ML20154E533

Suggests CRGR Meeting on 880127 to Review Proposed Compliance Bulletin Re Defects in Westinghouse Circuit Breakers

ML20154E533
Person / Time
Issue date:	01/20/1988
From:	Murley T Office of Nuclear Reactor Regulation
To:	Jordan E Committee To Review Generic Requirements
Shared Package
ML20151L073	List: ML20151L068 ML20154E459 ML20154E498 ML20154E533 ML20154E822 ML20195G946 ML20195J296 ML20196E010
References
NUDOCS 8809190035
Download: ML20154E533 (26)

Text

{{#Wiki_filter:_____ __ ((> UNITED STATES D g, NUCLEAR REGULATORY COMMISSION .j WASHINGTON, D. C. 20555 \\..v.../ JAN 2 01989 NEMORANDUM FOR: Edward L. Jordan, Chairman Comittee to Peview Generic Reauirements FROM: Thomas E. Murley, Director [ Office of Nuclear Reactor Repulation

SUBJECT:

REQUEST FOR CRGR REVIEW OF PROPOSED COMPLIANCE BULLETIN 88-XX: DEFECTS IN WESTINGHOUSE CIRCUIT BREAKERS We have received your memo dated December 24, 1987, responding to our rtquest dated December 16 for waiver of CRGR review of the sub.iect proposed compliance bulletin. You found that this proposed bulletin represents a new staff posi-tion recuiring CRGR review. Accordingly, please schedule a CRGR review of the proposed bulletin. We suggest a CRGR meeting to consider this issue on January 27, 1988. This bulletin is considered to be a Category 2 item as s pcified in the CRGR charter. The sumary infonnation contained in our December 16 memo remains essentially unchanged, except that now we have determined that it is not necessary to issue a temporary instruction to regional o-/fices to delineate the extent of audits to be perfomed. This is because the prop 9 sed bulletin would require address-ees to perfom the inspections and needed corrective actions recomended by the vendor, with the previously described rodifications, and to affirm completion of those actions to the NRC. The regional offices would not be requested to confirm those actions by means of special inspection effort. l We previously enclosed the proposed bulletin in our December 16 memo. Subse-quent discussion with OGC has caused it to be modified. Along with the modified proposed bulletin, enclosed in this memo is the infomation required 1 l l l CONTACTS: C. Vernon Hodge, NPR 492-8196 i Darl Hood, NRR 492-8961 000919o03500@$$$a REVGP N PNV PDR MEETINo120 J

t Edward L. Jordan 2-1938 3 1 I by the CRGR charter to support issuance of the bulletin (Enclosures 1-8). C. E. Rovi; Di.-actor, Division of Operational Events Assessment, is the i sponsorino Divi >for Director. \\ lcwru h

• f Fu Tho s E. Murley, Directbr /

Offihe of Nuclear Reactor Reaulation s)

Enclosures:

1. Proposed Bulletin 2. CRGR Package 3. NRC IN 87-35 4. NRC IN 87-35, Supplement 1 5. AIT Report on McGuire Ever.t 6 LER on McGuire Event 7. Calvert Cliffs Report 8. Seouoyah Report i 0

I ENCLOSURE 1 Proposed NRC Conollance Bulletin No. 88-XX: ' Defects in Westinghouse Circuit Breakers' l e t. 1 h i i I i k I ~

OMB Noo: 31500011 NRC Compliance Fulletin E8-XX UNITFD STATES NUCLEAR REGULATORY COWISSION OFFICE OF NUCLEAR REACTOR REGULATION WASHINGTON, D.C. 20fE5 January xx,1988 NRC COMPLIANCE BULLETIN NO. 88 XX: DEFECTS IN WESTINGHOUSE CIRCUIT BREAKERS Addressees: For Action - All holders of operating licenses or construction permits for nuclear power reactors. purpose: The purpose of this bulletin is to provide information on Westinghouse series 05 circuit breakers and safety concerns associated with their use and to request that addressees using these breakers in Class IE service perform and document inspection of the welds on the pole shafts and inspection of the clignment in the breaker closing mechanism. Description of Circumstances: The following occurrences have raised concerns about the use of these circuit breakers: McGuire 2: On July 2,1987, a 05-416 reactor trip breaker (RTB) failed to open Tn response to manual trip demands from the control room. The RTB had bound mechanically in the closed position because the main roller had become wedged between a raised seament of the close cam and the nearby side frame plate. Excessive lateral movement of the main drive link and a broken center pole lever to pole shaft weld pennitted the binding to occur. The failure was reproduced once by the Itcensee during bench tests of the RTB at McGuire and several times during detailed laboratory investigations performed by Westinghouse. Substandard welding during fabrication (i.e., porosity, lack of fusion, inadequate extent of welding) caused the weld to break. Details of this failure mode are given in Information Notice No. 87-35, Supplement 1 dated December 16, 1987 The licensee visually inspected the remaining pole shaft welds of the defective McGuire breaker and the other McGuire RTBs and found indications of lack of fusion (i.e., lack of characteristic weld bead ripple, notches pt the edges of s the weld beads, and small evidence of base metal melting). t J

NRC Compliance Bulletin 8P-yX January xx, 1938 Page 2 of 5 Catawba 1 and ?: The licensee inspected all 05-416 RTBs and found a pole shaft with a crack about 1/4 inch long at the finish end of the antibounce lever weld. The licensee also observed lack of fusion at the start ends of the center pole lever and antibourice lever welds. Subsequent examination under magnification of the Catawba pole shaft in the laboratory of an MC contractor revealed two additional cracks, one at each end of the center pole lever weld. After the cracks from the center pole lever weld were retroyed, about half (i.e., 67 degrees of weld arc) of the original 120 degrees of weld are remained. Similarly, 86 degrees of weld are remained after the antibounce lever weld defects we'e removed. The licensee's inspection of the RTBs included checking the alignmellt of the main roller on the close cam surface. Two RT8s were found with excessive lateral tolerance, allowing the roller to strike the side frue plates located adjacent to the close cam, even though toe pole shaft welds were - observed to be intact. The licensee also noted that some pole shaft welds of this type of circuit breaker used in its hydroelectric plants had failed several years ago but that they have performed satisfactorily since they were repaired by additional welding on the opposite sides of the levers. Seouoyah 2: In April 1987, two fillet welds broke on the pole shaft assembly of a D5-416 circuit breaker that energizes the emergency fire protection purps. The weld failures appirently) freed the center mnving contact asserrbly ( shaft that drives the other two moving cont 6ct assemblies, as evidenced by an electrical phasing problem and erratic operation of the fire purp. The two failed welds joined adjacent levers (the center pole lever and the antibounce lever 1 to the pole shaft. The two levers ara connected by a pin. On the basis of engineering analysis, the licensee conclucied that the center pole lever weld failed first because of excessive porosity; the antibounce lever weld then failad because it was inadequately sired and could not accomodate the load nomally supported by the center pole lever weld that was thrust upon it through the connecting pin, i Calvert Cliffs L: In September lo86, a broken weld connecting the center pole lever to the pole shaft in a DS-206 circuit breaker used in Class IE service fcr the control room habitability system was detected during routine mainte-nance surveillance. No adverse effect on breaker perfomance had been noted; the weld for the adiacent antibounce lever was observed to be intact and carrying the load of the broken weld. The licensee's measurements showed that the leg site on the pole shaft side was 0.3 inch and the leg size on the lever arm side was 0.1 inch. On the basis of analysis, the licensee concluded that the failure was due to extensive lack of fusion of the weld to the lever as a result of improper weld technique. The licensee examined an additional 10 welds on this pole shaft and another pole shaft and found that the start ends of the welds in general were not fused properly to the levers and that the weld legs generally exhibited mismatches. Cracks were detected in the start ends of 2 of the 10 welds.

NRC Compliance Pulletin 88-TX January xx, 1988 page 3 of 5 Westinghouse: Both comercial grade and Class IE circuit breakers of the DS series use similar pole shafts or po;sess features associated with the observed binding and electrical phasing problems. Specifically, Model Nos. D5-206, DSL-206, 05-416, DSL-416, and DS d?0 are susceptible to these types of fail-ures. The welds of these pole shafts were randomly inspected during manufac-inre. However, no documentation confims either that in-process insrections were perfomed when the pole levers were welded to the pole shafts or that inspections were perfomed during the dedication of the comercial grade breakers to Class IE service. 4 Discussion: As a result of the operating experiences arid observations discussed above, there is a question concerning the operability of PTBs and other Class 1E circuit breakers of the Westinghouse DS series. Some DS series circuit break. ers mey not have been fabricsted in cortpliance with General Design Criterion (GDC) 1 ard Appendix B 10 CFR 50, and have inadequate welos joining levers to pole shafts. Excessive misalignment of the main rollers on the close cam also can occur. GDC 1 and Appendix B recuire, in part, that components inportant to safety be fabricated to quality standards comensurate with the importance of the safety functions to be perfomed. Consequently, Itcensees should take action to confim compliance with GDC 1 and 10 CFR 50 Appendix B and to inspect all relevant welds and roller clearances according to the manufacturer's specifications and to te,ke appropriate remedial actions to correct deficiencies. On December 1, 1987, Westinghouse issued Technical Bulletin NSID-TB-87-11 (Attachtent 1) as a result of its investigation of the McGuire 2 RTB failure. It recomended 19spection of the pole shaft welds and of the alignnnt in the breaker closing mechanism according to specific criteria and provided guidance for corrective actions if required, including a procedure for the removal and installation of pole shafts. The NRC has reviewed the Westinghcuse technical bulletin and finds that it adequately addresses the NRC concerns subject to t;ertain changes discussed below. Specifically, the NRC has concluded that RTBs should be inspected expeditiously, that in view of the Sequoyah 2 wehi failures welds should be inspecte) for porosity, and that a bypass breaker not meeting the weld criteria in the Westinchoose technical bulletin should be removed from s service. l j Recormended Action for Addressees: The phrases "short-term inspection" and "long-tem inspection" used in this NRC Bulletin are consistent with the phrases as used in the Westinghouse technical bulletin. Specifically, short-term inspections refer to inspections of the three main pole levers (the left pole lever, the center pole lever, and the right pole lever). These short-tem inspections should be perfomed on break. ers at the next available opportunity (e.g., a maintenance outage) or during the next surveillance test for the breaker, whichever is earlier. Long-tem inspections refer to inspections of the four remaining welds on the pole shaft and to the direct check of the alignrrent of the breaker closing mechanism. These lor.g-tem inspections should be perfortred on the breaker prior to restart i following the next refueling outaoe. However, for plants that have not yet i j

NRC Compliance Bulletin 88-XX January 2x, 1988 Page 4 of 5 received an operating license, the recomended implementation periods for the short-tem and long-tenn inspections are modified by this NRC bulletin to mean before fuel loading. As used in this NRC bulletin, the phrase "replacement pole shaft" may include a repaired pole shaft. However, since welding of a pole shaft lever may cause distortion and misalignment of the lever, such repairs should be attempted only after consultation with Westinghouse. Any repaired pole shaft weld should meet the criteria in Section 6.1.1 of the Westinghouse technical bulletin, as supplemented below. Addressees using Westinohouse 05-206, DSL-206, 05 416. DSL-416, and 05-420 circuit breakers in Class IE applications, including RTRs, are requested to perfom short-tem e'id long-tem inspections in accordance with the Westinghouse technical bulletin, except that the following changes should be made~to the following sections: 6.0 Add the following: However, inspection of the 3 main pole shaft welds for all RTBs (both miin and bypass) should be completed within 30 days of receipt of this NRC bulletin. 6.1.1, 6.1.2, and 7.1 Add the followinc: e) porosity - surface pin holes with cumulative diameters less than 1/16 inch in each inch of weld 6.2.4 Delete this section and the reference to it in Section 6.2.3. With regard to Section 6.2.4, any RTB with a pole shaft that does not meet the criteria in Section 6.1.2 should be deemed iroperable and should not be used in the operating or bypass breaner position in the reactor trip switchgear. Such pole shafts should be removed from service and a replacement Dole shaft in-stalled in the breaker before returning it to service. The replacement pole shaft should meet the criteria in Section 6.1.1. Requested Reporting Requirements: If addressees cannot meet this suggested schedule for short-tem and long-tem hspections, they should justify to the NRC their proposed alternative schedules. Records of inspections and corrective actions in response to this NRC bulletin should be documented and maintained in accordance with plant procedures for Class IE equipnent. Additionally, addressees should provide letters of confir-mation to the NRC of the completion of the inspections, including a description of any pole shaft welds or mechanism alignments failing the acceptance criteria 1 and any r.eeded corrective actions, within (1) 30 days of completion of the short-tem inspections and (2) 30 days of completion of the long-term inspections.

NRC Complianct Bulletin 88-XX January xx 1988 Page 5 of 5 Since inspection of the three main pole shaf t welds for all RTBs should be completed within 30 days of receipt of this bulletin, a letter of confirmation of completion of these inspections is requested within 60 days of receipt of this bulletin, The letter of confirmation should be submitted to the appropriate Regional N ministrator under oath er affirmaticn under the provisions of Section 182a, Atomic Energy Act of 1954, as arnended. in addition, the original copy of the cover letter and a copy of any attachment should be transmitted to the U. S. Nuclear Regulatory Comission. Document Control Desk, Washington D.C.

20555, for reproducticn and distribution.

For purposes of NRC accounting, all corre-s pondence associated with this bulletin, including the letter of confimation, siould bear the identifying number TACS 65965/65956. j This request for information was approved by the Office of Management and Budget under blanket clearance number 31500011. Corrinents on burden and dupli-cation should be directed to the Office cf Management and Budget. Reports Management, Room 3208, New Executive Office Building, Washington D.C. 20503. ..Ithcuch no specific request or requirement is intended, the following infoma-tion would be helpful to the NRC in evaluating the cost of complying with this bulletin: (1) staff time to perfom reouested inspections, corrective actions, and associated operability testino (?) staff time to prepare requested documentation (3) additional cost incurred as a result of the inspection findings (e.g., costs of corrective actions, costs of down time) If there are any questions regarding this matter, please contact the Regional Administrator of the appropriate NRC regional office or this office. Charles E. Rossi, Director Division of Operational Events Assessment Office of Nuclear Reactor Regulation Technical Contacts: Darl S. Hood, NRR K. R. Naidu, NRR (301) 492-8961 (301)492-9659 C. Vernon Hodge, NRR C. D. Sellers, NRR (301) a92-8196 (301) 492-8301 Attachrents: 1. Wattinghouse Technical Bulletin NSID-TB-87-11 December 1, 1987 2. List of Recently Issued NRC Bulletins l l l e

? e I INCLOSURE 2 Contents of Package Submitted in Support of Proposed Bulletin on Defects in Westinghouse Circuit Breakers e

---

p----.

n

CRGR Item IV.B. Contents of Packages Submitt:d to CRGR (Rev 4. Stello to List 042387, des 41e(0 342 ff) The following requirements apply for proposals to reduce existing requirements or (regulatory) positions as well as proposals to increase requirements or (regulatory) positions. Each package submitted to the CRGR for review shall include fif teen (15) copies of the following information: WEJECT: DEFECTS IN WESTINGHOUSE CIRCUIT BREAVERS Question: 1. The proposed generic requirement as it is propcsed to be sent out to licensees.

Response

A. The proposed generic requirement is spelled out in the proposed bulletin. In sum, all Itcensees and construction permit holders are requested to follow Westinghouse Technical Bulletin NSID-TB-87-11 (Attachment 1 of the pro)osed bulletin) in in-specting and correcting if necessary the pole shaft welds and the mechanism clignments of the subject breakers. The subject breakers may be used in reactor trip breaker (RTP) or other Class IE applications. The affected Models Nos. are DS-206, DSL-206, DS-416, DSL-416, and 05-420. The proposed method of implemntation for pole shaft welds verification is by visual inspection. The inspections must be performed by a qualified welding inspector, certified per the station's requirements for welding inspection. In situ visual inspection is acceptable; however, tools to aid good visual examination, e.g., enhanced lighting, mirrors, etc., are Ncom-mended. If inadequate weld dimensions, excessive weld porosity, or weld cracks are observed, the weld is to be corrected. Marginally acceptable welds may be used under certain conditions for limited times. The acceptance criteria for continued use are 1. Weld size - 3/16 inch fillet 2. Weld length - 180 degrees continuously around the pole shaft 3. Weld fusion - fused to lever and shaft for the above length 4 Weld cracks - no cracks are allowed 5. Weld porosity - surface pin holes with cumulative diarreters less than 1/16 inch in each inch of weld Poller misalignm nts are to be inspected and corrected if the main roller cam contacts either side frame plate. Correction will probably require replacement of the pole shsft. B. The work is divided into short-tenn and long-tenn inspections. Necessary corrective actions are deemed to be part of these descriptions. In the short-tem inspections, the welds on the 3 central pole levers are to be examined. In the long-tem 1

inspections, the remaining 4 lever welds and the mechanism alignment, specifically whether the main roller is riding evenly on the close cam surface and does not rub against the side frame plates, are to be addressed. C. The NRC bulletin modifies or supplements the Westinghouse bulletin by 1. Reconrnending the short-term inspections for RTBs to be completed within 30 days of the date of the NRC bulletin and confirmed by letter to the NRC within M additional days later, 7. Prohibiting the use of bypass RTB breakers with pole shaft welds that have cracks. 3. Introducing an inspection requirement and acceptance criterion on weld porosity, and 4 Providing inspection and reportino tsbedules for plants that have not yet received an operatieg license. D. Addressees are further asked to report their findings of defi-ciencies and corrective actions and to confirm completion of the inspections to the NRC. Question: II. Draft staff papers or other underlying staff documents supporting the requirements or staff (regulatory) positions. (A copy of all materi-als referenced in the document shall be made available upon request to the CRGR staff. Any comittee member may reonest CRGR staff to obtain a ;opy of any referenced material for his cr her use.)

Response

A. Proposed NRC Compliance Bulletin No. 87-XX: "Defects in Westinghouse Circuit Breakers" 8. NRC Information Notice No. 87-35: "Reactor Trip Breaker. l Westinghouse Model DS-416. Failed to Open on Manual Initiation from the Centrol Room," 30 July 1987 C. NRC Infomation Notice No. 87-35, Supplement 1: "Reactor Trip Breaker, Westinghouse Model 05-416, Failed to Open on Manual Initiation from the Control Room," 16 December 1987 l D. NRC inspection Reports 50-369/87-22,50-370/87-22(Auomented Inspectirn Team Report on McGuire 2 event), 31 fugust 1987 E. Licensee Event Report 50-370/87-009 on McGuire ? event, 3 August 1987 I t l 2 (

F. "Notice of Meeting with Westinahouse Regardino Class IE Switchgear Models DS-416, DSL-416, DSL-420, 05-206, and DSL-206," 21 September 1987. Enclosure 2 (Calvert Cliffs report) G. "Surmry of September 23, 1987, Meeting on Westinghouse Switch ear Failures," 2 October 1987, Enclosure ? (Sequoyah report Ouestion: III. Each proposed requirement or staff (regulatory) position shall contain the sponsoring office's position as to whether the proposal would increase requirerents or staff (regulatorv positions, imple-ment existing requirements or staff (regulatory)) positions, or would relax or reduce existing requirements or staff (regulatory) positions.

Response

A. The proposed bulletin would ensure compliance with Genera' Design Criterion (GDC) 1,10 CFR 50, ard the cuality assurance requirements imposed by Appendix B of 10 CFR 50. The subject breakers are used in various Class IE systems for which the safety-related function of the breaker may be either to open to deenergize a circuit (e.g., a fault current protection or a reactor trip breaker) or to close to energi7e a circuit (e.g., a pump motor breaker or an air fan motor breakerl. Thus plant technical specificationi require the subject breakers to be operable subject to varying action statements and allowed outage times. The proposed bulletin would ensure the operability of the breakers. Accordingly, we conclude that all the require-me9ts of the proposed bulletin implement regulations and do not exceed them. B.- Staff regulatory positions are not altered by this proposed bulletin. 0,uestion: IV. The proposed method of implementation along with the concurrence (and any comments) of OGC on the method proposed. _ Response: I The staff proposes to promulgate this proposed reoutrement by means of a compliance bulletin. This method has been effective in the l past. OGC has no legal objection. Question: t ( V. Regulatory analyses generally conforming to the directive's and guidance of NUREG/BR-0058 and NilREG/CR-3568 (Make sufficient to address the Paperwork Reduction Act, the Regulatory Flexibility Act and Executive Order 12291.) l l l 3 l l l

Response

A. This request for information was approved by the Office of Management and Budget under blanket clearance number 3150-0011 as mreting requirements of the Paper Reduction Act and Executive Order 12291. Suffichnt hours are included in the NRC budget for this request. B. Since this is not a rulemaking action, the Regulatory Flexibil-ity Act does not apply. Question: VI. Identification of the category of reactor plants to which the generic requirement or staff position is to apply (that is, whetber it is to apply to new plants only, new OLs oniv, Ols after a certain date. OLs before a certain date, all Ols, all plants under construction, all plants, all water reactors, all PWPs only, some vendor types, some vintage types such as BWR 6 and 4. jet pump and nonjet pump plants. etc.).

Response

The proposed requir6 ment applies to all nuclear power plants. The subject breakers may be found in any Class IF application. Quest. ion: VII. For each such category of reactor plants, an evaluation which demon-strates how the action should be prioritized and scheduled in light of other ongoing regulatory activities. The evaluation shall docu-ment for consideration infomation available concerning any of the following factors as may be appropriate and any other infomation relevant and material to the proposed action: A. Statement of the specific objectives that the proposed action is designed to achieve;

Response

i 1. The objective is to avoid mechanical binding of Class IE breakers, such as that which occurred at McGuire 2 and elsewhere, because of defective pole shaft welds and excessive misalignment of the closing mechanism. The proposed action would provide ressonable assurance that the subject breakers do not exhibit the prerequisite for ecchanical binding. i 2. The problem that this proposed bulletin addresses is considered to be of high priority. This is based on the l NRC view of the importance of the reactor protection system, emergency core cooling systems, emergency electri-cal power generation systems, and other safety related systems. All these systems potentially contain the subject breakers at any nuclear power plant. l 4

Ouestion: 'B. General description of the activity that would be required by the licensee or applicant in order to complete the action;

Response

1. The requested actions are separated into short-tem inspec-tions (including any required corrective actions) and long-term inspections (including any required corrective actions). 2. The short-tem inspections have two components: those for RTBs and those for circuit breakers in other Class lE applications. Question: 3. Fow does this requirement affect other requirements? Does this requirement mean that other items or systens or prior analyses need to be reassessed?

Response

a) This proor,ed bulletin would not affect other require-ments, n, alyses, or items. It would not affect prior analyses. It does relate to Generic Letter 83-28 with respect to implementing the vendor instructions. The proposed NRC b9lletin would reinforce the instructions of Westinghcuse, the vendor of tFe subject breakers, on actions its customers should undertake, b) No other impr.ct is identified. Question: 4 What plant conditions are needed to install, conduct preoperational tests and declare operable? i

Response

al Most of the affected breakers can be inspected and corrected if necessary while the plant is operating at full power. I L b) The proposed bulletin does not recomend any addition-al periodic testing of the breakers beyond normal surveillance testing. l Question: t S. Is plant shutdown necessary? How long? l l l F

Response; a) Inspections can be conducted without requiring plant shutdown. However, if defective welds are found, the effect of the proposed bulletin may indeed result in plant shutdowns. The NRC staff does not expect many welds to be found to satisfy the unrestricted accep-tance criteria of Section 6.1.1. of the Westinghouse technical bulletin, as modified by the proposed NRC bulletin. However, the NRC staff accepts the limited use Westinghouse acceptance criteria of Section 6.2 except for Section 6,2.4 on use of cracked welds in RTB bypass applications. Employing these ifmited use criteria should avert unnecessary shutdowns. b) Estimates of times required for both the short-term and long-tenn inspections of a single breaker are given below. (1) Short-tenn Inspections (in situ 3 central welds,donebyaskilledtechnician) (a) Administrative controls 2 hrs (b) Visual inspection 1 (1) No problem found with welds (ii) Closer examination or replacement of pole shaft required 3-7 (c) Replacement and test of breaker 3 (d) Total 6-12 hrs (2) long-tem Inspections (bench. 4 outer welds, alignment, done by a skilled technician) (a) Administrative controls 2 hrs (b) Visual inspection (1) No problems found with welds, but disassembly required for alignment inspection 2 (ii) Closer examination or replacement of pole shaft required 3-7 (c) Replacement and test of breaker > 3 (d) Total 7-12 hrs 6

Question: 6. Does design need NRC approval?

Response

a) The proposed corrective actions do not need NPC approval before taking effect. However, a licensee would need to consult Westinghouse in 2 instances: (1) Before returning to service those breakers found with main roller-close cam misalignment--see Festinghouse Section 7.2.1.7 (?) Refore welding pole shaft levers--see NRC pro-posed bulletin, page 3 last paragraph b) The proposed bulletin requests that the licensee inforr. NRC by letter of confirmation that the inspec-tions have been completed. This letter of confinna-tion is to include reporting of deficiencies found in the inspections and needed corrective actions. Question: l 7. Does it require new equipment? Is it available for pur-chase in sufficient quantity by all affected licensees or must such eculpment be designed? What is the lead time for availability?

Response

a) The simplest corrective action is to replace existing breaker pole shafts not meeting acceptance criteria with good ones. The NRC understands that Westinghouse has about 70 spare pole shafts and have several thousand on order. This shculd be adequate. l b) Otherwise, licensees and permit holders will have to repair existing pole shafts. The NRC staff believes the capability for adequate inspection and adequate repair of existing pole shafts does reside at each plant. However, warranty considerations may preclude repair as a corrective action unless acceptable to both addressees and Westinghouse. Question: 8. Fay it be used upon installation or does it need staff approval before use? Does it need Tech Spec changes before use? i 7 i

Response

a) A breaker verified or made operable according to the proposed bulletin would be usable upon installation. NRC appruval is not required. b) No changes in the technical specifications are required. Mstion: 9. Consistent with the first two items above, provide the basis for requiring or permitting implementation by a given date or on a particular schedule.

Response

a) The item of first concern is operability of the RTPs. The failure of 2 RTBs could result in an anticipated transient without scram (ATWS) accident. To indicate the seriousness of this natter, all RTPs are to be inspected and made operable within 30 days. Reporting to the NRC of any deficiencies found would be required within 30 additional days, b) The 3 welds experiencing the highest loads in those breakers in other Class IE applications are also of concern. The proposed bulletin would have the licensee perfom all short-tem inspections at the earliest opportunity, such as a maintenance shutdown, or the next surveillance test, whichever is earlier. In all cases a short-term inspection should be done by startup following the next refueling shutdown. All long-tem inspections should be done prior to startup following the next refueling shutdown. The long-tem inspection schedule is provided because the 4 outer welds experience lesser loads and are there-fore less likely to fail. This schedule affords enough time to do the,iob well but is not so long that the work would go unattended. All reporting would be due within 30 additional days from the time the associated work is completed. This period is typical of report preparation schedules. Question: 10. Other acceptable implementation schedules and the basis therefor. This should include sufficient infomation to demonstrate that the schedules are realistic and provide sufficient time for in-depth engineering, evaluation, design, procurement, installation, testing, develoreent of operating procedures, and training of operators. 1 8 h:

Response

a) The NRC sta'f does not foresee a real need for an alternative schedule, but would consider any address-ee's request for additional time on its own merits. b) The main effect of the proposed bulletin is to but-tress the intent of the surveillance testing reouired by technical specifications. A breaker with an inadequata weld, as defined by the relevant Westinghouse shop drawing, may pass the surveillance test and thus be declared operable. However, the NRC staff has no basis for assurance that the inadeouate weld will not cause the breaker to fail on demand. The implementation schedule is realistic since no in-depth engineering, evaluation, desion, development of operatino procedures.nor training of operators are required, the Westinchouse technical bulletin has alretdy been completely developed and details the inspections and corrective actions needed. Question: C. Potential change in the risk to the public from the accidental offsite release of radioactive material

Response

The proposed bulletin would ensure that the safety-related circuit breakers meet minimum standards as specified by the vendor. Thus, the bulletin is considered necessary to ensure an acceptable level of public health and safety. Question: D. Potential impact on radiological exposure of facility employees and other onsite workers;

Response

1. No additional occupational radiation exposure is expected i from the recomended actions of the proposed bulletin. The breakers are located in nonradioactive areas of the plant. P. No additional public exposure is expected from the recom-mended actions of the proposed bulletin. Question: E. Installation and continuing costs associated with the action, including the cost of facility downtime or the cost of construc-tion delayi 9

Response

1. Estimates of minimum inspection costs are sumarized as follows (assuming $100 per person hour): (1) Inspection, hrs / breaker 6 (2) Cost, $/ breaker 600 (3) Cost, $K (a) Westinghouse plants (50 brkrs/ plant; 54 plants) 1,6?0 (b) Sequoyah plants (150 brkrs/ plant; P plants) 180 e (c) Calvert Cliffs plants (?00 brkrs/ plant; 2 plants) 240 ~ (d) Other plants (10 brkrs/ plant; 42 plants) 252 (e) Total $2,292K 2. No continuing costs are envisioned. Question: F. The potential safety impact of changes in plant or operational complexity, including the relationship to proposed and existing regulatory requirements and staff positions;

Response

1. The inspections and possible replacements or repairs of faulty components will not increase plant or operational complexity. 2. The relation of the proposed bulletin to the addressee-vendor interface with respect to Generic Letter 83-28 has been addressed above. Question: 3. Is it only computation? Or does it require or may it entail engineering desion of a new system or modification of any existing systemt/ System design would not be affected by the proposed bulletin. 10 4 0

Ouestion: G. The potential impact of differences in facility type, design or age on the relevancy and practicality of the proposed action;

Response

No effects of this nature are identified. Question: H. The estimated resource burden on the NRC associated with the proposed action and the availability of such resources; 1. An assessment of costs to NRC,

Response

The cost to the NRC consists of preparing the bulletin. ensuring that licensees respond (confirmation by project managers), perfoming limited routine inspections of licensee implementations, and closing the bulletin (by project managers). These costs together are estimated to require minimum staff resources and no contractual assist-ance. Question: 2. Schedule for staff actions involved in completion of requirement (based on hypothesired effective date of approval).

