ML20247R751
Jump to navigation
Jump to search
| ML20247R751 | |
| Person / Time | |
|---|---|
| Site: | Catawba  |
| Issue date: | 07/24/1989 |
| From: | Lesser M, William Orders, Shymlock M NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| To: | |
| Shared Package | |
| ML20247R704 | List: |
| References | |
| 50-413-89-16, 50-414-89-16, NUDOCS 8908080095 | |
| Download: ML20247R751 (20) | |
See also: IR 05000413/1989016
Text
{{#Wiki_filter:._. _
__ _ - _ - _ _ _ _ _ - _ _ _ _ _ _ . .. . _
. . v . ,; ,
y
I
. -
eatog
3" -
UNITED STATES
l'
o,
lj
t
NUCLEAR REGULATORY COMMISSION
REGION ll
,
1
.101 MARIETTA ST., N.W.' ;
g ,. g, ATLANTA, GEORGIA 30323
.i
Report Nos. ;50-413/89-16 and 50-414/89-16
Licensee: Duke Power Company :
422 South Church' Street !
Charlotte, NC 28242-
Docket Nos.: 50-413 and 50-414 License Nos.: .NPF-35 and NPF-52
.i
Facility Name: Catawba 1 and 2
Inspection Conducted,: M_ay 28, 1989 - July 1, 1989 I
Inspectors:/ //f/ / Z 9 a
M. T. (Trddrs Cate/ Signed l
/M.'S.'Le 7)Y// / ?kk ff
er 'Datt Sig~ned
['
Approved by: /M / -
4, 7[,2f[8
M. B. ShymlocT r Sectibh Chief Sat ( Signed
Projects Branch 3
Division of Reactor Projects
. SUMMARY .
.
Scope:> -l
This routine, resident inspection was conducted on site inspecting in the areas
of review of plant operations; surveillance- observation; maintenance-
observation; review of licensee . nonroutine event reports; ' and folicwup of'
previously identified items; Part 21 Inspections, and Facility Modifications.
Results:
In the areas inspected the licensee's programs were determined to be
acceptable. Two violations and one deviation were identified.
One violation was identified invobing an uncontrciled door to a high
radiation area,
o
One violation was )dentified involving an . unlocked and mispositioned ]
Component Cooling Valve. ;
One deviation w.s identified involving tMe licensees Fire Protection Pump )
testing mnthod. .j
-i
.
89U8080095 890725
PDR ADOCK 05000413
Q PDC
.I
_' -;
j
-
'
i. .. '
-
..
2 < ,
,
.
. 3
I,
REPORT DETAILS
1. Persons Contacted !
-Licensee Employees j
*H. Barron, Operations Superintendent
W. Beaver, Performance'Enginesr. .
l'
*T. Crawford,' Integrated Scheduling Superintendent '
1
*J. Forbes, Technical 1 Services Superintendent
*R. Glover, Compliance Engineer
T. Harrall, Design Engineering
R. Jones, Maintenance Engineering Services Engineer
F. Mack, Project Services Engineer -
W. McCollough, Mechanical Maintenance Engineer
*W. McCollum, Maintenance Superintendent
*T. Owen, Station Manager
*J. Stackley, Instrumentation and-Electrical Engineer
B. Caldwell, Stction Services Superintendent
Other licensee employees contacted included techr.icians, operators,
mechanics, security force members, and office personnel.
NRC Resident Inspectors
*W. Orders
*M. Lesser
l * Attended exit interview.
2. Unresolved Items
An Uaresolved Item is a matter abnut which more information is required to
determine whether it is acceptable or may involve a violation. There was s
one unresolved item identified in this report. (paragraph 10 b)' )
3. Plant Operations Review (71707.and 71710)
-!
a. The inspectors reviewed plant operations throughout the reporting. i
period to verify conformance with regulatory. requirements, Technical :
Specifications (TS), and administrative controls. Control room logs, !
TS Action ~ Item Log. . and . the removal and restoration log. were. j
routinely reviewed. Shift turnovers were observed to verify that' j
they.were conducted in accordance with. approved procedures. Daily -!
plant status meetings were routinely attended. l
The inspectors verified by observation and interviews- that the' .!
measures taken .to assure physical protection of the facility met- ;
current requirements. Areas inspected included the security
organization, the establishment and maintenance of- gates,; doors, and
l
'l
1
- - - - - - - _ _ _ - _ _ _ _ _ - _ _ _ . - . _ _ - . _ - . __ _. _ _ . _ . ___>
., 4., ,
'
.
. .
2
i
isolation zones in the proper conditions, and that access control and
and badging were proper and procedures followed.
In addition to the areas discussed above, the areas toured were
observed for fire prevention and protection activities and
radiological control practices. The inspectors reviewed Problem i
Investigation Reports to determine if the licensee was appropriately
documenting problems and implementing corrective actions.
b. Unit 1 Summ ry
The Unit started the period operating at 100% power. On June 16, ,
1989, the unit was shutdown to mode 3 in order to repair an 011' leak i
on the 2D Reactor Coolant Pump Mctor. The unit was started up on
June 19. On June 26 the unit was manually tripped from 86% power
after a runback from 100% when ICF-28, Feedwater Regulating Valve to
Steam Generator 1A, failed closed and steam generator level
approached the low low level setpoint. The failure was caused by an
air leak on the valve's positioner. On June 27 the unit was started
up and operated at 100% for the remainder of the report period.
c. Unit 2 Summary
The unit began the report period in mode 5, repairing a leaking
Reactor Coolant Pump Seal. The unit entered mode.4 on May 31 and
Mode 3 on June 2. The start-up on June 7 completed the unit's second
refueling outage. On June 22 the unit was shutdown to mode 3 when
excessive cavitation was noted downstream of two of the four main
feed line (CF) flow orifices. The downstream check valves were
suspected tc be cycling. The four orifices.had been replaced during
the outage to minimize the effects of fouling. The problem occurred
on the C and D CF lines due to the close proximity of the orifices
and check valves. The C and D orifices were replaced with the
originals and are.now being monitored for the fouling problem. The
unit was started up on June E5 and operated at 100% for the remainder
of the pericd.
d. Verification of Quality Assurance Request Regarding Diesel Generator
Fuel Oil (TI 2515/93)
lhe purp9se of the inspection was to verify that the licensee has
included Diesel Generator fuel oil in its quality ascurance program.
