ML20155K021

Partial Response to FOIA Request for Documents.Forwards App D Documents.App C Documents Available in Pdr.App E Documents Completely Withheld (Ref FOIA Exemption 5)

ML20155K021
Person / Time
Site:	Davis Besse, Palisades, Perry, Fermi, Kewaunee, Point Beach, Brunswick, Cook, LaSalle, 05000000
Issue date:	06/13/1988
From:	Grimsley D NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
To:	Maxwell B COSMOS BROADCASTING CORP.
Shared Package
ML20155K026	List: IA-88-165, Discusses Topics for Senior Mgt Meeting & Submits Partial Listing of Facilities Where Vendor Support Has Been Problem,Vendor Involved & Component/Issue of Interest. W/Copyrighted Matl ML20155K021 ML20155K074 ML20155K102 ML20155K120 ML20155K139 ML20155K174 ML20155K200
References
FOIA-88-165 NUDOCS 8806210087
Download: ML20155K021 (9)

Text

{{#Wiki_filter:U.S. NUCLEAR REGULATORY COMMISSION %*c roa noutst %vveinisi f(r Es,'h, [g/ a FOIA 88-165 a I s n$mSitui C i R ESPhN E TO FREEDOM OF l ' **' M I ''

• g,*., / '

JUN 13 2 . / INFORMATION ACT (FOIA) REQUEST owstWaWd%.wr/dal d RA -- LEQUE Sit a Mr. Bruce Mansell PART 1.-RECORDS RELE ASED OR NOT LOCATED ISee checked bonest No agency records subject to the request have been located No additonal agency records subrect to the request have been located Agency records subrect to the roovest that are cerittfed in Appendis b are already evelaNe for WN< inspecten and copq e the NRC Pubhc Document Room. X 1717 H Street. N W.. Washington DC Agency records sobr ct to the request th.t are identifed e Appenda. O* are beeg made ava4aNe for pubhc especton and coppg a the NRC Pubhc Documeet e X aoom. '717 H St eet N w. wassegton. DC. - a foide, und.t the FOtA numw a*c reavester name The nonproprietary verson of the proposaHsi that you agreed to accect e a te,echcre :cmersaton weh a member of ray staff a now beeg enaoe avadatae for pubhc inspecton and coymg at the NRC Pubbc Docurrent Room.1717 H Street. N W. WasNrigton. DC. in a foeder under tNs FOIA number and requestet name. Enclosed e eformaton on how you rnay obtain eccess to and the charges for copyeg 'ecords placed e the NRC PutWic Document Room.1717 H Street. N W., Washington DC. Agency records subrect to the request are enclosed Any appl <abie charge for copes of the records provided and paywent procedvies are noted in the comments secton Records subsect to the request have been referred to another Federal agencytest for reww and direct respoese to you in vow of NRC's response to the request, no further acton a beeg taken on appeal eer cated PART ll. A-INFORMATION WITHHELD FROM PUBLIC DISCLOSURE Certam informaton m the reques'ad records e being wethheld from publ< drselosu e pu suant to the FOIA enemptons descrted in and for the reasoes stated in Part it, sec-r r t:ons 8. C. and D. Am re' eased portons of the documents for *Nch or9y can of tN ecord is beeg arthheid are beirg made ava4abie for pubhc inspecton and copyeg in y the NRC PuNic Document Room.1717 H Street. N W. Washrgton DC, in a fo der unoer the FOIA number and requester name Comments

• Copies of Appendix 0 records are being made available at the Local Public Document Room maintained in the Perry Public Library, 3753 Main Street, Perry, Ohio.

) i l i 8806210087 880613 PDR FOIA MAXWELL 88-165 PDR A .J Sic TL DiMCTOA. DYSc% e a ESANO c5 0 A4d*2 _ N*.C FOR M a64 Pri o 3 ss t.

i JUN 131989 FREEDOM OF INFORMATION ACT RESPONSE FOiA NUMBERIS) 88-165 om PART 11 B-APPLICABLE FOIA EXEMPTIONS Records subject to the request that are described in the enclosed Appendices E a,e be,ng,inneid ;n the,, entrety or in part under FOIA i Exemptons and for the reasons set forth below pursuant to 5 U.S.C. 552(b) and 10 CFR 9 5(a) of NRC Regulatons. l 1

1. The wthheld anformation a propedy clasaded pu suant to Enecutive Order 12356 (EXEMPTION 1) r 1

The wthhe6d informanon redates soiety to the otemat personnel rules and procedures of NRC. 4 EXEMPTION 2) 1 ) 3 The wthbend oformation e specAca#y esempted from put*c d.sclosure by statute edicated IEXEMPTION 3) Secten 141145 of the Atomic Energy Act *Nch proNbets the disclosure of Restncted Data or formerty Restncted Data (42 U S C 21612165L i tx ten 147 of the Atom.c Energy Act wNch proNtzts the disclosu e of Unclassifed Sa'eguards information 142 U S C. 2167L r

4. The wthheld informabon e a trade secret or comtnercel or financel informaton that ts being withheld for the reasor4sl endicated IDEMPTION di The mformatcn 4 coredered to be conidential busmess (propretary) mformatson.

The informaten is cor*dered to be propretary eformation pursuant to 10 CFR 2.790tdtt1L The information was sutetted and recewd m contdence from a fore.gn source pursuant to 10 CFR 2.730tdH2L

5. The wthbeid information cor sists of eteragency or intraagency recordt that are not available through discovery dunng litgation. Deirere of predeceonal mformaten would tend to inNbet the ocen and frare enchanQe of dess essentel to the deliberstw process. Where records s's wthheld in thew ermrety, the facts are ineutricably etertened wth the predecisenal eformaten There arso are no reasonably segregable f actual portens because the reesse of the facts would permd an anderect inquiry ento the predeceonel process of the egency (EXEMPTION 51 L

6. The wthbe6d eformation a enormted from putAc dedosure because its disclosure would result m a clearty unwarranted invasion of personal privacy. (EXEMPTION 6)

7. The wthbed informatson Consists of arwstgatory records compded for law enforcement purposes and a bemg wthheld for the reasorbsi mdscated. (EXEMPTION 7)

Desclosure would eterfere wth an enforcement proceeding because et could reveal the scope. detectson, and focus of enforcernent 9"orts and thus Could possery enow them to take acten to shieid potential wongdoing or a violaten of NRC requirements from evestgators IEXEMPT60N 7tAH Dociosure would constitute an urwerranted invasen of personal pnvacy (EXEMPTION 7tCH The snformaton consets of names of edividuals and other eformation the disclosure of wAch would reveal identrnes of confeder t.e sou ces. (EXEMPTION 7tDH r PART 11 C-DENYING OFFICIALS Pursuant to 10 CFR 9 9 and 'or 915 of the U S. Nucer Regulatory Ccrwemsson regulatons, et has been octermeed that the informaton withhe6d e eienwt from producten or decknure, and that its producten or dociosure e contrary to the putAc interest. The persons responsee for the densal a*e those OHicels identrfed below as denymg officials and the Director. Dmson of Rules and Records. Office of Admestraten. for any cenws that may be appeas d to the Enecutne Director for Operatons <Eoo-e DENylNG OFFICIAL TITLE OFFICE RECOROS DENIED APPELLATE OFFICIAL SECAFTARY too Mr. Bert Davis Regional Administrator App. E X fy A A A PART 110- APPEAL KIGHTS The onial by each denying officialident/ed in Part IC rnay be appealed to the Appettate Official identfied in that section. Any such appeal must be in writing and must be made within 3) days of receipt of thrs response. Appeals must be addressed as appropriate to the Executrw Director for Operations or to the Secretary of the Comrnisson, U.S. Nucker Regulatory Commission, Washington, DC 20555, and should clearly state on the envelope and in the letter that it is an "Appeal from en initel FOLA Decision." N AR R RY COMSSION FOlA RESPONSE CONTINUATION

1 Appendix C F01A 88-165 Records Available in the PDR Date Description 1. 11/09/87 Ltr Cleveland Electric illuminating Company to NRC ACCESSION N0. 8805130239 ] 2. 11/12/87 Ltr Cleveland Electric Illuminating Company to NRC ACCESSION N0. 8711240094 i 3. 11/13/87 Ltr Cleveland Electric Illuminating Company to NRC ACCESSION NO. 8711200278 4 11/23/87 Ltr Cleveland Electric Illuminating Company to NRC ACCESSION N0. 8712040317 5. 11/25/87 Ltr Cleveland Electric illuminating Company to NRC ACCESSION NO. 8712030169 ) 6. 11/30/87 Confirmatory Action Letter ACCESSION NO. 8712070041 7. 11/30/87 Ltr Cleveland Electric illuminating Company to NRC i ACCESSION N0. 8712070358 8, 12/04/87 Ltr Cleveland Electric Illuminating Company to NRC f ACCESSION N0. 8712110227 9. 12/08/87 Ltr NRC to Cleveland Electric Illuminating Company ACCESSION N0. 8712140480

10. 01/22/88 Inspection Report No. 50-440/87024(DRS)

ACCESSION NO. 8802030312

11. 02/10/88 Inspection Report Nc. 50-440/87027(DRS)

ACCESSION NO. 8802160330

12. 02/12/88 Ltr Cleveland Electric Illuminating Company to NRC ACCESSION NO. 8802180163

13. 02/19/88 Inspection Report No. 50-440/87027(DRS)

ACCESSION NO. 8802160330

14. 03/18/88 Ltr Cleveland Electric Illuminating Company to NRC ACCESSION N0. 8803220099 15.

11/13/87 Ltr NRC to Cleveland Electric Illuminating Company ACCESSION NO. 8711190227 i

Appendix 0 FOIA 88-165 Records Being Placed at the PDR & LPDR Date Description 1. Various dates Various work orders, logs and reports (328 pages) 2. 10/29/87 Report Evaluation Form / Rill Tracking System (1 page) 3. 10/29/87 Condition Report (5 pages) 4. 10/29/87 Event Notification Worksheet (2 pages) 5. 10/30/87 Daily Report Region III (1 page) 6. 10/30/87 Reportable Event Number 10515 (1 page) 7. Undated Handwritten Partial Sequence of Events (2 pages) 8. 11/87 Perry Power Plant Photos of MSIV Inspection Due to Slow Valve Closure (44 pages) 9. 11/87 Sequence of Events and Operator Actions (2 pages)

10. 11/02/87 Daily Report Region Ill (1 page) 11.

11/02/87 Daily Report Region III (1 page)

12. 11/03/87 Preliminary Notification re: Shutdown Because of Excessive MSIV Closure Times (1 page)

13. 11/03/87 Event Notification Worksheet (2 pages)

14. Various dates Slow Strokes /MSIVs w/various documents from Cleveland Electric (54 pages) 15.

11/04/87 Confinnatory Action Letter A. B. Davis to M. R. Edelman (1 page) 16. 11/05/87 Quarantine List of Main Steam Isolation Valves (1 page) 17. 11/05/87 Memo J. P. Eppich to K. R. Pech re: Relationship of MSIV Air Pack Vendors w/ attachment (2 pages) 1 18. 11/05/87 Sequence of Troubleshooting Plant (6 pages)

19. 11/05/87 Memo E. G. Greenman to R. D. Lanksbury re: AIT Charter w/AIT Charter (3 pages) 20.

11/05/87 Perry Nuclear Power Plant Work Orders (16 pages)

Appendix D F01A 88-165 Date Description 21. 11/05/87 MSIV Forced Shutdown Recovery Plan (1 page) 22. 11/05/87 Draft evaluations and actions (7 pages) 23. 11/5-6/87 Plan of the Day (15 pages) 24. 11/06/87 AIT Action items (3 pages) 25. 11/06/87 Perry Nuclear Power Plant Work Orders (114 pages) 26. 11/06/87 Nuclear Quality Assurance Dept inspection Rpt (2 pages) 27. 11/06/87 Nuclear Quality Assurance Dept inspection Rpt (6 pages) 28. 11/06/87 Nuclear Quality Assurance Dept Inspection Rpt (3 pages) 29. 11/06/87 Nuclear Quality Assurance Dept Inspection Rpt (2 pages) 30. 11/06/87 Memo F. A. Kearney to M. W. Gmyrek re: Training Procedures Relative to Failure of MSIV Closing Events w/attachn.ents (12 pages) 31. Undated Analysis of loading on steamlines w/ attachments a. 11/6/87 ltr T. R. M:Intyre to J. Eppich re: effects of isolation of 3 main steam lines b. 11/6/87 memo J. E. Meyer to J. P. Eppich re: main steam line design (4 pages)

32. Undated Perry Nuclear Power Plant Evaluations of Single MSly Slow Closure w/ attachments a.

11/6/87 ltr D. D. Jones to G. Rhoades re: estimate of mass flows for treak outside of containment b. 11/6/87 ltr M. M. Waselus/J. Icannidi to X. R. Pech evaluation boundary dose with a single MSIV closure at 18 seconds w/ calculations c. 11/4/87 memo L., S. Burns to T. R. McIntyre re: effect of isolation delay or failure in one steamline d. 11/4/87 meno T. R. McIntyre ts J. P. Eppich re: MSIV closure testing (20pages) 33. 11/07/87 Perry Nuclear Power Plant Work Orders (13 pages) 34 11/07/87 Nuclear Quality Assurance Dept inspection Rpt (2 pages) 2

Appendix 0 F0IA 88-165 Date Description 35. 11/07/87 Step Description (12 pages) 36. 11/07/87 Sequence of Troubleshooting Plan (15 pages) 37. 11/08/87 Preliminary Results Summary: Instrument Air at MSIVs (4 pages) 38. 11/08/87 Nuclear Quality Assurance Dept inspection Rpt (2 pages) 39. 11/08/87 Nuclear Quality Assurance Dept inspection Rpt (2 pages) 40. 11/08/87 Nuclear Quality Assurance Dept inspection Rpt (3 pages) 41, 11/08/87 Draft Root Cause Analysis Executive Sumary (61 pages) 42. 11/09/87 Perry Nuclear Power Plant Work Order (30 pages) 43. 11/09/87 Memo K. Matheny to V. Concel re: MSly ASCO Solenoid Valves (1 page) 44. 11/09/87 Memo J. P. Eppich to G. G. Rhoads re: MSIV Closure / Scram Test w/ attachments a. 11/6/87 ltr T. R. McIntyre to J. Eppich b. 11/6/87 memo J. E. Meyee to J. P. Eppich re: Main Steam Line Design (4 pages) 45. 11/09/87 AIT Action items (3 pages) 46. 11/09/87 Memo K. Matheny to W. Kanda re: MSly Temperature Monitoring w/ handwritten notes and sketches (6 pages) 47. 11/09/87 Ltr M. R. Edelman to A. B. Davis w/ Executive Summary (6 pages) 48, 11/09/87 Root Cause Analysis Executive Summary (69 pages) 49. 11/10/87 Memo E. G. Greenman to H. G. Miller re: Transfer of Lead Responsibilities - Main Steam Isolation Valves (MSIV)atPerry(1page) 50. 11/11/87 Oraft evaluation and actions w/ fax request (9 pages) 51. 11/12/87 Draft evaluation and actions w/ fax request (9 pages) ~ 52. 11/12/87 Draft evaluation and actions w/ fax request (5 pages) 53. 11/13/87 Ltr M. R. Edelman to A. B. Davis w/ enclosures (10 pages) 3

Appendix D F0IA 88-165 Date Description 54 11/23/87 Perry Nuclear Power Plant, Task 3 Solenoid Valve Environmental (Thermal Endurance) Test Plan (26 pages) 55. 12/02/87 Report on Particles - Instrument Air at MSIVs (4 pages) 56. 12/30/87 Ltr A. Kaplan to A B. Davis w/ Final Report MSIV 3-Way Dual Solenoid Valve Failures (82 pages)

57. Undated ISEG Review of R61 & C91 for MSly Failure (78 pages)

58. Undated Figure M47-1 Steam Tunnel Cooling System (1 page)

59. Undated Class 1E Solenoid Valves (1 page)

60. Undated Handwri. ten note (1 page)

61. Undated ASCO Installation and Maintenance Instructions General Purpose and Explosion-Proof Solenoids (5 pages)

62. Undated ACCO Installation and Maintenance Instructions 3-Way Nuclear Power Plant Solenoid-0perated Pilot Valves (6 pages)

63. Undated Description of valves (2 pages)

64. Undated Handwritten Sequence of Events (33 pages)

65. Undated STI package for test on 10/24/87 (21 pages)

66. Undated Reportability Review (1 page)

67. Undated History (1 page) 68.

Undated Schematic Control Diagram (1 page) 69. Undated Valve drawings (2 pages)

70. Undated Work order description and sumary (3 pages)

71. Undated Handwritten outline (5 pages)

72. Undated System design (1 page)

73. Undated ASCO 1E printouts w/ drawing and bill of inaterial (13 pages) 74 Undated Various drawings (46 pages) 4

Appendix D F01A 88-165 Date Description

75. 02/17/88 Memo C. E. Rossi to H. J. Miller re: Transfer of Operating Event Long-Term Followup w/ enclosures EFR 87-172, EFR 88-02 and TAC 67092 (14 pages) 76.

03/09/88 Memo E. G. Greenman to A. B. Davis re: Topic for Senior Management Meeting (3 pages) 77. 10/30/87 Morning Report - Region II (2 pages) 78. 10/30/87 Daily Report - Region III (1 page) 79. 10/30/87 Reportable Event Number 10515 (1 page) 80. 11/2/87 Daily Report - Region III (1 page) 81. 11/3/87 Reportable Event Number 10560 (1 page) 82. 11/3/87 Preliminary Notification of Event or Unusual Occurrence PNO-III-87-138 (1 page) 83. 11/4/87 Letter to Murray Edelman from A. Bert Davis (2 pages) 84 11/5/87 Memorandum for R.D. Lanksbury from Edward Greenman,

Subject:

AIT CHARTER (3 pages) 85. 11/6/87 AIT ACTION ITEMS (3 pages) 86, 11/9/87 Letter to A. Bert Davis from Murray Edelman (6 pages) 87. 11/10/87 Memorandum for Hubert Miller from Edward Greenman,

Subject:

Transfer of Lead Responsibilities - Main Steam Isolation Valves (MSIV) at Perry (1 page) 88. 11/10/87 Notice of Significant Licensee Meeting (1 page) 89. 11/13/87 Letter to A. Bert Davis from Murray Edelman (10 pages) 90. 11/30/87 Event followup Report 87-172 (2 pages) 5

1 Appendix E FOIA 88-165 To be withheld entirely Date Description Exemption 1. Undated "Pre-Decisional Talking Paper"(3 pages) 5 l _,-,y p,-. s-m--- W

4 4 e e m Ccc O .'\\ W= ) 8 h 4 ..._---___._,-,..-...,_...-.n.. ,._.w., ,w..,,, ,,.,,._.-,e--,n-+,n,..an_, n.,

dTOL-TV no n w w 'IOLEDO 1-c~ aWl feiephone (419) 241:ti Costnos Broadmong Corpo,adon ih A CBS Amlwe $ 1 FREEDOM 0F INFORMATION ACT REQUEST f March 7, 1988 FozA-ff-/8 Freedom of Information Officer b 18 3 -// -(( U.S. Nuclear Regulatory Commission Washington, D.C. 20555 FREEDOM OF INFORMATION REQUEST

Dear Sir or Madam:

This is a request for information under the provisions of the Freedom of Information Act as amended (5 U.S.C. 552). I request a complete and thorough search of all filing systems for all records or documents maintained by your agency pertaining to an incident that occurred at the Perry nuclear power plant in Ohio. The incident involved problems with closure of main steam isolation valves. The test that found the problem started on Oct. 29, 1987, and the plant was shut down on Nov. 3,

1987, because of the valve problem.

I am seeking copies of any and all documents related to this incident. These documents include, but are not limited to, utility reports on the incident and its causes; daily operating reports for the plant for the days in question; correspondence between the utlity and the NRC about the incident; any internal NRC reports or documents about the incident, including minutes of any internal NRC meetings at which the incident was discussed; and any reports by outside consultants about the incident. I am making this request in my capacity as a news reporter for WTOL-TV. I hope to use the requested information for a i news story. l If all or any part of this request is denied, please cite l the specific exemption (s) which you think justifies your refusal to release the information and inform me of your agency's administrative appeal procedures available to me under the law. If there are any fees for searching for, or copying, the records I have requested, please supply the records without informing me if the fees do not exceed $50.00. - fr~~ 's a:> v v - ,py p

l-Freedom of Information Officer Page 2 As you know, the Act perm.ts you to reduce or waive the fees "if disclosure of the information is in the public interest because it is likely to contribute significantly to public understanding of the operations or activities of the government and is not primarily in the commercial interest of the requester." I believe that my request fits this category, and I therefore ask that you waive any fees. I would appreciate your handling this request as quickly as possible, and I look forward to hearing f rom you within 10 working days, as the law requires. Sincerely, OJ Bruce Ma well Reporter

1 PREllHINARY NOTIFICATION OF EVENT OR UNUSUAL OCCURRENCE--PNO-Ill-87-138 Date November 3, 1987 j This preliminary notification constitutes EARLY notice of events of POSSIBLE safety /2 or public interest significance. The infortnation is as initially received without verification or evaluation, and is basically all that is known by the Region III staff on this date. Facility: Cleveland E:ectric Illuminating Licensee Emergency Classification: Company Notification of an Unusual Event Perry Alert Perry, OH 44081 Site Area Emergency General Emergency Docket No. 50-440 X Not Applicable

Subject:

SHUTDOWN BECAUSE OF EXCESSIVE MSIV CLOSURE TIMES At 1:38 p.m. (EST) on November 3,1987, the licensee initiated en orderly reactor shutdown fromg3 percent power after two Main Steam Isolation Valves (MISVs) failed to close during testing within the required Technical Specification limits. The remaining six MSIVs closed within the 3 to 5 second requirement. There are two valves in sequences'on each of the four Main Steam Lines, one valve inside the reactor containment and one valve outs (de the containment. The valves that did not meet the closure time were both on the "D" steam line -- the inboapd valve closed in 18 seconds and the outboard valve had not closed within the two minutes %e control switch was in the close position. The tests were observed by the Resident' Inspectors. Subsequent tests of the two valves resulted in closure times within the Technical Specification limit. During testing on October 29, three valves did not meet the closure limit -- the two valves on the "D" line and the outboard valve on the "B" line. Subsequent cycling of the valves provided acceptable time respons s. (MT An Augmented Inspection Team 3 s eing dispatched to the plant site to review the circumstances 1 and possible causes of the MSIV closure problems. The team will consist of the resident inspectors and personnel from Region Ill (Chicago) and the Office of Nuclear Reactor Regulation. A Region III Supervisor will head the team. Region 111 will issue a Confirmatory Action Letter to the licensee documenting the licensee's agreement not to resume operation of the plant without concurrence of the Regional Administrator. l The State of Ohio will be notified. Region 111 was informed of the test results and shutdown at 1:15 p.m. (EST) by the licensee. This inforination is current as of 2:30 p.m. (EST). 1 CONTACT: R. Knop (FTS 388-5547) M. Ring (FTS 388-5602) DISTRIBUTION: H. St. EDO NRR E/W Willste Mail: ADM:DM8 Chairman Zech PA IE NMSS D0T:Trans only Com. Roberts ELD OlA RES Com. Bernthal AEOD NRC Ops Ctr Com. Carr Com. Rogers ACRS SP Regional Offices SECY CA INPO NSAC POR Rill Resident Office Licensee: (Corp. Office - Reactor Lic. Only) Region !!! i 2. D.O /g__, om nunce inn 7 n^ p u.sv

1 1 6 t u s. m... ...., s...... UN11LD ST Alt s

• *' N 4,b, NUCLE AR REGULATORY COYMisstoN 4

i / ' - S mtutoN iii i n, noeuvio ono

,g e,/ j t

e s e e m.,.ui. i ..o, . y~ f NOV 4 19 0 Docket N:. 50-44C Docket No. $0 a41 The Cleveland Electric 111ur.inating Company ATTN: Mr. Murray R. Edelman Vice President Nuclear Group Post Office Box 5000 Cleveland, O'! 44101 Gt. tie e: This letter confirms the telephone conversation on hoveeber 3,1957, bit ee Mr. Greent.an and others of this office and Mr. A. Kaplan of your staf f regarding the Main Stcat, Isolation Yalve (MSIV) f aGures occurrin; at t' !With regard Perry Nuclear Power Plant Unit 1 on hove her 3, ISE7. j r.atte s discussed, we understand that you will: ) Take those 5:tions necessary to ensure that complete d0cu.entary evicen:c 1. of the *as found* condition of equipment being inspected is maintained. J Provide a step by step troubleshooting program to establish the root 2. cause of the M51Vs failure to meet acceptance criteria. het disturb any cotipotents that of fer a potential for being the ro;; 3. cause including power sources, switches, scler.ofds, and the air syste directly feeding the MISYs ur.til that action is approved by the M0 MT tear iester. Except as dictated by plant safety, advise the NRC AIT Leader prior te, 4 Such notification should conducting any troubleshooting activities. be provided soon enough to allow time for the team leader to assign an inspector to observe actisities. Submit to NRC Region Ikl a forr.a1 repert of your findings and conclusions 5. within 30 days of isteipt of this letter. None of these actions shot 1d be construed to take precedence over actions which you feel necessary to ensure plant and personnel safety. We also understand that Perry Nuclear Power Plant Unit I will not be made critical withcut the concurrence of the Region Ill Regional Ad:ninistrator or his designee. .y ' 5 I C - =

