ML20236X139

FOIA Request for Documents Re All Info,Including Team Observation Repts,Minutes of Team Meetings & Other Team Shared Info Re 980209-20 NRC Team Insp 50-423/97-82

ML20236X139
Person / Time
Site:	Millstone
Issue date:	06/15/1998
From:	Del Core D AFFILIATION NOT ASSIGNED
To:	NRC OFFICE OF ADMINISTRATION (ADM)
Shared Package
ML20236X128	List: ML20236X126 ML20236X139
References
FOIA-COR98-238 50-423-97-82, NUDOCS 9808070164
Download: ML20236X139 (115)

Text

06/15/1998 14:'25 8608482020 ORBIT ItC PME 02 r, , -f '

- pagwn l .

92- 32

! l.. ^%k d 3D gle M _ G -M -9 9  ;

,AdkBG Re-Donaki W. Del Core, Sr. ,

F

{MCtm Freedom ofsnformation Request -

..[,

1~- s i F01A7NMOfBeer $ .:

U. S. hTyloar Regulatory Commission ~

Washington, D. C. 20555-0001  ;

3

c June 15,1998

+ s FACSIMILE  ;

i j

g  ;.

i

Dear $id/ Madam:

. .) . ,

q . g'

. i would like to formally request the releas, of fjl1information, including Team '

- obsersatiop reports, minutes of team meetings, and'other team shared information '

regardisthe February 9,1998 thmugh Febmary 20i998, NRC Teen Inspection 50-423/97-82 Additionally, please provide copies ofscaidhared personal notes, inter-office ,.

memos,{ tapes, c( mputer records, and other documenteion associated with the visit visits of all;takn members, and others associated vith theteam. I would also like any .

informakh' utilized by NRC management at the Millstofte site, or inhe Region 1 office. #

Also pl$ise provide all team shared information utilig*fby the NRC~ team at the

'l prelinjiriary exit meeting with Millstone managemen( Jkcause the Team utilized NRC O

InspectbtrProcedure 40500, please forward a cop'y of the proceduri utilized by them also.

i l ', . ..

Yihis request is made to you, under the proidsfhns of the Fmedom ofInformation

~

)

Act (Fh4 ). In view of the fact the Northeast Utilities has just recently revealed it's intentio,n to. request a restart vote. I wouki expect thK4bb informatbn be provided as 7;;

cxpedd busly as possible. This information wal te-atilg to infonnland educate the public

• on these.in)portant rcstart issues, and therefore sbob)d pe exempt from any fees. As you are aM'l have already applied for a fee waiver, myWtus with regard to that issue has not Mod,.and therefore shouki be considered withy request also. It is important 3 {

that'thi$ information be addressed as soon 43 possible4e prevent theless ofimportant i

.i public3bclosure. Again, I thank you in advance tbr filut prompt attention in this hnportdntpublic matter. n

.s 7 Sincerely,  !

~1 .

n

-[

~  :.

r ..

rm .

D c / c : Dr. Shirley Arm Jackson ;j Mr. Hubert Bc!1. O!G ce Susan Perry Luxton, CRC . y 9800070i64 980803 D '

PDR FOIA .-

DEL COR98-238 PDR ..

1; e

1

.. I s

INSPECTION PLAN Millstone Nuclear Power Station Unit 3 Effectiveness of Northeast Utilities' Controls in the identification Resolution and Prevention of Problems Inspesion Report No: 50-423/97 82 Inspection Dates: February 2 through 13,1998 Prep week of December 15 t Inspection Team Members:

Jacque Durr, Team Manager, Special Projects Office - Region i John T. Shediosky, Lead, Senior Reactor Analyst, Region i Norman J. Blumberg, Special Projects Office - Region i Edward J. Ford, Quality Assurance and Maintenance Branch (HOMB),

Division of Reactor Controls and Human Factors (DRCH),

Office of Nuclear Reactor Regulation (NRR)

Robert M. Latta, NRR, DRCH, HOMB Richard J. Rasmussen, Senior Resident inspector - Maine Yankee Dr. Garmon West, Jr., Human Factors Assessment Branch, DRCH, NRR Donald A. Beckman, Consultant James C. Higgins, Consultant, Brookhaver. National Lab I inspection Objective:

To evatunte the effectiveness of the Northeast Utilities' (NU) contro's in identifying, resolving and preventing issues that degrade the quality of plant operations or safety.

These controls include: safety committees, root cause analysis programs, corrective action programs, self assessment programs and industry and Millstone Station operational experience programs. The evaluation is to review performance information from the previous twenty-four months. I inspection Area:

Managemert

• Managerm nt - MGMT-1 l Assess the effectiveness of the process by which NU management provides direction to the plant staff necessary to prevent problems. Evaluate the organization's high level goals

and expectations. Conduct interviews.

(WEST 1.11 IBECKMAN 2] (0350-C.2.1.a. C.2.2.a. C.3.1.c)

January 27.1998

1 2

• Management - MGMT-2 Evaluate organizational communications and teamwork, including interdepartmental relationships and interfaces. This willinclude an assessment of both vertical and lateral i

communications. Verify that communications are adequate to properly identify and

! characterize safety significant issues. Determine if the communications between l organizations is adequate to properly address safety issues. Conduct interviews.

l IWEsT 1.21 (0350-C.2.1.g, C.2.2.f) l

• Management - MGMT-3 Determine if managers and supervisors encourage employees to identify problems.

Determine if the staff feels management is receptive to problems being brought forward.

l Conduct interviews.

[ WEST 1.3] (40500-02.03.c)

(0350-C.2.1.f)

• Management - MGMT-4 Review the process to prioritize corrective action activities based on risk, i [ WEST 1.41 (40500-02.03.el I (0350-C.1.4.f. C.2.1.f)

• Management - MGMT-5 Evaluate the performance monitoring (performance indicators), including management information systems, employed to evaluate the following programs: Corrective actions, root cause analysis, self-assessments, independent oversight and operating experience.

Evaluate the effectiveness of the performance measures process. Assess the quality of the information on performance which is provided to management. Verify that any of the performance indicators which identify areas that should be addressed have action taken on them.

[ WEST 1.5]

! [BECKMAN 7]

[ FORD /LATTA 11 (40500-02.03.d, 02.04b, 02.05d, 02.06d)

(slL-41.2, 0350-C.2.1.h)

• Management - MGMT-6 Evaluate management commitment to resolve safety committee recommendations and

audit and assessment findings.

I IWEsT 1.6) (40500 02.06.d)

I (0350.C.2.1.c C.2.1.d)

Corrective Actions

• Corrective Actions - CA-1 Assess the adequacy of the corrective actions program. This assessment willinclude evaluation of programs for the identification, analysis and resolution of plant deficiencies.

IFoRD/LATTA 21 140500-02.01, 02.02.a. 02.02.b. 02.03.a. 02.03.cl (0350-C.1.4.d)

January 27,1998

3

• Corrective Actions - CA-2 Review the effectiveness of the NU programs for corrective action taken in response to a sample of employee concerns issues. Conduct interviews to assess staff understanding and willingness to report problems, IBECXMAN 2]

[ FORD /LATTA 31 (0350 C.1.4.e)

• Corrective Actions - CA-3 Review a sample of root cause analysis and equipment failure evaluations to determine adequacy of the process. For less significant issues review a sampling of the apparent cause determinations. Independently verify the significance of condition reports for significance and that apparent cause or root cause determinations have been performed where required.

[ FORD /LATTA 4] (40500 02.02.b,02.02.c 02.03.d)

(0350-8.4.11

• Corrective Actions - CA-4 Evaluate the resolution of a sample of safety significant issues for timeliness and effectiveness of corrective actions.

[ FORD /LATTA 51

[HIGG' INS 2]

[BECKMAN 2] (40500-02.02.a)

(slL-41.2, 0350-C.2.1.h)

• Corrective Actions - CA-5 Review the backlog of open problem reports to verify that safety significant issues are being tracked to completion. Also, evaluate their action commitment tracking program, Action Requests. Review the process to prioritize corrective actions based on risk.

[ FORD /LATTA 6) (40500-02.03.e)

(0350-C.1.3.c, C.1.3.f, C.1.4.g)

• Corrective Actions - CA-6 Evaluate the NU process for assessing and the effectiveness of corrective actions.

[ FORD /LATTA 7] (40500-02.03.b)

IHIGGINsl IslL-41.2,0350-C.1.3.d. c.1.3.g. C.2.1.h. C.3.1.d)

Self-Assessment

• Self-Assessment - SA-1 Evaluate the Millstone site and departmental self-assessment programs and evaluate their focus on safety. Evaluate the effectiveness of management to address the findings of self-assessments and performance improvement programs. Verify that safety significant issues that could impact plant safety are being been addressed.

[ FORD /LATTA 8] (40500-02.01, 02.02.a. 02.0 5)

(stL-41.2)

• Self-Assessment - SA-2 l Evaluate the effectiveness of the self improvement programs. Compare the self-improvement programs deficiencies to the recently identified plant deficiencies to verify that the NU self improvement programs have adequately addressed the right problems.

[r0RD/LATTA 9] (40500-02.05)

January 27,1998

(

I

\ .

l I l 4 I

! (slL-41.2) l

• Self-Assessment - SA-3 Review the recent self-assessments made by NU and bench mark their conclusions with those of the NRC team to assess the quality of the NU self-asses nent capability.

[ FORD /LATTA 10)

[HIGGINS]

l [BECKMA N 31

• Self-Assessment - SA-4 Review the effectiveness of the NU program to identify and correct operator " work arounds."

l

[ FORD /LATTA 11]

• Self-Assessment - SA-5 Evaluate NU's actions taken in response to ACR-7007 - Event Response Team. This issue concerned the inaccuracies in the UFSAR and design basis documents.

[HIGGWS 1] (slL-41.1) 1 Independent Oversight I

• Independent Oversight 1 Assess the NU Quality Assurance (QA) involvement in operations, maintenance and {

surveillance and engineering. Verify that the QA organization adequately implements the l

l QA program.

, [ NORM 1] (slL-41.2, slL-73.1, slL 73.2,0350 C.1.4.a)

)

• Independent Oversight 2 Verify that significant QA audit findings, with safety implications, have been resolved in a high quality manner. Verify that reviews have been made to verify the effectiveness of the resolution of previously identified issues. Review audits of corrective action program deficiencies.

[ NORM 21 (40500-02.02.a, 02.03.b)

(stL-73.2)

• Independent Oversight 3 Evaluate the effectiveness of the offsite independent safety committee and other independent groups designed to oversee plant safety. Verify that the oversight provided by these groups is adequate to identify adverse plant performance and to initiate proper corrective actions to have the issues addressed. Also, verify that the outstanding items identified by these groups are being tracked to resolution. Verify that programs have been l estcalished to verify the effectiveness of actions for previously identified issues.

l

[ RICK 11 (40500-02.06.a, 02.06.b, 02.06.cl l (0350-C.1.4.c. C.2.1.d) l l

January 27,1998 I'

5

• Independent Oversight 4 Evaluate the effectiveness of the onsite independent oversight groups. Verify that they are capable of conducting reviews required by the Operating License Technical Specifications.

Also, verify that the outstanding items identified by these groups are being tracked to resolution. Verify that programs have been established to verify the effectiveness of actions for previously identified issues.

[ RICK 21 (40500-02.06.a, 02.06.b)

(0350-C.1.4.c. C.2.1.d)

• independent Oversight 5 Evaluate the effectiveness of the operating experience program, this includes the programs for evaluation industry data and site experience (Independent Safety Reviow Group),

dissemination of information or assignment of required actions, program effectiveness reviews and issue back log. Includes NRC IN 97-78

[ RICK 3] (40500-02.04)

(0350-C.1.4.b)

(Allegation: 97-A-0241)

Issues Scheduled for Closure During the 40500lnspection:

The following Significant items List (SIL) and MC-0350 items were identified in the NRC's Restart Assessment Plan, dated July 24,1997, and were assigned to the scope of the MC-40500 inspection.

SIL 41, Review Self-Assessment Root Causes and Verify Corrective Actions SIL 41.1: ACR 7007 for Unit 1 and the corresponding CR M3-97-1839 for issues applicable to Unit 3, addressed the fundamental causes for the Millstone Unit 1 UFSAR inaccuracies.

SIL 41.2: NRC Inspection 423/95-81 identified Unresolved item 95 91-01, Non-conformance Reports which were being processed by the QA department were not being trended for identifying possible adverse conditions to management as required by the QA Program.

SIL 73, Quality Assurance and Oversight Program SIL 73.1: NRC Inspection 423/96-05 identified a Violation item 96-05-12, failure to Audit the Technical Specification 6.8.4.e, " Accident Monitoring Instrumentation," within a five year interval.

SIL 73.2: NRC Inspection 423/96-17 identified Unresolved item 96-06-17, that the 1996 Joint Utilities Management Assessment LIUMA) review of the NU Quality Assurance i

Program concluded that Nuclear Oversight was not effectively implementing its audit, surveillance, and inspection program. Additionally, Nuclear Oversight was not ensuring l

timely resolution of identified problems.

MC-0350 - Restart Approval Checklist item B.4.1 Root Causes and Corrective Actions Evaluate the findings of the special team inspection, and evaluate the NU root cause determination and corrective action plan.

January 27.1998

I l MC-0350 - Restart Approval Checklist item C.1.3 Corrective Actions

! Issue detailed items c, d, f, and g require review of corrective action item tracking, effective corrective actions, control of long term corrective actions and effectiveness of the corrective action verification process, respectively.

[ MC-0350 - Restart Approval Checklist item C.1.4 Self-Assessment Capability i Issue detailed items a, b, c, d, e, f, g require effectiveness of QA Program, effectiveness of the indus try experience review program, effectiveness of the licensee's independent review groups, effectiveness of the deficiency reporting system, staff willingness to raise concerns, effectiveness of PRA usage and effectiveness of the commitment tracking

program, respectively.

l MC-0350 - Restart Approval Checklist item C.2.1 Management Oversight Effectiveness issue detailed items a, c, d, f, g, and h require that goals and expectations be communicated to the staff, management involvement in self-assessment and independent assessment capability, effectiveness of management review committees, management's ability to identify and prioritize significant issues, management's ability to coordinate resolution of significant issues, and m nagement's ability to implement effective corrective actions, respectively.

MC-0350 - Restart Approval Checklist item C.2.2 Management Support issue detailed items a, and f require impact of any management reorganization, and participation in industry groups respectively.

MC-0350 - Restart Approval Checklist item C.3.1 Assessment of the Staff issue detailed items c, and d require understanding of management's expectations and goals, and understand;ng of plant issues and corrective actions, respectively.

Allegation 97-A-0241 Operating Experience Reviews have been inadequate it was alleged on October 1,1997, that due to low staffing levels within the Nuclear Safety Engineering organization and the large backlog of operating experience (OE) issues that reviews have been inadequate, especially during the recent large reduction in the OE backlog.

1 i

January 27.1998 l

I l

l t

7 Inspection Procedures:

40500," Effectiveness of Licensee Controls in Identifying, Resolving and Preventing Problems" Additional background information can be found in:

40001," Resolution of Employee Concerns" 40301," Safety Committee Activity" 40501, " Licensee Self-Assessments Related to Team inspections" 93802, " Operational Safety Team inspection (OSTI)"

93806," Operational Readiness Assessment Team inspections" 93804, " Risk Based Operational Safety and Performance Inspection"

References:

NRC Restart Assessment Plan, dated July 24,1997 l

l January 27,1998

t c -

F Garmon Debrief Bullets 2/23/98 MGMT-1. Assess the effectiveness of the process by which NU management provides direction to the plant staff necessary to prevent problems. Evaluate the organization's high level goals and expectations.

+ Meetings between upper management and plant personnel were effective

+ Various methods used by plant management to communicate expectations was a strength

+ Common understanding of management's expectations by plant personnel I

(

Strategic plan and visual on where plant is heacea are in draft was considered a ]

ws.akness

+/- Overall, the " Nuclear Group Polices and Standards," were considered good; the lack of a {

" Nuclear Group Mission and Vision" statement because it is under development was considered a weakness j l

MGMT 2. Evaluate organizational communications and teamwork, including interdepartmental relationships and interfaces. This will include an assessment of both vertical and lateral communications. Verify that communications are adequate to properly identify and characterize safety significant issues. Determine if the communications between organizations is adequate to properly address safety issues.

+ Vertical communications are considered good Lateral communications need improvement

+ Communications were considered adequate relative to identifying and addressing safety issues  !

+/- Teamwork initiatives at the first line supervisor and above levels need to continue to be reinforced Teamwork at the plant worker level needs improvement MGMT 3. Determine if managers and supervisors encourage employees to identify problems.

Determine if the staff feels management is receptive to problems being brought forward.

+ Observations and interview results, and plant surveys show thct managers and supervisors encourage employees to identify problems l

+ Interview results indicated that the plant staff feel management is receptive to problems being brought fotward

- Interview results indicated that a limited number of plant staff may not fully trust plant rnanagement: have a " wait-and-see" view of management; therefore, pursue other e

I

1 I l

r 2

alternatives to raising problems rather than with their management MGMT 4. Review the process to prioritize corrective action activities based on risk.

[ Tom will address this item.)

MGMT 5. Evaluate the performance monitoring (performance indicators), including management information systems, employed to evaluate the following programs: corrective i actions, root cause analysis, self-assessments, independent oversight and operating experience. Evaluate the effectiveness of the perfonnance measures process. Assess the quality of the information on performance which is provided to management. Verify that any of the performance indicators which identify areas that should be addressed have action taken on ,

them. l l

+ Performance indicators to evaluate the subject programs were considered excellent

+ The effectiveness and the quality of information on performance was considered 1 satisfactory

)

Personnel errors [e.g., five fire protection surveillance (M3-97-3035, 97-3981, 97-4246, 97-4394, and 97-4618)] that were intentionally not performed was a weakness )

- The usefulness of personnel error data provided to line management needs improvement (e.g., total monthly data versus broken down by type of error or normalized to provide information i l'

The high number of LER-related human errors was considered a weakness and needs to be further examined MGMT 6. Evaluate management commitment to resolve safe committee recommendations, and audit and assessment findings. i l

+ Self-assessment recommendations on reactor engineering procedures had been addressed

+ PORC recommendations on a contingency plan to repair a valve were resolved Memo dated 12/23/97 re: ISEG independent verification of human errors are reduced as l much as practical does not meet TS 6.2.3.3 ,

1 Memo dated 7/7/97 (CRM3-97-1875) re: independence of radiation protection manager does not meet TS 6.2.1.d and recommendation in audit report MP-97-A06-02 1

+ SPDS-related issues raised by the team were resolved

- " the failure to fulfill ANSI /ANS 3.5 standards requirements.. " was considered a weakness regarding simulator and plant fidelity (re: self-assessment 97-004)

- it was considered a weakness that design engineering (re: NGP 5.25) determines

1 1 1

i I i

l l

I l 3 l whether modifications require a human factors review i

1 l

l l

1 i

i

r Garmon Debrief Bullets 2/25/98 MGMT-1. Assess the effectiveness of the process by which NU management provides direction to the plant staff necessary to prevent problems. Evaluate the organization's high level goals and expectations.

+ Meetings between upper management and plant personnel were effective

+ Varione methods used by plant management to communicate expectations was a strength

+ Common understanding of management's expectations by plant personnel Strategic plan and visual on where plant is headed are in draft was considered a weakness

+/- Overall, the " Nuclear Group Polices and Standards," were considered good; the lack of a

" Nuclear Group Mission and Vision" statement because it is under development was considered a weakness MGMT 2. Evaluate organizational communications and teamwork, including interdepartmental relationships and interfaces. This will include an assessment of both vertical and lateral communications. Verify that communications are adequate to properly identify and characterize safety significant issues. Determine if the communications between organizations is adequate to properly address safety issues.

+ Vertical communications are considered good

- Lateral communications need improvement: interview results noted the following:

a. time co..tramts

b. communications are often times issue driven: something needs to be resolved

c. Impediments: cultural issues, respect issues

d. Inter-department communications need improving

o. more interface between groups needed

+ Communications were considered adequate relative to identifying and addressing safety issues

+/- Teamwork initiatives at the first line supervisor and above levels need to continue to be reinforced 1

- Teamwork at the plant worker level needs improvement: interview results indicated the following-

a. some groups in conflict (most notably oversight and maintenance) i

b. not everyone familiar with conflict resolution

c. sometimes interpersonal conflicts don't get resolved

d. no formal process for rotational assignments

(

l

, YJD

I I

2 MGMT 3. Determine if managers and supervisors encourage employees to identify problems.

Determine if the staff feels management is receptive to problems being brought forward.

+ Observations and interview results, and plant surveys show that managers and supervisors encourage employees to identify problems

+ interview results indicated that the plant staff feel management is receptive to problems being brought forward i I

- Interview results indicated that a limited number of plant staff may not fully trust  !

plant management: have a " wait-and see" view oif management; therefore, purcue i other alternatives to raising problems rather than with their management Note: This lack of trust appeared to be a carryover of the previous negative perception of past management and questioning whether they cared about plant

, personnel; however, there was common agroement (with examples to support this contention) that current management does care about plant personnel. ]

MGMT 4. Review the process to prioritize corrective action activities based on risk.

[ Tom will address this item.]

MGMT 5. Evaluate the performance monitoring (performance indicators), including

[

management information systems, employed to evaluate the following programs: corrective actions, root cause analysis, self assessments, independent oversight and operating experience. Evaluate the effectiveness of the performance measures process. Assess the quality of the information on performance which is provided to management. Verify that any of the performance indicators which identify areas that should be addressed have sction taken on

! them.

+ Performance indicators to evaluate the subject programs were considered excellent i

+ The effectiveness and the quality of information on performance was considered satisfactory

- Personnel errors (e.g., five fire protection surveillance (M3-97-3035, 97-3981, 97-4246, 97-4394, and 97-4618)] that were intentionally not performed was a weakness l - The usefulness of personnel error data provided to line management needs  ;

improvement (e.g., total monthly data versus broken down by type of error or normalized ,

to provide information  ;

- The high number of human-performance-related LERs for 1997 was considered a weakness and needs to be further examined: the team's analysis of LERs found Unit 3 had 16 human-performance-related LERs in 1997 versus a national average of 6 human-performance-related LERs MGMT 6. Evaluate management commitment to resolve safe committee recommendations, l

i

- - _-_ _ -_ __- __-___-______ _a

I a

3 and audit and assessment findings.

+ Self-assessment recomrr.endations on reactor engineering procedures had been addressed

+ PORC recommendations on a contingency plan to repair a valve were resolved Memo dated 12/23/97 re: ISEG independent verification of human errors are reduced as much as practical does not meet TS 6.2.3.3

__ Memo dated 7m97 (CRM3-97-1875) re: independence of radiation protection manager does not meet TS 6.2.1.d ard recommendation in audit report MP-97-AOS-02

+ SPDS-related issues raised by the team were resolved

+ Overall, the simulator update process which ensures the fidelity of the simulator is maintained with regard to the reforence plant was considered a strength

a. strengths: the licensee's documentation stated:

(1)"At this time, there are no design changes with simulator impact that have been installed in the plant for greater than 30 days that have not been incorporated in the simulator." ,

I (2) "To support the current restart training needs, we have modified our target for i incorporating those plant design changes identified by Operations and Operator Training as having simulator impact to have them installed within 30 days of plant ,

installation. This is beyond what is required by ANS-3.5, the standarff to which I the Millstone 3 simulator is certified to. The standard allows 24 months to

~

incorporate such changes." I

b. weakness:

(1) the licensee's self-assessment (97-004) stated:

l "This assessment has demonstrated the absolute need for coordination between site and the simulator support group to insure that all plant modifications are accounted for in the simulato, upkeep." l it was considered a weakness that design engineering (re: NGP 5.25) determines i whether modifications require a human factors review l

l l

l i

Management Exit Meeting for Millstone Unit 3 Corrective Actions inspection l Inspection Report 50-423/97-82 '

Introduction:

This meeting is a management exit for the Millstone Unit 3 Corrective Actions Team inspection, which was conducted on-site February 9 through 20,1998. Our objective was to evaluate the effectiveness of the Northeast Utilities' (NU) controls for identifying resolving and preventing issues that degrade the quality of plant operations or safety.

This meeting is open for public observation; but is a working meeting between NU and the NRC. There is a second Exit meeting conceming an Emergency Planning inspection immediately following this meeting, the NRC staff will be available following that second Exit to answer questions from the public, inspection Dates: February 9 through 20,1998 Inspection Objective:

To evaluate the effectiveness of the Northeast Utilities' (NU) controls in identifying, resolving and preventing issues that degrade the quality of plant operations or safety.

Inspection Team Members:

I would like to introduce Wayne Lanning, Deputy Director for inspections of the Special Projects Office. And, Jacque Durr, Chief, inspections in the Special Projects Office, Jacque is also the Team Manager. Other team members, not present today are:

Jacque Durr, Team Manager, Special Projects Office - Region l l John T. Shedlosky, Lead, Senior Reactor Analyst, Region l Norman J. Blumberg, Special Projects Office - Region I Edward J. Ford, Quality Assurance and Maintenance BrLnch (HOMB), j Division of Reactor Controls and Human Factors (DRCH), 1 Office of Nuclear Reactor Regulation (NRR)

Robert M. Latta, NRR, DRCH, HOMB Richard J. Rasmussen, Senior Resident inspector - Maine Yankee l Dr. Garmon West, Jr., Human Factors Assessment Branch, DRCH, NRR  !

Donald A. Beckman, Consultant James C. Higgins, Consultant, Brookhaven National Lab i wish to thank Northeast Utilities management for the cooperation extended to the team, and in particular thank the NU staff coordinations who supported the individual +eam l members. Their efforts allowed us to cover quite a bit of material over the two weeks on site, which ir.cluded their working the intervening weekend and holiday with us.

l February 26,1998 lL j ll

Manageraent Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 Our inspection was organized over four areas:

Management Processes and Systems Corrective Actions Self Assessment independent Oversight I would like to start by briefly addressing our findings which are under consideration as ENFORCEMENT ISSUE 3. And, then review our inspection findings in more detail within the four areas which we covered.

Technical Specification Section 6 - Organization Independence TS 6.2.1.d, the FSAR and Reg Guide 8.8 specify organizational independence for the Radiation Protection Manager. Contrary to these requirements, the Radiation Protection Manager reports to the Maintenance Manager, a configuration specifically prohibited by the Reg Guide.

TS 6.2.3.3 requires that the ISEG perform independent review to reduce human error, the ISEG has obtained the service of Human I-actors practitioners who are assigned to the NSE organization. Contrary to the TS the HF practitioners have been called upon to review the work of their peer. [There is an administrative instruction (memo) in place, which prohibits review of their own work, but allows ISEG reviews of work performed by the peer in the organization.

j l

Condition Report Significance Level Classification )

l RP-4, Corrective Actions, establishes criteria for CR significance level. Several examples l were discovered where classifications were lower than required.

CR's written issued becausa of incomplete action on GL 90-03, Ven6r Technical information Program, and GL 89-13, Service Water Fouling. These were classified as Level 3, an Enhancement item, instead of Level 2, a condition adverse to quality. RP-4 requires that NRC commitment issues be assigned Level 1 or 2. Additionally, the CRs were coded as ' Deferred' until after start-up, and included on the Deferred items List submitted to the NRC. [There were actually three CRs for the GL 89-13 issue, two of the three were Lsvel 3, one was Level 2, all were deferred post startup. There was one CR for the GL 90-03 issue, it was coded Level 3 and was deferred post startup when written in August 1997, l the action to complete the CR was done in December].

I i

l l

2 February 18.1998

Management Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 i

A CR (M3-97 0530) which addressed electrical separation issues in the Control Room Main Control Board was originally categorized as Level 1 but changed to Level 2. There was no basis for the change.

The ISEG intervened in Main Transforrr,er switchyard work for industrial safety and risk to off-site power supply reasons. However, the CR which recorded the problems was classified as Level 3, thb was not identified or challenged by the ISEG group.

A series of problems were documented in CR M3-98-0309[and 98-0641 and 0642] about CRs that were written throughout 1997 that documented problems with the NCR process.

Some of those CRs [ Level 1 and 2] were closed to lower level CRs which were j improvement items and therefore not requiring action. This issue is also a program weakness because the combined issues may not properly carry the issue forward. ]

I Condition Report Closure RP-4, Corrective Actions, requires effective actions to prevent recurrence A CR, M3-97-2943 concerning air binding of the Boric Acid Transfer Pumps [3CHS-P2A and -P2B] was closed with essentially the same corrective actions taken earlier. There have been a series of five events of this type from 1992, three of which occurred in 1997.

This issue also demonstrates inadequate RCA. Also, the corrective action taken for 97-2943 was not recognized as a USO because it concluded that a more conservative specification for Boric Acid Tank level was required.

One Action Request (AR) under CR M3 97-0506was closed improperly without completing all reviews. The action was to review administrative DCNs for changes that should have received safety evaluations, but did not. The review was to be made for the past 5 (or 7) years, the action was closed, however only one year was reviewed. Completion of this action was an ACR 7007 and CMP restart commitment.

Audit findings that the ISEG Operating Experience procedure was not reviewed by SORC were responded to by including the review with a new Operating Experience procedure being developed. However, the ISEG procedure also needed SORC review.  !

, i l There was a lost opportunity for earlier detection of material condition deficiencies in High  !

Energy Line break doors because of narrow corrective actions for CRe concerning finding these doors open or not latched.

Although not categorized as a CR, an operator " work around" that involved service water flow indication, was closed before all actions were complete (Reference OWA 96-003). A CR was issued.

l 3 February 26,1998

Managemere Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 Other issues categorized as Failure to Follow Procedures NGP 5.28 requires that DCNs be used to update the Safety Functionti Requirements Manual. The SFRM documents system characteristics for input to the FSAn Chapter 15 analysis. Several revisions were not processed correctly, DCNs were not issued and entered into the Generation Records Tracking System (GRITS).

NGP 5.23 requires that the Master Setpoint List contain the setpoint an,1 the calculation reference. Problems were identified with two of ten items checked. The basis calculation for the set point for saturation subcooling margin was not available. It was stated to be 15*F, this 'is nonconservative low compared to the Safety Parameter Display Systein set point established at 32*F. In a second issue, the value for pressurizer level setpoint was blarL. Also, a general question of Set point controlis open.

PI-20 requires Design Basis Summary (DBS) documents for risk significant systems. The DBS have not yet been updated although they have collected DCNs for the last two years.

A Gap Report was construe:ted. DBS do not exist for two Maintenance Rule (MR) systems (Charging Voluma Control (CVCS) and for Emergency Lghting.

Other Findings that are Unresolved Long term compensatory measures are in effect for fire protection systems because surveillance testing which verifies operability of these systems has been suspended.

Compensatory measure are taken to allow restoration of a fire protection system. But, not intended to substitute for long term system inoperability. This issue will be referred the Fire krotection inspection which is scheduled for the near future.

Corrective actions to verify that 480 volt molded case circuit breaker trip settings are appropriate for their motor loads has been delayed until the end of the next refuel outage.

We have requested information related to the risk and safety significance of the motors remaining to be verified. With t.his additional information we will be able to judge tha l appropriateness of delt. ,'ing this work for that period of time.

l The ISEG reviews of plant operation have dropped from 24 in 1996 to 12 in 1997. The

ISEG group has reduced the backlog of operat.ng events significantly. However, the amount of work represented by the remaining issues is significant, and several may identify safety issues.

4 February 26,1998 I

! Management Exit Maeting Millstone Unit 3 - Corrective Actions February 26,1998 Within the first area, Management Processes and Systems we examined six areas:

Management Directions, Goals, Expectations Organizational Communications and Teamwork Managemont.and Supervisory Encouragement of Problem identification Process to Prioritize Corrective Actions Performance Monitoring l Management Commitment to Resolve issues 1

i l Management Directions, Goals, Expectatieres (40500) (0350-C.2.1.a, C.2.2.a, C.3.1.c) ~

l Assess the effectiveness of the process by which NU management provides direction.to

, the plant staff necessary to prevent problems. Evaluate the organization's high level goals

!- and expectations. Conduct interviews.

1 l

[ Evaluate the subject process against relevant criteria in the NRC'c Human Performance j Investigation Process (HPIP, NUREG/CR-5455). Assess the ability of the plant's  !

management team to direct the plant staff by evaluating meeting effectiveness and priority 1 f

setting. Review " Nuclear Group Policies and Standards." Interview the managers described in Attachment B.

l Evaluation criteria will include the following: what is expectation on preventing and l correcting problems, what is expectation on identifying and addressing safety issues, not strict enough, confusing or incomplete, communication of process / goals / expectations

! LTA, recently changed, enforcement LTA, no way to implement, accountability LTA,. does upper management have a plan for phasing out temporary employees who are managers (contractors and on loan from another plant) and phasing in permanent employees after startup, support of corrective actions, and what are expectations relative to communications.

Our findings in this area are:

a Management communications methods were a strength.

! A There was a common understanding of management's expectations by plant personnel.

l v Strategic plan and vision statement on where the plant is headed are in draft. This is a weakness.

1. Overall, the " Nuclear Group Policies and Standards," were considered good; the lack of a " Nuclear Group Mission and Vision" statement, under development, was a weakness.

5 February 26.1908 l

I _

. l l Management Exit Meeting Millstone Unit 3 - Corrective Actions i i February 26,1998 Organizational Communications and Teamwork (40500)(0350-C.2.1.g,C.2.2.f)

Evaluate organizational communic 6tions and teamwork, including interdepartmental relationships and interfaces. This willinclude an assessment of both vertical and lateral communications. Vecify that communications are adequate to properly identify and characterize safety significant issues. Cetermine if the communications between l organistions is adequate to properly address safety issues. Conduct interviews.

With regard to organizational communications and teamwork, assess the following: (1) organizational communications by evaluating the staffs understanding and implementation of management expectations, especially regarding identifying and preventing problerns and identifying and addressing safety issues; (2) the effectiveness of meetings relative to interdepartmentalinterfaces and teamwork, and (3) evaluating the interdepartmental 3 effectiveness of activities in closing issues [ identify 1-2 issues). Interview the individuals '

described in Attachment 8. Attend the following meetings: Unit 3 status ineetings and interdepartmental meetingc.

Our findings in this area are:

0 Vertical communications are considered good 0 Lateral communications need improvement: interview results noted the following:

a. time constraints

b. communications are often times issue driven: something needs to be resolved

c. impediments: culturalissues respectissues

d. inter-department communications need improving

e. more interface between groups needed a Communications were adequate relative to idersifying safety issues.