Response

After issuing the proposed bulletin, the originators would be available to answer questions by telephone. Pro.iect managers will verify that each licensee submits confirra-tion that the bulletin actions are completed. Receipt of this information will constitute closure of the bulletin. The regional offices do not need to do plant 'oy plant inspections, but they should take appropriate actions if, during the course of other inspecticns, followups to events, or reviews of equipment malfunctions, it is found that a licensee did not perfom the actions contained in the bulletin. The regional offices may, however, at their discretion do audit inspections of selected licensees. However, no regional office actions are recuired for bulletin closcout. Question: t I. Whether the proposed action is interim or final, and if interim, the justification for imposing the proposed action on an interim basis. 11

Response

The proposed bulletin represents a sinole action that would not be repeated or extended in the future. Question Vl!!. For each evaluation conducted pursuant to 10 CFR 50.100, the l proposing office director's determination, together with the rationale for the determination based on the considerations of the above, that ) A. there is a substantial increase in the overall protection of public health and safety or the coeren defense and security to be derived from the proposal; and

Response

The proposed bulletin would ensure that the safety-related circuit breakers meet minimum standards as specified by the vendor. Thus, the bulletin is considered necessary to ensure an acceptable level of public health and safety. l Question: B. the direct and indirect costs of implementation, for the facilities affected, are justified in view of this in-creased protection. l

Response

1 The direct and indirect costs, discussed above for question YII.E., are justified in view of this increased protection. f Question: IX. For esch evaluation conducted for proposed relaxations or decreases in current requirements or staff positions, the proposing office director's determination, together with the rationale for the deter-mination based on the considerations of the above, that the public health and safety and the common defense and security would be adequately protected if the proposed reduction in requirements or (regulatory) positions were implemented, and the cost savings attrib-uted to the action would be substantial enough to justify taking the action.

Response

This item is not applicable to the proposed bulletin because no relaxation or decrease in current reovirements is being proposed. i 12 t l

c ENCLOSURE 3 NRC Inforeation Notice he. 87-35: "Reactor Trip Ereaker, bestinghouse Model DS-416. Failed to Open on Fanval Ins tt ation f ree the Centrol Roon," 30 July 1957 ol' l I 6 l

SSINS No.: 6835 IN 87-35 I UNITED STATES NUCLEAR REGULATORY C0991!SS!0N OFFICE OF NUCLEAR REACTOR REGtrLATION WASHINGTON, D.C. 20555 July 30, 1987 NRC INFORMATION NOTICE NO 87-35: REACTOR TRIP BREAKER, WESTINGHOUSE MODEL 05-416. FAILED TO OPEN ON MNUAL IN!TIAT!0N FROM THE CONTROL ROOM Addressees: All nuclear power reactor facilities holding an operating license (OL) or constrvction permit (CP) employing Westinghouse DS-416 reactor trip breakers, purpose: This notice is provided to alert recipients to a potentially significant safety problem associated with a reactor trip breaker (PTB). The NRC expects that recipients will review this notice for applicability to their facilities and consider actions, if appropriate, to preclude a similar problem. However, suggestions contained in this notice do not constitute PDC requirements; therefort, no specific action or written response is reouired. I Description of Circurstances: On July 2,1987, McGuire Nuclear Station l' nit 2 was perfoming control rod drop tests af ter its recent refueling outage. This test was in progress with the plant in Ede 3 (hot shutdown). With all control rods inserted and the RTBs closed for testing the next bank of control rods, station personnel smelled smoke in the vicinity of the RTBs. A manual trip of A and B train RTBs was initiated from the control room. Only the A train RTB opened. The R train RTB was eventually tripped manually at the breaker panel. The smoke had come from the B train breaker snunt trip coil, which had burned and shorted while trying to open the breaker. The coil is designed for intermittent duty and to carry current only until the breaker opens. Failure of the breaker to open resulted in a prolonged and damaging current. Operator; in the control room stated that open indications for both the A and B train redundant RTBs were observed for all attempted breaker opening evolutions during the control rod drop testing process. However, the event recorder indicated that the B train RTB failed to open on a previous manual trip attempt (approximately 4 minutes before) when operators were setting up for the control rod drop test on the last bank of rods. 4707t?0073 [p-

IN 87-35 July 30, 1987 Page 2 of 3 An NRC Augmented Inspection Team (alt) evaluated the licensee's investigation into the reactor trip breaker problem. Abnormal wear and a broken weld were found in this early vintage of Westinghouse DS-416 breaker (see Westinghouse Figure, Attachment 1). The broken weld was on the main drive link between the I center pole lever and the pole shaft. Except for the shunt trip coil that had burned and shorted while trying to open the breaker, the breaker's electrical controls and auxiliary contacts were verified to be properly wired and operat-ing as designed. The cause for the anomalous breaker status indication is still under investigation. Attempts to repeat the condition, where the breaker was mechanically binding in the closed position, were minimally successful. Preliminary conclusions of the alt are that the breaker's mechanical bindin wear (greater than 2000 cycles of operation)g was caused by a combination of , manufacturing tolerances in this early vintage breaker, and the broken weld. These factors may have combined to allow sufficient lateral movement of the main linkage to cause it to jam at or near full breaker closure and thus prevent the breaker from opening. Since the control room operating personnel stated that they observed the open indication on the closed B train RTB, the field wiring is being verified by the licensee to ensure that wiring is as designed. The shorted shunt trip coil had allowed 125 volts de between the positive teminal and the chassis; a "sneak" circuit is possible. D,iscussion: ) Final conclusions for the cause of this event have not been reached. Further investigation and dismantling of the breaker will be conducted in Westinghouse laboratory facilities. The licensee and NRC will participate in this investi-gation. If the results indicate findings different than the above preliminary conclusions, a supplement to this notice will be issued, t i The licensee is inspecting all of the RTBs for signs of abnormal wear, cracks in welds, and excessive lateral play (greater than 1/8 inch) in the roller end of the main drive link where it contacts the close cam. This measurement had I not previously been part of the periodic preventive maintenance for the RTB. l Moreover, following any reactor trip, the licensee is ensuring the open post-l tion of both RTBs by inspecting the breaker before reclosure. These are short-term corrective actions until the detailed analysis of the deficiencies i is completed. A significant number of generic comunications have been issued with regard to reactor trip breakers (RTBs) and similar circuit breakers used in safety-i related systems. Such coesnunications that may be related to the matter in this t i information notice are listed in Attachmnt P. I i l i l

IN 87-35 l July 30, 1987 Page 3 of 3 No specific action or written response is required by this infonnation notice. If you have any questions about this matter, please contact the Regional Administrator of the appropriate NRC regional office or this office. f harles E. Rosst Director Division of Operational Events Assessment Office of Nuclear Reactor Regulation Technical Contacts: C. Vernon Hodge, NRR (3011492-819e I T. Peebles, R!! j (404) 331-4196 A. Ruff, R!! 1 (404)331-5540 ,1 l Attachments: 1. Power Operated f Stored-Energy) Mechanism Graphic Details 2. Generic Comunications on Reactor Trip Breakers and Similar Circuit t i Breakers ( 3. List of Recently Issued NRC Infomation Notices a r 4 I ( l i r i i i 1 J l

Center Po12 Lever @)q-Croke. .1d l =i L v -eh A' n [A, kW.. o g g NN d( _ s i

. [

I G 3.- ( ( c ['g; 1 N T,C f e / / ~ /, e ~ A Y,? N g q ~ j l l

1. SHUNT TRIP DEVICE

10. RATCHETWHEEL

19. RESET SPRING I

2. TRIP SHAFT

11. HOLD PAWL

20. CLOSING SPRING ANCHOR

3. ROLLER CONSTRAINING UNK

12. DRIVE P1 ATE

21. POLE SHAFT

4. TRIP LATCH

13. EMERGENCY CHARGE PAWL

- 22. MOTOft

5. CLOSE CAM

14. OSCILLATOR

23. EMERGEBICY CHARGE HANDLE 1

S. STOP ROLLER

15. CRANK SHA7

24. MOTOft CIIAIOK ASIO taamaru F

7. SPRING RELEASE LATCH

15. EMERGENCY CHARGE DEVICE

25. ABOVISIG CONTACT ASSEGASLY l

S. SPRING RELEASE DEVICE

17. CRANK ARM

26. ISISULATING LINK l

S. OSCILLATOR PAWL IS. CLOSING SPRISIG

27. MAIN DRIVE L8NK i

I O 1 Figure 2-16. Power-Operated (Stored-Energy) Medianism Graphic Details (Oose Spring Shown in the Charged Position) { j ] f Itroken weld from center pole lever to pole shaft (21). Another lever not shown in diagram 3 had an intact veld. The combination of broken weld, manufacturing tolerance, and high y cycle wear are considered to be factors in the mechanical bindiig of the breaker. (Figure f ra Maintenance Program Manual PfM-WOCRTSDS 416-01 for li type DS-416 reactor trip i circuit breakers.) I l ~ IN 87-35 July 30, 1987 Page 1 of 1 GENERIC ComuMICATIONS 04 REACTOR TRIP BREAKERS AND SIMILAR CIRCUIT BREAKERS Circuit Breakers Equipped With a Shunt Trip," July 31, yhouse Mn1ded Case Information Notice 86-62, "Potential Problems in Westin 198J. Infomation Notice 85-93, "Westinghouse Type DS Circuit Rreakers, Potential Failure of Electric Closing Feature 8ecause of Broken Spring Pelease Latch Lever," Deced er 6, 1985. Bulletin 85 02, "Undervoltage Trip Attachments of Westinghouse 08-50 Type Reactor Trip Breakers," November 5,1985. Infomation Notice 85 58, "Failure of a General Electric Type AK-2-?5 Reactor Trip Breaker," July 17, 1985.

---

Supplement 1. Noveder 19, 1985.

Information Notice No. 83-76, "Reactor Trip Breaker Malfunctions (Undervoltage Trip Devices on GE Type AK-2-25 Breakers)," November 2,1983. Generic Letter 83-28. "Required Actions Based on Generic !q1(cations of Salem ATWS Events," July 8, 1983. Infomation Notice 8318. "Failures of the Undervoltage Trip Function of Reactor Trip System Breakers," April 1, le83. Bulletin 83-04, "Failure of the Undervoltage Trip Function of Reactor Trip Breakers " March 11,19A3. Fulletin 83-01, "Fallure of Reactor Trip Breakers (Westinghouse 0R-50) to Open on Automatic Trip Signal," February 25, 1983. Circular 81-12. "!nadequate Feriodic Test Procedure of PWR Protection System," July 22, 1981. Bulletin 79-09, "Failures of GE Type Circuit Breaker in Safety Related System," April 17, 1979.

IN 87-35 July 30, 1987 LIST OF RECENTLY !$$UED INFORMATION NOTICES 1987 Information Date of Notice No. Subject Issuance Issued to 87-34 Single Failures in Auxiliary 7/24/87 All holders of an Feedwater Systems OL or'a CP for pressurized water reactor facilities. 87-33 Applicability of 10 CFR 7/24/87 All NRC licensees. Part 21 to Nonlicensees 87-3? Deficiencies in the Testing 7/10/87 All nuclear power of Nuclear 3rade Activated reactor facilities Charcoal. holding an OL or CP. i 87-31 Riocking, Bracing, and 7/10/87 All NRC licensees. Securing of Radioactive Materials Packages in Transporta tit.n. t 87-30 Crr.cking of Su ge Ring 7/2/87 All nuclear power Brackets in Lasge General reactor facilities Electric Compan/ Electric holding an OL or CP. Motors. 87-29 Recent Safety Related 6/26/87 All NRC licensees Incidents at large authorized to possess L frradiators, and use sealed sources in large irradiators. 87-28 Air Systems Problems at 6/??/87 All nuclear power U.S. Light Water Reactors, reactor facilities ( j holding an OL or CP. 87-27 Iranian Official Implies 6/10/87 All nuclear power Vague Threat to U.S. reactor facilities Resources. holding an OL or CP, research and nonpower reactor facilities, i and fuel fabrication and processing facilities using or possessing formula quantities of special nuclear material. OL = Operating License CP = Construction Fermit

UNITED STATES NUCLEAR RESULATORY COMMISSION = sa =**

• • • e******

WASHINGTON, D.C. 20006 E e.. = oniciat auswass 3 PtNALTV F04 PervAft ust,6300 i / I t t t l [ l i I J l I a + I J f i i i r [ I e I c l t i i I T

+ t ENCLOSURE 4 NAC Inforestion Notice No. 67-35. Supplement 1 ' Aeactor Trip Ereak er, Westinghouse Mode.' 05-416, Failed te k.en on Manual Initiation from the Control Acos," 16 December 1987 h 9 4

== ( 8 h I l l i I i L i l i i i i w....

IN 87-35, Supplement I UNITED STATES NUCLEAR REGULATORY COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION WASHINGTON, D.C. 20555 -December. 16, 1987 NRCINFORMATIONNOTICENO.8(-35, SUPPLEMENT,1: REACTOR TRIP BREAKER, WESTINGHOUSE H0 DEL DS-416, FAILE0 TO OPEN ON MANUAL INITIATION FRON THE CONTROL ROOM Addressees: All holders of operating licenses or construction pemits for nuclear power reacturs.

Purpose:

This infomation notice is being provided to alert addressees to the determi-nation of the cause nf the mechanical binding that resulted in the failed reactor trip breaker (RTB) described in Information Notice No. 87-35. The NRC is considering the need to request action by licensees usina Westinghouse DS series breakers in Class IE applications. This supplementai notice also discusses other concerns that arose during investigations of the RTB failure but that did not contribute to the binding of the RTB. It is expected that recipients will review the infomation for applicability to their facilities and consider actions, as appropriate, to avoid similar problems. However, suggestions contained in this information notice do not constitute NRC require-ments; therefore, no specific action or written response is required.

Background:

i Infomation Notice 87-35, dated July 30, 1987, disctssed the July 2, 1987, event at McGuire 2 in which an RTB would not open upon receipt of an electrical comand signal. The RTB had bound aechanically. The shunt trip coil burned and shorted during the attempt to open the breaker. Operators in the control room stated that they observed open indications for both redundant RTBs, but the event recorder showedithet only one had opened. The licensee's investiga-tion, observed by an NRC Auamented Inspection Team, revealed abnomal wear of the pole shaft assembly and a broken weld joining the center pole lever and the pole shaft, but did not' identify the specific cause of binding. Because the licensee's facilities for further investigation were limited, further investi-gation was to be conducted in a Westinghouse laboratory. i i l 6MM004H hp. 1

IN 87-35, Supplement 1 December 16, 1987 Page 2 of 4 Failure Mode: The McGuire 2 RT8 failed to open because the main roller was wedged between the raised edge of the close cam and the right-hand side frame plate (viewing the RTB from the rear). A labeled view of the RTB mechanism is shown in Figure 1. When the RTB is closed, the position of the mechanism is as shown in Figure 2. A conceptual sketch showing the main roller wedged between the right side frame plate and the left cam segment, as viewed from the rear of the RTB, is shown in Figure 3. Similar wedging could occur between the left side frame plate and the right cam segment. In the Westinghouse DS series breakers, the close cam (item ? in Fic9re 1) is composed of four steel segments that are sandwiched together and held by three rivets. The two outer segments are heat-treated steel; the two inner segments are non-hardened steel. The surface of the seg;nents is supposed to be of uniform shape. However, on the McGuire RTB, the two outer seoments are slightly(item 15). larger than the inner segments, providing the edge to catch the main roller Only a slightly raised cam edge is necessary to allow biliding. In addition, the distance.between the inner surface of the close cam edge and the side frame plate (item 19) must be close to the width of the main roller. The mair, roller can become wedged durina the closing action of the breaker. As the close cam rotates, the edge of the main roller is caught, as shown in } Figure 4. Continued rotation of the close cam causes the main roller axis to straighten. This action causes the edges of the main roller to attempt to separate the close cam and the side frame plate. However, the close cam and side frarne plate are not free to move and, therefore, they wedge the main roller in place. When an attempt is made to trip the breaker, the wedging of the main roller prevents the main roller from rolling down the close cam face to allow the circuit breaker to open. The wedging of the main roller also prevents full discharge of the closing springs.not shown in Figure 1), leaving the close cam 18 degrees from a fully rotated position. Both lateral displacement of the main roller end of the main drive link (iten 14) and a small rotation (3 to 5 degrees) of the main roller axis are necessary to allow wedging. If the weld joining the center pole lever (item 9) to the pole shaft (item P) is sound, the main roller end of the main drive link could still move laterally and even allow the main roller to strike the side plate. However, a sound weld would not allow sufficient rotation of the axis for wedging to occur. A large number of cycles of operation (3000 or more), however, could cause wear that would allow the necessary rotation of the axis. Additional details on the failure mode and the Westinghouse tests are contained it: References 1 ?, and 3. Other Concerns: Stop Poller Binding. The inspection of the McGuire 2 RTB components at Westinghouse revealed that the close caa surface had been peened. The peening l

IN 87-35, Supplement 1 December 16, 1987 Pace 3 of 4 1 flattened and laterally expanded the surface of the outer cam segments, creat-ing a mushroom shape. Of key concern was mushrooming in the area of the stop roller (item 1), which holds the mechanism in readiness for release of the spring release latch (iten 16). The extreme mushrooming impeded rotation of the stop roller. It is possible that sufficient mushrooming could totally prevent operation of the stop roller, which could prevent closure of the circuit breaker upon demand. While not of safety concern for an RTB, this failure to close condition would be of concern for a breaker in a Class IE application requiring energization of the connected loads. RTB Position Indicating Light. At McGuire, red and green lights placed below each of the two RIB spring-loaded manual control switches in the control room serve to indicate whether the associated RTB is closed or open, respectively. Operators are trained to operate both control switches simultaneously, one with each hand, and to interpret each set of lights as representing the actual status of each breaker. However, the red lights serve the additional purpose of indicating continuity of the shunt trip circuit for each breaker. The design circuitry is such that the absence of the red light can mean either that the breaker has Opened as intended or that the associated shunt trip circuit has been interrupted. Determining which is the case cannot be done from the red light behavior alone; rather, the red light behavior must be interpreted in combination with the green light behavior and other control room . Indications, such as rod position displays. Operators need to understand that a "malfunction" of the red licht may in reality be a valid Indication that the associated shunt trip circuit is inoperable. Additionally, it is important that operators understand that the combined absence of the red and areen lights after release of the spring-loaded manual control switch may mean that the associated breaker has failed to open in response to the electrical demand and that immediate local verification or trip action is needed. At McGuire, pressing the manual trip plate at the RTE did not open the breaker, but manipu-lating the manual spring-charging handle did open it. Since the McGuire event, the licensee has modified the requalification training program for operators to ensure that they understand the potential meaning of the various combinations of RTB indications and that they follow appropriate verification procedures for suspected "malfunctions" of these indications. Trip Latch Pivot Pin. During an NRC inspection at Braidwood ? in 1.1te Septem-ber 1987, the licensee reported that a Westinghouse D5-416 PTB at Braidwood 1 failed to close because im to the trip latch (item 5) proper brating nf the pivot pin (item 4 in Figure 1) resulted in disengagement of the twc. While not of safety concern for an RTB, this failure to close condition would be of concern for a breaker in a Class IE application requiring energization of the connected loads. )

IN 87-35, Supplement 1 December 16, 1987 Page a of 4 No specific action or written response is required by this information notice. If you have any questions about this matter, please contact the technical contact listed below or the Regional Administrator of the appropriate regional

office, f

f% Charles E. Rossi, Director Division of Operational Events Assessment Office of Nuclear Reactor Regulation Technical Contacts: Vern Hodge, NRR (301)492-8196 Darl Hood, NRR (301)492-8961 K. R. Naidu, NRR (301) 492-9659

References:

) 1. Interim Report on McGuire 2B Reactor Trip Breaker Failure Evaluation and Reconinended Corrective Actions from Westinghouse," Franklin Research Center, September 30, 1987, enclosure to letter from D. S. Hood, NRC, to H. B. Tucker, Duke Power Company, October 16, 1987, NRC Docket No. 50-370 ?. NRC Inspection Report Nos. 50-369/87-22 and 50-370/87-22, August 31, 1987 3. "Reactor Trip Breaker Failure Due to Mechanical Failure," Licensee Event Report 50-370/87-009 Duke Power Co., Aug%st 3, 1987 l l Attachments: 1 1. Figure 1. Linkages of DS-416 Breaker Mechanism 2. Figure 2. Position of Mechanism with RTB Closed 3. Figure 3. Roller Wedged Between Left Cam Segment and Right Side Frame Plate 4. Figure 4. Binding of Roller 5. List of Recently Issued NRC Infomation Notices 1 1

t < 1 At 1 IP,t a c hm e n t 87-35, Supp. 1 December 16, 1987 Page 1 of 1 2$ 1 e

3. Roller Constraining k

Link 4 Pivot Pin

5. Trip Latch

6. 1rio Shaft Latching Surfae.e

7. Trip Sha f t

8. Pole Shaft 1

9. Center Pole Lever

/ h IC. Pole Lever Pin -1~Q-D ',

11. Moving Contact Arm

'"s

12. Stationary Arcing 4

b,8 i e Contact /

13. Moving Contact 5

$,,, q Pivot Pin 15l !n of he 14hh h f 1 e ,,s 1

16. Spring Release L.

4 Latch L

17. Insulating Link

[ 2 Me h Adjusting Stud j f and Locknut Md /g h .ff la. Insulating Link 1?. Mechanism Side Front of Rear of Breaker Breaker Figure 1. Linkages of 05-416 Breaker Mechanism Shown with CB Open and Springs Charged (Source: Instructions for Low-Voltage Power Circuit Breakers Types 05 ar.d DSL. Westinghouse Electric Corp., Instruction Bulletin 33-790-!E) ) i

Attcchment 2 IN 87-35, Supp.1 "December 16, 1987 Page 1 of 1 O..i.. " 9 t ,~ f ~ / / p ,p k # ' 'Q \\ i O ) ,J s . 's s 9' ..y. j e# w Front of Rear of Breaker Breaker l Figure 2. Position of Mechanism with RTE Closed (Source: Instructions for f Low-Voltage Power Circuit Breakers. Types DS and DSL, Westinghouse Electric Corp., Instruction Sulletin 33-790-IE) A

At tachmen t 3 IN 87-35, Supp. 1 December 16, 1987 Page 1 of 1 Left Side Frame Plate (When Viewed from Rear of RTB) . Roller Right Side Frame Plate / (When Viewed from Rear of RTB) / Roller Axis / A / / / l / / Point of Wedging Between Side Frame j[ and Raised Edge of Close Cam / / f / / l / l / l / / r' Cam Segments r l l l / l / l / l / / / Spacers l / j l Crankshaft / l l t[ t j u Figure 3. Roller Wedged Between Lef t Can Segment and Right Side Frame Plate (Conceptual Drawing, Not Fully to Scalei Source: Franklin i Research Center, Interim Report, September 30, 1987) IN 87-35, Supp. 1 December 16,1987 Page 1 of 1 i e' / N (A) Edg3 of Roller / g caught by p/ N Cam segment j w j\\ '# "7 %j L h Direction of Cam Surface Rotation Right Side Frame Plate [ (Viewed from Rear of RTB) / / (B) i f N Roller Axis Straightens, l p N Causing Sideways Displacement.of Cam l f and Right Side Frame Plate. / . Displacement Results in Force That I / Pinches and Retains Roller. / h -m ~ / l / i N \\ k / \\ / / p / / Figure 4. Einding of Roller f (Source: Franklin Research Center, Interim Report: September 30, 1987) --n-_ -_,--,.--,.--,n _,,,,,,,._ne_ _-_ _,,,.,,,, - IN 87-35, Supplement 1 December 16, 1987 'Page 1 of 1 LIST OF RECENTLY ISSUED NRC INFORMATION NOTICES 1987 Informat16n Date of Notice Nc. Subject Issuance Issued to 87-63 Inadequate Net Positive 12/9/87 All holders of OLs Suction Head in Low Pressure or cps for nuclear Safety Systems power reactors. 87-62 Mechanical failure of 12/8/87 All holders of Ols Indicating-Type Fuses or cps for nuclear power reactors. 87-61 Failure of Westinghouse 12/7/E7 All holders of OLs W-2-Type Circuit Breaker or cps for nuclear Cell Switches. power reactors. 87-60 Depressurization of Reactor 12/4/87 All holders of OLs Coolant Systems in or cps for PWRs. l Pressurized-Water Reactors ~ 86-108 Degradation of Reactor 11/19/87 All holders of Ols Supp. 2 Coolant System Pressure or cps for nuclear Boundary Resulting from power reactors. Boric Acid Corrosion 87-59 Potential PHR Pump Loss 11/17/87 All holders of OLs or cps for nuclear power reactors. 87-58 Continuous Comunications 11/16/87 All nuclear power Following Emergency reactor facilities Notifications holding an OL and the following fuel facilities that have Emergency Notification Systems: Nuclear Fuel Services, Erwin, TN; General Atomics, San Diego, CA; UNC, Montville, CT; and B & W LRC and 8 8 W Navy, Lynchburg, VA. OL = Operating License CP = Constnaction Pemit

i 2 O ENCLOSURE 5 NRC Inspectdon Reports 50-369/87-22, 50-370/07-22 (Augmented Inspection Team Report on McGuire 2 event), 31 August 1987 O se h is 1 l 4 l E

8 UNITED STATES [po c.tcy% NUCtEAR REGULATORY COMMISSION 3* REG 10 Nil j ,i ( j 101 MARIETT A STREET.N.W. e ATLANT A. GEORGI A 30323 %,*****,/ AUG 31 1987 Docket Nos. 50-369, 50-370 License Nos. NPF-9, NPF-17 Ouke Power Company ATTN: Mr. H. B. Tucker, Vice President Nuclear Production Department 422 South Church Street Charlotte, NC 28242 Gentlemen:

SUBJECT:

NRC INSPECTION REPORT NOS. 50-369/87-22 AND 50-370/87-22 This refers to the Nuclear Regulatory Comission (NRC) augmented inspection conducted by T. A. Peebles on July 7-10 and 29-30, 1987. The inspection included a review of activities authorized for your McGuire facility. At the conclusion of the inspection, the findings were discussed with those members of your staff identified in the enclosed inspection report. Areas examined during the inspection are identified in the report. Within these areas, the inspection consisted of se',ective examinations of procedures and representative records, interviews with personnel, and observation of activities in progress. Within the scope of the inspection no violations or deviations were identified. In accordance with Section 2.790 of the NRC's "Rules of Practice", Part 2, Title 10, Code of Federal' Regulations, a copy of this letter and its enclosure will be placed in the NRC Public Document Room. Should you have any questions concerning this letter, please contact us. I Sincerely, .;-K J. Nelson Grace Regional Administrator

Enclosure:

NRC Inspection Report w/ Figures 1, 2, 3, 4, & A cc w/ enc 1: T. L. McConnell, Station Manager Senior Resident Inspector - Catawba l l #767A%/W /P. t z

UNITED STATES [9 t e s,#%, NUCLE AD REGULATORY COMMIS$10N CEGicNli d I^ 101 MA:!lETTA STRE ET, N W. f ATLANTA GEORGI A 30323 %,..v p# Report Nos.: 50-369/87-22 and 50-370/87-22 Licensee: Duke Power Company 422 South Church Street Charlotte, NC 28242 Docket Nos.: 50-369 and 50-370 License Nos.: NPF-9 and NPF-17 Facility Name: McGuire 1 and 2 Inspection Conducted: July 7-10 and July 29-30, 1987 Inspectors: D. S. Hood W. T. Orders A. B. Ruff A. L. Toalston Contributing Inspectors: K. R. Naidu P. K. Van Doorn Censultant: G. J. Toman, Franklin Research Center Approved by: [ 4 II:s/f') .Peeb)ef,TeamLfft6er Date Sfgned . vision W Reactor W ojects

SUMMARY

Scope: This special announced augmented inspection was conducted for the McGuire roactor trip breaker (RTB) failure of July 2,1987. The inspection team was charged with investigation of the RTB failure in order to determine the root cause and what corrective actions are needed. The areas inspected included: sequence of events, description of failure, breaker evaluation and testing, human factor aspects, safety significance, equipment history and j regulatory compliance and conclusions. Results: In the areas inspected, no violations or deviations were identified. l l l l n v m ss., s a i, r t> orQturut a v pp

REPORT DETAILS 1. Persons Contacted Licensee Employees &*T. McConnell, Plant Manager

• B. Travis, Superintendent of Operations
• 0. Rains, Superintent nt of Maintenance "B. Hamilton, Superintendent of Technical Services "N, McCraw, Compliance Engineer "M. Sample, Superintendent of Integrated Scheduling

&*N. Atherton, Compliance "G. Gilbert, Operations Engineer "J. Snyder, Performance Engineer

• W. Reeside, Operations Engineer
• E. Estep, Project Engineer "R. Blake, MNS/0PS
• R. C. Futrell, Nuclear Safety Assurance, Manager, GO
• C.

S. Geerken, MNS/MSRG

• A.