...u m e.un Atlich 1.E7TER i n The Cleveland Electric Illuminating 2 NOV 4 iM7 Company ( Please let tr.e kno= imediately if your understanding differs from that set ab*vt. $1ncerely, i A-d A. Bert Davis Regional Administrator F. R. Stead, Manager, Perry cc: Plant Technicel Depe-tnent M. D. Lys ter, kanager, Perry Plant 0;tra*. ices Depa rtinent Ms. E. M. BJ:relli, General Supervisirg Engineer, Licensing and Corsitance Section DCD/DCB (RfCS) Licensing Fe( Ma'agecent Branch F.esident Inspecter, RJII Harold W. Kohr., Ohio EPA Terry J. Lodst, Esq. James W. herris, State of Ohio ( Robert M. Quillin, Ohio Depertment of Health State c' Otte, Pwbilt utilities Co rissien J. M. Taylor. DEDO

7. E. Marley, hRR J. Lieberman, CE R. Cooper, EDO k'. Lanning, hER F. Miraglia, NRR G. Holahan, NRR i

J. Partlow,,NRR M. Virgilio hRR K. Connaughton, SRI J. Strasma, Rll! CONFIRMATORY ACTION LETTER

1. N e

e

s /f UNITED sf A1ES d.Mo oq\\ g NUCLE AR REGULATORY COMMi$slON f e qY $ REGloN lll 799 ROOSEVELT ROAD 5 [ OLEN ELLYN. ILLINoll 60137 NOV 5 id MEMORANDUM FOR: R. D. Lanksbury Team Leader, Perry Augmented Inspection Team (AIT) FROM: Edward G. Greenman, Deputy Director, Division of Reactor Projects

SUBJECT:

AIT CHARTER Enclosed for your implementation is the Charter developed for tha inspection of the events associated with the Perry MSIV failures which occurred on October 29 and November 3, 1987. This Charter was prepared in accordance with the NRC Incident Investigation Manual and the draft AIT implementing procedure issued for use on October 2,1987. As stated, the c' 9ctives of the Ali are to communicate the facts surrounding this event gional and headquarters management, to identify and comunicate any ric safety concerns related to this event to regional and headquarters management, and to document the findings and conclusions of the onsite inspection. If you have any questions regarding these objectives or the enclosed Charter, please do not hesitate to contact either myself or R. Knop of my staff. &k $mnd Edward G. Greentran, Deputy Director Division of Reactor Projects

Enclosure:

AIT Charter cc w/ enclosure: A. B. Dcvis, R!!! C. J. Papertello, Rll! F. Miraglia, NRR J. Partlow, NRR C. Rossi, NRR G. Holahan, NRR W. Lanning, NRR M. Virgilio, NRR R. Cooper, EDO K. Connaughton, SRI

Perry MSIV_ Stroke Time Failure AugmentedInspectionTeam(AIT) Charter Investigate: 1. Failure of MSIVs to close/close within Technical Specification limits. 2. Safety Significance Root Cause(s). 3. Interaction of prior maintenance activities to the event. 4. Safety implications if actual Group 1 isolation signal had been present. 5. History of any previous problems. 6. Broader Implications e.g. Other systems, other valve / components. 7. Event Reporting. ~ 8. Conclusions. Questions for Perry AIT 1. Failure of MSIVs to close/ clos within Technical Specification limits. (10/29/87 and 11/03/67) KC 1.1' What was the sequence of events? ' 1 1.2' What were the closure times generated during the surveillance? - f 1.3' What operator actions were taken during the event? Were they ' appropriate? 1 1.4' 15 there a history of any previous problems (e.g.10/29 event, etc) / with the tiSIVs? W0id T.he id5 iogic manup per design Jur 169 the surve111a17 tem h 1.6' What additional testing was being performed? 2. Safety Significance Root Cause(s). M 21 Was there any immediate safety significance from this event? If ' ! g o so, what was significant? 4 2.2 What was the root cause of the event? / 3. Interactions of maintenance activities to the event. SE 3.1 What is the past and present maintenance history of the MSIVs? 3.2. What is the maintenance history of the Service Air (SA) and - InstrumentAir(IA). v 3.3 What testing was perfomed as the result of maintenance activities? ' MA53.4 What is the material condition of the effected valves and inter-connected instrument air and control systems as it would affect the valve closure function? 4. Safety implications if actual Group I isolation signal had been present, kC 4.1 Does the licensee have procedurefin place to handle this event?- l 4.2 Are they adequate? / P 4.3 Have the operators been trained on them?

s' i .15 4.4 Does the accident analysis bound this event? g 4.5 What actions were taken by the operators? < 4.6 Was the event properly categorized? ' tt.V An4, okyfw6, reported as required? 'mp os du&eywctL 3, e.lodg+( 4.7 Wcs the nt i 35 5. History of any previous problesis. %) J5/F0 5.1 Have there been previous events similar to this? ' 5.2. If there were preyious events was the licensee aware of them? - 5.3 If not, why not? 5.4 Is there infonnation available on c,ther similar events? ' 5.5 Have there been any IEIN's or IEB's issued or s411er sub,jects?' / 5.6 Is there infonnation avaiable from other sites of similar problems? i 6. Broader Implications. C% f 6.1 !s a IE!N or IEB warranted or a result of this event? b t 6.2 Are there other valves or instruments that require investigation? O 6.3 If the problem lies external to the MSIV's, are there generic g implications? e.g. for other plant systems or other plants with same components, yo 7. Conclusion. DL 7.1 What corrective actions are proposed, and are they adequate? 7.2 Examine generic isnplications to other plants and advise NRC r.anagement subsequent to the site inspection. 7.3 Document inspection findings in accordance with drcft manual chapter 0325. i i

r PO box 97 e PERRY. OHIO 44o81 e TELEPHONE (216) 259 3737 e AoDRESS io CENTER AOAD 1 Serving The Best Location in the Nation Murray R. Edelman PERRY NUCLEAR POWER PLANT E YtG PS2JIDE%7 WC11AR November 9,1987 PY-CEI/01E-0288 L Mr. A. Bert Davis Regional Administrator, Region III U.S. Nuclear Regulatory Commission 799 Roosevelt Road Glen Ellyn, Illinois 60137 Perry Nuclear Power Plant Docket No. 50-440 Augmented Inspection Team (AIT) Status and Commitments

Dear Mr. Davis:

This letter provides a preliminary description on the sequence of events, troubleshooting, and conclusions surrounding the MSIV slow closure problem. It also contains corrective actions and commitments made to the AIT on November 9, 1987. The formal report required by your Confirmatory Action letter dated November 4, 1987 will be submitted on or before December 4, 1987. Based upon the information provided to date, we plan to restart the plant on November 10, 1987 with your concurrence. If you have any questions, please feel free to call. Very trul urs i Murray R. Edel n Senior Vice President Nuclear Group MRE:njc Attachment cc: K. Connaughton T. Colburn Document Control Desk D ^, ps n W n>

lC 1 I. Executive Summary on October 29, 1987 the Perry Nuclear Power Plant was completing the final stages of the Startup Test Program.- One of these tests involved fast closing one Main Steam Isolation Valve (MSIV). During this test, the valve (1821-F028D) failed to stroke closed within the required time. All other MSIVs (7) were cycled in order to verify adequate stroke times. Two of the other MSIVs failed to satisfy the required stroke time. The three valves that initially failed were stroked satisfactorily upon subsequent demand. Based on industry experience involving MSIV control air, the problem was attributed to a one time deposit of debris in the respective solenoids which was exhausted as shown by the subsequent successful stroke. The debris was believed to have caused a delay in the solenoid responses. Based upon the satisfactory stroke, the valves were considered operable and startup testing resumed. Plant management decided to perform the MSIV stroke tests again prior to the last startup test, the full MSIV isolation scram. This approach was discussed with NRC Region III. On November 3, in preparation for the final startup test, additional stroke timing testr -f MSIVs vere performed. During the first stroke attempt, two of the same LoIVs (IB21F022D, 1B21F028C) that previously stroked slovly, again failed to close within the required time. The valves vere subsequently recycled satisfactorily within minutes of their first tests. However, because the valves again failed to meet the required closure time on the first attempt, the basis for an isolated failure was no longer considered valid. RRC Region III was informed of the problem. The decision was made to shutdovn the plant and troubleshoot the problem. On November 4, an NRC Augmented Inspection Team (AIT) arrived onsite. A troubleshooting plan was established and implemented. The air actuators of the three valves which had exhibited slow closing times were disassembled ) and the inspection results documented. The conclusion drawn is that the 1 dual solenoids exhibited sluggish action after operating in localized high temperature conditions. It is felt that steam leaks caused a raised l temperature environment in the vicinity of the solenoids. The raised temperatures degraded the Ethylere Propylene Diene Monomer (EPDM) material causing the solenoid to stick or to sluggish. Corrective actions included disassembling all 8 MSIV's dual solenoids, and replaci.sg or rebuilding the solenoids as applicable, f 9 e m. ,2 .,e .,,.~,.,,,,.m,.

r l t 1 II. Chronology of Events On October 29, 1987 at 1837 Startup Test Instruction (STI)-B21-025A, "Main Steam Isolation Valve (MSIV) Functional Test" was being performed on 1B21-F022D, the steam line D inboard MSIV. This valve closed in 22.14 seconds. Technical Specification 3.4.7 requires the MSIVs to close in 2.5 to 5.0 seconds. At 2103 and 2106 the D inboard MSIV was cycled with closure times of 3.24 and 2.94 seconds, respectively. All other MSIVs were then cycled to verify closure times. The B outboard MSIV closed in 11.9 seconds and the D outboard MSIV closed in 77 seconds. Each was cyclef' again with satisfactory results. Since initial conditions causing MSIV u ov closure could not be repeated, all MSIVs were declared operable and plant startup testing continued. On November 3 at 1150, MSIV fast closure timing was commenced in preparation for the MSIV fast closure scram test in accordance with agreements made with the NRC on October 30. At 1157 the D inboard MSIV closed in 18 seconds and vas cycled again at 1159 with a closure time of 3.0 seconds. At 1208 the D outboard MSIV failed to close. A second attempt was satisfactory at 1213 vith a closure time of 3.4 seconds. The D inboard and outboard MSIVs vere declared inoperable and placed in the closed position in accordance with the requirements of Technical Specification 3.6.4.a. Based on repeat failures a plant shutdown commenced at 1330. The reactor was manually scramme' at 1819. On Noveinber 4, the Nuclear Regulatory Commission (NRC) issued a Confirmatory Action Letter (CAL) detailing various steps Perry management was to take and not to take in preparation for an NRC Augmented Inspection Team (AIT). The team arrived onsite November 4. III. Troubleshooting Activities Prior , performing any work in the field, a troubleshooting plan was written. Based on the symptoms shown on October 29 and November 3, it was felt that the component with the highest probability of causing the slov closures vts thc ASCO model number NP-8323A20E dual solenoid found on each MSIV air actuator. Numerous possibilities existed which could have somehov affected these solenoids. The troubleshooting plan was set up to determine what the root cause was and whether any secondary problems had an impact. On November 5 "As Fouad" conditions were documented and a more detailed troubleshooting plan was developed to establish the root cause of the MSIV failures and corrective actions necessary to restore the valves to operable condition. The troubleshooting plan was agreed to by the NRC AIT. On November 5 through November 8 various troubleshooting activities were i carried out.

l.. l The first MSIV investigated was the 1B21-F022D valve (inboard MSIV on "D" line). Solenoid voltages and solenoid air exhaust port samples were taken ss the valve was cycled all results were satisfactory. Next the field viring and air lines were disconnected from the air actuator or air pack. All connections and pipe openings were inspected and any disetepancies noted. The air pack was then removed from the valve actuator and taken to 1 the I&C hotshop for disassembly. The above steps were repeated for the 1B21-F028B outboard valve and then the i 1B21-F028D outboard valve. Any discrepancy no matter hov small was f documented for further evaluation. Whenever possible pictures vere taken of what was found. The major discrepancies appear to be the following: 1. All dual solenoids disassembled have impact marks on the star shaped disk subassembly and a deep depression (dimple) on the disc holder seal (EPDM), with the solenoids of the B21-F028D indicating the most degradation. 2. Many of the EPDM Body Assembly 0-Rings were hard, flatter.ed, and adhering to metal surfaces. 3. In the 1B21-F022D valve rust was found inside the solenoid valve body, and the B solenoid coil was badly corroded. In addition to the component disassembly three types of air analyses vere performed to determine what contribution, if any, instrument air quality may have had in the failure of the MSIV valves. Filter samples were collected to determine particulate matter present in the instrument air system at the solenoid and actuator supply points. Various unknown substances observed in or collected from internal component surfaces were analyzed using infrared j spectrophotometry to deduce origin of.naterials found. Grab samples of the air supply were analyzed by gas chromatography for hydrocarbon content and j quantification of organic contaminants if present in significant quantities. J The samples collected on filter paper for particulate vere analyzed under a microscope. Very small quantities of particles greater than 40 micron vere identified which indicates acceptable air system quality. Therefore, it is a very lov probability that the particles had an adverse effect upon the solenoid valve operation. Analyses of the substances collected during disassembly identified the presence of thread sealant and silicone lubricant, both of which are normally used during assembly of solenoid valves and air lines. Air supply grab samples indicated no hydrocarbons present in the instrument air supply. j Based on all the information it appears that the EPDM material used in various parts of the solenoid was interfering with solenoid valve movement. Thus, the decision was made to disassemble the dual solenoids on all 8 MSIVs, and refurbish as necessary. ,.,,,.,_..-m._

e 6 IV. Root Cause The cause of the HSIV delayed closures has been isolated to a failure of the ASCO dual solenoid valves. This failure is attributed to EPDM elastomer degradation due to elevated temperatures in the vicinity of the air packs resulting from steam leaks. The observed hardened dimples on the dise holder assembly and core assembly hardened elastomer seals is consistent with high temperature conditions. Other evidence of localized steam effects include degradation of the solenoid valve 0-rings and observed rust / moisture discoloration of the 1821-F022D solenoid coil. Localized high temperature conditions existed during the plant cycle due to steam leakage and elevated area temperature indications. Steam leakage is known to have occurred in MSIV IB21-F0228 packing and the MSIV leakage control system isolation valves. This leakage was in the direct vicinity of those MSIV's which exhibited slov closure. Steam in excess of 300 degrees F l is suspected of leaking in the direct location of the subject MSIV air packs based upon the degredation of the EPDM. V. Corrective Actions The following evaluations and actions have been or vill be completed prior to plant startup 1. For the dual (fast closure) solenoid the total air pack vill be replaced for the 1821-F028D valve, and the whole dual solenoid vill be replaced on the 1B21-F022D valve. No other solenoids shoved significant degradation or required replacement. All of the other MSIV dual solenoids have been rebuilt. 2. For the single (slow closure) solenoid the solenoid vill be replaced on the 1B21-F028D, since the whole air pack is being replaced. Based on the inspection results above, no other replacements were necessary. 3. A evaluation has been performed of other ASCO solenoid Class 1E harsh environment applications in the plant, including those which may have been subject to the steam leak environment which af fected the MSIV solenoids. The review identified two normally deenergized rolenoids which do not serve an active safety function. Vork history review of all other applications has shovn no solenoid failures. 4. An evaluation vill be made of other equipment in the v!cinity of the 1B21-F022D, 1B21-F028D, and 1B21-F028B valves, to atsess any impact that the steam leaks may have had on these comtecents. 5. Additional temporary temperature monitoring vill be installed in the steam tunnel on the preselected sample points in the HSIV area including the dual and test solenoid bodies. This monitoring vill be used to evaluate the actual temperature profile of the complete MSIV actuator assembly and the surrounding area. Following completion of the Startup Test Program, temporary ttaperature indication vill also be installed in the dryvell for monitoring of the inboard MSIVs. i-, w c- ,--,c.. ---.-,,,-.,-.c., 3

1. l l The following additional evaluations and actions vill be performed: 1. Further evaluation vill be performed on the exsisting industry experience and efforts on ASCO solenoid valve failure investigations. This evaluation vill include such areas as using different metal, and non-metal materials, and the effect of hydrocarbons. Possible design improvements, including an exhaust port screen vill be evaluated. Based on these evaluations a determination vill be made on future actions including replacement frequencies. l 2. A sampling plan for the solenoid elastomer components vill be f established. Analyses of these components.are expected to confirm i that hydrocarbons did not contribute to the EPDM degradation. Dev point and particulate sampling of the instrument air system vill continue at the existing test frequency. A preventive maintenance requirement vill be established for periodic re' placement of the instrument air system prefilters. The maintenance frequency vill be censistent with replacement of the instrument air system af ter filt ers. Additionally a generic precaution vill be added into air system vork orders regarding the use of thread lubricants and sealants. J An evaluation vill also be made of the relative physical location of the air compressors reduction gear vents, and the compressor air intake, to determine the need for modification, and/or periodic replacement of the intake filter. 3. Until the first refueling outage the full closure dual solenoids vill be checked for proper operation during the monthly slow closure check. This vill be performed by fully closing each MSIV individually utilizing the test solenoid, followed by taking the control switch to close, thus verifylt.g the proper operation of the dual solenoid. Also during this time frame the MSIVs vill be cycled individually on a quarterly basis regardless of plant j operating conditions, and the fast closure time verified. On an interval not to exceed six months an inspection vill be perforced on a dual solenoid during an outage of opportunity. This inspection vill verify no degradation of the solenoid valve internals.

l RCCT CAUSE ANALYSIS EXECUTIVE

SUMMARY

This document describes the evaluations performed te deter-mine the cause of events on October 29 and November 3,1987 when Perry Cnit 1 Main Steam Isolation Valves (MSIVs) failed to fast close on command. The most probable root cause, based on data currently available, i.e failure of an Automatic Switch company (ASco) Model 8323 3-way dual solenoid valve. The primary suspected causa is hardening and dimpling of the EPDM rubber disc seat material and other EPDM seals, causing the disc holder assembly to wedge in place when the solenoid was de-energized. Several mechanisms have been proposed that could lead to EPDM degradation, the most probable of which is a local high temperature environment. This document is organized into four sections. Section 1 describes the most probable root cause, and the basis for its selection as such. Section 2 gives an overview of how the root cause analysis team reached its conclusions. Section 3 describes potential component failure modes that could lead to MSIV failure to close, and finally, Section 4 describes specific failures within the ASco Model 8323 valve that could lead to the observed conditions, and discusses environmental conditions that could lead to the failure. 11/09/87 /

1-1 SECTION 1 MOST PROBABLE ROOT CAUSE The most probable root cause of the observed MSIV failure to close is failure of the Automatic signal Company (ASCO) Model 8323 3-way dual solenoid valve to shift from the energized to de-energized position. Within the component, the Ethylene Propylene Diene Monomer (EPDM) rubber disc seat material was found to be deformed. A "dimple" (see figure 1 and 2) was found in the EPDM seat material on the disc holder. This is also indicative of a general hardening and degradation of the rubber seals within the valve. If the disc holder sticks to the orifice the MSIV will not close. Delayed closure is consistent with de-energizing of the solenoid, followed by sticking of the disc holder to the orifice for some period of time, when the disc holder breaks loose and allows the air pressure to relieve through the orifice. Once the air pressure is relieved, the MSIV will close. Failure of this component is the only failure that is consistent with the observed f ailure. No other single component failure will result in a delayed MSIV closure. The EPDM degradation is most probably caused by exceeding the temperature limits of the EPDM material. EPDM was chosen for this application because of its radiation resistance from an equipment qualification standpoint. Perry has experienced bulk ~ drywell and steam tunnel temperatures which have approached tech spec limits during much of the startup test program. Additionally, steam leaks have occurred in the vicinity of the affected MSIV solenoids. While no data exists to actually confirm that the local temperatures have exceeded the capability of the EPDM rubber, a good correlation exists between the location of steam leaks and the affected valves. Several other mechanisms have been postulated for the EPDM degradation, and sufficient data does not currently exist to absolutely prove or disprove any hypothesis. It is true, howev-er, that the temperatures near the valves have been close to the maximum allowable for EPDM material, and this is the most likely cause.

l [,,~. ...,.. -, n., z..,;...

s. ;....

.:..~:.y~ ~,. ; ~ .*... ;. ~., p p~ m;:< ...ps =....;,...._.;y,,,,, - p.. ~.., 3. r. : . z.;. : n..:-t y . L.. s.. ..+..y .n ^s *.-- -... - - ~ - ..'l. ' <e +.';, . n. n: .~.. :. .., : w, r, y-1 .4 -.. a -: y .c

v. - :

t ..):,Q., 2 y ..~ ] z__\\,,*.; - f 1 [: 4 f_ . f.:.fi: : Q. } ? i y' .~ .,s L I m. _ 3, y' Y,, *.' -...? ~ ' l.. p *. :. '.;',.y~. ,__.[ 3, l} l

.

. q J.';,;;. M <

..g,N p,o__3:... K,:~.;:.3._ - z y g,_; ag ,7 y,.. y. -.m

y.

g -r .,x l 'f q.< ;'.-"~~[ b - -.e**'T ' e; ;,,$., .: ' 5.'..,,, - A. ' * ['. (E.,*, s ')o r ', ' c ;; ^I-k' ; s' ,1,.. -s-1"' f- ,J _. r ;-. :. :. ~ ' s. :: y;py., ;.:. ..,3,;. : y

v. ; ; - p... ;

~. p..:..;. :.:. :.. -. n.~;; :; -, c.- 3.; x ?... 3:s em l 9.:. .. z u. .. n s.... e.;:. x p s. n.. p#> y *. 1.,;*** M. j; p ). - f " hl 3 }@,.ary_ i,-- - w- ,., c. ~'. ;.. .' ' : u (.' ^.,.; n.<:.... i i l i i I &y; -.,'?

y,i., i,.'7,h

.cA ) g 4..; 'z.. ;- :- .e p . m;, d' m m-.m y w-- a ___ _, - = = - - "_$ ! i M lM l ) .4-. 2su"'"~' hi g y.. l! ^I Ou l h) -7 C..y.)4., W ~ a f +. m I O 1 T.s

n a

I

3.,. m.

i r .) .h7.R@y{.F3,! <. ~ T 0 d ~ .Tc. ,.h-. ..r. Wk':, E h) D.:p. i .c.- -a. m.,,.m N ~'t.W..m,*. s,.;f. v a.~ e c ,,y y 7.. - a.c. a. - ,s,. yr .,,.u \\/) or l..v. t t a .:.?.j,. ~.nc% .a A. ~C E~:2W#n .~ 4,W....&.;' 4,3. GtAL & pS$hV.:.t;f. bk u G tr ..:2' R , p; 2 V W, ~< m,m d & w.M w g'" Q . %, G.Y'%.m Y'. h .w V N 'v - $~n

>yp i%(

MX;.%.gsQ. U .t:~. M

m,

.,-g. M #fM N$ b i/ @Q....m .. :. :... a. :a % :V % w c4 .7 ~ O n% w.u.<

y. 2.x%w.e.n.::;.-n< g.

N Qen.2q@w-:.po.. %n@i&5%ik.n:c 3 4z2W:(^" .:.u. s.. n 1

. c

4. W Q

W2 M %;G.P '..u'. m.: "M n .u. W.r:.n..c:w:c w.. ~.ve,....<:. w:.. :.,..:. :.. a. n.2. u v. w... e2 .W"i. s.j.c.. ,...s... w....* W *' G Ci d Q. G..v %.* W.:.,,r A... ?. :.,. - - ......,. n.. u.. ys,.._. ...- s,.- .n. -2 .#..,.'.-*:.4.::,;;g.: s, >Q. W~.. . 3 ',1 < G:* :?. X '. f _*,9.C ':.. ';.:* '." \\,.. ' =:. .a s .-5.W,...

e

-N C - . ' *^ ~; .. b,. ~7. $:;,y?; n#*% 7 w f.M.,. ~.. &.Il &::. ~. p.*W:'. F. -..:: 3l. " ', 3. ..c- - r...

q...

v.,;:-:.,,s. y.;,p..: .h. .C y:. '... ' ).:Hc..... .e 4.,t: .;.-. / p %. -9 y, i :.': :.J;;.'.,'l; M..,\\,;;..:'.:,* ~:;A .:. :.<.',.y .n. ".;. t ~..,a ,,.h... %.;. t.x;9- .e ~, ' 2. (. : ~.~~* *: ,.r. a %.... ,yy . ?! ?.h .l. Fh:3. 6 L.;b. ' '\\' ? '. '.. : ';?k.:; '~; hl *aN' ~i' ..' &.,s."T.;f(W 'y '... G.. ..../... V-s.;.p,.. h.%-)4, S.;er - * < t .. '. -c... s ; t.1.9 ' .,4...#. :?,i.. p, :: /... t -. . i ss 7 ,c,t. a' c.,. - *, .A. .a . q.p v.,. u..,. v .. v. M.q%- ;[.rgt.d,: p.. .',.3:I. @2.-l.7 ih-49." h,.'.'. ], *. ~3.4.: g::. +.g.s l. %qe.. ;.;4. c.".. .1 .i.,.,' . p. ' Tar h.l.: LL.~.::.M./. ll. ~ i.;4 - g -- 6._ 2' e ~..

'r s g LF 4 g 4t+ 9;ks 4 g.9

1. $ f g,.js//,pf p)sc H0ty gg

,,,,.. e Q SK ETCH SJtoaGN& DISC jfftcgg ggggyp(,ypgpggp(g. gAa.s+ h> cyI,'nele< fefed ) y p3g pl5C I' #A%W HO' g w 1 9 3# psic ~ .~" JKK.TCH SRouDNf, cross. 'SEcrio/v oP D6C lM a i.5 sERTEb /0.5JTION h t.' f 4

v c I i 2-1 SECTION 2 ANALYSIS TECHNIQUES AND OVERVIEW ~ Following the f ailure of the B21-F022"B" and "D" Main Steam Isolation Valves, a multi-discipline team was convened with the charter to determine the most likely cause of the problem. This activity would be useful prior to actuator disassembly and inspection. The team consisted of senior engineers from the CEI mechanical and electrical engineering and technical departments, as well as the architect engineer (Gilbert) and NSSS supplier (General Electric). Problems analysis proceeded using standard Kepner-Tregoe i (KT) Problem Analysis techniques. The initial thrust of the team was to determine which equipment failures would cause the failure of a MSIV to close in the delayed manner observed. An-initial brainstorming session was held to determine potential component failure which might cause the observed behavior. These potential failures were then compared with known f acts and design condi-

tiens, using "is/is-not" techniques to rate the postulated failures as to probability.

