IR 05000387/1985031
| ML17139D301 | |
| Person / Time | |
|---|---|
| Site: | Susquehanna  |
| Issue date: | 11/26/1985 |
| From: | Jacobs R, Plisco L, Strosnider J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML17139D300 | List: |
| References | |
| 50-387-85-31, 50-388-85-26, NUDOCS 8512090334 | |
| Download: ML17139D301 (27) | |
Text
U.S.
NUCLEAR REGULATORY COMMISSION
REGION I
Report Nos.
Docket Nos.
50-387/85-31'0-388/85-26 50-387 CAT C
50-388 CAT C
License Nos.
NPF-14 NPF-22 Licensee:
Penns lvania Power and Li ht Com an 2 North Ninth Street Allentown Penns 1 vania 18101 Facility Name:
Inspection At:
Inspection Conducte
Sus uehanna Steam Electric Station Salem Townshi Penns lvania Se ember
1985 - November
1985 Inspectors:
y R
Approved By:
Jacobs, enior Resident Inspector 1'sco, esident Inspector ps'g date
/~,Zb SK date
//zb SS J. Strosnider, Chief Reactor Projects Section 1B, -DRP date Ins ection Summar Areas Ins ected:
Routine resident inspection (U1 201 hours0.00233 days <br />0.0558 hours <br />3.323413e-4 weeks <br />7.64805e-5 months <br />; U2 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br />)
of plant operations, licensee events, open items, surveillance, maintenance and ESF walkdown.
Results:
ESF walkdown of Unit 1 Core Spray System identified no substantive di screpancies (Detail 2.3); review of 18 month diesel surveillance requirement identified several discrepancies (Detail 5.3); licensee actions followingj RWCU spill on October ll were thorough and conservative (Detail 6.0).
No violations were identified.
8512090334 851202 PDR ADQCK 05000387
p+
L,
DETAILS 1.
Follow-u on Previous Ins ection Items Closed Ins ector Follow-u Item 387/83-12-04
- Control of Control Structure Boundar Doors In May 1983, the licensee discovered that a control structure bounda-ry door had been blocked open by cables to facilitate Unit 1 battery testing.
This violated the control structure boundary which would inhibit the proper performance of the CREOASS system in a post-accident condition.
In Inspection Report 50-387/83-12, the in-spector discussed concerns about the controls over the control struc-ture boundary and training of station personnel.
1.2 As a result of the incident, the licensee marked each affected door as a control structure boundary door and added a daily check of the doors on an operator rounds sheet.
On December 17, 1984, the licensee revised LER 83-076 to require that only non-security control structure boundary doors be checked on operator rounds'ecurity doors are monitored continuously.
The inspector reviewed S0-100-007, Rev.
5, "Daily Surveillance Operating Log" to verify that the doors have been included in the operator's rounds sheets.
Closed Ins ector Follow-u Item 387/83=S1-03 388/83-31-03
No Reference to Emer enc Procedures in Alarm Res onse Procedures In December 1983, the inspector noted that the alarm response proce-dure (ARP) for primary containment high pressure trip (AR-204-001)
had no reference to the appropriate Emergency Operating (EO) proce-dure to be used during the alarm condition.
The licensee reviewed all ARPs and included, where appropriate a
reference to the associated Emergency Operating or Off-Normal proce-dure.
The inspector reviewed several ARPs to verify that this action was completed.
1.3 Closed Ins ector Follow-u Item 387/84-38-01 388/84-47-01:
Ac-tions to Ensure Cleanliness of CRD Air S stem In December 1984, the scram discharge volume (SDV) vent and drain pilot valve malfunctioned preventing the SDV vent and drain valves from closing.
Investigation revealed a small piece of pipe dope was trapped between the valve inlet seat and disc of the pilot valve.
Because of the concern of loose particulates in the CRD air system, the licensee performed a freon flush of the CRD air system and during the first refueling outage, cleaned system red brass piping where pipe dope was used.
In addition, the licensee implemented PMR 85-1004 on Unit 1 which disassembled and replaced instrument air tub-ing from the CRD air header to each of the scram pilot solenoid
valves (SPSV).
Pipe dope had been used in the original tubing'o the SPSVs.
The flush results on Unit 1 did not identify any significant contamination of the air system.
Hence, the licensee does not: intend to perform a flush of the Unit 2 air system.
Both units now have redundant SDV vent and drain valves and pilot valves, so the l,ikeli-hood of both SDV pilot valves fai ling is reduced.
PMR 85-1005 will replace instrument tubing at the SPSV on Unit 2 during the first re-fueling outage.
The inspector had no further concerns.
Closed Ins ector Follow-u Item 387/83-11-05
Missin Surveillance Re uirements in Procedures In Apri 1 1983, the licensee identified that six instruments had not been included in Operations surveillance procedures.
These instru-ments were associated with containment isolation and ATWS recircula-tion pump trip.
In LER 83-45 the licensee committed to review the operator's 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />, 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and weekly surveillance procedures'to ensure all channel check requirements were included.
The licensee performed the above review and added the missing instru-ments to the shiftly surveillance procedure.
The inspector verified that the instruments were added.
Following this occurrence, the licensee committed to perform a comprehensive review of the survei 1-lance program, which included a verification that all surveillance requirements are included in appropriate procedures.
This task has been completed and will be reviewed with Violation 387/83-20-01.
The inspector also reviewed AD-gA-422, Surveillance Testing Program, to verify that the licensee ensures that surveillance procedures are revised as appropriate when the Technical Specifications are modified.
0 en Ins ector Follow-u Item 387/84-26-01
- Diesel Generator Turbochar er Failure On September 10, 1984, during monthly surveillance testing, the',
'D'mergency diesel generator (EDG) tripped on high vibration.
During three subsequent troubleshooting starts, maintenance personnel noted a whining noise from the turbocharger area and the EDG tripped on high vibration several minutes after starting.
