ML20147B247
| ML20147B247 | |
| Person / Time | |
|---|---|
| Site: | Salem  |
| Issue date: | 07/21/1994 |
| From: | Eugene Kelly NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| Shared Package | |
| ML20147B009 | List: |
| References | |
| FOIA-96-351 NUDOCS 9701300112 | |
| Download: ML20147B247 (1) | |
Text
__
.p euru 8
[*g UNITED STATES NUCLEAR REGULATORY COMMISSION o
- U REGloN 1 e'['
475 ALLENDALE ROAD
,,g KING OF PRUSSIA, PENNSYLVANIA 19406 1415 JUL 211994 i
Docket Nos. 50-272 50-311 MEMORANDUM FOR:
Distribution FROM:
Eugene M. Kelly, Chief, Systems Section, Division of Reactor Safety
[
SUIUECT:
SALEM AIT ESCALATED ENFORCEMENT -
BRIEFING PACKAGE 1
4 Attached is background information for the open conference scheduled at 10:00 a.m. on July 28,1994.
~
l We plan to discuss NRC positions and strategy in a prebrief on Wednesday, July 27,1994, at 3:00 p.m. in the Executive Conference Room.
Questions should be directed to me (610-337-5111) or Peter Drysdale of my staff (X5289).
d'
(
A Eugene M. Kelly,/.#
i Chief
/ Systems Section J
Division of Reactor Safety Attachments:
- 1. Meeting Notice e
- 2. July 6,1994, letter (Martin to Miltenberger)
- 3. Escalation / Mitigation Analysis
- 4. Salem Enforcement History
- 5. Analysis of Previous AITs (June 3,1993; December 13,1992; and November 9,1991)
- 6. Salem AIT Report 94-80 (Cover and Executive Summary) 4
- 7. Excerpts from Reports 94-11 (Section 4.2) and 94-13 (Section 4.1) t 9701300112 970124 PDR FOIA O'NEILL96-351 PDR
U. S. NUCLEAR REGULATORY COMMISSION 44 REGION I No. 94-86 NOTICE OF SIGNIFICANT LICENSEE MEETING EA 94-112 i
Licensee:
Public Service Electric & Gas Company E
f Facility:
Salem Generating Station i
Docket No:
50-272 Date and Time:
Thursday, July 28,1994 at 10:00 a.m.
location:
NRC Region I,475 Allendale Road, King of Prussia, Pa 4
Main Conference Room
[
Purpose:
The purpose of this Enforcement Conference is to discuss the apparent 4
violations identified from NRC's review of the April 7,1994, automatic I
shutdown of Salem Unit 1, including cause, safety significance, and corrective actions. The meeting was requested by the NRC as a result of findings identified by the Augmented Inspection Team and other NRC inspection acuvities related to the event.
[
NRC Attendees:
T.T. Martin, Regional Administator, Region I J. Wiggins, Division Director, Division of Reactor Safety (DRS)
J. Grey, Deputy Director, OE E. Wenzinger, Chief, Projects Branch 2, DRP J. White, Chief, Reactor Projects Section 2A, DRP C. Marschall, Senior Resident Inspector, Salem / Hope Creek E. Kelly, Chief, Systems Section, DRS
[
J. Stone, Project Manager, NRR Other Members of Senior NRC Staff Licensee Attendees: S. E. Miltenberger, Vice President and Chief Nuclear Officer J. J. Hagan, Vice President-Operations / General Manager-Salem Operations F. 'Ihomson, Manager, Licensing and Regulation Other Members of PSE&G Senior Staff Note: This meeting is open for public observation. Handicapped persons requiring assistance to attend the meeting shall make their requests known to John White, Chief, Reactor Projects Section 2A, U. S. Nuclear Regulatory Commission, Region I,475 Allendale Road, King of Prussia, Pa., 19406, (610) 337-5114. Attendance by NRC personnel at this meeting should be made known by 4:00 p.m., July 27,1994, via telephone call to John White, Region I.
Prepared by:
[
Edy(trd C. Wenzinger, Chief Jtfactor Projects Branch No. 2 Cqopagp p,
JUL 6 1994 9
Steven E. Miltenberger 3
In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy of this letter and its enclosure will be placed in the NRC Public Document Room. No response to this letter is required at this time, and your cooperation is appreciated.
Sincerely, pmes T. Wiggins, Director Division of Reactor Safety Enciosures:
1.
Apparent Violations Considered for Escalated Enforcement Action (Salem AIT Repon 94-80) 2.
Federal Register Nxice (Vol. 57, No.133, July 10,1992); "Two-Year Trial Program for Conducting open Enforcement Conferences - Policy Statement" cc w/ encl:
[
J. J. Hagan, Vice President-Operations /Ge&ral' Manager-Salem Operations S. LaBruna, Vice President - Engineering and Plant Betterment C. Schaefer, External Operations - Nuclear, Delmarva Powei & Light Co.
R. Hovey, General Manager - Hope Creek' Operations F. Thomson, Manager, Licensing and Regulation R. Swanson, General Manager - QA' and Nuclear Safety Review J. Robb, Director, Joint Owner Affairs 1
A. Tapert, Program Administrator R. Fryling, Jr., Esquire M. Wetterhahn, Esquire P. J. Curham, Manager, Joint Generation Department, Atlantic Electric Company Consumer Advocate, Office of Consumer Advocate William Conklin, Public Safety Consultant, IAwer Alloways Creek Township K. Abraham, PAO (2)
Public Document Room (PDR) local Public Document Room (LPDR)
Nuclear Safety Information Center (NSIC)
NRC Resident Inspector State of New Jersey State of Delaware I
?
L1 l
ATTACHMENT 2
!(
[fugk UNITED STATES i%
g NUCLEAR REGULATORY COMMISSION a
En REGION 1
'g 475 ALLENDALE ROAD KING OF PRUSSIA. PENNSYLVANIA 19406-1415 l
g....*
l JUL 6 1994 i
- Docket Nos.
0-272
- 11 l
Mr. Steven E. Miltenberger l
Vice President and Chief Nuclear Officer Public Service Electric and Gas Company
[
P.O. Box 236 i
j Hancocks Bridge, New Jersey 08038 l
Dear Mr. Miltenberger:
1 SURIECT:
APPARENT VIOLATIONS RELATIVE TO THE APRIL 7,1994 EVENT The NRC inspection findings relative to the circumstances surrounding the April 7,1994, ll partial loss of plant circulating water flow and subsequent plant trip and safety injections, wEre detaiteil in the Augmented Inspection Team (AIT) Report 94-80, jinivided'to you by our letter, dated June 27,1994. Further discussion of several of those findings is addressed in NRC Inspection Reports 94-11 and 94-13. These reports identify findings that we consider important to the safe operation of the Salem units, and either contributed to or complicated recovery from this event, or were indicative of weak management controls of important plant activities.
1 The April 7th event is of concern to the NRC since it resulted in multiple challenges to the Salem operators and plant safety systems. The control room command function was transferred or relinquished for pivotal parts of the event, and appropriate actions to counter the reactor cooldown in progress and prevent the ensuing safety injection transients did not occur. Longstanding equipment deficiencies that also led to, and complicated the event, specifically, the atmospheric relief valve controls and steam line flow sensors, were tolerated and corrective actions were not aggressively pursued. Communications with the NRC during the early stages of the event were ineffective in characterizing the significance of the transient, including the scope of equipment failures, the causes of safety system initiations, the resultant condition of the plant, and the planned recovery efforts. Further, procedures for dealing with the abnormal plant conditions (namely, grass intrusion, rapid power i
reductig and recoveryfrom a water-filled pressurizer ofi511d plant" c6iidition) were Eequate, insufficiently detailed, or nonexistent. FiMiy, design and material controls were found to be inadequate,'in that the pressurizer power operated relief valve (PORV) internal component materials, although satisfactory for use, were not as specified by the design; and a solid state protection system (SSPS) logic card was subsequently determined to be of the wrong type.
i
Pl 3
10 CFR 50, Appendix B, Criterion VIII, Identification and Control of Materials, E.
Parts, and Components, requires in part, that measures be established for the
-identification and control of parts and components. These measures shall assure that identification of the item is maintained throughout installation and prevent the use of incorrect parts. The following are two examples that demonstrate a failure to maintain configuration control:
- 1) During the 1993 Unit 2 outage, power operated relief valve (PORV) internals made of 17-4PH stainless steel (original design material) were installed in valves 2PRI and 2PR2, in lieu of internals made of type 420 stainless steel (the vendor-
,l t!
recommended and licensee-approved design change replacement material).
- 2) The post-trip investigation for the April 7,1994, event identified that the installed summator module for the high steam flow setpoint did not have the proper identification and contained an incorrect electronic part. Although the licensee determined this component did not affect the plant response during the event, it is an example of failure to properly identify components prior to installation.
F.
Technical Specification (TS) 3.5.2 requires two operable emergency core cooling system (ECCS) injection systems, or a plant cooldown to below 350*F within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. During the event on April 7,1994, and following reset of the safety injection actuation signals, automatic actuation capability was not available (and not reinstated) because the reactor trip breakers were not cycled. This occurred because there was no procedural guidance for re-establishing the safety injection logic, nor was it clear that the logic was operable, and because a cooldown could not be completed within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> due to the time required to re-establish a pressurizer steam bubble.
i This was recognized by the licensee, in that, TS action 3.0.3 was entered, but enforcement discretion was later granted by the NRC in recognition that the plant would not be below 350'F within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> (or by approximately 5:00 p.m.) of the safety injection actuations that occurred on April 7,1994.
j
i
~d ENCIDSURE 1 APPARENT VIOLATIONS CONSIDERED FOR ESCALATED ENFORCEMENT ACTION (AIT REPORT 50-272,50-311/94-80; AND INSPECTION REPORTS 54 272, 50-311/94-11 AND 50-272, 50 311/9413)
A.
Technical Specification 6.1.2 requires that the Senior Nuclear Shift Supervisor (SNSS) or, during his absence from the control room, a designated individual, shall be responsible for the control room command function. Technical Specifications require administrative procedures, as referenced in Regulatory Guide 1.33. PSE&G Administrative Procedure NC.NA-AP.ZZ-002(Q), Attachrnent 32, Shift Management
[!
Responsibility for Station Operation, requires, in part, that the SNSS shall remain free to survey and analyze all operating parameters. The procedure further states, " intense involvement in any particular detail may run the risk of losing control and perspective of the overall operation." The following two examples on April 7,1994, demonstrated a loss of perspective regarding overall plant safety and the command function of the senior nuclear shift supervisor:
[
1)
The SNSS left the control room during the loss of circulating water transient to override a circulator pump protective interlock, and thereby relinquished the command function in the midst of a significant plant transient. During his absence, operators caused reactor coolant temperature to decrease below the minimum temperature for criticality; and 2)
While the SNSS was absent from the control room, the nuclear shift supervisor (NSS), designated as responsible for the control room command function,
[,
assumed the duties of a reactor operator by performing control rod movements. As a result, for the period of time the NSS was manipulating the controls, no individual was properly exercising the control room command function.
B.
10 CFR 50, Appendix B, Criterion XVI, Corrective Action, requires in part, that licensees identify significant conditions adverse to quality, determine their causes, and take corrective action to preclude recurrence. Two examples of failure to meet this requirement occurred:
- 1) During previous reactor / turbine trips on June 10,1989, July 11,1993, and February 10,1994, the licensee failed to identify and correct the cause of spuriou:: high steam flow signals. As a result, the licensee failed to preclude an unnecessary safety injection actuation on April 7,1994, in response to a reactor trip; and l
~____ _ _ ___.._._._
l i
l Fedens Itagisest / Vol Sr. No. 2as / Prider July to, taas / Nodses as I
i emblems topasasemeleersenha,that l. LQuetuteForsenestesOpus tisse estegories of hoeseene weg be Essammentcessamese commsemal
- reactem, eigen, bessern, peseus, eaa. net larger Wher wh64 thes 3r he pesadsted,and eat i
Enforcement onefemeces wiu not be l
opes t2 the ymblesif she esferosesset osmanof theeme&aiogtypesof disreyeve pemene may be removed.
licensees.
Each regleseleSee esGIosamese to assion
- eomasepleted>--
coedmet the enfassementemelaremos (1) W be taham apenest as R. Asesmusing Open Befuessenes indiv6dmak or if the acosa, theech est Condnesses proceedings As amendemos wie reposal preceaa.no enfesammentaseference taken against anindividual turns as As esos as11is detarmined that as wiu condene a to a needes h wbother anindividualhas e===au'a.'l
'8f8'emment confersace wGl be open to eeNRC ades h m b public observ os, the NRC ertu orsUy enforcement assierenceis open sw
'davo6es significant personnel noth t se anforement public observeten,itle estopen for
(
fansree where the NRC has requested confance wfD be open to puhuc puhhc participensa.
I th:t the ladividual(s) involved be suon e se eW pareens attamens open esferoement present at the conference:
and
& Bconseea copy d l
(3)la based on the Andings of an NRC 7FedeelRegiser artice that outees conferences en reateded that p) the apparent violoalese disonesed at open l
OSee ofleveedganoes(OT) report;or I
wS be asked 2 anfJemama=8 esadescones are ashvect to g,%, g'Iyg (4) Involves safegannie tafaamanen, f
og,Q furtherrestow and may be ambient to
{
l Prtvecy Actinformanlaa or other
,ggg e angeprlwwanymeannes lafomanen which could be canaldered so that the NRC can scheduleas enforcementaedom and p)the proprietary.
- I*"""'***'
i Enforcement conferences lavolvtag
'gg e'a'a==a*= of views or expressione of
- pyropnees opinion made by NRC employees at medical misadmialstredone or hteDahoeoscus m
opea saforcement onaferences or the overszposures willbe open asemmles odorsement assieremos bus been lack esmef.em em totaded e i
the conferemos ons be conducted scheduled and thatitla opes a puhus represent Real deteradastlose or bebefs.
I withoet disclosing the exposed observeden.
In addnion e providleg comusets on l
ladividual's name.In addition.
De NRClatends to mananae=
enfoneessem osoferesese wit not be enforcemam enaderances e es y the seency's trial proyem in econdlance wie es guidassein tide no6ce.
l cpas tf es confeness wG normaDy atleast to working days in enemhes apes enaonemem e
be by telephone er ee advance of the enforcement conferosos will be provided as opporteelty to conference wtB be oceducted at a g, g,g,,,,,,m subett writtes commente seenymously relatively emailliosasee's facthty.
gy p,,,g g g, p,gg, to the regloaalafBon. Sees comments Finally, with the approvalof the na,
==t Rases Executive Director for Operations.
(2) Tee.dree *=n=ta========p= and wiU embosquesdy be forwarded to the Dtrocaer of the OlBoe af Enforcensent for enfore====e confersaces w15 act be (3)you. free elecognic beusen board open to the pulsein opadalcasse review and esealderados.
wham good anaes has been aboue aAar establishmentof thetoOeree Deted as asakiese, is, shie rit der et July balancing the beneRt of pehlis mesesse systeese,the pub 8e may can test ebeervanes assiest the pasadalimpaat (301) esB 4F3s to obtain e recording of For theNedeerRaydesaryc-e i
- 3 the asemer's seiercoment actian ia a opcomtegopen esforcement sessell. Chm Pardcolar caan, conferemose.The NRC win leses emother 3,c,eemyeMr Commamanes.
Tbs NRC will strive to ooeduct opee Fedeel Regleter netos enwes tou.tre p um eq w 4 enforesenent conferences dertog th**
message eyesens we estAh h=&
two year arm!programin accordamos To eselet he NRC la aseidag-l with tha fouowsag three goaler-a to errsagemente to support t
(1) Apprenansaaly as percent of aR i
eugible enforomment conf.esaw mese.am,,enen of endensesse,,
observ ains me,tdeais i.,,e.,
conducted by ee NRC wul be opse der stament a peremin enforemens j
pubueab" " " "'
codman doeld noe se mevidual Correctl0n8 'wel a+sw (2) At least aan open enfareement IdeutlSedin the emeeg mottos Vol. sr. No.13e conference wdl be conductedin ensk of anseendag the spea enfoneesem
~
ths onaloffices and
. h me low thaa tve banisms Fredey July 17.10!st '
(3) n enforoeseent somiensons days prior to the enforonmem t
wiu be conductedwith averlety of the conferseas.
