Text

Insp Repts 50-245/97-02,50-336/97-02 & 50-423/97-02 on 970311-0519.Violations Noted.Major Areas Inspected: Operations,Maintenance,Engineering & Plant Support
	ML20141E341
Person / Time
Site:	Millstone
Issue date:	06/24/1997
From:	; NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:	;
Shared Package
ML20141E286	List: ML20141E283; ML20141E310; ML20141E341;
References
	50-245-97-02, 50-245-97-2, 50-336-97-02, 50-336-97-2, 50-423-97-02, 50-423-97-2, NUDOCS 9707010034
	Download: ML20141E341 (102)
	v • d • e

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

'

l

.

U.S. NUCLEAR REGULATORY COMMISSION

OFFICE OF NUCLEAR REACTOR REGULATION

SPECIAL PROJECTS OFFICE

i

l

Docket Nos.: 50-245 50-336 50-423

l- Report Nos.: 97-02 97-02 97-02

License Nos.: DPR-21 DPR-65 NPF-49

!

Licensee: Northeast Nuclear Energy Company l

P. O. Box 128

Waterford, CT 06386

Facility: Millstone Nuclear Power Station, Units 1,2, and 3

.

Inspection at: Waterford, CT

Dates: March 11,1997 - May 19,1997

Inspectors: T. A. Easlick, Senior Resident inspector Unit 1

D. P. Beaulieu, Senior Resident inspector, Unit 2

A. C. Cerne, Senior Resident inspector, Unit 3

A. L. Burritt, Resident inspector, Unit 1

R. J. Arrighi, Resident inspector, Unit 3

L. L. Scholl, Reactor Engineer, SPO

N. J. Blumberg, Project Engineer, SPO

R. J. Urban, Project Engineer, SPO

D. T. Moy, Reactor Engineer, Region 1

J. E. Carrasco, Reactor Engineer, Region I

D. A. Dempsey, Reactor Engineer, Region i

Approved by: Jacque P. Durr, Chief

Inspections

Special Projects Office

Nuclear Reactor Regulation

l

{

,

9707010034 970624

PDR ADOCK 05000245

G PDR

l

._

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

.

I

TABLE OF CONTENTS

<

1

EXECUTIVE SUMMARY . . . . . . . ........................ . ...,...... iv

U101 Conduct of Operations . ...................... ... . 1

U103 Operations Procedures and Documentation ............... 11

! U106 Operations Organization and Administration .. . ......... . 12

4

U108 Miscellaneous Operations issues (92700) ....... . .. .. 13

U 1.ll Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .............. 14

U1 M1 Conduct of Maintenance ............................ 14 l

U1 M8 Miscellaneous Maintenance issues .............. ..... 16

!

U1.lli Engineering . . . . ......................... ......... ....... 18

U1 E1 Conduct of Engineering ............... . ........... 18

.

t

U1 E8 Miscellaneous Engineering issues . . . . . . . . . . . .. ... . 33

'

U2.1 Operations ................... . . ... . ........ ....... .. 36

U2 01 Conduct of Operations . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . 36

U2 08 Miscellaneous Operations issues . . . . . . . . . . . . . . . . . . . ... 39

'

U2.ll Maintenance . . ......................................... ... 40

.

U2 M1 Conduct of Maintenance .............. ............. 40

U2 M3 Maintenance Procedures and Documentation . . . . . . . . . ... 41 j

U2 M8 Miscellaneous Maintenance issues ........... ......... 43

U 2.lli Enginee ring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......... 49

U2 E2 Engineering Support of Facilities and Equipment . . . . . . . . . . . . 49

U2 E8 Miscellaneous Engineering issues . . . ........ . ....... 51

j U3.1 Operations .................... ........ ... .. . ......... 59

1

U3 01 Conduct of Operations . . . . , .......... .. .......... 59

U3 07 Quality Assurance in Operations (40500) ............ . . 63

U3 08 Miscellaneous Operations issues (92700) ............. .. 64

3 U3.ll Maintenance . . . . . . . . . ... ............. .... ...... ........ 65

U3 M1 Conduct of Maintenance .... ....... .. . .......... 65

l U3 M2 Maintenance and Material Condition of Facilities and

Equipment .. ......... ........ ............. . 66

U3M7 Quality Assurance in Maintenance Activities ..... ...... . 68

U3 M8 Miscellaneous Maintenance issues ...... ........ ..... 70

U 3.Ill Enginee ring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... 72

U3 El Conduct of Engineering ......................... .. 72

U3 E2 Engineering Support of Facilities and Equipment . . . . . . . . . .. 76

U3 E8 Miscellaneous Engineering Issues . . . . . . . .. ... .. .. 77

i

1

-

l

.

1

.

IV Plant Support ...................... ....... .. ....... ....... 84

! R1 Radiological Protection and Chemistry Controls .......... 84

V. Management Meetings . . . . . . . .................. .............. . 87

X1 Exit Meeting Summary . . . . . . ....................... 87 ;

X3 Management Meeting Summary . . ... ................ 87 !

l

1

l

i

! iii

'

i

l

!

l EXECUTIVE SUMMARY

! Millstone Nuclear Power Station

Combined Inspection 245/97-02:336/97-02;423/97-02

l

Operations

At Unit 1, operator response to the ESF actuation was excellent. Operator control !

and monitoring of plant parameters following the event and throughout the systems

restoration were appropriate. Shift supervision maintained good command and I

'

control throughout the event. The assessments and verifications performed by the

shift technical advisor were outstanding. (U1.01.2)

The reportability determination (RD) performed at Unit 1,in response to a condition

report (CR) concerning a design basis for the EDG with respect to outside air l

temperature, did not contain a sufficient technical basis for the conclusion that the

CR was not reportable. This is a particular concern since the RD had received three

levels of review. (U1.01.3) l

Overall, the implementation of procedure RP-4, " Corrective Action Program,"

revision 4 has not resulted in significant improvements in the corrective action I

process at Unit 1. The revision of the CR process was poorly implemented in that

specific guidelines were not put in place to ensure the initiation and appropriate

processing of CRs for conditions adverse to quality. However, a number of

management initiatives were implemented during this period and have resulted in

improvements, such as corrective action plan quality, in general, improvement was

noted throughout the period on many of the weaknesses discussed in this report.

However, CR trending reports were not being issued as required by procedure and is

an apparent violation of NRC requirements. The interface between RP-4 and other

lower tier deficiency reporting processes and the proper implementation of those

processes are considered unresolved pending the licensee's evaluation and

implementation of corrective actions as appropriate. (U1.01.4)

Condition reports are not being initiated consistent with the program requirement

described in procedure RP-4 at Unit 1. Based on discussions with the Unit 1 staff,

and also upon the number and significance of some of the examples either not

reported, reported after NRC prompting, or delayed without alternative reasons,

there appears to be some apprehension to initiate CRs. Although, corrective action

process implementation weaknesses have been identified with procedure controls,

training and inconsistent expectations, there also appears to be some reluctance for

personnel to document problems. Several deficiencies were also noted in the initial

operability and reportability screening for a number of the CRs reviewed. (U1.01.5)

The process for training and/or familiarization prior to the implementation of Nuclear

Group Procedure 2.25, Revision 10, Reportability Determinations, at Unit 1,is not

clearly defined, particularly in the area of familiarization training. Th!s issue is

unresolved pending completion of additional NRC and licensee revir.w. (U1.03.1)

iv

<

.

i

'

.

Unit 1 har not implemented appropriate controls for the replacement of main control

panel indicating lamps that are required to be OA category 1. The bulbs in a

number of these applications are typically replNed with non-QA and uncontrolled

bulbs. The use of non-OA bulbs in a QA aps.;ication is unresolved pending the

-licensee's review of the issue and implementation of appropriate corrective actions.

(U 1.08.1 )

The Unit 2 backlog of 828 condition reports (CRs) that are greater than 120 days

old indicates that timeliness for completing corrective actions continues to be a

concern. Although the 120-day-old CR backlog has increased from 798 since the

last inspection period, the total number (all ages) of open CRs has declined slightly

from 1335 CRs in January to 1215 CRs in April 1997, which is a positive trend.

(Section U2.01.2) '

At Unit 2, a comparison of the inspection results of 16 open items [ Licensee Event

Reports (LERs), Escalated Enforcement item (EEI), Violations, and Unresolved items 1

(URis)) reviewed in NRC Inspection Report (IR) 50-336/96-08 and 15 open items I

reviewed in this inspection report indicates that the licensee had made some

progress regarding the quality of corrective actions. In this report, the corrective

actions for 12 of 15 open items were acceptable while only 4 of 16 were

acceptable in IR 50-336/96-08. In this report, a violation was issued for 1 of 15

open items while 7 Eels and 2 violations were created associated with the 16 open

items discusced in IR 50-336/96-08. (Section U2.01.2)

The NRC followup of Unit 3 operational controls related to the Safety Grade Cold

Shutdown (SGCS) design features of the unit identified an event that was

determined to be reportable in accordance with 10 CFR 50.73. A technical

specification revision is being developed to address some of the concerns regarding

SGCS equipment controls. However, the need for comprehensive action to address

the NRC unresolved item on this topic is further highlighted by the fact that the

identified event would not have been reported without NRC questioning in this area.

(U3.01.2)

Several Unit 3 LERs discuss conditions prohibited by technical specifications (TS).

1 Individually, the issues were of low safety significance and are being treated as

Non-Cited Violations. However, the closure of the LERsdoes not address the

generic concern for TS compliance. A review of LERs issued since April 1996

revealed that there have been a number of LERs that have dealt with TS compliance

problems relating to questionable interpretations. This area is of current interest for

further NRC review and is included as an NRC followup activity; documented as

Significant items List (SIL) item 70. (U3.08.1)

Maintenance

At Unit 1, the licensee has successfully implemented the FIN team concept, a multi-

discipline, independent, and self-sufficient work team. This t6am has made a

positive contribution to the work effort, completing over 1000 AWOs since the

team was implemented ir. November 1996. A significant number of process

v

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

-

.

'

assessments, worker observations, and a self-assessment on the team have

resulted in improvements in the process. The concept of cross-training within the

disciplines on the team was considered a strength. (U1.M1.1)

At Unit 2, the operator's decision to unisolate the "B" emergency diesel generator

(EDG) starting air prior to filling and venting the lube oil system was considered to

'

be a significant weakness, particularly in light of the fact that the "B" EDG was

extensively damaged last year as a result of insufficient lubrication during routine

,

engine fast starts. The performance of the maintenance technician was excellent in 4

1

identifying this condition. The associated root cause analysis was of high quality. l

(Section U2.M1.1)

'

Over the last six months, NRC inspection reports have discussed 17 licensee event

reports (LERs) at Unit 2 involving inadequate surveillance procedures. For five of

,

'

the earlier LERs, the NRC either identified the issue or NRC intervention was

necessary to achieve satisfactory corrective action. In the response to Violation

! 336/96-08-07, which addressed inadequate containment integrity valve lineups, the

licensee committed to review all TS surveillance procedures for adequacyc Several

, more recent examples of inadequate surveillances were identified by the licensee as

i

a result of this commitment. Other examples are the result of reviews conducted as

part of their 10 CFR 50.54(f) effort and the reviews for Generic Letter 96-01,

" Testing of Safety-Related Logic Circuits." The more recent examples were

generally licensee-identified, however, these appear to be repetitive violations and

are being considered collectively as an apparent violation. The NRC Significant

i items List, item 8, lists surveillance procedure adequacy as an issue that the

licensee must satisfactorily address prior to restart. (Section U2.M3.1)

! *

At Unit 2, a previous violation regarding an inadequate containment integrity valve

lineup could not be closed because the revised valve lineup still did not provide

j' sufficient guidance to operators. The NRC found that for valves in systems that

must be in service (such as the shutdown cooling system in Mode 4,) the valve

position specified was Open/ Closed or Locked Closed /Open without any notes or

instructions to explain when or under what conditions, the open position would be

acceptable. .The licensee is revising the associated technical specification

amendment to address the concern. (Section U2.M8.3)

At Unit 2, the location specified in the shiftly surveillance for measuring the ultimate

heat sink temperature was not in strict compliance with technical specifications.

This licensee-identified concern was characterized as a non-cited violation. (Section

U2.M8.6)

A review of Unit 3 maintenance activities rever/3d that iho tagging boundary and

retest requirements for observed work activitics were adequate. (U 3.M 1.1 )

~

The Unit 3 licensee's handling of design change documents related to one as-built

"

pipe whip assembly inspected by the NRC represents a concern. Additionally, the

{ licensee's handling of ASME Code references and Code Case usage as design input

l information has resulted in some condition reports requiring followup. These issues

.

vi

. . .

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

4

.

I

information has resulted in some condition reports requiring followup. These issues

'

merit further attention to ensure that they are not ref!ective of programmatic

problems with the licensee's procurement, modification and configuration

management processes. (U3.tvl2.1 and M7.1)

Engineering

For Unit 1, the licensee identified long-standing examples of noncompliance with

the primary containment leak rate test provisions of 10 CFR 50, Appendix J.

l System configurations and test procedure deficiencies resulted in the inability to

demonstrate primary containment integrity. The magnitude and variety of problems

indicated a programmatic breakdown of the Appendix J program. Failure to

implement an effective containment leak rate test program properly is an apparent

violation of 10 CFR 50, Appendix J. (U1.E1.1)

.

For Unit 1, several examples were found of missed opportunities to havc identified

and corrected containment leak rate test deficiencies prior to the Appendix J

,

program review conducted in 1996. This is an apparent violation of the corrective

action requirements of 10 CFR 50, Appendix B, Coterion XVI. (U1.E1.1)

For Unit 1, failure to maintain drawings consistent with the plant configuration and

to apply administrative controls to valves 1-FW-107 A and B commensurate with

l those applied to similar manual containment isolation valves is a violation of NRC

design control requirements. (U1.E1.2)

In the context of low pressure coolant injection system heat exchanger capability

for Unit 1,9pparent violations of NRC requirements were identified pertaining to

performance of safety evaluations per 10 CFR 50.59 and extended operation

beyond the plant licensing basis, operability of the containment cooling system, and

corrective action for heat exchanger tube fouling. (U1.E1.3)

Failure to maintain the Unit 1 Updated Final Safety Analysis Report consistent with

the current plant licensing basis was an apparent violation of 10 CFR 50.71. '

(U1.E1.3)

The licensee planned to perform a test of the Unit 1 core spray system in the

recirculation mode, using normal surveillance procedures. Since a 10 CFR 50.59

review / screening had not been performed in preparation for this test, the

intervention of the inspector prevented a potential violation of NRC requirements.

The NRC is concerned that a vulnerability exists, which would have allowed the

performance of an unreviewed test to occur. This item remains unresolved pending

further licensee evaluation and NRC review. (U1.E1.4)

At Unit 1, the licensee's corrective actions for the replacement of CU-29, and

performance of the required localleak rate testing was found to be acceptable. The

failure of the as-found localleak rate test and evidence of the longstanding leakage

of CU-29 is contrary to Technical Specification 4.7.A.3.e.(1)(a), which requires a

combine leakage rate of less than 0.60 La (300.3 scfh) for all penetrations and

vii

r

.

I

I l

l

l

requirements. Additionally, the licensee failed to consider all recent discrepant I

conditions, related to containment integrity and evaluate the aggregate impact. The l

safety implication appears more significant than was discussed in LER 96-? 2.

l

,

'

(U 1.E8.2)

At Unit 2, bulges in several containment liner plates that have existad since original

construction were found to have been adequately evaluated by the licensee. 1

(Section U2.E2.1)

A Unit 2 unresolved item was reviewed concerning an enclosure building ventilation

damper with a single failure vulnerability that could result in exceeding 10 CFR 100

limits for offsite doses. The NRC decision to not backfit the licensee to address this

vulnerability was based heavily on the operator compensatory action described in

LER 50-336/94-40-02 involving securing the main exhaust fans in response to the

Unit 2 stack radiation monitor alarm. However, the inspector found that the ;

licensee's corrective actions were inadequate in that the alarm response procedure l

fai!ed to ensure the main exhaust fans were secured. This was characterized as a i

violation. (Section U2.E8.2)

The inspector reviewed two LERs which discussed conditions where installed

equipment or actual plant configeration differed from the Final Safety Analysis

Report descriptions. These errors did not directly impact the safe operation of Unit

3. The specified design concerns have either been corrected, or are scheduled to

be fixed prior to plant startup. The effectiveness of the design control process at

Millstone Station is under current NRC review and is included as an Independent

Corrective Action Verification Program followup activity that is documented as

Significant items List (SIL) item 79. (U3.E1.1)

Adjustments made to the inservice inspection (ISI) schedule due to the extended

shutdown were reviewed. The planned ISI schedule was prepared in accordance

with an extension allowed by section IWA-2430 of the ASME Code,Section XI. As

a sample inspection for ASME Class 1, Class 2, and Class 3 components, the

inspector reviewed the Reactor Pressure Vessel (RPV) ISI, welds in the chemical

volume and control system, and component conditions in the component cooiing

and service water systems. The inspector noted that the results of the most recent

RPV ISI will be submitted to NRR, along with an alternative for an augmented

examination of the subject welds. The Class 2 piping ISI observations made by NRR

are being addressed by the licensee in accordance with the ASME Code. With the

exception of some extemal corrosion of the Class 3 components, which was

appropriately addressed by the licensee, no deficiencies were identified. (U3.E1.2)

The inspector discussed with the licensee whether loss of an emergency diesel

generator vice a loss of a single main steam pressure relieving bypass valve was the

most limiting failure for the Unit 3 steam generator tube rupture margin to overfill

analysis. Continued NRC review of this issue is planned as an inspector followup

item. (U3 E2.1)

viii

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

- . . - . - = . . . . . . ..

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

=

t

l

!

'

Plant Support

The licensee has demonstrated a significant increase in management attention

towards the liquid radwaste systems at Unit 1. However, management oversight

for liquid radwaste systems at Unit 3 still warrants attention.

A repeat violation was issued for radiation workers failure to wear dosimetry

{ pursuant to written radiation protection program instructions. The violation is

discussed in detailin the Plant Support section of this report. (IV.R1.1)

,

I

i

,

l .

l IX

l

1

.. _- .,

.

e

Report Details

Summarv of Plant Status

Unit 1 remained in an extended outage for the duration of the inspection period. The

licensee continues to implement configuration management program activities, engineering

reviews, and docketed correspondence assessments to verify compliance with the

established design and licensing basis of the unit. The successful completion of these

activities is required by NRC order prior to restart of the unit. During this period, the

licensee announced that Unit 3 was designated as the lead unit and would be the first unit

ready for external review under the provisions of the independent Corrective Action

Verification Program (ICAVP). In addition, the Units 1 and 2 recovery officers will assist

Unit 3 in its recovery process, and will assume additional responsibilities in the area of

physical plant readiness and regulatory readiness respectively. While there will be a

reduction of restart activities at Unit 1, through the end of this year, configuration

management program activities will continue.

U1.1 Operations

U101 Conduct of Operations

01.1 General Comments (71707)

Using Inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing

plant operations. During this period the inspectors noted that an operating crew identified

and stopped planned work that was not properly reviewed for shutdown risk implications.

Specifically, while the crew was reviewing work activities for their upcoming shift, they

identified that the 12F transformer replacement plan had not fully considered the potential

single failure vulnerability created during the work activities. This demonstrates an

appropriate questioning attitude and maintaining positive control and oversight of ongoing

work activities by the operations staff.

01.2 Operator Response to an inadvertent ESF Actuat'on

a. Inspection Scone (71707)

On April 9,1997, during l&C activities, a perturbation occurred on an ESF instrument rack.

This anomaly caused an initiation of the reactor low ad low-low water level emergency

safeguard feature (ESF) logic and resulted in a loss of emergency service water, reactor

building ventilation, reactor water cleanup, along with a start of the gas turbine emergency

generator. The inspector observed the operator response to this event.

b. Observations and Findinas

The operators verified that the plant responded as expected including a subsequent

independent verification. The fuel pool and reactor building closed cooling water systems

were closely monitored and temperatures recorded for trending. Procedures were used as

appropriate. The operators at the controls effectively used the control room annunciators

to reconstruct and diagnose the event.

E_

1

.

2

The shift manager had non essential personnel leave the control room area during the

event. The senior control operator appropriately established the priorities for system ,

restorations. For example, the emergency service water was returned to service in a '

controlled manner within 16 minutes of the event initiation, allowing only a 1 degree

Fahrenheit increase in spent fuel pool temperature. The shift manager also stopped all I&C

work activities pending an investigation into the cause of the event. An off duty shift I

manager helped coordinate additional resources such as engineering support to confirm the I

cause of the event, j

l

A shift technical advisor (STA)in the cuntrol room at the time of the event assisted the

operators with the monitoring of the plant response and key parameters. The STA

anticipated what assessments were needed, and typically completed the task before his

assistance was requested. In addition, the STA backed up the initial operator assessments

such as expected plant response and a verification of isolation valve status.

l

Throughout the event the telephones at the operators desk were continuously ringing. )

Although most of the essential communications were performed via radio in some cases

l

the phone system was used to provide equipment status to the control room. Many of the I

calls received by the control room were non-essential and had the potential to distract the i

operators. The potential for operators to be distracted by non-essential calls was

discussed with operations management. The operations manager stated that the issue

was already being reviewed.

c. Conclusion

Operator response to the ESF actuation was excellent. Operator control and monitoring of

plant parameters following the event and throughout the systems restoration was

appropriate. Shift supervision maintained good command and control throughout the ;

event. The assessments and verifications performed by the STA were outstanding.

However, throughout the event the phone system was a potential distraction for the

operators.

01.3 Reoortability Determination Review

a. inspection Scope (71707)

The inspector reviewed the reportability determination (RD) performed in response to

condition report (CR) M1-97-0954, which questioned if the emergency diesel generator

(EDG) was historically operable with outside air temperature below zero degrees

Fahrenheit,

b. Observation and Findinas

The reportability section stated that "this CR has been screened for reportability against

the outside design basis criteria, and found to be not reportable." The RD referenced the

Unit 1 and Unit 3 FSAR sections which discussed the design basis outside air temparature

for the plant. Based on that discussion, the RD concluded that "the EDG design meets the

design basis outside air temperature for Millstone." This RD did not answer the obvious

___. . . _ _ _. . _ _ _. _________ ._ _ _ ___ _ _ _

i

l

1

f

3

1

question of whether or not the EDG was outside its design basis if the outside air

temperature dropped below zero degrees Fahrenheit. This issue was discussed with 1

operations and engineering management, and a revised RD was completed on May 9, !

1997. While the new RD arrived at the same conclusion, that the issue was not ,

l reportable, it did provide the requisite technical basis to support that conclusion, including )

calculations that indicated that the air intake piping contained sufficient prewarmed air to

i

start the dieselindependent of outside air temperature.

c. Conclusion . ;

The reportability determination performed to aodcass the EDG's capability to start at

temperatures less than zero degrees Fahrenheit, did not contain a sufficient technical basis ,

for the conclusion that the CR was not reportable. This is a particular concern since the

RD had received three levels of review with signoffs from a peer / supervisor, a licensing j

manager, and a unit director or designee. In this case the designee was a shift manager.

01.4 lmolementation of the Condition Report Process (SIL 17 UPDATE)

a. Insoection Scope

I

On February 25,1997, a revision to RP-4, " Corrective Action Program," was implemented. '

The purpose of the change was to simplify the corrective actions process and allow the

incorporation of improvement items in the process instead of limiting the process to

adverse conditions only. The inspectors rev!ewed various aspects of the implementation to

assess the licensee's corrective action process,

b. Observations and Findinas

Based on discussions with the licensee, the key improvements of the revis!on were to

restructure the CR levels for simplification and to provide the inclusion of enhancement

issues into the corrective action process. The revision also provided formal expectations

on reportability and corrective action timeliness along with the requirement for measures of

effectiveness for the most significant issues. In addition, the procedure requires more

involvement by the corrective action group in assessing the evaluations performed to

resolve CRs and the associated corrective actions. The licensee also stated that the

procedure change requires more management oversight of the process.

Trainina

Prior to implementation of the revision to RP-4, a formal training presentation was provided

on the changes to the directors, managers, and selected supervisors. The directors and

managers in turn provided a familiarization overview of the changes to the rest of the unit

staff, during departmental meetings and via a familiarization handout. A review of the

attendance of the formal training performed identified that half of the shift managers and

ceveral senior control operators were not trained on the RP-4 revisions. Further, none of

the maintenance or instrument and control supervisors attended the formal training, instead

they attended a familiarization course. According to the corrective action manager, most

of the changes to RP-4 did not effect the portions of the process that supervisors are

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

.

4

typically involved with, so mandatory training was not necessary. However, the inspector

determined that one of the most significant changes had to do with the CR initiation

threshold and the associated CR levels. In addition, RP-4 does not contain clear guidelines

or criteria as to when a CR should be initiated nor were the expectations clearly provided

via training based on a review of the associated records. The inspector also noted that RP-

4 made a special note of the importance of the supervisor in the corrective action process.

, Specifically, the supervisor determines the appropriate reporting method and serves as a

'

, quality check to determine if the condition meets the CR threshold. During this inspection

period a lack of familiarity with CR severity levels was also noted at the moming meeting.

Following a discussion with the inspector, the licensee agreed to revaluate the adequacy of

. the training provided following the completion of a planned self assessment. Additional

l training would be prm6ied as appropriate.

4

)

l The training provided for the revision of RP-4 was not commensurate with the limited

criteria that describe the threshold for CR initiation. In addition, the failure to provide

q

'

detailed training to all supervisors resulted in the loss of a barrier in ensuring the prompt

and appropriate initiation and processing of CRs.

Timeliness

A review was performed to determine if adverse or discrepant conditions, identified via a

I CR, were being promptly addressed for operability and reportability. RP-4 requires that the

supervisor and operational reviews occur within 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months of CR initiation. A review of the I

intervals between event identification and initial assessment of operability and reportabilit'/

determined that based on the method in which the CR dates are recorded and tracked,

there is no way to assess the processing of CR beyond reviewing each CR individually.

Based on daily observations of the CR process, numerous CRs required several days to a

. week from the initiation through the operational review by the shift manager or designee.

The licensee initiated a CR to address the processing delays and the lacia of event dates

and times. In addition, the licensee plans to revise the convention for input of dates into

the computer system such that evaluations can be performed of CR operational redew

timeliness.

.

The revision to RP-4 included guidelines for the development of the corrective action plans

within 30 days of the assignment to a responsible individual. A review of CR initisted

following the implementation of the RP-4 revision, identified that approximately 80% of the

63 CRs (past the initial 30 day point) did not have corrective action plans developed. The

licensee initiated a CR to address the high percentage of CRs not meeting the 30 day

guideline.

Delays in identifying and processing CRs has resulted in delays in assessing operability and

reportability of numerous issues; however, no consequential issues were identified. A high

percentage of CRs generated during the inspection period did not result in the developrnent

of corrective action plans consistent with the procedural expectations.

.

_

l

.

5 1

Sionificanco Lovej

J

A review rformed to determine if CRs were being assessed at the appropriate level

and there. xoiving the proper level of management attention. A sample of 19 of a total

of 261 CRs of the more significant level 2 and 3 CRs were selected for specific evaluation.

The review identified that 7 of the 19 CRs were incorrectly assessed at too low of a level.

The licensee subsequently upgraded the CR levels as appropriate. An additional 2 of the

.

19 CRs were dispositioned at a level appropriate to the reported condition; however, the l

licensee failed to perform a followup trend analysis to identify a broader issue that would

have increased its sign!ficance to the level 1 threshold. Aside from receiving additional

management attention, adverse conditions assessed to be level 1 CRs require the

implementation of measures to ensure corrective action effectiveness and are forwarded to

the Nuclear Safety Assessment Board for review.

The review of CR level dispositions also identified that a number of adverse conditions

were dispositioned at a level higher than warranted by the reported condition. For

example, security issues involving improper control of safeguards information, keys left in

unattended vehiclec ;aside the protected area, and an improperly controlled security key

card were determinad to be level 1 issues, but did not appear to meet the related I

threshold. Based on discussions with the licensee's corrective action staff, these adverse I

conditions had been assessed as leval 1 as a result of trends; however, the licensee had

not performed the trend analysis to support the assumption that a broader issue existed.

1

A review of level 1 CR issues was performed to determine if root cause evaluations were

l being performed as appropriate. A sample of 19 level 1 CRs was reviewed and found that

root cause evaluations were performed for all but 5 of these CRs. For the 5 cases in which

a root etuse evaluation was not performed, the licensee had an adequate bases for not

performing the more detailed evaluation.

A general lack of familiarity or understanding of CR severity level classification was

observed. A number of CRs sampled were inappropriately assessed at too low of a level,

resul ting in the potential for inadequate msnagement attention and oversight. In addition, l

leve! 1 issues that are incorrectly classified would not require measures to assess I

effectiveness of the corrective actions. Problems were also noted with the licensee's

staffs understanding of how the initial operability and reportability screening should be

addressed for equipment that is out of service at the time of discovery.

