ML20153C996

Forwards Insp Rept 99900403/88-01 Addressing Balance of Allegations & Concluding Insp Efforts Per 870514 Request

ML20153C996
Person / Time
Issue date:	08/19/1988
From:	Hoyle J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
To:	Bradburne J NRC OFFICE OF GOVERNMENTAL & PUBLIC AFFAIRS (GPA)
References
REF-QA-99900403 NUDOCS 8809020002
Download: ML20153C996 (1)

Text

r j I',, UNITED STATES

[ g NUCLEAR REGULATORY COMMISSION

• O :j WASHING TON. D. C. 20666

• s...../

i, -

bl10191999 Docket No. 97300403/88-01 5 MEMORANDUM FOR: John C. Bradburi.e Director , ,

Congressio.ul Affairs '

Office of Goyernmental and Public Affairs FROM: John Hayle, Acting Assistant for Operations Office of the Executive Director for Opeiations s

SUBJECT:

INSPECTION REPORTS PROMISED TO CONGRESSMAN MARKEY

REFERENCE:

Chainnan Zech's May 13, 1987 Letter to Congt.'ssman Markey i

? Your mem randum of May 14, 1987 requested that copies of two NRC inspection reports concerning allegations made by a former General Electric Company (GE) employee be forwarded to Mr. Markey through your office upon completion.

Report Nos. 99900403/87-01, 87-03 and 87-06 have been forwarded to ycur office previously. Since then, the final inspection concerning this allegation has been performed. Enclosed is a copy of NRC Inspection Report No. 99900403/88-01, which describes the results of this inspection. The report was recently reviewed by GE for proprietary informatinn pursuant to 10 CFR 2.790.

No proprietary information was identified.

Inspection report No. 99900403/88-01 addresses the balance of the allegations and concludes our inspection efforts in this area.

hW 5"J'd W John Hoyle Acting Assistant for Operations Office of the Executive Director fcr Operations

Enclosure:

NRC Inspection Report No. 99900403/83-01 h

I l CONTACT: R. Pettis (DRIS/NRR) 492-3214 eso9020002 C80919 I PDR QA999 EMVGENE 79900403 FDC

r

(

'.'[ UMTE] STATES 8 3 {,\,g NUCLEAR REGULATORY COMMISSION WASHis*GTON, D. C. 30666 l

T. '

s l S.....

MG 0 4 Iggg l

l Docket No. M900403/88-01 ,

Dr. Bertram Wolfe, Vice President and General Manager GE huclear Energy 175 Curtner Avenue San Jose, California 95125

Dear Dr. Wolfe:

This letter transmits the report on the inspection of your facility at San Jose, California conducted by Messrs. R. L. Pettis, R. P. McIntyre, and W. P. Haass of NRC on February 1-10, 1988, and the discussions of their findings with Mr. J. J. Fox and members of your staff at the conclusion of the inspection, s.

The inspection was conducted as a follow-up to NRC Inspection Report No. 99900403/

86-01, in response to allegations received by NRC in a letter dated October 5, 1985 from the Government Accountability Project (GAP). The letter and related documents alleged deficiencies in the oesign control activities within the Quality Assurance (QA) program at General Electric's (GE) San Jose, California, facility, during the period March 1978 to April 1982. The allegaticns were based on the work record of Mr. Sam A. Milam III, a fonner GE employee, and a on a review of Mr. M11am's work recora verfonned by Mr. Charles E. Stokes, a consultant for GAP. GAP made documenti concerning these allegations avaiiable for NRC review in February 1986. In March 1986, NRC obtained copies of the documents referred to, including a copy of the report prepared by Mr. Stokes.

, In July 1986. April 1987. August 1987, and November 1987 NRC conducted inspec-tions at San Jote to review these allegations. These inspections were docu-mented in NRC Inspection Report Mos. 99900403/86-01, 87-01, 87-03, and 87-06, respectively.

Areas examinea during the February 1988 inspection and our findings are discussed in the enclosed report. The inspection consisted of an examination of procedures and representative records, interviews with personnel, and observat uns by the inspectors. During this inspection,-it was found that the implementation of your QA program failed to meet cartain NRC requirements, specifically in the area of controlling material procurement as described in GE QA Program Topical Report NEDO-11209-04A and Quality Assurance Procedures 2.4 and 2.7. The specific findings and references to pertinent requirements are identified in the enclosures to this letter.

Please provide us within 30 days from the date of this letter a written state-ment addressing Nonconformance (88-01-01). The statement should contain (1) a description of steps that have been or will be taken to correct this d jU D fv1l7 Y

i . 1

.s Dr. Bertram Wolfe -2 AUG 9 41988 1

item; (2) the extent to which similt.r purchases may have been made for components purchased by GE which may be used in safety-related equipment; (3) a description l of steps that have been or will be taken to correct or evaluate the safety significance of any such purchases and to prevent recurrence and(4}the dates your corrective actions and preventive measures were or; will be completed.

We to dowill so.consider extending the response time if you can show good cagse fpr us The February 1988 inspection concludes our review of the allegations discussed above. A sunenery of each allegation reviewed and its final disposition is presented in Appendix C of this report, in accordance with 10 CFR 2.790 of the Consnission's regulations, a copy of this letter anc the enclosed inspection report will be placed in the NRC's Public Document Ruom. If thit, report contains an exempt from disclosur under 10 CFR 9.5(a)y (4),information you must (a)thatnotify you t,elieve to be this office by telephone within 5 .a.) frun the date of this letter of your intention to file a request for witV ,Iding and (b) submit within 25 days from the date of this letter a writter. p11 cation to this office to withhold such information, if your receipt of this letter has been delayed so that you have less than three days for your review, please notify this office promptly so that a new due date may be O tablished. Consistent with 10 CFR 2.790(b)(1), any application for withholding n.ust be accompanied by an af fidavit, executed by the owner of the infonnation that identifies the document or part sought to the withheld and that contains a full statement of the reasons which are the basis for the claim that the information should be withheld from public disclosure. 10CFR2.790(b)(1) further requires that the statement specifically address the considerations listedin10CFR2.790(b)(4). The information sought to be withheld shall be incorporated as far as possible into a separate part of the affidavit. If we do not hear from you in this regard within the specified periods noted above, the enclosed report will be placed in the Public Document Room without changes.

The response requested by this letter it. not subject to the clearance procedures of the Office of Management and Budget as required by the Paperwork Reduction Act of 1980 PL 96-511.

Should you have any questions concerning this inspection, we will be pleased to discuss them with you.

Sincerely.

! E. William rach, Chief l

l Vendor Inspection Branch Division of Reactor Inspection and Safeguarcs I

Office of Nuclear Reactor Regulation l

Enclosures:

1. Appencix A-Notice of Nonconfonnance

2. Appendix B-Inspection Repurt No. 99900403/88-01

3. Appendix C-Allegation Sumary i

l

APPENDIX A GE Nuclear Energy Docket No. 99900403/88-01 e NOTICE OF NONCONFORMANCE e

• GE QA Program Topical Report, NE00-11209-04A, Section 4 "Procurement Document Contrul," states that measures are established for the preparation, review, approval, and control of procurement documents to provide assurance that reguletory design bases, and other requirements which are necessary to assure the requisite level of quality are included or referenced in the documents for procurement of items and services, including spare and replacement parts.

Section 7. "Cont 91 of Purchased Material Equipment and Services," states that procedures ano practices are established and documented to provide assurance that purchased items and services, whether purchased directly or through subcontractors, conform to procurement docu:nent requirements. These measures include appropriate provisions, objective evidence of quality furnished by the supplier, and examination or review of items or services upon delivery or completion. Documentation of such requirements is originated and maintained by the cognizant QA organizations.

GE Quality Assurance Procedure (QAP) 2.4, Revision 3. "Procurement Document Control." states that measures havc been provided that assure that applicable regulatory requirements and quality assurance requirements are included or referenced in procurement documents.

Contrary to the above, the GE responsible engineer personally procured 100 replacement metric Phillips flat-head screws for use in the reactor mode switch hcusing assembly withaut purchase documents being processed and with no verit cation of the quality of the material by incoming receipt inspection.

The screws were purchased (over the counter) from Mr. Metric in Mountain View, California, with no specification, material request or purchase order. In addition, no receipt inspection for material quality was performed. (See Report Section E., p. 13.)

e

• ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REGRT INSPECTION INSPECTION NO.: 99900403/88-01 DATES: 02/01-10/88 ON-SITE HollRS- 71A j

. 1 CORRESPONDENCE ADDRESS: GE Nuclear Energy ..

ATTN: Dr. Bertram Wolfe, Vice President and General Manager -

175 Curtner Avenue e -

San Jose, California 95125 ORGANIZATIONAL CONTACT: Mr. J. J. Fox, S.alor Program Manager TELEPHONE NUMBER: (408) 925-6195 _

NUCLEAR INDUSTRY ACTIVITY: GE Nuclect Energy is engaged in furnishing engineering services for domestic and foreign nuclear power plants.

ASSIGNED INSPECTOR: .

/

_R. L. Pettis, Special Projepys Inspection Section Date (SPIS)

OTHERINSPECTOR(S): R. P. McIntyre, SPIS W. P. Haass, SPIS P. Es eman nsultant APPROVED BY: UNib1VP ~7 - -$

U. Potapovs, Chief, SPI E, Vendor Inspection Branch INSPECTION BASES AND SCOPE:

A. BASES: 10 CFR Part 21 and 10 CFR Part 50.

B. SCOPE: The purpose of this inspection was to review and disposition iTWgations involving potential deficiencies in design control activities within the Quality Assurance program at GE San Jose, during the period March 1978 to April 1982. In additien, the status of previous inspection findings was reviewed.

PLANT SITE APPLICABILITY: Potentially n.ultiple plant sites, including River Bend, TVA Units 17-22 (identified by GE as cancelled), Perry 1/2, Nine Mile Point 2. Hope Creek 1/2, Grand Gulf 1/2, Limerick, Clinton, and Susquehanna 1/2.

ORGANIZATION: GE NUCLEAR ENERGY'

. SAN JOSE, CALIFORNIA REPORT INSPECTION NO : 99900403/88-01 RESULTS: PAGF 2 of 6?

i A. VIOLATIONS: -

None. ,

B. NONCONFORMANCES:

Contrary to Sections 4 and 7 of GE Topical Report NEDO-11209-04A, and GE Quality Assurance Proceoure QAP 2.4 and 2.7, GE procured 100 replacement metric Phillips flat head screws for use in the rea: tor mode switch housing essembly without purchase documents and with no verification of

~

the quality of the material received by incoming receipt inspection.

(Refer to Nonconformance No. 88-01-01.)

C. UNRE50'.VED ITEMS:

~

None.

- D. STA'US OF PREVIOUS INSPECTION FINDINGS:

Background Information In Oc.tober 1985, allegations were presented to the Nc: lear Regulatory Commission (NRC), Office of Inspection and Enforcement, in a letter from the Government Accountability Project (GAP) identifying deficiencies in the design control and Quality Assurance (QA) program at the GE facility l in San Jose, California. A consultant for GAP conducted & review and analysis of this information and prepared a report of the major concerns. This report and the employee's wor? record were then furnished to the NRC for appropriate action.

The backgrouna information provided to the NRC by GAP was voluminous and addressed a number of issues. The initial NRC inspections were performeo at CE, San Jose in September 1983 in response to allegations which were the same as the basic allegations made by GAP to the NRC in 1985. The previous NRC review of these issues did not include the background information (the alleger's work recdrd and the report prepared by Mr. Charles E. Stokes, a consultant fo GAP) provided to the NRC by GAP in March 1986. A chronology of the major milestones is as follows:

October 21, 1983, NRC inspection at GE in response to allegdtions node by a former GE employee. -

hovember 21, 1983 GE responds to NRC Inspection Report No.. 99900403/

83-03, cated October 21, 1983.

c

-e

i

' ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: .LArJ S of e i October 5, 1985, letter from GAP which highlights concerns' raised by the alleger and the GAP consultant; requests NRC/ GAP meeting.

October 29, 1985, NRC responds to GAP letter and requests all pertinent infonnation, specific allegations, and consultant's report be sent to NRC for review.

Between October 29, 1985 and February 1986, discussions were belo among GE, GAP and NRC to work out details to obtain NRC access to allegers records held by GAP.

February 1986, NRC inspection team reviews records at GAP's Wshington, D.C. office (approximately two weeks).

March 10, 1986, Proprietary Iriformation Agreerrent executed between GE and GAP relating to non-prosecution of the alleger and non-

?

release of document considered proprietary by GE.

March 1986 NRC receives recoros from GAP.

April 1986, telephone conversation between Mr. R. Pettis (NRC) and the alleger to set up interview in San Jose, California, during the week of April 14.

April 16, 1986, NRC interviews alleger in San Jose.

April 14-18, 1986, NRC conducts first inspection at GE, San Jose (NRC Inspection Report No. 99900403/86-01).

May 6,1986, NRC sends alleger copy of transcript for review and correction, where appliatble.

June 1,1986, alleger responds to NRC letter of May 6 by sending corrected copy of interview.

July 14-18, 1986, NRC conducts second inspection at GE, Scn Jose.

October 2, 1986, telephone conversation between Mr. R. McIntyre (NRC and Mr. A. Jackson of GAP concerning GE inspection results and status of resolution of allegations. ,

October 8,1986, telephone conversation between Mr. J. Craig (NRC) and Mr. A. Jackson of GAP concerning GE inspection results and status of resolut1ori of allegations.

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.? 99900403/88-01 RESULTS: '

PArJ a nf A7 October 24, 1986, GAP sends letter to NRC requesting information concerning NRC's action to resolve concerns raised by 1.he alleger.

November 4, 1986, copy of corrected transcript sent to the al?ejer by NRC.

December 2,1986, GE representatives performed initial proprietary review of allegation documents at NRC in Bethesda, MD.

April 6-9, 1987 NRC conoucts second inspection at GE, San Jose (NRC Inspection Report No. 99900403/87-01).

April 10, 1987, Letter from Congressmen Edward J. Markey, and Dennis E. Eckart to the Honorable Lando W. Zech, Jr., requesting NRC to respc,nd to several questions concerning the General Electric allegation issue.

May 13, 1987, NRC responds to the above request.

June 15-18 ano July 27-August 6,1987, NRC conducts third inspection at GE, San Jose (NRC Inspection Report No. 99900403/87-03).

November 9-12, 1987, NRC conducts fourth inspection at GE, San Jose (NRC Inspection Report No. 99900403/87-06).

February 1-10, 1988, NRC conducts fifth inspection at GE, San Jose (NRC Inspection Report No. 99900403/88-01).

The allepuons inspected are sunnarized below, along with the results of the NRC review of each item. The inspection was comprised of personnel interviews, examination of applicable files, records, and procedures.

References to nuclear power plants identified as TVA (17-22) represent cancelled plants. The section references and coments referenced below have been restated vert,atim and correlate directly to those listed in Mr. Stokes' report; and as such do npt constitute an NRC interpretation of such allegations.

1. (Closed) Stokes Report Section 1.6 Engineering Review Memorandums (ERMs) .

"In the first week of November 1978, the following line was part of an entry: Bill Millard said either he would sign the ERMs or I (Sam) could forge his signature to them." (Clarificationaddedby Mr. Stokes.)

~

.c

, ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORhlA REPORT INSPECTION NO,: 99900403/88-01 RESULTS: PAGE 5 of 62

~

Ins pection Findings - During the inspection, e ascussion was held witi Mr. Millard, in the presence of Mr. Barton Smith -GE counsel, at which time Mr. Millaro denied any such statement about Morging" his signature. Mr. Millard, then the Project Manager (PM), stated that it was comen practice for PMs to authorize other individuals to sign for them in instances where, for example, logistics did not permit the PM easy access to sign such documents. The NRC inspector reviewed the responsibility given to the PM as outlined in GE E0P 42-6.10, Section 4.8a, to determine the significance placed on the PMs approval of such documents. The E0P states that the PM has tne responsibility to approve ERMs for the purpose of authorizing the application of the specific document to the assigned project and to supply project information to the initiating respontible engineer as requested. The PM is not responsible for verifying or checking the technical adequacy of the document, because this has already been performed by the design engineer and the verifier. Fecause specific detail was not available as to the ERM referenced by Mr. Milam's work record entry in November 1978, the inspector was unable to pursue the signature question.

A review of signature related allegations identified by Mr. Milam and CongroIsman Edward Markey's staff has been perfonned over the past sevetal inspections. During these inspections, the NRC inspec-tiun team did not identify any safety concerns within this area. In the absence of further infonnation, this item is closed.

2. (Closed) Stokes Report Section 1.7 Elementary Diagram Drafting Effort "Continuing with a problem of similar nature on November 14, 1978, a letter to C. W. Hart on the subject of the CNV connection has an interesting paragraph. It seems that the CN) ilementary diagram drafting effort was subcontracted to an out side firm, the Power Division of C. F. Braun & Company, in Alhambra, California. When completed, the diagrams were provided to the General Electric System Engineers for signature. The system engineers felt that they were not being giver sufficient time for review and refused to sign the documents. The documents were later signed by the C&EE CNV -

engineer, without review." ,

Ins)ection Findings During the inspection, discussions were held witiMr. C. W. Hart, Mr. Milam's supervisor during this period, who statec he had never received the hovember 14, 1978 letter, in addition, specific examples of insufficient review times could not

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA

. l REPORT INSPECTION NO.? 99900403/88-01 RESULTS: PAGF A nf 69 be identified from the connents contained in the Engineering Review Memorandums (ERM) or the discussion with Mr. Hart. The NRC inspec-tion team reviewed hu dreds of ERMs during these inspections, and have not identified any instance that would indicate an insufficient review had been performed. As a result, this item is closed.

3. (Closed) Stokes Report Section 6.2 Unauthorized Signature Changes "Mr. Milam wrote a letter to W. M. Barrentine on April 14, 1982 about unauthorized post signature changes. In this letter, Mr. Milam states that R. L. Reghitto made an unauthorized change to ERM AML-2997 without Mr. Milam's knowledge and in direct conflict with specific instructions."

Inspection Findings - A disec ston during the inspection with Mr. Barrentine in the presunce of Mr. Barton Smith, GE counsel, inquired as to what actions were taken concerning this subject.

Mr. Barrentine stated he had not received u.a. Milam's letter of April 14, 1982. He also stated that he w'.', not aware of anyone else who might have known about the letter an. also might have acted on it in his (Mr. Barrentine's) place whilt. he was on business travel.

• A further review of this issue revealed the changes in question related to administrative annotations made after Mr. Milam signed the document as the responsible engineer, and as such had no technical basis or impact on safety. Annotations such as these noted by Mr. Milam are described and authorized under GE Procedures.

This item is closed.

4. (Closed) Stokes Report Section 6.3 Letter to Management "On May 22, 1982, Mr. Milam wrote Mr. Barrentine a letter and included a copy of his work record while working for Mr. C. L. Cobler, in this letter Mr. Milam requested Mr. Barrentine to read about the on-guing underworld of C&!D and says he tried to communicate some of these things to Mr. Barrentine on several occasions but was discouraged by Mr. Barrentine's managers and attitude. Mr, Milam says:

.s

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSFECTION NO.? 99900403/88-01 RESULTS: PAGF 7 of 6?