I types of hcanaeos.
EL Condent of Open Enferomment Ifuct E.A.f.t ftE, OUI.AT0ftY To avoid potentialbiasla the en n
Confueses selection process and to attempt to meet ths three goals etsted above,every in econdemos with currentproctice.
Thee YearTfteiprogress for fourt shgible =8a=====* conference enforcement conferences will conttaue Conducting Open Enferoement involving one of three categories of to na== ally be bald at the NRC Conferenoses posey statemem licensees wGlnormeDy be eyes to Ibo omces. Members of the pubbe be i
public dertog the trial propea.
allowed access to the NRC. 78 i
Hrwever,in cases where there is en omens to eased open eedorcommet la motice deemeent es teass beginning i
ongolaC edtudicatory proceeding with
' conferences in accordamos with the on peas sorst la the lesse of Friday, Ene or more totervenore, soforcement
" Standard Opereting procedern For July.10, test, os page aorst in the conf:rences tavolving tesame related to providiesSeoartty Sapport For NRC seceed solmaa, sader Dates, beginalag the subject matterof the ongoing He And Meetings" published is the afth time. " July 1L 1982" abould l
adfudication may alas be opened.For No
),Imrt(se FR sazsn.These reed " July 1L tees *.
{
the purposes of this trielprogram, the procederse ptovide that viettore may be naame e um,e i
l
r B.
Spurious High Steam Flow Signal 1.
Identification:
The licensee failed to correctly identify the root cause of spurious high steam flow signals identified during three previous reactor / turbine trips. The high steam
. flow signal during the April 7,1994 event was self evident in that it provided the coincident logic (with low low Tave) for a safety injection signal.
2.
Corrective Action:
Prior to the restart of Unit 1, PSE&G installed a modification to' dampen the signal from the steam flow transmitter. The licensee's preliminary evaluation of I
the modification's performance following the April 10,1994 trip indicates that the
, modification did not completely correct the problem.
3.
Licensee Performance:
l Engineering and Technical Support were rated as Category 2 during the SALP l
assessment period. Three previous significant occurrences involving nonsafety-related equipment failures'have occurred since'1991. Specifically, the three previous AITs were in response to the following:
1991 Turbine Overspeed Event 1992 Loss of Overhead Annunciators 1993 Unexpected Rod Withdrawal 4.
Multiple Occurrences:
The rod control system and MS-10 control system are two examples where PSE&G failed to correct known deficiencies in nonsafety-related equipment. The inadequate corrective actions for these deficiencies complicated the April 7,1994 event, contributed to the challenges of safety-related equipment, and caused a reduction of the margin to safety.
5.
Prior Opportunity to Identify:
The licensee had at least three previous opportunities to identify and correct the root cause of the spurious steam flow signal.
6.
Duration:
This issue concerns the original system installed by the vendor. Post trip data from a June 1989 trip shows the spurious signal was present at that time.
4 l
~ ~
n,, i......
~
i
- ESCALATION / MITIGATION ANALYSIS cc Evaluation per 10 CFR 2, Appendix C, VI.B.2 Civil Penalty Adjustment Factors The adjustment factors below are for the two most significant apparent violations described in the enclosure to Mr. Martin's letter to Mr. Miltenberger, dated July 6,1994.
I A.
Control Room Command Function 1.
Identification:
f The inappropriate actions of the SNSS and NSS were identified by the AIT during interviews with individuals present in the control room area during the April 7, t
1994 event.
2.
Corrective Action:
l
[
Prior to the restart of Unit 1 the licensee provided operator training on the J
scenario for all shifts. Actions were taken to counoff the individuals involved in the event on expectations for crew performance. $*
j i
3.
Licensee Performance:
Plant Operations received a Category 2 rating during the last SALP assessment period. No violations involving plant operators and the control room command 4
function were identified in the last two years, 5
i 4.
Multiple Occurrences:
The SNSS and NSS both relinquished the command and control responsibility during the event. This indicates that the operations staff did not fully understand the intent of management expectations described in the administrative procedure.
5.
Prior Opportunity to Identify:
None was identified.
6.
Duration:
The SNSS left the control room during the loss of circulating water and returned just prior to the reactor trip. During transfer of house loads, the NSS stopped directing the procedure, performed the duties of an NCO (by moving rods), and then returned to directing the transfer of house loads.
SALEM VIOLATIONS APRIL 1,1992 TO JULY 12, 1994 REPORT ISSUED SEVERITY AREA DESCRIPTION 92-07 7/2/92 IV SA/QV Control room habitability, storage of ammonium hydroxide not communicated to NRC for evaluation.
4 i
a 92-11 7/30/92 IV M/S Failure to follow procedures regarding control of measurement and test equipment.
l 92-16 12/16/92 IV E/TS Inadequate corrective actions by engineering for identified deficiency regarding fire damper.
j 93-08 5/5/93 IV SA/QV Weaknesses identified in the licensee's implementation of the 10 CFR 50.59 program.
93-11 5/7/93 IV SEC FFD supervisory trr.ining issues.
93-15 7/12/93 IV -
OPS Availability of steam driven AFW during i
entry into Mode 3.
93-21 11/3/93 IV OPS Failure to initiate a timely shutdown of J
Unit 1 following failed surveillance test.
(Licensee identified) l 93-23 1/10/94 IV M/S Diesel generator air start system j
operability.
93-23 3/9/94 III M/S Failure to follow maintenance 4
(Violation resulted from Enforcement Action - Civil Penalty) procedures.
i 1
93-82 11/30/93 IV OPS /E/TS Failure to follow station procedure for l
measuring battery cell voltages.
93-27 2/10/94 IV M/S/ OPS Inadequate control of troubleshooting and corrective actions regarding RHR check valve leakage.
94-06 4/26/94 IV OPS Failure to comply with TSAS regarding PORVs.
94-07 6/29/94 IV E/TS No written safety evaluation for replacement 460V vital bus transformer.
94-16 6/30/94 IV P/S Access Control of Vehicles.
i Troubleshooting procedure has been significantly revised to provide more comprehensive guidance relative to root cause and troubleshooting processes.
Vendor supplied materials are required to go through Procurement and Material Control inspection prior to use in the plant.
Component and circuitry modifications and repairs completed and tested satisfactorily; design changes instituted for digital demand step counters.
r 1
I nd i
)
4 k
____ _____ j
I 4
50-311/93 81 JUNE 5,1993 REVIEW OF CIRCUMSTANCES SURROUNDING SEVERAL ROD CONTROL SYSTEM FAILURES AT SALEM UNIT 2 DURING START UP ACTIVITIES FROM MAY 25 i
THROUGH JUNE 3,1993 3
I BACKGROUND:
l During startup, Salem Unit 2' experienced multiple failures of the control rod control system 4
(CRCS). Five plant startup attempts were made from May 25 to June 3,1993. The failures involved failure CRCS to move the rods correctly and maintain them in the proper position. The most serious event involved erroneous indications on rod position indicators as compared to step counters, and withdrawal of a rod from the core without a demand signal and without response to operator actions to control the movement until power was removed form the rod. Due to the numerous abnormal occurrences in such a short time, NRC was concerned about the approach the licensee was using to determine cause and establish effective corrective actions.
ROOT CAUSE:
Process causes:
Restart process (station policy and procedures) did isot provide for programmatic determination of root cause of system failure, i.e., there was no clear policy on when, how, and to what extent, to perform root cause analysis for component failures.
Hardware causes:
Multiple integrated circuit and output transistor failures due to unsuppressed voltage spikes.
Regulation board short circuits created during manufacturing or maintenance activities.
Slave cycler logic card failures due to power supply short circuit or unsuppressed voltage spikes.
Q9 transistor failures due to jumpering activities performed for troubleshooting.
Wrong resistor installed on failure detector card due to manufacturing error.
CORRECTIVE ACTIONS:
(Applies to all Salem units)
CPAT effort is expected to address the lack on organized approach to root cause analysis and troubleshooting activities.
Standing Orders have been developed to provide additional guidance to operators relative to operability determinations (CPAT actions are also expected to enhance operability guidance).
3 m
.- ~ -.
i i 50-311/92-81
~
DECEMBER 14, 1992 REVIEW OF CIRCUMSTANCES SURROUNDING LOSS OF THE OVERHEAD ANNUNCIATOR SYSTEM AT SALEM UNIT 2 ON DECEMBER 13, 1992 BACKGROUND:
On December 13,1992, a Salem Unit 2 operator discovered that the overhead annunciators had not been updating alarms for about 90 minutes. The operator reset and restored the system 1
within two minutes. Subsequent investigation revealed that a member of the operating crew had been using the Remote Configuration Workstation (RCW) computer and had inadvenently entered a keystroke combination that, when input through the wrong system pon, prevented the proper operation of the overhead annunciator system.
ROOT CAUSE:
An operator failed to follow procedures relative to the operation of the Remote Configuration Workstation (RCW) which affects the function of the Sequence of Events Recorder and its interface with the overhead annunciator.
l\\
No alarms or warning were available to alen the operator of mis-positioned switch at the.RCW, therefore operators were not aware of OHA status, ae.
Design specifications for OHA weretTA relative to alarm / warning features Other items:
,I No loss of annunciator procedure.
i
\\
Operators were not trained on routine verification of proper system operation.
LTA software review.
Operators knowledge of need to declare Alen on system failure.
~
General communications with NRC CORRECTIVE ACTIONS:
Design change to incorporate alarm / warning feature to warn of OHA system malfunctioning.
New procedure for loss of OHA 6
i CORRECTIVE ACTION EFFECTIVENESS:
Recurrence of problems of this type have not been observed since this event.
4 i
5
50-311/91-81 NOVEMBER 10,1991 REVIEW OF CIRCUMSTANCES INVOLVING THE CATASTROPHIC FAILURE (DESTRUCTION AND FIRE) OF THE SALEM UNIT 2 TURBINE-GENERATOR ON NOVEMBER'9,1991 BACKGROUND:
The Salem Unit 2 Turbine-Generator was severely damaged, while at 100% power, on November 9,1991, during performance of Front Standard testing of autoinatic mechanical turbine trip features. During the test, normal turbine trip functions are isolated and reliance is placed on the ET-20 emergency trip solenoid and two overspeed protection solenoids to function.
[
While testing, a momentary AST oil pressure perturbation resulted in momentary closure of turbine steam admission valves, reactor trip, and subsequent load drop from the generator.
However, the ET-20 and the overspeed protection solenoids failed to function due to' mechanical binding. Consequently, once AST oil pressure returned to normal, hydraulic fluid which normally would have been drained by the functioning of the turbine control solenoid valves, was pressurized to reopen the steam admission valve to the turbine. Subsequently the turbine experienced severe overspeed without any abatement until the operators restored the Front Standard test controls to the normal position, at which time AST-20 functioned to close the
{
turbine steam admission valves. The turbine generator was subject to severe damage and fire.
ROOT CAUSE:
Three separate solenoid valves failed to function as designed to control turbine overspeed and effect turbine trip.
Insufficient preventive maintenance afforded to the turbine overspeed protection system.
Inadequate surveillance testing of solenoid valve-actuated turbine control systems.
Delayed replacement of Unit 2 solenoid valves after previous component failures were identified at Unit 1.
Failure of management and operating personnel to follow procedures and effectively resolve a failed test result involving the turbine overspeed protection system during a previous turbine startup procedure on October 21,1991.
CORRECTIVE ACTIONS (applies to both Units,):
Disciplinary action taken against individuals who failed to follow the requirements of the turbine stanup procedure on October 21,1991.
8 L
New security features added to RCW to prevent inadvertent or non-authorized manipulation of system diagnostic computer.
Operator training lesson plans developed and incorporated in training program; OHA malfunction added to simulator training program, t
Personnel disciplinary action taken.
NRC reporting and ECG criteria clarified relative to loss of OHA New procedures developed and issued rela
.o specifications for software and software verification and validation processes; revisiva of vendor technical manual by vendor; lesson plans developed for system training.
i PSE&G and vendor developing less complicated system testing process.
Software virus checks perform'ed.
Managements expectations relative to communicating concerns with faulty system operation conveyed in series of roll-down meetings.
I' CORRECTIVE ACTION EFFECTIVENESS:
No similar problems have been noted since 'his event.
7
i '
OVERALL CORRECTIVE ACTION EFFECTIVENESS:
Recurrent instances of failure to follow procedures despite numerous efforts by management to ameliorate this common problem.
Management effectiveness is questionable and a contributor in each occurrence.
Operator training is usually an identified item that requires remedial action.
Communication effectiveness is a common feature in these events.
F
-4 l
Ee i
s 10 i
l
1 Management expectations communicated to staff relative to the importance of procedure adherence.
Operator training revised to include simulator and classroom instruction relative to turbine control system functioning and testing; all operators trained; operator effectiveness evaluated.
Turbine startup procedures revised and upgraded.
" Conduct of Operations" training, relative to procedural adherence, provided to station staff.
Human performance analysis initiated with lessons-learned to be incorporated.
i Commitment Management procedure developed and issued to assure that all commitments and previous LERs are tracked and completed in a timely manner; several other planning and scheduling procedures were revised to assure the proper assignment of priority to items that require action.
Preventive maintenance program established for orifices in the AST system.
Design change for filter installation on AST orifices completed.
Front standard test procedure upgraded and revised.
INPO Awareness Training conducted, Resource Management, and Supervisory and Management Effectiveness Solenoid valves replaced; turbine EHC systems and components have been incorporated in Reliability Centered Maintenance program to assure PM performance; vendor identified PM specifications.
Surveillance program for solenoids revised to independently test each unit and verify function.
Tachometer and recorders reconnected.
Design change completed to provide backup to AST-20 so that it would not be isolated during Front standard testing procedures; electrical overspeed channel was added by design change.
Tech Spec have. been amended to clarify operability determinations relative to turbine system surveillance requirements.
CORRECTIVE ACTION EFFECTIVENESS:
No recurrence of similar problems noted.
9
i.
9 U. S. NUCLEAR REGULATORY COMMISSION REGION I h
Report Nos. 50-272/94-11 50-311/94-11 50-354/94-09 License Nos. DPR-70 DPR-75 NPF-57 Public Service Electric'and Gas Company Licensee:
P.O. Box 236 Hancocks Bridge, New Jersey 08038 Facilities:
Salem Nuclear Generating Station Hope Creek Nuclear Generating Station l
3 March 27,1994 - April 30,1994 Dates:
C. S. Marschall, Senior Resident Inspector Inspectors:
S. T. Barr, Resident Inspector J. G. Schoppy, Resident Inspector T. H. Fish, Resident InWar P. P. Sena, Resident Inspector r
A T. J. Frye, Resident laWar B. J. McDermott, Reactor Engineer M. J. Buckley, ess SpMalist J.1.aughlin, GY h
I v
Approved:
WChief, Projects'W 2A Date J. R.
Insoection Summary:
d f
d ing day
'Ihis inspection report documents inspections to assure public hea and bWhift hours of station activities, including:
maintenance and surveillance testing, emergency preparedness, security, i
engineering / technical support, and safety ="eument/ quality verifica Summary delineates the inspection findings and conclusions.
e
t p* mc oq[C UNITED STATES g
NUCLEAR REGULATORY COMMISSION e
i I
REGION I 475 ALLENDALE ROAD
,E KING OF PRUS$1 N
VA 19406 1415
- Dut:ket Nos. 50-272 50-311 50-354 Mr. Steven E. Miltenberger Vice President and Chief Nuclear Officer Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, NJ 08038
[
Dear Mr. Miltenberger:
SUBJECT:
SALEM AND HOPE CREEK RESIDENT INSPECTION NOS. 50-272/94-11; 50-311/94-11; 50-354/94-09 The enclosed report documents an inspection for public health and safety, conducted by Mr.
C. Marschall, Senior Resident inWor and other members of the NRC resident and regional staff at the Salem and Hope Creek Nuclear Generating Stations for the period
['
between March 27,1994 and April 30,1994. 'Ihe inWors discussed the findings of this inspection with Messrs. J. Hagan and R. Hovey of your staff.