The licensee's Quality Standards Manual for Systems, Structures and
Components lists diesel fuel oil as QA Condition 1, Nuclear Safety
Related. Portions of the system are also QA Condition 1. The
licensee is required by TS to perform various analyses on the fuel
oil after receipt and periodic chemistry sampling. Based on this i
inspection the TI is closed.
I
l
_--
<
' * .
.
. .
,
I
3
e. Uncontrolled High Radiation Area Door
On June 14, 1989 at approximately 2:45 p.m., the resident inspector
discovered door 316 to the "B" I!ecycle Holdup Tank Room, a posted
high radiation area, to be uncontrolled. The door was closed,
however, the latch mechanism was not latched. The inspector properly
closed h.;d locked the door and informed both the Health Physics
Supervisor and the Shift Supervisor. Licensee personnel immediately
inspected the door by allowing it to close by itself after opening
it. The door failed to latch on eight of ten attempts.
The licensee performs a daily check of all locked high radiation
doors pursuant to HP/0/B/1000/07, Duties of Health Physics Shift
Personnel. Door 316 had been verified locked earlier that day at
8:45 a.m. The licensee's investigation determined that the keys for
the room had not been issued to anyone within the last few days,
however,- a Nuclear Operating Technician (NOT) performing routine
rounds had accessed the room at approximately 11:30 a.m. The
licensee concludeo the door latch was faulty and the NOT failed n !
ensure the door was properly closed. The licensee inspected all !
locked doors the next day and discovered one additional door with a i
faulty latch. Work requests were initiated to repair the doors.
The inspectors asked the licensee to review the results of past daily
surveillance on the high radiation doors. The review revealed that
high radiation doors had been discovered uncontrolled by licensee l
personnel on 10 occasions over the past year, not ncluding this most !
recent event. Door 316 was not one which had previously been
discovered uncontrolled, however, three of the ten ccccsions last
year involved repeat occurrences on certain doors.
NRC Inspection Report 413, 414/88-33 documented one of the events
which occurred on August 2,1988 involving an uncontrolled high
radiation door. The svent was classified as a licensee identified
violation (LIV 413/88-33-02). Licensee corrective actions do not
appear to have been effective in. eliminating the evident trend of
personnel errors and/or faulty latch mechanisms,
Procedure, HP/0/8/1000/25 High Radiation Area Access, section 3.5
requires that each entrance to a high radiation area sha'll be locked
to prevent unauthorized entry. The area shall remain locke., except
during periods of access by personnel. This is identified as a I
violation of the requirements of HP/0/B/1000/25, violation
Failure to Maintain Door to High Rtdiation Area
)
413/89-16-01:
l Locked-
f. Unlocked and Mispositioned Valve '
On June 14, during a routine plant tour, the inspectors observed
valve 2KC05, Train A Post Accident liquid Sample Header Cooling
!
1
-
._
.. s. c ,
. .
,
. .
4
Isolation Valve, to be unlocked. OP/0/A/6200/21, Operating Procedure !
for the Post Accident Liquid Sample System, requires the valve to be
locked closed when not in use. A review of procedures in progress
determined the valve was not in use at thct time. The licensee
subsequently determined the valve was actually mispositioned open.
The licensee immediately rectified the problem and locked closed the
valve. The valve is required to be locked closed because it has the
potential to cross connect the two independent trains of Component
Cooling (KC) as described in FSAR section 9.2.2.2.
The licensee's review of the incident determined that on June 2,
1989 PT/2/A/4208/08, Post Accident Liquid Sampling System Periodic
Test, had been performed by chemistry technicians which lined up
Component Cooling (KC) to the Post Accident Liquid Sample Cooler by
opening 2KCD5. <
Step 12.20 PT/2/A/4208/08 restores the system upon completion of the
test. The step reads "The following valve nas to be locked closed by
Operations: 2KCDS." The step was initialed as having been completed
but in reality 2KCD05 remained open and unlocked until discovered by
the inspectors.
The licensee determined that confusion and miscommunication resulted !
between Chemistry and Operations as to which group would operate the ;
valve. The procedure requires the valve to be manipulated by !
Operations, however, on occasions Operations apparently has given
Chemistry the authorization. The valve is ,itually under control of
the Chemistry Departrant and is operated by oemistry personnel when
using OP/0/A/6200/?1.
The inspectors were also concerned with the method in which the
restoration steps are worded in that clear and concise action verbs
are not used to state the step requirements. Examples of this are: ;
-
" Inform Operations that the following valves will be closed..."
-
"The following 31ves have to be locked closed by Operations..." ;
-
"Obtain permission from Operations to open the following
valves..."
These observations were forwarded to thr licensee for consideration.
,
Since 2KCD5 is a chemistry controlled valve PT/2/A/4208/08 implied
the valve was to be controlled by Operations and led to the valve
being out of position, it is concluded the procedure was inadequate
and is identified as a violation of TS 6.8.1 Violation 414/89-16-02:
Inadequate Test Procedure to Ensure 2KCD5 Was Locked Closed.
Two violations were identified in paragraphs 3e and 3f.
1
_ _ _ - _ - - _ - _ _ _ . _ _ . _ _ _ _ _ _ - _ _ _ - - - - _ - _ _ _ _ _ - -_ _ -. ,
_
<'c .
