IR 05000352/1998005

Insp Repts 50-352/98-05 & 50-353/98-05 on 980601-24. Violations Noted.Major Areas Inspected:Backlogged Engineering Activities

ML20237A312
Person / Time
Site:	Limerick
Issue date:	08/07/1998
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20237A307	List: IR 05000352/1998005 ML20237A305 ML20237A309
References
50-352-98-05, 50-352-98-5, 50-353-98-05, 50-353-98-5, NUDOCS 9808130265
Download: ML20237A312 (29)

Text

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

.

C U.S. NUCLEAR REGULATORY COMMISSION

REGION I

Docket Nos: 50-352,50-353 License Nos: NPF-39, NPF-85 Report Nos: 50-352/98-05 50-353/98-05 Licensee: PECO Energy Facility: Limerick Generating Station, Units 1 & 2 Location: Limerick, Pennsylvania Dates: June 1 - June 24,1998 Inspectors: A. Lohmeier, Senior Reactor Engineer, Team Leader B. Buckley, Senior Project Engineer A. Della Greca, Senior Reactor Engineer S. Hansell, Jr., Resident inspector L. James, Reactor Engineer J. Yerokun,' Senior Reactor Engineer

.

Approved by: Glenn W. Meyer, Chief, Civil, Materials and Mechanical Engineering Branch Division of Reactor Safety I

<

l

)

I 9908130265 990007 PDR r"

0 ADOCK 05000352 pg u

L___________________.___ . _ _ _ __ _ _ .

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

,

d TABLE OF CONTENTS PAGE!

. EX EC UTIVE SU M M ARY- . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii E2 Engineering Support of Facilities and Equipment .......................1-E Standby Liquid Control Event ............................... 1 E2.2 Operability Determinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 E2.3 Eng'Aeering Work Bac klog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

-

E3 Engineering Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 E3.1 Agastat Relays in Mild Environments . . . . . . . . . . . . . . . . . . . . . . . . . . 3 E3.2 ' Seismic Response of Agastat Relays . . . . . . . . . . . . . . . . . . . . . . . . . . 5 E3.3 Agastat Relays in a Harsh Environment ........................6 E3.4 Safety Evaluation Program .................................8 E3.5 Safety Evaluation Program implementation . . . . . . . . . . . . . . . . . . . . . . 9 E3.6 Procurement of Commercial Grade Parts . . . . . . . . . . . . . . . . . . . . . . . 11 E3.7 D e sig n Ba se s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2 E4- Engineering Staff Knowledge and Performance . . . . . . . . . . . . . . . . . . . . . . . 13 E Engineering Knowledge and Performance . . . . . . . . . . . . . . . . . . . . . . 13 E7 Quality Assurance in Engineering Activities . . . . . . . . . . . . . . . . . . . . . . . . . . 14 E7.1 Problem Identification and Resolution . . . . . . . . . . . . . . . . . . . . . . . . . 14 E7.2 ' Engineering Controls and Self-Assessment . . . . . . . . . . . . . . . . . . . . . 18 E8 Miscellaneous Engineering Issues (IP 9 2 902) . . . . . . . . . . . . . . . . . . . . . . . . . 19 E8.1 (Open) Licensee Event Report 50-352; 353/97-010:

Potential Containment Bypass Path Resulting in a Condition Outside the Design Basis ................................19 E8.2 (Closed) Licensee Event Report 50-35 2/98-002 . . . . . . . . . . . . . . . . . . . 20 E8.3 (Closed) Licensee Event Report 50-3 5 2/98-005 . . . . . . . . . . . . . . . . . . 20

'

E8.4 - (Closed) Licensee Event Report 50-3 5 2/98-007 . . . . . . . . . . . . . . . . . . 21 E8.5 (Open) Licensee Event Report 50-3 5 2/98-008 . . . . . . . . . . . . . . . . . . . 2 2 E8.6 (Closed) Licensee Event Report 50-352; 353/98-009 . . . . . . . . . . . . . . 22 X1 Exit M e e ti ng . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 l

l-i ii

'

l

_

_ i

.

P EXECUTIVE SUMMARY Enaineerina j

.The team concluded that troubleshooting activitiec and the root cause determination for the SLC unplanned actuation were well planned and effective. Activities were conducted safely by technically competent individuals. The root cause determination was correct and insightful. (E2.1)

Operability determinations were acceptable and timely. Applicable root cause analyses were detailed and acceptable, and the depth of evaluations was commensurate with the significance of the issues. (E2.2)

The team concluded that backlogged engineering activities, both in the plant system and design engineering areas, were manageable and properly controlled. A sample review of the description of the activities found no items with potential safety significance on which resolutions lagged. (E2.3)

PECO had established a generally effective program to evaluate and control the replacement of safety-related Agastat GP/EGP type relays, including detailed monitoring of relay performance. However, the team concluded that corrective actions regarding replacement of Agastat relays in three panels which perform at higher temperatures (95 F)

had been ineffective and represented a violation. These relay replacements were deficient in that relays remained in service longer than intended, experienced increasing failure rates, and caused the CREFAS system to be unable to respond to an actuation signal on April 4,1998. (E3.1)

The team concluded that the qualified life calculation for Agastat series GP/EGP relays in a harsh environment included several discrepancies that primarily affected normally energized relays and would be reviewed by NRC following actions to resolve these discrepancies. (E3.3)

In general, the procedures, controls, and training supporting the 10 CFR 50.59 safety evaluation process provided comprehensive guidance and were found to be acceptable.

l (E3.4)

10 CFR 50.59 safety evaluations were of good quality and performed in accordance with j

'

the requirements of 10 CFR 50.59 and the applicable PECO procedures by trained, qualified personnel. The 10 CFR 50.59 Program was well established and was being implemented well. No indications of unauthorized changes were identified. (E3.5) j i

l l

i lii i

,

O l The team concluded the procurement of commercial grade parts and the dedication j

'

program for such parts in nuclear safety-related applications were acceptable. Based on the packages reviewed, the team also found that the program was satisfactorily implemented. (E3.6)

Based upon the review of DCRs and PIMS, the team concluded that PECO had revised design basis documents in an acceptable and timely manner. (E3.7) l

'

l l

'

System managers were actively involved in plant activities and their support of such activities was acceptable. Evaluations of events in different systerns had been well documented, thoroughly evaluated and comprehensive. System managers had a good general knowledge of their assigned systems, the current issues, and the technical resources available to them for resolution. (E4.1)

The team concluded that the process for corrective action (PEP), including deportability and I operability determinations, operating experience feedback, self-assessments, and oversight committees were generally appropriate and effective. A range of issues and problems had

,

been documented, categorized, analyzed, and addressed. Nonetheless, the team noted an l operability determination that was inconsistent and potentially confusing to operator Further, the team found that numerous operational challenges had occurred due to repetitive equipment and system design problems and the above processes did not appear to have been effective in reducing the challenges. (E7.1)

The team concluded that PECO's self-assessments of engineering practices were acceptable. Also, QA was effective in identifying and documenting unacceptable engineering practices. (E7.2)

t i 1

'

l

!

iV

- _ - - _ - _ _ - - _ _ - _ _ _ _ _ _ - _-_ _ _ - _ _ _

..

..