Raw on the results of this inWon, it appears that a number of your activities were not conducted in full compliance with NRC requirements, as set forth in the Notice of Violation.
These activities involved at least three occurrences of failure to follow procedures. In accordance with 10 CFR 2.790 of the NRC's ' Rules of Practice, a copy of this letter and its enclosures will be placed in the NRC Public Document Room. 'Ibe responses directed by ll this letter and the enclosed Notice are not subject to the clearance procedures of the Of5cc of Management and Budget as required by the Paperwork Reduction Act of 1980, Public Law No. 96.511.
In addition, the inWors identified a non-cited violation at Hope Creek involving failure to monitor reactor water chemistry in accordance with the frequency required by Technical Specifications.
Your cooperation with us is appreciated.
. Sincerely, 9
ll G
T 4
l
\\)
i Edward C. Wenzinger, Chief Projects Branch No. 2 b
Division of Reactor Projects
'ild 6:? E i./ f f
a 17 i
B.
Solid State Protection System Steam Mow Input Relays During the initial investigation of the short duration high steam now signal the licensee i
identified that twelve of sixteen steam flow input relays were discolored. Subsequent time response testing demonstrated that the relays were all functioning acceptably.i Test data for j
Train A shows that the actuation signal duration must be at least 16 msec for the master j
relay contact to close. Train B requires an actuation signal of 31 msec. De variance in the j
response time of these relays had minimal impact on the ability of the system to meet overall i
design requirements. The design specification for time delay of the high steam flow SIS j-actuation is less than 600 msec.
After the input relay and TS channel time response testing were completed the licensee 1'
F i replaced the discolored relays. De apparent cause of the discoloration was the accumulation i
of a powder-like carbon buildup due to cycling of the relays. The licensee elected to replace i
the relays based on their appearance, even though the test results indicated the time response i
of all the relays was acceptable.
J The inspector reviewed the licensee's evaluation, test results, and the discolored relays. De
)
ia==*ar made a visual naminatian of the relays and compared the extent of the carbon
[
j buildup to the timeiesponse test data. No correlation was observed bem the apparent j
carbon buildup and the time response performance. De licensee stated that discolored relays l
will also be replaced in Unit 2 during the next outage and that they have no reason to j
question their operability, ne in=~*ar concluded that the licensee's actions regarding the input relay evaluation were y=wable.
I.
i C.-
Steam Mow Summator i
r i
{
ne licensee examined the output of the steam flow mimmntars during Maintenanz and i
Controls troubleshooting to hmice if the mimmatar had caused the spurious signal. De i
test simulated the auxiliary contacts of the reactor trip breakers opening and monitoring the j
output of the summator module. De high steam flow function compares actual steam flow 1
to a programmed setpoint of 40% of full steam flow between 0 and 20% load and then i
increasing linearly to 110% of full steam flow at full load. When the reactor trip breakers
{
open the output of the setpoint should decrease to the 40% steam flow value (1.34 VDC).
Strip chart data shows the setpoint actually dipped below the 40% value by 100 mVDC before returning to the low value setpoint.
i He licensee is currently investigating the phenomenon however, there is no safety significance because the lower setpoint output would g,ngosd to a lower reference value l
being used to determine high steam flow. Based on the minor extent of the phenomenon and j
its safety significance, the inspector determined that no further NRC review of the issue is necessary at this time.
4 i
I i
7 i
i 16 4.2 Salem - Restart Inspection Activities ne inspectors reviewed licensee corrective actions for the following items to insure i
adequacy for Salem Unit I restart.
')
A.
Steam Row Transmitter Damping Cin: ult Modification On April 7,1994, a spurious high steam flow signal was sensed by the Rosemount flow transmitters and resulted in an "A" train safety iqjection signal. Subsequent to this event, these steam flow transmitten were modified per design change package (DCP) IEC-3328 to
,j add a damping adjnemant Previous to the April 7 event, transmitter thne responses ranged I
j from 0.020 to 0.110 seconds. After modification, the transmitter time rg== was adjusted i
to 0.225 i 0.025 seconds. His transmitter time response setting will prevent spurious i
steam line pressure spikes from tripping the steam flow bistable when the reactor is tripped and actual main steam flow is below 40%.
i 3
j De inWors reviewed modifiction package DCP IEC-3328 and the work orders used to j
modify the eight Reant flow transmitters. De !=g+:Mi. noted that the DCP contained i
a good evaluation of how the modification affected the design basis of the affected pisction circuits. The la&w also reviewed the sensor calibration and time response test results and concluded that the transmitters were adequately tested. De Technical Specifications and FSAR were reviewed to evaluate what limits and design basis may be affected by the Mikah De DCP contained a thorough design analysis of the modification which was SORC reviewed and approved. De ia*=*M agmed with the licensee conclusion that implementation of the modification did not constitute an unreviewed afety question, since i
l the transmitter response after the modification was less than the time assumed in the design basis. The ing+3-. concluded that the modification was we!! performed and adequately addressed the =*wiew Unit I restart issue.
O
d l
19 i
During valve reassembly, the licensee installed new valve internals for both PORVs using the i
revised procedure. No anomalics were noted. Further, as a one-time check to ensure the i
IPR 2 valve body did not contribute to the unusual wear pattern, the licensee cycled the valve i
10 times using the air operator. When plant staff subsequently disassembled the valve for ia= don they found no signs of wear. Plant staff again reassembled the valve using the revised procedure.
i In summary, licensee engineering determined that the galling and scuffing found in IPR 2, j
although heavy, was acceptable. They attributed the wear to miulignment of the valve internal during installation. To prevent future misalignments, the licensee changed the l
procedure for reassembly. The modified procedure includes manually stroking the valve during various stages of reassembly to insure that the plug moves freely within the cage.
l He team noted that the valves operated more than 300 times during the transient, and then j
passed the operability test. In addition, the vendor considered the observed wear acceptable.
l De licensee installed new internals in IPRI and IPR 2.
1 At the end of the inspection period the licensee had not determined the cause of the crack.
In addition, the licensee had not yet completed the safety evaluation of PORV operability.
He inee~s will review licensee root cause and corrective actions in a future iaea~* ion report. (IFI 50 311/94-11-01)
E.
Atmospheric Steam Relief Valve (MS-10) Modification During the April 7 transient, one or more steam generator code safety valves lifted on high steam prw. The MS-10 valves did not automatically operate as dacignad o prevent t
challenges to the steam generator code safeties. Although the Final Safety Analysis Report (FSAR) does not take credit for operation of the MS-10 valves in response to an accident, the design of the MS-10 valves was intended to provide steam pressure control, whereas the main steam code safety valves were designed to protect the steam generators and piping from failure due to overpressure. As a result of the lack of MS-10 automatic operation a steam generator code safety valve lifted. His ctused the reactor coolant system (RCS) temperature to reduce sufficiently to result in a RCS pressure drop. Inw RCS pressure initiated a safety injection.
The licensec found that the MS-10 valves did not open because of saturation of the control circuit, as a result of a modification performed in 1977. De licensee consulted with Westinghouse in 1977 to identify a solution for inadvertent MS-10 opening during load transients. Rey concluded that removing a " drain circuit," consisting of two diodes and a resistor, would prevent inadvertent MS-10 opening. The Liraawa and Westinghouse did not identify that removal of the " drain circuit" created the circuit saturation conditions which caused considerable delay in MS-10 response to rapid changes in steam pressure. In response to the April 7 transient, the licensee reinstalled the drain circuits to provide the correct MS-10 response to pressure changes. In addition, the licensee adjusted settings for control circuit response to insure adequate MS-10 response to rapid transients, while
i 18 D.
Power Operated Relief Valve (PORV) Operability During the plat response to the Salem Unit I transient on April 7, the PORVs operated a total of at least 300 times. The valves functioned as designed, and during the cooldown at the conclusion of the event, plant staff tested both valves for stroke time and seat leakage He valves were judged operable. Details of PORV operation during the transient are provided in NRC Inspection Report 50-272/94-80. Upon valve diunembly, mechanics did i
not experience difficulty indicative of severe binding while removing valve internals. As a result of the numerous valve operations, the licensee opened IPRI (cycled more than 100 times) and IPR 2 (cycled more than 200 times) for inmerion. ney discovered that IPRI had a small amount of wear randomly distributed around the plug and the interior of the j!
cage. The licensee concluded that IPRI exhibited typical wear for the circumstances. The i
licensee also found that IPR 2 had heavy scuffing on the outlet side of the plug and the cage,
[
and galling on the corresponding side of the valve stem. Additionally, the licensee found l
small cracks in the plugs for both valves radiating from a hole drilled in the plug.
I l
Because of the degree and orientation of galling wear noted on IPR 2, the licensee assembled i
a team of engineers to determine the root cause of the wear patterns and to assess the effects of the wear on operation. He internals were shipped to a Westinghouse materials j
laboratory for analysis of the cracks and assay of the materials on the scuffed surface of the l
plug. Correspondence from the valve manufacturer, Copes-Vulcan, indicated that the wear on IPRI was normal and IPR 2 was more than desired but not unusual. The team preliminwily determined that several factors contributed to the wear observed in IPR 2. He j
manufacturer designed the valve with small clearances between the plug and cage.
Installation of the valve internals was performed using a procedure which may have contributed to minlignment ofintemal components. Chrome coating from galling of the
[
stem may have migrated to a location between the plug and the cage, contributing to the scuffing of the plug. h hydrodynamic forces present in the valve during valve lifting under operating conditions caused a differential pressure across the valve plug pushing it
]
against the outlet side of the cage.
I l
De licensee team noted that the cage-guided design of these PORVs assumed a certain l
degree of contact wear in the plug and the cage. Dimensional tolerances of 3 to 5 mils th the plug and cage provide lateral support to the plug which otherwise would only be provided by the stem guide near the top of the valve, h team also noted that the PORVs l
were reassembled during 1R11 by installing the bonnet of the valve into the body with the i
stem, stem packing, and plug pre-assembled into the bonnet. His method was used primarily as a means to reduce jwieunel radiation exposure during valve maintenance, but i
did not allow for an adequate check for valve binding or misalignment during installation.
l Based on the investigation, the licensee revised the re-assembly procedure to include a i
vendor recommended piece by piece assembly process, including checks for clearance and j
freedom of movement at each stage ofinstallation.
i j
i
20 minimizmg the undesirable response to normal load changes, such as inadvertent opening.
Plant staff modified a spare control card and performed response tests of the card in the simulator control room. The tests demonstrated that reinstalling the drain circuit eliminated the saturation condition experienced on April 7, ne licensee planned to confirm the adequacy of the corrective action through startup testing.
The inspectors reviewed the circuit modification, supporting documentation, the safety evaluation, and results of simulator testing. He inWars concluded that the modification addressed the saturation condition. The inspectors will monitor startup testing to confirm the that the modified cimuit permits proper operation of the MS-10 valves.
-i Mr. Steven E. Miltenberger EXECUTIVE
SUMMARY
Areas Inspected: An AugmenteilInspection Team (AIT), consisting of personnel from Region I AEOD and NRR, inspected those areas necessary to ascertain the facts and determine probable
.I causes of the automatic reactor shutdown and multiple automatic initiations of the safety injection system that occurred on April 7,1994. The team assessed the safety significance of the event, including the resultant plant operation with a water (liquid) filled pressurizer and its cl.allenge to the primary coolant boundary integrity and the potential vulnerability of the ultimate heat sink i
to the same marsh grass intrusions that challenged the plant normal heat sink, which was the initiating event for the sequencetf events on April 7. The adequacy of the licensee's design, i
maintenance and troubleshooting practices relative to the safety injection system was reviewed.
[l l
The possibility for any potential generic implications posed by the Salein event was assessed.
1 Results: The Augmented Inspection Team (AIT) developed a sequence of events detailing the circumstances surrounding a Salem Unit 1 plant trip and a series of safety injection system actuations. It was found that the events led to the loss of the pressurizer steam bubble and the normal reactor coolant system pressure control system, and an Alert declaration. The AIT noted through an event sequence and causal factor analysis that the root causes of key events generally included a combination of component failure and human error. Additional procedural guidance
{
for, and prioritization of work activities of control room operators would have resulted in a better response to the event. The AIT found in general that the licensee response to the ' lmost a
daily event of grass clogging of the circulating water screens was very well planned and coordinated for the a6 Itional workload at the circulating water structure. However, as indicated by the performance of personnel and equipment in respoax to the April 7 event, the licensee did not adequately plan for, and coordinate, the activities corresponding to the additional workload in the control room resulting from the same event.
l, Finally, even though some equipment and licensed operators performed poorly during the ensuing transient on April 7, the core and its primary protective barriers were maintained throughout the event.
In addition, the following conclusions were developed as a result of the AIT review and discussed at a public exit meeting held on April 26,1994:
J Summary of
Conclusions:
1.
No abnormal releases of radiation to the environment occurred during the event (Section 3.4).
2.
The April 7,1994 event challenged the RCS pressure boundary resulting in multiple, successful operations of the pressurizer power operated relief valves and no operations of the pressurizer safety valves (Section 3.2).
i 3.
Operator errors occurred which complicated the event (Section 4).
l ii b
l.
ATTACHMENT 6 1*
I JUNE 24,1994 l
Docket Nos. 50-272 50-311 EA No.94-112 4
Mr. Steven E. Miltenberger Vice President and Chief Nuclear Officer l
Public Service Electric and Gas Company P. G. Box 236 Hancocks Bridge, New Jersey 08038
[
i
Dear Mr. Miltenberger:
SUBJECT:
NRC AUGMENTED INSPECTION TEAM (AIT) REPORT NOS.
56 272/94-80 AND 50-311/94-80 The enclosui report refers to a special onsite review by an NRC Augmented Inspection Team l
(AIT) from April 8 through April 26,1% The team reviewed the circumstances surrounding the automatic ~reacArJhutdown and two automatic actuations of the "_ safety injection" system that occurred at Salem Unit 1 on April 7,1994_._
The report discusses areas examined during the inspection. The inspection faculwas on the poteniaAl-significance of the events, and included detailed fact-finding, determination of root causes, and evaluation of operational and managerial pe_rformance. 'Ihe inspection consisted of selective exammation of procedures and representative records, observations, and interviews j
with personnel.
The AIT determined that the predominant canse of the event was the combination of pre-existing equipment problems or vulnerabilities and the resultant challenges to the operators, and operator errors thtt occurred during the transient. Other failures and their causes were reviewed and are discussed in the attached report. The AIT concluded that both the equipment problems and operator errors could, and should have been avoided by licensee management through a closer review of the operator needs in response to the frequent and expected transient conditions a
i resulting from the grass intrusions at the circulating water structure.
The AIT found the licensed operator response to the initiating event, a loss of circulating water, was weak. Operators did not take some actions that they were trained to perform. However, overall operator response was successful in achieving a stable plant condition; unfortunately, much later in the event sequence than expected, and too late to avoid a significant challenge to the pressurizer power operated relief and safety relief valves.
While we note the actions of PSE&G to improve plant hardware and procedures prior to the event, both hardware deficiencies and inadequate procedures played key roles throughout the event sequence. Also, the actions taken by PSE&G before and during the event to mitigate the frequent grass intrusions at the Salem circulating water structure were both well conceived and
,. l t
h
i Mr. Steven, E. Miltenberger 2
generally well performed. However, these initiatives were not accompanied by a similar review of task performance and procedural guidance in the control rooms to ensure that licensed operator response to the potential or actual loss of circulating water would also be successful.
It is for these reasons that the NRC views the relatively poor performance of the operating crew during the April 7,1994 event to indicate not just weak performance of certain licensed operators; but rather, and more importantly, an inadequate assessment by management of the prevalent operating " conditions at the plant and subsequent development of an appropriate i
operating philosophy to meet the expected needs.
l l
It is not the responsibility of an AIT to determine compliance with NRC rules and, regulations or to recommend enforcement actions. These aspects will be developed following additional NRC management review of this report.