.
. .
,
. .
5
4. Surveillance Observation (61726)
a. During the inspection period, the inspector verified plant operations
were in compliance with various TS requirements. Typical of these
requirements were confirmation of compliance with the TS for reactor
coolant chemistry, refueling water tank, emergency power systems,
safety injection, emergency safeguards systems, control room
ventilation, and direct current electrical power sources. The
inspector verified that surveillance testing was performed in
accordance with the approved written procedures, test-instrumentation
was calibrated, limiting conditions for operation were met,
appropriate removal and restoration of the affected equipment was
accomplished, test results met acceptance criteria and were reviewed
by personnel other than the individual directing the test, and that
any deficiencies identified during the testing were properly reviewed
and resolved by appropriate management personnel.
b. (Closed) Inspector Followup Item 413/86-39-02, 414/86-42-02:
Conduct 18 Month Capacity Tests of Fire Pumps at 100% and 150% of
Rated Volura. The inspector reviewed changes made to
PT/0/A/4400/01A, Exterior Fire Protection Functional Capability Test,
and determi.ned that the licensee is now testing the fire pumps at
100% and 100% of rated volume. Based on this the item is closed.
Review of this item revealed other areas of concern. TS 4.7.10.1.f.2
requires the pumps be tested at three points on the pump performance
curve. Section 11.0 of PT/0/A/4400/01A specifies the acceptance
criteria for the 100% and 150% flow capacity points, however, the
third point on the performance curve is not specified, nor is the
obtained value compared to any acceptance criteria. A review of
actual data obtained while conductina the test on three Fire
Protection (RY) pumps in January _1989 Indicated the third point to be
approximately 134 to 138% of rated flow, although explicit
requirements were nnt provided. This does not appear to meet the
intent of pump testing in that acceptance criteria is not specified.
Further review indicated that the three points on the pump
performance curve are intended to meet the requirements of National
Fire Protection Association (NFPA) Code 20, Centrifugal Fire Pumps
which stipulates in Section 11-3 that tests shall be perfonned to
determine the pump's ability to attain satisfactory performance at
shutoff, rated and peak loads. The licensee is committed to NFPA 20
through the Final Safety Analysis Report (FSAR) section 9.5.1.3 and
their Response to Appendix A to Branch Technical Position 9-5.1.
The licensee apparently is not meeting the commitment in that the RY
pumps are not tested at shutoff heed. The licensee has stated that
I testing at shutoff head has the potential to damage the pumps in that
minimum flow requirements are not met.
.
.pv., ,
. .
,
. ,.
I
6
The RY piping is designed to 150'psig with a relief valve to protect
it. Testing at shutoff head (220 'psig) will lift the relief and
place the piping system at risk. There.is no minimum flow line. In
a letter from H. D. Brandees to T. F. Wyke dated August 15, 1988 test
acceptance criteria for the RY pumps is specified. In lieu of
testin; at shutoff head,. a test point of 1000 gpm is specified to
meet minimum flow requirements. This. - however, has not been
incorporated' into the licensee's test procedure. This is identified'
as a deviation. of.the licensee's commitment to NFPA Code 20, DEV
413/89-16-03: Shutoff' Head Testing of Fire' Pumps.
One deviation was identified in paragraph 4b.
5. Maintenance Observations (62703)
Station maintenance activities of selected systems and components. were
observed / reviewed to ascertain that they were conducted in accordance with'
the requirements. The inspector verified licensee conformance to the
requirements in the following areas of inspection: the activities were-
accomplished using approved procedures, and functional testing and/or
calibrations were performed prior to returning components or systems to
service; quality control records were maintained;' activities performed
were accomplished by qualified personnel; and materials used were properly
certified. Work requests were reviewed to determine status of outstanding
jobs and to assure that priority was assigned to safety-related equipment
maintenance which may effect system performance.
No violations or deviations were identified.
6. Undersized Valve Actuators
Since March of 1988, problems' associated with the inability of some Rotork
actuated Borg-Warner valves to close under design differential pressure
conditions have been evolving. To present the problem accurately the
scenario to date is recapped in this report.
Background
The Auxiliary Feedwater (CA) System assures sufficient feedwater. supply to
the Steam Generators (S/G) for decay heat removal in the event of loss, of-
Main Feedwater. The CA System is independent for each unit. Each unit
has two 100% capacity Motor Driven CA Pumps and one 100% capacity Turbine
Driven CA Pump (CAPT).
Each CA pump is normaliy aligned to supply two of the four S/Gs per unit.
Each CA pump discharge line is' equipped t.ith a motor operated isolation
valve, an air operated fail-open control valve, and a check valve. To
mitigate any condition which could cause the CA pumps to operate beyond
their design capacity, runout protection is ' provided to automatically-
close the Motor Driven CA Pump Discharge valve to S/G B or C if the
Turbine Driven CA Pump is operating and the Motor Driven CA' pump of the
opposite train fails to start'after a 30 second time delay. These valves
______ _ -___ _- - _ - -
,, s. ,
,
.
,
. .
7
i
will also individually and automatically close on a high. pump discharge
.
4
flow indicating pump runout. The CAPT discharge isolation valves to S/G A I
and D are . closed when the CA System is aligned for Standby Readiness. l
These measures are intended to ensure that in the case of failure of a
Motor. Driven CA pump, the remaining motor. driven CA pump and CAPT would i
not be ,affected by a common mode failure caused by depressurization.of a
shared S/G. There are eight motor operated CA pump discharge valves per
Unit. Six of these valves are normally open when the CA System is' aligned
for standby readiness. Two of these six valves, CA '46B and' CA 58A,
receive an automatic signal to close for runout protection as described
above. The remaining motor operated CA pump discharge valves are provided
with essential train related power, but do not receive automatic signals
to reposition.
