IR 05000293/1990020
| ML20058G988 | |
| Person / Time | |
|---|---|
| Site: | Pilgrim |
| Issue date: | 11/07/1990 |
| From: | Rogge J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20058G981 | List: |
| References | |
| 50-293-90-20, NUDOCS 9011140173 | |
| Preceding documents: |
|
| Download: ML20058G988 (43) | |
Text
,
%
%
$
!
.
.
'
U. S. NUCLEAR REGULATORY COhihilSSION i
REGION I
Docket No.: 50-293
'
Report No.: 50 293/90-20 P
Licensce:
Boston Edison Company i
800 Boylston Street Boston, Massachusetts 02199
'
Facility:
Pilgrim Nuclear Power Station Location:
Plymouth, Massachusetts Dates:
August 16 - October 8,1990 Inspectors:
J. Macdonald, Senior Resident inspec'
A. Cerne, Resident Inspector
.
W. Olsen, Resident Inspector
.
D. Kern, Reactor Engineer
/ 7!
Approved by:
4,. G-e__.
gYogge, Chief, R(a6tffr' Projects Section 3A ate
!
i Insocetion Summary: Insocction on August 16 - October 8.1990 (Recort No. 50-293/90 20)
.
Areas Inspected: Routine safety inspection in the areas of plant operations, security, mnintenance and surveillance, engineering and technical support, radiological controls, emergency
'
preparedness, and safety assessment and quality verification.
y Results: Inspection results are summarized in the attached Executive Summary. One violation
!
in the area of engineering and technical support was identified for failure to perform appropriate.
evaluation of two revisions to a design change for the temporary leak seal repair of a shutdown
cooling suction valve.
l t
,
i s
9011140173 901107 PDR ADOCK 0500
o
\\
.
!
-
,
.
.
i
,
.
'
EXECUTIVE SUMMARY i
!
1.
Plant Onerations: Operators displayed excellent transient response knowledge during the September 2 manual reactor scram. Appropriate procedures were utilized and the plant was maintained in a safe condition throughout the event.
The operators also effectively ensured positive control over all activities during
!
extended plant startup testing.
[
Maintenance and Surveillance: The component failures and system malfunctions wnich presented operational challenges during the September 2 event were partially attributed i
to inadequate maintenance program implementation. Although the plant was maintained in a safe condition, the diverse equipment complications cause NRC concern.
Emergency Preparedness: The September 2 event was appropriately reviewed with l
respect to Emergency Plan emergency action level criteria, proper state and local notifications were completed in accordance with administrative procedures.
Safety Assessment and Ouality Verification:
The multi-disciplinary analysis team (MDAT) investigation of the September 2 event was effective and well focused. The
'
MDAT was fully supported by senior management. The MDAT report and subsequent
-
operations review committee event review demonstrated continued improvement in the j
licensee self identification, assessment, and corrective actions capabilities.
'
Engineering and Technical Support: The engineering analysis of the reactor core isolation cooling system suction line pressurization event was comprehensive and utilized conservative assumptions. However, engineering support and disposition of revisions to a design change for the temporary leak seal repair of a shutdown cooling suction valve was inadequate and resulted in the design change safety evaluation bases being adversely
,
impacted.
,
2.
Violation:
,
One violation was identified as a result of the failure of the licensee to perform appropriate evaluation of two revisions to a design change for the temporary leak seal repair of a shutdown cooling suction valve.
,
i l
3.
Unresolved item:
One unresolved item was identified to review and assess results of the increased high i
pressure coolant injection system surveillance testing periodicity and data acquisition i
capability.
'
,
i
?
_
-
....
!
i
'
.
.
'
l TABLE OF CONTENTS 1.0 Summary of Facility Activities...............................
I
!
!
l 2.0 Plant Operations (IP 71707, 93702, 92702, 90712)...................
-
2.1 Plant Operations Review............ _,.................
2.2 Safety System Reviews...............................
'
2.3 Review of Tagging Operations...........................
2.4 Inoperable Equipment................................
t 2.5 Operational Safety Findings............................
,
2.6 Manual Reactor Scram
...............................
,
'
2.7 Plant Procedure Revisions as a Result of the September 2,1990 Event
..
2.7.1 Feedwater Regulating Valves.......................
6-
+
2.7.2 Reactor Core Isolation Cooling System.................
2.7.3 Startup Feedwater Regulating Valve...................
2.7.4 High Pressure Coolant injection
}
.....................
2.7.5 Shutdown Cooling System.........................
2.8 Reactor Restart from Forced Outage....,...................
3.0 Maintenance and Surveillance (IP 37828, 61726, 62703, 93702)
..........
3.1 Malfunctions and Failures Associated with September 2 Scram,,.....
3.1.1 1.oss of Feedwater Regulating Valve Position Control........
,
3.1.2 Loss of Control of the Startup Feedwater Regulating Valve.....
"
l 3.1.3 Repetitive Reactor Core Isolation Cooling System Trips......
,
3.1.4 High Pressure Coolant Injection System Overspeed Trips and l
Erratic Automatic Operation.......................
,
3.1.5 Shutdown Cooling System Suction Valve Failure to Open and Automatic System Isolation Actuation...
...........
.
3.1.6 High Safety Valve Tailpipe Temperature...............
3.2 Maintenance Program Improvements......................
4.0 Emergency Preparedness..................................
5.0 Safety Assessment and Quality Verification..._.......'..........
.
5.1 Multi-Disciplinary Analysis Team........................
5.2 LER 90-10
......................................
5.3 LER 90-Il
'18
.....................................
5.4 LER 90-12
.....................................-
5.5 LER 90-13
.......................................
5.6 LER 90-14
..................-...................
5.7 Review of Periodic and Special Reports.................... -
,
.i
_
-
.
..
,
I
.
.
i Table of Contents
6.0 Engineering and Technical Support
...........................
6.1 Reactor Core Isolation Cooling Discharge Check Valve Design-De fi ciency......................................
6.2 Analysis of Reactor Core Isolation Cooling System Suction Piping Pressu rization....................................
6.3 Leak Repair of Shutdown Cooling Suction isolation Valve MO-1001-50
.
7.0 Management Meetings (IP 30702, 30703)
.......................
7.I'
Routine Meetings..................................
7.2
"A" Recirculation Pump Motor Generator Set Conference Call......
7.3 Conference Calls and Meetings in Response to the September 2 3 Event
- The NRC Inspection Manual inspection procedure (IP) that was used as inspection guidance.
l Attachment I: September 12, 1990, Meeting Attendees, Boston Edison Company NRC Attachment 11: September 12,1990, Management Meeting BECo Slide Presentation q
.
L
!
!
l j
i l
l
l l
i.
_
I
'
l
.
..
'
.
!
'
q
'
DETAILS i
1.0 Summary of Facility Activities
-
,
t Pilgrim Nuclear Power Station was at 100% power at the beginning of the report
.
period. On August 29,1990 a previously identined minor packing leak on the
>
"B" feedwater regulating valve (FRV) increased significantly. On August 30,
-
reactor power was reduced to approximately 46% to facilitate repair of the "B" i
FRV and to accomplish a backwash of the main condenser. Reactor power
>
returned to 100% on September 1.
I On September 2, at 10:33 p.m a manual reactor scram was initiated due to an increasing
'
reactor vessel water level caused by a component failure in the feedwater regulating valve
,
control air system. The shutdown following the reactor scram was complicated by several component failures and system malfunctions which are discussed in detail in this report.
The licensee entered a fifteen day forced outage after plant shutdown to investigate equipment challenges during the event and to implement corrective measures.
On September 17, plant startup was initiated.
The reactor was maintained at
,
approximately 120 psi and 1% of rated power for several days to facilitate post maintenance and operability testing of steam turbine driven core cooling systems and to perform a temporary leak sealing technique on a shutdown cooling system valve.
Following completion of these activities power ascension was commenced. The turbine a
generator was synchronized to the offsite distribution system on September 25. The
-
reactor attained 100% of rated power on September 28. At the conclusion of the inspection period, the plant was operating at 100% power.
