IR 05000277/1990001
| ML20012E447 | |
| Person / Time | |
|---|---|
| Site: | Peach Bottom  |
| Issue date: | 03/26/1990 |
| From: | Doerflein L NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20012E444 | List: |
| References | |
| 50-277-90-01, 50-278-90-01, NUDOCS 9004050167 | |
| Download: ML20012E447 (31) | |
Text
_ ____ __________
c
.
'
.
!
-
!
i U. S. NUCLEAR REGULATORY COMMI$$10N l
REGION I
Docket / Report No. 50-277/90-01 License Nos. DPR-44 50-278/90-01 DPR-56
[
Licensee Philadelphia Electric Company Correspondence C.'ntrol Desk J
P. O. Box 7520 r
Philadelphia, Pennsylvania 19101 t
i Facility Name:
Peach Bottom Atomic Power $tation Units 2 and 3
!
Dates:
January 1 to February 20, 1990 f
Inspectors:
J. J. Lyash, Senior Resident Inspector
!
R. J. Urban, Resident Inspector
L. E. Myers, Resident Inspector
[
0. T.
Operations Engineer
'
Approved By:
[. / oc 3/M/9O
'
oe L. T. Doerflein? Chief Date
'
Reactor ProjectJ Section 2B
!
!
Division of Reattor Projects l
Areas Inspected:
[
,
Routine, on-site regular, backshift and deep backshift inspection of accessible portions of Vait 2 and 3.
The inspectors reviewed operational
!
safety, radiation protection, physical security, control room activities,
'
licensee events, surveillance testing, engineering and technical support
activities, and maintenance.
The inspection identified two violations of NRC i
requirements, t
i l
l
i f
!
!
,
f
'
i
i I
i l
I
"
9004050167 900327 i
{DR ADOCK03OOg77
'
r
.,.
_
_. __
i
_ -.. _ _.. _ _ _ _ _ _ _ _
. _ _.
. _ -
_ _ - _ _ _ _ _ _ _ _ _
.
.
.
.
,
Executive Summary PeachBottom Atomic power Station
>
Inspection Report 90-D1 l
,
Operations
[
,
1.
Performance of the control room staff during the prompt plant shutdown f
following discovery of a Unit 3 electro-hydraulic control system fluid j
leak was commendable. The transient revealed several training and
,
hardware weaknesses which will be addressed by the licensee ($ection 2.4)
L
!
2.
Operations shift management did not implement actions required by l
Technical $pecification (T$) 4.5.E.2 following the failure of an
,
i automatic depressurization system (AD$) surveillance test.
Test
!
engineers and operators involved did not recognize the failure as
!
affecting system operability, and therefore did not implement the i
required compensatory measures (NC4 90-01-01, Section3.0),
j 3.
The control room staff did not trip the associated primary containment
}
isolation system logic subsystem for 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> following discovery of an
>
inoperable instrument. While Technical Specifications do not include a (
specific time limit, 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> is longer than necessary. The licensee
!
will evaluate the incident and find the root cause (UNR 90-01-03,
'
Section 6.1).
l Maintenance / Surveillance:
f i
'
1.
The licensee's system for scheduling of surveillance testing may not recognize and highlight some tests completed outside the required TS
interval. The inspectors did not identify any examples of actual
!
overdue tests.
Licensee review and documentation of the basis for i
acceptability of the present scheduling system is ongoing (UNR 90-01-02, t
Section 5.1).
i
2.
A Maintenance Failure Analysis Report concerning the failure of main
!
steam line low pressure transmitters did not include consideration of l
potential generic implications of the failt e. Also, the licensee did
!
not update the anal,.is to reflect addition;1 failures occurring at the
'
I plant.
Licensee sta'f stated that they had considered these. factors,
i although they did not documented it ($ection 6.1).
'
t Engineering / Technical Support:
1.
The licensee performed comprehensive, high quality technical evaluations l
in support of the emergency change to the ADS Technical Specifications l
,
($ection 3.0).
- l l
t
f,
'
,
.
!
,
I
_ _ _ _ _ _ _ - _ _ _
_ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _.
. _______ _ ___ ______
_ _ _
.
.
.
.
.
Executive Summary Continued
Radiolootcal Controls:
1.
During the inspection two licensee engineers entered a high radiation area without adhering to the requirements contained on the applicable radiation work permit. They received an unexpected dose of about 200 millirem (mr) per individual.
The licensee's follow-up and corrective actions were comprehensive (NV 90-01-04, Section 7.1).
2.
The inspector identified that the licensee procedure for establishment and control of portable in plant air monitor setpoints was incomplete and not effective. This may be indicative of other weaknesses with the completeness and clarity of radiation protection program procedures (Section 7.2).
Assurance of Quality:
1.
The inspectors identified that the licensee's Annual Report of Safety Relief Valve Challenges for calendar year 19f>9 was inaccurate. The licensee submitted a supplemental report and implemented corrective actions to prevent recurrence (Section 9.2),
i I
h
_ _ _ _ _ _ _ _ _
_ - _.
_.
.__
. - - _ - _ _ _ _.
.
!
.
.
.
.
TABLE OF CONTENTS
.P *M
-
1.0 Plant Operations Review (71707)
1.1 Unit 3 Thermal Limit Operating Restrictions.................
1.2 Unit 3 Off Gas Activity Increase.............................
2.0 Follow-up of Plant Events (93702, 40500, 35502)
2.1 10 CFR 50.9 Report..........................................
2.2 Unit 3 Shutdown Due to Emergency Core Cooling Systems Inoperable..................................................
2.3 Unit 2 Shutdown Due to Feedwater System Instrument Line Leak
2.4 Manual Unit 3 Scram Due to an Electro-Hydraulic Control Fluid Leak........
.................................................
3.0 Emergency Technical Specification Change for the A05.............
4.0 Engineering and Technical Support Activities (37828, 40703).
5.0 Surve111 ante Testing Activities (61726, 71707)
5.1 Surveillance Test Scheduling Controls.......................
5.2 Routine Observations........................................
6.0 Maintenance Activities (62703)
6.1 Unit 3 Main Steam Line Transmitter Failure..................
6.2 Routine Observations........................................
7.0 Radiological Controls (71707, 83750, 84750)
7.1 Unplanned Personnel Exposure...............................
7.2 Routine Observations.......................................
8.0 Physical Security (71707,81700)...................................
9.0 Review of Licensee Reports (30702, 90712, 92700)
9.1 Licensee Event Reports.......................................
9.2 Routine Licensee Reports....................................
10.0 Licensee Response to NRC Generic Letter 88-20, " Individual Plant Examinations for Severe Accident Vulnerabilities" (92703).........
11. 0 Mana gement Meeti ng s ( 30703).......................................
i
.
.
.
O
.
DETAILS 1.0 plant Operations Review Unit 2 entered the report reriod operating at 100% power. The licensee completed one forced shutdown due to a leaking instrument line on the feedwater pump discharge header on January 27.
Following restart and ascension to 100% power, a feedwater heater leak required power reduction to 74% and isolation of one feedwater heater string. The unit will remain at that power level until the planned maintenance outage in March.
Thermal lirnit restrictions required Unit 3 power to be held at 99% for much of the inspection period.
On January 28 a forced shutdown was caused by an electro-hydraulic control system leak.
Following repair of the leak the unit ascended to 100% power. Attachment I is a detailed chronology of plant events occurring during the inspection period.
The inspector completed NRC inspection procedure 71707, " Operational Safety Verification." This was accomplished by direct observation of activities and equipment, tours of the facility, interviews with licensee personnel, independent verification,of safety system status and limiting conditions for operation, and review of corrective actions and facility records and logs.
The inspectors performed 190 total hours of on site backshift inspection, including 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> of deep backshift and weekend tours of the facility.