Since you no longer hold my form 38 (a standard threat'),

I have nothing fu-ther to fear from either you or your conspiratoral managers. I hope, by sending you this e ,

record, to give you a glimpse into that hidden world of uncontrolled lootleg activity we all know so well."

Mr. Stokes also stated that Mr. Barrentine was the ianager of the Nuclear Control & Instrumentation Product Design Operation of the Control and Instrumentation Department (C&ID). He was Mr. Hart's, Mr. Cobler's, Mr. Reghitto's, Mr. Strambach's, Mr. Koslow's, and Mr. Wortham's supervisor. Mr. Milam had been notified of his layoff when this last letter was written and his reference to form 38 had to do with the constant threat of layoff if you did not go along with

) the system. He did not. (Emphasis addec by Mr. Stokes.)

Inspection Findings - A discussion ciuring the inspection with Mr. Barrentine, in the presence of Mr. Barton Smith, GE counsel, inquired as to what action was taken by Mr. Barrentine when he received Mr. Milam's letter and work record, which document problems Mr. Milam felt existed within the C&ID. Mr. Barrentine stated he never received the letter nor the portion of Mr. Milam's work record while Mr. Milam was assigned to Mr. C. L. Cobler. He stated that i

he had received information concerning problems within C&ID, he would have met with the managers and thoroughly researched the issues. Mr. Barrentine also stated that a Pot,tntially Reportable Conditia (PRC) evaluation of Mr. Milam's concerns would have been I initiated if warranted.

A review of Mr. Stokes' summary of Mr. Milam's work record, while working for Mr. C. L. Cobler, revealed 83 items of concern repre-senting approximately 50 percent of the total allegations summarized in Mr. Stokes' report. During the NRC's review of these allega-tions, several nonconformances were identified (refer to Appendix A) which were related to activities described in Mr. Milam's work record while working for Mr. C. L. Cobler during the period March 31,1980 through November 24, 1980. These nonconformances related to the reactor mode switch, deferred verification, processing of ECNs and CARS, Electrical Metallic Tubing, and the GE Engineering Information System. Although items of nonconformance were identified and respondeo to by GE, it was concluded that the concerns raised by Mr. Milam were typically of an administrative nature and 'as such may not have resulted in a Potentially Reportable Condition even if Mr. Barrentine had received the letter referred to by Mr. Milam. As a result, this item is closeo.

.i

• ORGANIZATION: GE NUCLEAR ENERGY' SAN JOSE, CALIFORri!A REPORT INSPECTION NO.: 99900403/88-01 RESVLTS:  ! PAGE B of 62 (Closed) Stokes Report Section 5.13 5.

Excluded Equipment List  ; e "Mr. Milam's work record includes a nonapproved fonn titled PWA No.

1229LD, Revision 1J for River Bend. This document, which is dated February 5, 1982, was caused by an Excluded Equipment List (EEL) which was sent to Gulf States Utilities Company by the NRC. The sticond page of this document states that there is no controlled tracking system for vendor identification of these devices and that a complete item by itena search of the entire River Bend database would be necessary. GE felt that the scope of such a search was prohibitive and furthermore was not considered to be necessary.

Excluded equipment as referred to in this list is equipment wl.ich

' has been found at other facilities to be so deficient that plant i safety is seriously in question. GE neither admitted nor denied that this equipment was installed at River Bend.

Mr. Stokes' Coment: Since GE believes the cost of the search to ensure that it is not installed is prohibitive, the plant should not operate. This is because all necessary requirements of 10 CFR 50, Appendix B have not been met justifying the granting of the operating license."

l Inspection Finding - A review of the EEL indicated that it consists of a compilation of components found to be defective by licensees

+

and/or vendors and reported to NRC. Such reporting, with appropriate reviews and analysis, resulted in the issuance of NRC Information Notices, Bulletins, and Generic Letters to inform licensees and construction permit (CP) holders so that proper corrective action could be taken if applicable. Apparently, some segments of the nuclear industry compiled information from these NRC documents in  ;

an abbreviated form and issued the EEL that is presently the subject of this concern. The EEL was not prepared or issued by hRC.

As noted above, information didseminate'd by NRC is sent to licensees who are responsible for implementation of the NRC recomendations regarding potentially defective components. For licensees, the proper action entails the review of plant hardware to determine the presence of the defective component; for CP holders, the proper actions involve the review of plant hardware already delivered and t informing vendors and designers of hardware not yet delivered.  !

Untier the provisions of 10 CFR Part 21, vendors who identify a defective basic component are required to inform their customers and to include a recommendation for corrective action.

l a

~

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: i PAGE 9 of 67 Another potential path for removal of defective components'from nuclear plant hardware is the receipt of NRC documents; by vendors and manufacturer's, and the implementation of NRC recomendations during the hardware fabrication process. The method nonnally used by General Electric to ensure the removal of defective components during the design and production processes was notification of a potential problem by the CP holder. The inspectors reviewed several instances on the Perry and Riverbend projects in which General Electric was requested by the utility to review its design to deter-mine whether certain components identified as potentially defective by NRC were included. General Electric had no specific procedure for accomplishing this type of activity other than requirements for the General Electric project manager to respond to customer requests,

~

and for General Electric to process Deviation Disposition Requests submitted by suppliers in accordance with Engineering Operating Procedure E0P 45-6.00 "Deviation Disposition Requests From Suppliers,"

Revision 6 dated July 20, 1987.

It was concluded that Mr. Milam's concern was adequately addressed by the existing process of handling recontrendations regarding defective components through licensees eno CP holders. This item is closed.

E. OTHER FINDINGS AND OBSERVATIONS:

I Stokes Report Section 1.8 "During the second week of March 1979, Mr. Milam recorded his discovery of an error in his, Russ's, and Don's work. They had assigned connector numbers in the E51A elementary diagram which duplicated connector numbers that were previously assigned. Threa

- panels (H11-P618. H11-P621, and H11-P640) were affected. Mr. Milam and Don attempted to perform corrective action.

Mr. Stokes' Coment: The problens here was the absence of any clearly defined authority for the assignment of connector numbers. The responsi-

, bility would normally lie with the design engineer, unless he delegates it to another engineer, designer, or draftsman. The positions which P.uss. Don, and Mr. Milam held were not noted."

l Inspection Findings - Apparently wrong connector numbers were assigned

, during the development of the E51 elementary diagram. The error was

discovered during a drawing review performed prior to issuance of the diagram when potentiel errors are expected to be discovered and corrected by the responsible design engineer. The record shows that the errors were corrected in confonnance with established procedures. This item is closed, l

. s l

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: PAGE 10 of 62 Stokes Report Section 1.12 "A letter dated October 25, 1979, to D. H. Currie from R. J. Giel voices Giel's concern for what she sees happening within Cal 0. She says that after looking at various procedures she did not find outright conflicts but a cumbersome redundancy of proce-dures. The problem originated from the previous issuance of incomplete, poorly researched and oocumented proposed ECNs on the part of engineering, thereby generating the need for additional review / signatures per each ECN approval. She stated that recently there had been a proliferation of concern on this subject, which resulted in two new procedures and their respective boards -- the ECN Coordination Board (ECB) and the Change Evaluation Board (CEB). She advocated a simplification rather than a proliferation of procedures. She said the present system encourages a let someone else check that aspect attitude."

Inspection Finding - An interview conducted with Ms. Giel, during the inspection, inoicated that she was assigrad the responsibility to review

• Engineering Change Notices (ECNs) for impact on Quality Assurance (QA) by Mr. H. Currie of General Electric. The letter in question was written as a suggestion to help improve the present system and indicated no problem with the existing system. Ms. Giel stated that she left QA shortly after writing this letter and to her knowledge could not recall whether any GE procedures were changed or modified as a result of her letter. This item is closed.

. Stokes Report Sections 1.17, 2.30, 2.57, 2.58, and 3.7 4

Section 1.17 "Mr. Milam requested an EMPIS information manual from Sarah Sanders in Schenectady during the first week of 1980."

Section 2.30 "During the second week of June, Mr. Milam noted that errors were still being made at a very high rate in the G36.P002 drawings. See Section 2.27."

Section 2.57 "During the third week of August Mr. Milam wrote the following comments in his work log. ,0ur purchased part drawings are terrible. All af the Material Reviek Board problems I handled dealt with purchased part o.awings."

Section 2.58 "On August 19, 1980 Mr. Milam wrote that there was no action yet on the CNV Category III design change problem. See,Section 2.48."

. s 1

• ORGANIZAT10h: GE NUCLEAR ENERGY SAh JOSE, CALIFORNIA REPORT INSPECTION

, NO.! 99900403/88-01 RESULTS: '

PAM 11 of a Section 3.7 "Mr. Milam addressed a letter to G. A. Senn on February 9, 1981 on an issue where he felt a problem existed with mispliced priorities.

He asked the following question: Why are we more concerned uith' the' quality of the meetings (than) with the quality of the product?" -

(Correction made by Mr. Stokes.)

Inspection Findine - These items were not reviewed during the inspection since Mr. M11am's statements appear only to reflect his day-to-day work activity and not a direct allegation or concern.

Stokes Report Section 1.20 "Mr. Milam made a connent concerning mode switch assemblies also during the first week of February. He stated that the assembly was not being properly documented. The switch was purchased as a nun-safety-related part and should have been assembled, tested, etc.,

with its own work order and other paperwork, the same as a panel. This did not happen on Cofrentes. Instead, the switch was delivered to the

' plant in a disassembled condition with no certification. The switch was assen, bled and disassembled many times by many people ano eventually shipped with a bootleg work order. Mr. Milam believed that the same thing happened to Kuo Sheng.

Comment: This is a violation of 10 CFR 50 Appendix B Section VII.

Control of Purchased Material. Equipment, and Services."

Inspection Finding - The issues concerning problems with Cofrentes and Kuo Sheng reactor mode switches were addressed during the October 10, 1980 meeting of the GE Problem Review Board (PRB) of which Mr. Milam was a member. Mr. Milam was assigned the task to review the panel design for these plants and take appropriate action to ensure no problems existed.

Mr. Milam was not satisfiet' with the suggested design fix for the Cofrentes and Kuo Sheng mode switches. He suggested an alternative fix, but was instructed by GE management that the fix suggested by the PRB on Octoter 10 was acceptable and should be implemented. There was no evidence that the switch was shipped using a "bootleg" work order. For additional infoma-tion involving the reactor mode switch, refer to Section E.6 of NRC Inspection Report No. 99900403/87-03.

Stokes Report section 1.26 "On March 29, 1980 Mr. Milam wrote C. W. Hart 6 letter about a problem with reactor mode switch.163C1487.

He stated that this switch was assembleo using cam switch 195E9497 and that the parts list shown on this drawing contains two vendor documents with status "LTR." Four switches had been assembled and shipped as of this letter. He said, that he nad notified Control and Electrical Engineering subsection personnel and requested clarification of the

document status but corrective action had not taken place. He opposed

shipping these switches until the documentation problem was resolveo.

I l

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA e

REPORT INSPECTION h0.: 99900403/88-01 RESULTS: i PAGE 12 of 62 l Mr. Stokes' Comment: Shipping these switches prior to the resolution of a documentation problem is a violation of 10 CFR 50 Appendix B fection VI, Document Control." Special reference should be given to the following i documents discussed in this section:

1. Letter from J. R. Cintas-Problem Review on Re(ctor Mode Switch Housing, November 17, 1980.

2. Letter from Sam A. Milam, !!! to (15 involved f ..lduals) Reactor Mode Switch in Kuo Sheng and Cofrentes panels, november 21, 1980.

3. Letter from C. W. Hart to S. A. Milam, III, Reactor Mode Switch ls Design, December 1980.

4. Perfonnance appraisal for January 1982 (Sam A. Milam,111).

, 5. Reactor Mode Switch work record."

Inspection Finding - This problem was addressed by the GE PRB during

'; their June 10, 1980 meeting. Drawing 195B9497 is the purchased part drawing for the cam Switch and the parts list drawing was referenced on it. The parts list drawing provides no information necessary for fabrica-tion, test, or use of the switch. GE had no explanation of why drawing i

386X240 was established. The reference to the parts list drawing on drawing 195B9497 was therefore deleted. (See Stokes Report Section 2.35.) Parts list *,86X240 Revision 1 changed the document status of the two vendor documents from "LTR" to "RVW." as of December 22, 1980.

This was accomplished via ECN NJ20254. This item is closed.

StokesgRa ort Section 2.1 *During the first week of April 1980, Mr. Milan made a purchase of some 100-M3X16 phillips flat head screws for use on reactor mode switches. These screws were purchased from Mr. Metric, 2257 Old Middlefield Way, Mountain View, California. No specification was written, no purchase order issued, nor any documentation received on material quality.

~

Coment: This action was a violation of 10 CFR 50. Appendix B. Section Vll, Control of Purchased Material Equipment, and Services."

Inspection Finding - Unavailability of the required length screws, as cefined by the switch inanufseturer (Gould Rundel), initiated Mr. Milam, the responsible engineer, to personally procure the required screws and control their application to the switch. This deviation from the nonnal procurement process violates 10 CFR 50 Appendix B, GE QA Program Topical

e6 ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.! 99900403/88-01 RESULTS: pirJ 11 ni As t

Report NEDO-11209, and GE Quality Assurance Procedure requiremen'ts.

Nonconfonnance (ho. 88-01-01) was identified during this past of the inspection. , ,

Stokes Report Section 2.4 "During the second week of April 1980, the M3X16 reactor mode switch screws are back. Mr. Jack B. Levy in Schenectady will provide a metric standard for the M3X16 screws. This standard will then be used to document the screws previously purchabed by Mr. Milam.

See Section 2.1."

Inspection Findino - The inspector reviewed the GE interim drawing (metric series) for steel flat head screws. This drawing was issued on April 11, 1980 and its purpose was to establish a GE machine part designation for i parts specified herein and to inform suppliers on the interpretation of this machine part designation when it appears on orders. These drawings are specification type drawings for metric flat head machine screws.

These drawings could not be used to document the r, crews previously

y. furnished as described in Mr. Stokes Report Section 2.1. These drawings shruid have been specified on the purchase order for the 100 screws had the order been properly written and the screws properly procured. Stokes items 2.1 and 2.4 above are closed due t.o the inspector's opinion that these issues do not represent a safety hazard.

Stokes Report Section 2.5 "On Friday of the second week of April 1980 Mr. Milam discovered that it was possible to desynchr0nize the mode i switch contacts by partially remving the square drive and turning it.

He verified that Quality Assurance was not at that time testing for contact closure of the mode switch, c Mr. Stokes' Coment: This was but one more problem with the controls over the purchase and use of the reactor mode switch. See also Section 1.19.

Inspection Finding - The issue of desynchronization of the reactor mode switch was covered and closed in NRC Inspection Report No. 99900403/

87-03, Section E.6, "Reactor Mode Switch."

Stokes Report Section 2.7 "Mr. Milam decided that his interpretation l

of the Engineering Operating Procedures (EOP) had been wrong and that source responsibility for the ECN verification statement is with the responsible manager, not the responsible engineer. -

Mr. Stokes' Comment: P.r. Milam had been with GE for four or five years.

The issue raises r. merous questions. Why hadn't this been a point of i

debate between designer / checker or approver? Why wasn't Mr. Milam taught l

t 1

i l

i

.s

. ORGAN!ZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA 4

REPORT INSPECTION NO.: 99900403/88-01 RESULTS: '

PAGE 14 of 62 this in a training session? If indeed Mr. Milam has been wrong,'then undoubtedly others who were in the chain of responsibility were equally in error or deliberately avoided this issue. In all cases, all' inst %nces where someone other than the responsible manager signed the verification ,

statement, a violation of 10 CFR 50 Appendix B.Section VI. Document l Control, has occurred."

Inspection Finding - General Electric E0P 55-2.00 requires the respon-sible engineer to assurs that each design change specified by an ECN is verified per E0P 42-6.00. It also requires that the responsible manager assure that the required design verifications have been completed in accordance with E0P 42-6.00. E0P 42-6.00 required, <turing this time, that the responsible engineer select the verifier and that the responsible

~ manager approve the verivication. Mr. Milam's interpretation si med from the ECN form completion guide which lists the "source" for the verifier and verification statement to be the responsible manager. The "source" responsibility is assigned to the responsible manager because of the requirement to assure verification of all design changes.

Whenever Mr. Milam or any employee has a question on a procedure, it is that person's responsibility to seek advice and counsel. The interpreta-tion of the forms completion guides and procedures are provided by the assigned counsel for each procedure. The counsel is identified in the E0Ps and available for interpretation to all NEB 0 employees. This item is closed.

Stokes Report Section 2.9 "Mr. Milam also found that ECN NJ17584 was released with deferred verification thus requiring EI's (Engineering Instruction) for Conditional Shipping Release on the panels affected.

Two panels were shipped already and six scheduled to ship during fiscal week 8017. GE started documenting deferred verifications as a result of

• finding out that verifications had been deferred violating 10 CFR 50, Appendix B.Section III, Design Control.

Mr. Stokes' Cwenent: If a compartsqn were performed of procedures before and after this practice started, one would find that the number of proce-dures and associated control requirements would have at least doubled.

If a comparison were also made of the number of audit findings and other documents recording 10 CFR 50 Appendix B violations, one would see that the number of errors made by the design organizations increased consider-ably. The fact that verifications are deferred may not be a violation of 10 CFR 50 if documented and procedurally controlled, but the fact remains that the entire project suffers from this practice. Not only does the quality of the organizations' work suffer, but the possibility of a piece 7

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA 1

REPORT INSPECTION

_ NO.! 99900403/88-01 RESVLTS: PAGE 15 of 62 of safety-relateo equipment causing or contributing to a nuclear accident increases unnecessarily all because management made a bad decision (to start documenting deferred verification, a practice which should'have beenstopped)."

Inspection Finding - The use of EI's for Conditional Shipping Release on panels with deferred verification is an acceptable method as described by GE prcceoures for accomplishing this activity. The electrical systems engineer revised the elementary diagram by ECN NJ17584 before the Piping and Instrumentation Diagram (P&lD) was revised and accordingly deferred the verification. The need for deferring the verification was question-able, since the elementary diagram was being brought in line with the design specification. P&l0 10504917BA was redrawn and renumbered by Engineering Change Authorization ECA 800425-1 to drawing number 795E805 and verified on ERM DMC-1496 issued February 20, 1981. The ECA and subsequent P&ID drawing changes included the changes requested concerning the volve positions. Once the work associated with the ECA was completed and the PalD verified DVSCN 00011 closed the associated ECNs that hao been deferred. The general topic is explained in NRC Inspection Report No. 99900403/87-06, Section E.2. This item is closed.

Stokes Report Section 2.11 "During this week Mr. Bob Gordon (QA) and Mr. Milam replaced 4-18 mm long screws on Hope Creek mode switches.

Mr. Stokes' Comment: The mode suitches as purchased did not have the proper documentation for use in safety-related systems."

Inspection Finding - The mode switch is considered and identified as nuclear safety-related on parts list drawing PL163C1487 and assembly drawing 163C1487. The Purchased Part Drawing (PPD) 195B9497 for the cam switch is the drawing that lists the design specifications and provides thedetaildrawingofthecamswitchforthemanufacturer(Gould-Rundel) to follow. The PPD is what governs the manufacture of the mode switch.