1. Although, teamwork initiatives at the first line supervisor and above were developed, strongest at the upper levels. Need to be extend this to the worker level. For example:

v [Beckground information: some groups in conflict (moet notably oversight and maintenance)]

O Not everyone familiar with conflict resolution 0 Sometimes interpssonal conflicts don't get resolved 0 No formal process for rotational assignments 6 February 26,1998

Management Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 Management and Supervisory Encouragement of Problem Identification (40500)(40500-02.03.c)(0350-C.2.1.f)

Determine if managers and supervisors encourage employees to identify problems.

Determine if the staff feels management is receptive to problems being brought forward.

Conduct interviews.

The team conducted employee interviews at all levels of the organization to determine the worker's perceptions of management's efforts and communications intended to enhance problem identification. Performance measurement data for problem identification and documentation associated with individual-identified problems (Condition Reports, Employee Concern Case Files, Self Assessment results and others) were reviewed The team conducted interviews, reviewed corrective action program data, and reviewed the licensee's technical resolutions, SCWE-related responses, and long term follow-up for problems identified by condition reports, employee concerns and Employee Concerns Oversight Panel issues, and NU department self assessments. This included a sample of condition reports and their resolution activities that resulted from employee concern cases, self assessments.

Our findings in this area are:

A Observations and interviews show that managers and supervisors encourage employees to identify problems.

4 The plant staff feels that management is receptive to problems brought forward (West).

Individuals generally characterized the environment as improved and currently receptive to problem identification (Beckman).

Safety Conscious Work Environment Programs:

We examined the organizational structure supporting the Safety Conscious Work Environment (SCWE), the Employee Concerns (ECP) and the Human Resources (HR) groups, for the effects of recent re-alignments. We understand the reasons for these actions. We recommend that efforts be made to stabilize the your organization in this area to promote the effectiveness of these groups. 9

1. individuals generally characterized the environment as improved and currently receptive to problem resolution. However, interview results indicate that some plant staff may not fully trust plant management. They have expressed a " wait-and-see" view of management. They therefore pursue other alternatives to rasing problem rather than with their management.

7 February 26,1998 i

i

l Management Exit Meeting Millstone Unit 3 - Corrective Actions

! February 26,1998 4 There is no reluctance or reservations expressed by individuals to identify problems through the Corrective Action Condition Report Process, to ECP or the NRC.

I v The SCWE processes have not yet been formalized. [This is the same finding as MC 40001 inspection). This has resulted in some of the management actions being handled personally by the Recovery Managers and other senior managers.

a 'The handling of individual HIRD cases by the Employee Concerns Program and the Safety Conscious Work Environment program is adequately responsive to specific case i

needs. Both technical and human-side problems are generally well addressed.

4 The Employee Concerns Oversight Panel (ECOP) oversight activities and surveys used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems are positive contributions to the overall process, v However, NU has not been effective in addressing the high incident rate of HIRD issues in ECP system. The incident rate of HIRD allegations and management-related or induced chilling effect events has not diminished significantly.

A Performance indicators and Corrective Action program statistics reflect adequate levels l of CR participation.

4 The identification methods for HIRD issues include the culture and leadership surveys, ECP case intakes, and ECOP identification and referrals. These mechanisms are effective especially for the more egregious issues identified as problem areas.

• Documented action plans are implemented for the Problem Areas. Sampling determined that the plans are generally effective at remediating both the technical and human performance and behavior issues.

A SCWE problem area corrective actions for Harassment, Intimidation, Retaliation and Discrimination (HIRD) issues are adequate and effective.

p Other HIRD issues that are perceived by NU management to be less egregious or are minor problems having the potential to become Problem Areas are handled less formally by senior and middle management.

4 NU is monitoring a number of units which pose the potential of becoming problem areas. Monitoring is done by the line management via normal activities, review of ECP intakes, and SCWE daily meeting which include dialog with ECP, ECOP and other program participants.

8 February 26.1998 i

i .

l l -

l Management Exit Meeting Millstone Unit 3 - Corrective Actions >

l February 26,1998 v No documented action plans are used but management demonstrated that actions have been taken in response to such HIRD potentials.

l Process to Prioritize Corrective Actions (40500)(40500-02.03.e)(0350-C.1.4.f,C.2.1.f)

Review the process to prioritize corrective action activities based on risk.

Review the relevant procedure for prioritizing corrective action activities based on risk.

Interview a recent user of this process.

v NU has recently begun using an additional risk significance classification for CRs.

However, this classification is not tied to the PRA, Maintenance Rule risk significance or information from the IPE.

Performance Monitoring (40500)(40500-02.03.d,02.04b,02.05d,02.06d)(SIL-41, 0350-C.2.1.h) l Evaluate the performance monitoring (performance indicators), including management information systems, employed to evaluate the following programs: Corrective actions, j root cause analysis, self-assessments, independent oversight and operating experience.  ;

Evaluate the effectiveness of the performance measures process. Assess the quality of the information on performance which is provided to management. Verify that any of the I performance indicators which identify areas that should be addressed have action taken on them. i Review the quarterly and weekly performance indicators to identify areas (e.g., corrective action performance indicators) that should be addressed have action taken on them.

Include interview questions with managers to determine what information they use to evaluate the following programs: corrective actions, root cause analysis, self-assessments, independent oversight and operating experience.

• Performance indicators to evaluate the subject programs were considered excellent. 9 A NU has appropriately addressed adverse trends identified in the Fourth Quarter Report concerning: Maintenance department procedure compliance, Operations department valve and breaker alignment issues and tagging errors, and Surveillance testing.

v The high number of LER-related human errors was considered a weakness and needs to be further examined. The team's analysis of LERs found Unit 3 had 16 human-performance-related LERs in 1997 versus a national average of 6 human-performance-related LERs 9 February 26,1998

• E l

Management Exit Meetir.g Millstone Unit 3 Corrective Actions February 26,1998 Management Commitment to Resolve issues (40500)(40500-02.06.d)(0350-C.2.1.c, C.2.1.d)

Evaluate management commitment to resolve safety committee recommendations and audit and assessment findings.

Evaluate management commitment to resolve the following: (1) PORC recommendations relative to PORC Number 3-97-256," Contingency Plan for Orange Path Inventory Due to 3CHS*HCV190A Repair," and PORC Number 3-97-25," Controller Modification for 3CHS*&HCV190A;" (2) audit recommendations relative to " Northeast Utilities Audit Report MP-97-A06-02, Radiation Protection," and (3) self-assessment findings relative to Millstone Unit 3 reactor engineering procedures.

A Self-assessment recommendations on reactor engineering problems had been addressed.

(Not discussed at management debrief).

A PORC recommendations on a contingency plan to repair a valve were resolved. (Not discus. sed at management debrief).

A SPDS-related issues raised by the tearn were resolved. (Not discussed at management '

debrief).

A Overall, the simulator update process which ensures the fidelity of the simulator is maintained with regard to the reference plant was considered a strength Corrective Actions Our observations are organized along with the process for which a Condition Report reaches resolution.

Identification and Classification Root Cause Analysis Appropriate Corrective Actions Completion and Closure of Condition Report l Identification and Classification:

A We observed a low threshold for identification of issues, that is Significance Level 3, Enhancement issues.

l A The Management Review Team (MRT) activities processing of Condition Reports (CRs) was judged to be effective by NRC Team members attending meetings.

10 February 26,1998 i

l

Management Exit Meeting Millstone Unit 3 - Corrective Actions j February 26,1998 v However, there were several examples where the Significance Level had been classified 1 lower than assessed by the NRC team, or the MRT had approved changing Level to a level that was inappropriately low. Examples of this are:

l v Condition Reports (CR's) issued for incomplete action on GL 90-03, Vendor i l

Information Technical Program (VITP), and GL 89-13, Service Water Fouling, were classified as Significance Level 3, an enhancement item, although they both

]

concerned NRC commitments for previous inspection findings. The team also noted that these issues were also included on the Deferred items List v A CR (M3 97-0530) addressing electrical separation issues in of electrical separation in Control Room, Main Control Board, was changed from Significance Level 1, a significant condition adverse to quality, to Level 2, a condition adverse to quality. There was no basis for the change.

v Recently ISEG intervened in Main Transformer switchyard work because l equipment was found running unattended, safety questions on equipment lifting practices and because of degraded backup power for Unit 3. However the CR was classified as Significance Level 3, an improvement item, which is j inappropriately low. This was not identified or challenged by the ISEG group.

v 'CRs were written on issues involving the NCR program, the CRs addressing these issues were classified as either Significance Level 2 or 3. In handling these CRs after initial processing, the MRT had assigned the Root Cause Analysis to be completed with another CR, and also had combined the Level 1 action items with a Level 3 CR, which was contained only improvement items. The MRT subsequently recognized the error and changed the CR the Significance Level 2.

The Corrective Actions program does not limit the combination of CR issues, nor limit CRs of greater significance being combined into CRs of low significance.

The NRC team considers the absence of controls in this matter to be a program weakness.

v NU has recently begun using an additional risk significance classification for CRs.

However as I mentioned previously, this classification is not tied to the PRA, Maintenance Rule risk significance or information from the IPE.

Root Cause Analysis:

A Review of Level 1 CR Root Cause analysis has observed an improving trend from mid-1996 to the present. Significant improvement was noted in analysis performed about a year ago. The additional review and grading program for RCA has added quality. This is an area which requires continued attention, we have observed some weaknesses in RCA and with RCA being deferred. We understand that program improvements are in place.

I 11 February 26,1998 i

. _ , _ , _ _ _ _ , _ _ _m____mm__ _ m . . - . _ _ _ _ _ . . . _ _ . _ _ _ _ . . . - - - _ - - - - __ -

i 0

- j Managernent Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 Inappropriate Corrective Actions:

v Actions to complete the verification of 480 volt molded case circuit breaker overload current magnetic trip settings have been deferred until the end of the next refueling outage, RFO 6, for inaccess ble motors. Additional information on the risk significance of those electrical motors is needed for final evaluation of this issue.

Inappropriate Closure of Condition Report:  ;

i v A CR concerning trips of the Boric Acid Transfer Pump was improperly ::losed. The j issue has been recurrent. Therefore we question the quality of the current and past corrective actions. Also, the NRC team questions the completeness of the Root Causu analysis because it did not consider possible air or gas intrusion paths, other than tank level.

We found that the specified corrective action, that is, increasing the minimum tank level administratively if correct, is also an Unroviewed Safety Question, because the Technical Specification for minimum tank level was less conservative than the new

]

administratively controlled minimum level. (The Technical Specification is for minimum volume for reactivity purposes, if the system required a higher level for reliable operation, that is volume above the minimum for reactivity purposes).

We also questioned the thoroughness of the operability and deportability determination which concluded that when found there were alternate bc ic acid flow paths available.

It did not consider previous events where the RWST may have been isolated.

• A CR was issued for the ISEG/OE procedure not being SORC approved. After a separate  ;

OP procedure was written for Operating Experience review, the CR action became l uncoupled from the fact that the ISEG procedure also required SORC review and approval.

• A CR related to procedures, tools anc; equipment needed to support EOPs, which were not available in the plant. The corrective actions for this CR were not tied to a key  ;

event and were outside of the restart at the time that it was approved. This issue was l also identified during an NRC review of the deferred items and was then designated as a  !

Mode 2 issue.

v Yhere was a fost opportunity for earlier detection of material deficiencies in High Energy Lne Break (MELB) doors because of narrow corrective actions for CRs which a series of i CRs HEi.B door closure issues.

• Although not categorized as a CR, an operator " work-around" that involved service water flow indication, was closed before all actions were complete (Reference OWA 96-003). A CR was issued M3-98-0942.

l 12 February 26.1998 l l

I  !

e

~

t Management Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 Conclusion Although we have seen evidence of a corrective action program beginning to function, it is clear that the program will require careful monitoring by NU management. We have found a disproportional number of issues in a relatively small sample size of CRs. Our observations were made despite your extensive review of the corrective action process by your self-assessment team Your team reviewed approximately 275 Level 1 CRs prior to our inspection.

We feel that on going attention is needed to facilitate program quality.

Self Assessme'its  ;

A The team found that the self assessments were generally of high quality. You have taken a strong initiative in areas like the Nuclear Oversite Restart Verification Assessment.

v However, we have observed the need for continualimprovement in assessments perforrned by the line departments. Continued attention should be applied to the line l departmental self assessments in order to improve their quality.

Significant items List issues SIL ltem 41-1 concerning the findings of the root cause team for FSAR and licensing basis inaccuracies ACR 7007 (ACR 7007 identified 104 issues,13 were designated Unit 1 only) and two items related to M3-97-1839 and ACR 13302 were reviewed. Based on the results of the review we are recommending to NRC management that the SIL item be closed.

in addition the process for UFSAR changes was reviewed and determined to be functioning. We observed that process functioning, and has been proces:ing changes to the Safety Analysis.

However, our inspection identified the following issues which remain open:

• The Safety Functional Requirements Manual was discovered lacking configuration l

control. This documents system characteristics for input to the FSAR Chapter 15 analysis. Several revisions were not processed correctly, DCNs were not issued and ,

entered into the Generation Records Tracking System (GRITS). l l

\

1 ,

1 \

l 13 rebruary 26,1998 {

l

o 4

Management Exit Meeting Millstone Unit 3 - Corrective Actions February 26,1998 v Errors were found in the Annunciator Response Procedures and the Master Set point List (MSL); also the MSL did not contain all necessary information, that is set points.

Problems were identified with two of ten items checked, the Set point for saturation I subcooling margin, which was found at 15 F, this apparently is a nonconservatively low Set point because the Safety Parameter Display System Set point is established at 32 F.

The second item concerned the empty data base value for pressurizer level. Also, a general question of Set point control is open. j v Design Basis Summary (DBS) control is deficient. DBS have not yet been updated although they have collected DCNs for the last two years. A Gap Report was constructed. DBS do not exist for two Maintenance Rule (MR) systems (Group 1 & 2).

Charging Volume Control (CVCS), there is a DBS for the Safety injection portion. A DBS for Emergency Lighting was not written when included as a MR system. Also, there is no mechanism to update the DBS when the MR systems change.

v There is a large backlog of drawing changes, approximately 3,000, and plans to work off this backlog extend out two years. This issue was passes to the OSTI to evaluate.

v One Action Request (AR) under CR M3-97-0506 was closed improperly without completing all reviews. The action was to review administrative DCNs for changes that should have received safety evaluations, but did not. The review was to be made for the past 5 (or 7) years, the action was closed, however only one year was reviewed.

This is a Restart item, also.

1. A few areas of the ACR/CR closeout package were found not to fully address the l concerns. The documentation was upgraded satisfactorily during the inspection. l For example, However, there was no justification to eliminate some of these 13 which appeared to be generic to Unit 3. However, all of these issues were found to I be overlapped by actions in Unit 3 issues.

SIL ltem 41-2 also tracked as URI 95-81-01, concerned trending of NCRs. This issue is Closed. Based on the results of the review we are recommending to NRC management that the Sllitem be closed.

SIL ltem 73-1 also tracked as NOV 96-05-12, concerned failing to have an effective five year Technical Specification audit program. Based on the results of the review we are l

recommendirc to NRC management that the SIL item be closed. However, the Violation remains Open pending corrective actions regarding the audit planning and tracking matrix.

SIL ltem 73-2 aiso tracked as IFl 96-06-17, concerning an inadequate Oversight Program

! based on the 1996 Joint Utilities Management Audit. Our review was of selected areas of the audit, it was not our intention to follow-up all the issues within the JUMA audit, and of course we are not acting for JUMA in closing any of their findings. That is an independent organization which has its own audit follow-up and closure process. Based on the results 14 February 26,1998

l l

\

l Management Exit Meeting Millstone Unit 3 - Corrective Actions l February 26,1998 j of the review we are recommending to NRC management that the Silitem be closed.

A The team found that there has been an improvement within the traditional QA/QC funttion of the Nuclear Oversight Organization.

A The number of Auditors has significantly increased; their qualifications and knowledge I level has increased.

A Audit program procedures are acceptable; audits and audit checklists are acceptable.

There are four new audit managers.

i A Oversight has tile opportunity to concur with the corrective actions taken for audit j findings and NCRs . ]

i A There is good follow-up on Audit Findings, QA re-audits for an effectiveness review during the next audit.

A There is good interface with the line organization concerning the Nuclear Oversight Restart Verification Plan.

A The NRC team observed initiatives being taken by OC inspecto- during a hold point l inspection. j Safety Committees NRC team members attenced meetings of the Site Operations Review Committee (SORC),

the Unit 3 Plant Operations Review Committee (PORC) and the Nuclear Safety Advisory Board (NSAB) .

A All three safety committees operated effectively, the members were prepared for the meeting and added quality to the i.ssue being addressed. SORC added quality to issues under review during the meetmg that we attended.

v The Nuclear Safety Advisory Board (NSAB) qualifications matrix is an example of non-conservative interpretation and implementation of the technical specifications.

v TS 6.5.2.2 specifies that the Senior Vice President and Chief Nuclear Officer (CNO) is the SORC Chairperson; however the meetings have routinely been chaired by the Alternate Chairperson. The Chief Executive Officer (CEO) has been filling the CNO position organizationally, but not attending the SORC meetings. This caJses a conflict l with the TS requirements for composition of the SORC, and TS 6.5.2.7.b that the SORC will provide written notification of a disagreement between the SORC and the Senior Vice President and CNO.

[ 15 February 26,1938

l =

l 9

Management Exit Meeting Millstone Unit 3 - Corrective Actions l February 26,1998 v These two observations on administration of the safety committees along with the two issues of organizationalindependence are examples of where Section 6 of the Technical Specification was interpreted for expediency.

Independent Safety Engineering Croup j

! A The ISEG has performed high quality plant reviews and Operational Experience (OE) reviews.

v The number of ISEG reviews done in 1997 was only 12, down from 24 the previous year. This appears to be e result of the OE workload being performed by the group.

1. The ISEG group has reduced the backlog of OE issues from several hundred to l approximately 40 for Unit 3. However, the amount of work represented by the remaining issues is significant.

I c Site implementation and use of OE was mixed. OE is not consistently being used by the working groups at this time. A key reason is that the site wide procedure tn l establish expectations for the use of OE was not yet issued. Once issued the departments will still have to develop implementing procedures. An exception is Unit 3 Health Physics which has established access to the nuclear network safety database.

Conclusion i

That concludes my summary, as I've indicated several issues are under review currently.

For purposes of this inspection report, we will receive new information through the end of

! next week. Thank you for your cooperation. Jacque has a few overview remarks.

i l

16 February 26.1998

l i

. Millstone Unit 3 iP 40500 Corrective Actions inspection

i

l. I POTENTIAL ENFORCEMENT ISSUES: I l 10 CFR 50, Appendix B, Criterion Ill:

NGP 5.23 requires that the Master Setpoint List contain the setpoint and the calculation l reference. Problems were identified with two of ten items checked. First, the calculation l

which establishes the basis for the saturation subcooling margin annunciator set point was not available. In addition, it was stated to be 15 F, this is nonconservative low compared

to the Safety Parameter Display System set point established at 32 F. In a second issue, f- the value for pressurizer level setpoint was blank in the Master Setpoint List.

l Also, a general question of the adequacy of set point control is open. l I I Technical Specification Section 6 - Organization independence:

TS 6.2.1.d, the FSAR and Reg Guide 8.6 specify organizational independence for the Radiation Protection Manager. Contrary to these requirements, the Radiation Protection )

l Manager reports to the Maintenance Manager, a configuration specifically prohibited by the '

Reg Guide.

l TS 6.2.3.3 requires that the ISEG perform independent review to reduce human error, the iSEG has obtained the service of Human Factors practitioners who are assigned to the NSE .!

organization. Contrary to the TS the HF practitioners have been called upon to review the I work of their peer. I Failure to Follow Procedures - Design Basis Documents: l l

i NGP 5.28 requires that DCNs be used to update the Safety Functional Requirements i Manual.' The SFRM documents system characteristics for input to the FSAR Chapter 15 l analysis. Several revisions were not processed correctly, DCNs were not issued and l entered into the Generation Records Tracking System (GRITS).

! PI-29 requires Design Basis Summary (DBS) documents for risk significant systems. The i DBS have not yet been updated although they have collected DCNs for the last two years.

l A Gap Report was constructed. DBS do not exist for two Maintenance Rule (MR) systems (Charging Volume Control (CVCS) and for Emergency Lighting.

Failure to Follow Procedures - Corrective Actions Condition Reports:

Significance Level Classification RP-4, Corrective Actions, establishes criteria for Condition report significance level.

Several examples were discovered where classifications were lower than required.

March 11,1998

• l v -

Millstone Unit 3 IP 40500 2 Corrective Actions inspection Condition reports were issued because of incomplete action on GL 90-03, Vendor Technical Information Program, and GL 89-13, Service Water Fouling. These were classified as Level 3, an Enhancement item, instead of Level 2, a condition adverse to ,

quality. RP-4 requires that NRC commitment issues be assigned Level 1 or 2.

Additionally, the Condition reports were coded as ' Deferred' until after start-up, and included on the Deferred Items List submitted to the NRC.

The ISEG intervened in Main Transformer switchyard work for industrial safety and risk to off-site power supply reasons. However, the Condition report which recorded the problems was classified as level 3, an improvement item. However, a Level 3 Condition report lacks tracking and trending, which occurs for a level 2 Condition report.

A Level 1 Condition report (M3-97-3710),which had been written to identify significant deficiencies i6 the NCR process, was closed to a Level 3 Condition report. This was an improvement item written to track self-assessment findings and therefore not requiring action, tracking and trending.

This issue also demonstrated a corrective actions program weakness. The program has inadequate controls over combining condition reports to preserve the issues and maintain their appropriate Significance Level.

Condition Report Closure RP-4, Corrective Actions, requires effective actions to prevent recurrence.

A CR, M3-97-2943 concerning air binding of the Boric Acid Transfer Pumps [3CHS-P2A and -P28] was closed with essentia5y the same corrective actions taken earlier. There have been a series of five events of this type from 1992, three of which occurred in 1997.  !

This issue also demonstrates inadequate root cause determinations, inadequate l deportability evaluations, inadequate consideration of operating exp_;ience in the evaluation l of the problem. Additionally, NU failed to recognize the potential unreviewed safety l l question (USO) which resulted from their assumptions of a non-conservative Boric Acid i Tank level technical specification. l I

RP 4, Step 1.12.4, states that actions are to be closed out "when assignment is j complete." Contrary to that requirement, one Action Request (AR 97003960-5)under CR  !

l M3-97-0506 was closed improperly without completing all corrective actions. The two

! actions that were not complete were: performing an MSEE reconciliation of 199 DCNs that  ;

document as-built conditions, and reviewing the 142 DCNs that initiated work and then i generated an MSEE, MMOD, or DCR as appropriate.

Completion of this action was an ACR 7007 and CMP restart commitment.

Also, procedure WC1, Unit 3 Work Management, Section 1.8.7, requires completion of March 11,1998

Millstone Unit 3 IP 40500 3 Corrective Actions Inspection work activities prior to closure of an automated work order. Contrary to that requirement the AWO to correct service water flow instrument anomalies for 3SWP-F1-059, A, B, and C, was closed prior to completion of all the specified work. This action was taken to correct operator work-around 96-03.

Other Significant Findings that are Considered Unresolved issues:

Long term compensatory measures are in effect for fire protection systems because surveillance testing which verifies operability of these systems has been suspended.

Compensatory measure are taken to allow restoration of a fire protection system. But in this case, are being used to substitute for long term system inoperability. The team found that the failure to conduct the surveillance was inappropriate because it sets a low standard of performance for plant personnel. The tea.n also found that the reliance exclusively on the 1-hour roving fire watch as an interim compensatory measure without other compensatory measures was a weakness, as noted in Information Notice 97-48, inadequate or Inappropriate Interim Fire Protection Compensatory Measures.

Corrective actions to verify that 480 volt molded case circuit breaker trip settings are appropriate for their motor load , has been delayed for up to eighteen months following plant restart. This action originated with an Engineering Department self assessment that found a sample of setpoints to be incorrect. The thirty three unverified breaker load trip settings that are being delayed following plant restart include emergency diesel generator support systems.  ;

I The ISEG reviews of plant operation have dropped from 24 in 1996 to 12 in 1997. The ISEG group has reduced the backlog of operating events significantly. However, the amount of work represented by the remaining issues is significant, and several may identify safety issues.

NRC Team Observations:

Management Directions, Goals, Expectations:

Management communications methods were a strength.

There was a common understanding of management = s expectations by plant personnel.

Strategic plan and vision statement on where the plant is headed are in draft. This is a weakness.

l l Overall, the Nuclear Group Policies and Standards, were considered good; the lack of a Nuclear Group Mission and Vision statement, under development, was a weakness.

March 11.1998

)

I L

i i

Millstone Unit 3 IP 40500 4 Corrective Actions inspection Organizational Communications and Teamwork:

Communications were adequate relative to identifying safety issues.

Although, teamwork initiatives at the first line supervisor and above were developed, strongest at the upper levels. There is a need to be extend this to the worker level. For example:

Not everyone familiar with conflict resolution Sometimes interpersonal conflicts don =t get resolved No formal process for rotational assignments.

Management and Supervisory Encouragement of Problem Identification:

Observations and interviews show that managers and supervisors encourage employees to identify problems.

The plant staff feels that management is receptive to problems brought forward, and individuals generally characterized the environment as improved and currently receptive to problem identification.

l Safety Conscious Work Environment Programs:

Individuals generally characterized the environment as improved and currently receptive to problem resolution. However, interview results indicate that some plant staff may not fully trust plant management. They have expressed a Await-and-see view of management.

They therefore pursue other alternatives to rasing problem rather than with their management.

There is no reluctance or reservations expressed by individuals to identify problems through the Corrective Action Condition Report Process, to ECP or the NRC.

The SCWE processes have not yet been formalized. The lack of SCWE program structure,

. that is procedures, formal processes and documentation requirements, resulted in many of the management actions being handled directly, on an ad hock basis, by Recovery Managers and senior managers.

The handling of individual HIRD cases by the Employee Concerns Program and the Safety Conscious Work Environment program is adequately responsive to specific case needs.

Both technical and human-side problems are generally well addressed.

The ECP case intakes, the Employee Concerns Oversight Panel (ECOP) oversight activities

! and surveys used to identify potential or actual HIRD problems or organizational units l which exhibit barriers to free identification and reporting of problems are positive contributions to the overall process. These mechanisms are effective especially for the more egregious issues identified as problem areas.

March 11,1998 i

t L

Millstone Unit 3 iP 40500 5 Corrective Actions inspection Documented action plans are used to define and manage the remedial actions for Problem Areas. Sampling determined that the plans are generally effective at remediating both the technical and human performance and behavior issues.

Other HIRD issues that are perceived by NU management to be less egregious or are minor problems having the potential to become Problem Areas are handled less formally by senior and middle management. These issues are identified by line management via normal management oversight activities, review of ECP intakes, and SCWE daily meeting which include dialog with ECP, ECOP and other program participants. No documented action plans are used but the team found that management had taken actions in response to such HlRD potentials, and were monitoring their effectiveness.

However, NU has not been effective in preventing the ongoing emergence rate of HIRD allegations to ECP. The incident rate of HIRD allegations and management-related or induced chilling effect events has not diminished significantly. Except for specific HIRD ECP cases or explicitly identified SCWE Problem Areas, management does not appear to be taking actions focused to reduce the overall incidence of HIRD allegations.

The Corrective Actions Program:

Conclusion Although we have seen evidence of a corrective action program beginning to function, it is clear that the program will require careful monitoring by NU. We have found a disproportional number of issues in a relatively small sample size of CR. Our observations were made following extensive review of the corrective action process by your self-assessment team which reviewed 275 Level 1 condition reports prior to our inspection.

l In addition to those listed above as potential enforcement issues, the NRC team had the j following observations:

Classification of Corrective Actions NU has recently begun using an additional risk significance classification for condition reports. However, this classification is not tied to the PRA, Maintenance Rule risk significance or information from the IPE. It is based on industrial safety and production factors.

Examples of Poor Review of Condition Reports  !

CR M3-97 2898, dated September 2,1997, documented that a nuclear oversight audit identified procedures, tools, and equipment needed to support emergency operating procedures were not available in the plant. The corrective actions for this CR were not tied to a key event and were outside of the scheduled restart date at the time they were approved.

March 11,1998 i

L

Millstone Unit 3 IP 40500 6 Corrective Actions inspection CR M3-97-3974, dated November 11,1997, documented an audit finding that the ISEG/OE procedure, NOOP 3.04 was not reviewed by SORC as required by technical specifications. The ISEG response was to get SORC review of the new OE procedure  ;

which was being developed. However, this response was inadequate because the ISEG procedure also needed SORC review by the same technical specification.

CR M2-98-0419 documented seven pieces of tape that were identified by the NRC during  ;

a QC foreign materialinspection of the unit two spent fuel pool. The CR was assigned to '

reactor engineering as a technicalissue and was not evaluated by OC as an inspector performance issue.

The NRC team found several examples of narrowly focused or waved Root cause analysis.  !

ACR M3-97-0558, documented that the Chemical and Volume Control System had non-conservative assumptions related to maximum temperatures for the ietdown heat exchanger and charging flow. The root cause was waived and referred to other condition reports with related similar situations. The inspectors regard this waiver as a missed opportunity to thoroughly evaluate the issue. However, the corrective actions appear to adequately address the design deficiency.

Anoe instance involved ACR M3-97-0409,that documented concerns for sump water leve alculated head losser. Although the cause of the event was addressed in an LER, no root cause was performed for the CR. Nevertheless, the proposed corrective actions appear adequate to resolve identified design deficiencies and a modification review is scheduled following reanalysis with actions to be completed prior to mode change.

The NRC team also found examples where corrective actions for condition report were narrowly focused. For example, there was a lost opportunity for earlier detection of material deficiencies in High Energy Line Break (HELB) doors because of narrow corrective actions for condition reports which documented earlier issues of HELB doors not being fully closed.

Self Assessments  !

The team found that the self assessments were generally of high quality. You have taken l a strong initiative in areas like the Nuclear Oversight Restart Verification Assessment. .i However, we have observed the need for continual improvement in assessments performed by the line departments. Continued attention should be applied to the line departmental self assessments in order to improve their quality.

Independent Oversight The .NRC team found that there has been an improvement within the traditional A/QC function of the Nuclear Oversight organization's Audits and Evaluation Group. The number March 11.1998

Millstone Unit 3 IP 40500 7 Corrective Actions inspection of Auditors has significantly increased; their qualifications and knowledge level has increased. Audit program procedures are acceptable; audits and audit checklists are acceptable. There are four new audit managers. Oversig.it has the opportunity to concur with the corrective actions taken for audit findings and NCRs. There is good follow-up on Audit Findings, A re-audits for an effectiveness review during the next audit. There is good interface with the line organization concerning the Nuclear Oversight Restart Verification Plan. The NRC team also observed initiatives being taken by OC inspectors during their in plant activities.

However, the Audits and Evaluation Group's Audit Commitment Database, is presently configured, does not meet the needs of the Audits and Evaluation Group, independent Safety Engineering Group The NRC team found that the ISEG has performed high quality plant reviews and Operational Experience (OE) reviews. However, the number of ISEG reviews done in 1997 was only 12, down from 24 the previous year. This appears to be a result of the OE workload being performed by the group. The ISEG group has reduced the backlog of DE issues from several hundred to approximately 40 for Unit 3. However, the amount of work represented by the remaining issues is significant.

Safety Committees NRC team members attended meetings of the Site Operations Review Committee (SORC),

the Unit 3 Plant Operations Review Committee (PORC) and the Nuclear Safety Advisory Board (NSAB). All three safety committees operated effectively, the members were prepared for the meeting and added quality to the issue being addressed.

However, the team observed two examples of non-conservative interpretation and implementation of the technical specifications. The first concerned the NSAB member qualifications matrix, credit was taken for an alternate member because no regular member possessed metallurgy experience. The second example concerned TS 6.5.2.2 specifies that the Senior Vice President and Chief Nuclear Officer (CNO) is the SORC Chairperson; owever the meetings have routinely been chaired by the Alternate Chairperson. The Chief Executive Officer (CEO) has been filling the CNO position organizationally, but not attending the SORC meetings, as assumed within the technical specifications. This causes l- a conflict with the TS requirements for composition of the SORC, and TS 6.5.2.7.b that the SORC wW - < ovide written notification of a disagreement between the SORC and the Senior Vice President and CNO.

The NRC team considers these two observations on administration of the safety committees along with the two potential violations concerning organizationalindependence are examples of where Section 6 of the Technical Specification was interpreted for expediency.

March 11,1998 I

l t

Millstone Unit 3 IP 40500 8 Corrective Actions inspection Significant items List Issues The NRC team reviewed the NU actions taken on several SIL items. The first, SIL Item 41-1 concerned the findings of the root cause team for FSAR and licensing basis inaccuracies ACR 7007, and two items related to M3-97-1839 and ACR 13302. The team found that NJ had taken appropriate action to address this issue. However, several new open items were identified. Three were previously identified as potential Violations concerning design basis document control and setpoint control. One additional observation concerns uncompleted drawing changes. There is a large backlog of drawing changes, approximately 3,000, and plans to work off this backlog extend out two years. This issue was passes to the OSTI to evaluate. The NRC team concluded that based on the results of this review, to forward a recommendation t o NRC management that the SIL item be closed.

In addition to ACR 7007, the NRC team reviewed the process for UFSAR changes. That process was found functioning, and has been processing changes to the Safety Analysis.

Additionally, NU actions in response to SIL ltem 41-2 was reviewed by the NRC team.

This issue is also tracked as URI 95-81-01, and concerned trending of NCRs. The team found a process in place that adequately performed this fenction. Based on the results of the review the team is recommending to NRC management that the SIL item be closed.

SIL ltem 73-1 is also tracked as NOV 96-05-12, and concerned failing to have an effective five year Technical Specification audit program. Based on the results of the review the NRC team is recommending to NRC management that the SIL item be closed. However, the Violation remains Open pending corrective actions regarding the audit planning and tracking matrix.

SIL ltem 73-2 is also tracked as IFl 96-06-17, concerning an inadequate Oversight Program I based on the 1996 Joint Utilities Management Assessment (JUMA). The team reviewed selected areas of the audit, it was not the team's intention to follow-up all the issues within the 1996 JUMA, or act for that organization in closing any of their findings. That is an independent organization which has its own audit follow-up and closure process. Based on the results of the review the NRC team is recommending to NRC management that the SIL item be closed.