On September 3, the licensee notined the NRC Operations Center via the Emergency Notification System (ENS) at 1:02 a.m. to report the manual reactor scram and to report that the reactor core isolation cooling system had malfunctioned following the scram and had been declared inoperable. Additional notifications to the NRC Operations Center via the ENS were made on September 3, at 3:17 a.m. to report automatic Group 1, Group Il and Group VI primary containment isolation system (PCIS) actuatloas during plant shutdown and at 4:38 p.m. to report an automatic Group IP "CIS actuation following the initiation of the shutdown cooling system. These notifica,ns were made in accordance
'
with the requirements of 10 CFR 50.72. Notification via ENS to the NRC was also
,
made on September 13 to report a partial PCIS actuation of portions of the primary atmospheric sample system during a maintenance activity and on September 17 to report i
an automatic partial Group I PCIS actuation which occurred when the shutdown cooling system was secured.
.
-
.-..
'
.
.
!
'
On September 27, the licensee announced several management changes.
Effective
!
December 1,1990, Mr. Stephen J. Sweeney will be retiring as the BECo Chief Executive
,
Officer (CEO) but will remain Chairman of the Board of Directors. Mr. Bernard W.
Reznicek, currently the DECO President and Chief Operating Officer, will succeed Mr.
r Sweeney as CEO. Also effective December 1,1990, Mr. Ralph G. Bird, currently the
!
BECo Senior Vice President of Nuclear Operations (SVP-N), will assume an Executive Vice President position responsible for all engineering and production operations. Mr.
George Davis, currently the BECo Vice President of Nuclear Administration, will
,
succeed Mr. Bird as Senior VP-N.
Additionally, the licensee announced that effective September 27, Mr. Roy Anderson, current Plant Manager, was selected to succeed Mr. Kenneth Highfill as Vice President of Nuclear Operations and Station Director. Mr. Edward S. Kraft, current Deputy Plant l
Manager and Acting Plant Manager, was selected to succeed Mr. Anderson as Plant.
(
Manager.
On October 3, the NRC Region 1 Regional Administrator was onsite to meet with the resident inspectors, to tour the facility, and meet briefly with licensee management.
On September 10, the NRC Region 1 Director of Reactor Projects was onsite to meet with the resident inspectors and to tour the facility.
An NRC Region I Special Inspection was conducted in response to the September 2-3 manual reactor scram and shutdown to; evaluate licensee performance during the event; i
evaluate the effectiveness of the licensee investigation of the event; and evaluate
>
maintenance program contribution to the event. The special inspection was conducted on September 5-7 (Inspection Report 50-293/90-21).
l 2.0 Plant Operations (IP 71707, 93702, 92702, 90712)
.
2.1 Plant Operations Review The inspectors observed plant operations during regular and backshift hours of the following areas:
Control Room Fence Line Reactor Building
.
(Protected Area)~
Diesel Generator Building Turbine Building
'
Switchgear Rooms
.
t t
,
f
r
.
.
.
.
l
'
.
-
,
'
Control room instruments were observed for correlation between channels, proper functioning and conformance with Technical Specincations. Alarms received in
'
the control room were reviewed and discussed with the operators. Operator i
awareness and response to these conditions were reviewed. Operators were found
cognizant of board and plant conditions. Control room and shift manning were compared with Technical Specincation requirements. Posting and control of l
radiation, contamination and high radiation areas were inspected. Use of and i
compliance with radiation work permits and use of required personnel monitoring
{
devices were checked.
Plant housekeeping controls, including control of
flammable and other hazardous materials, were observed. During plant tours,
logs and records were reviewed to ensure compliance with station procedures, to i
determine if entries were correctly made and to verify correct communication of
equipment status. These records included various operating logs, turnover sheets,
'
'
tagout, and the lifted lead and jumper logs. Inspections were performed on backshiftsincluding August 15,21-24, and 27 31, Sep' ember 4 6,10,17 20,24 and 25, October 1, and 3-4,1990. Deep backshift inspections were performed as follows:
D31c Time September 2 2:30 pm - 9:45 pm September 3 8:30 am - 2:15 pm September 5 3:30 am - 5:00 am
Pre evolution briefings were noted to be thorough with appropriate questions and l
'
answers. The operators appeared to have good knowledge of plant conditions.
No unauthorized reading material was observed. Food, beverages and hard hats
>
were kept away from control panels.
[
f 2.2 Safety System Reviews
Portions of the emergency diesel generators, reactor core isolation cooling, high pressure coolant injection, residual heat removal and safety related electrical systems were reviewed to verify proper alignment and operational status in the standby mode. The review included verification that accessible major flow path valves were correctly positioned, power supplies were energized, lubrication and component cooling water was proper, and components were operable based on a visual inspection of equipment for leakage and general conditions. No violations
'
or safety concerns were identified,
,
o l
l
!
.
i
-
j
.
..
'
i
.
.
2.3 Review of Tagging Oncrations The following tagouts were reviewed with no discrepancies noted:
Tagout Description
.
90-6-42 Startup Feed Regulating Valve; Perform air tightness test on AO 643 90-6 43 Feedwater Regulating Valve AO-642At Implementation of PDC-90 56 90-6-32 Feedwater Regulating Valve AO-642B; Investigate cause of failure of the valve 90 13-26 Reactor Core Isolation Cooling System; Repair pressure indicator PI 1360 20-
-
90-23 33 High Pressure Coolant injection System;
.
Investigate cause of turbine trip 90-10-54 Shutdown Cooling Suction Valve MO 1001-50; Investigate cause of valve not stroking
'
open (on September 3)
,
2.4 Inoperable Eauipment Actions taken by plant personnel during periods when equipment was inoperable
'
were reviewed to verify that technical specification limits were met, alternate
'
surveillance testing was completed satisfactory, and equipment was properly returned to service upon completion of repairs. This review was completed for the following items:
Date Out Date In Sylkm 9/2 9/24 Reactor Core Isolation Cooling System
'
9/2 9/24 Feedwater Regulating Valves AAB 9/2 9/25 Startup Feed Water Regulating Valve 9/2 9/24 High Pressure Coolant Injection System 9/2 ROOS *
Shutdown Cooling Suction Isolation Valve (MO 1001-50)
- ROOS - remained out of service. In accordance with TS 3.7.A, downstream
.
isolation valve MO 1001-47 was administratively controlled in the closed position.
e
,
,.-
L
i
.
..
,
s
,
2.5 Onerational Safety Findings
'
,
Licensee administrative control of off normal system configurations by use of temporary modifications and tagging procedures was in compliance with
!
procedural instructions and was consistent with plant safety.
Overall plant i
cleanliness and material condition continued to be good, l
2.6 Manual Reactor Scram l
l On September 2,1990 at 10:33 p.m. the licensee initiated a manual reactor scram from approximately 100% power due to increasing reactor vessel water level.
'
I Approximately a half hour before the scram, operators began to receive high
.
reactor vessel water level alarms and experienced difficulty with feed water flow
control. Operators successfully stabilized vessel level by cycling feed pump recirculating valves. However, it became apparent that the main feed regulating
-
valves were not responding properly to signal demands and the operators initiated
,
,
a reactor scram.
Following the scram, operators attempted to initiate reactor core isolation cooling
'
(RCIC) to provide vessel water level control. -The RCIC turbine immediately tripped on overspeed; it was restarted twice, tripped again both times, and i
remained unavailable. Operators utilized the feedwater start up feed regulating
,
valve (8-10% capacity) and a single feed pump to control vessel water level. This valve operated erratically and appeared to be fully open, providing excessive water to the vessel. The inability to gain fine vessel water level control resulted in a Group I isolation (Main Steam isolation Valve (MSIV) closure) on high reactor vessel water level. Operators manually initiated the high pressure coolant injection (HPCI) system to provide reactor vessel water level control. However, HPCI was designed to deliver approximately 4000 gallons per minute to the reacter vessel in the event of a loss of coolant scenario and also provided
.
excessive water to the vessel.
Reactor pressure control was accomplished by cycling safety relief valves and by l
operation of HPCI in full flow test. At approximately 2:00 a.m on September
'
3, the MSIVs were reopened. The plant was then cooled down and depressurized normally, j
At approximately 1:00 p.m. on September 3, while attempting to establish shutdown cooling system (SDC) flow, SDC suction valve SDC-50 did not stroke full open. Troubleshooting revealed a failed seal in relay. The valve was opened and SDC was initiated at 4:41 p.m.
Immediately following initiation, SDC automatically isolated due to a pressure transient. After SDC systern venting and inspection, SDC was successfully established at 5:33 p.m.