The vendor must certify that the design and material specifications have been achered to. This is part of the documentation for use in safety-related systems. This item is closed.

Stokes Report Section 2.13 "On April 15, 1980, Mr. Milam sent a report about the oesynchronization of reactor mode switches to Mr. C. W. Hart.

In his conclusions Mr. Milam stated thdt desynchronization could occur with the correct key positions being obtained although contact closure was incorrect. -

Mr. Stokes' Coment: This problem is a result of the lack of a functional test being perforced. What other equipment may be defective? Per Test Instruction 2244. Revision 2, pg. 23 of 24 the following types of 1

ORGANIZATION: GE NUC1. EAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: 1 PAGE 16 of 67 n

devices do not receive a functional test as part of the panel te'st:

protective relay circuits, equipment supplied by others, and packaged systems built by others. This violates 10 CFR 50, Appendix B. Sections X, XI, XIV, and XV."

Inspection Findins - The issue of desynchronization of the reactor mode switch was coverec and closed in NRC Inspection Report No. 99900403/87-03, Section E.6. The mode switch receives functional testing by GE as well as numerous tests by the licensee after installation at the plant and prior to operstion. This item is closed.

Stokes Report Section 2.16 "On May 2, 1980 Alice Lacombe changed the wording on ECN NJ-13571 without authority and against Mr. Milam's instruc-tions. See 2.14."

Inspection Finding - A review of this item together with Mr. Stokes'

- items 1.14, 2.14, 2.28, 2.39, and 5.16 could not identify technical modification to any document. Administrative additions or changes were identified and utilized to keep the documents within the GE document cuntrol system procedures. All the changes were easily identifiable but were not labeled as to the source of the modification. An engineering division memorandum later required additions to ECNs to be initialed for the identification of the modifier. Later ECNs were examined and provideo this additional capability to the document control system. This review did not identify deviations to GE E0P 55.02 for document control.

This item is closed.

Stokes Report Sections 2.22 and 2.51 Section 2.22 "Mr. Milam, during the third week of May, af ter having a conversation with Bob Gaetani and Mike Howley came to the following conclusion. I (Milam) have always assumed that an ICER code of 'C' indicated a complete crawing with a complete verification. Not so. The

'C' means the drawing is complete, but says nothing about the verifica-tion. According to the Engineering information System, if the verifica- '

tion is not complete the ICER code it 'U,' but on the drawing it will be

'C'.

Mr. Stokes' Coment: If this is true, then the procedures are wrong, it is not possible for a drawing to be finished before the design, process has been completed and checked, since the drawing relies on the design.

l At GE with their deferred verification, it appears that the cart has gotten before the horse. The drafting department has assured along with others that when a drawing is drawn up per engineering's request that it 1

i

l ORGANIZATION: GE NUCLEAR ENERGY ,

SAN JOSE, CALIFORNI A l l

REPORT INSPECTION NO.- 99900403/88-01 RESULTS: PAGE 17 of 62 i

is complete. This is not so. It is not complete until the desi'gn has been verified and the drawing verified from the design. Changes may be required to the drawing right up to the time of issuance. No drawing before issuance should have an ICER code of 'C' on it. Any drawing which does not indicate the design status upon which the drawing relies is incorrect. This is in violation of 10 CFR 50 Appendix B. Section !!!,

Design Control."

Section 2.51 "During the last week of July, it seems that some of the production control expediters are afraid to use the ICER and SUN codes.

They were told by Clark Canhan that they could not rely on codes that appear on drawings and EIS. See Section 2.22."

3 Inspection Finding - Mr. Milam's coment refers to the inconsistent use of terms for engineering documentattun practices. If a drawing is marked "C" (complete), this does not indicate whether the drawing is verified or unverified. The EIS indicates whether a document is complete or not and also the verification status. Both the document and the EIS are controlled by GE operating procedures. Mr. Milam indicated a desire for system ,

modifications for ease of use, not error corrections. Mr. Stokes has ioentified an expanded application of the drawing codes which is not a requirement under GE procedure. This item is closed.

Stokes Report Section 2.23 "ECN NJ-11071, when revised from Revision 2 to 3 hao a notation that "PL ISSUED" removed without authorization.

. Mr. Milam encountered a wall of apathy in trying to correct this error.

Mr. Stokes' Coment: This notation is important because the PL (Parts List) contains LTR's (indicating that the item would be completed or added at a later cate) which requires a conditional shipping release for the panel to be shipped. Mr. Milam states, we are continuing to ship reactor mode switches without conditional shipping releases, with LTR's on the parts list."

Inspection Findings - ECN NJ-11071 was written for changes ude to purchased part crawing 195B9497 for the reactor mode switch. Purchased part drawings do not have separate parts lists. The notation "PL ISSUED" was in error and was removed in accordance with established procedures.

The erroneous notation was identified as probiam No. 6 at the Problem Review Board meeting of June 10, 1980. (See Stckes Report Section 2.35.)

ECN NJ-11071 issued un July 20, 1979, deleted "PL ISSUED" under general changes. This item is closed.

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA e

REPORT INSPECTION NO. 99900403/88-01 RESULTS:

PAGE is of 67 Stokes Report Section 2.25_ "The fourth of May brought an interesting problem. Mr. Milam was responsible for responding to CAR's'(Corrective Action Response) requesting addition of surface finish requirem6nts 'to fabrication drawings. Most of these drawings seem to have been around for years and some of them indicate dozens of project applications.

Mr. Milam wrote down his three choices. The third is interesting - come up with a story on how every part we ever made satisfied the requirement even though the requirement didn't appear on the crawing. Even more interesting is the fact that his supervisor, Mr. Cobler, accepted it.

Mr. Stokes' Connent: This is but one more instance of drawings used over and over without any reverification that it meets plant specific require-ments or material availability or current standards. See Section 2.2."

Inspection Finding - The fabrication drawings were utilized as the basic requirements for the manufacture of each panel. If a client had a special requirement or if the manufacturing process changed, the change would be sent to the engineering departJnent for their review of applicability to each plant. The GE approach was to have a standard design and apply the requirements of this design. The engineering documents would only identify deviations to the standard design. No specific cases of redesign were identified during the inspection. This item is closed.

Stokes Report Section 2.26 "On May 29, 1980 Mr. Milan wrote a PRC (Putential Reportable Condition) on the problems with the reactor mode

,' switches. During FW8029 (third week of July), Mr. Milam was informed

! that as of July 10, Licensing had not received his mode switch PRC. A l check with Barrentine (by Currie) confinned that it has been sent to l

Licensing and was apparently lost. See Section 2.13."

Inspection Finding - It is true that the Licensing section did not receive anything official on the PRC until August 26, 1980, because the issue was being evaluated internally by the Engineering Departunt.

Af ter receiving Mr. Milam's letter, the Manager of Quality Systems, Audits and Records referred the PRC to the Manager of Product Design Engineering on May 29, 1980. On June 2,1980, a letter to Nuclear Licensing was written and signed by the Manager of Product Design Engineering and the Manager of Quality Assurance NC&lD. During the time span in question, the Problem Review Board met on June 10, 1980 to address the mode switch problems, which was being evaluated until fonnal transmittal of the PRC to Licensing in August. This item is closed.

Stokes Repurt Section 2.32 "Another problem concerning nuclear safety-relate:o or not Evolves saddle clamps. Per a conversation with Jim Morrissey and Ken Seibert, vendor (Charles Witter) will not continue vendor certification if the saddle clamps are removed.

ORGANIZATION: GE huCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.? 99900403/88-01 RESULTS: '

PArJ 10 nf A9 Mr. Stokes' Connent: This means many CR2940's are currently in the field in nuclear safety-related applications without certification. A similar situation exists for MSP 15.004 (pushbutton CR2940) and MSP 15.dD5 (130 pushbuttonswitch851E392). See Section 2.31."

Inspection Finding - The procedures governing this activity were reviewed and the following was tietermined:

Procedure MSP 14.017 Revision 13. Paragraph 5.1.1 calls for removal of "C" clips (saddle clamps) for switches used in non-safety-related applications only. (An MSP is a GE Manufacturing Standard Practice.)

MSP 15.004, Revision 3, which supercedes MSP 14.017, simply calls for removal of "C" clips and replacement with spring lock washers for terminal lugs for switches less than 4 in, apart.

Apparently, manufacturing followed the latest procedure (MSP 15.004, Revision 3) that did not distinguish between safety-related and non-safety-related applications. The standard practice in the General Electric NEB 0 shop is to terminate wires with lugs by crimping.

In response to a GE letter, Cutler Hanner, the switch vendor, sent a letter to GE indicating acceptability of using lugs in lieu of the saddle clamps. The GE vendor, General Purpose Controls, would not, however, authorize their qualification data to be used for the assembly without saddle clamps. Therefore, under PRC 81-35, GE verified by test that the use of lugs was acceptable; that is, no loose connections were identified nor has any equipment failed a qualification test. The switch was qualified for Class 1E service with or without the saddle clamps. The tests conducted considered amplification of the seismic response at higher locations on the panels. The fragility limits for testing were as follows:

Allowable Levels Calculated Levels Front to Back 12.68 g 11.70 g Side to Side 10.00 g 9.57 g Vertical 24.70 g 5.00 g Testing was conducted in accordance with IEEE 344-1975. It was concluded that Mr. Milam's concern was substantiated; however, appropriate corrective action was performed by GE to resolve the matter. This item is closed.

I

(

1 ORGANIZATION: GE NUCLEAR ENERGY' SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: ,

PAGE 20 of 62 Stokes Report Section 2.33 "In a work meeting on EAP's (Engine'ering Applivo Practices) with Jerry Willis, Ray Loui, and Paul Hopkins, it was discussed how to achieve traceability for as-shipped configuration w4th respect to EAP revision.

Mr. Stokes' Coment: This was because EAP's are part of the engineering documentation. Is this a new case of deferred verification?"

Inspection Finding - During an interview, Mr. Jerry Willis, who attended the EAP aseeting referenced by Mr. Stokes in June 1980, stated that Mr. Barrentine initiated this working group to extract engineering infor-mation from the MSPs and incorporate it into the EAPs. The MSPs describe the "how to" aspect and the EAPs describe the "what to do" aspect. Prior to 1980 there were no EAPs.

It was pointed out by Mr. Willis that EAP revisions are totally inter-changeable, and thus future revisions to an EAP would have no affect on the ones shipped. If the change were not interchangeable, a new EAP drawing number would be established. Traceability is controlled in either case by the Engineering Change Notice / Field Disposition Instruc-tion (ECN/FDI) process. This item is closed.

Stokes Report Section 2.34 "During this week, Mr. Milam noted that CNY G41-P002 and G41-P003 have terminal boards bridging the wire grill gap.

His recordeo options were to fix the drawings or get the customer to accept as is."

Inspection Finding - Mr. Milam issued ECN NJ-13554 to resolve the problem notec in his work record in accordance with GE procedures.

This issue is closed.

Stokes Report Section 2.35 "In a letter documenting the decisions made in a meeting on the reactor mode switch 195B9497, the following problems and solutions were decided upon.

Problem 1: Parts List PL386X240, Re' vision 0. Cam Switch contains "LTRs" and there is no plan in place to complete the document.

Solution: Parts List PL 386X240 is referenced on drawing 195B9497. It was agreed to delete parts list PL386X240 and the reference to it on the referenced drawing.

Problem 2: After renoving the switch neck assembly to install the switch into the isolation can, it is possible that the switch contacts are desynchronized.

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 21 of 62 Solution: To change the mounting design of the switch so that it is not necessary to do any disassembly of the switch. '

e

• Note: His caution against any disassambly does not correct or verify the correct position of the units previously disassembled and installed.

Problem 3: The replacement of the four mounting screws is not documented in any procedure.

Solution: The implementation of the solution of Problem #2 negates this problem.

Note: Implementation of the solution of Problem #2 does not negate the lack of documentation for those units already modified.

Problem 4: The screw fasteners in the base of the Isolator can become loosely coupled on some assemblies and tend to twist. It is virtually impossible to torque the screws to an acceptable level.

Solution: The base material will be changed to a thicker gauge. The screw fastener will be analyzed in the new base material to ensure that an acceptable procecure is in place to provide a high yield of acceptable assemblies.

Note: What about existing assemblies?

Problem 5: The switch neck assembly was fractured at least once during the installation of the switch into a bench board.

Solution: Implementation of the solution to Problem #2 will prevent stress to be placed on the switch neck at all times.

Problem 6: Drawing 195B9497, Revision 3, rernoved a note "PL issued,"

without benefit of an ECN. -

Solution: All drawing changes must be documented, however, the impact of this problem is negated by the fact that the solution of Problem #1

, removes all references to parts lists.

Note: The removal of any reference to the Parts List coes not resolve this problem, it only hides it. The Parts List must be referenced unless all information on the Parts List is either on the drawing or a second drawing which is referenced on the drawing from which the reference was removed.

' ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: "

PAGE 22 of 6?

Additional Concerns: It was generally believed that inadequate documen-tation is retrievable to document the true condition of the' switches' shipped in bench boards or as separate devices. "

Solution: The new assembly should replace those previously shipped.

These include: Cofrentes, Kuo Sheng 1 & 2 Perry 2, and the individual switches shipped to Hope Creek and Leibstadt. Grand Gulf 1 & 2. Perry 1, and Susquehanna 1 & 2 are FD1 (Field Disposition Instruction). Clinton is scheduled to be done by FDI. CNV 1 & 2, and TVA will be corrected in C&l before shipment.

Note: No mention was made of any notification to the owners of these facilities about the possible defective switches as required by federal regulations."

Inspection Finding - The meeting referred to is the June 10, 1980 Problem Review Board meeting on the mode switch. It should be noted that a follow-up review was held on October 10, 1980 to determine the adequacy of the housing design change proposed to resolve the problems identified during the June review. This followup review identified specific modifications to be u de to the mode switch housing assembly. Mr. Stokes' notes for Problems 2, 3, and 4 identify GE actiuns concerning the installed and previously shipped assemblies. GE stated that the projects group for each plant was contacted to notify them of the mode switch housing

' redesign changes. The only plant that experienced any problems was Susquehanna 2. The mode switch was replaced with a switch of the modified design by using appropriate procedures. The replacement was accomplished via ECN NJ 21793 and FDI MDCA.

The inspector interviewed the GE responsible engineer for the mode switch during the 1980 time frame. When the design modifications were agreed upon in the October 10, 1980 Problem Review Board meeting, all applicable plants were notified of the design modification via ECN's 21792 and 21793.

Mr. Stokes' note stating that owners of facilities were not notified about possible defective switches is' incorrect. The switches were never determined to be defective. The utilities were notified verbally and design modifications were identified via ECNs. Problems listed as one and six were previously discussed in Stokes Report Sections 1.26 and 2.23, respectively. This item is closed.

Stokes Report Sections 2.36, 2.37, and 2.38

! Section 2.36 "On July 11, 1980, Mr. Milam had a meeting with Chuck Hart

( and John Cintas. John was not aware of open items and work connitments on the reactor mode switch problems."

ORGANIZATION: GE NUCLEAR ENERGY ,

SAN JOSE, CALIFORNIA REPORT INSPECT!CN h0.- 99g00403/89-01 RESULTS: PAGE 23 of 62

~

Mr. Stokes' Comment: "Mr. Milam's concern over manpower to resolve the problems is well founded. These problems could slip through a crack and never get fixed. See Section 2.35." " '

Section 2.37 "On July 21, 1980 Mr. Milam checked again with Currie and he said that Roger Waldman told him that Mr. Milam was going to get a letter saying the PRC was not reportable. During the last week of July, Mr. Milam heard a rumor that Licensing might report the PRC after all." ,.

i Section 2.38 "Mr. Milam followed up on July 29, 1960 by asking Chuck Hart about the eode switch resolutions. He was told that Jerry Willis was no longer responsible. The new person was Flor Cadigal. On July 31, 1980, Mr. Milam called Flor to offer his aid, if needed. In a discussion with Flor and John Cintas, he was told that the new design was complete, and the Development Shop was building one which would be given a shake test.

See Section 2.36."

Insoection Findings - These are general coments made by Milam in his wort record concerning the mode switch evaluation. Section 2.26 and 2.35, in this report, discuss these coments and the evaluation by engineering. This item is closed.

Stokes Report Section 2.41 "A June 26, 1980, letter points out another problem. This letter states that 17.8 percent of the ECN's initiated by product design engineers were rejected by J. Cooper /D. E. Lee.

Mr. Stokes' Coment: This indicated the lack of a quality training program. This is evidenced by other problems noted in this report. See also Section 1.5."

Inspection Finding - Mr. Stokes' report identifies what is considered to be an excessive number of corrections to ECNs issued by product design engineers. This problem is identified as a basis for an improved training program at GE. (Refer to NRC Inspection Re discussion of Nonconfort..ance No. 86-01-05.) portThisNo. 87.01 item for the is closed.

Stokes Report Sectioi1 2.42 "During the first week of July, Mr. Milam wrote in its work record, does our component have an Engincaring Change Request log? Also noted was this coment, per later conversation, we'd rather not. (Emphasis added by Mr. Stokes.)

Mr. Stokes' Coment: This was in violation of GE's procedures and 10 CFR 50 Appendix B. Section !!!, Design Control. See Section 2.39."

. ORGANIZATION: GE NUCLEAR ENEAGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: 1 PAGE 24 of 62 Inspection Finding - Mr. Stokes' entry indicates that Mr. Milam's work cunpenent does not have an Engineering Change Request (ECR) Log and that this constitutes a violation of GE design control procedures. This state-ment is incorrect. GE employed ECRs for design changes external to GE NE80 Engineering and Engineering Change Notices (ECN) for internal changes. The method used to connunicate with various groups at GE was to useCorrectiveActionRequests(CAR)andhavetheresponsibleengineer prepare an ECN to resolve the CAR. This item is closed.

Stokes Report Section 2 43 "While researching for a solution to CAR 5J-53827. Mr. Milam dis E ered other problems with Hope Creek H11-P617 PL913E971, Revision 0. Terminal Board (137C6387P030, Revision 0) doesn't fit Teminal Module Housing (11201935G003, Revision 2). Hole positions don't match. This p'oblem exists for other panels. Assembly parts list (PL913E917 Resision 0) calls for Terminal Board Assembly (112019356006).

Teminal Board Assembly Parts List specifies that G006 should not be used.

Mr. Stokes' Comment: With 12-digit codes and identification numbers, it is understandable that so many errors are missed during the verification process . There was a problem editing these numbers to get them correct, if indeed they are correct."

i ins)ection Finding - Mr. Milam identified a particular terminal board fit

pro)lem. ECh hJ-17969 was prepared to correct the fit problem by the responsible engineer. This was in accordance with GE procedures, and there was no evidence of incorrect use of the 12-digit code or that it was too complex to be used by the staff. Documentation was not available to support Mr. Stokes' coment. This item is closed.

Stokes Report Section 2.44 "On July 7, 1960. Larry Harper promisto to show Mr. Milam the Quality Assurance Procedure (QAP) that supports his contention that he need not check elementary diagrams. Ken Seibert l supports this belief, also. Mr. Milam disagrees based on MP3.11, which says the elementary diagram is a controlled document for Dispiay Control System (DCS), see Section 2.41.