{

March 11,1998

d Inspection Report input Garmon West, Jr.

March 11,1998 1.0 MANAGEMENT PROCESSES AND SYSTEMS The team evaluated the processes and systems that Millstone Unit 3 managers use to identify, correct, and prevent problems. The team reviewed management directions, goals, and expectations; organizational communications and teamwork; managerial and supervisory encouragement of problem identification; nnd performance monitoring. The inspectors conducted interviews, attended meebngs, and reviewed licensee documents. The interviews were with all levels of plant personnel, including senior managers, middle managers, supervisors, and nonmanagerial personnel.

1.1 Management Directions, Goals, and Expectations

a. Inspection Scope (40500) (0350-C.2.1.a, C.2.2.a, C.3.1.c)

The team assessed the effectiveness of the process by which Northeast Utilities management provides the necessary direction to prevent problems to the plant staff. The team evaluated the organization's high-level goals and expectations.

l. b. Observations and Findings

Licensee management's most formal direction to plant personnel is a strategic plan (using a l top-down approach) and an associated picture that indicates where the plant is headed.

Both of these items are currently in draft form. The ficansee explained that it expects to i issue its strategic plan after recovery efforts are completed for Units 2 and 3. The l licensee has issued a draft "Long-Term improvement Plan" (which uses a bottom-up approach), along with vision, mission, and strategic focus area statements, as part of its periodic report to the NRC titled " Progress Toward Restart Readiness and Long-Term l Improvement at Millstone Station - Northeast Utilities Briefing for the U.S. Nuclear l Regulatory Commission," dated February 11,1998. The licensee nas also issued " Nuclear l Group Policies and Standards." The " Policies and Stnadards" document lacks a " Nuclear i Group Mission and Vision" statement because it is still being developmed. The tea team l considered the draft status a weakness of the otherwise good document. The team also considered it a weakness of management direction that the strategic plan is currently in draft form because such a plan helps to guide both activities and decisionmaking.

Licensee management has used numerous means to communicate and reinforce its expectations of plant personnel. The various means included " Nuclear Group Policies and Standards;" a daily newsletter, which publishes items of interest, including operating experience items; posters enumerating management's expectations, which were visible throughout the plant; and daily, weekly, and monthly meetings. The information covered in daily meetings included condition reports, emerging issues, plant status, management expectations, organizational changes, plant modifications, and priorities. The team observed that daily meetings showed interdepartmentalinteractions, a questioning attitude by participants, and command and control by senior managers.

l s

A, . 1

2 The interview results showed a common understanding of management's expectations pertaining to identifying and correcting problems and identifying and addressing safety issues. Plant personnel at different levels of the plant organization demonstrated a high degree of compliance with areas of management's expectations, especially in the area of problem identification,

c. Conclusions The team concluded that management's long-term direction of plant personnel needed improvement because the strategic plan (which represents long-term direction) was in draft form. The team concluded that the plant staff's clear understanding of management's expectations was considered a management strength.

1.2 Organizational Communications and Teamwork

a. inspection Scope (40500)(0350-C.2.1.g,C.2.2.f)

The tean evaluated organizational communications and teamwork, including interdepartmental relationships and interfaces. The team assessed both vertical and lateral (horizontal) communications. It verified that communications are adequate to properly identify and characterize safety-significant issues. It also assessed whether communications between organizations were adequate to properly address safety issues,

b. Observations and Findings (1) Organizational Communications (a) Interdepartmental Relationships and Interfaces The team found that interdepartmental relationships and interfaces needed improvement (e.g., two departments had past and current conflicts, that is, maintenance and oversight).

The team also found that senior plant management was aware of these conflicts and had 1 not only taken effective immediate actions but also had identified long-term actions to resolve the conflicts. '

(b) Vertical Communications The team found that vertical communications were a strength of the plant organization.

Interviews indicated that vertical communications were especially strong with respect to using the corrective action system and sending constructive inputs up the chain of l command. The team also found that upper management's ability to effectively l communicate its expectations down the chain of command was a management strength.

I l

(c) Horizontal Communications The team found that horizontal communications were not as effective or as free flowing as vertical communications. The team found that communications within groups were more effective than communications between groups. Interviews indicated that communications

3 between groups were greatly dictated by the necessity of completing a task and communications were more free flowing when individuals had some prior relationship with one another. Some impediments to horizontal communications included the following:

time constraints, busy schedules, the fact that communications are often issue driven, the need to resolve an issue, cultural issues, and respect issues.

(2) Teamwork The team determined that team-building initiatives had been completed beginning with officers, then directors, followed by managers and first-line supervisors. The team saw evidence of teamwork in several regularly scheduled interdepartmental meetings. Interview results indicated that teamwork initiatives are currently on hold until after startup of Unit 3.

Interview results did not identify any teamwork initiatives at the level of the plant worker.

Interviews indicated a need for teamwork initiatives at the level of the plant worker for the following reasons: some groups are in conflict (most notably oversight and maintenance),

not everyone is familiar with conflict resolution, sometimes interpersonal conflicts are not resolved, and no formal process exists for rotational assignments. 1 The team concluded that communications between groups and departments needed to be mproved. The team concluded that team work initiatives at the level of the first-line cupervisor and higher need continued reinforcement. The team also concluded that l teamwork at the level of the plant worker level needs to be improved.

I i 1.3 Encouragement of Prclem identification by Managers and Supervisors j a. Inspection Scope (40500) (40500-02.03.c)(0350-C.2.1.f)

The team evaluated whether managers and supervisors encourage employees to identify problems and whether the staff believes that management is receptive to the problems being brought forward.

l

b. Observations and Findings l Observations, interview s and plant surveys show that managers and supervisors l encourage employees to identify problems, interviews also indicated that the plant staff believes management is receptive to problems being brought forward.

l

c. Conclusions The team concluded that plant management was effective in its efforts to encourage plant j personnel to identify problems. 1 1.4 Performance Monitoring I

a. Inspection Scope (40500) (40500-02.03.d,02.04b,02.05d,02.06d)(SIL-41,0350- l C.2.1.h) j i

l 4

The team evaluated performance monitoring (performance indicators), including management information systems, employed to evaluate the following programs:

coirective action, root cause analysis, self-assessment, independent oversight, and operating experience. The team evaluated the effectiveness of the performance measures process. The team also assessed the quality of the information on performance that is given to management. The team verified that the licensee takes action when any of the performance indicators identify areas that should be addressed.

b. Observations and Findings The various performance indicators used to evaluate the subject programs were considered j excellent. The licensee has appropriately addressed adverse trends identified in the fourth I quarter report concerning compliance of the maintenance department with procedores, valve and breaker alignment issues and tagging errors of the operations department, and surveillance testing. The team's analysis of licensee event reports (LERs) found that Ur'it 3 had 16 human performance related LERs in 1997 versus a national average of six human perfromance related LERs. The team determined that the licensee intentionally does not perform five fire protection surveillance because the operations department does not have the manpower to conduct the surveillance. The following condition reports (CRs) have been written in conneciton with the surveillance: M3-97-303 5, M3-97-3981, M3 4246, M3-97-4394, and M3-97-4618. The licensee has implemented hourly patrols as a compensatory measure. The licensee stated that the subject fire protection surveillance would be performed by Station Fire Brigade personnel when they are trained. The team found that the failure to conduct the surveillance was inappropriate because it sets a low standard of performance for plant personnel. The team also found that relying exclusively i on the 1-hour roving fire watch as an interim compensatory measure without other compensatory measures was a weakness, as noted in Information Notice 97-48, l " inadequate or inappropriate Interim Fire Protection Compensatory Measures."

c. Conclusions

! The team concluded that the licensee's performance monitoring program was good. The team also concluded that the high number of LER-related human errors was a weakness  ;

that the licensee needs to examine further. The team concluded that the licensee's failure j to conduct five fire protection surveillance and exclusive reliance on the 1-hour roving fire I watch were inappropriate fire protection compensatory measures.

l 1.5 Management's Commitment To Resolve issues l

a. Inspection Scope (40500)(40500-02.06.d)(0350-C.2.1.c,C.2.1.d)

The team evaluated management's commitment to resolve safety committee recommendations, audit findings, assessment recommendations, and open issues.

b. Observations and Findings (1) Resolving Safety Committee Recommendations

E 5

Plant Operating Review Committee (PORC) Number 3-97-256 contained six

, recommendations, including preparation of a temporary modification and safety evaluation I report pertaining to the repair of a valve. The staff determined that menagement had addressed all of the subject recommendations.

(2) Resolving Audit Findings Audit Report (AR) MP-97-A06-02 noted one deficiency regarding reporting by the Unit 3 Radiation Protection Manager (RPM) to the Maintenance Manager. Unit 3 Final Safety Analysis Report (FSAR) Section 12.5.3 states that ail health physics procedures and methods for ensuring that occupational radiation exposure is as low as reasonably _

achievable (ALARA) follow the provisions and suggestions of Regulatory Guide (RG) 8.8, l Revision 3; RG 8.10, Revision 1-R; and RG 1.33, Revision 2, as applicable. RG 8.8,

( Section C.1.b(3), states, in part, as follows:

l "The Radiation Protection Manager (RPM, (onsite) has a safety function and responsibility to both employees and management that can best be fulfilled if the individual is independent of station divisions, such as operations, maintenance, or technical support, whose prime responsibility is continuity or improvement of station operability."

Millstone Unit 3 Technical Specification Section 6.2.1.d states, in part, as follows:

"The individuals who train the operating staff and those that carry out health physics and quality assurance functions may report to the appropriate onsite manager; however, they shall have sufficient organizational freedom to ensure their independence from operating pressures."

The licensee closed CR M3-97-1875, noting that the Unit 3 RPM has a direct reporting capability to the Unit Director on radiological issues as noted, by asterisk, on the organization chart. The team believes that having the RPM report to the Maintenance Manager is inconsistent with the Unit 3 technical specifications. The licensee stated during the onsite inspection that this issue would be evaluated along with other organizational changes that are currently being considered. Therefore, the team considers this item to be an open issue pending resolution (unresolved item 50-423/97-82-01).

I (3) Resolving Self-Assessment Recommendations i

Self-assessment Number 3TS-SA-97-02 involving Unit 3 reactor engineering procedures made five recommendations. The team found that the licensee had addressed each of these recommencatiions.

(4) Resolving the Open Issue on Independent Verification That Human Errors llave Been Reduced Section 6.2.3.4 of the Unit 3 technical specifications made the following statement:

~

i l

l 6 l "The ISEG [lndepenedent Safety Evaluation Group) shall be responsible for maintaining surveillance of unit activities to provide independent verification that these activities are performed correctly and that human errors are reduced as much as practical."

The licensee's memorandum dated December 23,1997, made the following statement:

"In the event that the HF/E [ human factors / engineering) review effort needs to be audited or become a part of a " Nuclear Oversight" assessment, an independent party will be assigned to ensure that independence is maintained. HF/E personnel may participate in Nuclear Safety Engineering ISEGs and in Nuclear Oversight audits and assessments that do not include previous HF/E involvement in design changes, MCB

[ main control board] changes, procedure reviews, etc."

j The team believes that the licensee's position on this issue is inconsistent with Technical Specification 6.2.3.3. Therefore, the team considers this item to be an open issue j pending resolution (unresolved item 50-423/97-82-02). '

(5) Resolving Open issue on Establishing or Applying Appropriate HFE Guidance for Advanced Computer-Based Displys in the Control Room The team determined that Specification SP-EE-140A details the design requirements for the Safety Parameter Display System (SPDS). The team found that minor modifications to the SPDS have been made over the years. However, the guidance references in the specification are out of date in that they do not contain current guidance on computer-based displays. The licensee stated that AR Number 97027958 titled " Review of and Update of Plant Process Computer Specifications" would update the guidance references by December 1,1998. The team found the licensee's commitment acceptable.

(6) Resolving Open issue Regarding the Post Loss-of-Coolant Accident Cooling Status l Tree The tearr. determined thrst the post-LOCA cooling (PLC) status tree is not addressed in the Emergency Operating Procedure (EOP) User's Guide, Section 1.6, " Monitoring Status Trees," or Attachment 4, " Control Room Usage of Status Trees." The licensee stated that this issue would be addressed by AR Number 07031064 titled "PORC Commitment to investigate the Overall Feasibility of the Post-LOCA Processing" by December 1,1998. 1 The team found the licensee's commitment acceptable.

The team determined that the navigation scheme for the PLC tree was not the same as for i other trees and did not agree with the SPDS specification. The licensee provided l documentation to show that the SPDS specification (i.e., Section 6.6.6 of SP-EE149A, Revision 6) had been changed to match the navigation scheme for the PLC tree. The team l

1 found this change to the specification acceptable for closing this open issue.

(7) Resolving Engineering Review Recommendations

l

.. j

. . \

l 7

l The team evaluated the effectiveness of the licensee's management team to resolve  !

recommendations made by Engineering Review 3-ESAR-97-008. I One recommendation concerned revising NGP [ nuclear group plan) 5.25 and SP-EE-261.

The licensee documented that the revision to SP-EE-261 will be completed by March 30, i 1998.through AR 96000103 and willinclude a section of standards for computer displays and an update pertaining to Revision 1 of NUREG-0700. The licensee also documented that the revision to NGP 5.25 will be completed of AR by June 30,1998 which includes a {

l change to NGP 5.25 that would require the Human Factors Specialist rather than the ]

Project Engineer to determine whether a detailed control panel design review is warranted and an update pertaining to Revision 1 of NUREG-0700. Another recommendation was to add another Human Factors Specia!ist in view of the workload. The licensee rejected f1 the proposed increase in the human factors staff.

The engineering review also raised the question of whether there was a match between the simulator and the control room designs. The licensee's documentation noted the following strengths of the simulator update process:

"At this time, there are no design changes with simulator impact that have been instaHed in the plant for (more] than 30 days that have not been incorporated in the simulator."

l "To support the current restart training needs, we have modified our target for incorporating those plant design changes identified by Operations and Operator Training ,

as having simulator impact to have them installed within 30 days of plant installation. )

This is beyond what is reovired by ANS-3.5, the standard to which the Millstone 3 I simulator is certified to. The standard allows 24 months to incorporate such changes." l l

The licensee's self-assessment (97-004) identified the following weakness of the simulator I update process: j l

"This assessment has demonstrated the absolute need for coordination between site and the simulator support group to insure that all plant modifications are accounted for in the simulator upkeep."

Overall, the team found that the simulator process, which ensures that the fidelity of the simulator is maintained with regard to the reference plant, was considered a strength.

The team found that there have been several advanced systems added to the control room without appropriate consistent design guidelines (e.g., Foxboro intelligent automated (IA) system for the moisture separator reheaters (MSRs), fire protection, enviornmental j qualification (EO) temperature monitoring, the aut# log system, and SPDS upgrades). l There are currently eight different computer-based systems in the main control room.

Since many of these systems are significantly different in their human systems interf ace, j displays, and alarms,it was the team's observation that the licensee should evaluate these l different systems with respect to unnecessary operator burden and the potential for l l increasing operator error. Further, it was the team's observation that such an evaluation l should include the criteria in the licensee's revision to SP-EE-261. I 1

l l

l

se 8

c. Conclusions Overall, the team concluded that management's commitment to resolving safety committee recommendations, audit findings, assessment findings, and open issues was l satisfactory. However, the licensee's failure to resolve resolve the two open technical specification related issues concerning the independence of the RPM and the independence of the ISEG relative to reducing hu. man errors was considered a weakness.

1 l

l I

i

$ DRAFT '

. D. Beckman Millstons - IP 40500 R:: port input June 26,1998 I

1. Management's Encouragement and Receptiveness in Problem identification by individuals

a. Objec2ives:

i. Evaluate the NU problem resolution processes (line management practices, corrective action program and others) to determine their effectiveness providing a vehicle for encouraging identification of, i readily accepting and adequately resolving problems and issues identified by individual employees as pathways preferable to ECP or NRC allegations. (SPO Oversight Program Plan, 9/10/97) ii. Determine if managers and supervisors encourage employees to identify problems. Determine if the staff feels management is receptive to problems being brought forward.. (MGMT-3)

b. Scope:

The team conducted employee interviews at all levels of the organization to determine the worker's perceptions of management's efforts and communications intended to enhance problem identification. Performance measurement data for problem identification and documentation associated with individual-identified problems (Condition Reports, Employee Concern Case Files, Self Assessment results and others) were reviewed

c. Observations and Findings

i. bterview results indicate no major barriers to problem identification l (1) Interviewed personnel from organizational units both with and without histories of HIRD or other barriers to problem identification.

(2) Individuals generally characterized environment as improved and currently receptive to problem identification.

(3) No reluctance or reservations expressed by individuals re identification of problems to line, CR process, ECP or NRC.

ii. Reviewed ECP Program, SCWE Progra and management initiatives for NU activities and response actions taken for prior HIRD problems above.

(1) IP40001 found that SCWE processes not formalized. Same results here. Resulted in many of the management actions I being handled personally by Recovery Officers and senior managers.

(2) Handling of individual HIRD cases by ECP, SCWE, appears adequately responsive to specific case needs. Both technical and human-side problems are generally well addressed.

(3) However, NU is not effective in addressing high incidence rate of HIRD allegations in ECP. Incidence rate of HIRD allegations DRAFT l

3

e. - o,

1 i DRAFT

. D. Beckman Millstone - IP 40500 Report input June 26,1998 and management-related or induced chilling effect events has not diminished significantly.

iii. Performance indicators and CR program statistics reflect adequate i

levels of CR program participation.

l l

l' l

l r

i t

l DRAFT 1

~

DRAFT

, D. Beckm:n Millstona - IP 40500 R:: port input June 26,1998

2. Effectiveness of Problem Resolution Processes in Resolving Employee identified Problems

a. Objective:

Assess the effectiveness of the process by which NU management provides direction to the plant staff necessary to prevent problems. Evaluate the organizations high level goals and expectations. (MGMT-1) ,

b. Scope:

I The team conducted interviews, reviewed corrective action program data, and reviewed l the licensee's technical resolutions, SCWE-related responses, and long term follow-up j for problems identified by condition reports, employee concems and Employee Concems Oversight Panel issues, and NU department self assessments. This included a sample of condition reports and their resolution activities that resulted from employee concem cases, self assessments,

c. Observations and Findings Review of ECP Cases, SCWE Problem Area response plans, and sample of CR corrective actions determined licensee actions generally responsive to probWms and adequate with some exceptions:

i. SCWE Problem Areas and Potential Problem Areas (1) Identification methods for HIRD issues include the NU Leadership Surveys, Culture Surveys, ECP case intakes, and ECOP identifications and referrals. These mechanisms appear to be effective, especially for the more egregious issues identified as " Problem Areas".

I (2) Documented actions plans are implemented for the Problem Areas. Sampling review determined that action plans are generally effective at remediating both the technical issues and human performance and behavior issues.

(3) Other HIRD issues that are perceived by NU management to be less egregious or minor problems having the potential to eventually become Problem Areas are handled more informally by senior and middle management.

(a) Licensee is monitoring a number of organizational units and activities which pose the potential of becoming Problem Areas.

(b) Monitoring is being done by line management via morning normal line activities, review of ECP intakes, and SCWE daily meeting which include dialogue with DRAFT

DRAFT

. D. Beckman Millstons - IP 40500 Report input June 26.1998 l

ECP, ECOP, HR, Legal, LHC and other program

! participants.

(c) No documented action plans are used but management demonstrated that actions have been taken in response to such HIRD potentialities.

(4) Actions taken to date for HIRD-related problems has been unsuccessful in reducing the emergence rate of HIRD allegations to ECP.

(a) An average of about 54% of all ECP allegations included HIRD and about 26% of the total involved HiRD + 10 CFR 50.7 nuclear safety related protected activities. These occurrence rates appear to be l continuing with the 2/98 data.

l (b) NU has performed extensive re evaluation of the data to ensure their categorization is correct with no i

substantive changes made in results.

l (c) Ooif three 1997-98 ECP cases have involved confirmed 10 CFR 50.7 HIRD issues (and they all involved the same organizational area of MOV activities).

l (d) The fraction of the total cases not involving 10 CFR

' 50.7 typically involve other forms of HIRD, e.g. age, sex or other discrimination, harassment or intimidation over work rules, compensation, benefits, etc.

(e) NU acknowledges that HIRD in non-50.7 environments j represents a real and present potential of 50.7 HIRD.

(f) Discussed management actions planned /taken with M.

Brothers, D. Amerine, M. Gentry, A. Elms, J. Streeter, and others. Except for specific cases or explicitly identified problem areas, management does not appear to be taking actions focused to reduce the overall incidence of HIRD allegations.

I- 11. Corrective Actions for Chronic Performance or Programmatic issues / Trends are Less Than Adequate (1) M&TE Program issues (The Fedex with my Rites never arrived on Thur-Fril WMf write up l from my notes ifit doesn't arrive in a\me but renty need the packspel lii. Adequacy of Corrective Actions / Technical Resolution (1) BA Transfer Pump - CR M#-97-2943 dated 9/4/97 (a) BA Transfer Pumps (3CHS-P2A & -P28) have history of air binding bedly enough to render them inoperable.

DRAFT l

DRAFT .

D. Beckm n Millstons - IP 40500 R:: port input June 26,1998 (b) Previously identified in 1992 (PIR 3-92-210); 1995 (ACR 3617); and earlier in 1997 (CR M3-97-0715, CR M3-9701011, CR M3-97-0954) but not corrected.

(i) NU concluded that air entrainment caused the 1992 event, vented the pump and required no further action as it was considered an isolated occurrence.

(ii) No cause was determined for the 1995 event other than the presence of trapped air in a horizontal pipe run. Internals of the BAST batch tank discharge check valve were removed and a modification was proposed to add an isolation and manual vent valve to provide improved venting were proposed.

(c) CR M3-97-954 identified problem and recurrent and initiated actions to:

(i) Revise BAST low tank level operating limits and alarm setpoints upward to prevent air entrainment.

(ii) Revise system OP to vent pumps if BAST levels drop below the increased low limits.

(iii) Complete the mod proposed in 1995.

(iv) Correct an erroneous root cause analysis.

(d) The corrective actions for M3-97-2943 were essentially j the same as above except that they included evaluation of a mod to re-route the Boric Acid Gravity Boration Piping.

(e) Non-Conservative Technical Specification (i) TS (#Later) require BAST's to be maintained

>23,620 gals total for Modes 1-4 and >6,600 gals for Modes 5-6.

(ii) M3-97-2943 recommends administrative controls to maintain individual tank volumes >

14,000 gallons (>28,000 gals total) to prevent BA Pump inoperability.

(iii) Leads to conclusion that TS are non-conservative for BA Pump air entrainment resulting from suction tap elevation in tanks.

(iv) Appears to meet test of 50.59 for USQ (v) NU had planned to submit routine TS change request to NRC and operate IAW administrative limits pending that change.

(vi) Discussed with D. Smith, ll3 Licensing Mgr, who did not agree that TS non-conservatism required declaration of USO.

DRAFT w___-____ .. . - - - _ _ _ -

$ DRAFT

. D. Beckm:n Millstona - IP 40500 Report input June 26,1998 (vii) CR engineering evaluation activites and root cause analyses relied heavily on inspection of piping, review of drawing dimensions, UT of airbound piping, and engineering deduction to determine causes (confirmed by CHS System Engineer P. Tirinzoni). No hard data available to confirm actual source of air.

1) CRs cross-reference other cases where charging pumps also cavitated due to apparent air binding. (Turned OE issue 1 over to Rick Rasmussen)

2) Did not consider possible migration of j hydrogen generated by CHS/HHSI pump mini-flow recire orfices as

{

reported by INPO SOER 97-01 and Beaver Valley. l

3) Piping configuration appeared to i support such migration.

4) CHS System Engineer not even aware of prior OE or phenomenon of H2 generation.

(2) 480V Molded Case Circuit Breaker Magnetic Trip Setpoints (CR M3-97-3095)

(a) NU Erigineering conducted a very effective self assessment of the " Control of Magnetic Trip Settings on 480 VAC Motor Starters" and issued the CR to document the findings. (That part was GOOD).

(b) Self assessment found a sample of setpoints to be incorrect. Proposed corrective actions were generally comprehensive but required verification of motor nameplates to determine actual required trip settings.

(c) Walkdowns were deemed impractical for a "large' (but undefined) number of breakers due to their difficult accessibility.

(d) As an attemate to the waikdowns, NU determined that most of the erroneous setpoints occurred when motors l were changed with new operating currents but the  ;

settings were not appropriately adjusted. 1 (e) Consequently, NU reviewed motor maintenance files and confirmed that none of the unverified motors had been changed. This caused them to conclude that the setpoints are probably correct and further confirmation is not needed prior to restart.

DRAFT

(_

DRAFT s D. Beckman Millstona - IP 40500 Report input June 26,1998 1

(f) Long term corrective actions include:

(i) revision of the PM procedure to include verifications by RFO6 (10 mos approx after restart)

(ii) eventualimplementation of the PM procedure I as revised to actually confirm the conclusions ,

of acceptability. 1 (g) Requested but have not received, as of 2/21/98:

(i) Identification of specific motors / breakers not actually verified (ii) NU assessment of risk significance with regard to Maintenance Rule /PRA considerations (iii) Additional actions needed to ensure motors not changed without setpoint changes being

. accomplished, i.e. additional barriers to prevent recurrence (iv) Planned content, normal frequency, and expected schedule for actual performance of PM expected to verify the correct settings.

l 1

l l DRAFT

i l

l I*

t*

l DRAFT l D. Beckman Millstone - IP 40500 Report input June 26,1998 Management's Encouragement and Receptiveness in Problem identification by individuals

1. Objectives:

= Evaluate the NU problem resolution processes (line management practices, corrective action program and others) to determine their effectiveness providing a vehicle for encouraging identification of, readily accepting and adequately resolving problems and issues identified by individual employees as pathways preferable to ECP or NRC allegations. (SPO Oversight Program Plan, 9/10/97)

• Determine if managers and supervisors encourage employees to identify problems.

Determine if the staff feels management is receptive to problems being brought forward.. (MGMT-3)

2. Scope:

The team conducted employee interviews at all levels of the organization to determine the worker's perceptions of management's efforts and communications intended to enhance problem identification.

Performance measurement data for problera identification and documentation associated with individual-identified problems (Condition Reports, Employee Concem Case Files, Self Assessment results and others) were reviewed

3. Observations and Findings The employee interview results indicated that previously existing barriers to problem identification had been largely eliminated and no major barriers to problem identification were found. Personnel from organizational units both with and without histories of such barriers to problem identification were interviewed and generally characterized the environment as improved and currently receptive to problem identification. No reluctance or renovations were expressed by the individuals with regard to fheir identification of problems to line, CR process, ECP or NRC. Most of the individuals interviewed indicated that they had initiated CRs either personnally or through referral to their supervision. Several

, the individuals stated that the current working environment also supported escalating concems and i

Page 1 of t

f db _ .'

l problems to management above their direct supervision if believed by the employee to be necessary.

The NU performance indicators and condition report program statistics reflect reasonable levels of employee participation which corroborated the interview results.

The team also reviewed the Employee Concerns Program (ECP) Program, the Safety Conscious Work Environment (SCWE) Program, the Employee Concerns Oversight Panel (ECOP) and other management initiatives for NU activities and response actions taken for previously existing barriers to problem identification above. The licensee's handling of individua!, alleged cases of harassment, intimidation, retaliation, or discrimination by the ECP and SCWE organizations appears adequately j responsive to specific case needs. Both technical and human behavior and performance problems were generally well addressed in the case files reviewed.

However, a recent NRC inspection (Need to insert IR# - Unknown of these programs in accordance j with NRC IP-40001 found that the SCWE processes were not formalized. The lack of SCWE program structure (procedures, formal processes and documentation requirements) resulted in many of the mariagement actions being handled directly, on an ad hoc basis, by Recovery Officers and senior managers.

l The licensee uses incoming ECP allegations, employee culture surveys, leadership surveys and ECOr-1 activities to identify barriers to problem identification and HIRD. ECOP oversight activities and surveys !

i used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free l identification and reporting of Voblems appear to be positive contributions to the overall process.

J i

While the licensee has become effective in identifying these barriers, they have not been effective in addressing a continuing high incidence rate of HIRD allegations coming to the Employee Concems Program. The monthly HIRD allegation incidence rate and the frequency of actual and potential chilling effect events at Millstone has not significantly diminished during 1997-1998 and continues to represent potentially serious barriers to a suitably self<ntical organization and appropriate safety conscious work environment. The licensee's effectiveness in acting to turn the trend of HIRD and i prevent its recurrence is further discussed below.

Page 2 of i L____________________________.__________.____.______.__

l l Effectiveness of Problem Resolution Processes in Pesolving Employee identified Problems

1. Objective:

l Assess the effectiveness of the process by which NU management provides direction to the plant staff necessary to prevent problems. Evaluate the organizations high level goals and expectations. (MGMT-1)

2. Scope:

The team conducted interviews, reviewed corrective action program data, and reviewed the licensee's technical resolutions, SCWE-related responses, and long term follow-up for problems identified by

~

condition reports, employee concerns and Employee Concerns Oversight Panel issues, and NU

department self assessments. This included a sample of condition reports and their technical and human performance / behavior resolution activities that resulted from employee concern cases, self assessments.

3. Observations and Findings Review of ECP Cases, SCWE Problem Area response plans, and sample of CR corrective actions determined licensee actions generally responsive to problems and adequate with some exceptions:

1

a. SCWE Problem Areas and Potential Problem Areas identification methods for HlRD issues include the NU Leadership Surveys, Culture Surveys, ECP case intakes, and ECOP identifications and referrals. These mechanisms appear to be effective, ecpecially for the more egregious issues identified by NU management as HIRD

• Problem Areas". Documented actions plans are used to define and manage the remedial actions for the Problem Areas. A sampling review of these plans by the team determined that action plans are generally effective at remediating both the technical issues and human performance and behavior issues.

l Other HIRD issues that are perceived by NU management to be less egregious or that are perceived as minor problems having the potential to eventually become Problem Areas are handled more informally by senior and middle management. These issues are identified by l

line management via normal line management oversight and activities, review of ECP intakes, i

Page 3 of t

and SCWE da;ly meetings which inc'ude dialogue with ECP, ECOP, Human Resources, Legal, the ECP/SCWE Independent Third Party Oversight Contractor (Little Harbor Consultants, Inc.)

and other program participants. No documented action plans are used but the team found that management had taken actions in direct response to such HIRD potentialities and were monitoring the effectiveness Gf the actions.

Notwithstanding the specific HIRD and HIRD-potential management response actions taken to date for HlRD-related problems, NU has been unsuccessfe!in preventing the ongoing emergence rate of HIRD allegations to ECP. An average of about 54% of all ECP allegations included HIRD and about 26% of the tota allegations involved HIRD associated with nuc! car safety related protected activities subject to 10 CFR 50.7. The fraction of the total cases not !

involving 10 CFR 50.7 typically involve other forms of HIRD, e.g. age, sex or other discrimination, haraument or intimidation over work rules, compensation, benefits, etc. The overall frequency of personnel behaviors that can be construed as HIRD, regardless of their source or subject has been identified by NU and NRC as a serious concern relative to the establishment of a healthy SCEW. While NU notes that only three 1997-98 ECP cases have involved confirmed 10 CFR 50.7 HIRD issues, investigation c'f many of the alleged cases remains open.

NU has performed extensive re-evaluation of the data to ensure their categorization is correct; re-evaluation resulted in no substantive changes in categorization. NU acknow' edges that HIRD in non-50.7 environments represents a real and present potential of 50.7 HIRD. The team discussed and evaluated the management actions planned and taken with the Unit 3 Vice President (M. Brothers), the Vice President-Engineering (D. Amerine) responsible for SCWE, the SWCE management team, the ECP assistant director.(M. Gentry, A. Elms, J. Streeter),

and others. Except for specific HIRD ECP cases or explicitly identified SCWE Problem Areas, manaaement does not appear to be takina actions focused to reduce the overall incidence of HIRD allegations.

I

b. M&TE Program issues Several of the ECP case reviewed by the team identified programmatic and implementation problems in the NU programs for control of measuring and test equipment (M&TE). The inspector initially reviewed the two M&TE related CRs (CR-M3-971292,5/2/97 and ACR M3-97-0150,1/1K/97) issued in response to the ECP cases, finding that they also related a several other related CRs. Further inspector review of these and a several year history of M&TE-related CR data found a numerous NU-identified examples of M&TE program breakdown.

l Page 4 of l

I Discussions with NU and Region I personnel indicated that the adverse performance history at least from 1992.

NU was requested to provide further information on actions taken to address apparent adverse l trend and a meeting was held on February 18,1998 with the newly appointed Metrology Lab i Supervisor, the new program cwner. The supervisor advised that all M&TE issues had been rolled up into ACR M-1-96-0614 which has been used as a vehicle to address all contemporary problems and drive development of a completely new M&TE program.

On February 25,1998, following the end of the inspection, NU provided the inspector with ACR M-1-96-0614 and its associated corrective action docuntentation which indicated that the M&TE Program had been completely re-written and had just become effective in early January 1998. The new program provided for a new Metrology Lab to provide central control for equipment and activities, a new training for all program implementors / users. Full implementation had not yet been achieved but was expected to complete soon with an Effectiveness Evaluation planned for mid-1998.

JTS This ACR is actually pretty messy but appears to broadly bound all the problems of the l

past...though it's hard to tell without more extensive discussions with NU. I am not comfortable " signing off" on it based on a table top review go i recommend either an IFl or URI.

I know it's a lousy " bomb" to drop on some other poor, unsuspecting inspector but Norm &

Tony said tney'd been having badproblems for 5-6 years or more. Might be worthwhile to make sure it's ok sometime after the licensee says it's mature.

Based on the unacceptably long history and broad extent of problems with the M&TE program and the comprehensiveness of the changes in the new program, the licensee's specific corrective actions and full implementation of the program will be reviewed during a future inspection.

c. Boric Acid Transfer Pump Air Binding Boric Acio (BA) Transfer Pumps (3CHS-P2A & -P28) are part of the Technical Specification required reactivity control systems and provide a boron injection flow path to the Reactor Coolant System. CR M3-97-2943, dated 9/4/97, identified that a chronic air binding problem with the pump that periodica3y renedered them inoperable. The CR noted that the condition had previously been identified in 1992 (PIR 3-92-210); 1995 (ACR 3617); and earlier in 1997 (CR M3-97-0715, CR M3-9701011, CR M3-97-0954) but not corrected.