The CA pump motor operated discharge isolation valves are Borg-Warner item
number 6J-219, 4 inch 1500 psi gate valves with a flexible wedge gate.
These valves are equipped with Rotork motor actuators which are set up to
stop closing at a specified torque output of the motor which relates to
the stem thrust required to seat the valve. The motor opening circuit has
a specified torque switch setting to prevent damage to the actuator and
motor during the opening cycle and a 95% open limit switch which. prevents
the valve from backseating. In certain applications where the valve
opening is necessary to insure safety of the plant, the valve open torque
switches are bypassed.
Duke is using valve signature analysis Ltests to obtain data to properly
set the actuator torque switches to the manuf acturer's specified thrust
values for opening and closing valves. Development'of thrust values for
gate and globe valves is based primarily on the standard formula:
Thrust Differential Pressure Seat Load + Steam Rejection Load-+ Packing
Load.
The packing load is due to frictional resistance on the valve stem from
the valve packing. The packing load is verified by cycl.ing the valve
under zero differential pressure (D/P) conditions and subtracting the stem
rejection lood. The stem rejection load is calculated by multiplying the
upstream prescure by the valve stem cross-sectional area. This force aids
in valve opening and resists valve closing. The differential ' pressure
seat load is the largest variable in the thrust equation and uses an
empirically derived valve factor for each valve type. Flex wedge gate
velves typically use a valve factor of 0.3 and globe valves use a valve
factor of 1.1. The valve factor is multiplied by the valve seat area and
the differential pressure across the valve to approximate the sliding disk.
friction component against the seat rings.
NRC Bulletin 85-03, Motor Operated Valve Common Mode Failures During Plant
Transients Due to Improper Switch Settings was issued on December 15,
1985. This bulletin requests licensees to develop and implement a program
to ensure that switch settings on certain safety-related motor-operated
!
_,
:..a.
.
.
-
,
#
.
. .
8
valves are selected, set and maintained correctly to accommodate the
maximum differential pressures expected on these valves during normal ano
abnormal events:within the' design basis. This bulletin was issued by the
NRC due' to events' in the nuclear industry.where motor operated valves
failed due to improper torque switch settings. Duke expanded the scope of'.
its reply due to an investigation initiated in response to bulletin 85-03.
Event Description
. 0n March-14,.1988, Unit 2 was in Mode 4, Hot Shutdown. The Control Room-
Operator (CRO) was performing a flush of the Auxiliary Feedwater (CA)
piping' to S/G 2A using Motor Driven CA. Pump ?n, At 11:54 a.m.,'the
operator attempted to close' 2CA62A; CA Pump 2A Discharge.to-S/G 2A
Isolation valve, to prevent overfilling S/G 2A. 2CA62A failed to
completely close. This in turn contributed to the overfill of.S/G 2A and
resulted in a Feedwater Isolation that was addressed in LER 414/88-13.
On April 1, 1988, a Problec. Investigation Report (PIR) 88-143 was
initiated to determine the cause of. 2CA62A failing to completely close.-
On April 6, the Catawba Compliance group issued an Intrastation Letter
stating that 2CA62A was operable pending a Design Engineering operability'
evaluation. On April 20, 1989,-Design Engineering' issued an operability
evaluation for 2CA62A that stated the valve was operable based on .the
following reasons:
1
(a) 2CA62A does not receive any automatic signal to close, l.
(b) The function of.2CA62A is to isolate Motor Driven CA Pump 2A -{
from S/G 2A and this is done by Operator action in the event of j
a faulted S/G 2A, 4
:
(c) If the valve did not close during a design basis event with a
.
l
'
faulted S/G 2A, a condition would not result where insufficient
flow was delivered to the intact S/Gs; even with the single
failure of either of the two remaining CA pumps. . Also, the CA l
flow from CA Pump 2A to containment through the faulted S/G 2A
is bounded by the existing FSAR Chapter 15 analysis. :
On April 21, 1988, the Catawba Performance group issued Work Request 6398 i
PRF to investigate and repair the actuator on 2CA62A after performing:a ;
review of the PIR. On April 27, 1988, the valve actuator was tested per
the work reauest and the~ torque switch settings were found to be correct !
'
, per the design documents. On April 28, 1988, 2CA62A was retested at 1800 l psi D/P (design D/P), and failed to close. Instrumentation and Electrical L (IAE) technicians adjusted the torque. switches on 2CA62A to the maximum- l
allowed by the manufacturing design tolerance. On'May 30, 1988, after the ,
adjustment of the torque switches, 2CA62A was retested and fully closed. i
under 1800 psi D/P. ;
!
l
i
'
_ _ _ _ _ _ _ _ - - - - - .- -
- - _ _ - _ _ - _ _ _ _ - _
.. 4. ,
,
*
.
. .
9
On July 18, 1988, Design Engineering revised the Proposed Problem
Resolution on the PIR to have the internals of 2CA62A inspected during the
next outage to determine if excessive tolerances are present in the valve
disk guides that might cause binding and thus the higher required open to
close valve thrust.
On November 4, 1988, Duke performed valve signature analysis testing on a
Borg-Warner 6J-219 valve on the boiler feed loop at Riverbend Steam
Station. This test resulted in higher than anticipated required seating
loads at high D/P conditions. The valve failed to close on two of the
tests completed with D/P greater than 1500 psi.
