IR 05000298/1997018

Insp Rept 50-298/97-18 on 971020-1117.No Violations Noted. Major Areas Inspected:Maint & Engineering

ML20203J782
Person / Time
Site:	Cooper
Issue date:	12/17/1997
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
To:
Shared Package
ML20203J774	List: IR 05000298/1997018 ML20203J771
References
50-298-97-18, NUDOCS 9712220092
Download: ML20203J782 (24)

Text

. ._ . _ ~ . _

.

t

.*

ENCLOSURE

..

U.S. NUCLEAR REGULATORY COMMISSION

REGION IV

Docket No.: 50-298 License No.: DPR-46 Report No.: 50 298/97-18 Licensee: Nebraska Public Power District

- Facility: Cooper Nuclear Ctation Location: P.O. Box 98 Brownville, Nebraska Dates: October 20 through November 17,1997 -

Inspectors: J. Shackelford, Senior Reactor Analyst C. Skinner, Resident inspector, Project Branch C A. Pal, Office of Nuclear Reactor Regulation S. A:exander, Office of Nuclear Reactor Regulation Approved By: Arthur T. Howell lit, Director .

Division of Reactor Safety ATTACHMENT: Supplemental Information s

s L

E Y-971222OOV2 971217

.PDR ADOCK 05000299

.O PDR

,..

- . - - -. - - .

-,-

'

2 EXECUTIVE JUMMARY Cooper Nuclear Station NRC Inspection Report 97-18

-

This team inspection reviewed the causes, circumstances, and corrective actions associated with the lack of certain maintenance activitiei associated with the licensee's safety-related 4160V circuit breakers. Additionally, a review was conducted of the licensee's 125Vdc battery load and voltage calculatio _,

Maintenance .

• - The review of several Magne Blast-related NRC information notices, and a lack of resolution of associated service advice letters were indicative of a superficially implemented industry oper..ng experience review program (Section M1.1).

• A history of 4160V breaker problems and failures had existed at the facility. Numerous breaker failures and deficiencies were first discovered in 1987, and a process to overhaul the breakers was instituted. During the period from 1989 to 1994,18 of the 24 safety-related breakers were overhauled. None of the nonessential breakers had been overhauled. In August of 1994, 'he overhaul process was terminate 1 The final evaluation of the licensee's root-cause assessment and the adeouacy of the program to maintain the 4160V breakers will remain open pending further review by the NRC (Sections M2.1, M2.3).

• _ A number of inconsistencies betwt en the licensee's procedures and the vendor recommendations were identified and the licensee had either weak or insufficient justification for these differences. This was characterized as a weakness in the maintenance program (Section M3.1).

Enaineerina

• The 125Vdc load profile calculation contained a number of nonconservative errors. The licensee's operability assessment indicated that the 125Vdc system remained capable of performing its design function. Additionally, the surveillance tests associated with the

125Vdc and 250Vdc systems lacked acceptance criteria. The overall adequacy of the

.125Vdc load profile calculation will remain open pending further NRC review (Section E2.1).

• The team identified weaknesses in the licensee's operability assessments associated wnh the potential degradation of the 4160V breakers (Sections E2.1, E2.3).

. .

rr

.

s 3-

• The team concluded that the licensee's approach to evaluating the potential risk associated with the lack of breaker maintenance was appropriate. The use of a risk-ranking methodology to prioritize plant maintenance efforts was regarded as a strength in the planning process (Section E2.2).

F

< .

o-4-Reoort Details Summary of Plant Staty3 The unit remained at 100 percent power throughout the duration of the inspectio ll.MaiQtenance Backoround General El3ctric recommends overhauling 4160V ac Magne-Blast circuit breakers on a 5-year interval to assure maximum reliability. Other industry sources recommend overhauls on schedules which range from 8 to 12 years. Cooper Nuclear Station had experienced failures and deficiencies on certain safety-related 4160V ac circuit breakers in the late 1980's time frame. As a result of these failures and deficiencies, the licensee instituted a process to systematically overhaul all of the safety-related 4160V breakers. Eighteen of the plant's 24 safety-related breakers were overhauled in the time period between 1989 and 1994. The process was terminated in 1994. On October 5,1997, a failure was noted on a nonsafety related 4160V breaker, which brought to management's attention the fact that 6 of the plant's safety-related breakers had not been overhauled in over 23 years of plant operatio This special te am inspection reviewed the licensee's program for performing preventive maintenanca and refurbishment for certain safety related circuit breaker M1 Conduct of Maintenance

.

M 1.1 Evaluation of Maintenance Activities Associated with 4160V Breakers Insoection Scoog The inspectors reviewed the h. story, scope, and extent of the maintenance activities associated with the GE Magne- Blast 4160Vac cirCJit breakers Particular emphasis was placed on whether the licensee had instituted a maintenance program that conformed to vendor and industry standards and whether the applicable NRC information notices had been appropriately dispodiione Observations and Find.093 Review of Vendor Recommendations The heensee had not implemented a mechanism for systematically tracking and dispositioning General Electric (GE) service advice letters (SALs). The licensee response to NRC Generic Letter 90-03, " Relaxation of Staff Position in Generic

_

.

.