Partial disassembly of the Cooper-Bessemer ET high pressure turbocharger revealed that the blower end journal bearing and the adjacent thrust bearing were severely damaged.
The turbocharger was replaced with a spare and the EDG was declared operable on September 12, 1984.
A previous turbocharger thrust bearing failure occurred on the 'C'DG during surveillance testing in February 1984.
(See Inspection Report
50-387/84-07).
The licensee shipped the damaged turbocharger from the September failure to the vendor (Cooper-Bessemer)
for evaluation.
The vendor concluded that the lack of prelubrication to the thrust bearing was a
major contributor to the accelerated bearing wear.
The thrust
bearing wear progressed with each "dry start" unti 1 the bearing geom-etry was altered to the point that the bearing was no longer capable of sustaining normal load even under normal lubrication conditions.
The journal bearing failure was determined by the vendor to be conse-quential.
The vendor recommended that the licensee pursue the pro-posal of adding a turbocharger prelubrication sequence to the 'test startups to eliminate the problem.
The vendor stated the turbocharger may have been within minutes of impeller-casing contact and subsequent gross failure.
The turbocharger bearings are not prelubricated in the standby mode as are the remainder of the engine bearings.
Due to the design of the labrynth bearing seal, lube oil cannot be continuously circulated through the turbocharger when the unit is shutdown because oil', would flow past the seals and into the air inlet system.
The bearings may not reach rated lube oil pressure until 5-10 seconds after sta'rting.
The licensee has now incorporated a turbocharger bearing prelubrication sequence prior to starting the diesel during scheduled surveillance testing.
Operating Procedure OP-024-001, Diesel Genera-tors, was revised in February 1985 to allow manual initiation of turbocharger lubrication prior to manual starts.
The surveillance test procedures have also been revised to incorporate the prelubrication sequence.
The licensee has additionally issued'a Re-quest for Modification (RFM) to change the non-emergency mode start-ing logic to automatically initiate turbocharger lubrication prior to a diesel start.
The licensee presently intends to implement this modification in 1987.
The 'A'nd 'B'DG turbochargers have not been inspected since the two bearing failures.
The inspector reviewed the periodic mainte-nance requirements for the diesel generator (See Detail 5.3)
and not-ed that the turbocharger is only inspected every six years (every fourth 18 month inspection).
During the 18 month inspections, a vi-bration survey is performed, the turbocharger coast down time is checked and the turbocharger lube oil filters are inspected for me-tallic particles.
These checks may not detect a degradation of the turbocharger thrust bearing.
The last 18 month inspection was per-formed in January 1985, and the next inspection due date is September 1986.
This inspection will be the fourth inspection, so in accor-dance with the surveillance procedure, SM-024-002, and the vendor recommendations, the turbochargers will be removed and inspected in 1986.
This item will remain open pending NRC review of the results of the
'A'nd 'B'DG turbocharger inspection and final corrective actions.
Closed Ins ector Follow-u Item 388/85-09-01:
RCIC S stem Walkdown Discre ancies In March 1985, the inspector identified several minor discrepancies with the Unit 2 RCIC system during a system walkdown.
These
discrepancies were that 1) several RCIC valves were locked closed but the PAID did not indicate that they were required to be locked closed, 2) two valves were indicated locked open on the P&ID but they are normally open, motor operated valves, and 3)
some buckets 'of oil were left unattended in the RCIC room.
The licensee issued Dr'awing Change Notice (DCN) 85-3262 to correct the above drawing discrepan-cies and the buckets of oil were removed from the room.
The inspec-tor verified the above actions.
Closed Ins ector Follow-u Item 387/84-35-03
- Scram Dischar e Vol-ume SDV Vent and Drain Valve Missed Surveillance In October 1984, the licensee identified that surveillance procedure S0-155-003, SDV Vent and Drain Valve 18 Month Operability, was'ver-due by approximately 15 months.
The cause of the overdue surveil-lance was an administrative error.
The completion date of the'revious surveillance was listed as June 2,
1983.
The surveillance was actually'ast performed during preop testing in January 1982.
June 2,
1983 was the date that a
new surveillance authorization cover sheet was completed since the original surveillance document was misplaced.
The licensee performed a
100 percent verification of the completion dates for all surveillances whose frequency was quarterly or longer.
No additional late surveillances were identified.
An individual was assigned to conduct an on-going independent review of the surveil-lance program implementation.
0 en Unresolved Item 387/82-36-02
- Preservice Ins ection of Re-circulation S stem Pi e Melds Because of ALARA considerations the licensee has proposed to perform volumetric examination of recirculation system sweepolet to riser pipe welds using an ultrasonic examination technique in lieu of the currently mandated radiographic technique using MINAC which was used during the last outage.
Previous attempts to ultrasonically examine the welds proved unsuc-cessful because of numerous indications which were detected.
Those indications were attributed to the geometric and metallurgical condi-tion of the materials comprising the welds, base material and clad-ding through which the ultrasonic beam must pass.
The great number of indications precluded a meaningful interpretation of the data.
The technique currently being proposed by the licensee incorporates the use of a dual element, 45 degrees refracted longitudinal beam focused transducer, and is similar to the technique which was recent-ly developed for the examination of piping which contains corrosion resistant cladding.
The transducer active elements were designed to
f
match, as closely as possible, the acoustic impedance of austenitic stainless steel.
The previously attempted ultrasonic technique was demonstrated on a
weld mock-up containing ID and OD notches.
The mock-up was re-cently modified by the addition of two artificially induced cracks in the weld heat affected zone.
A demonstration of the new technique was performed by a Southwest Research Institute Level II individual, and was witnessed by the in-spector.
The purpose of the demonstration was to show that the two cracks can be detected, and that the crack indications are discern-ible above the noise level.
Equipment calibration was performed using a calibration block con-taining notches 10% of material thickness in depth, and system','sensi-tivity was then increased by 6 dB.
The resulting noise level )n the mock-up maximized at approximately 25% of full screen height at the inside surface of the material.