Page 5 of L_-_______

NU's root cause analysis concluded that air entrainment causeci t.he 1992 event, vented the pump and required no further action as it was considered an isolated occurrence. No cluse was determined for the 1995 event other than the presence of trapped air in a horizontal pipe run.

Intemals of the Boric Acid Storage Tank (BAST) batch tank discharge check valve were removed and a modification was proposed to add an isolation and manua! vent valve to provide improved venting were proposed. CR M3-97-0954 identified the recurrence of the problem and initiated actions to revise BAST low tank level operating limits and alarm set points upward to prevent air entrainment; revise the system Operating Procedure to vent pumps if BAST levels drop below the increased low limits; and to complete the valving modification originally proposed in 1995.

l The corrective actions for M3-97-2943 were essentially the same as the earlier 1997 actions above j except that they included evaluation of a possible modification to re-route the Boric Acid Gravity Flow Boration Piping. The licensee's root cause had concluded that the height of the BA Pump and gravity feed piping connections to the BASTS relative to the minimum TS-allowable tank levels had resulted in air entrainment and pump binding. The proposed corrective actions had been approved by management and were awaiting full implementation at the time of the inspection. l l

It appeared that the NU root cause and corrective actions required implementation of BAST Level )

administrative controls more conservative that the minimum levels required by Technical Specification 3.1.2.5.a, " Borated Water Sources - Shutdown, " or 3.1.2.6.a, " Borated Water i Sources - Operating." The inspector requested the licensee's 10 CFR 50.59 safety evaluation and 10 CFR 50.72 or 50.73 reportabliity evaluations for these actions in that they appeared to l constitute an unreviewed safety question. Specifically, it appeared that the if the pumps and tanks were operated in accordance with the Technical Specifications, they might be unable to perform l

their analyzed safety functions as described in the Final Safety Analysis Report.

l The licensee advised that the M3-97-2943 proposed administrative controls had not been completely prepared and published, consequently the 10 CFR 50.59 evaluation had not yet been completed and was therefore unavailable. Further the NU Unit 3 Licensing Manager stated the licensee's belief that the above conditions did not represent an unreviewed safety question and that a routine Technical Specification Change Request was planned for later submittal to NRC.

Operation would proceed in accordance with the administrative tank level limits pending NRC approval of the change, Additionally, the inspector reviewed potential air or gas infiltration paths with the NU System Engineer. NRC Information Notice 88-23, INPO " Red" Significant Operating Experience Report 97-009 and recent events at a Unit 3 sister-plant, Beaver Valley Power Station, had each identified the potential for accumulation of substantial amounts of hydrogen gas in charging pump Page 6 of

i suction lines. The accumulation resulted from the large pressure drop across the charging pump mini-flow orfices stripping dissolved hydrogen from the charged coolant and retuming it to the pump suctions. The phenomena has the potential for severe charging pump damage. Because of the location of the BA Pump piping connection to the charging pump suction piping, the inspector had inquired whether the licensee had considered hydrogen accumulation and rJ. puion j into the affected BA piping.

1 The NU System Engineer, responsible for both the boric acid and charging systems, was unaware of the operating experience information referenced above and the information had apparently not been considered in the root cause determinations for the Crs. It was later determined that NU l Design Engineering had been assigned responsibility for the operating experience items and had prelimir';.ory determined that Unit 3 did not have the hydrogen problem. The basis for the determinations was not reviewed; the matter was referred for further follow-up to the NRC l Operational Safety Team that was concurrently onsite.

On February 19,1998, NU issued CR M3-98-0975 which documented inadequacies in the root l cause evaluation for CR M3-98-0954 in that it did not consider potential introduction of air from the BAST batching process.

The licensee further advised that the deportability evaluation performed with the CRs had evaluated the deportability for plant conditions at the times of discovery of the pump inoperabilities but had not adequately evaluated past plant conditions for potential, historical inoperabilities of the BAST /BA Pump boration flow paths. On February 18,1998, the licensee issued CR M3-98-0952 which documented the inadequate historical deportability evaluations; disposition of the CR was in progress at the end of the inspection.

Tom - not sure how you want to handle the conclusion....here's what I recommend as a

" maximum" l

l l 1. Inadequate root cause determinations and corrective actions for recurrent BA Pump problems, including two cases under the contemporary CR system.

ii. Inadequate deportability evaluations l

l iii. NU did not adequately consider OE in their evaluation of the problem iv. NU did not recognize the potential USO resulting from non-conservative BA Tank level TechSpecs

d. 480V Molded Case Circuit Breaker Magnetic Trip Setpoints ()  !

I Page 7 of u_________________. _.

i

l 4 Q

l I

The inspector reviewed NU Engineering Self Assessment 3DE-SA-97-03, " Control of Magnetic Trip l

Settings on 480 VAC Motor Starters

• which resulted in CR M3-97-3095 documenting that incorrect magnetic trip settings and thermal overloads were found in safety related applications. The self assessment was comprehensive and rigorous and represented a quality effort to evaluate the i activities.

l Self assessment found a sample of setpoints to be incurrect. Proposed corrective actions were generally comprehensive but required verification of motor name plates to determine actual required trip settings. Name plate inspections were deemed impractical for a "large" (but undefined) number of breakers due to their difficult accessibility. As an attemate to the inspections, NU determined that most of the erroneous setpoints occurred when motors were replace, resulting in new operating currents, but the trip settings were not appropriately adjusted. l L Consequently, NU reviewed motor maintenance files and confirmed that ncne of the unverified motors had been changed. This caused them to conclude that the setpoints are probably correct and further confirmation is not needed prior to restart.

Long term planned corrective actions included: revision of the 480V MCB Preventive Maintenance procedure to include verifications by Unit 3 Refueling Outage 6 (scheduled approximately 10 months after Unit 3 restart) and eventual implementation of the PM procedure as revised to perform the inspections and calculations necessary confirm the conclusions of acceptability. The CR documentation did not identify the equipment affected by the incorrect l

settings, did not include the values for the incorrect as-found magnetic and thermal overload settings, the amount of deviation from as-required settings, nor the evaluation of the impact of l

the deviations on the operability of the affected motors. Further, the documentation did not identify the motors which had not been inspected and verified.. Further the inspector requested the licensee's basis for the acceptability of delaying n in and setpoint verificatoin until at least l

sometime in 1999.

f i

This information was requested by the inspector and received on February 25,1998 following the end of the inspection.

• The reported deviations between the as-required and as-found settings were, in each case, minimal and in the low (conservative) direction to protect the equipment from overcurrent. The licensee further advised that they had examined their records for possib!e nuisance equipment trips

)

Page 8 of I l

resulting from the slightly low trip settings and had found no indication of problems.

• NU advised that thirty seven of seventy uninspected motors would be inspected before restart.

• The remaining thirty three unverified breaker loads / settings included four emergency diesel generator (EDG) fuel transfer pump motors; various control building, EDG building, and intake structure HVAC motors; and other similar loads. The loads are generally support system accessory loads that needed for both immediate and long term recovery responses for design absis accidents. Upon further review, NU advised the inspecter that the plans still include motor name plate verification during the next scheduled PM. However, in those cases where the next PM is more than 18 months from restart, the inspections will be scheduled to occur during a sooner system outage.

Tom - I'm not ree"" omfortable with this. I realize these are 2"d and 3"' order TS support system i eut as long as their not finding any settings that are way-out, I don't have a basis for pushing the issue. Let's discuss further if you need to make a different call than routine followup!

l I

Page 9 of

l F

Mr. Martin L. Bowling Recovery Officer - Millstone Unit 2 c/o Mr. H. Miller Northeast Nuclear Energy Company P.O. Box 128 Waterford, CT 06385-0128

Dear Mr. Bowling:

This letter provides the preliminary results of the NRC Region I team inspection of Northeast Utilities (NU) controls in identifying, resolving and preventing issues that degrade the quality of plant operations or safety at Millstone Unit 3. This team inspection was performed onsite from February 9 through February 20,1998. The detailed findings of uw team inspection will be documented in inspection report 50-423/97-82. The inspection team leader provided you with the results of the inspection at a public meeting on February 26,1998, inspection Scope Our inspection examined the management processes used to provide direction to the plant staff to facilitate effective and safe plant operations. This was accomplished by reviewing your goals and expectations, communications and teamwork, receptiveness to problems brought forward, performance monitoring, and their commitment to resolve safety committee recommendations and audit and assessment findings.

Our inspection also assessed the adequacy of your corrective actions program including processes for identification, analysis and resolution of plant deficiencies. We also evaluated your organizations responsiveness in dealing with issues brought forward by employees through various channels including your employee concerns program. We examined the backlog of open problem reports to verify that safety significant issues are being tracked to completion. We reviewed the process to prioritize corrective actions based on risk, and evaluated your process for assessing the effectiveness of corrective actions.

The inspectors evaluated your process for site and departmental self assessments. We reviewed your corrective actions in place for several significant self assessments and third party audits including actions applicable to Unit 3 from ACR-7007 Event Response Team Report, and the a,,tions to improve the Nuclear Oversight Department taken in response to the 1996 Joint Utilities Mana0ement Assessment. In addition, we examined the effectiveness of your Performance and Evaluations Group in their audit, surveillance and quality control function.

We also observed the Nuclear Safety Advisory Board (NSAB) and also the on-site safety committees, the Site Operations Review Committee (SORC) and the Unit 3 Plant Operations Review Committee (PORC). We also reviewed your operating experience program, including the programs for evaluation of industry data and site experience.

Overall Assessment of the Corrective Actions Program l

Q 4

l 7 I Ni -?

O 2

M. L. Bowling The team found a structured framework in place that provided a strong definition for the corrective actions program. Also, there is evidence of a good deal of management attention applied to infuse quality into the program implementation process. The team observed a general improving trend in quality over the last year for most program aspects, that is, issue identification, classification, analysis and actions to prevent recurrence.

However, deficient conditions were found to exist in some root cause analysis and in some corrective actions.

1 Our overall assessment of the corrective actions program is, although we have seen j evidence of the corrective action program beginning to function, it is clear that the program will require careful monitoring by NU. For example the team found a disproportional j number of issues in a relatively small sample size of Condition Reports after NU had J

completed an extensive self assessment.

Preliminary Findings identified by the NRC Team i The NRC inspection team identified regulatory issues within the scope of the areas e.xamined. Our preliminary evaluation indicates that the following findings are being considered as potential violations.

In the area of Design Control, we found problems in the Master Setpoint List (MSL), in that it did not contain the setpoint and the calculation reference as required by NGP 5.23.

Also, a potential deficiency with the sub-cooling margin setpoints was found, in addition, the team questioned the adequacy of setpoint control based on our observations. l 1

Thete were two procedural problems identified with maintaining accurate design basis documents. The Safety Functional Requirements Manual and the Design Basis Summaries were not maintained as required by NGP 5.28 and PI-29, respectively.

The team also found problems with meeting the organizationalindependence required by Section 6 of the Millstone Unit 3 Technical Specifications because of the reporting relationship between the Radiation Protection Manager and the Maintenance Manager.

These requirements are established in TS 6.2.1.d, the Updated Final Safety Analysis Report and Regulatory Guide 8.8 Also, the independence required for persons performing the Independent Safety Engineering Group (ISEG) reviews for human performance issues did not meet the TS 6.2.3.3.

The team found severalindividual problems within the Millstone Corrective Action process.

These problems include incomplete corrective actions and root cause analyses, as

illustrated by the repetitive air binding of the Boric Acid Transfer pumps, issue closure l without completion of all corrective actions, as was the case with Design Change Notice (DCN) review required by CR M3-97-0506, and also the closure of an Automated Work Order (AWO) to correct service water flow instrument anomalies that was associated with Operator Work Around 96-03. We note that completion of the DCN review was an Adverse Condition Report (ACR) 7007 and Configuration Management Program restart commitment.

i l

I i

3  !

M. L. Bowling {

The team also identified procedural problems with assigning inappropriate significance levels for Condition Reports. These included the following: incomplete action on Generic Letter {GL) 89-13, Service Water Fouling, and GL 90-03, Vendor TechnicalInformation Program. These are commitments to the NRC and should have been classified as at least Level 2. We also observed that the Condition Reports for the incomplete Generic Letter issues had been inappropriately coded as ' Deferred' until after plant restart.

Additional Findings in the management area, the team found that management communications methods with the plant staff were a strength. There was a common understanding of management's expectations by plant personnel. However, it was noted that a strategic plan and vision statement on where the plant is headed are in draft. This is considered a weakness in view of the f act that the current management has been in place since late 1996. Overall, the Nuclear Group Policies and Standards, were considered good. Although, teamwork initiatives at the first line supervisor and above were developed, there is a need to. extend this to the worker level.

Observations and interviews show that mancgers and supervisors encourage employees to identify problems. The plant staff feels that management is receptive to problems brought forward, and individuals generally characterized the environment as improved and currently receptive to problem identification. There is no reluctance or reservations expressed by individuals to identify problems through the Corrective Actions Condition Report process, to the Employee Concerns Program (ECP) or to the NRC.

The handling of individual HIRD cases by the Employee Concerns Program and the Safety Conscious Work Environment (SCWE) program is adequately responsive to specific case needs. Both technical and human-side problems are generally well addressed. The ECP case intakes, and the Employee Concerns Oversight Panel (ECOP) oversight activities and surveys, are used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems. These are positive contributions to the overall process. These mechanisms are effective espotially for the more egregious issues identified as problem areas. However, NU has not adequately dealt with trends, common causes and the overall occurrence rate of HIRD allegations to ECP organization-wide. A significant backlog of HIRD allegations were pending investigation and the backlog and emergent HIRD allegations had not been analyzed by NU for broad trends or patterns, and common causes. As a result, the actions taken to date had not been effective in curtailing an adverse trend in the incidence of HIRD allegations. Further, the SCWE processes have not yet been formalized, that is the program lacks structure in the form of procedures, formal processes and documentation requirements.

l NU has not been effective in preventing the ongoing emergence rate of HIRD allegations to ECP. The incident rate of HIRD allegations and management-related or induced chilling effect events has not diminished significantly. Management actions to reduce the overall rate of HIRD allegations have not been effective, except for specific HlRD cases or explicitly identified SCWE Problem Areas.

l

! 4 l M. L. Bowling Concerning corrective actions, we noted a generally low threshold for recording issues but also noted a tendency to assign a lower Significance Level classification and to wave root cause analysis to similar issues. We are concerned that this practice may overlook

! ~ detecting earlier ineffective corrective actions.

In the issue addressed earlier concerning air binding of the Boric Acid Transfer pumps, NU l has re-evaluated their earlier root cause analysis and conclusions and began re-analysis of j the issue at the conclusion of the inspection period. However, the team noted that NU i friled to recognize the potential unreviewed safety question (USQ) which resulted from

!' their initial conclusion of a more restrictive Boric Acid Tank level requirement relative to the Technical Specification. The team also observed that the reportabihty evaluation of the event was incomplete and that NU failed to consider industry operating experience in their evaluation of the problem.

The team also found that your Corrective Actions program had inadequate controls over combining Condition Reports such that their issues were preserved and that they )

maintained their appropriate significance level. This weakness was evident in the hand!ing of multiple Condition Reports concerning deficiencies in the Nonconformance Report process. We consider the poor control of this activity as a program weakness.

We found that long term compensatory measures are in effect for fire protection systems because surveillance testing which verifies operability of these systems has been suspended. Compensatory measures are taken to allow restoration of a fire protection ,

, system. But in this case, are being used to substitute for long term system inoperability. l l The team also found that the failure to conduct the surveillance was inappropriate because it sets a low standard of performance for plant personnel.

l The team found that the self assessments were generally of high quality. NU has taken a l strong initiative in areas like the Nuclear Oversight Restart Verification Assessment.

The NRC team found that there has been an improvement within the traditional quality assurance and quality control function of the Nuclear Oversight organization's Audits and Evaluation Group. The number of Auditors has significantly increased as well as their

, qualifications and knowledge level has increased. Audit program procedures are i acceptable. There are four new audit managers. Oversight has the opportunity to concur with the corrective actions taken for audit findings and nonconformance reports. There is good interface with the line organization concerning the Nuclear Oversight Restart Verification Plan.

The NRC team found that the Independent Safety Engineering Group (ISEG) has performed high quality plant reviews and Operational Experience (OE) reviews. However, the number of ISEG reviews done in 1997 was only 12, down from 24 the previous year. This appears ,

to be a result of the OE workload being performed by the group. The ISEG group has  !

reduced the backlog of OE issues from several hundred to approximately 40 for Unit 3.

NRC team members attended meetings of the Site Operations Review Committee (SORC),

the Unit 3 Plant Operations Review Committee (PORC) and the Nuc! ear Safety Advisory

. I I

M. L. Bowling ,

i l

Board (NSAB). All three safety committees operated effectively, the mernbers were '

prepared for the meeting and added quality to the issue being addressed. The team observed two minor examples of nonconservative interpretation and implementation of the technical specifications regarding staffing qualifications and delegation of responsibility, i

The team reviewed NU actions on several Significant items List (SIL) issues, and has recommended that all four be closed. These issues include SIL ltem 41-1, ACR-7007 issues relevant to Unit 3, SIL ltem 41-2, concerning trending of NCRs, SIL ltem 73-1, concerning the Technical Specification audit program, and Sllitem 73-2, concerning the adequacy of the Nuclear Oversight Program.

The SPO staff willinclude these findings within NRC inspection Report 50-245/97-82, j which will provide the final observations, findings, and any enforcement actions to which I you will be required to respond based on the results of the subject inspection. No response to the issues discussed in this letter are required at this time; however, any potential enforcement items which warrant prompt corrective actions should be addressed in a timely manner rather than waiting for the final report. l Should you have any questions or comments regarding the issues discussed in this letter, please contact me at (610) 337-5126. i Sincerely, Wayne D. Lanning l Deputy Director, inspections Special Projects Office Office of Nuclear Reactor Regulation Docket No. 50-423 cc: See next page g:\tempfile\405quic.wpd i

L___ ___-____ _____________________----------__.- _ ----------------- --_____ _ ________________ - - - - - - - - - . _ _ _ _ _ _ _ _ _ _ _ - - - - - - - - - -

. l 1

Mr. M. L. Bowling, Recovery Officer, Unit 2 C

/o Patricia Loftus, Director - Regulatory Affairs for Millstone Station ,

NORTHEAST NUCLEAR ENERGY COMPANY P.O. Box 128 Waterford, Connecticut 06385 j

Dear Mr. Bowling:

\

SUBJECT:

NRC TEAM INSPECTION 50-423/97-82 and NOTICE OF VIOLATION l

On February 26,1998, the NRC completed a team inspection of the Northeast Nuclear Energy Company (NNECO) controls in identifying, resolving and preventing issues that  ;

degrade the quality of plant operations or safety at Millstone Unit 3. The enclosed report )

presents the results of that inspection. '

The team inspection was performed onsite from February 9 through February 20,1998, l using NRC Inspection Procedure 40500," Effectiveness of Licensee Controls in Identifying,  !

Resolving and Preventing Problems." Overall, the team found evidence of improving  !

performance. The corrective actions program is functioning, but it is clear that the  !

program will continue to require careful monitoring by NNECO management to ensure sustained performance. Self-assessments were typically of high quality. The quality assurance and quality control functions of the Nuclear Oversight organization have <

improved as a direct result of improved staffing, qualifications, and knowledge level. Of  ;

note is the Nuclear Oversight Restart Verification Assessment, which the team considered I to be a strong initiative. All three safety review committees (Nuclear Safety Assessment Board, Site Operations Review Committee, and Plant Operations Review Committee) operated effectively and provided quality input in addressing the issues before them. The team also noted that the Independent Safety Engineering Group has made considerable progress, alheit at the expense of performing safety reviews, in reducing the backlog of operating experience (OE) reviews. The team found that management communications methods with the plant staff were a strength. The plant staff feels that management is  ;

receptive to problems brought forward and individuals generally characterized the  !

environment as improved and currently receptive to problem identification.  ;

Nonetheless, our inspection identified three violations of NRC requirements. The first  :

instance, involves failure to take corrective actions: for recurrent problems with the boric  ;

acid transfer pumps; for failing to complete all of the work specified in an automated work order concerning modifications to service water flow instrumentation; for failing to complete the specified corrective actions in a daign change notice review and for failing to review the Independent Safety Engineering Group procedure. The second violation

} concerns the failure to follow procedures: concerning the assignment of significance levels to condition reports and for changes to the Safety Function Requirements Manual and design basis summaries. The third violation concerns f ailure to have the organizational independence required for the Radiation Protection Manager. This is a concern because it b

,, n M ?? w h e -

'~

~

" - ll 4

Mr. M. L. Bowling 2 and other examples found during the inspection reflect a lack of attention to compliance with Section 6.0 of the plant's Technical Specifications.

These violations are cited in the enclosed Notice of Violation and the circumstances surrounding them are described in detailin the enclosed report. Please note that you are required to respond to this letter and should follow the instructions specified in the enclosed notice when preparing your response. The NRC will use your response, in part, to determine whether further enforcement action is necessary to ensure compliance with regulatory requirements. Furthermore, in addition to the enclosed enforcement actions, your attention is directed to specific areas identified in the body of the report.

NRC review of SIL closura packages has closed SIL ltems No. 41 and 73.

In accordance with 10 CFR Part 2.790 of the NRC's " Rules of Practice," a copy of this letter and its enclosures will be placed in the NRC Public Document Room (PDR).

Sincerely, Wayne D. Lanning Deputy Director of Inspections Special Projects Office Office of Nuclear Reactor Regulation 1

Docket No.

50-423

Enclosures:

1. Notice of Violation l 2. NRC Inspection Report 50-423/97-82 l

Mr. M. L. Bowling 3 cc w/ encl:

B. Kenyon, President and Chief Executive Officer M. H. Brothers,.Vice President - Operations ,

J. McElwain, Unit 2 Recovery Officer

]

J. Streeter, Vice President, Nuclear Oversight '

E.- J. Harkness, Unit Director, Millstone Unit 1 G. D. Hicks, Unit Director - Millstone Unit 3 J. A. Price, Unit Director - Millstone Unit 2 D. Amerine, Vice President for Engineering and Support Services P. D. Hinnenkamp, Director, Unit 1 Operations ,

F. C. Rothen, Vice President, Work Services J J. Cantrell, Director - Nuclear Training S. J. Sherman, Audits and Evaluation  !

L. M. Cuoco, Esquire i J. R. Egan, Esquire 1

. V. Julisno, Waterford Library ,

J. Buckingham, Department of Public Utility Control l S. B. Comley, We The People State of Connecticut SLO Designee D. Katz, Citizens Awareness Network (CAN)

R. Bassilakis, CAN - I J. M. Block, Attorney, CAN l S. P. Luxton, Citizens Regulatory Commission (CRC)  !

Representative T. Concannon E. Woollacott, Co-Chairman, NEAC l

4

)

i i

i i

I I

Mr. M. L. Bowling 4 Distribution w/ encl:

Region i Docket Room (with p_ony o of concurrences)

Nuclear Safety Information Center (NSIC)

PUBLIC FILE CENTER, NRR (with Oriainal concurrences)

NRC Resident inspector W. Axelson, DRA (Inspection Reports)

B. Jones, PIMB/ DISP M. Kalamon, SPO, RI D. Screnci, PAO.

W. Travers, Director, SPO, NRR -

i 1

s

~

\

Mr. M. L. Bowling 5 Distribution w/enci (VIA E-MAIL):

J. Andersen, PM, SPO, NRR M. Callahan, OCA

.R. Correia, NRR B. McCabe, OEDO S. Dembek, PM, SPO, NRR E. Imbro, Deputy Director of ICAVP Oversight, SPO, NRR  ;

D. Mcdonald, PM, SPO, NRR P. McKee, Deputy Director of Licensing, SPO, NRR S. Reynolds, Chief, ICAVP Oversight, SPO, NRR i D. Screnci, PAO Inspection Program Branch (IPAS)

DOCDESK (Inspection Reports Only)  ;

i i

l i

l l

l l

l DOCUMENT NAME:

T3 receive a copy of this document. Indicate in the box: "C" = Copy without attachrnent/ enclosure "R" = Copy with attachrnent/ enclosure "N" = No j i

copy 0FFICE Rl/DRS l RI/SP0 lE Rl/SP0 l l 1 NAME Shedlosky Durr ' Lanning i DATE 05/05/98 06/ /98 06/ /98 06/ /98 06/ /98 I 0FFICIAL RECORD COPY  ;

i

l ENCLOSURE 1 NOTICE OF VIOLATION Northeast Nuclear Energy Company Docket No.: 50-423 Millstone Nuclear Power Station, Unit 3 License Nos.: NPF-49 I

During an NRC inspection conducted from February 9,1998, through February 20,1998, violations of NRC requirements were identified. In accordance with the " General  ;

Statement of Policy and Procedure for NRC Enforcement Actions," NUREG-1600, the l violations are listed below:

I A. Criterion XVI of 10 CFR 50, Appendix B, requires, in part, that measures must be  !

established to assure that conditions adverse to quality, such as failures,  !

malfunctions, deficiencies, deviations, defective material and equipment and nonconformances are promptly ident'fied and corrected. '

Contrary to the above, appropriate corrective actions were not taken for the following four j issues: j i

1) The boric acid transfer pumps are part of the Technical Specification required reactivity control systems and provide a boron injection flow path to the Reactor-Coolant System. 1 There has been a chronic air binding probiem with the pumps that periodically rendered the l sub-system inoperabl6. The condition had been identified six times, including as early as 1992, but not corrected. .

2) An automated work order (AWO) associated with a r". modification to correct flow indication anomalies on service water instrumentation 3SWP-FI-059 A, B and C was inappropriately closed, prior to the completion of all specified work. Specifically, the final ,

setpoint calibrations for flow indicators 3SWP-FI-059 A, B and C had not been accomplished prior to closing the AWO. l

3) An Action Request, AR 97003960-05, associated with Adverse Condition Report ACR M3-97-0506 was closed, but all of the actions for the assignment, requiring review of Design Change Notices, had not been completed.

4) The independent Safety Engineering Group procedure was not reviewed by the Site l Operations Review Committee (SORC) to complete corrective actions associated with Condition Report (CR) M3-97-3974.

This is a Severity Level IV violation (Supplement 1) for Docket No. 50-423.

l B. Criterion V of 10 CFR 50, Appendix B, requires, in part, that activities affecting quality be prescribed by instructions or procedures of a type appropriate to the circumstances. Further, it requires that these activities be accomplished in accordance with these instructions or procedures.

0@$ b b Acua.e e >u m

2 Technical Specification 6.8 requires, in part, that procedures shall be established, implemented maintained covering the activities referenced in ... Appendix A of 1 Regulatory Guide 1.33.

1 RP4, Corrective Actions Program, Rev 5, Attachment 3, CR Initiation and Classification Guidelines, includes in the Level 2 guidelines: an external station commitment not  ;

adhered to; or a deficiency in material that, if left uncorrected, could affect safe reliable {

plant operation. {

l Contrary to the above, written procedures were not followed for the following:

l l

1) Changes to the Safety Functional Requirements (SFR) Manual, which was developed to  !

identify the key system level requirements that are reflected in the safety analysis, were '

not documented as Design Change Notices (DCN) and the DCN numbers entered into the Generation Records Information and Tracking System as required by station procedure NGP 5.28, Design Basis Documentation Packages, Rev. 3,10/15/97, Step 1.1.2.

2) There was no Design Basis Summary (DBS) for the Emergency Lighting System, or for 1 the full Chemical & Volume Control System as required by station procedure PI 29, Development of Millstone Unit 3 Design Bases Summary Documents. l

3) Condition Report (CR) M3-97-4672 is related to NRC Generic Letter GL 89-13, " Service Water System Problems Affecting Safety-Related Equipment" and was inappropriately classified as Level 3. CR M3-97-4346 contains a deficiency in material (inadequate corrosion control) that, if left uncorrected, could affect safe reliable plant operation. It is related to NRC GL 89-13 and was inappropriately classified as Level 3. Station procedure RP 4, " Corrective Actions Program," Rev. 5, Attachment 3, "CR Initiation and Classification Guidelines," includes in the Level 2 guidelines as: an external station j commitment not adhered to; or a deficiency in material that, if left uncorrected, could affect safe reliable plant operation. l l

This is a Severity Level IV violation (Supplement 1) for Docket No. 50-423.

C. Technical Specification 6.2.1.d requires, in part, that those who carry out health physics functions have sufficient organizational freedom to ensure their independence from operating pressures.

Unit 3 Final Safety Analysis Report (FSAR) Section 12.5.3 states that all health physics procedures and methods for ensuring that occupational radiation exposure is as low as reasonably achievable (ALARA) follow the provisions and suggestions of Regulatory Guide (RG) 8.8, Revision 3; RG 8.10, Revision 1-R; and RG 1.33, Revision 2, as applicable. RG 8.8, Section C.1.b(3), states, in part, as follows: The Radiation Protection Manager (RPM) ,

(onsite) has a safety function and responsibility to both employees and management that j can best be fulfilled if the individuals independent of station divisions, such as operations, '

maintenance or technical support, whose prime responsibility is continuity or improvement of station operability.

l l

L

3 Contrary to the above, the Radiation Protection Manager reports to the Maintenance Manager.

This is a Severity Level IV violation (Supplement 1) for Docket No. 50-423. j For the violations listed above and pursuant to the provisions of 10 CFR 2.201, Northeast '

Nuclear Energy Company is hereby required to submit a written statement or explanation within 30 days of receipt of the letter transmitting this Notice of Violation (Notice) to the U.S. Nuclear Regulatory Commission, ATTN: Document Control Desk, Washington, D.C.

20555 with a copy to the Director, Special Projects, Nuclear Reactor Regulation and a )

copy to the NRC Resident !nspector at the facility that is the subject of this notice. This  !

reply should be clearly marked as a " Reply to a Notice of Violation" and should include for each violation: (1) the reason for the violation, or,if contested, the basis for disputing the violation; (2) the corrective st3ps that have been taken and the results achieved; (3) the corrective steps that will be taken to avoid further violations; and (4) the date when full compliance will be achieved. Your response may reference or include previous docketed correspondence, if the correspondence adequately addresses the required response. If an adequate reply is not rer.eived within the required time specified in this notice, an order or a Demand for information may be issued as to why the license should not be modified, suspended or revoked, car why such other action as may be proper should not be taken.

Where good cause is shown, consideration will be given to extending the response time.

Because your response will be placed in the NRC Public Document Room (PDR), it should not, to the extent possible, include any personal privacy, proprietary or safeguards information so that it can be placed in the PDR without redaction. If personal privacy or proprietary information is necessary to provide an acceptable response, then please provide ,

a bracketed copy of your response that identifies the information that should be protected (

and a redacted copy of your response that deletes such information. If you request withholding of such material, you must specifically identify the portions of your response that you seek to have withheld and provide in detail the bases of your claim of withholding I I

(e.g., explain why the disclosure of information will create an unwarranted invasion of personal privacy or provide the information required by 10 CFR 2.790(b) to support a request for withholding confidential commercial or financialinformation). If safeguards information is necessary to provide an acceptable response, please provide the level of protection described in 10 CFR 73.21.

Dated at King of Prussia, Pennsylvania this XXth day of May 1998 l

)

l

, l l

\

I

N l

1 Mr. Martin L. Bowling Recovery Officer - Millstone Unit 2 c/o Mr. H. Miller Northeast Nuclear Energy Company

- P.O. Box 128 Waterford, CT 06385-0128

SUBJECT:

Pr4ELIMINARY INSPECTION RESULTS This letter provides the preliminary results of the NRC Region I team inspection of Northeast Utilities (NU) controls in identifying, resolving and preventing issues that degrade the quality of plant operations or safety at Millstone Unit 3. This team inspection was performed onsite from February 9 through February 20,1998, using NRC inspection procedure 40500," Effectiveness of Licensee Controls in Identifying, Resolving and

' Preventing Problems." The detailed findings of the team inspection will be documented in inspection report 50-423/97-82. The inspection team leader provided you with the results of the inspection at a public meeting on February 26,1998, inspection Scope Our inspection examined the management processes used to provide direction to the plant staff in order to f acilitate effective and safe plant operations. ' This was accomplished by reviewing your goals and expectations, communications and teamwork, receptiveness to problems brought forward, performance monitoring, and your commitment to resolve safety committee recommendations and audit / assessment findings.

l Our inspection also assessed the adequacy of your corrective actions program including

! processes for identification, analysis and resolution of plant deficiencies. The inspectors evaluated your organization's responsiveness in dealing with issues brought forward by

employees through varicus channels including your employee concerns program. The team l examined the backlog of open problem reports to verify that safety significant issues are being tracked to completion, reviewed the process to prioritize corrective actions based on risk, and evaluated your process for assessing the effectiveness of corrective actions, t-The inspectors evaluated your process for site and departmental self-assessments. The team reviewed the corrective actions that were implemented for several significant self-assessments and third party audits including actions applicable to Unit 3 from the ACR-7007 Event Response Team Report, and the actions to improve the Nuclear Oversight Department taken in response to the 1996 Joint Utilities Management Assessment. In addition, the inspectors examined the effectiveness of your. Performance and Evaluations 51&W~N~ ~)y. .

l ' f13

1 V

2 M. L. Bowling l l

Group in their audit, surveillance and quality control function.

l The team observed activities involving the Nuclear Safety Advisory Board (NSAB), the Site Operations Review Committee (SORC) and the Unit 3 Plant Operations Review Committee (PORC). The inspectors reviewed your operating experience program, including the programs for evaluation of industry data and site experience.

Preliminary Assessment of the Management Process The team found that management communications methods with the plant staff were a strength. There was a common understanding of managements expectations by plant J personnel. However, it was noted that a strategic plan and vision statement on where the plant is headed were still in draft. This is considered a weakness in view of the fact that I the current management has been in place since late 1996. Overall, the Nuclear Group policies and standards were considered good. Teamwork initiatives at the first line supervisor and above were developed, but have not been fully extended to the worker level. ,

I Observations and interviews show that managers and supervisors encourage employees to identify problems. The plant staff feels that management is receptive to problems brought forward and individuals generally characterized the environment as improved and currently  ;

receptive to problem identification. There is no reluctance or reservation expressed by i individuals to identify problems, either through the Corrective Actions Condition Report process, the Employee Concerns Program (ECP), or to the NRC.