. Letter 83 28," indicated that the facility had a vendor interface program that complied with the relevant aspects of the generic letter. However, the licensee operating experience review (OER) supervisor stated that GE SALs were not part of the OER program and that he was unsure of what oiganization would be responsible for reviewing them. Contrary to this position the inspectors determined that Administrative

"

Procedure 0.10. " Operating Experience Piogram, Revision 2.1, dated July 9,1996, in paragraph 7.1.1.3.d indicated that GE " service advisory"information should be reviewed in the OER program. Subsequently, the OER supervisor acknowledged that his group should have been reviewing GE SAL In response to this finding, the OER supervisor initiated Problem Identification Report #2-20820 on October 21,1997. The problem identification report indicated that the applicable SALs had been obtained from the maintenance engineer and entered the OER program on October 16,1997, yet the licensee's status document provided to the inspectors, dated October 22,1997, which was supposed to show all Magne-Blast-related SALS in the OER program, listed only three documents: SALs 073-325.1,352.1, and 354.1. A complete file of all Magne-Blast-related SALs had been obtained recently by the maintenance engineer from GE at an industry meeting. Following discussions with the OER supervisor and the maintenance engineer, the additional applicable SALS were added to the databas The licensee's preliminary review of GE recommended maintenance indicated that several modifications or upgrades (planned for current and future overhauls) remained to be accomplished on nonoverhauled breakers, as well as, several others on previously overhauled breakers. These modifications included:

• Replac3 ment of Tuf-Loc bushings (including prop bushings) with aluminum-bronze type in accordance with SAL 31 Replacement of the original prop springs with improved prop springs in accordance with SAL 34 Adding an additional prop spring in accordance with SAL 35 * Replacement of square shaft bushings with cast iron type; replacement of tension close latch springs with torsion types; replacement of latching pawl hinge pins and the driving pawl crank assembly with new designs; and replacement of the stop block, in accordance with SAL 352.1. The overall adequacy of the licensee's program to review and disposition the GE SALs is characterized as an inspection followup item (50-298/9718-01).

.

.

.

6-b.2 Review of NRC Information Notice Licensee Letter NLS940026, dated August 8,1994, to NRC Confirmatory Action Letter dated August 2,1994, described the commitments related to the OER program. The response indicated that the licensee believed that, although some problems existed, an adequate OER program was in place. The licensee commitment indicated that a complete review of all OER responses would be reviewed for adequacy within a 2 year period. Additionally, the response indicated that a 100 percent review of all pre-1987 OER responses would be reviewed for adequacy due to the licensee's perception that the program lacked an appropriate level of formality. This review apparently failed to identify that review of SALs were not being included in the OER progra The inspectors determined that the licensee response for NRC Information Notice 84-29,

" General Electric Magne-Blast Circuit Breaker Problems," was inadequate. The information notice was related to problems associated with excessive wear of the Teflon / fiberglass ("Tuf-Loc") sleeve bearings or bushings in the ML-13 mechanism of the Magne-Blast brmkers. This issue was also addressed by GE SALs 318.1,318.1 A and 318.2 in 197 , ccording to the licensee's response document, the licensee had determined that the information notice was not applicable to Cooper Nuclear Station because of the breakers being AMH type breakers, had Type ML-13A operating mechanisms. Neither the NRC information notice or the SAL explicitly mentioned ML-13A mechanisms, although the basic mechanisms are the same, and GE has stated that, in most cases, they used the term ML-13 as a generic type, meant to include the ML-13A. The licensee's error was compounded by an informal communication with a GE representative, who stated that the information notice mentioning only ML-13 mechanisms would not be applicable to ML-13A Information Notice 90-41, " Potential Failure of General Electric Magne-Blast Circuit Breakers and AK Circuit Breakers," addressed various deficiencies and operational issues associated with GE 4160V Magne Blast breakers. Most notably, it addressed the problem of breaking prop springs in Magne-Blasts, which was later addressed by GE in SAL 348.1. The replacement of the original, inferior-design prop springs discussed in Information Notice 90-41 with the new prop springs recommended in SAL 348.1 could be accomplished by competent licensee technicians. These new prop springs would presumably enhance the reliability of the breaker. The other issues addressed by this information notice could generally be covered by a periodic overhaul program. The licensee's response to this information notice credited the " overhaul program" with implementbg corrective actions to address these issues. The inspectors acknowledged that the licensee had a task order in place that authorized GE to perform breaker overhauls at the time the information notice was reviewed, but the overhaul program was never instituted as a formal plant process and this issue was not recognized in the OER program reviews. (As discussed in section M2.1 of this report. the overhaul program was terminated in 1994.)

.-

.

Information Notice 94-02,"lnoperability of General Electric Magne-Blast Breaker Because of Misahgnment of Close-Latch Spring," alerted addressees to a potential problem with Magne-Blast breakers in which the close latch reset spring of the tension type (used in breakers built before 1980) could get caught on the close latch monitoring switch bracket. This condition effectively disables the closing spring recharging function and the charging motor can attempt to recharge the spring repeatedly until the closing latch reset spring dislodges itself with the shock and vibration of the charging operation or with operator intervention. The licensee's determination that the problem was not eplicable to Cooper Nuclear Station was inadequate, it was based on information from GE, who determined that it was not applicable because the information notice mentioned

"AM" type Magne Blasts (meaning " air-magnetic") and did not mention AMH types (air magnetic, horizontal drawout), such as those at Cooper Nuclear Station; yet SAL 352,1 states that the replacement of tension type springs with torsion type springs in 1980 is applicable to both ML-13 (AM) and ML.13A (AMH) mechanisms (the Cooper Nuclear Station breakers were built before 1974). The licensee's electricians had only consulted breaker drawings and could not find a " tension-style" spring. There was no indication in the record that the licensee inspected the breakers to determine whether there was any actualinterference between the close latch reset springs and the close latch monitoring switch mounting brackets. The summary of parts replaced during the 19891994 overhaul project indicated that the spring was replaced with a torsion type coring on only two of the breaker Finally, Information Notice 97-08 involved a number of Magne-Blast component problems. The licensee had addressed some of the issues satisfactonly, but had not inspected its breakers to determine if the weld reinforcement on the trip crank pins had been ground flush, or if any cracking in the weld was evident. Some breakers inspected during the inspection did have the welds ground flush and at least one showed signs of incomplate fusion. This deficiency did not render the breakers inoperable because the amount of weld below the surface is variable depending on pin penetration and chamfer that is not visible. However, according to GE, trip cranks in this condition could potentially faii and should be replace Conclusions The licensee's review of many of the Magne Blast-related NRC information notices, and lack of resolution of associated SALs, review and were indicative of weaknesses in the OER program. Determinations of inapplicability were often based solely on minor differences in equipment-type designations instead of verifying that the equipment was actually not susceptible to the types of problems cited in the information notices and SALs. In some cases, it appeared that the information of the information notice that had genene applicability was not considered further when the equipment-type designation (regardless of inherent similarities) did not exactly match the examples used in the information notices. The licensee's process for review and disposition of SALs will be revic.ved further during a future inspection. The inpact of the specific examples noted

- will be evaluated as part of the unresolved item discussed in Section M .