R. Sipe, MNS/MSRG

• A. Rose, CNS/ Transmission
• K. O. Leuschner, Transmission (GSS)/McGuire
• G. W. Hallman, Nuclear Maintenance, GO
• R. Wilkinson, Nuclear Rel. Assurance, GO "G. A. Copp, Maintenance, Planning Engineer, MNS

&*J. E. Thomas, Design Engineering

• 0. W. Murdock, Design Engineering
• R. L. Gill, NPD, NTS, Licensing
• R. F. Banner, HPD, MNS, Compliance
• R. B. White, Jr., NPD, MNS, IAE Engineer
• W. H. Messar, Nuclear Maintenance, Mnt, Engineer, GO
• L. S. C.arles 0/E, Technical Specification, GO
• A. F. Batts, QA, McGuire "M. D. McIntosh, NPD/ General Office
• T. Cline, MNS/IAE

&K. Wilkinson, Transmission-McGuire &R, G. Bledsoe, Transmission Department Westinghouse Employees

• John M. Roth, McGuire Station Representative A. K. Deb, RCS Engineering G. C. Steinel, RCS Engineering B. J. Metro, Nuclear Safety Other licensee employees contacted included transmission circuit breaker technicians, operators, quality control technicians, and office personnel.

s a 2 Resident Inspectors

• W. T. Orders

&"S. F. Guenther "O. J. Nelson

• Attended exit interview July 10

& Attended exit interview July 29 2. Exit Interview The inspection scope and findings were summarized on July 10 and July 29, 1987, with those persons indicated in paragraph 1 above. The licensee did not identify as proprietary any of the information reviewed by the inspectors during the course of their inspection. No dissenting comments were received from the licensee. 3. Unresolved Items No unresolved items were identified. 4. Augmented Inspection Team (AIT) Activities The NRC was promptly informed of the Emergency Safety Feature (ESF) actuation and reactor trip breaker (RTB) failure at 1:12 a.m. on July 3. The licensee continued prompt corrective actions to assure the safety of the plant and in the process localized the problem to the specific RTB at 4:00 a.m. A discussion between the Region, the Senior Resident Inspector and the licensee on. July 3 and 6 concluded that a quarantine of the RTB was warranted. The licensee finished successful testing of the installed Unit 2 RTBs at 2:10 p.m. en.'iuly 3 and startup testing commenced at 2:20 p.m. The AIT arrived at the site on July 7 at 9:00 a.m. and began a cooperative evaluation of the failed breaker with the licensee and the RTB vendor (Westinghouse Electric Corp.). The AIT exited on July 10. I 5. Overview During the evening of July 2, 1987, reactor control rod drop timing tests were being conducted on McGuire Unit 2 as part of startup testing following the refueling outage. After several rod banks had been successfully tested, the B RTB failed to open during a manually initiated trip from the main control panel. The coil of the shunt trio attachment (STA) had overheated, shorted, and opened; and the fuse to the STA circuity had opened. Three operators in the control room each stated that the RTB indicating lights (red and green) showed that the RTB had tripped. Subsequent observation at the RTB and the events recorder printout both l indicated that it had not opened.

3 The f ailed RTB was found.when a technician observed that the B RTB was still closed and there was smoke in the area. pressing the manual trip plate at the RTB did not open it. The RTB was subsequently opened by manipulating the manual spring charging handle. The RTB was then removed i from its cubicle. The licensee continued attempts to isollte the problem and to assure that the plant was in a known safe condition. The failed RTB was taken to the test room and placed on a bench. The inspection of the RTB showed that the STA coil was burned and had opened. The RTB was then electrically closed, and successfully opened by de-energizing the undervoltage trip attachment (UVTA). The RTB was again electrically closed, but mechanically bound and failed to trip during a manual trip force test using a push force gauge. Then continuing to duplicate the original cenario, the technician manipulated the -anual spring charging handle and the breaker opened. One further test of the the RTB for opening was successful. The RTB was then quarantined. A McGuire Unit 1 bypass RTB was used to replace the failed breaker. This RTB and the associated circuity wis tested in place satisfactorily and no abnormal conditions were found. The other three breakers on Unit 2 were then tested in place satisfactorily. The licensee determined that the problem was isolated to the one RTB and the other RTBs were operable. The red drop tests were then resumed. A methodical procedure for inspecting the breaker for the mechanical binding and erroneous position indication was developed by representatives from Duke Power Company, the RTB vendor and the AIT. The mechanical inspection revealed a looseness of the center pole lever of the pole shaft and in particular a broken weld between the pole shaf t and center pole lever. This broken weld and a general looseness contributed to an approximately 5 degrees of angular rotation and a skewing of the the main drive link and its roller with respect to the closing cam and its mounting frame. It was hypothesized that the skewing of the main drive link such that it could centact the mounting frame could result in jamming of the RTB in the closed position. The main drive link is the only substantial component that is oriented such that its jamming could hold the breaker closed; therefore, this was tentatively concluded to be the probable cause of the breaker's failure to open and the focus of further efforts of the investigation. The electrical inspection of the internal breaker wiring was completed with no abnormalities found. Subsequent attempts to duplicate the failure to trip were minimally successful. All parties agreed that dismantling of the breaker and further testing should be conducted under laboratory conditic?s. Tha breaker has been shipped to the vendor's laboratory where testing is to begin on August 24, 1987. There the causes of failure will be further investigated under the direction of the licensee and with participation by the NRC.

i 4 6. Description of Event On July 2, 1987, McGuire Unit 2 was in mode 3 (Hot Standby) undergoing control rod drop timing tests. At 11:41 p.m., B RTB, one of two in series, failed to open when a planned, test induced, manual reactor trio was initiated in the control room. Control room RTB position status lights erroneously indicated that both RTBs had opened. (Green Lights). At this point in time, the reactor control rods were inserted and there was no indication of a problem. At 11: 43 p.m., both RTBs were given a manual demand to close from the control room in preparation for the continuance of rod timing tests. RTB A closed and B remained clcsed. Control room indication was normal; both RTBs indicated closed. (Red Lights). As the reactor operator (RO) attempted to withdraw control red bank E, a demand counter (rod position) problem was encountered and the operator reinserted the bank of rods. An electronics technician (IAE) was dispatched to the rod control logic cabinets, which are in the vicinity of the reactor trip breakers, to troubleshoot the problem. The IAE technician informed the Shift Supervisor (55) of smoke in the vicinity of the reactor trip breakers. The 55 then directed the RO to initiate a manual reactor trip. At 11: 46 p.m., the control room indication reflected both RTBs open. At the RTB cubicle, smoke was found coming from the B breaker which was found to be closed. The operator at the RTB informed the RO that the B RTB was closed. The R0s in the control room did not verify that the breaker open indication was still present. Attempts to open the breaker locally by depressing the trip button on the breaker were unsuccessful. Manipulation of the manual spring charging ratchet handle on the front of the breaker resulted in the breaker opening at 11:58 p.m., some 12 minutes af ter the reactor trip had been initiated from the control room. When the breaker opened, it resulted in a train B feedwater isolation since the operators in the control room were not expecting the reactor trip signal generated by the RTB opening and had not blocked the feedwater initiation signal as is normally the case during testing. The feedwater isolation was of no major consequence and was quickly recovered. During troubleshooting of the breaker on the morning of July 3,1987, it was determined that the breaker had mechanically bound. Technicians at that time were successful in reproducing the binding which had led to the failure to trip. The binding had resulted in the electrical failure of the STA coil which remains energized subsequent to the initiation of a reactor trip signal until the breaker actually opens. The overheated coil was the source of the smoke.

f f Ultimately, a unit 1 bypass breaker was placed in the Unit 2 8 breater cubicle, all four Unit 2 breakers (2 trip breakers, 2 bypass breakers) were functionally tested satisf actorily, and rod drop timing tests.ere resumed. No further difficulties were encountered. Extensive testing perfor.ted on the B breaker during the week of July 7-10, 1987, failed to reproduce the binding which had been previously observed. 7. Sequence of Events July 2, 1987 Manual reactor trip initiated in control room at 11: 41 p.m. completion of testing control rod bank D Reactor trip breaker A opens Reactor trip breaker B remains closed Status lights in control room indicate both A and B reactor trip breakers open (operator statements) Reactor trip breakers A and B given close signal to 11: 43 p.m. begin testing of control rod bank E Reactor trip breaker A closes Reactor trip breaker B remains closed Status lights in control room indicate both breakers closed (operator statements) Reactor operator begins withdrawing control rod bank E 11: 44 p.m. for testing, step counter problem observed RO reinserted control rods IAE technician dispatched to control rod logic 11: 44 < cabinets to troubleshoot step counter problem. smells smoke... informs Shift Supervisor. (SS) Reactor operator initiates manual reactor trip as 11:46 p.m. directed by SS Both breakers indicate open in Control Room (multiple operator statements) IAE technician and one reactor operator respond to 11: 46 + area of reactor trip breakers, find B breaker still closed and smoking Attempts to open breaker by depressing trip button on braaker futile Manipulation of manual spring charging ratchet handle 11:58 results in breaker opening Train B feedwater isolation initiates due to breaker opening NRC notified of ESF actuation (feedwater isolation) July 3, 1:12 a.m. and breaker failure Testing of B breaker in shop reveals STA coil July 3, 4: 00 a.m. j overheated; mechanical binding reproduced l l l I

6 Unit 1 bypass breaker B placed in Unit 2 reactor trio July 3, 2:10 p.m. breaker B cubicle. Both reactor trip breakers and both bypass breakers tested satisfactorily Rod drop timing testing recommenced July 3, 2:20 p.m. 8. Discussion of Equipment Failure The operator tripped the 2A and 2B RTBs from the manual control switches at 11:46 p.m. on July 2. By observation of the green light, the operator received indication that both RTB had opened. The source of the smoke wa<, traced to the 28 RTB, which was found to be closed, counter to the previously indicated status on the control board in the control room. (0 pen indication was not verified at this time). Attempts to open the 2B RTB by pushing the manual trip button on the RTB were unsuccessful. Af ter approximately 12 minutes, the RTB opened when the lever for manual ch *ging of the closing spring was pulled approximately five times. The source of the smoke was the STA coil that had shorted to ground and open circuited. The STA coil is normally deenergized when the circuit breaker opens and, therefore, is energized only for approximately 100 milliseconds, from the time the trip signal is received until the circuit breaker opens. However, during the period following the 0 bank rod drop tests until the E bank tests were started, the coil was energized (approximately 2 minutes); and again, after the tripping of the RTBs subsequent to detection of smoke, it was again energized until it burned open. Two concerns arose from the failure of the 2B RTB to open. First, what was the cause of the jamming of the RTB and, second, what was the source of the erroneous open indication in the control room? Evaivation of alarn printer and events recorder data that receive status indications from the same set of 28 RTB auxiliary switches correctly indicated that the RTB did not open after the D rod drop tests or subsequent to the detection of smoke. When the 2B RTB finally opened, the alarm printer and events recorded the position of the RTB correctly. It should be noted that illumination of the control room green light fer the 28 RTB failure was observed by three different operators. During subsequent interviews, these operators were sure that they had indeed observed illumination of the green light. 9. Discussion of Evaluation and Testino of the Unit 2 RTB Overview of Test Proaram On July 7 through 9, 1987, inspection and testing of the 2B RTB were performed. The breaker mechanism was not dismantled during these efforts. The goal of this testing was to determine the cause of the 2B RTB failure to open and to determine if the erroneous open indication occurred due to improper operation of the auxiliary switches, or failure of or improper connections in the RTBs wiring harness. Between the start of the testing - ~,,,.

7 on July 7 and the time of the f ailure on July 2, a Duke Power circuit breaker technician had replicated the f ailure to open by slowly pushing the manual trip pin on the trip eechanism. Prior to this, the RTB tripped correctly by operation of the undervoltage trip attachment. The j amm.ing of the RTB was again cleared by partially charging the closing springs by means of the manual charging lever. One further attempt at replicating the failure to open was unsuccessful. Further tests were not attemptec until the July 7 tests done in conjunction with the manuf acturer and observed by the NRC. The RTBs are Westinghouse model 05-416 drawout circuit breakers. These are stored-enwrgy breakers in which the closing force is supplied by a set of springs that are charged by means of a motor and a ratchet system. Af ter release of the trip latch, the contacts are opened by means of another spring. The tests and inspections that were to be performed on July 7 and 8 were: 1. Determination of the electrical status of the internal control circuits within the RTB including shorting between the circuit for the green light indication and other circuits in the RTB; and the condition of the STA coil. 2. Evaluation of the condition of the circuit breaker operating mechanism, and the auxiliary switch and associated linkages without operating the circuit breaker. 3. Evaluation of the circuit breaker after it was closed. 4. Simulating a trip condition by pushing on the pin for the manual trip in an attempt to replicate the jamming condition. During the closing and opening (if any), the close and open indications from the auxiliary switch would be monitored by a digital recording t oscilloscope to verify proper auxiliary switch operation. Thereafter, further efforts would be taken as needed. The results of the electrical tests with a digital ohmmeter were as follows: No shorting or abnormal conditions were detected between the green light circuit and any other control circuit in the RTB. The STA coil was open circuited and the positive leg was shorted to the RTB chassis with a 2.38-ohm resistance (a properly working coil ir insulated from the chassis). The coil was observed to be charred. Subsequent testing on June 9 with a meggar and by applying 125 Vdc to the shunt coil reconfirmed that there were no short circuits between the green light circuit and the remainder of the control circuits. I The undervoltage trip attachment (UVTA) was confirmed as working properly.

8 The visual inspection of the mechanism revealed that the weld between the pole shaf t (Item 8 in Figure 1) and the center pole lever (Item 9) was completely cracked. Figure 2 shows an overall view of the main mechanism. In this figure, the pole shaf t is Item 21 and the center pole lever has been labeled "A". This weld is critical to correct breaker operation in that the bulk closing force exerted by the closing cam is transmitted by this lever to the pole shaf t which causes the main contacts to close. Although the weld was broken, motive force was still being supplied to the pole shaft from the main drive link (Item 27 in Figure 2) by means of a parallel lever, the anti-bounce lever (which is not shown in the figures). The anti-bounce lever is welded to the pole shaf t and connected to the upper pin of the main drive link that connects to the center pole lever. With the cracked weld on the center pole lever and the good weld on the anti-bounce lever, the center pole lever will skew to one side of the RTB and cause the bottom of the main drive link to cant towards the latch side of the RTB when the circuit breaker is closed. Evaluation of the close cam (! tem 5 in Figure 2) detected flattened (mushroomed) spots on the laminations, especially in the area of the stop roller (Item 6 in Figure 2). These flattened areas indicated that the roller on the main drive link was riding well of f of center towards the trip latch side of the RT8. Visual inspection of the auxiliary switches (see Figure 3) indicated no apparent problems. There are thete auxiliary switches stacked upon one another. The linkages between the pole shaf t and the auxiliary switches were found to be correctly connected and set up for proper operation. No loose or misaligned contacts were observed. Following the visual inspection, the RTB closing springs were charged and the RTB was closed by energization of the electrical closing circuit. Correct operation of the auxiliary switch was observed. The green light circuit opened; then, the shunt coil circuit closed. With the circuit breaker closed, the center pole lever was observed to be canted at the bottom towards the charging motor side of the RTB, which, in turn, caused the bottom of the main drive link to be canted towards the trip shaft side of the RTB by approximately 1/4 in. Subsequent observation showed that the roller on the main drive link was forcing the roller constraining link (! tem 3, Figure 2) to the trip shaf t side of the RTB, which, in turn, was exerting a similar force on the trip latch (Item 4, Figure 2). In an attempt to replicate jamming the manual trip pin was slowly pushed to see if the circuit breaker would jam. It opened i properly. Correct auxiliary switch operation occurred with a 9 milliseconi delay between shunt coil circuit opening and closing of the green light circuit. The closing of the green light and event recorder l circuit were nearly simultaneous. l l l 1

9 Fourteen more attempts were made to allow the circuit breal.er to jam. It did not. On July 9, purposeful, but nondamaging attempts of manipulating l mechanism components were made to cause the circuit breaker to jam but none were successful. Weld Inspections The inspector visually inspe:ted welds between the levers and the pole shaft on the defective breaker and the two spare breakers. The welds were a single pass for approximately 180' of each lever (six per breaker). Westinghouse standards and documentation were not av411able for this inspection, and in addition the surfaces were apparently plated after welding, inhibiting a proper visual inspection of the welds. The welds were therefore inspected for obvious defects and general appearance. The weld of the center lever on the defective breaker was separated between the weld and the lever for approximately one-half the weld length. At the end of the separation a crack had propagated across the weld bead thus allowing the lever and pole shaf t to move independently to some degree. No obvious defects were identified in the other welds of the defective breaker; however, portions of the other welds appeared to be of poor quality exhibiting a high probability of poor fusion or lack of fusion of the welds to the base material. This was evidenced by a lack of characteristic weld bead ripple, little evidence of base metal melting and notches at the edge of the weld beads. The welds on the newer vintage spare breakers appeartd to be of better quality. The welds were larger in bead size with only a few small areas where lack of fusion might be questioned and some porosity was noted at several weld starts and stops. A licensee welding inspector also visually inspected the welds. His opinion paralleled that of the NRC inspector. It is recommended that a failure analysis be conducted of the failed weld with NRC observation and that the quality control program for these welds be reviewed. Consideration should be given to inspection of these welds, especially on older vintage breakers, utilizing a qualified welding inspector. Green Light Anomaly Investigation Three operators in the control room stated that the two RTB open green lights on the control pinel had come on, indicating that both RT8s had tripped, even though one had not. The manual control switch is a three position handle-operated, spring-return type switch located on the vertical panel in the main control room. The handle's normal position is top dead center. It is turned to the left about 30' (counterclockwise) to trip the reactor trip breakers, and to the right about 30' to close the associated reactor trip breaker. When released it returns to the coeter position. The two switches are on the same vertical plane about three feet apart. Each control switch will trip its reactor trip breaker (this

10 is a McGuire specific design) and the bypass breaker if it is racLee-ia 10 the operate position. The operator is trained to operate octh :cv rol switches simultaneously, one with each hand. Directly below each c:. 31 switch are red and green indicating lights. The operator is tra'aec :o interpret the lights to indicate the actual position of the rea: tor trio breaker contacts. The red light indicatgs that the associated rea: tor tr.o breaker is closed, and the green light indicates it is open. In reality, this is not always true and may raise some human factor conceres as discussed elsewhere in this report. Both the red and green lights are energized from the same 125V de source. In series with the green light is an auxiliary "b" contact. An 'a' contact follows the breakers movement and a 'b' contact acts opposite to the breaker. (See Figure A). The auxiliary contacts are mechanically operated by linkage from the breaker pole shaft. The moving main contact blades are connected to this same shaf t. Thus, the auxiliary switch position reflects the actVal position of the main contact blades. Each auxiliary switch is a rotary switch with wiping type contacts. A fiele measurement made as the main contact blades were slowly opened showed that a main centact blade opening of 5/8 of an inch was required before the auxiliary "o" contacts began to close. The red light is in series with another contact on the same auxiliary switch as the green light. A field measurement made as the main contact blades were slowly opened showed an opening of 1/16 of an inch before the "a" auxiliary switch contact began to open. Thus, the "a" contact began to open before the "b" contact began to close. Oscilloscope measure eats during the testing confirred tnis. The STA coil is in series with the red ligFt and the "a" auxiliary : *ta:* (See Figure A). The red light has a dual function of indicating tae continuity of the shunt trip circuit as well as the breaker status. Wht'e the RTB is closed, a small current flows through the red light and tae shunt trip coil. When the manual control switch is rotated to the trip position, a contact (CS-Trip, top of Figure A) closes to bypass the red light and energize the STA coil. Release of the manual control swit:n to l the center position opens this contact. 1 Another contact in the control switch (CS-Trip, bottom of Fig.re A) interrupts current to the UVTA and also to the (S) relay. When the I control switch is released, the design is such that the UVTA cirt it l remains deenergized, and thus the (S) relay remains deenergized until tte RTB is closed. When the (S) relay is deenergized, it closes contact (S) (Top of Figure A). Closure of this contact energizes the STA and also bypasses the red light. If the breaker fails to open, the STA remains energized. The STA is not designed to carry full current for a Iceg period of time and will buri out if the breaker's auxiliary contact does i not open to interrupt the current. In this case, the RTB did not esen, the STA coil burned out and shorted to the frame, and the fuse blew. l l 1 J

11 The initiation of the event recorder is by a 'b' contact on a differeat auxiliary switch then the auxiliary switch which controls the red a d green lights. However, this switch is mechanically ganged to the ct e, and both operate from the same pole shaft. Visual inspection revea Nd ao abnormality-of the ganging mechanism. Field measurements, confi mec by the oscillograph plots, showed that the contact initiations to the gresn light and to the event recorder were simultaneous. Thus, the green lig>t indication at the control panel should have been consistent with *.he event recorder printout. The RTB failed to open, as evidenced by the events recorder, the alarm typer, and the STA coil burnup. Yet, three different operators state that the green light was on even though the RTB was still closed. This would suggest that something was wrong with the wiring to the green light. = However, the green light was operating properly during every other trip prior to the event. It was also tested to be operating properly after tre failed RTB was replaced by another RTB, without any wiring changes. No measurements or inspections at the failed RTB explained the ancmaly. Comprehensive measurements and checking of the circuitry external to the RTB could not be safely made with the reactor in operation.

However, drawings are being checked &-d an extensive checkout of the actual wiring is being pursued by Duke Power Company personnel as conditions permit.

Discussion of Results of Testing While the jamming of the circuit breaker could not be replicated duMeg the testing of July 7 to 9,1987, the observation of the failed weld and the associated misalignment of the main drive link, coupled with the near marks on the close cam and frame, st ongly suggested that this ccmbiratien was responsiblo for the jamming of the circuit breaker. The testing of the electrical control circuits and the auxiliary switch showed no abnormal condition that would result in err tous operation of the green open position indication in the control rt..a. The shorting of the STA coil to ground should not have been the cause of the erroneous indicction because the operators stated that, upon the call for opening of the RTBs following the D rod tests, there was an immediate green light. At that time, the STA coil probably had not yet failed and the B RTB had not properly opened. The electrical control circuits within the RTB were found to be workirg correctly. Therefore, the incorrect control status appears to be ccused by a problem in the circuits between the reactor trip breaker cubicle and the control room indicating light and control switches. Duke Power is evaluating these circuits. However, only limited testing can be performed with the unit at power.

12 9. Hu-an Factor Aspects The principle human factor concern disclosed by the event was the f a 'u e by the operatur to discern that one of the RTBs had failed to open. If the green indicating light had indeed come on even though the RTS d'.d not l open, this would indicate a wiring deficiency, and not a human factoe l deficiency. This is still under investigation. If the green light d4d not come on, as designed, and the operator failed to notice that it had not, or thought it had come on when indeed it had not, this could indicate a human factor deficiency. l If an operator is performing a series of consecutive tests every f e., minutes, with each test requiring a series of steps and checks, and the checks are successful time after time as they were in this case, it can be e postulated that there would be a possibility of missing one of the checks, particularly it the checks were not individually recorded and that the check was merely confirming what the previous checks had already shown. In accordance with procedures and training, in the event of a reactor trip, the operator first observes rod insertion on the rod position indication display, next, that the reactor flux has decreased, and next that the manual reactor trip annunciator is on and that the green breaker indication lights are lit. Further, there are two green lights to observe (one for each RTB) about 3 feet apart, one to each side of where tre operator was standing to manually trip the RTBs. The manual reactor trip annunciator is initated by the manual reactor trip switch. l Another possible human factor concern is that the circuit design ca.:ses the red light to go out before the RTB physically opens, and even if tne l RTB fails to open. First the manual trip switch centacts bypass (sorts) the red light, and then the $ relay contacts bypass the red light. This falsely indicates to the operator that the RTB has opened, when indeed it j may not have as in this case. Consideration should be given to l redesigning the circuitry so that the red light truly indicates the closed [ i position of the RTB. A dif ferent color light could be added to eenitor l the continuity of the shunt trip circuitry. l Safety Significance and Radiological Consequences The reactor trip breaker and its associated switchgear provides the sefaty r related function of interruption of power to the control red de!ve i mechanisms (CRDM) 1a response to an automatic reactor trip signal or manual command. Interruption of the power causes the magnetic latches of t the CRDMs to be denergized and the fuil-length control rods to f all by i gravity into the core, decreasing reactivity to shut down the reactor. i Two main reactor trip breakers, each receiving separate control power and an independent reactor protection system trip signal, are connected in i l l l I \\

13 series. Thus, failure of a single breaker to open does not prevert a reactor trip; failure of both RT8s to open in response to an ant'c4ca:ec l operational occurrence could result in an anticipated transient fm. t scram (ATWS). ATWS mitigation actuation circuitry required b 10 CFR 50.62 is scheduled for installation during the 1988 refuetin; outages of each McGuire unit. The specific event had small safety significance and no radiological consequences because the reactor was in hot standby with all reds inserted, and one of the RTBs operated properly. The event would have 5ad more safety significance if a significant common cause f ailure mechanism for this vintage of RTB is discovered during the continaation of the analysis and investigation. No common cause f ailure has been discovered to date. An erroneous green light indicating that the breaker is open when it is closed can reduce the probability of detection or delay timely corrective action by the reactor operator. This specific green light anomaly, ard the failure of the operator to detect that the RTB had failed to trip had no safety significance in this instance because the problem was discovered I through the smoking coil. If the smoke had not been detected, the ocening of the coil, or the absence of voltage across the coil due to the blown fuse, should have been detected by the absence of the red indicatirg light. Thus, the event discloses that the actual position of the RT35 should be checked prior to resetting them and positive indication of breaker position is highly desirable. 11. Eeviement History and Review of Licensee and NRC's Recuirements The four RTBs (main and bypass) for each McGuire unit are Westingreuse type 05-416 circuit breakers of pre-1984 vintage. The breakers were delivered to the site in 1973 and are some of the earliest of this vintage breaker. The licensee estimates that McGuire RTB B had experienced about 3000 cycles as of July 2, 1987. The actual number of cycles could not be determined because counters were not initially included on the breakers. j The counters are not required to remain functional and the counter for RTB j B was not operating. Many of the accrued cycles are associated with periodic surveillance / maintenance of the breakers which is performed on ) I each McGuire unit monthly and at 6 month intervals in accordance with Unit 2 License Condition 2.C(12)c, Table 1. I Each breaker accumulates 250-300 cycles per year due to testing aed maintenance af the breaker and associated equi;Weent. Although no overall limitation on the numoe-of cycles is specified in the Westinghouse Maintenance Program Manual for the 05-416 breakers, replacement recommendations are included for certain components. A recomended service life of about 2500 operations is specified for both the Uv1A and STA. Certain parts such as breaker arcing and main contacts are specified I for replacement based upon specified dimsnsional checks. i I

14 The inspectors reviewed the following document sections and procecu'es: 1 a. Technical Specifications Section 3/4.3.1 ( b. License Amendment No. 2 for McGuire Unit 2, Reactor Trip Breaker i Section c. Maintenance and Periodic Test Procedures: (1) MP/0/A/2001/06', Westinghouse 05-416 Air Circuit Inspecticn and 4 Maintenance l (2) PT/2/A/4600/56, Manual Reactor Trip Function Test i >(3) PT/0/A/4601/07A, Response Time Testing of Reactor Trip Breakers RTA and/or BYB (4) PT/0/A/4601/078, Resronse Time. Testing of Reactor Trip Breakers RTB and/or BYA j (5) PT/0/A/4601/088, Solid State Protection System (5505L Train "B" Periodic Test With NC System Pressure Greater Then 1955 psig (6) PT/0/A/4601/09, Solid State Protection System (SSPS) Train "B" i Periodic Test With NC System Pressure Less Than 1955 psig. The procedures implement the Technical Specification and licensing i amendment (a. and b. above). These documents require functional tests of i the VVTA the STA manual reactor trip from the control room, the response I time testing of the UV/ breaker on signal from the reactor protection system and the periodic surveillance /maintenanse of the reactor trip (RT) ( and bypass (BY) breakers. The procedure appeared to be complete, j j accurate, and when applicable, the vendor's requirements (Westinghouse 1 Program Manual MPM-WOGRTS05416-01) were incorporated. A review of recent work requests for breaker, S/N 24Y9850B4, that failed to open was made. The following tabulation shows.the various locations of l this breaker and the tests and maintenance that were performed over tne r j past 18 months. Breaker Location Procedure From c. Above Date l

• 2BYA 4

1T/M i 2BYA 1 01/86 i l 28YA 4 01/86 28YA 1 06/86 l 2BYA 4 06/86 [ j 28YA 1 12/86 l 28YA 4 12/86

1. 2RTB 4

12/86 l 2RTB 4 12/86 2RTB 2&5 01/87 2RTB 5 02/87 t 2RTB 5 03/87 2RTB 5 04/87 2RTO 5 05/87 l 2RTB 2&6 06/87 Breaker failed July 2 L 1