Inspection Finding - The applicable Quality Control rrc:edure (QAP) 12.1 i refers to the following controlling Manufacturing Pror.edures: MP 3.11, MP 3.32, and MP 5.09. MP 3.11 lists the elementary d?agram as a document to be controlled, as stated by Mr. Milam. However, it is only applicable if called for by Engineering Instructions (EI's) or the Parts List (PL).

Neither C these documents specified elementary diagram 828E256AK in the manufacturing of CNV 1. G41-1050. The elementary diagram was not listed on the drawing control sheet referred to as the Display Control System (DCS). As a result, Larry Harper and Ken Seibert's statements were correct.

t

1

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIF 6RNIA 1

REPORT INSPECTION NO.- 99900403/88-01 RESULTS: PAGE 25 of 62 During the shipping inspection process, the check by manufacturing and quality control to assure design verification status did not include the elementary diagram or ECN NJ-07991 which became Revision 1 of the equip-ment list because it was not a required document. Ther. fore, the panel was shipped correctly per procedural requirements. The deferred status of elementary diagram ECN NJ-07991 was cleared by Defarred Verification Status Change Notice 00316 on November 10, 1983. This item is closed.

Stokes Report Sections 2.45 and 2.46 Section 2.45 "An" example of the problems within this department of GE occurred whc .am deviated from his normal practice of just getting the stat 6 .A production control. This time he asked to see the enclosures (Ho p Creek 1 & 2 H11-P613 and Hope Creek 1 & 2 H11-P619).

Production Control did not know where the enclosures were, or if they had been built. Three were finally found by Mr. Milam. One H11-P613 was labeled Unit 2; he had been told only Unit I was built. The two H11-P619 enclosures were botn labeled Unit 1. not Unit 1 & 2. None were where he had been told they were. The fourth was not located.

Mr. Stokes' Coment: There labeling errors are a violation of 10 CFR 50, Appendix B.Section VIII, Identification and Control of Materials, Parts, and Components."

Section 2.46 "The enclosure problam continued into the second week of July. At tTe production meeting on July 8,1980, Mr. Milam was told by Phyllis Fitzpatric that he shouldn't worry abai~ what it says on the metal tag on the panel. He should just go by the work . Je r . See Section 2.45."

Usection Findings - The enclosures were labeled by production for their di local use but were not identified as e designated item with a label and tracked until the assedly work began. The review of the ECNs, l deawings, and parts list for Hope Creek I and 11 panels H11-P613-P619 verify that the items were identified and tracked by .he GE system. The local GE work unit labeling may have been ircorrect but the documented GE tracking systems identified and tracked the items in accordance with established prucedures. This item is closed.

Stokes Report Sectico 2.49 "This same week, Mr. Milam wrote a letter to D. l. 3 perry concerning the procedure requirements of E0P 55-2.00, Engineering Change Notice anc MP 2.28 Change Request Procedure. Per E0P 55-2.00, paragr ph A3.1, until approval and issue, the ECN is a Change Request, some:imes referred to as an Engineering Change Request (ECR)....

This statement requires that all ECNs writSen by engineering be processee

ORGANIZATION: GE NUCLEAR ENERGY' SAN JOSE, CALIFCdNIA REPORT INSPECTION l NO.- 3900403/88-01 RESULTS: pAGE 26 of 62 l per MP 2.28 as ECRs prior to normal processing as ECNs. To sati'sfy requirements of rip 2.28, the Product Design Engineering Section thould have a Change Request Coordinator, responsible for maintaining a Change Request Log, logging all Change Requests, and issuing a quarterly report'of outstanding Change Requests. Since our ECNs are never treated as ECks, we are far out of compliance with the requirements. See Section 2.42."

Inspection Findinj - Mr. Milam wanted to employ a change request coordinator ano '.ag for documents created external to GE Engineering. l GE E0P 55-2.00 allows the use of a unique ECR number when the fonn is generated external to Engineering. MF 2.28 was a manufacturing procedure not used by Engineering. Mr. Milam was required to provide an ECN for each ECR he r:ceived with tracking and control of the document provided via the Engineering Information System. The system identified by Mr. Milam suggests an improvement over the existing system and not a C

requirement. This item is closed.

Stokes Report Section 2.50 "Continuing, Mr. Milam says, I am having trouble with CAR SJ-5619f.' The panel can't be built the way it's designed.

Three panels already shipped, I assume ' bootleg' modifications were made by the shop and accepted by QA but I have no idea what was done. CAR SJ-56192 was written against item G36-P002 for Grand Gulf I'.. See also Section 1.16."

Ir.spection Finding - This item refers to CAR SJ-56192, item 198. Produc-tion requested more information on the assembly of this item from engineering. The available drawing was unclear and the panels in produc-tion were completed and shipped without documentation changes. The remaining panels werc modified to an ECN prepared by engineering. No safety implications were noted and the assembled unit is considered to be the same as the drawing change per the ECN since the quality c.ontrol verif t:ation of the panels did not identify an error prior to their release for shipment. The first panel was completed by an obvious completion of task not clearly identified on the drawing. The problem was corrected and

! no panel assembly error was ever idqntified. The ECN provided only clarification. This item is closed.'

Stokes Report Section 2.53 "On Augnst 7,1980, a Quslity Assurance News Letter was circulated. The first item was of interest. It concerneo the Handatory Distribution List (MDL). The article stated that recently a list was discovered which was three years out-of-date. To correct this use of uncontrolled documents, the MDLs will be dated.

1 -

ORGANIZATION: GE NUCLEAR ENERGY {

S.\N 00SE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 27 of 62 Mr. Stokes' Comment: It was still up to each individual to ver1'fy that his document was the latest issue. This method does not ensure that out-of-date documents are not used. This loophole violates 10 CFR 50, ,

Appendix B.Section I.

Synopsis: Not only Fas each MDL uncontrolled, but each responsible

' engineer had to assure that the person checking their design documents was not involved with the original design when this checking function serves as independent design verification.

Mr. Stokes' Coment: This was not possible when the work had been assigned to many people in the past. But management knows or should know who has worked on a project, since they made these assignments."

Inspection Finding - Mr. Milam's work record identifies the addition of an issue date to the MDL to ensure that only current lists would be utilized. The additional Stokes coments refer to the design effects on the manufactured parts. The MDL is a managenent tool to ensure that correct design and design reviews have been scheduled but does not affect the design detailr. or the control practices. A 10 CFR 50 control program provided the necessary design controls. This item is closed.

Stokes Resort Section 2.54 "On August 11, 1980, El MY-H13-P604 released T15D6358A)G001, Revision 4 for manufacturing for CNV 1. However, the drawing was not applied to CNV in EIS. Per a conversation with Bob Gaetani, an ERM for MPL application only should have been written concurrently with the EI, but apparently it was cet.

Mr. Stokes' Coment: Here two problems exist. (1) EIS is not set up for j the necessary recoros retention for the level of quality assurance dependency placed on it and (2) holes in the procedures allow information to slip through unrecorded. Both violated 10 CFR 50, Appendix B. Section j t

VI, Document Control. See Section 2.49."

, Inspaction Findina - The initial release of drawing 115D6358ADG001 for

panel H13-P604 was for the purpose of advance accumulation of parts.

Final release for manu scturing was made by page 4 of Engineering Instruc-tion NY H13-P604 and the drawir j number was changed to 14707545G001 on Octobr B,1980 and applied to CNV 1 by ERM AML868. The conditional

(# ted) release of engineering requirements is procedurally addressed ir '0P 42-5.00 "Engineering Requirements Document Release," Revision 1 Ju.. A7. Paragraph A2.2. It was concluded that Mr. Milam's concern wat aver, it addresses only a portion of the process which was ult)..:ateiy carried to proper completion. This item is closad.

i

\

. ORGANIZATION: GE NUCLEAR ENERGY  ;

SAN JOSE, CALIFORNIA 1 REPORT INSPECTION NO.: 99900403/88-01 RESULTS: "

PAGE 20 of 62

~

Stokes Report Section 2.55 "On August 13, 1980, Mr. Milam rejected two ECN's written by P. DeBenedetto. In nis letter, Mr. Milam Stated, ...the reasons are too numerous to include here, but the majority of pfbblefas were the result of noncompliance with MP 2.2G.

Mr. Stokes' Coment: This was not the only reference to problems with MP 2.28. Without additional facts, the reason for this problem cannot be precisely determined, but one cause of it might be lack of training.

Mr. Milam said that he once told Mr. Senn (Manager of Cal) that he learned to do ECN's by doing them wrong,"

Inspection Finding - Mr. Milam rejected two ECNs written by a production staf f member, due to incorrect completion of the fonn; not for technical reasons. Mr. Stokes' :omaents are directed to a lack of training; however, GE procedures state that trafning courses are established by the unit manager as needed. A detailed examination of the work packages indicated

, only a deby in wors progress resulted, since the nacessary fonns were eventua'ly completed by Mr. Milam and the necessary document changes were identified. This item is closed.

Stches Report Section 2.56 "A quickletter from Denris York on July 28, ,

1980, to Mr. Milam requested the definite location for cutout J18, drawing 14707020 for enclosure H13-P629 CNV 1.

Mr. Stokes' Comment: The infonnation was supplied by Mr. Milam, but his reply is troubling: While this is acmittedly an example of sloppy draf ting, I do not feel that I can justify the cost of an ECN to correct it. Was this sloppy drafting or shoddy verification by engineering? In either case, 10 CFR 50 Appendix B does not state that cost should enter into consideration where the error involves safety-related equipment.

Tailure to correct this error is a violation of 10 CFR 50, Aprendix B, Section 111."

Inspection _ Find't9 - Drawing 14707020 for Enclosure H13-P629 for CNV I was revieweu. Cutout J18 is not shown c Good draf ting practice would indicate a d}early but is not rawing change forincorrect.

this portion of the drawing. It was noted that poor drafting practices and checking existed; however, no violation of 10 CFR 50, Appendix B was 1Jentified. This item is closed.

Stokes Peport Section 2.59 "On August 20, 1980 Mr. Milam wrote, CAR SJ-51646, CNV isolator problem was still open. I am sorry I omitted the first record for this entry, but this infonnation is contained in the work recoros attached to this report."

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA

- l REPORT INSPECTION NO.- 93900403/8_8-01 RCSULTS: PAGE 29 of 62 Inspection Finding - A review of Corrective Action Request (CAR) No.

5J-51646 indicated it opened on July 18, 1980 with a reply scheduled for i July 22, 1980. On September 18, 1980, the CAR was closed out arid signed by Mr. C. J. Cobler. The CAR stated that the work referenced wa's to be performed by FDI; however, a further review revealed chat it was actually closed by an ECN. ECN's NJ P1586, dated January 15, 1981, and ECN NJ 20290, dated November 4,1980, were generated to incorporate the new isolator design. This item is closed.

Stokes Re3 ort Secti,n 2.60 "On August 21, 1980, Mr. Milam supplied 57 pages of ECNs from his files to the shop, who were having trouble getting ECNs together for TVA 17 G36-P002. Some were apparently not logged by Production Control.

Mr. Stokes' Comment: Apparently procedures are not in place for the control and distribution of ECNs. This violates 10 CFR 50, Appendix B, Section VI. Document Control."

Inspection Finding - Review of manufacturing and engineering procedures indicated that all procedures for the distribution of ECNs were in place at this time. It is possible that Mr. Milam had copies of the ECNs prior to their being logged by manufacturing to perform the work. Also, Mr.

Milam's work record indicates that the print room had backlog problems.

It is concluded that, although a system was ciready in place, Mr. Milam expedited the distribution of ECNs. This item is closed.

t Stokes eport Section 2.61 "According to Mr. Milam's work record, E0P 55-2.10, Floor Engineering Changes, issued July 9,1930, Paragraph 4.1 states Floor ECh coordinators are needed to write Floor ECN's. It also states in Paragraph 4.3 that the responsible manager is to appoint Floor ECN Coordinators. Mr. Milam did not know of any.

Mr. Stokes' Coment: Having a procedure which is unnecessary is not a violation of 10 CFR 50, but not following a procedure which is in place is a violation of 10 CFR 50, Appendix B Section II, Quality Assurance Prog ram. "

Inspection Finding - GE Procedure E0P 55-2.10 identifies the appointment of a floor ECN coordinator but no formal procedure is required. The responsible manager for each project selected and verbally assigneri this responsibility on a project-by-project basis. No violation of GE proce-dures was identified. This item is closed.

l t

^

• ORGAhlZATION: GE h0 CLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION

_ NO.! 9960NO3/88-01 RESULTS; i PAGE ?O of p 3

Stokes Report Section 2.62 "In the last week of August, the fo'llowing problem was noted, r.eferencing 913E345, Revision 0, Connection Diagram for CNV, TVA, Clinton, and Black Fox, Mr. Milam states that thir drawing derives from different elementary diagrams, depending on the project.

Experience indicates that the different elementary diagrams (for the same system) will not be identical. These differences will appear as discre-pancies when panels built to the referenced drawings are tested to the various elementary diagrams. Additional references: Elementary Diagrams 914E108, Revision 2, Elem(ntary Diagram 386X994-003, Revision 1, and CAR SJ-56806. See Section 2.8."

Inspection Finding - A single connection diagram, 913E845, Revision 0, for ChV, TVA, Clinton, and Black Fox projects was utilized. As deviations due to customer regrests and for other reasons arose on the different elementary diagrams, General Electric did, in fact, customized the connection diagrams to meet the specific requirements of each plant elementary diagram and gave each a unique identification. This process is consistent with General Electric's design control procedures. The connection diagrams are 913E845AK for CNV, 913E845CE for TVA, and 913E845 for Clinton. (Black Fox was later cancelled.) It w3s concluded that Mr. Milam's concern was valid; however, General Electric prepared unique drawings for each project in accordance with established procedures.

This item is closed.

Stokes Report Section 2.63 "The mode switch PRC which was written May 28 1980, was delivered to licensing on August 25, 1980. Mr. Milam feels that this delay is in violation of: Section 206 of the Energy Reorganization Act 1974, Paragraph (a), Subparagraph (2); Title 10. Code of Federal Regulations, Part 21. Paragraph 21.21, Subparagraph (2) (10 CFR 21) (Note Paragraph 21.61 Failure to Notify); Nuclear Energy Business Group Proce-e.fre (OPG) 70-42, Application Section, Paragraph IV, Subparagraph A (Note Section VII, Fines); and MP 5.08, Paragraph 4.6. Mr. Milam notified his manager, Mr. C. L. Cobler, of the above cn August 27, 1980."

Inspection finding - The reason why the mode switch PRC was not forwarded to licensing until August 26, 1980 was discussed previously in the inspec-tion finding following Stokes Report Section 2.26.

The comment concerning a violation of a 10 CFR Part 21.21 is not correct because the time limit for notification begins after it has been determined by a respunsible director or officer that a reportable condition does exist.

This has nothing to do with internal organization notifica ions.

This item is closed.

l l

l i

.s

, ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA 1 _

REPORT INSPECTION NO, ; 99900403/88-01 RESULTS: '

pAGF 31 of 62 <

i Stokes Report Section 2.64 "On August 28, 1980 Mr. Milam drafted some ECN's from mark-ups made by Russ Thompson, et al. He later found he was expected to sign the ECN's as responsible engineer without knowthg why the changes were required or whether they were correct.

Mr. Stokes' Connent: Similar problems are reported in Part I."

Inspection Finding - Mr. Milam drafted some ECNs from information supplied by others and then issued the documents in accordance with GE procedures.

Mr. Milem appeared to be satisfied with the information by the time the documents were issued. This item is not a violation of procedures, just a concern expressed by Mr. Milam of the initial work assignment.

This item is closed.

Stokes Report Section 2.65 "In a document dated August 29, 1980, entitled, Multifunctional Manufacturing Failure Analysis Team Minutes, the following topics were of interest: first, the removal of saddle clamps on CR2940 contact blocks. According to this document, engineering decided to qualify the CF2940 switches (without saddle clamps) as part of the BWR/6 requali-fication program. These tests were to begin in November 1980. See Section 2.32."

Inspection Finding - This item is a continuation of Stokes Item 2.31. The engineer decided to qualify the switches without "saddle clamps" for the BWR6 qualification program, and a standard GE work procedure is referenced.

This item is closed.

Stokes Report Section 2.66 "The second topic of interest concerned the lack of a revision number in all references to Manufacturing Standardiza-tion Procedures (MSPs) on drawings. It was stated th.t per C. L. Cobler (Engineering), M. Hurn, Manager of Drafting, had been directed to delete any reference to MSPs on drawings.

Mr. Stokes' Comnent: If information shown on the MSPs is utilized in the design onsafety-related hardware, then MSPs should be referenced somewhere in the design documents. See Section 2.52."

Inspection Finding - The MSPs were considered standard shop practices l

which were updated and followed pe: the latest dated instruction. These MSPs are not a part of the design control procedures. Use of a dated MSP is noted on the manufacturing division work recoro so process procedures used for any work item can be identified. This item is closed.

)

ORGANIZATION: GE NUCLEAR ENElGY SAN JOSE, CALIl0RNIA 1

REPORT INSPECTION NO.: 99900403/88-01 RESULTS: pAGE 32 of 62 Stokes Repurt Section 2.67 "On September 2,1980, J. E. Morrissey wrote a letter to F. Sterzinar about MSP revision numbers. In this letter he returned two work orders which did not have blanks for the inseFtion',of revision number of the MSP actually used. He requested that the two work orders be corrected and that a blank be provided for the revision number on all issued planning documents, where applicable.

Mr. Stokes' Connent: It appears that Mr. Cobler did not discuss his action in Section 2.66 with his superiors, since Mr. Narrissey feels that the revision number is important enough to demand that spaces be provided for them."

Inspection Finding - The u:e of revision nual.rs for MSPs is noted only on the manufacturing work records, not the engineering documentation for which Mr. L. Cobler was responsible. The confusion existed during a transaction in the noting of these revision numbers. The new procedures deleted the revision numbers, with the information available as a dated procedure and the date of work. No work outside GE procedures was iden-tified. This item is closed.

Stokes Report Section 2.68 "It is recorded in the work record for September 3,1960, that Mr. Milam was correcting more errors in G36-P002 design for Kuo Sheng, Grand Gulf, Perry, and Nine Mile Point. The panels had already been shipped to Kuo Sheng 1 and 1 and Grand Gulf 1. See Section 2.60."

Inspection Finding - ECH NJ-18237 was prepared to correct the minor errors identified on panel G36-P002 during the assembly for Grand Gulf 2 Perry 1, and Nine Mile Point 2. No errors were identified on the shipped panels.

The m': or errors were corrected to reflect the as-Luilt condition of the panels. It was concluded that Mr. Milam's concern was valid and that appropriate corrective action was taken. The shipped panels were unaffected. This item is closed.

Stokes Report Section 2.69 "Recorded under. FW8037 (second week of September) is a novel practice calle'd mortgaging ECN's. Mr. Milam says

! he mortgaged 2 ECN's on 2 of the IR'.s."

Inspection Finding - A mortgaged ECN refers to the practice of )reassigr.ing a control number to the ECN; however, all controls used to tract and issue the ECN remain the same regardless of the single track or multiple track

work progress (mortgeged). No violations of design control procedures could be identified. This item is closed.

l I

i

.o a URGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS:

PAGE.33 of 62 Stokes Report Section 2.70 "On September 10, 1980, Mr. Milam w' rote a letter to CL Cobler about fab shop design changes. In this' letter, he says the feb shop sometimes invents its own fixes to design problems'instead of asking engineering for clarification or design changes. There was no documentatiun for these changes nor was the drawing changed. See Section 2.49."