.

8-M2 Maintenance and Material Condition of Facilities and Equipment M2.1 Assessment of Material Condition of 4160V Breakers lDipection Scope The inspectors reviewed the materias condition and maintenance history associated with the safety related 4160V breakers. Reviews were conducted of the as-found results of previous breaker overhauls that aad been conducte Observations and Findings Six of the 24 essenfr.nl 4160V breakers at Cooper Nuclear Station had not been cverhauled in the 23 years of p! ant operation. These were 2 emergency diesel generator output breakers (EG1 and EG2),2 residual heat removal pump breakers (RHRP1 A and RHRP1C),1 service water pump breaker (SWP1D), and 1 of the 4160/480V transformer supply breakers (SS1G). The fact that these breakers had not received the vendor and industry recommended overhauls were recognized by the licensee while investigating the failure of a condensate pump breaker. The condensate pump breaker was one of the plant's 29 nonsafety-related Magne-Blast breakers, none of which had been overhaule It was originally believed that only 5 of the 24 safety-related, essential breakers had not been overhauled. (The licensee believed that the SS1G b eaker had been overhauled.)

However, subsequent reviews by the licensee determined that of the 19 breakers sent in for overhaul during the period between 1989 and 1994, only 18 separate serial numbers had been used. This indicated that 1 of the 19 breakers had been overhauled twic The licensee determined that the SS1G breaker had not, in fact, been overhaule Problem Identification Report #2-19247 was generated to document the problem and the licensee's operability aseessment associated with the breakers was revised to incorporate this informwo The as-found reports were reviewed for 17 of the 19 4160V breakers that had been previously overhauled. (The licensee was unable to retrieve the reports for 2 of the overhauled breakers.) GE reported that one of the breakers was received in an inoperable condition. Ten of the breakers were reported as having evidence of stiff grease and/or a stiff closing mechanism. Additionally, a variety of other material condition problems were documented on the remaining breakers that were sent in for overhaul. The licensee's operability determination noted that the plant had not experienced operational failures due to inadequate lubrication during the time period since the overhaul process was terminated in 199 .- .- . - . . .. .- - . . - --

.

.

. The inspectors ieviewed the work and maintenance history associated with safety -

related 4160V breakers at the facility, it was determined that a number of breaker ,

problems had occurred in the mid to late 1980's time frame. Specifically, there were six cases of breaker failures documented due to less than adequate lubricatio Additionally, a number of other significant deficiencies were noted in that same time period. As a result of these problems, the licensee instituted a process (via a task ,

authorization order) to systematically overhaul the breakers. During the time period

'

between 1989 and 1994 all but six of the essential breakers were sent in for overhau ,

Licensee records indicated that the overhaul process was terminated in August of 199 Preliminary indications are that other, higher priority items associated with an emphasis on restart issues coupled with organizational changes contributed to the breaker overhaul program being dropped. (Cooper Nuclear Station was in an extended shutdown, which lasted from May 1994 through February 1995,) The issue of not completing the breaker overhaul had been brought up several times in the last few years, but never rose to the level of being identified via the corrective action process. The issue was finally brought to a high level of visibility on October 5,1997, following the failure of the charging motor to de-energize on the 1 A condensate pump breake As a result of this condition, the licensee conducted visualinspections on five of the six breakers that had not been overhauled. (Inspections were not conducted on the SSIG breaker that remained energized and under load throughout the inspection period.) In addition to the visual inspections, the breakers were manually cycled t3 check for abnormal stiffnass. The results of the inspections and manual cycling did not indicate that any immediate operability issues existed. The licensee took measures to procure additional breakers to replace those that had not been overhauled and to ship the remaining breakers to the vendor for servicing. This process had not been completed as of the end date of the inspection period. The overall adequacy of the licensee's program to perform preventive maintenance on safety-related,4160Vac circuit breakers ic characterized as an unresolved item (50-298/9718-02) pending the results of the as-found condition of the remaining breakers to be overhauled and further NRC revie c. Ognclusicas The inspectors concluded that a history of 4160V breaker problems had existed at the facility. Numerous breaker failures and deficiencies were first discovered in 1987, and a process to overhaul the breakers was instituted as a measure to correct this conditio During the period from 1989 to 1994,18 of the 24 safety-related breakers were overhauled In August of 1994 the overhaul process was terminated. Visualinspections and manual cycling of five of the six remaining breakers indicated that the breakers that had not been overhauled remained functional. The adequacy of the licensee's program to maintain the 4160Vac safety related breakers was characterized as an unresolved .

ite ._

-

, - ._ .,

- . _ . _ . __

-.

P

.

.10

.

- M2.2 Compensatory Measures a, insoection Scooe (62700)

The inspectors evaluated the licensee's compensatory measures associated with the potentially degraded reliability of the 4160V breakers that had not been overhaule Observations and Findinas -

The licensee liaplemented immediate compensatory measures to address the reliability

. concems associated with the 4160V breaker issue. These actions were:

1) Establishment of Night Order 97 017 to have operations visually verify proper actuation of all 4160V breakers when placing the equipment in service or removing it from service 2) Brief the plant operators on the current operability assessment, compensatory measures, and procedures to conduct local breaker operations 3) Require operators to verify circuit continuity twice per shift 4) Perform preventive maintenance on four of the breakers which had not undergone recent preventive maintenance (Residual Heat Removal Pump A and C breakers, Service Water Pump D breaker, and Diesel Generator 2 breaker)

It was noted that preventive maintenance had been recently performed on the output breaker for Diesel Generator 1. The preventive maintenance on the SS1G breaker could not be performed because it was in service and under load. The condition of the SS1G breaker was inferred on the basis of the results of the other breaker inspection Conclusions The inspectors concluded that the licensee's compensatory measures were adequat M2.3 Root-cause Assessment of Failure to Perform 4160V Breaker Maintenance in Accordance with Vendor and Industry Standards Inspection Scapg x

The inspectors reviewed the licensee's root-cause-assessment of the failure to perform 4160V breaker maintenance in accordance with vendor and industry standards. The licensee's overall process was evaluated, as well as, the reasonableness of the conclusions.