15

• Unit 2 reactor trip bypass breaker Train A

1. Unit 2 reactor trip breaker Train B During these recent inspections, no major anomalies were recc-:ec f:-

the RTB that failed. These inspections are performed for tese tne Catawba and McGuire stations by the Duke Power Company TransMssien ~ Department personnel. These craftsmen are excepticeally knowledgeable of the breaker's construction and working mecha*is 5 The inspections include an examination for cracked or broken welcs None were recorded and in a discussion with these craf tsmen, they stated that if a broken weld had existed during the last inspection period, they would have seen it and it would have been noted and corrected. However, welds with partial cracks would have been i difficult to notice. The time response testing of the breaker over this time period showed that the breaker was functioning as designed and no adverse treecs were evident. The opening times were random between 60 and 93 milliseconds which is well within the acceptance criterion of 150 milliseconds. In order to assure that no major malfunctions had occurred, the inspectors reviewed the overall maintenance records en the failed RTB and determined the fo11 ewing. Breaker Serial No. SN24Y9350 B-4 Type 05416 Initially installed as Bypass breaker A in Unit 2 i Designated CR 9 Initial Inspection July 26, 1979. A. March 26, 1983 As a result of NRC Bulletin 83-02, Duke inspected and determined that the clearance between the operating arm and the moving i i plunger of the Undervoltage Trip Attachment (UVTA) was.037 i inches. The trip shaf t was removed for lubrication and it was observed that one end of the shaft bad been ground on 4 ;"incing machine. B. April 16-18, 1983 The trip shaf t was replaced and was coated with Poxy16be aed molybdenium disulphide and alchohol mixture. Subsequently, the breaker was operated for 25 cycles and the UVTA operability was successfully verified. j 7-26-83 Regular preventive maintenance (RPM) was performed and the breaker was tested on the test bench. No i abnormalities wera documented. i l i

] s 16 y 1-7-84 RPM was performed, followed by routine bench tests. No abnormalities were documented. ~ 7-18-84 New trip latches were installed and the operability of tne breaker was verified. 1 1-7-85 RPM was performed follcwed b) routine bench tests. No ) abnormalities were documented. 7-10-85 RPM was performed followed by routine bench tests. No abnormalities were ducumented. 1-2-86 RPM was performed and counters were installed on the breakers. The reading was 99749. i 6-24-86 RPM was performed and the breaker was operated 95 times. 12-18-86 RPM was performed. Triplatch overlap adjustments were made and locAtite was applied on the threads. 7-2-87 The breaker did not open on signal from the control room subsequently, on July 8,1987, it was determined j that mechanical bindit1 occurred due to a weld separation. The weld ccnnecting the pole shaf t and tr.c canter pole shaft separated. l Conclusion The inspectors found the testint), inspections, and mainterance of the breakers to be satisfactory, and considers that NR; and the licensee's requirements were mat. 12. Evaluation of Failure to Trip of McGuire Reactor Trip Breaker No. 2AY9349B Durino Periodic Testino of UVTA on July 28, 1987 Inspection conducted on July.9 and 30, 1987 Location: McGuire and Catawba Nuclear Plants Personnel Involved: T. Peebles,' NRC Region !! D. Hood, NRC K. Naidu, NRC J J. Thomas, Duke Power R. Bledson, Duke Power G. Toman, Franklin Research Center l

17 Background

On July 28, 1987, during periodic maintenance and inspection cf sne McGuire Unit 1 RTBs, RTB No. 24Y98498 f ailed to trip during post-t a<e' margin testing of the associated UVTA. The RTB failed to trip because t e UVTA did not cause the trip shaft to rotate far enough to release the trip latch. The RTB was laf t in the "f ailed to trip" condition until arri,a1 of members of the NRC AIT that had previously evaluated the RTB fatture of July 2, 1987 at McGuire. Discussion of Findings Subsequent to the Salem ATWS events in February 1983, which were due to problems with the UVTAs of Westinghouse 0B-50 RTB, problems were recognized on the UVTAs of the Westinghouse 05-416 RTB at McGuire. The evaluation of these problems led to development of acceptance criteria for the VVTA and its interf ace with the RTB trip shaf t. No criteria were necessary for the STA. These acceptance criteria for the six month inspections are: 1. The UVTA must have a static output force of at least 3 lb. (A new UVTA will produce approximately 6 lb.). 2. The UVTA must operate (trip the circuit breaker) when the applied voltage is reduced to no more than 28.8 V and no less than 14.4 V. 3. .The force required to trip the breaker via the trip shaft shall be no more than 2 lo (normal values are generally 1 to 1 1/4 lb). 4. In the reset and energized state, with the RTB closed, a gao of at least 0.030 in must exist between the trip lever of the UVTA cad the associated pin on the trip sha t. 5. The UVTA must be able to trip its RTB with a 0.070 inch shim in place that Ilmits full travel of the UVTA trip lever. This is a go/ne go test that proves that margin exists in the travel of the UVTA trip lever such that a UVTA is assured of causing the trip shaft to rotate far enough to release the RTB trip latch. 6. The trip shaf t adjustment must be set by turning its screw cleckwise until the trip latch releases (trips the RTB) and then it is backed off four complete turns. The trip shaft adjustment determines the distance that the trip shaf t will rotate beyond the trip point when the trip shaft resets. If this 56tting is backed off too far, a larger rotation of the trip shaft will be required before the breaker trips. If *he trip shaf t adjustment is too little, the trip latch could fail to hold and the circuit breaker could trip spuriously due to vibration or at the time of closing (trip frea).

18 The f ailure of the RTB on July 28, 1987, at McGuire is related to the ::st travel margin test of Item 5 above. With the 0.070 in shim in place, we UVTA fa11ec to trip the RTB when the UVTA was deenergized. The vi nal inspection of the UVTA and RTB by the AIT verified Duke Poner's observation that a gap existed between the trip lever of the UVTA ar.d tne trip shaf t pin, indicating that the UVTA trip lever had struck the t ip sha't pin and has caused the trip shaft to rotate. However, tne combination of the travel and force of the UVTA trip lever was insufficient to cause adequatt rotation of the trip shaf t to trip the RTB, Duke Power personnel proceeded to evaluate the actual margin of UVTA travel under static actuation (slowly allowing the UVTA te operate rather than allowing normal rapid operation as wJuld occur when the VVTA was deenergized). The static trip point occurred with a shim of 0.019 inches, but did not occur with a shim of 0.031 inches. Subsequent to these tests, the RTB was tripped successfully three times with the 0.070 inches eim in place when the UVTA was deenergized. These results indicated that while eargin existed, the UVTA was at the borderline of the acceptance criterion. The force required at the trip shaft to trip the RTB was verified to be 11/4 lb. Duke Power personnel replaced the UVTA are demonstrated that the new UVTA mat the acceptance criteria with much improved tolerance margins. 13. Catawba RTB Inspection on July 30, 1987 ')uring the visit to McGuire on July 29, 1987, Duke Power personnel stated that cracks had been found in a Catawba RTB center pole lever to pole shaft weld. Since this is the weld that had failed on the RTB that did not trip at McGuire on July 3,1987, the leader of the AIT chost to visit Catawba and observe the RTB. During periodic testing of the Catawba RTBs subsequent to the July 3, 1987 McGuire failure, Duke personnel had observed that the weld of the center pole lever to pole shaf t on one of the RTBs appeared to be poorly made. Hcwever, the upper frame of the rib i obscured complete observation of the weld. The Duke personnel removed the upper frame and performed dye penetrant irspection of the weld and verified that the previously obscured portion of the pole shaf t weld was indeed cracked. It should be noted that removal of the upper frame is not part of normal inspectior procedures. The crack was not completely through and the weld was still holding the shaft to the lever. Duke Power will replace the pole shaf t and the cracked weld will be examined by an outside laboratory to determine if the crack was propagating during operation. Conclusions 1. The VVTA probleth at McGuire of July 28, 1987 is unrelated to the July 2, 1987 RTB failure. During the July 2 RTB failure, the RTS l trip latch had actually released and the main operating eechanism jammed. In the July 28, 1987 UVTA event, the UVTA did not cause i sufficient travel of the trip shaft to allow the trip latch to relaase. l \\ l t

19 2. The observation of the crack in the Catawba RTB is interesting in that it shows that detection of such cracks may be dif ficult since the frame of the RTB partially obscures the welds. A one-time partial disassembly of the RTBs may be necessary to verify t*.at ro manufacturing flaws exist in the welds. Subsequently, evaluations eay be limited to the readily observable portions of the welds ducirg I periodic inspections. 3. Of the above criteria, only the first two are applicable to the initial RTB failure to trip on July 2 and both of these criteria appear to have been met at the time of the event. During inspection of the 2B RTB subsequent to the event, the trip force required to trip the CB was measured twice and found to be 1 1/4 lb, Therefore, the 2-lb limit was met. During Duke Power's initial evaluation of i the event, when the technician actually succeeded in having the RTS jam a second time, the latch of the RTB was observed to hsve been

released, indicating that the trip shaft adjustment was net responsible for the problem.

Failure to meet the established tolerance criteria for the trip shaf t and UVTA was r.ot a cause of the 1 malfunction of the RTB but is to be rechecked before the RIB is 2 dismantled. r 14 Root Cause of the Event 1 The root cause of the RTB f ailure will be determined by the ongoing inspections and evaluations described below. I 15. Findings and Conclusiens j The broken weld betw' en the pole shaf t and center pole lever resulte in e unsymetrical forces at several points of the RTS mechanism includirg the l l interfaces between the closing cam and the main drive link roller. These i unsymmetrical forces occurred during both the closteg and tripping of the RTB causing uneven wearing of the parts and loosening at the various pivot i points. Manufacturing tolerances for this pre-1984 vintage 05-416 breaker, and normal wear resulting from the 3000 cycles of operation (estimated by r the licensee) undoubtedly also contributed to the twisting and lateral play cbserved in the mechanism. It is possible that this permitted the t roller to shift sufficiently off center of the closing cam to cause it to bind against the frame, or for the trip latch to bind against other parts. $1nce the inspection team was not able to observe such a binding, it was [ not popsible to determine the location or cause of binding. The fact that the bincirg phenomenon could not be repeated during the bench testing suggests that the original point at which the bindirg occurred had been worn to a smoother or more rounded surface by the subsequent tripping / closing of the breaker, The decrease in the measured force required to trip the breaker after the breaker was closed and tripped several.imes may not be significant, since this force would l appear to be more closely associated with the trip shaf t than wit.h the parts subject to binding due to the unsymmetrical forces. I c 1 l I r L_

20 The erroneous breaker open indication cannot be presently explai ec. Physical inspection confirmed the correct operation of the RTB an*ary switches. Internal RTB wiring checks for grounds and circuit coat'sv ty disclosed no problems with the RTB wiring other than the open STA :c.. Correct cperation of the indicating lights prior to and subsequent to the 6 vent indicates that any problem in the wiring external to the RTB rust te I unusual. It is possible that additional wiring cnecks, which Duke to.er Company is performing as conditions permit, will disclcse sucP a situation. If no circuitry problems are found to exist, it would seem reasonable to conclude that the operators were probably remembering a-d referring to the green light indication on the immediate previcus norn41 trip operation of RTB rather than at the time of the failed operation. Recommendations and Leno Term Actions a. Consideration should be given to redesigning the breaker indication circuits so that an indicating light clearly indicates the position of the TB movable blades. If tnis design had been, a light woJld [ have stayed on and the operator would have been alerted that the breaker had n

• opened even if the green light did incorrectly i

indicate that it had opened. An alternative is to check the RTB opened by observation at the breaker cubicle prior to re-closing the breaker. i 1 b. NRC Information Notice No. 87-35 issued July 30, 1987, stated tnat i enhanced visual inspection of the RTB is appropriate. This visual inspection should consist of inspection for cracked welds or ab9er al wear such that the main drive link roller could contact the support t frame at the close cam or bind against the cam. c. A one time inspection of the pole shaft welds to assure their adequacy is appropriate. The Catawba pole shaf t is being inspected for the NRC by an independent consultant. Recommendations for l inspection criteria are forth coming from the inspection. d. The licensee submitted Licensee Event Report (LER) 370/87-09 dated ) August 3, 1987, describing the failure of the RTB to open. The results and recommendations of the Duke Westinghouse - NRC inspection and testing of the breaker will be submitted in a supplement to that LER. i \\ I I l I

I GG@ Qu /1 rh 4{1-- - 4

i. sy ne.e,

2. comeCem l

3. Momer Cemeereining Link

4. PlweePin E Trip Laech

_O ~ S. Trip Shefa Laeching Surfeen \\./ i h ',,J g,, @

7. Trip Shaft

.a

s. Peseshen

s. Cen or Pese La.e, 4

. (. ). n ,.g 1; -y [.[' l)

10. Pete Lower Pin 5

\\x /

11. Moving Ceneoct Arm j

g l ,2.s. net.n.r, - - h,[!

• O

' (,'< 34 . 13. Moving Ceneoct Pives Pin g g i

14. Main Deine Link

~ i.,, j i s Mein nene, <o l i 'h l

18. Spring Amtesse Leech

( L ' j FL lamuteeing Link Aguseing .y fjA,.. (f)S i Seesd and Lectuent (-}.p ~ l

18. Imendesing Link g

d 1 1s. N sMe Fresse v f = -- - - r-- j

20. Herdened Leech Surfaces l

l t I, 1 Figure 1. Linkages of DS-416 Breaker Mechanien Shown with CB Open and Springs Charged (Source: Instructione for Low-Voltage Power I Carcuit Breakers Types DS aeed DSL Meetinghouse Electric Corp., Instruction Bulletin 33-790-IE)

9 q@ s j .O N ~ f \\ h NA%, O r N s N b g' 4 [ t y <h {, k 19 w 4 N ra 4 'il N / ) ,c 4q'e}m / a ~ /, y /, i d'gy IO [ k M# /- f @@@h@ a A. Center Pole Lever i

1. Shont Teep Device

10. Raschet V"heel

19. Reset Speer.g

2. Trop Shef a

11. Hood Pawl

20. Clasang Speeng Anchoe

3. Roller Coastseeneng t_enk

12. Deeve Plate

23. Fo8e Shaf t

4. Trep Lasch

13. Emespency Chaege Pawl

22. Motos e

5. Close Cam

14. Oscobsor

23. Emergency Change ifandie 4

6. Stop Rollee

15. Ceank Shatt

24. Mono, Ceank and ifandse 1

7. Speeng Release Latch

16. Eeneagency Charge Device

25. Moveng Contact Assendily

8. Speang Release Device

17. Crank Aem

26. Insulesang Link

9. Oscellator Pews

18. Clowns Spring

27. Maen Dreve Link s.,.,.7 pr.o,-.n.s p2,,.

ar ne.aer. n,. 6.- e %..,... e.

FIGURE 3 ]' 24 \\f Y.n) / i M: h.d.' / t n/ y ?.~/ Y,. s s

• / BrC Ut0>tnSprince M Yled

) c) BresketCloudspy ~ Dixhsrged _N% k '.UY,.dh \\ 9 QU ) K \\ \\ $ml, 4 / h -+ G 4, m y Dip Latch H M ntplatchRelement el Dip Ae/tlatchDrande @ @fi K G i

1. Seep Reser

2. CheseCam

3. Messer Career mieg Lhk

/

4. Povet Pm i

& Trte W S. Trte Sheet LeesNag Sur%se pO

7. Trte Shoft h

e. Poe e.ot N

g,,' ( \\w%-r.J..

9. Canner Pee Lauer i

l 1OL Pode Laver Pm

• /

j

11. noverg C.meest Arm

,9 * / is.so ,, m c

13. u.*g c aan Pe P=

u C is. uma m uan @ g 1L umm mener ) e,G' itsprigam Laem

17. im Lm Ae g

' g1 @V sess mis Lmesive It le LM N /

19. Meshandam two Prune

{ --marusessanses att Mweser?

• sesh turfenes 61 * =heropm Dr*iss Owsed/J>r*w owraf hanm W ::; To nnt a;mw once hatten Shomee be ?> N).

Fig. 3 These 54s.' case Show the row Asde har&ms ofhwker not Un4qpe Wth EMarged Hew of THp.Wft endlasch i L t. 3 M 90 II

FIGllRE 4 Acactor Trip Brecker Arrangemsnt ROD CONTROL SYSTEM

• 3, gYA r- - -

i I L., _ _ _.. j D L. ,.. J TRAIN A E PROWCTION SYSTEM TR AIN E p-- q i i i RS l L ____) -l t ST i BYB L. .J_ 1 f r 9 ) ) ne s.: Ostput Sreaker I f G G M M e

.._.a w FICURE A ILLUSTRATIVE DIAD W REACTOR TRIP BREAXIA CIRCUITRY (S) MR& srh c-Tw"rAr y awr>er 5-y' <r/r &r l 1 h u# *# l l S I i f l L \\ ~ _.. -.. ~ ~ - ,, -. - - _. _ _, - _ -.. _. ~ _. _ _ - -..,,,,.. - - - -.

i ENCLOSURE 6 Licensee Event Report 50-370/07-009 (on McGuire 2 evnnti, 3 August 1987 s t ? a l l lb I t>s L P .F l [ l l l 5,

J B mvChe au ettwkatoa. geese gg.og &8*80vl3 Cut see lit 040a a LICENSEE EVENT REPORT ILER) ooc it=v . m i ..ei 4 o [s I o [o I o l 3l 7 to ! 1 loFl1 l 2 .aca.,,.a. n, McGuire Nuclear Station - Unit 2 ti1L4 see REACTOR TRIP BREAKER FAILURE DUE TO MECHANICAL FAILURE otmg a p aciLitill Hev0Lv 80 it, espom? pars #M tvle? Daf t its ts e evn.sg a 168 WO';Ta 047 vtaa vgas s e g.g,,a. .,y wogtu car team 8ac'.it.mawes Coen st g,va ge g. + oistoloici ii q7 O 2 87 87 O09 O i O O8 0:3 ai 7 i O i 5io j o i 0 i l i i i i i I I I r I i I I e......,.. m.,v. a =,,.,. ov....., o, i. C.. i <c.. -., -., ~,. -,a n i, 3 n n., S . n.,a,o.,

==*i + n n... .... m . n... a., =.. m. y ....ani . n.,a,. .,. gig.,g O oi .,n n.i i n. .,o n., . n.ia,. . n.io,. nai -a, ..m . n ain., = n ns, .n. n., .n...... . n.a a., L6Cleellt COegiaCT SCR fut$ Llh tili v.L...e uw. 3,7,3, ,7,0,3,3 J.B. DAY - LICENSING 70 4 CD s*Lets ont Line,04 tac. Coe.804t47 pa, Lust Osecaisse en fleis a ssont insi Cawlt 3v 578 C0w'048%, N gh eY C.av$ 8 dvlttu C0t**0% I %f T TO pa g Vl2p Y I f I f f I I B JlC B; M Ri i I I l 1 1 I I I I I I i 1 1 w0%'= Cav .gae Sve*Lautartat elece? tapICTIO 1948 xxl,sim _. :e,*rerrosvev e0,eart, ~] o 0 12 011 81 8 amer _ac,,i-..x n. At 2341 on July 2,1987 during performance of the control rod drop timing tests, personnel detected smoke in the area of the reactor trip switchgear. he control room was notified and operators manually tripped the reactor trip breakers (RTBs). Control room status lights indicated both breakers had opened, though investigation revealed RTB 2 to be closed. D e breaker could not be opened locally until an attempt was made to manually tension the breaker closure spring. The operators were not holding the feedvater isolation reset button when the breaker did open so a Train B Teedvater Isolation occurred, though it did not cause any adverse effects. I ne failure of the breaker has been classified as a manufacturing deficiency due i j to a fabrication deficiency causing the failure of a weld inside the breaker. De investigation has revealed that the breaker failed to automatically open due to a mechanical binding of the breaker. A veld failure and vorn components of the i breaker closure mechanism are suspected of causing the binding. but nothing conclusive has been found during the investigation at McGuire vSich pinpoints the The breaker will undergo further inspections and tests at Westinghouse. cause. 170

• MWU

.....o -.,. l

~ .4 s........ ;..... ve..m LICENSEE EVENT REPORT (LER) TEXT CONTINUATION .e e i.,,,,, s;;..y,,, f ~.> -~~ r McGuire Nuclear Station - Unit 2 o 15Io!oIoi ) 7 0 8, 7 ' 010 '9 0 0 0 2 c0 12 ....e-._.. .e -.- INTRODUCTION: On July 2, 1987, at 2341 during performance of the Control Rod Drop Timing tests, personnel detected smoke in the area of the Reactor Trip Switchgear [EIIS JC). Control room operators were notified of the smoke and manually tripped the Reactor Trip Breakers (RTB). Control room status lights indicated both breakers had apparently opened. When investigating the cause of the smoke, it was determined that the Westinghouse DS-416 RTB, installed in the 2RTB cubicle, was in the closed position. Operators made several unsuccessful attempts to locally trip the breaker. When an attempt was made to manually tension the breaker closure springs, the breaker opened. Operators were not holding the Train B Teodwater [EIIS SJ) Isolation Roset button at the time the breaker opened which resulted in a Train B Feedvater Isolation. The breaker was removed from its cubicle for testing to determine the reason it did not open. A Bypass RTB was removed from Unit 1 1BYB cubicle and temporarily installed in the vacant Unit 2 2RTB cubicle to complete the Rod Drop Timing teJting. The Control Room indicating lights for 2RTB cubicle have functioned properly during all subsequent tests. Unit 2 was in Mode 3. Hot Standby, at the time the breaker failed to open. The failure of the breaker has been classified as a manufacturing deficiency due to a fabrication deficiency causing th6 failure of a weld inside the breaker. The investigation has revealed that the breaker failed to automatically open due to a mechanical binding of the breaker. A veld failure and worn components of the breaker closure mechanism are suspected of causing the binding, but nothing l conclusive has been found during the investigation at McGuire which pinpoints the cause. The breaker vill undergo further inspections and tests at Westinghouse in an attempt to determine the cause of the binding. An investigation into the cause for the apparent erroneous Control Roon breaker position indicator light revealed all circuits functioning properly. No abnormalities have been discovered which would have caused an open indication when the breaker was still in the closed position. D e erroneous breaker open indication in the Control Roon vill not be assigned a cause code until the final analysis on the breaker is completed by Westinghouse and all wires associated with the open indication light have been fully checked. EVALUATION:

Background

here are four identical RTBs for each unit's Rod Control system. D e normal alignment uses two main breakers while two bypass breakers are used to support testing and allow continuous operation of the systen during periodic maintenance. Cubicles which house the breakers are labeled as RTA, RTB, BYA, and BYB. ne four breakers are arranged in a series-parallel network (See Figure 1), which allows a main breaker and the opposite train bypass breaker to be deactivated and isolated t 1 i ene(.G ew seee

N LICENSEE EVENT REPORT (LER) TEXT CONTINUATION e

• • ""**""4 I

a t

• w' H ' '

'+aa ..eg,....., I "' N ' *" ' McGuire Nuclear Station'- Unit 2 3 7 0 l8 7 i !009 00 03 l12 o ts t o t o 10 I I ! i I l -: I ' C -n... . car m. m for testing or maintenance. These breakers may be moved from cubicle to cubicle as required, The RTBs connect the power from the Motor / Generator Sets to the Reactor Control Rod Drive (EIIS AA) sechanisms. When either of the two operable breakers, which are aligned in series, opens, the power is cut off to the control rod drives releasing the rods, and the Reactor is tripped. The McGuire Unit 2 Operating License specifies that all four Reactor Trip Breakers be tested 7 deys prior to unit startup and have similar testing plus time response testing every 31 days. Every 6 months, the breakers are thoroughly tested and serviced according to Westinghouse specifications. Maintenance of this fashion is also performed on all Unit 1 Reactor Trip Breakers, though not required by the Unit 1 License. ne Westinghouse DS-416 Air Circuit Breaker Inspection and Maintenance procedure provides for inspection and maintenance of the RTB and the connection hardware inside the breaker cubicle. Tests are performed on the Under Voltage (UV) trip solenoid and the Shunt trip solenoid to verify proper operation. Also, a force test is performed on the bar (trip bar) which is rotated when the breaker is tripped by either the UV trip solenoid, the Shunt trip soleniod, or the manual trip lever. A maximum trip force of less than 2 pounds to rotate the trip shaf t indicates satisfactory operation. Typical values foc trip force on a new breaker are in the 0.5 pound range. The breaker should trip via the UV trip solenoid when the voltage supplied to the UV coil decreases to between 28.8 volts and 14.4 volts vben the UV solenoid deenergizes. nis UV trip solenoid action is tested to assure it provides a mintmum force of 3 pounds to rotate the trip shaf t. Typical values for a new UV trip solenoid force acting on the trip shaft is in the 4 to 5 pound range. ne breaker inspection procedure is performed every 6 months for each breaker to inspect for damaged wires: tightness of the connecting nuts and bolts; the operating mechanism for binding, loosenass, worn or defective parts; and cracked welds. During these inspections, the breaker is normally cycled on the order of 50 times to perform the inspection according to the Westinghouse Maintenance Program Manual for DS-416 RTBs. This procedure was last performed for breaker B-4 on December 18, 1986 land documented that no problema were found prior to placing the breaker back into service, breaker Components and Operation ne RTBs at McGuire are Westinghouse DS-416,1600 Aap 600 Volt, three phase breakers that can be opened or closed either automatically or manually. Figure 2 Reactor Trip Breaker Shaf ts And Components, shows there are three main shaf ts fuside the treakers the pole shaft (21), the trip shaft (2), and the crank shaft (15). The crank shaft is used to tension (charge) the close spring (18) by crank shaf t rotation driven by either the motor or the manual energency charge handle (23). Rotation of the crank shaf t and the connected close can (5) is always in the counter clockwise direction. Once the close spring is charged, the spring release latch (7) holds the spring open by preventing the rotation of the

.. s................ E '" * ' LICENSEE EVENT REPORT (LERI TEXT CONTINUATION .u: u i ....y,

• "'"'"8

..um,,u...... 6""'" ""2

a..

F g. - 9,,. McGuire Nuclear Station - Unit 2 o 3 o l c l o, 3 7.0 S 7 !._! O,O 9 0.0

1. 2 9

i 0a e. ..,.._.c. .., a. wa-ma n m n close can stop roller (6) until the spring release device (8) rotates the spring release latch down around the stop roller (6). The pole shaft (21) supports the small angular rotation of the center pole lever (28), which ties the movement of the Y-Phase moving contact assembly (25) with the main drive link (27). Two other similar pole shaf t levers (not shown) are found to the far right and left of the center pole lever. These levers operate the X-Phase and the Z-Phase moving contact assemblies in unison with the Y-Phase due to the comon pole shaf t. An anti-bounce lever (not shovn) is similarly sounted to the pole shaft adjacent to the center pole lever and tied to the main drive link via its extended upper connecting pin. The anti-bounce lever acts as a dashpot to slow down the breaker only during opening to prevent any momentary reclosure action of the breaker. Another pole shaft laver mounted on the pole shaft (not shown) is connected to a rod which is directed upward and is attached to the botton block of auxiliary contact switches. Two additional auxiliary contact switch blocks are stacked on top of the bottom auxiliary contact block switches for a total of three (see Figure 3). Two short rods connect the movement of the bottom auxiliary contact block with the upper two. The rods tie the movement of the pole uhaf t with the positita of three stacked auxiliary switch blocks. Each of the three auxiliary switches contain four sets of contacts ("A finger" contacts and "B finger" contacts) for use in breaker position indication and control functions. The A finger contacts are open and, the B finger contacts are closed, when the breaker is open. Referring to Figure 4A. when the main drive link (14) moves upward, and to the right, the connected center pole lever (9) rotates counter clockwise with the pole shaf t driving the moving contact assemblies (11) (only one of three shown) forward, closing the breaker. In order for the main drive link to move upward and close the breaker, the pivot point (4) of the roller constraining link (3), tied to the trip latch (5), must be in a fixed position. This allows the opposite end of the roller constraining link, which is attached to the roller end of the main drive link, to travel upwards on the lobe on the close can (2) during rotation. The roller constraining link prevents the main drive link from traveling down the lobe on the close can as the can rotates. The close can always rotates in the counterclockwise direction. Referring to Figures 4 and 5. the trip shaf t (7) performs a function for both the closing and the opening (tripping) of the breaker. Prior to closing the breaker, the trip latch (5) is in the tripped position (See Figure 5A) and must be reset after the breaker has been opened to allev breaker closure. During the rotation of the crank shaf t to charge the close spring, the trip latch moves down through the notch in the trip shaf t, under the direction of the roller constraining link (3) to allow rotation of the trip shaft. The trip shaf t then rotates to a position which prevents the trip latch from moving back upward through the notch in the trip shaft (See Figure 5B). When the trip shaft (7) and the trip latch (5) ate in this position, the spring is charged and the can is in the position to

r. lose the breaker (see Figure 4A). Af ter the spring release latch (16) drops A

NO

e r.= m .6 4....** *t. o c.. n.3 UCENSEE EVENT REPORT (LERI TEXT CONTINUATION y. ca.o w en : i a. %..o. ..n, ..u.......... ,ug-i i 4 McGuire Nuclear Station - Unit 2 o 15Io1010I 3' 7 :0 ; 817 ; f 0 t0 9 00 0' 5 c" 12 -o ~...<.. .... o c ar mu.. + 1 below the stop roller (1). the close springs will rotate the close cam lifting the main drive link under the direction of the roller constraining link, closing the breaker contacts (see Figure 4B). If at any time the trip shaft is rotated to allow the upward movement of the trip latch, the breaker will open. If this action occurs while the breaker is attempting to close, the breaker will trip and remain open. Referring to Tigure 6. tripping the breaker is accomplished by any one or combination of three methods: manually at the breaker, the Under Voltage (UV) solenoid operation coil, and the Shunt trip solenoid operation coil. All three of these methods rotate the trip shaft (8) directly which trips the breaker. When the breaker is closed, the open spring (Tigure 2. Component 19) and the springs loaded behind the fixed contact assemblies (not shovn) maintain a clockwise rotational force on the pole shaft. n e pole shaft is prevented from rotating clockwise by the tension of the roller constraining link (3) via the compression in the main drive link (10) and the center pole lever. Motion of the roller constraining link (3) is prevented by the position of the trip shaft (8) which prevents upward travel of the trip latch (1). When the trip shaft (8) is rotated allowing the trip latch (1) to move upward through the trip shaft notch, the roller constraining link allows the main drive link roller (11) to roll down the close can, opening the breaker. Figure 8 is provided to show a sectional view of the breaker components and to detail the components on the crank shaft. Description of Incident On the night of July 2, 1987, the Control Rod Drop Tests were being performed. Testing had been completed for Shutdown Banks A through C. Testing was being concluded on Shutdown Bank D and the RTBs were required to be opened as directed by the test procedure. Operators opened the breakers at 2343:46 and, while [ holding the Teodwater Isolation Reset buttons as directed by procedure, observed i the breaker position lights in the Control Roon change from closed to open. The Events Recorder indicated the Train A breaker had opened, but it did not indicate a change from Closed to, Open for the Train B 2RTB Breaker, unknown to the Control Room Operators. However. Operators in the Control Roon did observe the illuminated Open status indicator light for the Train B 2RTB Breaker (here.ifter called breaker B-4). Operators closed the RTBs to allow continuation of the test with Shutdown (S/D) Bank E Control Rods. Operators observed the breaker position lights change from open to closed and shortly thereafter, began to withdraw S/D Bank E rods. Th e Events Recorder again did not show a change of breaker position for breaker B-4 As Operators were withdrawing S/D Bank E. they noticed the demand counter for this bank was not counting up from aero. The operators notified the personnel who were vorking with them on this test in an adjacent room which contained the RTBs. At the same time, test personnel detected smoke which appeared to be coming from the l RTB cabinets and informed operators who immediately opened the RTBs (at 2345:55). i l e sae.. ne.