The examination makes use of the half arc technique, so that reflections are disregarded which appear beyond the I.D. location on the cathode ray tube.
The licensee successfully demonstrated that the two cracks can be detected ultrasonically, and that the crack signal to noise ratio was great enough to clearly display the crack signals.
The licensee plans to send the mock-up to the EPRI NDE Center at Charlotte, North Carolina, where it will be radiographed with the MINAC system.
The results will be used to assess the relative, merits of radiography and ultrasonics regarding detection of the cracks in the mock-up.
Additionally, the licensee wants to use the ultrasonic technique on the field welds in the plant during the next outage.
The inspector stated that he agreed with the licensee's plan.
He further stated that the preliminary results were promising, but'hat final judgement must be reserved pending examination of the field welds.
2.0 Review of Plant 0 erations 2.1 0 erational Safet Verification The inspector toured the control room daily to verify proper manning, access control, adherence to approved procedures, and compliance with LCOs.
Instrumentation and recorder traces were observed and the sta-tus of control room annunciators were reviewed.
Nuclear Instrument panels and other reactor protection systems were examined.
Effluent monitors were reviewed for indications of releases.
Panel indica-tions for onsite/offsite emergency power sources were examined for automat,ic operability.
During entry to and egress from the protected area, the inspector observed access control, security boundary
!
I
integrity, search activities, escorting and badging, and availability of radiation monitoring equipment.
The inspector reviewed shift supervisor, plant control operator, and nuclear plant operator logs covering the entire inspection period.
Sampling reviews were made of tagging requests, night orders,
~the bypass log, Significant Operating Occurrence Reports (SOORs),
and gA nonconformance reports.
The inspector observed several shift turn-overs during the period.
il 2.2 Station Tours The inspector toured accessible areas of the plant including, the con-trol room, relay rooms, switchgear rooms, cable spreading room's, pen-etration areas, reactor and turbine buildings, security control center, diesel generator building, ESSW pumphouse, and the plant pe-rimeter.
During these tours, observations were made relative to equipment condition, fire hazards, fire protection, adherence to pro-cedures, radiological controls and conditions, housekeeping, securi-ty, tagging of equipment, ongoing maintenance and surveillance
'and availability of redundant equipment.
2.3 ESF Malkdown - Unit 1 Core S ra S stem
~r On October 9 - 10, 1985, the inspector independently verified the operability of Unit 1 Core Spray System, Division II.
This ver~ifica-tion included a complete walkdown of all accessible system mechani-cal, electrical and instrumentation equipment to assess:
!
conformance of the "as-built" configuration to the system )draw-ings and FSAR; adequacy of the system mechanical and electrical equipment'heck-off lists; proper position, and remote and local position indication,f of valves and breaker s, including valve locking devices and end caps; hi operability of system instrumentation; and conditions which might degrade system operation or performance.
f!
In addition to the above, the inspector reviewed selected sections of Core Spray Logic Functional Test SE-151-001.
The review was conduct-ed to ensure that the test, as written, meets the intent of the",tech-nical specification and surveillance requirements.
Test sections were compared to Bechtel Electrical Schematics and General Elec(ric Elementary Diagrams to determine their accuracy.
lI
The inspector determined that system alignment was consistent with the system check-off lists and the current mode of operation.'ystem equipment was reasonably well maintained and identified.
Portions of the surveillance reviewed were technically adequate.
A few minor discrepancies were noted and discussed with the licensee.
The licensee initiated action to correct the discrepancies'.
Summar of 0 eratin Events 3.1 Unit
3.2 At 3:33 p.m.
on October 28, the Unit 1 reactor scrammed from 100 per-cent power due to low reactor water level.
While a Division II half-scram was inserted during routine surveillance testing of reac-tor vessel level instrumentation, a fuse blew to one of the four groups of Division I scram pilot solenoid valves, causing the
'rod group to insert.
The resulting level transient caused a full reactor scram on reactor vessel low level.
After investigation of the'blown fuse, the Unit was restarted and criticality was reached at 5: 10 p.m.
on October 29.
The scram ended a site record 138 day continuous run.
(See Detail 3.4)
At 7:36 p.m.
on October 30, the Unit 1 reactor scrammed from about
percent power on control valve fast closure due to a turbine trip.
The cause of the turbine trip was high level in the 'B'oisture sep-arator.
After investigation and instrumenting the moisture separa-tor, the unit was restarted, reaching criticality at 6:02 p.m.
on November 5.
After power escalation, the unit operated at full power for the remainder of the period.
(See Detail 3.5)
Unit 2 3.3 The Unit 2 reactor scrammed at 8:37 a.m.
on October 5 due to a 'ground fault in the 500 KV system grid caused by a lightning strike.
During the switchyard transient one of the Unit 2 synchronizing breakers (2T) was slow to clear the fault, initiating a breaker failure relay scheme.
This initiated a generator lockout and subsequent turbine trip.
The reactor scrammed due to turbine control valve fast closure.
After investigation of the switchyard transient, the startup was ini-tiated at 12:20 a.m. October 6 and criticality was reached at 10: 18 a.m.
the same day.
The Unit operated at or near full power for",,the remainder of the inspection period.
Turbine Bui ldin S ill At 10: 15 a.m.
on October 26, 1985, the on-shift STA identified water flowing out from under the door of the 'B'ondensate Demineralizer room on the 676 foot elevation of the Unit 1 Turbine Building.
, The STA notified the control room, and operators were dispatched to,the scene to isolate the leak.
On initial entry into the room by an HP
technician, it was noted that water and resin were blowing out of a vessel vent line flange, three of the four flange bolts were missing, and the flange gasket was blown out.
Attempts to isolate the'demin-eralizer vessel at the local control panel were unsuccessful and the manual block valves had to be closed to slow the leakage.
The leak was not completely stopped until approximately 12:30 p.m.
when a
new gasket was installed.