1.1 Unit 3 Thermal limits Operatino Restrictions When the Peach Bottom facility was shutdown by NRC order in March i
1987, Unit 3 was operating at 100% power.
Unit 3 was not i
scheduled for a refueling outage until later that year.
Due to the early cycle 7 shutdown, complete utilization of the fuel was
.
'
not possible.
The fuel reload was more reactive than a normal
core reload and the licensee expected that thermal limits would be encountered before the unit could reach 100% power, t
On January 5,1990, Unit 3 reached about 100% power. The next P1 l
program (Periodic Core Evaluation-Thermal Limits) showed that two
.
.
of the three core thermal limits were greater than allowed by
!
'
Technical Specifications (TS).
The core maximum fraction of i
!
limiting power density (CMFCPD) was 1.001 and the core maximum
!
~,
average power ratio (CMAPR) was 1.004; both TS limits are 1.000.
!
I l
As required by TS 3.5.1 and 3.5.J. the licensee reduced reactor
!
i power to about 98%, to lower the core thermal limits below 1.000.
l l
l I
i
,
.
.
.
.
.
The licensee Reactor Engineering and Fuels Management organizations issued a memo to operations personnel that explained the reason for the thermal limit restrictions and recommended running a P1 program every 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />.
The memo also recommended keeping the thermal limits below 0.995 so that an adequate margin to the TS limit would be maintained.
By the end of January, fuel burn-up and a slightly different rod pattern reduced the reactivity peak of the core. The unit reached full power while maintaining core thermal limits below 1.000. The inspector concluded that 11censeo response and verification actions were prudent and effective.
1.? Unit 3 Off Gas Activity increase The licensee is required to monitor radioactivity in various systems, including the reactor coolant and the off gas systems.
During startup and power ascension of Unit 3, the licensee compared reactor coolant activity with off gas activity to verify agreement. The off gas activity was fluctuating between 2000-4000 microcuries per second, and showed a lower activity level than implied by the reactor coolant samples.
The licensee initiated troubleshooting of the off gas sample system.
A leak was found in a rotometer on an off gas monitoring line.
Once the repair was complete on January 19, off gas activity increased to about 6500 microcuries per second.
The control room received an "Off Gas Hi Radiation" alarm and operators entered Off Normal Procedure ON-102, " Air Ejector Discharge High Radiation."
Since the radiation levels were well below the action setpoint, operators monitored off gas and ventilation stack radiation levels as described in the procedure. The previously fluctuating low values were due to dilution by air in-leakage; once the licensee sealed the leak off gas level increased.
The operators exited the procedure when they determined the cause of the alarm.
Between February 7 and 11 off gas activity increased to approximately 14,000 and later subsided to 10,500 microcuries per second. The licensee believes that one and possibly several fuel pins are leaking.
Licensee analysis suggests that the activity is a result of about 75% diffusion and 25% equilibrium.
The licensee raised the off gas alarm setpoint to 1.5 times the background to clear the alarm.
This ensures that an additional alarm will occur if activity continues to increase. Also, the licensee formed a task force to track and trend the apparent deteriorating fuel leak.
In addition, off gas system grab samples are analyzed daily.
The licensee is taking prudent action.
The inspector will continue to follow this area as part of the routine inspection program, f
.
.
_ _ _ _ _ _ _ _ _ _. _ _ _. _
.
.
O
'
2.0 Follow-up Of plant Events The inspector evaluated the licensee response to plant events to ensure that analysis was performed, reasonable root causes were identified, and appropriate corrective actions were implemented.
In each case, the inspector reviewed applicable administrative and technical procedures, interviewed personnel and examined the affected systems and equipment.
2.1 10 CFR 50.9 Report During follow-up to a design basis weakness identified at the
'
Limerick Generating Station, the licensee determined that transient response procedures implemented following a reactor
,
scram potentially place the plant outside the design basis.
The Peach Bottom FSAR describes the plant response to a main turbine electro-hydraulic control (EHC) system failure in the high direction.
The FSAR states that a main steam line (MSL) low pressure signal is generated, closing the main steam isolation
>
valves (MSIVs), causing a reactor scram and terminating the
.
depressurization.
Recent analysis by the licensee determined that the EHC failure would result in reactor vessel level swell, a high reactor vessel level turbine trip and a reactor scram.
The scram would occur well before reaching the MSL low pressure isolation setpoint.
Scram response procedures direct the operator to place the mode switch in shutdown immediately, bypassing the MSIV isolation on low MSL pressure.
The turbine bypass valves would remain open due to the postulated EHC failure, continuing the reactor depressurization.
If the operator takes no additional action the vessel would quickly depressurize and, due to condensate system injection, would overfill.
The effect of subcooled liquid in the main steam lines, the potential for subsequent two-phase flow through the safety relief valves, and the reactor vessel stresses resulting from the rapid cooldown are of concern.
The FSAR describes the EHC failure as an
" abnormal operational transient."
Exceeding the code allowable reactor vessel stresses for this type of transient is not an acceptable result.
Preliminary licensee analysis shows that the effects of the overfill /depressurization event is bounded by the design basis event analysis. The licensee prepared a justification for continued operation (JCO) to support operation until a comprehensive analysis is completed. The licensee notified the NRC of the problem under 10 CFR 50.9.
The inspector reviewed the JC0 and did not identify any concerns, f
'
.
___-_ __
.
.
.
.
2.2 Unit 3 $hutdown Ove to Emeroency Core Coolino Systems _Ineperable
>
'
On January 8,1990 Unit 3 entered a Technical Specification (TS)
required shutdown because of inoperable reactor core isoittion
-
cooling (RCIC) a..J high pressure coolant injection (HPCI) systems, o
TS 3.5 C.3 required the reactor to be in cold shutdown within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
Surveillance test ($T) 6.11F-3, "RCIC Pump, Valve, Flow and
,
Cooler Functional flow Test," performed late on January 7, showed
,
that pump differential pressure was high and in the Inservice Test
(!$1) Program action range. According to the ST, test measurements in the action range require that the system be declared inoperable and returned to service only after determining the cause of the deviation.
'
According to TS 3.5.0.2, if HPCI is tested operable, continued
reactor power operation is permissible for 7 days.
$T 6.5-2,
.'
!
"HpCI Pump Valve, Flow, Cooler," was performed on January 8
,
shortly after the RCIC ST failure.
HPCI didn't reach design flow
!
a
'
and pressure within 25 seconds and was also declared inoperable.
The licensee initiated a shutdown at this time.
In addition, the -
'
HPCI gland seal condenser (GSC) vacuum pump didn't operate and the
,
GSC condensate pump tripped due to a thermal overload.
,
y
,
RCIC System
.
The RCIC pump differential pressure is highly dependent on pump speed. The indicator in the control room used by the operator to
,
'
adjust speed is only accurate to about +/- 100 RPM, making precise
,
assessment of speed difficult. The licensee reperformed the $T on
!
January 8 using a calibrated strobe light to accurately set pump
speed at 3600 RPM. At that point the control room indicator was
'
reading 3500 RPM. RCIC pump differential pressure was within the i
normal range. The licensee declared the RCIC system operable and
.;
terminated the TS required shutdown.
The licensee revised the $T i
to require the use of a strobe to set pump RPM accurately, i
Therefore, more consistent data will be ensured.
The inspector
.
'
reviewed the failed and successful performances of $T 6.11F-3, the
{
pump data sheet, and the revised $T and had no further questions
!
in this area, i
With the HPCI system still inoperable TS action statement 3.5.C.2 I
allowed reactor operation for 7 days if testing showed that other i
core and containment cooling systems (CCCS) were operable.
The
!
CCCS testing completed on January 8 was successful.