Inspection Finding - Mr. Milam noted feb shop fixes unauthorized by engineering. A review of the panels referenced in the letter, G36-P002, as applied to Clinton, Black Fox, CNY and TVA was undertaken. The review produced ECNs authored by Mr. MOam to correct the stated problems. No evidence of unauthorized fab sh y fixes was identified. This is nnt a safety-grade panel, and no woik outside the GE procedures was identified

, by shop planning cards. This item is closed.

Stokes Report Section 2.71 "On September 15, 1980, Production Control couldn't find enclosure (CNVI H13-P632). They thought it was completed and painted. Mr. Milam found it in the fab shop. See Section 2.45."

Inspection Finding - Panel enclosures, which are the front end building block of the panel assembly, are often the same for many projects at this drawing level of assembly. Later, specific panel enclosures are labeled by controlled drawing irientification for the higher level of assembly and, ultimately, for the final panel level assembly. Correct enclosures were

' used in assembly, testing, and certification of the panels shipped to the customer. Product Quality Certifications were reviewed that identify conditions of shipment (i.e., shorts, ECNs, release conditions, etc.). It was concluded that Mr. Milam's concern was valid, but the resolution was

, part of the normal production process. This item is closed.

Stokes Report Section 2.72 "During the third week of September, Mr. Milam found that an ECh had not been placed in the Engineering Information System (EIS). He corrected it."

Inspection Finding - No further information is available to indicate how of ten this situation occurred or whether the EIS verification process missed this entry. This item is closed.

Stokes Repo_rt Section_2.73 "Marge Zych insisted that TVA 22 C71-P001 required an El for Conditional Shipping Release. Mr. Milam discovered that an Engineering Instruction (EI) had already been written."

Mr. "itokes' Coment: Why didn't Marge find the EI?"

, ORGAHlIATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: <

PArJ to nf A9 Inspection Fir, ding - The GE expediter was properly perfor1ning her duties by attempting to assure the availability of an El prior to shipment of a panel to TVA. The responsible engineer made the El available te the' expediter to permit shipment of the panel. This item is closed.

Stokes Report Section 2.74 "Mr. Milam discovered an unauthorized change was made to Connection Diagram 866E151 for CNY H22-P028 when ECN NJ-11332 was incorporated. He says, I must now change it back."

Inspection Finding - An unauthorized change was noted to the diagram when ECh NJ-11332 was implemented. Evidently the draftsperson made changes not authorized by the ECN. This was noted in the field by CAR SJ 59003, and an ECN was issued by Mr. Milam to correct the error. The GE processing system identified the error and corrected the problem to this non safety-related par,el. It appears that the drafter's training did not make it clear that changes outside the ECN were not to be added. No other examples were identified during the inspection. This item is closed.

Stokris Repurt Section 2.75 "About EIS, Mr. Milam said that this informa-tion system is reliec upon in all sections of GE and by vendors to trace a documents status and open documents tied to a particular document. This system is required because the status of a document does not appear on the document. Thus, many documents which have inadequate verification

are used in the verification process of other nuclear safety-related I design documents and drawings. This system is set up to maintain only the l

most cm rent document information. It does not maintain a permanent record, ane which can be used by QA to verify the past. If it did contain a permanent chain of events (i.e., documents issued) then piecing together the past would be possible. The use of this system to be a source of status level for safety-related documents is a violation of 10 CFR 50 Appendix B. There is no reason for the system not being used as a cross-reference ' record system, but should not be used in final verification.

The status of a document should be obvious on the document itself. This is the most fool-proof method of control over document use."

Inspection Finding - The Engineerin Information System (EIS) is an on-line status list foF engineering design documents. This listing does not contain design information and therefore is not part of the design contrul portion of the 10 CFR 50 Appendix B program. The design verification process requires the design documents for final verification and EIS would serve to identify the listing and status of these documents. A history of the documents is not provided nor was it ever intended. This ite.n is closed.

1 \

l l \

l l

l l l 4

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: PAGE 35 of 62 Stokes Report Section 2.76 "CAR SJ-59004 initiated by Pho Le, CAR SJ-59006 initiated by George Rea, and CAR SJ-59007 initiated by Samuel Aveno all Indicate that these people do not understand the use qf EIS and/

or other cocument research methods to be used to find the necessary reference documents. Here the documents needed were ECN's."

Inspection Finding - It was discovered that Mr. Milam had referenced the wrong drawing number on ECN NJ12997 as being affected (drawing 828E535AK was referenced instead of the correct drawing 828E537AK). The EIS there-fore yielded the wrong information. Mr. Milam subsequently realized his mistake and inued CAR SJ-59022 to correct the error. This item is closea.

Stokes Report Section 2.77 "Several problems were noted during the first week of October. The first was an old one. Production control was still having problems keeping track of panels. This time, two G36-P002 panels were both labeled Grand Gulf. The Grand Gulf panel was shipped already.

One of the two panels must be for Perry. See Section 2.71."

Inspection Finding - Production control utilized an informal tracking system for work within their unit. Two panels per drawing 851E767 ABG002 were verified to be correct by Production Quality Control (PQC) documents.

Also, Perry Panel PQC documentation dated September 17, 1981 was examined, and correct completion of this panel was indicated. It appears the formal

. tracking system which is utilized as the panels are assembled was correct in its tracking of these panels. This item is closed.

Stokes Report Section 2.78 "The second problem involved the hold point for separation sign-off on panels. The separation sign-off (by engineering) for CNV H22-P028 did not happen until final ship preparation. Then separation sign-off for Cofrentes H13-P620 did not happen until final ship preparation.

Mr. Stokes' Coment: This indicates the sign-off hold points were missed by QC, This is a violation of 10 CFR 50 Appendir B Section X, Inspection."

Inspection Findins - A Review of Inspection Instruction II PA-006, Paragraph 3.4, incicates that the final entry on the Records Checklist (RC-394) to be completed is item fl. "Records Check List." This must be signed-off by the quality control inspector after all other items have been signed-off. Review of the dates on the two sign-off sheets indicated that both were signed-off last in accordance with the procedure. It was concluded that Mr. Milam's concern was not valid. This item is closed.

.. l

~

ORGANIZATION: l GE NUCLEAR ENERGY '

Sari JOSE, CALIFORNIA o I REPORT INSPECTION NO.: 99900403/88-01 RESULTS: '

PAGE 36 of 62 Stokes Report Section 2.79 "On October 1, 1980. Thomas Regenie and Eleanore Schock of the Safety and Licensing operation stopped to,,see, Mr. Milam. They asked some questions about the mode switch relating to assembly and contact desynchronization. Regenie requested a package of data, letters, etc., relating to the mode switch problems. Mr. Milam was instructed by W. Marklein and C. L. Cobler to giver Roger Waldman a copy of the Tamper Test Report for transmittal which was not what was requested.

Mr. Stokes' Conment: The people responsible for the notification of the NRC and customers of safety problems should be given all relevant infonna-tion on the subject known by the company. The act of preventing this information from being given to these people is a violation of 10 CFR 50, Appendix C." See Section 2.84.

Inspection Finding - Mr. Stokes conment about preventing information from being given to the people responsible for notification to the NRC is incorrect. Mr. Milam never stated this about his management; in fact, his work record further stated that he later contacted (fiscal week 8040)

Schock and advised her that the information would have to be requested through Roger Waldman in the future. This does not indicate withholding of information. This item is closed.

Stokes Report Section 2.80 "On October 9, 1980, Mr. Milam attended a Design Review Meeting on the redesign of the reactor mode switch 195B9497.

According to this work record, two items were not resolved in accordance with the action proposed in the letter of June 13, 1980. See Section 2.36."

Inspection Finding - The October 9,1980 meeting of the Problem Review

, Board (PRB) was a follow-up to the June 10, 1980 meeting. All six items from the June PRB were addressed. In some cases the actions were not exactly the same as proposed in the June 13, 1980 letter concerning the PRB meeting. This can be attributed to the input from the evaluation which

, was conducted during the June 10 - October 10 time frame. All items docu-l menteo as part of the October 9 meeting were agreed to by the entire PRB, except for the two issues with which Mr. Milam disagreed. All issues were resolved satisfactorily by the PRB, with Mr. Milam dissenting on the two l issues. This item is closed.

Stokes .leport Section 2.82 "During the third week of October, C. L. Cobler told Mr. MiTam to remove all production folders for shipped panels from the production files. He said he did not want the responsi-l bility for shipped panels and that Mr. Milam should either throw them away or give them to the FD1 people if they wanted them. Documents were removed for the following plants: Cofrentes, Fermi, Hatch, Kuo Sheng, Laguna Verde, LaSalle Liebstadt, Lilco, and Zimer.

l l

l l

l

• ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO. - 99900403/88-01 RESULTS: PAGF 37 of 6?

Mr. Stokes' Coment: If important information was contained such as original signatures of authorizations and hold-points, etc., were destroyed, this action violated 10 CFR 50 Appendix B Section II, Quality Assurance Program."

Inspection Finding - General Electric procedures require retention and maintenance of original controlled documentation consistent with the requirements of 10 CFR 50 Appendix B. The production folders Mr. Milam was directed to discard were only copies of the original documents and were not required to be retained. Therefore, no NRC regulations were violated. This item is closed.

Stokes Report Section 2.83 "On October 22, 1980, Mr. Milam sent a letter to C. L. Cobler concerning Filter Demineralization Control Panels. He referenced drawings 913E844, and 913E845 for Clinton, CNV, Black Fox, and TVA as well as 913E844A, and 913E845AA for River Bend. He said that many changes were made on the first set of drawings, but not all cf them were made before the second set were copied from the first. These are contained on the following ECNs: NJ-18216, NJ-18234, NJ-19311, NJ-19313, NJ-19314, NJ-19315.

Mr. Stokes' Comment: This notification occurred because Mr. Milam was changing jobs and felt that someone replacing him should be aware of this."

Inspection Findings - ECNs for River Bend were reviewed and it was deter-mined that the design changes were in fact incorporated into the applicable drawings. A check of the dates of incorporation indicated that some of the changes were made on the day that Mr. Milam wrote his letter, thus corroborating the statement that not all changes had been made. It should be nuted that the River Bend drawings were copied from the drawings for the other four projects. It was concluded that Mr. Milam's concern was valid, but ultimate completion of the drawing change process resulted in proper resolution of all changes. This item is closed.

Stokes Report Section 3.1 "On October 28, 1980 Mr. Milam discovered a folder containing the following:

1. ECN NJ-05095, original copy, unissued

2. Enclosure Drawing 159C4563, Revision 2 (2 copies)

3. Enclosure parts list PL159C4563 Revision 1, revision status sheet, and vellums for sheets 0 and 2, all originals, unissued

.s .

. ORGANIZATION: GE NUCLEAR ENERGY' SAN JOSE, CALIFORNIA

~

REPORT INSPECTION N0.: 99900403/88-01 RESULTS: '

PAGF 38 of 62

4. Copy of ECN NJ-07522

5. Copy of PL615904563, Revision 0. Sheet 0 e -

6. Blue card for 14707150 two door enclosure

7. ERM AMC-3173, for standard plant H22-P008, 147D-7150 original, unissued

8. Copy and check print of 14707150, Revision 0, unissued

9. ERM AMC-3083, for general use H22-P008, 169C9423, original, unissued

10. Blue card for 169C9423, purchased part, enclosure

11. Copy of strike, latch drawing 272A6875. Revision 1

12. Check print for strike, latch drawing 272A6875, Revision 2

13. Copy of CAR SJ-39222

14. Various EIS output and hand written notes Mr. Milam did not know how this work got into his boxes, nor the status of the requirements for the work. These documents were turned over to C. W. Hart on October 29, 1980.

Mr. Stokes' Coment: This represents a gross example of carelessness, if nothing else."

Inspection Finding - Mr. Milam picked up a folder by mistake from his supervisor's desk and then returned it the next day. The work entry identifies his original concern when he first discovered the records.

The folder was returned and the items processed by the appropriate persons.

This item is closed.  ; , ,

Stokes Report Section 3.2 "Mr. Milam found what he considered a routine noncompliance with MP 2.28 during the first week of November 1980. MP 2.20 required a Change Request (CR) coordinator; no CR coordinator existed, to Mr. Milam's knowledge ur anyone else's. When he approached his boss, Mr. Reghitto, Mr. Milam was bothered because Mr. Reghitto wanted him to explain why they should change their ways -- that is, get a CR coordinator and work per procedure. See also Sections 2.49, 2.55, and 2.61."

o ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 30 nf 6?

~

Inspection Finding - Mr. Milam identified GE manufacturing procedure MP 2.28 which specified the Change Request Coordinator. Engineering did not, however, have this procedure as an Engineering Operating Pfbcedbre (E0P). Mr. Reghitto suggested the use of existing engineering procedures to identify why a change in the existing procedures should be implemented, it appears Mr. Milam was addressing his personal views and not evaluating GE confonnance to existing procedures. This item is closed.

I Stokes Report Section 3.3 "Mr. Milam received a letter dated December 5, 1980, on the topic of the reactor mode switch marked COMPANY PRIVATE stating that based on a letter evaluating the PRC which Mr. Milam filed against the mode switch, GE does not have a reportable condition. In any case, the problems pointed out would be corrected. The clenching state-ments are taken from the attached letter, page 2: None of the concerns of reference 1 (Letter, Mode Switch, S. Milam/D. Taylor to D. H. Currie /

C. L. Cobler, May 28, 1980 and refs.) are safety problems except for itam

2. The item 2 concern is not a safety problem as long as system and QA checks are completed and documented. Note: Non-PGCC panels are not tested before shipment. Even though not tested, this is not a safety problem since, as discussed previously for item 2, the contacts would be in step between divisions with possible incorrect indexing. Incorrect indexing is not a safety problem. To get contact out of step between divisiens would require deliberate action.

Mr. Stokes' Coment: The act of disassembly and reassembly by a repairman without adequate documentation and warnings for that part is deliberate action not considered or ensured against by this evaluation."

Inspection Finding - Mr. Stokes comment once again deals with the potential for the mode switch to be positioned out of phase during disassembly ano subsequent reassembly to the switch isolation housing. This concern was previously addressed in NRC Inspection Report No. 99900403/87-03, Section E.6. This item is closed.

Stokes Report Section 3.4 "On December 15, 1980, Mr. Milam found out that Engineering was using a form called a Program Information Request / Release (PIR). A PIR was being used in place of the Engineering Change Request.

No procedural change or basis was recorded in Mr. Milam's work log for the use of these documents.

Mr. Stokes' Coment: This is a violation of 10 CFR 50, Appendix 3 Section II."

1

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: '

PAGE 40 of 62 Inspection Finding - The PIR was a "quick letter" type of form f'or an information system prog am which was utilized within GE for-a brief period. The PIR form was not a substitute for the Engineering Ghange l

Request and was nnt a part of the GE E0P (10 CFR 50) program. No lack of conformance to the GE-controlled procedures was identified. This item is closed.

Sto_kes Report Section 3.5 "Mr. Milam took the initiative again during NS101 with respect to the reactor moce switch resolution.

Reference:

1. Item 5 FW8022

2. Letter, W. M. Barrentine and D. H. Ferguson to S. A. Milam and D. C. Taylor, Reactor Mode Switch, December 5, 1980

3. Letter, W. H. Hendrix, B. P. Grim, and D. W. Reigel to W. M. Barrentine Reactor Mode Switch PRC evaluation, December 5, 1980 On December 30, 1980, Mr. Milam called S. Hassan and expressed concerns with Reference 3. The letter states that in at least three separate locations problems do not exist because the existing configuration is the same as the qualified configuratio1 Mr. Milam stated that he was aware that the mounting configuration used in Kuo Sheng and Cofrentes was, in fact, different than that qualified. As of December 30, 1980, no analysis or testing had been performed to assure that the existing configurations were the same as those qualified.

Mr. Stokes' Coment: If Mr. Milam is correct, it is apparent that the problems would be reportable. See Section 3.3."

Inspection Finding - The seismic qualification of the horizontally mounted Feactor mode switch for Cofrentes and kuo Sheng is documented in Wyle Test Report No. 58459 dated August 27, 1979.. This report documents that the existing configuration at both plants is the same as that shown in the test report. Design Record File. A00-696 (Drawing No.163C1487) contains this qualification documentation.

The inspector reviewed information contained in GE's seismic qualification test report H13-P680, ' Prototype Center Enclosure Compact Principal Plant Control Console." This infonnation was part of Design Record File H13-50, dated July 28, 1978 and was reviewed to determine acceptability for mode l switch configurations other than Cofrentes and Kuo Sheng (horizontally mounted). The test docunented that all switches operated correctly before, i

l l

l l

i

, ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: PAGE 41 of 62 during and after the test and demonstrated no structural changes" or discoloration from exposure due to temperature variations applied during the test. During the inspection a Memo of Record dated August &,1987 was revised by GE to include a reference to reactor mode switch drawing 163C1487. This revised letter, dated February 8,1988, now properly references both mode switch drawings and provides a more complete qualifi-cation of the reactor mode switch. This item is closed.

Stokes Report Section 3.6 "All NC and ID salaried employees were invited to a meeting with G. A. Senn, General Manager, to be held February 11. The invitation requested that questions of concern be sent to Dick Jones.

Mr. Milam was concerned by what he observed when employed in the production a rea . He felt that even though QA tried hard, that they appeared to be somewhat handicapped by their position in the organizational chain. That is, they worked for George Senn, whose main objective was to ship the product. Mr. Milam felt this fact reduced QA's effectiveness.

Mr. Stokes' Consnent: 10 CFR 50, Appendix B Section I, Organization, states at the end of the first column, "Such persons and organizations performing quality assurance functions shall report to a management level such that this required authority and organizational freedom, including sufficient independence from cost and schedule when opoosed to safety considerations, are provided. See also Section 1.25."

Inspection Finding - The organizational arrangement is presented in the General Electric Quality Assurance Topical Report, NE00-11209, "Quality Assurance Program Description" which has been reviewed and found acceptable by NRC in compliance with 10 CFR 50, Appendix B. Mr. Senn is a department general manager with many sections reporting to him, in addition to productiun. The NRC review detennined that the quality assurance section has adequate freedom of access to him at any time to assure proper attention to quality assurance matters. There is also freedom of access to the Operations Vice President. This item is closed.

Stokes Report Section 3.8 "A letter from R. Reghitto/B. Grim and P. Wang to all applicable Technical Licensing Engineers, dated February 4, 1981, concerning the Susquehanna FCD/ELEM updates made this point: In response to the NRC licensing questions on FSAR Chapter 7. C&I has coninitted to correct its Susquehanna drawings, mainly FCD's and elementary diagrams.

Mr. Stokes' Coninent: In the past, the NRC has issued NUREG's 79-02 and 79-14 requiring verification and as-building of safety-related systems in nuclear plants. It appears the NRC should take action requiring a

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESVLTS:  ! PAGF 47 nf A?

complete verification of equipnent and documentation of components supplied by GE and/or their subcontractors, etc., since GE ts not main-taining up-to-date records of their own initiative or as requiret! by' federal law. See 10 CFR 50 Appendix B."