Reoort Details E2 Engineering Support of Facilities and Equipment E Standbv Liould Control Event Inspection Scone (37550)

The team evaluated the engineering follow up activities following the unplanned actuation of standby liquid control (SLC) system pump C on June 3,1998. On June 3,1998, while in cold shutdown, Unit 2 experienced an unplanned actuation of the SLC system. The system's train C (pump C and associated squib valve)

actuated for no apparent reason After about 6 minutes of operation, operators were able to secure the pump and terminate the injection of boron into the reactor coolant system. PECO immediately established an investigative team. The NRC team assessed activities primarily in the troubleshooting and root cause determination area Observations and Findinas

'

The team found the root cause determination to be correct and insightful. PECO identified the root cause of the actuation to be due to an electromagnetically induced interference on a redundant reactivity control system (RRCS) cable. A six foot stretch of the RRCS 140 VAC cable, associated with the SLC pump C logic, was discovered in the proximity of a main steam relief valve (MSRV) 125 VDC cable. At the time of the SLC actuation, operators had been operating the MSRV for planned testing. PECO was able to repeat the event on a test basi The team attended a PORC meeting on June 7,1998, during which the results of the investigative team's efforts were discussed. PORC members were thorough and plant personnel were well prepared. There was good discussion 'of the issues with good safety focu The team determined that troubleshooting activities were comprehensive and well controlled. The packages used for the activities included the acceptable 50.59 safety evaluation Conclusions The team concluded that troubleshooting activities and the root cause determination for the SLC unplanned actuation were well planned and effective. Activities were conducted safely by technically competent individuals. The root cause determination was correct and insightfu !

l

.

.

E2.2 Operability Determinations Insoection Scope The team reviewed a sample of operability determinations (ODs) selected from a list of nonconformance reports issued since April 1997 to assess the adequacy of the supporting technical and regulatory bases. The team reviewed approximately 15 operability determinations addressing a variety of issues and equipment, e.g.,

Engineering Change Request (ECR) 97-01099,"Miscalibration of Flow Loops"; ECR 97-02636," Discrepancies Between Schematics and LGS Calculation 6900.E21";

ECR 98-00180," LGS U1&2 HCU Scram Pilot Valve Electrical Termination Deficiency"; ECR 98-00648," Degraded Barrier Between Class 1E and non Class 1E Circuits"; and ECR 98-01063," Separation Criteria Violation in Panel 20-C790." Findinas and Observations The team found the documentation of the issues to be detailed with a good assessment of the condition. Operability was reviewed in terms of the ability of the components to perform their safety functions and within the context of the design and licensing bases. Root cause analyses were usually detailed and provided good insights. The depth of the analyses was usually appropriate and commensurate to the significance of the issue. When applicable, appropriate followup actions were delineated. Reviews were conducted in a timely manne Conclusions Operability determinations were acceptable and timely. Applicable root cause analyses were detailed and acceptable, and the depth of evaluations was commensurate with the significance of the issue E2.3 Enaineerino Work Backloa Inspection Scoce The team reviewed backlogged engineering activitie Observations and Findinas The team found the quantity of backlogged engineering activities was reasonable r and manageable in site design engineering a total of 568 engineering change

!- requests (ECRs) were awaiting engineering disposition. Of these,404 were not nonconformance type issues and the number had been reduced by approximately 25% in the last 12 months. Nonetheless, PECO considered the design change request type ECRs (164) to be too high and a problem. The backlogged volume had been recently reduced slightly, but it had remained nearly constant during the last 12 months. To address this concern PECO had established a team to reduce the backlog and to identify areas of improvement.

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

(

.

l

l 3 i l

in addition to the above, PECO listad 86 material evaluations requiring disposition and an additional 154 ECRs that were awaiting physical work for closure. Both of these groups showed a downward trend for the last 12 mnnths. The team's sample review of the backlogged activities descriptic, s in the design area identified no issues that had potential safety significanc In the system engineering area, the team reviewed on a sempting basis the ,

description and due dates of activities assigned to the electrical engineering grou !

The team determined that of the 48 PEPS only six were more that six month ol ]

Based on the item descript. ion, none of the items involved safety concerns. The )

same was true of the approximately 140 backlogged action requests also assigned to the electrical grou Conclusions The team concluded that backlogged engineering activities, both in the plant system ,

and design engineering areas, were manageable and properly controlled. A sample review of the description of the activities found no items with potential safety significance on which resolutions legge E3 Engineering Programs E3.1 Aaastat Relavs in Mild Environments Insoection Scope (37550)

The inspection evaluated actions to address service life failures of Agastat relays in mild environments and the ability of safety-related systems to perform their functions consistent with Information Notice (IN) No. 84-20, " Service Life of Relays in Safety Related Systems." (IN 84-20 addressed that the service life of all relays in the normally energized state is significantly shorter than when used in a cycled or normally deenergized application). [b_pyvations and Findinas A computerized database existed for all Agastat relays (approximately 3400 total with about a third being safety-related). The team found the database to be comprehensive in information and to contain impc tent details like installation dates, relay functions, and failure modes. The database had been used by PECO in a variety of ways to evaluate the relays' performance, including development of failure ate plot The plots showed that in the last three years the failures per year had been increasing considerably. For instance, for Unit 1 and common, Limerick experienced 14 failures in 1993,20 in 1994,20 in 1995,31 in 1996, and 40 in 1997. In comparison, the same plot showed that based on a rate of 1.33 failures per million hours, as specified in IEEE Standard 500-1984,the expected quantity of r

- _ - _ _ _ _ _ - ______ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ . ._ _ _ _ _ - _ _ _ , _ _ _ _ _ - - . _ - _ _ _ - _ - - _ _ _ _ - _ _ _ _ _ _

.

.

failures per year should have been approximately 12. Although the quantity of failures specified above include both age and nonage-related failures in approximately equal amounts, the plots showed that both types of f ailures were increasing at approximately the same rat The team Judged that PECO was aggressive in replacing relays due to aging conccins. PECO replaced 284 relays during the Unit 1 refueling outage earlier this yeu. Previously,73 relays had been replaced in 1996 and 85 in 199 Approximately 400 safety-related relays in Unit 1 and common applications remtJned to be replaced. Discussions with PECO bdicated that the responsible engin9er had solicited recommendations from system managers regarding relay replacements and approximately 231 are scheduled to be replaced later in 1998, with al additional 113 in 1999 and 50 in the year 200 Howes er, in a January 19,1990, letter to the NRC, PECO stated that using availabla test data and the Arrhenius technique, a time versus temperature plot had been calculated that would be used for establishing the service life of the safety- 1 related Agastat relays in mild environments. Using this plot and a maximum panel internal temperature of 95* F, they established the minimum service life of normally-energized relays to be eight years. For Unit 1, replacement of the relays should have begun in 199 i Despite the comprehensiveness of the relay monitoring program in general and the l aggressiveness with which relays had been replaced of late, the team expressed a concern regardmg the relays that had yet to be replaced in higher temperature i conditions. PECO had not replaced some safety-related relays that had operated for more than 14 years in relay panels with an internal temperature of 95 F or mor ;

Several of these relays were not scheduled for replacement for two more years. For

'

instancs, for relays in panels OA, OB, and OC-C124 panels, with internal temperature of approximately 95 F,(temperature measurements taken on March 30,1998) PECO established (Issue Number 10008186) service lives of 7.7, 7.0, and 8.0 years, respectively. For these panels, as predicted by the calculation, the failure rates were well above the overall average. Thus, in panel OA-C124 six of the 36 Unit 1 and common relays had failed at an average age of 10 years. At least five of these failures were classified by PECO as age-related. Similarly,in panel OC-C124, seven of the 22 relays had failed with at least five of the seven relay failures due to age degradatio The team was also concerned that despite the current experience with the relays in these same cabinets and the calculations which indicated service lives between seven and eight years, for those relays that already had been replaced, PECO had set a new replacement date approximately 14 years from their last replacement date, without a technical basis or justification for extending the service life beyond the calculated values.