8 j
A representative from the State of New Jersey, Department of Emironmental Protection and i
Energy (DEPE), observed parts of the onsite AIT inspection activities. A copy of a letter from Mr. Anthony J. McMahon, Acting Assistant Director, Radiation Protection Element, NJ DEPE to NRC is enclosed with this letter. That correspondence describes three issues not specifically addressed in the AIT report. Also enclosed is the NRC reply letter describing our plans to address those concerns.
In accordance with 10 CFR 2.790 of the Commission's regulations, a copy of this letter and the enclosures will be placed in the NRC Public Document Room.
We will gladly discuss any questions you have concerning this inspection.
Sincerely, ORIGINAL SIGNED BY:
James T. Wiggins, Acting Director Division of Reactor Safety
Enclosures:
- 1. Inspection Report Nos. 50-272/94-80
- 2. I.etter, dated May 20, 1994, from A. J. McMahon, NJ DEPE to J. T. Wiggins, NRC
- 3. Letter, dated June 24,1994, from J. T. Wiggins, NRC to A. J. McMahon, NJ DEPE
L.,
a Mr. Steven E. Mittenberger EXECUTIVE
SUMMARY
(CONT'D)
P 4.
Management allowed equipment problems to exist that made operations difficult for plant operators (Section 7.2).
5.
Some equipment was degraded by the event, but overall, the plant performed as designed (Section 3).
4 6.
Operator use of emergency procedures was good. However, procedural inadequacies j
j were noted with other operating procedures (Section 4).
5 7.
Licensee's investigations and troobleshooting efforts were good (Section 5).
i i
i I
4 i
1 1
i i
)
I i
)
l 1
i t
1 iii
. - -.,~~
ATTACHMthi /
i assog hg UNITED STATES 6
j'
[,]
g NUCLEAR REGULATORY COMMISSION a i 8
REGION i I
9,
/
475 ALLENDALE ROAD j
g j'
KING OF PRUSSIA. PENNSYLVANIA 19406 1415 i
g-j p*o.
l Docket Nos. 50-272
- ~
50-311 4
50-354 i
EA No.94-112 Mr. Steven E. Miltenberger
}
Vice President and Chief Nuclear Officer I
j Public Service Electric and Gas Company i
P.O. Box 236 l
Hancocks Bridge, NJ 08038
]
Dear Mr. Miltenberger:
i
SUBJECT:
SALEM AND HOPE CREEK RESIDENT INSPECTION NOS. 50-272/94-13; 1
50-311/94-13; 50 354/94-11 l
i The enclosed report documents an inspection for public health and safety, conducted by Mr.
l C. Marschall, Senior Resident Inspector and other members of the NRC resident and i
regional staff at the Salem and Hope Creek Nuclear Generating Stations for the periM l
between May 1,1994 and June 25,1994. The inspectors discussed the findings of this inspection with Messrs. J. Hagan and R. Hovey of your staff.
l During the inspection period we noted that you operated the Salem and Hope Creek plants safely overall. However, we also noted that the quality of operability determinations at Salem and Hope Creek could be improved through documentation of the basis for the determinations. Matters penaining to our followup inspection on issues relative to the April 7,1994 trip of Unit 1 are provided for your information. In accordance with 10 CFR l
2.790 of the NRC's " Rules of Practice," a copy of this letter and the enclosure will be l
placed in the NRC Public Document Room (PDR). No response to this letter is required.
Your cooperation with us is appreciated.
Si ly, 3
s o
s Edward C. Wenzinger, Chief Projects Branch No. 2 Division of Reactor Projects
Enclosure:
NRC Inspection Report Nos. 50-272/94-13; 50-311/94-13; 50-354/94-11 b
\\ -
g
i j
A j
U. S. NUCLEAR REGULATORY COMMISSION l
REGION I Report Nos. 50-272/94-13 50-311/94-13 50-354/94 11 4
i License Nos. DPR-70 DPR-75 NPF-57 Licensee:
Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, New Jersey 08038 e
Facilities:
Salem Nuclear Generating Station Hope Creek Nuclear Generating Station
{'
4 Dates:
May 1,1994 - June 25,1994 Inspectors:
C. S. Marschall, Senior Resident Inspector S. T. Barr, Resident Inspector J. G. Schoppy, Resident Inspector T. H. Fish, Resident Inspector J. Laughlin, Emergency Preparedness Specialist
]
R. K. Ierson, Resident Inspector S. M. Pindale, Resident Inspector l
B. C. Westreich, Resident Inspector N. S. Perry, Senior Resident Inspector T. Y. Liu, Project Engineer B. J. McDermott, R ctor gi r
P.F.Bo Res' ent
_r Approved:
/
M/
,_ 7 /
J. R.
'te, Chief, Pfojects Section 2S Eate' Jnsnection Sum
'Ihis inspection report documents inspections to assure public health and safety during day t
and backshift hours of station activities, including: operations, radiological controls, j
maintenance and surveillance testing, emergency preparedness, security, engineering / technical support, and safety assessment /quali'y verification. The Executive Summary delineates the inspection findings and conclusions.
1
-qqo=Msere-AC 5(T'
l i
14 J
i I
.i 4.0 ENGINEERING 4.1 Salem NRC Followup To Unit 2 Power Operated Relief Valves (PORVs) Materials A.
In response to the licensee's discovery that incorrect internals for the 2PRI a potentially been installed during the last Unit 2 refueling outage (2R7), the inspect investigated the history of the materials that had been used by PSE&G for the 2PRI and 2PR2 and reviewed the work process that had been used by the license the valves during 2R7.
The inspectors determined that the original material used for the Unit 2 PORVs whe unit went in service in 1981 was 17-4 PH stainless steel. As a result of industry expe with seat leakage and susceptibility to galling, and in response to TMI Action Item II.D PSE&G implemented a design change in 1982 w! installed PORY internals (plug made with 304 stainless steel with a stellited plug. Previously, the licensee removed l 1
1
i 15 seals from the PORV installations. Based on vendor recommendations, the licensee intende t
I to upgrade this material during 2R7 with 420 stainless steel to accommodate the different environment with loop seals removed.
a De licensee prepared two work orders (WOs) for the Unit 2 PORVs to be accomplished during 2R7: one for the repetitive task of inspecting the valve internals and replacing t with 304 stainless components if neenary, which is routinely done at every refuelir.g outage; and another WO for the installation of 420 stainless steel internals to be accomplished per the approved design change package. As part of the WO for the routine inapa don of the valves, the licensee pre-staged a new set of internals in case the pa j
needed to be replaced. Licensee records indicated that the pre-staged set ofinternals was i
made of 17-4 PH stainless steel; the licensee could not explain why 17-4 PH and not 304
{
stainless steel had been pre-staged other than the recognition that 17-4 PH parts had still l
in the PSE&G inventory. Warehouse material issue tickets indicated that it was this set o pre-staged parts that was mistakenly issued for the performance of the WO that was to i
accomplish the upgrade to 420 stainless steel.
i While the licensee recognizes that the 17-4 PH parts were not intended to be installed in the Unit 2 PORVs, their evaluation concluded that use of the parts was not unsafe, even other material was considered better for this application. The licensee's conclusion was i
based on: the acceptability of 17-4 PH material performance supported by the analysis o results as presented in the Salem FSAR for original licensing; the continued supply of the j
material by the vendor for other licensees' use; and the fact that, while 17-4 PH stainless steel is more susceptible to long term wear, the current internals will only be used for one fuel cycle (PSE&G plans to replace the internals at the next Unit 2 refueling outage). The I
licensee performed a 10 CFR 50.59 Safety Evaluation tojustify the continued use of the current installed valve internals for the remainder of the current fuel cycle. The Safety j
Evaluation was approved by the Station Operations Review Committee.
'a The inspectors reviewed the licensee records for the procurement and disposition of the 17-4 PH valve materials and for the performance of the two applicable WOs, and discussed the results of the licensee's investigation into the matter with licensee management. De i
inspectors determined that the 17-4 PH stainless steel valve internals manufactured by C j
Vulcan had been issued by the PSE&G warehouse for the material upgrade WO, as evidenced by component tagging and warehouse receipts. He inepa 'ars also identified that the component tags for the 17-4 PH parts had been included in the paperwork that had accompanied the WO for the material upgrade design change. He inspectors were unable to positively determine if it had been those components that had been installed into the PORV valve bodies during the performance of the WO, although the inepa-' ors did identify three separate ats of hand-written notes that indicated that the internals that had been removed were separately removed from the work area. All parts that were removed as part of that WO had been dhposed of by the licensee, and the inspectors concluded that it will not be until the next outage when the valve is opened for ersmination that the identity of the valve internals w' ill be positively determined.
aI 16 In addition, the inspectors determined that the licensee had procured a total of six sets stainless steel internals. The inspectors determined that the accounting of those six sets supported the belief that 420 stainless steel pans had not been installed in the Un
[lt
'Ihe inspectors noted that only two quality control hold points had been required for the installation of the valve internals: one to witness a blue contact check of t for a cleanliness inspection of the valve prior to valve closum.
By examination of available 17-4 PH and 420 stainless steel valve internals in the PS warehouse, the inspectors determined that thne was no readily identifiable means of distinguishing the two different materials ar.d that the licensee explanation of the events 2R7 was plausible. De inspectors determined through their review of the available reco
[
(including the 10 CFR 50.59 Safety Evaluation) and discussion with licensee personnel tha there was a reasonable level of assurance that the current internals of the Unit 2 P made of 17-4 PH stainless steel and that these internals do not compromise the ability of the valves to perform their design function. However, the inspectors also noted that the licensee failed to prevent the use of incorrect pans (i.e., parts other than those specified by the specifications of the modification).
B.
Atmospheric Steam Relief Valves (MS-10s) and Steam Row Transmitters
[
The effect of the atmospheric steam relief valves (MS-10s) on the Salem Unit 1 event of Licensee action j
April 7,1994 is documented in NRC Inspection Report 50-272&311/94-80.
to correct the " reset windup" (inability to respond to steam pressure increases) was
]
documented in NRC Inspection Report 50-272&311/94-11. As documented in that report,
~
the MS-10 valves were designed to prevent challenges to the main steam code safety valves.
In March 1977, the licensee modified the control circuits for the MS-10s to prevent undesired and inadvertent opening of the MS-10s. That modification contributed to the 1l failure of the MS-10s to function appropriately by remaining closed in a condition (on April l
7,1994) which required the valve to open automatically. Consequently, a main steam code safety opened, resulting in a Safety Injection.
The licensee corrected the circuit by restoring the control circuits to their 1977 configuration and making adjustments to insure that the valves operated to control steam pressure as originally intended. De inspectors noted that, in response to the June 10 Unit I trip, the
~
MS-10s worked as designed.
The licensee found that their analysis for the 1977 modification did not identify the potential that the MS-10s would not respond properly to increasing steam pressure. The licensee subsequently identified the lack of MS-10 response to pressure increases. An approved digital feedwater modification (scheduled for implementation in 1993 at Unit 1) was intended to resolve the deficiency. However, the licensee delayed the implementation of the modification. Consequently, the licensee allowed the MS-10s to function in a degraded n
L
1 17 failed to promptly identify and correct the m a t e censee until April 7,1994.
- s rom March 1977, During the April 7th transient, the first. safety injection (SI) sig temperature coincident with a spurious'high steam flow sig l f hwas equipment that injected water into the vesse na o s ort duration. The e one train of that event sequence records for reactor and turbine trips on Ju
. The inspectors noted and February As noted above, the licensee failed to promptly i a
een present.
least, those three previous occasions.
e spurious signal on, at i
k j
I s
r
T.
~ ~ ~ -
h 2%
UNITED STATES
, 4 ['
p, NUCLEAR REGULATORY COMMISSION U
![
REGION I D
.3 g
475 ALLENDALE ROAD KING oF "9USSIA, PENNSYLVANIA 19406-1415
"'44.....
d JUL 2 G Y394 Docket Nos. 50-272 50-311
[
MEMORANDUM FOR:
Distribution I
FROM:
Eugene M. Kelly, Chief, Systems Section, Division of Reactor Safety
SUBJECT:
SALEM AIT ESCALATED ENFORCEMENT -
r BRIEFING PACKAGE i
ha A tT ncilwu -
Attached is reissued background information for the open conference scheduled at 10:00 a.m.
on July 28,1994, adding a missing page and reordered attachments.
j We plan to discuss NRC positions and strategy in a prebrief on Wednesday, July 27,1994, j
at 3:00 p.m. in the Executive Conference Room.
[
Questions should be directed to me (610-337-5111).
Eug M. K y,Ch*f t
Systems Section Division of Reactor ety j
l Attachments-l
- 1. Meeting Notice
- 2. July 6,1994, letter (Martin to Miltenberger)
- 3. Festation/ Mitigation Analysis "l
- 4. Salem Enforcement History l
- 5. Analysis of Previous AITs (June 3,1993; December 13,1992; and November 9,1991)
- 6. Salem AIT Report 94-80 (Cover and Executive Summary)
- 7. Excerpts from Reports 94-11 (Section 4.2) and 94-13 (Section 4.1)
J'
/Q-7-Gk3-@t%
Ip-i
U. S. NUCLEAR REGULATORY COMMISSION S
REGION I No. 94-86 NOTICE OF SIGNIFICANT LICENSEE hEETING EA 94-112 Licensee:
Public Service Electric & Gas Company Facility:
Salem Generating Station
[
Docket No:
50-272 Date and Time:
Thursday, July 28,1994 at 10:00 a.m.
Location:
NRC Region I,475 Allendale Road, King of Prussia, Pa l
Main Conference Room r
1
Purpose:
The purpose of this Enforcement Conference is to discuss the apparent violations identified from NRC's review of the Apdl 7,1994, automatic shutdown of Salem Unit 1, including cause, safety significance, and corrective actions. The meeting was requested by the NRC as a result of findings identified by the Augmented Inspection Team and other NRC inspection activities related to the event.
I t
NRC Attendees:
T.T. Martin, Regional Administrator, Region I J. Wiggins, Division Director, Division of Reactor Safety (DRS)
J. Grey, Deputy Director, OE E. Wenzinger, Chief, Projects Branch 2, DRP J. White, Chief, Reactor Projects Section 2A, DRP C. Marschall, Senior Resident Inspector, Salem / Hope Creek l
E. Kelly, Chief, Systems Section, DRS j
J. Stone, Project Manager, NRR Other Members of Senior NRC Staff Licensee Attendees: S. E. Miltenberger, Vice President and Chief Nuclear Officer J. J. Hagan, Vice President-Operations / General Manager-Salem Operations F. Thomson, Manager, Licensing and Regulation Other Members of PSE&G Senior Staff
_l Note: ' Itis meeting is open for public observation. Handicapped persons requiring assistance to attend the meeting shall make their requests known to John White, Chief, Reactor Projects Section 2A, U. S. Nuclear Regulatory Commission, Region I,475 Allendale Road, King of Prussia, Pa., 19406, (610) 337-5114. Attendance by NRC personnel at this meeting should be made known by 4:00 p.m., July 27,1994, via telephone call to John White, Region I.
Prepared by: '
[
Ed d C. Wenzinger, Chief Jt ctor Projects Branch No. 2 7qo % cree M-m.
__ _. _ -. _ _ _ ~
. ~ _..... _ - - _
y~
k UNfTED STATES 4
y NUCLEAR REGULATORY COMMISSION g
g l.
Im'
'j REGION I 475 ALLENDALE ROAD 1,
%,..... [4 KING OF PRUSSIA, PENNSYLVANIA 19406 1415 i
JUL 6 1994 f
[ve,.2.. a vM] @M @ {
Docket Nos. 50-272 50-311
((c Wmt 4
- Mf I aditv Mr. Steven E. Miltenberger Vice President and Chief Nuclear Officer Public Service Electric and Gas Company r
i P.O. Box 236 I
Hancocks Bridge, New Jersey 08038
Dear Mr. Miltenberger:
SUBJECT:
APPARENT VIOLATIONS RELATIVE TO THE APRR.7,1994 EVENT The NRC inspection findings relative to the circumstances surrounding the April 7,1994, j
partial loss of plant circulating water flow and subsequent plant trip and safety injections,
[
wem detailed in the Augmented Inspection Team (AIT) Report 94-80, provided to you by our letter, dated June 27,1994. Further discussion of seveal of those findings is addressed l
in NRC Inspection Repotts 94-11 and 94-13. Dese reports identify findings that we consider important to the safe operation of the Salem units, and either contributed to or j
complicated recovery from this event, or wem indicative of weak management controls of important plant activities.