, ,

.'

X;

,- - -

. _m

.

.

-11- QhieJXalsns and Findings The licensee documented the failure to implement a maintenance program that conformed to vendor and industry standards in Problem identification Report #2-1780 The licensee had characterized this deficiency as a "SCAQ," or significant condition aciverse to quality, which is the highest level of deficiency in their corrective actions process As such, this condition was required by the licensee's program to be reviewed via a formal root-cause assessment. The licensee formed a root-cause assessment team to review the issues associated with the failure to implement the breaker maintenance program. The final review of the findings and adequacy of the licensee's root cause assessment is characterized as an additional area of review associated with the unresolved item to evaluate the adequacy of maintenance activities associated 4160V breakers (50-298/9718-02). C_onclusions The licensee had instituted a formal root cause assessment associated with the failure to institute a maintenance program for the 4160V breakers in conformance with vendar and industry standards. The licensee's assessment process was ongoing as of the end date of the inspection. The final evaluation of the licensee's root cause assessment is characterized as an unresolved item pending the completion of the process and further review by the NR MS Maintenance Procedures and Documentation M3.1 Evaluation of Maintenance Procedures and Documentation of Circuit Breakers Inspmt. on Scoce (62700)

The inspectors reviewed the vendor manuals for the General Electric Magne-Blast Circuit 4160V breakers, the Westinghouse 08 50 480V breakers, and the GE-Type AK breakers. A comparison was conducted between the recommended maintenance with that performed in accordance with the licensee's procedures and preventive maintenance progra Observations _and Findings b.1 GentIdi Electric Magne-Blast Circuit 4160V Breakers The inspectors conducted a comparison between the vendor manual and the licensee's Procedure 7.3.17,"4160V Breaker Maintenance and Testing," Revision 16. There were a number of items listed in the vendor manual that were not being performed during the performance of the licensee's preventive maintenance procedure. The vendor manual recommended that the breaker be operated slowly with the manual charging wrench and to verify no excessive binding or friction and that the breb er should be moved to the fully

_ _ ._ _ . _ _ . _ . _ . - * l i

.

I

!

opened and closed positions. Procedurs 7 3.17 dici accomplish the cychng of the breaker. However, the procedure did not ha,e specific steps that required the operator j

>

to check for excessive binding and friction nor was there a step to check that the breaker full opened and c osed Also, the vendor manual recommended that the trip coil l plunger ard the release coil plunger be checked for free movement. The procedure did not specifically have a step to check movement, but that the two coils were cycled during the performance of the procedure. The licensee could not produce documentation or justification for these discrepancie Additionally it was noted that the vendor manual and Service Advice Letter 35 recommended the use of a light, synthetic lubricating oil during inspections on certain breaker components (for example (roller / needle bear, Al/ bronze prop bushings, trip shaft split sleeve bearings, etc.). Procedure 7.3.17 required that grease be boed on these ,

same breaker components duiing the periodic inspection. The licensee could not justdy why grease was acceptable to be used versus the oil. The inspectors determined that oil was recommended, because as grease ages it hardens and adding the o.: replenishes the lubrication properties in the grease (proventing the grease from hardening). Adding new grease on top of oH grease does not ensure adequate lubrication as would the proper application of the vendor recommended oil. The licensee documented this issue in Problem Identification Report #2 2016 ,

The inspectors also observed that the vendor manuallisted a number of adjustments that should be checked dunng inspections, and the vendor manual stated that the adjustments were listed in the order in which they were to be checkrt. The inspectors identified that the procedure did not perform the adjustments in the order specified in the vendor manual. The licensee was unaware of the vendor manual statements and, therefore, could not justify why the site procedure performed the adjustments in a different order than that listed in the vendor manual. A maintenance engineer contacted a vendor representative and documented that the vendor representative concurred that the ordet is not critical to the performance of the adjustments. The inspectors, later contacted tne same vendor representative and received conflicting information as to the importance of performing the steps in the order specified in the vendor manua Consequently, the inspector observed a maintenance technician performing the stDps in the pacedure. The order in which the adjustments were performed did not appear to affect the other procedural steps or adjustments. At the end of the inspection period, the licensee had not documented why their order of performing the steps was not affecting other adjustments that were performed in the procedur Finally, the inspectors noted that the vendor had recommended an overhaul frequency for 4160V breakers of once every 5 years in order to assure maximum reliability. i a inspectors questioned the licensee as to the specific overhaul frequency in use at tne

.

~ w

. - _ _ _ _ - - - -_-_ _ . _ - - - . ___ _ _ _ - _ _ . -

l l

'

.

13-

,

facihty. The licenses stated that the facikty intended to implernent a 9 year overhaul frequency. However, the licensee's standard for preventiv,s maintenance for breakers and control circuits, (which had not yet been implemented), stated that a 10 year overhaul frequency would be implemented. The inspectors requested the licensee's justification for a 9 or 10 year overhau; interval as opposed to the vendor recommended 5 year interval. The licensee was not able to locate the documentation that justified a 9 or 10 year overhaulinterva b.2 Weni.oghmse DB 50 480VBJAakEs The inspectors conducted a comparison between the vendor manual and the licensee's Procedure 7.3.2.1, * Westinghouse DB 50 Breaker Maintenance and Testing,* Revision It was noted that Procedure 7.3.2.1 did not require that the auxiliary switch contacts actuate before the end of travel as recomrnended by the vendor manual. Instead, the procedure required that resistance measurements be taken to ersure that the contact';

were actuating. The licensee had removed the overtravel check from the procedure in Revision 6, without written justification as to why the resistance checks were equivalent to ventying the overtravel. At the end of the inspection period, the licensee was still evaluating whether the resistance ecks were equivalent to verifying overtravel. The inspectors noted that the venficatk of overtravel was important because this motion wiped clean the contacts after each breaker cycle and performing resistance checks would not venfy that there would be enough overtravel to accomplish this actio Adddionally, the inspectors determined that the vendor manual recommended a 6 month inspection frequency for the breakers. Westinghouse NSD Letter 74 02 recommended that inspections be initially performed on a semiannual basis and that the period, based on experience, might be extended to 9 or 12 months. In February 1987, the licensee provided justification for extending the frequency to every refueling outage based on