!

- _ _ _ _ _ _ _ _

- _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

.

.

The above concerns were further heightened by an April 4,1998, age-related failure of a relay (03-DV138Y)in the B train logic of the control room emergency fresh air supply system (CREFAS). The failure of this relay prevented the redundant B train from automatically starting on demand while the A train was out of service for different reasons. The following day, the corresponding relay (03-CV138Y)in the A CREFAS train failed to initiate the system due to a contact failure (nonage-related).

The concern with this issue was that a random relay failure (A train) combined with a degraded relay failure (B train) prevented the redundant CREFAS trains from performing their safety functio Accordingly, the team determined that PECO's corrective actions on safety-related relays were ineffective and represented a violation of 10 CFR 50, Appendix B, Criterion XVI, Corrective Action. Specifically, PECO permitted safety-related Agastat relays in higher temperature (95*F) conditions to remain in service beyond their intended service life and to experience corresponding higher failure rates than expected. (VIO 50-352;353/98-05-01) Conclusions PECO had established a generally effective program to evaluate and control the replacement of safety-related Agastat GP/EGP type relays, including detailed monitoring of relay performance. However, the team concluded that corrective actions regarding replacement of Agastat relays in three panels which perform at higher temperatures (95*F) had been ineffective and represented a violation. These relay replacements were deficient in that relays remained in service longer than intended, experienced increasing failure rates, and caused the CREFAS system to be unable to respond to an actuation signal on April 4,199 E3.2 Seismic Response of Aaastat Relavs During the review of age-related failures of normally energized relays in a mild environment, the team reviewed the ability of the safety-related relays that had yet to be replaced to perform their safety function during a seismic event. Specifically, the team was concerned that contact chatter in a relay aged beyond its service life might, during a seismic event, inltiate or prevent an action that would interfere with the normal shutdown of the plant.

l PECO provided the results of seismic tests performed by South West Research Institute (SWRI), Report No. 04-1738-001, dated December 1988. This report was l not part of the Limerick documentation package and the licensee had not verified the applicability of the seismic test response spectrum to the Limerick required .

response spectrum. Therefore, the team did not review the report, nor did it ascertain the test purpose, scope, or methodology. However, PECO stated that, based on the results of the SWRI report, contact chatter occurred only in normally closed contact of relays in the de-energized o ate. PECO also stated that the contact chatter of aged reiays (15 milliseconds) was comparsble with that of unaged relays ,

(20 milliseconds). I i

.

Q_____________

_

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

.

.

The team was unable to determine whether seismic accelerations of the magnitude expected at Limerick could induce contact chatter of the durations measured by SWRI, and whe? v '.nis chatter of normally closed contacts of de-energized relays could be toleratet n all Limerick relay applications. This issue will be evaluated further by the NRC, following appropriate review by PECO. (IFl 50-352:?53/98-05-02)

E3.3 Agpstat Relavs in a Harsh Environment Inspection Scoce The team reviewed calculations and other applicable documents to determine the replacement program for normally energized Agastat GP/EGP type relays in harsh environment Observations and Findinos PECO established the qualified life of Agastat GP/EGP series relays in Calculation No. LE-089, Revision 1, dated April 7,1997. The methodology used in the calculation was the industry-accepted Arrhenius regression analysis. While reviewing the inputs to this calculation, the team observed that the operating temperature used was that of the mounting area of the relay panels and did not account for panelinternal heating. In addition, for post-accident temperature the calculation did not include any margin, as recommended by NUREG-0588. The team was concerned that both of these issues might negatively impact the qualified life of the relays. For instance, Table D of the above specified calculation showed that a normally energized relay in an 85*F arnbient temperature has a qualified life of 12.7 years. The same relay in a 90 F ambient temperature has a qualified life of only 9.9 years. A panelinternal heat rise of 5 F or more is not an uncommon occurrenc Regarding the normal operating temperature, PECO stated that the one used in the calculation included panel heat rise and provided records of panel internal temperatue measurements that had been taken since January 1996. The team's review of the data confirmed that the temperature used in the calculation was comparable to that measured in several panels. The review also determined that the measured temperature was highly dependent on the external temperature and on the power output of the reactor. Therefore, the team asked PECO whether: (1)

one measurement per month was sufficient to establish the panel internal temperature; (2) the measurements taken in certain panels could be extended to other panels that had not been monitored; and (3) the measurements had been taken with door closed and had allowed sufficient time for the internal temperature

.

to stabilize after closure of the doors.

(

)

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

. _ _ _

.1

..

To address the team's questions, PECO took temperature measurements of two panel internals and of the ambient surrounding the panels. Two methods were used for these measurements. The first method involved a one time measurement of the temperature of the targeted areas using the same instrument and procedure used in the past. The second method involved the use of recorders that took temperature measurements every 10 minutes for seven days. The data showed that: (1) outside j the panel, the two methods yielded the same initial temperature; (2) inside the i panel, the initial temperature registered by the recorder was approximately three degrees higher than that measured with the original method; and (3) ths temperature both inside and outside the panels changed several degrees during each day, further displaying their dependence to the outside ambient temperatur Regarding the post-accident temperature, PECO provided time-temperature profiles j constructed in calculation No. M-76-276, Revision 1. These profiles indicated the existence of some margin, though it was not evident that this was sufficient to account for panel temperature rise as well as for the conditions (e.g., variations in commercial production of the equipment and test equipment inaccuracies) for which margin is specified in NUREG 058 i To identify margin availability PECO took measurements of the relay coil- i temperature rise while the relay was being cycled off and on once every five  !

minutes, in the'same manner the temperature had been cycled during the test used to calculate the qualified life of the relay. PECO determined that during the off time, the temperature never dropped to the value assumed in the calculation. Therefore, some margin could be extracted from the test result The acceptability of the qualified life of these relays represents an inspector l followup item pending the NRC review of actione to resolve the above d. discrepancies, including measurement of applicable panel internal temperatures, I determination of available margin in the accident profile, and applicability of these j determinations to the qualified life calculation. OR 50-352;353/98-05-03) j i

c. Conclusions l The team concluded that the qualified life calculation for Agastat series GP/EGP relays in a harsh environment included several discrepancies that primarily affected normally energized relays and would be reviewed by NRC following actions to

. resolve these discrepancie l l

I l i

I

l

.

. _ _ _ _ _ _ _ _ _ .. . _ _ . _ . .

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _

.

.

E3.4 Safetv Evaluation Proaram Insoection Scooe (37001)

The team reviewed selected program procedures and held discussions with PECO representatives to determine: (1) If proper procedural guidance had been established for implementing the requirements of 10 CR 50.59 for proposed changes, tests and experiments (CTE); and (2) if proper procedural guidance has been established for updating the Final Safety Analysis Report (FSAR), as required by 10 CFR 50.71(e).