Twenty four (24) potential component failures were initially postulated. Of these, 19 were rated as unlikely, one (1) as potential, and four (4) as probable causes. All five of the potential and highly likely candidates involved either the ASCO Model 8323 3-way Dual Solenoid Valve, or the air supply to these components. Specific work items and inspection steps were thus incorporated in other site action plans to address these components in detail. Section 3 of this report documents each of the 24 postulated component failures. It is organized in order of highest to lowest probability. Each potential cause is described, discussed and conclusions drawn with regard to root component failure. \\ 11/9/87 -r-n -e -.e- ---r,,...n- - - -. +,-- e --,p -w.- --ap-~e--

---

n-w w-*

w

g - l i 2-2 Following disassembly of the actuator air packs and diagnos-tic tests on the air supply system, it was determined that the most likely failure mode was, in f act, the Asco Model 8323 3-way dual solenoid valve. The suspected cause was dimpling of the EPDM rubber disc seat material, causing the disc holder assembly to wedge in place when the solenoid was de-energized. The team was again convened, this time to evaluate the environmental and design conditions which could be responsible for the observed component failure. Analysis techniques similar to those utilized in the compo-nent evaluation were used to screen the potential causes. Absolute determination of the root cause is difficult.

However, the most likely condition leading to the f ailure was local high temperatures leading to EPDM degradation.

Analysis results are given in Section 4, again describing each of the nine (9) postulated rect cause conditions and discussion of the.1dence to confirm or deny the postulated condition as root cause. O 11/9/87 l

t SECTION 3 COMPONENT FAILURE DESCRIPTIONS e t 11/9/87

g 4 + I 3-1 Potential CAgn Failure of the Part #4 ASCO Model 8323 3-way Dual Solenoid Valve Discussion Failure of the ASCO Model 8323 3-way dual solenoid valve to chift from the energized to de-energized position could cause the delayed closure event experienced by Perry. This failure mode has happened in the past due to various reasons as evidenced by IE Notices 85-17 and 86-57, (copies attached) and INPO SER 57-85. Conclusion This failure mode is the most likely candidate for root component failure of the problem. The post-disassembly q inspection has found dimpling of the EPDM rubber disc seat material. This could cause the disc holder assembly to wedge in place when the solenoid is de-energized. This would in turn not allow air pressure to relieve through the

1. 3 air port, and preclude MSIV closure.

i e 11/9/87 4

o / s SSINS No.: 6835 IN 85-17 UNITED STATES NUCLEAR REGULATORY COPHISSION RECEIVED OFFICE OF INSPECTION AND ENFORCEMENT WASHINGTON, DC 20555 MAR 121985 March 1, 1985 R & DAS IE INFORMATION NOTICE NO. 85-17: POSSIBLE STICKING OF ASCO SOLEN 0ID VALVES Addressees: All nuclear power reactor facilities holding an operating license (OL) or construction permit (CP). Puroese: This notice is provided to inform recipients of a potential problem with ASCO + solenoid valver. that may prevent the main steam isolation valves (MSIVs) from closing on BWRs and may prevent other safety functions on BWRs or other types of plants. It is expected that recipients will review the information for applicability to their facilities and consider actions, if appropriate, to preclude a similar problem occurring at their facilities. However, suggestions contained in this information notice do not constitute NRC requirements; therefore, no specific action or written response is required. Descriotion of Circumstances: On February 10, 1985, at Grand Gulf Unit 1, one inboard and two outboard MSIVs - failed to remain shut following manual closure. The problem is now attributed to failure of the ASCO solenoid valves used to provide fast closure of the MSIVs in the event of an accident. At the time of the event, the plant was using a slow-closure procedure to close the MSIVs prior to a drywell entry following a reactor scram. The procedure calls for one set of solenoid valves to slowly close the MSIVs and another set of solenoid valves, which are the fast closure solenoid vsives used in an accident, to hold the MSIVs closed. Evaluations and tests performed subsequent to the event have attributed the problem to the second set of j solenoid valves (type HTX 8323-20V). Each of these solenoid valves contains two solenoids, one solenoid rated at 16 watts and the other solenoid rated at 6 watts. Preliminary test results, to date, have reproduced spurious sticking of the solenoid actuator of the solenoid valve when the solenoids are subjected to temperatures of about 180*F. However, these tests have not been definitive. 8503o10448 4 e ,,,,,,_w

e / IN 85-17 March 1, 1985 Page 2 of 2 Discussion: Although the NRC is not aware of all applications of this type of dual solenoid valve in all types of plants, a list provided by General Electric of BWRs that are. believed to use this valve for closing the MSIVs includes: Brunswick 1 & 2 Clinton Cooper Ouane Arnold Fitzpatrick Grand Gulf 1 & 2 Hatch 2 LaSalle 1 & 2 Shorehar No specific action or written response is required by this information notice. If you need additional information about this matter, please contact the Regional Acministrator of the appropriate NRC reg'onal office or the technical contact listed below. / owa Jordan, Director Div en of Emergency Preparedness a Engineering Response Of fice of Inspection and Enforcement Technical

Contact:

Eric W. Weiss, IE (301) 492-9005 Attachment-List of Recently Issued IE Information Notices 4 w ,-----.----ms

e 4 IN 85-17 March 1, 1985 LIST OF RECENTLY ISSUED IE INFORMATION NOTICES Information Date of Notice No. Subject Issue Issued to 85-16 Time / Current Trip Curve 2/27/85 All power reactor Discrepancy Of ITE/Siemens-facilities holding Allis Molded Case Circuit an OL or CP Breaker 85-15 Nonconforming Structural 2/22/85 All power reactor Steel For Safety-Related facilities holding Use an OL or CP 85-14 Failure Of A Heavy Control 2/22/85 All power reactor Rod (84C) Orive Assembly facilities holding To Insert On A Trip Signal an OL or CP 85-13 Consequences Of Using 2/21/85 All BWR and PWR Soluble Dams facilities nolding an OL or CP 85-12 Recent Fuel Handling Events 2/11/85 All power reactor facilities holding an OL or CP 85-11 Licensee Programs For 2/11/85 All power reactor Inspection Of Electrical facilities holding Raceway And Cable Installation a CP 1 85-10 Posttensioned Containment 2/6/85 All power reactor Tendon Anchor Head Failure facilities holding an OL or CP 85-09 Isolation Transfer Switches 1/31/85 All power reactor And Post-Fire Shutdown facilities holding Capability an OL or CP 85-08 Industry Experience On 1/30/85 All power reactor j Certain Materials Used In facilities holding Safety-Related Equipment an OL or CP 85-07 Contaminated Radiography 1/29/85 All NRC licensees Source Shipments authorized to possess industrial radiography sources OL = Operating License CP = Construction Permit f i e . ~.-.

l y SSINS No.: 6835 C IN 85-17, Supplement 1 UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT ....lV E D nL*' WASHINGTON, D.C. 20555 CCT 0 3 G85 October 1, 1985 R & DAS IE INFORMATION NOTICE NO. 85-17, SUPPLEMENT 1: POSSIBLE STICKING OF ASCO SOLEN 0IO VALVES Addressees: All nuclear power reactor facilities holding an operating license (OL) or a construction permit (CP). purpose: This notice is to inform recipients of the results of followup investigations regarding the reasons for sticking of Automatic Switch Company (ASCO) solenoid valves used to shut main steam isolation valves (MSIVs) under accident conditions. Recipients are expected to review the information for applicability to their facilities and consider actions, if appropriate, to preclude similar problems occurring at their facilities. However, suggestions contained in this information notice do not constitute NRC requirements; therefore, no specific action or written response is required. Description of Circumstances: Information Notice No. 85-17, "Possible Sticking of ASCO Solenoid Valves," described the problem that occurred with ASCO Model HTX 832320V solenoid valves at Grand Gulf Unit 1. General Electric (GE) and ASCO conducted tests to determine the cause and any further corrective actions. The following is a summary of the GE and ASCO tests and analyses: Two of the three solenoid valves which failed at the Grand Gulf Nuclear Station also sporadically failed to transfer during testing at' elevated temperatures. These two valves were the only valves that failed during these tests. However, these failures were not predictable. Subsequently, five valves from Grand Gulf service were disassembled and inspected. This inspection identified a microscopic foreign substance on the lower core / plug nut interfaces on all five valves. Further evaluations of this microscopic substance were inconclusive because of the small foreign substance sample size. After cleaning and reassembly of these valves, tests were conducted on four of these five valves at elevated temperatures. These four valves functioned normally. ASCO felt certain that the valve failures resulted from high-temperature sticking of the lower core-to plug nut faces resulting from a foreign substance or combination of substances collected at this interface. 8509260369

D r IN 85-17, Supplement 1 October 1, 1985 Page 2 of 2 A detailed dimensional analysis and comparison among the valves returned from Grand Gulf indicated that all parts were within allowable limits and differences *were not enough to cause a failure to shift. Therefore, this examination tended to relieve concerns related to a generic design defect. GE has attempted to locate additional foreign substance from other valves of the same type in use at Grand Gulf to determine how the foreign substance got in the valve, or where it originated. GE was able to scrape some small amounts of foreign substance from the lower core-to plug nut interface. However, there was not enough residue to make a definitive identification of the nature of the foreign substance. GE has recommended that the licensee replace the potentially contaminated MSIV solenoid valves and institute a periodic examination and cleaning of the MSIV solenoid valves. Grand Gulf has replaced the eight MSIV HTX832320V dual solenoid valves with fully environmentally qualified ASCO Model NP 8323A20E dual solenoid valves. The environmentally qualified valve Model NP 8323A20E was included in a control sample placed in the test ovens with the solenoid valves that stuck at Grand Gulf. The environmentally qualified model did not stick under the test conditions that cause sticking in the other solenoid valves. No specific action or written response is ' required by this information nutice. If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office. bian, Director awarc Divisi of Emergency Preparedness and gineering Response Office of Inspection and Enforcement Technical

Contact:

Eric Weiss, IE (301) 492-9005

Attachment:

List of Recently Issued IE Information Notices

) IN 85-17 Supplement 1 October 1, 1985 LIST OF RECENTLY ISSUE 0 IE INFORMATION NOTICES l 1 Information Date of -Notice No. Sub.iect Issue Issued to j 85-79 Inadequate Communications 9/30/85 All power reactor Between Maintenance, facilities holding Operations, And Security an OL or CP; research l Personnel and nonpower reactor facilities; fuel fabrication and processing facilities 85-78 Event Notification 9/23/85 All power reactor facilities holding an OL or CP 85-77 Possible Loss Of Emergency 9/20/85 All power reactor Notification System Due To facilities holding Loss Of AC Power an OL or CP 85-76 Recent Water Hammer Events 9/19/85 All power reactor facilities holding an OL or CP 85-75 Improperly Installed Instru-8/30/85 All power reactor mentation, Inadequate Quality facilities holding Control And Inadequate Post-an OL or CP modification Testing ^ l 85-74 Station Battery Problems 8/29/85 All power reactor facilities holding an OL or CP 84-70 Reliance On Water Level 8/26/85 All power reactor ) Sup. 1 Instrumentation With A facilities holding i Common Reference Leg an OL or CP 85-73 Emergency Diesel Generator 8/23/85 All power reactor Control Circuit Logic Design facilities holding an OL or CP Error 85-72 Uncontrolled Leakage Of 8/22/85 All power reactor Reactor Coolant Outside facilities holding Containment an OL or CP i l I OL = Operating License CP = Construction Permit j I I

g s SSINS No.: 6835 IN 86-57 UNITED STATES NUCLEAR REGULATORY COM4ISSION R E C E,W E,0 0FFICE OF INSPECTION AND ENFORCEMENT JUL 18'lW0 WASHINGTON, D.C. 20555 R & DAS July 11, 1986 j IE INFORMATION NOTICE No. 86-57: GPERATING PROBLEMS WITH SOLEN 0ID OPERATED VALVES AT NUCLEAR POWER PLANTS Addressees: All nuclear power reactor facilities holding an operating license or a construction permit. i Purcose: l This notice is to advise recipients of a series of valve failures that have occurred recently at several nuclear power plants. It is expected that recipi-ents will review the events discussed below for applicability to their facili-ties and consider actions, if appropriate, to preclude similar valve failures 1 occurring at their facilities. However, suggestions contained in this notice l do not constitute NRC requirements; therefore, no specific action or written response is required. Description of Circumstances: The NRC has received reports from licensees of operating nuclear power plants i involving failures of certain valves that are actuated by solenoid operated valves (SOVs) to operate properly. These failures have adversely affected the f intended functions of tne sain steam isolation system, pressure relief and t / fluid control systems. Attachment 1 to this information notice describes the failure events and the corrective

• actions taken, i

i Discussion: In most of the cases described in Attachment 1, the cause for triggering the event was attributed to a malfunctioning SOV that served as a pilot valve. This in turn resulted in the malfunction of the associated main valve. The failures of the SOVs can be traced to the following different causes: (1)potentially I high-temperature ambient conditions are not being continuously moni.tored in areas where SOVs are installed and operating in an energ(ized state, (2) hydrocarbo contaminants, probably because backup air systems e.g., plant service or shop l air systems) are being used periodically and are not designed to "oil-free" 4 g specifications as required for Class IE service (3) chloride contaminants 4 causing open circuits in coils of the SOVs, possibly as a result of questionable handling, packaging, and storage procedures, (4) an active replacement parts program associated with the elastomers and other short-lived subcomponents used in SOVs has not been adequately maintained, and (5) lubricants have been used excessively during maintenance. ASCO provides installation and maintenance t I aso70so425 j i

-} 'E IN 86-57 July 11, 1986 Page 2 of 2 ..i 9 sheets with all its valves and rebuild kits. For additional information ASCO should be contacted. Because of the recurring 50V failures discussed above, NRC's evaluation of the l problem is continuing. Depending on the results of the evaluation, specific actions may be requested. 1 No specific action or written response is required by this information notice. If you have any questions about this matter, please contact the Regional Administrator of the appropriate regional office or this office. 0 e ft v dward L. ordan, Director 4 Division Eserger.cy Preparedness j and Eng neering Response Office of Inspection and Enforcement j Technical Contacts: Vincent D. Thomas, IE 1 (301)492-4755 4 b George A. Schnebli, Region II (404)331-487.5 Attachments: e, 1. Examples of Solenoid-Operated Valve Failures at Operating Nuclear Power Plants d 2. List of Recently Issued IE Information Notices 3-1, d t L k c

1-o / e IN 86-57 l July 11, 1985 Page 1 of 4 EXAMPLES OF SOLEN 0ID-OPERATED VALVE FAILURES i AT OPERATING NUCLEAR POWER PLANTS Brunswick Station 1. Main Steam Isolation Valve (MSIV) Solenoid Failures On September 27, 1985 at Brunswick Unit 2, during the performance of a. periodic test to demonstrate operability of the MSIVs, three out of eight isolation valves failed to fast close as designed. The fast-close test was required before returning Unit 2 to full power operation after the plant had been placed in cold shutdown on September 26, 1985. Two of the l three affected valves were installed as inboard and outboard MSIVs in the same main steam line, wnich would be a significant safety problea in the event of a failure of that steam line. The licensee's initial investigation isolated the cause for the MSIV i t failures to the dual SOVs that serve as pilot valves that supp y operating air to the MSIV operators to open or close the MSIVs. The fau ty SOVs were identified as Automatic Switch Company (ASCO) Model NPL8323A36E. A j more detailed review of the problems determined that the causes for failure were attributed to valve disc-to-seat stickin of the 50V and portions of the elastomer dit: material plugging the V exhaust port. These failures prevented closing the associated MSIV. Ethylene propylene f. (EP) was the elastomer substance used for seals and valve disc saterial in this model 50V, i ) The licensee's failure analysis of the SOVs included technical assessments of the problems from the valve manufacturer (ASCO), the supplier of the EP material (Minnesota Rubber),.ind Carolina Power and Light's (CP&L's) research center (Harris Energy and Environmental Center, Raleigh, North Carolina). The findings resulting from this joint effort indicated that the SOV failures could heve been caused by a combination of hydrocarbon j contamination of the air systes ard high ambient temperaturt conditions' ) i causing degradation of the EP valve seating and seal material. i The ASCO Model NPL8323A36E SOVs are installed in Brunswick Unit 1 in " June 1983 and in Unit 2 in August 1984 to meet the Environmental Qualifi-cation (EQ) Program requirements. The Unit 1 50Vs were subsequently replaced during the 1985 outage when modifications were being made to the MSIVs. The new SOVs (NP8323A36V) were identical to the old ones except the valves contained Viton seats and seal materials in lieu of EP. i Additionally, the information provided from ASCO shows the following: Ethylene propylene is resistant to higher levels of radiation a. J i (200 megarads) than Viton. However, EP absorbs h cause swelling and loss of mechanical properties.ydrocarbons that can It is unsuitable in applications where the air system is not designed to "oil-free" 1 specifications, i --,,,,--r yy,.,,--, ,--v w. _n -,---,n-, ,,w,c-y-o n g - w r e rw, -,,, ,_-m,,9,-c

T IN 86-57 July 11, 1986, Page 2 of 4 b. Viton has superior high-temperature performance when compared to EP and is impervious to hydrocarbons. Its' major disadvantage is that it is less resistant to radiation than EP by a factor of ten. ASCO j recommends Viton for applications that are not oil-free and where radiation levels do not exceed 20 megarads. '<i On the basis of a licensee review of the Brunswick Station maintenance l history, which showed the performance of Viton to be satisfactoiy in ASC,0 valves, and the available literature and industry experience,'the licensee ' replaced all Unit 2 dual solenoid valves with valves having Viton seats and seals. Because Viton has a 20-megarad limit, the licensee plans to l replace these elastomers every 3.3 years to meet environmental qualifica-tion requirements for the MSIV application. i After replacing the M ty valves with valves having Viton disc and seal j material, the licanu experienced several SOV failures resulting from open circuits of coils on Unit 2. (Brunswick Station employs ASCO NP8323A36V vah 4t use one ac coil and one de coil in applications using the subje;.... solenoid valve.) On October 5,1985, the de coils of two MSIVs failed during the perfor-mance of post-maintenance testing of the MSIVs. Investigation into the failures indicated an open circuit in the de coils. The coils were replaced and the valves subsequently ratested satisfactorily. On October 13, 1965, an unplanned closure of an MSIV occurred while Unit 2, was operating at 99 percent full power. 01osure of the MSIV occurred when the ac solenoid coil portion of the HSIV associated 50V was de-energiz'ed in accordance with a periodic test procedure. It was not known then that there a s an open circuit in the associated de solenoid coil portion of the M SOV. Consequently, when the ac coil was de energized, closure of i the W.IV resulted. The failed de coil was replaced and then retested satufactorily. Investigation into the failures of the de coil by the licensee determined that the failures appeared to be separation of the very fine coil wire at the junction point where it connects to the ruch larger field lead. This connection point is a soldered connection that is then taped and lacquered. I l' Further analysis of the coils (two failed de coils plus five spares from storac2) by the CP&L Research Center indicated the separation sight be corrosion induced by chloride contaminants. To date, the licensee and ASCO are unable to determine the source of the chloride. However, ,., followup investigation by the NRC revealed that lC0 had previously experienced similar de coil open circuit anomalio after a surface ship-ment of SOVs overseas to Japan. At that time, ASCO believed that the salt water ambient conditions during shipping may have been the source of the chlorine-induced failures. 'ASCO recomends specific handling, packaging, and storage conditions for spare parts and i,alves at facilities. ~-____,__.__. _ _.._._

,I IN 86-57 July 11, 1986. Page'3,of 4 The licensee initiated a temporary surveillance program to monitor opera-bility of the solenoid coils on October 16, 1985. A modification was performed to install a voltage dropping resister in the individual coil circuits so that they can be monitored directly from cabinets in the control room. This allows continuity of the coil circuitry to be verified by measuring a voltage drop across the resistor. According to the lic:nsee, until the cause for failure can be determined, plans are to check the coil circuitry for continuity on a daily basis. 2. Scram Discharge Solenoid Valve Failure In November 1985, Carolina Power and Light's Brunswick facility experi-enced problems with several scram discharge SOVs. The problems were identified during periodic surveillance testing to determine the single rod insertion times and resulted in several reds with slow insertion times. Initial troubleshooti,ng isolated the problem to the SOVs in the scram discharge line for two of the control rods, which were subsequently replaced and tested satisfactorily. The licensee disassembled the failed SOVs, which were manufactured by ASCO (Model HV-90-405-2A), for failure analysis. Eien the valves were disas-sembled, it was noted that cepicus amounts of silicone lubricant had been applied by the licensee to all gaskets, seals, and diaphragms internal to the valves during previous routine maintenance. The licensee believes that the excessive amount of lubricant may have blocked some of the t . valves' internal passages or caused sticking of the diaphrages, thereby contributing ta the slow insertion times. The technical manual for the i subject valves states that body passage gasket.s should be lubricated with moderate amounts of Dow Corning's Valve Seal Silicone Lubricant or an i equivalent high grade silicone gnase. The licensee conducted successful scram tests on all other rods. A periodic retest of 10 percent of the control rods every 120 days as required by the Technical Specifications provides sufficient assurance that this problem does not exist in other SOVs. In addition, the licensee stated that maintenance proceduras and practices would be revi'ewed and modified, as required, to prevent the apolication of excessive amounts of lubricant during repair or overhaul of components. Haddam Neck Nuclear Power Plant On September 10, 1985, the Haddam Neck Nuclear Power Plant was operating at 100 percent power when one of che six SOVs in the auxiliary feedwater system ( AFW) failed to changa state when de energized. This failure was detected during the performance of a preventive maintenar,:.e procedure developed to periodically cycle each of the si) SOVs to prevent a sticking problem similar to 50V fail-ures previously expe rienced on Novembir 2,1984. In that earlier event. two feedwater bypass valves failed to open automatically and the cause was deter-mined to be sticking SOVs. The faulty SOY was ASCO Medel NP8320A-185E ind tha licensee has bean unable to determine the cause of the malfunction. The

v IN 86-57 July 11, 1986 Page 4 of 4 licensee's plans are to periodically cycle the SOVs until they are either replaced with an upgraded model or the specific cause of the existing sticking problem is determined and corrected. Millstone Nuclear Power Station, Unit 1 On December 24, 1985, while performing' a control rod scram time test at Mill-stone Unit 1, three centrol rods failed to insert during the performance of j single rod scram time t:32.ing. In all cases, the control rod was immediately. inserted and electrically disabled. Investigation into the failures revealed that in the first case the cause for failure of one sticking SOV was attributed to deterioration of the BUNA-N valve disc material within the valve. According to the licensee, this type of failure had been identified by General Electric in tneir Service Infctmation Letter No. 128, Revision 1, dated March 2, 1984. The licensee's investigation of the other two control rod drop failures failed to reveal the causes for failure other than a misalignment problem of one 50V's internals, which prevented proper movement. However, in each case, the SOVs were disassembled, overhauled, ratested satisfactorily, and returned to service. Grand Gul' Nuclear Station, Unit 1 Another failure of sticking SOVs occurred at Grand Gulf Unit 1 on February 10, 1985, and was the subject of IE Information Notice No. 85-17, entitled "Possi-ble Sticking of ASCO Solenoid Valves." h -.,,-w--,,..-- .---.v.-,,,

t 4 5 3-2 ) Potential Causq Instrument Air System Quality (oils, moisture, particulates) Discussion This potential cause has been experienced at other plants. This is evidenced by IE Information Notices No. 86-57 and 85-17. In the likelihood that poor instrument air quality, such as the presence of moisture, particulates, and/or oils, the possibility of failure related to several Main Steam Isola-tion Valve components would be highly likely. The main concerns would resolve around the Automatic Switch Company (ASCO) solenoid valves. Since the sesl and discs internal to these valves are Ethylene propylene, any intrusion of oil j into the instrument air system could cause degradation. Degradation of the seals and discs would, in this case, be caused by hydrocarbori contamination that would distort them and could result in malfunction of the valves.

However, at Perry this is unlikely because of the "oil free air" compressors.