1 I

The handling of individual HIRD cases by the Employee Concerns Program and the Safety l Conscious Work Environment (SCWE) program is adequately responsive to specific case needs. Both technical and human-side problems are generally weli addressed. The ECP case intakes and the Employee Concerns Oversight Panel (ECOP) oversight activities and surveys are used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems. These are positive i contributions to the overall process. These mechanisms are effective, especially for the more significant issues identified as problem areas. However, NU management has not been fully effective in addressing trends and common causes for HIRD allegations generated organization wide to ECP. A significant backlog of HIRD allegations was pending investigation and the backlog and emergent HIRD allegations had not been analyzed by NU for broad trends or patterns, and common causes. As a result, the actions taken to date had not effectively assessed the nature and substance of the continuing high incidence rate of HIRD allegations at the time of the inspection. Further, the SCWE processes have not yet been formalized. That is, the program lacks structure in the form i of procedures, formal processes and documentation requirements.

l l

Preliminary Assessment of the Corrective Actions Process I l

Overall, the team saw evidence that the corrective actions program is functioning, but it is l clear that the program wil! continue to require careful monitoring by NU management to ,

I ensure sustained performance. For example, the team found a notable number of 1

t i

l 3

M. L. Bowling l additionalissues in a relatively small sample size of Condition Reports, after NU had completed their own extensive self-assessment preparing for this team inspection. In general, the findings represent minor process problems with the exception of the boric acid tank level finding discussed below.

The team found a structured framework in place that provided a strong definition for the corrective actions program. There is evidence of a good deal of management attention applied to infuse quality into the program implementation process. The team observed a general improving trend in quality over the last year for most program aspects, including issue identification, classification, analysis and actions to prevent recurrence. However, deficient conditions were found to exist in some specific root cause analyses and corrective actions.

The team found that the Corrective Actions Program lacked controls when it combined Condition Reports such that they did not preserve issues and that they did not maintain their appropriate significance level. This weakness was evident in the handling of multiple condition reports concerning deficiencies in the Nonconformance Report process. The team considers this activity to be a program weakness.

The team noted that the licensee had established a generally low threshold for recording issues but that there was a tendency to assign a lower Significance Level classification than appropriate and to waive root cause analysis to similar issues. The team is concerned that this practice has the potential to cause trends to be missed and possibly result in ineffective corrective actions, incomplete corrective actions and root cause analysis were identified during the team's l

review of several Condition Reports including the repetitive air binding of the Boric Acid 1 Transfer pumps. The team noted that NU engineering failed to recognize the potential  !

unreviewed safety question (USQ) which resulted from their initial conclusion of a more l restrictive Boric Acid Tank level requirement relative to the Technical Specification. The )

team also observed that the deportability evaluation of the event was incomplete and that l NU failed to consider industry operating experience in their evaluation of the problem.

The team found that long term compensatory measures are in effect for fire protection systems because surveillance testing which verifies operability of these systems has been suspended. Compensatory measures are being taken to allow restoration of a fire I protection system, but in this case, are being used inappropriately to substitute for long term system inoperability.

Preliminary Assessment of Site and Departmental Self-Assessment Process i

i l The team found that self-assessments were typically of high quality. It noted that the I quality assurance and quality control functions of the Nuclear Oversight organization have  !

improved as a direct result of improved staffing, qualifications, and knowledge level. Of l note is the Nuclear Oversight Restart Verification Assessment, which the team considered l to be a strong initiative. That Nuclear Oversight now has the opportunity to concur on corrective actions taken as a result of its audit findings and nonconformance reports is also  ;

i L____.__-_____.__.______.___ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _

4 M. L. Bowling considered to be a substantive improvement.

Preliminary Assessment of Safety Review Committee / Operating Experience Programs The team found that all three safety review committees (Nuclear Safety Advisory Board, Site Operations Review Committee, and Plant Operations Review Committee) operated effectively, were well prepared for their meetings, and provided quality input in addressing the issues before them. The team also noted that the Independent Safety Engineering Group has made considerable progress, albeit at the expense of performing safety reviews, in reducing the backlog of operating experience (OE) reviews. The backlog of OE reviews on Unit 3 has been reduced from several hundred to approximately 40. However, the number of ISEG reviews done in 1997 was only 12, down from 24 the previous year.

Preliminary Findings identified by the NRC Team The NRC inspection team identified regulatory issues. Based on our preliminary evaluation, the following findings are being considered as potential violations:

• In the area of design control, the team found problems in the Master Setpoint List (MSL), in that it was inconsistent and did not contain all of the setpoints and the calculation references as required by NGP 5.23. Also, a potential deficiency with the sub-cooling margin setpoints was found, in addition, the findings callinto question the adequacy of setpoint control.

• There were two procedural problems identified with maintaining accurate design basis documents. The Safety Functional Requirements Manual and the Design Basis Summaries were not maintained as required by NGP 5.28 and PI-29, respectively.

l

• The team found problems with meeting the organizationalindependence required by I Section 6 of the Millstona Unit 3 Technical Specifications because of the reporting relationship between the Radiation Protection Manager and the Maintenance Manager. l These requirements are established in TS 6.2.1.d, the Updated Final Safety Analysis Report and Regulatory Guide 8.8. Also, the independence required for persons performing the Independent Safety Engineering Group (ISEG) reviews for human

! performance issues did not meet the intent of TS 6.2.3.3.

• The inspection identified several individual problems within the Millstone Corrective Action Program. These problems include: (1) incomplete corrective actions and root cause analyses, as illustrated by the repetitive air binding of the boric acid transfer pumps, and (2) issue closure without completion of all corrective actions, as was the case with Design Change Notice (DCN) review required by CR M3-97-0506 and the closure of an Automated Work Order (AWO) to correct service water flow instrument anomalies that was associated with Operator Work Around 96-03. The team noted that completion of the DCN review was an Adverse Condition Report (ACR) 7007 and Configuration Management Program restart commitment.

• The team identified some procedural problems with assigning inappropriate significance l

l h _

m 5

M. L. Bowling levels for Condition Reports. These included the following: incomplete action on Generic Letter (GL) 89-13, Service Water Fouling, and GL 90-03, Vendor Technical Information Program. These are commitments to the NRC and should have been classified as at least Level 2. The team also observed that some actions on the Condition Reports for the incomplete Generic Letter issues had been inappropriately coded as ' Deferred' until after plant restart.

The team reviewed NU actions on four issues related to two NRC Significant items List (SIL) items (SIL ltems 41 and 73), and has recommended that all four be closed. These four issues include: ACR-7007 issues relevant to Unit 3; trending of NCRs; the Technical Specification audit program; and the adequacy of the Nuclear Oversight Program. The closure of these issues also results in the final closute of SIL itema 41 and 73.

The NRC's overall assessment of the effectiveness of your corrective action program will be based on the results of this inspection, as well as additional evaluations such as the ongoing inspection of ICAVP corrective actions, the NRC's review of the Employee Concerns Program, and the upcoming Operational Safety Team inspection (OSTI).

The SPO staff willinclude these findings within NRC Inspection Report 50-245/97-82, which will provide the final observations, findings, and any enforcement actions to which you will be required to respond based on the results of the subject inspection. No response to the issues discussed in this letter are required at this time; however, any potential enforcement items which warrant prompt corrective actions should be addressed in a timely manner rather than waiting for the final report.

Should you have any questions or comments regarding the issues discussed in this letter, please contact me at (610) 337-5126.

Sincerely, Wayne D. Lanning  ;

Deputy Director of Inspections Special Projects Office Office of Nuclear Reactor Regulation Docket No. 50-423 j i

l i l

6 6

M. L. Bowling cc:

B. Kenyon, President and Chief Executive Officer M. H. Brothers, Vice President - Operations J. McElwain, Unit 1 Recovery Officer

~

J. Streeter, Recovery Officer - Nuclear Oversight G. D. Hicks, Unit Director - Millstone Unit 3 J. A. Price, Unit Director - Millstone Unit 2 D. Amerine, Vice President for Engineering and Support Services P. D. Hinnenkamp, Director, Unit 1 Operations F. C. Rothen, Vice President, Work Services J. Cantrell, Director - Nuclear Training S. J. Sherman, Audits and Evaluation L. M. Cuoco, Esquire J. R. Egan, Esquire V. Juliano, Waterford Library J. Buckingham, Department of Public Utility Control S. B. Comley, We The People State of Connecticut SLO Designee D. Katz, Citizens Awareness Network (CAN)

R. Bassilakis, CAN J. M. Block, Attorney, CAN S. P. Luxton, Citizens Regulatory Commission (CRC)

Representative T. Concannon E. Woollacott, Co-Chairman, NEAC l

i j

l 1

l l

a 7

M. L. Bowling i Distribution w/ encl:

Region i Docket Room (with .cJ2p_Y of concurrences)

Nuclear Safety information Center (NSIC) l PUBLIC l

FILE CENTER, NRR (with Oriainal concurrences) l SPO Secretarial File, Region l l NRC Resident Ir.spector

! B. Jones, PIMB/ DISP l W. Latining, Deputy Director of Inspections, SPO, RI l D. Screnci, PAO W. Travers, Director, SPO, NRR l

Distribution w/enci (VIA E MAIL):

J. Andersen, PM, SPO, NRR M. Callahan, OCA l R. Correia, NRR B. McCabe, OEDO S. Dembek, PM, SPO, NRR G. Imbro, Deputy Director of ICAVP Oversight, SPO, NRR D. Mcdonald, PM, SPO, NRR l P. McKee, Deputy Director of Licensing, SPO, NRR S. Reynolds, Chief, ICAVP Oversight, SPO, NRR D. Screnci, PAO l Inspection Program Branch (IPAS)

DOCUMENT NAME: G:tempfile\405quc2.327 T3 receive a copy of this documint. Indicato in the box: "C" = Copy without attachment / enclosure "E" = Copy with attachment / enclosure "N" = No copy 0FFICE RI/DRS l Rl/S:'O lE RI/SPO l l l NAME Shedlosky Durr Lanning DATE 04/04/98 06/ /98 06/ /98 06/ /98 OI,/ /98 0FFICIAL RECORD COPY l

l i

l l

l i

U.S. NUCLEAR REGULATORY COMMISSION OFFICE OF NUCLEAR REACTOR REGULATION SPECIAL PROJECTS OFFICE Docket Nos.: 50-423 Report Nos.: 97-82 l License Nos.: NPF-49 Licensee: Northeast Nuclear Energy Company P. O. Box 128 l Waterford, CT 06385 l Facility: Millstone Nuclear Power Station, Unit 3 l l

Inspection at: Waterford, CT i

I Dates: February 9 through February 20,1998 Inspection Team: John T. Shediosky, Lead, Senior Reactor Analyst, Region ! )

Norman J. Blumberg, Special Projects Office - Region i Edward J. Ford, Quality Assurance and Maintenance Branch (HOMB),

Division of Reactor Controls and Human Factors (DRCH),

. Office of Nuclear Reactor Regulation (NRR)

Robert M. Latta, NRR, DRCH, HOMB Richard A. Rasmussen, Senior Resident inspector - Maine Yankee Garmon West, Jr., Human Factors Assessment Branch, DRCH, NRR Donald A. Beckman, Consultant )

James C. Higgins, Consultant, Brookhaven National Lab l Approved by: Jacque P. Durr, Chief Team Manager Inspections Branch Special Projects Office, NRR i

l i

--h/CGJd666 _ _ _ _ - _ _ _ _

Qyg - _ _ _

l l

TABLE OF CONTENTS I EX EC UTIV E SU M M A RY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv 1.0 MANAGEMENT PROCESSES AND SYSTEMS . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 l 1.1 Management Directions, Goals and Expectations .................... ... 1 1.2 Organizational Communications and Teamwork . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Encouragement of Problem Identification by Managers and Supervisors . . . . . . . . . 3 1.3.1 Management's Encouragement and Receptiveness in Problem Identification by Individuals ...................................... ....... 4 1.3.2 Effectiveness of Problem Resolution Processes in Resolving Employee Identified Concerns................................................6 1.4 Pe rf or m a nce M o nitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.5 Management's Commitment To Resolve issues .........................9 2.0 CO R R ECTIV E ACTI O N S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.1 Corrective Action Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 l

2.2 Corrective Actions - Classification and Root Cause Analysis ...............11 )

2.3 Corrective Actions Effectiveness ..................................14 2.3.1 Boric Acid Transfer Pump Air Binding ...........................17 )

2.3.2 480 Vac Molded Case Circuit Breaker Magnetic Trip Setpoints ......... 19 2.3.3 Actions on Condition Reports - Failure to Conform to NRC Commitments . . 21 2.3.4 Resolution of Audit Findings - OrganizationalIndependence of Radiation Pro t e c tio n M a n a g e r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 2 3.0 S ELF-ASS ESS M ENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 3 3.1 Self-Assessment Program .......................................23 3.2 Operator Work Arounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.3 Self-Assessment of Design Basis issues .............................26 3.4 Engineering Assessment of Control Room Design Review .................33 4.0 INDEPENDENT OVERSIG HT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.1 Effectiveness of Nuclear Oversight - Audits and Evaluations ...............34 4.2 Effectiveness of Nuclear Oversight - Quality Control . . . . . . . . . . . . . . . . . . . . . 41 I 4.3 Followup of Previously Inspection Findings - Nuclear Oversight .............42 ii

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

t i

l i

4.4 Performance of the Nuclear Safety Assessment Board ...................45 t

4.5 Performance of the Plant Operations Review Committee . . . . . . . . . . . . . . . . . . 46 i

4.6 Performance of the Site Operations Review Committee . . . . . . . . . . . . . . . . . . . 47 4.7 Performance of the Independent Safety Engineering Group ................48 Management Meetings . . . . . . . . . . . . . . .. . . . . . . . . . . ...................50 l

X1 Exit M e e tin g S u m m a ry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 0 l

l iii

l l

l l

EXECUTIVE

SUMMARY

Millstone Nuclear Power Station Corrective Actions Team Inspection 50-423/97-82 Operations - Corrective Actions Processes Management Processes and Systems

• The plant staff's clear understanding of management's expectations was considered a management strength. However, management's long term direction of plant personnel is incomplete because the strategic plan (which represents long-term direction) was in l draft form (Section 1.1). l I

• Communications between groups and departments in formal meeting settings showed a questioning attitude and command and control by senior managers. Teamwork training at the level of the plant worker has not been developed (Section 1.2). l

• Plant rnanagement was effective in its efforts to encourage plant personnel to identify problems and the plant staff feels that management is receptive to problems brought forward. Individuals generally characterized the environment as improved and currently l receptive to problem identification. There was no reluctance or reservation expressed by individuals to identify problems (Section 1.3.1).

1 I

• The licensee is adequately responsive to specific harassment, intimidation, retaliation or .

discrimination (HIRD) case needs. The Employee Concerns Program, the Employee Concerns Oversight Panel and the Safety Conscious Work Environment programs are positive contributions to the overall process. However, NNECO management has not l been fully effective in dealing with trends and common causes for HIRD allegations l generated organization-wide to Employee Concerns Program. The Safety Conscious l Work Environment processes have not yet been proceduralized (Section 1.3.1).

• The performance monitoring program was good. The high number of human errors was a weakness that the licencee needs to examine further. The licensee's failure to conduct five fire protection surveillance and exclusive reliance on the 1-hour roving fire watch were inappropriate long term fire protection compensatory measures. The failure te conduct the surveillance sets a low standard of performance for plant personnel (Section 1.4).

• Management's commitment to resolving safety committee recommendations, audit i

findings, assessment findings and open issues was satisfactory (Section 1.5).

Corrective Actions I

• Overall, the corrective actions program is functioning, but it is clear that the program will continue to require careful monitoring by NNECO management to ensure sustained performance. The team's general conclusions regarding the adeqecy of the licensee's corrective action program were that the program elements concerning identification, condition report (CR) initiation, and CR processing were performing vee:1. The threshold iv t_________________________________________________________. _ _ _ ._ _ _ _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _

l l i 1 l l l

l for identification of issues, deportability reviews, generally the assignment of severity i level and corrective actions were timely and appropriate. The Corrective Action Program i

! elements concerning root cause, corrective actions and f ailure recurrence were considered to be operating at an acceptable level, but with room for improvement. )

However, the team found a notable number of discrepancies in a relatively small sarMle size of CRs, after NNECO had completed their own extensive self-assessment preparing l for this team inspection. For example:

l i

1) The Corrective Actions Program lacked controls over combining similar Condition l Reports such that their issues were preserved and that they maintained their I appropriate significance level. This a program weakness (Section 2.1).

2) Several Condition Reports (CRs) were inappropriately classified at a Significance l Levellower than that prescribed by station procedures. Root cause analysis was often narrowly focused and sometimes waived. The root cause analysis quality has J

improved through 1997, but overall performance is somewhat mixed (Section 2.2), i i

3) Some corrective actions were narrowly focused and missed the opportunity to j detect additional existent problems. All corrective actions were not taken regarding l Site Operations Review Committee review of Independent Safety Engineering Group l procedures. This is a violation. The condition report risk significance classification process is not linked to probabilistic risk analysis or maintenance rule risk ranking.

There was no evidence that risk information was used in the prioritization of issues j (Section 2.3). )

4) Condition Reports (CRs) which were written for failed NRC commitments were not classified appropriately. This is a violation (Section 2.3.3).

• Root cause determinations and corrective actions for recurrent boric acid transfer pump problems were inadequate. The deportability evaluations were incomplete. Operating experience was not considered. A potential unreviewed safety question resulting from non-conservative boric acid tank level technical specification wes not identified. The failure to identify and correct the air binding of the boric acid transfer pumps is a violation (Section 2.3.1).

• The licensee f ailed to take corrective actions for a technical specification related issue concerning the organizational independence of the Radiation Protection Manager (RPM).

This is a violation (Section 2.3.4).

• Operator work-around issues, which involve plant material deficiencies, were not included in the corrective actions program (Section 3.1).

Self-Assessments

• The self-assessment program was being adequately implemented and the associated l

i recommendations were beneficialin identifying areas for enhancement and improved

! performance. An Automated Work Order (AWO) associated with a modification was I closed prior to completion of all specified work. This is a violation (Section 3.1).

v

• Acceptable processes for Final Safety Analysis Report change control are being applied (Section 3.2).

• A Condition Report Action Request concerning review of Design Change Notices was closed without accomplishing the specified corrective actions. This is a violation (Section 3.2).

• The Master Setpoint List was found to contain incorrect information. Additionally, the methods for control and documentation of setpoint information appeared inconsistent, difficult to retrieve at times and had the potential for allowing incorrect information to persist. The findings of an engineering self-assessment of this area have not been addressed to date. This is a follow-up item (Section 3.2).

• Violations were identified concerning Design Basis Summary documents and concerning the Safety Functional Requirements Manual (Section 3.2).

• Updates of 3,000 Category 2A drawings are planned to take two years. This drawing category has a 90-day guideline for incorporating outstanding changes (Section 3.2).

• SIL ltem No. 41, concerning the ACR-7007 - Event Response Team findings is closed (Section 3.2).

Independent Oversight

• Nuclear Oversight was effective in performing audits, general plant oversight and work surveillance activities. Considerable improvement was noted since independent assessments identified considerable weaknesses two years ago in the performance of OA activities (Section 4.1).

• Procedures and audits have improved. Audit findings are more meaningful and there is good control in the follow-up of audit findings (Jection 4.1).

• Audit scheduling has improved and is adequate but there are still weaknesses that have to be resolved (Section 4.1).

• There is much better communication between the line organization and nuclear oversight (Section 4.1).

• Quality Control was generally effective in performing the required in-plant inspections.

The OC support group was effective in establishing and standardizing the use of QC hold points in work packages (Section 4.2).

• SIL ltem No. 73, concerning audits of the technical specifications, is closed (Section 4.3.1).

I

• SIL ltem No. 73, concerning the effectiveness of the Nuclear Oversight Organization, is closed (Section 4.3.1).

• Sll item No. 41, concerning trending of Nonconformance Reports (NCR) reports, is closed (Section 4.3.2).

The Nuclear Safety Assessment Board (NSAB) was effective in reviewing activities on-site and identifying potential nuclear safety issues. The implementation of the NSAB met the technical specification requirements. Hovvever, the 1997 resolution of an issue of membership qualifications did not fully satisfy the technical specifications intent as presented. This was a weakness in the implementation of technical specification section 6.0 requirements (Section 4.4).

• The Plant Operations Review Committee (PORC) and the Site Operations Review Committee (SORC) were effective in accomplishing the reviews required by technical specifications (Sections 4.5 and 4.6).

• The Independent Safety Engineering Group (ISEG) has made considerable progress, albeit at the expense of performing safety reviews,in reducing the backlog of operating experience (OE) reviews. The backlog of OE reviews on Unit 3 has been reduced from several hundred to approximately 40. However, the number of ISEG safety reviews done in 1997 was only 12, down frorn 24 the previous year (Section 4.7).

vii L____________________-_______________

Report Details l

l. Operations l 1.0 MANAGEMENT PROCESSES AND SYSTEMS The team evaluated the processes and systems that Millstone Unit 3 managers use to identify, correct and prevent problems. The team reviewed management directions, goals and expectations; organizational communications and teamwork; managerial and supervisory encouragement of problem identification; and performance monitoring. The NRC team members conducted interviews, attended meetings and reviewed licensee documents. The interviews were with alllevels of plant personnel, including senior managers, middle managers, supervisors and nonmanagerial personnel.

1.1 Management Directions, Goals and Expectations

a. Inspection Scope The team assessed the effectiveness of the process by which Northeast Nuclear Energy Company managernent provides the necessary direction to the plant staff to prevent problems. The team evaluated the organization's high-level goals and expectations.

1

b. Observations and Findings Licensee management has used numerous means to communicate and reinforce its expectations of plant personnel. The various means incleded " Nuclear Group Policies and l Standards"; a daily newsletter, which publishes items of mterest, including operating experience items; posters enumerating management's expectations, vyhich were visible throughout the plant; and daily, weekly and monthly meetings. The information covered in daily meetings included condition reports, emerging issues, plant status, management expectations, organizational changes, plant modifications and priorities. The team observed that daily meetings showed good interdepartmentalinteractions, a questioning attitude by participants and command and control by senior managers.

The interview results showed a common understanding of management's expectations pertaining to identifying and correcting problems and identifying and addressing safety issues. Plant personnel at different levels of the organization demonstrated a high degree of compliance with areas of management's expectations, especially in the area of problem identification.

Licensee management's most formal direction to plant personnelis a strategic plan (using a top-down approach) and an associated picture that indicates where the plant is headed.

Both of these items are currently in draft form. The licensee explained that it expects to issue its strategic plan after recovery efforts are completed for Units 2 and 3. The licensee has issued a draft "Long-Term improvement Plar" (which uses a bottom-up approach),

along with vision, mission and strategic focus area statements, as part of its periodic report to the NRC titled " Progress Toward Restart Readiness and Long-Term improvement at Millstone Station - Northeast Utilities Briefing for the U.S. Nuclear Regulatory Commission," dated February 11,1998. The licensee has also issued " Nuclear Group

2 Policies and Standards." The Policies and Standards" document lacks a " Nuclear Group Mission and Vision" statement because it is still being developed. The team considered the draft status a weakness of an otherwise good document. The team also considered it a weakness of management direction that the strategic plan is currently in draft form because such a plan helps to guide both activities and decision-making.

c. Conclusions The team concluded that the plant staff's clear understanding of management's expectations indicated management strength. The team concluded that management's long-term direction of plant personnel needed improvement because the strategic plan (which represents long-term direction) was in draft form.

1.2 Organizational Communications and Teamwork

a. Inspection Scope The team evaluated organizational communications and teamwork, including interdepartmental relationships and interfaces. The team assessed both vertical and lateral (horizontal) communications. It verified that communications are adequate to properly identify and characterize safety-significant issues. It also ast.essed whether communications between organizations were adequate to properly address safety issues.

b. Observations and Findings (1) Organizational Communications (a) Interdepartmental Relationships and Interfaces As noted in paragraph 1.1 above, the team observed that daily meetings showed good interdepartmental interactions, a questioning attitude by participants, and command and control by senior managers. However, the team found that horizontal communications were not as effective or as free flowing as vertical communications. The team found that communications within groups were more effective than communications between groups.

Interviews indicated that communications between groups were greatly dictated by the necessity of completing a task and communications were more free flowing when individuals had some prior relationship with one another. Some impediments to horizontal communications included the following: time constraints, busy schedules, the fact that communications are often issue-driven, the need to resolve an issue, cultural issues and respect issues.

Also, there are pockets within interdepartmental groups that the team found some interdepartmental relationships and interf aces that were under stress (e.g., maintenance and oversight, had past and current conflicts). The team also found that senior plant management was aware of these conflicts and had taken effective immediate actions to include long-term actions to minimize the conflicts.

l 3

(b) Vertical Communications The team found that vertical communications were a strength of the plant organization.

Interviews indicated that vertical communications were especially strong with respect to using the corrective action system and sending constructive inputs up the chain of  ;

command. The team also found that upper management's ability to effectively communicate its expectations down the chain of command was a management strength.

(2) Teamwork The team determined that team-building initiatives had been completed beginning with officers, then directors, followed by managers and first-line supervisors. The team saw evidence of teamwork in several regularly scheduled interdepartmental meetings. Interview J results indicated that teamwork initiatives are currently on hold until after startup of Unit 3. l Interview results did not identify any teamwork initiatives at the level of the plant worker.

~ The absence of teamwork initiatives at the level of the plant worker may have resulted in l unfavorable conditions, including: some groups are in conflict (most notably oversight and maintenance), not everyone is familiar with conflict resolution, sometimes interpersonal i conflicts are not resolved and no formal process exists for rotational assignments.

l

c. Conclusions The team concluded that communications between groups and departments was effective in formal settings, but needed to be improved for day to day interactions. The team concluded that teamwork initiatives at the first-line supervisor and higher were good initiatives and needed continued reinforcement. The team also concluded that teamwork at the level of the plant worker needs to be improved.

1.3 Encouragement of Problem identification by Managers and Supervisors

a. Inspection Scope The team evaluated whether managers and supervisors encourage employees to identify problems and whether the staff believes that management is receptive to the problems being brought forward,

b. Observations and Findings Observations, interviews and plant surveys show that managers and supervisors encourage l

employees to identify problems. Interviews also indicated that the plant staff believes

! management is receptive to problems being brought forward,

c. Conclusions The team concluded that plant management was effective in its efforts to encourage plant personnel to identify problems.

l

t 1.3.1 Management's Encouragement and Receptiveness in Problem Identification by Individuals

a. Inspection Scope The team evaluated the NNECO problem resolution processes (line management practices, corrective action program and others) to determine their effectiveness in providing a vehicle for encouraging identification of, readily accepting and adequately resolving problems and issues identified by individual employees as pathways preferable to Employee's Concern Program (ECP) or the NRC allegation process. The team also attempted to determine if managers and supervisors encourage employees to identify problems and if the staff feels management is receptive to problems being brought forward.

b. Observations and Findings The team conducted employee interviews at all levels of the organization to determine workers' perceptions of management's efforts and communications intended to enhance problem identification. Performance measurement data for problem identification and documentation associated with individual-identified problems (condition reports, employee concern case files, self-assessment results and others) were reviewed.

The employee interview results indicated that previously existing barriers to problem identification had been largely climinated and no major barriers to problem identification were found. Personnel from organizational units both with and without histories of such barriers to problem identification were interviewed and generally characterized the environment as improved and currently receptive to problem identification. No reluctance or reservations were expressed by the individuals with regard to their identification of problems to line management, condition reports (CR) process, Employee Concerns Program (ECP) or NRC. Most of the individuals interviewed indicated that they had initiated CRs either personally or through referral to their supervision. Several of the individuals stated that the current working environment also supported escalating concerns and problems to management above their direct supervision if they believed that to be necessary. The NNECO performance indicators and condition report program statistics reflect reasonable levels of employee participation, which corroborated the interview results.

The team also reviewed the Employee Concerns Program (ECP) Program, the Safety Conscious Work Environment (SCWE) Program, the Employee Concerns Oversight Panel (ECOP) and other management initiatives for NNECO activities and response actions taken for previously existing barriers to problem identification. The licensee's handling of individual alleged cases of harassment, intimidation, retaliation or discrimination by the ECP and SCWE organizations appears to be adequately responsive to specific case needs. Both

technical and human behavior and performance problems were generally well addressed in I the case files reviewed.

However, a recent NRC inspection 50-423/97-2120f these programs in accordance with NRC IP-40001 found that the SCWE processes were not formalized. The lack of SCWE program structure (procedures, formal processes and documentation requirements) resulted

in many of the management actions being handled directly, on an ad hoc basis, by ,

recovery officers and senior managers.

The licensee uses incoming ECP allegations, employee culture surveys, leadership surveys and ECOP activities to identify barriers to problem identification and harassment, intimidation, retaliation or discrimination (HIRD). ECOP oversight activities and surveys are used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems appear to be positive contributions to the overall process.

While the licensee has become effective in identifying these barriers, it has not been fully effective in addressing a continuing high incidence rate of HIRD allegations coming to the Employee Concerns Program. The ECP, ECOP and SCWE mechanisms are effective,  !

especially for the more significant issues identified as problem areas. However, as I discussed in report section 1.3.2, NNECO management has not been fully effective in i dealing with trends and common causes for HIRD allegations generated organization-wide l to ECP. A significant backlog of HlRD allegations was pending investigation and the i backlog and emergent HlRD allegations had not been analyzed by NNECO for broad trends i or patterns, as well as common causes. As a result, the actions taken to date had not been effective in assessing the nature and substance of the continuing high incidence rate of HIRD allegations. The monthly HIRD allegation incidence rate and the frequency of l actual and potential chilling effect events at Millstone has not significantly diminished during 1997-1998 and continues to represent potentially serious barriers to a suitably self-critical organization and appropriate safety conscious work environment. The licensee's effectiveness in reversing the trend of HIRD and preventing its recurrence is further .

discussed below.

c. Conclusions Observations and interviews show that managers and supervisors encourage employees to identify problems. The plant staff believes management is receptive to problems brought forward and individuals generally characterized the environment as improved and currently receptive to problem identification. There is no reluctance or reservation expressed by individuals to identify problems.

The handling of individual HIRD cases by the licensee is adequately responsive to specific case needs. Both technical and personal problems are generally well addressed. The ECP and the ECOP oversight activities are used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems.

These are positive contributions to the overall process. These mechanisms are effective, especially for the more significant issues identified as problem areas. Further, the SCWE processes have not yet been formalized. That is, the program lacks structure in the form l of procedures, formal processes and documentation requirements.

l 1.3.2 Effectiveness of Problem Resolution Processes in Resolving Employee identified Concems

a. Inspection Scope

6 Assess the effectiveness of the process by which NNECO management provides problem resolution to employee-identified concerns.

b. Observations and Findings The team conducted interviews, reviewed corrective action program data and reviewed the licensee's technical resolutions, SCWE-related responses and long-term follow-up for problems identified by condition reports, employee concerns and Employee Concerns Oversight Panelissues and NNECO department self-assessments. This included a sample of condition reports and their technical and human performance / behavior resolution activities that resulted from employee concern cases and self-assessments.

The team also reviewed ECP Cases, SCWE Problem Area response plans and a sample of CR corrective actions to determine if licensee actions were generally responsive to problems and adequate. However the e were some exceptions:

(1) SCWE Problem Areas and Potential Problem Areas identification methods for HIRD issues include the NNECO Leadership Surveys, Culture Surveys, ECP case intakes and ECOP identifications and referrals. These mechanisms appear to be effective, especially for the more egregious issues identified by NNECO management as HIRD " Problem Areas." Documented actions plans are used to define and manage the remedial actions for the Problem Areas. A sampling review of these plans by the team determined that action plans are generally effective at remediating both the technical issues and human performance and behavior issues.

Other HIRD issues that are perceived by NNECO management to be less egregious or that are perceived as minor problems having the potential to eventually become " Problem Areas" are handled more informally by senior and middle management. These issues are identified by line management via normal oversight and activities, review of ECP intakes and SCWE daily meetings which include dialogue with ECP, ECOP, Human Resources, Legal, the ECP/SCWE Independent Third Party Oversight Contractor (Little Harbor Consultants, Inc.) and other program participants. No documented action plans are used, but the team found that management had taken actions in direct response to such potential HIRD issues and were monitoring the effectiveness of the actions.

Notwithstanding the specific HlRD and HIRD-potential management response actions taken to date for HIRD-related problems, NNECO has been unsuccessful in reducing the ongoing emergence rate of HIRD allegations to ECP and eliminating the causes of HIRD. An average of about 54% of all ECP allegations included HIRD and about 26% of the total allegations involved HIRD associated with nuclear safety-related protected activities subject to 10 CFR Part 50.7. The fraction of the total cases not involving 10 CFR Part 50.7 typically involve other forms of HIRD, e.g., age, sex or other discrimination, harassment or intimidation over work rules, compensation, benefits, etc. The overall frequency of personnel behaviors that can be construed as HIRD, regardless of their source or subject, has been identified by NNECO and NRC as a serious concern relative to the establishment of a healthy SCWE. While NNECO notes that only three 1997-98 ECP cases have involved l

l

7 confirmed 10 CFR Part 50.7 HIRD issues, investigation of many of the alleged cases remains open.

NNECO has performed extensive re-evaluation of the data to ensure their categorization is correct; re-evaluation resulted in no substantive changes in categorization. NNECO acknowledges that HIRD in non-50.7 environments represents a present potential of 50.7 HIRD. The team discussed and evaluated the rnanagement actions planned and taken with the Unit 3 Vice President, the Vice President-Engineering responsible for SCWE, the SCWE management team, the ECP assistant director and others. Except for specific HlRD ECP cases or explicitly identified SCWE Problem Arcas, management does not appear to be taking effective actions focused to reduce the overallincidence of HIRD allegations.

(2) Measuring and Test Equipment Program issues Several of the ECP cases reviewed by the team identified programmatic and implementation problems in the NNECO programs for control of measuring and test equipment (M&TE). The team initially reviewed the two M&TE related CRs (CR-M3 1292,5/2/97 and ACR M3-97-0150,1/15/97) issued in response to the ECP cases, finding that they also related to several other related CRs. Further team review of these and a multi-year history of M&TE-related CR data found numerous NNECO-identified examples of M&TE program deficiencies. Additionally, Nuclear Oversight audit findings substantiated the CR issues. Discussions with NNECO and NRC Region I personnel indicated that the adverse performance history existed at least from 1992. The issues concerned mis-handled M&TE, for example f ailure to perform impact reviews for M&TE found out of calibration.

NNECO was requested to provide further information on actions taken to address apparent adverse trends and a meeting was held on February 18,1998, with the newly appointed Metrology Lab Supervisor, the new program owner. The supervisor advised that all M&TE issues had been rolled up into ACR M-1-96-0614,which has been used as a vehicle to address all contemporary problems and drive development of a completely new M&TE program.