Administrative Controls

A review of the CR backlog was performed and identified an increasing trend with 1943

open CRs at the beginning of the inspection period and 2193 open near the end of the

period, with 558 new CRs generated. Notwithstanding the high number 9f CRs generated

during the period, CRs were not being consistently initiated in accordance with program

requirements as discussed in Section 01.5. Two initiatives were implemented to f acilitate

the reduction of the CR corrective action plan backlog. The first was "CR day," which was

a unit wide work focus for two, one-day sessions to develop corrective action plans for

open CRs. This initiative resulted in an approximately 25% reduction in the backlog of CRs

awaiting evoluation and development of corrective action plans. The second initiative was

__ _ .

l

l

, 4

.

l

\ 6

the development of a "CR team" which screens the previous days CRs, gathers additional

information when necessary, and makes an initial assessment. The CRs are then presented

to the management tearn with appropriate recommendations, in many cases, the CRs are

subsequently assigned to the CR team for development of the corrective action plan within

a couple of days. Although, these initiatives have caused a reduction in the backlog of

needed CR evaluations, and may improve the timeliness of development of corrective

action plans in the future, the CRs will remain open until all corrective actions are

implemented. An accompanying increase in the open corrective actions requiring

implementation, indicates that although the open CR backlog is increasing, a significant

percentage of those CRs have moved further in the process and are closer to

implementation. l

Although a number of initiatives such as CR day and the establishment of a CR team have i

been implemented during this inspection period, these efforts have not turned the '

increasing trend in the open CR corrective action backlog.

l

Curing a review of specific CRs, a number of administrative controls weaknesses were i

noted. Specifically, RP-4 does not require the documentation of the bases for the l

reportability assessment for section 2 or section 3 of form RP4-1, if no is checked. The l

lack of specific details led to a significant amount of research, by the licensee, in order to

establish the bases for the assessments, in addition, some of the shift managers were l

addressing the questions in sections 2 and 3 of the CR form differently, depending on l

whether or not the system or component was in an operable status at the time of

discovery. However, RP-4 does not provide guidance to support different approaches

depending on the equipment operability status. In addition, RP-4 does not describe how

and when personnel questionnaires should be used and controlled. Personnel

questionnaires are used for future evaluations and to record the conditions, circumstances

and actions taken during an event, i

Trendina

In October 1996, the sitewide corrective actions group was disbanded, as Northeast

Utilities moved to a unitized approach for implementation of most processes. Shortly after

that time, the licensee stopped performing trend analysis as required by RP-4 revision 2, in

effect at that time. The procedure required analysis of data to identify trends. Specifically,

the procedure states on a monthly basis, issue a report covering apparent adverse and

positive trends and other topics as appropriate. The QA topical report also discusses the

use of trend analysis reports as a means for meeting 10 CFR 50, Appendix B, Criterion

XVI, " Corrective Action." The last trend report performed was for the month of October

1996. CR M1-97-0258, initiated by the nuclear oversight organization, identified this issue ,

in February 1997. The CR corrective actions discussed a revisic,n of the trend requirement

to a quarterly periodicity, the development of guidelines for trending and to perform interim

trend analysis until the first quarterly report is issued. However, at the end of the

inspection period the licensee had not completed the interim trending. This is an apparent

violation of the RP-4 procedure and is viewed as another example of the previously

l

t identified programmatic breakdown of the licensee's corrective action process discussed in

NRC report 96-04. (eel 245/97-02-01)

,

.

7

The current revision of RP-4 requires quarterly trending of CRs; however, no program

guidelines or expectations have been established to address how and what areas should be

trended at this periodicity. Additionally, no guidelines or protocols have been established

for as needed trending based on repeat occurrences observed during the management

review of CRs. However, as a result of management initiatives, a number of trends have

been identified. For example, a clearance and tagging problem, personnel performance

issues related to security, radiation and worker practices, automated work order

compliance, and use of personnel protective equipment, have been identified.

Other Deficiency Processes

Procedure RP-4, " Corrective Action Program," is one of the processes listed in the topical

report to address the corrective action requirements. RP-4 lists design deficiencies among

other things that warrant a condition report (CR). As a result of the extensive design

reviews, the licensee has established another process to address potential design

deficiencies, namely the unresolved item report (UIR). The process is described in Project

Instruction (PI) 14 but is not a process described by the QA topical report as a process

used for compliance with Appendix B Criterion 16.

CR M1-97-0824, identified reactor protection instrumentation not in conformance with

design requirements. The licensee planned to close this CR with no action and reference

the associated UlR for corrective action. The licensee immediately suspended the practice

of closing CRs to UIRs in part, based on discussions with the inspector. The licensee staff

a

stated that a review of the QA Topical Report and RP-4 had been performed and found the

,

use of the UIR process as an extension to the CR progress acceptable. However, a

subsequent review determlned that the practice was unacceptable. The licensee's staff did

not elaborate on bases for the change in assessment.

The nuclear oversight organization performed surveillance MP1-P-97-018 to assess trouble

reports and corrective maintenance automated work orders for adequacy of condition

reporting on plant equipment. This review was performed as a followup to an early

surveillance, MP1-P-97-011, which identified that the Fix It Now team was not

appropriately writing CRs for adverse conditions and improvements as required by RP-4.

Surveillance MP1-P-97-018 resulted in a CR concerning the inappropriate use of the trouble

reporting (TR) system to identify adverse equipment conditions.

As a result of the NRC inspection activities concerning the use of TRs verses CRs, CR M1-

97-1119, was initiated and identified that multiple site deficiency identification programs

existed at Millstone. The implementation of the these other deficiency reporting processes

could circumvent the RP-4 corrective action process. At the end of the inspection period

the licensee was evaluating the administrative controls and implementation of these other

deficiency programs.

Self Assessment

The licensee performed a station wide self assessment that included the corrective action

program as a common area to be assessed. The preliminary findings included many of the

same issues identified in this report; however, the assessment was not finalized at the end

,

-

t

l

.

l

l 8

1

of the inspection period. In addition, a " Millstone Unit Comparative Assessment Readiness

, for Restart," performed by nuclear oversight, also identified problems with the corrective

l action process. The most significant assessment was that "the threshold for CR initiation

on Unit 1 is too high, thus preventing adequate corrective action investigation of the

issues."

,

'

c. Conclusion

Overall, the imp 6mantation of procedure RP-4, Revision 4, has resulted in limited

improvements in i5e corrective action process. The revision of the CR process was poorly

implemented in that specific guidelines were not put in place to ensure the initiation and

appropriate processing of CRs for conditions adverse to quality. However, a number of

management initiatives were implemented during this period and have resulted in

improvements such as corrective action plan quality. In general, improvement was noted

throughout the period on many of the weaknesses discussed above. The licensee plans to

revise RP-4 and develop detailed department instructions to resolve the issues discussed

above, as well as many other issues identified in their self assessment.

The interface between RP-4 and other lower tier deficiency reporting processes and the

proper implementation of those processes is unresolved (URI 245/97-02-02) pending the

licensee's further evaluation and implementation of corrective actions.

01.5 Initiation of Condition Reoorts (SIL 17 UPDATE)

a. Inspection Scoce

The inspectors reviewed the threshold at which CRs were being initiated through daily

observations. This inspection activity was performed in part, as a result of previous

interaction with the licensee's staff, which indicated that they were unclear on the CR

threshold or did not want to initiate CRs for other reasons. These issues included cable tray

deficiencies, service water system macroscopic fouling, and service water system internal

pipe coatings issues. Although CRs were ultimately initiated for all of these issues, a

significant number of discussions between the inspector and licensee management were

necessary before these issues were put into the corrective action process.

b. Observations and Findinas

During a December 1996 meeting with the licensee concerning the controls used to

maintain isolation between the high to low system pressure interface, the inspector

identified a procedure deficiency. Procedure 305A, " Operating Shutdown Cooling with

Fuel Pool Cooling," allowed the use of the shutdown cooling system to cool the spent fuel

pool. However, in one of the configurations specified, adequate separation between the

reactor coolant pressure boundary, and the spent fuel pool was not maintained during

conditions postulated to comply with 10 CFR 50 Appendix R. During a followup review,

l

the procedure was verified to have been corrected; however, no CR had been initiated for

i this deficiency. The inspector discussed the failure to initiate a CR for the deficient

! condition with the licensee's staff; however, at the end of the inspection period the

licensee had not initiated a CR.

_ _

.

9

CR M1-97-0855, initiated on April 23,1997, identified that on February 4,1997, an NRC

commitment related to safeguards information had not been properly implemented.

Although the condition was identified by a security supervisor, observed by a nuclear

oversight representative, and reported to two levels of management and an employee

concerns representative, no CR was initiated. This CR and an earlier related CR were

initiated as a result of NRC inspection activities. The previous CR, M1-97-0580, initiated

March 25,1997, discussed the safeguards issue, but did not address the delay in

identifying the problem. Further, when this CR was initiated the date of the event was

listed as March 25,1997, thus indicating the event had just occurred and there was no

delay in initiating the CR.

. CR M1-97-0694, identified that NRC inspectors observed a security officer improperly

performing personnel searches. This observation occurred during a routine security

inspection in early February 1997; however, the CR was not initiated until April 1997 by

licensing personnel, when the NRC report was issued. Although prompt corrective actions

, were implemented to resolve improper search concern, this issue was not put into the

licensee's corrective action process through the initiation of a CR. At the end of the

inspection period the licensee had,.not initiate a CR for the failure to initiate the CR

following identification of the security search issue.

CR M1-97-0842, identified that two fuel assemblies stored in the spent fuel pool did not

agree with the computerized tracking systern or the special nuclear material records. The

CR was initiated on April 22,1997; however the discrepant condition was identified

several months earlier. Although a CR was promptly written when the reactor engineering

supervisor became aware of the problem, at the end of the inspection period, the licensee

had not initiated a CR on the several month delay in identifying the discrepant condition.

CR M1-97-0689, identified a breach in the gas turbine enclosure which provides a barrier

for the CO2 fire suppression system. The inspector determined that the condition had

been previously identified, approximately 10 months earlier, but was not put into a

corrective action process.

CR M1-97-0507, identified that CRs were not being generated for adverse conditions or

improvements identified during work performed by the Fix-It-Now group. This CR was

originally assessed as a level 3 issue, indicating that the failure to initiate CRs consistent

with RP-4 requirements was an enhancement and not an adverse condition. Subsequently,

the licensee found that one of the specific deficiencies used as an example of the problem

was not a valid issue. Corrective actions were implemented to resolve the second

example; however the broader problem of not appropriately initiating CRs went

unaddressed. The licensee upgraded the CR to a level 2 issue.

CR M1-97-0975, identified that during a review of conditionally released position papers,

licensing commitments were inadequately addressed and resulted in a conflicts associated

with systems or components described by the FSAR. However, these discrepancies were

not identified via a unresolved item report (UIR) or CRs as appropriate.

CR M1-97-0688, identified unacceptable indications on a recirculation system weld. The

shift manager determined that the condition had no effect on operability and reportability.

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

.

a. .

i

10

,

The CR was subsequently presented to the management review team (MRT) and

operability and reportability were not questioned. The engineering representative accepted

a

the assignment to resolve the CR as a level 2 issue. The inspector discussed the issue

with the engineering representative at the MRT and other engineering personnel, and

,

'

determined that the issue may be reportable depending on the size and depth of the

cracks, which had not yet been determined. The inspector discussed this assessment with

i the licensee's staff and the concern that implementation of the CR process in this case

'

missed a potential operability /reportability issue. At a subsequent MRT meeting, an

engineering representative discussed the issue and the two barriers which failed to catch

the potentially reportable issue. However, following a discussion of whether a second CR

i

was appropriate to address the missed operability /reportability assessment, the licensee

5

elected to address the process issue via the existing CR. A subsequent review determined

that the missed operability /reportability assessment was not identified in the existing CR,

nor was a new CR generated. The licensee initiated a CR to address the missed

reportability evaluation generically; however, the CR initiated did not discuss the specific

'

!

events referenced above, instead it discussed similar issues and referenced several other i

i CRs. )

i 1

l CR M1-97-0621 identified the lack of technical specification requirements for gas turbine

i battery testing. The issue,was identified on February 1,1997; however a CR was not

)

initiated until March 27,1997. The licensee attributed the delay in initiation to a personal :

error. In addition, a number of other CRs delayed by three weeks or more were attributed

i to mis-communications and the thought that CRs were already initiated to address the

issues that were identified.

i

j C5 M1-97-0762 identified disconnected tie rods in a safety related heat exchanger. The

.

CR was determined to be not reportable based on additional tie rods and that the heat

! exchanger tubes provided adequate support to the heat exchanger internal components.

l The shift manager that reviewed the CR accepted the position without challenge.

j However, based on discussion with the licensee staff this conclusion was not supported by

j an engineering evaluation. The engineering staff stated that they planned to analyze the

condition and determine if it effected operability or was reportable, but during a

<

subsequent inspection they determined that the tie rods were in place and there was no

j deficient condition. However, the RP-4 controls to assure timely operability /reportability

evaluations were not implemented and licensee management was unaware that the

condition described by the CR may represent a significant historical deficiency.

I Subsequent to a discussion with the inspector, the licensee initiated a second CR to

'

address the missed operability reportability evaluation.

CR M1-97-0805 identified a potentially degraded backup air source to the diesel generator

starting air system. The issue was determined to be not reportable since the validity of the

problem was unknown. The CR stated " if the condition is determined to actually exist, a

CR should be written to document the condition and a reportability determination initiated

at that point." CR M197-0497 identified seven Millstone Unit 1 safety related actuators

that were susceptible to failure as discussed in an NRC information notice. The issue was

determined to have no actual or potential effect on operability or reportability. The CR

stated "if any diaphragms are found with the wrong sizing during the investigation,

separate CRs will be written to address specific operability /reportability reviews." Based

-

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

.

11

on discussions with the corrective action manager, these conditions should have been

promptly evaluated under the original CR consistent with the expectation of a 24-hour

investigation time interval. RP-4 does not allow the latitude to delay addressing

operability /reportability issues.

In response to the inspectors concerns that the licensee's staff was not consistently

initiating CRs for significant conditions adverse to quality, Nuclear Oversight surveillance

was initiated at the request of Unit 1 management. The objective of the surveillance was

to assess the corrective action initiation process through questionnaires and interviews.

The surveillance identified that the majority of personnel were adequately identifying l

problems and writing CRs. However, the surveillance also found that some personnel are

reluctant to write CRs for various reasons such as, the types of items that are reportable

are unclear, procedure use identification was poor, and training received to execute the )

program was not standard or clear among groups. A CR was subsequently initiated to !

address these issues. i

1

c. Conclusion

i

Condition reports are not being initiated consistent with the program requirement described

in procedure RP-4. Based on discussions with the licensees staff, the number and j

significance of some of the examples either not reported, reported after NRC prompting or l

delayed without alternative reasons, there appears to be some reluctance to initiate CRs.

Several deficiencies were also noted in the initial operability and reportability screening for

a number of the CRs reviewed. Specifically, three examples were identified in which the

issue identified was not promptly evaluated consistent with the expectations applicable for i

potential operability /reportability issues. In two of the cases, the licensee's staff '

documented that if a deficiency was identified during the CR closeout investigations, that

resulted in inoperable equipment, another CR was to be written. This approach

circumvents the RP-4 and other controls that ensure prompt evaluations of deficiencies

that could adversely affect operability or may require prompt reporting to the NRC.

U103 Operations Procedures and Documentation

03.1 NGP 2.25, Reportability Determination and LER Processino

a. Inspection Scone (71707)

The inspector reviewed the training and implementation process for Nuclear Group

Procedure, (NGP) 2.25, "10 CFR 50.72,10 CFR 50.73, and 10 CFR 50.9(b) Reportability

Determination and Licensee Event Report Processing," revision 10. The procedure revision

incorporated and expanded guidance for reportability determinations by the inclusion of a

revised version of "NU Reporting Guidance" (Redbook) as an attachment.

b. Observations and Findinas

NGP 2.25, revision 10, became effective on April 23,1997. During the control room

walkthrough that morning, the inspector discussed the procedure change with the shift ,

managers to ensure that he was aware of the new procedure. This procedure had a direct l

_ _

- _

_

, .

f

'

12

impact on the operation staff since they perform the initial review for reportability for any

condition report that is processed through the control room. The shift manager informed

'

the inspector that he was aware of the procedure change but as of yet had not received

any training on it. Discussion with the operations manager indicated that the shift

managers were going to receive familiarization training that evening, at the weekly shift

manager meeting. A review of the " Documentation of Training Requirements," section 3,

.

for this change, indicated that only familiarization was required and it was not needed prior

to the effective date of the new revision. The methods to provide familiarization included:

! department meetings, pre-shift briefing, pre-work briefing, or document acknowledgement

sheets.

During discussions concerning the appropriateness of only familiarization training for this

procedure revision, the inspector was informed that previous Revision 9 of this procedure

had been initially approved by SORC with an effective date of March 7,1997, but

subsequently postponed implementation due to formal training not being completed. This

,

revision had already been distributed, when on March 6, the plant staff discovered that the

training had not been completed. Revision 9 of NGP 2.25 was never formally issued.

Revision 10 was approved by SORC on April 16, following station reviews, a detailed

independent review, and a quality assurance review. At that time, the training required for

implementation was reduced from formal training to familiarization training "with no impact

on effective date." The inspector discussed this issue with the licensee personnel

responsible for procedure implementation and could not determine why Revision 9 required

formal training and why that training was not completed. Additionally, there was no

record to verify that familiarization training for Revision 10 was completed as required.

~

c. Conclusion

The inspector concluded that the process for training and/or familiarization prior to the

implementation of a procedure revision is not clearly defined, in addition, in the area of

familiarization training, there is no record to ensure completio'n of this type of training.

This issue will become more significant as the unit completes its design basis

reconstruction and a large number of procedures will require revisions. This issue will

remain unresolved pending completion of additional NRC and licensee review (URI 245/97-

02-03).

U106 Operations Organization and Administration

06.1 Overtime Controls for Operations Personnel

The inspector viewed the overtime controls and records for the operations department

personnel during the first quarter of 1997. NGP 1.09, " Overtime Controls for Nuclear

,

Group Pe. sonne ," states that affected plant staff shall verify proper authorization has been

obtained prior to exceeding any overtime limits. "Affected plant staff" is defined as plant

staff who perform safety related functions such as operators, health physics personnel,

chemistry personnel, key maintenance personnel, and the first line supervisors of these

personnel. The initial review of the time sheets from that period indicated approximately 6

individuals had exceeded the overtime limits without proper authorization. The Operations

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

,

<

.

13

i

,

Manager provided this additional information to the inspector, which indicated that those

individuals had not performed safety related work, and therefore, were not subject to NGP

i 1.09 limits.

l

j The inspector noted that an operator, working in the operations work control center, had l

exceeded the overtime limits, including working an 18 hour2.083333e-4 days 0.005 hours 2.97619e-5 weeks 6.849e-6 months day. The inspector was

-

1

!' informed that there was some discussion about whether or not NGP 1.09 applied to this j

1

individual since he was not working in the control room at the time. While he was not i

actually manipulating equipment that provided a safety-related function, he was involved

i with tagging of a safety-related system. The licensee determined he met the definition of

j an "affected plant staff" and completed an authorization to exceed overtime limits prior to

exceeding the established limits. The authorization stated that "no manipulation of

i equipment that provides a safety function is authorized," and was signed by the

l Department Manager and the Director of Unit Operations. The inspector concluded that

j operations personnel were in compliance with NGP 1.09 for the first quarter of 1997.

U108 Miscellaneous Operations issues (92700)

j 08.1 Use of Non OA liaht Bulbs in a OA ADolication

a

j a. Insoection Scope (92903)

l A review of the controls for replacement of main control board indicating lights for safety

,

related equipment was performed.

4

j b. Observations and Findinas

l Adverse condition report (ACR) 01447 was initiated to address the replacement of an

j anticipated transient without scram (ATWS) panel light bulb with a spare from the same

i panel. The ACR questioned the practice of using spare installed bulbs as replacements for

active indicators. The corrective actions to resolve the ACR discussed the need to use the

i

automated work order (AWO) process to replace indicator lamps classified as QA Category

1 equipment.

Main control panel indicating lights classified as OA Category 1 equipment, are replaced by

both operations and l&C personnel. Based on discussions with the operations staff, the

AWO process is not used by operators to replace these indicating lights. Further, there is

no alternative control to ensure that bulbs with the proper quality attributes are used.

Typically spare non OA bulbs maintained in the control room are used. Based on

discussions with the l&C staff, the AWO process is used by the l&C group when replacing

OA Category 1 indicating lights. However, when the inspector requested some examples,

the licensee only found two cases since 1984,in which an AWO was used to replace

control room indicating lamps.

A subsequent review found that numerous control board indicating lights are classified as

OA Category 1 equipment. For example, core spray and isolation condenser valve

indications, source range monitor trip lights, and numerous process radiation monitoring

status lights. Previous evaluations had been performed and determined that numerous

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

*

.

l

4

14

indicating lights in the main control room panels should be QA category 1, typically based

on the need for that indication for post accident monitoring. The licensee initiated

j condition report (CR) M1-97-0423 to address the issue. The licensee also plans to re-

i evaluate each of the OA bulb applications and believes they can be downgraded to a non

OA status,

c. Conclusions

The licensee has not consistently implemented the AWO process or other appropriate

controls for the replacement of main control panelindicating lamps that are required to be

OA category 1. The bulbs in a number of these applications are typically replaced with

non OA and uncontrolled bulbs. The use of non OA bulbs in a QA application is

unresolved (URI 245/97-02-04) pending the licensee's review of the issue and

implementation of appropriate corrective actions.

U1.ll Maintenance

U1 M1 Conduct of Maintenance

M 1.1 Unit 1 FIN Team Process (SIL 30 UPDATE)

a. Inspection Scoce (62707)

In November 1996, the Unit 1 maintenance department developed and implemented a

multi-discipline, independent, and self-sufficient work team called the Fix-It-Now (FIN)

Team. The purpose of the team was to perform a variety of maintenance work including

troubleshooting and repair work. This process was intended to supplement the normal

work control program. The inspector reviewed the FIN procedure, interviewed team

members, reviewed audit reports, training records, and observed the team's interaction

with other station organizations.

b. Observations and Findinas

The FIN Team consisted of a mechanic, an electrician, a maintenance planner, a health

physics technician, an engineer, two instrument and control (l&C) technicians, and two

operators. In addition, the Fin Team Supervisor (FTS) is a senior licensed individual who

maintains responsibility for implementation of the FIN procedure, U1 WC1 A, Unit 1 Fin

Process. Each FIN team member is a team qualified expert (TOE) and assumes

responsibility for all job functions under their discipline or qualification. Additionally, each

of the team members completed a basic maintenance fundamentals course and specific

training pertaining to the FIN process. Cross-training within the disciplines is strongly

encouraged. Individual training / qualifications were maintained current with the use of a

training matrix for each discipline, which is updated and maintained by the FTS.

As stated in the FIN process procedure, the scope of the plant work performed by FIN

includes all plant equipment provided that the completed work does not affect the design

function of components or structures. Troubleshooting may be performed utilizing the FIN

process as determined by the shift manager and the FTS, provided that the guidance in

- _ .

.

15

"U1 WC 1, Work Control Process," is followed. Team members are not permitted to

perform welding or work that is determined to be ASME,Section XI R&R (Repair and

Replacement). Since the team is independent and self sufficient, team members generate

their own automated work orders (AWO) and work under a FIN team tagging system.

The inspector attended the FIN team morning meeting. Each morning the FIN team

members, led by the FTS, review the trouble reports (TRs) that were generated in the

previous 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months and determine which items can be performed by the team. TRs on

related equipment were grouped together in order to improve efficiency. The FTS attended

the unit work control morning meeting and informed the group which TRs would be

worked that day by the FIN team.

Since the implementation of the FIN team, a number of assessments have been performed

on the process. FIN team process assessment are conducted by the team members each

Friday and provide a critical, detailed look at the previous weeks activities. Worker

Observations Checklists are completed by the FTS, at a minimum of ten per month. In l

March 1997, the nuclear oversight group performed a surveillance to assess the FIN work

control process. This surveillance resulted in procedural changes that included: 1) a {

clarification of the work scope to include repairs, as well as replacement; 2) a change to i

the FIN tagging process to allow use of all tags instead of only using blue tags; and 3) the l

removal of the 24-hour requirement for FIN work. In April 1997, a self-assessment report

was completed as part of the Unit 1 self-assessment program. One enhancement to the

team, which was identified in the assessments, was the need for a dedicated procurement

individual to handle parts acquisition for the team. This function was bein0 performed by

the team planner and was inhibiting his overall effectiveness. At the end of the inspection

period, a procurement person was added to the team.

c. Conclusions

The licensee has successfully implemented the FIN team concept at Unit 1 by establishing

a multi-discipline, independent, and self-sufficient work team. This team has made a

positive contribution to the work effort, completing over 1000 AWOs since the team was

implemented in November 1996. A significant number of team process assessments,

worker observations, and a self-assessment on the team have resulted in improvements in

the process. The concept of cross-training within the disciplines on the team was

considered a strength.

_

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

!

l

i-

i

.- l

i 16

i1 U1 M8 Miscellaneous Maintenance lasues

!.

E

M8.1 (Uodate) Unresolved 50-245/96-06-02: Soent fuel Pool Tri-Nuclear Filter Removal

(SIL 37 UPDATE)

a, 'Insoection Scoce (92902)

On July 16,'1996, maintenance and health physics personnel were attempting to remove a

portable "Tri-nuc" filter assembly from the spent fuel pool floor when a 1/8 inch wire rope,

attached to the filter assembly, caught on the bottom of three used control rod blades that-

were stored along the east wall of the spent fuel pool. This caused the control rods to

move, resulting in five control rods shifting position, moving away from the wall clustering

together, and coming to rest against an adjacent spent fuel storage rack. On October 1,

-1996, the inspector observed the pre-job briefing for the Tri-Nuclear filter removal. As

documented in NRC Inspection Report 50-245/96 08, Section U1.M1.1, the inspector

concluded that the maintenance activities associated with the restoration of control rod

blades and the removal of a Tri-Nuclear filter assembly from the spent fuel pool were well e

'

planned and carefully executed. Team work and procedure' adherence was emphasized by

a strong management presence throughout the work. During this inspection period the

inspector reviewed two issues concerning management's intervention in the adverse-

condition report (ACR) and the event review team (ERT) processes, following the filter

removal event.

b. Observation and Findinos

Following the July 16,1996 event, senior management identified a potential weakness

concerning a lack of radiological data and information regarding to the contaminated

individuals during that event. Six of the eight individuals involved in the evolution were

contaminated as a result of the filter removal event. They were successfully

decontaminated and whole body counts indicated no internal dose was received. At that

time, the Unit Director asked a health physics (HP) supervisor to document this issue in an

adverse condition report (ACR). The ACR was written on July 23, by the HP supervisor,

and submitted to the control room for processing. The following day, the HP supervisor

was off and when h'e returned to work, he was informed that the Unit Director had

rewritten the ACR to better document his concern. Interviews conducted by the inspector

with the HP supervisor had indicated that he was surprised that no one had informed him

that his ACR was being changed until he returned to work the following day.

The inspector reviewed both versions of the ACR and determined that there was no safety

significant changes between the original and the final ACR. ' The original stated, in part,

that " Senior management was unaware that the measured contamination on the 6

individuals was by industry standards considered insignificant, and was not required to be

reported." The final version stated "A mechanism to communicate radiological statistical

data and Key Performance indicators, on a regular basis from the Unit 1 HP Department to

Unit 1 management, does not exist. As a result, Unit 1 management may not have all the

necessary radiological data available at any given time. This creates the potential for

actions to be taken without all the relevant facts being known." The inspector reviewed

the revision of RP4, Adverse Condition Resolution Program, in effect at the time of the

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

.

17

event. While the procedure is silent as to when an ACR can be changed, it does state that

the supervisor shall review the ACR for completeness, "Is the problem statement clearly

written to provide management and future reviewers with sufficient information to fully

understand the nature of the issue?"

In this case, the inspector concluded that there was clearly an opportunity for better

communications between the Unit Director and the HP supervisor. The HP supervisor was

not aware that the ACR was changed, and may not have wanted to remain as the initiator

of record on the revised ACR. The supervisor was given the opportunity to review and

sign the revised ACR when he returned to work the following day, which he did.

The inspector reviewed a second issue dealing with control rod entanglement, which

concerned the ERT process for documenting results, as well as the appearance of

n anagement's influence over those results. The ERT was formed to perform a root cause

investigation, the scope of which was to determine the root cause and causal factors for

the Tri Nuclea event, and to recommend corrective action that would prevent recurrence.

