragraph 2.1.1
requires that a cualitative assessment of each data channel's behavior
be performed to determine if that channel is exhibiting normal charac-
teristics fo,r the system's current configuration based on previous
operating experience and sound engineering judgement.
paragraph 2.1.2 requires that if a data channel is determined to be
exhibiting other than normal characteristics for the systems current
configuration, the eff-standard values shall be circled and immediate
reference be made to the " Action Statement" of the applicable Stanoard
Technical Specification paragraph referenced.
No circles were noted
on the RC Pump Seal Data Sheets from the start up, after the summer
1985 outage, to January 1,
1986.
Conversely, the inspectors noted
the following apparent off-standard data: the Radial Vibration (mils)
ranged from 2 to 7 mils in the X direction and ranged from 1 to 6 mils
in the Y direction except on the following dates 9/28, 9/30, 10/2,
10/4, 10/25, 11/11-14, 12/28 and 12/31, 1985, when the X direction
vibration was 15 mils and the Y direction vibration was 13 to 14 mils.
'
There was no evidence that the personnel, reporting to the Shift
Supervisor, identified the elevated vibration data as off-standard,
and hence, no evaluation was made.
The inspectors made an inspection in the control room of the Reactor
Coolant Pump 1A Radial Vibration Indicator, the instrument used to take
!
the vibration readings discussed above, and conducted interviews with
!
licensee's engineering personnel. The licensee's engineering personnel
indicated that they expected vibration data to be less than 10 mils for
normal operation.
In fact, the " Alert" and " Danger" set points for
these instruments were set on the order of 20 mils and were nominally
within 2 mils of each other.
In one case, the inspectors noted the
" Alert" and " Danger" set points were the same.
Insufficient time was
available for a comprehensive inspector review of calibration
procedures and cata associated with the setting of RCP vibration
instrumentation, however, it appears that the instrumentation may be
improperly set and inconsistent with engineering analysis.
Although
these concerns are not considered to have been of any impact with
regard to this specific event, the inspectors noted that this equipment
has the potential for early warning of erratic RCP operation which
could preclude a pump failure from a progressive degradation.
The
inspectors discussed the above with both the licensee's operations and
engineering personnel.
The following appear to be areas of concern:
The off-standard values were not identified as such because the
-
RCP seal data sheet did not contain acceptance criteria for
comparison.
o
-1
.
.,
.
9
The off-standard values were,'most likely, the result of improper
-
operation and reading of the Reactor Coolant Pump Radial Vibration
Indicators by personnel reporting to the shift supervisor.
The " Alert" and " Danger" set points on the Reactor Coolant Pump
-
Radial Vibration Indicators appear to be too high to be an
effective harbinger of excessive pump vibration. At the time of
this inspection the basis for the selection of set point levels
was not available.
Disscussions with control room personnel indicated that they were
-
only collecting data on the Reactor Coolant Pump vibration fer
others to evaluate because the data sheet contained no acceptance
'
criteria.
Discussions with the responsible engineer indicated
that he assumed that the control room staff would make a qualita-
tive assessment of the data to determine whether the data was
exhibiting normal characteristics for the current configuration
based on previous operating experience and . sound engineering
judgement consistent with paragraph 2.1.1
of SP-300.
The
responsible engineer also stated that he assumed if off-standard
data was encountered, he would be notified.
This is indicative
that confusion exists with regard to which licensee organization
is responsible for pump operational data evaluation.
Resolution of this concern is identified as an unresolved item
pending further NRC review of calibration procedures and data
associated with this equipment.
(302-86-04-05)
e.
Review of Past Machinery History Associated with the Secondary Relief
Valve Problems
The inspectors also reviewed the below listed procedures and their
associated records for technical adequacy ard c'mpliance with applic-
able codes, procedures and regulatory reo Ji remen 's .
The applicable
code for Inservice Testing (IST) of pumps and valves is ASME Boiler and
Pressure Vessel (B&PV) Code Section XI, 1974 Edition eith Addenda
through Summer 1975. Specific attention was directed in tne following
areas:
number of valves to be tested and test frequency; set point
test procedure in accordance with ASME requirements; and provisions
for increased testing if valves fail to function properly.
Procedure No.
Title
SP-600, Rev. 12
" Pressurizer Code Safety Valves Test"
SP-602, Rev. 11
"ASME Section XI Relief Valve Testing"
SP-650, Rev. 12
" Main Steam Code Safety Valve Test"
- . -
- -
. .
- -
f,
W
'
e
-
.
10
r.
i
L
(1) With regard to the inspection above, the inspectors noted that
!
procedure PS-602, Rev. 11 directs the reinstallation of piping and
supports af ter relief valve testing with no reference to post
l
installation inspection for safety related supports. At the time
of this inspection it could not be determined whether safety-
!
related -ipe supports had been removed or disassembled and
l
reinstalled or reassembled without required quality control
l
inspections. Pending further NRC review in this area, this matter
will be identified as an inspector followup item (302-86-04-06).
(2) During the course of this inspection, the licensee informed the
!
inspectors that a non-related valve, EFV-14, was included in a
!
fire resistant sealed enclosure pursuant to 10 CFR 50 Appendix R.
l
Valve EFV-14 is required to be inspected once every two years to
l
verify that valve operation is accurately indicated, pursuant to
ASME Section XI paragraph IWV-3300. Pending NRC inspection of the
'
valve for testability and Inservice Test Program assurance that
the valve is adequately addressed, this matter will be identified
as inspector followup item (302/86-04-07).
l
f.
The inspectcrs reviewed the valve test records indicated below to
l'
verify adequate implementation of procedure SP-650.
Valve No.
Insoection Date
MSV-35
3/8/85
MSV-36
8/12/85
'
L
MSV-41
8/12/85
Within this area examined, no deviations or violations were identified.
!
!
l
(
l
!
l
+
.
.
.
.
.
.
.
.
.
.
.
. .
m