Inspection Finding - In the review of this issue, GE stated that during the design process various documents including the plant Final Safety Analysis Report (FSAR) may contain some inconsistencies for a short period of time. In the above example, a Function?1 Control Diagram (FCD) update program was undertaken by GE at the request of the utility. During the inspection, a review was conducted of the GE process to assure that plant aesign documentation and FSAR/ system drawings are in agreement.

The review showed that GE NEDO-11209 (September 3, 1986), NEB 0 Quality Assurance Program Description, Section 3.10, Design Change Control, defines the requirement for accurate drawings and documents for each design. The responsible engineer is charged with the implementation of this requirement. Sections 3.10 and 3.11 Field Change Control, state that all changes are to reflect the accurate conditions of the installed equipment. GE E0P 55-5.00 states that all changes shall be in accordance with the Project Work Authorization (PWA) for each design unit. The documents are then turned over to the licensee in the forms requested, which might include outstanoing EChs. This documentation update to the PWA requirement is audited by GE in accordance with the GE 10 CFR 50 Appendix B approved program. This item is closed.

Stokes Reoort Sections 3.11 and 3.12 "Mr. Milam wrote the following question on April 21, 1981 during his QAEE&I Audit Training Class; I understand auditing to be for the twofold purpose of both assuring that all is wall and providing for the initiation of corrective action when necessary. Ultimately, everyone benefits when errors are discovered and corrected. This being the case, why am I so consistently cautioned against saying more than is absolutely necessary b answer an auditor's question?

If I accidentally reveal an error he didn't know about, so what? Once the error is out in the open, we can correct it."

Section 3.12 "A special trair.ing class was seneduled on April 30, 1981.

The invitation fonn stated on the first line that you have been selected to attend the subject training class along with other managers and key contributors from the department. (Emphasis added by Mr. Stokes.) The l training was to acquaint those in attendance with the department's new quality system, I

(

- - _- - - .n

e4 ORGANTZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.! 99900403/88-01 RESULTS:  ; pAGE 43 of 62 Mr. Stokes' Coment: A concern here is whether all personnel attended a

^

training session on this subject. The only people who could be omitted from this training, per Section II of 10 CFR 50 Appendix B, would be the '

employees of GE who work 100 percent of the time on non safety-related components, etc." 1 Inspection Finding - The training class referred to in this item had the specific purpose of introducing all C&ID Design Engineering personnel to a new procedural system called Engineering Internal Procedures. All C&ID Design Engineering personnel were invited to attend. Contrary to Mr. Stokes' assertions, the synopsis / work record does not describe a violation of 10 CFR 50 Appendix B. This item was also covered under Nonconformance No. 86-01-05 of NRC Inspection Report No. 99900403/86-01.

This item describes the GE managerial responsibility for training and docu..:enta tion. This item is closed.

Stokes Report Section 4.1 "On June 2, 1981, Mr. Milam's work record indicates that he was submitted for a Design Record File class on Tuesday, June 16, 1981.

Mr. Stokes' Comment: This entry is not a problem. It is included as documentation of corrective training measures taken, but it is doubtful that this measure will resolve any problems."

Inspection Finding - A primary mode of training is for the supervisor to oversee an employee's work perfonnance to assure that the employee is performing the job correctly. In addition, formal classes are offered, as cited in this concern, to aid management in their assigned responsi-bility to assure that employees are fully indoctrinated in the use of procedures applicable to their work. The latest edition of General Electric Procedure 70-30 "Personnel Proficiency in Quality Related Activities," addresses this mode of training. The prior editior,of this procedure indicated that a new employee shall be indoctrinated / instructed i

in the system procedures within three months of assuming the assignment but does not require fonnal classes. In either event, these approaches to training are acceptable under 10 CFR 50, Appendix B, Criterion II, and are consistent with the comitments made in General Electric Topical Report NED0-11209, "Quality Assurance Program Description," fourth paragraph of Section 2.1 on page 2-1. This item is closed.

, Stokes Report Section 4.4 "During the third week of June, an' entry was made that Mr. Milam gut 7 copy of NEDE 24605, Program Control System Procedure. This document did not have a number on it anywhere."

Mr. Stokes' Coment: "This document should have a control nur.ber for each of its pages."

l l

l

.s ORGANIZATION: GE NUCLEAR ENERGY' SAN JOSE, CALIFORNIA REPORi INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 44 of 62 i

Inspection Finding - This document was a training aid and not an operating procedure. In this capacity, it was not under control for (istribution to individual users. The NEDE series and other noncontrolling docupents are not required to have a control number for each page. This item is chosed.

Stokes Resort Section 4.5 "Mr. Milam also got tired of the poor main-tenance tie oesign manuals were getting in his unit and volunteered to take over responsibility for keeping them up to date.

Mr. Stokes' Coment: It is unlikely that the procedures allow this action.

This should be one of QA's duties."

Inspection Finding - Responsibility for updating the procedure manuals lies with the maoual assignee or delegate. These manuals were the Manu-facturing Practices & Procedures Manual (MP&P) which were not needed in this work unit and were returned by G. Strambach when he took over the leadership of this work group. The assignment of this task appears to be in accordance with GE practice. This item is closed.

Stokes Report Section 4.6 "A latter dated May 18, 1981 to MP&P manual FoTders, statea that many MP's were being deleted from the manuals.

Mr. Stokes' Coment: This is reminiscent of the problem with the lack of revision control over MSP documents. Do the MP&P's have revision numbers?

See Section 2.67."

Inspection Finding - Each MP procedure was maintained by indicating the issue date on each MP and in the table of contents for the manual. The letter referenced in the cement identifies the replacement of these proceoures with new NC&ID Practices and Procedures. The use of dated procedures and the table of contents listings enables the tracking of a current procedure. This item is closed.

Stokes Report Section 4.7 "During FW8126 Mr. Milam details a new problem, the use of NEDE documents in design documents without correct revision s ta tus .

Reference:

1. NEDE 22000, Publications Handbook for the Nuclear Energy Business Group August 1079. -

2. NEDE ?4584 Engineering Documentation Practices Manual (issue date unknown).

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS: PAGr 45 of p

3. NEDO 11209-04A, Nuclear Energy Business Group Boiling Water' Reactor.

4. NEDE 24605, Program Control System Procedure, September 19Z9. ,

l The revision status information shown on the document for reference 3 is not in accordance with the method described in reference 1. Reference 4 does not contain a document number, yet it was referenced by such in E0P 25-4.00, Paragraph 4.1.1.

Mr. Stokes' Coment: Control of documents dictates that when they have inadequate reference information, the document control department assigns their own control numbers, namely part of QA. This ensures document traceability and would meet the requirements of 10 CFR 50, Appendix B."

Inspection Finding - Review of the General Electric practices indicated that the non-design disclosure documents, such as those listed above, are not subject to configuration control procedures because they do not contain technical information used in the design process. These documents are under individual document control as governed by the issuing organization and their preparation and numbering is addressed in NEDE-22000. Design disclosure docunents that do contain such infonnation are subject to conhguration control and fall within the configuration management system.

This item is closed.

Stokes Re3 ort Section 4.8 "While inserting package 68 into E0P Manual 264, Mr. MHam noticed that the old authorizing letter required traceable recoros, and the new authorizing letter did not. Mr. Milam was referred to Dave Lee about the change. He said no ulterior intent (Milom's words)

existed, they had just tried to shorten the letter. Lee promised to put a note in the file to consider the char.ge at the next revision.

Mr. Stokes' Coment: Copies of the two letters were reviewed, and the omitted statement amounteo to "and traceable records." This could have been added to the new revision, without increasing the letter's over-all length since the second paragraph last line only had the phrase "ment controls" written on it and the addi'tional three (3) words would fit on it without causing any increase in the letter's length. Mr. Lee's excuse does not appear to be factual. No QA program can succeed without trace-able records for without traceability the problem cannot be traced to the cause. This violates 10 CFR 50, Appendix B."

Inspection Finding - The administrative letter was a policy type of state-ment and did not reflect detailed requirements. GE's position is that the phrase "appropriate management controls" was meant to include a reference to traceability. No change in the implementation of the E0Ps was intended t,y the letter. This item is closed.

- - - - -_. _ - - _ . . - _ . . - _ _ . , _ _ . - . _ . _ . - . _ _ - ---..m_ . -

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT NO : 99900403/88-01 IllSPECTION RESULTS:

PAGE 46 of 62 L Stokes Report Section 4.9 "On June 29, 1981 responsible for the system.as responsible engineer on an ECN he had neve e,-nor was he Lee his job description, it wasn't HeavailableStrambach then asked to abo then called T. R. Dankmeyer and of the Atomic Energy Act of 1954.

asked a e either. He for more co i p es of Section 223(b) plus Employee Bulletin 35, datedHe and George Strambach.

August distributed copies of Section 223(b) 26, 1980 to Paul S h c erer Tom Casey Mr. Stokes' Coment:

Section 1.1 and they still have not been remedied "Similar problems ar

_ Inspection Finding - No written job definition was found however, GE feltlob assignment to do specific tasks responsibility.

was a management work material prior to signing the ECN would be appropriatA numbers or work details were provided. e This item is closed.e; however, no

_ Stokes Report Section 4.10 "On June 29, 1981, Stramback concerning references to Engineering OperatiMr.

. B. Milam wrote M his letter, Mr. Milam It seems that a letter on May was requesting more specific ng Procedures. refe In 20, 1981 P. R. Schere unspecified E0P's. to B. P. Grim /J. B. Moritz/G. u ority....

BonfromTransfer of A thrence Mr. Milam requested tha including the issue date of each reference.t specific references be use Mr. Stokes' Coment:

ment, there is a reason for it.When a reference is used in a letter or other docu -

subject which the reader should consult.It contains related information on the the referenced document. documents, it is impossible in many cWithout specific refere ases to locate the employees, then no 10 CFR 50, Appendix B.Section VI " ;

one should expect more of the emp e

This violated Inspection Findinos - The transfer of auth ment personnel who understood the message.ority letter was between manage-was transferred, should apply. it would appear that all E0Ps which weSince the design respon re appropriate from the information presented.A violation of 10 CFR 50, Appendix B could n e ermineo This item is closed. '

Stokes Report Section 4.11 "One of Mr again o~n July V 1981 when Ed Chin called Mr. Milam. Milam's old projects came u was following up on Mr. Milam's attempts to Itobtain .

seems a that Ed computerized tracking

- _ - _ _ - - - - ~ ~ '

l i

ORGANIZATION: GE NUCLEAR ENERGY -

SAN JOSE CALIFORNIA l 1

REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 46 of 62 Stokes Report Section 4.9 "On June 29, 1981, Mr. Milam was ask'ed to sign as responsible engineer on an ECN he had never seen before.-nor was he responsible for the system. He responded by complaining to Geonge ,

Strambach about having his responsibilities juggled. He then asked to see his job description, it wasn't available. He then asked to see the letter assigning oesign responsibility, it wasn't available either. He then called T. R. Dankmeyer and asked for more copies of Section 223(b) of the Atomic Energy Act of 1954. He distributed copies of Section 223(b) plus Employee Bulletin 35, dated August 26, 1980 to Paul Scherer, Tom Casey and George Strambach.

Mr. Stokes' Coment: Similar problems are recounted starting with Section 1.1 and they still have not been remedied."

Inspection Finding - No written job definition was ibund to be available; however, GE felt job assignment to do specific tasks was a management responsibility. A request by Mr. Milam for adequate time to review the work material prior to signing the ECN would be appropriate; however, no numbers or work details were provided. This item is closed.

Stokes Report Section 4.10 "On June 29, 1981, Mr. Milam wrote Mr. G. B.

Stramback concerning references to Engineering Operating Procedures. In his letter, Mr. Milam was requesting more specific references to E0P's.

It seems that a letter on May 20, 1981 on Transfer of Authority.... from P. R. Schere to B. P. Grim /J. B. Moritz/G. B. Strambach made reference to unspecified E0P's. Mr. Milam requested that specific references be used including the issue date of each reference.

Mr. Stokes' Coment: When a reference is used in a letter or other docu-ment, there is a reason for it. It contains related information on the subject which the reader should consult. Without specific references to the referenced documents, it is impossible in many cases to locate the document. If management doesn't set a good example for the rest of the employees, then no one should expect more of the employees. This violated 10 CFR 50, Appendix B, Section VI."

Inspection Findings - The transfer of authority letter was between manage-i ment personnel who understood the message. Since the design responsibility was transferred, it would appear that all E0Ps which were appropriate should apply. A violation of 10 CFR 50. Appendix B could not be determineo l from the information presented. This item is closed. -

1 Stokes Report Section 4.11 "One of Mr. Milam's old again on July 7,1981 when Ed Chin called Mr. Milam. projects came It seems that up Ed was following up on Mr. Milam's attempts to obtain a computerized tracking l

i I

.6 ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION i NO,: 99900403/88-01 RESULTS: 1 PAGE 47 of 62 system for CAR's. Mr. Chin requested copies of Mr. Milam's records of what was purged from the computer in February, per C. L. Cobler'i instructions.

See also Section 2.82." " '

Inspection Finding - The required in-process tracking and closeout of issues in controlled docununts is administered under established proce-dures. Therefore, the use of personal computer programs for the tracking of documents such as CARS on an interim basis is discretionary and not necessary. It was concluded that this concern invnived management judg-ment of whether a computerized tracking system for CARS was necessary or not. This item is closed.

Stokes Report Section 4.12 "Mr. Milam assumed responsibility for Alto Lazio coordination tasks, including clche-out of unneeded DRF's and issue of mortgaged ECN's on July 17, 1981. In the process, he was informed that many promises were made, and permission given to do things based on promises by Paul Scherer to issue ECN's that had not been issued at the time of Mr. Milam's assuming the task."

Inspection Finding - Mr. Milam noted his new work assignment in his work record. The coordination task responsibility would include the tasks described in the work record. The tenns "mortgaged ECNs" and "doing things based on promises" refer to the time comitment for document issue. No violation of existing procedures was identified in the work activities.

This item is closed.

Stokes Report Section 4.13 "On J"ly 17, 1981 Mr. Milam wrote a letter to G. A. Senn concerning the subject: Questions for Salaried Employees Meeting. According to the letter, Mr. Milam had six months before responding to an invitation to submit questions for the meeting by asking why GE allowed the QA section to report to the nianager of the production organization? Mr. Senn haa responded by saying that whenever he felt that it was necessary, the QA manager could go around his immediate manager for the resolution of a quality problem. Mr. Milam felt this did not address the question completely.

i Mr. Stokes' Coment: The QA manager and personnel cannot go around the imadiate manager at promotion / review time. This is why 10 CFR 50 requires a different organization. See also Section 3.6 and Mr. Milam's performance appraisal of January 1982." .

Inspection Finding - A review of the organizational arrangement indicated that Mr. 5enn was the cepartment general manager to whom reported many sections with different responsibilities, including quality assurance.

This is consistent with the requirements of 10 CFR 50, Appendix B (refer r to Stokes Report Section 3.6). This item is closed, t

l

• ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.- 99900403/88-01 RESULTS! par.r ao n< r <,

Stokes Report Section 4.14 "Mr. Milam wrote G. B. Strambach a letter on July 20, 1981 about the use of uncontrolled documents. The. letter was to report the existence and use of uncontrolled copies of NEDO-10466-A., Power Generation Control Complex Design Criteria & Safety Evaluation, February 1979. Mr. Milam had obtained a copy of the microfiche from the Document Distribution Service of the Technical Support Services organization.

Mr. Stokes' Coment: Since no record was made, Fr. Milam is not going to receive addenda or new revisions of the document, nor will the other users with uncontrolled copies.

Synopsis: Mr. Milam says that since revision status information for this type of document is not readily available, that he cannot assume his docu-ment is current, nor can he use it as a formal reference."

Mr. Stokes' Comment: What guarantee is there that the other employee with uncontrollea copies will decide not to use their documents? See also Section 3.10."

Inspection Finding - The document referenced in this coninent is not a design control document. The user is to verify by contact with the library that this is the latest version of the document. No deviation of design control procedures was identified by this comment. This item is closed.

Stokes Report Section 4.15 "On July 31, 1981, Mr. Milam wrote Sid Smith concerning Airendment 16 to the WPPSS Nuclear Project No. 2 Final Safety Analysis Report. The letter was to inform Mr. Smith of two errors in the instructions for inserting Amendment 16.

Mr. Stokes' Coment: Were these corrected? And even more important, if simple instructions can be carelessly written and issued then how well wds the change reviewed? See Section 4.2."

Inspection finding - The question of errors on materials generated by othcrs (utilities in this case) is not controllable by GE procedures.

Here the error was identified to the utility when noted, but this was outside the GE organization and therefore not controllable. This item is closed.

Stokes Report Section 4.16 "GE's subcontractor's work becomes a concern on August 21, 1981. Mr. Milam wrote a letter to George Strambach about Limerick FSAR Annotation, Section 7.7.1.3, Recirculation Flow Control System Instrunentation and Controls. Mr. Milam had been reviewing this l

l f

i

~

l

.s

. ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA j

REPORT INSPECTION NO. 99900403/88-01 RESULTS: 1 PAGE 49 of 62 l

annotation and had come to the conclusion that there was an erro'r rate in the subcontractor annotation of approximately 47 percent. He requested that GE's engineers be alerted to the possibility of open itemselurkeg within the subcontractor's references.

Mr. Stokes' Comment: GE's subcontractors cannot be expected to be any better than GE themselves."

Inspection Finding - A review indicated that Mr. Milam was performing his assigned job function. The subcontractors were to identify disc epancies in the FSAR where they noted them and have the GE reviewing engineer confirm, resolve and correct the errors. Mr. Milam was correctly performing his job. This item is closed.

Stokes Report Section 4.17 "A second letter was written to George Strambach on the 21st of August, but it was on a different subject. This letter concerned Engineering Change Requests. The Nuclear Control &

Instrumentation Department Practices & Procedures Manual, Procedure 4.42, Design Document Change Control, May 18, 1981 was referenced. The main point made was that according to the reference document only manufacturing was authorized to initiate change requests. Apparently, C&1 engineers lacked authority to initiate change requests even though they had been doing so.

Mr. Stokes' Connent: This is a violation of 10 CFR 50, Appendix B, Section III."

Inspection Finding - An Engineering Change Request (ECR) is any unistued ECh, as defined in the Engineering Operating Procedures. The ECR is to l be used by personnel outside of engineering. Engineering personnel are l to directly fill out the ECN and implement the required change. This

(

item is closed, l Stokes Re sort Section 4.18 "On September 1,1981, Mr. Milam wrote Mr. Stram)ach a letter on Document Status. He was concerned that many of GE's documents were issued with deferred verification, and that these

, documents might be used for design verification of other documents

( incorrectly. He noted that the ICER code near the revision block of the i

document does not indicate verification status. See also Sections 2.22 j and 3.9. .

Reference:

1. Engineering Documentation Practices Manual Section C1, Completion

! Status Codes, August 24, 1979.

l

(

.s

, ,0RGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: '

PAGE 50 of 62

~

2. NEDE-24563, Engineering Information System User's Guide, December 1978." -

Inspection Finaing - General Electric's system for tracking the verifica-tion status of a document is to enter a code in the Engineering Information System (EIS) rather than on the drawing, which accomplishes the objective.