I 1l j
The April 7th event is of concern to the NRC since it resulted in multiple challenges to the Salem operators and plant safety systems. De control room command function was i
transferred or relinquished for pivotal parts of the event, and appropriate actions to counter the reactor cooldown in progress and pmvent the ensuing safety injection transients did not occur. Longstanding equipment deficiencies that also led to, and complicated the event, l
specifically, the atmospheric relief valve controls and steam line flow sensors, were tolerated and cornetive actions were not aggressively pursued. Communications with the NRC during the early stages of the event were ineffective in characterizing the significance of the transient, including the scope of equipment failures, the causes of safety system initiations, the asultant condition of the plant, and the planned mcovery efforts. Further, procedures for dealing with the abnormal plant conditions (namely, grass intrusion, rapid power reduction, and recovery from a water-filled pressunzer or " solid plant" condition) were inadequate, insufficiently detailed, or nonexistent. Finally, design and material controls were found to be inadequate, in that the pressurizer power-operated relief valve (PORV) internal component materials, although satisfactory for use, were not as specified by the design; and a solid state protection system (SSPS) logic card was subsequently determined to be of the wrong type.
P+
~
V
/
JUL 6 1996 Steven E. Miltenberger 2
i This event and the concerns outlined above are more troubling given the relative high frequency of hardware initiated events at Salun over the past four years along with recurring problems with procedural adherence, depth of root cause assessment of problems, and degree 2
of long-term effectiveness of corrective actions.
Apparent violations, as described in the enclosure, are being considered for escalated enforcement action in accordance with " General Statement of Policy and Procedure for NRC i
Enforcement Actions" (Enforcement Policy),10 CFR Part 2, Appendix C. Accordingly, no Notice of Violation is being issued for these inspection findings at this time. Further, the i
number and characterization of the apparent violations may change as a result of further NRC review.
5 l
The apparent violations and the need for an enforcement conference were discussed with you j
in a telephone conversation on July 6,1994, and the conference has been scheduled in the 1
}
NRC Region I office for July 28,1994. This conference will be open to public observation in accordance with the Commission's trial program as discussed in the enclosed Federal Register notice. Although not required, we encourage you to provide your comments on j
how you believe holding this conference open to public observation will affect your l
presentation and your communications with the NRC. The decision to hold an enforcement conference does not mean that violations have' occurred, or that enforcement action will be i
taken. De purposes of this conference are: (1) to discuss the apparent violations, including cause and safety significance; (2) to provide you with an opportunity to point out errors in i
our inspection report, and identify corrective actions, taken or planned; and (3) to discuss any other information that will help us determine the appropriate action in accordance with the Enforcement Policy. De conference is also an opportunity for you to provide any information concerning your perspectives on the severity of the apparent violations, and the application of the factors that the NRC considers when it determines the amount of a civil
'l 4
j penalty that may be===d in scwidsnce with Section VI.B.2 of the Enforcement Policy.
e i
e i
1 i
i i
~
JUL 6 1993 Steven E. Miltenberger 3
l In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy of this letter and its enclosure will be placed in the NRC Public Document Room. No response to this letter is required at this time, and your cooperation is appreciated.
Sincerely, W$-
3ames T. Wiggins, D[irector Division of Reactor Safety i
Enclosures:
1.
Apparent Violations Considered for Escalated Enforcement Action (Salem AIT Report 94-80) 2.
Federal Register Notice (Vol. 57, No.133, July 10,1992); "Two-Year Trial Program for Conducting open Enforcement Conferences - Policy Statement" cc w/ encl:
[
J. J. Hagan, Vice President-Operations / General Manager-Salem Operations S. I.aBruna, Vice President - Engineering and Plant Betterment C. Schaefer, External Operations - Nuclear, Delmarva Power & Light Co.
R. Hovey, General Manager - Hope Creek Operations F. Thomson, Manager, Licensing and Regulation R. Swanson, General Manager - QA and Nuclear Safety Review J. Robb, Director, Joint Owner Affairs A. Tapert, Program Administrator i
F Fryling, Jr., Esquire M. Wetterbahn, Esquire P. J. Curham, Manager, Joint Generation Department, Atlantic Electric Company Consumer Advocate, Office of Consumer Advocate William Conklin, Public Safety Consultant, IAwer Allow 2ys Creek Township K. Abraham, PAO (2)
Public Document Room (PDR)
I4 cal Public Document Room (LPDR)
Nuclear Safety Information Center (NSIC)
NRC Resident Inspector State of New Jersey State of Delaware
~
. tut o is i
Steven E. Miltenberger 4
bec w/ encl:
Region I Docket Room (with concurrences)
Kay Gallagher, DRP D. Holody, RI J. Lieberman, OE DRS File (2) bec w/ encl: (Via E-Mail)
J. Stone, NRR i
W. Dean, OEDO C. Miller, PDI-2, NRR M. Shannon, ILPB M. Callahan, OCA t
W
.?*
I i
ENCLOSURE 1 A
APPARENT VIOLATIONS p
.n CONSIDERED FOR ESCALATED ENFORCEMENT ACTION gy p'
(AIT REPORT 50-272,50-311/94 80; AND INSPECTION REPORTS i
j e.y U
50-272,50 311/94-11 AND 50-272,50 311/94-13) x[e A.
Technical Specification 6.1.2 requires that the Senior Nuclear Shift Supervisor (SNSS)
[
or, during his absence from the control room, a designated individual, shall be responsible for the control room command function. Technical Specifications require administrative procedums, as referenced in Regulatory Guide 1.33. PSE&G I
1i 8
Administrative Procedure NC.NA-AP.ZZ-002(Q), Attachment 32, Shift Management
[
^7 [
Responsibility for Station Operation, requires, in part, that the SNSS shall remain free l
- f to survey and analyze all operating parameters. He procedure further states, " intense py involvement in any particular detail may run the risk of losing control and perspective j4 d
of the overall operation." The following two examples on April 7,1994,
- d demonstrated a loss of perspective regarding overall plant safeg and the command oM" 34 A$U8yNYQ lcdf function of the senior nuclear shift supervisor:
&&Af R p laMMM d
ne SNSS left the control room during the loss of circulating water l
1) 4 override a circulator pump protective interlock, and thereby relinquished the p gM l
t f.
command function in the midst of a significant plant transient. During his gr>-
absence, operators caused reactor coolant temperature to decrease below the j yd,[*g>,
minimum temperature for criticality; and w ss uq+ md t.c l [ cum A g/
i j fpf)\\.y
,p While the SNSS was absent from the control room, the nuclear shh 2)
(NSS), designated as responsible for the control room command function,
[
.qe
'gf assumed the duties of a reactor operator by performing control rod N%,
I movements. As a result, for the period of time the NSS was manipulating the WC M controls, no individual was properly exercising the control room command hg, b
function.
(frwW MD, 44 N*N B.
10 CFR 50, Appendix B, Criterion XVI, Corrective Action, requires in part, that licensees identify significant conditions adverse to quality, determine their causes, and J
take corrective action to preclude recurrence. Two examples of failure to meet this requirement occurred:
p~ p
- 1) During previous mactor/ turbine trips on June 10,1989, July 11,1993, and February 10,1994, the licensee failed to identify and correct the cause of spurious high steam flow signals. As a result, the licensee failed to preclude an unnecessary safety injection actuation on April 7,1994, in response to a reactor trip; and
r i'
2
- 2) In March 1977, the licensee modified the control system for the main steam t
atmospheric relief valves (MS-10s) and has since failed to correct deficiencies intmduced during the modification. As a result, the licensee failed to preclude a
('.
second un-ry safety injection actuation on April 7,1994, (precipitated by opening of the main steam safety valves in lieu of the MS-10s) in response to a 1
reactor coolant system heatup following the initial safety injection actuation.
i y
C.
10 CFR 50.57 sequires, in part, that emergency plans and procedures for event dmification and notification of offsite authorities be implemented. Salem Emergency Plan and Event Classification Guide, Attachment 8, NRC Data Sheet, requires that specified information regardmg the event description be completed, approved, and -
[~
provided to the designated communicator for transmission (to the NRC) within 60 minutes. The specified information includes, "... systems affected, actuations and their initiating signals, causes, effect of event on plant, actions taken or planned, etc.
Note anything unusual or not understood..."
On April 7,1994, the following information was not communicated or adequately described: (1) the apparent logic mismatch of the protection system and resultant
. unexpected operation of the emergency core cooling system (ECCS) flow path valves
(
,and the unexpected condition of the main steam and feedwater isolation systems; (2) the cause of the reactor trip was not described; (3) the effect of the event on the plant (namely, the resultant filled pressurizer or " solid" RCS condition); and (4) the operator plans to recover from the solid RCS condition.
D.
Technical Specification 6.8.1, requires, in part, that the licensees implement written procedures referenced in Appendix A of Regulatory Guide 1.33, Revision 2,
['
February 1978. Appendix A of Regulatory Guide 1.33, Section 6, requires procedures for combating emergencies and other significant events, including expected transients and acts of nature.
During the April 7,1994, event, the initial response to the grass intrusion at the intake structure and recovery from the subsequent transient were complicated for operators. Procedural guidance was inadequate or nonexistent for the following activities:
a critio% of RCS temperature from below the minimum temperatur 1)
Reco klity; f
)
- 2) Rapid power reductions due to grass intrusion;
' s >l jj j 3) Recognition of and response to safety injection train logic disagreement; and d
h.'
p w
- 4) Recovery from " solid" plant conditions.
N
&g b
wh j
o al$2~f W3'g.().
6 y
[
'> s
i 3
10 CFR 50, Appendix B, Criterion VIII, Identification and Control of Materials, Parts, and Components, requires in part, that measures be established for the E.
identification and control of parts and components. These measures shall assure Y
identification of the item is maintained throughout installation and prevent the use 3
incorrect parts. The following are two examples that demonstrate a failure to s.O maintain configuration control:
During the 1993 Unit 2 outage, power operated relief valve (PORV) internals 1) made of 17-4PH stainless steel (original design material) were installed in valves 2PRI and 2PR2, in lieu of internals made of type 420 stainless steel (the vendor-recommended and licensee-approved design change replacement matedal).
i The post-trip investigation for the April 7,1994, event identified that the installe 2) summator module for the high steam flow setpoint did not have the proper identification and contained an incorrect electronic part. Although the licensee Jetermined this component did not affect the plant response during the event, it is an example of failure to properly identify components prior to installation.
Technical Specification (TS) 3.5.2 requires two operable emergency core cooling j
F.
system (ECCS) injection systems, or a plant cooldown to below 350*F within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. During the event on April 7,1994, and following reset of the safety injection actuation signals, automatic actuation capability was not available (and not reinstated) because the reactor trip breakers were not cycled. This occurred because there was no procedural guidance for re-establishing the safety injection logic, nor was it clear that the logic was operable, and because a cooldown could not be completed within hours due to the time required to re-establish a pressurizer steam bubble.
This was recognized by the licensee, in that, TS action 3.0.3 was entered, but enforcement discretion was later granted by the NRC in recognition that the plant would not be below 350*F within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> (or by approximately 5:00 p.m.) of the safety injection actuations that occurred on April 7,1994.
) (, U. 6 hS N
$6 hqb.svo elh ayp6eb hdjterst
LNCLUbukt 4 I
. s. dant assheer / vol. g..nw 133 / Pdder, fair stk tear / Notices
.seyes.
i amenassem8end comments to:ne Secretaryof theComedesten.U.S.
Nedeer Regulatory c===a--'a=
l Washirston.DC 30854. ATTN:
1 Docketlag and Serotos Breach.
Hand deliver commeste to:One White il Flint North.11555 Rockville pike.
Rockvilla. MD between 7:48 am. to 4:1s pm.Federalworkdays.
l Copies of commente may be examined i
at the NRC public Document Room.2120 L Street.NW.(14wer level).
i Washingesa.DC i
een namesosomanoscowract:
James Lieberman. Director. Oflice of Enfesosoment. U.S. Neclear Regulatory j
Commiselos.Wealdagten.DC 20555 i
(m ao+.am).
eumamouramesonnumose i
Badgound l
De NRCs count policy on enforcenset conferences le addressed in j
Section V of the latest rettelon to the
- Genere! Statement of poligy and procedere for Enforcement Actions."
(Enforcement policy)to CHL 2.
[
3 a
Cthatwas on j
F 14.1982(57 8791).The Ederesseemt etesse that.
enforcement
~ eresses will not normally be open to the public."
However, tbs
===aaala= bas decided c
toimplementa trialprogram to detensine t '
rto malatain the currentpol'.. m.hsogerdto enforcemeek rences or to adopt a I
Two year 7tted program for new policy that would allow most yw enforcement osoferences to be open to poesy seenement, ettsodemosbyallmembers of the public.
WN AamsovsNedeerRegulatory cm Pneition asmosapolicy statement.
He NRCleimplementing a twofear trialprooms to allow public suussadmneNedearRegulatory observatism of seleased enforcement Comademian (NRC) le leasing this poucy confereseen.no NRC willmonitor the statement en the taapleasemantica of a tweteer trial progress to eBow selected program and determine whether to establish a permansat policy for enforsommet conferences to tu open to
% epen enforossent attendance by au mesabers of the conferesses based on an asmsment of seeeralpubhc.nis policy stetmeest the fouowing criteria:
describes the two year trialprogram (1)Whether the fact that the andlaforms the public of how to get conferemos wee opealapected the information on upcoming open NRC's abuity to conduct a meaningful enforcement conferences.
conference and/or haplement the NRC's catsa:%1e trial progrem ie effective on P"W8" July 10.1982.while comments on the (2)Whetherthe open conference prograsa are twing received. Submit
'*"**d the hama's participation m commente ce or before the completion the confweses of the trial program scheduled for July (3) Whether the NRC' expended a 2L 1992. f%=mants received after this date will be considered if it is practical signifleant amount of resources in to do so, but the Comnussion is able to "baf the conference public; and assure considerstica only for comments (4) ne eximot of public interesi m received on or before this date.
opening the enforcement conference
I Federed Register / Vol s7. No.13s / PHder. July to saat / Noelses at l
L Cdesde Far.a marOpen thrw asespertes elIW wiu be en$ce to pememmel emmenlag,that
{
m
~i. -^r--* Camiseense
-t
- reactem, signe, beenere, poseem, etc. nouarger "A
o&er M==aa== which thee 3r be p== dan =d and ha Enforcement conferuname wW w be widconstet of the s'=aaW type of disrepow pemene may be removed.
l open to the puhusifes endernement bonneeen.
Each res6amalafhoe wG connaue to ua sega.
=
condecs the emfaseement onsference (1) W be takaaassinet sa II. Amenemedag open angemement Andividual or if the acnen, though met Comieresses proceediese la acomedense atth regional procdos he sederoomset conference i
taken against an individual, terns on As ecom asitis deteradood that an wG aand==a a be a mesdag h wbotheranindividoalhas eaawnm d enforcement conference wu! be open to se NRCand se uneneen.Whus e pubuc observation, the NRC erm orally saforcanentosadoremos te ope for i
(2) volym ='=a*=at personnal non y &e h &at es enfacement public observe $on.it is not open for faDures where th's NRC has requested confame wm be oport to pubBe publac parthspetnam.
thst thsindividual(s) involved be 80" 88 P881 d Os W8 persons attendlag open enforcement i
pruent at the caferencs; pmgresa and send he h a copy M conferences ase readaded that (2) the (3)la beeed onthe Andings of en NRC 6 18 R88teternodes setco@ne8 apparent violeelone disamened at open OSom ofInvad aham(OT) report;or 6e progrenha== we be asked to enforcement couderesses are sebket to f
(4) Involves safayeenis infan==uan.
numam 6e monbadperdcipene furthertweew and may be ambtect to Priv:cy Actinformanen or other
- d
- O' 888I*"8" change prior to any resulting infortnation erhich conld be considered so that the NRC can schedule en enferoamentsedes and 4)b proprietary.