'Y" 4............. UCENSEE EVENT REPORT (LERI TEXT CONTINUATION mn ..a......n, "'***""8 6" w'H ' ' '* u 2 pu q;e-2.. i l McGuire Nuclear Station - Unit 2 o 15 I o I o I o i 3! 7 'O ! Si 7 !! O 10 9 00 0 6 C8 l' 2 .. w. ..c.,- w. o n while holding the Teodwater Isolation Reset buttons, and observed the status lights change from closed to open. Unknown to the Control Roon Operators the Events Recorder again did not indicate that breaker B-4 had opened. Control Roon personnel notified Supervisors to investigate the smoke and found that the breaker in 2RTB (B-4) had not tripped and was the source of the smoke. The Supervisor informed Control Roon personnel that a local (manual) trip of the breaker was to be attempted, and that they should hold the Feedwater Isolation Roset button. make several pushes on the breaker's manual trip lever, butSince the breaker was the lever would not move any appreciable amount. did not open the breaker. These manipulations of the local manual trip lever began to manually charge the close spring.In an attempt to cycle the breaker, the Supervisor As the close spring was being tensioned, the breaker opened at 2358:01. Due to the difficulties and long delay (approximately 12 minutes) in opening the breaker, Control Room operators were not holding the Train B Teodwater Isolation Reset button when the breaker finally opened. This resulted in a Train B Teodwater Isolation signal that was generated by the Solid State Protection System (SSPS). Closure of the Condensate Teodwater valves under the direction of the Teedwater Isolation signal did not cause adverse affects. Operators made an entry in the Unit 2 Technical Specification Action Item Logbook declaring 2RTB inoperable. Breaker B-4 was removed from 2RTB cubicle and placed on the floor. Operators notified the Shif t Engineer (Shif t Technical Advisor) and the Unit 2 Coordinator of the situatian. The Shift Coordinator advised operators to contact Transmissions Generation Station Support (CSS) personnel to determine if the Unit 1 Bypass Breaker from IBYB cubicle could be used in the 2RTB cubicle replacing the failed breaker B-4 Operators wrote a work request to investigate / repair the problem with RTB B-4 Operators also implemented the NRC Immediate Notification Requirements procedure, and informed the NRC of the breaker failure and the Teodwater Isolation (an Engineered Safety Teature actuation) at 0112. The Station Manager inattveted operators not to withdraw any Unit 2 Control Rods without his permission. McGuire Management discussed the situation with NRC personnel and at 1410 on July 3,1987, appropriate permission was given to operators to resume Rod Drop Timing tests with the use of a Unit 1 Bypass RTB. tests were completed. At 1730 that same day. Rod Drop Timing The problem associated VAth the demand counter not counting up from aero was attributed to a sticky release latch which prevented the counter wheels from moving. In no way did the demand counter probles cause or assist in the failure of breaker B-4 to open. .u.... m.

.ss......s. UCENSEE EVENT REPORT (LER) TEXT CONTINUATION . " - i:. ,,.,,i.. .,n, m ni -,ia, ..w.,...... ,inig;.. l I i i 0*0 0 7 ost1 2 McGuire Nuclear Station - Unit 2 0 41 o I e i o 13! 7 :0 l 817 : - 0 +0 9 -n.. .m.., ...m..v.- w. m Descriptf or. Of Breaker Inspectig CSS took breaker B-4 to their shop to begin an initial assessment to determine the cause of the failure. The breaker was cycled a total of three times during their inspection. GSS personnel closed the breaker electrically, and continuity checks of the auxiliary switch contacts used in the Shunt trip circuit indicated that they were closed. ne breaker was tripped by use of the UV trip attachment and the breaker opened. The breaker was closed electrically a second time and a trip force test was perferned on the trip shaft, but rotation of the trip shaft did not trip the breaker. This constituted the second failure of the breaker. Repeating what operators had previously done to open the breaker, they began to manually tension the close spring. As soon so the crank shaft (and close cas) began rotation with the manual emergency charging device the breaker jarred open. The breaker was closed electrically a third time and GSS personnel made a second attempt to perform a trip force test. On this attempt, the breaker opened successfully with a trip shaft tripping force of 1.25 pounds. Testing was stopped ct 1030 on Friday, July 3, 1987, and the breaker was voluntarily quarantined until all groups concerned with the failure could be present. On Tuesday, July 7,1987, an NRC Inspection Team arrived on site and a meeting was held with Duke, Westinghouse, and NRC personnel at 1000 to determine a planned course of action to methodically inspect and test breaker B-4. Attention of the meeting focused on attempting to preserve any evidence which would show why the breaker failed to open. Test'.ts of the breaker resuasd on Tuesday, July 7, with Duke, h'RC, and Westinghouse personnel overseeing all aspects of the inspection. Inspections of the breaker mechanisms revealed the weld which connects the center pole lever to the pole shaft had failed. This fillet veld had cracked along the cntire length of the weld. Welding is performed on only one side of the center pole lever to attach it to the pole shaft and was made approximately half way around the pole shaft. Inspections of other pole shaft levers revealed their veld lengths to be on the order of 3/8 to 3/4 the circumference of the adjoining pole shaft. Corresponding abnormal wear markings were found at two dif ferent locations in the breaker. An indentation burr of approximately 3/32 inches long was found on the notch of the trip shaf t which mated with a small wear mark on the trip latch. Subsequent testing to determine the force required to rotate the trip shaft, allowing the trip latch to pass through the notch on the trip shaf t, was within acceptable limits. De second area of excessive wear was the far left side steel 1aminant of the four piece laminated close can surface, which contacts the roller on the main drive link. Wear was also found on the right most steel plate of the close cas, but was not as excessive as the left most plate. i In an attempt to recreate the binding situations which had twice prevented the breaker from opening, thirty-one trips of the breaker were performed at McGuire. Several attempts were made to artificially produce a similar binding of the l l h

.c w. 4.a UCENSEE EVENT REPORT (LERI TEXT CONTINUATION .. ".. o c .".w.*.'v".'.

• • "***"8

'm 2 ..c.w........ c i i ' O' 8 c l 2 ,o is to lo io i 3t7 0 !8t7 - O 10' 9 - O'O McGuire Nuclear Station - Unit 2 -i o m. .e. .oa-nu. a breaker mechanisms, but each time the breaker was directed to trip, the breaker opened successfully. During the times the breaker was closed, inspections revealed the interfaces between the sain drive link top pin, main drive link, roller constraining link, the trip latch, and the center pole lever with the broken veld, vore able to twist and subsequently create a bi ains situation. All other trip system components inspected did not reveal any abnormal conditions. The collective inspection and testing of the breaker wechanisms at McGuire was concluded on Thursday, July 9, 1987. Inspections began on the remaining three Unit 2 RT3s on July 13, 1987. During the inspection of the breaker from cubicle 2RIA (breaker B-1) QA personnel found a questionable veld on the pole shaf t lever which operates the auxiliary switch linkage. The inepector performed a liquid dye penetrant test on the veld and discovered what appeared to be a hair line crack in the veld. A force test was performed on the lever arm with a safety factor greater than 6. The lever satisfactorily passed the test, and the inspection of the breaker was complete the breaker was cleared to return to service. During the inspection of the breaker from cubicle 2RT3 (breaker B-2), visual inspection revealed heavy wear on the left side of the close can. The main drive link appeared to be tilted approximately 3' to the left from the lateral plane. Clearance of the mechanisms appeared to be adequate and velds were good. The same type of wear was observed on breaker B-3 and all velds were determined to be acceptable. Description Of Breaker Indication Inspection The Control Room indications for the Reactor Trip Breakers are comprised of a green light to indicate an Open breaker position and a red light to indicate a closed breaker position. Operation of the green light (0 pen) is controlled by the breaker's auxiliary switches. These switches are operated by a connection rod to a lever on the pole shaft and by design provide a positive position indication for the breaker. The breaker main contacts position is determined by the pole shaf t position. The red Control Roon status indicator light for breaker Closed does not provide a positive indication of true breaker position. There is a switch contact in parallel with the 11'ght which "shorts" around the light to extinguish it when the Control Roon manual trip pistol grip switch is operated. Electrical testing performed on the B-4 breaker indicated that all switch and internal wiring were satisfactory except for the burned out Shunt trip coil. The Shunt trip coil burned out approximately 2 to 5 minutes af ter the coil was i energized at 2341:16 but did not deenergine because of the f ailure of the breaker to open. This coil is not rated for continuous duty and is normally deenergized by another auxiliary switch as the main breaker contacts open. Testing of the breaker main contacts, as related to auxiliary switch contacts, revealed the main contacts aust physically separate to deenergine the Shunt trip coil. The breaker main contacts aust move 5/8 of an inch further open for the auxiliary switch contact to close sending a signal to the Events Re: order indicating that the

.. t............ Y UCENSEE EVENT REPORT (LERI TEXT CONTINUATION +u i e*u, emi' e m a si w eie e ..n., 0.... o. i*9.,.- McGuire Nuclear Station - Unit 2 o a t o t o 10 3t 7 0 [8i7 ; l0 019 % I 00 0 9 ca 12 s.

w. s c e m a. ~ m breaker had opened. The breaker main contacts move a total of approximately 3.5 l

I inches during one operation. ne Shunt clearing contact opens well before the green light (Open) and the Events Recorder contacts. This indicates that the Shunt trip coil could not have burned l out if the auxiliary switch contacts for the green light (Open) had made contact. Since both the green light (Open) and the Events Recorder operate simultaneously. this would provide printed Events Recorder data that the breaker did not open and likewise, the auxiliary switch contacts for the green light (open) could not have ande contact to directly illuminate the green light unless an external circuit viring probles existed. The external wiring circuits were checked for circuit polarity and determined to 1e correct in all cases. Circuit grounds on the 125 Volts DC battery power systen J were investigated. A ground was found on the positive side of the DC system but was not located inside the 'oreaker compartment. After failed breaker B-4 was replaced with an operable breaker, the replacement breaker was cycled five times in cubicle 2RTB. During this testing, all breaker position light indicators and operations of the breaker performed correctly. Conclusion The f ailed RTB B-4 was thoroughly inspected and tested in December 1986. The inspection procedure documents that no problems were fount during that inspection. Since that inspection, breaker B-4 has been cycled to satisfy the testing i requirements of the Unit 2 License and during other related tests. Each time, the breaker performed as required and passed the time response tests required every 31 days without exceeding the maximum 150 stilisecond opening time limit. Time response testing verified sound operation within all interacting trip components. i Preliminary investigation results obtained at McGuire concurrently with Duke. Westinghouse, and the NP.C Inspection Team, could not determine a definitive reason i i to explain why the breaker stuck closed and failed to open. Three areas are prer,ently considered factors which may have contributed to the f ailure of the breaker to open as directed weld failure, aanufacturing tolerances of the breaker components, and the cumulative effect of the high number of cycles on the breaker. The actual number of cycles the breaker had could not be determined to any degree of accuracy since the breaker was retrofitted with a counter after the initial installation. Best present estimates place the number of cycles above 3000. Closure mechanism parts, particularly the close cas, showed signs of abnormal and excessive wear, but the exact point of the binding, which resulted in the breaker failing to open, has not been determined. The failed weld on breaker 3-4 is located on the pole shaf t (21) where the center pole lever (28) passes through. The majority of the weld remained a the pub ]l i shaf t due to a more complete fusion of the weld to the pole shaf t chn ta tna center pole lever. Operational ability of the f ailed breaker was ard 7 during cycling due to the common top connection pin in the sain d-17) r r sgA W 'Y e j

t.c e,- au s.............. LICENSEE EVENT REPORT (LER) TEXT CONTINU ATION ax="'" I s o w~ u '

• ai j...

.m. ,; c - 1 McGuire Nuclear Station - Unit 2 015101010I3!7 0 !8i7 009-l' O cs' l' 2 -v... .na.. . m. c n-nu e, and the adjacent and similar anti-bounce lever mounted on the pole shaf t (21) transferring the loads. Terces opening and closing the breaker were transmitted through the anti-bounce lever and the top connection pin in the main drive link. Investigation into the Shunt trip coil probles determined it failed sometime af ter the breaker B-4 was cycled at the conclusion of Control Rod Drop Timing tests of Shutdown Bank D and before rod drop testing was started on Shutdown Bank E. The Events Recorder indicated the breaker did not open during these attempted breaker cycles and the Shunt trip coil remained continuously energized in its attempt to open the breaker. Due to overheating, the Shunt trip coil burned up and shorted itself to ground. The resultant smoke is what alerted IAE and OPS personnel to the breaker problem. The only remaining testing to be performed at McGuire is the inspection of the Unit 1 RTBs and a wire to vire resistance checkout of the external breaker Open position indicator circuits. This checkout will deturnine if a viring error exists, external to the brSaker, which may have caused the erroneous green light (Open) indication when the breaker failed to open. A review of past incidents at McGuire revealed one incident in which an P.TB failed j to open during testing. 1.ER 370/83-03 describes W tests in which several dif ferent failures prevented automatic opening of the breakers under W trip force only. Dese failures were attributed to bindings of the W trip arm lack of a 4 t gap between the trip arm and the trip shaft pin, and missing retainer rings. The breakers would open subsequently by use of the Shunt trip solenoid. Since there were no excessively worn areas, and the breakers did not bind and prevent the breaker from tripping, the failure of breaker B-4 to open is not considered recurring with respect to these past events. Turther investigation into Westinghouse breaker failures, with the assistance of the Nuclear Plant Reliability Data System (NPRDS). revealed many failures of DS-416 breakers used in various applications. Primarily, the failures were failures to close due to electrical problems in auxiliary swit.:hes and external control circuitry. However, three problem 6 were associated with broken welds. Two cccurrences at other utilities involved weld failures associated with the pole shaft. The third occurrence was associated with the secondary disconnect support bracket. Most recently, a DS-416 Reactor Trip Breaker in use as a Bypass breaker at Catawba Nuclear Station ves being inspected due to this incident at McGuire. Welds associated with the pole shaft appear to have cracks in the welds. Further information from the inspection of these velds and f rom the remaining breaker inspections will be included in the addendum to this report describing the results from the Vestinghouse lab inspection of breaker B-4. Also, the three veld failures at other utilities will be considered by Westinghouse, n ere were no personntl injuries, radiation overexposures, or releases of radioactive material as a result of this incident. "5'*""

.s s............. Y LICENSEE EVENT REPORT (LERI TEXT CONTINVATION ."*- 4: + q.a se g 'o""'"*"* . I * ' w' " ' '

• • ' 8

...a, u...,,, i.... c ' a r;, t - 9 ',.. _ i 015 t o l 0101317 l0 l 817 !--li d O' 9 - 0 ' O l'1 C" l' 2 McGuire Nuclear Station - Unit 2 .s o a m ou. a e.. m s*< *e m s an CORRECTIVE ACTIONS: Lamediate: Operators initiat6d a manual Reactor trip after they were alerted to seoke in the area of the breakers. Subsequenet 1) Operators tripped breaker B-4 locally and removed it from the 2RTB cubicle. 2) The breaker was declared inoperable and secured from the Rot Drep Timing tests. 3) The NRC was notified of the failed brasker and the Teodwater Isolation (ESF Actuation). 4) All Unit 2 Reactor Trip Breakers were thoroughly inspected and a new breaker vna installed as a result of the failed breaker. 5) Operators have issued a Special Order directing personnel in the Control Room to verify both Reactor Trfp Breakers have opened in the time period af ter a Reactor trip and before reclosing the breakers. This Special Order is applicable to both units. Planned: 1) Reactor Trip Breaker B-4 vill be sent to a Westinghouse lab in Pittsburgh, Pennsylvania, and a proposed faspection with testics will be performed af ter Duke Power. Westingnouse, and the NAC have concurred es to what inspections and tests are to be done at the lab; Duke and NRC persoanel will be present. 2) All Unit 1 RTBs will be thoroughly inspected to determine the quality of the velds in the breakers and to detect any areas of abnormal wear. 3) Ope,rators will make a change to the Reactor Trip Procedures, to physically verify RTA status af ter a Reactor trip, and to maintain the neatus of a single breaker if it fails to open. 4) Changes will be made to procedures which involve RTB operations to physically verify the breaker status af ter tripping. 5) Checks will be performed an the external wiring circuits for the breaker Open indication light. 6) One new RTB of the new vintage will be installed in Unit 1 replacing an older vintage breaker, to further increase breaker reliability for Reactor trips. m ucm e -.

........ 77'..... N LICENSEE EVENT REPORT lLER) TEXT CONTINUATION .

• i: e 8'**"

'"' 8 '...,e,u...... ,m 'L :- t ' +., ;; l I i i McGuire Nuclear Station - Unit 2 o p t o j 010 8 3! 7 'O l 8 7 p : 0 0 t 9 -- 0 0 l* 2 08 1'2 ..c.m. m SATETT ANALYSIS: Unit 2 was in Mode 3. Hot Standby, and had not been critical for 63 days due to a refueling outage. Control Rod Drop Timing tests were being performed which allowed only one bank of Control Rods to be withdrawn f rom the core at any given t ime. To verify that the rods of the bank being tested have all been dropped for that section of the teste both RT3s are opened ensuring all rods are at the bottom of the core prior to withdrawing another bank. In this incident, one of the RTBs failed to open, but the second redundant breaker, did open. The failure of this breaker was discovered due to the smoke which resulted when the Shunt trip coil overheated and burned up as a result of not being reset after energining. If the Shunt trip coil had not burned up or if the smoke had not been noticed the time at which the stuck closed breaker would have been discovered is not certain, assuming the condition which provided the erroneous indication in the Control Room continued to exist. It should be noted that the maximum time period a breaker could be in an unknown failed condition is 31 days due to the surveillance requirements. This condition of one breaker continuously stuck closed would not have af fected the timing tests being performed since the redundant operating breaker would have performed the opening function. Had this incident occurred in Mode 1. Power Operation with the unit at 100% full power, the recundant RT3 should have performed the opening action required to cut power to the Control Rod Drives tripping the Reactor. Postulating an incident i where the stuck c1.osed breaker condition previously existed, a failure of the second breaker would involve both RT3s failing to shut down the Reactor. This scenario of failura to shut the Reactor down is addressed by the Emergency Procedure EP/5000/11 Suberiticality. If a Reactor trip has nor occurred, three Operator actions would be carried out which would cause the Control Rods to be inserted. First. Operators would b6 sin to manually insert thi Control Rods. Another Operator would go to the adjacent room which contains the Motor / Generator (M/G) sets and open tha output breakers from the M/G sets and aise open the supply breakers to the motors of the M/G sets. This action can be completed less than 1 I minute af ter observing the Reactor as still critical. It should be noted that during an actual Reacto,r trip event. Operators in the Control Roon vill usually first look at the Digital Rod Position Indicator lights for confirmation that the rods have dropped af ter opening the RT3s. Therefore, sufficient capability to trip the Reactor would exist at all times. I The health and safety of the public were not affected by this incident. I i l l

Tigure 1 Reactor Trty Breaker Arrangoes,- ROO CONTROL SYSTEM 1 t--- +] gyA t e i I f 8 !i RT(/ i ( 'I i UV I i i UV i I'.--<J Lo l- _J Sotto TRAIN A L__ STATE e l PROTECTION ~~] ~~ l q SYSTEM TR AIN 8 p-UV I '\\ l '\\ a * ')1 I UV !. _) _ _ _ ;g l SYS l i ST u__. .. _s f) I) i no set i Output treaker l 1 l l G G ? i M .i M I i l l t l l \\ l l

7__ 9 = ce h[ / - \\;;Q[, ', m n A 9 7 l J I ( g N /, Q b y-g f } '/- / Af 5 ~ y 3l 19.1 RIP /OPEN SPRING =

10. RACHET 1pHEEL

20. CLOSING SPRING ANCHOR C

1. SHUNT TRF DEVICE

11. HOLD PAR

21. POLE SHAFT

2. TRP SHATT

12. DRIVE PLATE

22. MOTOR

3. ROLLER CONSTRAMING UNK13. EMERCENCY CH RGE PAK

23. EMERGO4CY CHARCE HANDLE

4. TRIP LATCH

14. OSQLLATOR

24. MOTOR CRANK ANO HANDLE

5. CLOSE CAM

15. CRANK SHAFT

25. MOvMG CONTACT ASSEWOLY

6. STOP ROLLER

16. EMERGENCY CHARGE DEVICE

25. MSULATINC UNK

7. SPRING RELEASE LATCH

17. CRANK ARM

27. MAIN DRIVE UNK

8. SPRNC RELEASE DEVICE

18. CLOSMG SPRNC

28. CENTER POLE LENER

9. OSCILLATOR PAE

l Figure 3 l l Reactor Trip Breaker With Tront Panet Recoved f 1 I i 1 1-BRE AKER AUXIUARY $ WITCH 4-LEVERING MICHAMSM MOUNTING ASSEMBLY (TOP AND Ml00LE I l SCREWS LOCAftD ON TOP Pt.ATFORM FRONT COVERS REMOVE 0) l i b, l l c-l te d l I 4 "o-6-i:REAKER 0'ER A7)ON 3*8 COf,'NT E R 8 ', 's' 2- $PRING l l CHARGING MOTOR l I Y. l 1 fI j c l f e. l I b l hs. er l l ~ l \\ 3-POWER OPER.ATIO [

5. 5 HUNT TRtp (570 RED.tNERGY)

ATTACHMENy MECHANthM (STA) i l. i l l l l l I l l

Figuro 4 Breaker Open Coing Closed Action m y { l, ,C D g D 3 '%c / 5 a!... o., ,, s } f ,1 l ~ ' hVa__ /@ja g N 6p y PIGURE A: SRE AKER OPEN-SPRINGS CH ARGE0 l-O 6:o r s-1 0,,

• 4,,T 9

O I FIGURE 8: BRE AKER CLO$f D-SPRING $ Ol5 CHARGE 0 l l { m.--

Tigure 5 Trip Shaft Latch Details O 0;, - ( io D m- 'b FlOURE A: BRE AKER OPEN-SPRLNGS DISCH ARGED Blus t b / I / \\ i N 7 S I I ) e i FIGURE 5: TRIP SH&FT LATCH QtTAILS l Tmtr LATCH mtLE Alto Trip LATCH HELO gTRIPP(0) tRESET) F N470181 i

Figuro 6 o i-Brer.ker Cosporert Trip Actica l t P @/D Q -O- { &t se s%c ', k N I s\\ / c p) s ee/ e i L STA/SREAKEM INTERACTION l

7. TRIP SH AFT LEVER

1. TRIP LATCH S. TRIP SH AFT

2. TRIP LATCH PlVOT PIN

9. TRIP SHAF1 LATCH $URF ACE

3. ROLLER CONSTRAINING LINK

10. M AIN DRIVE LINK l

4. SHUNT TRIP DEVICE j

5. SHUNT TRIP ARMATURE

11. Mus curt Llut acLLas

4. SHUNT TRIP COLL i

j

Figure 7 Indicator Light Schematic 0 hf *. \\ . :.1% $ >y . - '. y. :. ~....., as g 'lIs [Q2 y g w a $f8 om $

M p

N N Y E fe eb Rf_ %y 4 6. K a1 Id31 ks e.s a 1 1 n 't t(. 4t t , k. t 9 v, g gj 4 .a a t. ll w. . n' ss 2 s T $%5 9 l 'f.Qf. E v& h l f 'gg C, ! 'i i m3s g 1, s . ansi 0, I L s. M# man 3s e e a ~ ~ ~ g --( 3 M[ I E 33, he pu 08 m E1 y t o, t M S nai Q/ u O 4 % g i e% s,( j >$A f i Ae r m 2 *1 4 $ 4H F+W' '; 4 w %5 [ E w ir c O k. 1 E -i j R i 1 I5 g 't I43' go g 3 .I r ,q 2 t, Y u -r. _*3 I w < r e ~ 4 o ? -6, - < r- ..is s a. I w I 7 -1 e

• d' N

^ 0 j 8 e D 3 *1 I {\\ --< H H > gen p 1

{

wa h 4, 6! s. e ,n d d" 5 g* El %'4 Ep d 1 v -w f

1. s (_s\\ W 2

u o e

Figure 8 Sectional viev Of Breaker Components E ii es i e gssaa3I t 55 s 2 WW ~ + 624dd ~ e s () .p:t I W / \\ilWd+!B/e [si cg a n 8 l_2 !, p:, k h . - Ad !'m 8 ( _,WT K-na .ms u-diti66 ?....?2%'\\g - n" ~ _ 9 E 8 " 1111 'iziliiB rn 8 A 3d! L l t! p gARo J S

p__g I

} N IW w*' r g 53855o I ..l .i. i o l, s, g g ~ j s, v 2 a. -es->-- -:k a = s $$ U A ei e ( + 3 e, g A o t= s 3 5 $ P, !

• g!1g1 l

$686 0 6 s a s ,o i l l l l o ~ ~ ~ ~ ~ ~ - - - -.

4 DmCE POWER GOMPANY .E P.O. BOX 33189 CHAmLOTTE. M.C. 98949

• • ";."N?