The spill volume was estimated to be approxi-mately 12,000 gallons.
The water and resin covered the 'B'eminer-alizer vessel room and the adjacent passageway in the Turbine Building, an area of about 3,000 square feet.
A small amount of the water also drained down an equipment access hatchway to elevation 656.
The water and resin were collected by the Turbine Building drains and processed by the Liquid Radwaste System.
General area radiation levels in the demineralizer room were 15-50 MR/HR immedi-ately following the spill.
The area was promptly decontaminated and unrestricted access to the area was restored.
The cause of the leak has been attributed to the failure of a 1-1/4 inch vent line flange gasket on the 'B'emineralizer vessel.,', Main-tenance had been performed on the demineralizer the week before, and the work included disassembly of the flanged connection.
Reactor power had been reduced to 52 percent from 100 percent at 6:50 a.m.
on October 26 in order to perform a rod sequence exchange.
The power decrease caused an increase of approximately 150 psig in condensate pump discharge pressure, and may have initiated the gasket fai lure.
On October 18, 1985 Work Authorization (WA) S55179 was performed to open, clean and inspect the 'B'ondensate demineralizer.
The'vessel was reassembled at approximately ll:00 p.m. October 18, 1985 after the inspection was completed.
The work plan specified cuttingiand installing new manhole gaskets with "GARLOCK 3200 SBR, 3/32 inch gas-ket material or equal".
The work plan was not specific concerning the material to be used for the flange gasket.
The piping isometric drawing (SP-GBD-107-7) specifies a 1/16 inch compressed asbestos gas-ket for the flange.
The licensee investigation following the s'pill determined that a red rubber gasket was actually installed during the flange reassembly.
The construction mechanics installed a red ',rubber gasket based on previous practice with the other demineralizers.
Two operational leak tests (approximately 470 psig) were performed ion the system after reassembly and no leaks were found.
At 4:47 p.m. Octo-ber 24, the demineralizer was placed in service, and remained i,n ser-vice for approximately two days before the leak occurred.
Licensee evaluation following the event determined that the red rubber w'as an acceptable material for this application.
A similar spill occurred on July 10, 1984 when the 'A'ondensate demineralizer blew a gasket and 3,500 gallons of water and resin spilled on the floor and adjacent hallway (SOOR 1-84-278).
The', root cause was determined to be poor alignment of the vent lines, which required multiple (stacked)
gaskets tb obtain an adequate seal.
Based on this occurrence and previous leakage problems, the vent
lines were reworked to provide a proper flange alignment in March 1985.
According to the associated WAs, compressed asbestos gaskets were installed following completion of this work.
The Unit 2 demin-eralizers have the same alignment problem and the WAs have been written, but the work is currently scheduled for the Unit 2 First Refueling Outage (August 1986).
The inspectors toured several demineralizer rooms on both Unit 1 and Unit 2 and found that various materials are being used as gaskets for the flanges and manhole covers on the demineralizer vessels.
The results of the inspection were discussed with the Maintenance Supervisor.
It appears that although this piping in non-g, stronger controls need to be in place to ensure the correct gaskets are used for each application.
Currently the only station guideline used is to "replace in kind".
Additionally, the inspectors stated that re-view of the schedule for correction of the Unit 2 flanges may be war-ranted based on this event.
The Maintenance Supervisor stated the Unit 2 work schedules would be reviewed for possible upgrade in pri-ority, and an evaluation will be conducted concerning the need for a
gasket control program.
Unit 1 Reactor Scram Due to Blown Fuse At 3:33 p.m.
on October 28, the Unit 1 reactor scrammed from 100 per-cent power due to low reactor water level.
While a Division II half-scram was inserted during routine surveillance testing of the
'D'eactor vessel level instrumentation, a fuse blew to one of the four groups of Division I scram pilot solenoid valves, causing the rod group to insert.
The resulting level transient caused a full reactor scram on vessel low level.
No ECCS actuations occurred dur-ing the scram and no safety relief valves lifted.
Just prior to the scram, IEC was performing quarterly surveillance test SI-180-305 on the 'D'eactor vessel level instrument (LIS-B21-N024D) which caused an expected half-scram on Division II.
While the half-scram was in-sertedd, fuse C72A-18A to the Group 1 scram pilot solenoid valves blew causing 1/4 of the control rods to scram.
The partial scram caused a
power decrease and rapid level decrease to the scram setpoint of 13 inches.
The operators performed the immediate actions of emergency procedure EO-100-101 and stabilized the plant.
The inspectors observed the operator response and recovery from the control room and attended the post-scram meeting.
All systems re-sponded as designed during the scram except for the 'O'RM which would not drive in. It was later found to have a blown fuse and was returned to service.
Review of the sequence of events printout and GETAR's traces indicated that all systems responded as designed.
A similar scram occurred on Unit 1 on July ll, 1983.
During IKC sur-veillancee testing of the primary containment high pressure channel
'A',
a fuse blew on the Group 4, System B control rod scram pilot
The blown fuse deenergized the Group 4, System B
scram pilot solenoid valves and the I&C surveillance deenergized the System A scram pilot solenoid valves, resulting in the insertion of only Group 4 control rods.
This 1/4 scram caused reactor power to decrease rapidly from 66 percent'to 10 percent, which resulted in a rapid decrease in reactor vessel water level causing a full scram on low reactor vessel water level.
In response to the July 1983 scram an operations instruction was is-sued which requires the Plant Control Operator to station observers in the upper and lower relay rooms during RPS surveillance tests to verify RPS status light indications during the reset of half scrams.
The cause of the blown fuse was found to be that an incorrect fuse type had been installed.
All of the other fuses were inspecte'd and found to be correct.
At least three occurrences of blown fuses to the scram pilot solenoid valves have been documented.
The two occur-rences in Unit 1 noted above resulte'd in reactor scrams.
One blown fuse was identified in Unit 2 prior to the start of a surveillance test in July 1985, and a scram was aver ted due to a precautionary check by the PCO.