'i i
l Enforcement conferences involving
- [NE alm 8 '"aate of views or expressaans of opinion made by MRC
% et medicalmisadadalstrations or l
overexposeres wW be open assuming gagg g dg open enforcement conferences or the g,g
- edorcensent besban the confersone can be conducted scheduled and that it is P' to pubhc tvpresent anni deteradast6cas or behis.
withoet dieciosing the avaamd observenon.
in addHim ego,_
s e,ds to annoanos open Ge.g cr.
poo.an o.a ts cc manaan ea-os e.In addition.
individual's nam The NRCinten
.e.o.e we.ot be n.
e,s e o &e,um,c wh es guidance in &ie nodos.persas CPen to the public if the conference wm normally atleast to woridesdeye la 4
8 oPen h conferences
[
be ocadacted by telepboos or the advance of the enforcement conference wG be prntded an 4, aty to conference wG be conducted et a through 6e fo8 owns nochemiens entonit written commente sansynomaly relatively smalllicensee's facility.
0)Nodoes postedin &e pebbs to &e ngloaal oSon.Deee comnients Finauy.with theapprovalof the Docmanat Room:
Execadvs Director for Operations, p) yog4ree *=W meenages and wG # n % be forwarded to the Director of the OSee af Enforcement for enforcessant conferences wm not be 9)Tou-free electronic beusen board
'"$ew and consideredon.
cpen to the publicin specialcases a
where good anaeshas been shown aAer establishment of the to64ree Deud as Andrielle. MD. ah6e rik der et J.ly I
h='-* the benent of pubhc mesesse systems,the pubec may can toer.
cbeervason against the potential impact 001)4sMrst to obtete e recording et For de Noaiser = f are===r==--
)
on the asancy's enforcament action la a opcondag spee esforcement sammel 5.Gek
[
particular essa-conferemose. De NRC wtD isees eno&ar secreevyofyw e--
ne NRC wulstrive to conduct open Federal Register mo6ce afterthe tolMree P Dec. m-14 ass Fund FM thj enforcement conferences dartng the" messege systems are som
.3,,,,,,-
two year trielprogreesin a.ordance To oeshot the NRC to meidag-j with the fouewing three goals appropriate arrangements to sagsport i
(2) Appresamasely as percent of aB pubbe observenom afendeseement U 754 l
cligible enforcementconferenese condesemase. dadhtdmale INessesed in conducted by the NRCwW be open for ettending a perdosiar enforomeeset a
puhuc a=====
o. fere.no ebo. w.o e s e e e, desi CorrectlOnS
- '.mi n+
(2) At Laast one open enforcement ta en=d ta the mostlag antico yet y, n a 1 e confervoos will be condacted ia ensk of
===a=went the open enforcumsat ths regional omcas;and
. conferemos ne aster thea Svu bastones Pr+ der. Jdy 17.1882 (3) Open enforcessant conferences days pr6er to the enforcement wiU be conductedwith a vertety of the conferemos.
d
}
types cf bcensees.
III.Condus of Open Enforcemet NUCL.EA.R RE,GULATORY To evoid potentialbiasin the mn.
ne Conferomoes i
selection process and to attempt to meet the thrw goals stated above, every in eooordemos with corrent preedes.
Two. Year THet proyam Mr fourth shgib&e aaL====at caniacence enforcement coderences wW continue Corickseterte Open ardorcement to normeu be held at b NRC regional Corderenose; potoy Statement involving one of thrw categones of r
j beansees wdi normeDy be open to the emces. Meanbers of the public wm be pubhc daring the tr al program.
allowed access to the NRC regional i
However. in cases where there le en emose to attend ope enfor-man
- In notice Wt 921a:33 beginrung i
ongoing adfudicatory p.vs.
by with conferences in accordsoos wrth the on pese 30752 in the ime of Priday.
onis of more tetervencrs, edorcement
" Standard Opereting hu For
)dy. ia 1982. on page 3c?t2. in the j
conferences invoking issues re. lated to Eveme Securtty Scyport For NRC
, mnd mhann under nam. bet"""W i
the subhet matter cf the ongo.ng Heannge And Meenngs" pubbhed to the finh haa. luly it w" shodd i
adludicetion may also be opened. For November.L tw1(se G sc1).Thes.
med % 11. N".
j the purposes of thja trial program. the procedurve provide that visitors may be
== - coes -
i J
l
p@
al i
g~
Y @N
- ESCALATION /MITIGA ON[ ANALYSIS **
IY G
\\
i*.
j Evaluation per 10 CFR 2, Appendix C, VI.B.2 Civil Penalty Adjustment Factors 4
1 The adjustment factors below are for the two most significant apparent viola,tions described in
}
the enclosure to Mr. Manin's letter to Mr. Miltenberger, dated July 6,199'.j i)
(
1 A.
Control Room Command Function c_. l u dWP
~
" [NL 1.
Identification:
up evMLG l
The inappropriate actions of the SNSS and NSS were identified by the during._ mdlm
)
interviews with individuals present in the control room area during the April 7, Su'u I
1994 event.
ilth, h es
/
l(/
Nlu pk S N 2.
Corrective Action:
\\
Prior to the restart of Unit 1 the licensee provided operator training on the p C
scenario for all shifts. Actions were taken to council the individuals involved in qt the event on expectations for crew performance.
& ;E
\\
i Y
[fi
'2 wspd % ( l l%
Licensee Performance:
- u
,y 4C
.7 6
Plant Operations received a Category 2 rating during the last SALP assessment l'
period. No violations involving plant operators and the control room command R[)y
/
function were identified in the last two years.
'Q( g t'
/
4.
Multiple Occurrences:
j' l
The SNSS and NSS both relinquished the command and control responsibility during the event. This indicates that the operations staff did not fully understand the intent of management expectations described in the administrative procedure.
5.
Prior Opportunity to Identify:
}'
.(d None was identined.
/
ti N
6J %
t ii 6.
Duration:
(f
~
'h4 The SNSS left the control room during the loss of circulating water and returned h
ko. p just prior to the reactor trip. During transfer of house loads, the NSS stopped directing the procedure, performed the duties of an NCO (by moving rods), and
- , t p then returned to directing the transfer of house loads.
i p
iC((^f f
h t,h k
1Y b
.. gW
- 'C
\\#
I #
l s
i B.
Spurious High Steam Flow Signal i
a-1.
Identification:
N The licensee failed to correctly identify the root cause of spurious high steam flow signals identified during three previous reactor / turbine trips. The high steam flow signal during the April 7,1994 event was self evident in that it provided the i
coincident logic (with low low Tave) for a safety injection signal.
I 2.
Corrective Action:
Prior to the restart of Unit 1, PSE&G installed a modification to dampen the signal from the steam flow transmitter. De licensee's preliminary evaluation of i
the modification's performance following the April 10,1994 trip indicates that the modification did not completely correct the problem.
3.
Licensee Performance:
Engineering and Technical Support were rated as Category 2 during the SALP assessment period. Three previous significant occurrences involving nonsafety-l related equipment failures have occurred since 1991. Specifically, the three previous AITs wem in response to the following:
1991 Turbine Overspeed Event 1992 loss of Overhead Annunciators 1993 Unexpected Rod Withdrawal 4.
Multiple Occurrences:
[
The rod control system and MS-10 control system are two examples where PSE&G failed to correct known deficiencies in nonsafety-related equipment, ne inadequate corrective actions for these deficiencies complicated the April 7,1994 event, contributed to the challenges of safety-related equipment, and caused a reduction of the margin to safety.
'l 5.
Prior Opportunity to Identify:
-l The licensee had at least three previous opportunities to identify and correct the root cause of the spurious steam flow signal, i
6.
Duration:
This issue concerns the original system installed by the vendor. Post trip data from a June 1989 trip shows the spurious signal was present at that time.
ATTACHMtni
- 8 SALEM VIOLATIONS APRIL 1,1992 TO JULY 12, 1994 REPORT ISSUED SEVERITY AREA DESCRIPTION 92-07 7/2/92 IV SA/QV Control room habitability, storage of ammonium hydroxide not communicated to NRC for evaluation.
92-11 7/30/92 IV M/S Failure to follow procedures regarding control of measurement and test equipment.
92-16 12/16/92 IV FRS Inadequate corrective actions by engineering for identified deficiency
[
regarding fire damper.
93-08 5/5/93 IV SA/QV Weaknesses identified in the licensee's implementation of the 10 CFR 50.59 program.
93-11 5/7/93 IV SEC FFD supervisory training issues.
93-15 7/12/93 IV -
OPS Availability of steam driven AFW during entry into Mode 3.
93-21 11/3/93 IV OPS Failure to initiate a timely shutdown of Unit 1 following failed surveillance test.
(Licensee identified)
[
93-23 1/10/94 IV M/S Diesel generator air start system operability.
93-23 3/9/94 III M/S Failure to follow maintenance (Violation resulted from Enforcement Action - Civil Penalty) procedures.
93-82 11/30/93 IV OPS / FITS Failure to follow station procedure for measuring battery cell voltages.
93-27 2/10/94 IV M/S/ OPS Inadequate control of troubleshooting and i
corrective actions iegarding RHR check valve leakage.
94-06 4/26/94 IV OPS Failure to comply with TSAS regarding PORVs.
94-07 6/29/94 IV FRS No written safety evaluation for replacement 460V vital bus transformer.
94-16 6/30/94 IV P/S Access Control of Vehicles.
L
ATTACPMENT 5
SUMMARY
OF THREE PREVIOUS AIT's 50 311/93 81 A. JUNE 5,1993 REVIEW OF CIRCUMSTANCES SURROUNDING SEVERAL ROD CONTROL SYSTEM FAILURES AT SALEM UNIT 2 DURING START UP ACTIVITIES FROM MAY 25 THROUGH JUNE 3,1993 i
BACKGROUND:
During startup, Salem Unit 2 experienced multiple failures of the control rod control system (CRCS). Five plant stanup attempts were made from May 25 to June 3,1993. The failures involved failure CRCS to move the rods correctly and maintain them in the proper position. The most serious event involved erroneous indications on rod position indicators as compared to step
}
counters, and withdrawal of a rod from the core without a demand signal and without response to operator actions to control the movement until power was removed form the rod. Due to the numerous abnormal occurrences in such a short time, NRC was concerned about the approach the licensee was using to determine cause and establish effective corrective actions.
ROOT CAUSE:
d Process causes:
Restan process (station policy and procedures) did not provide for programmatic determination of root cause of system failure,.i.e., there was no clear policy on when, how, and to what extent, to perform root cause analysis for component failures.
Hardware causes:
Multiple integrated circuit and output transistor failures due to unsuppressed voltage spikes.
1 Regulation board shon circuits created during manufacturing or maintenance activities.
Slave cycler logic card failures due to power supply short circuit or unsuppressed voltage spikes.
Q9 transistor failures due to jumpering activities performed for troubleshooting.
Wrong resistor installed on failure detector card due to manufacturing error.
4 CORRECTIVE ACTIONS:
(Applies to all Salem units)
CPAT effort is expected to address the lack on organized approach to root cause analysis and troubleshooting activities.
)
Standing Orders have been developed to provide additional guidance to operators relative to operability determinations (CPAT actions are also expected to enhance operability guidance).
4
.. =
= _.. -. _ -.
i Troubleshooting procedure has been significantly revised to provide more comprehensive guidance relative to root cause and troubleshooting processes.
Vendor supplied materials are required to go through Procurement and Material Control mspecuon prior to use in the plant.
Component and circuitry modifications and repairs completed and tested sadsfactorily; design changes instituted for digital demand step counters.
New security features added to RCW to prevent inadvertent or non-authorized manipulation of l
system diagnostic computer.
Operator training lesson plans developed and incorporated in training program; OHA f
malfunction added to simulator training program.
Personnel disciplinary action taken.,-
NRC reporting and ECG criteria clarified relative to loss of OHA New procedures developed and issued relative to specifications for software and software l
verification and validation processes; revision of vendor technical manual by vendor; lesson plans developed for system training.
-PSE&G and vendor developing less complicated system testing process.
Software virus checks performed.
{
Managements expectations relative to communicating concerns with faulty system operation conveyed in series of roll-down meetings.
CORRECTIVE ACTION EFFECTIVENESS:
No similar problems have been noted since this event.
dl L
r L
50-311/92 81 DECEMBER 14, 1992 B*
REVIEW OF CIRCUMSTANCES SURROUNDING LOSS OF THE OVERHEAD ANNUNCIATOR SYSTEM AT SALEM UNIT 2 ON DECEMBER 13, 1992 BACKGROUNIh On December 13,1992, a Salem Unit 2 operator discovered that the overhead annunciators had not been updating alarms for about 90 minutes. The operator reset and restored the system within two minutes. Subsequent investigation revealed that a member of the operating crew had been using the Remote Configuration Workstation (RCW) computer and had inadvertently entered a keystroke combination that, when input through the wrong system port, prevented the proper operation of the overhead annunciator system.
[
ROOT CAUSE:
An operator failed to follow procedures relative to the operation of the Remote Configuration Workstation (RCW) which affects the function of the Sequence of Events Recorder and its interface with the overhead annunciator.
[
No alarms or warning were available to alert the operator of mis-positioned switch at the RCW, therefore operators were not aware of OHA status.
Design specifications for OHA were LTA relative to alarm / warning features i
Other items:
[
No loss of annunciator procedure.
Operators were not trained on routine verification of proper system operation.
LTA software review.
Operators knowledge of need to declare Alert on system failure.
General communications with NRC CORRECTIVE ACTIONS:
Design change to incorporate alarm /warmng feature to warn of OHA system malfunctioning.
New procedure for loss of OHA e
l a1, 50-311/91-81 C.
NOVEMBER 10,1991 OF CIRCUMSTANCES INVOLVING THE CATASTROPHIC FAILURE REVIEW (DESTRUCTION AND FIRE) OF THE SALEM UNIT 2 TURBINE GENERATOR
'l i
NOVEMBER 9,1991 i
i BACKGROUND:
The Salem Unit 2 Turbine-Generator was severely damagu$, while at 100% power, on November 9,1991, during performance of Front Standard testing of automatic mechanical turbine trip features. During the test, normal turbine trip functions are isolated and reliance is placed on the ET-20 emergency trip solenoid and two overspeed protection solenoids to function.
j While testing, a momentary AST oil pressure perturbation resulted in momentary closure of
[
turbine steam admission valves, reactor trip, and subsequent load drop from the generator.
However, the ET-20 and the overspeed protection solenoids failed to function due to mechanical i
binding. Consequently, once AST oil pressure returned to normal, hydraulic fluid which normally would have been drained by the functioning of the turbine control solenoid valves, was pressurized to reopen the steam admission valve to the turbine. Subsequently the turbine experienced severe overspeed without any abatement until the operators restored the Front Standard test controls to the normal position, at which time AST-20 functioned to close the l
turbine steam admission valves. The turbine generator was subject to severe damage and fire.
[
ROOT CAUSE:
'Ihree separate solenoid valves failed to function as designed to control turbine overspeed and 1
effect turbine trip.
Insufficient preventive maintenance afforded to the turbine overspeed protection system.
I 4
i Inadequate surveillance testing of solenoid valve-actuated turbine control systems.
j Delayed replacement of Unit 2 solenoid valves after previous component failures were identified I
at Unit 1.
Failure of management and operating personnel to follow procedures and effectively resolve a
~
failed test result involving the turbine overspeed protection system during a previous turbine startup procedure on October 21,1991.
CORRECTIVE ACTIONS (applies to both Units,):
Disciplinary action taker. against individuals who failed to follow the requirements of the turbine startup procedure on October 21,1991.
4 i
4 J
J
_ ~. -
Management expectations commuricated to staff reladve to the importance of procedure adherence.
Operator training revised to include simulator and classroom instrucdon relative to turbine control system funcdoning and testing; all operators trained; operator effectiveness evaluated.
Turbine startup procedures revised and upgraded.
" Conduct of Operations" training, relative to procedural adherence, provided to station staff.
Human performance analysis initiated with lessons-learned to be incorporated.
Commitment Management procedure developed and issued to assure that all commitments and previous LERs are tracked and completed in a timely manner; several other planning and scheduling procedures were revised to assure the proper assignment of priority to items that require action.