S August 3,1987 U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555

Subject:

McGuire Nuclear Station, Unit 2 Docket No. 50-370 LER 370/87-09 Centlemen: Pursuant to 10CTR 50.73 Sections (a)(1) and (d), attached is Licensee Event Report 370/87-09 concerning a Reactor Trip Breaker Failure due to a mechanical failure. This report is being submitted in accordance with 10CFR 50.73(a)(2)(iv). This event is considered to be of no significance with respect to the health and safety of the public. Very truly yours, s e---- h,- d}4-o Hal B. Tucke.' SEL/94/j ge Attachment xc: Dr. J. Nelson Crace American Nuclear Insurers Regional Administrator, Region II c/o Dottie Sherman, ANI Library U.S. Nuclear Regulatory Cormission The Exchange, Suite 245 101 Marietta.St., NW, Suite 2900 270 Farmington Avenue i Atlanta, CA 30323 Fermington, CT 06032 INPO Records Center Mr. Darl Hood Suite 1500 U.S. Nuclear Regulatory Commission l 1100 Circle 75 Parkway Office of Nuclear Reactor Regulation Atlanta, CA 30339 Washington, D.C. 20555 M&M Nuclear Consultants Mr. W.T. Orders 1221 Avenue of the Americas NRC Resident Inspector l New York, NY 10020 McGuire Nuclear Station

l ENCLOSURE 7 ' Notice of MJeting with Westinghouse Regarding Class IE Switchgear Models DS-416, DSL-416, DSL-420, DS-206, and DSL-206,' 21 Septer.ber 1987, Enclosure 2 (Calvert Cliffs report) kr .o o k: M, l-9' s Y-------- ________x____________

b enas'o, UNITED STATES ~, ?" NUCLEAR REGULATORY COMMISSION n WASHINGTON. D. C. 20555 9;. /Y' ~ P 1 SEP 1987 Docket Nos: 50-269 and 50-370 MEMORANDUM FOR: T. Murley* J. Partlow R. Capra J. Sniezek* F. Congel W. Butler F. Miraglia* H. Miller V. Nersos R. Starostecki* 5. Black

• J. Stolz S. Yarga*

B. Boger* E. Adensam D. Crutchfield* G. Lainas* H. Berkow L. Shao*- F. Schroeder K. Jabbour cf.~ Rod G. Holahan A. Thadani J. Richardson W. Lanning W. Troskoski THRU: Kahtan N. Jabbour, Acting Director . Project Directorate II-3 Division of Reactor Projects - I/II i FROM: Darl S. Hood, Project Manager Project Directorate II-3 Division of Reactor Projects - I/II

SUBJECT:

NOTICE OF MEETING WITH WESTINGHOUSE REGARDING CLASS IE SWITCHGEAR MODELS DS-416, DSL-416. 05-420, DS-206 AND DSL-206 TIME & DATE: September 23, 1987 9:00 am - 4:30 pm LOCATION: Air Rights Building Conference Room AR 2242 PURPOSE: To discuss the technical basis for Westinghouse's g. recommended actions, criteria and conclusions in Enclosure 1 pole shaf t assembly at Calvert Cliffs 1 (quate welds of the and related breaker failures due to inade 9 ) and Sequoyah 2. PARTICIPANTS: 1/ NRC WESTINGHOUSE ~ Dichardson P. Morris J. Stone ' W. Bamford, et al. D. Hood, e t al. l &arl S. Hood, nL ,D ject Manager ? . Project Director 11-3 ,. Division o,f Reactor Projects f(II p i cc: Ace next page x.' ?-4 4 - -t. ' V. D J

Enclosures:

H] ' '. ' - (1) Westinghouse letter of. tamber 11, 1986 t' A. Li3.(2)..Calvert41 ~.4 aseo W tader.18 -1986 ..y n

• gi,, :.. <. 8,. c:.R f,y,..~ ( p

.N to: ~ fE.hterested members of'the pdb'lic to attend as observers ' s'. ~ Ituiasett f:' i ling.ead,$tatement of IWtc Staff Policy.* 43 Federal Register is.i.. ' ., 4 'g. 42 a 'wr% %. ,~ nN 7g, .! 4 d.t. m. .s

t'r. H. B. Tucker Duke Power Comoany McGuire Nuclear Station ec: Mr. A.V. Carr, Esq. Dr. John M. Rarry Duke Power Comoany Department of Er.vironmental Health P. O. Box 33189 Fecklenburg County 422 South Church Street 1200 Blythe Boulevard Charlotte, North Carolina ?8242 Charlotte, North Carolina 28203 County Manager of Mecklenburg County Mr. Dayne H. Brown, Chief 720 East Fourth Street Radiation Protection Branca Charlotte, North Carolina 28202 Division of Facility Services Department of Human Resources 701 Barbour Drive Mr. Robert Gill Raleigh, Nortn Carolina 27603-2008 Duke Power Company Nuclear Production Department P. O. Box 33189 Charlotte, North Carolina 28242 J. Michael McGarry, III, Esq. Bishop Libennan, Cook, Purcell anc Reynolds 1200 Seventeenth Street, N.W. Washington, D. C. 20036 Senior Resident Inspector c/o U.S. Nuclear Regulatory Comission Route 4. Box 529 Hunterville, North Carolina 28078 Regional Administrator, Region II U.S. Nuclear Regulatory Comission, 101 Marietta Street, N.W., Suite 2900 Atlanta, Georgia 30323 L. L. Williams Area Manager, Mid-South Area ESSD Projects Westinghouse Electric Corporation MNC West Tower - Bay 239 P. O. Box 355 Pittsburgh, Pennsylvania 15230

^ / L D eLosunz 1 I 1 i l I The following information and reconnendations are provided for your use. if l you havo Westinghouse Models D5 416 DSL-416, DS 420. D5-206 and DSL-206 switchgoarinstalledinyourplantInIEservice. 8ACKGROUND i On July 2,1987, it was remrted that a DS-416 raractor trip breaker did not l open on demand at McGuire 'Jnit 2 during rod drop testing following a refueling ~ j outage. This malfunction was determined when in the vicinity of the reactor trip switchgear. plant personnel observed smoke Since the breaker had not 4 opened 9n demand the shunt coli current was not interrupted resulting in a 1 The breaker could not be tri damagea, coil. the mangal charging handle was manipulated.pped manually, but did trip when During subsequent cycling on the test bench, the breaker jammed again. An inspection was conducted at the site jointly by Westinghouse, Duke Power and the NRC during which the breaker was e cycled "or 37-38 times. 'It operated successfully each tlas. Visual inspect on noted wear (nearly 3000 cycles of onration) and separation of the weld whhch attached the center pola lever to tw pole shaft. The NRC issued InformaSion Notice 87-35 on July 30, 1987 reporting this event. INVEST 1dAT!0NRESULTS The breaker was subsequently shipped to Westinghouse where a detailed investigation following the guidelines jointly developed by Duke Power, the NAC and Westinghouse. The breaker maffunctioned after some 130 operations. J 9 8 4 a 6 k l 0 ---n.- n,_,,_

( I I Se,7tembo'r 11, 1987 Page 2 After observing the condition it was found that the jawning could be repeated by manually forcing the close cam and main drive link into a unique l constrained position. The breaker did not assume this unique position on its l own through about thirty subsequent operations. The scenario at McGuire can be explained as follows: The roller attached to the main drive link normally rests on the outer close cam laminations. The broken veld pennitted lateral movement of the main drive link which moved the voller close to its tolerance limits. In the jammed position, the roller had slipped off the outer laminate of the cam. The force exerted by the breaker cicsing action induced a twisting motion which caused the roller to we' ge d between 'the close cam lamination and the side frame. Although it was established that the stacking of part tolerances pityed a part in the jam:ning of the breaker, it was also concluded that the breaker would not jam unless a broken w, eld was present to permit the twisting action that allowed the roller to wedge.. Subsequeht evaluation of the broken weld revealed that the weld had about 25% fusion. 'The mechanism producing the weld separation was low cycle fatigue with tho' fatigue striations indicating separation after about 2,500 cycles (consist'ent with Duke.'s estimate of operating cycles). A conservatively calculat'ed load on the weld was determined to be 10.000 psi. The designed weld str'pngth is 35,000 psi giving a ' safety factor" of 3.5. POTENTI Al SAFETY IWPACT s Westingh'oute considers this malfunction of the DS-416 Reactor Trip Breaker to DS-416 breakers have operated through many thousands i be a random occurren:e. of cycle's without any malfunction similar to that reported at id:Guire. Despite )000 cycles confirming inat the weld as designed is conservat It about 3, was also,, evident that while it is necessary to have a weld separation to initiate the occurrence it also requires other part tolerances to be naar maximum.; For thes'e reasons, Writinahouse does not reconsnand_that any i==aditta metWe be taken'. This, however, does not preclude recommended actions in line with ~ .horual surveillance a W aaE ti Wnce practices. RECOMMEGED ACTIONS' i 1 Primary ttention has been focused on 'the nld separation with contributing roen tolerance LJild-up. Because Westinohouse performed a random _ factors inspection of the pole' shaf ts (welds) during manufacture and because one i l I I t 4 l __ j

t 4 Septer.be'r 11, 1987 Page 3 I i i_nstance of the roller rubbina the side frame surfaced durina the,. inves tigifi6n~We s tinghouse. reconseMs_the_f6TTEjns_actiaan far it. gpl.ica_t,Lon,.s of DS-416 switchoear: a A.jShortTernInspection(NextSurveillance) IWeld Insee'et on (On Three Pole Lever Welds) 'This inspection may be performed with the breaker disconnected and ' racked out fully on the cell ratis, or on a bench. as is suitable to the user. Minimum tools are small mirror, fillet gauge (1/8" and i ,3/16"), flash light, screwdriver, socket wrench and long handled l pliers. Proe'edure 1. Trip the breaker if energized and closed. Rack it out on cell rails fully extended, or transfer to bench. j 2. Remove front panel. 3. Disconnect motor leads, and the link for the auxiliary

switches, i

4. Remove the top cover towards the front of the breaker, i making sure tsat wires ir the harness are not damaged. l 5. Inspect the weld (s) visually to the criteria given below. '6. Reinstall all items removed or disconnected. Criteria and Actions 1. Wald Separation Action: If separated welds are found, remove frors service as main or typass areater. I 2. Cracked Weld _ For checking the presence of weld cracking, exclude the ends l which may show evidence of cold start. Action: If cracks are found, use only_as bypass breaker _ until weld condition can be corrected. i 4 6 4

1 I Septembdr 11, 1987 i Page 4 l l t i 3. Size and Length of Weld Exclusive of the ends of the weld, which may show e'vidence of cold start, the weld should have at least 3/16' fillet for 90' continuously around the pole shaft. If the fillet is under 3/16', then the weld must be at least 1/8' fillet for 120' continuously around the pole shaf t. Either size weld provides a ' safety factor" in excess of 1.5. Action: If dimensions are not set, use only as bypass breaker until weld condition can be corrected.

8. ' Long Tern Ingpection (Next Rsfueling) i 1.

{xamine Welds for Separation, Cracks or Size Inspect Nmainder bf pole shaf t welds ~with'the exception of stop levers which do not perform a safety function. Replace pole i l 1 shaft if necessary. ' 2. AlignmenkofBreakerNechanism Refer to Fi ure 1. This tolerance check should be performed on 1 l the bench w th the closing springs disconnected from the cam-shaft (common shaft going through the close cas). I l Frocedure 1. Remove front panel of the breaker. 2. Disconnect the closing springs from the cam shaf t. The other end may be lef t undisturbed., 3. De-energize control powers to the breakers. if wired to power supplies. Breakers should be open with springs discharged. a I 4. Restrain the WTA with a wire loop so that the breaker is not in.a trip-free er.de. 5. Simulate manual charge of the closing sgrin s to the charged position, to turn the 'close can to the Rea to Close" position. I i i s

7 Septe..ber 11, 1987. - Page 5 I t ~ 6. With pressure applied to roller as indicateo in Figure 1, slowly turn the closing cam manually by the a eing charging handle. (Note: To release the can to turn, press both manual tr$p and close buttons simultaneously. Continue to turn the cam until the breaker contacts reach the closed position. At this time, the maximum lateral play of the roller is in effect. 7. Through the front of the breaker, sight the close cam, the roller and the side frames. Using a flashlight, check to see that - a. roller is making contact with the two outer laminates of the close cam. It is not required to be centrally p1 aced. b. there is visible gap between the side frame and the roller side at each end of the mechanism. If either of the two checks are not satisfactory, contact Westinghouse. 8. Reinstall all couponents removed. !OtherSwitchaearModels I Other switchgea'r models which utilize the identical pole shaft and mechanism s uld also be inspected. 1. DSL-f16 and 05-420 Inspection schedule should be identical to that outlined l abov, for DS-416. l 2. 05-206andDSL-206 I Since the stresses on these welds are considerably less than thost on the DS-416 application,ded that all the above(resulting in a much large ' safety f6ctor"), it is.recommen inspections be accesplished at the utilities' convenience in a time frame not to exceed the next refueling outage. l t h 9

DOOG Pv0 i l l i ( September 11, 1937 i Page 6 j CONCLUSIONS Westinghouse believes that the above actions are prudent and when accomplished on a one-time basis wl.ll provide assurance that a similar :frcumstance will not be repeated. I ~ i Sincercly, i H. C. Walls, Manager 6 Wid-America Region Projects Department Attac 'nt - Figure 1 HT/3277 cc: G. J. Plial T. A. Rieck F.'G. Lentine J. A. Johnson W E. 'J. Fuerst D. L. Farrar J.I.Usem)! J. Marianyt WOG Rep. A s i ~ i 1 1 i i a l t 1

l FIGURE 1 I i I s MECHdNISMSIDEFRAME ~ o g MAIM DRIVE LINX l l c. L. Ax

1.  ;/

/ MOVE ROLLER LEFT ROLLER CONSTRAIN!Nk 4 RIGHT g LINK tl y s. q-g ~ i g ,4 i N N + I INSPECT FOR VIS!BLE d INSPECT 7 '/ FOR VISIBLE, GAP j GAP l i CAM 1 MAIN ROLL / r r l i I I l j 4 i / I t G / g e e f .i l p i t- / = /0Y / FRONT VIEW POWER OPERATED (STORED ENERGY) MECHANISM i l l 6

%f Nh Septenbar 28, 1986 'Io: .7. P. M \\'ickar Frtrn: D. A. Wright subject: CO&P 450 Volt Breakar No. 52-1108 Nm 3894 turirg 2tutine maintenance of the subject breaker, it was disocuared that the veldn.snt connectirg the contar pole levar am to the pole shaft war, broken. 'it.e pole shaft with the broken weldment and a pole shaft rem:ned frtan anothar breakar were sent to the Materials Engineering and Analysis Unit (MISAU) for evaluation. A visual examination of the failed breakar indicates that the failure occurred on Am Ho. 3 shown in Fig.are No.1. At hipar magnification, shcWn in Fig.tre No. 2 it is observed that the fillet weld separated fran the lever arm at all but a small area at the start of the wald. cide along its er. tire length. '!he fillet wald rinnained attached to the pole shaft [ Darr.ination of the separated marface reveals that approximately 70 per:: ant of the wald had not fused to the lever am. '! hose unfused areas have a flat, anooth and relatively featursless surface. 'the runnining 30 parount represents the areas that fractured our aparatim of the breakar. Macroscopic exar.inaticr) of the fractured areas e a rough woody texture characteristic of an overload fructure in a wald. All fractured areas have a similar appearance and there is no evidence cf beach marks iruiicative of a fatigue fracture. 'therwfore, we assme that the fracture of the fused areas was the result of cwarload. Since the cyclic operating lands of this katakar are mamaned to be consistant the failure unst prttably coeurred during the first cycle (s) of operation. 'the failed lever arm was ocnnectai to the anti-totational lwer arm beside it, therefore, the load was transferred to this levar am after the failure and the breakar ocritinued to aparata. A measurenant of the fillet legs indicatas that the leg on the pole shaft side was 0.30 inch and the leg on the lever arm side was 0.10 inch. ' Ibis mismatch of leg sises of 3 to 1 and lack of fusien as the lever arm side indicates an ispgar wal tactinique in marantactuts. We maapact faum the wald W p.clogy and spattar that arms were mwd to the pole shaft in producticm using ons Metal Arc Walding (eguf) process. It appears that the walder did not preparly position the electrode in the $oint. 1he alactrode was positioned more toward the pela shaft zesulting in lag mismatds ard lack of fuelen on the lever arm side (see Figure No. 3). A nondestructive ameninatien (MtE was performed ces all pole shaft to lever arz walds for both pole shaft wiias)'sent to ManU.1he MtX ocrisistad of vimal emutinatirm ard wet fluorescent angnetic partiola.ausadnation, masults of these esaminations are ahtam J.n NE hsport No. 85-1445 included as Attadment No.1, and are mamarised as follows: 1. Visual enemiatics) indiastas that in general the starta af the walds are not fused properly to the lever arm side. Vimal enemination also indicates that in general a fillet lag mismatd) 2.

exista, s

I

g Magnetic particle examinatien inilcatas that with the exception of the 3. start, the welds a@aar to be fused to the levar arm side. 'Ihe ocrditico of the unfailed welds were probably adegaata for intended service, hcuever, flaws present includirg the failed weld irdicatas that there es Anadagaate centrol of weldirg durirg production. To further investigate the problen of welder techniqJe a section was rwcVed fra the cantar of all unfalled welds. 'Ihe sections were greurd to a 240 grit finish and the walds were ancroetched with anrnoniurn parralfate to reveal the dep*2 cf fusion. Depth of fusion measurenants were made for all sections and the results recorded in Table No. 1. A photograph representative of Walds exanined is ahcun in Figure No. 4. 'mase results zwveal that the depth of fusion en the pole s. ft was M significantly greater than the depth of fusion on the levar arm It is irlportant to rota that the depth of fusion does not dotarmine the adegaacy of the ;oint. If ocuplete fusicm is present the joint is edegaata. 'Iha results iniicate that the welder's technique was poor, but in most cases thart was adegaate fusion cutside of the start of the wald. We canedu$e frm the analysis that the failure was caused by lack of fusion of the wald as, a result of impxtpar walding technigas. Inspection of the welds on both assemblies sent to MEsAU indicatas a potential exists that scre failurus may have occurred or will occur cm the pole shaft assemblies new in servios. ME4AU roccan-mands that prorisions be unde to rce-es.Mvaly examine and/or repair all breakare that ars new operating with similar pole shaft aseenblies. We believe the most afficient fix wculd be to fulet veld the backside of the joint. For scre details on Wald repair ocmtact R. E. Cantrell cm 787-5505. ./1<} DLZ

7. A. Nritit Approved M/

rkMllPldi1 M VP'W attactmenta oct C. M. cruse W. J. ld J. A. crunkletcm K. A. G. R. ht1runn T. I Sydnce Fila No.t 86-30-038 l I w-,-

Ii a p. s l Figurs No. 1 3WrW of Pole Shaft Assembly I )pi$nthiIiiih$ ninon .oppinpninpn illill.l Ild.1 2 3 4 ,= e 1.l.i.tlt.1111..l.lt 11111111111111illIJ.li 1111111 ( $Y. ' ' .,.?f r.. 9 I %2 Werm the zeer Armw eto. sten. ear = m.

==t,e,.rti=or 1

1 1 l 1 TWe A.r.on l / s \\ 3 I W (5) l l Fiv e No. 3 Time: Wald Profiles Shcuirg Accerkle (A) an$ i t#aomer41e (3) Profues l l l ie y: ..g ' 'f,.3 f - J 5 :j. i., d...it..,- s... t.,. .r.4: ! % ;.c o t - ~. f o I .. b.,T ,.i* ',fi,i.,... '. .s, i'1', ;r.. .,s-i .g '.. *,*s.._.t. o. ta,,.,s, u.:g.e. l s..f, ', lf,\\ p. gI V,.' ::. T ..y y.,.. f - -,r y,. ,, l s j .f. 'g,, ' l' h. J.i . l t.i l'.. .c

• .~;-

I a o l e l .'. is. } f j 3 -l. I. l [ l ~ 1 l l j Figure No. 4 A Talet Wald Profile Wtive of the M j crees sections I i l b ) 4

O %s.wme.n1 t4:.1 0 i 'r. 65l449.- t no. _ NONDESTRUCTlYE TESTING REPORT

• ' 'ky..' 3 i.

g 5870-2.003[Ak$.b8N M. R. No. Q. C. Cell No._8e ' 3 Id 0 W. A. T. No. Al A Sy a tem No. 8 C-/- 01ce ND c.4 I 8 vnt 3 De t.

se,l e a-at n a &

.*Y- ~, pl.., E.L,s. A t,\\: G r Compement %eUw 0k 6L A s Co.,.nen, t...eti.a T 5 9 c. As.k e,..il 3 Em

c. J A ~d.,.w s M MT 2.e~

4,ee iomi ed Mr. 5 L. 4 4 L t --4 b 7 ikt d s 4 4,,e a. ; ; L

t. b_ s. h e4 Ta'ild

~ M Me,or.elT8n.--w al.2 Met'lTemp. 7P F Precedwre Ns.

6. t o L Rev.

3 Svefece Conditlen W.vl b.e f M C,k a-- ,G r., J t) O visk (v.T.) Dole shek No. \\ O Ultroshrs (U.T.) golibestion Rece'd No. / / Dere Shes' No. OEddyfwerent(Y..T.) Date 58h No. [Colibre,sen Rece d No. [ [ Q Re[egraphy (R.T.) Dem etNo.. Technique $heet No. ,/ ~ EMegnetic Particles (M.T.) Deve Sheet No. 86 l'Nf" A28tenderd No. E c' C.A.ste Mothed: [ Continuous C Residwel / Current: [AC C DC C HWDC l fewtpseent: ICeil C Ye4e O Prede O Centrol Condwstw [ Heed Shot C Pwm. Megnet l Particle.: C Dry [ Wet / C Yisible hluwescent / Color: gde b,N ' M Liquid Penetrer/t (P.T.) DateSewefNo. / Pw>trent TrPe: Olelvent R*aenble f C Water weshable Q Pest Esiulatf4d ~ O Visible Q Fivwescent / f enetrent - Developer tend Heme:, Dev leper Type: O ry C Wet CQNonequesea \\ a4 teiv wo. C - Enporetion e \\ Penetreat \\ Peee nt - Dwell fle _ \\ ___De mi per \\ I Devel at Time \\ \\ Cl*ener \\ N.. ite. e i, d:li di No. it s Aese),d: J.-a1 N., item Reiseted: ro aat. No. Items utth Reeeedsble ladiestiens > 0 Romerk :,34 Yh! hek %d, = an-- e r' Eaesil.or ) ab LevelT Report A* vie e Dete 4 ,04,,,_ u R.ee, e J u w W& .f CNece ame.14 93 -l . 's ' < s t. g

-o V N w 4 g O I. i !A m l sv d ':6,' O 1 h 'Y I 1 t ~6- $k 44 &Y e

( C _. _.. ~ htmWr IJ>.1-He 66 Idd$ 0 ..,.c . z, p BWeiDATA 5HEET' .,2

u. Qsb 70 - teos d: " '

w w s.1o2 .M'i wo e n.,,x.._a s. m u r. a c-p ..J. bd'% 9ok 5L.4 Y, J. d

e. p.

. M.. p.f..--<.J p y.. 6 %da Ps' I ' d i

4. id.~.

J of L.L ~% ~. \\ c.) d. R M o(' e, M,.( R. re-, 4 s - I a ~,.<.J, W 1L M )$. x G,'4 .hd sq/r,;n'.f4, 4,C e 9 w I-Ad,oog.Au,( mt. m j, 'y is vak c4

4. 4k 4,e,( g.4,A.2. g v-h-eA. wi. Q.

.'..J.4 H.. .A,4 pk r M '.^'* L r .n .sr- '.D J d v k,ey, k a gl > -frk (k.r :. k q. f & g.... g y q ;, + y....... q .a. p ~. A..k..:, 6. ".4.- 3..Mi ....i +- e.hqQ S /, -';.~"B...*-4. %:.. w. r.,.g s. ~; &.m 4 "t. v. b, s n e.3.... 1,.*. .. r.a A>.,o,;!. u...g '11 t' .....P+::k, u '.s m a .. x,:,: ..r 1. s.-. llc..;. g. . y. y *. p.

a.... :...

• :. t.

"Q:.*. n.. l l

~

r.

\\ r. o . s,g ':' )I.

• '.~,....,. y ;

.A..,?,,..... :p{-l, ..,4 ~. ..>,n. H.!. a, ., e .i,, s,- .~; ir 7. ; -.PJ:f r,-{ .t - /. m. y.r. "e,,. * (. q, ), p. ,. g s .g ..s . ',. ':\\*: p..:t,s :t r.. t t.1 q 4 ?,c, J: G. .l.:pjy '; ,. :.; ~. ..,,.g.-, .. ).* I

• k.., )'

? @.. ^. o. ..... (;..c.f:i-.. ;..... :. ;. :, a... +. '.. 14 .'..T,... '.';,;.,,. g .1,- t. .. 4 .4' :, ..h

y.

. u.. 3....: c.

s...

. - )... w ,. ;;/c,..g,. . a.., ' mt '... i. .....s i.. i t.;N, n. .i..p, l: e,.,; . * <yW,'.p.s e.o !..X,...%.pri. 4, 1 .w u.a .f.N.c N, i .: :, t.I -4.,. c,.,

p..,.,, ',

d yt e ? .y.*,,.. m. . n.. 1, , u, u.: :. t. t 4,, .. m...

...

. r.,

+,s.5..P ~.o. n. s1 ~s ?o, 1,, 5'/s-T.

1. 8&...8W. d.. *

. L . 1 e > - vA. e 4 1

.-v... .xmmmer. O~. '.. w'.t. x . : qcra.. ~, - '.:.if g.;fWk:-M~.W'-3, W E'> Q"'k i ~ ' ' '4 e~

• z l

^ ' ' ~ n nl~ R H n r. y n w M &&wn i c g:. n 7... TY 't -t c'ni s* y, m. 9 \\ my .<a 'MV-F~ M IP" 4 'IP P 7 7" " Y F p men p p,*,g.n ... se{ ve, } 'itVt ~ v3 y-+ m.y.d t~~ g m <w w > p =4 e sp s~<" stdS t r y p p p r*g pa~,= u - g g .t y je p g.<y p@ t~^ ".'f ~ ' ' **fe'mM3~y c'14r*e:'p'yc*nyyy W <e* y ~ ?yr'yed '% %on o %,y. A 9.g *, y~A n4 m it\\~,t w, u.a m-s y e, \\ ,..,{ i f e. y,.r.n w s~ ~ y ,i 9 ~ ~ e g 9.s p o y s << q p ' f * ^ t " o s.' S. K.s zy n 3 p 3 g 4 -Yo p(wg < ,,,y. e-, %,,f yd <-~n py ,,- +<. c...pa.4 n2.,s f'y~~'* $Tpd 7 P~"W 0 syy.f 2 p$ w m ta yy y. y /itew a + v nm ~ e*a + +o n s y p i, cmnp p-4~,s.r7m es2 y y n 5 7p., -,9.). M., c hy. -r. n 3. 4.- <.n, c.g p~. -. n .r g 3., p y 0. ..y~ wg y;<q; tPW Wd T"4 g n i.gg n ~.n un ..u.n...u gco2 -ozqs '"*uv 133HS Y1YO 30M 9 Shnt S8N 3 4 pwgy

Table No. 1 Depth of Rision Maasuremants a. Depth of Pusion Depth of Fusico h Pole Sidt- (m) Inva* Arm E4da fmi 1 2.40 0.1 2 2.0 0.5 3 2.25 1.0 4 2.25

2. 0 5

2.0 0.3 6 2.0 1.0 le 1.5 0.3 2* 1.5 0.2 4* 2.25 0.4 5* 1.0 0.2 6* 1.5 0.4 = T-; 4 . -.j. p 1 N: f<;,.

b.. ',

c. .2 ,a

e., :

.r t.,. - 1 **,r.: n - m.u.. c%..h. y;,9 c..... yn- . s e

• .pi

.nh (. ?f eA ..c .,4

7. 'r M zw,(gk;<%,. $f.y;.W. ]Q-}...

b b g,y** l:' ..R ~,4 ' ?"e.s u,,a;o* v . 3 {

• ,L,

& g;LM'

d> 4 f. '

• a.~

~ l .,3.. '

f ENCLOSURE 8 "Suasary of September 23, 1987, Meeting on Westinghouse S w i t c h g e r,t failures," 2 October 1987 Enclosure 2 (Sequoyah reportl t em e fw h' e

~ MEETING SUMPARY DISTRIBUTION Docket File NRC Participcnts NRC PDR D. Hood E. McKenna C. Berlinger NSIC E. Merschoff PRC System J. Stone PD#Il-3 Rdg C. Y. Cheng Project Manager D. Hood A. Toalston M. Duncan W. Hazelton

8. Kolostyak C. Sellers W. Troskoski (MNEB 6113)

J. Richardson OGC-Bethesda K. Naidu ACRS (10) K. Wichman B. McIntyre, W S. Hou J. Jelovich, W V. Hodge S. McNeil T. Peebles S. West e e r I l t

a ae% [o o UNITED STATES l NUCLE AR REGULATORY COMMISSION i j WASHINGTON, D. C. 20555 a '%,,,,, / october 2, 1987 MEMORAf;DUM FOR: Kahten N. Jabbour, Acting Director Project Directorate !!-3 Division of Reactor Projects I/II FROM: Darl S. Hood, Project Manager Project Directorate 11-3 Division of Reactor Projects I/II

SUMMARY

OF SEPTEMBER 23, 1987 MEETING ON WESTINGHOUSE SWITCHGEAR FAILURES (TACS 65955/65956) On September 23, 1987 the NRC staff met in Bethesda, Maryland with Westinghouse i to discuss the technical basis for Westinghouse's recorzended actions, criteria l and conclusions in its letter of September 11, 1987 (Enclosure 1) to utilities of the Westinghouse Owners Group (WOG) that use its Models 05-416. DSL-416 05-420. 05-206 and DSL-206 switchgear in Class IE service. The letter is based upon the D5-416 reactor trip breaker (RTB) which would not open at McGuire Unit 2 on July 2,1987, because of a failed pole shaft to center pole lever weld (Referencesa,b&c). Related breaker failures at Calvert Cliffs Unit 1 (05-206) in September 1986 (Reference d) and at Sequoyah Unit 2 (Enclosure 2) in June 1987 were also discussed. Meeting attendees are listed in Enclosure 3. I. WESTINGHOUSE PRESENTATIONS k!stinghouse provided presentations in 5 areas (Erclosure 4): (1) Design Adequacy of the Pole Shaf t-lever Assembly ~ To show that the design weld is adequate, Westinghouse described calcula-tions showing th6t the working torque of the 3/16" fillet weld of the 3/4" diameter pole shaft exceeds the peak cticulated actuation (closing) torque. Calculations of the failure torque were perfonned to show that the pnle shaft would deform before the design weld. The calculations were confirmed by three tests for the failure torque of a shaft lever. Tests also showed that only 10% of the length of the design weld would still carry the actuation load. Other calculations were described to show that fatigue would not occur in the design weld. (2) Inspection Program Westingbouse described the failure torques resulting from tests of several 0.19" and 0.125" fillet weld specimens of varying lengths used to establish its inspection criteria. A general discussion of desirable and undesirable indications for visual inspections of fillet welds was presented. N/h0$OYl?" 5h$.

} 2 (3) Design, Manufacturing and Test Criteria kestinghouse described the various uniform industry standards used for the design and endurance tests of the breakers. Westinghouse has performed about 90,000 total breaker test operations on various DS-416 tests to satisfy different requirements. The highest number of operations on a given breaker was 10,964 and occurred during WOG tests. No failure of a pole shaft was known to have occurred during any of these Westinghouse tests. (4) Westinghouse's Comercial Dedication Process The breakers are manufactured as a commercial-grade device and subsequently certifi;d (dedicated) as qualified for use in a safety-related application. This process of cocinercial dedication is presented in Enclosure 4 (5) Quality Assurance Process for Procured Hardware Westinghouse discussed the QA process for vendor control and vendor vograms at its East-Pittsburgh, Puerto Rico, and North Carolina facilities.