As a result of the November 4 testing at Riverbend, the operability
evaluation for PIR 88-0143 was revised. Revision 1 expanded the scope of
the operability evaluation to address the ability to close of all the
motor operated CA pump discharge valves. This operability evaluation
concluded that the CA pumps motor operated discharge isolation valves were
operable if they could close against a D/P of 1425 psi based on design !
basis accident conditions. The operability evaluation further stated that J
valves CA58A and CA46B are the only valves that receive an automatic
closure signal and must be able to close when D/P across the valve is 1425
psi. The remaining valves CA38A, CA428, CA50A, CAS48, CA62A, and CA66B,
are closed by Operator action and the failure of any or all of the
remaining valves would not result in a condition where insufficient flow
is delivered to the S/Gs. The operability evaluation also states that
review of the Riverbend test data and the as-left valve signature analysis
test data for the affected valves indicates all valve operators provide
sufficient thrust to close against 1425 psid.
On November 15, 1988, further testing was conducted on the flow loop at
Riverbend Steam Station after the valve tested .on November 4,1988 had
been disassembled and metal spacers had been installed between the wedge
guide ring and the bonnet. The results of this test were not appreciably
different than the results of the test performed on November 4.
On Movember 26, 1988, during the Unit 1 End of Cycle 3 (EOC-3) refueling 1
outage, ICA42B, ICA46B, ICA58A, and ICA62A were tested at 1800 psi D/P
conditions. ICA42B was the only valve that would not close enough to
isolate flew, however, all four valves indicated intermediate position
after valve closing and were from 1/8" to 1/2" from their initial closed
position when the valves were closed with zero D/P. ICA46B fully closed
the secund time it was stroked under 1800 psid conditions. On March 13,
1989, while Unit 2 was beginning the E0C-2 refueling cutage, 2CA42B,
2CA468, 2CA58A and 2CA62A were tested using valve signature analysis tests
to determine the thrust required to seat these valves under D/P conditions
of approximately 1800 psi. During this test, three of the four valves
failed to completely isolate the flow during the test and all four failed
to wedge completely closed. Test data yielded closing valve factors
ranging from 0.38 to 0.74 which is considerably higner than the 0.3 valve
factor supplied by the valve manufacturer.
_ -__
. , , , , . ,
,
- '
10
l
As a result of the Unit 2 CA valve testing data, revision 2 of the
Operability Evaluation for PIR 2-C88-0143 was issued on March 17, 1989.
This evaluation stated that ICA46B is operable based on the Unit 1 testing
completed on November 28, 1988, where ICA46B fully isolated flow.
However, since D/P testing had not been completed on ICA58A, its torque
switch settings were changed to correspond to the worst case valve factors
deternined during the Unit 2 testing.
On .4 il 6,1989, 2CA62A was disassembled and inspected during the Unit 2
refuel g outage. Inspcction of 2CA62A did not reveal any significant
damage that could be judged to have caused binding in the valve.
Conclusion
The failure of 2CA62A to fully close with an 1800 psi differential
pressure across the valve has been attributed to design deficiency. The
actual valve factors obtained in field D/P testing of Borg-Warner 6J-219
valves are 30% to 146% greater than the manufacturers supplied valve
factors which are used to calculate the requiuc actuator thrust to close
the valves. The valve factor is used by the manufacturer to size the
actuator and to calculate the required thrust for the actuator torque
switch settings.
Duke conducted 0/P testing of Borg-Warner valves of similar design as the
6J-219 valve to determine if design deficiencies exist in other
Borg-Warner pinned guide ring, flex wedge gate valves. Testing confirmed
higher than expected valve factors. Design Engineering evaluated the
results of this testirg. From this evaluation, the other Borg-Warner
safety related valve types onsite affected are as listed below:
Grous A (Stainless Steel)
1. Safety Injection Pump to Cold Leg Injection Isolation valves
2. Centrifugal Charging Pump to Cold Leg Injection Isolation valves
3. Residual Heat Removal System to Safety Injection Pump and
Charging Pump Suction Supply (two different valve types)
4. Safety Injection Pump to Hot Leg Injection Isolation valves
5. Safety Injection Pump to Cold Leg Injection Header Isolation
valve
6. Steam Supply Isolation valves to the Auxiliary Feedwater Pump
Turbine
Group B (Carbon Steel)
1. Steam Generator Blowdown Isolation valves
2. The Steam Generator Main Feedwater (CF) Bypass to CA Nozzle
valves which are air operated.
_ .. _ _ - _ - _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ -
-___
.. .. . .
'..
-
l
.
i
11 l
1
J
The Group A valves were evaluated by Design Engineering. The findings
show the valves can perform their functions in accident conditions even
with the assumed higher valve factors. No modifications are planned at
this time for this group of valves.
The Group B valves ability to. close fully under design D/P conditions
could not be justified solely from an engineering calculation. Testing
was conducted on the Unit 2 CF to CA Nozzle valves at D/Ps greater than
the required accident D/P and.the valves isolated flow properly. Thus,
they were considered fully operable. The 5 team Generator Blowdown 4
Isolatit valves are considered conditionally operable with their torque
switches anmodified, with compensatory action to be taken by the Operators
in a seismic event to manually isolate blowdown flow. Unit 2's valve
actuators are now modified to provide sufficient closing torque, assuming i
the 0.74 valve factor for carbon steel valves. Unit l's valves are
awaiting modification at the next shutdown of sufficient duration. In the
interim, the compensatory action is being used to supplement the automatic j
function. i
i
Corrective Actions Taken
1
(1) PIR 2C88-0143 was initiated to determine cause of 2CA62A not stroking '
fully closed.
(2) Work Request 6398 PRF was written to investigate / repair the actuator
on 2CA62A and the valve was found to be setup within design
tolerances.
(3) After 2CA62A was retested and failed to close under full D/P J
conditions, the valve was setup at the maximum design tolerances and
closed when retested under full D/P conditions.
(4) Testin1 prformed on a Borg-Warner 6J-219 v.alve at Riverbend Steam
Static under D/P conditions found that the valve would not close at !