Disassembly and inspection of the ASCO NP8323-20E dual solenoid valve from MSIV F022D did not reveal any hydrocarbon substance which could have been borne from the instrument air (as described below). A visual inspection of the EPDM parts of the ASCO solenoid valves was conducted. This inspection indicated that the EPDM disc was hard and brittle versus a new EPDM disc which is pliable and resilient. In addition, the discs were handled with white cotton gloves, and no residual was left on the white cotton gloves. The surface of the disc also did not appear to be sticky or tacky while it was being handled with the gloves. This is important since any rcsidual would be an indication of the EPDM breaking down due to hydrocarbon contamination. The possible intrusion of water or moisture into the air system could cause residue to form on the ASCO valva inter-nals and cause malfunction of the valves over a period of time. The moisture may collect during outage periods and form residue during plant operation when the ambient temperatures are higher. Dewpoint measurements were performed for the supply air to both the inboard and outboard MSIVs. Measured dewpoints were minus 55 F or lower 0 indicating that intrusion of moisture into the air system is not a concern. Grab samples frca the instrument air supply to conte.inment were analyzed for hydrocarbons using gas chromatography. Neither sample analyzed revealed detectable condensable hydrocarbons greater than 0.1 ppm. 11/9/87 t 1

1 0 3-2 Twelve particulate air samples were also obtained. The results from all the samples have not been completed. The results that are available show a trend of very low total particle counts with relatively few particles above 40 microns. Results from past air analysis have shown numerous i counts for particles below 40 micron with relatively few indications of particles greater than 40 micron. The disassembly and inspection of the ASCO NP8323-20E dual solenoid valve revealed no traces of moisture or particulat s contamination. There was no wear on either the EPDM or the metal components of the nolenoid valve. This result, in addition to the very low number of total particles in the air ".ystem and the low dewpoint temperatures would indicate that the root cause is not associated with the instrument air quality. The concerns addressed above also apply to the C.A. Norgren Shuttle Valves; however, the shuttle valves are much more tolerant to poor instrument air quality. Conclusion The air samples taken do not reflect a problem with hydro-carbon contamination. The presence of this type of interac-tion between the valve materials und hydrocarbon contamina-tion would be seen as a swell of the material. This is not the case where the material has been found to be embrittled. The investigation as the cause of failure will be pursued with conversations to be held with Susquehanna, Brunswick ' and Riverbend. Possible causes of the failure could be related to elevated temperature of the valves due to steam leaks in the vicinity. There will be further investigation to determine the root cause of the failure which could 9 involve destructive tasting of the components. The air sample counting will be completed and particles greater than 40 micron will be quantified. Our past experience with higher distribution of total particles indicates that the failure of the component was not attributed to the particle size or quantity since our total results nad been low. ASCO l has determined that particles less than 50 micron are acceptable for reliable operation of their valves. We plan to continue our inves-igation as to the root cause through analysis of the EPCM components. With the technical information we will obtain from the plants mentioned above, our plans are to formulate a testing plan that addresses both embrittlement and hydrocarbon contamination as the failure mechanics. The existing data obtained will allow us to envelop and quantify our failure analysis. 11/9/87

o 6 3-3 1 Potential Cause Obstructions / Foreign Materials in Air Lines / Accumulators Discussion This potential cause has been experienced r" other plants as evidenced by IE Information Notice 86-57 <nd 85-17. Ob-1 structions/ Foreign Materials in the air lines / accumulators is a likely cause since it would permit valve failures as experienced. The obstructions may permit periodic operation of the valves and depending on the instrument air cycling l could temporarily become dislodged. This could result in the same characteristics discussed in the write-up on "Poor Air Quality". Conclusion This item was initially considered to have a high potential as root component failure. Inspections of the air lines and accumulators found no defect that could cause the observed operational pattern, however, so this potential cause is unlikely to be a root component failure. t 11/9/87

t e m 0005 I N 3-swww I o N g p N OOCh = .g o a .e

o..

o. l = =g<s a m-e: =ga,. r-a <= -oo = _, = < ro zo* <<~> I ..e = u i b w u / ( e u 2-m .N .E w L 5 ( ::, -- LL.J J. V J_. ? m$ )c.- 2 .e..

n. ~g;._

\\ es. Eb aJ 2 l y v i

• C Q

jij F 3 8,. c i I 94 f l 2 < \\ \\ = N ( e l l .=. . =_ l_._ .sa c c

e. :. c

) p2= p.,. <=.} 3 es t .) I = ,a. e- .a. sv u _e

i l l DRAWING UNE CONFIGURATION AS OF REV. TAR.TiAL ~b4 *o Tz oS - o is dect z-g g NUCLEAR St NOTES.- l

1. FOR GENERAL NOTES, SEE ONG. D-209-001, SHEETS 1 AND 2.

1

2. IERMINAL BOX LAYOUT, SEE DWG. D-209-001, SHEET 1.

4 ~

3. FOR PHYSICAL LAYOUT AND WIRING INFORMATION, SEE ATWOOD L MORRILL COMPANY DWG.13560-01-H SHEETS 1 THRU 4.

(FLD40-003, SHEETS 1 THRU 4) OR 2 G.E. 0WG. 10505228. DATE ~

4. WIRED BY VENDOR.

5. JUMPERS FIELD TO. INSTALL. COLOR PER FIELD CABLE Div'lSION.

TI-

6. FOR CONAX C.O N DUlT SEAL ASST:inoty WIRING n Wr-n-209-001_ Sif EETS 7J,9 i 10.

RERE gt9_,3F_F. 7. BISCO TYPE LOC A SEAL IN STALLEb IN CON bulT A55 EMfblY FOR ELE' ( EACH 50LEN0lb To SERVE A5 AN ENVIRON MENTAL MOISTU RE (SEAL C. D C P B S-O (oI8 3. j i l$ 0b $ k? S$NSN$W b. i .o m ,.,,-__________-_____~-__.<,,,____nw

1 Bal.Tocrzz. A Sotecom uM._0C5 NOT E b/ N U l't O Sol E/VO/D I WlRE COL 0R T4,tnAt'h

• MARK CODE c> zoS. o rs skeeA e_.

i 7 b i B2 t H3G i3 A G i 2 621 H 3G I S A 5 l 26 Soc ENo/o# 2 N O,T E 4 ~ (NOTE 7 A N ts WIRE COLOR i h MARK CODE v i BZl H 3613 A 4 l 2 B21 H 3G il A 3 2 Q NOTE 4 SOL ENO/D 3 /su f ( NOTE l WIRE C O LO R \\g J MARK CODE L i e2iH3eo3A 2 2 621H SG o t A zu

s 0 3-4 Potential Cause one or both of the pilot solenoid valves for each of the MSIVs failed to decouple (mechanically separate) upon de-energization. Discussion Electrical control circuits identify positive de-energization of the respective pilot solenoids. This is verified via the indicating light and any meters as shown per elementary diagrams per B-208-013 H0ll and H036. The testing sequence and visual verification has identified that-the solenoids have been de-energized, although the MSIVs failed to open or delayed opening. If either solenoid fails to decouple, the MSIVs will not operate. No method exists to remotely determine whether one or both of the solenoids for a particular valve failed to decouple. The mis-operation (erratic) closure or deferred closure may possibly be attributed to this occurrence. As such it may be a highly susceptible cause. Further evaluation identi-fled that each of the pilot solenoids were sealed with Bisco Locaseal at the conduit entry point. This design change implemented per DCP 850618 is the only change initiated recently. The degradation and/or migration of foreign matter could also be a cause to prevent decoupling of the. solenoids. conclusion This item was initially classified as a high potential, and condition of the Bisco Locasaal was evaluated upon solenoid disassembly. Since no interference with the valve operation vas noted, this cause has been eliminated from considera-tion. 11/9/87 g.~, ,yr-e--*~m----? v-*-- = = - *

• • • -t-

'-+-t-*-r* e--vw--v------ww-w-v-v-'*,-+-ev '*-"+v----w---r-r

0 3-5 Potential cause Solenoid valve exhaust port blocked. j Discussion 1 1 Blockage of the exhaust port could occur through internal or external contamination. The port is open to the ambient. Particles may fall below the disc preventing shifting of the solenoid valve from its normally energized to normally de-energized position. Subsequent actuation could blow the blockage out of the valve allowing normal operation thereaf-ter. This is considered a potential cause for the Perry delayed MSIV closure experience. i Conclusion This was initially considered to be a potential cause for the Perry delayed MSIV closure experience. Inspection for blockage was performed, and on one solenoid a piece of tape was discovered to be blocking one port. Subsequent testing determined that this blockage was insufficient to preclude MSIV actuation. i 11/9/87

e 3-6 Potential Cause Failure of the Part f3 Norgren Model B0004A 2-way shuttle valve. Discussion The 2-way shuttle valve works in conjunction with the Part il 4-way shuttle valve to open and close the MSIV. The 4-vay shuttle valve provides the primary.*e.ogic for pressur-ization and venting of the actuator cylinder. The potential failure mode description is the same as that for the 4-way shuttle valve operation. j The 2-way shuttle valve cannot by itself open or maintain the actuator in the cpen position unless the 4-vay valve is energized or stuck in the energized position. onclusion The delayed closure event experienced at Perry is unlikely to have been caused by the 2-way valve failure, since it requires dual mode failure. 1 b l 1 11/9/87

3-7 Potential Cause Hydraulic Speed Control Failure Discussion The hydraulic cylinder function is to slow the closing speed of the MSIV to specification limits under a wide variation of applied forces. The closing speed of the MSIV is accomplished through adjustment of the Monatrol needle flow control valves Parts

1. 6 and #7 as shown in the drawing 13560-01-4 hydraulic flow logic schematic.

J Should either or both flow control valve (s) and also all other fluid leak paths (e.g. ring gaps in piston) become totally blocked, motion would be prevented. Such a situation is unlikely because: 1. The amount of contamination would need to be so large that it would not disappear after one cycle. 2. The hydraulic fluid was installed under clean controlled conditions. The system is closed and pressurized, preventing contamination from external sources. 3. Such a failure mechanism is not supported by historical experience. NOTE: The flow control valves are designed to provide a flow path even at the maximum choked condition. Conclusion Unlikely to be occurring. 11/9/87

3-8 Potential Cause MSIV internal binding. Discussion Poppet binding against the upper body ribs due to poppet rotation is very unlikely due to poppet concentricity and long length of rib engagement. Binding of the stem against the packing gland edge is considered extremely unlikely by the valve manufacturer. Potential for the lantern ring to cock and bind to the stem is a possibility with inadequate packing compression but is also considered unlikely. The packing compression used in the reassembled valves is esti-mated to be adequate to prevent lantern ring movement. Conclusion The low probability of binding and lack of reported industry cases, is inconsistent with the multiple valve failures or the time factor seen in the free up of some valves. This is unlikely to be occurring. e a 11/9/87 ~- .~

3-9 Potential Cause Swagelok fittings improper installation / assembly / leakage Discussion Excessive fitting leakage would not cause an irregular operation of the valve. This type of leakage would induce a constant operational characteristic, i.e. slow rate of change. Likewise, the accumulator would close the valve in case of leakage on ASCO pilot control valve tubing. ConclusiQD Unlikely to be occurring. T 9 11/9/87

3-10 Potential Cause Failure of the Part #5 ASCO Model 8320 3-way solenoid valve. Discussion The model 8320 3-way nolenoid valve is used to slowly stroke the MSIV (close MSIV when energized). When the solenoid valve is energized (opened), pneumatic pressure is routed to the Part

1. 2 3-way air valve.

This causes the 3-way air valve to vent the rod side of the actuator through a flow control orifice, while blocking the inlet air from air valve Part fl. The gradual less of pressure from beneath the piston allows the actuator springs to slowly close the MSIV (up to 60 seconds). The potential failure modes of the valve are: a. Stuch open (failure to close whan de-energized) j b. Stuck closed (failure to open when energized) i c. Stuck partially opened d. Catastrophic failure of valve body The effects of these failure modes are as follows: a. A stuck open valve prevents reopening of the MSIV. i b. A stuck closed valve prevents operation of the MSIV in the slow closure mode. This is the normal 1 (nontest) mode of the valve and does not affect the normal closure functions of the other subcomponents. c. A partially opened valve will tend to close the MSIV; however more slowly than the normal fully opened condition. This affect can be visualized in the drawing 13560-01-H schematic. The 3-way solenoid valve, partially opened, would bleed inlet air from the system, e.g., exhausting it. Additionally it could pressurize the 3-way air Jalve resulting in further exhausting of both inlet and air pressure. d. A catastrophic failure of the valve body would result in loss of pneumatic pressure resulting in MSIV closure. None of the above failure modes support the delayed closure event at Perry. Conclusion Unlikely to be occurring. 11/9/87

0 3-11 Potential cause Valve packing too tight. Discussion Grafoil packing has replaced earlier asbestos packing on 7 of 8 MSIVs. While it is likely that the grafoil packing has greater breakaway friction due to increased compression of the softer material, the circumstances of.the events showing quick closure after initial release make this somewhat unlikely as the cause. I conclusion Because other valves with grafoil packing and equal packing compression requirements showed no effect during fast or i slow speed testing and the lack of industry experience of an MSIV being held up due to packing, this cause must be considered unlikely. I I 11/9/87 w-- m-- = ~ w--- -~ -,e ,.,4 _y a.

? 3-12 Potential Cause Failure of the Norgren Model F0013A 4-way shuttle valve. Discussion The 4-way shuttle valve is energized by the Part #4 3-way dual solenoid valve. Upon energization it routes pneumatic l pressure to the rod (bottom) side of the actuator cylinder piston and vents the blind (top) side of the piston. The resulting pressure differential across the piston forces the ] rod up, opening the MSIV. The 3-way dual solenoid valve when de-energized, vents (de-energizes) the 4-vay shuttle valve, venting the rod side and pressurizing the blind side. The resulting pressure differential across the piston in conjunction with the springs forces the MSIV closed. The Part #3 2-way air valve is provided in the circuit to eliminate a single mode failure of the 4-way valve. The failure mode of interest concerns failure of the MSIV to close when the 3-way dual solenoid valve is de-energized. Should the pressure leg of the 4-way valve

stick, the pressure is still vented by the Part (3 2-way valve.

If the exhaust leg sticks upon de-energization of the valve, the j springs alone are capable of closing the MSIVs although at a, i slower rate. If either leg partially

sticks, the inlet pressure is exhausted, promoting closure of the MSIV.

Conclusion Ths only failure of the 4-way valve which can res' ult in delayed closure of the MSIVs as experienced at Perry is sticking of the pressure leg with a concurrent failure of the Part #3 2-way air valve. This is unlikely as it is double mode failure - requiring failure of two separate subcomponents. Thus this is unlikely to be occurring. 11/9/87

- _ ~ 3-13 Potential Cause Valve line-up of instrument air header system. Discussion Had an improper valve line-up in the instrument air header system occurred, numerous other air users throughout the plant would have been affected. Key valves and the possible consequences had they been advertently closed are listed below. l 1) 1P52-F640 (manual drywell isolation). Improper line-up of this valve would have prevented repeated actuation of B21-F022A, B, C, and D. This valve would also isolate the MSR valves as well as the personnel air lock at 599'-0" Elevation. 2) 1P52-MCV-F646 (drywell isolation). Had this valve closed, it would have been indicated by status lights on both H13-P601 and H13-P870 panels in the control rocm. ERIS points EC-007 and 008 would have also indicated closed. 3) 1P52-MOV-F200 (containment isolation). (A) Had this valve been closed the entire air supply into contain-ment would have been isolated which in turn would have affected instrument air supply to all the air users off' of the air distribution manifolds P52-J600, 601,

602, 603, 604, 605, 606, 607, 608, 609, 610, 611, and 612.

(B) Also, had this valve been closed it would have been indicated by status lights on both the H13-PS01 'and H13-P870 panels in the control room. 4) Manual valves P52-F554 and F605. Had these valves been closed they would have isolated a large number of; tne air users throughout the containment. With all of the discussion above the fact remains that the valves did operate as observed. This would not have been the cause since the MSIVs would not have repeatedly func-tioned. Conclusion Unlikely to be occurring. 11/9/87

e e 3-14 Potential Cause Air pack wiring and termination failure resulting in a hot short. Discussion The air pack units are self contained for each solenoid and wired to a common junction box. This wiring and itssociated hardware is provided by the manufacturer. The 'tield wiring I is terminated at the respective solenoid valve junction boxes. Refer to drawings D-209-013 Sheets 2 througe 9 fer i each of the MSIV assemblies. Per review of the interconnection wiring diagrams and corresponding elementary schematics, the wiring and termina-tion information is correct. The control schematic for operation of the respective solenoids is "fail safe" by design basis, which requires the solenoid coil to be energized to prevent an isolation. De-energization ould result in closure of the valve. The wiring to each varte,is classified as Class 1E. Al-though the 120VAC power to each of the A & B pilot solenoid valves pairs is contained in a common cable, each conductor is properly sized and meets the separation requirements. The cables are rated for 600 volt insulation, besides having, minimum current draw. Therefore, the potential for a hot i short is improbable. ~ References D-209-013 Sheets 2 through 9. Conclusion Unlikely that wiring or hot short is a potential cause. ~ 11/9/87

~. I e, NOTE 5 NOTE G i SOL ENO/D #l (1821 -F 460) WIRE COL 0R g sz t Fo 2z.A { VARK CODE

1. M~

%,om ogwe l t BZiH36:3A G 2 eziH36isA 5 i n ] SOL ENO/D 2 Nolk A 6 B2 i i[N OTE 7 ANb N 5 B2. l WIRE COLOR _../ 4 62 l MARK CODE' 3 ga i BZl H 3613 A 4 l T g2 2., 8 9.1 H 3 G il A 3 , pp n l JE 4 ] I Nf^Nb SOL ENO/D 3 N OT E 7 WIRE COLOR MARK CODE i e2:y seo3A 2 'DM 2 62iy3Go: A I 7zo9 -ol3, SUCY2 8

O J MOTE 2 (I B7 I-F460) SOLENOID VALVE JCT Box l J'UNCT/OA/ EOX A I Ezi von. A Mwova ukuoE \\ ] NOTE 4 6 B2/N36 /3A G s s 5 82/H36l5A O ~ mote 7 i 1 J 4 82/H36 /3A i J , klO3 3 82/H3 Gila W

2. 82/N3603A R

i B2/H360 / A BK n i N I NOTE 4 M AN D NOTE f j AttTML.30G E ~~D z o9 - O G Skd E-l ..^ unTCC -ew w...-* - - - ---w- -e-e-- +

ilb b 71 79 0623 F33^ 2XI 7 3* - f ] p sREF) () 6 l C w 10 5 C i J.JL3? f K b S2N 1 ? \\s i l l I}:252A [ l -l l l_ ')_' 12OVAC MI A j SH 5 FIG 4 NC l l .;lo?_I hi M ti H 1 3 - P 6 0 1 [ i J 2.b b l L

• 10 7 C

O O I Bzi voz.sA i

1. 3C C mRot. Stovic.C i

i ~ E of 25 gi / K74A / K MC-KT5A 64' B4 i 9 -c ~ - - - _ _ ~ - - l H13-P623 7 '~ TART iAL DLOG *. (05\\K7 12os ors s k.ect \\\\

~~/ e \\ /1 V A LV E IB 21 -Foz.8A my mc 7 V 1 3-ZO&O96 ~ F480 - i C95-o32_ PILOT SOL. TEST SOL. LS-3 ISA r l H O NOFC D O NCFC MN 2 ) ' 050 -C' .O lG w tyc 93 44 05 os 26 ~~ IH13-F713E hil 12 n' a,3 j ( >- I 4) ( i o 6 BB-20 n _ o 'd[ M I '~ 1B2.1 Fo2B f\\. St Bl + T' lot Secooto .Al27A N14A Al i B4 og a & 07 H p m pne nm g l 0 C luo s(Myo t uurT o M4 Kl4 A j A uiT cv =- 1 f c5 3'a [9 - 0 7J I ( R3 ~~P a r n t ' D m 4 i 8 ~d K 74A BZoS-o G skeet )1 ( ) yIX1 I } p EE 21- - c 5, O O

JZOZ_ 1 A B C. t t_)- I O E t 9[ o r s 2) v o z.a. A Rz i gg74g 3,7 som mo,o a (o -5 i 'L 2 R;,. O'[ h "A ~~ vU 4 ' 1 B' Ac1P !1sme. 4 I , %g (W 100 IlhT1A t 1 loI\\T A( Al j Ki3Ag 56 y ls t / K51A ( -- i __ K SIA i 1 gg t p y r, t i e-o o 'N Q RG61 A' y (y 6 i 58 Pt z;) o (4 st w 2 2. g 4 ,x x _o7Eo%, 4 na 1 H -l1-- w o34 6 ao o e c 6 i NOFC l A .i 8 8 li \\ PILOT SOL. LS-2 LS-1 ' Pae.T a t, % w n - 'Bzos ol3 slect u 3 i. 4

71 73 i F33B r-D I !H 13-PS2o c E] O i i 4D j (REF) D 0 u '~ t > 66 ) w EST l a_: SIA 76 't l i i I

SI A l

L l g + 120 VAC l v II A l ~ ) su.5 pic e ' o% a w il uis-P601 34 a(, v c -_)- (o la iBr.1 D zz4 O O c ooraot. SasTc.M v 'l 3 C. [ 3 8I - bl K74B g g;40 g K7 B4 V lHlhP622

• (05 3%7_

-~~ 7 "^'

• ded TZoB-o63 to 1

.)

i ,1. O wN O h [ 1%2 g Tozz. A Rz g g743 7t,c %<,mo,3 h 07,fA "A" 1 \\ M 1 j ./ " Ane rterwe. ( B f leice<- o t Cr l ' DSZBB [) A2 I B1 BI' K61g A [-) l <l31B/ / TI 7A 64 54

K S I I

i O j j L h, ( y %sm 9 0 -4 l 55 (,5 3 j>_9 qs q'c 7c > 6 2. \\ l s 05A 07E">09A">cw 1.1A o o o o o l x / \\" / V V l ckf8 h ~~Pe+ %A - O >1e t_coc g m_ l 7

• • 3 6 1

O { j M ni 8 (.' f my a 6 2 p z.s ,) 'n O O U N 4 2 P ad, o w w 0 -W h i o p e s. CO ( 4l l V o + f / T k 4 0 4. n. s e t' x f \\ N N <t - CD cc 0 g'

n s

n -+) ) y i+ -l i .- y h r a u m N b' 7 Ci d I c P-O o. h c i P g < U co m a e / tn m as o / O g 2 O r b 3 x c ._ s Q b -g i 'A i to c ,o

e s 3-15 Potential Cause Glaked contacts on centrol and relay components creating a high resistance which would result in discontinuity and potential mis-operation of the MSIV circuitry. Discussion Contact integrity and circuit continuity of the respective solenoid valve coils is constantly monitored by measuring the coil circuit current, in addition to an indicating light (white) which relies on actuating contact integrity to remain energized. Refer to attached partial of drawing B-209-013, Sheet 10 and Sheet 11. ~ The isolation control circuit (s) are a "fail safe"

design, which requires the solenoid coil to be energized to prevent an icolatio.i.

If contact glazing had occurred resulting in a discontinuity (L gh resistance at connect >on or

3ntact points) in the control circuit (s), the resulting effects would causa the lack of voltage to the coil (s).

This condition, due to the "fail safe" design, would cause an undesirable isolation (closure of the MSIV valves), rather than a failure to isolate. Eiferences D-206, Gheets HOS, H10, H11 and H36. Sqnc.1usion E 'ider.ce of repetitive tasks to cycle those valves along with the proper configuration for power and control indica-tion does not sugaest any potential failure.

Also, the control circuitry and electrical components for each of the inboard and outboard MSIVs are identictl.

In that thsre is no past or present evidence to support thAs cause scenario, it is high1) unlikely that this is the root cause of the problem. 11/9/87

m 3-16 Eptential cause Relay failure or incurrect oporation resulting in mis-operation of the MSIV valves. Discussion The associated control and relay componenta are lochted ih the PGCC which is designated as a non-harsh environment and is also seismically designed. Furthermore, this area is controlled for relative humidity and temperature. The likelihood of a failure or incorrect operation due to component failure is highly improbable in that this failure would have to occur on three (3) different MSIV log-ic/ control circuits. The proper operation and closure of these valves and repetitive testing positively indicates that relay failure is not the cause. Also, as shown through testing and verification, the control functions and indica-tion was correct. l l conclusion Unlikely and highly improbable that relay failure is a potential cause. i l 11/9/87 .--we -g---g -,w., ,w,m,-,- - -s- -+r-, yw w--e', ---r ww-t-v-- --w - ~ ~ r-

• • • 3--t--~r-e----"---r---v--

v7 -- -

f 3-17 i Potential Cause Panel control switch failure or mis-operation. l Discussion I The control switches nos. SLA-D and S2A-D are General Electric type CR2940, 3 position maintained contact switch-es. All of these are located in the PGCC. The control schematics, as shown per drawing B-208-013 Sheet 10 (in-board) and B-208-013 Sheet 11 (outboard), are id3ntical. No test data or evidence has been identified to suggest a failure of the switches. Repetitive testing has demonstrat-i ed the proper operation of each of these control switches. References B-208-013 Sheet H04, H10, and Hil. f Conclusion Evidence of repeated acceptable testing to cycle these l valves does not suggest any potential failure. As such it l is highly unlikely that this is a potential root cause o f-the problem. l r I i i i 11/9/87 l 1

3-18 Potential Cause Limit switch settings incorrect or inoperable. Discussion The limit switches (total of 6 each) for each of the MSIV inboard and outboard valves are NAMCO type, as furnished by Atwood & Morrill Company. These limit switches are not an active component in the control schemo which initiates opening or closure of the respective MSIV valves, rather they monitor and provide local indication in the control room for valve position. Refer to elementary drawings B-208-013 Sheets H10, Hil, and H36. The potential for inaccurate limit switch settings is possible, but other independent sources can verify and provide indication for closure or opening of the valves via instantaneous steam flow and steam line pressure.

Again, this issue wculd not impact tha actual operation of the valves.

References B-208-013 Sheets H10, Hil, and H36. Conclusion In that the limit switches are not part of the contrel circuits, mis-operation would not affect valve closure. 11/9/87

3-19 Petertial Cause Miswiring for indication of instrumentation or switches. Discussion This potential cause was recently a problem wherein the "A" and "B" solenoid valves were wired to a common Reactor Protection System (RPS) bus. The basis of the design requires that sach of tho trip solenoids A and B for each of the MSIVs be wired to different RPS buses. This issue was corrected via the preparation and issue of Design Change Package (DCP) 870414. As part of this design package and a prerequisite for start-Lp, each of the MSIVs were verified and tested for applicable power sources and functio'ial operations. The probability of additional viring errors is highly unlikely in that repetitive testing of these valves did not indicate mis-operation. References B-208-013 Sheets HOS, H10, Hil, and H36. Conclusion Although this item was a problem previously, it is highly unlikely that a similar type of problem could be the root cause. The efforts to resolve this RPS problem, ratesting and management exposure significantly rule out this poten-tial cause. Also, recent testing of the specific valves in question indicate that the instrumentation and switches are correct. 11/9/87

3-20 Potential Cause Data acquisition failure. Discussion Failure in the data acquisition and recording system coule lead to improper assessment cf closing speed. Valve speed data is taken and recorded using the TRA subsys-tem of ERIS. This system has tha capability to sample data from a wide variety of signals for later analysis. Data on reactor power, steam flow, reactor pressure, limit switch position, and solenoid current are all consistent. Measure-ments exterior to ERIS, main control panel and back panel indicating lights, for example, are also consistent with the ERIS data. In summary, multiple concurrent failures neces-sary for this scenario to occur make it incredible. Conclusion H4.ghly unlikely to be occurring. 11/9/87 -4 ,,-m- ..,~s -,y r,,-,,, ,-a,,n- -.pe,-,. -. - ~p ,---me -r.w-e. m.