DraR reports and a final report were issued between July 26 and August 27,1996. Each .

time a draft report was issued it became longer and more detailed following comments '

from plant rr anagement. The inspectors were concerned about the team remaining

independent of line management during their investigation. Interviews were conducted

with etch of the ERT members to determine what influence, if any, management had over j

the ERT results. The inspectors were informed, by the team leader, that the draft reports l

were more like a briefing paper meant to update management on the progress of the team. !

He statad that they were never meant to be a final report, but that he felt compelled to

give them "something" each week. The consensus of the group interviewed was that

management was looking for additional infoimation after each draft was completed. The

feeling was that as more facts about the event were identified, more questions were being

asked. They stated that the scope of their review kept increasing. No one felt that

management was trying to adversely influence the outcome of the review. However, the

team stated that they felt rushed to get done as early as possible, particularly as the scope

of the review increased and more time was needed to complete the work.

The inspectors concluded that the ERT maintained their independence from line

management, and the results stated in the final report were the conclusions of the team

and not the result of management's preconceptions or influence. The inspectors did note,

however, that the practice of issuing draft reports, or in this case briefing papers, lead to

the perception of management influence over the teams findings and conclusions.

Additionally, the insnector concluded that the team members lacke,d formal root cause

methodology training. Management should also be sensitive to any perceived urgency on

the part of ERT members to get resuits quickly.

- .-- - - - . -. .. .. . . . - .- . . - ~.

.

I 18

l

U1.Ill Enaineerina

U1 E1 Conduct of Engineering

E1.1' Leak Rate Testina of the Primary Containment

a. Insoection Scoce (92903)

Licensee event reports (LER) 50-245/96 26 and 96-46 contained discussions of various

system configurations that did not allow piping to be drained adequately for local leak rate

testing (LLRT) per 10 CFR 50, Appendix J. Additional discrepancies involving

pressurization of valves in a nonconservative direction, and failure to test certain

containment penetrations also were discussed. The inspector reviewed the following

documents with regaid to the conduct of containment leakage rate testing at Millstone 1:

Millstone Unit 1 Technical Specifications

LLRT procedures

Piping and instrumentation diagrams (P&lDs) and system drawings

Letters and other documents pertaining to the Systematic Evaluation Program (SEP)

and 10 CFR 50, Appendix J program commitments

Licensee Independent Root Cause Evaluation of Millstone 1 Feedwater System

Configuration

Quality Assurance Audit of 10 CFR 50, Appendix J program

Operations Department memoranda and Engineering Work Requests

10 CFR 50, Appendix J Program Review - Testing Program Requirements, General

Physics Corporation, dated October 28,1996

b. Observations and Findinas

Feedwater Check Valves 1-FW-9A(B) and 1-FW-10A(B)

In LER 96-26, the licensee documented that the feedwater system configuration did not

allow water to be drained completely to expose the isolation valve seating surfaces to the

LLRT medium (air). This condition is contrary to Section ill.C.2(a) of 10 CFR 50, Appendix

J, which requires that valves, unless pressurized with fluid from a seal system, shall be

pressurized with air or nitrogen. Failure to perform a valid LLRT results in inability to

demonstrate primary containment integrity per Technical Specification 4.7.A.3.

The containment isolation provisions of feedwater penetrations X-9A and X-98 consist of

two check valves in series; valves 1-FW-9A(B), respectively, in the reactor building steam

tunnel outside of the primary containment, and valves 10FW-10A(B), respectively, inside

the drywell. The valves are 18-inch,1500 psi, Anchor-Darling, carbon steel, check valves

with resilient seats. This configuration does not meet the containment isolation provisions

of 10 CFR 50, Appendix A, General Design Criteria 55 and 56, which require at least one

check valve inside the containment and a remote manual isolation valve outside of the

containment.

l

,

-- _

-.

_ _ _

l

I

.

l

r

19

l

(n 1983, the NRC evaluated the configuration under the Systematic Evaluation Program,

and accepted the use of two check valves in series,in part, because of the LLRT

l provisions of Appendix J. (Refer to Section 4.20.6 of NUREG 0824, Integrated Plant

I

Safety Evaluation - Systematic Evaluation Program for Millstone Unit 1.)

The inspector walked down the feedwater system piping and reviewed surveillance

procedure SP 623.14 (in effect prior to 1981) and current procedure SP 623.141,

! Containment Isolation Valve Leak Rate Test, to evaluate the method used by the licensee

to drain the piping prior to testing the feedwater check valves. Downstream of inboard

valves 1-FW-10A(B), the piping was drained through a one-inch bypass line around manual

isolation valves 1-FW-11 A(B). The drain path is approximately three and one-half feet

downstream and two feet above the check valves. The inspector estimated that the

inboard check valves thus would be sealed by about 50 gallons of water during the LLRT.

The piping between the inboard and outboard check valves is drained through a one-inch

test connection in the steam tunnel. The test connection is located on the mid-plane of the

feedwater piping. As a result, approximately one-half, or nine inches, of the outboard

valve seating surfaces remained sealed with water during the LLRTs. The inspector

determined that the draining method specified by the procedures had not changed since

LLRT started at Millstone 1 in 1976.

The licensee's independent root cause evaluation team reviewed plant drawings and a

modification package under which the feedwater check valves and certain piping sections

were replaced in 1981. The licensee concluded that the system had been constructed

originally with a one-inch bottom drain line about six inches downstream of inboard check

valves 1-FW-10A(B). The inspector confirmed that several drawings issued prior to the

1981 modification showed the drain lines. (See Section E1.2 of this report) However, the

weld sketches, drawings, and other documents in the plant design change record (PDCR 1-

75-80) do not reflect the existence of the drains, and they were not identified by the

licensee during walkdowns of the modification at the time. Finally, the pre-modification

version of test SP-623.14 did not reflect the existence of the drain lines. The inspector

noted that had the bottom drains existed, they would have been more readily accessiNe

and technically correct to use during the test. The inspector concluded that, while the

bottom drains may have been inadvertently removed during the 1981 modification, it was

equally likely that they had never been installed in the piping.

The inspector reviewed test data for the feedwater penetrations for the period 1978 to

1994. Valve 1-FW-9A failed "as-found" LLRTs in 1991 and 1994. Following repairs, for

which the system piping and valves would have been drained, the valve was successfully

retested. Thus two valid "as-left" LLRTs were performed on this valve. However despite

these valid tests, the licer;see could not demonstrate that the combined LLRT limit of 0.6

La specified by Section ll'.C.3 of Appendix J and TS 4.7.A.3.e(1)(a) was rnet due to other

LLRT discrepancies. In addition, because penetrations X-9A and X-9B were not drained

and venbd during the periodic crintainment integrated leak rate tests (ILRTs), the ILRT

i results cannot be med qualitatively to demonstrate the leak tight int'e grity of the feedwater

l penetrations.

l

The inspector ncted prior opportunities for the licensee to have identified the inability to

drain the feedwater lines. A Quality Assurance (OA) Department audit of the Millstone 1

l

- . _ __

'

.

1 s .

20

l Appendix J program was documented in report OSD-91-5448, dated December 30,1991. l

The audit was performed to verify the effectiveness and implementation of the Appendix J l

program by (a) verifying proper establishment of containment boundaries, and (b) verifying l

that ILRT and LLRT procedures implement Appendix J requirements. The inspector found i

that the auc'it scope was limited in that it did not explore whether the test prerequisites,

such as system draining, were achieved or whether the test methods met other Appendix J

requirements. As evidenced by the number and variety of program deficiencies identified

by General Physics Corporation's 1996 review (discussed below and in Sections E1.3), the

inspector concluded that the OA audit was ineffective in this respect, in assuring the i

program's compliance with the provisions of Appendix J.

l

In an August 8,1991 memorandum to engineering, an operator stated a need for high '

point vents to be installed on both feedwater lines in the reactor building steam tunnel to

facilitate draining of the piping for LLRT. The memorandum indicated that the piping was

walked down and that the issue was discussed with the system engineer. No action was

taken at that time. Operations again raised the issue with engineering in memorandum

MP1-OPS-93-8, Unresolved LLRT Design Modifications, dated January 12,1993. In that

memorandum, engineering was requested to evaluate several deficiencies and, as

appropriate, to initiate design changes. The items were identified in the memorandum

either as personnel safety issues or as potentially affecting the results of LLRT. The i

inspector discussed the memoranda with the operations and engineering managers

identified in the memoranda. The managers recalled the issues generally as involving

enhancements, with no impact on the validity of the LLRTs. The inspector was unable to

find any documentation that the licensee recognized the inability to drain the feedwater

piping or if so, considered the condition to be acceptable through a misinterpretation of

Appendix J requirements.

The inspector concluded that due to the inability to drain the piping adjacent to the

feedwater containment isolation check valves, the LLRTs conducted since 1976 (with two

exceptions) were performed with the valve seating surfaces sealed, or partially sealed, with

water. This is contrary to 10 CFR 50, Appendix J, Section Ill.C.2(a), which requires that

the valves be pressurized with air or nitrogen.

Main Steam Drain Valve 1-MS-5 and 1-MS-6

In LER 96-26 the licensee documented the inability to drain the piping between main

steamline drain valves 1-MS-5 and 1-MS-6. Similar to the feedwater check valves, inability

to drain the piping would invalidate the LLRT results. The licensee identified that the

normal drain path specified by the procedure was to the main condenser. However, this

path was inadequate because the piping is at a higher elevation than the valves. An

alternate drain path via a main steam line drain level switch was used by the operators

during LLRT.

In memoranda to engineering dated May 16,1990, and August 8,1991, an operator

questioned the ability to drain the piping through the level switch. The memoranda

,

indicated that the piping was walked down with the system engineer, but that no further

l action took place. The issue was re-identified in memorandum MP1-OPS-93-8 (dated

l January 12,1993) to engineering as a deficiency potentially affecting the results of LLRT.

! Engineering work request (EWR) 1-93-A104, dated September 13,1993, stated, "Need a

low pt [ point] drain on line 4"-MD-51 - this drain is required to ensure valid LLRT results of

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

.

21

> valves 1 MS-5 and 1-MS-6 and to be abie to maintain secondary containment during

testing." A due date of December 15,1993 was assigned to the EWR, The documented

EWR resolution addressed the secondary containment implications of the testing and stated

that "...the present method requires careful coordination to take advantage of proper plant

conditions when they occur." No reference was made regarding the ability to drain the

'

piping between the isolation valves. The inspector discussed the memoranda and EWR

.with the Operations and Engineering managers, neither of whom recalled any concern

regarding the validity of the LLRT.

!

During a system walkdown, the inspector was unable to discern a slope that would allow

the piping between the valves to drain through the level switch piping. However, on the

basis of a seven-inch change in the pipe elevation shown on main steam system isometric

,

drawing 25202-20297, the licensee concluded that adequate slope existed to drain the

!

line. The inspector agreed with the licensee's conclusion, but considered that it would be

l prudent to verify that the slope actually existed as depicted in the drawing, and to confirm -

through discussion with the operators that no anomalies indicating water in the main steam

l drain line have occurred during the conduct of past LLRTs.

l

In LER 96-46 the licensee reported that valve 1-MS-5 was leakage rate tested in a direction

opposite to an accident condition without verifying that the reverse direction test was

valid. The condition was identified during a review of the Appendix J program conducted

for the licensee by General Physics Corporation. Section Ill.C.1 of Appendix J requires

,

pressure to be applied in the same direction as that when the valve would be required to

perform its safety function. Test pressure may be applied in a different (reverse) direction

'

if it can be determined that results provide equivalent or more conservative results.

The licensee informed the NRC in a letter dated November 14,1975, that valve 1-MS-5

(then a two-inch gate valve) was tested in the reverse direction, and that the valve

"...should seat the same regardless of direction of pressure application." In a letter dated

March 3,1977, the NRC requested additional written justification for reverse direction

testing at Millstone 1. The licensee responded on September 20,1978 that, for reasons

, unrelated to Appendix J testing, the valve would be replaced with a larger valve that would

l meet NRC criteria for reverse direction testing, viz. (a) that test pressure tends to unseat

! the valve making the results more conservative, or (b) that the seating force on the valve

disk is some factor greater than the force on the disk due to accident pressure. Installation

of the current valve was documented in a letter dated November 6,1980. In the letter,

the licensee stated that valve 1-MS-5 satisfied the criteria. NRC review of the licensee's

justification was documented in Franklin Research Center Technical Evaluation Report TER-

C5257-29, dated May 26,1982. The report was appended to the NRC's Safety

Evaluation Report (SER) on the licensee's Appendix J program, dated May 10,1985. The

l TER stated that the licensee should replace valve 1-MS-5 and take other actions as

necessary to ensure that the NRC criteria for reverse direction testing are met. The 1985

l SER concluded that the licensee's proposed actions with regard to reversing certain valves

j in order to consorvatively perform reverse direction testing met the requirements of

'

Appendix J anr. were acceptable. The inspector was unable to determine whether the

( rep!acement 0; valve 1-MS-5 discussed in the TER referred to the original two-inch valve or

to the current valve that was installed in 1980.

l

t

I

l

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

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

,

.

22

NRC Information Notice 94-30, Supplement 1, " Leaking Shutdown Cooling Isolation Valves

at Cooper Nuclear Station," dated August 19,1994, discussed that licensee's

determination that reverse direction testing of flexible wedge gate valves did not always

l meet the Appendix J requirement in that tests conducted in both directions showed that a

l reverse direction test did not consistently produce equivalent or more conservative results.

The licensee at Millstone 1 did not verify through testing that the results of reverse

direction test of valve 1-MS-5 satisfied the Appendix J criteria. The inspector concluded

l that reverse direction tests of valve 1-MS-5 were contrary to 10 CFR 50, Appendix J,

l Section Ill.C.1. This is the first example of failure to satisfy the corrective action

requirements of 10 CFR 50, Appendix B, Criterion XVI. (eel 245/97-02-05) In LER 96-

l

46, the licensee committed to implement a modification and to test the valve in the reverse

direction prior to startup of Millstone 1.

i

'

Other Local Leak Rate Test Deficiencies

l In May 1996 the licensee retained General Physics Corporation to develop an Appendix J

'

program basis document for Millstone 1. '3eneral Physics conducted a comprehensive

review of the program including test methods, procedures, technical specification

! requirements, and regulatory commitments. The results of the review were documented in

a report dated October 28,1996. Subsequently, the licensee reported the deficiencies and

proposed corrective actions to the NRC in revisions to LERs 96-26 (Supplement 1) and 96-

46 (Supplement 3). The inspector reviewed the General Physics report and noted the

following additional examples of noncompliance with Appendix J requirements.

l (1) Primary containment penetrations could not be drained, or adequate assurance of

full draining could not be established.

Penetrations X-30f and X-34f Recirculation pump seal flush valves 1-RR-

25A(B)

Penetration X-211 A Post-accident sample valves 1-PAS-241-PAS-25

Penetration X-14 Reactor water cleanup inlet isolation valve 1-CU-

2A

Penetrations X-16A and X-16B Core spray injection valves 1-CS-5A(B)

Penetration X-42 Standby liquid control valve 1-SL-7

Penetration X-47 Recirculation loop sample valve 1-RR-37

This is contrary to 10 CFR 50, Appendix J, Section Ill.C.2(a), which requires valves to be

tested with air or nitrogen.

l (2) An NRC-approved exemption allowing reverse direction testing of atmosphere

l control system valves 1-AC-9 and 1-AC-12 was based on the valves being butterfly

l valves. However, they are plug-type valves for which reverse direction testing is

! nonconservative. This is contrary to 10 CFR 50, Appendix J, Section Ill.C.1, which

!

.

I

l

23 l

l

l requires test pressure to be applied in the same direction as that when the valves

l

would be required to perform their safety functions. l

l

(3) Type B tests by local pneumatic pressurization at a pressure not less than Pa was '

not performed and the combined leakage rate of all penetrations and valves subject

to Type B and Type C tests was not confirmed to be less than 0.6 La as listed

l below.

Penetrations X-25 and X-202D Integrated leak rate test pressure and flow .

connections l

Penetrations X-202A through H Testable packing glands on both sides of the

operating arm and stuffing box of atmosphere

control system valves 1-AC-1 A through J

Penetrations X-25 and X-202D Flange gaskets on inboard side of atmosphere

control valves 1-AC-9 and 1-AC-12

Penetration X-17 Pipe flange between reactor vessel head spray

valves 1-HS-4 and 1-HS-5

Penetrations X-7A through D Combined main steam isolation valve leakage l

rate determined at 25 psig was not corrected to l

Pa when added to the totalleakage limit of 0.6

La l

This is contrary to 10 CFR 50, Appendix J, Sections Ill.B.2 and Ill.B.3(a).

(4) The acceptance criterion for containment air lock testing was not stated in the

Millstone 1 Technical Specifications. This is contrary to 10 CFR 50, Appendix J,

Section Ill.D.2(b)(iv).

Primary Containment Intearated Leak Rate Testina

Section Ill. A.1(d) of 10 CFR 50, Appendix J requires that those portions of the fluid

systems that are part of the reactor coolant pressure boundary and are open directly to the

containment atmosphere under post-accident conditions and become an extension of the

boundary of the containment shall be opened or vented to the containment atmosphere

prior to and during the ILRT. Portions of closed systems inside containment that penetrate

containment and rupture as a result of a loss of coolant accident shall be vented to the l

containment atmosphere. All vented systems shall be drained of water or other fluids to

the extent necessary to assure exposure of the system containment isolation valves to

containment air test pressure and to assure they will be exposed to the post-accident

differential pressure. l

The inspector found that prior to the ILRT in 1994, General Physics had submitted to the

licensee several recommended changes to procedure T-94-1-01, " Primary Containment

Integrated Leakrate Test." Based on the system lineups contained in the procedure, l

w

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

1

'

r 1

l

t

[ 24

General Physics listed 24 penetrations for which LLRT penalties needed to be added to the j

ILRT results. The licensee did not implement the recommendation for 20 of the

i penetrations. This is the second example of failure to identify and correct conditions

l adverse to quality per 10 CFR 50, Appendix B, Criterion XVI.

In its 1996 review, General Physics identified 42 primary containment penetrations I

involving about 18 systems in which piping was not drained and vented (or the degree of

compliance could not be determined) and localleak rate test leakage penalties were not

added to the 1994 ILRT results, included in the list were the penetrations (bold print

below) identified prior to the 1994 test. (An asterisk denotes those penetrations in which

LLRTs also were not conducted correctly.) .

l

Penetration System I

l

*X-9A(B) Main feedwater

X-10A and X-11B Isolation condenser i

a X-14 and *X-15 Reactor water cleanup I

X-16A(B) Core spray l

'

X 17 Reactor vessel head spray

X-18 and X-19 Drywell floor and equipment drains

X-23 and X-24 Reactor building component cooling water

X-25 and X-202D Hydrogen / Oxygen analyzer

*X-30f and X-34f Recirculation pump seal flush

X-35A through E Traversing incore probe

X-36 and X-205 Drywell nitrogen

X-37A through D Scram discharge

X-38A through D Scram discharge

X-39A(B) Containment spray

l

* X-42 Standby liquid contral '

X-43 Low pressure coc".% injection inlet

* X-47 Recirculation sarnple ,

'

X-211 A(B) Suppression hoeder spray / PASS

X-2008 Torus manway B

---

lLRT pressure and flow test connections

The failure to drain and vent the penetrations during the test or to apply the LLRT penalties

to the ILRT results was contrary to Section Ill.A.1(d) of 10 CFR 50, Appendix J.

Technical Specification 4.7.A.3.a requires that containment leakage rates shall be

determined in conformance with the provisions of ANSI N45.4-1972, " Leakage Rate

Testing of Containment Structures for Nuclear Reactors," BN-TOP-1, " Testing Criteria for ;

integrated Leakage Rate Testing of Primary Containment Structures for Nuclear Power i

Plants," and/or the Mass Point Method. Section 7.4 of ANSI N45.4-1972 requires that

area surveys within the containment structure shall be made in advance of leakage rate

testing to establish any tendencies to regional variations in temperature. General Physics

found that temperature surveys prior to ILRT had not been documented and may never )

, have been performed. Failure to perform the required area tempereture surveys was I

! contrary to the Millstone 1 Technical Specifications. In LER 96-26 the licensee committed

I

. . . . . . . .- . . . . . . . - - - . - .- - . . .

.

25

3

'

to perform an ILRT prior to startup of Millstone 1. Additional commitments contained in

LER 96-46 address the specific deficiencies identified above, and are to be implemerced

. prior to conducting the ILRT.

}

c. Conclusions

The inspector found several missed opportunities for the licensee to have identified and

corrected significant and long-standing conditions adverse to quality pertaining to Appendix

J leak rate testing. During an Appendix J program review in 1996, the licensee identified

unperformed and improperly perfarmed leakage rate tests that resulted in the licensee's

inability to demonstrate prirnary containment integrity as required by the Millstone 1

Technical Specifications. The megnitude and variety of problems identified were indicative

of a programmatic breakdown of the Appendix J program. In the aggregate, the failure to

properly implement an effective containment leak rate test program is an apparent violation

of 10 CFR 50, Appendix J. (eel 245/97-02-06)

E1.2 Walkdown of Feedwater Containment Penetrations ]

a. Inspection Scooe

i

The inspector walked down the feedwater system piping between the outboard isolation

valves in the reactor building steam tunnel and the manual isolation valves inside the ,

primary containment. The purpose of the walkdown was to verify that the system was i

depicted correctly in plant drawings, including:

25202-26013 Operations Critical Piping and Instrumentation Diagram

(P&lD)- Feedwater/ Condensate System

25202-29119 P&lD - Nuclear Boiler (GE Dwg. 718E831)

25202-20290 IE Bulletin 79-14 Inspection & Measurement Program - ;

Reactor Building Feedwater System Line No.18 i

25202-20002 Main Steam and Feedwater Piping Sections (Ebasco

Services Dwg. G187497)

25202-29103 DRAVO Corporation Pipe Fabrication Sketch E-2362-lC-

48 (Ebasco Dwg. 5385-8147)

b. Observations and Findinas

The inspector found that P&lDs 25202-26013 and 25202-29119 correctly showed no

bottom drain connections downstream of the check valves inside the drywell. Thus,

drawings 25202-20290,25202-20002, and 25202-29103 were inaccurate in showing a

drain line in these locations. The inspector also noted that one of the inaccurate drawings

(25202-29103) was included in the Updated Final Safety Analysis Report as Figure 6.2-15.

.. -

.

26

in the reactor building steam tunnel, the inspector noted a 3/4-inch line located on the mid-

plane of each feedwater pipe directly opposite of the one-inch LLRT connection. The

connections terminated in valves 1-FW-107A(B), a reducer,1/8-inch O.D instrument

tubing, an instrument root valve, and 1/4-inch O.D. tubing. These branch lines were

shown only on the P&lD (25202-26013) that is used by the operators in the control room.

The inspector also observed that valves 1-FW-107A(B) were not locked in position as is

the case with other small manual primary containment isolation valves. Operations

Instruction 1-OPS-10.11, " Locked Valve List," provides the administrative controls

implemented by the licensee to comply with the containment isolation General Design

Criteria oi O CFR 50, Appendix A, as discussed in Section 4.20 of NUREG-0824,

"Integrater' Plant Safety Assessment - Systematic Evaluation Program - Millstone Unit 1."

Valves 1-FW-107A(B) were not included either in the NUREG or the Operations Instruction,

c. Conclusion

10 CFR 50, Appendix B, Criterion Ill, " Design Control," requires measures to be

established to correctly translate the plant design basis into specifications, drawings,

procedures, and instructions. The inspator concluded that failure to maintain plant

drawings consistent with the plant configuration and to apply procedure controls to valves

1-FW-107A(B) commensurate with those applied to similar manual containment isolation

valves was a violation of this requirement. (VIO 245/97-02-07)

E1.3 Low Pressure Coolant Iniection Heat Exchanaer Foulina

a. inspection Scope

in November 1995 licensee engineers identified scale deposits on the tubes of the "B" train

low pressure coolant injection (LPCI) heat exchanger during a post-cleaning inspection. A

similar condition subsequently was identified on the "A" train heat exchanger. On March

26,1996, Adverse Condition Report (ACR) 9801 was initiated to document the tube

fouling. Since the effect of the scale on heat exchanger thermal capability was

indeterminate, the licensee declared both heat exchangers inoperable. Since the plant was

in the refueling mode, the inoperability of the heat exchangers had no immediate adverse

safety consequences.

The inspector reviewed the Millstone 1 Technical Specification bases and the Updated Final

Safety Analysis Report (UFSAR) to assess the potential impact of the scale deposits on the

ability of the heat exchangers to remove post-accident decay heat from the primary l

containment wetwell (torus) as described in the plant design and licensing bases. The !

following additional documents were reviewed:

Calculation W1-517-921-RE, MP1 Low Pressure Coolant Injection Heat Exchanger

Performance, Revision 4, dated July 22,1994

General Electric Calculation GENE-523-A013-1295, Evaluation of Post-LOCA Net

Positive Suction Head Margin for LPCI and CS Pumps and Suppression Pool

Temperature for the Millstone Unit 1 Nuclear Power Station, dated February 1995

Licenst ' Event Reports 50-245/90-014,90-002, 91-002, and 94-13

_ _

.

l

.

27 i

Millstone 1 Operating License Amendments 46 (Containment spray interlock

setpoint change) and 84 (ANSI /ANS 5.1-1979 Decay Heat Model)

Root cause evaluations regarding LPCI heat exchanger scale deposits, dated July

11,1996 and December 10,1996

b. Observations and Findinas l

l

System Description and Desian/Licensina Basis

l

Two redundant containment cooling subsystems are provided to remove heat energy from I

the primary containment following design basis accidents. During plant operation,

Technical Specification (TS) 3.5.B requires both LPCI containment cooling trains to be i

operable, In terms of peak torus temperature and minimum available emergency core l

cooling pump net positive suction head, the most limiting accident is a small (.01 square

foot) steamline break within the containment. Each subsystem consists of two emergency ,

service water (ESW) pumps, one 5000 gallon per minute (gpm) heat exchanger, and two i

5000 gpm LPCI pumps. When LPCI is initiated, an open heat exchanger bypass valve I

injects 10,000 gpm to the reactor vessel for core floodup. When the reactor core is at l

least two-thirds covered, reactor vessel level is stable or increasing, and containment

sprays are terminated (at 9.0 psig in the drywell), the LPCI system is switched manually to

the containment cooling mode by shutting tha heat exchanger bypass valves. Due to flow-

induced vibration, LPCI flow through the heet exchangers is limited in the emergency

operating procedures to 5000 gpm by stopping one LPCI pump per train. During

containment cooling, LPCI flow may need to be throttled to ensure adequate net positive

suction head (NPSH) to the LPCI and core spray pumps as the water in the torus heats up j

and containment pressure decreases. Consequently, in order to maintain a 15 psid l

differential pressure between the LPCI and ESW systems (to prevent release of radioactive i

material to the environment due to tube leakage), ESW flow may elso be throttled.

The LPCI heat exchangers are vertically mounted, single-pass, shell-and-tube heat

exchangers with LPCI flow through the shell and ESW flow through the tubes. The heat

exchanger duty described in Section 6.2.1.1.3 of the UFSAR is 40 X E6 BTU / hour at 5000

gpm each of LPCI and ESW flow, shell-side inlet temperature of 165 F, and ultimate heat

sink temperature at the TS maximum of 75 F. Per UFSAR Tabk,6.3-4 and the heat

exchanger manuf acturer's data sheet, a design fouling factor of .0005 is assumed. The

fouling f actor is a typical design value for saltwater derived from industry (TEMA)

standards.

Prior to initial plant operation, the design-basis peak torus water temperature was 165 F,

and LPCI and core spray system piping and components were analyzed and qualified to this

value. Prior to commercial operation, FSAR Amendment 18 raised the peak torus water

temperature limit to 203 F in November 1969, and this value is reflected in the basis of TS

3.5.B. The basis states that "...the heat removal capacity of a single cooling loop is

adequate to prevent the torus water temperature from exceeding the equipment

temperature capability which is specified to be 203oF. It also provides sufficient

subcooling so that adequate NPSH could be assured without reliance on containment

pressure except for short intervals during the postulated accident." The licensing basis

remained at 203oF until the licensee completed more refined thermal-hydraulic analyses of

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

i ,

.

28

1

LPCl/ESW system performance and re-evaluation of peak torus water temperature in

1994-1995. Operating license amendment 84 (dated July 24,1995) authorized the

l ' licensee to use the ANSI /ANS Standard 5.1-1979 decay heat model for post-LOCA

l containment cooling analysis. Using this model, a new peak torus water temperature of

194.4 F was calculated. The inspector observed that the analysis assumed that five

! percent of the LPCI heat exchanger tubes were plugged, and that LPCI pump performance

was degraded to the maximum amount (10 percent) permitted by the inservice test

program.