Licensee
troubleshooting of the three HPCI system problems is discussed l
below.
'
i l
l f
i (
'
..
l
.
.
.
.
HPCI Flow The licensee believed that the HPCI system didn't reach design i
flow and pressure within 25 seconds due to problems with the ramp generator signal converter (RGSC).
Using a troubleshooting control form, the licensee measured the RGSC starting voltage and the ramp time (18 sec.).
Both values were within specifications.
The licensee performed a complete diagnostic check of the turbine governor using surveillance test SI 3M-23-RGSC-XXCO, " Calibration Check of HPCI Turbine Governor (EGM, RGSC)." The check identified several performance irregularities and the licensee replaced the electronic governor mechanism (EGM) and RGSC.
During troubleshooting activities the licensee also questioned the I
established RGSC ramp range of 15-20 seconds.
In 1983 a lube oil bypass line to the HPCI governor was installed, Its purpose was to smooth the HPCI start by producing a more linear acceleration rate. This lengthened pump start times slightly.
At that time General Electric Service Information Letter (SIL) 351 recommended a RGSC ramp range of 10-15 seconds.
The licensee used this range until February 1989, when the time range was changed to 15-20 seconds due to an error during revision of the procedure.
Later, reyision 1 to SIL 351 recommended a range of 15-20 seconds.
This recommendation was intended for utilities that had not implemented the oil bypass modification, but the SIL did not discuss this aspect. This reinforced the previous error. When the licensee set the new ramp range pump starts were very near 25 seconds, and the most recent HPCI start occurred in greater than 25 seconds.
!
In a January 8,1990 letter to PEco, GE returned the recommended RGSC range to 10 to 15 seconds.
RGSC recalibration set the ramp time to 12 seconds.
The HPCI system reached design flow and pressure within 25 seconds when tested on January 9.
To declare the HPCI system operable, the GSC vacuum pump and i
condensate pump needed to be repaired or dispositioned. The
!
licensee produced a safety evaluation report (SER), an engineering work request (EWR), and text from the FSAR that stated the HPCI system can perform its design function without the GSC subsystem.
Therefore, the licensee declared HPCI operable on January 9.
Based on a review of the licensee's evaluation the inspector concluded that-the HPCI pump and turbine could function without this subsystem. The inspector was aware, however, that failure of the GSC at other facilities had been determined to cause room temperature and humidity in excess of the equipment. qualification values.
The inspector requested that the licensee review this potential.
The inspector referred further review of this issue to the NRC Safety System Functional Inspection Team, j
>
.
.
.
'
I i
The inspector reviewed P& ids, SI 3M-23-RGSC-XXCO, electrical schematics, GE SILs, EWR P51230, the HPCI and RCI'.' system operability SER, and the FSAR. The inspector also spoke with
,
licensed operstors and the system engineer, and observed I
troubleshooting of the RGSC. The adjustment to the RGSC ramp
'
should prevent this problem from recurring on Unit 3.
This
!
adfustment will be performed on Unit 2 during the unit shutdown i
for its mid-cycle outage in March.
The inspector had no further
'
questions in this area.
,
GSC Condensate Pump i
The licensee believed that the overload problem associated with i
the GSC condensate pump was a worn thermal overload device.
It
,
was replaced and the condensate pump was post-maintenance tested r
on January 9 by running the pump several times with the GSC full of water. During HPCI system surveillance testing on January 17 the GSC condensate pump again operated satisfactorily. But, on
'
January 28 during the Unit 3 shutdown (see section 2.3) during HPCI operation, the GSC condensate pump tripped due to a thermal overload. On January 29, investigation found the GSC condensate pump seized.
The licensee removed, cleaned, and refurbished the pump.
It was reinstalled and subsequentiv M M d op s le Ch February 4.
Apparently, the the m.zi overload problem on January 8 was due to an intermittent seizing problem besides a worn, thermal overload device.
Thermal overload devices in power suppiy breakers are currently on a 5 year preventive maintenance (pM) replacement schedule. Most i
HPCI thermal overload devices are located in their respective power supply breaker compartments. The motor starter and thermal
,
overload devices for several HPCI components (GSC condensate pump,
'
vacuum pump and the auxiliary oil pump) are located in the HPCI room.
The licensee PM program overlooked the thermal overload devices in these breakers due to their unusual location.
RCIC is the only other system with a similar configuration. The system engineer added these HPCI and RCIC thermal overload devices to the PM program. They will be replaced on an 18 month schedule due to the harsher environment in these rooms.
The inspector concluded that placing the GSC condensate pump
'
thermal overload device on an 18 month PM replacement schedule, besides refurbishment of the GSC condensate pump, should prevent similar problems from occurring in the future. The inspector questioned whether the GSC condensate pump was currently in the PM Program. The system engineer determined that it was not and agreed that it would be prudent to add the GSC condensate pump to the PM Program.
The inspector had no further questions in this area.
l
.
--
-
-
.
.
.
,
'
.
GSC Vacuum Pump A short in the motor caused the GSC vacuum pump problem. To
!
repair the motor, the licensee removed the entire blower unit and sent it off site to be rewound.
To allow HPCI to remain operable, the licensee installed Temporary Plant Alteration (TPA) 2301 to replace the blower with hard piping so that vacuum on the GSC would be provided by the standby gas treatment system. After the TPA was in place, HPCI was successfully run to check the hard pipe installation on January 17.
The GSC vacuum pump was reinstalled and tested operable on February 2.
.
The inspector observed maintenance activities on the GSC vacuum pump, walked-down the HPCI system and reviewed nonconformance report (NCR) P90-015, maintenance request form (MRF) package 9000194 and TPA 2301.
The inspector concluded that the TPA was adequate to allow the HPCI system to remain operable with the
,
blower unit removed.
During review of the MRF package for replacement of the GSC vacuum
,
pump, the inspector noted a discrepancy in the safety evaluation
'
- +ta t d t: NCR P90-015. This NCR questioned the removal the GSC condensate pump starting coil from the Q-iist.
The SER stated that if the GSC condensate pump failed, condensation overflowing the GSC would be removed from the HPCI room by the floor drain
,
system. But, a modification completed before restart plugged all the floor drains in the ECCS rooms due to potential common mode flooding concerns.
The inspector noted that the NCR was not final because the Quality Assurance (QA) group review was incomplete.
The licensee stated that the engineers who researched the NCR were not familiar with the flooding modification at the site. QA stated that their review would have found the discrepancy.
During a control room walkdown on January 18, the inspector noted that the "HPCI DC Motor Power Loss" alarm was illuminated. This was due to the blocking permit applied on the GSC vacuum pump.
The inspector pointed out that in this condition a loss of DC power to either the auxiliary oil pump or GSC condensate pump would not annunciate. The start /stop lights for the motor would
,
extinguish providing some indication to the control room operator that a loss of DC power may have occurred. The licensee agreed and revised the blocking permit to disable the GSC vacuum pump input to the alarm.
The inspector concluded that corrective
'
actions associated with repairs to the HPCI system were adequate and had no further questions.
.
.
..
.
.
,
.
,
,
2.3 Unit 2 Shutdown Due to Feedwater System Instrument Line Leak
-
On January 27, 1990, at 3:00 a.m. a nonlicensed operator discovered a leak coming from an instrument line for the flow indication on the "B" feedwater pump discharge line.
The licensee closely monitored the leak for change.
Increased leakage caused
,
the "B" reactor feedwater pump (RFP) discharge indicated flow to
!
increased, and a mismatch between feedwater and steam flow. The
- '
control room operator placed the feedwater byel control system in single element mode, reduced reactor power to 73%, and removed the
"B" RFP from service.
The failure was a crack in the heat
!