The team also evaluated the 50.59 training and qualification program for consistency with PECO's commitments for 50.59 safety evaluation (SE) personne The team reviewed some of the procedures which provide guidance and responsibilities related to 10 CFR 50.59 safety evaluation for activities at Limerick Generating Station. LR-C-13,10 CFR 50.59 Reviews, contains the requirements for performing 10 CFR 50.59 reviews. It covers all aspects of changes, tests and experiments, and addresses the preparation, review, and annual reporting requirements associated with the program. LR-CG-13, Performing 10 CFR 50.59 Reviews, provides further instructions and guidance for implementing procedure LR-C-1 Observations and Findinas The team found that the procedures provided appropriate guidance for performing SEs, including a screening process to determine if the CTE constitutes a change to the SAR and if applicable, the preparation of a safety evaluation (SE) to determine whether or not an unreviewed safety question (USQ) exists. The team determined that program procedures were acceptable and up to date regarding guidance. The procedures adequately delineated the responsibility of the various individuals who prepared, process, and approve SEs. However, it was not clear what departm6nt had the overall ownership of the 50.59 progra PECO's 50.59 training program specified that all 50.59 SE personnel, as well as peer reviews and vendors, be qualified to conduct this activity. The team examined the training material and attended an ongoing 50.59 training lesson. The team found that the training program provided adequate guidance for the preparation of 50.59 SEs. The team noted that no written test or examination was required to become qualified. There was also no requalification requirement. The team did not identify any performance problem with these aspects, and PECO indicated that these aspects were being evaluate The team reviewed a sample of changes and concurred with the determination that no USQs was involved. The team reviewed the anaual 10 CFR 50.59 and Commitment Revision Report for July 1,1996 through June 30,1997. The report contained a summary of changes to the facility and procedures as described in the Safety Analysis Report, tests, and experiments that were implemented between July 1,1996 and June 30,199 ._

_ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ -

.

.

'

9 Conclusion in general, the procedures, controls, and training supporting the 10 CFR 50.59 safety evaluation process provided comprehensive guidance and were found to be acceptabl E3.5 Safety Evaluation Proaram implementation Inspection Scope (37001. 37550)

The team evaluated implementation of the safety evaluation program. The team reviewed a sample of safety evaluations to determine if the safety evaluations for permanent plant modifications, temporary plant alterations, and procedure changes addressed all safety issues pertinent to the associated modifications or changes, did not involve an unreviewed safety question (USO), or require a change to the technical specification. The team also verified that the changes described in the selected SEs had been appropriately incorporated into the update final safety analysis report (UFSAR) pursuant to the requirements of 10 CFR 50.71(e) or were being processed for incorporation. The team also reviewed the retention records of

,

selected CTEs to determine compliance with 10 CFR 50.59(b)(3), Observations and Findinos The team found that safety evaluations were prepared and reviewed by individuals who had received appropriate training. The SEs addressed pertinent documents and were technically sound. Some of the SEs reviewed are discussed in detail below:

Permanent Plant Modifications SEs for permanent modifications were reviewed at various depths. In general, they were accomplished or were being accomplished well. Specific examples included:

ECR Numbers LG 97-02093 and LG 97-03034 for Modification P000703. Unit 1 MSRVS A.C. and N Local Control The modification was needed to support the fire safe shutdown re-analysis under a thermo-lag reduction project and involved the rerouting of cables associated with the main steam relief valves (MSRV), and reactor vessel and suppression pool instrumentation.

'

The team found that the safety evaluation was detailed, and applicable calculations were included which addressed various issues satisfactorily. Affected system operating procedure changes were also addressed and included in the change c package. The team conducted walkdowris of portions of the cable reroute, the i remote shutdown panel area and the new PCIG and EDG reservoir connections. No physical discrepancy was identifie l

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - _ _ _-___- _ _ __ _ _ - _ _ _ _ __ - _______ _ _ _ _ _ - - _

-

.

.

ECR LG 96-03220.1 A Core Sorav Vent Line This modification installed a bypass line (with valves) around the 1 A Core Spray (CS) pump discharge check valve to create a controllable vent path to the suppression pool. The 10 CFR 50.59 safety evaluation was detailed and had the proper review and approval, ,The package included appropriate post modification tests, design considerations, and procedure revision Modification P781. (ECR LG 98-00124) ADHR --FPC & RHR System Chances The alternate decay heat removal (ADHR) modification was installed to provide for decay heat removal from the RPV and SFP during the common RHR shutdown cooling supply valve local leak rate tes The affects on RHR and FPC systems were properly considered. Technical specifications and UFSAR implications were properly addresse Temocrarv Plant Alterations Temporary Plant Alterations (TPA) are controlled by procedure MOD-C-07. The procedure contained the proper references for 50.59 considerations in implementing TPAs, including the appropriate reviews (initial and periodic) and approvals. The number of outstanding TPAs was relatively low (9), possibly an indication of a strong engineering support to operations. A sample of TPAs was reviewed in more .

detail as discussed belo TPA' 98-OO10 for Gaoaina RWCU Hx. Relief Valve PSV-044-209 The 50.59 screening appropriately determined that a SE was required. The SE was completed properl TPA 97-01183 for Drvwell Drain Cooler Flow Monitorina System The TPA removed one of six flow transmitters (FTs) from the summing circuit until the FT can be repaired or replaced. The system normally provides a control room alarm when the sum of the flows reaches 1 gpm above the current value of drywell leakage as determined by the system. Due to the loss of the failed FT, a new setpoint of 0.45 gpm was determined to be appropriate for the system to still perform its intended function.

I: The SE was detailed and technical. However, the team identified that the technical justification used for assigning the maximum amount of lost flow due to the failed transmitter did not include considerations for possible steam leaks from two reactor coolant pressure boundary pipes in the vicinity of the register of the broken F PECO generated PEP 10008571 to address the discrepancy performed an evaluation i- that included the possible steam leaks from the two pipes, and determined that the statement in the 50.59 SE was still correct because no more than 50% of a postulated leak can be drawn into the register.

l

. _ _ _ _ _ _ _ _ _ _______________________- ________ ___________

.

.

I 11

!

l De Facto Chanaes and other 50.59 Safety Evaluations j-The team reviewed safety evaluation tracking list and selected a sample of safety evaluations for reviews. Some PEP issues involving 50.59 evaluations were reviewed. PEP 10008160, Control Pin on #4 Fuel Pump for EDG 1B was reviewed I and found to be appropriate. PEP AR A1153404for troubleshooting and testing activities to address inoperable position indication for three drywell/ suppression pool r

'

vacuum breakers was also reviewed. The team found no safety issue with it, however, the team noted that there was a lack of consistency in the licensing basis i documents regarding the maximum operating drywell and suppression pool pressure. Action statement #1 of TS 3.6.4.1.c calls for a test on the vacuum breaker at greater than 0.7 psi. TS 3.6.1.6 allows drywell and suppression pool pressure to be maintained at -1 to 2.0 psig. UFSAR 9.4.5.1.2.3 states that during j reactor operation, containment pressure is normally maintained in the range of psig to 0.7 psig. PECO had a NCR (LG 92-00081) generated to address this conflict. The team did not identify any immediate safety concern since the plant j was maintained within the limits of all the document Assessments The team reviewed some self assessments implementation of the 10 CFR 50.59 l program. A recent PEP item (10007325) had been initiated to track and trend PEP issues related to 50.59 safety evaluations. This effort was good because it i reflected that not only was PECO reviewing each issue, they were also looking at j: ths potential trends. .The team also reviewed a sample of Site Engineering Self i l Assessment of 10 CFR 50.59, and some PECO QA surveillance reports on plant l l modifications, temporary alterations, temporary change process and procedure l p changes. These self assessment efforts were appropriate and demonstrate

satisfactory.

Conclusion  !