Historical Operability Issues

Millstone 1 originally was designed with two heat exchangers per LPCI train, providing a

single train heat removal capacity of 80 E6 BTU / hour. However, the plant was constructed

and licensed with only one heat exchanger per train. As a result, the LPCI heat exchangers

historically have had very little margin with respect to design-basis capability.

On September 8,1990, the licensee shutdown Millstone 1 when both LPCI heat

exchangers were decle.ed inoperable. An inconsistency existed between the emergency

operating procedures, which specificd maximum LPCI flow of 10,000 gpm, and the heat

exchanger design limit of 5,000 gpm. In the short term, administrative limits (lower than

those permitted by the TS) were placed on drywell, torus, and ESW temperatures. A TS

amendment (No. 46) that increased the containment spray permissive interlock setpoint

from 4.5 to 5.5 psig to 9.0 to 10.0 psig was approved by the NRC to ensure adequate

LPCI pump NPSH An interim containment analysis conducted by General Electric

Company (GE) calculated a peak torus water temperature of 209oF. In April 1991, using

design-basis assumptions for drywell, torus, and ESW temperatures, and a revised

hydraulic model, GE preliminarily calculated a peak torus temperature of 205 F. On

December 27,1991, another GE analysis for throttled LPCl/ESW flows predicted a peak

torus water temperature of 205.7o assuming a heat exchanger ESW inlet temperature of

72 F (2oF less than the TS maximum). GE completed post-LOCA containment analysis

DRF-T23-00642 in November 1992. The analysis concluded that the peak torus water

temperature was 206.8 F for the limiting accident.

In March 1994, the licensee completed hydraulic modeling and analysis of the LPCl/ESW

systems in support of NRC Generic Letter (GL) 89-13, " Service Water System Problems

Affecting Safety-Related Equipment," The analysis indicated that ESW flow may be less

than previously assumed in the containment analysis; that is, that the required 15 psid

differential pressure between the LPCI and ESW systems might not be maintained. The

analysis was predicated on a descending spiral of throttling LPCI flow to maintain adequate

pump NPSH resulting in the need to throttle ESW flow to maintain the required differential

pressure. This in turn would increase torus water temperature requiring additional

throttling of LPCI flow, et cetera. The licensee imposed an administrative ESW

temperature limit of 60 F, and in May 1994 submitted to the NRC a license amendment

request to remove the 15 psid requirement (thus maintaining ESW flow at 5,000 gpm).

The NRC wjected the request and recommended that the licensee evaluate using

ANSl/ANS Standard 5.1-1979 to predict post-accident decay heat generation rates.

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

,

.. !

!

. 1

l

L 29

l

In February 1995, in Order to explore corrective action options, the licensee tasked GE with !

re-analyzing peak torus water temperature for various conditions, with the following

'

results:

l

No ESW throttling (design ESW flow), no heat exchanger tubes plugged and no

LPCI pump degradation: 206.3 I

I

ESW throttled (15 psid maintained),5% tubes plugged,10% pump flow I

degradation: > 213.3

l * ESW throttled, 5% tubes plugged,10% pump flow degradation, ANSI /ANS

Standard 5.1-1979 model: 194.4oF

l

! On July 24,1995, the NRC approved operating license amendment 84, which authorized

the licensee to use the ANSI Standard, and established a new licensing basis torus water

temperature limit of 194.4 F.

From September 1990 to July 1995, Millstone 1 was operated with predicted peak torus

water temperature in excess of the plant design and licensing basis limit of 203 F. On

j mtwo occasions it became necessary to impose administrative limits more rostrict:ve than - '

those permitted in the TS in order to maintain the maximum torus water temperature below -

the higher analyzed values. Operability determinations were needed to permit continued

power operation with unqualified piping, pipe supports, and emergency core cooling

components (LERs 91-02); core spray and LPCI pump motor cooling deficiencies; potential I

pump cavitation; and inability to maintain a 15 psid LPCl/ESW differential pressure (LER

94-13). Although the licensee kept the NRC informed in each instance, in none of these

cases did the licensee perform a safety evaluation of procedure changes, administrative

limits, or exceeding the plant licensing basis per 10 CFR 50.59 to determine if an l

unreviewed safety question existed. The inspector concluded that an unreviewed safety J

question did exist in that the margin of safety as defined (for peak torus water

temperature) in the basis for TS 3.5.B was reduced. Extended operation of the plant in ;

this condition was an apparent violation of 10 CFR 50.59. (eel 245/97-02-08)

LPCI Heat Exchanaer Foulina

Generic Letter 89-13 recommended thermal performance testing to verify

the performance of heat exchangers in open cycle cooling water systems. The licensee

has not performed the testing on the LPCI heat exchangers. Instead, the licensee relied on

periodic inspection and cleaning, and eddy current testing (ECT) of the heat exchanger

tubes each refueling outage.

The licensee has observed over many years tube scale deposits following cleaning

(hydrolazing) of the LPCI heat exchangers. The inspectors reviewed ECT reports from

. Craemer and Lindell Engineers, Inc. (C&L) documenting heat exchanger conditions

I observed during outages between 1980 and 1995. C&L reported deposits and scaling on

the inside of the heat exchanger tubing exposed to ESW, and tubes typically required

hydrotazing prior to ECT in order to permit the passage of inspection p;obes. Some tubes

were found blocked or restricted such that smaller probes had to be used to conduct the

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

_

.

30

inspections. A notable example involved the "B" heat exchanger in December 1995 where

C&L documented a "...significant amount of scale and deposits..." in the tubes, and

recommended monitoring to ensure that the fouling was not affecting overall heat

exchanger efficiency. In 19ES, C&L stated that although no tubes were found to be

blocked, at some future timo a very diligent cleaning should be conducted to remove the

very adherent remaining scale in the heat exchanger tubes.

Licensee heat exchanger performance calculation W1-517-921-RE assumes the design

basis tube and shell side fouling factor of .0005 documented in UFSAR Table 6.3-4 and

the manufacturer's specification sheet. The purpose of the calculation is to derive heat

transfer cos.fficients for various LPCl/ESW flow conditions used by GE as inputs to the

torus water temperature and pump NPSH analysis. The fouling factor is a principal

parameter affecting heat exchanger capability, and is influenced by the presence of scale ,

or deposits on the tube surfaces. Given the partial tube blockages and scale deposits l

observed after hydrolazing during the past 15 years, it is unlikely that LPCI heat exchanger

efficiency was maintained during the plant operating cycles at the values needed to i

support the current post-accident torus water temperature and LPCI pump NPSH analysis. I

For example, the licensee estimated that increasing the assumed fouling factor to .001 l

would reduce the heat transfer rate to 33 X E6 BTU / hour (a 17% reduction), resulting in !

- higher peak torus water temperature and reduced available NPSH. There is a limit to v,hich !

LPCI and ESW flow can be throttled while still removing sufficient heat and maintairing an

acceptable peak torus water temperature. The February 1995 GE analysis inaicated that at

213 F additional reduction of LPCI flow would be ineffective in maintaining adequate j

NPSH. l

The inspector concluded that the licensee's practices 6d not assure that the LPCI heat

exchanger capacity described in the UFSAR and the TS basis was preserved. This is an

apparent violation of T9 3.5.B, which requires two independent containment cooling

subsystems to be opemale during power operation. (eel 245/97-02-09)

Corrective Actions

The licensee has hydrotazed LPCI heat exchanger tubes prior to ECT for several years. '

However, the inspectors found that no procedure guidance existed for the conduct of tube

inspections, and no acceptance criteria for fouling were provided, in addition, the

inspector found that the licensee previously recognized the need for quantitative visual

inspection acceptance criteria. In a memorandum discussing NRC expectations pertalning

GL 89-13 programs, dated February 18,1994, the licensee stated that "Some [ industry]

approaches. . rigorously examine the design calculations and in-place construction to

evaluate the margin on each of the heat exchangers and make an estimation of an

acceptable level of fouling in terms that would permit a tangible surveillance limit." The

licensee also stated that "...we inspect heat exchangers frequently...and qualitatively

document the condition of each heat exchanger, such as: " clean". Acceptance criteria for

these visual inspections has been questioned in the SWSOPIs (NRC service water system

operational performance inspections)." The inspector reviewed the root cause evaluations

performed by the licensee under ACRs 9018 and M1-96-0107 in July and December 1996,

respectively, concerning heat exchanger tube scaling concerns at Millstone 1. The licensee

concluded that personnel had become accustomed to seeing tube scale and assumed that

.

'

l

r 31

! cleaning prior to ECT kept buildup to acceptable levels. Therefore, the effect of the fouling

was never critically assessed by engineering. The root cause identified by the licensee

was lack of a cornprehensive heat exchanger testing, monitoring, and visual inspection

l program with clearly defined acceptance criteria. The NRC independently confirmed the

findings during an inspection of the licensee's GL 89-13 program and commitments

documented in Combined Inspection Report 50-245,336,423/96-09. The licensee has

committed to develop and implement a comprehensive heat exchanger inspection program

and to test heat exchangers to establish performance baselines and verify design margins.

The licensee did not identify as a condition adverse to quality the potential for degraded

heat exchanger performance due to tube fouling, and did not assess heat exchanger

operability. This condition existed from at least 1980 until November 1995, and was the

third example of an apparent violation of 10 CFR 50, Appendix B, Criterion XVI, which

requires that significant conditions adverse to quality be identified and corrected.

UFSAR Discreoancies

The inspector found that the dear:ription of the LPCI heat exchangers contained in the

UFSAR is not consistent with the current design antl licensing basis for peak torus water

temperature (circa November 1969). For example, rable 6.3-4, "Summarv of Low

Pressure Coolant Injection Component Design Pararneters," was not updated to show that

the LPCI heat exchanger shell side inlet temperature (peak torus water temperature) was

203 F (vice 165 F). The same error exists in Section 6.2.1.1.3, "Desigr> Evaluation,

Primary Containment Response Pipe Breaks," which states that the conteinment cooling

heat exchanger will remove 40 X E6 BTU / hour at "...the design condition of a hot inlet

temperature of 165 F." The inspector concluded that the licensee's failure to update the

UFSAR to reflect the correct licensing basis heat exchanger inlet temperature

was an apparent violation of 10 CFR 50.71(e), which requires periodic revision of the

updated FSAR to include changes made in the facility or procedures as described in the

FSAR. (eel 245/97-02-10)

c. Conclusions

The impact of LPCI heat exchanger tube fouling on the system's ability to cool the primary

containment and to maintain post-accident LPCI pump NPSH were reviewed. The inspector

identified apparent violations of NRC requirements pertaining to performance of safety

evaluations per 10 CFR 50.59 and extended operation beyond the plant licensing basis,

operability of the containment cooling system, and corrective action for heat exchanger

tube fouling. In addition, failure to maintain the UFSAR consistent with the current plant

licensing basis was an apparent violation of 10 CFR S0.71(e).

E.1.4 Core Sorav Recirculation Heatun Test

l a. Insoection Scoce (37551)

l

j The inspector reviewed the preparation for a test involving the core spray system in the

l

recirculation mode using normal surveillance procedures. The test was designed to

determine if the core spray pumps could add sufficient heat to increase the torus water

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

.

9

4

.

4

1

32

, temperature for a low pressure coolant inje^. tion (LPCI) heat exchanger therma:

,

performance test. The thermal performance test is required prior to plant ptarten as part of

l an NRC commitment.

!

l b. Observations and Findinos

i

j On Wil 25,1997, during the daily control room walk-through, the inspector discussed

4

with %e :ontrol room operators a planned performance of a 48-hour coro spray pump run.

The irgector was informed that the system was being placed in service as part of a test

to deterrnine if the core spray pumps could provide sufficient pump heat to increase the

2

torus water temperature. Further inspection identified that the test was being controtted

via a one page AWO that referec.ceo the normal sunfoiilance procedure and a

j memorandum, dated April 24,1997, to the shift manager. The subject of the

4

memorandum (MP1-TS-97-0099) was " Core Spray Pump Operation Technical Guidance"

and provided technical objectives, pre-start and operating conditions, and abort criteria.

The inspector questioned the lack of a special test proccdure, safety evaluation, or 10 CFR

j 50.59 review / screening. The shift manager informed the inspector that they were just

3 going to run the pumps using the noimal procedure and take some data, a special test

2 procedure wasn't needed.

<

The inspector immediately discussed the issue with both the Operations Manager and the i

Director of Unit Operations. The test was terminated prior to actually placing the pumps in

service. Other than the control room operators, it was not clear to the inspector that

( operation management understood what the test involved. The engineering staff did not ,

'

clearly communicate their expectation to operations management. Condition report M1-97- l

3

0914 was initiated to document the incident, and stated that management concluded that

j a memorandum was not the appropriate vehicle for giving additional guidance to operations

for the test.

l Subsequently, on May 6,1997, the inspectors had a discussion with the core spray

.

system manager about the heatup test. He suggested that perhaps a troubleshooting plan

1

could be used to perform the evaluation, insisting it was not a test at all, but an attempt to j

gather data on an operating system. The inspector determined that this was not I

appropriate, since the core spray system did not have a problem that troubleshooting

I would be tryirs to correct.

c. Conclusion

The licensee planned to perform a test of the core spray system in the recirculation mode,

using ntemal surveillance procedures. 10 CFR 50.59, " Changes, Tests, and Experiments,"

i states, in part, that a licensee may conduct tests or expsriments not described in the ;

safety analyds rep 0(t without prior Commission approval, unless the proposed test or

experiment involves a change in the technical specifications or an unreviewed safety

question. Since a 50.59 reviev / screening had not been performed in preparation for this

test, the intervention of the inspector in this case, prevented a potential violation of NRC

.- - . - . . - - --.- - - - - - .. -~ -. - - . - - __

.

{

l

[ 33

requirements. The NRC is concerned that a vulnerability exists, which would have allowed

the performance of an unreviewed test to occur.

U1 E8 Miscellaneous Engineering issues

l

l E8.1 Electrical Bus Contral Power Transfer Switches

!

a. Inspection Scope (92903)

l

'

The inspector reviewed the actions being taken by the licensee to address problems with

the control power transfer switches.

b. Observations and Findinas

Each of the 4160 and 480 Volt electrical buses has a transfer switch that permits the

control power for the circuit breakers to be supplied from one of two dc distribution panels.

For each bus, one of the sources is designated as the normal source and the other as an

emergency source. The operation of the circuit breakers and transfer switches is

controlled by Operating Procedure OP 344A,125 Volt dc Electrical System. During normal

plant operation the emergency supply is deenergized by opening the supply circuit breaker

at the dc distribution panel. The transfer switches tre configured for manual operation and

require operator action to switch the control power from normal to emergency power.

During a plant walkdown, the inspector noted that all of the control power was being

supplied from the normal supply and the emergency supply circuit breakers were open.

The licensee was in the process of developing a preventive maintenanco FM) program for

the transfer switches and planned to implement the program during the next refueling

outage. Engineering Work Request (EWR)96-207 was written to develop as built drawings

for each of the transfer switches and to generate specific component identification

numbers. The shift manager has written EWR 97-0116 to request engineering to generate

setpoint design bases information and to have Material, Equipment, and Parts List

evaluations performed for each of the components. As a result of recent problems with

several transfer switches, documented in condition report M1-97-0670, the licensee now

plans to implement a program during the current shutdown.

c. Conclusions

The inspector noted that the two dc electrical power divisions were isolated from each

other by both the contacts in the transfer switches and the open circuit breaker for the

emergency source. This would prevent a failure from affecting both divisions. Also, the

problems that have been experienced occurred while operating the transfer switches during

the current outage. During power operation the transfer switches are not operated and the

failures would not have resulted in the loss of control power to any of the circuit breakers.

The inspector concluded that the licensee is taking appropriate actions to address the

problems with the transfer switches.

_ _ _ . . _ . _ _. ~. .. _ _ __ _ _ . _. . _ _ _ _ _

,

.

l

34

E8.2 (Closed) LER 50-245/96-12, Containment Isolation Valve CU-2J Exceeded the

Maximum Leak Rate Durino Ooeration

(Open) eel 50-245/97-02-10

a. Insoection Scope (37551)

The licensee failed to perform local leak rate testing (LLRT) on a number of containment

isolation velves including the reactor water cleanup return inboard isolation valve CU-29 for

a number of years following the establishment of the LLRT requirement. The failure to

perform the required testing and related issues was addressed in NRC report 245/95-07

and in the attached violation for failure to perform the required LLRT. The operability of

1 CU-29 was discussed again in NRC report 245/95-20. NRC report 245/95-28 also

l discussed the historical operability of valve CU-29, but could not determine if CU-29 would

have performed its intended function. This report also questioned the ability of

containment to perform its intended function following a design basis accident together

i with a single failure of the redundant containment isolation valve, CU-28. In December of

'

1995, valve CU-29 was replaced, resolving the longstanding testing and operabdity issues.

Following removal of the original CU-29, an LLRT was performed and determined that CU-

29 had exceeded the maximum allowable leakage rate. In February 1996, the licensee

+. determined that the observed leakage was contrary to technical specification requirements .

and reported the issue in LER 50-245/96 12. The inspectors reviewed the LER and

associated corrective actions.

b. Observations and Findinas

A review of the design documentation, installation, and testing for the replacement CU-29 ;

valve was performed and found to be adequate. An improved valve design was used to l

more closel/ match the typical operating conditions of the reactor water cleanup system.

A walkdown of the valve, following installation, revealed the addition of the test taps

necessary for local leak rate testing. The post installation LLRT test results were found to

be acceptable,

In December 1995, during the cycle 15 refueling outage, the original CU-29 was iemoved

and a bench test LLRT was performed. However, during the reverse flow air test, the test

pressure of 43 psig could not be achieved as a result of excessive seat leakage. The

licensee subsequently assumed the leakage to be in excess of 300.3 standard cubic feet

per hour (scfh). The licensee stated that a(though a specific calculation was not

performed, the high leakage assumption was supported by measured leakages, at test

pressure, through other large bore penetrations as reported in LER 94-004. Technical

Specification 4.7.A.3.e.(1)(a) requires a combine leakage rate of less than 0.60 La (300.3

scfh) for all penetrations and valves subject to local leak rate tests.

The original CU-29 valve was subsequently disassembled and degradation of the main seat

seal was identified The 8!censee determined that the valve disc and seat were eroded

l- such that the valve would not have prevented backflow leakage and consequently would

'

not have performed its containment isolation function. Disassembly also revealed that the

seat assembly had not been firmly attached to the valve body; however, the disk still had

,

freedom of movement and thus the a%ity to close on reverse flow conditions.

1-

_ ..

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

-

j !

.

'

o

4

35

, . Gross water leakage tests to meet inservice testing requirements had been performed

j 'during the previous three refueling outages. These leakage evaluations were reverse flow

4

water tests using only the elevation head of the water in the reactor vessel and reactor

cavity, approximately 30 psig. Although the tests were performed under different

i

'

conditions and were only intended to verify that valve CU-29 would close, the tests also

provided evidence that the valve was leaking for several cycles. Specifically, during the

,

'

' cycle 13 refueling outage the gross water leakage test results indicated approximately 1

gallon per minute water leakage through CU-29. The previous two tests records only

. indicated that the leakage was less than the acceptance criteria of 1.5 gallons per minute.

V

The LER stated that there was no safety consequence as a result of the event, since

l containment isolation would have been maintained via the redundant isolation valve CU-28.

j The LER also discussed that CU-29 would not have performed its containment isolation

i safety function if a single active failure had occurred in the second barrier CU-28. A

! review of related LERs and other information identified that other vulnerabilities existed

durirg previous operating cycles that could have increased the plant's risk as a result of

the integrated effect of these deficiencies. Specifically, ptior to the cycle 14 refuel outage,

j CU 23 was not c;ualified for harsh environments consistent with the conditions associated

'

with e high energy line break. Additionally, the reactor water cleanup high temperature

. isolation logic, used to detect a high energy line break, was not installed until the cycle 14 j

1 . refueling outage. The leak detection logic deficiency was documented in LER 94-07. The i

!

net impact of these deficiencies could have resulted in the inability to isolate the

j . containment as a result of a high energy line break in the reactor water cleanup system.

j Specifically, CU-28 may not have operated as a consequential effect of the postulated

!

environmental conditions in the vicinity of CU-28.

!

-The review of related LERs also revealed that the combine leakage requirements have not

been met for past operating cycles without consideration of leakage through CU 29. For

,

example, LER 94-04 reported that the three penetrations with the most leakage, more than

doubled the allowed combine leakage.

I

C. Conclusions

The licensee's corrective actions for this issue, replacement of CU-29 and performance of

j the required localleak rate testing was found to be acceptable. The failure of the as-found

localleak rate test and evidence of the longstanding leakage of CU-29 is contrary to

j Technical Specification 4.7.A.3.e.(1)(a), which requires a combine leakage rate of less than j

'O.60 La (300.3 scfh) for all penetrations and valves subject to local leak rate tests. This '

, is an apparent violation (eel 245/97-0211) of NRC requirements.

.

The licensee fai!ad to consider all recent discrepant conditions, related to containment

integrity and evaluate the aggregate impact. Specifically, the LER characterized the

leakage through CU-29 as a single failure vulnerability to containment integrity. However,

prior to cycle 15, containment integrity would have been challenged as a result of the

consequential effects of a high energy line break in the reactor water cleanup system. The

safety implication appears more significant than was discussed in LER 96-12. The licensee

- is planning to submit a supplement to this LER.

'

l

3

-

l

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

l *

I

l

l l

.

36

Report Details

Summarv of Unit 2 Status

Unit 2 entered the inspection period with the core off-loaded. The unit was initially shut

down on February 20,1996, to address containment sump screen concerns, and has

remained shut down to address an NRC Demand for Information [10 CFR 50.54(f)] letter

requiring certification by the licensee that future operations are conducted in accordance

with the regulations, the license, and the Final Safety Malfsis Report.

U2.1 Operatior!s

U2 01 Conduct of Operations

01.1 General Comments (71707)

Using inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing

plant operations to ensure that licensee's controls were effective in achieving continued

safe operation of the facility while shut down. The inspectors observed that proper control

room staffing was maintained, access to the control room was properly controlled, and

operator behavior was commensurate with the plant configuration and plant activities in

progress. In general, the conduct of operations was professional and safety-conscious.

However, one concern was noted regarding the clearing of a danger tag from the air start

valve of the "B" emergency diesel generator before the lube oil system was filled. This is

described in detail in Section U2.M1.1 below, in addition, during a containment tour, the

NRC noted areas of blistering paint on the containment liner. This issue is discussed in

detail in Section U2.E2.1 below.

!

01.2 Timeliness of Corrective Actions - Condition Report Backloa

a. Inspection Scope

The NRC evaluated the timeliness in which the ficensee completed corrective actions

associated with Unit 2 condition reports (CRs),

b. Observations and Findinas

Timeliness for completion of corrective actions has been a longstanding concern at

Millstone. Having a CR backlog in itself is not a reflection of poor performance because as

the threshold for writing CRs decreases, the CR backlog willincrease accordingly. The

concern is the number of CRs that are not closed in a timely manner. To help provide the

NRC some sense of the licensee's progress in addressing the timeliness cor c,ern, the

licensee was asked to provide the number of CRs having outstanding corrective actions

t.1at are greater than 120 days old. Although the NRC does not consider 120 days a level

cf excellence nor is it acceptable when addressing immediate safety concerns, it does

provide some understanding of licensee management effectiveness in addressing the

corrective action timeliness issue.

._ -

__. . .-..

.

37

The 120-day CR backlog for January, February, March, and April, was 738,787,735,and

810 respectively. At the end of the current inspection period, there were 828 CRs greater

than 120 days old that have not been closed reflecting an increase from 798 CRs at the

end of the last inspection period.

DEPARTMENT CRs OLDER THAN

120 DAYS

Operations 53

Design Engineering 258

Technical Support (System 221

Engineering)

Work Planning 22

Maintenance 42

I&C 41

Safety / Licensing 45

Other 139

TOTAL 828

c. Conclusions

4 The backlog of 828 CRs that are greater than 120 days old indicates that timeliness for

completing corrective actions continues to be a concern. Although the 120-day-old CR

backlog has increased from 798 since the last inspection period, the total number (all ages)

of open CRs has declined slightly from 1335 CRs in January 1997, to 1215 CRs in April

<

1997, which is an improving trend. As discussed in NRC Inspection Report 50-336/96-04,

i timeliness and effectiveness of corrective actions is an area in which the licensee must

demonstrate susiained improved performance before the NRC will allow the unit to restart.

!

01.3 Ouality of Corrective Actions

a. Inspection Scoce

This inspection involved a collective evaluation of the 15 open items [ Licensee Event

Reports (LERs), Escalated Enforcement items (Eels), Violations, and Unresolved items

(URis)) that were reviewed as part of this inspection report.

i

4

-. .

.

38

b. Observations and Findinos

For comparison purposes, the following table summarizes the inspection results of the 15

open items that were reviewed in this inspection report and the 16 open items that were

reviewed in Inspection Report (IR) 50-336/96-08, which covered the inspection period from

August 27 to October 25,1996.

Inspection Inspection Report

Report 50- 50-336/97-02

336/96-08

Corrective Actions Acceptable and 4 12

Complete (includes Non-Cited Violations)

Corrective Actions identified But Not 2 2

Sufficiently Completed to Allow Closure

eel Created 7 -

Violation Created 2 1

URI Created 3 -

NOTE: Several LERs were administratively closed in Inspection Report 50-336/96-08

because the specific issue was being tracked by an eel, URI, or violation from a previous

report. To accurately reflect performance, these LERs were counted as an " eel Created,"

"URI Created" or " Violation Created" as appropriate. In addition, the column for Inspection

Report 50-336/96-08 totals 18 even though only 16 open items were reviewed because

the inspection of one URI resulted in 3 Eels being created.

c. Conclusion

A comparison of the inspection results of 16 open items [ Licensee Event Reports (LERs),

Escalated Enforcement Item (EEi), Violations, and Unresolved items (URis)) reviewed in

Inspection Report 50-336/96-08 and 15 open items reviewed in this inspection report

indicates that the licensee has made some progress regarding the quality of corrective

actions. In this report, the corrective actions for 12 of 15 open items were acceptable

while only 4 of 16 were acceptable in Inspect lon Report 50-336/96-08, in this report, a

violation was issued for 1 of 15 open items while 7 Eels and 2 violations were created

associated with the 16 open items discussed in Inspection Report 50-336/96-08.

.

39

U2 08 Miscellaneous Operations issues (92700)

08.1 (Open) Escalated Enforcement item 50-336/96-08-06: Failure to Ensure

Refuelina Pool Drain Valves are Locked Open Durina Operation (SIL 9 & 34

UPDATE)

a. Inspection Scope

The scope of this inspection included a review of Escalated Enforcement item (EEI) 50-

336/96-08-06.

b. Observations and Findinas

Final Safety Analysis Report (FSAR), Section 6.4.3.1, which describes the operation of the

containment spray system during emergency conditions, states that the refueling pool drain

line isolation valves (2-RW-24A&B) are locked open during the operating cycle to prevent

the refueling pool from capturing water. Following a loss of coolant accident, the ability

to cool the core would be lost if a sufficient amount of inventory accumulated in the

refueling pool rather than draining the containment sump where the emergency core

cooling system pumps take suction. Operating procedure OP 2305, " Spent Fuel Pool

Cooling and Purification System," provides instructions for draining the refueling pool

following refueling activities, and therefore, is the procedure that positions the drain valves

prior to operation. The NRC found that the refueling pool draining instructions, as well as

the refueling water purification system valve lineup (OPS Form 2305-2) leaves the valves

in the open but not locked open position as required by the FSAR.

This concern was discussed with the licensee who changed Section 5.27 of procedure OP

2305, which provides instructions to drain the refueling pool to the refueling water storage i

tank (RWST), as well as the valve lineup, to lock open the valves. However, the NRC '

found that licensee failed to change Section 5.28 of the procedure, which drains the

refueling pool to the liquid radiological waste system. Since Section 5.28 makes no

reference to valves 2-RW-24A&B, the valves could have been left in the closed position.

The failure to change Section 5.28 is significant because this section, rather than Section

5.27, is normally the last section to be performed (thereby dictating the final position of

the drain valves) because water remaining in tho refueling pool after the RWST is full must

be drained to the liquid radiological waste. eel 50-336/96-08-06 was created concerning

the failure of the licensee to take adequate corrective actions even after being informed of

the deficient condition.

As corrective mtions, procedure OP 2305, Section 5.28, was changed (and is now Section

4.29) to enuao that valves 2-RW-24A&B are locked open at the end of the draining

operation to the radwaste system. The licensee noted that another cause of the problem

was that system drawing (P&lD), 25203-26023, " Spent Fuel Pool Cooling and Cleanup

Sy tem," did not show valves 2-RW-24A&B in the locked open position. Design Change

Notice (DCN), DM2-00-0089-97, was issued on March 24,1997, to correct the valve

positions on the P&lD. The inspector reviewed Operations Critical Drawings in the control

room and the Unit 2 work control center and verified that the DCN had been entered on

these drawings on March 27,1997.