On February 25,1998, following the end of the inspection, NNECO provided the team with ACR M 1-96-0614 and its associated corrective action documentation, which indicated that the M&TE Program had been completely rewritten and had just become effective in early January 1998. The new program provided for a new Metrology Lab to provide central control for equipment and activities and new training for all program implementors / users. Fullimplementation had not yet been achieved but was expected to be completed soon with an Effectiveness Evaluation planned for mid-1998.

Based on the unacceptably long history and broad extent of problems with the M&TE program and the comprehensiveness of the changes in the new program, the licensee's specific corrective actions and fullimplementation of the program will be reviewed during the NRC Operational Safety Team Inspection. At that tirne, the M&TE program performance may be reviewed relative to the Nuclear Oversight audit findings,

c. Conclusions l

l

8 The handling of individubl HIRD cases by the Employee Concerns Program and the Safety Conscious Work Environment (SCWE) program is adequately responsive to specific case needs. Both technical and human-side problems are generally well addressed. The ECP case intakes and the Employee Concerns Oversight Panel (ECOP) oversight activities and surveys are used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems. These are positive contributions to the overall process. These mechanisms are effective, especially for the more significant issues identified as Problem Areas. However, NNECO management has not been fully effective in dealing with trends and common causes for HIRD allegations generated organization-wide to ECP.

1.4 Performance Monitoring

a. Inspection Scope The team evaluated performance monitoring (performance indicators), including management information systems employed to evaluate the following programs: corrective action, root cause analysis, self-assessment, independent oversight and operating experience. They evaluated the effectiveness of the performance measures process, and assessed the quality of the information on performance that is given to management. The team examined the licensee action plans for correcting areas identified by performance indicators that warrant management intervention.

b. Observations arv' Findings The various perf t.mance indicators used to evaluate the programs of interest were considered excellent. The licensee has appropriately addressed adverse trends identified in the fourth quarter report concerning compliance of the maintenance department with procedures, valve and breaker alignment issues and tagging errors of the operations department and surveillance testing. The team's analysis of licensee event reports (LERs) found that Unit 3 had 16 human performance-related LERs in 1997 versus a national average of six human performance-related LERs.

The team determined that the licensee intentionally does not perform five fire protection surveillance because the operations department does not have the manpower to conduct the surveillance. The following condition reports (CRs) have been written in connection with the surveillance: M3-97-3035, M3-97-3981, M3-97-4246, M3-97-4394 and M3 4618. The surveillance which expired were: fire protection water deluge and sprinkler systems on September 11,1997, Fire Zones panels 6E and 6F and detectors on November 11,1997, Fire Zone panel 4E on November 25,1997, the computer room on December 4, 1997 and fire seals in the Auxiliary Building, Control Building, Control Room, Emergency Safeguards Features Building, Emergency Diesel Generator Building and Main Steam Valve Building.

The licensee has implemented hourly patrols as a compensatory measure prescribed by the Technical Requirements Manual ' Action' statement for the applicable ' Limiting Condition for Operation.' The licensee stated that the subject fire protection surveillance would be performed by Station Fire Brigade personnel when they are trained. The team found that

9 the failure to conduct the surveillance was inappropriate because it sets a low standard of performance for plant personnel. The team also found that relying exclusively on the 1-hour roving fire watch as an interim compensatory measure without other compensatory measures was a weakness, as noted in loformation Notice 97-48," Inadequate or inappropriate Interim Fire Protection Compensatory Measures."

c. Conclusions The team concluded that the licensee was responding to the performance monitoring indicators. The team observed that there are a high number of LER-retated human errors.

They concluded that the licensee's intentional suspension the conduct five fire protection surveillance was inappropriate because it conflicted with their stated expectations for

, performance and it also sets a low standard of performance for p! ant personnel, l

1.5 Management's Commitment To Resolve issues

a. Inspection Scope The team evaluated management's commitment to resolve safety committee recommendations, audit findings, assessment recommendations and open issues.

l

b. Observations and Findings l Management's commitment to resolve issues was ase.essed by sampling safety committee

( recommendations, audit findings, self-assessment recommendations and engineering

! evaluations. The team then evaluated management support for resolution of these issues.

Particular emphasis was placed on issues which management directly influenced. The findings are stated within report sections on corrective actions, self assessments and l

independent oversight.

The team found that NNECO management generally offered strong support for issue resolution. However, two issues were identified where corrective actions failed to resolve identified conflicts with the technical specifications. These related to the organizational l independence required for the Radiation Protection Manager and also for the members of an Independent Safety Engineering Group review. They are addressed in report Sections 2.3.4 and 4.7, respectively. l l

c. Conclusions l

l Overall, the team concluded that management's commitment to resolving safety committee recommendations, audit findings, assessment findings and open issues was satisfactory.

2.0 CORRECTIVE ACTIONS l

2.1 Corrective Action Program

a. Inspection Scope I

l

10 The team assessed the adequacy of the corrective action program. This assessment included the evaluation of programs for the identification, analysis and resolution of plant deficiencies,

b. Observations and Findings The team conducted discussions with the licensee and performed document reviews of the Corrective Actions Program contained in the recently issued program procedure (September 30,1997) revision 5 to RP 4, " Corrective Action Program." Prior to 1995, the program, l employing what was known as the Plant incident Report, failed to perform satisfactorily.

l This was a program that captured an average of only 300 !tems per year for Unit 3 at a l high threshold level for events or reportable conditions. This program was superseded by the site-wide, Adverse Condition Report (ACR) which improved the " capture" threshold to approximately 4,000 items per year. In 1996 and 1997, the licensee's program was moved closer to industry practices with revisions 2,4 and 5 (revision 3 was never issued).

These revisions, among other changes, resulted in a multi-disciplined management review, as well as the requirement that the shift manager review discovered conditions for l operability and deportability. The revised condition reporting program also strengthened i accountability, provided for enhancement items as well as adverse conditions and defined management expectations.

The team's general conclusions regarding the adequacy of the licensee's corrective action program were that the program elements concerning identification, CR initiation and CR

processing were performing at a goed performance level. The CR program elements concerning root cause, corrective actions and f ailure recurrence were considered to be operating at an acceptable level but with room for improvement. The remaining element, effectiveness, involves trending and self-assessment, as wel1 as effectiveness review. The first two attributes (trending and self-assessment) show indication of being performed at an acceptcble level. While the effectiveness reviews cannot yet be fully evaluated due to the newness of the current program requirements.

The team evaluated the performance of the licensee's program for identification of adverse conditions. As indicated by the range of CRs reviewed and discussions with a broad spectrum range of plant personnel, the team concluded that the licensee has attained a low l threshold for initiating CRs. The fourth quarter continued the high volume trend of CRs initiated for Unit 3, with a total of 1,621 CRs. An increased awareness by plant staff and a decreased threshold for initiating a CR, appear to be the reasons for the high volume of CRs initiated. An additional reason for the high volume of CRs is the generation of numerous findings from the Independent Corrective Action Verification Program.

It was also the team view that the licensee's analysis of CRs, while generally adequate, tended to be narrowly focused. Several examples to support this are discussed below.

Among those examples is the resolution of closure problems with High Energy Line Break (HELB) doors.

The team also found that the Corrective Actions Program lacked controls over combining similar Condition Reports such that their issues were preserved and that they maintained their appropriate significance level. This weakness was evident in the handling of multiple

11 condition reports concerning deficiencies in the Nonconformance Report process. This controlissue was documented in great detail as CR M3-98-0309. In this case, a series of CRs documenting implementation issues within the Non-Conformance Report (NCR) process were combined into a single Significance Level 1 CR, M3-97-3710. No specific root cause was performed for the CR, but the analysis was transferred to another Level 1 CR, M3-97-0845, which did not concern NCR implementation issues. Additionally, CR M3-97-3710 corrective actions were closed to a Level 3 CR, M3-974468, which was written to track recommendations from a self-assessment.

The team found that the controlling administrative document, RP 4, Revision 5, does not establish controls over cornbining CRs to preserve their subject material and Significance Level. These issues were discussed with licensee management who outlined their plans to revise the program requirements stated in RP 4. The team considers this as a program weakness which will be reviewed during a future inspection (IFl 50-423/97-82-01).

c. Conclusions The licensee has attained a low threshold for initiating Condition Reports. An increased awareness by licensee personnel and a decreased threshold for initiating a CR have resulted in a high volume of CRs initiated.

The team also found that the Corrective Actions Program lacked controls over combining i similar Condition Reports such that their issues were preserved and that they maintained their appropriate significance level. This a program weakness.

2.2 Corrective Actions - Classification and Root Cause Analysis

a. Inspection Scope The team reviewed a sample of root cause analysis and equipment failure evaluations to determine the adequacy of the process to classify and analyze condition report issues. For less significant issues, the team reviewed a sampling of the apparent cause determinations.

They also independently verified condition reports for significance and that apparent root cause determinations had been performed where required.

b. Observations and Findings For the Level 2 and Level 3 CRs reviewed, the team did not disagree with the apparent causes. Although generally adequate, the team's review of Level 1 CRs disclosed a tendency for root cause analysis to be narrowly focused and several cases in which the root cause analysis was waived. Although exercising waivers of root cause analysis are within procedural constraints, their over use can also lead to missed opportunities for case-specific, in-depth evaluation. Several examples are:

• In the case of CR M3-97-0652,which described how design interface distribution and transmittal control of design information did not meet the requirements of Criterion ill and V of Appendix B, the root cause analysis was waived by the Multi-Discipline Management Review Team (MRT) with no accompanying explanation. The corrective

12 actions involved training Unit 3 design engineers on design change record / minor modification (DCR/MMOD) requirements and emphasized the need for attention to detail.

As determined by the team, the corrective actions appeared to be narrowly confined to the items associated with the DCR deficiencies.

• For ACR M3-97-0558, dated February 20,1997, the licensee's design basis verification program document for the Chemical :d Volume Control System had non-conservative assumptions related to maximum temperatures for the letdown heat exchanger and charging flow. The root cause was waived and referred to other CRs with similar situations. The team regard this waiver as a missed opportunity to thoroughly evaluate the cause of the incorrect assumptions.

• Another instance involved ACR M3-97-0409, a Significance Level 'B' ACR, dated February 4,1997, that documented concerns for sump water level calculated head losses. Although the cause of the event was addressed in LER 97-015, no root cause was performed for the CR Nevertheless, the proposed corrective actions appear adequate to resolve identified design deficiencies and a modification review is scheduled following reanalysis with actions to be completed prior to mode change.

• The root cause analysis for a diesel low lubricating oil pressure trip was addressed in ACR 10428, dated June 6,1996. The analysis failed to identify a root cause.

However, there were obvious implications that a valve had been misaligned or leaked because 58 strainer was found half full of oil. Valve manipulation scenarios that would have caused an engine trip were not developed as part of the analysis.

• The root cause for MOV calculation errors addressed in CR M3-96-0833, dated February 11,1997, was thorough. However the CR closure, dated December 31,1997, waived the effectiveness review to the MOV program periodic self-assessments. The team observed that this approach does not ensure that a corrective action effectiveness review will be completed during the next self-assessment, for example there was no specific review plan developed which considered the weaknesses and other f actors identified in the root cause analysis.

• The root cause analysis for loss of the component cooling water system addressed in CR MP3-96-0919, dated December 15,1997, failed to analyze and address corrective actions for failed barriers concerning informalinstallation testing by the I&C group without a procedure, lack of post maintenance / modification testing and a procedural  !

non-compliance of failure to have vendor participation. Loss of all three of these barriers I was evident from the information in the root cause analysis. However corrective actions were not developed for these issues.

l l

l

• The material within seven CRs, concerning multiple electrical separation issues, was gathered under one root cause analysis, CR M3-96-1337, dated January 2,1997. The l

documentation of that analysis failed to demonstrate thorough treatment of the issues j encompassed by the analysis. For example, the root cause analysis for item number 3 of ACR M3-96-1287 stated that the electrical designer responsible for a design change  ;

examined one of the deficient conditions and "he stated that it looks like it was missed."

1

[ l 1

l

13 However, there were no corrective actions to establish standards for design engineering field inspections.

• The team noted that some of the root cause analysis associated with the CRs selected for review were well done. Specifically, the team found that the root cause analysis were good for: CR M3-96-Oc '3, dated February 11,1997, CR M3-97-0066 and M3 0132, dated January 31,1997, M3-97-0119 and M3-97-0161, dated February 14, 1997, M3-97-0709, dated April 1.,1997, M3-97-0908, dated April 21,1997, M3 1821, dated July 8,1997 and M3-98-0200, dated February 9,1998. This reflected a generalimproving trend of root cause analysis through 1997.

The team also noted a tendency for classification of CRs at a lower Significance Level. In addition to the issue addressed in Report Section 2.3.3, they also found that CR M2 0510, dated April 1,1997s documented an Independent Safety Engineering Group (ISEG) activity that identified significant work control issues in the high-voltage switchyard.

These issues involved personnel safety and potential loss of off-site power. ISEG promptly l addressed the issue by stopping work in the switchyard. However, the CR was not I initiated until two weeks later. Once initiated, the CR was downgraded to a Level 3. The justification for the Level 3 assignment was that the facts associated with the issue had been reviewed and actions taken prior to the issue of the CR. Further, this previous review determined that no procedural violations actually occurred. The evaluation done in response to the CR minimized the significance of the issues. Based on the potential safety significance of the issues and the fact that the ISEG, an oversight group, developed diverging opinions of procedural requirements, the NRC concluded that the assignment of the Level 3 evaluation was inappropriate and served to further minimize the issues. The ,

amount of review ectually performed was wellin excess of the review required for a Level 3 issue and supports the NRC conclusion,

c. Conclusions The team's review of CRs disclosed several CRs that were inappropriately classified at a lower significance level. The team also noted a tendency for the root cause analysis to be narrowly focused and identified several cases where the root cause was inappropriately waived. Although the quality of the root cause analysis had improved through 1997, overall performance was somewhat mixed.

2.3 Corrective Actions Effectiveness

a. Inspection Scope The team evaluated the technical resolution of a sample of tafety significant issues for timeliness and effectiveness of corrective actions, as well as the backlog of open condition reports, to verify that safety significant items were being tracked to completion. The team's evaluations also included interviews with supervisors regarding the closure rate and reviewing the process to prioritize corrective actions based on risk,

b. Observations and Findings I _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

14 During the inspection preparation trip in December 1997, the team questioned the apparently low closure rate associated with the licensee-defined, high-priority Level 1 CRs.

At that time, a review of the approximately 260 items classified as Level 1 and by the licensee's definition, a "Significant Condition Adverse to Quality," showed only 25% of the items to be closed. Subsequently, various supervisors, including the root-cause supervisor and several QA supervisors, were interviewed regarding the CR process in general and the apparent low closure rate in particular. From these discussions and a review of the current status of the Unit 3 CR backlog (as of February 16,1998), the following information was obtained:

Section 1.8,5 of the licensee's procedure, RP 4, Corrective Action Program, states that evaluation due dates of 30 days from determination of assignment duties are established.

The number of overdue evaluations in January 1997 was 825 items. This number had been reduced to as low as eight items by July of 1997. From November of 1997 to present, the average backlog number has been about 20-30 items. This reflects a partial recovery of the management of this portion of the process. A very high volume of CRs has been and currently are being initiated; as the volume of CRs being generated moderates, it is anticipated that this average backlog number will be further reduced.

Good efforts also have been made in the reduction of restart-related CRs. Starting in the same period, January 1997, restart CRs were at 575 items. Due to the licensee's emphasis and resource focus applied to the reduction of the " overdue evaluation" backlog, these CRs increased to more ! ban 1,600 items durjng the period when the evaluations were being driven down. Starting from July 1997, with the " freeing up" of resources from the efforts to reduce the "over-30-days evaluations," the licensee was able to drive down the restart CRs to less than 400 items cunently.

The teams selected a sample of CRs and reviewed the resolution of these CRs to evaluate both timeliness and effectiveness. The team noted that the CR process generally appeared effective and served its purpose. Several areas were noted for further follow-up and are discussed below:

• Many of the CRs had all of the Action Requests (ARs) completed, but the overall CR was still open. The licensee stated that due to the high volume of CRs, as the CMP program is coming to a conclusion, there is presently a backlog of 600 to 700 CRs of this type in the system awaiting closure by the Corrective Actions Department.

• In the computerized on-line CR system, all of the pertinent ARs do not appear as associated with their particular CR. The licensee stated that this is an historical problem that is in the process of being corrected but will still take more time to backfit the appropriate links.

• There were noted to be some gaps in the tracking of Non-Conformance Reports (NCRs) and the ARs associated with NCRs. The licensee was aware of this problem and is in the process of addressing it (reference Memo MP3-CAD-98-004).

The licensee's procedure RP 4 requires CRs to be assigned to one of three significance levels: Significant Adverse Condition (Level 1), Adverse Condition (Level 2) and

l 15 Improvement item (Level 3). The licensee has in use a risk significance classification system that further subdivides risk into four categories. Specifically, Attachment 4, Risk Significance, to RP 4 states that in addition to a determination of whether a condition I adverse to quality is significant, further attribution is warranted to assist the CR Owner and MRT in prioritizing the corrective actions. The MRT is tasked by RP 4, Attachment 9, to review CRs to determine significance levels, risk significance, evaluation criteria, if assigned evaluations have been properly conducted, if investigations and corrective actions are conducted in a timely manner and to validate the effectiveness of Level 1 CRs.

l The four risk significant categories are defined as: Priority 1, Risk Critical -This category would result in consequences that are severe and unacceptable in either human, societal, i

- political, or monetary terms; Priority 2, important to Safety - This category can result in risk to the reactor systems, industrial safety, public health and safety, or the environment; Priority 3, Compliance - This category corrects common and repetitive non-complisnce with regulations, procedures, or normally accepted standards and expectations as defined by station standards and regulatory agencies; Priority 4, Good Management - this category identifies an isolated condition adverse to quality or an opportunity for program enhancement.

The team noted that the CR risk significance is not coupled to the PRA results or to risk ranking by the Maintenance Rule Program. Also, the team did not discover any evidence, through review of RP 4 or interviews, that the licenset was availing itself of information from their Individual Plant Evaluation (IPE), nor from maintenance rule risk rankings, to assign CR classifications. Additionally, there was no evidence that risk information was  ;

utilized for approving extensions to action due dates.  !

l The MRT members review and evaluation of CRs showed that both the level categorization j and the risk-significance priority classification were assigned to Cks generated each day.

Ilowever, discussions with the Corrective Actions Department and others revealed that the risk classification has not been acted upon beyond the assignment of the classification by the MRT. The licensee envisions further implementation of this concept in the future, in  !

that this concept is not a "hard" requirement according to the procedure. However, at this l stage of implementation, it is ineffective in accomplishing its intent and may create a j l wrong mindset regarding the spirit, as well as the rule, of ptbcedural complignce. It also was noted by the team that the nomenclature has the potential to create program confusion in thrq the title "rish significance" is already a defi.ned term with respect to the maintenance rule.

Team findings of several corrective actions issues are in Report Sections 2.3.1 through 2.3.4. In addition to those detailed discussions the team made the following observations:

• There appeared to be some tendency tc schedule some required corrective actions far into the future and to delay or postpone corrective actions.

• That corrective actions for unclosed HELB (High Energy Line Break) doors were n91 timely and were nanowly focused. During January 1997, six Lnel 1 CRs were written on problems involving HELB doors and the failure to close them properly. Corrective actions focused on correcting personnel errors as the resolution to the recurring prob:am

16 of unclosed doors. Subsequently, in August of 1997, CR M3-97-2567 was written.

This CR addressed and described the physical deterioration of the doors and stated that problems ranged from missing gasket sections, damaged thresholds and damaged gasket bars to large gaps around the gasket seals.

The origination of this CR (M3-97-2567)was not driven by the HELB door closure prvblems but rather by a different review (CMP Pt 21, " Engineering Topical Areas Reviews"). This later CR pointed out the lack of a PM program, the lack of regular inspections and the wck of inspection criteria.

• CR M3-97-2898, dated September 2,1997, documented that a nuclear oversight audit identified that procedures, tools and equipment needed to support emergency operating procedures were not evailable in the plant. The corrective actions for this CR were not tied to a key event and were outside of the scheduled restart date at the time they were approved. This issue was also picked up in the NRC review of deferred items and was changed by the licensee to be a mode 2 issue.

• CR M3 97-3974, dated November 11,1997, documented an audit finding that the ISEG operating nperience (OE) procedure, NOOP 3.04 was not reviewed by SORC, as required by technical specificauons. The ISEG response was to get SORC review of the new OE procedure being developed. However, this response was incomplete because the procedure for conduct of the ISEG also needed SORC review by the same technical specification. Therefore this issue is a Violation of 10 CFR Part 50, Appendix B, Criterion XVI (VIO 50-423/97-82-02).

• CR M2-98-0419 documented seven pieces of tape that were identified by the NRC during a Quality Control (OC) foreign materialinspection of the Unit 2 spent fuel pool.

Although, the CR was assigned to reactor engineering as a technicalissue, it was not evaluated by QC as an inspector performance issue.

• Specification SP-EE-149A details the design requirement? for the Safety Parameter Display System (SPDS). The team found that minor modifications to the SPDS have been made over the years. However, the guidance references in the specification are out of date in that they do not contain current guidance on computer-based displays.

l The licensee stated that AR Number 97027958 titled " Review of and Update of Plant Process Computer Specifications" would update the guidance references by December 1,1998.

• The team determined that the post-LOCA cooling (PLC) status tree is not addressed in the Emergency Operating Procedure (EOP) User's Guide, Section 1.6, " Monitoring Status Trees," or Attachment 4, " Control Room Usage of Status Trees." The licensee stated that this issue would be addressed by AR Number 07031064 titled "PORC Commitment to investigate the Overall Feasibility of the Post-LOCA Processing" by December 1, 1998.

c. Conclusions L_.____

17 )

Efforts have been made to reduce the number of CR evaluations open over 30 days. This reflects a partial recovery of the management of this portion of the process. Good efforts also have been made in the reduction of restart-related CRs. Progress on closing Level 1 )

CRs has accelerated since late last year. However, some corrective actions were narrowly focuced that missed the opportunity to detect additional existent problems. The failure to take full corrective action for SORC review of ISEC procedures is a violation.

The CR risk significance classification process is incomplete. At the current state of implementation, it is ineffective in accomplishing its intent. It was also noted by the team that there is the potential to create program confusion in that the term " risk significance" is already in widespread use in the industry for an entirely different program: Probabilistic Risk Analysis and Risk Ranking in the maintenance rule. Furthermore, there was no evidence that the licensee was using risk information in the prioritization of issues in the corrective action process.

2.3.1 Boric Acid Transfer Pump Air Binding

a. Inspection Scope The team assessed the adequacy of the licensee's actions relating to CRs concerning air binding of the Boric Acid Transfer Pumps, which has been a recurrent issue.

b. Observations and Findings Boric acid (BA) transfer pumps (3CHS-P2A & -P2B) are part of the technical specification-required reactivity control systems and provide a boron injection flow path to the Reactor Coolant System. CR M3-97-2943, dated 9/4/97, identified a chronic air binding problem with the pump that periodically rendered them inoperable. The CR noted that the condition had previously been identified in 1992 (PIR 3-92 210);1995 (ACH 3617); and earlier in 1997 (CR M3-97-0715, CR M3-9701011,CR M3-97-0954) but not fully resolved. J NNECO's root cause analysis concluded that air entrainment caused the ? 992 event, vented the pump and required no further action, as it was considered an isolated occurrence. No cause was determined for the 1995 event other than the presence of l trapped air in a horizontal pipe run. Internals of the boric acid storage tank (BAST) batch l tank discharge check valve were removed and a modification was proposed to add an

isolation and manual vent valve to provide improved venting. CR M3-97-0954 identified the recurrence of the problem and initiated actions to revise BASTlow tank level operating l limits and alarm setpoints upward to prevent air entrainment; revise the system Operating Procedure to vent pumps if BAST levels drop below the increased low limits; and to  ;

complete the valving modification originally proposed in 1995. 1 The corrective actions for M3-97-2943 were essentially the same as the earlier 1997 actions above except that they included evaluaticn of a possible modification to re-route the boric acid gravity flow boration piping. The licensee's root cause had concluded that the height of the BA Pump and gravity feed piping connections to the BASTS relative to the  ;

minimum Technical Specification-allowable tank levels had resulted in air entrainment and

18 pump binding. The proposed corrective actions had been approved by management and were awaiting fullimplementation at the time of the inspection.

The NNECO root cause and corrective actions required implementation of BAST Level administrative controls more conservative that the minimum levels required by Technical Specification 3.1.2.5.a, " Borated Water Sources - Shutdown," or 3.1.2.6.a. " Borated Water Sources - Operating." The team requested the licensee's 10 CFR Part 50.59 safety evaluation and 10 CFR Part 50.72 or Part 50.73 reportabliity evaluations for these actions in that they appeared to constitute an unreviewed safety question. Specifically,if the pumps and tanks were operated in accordance with the Technical Specifications, they might be unable to perform their analyzed safety functions as described in the Final Safety Analysis Report.

The licensee advised that the M3-97-2943 proposed administrative controls had not been completely prepared and published. Consequently the 10 CFR Part 50.59 evaluation had not yet been completed and was therefore unavailable. Further, the NNECO Unit 3 Licensing Manager stated the licensee's belief that the above conditions did not represent an unreviewed safety question and that a routine Technical Specification change request was planned for later submittal to NRC. Operation would proceed in accordance with the administrative tank levellimits pending NRC approval of the change. The failure to correct

he air binding of the boric acid transfer pumps and take actions to prevent recurrence of a significant deficiency is a Violation of 10 CFR Part 50, Appendix 8, Criterion XVI (VIO 50-423/97-82-03).

Additionally, the team reviewed potential air or gas infiltration paths with the NNECO System Engineer. NRC Information Notice 88-23, institute of Nuclear Plant Operations (INPO) " Red" Significant Operating Experience Report 97-001 and recent events at a Unit 3 sister plant, Beaver Vailey Power Station, had each identified the potential for accumulation of substantial amounts of hydrogen gas in charging pump sucticn lines. The accumulation resulted from the large pressure drop across the charging pump mini-flow orifices, stripping i dissolved hydrogen from the charged coolant and returning it to the pump suctions. The phenomena has the potential for severe charging pump damage. Because of the location of the BA Pump piping connection to the charging pump suction piping, the team had inquired whether the licensee had considered hydrogen accumulation and migrati3n into the aHected BA piping.

The NNECO System Engineer, responsible for both the boric acid and charging systems, l

was unaware of the operating experience information referenced above and the information had apparently not been considered in the root cause determinations for the CRs. It was l

later determined that NNECO Design Engineering had been assigned responsibility for the j l operating experience items and had preliminarily determined that Unit 3 did not have the I hydrogen problem. The basis for the determinations was not reviewed; the matter was )

referred for further follow-up to the NRC Operational Safety Team thst was concurrently onsite.

On February 19,1998, NNECO issued CR M3-98-0975 which documented inadequacies in the root cause evaluation for CR M3-98-0954in that it did not consider potential  :

introduction of air from the BAST batching process. l l

l

1 19 The licensee further advised that the deportability evaluation performed with the CRs had evaluated the deportability for plant conditions at the times of discovery of the pump inoperabilities but had not adequately evaluated past plant conditions for potential, historical inoperabilities of the BAST /BA Pump boration flow paths. On February 18,1998, the licensee issued CR M3-98-0952, which documented the inadequate historical deportability evaluations. Disposition of the CR was in progress at the end of the inspection.

c. Conclusions t

i The team found that there had been inadequtte root cause determinations and corrective actions for recurrent BA Pump problems, including two cases under the contemporary CR l system. The licensee's deportability evaluations did not evaluate the availability of boric acid flow paths required by the Technical Specifications (TS) for the past events. NNECO did not adequately consider operating experience at other reactor facilities in their

[ evaluation of the problem. The team also determined that NNECO did not recognize the potential USQ resulting from non-conservative BA tank level changes to the TS. The failure

, to correct the air binding of the boric acid transfer pumps and take action to preclude

! recurrence is a Violation of 10 CFR Part 50, Appendix B, Criterion XVI.

2.3.2 480 Vac Molded Case Circuit Breaker Magnetic Trip Setpoints

a. Inspection Scope The team assessed the adequacy of the licensee's actions relating to CRs concerning the i

overload magnetic trip setting for 430-volt molded case circuit breakers, which has been a recurrent issue.

b. Observations and Findings The team reviewed NNECO Engineering Self-Assessment 3DE-SA-97-03," Control of Magnetic Trip Settings on 480 VAC Motor Starters" which resulted in CR M3-97-3095, j documenting that incorrect magnetic trip settings and thermal overloads were found in I safety-related applications. The self-assessment was comprehensive and rigorous and represented a quality effort to evaluate the problem.

The self-assessment found three of 84 setpoints to be incorrect. Proposed corrective .

actions were generally comprehensive but required verification of motor name plates to determine actual required trip settings. Name plate inspections were deemed impractical for a "large" (but undefined) number of breakers due to their inaccessibility. As an 1 alternate to the inspections, NNECO determined that the erroneous setpoints occurred j when motors were replaced, resulting in new operating currents, but the trip settings were not appropriately adjusted. Consequently, NNECO reviewed motor maintenance files and )

confirmed that none of the unverified motors had been changed. This caused them to j conclude that the setpoints are probably correct and further confirmation is not needed prior to restart.

i

20 Long-term planned corrective actions included: revision of the 480 Vac MCB Preventive l

Maintenance procedure to include verifications by Unit 3 Refueling Outage 6 (scheduled approximately 10 months after Unit 3 restart) and eventualimplementation of the PM procedure as revised to perform the inspections and calculations necessary to confirm the l conclusions of acceptability. The CR documentation provided initially did not identify the l

equipment affected by the incorrect settings, did not include the values for the incorrect as-found magnetic and thermal overload settings, the amount of deviation from as-required settings or the evaluation of the impact of the deviations on the operability of the affected motors. Further, the documentation did not identify the motors which had not been inspected and verified or the licensee's basis for the acceptability of delaying inspection and setpoint verification.

This information was subsequently provided by the licensee between February 25 and l March 18,1998, following the end of the inspec+ ion. Only the original three mis-set l breakers were found. The reported deviations between the as-required and as-found l settings were, in each case, minimal and in the low (conservative) direction to protect the

equipment from overcurrent. No other discrepancies were identified in the 121

!- motor / breaker sets inspected to date. The remaining unverified 33 breaker loads / settings

! included four emergency diesel (EDG) fuel transfer pump motors; various control building,

! EDG building and intake structure heating, ventilation and air-conditioning (HVAC) motors; I and other similar loads.

l The licensee concluded that since the incorrect settings appeared to be the result of l improperly documented motor replacements and no other motor replacements had taken l place for the 33 unverified breakers, the probability of incorrect settings is negligible.

! NNECO further advised that it had examined its records for possible nuisance equipment i

trips for the affected loads that might have resulted from possibly low trip settings and had found no indication of problems. Most of the equipment was either in continuous or frequent intermittent operation and was functioning properly Upon further review, NNECO advised the team that the plans stillinclude motor name plate j verification during the next scheduled PM. However, in those cases where the next PM is l more than 18 months from restart, the inspections will be scheduled to occur during a more timely system outage,

c. Conclusion

The team concluded that the actions proposed by NNECO for this issue were reasonable.

2.3.3. Actions on Condition Reports - Failure to Conform to NRC Commitments

a. Inspection Scope The team assessed the adequacy of the licensee's actions relating to CRs concerning the previously identified failure to conform to NRC commitments,

b. Observations and Findings

21

.The team also noted four CRs that involved failure to conform to NRC commitments made in Generic Letters (GLs) and that were apparently misclassified as a Level 3 versus a Level 2 CR and/or contained ARs that were deferred post-startup. RP 4, Rev. 5, Attachment 3, CR Initiation and Classification Guidelines, is included in the Level 2 guidelinee as: an extarnal station commitment not adhered to; or a deficiency in material that, if left uncorrected, could affect safe reliable plant operation. The specifim were as follows:

• CR M1-97-1914is a Level 1 CR (related to GL 90-03 cad applicable to all three Units).

Corrective action #1 (AR 97020123-02)for this CH was inappropriately classified as deferrable to post-startup. However, the portions of this AR that related to compliance with the commitment were previously completed in December 1997.

• CR M3-97-4672 is related to GL 89-13 and was inappropriately classified as Level 3.

Further, none of the actions were coded as needing to be completed prior to startup. At the time of the inspection, allitems but one were noted as complete. The one remaining item appears that it should be completed prior to startup.

l

• CR M3-97-4346 contains a deficiency in material (inadequate corrosion control) that, if l left uncorrected, could affect safe reliable plant operation. It is related to GL 89-13 and l was inappropriately classified as Level 3. Some of the corrective actions appear that  ;

they should be completed before startup. CR M3-97-3501is also related to GL 89-13 I and documents an adverse condition. It is classified as a Level 2. However, the l corrective actions were not tied to startup. Some of the corrective actions appear that j they should be completed before startup. 1 CRs M3-97-4672 and M3-97-4346 were inappropriately classified as Significance Level 3, contrary to the requirements of procedure RP 4, Revision 5, Section 1.4 and Attachment.

This is a Violation of 10 CFR Part 50, Appendix 8, Criterion V (VIO 50-423/97 82-04).

The licensee issued CR M3-98-0933 to address these findings and performed a preliminary  :

review to determine the cause. They noted several breakdowns in their program related to l corrective actions for NRC commitments and began actions to both correct current  !

instances and prevent further such occurrences. On March 9,1998, the licensee rects sdfied CR M3-97-4346 and CR M3-97-4672 as Level 2 CRs.

I

c. Conclusion

The team concluded that NNECO had failcd to appropriately classify the CRs which were written for f ailed NRC commitments, which is a Violation of the procedural requirements of RP 4.

2.3.4 Resolution of Audit Findings - Organizational Independence of Radiation Protection Manager

a. Inspection Scope The team assessed the ad3quacy of the licensee's actions relating to audit findings concerning Section 6 of the Technical Specifications.

l

l 22

b. Observations and Findings Audit Report (AR) MP-97-A06-02 noted one daficiency regarding the organizational reporting by the Unit 3 Radiation Protection Manager (RPM) to the Maintenance Manager.