3-21 Potential Cause Procedural error for testing. Most previous fast speed MSIV closures have been performed using SVI B21-T2001. The first failure was noted while performing the test per STI-B21-025A section 8.3 and the remaining failures were noted while performing the MSIV strokes using the system operating instruction (S.O.I.) Discussion Although most previous tests have been performed using the SVI, this is not the first time that an STI has been per-formed. As early as 10/12/86, STI-B21-025A section 8.1 was used to fast stroke the valves. Additionally, the use of the SOI has been demonstrated before and after the failures. During the B21-F022D, B21-F028B, and B21-F028D failure on 10/29/87 and the B21-F022D and B21-F028D failure on 11/3/87, the SOI was used. However, this is the arme SOI that was used for the remaining valves which passed their stroke time. Conclusion It is highly unlikely that there is a procedure problem. l 11/9/87

3-22 Potential Cause High Steam Flow /High Reactor Power Interaction. All previ-ous low and high speed MSIV closure tests have been per-formed at low to medium reactor power. The potential exists that the higher steam flows associated with high reactor power could interfere with MSIV closure. Discussion Although all previous tests have been run at low power, the valve design basis is closure at full flow, and the capabil-ity of the valve to close under full power conditions has been demonstrated numerous times at numerous operating BWRs. The valves that showed delayed closure are identical in design to valves that closed within specifications, and the affected valves closed successfully following cycling. The valve design is such that pressure drop associated with steam flow will actually assist in closing the valve. conclusion It is highly unlikely that this is the cause of the problem. i \\ 11/9/87

l 1 3-2' Potential Cause Incorrect reassembly and installation of the air pack. The air packs were all removed, but not disassembled, during the September 22, 1987 forced MSIV outage. The purpose for removing all of the air packs was to allow for temporary air supply to be installed and allow local stroking of the MSIV to check stroke measurements. Discussion During the September 1987 outage all air packs were removed fram the MSIVs to facilitate local stroking of each valve to set the stroke length. After final reinstallation of the air packs there were several fast and slow strokes per-formed. These strokes were performed using SVI C71-T0039 and SVI B21-T2001. Even though SVI C71-T0039 (slow stroke testing) does not test the same valves as SVI B21T2001 (fast stroke testing) the same air pack is used and the mating surface between the air pack and actuator remains the

same, as do all hose connections.

conclusion It is highly unlikely that this is the problem due to the' number of strokes performed after reassembly. 1 11/9/37 l

3-24 Potential Cause Actuator binding / stem binding Discussion Binding of the actuator internals for both the hydraulic and pneumatic assemblies is highly unlikely. Neither assembly is subject to external loads to cause stem bending. The hydraulics are not subject to external particulate contami-nation and contamination within the main-air cylinder may score the cylinder but could not likely stop the movement by resisting the air pressure force. conclusion This cause would likely have shown up during prior history of stroking the valves and would not likely apply to multi-ple valves at one time. Nor would such binding likely apply to the top of stroke only. Thus this cause is estimated to be highly improbable. i 11/9/87

1 l l 1 1 I i SECTTON 4 FAILURE ROOT CAUSE DESCRIPTIONS I J i l { l { 11/9/87 -.c

=. 4-1 Potential Cause Local High temperature has caused deterioration of EPDM seal materials. Discussion Perry has experienced drywell' and steam tunnel temperatures which have approached the Tech spec limits during much of the startup test program. Figure 1 gives a history of the bulk drywell temperature since June of this year. In

addition, localized temperatures in excess o' the bulk drywell temperature during the past year.

It

addition, localized temperatures in -excess of the bu '.k drywell temperature can be postulated to have occurred due to - steam -

leakage from several valves. In particular, main steam isolation valve B21-F022B had a major steam leak just prior to the actuator / stem separation incident during September 1987. Leakage control system valves E32-F001N has also experienced several body to bonnet steam leaks. Figure - 2 shows the physical location of these valves relative to the location of MSIVs B21-F022D, B21-F028B, and B21-F028D. One of the solenoids from B21-F028D was found to have rust and corrosion, indicative of a steam environment. Adiabatic expansion of steam from 1000 psia to 15 psia will result in a steam jet temperature of about 300 F. This jet will, of course, rapidly cool and condense to saturation at drywell conditions. This local condition, along with the proximity of the leaking valves to the MSIVs which failed to close is indicative of a temperature related cause. Discussions with Automatic Switch company (Asco), the manufacturer of the failed component, has indicated that elevated temperature is a potential cause of the hardening of the proprietary EPDM rubber compound used for the valve seals and o-rings. Seals and 0-rings taken from MSIV solenoids for valves that had not demonstrated delayed closing do not have the level of degradation seen in seals from the failed valves. In particular, preliminary inspec-tion of the seals from MSIV B21-F028C indicated this valve to have seals in a near-new condition. As shown in Figure 3 arrangement of ventilation in the steam tunnel is such that this valve would be expected to see the lowest ' ambient temperature, and conversely, F028D & B would see the highest ambient temperature. In combinatin with the previously discussed steam leakage, it is clear that F028D & B have been exposed to higher than expected ambient temperature. 11/9/87 ~n. m, g re en -m-we--m---v---o--e,-,-e-r-,-- ,,,-,---+,wv.,w.we-r---,.+re,-,-o,-r----r--rwre-,,-,,re-me,r g w y n,-- pemw-w- - e v,-,n, vm m,e-w ee-w s e-w o r n a,m v --

? o 4-1 Conclusion Elevated local temperature is the most probable cause for degradation of the EPDM seals in tte ASCO Model 8323 pilot solenoid valve. The material is known to be temperature sensitive, the potential for elevated temperature has been shown to exist, and the best performing valves are in the lowest temperature locations. i 9 3 11/9/87 --y-n+-nrw* -.--&y,- ,--m--g-,,_,,ygm., -m. %,.,yyme r.,-y--,,,%9,- ym., yg .,w-g,yyy mm, c -my -g-.ygg,.,m-y---em-.-e-a

4 O l y>vwa v %, '~ ss \\ . M n = ~ 39 %.00 2 e }\\ N q o s no 1/ O 00 V4 et dg P - at L M co e c W .u, J g W V, 3 F r -r cd c2 k t 3 0 k kp 5 0 I i if o M i '] Uf i '6 e (,3.) 3 a ni.w-4 saws L j

*\\l'* I'* i F' 8 I r TM rH 14 z wiao t ivu> ) f..,g w... ,.,w,.3.,.,,,,, .. - 1 e. i 1 rt2-2 Hiu s " j %.....,.... x b1202T j ot2a N.. Hgs"# % Fmuee 2 5% LeMdC- + ~, Lhe -b r 5 -- -d \\ '"._'_y-MO;i 97t.?,"" l.oces oe) (6 r ao%a.) L., m- - - - - n, _3 2... D]425' 5"{ 7 w,p, gg,43g'.3}8 W2t1 W.R E L. e stora IS' w e, gg,y W.h EL Git'a p,k"(TOP) / L j I W.P. EL G69'.M(BOT) MHl03 g f 9' 2s F50GE TO P) w, p. E L G'3Ok 3*(TO P) ] I-SOT)

s. W. P. EL G29' Ok'(50T)%

l 'H09' g pe I ) SI N y:pg gg-) g'M .,.s + 'N to ,,: # g,- p g2r i z _7, g Ii IEl ,6 6 ~ l 7 - y ST E Av l \\ W.P. EL. 620'. 9 {. /gHill, .ws g["< t32I ko28B l W.P. E L. G2 O'.94N i.or g

e. i' e (Pu Et Gi g n

c.<) I j 1 j.c 5 *. o[ , c)" N Y T I- g'. e, l

• i'.o{

-+'g_3 2 W , i - * '.% Hil3 g g [ l, 34 ll',1 PL AT F, E B24 FO28o 4 Hl 450'.o / ' ' j 4 3,.., .)STEAh f 9 Y N Hl $ Hjog i ,ns_ 6%,, p u.- l 24.H il0 y f HIM i 2,.n3 m.i,"c" w-c m o,n 4M b4 E 6 r., t.. u.. _ ~ O l*5} N ! 9" -w,p,Et 43o'.3 1" W. P. EL Glo' 2 k "(top) jd' { Hl36 4 ~ W. P. E L G 2 9 '. 21L = ( c oT ) 'g l f / ST E AP/ 8* I l Pw AT F. E t.,- Hi?Bg lf @6,. U w w-o s u o I Hii7 s pHiis ) s Bli F026 f, 2 Y %A NIf 4 X Hitt 2jHOSer 0 2 '- i t g , l'. i" I i l<w-j-[o \\ C I <' l i" f MCV root A "Hil5 t b II'$ MCV F001 A 9 -" Y. Hil8 - W. P. E L G 30'. b [ jj W.P. EL G 30,' t h"("top) -H08 7 2 {* I* S k_ W4ELG2 .q g g T) -Hl3 5 4' 'l' I {"S.W. cotaa, i /Ng ~ /. 7 (C.S.)(1vP 4T, H094 3,# gh I l SE E DE T Ait" r __ HIT $ M NII4 MtvFOO2J 'g,n MCyn e242 Q \\ ;025J k =5 Ni W.P. u ez e.,p,(Top)[2, hh,J ,..,.4,,.N 'O L = ,L W.P.NG30'n" %i{"oi i ~ c 3ga q G2p-O,, m c-

'e l l p [f(y(jgf _-f $7f~ arm lYA/A/f/- Y[/V7/L A l'/CAJ /A " l l AUXLIIRY BUILDWG } TURDNE POWER COMPLEX 1 l' 1 tboo <L. A r.,w g. 00M AIR Y EAmust f %w rist Y s / ~~g. p ~~ ~ ~ ~ E31TYl' OP 4 ~ '" t =* O-Fggt$v/ATER. i f f ^Yed* e i ^ TyP 07 7. = = o :: a g g es y V-F:EACTOR Bzarozwn Pilp/ V V BUILDlHG b 35"*S~ TO TUR9WE.BULDWs BOlfoteA \\ ) > MS VALVE AREA

o=

~O I -- "Ws -- m - f fBZlrom f ESI TYP OF f f T A TELw T NT'7#*~ cott> me i xcx_ -Q$'l ,[#h .C00lA O

---

6--G- -> ' CON 1ROI. VALVE

>c>- 7 ~~ ~ M'I StJPft.y 6tsim SF*F AUXRJARY NNe g e = r-r NORTilEAST IIALLWAY L5o'o <s-p g g g oowtv a e r-g h.-" 5 __-_g COOld pQ~ b

4-2 Potential Cause Blockage of the dual solenoid valve exhaust port with tape. D!scussion During the previous MSIV refurbishment where the air packs were removed, duct tape was used to cover exposed ports, including the solenoid valve exhaust port. On F028D the exhaust port tape had apparently not been removed following the refurbishment. Blockage of the solenoid valve exhaust port could delay the closure of the MSIV. However, the strength of the tape adhesive is considered weak compared to the pneumatic pressure forces. Typically, the tape will blow outward, reiaining connected on one side during de-energization and fall back in place like a flap. Further tests of the F028D valve has verified the tape is not an effective bicek. Conclusion very unlike;y to be occurring. e 11/9/87

e 4-3 Potential cause Jamming of kinematic components. Discussion In order for the valve to shift to the de-energized condi-tion, both solenoid movable cores must slide within their guides. The disc holder assembly is also a guided component which must shift for the valve to operate. Failure of the components to shift may be caused by foreign material contamination of the sliding surfaces, either particulate or fluid (adhesive in nature), or by physical damage to the valve parts. Examination of the F022D valve, and the air supply system has not identified any unusual substances or damage which could explain the MSIV delayed closure condition. Consider-ing the proportion of valves which demonstrated the delayed closure (3 of 8), an extremely dirty system would be expect-ed for this effect. ponclusion Unlikely to be occurring. l l 'T 11/9/87 ---,,--.-.-------,,,n,-, ,,n,,e.,, .-.-----.,,,,y, ,,w,,..- ,-,__.w-.,,,,,~.,.y W- - - + ,m-, ,-g .,,e ..-w,,r,a,

i e 4-4 Potential Cause Oxidation of EPDM rubber compound used in gaskets, seals and disc seal materials. Discussion Oxidation of EPDM rubber in the presence of a brass catalyst has been suggested as cause foi a similar incident at Brunswick-2. This has been documented in INPO Significant Event Report 57-85. Review of SER 57-85 indicates : hat although catalytic oxidation is a potential cause for the Brunswick situation, that utility was never able to determine the exact cause for EPDM degradation. There is, however, a relatively large data base for use of EPDM elastomer in brass valve bodies with acceptable results. The solenoid valve supplier has stated that there is no evidence to suggest that catalytic oxidation has ever occurred. The condition of other Perry valves would be expected to be similar if oxidation were at fault. This has not occurred. Conclusion' l Catalytic oxidation of EPDM in the presence of brass cannot be completely ruled out as the root cause for pilot valve failure. While postulated as a failure mechanism, its validity has not been proven. If catalytic oxidation does play a part, it is most likely as a contributing factor, in the high temperature scenario, for example, \\ 4 b I 11/9/87 ) )

e 4 s 4-5 i Potential Cause Residual magnetism following coil de-energization. Discussion i Sufficient residual magnetism of the ferritic steel materi-als in the region of the coil could cause the valve to l remain open following de-energization. No similar experience has been found elsewhere. The ASCO valve representative has identified that the solenoid valve return spring is sufficiently strong to overcome residual magnetism of the ferritic steel components. Any residual i magnetic forces would be low compared to the closure force unless additional magnetic mass was added to the coil vicinity. conclusion Unlikely to be occurring. '1 \\ l 11/9/87

e 4-6 Potential Cause Wrong materials. Discussion This failure root cause description considers the use of wrong materials for the disc holder elastomer seal. The potential for wrong lubricant is considered separately. Dimpling of the disc holder seal in the dual solenoid valve is postulated to result in wedging of the seal in the exhaust to cylinder port. The use of a wrong material could result in the observed dimpling. The proper disc material is an ASCO proprietary EPDM, utilized in their nuclear qualified valves. Material problems may include the follow-ing: Wrong material of lower strength or thermal capability. Improperly cured EPDM. Improperly formulated EPDM. An analysis of the disc material may be performed to identi-fy the material or formulation; however, it is unlikely to determine the relative cure of the compound. Conclusion This is not expected to be occurring, and will 1e co'nfirmed by material analyses. i 11/9/87

Y + 4-7 Potential Cause Locaseal vapors Discussion In order to seal the solenoid housings on the solenoid valves a Locasaal is poured in the opening and allowed to cure. The compounds contain no oils, solvents or reactive materials. Also, the alkylated phenols and aromatic amines are highly cross-linked and polymerized. This configuration does not allow the release of hydrocarbons. Furthermore, the lowest temperature that decomposition takes place would be approximately 500 degrees F. The ambient temperature of the air pack assemblies are greatly below 500 degrees F. conclusion Unlikely to be occurring. 12/9/87 -~.

a w 4-8 Potential Cause 0-ring / lubricant interaction ' Discussion During the disassembly and inspection of the ASCO dual solenoid valves, the three body gaskets (o-rings) were found to be significantly degraded. Degradation included harden-ing, flattening and adherence to the mating valve body. The observed condition of the gaskets could be caused by an improper lubricant. The EPDM gaskets are susceptible to hydrocarbon oils. Normally a silicone oil (Dow Corning 550) is used as a gasket lubricant. EPDM is compatible with silicone fluids. The degradation of the gaskets could not affect the valve itself, as they are located away from the moving components.

However, vapors from the lubricant (no signs of fluid migration were observed) could result in sof tening of the disc pads resulting in the dimple ef fect suspected as being the physical cause of adherence.

Conclusion Possible but unlikely since similar valves have not shown the same condition. The o-rings will be investigated for proper material and lubricant. i k 11/9/87

s s' w C 4-9 Potential Cause corrosion within solenoid enclosure. Discussion The "B" coil housing in the F028 MSIV dual solenoid valve was found to contain moisture and corrosion. Corrosion within the solenoid coil housing cannot affect the valve internals as the valve body is protected from external contamination through body gasket seals in the vicinity of the coil. The subject coil ("B" side) is the lower

coil, such that any corrosion products escaping the coil enclosure would fall down away from the solenoid valve body.

Addi-tionally, corrosion products were not found within the valve body. conclusion very unlikely to affect performance. ) 1 1 e 11/9/87

9 ~ UNITED ST ATES l ,'p* *8 c g'o, g NUCLE AR REGULATORY COMMissl0N [; ' ; e(,' $ REGION iil s E 799 RooltvtLT ROAD OLEN E LLY N, ILLINOIS 40137 NOV 5 W MEMORANDUM FOR: R. D. Lanksbury, Team Leader, Perry Augmented Inspection Team (AIT) FROM: Edward G. Greenman, Deputy Director, Division of Reactor Projects

SUBJECT:

AIT CHARTER Enclosed for your implenentation is the Charter developed for the inspection of the events associated with the Perry MSIV failures which occurred on October 29 and November 3, 1987. This Charter was prepared in accordance with the NRC incident Investigation Manual and the draft AIT implementing procedure issued for use on October 2, 1987. As stated, the objectives of the AIT are to comunicate the facts surrounding this event to regional ar.d headquarters management, to identify and comunicate any generic safety concerns related to this event to regional and headquarters management, and to document the findings and conclusions of the onsite inspection. If you have any questions regarding these objectives or the enclosed Charter, please do not hesitate to contact either myself or R. Knop of my staff. MM Edward G. Greenman, Deputy Director Division of Reactor Projects

Enclosure:

AIT Charter cc w/ enclosure: A. B. Davis, RIII C. J. Paperiello, Rill F. Miraglia, NRR J. Partlow, NRR C. Rossi, NRR G Holahan, NRR W. Lenning, NRR 1 M. Virgilio, NRR R. Cooper, EDO K. Connaughton, SRI s s /9 /

Perry MSly Stroke Time Failure Augmented Inspc tion Team (AIT) Charter Investigate: 1. Failure of MSIVs to close/c hse within Technical Specification limits. 2. Safety Significance, Root Ce te(s). 3. Interaction of prior mainten. activities to the event. 4 4. Safety implications if actual o. kolation signal had been present. -5 5. History of any previous problems. 6. Broader 1mplications e.g. other systems, other vtive/ components. 7. Event Reporting. 8. Conclusions. Questions for Perry AIT 1. Failure of MS!Vs to close/close within Technical Specification limits. (10/29/87and11/03/8J) 1.1 What was the sequence of events? 1.2 What were the closure times generated during the surveillance? 1.3 What operator actionc were taken during the event? Were they 1 appropriate? 1.4 Is there a history of any previous problems (e.g.10/29 event, etc) with the MSIVs? 1.5 Did the RPS logic makeup per design during the surveillances? 1.6 What additional testing was being perfonned? 2. Safety Significance, Root Cause(s). 2.1 Was there any immediate safety significance from this event? If so, what was significant? 2.2 What was the root cause of the event? 3. Interactions of maintenance activities to the event. 3.1 What is the past and present maintenance history of the MSIVs? 3.2. What is the maintenance histcry of the Service Air (SA) and InstrumentAir(IA). 3.3 What testing was perfomed as the result of maintenance activities? 3.4 What is the material condition of the affected valves and inter-connected instrument air and control systems as it would affect the valve closure function? 4. Safety implications if actual Group 1 isolation signal had been present. 4.1 Does the licensee have procedure in place to handle this event? 4.2 Are they adequate? 4.3 Have the operators been trained on them?

4.4 Does the accident analysis bound this event? j 4.5 What actions were taken by the operators? 4.6 War the event properly categorized? 4.7 Was the event reported as required? 5. History of any previous problems. 5.1 Have there been previous events similar to this? 5.2. If there were previous events was the licensee aware of them? l 5.3 If not, why not? 5.4 Is there infonnation available on other similar events? 5.5 Have there been any IEIN's or IEB's issued or similar subjects? 5.6 Is there information avaiable from other sites of similar problems? 6. Broader Implications. 6.1 Is a IEIN or IEB warranted er a result of this event? i 6.2 Are there other valves or instruments that require investigation? 6.3 If the problem lies external to the MSIV's, are there generic -implications? e.g. for other plant systems or other plants with same components. 7. Conclusion. 7.1 What corrective actions are proposed, and are they adequate? 7.2 Examine generic implications to other plants Ond advise NRC management subsequent to the sits inspr,ction. 7.3 Document inspection findings in acc rdance with iraft manual chapter 0325. 4 ---_,,---,y. ,, ~,, - ,e

f.YN 11/06/87 PAGE 1 0F 3 AIT ACTION ITEMS RESPONSIBLE SECTION XTEM DELIVERED DESCRIPTION OPS /LCS 1. SEQUENCE OF EVEhtS X o OPS CHRONOLOGY A. CLOSURE TIMES X o UNIT LOGS B. OPERATOR ACTIONS TAKEN X o STA LOG o CONDITION REPORTS o

SUMMARY

o STI DATA LCS 2. ADEQUACYOFREPbRTINGAND X o

SUMMARY

VRITE UP CATEGORIZATION OF EVENT NED/LCS 3. IMMEDIATE SAFETY X o HISTORY OF EVEtRS SIGNIFICANCE

SUMMARY

OPS /LCS 4. ADDITIONAL TESTING X o SVI LIST ACTIVITIES IN PROGRESS o V.0. LISTS /VARIOUS UNITS I&C/LCS 5. RPS ACTUATION SIGNALS X o SVIs DUPTNG SURVEILLANCES LCS 6. MANAGEMENT DECISION MAKING X o

SUMMARY

VRITE UP PROCESS-INFORMATION AVAILABLE LCS 7. PREVIOUS MSIV TIMING X o

SUMMARY

VRITE UP PROBLEMS o CANTLIN HEMO TECH /LCS 8. MSIV MAINTENANCE HISTORY X o VO LIST - V0's (OTHER THAN STI/SVI) PROVIDED A. RETESTING PERFORMED X TECH 9. AIR SYSTEMS MAINTENANCE X o VO LIST - V0s NOT HISTORY PROVIDED (NOT IN BOOK) A. RETESTING PERFORMED X o VARIOUS PS2 V.O.s/CRs B. VENDORS MANUALS X o 3 VENDOR MANUALS PROVIDE TO NRC OPS

10. ADEQUACY OF PROCEDURES IN PLACE TO HANDLE EVENT X

o OPS

SUMMARY

A. OPERATC3 TRAINING NED S

11. SAFETY SIGNIFICANCE OF o HISTORY OF EVENT

SUMMARY

INCIDE!G (ACCIDENT o GE; MSIV CL URE TESTING \\ ANALYSIS) / a

9 s i 11/06/87 PAGE 2 0F 3 NED/LCS

12. ANALYSIS OF LOADING ON X

o GE -EFFECTS OF ISOLATION STEAHLINES (3 CLOSED, 1 OPEN) LCS

13. PREVIOUS SIMILAR INDUSTRY X

o NFRD PRINT 0tTT (NOT IN BOOK) EVENTS o SERs 36-84, 57-85, A. LER 86030 o RELATED LER SUnMARIES o PERRY LER 86030 LCS

14. PREVIOUS NRC INFORMATION-X o IENs; 80-11,81-29,82-52, BULLETINS, CIRCULARS, 83-57,84-23,84-68,85-08, INFORMATION NOTICES 85-17,85-17-01,85-84, 86-57,78-14 o IEB; 78-14,79-01A NED/LCS

15. OTHER APPLICATIONS OF X

o EO LIST ASCO VALVES TECH /I&C

16. TROUBLESHOOTING PLAN X(REV. 0)o TROUBLESHOOTING PLAN, A.

MATERIAL CONDITIONS AIR SYS. AFFECT ON CLOSURE o POINTS SAMPLED B. ANY FURTHER o SEQUENCE OF TROUBLESHOOT INVESTIGATIONS PLAN o PARTICLE COUNTS

17. GENERIC IMPLICATIONS TECH

18. ROOT CAUSE o GE; PRELIMINARY ANALYSIS TECH

19. CORRECTIVE ACTIONS TECH

20. PLANS FOR STARTUP LCS

21. CLOSURE INFORMATION ON X

o CEI/NRR LTR 0306 1985 OPEN ITEM ON FSAR o CEI/NRC LTR NOV. 9, 1984 AIR QUALITY CHANGE (3 o VIOLATION FROM 84-15 TO 40 MICRONS) o IER 85-039 o IER 85-066 o IER 85-088 o SSER SUPP 7 - 9.3.1 TECH

22. MESH SIZE OF FLUSH CLOTHS CLOSED PER DISCUSSIONS USED ON AIR SYSTEMS TESTS RPS

23. ANALYSIS OF AIR SYSTEM CLOSED PER DISCUSSIONS FLUSH CLOTHS TO VERIFY LESS THAN 40 MICRON PARTICLE SIZE (OIL, VATr.,j i

s

'/ 11/06/87 PAGE 3 0F 3 NED-MDS

24. BRIEF

SUMMARY

DESCRIBING X o

SUMMARY

VRITE UP/ LIST l RELATIONSHIP BETVEEN COMPONEffT SUPPLIERS AND MSIV CONTROL AIR PACK ASSEMBLERS (i.e. HILLER SHEFLER, NORGREN, ETC.)_ OPLS