- _ -

.

I

!

l 40

l

l c. Conclusions

l

l Licensee corrective actions to address the specific FSAR discrepancy were determined to

'

be acceptable. The broader issue associated with failure of the licensee to operate the

facility in accordance with regulations, the license, and the FSAR is the subject of the

March 7,1996, NRC 50.54(f) letter and is considered an NRC restart issue. The broader

issue regarding the failure of the licensee to effectively implement a corrective action

program is also an NRC restart issue. The proposed violation and potential escalated

enforcement action for this item is still under review by the NRC.

U2.ll Maintenance

U2 M1 Conduct of Maintenance

M 1.1 Isolation Taa Cleared Before "B" Emeraency Diesel Generator Lube Oil

System was Filled and Vented

a. Inspection Scope

This inspection involved interviews with licensee personnel, as well as a review of the

licensee's root cause analysis associated with restoration of the "B" emergency diesel

generator (EDG) following maintenance.

b. Observations and Findinas

On March 4,1997, the licensee was preparing to run the "B" EDG following maintenance.

During the pre-job brief, a maintenance technician noted that the danger tag from the diesel

air start valve had been cleared and the lube oil system had not yet been filled and vented.

Had an automatic start signal been received during the 45 minutes that the air start valve

tag was cleared, this could have resulted in the failure of the "B" EDG due to insufficient

lube oil to the upper crankshaft. This is significant particularly in light of the fact that

extensive damage to the "B" EDG occurred in April 1996, when several upper crankshaft

bearings failed due to insufficient lubrication during engine starts.

The licensee's root cause analysis of the March 4 event revealed that Maintenance

Procedure MP 2719K, "ELG Lube Oil System Maintenance," provided restoration steps for

filling and venting the lube oil system. However, the Senior Reactor Operator (SRO)

assigned to work control wlio was coordinating the EDG restoration failed to recognize the

need to defer clearing the air start tag until completion of the fill and vent activity. In

addition, the automatic work order process, the tagging process, and the Operations Work

Control practices collectively did not adequately control the sequencing of the fill and vent

evolution.

Corrective actions taken for this event included: (1) Operations personnel were briefed; (2)

Standard EDG work orders were revised to specify that subsystems are restored prior to

making the EDG available for operation; (3) As an interim measure, the standard EDG

tagout was changed to reflect the need for Shift Manager approval for clearing the EDG air

start valves; and (4) As a longer term measure that was already under development, the

.

41

licensee plans to standardize the process for the removal and restoration of components

and systems. This includes formalizing standardized tagouts and utilization of new tagout

software.

c. Conclusions

The performance of the maintenance technician was excellent in identifying that the "B"

EDG could automatically start, potentially resulting in engine damage. Although this event

did not involve any violation of NRC regulations, the SRO's decision to unisolate the "B"

EDG starting air prior to filling and venting the lube oil system was considered to be a

significant weakness, particularly in light of tbs f act that the "B" EDG was extensively

damaged in April 1996 as a result of insufficient lubrication during routine engine fast

starts. Although licensee corrective actions from the April 1996 EDG failure appropriately l

focused on ensuring adequate lubrication during routine surveillance testing, the EDG !

failure should have raised operator awareness during the March 1997 post-maintenance ;

restoration to ensure the diesel could not start without sufficient lubrication. The root ;

cause analysis for March 1997 event was of high quality and was comprehensive in l

addressing not only the operator performance issues but also the work control process

enhancements associated with equipment restoration.

U2 M3 Maintenance Procedures and Documentation

1

M3.1 Numerous Examples of inadeauate Surveillance Procedures (SIL 8 UPDATE) l

l

a. inspection Scone

This inspection involved a review of NRC inspection reports and licensee event reports

(LERs) that have been issued over the last six months to identify and evaluate any trends in

licensee performance,

b. Observations and Findinos

The review of inspection reports and LERs revealed the following 17 examples of

inadequate surveillance procedures (SPs).

LER 336/96-23 and LER 336/96-26 - Operations Form 2605A-1, " Verifying

Containment Integrity," failed to satisfy TS 4.6.1.1.a in that all of the required

valves were not included on the valve lineup and valves located inside containment

were being marked N/A while at power. Although this condition was initially

identified by the licensee, followup inspection by the NRC, which is documented in

Section M.8.3 of this inspection report, found that the procedure change to

Operations Form 2605A-1 was inadequate.

LER 336/96-24 - Surveillance procedures failed to satisfy the requirements of TS

4.3.1.1.3 and 4.3.2.1.3 for reactor protection system and engineered safety feature

actuation system response time testing. LER 336/96-24, Rev. O, discussed that the

SPEC 200 electronics were not included in the response time testing. Unresolved

Item 336/96-08-09 was created because the NRC found that portions of the circuit

1

.

.- l

42

other than the SPEC 200 electronics were also not being tested. The licensee i

issued LER 336/96-24, Rev.1, which stated that various cabling, wiring, connector

pins, interposing relays, and plant end devices were also not included in the

response time testing. !

LER 336/96-25 - SP 2514D, " Auxiliary Exhaust Actuation System," (AEAS) failed

to test the interlock between AEAS and the enclosure building filtration actuation

system.

'

LER 336/96-30 - The NRC found that SP 21136, " Safety injection and Containment

Spray System Valves Operational Readiness Test," was inadequate in that it failed

to test the high pressure safety injection pump discharge check valves in the closed

direction as required by TS 4.0.5.

'

LER 336/96-35 - SP 2402P, " SPEC 200 Safety Parameters Functional Test," failed i

to test the interlock function associated with the main steam isolation system as i

required by TS 3.3.2.1.

LER 336/96-37 - SP 2619A, " Control Room Shift Checks," did not satisfy the

Technical Specification (TS) 4.7.11 requirement for verifying ultimate heat sink .

temperature.

LER 336/96 38 - SP 2613A/B, " Diesel Generator Operability Tests" used an

inaccurate and non-conservative " Ready to Load" annunciator for timing when the . ;

diesel achieved rated voltage and therefore, failed to satisfy TS 4.8.1.1.2.a.2. The

associated voltage relay was found to be miscalibrated.

LER 336/96-39 - SP 2505H, " Containment isolation Valve Operability Test -

Shutdown," failed to test the containment purge isolation function as required by ,

TS 4.9.10.

LER 336/96-40 The NRC identified that SP 2619A, " Control Room Shiftly Checks"

did not satisfy the requirements of TS 4.1.2.3.2, 4.1.2.3.3, and 4.4.1.4 for

verifying the motor circuit breaker positions of high pressure safety injection pumps,

charging pumps and reactor coolant pumps respectively.

LER 336/97-03 - SP 2618C, " Fire Protection System Smoke Detector Test," failed

to satisfy TS 4.3.3.7.1 (prior to 1995) or Technical Requirements Manual (after

1995) requirements associated with verifying the trip functions of components

associated with the smoke detectors.

LER 336/97-07 - SP 2605N, " Reactor Head and Pressurizer Vent Solenoid Valve

Operability Test," did not adequately verify flow through the vent path as specified

in TS 4.4.11.3.

LER 336/97-08 - Surveillance procedures failed to test some relays in the reactor

protection system logic circuitry.

!

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

.

l

43 l

LER 336/97-09 - Surveillance procedures f ailed to test the delay circuit actuation

modules and downstream circuitry for the containment spray actuation system.

)

LER 336/97-12 - SP 2613A/B, " Diesel Generator Operability Tests" used an

inaccurate and non-conservative " Ready to Load" annunciator for timing when the

diesel achieved rated cpeed and therefore, failed to satisfy TS 4.8.1.1.2.a.2. The

i

licensee was unaware that the spead sensing circuitry also had a lube oil pressure

input that could cause ths " Ready to Load" light to turn on early.

LER 336/97-13 - SP 2404AK2, " Containment Gaseous Process Radiation Monitor

RM 8123B," failed to perform TS Table 4.3-13, Item la, Notation 2, which requires

that the operability of the associated alarm be verified. The procedure inadvertently

jumpered out this alarm rather than the alarm for the stack gaseous radiation

monitor.

LER 336/97-16 - SP 2610B, " Turbine Driven Auxiliary F9edwater Pump Operability

and Operational Readiness Tests," failed to run the pump for 15 minutes or start the

pump from the control room prior to entering mode 3 as regired by TSs

4.7.1.2.a.3 and 4.7.1.2.a.1. j

c. Conclusions

Over the last six months, NRC inspection reports have discussed 17 ffRs involving

inadequate surveillance procedures. For five of the earlier LERs, the f.RC either identified

the issue or NRC intervention was necessary to achieve satisfactorv :orrective action in

the response to Violation 336/96-08-07, which addressed inader;oate containment integrity l

valve lineups, the liceasee committed to review all TS surveil!ance procedures for i

adequacy. Several more recent examples of inadequate surveillances are the result of this

commitment. Other examples are the result of reviews conducted as cart of their 10 CFR

50.54(f) effort and the reviews for Generic Letter 96-01, " Testing of !;afety-Related Logic

Circuits." The more recent examples were generally licensee-identificJ, however, these

appear to be repetitive violations and are being considered collectively as an apparent

violation. The NRC Significant Items List, Item 8, lists surveillance procedure adequacy as

an issue that the licensee must satisfactorily address prior to restart. This is an apparent

violation (eel 336/97-02-12)..

U2 M8 Miscellaneous Maintenance issues (92902)

M8.1 (Closed) Violation 50-336/96-01-06: Anticioated Transient Without Scram (ATWSJ

System Testina

a. Inspection Scooe

The inspectors reviewed the corrective actions implemented in response to the subject

Notice of Violation.

.

44

b. Observations and Findinas

This violation concerned the failure of the licensee to adequately establish and implement a

test procedure to ensure that the ATWS mitigation system actuation circuitry (AMSAC)

performed its function in a reliable manner. Procedure SP 24020 "ATWS Setpoint

Functional Test," Revision 3, was inadequate in that it failed to reflect changes in contact

positions that occurred as a result of system modifications. Also, on July 16,1995, the

procedure was not properly implemented in that 4 of 12 contacts that were specified for ;

checking the position of the AMSAC relays were not in the specified position, yet the

procedure was signed off that the acceptance criteria had been met. l

Corrective actions taken by the licensee included:

Procedure SP 24020 was revised to correct the deficiencies and to clarify the

specific contact verifications required to be performed and documented by the

technicians. l

The revised procedure was validated to ensure the system operability. l

i

a l&C personnel involved in procedure changes received training to ensure they were

aware of how to use the Generation Records Information Tracking System to ensure !

that the latest drawing revisions and design change notices (DCNs) are utilized

when changing procedures.

Enhancements were made to the Design Control Manual and Millstone Procedure

Writer's Guide.

O

The inspector reviewed the associated procedures and drawings, including the results of

the revised test, and found them to be satisfactory. 1

a

c. Conclusions l

The licensee's corrective actions were appropriate and had been implemented to resolve

this issue. This item is closed.

M8.2 (Closed) Violation 50-336/96-04-08: Failure to Adeauatelv Retest Solenoid Valve 2- )

SI-618 Followina Valve Replacement i

l

a. inspection Scope

!

The inspectors reviewed the corrective actions implemented in response to the subject !

Notice of Violation. j

i

b. Observations and Findinas

This violation concerned the fact that the retest of a safety injection system solenoid valve, ;

2-SI-618, following its replacement in 1989 was inadequate in that it failed to identify the !

valve was inoperable due to a missing part. Other associated concerns involved working i

s

i

- - . - ._ - . - - -

-

l

.

45 l

i

outside the scope of the work order, and the failure of routine surveillance tests to

'

collectively verify valve operability. During the short time period this valve is opened each

month, a portion of safety injection flow to the reactor coolant system would have been i

diverted.

The inspector verified that the licensee has taken the following corrective actions: i

1

l

Valve 2-St-618 has been repaired and retested. i

Procedure CWPC-3, " Post Maintenance Testing," has been revised to more

clearly state post maintenance testing requirements.

Surveillance Test Procedure SP 2004P, " Engineered Safety Features ,

Equipment Response Time Testing," has been revised to require verifying )

that 2-SI-618,2-SI-628, 2-SI-638, and 2-SI-648 individually close to satisfy 1

technical specification time requirements. In addition, each valve closure is

timed.

The standard automated work orders for all 94 safety-related air operated

valves contain a caution that precludes disacsembly of the solenc>id-operated j

valves unless engineering is first contacted. i

Because of many other problems in the work control process identified since

1989, the licensee has made significant changes to their work control

process. Procedural controls, personnel training, and the establishment of a l

more effective work control group has minimized the chance of work being

performed outside the scope of an automated work order.

l

Additional training has been given to the maintenance workers.

'

c. Conclusions

The licensee's corrective actions were appropriate and had been implemented to resolve

this issue. This item is closed, j

M8.3 (Ocen) Violation 50-336/96-08-07: Containment Intearity Surveillance Test

i

a. jnsoection Scope

The inspector reviewed the corrective actions that were taken in response to the subject

technical specification (TS) violation,

b. Observations and Findinas ,

,

in December 1996, the licensee identified a violation of the containment integrity plant l

technical specification. Specifically, surveillance procedure (SP) 2605A, " Verifying

Containment Integrity," did not include all of the applicable valves on the associated OPS

Form 2605A-1. The TS requires the valve positions to be verified at least once per 31

. --. . .

.. - -

\

l

!

.

j 40

days. The licensee also identified that when the survoillance was performed with the

reactor operating at power, the operators performing the procedure were annotating the

lineup as "not applicable" for the valves located inside the containment. However, the TS

did not allow for this exemption.

l

The licensee's corrective actions included issuance of Revision 18 to OPS Form 2605A-1

to include the valves that were previously overlooked and the submittal of a TS

amendment request to address the verification requirements for valves inside containment.

During a review of Revision 18 of the surveillance procedure, the inspector noted that in

some cases the required position was listed as OP/CL (open/ closed) or LC/OP (Locked l

closed /open) without any notes or instructions to explain when or under what conditions,

the open position would be acceptable. This was the case for three of the valves added by

Revision 18 and for four valves previously included in the procedure.

The licensee informed the inspector that the reason for this change was to allow selected

valves to be operated as necessary to support plant operations, such as the operation of

shutdown cooling system while in Mode 4. However, the procedure as written lacked the

guidance to ensure that a valve that may be out of the desired position would be identified

and corrected during the monthly performances of the procedure.

The licensee acknowledged this concern and has revised the TS amendment request to

include a detailed discussion in the TS bases regarding the administrative controls for

having one or more of the specified containment isolation valves open.

The inspector also noted weaknesses in the procedure revision process in that the

personnel preparing and reviewing Revision 18 decided that a safety evaluation was not

necessary for the changes being implemented. Also, the Plant Operations Review

Committee (PORC) Cover Sheet for the procedure revision noted that the required position

for valves 2-SI 651 and 2-SI-709 were affected by the revision. However, other affected

valves were not discussed in the cover sheet. The licensee issued Condition Report M2-

97-0787 to address these concerns.

c. Conclusions e

The NRC concluded that procedure SP 2605A does not provide adequate direction to the

plant operators performing the containment integrity verification. This item remains open

pending issuance of the technical specification amendment and associated procedure

changes.

M8.4 (Open) Escalated Enforcement item 50-336/96-08-10 Heavy Load Concern and

Corrective Actions (SIL 36 UPDATE)

a. inspection Scope

The inspector reviewed the corrective actions implemented in response to the subject

escalated enforcement item.

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

.

!

.

t

47

b. Observations and Findinas

This item addresse'd the fact that although the licensee reported that Unit 1 heavy loads,

j as well as Unit 2 heavy loads, have been lifted over a Unit 2,480 Vac vital switchgear

'

room, neither the licensee event report nor the corrective action tracking system reflected

the need for Unit 1 corrective actions. In addition, Unit 1 personnel indicated they were

not aware of the need to change their crane operating procedures.- The failure of the

licensee'to implement adequate corrective actions to address the Unit 1 heavy load

vulnerability was characterized as an apparent violation. l

1

The inspector reviewed the licensee's corrective actions and verified that Unit 1 procedures

MP 791.2, " Turbine Generator Major Components Laydown," and MP 790.4, " Control of

Heavy Loads;" and Unit 2 procedures OP 2352, " Crane Operations," MP 2703H5, i

"Turbino Generator Laydown," and MP 2712B1, " Control of Heavy Loads" have been

revised to recognize and add precautions concerning the moving of heavy loads over

safety-related equipment using the turbine building cranes. As discussed in Inspection

Report 50-336/96-08, areas of the Unit 2 turbine building which are above safety-related

equipment have been clearly marked,

c.- Conclusions

Licensee corrective actions to address the specific concern regarding the control of Unit 1 ,

heavy loads over Unit 2 safety-related switchgear were determined to be acceptable. The '

broader issue regarding the failure of the licensee to effectively implement a corrective

action program is considered an NRC restart issue and will be the subject of future of NRC

inspection activity. The proposed violation and potential escalated enforcement action for

this item is still under review by the NRC.

M8.5 (Closed) Licensee Event Report 50-336/96-16: Non-Functional Circulatino Water

Pumo Trio Function

a. Inspection Scone l

!

The insoectors reviewed the licensee's findings and corrective actions associated with the i

subject LER. l

b. Observations and Findinas

in March 1996, the licensee found that the power supply to the condenser pit level

switches was incorrectly connected. This error would have prevented the automatic trip of

the circulating water pumps on high pit level in the event of a pipe rupture and flooding.

The purpose of this trip is to mitigate the rupture of a circulating water system piping to

prevent flooding of the auxiliary feedwater pumps.

The licensee corrective actions included the following:

Correction of the wiring to the level switches;

. _ -. _ _._ _ _. _ _. ~ > _ . _ . . . . - ~._ _ _ _ _ __ _

-

l

.

1

48

1

l * Development of procedure IC 2440, " Circulating Water Pump Trips Functional !

l Test," to perform periodic functional tests;

l

i

Satisfactory performance of the functional test of the trip functions and;

j '* Addition of a note to the standard automated work order for the level switches to

l identify the need for a functional test following maintenance and included a

reference to this LER.

The inspector reviewed the documentation associated with the above noted corrective i

actions and performed a field inspection of the affected switches. No discrepancies were j

noted during the document review and the switches were found to be in good material

condition. '

c. Conclusions

Licensee corrective actions associated with LER 50-336/96-16 were found to be thorough.

This item is closed.

M8.6 (Closed) Licensee Event Report (LER) 50-336/96-37: Ultimate Heat Sink i

Tomoerature Surveillance Te.st

a. Inspection Scope

The inspectors reviewed the licensee findings anti corrective actions associated with the

subject licensee event report.

b. Observations and Findinas

In December 1996, the licensee identified that surveillance procedure 2619A, " Control

Room Shift Checks," did not verify the ultimate heat sink temperature specifically as

required by the technical specification surveillance requireme The TS specifies that the

ultimate heat sink should be determined to be operable by vedpig the average water

temperature at the Unit 2 intake structure. Procedure 2619,A specifies the use of a single

instrument to monitor temperature up to 70oF. When the temperature reaches 70 F, the

procedure specifies using service water header temperature instruments located in the

turbine building. Thus, while the intent of the requirement was being met, the specific

surveillance requirement to measure the " average" temperature, "at the Unit 2 intake

structure" was not being met.

On March 27,1997, the licensee submitted a proposed revision to the technical

specifications and the technical specification bases. The revision would not change the

ultimate heat sink temperature limit of 75Y, but would clarify the temperature

measurement requirements.

l

.

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

.

'

l

. .

1

49 '

c. Conclusions

4

L The NRC concluded that the licensee's submittal of the technical specification amendment

would address this issue. This item is closed. This licensee-identified technical

specification minor non-compliance is being treated as a Non-Cited Violation, consistent i

! with Section IV of the NRC Enforcement Policy. I

i U2.lli Enaineerina

U2 E2 Engineering Support of Facilities and Equipment

E2.1 Blisterina Paint on the Containment Wall

a. Insoection Scope

4

The NRC reviewed the licensee's evaluation of the blistering paint observed on the

'

containment liner and its potential adverse affects during accident conditions. The I

licensee's evaluation of other containment liner plate anomalies was also reviewed. i

1

b. Qbservations and Findinas

i

During a tour of the Unit 2 containment, the inspector noted areas of blistered paint on the I

containment wall. The inspector asked the licensee if this peeling paint had been analyzed

for its effect on clogging the containment sump screen. The licensee produced Engineering l

Record Correspondence ER-95-0014, (Rev.1, dated 2/9/95) which was written to address l

Trouble Report No.11M2092923 that reported paint blisters i.. the coating applied to the

containment liner plate inside containment. The ER evaluated: 1) whether there would be

increased hydrogen generation due to exposed ziric-based primer; 2) possible containment

sump screen clogging due to dislodged paint during an accident; and 3) potential l

containment liner plate degradation due to containment spray.

The containment liner plate is painted with a zinc-r:ch primer, which is then covered with

Phenoline 305, an epoxy paint finish coat. The ER discusses a hydrogen generation

calculation (NUSCO Calculation W2-517-1043-RE, dated 11/17/92) that had already

conservatively assumed that all of the zinc-based primer inside containment was exposed

to containment spray. Therefore, postulated failure of the blistered paint is not an

unanalyzed event and will not further contribute to the estimated hydrogen generated.

The ER also discusses the possibility of the paint chips clogging the containment sump

screen. Based on the specific gravity of the paint, expected amount of paint chips, mesh

size of the screen covering the containment sump and sump face velocity, the ER

concludes that clogging of the sump screens would be negligible. Due to generic concerns

with containment sump screen mesh size and emergency core cooling system (ECCS)

pump throttle valve clogging, the NRC has a restart item (Significant items List #22)in this

area to inspect and evaluate prior to Unit 2 restart.

. -

._ _ . --

'

l

.

50

The ER further describes the zinc * rich primer being chemically resistant to alkalies. Since

the containment spray solution is slightly alkaline, the general corrosion rate of the I

containment liner plate steelis negligible, and only localized pitting, at very small rates,

would occur. ,

I

A memorandum dated March 29,1995, (DE2-95-227) discusses repainting the

containment liner plate. However, since containment liner paint is a factor in the

containment response analysis, a thorough evaluation of existing conditions should be

done with a proper safety evaluation prior to any repainting efforts. The memorandum

l

recommended establishing a project to fully analyze the details of a coating replacement, '

and if practical, schedule resources to recoat, replace and/or repair the coating over a

number of future outages.

A containment liner plate anomaly was also reviewed by the inspector. This anomaly was

discussed in a plant incident report written in December 1981, and described two bulges in

the same panel of the containment liner plate. An engineering report investigated the

cause of the bulges, and evaluated their existing condition with respect to future operation.

The report discussed typical loading conditions, including thermal expansion. During

normal operation, the superposition of stresses loads the containment in the hoop direction

in compression. In other words, the concrete containment tends to shrink relative to the ;

containment liner plate, which would try to bulge the liner plate inward. However,

calculations show that the liner plate is stronger than the stresses involved. Therefore, this

particular panel of the liner plate was probably originally installed with a slight inward

bulge, which was allowable during construction. This panel, when subject to concrete

shrink during normal operation, would then further deflect inward, creating two noticeable

bulges in a panel of the containment liner plate.

The Millstone Unit 2 Final Safety Analysis Report, Section 5.2.4, " Steel Liner Plate and

Penetration Sleeves," describes isolated areas where the liner has an initial inward curve.

Inward deformation of the liner between anchors may occur under both operating and

accident conditions. The liner and anchors are designed with sufficient ductility to undergo

displacement to relieve the loads without rupturing under these conditions.

A Bechtel topical report also allows the containment liner plate to distort without any

detrimental effects as long as anchor integrity is maintained. The licensee confirmed that

the bulges were contained between two adjacent anchors. A magnetic particle test and a

negative pressure test of the bulge found no cracks or breaks in the containment liner

plate,

c. Conclusion

Based on the inspector's review of the ER, the blistering paint on the containment wall

presents no apparent safety concern in its present or expected future form. The licensee is

also considering whether to recoat, replace and/or repair the paint. Bulges in the

containment liner plate wall are not abnormal occurrences and are analyzed to have no

detrimental effect on future plant operation. It does not appear that bulges are causing the

paint to blister since the blistering paint is found in other places in containment rather than

just in the area of the bulges, which are limited to two - 1' x 10' areas within the same

_.

,

. l

l

i

l

l j

l

51

section of the liner plate. The NRC has a restart item to inspect and evaluate containment

sump mesh screen size and ECCS pump throttle valve clogging.

U2 E8 Miscellaneous Engineering issues (92903)

E8.1 (Closed) Violation 50-336/95 11-01: Failure to Promotiv Identify and Correct i

Enaineered Safeauards Actuation System Desian Deficiency '

a. Inspection Scope (92903)

The inspectors reviewed the corrective actions implemented in response to the subject

notice of violation.

b. Observations and Findinos

The violation cited the licensee for failing to promptly identify and correct a design I

deficiency that resulted in six failures of undervoltage modules in the engineered

safeguards actuation system (ESAS).

Corrective actions taken by the licensee included returning all of the modules to the vendor ' . -

for modifications to correct the design deficiency. Also, a root cause evaluation of the

ESAS design and control process and a self-assessment of ESAS related issues were

perfomied. The licensee has taken actions to support replacement of the ESAS equipment

whch is becoming obsolete and difficult to maintain due to the difficulties in obtaining

rep'acement parts.

Overall, the NRC had found the corrective action program to be weak and, as discussed in

NRC Inspection Report 96-04, the corrective action program must be demonstrated to be

effective prior to restart of any of the Millstone Units.

Other actions associated with the overall corrective action program included the

development and issuance of procedure RP-4, " Corrective Action Program." Revision 4 to

this procedure was issued in February,1997 in an ongoing effort by the licensee to

implement broader based improvements to the corrective action process.

The licensee has also implemented the Maintenance Rule program which should also

identify repetitive equipment problems,

c. Conclusions

Licensee corrective actions to address the specific concern regarding the ESAS

undervoltage module failures were determined to be acceptable, and therefore this item is

closed. The broader issue regarding the failure of the licensee to effectively implement a

corrective action program is considered an NRC restart issue and will be the subject of

future of NRC inspection activity.

l

r- --

'

l

. 1

l

.

l

! 52

l E8.2 (Closed) Unresolved item 50-336/95-25-03: Enclosure Buildina Filtration System

Sinale Failure Vulnerability l

a. Insoection Scope

The scope of this inspection included a review of Unresolved item 50-336/95-25-03.

l

b. Observations and Findinns

This unresolved item discussed a single failure vulnerability associated with the i

containment and enclosure building purge system (CEBPS). The Enclosure Building is a

secondary containment that is credited in the accident analysis for ensuring that leakage

from the primary containment following a loss of coolant accident is directed to and treated l

by the safety-related enclosure building filtration system (EBFS) prior to release. CEBPS is l

a non-safety-related system in which the main exhaust fans can be aligned to take suction

from the enclosure building to reduce the temperature for personnel comfort. Unlike

CEBPS, EBFS contains charcoal filters that remove lodine, which is necessary for l

maintaining offsite doses to less than 10 CFR Part 100 limits at the site boundary. The l

single failure vulnerability involves the one component in CEBPS that is safety-related, the

purge damper AC-11, which is designed to close on a containment isolation actuation

signal (CI AS). Because CEBPS does not have charcoal filters,if purge damper AC-11 failed

to close due to a mechanical failure or CIAS signal failure,10 CFR Part 100 limits for l

offsite doses could be exceeded.

Subsequent licensee reviews determined that the original licensing basis did not require

damper AC-11 to meet the single failure criteria. The NRC also reviewed early licensing

basis records, and found there was insufficient information available to determine whether j

single failure reliability was required during enclosure building purging operations. Due to ;

this uncertainty, the NRC determined that correction of this single failure vulnerability '

would be considered a backfit. The NRC decision on whether to require the licensee to )

take backfit corrective actions was based, in part, on the information the licensee provided l

in Licensee Event Report (LER) 50-336/94-40-02 which stated: (1) The vulnerability only l

exists while purging the enclosure building at power, which is an infrequent evolution that

is performed when the enclosure building becomes too hot for workers; (2) If damper AC-

11 failed to close, " Radiation Monitoring alarms and trends would indicate an abnormal

condition and alert the operators to take corrective actions to quickly terminate the event." I

In addition, Attachment 1 of the LER discusses the operator actions that can be expected l

"without procedure changes" and states that Unit 2 stack radiation monitor would alarm in !

the control room and would "wll the operators of the unfiltered release condition and they

will secure main exhaust fans." The NRC reviewed this information and determined that a

backfit to correct single failuro vulnerability with damper AC-11 was unnecessary.