Preventive maintenance program established for orifices in the AST system.
Design change for filter installation on AST orifices completed.
c*
i Front st'andard test procedure upgraded and revised.
INPO Awareness Training conducted, Resource Management, and Supervisory and Management Effectiveness Solenoid valves replaced; turbine EHC systems and components have been incorporated in Reliability Centered Maintenance program to assure PM performance; vendor identified PM specifications.
Surveillance program for solenoids revised to independently test each unit and verify function.
Tachometer and recorders reconnected.
Design change completed to provide backup to AST-20 so that it would not be isolated during Front standard testing procedures; electrical ovssycsi channel was added by design change.
Tech Spec have. been amended to clarify operability determinations relative to turbine system surveillance requirements.
CORRECTIVE ACTION EFFECTIVENESS:
Recurrent instances of failure to follow procedures despite numerous efforts by management to ameliorate this common problem.
Management effectiveness is questionable and a contributor in each occurrence.
Operator training is usually an idendfied item that requires remedia] action.
Communication effectiveness is a common feature in these events.
L a
j.."
ATTACHMENT 6 4
i.
j JUNE 24,1994
)
4 I
Docket Nos. 50-272 i
50-311
}
I Mr. Steven E. Miltenberger Vice President and Chief Nuclear Officer j
Public Service Electric and Gas Company P. O. Box 236 Hancocks Bridge, New Jersey 08038 J
Dear Mr. Miltenberger:
l SURIECT:
NRC AUGMENTED INSPECTION TEAM (AIT) REPORT NOS.
50-272/9440 AND 50 311/94-80 l
He enclosed report refers to a special onsite review by an NRC Augmented inWon Team l
(AIT) from April 8 through April 26,1994. The team reviewed the circumstances surrounding i
the automatic reactor shutdown and two automatic actuations of the " safety injection" system that occurred at Salem Unit 1 on April 7,1994.
t l
The report die-1 areas examined during the inspection. 7he inspection focus was on the l
potential safety signi6eance of the events, and included derniled fact-finding, determination of
{
root causes, and evaluation of operational and manageria1 performance. De in= pan *iaa consisted of selective examination of procedures and representative records, observations, and interviews
{
with personnel.
i j
De AIT determined that the prahminant cause of the event was the combination of pre-existing i
equipment problems or vnlaarahilities and the resultant ch.naages to the ein.iers, and operator i
errors that occurred during the transient. Other failures and their causes were reviewed and are l
di=en9ad in the attached report. De AIT concluded that both the equipment problems and operator errors could, and should have been avoided by licensee management through a closer l
review of the operator needs in response to the frequent and er=~*ad trannient conditions
~
resulting from the grass intrusions at the citedadag water structure.
The AIT found the licensed operator response to the initiating event, a Io;s of circulating water, was weak. Operators did not take some actions that they were trained to perform. However, overall operator response was successful in achieving a stable plant fondition; unfortunately, much later in the event sequence than expected, and too late to avoid a significant challenge to the pressurizer power operated relief and safety relief valves.
While we note the actions of PSE&G to improve plant hardware and procedures prior to the event, both hardware deficiencies and inadequate procedures played key roles throughout the event sequence. Also, the actions taken by PSE&G before and during the event to mitigate the frequent grass intrusions at the Salem circulating water structure were both well conceived and
1 i
Mr. Steven E. Miltenberger 2
generally well performed. However, these initiatives were not accompanied by a similar review 2
of task performance and procedural guidance in the control rooms to ensure that licensed operator response to the potential or actualloss of circulating water would also be successful.
It is for these reasons that the NRC views the relatively poor performance of the operating crew during the April 7,1994 event to indicate not just weak performance of certain licensed j
operators; but rather, and more importantly, an inadequate assessment by management of the preyalent operating conditions at the plant and subsequent development of an appropriate operating philosophy to meet the expected needs.
It is not the responsibility of an AIT to determine compliance with NRC rules and regulations r
I or to recommend enforcement actions. These aspects will be developed following additional NRC management review of this report.
A representative from the State of New Jersey, Department of Environmental Protection and Energy (DEPE), observed parts of the onsite AIT inWon activities. A copy of a letter from Mr. Anthony J. McMahon, Acting Assistant Director, Radiation Protection Element, NJ DEPE to NRC is enclosed with this letter. That corrampandance describes three issues not W&mily addressed in the AIT report. Also enclosed is the NRC reply letter describing our plans to
[
i address those concerns, In accordance with 10 CFR 2.790 of the Commission's regulations, a copy of this letter and the enclosures will be placed in the NRC Public Document Room.
We will gladly discuss any questions you have concerning this inMon.
Sincerely, ORIGINAL SIGNED BY:
James T. Wiggins, Acting Director Division of Reactor Safety f
Fela=nns:
- 1. Inspection Report Nos. 50-272/94-80
- 2. Ietter, dated May 20,1994, from A. J. McMahon, NI DEPE to J. T. Wiggins, NRC
- 3. Ietter, dated June 24,1994, from J. T. Wiggins, NRC to A. J. McMahon, NJ DEPE O
I a
~
Mr. Steven E. Miltenberger U. S. NUCLEAR REGULATORY COMMISSION REGION I P
L REPORT / DOCKET NOS. 50-272/94-80 50-311/94-80 LICENSE NOS.
DPR-70 DPR-75 LICENSEE:
Public Service Electric and Gas Company 11 P.O. Box 236 Hancocks Bridge, New Jersey 08038 FACILITY:
Salem Nuclear Generating Station INSPECTION DATES:
April 8-26,1994 INSPECTORS:
Stephen Barr, Resident Inspector, Salem, DRP (Asst. Team
[
Leader)
J. Scott Stewart, E= miner, DRS Iqbal Ahmed, Senior Electrical Engineer, NRR Warren Lyon, Senior Reactor Systems Engineer, NRR John Kauffman, Senior Reactor Systems Engineer, AEOD I.arry Scholl, Reactor Engineer, DRP Richard Skokowski, Reactor Engineer, DRS i
Howard Rathbun, NRR Intern STATE OBSERVER:
Richard Pinney, New Jersey Department of Environmental Protection and Energy ORIGINAL SIGNED BY:
6/23/94 TEAM LEADER:
R. J. Summers, Project Engineer Date Projects Branch 2, DRP ORIGINAL SIGNED BY:
6/23/94 APPROVED BY:
James T. Wiggins, Acting Director Date Division of Reactor Safety
-%eweosc see
l Mr. Steven E. Miltenberger EXECUTIVE
SUMMARY
Areas Insnected: An Augmented Inspection Team (AIT), consisting of personnel from Region I AEOD and NRR, inspected those areas n==y to ascertain the facts and determme probabic causes of the automatic reactor shutdown and multiple automatic initiations of the safety injection system that occurred on April 7,1994. The team ammi the safety significance of the event, including the resultant plant operation with a water (liquid) filled pressurizer and its challenge to the primary coolant boundary integrity and the potential vulnerability of the ultimate heat sink
~
to the same marsh grass intrusions that challenged the plant normal heat sink, which was the initiating event for the sequence of events on April 7. The adequacy of the licenser's design, maintenance and troubleshooting practices relative to the safety injection system was reviewed.
The possibility for any potential generic implications posed by the Salem event was ases-i.
I Results The Augmented Inspec' tion Team (AIT) developed a sequence of events detailing the circumstances surrounding a Salem Unit 1 plant trip and a series of safety injection system actuations. It was found that the events led to the loss of the pressurizer steam bubble and the i
normal reactor coolant system pressure control system, and an Alert declaration. The AIT noted through an event sequence and causal factor analysis that the root causes of key events generally included a combination of component failure and human error. Additional procedural guidance
['
for, and prioritization of work activities of control room operators would have resulted in a l
better response to the event. The AIT found in general that the licensee response to the almost daily event of grass clogging of the ciradadag water screens was very well planned and coordinated for the additional workload at the ciradadng water structure. However, as indicated by the performance of personnel and equipment in response to the April 7 event, the licensee did not adequately plan for, and coordinate, the activities sieg-:--Aing to the additional workload in the control room resulting from the same event.
I Finally, even though some equipment and licensed operators performed poorly during the ensuing transient on April 7, the core and its primary protective barriers were maintained tiuuughout the event.
In addition, the following conclusions were developed as a result of the AIT review and discussed at a public exit meeting held on April 26,1994:
P Summary of
Conclusions:
1.
No abnormal releases of radiation to the environment occurred during the event (Section 3.4).
2.
The April 7,1994 event challenged the RCS pressure boundary resulting in multiple, successful operations of the pressurizer power operated relief valves and no operations of the pressurizer safety valves (Section 3.2).
3.
Operator errors occurred which complicated the event (Section 4).
ii L
1
a I'.
~
Mr. Steven E. Miltenberger EXECUTIVE
SUMMARY
(CONT'D) r i
4.
Management allowed equipment problems to exist that made operations difficult for plant operators (Section 7.2).
5.
Some equipment was degraded by the event, but overall, the plant performed as designed (Section 3).
4 6.
Operator use of emergency procedures was good. However, procedural inadequacies
[
were noted with other operating procedures (Section 4).
7.
Licensee's investigations and troubleshooting efforts were good (Section 5).
[
i se e
iii r
L
. _ - -.. =
n i i nunnun 7A n
[
[om cg'C UNITED ST ATES 1
NUCLEAR REGULATORY COMMISSION e
f REGION t
/
475 ALLENDALE ROAO o
lung OF PRUSS N
VA 19406 1415 l
g
- Dat:let Nos.
j s
50-272 l
50-311 l
50-354 r
n 4
l Mr. Steven E. Miltenberger l
Vice President and Chief Nuclear Officer l
Public Service Electric and Gas Company P.O. Box 236 i
Hancocks Bridge, NJ 08038 2
r Dear Mr. Miltenberger.
i SURTECT:
SALEM AND HOPE CREEK RESIDENT INSPECTION NOS. 50-272/94-11; 50-311/94-11; 50-354/94-09 h
ne enclosed report documents an inspection for public health and safety, conducted by Mr.
i C. Marschall, Senior Resident rnewtor and other members of the NRC resident and regional staff at the Salem and Hope Creek Nuclear Generating Stations for the period
[
between March 27,1994 and April 30,1994. De inspectors dins =4 the findings of this l
inspection with Messrs. J. Hagan and R. Hovey of your staff.
)
Based on the results of this inspection, it appears that a number of your activities were not l
made~i in full compliance with NRC requirements, as set forth in the Notice of Violation.
j nese activities involved at least three occurrences of failure to follow procedures. In accordance with 10 CFR 2.790 of the NRC's ' Rules of Practice, a copy of this letter and its j
enclosures will be placed in the NRC Public Document Room. He responses directed by l
this letter and the enclosed Notice are not subject to the clearance procedures of the Office of i
Management and Budget as required by the Paperwork Reduction Act of 1980, Public law l
No. 96.511.
l In addition, the iamavars identified a non-cited violation at Hope Creek involving failure to monitor reactor water d=%y in accordance with the frequency required by Technical j
Specifications.
l Your cooperation with us is appreciated.
i Sincerely;-
, Q v
l e
i
\\/
I Edward C. Wenzinger, Chief. [
l
'u l
Projects Branch No. 2 Division of Reactor Projects F
k
~
h
~b
- ^ f-
=
a i
i U. S. NUCLEAR REGULATORY COMMISSION REGION I i
i Report Nos. 50-272/94-11 50-311/94-11 l
50 354/94-09
?
j License Nos. DPR-70
~
DPR-75 NPF-57 I
i Public Service Electric'and Gas Company Licensee:
l P.O. Box 236 Hancocks Bridge, New Jersey 08038 Salem Nuclear Generating Station Facilities:
Hope Creek Nuclear Generating Station March 27,1994 - April 30,1994 Dates:
C. S. MarschaH, Senior Resident Inspector Inspectors:
S. T. Barr, Pddant Iaeaaafar J. G. Schoppy, Paeidant Iaaaaa'ar T. H. Fish, Petident !aeaaatar P. P. Sena, Resident Inspaa'ar T. J. Prye, Pddant f==+::=
B. J. McDermott, Reactor Engineer M. J. Buckley, Specialist J. laughlin, EA - =f
/I/J v
I Approved:
[
Date J. R.
f(Chief, ProjectsJ+n= 2A inenection Summary:
This inspection report documents iaWons to assure public he operations, radiological controls, and ha+4if hours of station activities, indading:
maintenance and survei!!ance testing, emergency p@wiscas, security, t
i engineering / technical support, and safety me== ment / quality Summary delineates the inspection findings and conclusions.
% h[b h
[
16 r
L 4.2 Salem - Restart hupection Activities The inspectors reviewed licensee corrective actions for the following items to insure adequacy for Salem Unit I restart.
r A.
Steam Mow Tr====8'*er Damping Clreult Modification On April 7,1994, a spurious high steam flow signal was sensed by the Rosemount flow transmitters and resulted in an "A" train safety iqjection signal. Subsequent to this event, these steam flow transmitters were modified per design change package GX'P) 1EC-3328 i
{
add a damping adjustment. Previous to the April 7 event, transmitter time responses from 0.020 to 0.110 =v=de i'
After modification, the transmitter time response was adjusted j
to 0.225 i 0.025 erande This transmitter time response setting will prevent spurious j
steam line pressure spikes from tripping the steam flow bistable when the reactor is trippe i
and actual main steam flow is below 40%.
l The in~*s reviewed modification package DCP IEC-3328 and the work orders used to i
modify the eight Pmant flow transmitters. The '=Y+c-5 noted that the DCP contained l
a good evaluation of how the madihtion affected the design basis of the affected pnscdon [
circuits. The i=Yuw also reviewed the sensor calibration and thne response test results and concluded that the trannmitters were adequately tested.1he Technical Se% and FSAR were reviewed to evaluate what limits and design basis may be affected by the
{
madihtian The DCP contained a thorough design analysis of the modification which was l
SORC reviewed and approved. The laeaae*ae agreed with the licensee conclusion that implementation of the modification did not raavia* an unreviewed afety question, since i
the transmitter response after the madificatina was less than the thne assumed in the design basis. The inspectors concluded that the modification was well performed and adequately addressed the seenciatert Unit I restart issue.
1
{
1 V'
l l
'f
~
17 B.
Solid State Protection System Steam Flow Input Relays 4
During the initial investigation of the short duration high steam flow signal the licensee identified that twelve of sixteen steam flow input relays were discolored. Subsequent time response testing demonstrated that the relays were all functioning acceptably. Test data for Train A shows that the actuation signal duration must be at least 16 msec for the master relay contact to close. Train B requires an actuation signal of 31 msec. De variance in the roonse time of these relays had minimal impact on the ability of the system to meet overall ddgn requiremer,ts. ' De design specification for time delay of the high steam flow SIS actuation is less than 600 msec.
After the input relay and TS channel time repan_w testing were completed the licenw j
replaced the discolored relays. De apparent cause of the discoloration was the accumulation of a powder-like carbon buildup due to cycling of the relays. The licenw elected to replace the relays based on their appearance, even though the test results indicated the time response of all the relays was acceptable.
De la== *m reviewed the licensee's evaluation, test results, and the discolored relays. De la== *w made a visual ernminatian of the relays and compared the extent of the carbon buildup to the time response test data. No correlation was observed between the apparent
[
carbon buildup and the time response performance. The licenw stated that discolored relays will also be replaced in Unit 2 during the next outage and that they have no reason to question their operability. De inen-M concluded that the licensee's actions regarding the input relay evaluation were acceptable.
C.
Steam How Summator i
De licensee ernmined the output of the steam flow mimmatars during Maintenance and Controls troubleshooting to determine if the mimmatar had caused the spurious signal. De test simulated the adary caatmen of the reactor trip breakers opening and monitoring the output of the mimmatar module. De high steam Sow function compares actual steam flow to a gc==+1 setpoint of 40% of full steam flow between 0 and 20% load and then increasing linearly to 110% of full steam flow at full load. When the reactor trip breakers open the output of the setpoint should decrease to the 40% steam flow value (1.34 VDC).