25. EQUIPMEfft QUARAlfrINE LIST X

o POD, NOV. 5 s

P.O. BOX 97 e PERRY, OHIO 44081 e TELEPHONE (21s) 259 3737 e ADDAESS.to CENTER ROAD Serving The Best location in the Nation Murray R. Edelman PERRY NUCLEAR POWER PLANT m vxa ratsoo<t NUCLEAR November 9, 1987 PY-CEI/01E-0288 L Mr. A. Bert Davis Regional Administrator, Region III U.S. Nuclear Regulatory Commission 799 Roosevelt Road Glen Ellyn, Illinois 60137 8 Perry Nuclear Power Plant Docket No. 50-440 Augmented Inspection Team (AIT) Status and Commitments

Dear Mr. Davis:

This letter provides a preliminary description on the sequence of events, troubleshooting, and conclusions surrounding the MSIV slow closure problem. It also contains corrective actions and commitments made to the AIT on November 9, 1987. The formal report requited by your Confirmatory Action letter dated November 4,1987 will be submitted on or before December 4,1987. Based upon the information provided to date, we plan to restart the plant on November 10, 1987 with your concurrence. If you have any questions, please feel free to call. Very trul urs d Murray R. Edel Senior Vice President Nuclear Group MRE:njc Attachment ec: K. Connaughton T. Colburn Document Control Desk s o' r %', j j e s lJ b (tyy

i g I. Executive Summary On October 29, 1987 the Perry Nuclear Power Plant was completing the final stages of the Startup Test Program. One of these tests involved fast closing one Main Steam Isolation Valve (MSIV). During this test, the valve (1B21-F028D) failed to stroke closed within the required time. All other MSIVs (7) were cycled in order to verify adequate stroke times. Two of the other MSIVs failed to satisfy the required stroke time. The three valves that initially failed vere stroked satisfactorily upon subsequent demand. Based on industry experience involvind MSIV control air, the problem was attributed to's one time deposit of debris in the respective solenoids which was exhausted as shown by the subsequent successful stroke. The. debris was believed to have caused a delay in the solenoid responses. Based upon the satisfactory stroke, the valves were considered operable and startup testing resumed. Plant management decided to perform the MSIV stroke tests again prior to the last startup test, the full MSIV isolation scram. This approach was discussed with NRC Region III. On November 3, in preparation for the final startup test, additional stroke timing tests of MSIVs vere performed. During the first stroke attempt, two of the same MSIVs (IB21F022D, 1B21F028D) that previously stroked slowly, again failed to close within the required time. The valves vere suosequently recycled satisfactorily within minutes of their first tests. However, because the valves again failed to meet the required closure time on the first attempt, the basis for an isolated failure was no longer considered valid. NRC Region III was informed of the problem. The decision was made to shutdovn the plant and troubleshoo't the problem. On November 4, an NRC Augmented Inspection Team (AIT) arrived onsite. A troubleshooting plan was established and implemented. The air actuators of the three valves which had exhibited slow closing times were disassembled and the inspection results documented. The conclusion drawn is that the dual solenoids exhibited sluggish action after operating in localized high temperature conditions. It is felt that steam leaks caused a raised temperature environment in the vicinity of the solenoids. The raised temperatures degraded the Ethylene Propylene Diene Monomer (EPDM) material causing the solenoid to stick or to be sluggish. Corrective actions included disassembling all 8 MSIV's dual solenoids, and replacing or rebuilding the solenoids as applicable. G ,,__c.- -m,.-----.,-.------ c.

i II. Chronology of Events On October 29, 1987 at 1837 Startup Test Instruction (STI)-B21-025A, "Main Steam Isolation Valve (MSIV) Functional Test" was being performed on 1B21-F022D, the steam line D inboard MSIV. This valve closed in 22.14 seconds. Technical Specification 3.4.7 requires the HSIVs to close in 2.5 i to 5.0 seconds. At 2103 and 2106 the D inboard MSIV was cycled with closure times of 3.24 and 2.94 seconds, respectively. All other MSIVs were then cycled to verify closure times. The B outboard MSIV closed in 11.9 seconds and the D outboard MSIV closed in 77 seconds. Each was cycled again with satisfactory results. Since initial conditions causing MSIV slow closure could not be repeated, all MSIVs were declared operable and plant startup testing continued. On November 3 at 1150, MSIV fast closure timing was commenced in preparation for the MSIV fast closure serem test in accordance with agreements made with the NRC on October 30. At 1157 the D inboard MSIV closed in 18 seconds and was cycled again at 1159 with a closure time of 3.0 seconds. At 1208 the D outboard MSIV failed to close. A second attempt was satisfactory at 1213 vith a closure time of 3.4 seconds. The D inboard and outboard MSIVs were declared inoperable and placed in the closed position in accordance with the requirements of Technical Specification 3.6.4.a. Based on repeat failures a plant shutdovn commenced at 1330. The reactor was manually scrammed at 1819. On November 4, the Nuclear Kegulatory Commission (NRC) issued a Confirmatory Action Letter (CAL) detailing various steps Perry management was to take and not to take in preparation for an NRC Augmented Inspection Team (AIT). The team arrived onsite November 4. III. Troubleshooting Activities Prior to performing any work in the field, a troubleshooting plan was written. Based on the symptoms shovn on October 29 and November 3, it was felt that the component with the highest probability of causing the slow closures was the ASCO model number NP-8323A20E dual solenoid found on each MSIV air actuator. Numerous possibilities existed which could have somehov affected these solenoids. The troubleshooting plan was set up to determine what the root cause was and whether any secondary problems had an impact. On November 5 "As Found" conditions vere documented and a more detailed troubleshooting plan was developed to establish the root cause of the MSIV failures and corrective actions necessary to restore the valves to operable 4 condition. The troubleshooting plan was agreed to by the NRC AIT. On November 5 through November 8 various troubleshooting activities were carried out. I

\\ The first MSIV investigated was the 1821-F022D valve (inboard MSIV on "D" line). Solenoid voltages and solenoid air exheast port samples sere taken as the valve was cycled all results were satisf ctory. Next the field 3 viring and air lines were disconnected from the alt actuator or eir pack. All connections and pipe openings were inspected and any discrepancies noted. The air pack was then removed from the valve actuator and taken to the I&C hotshop for disassembly. The above steps were repeated for the 1821-F028B outboard valve and then the 1B21-F028D outboard valve..Any discrepancy no matter how small was documented for further evaluation. Whenever possible pictures were taken of what was found. The major discrepancies appear to be the following: 1. All dual solenoids disassembled have impact marks on the star shaped disk subassembly and a deep depression (disple) on the disc holder seal (EPDM), with the solenoids of the B21-F028D indicating the most degradation. 2. Many of the EPDM Body Assembly 0-Rings were hard, flattened, and adhering to metal surfaces. 3. In the 1821-F022D valve rust was found inside the solenoid valve body, and the B solenoid coil was badly corroded. In addition to the component disassembly three types of air analyses were performed to determine what contribution, if any, instrument air quality may have had in the failure of the MSIV valves. Filter samples were collected to determine particulate matter present in the instrument air system at the solenoid and actuator supply points. Various unknown substances observed in or collected from internal component surfaces were analyzed using infrared spectrophotometry to deduce origin of materials found. Grab samples of the air supply were analyzed by gas chromatography for hydrocarbon content and quantification of organic contaminants if present in significant quantities. The samples collected on filter paper for particulate vere analyzed under a y microscope. Very small quantities of particles greater than.40 micron vere identified which indicates acceptable air system quality. Therefore, it is a d1#g ~ 'very low probability that the~ particles h'ad '~n adverse"effect upon the ~ a solenoid valve operation. Analyses of the substances collected during disassembly identified the presence of thread sealant and silicone lubricant, both of which are normally used during assembly of solenoid valves and air lines. Air supply grab samples indicated no hydrocarbons present in the instrument air supply, obstru s hm cdo 4 plas Based on all the info & fw.sae Gu.( mation it appears that the EPDM material used in various parts of the solenoid was interfering with solenoid valve movement. Thus, the decision was made to disassemble the dual solenoids on all 8 MSIVs, and refurbish as necessary. 9

t IV. Root Cause The cause of the MSIV delayed clostres has been isolated to a failure of the ASCO dual solenoid valves. This failure is attributed to EPDM elastomer degradation due to elevated temperatures in the vicinity of the air packs resulting from steam leaks. The observed hardened dimples on the dise holder assembly and core assembly hardened elastomer seals is consistent with high temperature conditions. Other evidence of localized steam effects include degradation of the solenoid valve 0-rings and observed rust / moisture discoloration of the 1B21-F022D solenoid coll'. Localized high temperature conditions existed during the plant cycle due to steam leakage and elevated area temperature indications. Steam 'eakage is known to have occurred in MSIV 1B21-F022B packing and the MSIV leakage control system isolation valves. This leakage was in the direct vicinity of those MSIV's which exhibited slow closure. Steam in excess of 300 degrees F is suspected of leaking in the direct location of the subject MSIV air packs based upon the degredation of the EPDM. V. Corrective Actions The following evaluations and actions have been or vill be completed prior to plant startup: 1. For the dual (fast closure) solenoid the total air pack vill be replaced for the 1B21-F028D valve, and the whole dual solenoid vill be replaced on the 1B21-F022D valve. No other solenoids shoved significant degradation or required replacement. All of the other MSIV dual solenoids have been rebuilt. 2. For the single (slow closure) solenoid the solenoid vill be replaced on the 1B21-F028D, since the whole air pack is being replaced. Based on the inspection results above, no other replacements were necessary. 3. A evaluation has been performed of other ASCO solenoid Class 1E harsh environment applications in the plant, including those which may have been subject to the steam leak environment which affected the MSIV solenoids. The review identified two normally deenergized solenoids which do not serve an active safety function. Vork history review of all other applications has shown no solenoid failures. 4. An evaluation vill be made of other equipment in the vicinity of the 1B21-F022D, 1B21-F028D, and 1821-F0288 valves, to assess any impact that the' steam leaks may have had on these components. 5. Additional temporary temperature monitoring v'11 be installed in the steam tunnel on the preselected sample points in the MSIV area including the dual and test solenoid bodies. This monitoring vill be used to evaluate the actual temperature profile of the complete MSIV actuator assembly and the surrounding area. Following completion of the Startup Test Program, temporary temperature indication vill also be installed in the dryvell for monitoring of the inboard MSIVs. 9 . - _.._,,-.,..- - ~ __,_,

l \\ The following additional evaluations and actions vill be performed 1. Further evaluation vill be performed on the exsisting industry experience and efforts on ASCO solencid valve failure investigations. This evaluation vill include such areas as using different metal, and non-metal materials, and the effect of hydrocarbons. Possible design improvements, including an exhaust port screen vill be evaluated. Based on these evaluations a determination vill be made on future actions including replacement frequencies. 2. A sampling plan for the solenoid elastomer components vill be established. Analyses of these components are expected to confirm that hydrocarbons did not contribute to the EPDM degradation. Dev point and particulate sampling of the instrument air system vill continue at the extsting test frequency. A preventive maintenance requirement vill be established for periodic replacement of the instrument air system prefilters. The maintenance frequency vill be consistent with replacement of the instrument air system after filters. Additionally a generic precaution vill be added into sir system vork orders regarding the use of thread lubricants and sealants. An evaluation vill also be made of the relative physical location of the air compressors reduction gear vents, and the compressor air intake, to determine the need for modification, and/or periodic replacement of the intake filter. 3. Until the first refueling outage the full closure dual solenoids vill be checked for proper operation during the monthly slow closure check. This vill be performed by fully closing each MSIV individually utilizing the test solenoid, followed by taking the control switch to close, thus verifying the proper operation of the dual solenoid. Also during this time frame the MSIVs vill be cycled individually on a quarterly basis regardless of plant operating conditions, and the fast closure time verified. On an interval not to exceed six months an inspection vill be performed \\,93 on a dual solenoid during an outage of opportunity. This (kg$3 inspection vill verify no degradation of the solenoid valve internals. i

ff,t l

1

ev 0 * ' O i l l 4 a 3, { w %e I w w, a i. a s o s : m 4.p y g. n u. c. p g :tfr., 7..: r. 7 ;.p, g j, y 7.; .jbf.. . w<.; A '.e .. - -~ .c ' ~+e,J [s , i.. g-3 ','.'.,:. / '.. s ...,,. - u ' t -....f (h',)..% { .'s. S. ", %R,r i )4.r. ' y sg.g. Jo, :- ' .c,,. ,a - r ... s ,. 4.I n '4 h.

• g.[ y w.,,-.

'&.f, - r.y. s. '-. e: $ r. 1 . ~.,u m s w w...- g 4 :,. ; ,f.. n a'Wh,h'Q{p,tf~~ [ .n '} n. c /aw. K,f. -gg ~M u;t Q,$..ym", 4 - r 3 m..M - 4.C Q%.a :g gfg +amy,aw o w* 4,14 ny.., a w p M r.1 #, . 1 .s ,*i-! T, ~s )ti.e,. r k.,h - $~ Y f ff*. ".3c j f;:, h _ k,[P..d ?'- n.g.W,M,1. C d %)s,,;,,..,r.,. ;,. 4, q.:.,,. u t. ,.,..,,, g a..- ,.is e ya., v, g. . g.hg;,, ,,.v. Qa. :g.. _. r ; A ' #. v, " .,.,.. :. w, . y. ,..; ~ a. y g.,fi;~y. ..q - *

k...c.

• r.( r..*,

. + +,, .e J' s +.

• .. +

,s ,g l kl.* 4~ .4 4 .. n . j!~ o t' ,..s...,..a es 3 4 4 <1 4 <a e K * :f,; J.s v

• 4

. ~ .' Q l,. ~ :A -..? ,,k t' * :' m 'h,, l K ~ y %. ' '? -i. i a,. ., n..

1.. u v.t.<

...... m., e, w, y;.u,...#..* + ,3, s ~.. gv , ' y, [ * ' f..., s ,., u. _., ~ .., u .m o ,c

gg

• / ; ^.

I a'. [ T,., ;. ', y %2 ',. , 81,., 4, ;,. k

• ' r,v.,3

.s ,f# ' W* In ,. +. ,e '. - y.' sy -...g ' E Y,.,{ r;.o- .l, 5~-..- +,' > ~.l -, s ' =, p.2.. ..Q - 7', .t h. i .. j.,', ( ;

.~, [

. j, '. ", _, ..',. '.,j _'. Q,, L 5 '..., ~ -,,,,'.., ,,t ,'. ? ,,.. L '.$ 4,,..V', , _ = 9 , 3. f., ',.\\ ) ,t 3 T .**r.. .a .g.

• '.y c.I,.,

'e . i. % f, " i,.; ',.- '. f y s d, r* '.g[. r

• 3-.,

g. 'c, & s. - - s,'.1.f :xc-T, . jy.. ', :. 0,, $.h. ' .e A '.II .s .. q 9. s, '.a, wp :.

g..;. f, /,

e p ;.v.",'; Q; J.s".. -. -b:3 r9 ^ i. t,5,, -{. r s.; x g 4-s g, '< s eM y, .t.+. .: u + - T : g -;. yp y k b , d. t . Y M! -).. ky,,.. rem.wg ay......y.,a.g,g 3 x. ... m.. .-m

= MSIV EG VALIDATION n TASK 3. SOLENOID VALVE ' \\/ ENVIROMENTAL (THERMAL ENDURANCE) TEST PLAN PERRY NUCLEAR POWER PLANT I I J t J L I e m fl TEST STAND PICTURE FIG.i A A HW O / u NOVEMBER 23,1987 ___w. --.....,.._,-_,,,,___,,,.,__.,__.n,

MSIV E.Q. VALIDATION TASK 3, 50lm0ID VALVE ENVIRONMENTAL (THERMAL ENDURANCE) TEST PLAN POR PERRY NUCLEAR POWER PLANT NOVEMBER 23, 1987 fl. 'd.l 67 Prepared by ry Elam n & hdp Yb.[fh W k

1. -J 3-8 7 Prepared by A Ed ie Thomis M !I O 73 d Reviewed by t

I Stuart y tchfield ' ' ~ Approved by

• bW a

Ken Rathen

MSIV EO Validation Task 3, Solenoid Valve Page i November 23, 1987 TABLE OF CONTENTS I.ag Section 1.0 SCOPE 1 2.0 PURPOSE 1 3.0 REFERENLES 1 i 4.0 TEST EQUIPMENT 4.1

• est Specimen Description 1

4.2 Test Specimen Marking and Labeling 2 l 4.3 Air Supply 2 4.4 Pover Supply 2 4.5 Instrumentation 2 4.5.1 Measurements and Tolerances 2 50 ACCEPTANCE CRITSRIA 3 f.0 TEST PROGRAM 6.1 Test' Sequence 3 6.2 Baseline Tests 3 6.2.1 Visual Inspection 3 6.2.2 Functional Tests 3 6.2.2.1 Leakage Measurement 4 6.3 Thermal Endurance Test 4 6.4 Post-thermal Functional Test 5 6.5 Test Documentation 5 7.0 POST-TEST INSPECTION 5 8.0 QUALITY ASSURANCE 5 8.1 Notices of Anomaly 6 Tables 1. Oven Temperature Requirements 7 i \\

i MSIV 20 Validation Task 3, Solenoid Valve j Page li i November 23, 1987 Figures 1. Test Stand 8 2. Thermocouple Mrunting Instructions 9 3. Thermal Endurance Test Setup 10 4. Le'akage Tes'; Setup 11 5. Typical Pressure Decay Plot 12 Appendices I. NP 8323 A20E Dravings, Installation, and Maintenance 13 Instructions II. NP 8323 A20E Temperature Data 19 1 l 1 i

MSIV EQ Validation Task 3, Solenoid Valve Page 1 November 23, 1987 1.0 SCOPE This test plan describes the requirements, procedures, and acceptance criteria for thermal endurance testing of ASCO NP 8323 A20E solenoid valves for use as MSIV pilot valves at the Ferry Nuclear Pover Plant (PNPP). This environmental test is required per reference 3.1, page 4, item 1. 1 2.0 PURPOSE The purpose of this program is to demonstrate that the solenoid valves vill perform their safety function at a variety of elevated temperatures for a specific length of time in a continuously ener-gized state. The ASCO pilot valve's safety function is to vent air pressure and close off supply air from the two-vay and four-vay Norgren air control valves which causes the 20-inch air cylinder to extend and thus close the HSIV.

3.0 REFERENCES

3.1 Letter PY-CEI/01E-0289L November 13, 1987 Murray R. Edelman (CEI) to A. Bert Davis (NRC, Region III). 3.2 ASCO Draving FV 236843, July 10, 1987, Maximum Temperature of Coil and Critical Elastomers with Valve Energized and No Flov. 3. Maintenance and Instruction Manual for Model SA-A068 Valve Actuator, Ralph A. Hiller Company. 3.4 ASCO Bulletin 8323, Installation and Maintenance Instructions, form V5972R1, 1981. 4.0 TEST EQUIPMFXI 4.1 Test Specimen Description Six new ASCO NP 8323 A20E solenoid valves shall be utilized for testing and shall be designated as test specimens 1 through 6. The valve description is as follows (see Appendix I): Three-way, normally closed valve. Dual solenoids, 120 VAC each. EPDM seats and elastomers. Individually serialized.

MSIV EQ Validation Task 3, solenoid Valve Page 2 November 23, 1987 The valves shall be supplied by The Cleveland Electric Illuminating Company (CEI) complete with pipe fittings installed in the solenoid housing conduit connection and a conduit seal of Bisco sealant. The pipe fittings and conduit seal shall be representative of the valves installed on the MSIVs at PNPP. 4.2 Test Specimen Marking and Labeling At receipt inspection, each specimen vill be labeled with en aluminum or stainless steel tag indicating specimen number. The test lab shall also keep records of the valve serial number versus specimen number. 4.3 Air Supply The compressed air used for this testing shell be clean, dry, and free of hydrocarbons as follows: maximum particle size, 40 micront dev point, 40 degrees Farenheight; and hydrocarbon content, less than 1 parts per million. A 50 micron filter shall be installed up-stream of the test specimens. The compressed air shall be sampled i for dev point and hydrocarbons prior to testing. 4.4 Power Supply The specimens '.1 be supplied with 120 + 5 - 0 VAC uninterruptable pover. 4.5 Instrumentation All test equipment and instrumentation to be used in the performance of this test program must be calibrated in accordance with the test facility's OA program which conforms to ANSI N45.2,10CFR50/ Appendix B 10CFR21, and Military Standard MIL-C-45662A. Standards used in performing all calibrations shall be traceable to the National Bureau of Standards. 4.5.1 Measurements and Tolerances Instrument Tolerance Pressure Transducer + 3 psi Thermocouple (Type K) {4*F Multimeter + 2% Note: Whenever possible, use instrumentation within middle third of specified range. 4

---

,-.,-..,_n,

MSIV EO Validation Task 3, Solenoid Valve Page 3 November 23, 1987 5.0 ACCEPTANCE CRITERI A The valves shall operate upon demand without sticking or binding. When do-energized, the valves shall vent the 27-cubic-inch air tank from 90 to 30 psig in 2.0 seconds or less. Otherwise, the data collected vill be analyzed by CEI personnel in accordance with the requirement for the MSIV to fast close within 5 seconds. 6.0 TEST PROGRAM 6.1 Test Sequence Testing shall be conducted in the following sequence: Baseline Tests Thermal Endurance Test Post-thermal Test Disassembly and Inspection 6.2 Baseline Tests, 6.2.1 Visual Inspection j i The valves shall be visually inspected and photographed. Any I damage shall be noted and reported to the CEI representative. Verify specimens are properly tagged. 6.2.2 Punctional Tests Configure the specimens as shovn in Figures 1, 2, and 3, except do not heat the oven. Clean and blov out all air system components prior to test to reduce the risk of contat-ination entering the specimens. Consult Appendix I for installation requirements. Energize both solenoid coils for each specimen, allowing the air tanks to pressurize to 90 to 100 psig. Measure the voltage, in-rush current, and the steady-state current for each solenoid. Allow the specimens to remain energized four hours in order to stabilize their temperatures. Record the specimen and ambient temperatures (Figure 2) hourly. Measure leakage at the vent in accordance with paragraph 6.2.2.1. De-energize both solenoids together and plot the tank's pressure decay curve versus solenoid svitch signal. The tank should be exhausted in 2 to 4 seconds. Time shall be sensured to within 0.05 seconds and pressure shall be accurate to 1 3 psi.

MSIV EQ Validation Task 3 Solenoid Valve Page 4 November 23, 1987 i Immediately folloving the initial de-energization, repeat the above process twice in the following sequence: 1. Energize both solenoids. Measure voltage, in-rush current, and steady-state current for each solenoid. 2. Perform leakage measurement at vent per paragraph 6.2.2.1. 3. At 30 minutes after energization (step 1), de-energize both solenoids. Plot pressure decay versus solenoid switch signal. If a specimen exhibits leakage greater than 1 cc per minute, that specimen must be repaired or replaced and the baseline test repeated. 6.2.2.1 Leakaze Measurement Leakage measurements are performed as an indicator of specimen degradation. Leakage measurements shall be performed using an inverted graduated cylinder as shovn in Figure 4. Leakage shall be measured in el per minute (cc/ min) of air and shall be of five-minute duration. Leakage measurements are for information only; therefore, calibrated equipment for leakage measurements is not required. 6.3 Thermal Endurance Test The six test specimens shall be tested in three groups of two specimens each per Table 1. Each set of two specimens vill be placed in a separate oven. The oven temperature vill be maintained continuously for 92 days. Specimen and even temperatures (Figure 2) shall be recorded hourly. The air pressure to the specimens shall be maintained at 90 to 100 psig during the entire 92-day test. The solenoids vill remain energized at 120 + 5 - 0 VAC during the entire 92-day test except for de-energitation during cycle tests. The odd-numbered specimens vill be cycled (de-energized /energised) every 30 days to simulate the monthly surveillance tests being conducted at PNPP. Specimen 4 vill be cycled (de-energized / energized) at 42 days and once every seven days thereafter. Specimens 2 and 6 vill not be cycled (de-energized / energized) until after the 92nd day. l 4 - - - - - - - -, - -, -, - -,, - - - - ~, -

MSIV EQ Validation Task 3 Solenoid Valve Page 5 November 23, 1987 Air tank pressure decay plots vill be generated for each de-energization as described in paragraph 6.2.2. Perform a leakage test per paragraph 6.2.2.1 on each specimen each vorking day (Honday through Friday) and after each de-energization/energization cycle. At the end of the 92 days, notify CEI for approval to proceed to paragraph 6.4, Post-thermal Functional Test. If a specimen f ails to operste upon demand or othervise exhibits anomalous behavior, contact the CEI representative for disposition. 6.4 Post-thermal Functional Test At the conclusion of the testing specified in paragraph 6.3, turn off the ovens, open the oven doors, allow the ovens to cool to room ambient conditions, and repeat the baseline tests specified in paragraph 6.2. 6.5 Test Documentation A test log and data sheets shall be maintained to provide a complete history of testing. Each log sheet and data sheet shall be identi-fled by the project number and shall be signed and dated by the person making the measurements. Interim test results vill be provided to CEI biveekly. CEI must approve the data sheets prior to testing. A final report vill be issued providing a complete description of the requirements, procedures, and results associated with the test program. 7.0 POST-TEST INSPECTION At the conclusion of testing, the test specimens vill be disassembled, inspected, and photographed. In addition to overall specimen photos, closeup photos of the EPDM discs and seals and their mating metal parts vill be made. The use of oil, lubricant, or any potential contaminant is prohibited during the disassembly and inspection. All personnel handling the valve parts shall wear clean rubber gloves. Individual valve parts vill be labeled and stored in plastic bags. All specimens are to be returned to CEI for possible further chemical and physical analysis. 8.0 OUALITY ASSURANCE The test program vill be conducted in accordance with the test f acility's OA program which shall meet the requirements of ANSI N45.2,10CFR50/ Appendix B, and 10CFR21.