D/P greater than 1500 psi. I
(5) Testing performed on the Unit 1 CA Pump' Motor Operated Discharge
Valves at 1800 psi D/P condition during E003 found that 1 of the 4
valves tested would not fully stop flow. All four valves failed to
wedge fully closed.
(6) Testing performed on the Unit 2 CA Pump Motor Operated Discharge ,
Isolation Valves during E0C-2 yielded valve factors 27% to 147% i
greater than those supplied by the valve manufacturer.
(7) As a result of the tssting results on Unit 2, the torque switch
settings on ICA58A were increased to account for the worst case valve ,
factor found on Unit 2. The torque switch setting on ICA46B was not
changed since it had previously been tested and closed at 1800 psi
D/P on November 26, 1988.
,
___-__m.____m._ m
'
f. < i:. . *. . .,
. .
,
. 4
12 ;
I
Borg-Warner has been. contacted to respond to the ' failure of the
(8). 6J-219 valve to not operate as designed against a' design differential
' pressure of 2000 psi.
(9) Emergency Procedures have been revised to direct the CR0s to close
the CA flow regulating valves upstream of the Motor Operated CA pump
discharge valves in the event that a Motor Operated CA Pump Discharge .
Valve fails to- close. This is done .to reduce the differential A
pressure sufficiently to close the Motor Operated CA Pump Discharge
Valve on a faulted S/G.
(10) 2CA58A, which failed to fully close during the Unit'2 stroke' test at
the beginning of the E00-2 outage, was retested and adjusted prior to
the completion of the E00-2 refueling outage per Work Request 1015
>
MES. ]
Planned
(1) Further D/P testing was performed on Borg-Warner 6J-219 valves and
other similar valves to determine if the design supplied. valve
factors are adequate. The results of this testing will be reported
on the Duke Power Company's response to Bulletin 85-03.
(2) Duke Power will determine a course of action based on.the following
options.
-
Contract with a vendor to determine. the cause of the higher
valve factors and propose modifications to decrease the valve
factor.
-
Pursue in-house testing to determine'~ the cause of the higher. ;
valve factors and modify the valves to decrease the valve i
factor.
-
Pursue increasing the size of the valve act:stors ta be more. I
consistent with the prosent valve facto rs. !
!
- Maintaining the present actuators in operable status. I
l~ The NRC is currently developing an Information Notice to bring the lessons- )
learned at Catewba to the attention of the industry. ;
7. 10 CFR Part 21 Inspection (36100) .:
!
(Closed) P2188-06: .Inconel 600 Stear,,- Generator TubeL Plugs Susceptible to j
Stress Corrosion Cracking. Supplied for. B&W OTSG end.RSG DCS 88. The ,
licensee documented corrective action for this 10CFR21 notification on~PIR !
2-C88-0328. Catawba Unit 2 had six'of the suspect rolled plugs installed j
(Heat No. W592-1); two hot leg plugs and four cold leg plugs on Steam-
'
i
j
_ _ _ _ _ _ _ _ _ - _ _
. . . .
. .
,
. .
!
13
Generator 'A'. All six plugs were removed during the second refueling
outage which ended . in June 1989. The tubes were then plugged with
acceptable material. Based on this the item is closed.
8. Facility Modifications (37701)
The licensee implemented Nuclear Station Modification (NSM) CN-50391. The
modification relocated Nuclear Service Water (RN) supply to the RN pump
motor coolers and motor upper bearing oil coolers upstream of the lube
injection strainers. The licensee determined that the previous design was
such that the coolers would rob cooling flow from pump seal and bearing
and had experienced flow balance problems with lube injection. The lube
injection strainers are subject to clogging with microorganisms known as
copepods (!FI 413, 414/88-22-02) which has - the potential to damage the
motor due to lack of cooling.
The inspector reviewed the modification to ensure that the licensee's
safety evaluation was adequate, drawings were revised and the system was
properly retested. The licensee additionally intends to install parallel
lube injection strainers to allow on line cleaning. Licensee actions
appeared acceptable.
No violations or deviations were identif'ed.
9. Startup From Refueling Outage (71711)
During this report period, the activitie:, associated with the startup from
the Unit 2 E0C-2 refueling outage were reviewed to ascertain whether
systems disturbed or tested during the outage were returned to an operable
status prior to startup, and whether plant startup, heatup, approach to
criticality, and core physics testing were conducted in accordance with
approved procedures.
A walkthrough of portions of four systems was performed in order to
independent 1y ascertain that they had been returned to service properly
and in accordance with approved procedures. The systems selected were:
Reactor Coolant
Safety injection
Cold Leg Accumulators
Containment Air Return
Some problems were noted with the licensee's technique of securing certain
Safety Injection throttle valves. The licensee subsequently secured the
valves prior to unit startup.
Portions of the reactor startup were witnessed in order to ascertain that
the startup was performed in accordance with technically adequate,
approved procedures, and the startup activities were conducted in
accordance with the technical specifications.
_ _ - - -
,;.7..;. .
e e'
,
. .
14
Portions of three core-physics tests were witnessed in order to verify
that they were performed in accordance with technically adequate, approved-
procedures and, the requirements of the technical specifications. Those
.
tests were:
Core Power. Distribution'
Core Thermal Power
Rod Worth Measurement-
No deviations or violations were identified..
10. Review of Licensee Non Routine Event Reports (92700)
a. The below listed Licensee Event Reports- (LER) were reviewed to-
'
determine if the information provided met NRC - requirements. The
determination included: adequacy of description, verification of-
compliance with TS andl regulatory requirements, corrective action ,
taken, existence of potential generic problems, reporting ;
requirements satisfied, and the relative safety significance.of each j
event. Additional inplant reviews and discussion with , plant
personnel, as appropriate, were conducted for those reports indicated
byan(*). The following LERs are closed:
413/87-37 Rev 2 Failure of ITT Grinnel Mini Stiff
Pipe Clamps
*413/89-11 Inoperable Fire' Detectors
414/87-13 Rev 1 Auxiliary. Feedwater Auto Start on !