"f avoreble operating history." The justification did not include any quantitative observations as to the results of breaker inspections. The inspectors reviewed the maintenance records from 1982 to 1997 it appeared that a considerable number of these problems might have been identified or prevented with a shorter preventive maintenance frequency. The licensee's inspection frequency specified some brea' ers being inspected every refuehng outage and others every socond refueling outage. The inspectors concluded that more specific details should have been provided for the 1987 justification on extending the frequency and that there is no justification for extending the frequency an additional 18 months (total of D6 months) for some breakers. The inspectors concluded that a weakness existed in the licensee's inspection program associated with 480V circuit breakers. This issue is characterized as an inspection followup item (50-298/9718-03) for further NRC revie . Gennauilectnc Type AK Breakes The vendor manual recommended an inspection frequency of every year, and the

- licensee performed the inspections every 18 months. The inspectors requested the

_ ._

_

- . - , _ .. . . - . - - . - . . _ - - . - . - - - - . . - ~ _ - .- - - ._ - . - - -

.

,

,

justification for extension of the inspection frequency to a periodicity that exceeded the j manufacturer's recommendations. As of the end date of the inspection, the licensee .

~ had not provided adequate justification for an inspection interval that was inconsistent with that recommended by the manufacture Conclusions The inspectors identified a number of inconsistencies between the licensee's procedures i and the vendor recommendations. It was determined that the licensee either had weak or no written justification for these dinerences. - The inspectors concluded that while none of these inconsistencies resulted in a breaker failure, they could potentially affect ureaker reliability and performance. In particular, it was noted that the licensee's inspection and overhaul schedule was significantly less conservative than that recommended by the manufacturer for both 4160V and 480V circuit breakers. These extended schedules, *

coupled with at lack of appropriate justification represented a significant weakness in the licensee's preventive maintenance program associated with the breaker ?

Ill. EnalDie E2 Engineering Support of Facilities and Equipment E Vdc Load and Voltage Study Evaluation

, InspachotLSgcoe (93809)

As a result of the October 5,1997 condensate pump breaker failure, which was due to a failed de charging motor, the team reviewed 125Vdc Battery Load and Voltage Calculation NEDC 87131D, Revision 6, to evaluate the adequacy of the methodology, the accuracy of the duty cycle load profile, and equipment terminal voltages, Ob_stIYalians and FindiD93

'

The cal:u!ation documented the sequence of events on the 125Vdc battery system for the combined loss of offsite power / loss of coolant accident, station blackout, and safe shutdown fire scenarios at Cooper Nuclear Station. In addition, the calculation determined the tenninal voltage at the devices of the 125Vdc system to verify that adequate voltage is available for the affected devices to perform their safety fonctio The team identified the following errors associated with the 125Vdc Battery Load and Voltage Calculation NEDC 87-1310 Revision 6:

a t

_ . . -.. . - - - -

._ . - - -- - - - -

. -

- . . - _ -

.

l

<

.

15-i i

+ Input errors were identified with respect to certain motor-operated valve running ;

- current loads. Valve HPCI MO 16 was reflected in the calculation as  !

representing a running load of 8.8A: however, the nameplate data for the motor ;

I 1

! indicated that the actual running current would be 24A. This error resulted in i nonconservative estimates for the totalloading in the time intervals from  !

26 to 58 seconds as modeled in the calculatio l

!

i

• ' The loads associated with the 480V shunt trip loads associated with the Division 2 battery had not been modeled in the load calculation. This error l

resulted in a nonconservative estimate of approximately ',2A for the 4 to 8 second j i intervalin the calculatio !

• The impact of modifications had not always bem adequately incorporated into i I

- the de load profile. Calculation EJ 96115. Revision 2, was implemented, which

"

increased the size of the motor for Valve MS MO77 The impact of this i modification had the effect of ir. creasing the load on the de system and was not considered in the de loading calculation. The licensee shut and doeneiglzed t Valve MS MO77 as a compensatory measure to address this deficienc f Additional'y, the licensee identified the following error;

• The 4160V breaker closing amps considered in the de loading calculation were assumed to have a value of 14A for the spnng charging motor current. This value i

. represents the steady state value of the current and not the in rush value that is l approximately 75A. This observation applied to all 4160V breaker operations that were evaluated in the load profile calculatio ;

Based on the multiple errors and nonconservatisms that were identified, the licensee initiated Problem Identification Report #219750 to document the deficiencies associated i with the 125Vdc load profile. An operability assessment was performed, which acknowledged that the existing calculation of record may not have demonstrated the ,

adequacy of the de system under all required scenarios. The operability assessment  !

used qualitative reasoning to determine that the de system would remain operable and  ;

'

provide sufficient capacity for all design basis scenarios. In followup investigations, the licensee quantified the impact of the various errors that had been identified on the calculati_on. It was determined that the impact of the ident fied errors resulted in a nonconservative estimate of battery loading during the 4 to 8 second and 11 to i 14 sec snd intervals of the LOCA LOOP scena,io. (The magnitude of these nonconservatisms were approximately 3.4A and 52.45A, respectively.) However, it was >

l

$

l h

, i

. . . - - _ -. - . - - - - - -. ... ,- - -- - . a -,.--.= - .--

= .__

.- -- .- _.- ._ _-- . - - - - . _ . - . - - -

.

.