3.5 The licensee is evaluating the possible causes for the fuse failures and is also investigating several spurious half-scrams and chattering relays that have occurred recently on Unit 2.
The F18 fuses are slow blow type FNM15 (15 amp),
and electrical maintenance measurements have determined that inrush current of 15 amps will occur occasional-ly when a half-scram is reset.
Licensee review of the FNM time cur-rent curve found it suitable for the required slow blow service.
The results of the licensee evaluation will be reviewed in a subse-quent inspection.
(387/85-31-01)
Reactor Scram Oue to 'B'oisture Se arator Hi h Level At 7:46 p.m.,
October 30, 1985, Unit 1 scrammed from 64 percent power due to turbine control valve fast closure on a turbine trip.
The tur-bine tripped on high level in the 'B'oisture separator (MS).
Con-ditions during the scram were normal.
'ifted.
The recirculation pumps tripped on the end of cycle recircu-lation pump trip (EOC-RPT)
as designed.
Lowest vessel level reached was 8 inches.
The inspector reviewed post-trip data, and monitored licensee's'or-rective actions.
Plant computer data indicated that the
'B'. MS'rain tank level rose very rapidly from the normal level to the trip, setpoint of 68 inches and remained above that level for greater,'han 10 seconds which causes a turbine trip.
This corresponds to several hundred gallons of water.
Other computer data (i.e. feedwater heater alarms, operation of MS drain valves, etc.)
supports the licensee's conclusion that the level rise was due to actual water and not flash-ing in the drain tank level control system.
I5C personnel performed
calibration of the MS drain tank level control system which identi-fied that one MS tank valve to the 4B feedwater heater was inopera-ble.
However, this discrepancy does not account for the note'd level rise.
During the investigation, it was determined that there was excessive spring can loading on the piping hangers for one of the 42 inch cross-around pipes between the HP turbine and the 'A'S, indicating that there was water in the line.
Manways for all six cross-around pipes were opened and the lines inspected.
Some water was found in two of the cross-around pipes and the pipe which had its spring cans displaced was drained prior to opening it.
The spring cans for this line returned to normal after draining.
The drain valves on these six lines were verified to not be clogged'he licensee was unable to positively identify the cause of the high MS drain tank level.
It is surmised that the source of the water was one or more of the cross-around pipes, but the licensee wa's un-able to identify a forcing function which would have caused the water to enter the MS at such a rapid rate.
The plant was at steady st'ate power at the time of the event and no anomalies were noted in ~plant parameters just prior to the event.
The accumulation of water',, in the cross-around pipes was probably due to condensation which was 'not fully drained during the unit startup which began on October 29.
It could not be determined, based on operator interviews whether, the cross-around drain valves were opened and remained open when the tur-bine was shutdown following the October 28 scram.
The licensee instrumented the MS drain tank level control system to allow monitoring by the G.E. Transient Analysis Recording (GETARS)
System.
The general operating procedure, G0-100-003, was revised to specify opening the cross-around piping drains for 15 minutes at about 30 percent power, to ensure the pipes are drained.
Thermocou-ples were attached to the drain lines to enable verifying thatjthe drain valves open.
Boosters were also installed on the MS emergency dump valves (LV-10232A and B) to enable quicker opening to help con-trol level transients.
This modification had been previously per-formed on Unit 2 following moisture separator level problems.
( Inspection Report 50-387/84-38; 50-388/84-47).
The licensee returned the unit to operation on November 6 and esca-lated to full power.
No,further anomalies were noted on moisture separator drain tank levels.
I 4.0 Licensee Re orts 4. 1 In-Office Review of Licensee Event Re orts The inspector reviewed LERs submitted to the NRC:RI office to verify that details of the event were clearly reported, including the accu-racy of description of the cause and adequacy of corrective actio The inspector determined whether further information was required from the licensee, whether generic implications were involved, and whether the event warranted onsite followup.
The following L~ERs were reviewed:
85-029-00, Inservice Inspection Surveillance Requirements Not Met Unit 2
- 85-025-00, Reactor Scram Caused by Lightning Strike on 500KV Transmission Linc'Discussed in Detail 3.2.
4.2 Review of Periodic and S ecial Re orts 4.3 RCIC Injections (Unit 1) dated September 27, 1985 Upon receipt, periodic and special reports submitted by the licensee were reviewed by the inspector.
The reports were reviewed to jdeter-mine that the report included the requi red information; that test results and/or supporting information were consistent with design predictions and performance specifications; that planned corrective action was adequate for resolution of identified problems, and wheth-er any information in the report should be classified as an ab'normal occurrence.
The following periodic and special reports were reviewed:
Monthly Operating Report September 1985, dated October 14, 1985
CFR Part
Re ort Yarway Valves 1r On September 26, 1985, Yarway Corporation provided a notification of a defect to Region I in accordance with 10 CFR Part 21.
The notifi-cation stated that a cracked stem assembly was detected in a 3/~4 'inch Yarway Welbond valve at a non-nuclear facility.
Five other valves exhibited leakage and Yarway concluded that the leakage was caused by a void in the bar stock used to manufacture the stems.
The bar~ stock used to manufacture the stems is 5/8 inch round bar, AL Tech stain-less steel type 416, ASTM A-562-75 condition T, heat number 93876.
The report indicated that valves with stems from this bar stock~ were sent to PP&L.
The inspector provided the Part 21 report to PP&L.
On October 8, 1985, Yarway Corporation notified PP&L of the defective valve stems.
Yarway indicated that 21 1/2-inch and 3 3/4-inch valves with defective stems were supplied to PP&L.
The licensee revieked the information and determined the location of the 24 valves.
Of the 24 valves, eleven are in the warehouse, three are installed in the
I
Unit 2 RHR system, four are installed in Unit 2 RHR pump room,'ooling piping, four are installed in the Unit 2 RWCU system, one is Iin-stalled in the Unit 1 offgas system, and one is unaccounted for.