'
Loss of. Main Feedwater Pumps
*414/87-21 Rev 2 Reactor Trip Resulting From a
. Condensate Transient of Unknowh -
Cause: j
;
*414/88-21 Rev.1 Manual, Reactor Trip Due to Main j
Feedwater Pump Low Pressure Steam- !
Isolated j
*414/88-26 Rev 1 Shutdown Dua to Inoperable
Chemical Volume Control 1
Centrifugal Charging Pump. j
*414/88-30 Rev 1 Inoperable- Power Range ' Nuclear l
Instruments Due to Inadequate -
. Procedures !
*414/89-04 Safety injection on ' Rate
Compensated Low Steam Line i
Pressure
I
l
1
. . * . .
.
.
.
. .
15
*414/89-07 Containment Air Return Fan Start '
Oue to Inadequate Policy
Concerning Sliding Links
b. Inadequate Compensatory Sample Collection
During a review of LER 414/89-05, a problem was detected in the
philosophy employed by the Catawba Health Physics organization with
respect to the performance of compensatory sample analysis.
As stated in the LER, on March 21, 1989, the compensatory sgmple due
at 2:00 a m.. for 2 EMF 31, Unit 2 Turbine Building Sump Radiation
Monitor, was,'issed
m due to an inadequate policy to ensure compensatory
samples are obtained in the required time frame. The sample was
obtained at 2:10 a.m. 2 EMF 31 was initially declarr.i inoperable on
March 1, 3989, and all required 24 hour compensat 'v samples were
obtained prior to thic incident. A Health Physics 'alist arrived
at the sampling point prior to 2:00 a.m., however, due 6 a hydrazine
warning sign which required that a self contained breathing apparatus
(SCBA) be worn, he was not able to obtain the required compensatory.
sample within the 24 hour period. The present Health Physics policy
to obtain a compensatory sample 15 to 20 minutes prior to TS violation
does not take into consideration unforeseen delays.
TS 3.3.3.10 specifies that when EMF 31 is inoperable, effluent
releases may continue for up to 30 days provided grab samples are
analyzed for radioactivity. These grab samples must be analyzed at
least once per 24 hours if the specific activity of the secondary
coolant is less than or equal to 0.01 microcuries per gram dose
equivalent iodine. If the specific activity is greater than 0.01
microcuries per gram, the grab samples must be analyzed at least once
per 12 hours.
Having read the LER, the inspectors became concerned that HP operated
under the impression that the act of collecting the compensatory
somple fulfilled TS requirements.
The inspectors contacted the HP Supervisor to obtain actual sample
collection and analysis times for EMF 31 covering the period between
March 1 and June 14, 1989. The data revealed that on 52 occasions
during that period, the time between analyses exceeded 24 hours.
Discussions with the HP Supervisor revealed that indeed the focus has
been on sample collection, not the analysis of the sample.
The above appears to constitute a violation of the requirements of TS
3.3.3.6 in that during the period spanning from March 1 to June 14
samples taken as a compensatory measure, for the inoperability of
Radiation Monitor EMF 31 were not analyzed once per 24 hours as
l
t
.. .. o ..
. . .
,
a 0
16
required. Hswever, continued discussions with licensee management
revealed that accepted industry practice entails the interpretation
that pulling the sample followed as soon as possible by the analysis
constitutes an acceptable technique. In consideration of this
information this item will remain ' unresolved until such time as
Region II HP effluent inspectors review the event. This item is
identified as Unresolved Item 413/89-16-04: Interpretation Of
Requirement To Analyze Compensatory Samples Once Per ?4 Hours.
c. On May 15, 1989, Problem Investigation Report (PIR) 0-C89-0198 was
initiated identifying a potential violation of the intent of TS
4.9.4.1, due to operating the Incore Instrument Room Purge subsystem
during core alterations or movement of irradiated feal in containment
without the capacity to isolate the system upon h@ humidity. The
main portion of the Containment Purge system is equipped with heaters
which isolate the system upon a high relative humidity of 70 percent,
or upon a heater failure. The Incore Instrumentation Room Purge
subsystem does not include this feature. TS 4.9.4.1 implies the
Incore Instrumentation Purge tubsystem should include these heaters.
Based upon ar, evaluation by Design Engineering, it has been concluded
that heaters are not required for the Incore Instrumentation Purge
subsystem, and TS Surveillance Requirements 4.9.4.1 and 4.9.4.2 do
not apply to the Incore Instrumentation Room Purge subsystem. TS
Limiting Condition for Operat%n. 3.9.4 applies to the Incore
Instrumentation Room Purge - ;ystem. However, there are no
surveillance requirements which will verify operability of the Incore
Instrumentation Room Purge subsystem.
The licensee will submit a courtesy LER to describe this event.
No violations or devietions were identified,
i
11. Followup on Previous Inspection Findings (92701 and 92702) l
a. (Closed) Inspector Followup Item 414/89-07-02: 2A VX Fan Found
Operating Due to Sealed In Start Logic. The licensee submitted a i
courtesy Licensee Event Report (LER 414/89-07) describing the event. l
1
On March 16, 1989, at 4:33 a.m. Unit 2 Containment Air Return Fan 2A, i
CARF-2A, started in response to a Containment Pressure Control System l
(CPCS) permissive for high containment pressure. At the time of the ;
incident containment pressure was increasing due to Operations ,
personnel preparing for the Containment Integrated Leak Rate Test
(ILRT). However, the high-high containment pressure (Sp) setpoint, ,
'
which is also needed to start the fan, should not have been actuated.