11. NRC COMMENTS AND CONCERNS The NRC expressed several concerns during the meeting. These were not resolved.

(1) The staff disagrees that a RTB with a cracked center pole lever weld can he used as a bypass.ireaker. Such a breaker should be considered inoperable until the wid is istinired or the pole shaft replaced. (2) Cracks in the fillet w Id may be missed unless the pole shaft is removed from the breaker for W s visual inspection. Also, the zinc coating on the welds can mask fabrication cracks. (3) Weld inspection criteria and qualifications of the welding inspector are not included in the letter. (4) A weld "safety factor" of 1.5 is not sufficient for Class IE application. ~ (5) Westinghouse's reconinended actions, in essence, accept as satisfactory a weld less than the requirements of the original drawing. Such a weld has not been tested to ANSI standards. (6) Westinghouse has not effectively comunicated the endurance limits of the pole shaft to its Class IE users. (7) Westinghouse's letter was sent only to WOG users. Other utilities also use these breakers for Class IE applications. All DS Model Class IE breakers need to be inspected. (8) Westinghouse does not have records for individual Class IE breakers. Its records are based upon a given purchase order which usually covered several breakers ordered by a utility. The number of breakers per purchase order is unknown.

f . (9) Considerable peening of th, close cam surface surrounding the'stop roller was observed to have occurred at McGuire Unit 2. Binding of the stop roller inside the close cam could impede closing of the breaker. This is of concern for those Class IE breaker applications in which the safety function is to close, l (10) The staff remains skeptical about Westinghouse's weld stress calculations which are based upon static rather than dynamic ' factors. Westinghouse suggested and the NRC agreed that fatigue testing of known weld lengths would be useful here. (11) Licensees are not required to report Class IE breaker failures to Westinghousat or the NRC. Westinghouse had limited knowledge of the Calvert Cliffs 1 or Sequoyah 2 failures. The NRC learned of these failures by chance and after the McGuire failure. !!!. STAFF CONCLUDING REMARKS The NRC staff stated that in view of its above, oncerns, consideration will be given to separate regulatory actions in order to assure that Class IE breakers are designed, fabricated, erected and tested to quality standards and will operate comensurate with the importance of the safety functions perforired. IV. REFERENCES a. Licensee Event Report 370/87-09, "Reactor Trio Breaker Failure Oue to Mechanical failure " McGuire Nuclear Station - Unit 2, Duke Power Comoany, dated August 3, 1987 b. NRC Inspection Report Nos. 50-369/87-22 and 50-370/87-22, dated August 31, 1987 c. NRC Information Notice 87-35, dated July 30, 1987. d. Memorandum from B. A. Wright to J. P. McVicker, "Calvert Cliffs Nuclear Power Plant 480 Volt Breaker No. 52-1108-NCR 3894", Baltimore Gas and Electric Company, dated September 18, 1986 (same as Enclosure 2 of Meeting Notice, 9/21/87, D. Hood). ALhoo Darl S. Ho , Project Manager Project Directorate 11-3 Division of Reactor Projects, I/l! Enclosu res : As stated .)

g

• • 1

,/.f. i l i l l i 1 l The following information and recommendations are provided for yor use. if DSL-416. 05-420, DS-206 and DSL-20C i you havo Westinghouse Models 05-416ewitchgoarinstalledinyourplant i 8ACKGROUND On Julyh,1987, it was remrted that a 05-416 reactor trip break Jnit 2 during rod d/op testing following a refueling i ~ This malfunction was detsrained when plant personnel observed smoke l 4 open on Since the breaker had not outage. in the yi:inity of the reactor trip switetgear. ina i opened $n demand the shunt coil current was not interrupted resultind tr p when i the mangal charg ng handle was manipulated.pped manually damaged. coil. The breake* could not be tri During subsequent cycling on the test bench, the reaker jammed again. An inspection was conducted at the site during which the breaker was jointly by Westin house Duke Power and the NRCratedsuccessfullyeachtime. */isual -cycled for 37-38 ines. It inspection noted wear (near) 3000 cycles of o>eration and separation of the weld wMeh attached the center pole lever to t u pole haft. The WRC issued' Information Notice 87-35 on July 30, 1987 reporting this avant. INVESTidATION RESULTS i The bre6ker was subsequently sh to Westi house where a detailed inesjointi developed by Duke Power the l investigation following the gutThe breaker malfunctioned af ter some 130 operation i NRC and Westinghouse. ) i d. ~

---

.m_.-,_..c-m--.-mm_.

.... -.--r -

=. r I I Septembek 11, 1987 + Page 2 td Af ter obrerving the condition it was found that the jaaning cou by manually forcing the close can and main drive link into a constrained position. own through about thirty subsequent operations. The roller attached to The scen'ario at P:Guire can be explained as follows: i The the main drive link normally rests on the outer close cam laminat h moved the broken weld permitted lateral movement of the main drive link wh i roller close to its tolerance limits. The force exerted by the breaker ller to Va'dge slipped off the outer laminate of the cam. closing action induced a twisting motion which caused the ro Although it was between the close cam lamination and the side frame. t in the jaming established that the stacking of part tolerances played a par jam unless a of the b'reaker, it was also concluded that the breaker would not h ller broken weld was present to permit the twisting action that allow to wedge.. d about 25% Subsequeht evaluation of the broken weld reve fatigue ( 2,500 cycles with the' fatigue striations indicating separation af ter about A conservatively (consist'ent with Duke's estimate of operating cycles). The designed calculat'ed load on the weld was determined to be 10.000 psi. 3.5. weld str'pngth is 35,000 psi giving a "safety factor" of POTENTIAI. SAFETY IWPACI i Breaker to Westingh'ouse considers this malfunction of the DS-416 Re d at McGuire. be a random occurrence.of cycle's without any malfunction similar to l d for Despite the quality of the weld in the McGuir tive. It ld separation to was also'. evident that while it ir necessary to have a l s to be near l l' iinns For thes~e reasons, Wejlingho_u_se_do_es_no maximum.; s T be taken'. practice:. . normal _s arniUsattJnfmathtsE4de6 ? RECO64t@ED ACTIONS ' ', ting ttention has been focused on the weld separation wit d a random _ I ! e one _ Primary inspectipn of the pole' shaf;s (weli5s) during manufacture and rom tolerance build-up. ' factors l j \\ I t

1 I Septembe'r 11, 1987 - i i I Page 3 i j. i I instance of the roller rubbino the side frame surfaced Jurina tha_.. for 1E inves{igaMon Westinghouse.recoupeDd1_the_f61 Towing _atifont g p]1 cations of DS-416 switchgear: A. l Short Ters 16spection (Next Surveillance) IWeld Inspe'etion (On Three Pole Lever Welds) ! This inspection may be performd with the breaker disconnected and ~ or on a bench, as is suitable to ' racked out fully on the cell rails,l mirror, fillet gauge (1/8" and Minimum tools are - smal ,the user. 3/16'), flash light, screwdriver, socket wrench and long handled , liers. i p Proe'edure Trip 1.he breaker if energized and closed. Rack it out on 1. cell rails fully extended, or transf': to bench. i 2. Remove front panel. Disconnect motor leads, and the link for the aux 111ary 3. i-switches. Remove the top cover towards the front of the breaker, i 4. making sure that wires in the harness are not damaged. I Inspect the weld (s) visually to the criteria given belcw. 5. Reinstall all items remed or disconnected. 6. l Crite:_ta and Actions j 1. Wald Separation If separated welds are found remove from service Action: W ypass creaaer. as main i 3 j 2. Cracked Weld _ ) For checking the presence of weld cracking, excitde the ends which may asow evidence of cold start. If cracks are found, use only as byoans breaket Act16n: untti weld condition can be corrected. i i i t i p. s

s

t i

i Septorb4r11.1987 Page 4 l l 3. Size.and Length of Weld I Exclusive of the ends of the weld, which may show e'vidence l of cold start, the weld should have at least 3/16' fillet for 90' continuously around the pole shaft. If the fillet is under 3/16', then the weld must be at least 1/8' fillet l for 120' continuously around the pole shaft. Either size weld provides a "safety factor" in excess of 1.5. i Action: If dimensions are not met, use only as bypass breaker until weld condition can be corrected. l l l B. LongTeraIngpection(NextRefueling) l 1. Examine Welds for Separation, Cracks or Size Inspect tenainder bf pole shaft welds ~with"the exception of stop levers which do riot perform a safety function. Replace pole shaft if necessary. i i i i ' 2. AlignmenkofBreakerMechar. ism Refer to Finure 1. This tolerance check should be performed on i i the bench with the closing springs disconnected frca the cam-shaf t (common shaft going through the close cas). I I 1 Procedurs 1. Remove fr>nt panel M the breaker. f The 2. Discpnnect the cle:1pg springs from the can shaf t. other.end may be lef t undisturbed., j l 3. De-energize control powers to the breakers, if wired to power supplies. Breakere should be open with springs discharged. j i Restrain the WTA with a wire loop so that the breaker is 4. not in a trip-free moda. rings to the charged SimulatemanualchargeoftheclosingagReadytoClose" 1 5. position to turn the close can to the position. j i i s ^

~ i i i c Septeeber 11, 1987 Page 5 With' pressure applied to roller as indicated in Figure 1. 6. slowly turn the closing cam manually by the a ing charging handle. (Note: To release the can to turn, press both manual trip and close buttons simultaneously Continue to turn the cam until the breaker contacts reach the closed posttion. i At this time, the maximum lateral play of the roller is in effect. 7. Through the front of the breaker, sight the close car, the 1 roller and the side frames. Using a flashlight, che:k to see that - roller is making contact with the two orter lamleates of a. the close cam. It is not required to % centrally placed. l j l b. there is visible gap between the side frame and the roller side at each and of the mechanism.

!f either of the two checks are not satisfactory, contact Westinghouse.

8. iteinstall all cosponents removed. 8 Other Switchoear Models l Other switchgea'r models which utilize the identical pole shaft and mechanismsh9uldalsobeinspected. 1. DSL-jl6 and 05-420 Inspection schedule should be identical to that outlined abov9 for D5-416. 2. 05 2Q6 and OSL-206 l Since the stresses on these welds are considerably less than those on the DS-416 application,ded that all the above(resulting in a much la

• safety factor"), it is recommen inspections be accasplished at the utilities' convenience in a time frame not to exceed the next refueling outage.

I l S 1 1 s Y

voou v u /. Septembe;r 11, 1987 1 i Page 6 l t CONCLUS!0NS I Westinghouse believes that the above actions are prudent and when accomplished on a one-time basis wl.11 provide assurance that a similar circumstance will not be~ repeated. I Sincerely, l f I f Manager H. C. Walls, Region i Wid-America i ProjectsDepartment Attac 'nt - Figure 1 HT/3277 T. A. Rieck cc: G.,J. Plial J. A. Johnson W F.lG.Lentine ~ E. J. Fuerst D. L. Farrar J. A. Usem )! J. Marianyt WOG Rep. I I i i 1 i i i 1 l

i A I FIGURE 1 l. i 2 1 MECH 5N15MSIDEFRAME MAIN DRIVE LINX t. / A.- As /' / / ROLLER LEFT [ g CONSTRAININb-ROLLER LINK I j f - l,, r i z' / sj e e w L N J INSPECT FOR VIS!BLE 6 ' [* l' A GAP ( .I 7 ggg' ' .P W MAIN ROLL / l r I s / i, 9 t l i I { l I l t e t e l 1 e -tu t% 6 f / /h/ t FRONT VIDi pcWER OPERATED (STORED ENERGY) MECHANISH t' ~ = _.

QA Record Enclosuro 3 L'N!TED STATES COVERNMENT 0710 028.. Memorandum TENNESSEE VAILEY AUTHORITY 870721S0629 L. H. Nobles, Plant Manager, POB-2, Sequoyah Nuclear Plant l To e J. B. Hosser. Project Engineer Sequoyah Engineering Project. DSC-E, fro.4 t Sequoyah Nuclear Plant JUI.101987 om s t'fu rcT. SEQV.,f AH NUCLEAR PLANT UNIT 2 - PAILURE ANALYSIS OP DtERGENCY TIRE PROTECTION PUNP CIRCUIT BREAKER PC1.E SHAFT ASSD*.tLY This memorandum supersedes my memorandum to you dated June 15, 1987 (225 870615 026). Attached for your review is an angineering analysis of the broken pole shaft assembly. The Dreliminary results indicate that the failure is not generic and is probably an iac, lated case. However, the Civision of Nuclear Engineering is coordinating with Westinghouse to review all weld documentation and procedures to make this c'atermination. After an evaluation of this information is completed, my recommendations for revisions w maintenance and surveillance instructions will be submitted to you for incorporation in the appropriate plant instructions.

1. J.a.Hoamer tbB PtHC Attachment cc (Attachment):

l RIM 3, S_lj_C-r \\

0. P. Cooper, 5-215 SbK l

C. T. Mc11. DSC-P Sequoyah R. M. Jessee W9 A60 C-K J. A. Roach, W9 C135 C-K l H. A. Skartinski, P0b 2, Sequoyah (Attn T. Eonovitch) R. C. W1111ama. DSC-P, Sequoyah i l \\ i DE05;H071&2.05

PRE 1.1MINARY FINDIHCS REPORT ABSTR ACT This report briefly covers the laboratory results of the examination of two _ failed fillet welds on the pole shaft assembly of a circuit breauer that ~ energises the emergency fire protection pumps located in the plant. It describes the plain carbon steel materials used, ytherosity.*the crater _ cracks, and the fillet weld size and concludes that more class i circuit ~ Dreauers be examined on a routine basis in accordance with a revision to Electrical Maintenance Instruction (D(!) 10.5 9d Surveillance Instruction (SI) 275.1 and SI 275.2 in. ster to look for__ porous or cracked welds. This is not a generic probles based on data to Da supplies rphe vencor, l Westinghouse and the observations of the Materials Eng eering section as j descr! bed in this Q rt.

• }

INTRODUCTION A pole shaft assembly with cans wided coto it was brought to the Materials Engineering Section for examination of two fillet welds connecting the can and the rod, which are the principal ocoponents of the assembly (see figures la and B). It was determined at that moment that it should be sent to the TVA nota 11urgical laboratory for further destructive emanination. PESUI.TS AND DISCUCSION X-Ray spectrographic analysis for cheoloal eeeposition shows a plain earWn

• steel rod and a ocrtosion rwaisting oopper tearing plain carbon steel plate

'for the saa (see Data Sheet 1). Wald metalacheetcal analysis by induction furnace combustion techniques and a visual eFaminatifn of the weld and adjacent metal for are atrikes, weld spatter, or alq ahow evidonoe of a alid steel esopo*1tf on, probably an AW3 SFA 5.18 5703-6 saa ahtelded arc welding tiller satorial (see Data Sheet 2). Average hardness values of the rod, war and weld were taken and found to be steeptable (Data Sheet 4). The two failed welds, as visually canained, give more detail to the mode of failure than ehesteal eenposition. Visible poreetty in shotos 2A and 28 i shows almost a 20 percent reduetten in surface area of the eroes asetten photo (see figure 23), whteh eeu14 be assumed to M a 25 pereent volumetrie reduetten of metal along the length of the weld Etnee the line of ffacture runs irregularly free veld te veld the entire 1erath of the weld. In figure ira, it to also noted that the weld was of multiple passes, as if they tried to *)um aut* tha mama t t v. Figure 3A is equalidateresting although evidence of pereatty is not demonstrated. The weld freatured Gown the center of the weld in a atraight_ line wish steoat equal volumes of metal dessaited en hath the har and tna plate. This led to the belief that while a sound weld was made to join the plate and red it may not have been of suffletent aise to earry the loaa that the assembly was designed to carry. After exastnin; new the pole shaft assembly _funettoes in the nippuit hankar. it was antamants that airter thi porwa weld fatted the see ne:L to it failed beesuae all of the l

t.s -4 ' i.,: p '; s j y 2 1_oad for the two welds was now placed on it (these two cams were connected at the end by a pin throuAh the holes seen in' fizure 11). 31nce the pin was connected about 1 1/2 inches away from the shaft, a twistina force would have also been placed upon it adcing to its shear and subsecuent rallure. The failed pole shaft assembly was repaired with a replacement from arare parts, and four more have been ordered. Any future welds encountered with eacessive perosity or cracking can either be replaced or reweided a'. the direction of the Materials Engineering Section. CONCLUSIONS After seeing several of these welds on existing pole shaft assemblies in the field, it is recoeunended that revisions be made to EMI 10.5 and l 51 27! 1 and 31 275.2 to look for porous or cracked welds on all circuit breakers using this type of pole shaft assembly. This is not a gensrit: probles based on data to be supplied from the vendor, Westinghouse, and the observations of the Heterials Engineering Section. ADD!?!CH4t ACTION REQUIRED the Materials Engineering Section will request Westinghouse to supply the l following 40cumentation, which further supports the belief that a generio condition dos; not exist with regard to these fillet welds breaking ca pole l shaf t asseeb11es6 j 1. Data about related failures reported to Westinghouse to date. i 2. Process and mate,'lat specifica.1ons relating to the fabrication of the pole shaft assembly and any applicable industrial codes, such as AVS welding standards, naod in its fabrication. 3. Engineering dat? reflecting the weld site and length, slid to be 180o around the pole and the 1/8. inch fillet weld site. 4 Alternate engineering data regardinj weld stae and length if said length and site on the pole shaft asseebly proves to be less than 1800 but greater than the specified 1/4.!nch fillet weld atte. l Upon receipt of this information, the Materials Engineering Section will 1saue a final report deta11&ng the changes to be made in the EN! and 31 procedures. t l Edward W. Pugh Codes. Standards. Welding, and NDE i l H0715).05 l l ( i

-.---m-._ .-a-m .=wa m_ _ _. -- mar.a-m aa m.as c - -u._ , ma .m.s a A l I i ( e I l! t I i t l i L I

• -_ __ q n:.

\\ i 3 c, EhfCOPY F18ure 1 A - Circuit treaker Pole haft hambly g J ft f ?i ? W w.1 i ~ n e g g v. s r. tt ') g e,s:s.3 gi i :=-l. - .k e? gl';.,.,=:rg;" . A ; 9' .'i. 3 k', l

.3p],, j 4,

s3

%..C) e,) s'. s ' E2' Gl'] i e. i @Ml, ~ t ripre 18 - Pois mart asambly Tractum ritm m l Fase 3 of a J i 'b

• 1 l

w.mw c, y.m

} <

?,. .p ~0 .i. i a rc...t. ,.Ff'6 = -1 i ED FROM BsST AVAILABLE COPY re u - vi. int. ror..it, in r.o.o rin.i v.a ~. Q.':7 M.u... E.?)f. g F,/.}$ < ' s. .. % J.'..n. " '. 7<. . D

y. 'r.. t'

., s. : . <:.;:_.; (,* n 4'.1,9 .;; :.'4 - .. ' j.+.. y' .w;., *g A'^*'. :,." .*Q.', c. cs.. ._' 1 '. '

9., q.... ws..:.

.,..,: y. ;. A.t *b..

• ?

'1 5

g.. <,d f '

w. 5 - t

g.,,,/ a, 4 8

Y f s.*....-ls P,q. s.... v. + r.e v y, c. : .. u e. .3 g.:..'. ( $l1 4. ' h yh N 3,#~.! Y'SN dpl$ $ ): ', $f,. v$,ec$.I.8

p...

2f,,...f. ' ygj5-Yg[I ' /O .s.. + -. ',[u;,.L.g f N-s. 7, + ' g.

4Q Y

4 Figure 28 - Y! s t b le Poroa l t y in Cros s Se c t ico of Fa t t e d F111st Weld Page 4 of 4 l

et ;i...., W ;-> # L. + --- . -(~ 1 , 4. t 4 e ' : '-.1.,(, ',; e,..;.'.,'N<.~S t- ? O g w c- [_ 'M O M%. ;W.i :wg"KP u -, s..em.gv.g.v. =.

g. w. W_, ~ 2_.. u. - q,...... y, -

g.- -w r r.*, ~. e y,w-s.u ~w: p~ y..... p-. ; y.s.a %. =W,.. .= ' -.sr (..- t .e ,ge - +, 4 i - 57 ...m ,.g,s,i'. %./. .pi,y j...3 e.. ., e, ,.-i. p. - E,- - ,N. ;l. Y,f fy Ti.' fel*:.N'l,k $' l' ,~ W, . k l,,,.e- [ T.'".p, ;rs4..;"%tJb j-r- % p. % j ', %.*,A.vi.4- i-3 ?

n.p * -

~p

• (%g.. ;3"..,,.,-

( g* 4 4, g ) .,s. j?* ,y$,, ,,* *, ' h.. L :.,

t..,y /. g t.[;[g41.g [

44 .* vh :. y,s t . A '.. ,8- .t,, a .(..'[.'. h,7: I K' f, * /4'.j.. . a- .e ,. s :.

• (- }, './,

v n , '.*~.7 r 4,.' l- , M' Q.l,. : - 7 ~ ?- -s 4,i -.... .,,y.,e... g '., g '*. - s n},.' a. :.%. - ' g ;..r'.1 '.. C /. e *. -w

1. ..~.

sef, m p' '-' '.- a.. .s . p ,.r.. L. w.~1h g s .c l - v-t %w :. A.;. 3f 6 = g

l..

1.,..Q..,. s.c it-

. g.

u...

r .' s s e; . *y. s. .*s..- . a. g

2.,g.,.,,g e.

c I F8sm u - ratt.a ritt.s wie.

• nie=.. or e.mts, FILMED FRfCOPY BEST AV AI'b8k hF r.

c a F r k u F ue S of 8 r

Itrc c 0F CHERICAL COMPOSITICW BT I tAY SPECTFOCRAPHrC ANALTsfs ~ 3tPot7 Not e bgtMEDpa0M v'/. 2//N/ v* BE5\\ 3"""ABLE CC.q Program Standard (s) ..Au t/ $1MPLE Wo. cue 0 ro. % Coppts, % 0.!)) fufl shy ? ItoN. % LCAD.% KANCuist, % bo Y D. $5 N0LTBDENVM. % ~ ~ ~ NICrtt. % FNOspHORus. % 0*bl$ b bll ,LICON. %

d. l f (lo b f

'~ f!N. % ~ ffTAN!W. % TANADIW. t 11NC. % CEWOttNT3: FILS 3.0. Amat,ytst 3r SDOA 4/8/86 Deta Sheet ! Paste 6 of 8 y .M

+ IEPORT OF CHEMICAL COMPOSITICW BY JNDUCTION FilR.ANCE COM5UST!oW TECHN! QUES Esport ko._ d ~d ' [,#! Standardts) ~ fl. t.Y - su,,1, u. Carben, t solfor. t E Analysed By .-n cROS g K b REPCAT OF CHEMICAL COMPo$f7!0W ST ENERGY DISPERSIVE I. RAT ANALYS!$' (80%) Report No. [(!/I!b Iasrele We. A, ta h.. 6, !.'. ' - ff %rl//r, b's

• D!< n.iis

/. 7 , / A,, : c/7 s 'EDI analFels is a seel. quantitative techalt'se which sees se steedstds, analysed By O g

{,- ....~. } AyrtAct na:Dwtss vat.Ut3 ROM I! i E g t.E C O P i' roi errorr wo. ' 't l EAMPLE 50., MAIDNE.SS ' g, /u<d. M 4-I /We Bt.r s l p,.4. jig 9 t v t l I .- = =. e t I ~_ l IIfLMCHIRLVf1L1112t l \\ Rockwell _,. i [ Seperfleist Microbardaess i et m ( 4/8/84 i l r t s \\ ..*( l .e e l l l Data Sheet 3 Fage 8 of 8 l J r r ~ -

w e Meetino Attendees September 23, 1987 4 NAME ORGANIZATION Darl Hood NRR/PDv!!-3 Carl H. Berlinger NRR/DOEA/0GCB Ellis W. Merschoff NPR/DRIS/VlB James C. Stone NRR/DRIS/VIB C. Y. Che:ig NRC/NRR/EMTB A. Toalston NRR/ DEST /SELB r W. S. Hazelton NRR/ DEST /EMTB C. D. Sellers NRR/ DEST /EMTB J. E. Richardson NRC/NRR/EAD r K. R. Naidu NRC/NRR/VIB X. R. Wichmin NRC/NRR/EMTB NRC/NRR/EMEB

5. Hou Vern Hodge NRC/NRR/0GCB W/GTSD/MT Cindy Pezze Barry Barnett W/NSID/QA Chuck Geis W/NSID/EOP Anup Deb W/NSID/EOP Dick Miller W/NSID/NS r

Warren Bamford W/GTSD/ Material Technology Joe Jelovich W/NSID/EQP Franklin Research Center Howard Fishman Franklin Research Center Gary Toman Westinghouse Nuclear Safety Brian McIntyre Westinghouse Nuclear Safety Pete Morris i

DELOSURE 4 Viewgraph Slides and Handouts for Westinghouse Presentation September 23, 1987 (

9/t>/y7 u-A-ys Design Adequacy of the Pole Shaft - Lever Assembly e Actuation Torque e Calculation of Working Torque e Calculation of Failure Torque e Determination of Allowable Weld Degradation

• Experimental Volidation e Fatigue Evoluotion e Conservatisms

Actuation Torque e Linkoge for Opening and Closing Contains 5 Wembers e Shaft Actuation is Soley Through Pole Shaft Lever e 1.ooding is Essentially Pure Shear on the Weld l l e Actuation Torque Weasured by Duplicate Strain Gage Rosettes e 12 Tests Were Wode, and Results Were Very Reproduceable e Peak Torque Was f 814 in. Ib. on Closing + 615 in.16. on Opening e Range of Torque Wos 814 in. Ib, to -692 in, ib. i f

. l.

SUMMARY

OF ACTUATION TORQUE MEASUREMENTS BRIDGE A BRIDGE B MAX MIN MAX ' MIN D1 Track 5 800 -700 620 -360 D1 Track 6 840 -600 615 -350 D1 Track 7 830 -710 620 -350 D1 Track 8 840 -730 620 -370 D2 Track 1 800 -730 610 -380 D2 Track 2 815 -705 610 -400 D2 Track 3 805 -705 605 -405 D2 Track 4 810 -730 620 -405 D2 Track 5 800 -700 620 -400 D2 Track 6 815 -700 605 -390 D2 Track 7 805 -6 80 620 -360 D2 Track 8 805 -620 610 -350 -@?!'7 AVERAGE: 814 -692 615 - 4.- I RANGE: 1506 IN.LB. 1292 IN.LB. i I i o I 1

", O

• 2 L

a lI 8I 1 9 *1 h y *I 1 A ie g.7 e .u 0.I w h', m d I " 8 'O A 7* H .. g.0 p.g g.0 0 *O a a o G s a G G G G G s G G G G G G G G G S D D 4 N N D D G l i I I ~ l 87-NI '3n0BO.L

",ga g.7 g.7 t 'I J P M g.7 .U LD 0 'I m >k, td I 8 'O s g.g p.0 g.g 0 *O G G G G G G G G G G G G G G G G G G G G G G (D to 4 N N 4 O CD G e i I i 1 l 87-NI '3n0801

Calculation of Working Torque T LtyER 9 I ~ CQ %#'4 +-3/16 \\ l 1 SHAM .750 j i I From AlSC Guidelines, For A 70XX Electrode l %w. = os % B M hl* I 1 I Working Torque: T = Lo. (v e) A. l e t . L B o 64.L%. [ Actuation Torque is 814 in. Ib. maximum, t. Weld is Adequately Designed. l ( t

4 Calculation of Failure Torque For the Weld: T = T...JaL,T n.'I R $. we,mssa jw. J 4., w to r " = 3 2. T,3 us i., u. T m., = YS ( Ty,,a= b : l' T,, 2aso iu.a. l 73 6 er the Shaft: Ts T=== T d' st. E 5'o l T" = W ' W T% '- 14o0 w.to. i T wa gy T,, i4 3 r iw. t. I i i l Conslusion: Shoft Will Deform Before The Weld f l I I t I

i Experimental Validation e failure Torque for o Shoft Lever r i e Three Tests Performed, Average foilure Torque = 4285 in. Ib. i l e for o Given Torque (900 in. Ib.), How Wuch Weld Degradotion Will Still Corry The Lood? e Result: 0.16 in. Length of Weld, or 10% of the Weld e This Resun is Consistent With the We Cuire foilure i l i i i I i 4 l I I r t 1

.__ 4 _.__ ___ _. __. _ g ..w.. .... - _- _ -...t. .l.__.___ _.. _.f j } _ j t } - :J -3 .-_ _1 ~ _ _ _ _, i , __.____4 i I i ' i g . ip -. 8 ..j. __ q -. 1 g --t b j. ..M_..- _ And.. ...j.. -.. -.9.__. .___.t~*~ ..4 3 _m l 4 __ .o. e J e i 1 4.;] j.., ~._ - 4, ; 7 ,,., j,_- -. y... .j 1, , 7 ) 9 j. .g 9... g 7 1 .._; ;_ u...,. _ ? t1 .a i _,._, , _. 6,

• M-

.m ~ -. ': :.- h.. !.1 ; ' : :' ' i i 2 ~~'

h. '

• • • • k N I

_J l ^! i 3 i.[ i u -4

1..

m,, J, 3, ' g .g -1 { { .) - ..k_. j c!_. ;.. "l4 y f ) A, .MT - 9, m. h. . y -}- y._ - q, 4 ?.,.. _-- p. i ~ i ? - '* 4 _i _ . b j j_' -~j J f

• 3

..~ l ~ . j ~ - r. em ) l' ~i ~ ,. _,. _..? l .N 4 o i j i y I D b, { 4 { -- i { = f. i i M- ^ ' ,j j -{.~T'._. p_ j _ . _9. __ (J. .,.b_- -- g.4_. _~ ..)- gi-9. j 7 4 an -- g NJ . 46 4 ' ' 4 4.. .i4.:' -1 ^ '7 L - .... 5 4_. -.}. .4 .1'- _j -"s .N ; g -*A', e ~~' ' - g -. . p. g y c e 1 3

-j ;..,j..: q.