MSIV EQ Validation Task 3. Solenoid Valve i Page 6 November 23, 1987 8.1 Notices of Anomaly If, at any time during the test program, an anomaly occurs, CEI vill be notified by telephone within 24 hours and in writing within five working days. The details of the anomaly shall be fully described including date, time, duration, etc. ) 1 1 1 i i l 1 .__ - - -._..-_ __ - - _ -_-. _-._.~.-

i MSIV EQ Validation Task 3, solenoid Valve Page 7 November 23, 1987 i TABLE 1 OVEN TEMPERATURE REQUIREMENTS Oven Specimens Temperature ('F) Comments 1 and 2 140 Maximum postulated PNPP ambient for mid-dryvell and steam tunnel (see reference 3.2, Appendix I). l , 3 and 4 190 This oven temperature should produce a ) core disc temperature of 284*F (see i Appendix II). Adjust the oven temper-ature as required to attain TC2 (valve body) equal to 284'F. This simulates the original qualification test aging temperature of 284'T for 42 days with solenoids de-energized (see NEDC-30800). 5 and 6 225 This temperature vill challenge the limitations of the specimens. NOTES: 1. Oven temperatures shall be maintained and recorded every hour within +5 to -0*F of temperature setting. y -w Ef- -m m --y--,, _.-_w y y-- +--,-- ev---

PRESSURE TRANSDUCER FITTING AIR T ANK (TYP.2 PLACES) NOTE: TANK MAY BE MOUNTEC ON REAR Or TEST STA'C m APPROX. 27 IN? " TEST STAND-1/4' STEE. P.A~i (OR EQJIVA.ENT: e y_ n.i n.u 3/8' STAINLESS STEEL TUBING SPECIMEN IN.ET AIR LINE CYLINDER (3/8* TUBE) j PORT [ SPECIMEN VENT LINE \\- ( (3/8' TUBE) SPECIMEN (TYP.2 PLACEE) = NPS323A20E HIGH TEMP. LEAD WIRE N 10E STEEL FLEX JUNCTION BOX CONTAINING CONDUIT r RAYCHEM SPLICES OR TERMINAL BLOCK AS AN ALTERNATE. [ .e E NOTES: 1.SOLENDID *A* MUST BE VERTICAL AND UPRIGHT.

2. CARE SHOULD BE T AKEN TO PREVENT THRE AD SEALANT (RECTORSEAL) FROM ENTERING AIR LINES.

3. TUBING FROM SPECIMEN CYLINDER PORTS TO AIR TANKS AND TUBING FROM SPECIMEN VENT PORTS MUST BE OF UNIFORM SIZE AND LENGHT,WITH GOOD GUALITY BENDS FREE OF CRIMPS OR KINKS, 4.THE TEST LAB WILL SUPPLY ALL MATERI ALS FOR THE TEST STAND WITH THE EXCEPTION OF THE ASCO NP83223A20E TEST SPECIMENS.

THREE TEST STANDS WILL BE REQUIRED, WITH THE TEST LAB MAKING ALL THREE TEST STANDS IDENTICAL. FIG.i TEST STAND W

1 TC fi j i COIL A HOUSING ASCO NP 8323A20E g i J SOLEN 0IO A ) k ] c y TC #2 BODY VLE-(APPLY TO FLAT SPOT)- BOOY 1 b ) TC #3 3 d g COIL B HOUSING SOLENDID B 9 ~ I i __ i b TC #4 3 FREE AIR ( APPROX. 6" ( FROM VA VE t t I' NOTES:

i. USE TYPE K THERM 0 COUPLES WITH AT LEAST 14' F ACCURACY

2. SPOT WELD (INTRINSIC METHOD) AS SHOWN ABOVE

3. THE ENVIRONMENTAL OVEN TEMPERATURE SHALL BE CONTROLLED FROM TC #4.

FIG. 2 THERM 0 COUPLE MOUNTING INSTRUCTIONS m

RV SE7 / 9 150 fjg;p33; AIR HOSE PRESSUAE

d

/ RELIEVIN3 4A REGU A1CA i ,r 50 t PT PT FILTEL VE,NT 3 . >=

2 SOLEN 0!C POWEA CIA: T!

M', r

1. 120 VA:

AIR AIR 9 0 TAN < TANK mmmmama / L L SPEC.2 ~ SW. / O O f% g i -- e 3 /% U L' VENT kl WT i A -0 {PNNNNAV7AVNINNW AV_ J W' .[ f%, c N} W< 'w \\ SPEC.1 Sm. O / - FOR ED AIR TEST STAND OVEN o o 120 VAC L J EN Sog0wER j NOTES: CIROVITS

1. VENT LINES TO BE SAME SIZE & LENGTH.

2. CONNECT SWITCH INDICATION AND PT'S TO PLOTTER TO RECORD SWITCH VS. PRESSURE DECAY.

3. SHROUD THE SOLEN 0 IDS FROM AIR ORAFTS.

FIG.3 THERMAL ENDURANCE TEST SET-UP l ~ pt g

FROM SDLENDID VENT PORT VENT TUBE (3/8'DI A.) INVERTED GRADUATED ~ [ CYLINDER FILLED WITH WATE; 4 0 O O M ~ O N O-k- m M BEAKER A m s FIG. 4 LEAKAGE TEST SET UP -~~ - j ~~ ~~

.I ~ ( t m i i W W G D mO (n Ow (A O W a N c c. g E O k )f U >- H< E O E W h C LD o H 4 LL W ~U I W s80 dl e O

• =

kkO OMI {W DU <E 5t -m a2 h EW -a 4 Q. ZO

MSIV EQ Validation Task 3 Solenoid valve Page 12 Novemb-- 23, 1987 APPENDIX I NP 8323 A20E DRAVINGS, INSTALLATION, AND MAINTENANCE INSTRUCTIONS e 1 l

INSTALLATION AND 8 23 M AINTENANCE INSTRUCTIONS l 3-WAY NUCLE AR POWER PLANT SOLENOIDCPER ATED PILOT VALVES NORMALLY CLOSED AND NORMALLY OPEN OPERATION g l 1/4 NPT - 1/16,3/32 AND 1/5 ORIFICES INSTRUMENT AIR SERVICE Form No V69W1 As an addauena) precevuon assinet maltwneuen on starewp.nsolung fror-DESCRIPTION Isrse parueles of pipe scaJe. weld splatter of ether debris in p@e kna these B ul:eun 8323 sals es with Prefn "NP" in the estalos number are 3.u av. teJves hast a large mesh screen (not a idters et injet Thas scenn is not a threct acuns solenoid sabes densned as pueleu pon er plant pdot sabes saboutwte for the straaners or fute=a recommended abose. whose h.neuor, Vahes are of brass constrwetion eith two sadependent solenoids each or 6s to provide contanwous straanant er Ritrauen of the hne fluid both espable of opersums the valve when emersared. **abe elastomers are othylene propylene t$vff 6a "E") for oil free or VITON e (Suffa %"-) for wiping non od fne tastrument att semce. Standard vabes base a %'aterusht. sinns must eoen ply with all appbeable Local and Nauona: E br e tn t.. 1 NE%3 A Type 6 Solenoid Encloevn. Vahes sna) also be eewpped with an Codes. Houangs are prouded with a 1/2 NP5 or ass NPT condmt c: e: ) Explosion. Proof waterusht Solenoid Enclosure which La deugned to sneet son. Connect sving throwth conduit of sutuba evahts for the especto r, j NE.tf A T ype d - E stertsaht. NEntA Tree i fC or D) Hazardows Locauons osvtronunent to a vented electricaJ fuscues bon located an the same area at - Class I. Groups C or D and NEM A Type 9 (E.F or G) Hasardows Loca, the valve. The condwJttlunemon bon erstem should b+ onented meet that tons - Class II. Groups E. F or G. lastaDamon and nialatenance lastrue, eay sceumulated powe"re se LOCA apray wtM nel rus Late W solenoid seas for this solenoid endosure are shown on Feria No. V&as0. enclosure. The waterusht solemoed enclomas may be rotated to sardatate OPE R ATION winns Refer to Form No. V6360 for W method used to route the i NORM ALLY CLOSED (Pnsaure at Commeeuon "2") emplouon proof ts stortaght solenced eneJorun Both Soleno ds De enersaed Dou u from Connection "1" to Connet. NOTE: Alternatias Curnat (A4) and Derett Current (D41 solenoids are budt dtfiervatJy. To esavert Solenoid "3" trorn one to the othas, it u son "3.* mecessary to change the complete soleno6d. EJther or Both Solenoids Energued: Flow is from Consecuon "2"to Con-SOLENOID TEMPE R ATURE NORM ALLY OPEN (Pnmure at Comaecuen "3") Standard catalos vajves are snappbed with cogs dessmed for tenuswows-Both Solenoids De enersued. Flow is from Cotateuen "3" to Comaecuon duty aernee. When the tolenoid W energ1 red for a less penod, the solenoid enclosure becomes hot and een be touched with the ha.ad only for an in- "1." Esther or Both Solenoids Emersaed: now la trom Connecuon "1"to Con. stant. Thas u a safe operauns temperature. Any eseesase heattas usD be led.cated by the smoke and oder of burruas cod har.dauon. secuon "2 7 NOTE: To chanse from normany cJosed to normally open operation. con-MAINTENANCE W AANENG: Turn of f electrical powet espp)7 and depreanuue va)w before sult ASCO. umaklas npaus. It is not necessary to remon the wake troe the pape has i FLOW DI AGR AMS fee repants. CLE ANING A pertodae *aa-af of all soarseid valves is destrable. The tune between NORM ALLY CLOSE D NORMALLY OPEN e6sninne viu vary depensas on snetum and ownee con &uona. La sen-PRES $URE AT 2 PRESSURE AT 3 oval. If the voltase to the eeu ts comet.shassen wasw operanoa.one. san moue or lashase wtU todatate that t)easans as retured. C6saa m)ve serainer ce tuter ' hen cleasans solenoid vabe. PREVENTIVE MAINTENANCE lsp j [ sp j 1. Keep the medauen Downas through thie vajw as frte from dart and for-1 t I sign statenaj as possable. Use taatrument quakt) aar, on) tree f or I t 1 I 1 Suf fa "E " 3 2. w hee in peruce, operate valve penodacaty to insure proper openit:s 3 3 and clostas

3. Penodic taspecuen (dependans upon mesur:: and serwee condiuonss j

of intera4J vahe parts for damass or aseenaat wear u recomer er.ced set le' sM se' Thoroughly cJeta s.U parts. Replace any paru that are u orr. or dam-a a e a I ased. sets f emer se sets sets f emer er sets

4. The vaJves may reestre periodic replacement of the cous and a:: resar Sene es.es scree ss saw+s.s s s eiees1si ont paru dunas thetr unstaued hfe to maataan quahhcauon. The esset De tee'lJ ee seersiref Degee']J ee s tartirrf nptacemeSt peDod wkU depend oO ambsent and ser%et road.tacru.

Spart Paru Esta and Cous an ordend separatelv (see Ordenas Inf or-l snauen). Consult ASCO for spec Sc recommendauons ta conneeuon l

• 'A 'h' " ' L*"* 'I M358 -

1 INSTALLATION Check nameolate for corrtet estalos number.oreerure, voltase sad serwee. IsAPROPER OPER ATION 1. Faulty Control Ctreult: Check the eleetneal eyrtem by energitans the poglTIO N IN G solesond. A metalhe clack sagasbes soienoid is operatias. Absence of Vahe must t e mounted with SoleLoad "A"la a vertataj and upnsht pon' the thek in& cates loss of power supply. Chath for loose or blown <nt son (see Fisure 2)- fuses, open etreuited oe grounded coC. broken lead wtres or apbce eennocuens. GAOUNTING 3. Out Cou: Cheek for open<arealted cod. Replace cou tf pec-For body bou mouattas duneasons refer to Fadure 2. PiptNG

3. Low Voltage: Cheth vehase across the cou leads Volttse must be at least SM of nameplate satsad for AaC and non-bittery operaud D-C Connect pipins or tubtas to valve accordans to anarkiass on valve body.

vaJns. Voltade must be at least 72% of sameplate raums for batterp Refer to Cow daasts.rns provtded. CAVTION: Valws mappbed for od free operated D C valves. Instrum ent aar eerwee are equipped with ethylene propylene elastomers

4. Lacomet Preeman: Cheth va)ve preemas. Preemm to valo murt be wtuch can be attacked by oas and grosaes. Wape the pape threads alaan of within raase spectSed on aseseplate.

outtans oils. This precauuon does not apply to vaWs with Viton alastomers

6. Raseestve Leakase: Dennamenable vthe and clean an parts. Repisce (Sultas "V"La catajos numbers).Ptatas to aD valve porta should be oriented worn or damaased paru witJi a complete Spart Paru Rat for best re-path that any aceuzaulated moisture (parucularly LOC A cheaucaJ sorty) sults' wiD not ester the Laternal areas of the vahe. Foe apputations whert 43 h.aust pipnas is not requatd. lastaD e downward 41rteted street elbow in COIL REPLACEMENT the valve exhaust port. Apply ptpe cosapound spartagly to saale ptM Refer to Foren No. V6340 for ErplostooProof/ Watertight sonenoid En-threads only:if apphed to valve thneda,it may ester the vtJve and cause olosun.

opersuonaj dAfbeulty. Pipe strain se va)w body abould be avo6ded by Tara off sleet:6 cal power anpply and disconneet toD lead wires. For A.C proper mapport and ahrment of ptytas. 5'Isen tashtentas coanteuens. do (Alternatlas Camat) Constreettoe nfer to Figures 3 or 3. For D C mot use vnhe bode or solenced as a lever, t renthes appbed to valve body (D,irec,t Current) Constructon, refer to Farart 1. Proceed ta the fouowan oe pintas are to be located as close as possable to con.neeuon potnt. ,,, y DdPORTANT: For the protectaen of the solesold va)ve,testaD a strainer of

1. Loosen cover screws (3) and somsw cent w th senus, cover gasket alter saltable for the semee tavoked as t:ne talet a6de as tioes to the esJve and nameplate.

as posaitde Fenodie essaams is reestred depedag on service condauona.

2. Umsenw sad remove retalains chp from solenoid base sabeasembly.

See Buueuas 8600 med 8601 for streamers.

3. For A.C Construenon remove yoke ecstaanans sprans waahet, eod i

and tasalauas waahers (2) For D C Construcuon. remove cod washers i eDoones,esmered mdem,k. A S C O Valves A5Co. Form No. V5972R1 PRINTE O IN U.S A. 1981_ Automauc Setch Co. uO-pm - mn mm se ca..e. m em. s amem

g ps.,5._,. ,, s......................i.. ,, r,g n _, !.,i x: = il i i ii ,1'lr,li:ill!- iitill is ei. mtitt 1j,;1 3i v... i i I I:! ! !!!!!!! . j j y,.i c . y ..............:...n... ! !! I !!I i -i gi: N e s's : : e c: !a i! O 7.i-

i o

i,1 2 .) I = II i' n- -c aI li ll 11 ~ f j' h 11 m I i i p. m.m.s_ m,i i. l

1.L.II!! l.L.1,1 illi.

!;!.'If ~ a 4 I .in i.- i .Il .s I 5! - t 17 e P r ' . partas tanful site:Osa Paru marted with am nasanak (e) are p.ppbed in Spare Parts Lu to espieded view prov$ded f ade:uhesuss e'id p':eement tim

8. Caten replectas retaantas the. ushten umul rotatrums ekp ts not spee ORDE RING INFORMATION to r: tate soprontmately 9'32 of an toch between ecnw head and aus.

4. TE# sue coser serens ena;7 to 10 anch pounds (1.1 newton meters)

POR SP ARE PARTS KITS ta nature proper saanet cosapresaaon. l When Orcenas Soare Ports Krts c. Coas CAVT.IO.N: Sen.eseed m.ust be faDy reassembled, es the besand had. blesses l so.cify veNe cowee Numeer, estu s een f end.mpa.u me maemeve eseews. Pteee as nn damas Serial Number, Vois and waaber es ossh end of sec,af veentrod.

• • A A D.

VALVE DIS AS$tMBLY (Re'e' tc F*gur s 1.2 and 3.1 e Depnesunse vain and essa off sleetnea! power supply.

1. Diamaaemble veJve is an orderly laahaoa pertas careM siteston to es; nodes tiews prended ler sdeauhtauen and placemer.t of paru.

2. Starkas mth solenood "A."loosen sever screes and remow cour with screws. cover gasket and maaneplate. For esplomon-proof t waterudht saleme64 eas4emare. refer to tastallauen and blaiatenaste lastrueuena. Forum No. V6380.

3. Unscrew and resnove totataans e!!p from solesold base evbessrseWy.

d. Sbp yoke tentairuas senas washer, cod had lasulatans washere (2) cff the solemond base sabeasembly. NOTE: lasalsung waabers (2) are er.reted when a molded cod na used.

6. Unacu m solemand base mahasembly with special wrench adapter mappbed ta Spare Parta Kat (wrench Adapter Order No. 304 400 8).

4, utmove saleno64 bene subessembly, upper solenoid base snaket, housad, retainer snaket, setatner and sees assembly with oore spetas and een su6de. 1. Hemove bodt tasket troen veJve body.

8. Fer Solenoid -3." loosen eene ocnws and remove cover with eenws.

eover tasket and namaanste.

9. Unacrew and remove rotataans clip from solenotd base sab4asemb!7.

10. Far A C (Altermanas Current) Construeuen. any yoke cosuintas sonns washer coq and tarelattas washers (2) off the solenoad base subseaembly. For D C (Dtreet Current) Construcuen. sup cod washers (2), tasulattas washev. coil and tasulatias waabet off the solemeld base subnasembly. MOTEa 3asulaua4 washers (2) are emitted when a smolded toQ ts used.

11. Umsens solenoid base Nbsasembly with special wTench adapter su;phed ta Spare Parts Lt.

12. 11esteve solenoid base matsasambly, uppet salenoid base gasket, heumad. tessiner saatet, retainer, eore. plusaut snaket, plusaut assem.

bly and stesa sasaged 13 solenoid base sub440esably. Remove adaptar paaket from adapter. adapter and remove asse bolder spring, disc holder sab-

13. vascrew assembly tad b*dy gasket.

14. A.u parts an new seeemable for cleanlag or replacement. Desa aD taternal paasnetware thoroughly before valw roannembly. Rep & ace worn er damased parts with a complete Spare Paru Ot for best remaJu. When tamausas e sempine Spare Parte Lt.It to recomasanded that the cous also be repLseed.

{es_ A

c. Z i"".!M '"

vALvt Rr Assrusty e-oe -e ~.s

1. nesmemhie in rewme.rder.: dammemur. ytas e.nw.tu.uon to exploded views provuled for identiheation and placement of paru.

V'4 " 2. Lybncate all saskets tescept toser staket) with a hsht coat of DOW mis anse

• i CORNINGS 560 Fluso tunnesat (r ppued in Spare Parts Lt).

I

3. Stanans with Solenoid "5." tasu11 body saaket, dise holder sub-asse mbly and adapter. Torque adapter to 116 2 25 Lachtounds va.'s..e. ac e==s tie

[1* 4 2 2.8 newtoa meters). 4. Repspe upper solenoid base saaket on solenoid base sub assembly. " * * "., N '.",N" U 5. Piauon solenoid base sub assembly in houstas and Lasull core. b* (sm all end in tLrst), plugnut gasket, plugnut masembly nad esem .f\\ (ernal! end tato core). w.o

• .us as ess =

6. Repiace retalatt sanket.tettiner. body gasket sad positjon dise bolde: er ee p es s *s* t-sprsns. W '*'4

• 48" '

1 Heldans the solenoid base sub-enerably seeurely, essage the solenoid C'h*,**" base sub-assembly tato the adapter usas spectaj wtenth adapter pro-7 I vided as Spare Fans Eat (% reach Adapter Order No. 304 600 1). ,w w w,y,g Df Tareue solenced base sabeasembly to 176 226 nach-pounda [19.8 2 po.se.e.e e s.u

1. 6

%g* me smae 2.S eewton meters].

• • • "**8""'

S. Far A C (Alternattag Cunest) Construction. Poettion tasulattas o-,7Eso see T washers (one at each end of coU). eet) and sonna waabet ts yoke. Y es a.nes Sha the yoke over the solenoid base sub4aserably. For D4 (Dtreet Curre nt) Constructica, nylaev tasulauas weaner, cou, hasulatlas vo.ese so.pme mass e6 washer and eou weabers (31. a.*

• a "e gy

9. Reptsee reta:anas she and tighten untu retaanlag clip is not tree to

,,,2 * 'wE"*s rotate, approstasately 9/32 of as tach between screw head and aut. u C -s eo-emee sau saast $ s

33. Replace cows saaket and covet with nameplate and screws. Torque cover screws evenJr to 10 tech pounda (1.1 newtoe aseters) to tasure A(;:

proper saaket compreemon. N

11. lasta21 soleno64 bass sasket on solenoid base subsaerably.

12. For Soleno6d "A." posuon solenold base sabeasembly la housias

\\,, ,,0 Z e,%l and tast&D setalaer gaskat retalaer at base of eonesotd base sub-a emwy. C maap awr-$

13. lastan body ensket and poetica eore assembly with tore speing est core rulde sato solemoed base sub4asembly.

p

• • • ' 8 tastall taleased base sabeasembly maas specb! wrench adapter pro.

14. ended la spare Paru Elt. Torque solenoid base sabessembly to 118 2 25 nach-sounds (19.8 2 3.4 newtoa saeters).

15. Postion inaa3 atlas washers (one at each end of eeu) top and spring washer to yoke She yoke over solenold base sub nasembly.

18. Replete retalains thy and tighten until retaaning clip ta act tree to route, approatmaul; 9132 of as tach between eenw band and out.

17. Replace cover saaket and covet with massepiste and actews. Tertue towt eenws evenly to 10 tach-pounds (1.1 aswton metersi to tarure proper sasket coves 4eeden.

18. After malatemance, operste the va)ve a few times to be sure of proper operation.

ab. SaDeesa 3 42 3 - Partial V6ews o c <D en es,nnii C.-ree A S C O Valyes rare 1. A5Co,, dui0lT10ijC Switch Co. no-ammimm mn *- vana'

• aiutio '~ u.s A.