A complete chronology and assessment of operator actions and plant responses to this event is documented in Inspection Report 50-293/90-21.
'
l t
'*
.
..
g f
1-
,
'
,
The inspectors, in conjunction with the special inspection team, determined that the operating staff performed appropriately in mitigating the operational challenges presented during this event.
,
2.7 Plant Procedure Revisions as a Result of the September 2.1990 Event
[
!
Review of this event, as documented in Special Inspection Report 50-293/90-21, identified several inadequate procedures which did not provide sufficient instruction to operators to mitigate an event involving increasing reactor water level. Additionally, the Special Inspection Report concluded that the instruction
,
of several procedures was incorrect and contributed to system malfunctions during the September 2 event. A summary of each procedure revision by the licensee in response to this event is provided below. The revisions were reviewed by
'
'
Special Inspection team members as well as the resident staff and were determined to be appropriate.
2.7.1.Sedwater Regulating Valves
,
Plant procedure 2.4.49, " Loss of Normal Feed and Feedwater Control
!
Valve Malfunction," was revised to include instruction to position reactor feed pump recirculation flow valves to control reactor vessel water level in the event of a feedwater regulating valve failure resulting in an increasing reactor vessel water level. The procedure also described the addition of indicating lights at the main control panels to indicate loss of power to the valve lockup circuit.
2.7.2 Reactor Core Isolation Cooling System On September 2, the initial reactor cort, isolation cooling system (RCIC)
turbine overspeed trip was the result of inadequate procedural instruction
'
for manual system initiation. The procedure directed the system be manually started with pump discharge flow directed through the minimum flow line. When RCIC pump discharge pressure equalled reactor vessel pressure, the procedure directed transfer to reactor vessel injection flowpath.
The RCIC flow control system' flow element is located
_
downstream of the minimum flow line. Therefore, when the system was manually initiated the flow element sensed essentially no flow and
demanded increased RCIC turbine speed to increase pump discharge flow until the turbine reached an overspeed condition and tripped. This event j
was the first manual initiation of the RCIC system since plant procedure 2.2.22, " Reactor Core Isolation Cooling System," was revised to reduce potential high/ low pressure system interface events. The procedure had -
been validated on the plant specific simulator, however the simulator
,
fidelity did not reflect flow element connguration. The procedure was i
.
_
.
.
.
_
.=..
,
.
'
revised to return to.the original system operating mode for manual initiation and injection to the reactor vessel. Also the procedure was revised to discuss the proper operation and resetting of the RCIC turbine overspud device and swap over from the vessel injection to full now test modes and vise versa.
j
\\
2.7.3 Startup Feedwater Regulatine Valve Plant procedure 2.4.49, " Loss of Normal Feed and Feedwater Control Malfunction," was revised to provide guidance to the operators upon the failure of the startup feedwater regulating valve in the open position with A
resultant increasing reactor vessel level. The use of reactor feed pump minimum How valves to control reactor level was also modified.
2.7.4 Hich Pressure Coolant inlection plant procedure 2.2.21,'"High Pressure Coolant Injection System," was revised to provide additional instruction to the operators concerning low system flow situations with flow controlin automatic.
2.7.5 Shutdown Cooling System J
Plant procedure 2.2.19, " Residual Heat Removal," is being revised to
-
include instructions for operation of the keep-fill system during startup of f
the shutdown cooling system to reduce the potential for hydrodynamic i
events.
Inspector review of the draft revision determined proposed
!
changes were appropriate, l
)
2.8 Reactor Restart from Forced Outage i
On September 17, the licensee made preparations for reactor startup from the unscheduled outage. At 7:23 p.m. an automatic Group I PCIS actuation occurred
'
immediately after the shutdown cooling system (SDC) was secured causing closure of the main steam isolation valves (MSIVs). The isolation occurred due to a high
,
reactor vessel water level ec.1dition which resulted from; initially high reactor i
vessel water level prior to securing SDC; higher than normal initial reactor
.
coolant system temperature; and the tripping of both residual heat removal (RHR)
I pumps and closure of the RHR injection valve in very close order.In response to the Group I isolation, the SDC system was returned to service and the event was '
reviewed to ensure all components responded as designed. Following satisfactory I
event review the isolation was reset, the SDC system was secured, and plant startup was resumed.
!
On September 18 at 3:42 a.m., reactor criticality was achieved and the reactor
-
.
..
!
.
'
!
'
i was stabilized for several days at approximately 120 psi and 1% of rated power to support post work and operability testing of the turbine driven HPCI and RCIC i
systems. Following successful completion of the testing, reactor pressure was
,
increased to approximately 400 psi to support temporary leak seal repair of the
!
.
MO-1001-50 SDC system suction valve. On September 24 repairs to MO 1001-
'
50 were completed and ascension to full power was commenced. The turbine -
'
,
generator was synchronized to the distribution system on September 25 and the.
-
.i reactor achieved 100% of rated power on September 28.
The operations staff performed well during. reactor restart.
Response and
corrective actions to the Group 1 isolation were appropriate. Plant conditions
'
were well maintained throughout HPCI and RCIC system testing evolutions.
l
3.0 Maintenance and Surveillance (IP 37828, 61726, 62703, 93702)
l 3.1 Malfunctions and Failures Associated with September 2 Scram
The September 2 manual reactor scram was init!ated as a result of a component failure in the non safety related feedwater regulating valve (FRV) control air-
'
system. Additionally, several system malfunctions and component failures created
!
operational challenges in maneuvering the reactor to a cold shutdown condition.
Each malfunction and failure including failure mechanism, causal analysis, and licensee corrective actions enacted or planned, is addressed below individually.
.
3.1.1 Loss of Feedwater Regulating Valve Position Control Loss of FRV position control which resulted in an increasing reactor vessel water level was the initiating cause of the event. The FRV contial was lost due to a blown fuse (1/8 amp) in a common FRV control air power supply (640 42). The blown fuse caused both FRVs to lockup in fixed
position. However, as control air decayed and bled from the FRV lockup system, the valves tended to slowly stroke open under the opening spring force.
,
f
!
t t
?
.
-
..
__
'
.
..
.
i
<
Additionally, the blown fuse caused control room FRV lockup light
indication to be inoperable. The fuse blew when moisture from the i
previous steam leak on the "B" FRV penetrated ajunction box on the "A"
FRV control air panel, migrated internal to conduit, and collected in a
pressure switch (PS-656A) causing it to ground and short. The pressure switch was powered by the circuit (640-42) protected by the fuse.
Beyond the previous "B" FRV packing leak which introduced abnormal
[
amounts of moisture into the surrounding area, the major contributing
!
factor to the fuse failure included the degraded material conditlen of the junction box seal. Visualinspection of thejunction box revealed evidence
of previous standing water indicating the degraded junction box was a long :
standing condition.
j As was previously noted, the feedwater control system and the FRVs are not safety related.
As such, power supplies are not required to be
independent and separate and junction boxes are not subject to
-
environmental qualification criteria.
The licensee replaced the failed pressure switch. The degraded,iimetion box was refurbished and weep holes were drilled in all FRV junction boxes to prevent water buildup. The junction box maintenance program
'
practices were also reviewed.
!
.
Additionally, the licensee implemented a modification (PDC 90-56) to provide separate power supplies (from the 640-41 power supply) for each i
FRV, with each output having an internal protective fuse (1/8 amp). This modification provides the capability for continued operation of one FRV
,
'
in the event a fuse should fail in the circuitry of the other FRV control circuit.
.
!
3.1.2 Loss of Control of the Startuo Feedwater Regulating Valve Following the reactor scram, operators attempted to utilize the startup
,
feedwater regulating valve (non safety related) to provide reactor vessel
,
water level control. However, a degraded internal diaphragm caused the air booster relay to fail which resulted in a startup FRV lockup, similar in
effect to the main FRV lockup but independent in cause. Feed pump
{
discharge pressure forced the valve full open thereby delivering approximately 8-10% of full power feedwater flow to the reactor vessel.
Probable cause for the booster relay failure was believed to be random end l
of life ' component failure with contribution from a harsh operational
. environment. However, exact in service and shelf life times for the
-
?
'f
- i
-
.
,_
_
!
l
.
..
.
.
.
.