During the final NRC review of Unresolved item 50-336/95-25-03, the inspector reviewed

Alarm Response Procedure (ARP) 2590H, " Alarm Response for Control Room Radiation

Monitor Panels, RC-14," Alarm RC-14C, " Unit 2 Stack Gaseous," to confirm the LER

statements indicating that operators "would quickly terminate the event" by securing the

main exhaust fans. The inspector found that procedure ARP 2590H does not specifically

direct operators to secure the main exhaust fans based on the failure of damper AC-11 to

__. .

< l

4

-

,

1

'

53

close. Rather, the procedure first directs that all three main exhaust f ans be started if it is

desirable to increase dilution (The main exhaust fans take suction from various sources

such as the auxiliary building.) The procedure then directs that one or more (and possibly

all) main exhaust fans be secured as necessary to maintain the release rate below the

95,000 microCuries per second limit. The release rate is a value that is manually

calculated by :.he operators using the stack radiation monitor reading and a conversion

factor based en the main exhaust fan combination. Although all three main exhaust fans

would be secured if necessary to prevent exceeding the release rate limit, sr curing these l

fans is not a certainty, which is inconsistent with the LER statement that cserators "will j

secure main exhaust fans." in addition, since securing the main exhaust fans involves a i

manual calculation by the operators to evaluate the release rate, the LER statement I

regarding "quickly terminate the event" is questionable. The need to perform the manual

calculation increases the possibility of operator error. !

The NRC discussed the concern with the licensee, vihe changed procedure ARP 2590H

(Rev. 2, Change 6) to state that if a loss of coolant .iccident has been diagnosed, stop all

main exhaust fans.

c. Conclusions

l

10 CFR 50, Appendix B, Criterion XVI, requires that measures be established to assure

that conditions adverse to quality, such as deficiencies, deviations, and nonconformances

are promptly identified and corrected. The NRC decision to not backfit the licensee to

address the damper AC-11 single failure vulnerability was based heavily on the operator

compensatory action described in LER 50-336/94-40-02 involving securing the main

exhaust fans in response to the Unit 2 stack radiation monitor alarm. The licensee's

corrective actions to address this single failure concern were inadequate in that the Unit 2

stack radiation monitor alarm response procedure failed to ensure the main exhaust fans

were secured and is considered a violation. (Violation 336/97-02-13)

E8.3 (Closed) Unresolved item 50-336/95-27-01: Review of Reload Safety

Analvsis Recort

a. inspection Scone

The inspectors reviewed the corrective actions implemented in r'sponse to the subject

unresolved item.

b. Observations and Findinas

This item concerned a weakness that was identified for Cycle 10,11,12, and the current

Cycle 13 Reload Safety Analysis Report at Millstone 2. The licensee apparently had no

records which documented that the reload safety analysis reports had been reviewed.

Paragraph 6.2.2 of NGP 5.05 procedure notes that "the result of the independent review

shall be documented on the Design Review Form (Figure 7.4) by the independent

reviewer "

e.

.

54

The inspector reviewed Nuclear Group Procedure, NGP 5.05, " Design input, Design

Verification, and Design Interface Reviews" and NGP 6.06, " Processing and Control of

Purchased Material Equipment, Parts, and Services." The inspector also reviewed Nuclear

Engineering Procedure, NFE 4, " Purchase of Analysis on a Sole Source Basis" step 6.4.9 to

ensure all deliverables were received, and to provide reasonable assurance that the

engineering service was accurate and free of enorc. Furthermore, the inspector reviewed a

sample of Fig. 7.4 Form " Review of Analysis Deliverable" and found it to be technically

accurate, and in compliance with NFE procedures.

c. Conclusions

The licensee's actions taken to address this unresolved item were acceptable. This item is

closed.

E8.4 (Closed) URI 50-336/95-81-01 and (Open) eel 50-336/96-201-29: Trendina and

Prioritization of Non-Conformance Reports (NCRs)

a. Insoection Scone (92903)

.

The inspectors reviewed the corrective actions implemented in response to the subject

i

unresolved item (URI) and escalated enforcement item (EEI).

b. Observations and Findinas

The licensee's quality assurance program requires that a trend analysis of non-

conformances documenting program / procedural problems be performed, and the trend

analysis reports identifying program /procedursi problems be periodically reported to upper

management by the organization responsible for controlling the problem report document.

URI 50-336/95-81-01 concerned the fact that non-conformance reports (NCRs) initiated by

the QA Department and Level "D" adverse condition reports (ACRs) were not included in j

the trend report nor was there any requirement to trend these items. Based on this finding, '

the lack of trending of NCRs and verifying the effectiveness, and adequacy of the ACR !

database by the Quality Assurance Department was considered an unresolved item.

l

In a related finding, eel 50-336/96-201-29, identified inadequate corre<#ive actions

concerning the resolution of existing NCRs. Many NCRs had remained op3n for years

without any corrective actions being performed,or, if performed, the NCR 'was not

adequately closed out. The lack of trending noted above and inadequa e corrective actions

to audits apparently helped lead to this problem. The ,aspector reviewed a listing in which

all outstanding Unit 2 NCRs have been identified and assigned responsible engineers for

closure.

The licensee has established procedure OAS 2.14, " Trend Review of Non-conformance l

Reports," which requires trending of NCR backlogs and adverse trend areas. The inspector l

verified that such trend reports have now been issued, in addition, QA has performed

several audits of the NCR and ACR (CR) pre:ess. Procedure RP4, " Corrective Actions

Program," has been issued to more formalize the corrective actions process. Each unit has

established a corrective actions group to resolve ACRs and other corrective action issues

_

.

.

i 55

and backlogs, in addition to the NCR trend report, as required by RP4, each unit issues a

quarterly " Corrective Action Trend Report." The inspector observed that QA has

established a program to periodically audit the corrective action program and to perform

,

periodic surveillances of the implementation of licensee corrective actions, and that such

l audits and surveillances are being conducted.

The licensee issued a memorandum on August 15,1996, titled " Interim Guidance on

Operability /Reportability Reviews of NCRs." This memorandum requires that an ACR (now

called a CR) be generated in tandem with the NCR to provide the operability /reportability

review by the shift rnanager. The only exception is NCRs generated by the receipt

inspection process. The memorandum further stated that appropriate procedures were to

be revised to reflect the requirements of the memorandum. The inspector noted that

neither procedure RP4 nor the NCR procedure, NGP 3.05, have been revised to reflect the

NCR-CR tie in. During t, licensee review of corrective actions to resolve this EEi, the

licensee observed that procedures had not yet been updated to reflect the memorandum.

On March 24,1997, Condition Report M2-97-0466 was issued to identify the fact that

procedures had not yet been revised.

c. Conclusions

Based on the above review, licensee corrective actions associated with URI 50-336/95-01-

01 were found acceptable and this item is considered closed. However, eel 50-336/96-

201-29 remains open because procedure RP-4 has not yet been changed to proceduralize

the practice documented in the August 15,1996, memorandum regarding generating an

ACR as a means of tracking NCRs. l

l

E8.5 (Closed) Follow-un Item 50-236/95-201-07: Control of Molded Case Circuit Breaker

Adiustable instantaneous Trio Settinas l

1

a. Insoection Scone

The inspectors reviewed the corrective actions implemented in response to the subject

follow up item.

b. Observations and Findinas

During the 1995 Restart Assessment Team inspection of Millstone Unit 2, the team

observed the testing of two molded case circuit breakers. The testing was observed to be

satisfactory except that there was no documentation of the trip settings for the breakers.

Breakers were routinely left at their high trip settings. This appeared to have the potential

for causing spurious tripping of safety related equipment, in a July 31,1996, response to

a letter from the NRC dated May 3,1996, the licensee stated that adjustable instantaneous

trip settings will be incorporated into a controlled plant procedure.

l

The inspector verified that the licenseo issued site common maintenance procedure C MP

!

751 A, " Molded Case Circuit Breaker inspection and Testing." This procedure superseded

Unit 2 specific procedure PT21421D. Essentially this procedure specifies that breaker

!

.

56

setting shall be left in its "as is" position for already installed breakers. Breaker settings for

newly installed breakers shall be per engineering instruction.

Licensee engineering identified all adjustable molded case circuit breakers, and determined

that all but one would be set at the high setting. This was done by a memorandum dated

June 8,1995. Hence, the high setting observed by the inspection team was the correct

setting for the breakers being tested. The inspector observed that engineering drawings

have been revised to specify adjustable breaker settings. With the exception of one

breaker set on low all other breakers are set on the high setting.

c. Conclusions

Licensee corrective actions to address this follow-up item were acceptable. This item is

closed.

E8.6 (Closed) Unresolved item 50-336/96-05-11 (IFS No. 96-05-17); Failure to Update

the FSAR in a Timelv Manner as Reauired by 10 CFR 50.71(e) (SIL 38 CLOSED)

a. Inspection Scone

The intpectors reviewed the corrective actions implemented in response to the subject

unresolved item.

b. Observations and Findinas l

1

This unresolved item noted that some detailed information relative to the cooling capability

of the spent fuel pool was provided to the NRC by the licensee in support of Technical l

Specification (TS) Amendment No.114 but was not included in subsequent updates to the j

FSAR. The supporting safety evaluation to the TS amendment stated that the spent fuel

pool cooling system may not be capable of removing the decay heat required to maintain I

the SFP temperature below 140 degrees Fahrenheit during the first 21 days of a core I

offload. This determination assumed a nominal one-third core offload and a single failure in

the cooling system. The amendment modified the TS to require a minimum decay time of

504 hours0.00583 days 0.14 hours 8.333333e-4 weeks 1.91772e-4 months (21) days for a one-third core offload. 10 CFR 50.71(e) requires that FSAR I

updates must be filcd annually or 6 months after each refueling outage provided the

interval between successiv9 updates does not exceed 24 months. Amendment 114 was

issued on November 14,1986, and as of the dates of inspection 50-336/96-05, the FSAR

had not been revised to reflect the SFP cooling capability as discussed above.

The absence of information in the current FSAR was considered a potential violation of 10

CFR 50.71(e). The item was called unresolved pending a more broader collective review

FSAR discrepancies. The inspector verified that the FSAR has now been revised to reflect

SFP capability, and the specific technical issue has been resolved.

The unresolved item is nM cited because it is an isolated example of a much broader issue

for wh. 5 potential esc';ated enforce actions have been issued concerning plant operation

that has tu. -sistent with the licensing basis. The licensee is currently addressing

these issues through its configuration management program. In addition there is an

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

.

57

independent Corrective Acho Program (ICAVP) by the licensee to verify their licensing

basis; and, there will be su aquent NRC verification of the ICAVP process.

The inspector reviewed the following licensee procedures:

(1) NGP 4.02, " Proposed Technical Specification Change Requests and Proposed

Technical Requirements Manual Changes," Revision 10, January 8,1997

(2) Design Control Manual (DCM), Revision 5, April 16,1997

(3) NGP 4.03, " Changes and Revisions to Final Safety Analysis Reports,"

Revision 9, April 8,1997

NGP 4.02 requires that the originator of a TS change determine if information in the FSAR

will require a change or updating, if the proposed TS change request is approved, and if

the NRC approves the amendment to the TS. If the FSAR will be affected, an FSAR

change request should be initiated. NGP 4.02 also provides a mechanism for processing

the proposed FSAR change request, if a design change is made, the DCM provides a

mechanism for reviewing the design change for potential FSAR changes. In addition to

- procedural requirements, DCM Form 3-28, Design Change Administrative Checklist

provides a mechanism for documenting a proposed FSAR change request, if required by

the design change. NGP 4.03 provides the mechanism for actually updating the FSAR.

Following the requirements of these procedures should assure that any current TS or

design changes are incorporated into periodic FSAR updates when required,

c. Conclusions

Based on the above licensee actions already in progress concerning configuration

management, required independent reviews to follow, and subsequent planned NRC

inspection activity; and because the FSAR has been revised to reflect the correction of the

specific item identified above, this unresolved item is closed.

E8.7 (Ocen) Escalated Enforcement item 50-336/96-09-10: Erneraency Diesel Generator

Corrective Actions (SIL 26 UPDATE)

a. Inspection Scope (92903)

The inspectors reviewed the licensee actions that were taken following an emergency

diesel generator (EDG) bearing failure in March 1996 for which the licensee failed to

identify the root cause and implement corrective actions,

b. Observations and Findinas

Subsequent to the EDG bearing failure in March 1996, a catastrophic failure of the "B"

EDG engine occurred. The engine failure resulted in an extensive review by an Event

Review Team (ERT) which identified numerous contributing causes to the failure, and

! provided a comprehensive cortactive action plan.

!

-

_ _ . _ . _ . . . _ _ . . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . . _ . _ _ _ . _ . _ _ _ . _ . ..m. _ _

..

4

..

"

58

. The failure of the licensee to identify the root cause of the March 1996 bearing failure and

implement appropriate corrective actions was one of numerous examples of inadequate

corrective action cited by the NRC in that time frame. As discussed in NRC Inspection

Report 96-04, issued June 6,1996, the overall corrective action program was not effective

in correcting identified deficiencies.

- The inspector noted that in addition to the specific corrective actions that were

_, implemented as a result of the engine failure, the licensee has taken actions to improve the

, corrective action program. This included a revision of procedure RP-4, " Corrective Action

! Program," in February 1997, The licensee provided Condition Report M2-97-0363 and the

initial root cause investigation as an example of improvements in the corrective action

process. The CR documents a problem where an EDG was aligned for automatic starting

following maintenance work but prior to filling and venting the lube oil system. An-

l' automatic start could have resulted in engine damage as a result of inadequate lubrication.

The condition was documented and a root cause investigation was performed even though

no automatic start or engine damage occurred.

c. Conclusions

4

l - This item remains open pending the completion of NRC considerations of escalated y

enforcement action for this issue. The overall effectiveness of the corrective action

program is being tracked in the NRC Restart Assessment Plan as Significant Issues List

Item Number 5.

!

1

k

4

l

,

2

J

1

.

l

.

!

59

Report Details

Summarv of Unit 3 Status

Unit 3 remained in cold shutdown (mode 5) status throughout the inspection period. The

licensee continued its implementation of the Millstone Unit 3 Recovery Plan and the

configuration management program activities in support of the milestones leading to the

readiness for the unit restart.

On May 5,1997, the licensee announced that Unit 3 was designated as the lead unit and

would be the first unit ready for external review under the provisions of the Independent

Corrective Action Verification Program (ICAVP). In addition, the licensee decided that the

recovery managers for Unit's 1 and 2 would assist Unit 3 in its recovery process. Mr. J. P.

McElwain, in addition to continuing as the Unit 1 Recovery Officer, assumed responsibility

at Unit 3 in the area of physical plant readiness; and Mr. M. Bowling, the Unit 2 Recovery

Officer, was assigned additional responsibility for Unit 3 regulatory readiness. Both

individuals report to Mr. M. Brothers, the Unit 3 Vice President and Recovery Officer.

On May 19,1997, Mr. M. Brothers declared Unit 3 ready for commencement of the

ICAVP. Mr. M. Bowling concurred with this decision after the conduct of an independent

contractor review of Unit 3 readiness. As of the end of this inspection period, the

Millstone Recovery Oversight group was continuing its review of Unit 3 readiness,

representing the last concurrence required for the recommendation to the executive level

for the ICAVP activities to commence. The Unit 3 " target" date for declaration of ICAVP

readiness to the NRC is May 27,1997.

U3.1 Operations

U3 01 Conduct of Operations

01.1 Operational Observations and Issue Followuo (71707,92901)

Using Inspection Procedures 71707 and 92901, the inspectors conducted frequent reviews

of ongoing plant operations, to include plant inspection-tours, control room observations,

and witness of licensee planned or exigent meetings and briefings. Where appropriate,

interviews were conducted with licensed operators and other support personnel to assess

the level of control and detail of knowledge being implemented with regard to observed

operational evolutions. During this inspection period, in addition to the routine tours and

observations, the following activities were specifically examined, either in progress, or to

address questions bearing on the final disposition of identified problem areas:

a mispositioned chlorination throttle valves in the service water system;

reference condition reports CR M3-97-0717 & 0731

The inspectoi interviewed workers, examined the subject valves, and

reviewed and discussed w;th the licensee CR investigator the description,

causal factors, and generic implications of this event. Licensee corrective

, _ _ .

I

4

60-

l

, measures were determined to be commensurate with the signibcance of the 1

identified problem. l

!

conduct of emergency core cooling system pump performance testing as an )

Infrequently Performed Test or Evolution (IPTE); reference IPTE procedures

{

IST 3-97-001 & 002 ;

The inspector reviewed the safety evaluation for the conduct of both IPTEs,

evaluating the use of dedicated operators during test performance. The

inspector also witnessed a portion of the charging pump testing (IST 3-07- )

001) and noted that the Shift Manager terminated the conduct of the IPTE q

activities on two occasions when a potential technical specification (TS) l

violation was perceived (CR M3-97-0908) and when a procedure problem

was ideniified (CR M3-97-0924). The inspector reviewed the reportability

evaluation tcr CR M3-97-0908 and concluded that Shift Manager had

correctly and conservatively suspended testing, even though the subsequent

assessment determined that no actual violation of TS had occurred. In both

cases, operator actions were deemed to have been prudant.

charging pump cavitation during valve operability testing; reference CR Mb

97-0934

i

The inspector reviewed the surveillance procedure SP 3604A 5 and re;ated

emergency and abnormal operating procedures. License operators on shift

,

were interviewed and the inspector verified that the affected pump was

maintained in an operable status during the conduct of testing and as a result

of subsequent evaluation of the CR. Inspector questions regarding design

flow requirements were clarified by cognizant engineering personnel. The I

shift manager took appropriate cautionary actions to ensure pump flows did i

not approach observed cavitation limits during continued performance

testing. !

"

1

control of locked closed manual containment isolation valves in the bypass

lines for the turbine driven auxiliary feedwater (TDAFW) steam supply lines;

reference Northeast Nuclear Energy Company letter B16364

The inspector discussed the purpose of these steam supply bypass lines with

a technical support engineer and confirrned that they had not been

inappropriately used for pre-warming the TDAFW pump prior to the TS

required, time response testing and quarterly pump performance surveillance

activities. The inspector examined the existing field piping configuration and

valve conditions during an inspection tour in the engineered safety features

(ESF) building. Licensee commitments, as documented in licensee event

report (LER)97-013 and relating to the revision of design and licensing

information on the subject bypass lines and valves, will be reviewed during a

future inspection in followup to the LER.

_ _

e i

l

.

l

l 61

loss of "A" train ESF building ventilation as a result of unscheduled work on

"B" train equipment; reference CR M3-97-1385

The inspector verified that the licensee entered the appropriate TS limiting l

condition for operation, governing residual heat rernoval system operability,

based upon the degradation to the supporting ventilation system. Unit 3

management ordered a stop work to the release of new safety-related,

scheduled work until an event review team assessed the identified concerns

and recommended corrective measures. The inspector observed an event

review team meeting on this s*.(ect and reviewed the Unit 3 Director

decision to allow for a work restart. The conditions placed on the new work

controle appeared to address the causal factors associated with this event.

Work D had already been released to the field under conditions similar to

the unschaduled ventilation work were subjected to further shutdown risk

and safety impact. Significant licensee management attention was devoted l

to this event and its acceptable resolution.

Overall, as determined by NRC followup to the issues and events identified above, the

inspector concluded that the licensee demonstrated an acceptable approach to the control

of planned operational evolutions and good response to problems that developed during the- ..- J

progress of those evolutions. Event review and investigation appeared thorough and

corrective measures appeared well directed. The inspector noted that the effectiveness of

the corrective actions to preclude similar problematic events can only be demonstrated

over time. Nevertheless, the licensee's short-term response and directed additional

controls provided evidence of proper Unit 3 management attention to emergent operational

areas of concern.

01.2 Ooerational Followuo of Safetv Grade Cold Shutdown Controls

a. Inspection Scope (71707,92700)

The Millstone Unit 3 Final Safety Analysis Report in section 5.4.7.2.3.5 delineates the

requirements for bringing the reactor to a cold shutdown condition under specified criteria

and assumptions. These design-basis critoria, termed " safety grade cold shutdown"

(SGCS) controls, have a regulatory r.exus to the guidance provided in USNRC Regulatory

Guide 1.139. Previous NRC inspect'ons have reviewed the operational control of SGCS

equipment, resulting in the documentation of some unresolved items, one of which (URI

423/96-01-07) remains open. During this inspection, a further review of SGCS controls

was conducted by conducting a followup of condition report CR M3-97-0835, which

described a potential design deficiency with the accident analysis involving certain SGCS

equipment.

b. Observations and Findinas

The SGCS components discussed in CR M3-97-0835 are the four main steam pressure

relieving bypass valves,3 MSS *MOV74A,B,C,& D. The condition being reviewed was the

postulation that one of these main steam pressure relieving bypass valves being out of

I

service would violate the design basis of the Millstone Unit 3 steam generator tube rupture

.

62

(SGTR) analysis. Further engineering evaluation of this design concern and the licensee's

analysis of the SGTR accident response is discussed in section U3.E2.1 of this inspection

report.

The inspector reviewed the licensee's original reportability determination for CR M3-97-

0835. The licensee had conducted a search of the inservice test records for the subject

main steam pressure relieving bypass valves, had found none of the 3 MSS *MOV74 valves

inoperable with the plant in mode 1, and had concluded that the plant had not been

operated outside its design basis. Therefore, the licensee determined that the event

considered in conjunction with CR M3-97-0835 was not reportable in accordance with the

criteria of 10 CFR 50.73.

The inspector indicated to the licensee, however, that in 1996 two block valves

(3 MSS *MOV 18A&B) upstream of valves 3 MSS *MOV74A&B had been de-energized

closed with the plant in mode 1 (reference: NRC inspection report IR 423/96-01). The

block valve closure had been effected to support maintenance on two atmospheric steam

relief valves, located downstream of the block valves and in a parabel flow path to the

main steam pressure relieving bypass valves. This configuration had not been evaluated in

the licensee's original reportability determination.

After discussions with the inspector, the licensee conducted another reportability

evaluation (M3-97-0835 R1), also considering another adverse condition report, ACR 935,

associated with the potential inoperability of the 3 MSS *MOV18 block valves. The licensee

concluded that local manual action to mitigate a SGTR event with the block valves closed

would not be feasible, and therefore that the conditions evaluated were reportable in

accordance with 10 CFR 50.73(a)(2)(ii)(B). The inspector reviewed the revised

reportability determination, noting that the licensee production maintenance management

system (PMMS) records had not indicated the 3 MSS *MOV18A&B valves to be inoperable

during the time period covered by IR 423/96-01.

The licensee in following up URI 423/96-01-07, had recognized a programmatic weakness

in the plant SGCS controls, and currently has a proposed technical specification change

request (PTSCR 3-25-97) being processed to address the concern over the need for more

rigorous control of the main steam pressure relief bypass valves. The problems

documented above, regarding both the lack of PMMS records indicating accurate block

valve positioning and the noted impact upon accident analysis and 10 CFR 50.73

reportability, further highlight the need for additional licensee control in this area,

c. Conclusion

Licensee review of all the events related to CR M3-97-0835 has resulted in the

determination that the identified condition is reportable to the NRC in accordance with 10

CFR 50.73. The NRC will conduct further inspection of this issue after the licensee event

report is submitted. Furthermore, URI 423/96-01-07 remains open pending review of the

PTSCR and the licensee's implementation of corrective actions, specifically with additional

focus on the facts developed during this inspection.

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

.

i

!

63

,

U3 07 Quality Assurance in Operations (40500) (SIL 37 & 41 UPDATE)

l 07.1 Operational Oversiaht Activities

The inspector met with several Nuclear Safety & Oversight (NS&O) personnel ta discuss

activities and initiatives in the areas of corrective action program enhancement, self

l

'

assessments, priorities for unit restart readiness,10 CFR 50.54(f) involvement, and

conduct of the Nuclear Safety Assessment Board (NSAB) meetings. The inspector n.oted

that the licensee met its established milestone to implement a new corrective action

program (RP-4) by March 31,1997. Evaluations are continuing for proper implementation

of this program and assessing its effectiveness over the next several months. Additionally,

the Inensee has hired a contractor with industry-wide corrective action program experience

to assess the quality of root cause analyses and the resultant reports and corrective ur: tion

conclusions.

,

The inspector also noted that the licensee has implemented a self-assessment program

improvement plan, including efforts involving process and procedural revisions, training

initiatives, implementation and monitoring activities, and the confirmation of results, lo

developing this improvement plan, the licensee has " benchmarked" similar programs at

other nuclear plants where the licensee's self assessment ir,;tiatives have-been deemed

effective.

In monthly meetings with NS&O performance evaluation personnel, progress on the

Nuclear Oversight Recovery Plan, assessment of the unit readiness to restart, and strategic

planning priorities were discussed. Performance evaluation participation in daily plant

meetings and event review team efforts, as well as in the conduct of surveillances directed

to perceived plant or ppgrammatic problem areas, have been noted.

Finally, the inspectors have been kept informed, through status reports and periodic

meetings, of the progress made by the Recovery Oversight organization in its assessment

of ongoing configuration management program work. During an NSAB meeting attended

by an NRC branch chief, differences between Recovery Oversight er.d line management n

the expectations for measuring progress of the 10 CFR 50.54(f) project were discussed.

This resulted in a subsequent meeting chaired by the Senior Vice President and Chief

Nuclear Officer to establish a common set of goals. The results of these licensee

deliberations on goals and expectations have been discussed in terms of a Recovery

Oversight " gap analysis" at public meetings conducted by the NRC.

Overall, as discussed above, while the licensee NS&O organization continues implementing

initiatives and efforts directed toward program improvements and enhancement, the final

NRC measure of NS&O success will be the effectiveness of its relationship with the line

organization in the unit recovery and proper plant operation thereafter. Progress toward

'his goal will continue to be monitored during future NRC inspections, and will be tracked

as appropriate, with respect to SIL items 37 and 41.

l

.

l

.

64

U3 08 Miscellaneous Operations issues (92700)

0.8.1 Technical Specification (TS) Noncompliance

a. Scope

Several licensee event reports (LERs) recently issued have dealt with TS noncompliance

issues. The inspector reviewed the LERs for root cause and safety significance

determinations, and adequacy of corrective actions. The inspector also verified that the

reporting requirements of 10 CFR 50.73 had been met.

b. Observations and Findinas

(Closed) LER 97-08: (SIL 5 UPDATE):

In accordance with TS 3.6.3 and 3.7.1.5, main steam isolation valves (MSIVs) are required

to be operable during Modes 3 and 4. However, below 350oF, the MSIVs are technically

inoperable because they cannot meet the required closure times. During a plant shutdown

and cool down conducted on April 15,1995, the MSIVs had been shut per operating

procedure. On April 16 adverse condition report (ACR) 01844 was written stating that a -

TS limiting condition for operation should have been entered per TS 3.03. An evaluation of

the ACR at that time determined that the TS had been met since all four MSIVs were

already shut and that the event was not reportable per 10 CFR 50.73. As a result of the

ACR, operating procedures were changed to assure that MSIVs would be shut prior to

going below 350 F. A TS change was submitted to the NRC on June 20,1995, to clarify

TS 3.6.3 and 3.7.1.5. This change has not yet been approved by the NRC.

On reviewing the status of the TS change request submitted in June 1995 the licensee, on

January 16,1997, reviewed the corrective actions for ACR 01844 and determined that

there had been a TS violation and it should have been reported to the NRC per 10 CFR

50.73. Based on this licensee determination, ACR M3-97-0170 was written and LER 97-

08 was submitted to report the April 1995, TS violation. The violation was of low safety

significance since the MSIVs had already been shut and the licensee was proceeding to

cold shutdown for a refueling outage.

The corrective action plan for ACR M3-97-0170 requires the implementation of the new

requested TS 3.7.1.5 and 3.6.3. The inspector verified that the licensee's current

checklist for entering Mode 4 from their current Mode 5 requires that the TS amendment

be approved.

LClosed) LER 50-423/97-06-01: This LER documents that the residual heat removal

suction containment isolation valves had not been maintained closed in mode 4 as required

by TS 4.6.1.1.a. Technical specifications requires that all penetrations not capable of

being closed by containment automatic isolation valves or operator action during periods

when containment isolation valves are under administrative control be secured in their

accident position. The subject valves had been opened in mode 4 in accordance with unit

operating procedures to provide a flow path for plant cool down to cold shutdown as

l

-

.

65

required by plant design. There was no specified administrative controls (operator action)

in place to control these valves. ,

As corrective action, the licensee performed a review to verify that those manual

containment isolation valves that potentially require intermittent operation in modes 1

through 4 were included in TS 4.1.1.a. The review identified four additional valves that

are operated in modes 1 through 4 that were not included in the TS. These valves are

normally opened to allow the performance of surveillance tests. The inspector verified that

the licensee reviewed manual containment isolation valves that require intermittent ;

operation in modes 1 through 4 and developed a proposed TS change request to include l

the valves in TS. I

J

c. Conclusion

The two LERs discuss conditions prohibiter] by TS. Further NRC review of each LER

established that while the licensee's or, err.tional activities were proper evolutions, literal l

compliance with the plant TS had not Leen maintained. Based on the above corrective J

actions and the low safety significance of the issue, these licensee-identified and corrected j

minor violations are being treated as Non-Cited Violations, consistent with Section IV of i

'

- the NRC Enforcement Policv. The listed LERs are closed. SIL ltem No. 5 is partially .

r:losed.