_-] T- . K _. O, _ - ' l-_

• ~

_.___3__ .g ,j L r=4 .=..e. t a-it t . d. - 1' .g.,( ,i sp 6.- 4

• 1

-i 1 i 1 _.] 11 !W 1 _j -~ L7 e j FB 7 4 S 4_ ..___a 2 _- .: y ~._, _ _.i._ -l 4 i . _ g _' ; ~ . j g --r a w a; 7i a._ y , j

s.n

-.. {. - l. f __i __j %l -t g...., _.r .t . u-a 2 ~~ ; N, }. 4 1 .i j _. i-j 3_r q -p _ :i a +.._ p 2. a ,_q .c 1 - 3 { i 4 .[ v i I .a____. ... _,_q.. 9.... _L.; . g.... .4-.. -t__ .d. i. 2. i 9-f { -1 2.

g _.;

>a: - or u-ms ....... om (

} j. . -.g"*._g*7 3 . } -i.. t ________o [ 3 I .. ___ _, _.. }4 , _._.,i _._.e._. . i _-. 1. i m ..i _.. _s .y_4 _m._ x l _F u y q_. g _. it, n4 ,y t. f*. _.q.. ..t

m. j

. _ g t .s, ~ I .,~. k,, r.1 _ _ _.- i ~ i i . %I .l i ,.e- ..e ee ...e._ eeme. e e gJ -.... l 3 l t .Q-..,1 9. c M.PMM, W .__.m.'..e_-__ 2 -' l. }.. 9 g s '. I ? -: 1.* 4_: j... ei e we u. 9. ,,.. -m. .,_p. ,.q q , M., .f. g J _j } } ;

4. st -- ;

1. c

j.-. -

i, . _ p 9... b

t.

4 ( j _ 'q - j . t. 4 .Lm .M,f j._., m, Jn -4 9, -_L e

m. w

-~

• ~ '
• *f

,,f__ ,_.2 3 ...s l j,.--.._.. j.,.. _.. e .., 3 .m_ . _.-.-,h.~, b ( ._ ~' i 4 . 1 1 y,.r. ..g-- ., - - j - 3 j n J 4 Jf --) _i.-t'b . } *' t 4,,._ y* e s I k h ~f"k 7 i ' l - 7{ j c T ry-_._ -.i _ _., t o

a.,.

m I - 6 .4 - r. L, -. n. 7.. 4 . m --- 4 , i g. g ~ g,, .~a__ _.l., ... ~,. S-l ...a.-.. -.4. w. g-. 'l -6 .. -j ~. -.. _ g' l 4 .] - Q-. -. _..) _3 .f... t ._ y... .s _ vu

=t-

=k. d -]. -g .} ._y_... g y . 2 ~; -- t ' .,.,a _.A _,g.y g__. j. a a: - .2 .t p .J .g e... M .o 2, 4 -f

• 9 *
  • '***T 7 --.--m y~- m'- ---;.3.

. = .Q,

• gg'

'l 4 .{. t -...-..q. A ..y .g. _. j.... ...J._. . A 4. _' ~~ ^h .- f -m> 1 __ -l 'f- ~QAQ __7-f'~._d, .a ~ I v ~ ' -- y ' ' _. E), _ a a 1 . i %1 ...J 4 i .o g .._ ]- .y w .a w = -a- ! ~ h :.A ?.'a j '4 4

• 7

b. ',

- i: _. }. (

• • - ' ~ '/

..5 4 o - 1 j -. T - e W... l _~7' m ' - i.,. I q -) - ,i ~ ee '.I J. E e. J + , p_, ' I E 'Ie _a_ _M i w o i e l .I _ i .- J. :'_. l~ {.. -j. i 1 ..- 3 l l .y \\ f -J - r v. ........ ~ a

-P \\{> Fatigue Evaluation r 'T = T Jw A m. T

• T T R8 A.

a T=T/(wet.) P Ss = b-h 157te pi s i i 4 t t r i 1 I L r l e from ASWE Section Ul, Anavoble Cycles = 55,000 l l P N HoW the Weld is Gone, Eowoble Cycles = 6,000 'I l l ,. - - ~ _ -. _ - _ -,.. - -,, _ -.

I Y, b .T .~

a.,

Conservatisms / i e AISC Guidelines Were Used e No Credit Take for Reinforcement from Adjacent Lever e Maximum Weasured Torque Used - No Energy Associated With the Peak I i I l l l 1

v s Inspection Program ) .e Criterio Development e Criterio Validation Tests e Capabilities for Finding Degradativ.

~,. INSPECTION CRITERIA VALIDATION. TESTS .? ' Weld J Weld Specimen Failure Torque Average ./f ') Fillet Size (IN.LB.) (IN.LB.) (Degrees) I

!p

.1875 180 2E 3784 4285 1 60 3D 4181 180 2D 4890 .1875 90 ,18: 2416 2883 90 2B 3322+ 90 3A 2913 .1875 45 4A 2667+ 45 2A 2126 2142 45 3F 1633 .125 120 1A 1873+ 2641 120 3B 3431 120 1F 2618 .125 90 4B 2938 90 4E 3699 3321 90 4F 3326 4 + I n - e r--,n. ,.----n- ,-n,_ -.-_,_n,,

'.r; '- t ' =: ._._:1 r: f._l, __~.;~ ~ t. r. F e + i. - - Ei. k, ' u-'-=-4,-._.-~; :_._ y-1;.1 _is. H_1__, t -- _.3 E j:-.u -l

--= ^::

z: r:_ :..- f -- .:. - y ~* -!==. -. r. ;- p.-- L.-t._ - - ~ - .d_ } :. _ ;

i. -

. ; b. - *:. .!.a . ~ -. -I.*~._. ..~#-. . f - _. i. -1 (-. .I"- ; [-[._J N_. .E' b~~~ M*Jd~-- '- * ' ~ 1 " 5 ~~._ d *_"N'" ' .9_ .=... : ;. 4.. _.;.y_:_ __ ., g,..;.c.g. _,. _.

,,..

:,,..c.;.

. J - :: i _

7. _. [;.

3_.

-( _. _j.. : g, _._. . -.gg-p,r

; p. = -=

.-

a- }; - .: ;.p.. ,. :.q .'._2..

.. L:.=.= -. : [..-...:.

...g._~.- . ; 5 _- f, -t.- _ w F -- .q ._._.3.._._ _....,;_,.- m p __.- - 4_,,. .{ 7 [

h.

.._-{.-- .: 4 ;;a . K. 'd.a:_; 5-Q.. .. j-__ _..g ~ "~~{-Z.;c' h g ;.;El--' . _'. ( - - - j. fi ;.::.:{.-- - ; _: t i. ;d _. i--=-' ~_ M-i._: i?. .~g q &. . ~. y; - ~} . - ~ y - =~ :. L.c.; Y.is $ w. __ -g.._... _. 9, y, . y.

l..

_y_._.__ _.._...j. .. ;.. _ f 7-

.;;_p : !=jf4

• r.2..-.-J'

. ]._ [. - t -.j. ~~; d.h, 'h':d:.h.'. ~{ b,j .; "* { ".. -* L,.i..j _..$.Z_.'

,,2,_-

y ? -:. -Q& ' ~. ' ". " 4 =. ! .. ~. .. - !..+~~.. _ .~

" :~ ; =~~:

.:.. ~ f'=T: -

==TT. - ~ ~~ __. V..-.i._ _"." ~~: ? L.. .4.. g. } - -. =. = J ';.' 1(_"l l: e C: * ' ~ .L.' ? -s' h. j- . ~ _ f r,-- 3_ _.a z-1.~..... _. = _ _ . _. t_ .7.-..;.._ -- -r,.. p.. __ .l: 2.;:-. t_ m ----

---

L. - -

-- r :_ - -.- e m m m- - = E 4+ 9 -.. = 2 i>--- =lWE:' -= r = =:--:::-*m 1 22.v * , ve * ?v p f..._- n.- t.__. _.__a__._ :_ 3 _ p _.. p_ p .,[.. . (h ~. _%. .- W d-C ~' ] -h _y_.q:.4 _b.

_ }

.

L_i_ . r} L~. ~ ._L.: .-}-

i;;.~.::. :. :-@..-j-.{ =:=: - =if f

- - t-T

==- y -f - - -t - - ~ 3 ' ~~ -~~:-~.^ ~;& G T-4_ __..-.-...,'y=- 3 - + r . !.. _.m..=

4. :.. :. --

.-.t..:. 3- - - r : - r_. ---i.+-.__ --j :: - .u. 22 5 ^ _ _S.4:..._ l . i

_.~._

.~r-i .n .y

u:L.-..

4. l.,., =+ p e } , n __, 9 t_ -h t:= I i .i _.. - g :;---- s i 7 n 6_'N E - ,2- ~ ~~ s I l ____ N i- -. l l 2 {. l g ~ Q o _... - [ __.2 -t i -.. - - f, ..a U V o .r i f n i z:..... . _\\ - i..,:. =r ~ _.t~ -- s. J V t l-y,- } . - i=. . n - 1

r. -.

s== - y 1- ^ g. s .._- ta l U f ..._j _.g. . _. - = _.._- r -I = :n-a; 4-4 --- "4-- L- ~~' ~ ' ~~~ e I W h - ---- -M i i-~ ' _ i ._,L_..~-~~t 1-l 'Q rf I_. -.. . _ _. - ~. _ _ 4_- i. p _. p. -. __ g.y _. g___ 1. W 2 g _ _. u.L _ $. _._- 3_-_.__. I g. g ~ -..--.I.-==.E--= i---- . + - _ -. y s j.. g. L. ,,.. (J t { q a s --,--r--

---

p-t m

-~ ~. - - - ~ ~ - - --- - _...L... &m i h t O h 3 0 9 Q ' - -, *e.g o f Ob3 %$ $ $ bl% @4 p T4 \\ Q ( l

~

c. f 9/z)/t?

1 A l Visual Inspection e Surface Cracks i e Crater Cracks e lack of Side Wall Fusion e Cold Weld e Weld Contour t

• Size of Fillet i

t i I 1 l l P f l I I

Fillet Weld Profiles b CONCAVE

• Desirable Fillet Weld l
• Indicative of Hot Weld
• Good Side Wall Fusion
• Indicative of Good Root Fusion CONVEX
• Desirable Fillet Weld
• Good Side Wall Fusion
• Indicative of Good Root Fusion HIGHLY CONVEX
• Undesiro'>le Fillet Weld
• Indicative of Cold Weld l

\\j

• Indicative of Poor Side Wall Fusion l
• Indicative of Poor Root Fusion l

I

9/'O/s 7 DESIGN, MANUFACTURING & TEST CRITERIA A. UNIFORM INDUSTRY STANDARDS NEMA SG L.V. POWER CIRCUlT BREAKERS ANSI /IEEE C37.13 -L.V. A.C. POWER CIRCUlT BREAKERS USED IN ENCLOSURES ANSI C37.16 - PREFERRED RATINGS, RELATED REQUIREMENTS, AND APPLICATION RECOMMENDATIONS FOR L.V. POWER CIRCUlT BREAKERS ANSI C37.17 - TRIP DEVICES FOR AC..L.V. POWER CIRCUlT BREAKERS ANSI C37.50 - TEST PROCEDURES FOR L.V. AC POWER CIRCUIT BREAKERS USED IN ENCLOSURES ENVELOPE OF ABCVE STANDARDS ARE: DESIGN TESTS DIELECTRIC WITHSTAND TEST i CONTINUOUS CURRENT TEST SHORT-CIRCUlT CURRENT INTERRUPTING TEST ENDURANCE TEST REQUIREMENTS OF ENDURANCE TEST CRITERIA (S THAT IT ... SHALL NOT NECESSITATE THE REPAIR OR REPLACEMENT OF ANY FUNCTIONAL PARTS PRIOR TO COMPLETION" (OF ENDURANCE TEST). MINIMUM LIMITS FOR ACCEPTANCE ON ENDURANCE TEST ARE SPECIFIEO IN THE STANDARDS. - PRODUCTION TESTS - DIELECTRIC WITHSTAND TEST - MECHANICAL OPERATION TEST - CALIBRATION TEST B. INDUSTRY CERTIFICATION BREAKER HAS BEEN CERTIFIED BY UNDERWRITERS COMPLETED PER UL) STANDARD 498 IN APRIL-MAY, LABORATORIES fUL AFTER TESTS WERE SUCCESSFULLY 1974. RETEST REQUIREMENTS FOR PRODUCT, AS DEFINED IN ANSI C37.50 HAVE BEEN COMPLETED. LAST SERIES OF TEST W8RE PERFORMED IN 1985.

I TEST PERFORMED ON DS-416 BREAKERS BY WESTINGHOUSE DESIGN VERIFICATION TEST (TEST DESIGN INTEGRITY) - SWITCHGEAR DIVISION I PROTOTYPE TESTS (TO ANS', SYD) - SWGR. DIVISION I TYPE TESTS (TO UL STD) - SWGR. DIVISION RETEST (TO ANSI STD) - fiWGR. DIVISION 1 QUALIFICATION TESTS (TO lEEE STD) - SWGR. DIV. l QUALIFICATION TESTS (TO lEEE STD) - NUC. DIV. WOG TESTS - NUC. DIVISION These tests were done tc satisfy different requirements as noted in parenthsis against eacli. The time period spans from 1968/69 to 1985. Different specimens were used in each of the tests, and different amount of cycles operations were induced in each. In total approximately 90 000 breaker test operations were enco,untered through this series of tests. Of the total 00,000, about 50,000 test operations were done during WOG tests. No failure of pole shaft occurred. Highest three numbers of operations on a breaker are 10,964, 10,505 and 8,469, all induced during the WOG tests. 1 l

9h3/8? h-ABSTRACT THE DEDICATION OF COMMERCIALLY PROCURED PARTS FOR USE IN SAFETY-RiLATED APPLICATIONS J. J. JELOVICH; J. R. HILL; F. B. HYLAND WESTINGHOUSE ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA 15230 In resolving the problem of diminishing sepport of original equipment manufacturers for supply of safety related parts to the nuclear industry, Westinghouse has developed a techt.ique which permits the use of parts manufactured under commercial procedures, in safety-related applications. The process transfers responsibility for 10CFR50 and 10CFR21 compliance from the primary manufacturer to the dedicating agency. Two years of experience with the formal process has demonstrated that adherence with certain key guidelines will assure qualificatien of the commercially manufactured parts for use in safety-related applications. Published in the American Nuclear Society Proceedings of the International Meeting on Nuclear Power Plant Maintenance March 1986 I

THE DEDICATION OF CONHERCIALLY PROCURED PARTS FOR USE IN SAFETY-RELA 1ED APPLICATIONS J. J. JELOVICH; J. R. HILL; F. B. HYLAND WESTINGHOU!E ELECTRIC CORPORATION PITTSBURGH, PENNSYLVANIA 15230, U.S.A. With the diminishing support of original equipment manufacturers for supply of parts to the nuclear industry, utilities are being faced with the problem of maintaining safety related equipment as the sources of qualified replacement and spare parts dwindle. From the equipment manufacturer's viewpoint, the evolution of the nuclear power industry has come to a point where the absence of new plant construction has depressed his market for qualified equipment to the level where he cannot justify the cost of maintaining the organizational structure and processes required for compliance with 10CFR50 and 10CFR21. With all aspects of the situation considered, and particularly the pragmatic aspect of the comercial world, the problem reduces to that of how to comply with the intent of the federal regulations without imposing the requirements of 10CFR50 and 10CFR21 upoa the primary manufacturer of the equipme.it. As an NSSS supplier, Westinghouse has a continuing 'nterest in i seeking *a viable solution to this problem and disseminating the results of our efforts and the apparent solution to the industry. l I Basic to the resolution of the probler, is acceptance of the fact ~ 4 that the part to be secured for use in the safety related application will originate as a comercial device, not intended especially for the nuclear industry, and when procured from the manufacturer, it will be no different from identical para provided to the industrial and comercial market in general. A rcview of the philosophy employed in the original design and developan.6 of the vast majority of safety-related equipment reveals that the designer selected "commercially" available compontnts for his design, built the prototype, proved his

design, and then qualified the oculpment by analysis and testing, finally imposing design control anc manufacturing control upon the production of the equipment to assure compliance with the regulations.

l l l l l

The Westinghouse approach parallels the earlier philosophy by again utilizing comercial components. However, today, that component l configuration has already been qualified and it remains merely to prove that the ccaponent pM :ured today is identical in all critical attributes to the cualified component spesinen, establishing the link to qualification,' snd accomplishing this without resorting to strict design control and manufacturing control on production of the commercial component. Once purchased, the specific part must be subjected to a process which will result in certification of the part as qualified for use in a safety-related application. That process, which will probably be l defined as "conditioning" by thi IEEE, has been described as "comercial dedication" in the Westinghouse program. This paper describes the comercial dedication process as developed by Westinghouse and presents it as a reference point for the dedication of parts by other agencies. KEY ELEMENTS OF COMiERCIAL DEDICATION In formulating a policy addressing commercial dedication. Westinghouse considered past experience and looked at the variations of procurement and supply anticipated in the long term. There are three categories of equipment whic) must be accotuodated in a comercial dedication program: a) supply of a component which is available et a comercial product; b) su) ply of a component to replace an obsolete component no longer availa sle, even as a comercial product and c) supply of components for new systems intended for safety related applications. The decision was made to establish a single policy and subsequently implement a program which would envelope all three of these categories. The commercial dedication process consists of establishing that the part is stellar in all critical attributes to the specimen part which was manufactured under a 10CFR50 program and qualified by actual testing to the requirements of IEEE 323 and IEEE-344. Essential to the process are certain key elements: 1. The component must be qualified by documented testing or analysis. Whether a component starts life as a commercial device or is subjected to stringent 10CFR50 controls from its inception, it must be proven qualifi%i for use in a safety related application by actual test or anaylsis in compliance with the regulatory guidelines. A key element is to identify that qualificatien reference and establish the physical configuration of the referenced specimen, i ,._,n, --,,,--------,-wen-

2. A second key element, the heart of the commercial dedication

process, and the one which most affects the quality of the dedication, is establishing that the component is similar.in all critical attributes to the specimen component which was qualified by actual testing to the requirements of IEEE-323 and IEEE 344.

This process is carried out through application of engineering evaluations, physical measurement, functional

testing, and material analysis, all relating back to the qualified design.

3. To distinguish the dedicated component from the straight commercial version in the field and facilitate compliance with 10C F.'21, the component must be uniquely identified to facilitate tracking. Techniques such as marking with special part numbers and serial numbers may be used. This third key element must also be documented in traceable format associated with the end user's purchase order. 4. Because rigorous design control is not employed in the commercial dedication process, material changes and subtle design changes, particularly in the area of tolerances, may be incorporated in the commercial part. To detect such changes, which could compromise the qualification of the component, periodic requalification of the component must be emgloyed. The frequency of this periodic requalification testing is established by engineering evaluation, taking into account the degree of sophistication of the specific product design and manufacturing techniques, the state of the art of commercial manufacturing in this area, reputation of the manufacturer and commercial version user experience. 5. The key element which is of prime importance to the components end user is the assumption of 10CFR21 responsibility and certification of that responsibility by the dedicating agency. The dedicating agency must comply with all aspects of the 10CFR21 I responsibility. The commercial manufacturing origin of tha component does not mitigate this responsibili'.y. l 6. The final key element is satisfying the intent of 10CFR50 by the application of a comprehensive quality assurance program to all aspects of the commercial dedicatior, process. I l

IMPLEMENTATION OF THE COP 91ERCIAL DEDICATION PRO The comercial dedication process is considered a significant departure it receives a great deal of visibility.from past practice by a

such, as itself is the control and documentation ofAlmost as important as the process The process must be structured to acconnodate detailed audit by the end the process.

user and the USNRC. It is essential that the objectives of the process, its' structure and procedure be formally described in a policy statement endorsed by the partir,ent cor> oration management. At Westinghouse this has been accomplished ay a formal report, the "Renewal Parts Dedication Process', which delineates the actions necessary to basic component application. enable the dedication of a ' commercial grade item" for In this document, the purpose, scope and dedication process are described the participatin along with the responsibilities of and Requisition; g working groups;, Engineering;Order Administration Test and Manufacturing Operations; Quality Assurance; and Nuclear Safety. In addition guidelines for the working structure are given, establishing the overa,ll program controlling vehicles, ' Engineering the Work Order' and the

detailed, product

specific,

' Engineering Control Instruction". The in use at Westinghouse's Monroeville,comercial dedication program has b and Pa. facility for two years. In

practice, we have found that three technical groups must work together in a closely-coupled relationship in order to responc' quickly and adequately to the user's requirements.

The Engineering group provides detailed technical direction by designing the the customer's request, program to produce the specific product in response to in the production process.and by defining the exact procedure to be used This is accompitshed by the issuing of Engineering Work Order which tells each party in the precess an what they exactly are to do in filling the customer order. The group is told exactly what saterials to buy and from whom to buyPurchasing The Test and them. Manufacturin group is told how to carry out the inspection, manufacturing, gassembly and testing operations. These i instructions define the critical aspects of the parts important to performance of the safety related function, such as dimensions, materials and material hardnesses and finishes, the tolerance stackups among related parts, etc. The Quality Assurance group is given the information necessary to show the tie between the product and the supporting qualification testing pro sampling and examination of materials, gram. Direction for periodic and directions for including samples in ongoing requalification seismic testing programs are also provided to Quality Assurance. The Engineering Work Order also provides the cross reference and information necessary to permit an audit of the process, beginning with the customer's order and ending with the quality release information at shipment. \\

l f In order to carry out this phase of the process, the Oedication l Agency's engineering group must have access to the specific qualification records, at least to the depth which allows the test specimen physical configuration to be established. Without infomation on a past related qualification test, the Engineering group must undertake to document the present comercial product, carry out a qualification test and thereby establish the qualification reference for future work where the part would be obtained from the primary manufacturer as a comercial part. This technique addresses the categories of obsolete component replacement and new system components. The Engineering group must also have access to the design information or, as a minimum, complete operational characteristic information sufficient to permit an analysis of the critical function and associated physical characteristics important to performance of the critical function. The spectfic technical procedure for dedication of the comt.onent is provided by meals of the "Engineering Control Instruction *, a formal document giving step by step detailed direction for the inspection, measurement, modification and testing of the comercial component. During the past two years, a library of several hundred of these component specific documents has been developed. They are written in response to the initial end user's request for a component and then utilized, intact, to satisfy all subsequent requests from all end users of that component. The majority of the actual work involved in the comercial dedication process is performed by the Test and Mantfacturing group at a facility established for the purpose of producing obsolete spare parts in support of the nuclear industry. The facility is organize a a light manufacturing operation with sophisticated inspection mad testing capabilities. The core personnel at the Assembly and Test facility consist of manufacturing engineers and technicians. However, because of the comercial dedication requirments, the Qualit/ Assurance group plays a major role. 10CFR50 AND 10CFR21 COMPLIANCE In a comercial dedication program all 18 criteria of Appendix B to 10CFR Part 50 must be addressed and applied to the process. The dedicating agency relives the primary manufacturer of 10CFR50 responsibility, as well as the 10CFR21 responsibility and assumes those responsibilities. The quality assurance aspects thus become crucial to the success of such a program.

l l l l l l In the Westinghouse program, Quality Assurance personnel are involved in the process from manufacturer evaluation to verification l that the dedication process has been faithfully followed. Inspection is part of the dedication process, and as such, Quality Assurance has a l verification role as well as a surveillance function establishing that others have accomplished their objectives, as required. Quality Assurance also initiates the "Certificate of Conformance/ Compliance" and retains auditable documentation supporting 10CFR21 reporting requirments. Because of our role as an NSSS supplier, with a major role in supporting the licensing of plants, Westinghouse has in place, a large Nuclear Safety and I.icensing department. The comercial dedication program draws upon this existing organization for regulatory support. It is the utilization of this existing capability which makes practical the transfer of responsibility for regulatory compliance from the primary equipment manufacturer to Westinghouse's Nuclear Services Integration Division (NSID). C009tERCIAL DEDICATION EXPERIENCE Because of the pressing need to continue support of the obsolete 'DB' line of Westinghouse switchgear in operating nuclear power plants, l the initial products undertaken in the comercial dedication program were DB air circuit breaker parts and complete DB circuit breakers. The original manufacturing division had comitted to produce some 'wnar" parts for a limited period to the comerf.ial market, but, not the nuclear market. NSID found it necessary to transition all aspects of this switchgear product line from the original division. This included design calculations; drawings; manufacturing procedures; processes; manufacturing and test fixtures, tooling and qualification test reports. During the past two years, we have produced qualified parts for all models of DB air circuit breakers, complete DB air i circuit breakers and have refurbished DB breakers for reactor trip sisitchgear application. Among the lessons learned, or rather among the suspicions verified, was that coernercial products are subject to evolution. Comercial products which are initially designed and manufactured to very stringent specifications may evolve over the years to a product significantly different from the original product, aarticularly in the area of materials used. This is due, of course, to tie need to optimize production and the related costs.

In the DB circuit breaker experience, this meant that we have had to manufacture certain parts or modify certain commercial parts so that we could control aspects such as dimensional tolerances, material

hardness, plating and surface finishes important to performance of critical functions.

Every safety related part provided by the dedicating agency must be subjected to the dedication process in order to identify manufacturing changes affecting critical functions. An example of a "standard" comercially dedicated product is the '01" type air circuit breaker. The circuit breaker is procured as a comercial product and subjected to a comorcial dedication process which includes the replacement of certain comercial parts with qualified, Class IE,

parts, such as wiring, shunt trip attachment, amptector and control relay.

In this product, computer access to the sianufacturing instructions permits NSID to identify design changes in a comercial product. No special requirements are imposed upon the primary manufacturar. The manufacturer treats NSID as he would any other comercial customer and NSID accepts the circuit breaker as a comercial product. A wide variety of products have been produced using the comercial dedication technique. They range from small parts such as teminal

blocks, switches, protection and control relays, up through more sophisticated devices such as printed circuit cards and molded case circuit breakers to switchgear lineups and dry type transfomers.

The 1000 KVA dry type transformer is typical of a new product which required seismic and environmental qualification of the product by actual testing and analysis and documenting the product such that subsequent transfomers in this family can be comercially dedicated using this qualification work as the reference. Other ratings in this

family, from 250 KVA to 2500 KVA will be qualified by analysis based upon the detailed instrumentation of the actual test and computer modeling of other ratings. All of ths dedication is perfomed by NSID.

The manufacturer will continue to provide the transfomers as comercial devices to NSID. StM4ARY After two years of use, the comercial dedication technique for meeting the utility need for acquiring qualified spare and replacement components for safety-related applications has been proven viable and successful. The comercial dedication process has been accepted by the utilities, equipment manufacturers, and the USNRC as evidenced by the purchase of Westinghouse dedicated components and the successful audit of the process, o

There are certain elements whic'n are essential for a valid commercial dedication program. The dedicated component must be demon:trated to De essentially identical to a qualified component which was qualified by actual testing, and the related qualification report must be referenced and traceable. The requirements of 10CFR50 must be applied by the dedicating agency. The dedicating agency must assume 10CFR21 responsibility and put into place the ceganizational structure required to support that responsibility. The overall program is based upon the sophistication of the dedicating agencies engineering, quality assurance and licensing support capabilities. 6 4 -n a-

QA PROCESS (HDWE PROCURED FROM E-PGH) - REQUIREMENT 10CFR50 Appendix B e -lNTERNAL QA PROGRAM e RESAR 3 WCAP 8370 (1974) e Operating Procedures e Departmental Manuals / Procedures e QA Engineering Purchasing Drafting Internal Audit Program e i VENDOR CONTROL Approved Vendor e Vendor Audits l QCS-1 t Vendor Surveillance e Test Witness Final Examination Quality Release Quality Procurement Specification e QPS 386-2 (March 77) ._-,-,.,_..,--..-__a. ___n_,,-, r.___.--- ,,,e,-

- VENDOR PROGRAM e Formal QA Manual Supplemental Procedures e Manufacturing MI's e Process Control Written Procedures Qualified Personnel e inspection Receiving in Process First Article Roving Random Sample e Final Product Test Inspection

QA PROCESS (HDWE PROCURED FROM E-PGH/P.R.) - REQUIREMENT e 10CFR50 Appendix B -lNTERNAL QA PROGRAM e WCAP 8370 Operating Procedures e Departmental Manuals / Procedures e QA Engineering Purchasing Drafting - VENDOR CONTROL Approved Vendor (East Pgh) e ~ Vendor Audits QCS-1 Approved Subtier (WICO - PR) e Subtler Audit QCS-1 Vendor /Sub-Vendor Surveillance e Test Witness Final Examination Quality Release Quality Procurement Specification e QPS-386-2 i

- VENDOR PROGRAM e Formal QA Manual Supplemental Procedures e Procurement East Pgh. Responsibility WICO Procured Misc. Services, e.g. Plating e Receiving Parts supplied via East Pgh. Control RI at PR Qty. Damage Correct Part e Manufacturing MI's e Process Control Written Procedures Qualified Personnel e in Process inspection First Article Roving Random Sample e Final Product Test inspection

QA PROCESS - COMMERCIAL DEDICATION - REQUIREMENT e 10CFR50 Appendix B t Internal QA Prograns e .WCAP 8370 WCAP 9245 Operating Procedures QA Manual and Procedures A&T Procedures Manual - VENDOR CONTROL e Vendor Audit QCS-2 e-Procured Commercial Grade - A&T CONTROLS 1 e Requirements Document EWO ECl e Manufacture Auditable Trail e Test Documented l l l -, - - - - - - -,,, ~.,, -, _, - - - -,. _ - - _,. - - - _. -. -

V O ' = r e inspection Documented QA and Engineering Certification e Package and Ship e QA Witness VENDOR PROGRAM e Formal QA Manual Supplemental Procedures Manufacturing e MI's e Process Control Written Procedures l Qualified Personnel inspection e Receiving in Process Roving Random Sample i 1 In Process Assembly l 100 % o Final Product i Test inspection ---=c-r- -s--~-}}