I

,

L 10 CFR 50.59 safety evaluations were of good quality and performed in accordance L

with the requirements of 10 CFR 50.59 and the applicable PECO procedures by trained, qualified personnel. The 10 CFR 50.59 Program was well established and was being implemented well. No indications of unauthorized changes were  !

identifie j g E3.6 Procurement of Commercial Grade Parts Insnection Scooe The team reviewed the procedures related to the procurement and the dedication of ;

commercial grade parts to evaluate the acceptability of the engineering review. The j team also assessed the implementation of such procedures, i

!

l l

I

_ _ _ _ _ _ _ _

.

.

12 Observations and Findinas The team's reviews of the procurement guidance procedures found the instructions sufficient for understanding of the process and program objectives. The team also found that the instructions were sufficient for the correct procurement, analysis, and dedication of commercial grade parts for safety-related application To address the program implementation, the team selected five recently prepared procurement and dedication packages in the electrical and instrumentation area, including gauges, fuses, temperature elements, and components no longer availabl The team found that the packages included acceptable descriptions of the application, the important characteristic of the equipment being replaced, acceptance criteria of evaluated parts, and the approved disposition Conclusions The team concluded the procurement of commercial grade parts and the dedication prograrn for such parts in nuclear safety-related applications were acceptabl Based on the packages reviewed, the team also found that the program was satisf actorily implemente E3.7 Desian Bases Scone (37550)

The team reviewed the maintenance of design basis information to evaluate the degree to which the engineering organization maintained the design basis documents (DBDs) current. Specifically, the team verified that design changes were being translated into DBDs, that the DBDs were being updated appropriately, and that changes between DBD revisions were posted in Plant Information Management System (PIMS). For this purpose, the review addressed the DBD change procedure, and the design change reports (DCRs) and PIMS for the reactor core i .olation system (RCIC) and residual heat removal (RHR) syste Obse rvations The '.eam verified through sampling that for several DCRs associated with the RCIC and 9HR DBDs, design changes were translated into the RHR and RCIC DBDs acceptably, updating of RHR and RCIC DBDs was timely, and changes between

revisions of the RHR and RCIC DBDs were posted in PIMS.

l Conclusion Based upon the review of DCRs and PIMS, the team concluded that PECO had revised design basis documents in an acceptable and timely manne _ _ - _ _ - _ _ _ _ _ - _ - _ _ - - _ - _ _ _ _ _ - _ _ _ _ _ _ - - _ _ _ _ . _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ _ _

.

.

E4 Engineering Staff Knowledge and Performance E Enaineerina Knowledae and Performance Inspection Scope (93802)

The team interviewed selected system managers, conducted system walkdowns, and reviewed applicable documents to evaluate the experience of the responsible engineers with the assigned systems and to assess their effectiveness in assuring the reliability of the systems. The team also evaluated their coordination with other department Observations and Findinas PECO has published quarterly reports of system performance based its availability and reliability. The team selected several systems that in the last report had been coded as having unacceptable performance. Discussions with the responsible system engineers determined that the unacceptable performance status of a system was typically tied to specific events that had rendered one or more component in the system unavailable or otherwise unreliable. The team's review of the documentation packages found that the events had been well documented, thoroughly evaluated, and comprehensive, including both equipment and human performance considerations. The team determined that PECO had appropriately reviswed each issue for its impact on system and component operability. For instance, the team found effective evaluations of an unexpected overloading of emergency diesel generator D22 during a diesel surveillance testing and the effects of a bus overvoltage condition resulting from the failure of emergency diesel D21 voltage regulator.

With these as with other examples of plant issues, the team found the system managers to be responsive to site needs and directly involved in troubleshooting, coordinating the resolution of technical issues, and interfacing with other

!

departments, as needed. Resolution of specific issues showed that the interfaces among engineering groups were good and the communication between engineering and operations was generally effective.

l

'

The team discussed with the system managers their experience with the systems assigned to them, the length of time they had been responsible for the systems, and the system training they had received. The managers usually displayed a good general knowledge of their assigned systems, current issues, and the technical resources available to the Conclusions System managers were actively involved in plant activities and their support of such activities was acceptable. Evaluations of events in different systems had been well documented, thoroughly evaluated and comprehensive. System managers had a good general knowledge of their assigned systems, the current issues, and the technical resources available to them for resolutio _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - -

.

,

E7 Quality Assurance in Engineering Activities

5 Problem identification and Resolution Insoection Scone (40500) -

The team conducted a performance based and programmatic evaluation to k determine the effectiveness of the Limerick Generating Station's (LGS's) controls in identifying, resolving, and preventing problems that affect station safety or are adverse to quality. The inspection was performed using the guidance of inspection

Procedure 40500," Effectiveness of Licensee Controls in Identifying and Resolving L Problems." i The team compared of the control room narrative log to the index of Performance Enhancement Program (PEP) issues for the past year, focusing on documentation of problems that met the PEP procedure threshold, the proper PEP classification level, i operability determinations, deportability determinations, and effectiveness of ,

corrective actions. The review included the experience assessment (EA) i department's involvement in the PEP process, various self assessments and oversight function . Observations and Findiras

- Performance Enhancement Proaram

,

.The primary process used at LGS for the identification and resolution of conditions

adverse to quality was the PEP. This process was also used to identify other l- performance enhancement opportunities. PECO procedure LR-C-10," Performance

!- Enhancement Program," provided direction for the implementation of the PEP program. Identified concerns are classified according to significance. The Level 1

issues are the most significant conditions adverse to quality that could result in a l major impact on plant safety or substantial hazard to the safety and welfare of the l' public or plant personnel Level 2 issues involve moderate challenges to plant or

[ personnel safety. Level 3 issues represent minor challenges to plant or personnel safety. Examples of Level 1, 2, and 3 issues are described in PEP procedure exhibit f, LR-C 10-3. In the past year, the PEP process had been revised to include more

! ; involvement from the line organizations (less direct involvement from EA) and to

!

reflect organizational change The team determined that the assignment of the PEP significance level and classification was generally appropriate and in accordance with the program guidance. Consistently, the more significant issues required a detailed review

, including a root cause analysis, while those of lower significance required less investigation.

L _ _ _ _ _ .- _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _

.

.

The team found communication and overview regarding PEP issues to be satisfactory. New PEP issues were presented at the deily management leadership meetings. The issues were classified by the responsible department managers, and discussion of the PEP topics was usually good. EA representatives attended the ,

morning meeting and continued to ensure the proper implementation of the PEP

{

program and related activities. The PEP performance indicators provided an '

excellent overview of the health of the PEP program and an indication where management attention is warranted to ensure the program effectivenes One procedure inconsistency was noted for the PEP classification level of a seriously injured worker and corrected. The overall program procedure, LR-C-10, sechon 7.4.2, listed examples of Level 1 PEP issues including fatality or serious injury. LR-C-10-3, " Examples of issues and Their Significance," listed a fatality as a Level l PEP and a serious injury as a Level 2 PEP. Following the team's identification of this inconsistency, LR-C-10 was revised to be consistent with LR-C-10- PEP Documentation Review and Root Cause Analysis / Corrective Action Proarams The team determined that PEP issues contained good documentation and the initiation threshold was appropriate. An example of an excellent root cause determination and timely corrective actions was noted for PEP No. 10008185, RWM surveillance test procedure discrepancy for the Unit 1 shutdown, on April 3,199 Some PEP narrative descriptions did not provide enough information to understand the actual proble Infrequently, PEPS were classified at a lower level when compared to the procedure classification criteria. For example, PEP Nos. 10007876, Late 1-hour report to the NRC for a TS required plant shutdown, and 10008407, Missed Unit 2 30 day CREFAS TS LCO for the D13 diesel outage, were classified as level 3B versus 2A/B as described in the procedure. The missed CREFAS TS PEP corrective actions were narrowly focused and missed an opportunity to discover a missed 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> CREFAS TS LCO which was ultimately discovered by the NRC two weeks late Reoortability and Operability Determinations The team reviewed a sample of PEP documents and the control room log during the past year to verify that deportability and operability concerns were addresse Overall, the deportability and operability determinations were addressed appropriately in the PEP documents. However, there were a few problems note On April 4,1998, a reactor water cleanup (RWCU) system engineered safety feature (ESF) actuation occurred and was not reported to the NRC as required by 10 CFR 50.72(b)(2)(ii) and 50.73(a)(2)(iv) for a four-hour non-emergency notifications and 30 day licensee event reports (LERs). The ESF actuation had been questioned by the shift manager (SM) and the RWCU system manager but not reported due to the outdated information contained in the site deportability reference manual.