Unit 3 Final Safety Analysis Report (FSAR) Section 12.5.3 states that all health physics procedures and methods for ensuring that occupational radiation exposure is as low as reasont3ty achievable (ALARA) follow the provisions and suggestions of Regulatory Guide (RG) 8.'1, Revision 3; RG 8.10, Revision 1-R; and RG 1.33, Revision 2, as applicable. RG 8.8, Se ction C.1.b(3), states, in part, as follows:  !

"The Radiation Protection Manager (RPM) (onsite) has a safety function and responsibility to both employees and management that can best be fulfilled if the  !

indivi lual is independent of station divisions, such as operations, maintenance, or techrical support, whose prime responsibility is continuity or improvement of station operability."

Millstone Unit 3 Technical Specification Section 6.2.1.d states, in part, as follows:  ;

"The individuals who train the operating staff and those that carry out health physics and quality assurance functions may report to the appropriate onsite manager; however, they shall have sufficient organizational freedom to ensure their independence from operating pressures."

The licensee closed CR M3-97-1875, noting that the Unit 3 RPM has a direct reporting capability to the Unit Director on radiologicalissues as noted, by asterisk, on the organization chart. Having the RPM report to the Maintenance Manager is inconsistent with the Unit 3 technical specifications. There were inadequate corrective actions for this Condit'cn Report. The licensee stated during the onsite inspection that this issue would be evaluated along with other organizational changes that are currently being considered.

Therefore, the team considers this item to be a Violation of Technical Specification 6.2.1.d (VIO 50-423/97-82-05).

c. Conclusion

The licensee f ailed to take corrective actions for a technical specification-related issue concerning the organizationalindependence of the RPM, which is a violation.

3.0 SELF-ASSESSMENTS 3.1 Self-Assessment Program

! a. Inspection Scope l

l The team evaluated the Millstone site and departmental self-assessment programs to determine their focus on safety. They also evaluated the effectiveness of management to address the findings of self-assessments and performance improvement programs. The team verified that significant issues that could impact plant safety are being addressed.

?

l l

w_ _ _ _ _ _ __

7 23

b. Observations and Findings The stated purpose of the licensee's self-essessment program is to identify areas of concern and to improve performance utilizing the established Corrective Action Program.

Specifically, the programmatic controls described in unit procedure U3 OA 11, "Self-Assessment," Revision 1, direct the performance of pre-planned department self-assessments, b7 lualified individuals, with the objective of achieving higher standards of quality and performance at Millstone Unit 3. This procedure further states that the primary l responsibility for the performance of self-assessment activities resides within the line  ;

organizations, including identification and resolution of deficiencies.

In order to evaluate the effectiveness of the self-assessment program, the team reviewed a l selected sample of (20 out of approximately 110) recently completed U3 self-assessments.

i These self assessments included a cross-section of U3 department evaluations, as well as fol;ow-up reviews, which were performed during 1997 and early 1998. As a result of l

these reviews, the team determined that there was a significant variance in the quality and I

technical depth of these self-assessments depending on (1) the time frame in which the assessment was performed and (2) the organization performing the activity. In particular, the self-assessments which were performed earlier in 1997 tended to be narrowly focused with limited corrective actions specified. Examples of this category of self-assessments included the following reports:

• 3MNTC-SA-97-17,QC Hold point Requirements, February 5,1997 1

• 3MNTC-SA-97-01,In-Field Procedure Compliance, May 7,1997 l l

• 3 CAD-SA-97-03, Root Cause Evaluation Quality, April 30,1997

However, as noted by the team, the technical adequacy of self-assessments tended to improve in the latter half of 1997 and into the first quarter of 1990. Self-assessment reports which reflected this improving trend included the following examples:

• 3MNTC-SA-97-07, Health Poysics Department, June 24,1997

• 3TS-SA-97-13, Functional Requirements in Safety Systems Preoperational Testing, September 30,1997

• 3 CAD-SA-97-14, Assessment of MP3 and Site Level 1 Condition Reports, January l

22,1998 l

As determined by the team, the overe' improvement in the quality of U3 self-assessments

! was attributable to the establishment of definitive management expectations regarding the need for performance improvement, an emphasis on self-assessment training and enhanced procedural controls. Additionally, at the time of the inspection the licensee was in the process of implementing a self-assessment program that incorporated both station and unit specific elements. Relative to this issue the team ascertained that the station unit and support organizations have deve!oped departmental self-assessment plans with each major support group performing formal assessments using a common approach. Accordingly, L--___--__-_ . . _ _ _ .

24 department self-assessments consist of an in-depth evaluation of significant line and staff activities, which are performed by teams of knowledgeable individuals in accordance with defined assessment plans. The assessment results are used to identify program strengths, findings and areas for improvement. The licensee's program also has been expanded to evaluate the effectiveness of the self-assessments in order to improve the quality and consistency of these activities. The Nuclear Oversight Restart Verification Plan (NORVP),

which is discussed in R, port Section 4.1.b.6, evaluated the performance of self-assessments from October 1997 through this inspection period.

c. Conclusions The self-assessment program had established appropriate administrative controls which provided for the trecking of information to detect declining performance and adverse trends. The team concluded that, in general, the self-assessment prog.am was being adequately implemented and that the associated recommendations were beneficial in identifying areas for enhancement and improved performance.

3.2 Operator Work-Arounds

a. Inspection Scope Evaluate the effectiveness of NNECO's program to identify and correct operator " work-arounds."

b. Observations and Findings The team evaluated the effectiveness of NNECO's program to identify and correct operator

" work-arounds." The purpose of the work-around program is to identify and assess conditions that may adversely affect plant operations. These deficiencies are characterized, for example, as items that may degrade the operator's ability to react to l plant transients or may sirnply interfere with the operator efficiently performing normal tasks. The inspection effort involved the review of the procedural controls contained in OP 3260E, " Program for Resolution of Operator Work-Arounds," Revision 0, examination of Self-Assessment 30PS-SA-97-04," Effectiveness of The Operator Work-Around Program" and the evaluation of recently completed modification packages related to operator work-arounds.

The inspection team determined that discrepancies associated with operator work-arounds are not documented on CRs. Further, they are not required to be tracked on the licensee's Action item Tracking and Trending System (AITTS) system but are administered under a separate program. In particular, procedure OP 3260E directs that operator work-arounds initially identified in the shift turnover log, be screened for their cumulative impact, including operability and deportability considerations. This screening process is performed by the unit supervisor, the shift manager and finally by the operations manager. Based on this screening process, the items that are classified as operator burdens or work-erounds are separated from the category of non-conforming conditions which would require.

corrective actions in accordance with the CR program. Additionally, these items, which are tracked using the Operations Performance Data Base, are prioritized and periodica!ly i

l l

25 l

! reviewed by the Operations Department and they are reviewed on a biweekly basis by unit l management personne!.

Subsequent to reviewing the completed modification work packages for six operator work-arounds, the team performed a system walk-down in order to confirm the implementation of equipment modifications and the adequacy of the completed work activities. Based on the team's reviews and the results of system walk-downs, it was determined that the equipment modifications associated with five of the operator werk-arounds had been appropriately completed or were awaiting testing following component rework. However, during the review of the documentation related to operator work-around Number 96-03, (correction of flow indication anomalies on service water instrumentation 3SWP-Fl-059 A, B and C), the team determined that the Trouble Report (TR) tags had been removed from the flow indicators and that the automated work order (AWO) associated with this modification had been inappropriately closed, prior to the completion of all specified work.

Specifically, the final setpoint calibrations for flow indicators 3SWP-F1-059 A, B and C had not been accomplished prior to closing the AWO, which is contrary to the requirements of 10 CFR Part 50, Appendix B, Criterion XVI, Corrective Actions and also to the work control requirements of Procedure U3 WC 1, " Unit 3 Work Management," Revision 1, Section 1.8.7. This is a Violation (VIO 50-423/97-82-06).

The team also observed that the action taken by the AWO to correct an operator work-around item had not been identified as a CR. In this case, service water flow instrument operation was affected by system process valve position. Thus, the issue did not benefit from the controls that would have been imposed by the corrective actions process.

Following the identification of this issue, the licensee initiated CR M3-98-0942,in accordarce with procedure RP 4, to document and resolve this deficiency.

Two contributing factors related to the above noted violation involved the (1) the lack of formal training for operations personnel on precedure OP 3260E and (2) the fact that Action Requests (ARs) are not initiated to track the status of each operator work-around.

These specific issues had been previously identified in Self-Assessment Report 3 OPS-SA-97-04 and although a CR (M3-97-3632)had been initiated on October 21.1997, to address the assignment of ARs to track discrete operator work-around items, the corrective actions for this item were not scheduled for completion until September 30,1998. Given that both of these contributing factors had been previously identified in a self-assessment report, this missed opportunity to avert an item of noncompliance is identified as a weakness within the self-assessment program.

c. Conclusions One Violation that was identified related to the implementation of NNECO's program to i document and correct operator work-arounds, an AWO associated with a modification was

}

closed prior to completing all specified work. The team also observed that operator work-around issues, which involve plant material deficiencies, were not included in the corrective actions program.

3.3 Self-Assessment of Design Basis issues 1 \

i j l

l 26

a. Inspection Scope EvalJate NNECO's actions taken in response to ACR-7007- Event Response Team. This j issue primarily concerned the inaccuracies in the Updated Final Safety Analysia Report (FSAR) and design basis documents. It is designated by NRC as Significant Issue List (SIL) ltem 41.

b. Observations and Findings i

Background:

ACR 7007 was written by the licensee in January 1996 to address the high-level concern that "the UFSAR, system descriptions and design basis documents contain inaccuracies."

An Event Response Team was chartered to determine t* s causes of these inaccuracies.

This team used root cause ana' lysis methods to identify causes and contributing factors.

This team developed the ACR 7007-Event Response Team Report, dated February 22, 1996, that identified conclusions, corrective actions and comments in Sections 5,6, & 7 of the report. It also included a significant amount of backup material that led to these conclusions. The report presented a thorough analysis of the problems and a reasonable set of conclusions and corrective actions. The identified causes and contributing factors were broad and far-reaching, necessitating extensive corrective actions. These were further elaborated upon by the licensee as described below. .

In July 1996 the Fundamental Cause Assessment Team (FCAT) and the Nuclear Committee Assessment Team (NCAT) determined and reported to the Northeast Utilities (NU) Board of Trustees that the root cause of decline in Millstone performance was that senior executives at NU, from the CEO to senior ruclear site executives, were ineffective over a number of years in providing vision, direction and leadership necessary for the management of the NU nuclear power program.

Over the last two years, the licensee has embarked on a number of broad programs and changes to address these concerns and improve performance. These have included the configuration management program (CMP) and major changes in organization and management at the Millstone site.

In order to address the specifics of ACR 7007, the licensee extracted allidentified issues from the Event Response Team Report. These were compiled into a main item list of 104

> issues, dated Nov. 5,1996. In June 1997, Unit 3 isu M CR M3-97-1839 to address the

! Unit 3 aspects of ACR 7007 and the Event Response . am Report. This document sorts the 104 issues into five major areas in order to track and address them effectively:

l leadership, self-assessment, corrective actions, configuration control and oversight. These categories are the same that have been identified by the Corrective Actions Department.

Additionally, they are the same as five of the 111,y site-wide issues for restart that are being tracked by NNECO and that are addressed in the periodic letter to the NRC titled,

" Progress Toward Restart Readiness at Millstone Station" (NRC Briefing Book).

In order to provide more detailed tracking and specification of corrective actions, subcategories were established below the five categories. All of the 104 issues were then 1

27 assigned to the categories and subcategories. [However, three of the 104 issues were noted to have no action required and 15 were noted to be applicable to Unit 1 only). A succinct high-levelissue was developed for each of the subcategories and corrective actions were defined and approved to address each high levelissue. AR numbers were assigned for all corrective actions. An important point to note is that this ACR and the related corrective actions addressed primarily the configuration control aspects of the five major areas and did not try to fully address other aspects of leadership, self-assessment,  !

corrective actions and oversight. l As part of the CMP program, under PI 2, " Unit Specific Assessments," Unit 3 performed a self-assessment to determine any other areas similar to those in ACR 7007 that needed to be addressed. ACR 13302 was written to address the findings from that assessment. l ACR 13302 was included as part of the ACR 7007 package and was also reviewed for this ,

inspection. I Genera! Discussion:

The team reviewed the lists discussed above and verified that the issues had been appropriately extracted from the Event Response Team Report and that each of the 104 issues had been assigned to at least one of the categories / subcategories. The decision of

'no action required" on three issues was judged appropriate. For the 15 issues designated as applicable only to Unit 1, the documentation was not clear as to why they were not also applicable to Unit 3. The licensee acknowledged this, performed further review of these items and documented the review in Memo PES-98-055, Clarification for Corrective Action Plan for CR M3-97-1839, dated February 12,1998. This memo provided justification for l the 13 of the 15 Unit 1 items. The other two were determined to be generic to all units.

However,it was further determined that the Unit 3 corrective actions, already completed for ACR 7007, encompassed these two items as well.

The team also reviewed each subcategory and verified that the issues (assigned to the  !

subcategory's defined as high-level issue) and the specified corrective actions were appropriate. A few areas were noted, requiring further justification and questions were passed on to the licensee. With the exception of those noted below, all were satisfactorily  !

, recolved.

l The NRC also reviewed, on a sampling basis, the documentation provided with the CR for the completion of the defined corrective actions. Some of the reviewed areas were then selected for further in-depth follow-up. Additionaiiy, the team performed independent l review in the general area of the high-level concerns to verify the thoroughness of the corrective actions. Some areas addressed in more detail included: the FSAR update process, setpoint control, an independent contractor audit of the Design Control Manual l and the design basis summaries.

The team also verified that ell actions for each of the three CRs were appropriately closed or that remaining open items are not significant and scheduled for closure on an acceptable time frame. Some areas designated for completion following startup were questioned.

The licensee provided documentation that the work had been completed already or justification that its deferral was appropriate, except as noted below. Additionally, the team

r l

28 revie.wed an internal audit of the corrective actions associated with ACR 7007,"The ,

independent Review Team Report on the Effectiveness of Correctiveness Actions j Associated with ACR 7007, Rev.1,6/17/97,"and noted that the findings were addressed I

by extensive actions taken by the licensee on ACR 7007 since the audit.

l Specific Areas and Issues:

(1) Final Safety Analysis Report Changes The team discussed the area of Final Safety Analysis Report (FSAR) change identification and control with licensee representatives and reviewed procedure RAC 03, Rev. O, Changes and Revisions to Final Safety Analysis Reports. The team also reviewed: selected FSAR Change Requests, the computer tracking system for FSAR changes, FSAR changes in process, the process for initiation of a FSAP change, the review and approval process, the facilitatory / owner of the FSAR and specific FSAR sections, the trackin0 of in-process changes and the overall scheme for FSAR upuates during the current outage. The licensee has processed a large number of changes (e.gy 596 in 1997) during the current outage due to CMP issue identification (called an updating FSAR change) and due to the large

! number of modiNations being completed (called an up-front FSAR change). Approvals for the up-front change,s are processed together with the modification, ensuring that l appropriate reviews are given and that the FSAR is kept current. This large number of changes has resulted in several update submittals to the NRC during the past year and has measurably improved the content of the FSAR.

RAC 03, Attachment 7, contains a listing of the manager of the primary responsible discipline for every section of the FSAR, thus allowing proper review of change requests.

The procedure also has mechanisms for ensuring that in-process changes are consistent with each other. Time limits are set and, if processing time frames of the procedure are not met, a CR is issued. The team reviewed the licensee's tracking and trending data. The area of FSAR change control was judged acceptable.

(2) Design Control Manual Review

! One of the issues associated with ACR 7007 related to design control and the Design Control Manual (DCM). As a result of this, one of several actions was to contract MDM Services Corp to perform an independent review of the DCM. MDM issued the report, l " Strategic Overview of Millstone Configuration Control Processes by MDM Services Corporation, Final Report," on July 15,1997. Contained in this report were a number of recommendations, which the licensee addressed in Memo PES-97-412, dated Dec. 31, 1997. However, a number of the recommendations were rejected with no justification.

The licensen revisited the area and provided a new evaluation that adequately addressed all recommendations or provided justification for not addressing them.

t (3) NRC Commitments The area of " licensee commitments to NRC" was addressed in ACR 7007. The team also noted that Level 1 CR M3-971759," Trend identified in the area of NRC commitments,"

was issued in early 1997 to address then-current problems in this area. The licensee L_-----__--

29 placed considerable effort into the identification of past regulatory commitments made to the NRC, in docketed correspondence, through the implementation of ptocedure PI 6, Licensing Reviews. Ongoing controls for commitments were established in several procedures, including DC 18, "NRC Communication," U3RP10 Outgoing Regulatory Correspondence Processing and Validation," and RAC 06, " Regulatory Commitment Management Program." The team reviewed the list of CRs generated as a result of the PI 6 leviews and sampled the actions taken to resolve them. No problems were identified.

(4) Updating of Drawings The team discussed the area of drawing control and updating with the MP3 design engineering personnel. The updating process is controlled by the DCM, Chapter 7 and by EDI 30250. These documents define categories of importance for incorporation of changes into the more important drawings more quickly. Category 1 or Operations Critical drawings require incorporation within 30 days of the DCN being released. Category 2A drawings have a 90-day guideline for incorporating outstanding changes. However, there is ' urrently a backlog of about 3,000 drawings. The licensee had no concrete plan for eliminating this backlog. In response to team questions, the licensee issued Memo M3-DE-98-0090 that forecast a plan to work off the backlog over a two-year period. This area will be reviewed during the NRC Operational Safety Team inspection (OSTI).

(5) Design Change Controlissues issue 4 and the similar issues 45 and 59, were not specifically addressed in the ACR 7007 close-out package. These issues related to a potential weakness in programs that may l have allowed drawing changes without generating a plant design change request (PDCR) or a DCR. The licensee reviewed this area and prepared a documentation package to demonstrate that other controls existed to prevent such an occurrence. Further, the CMP utilized PI-29, Unit 3 Piping and instrument Drawing (P&lD) Walkdowns, to review P&lDs for discrepancies and then evaluato and correct them.

One closely related concern, documented in ACR 7007, was that the process for controlling dawings had been weak, allowing drawings to be changed without a DCN or DCR and not ensuring that other design documents were updated. As a follow-up to this concern, an Engineering Self-Assessment Report (3-ESAR-97-001)was performed to assess Category 8 Administrative DCNs for this general problem. The ESAR found that certain DCNs exceeded allowable criteria in that an MMOD or DCR may have been required to properly approve and document the change in question on the administrative DCN. ACR M3-97-0506 was written to take corrective actions on this specific finding. The team reviewed the corrective actions plan (CAP) and the documented closeout actions to address the CAP. The team noted that not all of the required actions were clearly documented as being complete. The team requested additional documentation from the l licensee to show that the actions had been completed. The licensee's review then found that all of the actions for one assignment, AR 97003960-05,had in fact not been completed, even though the AR had been closed. This is contrary to RP 4, Step 1.12.4, which states that actions are to be closed out "WHEN assignment is complete." The two actions that did not appear to be completed were: performing a maintenance support engineering evaluation (MSEE) reconciliation of 199 DCNs that document as-built L

30 conditions and reviewing the 142 DCNs that initiated work and then generating an MSEE, MMOD or DCR as appropriate.

As a result, the licensee issued CR M3 98-0921,which documents the incorrect closure.

AR 97003960-05 specifies four actions. This is a violation of the requirements of 10 CFR Part 50, Appendix B, Criterion XVI and also of the requirements of RP 4 (VIO 50-423/97-82-07).

(6) Setpoint control The team evaluated the design control area of setpoint control. The licensee does not maintain all of the setpoints in one common data base or area. However, Specification SP-ST-EE-329, Use and Control of Master Setpoint index, Rev. 2,10/15/97, provides a roadmap of the various documents, control mechanisms and data bases involved. Change control for the setpoints is typically via the DCM. NGP 5.23, Plant Design Data System (PDDS) Data Packages states that the Master Setpoint List provides the setpoint calculation number and process and instrumentation setpoints for components designated by an instrument mark number. In order to evaluate this area, the team selected 10 annunciator windows and their respective setpoints, as noted in the corresponding alarrr response procedure (ARP). The team requested the licensee to provide the design information for each of the 10 instruments involved, including the setpoint value and the basis for the setpoint (e.g., the calculation).

Based on the licensee's response, the team noted that the information for the 10 selected setpoints was not contained in one consistent location but rather was in the Master Setpoint List, the Technical Specifications, the Westinghouse Precaution, Limitations and Setpoints (PLS) document (WCAP-10072), a Surveillance Frocedure and a calculation.

Further, one of the ARPs was found to contain an incorrect instrument as the initiating device for the alarm. The licensee wrote CR M3-98-0805,"Pzr Level alarm initiating device incorrect in ARP," to address this finding. Also, for one of the alarms, the saturation trouble alarm for the Inadequate Core Cooling System, the Master Setpoint List referenced an incorrect calculation. The licensee was unable to find a calculation that provided the basis for the setpoint of 15 degrees. Upon further exploration,~ the licensee determined that the setpoint should have been changed to 32-degrees (which would agree with a similar alarm in the Safety Parameter Display System - SPDS), as part of PDCR M3- i 93-121 (and as recommended by Memo NE-93-SAB-263 dated 6/14/93). This 32 degree setpoint is derived in the Millstone 3 Emergency Operating Procedure (EOP) Setpoint l Documentation, Calculation # W3-517-981-RE, Rev. 6, dated 9/17/97. The licensee l issued CR M3-98-0935 to address this finding. The saturation trouble alarm in question is l described in the FSAR, Section 4.4.6.5, instrumentation for Detection of inadequate Core l

Cooli'ng, as backups to the primary subcooled/superheat display on SPDS.

The team noted that the methods for control and documentation of setpoint information appeared inconsistent, difficult to retrieve at times and had the potential for allowing incorrect information to persist. In addition, the licensee performed an Engineering Self-Assessment of the Setpoint Control Topical Area,3-ESAR-97-015, dated May 2,1997 and this self assessment also noted weaknesses in the Setpoint Control Program that have not l t

31 been addressed to date. The team has identified an issue to review the quality of setpoint controlin the Tuture (IFl 50-423/97-82-08).

(7' Design Basis Summaries As part of the changes to the design control process made in addressing ACR 7007, the licensee eliminated their Design Basis Documentation Packages (DBDPs) and System Descriptions and established Design Basis Summaries (DBSs). These were created per station procedure U3 PI 29, Development of Millstone Unit 3 Design Bases Summary Documents. U3 P! 29 states that the objectives of the DBSs include providing a documented reference: for use in the design process for future modifications, to support technical reviews and safety evaluations, to support operability evaluations and determinations for continued operations and to support review of technical specification changes and FSAR changes. It further states that the DBSs are integral to the Unit 3 restart program and are prepared for the Unit 3 Maintenance Rule (MR) Group 1 and 2 systems. The team reviewed the set of Dl3Ss against the list of MR systems and reviewed portions of selected DBSs. The team also reviewed methods for updating the DBSs, which are contained in the Design Control Manual (DCM), Chapter 11, Attachment 5, Instructions for Controlling and Revising Design Basis Summaries, Rev. 6, Change 2.

The team noted that there was no DBS for the emergency lighting system that was moved from MR Group 3 to Group 2 during the summer of 1997 after the original DBS list was developed. He also noted that the licensee was unable to ensure that any changes to the list of MR systems was reflected in the DBSs. The licensee initiated CR M3-98-0892 to address this issue. Further inspection disclosed a lack of full coverage for the Chemical &

Volume Control System (CVCS)in a DBS since this was a MR Group 1/ Group 2 system.

Portions of the CVCS were included in the Emergency Core Cooling System (ECCS) DBS, but much of the CVCS system was not included in any DBS. The failure to include the emergency lighting system and the full CVCS in the design basis summaries is a Violation of 10 CFR Part 50, Appendix B, Criterion V of the requirements of PI 29 (VIO 50-423/97-82-09).

The team noted that the original DBSs were prepared based on design information that had a freeze date of June 1996. The DCM update controls began sometime after that, resulting in a gap of coverage of about one year. This was identified in Memo MP3-DE 1616 and is being tracked by Design Engineering, CRs and Oversight. While not yet updated, the gaps are clearly identified and are being satisfactorily tracked.

The team examined the overall control of DBSs, since the PI procedures are being phased out and the DCM only addresses a revision process. The licensee presented the Millstone l Nuclear Power Station, Programs and Engineering Standards, Configuration Management l

Plan, Revision 1, September 23,1997, which addressed the transition process from the CMP and the PI procedures to a permanent organization. This document is to ensure no gaps in processes and that the going-forward products are clearly identified.

(8) Safety Functional Requirements Manual J

u__-_-_______________ . _ _ _

l l

1 l 32 The MP-3 Safety Functional Requirements (SFR) manual was developed to identify the key l system level requirements that are reflected in the safety analysis. This provides design l input and assumption information, primarily for the nuclear steam system supply (NSSS) equipment. The actual plant NSSS calculations are proprietary and are maintained by Westinghouse. The SFR was developed by NNECO and was reviewed and commented on ,

by Westinghouse. Chapter 2 of the manual addresses systems, Chapter 3 addresses the l FSAR Chapter 15 safety analyses, and Chapter 4 addresses three programs (fire protection, station blackout (SBO) and safety grade cold shutdown). The manual provides a valuable tool for the plant design personnel.

The SFR manual, (MP Unit 3, Design Basis Documentation Package, Safety Functional Requirements, DBDP-MP3-SFR)is controlled by procedure NGP b.28, Design Basis Documentation Packages, Rev. 3, October 15,1997. Step 1.1.2 of NGP 5.28 states that if changes are required they must be documented as DCNs and the DCN numbers entered into GRITS (Generation Records Information and Tracking System). GRITS is the on-line interactive data base system that provides the current design and revision status for the Millstone facility. The original Revision 0 of the SFR was issued December 30,1994.

Revision 1 addressed Westinghouse's review comments and was issued December 11, 1996. Rev. 2 was issued on November 20,1997 to incorporate CMP changes. However, Revisions 1 and 2 were issued with Engineering Record Correspondence per NGP 5.31 rather than NGP 5.28 and as a result no DCNs were issued and GRITS was not updated.

As of February 1998, GRITS still showed Rev. O as the latest version. This constitutes a failure to follow procedure NGP 5.28 and is a violation (VIO 50-423/97-82-10). The licensee issued CR M3-98-0861 to address this problem.

c. Conclusions Acceptable processes for FSAR change control are being applied. An AR was closed without accomplishing the specified corrective actions. This is a violation. The Master Setpoint List was found to contain incorrect information. Additionally, the methods for control and documentation of setpoint information appeared inconsistent, difficult to retrieve at times and had the potential for allowing incorrect information to persist and the findings of an Engineering Self-Assessment of this area has not been addressed to date.

This is a Follow-Up item. Violations of design control procedures were identified concerning Design Basis Summary documents and concerning the Safety Functional Requirements Manual. Updates of 3,000 Category 2A drawings are planed over a two-l year period. This category has a 90-day guideline for incorporating outstanding changes.

SIL ltem No. 41, concerning the ACR-7007 7ent Response Team findings, is clo ed.

l 3.4 Engineering Assessment of Control Room Design Review I

a. Inspection Scope  !

The team evaluated the effectiveness of the licensee's management team to resolve l recommendations made during a control room design review by Engineering Review 3-  !

ESAR-97-008 dated January 31,1997.

b. Observations and Findings <

i i

33 The assessment recommended revising procedures NGP [ Nuclear Group Plan] 5.25 and SP-EE-261. The licensee documented that the revision to SP-EE-261 will be completed by March 30,1998, through AR 96000103 and willinclude a section of standards for computer displays and an update pertaining to Revision 1 of NUREG-0700. The licensee also documented that the revision to NGP 5.25 will be completed of AR by June 30,1998, which includes a change to NGP 5.25 that would require the Human Factors Specialist rather than the Project Engineer, to determine whether a detailed control panel design review is warranted and an update pertaining to Revision 1 of NUREG-0700. Management rejected a recommendation to add another human f actors specialist in view of the workload.

The engineering review also raised the question of whether there was a match between the simulator and the control room designs. The licensee's documentation noted the following strengths of the simulator update process:

"At this time, there are no design changes with simulator impact that have been installed in the plant for [more] than 30 days that have not been incorporated in the simulator."

"To support the current restart training needs, we have modified our target for incorporating those plant design changes identified by Operations and Operator Training as having simulator impact to have them installed within 30 days of plant installation.

This is beyond what is required by ANS-3.5, the standard to which the Millstone 3 simulator is certified to. The standard allows 24 months to incorporate such changes."

Overall, the team found that the process, which ensures that the fidelity of the simulator is maintained with regard to the reference plant, was a strength. The team also found that there have been several advanced systems added to the control room without appropriate consistent design guidelines (e.g., Foxboro intelligent automated (IA) system for the moisture separator reheaters (MSRs), fire protection, environmental qualification (EO) temperature monitoring, the auto-log system, and SPDS upgrades). There are currently eight different computer-based systems in the main control room. Since many of these systems are significantly different in their human systems interf ace, displays and alarms.

The team observed that the licensee had not evaluated these different systems with respect to unnecessary operator burden and the potential for increasing operator error,

c. Conclusions The self-assessment was thorough, the simulator fidelity process was in place. However, i several advanced systems added to the control roorn without appropriate consistent design l

guidelines, j 4.0 INDEPENDENT OVERSIGHT 4.1 Effectiveness of Nuclear Oversight - Audits and Evaluations

a. Inspection Scope I

l L_ - - - - _ - - - - - - _ - - _ _ - - - - - - _ _ - - _ . - - - _ _ - _

34 l l

The team reviewed the activities of the Audits and Evaluation Group of the Nuclear l Oversight Organization. This review involved discussions with auditors, managers, the l director of the audit group and the Vice President, Nuclear Oversight. Documents  ;

reviewed included, but were not limited to, the Nuclear Oversight Group procedures, a '

sampling of recently performed audits, tracking and review of audit findings, adequacy of i audit findings, scheduling of audits, training and qualifications of auditors, adequacy of l staffing, adequacy of audit corrective actions and Nuclear Oversight Group self-assessment. In addition to the Audit and Evaluations Group, the following activities were also reviewed:

I l

• Nuclear Oversight Group actions to improve on deficiencies identified by independent j outside assessments of quality assurance (QA) performed during 1996. Specifically -

addressed were a sampling of actions taken by the licensee to respond to criticisms i identified in the Joint Utilities Management Assessment (JUMA) performed in July l 1996; l

i

• The Recovery Plan for Nuclear Oversight, Revision 2 dated May 27,1997; i

• Surveillance and quality controlinspections performed by the Performance Evaluation Group concerning repairs made to a Unit 3 emergency diesel generator; and,

• Various activities performed by Oversight implementing the Nuclear Oversight Restart Verification Plan (NORVP).

t

b. Observations and Findings s l

The team reviewed the actions identified within the Nuclear Oversight Recovery Plan taken  !

to complete the plan objectives. Some of these, as the development of an effective ECP and enhance the effectiveness of the Nuclear Safety Assessment Board (NSAB), are addressed within other report sections.

l The review of Nuclear Oversight involved personnel, programs, procedures, audits, audit

! follow-up, audit scheduling, certain aspects of Performance Evaluation and NORVP. The observations and findings for each area are discussed separately below:  ;

i i (1) Personnel l l

Since 1996, the auditor staff has been increased from five to 20 auditors. The audit staff has a significant background in dealing with various technical areas, including operations, in addition, there is an auditor qualification and training program in place. There are now l four audit managers over the areas of operations, maintenance, engineering and technical support and plant support; and an overall director of audits and evaluations. This increase in audit . management attention and audit personnel has improved the quality of audit findings. There is now a stronger interface with the line organization. Licensee audit  ;

teams typically have four or five members and take up to two weeks to perform. For this i reason, support for the audit team is periodically obtainert from contractors, the line organization or other groups within the Nuclear Oversight organization.

l

35 (2) Procedures The tearn reviewed the following procedures:

NOOP 1.05,'Self Assessment Process, Revision 0, June 30,1997

- NOQP 1.06, Nuclear Oversight Resolutions issues, Revision 0, November 12,1997 l -

NOOP 2.01, Nuclear Oversight Audits, Revision 2, November 20,1997 NOOP 2.02, Qualification / Certification of Audit team Leaders and Orientation of Team Leaders, Revision 2, November 20,1997 NOQP 3.02, Analysis of Quality Program Performance, Revision 0, August 29,1997 NOOP 3.03, Nuclear Oversight Assessments, Revision 0, May 12,1997

- NGP 3.19, Procedure to Stop work, Revision 1, July 30,1997

- RP 4, Corrective Action Program, Revision 5, September 5,1997 The team observed that all the procedures listed above were established or significantly revised during 1997. However, not all were new procedures as they may have existed in another form in the previously established QA program. The procedures appeared to be comprehensive, clearly written and user friendly. There were procedures for all aspects of the program. It was determined from the review of the procedures that a new audit program had been established with more strength than the previous OA audit system.

(3) Audits l

The team reviewed the following audits and audit checklists:

1 MP-97-A11-03, Software QA, performed November 3 - 14,1997

- MP-97-A09-01, Fire Protection, Performed September 8 - 26,1997 MP-97 A05-02, Chemistry, Performed May 27 - June 6,1997 l

l - MP-97-A10-07," Operating Experience" Program, November 10-17,1997 Audit checklist for Audit MP-97-A09-02, Security l

l

- Audit checklist for Aud:t MP-97-A06-03, Systematic Approach to Training Procedure NOOP 2.01 has strengthened the audit process by more clearly defining audit expectations, audit checklists and the makeup of audit teams. A review of the above audits and audit checklist indicates significant improvement in the audit process. Audits l

l

36 are generally two weeks in length and performed by a team. This has led to a more in-depth look into each area.

Significant audit findings are issued as Level 1 condition reports (CRs), because they receive a high level of attention as required by procedure RP 4. An initial 7-day audit report is issued with the CR findings. The audited organization must respond within 30 i days and the audit group must agree with proposed corrective action as stated on the .

l initial CR response. After agreement on the CR response to the findings, a more complete 45-day audit report is issued. This process ensures significant QA involvement in the  !

Corrective Action Process. Less significant audit observations are issued as Level 2 CRs.

1 In the past, audit exits were apparently given a low level of attention and were rarely attended by the line staff and rarely attended by appropriate management. The team l reviewed a sampling of audit exit attendance sheets and observed that the exits were well attended by both line staff and managers.