- 16 determined that the most recent service test (performed in 1995) showed that sufficient capacity existed to account for these additional loads that had not been previously evaluated. On this basis, the licensee concluded, and the team agreed, that the battery remained operable The failure to implement measures for adequately translating the design features associated with the 125Vdc system is identified as an unresolved item (50 298/9718 04) pending the licensee's revision to the calcelation and further NRC revie In addition to the observations concerning errors in the 125Vdc load profile calculation, ,

the team found that Surveillance Procedures 6 EE.603 and 6.EE.605, (the surveillance procedures for the 125Vdc and 250Vdc battery service tests) did not contain acceptance enteria. The procedures, as written, only required that the specified current is delivered and did not require that the corresponding voltages be checked. The licensee acknowledged these deficiencies and initiated Probiern Identification Report #219092 to document the condition. Additionally, the licensee conducted a review of the most recent surveillance test results and verified that the performance of the battery was within acceptability standards. The f ailure to incorporate acceptance criteria into the 125Vdc and 250Vdc surveillance tests is identified as an additional issue associated with the unresolved item partaining to the adequacy of the 125Vdc load and voltage study (50 298/9718-04).

The team reviewed Battery Charger Sizing Calculation 91094, Revision 4 (November 22,1996). The charger sizing calculation considered the battery load profile based on Calculation NEDC 87-131C, Revision 6, and -131D, Revision 5, that had been revised in 1997. The team determined that the charger sizing calculation had not been revised to reflect the newly revised de load profile calculations. The licensee acknowledged the load profile calculations represented inputs for the charger sizing calculation and initiated Problem Identification Report #219083 to document the deficiency. The licensee determined that the impact on the charger calculation would be minimal and that the charger remained operable based on the fact that actual charger current from field measurement is approximately 45 amps compared with the calculated current of 71.6 amps. The failure to update the battery charger sizing calculation when the input calculation had been revised is identified as an additionalissue to be addressed with respect to the unresolved item pertaining to the adequacy of the 125Vdc load and voltage study (50 298/9718-04).

he licensee s de voltage load calculation, assumed a minimum closing coil voltage of 26 volts for all 4160V circuit breakers. The licensee's Test and Maintenance

, ,edure 7.3.17.1, "4150V Breaker Examination," only verified a clouing coit voltage of 90V '90V is consistent with the manufacturer's recommendation). The 73.26V was based on a historical test performed on a spare breaker that indicated that the coils would operate at a voltage of 40V. The value of 73.26 volts was obtained by adding

P-m.- -r,

-u+ - -- , ww -- - -, - - . - em -

. -- _ --- -- __ .-. - .

t

.

17-

,

s margin for conservatism. The licensee indicated that the current calculations estimated that all of the safety related 4kV breakers could experience less than 90V at the closing coil under certain conditions. The values that were estimated ranged from 88.6 to 74.4Vdc with EG1 being the lowest vslue. Because the plant testing procedures had not verified proper breaker operation at these reduced voltages, the licensee initiated Problem identifi:ation Report #216753 to document the issu An operability determination was performed, wh;ch concluded that the breakers would be able to perform their design closing functions under the potentially degraded voltage conditions. The operability determination was based on a previous special test result (STP 8713), which was performed in 1987 and indicated that the closing coil (on a spare breaker) could operate at voltages of approximately 40V Additionally, the licensee had correspondence from GE (GE Letter G HPO 7 368), which indicated that breaker operation was possible at these reduced voltages for breakers that were in a new or refurbished condition. Additionally, the licensee's operability assessment stated that the resistance and inductance measurements that were recorded as part of Test and Maintenance Procedure 7.3.17.1 are trended for degradation and that the trend data would provide sufficient evidence that the closing coil would be in a condition consistent with that described in Procedure STP 8713 and GE Letter G HPO 7 368. However, the inspectors determined that no actual trending activities had taken place Many of the affected breakers had only a single data point associated with the resistance and inductance measurements. (This data had only been collected since 1995). Rather than a trending evaluation, the inspectors deten,iined that the licensee had, at most, conducted a qualitative engineering evaluation of the available data, which represented a weakness in the operability determination. Additionally, the inspectors disagreed with the licensee that the data would be reflective of that associated with a new or refurbished breaker as indicated in the GE position outlined in Letter G HPO 7 36 Subsequent to the operability evaluation, the licensee performed a special test on the EG1 output breaker and RHR1C supply breaker that indicated that these particular breakers would operate under the postulated design voltage conditions. The failure to implement appropriate measures to ensure that the 4160V breakers would operato under postulated design conditions is identified as an additional issue to be addressed with respect to the unresolved item pertaining to the adequacy of the 125Vdc load and voltage study (50-298/ 9718-04).

The 125Vdc load and voltage calculation used a design margin factor of 1.05 that gives a margin of 5 percent to accour t for any load variations found during subsequent calculation revisions. Howe'.cr. IEEE 485, to which the licensee had committed, secommends that additional margin be included for recent battery discharge. After the performance discharge test, the battery is recharged to approximately 95-98 percent and the battery is then declared operable. To compensate for this, IEEE 485 recommended this additional margin to account for the battery not being recharged to fully 100 percent, The team concluded that the licensee's failure to include this margin represented a weakness in the calculation.

l

'

. - ,-

.- . - - ..- .- - . _ - . ._- .. ..

.

.