Non-Conformance Report (NCR) 85-0462 was prepared to documentI the condition of these valves.
Yarway indicated that new valve stem/disc assemblies will be provided.
PPKL intends to replace the installed valve stem/disc assemblies at the next available outage.
TheIremain-ing valves in the warehouse were tagged with NCR tags and will be returned to Yarway.
This issue will be reviewed following disposi-tion of the valves.
(388/85-26-01)
5.0 Honthl Surveillance and Maintenance Observation 5. 1 Haintenance Activities The inspector observed portions of selected maintenance activi,,ties to determine that: the work was conducted in accordance with approved procedures; regulatory guides, Technical Specifications, and industry codes or standards.
The following items were considered during this review:
Limiting Conditions for Operation were met while components or systems were removed from service; required administrative 'approv-als were obtained prior to initiating the work; activities were ac-complished using approved procedures and gC hold points were established where required; functional testing was performed prior to declaring the particular component operable; activities were accom-plished by qualified personnel; radiological controls were imp)e-
'mented; fire protection controls were implemented; and the equ'pment was verified to be properly returned to service.
Activities observed included:
'O'SW pump discharge check valve repairs (WA-S55279) performed on October 10, 1985
'C'iesel generator maintenance performed on October
-, 18, 1985 Replacement of the Unit 2 'C'eactor Steam Dome high pressure switch (B21-N023C) performed on October 30, 1985.
The work was performed under WA-U57406.
5.2 Surveillance Activities The inspector observed the performance of surveillance tests to" de-termine that: the surveillance test procedure conformed to technical specification requirements; administrative approvals and tagout's were obtained before initiating the test; testing was accomplished by qualified personnel in accordance with an approved survei llanceI pro-cedure; test instrumentation was calibrated; limiting conditions for operations were met; test data was accurate and complete; removal and restoration of the affected components was properly accomplished;
test results met Technical Specification and procedural requirements; deficiencies noted were reviewed and appropriately resolved; and the surveillance was completed at the required frequency.
These observations included:
S0-100-006, Shiftly Surveillance Operating Log, performed on October 4, 1985 No unacceptable items were identified.
5.3 Emer enc Diesel Generator Survei llances a)
Engine over speed The inspector conducted a review of selected emergency diesel genera-tor (EDG) surveillance procedures to ascertain whether the survei 1-lance of safety-related systems and components is being conducted in accordance with approved procedures as required by Technical Specifications.
18 Month Lockout Feature Surveillance Unit 1 and Unit 2 Technical Specification Surveillance Re-quirement 4.8. 1. 1.2.d. 13 states that at least once per
months, each of the diesel generators shall be demonstrated operable by verifying that the following diesel generator lockout features prevent diesel generator starting and/or operation only when required:
b)
Generator differential c)
Engine low lube oil pressure The Unit 1 and Unit 2 Technical Specification/Surveillance Procedure Cross Reference Matrix prepared by the lice'nsee states that procedures SI-024-301 through 304 and SM-024-A02 through D02 implement the requirements of !Tech-nical Specification 4.8. 1. 1.2.d. 13.
The inspector reviewed the designated surveillance procedures to determine whether the Technical Specification surveillance requirement was adequately implemented.
The following items were identified:
Surveillance procedure SI-024-301, Revision 1, "18 Month Calibration of Diesel Generator 'A'ube Oil Low Pressure PSL-03468A1, A2, A3, A4", performs testing on the low lube oil pressure switch for the diesel gener-ator.
The automatic circuitry is not functionally
checked for proper operation (i.e. repeater relays and associated contacts).
Surveillance procedure SM-024-A02, Revision O,I,"18 Month 4KV Diesel Generator 'A'ifferential Relay Cal-ibration", performs testing of the 4KV Diesel Genera-tor differential lockout relay (87AG) by removing the relay and reinstalling it after a bench calibration.
The automatic lockout feature circuitry is not'tested during the surveillance.
The inspector could not identify any surveillan'ce testing performed where the engine overspeed lockout feature has been functionally tested, or verifi'ed to be operational.
The Technical Specification cross-reference also did not note any procedure to meet this requirement.
During discussions concerning the surveillance requiirements with the licensee, the inspector learned that an NQA~ audit finding (Finding No. 0-84-33-01)
was issued April 16, 1985 which apparently addressed the same issue.
The finding stated that current plant surveillance procedures doI not functionally test the three main trip relays for each start circuit as required by the Technical Specifications.I',
At the close of this inspection period, the audit finding was still open, although some preliminary responses haveI been submitted by the responsible departments.
The plant staff believes that the current calibration procedures are",suffi-cient for the low lube oil pressure switches and differen-tial relays.
Additionally, the staff believes that
~',
verification of completion of the diesel generator load reject test (SE-024-A05) without an overspeed trip is suf-ficient to show that the over speed trip will not occur when it is not required.
Since this apparent violation was previously identified by the licensee, a Notice of Violation will not be issued.
The, item will remain unresolved pending closure of the au-dit finding and completion of corrective action.
(387/85-31-02)
5.4 Surveillance Procedure Review - SO-100-005 The inspector reviewed S0-100-005, Weekly Electrical Power Systems Distribution Operability Verification, to determine if all requ'ired AC and DC power distribution system divisions are verified to be en-ergized as required by Technical Specification (TS) 4.8.3. 1. 1 and 4.8.3.2. 1.
Amendment 48 to the Unit 1 Technical SpecificationsI'and Amendment 14 to the Unit 2 Technical Specifications, which were',is-sued in July 1985, added the 480 volt AC swing bus motor generator
(MG) set, transfer switch and preferred and alternate power sources to the Technical Specifications required for power distribution.
When operations reviewed the amendments on July 12, 1985, they indi-cated that no surveillance procedure changes were required.
Inspec-tor review of the applicable surveillance, i.e.
S0-100-005, indicated that the preferred power source, the MG set and the transfer switch were not specified in the procedure.