This item is closed.

Stokes Report Section 4.19 "The following problem was detailed on a PrWuctivity Improvement Proposal form of Mr. Milam. It seems that Design Verification Status Change Notices (DVSCN) were issued with no number on them makin These documents were required to confirm closure o'g deferred them non-traceable.

verification. See Section 4.8."

Inspection Finding - The initial DVSCN Form SD-028 did not have a pre-printed control number or a specific area for manual entry of a control number. During the logging and recording process performed by Plant Definition and Release Control, a control number identical to the log number was identified on each DVSCN; for example, DVSCN 00001, 00002, and 00003. This identification preceded issue, distribution and microfilming.

The current DVSCN, Fom SD-028A, has a pre-printed DV number as of January 1984. All DiSCNs, past and present, are traceable and available via microfilm. This item is closed.

Stokes Report Section 4.20 "During FW8138 or the third week of September 1981, Mr. Milam wrote in his work record that the Nine Mile Point 2 FCD/IED (Functional Control Drawing / Instrument and Electrical Diagram) update was being done backwards. The elementary diagram is Leing used as the source document to update the FCD. Normally, the FCD is considered to control the elementary diagram. Instri!ctions, guidelines, etc., have been provided in references 1 and 2, which are inadequate authorization to use irregular procedures, in that they were written by an engineering assistant. Such instructions should come from a subsection manager.

Reference:

1. Letter, C. M. Parks to W. Gibford, B. P. Grim, G. B. Strambach.

Transfer of Authority.... August 24, 1981.

2. Letter, C. M. Parks to Distribution, hine Mile Point 2 FC.0/IED Effort.

August, 1981.

Coment: Is this an example of the "work-around" practices referred to in Section 1.187" I

.s ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA

~

REPORT IhSPECTION NO.: 99900403/88-01 RESULTS: " pAGE 51 of 62 Inspection Finding - Originally, the IED/FCD drawings were source documents for the Elementary Diagrams (EDs). Once developed, the EDs'can stand alone as a complete depictiun of the actual functional implementation'of e'ach system. Subsequently, the EDs were updated to reflect the latest design changes to respond to customer need for regulatory and standardization design improvements in advance of comparable changes to the FCDs and IEDs.

Sufficient detail was available ir. the design inputs to perfom the design update directly in the EDs, and priorities required the updating of EDs well before updating the FCDs and IEDs. Consequently, the EDs were used as input for updating the FCDs and IEDs. It should be noted that the FCDs and IEDs were the responsibility of the same person as the EDs in the Control and Instrumentation organization and the need for changes to be made on all three documents was well understood. Further, since the IEDs, FCDs and EDs were all supplied to NRC for review along with the FSAR, the updating of the IEDs and FCDs became tied to FSAR submittals rather than manufacturing schedules. The authorization to proceed in this manner was provided by the referenced letters, which had the full approval of management. This item is closed, Stckes Report Section 4.22 "It was also noted that Mr. Milam requested correction of a generic terminology inconsistency. Mr. Milam stated that every project he had investig, 'd shows this error. He wanted to write a CAR to B. P. Grim or W. Gibir.4a. George Strambach said to write a Quick Letter to LSE. See Quick Letter from Sam A. Milani, III to K. M. Tao and O. J. Foster, Water Level Range Terminology, October 7, 1981, filed FW8141."

_ Inspection Finding - Mr. Milam's concern here was that a feedwater level signal had two names. C.o .isme was "upset" and the other "wide range" for the 0 - 180 in, feeds ter level indication. The signal is not safety-related, and the two terms appear only in GE documents. Terminology is "wide range" when used in procedures forwarded to plants. No action is required by users of the equipment. This item is closed.

Stokes Report Section 4.23 "Mr. Mi; lam made the following observation j j ctring FW8141.

f There appears to be a great reluctance among C&I management to allow engineer: to use CAP,'s. CAR's to recipients external to CSI might be reasonably deprecated, but I can see no justification for a reluctance to use them internally. Thus far I have not experienced an overt refusal of permission (which would be in conflict with reference 6). Rather, I have been ' instructed to try something different.

Reference 6 referred to the NC&ID Practices & Procedures No. 5.45, Corrective Action. See Section 4.22."

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA m

REDORT INSPECTION  !

M. : 99900403/88-01 RESULTS: PAGE 5? of 6?, l l

Ir.spection Finding - ECNs are primarily used by engineering to fncorporate design changes encounteted during the design engineering phase whereas CARS are primarily useo by personnel outsdde of the engineering mork, unit.

The GE position is that the use of ECNs by engineering personnel provided a prompt direct action for the item to be corrected. The use of CARS was ,

not prohibited, ju;t discouraged as iaefficient. This item is closed. i Stokes Report Sectivg 4.?4 "During FW8142, the third week of October, there is a proolem with the lack on authority to write change requests.

Mr. rlilam wrote ei. ECA (Engineering Change Authorization) for Hanford.

This was rione so that engineering would be given authority to make engineering c inge3. However, Ray Moore did not want to issue the ECA.

He felt tisat ..e Project Work Authorization (Pk'A) had sufficient auttsrity to get the change requests done.

Mr. Stokes' Coment: The problem boils down to: 1) Do the PW', provide suffb:ient aushority for change request work? If sn, what is the purpose j for the ECA e M is ai. ECA needed? It seems GE has to many y >cedures dupliceting functions of other procedures that the procedures themselves beconc probl .ms. See Section 4.21."

Inspection Finding - The preparation of an ECR for the subject work was considered unnecessary since a PW4, was open and provided authority to continue with outstanding tvork. No effect upon the quality of the design work was identified from the inspection of the work items. This item is only concermed with the p:per required for the work 110thorization.

This item is closed.

Stokes Report Section 5,1 'Mr. Milar wrote Mr. A. Koslow on Novembar 2, 1981 about. verification of Design Changes. According to the letter, e

changes inade by Component 913 engineers were being verified by engineers fiom Components 901 eno 909 who were responsible for the system being changed.

Mr. Stutes' Coment: This was 'n conflict with E0P 42-6.00. Inoependent Design Verification. April 30,1981,'which states that the verifier must be a person wSn is not directly responsible and accountable ?or the design. This also violates 10 CFR 50, Appendix B, Section III Design Cortrol. Quotir.; f M the third paragraph, the verifying or checking process shall be pe- Fh/ individuals or groups other than those who performed the orig' ' -

gn, bu; ney te from the same organfzation."

Inspection Findingn evit:ed * . rion stating that engineers in components 901 6no e m a.;k whicn they had designed. A saniple .,f the subju a u +'ined thal incl +d PWA 2123LD,

- + . , - . -

1 I

.o ORGANIZATION: GE NUCLEAR ENEPGY SAN JOSE, CALIFORNIA I

REPORT INSPECTION

_ NO.: 99900403/88-01 RESULTS: i PAGE 53 of 62_

Revision 5, ECNs NJ27996, NJR6936, NJ27932, NJ18314 NJ31445,' N816466, hJ21436, drawings 828 E446 AA, Revision 1 through 20, 866E30, Revisions 1 through 4, and drawing 865 E 352, Revisions 1 and 2. No comonenntr'ies for design were noted during ti.e inspection. This item is closed.

Stokes Report Section 5.2 "On November 4,1981 Rick Smetka brought Mr. Milam a Laguna Verde review package that war overlooked. Another document handling problem."

Inspection Finding - A review of Mr. M11am's work record indicated that the package contained sections that Pr. Milam was to annotate. This infonnation did not indicate any concern over document handling. This item is closed, 7

y Stokes Report Section 5.3 "A letter from D. H. Currie, Manager Quality

.< Systems Audits & Records to C. Smith, J. Cocper, and A. Sagen on

' November 12, 1981, pointed out a probles with the procedures for using the ECN interndi sheet. "On ECN internal sheet, there are spaces indicating (for example) FDI required MR, El affected, etc. E0P forms index (Section 85,55-2.00) states that these spaces must be signed by the person responsible for the activity (FDI, MR, EI, etc.) when it is e different ptrson than the engineer who is responsible for the ECN. P.outine practice was and always has been, in C&ID, to have the engineer issuing the ECN also sign the aforementioned activities. This was in conflict with the E0P U1d involves tbv risk that pecple who need to do things (w;'ite FDI's, MR's, EI's, etc.) won't know it, and won't do it, causing many problems downstream."

Mr. Stokes' Comer.t This is just another example where even though a procedure exists etiner the training session was inadequate o. the

^^ written instructions were not clear or there were neither training or

'4ritten irstructions. All violate 10 CFR 50, Appendix B. Section 11 requirements."

Inspection Finding - A review of GE E0P 55-2.00, Engineering Change Notice, I indicated that signatures and initials for certain activity blocks are required when the activity is perforsned by other than the Responsible Engineer (EOP 55-2.00, Supplement D, page 6 of 6. Revision 0, issued June 20, 1982 Note 1). This step assures that the sigr : will receivt a copy of the Ei., and thereby officially infonn him that the signed activity i must be performed. The alternate approach is for the Respu.i:ible Engineer to sign the activity block and subsequently assure that the individual required to perform the activity is informed by placing that individual's name on the distribution for the ECH [EOP 53-2.00, page 4 of 6, issued l

l f

, ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS:  ;

PAGE 54 nf r3 Augt.n 24,1979, Item U (circled)). According to General Electr'ic, the letter approach was predominantiv used in the Control and Instrumentation oraahization. It was concluded that Mr. Milam's concern was vaMd; however, alternate means were available anc were used to accomplish the same objective. It was suggested to GE that a procedure change in Note 1 of E0P 55-2.00 Supplement D would eliminate any ambiguity. This item is closed.

Stokes Report Section 5.4 "On November 20, 1981 Mr. A. Koslow received a letter from J. G. Puls, Omtec, and attached were the responses to MP&L's concerns regarding the differences between the Grand Gulf PGCC design and NEDO-10466A per PWA-1285JB, Revision 1AC. Also, enclosed were copies of the reference drawings and GE's copy of the NED0 document. Mr. Milem had the following coments about NEDO-10466A and the sumary of questions /

comnents te;.ich were returned with the NEDO document returned by Omtec.

1. Mr. Milam says the copy of NE00-10466A is an old revisior.. At least 2 errata and addenda sheets have been issued. Were these sheets used in this review? They were not included. Known changes affect:

a. Connectors

b. QA Procedures

c. Factory Process Controls

d. Crimping Process Controls Mr. Milam had not examined the changes in detail. See his letters about uncontrolled copies of NE00-10466A in use upon which no action was taker.

See also Section 2.14.

2. Mr. Milam says that the 'whole series of questfons should have been answerud by reference to the notice on sheet 11 of NED0-14066A which states that neither GE ror any other party "makes any wa.ranty...with respect to the accuracy...of the information contained in this docer ant...." See also Section 3.10.

?. MP&L's sumary number 5 states that Grand Gulf Panels did not receive full scale factory testing. Omtec says that the Grand Gulf Document Control Sheets in GE Q/C Records .ndicate full scale factory testing.

Mr. Milam says full scale testing is not necessarily complete testing.

Response indicates anonymous records in the Q/C record center.

Are all T1's signed off? Response should include a list of them.

How about ship short items, which are not tested, but either simulated or bypast.ed dering test? See also Section 2.13."

ORGANIZATION: GE hUCLEAR ENERGY SAN JOSE, CALIFORNIA

.c REPORT JhSPECTION NO : 99900403/88-01 RESULTS: 4 PAGE 55 of 62 Inspection Finding - A review of applicable records r2sulted in 'the To11owing: -

1. Generel Electric indicated that Mr. Milam's coments on the Omtec response to Missis >ippi Power & Light (MP&L) along with the addenda and errata to the NED0 report were submitted to Omtee who updated the responses as required. A record of the final response to MP&L could not, however, be found by General Electric.

2. In his letter to G. Strambach dated July 20, 1981 Mr. Milam indicated that issuance of NEDO-10466A should be noted as "uncontrolled" or a controlled distribution should be established to assure that copies include current information. General Elect ric chose not to implement this suggestion.

3. Mr. Milam's suggestion to respond to the series of questions by stating that neither General Electric nor any other party "makes any warranty..." was not adopted by General Electric.

4. Factory testing is a specific controlled set of tests perfonned prior to shipment. Quality assurance records describing performed tests were prepared during factory testing and are maintained in the Quality Assurance Record File. "Complete testing" involves several levels of test, including factory and onsite testing. Test instructions are prepared in accordance with Quality Assurance Procedure 6.7 and test control is covered under procedure QAP 2.11 The handling and docu-menting of "ship-short" items are addressed under Manufacturing Procedure 3.03. Engineering Operating Procedure 55-3.20. Field Disposition Instruction, which includes the controls for completing "ship short" items, includes acceptance testir.g.

It was concluded that Mr. Milam's concerns were generally valiJ and that General Electric (1) assured proper responses to the MP&L concerns by transmitting the necessary documents to Omtec (2) chose not to implement Mr. Milam's suggestions in Items 2 and 3 above, and (3) conducted the appropriate factory and onsite testing, considering ship-short items, in accordance with governing procedures. No evidence cwid be found regarding anonynous records in the QA Rec.ords Center. Tbla item is closed.

Stokes Report Section 5.5 "In Mr. Milam's work re:ord is a letter dated November 25, 1981 concerning the weekly report, Sys'.eas Engine'ering, FW48.

Item rupber 7 was of interest. It seems that Barksdale pressure switches installed in the LaSalle RCIC system have a much lower limit of operation than that of the plant. These switches should be replaced before the plant goes critical.

I

.o ORGANIZATION: GE NUCLEAR ENERGY' SAN JOSE, CALIFORNIA REPORT INSPECTION NO. : 99900403/88-01 RESULTS: PAGE 56 of 62 Mr. Stokes' Consnent: It is good that this was caught before the' plant went critical, but ware they replaced? There seems to be too much dependence on operational checkouts on systems and equipnent in the field. ,If the people performing the systems checkouts have as little regard for quality as the desigr.ars and verifiers in the office, then there is little assurance of the saf: operation of the plant."

Inspect. ion Finding - The original Barksdale switches were shipped conditionally based on a lack of available qualification data. When the switch could not be qualified, new switches were provided and qualified.

FDDRs were examined which identified the necessary corrective action and indicated that GE procedures were operating effectively. This item is closed.

Stokes Report Section 5.6 "On December 12, 1981 Mr. Milam entered on a HEMS PD Planning 5heet that the verification of ECH NJ-27967 was deferred due to deferred verifications on ECN's NE-97217. NE-97298, NE-97292 NE-97301, written against 761E791TY and 807E160TY, source documents.

Mr. Stokes' Coment: Did Milam catch all the deferred documents? Will one document be released by mistake with a deferred verification? The bottom line of GE's Deferred Verification Program is that none of the plants built by GE are in compliance."

Inspection Finding - Mr. Milam was follo.4ing procedures when he entered the ECN with deferred verification into the GE NEMS work tracking system.

This system required that the Respansible Design Engineer (RDE) would, by 1he date indicated, review the drawing and remove the deferred verifica-tion if the original reason for deferral had been removed. GE F0P 42-6.00 also required the processing of a DVSCN. The DVSCN included the DRF reference to where the traceable verification was filed. The RDE for the affected drawings was required by procedure to assure that all deferred related drawings were reviewed for closure.

For a complete explanatian of the deferred verification system, refe i NRC Inspection Report Nos. 99900403/86-01, Section E.1; 99900403/87-01, Section D.7; 99900403/87-03, Section D.2; and 99900403/87-06, Section E.2.

This item is closed.

Stokes Report Section 5.7 'The following documents were also noted as being unverified: Reactor Protection System elementary diagralu 807E16ATY, Revision 5 including the following ECNs: NJ-20107 NJ-15627, NJ-26456 NJ-26855 NJ-26897, NJ-27600, and NJ-28030. Reactor Recirculation system elementary ciagram 761E/91TY, Revision 4 including the following ECNs:

NJ-25828 and NJ-26955.

A _ _ _ _

.s

. ORGANIZATION: GE i.UCLEAR ENERGY SAN JOSE, CALIFORNIA I REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 57 of 6?

Mr. Stokes' Comment: It is obvious the Deferred V:rification Pr'ogram is out of control. See Section 5.6." -

e

• Inspection Finding - A review of the referenced drawings and ECNs indicated that verification had been performed. The verificattun status of the above identified drawings was stated on Design Verification Status Change Notice sheets as complete on DV00306. DV01862. DV01863, and DV00072. Design Verification Summary sheets also indicated the verification status of the drawings for m ch revision. The need for deferred verification arose because of the use of the Space Control Systems Department (SCSD) of General Electric in Florida as a supplementary production f acility. SCSD maintained the elementary diagrams in order to keep the hardware drawings current. Because the reference documents for these elementary diagrams were maintained in San Jose, all revisions made by SCSD were carried as deferred verifications. When the panels were shipped to San Jose for final assembly, a design review was perfonned by the responsible electrical systems engineers and the deferred verifications were cloled out. The ECNs were prepared in San Jose and the verifications were completed for each on a nondeferred basis. In each case, the changes could be independently verified. All deferred verifications were cleared. It was concluded that Mr. Milam's concerns of unverified elementary diagrams were valid; however, that of unverified ECNs was incorrect. All documents were verified in accordance with GE procedure E0P 42-6.00. Independent Design Verification. This item is closed.

Stokes Report Section 5.8 "A letter written on December 22, 1981 to D. H. Currie from A. Koslow requested a QAP Manual be assigned to Component 913, C&I Technical Licensing. The letter stated that no QAP Manual was available at that location.

Mr. Stokes' Comnent: Do other areas exist that do not have Quality Assurance Maiuals available for the employees to use?"

Inspection Finding - A review of the referenced December 22. 1981 letter inoicated that one of Component 913'.s primary functions included the performance of quality reviews of FSARs and responding to NRC and customer questions. Quality Assurance Procedure Manual No. 70 was assigned to Mr. Milam on January 6, 1982 frcm Connie Howard and receipt acknowledged by Mr. Milam on January 8.1982. This item is closed.

Stokes Report Section 5.9 "During the last week of December'1981 Mr. Milam inserted Amencment 18 into the Hanford test. Amendment 17 o missing. It was requested by Mr. Milam. See also Section 4.2."

, ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO. 99900403/88-01 RESULTS: PAGE EA of 62 Inspection Finding - The action identified is within the GE proc'edures and was a normal work function. The licensee was contactea to provide the missing information. This item is closed. e ,

Stokes Report Section 5.10 "On January 5, 1982, Mr. Milam received his performance appraisal. It is included here because Mr. M11am's former supervisor, George Strambach, thought Mr. Milam's quality of work was

?o high. (Emphasis added by Mr. Stokes.)

Mr. Stokes' Coment: It would seem that no one's work quality could be too high, unless their production was zero. This is especially true of personnel trying to correct problems within the system. This attituoe says a lot about the quality of the group's work. Canp11ance with 10 CFR 50, Appendix B with the built-in checks and corrective action promotes the highest quality."

Inspection Finding - A review of Mr. Milam's 1981 Performance Appraisal dated January 5,1982 indicated overall satisfactory perfonnance. Factors limiting performance during the appraisal period were noted as "accuracy to an extreme" and "outside unit preoccupation." It was further stated that Mr. Milam should place his emphasis on the standard of reasonable confidence in the work product instead of absolute certainty, leave alone past work problems, and keep attention on present unit's goals and responsibilities. There was no statement that Mr. Milam's quality of work was too high. This item is closed.