$..t ? Aammrlm $cithCa es ...... i.m.. n

TORCUtChui 7%RQM VAm TORQUE VAm FART M Abgg INCH PSL*ND5 NtuTON METE AS Soksoid Base j t I##""" !!& : 25 It 6 : 2 6 8 A dapter Ceee Screws 10 { 1.1 l f ^A COVE R WITM $C Rtwi t3'. A*

• f N AWEPLATE ATT ACMEO TIGHTIN VNTIL RtT AINING CLIP

'l l$ N0* F RE E TO ROTATE Qfp/ I APPROXiv Att Ly s'37' tETWE E N J/ $C8t A NE AD AND NUT) Covt R GAsut1

• P ART $ steCLUDED IN RETAININO CLIP SPARE P ARTS KIT 5$

g 3 6 'N C $PRING % A5Mt m

• I
• /

s (CM 10MITTID WHEN WCit tD D m!vt ATiNG w Asat s C0iLisustDi $0 LINO:D B ASE $UB AS$1V8LY % l hd Colt b [ -lNSULATING W A$ME R 'gg7[ 10 WITTED WHEN MOLDED gg = Coll 5 UstD1 HOU$ING 'l A 'T 1/1 sors OR 3'8 NPT COND7T CONNf CTiON ig h RtTAINt R GASKET RET AINIR $ ,...e D CORE GUIDE $ l i ce s .u CORE ASStWBLY$ i l y MRt $PRING$ 1 ce., .. g g, C BODY GA'Kt7 $ l p p...,

• g,.
  • -a 's ** *: $

R. = SOLENOID T - ' s' COM P LETE LY l* SEW 8 Lt 0 $PECI AL WRENCH ADAPTER $ \\ Cg FOR $0LENotD B Ast yS s gy $US A55tW8LY I g E 'a i (ORDE R No. 20&4DMI >=inCMts=e4 y I hb hh I \\ / k-i h t v r Twolt) WOUNTING HOLES S/IE @,t] DEEP FOR NO a THREAD CUTTING SCREW. saneena sans - Parual vbew showtas senemow "A." wstarusht sownow teelemar. shows. For Esplomoe#roef twaterucht Sokno64 Entleeus.see Fosse No.V4M0. Fwure 2. C Auiomen 6Cth (6 *** * *<, 's etm.t;

TOR QL'E CH ART TORQM VAWE TCRQUE VAWE PART NAME IN C H.POUN D S NEWTON SIETER$ Sol.noid Base 54 Ammta STS :ll !.B : 2.6 A 4.ete r l Cevet Seews to 1.1 / 4*' 50LINotD A" C0VP i*f LY ASSEVILED L [ () -BODY G ASKE7% DISC HOLDie $US ASSEMSLY$ DISC HOLDE R $PRING% PARTS INCLUD E D IN EP A.R E P A PT5 Ki 5 $ AD APTE R j () AD APTE R G ASKET $ 'l STEM $ ($ MALL END INTO Cop ti PLUCNUT A$$tM8LY $ PLUGNUT GASKET $ COR E $ SPtCI AL wmENCH ADAPT!R$ / (SMALL E ND DOWN1 FOR $0LIN010 8 ASE rf"2 RET AiN t R* sus ustusay (ORDIR AO. 206-400 ti n RET AINE R G ASKET$ i 'NU MOU$1NG s E 1/2 NPS OR 14 NPT CONDulf CONNECTION C unea soLiNm0 B ASE G ASKET $ TIGHTEN UNTIL 8'ETAINiNC Clip 15 NOT PRif 50 ROTATE. ( APPROXIM ATILY $ W' 8ttw E E N 50LIN0'D B Ast SCREW ME AD AND NUT) J $USASSEW8LY$ D YOKE 'v, N ai=~>~o cup

• c

,N,UaT,No.o 1 c,. M (OMITTED WHEN MOLDED CoiLl$ U51DI 4 gg CovlR GASKET c' Coit s ~ INSULATING W ASH E R s Ch 10 WITTED WHEN MOLDE D + COVE R WITM $ CREWS (3) G PRING W ASHI R AND N AvtPLATE ATTACMED / r., ..i.....Pr,.3 2 3 - Partial V6ew Showies son.no64 "5." SaDesia 0 W t rusht 3 14. 64 E ea.eure sise.w.s. tw..r ai s i... .o .. r.. w.. v.... r.... A S C O Valves g-Automuttt Switch Co. no Miu,AR K. ~io nii, m> e-a.. vimRi eai~TEO i~ u A 2, t Amamctit SGth Ca. m e.,s.esi..ie

I MSIV EQ Validation Task 3, Solenoid Valve Page 18 November 23, 1987 APPENDIX II NP 8323 A20E TEMPERATURE DATA O

Yr) cATAtoca NUMBER NP 8323 A2or Ac/Ac 1 ELENotD ' ' A j D / / ~ Colt W J c CORE OtSC g-LOWER DISC. i r g Colk w 1 i 1 I SOLENoto'B' ISI*C 219'C 2ET 'C lTC*c 170 *C GG*C ( (,5

• C 175 *C IEl*C i15'c 49'C 147'c f (so *C 108'c 99'c 25*C 12G'c 141'C 86 *C 77 *C AMBIENT COIL SQL. *A'
• Coll-bCL.'S*4 CORE. DISC LOWER DISC TEMR A%%INUM TEMPERATVRE i

I I I um 6 't. Aposo Fee M $9ef" I 1 IlII '~ i i i MAXIMVM TEMPERATURES OF Colt AWD CRITICAL ELASTOMERS WITH \\ALVE ENERGIZED AND NO FLOW sa no. av, cars 4" o "' O "" O U "" U wANvf ActuniNo totEmAmets 10 E

• • • ww t9NS 7-M7 at w AccomoANCE witd ASCO E

2(0651 ca O av daa O ** Ol.n N 8" " " ( l peeJ. aw. scata assem. ase. no as O d O d _ c n.c m o cf.V. 7737 . nary w mamanc a.,cm m m M.usett.as m..M.. - \\/ 2Md l sit v C#t m a on. .m. a i,0-3c5-Who Automuttt Switch Ca.. -r a ces,w, no....... .mn m. urri. t

APP. II TITLE ptoT F CORE DISC TDF. VS. Me. MP. 400 - C 300 - ) i W t:HECKPOINTS B 200 - 100 - DATA FROM ASCO DWG. FV236843 AMB. TEMP. CORE DISC TEMP. 77 190 120 226 151 250 268 349 I i i 0 100 200 300 Abe. TDR N.JE (*F) ~-- - - - - - + - - -

PREllMINARY NOTIFICATION OF EVENT OR UNUSUAL OCCURRENCE--PHO-Ill-87-138 Date November 3, 1987 This preliminary notification constitutes EARLY notice of events of POSSIBLE safety or public interest significance. The information is as initially received without verification or evaluation, and is basically all that is known by the Region !!! staff on this date. Facility: Cleveland Electric illuminating Licensee Emergency Classification: Company Notification of an Unusual Event Perry Alert Perry, OH 44081 Site Area Emergency General Emergency Docket No. 50-440 X Not Applicable

Subject:

SHUTDOWN BECAUSE OF EXCESSIVE MSIV CLOSURE TIMES At 1:38 p.m. (EST) on Nov' ember 3,1987, the licensee initiated an orderly reactor shutdown fromg3 percent power after two Main Steam Isolation Valves (MISYs) failed to close during testing within the required Technical Specification limits. The remaining six MSIVs closed within the 3 to 5 second requirement. There are two valves in se'quences'on each of the four d 3 Main Steam Lines, one valve inside the reactor containment and one valve outside the containment. The valves that did not meet the closure time were_both on the "D" steam line'-- the inboapd valveclosedin18secondsandtheoutboardvalvehadnotclosedwithinthetwominutes%e control switch was in the close position. The tests were observed by the Resident ~ 1nspectors. 'l Subsequent tests of the two valves resulted in closure times within the Technical Specification limit. During testing on October 29, three valves did not meet the closure limit -- the two valves on the "D" line and the outboard valve on the "B" line. Subseq'uent cycling of the valves provided acceptabis time respons's. (M l An Augmented Inspection Team is eing dispatched to the plant site to review the circumstances a and possible causes of the H51V closure problems. The team will consist of the resident inspectors and personnel from Region 111 (Chicago) and the Office of Nuclear Reactor Regulation. A Region III Supervisor will head the team. Region III will issue a Confirmatory Action Letter to the licensee documenting the licensee's agreement not to resume operation of the plant without concurrence of the Regional Administrator. The State of Ohio will be notified. RegionIIIwasinformedofthetestresultsandshutdownat1:15p.m.(EST)bythelicensee. This information is current as of 2:30 p.m. (EST). CONTACT: R. Knop (FTS 388-b547) M. Ring (FTS 388-5602) DISTRIBUTION: H. St. EDO NRR E/W Willste Mail: ADM:DM8 Chairman Zech PA IE NMSS DOT:Trans only Com. Roberts ELD OIA RES Comm. Bernthal AEOD NRC Ops Ctr Comm. Carr Com. Rog rs 4 8y ACRS SP Regional Offices SECY } CA INPO NSAC PDR Rl!! Resident Office icensee: (Corp. Office - Reactor Lic. Only) / ,im W . Dan 4nn 117 y

CONFIRMATORY ACTION LETTER f p* *lco UNITED STATES y f C, NUCLEAR REGULATORY COMMisslON e( REGION lil 5* E 70s moostvtLT noAD 'g CLEN ELLvN, ILLit oss son st NOV 4 19 0 Docket No. 50-440 Docket No. 50-441 The Cleveland Electric Illuminating Company ATTN: Mr. Murray R. Edelman Vice President Nuclear Group Post Office Box 5000 Cleveland, OH 44101 Gentlemen: l This letter confinns the telephone conversation on November 3,1987, between i Mr. Greenman and others of this office and Mr. A. Kaplan of your staff regarding the Main Steam Isolation Valve (MSIV) failures occurring at the Perry Nuclear Power Plant Unit 1 on November 3,1987. With regard to the matters discussed, we understand that you will: 1. Take those actions necessary to ensure that complete documentary evidence of the "as found" condition of equipment being inspected is maintained. j 1 2. Provide a step by step troubleshooting program to establish the root cause of the MSIVs failure to meet acceptance criteria. 3. Not disturb any components that offer a potential for being the root cause including power sources, switches, solenoids, and the air system directly feeding the MISVs until that action is approved by the NRC AIT team leader. 4. Except as dictated by plant safety, advise the NRC AIT Leader prior to conducting any troubleshooting activities. Such notification should i be provided soon enough to allow time for the team leader to assign an inspector to observe activities. 5. Submit to NRC Region III a fonnal report of your findings and conclusions within 30 days of receipt of this letter. None of these actions should be construed to take precedence over actions which you feel necessary to ensure plant and personnel safety. We also understand that Perry Nuclear Power Plant Unit I will not be made eqitical without the concurrence of the Region III Regional Administrator or hi!Niesignee. s 's /r CONFIRMATORY ACTION LETTER i v

m

1 ~ CONFIRMATORY ACTION LETTER The Cleveland Electric Illuminating 2 NOV 4 1967 Company Please let me know imediately if your understanding differs from that set out above. Sincerely, A. Bert Davis Regional Administrator cc: F. R. Stead, Manager, Perry Plant Technical Department M. D. Lyster, Manager, Perry Plant Operations Department Ms. E. M. Buzzelli, General Supervising Engineer, Licensing and Compliance Section DCD/DCB (RIDS) Licensing Fee Management Branch Resident inspector, RIII Harold W. Kohn, Ohio EPA Terry J. Lodge, Esq. James W. Harris, State of Ohio { Robert M. Qut'lin, Ohio Department et Health State of Ohio, Fablic Utilities Con ission J. M. Taylor, DELD T. E. Murley, NRR J. Lieberman, OE R. Cooper, ED0 W. Lanning, NRR F. Miraglia, NRR G. Holahan, NRR M. Virgilio, NRR J. Partlow, NRR K. Connaughton, SRI J. Strasma, RIII CONFIRMATORY ACTION LETTER

1 . ~..., o. 6. oo P.2 l 1.:]F, h 1. t. i['i,[ 0 '# S.'

. f$." r F15 C i L. '. n :! XI k ' F '

%U ;5 K7" P.O SOX s7 e PERRY. CHIC 44001 e TELEPHONE (21s) 260 3737 a AoDREst to Murray R. Edelar.n Serving The Best locotton in the Nation a vxx m aoewt PERRY NUCLEAR POWER PLANT hVQ441 Novembe r 13, 1987 PY-CEI/01E-0289 L Mr. A. Bert Davis Regional Administrator, Region III U.S. Nuclear Regulatory Commission 799 Roosevelt Road Glen Ellyn. Illinois 60137 Perry Nuclear Power Plant Docket No. 50-440 Updated Information to ) i letter FY-CEI/OIE-0288 L Dear Mr. Davia t This letter provides additional information regarding the commitments nada in our letter, PY-CEI/01E-0288 L, dated November 9,1967. discussions held with members of you etaff on November Based upon the i 10, 1987, enclosed is a description of committed actione, established parameters to be monitored and appropriate action statements if predetermined threshold values are exceeded. i Following receipt of your concurrence, we plan to restart the plant to complete the remaining tests in the Startup Test Program. If you have any questions, please feel free to call. Very truly yours, o Hurray R. Edelsen Senior Vice President i Nuclear Group HRE: cab Enclosure ce: K. Connaughton T. Colburn sDocument Control Desk N x ~ / W'& ca, 7 ts; t .) ;)

)

i i \\

] M 1 .~. u., 2. 4~.4s P.3 1 t \\ Enclosure Page 1 of 6 The following evaluations and actions have been or vill be completed prior to plant startup: 1. As previously stated in FY-CEI/0IE-0288 L, for the dual (fast closure) solenoids, the total air pack has been replaced for the 1821-F028D valve, and the whole dual solenoid has been replaced on the 1821-F022D valve. Additionally, the 1821-F' eA solenoid valve has been replaced tue to a frayed wire at the termination. No other solenoids showed.ignificant degradation or required replacement. All of the other MSIV dual solenoids have been rebuilt. 2. As previously stated in PY-CEI/0!E-0288 L, the single (slow closure) solenoid was replaced on the 1821-F028D valve since the whole air pack was replaced. Additionclly, the 1821-F0288 solenoid valve has been replaced due to d frayed vire at the termination. Based on the inspection resulta above, no other replacements were necessary. 3. As previously stated in PY-CEI/0II-0288 L, an evaluation has been performed of other ASCO solenoid Class 1E harsh environment applications in the plant, including those which may have been subject to the steas leak environment which affected the HSIV solenoids. The review identified two normally deenergized solenoids which vere subject to the same conditions as the MSIV solenoids. Since the solenoids are in a normally doenergised state, no further action was considerad necessary. The two solenoids were 1821-F0451 (solenoid for valve IB21-F0069) and 1M14-F0063A (solenoid for valve 1M14-F0060A). Vork history review of all other applications has shown no solenoid failures, indicating the ASCO solenoid degradation appears to be limited to the HSIV solenoid valves. Further reviews are described in item 4 of the post startup act!ons. The IM14-F0060A valve is a normally closed valve and has no safety function to mitigate an accident. It is associated with a Dryvell Purge system damper that is closed during normal operation with a water seal in place for shielding purposes. The 1821-F0069 valve is a one inch before seat drain valve that is clcsed at greater than 50% main steam flov (at which time the sole':oid is doenergised). The valve was recently cycled on November 7, 1987 with no deficiencies identified. In addition, this valve vill be cycled again during this plant startup. 4. An evaluation has been performed of other equipment in the vicinity of the 1821-F022D, 1B21-F028D, and 1821-F028B valves, to assess any impact that the steam leaks may have had on other components. This evaluation revealed that there vere six valve actuators in the steam tunnel and two in the dryvell that vere in close proxicity to the knovn steam leaks. These actuators were inspected and no steam / heat degradation was observed. Viring, terminal blocks, torque svitches, limit switches, splices, gaskets, and limit switch gear box lubricants were inspected by a team that included E0 pers3nnel. s war e ---,,.-c,-me.---,.m.w.- --,.c_,, -.-m,---.y. .--,-3 ---,__.-._g.

l s o. .,.6, p,5 s Enclosure Page 2 of 6 There was no evidence of a thereal degradation from a ste".m environment that would affect valve operability. It should be noted that the Limitorque actuators do not contain EPDM material. A reviev for qualified life adjustment vill be included in the further reviev described in the post startup item 4. There is no short term concern of Limitorque motor operators qualified life in the dryvell or steam tunnel areas. 5. The historical rsadings of the existing permanent steam tunnel and dryvell temperature elements in the vicinity of the MSIVs have been reviewed, and a baseline has been determined for each element (see ). Until the temporary temperature monitoring baseline values have been detesmined, the existing permanent temperature. elements vill be used. It has been determined that a 10% rise above these baseline values may be indicative of a localized steam leak and would require investigation. This value was conservatively selected since it is approximately one half of the temperature rise expected for the Technical Specification trip value for leak detection. It is suffielently conservative for the interim period until the MSIV area and surface temporary temperature element readings have been fully baselined. This temperature rise vould have indicated the steam leaks which impacted the inboard MSIV ,(24 degrees P differential temperature) A lover threshold temperature rise could result in unnseessary actions or reduction in pover operation due to minor temperature fluctuations. (5a.) A procedure vill be establishco specifying necessary actions to be taken upon exceeding the interim temperature values. The interim temperature thresholds are, area temperature plus a 10% rise or a selected 225 degrees F for the temporary temperaturs elements in the area surrounding the MSIVs for both the steam tunnel and dryvell. The Senior NRC Resident Inspector vill be notified if any of the following corrective actions are to be taken: Reduce power, as necessary, to perform a visual inspection to o determine the equipment affected, Immediately repair the leakage or shield the adjacent Class 1E o components to limit the impact until a repair is possible. Note components being affected and assess the thermal impact o (EQ). Evaluate and determine the necessary time frame for taking additional action, such as increasing surveillance frequency or changing replacement interval. e ,,,,-,,,,,.-,-,,wy- ,,v,-+-. , - - -, _m,-+ .,y, e 4 -e a.2,--.,,

NOV 13 '87 14: 15 P.6 s Enclosure Page 3 of 6 o At least 1 temporary temperature element in the area of each MSIV will be maintained in service in Operating Conditions 1, 2 and 3. If all temporary temperature elements fail for a specific MSIV, the adjacent temperature elements vill be utilised in an interim period not to exesed 7 days. In the interim a correlation vill be established between the adjacent temperature elements and the specific MSIV vithout individual monitoring. After 7 days, reactor power vill be reduced in order to repair / replace the failed element within 24 hours or the plant vill be placed in Hot Shutdown within 12 hours and Cold Shutdown within the following 24 hours, o If the local temperature monitoring in the area of an MSIV exceeds 284 degrees P, the affected MSIV vill be declared inoperable in a6eordance with Technical Specification 3.6.4.a or cycled daily consistent with the 30 test parameters. This remains in effect until the additional environmental testing is completed (see Attachment 2). 6. Additional steam tunnel temporary temperature monitoring has been installed on the preselected sample points in the MSIV area including on the dual and the test solenoid bodies. Baseline data vill be obtained on the temporary temperature elements in the steam tunnel during the next' full operating period of sufficient duration to allov temperatures to stabilise..From our experience, this vill be several days after the plant is at full power. Until the baseline data is established, a value of 225 degrees F vill be l utilised for the temporary temperature elements in the areas surrounding the MSIV to initiate the actions described in 5s. Inspections vill be performed during startup to assure that the initial temperature reading are not being effected by steam leaks. Once it has been determined that the readings have stabilized, the procedure outlined in item (Sa) above vill be revised to use the temporary temperature elements in lieu of the permanent elements. The temporary temperature monitoring program vill continue until the final analysis results of the environmental testing (see Attachment

2) is fully evaluated. At this time, possible design improvements vill be evaluated and a determination vill be made on future actions, including replacement frequencies or correlation to permanent area temperature elements. The NRC will be notified prior to removal of the temporary temperature elements.

Nine dryvell temporary temperature elements have been installed with at least one on each of the dual solenoids on the inboard MSIVs, typical of what was done with the temporary steam tunnel temperature elements. A baseline vill be established after the startup following the outage as described above for the temporary steam tunnel temperature elements. These baseline values vill then be incorporated into the program, along with the respectiva acceptance criteria. In the interim, a selected threshold of 225 degrees F l vill be used for temperature elements in the area surrounding the inboard MSIVs to initiate the actions described in Sa. r

KDV 13 '87 14 10 P.7 s Enclosure Page 4 of 6 l 7. A test has been performed which verified that air does not flov between the air compressor reduction gear vents and the air compressor intake. Consequently, it was determined that there was no need for any equipment modification, or change in the intake filter replacement frequency. l Following startup,'these additional evaluations and actions vill be performed: 1. To further substantiate the high temperature root cause, laboratory analyses vill be performed to confirm the failure sechanism of the EPDM degradation. A review of industry experiences and discussions with various industry sources vill continue to be conducted in order to input into our analysis plan. Our preliminary analysis plan, which included these industry contacts, is completed, and a summary is provided in Attachaient 2. Ve have completed an initial evaluation of industry experience. The initial industry review did not change our preliminary conclusion that the root cause of the problem was primarily localised elevated temperatures near the ASCO solenoid valves. The visual inspection of the EPDM did not exhibit the normal signs of hydrocarbon degradation (stickiness, sponginess, or svalling), however, we have not eliminated the potential of hydrocarbons having a deleterious effect. Ve plan to use data obtained from other plant experiences as described in IEN 86-57, along with our own analysis, to confirm the root cause, Our preliminary schedule is to have initial infrared analysis for hydrocarbon degradation by the end of January 1988 with the remaining results and analyses by end of the first quarter 1988. I Any further analyses required vill be determined at that time. Ve plan to use a local research laboratory, as our primary analyses contractor. Results vill be provided to the NRC. Vith respect to environmental testing, a test plan vill be provided to the NRC by November 23, 1987. Interim test results vill be provided to the NRC as they become available during the 92 day test duration. Following completion of the analysis program, possible design improvements, vill be evaluated and a determination vill be made on future actions, including replacement frequencies. j 2. Presently, in order to minisine the potential for introducing hydrocarbons to the air system, a preventive maintenance requirement will be established for periodic replacement of the instrument air system profilters. The maintenance frequency vill be consistent with replacement of the instrument air system after filters. Additionally, a generic precaution vill be added into air system work orders regarding the use of thread lubricants and sealants. If the outcor.e of the chemical analyses indicates the presence of hydrocarbons, ve vill immediately implement an appropriate l

t ,an u ar usn P.S I Enclosure Page 5 of 6 hydrocarbon sample and analysis program for the instrument air system. This vill include weekly mampling of the supply lines to the HSIV's at the containment penetration connection as well as other main J-headers throughout the air supply system. The senior NRC Resident Inspector vill be notified upon taplementation of this action. Dev point and particulate sampling of the instrument air system vill continue in accordance with the existing plant administrative procedure. Any unseceptable results vill be evaluated and system blevdowns vill be conducted until satisfactory results are obtained. Until the first refueling outage, the fast closure dual solenoids 3. vill be checked for proper operation during the monthly slow closure check. The existing monthly surveillance instruction vill be revised prior to startup to reflect the following test procedure. The test vill be performed by fully closing each MSIV individually utilizing the test solenoid, followed by taking the control svitch to close. Perforr.ance of this test vill verify the proper operation of the dual solenoid, since the HSIV vill only remain closed if the dual solenoid deenergizes and properly repositions. If any MSIV should reopen during the test, indicating failure of a dual solenoid, the associated MSIV vill be declared inoperable and the plant vill be placed in Hot Shutdovn within 12 hours and Cold Shutdovn within the following 24 hours. The NRC vill be notified upon discovery of such a failure. Also during this time frame the MSIVs vill be cycled individually on a quarterly basis regardless of plant operating conditions, an3 the fast closure time verified. As a result of a failure of this quarterly test due to temperature related problem with a dual solenoid, or other air pack component, the plant vill be shutdovn and the NRC vill be notified as described above. The sonthly test described above, vill not be performed during those months when the quarterly fast closure test is performed. Prior to exceeding a six month period an inspection vill be performed during an outage of opportunity, on the' dual solenoid experiencing the highest temperature profile. This inspection vill verify no degradation of the solenoid valve internals. ~If accelerated heat degradation is observed, a complete investigation vill be initiated and the NRC notified. l 9 - - - - - - - - - - - - - - - - - - ~ - ~ ~ ~ ' ' ~ ~ ' - ~ ' - ~ ~ - ' ~ ~

i .~.,, e. .,4, Pd 2 s Enclosure Page 6 of 6 4. A reviev has been completed of all known steam leaks in the plant which could have affected Class 1E equipment. For all of the Potentially affected equipment identified, there is no configuration where elsstomer compression set or degradation could result in the equipment not being able to perfort,its intended function.

However, these components vill be evaluated to untermine if there has been any affect on their long tera qualified life based on the environment under which they were subjected. The results of this evaluation vill be completed and submitted to the NRC by November 30, 1987. A further reviev vill be conducted for pr,rentially high temperature area environments of all Class lE solenoids and other equipment vith EPDM subcomponents where elastomer compression i not being able to perform its intended function. set or degradation could res This reviev vill be completed by the end of the first quarter 1988.

l l l ,y.-y,m --a, .--.- ---,,,-- --, - - - -. - -..,.,-- -, =, - -, - --

P9 TEMPERATURE MONITORING FOR DETECTION OF STEAM LEAKS TEMPERATURE NORMAL SENSOR OPERATIONAL ACTION PLAN NUMBER IMPLEMENTATION TE RE TEMPERATURE.- D23 K102 A 140'F 154'F UPPER D23 K102 8 140' F 154' F DRYWELL M13 R110 2 150'F 165'F AREA M13 R11016 135'F 148'F D23 K112 A 135'F 148' F D23 K112 B 131'F 144*F MIDDLE M13 R110 3 136'F 150' F DRYWELL M13 R110-4 124' F 136' F AREA M13 R11014 136*F 150' F M13 R110-15 127*F 140'F D23 K122 A 130'F 143'F D23 K122 8 128'F LOWER M13 R110 5 141' F 114'F 125'F DRYWELL M13 R110 7 122'F AREA M13 R110 8 134'F 122'F M13 R11011 134' F 110'F 121*F M13 R11012 127' F 140'F STEAM E31 N604 A 125' F 138'F TUNNEL E31 N604 8 134' F AREA E31 N604 C 147'F 130' F 143' F MONITORS E31-N604 D 128'F 141'F STEAM E31 N605 A 80' F 88*F TUNNEL E31 N605 B 80'F 88*F DELTA T E31 N605 C 82' F 90'F MONITORS E31-N605 D 82*F 90'F F - _ _ -.. - - _, _. _ _., _ _ _ -,.... -. _ _ _. - - _ _.. -, _., _ _.. _, _ _, _.., _ _ - _. _ _ _.. - - - _ _ -. - - _. _ _, _... ~. _ _ _. _ - - - -

P.10 \\ ANALYSIS PLAN FOR RPDN SOLEN 0ID COMPONENTS I. INTRODUCTION To determine the'cause for failure of solenoid pilot volves which resulted in the slow closing of M8IV85, two approaches ' vill be taken. Both approaches involve analyses of the IPDM elastomer gasket asterial. The physical properties of the elastomeric material which was in service vill be compared to new asterial to observe degradation, loss of material, deformation anomolies in surface characteristics, and reduced performance.,In addition, the gasket material vill be subjected to chemical analyses to discover changes from original material at the selecular level. Data obtained from the analysis regimen along with data from a similar failure experienced at Brunswick in 1985 vill be used to determine cause. II. PERSONNEL CONTACTED Interviews with the Barris Research Personnel and info'raation regarding analyses performed and resulting postulations. NRR provided j PNPP analyses vill include methods to confirm or deny these failure postulates. The full Brunavick Failure Analysis Report has been sent and vill be used as guidance. A meeting witn Ricerca, Inc. Personnel regarding this failure analysis program resulted in the following proposed course of testing. III. ANALYSIS PROGRAM A. Samples 1. Unused Blastomer Gasket asterial 2. Used 11astomer from pilot solenoids which did not fail. 3. Used, degraded Elastomer Material from failed pilot solenoids. 4. Pilot Solenoid valve bodies with elastomer residue. 4 s I

P.11 B. Physical Testing 1. Profilimetric analysis to compare indentations in IPDM discs (sample nos. 3, and 2) 2. Optical Microscopy to determine the presence of foreign material, or loss of material from surfaces. 3. Hardness testing to compare with original specifications. 4. Compression set to compare with unused material and note performance degradation. C. Chemical Testing 1. Infrared survey to determine carbonile content. This vill provide information about mode of attack (organic acida from the presence of hydrocarbons) and extent of oxidation. 5 2. Scanning Electrcn Microscopy /X-Ray dispersion spectrometry to confirm or negate copper-catalysed accelerated oxidation. (Vhich was a postulated Failure Mode at Brunsviek) D. Environmental Testing six new dual coil solenoids vill be sent to a laboratory for additional environmental testing. The solenoids vill be placed in three separate environmental chambers (two per chamber) at various elevated temperatures in an energized condition. The solenoids vill rossin energized for predetermined times in an attempt to determine the temperature and continuously energized time at which the solenoids do not perform their function. IV.

SUMMARY

The above analyses and their results vill provide evidence of failure mode and vill describe any further confirming analyses which may be needed. In addition, reconsendations vill be made in order to preclude recurrence. _}}