'
affected zone next to the 3/4 inch socket-weld at the pump discharge line. The failure was about 300 degrees around the pipe, and resulted in a leak of about 8 GPM.
Maintenance personnel recommended a plant shutdown to perform repairs.
The operators commenced a shutdown.using procedure GP-3, " Normal Plant Shutdown," and manually scrammed from 30% power. The plant responded normally and no isolations occurred. The reactor was depressurized to support the weld repair. QC personnel examined the completed repair, and other welds on the instrument lines connected to the "A " "B," and "C" feedwater pump discharge lines using dye penetrant. The inspections did not find any additional indications.
A leak on a similar weld on the '{" feedwater pump discharge line occurred in December, 1989 as described in Inspection Report 50-277/89-26, Section 5.0.
The lines are subject to vibration of medium to high frequency and small amplitude. The instrument lines are about 36 inches long and include two heavy root valves before reducing to 5/16 inch instrument tubing. Engineering studies initially concluded that the 3/4 inch piping required no support.
On February 7,1990, a crack on another "B" feedwater pump discharge line instrument line occurred after the unit returned to power. The crack was in the heat affected zone next to the butt weld at the root valve.
The crack was small and was repaired at power using leak seal.
Engineering has reconsidered the use of supports far the instrument lines.
They are studying several designs, one of which will be installed during the upcoming mid-cyc1( outage in March. The inspector had no further questions.
2.4 Manual Unit 3 Scram Due to an Electro-Hydraulic Control System Fluid Leak On January 28, 1989, an electro-hydraulic control (EHC) system fluid leak developed, prompting a rapid plant shutdown. At 8:45 a.m. an "EHC Fluid Reservoir Low Level" annunciator alarmed in the control room.
The Shift Manager (SM) immediately dispatched
.'
.
-[
.
.
.
%
.
,
,
auxiliary operators (AO) to assess EHC fluid reservoir level and to locate the leak causing the loss of fluid.
Nonlicensed operators reported a significantly low reservoir level and identified the source of the leak as the number 1 turbine control valve servo actuator. Based on this information the SM decided that a prompt plant shutdown was appropriate.
The Shift Manager directed initiation of a rapid power reduction using procedure GP-9-2, " Fast Reactor Power Reduction." Transient Response Implementation Procedure (TRIP) T-100, " Reactor Scram,"
was entered.
Following T-100, the operator manually transferred
!
house electric loads from the unit auxiliary (VA) transformer to i
the startup (SV) feed bus.
During the transfer the UA transformer
' '
feeder breaker to the #4 auxiliary bus didn't open automatically upon closure of the SU feed bus feeder breaker. The operator recognized the failure and manually opened the breaker a short time later. After reducing power to about 50 percent the mode switch was placed in shutdown.
Following the scram a primary
'
containment isolation system (PCIS) group II/III isolation
-
occurred due to the reactor vessel level shrink.
The operator manually tripped two reactor feedwater pumps immediately following the scram. As directed by TRIP T-99, two condensate pumps were also removed from service.
Although two of the three FW pumps had been tripped, reactor water level continued to increase due to the input of the remaining FW pump and the expected inventory swell.
The level quickly reached the reactor vessel high level trip setpoint of +45 inches' and the
"C" FW pump turbine tripped. As vessel level slowly decreased due to steaming via the turbine bypass valves to the main condenser, the operator attempted to reset the "C" FW pump turbine in preparation for its return to service.
The operator attempting to perform the reset perceived, based on interpretation of the control room indication, that the machine would not reset. The high pressure coolant injection (HPCI) and reactor core isolation
'
cooling (RCIC) systems were placed in service to provide reactor vessel makeup, and the "C" FW pump turbine reset effort was set aside. Also during this time the SM noted that the main turbine steam seal regulator had failed closed and condenser vacuum was
.
trending downward.
Loss of EHC pressure or loss of condenser vacuum would result in closure of the main turbine bypass valves, and a resultant loss of the condenser.
Because of the continuing EHC fluid leak the SM directed the operator to depressurize the reactor using the bypass valves, to about 600 psig.
Reactor pressure decreased from 980 psig to a low of about 420 psig over about a 20 minute period.
During this time RCIC, with periodic j
input from the HPCI system, provided reactor vessel makeup.
l
'
-
,
.
.
'
-
.
i
.
!
>
Level control with HPCI and RCIC while depressurizing was
'
difficult and several level oscillations occurred, resulting in two additional PCIS group II/III isolations.
It appears that the
!
HPCI system was used to " batch feed" during the transient. This
!
exacerbated the level control effort and contributed to
.
'
overshooting the 600 psig reactor pressure target.
In addition, the single inservice condensate pump provided no makeup flow when reactor pressure dropped below 600 psig.
Shortly after reaching 420 psig the SM directed closure of the main steam isolation valves (MSIVs), and the removal of the EHC system from service.
Reactor vessel level control using RCIC and pressure control using HPCI continued until sufficient decay heat had been removed to
,
allow use of normal heat removal systems.
Following the transient the inspector reviewed control room logs, the process computer sequence of events log, control room strip charts, safety parameter display system data, and operating.
procedures. The inspector discussed the event with the SM, the operators involved and licensee management personnel. The licensee's completed procedure GP-18, " Post-Scram Review Procedure," was also evaluated.
Control room staff performance during the transient was commendable. The SM made prompt, conservative assessments of the significance of the problem and established a clear course of action. Operators did not become distracted by the complicating problems arising, but focused on placing the plant in a stable
,
condition.
Described below are the equipment, procedural and l
training weaknesses identified because of the event:
The cause of the EHC fluid leak was failure of a sealing "0"
--
ring between the servo actuator and the body of the number 1 turbine control valve. The licensee's preventive maintenance program includes replacement of servo and "0" rings each
operating cycle. They were replaced during the recently
'
concluded outage.
Licensee evaluation of the failure cause is l
ongoing.
During the attempt to transfer house electrical loads, the
--
automatic trip function of the UA transformer feeder breaker to bus #4 didn't operate.
Logic testing prior to unit restart t
'
ident'ified no problems.
Functional testing of the transfer during plant startup was successful.
No conclusive root cause has been identified. The failure of the automatic trip function does not present a safety concern.
.
O
.
.
,
,
'
.
,
Following the trip of the
"C" FW pump on high level due to
'
--
inventory swell, the operator didn't recognize that attempts to reset the trip had been successful.
This prevented use of the pump for reactor vessel makeup.
Investigation revealed that the trip reset indication for the Unit 2 FW pumps was not operating as designed.
The indication was different from the corresponding Unit 3 and simulator indication. This confused the operator. The licensee initiated a maintenance request to repair the Unit 2 indication circuit, and provided a memo describing correct circuit operation to the plant operators.
'
Licensee TRIP procedures and training call for operators to
--
remove two FW pumps and two condensate pumps from service immediately following a plant trip. The third PW pump is tripped when no longer needed; but, the third condensate pump remains in service to provide vessel makeup when reactor pressure drops below about 600 psig. Operators are trained to close the pump discharge valves when securing the FW pumps, and to open the startup control valve block valve. Opening the block valve allows the startup control valve to regulate condensate pump discharge flow to the reactor vessel automatically.
The operators did not close the FW pump suction valves following removal of the pumps from service.
This resulted in
'
all three FW pump 6 inch minimum flow lines continuing to be a potential condensate flow path back to the main condenser.
Because only one condensate pump remained in operation, sufficient capacity was not available to feed the reactor vessel with these minimum flow lines diverting flow to the condenser.
The licensee determined that operator training in this area was not comprehensive.
The Operations Superintendent stated that additional training in this area would be included in the requalification program.
During recovery from the transient the operator controlling
--
the HPCI system had difficulty preventing " batch feeding."
This contributed to the level oscillations observed and to the magnitude of the reactor depressurization.