10 booster relay were not able to be determined. The failed booster relay was replaced. Additionally, the valve was repacked with live load packing
,
'
to provide enhanced stem leakage mitigation potential. Following these activities the valve was successfully stroke tested.
3.1.3 Recetitive Reactor Cog Isolation Cooline System Trios
!
Operators attempted to manually initiate the reactor core isolation cooling (RCIC) system three times following the manual reactor scram to provide
,
reader v==! viater level control. However, the RCIC turbine tripped
,
shortly after each attempted system initiation. The first turbine trip
'
occurred as a result of an actual overspeed condition due to an inadequate procedure to control manual initiation and is discussed in Section 2.7.2.
The second and third RCIC system turbine trips where different from the
-
first in that an actual civerspeed condition was not achieved. On the second start attempt, injection valve MO 1301-49 was opened and for approximately twelve seconds flow to the reactor vessel was established before the trip and throttle valve actuated. The injection valve has a fifteen second stroke time. While this valve was closing, the injection check valve 1301-50 remained approximately 1 inch off the close seat causing the RCIC pump suction line to be pressurized until the injection valve stroked full closed (analysis of this condition is documented in sections 6.1 and 6.2).
On the third and final start attempt, RCIC operated for approximately 80 seconds in the full flow test mode until a spurious overspeed trip occurred prior to opening of the injection valve.
The premature overspeed tripping of the RCIC turbine on the second and third start sequences were the result of excessive mechanical tolerances in the trip and throttle valve linkage caused by past multiple actuations and normal system vibrations and were exasperated by previous inadequate maintenance. Additionally, the internal tappet (valve) latch surface was l
slightly rounded. Also, the tappet guide area was affected by the intrusion i
'
of foreign material into the oil system which affected reset capability.
The licensee enlisted vendor support to the resolution of the RCIC turbine trip events.
The vendor representative immediately identified the
!
excessive trip and throttle valve linkage tolerances, as well as, foreign l
material intrusion into the tappet guide and rounding of the tappet nut latch -
surface.
t The trip and throttle valve was dissembled and cleaned as was the l
i
.
_
[
-
,
..
,
-
.i
.
,
.
'
i
-
,
l overspeed trip weight and reset spring assembly, and the tappet nut latch surface was machined to design specifications.
The oil system was
-
drained, the sumps were cleaned, and the system was refilled, i
Following plant startup with the reactor s:abilized at approximately 120 psi the licensee conducted extensive post maintenance RCIC testing. The RCIC turbine overspeed trip test verified proper function of the trip and throttle valve. Several delays were encountered while performing this test.
Initially the test procedure necessitated revision to provide a discharge path i
from the RCIC barometric condenser to the suppression pool.
"
Additionally, the EG M (electronic governor module) and a damaged i
multi point pin connector, which were contributing to erratic speed indication, were replaced.
Following completion of the RCIC turbine overspeed test, the revised.
RCIC system operating procedure (2.2.22) was performed and validated.
Additionally, de licensee performed a Temporary Procedure (TP 90-68)
,
which verified full flow test capability as well as verified full flow manual
!
reactor vessel injection capability. This test also verified that RCIC discharge check valve 1301 50 fully seated.
All RCIC testing was
<
-
satisfactorily completed and the system was declared operable on
,
September 24.
i 3.1.4 High Pressure Coolant Iniection System Oversoecd Trips and Erratic Automatic Oncration During recovery from the scram, operators manually initiated the high i
pressure coolant injection (HPCI) system two times. The first start was
!
initiated to provide reactor vessel level control and was utilized for l
l approximately two minutes. The second start was initiated to provide F
reactor vessel pressure control.
.
t
.
l
.-.-
s
_
'
.
..
'
e
.
.
During both HPCI system starts the operators noted flow oscillations in automatic flow control at less than 3000 gpm. The oscillations were eliminated when the flow controller was placed in manual control.
Additionally, during multi disciphnary analysis team review analysis of HPCI system data, the lleensee determined on both initiations that the HPCI turbine experienced actual overspeed trips which automatically reset during the start sequences.
The licensee, with vendor s'?pport, conducted extensive causal analysis of each of the observed anomatics in HPCI operation, however was unable to ascertain definitive root cause determinations. With respect to the overspeed trips, the licensec identified a hand operated valve in the turbine oil system in the full open position that the vendor recommended be approximately one turn open. The valve (HO-2301 123), which is the oil relay pilot supply block valve that ports control oil to affect control valve responsiveness to position demand signals, was repositioned and appropriate procedures were revised to reficct the new position.
Licensee review of the HPCI flow oscillations while at low flow
!
conditions (<3000 gpm) in automatic cor. trol identified that the EG R actuator needle valve was one full turn open vice one-quarter turn open as recommended by the vendor. Adjustment of the needle valve provided smoother automatic control at low HPCI flow conditions. The licensee also attributed the oscillations to the design limitation of the automatic controller to maintain stable flow conditions at the low end ofits operating range. Due to the presence of contaminants in the oil system, it was drained and flushed.
The licensee connected tempomry diagnostic instrumentation to the HPCI system to accumulate additional operational data during post maintenance testing and subsequent surveillance testing. The post maintenance testing conducted September 23 indicated improved HPCI low flow operation as well as the absence of overspeed trip conditions and the system was declared operable.
In order to expediently assess the long term i
effectiveness of the HPCI system corrective maintenance, the licensee has
'
changed the HPCI surveillance periodicity from monthly to every two weeks for the next two months. However, due to the inconclusive root cause determinations for the HPCI system operational anomalies this issue -
is identified as an unresolved item until the results of testing have been reviewed (50-293/90-20-01).
'
.
'I
_. -
- _ _ _ _ _ _ _ _ _ _ _ _
,
i
'
.
.,.
.'
'
<
i 3.1.5 Shutdown Cooling System Suction Valve Failure to Ooen and Automatic j
System Isolation Actuation
,
'
On September 3 while performing initial shutdown cooling system (SDC)
l alignments, the inboard suction isolation valve, MO 100150 did not stroke full open in response to control room operator signal. The valve
is designed as a seal-in valve and should stroke to full open position
following a single control switch manipulation..
The ' valve.was
'
subsequently opened by holding the control switch to the open position until full open position indication was received.
.
.
'
The licensee was unable to duplicate the scal-in circuit function failure during troubleshooting.
However, Suctuating seal in relay contact -
resistance values indicated potential contact degradation. The seal in relay was replaced. Additionally, the valve's circuit breaker open coil and auxiliary contractor was replaced.
Additionally, while entering shutdown cooling on September 3, an automatic Group Ill PCIS actuation occurred. The isolation occurred due to a sensed reactor pressure of greater than 100 psig upon SDC initiation.
The PCIS pressure sensing switch signal originates in the recirculation loop suction piping in close proximity to the SDC connection.1It should
'
be noted these isolations have been a recurring problem.
Licensee investigation, in conjunction with the' vendor, indicated the potential for steam bubble formation in the SDC suction piping between the recirculation loop and MO 100150 Collapse of a steam bubble upon SDC initiation could provide a momentary pressure pulse suf6cient to cause Group Ill actuations. This position is supported by generic industry (BWR) experience. Due to SDC/RHR piping con 6guration the potential
-
exists for incomplete system vent and Gil evolutions. Additionally, a pressure switch instrument line snubber was observed to be missing, which
,
may have contributed to the isolation events.
-i j
l
-
<,
.
..
N
.
'
The pressure switches were calibrated by the licensec and determined to be within acceptable limits.
Additionally, SDC operating procedure (2.2.19) was revised to provide improved vent and fill instruction and to provide slower suction piping backfill via the keepfill system.
The licensee is reviewing a 1976 vendor service letter addressing RHR/ recirculation system water hammer during plant cooldown.
This issue was addressed previously in NRC inspection report 50 293/90-15. The inspectors will continue to observe and review licensee activity during SDC initiation.
3.1.6 High Safety Valve Tailpine Temocrature Shortly before the September 2 reactor scram, operators noted that safety valve RV-203-4B tailpipe temperature recorders indicated an increase in temperature from 165 to 185 degrees F.
The safety valve actuation setpoint is 1240f.12 psig. Reactor pressure was 1036 psig prior to the scram and decreased following the scram. Following containment de-inerting the licensee visually inspected the safety valve and tailpipe without observing evidence of actuation or leakage. On September 22, a drywell tour with the reactor at approximately 120 psi was conducted. A minor packing leak on hand operated root valve HO-130190 was observed to be impinging on the safety valve tailpipe causing the elevated temperatures.