However, the closure of the LERs does not address the generic concern for TS compliance.

A review of LERs issued since April 1996 revealed that there have been a number of LER's l

'

that have dealt with TS compliance problems relating to questionable interpretations This

area is of current interect for further NRC review and is included as an NRC followup

activity, documented as S L ltem 70. )

.

O.8.2 (Closed) LER 50-423/97-09: documents a historical event where a high energy line

break (HELB) door was open in violation of the HELB design criteria. The licensee

considered this event to be the result of their failure to develop and implement an effective

HELB program. This issue was previously discussed in NRC Inspection Report 423/97-01.

No further NRC inspection followup is deemed necessary; consequently, LER 97-09 is

closed,

i

J

U3.Il Maintenance

U3 M1 Conduct of Maintenance

M1.1 General Comments

a. Inspection Scooe (62707)

The inspector observed / reviewed all or portions of the following maintenance activities: i

M3-97-06401, install restricting orifice 3SlH*RO40 i

e M3-97-06471, install freeze seal for isolation of restricting orifice 3SlH*RO40 i

- . _

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

-

___

.

1

[ 66

!

l b. Observations and Findinas l

l )

l The inspector noted that fire watches were established for the welding activities and ;

j personnel were monitoring the freeze seal temperature as required by maintenance 1

procedure MP 3709C. Contingency actions for the potential loss of a freeze seal had been

developed as part of the safety evaluation for the modification. Discussions with the Shift

Manager revealed that control room operators were aware of the required contingency

actions in the event of seal failure. However, the contingency actions did not address all

areas covered in procedure MP 3709C. With respect to the maintenance procedure,

requiring that a Plant Operations Review Committee (PORC) approved contingency plan be

developed and maintained on file in the control room, the licensee opted for less formal, i

verbal contingency planning. A condition report was generated to document this concern.

i

The inspector also reviewed the work activities and verified proper isolation and retest

requirements were identified. Although not specifically listed on work order M3-97-06401

or the ASME section XI repair and replacement plan for the installation of the restricting ;

orifice, the design change request correctly identified the need for a flow balancing test. i

This test is required by Technical Specifications 4.5.2.h following completion of i

modifications to the emergency core cooling system subsystem that alter the subsystem

flow characteristics.

c. Conclusions

l

The inspector concluded that the tagging boundary and retest requirements for the work

activities were adequate. Although a PORC approved contingency plan was not un file in

the control room, the Shift Manager and the control room operators were aware of the

required actions in the event of a freeze seal failure.

The inspectors determined that the maintenance and surveillance activities observed were i

properly performed and a CR was initiated to document the concern regarding contingency

planning.

U3 M2 Maintenance and Material Condition of Facilities and Equipment

M2.1 Maintenance and Confiauration Control of Pipe Suncorts

a. Inspection Scoce (62703,37551)

During the conduct of field inspection-tours, the inspector selected three types of

leedwater system (FWS) pipe support / whip restraint assemblies for detailed review. The

material condition of a sample of the support types, including snubbers, was examined;

component identification markings were noted; and the field configuration of each general

support type was evaluated against its respective design drawing and specification

requirements. The inspector selected welding, bolting, location, attachrnents, and base

plates and anchorages as the criteria for specific review. Where certain drawing

discrepancies were identified, the inspector assessed the licensee's follow-up and

i corrective actions.

=

1

l

.

l

67 l

b Observations and Findinas

The inspection of a support restraint assembly (3-FWS-4-PSR-050) and snubber /suppressor

configurations (3-FWS-PSSP-25 & 26) identified no hardware deficiencies or deviations l

from the original design details, augmented by the applicable Engineering & Design

Coordination Reports (E&DCRs). The inspector examined the Stone & Webster (S&W)

design drawings and spot-cneckeo the fabrication and field installation specifications for

consistency with the as-built assemblies. Acceptable material conditions and a proper field

configuration were noted for these sample support types.

l

On the third type of support, pipe rupture whip restraints (3-FWS-PfiR6 & 7, North & ,

South), the field inspection revealed some apparent design discrep&cies, as follows: )

S&W Specification 2280.000-940 and drawing EV-56E-4 both delineate specific

details for shimming the gap (a clearance of approximately 1/16") between the whip

restraint structural beams and the concrete wall embedments after the conduct of I

hot functional testing, to allow for the unrestrained thermal growth of the main '

structural members. The inspector noted that no shims had been installed on the

beam-to-wall connections. A design change record review revealed that while l

<

E&DCR T-J-03934 addressed changes to the end connection / shim details, it did not

authorize elimination of the shims or modify the specification requirements for

restraint beam-to-wall clearance dimensions.

A subsequent desigr . .odification, installed in 1989, implemented a welded

attachment of nonsafety-related electrical conduit supports to the main structural

support beam of whip restraint 3-FWS-PRR6N. The plant design change record,

PDCR MP3-89-002, for this modification documents in the 10CFR50.59 Safety

Evaluation that conduit and support installations will be performed in accordance

with S&W Specification SP-EE-076; also indicating that no seismic ll over i

condition is created by this design change. However, no evidence of engineering

approval for the conduit support attachment, as required by SP-EE-076, was

retrievable. Furthermore, field inspection revealed safety-related components in

proximity to the nonsafety conduits, indicating that a seismic 11 over i evaluation

should have been performed.

Subsequent tn the identification of these discrepancies, the licensee initiated condiden

reports, CR M3-97-1272 for the shim concern and CR M3-97-M61 for the attached

conduit supports. Immediate corrective actions included a review cf pipe whip restraint

calculations and the conduit support details. Th,s licensee concluded that component

operability had not been adversely affected because the subsequent design reviewr

demonstrated that the shims were not functionally required and the subject conduit was, in

retrospect, seismically supported. Additionally, licensee engineers conducted some further

field inspection of another group of pipe whip restraints, as design detailed or; S&W

drawing EV-46K. Based upon a sample of specifi: inspection criteria, no further design

discrepancies were found.

J

Based upon this expanded inspection sample of insaned pipe whip restraints, additional l

pipe support inspections implemented in accordance with engineering instruction 3DE-97-

_ - - -

.

'

68

003, ongoing configuration management re-evaluations, and prior inspections performed on

pipe support assemblies potentially affected by design temperature considerations, the i

licensee concluded that the discrepancies identified during this NRC inspection were not

l

representative of generic problems with the as-built pipe supports. I

!

c. Conclusion l

NRC field inspection identified one specific series of pipe whip restraints with the as-built )

configuration deviating from the existing design details with respect to certain criteria. The

'

noted discrepancies were provided further engineering evaluation by the licensee and

determined to not result in adverse impact upon the affected component functionality. l

Additional i; cense assessments have concluded that the identified discrepancies are not '

reflective of a more general problem. However, given that the identified deviations l

involved elements of questionable control of both a E&DCR and a PDCR, the licensee's l

handling of design change documents, at least on this one whip restraint assembly,

represents a concern. Pending both the final closura of CRs M3-97-1272 and M3-97-

1461, along with presentation of further evidence by the licensee that a causal problem

linkage does not exist in the processing of other pipe support design changes, this issue

'

remains unresolved. (URI 423/97-02-14)

U3 M7 Quality Assurance in Maintenance Activities

M 7.1 Maintenance Corrective Action /ASME Code Compliance

a. Inspection Scope (62703,40500)

The inspector reviewed the implementation of corrective maintenance on service water

system relief valve 3SWP*RV96A, as documented in nonconformance report NCR 397-

057. Inspection followup included a field inspection for material condition, ASME code I

compliance, and generic corrective actions, The inspector also assessed the licensee's !

controls in handling the identification of a concern involving the procurement of service i

water system (SWP) piping fittings and components. Discussions with cognizant licensee

'

maintenance and procurement personnel revealed questions regarding the issuance of parts

certified to the American Society of Testing and Materials (ASTM) standards for installation

in the SWP (i.e., an ASME Section Ill) system.

b. Observations and Findinas

During maintenance on relief valve 3SWP*RV96A, interior pitting was identified on the

valve bonnet and documented on NCR 397-057. Ultrasonic testing (UT) was performed by

the licensee to determine that the minimum valve wall thickness was acceptable and that

no repair was required. However, the NCR incorrectly identified both the pitting and

subsequent UT to apply to the valve body, vice the bonnet. A review of the technical data

sheet for this valve indicated that the body consisted of a copper-nickel baso material,

l while the bonnet was an aluminum-bronze casting.

,

! The inspector questioned this inconsistency since the American National Standard, ANSI

i B16.34-1981, documented in the "use-as-is" disposition of the NCR was found to not have

l

1

l

' l

!

69

endorsed the material type (SB148-954) used in the fabrication of the valve bonnet. While j

the minimum allowable wall requirements delineated in ANSI B16.34-1981, as referenced l

by later ASME Section 111 code editions, were technically adequate, the design specification !

(2472.110-186) for this relief valve referenced earlier ASME Section ill editions, and a I

different standard, i.e., ANSI B16.5. Further, during field inspection of maintenanco work

on the subject relief valve, the inspector noted a missing lock wire on a similar relief valve

(3SWP*RV96B) in the opposite train. Such a condition is contrary to the ASME Section 111

requirements for valve sealing and setpoint adjustments. ,

l

Upon raising these concerns to the responsible licensee personrel, the inspector was

provided a copy of condition report CR M3-97-1006, which had already documented

missing or broken lockwire conditions on several SWP relief valves, including

3SWP*RV968. With regard to the NCR 397-057 discrepancies, the licensee issued a new

NCR 397-095, documenting a new disposition to acceptance of the pitted valve bonnet in

accordance with applicable requirements of ANSI B16.5, with reference to the 1971

edition of the ASME section ill code and an applicable Code Case 1288. These documents l

were verified by the inspector to establish the governing endorsements for use of SB148 l

aluminum-bronze cast materialin the relief valve bonnet construction.

~ln a related question on ASME Code Case usage, the inspector was informed that small ,

piping parts and fitting material had been procured to ASTM requirements from a non-

ASME supplier and issued for installation in the SWP system. Such material usage was )

believed to be authorized by the intent of ASME Code Case N483; subsequently

determined in review by the NRC inspector and licensee procurement engineers to l

represent a code case that has not been endorsed by USNRC regulatory guidance. i

Upon discovery that the affected SWP items had been ordered, received, and issued to the

field with questionable certification to the intended ASME applications, the licensee

removed the subject material from inventory and placed it in a " hold" status. CR M3-97-

1089 was issued to document this problem and follow up and track the potential

nonconforming condition of improperly certified material being issued and installed in the

safety-related sections of the SWP system. The inspector reviewed CR M3-97-1089 and

discussed continued evaluation of this problem with cognizant licensee engineering

personnel.

c. Conclusions

The inspector identified a field concern and a discrepancy in the NCR disposition to a

questionable ASME relief valve material condition. Licensee corrective maintenance was

already in progress and a new NCR was issued to clarify the governing ASME/ ANSI

requirements for acceptance of the installed valve. The inspector noted that an existing

CR, M3-97-1006, had been issued to correct the missing relief valve lock wires. Also, the

licensee issued another CR, M3 97-1089, to follow up and correct any unauthorized non-

ASME material issued for usage in the SWP system. The inspector determined that

licensee corrective actions to address these issues have been appropriate and intends to

review the final resolution of CRs M3-97-1006 & 1089, tracking their closure as an

inspector followup item. (IFl 423/97-02-15)

L_

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

o

.

70

U3 M8 Miscellaneous Maintenance lasues

M8.1 (Closed) LER 97-002-00: Torquing of Battery Connections Not Performed

a. Insoection Scope (92903)

The inspectors reviewed the licensee findings and corrective actions taken regarding the

failure to check the tightness of the battery connections during surveillance testing.

b. Observations and Findinos

Technical specification surveillance requirement 4.8.2.1.c.2 requires that the licensee

verify that "The cell-to-cell and terminal connections are clean, tight, and coated with .

'

anticorrosion material." The licensee previously had considered the resistance check of the

cell-to-cell and terminal connections to be adequate to ensure connection tightness. A

recent review by the licensee identified that the technical specification bases reference

IEEE Standard 450-1980, " Recommended Practice for Maintenance, Testing, and

Replacement of Large Lead Storage Batteries for Generating Stations and Substations."

This standard recommends a check of the tightness of connections of bolted connections

to the torque requirements of the battery manufacturer. A check of connection torque- -

values was not being performed as part of the battery surveillance test program.

Although the torque values were not being checked in the past, resistance checks and I

battery discharge tests had been performed as required, with satisfactory results.

The licensee revised surveillance test SP3712NA, " Battery Surveillance Testing," to include

the check of bolted connection tightness and performed the procedure to ensure the

correct torquing.

The inspector confirmed that the procedure had been revised as stated in the LER.

c. Conclusions

The inspector concluded that the licensee had appropriately addressed this issue. This

licensee-identified and corrected minor violation is being treated as a Non-Cited Violation,

consistent with Section IV of the NRC Enforcement Poliev. LER 97-002-00 is closed.

M8.2 (Closed) LER 97-005-00: Circuit Breaker Testing With Gages Not in Measuring and

Test Equipment (MT&E) Program

a. Inspection Scoce (92903)

The inspectors reviewed the licensee findings and corrective actions taken regarding the

use of tools that were not controlled within the MT&E Program.

--

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

.

O

71

b. Observations and Findinas

in January 1997 the licensee identified that gages used during circuit breaker maintenance

to check dimensions of clearances were not controlled within the site MT&E Program. One

of the gages used to perform maintenance on the 4160 Volt circuit breakers was found to

be out of tolerance. The affected circuit breakers were declared inoperable until the

preventive maintenance (PM) procedures were performed using a qualified gage. When the

preventive maintenance was performed, all measurements were found to be acceptable.

l

'

Additional actions included a review of other tools and procedures to determine if similar

I conditions existed. Two other tools were thus identified; but when checked, were found

l to be with tolerance.

c. Conclusions l

The inspector concluded that the licensee had appropriately addressed this issue. LER 97-

005-00 is closed,

i

M8.3 LQlosed) ACR M3-96-0557 (Partial - SIL ltem 58); Safety injection System l

Hydrosatic Test l

(Closed) LER 96-032-00 (Partial - SIL ltem 81); !

(Undate) eel 96-201-33 (Partial - Sil item 58);

a. Insoection Scope (92903) l

l

The inspectors reviewed the licensee findings and corrective actions taken regarding the

adequacy of the high pressure safety injection system hydrostatic test.

l

b. Observations and Findinas

Adverse Condition Report (ACR) M3-96-0557 and Licensee Event Report (LER) 96-32-00

address an issue where high pressure safety injection piping was not tested in accordance

with ASME code requirements. Following a change in the setpoints for thermal relief

valves and an upgrade in the system design pressure a hydrostatic test was performed at a

l pressure of 1.1 times the relief setpoint. The correct test pressure should have been 1.25

tirr.es the relief setpoint.

l The licensee subsequently reperformed the hydrostatic test at the correct pressure and

provided trainina for the individuals responsible for performing hydrostatic testing.

1

'

c. Conclusions

The inspector reviewed the work orders and associated documentation for the performance

of the testing at the correct pressure and found that the licensee had properly tested the

system. ACR M3-96-0557 and LER 96-032-00 are considered closed. eel 96-201-33

<

remains open due to ongoing NRC considerations of potential escalated enforcement action

involving this issue.

-

.

l

72

M.8.4 (Closed) LER 50-423/97-12: This LER documented that temporarily installed

concrete blocks, serving as the weighted tornado missile restraints for electrical manhole

covers, had been removed for a period of three minutes during a maintenance activity.

This was done to allow removal and reinstallation of sealant on the covers. Personnel

reviewing and performing the maintenance activity failed to identify the bypass-jumper

requiring the blocks or recognize the safety-related function of the concrete blocks until

they had already commenced the work activity. The inspector determined that the

reported concern represented a minor issue and that LER 97-12 is closed.

U3.Ill Enaineerina

i

U3 E1 Conduct of Engineering

l

i

E.1.1 Deslan Control Weakness

a. Insoection S,. cone l

l

Several licensee event reports (LERs) recently issued have dealt with conditions that are I

outside the design of the plant. The inspector reviewed the LERs for root cause and safety

significance determinations, and adequacy of corrective actions. The inspector also )

verified that the reporting requirements of 10 CFR 50.73 had been met,

b. Observations and Findinas

(Closed) LER 50-423/96-27: documents that tornado restraints for five safety related

electrical manhole covers were not installed. The manholes contained safety-related

cabling which could be damaged by a missile in the event that a manhole cover was lifted

by the forces that could develo,p during a tornado. There was no evidence that hold-down

restraints for the manhole ccvers were ever installed. As corrective action, the licensee

installed concrete blocks of sufficient mass over the manholes to restore tornado

protection until a permanent design change is implemented. The inspector verified that

blocks were placed over the affected manholes, and that work orders and a design change

were generated to install permanent hold-downs for tornado protection. This condition is

scheduled to be resolved prior to unit restart.

(Closed) LER 50-423/96-41: documents that the gap between the tubes and the lower

tube sheet for the service water (SWP) pump strainers was not in accordance with design

requirements. The tube sheets had been replaced in 1988 due to galvanic corrosion

problems. The tube sheets and filter elements had erroneously been considered to be

nonsafety-related and consequently a nonsafety control process for the manufacture of the

internals was used. This issue was discussed in NRC Inspection Report 423/96-09. The

inspector verified that a temporary modification was performed for each SWP strainer to

restore the designed diametral clearances; and that an evaluation was performed that

concluded the SWP strainer was a safety-related component. A permanent design change

to restore the tube sheets to conform with design requirements is scheduled to be

performed prior to unit restart.

-- l

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

,

-

l

I

=

73

c. Conclusio2

l The two LERs discuss conditions where installed equipment or actual plant configuration l

!

differed from the Final Safety Analysis Report (FSAR) descriptions. These errors did not I

directly impact the safe operation of the plant. The specified design concerns have either I

been corrected, or are scheduled to be fixed prior to plant startup. The listed LERs are

closed. However, the closure of the LERs does not address the effectiveness of the design 1

control process at Millstone Station. This area is under current NRC review and is included

as an ICAVP followup activity that is documented as SIL ltem 79.

1

E1.2 inservice insoection (ISI) Proaram Review l

l

a. Inspection Scope (73753)

According to 10 CFR 50.55.a(g)4(ii), Millstone Unit 3 (the licensee) is committed to i

develop an ISI program to the 1980 Edition through the winter 1981 Addenda of Section i

XI. The licensee has upgraded this program to the 1983 Edition, including the summer l

1983 and 1985 Addenda as permitted by 10 CFR 50.55a(g)(4)(iv). !

,

iThe purpose of this inspection was to determine whether the licensee ISI of Class 1,2,

and 3 pressure-retaining components was performed in accordance with the requirements i

of ASME Boiler and Pressure Vessel (B&PV) Code,Section XI,1983 Edition, including the !

1985 Addenda,

b. Observation and Findinas

b.1 ISI Plans and Schedule

The licensee is adjusting the ISI schedule due to the extended outage, as follows: the

First Ten Year Interval began April 23,1986, and the licensee's current schedule for the

completion is at the end of refueling outage six (RFO 6), projected as December 1998.

The ISI plan shows 48 examinations which are required to be completed during RFO 6 to

closecut the first interval. This number may change as the licensee continues with their

self assessment and interval close out review.

The extension from April 23,1996, to December 23,1998, is allowed by the following: a

year extension per ASME Code IWA-2430 to April 23,1997, and the additional extension

for two long shutdowns. The first was 7 months long in 1991 and the current shutdown )

l (approximately 20 months) will bring the license to an expected closeout of December 23,

1998. l

l

An anticipated start of the Second Interval and Program Plan upgrade to the latest code is l

l April 23,1998. This program plan upgrade is currently under way with the anticipated

L submittal by October 23,1997 (six months prior to implementation as required by 10 CFR ,

I 50,55 a). This would mean the Second Interval ends July 23, 2008 (April 23, 2006, plus l

27 months of extended shutdown).

.

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

<

l

!

'

I

74

c.1 Conclusion

The Millstone Unit 3 ISI schedule was adjusted in accordance with the extensions allowed

by the ASME Code,Section XI.

l b.2 Proaram implementation

The inspector reviewed the licensee's ISI Program Manual, which established the

requirements for ASME Code Classes 1,2, and 3 and determined that the Class 1

requirements system boundary was developed in accordance with the requirements for

reactor coolant pressure boundary as defined in 10 CFR 50.2 and the Millstone Unit 3

FSAR. Class 2 and 3 requirements system boundaries were developed in accordance with

Regulatory Guide 1.26 and Millstone Unit 3 FSAR. 1

1

b.2.1 ASME Class 1 Components

As a sample inspection for Class 1 components, the inspector reviewed the Reactor

Pressure Vessel (RPV) ISI to ensure compliance with ASME Code requirements. The

licensee recognized there were cases where component configuration and/or interference 1

. prevented 100% coverage of welds as required by the code (volumetric or surface - s

examinations). In cases like welds 4,6,7, and 8 of the RPV where these limitations ,

existed, the licensee had submitted Relief Request 1 (RR-1), Revisions 1-3 to the NRC, i

documented as follows: I

The lower shell to-lower head weld #4. This weld is 100% accessible from the lower head

side and, in both circumferential directions, is obstructed by'six core lugs from the lower

shell side of the weld where coverage is limited to only 70% of the required weld volume. !

The upper shelllongitudinal welds #6, #7, and #8 examinations were limited to only 37%

of weld #6 and to only 47% of welds #7 and #8 due to nozzle geometry. The licensee

documented that these examinations will be performed to the maximum extent practical at

the end of the first Ten-Year Interval. !

Based on the coverage of the examined welds described above, the NRC granted a relief to

RR-1, Rev. 3, in letter A10880, dated March 3,1993. To document the present RPV ISI

inspection results, the licensee prepared RR-1 Revision 4 documenting the non-destructive

examination (NDE) results that exceeded the previous coverage documented in RR-1, Rev.

3. These results are still below the 90% coverage required by 10 CFR 50.55a(g)(6)(ii)(A).

Therefore, the licensee plans to submit, in addition to the RR-1, Rev. 4, an alternative for

the augmented examination for each weld where 90% coverage was not achieved,

c.2.1 Conclusion

Although the licensee's latest RPV NDE results of welds 4,6,7 and 8 show better

coverage of the examined welds than the previous NDE,90% coverage was not yet

achieved. Therefore, the licensee plans to convey these latest results as Revision 4 of RR-

1, along with an alternative for augmented examination for each of the welds with less

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

l .

.

75

than 90% examined length, to the NRC Office of Nuclear Reactor Regulation (NRR) for

their review.

b.2.2 ASME Class 2 Components

NRC letter of March 3,1993, documenting the review of Millstone Unit 3 first ten-year

interval ISI indicated that the licensee's program was upgraded in Revision 3 to meet the

'

requirements of ASME B&PV Code,Section XI,1983 Edition, including the 1985 Addenda

for Class 2, Examination Category C.F-1 and C-F-2 piping welds. However, the NRR

review of Revision 3 to RR-1 noted that the chemical and volume control system

containing 129 welds, and the intermediate pressure safety injection systeni containing 96

welds have been completely excluded from examination.

The inspector discussed the licensee's actions to address the NRR concerns as follows:

The requirements for selecting ASME Section XI Class 2 pipe welds to the ASME B&PV

Code,Section XI,1983 Edition, including the 1985 Addenda, is based on pipe size and

wall thickness. For the system of high pressure injection, the piping has to be greater than .

4" in diameter and greater than 0.375" in we" thickness to require any ISI NDE. The 129 l

welds in the chemical volume control system and the 96 welds in the intermediate safety -

injection system are located on 6" and 8" diameter, schedule 40 piping (pump suction

lines); this wall thickness is less than 0.375" and, therefore, no NDE is required.- However,

the code requires this size pipe to be included in the total Class 2 population as a 7.5%

sample of the weld population. Therefore, the licensee plans to perform volumetric

examinations for the 7.5% of the 225 weld population (seventeen) during the next

refueling outage (RF06), currently scheduled to commence in October 1998,

c.2.2 Conclusion

The inspector determined that the licensee actions, including the plans concerning

inspection of the previously excluded 129 piping welds in the chemical volume control i

system and the 96 piping welds in the intermediate safety injection system, were l

acceptable. The inspector noted that the licensee's Class 2 piping ISI observations, made

by NRR, were addressed by the licensee in accordance with the ASME B&PV Code,

Section XI,1983 Edition, including the 1985 Addenda.

b.2.3 ASME Class 3 Components

To assess the adequacy of the ISI performed for the ASME Class 3 components, the

inspector performed a system walkdown of segments of the component cooling and

service water systems pipe supports.

c.2.3 Conclusion

The inspector identified extensive external corrosion on the channel head bolting flange and

l cover of the Component Cooling Water System heat exchanger Nos. 3CCP"E1 A and

3CCP*E18.

i

4

, .

l

.

76 \

The licensee created a trouble report to have maintenance personnel clean and preserve

the affected bolting. No other significant deficiencies were identified.

U3 E2 Eng!neering Support of Facilities and Equipment

E2.1 Steam Generator Tube Rupture (SGTR) Analysis

a. Insoection Sco_pe (375511

As part of the 10 CFR 50.54f review of the design and licensing basis of the plant, the

licensee identified that one main steam pressure relievirig bypass valve (MSPRBV) out of

service would invalidate the design basis for the SGTR overfill analysis. The MSPRBV's are

used to cool down the reactor coolant system (RCS), on a loss of offsite power (LOOP),

and to maintain adequate margin to prevent overfill of the ruptured steam generator (SG). I

Condition report (CR) M3-97-0835 was written to document this condition. The inspector l

reviewed the Westinghouse WCAP-13002, " Margin to Overfill Analysis for a SGTR for l

Millstone Unit 3 Four Loop Operation," as followup to the design concern documented in

the CR, at well as to further evaluate the engineering aspects of the operational issue

discussed in section U3.01.2 of this inspection report,

b. Observations and Findinas

WCAP-13002 assumes that cool down of the RCS, during a SGTR event and LOOP,is

accomplished with two SGs via the MSPRBVs. Two SGs are unavailable due to both the

ruptured SG and an assumed singie failure of one MSPRBV. This assumed single failure is

consistent with the generic guidance documented in Westinghouse WCAP-10698, "SGTR

Analysis Methodology to Determhe the Margin to Steam Generator Overfill."

WCAP-10698 analysis methodology included the selection of a reference plant (three loop

plant) that has the potential for th; least margin to overfill. The worst case single failure

for this plant was determined to be the loss of one MSPRBV, since this decreased the

available steam dump capacity (for plant cool down) by 50 percent for the LOOP case.

The loss of an emergency diesel generator (EDG) was not considered the most limiting

case because, although power is lost to a MSPRBV, the rate of filling of the ruptured SG is

also reduced by the loss of components that automatically start to fill the RCS.

The inspector questioned whether the most limiting failure for the SGTR analysis, for a four

loop plant, was the loss of an EDG. A loss of the EDG could result in the inability to

operate two MSPRBVs. This would result in a decrease in the available steam dump

capacity by 67 percent. There was no discussion in WCAP-13002 addressing the loss of

an EDG.

l

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

.

l

.

77

l

c. Conclusion

i

The inspector discussed the concern regarding the most limiting failure for the SGTR

margin to overfill analysis with cognizant licensee personnel. The licensee plans to perform

a specific Unit 3 SGTR analysis to further address questions of this nature. This matter

l will be reviewed further by the NRC as an inspector followup item. (IFl 423/97-02-16)

l

.

U3 E8 Miscellaneous Engineering issues

!

E8.1 (Closed) SIL ltem 30: Auxiliary Feedwater Check (AFW) Valve Leakage

References: ACR M3-96-0855 Auxiliary Feedwater Valve Operability

NU Letter B15397, Dated November 1,1995

a. Inspection Scone (92903)

The inspector reviewed the licensee actions taken and actions planned to resolve a problem

with backleakage through the AFW system check valves.

b. Observations and Findinas

Check valve leakage in the auxiliary feedwater lines has resulted in elevated piping and

containment penetration temperatures. Prolonged operation at elevated temperatures could

result in damage to the concrete adjacent to the penetrations. To alleviate this concern the

operators monitored the penetration temperatures and, when necessary, operated an AFW

pump to cool the penetrations by pumping the relatively cool water from the demineralized l

water storage tank to the steam generators. During this evolution the turbine driven AFW l

pump was isolated resulting in a significant increase in the unavailability time of this pump. !