No direct indication of HPCI flow to the vessel is available when operating the system in this "nonemergency" mode. One indirect indication of the initiation of HPCI flow to the vessel is the opening of the injection check valve, as described in the system operating procedure.
This indication was inoperable at the time of the event making control more difficult.
The operator involved received additional siniulator training on this mode of HPCI operation.
.
O
.
.
,
C i
.
'
,
!
Licensee procedure GP-8, " Primary Containment Isolation,"
i
--
includes verification that all automatic safety functions i
operated as designed following an isolation. Completion of this verification is the responsibility of the Shift Technical Advisor (STA).
In this case the Shift Junior Technical Advisor (JTA) performed portions of GP-8.
The JTA mistakenly verified switch positions instead of the final actuated status
,
of secondary containment dampers and the standby gas treatment
'
trains. The licensee performed a special functional test of i
the equipment prior to unit restert to ensure that components
!
would operate properly.
Operations canagement discussed the responsibility for completion of GP-8 with all STAS and JTAs.
During review of the completed GP-18 the inspector noted that
--
the Plant Operations Review Committee (PORC) quorum approving i
the results consisted of the Technical Superintendent, the
>
Operations Superintendent and three Shift Managers.
This is allowable under the licensee's Technical Specifications and
,
procedures.
The inspector expressed concern that a PORC
'
quorum including four members from the operations organization, including 3 SMs, would not provide the balanced,
,
broad perspective envisioned by creation of the PORC.
The PORC Chairman stated that this was an unusual circumstance and was not typical of other PORC quorums.
T.mically only a single SM would participate. He also statea that the PORC i
review conducted was thorough and balanced. The inspector
'
reviewed a sample of other PORC minutes and did not identify any additional concerns, i
This plant transient was also the subject of a special human
factors-oriented review led by the NRC Office of Analysis of Events and Operating Data (AE00) during the week of January 30.
The results of this review will be documented by AE00.
I In summary, control room staff response to the transient was commendable.
Licensee management follow-up to problems identified was acceptable. The inspector had no further questions.
3.0 Emergency Technical Specification Change for the ADS Identification of the Problejn On February 7, 1990, licensee test engineers performed surveillance test ST 1.8, " Automatic Depressurization System (ADS) 'A' Logic System l
Functional," Revision 23, on Unit 2.
This test is performed once per six l
months to verify operability of the ADS actuation logic, and tests the I
various logic combinations needed to energize each ADS valve.
Pilot lamps l
installed in parallel with the two in-series final actuation relay l
-
.
l D
.
f
!
.
!
,
i contacts indicate the state of the relays during the test.
The initial test steps installed high resistance neon bulbs into the pilot lamp i
sockets, and established the logic combinations required to actuate one
'
of the two final relays.
The licensee proceeded through the test to
step 10, energizing the relay and thereby applying 125 V DC to the pilot lamp. The lamp circuit to the system grcund is completed through the applicable ADS valve soler.oid coil. The high resistance of the neon bulb prevents energizing the coil and lifting the safety relief valve.
Lighting the lamp proves that the logic combination is complete, that the final actuation relay contacts have closed, and that the circuit
,
through the ADS solenoid coil is complete. Test step 10.E requires verification that pilot lamp 2E-SD6 E, associated with the "K" ADS valve, is on.
During the test on February 7, the test engineers identified that pilot lamp 2E-SD6 E did not light when required.
The engineers didn't sign off Step 10.E aborted the test, and signed the test cover. sheet as i
unsatisfactory.
Surveillance test steps relied upon to demonstrate compliance with the Technical Specifications (TS) are identified in the
body of the procedure with an asterisk.
Section B of the cover sheet for procedure ST 1.8 indicates that the test is unsatisfactory if "One or more of the asterisked steps were completed UNSATISFACTORILY.
Refer to Tech. Spec. 3.5.E.1, 2, & 3, 4.5.E.2."
Step 10.E is an asterisked step. The Unit 2 Reactor Operator, the Shift Supervisor and the Shift Manager reviewed the results and also signed Section B as unsatisfactory.
The test engineers evaluated the problem using a Troubleshooting Control Form (TCF) and concluded that the pilot lamp socket was defective, preventing the bulb from lighting.
The troubleshooting activity consisted primarily of verifying that the final actuation relay contacts had closed.
Systems engineering supervision and the control room staff reviewed the results of the TCF. The K ADS
,
valve was not declared inoperable based on acceptance of the socket as the source of the problem.
A maintenance request form (MRF) was
'
initiated. The licensee didn't act on the MRF, reperform surveillance test ST 1.8, or declared the valve inoperable.
On February 12, the system engineer performed the surveillance test for the B ADS logic, ST 1.9, and the pilot lamp associated with the B_ logic
'
for the K ADS valve also didn't light when expected.
Follow-up r
troubleshooting identified that the pilot lamps had not lit on February i
'
7 and again on February 12 because the circuit ground through the K ADS solenoid coil was open.
The open circuit is inside the drywell and the licensee believes it is due to a failed ADS solenoid coil.
Interruption of the circuit or f ailure of the coil prevents the K ADS valve from performing its ADS function and from opening in response to a remote manual demand.
The valve's pressure relief function is unaffected.
'
.
f
.
l
.
t l
Technical Specification (TS) 3.5.E allows continued operation with one ADS valve inoperable for 7 days if the HPCI system remains operable. TS , , 4.5.E.2 states that with one valve of the ADS inoperable, the ADS i
subsystem actuation logic for the other ADS valves and the HPCI subsystem shall be demonstrated to be operable immediately and at least weekly thereafter.
Due to the inadequate treatment of ST 1.8 results on February 7, 1990 the compensatory testing required by TS 4.5.E.2 was not performed until after failure of the second ST on February 12. The
[
inspector informed the licensee that the above constitutes a violation t
of the TS (NC4 90-01-01).
The licensee's surveillance procedures represent the carefully reviewed and agreed upon method for demonstrating system operability and compliance with the TS.
Li.:ensee administrative procedures, and the t
surveillance procedures themselves, identify that tests are successful only if the acceptance criteria are met.
For most STs, ST 1.8 included,
'
the test acceptance criteria are identified in the body of the procedure t
by asterisks, and are linked to the appropriate TS action statement by Section B of the cover sheet.
In this case the licensee failed to meet the test acceptance criteria. The required compensatory actions where not implemented, nor were the acceptance criteria revised using the established process.
Instead the licensee staff performed an informal review and, by lack of additional action, accepted the results as sati s f actory.
The inspector expressed concern that this method of treating ST acceptance criteria was inappropriate.
Waiver of Compliance and Emergency Technical Specification (TS) Change
The discovery that the K ADS valve was inoperable beginning on February 7, 1990 placed Unit 2 in a 7 day Limiting Condition for Operation (LCO).
Licensee analysis demonstrated that the 4 remaining ADS valves would be
,
i capable of performing the system design basis function.
In addition, the unit was limited to operation below 74% power due te isolation of a feedwater heater string, adding additional margin. The licensee had previously scheduled a mid-cycle maintenance outage for March 3, and felt that continued operation with the inoperable valve could be justified until the planned shutdown.
The licensee submitted a request for an Emergency Technical Specification (TS) Change on February 14.
The change allowed a one time extension of the LC0 to 11:59 p.m. on March 3, 1990. The licensee committed to perform weekly testing of the remaining ADS logic, HPCI and RCIC during this time.
The licensee also stated that a plant shutdown would be initiated if any of these systems became inoperable.
The NRC granted the licensee a Temporary Waiver of Compliance on February 14 to allow continued operation while the request was being reviewed.
Following the close of the inspection period on February 23 the NRC issued the requested TS amendment.
~
.
.
,
,
!