The leak was terminated by tightening the valve packing gland and tailpipe '
temperatures returned to normal.
3.2 Maintenance Procram improvements As was documented in Special Inspection Report 50-293/90 21, insufficient maintenance supervision was present during the RCIC trip and throttle valve disassembly and the FRV packing removal to provide oversight and ensure as found data was properly recorded.
Additionally, instances of' inadequate
- i procedural instruction and work area lighting were also observed by NRC
'
inspectors during conduct of these activities. The report noted the recent licensee i
self-assessments of the maintenance program which identified several deficiencies that were evidenced during this event. Further, the report recommended the licensee accelerate programmatic improvements resultant from the self assessments.
!
i'l
.
g
.
.
.
.
_.
,
,
...
h
.-
,
.
'
15.
During the September 12 management meeting the licensee discussed conceptm aspects of the maintenance improvement program. Subsequent to the meeting the the program.
'
\\{
licensee provided the inspector with a detailed summary of the major elements of Major elements of the licensee maintenance program improvement plan completed or scheduled include:
Implementation of the Work Control Process as written to relieve the first
{
-
line supervisor of administrative functions.
To aid in this effort, two system engineers and three NED engineers were temporarily reassigned to the work control group.
'
l
-
First line supervisors v c directed to perform duties only as prescribed L I
- in their job description, w. a the major emphasis on job-site supervision.
The production work force has been restructured into teams, with each
-
,
team assigned to one specific first line supervisor. The supervisor will be responsible for team performance, training, qualification and personal.
~
y accountability.
Each team will be specifically assigned complete
,
responsibility for, routine repairs, preventive ' maintenance and (
l surveillances.
q A " quality of workmanship" training module is under preparation by the
-
i Nuclear Training ' Department.. The module. is -intended to_ provide l
specialized training in the completion of quality work activities and should
.j be ready early in 1991.
'
The " Rework-Program" effort between systems' and maintenance to
-.
identify rework activity, determine root cause and -initiate corrective actions is planned to be implemented before November,1990.-
'
Improved maintenance request scheduling on a quarterly basis.
-
~ A " Maintenance Quality' Improvement Program" is planned to 'be
-
developed and implemented in which anyone who had involvement:in
selected jobs will participate in critiques chaired by the maintenance.
manager. Strengths and weaknesses will be determined. Lessons learned will'be developed and incorporated into maintenance processes.
.
-
.
-
.
,
'
.;...
'<
..
,
c
.
'
I
- !
The Preventive Maintenance Program is planned to be upgraded. Four
-
predictive maintenance programs are planned to be in place in early 1991; including thermography,' lube oil ' analysis, vibration monitoring and analysis, and bearing temperature trending.-
!
The maintenance proced_ure upgrade program continues to make progress.
-
The program includes the normal technical verification and human-engineering as well Eas such items as a plant impact statement, =
identification repair and replacement parts for preventive maintenance and surveillances, proposed tagouts, radiological work permits, and special tools. This effort should complete in 1992.
In addition. to routine inspector ' review of maintenance performance,- the
- !
effectiveness of the improvement _ program will be comprehensively assessed I
during the upcoming NRC maintenance team inspection to be conducted ~
November 5-16, 1990 -
4.0 Emergency PrenarednCH j
The inspector conducted a complete review of the criteria of emergency plan
!
implementing procedure EP-IP-100,~ " Emergency Classification," Revision 1 with respect -
i to the events of the September 2 manual reactor scram to independently determine if any -
j emergency action level entry conditions were satisfied. The inspector concluded that the l
operaticna' occurrences of the-September 2 manual reactor scram did not present
emergen'y action level entry conditions and therefore the licensee was appropriate in 7t
]
activatig the emergency plan.
!
!
Additionally, the inspector concluded the licensee ' appropriately implemented the instruction of procedure. EP-AD-130, " Responsibilities Lof On Call Management l
_
Representatives," Revision 2. The procedure directs that the on call EP manager notify -
i Commonwealth and town officials of events which are significant enough to warrant
!
increased awareness of plant management but are: not serious enough to warrant
,
e implementation of the emergency plan via amergency l action level classification.
. {
.
Inspector review of procedure notification checkiists indicated all specified state and local-j officials or their alternates were either contacted or attempted to be contacted during the
!
morning of September 3.
l
i e
.1 i
.______.-_--.-_--.--_-._L._-------_
. -. _. - - -
.
.
.
-
.
.
.
5.0 Safety Assessment and Oualltv Verification
5.1 Multi-Disciplinarv Analysis Team
~
Immediately during recovery from the September 2 manual reactor scram the licensee classified the event as a Type 3 scram consistent with the instruction of'
procedure 1.3.37, " Post Trip Reviews." A Type 3 scram is an event in which the course of the scram is not positively lcrown and/or some safety related or important equipment functioned abnormally during the event. In the instance of
the September 2 event the course of the scram was understood, however, the high -
'
pressure core injec_ tion and reactor core isolation cooling system malfunctions
-;
were not readily understood.
H A Type 3 reactor scram classification t.lso requires that a multi-disciplinary j
analysis team (MDAT) be assembled to review each aspect of the event and j
identify root cause determinations, i
The licensee established a 38 member MDAT on September 3. The MDAT team l
leader was the station chief technical engineer who possesses a current ' senior
reactor operator license. The 37 team members were engineers and technical-
!
specialists that represented diverse plant disciplines. The MDAT was stationed in the technical support center and provided oversight during day and evening i
shifts. Extensive assessment of MDAT performance is~ documented in the Special l
Inspection Report (50-293/90-21). The MDAT continued to function effectively
!
through plant restart. The MDAT issued. a comprehensive report that was
~
-
submitted to the Operations Review Committee (ORC) for review and approval ~
prior to restart. The ORC provided extensive review and analysis of the MDAT report and subsequently recommended plant 'startup readiness to the Station Director. All MDAT report followup items were serialized and were required to be reviewed and approved by ORC prior to cic.;ure.
.j Subsequent to plant restart the MDAT amended _ applicable report sections to incorporate the results of HPCI post maintenance testing. However, amended j
MDAT report sub-sections were not presented to ORC in a timely fashion.. This inspector concern was brought to the attention of the MDAT team leadcr who
!
acted to ensure ORC was provided report updates in a timely manner.
{
<
,
y i
I
_ _ _ _.
_......
...
...
i
-,
,- ;.
,
'
.
.
,
18_
q
- .
t 5.2 LER 90-10
[
LER 90-10, " Completion of a Shutdown Due to One Inoperable Recirculation-Loop," addressed the July 3,1990, TS required plant shutdown necessitated by j
l the inability.to restart the "A" M-G set which had tripped the preceding day.-
This event was documented in NRC inspection report 50-293/90-15. The LER
'!
accurately detailed the event; provided extensive technical description of the
,
recirculation system and described corrective ~ actions enacted.
.
The ORC review of this LER was noteworthy The failure and malfunction report
(F&MR 90-204) which will document the event root cause determination was not'
complete when the LER was drafted. Therefoie, the initial LER presented to
' ORC lacked sufficient causal analysis. The ORC quorum appropriately noted this.
weakness and requested further causal analysis development be included in the
'
LER. The LER was revised to include documentation of a modification to the M--
_
G set speed control circuitry (PDC 190-14); and was preser..ed to ORC.
i Subsequently, ORC approved the LER'with the addition of a commitment to,
'
submit a supplement after the F&MR. investigation is. completed. The licensee expected the supplement to be submitted by November 2,1990. The inspector had no further concerns regarding this LERc 5.3 LER 90-11 LER 90-11, " Automatic Closing of the PCIS Group 111 Isolation Valves While Shutdown," addressed the July 3,1990 automatic isolation of the shutdown
.
cooling (SDC) system immediately following system initiation. This event was
.
documented in the NRC Inspection Report 50-293/90-'15. The LER appropriatbly addressed the reporting criteria as well as identified similar previous events. An s
additional similar event is documented in Section 5.6 of this report.
As discussed in the referenced inspection' report,(the inspector.will review the status of the plant design change intended to enhance SDC venting. The inspector had no questions regarding this LER.
q
.