Y 18-t The team reviewed the licensee's approach to adjusting the de load profile calculation to account for the effects of both temperature and aging. It was determined that the licensee had used a temperature correction factor of 1.04 to allow for a battery room temperature of 70 degree F. The battery room temperature was annunciated in the control room and plant procedures directed the operators to deploy a portable heater whenever battery room temperature dipped below 76 degree F. The calculation used an aging factor of 1.11 that would allow degradation of the battery to 90 percent of the nominal value to account for aging effects. IF.EE 485 recommends an aging factor of 1.25 that will allow degradation to 80 percent. The team reviewed the Technical Specifications 4.9.A.3.d.2, which states that a performance discharge test, in lieu of the above service test, once every 5 years is adequate to verify that battery capacity is at least 90 percent of the manufacturer's rating. Additionally, Surveillance Procedure 6.EE.607, '125V station battery Performance Discharge Test ' includes an acceptance enterion of a 90 percent. The team concluded that the oging factor of 1.11 and the temperature correction factor of 1.04 were acceptabl COEh1510DS The adequacy of the de load profile is indeterminate based on a number of errors that have been identified in the calculation. The licensee's operability assessment indicated that the 125Vdc system remains capable of performing its design function. Based on the review of the operability assessment, the inspectors concurred that the 125Vdc system was operable. In addition to the errors noted in the calculation, deficiencies were identified in the area of acceptance criteria for the surveillance tests associated with the 125Vdc and 250Vdc systems and in the testing associated with the 4160V breakers at minimum voltage conditions. The overall adequacy of the 125Vdc load profile calculation was characterized as an un,esolved item pending further NRC reVic E Risk Assessment Associated with Potential Degraded Reliability of 4160V Dreakers hispection Seppa The team reviewed the Risk Evaluation PSA-ES031, Revision 0, which assessed the potential core damage contnbution of continued short term operation with degraded 4160V breakers due to failure to perform the sendor recommended rnaintenanc Ohservations aQf Findinos The licensee performed a risk assessment using probabilistic safety assessment techniques to quantify the increase in core damage frequency associated with potentially degraded breakers. Additionally, the individual breakers were ranked with respect to risk importance measures to aid in developing a decision making strategy to prioritize breaker overhaul and refurbishmen . --- --. . - . _ _ - . __. . . . . _-

.

e e-19-The licensee increased the assumed failure rates for the six essential and all nonessential breakers by one order of magnitude and quantified the plant specific p obabilistic safety assessment modelin order to develop an estimate of the potential increase in risk associated with lack of breaker maintenance. This approach yielded an estimate of an increase in core damage frequency of approximately 1.4E 05/yr abono the base case value of 2.5E 05/yr. (This is an increase of approximately 78 percent.) '

Using the industry guidelines for * temporary" risk increases, the licensee determined that this postulated increase could be tolerated for approximately 30 days before actions would be necessary to reduce the risk to acceptable levels. The licensee then performed a risk ranking, based on the Fussel Vessely importance measure, to determine the prionty of circuit breaker overhauls. The resul'.s of the licensee's evaluation indicated that the emergency diesel generator output breakers (EG2 and EG1) would provide the greatest reduction in risk with respect to increases in reliability of the breaker Consequently, the output breaker for EG2 was replaced on November 3,1997, and the corresponding division 1 breaker was replaced onNovember 6,1997. The intpectors determined that the licensee's approach to evaluating the potential risk, as well as priontizing the plant efforts to correct the problem were appropriat Conclusions The team concluded that the licensee's approach to evaluating the potential risk associated with the lack of breaker maintenance was appropriate. The use of a risk-ranking methodology to prientize plant maintenance ports was regarded as a strength in the planning proces E2.3 Operability Assessroents of 4160V GE Magne-Blast Breakers Inspection Scope The inspectors reviewed the licensee's basis for the operability of the 4160Vac safety-related breakers that had not been overhauled. Additionally, the team reviewed the licensee's compensatory measures associated with the potentially degraded breaker Qbiervations and Findinos The inspectors reviewed the licensee's operability assessment associated with the potentially degraded condition of the six 4160V GE Magne-Blast breakers that had not been overhauled in 23 years of plant operation. The licensee's assessment of breaker operability was contained in Problem identification Report #217801, The liceitsee determined, and the inspectors agreed, that based on the information available, the breakers were operable but in a potentially degraded condition.

!

l

_ _

_ _ _ _ _ .

• i C

The team determined that the licensee's operability determination was adequate, but that certain waaknesses existed in the assessment in particular, it was noted that the l operability assessment did not incorporate the manufacturer's (GE) recommendation regarding opening and closing the breaker at the minimum voltages. In a letter dated October 15,1997, GE stated that, due to the fact that operating the breaker at the minimum voltages would represent the limiting condition, this would be the best indicator of the condition of the lubricant. The licensee's basis for operability included an assessment of the historical number of cycles to which the breaker had been subjected, visualinspections and manual cycling of the breaker for sigas of stiffness but did not include verification of breaker operation at the minimum expected closing coil voltage (Additionally, the inspectors noted that the GE report on the as found condition of the spare breaker that was sent in for overhaul reported " stiff-grease.") Conversations with GE representatives conducted on October 22,1997, indicated that GE believed that the breaker that was recently sont in for overhaul was operable but that it would be inadvisable for it to remain in service for more than about 2 months. Additionally, this determination was based on the assumption that the opening and closing operations had been venfied at the minimum voltages. This opinion was extended to include those breakers currently in service that had not yet received any overhaul in addition to the issue involving verifica%n of the breaker operation at minimum control voltages, the inspectors determined that the licensee's basis for operability with respect to the effects of continuously running spring charging motor was deficient. The licensee's Engineering Evaluation IEE 97-308," Affect of a Continuously Running Spring Charging Motor," was used as a supporting basis for operability. However, the licensee determined that the design calculation on which this evaluation was based, NEDC 87-131C,D, was deficient in that the inrush values for the spring charging motor currents had not been considered. (The calculation only evaluated the steady state impact of the motors.) As such, the engineering evaluation was based on a nonconservative design input. It was determined, however, that sufficient margin existed to account for this nonconservatism. Subsequently, the licensee revised the original operability assessment to address this issue, c. Conclusions The team concluded that the licensee's basis for the operability of the 4160V breakers that had not been overhauled was adequate, as revised. However, weaknesses were identified in the operability assessmen IV. Manacement Meetinos X1 Exit Meeting Summary The licensee and the NRC conducted an exit meeting on November 17,1997, to discuss the results of the inspection. The licensee acknowledged the findings and conclusions

. _ . _- -

O

.

- 21-that were presented. The bcensee indicated that n0ne of the materials that had been presented to the inspectors or discussed during the inspection were proprietary.