"
.
.
!
The licensee's TS change request and the supporting engineering evaluations were reviewed by the resident inspectors, Region I, and NRR.
The analysis was thorough, well developed and reflected high quality work, despite the time constraints imposed.
4.0 Engineering and Technical Support Activities
'
'
A utility made a Part 21 notification concerning misapplication of a Magnetrol level switch on the high pressure coolant injection (HPCI) and reactor core isolation cooling (RCIC) systems.
This level switch (Model
5.0-751-XXXX, Part No. 57-3003-006 without fin section) opens a bypass
~
valve around the steam trap on the HPCI and RCIC steam supply lines when a high level exists in the drain pot.
These switches were failing at a shorter interval than specified by the vendor.
The potential safety concern is the loss of HpCI and RCIC due to moisture buildup in the steam lines.
The inspector performed a document review and determined that the questionable Magnetrol switches are not used on the HPCI or RCIC steam lines drain pots at peach Bottom Units 2 and 3.
The HPCI systems use FCI level switches and the RCIC systems uses Robertshaw level switches.
The inspector also walked-down the systems and verified that the as-built configuration agreed with documentation reviewed. The inspector had no further questions.
5.0 Surveillance Testing 5.1 Surveillance Test Scheduling Controls During follow-up to a Licensee Event Report the licensee's Compliance Group recognized a potential discrepancy in the surveillance test scheduling program known as STARS.
The STARS system schedules testing within a one-week window. Any test performed within the window is considered completed on time, whether it is performed on day 1 or 7.
This practice can result in a test exceeding its scheduled interval without being considered overdue.
In addition, the interval plus 25 percent grace period can be exceeded without being recognized.
For example, if STARS schedules a monthly test required by Technical-Specifications (TS) for performance in week 1, and it is completed on day 1 of the week, it would be scheduled next during week 5.
If performed on day 7 of week 5, the test would be viewed as successfully completed within the TS required interval, although 5 weeks would have elapsed between performances.
Similarly, if not
'
completed in week 5 the test would be rescheduled for week 6.
If completed on day 7 of week 6 the test would be viewed as completed within the 25 percent grace period, although 6 weeks would have
.
.
.
.
.
!
i I
elapsed (a 50 percent grace period).
This approach when applied to weekly testing could result in 2 weeks between consecutive tests. This weakness also exists for longer interval tests but because the maximum unrecognized slip is 2 weeks it is less significant.
The scheduling system does ensure that three consecutive tests do not exceed 3.25 times the required interval. Other licensee work practices reduce the likelihood that this weakness will be i
frequently exploited.
For example, the licensee generally performs weekly tests on the same day each week.
The licensee tracks the percentage of tests using the grace period and this number remains very low.
The licensee stated that they cannot revise the present scheduling system to track test completion by date instead of scheduling
'
week.
In addition, the licensee stated that the current system is adequate and meets the intent of the TS. The inspector has not identified an instance in which the 25 percent grace period has been exceeded. But, it is possible that this has or will occur.
The inspector requested to review the licensee basis for
'
concluding that the present system complies with TS requirements, and the licensee management approved documentation of this assessment. This item will remain unresolved pending completion of this analysis by the licensee and review by the inspector (UNR 90-01-02).
5.2 Routine Observations The inspector observed surveillance tests to verify that testing had been properly scheduled, approved by shift supervision, control room operators were knowledgeable regarding testing in progress, approved procedures were being used, redundant systems
or components were available for service as required, test
'
instrumentation was calibrated, work was performed by qualified personnel, and test acceptance criteria were met.
Daily surveillances including instrument channel checks, jet pump operability, and control rod operability were verified to be adequately performed.
Parts of the following tests were observed or reviewed:
-- ST 6.5-3, "HPCI Pump, Valve, Flow, Cooler," performed on Unit 3 on January 9.
-- ST 6.11F-3, "RCIC Pump, Valve, Flow & Cooler Functional Flow Test," performed on Unit 3 on January 8.
-
e w
.
.
.
.
'
-- ST 1.8, " Automatic Depressurization System "A" Logic System Functional," performed on Unit 2 on February 7.
l
-- ST 1.9, " Automatic Depressurization System "B" Logic System i
Functional," performed on Unit 2 on February 12.
'
No concerns were identified.
6.0 Maintenance Activities 6.1 Unit 3 Main Steam Line Flow Transmitter Failure At 10:20 p.m. on January 5, 1990, a licensee shift supervisor touring the reactor building noted that the master trip unit
output for the 116B main steam line (MSL) flow transmitter (DP15
'
3-2-116B) indicated 0 psid.
A gross failure trip light on the master trip unit was not illuminated nor was a gross failure alarm received in the control room.
At 1:00 a.m. on January 6, technicians valved the 116B MSL flow transmitter out of service for troubleshooting. At 1:23 a.m. the control room operator inserted a "B" channel primary containment isolation. system (PCIS) group I trip to comply with Technical Specification (TS) Table 3.2. A.
Troubleshooting determined that the Rosemount transmitter (Model 1153) had failed and a maintenance request form (MRF) was written to replace it. At 11:50 a.m. the control room operator reset the "B" channel PCIS group I trip to allow calibration of 116B. At 2:10 p.m. the transmitter was declared operable and returned to service.
The licensee believes that the failure was caused by metal particles in the sensor oil that, over time, aligned to produce a short circuit.
Since mid-1989, five MSL flow transmitter failures occurred.
Two Unit 2 transmitters that failed in July 1989 were found to contain metal particles in the sensor oil, and two Unit 3 transmitters that failed in December were sent to Rosemount for destructive testing. The recently failed 116B MSL flow transmitter was also sent to Rosemount for destructive testing.
The inspector expressed concerned that operations took too much time (3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />) to place the affected PCIS channel in a tripped condition. Although Peach Bottom TS do not include a specific time limit, standard TS include a 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> limit.
The licensee agreed, and committed to evaluate and address the cause of the delay.
This issue will remain unresolved pending review of the licensee's response (UNR 90-01-03).
The inspector also questioned the amount of time allowable to test
-
.
@
.
.
,
-
.
.
i
'
and calibrate the instruments.
The area of concern is the amount of time the instrument channel is not in the tripped condition and is valved out of service.
Peach Bottom TS do not address a time limit during which an instrument'may be removed from service for
.
testing without tripping the affected channel..The bases state,
,
"a channel may be made inoperable for brief intervals to conduct required functional tests and calibration." Standard TS define this interval as not to exceed 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />.
Here about 2 1/2 hours
.
elapsed between the reset of the trip and the return of the instrument to service. 'This period is not excessive; but,
!
licensee procedures do not restrict nor track this out-of-service time.
The licensee is currently evaluating this issue to decide if a specific period of time and controls are warranted.
Since this MSL transmitter failure did not generate an alarm in the control room, the inspector questioned whether the licensee began monitoring of the local indicators on the master trip units, i
The licensee stated that they have been monitoring these instruments once per-day through the use of non-licensed operator round sheets.
The inspector also questioned other applications of the 1153 transmitters and whether their outputs were monitored by
'
indicators as well.
The licensee provided a listirg of all safety-related applications where the 1153 transmitters used.
None were blind applications, all provided output to indicators or recorders.
.
The inspector questioned whether the licensee had evaluated the MSL flow transmitter failures.
The licensee produced a failure L
i analysis report (FAR) dated December 28, 1989, concerning the two
'
MSL flow transmitters on Unit 2 in July 1989.
The maintenance organization did not issue the FAR until 5 months after the failures occurred, and did not update it to reflect the December or January failures. The FAR did not address 10 CFR Part 21 reporting, investigation of other applications of the 1153s and
.
whether they were blind applications, or a determination if increased monitoring of 1153 indicators would be appropriate. The cognizant licensee maintenance engineer stated that although he
>
did not document these elements, he did considered them during the failure follow-up.