5.4 LER 90-12
.
'
LER 90-12, "Two. Radioactive Sources Not' Leak Checked Within TS Required.
Interval," addressed the July 12,1990 licensee discovery that two of twenty-six radioactive sources had not been leak checked within the six= month periodicity
'
required by TS.
Upon discovery, the two sources, a 200 microcurie (Uci) cesium-137 source and '
a 5.8 Uci americium-241 source, were immediately inventoried and leak checked satisfactorily (less than minimal detectable activ,ity). The missed source check was -
>
q
>
Il
'
'
,
,
,
,
... _..
.
'
.. o..
=
.
,
.
'
identified while the licensee was preparing for a Quality Assurance audit. The event was the result of a procedural weakness. The radioactive test source procedure (6.6-010) did not provide a consolidated list of sources maintained by, the radiological operations support (ROSD) and chemistry divisions (CD). The procedure was performed by the ROSD and _the missed sources were used by the CD. The procedure is being revised to ensure effective control _ of radioactive
,
sources. The LER fully addressed the reporting criteria including a previous similar event. The inspector had no further questions regarding this LER.
5.5 LER 90-13 LER 9013, " Manual Scram Due to IAckup of the Feedwater Regulating Valves,"
addressed the September 2,1990 feedwater regulating valve control system failure which necessitated a manual scram from approximately 100% power and the subsequent equipment and functions that occurred when the licensee attempted to -
reach the cold shutdown condition. This event has been documented in the inspection report as well as Special Inspection Report 50-293/90-21. The LER-accurately detailed the event and provided accurate technical descriptions of the -
individual events. The inspector had no further questions regarding this LER.
5.6 LER 90-14 LER 90-14, " Automatic Closing of the Primary Containment System Group Ill Isolation-Valves While Shutdown," addressed the September 3,1990 automatic isolation of the shutdown cooling (SDC) system immediately.following system initiation. This event is documented in 'section 3.l'5 of this inspection report.
.
The LER appropriately addressed the reporting criteria as well as identified similar previous events. The inspector had no questions regarding this LER.
l 5.7 Review of Periodic and Soecial Reports.
Upon receipt, the inspector reviewed periodic and special reports submitted pursuant to Technical Specifications. This review verified, as applicable: (1) that the reported information was valid and incitided the NRC-required data; (2).that test results and supporting information; andL(3) that planned corrective actions were adequate for resolution of the problem. The inspector also ascertained whether any reported information should be classified as an abnormal occurrence.
The following reports were reviewed:
l l
b
-
..______..!.._...._._.
....
~.
.
.
'
.
.
20 Monthly, Operational Status Summaries for August and September 1990.
--
Operations Review Committee and Nuclear Safety Review and Audit
-
Committee Meeting Minutes.
6.0 Engineering and Technical Supoort =
'
6.1 E; actor Core Isolation Cooling Discharge Check Valve Design Deficiency
The reactor core isolation cooling discharge check valve CK-1301-50 failed to fully close following the second RCIC trip during the September 2 event causing a momentary pressurization of the RCIC pump suction piping. Engineering analysis of the affected piping is documented in Section 6.2.
Licensee as found Dek' observations indicated the check valve was approximately 1 inch from the close seat. - Minimal torque was applied to the external test shaft to fully seat the valve. The licensee, with vendor support, disassembled the valve, and conducted a detalied component identification and dimensional specification verification.
Check valve CK-1301-50 is a four inch carbon steel testable swing check valve manufactured by. the Anchor / Darling Valve Company. Licensee root,cause investigation identined a valve design deficiency as the failure mechanism which.
prevented full valve closure.
The valve shaft key which provides torque-translation capability for external testing was observed to be loose in its keyw'ayi and also was observed to be contacting the valve packing and bushing. This.
j interference caused increased frictional force contributing to the inability of the
"
valve to fully close. Additionally, a knob or dog which could be utilized to provide valve position indication (but is not utilized) was also determined to have -
the potential to impede valve closure.
The licensee modified the shaft key and securely staked it to the shaft. The key'.
length was, educed decreasing key to packing and bushing interference potential.'
The knob on the shaft bushing was also removed.
The vendor concurred with the licensee determination of a valve design deficiency.
as well as with the licensee modifications to the valve. The licensee is currently assessing the reportability of this design deficiency with respect to the criteria of;
,
10 CFR 21. The licensee indicated that as a minimum a letter is planned to bel
.
submitted to the NRC describing the observed deficiency, i
- l
>
I
.
<
'
.
,
,
'
b
.
21 During the previous April 12, 1989 RCIC pump suction piping pressurization event, the check valve also failed to properly seat. The cause of that failure was determined to have been shaft binding caused by the lodging of residual leak seal -
material between the valve body and shaft bushing. NRC Inspection Report 50 '
293/89 80 provided detailed documentation 1of the April: 12,1 1989 event.-
Therefore, although the failure of the CK-1301-50 valve to properly' seat occurred in both events, causal analysis determined the failure mechanisms of each event J j
were not closely related, j
The licensee demonstrated sound toot cause investigationiprocesses1in the l
development of the design deficiency deterntnation. As-found data'was properly; observed and recorded.
The-correctise modifications :were effectively-implemented and received vendor concurrence. Startup.RCIC system testing included full flow injection to the reactor vessel which successfully demonstratedL j
that the check valve properly opened and fully seated. ' The inspector had no l
further questions.
,
6.2 Analysis of Reactor Core Isolation Cooling. System Suction Piping Pressurization As was previously discussed, the reactor core isolation' cooling (RCIC) system suction piping experienced a momentary pressurization condition when the'RCIC
,
discharge check valve (1301-50) failed to fully seat after the second unsuccessful
'
system start attempt following the September 2-manual reactor scram. The -
overpressure condition was not noted during the event but mther was identified several days later during MDAT evaluation'of the emergency plant information computer (EPIC) charts of the September 2 RCIC start sequences. The RCIC
,
pump suction pressure trace indicated approximately a' 90 psi per second pressure
increase immediately following the RCIC turbine trip.,The pressure transient was
'l less than fifteen seconds in duration and was terminated by the closure of RCIC
'
injection valve MO-1301-49. Additionally, EPIC data indicated that the suction thermal relief valve PS-1301-31 lifted and relieved to thejfloor drain system as designed.
The low pressure RCIC system suction piping is six inch dia_ meter, schedule 40 i
A-106B-carbon steel with a minimum allowable stress of 15,000. psi, a minimum -
required yield strength of 35,000 psi and an ultimate tensile strength lof;60,000
psi. The suction piping design service conditions are 80 psig and 170 degrees F.
The RCIC system was designed to ASME B31.1' code criteria.
On April 12, 1989, the licensee experienced a similar but more severe.RCIC'
system suction piping pressurization event. ' Duringithe April 1989 event, the
suction piping was instantaneously overpressurized which resulted in a " sonic'
wave affect" or hydrodynamic pulse within the piping system. The engineering analysis of the April 1989 event assumed a 0.001 second pressurization time with a peak pressure of 900 psi. The current pressurization event pressure rise rate was 90.38 psi per second based on a calculation of the slope of the EPIC RCIC pump suction pressure trace. At this rate it would take'a pressurization-time of
..
.
_
_
._.
_
_ _ _ _ _ - _ _ _ _ _ _ _
.
x
...
'
y
)
t
.
- 22 M;
approximately ten seconds to reach the postulated 900 psi suction piping pressure.
l However, the licensee conservatively assumed full pressurization occurred in one-i half second (0.5 seconds) which 'would still result in a pressurization time that a
would be 500 times slower than that the of April 1989 event. Therefore, review of the current event data and related calculations clearly indicated a RCIC system
[
suction piping " sonic wave affect" from instantaneous overpressurization'did not.
!
occur. Additionally, licensee calculations determined that the circumferential and
?
longitudinal pressures combined with deadweight stresses were well within design j
basis code allowable stress.
-i in conjunction with the above calculations, the licensee conducted a visual-j inspection of the structural integrity of the RCIC system which revealed no J
damage. Magnetic particle examination was performed at the calculated highest '
stress weld area on the suction piping which identified only manufacturing defect indications, q
Based on the above calculations, analysis, and examinations the inspector concluded that RCIC system suction piping design. specifications were not i
exceeded by the September 2 pressure transient.'