!

l l

l L______.__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . __ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

- _ _____ _ _ . - _ _ _ _ _ - _

.

.

I The manual contained guidance from the draft NUREG-1022," Event Reporting Guidelines 10 CFR 50.72 and 50.73." Specifically, the SM assumed that because .

the possibility of a RWCU isolation was discussed at a pre-evolution brief, the subsequent closure of the primary containment isolation valves was not reportable in accordance with the draft guidance. NUREG-1022 had been revised in January 1998, to provide more detailed criteria for ESF deportability determinations and had deleted the draft guidance which would have permitted the RWCU isolation to not be reported. The deportability reference manual had not been updated to reflect the change in deportability criteri The deportability reference manual was revised on June 17,1998, to correct the error. A PEP was initiated to determine why the deportability manual was not updated when NUREG 1022 was revised. The PEP was also written to review additional procedures that may contain deportability guidance. In addition to the April 4 ESF actuation, two additional RWCU isolations on January 28, and March 23,1998, had not been reported to the NRC for the same reason On June 19,1998, PECO notified NRC of the missed ESF isolation notification Subsequent to the inspection, on July 20,1998, PECO issued LER 98-14 on these isolations. As the isolations were eventually reported and analyzed in an LER (i.e.,

corrected), the team determined that the late reporting violation should be treated as a non-cited violation, consistent with Section Vll.B.1 of the NRC Enforcement Policy. (NCV 50-352:353/98-05-04)

During control room log review, the team noted an example of inconsistent operability determination. Specifically, the operability standard provided to shift SROs from management was inconsistent for the degraded control room chillers and the chillers' impact on the CREFAS systems and potentially confusing to operator Technical Specifications do not contain an LCO for the control room chiller Indirectly, the control room chillers are needed to maintain control room air temperature below the TS limit of 85 degrees Fahrenheit. The operation policy was to enter a CREFAS 7-day TS LCO when a control room chiller was . inoperabl Typically, this guideline was followed. However, on two occasions for chiller equipment problems, the control room log noted that additional action would only start when the 85'F TS limit was reached. Operation management acknowledged the inconsistent operability guidance provided to shift SROs and stated their intention to provide a clear standard. The team noted that the determinations were always conservativ __ ______ _________________

,___ _ _ _ - _ _ - _ - __ _ _ ____ - _ - _ _ _ - _ _ _ _ _ - _ .

.

.

Operatina Exoerience Feedback The team judged Limerick's program for operating experience review to be very effective. The process provided the right amount of information from external sources to system managers, supervisors and training personnel to improve the plant processes on a daily basis, and there was ample evidence of corrective actions based on these insights. Significant industry experience was routinely

.

highlighted at the morning management meeting to ensure a broad audience can

, benefit from the information. Also, the nuclear network information was used for L the full evaluation of the Unit 2 standby liquid control (SLC) injection, in that l detailed information related to previous SLC injections at other sites was used to l resolve the issue at Limeric Self Assessment Activities and Oversiaht Groups (

'

The team reviewed various self assessment activities, including the annual station wide, individual department, and experience acessment evaluations, and nucicar review board and PORC activities.

l The team determined the nuclear quality assurance (NOA) group provided a positive l

'

contribution toward plant improvement. NQA had performed an objective and focused review of the PEP process. The detailed findings were documented in a j PEP that included multiple examples of PEPS classified at a lower level than the l criteria contained in the program procedure. A second NOA PEP was initiated to pinpoint the weaknesses in the corrective actions process related to the recent PEP program organizational changes.

i i The team found that the independent safety assessment group (ISEG) had performed an excellent and critical review of the Limerick operability process. The

, evaluation contained good bench marking of the Peach Bottom (PB) process and l other utilities with good operability determination programs. ISEG determined that

! unlike PB, Limerick did not adopt a concise operability guideline after Generic Letter (GL) No. 91-18 was issued. Since the ISEG review, Limerick has provided improved operability guidance to the shift senior reactor operator Members of the inspection team attended the June 4,1998, nuclear review board

'

L (NRB) meeting held at Limerick. NRB assessed the safety of plant operations and plant management's ability in identifying actual and potential problems, and in implementing corrective actions to prevent recurrence. From the review of the meeting minutes and observation of the meeting, the team judged that NRB provided an effective review of plant operations, engineering, quality assurance and PECO performance issue ,

i Operational Challenoes I Despite the apparent effectiveness of the above programs and activities, the team noted an apparent inconsistent outcome in the level of operational challenge During the review of the control room log of operational activities, the team found

.

that operators had been challenged to immediately to respond to repetitive equipment problems, such as Agastat relay failures, and known system design

.

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

_

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

..

.

problems, such as RWCU and control room ventilation, on numerous occasion These problems appeared to have caused entry into off-normal and emergency operating procedures (EOPs) at frequencies higher than expected or desirable. The team noted that operators had been effective in responding and resolving the operational challenge Conclusions The team concluded that the process for corrective action (PEP), including i deportability and operability determinations, operating experience feedback, self- 1 assessments, and oversight committees were generally appropriate and effective. A range of issues and problems had been documented, categorized, analyzed, and l addressed. Nonetheless, the team noted an operability determination that was inconsistent and potentially confusing to operators. Further, the team found that numerous operational challenges had occurred due to repetitive equipment and l system design problems and the above processes did not appear to have been effective in reducing the challenge E7.2 Enaineerina Controls and Self-Assessmgtrit Scope The team evaluated the effectiveness of engineering controls and self-assessments of the Updated Final Safety Analysis Report (UFSAR) verification program, DBD maintenance, and the process for identifying procedures affected by design changes. The team also reviewed two quality assurance (QA) surveillance reports, LSR-98-0111, " Engineering Change Request (ECR), Action Request (AR) Reviews" and LSR 97-0172, " Document Control / Procedure and Drawing Control," to evaluate the effectiveness of PECO's self assessment programs related to engineering activities. This evaluation verified completion of selected corrective actions recommended in the QA surveillance report ; Observations l

The team found the self-assessments to be generally effective and self-critica PECO's UFSAR verification program identified a variety of discrepancies and noncompliance. For instance, while reviewing an LPRM issue, itself a UFSAR discrepancy identified by the NRC (Inspection Report 98-02), PECO observed that a

'

calculation that support a UFSAR statement regarding spent fuel pool drain down doses was missing. This issue had been entered in the Plant Enhancement Process (PEP) program and also questioned the lack of identification by the UFSAR verification program. In another example, PECO identified some chemistry setpoint

. settings that were not as described in the UFSAR and had not been identified during the UFSAR verification proces )

I

_ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

- _ _ _ _ _ _ - - - - _ _ _ _ _ _ _ - _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ - _ - _ _ _ _ _ _ _ -

.