The inspector concluded that the licensee took an inappropriately long time to place the PCIS channel in a safe condition after it was found to be inoperable. The documentation for the FAR associated with the failures of the Rosemount transmitters was incomplete. However, the licensee indicated that the appropriate factors had been considered.
.'
.
O
i
.
.
.
h
.
'
6.2 Routine Observations i
The inspectors reviewed administrative controls and l
associated documentation, and observed portions of work on l
the following maintenance activities:
Document Equipment l
MRF 9000194 Unit 3 HPCI gland seal condenser system
'
MRF 9000519 Unit 3 "A" RHR heat exchanger cooling water discharge valve
.
TCF Unit 3 "D" CAD analyzer MRF 9000998 Unit 3 "C" reactor feedwater pump M/A station In addition, reviews of the following completed maintenance procedure was performed:
-- M-511.100, " Procedure for Determining MOV Post-Maintenance Testing."
Administrative controls checked included blocking permits, fire watches and ignition source controls, QA/QC involvement, radiological controls, plant conditions, Technical Specification LCOs, equipment alignment and turnover information, post-maintenance testing and reportability.
Documents reviewed included maintenance procedures (M), maintenance request forms
'
(MRF), troubleshooting control forms (TCF), item handling reports, radiation work permits (RWP), material certifications, and receipt inspections.
No concerns were identified.
.
7.0 Radiological Controls 7.1 Unplanned Personnel Radiation Exposure On January 16, 1990, two systems engineers entered the Unit 3
[
condenser /preheater room in the recombiner building to
troubleshoot off gas system problems.
The area was posted "High Radiation Area-Contact HP."
The radiation work permit (RWP) that the individuals had signed allowed them to enter.
But, it i
required a briefing from health physics (HP) personnel, alarming dosimeters, and positive HP coverage prior to entering the room.
The individuals ignored the posting and basic RWP provisions and entered the room in violation of the licensee's procedures.
.
N
.
.
.
i
,
e The individuals were in the room for approximately 52 minutes during which time one individual checked his self reading dosimeter (SRD) at least once, showing that he had picked up
'
approximately 50 millirem (mr) in about 15 minutes.
Upon exiting the room both individuals noted that their self-reading dosimeters (SRDs) were off-scale (>200 millirem), When exiting the power block the individuals notified HP that their SRDs were off-scale.
HP immediately initiated a radiological occurrence report (ROR)
and surveyed the work area.
Their thermoluminescent dosimeters were sent off-site for a reading, and they were den'.ed access into
.
'
any radiation areas.
The HP survey found general area dose rates between 60 and slightly less than 1000 mr/hr.
Therefore, the room did not ineet the criteria to be a locked high radiation area (>1000 mr/hr.).
The licensee conservatively posted and locked the room-on January 17.
The results of the TLD readings were that one individual i
received about 200 millirem while the other individual received about 230 millirem. The licensee suspended the individuals for one week without pay and required them to attend general employee training (GET) that week.
The Radiation Protection Manager (RPM)
interviewed both individuals after they passed the GET, and reinstated their access to the power block. In addition, all-hands meetings were held with station personnel concerning this issue and stressing the importance of obeying RWPs and HP postings.
The inspector reviewed ROR 90-0005, the radiation survey, TLD results, and spoke with HP personnel.
The inspector concluded that licensee corrective actions were adeouate and management involvement in this occurrence was evident.
The individuals'
exposure didn't exceed regulatory limits and the inspector had no further questions.
The inspector concluded that this violation satisfies the criteria for licensee identified violations as stated in 10CFR 2, Appendix C, Section V.G.20, and that no Notice of Violation is warranted.
Since licensee corrective actions were reviewed during the inspection period and found to be acceptable no additional specific NRC follow-up is planned (NCV 90-01-04).
7.2 Air Monitor Alarm Set Points On November 26, 1989, following a Unit 2 scram from 30% power, sections of the reactor building (RB) on the 165 foot elevation were contaminated near the reactor water cleanup (RWCU) pump rooms (see Inspection Report 50-277/89-26, Section 6.3).
During the event the continuous airborne radioactivity monitor (AMS-3)
located in the area did not alarm.
The instrument recorded an increase from 800 to 8000 counts per minute (CPM) in 15 minutes.
The alarm set point posted on the AMS-3 was 6000 CPM.
The i
licensee completed an investigation of the incorrect alarm set i
point and initiated corrective actions.
.
W
.
.
I
.
The licensee investigation revealed that an HP supervisor directed
!
increasing the alarm setpoint to 10,000 CPM on November 22, 1989, due to high radon levels in the plant. The supervisor noted in the r
log that the AMS-3 should be trended and the alarm reset as the
'
background decreased.
This instruction was not carried out nor did daily operational checks determine that the alarm set point had been changed.
The information sticker on the instrument did not reflect the set point increase.
Procedure HP-448, " Operation of the Eberline Beta Air Monitor, Model AMS-3," Revision 2,
addresses the setup and operation of the AMS-3.
The procedure
suggests that the information sticker should reflect the alarm set i
point. But, this is contained in the initial setup instructions for the instrument and it is unclear if it applies to readjustments.
The procedure allows readjustment of the setpoint but does not contain any provisions for tracking to ensure r
reevaluation. The licensee did not discover the problem until questioned by the inspector following the contamination event.
The immediate corrective actions included:
issuing a temporary I
change to procedure (TC-90-004) HP-448 requiring a weekly alarm setpoint check; issuing an instruction to HP technicians
.
specifying the AMS-3 normal alarm setpoint as 6000 CPM; specifying
when and how the alarm setpoints may be changed, and stating that changed alarm setpoints must be monitored each shift and returned to 6000 CPM when possible; and issuing an instruction to HP supervisors discussing the above.
The long-term corrective-action is to rewrite and clarify Procedure HP-448.
r The inspector noted that in the previous Inspection Report 50-277/89-26; 50-278/89-26, procedure HP-514, " Respiratory Protection Program," was also found to be unclear.
The inspector expressed concern that other HP procedures may contain similar weakness.
The licensee acknowledged this concern. The inspector also
.
pointed out that the Radiological Occurrence Report (ROR) took an unusually long time to complete, about 60 days.
This is an exception to otherwise timely completion of RORs. The inspector had no further concerns at this time.
i 7.3 Routine Observations l
During the report period, the inspector examined work in progress j
'
in both units and included a review of health physics procedures
'
and controls, ALARA implementation, dosimetry and badging, protective clothing use, adherence to radiation work permit (RWP)
requirements, radiation surveys, radiation protection instrument
use, and handling of potentially contaminated equipment and materials.
l l
.
Y
.
.
.
.
The inspector observed individuals frisking according to HP procedures. A sampling of high radiation area doors was verified to be locked as required.
Compliance with RWP requirements was verified during each tour.
RWP line entries were reviewed to verify that personr.el had provided the required information.
People working in RWP areas were observed to be meeting the applicable requirements.
No unacceptable conditions were identified.
8.0 Physical Security The inspector monitored security activities for compliance with the accepted Security Plan and associated implementing procedures, including: security staffing, operations of the CAS and SAS, checks of vehicles to verify proper control, observation of protected area access control and badging procedures, inspection of protected and vital area barriers, checks on control of vital area access, escort procedures, checks of detection and assessment aids, and compensatory measures.
No inadequacies were identified.
9.0 Review of Licensee Reports 9.1 Licensee Event Reports (LERs)
The inspector reviewed LER: submitted to the NRC and verified that I
the licensee took appropriate corrective action and assigned responsibility, and that continued operation of the facility was conducted according to Technical Specifications and did not constitute an unreviewed safety question as defined in 10 CFR 50.59.