.>
t 6.3 Leak Repair of Shutdown Cooling Suction Isolation Vaive MO-1001-50 :
'
During the course of the September 2 17, unscheduled outage, the licensee applied a temporary leak repair technique to the shutdown cooling system inboard suction
,
isolation valve, MO-1001-50.H Previously, the valve was experiencing a pressure
.l seal ring area leak. On. September 4,1990 the licensee issued maintenance
request (MR 90-1043) to initiate the repair activities. The temporary repair was
I to be accomplished ay the injection of a leak sealant compound in accordance with the directions of Phnt Design Change (PDC) 89-49 and its revisions, consistent l
'
with this associatec safety evaluation.
Two key parameters of the safety evaluation ~ included the initial reactor coolant system (RCS) pressure necessary to facilitate sealant compound injection and the p
maximum volume of leak sealant compound to be injected.- The. initial RCS.
,
I pressure was significant to prevent sealant compound from entering the RCS. The l _
The maximum volume of. sealant compound to be injected was significant to
y safety evaluation required a minimum RCS pressure of 200 psig prior to injection, ensure it would not exceed the volume of the voided pressure seal ring area and potentially infiltrate the RCS.
'
!
.!
,
1 i
'
-
._
_..
_
l=-
.1
.
.
-
.
.
.1
'
23'
Nuclear engineering department procedure, 3.0.3, " Field Revision Notices,":
' i'
L requires cognizant division managers or designated engineers to review field
,
l revision notices to PDCs to determine if such changes are major or minor and.
safety related or non-safety related. The procedure defines a major FRN as one -
!
that: changes the original conceptual design or the intent of the implementation '
document, or affects the bases on which the approved safety evaluation was made.
I Contrary to the procedure atiove, the licensee issued and dispositioned, as minor, l
two FRNs to PDC 89-49 that affected the bases of the original safety evaluation; -
On September 1;, the licensee issued minor field revision notice FRN 89-4915, which among othe.- things, authorized the injection of scalant compound under.
,
cold conditions of 25 +/- 5 psig. On September 23, the licensee issued minor j
FRN 89-49-19,- which authorized injection of an additional 100 sticks of leak scalant which caused thytotal authorized volume of sealant compound to be.
,
injected to exceed the voiced pressure seal ring volume Upon licensee identification of these issues, Failure 'andLMalfunction Report
<
(F&MR)90-340 was issued on September 23,;1990. The engineering department l
revised the safety evaluation to PDC 89-49 to analyze the impact of the two FRNs -
,
of concern. ' The evaluation utilized'the conservative assumption that all scalant compound injected entered the valve bonnet void' area'. The total: volume of-
.
l sealant injected;was 197 cubic inches and the voided bonnet dea volume was
'
approximately 8,000 cubic inches. The evaluation concluded that the relative ratios of the two volumes, in conjunction with close tolerance clearances the sealant compound would have to traverse to enter the bonnet void area, are such that the operation of the valve was not impacted.
The engineering department initiated' an aggressive investigation into this event
,
utilizing a department critique matrix. The initial investigation results determined.
"
that the major nature of the two FRNs:of concern was not identified due to personnel error on the part of the reviewing cognizant engineer. Engineering
.
management developed a lesson plan.to reinforce FRN. review criteria and a department. quality' memo was issued.- A deputy division manager was also assigned to the fluid system and mechanical components division to improve -
supervisory oversight.' Additionally, the NED division manager instituted a~ policy
,
to review all temporary leak seal PDCs and FRNs.
'
Notwithstanding licebsce response to this event, the failure to properly evaluate j
and approve this design change is considered a violation (50-293/90-20-02).
j J
i i
"
i
.
>
'
l f
-
,
,-
y
-
-
._
,
_
_
-._ _
_
-
-
-
.' '
.-,..
'
.'
'
'
7.0 Management Meetings (IP 30702,30703)
7.1 Routine Meetings At periodic intervals during this inspection, meetings were held with senior plant
.
management to discuss licensee activities and areas of concern to the inspectors.
On October 31, the resident inspector staff conducted an exit meeting with BECo
management summarizing inspection activity and findings for this report period..
No proprietary information was identified as being included in the report.
l 7.2
" A" Recirculation Pump Motor-Generator Set Conference Call
,
On August 23, a conference call was conducted between technical staff members :
t of NRR, NRC Region I and the licensee' to summarize licensee actions in response t
to ?.he repetitive automatic trips of the " A" recirculation pump motor generator set l
experienced since October 1989,~ and to update current status. An'NRC synopsis of this issue was provided in Inspection Report 50-293/90-15.
l t
7.3 Conference Calls and Meetings in Resoonse to the Scotember 2-3 Event.
,
On September 4, NRC Region I management initiated a conference call with the-
,
licensee to express concern with component failures and system malfunctions that :
!
presented operational challenges prior to and following the September 2 manual-
'
reactor scram.; Region I was provided a current status of the licensee multi-disciplinary analysis team -investigation.
Additionally,1NRC management
'
informed the licensee that a Region I special inspection team was being dispatched
.
to the site to review the event.
,
On September 7, following completion of the special inspection;.NRC Region 1 -
'
management initiated a conference call with the licensee to cicuss scheduling of
,
a Management Meeting to address the event and to gain an agreement from the
licensee to maintain the plant in a shutdown status-until the meeting was conducted. The licensee provided agreement and the management meeting was scheduled for September 12.
'
On September 12, NRC Meeting Number M-90-114 was convened in the Region
I' office to discuss BECo corrective actions and restart planning related to the
'
September 2-3 event. ' A -list of attendees is included as Attachment l~ to; this-L report. - Mr. Charles Hehl, NRC Region 1 Director of Reactor Projects, opened
!
the, meeting with a brief introduction of the topic.followed-by a detailed i
description of the NRC concerns relative to the event. :Mr. Ralph G. Bird, BECo Senior Vice President of Nuclear Operations, provided an overview of the licensee
!
planned presentation. BECo representatives distributed the prepared overhead displays which are included as Attachment II to this report.
.
!
,
.
_ - -
-
.
o u
q
. -
,
, _,
<
.
j
-
'
'
25-The licensee then made a presentation of the September 2-3 manual reactor scram ;
j and plant shutdown. The presentation included a detailed review of initial plant conditions, immediate operator actions and extended operator actions to maneuver :
!
the plant to cold shutdown._ The licensee described the MDAT composition?
charter, and activities and conclusions to that point in time. Additionally, the; j
licensee provided a comprehensive review of each component failure and system j
.
'
malfunction; identified failure mechanisms and causal determinations which
- included programmatic implications; and identified short and long term 'correctiveY j
~'
actions.
>
The licensee concluded the MDAT investigation continued:to be functioning effectively. - Additionally, senior licensee management stated that MDAT function,
and plant startup were dependent on satisfactory task completion'not by critical +
date scheduling. - The licensee projected plant readiness for restart'during.the'
u upcoming ~ weekend (September 15-16).. A period of NRC' questioning' and requests for clarification and explanation followed the licensee presentation, j
At the conclusion'of the meeting, Mr. Hehl requested the licensee extend the.
.!
agreement.to not restart the plant until a conference call could be conducted on September 14 at which time the licensee.would present a summation of the MDAT investigation and an Operations Review Committee recommendation to the
,
Station Director to restart. The licensee provided agreement and the meeting was adjourned.
'
On September 14, NRC Region I management initiated the scheduled conference
'
call 'with the licensee. > The licensee presented closure of the remaining key MDAT items and identified six minor items to be completed prior to startup. The
.
NRC Region I management concurred with licensee action and terminated the l
agreement to maintain the plant in shutdown.
The licensee subsequently j
completed the six open items and presented them to the resident inspector before -
[
'
plant startup on September 17.
,
-i F
l
4 i
)
i
'
.
o.,
j
,.
'
.
'
ATTACHMENT I September 12,~ 1990 Meeting Attendees
'
Boston Edison Company - NRC BECo
>
R. Anderson, Plant Manager R. Bird, Senior Vice President - Nuclear W. Clancy, Acting Technical Section Manager
I G. Davis, Vice President - Nuclear Administrator R. Fairbank,- Nuclear Engineering Department Manager.