Strip chart data shows the setpoint actually dipped below the 40% value by 100 mVDC before + 'ag to the low value setpoint.
De licensee is currently invaeignting the phenomenon however, there is no safety significance because the lower setpoint output would couwpoed to a lower reference value being used to determine high steam flow. Based on the minor extent of the phenomenon and its safety significance, the inspector determined that no further NRC review of the issue is rwamy at this time.
F<
l
1 18 D.
Power Operated Rellef Valve (PORV) Operability During the plant response to the Salem Unit I transient on April 7, the PORVs operated a total of at least 300 times. He valves functioned as designed, and during the cooldown at the conclusion of the event, plant staff tested both valves for stroke time and seat leakage.
He valves were judged operable. Details of PORV operation during the transient are provided in NRC ram ~* ion Report 50 272/94-80. Upon valve dianambly, mechanics did not experience difficulty indicative of severe binding while removing valve internals. As a result of the numerous valve operations, the licensee opened IPRI (cycled more than 100 times) and IPR 2 (cycled more than 200 times) for inspection. Bey discovered that IPRI had a small amount of weir randomly distributed around the plug and the interior of the P
cage. He licensee concluded that IPRI exhibited typical wear for the circumstances. De licensee also found that IPR 2 had heavy scuffing on the outlet side of the plug and the cage, and galling on the corresponding side of the valve stem. Additionally, the licensee found small cracks in the plugs for both valves radiating from a hole drilled in the plug.
Because of the degree and orientation of galling wear noted on IPR 2, the licensee assembled a team of engineers to determine the root cause of the wear patterns and to assess the effects of the wear on operation. He internals were shipped to a Westinghouse materials laboratory for analysis of the cracks and assay of the materials on the scuffed surface of the plug. Cvagondence from the valve manufacturer, Copes-Vulcan, indicated that the wear on IPRI was normal and IPR 2 was more than desired but not unusual. De team pr*1imi=ily determined that several factors contributed to the wear observed in IPR 2. De manufacturer designed the valve with small clearances between the plug and cage.
Installation of the valve internals was performed using a procedure which may have contributed to misalignment ofinternal components. Chrome coating from galling of the stem may have migrated to a location between the plug and the cage, contributing to the scuffing of the plug. De hydrodynamic forces present in the valve during valve lifting under operating conditions caused a differential pressure across the valve plug p=hia: it against the outlet side of the cage.
De licensee team noted that the cage-guided design of these PORVs assumed a certain degree of contact wear between the plug and the cage. Dimensional tolerances of 3 to 5 mils bem the plug and cage provide lateral support to the plug which otherwise would only be provided by the stem guide near the top of the valve. De team also noted that the PORVs were reanamhled during 1R11 by installing the bonnet of the valve into the body with the stem, stem packing, and plug pre-assembled into the bonnet. His method was used primarily as a means to reduce personnel radiation exposure during valve maintenance, but did not allow for an adequate check for valve b'mding or misalignment during installation.
Based on the investigation, the licensee revised the re-assembly procedure to include a vendor recommended piece by piece assembly process, including checks for clearance and freedom of movement at each stage ofinstallation.
f I
.t
)
19
)
l During valve reassembly, the licensee installed new valve internals for both PORVs using the revised procedure. No anomalies were noted. Further, as a one-time check to ensure the l
IPR 2 valve body did not contribute to the unusual wear pattern, the licensee cycled the valve 10 times using the air operator. When plant staff subsequently diammbled the valve for ia-c* ion they found no signs of wear. Plant staff again reassembled the valve using the revised procedure.
In summary, licenses enginaaring determined that the galling and scuffing found in 1PR2, l
although heavy, was acceptable. Hey attributed the wear to mitalignment of the valve j
internal during installation. To prevent future mieligamants, the licensee changed the l
procedure for reassembly. ne modified procedure includes manually ' stroking the valve l
during various stages of reassembly to insure that the plug moves freely within the cage.
j' The team noted that the valves operated more than 300 times during the transient, and then passed the operability test. In addition, the vendor considered the observed wear acceptable.
De licenw installed new internals in IPRI and IPR 2.
l At the end of the inspection period the licensee had not determined the cause of the crack.
In addition, the licensee had not yet completed the safety evaluation of PORV operability.
l The t=5+-=s will review licensee root cause and corrective actions in a future in=etion c
~
report. (IFI 50-311/94-11-01) l E.
- Atmwphric Steam Relief Valve (MS 10) Modificaflon i
During the April 7 trentient, one or more steam generator code safety valves lifted on high
]
steam pressure. The MS-10 valves did not autamatienny operate as da=ignad to prevent l
challenges to the steam generator code safetics. Although the Final Safety Analysis Report j
(PSAR) does not take credit for operation of the MS-10 valves in response to an accident, il J
the design of the MS-10 valves was intended to provide steam pressure control, whereas the j
main steam code safety valves were designed to protect the steam generators and piping from failure due to overpressure. As a result of the lack of MS-10 automatic operation a steam j
generator code safety valve lifted. This caused the reactor coolant system (RCS) 4W j
to reduce sufficiently to result in a RCS pressme drup. Inw RCS pressee initinted a safety sigection.
j ne liennwe found that the MS-10 valves did not open because of satmation of the control circuit, as a result of a modification performed in 1977. De licenw consulted with Wadaghaa-in 1977 to identify a solution for inadvertent MS-10 opening during load transients: Hey concluded that removing a " drain circuit," consisting of two diodes and a resistor, wocid prevent inadvertent MS-10 opening. De licenw and Westinghouse did not identify that removal of the " drain cire. tit" created the circuit saturation conditions which caused considerable delay in MS-10 isponse to rapid changes in steam pidiure. In response to the April 7 transient, the licensee reinstalled the drain circuits to provide the correct MS-10 response to pressure changes. In addition, the licensee adjusted settings for j
control circuit response to insure adequate MS-10 response to rapid transients, while i
l i
a 20 miniminng the undesirable response to normal load changes, such as inadvertent opening.
Plant staff modified a spare control card and performed response tests of the card in the simulator control room. He tests demonstrated that reinstalling the drain circuit eliminated the saturation condition experienced on April 7. He licensee planned to confirm the adequacy of the cormetive action through startup testing.
,l n
ne in=riors reviewed the cimuit modification, supporting documentation, the safety
~
evaluation, and results of simulator testing. He inWors concluded that the modification addressed the saturation condition. De inW_ars will monitor startup testing to confirm the that the modified cimuit permits proper operation of the MS-10 valves.
r
[
P m.e O
b L
b pa seco 04 UNITED STATES j
E
' g NUCLEAR REGULATORY COMMISSION y
8 REGION I t,
[
475 ALLENDALE ROAD g*
,oe KING OF PRUSSIA. PENNSYLVANIA 19406 1415 g
I p*o <
Docket Nos. 50-272 T
~
50-311 50-354 i
EA No.94-112 Mr. Steven E. Miltenberger l
Vice President and Chief Nuclear Officer
'j j
Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, NJ 08038
?
Dear Mr. Miltenberger:
i SUBIECT:
SALEM AND HOPE CREEK RESIDENT INSPECTION NOS. 50-272/94-13; l
50 311/94-13; 50-354/94-11 e
s
)
The enclosed report documents an inspection for public health and safety, conducted by Mr.
C. Marschall, Senior Resident Inspector and other members of the NRC resident and 4
i regional staff at the Salem and Hope Creek Nuclear Generating Stations for the period between May 1,1994 and June 25,1994. The inspectors discussed the findings of this inspection with Messrs. J. Hagan and R. Hovey of your staff.
l During the inspection period we noted that you operated the Salem and Hope Creek plants El l
safely overall. However, we also noted that the quality of operability determinations at Salem and Hope Creek could be improved through documentation of the basis for the determinations. Matters pertaining to our followup inspection on issues relative to the
}
April 7,1994 trip of Unit I are provided for your information. In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy of this letter and the enclosure will be placed in the NRC Public Document Room (PDR). No response to this letter is required.
q Your cooperation with us is appreciated.
Si ly,
,s o
s s
/
Edward C. Wenzinger, Chief Projects Branch No. 2 Division of Reactor Projects
Enclosure:
NRC Inspection Report Nos. 50 272/94-13; 50-311/94-13; 50-354/94-11 F-L
- M D7J3noy /f-
,o U. S. NUCLEAR REGULATORY COMMISSION REGION I Report Nos. 50 272/94 13 50 311/94-13 1
50-354/94-11 License Nos. DPR-70 DPR-75 NPF-57 t
Licensee:
Public Service Electric and Gas Company P.O. Box 236 Hancocks Bridge, New Jersey 08038 Salem Nuc' ear Generating Station Facilities:
l Hope Creek Nuclear Generating Station Dates:
May 1,1994 - June 25,1994 Inspectors:
C. S. Marschall, Senior Resident Inspector S. T. Barr, Resident inspector J. G. Schoppy, Resident inspector T. H. Fish, Resident Inspector J. I2ughlin, Emergency Preparedness Specialist R. K. Ierson, Resident nwar i
i S. M. Pmdale, Resident Inspector B. C. Westreich, Resident inwar N. S. Perry, Senior Resident Inspector T. Y. Liu, Project Engineer B. J. McDermott, ctor min-P. F.
' cat M Approved:
/
M/
7/ [
J. R.
'te, Chief, Pfojects Section 2S Mat (
Inmection Sum This iaWon report documents inWons to assure public health and safety during day and backshift hours of station activities, including: operations, radiological controls, maintenance and surveillance testing, emergency preparedness, security, engineering / technical support, and safety assessment / quality verification. The Executive Summary delineates the inspection findings and conclusions.
i i
fe s
_ _ _ ~
.. ~ _ _
~
b l'
14 i
4 f
s I
i j
r-l l
i 1
1 4
3 e 1 L
}
i l
i
[
4
)
l 4.0 ENGINEERING j
4.1 Salem NRC Followup To Unit 2 Power Operated Relief Valves (PORVs) Materials A.
i In response to the licensee's discovery that incorrect intemals for the 2PRI and 2 l
potentially been installed during the last Unit 2 refueling outage (2R7), the inspectors investigated the history of the materials that had been used by PSE&G for the l
2PRI and 2PR2 and reviewed the work process that had been used by the licensee to
}
the valves during 2R7.
i.
The inspectors determined that the original material used for the Unit 2 PORVs when unit went in service in 1981 was 17-4 PH stainless steel. As a result of industry ugience with seat leakage and susceptibility to galling, and in response to TM1 Action Item II.
PSE&G implemented a design change in 1982 and installed PORV intemals (pl i
made with 304 stainless steel with a stellited plug. Previously, the licensee removed l j
0 1
p.
i
.O-i 15 seals from the PORV installations. Based on vendor recommendations, the licensee intended j
to upgrade this material during 2R7 with 420 stainless steel to accommodate the different environment with loop seals removed.
The licensee prepared two work orders (WOs) for the Unit 2 PORVs to be accomplished during 2R7: one for the repetitive task of inspecting the valve internals and replacing them with 304 stainless components if a-*y, which is routinely done at every refueling outage; and another WO for the installation of 420 stainless steel internals to be accomplished per the approved design change package. As part of the WO for the routine inspection of the valves, the licensee pre staged a new set ofinternals in case the parts i
needed to be replaced. Licensee records indicated that the pre-staged set ofinternals was made of 17-4 PH stainless steel; the licensee could not explain why 17 4 PH and not 304 stainless steel had been pre-staged other than the recognition that 17-4 PH parts had still been in the PSE&G inventory. Warehouse material issue tickets indicated that it was this set of pre-staged parts that was mistakenly issued for the performance of the WO that was to accomplish the upgrade to 420 stainless steel.
While the licensee recognizes that the 17-4 PH parts were not intended to be installed in the Unit 2 PORVs, their evaluation concluded that use of the parts was not unsafe, even though i
other material was considered better for this application. 'Ihe licensee's conclusion was based on: the acceptability of 17-4 PH material performance supported by the analysis of test results as presented in the Salem FSAR for original licensing; the continued supply of the l
material by the vendor for other licensees' use; and the fact that, while 17-4 PH stainless steel is more susceptible to long term wear, the current internals will only be used for one
)
fuel cycle (PSE&G plans to replace the internals at the next Unit 2 refueling outage). The licensee performed a 10 CFR 50.59 Safety Evaluation tojustify the continued use of the current installed valve internals for the remainder of the current fuel cycle. The Safety Evaluation was approved by the Station Operations Review Committee.
The inspectors reviewed the licensee records for the procurement and disposition of the 17-4 PH valve materials and for the performance of the two applicable WOs, and discussed the results of the licensee's investigation into the matter with licensee management. The inWors determined that the 17-4 PH stainless steel valve internals manufactured by Copes Vulcan had been issued by the PSE&G warehouse for the material upgrade WO, as evidenced by component tagging and warehouse receipts. 'Ibe iaWors also identified that the component tags for the 17-4 PH parts had been included in the paperwork that had accompanied the WO for the material upgrade design change. 'Ibe inmeters were unable to positively determine ifit had been those components that had been installed into the PORV valve bodies during the performance of the WO, although the iaea~'ars did identify three separate sets of hand-written notes that indicated that the internals that had been removed were separately removed from the work area. All parts that were removed as part of that WO had been disposed of by the licensee, and the inspectors concluded that it will not be until the next outage when the valve is opened for examination that the identity of the valve i
intemals 'will be positively determined.
r-
o!
B 4
16 In addition, the inspectors determined that the licensee had procured a total of stainless steel internals. The inspectors determined that the accounting of those six s supported the belief that 420 stainless steel pans had not been installed The inspectors noted that only two quality control hold points had been requir
[
installation of the valve internals: one to witness a blue contact for a cleanliness inspection of the valve prior to valve closure.
By examination of available 17-4 PH and 420 stainless steel valve internals warehouse, the inspectors determined that there was no readily identifiable means distinguishing the two different materials and that the licensee explanation of 2R7 was plausible. The inspectors determined through their review of the avai (including the 10 CFR 50.59 Safety Evaluation) and discussion with licensee p y
there ms a reasonable level of assurance that the current internals of
- l made of 17-4 PH stainless steel and that these internals do not compromise the abil valves to perform their design function. However, the inspectors also noted that the li failed to prevent the use of incorrect parts (i.e., pans other than those specified by the specifications of the modification).
Atmospheric Steam Relief Valves (MS-10s) and Steam Mow Transmitters B.
h The effect of the atmospheric steam relief valves (MS-10s) on the Salem Unit 1 event of Licensee action April 7,1994 is documented in NRC Inspection Repon 50 272&311/94-80.
to correct the " reset windup" (inability to respond to steam pressure increases) was As documented in that repon, documented in NRC Inspection Repon 50 272&311/94-11.
the MS-10 valves were designed to prevent challenges to the main steam code safety valves In March 1977, the licensee modified the control circuits for the MS-10s to prevent undesired and inadvenent opening of the MS-10s. That modification contributed to the failure of the MS 10s to function appropriately by remaining closed in a condition (on Apr 7,1994) which required the valve to open automatically. Consequently, a main steam safety opened, resulting in a Safety Injection.
l De licensee corrected the circuit by restoring the control circuits to their 1977 configuration l
and making adjustments to insure that the valves operated to control steam pressure as
[
originally intended, ne inspectors'noted that, in response to the June 10 Unit I trip, the l
MS-10s worked as designed.
The licensee found that their analysis for the 1977 modification did not identify the potential that the MS-10s would not respond properly to increasing steam pressure. The licensee subsequently identified the lack of MS-10 response to pressure increases. An approved digital feedwater modification (scheduled for implementation in 1993 at Unit 1) was to resolve the deficiency. However, the licensee delayed the implementation of the modification. Consequently, the licensee allowed the MS-10s to function in a degraded F- '
a
,s 17 failed to prompdy identify and correct the m censee until April 7,1994, m March 1977, During the April 7th transient, the first. safety injection (SI) sign temperature coincident with a spurious'high steam flow signal f h aused by low RCS equipment that injected water into the vessel o s ort duration. The e ra n of that event sequence records for reactor and turbine trips on Ju and February As noted above, the licensee failed to promp een present.
least, those three previous occasions.
r ous signal on, at I
l k
f te b
h