Report accuracy, compliance with current reporting requirements and applicability to other site systems and conponents were also reviewed.
LER No.
LER Date i
Event Date Subject 3-89-10 Late performance of TS surveillance test due to 01/11/90 programmatic deficiencies 12/11/89 3-89-07 TS actions not performed as required due to 11/27/89 deficient procedure 10/27/89 2-89-16, Rev. 1 LPRM spike causing full scram while shutdown 01/11/90 07/22/89
.
O
-
.
.
.
.
2-89-27 TS actions not performed due to personnel error 11/15/89 10/16/89 No concerns were identified.
9.2 Routine Licensee Reports TS 6.9.1.C requires that the licensee submit an annual report listing the challenges to reactor vessel safety relief valves (SRV) and containing a description of the circumstances of each challenge.
By letter dated January 16, 1990, the licensee submitted this report, stating that no challenges had occurred during 1989. The inspector noted that on July 21, 1989 a primary containment isolation system group I isolation resulted in opening of two SRVs due to high reactor pressure. The licensee described the challenge in LER 2-89-015.
When the inspector pointed out the discrepancy, the licensee performed a second review of plant events, and submitted a revised report on February 7,1990.
The licensee identified that authorship of the report had not been assigned to the appropriate organization.
This responsibility was reassigned to the technical organization including the system engineer responsible for the SRVs.
Additional review of other similar routine report responsibilities identified no other misassignments.
The inspector had no further questions.
10.0 Individual Plant Examinations for Severe Accident Vulnerabilities Review of Licensee Response to NRC Generic Letter 88-20 On November 23, 1988, the NRC issued Generic Letter (GL) 88-20 requesting individual plant examinations (IPE) for severe accident vulnerabilities from all licensees.
The general purpose of this examination is to (1) develop severe accident behavior, (2) understand the most likely severe accident sequences that could occur, (3) gain quantitative understanding of the probabilities of core damage and fission product releases, and (4) reduce the core damage and fission product releases by modifying hardware and operating procedures intended to prevent or mitigate severe accidents.
The NRC issued NUREG 1335 in August 1989 to provide specific guidance for IPE development.
Supplement No I to GL 88-20 was issued on August 29, 1989, to announce the issuance of NUREG 1335.
GL 88-20 and its supplement requested the licensee to submit their proposed program for completing IPE to the NRC within 60 days of the publication of NUREG 1335 addressing the following:
,
i
,
.
!
e
.
.
-
i
.
!
identify the method and approach selected for performing the IPE, l
--
.
describe the method to be used for the examination, and
--
identify the milestones and schedules for performing the IPE and
--
,
submitting the results to the NRC.
.
Additionally, the Generic Letter requested the licensee to complete the l
IPE and submit the final report within 3 years of the issuance of NUREG 1335.
'
The Licensee submitted their program developed in response to Generic I
Letter 88-20 to the NRC on October 31, 1989. The inspector reviewed the Licensee's actions in response to the Generic Letter 88-20 and verified the following:
NUREG 2300/PRA Procedures Guide was used to develop Peach Bottom
--
Unit 2&3 IPE (NUREG 2300 is an acceptable method listed in NUREG 1335).
PECo will also use NRC-sponsored NUREG 1150 as a guide in developing a final IPE.
^
The final submittal for Peach Bottom IPE is presently scheduled for
--
September 1, 1992. This meets the 3 year schedule outlined in GL 88-20.
The licensee currently uses NUREG 1150 and WASH 1400 Level 1 PRA to
--
plan system outages and set priorities for maintenance work.
General concepts of core damage conditions have been included in
--
the licensee's technical training program.
The licensee has also implemented a Reliability Assessment Program.
The general scope of this Reliability Program is to maintain an updated risk i
model based on current as-built configuration and to increase the awareness of PRA concepts within station operations. This program has achieved the following:
developed and documented Peach Bottom computer risk model for
--
predicting the core damage frequency and the ability of containment to mitigate accident sequences, the Peach Bottom computer risk model is maintained in "Living"
--
format with a one year update interval.
the program assures that the Peach Bottom PRA is revised and
--
submitted as required in Generic Letter 88-20.
The inspector had no further questions.
-
O
!
...
.
.:
i i
'
.
i
i
!
11.0 Management Meetings-
!
!
!
The inspectors provided a verbal summary of. preliminary findings to the Peach. Bottom Station Plant Manager and members of his staff at the l
.
. conclusion of the inspection. During the inspection, licensee management t
was periodically notified verbally of the preliminary findings by the l
resident inspectors. No written. inspection material was provided to the
[
licensee during the inspection.
This report includes no proprietary information.
,
i s
I f
i
$
l
<
i
>
I-I f
s t
l l
l
.
,
,
,
%
..,__
. - _ _,.. _..
.
_
, _ _..... _ _ _,
. _ _, _....
.
-. - -
.. -
,. ~.
,
-
.
.
,
,
i f
ATTACHMENT 1 Facility and Unit Status
,
Unit 2
>
January 1 Reactor power increased to 100%.
January 27 Reactor shutdown due to unisolable leak on "B" reactor
-
feedwater discharge flow instrument line.
Scrammed from 30%
power (normal shutdown routine).
January 28 Leak repair on "B" reactor feedwater discharge flow instrument line completed.
January 29 Reactor startup, critical and at rated pressure.
January 30 Power ascension in progress.
,
<
February 2 Reactor power at 96%.
February 3 Reactor power reduced to 83% to adjust rod pattern, power ascension to 100% initiated.
February 5 Reactor power reduced to 66% due to removal of "A" feedwater heater string.
Removal of the string was prompted by a degrading 2A internal heater leak.
February 6 Reactor power increased to limit of 74% with one feedwater
heater string out of service.
'
February 12 Automatic depressurization system (ADS) safety relief valve (SRV) "K" declared inoperable from February 7.
Licensee requests an emergency TS revision to allow operation until March 3, 1990. Reactor power at 74%.
,
February 14 Plant shutdown initiated (power decreased to 72%) due to inoperable ADS SRV "K" LCO expiration.
Emergency TS change
'
received to allow continued operation until March 3 with ADS SRV "K" inoperable and with additional HPCI, ADS, and RCIC testing requirements. Reactor power increased to 73%.
February 20 Reactor maintained at 73%.
Unit 3 January 1 Reactor power increased from 57% to 74%.
January 2 Reactor power dropped to 55% to achieve 100% rod pattern.
January 3 Reactor power increased to 75%.
January 4 Reactor power increased to 91%.
January 5 Full reactor power is achieved.
January 6 Core thermal limits are reached and power is reduced to 98%.
January '.
Plant shutdown initiated (decrease to 97%) due to HPCI and RCIC inoperable; shutdown halted after RCIC tested operable.
January 9-18 Reactor power remains at 99%.
s e
i
'
.
,
,
,
!
.
l
'
'
i
{
l l
January 19
.0ff gas activity increases from 2000 to 6000 microcuries per I
second due to repair of sampling line air leak.
[
January 20-27 Reactor power remains at 99%.
j January 28 Unit shutdown due to EHC leak on #1 turbine control valve.
l January 30 Mode switch to startup and reactor critical.
!
January 31 Reactor at rated pressure.
February 1-5 Power ascension in progress.
,
February 6 Reactor power able to reach 100% due to new rod pattern
~
l relieving thermal limit restrictions.
- February 7-12 Reactor power remains at 100%.
February 13 Off gas activity increases from 6000 to 14000 microcuries per second due to. progressive leak.
,
February 14-20 Reactor power remains at 100%.
i
!
,
P t
,
I
!
l
,
i.
l'
i
'
i
K
_ -.. _
_.. _. ~.,
__.
-
..