.

The DBD maintenance and procedure change evaluations were generally acceptabl The team reviewed the PEP and corrective actions associated with the DBD maintenance evaluation and concluded that the evaluation was thorough, and corrective actions being implemented were comprehensive. The team reviewed the evaluation of the process for identifying affected procedures during design change The team verified that the corrective actions identified were implemented and found the corrective actions acceptabl The team found that QA was effective in identifying and documenting unacceptable engineering practices via QA surveillance. The team found that the QA surveillance of ECRs and ARs had identified concerns regarding these processes and

>

that similar concerns had been raised by QA in a previous surveillance, LSR 98-0021. Both surveillance report results were documented in PEP 10007277. PECO had identified this issue as an area of concern and was working toward improving the AR initiation proces Conclusion The team concluded that PECO's self-assessments of engineering practices were acceptable. Also, QA was effective in identifying and documenting unacceptable engineering practice E8 Miscellaneous Engineering issues (IP 92902)

E (Ocen) Licensee Event Report 50-352:353/97-010: Potential Containment Bvoass Path Resultina in a Condition Outside the Desian Basis On August 21,1997, PECO reported that engineering had discovered a potential suppression chamber steam bypass leakage path between the drywell and the suppression pool air spaces. The bypass path could exist in the six-inch nitrogen supply line to the drywell and suppression chamber which includes a single air-operated outboard isolation valve and two. air operated inboard isolation valves (one each to the drywell and to the suppression chamber). PECO postulated that during a LOCA a single hot short affecting the cables of both inboard isolation valves could result in their opening and interconnecting the two areas. Isolation of the two areas is necessary to ensure adequate pressure suppression during the event.

< PECO had not completed their evaluation of the event significance and, therefore, l

had not developed permanent corrective actions, in the interim, however, they had j disabled one of the two inboard valves by shutting off the air supply to the actuator. The team confirmed the accessibility of the air supply valve and that proper administrative controls were in place. The team concluded that PECO's interim corrective actions were acceptable. - The item, nonetheless, is unresolved j pending the NRC review of event evaluation and permanent resolution of the issu (URI 50-352; 353/98-05-05)

.

l

. 1

E8.2 (Closed) Licensee Event Reoort 50-352/98-002: ESF Actuation Due to Blown Fuse Durino Jumner Insbilation On February 6,1998, while the RWCU system was being placed in operation, a fuse blew and resulted in various actuations of the Units 1 and 2 primary containment and reactor vesselisolation control system and of the Unit 1 reactor enclosure secondary containment isolation. The team determined that the blown fuse was caused by the inadvertent grounding of a temporary jumper that was being installed in accordance with procedure S44,1,4," Reactor Water Cleanup Cold Startup." PECO attributed the event to personnel error and the use of a jumper wire ,

without a retractable insulating cover. To minimize inadvertent grounding of jumper ]

wires in the future, PECO initiated action to remove all jumpers without protective i Cover Apparently to minimize pump trips and RWCU system isolation due to flow and !

pressure transients; the procedure requires that during system startup jumpers be installed. Recognizing that procedural requirements to install jumpers were contributing factors, PECO revised the procedure to add caution statements and heighten personnel awareness. In addition, they initiated action to evaluate alternative design changes to eliminate the use of jumpers and minimize operator l burdens during the system startup. The team concluded that a comprehensive i evaluation was ongoing and that the actions to address the event were acceptabl The team also concluded that no NRC regulations were violated during the even '

)

This item is close .

E8.3 (Closed) Licensee Event Reoort 50-352/98-005: RWCU PCIV lsolation due to Steam Leak and Load Center Breaker Trio On March 22,1998, the supply breaker to load center 124B tripped unexpectedly causing the de-energization of all the loads supplied by the load center, including several reactor enclosure HVAC fans. The team determined an incorrectly wired (reversed polarity) current transformer internal to the breaker, had caused the breaker trip prematurely. The breaker had been placed in operation only four days s earlier as a replacement for existing breaker that had been scheduled for testin l Apparently, PECO becaine aware of the potential wiring errors affecting the K-line low voltage circuit breakers in Novembar 1996, as a result of a similar event at {

another nuclear facility. At that time, PECO initiated action to revise the breaker i calibration procedure. The procedure, had not been revised in 1997 when the manufacturer issued a Part 21 regarding the potential for polarity reversal and was still unrevised when the new 124B'ioad center supply breaker was calibrated prior to its placement into servic As a result of the event PECO conducted a thorough investigation and found two additional circuit breakers with an incorrectly wired current transformer. One of these breakers supplied power to an emergency motor control center (D224-R-C).

A third breaker, load center feeder 114A-22 was found with a reversed current transformer.

!

I J

- _ _ _ _ __

.

.

The team also found that the evaluations and corrective actiores were appropriate and acceptable. However, when PECO originally determined, in November 1996, that the issue was applicable to Limerick, the actions to address it were inadequate and not timely, in that they failed to revise the cali5 tion procedure, which could have prevented the event, until after the event. In addition, as acknowledged in their analysis of the event, had PECO recognized, prior to the testing on March 18,1998, the existence of an "open change paper against the PM procedure" the event could have been averted. This is a second example of a 10 CFR50, Appendix B, Criterion XVI, Corrective Action violation. (VIO 50-352; 353/98-05-01)

E8.4 (Closed) Licensee Event Report 50-352/98-007: Three Inocerable ITT Barton Model 580A Differential Pressure Switches Result in Two or More Independent Trains of a Sinale Safety System Beina Inocerable From a Common Cause During the scheduled surveillance testing of the Barton differential pressure switches in the relayed emergency trip system, PECO found that three of the four switches had fallen below the Technical Specification allowable value of 465 psi (The function of the Barton switches is to provide an anticipatory trip signal to the end-of-cycle reactor recirculation pump trip system and to the reactor protection system.)

PECO replaced the four Barton switches with equivalent Barksdale ones, evaluated all other Barton differential pressure switches in safety-related applications, and determined that no similar drifts had been experienced in such applications. The selection of the Barksdale switches was based on their performance in Unit The team determined that PECO had evaluated the switch drift experience in conjunction with the 24-month fuel cycle review. Based on their conclusion that drift was due to vibration, PECO had installed a modification to minimize such i vibration. The drift measured in the subsequent cycle showed acceptable result Therefore, no drift outside the TS limit was anticipated. Nonetheless, they evaluated tne impact of a maximum instrument drift of 200 psig. They concluded I that this drift would have delayed the trip action of only 3 milliseconds and that such delay would have had minimalimpact on the overall TS-required response time of the trip function. Based on the above review, the team concluded that PECO had properly addressed the drift issue by replacing the switches with different design l and that the technical consequences of the drift had been minimal. This item is j close '

This violat on of the plant TS surveillance requirements was identified by PECO and corrected when identified. Based on the above, this violation is being treated as a 3

non-cited violation, consistent with Section Vll.B.1 of the NRC Enforcement Polic i

'

(NCV 50-352/98-05-06)

i l

,

l _____________________________b

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

l

.