I

Engineering evaluations had been performed to assess the effects of the elevated

temperatures on the penetration concrete temperature and the pipe supports. These

evaluations concluded that operations with temperatures as high as 300 degrees for short

periods would not be detrimental to the operation of the piping or the concrete. However,

those evaluations did not specifically document an assessment of the effects of the

elevated temperatures on valves 3FWA*MOV35D and 3FWA*HV36D. ACR M3-96-855

was issued and documented an evaluation that concluded that the valve performance

would not be impacted by elevated temperatures.

In 1995 the licensee replaced check valve 3FWA*V47, located outside of the primary l

containment, in an effort to reduce the backleakage through the "D" AFW penetration. l

The licensee also planned to replace the two check valves in the same line that are located

'

inside of the primary containment during the sixth refueling outage. During that time

manual isolation valves were to be installed to facilitate future repairs. After entering the

current unplanned shutdown the licensee considered performing this work prior to restart,

but then decided to do the work in the sixth refueling outage as originally pla 1ed. l

l

f

l

.-.

.

78

The licensee decision was based on the following factors:

Surveillance procedure SP 3622.3, " Auxiliary Feedwater Pump 3FWA*P2

Operational Readiness Test," was revised to shut a manual discharge valve to

prevent the check valves from being lifted off of their seat during pump tests.

Experience has shown that once the check valves seat, they remain seated.

An operating procedure is in place that includes instructions on reseating the check

valves.

The AFW piping temperatures at the containment penetrations are checked and

recorded each shift as part of the operator rounds.

Procedures are in place to cool the piping if the temperature exceeds 150 F.

The piping has been analyzed to ensure elevated temperatures would not adversely

affect system operability.

Following the replacement of check valve 3FWA*V47 and the change to the

surveillance procedure there were no entries in the shift manger logs to indicate any

problem with leakage for the last eight months of power operations, prior to the unit

shutdown.

The replacement of the two check valves inside containment is still planned for

refueling outage (RFO) 6.

In addition to these considerations the inspector noted that the two check valves l

scheduled to be replaced had been inspected during the la.st refueling outage and no l

significant deficiencies were identified.

c. Conclusions j

The inspector reviewed portions of the associated documentation, drawings and

procedures and discussed the issue with the design and system engineers. The inspector

noted that the surveillance procedure change to shut a manual isolation valve during

testing did not affect the previously established testing flow path and did not involve a

significant additional burden on the operators. Also, the most recent operating experience

indicated that the need to frequently cool the penetrations had been eliminated, minimizing

the potential for the condition requiring significant operator attention . The inspector

concluded that the licensee had provided adequate bases for implementing further

corrective actions, as previously planned, during RFO 6. Sllitem 30 is closed.

E8.2 (Closed) SIL ltem 62: ACR 13788 - TSP Basket Modification Safety Evaluation

a. Insoection Scope (92903)

The inspector reviewed the licensee actions taken to resolve the issues documented in

ACR 13788.

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

.

?-

D

.

79

b. Observations and Findinas

'

Unresolved item Report (UIR) 107 raised several questions associated with the trisodium

phosphate (TSP) basket modification that was performed by Plant Design Change Request

(PDCR)94-135. A question as to the validity of the radiological safety evaluation in the

i PDCR was raised because the licensee safety evaluation was originally written to support

two changes; the addition of the TSP baskets and an increase in the allowable containment

leak rate. The licensee's safety evaluation stated that "This safety evaluation is not valid if

only one of the two changes is approved." The NRC reviewed and approved only the

addition of the TSP baskets in License Amendment No.115, and stated in the associated

NRC safety evaluation that the proposed increase in the containment leak rate would be

considered separately. NRC evaluation and approval of the TSP basket change was based

on the existing assumed containment leakage rate.

The licensee subsequently reviewed the effects on the safety evaluation based on only

having the one change approved. This review determined that the safety evaluation

i bounded the technical specification status as they existed for plant startup from the 1995

.

refueling outage and the safety evaluation was revised to reflect this conclusion.

Several FSAR discrepancies that were identified were resolved via FSAR change requests . !

that had been approved by the unit director.

c. Conclusions

The inspector reviewed the associated documentation and discussed the issue with the

responsible engineer and concluded that the licensee had appropriately addressed the

issues documented in the subject ACR. SIL ltem 62 is closed.

E8.3 (Closed) LER 96-047-00 : Seismic Qualification of 4 Kv Circuit Breakers

a. Insoection Scope (92903)

.

The inspector reviewed the licensee findings and corrective actions taken regarding a

potential seismic concern that the 4160 Volt General Electric (GE) circuit breakers may not

have been adequately restrained when they were not in the fully engaged position.

b. Observations and Findinas

,

The 4160 Volt switchgear was seismically qualified with the circuit breakers fully engaged

in the operating position. During maintenance or testing activities the circuit breakers may

be in the racked-down position, and would be free to move as a result of a seismic event.

This could impact the switchgear cubicle structure. Such an impact could result in the

inadvertent actuation of protective relays mounted on the cubicle doors. Inadvertent

actuation could result in the loss of power to safety equipment. Neither the seismic

qualification report or operating and maintenance manuals discussed this condition and the

plant staff had previously failed to recognize this potential.

__

- - . -- - - . . . . - - - .

.

O

80

- The licensee removed all safety related circuit breakers that were racked-down from their

associated switchgear cubicles while long term solutions were evaluated. Also, operating

procedure OP 3370A was revised to ensure that when the breakers are racked down that

i the breakers are restrained or removed from the switchgear cubicles. The method of

l restraining the breakers is by verifying that the elevator handle was forward and latched in

place. This method was based on discussions with GE. Additional reviews to confirm if

any additional actions are required are to be completed prior to restart of the plant.

The inspector reviewed procedure OP 3370A and inspected a breaker in the racked-down

position and found it to be in accordance with the procedure. A plant operator interviewed

at the switchgear was familiar with the requirement and the bases for the requirement for

positioning the elevator handle.

c. Conclusions

- ,

!

The inspector concluded that the immediate corrective actions appropriately addressed the

issue and that additional reviews are being tracked and scheduled for resolution prior to

plant startup. LER 96-047-00 is closed.

E8.4 (Closed) LER 97-001-00 125 Volt Battery and Charger Surveillance Testing i

a. Inspection Scope (92903) l

l

The inspector reviewed the licensee findings and corrective actions taken to resolve issues '

associated with the battery and battery charger surveillance testing.

b. Observations and Findinos

in January 1997, the licensee identified two examples where surveillance testing was not

being performed in verbatim compliance with the technical specifications. In one case

surveillance requirement 4.8.2.1.b.3 requires a verification that "The average electrolyte

temperature of six connected cells is above 60 F." in this case, the licensee was

obtaining the temperatures of all 60 cells and verifying the average temperature was above

60 F.

In the other case, surveillance requirement 4.8.2.1.c.4 requires a verification that "Each

battery charger will supply at least the amperage indicated in Table 4.8-2b at 125 Volts for

at least 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months ." The testing was performed with the battery charger voltage at the

normal operating voltage of approximately 129 Volts.

Although the testing methods appeared to be adequate to ensure equipment operability,

the licensee revised the affected test procedures to ensure verbatim compliance with the

technical specifications. The revised tests were performed with satisfactory results.

c. Conclusions

i

The inspector reviewed the associated surveillance requirement bases and the most recent

surveillance tests and found the licensee actions to be appropriate. These failures

.

81

constitute violations of minor significance and are being treated as Non-Cited Violations,

consistent with Section IV of the NRC Enforcement Policv. Therefore, LER 97-001-00 is

closed.

E8.5 (Ocen) SIL ltem 16: ACR 1935 - Dual Function Valve Testing

(Onen) Unresolved item 96-08-18: Inadequate Inservice Test (IST) Program

Controls

a. Inspection Scone (92903)

The inspector reviewed the licensee evaluation of the potential for Unit 3 to experience

problems similar to Unit 2 relative to containment isolation valves that have a function to

close at a system operating pressure that is higher than that at which the valves are tesud

for containment isolation purposes,

b. Observations and Findinas l

This issue was initially identified in 1993 by Unit 2 personnel when air operated valves in

.the Unit 2 letdown line did not fully shut against reactor coolant system pressure when

attempting to stop flow to perform a valve repair. The licensee found that the cause was )

that the spring preload on the air operators had not been properly set during maintenance. l

The immediate concern was resolved by adjusting the spring preloads and verifying closure l

of the affected valves by performing a seat leakage test at normal operating pressure. Unit j

2 engineers also committed to define retest requirements to verify isolation capabilities for

dual function valves at full system pressure.

Unit 3 personnel reviewed this issue and concluded that it was not applicable to Unit 3 for

the following reasons:

The Unit 3 actuator specifications included a valve specific data sheet which clearly j

identifies the maximum shutoff and operating pressure. i

The actuator setup was specified by the vendor for the maximum shutoff pressure

and the actuator was provided with a valve data nameplate which specified the air

pressure settings that correlated to the required actuator spring preload.

The Unit 3 actuator procedures record the spring setting, include a step to re-

establish the as found setting and to test the actuator pressure was in accordance

with the nameplate data.

During the review of the valve testing performed on the various valves, the inspector

reviewed the design bases requirements of the isolation valves in the letdown line for Unit

l 3. The letdown line flow path is from the reactor coolant system (RCS) loop piping

l through two letdown isolation valves (3-RCS*459 and 460), through the regenerative heat

i exchanger. After the heat exchanger, the piping splits into three parallel flow paths, each

with an orifice and a downstream orifice isolation valve (3-CHS* AV8149A, B, and C).

From the outlet of the orifice isolation valves, the piping connects to a common line which

l

.

!

.. . .- - - - . . - .. --

. >

l

. ,

,

'

l'

82 l

l

then exits the containment through the inboard and outboard containment isolation valves

(3-CHS*CV8160 and 8152).

1

-The inspector noted that Section 15.6.2 of the Final Safety Analysis Report (FSAR)

evaluates the effects of a failure of a smallline carrying primary coolant outside

containment. The analyses indicates that the most severe pipe rupture would be a

complete severance of the letdown line outside of containrrent. The FSAR states that the

operator would isolate the letdown line rupture by closing the letdown orifice isolation

! valves (3-CHS* AV8149A, B, and C) followed by the pressurizer low level isolation valves

(3-RCS*459 and 460). Ahernatively, the operator would close the containment isolation

i valves (3-CHS*CV8160 and 8152) to isolate the rupture,

The inspectors found tnat the licensee was in the process of adding the 3-

CHS* AV8149A,B,C valves into the IST program. However, the 3-RCS*459 and 460 l

valves weia not included in the IST program, nor had they been addressed by a recent

licenseo review of the IST program. The licensee reviewed this question and found that j

l

various design documents had conflicting information and also that all of the valves inside '

of containment were powered from the same train of electrical power. Condition report

M3-97-0866 was issued by the licensee on March 21,1997 to document these questions. l

l

These issues are similar to those addressed by NRC unresolved item 50-423/96-08-18

which was issued to track licensee actions concerning inadequate IST Program controls

and testing. License actions to address these additional CR M3-97-0866 questions will be

reviewed as part of the unresolved item review.

c. Conclusions

NRC review of the adequacy of the dual function valve testing is continuing and SIL ltem

16 remains open. NRC unresolved item 50-423/96-08-18 remains open pending NRC

review of the IST program issue resolution.

E8.6 (Closed) URI 96-201-17: Auxiliary Feedwater (AFW) Pump Lubrication Schedule

(Closed SIL ltem 18 - Partially Closed)

a. Inspection Scope (92903)

The inspector reviewed the licensee findings and corrective actions taken regardinh the

adequacy of the AFW pump bearing lubrication.

b, Observations and Findinas l

l

In April 1996, an NRC inspection team questioned the bases for caution statements in the

AFW surveillance tests that required the pump bearings to be manually prelubricated prior

to a pump start if the pump had not been operated in the previous 40 days. Since the

surveillance test only operates the pumps quarterly, the pump could receive an automatic

j start signal, after a period greater than 40 days without the procedurally mandated, manual

prelubrication.

1

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

.

l

..

83

The licensee reviewed this concern with the pump vendor and concluded that the pumps

could be started without prelubrication after an idle period of up to 113 days. This

conclusion was based on the light load on the bearings, experience with similar pumps,

and adequate residual oil on the bearings to provide lubrication for the initial one or two

i seconds until the shaft driven luba oil pumps provide considerable quantities of oil to the

bearings.~

The operating and surveillance procedures and the pump vendor manual have been revised

to reflect this evaluation.

c. Conclusions

The inspector concluded that the licensee had appropriately addressed this issue.

Unresolved item 50-423/96-201-17 (Partial of SIL ltem 18) is closed.

'

!

l

_ __ _ . _ _ .

.

l

-

84

!

IV Plant Support

(Common to Unit 1, Unit 2, and Unit 3)

R1 Radiological Protection and Chemistry Controls

R 1.1 Radioloaical Protection Proaram

a. Insoection Scope (83750)

The inspector reviewed the circumstances surrounding multiple examples of workers

entering the RCA without proper dosimetry. These instances occurred from March through

April 1997, and included one instance which occurred during the period of the specialist

inspection.

b. Observations and Findinas

Unit 1

. On March 1,1997, Unit 1 identified a new person to serve as Radiation Protection -

Manager (RPM). The inspector reviewed the designated individuals training and

qualifications against the requirements set forth in Unit Technical Specification 6.3.1, and

determined that the designated RPM met the qualification requirements.

On April 13,1997, a work group entered the liquid radwaste facility to perform work in the

"A" and "B" Concentrated Waste Tank cubicle. This area is a posted Locked High

Radiation Area (LHRA). One of the workers failed to obtain an electronic dosimeter as

required by plant procedures as required by unit Technical Specification 6.11 for entry into

an LHRA. After working in the LHRA for a period of time, the worker realized he was not

wearing electronic dosimetry, and exited the area. The workers exposure was calculated

to be 70 millirem, which was added to his official dose of record. Procedure RPM 5.22

requires radiation workers to comply with written instructions, including RWPs, from the

radiation protection staff. Failure to adhere to the licensee's radiation protection program,

specifically procedure RPM 5.22, is a violation of 10 CFR 20.1101. (VIO 245/97-02-17)

Unit 2

Since the last specialist inspection in this area, Unit 2 has experienced three additional

incidents involving workers i no RCA without electronic dosimetry. During the last

specialist inspection (50-336/9L01), three earlier examples of this violation had been

identified. Corrective actions taken by the licensee to address this issue have failed to

prevent a recurrence. Corrective actions taken at the time of this inspection included

reducing the number of access points into the RCA; posting of a radiation protection

technician near the main RCA access point to observe workers entering the RCA; and

l working with the training department on the development of an enhanced radiation worker

training program to include use of a mock-up RCA facility. Failure to adhere to the

l licensee's radiation protection program, specifically procedure RPM 5.22,is a violation of

! 10 CFR 20.1101. (VIO 336/97-02-17) This is a repeat violation.

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

o

.

85

_ Unit 3

i. During this specialist inspection, on April 29,1997, a contractor worker entered the RCA

'

without hi.s thermoluminescent dosimeter (TLD). The worker's TLD was later found

outside the RCA in the turbine building, where he had previously been working. Since the

TLD is utilized to determine dose of record, all workers assigned a TLD must wear it when

in the RCA. Failure to adhere to the licensee's radiation protection program, specifically

procedure RPM 5.22, is a violation of 10 CFR 20.1101. (VIO 423/97-02-17).

Site Health Physics

l Recently the licensee was sent a 10 CFR 21 notification by a vendor regarding the

operability of a condenser R-meter utilized to maintain traceability of the licensee's :

calibration source to the National Institute of Standards and Technology (NIST) primary i

calibration standard. The inspector reviewed the actions taken by the licensee, and

determined that appropriate corrective actions occurred, that survey instrumentation

utilized were properly calibrated, and that no erroneous data or calibrations occurred at the

licensee's f acility,

c. Conclusions

Licensee corrective actions for a previously identified violation involving radiation worker

practices have not been successfully implemented so as to prevent recurrence. Since the

last specialist inspection in this area, concluded on February 7,1997, five additional

examples have been identified, including one in which the worker entered a posted, locked I

high radiation area. 1

R1.2 Radwaste Proaram

l a. Insoection Scope (86750)

l

.

l The inspector reviewed actions taken by the licensee to remediate the conditions found in l

L the Unit 1 liquid radwaste f acility, previously identified by the NRC (NRC Inspection Report !

l 50-245/95-35), together with actions taken to ensure appropriate management attention is

focused on this program area in the future. Additionally, the inspector reviewed the

management focus on the liquid radwaste program at Unit 3.

b. Observations and Findinos

l

l Unit 1

l

The licensee recently completed the removal of six major processing vessels / tanks from

the liquid radwaste f acility. These included two vessels that were still in service in 1995,

although both were also actively leaking spent filter media. During this inspection, the

inspector toured all areas of the liquid radwaste facility; except sealed demineralizer/ filter

,

cubicles (radwaste demineralizers [2] and spent fuel pool filters [2)) and the spent resin

tank. Recent photographs of these areas, except the radwaste domineralizers, were

reviewed by the inspector rather than actual entry to these facilities, due to radiation

l

r

!

__ ,

. - - . . ~ - --. . - -- - - - -- - -

[ .

.

I

l

86 !

I

exposure considerations. Examination of the radwaste demineralizers has not yet been j

l started by_the licensee. l

In general, all previously identified areas of radwaste materials located on floors and other

vertical surfaces have been cleaned up, and in-service processing vessels have been

inspected and determined to be usable, inspection of the "A" and "B" floor drain collector i

tanks was not completed at the time of this inspection, although the inspector did examine

the interior of the "B" collector tank during 1.11s inspection. The inspector reviewed the

actions taken by the Engineering Department to inspect the tanks, pumps and piping in the

facility and determined that the conclusions reached on the continued use of certain

components was reasonable. The inspector was also apprised of the current project

status, its interim completion date (July 15,1997) when the facility would be ready to l

support plant operations, and items that would still need to be addressed after July 15, !

1997, including the purchase and installation of a new sludge tank.

Of particular interest to the inspector was actions taken by the licensee to prevent a

recurrence of the radwaste problems, specifically actions taken to ensure appropriate

management oversight of the systems. Previous unit management focused on the quantity

of radionuclides discharged in the liquid effluent from the unit, and determined that since

mthis number was trending down, the radwaste facility was functioning properly.:

Additionally, the person designated at the unit as responsible for the safe and appropriate

operation of the facility was an operations assistant, a position five levels below the Unit

Director.

The inspector discussed the current management of the unit's radwaste facility with the

Operations Manager assigned oversight responsibility, his supervisor, the Operations

Director, and other staff at the unit involved in the radwaste program. By placing

responsibility for the safe operation of radwaste under an Operations Manager (one of two

in the unit), the system is subject to more management attention and oversight than

previously. In addition, the focus within the unit appears now to be placed on operation of

the radwaste facility in a manner consistent with plant safety, operational reliability,

conformance with plant design documents and minimization of effluent discharge. The

inspector indicated that this would be an area of continued focus, especially when the

radwaste facility was fully operational again.

Unit 3

At Unit 3, while the amount of radioactivity discharged to the environment remains low, no

clear management oversight appears to exist During this inspection, the inspector

interviewed the unit Chemistry Supervisor, under whom the Primary Equipment Operator -

Radwaste, now works, and the Unit Director. Neither could identify a single point of

contact within the unit's management as being respont.ble for liquid radwaste. Since a

lack of this type of management focus has been identified as a principle root cause of the

radwaste problems which occurred in Unit 1, this similar lack of clear management

l oversight at Unit 3 is of concern to the NRC. While no violation is evident, this area

requires more managerial attention,

l

i

,, ., ,_

_. . _= . _ __ _ _ _ _ _ . . _ _

\ O

I

87

c. Conclusions

l Significant physical improvements ;n the Unit 1 liquid radwaste facility have been made,

although work has not yet been completed. The Unit has also placed heightened

,

management attention to this program area, although the effectiveness of this heightened

l focus will have to be evaluated once the physical remediation of the facility is completed.

At Unit 3, there continues to be a lack of clear management oversight for the liquid

radwaste system and program. !

l

V. Manaaement Meetinas

1

i X1 Exit Meeting Summary

The inspectors presented the inspection results to members of licensee management at

separate meetings in each unit at the conclusion of the inspection. The licensee

acknowledged the findings presented.

l

X I .2 Final Safety Analysis Report Review

A recent discovery of a licensee operating their facility in a manner contrary to the updated

final safety analysis report (UFSAR) description highlighted the need for additional

verification that licensees were complying with UFSAR commitments. All reactor

inspections will provide additional attention to UFSAR commitments and their incorporation

into plant practices, procedures and parameters. .

1

l

While performing the inspections which are discussed in this report the inspectors l

reviewed the applicable portions of the UFSAR that related to the areas inspected. l

Inconsistencies were noted between the wording of the UFSAR and the plant practices,

procedures and/or parameters observed by the inspectors, as documented in Sections

U1.E1.3, U2.E2.1, U3.E1.1, U3.E8.2. and RI

X3 Management Meeting Summary

On April 30,1997, the NRC staff participated in a publicly observed meeting with licensee

representatives to discuss the licensee's progress in facilitating the restart of all three

Millstone units. A summary of this meeting, to include licensee slides, was published on

May 12,1997 and has been made available in the NRC Public Document Room.

l

,

I

l

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

.

88

INSPECTION PROCEDURES USED

b

,

IP 37551: Onsite Engineering

f

IP 40500: Licensee Self-Assessments Related to Safety issues inspections

IP 62703: Maintenance Observations

} IP 62707: Maintenance Observations

IP 71707: Plant Operations

f IP 73753 Inservice Inspection

[

t

IP 83750: Occupational Radiation Exposure

IP 86750: Solid Radioactive Waste Management and Transportation of

j Radioactive Materials

IP 92700: Onsite follow-up of Written reports of Nonroutine Events at Power

j

Reactor Facilities

IP 92901: '!Iowup - Operations

,

IP 92902: Followup - Maintenance

,

IP 92903: Followup - Engineering

i

!

.

.. . - . - - . -

o

e

89

ITEMS OPENED, CLOSED, AND DISCUSSED

Opened

URI 50 245/97-02-01 U 1.01.4 Failure to Trend CRs

URI 50-245/97-02-02 U 1.01.4 RP-4 interface with lower

,

'

tier reporting processes

URI 50-245/97-02-03 U1.03.1 NGP 2.25 reportability

l determination and LER

.,

processing

URI 50-245/97-02-04 U1.08.1 Non OA lamp bulb usage

eel 50-245/97-02-05 U 1.E1.1 Inadequate corrective

action

}' eel 50-245/97-02-06 U 1.E 1.1 Appendix J leak testing

VIO 50-245/97-02-07 U1.E1.2 Inadequate procedure

control of CIVs

l EEI 50-245/97-02-08 U1.E1.3 Operation of LPCI system

I

beyond licensing basis

.

eel 50-245/97-02-09 U1.E1.3 Inoperable LPCI heat

exchangers

EEI 50-245/97-02-10 U 1.E1.3 UFSAR discrepancies

eel 50-245/97-02-11 U 1.E8.2 Leakage of CU-29 valve

eel 50-336/97-02-12 U2.E8.1 Inadequate Surveillance

Procedures

VIO 50-336/97-02-13 U 2.E8. 2 Corrective action failure for

single failure vulnerability

URI 50-423/97-02-14 U3.M2.1 Maintenance and

Configuration of pipe

supports i

IFl 50-423/97-02-15 U3.M7.1 Maintenance corrective

action /ASME code

compliance

IFl 50-423/97-02-16 U3.E2.1 Steam generator tube

rupture analysis

VIO 50-245/336/423

97-02-17 R 1.1 Failure to adhere to

radiation protection

program

Closed

VIO 50-336/95-11-01 U2.E8.1

URI 50-336/95-25-03 U2 E8.2

URI 50-336/95-27-01 U2.E8.3

URI 50-336/95-81-01 U2.E8.4

IFl 50-336/95-201-07 U2.E8.5

VIO 50-336/96-01-06 U 2.M8.1

VIO 50-336/96-04-08 U 2.M8.2

l

\

'

o j

!

i

I O

.

-90

.

URI 50-336/96-05-11 U2.E8.6

URI 50-423/96-201-17 U3.E8.6

'

Updated

URI 50-245/96-06-02 U 1.M8.1

eel 50-336/96-08-06 U 2.08.1

f VIO 50-336/96-08-07 U2.M8.3

2

. eel 50-336/96-08-10 U2.M8.4 -

eel 50-336/96-09-10 U2.E8.7

, eel 50-336/96-201-29 U2.E8.4

, URI 50-423/96-01-07 U3.01.2

eel 50-423/96-201-33 U3.M8.3

URI 50-423/96-08-18 U3.E8.5

. The followina LERs were also closed durina this inspection

,

Docket Number 50-245

!

96-12.

1

i Docket Number 50-336

i

! 96-16

96-37

i

Docket Number 50-423

1

96 27

] 96-32

,

'

96-41

j. 96-47

3

97-01

97-02

4

'

97-05

, 97-06-01

97-08

i 97-09

l , 97-12

i

i.

i

l

,

<!

.

f

_ _. - _ _ _ . _ .

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

! e

f 91

LIST OF ACRONYMS USED

ACR(s) adverse condition report (s)

AFW auxiliary feedwater '

AMSAC mitigation system actuation circuitry

ANSI /ANS American National Standards institute /American Nuclear

! ARP Alarm Response Procedure

ASME American Society of Mechanical Engineers

l ASTM American Society of Testing and Materials

l ATWS anticipated transie'nt without scram

AWO(s) automated work order (s)

CEBPS containment and enclosure building purge system

CFR Code of Federal Regulations

CIAS containment isolation actuation signal

CR(s) condition report (s)

DCM design chtnge manual

DCN design change notice '

EBFS enclosuro building filtration actuation system

ECP estimated control rod position

ECT eddy current testing

E&DCR(s) Engineering & Design Coordination Reports

]

EDG emergency diesel generator i

EEI escalated enforcement item

ER Engineering Record

ERT event review team

ESAS engineered safeguards actuation system

ESF engineered safety feature

ESW emergency service water j

EWR engineering work request

'

FIN Fix-It-Now

FSAR Final Safety Analysis Report

FTS Fin Team Supervisor

FWS feedwater system

GL Generic Letter

gpm gallons per minute

HELB high energy line break ,

HP health physics

I&C instrument and control

ICAVP Independent Corrective Action Verification Program

IFl inspector follow item

ILRTis) integrated leak rate test (s)

IP(s) inspection procedure (s)

IPTE(s) Infrequently Performed Test or Evolution (s)

lR(s) Inspection Reports (s)

ISI inservice inspection

IST in service testing

HP health physics

LER(s) licensee event report (s)

..

s' j

9

92

LHRA locked high radiation area

LLRT local leak rate testing ,

LOCA loss of coolant accident

{

LOOP loss of offsite power i

LPCI low pressure coolant injection j

MRT management review team j

MSPRBV main steam pressure relieving bypass valve

NCR(s) nonconformance report (s)

NDE non-destructive examination

NGP(s) nuclear guidance procedure (s)

NIST National Institute of Standards and Technology

NNECO Northeast Nuclear Energy Company

NPSH net positive suction head

NRC Nuclear Regulatory Commission

NRR Nuclear Reactor Regulation

NSAB nuclear safety assessment board

NSAL Nuclear Safety Advisory Letter

NSIC Nuclear Safety Information Center

NS&O nuclear safety and oversight

NUQAP Northeast Utilities Quality Assurance Program

NUREG Nuclear Regulation

NUSCO Northeast Utilities Service Company

NVLAP National Voluntary Laboratory Accreditation Program

OCA Office of Congressional Affairs

OP(s) operating procedure (s)

P&lD piping & instrumentation diagrams

PAO Public Affairs Office 1

PDCR plant design change record 1

PDR Public Document Room I

PMMS production maintenance management system l

PORC plant operation review committee l

PTSCR proposed technical specification change request {

OA quality assurance '

OAS Quality and Assessment Services

RCA radiologically controlled area

RCS reactor coolant system

RI Region I ,

RPV reactor pressure vessel I

R W P(s) radiation work permit (s)

SER(s) safety evaluation report (s)

SFP spent fuel pool

SG steam generator

SGCS safety grade cold shutdown

SGTR steam generator tube rupture

SIL significant items list

SORC site operations review committee

SPO Special Projects Office

SRO senior reactor operator

-

g ,

,

, --7

r6.

W-

93

STA shift technical advisor

SWSOPI(s) service water system operational performance inspection (s)

TDAFW turbine driven auxiliary feedwater

TEMA Tubular Exchanger Manufacturers Association

TER Technical Evaluation Report

TLD(s) thermo-luminescent dosimeter (s)

TOE team qualified expert

TS(s) technical specification (s)

TSC Technical Support Center

UFSAR updated final safety analysis repo t

UIR(s) unresolved indication report (s)

URl(s) unresolved item (s)

VIO violation

i

\

{ _ . _

ML20141E341

Text

Navigation menu

Search