Stokes Report Section 5.11 "A package was included in Mr. Milam's work record which included a PWA-1289LH for the Clinton project dated January 22,

( 1982. This documer.t was not approved for use. The cover letter stated that the responses would close the authorized PWA comitment to the Safety and Licensing Operation. One comitment was the environmental qualifica-tion of Clinton equipment. It said "environmental qualification is assumed to include consideration of the temperature, pressure and relative humidity conditions that exist external to the device or panel.

Mr. Stokes' Coment: What about radiation, or isn't some of this equipment exposed to radiation? The internal conditions may in some cases exceed the external. While employed as the stress analyst for .;he injector of the Advanceo Test Accelerator at the Livennore Laboratory, the author worked with internal conditions such as extremely large voltag'e and magnetic fields,' which far exceeded the external ones. Why were these '

not considered by GE if they make up the environment of the equipment during operation? Radiation modifies the molecular structure of all material. See also Section 2.13 (items that don't receive functional testing).

, ORGANIZATION: GE NUCLEAR ENERGY SAN , JOSE, CALIFORNIA REPORT INSPECTION '

NO.: 99900403/88-01 RESULTS: "

PAGF 59 of D l

a. On a separate page under the classification of Active Essential Devices B13-D124 Position Indicator Probe, it is state 5 that I have not discovered evidence of environmental qualification. " '

b. Another page for Active Essential Devices C51-N002 Start-up Range Detectors, states that no environmental qualification data exists for these devices,

c. No environmental qualification has yet been scheduled for detectors installed in dry tubes within the reactor vessel,

d. A separate sheet for Active Essential Devices C51-N011 to C51-N014 states that no environmental qualification data exists for these devices. All passive essential devices received material and seismic qualification but did not receive environmental qualification."

hr. Stokes' Coment: It is unclear how engineers can ignore radiation, electrical, magnetic field, effects of salt water and certain gases on metal when these alter or destroy the materials molecular structure in relation to exposure and may cause premature failure."

Inspection Finding - A review of the following documents indicated that radiation ef fects were indeed considered when establishing the environ-rental qualification requirements placed on particular devices used in GE BWR plants:

22A6926 Revision 0, BWR Equipment Environmental Interface Data 384HA178, Revision 0. Evaluation of Reed Switch for Use on Position Indicator Probe, MCSR 439 994-75-045, Performance' of Bottom Entry Detectors at Oyster Creek 994-75-044, NA 200 Power Range Detector Qualification Sunnary An internal memorandum (R. Strong to S. Tang, Licensing Position for the BWR/6 CRD/RC&IS Position Probe, September 29,1981) clearly indicates that a special "S" classification was assigned to the position indicator probe to assure that the probe is seismically qualified for the OBE condition and that qualification requirements end test results.are control documented. The memoranoum hiso discusses the exemption from the require-ments of IEEE 323-1974 and 344-1975 based on the analysis of no credible failure to disable execution of control tunctions. In addition to the environmental interface data (22A6926), the following documents include qualification testing data for the position indication probe:

.o l

l

, ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 60 of 62 1

Seismic Test Report Wyle f58470 (GE DRF B13-59, Section 25)'

November 19, 1979 -

e

• Final Test Report, "CRD Post-Seismic Qualification Functional Test,"

GE NEDE 24765, December 1979 "Test Results for Position Indicator Probe Test," GE NEDE 21959, September 1978 (shock testing to 1200-1400g)

Devices C51-N002 are Intennediate Ran called them Start-Up Range Detectors)ge whichDetectors (Stokes are installed incorrectly in a dry tube within the reactor vessel curing start up. Following start up, they are retracted, and, although safety-related, they serve no function during and after an accident (normal operating environment is high radiation and high temperature). The environmental requirements were considered in DRF

1. C51-00040, Clinton Potver Station Environmental Qualification Report Book

1. 2, for both the detectors and the connector.

Devices C51-N011 to h014 are Power Range Detectors used during full power operation. They are also installed in dry tubes within the reactor vessel dnd are not required to function during or following an accident (similar environment as the Intermediate Range Monitors). The environmental requirements were considered in DRF #C51-00040 Clinton PS EQ Report Book

1. 1. Test cocument 994-75-044 sunnarizes qualification testing of three improvements made to the Power Range Detectors.

It was concluded that Mr. Milam's concerns were not valid, since all devices referred to did receive environmental qualification consideration and testing appropria'.e to intended operating conditions. This item is closeo.

Stokes Report Section 5.12 "On January 27, 1982, Mr. Milam wrote a letter to T. R. Snerpey about proposed changes to EIS. In his letter, he offered several suggestions for changes. One was to clear up the following problem. Mr. Milam says that document application changes to project Master Parts Lists are made by Engineering Review Memoranda (ERM). The actual incorporation of the change into the MPL may occur several oonth; later. Thus, an ERM can be, in effect, an unincorporated change ar.d when the MPL is checked using the DI program, the change is not visible. No revision increment is made and no unincorporated ERM number is shown.

~

Thus, inappropriate drawings may inaovertently be used."

Mr. Stokes' Conment: For a system to be relied upon the way GE uses EIS, the data updates and entries should be required to b- delayed no more than the shortest time which a document might reasonably spend being

ORGANIZATION: GE NUCLEAR ENERGY SAN JOSE, CALIFORNIA REPORT INSPECTION h0.: 99900403/88-01 RESULTS: ,

PAGE 61 of 62 issued. This is ne:essary because once the document is issued 'it may reasonably be used by anyone. At this time the document should be available and retrievable from EIS, otherwise, inecrrect reference ,

documents may be used. See also Section 2.54 (example drawing left out of EIS) ana Section 2.72 (ECN not logged on EIS).

Inspection Findin timing necessary,3 - Mr. Milamchanges.

to incorporate was incorrect The in MPLhisisassessment of the updated in EIS within two days of ECN or ERM issue. The upcated version of the MPL is available through EIS by the DA program (not the DI program). Contrary to Mr. Stokes' assertion, the synopsis / work record does not describe a systematic or timing problem with EIS.

The configuration of documentation in EIS may differ from the issued and distributed documentation (hard copies) mainly because of the time required to issue and distribute hard copies. This includes documents applied to a Master Parts List via an ERM. The primary purpose of the EIS is to provide an electronic data base which reflects the latest engineering definition of the product. The entry of a document in EIS is made within hours of issuance of the ERM. Using the Document Applicatien program. EIS will show the latest application of issued cocuments. Users of engineering cocuments must query the EIS data base for the latest document status (revisions, outstanding changes and application). This item is closed.

Stokes Report Section 5.15 "Mr. Milam noted in his work record for the first week of March 1982 that he was involved in a fight over procedures.

He insisted that a DVSCN was required in addition to the ERM. He also insisted that a DRF reference was not required on the ERM. R. L. Reghitto was blocking the issue of the DVSCN and was requiring the DRF reference.

Mr. Stokes' Connent: From a review of E0P 42-6.00, Paragraph 4.1.2, it seems that per Paragraph 4.1.2(b) a DVSCN is necessary to notify engineering services and from Paragraph 4.1.2(a) etther an ERM, or ECN, or applying document, or DVSCN may be used to show the DRF containing verification, if application, Paragraph 4.1.2(b). Has Mr. Reghitto attended a training class on the issuance of DVSCN's? See Section 1.5."

Inspection Finding - During this perioc, a degree of variance was evident regarding individual interpretation of the design verification procedural requirements. Corrective action was taken in the form of user'information/

clarification and training. Quality Assurance Newsletters (numbers 61, 63, 64, 65 and 66) were circulated during the March 1983 to April 1984 time

l 4

, ORGANIZATION: GE NUCLEAR ENERGY' SAh JOSE, CALIFORNIA

=

REPORT INSPECTION NO.: 99900403/88-01 RESULTS: PAGE 62 of 62 l

frame addressing specific design verification topics. GE E0P 42-6.00 Change Notice B was issued December 5, 1983 to clarify the DVSCN require- I ments. Also, in March 1984 the QA Training Course on Design Verification was revised to reflect changes in the design verification process.

In the case of the DRF reference, Mr. Milam failed to make the distinction between what is required by procedure and what is permitted by procedure.

DFR references were not required by procedure; however, Mr. Reghitto desired such a reference notation in order to enhance traceability and was not in violation of any applicable procedure in insisting on its inclusion.

Paragraph 4.2.2a.4 permits the application of DRF reference on an ERM.

This item is closed.

Stokes Re sort Section 6.1 "On April 7,1982, a letter from W. H. hendrix

~

to D. W. Reigel was written on the subject of NCalD Fusing Policy.

Mr. Milam circled a line and included a coment of his own.

Mr. Stokes' Coment: The line circled states that no credit is taken for fuses protecting specific pieces of equipment, nor is it required by any NC&ID documents. Mr. Milam stated that in our control system failure analyses, we take credit for fuses preventing failure propagation. Does GE still take credit for 'uses as Mr. Milam said? It this acceptable or not?"

Inspection Finding - The GE NC&ID fusing desigr. policy as stated in the letter from Hendrix to Reigel is consistent with generally accepted electrical system design practice of 10cating and sizing fuses to protect wiring and power supplies from damage or loss due to faults in individual circuits or loads served by that wiring or power supply. Fuses with these functions only would be located to best perform that function and sized at current ratings that may be too large necessarily to protect individual loads. Accordingly, it would be correct not to take credit for specific equipment protection by these fuses as stated in the policy.

The letter went on to say that all safety systems are designed with redundancy to prevent single failures from compromising safety fue.ctions.

Mr. Milam's coment referred to special non-safety control system studies (in which he was involved) to analyze system response and interactions as a result of hypothetical power losses to selected control circuits. The analyses assumed no component failures or circuit electrical faults other than postulated open fuses where this was a logical initiating event.

The effected circuit would then be electrically isolated from its pov.tr rupply while other circuits remained unaffected. In this sense, fuses might be said to prevent failure propage. tion. However, these analyses were not related to fusing design policy and their methoJology was not contradictory to it despite what Mr. Milam's coment may have appeared to be implying. This issue is therefore closed, i

.s APPENDIX C ALLEGATION SUMARY I. GAP / Stokes' Concerns Stokes Closed Report NRC in NRC

• Item Ca tegory Allegation Title Report Nonconforinance 1.1 EA Design Verification 87-06 1.2 EA Drafting Package 87-06 1.3 TR Document Label 87-06 1.4 EA Drawing Errors 87-06 1.Sa TR Training 87-06 1.5b DV Design Verification 87-06 1.6 NR Forged Signature 88-01 1.7 NR Sign Without Review 88-01 1.8 EA Drawing Errors 87-06 1.9 EA Drawing Errors 87-01 See note 1 1.10 TR Circuit Labels 87-06 1.11 EA Design Verification 87-06 1.12 EA ECN Procedures 88-01 1.13 TR Training 87-01 See note 1 1.14 EA Altered ECN 87-00 1.15 EA Drafting Practice 87-06 1.16 EA Shipment Nonconformance 87-06 1.17 NR EMPIS Manual 88-01 -

1.18 EA Engineering Procedures 87-06 1.19 EA Product Improvement 87-06 1.20 MS Mode Switch 88-01 1.21 TR Computer Program 87-06 1.22 EA Utility Cutouts 87-06 1.23 EA Corrective Action 87-06 1.24 EA Bootleg Work 87-06 1.25 EA Ralay Interference 87-06 1.26 MS Mode Switch 87-06 2.1 MS Mode Switch 88-01 88-01-01 2.2 EA Specification Update 87-06 2.3 EA ECN Verification 87-06 2.4 MS Mode Switch 87-06 2.5 MS Mode Switch 87-06 2.6 DV Drawing Discrepancy ' 87-06 2.7 TR Training 88-01 S.C EA Incomplete ECN 87-06 See note 1 2.9 DV Deferred Verification 87-06 See note 1 2.10 MS Not Shipped Per Print 87-06 2.11 MS Mode Switch 87-06 -

2.12 EA Panel Testing 88-01 2.13 MS Mode Switch 87-06 I 2.14 EA ECN Changes 87-06 2.15 EA Separation Signoff 07-06 2.16 NR ECN Changes 88-01 2.17 DV Panel Changes 87-06 2.18 DV Project Changes 87-01 See note 1 l

Al

o

.4 4

Stokes Closed Report NRC in NRC

, Item Category Allegation Title Report Nonconformance 2.19 NR EIS Panel Testing 88-01 2.20 EA Marker Plate Changes 87-06 2.21 EA Enclosure Top Cover 87-06 2.22 EA ICER/ SUN Codes 88-01 2.23 MS Mode Switch 87-06 -

2.24 EA Mismarked Drawing 87-06 , ,

2.25 NR CAR Drawing Update 88-01 See note 1 2.26 MS Mode Switch 87-06 2.27 EA Drawing Errors 87-06 2.28 EA Altered EChs 87-06 2.29 EA EMT/Glyptal 87-03 87-01-01 2.30 NR Drawing Errors 88-01 2.31 DV Switch Hardware 87-00 2.32 DV CR2940 Switches 88-01 2.33 NR Deferred Verification 87-06 See note 1 2.34 EA Terminal Board 88-01 2.35 MS Mode Switch 87-03 87-03-01 2.36 MS Mode Switch 87-03 2.37 MS Mode Switch 87-03 2.38 MS Mode Switch 87 03 2.39 EA Altered ECN 87-06 2.40 EA Open CAR 86-01 2.41 TR ECN Rejections 87-01 See note 1 2.42 EA ECR Log 88-01 2.43 EA Jroduction Engineering 88-01 2.44 NR Urawing Checks 88-01 '

2.45 EA Mistagged Enclosures 88-01 2.46 NR Panel Tag Errors 88-01 2.47 TR Work Assignments 87-06 2.48 EA Drafting Practice 87-06 2.49 EA ECR 88-01 2.50 EA Bootleg Modification 88-01 2.51 NR ICER/ SUN Cooe Use 88-01 2.52 EA MSP Revision Control 87 06  ;

2.53 EA Outdated MDL 83-01 l 2.54 EA EIS Drawing Application 88-01 2.55 EA ECN Rejection 88-01 2.56 TR Cutout Correction 88-01 2.57 NR Purchase Part Driwings 'B8-01 2.58 NR Design Changes 88-01 i

2.59 NR CAR Misinforvation 88-01 2.60 TR ECN Procedure 88-01 2.61 EA Floor ECN Coordinator 88-01 2.62 EA Connection Diagrams 88-01 -

A2

1

,o Stokes Closed Report NRC in NRC

. Item Category Allegation Title Report Nonconformance 2.63 MS Mode Switch 87 06 .

2.64 EA ECN Signoff 88-01 -

2.65 EA Switch Qualification 88-01 2.66 EA MSP Control 88-01 -

2.67 EA MSP Revision Numbers 88-01 .

2.68 EA Drawing Errors 88-01 , ,

2.69 Mortgaged Echs EA 88-01 2.70 EA FAB Shop Changes 88-01 2.71 EA Lost Panel Enclosure 88-01 2.72 EA ECN Not is EIS 88-01 2.73 EA ECN Error 88-01 2.74 EA Unauthorized Change 88-01 2.75 DV EIS System 88-01 See note 1 2.76 TR Misuse of CARS 88-01 2.77 EA Panel Enclosure Label 88-01 2.78 EA Separation Signoff 88-01 2.79 Mode Switch 87-06 MS 2.80 MS Mode Switch 87-03

, 2.81 TR Training 86-01 2.82 EA Panel Documentation 88-01 2.83 TR Work Transfer 88-01 3.1 EA Found Folder 88-01 3.2 EA ECR Coordinator 88-01 3.3 MS Mode Switch 87-03 3.4 EA Program Info Request 88-01 3.5 MS Moce Switch 87-06 3.6 EA QA Organization 88-01 3.7 NR Quality Emphasis 88-01 3.8 EA FCD/Elem Update 88-01 3.9 FS FSAR Review 87-03 3.10 FS FSAR Review 87-03 3.11 TR Audit Training 88-01

. 3.12 TR Training 88-01 4.1 TR Training 88-01 4.2 FS FSAR Insert 87-03 4.3 NR Interlock for Clinton 87-01 4.4 EA Document Control 88-01 4.5 EA Manual Maintenance 88-01 4.6 EA MP Revision Numbers 88-01 4.7 EA Document Control 88-01 4.8 EA E0P Authorization 88-01 4.9 R ,

Job Responsibilities 88-01 4.10 (EA E0P Reference 88-01 4.11 EA CAR Tracking 88-01 -

4.12 EA Work Commitments 88-01 r

I A3

?

tb

.o e

Stokes Closed i Report NRC in NRC Item Category Allegation Title Report Nonconformance 4.13 EA Question to Mr. G. Senn B8-01 ,

4.14 EA Document Use 88-01 '

4.15 FS FSAR Insert 88-01

• 4.16 TR Subcontractor Error 88-01 4.17 EA ECR Initiation 88-01 -

4.18 DV ICER/ SUN Codes 88-01 , ,

4.19 DV Deferred Verification 87-06 See note 1 4.20 EA Work Around 88-01 4.21 NR Control System Failure 87-01 4.22 EA Termit. ology 88-01 4.23 EA CAR Procedures 88-01 4.24 EA Work Authorization 88-01 4.25 TR Training 86-01 4.26 TR Training 87-01 See note 1 4.27 DV Subcontractors 87-03 See note 1 4.28 TR Training 86-01 5.1 TR Verifier Selection 88-01 5.2 NR Document Handling 88-01 5.3 TR ECN Signature 88-01 5.4 TR Subcontractor Review 88-01 5.5 EA Pressure Switch 88-01 5.5 DV Deferred Verification 87-06 See note 1 5.7 DV Deferred Verification 87-06 See note 1 5.0 NR No QA Manuals 88-01 5.9 FS FSAR Amendment 88-01 5.10 NR Work Qual Too High 88-01 5.11 EA Environm6ntal Issues 88-01 5.12 EA EIS Changes 88-01 5.13 EA Excluded Equipment I* 88-01 5.14 NR NRC Deferred Inspection 87-01 5.15 DV Defe'rred Verification 87-06 See note 1 5.16 EA Altered ECN 87-06 See note 1

- 6.1 EA Fusing Policy 88-01 6.2 NR Post Sign Changes 88-01 6.3 NR 8arrentine Letter 88-01

11. Congressional Concerns N/A FP Fire Protection 87-06 N/A MS Reactor Mode Switch 87-06 87-03-01 l N/A FS FSAR Verification 87-03 N/A DV Deferred Verification 88-01 N/A NR Subcontractor Signatures 88-01 1

A4 ,

I

. _ ,_ _ _ _ _ _ O

,o O

4 m

legend DV-Deferred Verification EA-Changes to Documents '

MS-Hode Switch '

NR-NRC Reserved items TR-Training ,

Nonconformance Identifier , ,

i Example: (87-01-06)

NRCInspectionReportNo.(87-01)

Nonconformance No. (06)

Notes

1. These Stokes items, in general, weie the subject of six nonconformances involving deferred verificatisiis, training, drawing inconsistencies, and corrective action. NRC inspection Report No. 99900403/86-01, Noncon-

- formances 86-01-01 through 86-01-06 address these items.

l e

A5