L. Olivier, Operations Section Manager.
E. Robinson, Manager Nuclear Information Division G. Stubbs, Maintenance Section Manager R Swanson, Regulatory Affairs Manager l
'
NRC/NRR R. Conte, Chief BWR Section, Division of Reactor Safety (DRS)
l C Hehl, Director, Division of Reactor Projects (DRP)
W. Hodges, Director, DRS J. Macdonald,-SRI Pilgrim-E. McCabe, Acting Chief, Projects Branch 3, DRP.
T. Martin, Regional Administrator J. Rogge, Chief, Projects Section 3A, DRP R. Wessman, Project Directorate I-III, NRR
,
-
I
\\
l l
!
<:
. -... _ _ _ _ _ -
_ _ - _. - _ - - ____ - __ - _ __ __ __
..
_ _. -.
_
- _
.
. _.
,
l
-
_
J
.-:
i-
.
.
-.
g
...
- fr
,
.;
-
l:
-i
.
o i
.
I Hi i
,
i
!
Bosten Edison Company l
'
Pilgrim Nuclear Power Station j
,
i
,
,
,
-
a i
i g.
i as
o J
September 2,1990 - Plant Shutdown
=
s
~
-e+.
!.
i
.
'!
'
I.
I
!
]
'
.
.
..
-,
..
_-
.
....
_
._
,,.,,
, -...
, -
....
.,,
,
-- -. -- - ---- -.-- -- - _ - - - - -
- - _ -
_
-.
-
- __. -
_-_
--
.
.
.._
.
.
.-
,
,-
,
..
.
t
..
-
...
.
i
- :
I Discussion Topics
.
,
'
j i
.
-
i;
'
-
Findings of Multi-Disciplinary Analysis Team (MDAT)
-
a l
~ -
Basis for Confidence in Plant Equipment l
.
Conclusion and Projections for Restart
.
.
-
-
.
-
.}
-
-!
'
.l l
!
'!
-i i
.
i i
.
'a f
.4
--
.n
,. _
.
_,._. _,. ~
_
_-
.
.-.....,... a
._,,
.
., -
. -. _, _. _ _.. _,.~ _
.
--
- _
_ _
.
_
._
-
. - -.
.
..
,
e.
y
..
-
i
,
.
!
i i
Plant And Personnel Safety Was Maintained i
Throughout The Event
,
Water Level Transient Minor and Well Controlled i
-
!
,
..
-!
Operators Stabilized the Plant.and Responded Properly to
-
Equipment Malfunctions
._
..
,
Operstors.PerformedLWell
-
j
-
<
'
.
'
<
i
'
!
'
'
,
!
! _ _.
.. _ _
_
-._
_.
. _
.. _
.-
-.,
, _
_
,._,.
._.
-
..-,_~.,,,...m...
....
.....
_
....
-
..
..
.
.
MDAT Focus Is Appropriate
Identified Key Problems
-
Root Causes
-
Not Constre.ined.by Schedule.
-
~
Access to Available: Resources
' '
-
ORC, NMC; Oversight
-
'
,
c
,
.,.
. -.,
.,...R.
-.,:-'
,.-Y^.-.~..L-----
~ - - - - - - -
", - - -
' + - - - -
- - - - - - - - - - - - - - - * - + " - - * - ~ - - ~ - ' - - - ~ ~ ~ ' * * " " ' - " ' ' ~ ' ~ ' - ~ " ' ' '
-
-
-
- N
.**
- .
..
,
-
.
Feed Regulating Valve Control System
.
identified As Source Of Transients Feedwater RegulatingValve "B" Packing
-
Feedwater Reg Valve Lockup
-
.
Startup Feedwater Regulating Valve Air Booster Relay
-
.-
-
.
.
__
-
_
-
. _
_
... _..
__
-
.
_
..
,
-
.,
-
v
.-:.
-
i
-
..-
j
!
.
-
,
,
!
,
'h I
RCIC ISSUES
!
i Overspeed Trip MechanismY,y#
!
.
.
,
,
Trip Throttle Valve Linkage
-
.
p
.
.
Injection Check Valve).
.
.
d i
-
.
-
.
d#
a ~-
'
yf th '
I
'
i-l
'.
!
y
,
,.
l j-i
-
y..
. _._
-. i.. a :. _. _ ',
.-.
...y...
, _,
-
~.
_
., -.,.... _. _ _
_.__J.,
_ _ _ _ _ _ _ _ _ _ _ _, _ _ _ _. _ _. _ _ _ _.__
,
.
..
.
.
i,Il - '0.-
t o s
.
- -
.?
..,
- y
.
..
'
,
'
e l
i k
t l
l-(
i --
,
t q
I (1 (
t
>
l l-
,
'
4p to
,
l CC
m Mh
!
g a
.
.
.-
._.
M b-
o H
w-
- E I
C l;
>
$e
O
>
i
-
~
>
O
,
L e
e i
!
i l
-
i
.
'.
l
'
-e
.-
e n
,
- - -.,,,
-
.,,w-,--,
, -. +,,-,.
-
r,
. -,,,
,
...,
.
,.
.)
.
o
!
.-
.
-
i RHR/ SHUTDOWN COOLINGISSUES j
.
,
!
!
Manual Control Circuit - Inboard Shutdown Cooling
'
,
.
Suction Valve
i-l
.-
,
-Automatic Shutdown Cooling Isolation
.
..
i
'
.
_
.
- a
,
-
~
_.
I
h i
!
e h
.!
..
.
.:_._ _
_ ;.. -
-
_
. u. -
._
,
,_
., _
._._.,:--
, _.. _ _ _...
..
.:
.-
-
--
- - - _
-__
_-
__
- -_. _ _-
_-_ _ __ -
-
- _
.....
,. # :;
.
..
.
..
l
-
}
ACTIONS THAT HAVE BEEN/WILL BE
ACCOMPLISHED PRIOR TO RESTART
i
-
e
RCIC Oil Flush / Inspect
.
.
.
.J
.,
.
RCIC Linkage Disassembly / Adjustment
- l r
.
~~
HPCI. Oil Sump Clean / Flush '
' l
.
.
l
'
,
RCIC Check Valve Repair / Test -
i
..
I
'
I
'
t
Shutdown Cooling Valve Relay / Contact
{
.
j Replacement:
o
Correct FRV Water Damage
.
i.
l l
r i
.
..
,
..,,
,
-
--J.'.,
,
.*--y
-.
-
,.
...s,~,,,
g
,g
-___y
.
.
'
.
.
-
.
.- -
O
-
'
.
O ACTIONS THAT HAVE BEEN/WILL BE ACCOMPLISHED PRIOR TO RESTART (cont'd)
FRV Repacked
.
.
RCIC Suction PipeInspection
.
Procedure Revisions
.
Appropriate Restart Testing -
.
-
-
-
-
- -
-
. - -
-
.
-
.
=. -.
. -. - - - - -
_ -. _
_,,
_-
.
.
-
..
,...
.
..
~.
.
,
.
.
,
i We Have Confidence In System. Reliability l
i
'
i i
l
- -
-
!
Workmanship. Performance Review
,
-
i Successful Loss of Offsite Power Test
.
l
-
$
Maintenance Personnel Qualification Review
-
,
,
System Engineer Review L
-
-
i i
-
'
Augmented Testing-
-
.
I t
t T
- i
!
i
,
. i s
..
i
~
'
... -. -;.,.
,. -._.~ -...~.-.. :
.
.
..
-L
--t
. -..-
- -...
. _.
._.
-
_
_
. _.
_
_ _ _
_ -
.
l
..s
.
,.
i
'
~
-
t
~
,
'
.
,
!
Acccelerate Ongoing Long Term Improvements
.i
.
Predictive Maintenance
.
Preventive Maintenance
.
.
Technical Support
.
.--
_
.
= _.
. _.. ~. _.
.. _..,
.,
,
._ _ _ ___ _,__ __ ____
.
__.,_,__
m
.
,
-..
. _.
.
_
-.
. _.1 a
-
....e
'
,, _ ;
_.
'
.
,
.
.
.,
!
CONCLUSION
i
!
I
..
t MDAT Properly Focused and Thorough
.
,
i
..
Equipment Issues Being; Resolved jl
.
.
y
-
_
Confidence in System Reliability
.
.
l Accelerate On-Going Program Improvements '
.
l
!-
-
t
.,
.
-
-.i
,
~'
T.
....-.
-....
- _ -,,
-
.
.-
-
_
,.
_