The licensee was only verifying that the swing bus was energized and that its alternate source was available.
In response to the inspector's concern, the licensee re-vised SO-100-005 and SO-200-005 to specifically include the MG set, transfer switch and the alternate power sources.
6.0 Reactor Coolant S ill and RWCU Isolation Unit
6.1 Event Summar On October 11, 1985, at 10: 14 a.m.
the reactor water cleanup (RWCU)
system on Unit 1 isolated due to high differential flow and a',spill of reactor coolant and resin occur red on the 779 foot elevation of the Unit 1 Reactor Building.
Troubleshooting on the 'A'ilter de-mineralizer (F/D) effluent control valve (FV-14566A) was in progress at the time of the spill.
The spill was contained on the 779 'foot level.
A precautionary evacuation of non-essential personnel from both reactor buildings was ordered and the Technical Support C'enter (TSC) was manned.
The licensee did not activate the Emergency Plan nor did their Emergency Plan procedures require them to do so."
The precise source of the spill was not known.
General area radiation levels for the 779 foot elevation were 100-250 mrem/hour, with"levels up to 5 rad on contact with the resin.
Airborne activity levels remained less than
MPC (maximum permissible concentration) until the spill was nearly cleaned up.
Airborne levels reached 3.5 MPC for a short period as the resin dried out just prior to completion of the cleanup.
The recovery action consisted of 1) isolating the spill source and stabilizing the plant, 2) cleaning up the spill, and 3)
accessing the system damage, if any.
It was determined that there was no offsite release.
The shoes of one individual were contaminated but there was no other personnel contamination.
After the gross contamination was cleaned up, Tech Staff performed a system walkdown and observed no physical damage.
It was suspected that the spill came from a 150 psig relief valve in the precoat system.
On October 12, a test procedure (TP-161-007)
was performed to verify system integrity by introducing low pressure water to the low and high.
pressure portion of the system.
In addition, a full system lineup was performed and the system was walked down at full system pressure.
No problems were noted which could have led to the spill.
On October 13, at 5:34 a.m.,
the RWCU system was placed in service on the 'B'/D.
At 11:30 a.m., while attempting to place the 'A'/D in service, the 150 psig precoat relief valve (PSV-14561) lifted, (he RWCU system isolated and a minor spill (about 10 gallons) occurred.
There was no radiological hazard associated with this spill since the
'A'/D had been backwashe A few hours later, the RWCU system was restored on the 'B'/D and the 'A'/D was left isolated.
6.1
~6 10/11 5:08 a.m.
'A'WCU filter demineralizer (FD) taken out of service on permit to work on FD flow control valve (FV-14566A).
10: 14 a.m.
10:34 a.m.
(Time approx.)
Control room received RWCU hi-flow alarm followed by system isolation.
Spill of re-actor coolant and resin occurred on 779 foot elevation of Unit 1 Reactor Building.~
Control room received
"High Radiation Reactor Building" alarm due to an ala'rm on an area radiation monitor (ARM) on 779 foot elevation.
10:40 a.m.
11:15 a.m.
Personnel evacuated from 779 foot elevation of Unit 1 Reactor Building.
Evacuation of unnecessary personnel from the Unit 1 Reactor Building and 779 foot elevation of Unit 2 Reactor Building
,
ordered.
11:10 a.m.
11:50 a.m.
TSC ordered manned.
Unit 2 Reactor Building elevations 799, and 779 ordered evacuated of non-essential personnel.
jj 818, 12:05 p.m.
HP reported 100-250 mrem/hr general area radiation levels on 779; contamination lev-els of 10,000 dpm to 5 rads contact on res-in.
Airborne levels were about 75 percent of Maximum Permissible Concentration (MPC).
12:25 p.m.
12:25 p.m.
12:53 p.m.
The shoe of one non-licensed operator was determined to be contaminated.
There
'was no other personnel or clothing contamination.
ENS call to NRC.
HP reported that gross contamination (i.e.
resin)
was cleaned up.
13:33 p.m.
TSC deactivated, spill cleaned up.
No offsite release, RWCU remains isolate /13 6:34 a.m.
11:32 a.m.
RWCU system in servi ce on /D.
Prior to placing in service, system walkdown, valve lineup and pressurization test,,were performed to verify system integrity.'WCU isolation on high differential flow when placing the 'A'/D from precoat cycle to hold.
The 150¹ precoat feed relief lift-ed causing a minor spill.
Spill was "stopped when operator locally isolated the 'A'/D.
6.3 Details On October 11, just prior to the spill, the RWCU system was in ser-vice on the 'B'/D and the 'A'/D was tagged out to troubleshoot the F/D effluent flow control valve (FV-14566A) under WA S55381.
The valve appeared to be bound in the open position.
In order to 'istroke check the valve, an operator pushed the "Cleaning Cycle Start". push-button at the F/D control panel and manually stepped through the F/D cleaning cycle sequence by turning a thumbwheel on the programmer in the back of the F/D control panels Following completion of th'
cleanup cycle, the operator then pushed the "Precoat Cycle Start" pushbutton and manually stepped through the precoat sequence.
l, Fol-lowing completion of the precoat cycle (Step 15 of the programmer),
the operator turned the ISOLATE/DEISOLATE switch to DEISOLATE,, which causes the F/D block valves (HV-14531A and HV-14532A) to open.,
The operator and other individuals (maintenance mechanics and the Assis-tant Unit Supervisor)
heard flow.
They observed that the 'B'/D flow increased rapidly and the operator attempted to reduce the flow by operating the 'B'/D effluent flow control valve.
The operator then turned the ISOLATE/DEISOLATE switch to ISOLATE, pushed the HOLD pushbutton, and the individuals left the area as water was coming out of the floor drains.
The control room received the RWCU Hi-Flow alarm followed by system isolation and were notified of the leak.
A few minutes later,,a Re-actor Building High Radiation alarm was received which was determined to be due to an area radiation monitor at the sample station on the 779 foot elevation.