,

-,y

_ _ - - _ _ . . - - _ _ - _ _ . - - _ _ . _ _ _ _ _ _ . . _ . _ _ _ _ _ _ . _ _ _ _ _ . . _ _ _ . . .

,

• ;

,,

-

o AIIACHMENI  ;

i SUPPLEMENTAL INFORMATION

i PARTIAL LIST OF PERSONS CONTACTED  :

?

Licensee ,

!

M. Baldwin, Senior Electrical Engineer

.

D. Buman, AE Team Leader

- P. Caudill, Senior Manager, Safety Assessment l R. Field, Senior Core Cocling System Engineer '

R. Fosbinder, Electrician

- P. Graham, Vice President, Nuclear W. Hofmeister, Senior Maintenance Engineer -

B. Houston, Licensing Manager T. Jamar, Senior Electrical Engineer J. Lindinger, Core Cooling Engineering Supervisor .

R. Moberly, Senior Electrical Engineer C. Moeller, Senior Staff Engineer B. Newell, Assistant Maintenance Manager 0; Olson, Plant Engineering Manager M. Peckham, Plant Manager

!

J. Pelletier, Senior Manager, Engineering

'

D. Reeves, Senior Staff Engineer *

R. Rexroad, Electrical Engineering Supervisor G. Smith, Quality Assurance Manager

- K. Sutton, Senior Reliability Engineer ,

M, Unruh, Mechanical Supervisor R. Wachowiak, Reliability Engineering Supervisor NRC i

~

K. Brockman, Deputy Director, Division of Reactor Projects S. Matur, Team Leader, Office of Nuclear Reactor Regulation "

M. Miller, Senior Resident inspector, Cooper Nuclear Station

.

INSPECTION PROCEDURES USED

TI XXXX/XX - Medium Voltage Circuit Breakers -

- 62700: Maintenance implementation _ ,

62702 Maintenance Program =

93809 Safety System Engineering Inspection a

!

,

- n , - --,i-- ~ .--mv.- .+M1-- +a---*-9-m v w -= + + - -vm - < ---7- -v= w

o e

.

2 ITEMS OPENED

.QDftned 50/298/9718-01 IFl Adequacy of Licensee Program to review and Disposition GE Service Advice Letters (Section M1.1).

50-298/9718-02 URI Corrective Actions and Root-cause Assessment Associated With 4160V Breaker Maintenance (Sections M2.1, M2.3)

50-298/9718-03 IFl Adequacy of the licensee's maintenance program for 480V Circuit Breaker- (Section M3.1).

50-298/9718-04 URI Adequacy of the 125Vdc Load and Voltage Study (Section E2.1)

LIST OF DOCUMENTS REVIEWED Cooper Nuclear Station Standard for Preventive Maintenance, " Breakers and Control Circuits,"

Approved October 23,1996 Night Order 97 017, October 6,1997 Night Order 97-020, Octot'er 10,1997 Night Order 97 021,0,:tober 10,1997 Problem Ident:6 cation Report 217801, October 7,1997 Prob!cn identification Report 2 11681, October 20,1997 Problem identification Report 2-11682, October 20,1997 Problem identifica%n Report 216753, October 21,1997 Problem Identification Report 2-17171, October 22,1997 Problem Identification Report 217865, October 17,1997 Problem Identification Report 218271, October 16,1997 Problem Identification Report 118273, October 15,1997 Problem Identification Report 2-18332, October 24,1997 Problem identification Report 218375, October 24,1997 Prnblem identification Report 2-18948, October 21,1997 Problem Identification Report 218949, October 22,1997

'

Problem identification Report 218985, October 21,1997 Problem identification Report 2-19036, October 22,1997 Problem identification Report 2-19040, October 23,1997 Problem Identificatbn Report 219049, October 25,1997 Problem identification Report 219050, October 23,1997 Problem Identification Report 219056, October 21,1997 Problem identification Report 2-19060, October 16,1997 Problem Identification Report 219063, October 22,1997

.___ _ - -

o-

.

.

e 3-Problem identification Report 219065 October 19,1997 Problem Identification Report 219082, October 22,1997 Problem Identification Report 2-19086 October 21,1997 Problem identification Report 2-19092, October 23,1997 Problem Identification Report 219098, October 27,1997 Problem Identification Report 219235, October 21,1997 Problem identification Report 2-10247, October 22,1997 Problem Identification Report 219750, October 23,1997 Problem Identification Report 2 20160, October 22,1997 Problem Identification Report 2 20161, Octnber 21,1997 Problem Identification Report 2 20163, October 22,1997 Problem identification Report 2 20164, October 22,1997 Problem identification Report 2 20169, October 22, 'i997 Problem Identification Report 2 20183, October 22,1997 Problem identification Report 2-20407, October 20,1997 -

Problem identification Report 2 20408, October 20,1997 Problem Identification Report 2 20438, October 24,1997 Problem Identification Report 2 20820, October 21,1997 Problem identification Report 2 20822, October 22,1997 Problem identification Report 2-20823. October 23,1997 Problem Identifict. tion Report 2 21704, October 22,1997 Problem Identification Report 217816, October 10,1997 Problem Identification Report 2-03174, October 30,1997 Procedure 2.1.11, ' Station Operators Tour,' Revision 73c1 Procedure 7.3.2.1, "Westinghout e DB 50 Breaker Maintenance and Testing," Revision 6 Procedure 7.3.17, '4160V Breaker Maintenance ano Testing," Revision 16 Resolution Matrix, 'Magne Blast Breaker issue," October 20,1997 Service Advice Letter 073-3S4.1, August 25,1995 Vendor Manual 0234, ' General Electric Magne Blast Circuit Breaker (gel 88771)"

- Vendor Manual 0230, ' Westinghouse 480V Unit Substation and Low Voltage Switchgear*

'

NEDC 87-131C, "125VDC Load and Voltage Study " Revision 7 NEDC 87-131D, *125VDC Load and Voltage Study," Revision 6

'

, .

,+=

a I - -

-. - -

,

. .. - _ _ _ _ _ _ _ . - _ _ _ _ _ _ . . _