Three days prior to the incident a calibration was performed on a
timer in the CARF-2A control circuitry. It is believed that during :
i
this calibration the timer was not properly isolated by use of
I
l
.
__ .-_- _ _ _
*C
. '9. e ,
#- 4 .
. .
17
: sliding links which subsequently caused an Sp signal to be
unknowingly _ sealed-in .in the CARF-2A circuitry. :This incident hes
been attributed to a possible inadequate policy involving control of
sliding links. Sliding links were not previously required to be
secured in the OPEN position. If the isolation sliding link
unknowingly reclosed during the calibration the circuit seal-in would
have cccurred. Unit 2 was in Mode 5, Cold Shutdown, when the seal-in i
and the fan start occurred.
Corrective actions have been implemented to ensure sliding links are' ,
secured when operated. The licensee evaluated the consequences of' l
the fan being dead headed and concluded that there were no concerns
i associated with. motor overheating or vibrations. .The licensee
additionally determined that -fan operation had no impact on the
ability of the discharge damper'to function., Based on this the~ item
is closed,
b. (Closed) Inspector Followup Item 413/88-38-09: Revision of'1NF-233
Testing Method and Licensee Review of Stroke Time Data. The licensee-
reviewed Section XI IWV test data to identify significant differences
between stroke times obtained using limit switch to limit switch -
(LTL) and initiation signal ~ to limit switch (ITL)' test methods. In
an Interstation Memorandum for File dated February 22,;1989, six
valves were identified where ITL testing approached or equalled the
required stroke time. (INF-233, IBB-10, IBB-21, IBB-56, IBB-61,
IVY-17) The licensee committed to test these valves by the IT. 3
l method in the future. Procedures have been revised and this item is !
closed. l
c. (Closed) Violation 413,414/88-35-01: Failure to Have Adequate
Procedures and Drawings For Power Changes and . Maintenance. - The
licensec responded to the violation in correspondence dated January !
6, 1989. Concerning the use of uncontrolled manuals, the licensee j
has advised personnel that uncontrolled technical documentation will 1
not be allowed outside the shop areas. With. regards to the :
inoperability of nuclear instruments due to power ' changes, the i
licensee issued a revised TS Interpretation and improved mechanisms
'
to ensure operators are trained on such guidance. Based on this the
item is closed,
d. (Closed) Unresolved Item 413,414/88-13-02: Evaluation of Corrective- !
Actions Regarding Control of Sliding Links and Jumpers. In addition ~
tn corrective actions taken by the licensee as documented _in
Inspection Report 03,414/88-15, the licensee .has completed the
.
:
following: I
-
Inspected 50% of safety-related cabinets for open sliding links,
lifted leads; jumpers. Based on a larger than expected number
of problems the scope was expanded to 1001. The results were i
documented in an Intrastation letter dated July 15, 1988. No TS
violations were identified. ;
l'
__ _
7 -
. w . .._ w .,
.
,o
. .
i
18- 'I
-
All groups reviewed procedures _which position sliding links or
jumpers to ensure independent verification is required. i
- Modification Implementation Procedures are now used 'to provide
specific isolations required for the modification - and to -
describe the effects on systems.
- The Post Modification Testing program includes steps to ensure
that systems or components affected by the repositioning of
sliding links are verified operable' by testing .or. verified
closed.
- The Catawba Safety Review Group (CSRG) conducted a review of
jumper and sliding link problems and did not identify additioral.
problems and concluded that corrective actions appear to have
led to improvements.
- Training on . isolation controls was conducted with all ~ applicable i
groups.
1
Based on these actions and.no further significant problems having
'
;
been identified, this item is closed.
e. (Closed) Inspector . Followup Item 413,414/87-44-02: Review of i
Lice.nsee Actions to Replace Vent and Drain Caps. The licensee j
implemented a program to routinely inspect piping systems for, among
other things, the existence of pipe caps on vents and' drains. The
licensee's method is to assign responsibility of systems. to each
shift of operators, she would then conduct walkdown inspections of.
the systems. Based on this the item is closed. ,
i
No violations or deviations were identified. I
i
12.. Licensee Quality Assurance Program Implementation (35502) Units,1 and 2 i
An internal office evaluation of'the licensee's quality assurance program
implementation was conducted by reviewing recent inspection repcrts, SALP
reports, open items, licensee corrective actions for NRC inspection i
l findings, and licensee event' reports. Particular ' emphasis was placed on
all new items since the end of the last SALP period (August 1,.1988 to-
May 31, 1989).
'
All- functional areas appeared to be satisfactory. Strengths were noted.
again in site management and management involvement. A strength was also-
observed in the emergency preparedness ' area in that the last emergency
exercise - was fully successful. Engineering and' technical support
continues to improve. A weakness continues. in the area cf procedure
adherence, procedure adequacy and attention to detail.. Problems were alsc
noted in the area of' health physics work practices.
I' - - - - - - - - - - - - - - - - - - _ - - - - _
c, ~. s ,
.
,.s
. *
19
13. Exit Interview
The inspection scope and findings were summarized on July 5,1989, with
those persons indicated in paragraph 1. The inspector described the areas
inspected and discussed in detail the inspection findings listed below.
No dissenting comments were received from the licensee. The licensee did
not identify as proprietary any of the materials provided to or reviewed
by the inspectors during this inspection.
Item Number Description _ard Reference
VIC 413/89-16-01 Failure to Maintain Door to High Radiation
Area Locked (paragraph 3e)
V10 414/89-16-02 Inadequate Test Procedure to Ensure Valve
2KCD5 Was Locked Closed (paragraph 3f)
DEV 413/89-16-03 Shut-off Head Testing of Fire Pumps
(paragraph 4b)
UNR 413/89-16-04 Interpretation of Requirement to Analyze i
Samples Once Per 24 Hours (paragraph 10b) !
<
!
:
}}