The Control Room ordered evacuation of unneces-sary personnel from 779 foot elevation and later from all Unit
Reactor Building.
The control room also began vacuum dragging
~the RWCU system to the condenser to reduce system pressure.
Initial radiological conditions were determined to be 150 mrem/Pr general area and it was noted that resin was on the floor.
The".
licensee manned the Technical Support Center (TSC) but did not acti-vate the Emergency Plan (the event did not meet Emergency Plan entry conditions).
A precautionary evacuation of unnecessary personnel from the Unit 2 Reactor Building 818, 799 and 779 foot elevations was ordered.
Radiation levels increased to 150-250 mrem/hr general, area,
contamination levels up to 5 rad on contact with resin and airborne levels reached 75% of MPC for affected nuclides.
The isolation of the RWCU system stopped the spill.
Health Physics personnel entered and cleaned up the gross contamination.
There was no offsite release.
One shoe of the operator at the F/D contr'ol pan-el was contaminated, but there were no other personnel contami'na-tions.
Following cleanup of the spill, Technical Staff engineers walked down the system and verified that there was no system damage.
On October 12, the licensee performed a test procedure to introduce low pressure water from the condensate system to the low and high pressure portions of the system to check for damage.
None was', found, although leakage past one of the system isolation valves had already repressurized the system.
The system was vented at high points, and the 'A'nd 'B'/D were backwashed.
On October 13, the '8'/D was precoated and placed in service.
At ll:32 a.m. October 13, while attempting to place the 'A'/D in service, a
RWCU system high, leak-age alarm occurred, the system isolated and the precoat relief valve (PSV-14561) lifted causing a minor (10-15 gallon) spill.
There was no radiological hazard from this spill since the 'A'/D had been backwashed.
The operator had positioned the ISOLATE/DEISOLATE switch to DEISOLATE following the precoat cycle when this spill occurred.
The operator immediately returned the switch to ISOLATE when he heard flow.
The 'B'WCU F/D was returned to service and the 'A'/D re-mained isolated.
Plant staff and the Nuclear Safety Assessment Group performed an in-vestigation of the incident.
The cause of the October 11 spill was apparently due to a limit switch problem with the 'A'/D effluent flow control valve (HV-14566A).
The full closed limit switch on the HV-145066A energizes a relay if the valve is not fully closed, which causes the F/D influent valve (HV-14506A) to open.
The limit s'witch collar on the HV-14566A was found to be loose and it apparently caused HV-14506A to open.
When the operator deisolated the vessel, a
flow path existed to allow reactor coolant to reach the 150 psig precoat piping, lifting the 150 psig precoat relief valve.
Normally, prior to deisolating the F/D, a backwash cycle would have been
.com-pleted.
However, on October 11, the operator was manually stepping through the programmer which did not allow the backwash to be p'er-formed.
The 'A'/D had been backwashed and returned to service on October 10.
The activated resin, which spilled on October ll causing the radiological hazard, had only been in service for one day.
Nor-mally, a F/D is inservice for a longer period (nominally six to, eight days) prior to backwashing which implies that consequences of a,'spill could have been more severe.
The minor spill on October 13, was apparently due to gross leakage by HV-14566A.
This valve is an air operated globe valve manufactured by VALTEK, and the valve contains soft seats.
A leak check of thel valve on October 16, identified gross leakage.
When the valve was disman-
tied, the soft seats were missing, accounting for the gross leakage.
The valve was repaired.
The programmer design requires that the F/D vessel be deisolated (i.e. the isolation valves, HV-14531A and HV-14532A, opened) prior to returning to hold.
The HOLD pushbutton does not receive powerl until the vessel is deisolated following the precoat sequence, in the ex-isting design.
Therefore, the F/D is in precoat with low pressure piping connected to the vessel, when the vessel is deisolated.,
A failure of one valve (i.e.
the F/D influent or effluent valves) which is not designed as an isolation valve, will cause a spill.
The licensee modified the programmer circuitry under DCP 85-9054 to re-move the block valves and operation of the HV-14506A valve from the programmer.
This change ensures that two valves in series will iso-late the F/D precoat system from reactor pressure.
However, based on discussions with the licensee for Limerick, and Graver (the F/D con-trol system manufacturer),
this is a generic problem and will be pur-sued separately.
Following valve leak checks and repairs to HV-14566A, the 'A'/D was returned to service on October 22.
The Unit
RWCU operating proce-dure OP-161-001 was revised to reflect the modification and to',ensure manual isolation valves to low pressure piping (i.e.
F/D vent piping and precoat piping) are shut prior to deisolating the F/D vessel.
The Unit 2 procedure was also revised to ensure the manual isolation valves are shut, however, the modification has not been completed on Unit 2.
In the inspector's view, the licensee's response to this event,'n-cluding the immediate response, spill cleanup, and actions to assess RWCU system status and return the system to service, were thorough and conservative.
The spill was quickly isolated, the spread of con-taminationn minimized, and the spill was cleaned up with no personnel contamination and very little personnel exposure.
Several PORC, meet-ings were held on October
and 12 to determine recovery actions, a
system walkdown and complete valve lineup were performed, and t'est procedures prepared to verify system integrity.
These steps were taken prior to any attempt to restore the system.
The licensee, closely monitored reactor coolant chemistry using the post-accident sampling system while the RWCU system was out of service.
Although a second, minor spill occurred on October 13, the licensee took prudent actions to prevent this occurrence and there was no hazard from the second spill.
At the time of this report, the licensee's investigations had not been completed and further corrective actions have not been determined.
The inspector will review the LER and investigation reports when they are issued.
(387/85-31-03)
i
7.0
~Ei ft iI On November 15, 1985 the inspector discussed the findings of this Iinspec-tion with station management.
Based on NRC Region I review of this report and discussions held with licensee representatives, it was determined that this report does not contain information subject to 10 CFR 2.790 restrictions.
II
0