,

l E8.5 (Open) Licensee Event Reoort 50-352/98-008: Pressure Setooint Drift of Five Main Steam System Safety Relief Valves Primarily Caused by Corrosion Induced Bondina The Unit 1 TS requires recalibration or replacement of seven of the 14 main steam j safety relief valves (SRVs). However, based on industry generic drift issues, the licensee committed to replace or recati' orate all 14 valves. During the April 1998 refueling outage, PECO replaced all 14 valves. Subsequently, testing of five SRVs determined that their setpoint had drifted upward by as much as 6.1%. The Limerick TS requires that for 11 of the 14 SRVs the setpoint tolerance to be within i1%. The ASME Code allows a 3% drif The upward setpoint drift of Target Rock SRVs has been an industry issue that is

,

still under review. The drift is the result of corrosion-induced bond between the pilot disc and seat. Based on industry recommendations, PECO had modified half of the plant SRV population and installed a modified platinum doped pilot discs in the pilot assemblies. Two of the five SRV tested had modified disc. Therefore, the modification failed to provide the desired results. PECO was evaluating the replacement of the pilot assemblies with a three-stage design that has been successful at their Peach Bottom facility. Based on PECO's replacement of the pilot valve assemblies, the immediate actions were acceptable. In addition, PECO was evaluating alternatives to the current design to prevent future system deficiencie This issue is close This violation of the plant TS surveillance requirements was identified by PECO and corrected when identified. Based on the above, this violation is being treated as a ;

non-cited violation, consistent with Section Vll.B.1 of the NRC Enforcement Polic l (NCV 50-352;353/98-05-07) l E8.6 (Closed) Licensee Event Report 50-352: 353/98-009: Primary Containment Overcurrent Protection Circuits for Six Electrical Penetrations On May 6,1998, PECO determined that six electrical circuits entering primary containment (three in Unit 1 and three in Unit 2) were outside the design bases of j

'

the plant, in that the electrical containment penetrations did not have dual protection over the entire range of faults. The circuits pertained to inboard isolation valve The team found that PECO had determined that the deficiency was the result of poor communication between different engineering groups. Apparently, during the licensing phase of the plant, in response to NRC questions, regarding Limerick's commitment to Regulatory Guide (RG) 1.63, PECO stated that the backup overcurrent protection in the long/short-time protection zone was provided by the thermal overloads supplied with the motor starter. At the same time, to satisfy a l commitment to RG 1.106 and ensure that the valves completed their safety I function, the thermal overloads were bypassed. With the thermal overloads bypassed, the redundant overcurrent protection was no longer providad to the circuit in questio I

l I

_ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ . _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _____ - _____ _ ___ _ __

'

,

l l

<

,

23 l PECO identified the deficiency during an internal review of thermal overload of l

protection schemes review. Although several weeks passed since the problem was  ;

originally discovered, PECO's plans for investigation and resolution of the issue  ;

i indicated an acceptable approach. To resolve the identified deficiencies PECO l replaced the backup instantaneous magnetic 'reakers with appropriate thermal  !

magnetic ones. The change provided full reoundant protection to the six i penetrations.

l PECO-identified and corrected violation of the NRC requirement regarding dual  :

! protection of primary containment electrical penetrations is being treated as a Non- l Cited Violation, consistent with Section Vll.B.1 of the NRC Enforcement Polic ,

(NCV 50-352:353/98-05-08)

X1 Exit Meeting l

The inspection team presented the inspection results to members of PECO management led j by Mr. VonSuskill, site vice president at the conclusion of the inspection on i June 24,1998. PECO acknowledged the findings presented. No proprietary information )

was identified by PEC PARTIAL LIST OF PERSONS CONTACTED Limerick Generatina Statiqn V. Angus, Engineer, BOP Systems M. Alderfer, Senior Manager, Plant Engineering l T. Bell, Manager, Reactor Engineering l S. Bobyock, Manager, Mechanical Design F. Cook, Senior Manager, Design Engineering E. Cullen, Manager, I&C S. Gamble, Manager, Electrical /HVAC Systems i J. Gilbert II, Systems Manager l J. Grimes, Director, Site Engineering J. Hutton, Senior Manager, Operations T. Moore, Manager, Experience Assessment S. Muntzenburger, Systems Manager G. Reid, Manager, Design Chango T. Tonkinson, Engineer, Experience Assessment J. Von Suskill, Vice President, LGS Site R. Swenk, NQA/ LOD R. Weingard, Acting Mr...ager, Equipment Focus U.S. Nuclear Reaulatory Commission A. Burritt, Senior Resident Team G. Meyer, Branch Chief, Components J. Wiggins, Director, Division of Re actor Safety l

____ ________ - __ _ __ a

_ ___ _ ____ _ _ __

.

.

LIST OF INSPECTION PROCEDURES USED IP 37001 10 CFR 50.59 Safety Evaluation Program IP 37550 Engineering IP 40500 Effectiveness of Licensee Controls in Identifying, Resolving, and Preventing Problems IP 92903 Followup - Engineering LIST OF ITEMS OPENED, CLOSED, AND DISCUSSED

Open items in this report VIO 50-352:353/98-05-01- Ineffective corrective action on Agastat relays (E3.1)

W 50-352:353/98-05-02- Seismic response of Agastat relays (E3.2)

IFl 50-352;353/98-05-03- Agastat relays in a harsh environment (E3.3)

NCV 50-352:353/98-05-04- Late reporting of RWCU isolations (E7.1) I

.URI 50-352;353/98-05-05- Potential containment bypass (E8.1)

i NCV 50-352:353/98-05-06- Commons cause failure of Barton switches (E8.4)

]

NCV 50-352;353/98-05-07- SRV setpoint drift (E8.5)

NCV 50-352:353/98-05-08- Overcurrent protection of electrical penetration (E8.6)

LERs Open (Open) Licensee Event Report 50-352; 353/97-010: Potential Containment Bypass Path Resultingin a Condition Outside the Design Basis. (URI 50-352; 353/98-05-05) (E8.1)

(Open) Licensee Event Report 50-352/98-008: Pressure Setpoint Drift of Five Main Steam System Safety Relief Valves Primarily Caused by Corrosion Induced Bonding. (NCV 50-352/98-05-07) (E8.5)

LERs Closed

(Closed) Licensee Event Report 50-352/98-002: ESF Actuation Due to Blown Fuse During

'

Jumper installation (Closed) Licensee Event Report 50-352/38-005: RWCU PCIV isolation due to Steam Leak and Load Center Breaker Trip. (VIO 50 352: 353/98-05-01) (E8.3 )

(Closed) Licensee Event Report 50-352/98-007: Three Inoperable iTT Barton Model 580A Differential P, essure Switches Result in Two or More independent Trains of a Single Safety System Being inoperable From a Common Cause. (NCV 50-352/98-05-06)

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

_ _ - _ _ -__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _ _

,

e

(Closed) Licensee Event Report 50-352; 353/98-009: Primary Containment Overcurrent Protection Circuits for Six Electrical Penetrations.(NCV 50-352:353/98-05-08 (E8.6)

LIST OF ACRONYMS CREFAS Control Room Emergency FA System EA Experience Assessment ECR Engineering Change Request ESF Engineered Safety Feature GL Generic Letter IP inspection Procedure ISEG Independent Safety Evaluation Group LGS Limerick Generating Station MSRV Main Steam Relief Valve NQA Nuclear Quality Assurance NCR Nonconformance Report NRB Nuclear Peview Board OD Operability Determination PBAPS Peach Bottom Atomic Power Station PECO Philadelphia Electric Company PEP Performance Enhancement Program PORC Plant Operations Review Committee RRCS Redundant Reactor Control System RWCU Reactor Water Clean Up SE Safety Evaluation SLC Standby Liquid Control SM Shift Manager SRO Senior Reactor Operator SWRI Southwest Research Institute TS Technical Specification UFSAR Updated Final Safety Evaluation Assessment Report o_---  ;