(4) Control of Status of Audit Findings As stated above, audit findings are controlled by the CR process. It is the responsibility of the line organization to provide corrective actions. The corrective action group for each l

unit maintains a status of open CRs. However, they do not necessarily track CRs as audit findings. All due or overdue Level 1 CRs are treated equally. The team had a concern that audit findings may be lost in the large number of open Level 1 CRs.

The audit group has established its own computer tracking system of open audit findings.

l This information is provided to audit managers and certain line personnel. It is used as a l tool to track overdue or inadequate audit corrective actions. The audit open item is tracked i by both the CR number and associated assignment request numbers (ARs). Recently, the l report has been improved to clearly show a summary of the open or incomplete issue. l This audit finding tracking system appears to provide an effective mechanism to ensure l l

that audit findings are not lost in the " shuffle" and to ensure that audit managers have a tool for managing the follow-up of audit findings.

All audit findings receive a follow-up for adequacy of corrective action, although this may not be a 100% verification of each action. Audit observations are followed up on a sampling basis. Audit follow-ups may be done independently or during the next audit of the subject area, inadequate corrective actions are identified by the issuance of a new CR.

The team noted that this was done concerning corrective actions identified in a fire protection audit. The details of these audits are further discussed in NRC Inspection 97-84.

(5) Audit Scheduling and Planning (Open VIO 50-423/96-05-12) l l

The team observed that an audit schedule had been established for 1998 and a projected schedule for 1999. These schedules only covered general areas to be audited and not detailed audit objectives. Actual audit planning is done by the team leader in conjunction with the manager for that area. Audit planning is established in NOQP 2.01. The plan is

D 37 to include verification of corrective actions for previous findings within the audit scope.

The team verified that such audit planning is accomplished for each audit.

On May 29,1996, the licensee wrote an ACR that audits of TS 6.8.4.e, " Accident Motoring instrumentation," may have been missed. Based on this ACR, a review of audits of the technical specifications and a concern that TS audits were neither comprehensive nor well documented was previously detailed in Inspection Report 50-423/94-28, violation 50-423/96-05-12was issued stating that, "The failure to audit technical specification section 6.8.4.e within a five year period is a violation of TS 6.5.3.7..." in its response dated September 16,1996, the licensee stated in part that, "The process for the independent review of the TS Audit Matrix will be revised to require a review of "what has been done" versus "what has been scheduled." In the future, personnel will enter information into a tracking data base after the audit has been completed and after the independent reviewer has checked what has been done. Procedure ... will be revised by October 31,1996."

The team verified NOOP 2.01 has been revised to require an audit commitment data base.

A data base has been established for general audit commitments and a separate one for all technical specifications. Further discussions indicated that while a data base had been established to determine what technical specifications had been audited and what needed to be audited, it was difficult to use as a " tracking and scheduling" tool. The licensee had already discovered the same thing during a recent self-assessment of the audit program.

Memorandum AE-98-4037, dated January 19,1998, issued report no. 97-AE-08, "Self-Assessment of the Audits and Evaluation Group's Audit Commitment Tracking Process."

The self-assessment was performed December 1-5,1997. The executive summary of this assessment stated, in part, the following:

"The audit commitment data base, as it is presently configured, does not meet the needs of the Audits & Evaluation (A&E) Group. This data base is not a tracking data base but is strictly a data base for recording information. As a result, the A&E Group can not use this data base to track when a commitment was addressed, by what audit and the deadline for addressing the commitment again. There is not a high level of confidence that all audit commitments usted in the 45-day audit report have been completely addressed... ATLs

[ audit team leaders) and Managers are not consistently using the data base and in some cases, do not completely understand the purpose of the data base. Finally, the data base l

is missing some data and contains some data that is no longer relevant."

It should be noted that the commitment data base includes all audit commitment sources such as NRC regulations, the FSAR,10 CFR Part 50 Appendix B, ANS! Standards, regulatory guides, INPO, etc., and not just technical specifications. The licensee was ultimately able to generate a list of all TS in its data base and those already audited.

However, this was not being used as a scheduling tool. The Director of A&E stated there is commitment to NSAB to resolve this issue by June 1998. Although much of the corrective action for the violation has been accomplished, this violation remains open pending the development of an effective scheduling tool to ensure that all technical )

specifications will be audited within a fise-year period, in addition, the remaining l

l 1

i 1

j

38 commitment data base is lengthy and cumbersome. The licensee stated they intended to

" scrub" the data base of unnecessary and obsolets commitments.

(6) Nuclear Oversight Restart Plan (NORVP)

The NORVP was established to provide independent oversight for readiness to restart and for heatup (mode 4). The NORVP meets at least weekly and frequently interacts with the line organization. Among the documents reviewed were the following:

• Work Control and Planning Brief (February 5,1998)- Presents work control and planning for restart from a Nuclear Oversight perspective

• Nuclear Oversight Restart Verification Plan - Presents Nuclear Oversight numerical evaluation of progress in the areas of leadership, corrective action, NSAB/ oversight, configuration management, engineering, maintenance /l&C, regulatory compliance, radiation protection, conduct of operations, security, procedure quality & adherence, work control & planning, training, conduct of operations and fire protection

• Millstone 3 Mode 4 Attribute Summary, February 9,1998

• 50.54(f) Recovery Oversight status report for the week ending February 7,1998 Review of the above documents and discussions with Nuclear Oversight personnel indicate significant oversight involvement in the Millstone recovery process. The reports were indicative of an extensive Oversight review. Nuclear Oversight meets three times a week to discuss station progress toward restart. Deficiencies observed in the startup l process are discussed with the line organization. The NORVP contained 21 key issues that were intensively tracked by Nuclear Oversight that gauged the performance improvements being made by the line organization. This process appears to be effective to ensure that senior management is provided with an independent evaluation of the restart process.

(7) Self-Assessment Self-assessments have become a significant portion of the 'icensee's self-improvement program. All Mitlstone organizations are required to perfot.. self-assessments including the Nuclear Oversight Group. As noted above, the Oversight self-assessment procedure is NOOP 1.05, Self-Assessment Process, Revision 0, June 30,1997. The team reviewed the following self-assessments:

• AE-97-S2 Final Self Assessment of: Audits and Evaluations C' apliance With NOOP 2.01, Rev.0, " Nuclear Oversight audits", Dated May 16 1997 l

• AE 97-S2 Self-Assessment Report of Training and Certification of Lead Auditors / Auditors in Training and Technical Specialists, Dated May 19,1997

• Self Assessment Report of Performance Evaluation Strategic Plan Development and Assessment of Organizational Effectiveness t

l 39 l

l 97-PE Self-Assessment Report of Applying PE QC Hold Points and AWOs by PE QC AWO Reviewers, Dated January 6,1998 97-PE Self-Assessment Report of Performance Evaluation's Work Process, Dated i January 12,1998 l l

• 97-AE Self-Assessment of the Audits & Evaluation Groups's Audit Commitment l Tracking Process i

)

l The team observed that the self-assessments were in-depth and effective. The se'f-assessment process allows for process deficiencies to be identified and brought forward in a formal manner. A response to the NSAB is required and corrective action commitments must be made. The self-assessment process is proactive rather waiting for problems to i arise or to be discovered by outside assessment groups. As noted above self-assessment  ;

97-AE-08 noted significant weaknesses in the audit scheduling process. " DRAFT" Nuclear i Oversight Plans for Audits and Evaluation for 1996-2000, identifies future planned self-assessments. -l I (8) Performance Evaluation The Nuclear Oversight Performance Evaluation Group performs surveillance of work activities and quality control hold point inspectioris for specific work activities. The

! surveillance technicians and quality controlinspectors comprise two separate groups. The i aerm surveillance as described in this section applies to periodic oversight of various work l activities and does not refer to surveillance testing as required by the technical i specifications. Su.ve:llance activities are controlled by the following procedures

• 1

• NOOP - 4.02, Performance, Reporting and Follow-up of Surveillance Activities and Field Observations at the Millstone Station, Revision 1, Dated May 20,1997

• NOOP - 4.09, Planning, Scheduling and Administration of Quality Surveillance Activities, j Revision 1, Dated December 19,1997 There are 27 surveillance technicians, of which nine are assigned to Millstone Unit 3  ;

j activities. .There are 23 OC inspectors. For both the technicians and inspectors, there is a mix of permanent employees and contract specialists. Most technicians and QC inspectors were former maintenance workers or l&C technicians, thus giving them credibility with the personnel performing the maintenance.

The team reviewed documentation of the following completed survei!;ance activities:

l

• MP3-P-97-132, System Engineering Communication, September 23,1997 l

• MP3-P-97-105, Cortical Maintenance, Dated September 23,1997

• MP3-P-97-123, Shielding Program, Dated October 7,1997

40 l

• MP3-P-97-136, Control of Overtime, Dated November 3,1997 )

i

• MP3-P-97-149, Procurement - Vendor Control, Dated December 4,1997 '

• MP3-P-97-153, AWO Quarterly Review, Dated December 6,1997 l

• MP3-P-97-118, Material Condition - Field Walkdown - Housekeeping / Material storage, Dated December 30,1997

• MP3-97-154, Minor Modifications, Dated January 8,1998

• MP3-P-98-001, Conduct of Operations, Dated January 13,1998

• MP3-P-98-005, AWO Quarterly Review, Dated January 19,1998.

Surveillance activities for major areas and planned activities are scheduled approximately )

six months in advance. The team reviewed the current six-month schedule which called j for 27 surveillance for Unit 3 alone. Surveillance for emergent maintenance activities are i scheduled as they occur. Not all jobs are reviewed, but there is an attempt to provide some oversight of most major work activities.

c. Conclusions l

The team determined that the Nuclear Oversight organization is effective in performing audits, general plant oversight and work surveillance activities. Considerable improvement is ncted since independent assessments identified substantial weaknesses two years ago in the performance of QA activities. Procedures and audit quality have improved. Audit findings are more meaningful and there is good control in the follow-up of audit findings.

Audit scheduling has improved, but there are still weaknesses that have to be resolved.

There is much better communication between the line organization and Nuclear Oversight.

4.2 Effectiveness of Nuclear Oversight - Quality Control

a. Inspection Scope The effectiveness of the Quality Control (QC) Department was assessed by reviewing procedures, interviewing personnel and accompanying inspectors on inspections.

b. Observations and FindinOs l The QC inspectors interviewed were experienced and qualified in their areas c,f expertise.

l The licensee utilized a mix of staff and contract personnel to perform the QC inspection l function. QC Inspectors were knowledgeable of the site work control process and documentation. All of the inspectors stated they would stop work if required. During observations of inspections in the field, two inspectors stopped jobs due to questions with proposed signoffs. All of the interviewed QC inspectors stated that they now felt they had management support to stop jobs if required.

l

41 During this inspection, the team watched one surveillance activity in the field. Testing of the Unit 3 "B" Emergency Diesel Generator was stopped due to a leak on the air start line.

The leak had been caused by chafing between the air line and the fuel line. AWO M3 2900 was issued to replace and reroute both lines to eliminate the tubing contact. While there was no specific procedure for this job, generic procedure CMP 721 A," Installation of Instrument Tubing, Fittings and Supports" was used as the basis for the tubing installation.

QC hold points were estab!!shed by this procedure.

The arveillance was comprehensive and reviewed all aspects of the job. There was good interface between the surveillance technician and the maintenance workers performing the job. AlthougS not witnessed by the team, the surveillance technician briefed the workers as to the results of his oversight at Job ::ompletion. Field observation checklist MP3-P 004-F18 was issued on February 12,1998, giving the results of the surveillance. The surveillance identified one deficiency concerning the f act that a systems engineer at the job site did not have controlled copy of the drawing in use. However, the maintenance workers did.

The team also observed a QC inspection of the foregoing job. As part of the hold point the inspector verified tubing connections and tubing bend radiuses and then signed off on a verification sheet. The QC inspector stated that he also looks at more than just what it called for in the hold point. Overall, the QC inspection appeared to be acceptable.

The team's evaluation of both the surveillance and OC inspection activity is that they were effective and comprehensive.

Although the observed inspections were generally rigorous, the NRC observed one instance that indicated an inspection performed as part of the spent fuel pool foreign material  ;

exclusion (FME) program was not adequately performed. The purpose of the inspection was a monthly requirement to sign for a list of objects allowed within the FME barrier and to determine that no other objects existed within the exclusion zone. The NRC team noted a row of tape pieces hanging from the underside of the spent fuel pool bridge. The tape j was not listed as authorized and was not noticed by the QC inspector. Although the tape was not noted by the QC inspector, the team did raise other concerns with staging in the area that was not tracked with the list. A condition report was written to document the tape. However, the response went to the reactor engineer and was being treated as only a technical issue. Through interviews of the QC manager, the team identified that the QC inspector iss 'e was not being addressed. This is considered a minor, noncited violation in accordance wuh the NRC Enforcement Policy, NUREG 1600,Section IV (NCV 50-423/97-82-11). )

I The QC support group, which was developed to standardize the inspections performed by I the QC inspectors, was a strength. The QC support group reviewed all work packages for l QA Page work orders to identify hcid points prior to the packages going to the field.

Non-QA work packages optionally may receive review by QC. The group developed standardized inspection points for many routine work activities and group members coordinate closely with each other to assure standardization. Additionally, the group had a i rotational assignment that v as filled by an inspector from the QC inspection staff. The l purpose of this position was two-fold. It brought the QC inspector's perspective into the l I

l

\

42 process and trained the inspectors on the group's policies and procedures. This was essential because emergent work packages developed after normal working hours were processed by the field inspectors in the absence of the QC support group.

l c, Conclusions Quality _ Control was generally effective in performing the required in-plant inspections.

However, the spent fuel pool FME inspection was noted as an inspection that was not adequately performed. This is a minor, noncited violation. The QC support group was effective in establishing and standardizing the use of QC hold points in work packeges.

4.3 Followup of Previously inspect!on Findings - Nuclear Oversight 4.3.1 (Closed) Inspector Follow Item (IFI) 423/96-06-17, Joint Utilities Management Assessment (JUMA) Concerning the Effectiveness of QA; (Open) Violation 423/96-0512; Failure to Audit All Technical Specifications Within a Five Year Period; (Closed) MC 0350, Restart Checklist items C.1.4.a,b &c and C.2.1.c; (Closed - SIL ltem No. 73)

The JUMA was performed during June 1996. A gubssquent NRC inspection Report,96-06, referring to the JUMA results stated, in part, that:

" The JUMA team concluded that the audit, surveillance and inspection programs at l Millstona were not effective in the implementation of their mission statement and the i resolution of identified problems. The [JUMA] team attributed these problems to:

• Lack of support for the QA organization by the executive and line management

• Lack of an effective action program."

The NRC made the licensee JUMA corrective actions an IFl in IR 96-06.

l

( Section 4.1 of this report noted an increase in the effectiveness of the Nuclear Oversight

, organization and increased support for the Nuclear Oversight organization by the NU l President and CEO. There is increased interaction with the line organization and Oversight

! is involved with corrective actions to deficiencies identified by them. The licensee choned l

the team documentation that they had been responsive to all JUMA concerns. Section 21 of this report demonstrates improvement in the licensee's Corrective Action Process. It is not the intent of this inspection t make a judgment as to the adequacy of the Ucensee's response to the JUMA. However, because of NRC observations made during this inspection, IFl 96-06-171s considered closed.

l The licensee's response to Violation 96-05-12 concerning the scheduling of audits of the technical specifications is discussed in report section 4.1. Although the licensee has made considerable progress in this area, it is not yet effectively using its data base to schedule TS audits. As stated in section 4.1, this violation remains open pending completion of all l

i 43 licensee corrective actions, but based on corrective actions already taken, this aspect of Sil Item No. 73 is closed.

An issue identified by the JUMA audit team involved QC inconsistently assigning hold points in tvork packages. This issue was addressed by the licensee by establishing the QC 1 support group. The procedures for reviewing work packages and assigning hold points were formalized. The QC support group reviewed all work packages to identify hold points '

prior to the packages going to the field. The group developed standardized inspection i point for many routine work activities and work members coordinato closely with each l other to assure standardization. Additionally, the group had a rotational assignment that was filled by an inspector from the QC inspection staff. The purpose of this position was i two-fold. It brought the QC inspector's perspective into the process and trained the inspectors on the group's policies end procedures. This was essential because emergent work packages developed after hours were processed by the field inspectors in the absence of the QC support group.

The tearn reviewed several work packages and concluded that the inspection points were

appropriately selected. Based on the program improvements and implementation of the process by the QC support group, the team concluded that the issues identified in the l JUMA , report were adequately addressed.

l NRC Manual Chapter 0350," Restart Approval Checklist" include the following items:

l

• C.1.4.a Effectiveness of quality assurance program

• C.1 4.b Effectiveness of industry experience review program

• C.1.4.c Effectiveness of licensee's independent review group

• C.2.1.c Managemerit involvement in self-assessment and independent self-assessment capability Each of the above areas are addressed in various parts of this inspection report and have l

been found to be sufficiently acceptable to close these areas. Based on the above, SIL ,

j ltem No. 73 is closed. l l

l 4.3.2 (Closed) Unresolved item (URI) 50-423/95-81 Lack of Trending of Non-j' conformance Reports (NCRs) and Level "D" ACRs ; j l (Closed - SIL ltem No. 41) ]

i, inspection Report 95 81 stated, in part, the following: "... QA did not trend any NCR or look for adverse trends which may be discernible from such data ... the lack of Level "D" o ACRs may mask a recurring problem and its significance. Formalizing the practice of trending (JCRs and Level "D" ACRs would address a potential weakness in the program i and make the process less reliant on individual analysis and perception of the significance j of the recorded problem."  !

I

i 1

44 "The acceptability of the lack of trending of NCRs end verifying the effectiveness and adequacy of the ACR data base by the Quality Assurance Department remains unresolved.

Since Inspection Report 95-85 was performed, CRs have taken the place of ACRs and Level "D" ACRs have been eliminated. The team verified that NGP 3.05, "Non-Conformance Reports," has been revised to have NCRs trended in accordance with procedure RP 4, " Corrective Action Programs.' Also, a CR is written for each NCR issued.

Procedure RP 4, Paragraph 1.17, states "... at least quartorly, PERFORM trend analysis of AITTS data base related to CRs and ISSUE trend report within 30 days of the end of the of the quarter. Trend analysis shall include human errors associated with NOVs and LERs ...

BRIEF appropriate levels of organizational management of reisults of trend analyzed trend data ... ENSURE CRs are iriitiated for any adverse trends identified through periodic trending." The team reviewed recent trend reports and verified that CRs and NCRs (which are trended separately) are now trended at least quarterly. Based on the above review, URI 95-81-01 is closed.

The licensee's actions taken in response to the findings of its ACR-7007 task forcu was addressed in report section 3.2, above. This closes SIL ltem No. 41.

4.4 Performance of the Nuclear Safety Assessment Board

a. Inspection Scope The performance of the Nuclear Safety Assessment Board (NSAB) was evaluated by reviewing board meeting minutes, observing a board meeting and interviewing board members,

b. Observations and Findings The NSAB consisted of senior plant managers and two independent contractors. All of the board members met the TS requirements for education and background. Section 6.5.3 of technical specifications specifically stated the areas of expertise required for representation on the NSAB. To assure compliance with these requirements, the board secretary maintained a matrix of qualifications of the board members. The list, as originally presented to the NRC, did not indicate any regular members as having metallurgy l experience.

As a result, an alternate member with specific metallurgy experience was added to the list of attemate board members in February 1997. However, based on the review of the qualification, matrix, the NRC concluded that the technical specification was not met l

because no regular member possessed metallurgy experience.

The licensee re-evaluated the qualification matrix and concluded that its previous screening of qualifications was too conservative. The previous screening criteria required an academic degree or direct experience to be counted as a metallurgist. However, this screening was in excess of the technical specification requirements. Based on the new screening criteria, two current members were identified with metallurgy experience.

I k

1 i

l 45 l

Although the NRC did not contest the actual board member's qualifications, the evaluation I and resolution of this discrepancy in 1996/1997did not meet the literalinterpretation of l the technical specifications. This was an example of non-conservative interp.etation and j implementation of technical specifications. This is considered a minor, noncited violation in accordance with the NRC Enforcement Policy, NUREG 1600,Section IV (NCV 50-423/97- {

l 82-12). i 1

Portions of an NSAB meeting were observed on January 29,1998. The meeting consit.ted largely of presentations to the board by various managers regarding departmental readiness for restart. Several board members asked probing questions and displayed significant knowledge of the issues and obviously were prepared for the meeting; while in contrast, there was only limited participation by a few of the members.

The board identified a potential safety issue with fire protection systems that have l outstanding surveillance tests. The board questioned the acceptability of prolonged use of compensatory measures. The fire protection program manager was requested to respond to the board at the next meeting. Action items assigned during meetings were tracked by the NSAB secretary and the status of open items were included as part of the meeting minute packages. Closse of open items required consensus of the board membership. j Another example of the NSAB providing appropriate oversight of plant activities was weaknesses in the area of training. As a result of NSAB intervention, the training on-site was stopped pending program improvernents. Additionally, the board thoroughly probed the area of operational experience (OE). To validate a presentation by the OE program i manager, the board questioned various department managers on the use of OE by their departments during subsequent presentations.  !

l The NSAB audit program was appropriately implemented by the Audits and Evaluation 6 Department. The scope of the audit plan was reviewed by the board and audit results were presanted to the board by the auditors. Follow-up of a previously identified violation on the technical specification audit program is addressed in Report Section 4.3.1.

The NSAB also sponsors additional third party reviews. These are performed by members of the licensee staff or supported by external organizations. Examples of these are: the Northeast Utilities Nuclear Group Nuclear Oversight Assessment independent Assessment I Team report, dated July 1997, an INPO review of safety evaluation screens and of the safety evaluation process and safety system functional reviews. Additionally, the NSAB members are assigned systems to physically review and inspect with the system engineers, and make simulator ' walk-through' to review the operator tasks. ]

c. Conclusions

! The NSAB was effective in reviewing activities on-site and identifying potential nuclear safety issues. The implementation of the NSAB met the technical specification requirements. However, the 1997 resolution of an issue of membership qualifications did .

not meet technical specifications as presented. This was identified by the NRC as a  !

noncited violation in the implementation of Section 6 of the technical specifications, l

i i

i

i l

l 46 l

4.5 Performance of the Plant Operations Review Committee

a. Inspection Scope The performance of the Plant Operations Review Committee (PORC) was assessed by reviewing committee meeting rrinutes, observing board meetings and interviewing board members,

b. Observations and Findings l

l During observed PORC meetings, PORC membership met the technical specification l

requirements. PORC members were prepared for the issues on the agenda and asked technical questions of the presenters. The questions focused on safety and compliance with regulatory issues. The PORC meetings were conducted in a professional manner with an emphasis on clear communications between the committee and the presenters.

Station expectations for the conduct of PORC were high, as demonstrated by several assessments cf PORC performance. The PORC process was being evaiuated at the station level as including the performance of the committee as well as the personnel presenting

items to the committee. Instances such as presenters being unable to answer PORC questions and PORC-rejected documents were considered weaknesses by the licens a. An effort was under way to develop a feedback process to track and teduce poor presentations to PORC.

PORC issues were tracked in the plant action request tracking system by the PORC secretary. A review of the backlog indicated timely resolution of PORC issues.

I

c. Conclusions I PORC was effective in accomplishing the reviews required by technical specifications.

Station evaluation of PORC performance and the standards being demonstrated set high l standards for the documentation and review of technicalissues.

l 4.6 Performance of the Site Operations Review Committee i

a. Inspection Scope The performance of the Site Operations Review Committee (SOHC) was assessed by reviewing ccmmittee meeting minutes, observing board meetings and interviewing board members.

b. Observations and Findings l

During observed SORC meetings, SORC membership met the technical specification requirements. However, technical specification 6.5,2.2 designate that the Senior Vice President and Chief Nuclear Officer (CNO) as the SORC chairperson. A review of SORC  ;

meeting minutes demonstrated that the chairperson responsibility was normally delegated to the Director of Unit Operations. This causes a conflict with the TS requirements for

47 composition of the SORC, and TS 6.5.2.7.b that the SORC will provide written notification of a disagreement between the SORC and the Senior Vice President and CNO. The team did not take issue with delegation of SORC chairperson responsibilitit.-

SORC member.s were prepared for the issues on the agenda and asked te,hnical questions of the presenters. Members adequately represented the site-wide perspective of SORC.

This site presents a particular challenge with the differences in license requirements between the units. SORC members displayed a combination of knowledge to integrate site-wide license and technical requirements. This was evident in a discussion of fire protection issues which required detailed knowledge related to all three units. During one meeting, the team noted four occasions on which safety issues which required further l

evaluation were identified by the committee. l l

SORC issues were tracked in the plant action request tracking system by the SORC

]

secretary. A review of the backlog indicated timely resolution of SORC issues.  !

c. Conclusions SORC met the technical specification requirements and was effective in identifying potential safety issues.

4.7 Performance of the Independent Safety Engineering Group

a. Inspection Scope The performance of the independent Safety Engineering Group (ISEG) was evaluated by reviewing ISEG reports, reviewing the operating experience (OE) program implementation, reviewing ISEG backlogs and interviews of personnel.

1

b. Observations and Findings Millstone Unit 3 technical specifications required an ISEG consisting of four full-time personnel to perform independent reviews of plant operations. The unit 3 ISEG consisted of three full-time engineers, one part-time contractor and a supervisor. The ISEG charter required the group to use operating experience when reviewing plant operations and to make detailed recommendations to improve safety and reduce human errors. in implementing this charter, the licensee elected to make the ISEG the site group responsible for reviewing and implementing OE. OE typically comes in the form of industry group reports, vendor notifications and NRC information.

Over the past two years, the ISEG focus appears to have shifted from the ISEG activity of independent reviews of plant operations to the performance of OE reviews. A large backlog of unreviewed OE items was processed, which resulted in a reduction of the number of ISEG activities performed. The number of ISEG reviews done in 1997 was only 12, down from 24 the previous year, i

Reviews of plant operations performed by the ISEG group were documented in reports.

Reports reviewed by the team indicated that the ISEG performed critical reviews and made L_____---_--_-_--_________

4 1

48 appropriate recommendations. The independent reviews included human performance evaluations intended to improve safety through human error reduction. Examples of these included: An evaluation of personnel errors on entering the radiation control area (RCA) without proper dosimetry, an evaluation of human errors at another facility where liquid sealant migrated into the reactor vessel head vent system, an evaluation errors in the foreign material exclusion program, an evaluation of overtime controls for nuclear group personnel, an evaluation of work activities in the spent fuel pool, an assessment of the vendor information review program, an evaluation of high-voltage switchyard work and an l evaluation of the administrative controls for the safety injection accumulator isolation valves. The ISEG group used the AR process to track the implementation of recommendations and performed closecut inspections of each item.

The team followed up on the condition report from the ISEG review of the high-voltage switchyard work. As discussed in Report Section 2.2, this ISEG evaluation identified several significant issues which involved personnel safety and the potential loss of off-site power. ISEG promptly addressed the issue by stopping work in the switchyard. The initial response to the issue was good. However, the condition report was processed as only a ,

Level 3. This was not identified or challenged by the ISEG group. NRC review of this i issue indicated that the condition report should have been processed at a higher level.

The team also learned that the ISEG had not been involved in any reviews of the CR program, its status, or the impact of the backlog of high-priority CRs. This determination further reinforces the team view that the ISEG workload is unbalanced and has not been focused on plant and management controlissues.

By focusing on operating experience, the ISEG group reduced the backlog of OE issues from several hundred to approximately 40 for Unit 3. A sample of OE packages was reviewed and were found generally thorough and complete. However, one weakness was identified in the review conducted for NRC IN 97-14 related to spent fuel pool cooling. The information notice suggested reviewing siphon breakers used to prevent spent fuel pool draining. The OE review confirmed that the siphon breakers were installed. However, the reviewer stated no future inspection or maintenance was required. The team considered that the issue of potential fouling of the siphon break orifices or the need for periodic inspections was not adequately evaluated. The licensee was reviewing this issue at the I close of the inspection.

l Although the backlog of OE issues had been significantly reduced over the past year, the l amount of work represented by the remaining issues was significant. Several of the issues remaining in the backlog had a high probability of identifying safety significant issues. One example is NRC IN 97-78, which involved identifying unreviewed safety questions related to the use of manual actions in place of automatic actions in emergency operating procedures (EOP's). The review of this issue was completed for unit 2, which identified several areas of concern. Additionally, a previous NRC inspection identified an issue involving manual action for a control room ventilation system that cannot be accomplished within the time assumed in the safety analysis. Based on these factors the priority of evaluating this issue appears inappropriate.

l r i l

o-_----_---_-- _ - - - - _- J

49 l l

Site implementation and use of OE was mixed. OE is nnt consistently being used by the working groups at this time. A key reason is the' .ne site-wide procedure to establish the expectations for the use of OE was in development and not yet issued. Once a site-wide procedure is issued, the departments still have to develop implementing procedures. The OE Minute, a daily publication listing relevant OE items, appeared effective in disseminating OE to the site on a daily basis. However, the information was not easily retrieved after the fact. Industry and NRC information was sent to appropriate personnel (system engineers, operations) for information. However, the site responsibility for evaluating and identifying actions was with the ISEG group. Access to the industry nuclear network data base was improving, but was still lacking in some areas such es system engineeririg.

(1) Organizational Independence of the ISEG J

The team identified that the licensee had made the provision to use human factors / engineering personnel to support ISEG review of plant operations, but would allot them to review their own group's work. Independence is required by Section 6.2.3.3 of the Unit 3 technical specifications made the following statement:

"The ISEG [ Independent Safety Evaluation Group] shall be responsible for maintaining ,

surveillance of unit activities to provide independent verification + at these activities are l

performed correctly and that human errors are reduced as much as practical."  ;

However the licen.9e's memorandum dated December 23,1997, made the following statement which would allow personnel to participate in review of work previously performed by their own group:

"In the event that the HF/E [ human factors / engineering] review effort needs to be audited or become a part of a " Nuclear Oversight" assessment, an independent party will be assigned to ensure that independence is maintained. HF/E personnel may participate in Nuclear Safety Engineering ISEGs and in Nuclear Oversight audits and assessments that do not include previous HF/E involvement in design changes, MCB 1 main control board] changes, procedure reviews, etc."

The team concluded that the licensee's position on this issue is inconsistent with Technical Specification 6.2.3.3. The team did not find any instances where there was a deficiency in the required organizationalindependence. Licensee actions on this issue will be reviewed in the future (IFl 50-423/97-82-13),

c. Conclusions The ISEG was staffed and met the technical specification requirements. However, the implementation of the OE reviews by the ISEG was limiting the number of independent reviews of plant operations performed by this group. ISEG independent revie'as and ISEG reviews of OE were generally thorotgh. Corrective actions for ISEG items were tracked and verified by the ISEG vior to closure. The team identified the potential for a deficiency in the required organizational indepenaance of an ISEG review.

Management Meetings

i 50 l X1 Exit Meeting Surnmary I

The team members presented the inspection results to members of licensee management i and area coordinators at daily meetings during the inspection. The licensee acknowledged l the findings presented. The inspection findings were discussed with licensee l representatives at a public exit meeting on February 26,1998. I i

l l

l 1

(

51 INSPECTION PROCEDURES USED t IP 40500: Licensee Self-Assessments Related to Safety issues inspections IP 92903: Follow-up Engineering I

l I

i l

l l

o i

t i

52 ITEMS OPENED, Cl.OSED and DISCUSSED Ooened IFl 50-423/97 82-01 Corrective Actions Program lacked controls over combining CR to preserve issues and significance level (Section 2.1).

VIO 50-423/97-82-02 Failure to complete corrective action for SORC review of the ISEG procedure (Section 2.3).

VIO 50-423/97-82-03 Failure to identify and correct the air binding of the boric acid transfer pumps (Section 2.3.1).

l VIO 50-423/97-82-04 Inappropriate Significance level assigned to CRs M3-97-4672 and M3-97-4346 (Section 2.3.3).

VIO 50-423/97-82-05 Radiation Protection Manager lacks organizational independence (Section 4.1)

VIO 50-423/97-82-06 Failure to complete corrective actions for SW flow indicators 3SWP-F1-059 A, B and C (Section 3.1).

VIO 50-423/97-82-07 Failure to complete corrective actions for ACR M3-97-0506, AR 97003960-05(Section 3.2.b.5).

IFl 50-423/97-82-08 Quality lacking in the Setpoint Control Program Section 3.2.b.6).

VIO 50-423/97-82-09 Failure to follow *.he requirements of PI 29 and develop Design J Basis Summaries for the Emergency Lighting System and CVCS (Section 3.2.b.7).

VIO 50-423/97-82-10 Failure to follow the requirements of NGP 5.28 to maintain the Safety Functional Requirements manual (Section 3.2.b.8).

NCV 50-423/97-82-11 Minor, noncited violation concerning the effectiveness of FME control (Section 4.2).

NCV 50-423/97-82-12 Minor, noncited violation concerning the documented qualifications of the NSAB members (Section 4.4).

l IFl 50-423/97-82-13 HF/E personnel may lack organizational independence required by Technical Specification 6.2.3.3 for ISEG assessments ,

(Section 4.7). l Closed  !

IFl 50-423/96-06-17 4.3.1 l URI 50-423/95-81-01 4.3.2

i 53 Discussed VIO 50-423/96-05-12 4.1.b.5 l

r i

1 l

l l

{

p l

l 1

l I

1 I

j I

i.

t i

54 LIST OF ACRONYMS USED

- ACR(s) adverse condition repo.-t(s)

AITTS Action item Tracking and Trending System AR(s) action request (s)

ARP(s) alarm response procedure (s)

ALARA as low as reasonably achievable AOP(s) abnormal operating procedure (s)

ATL(s) audit team leader (s)

AWO(s) automated work order (s)

BAST (s) Boric Acid Storage Tank (S)

CAP corrective action plan CCP Component Cooling Water System CFR Code of Federal Regulations CMP configuration management plan CNO Chief Nuclear Officer CR(s) condition report (s)

CVCS Chemical & Volume Control System DBDP(s) design basis documentation package (s)

DCM Design Control Manual DCN(s) design change notice (s) I DCR design change record DRS Division of Reactor Safety ECOP- Employee Concerns Oversight Panel ,

ECP Employee Concerns Program EDG(s) emergency diesel generator (s)

EOP(s) emergency operation procedure (s) j EQ environmental qualification FCAT Fundamental Cause Assessment Team FME foreign material exclusion FSAR Final Safety Analysis Report l GL Generic Letter HELB high energy line break l HF/E human factors / engineering HVAC heating, ventilation and air-conditioning IFl inspector follow item INPO Institute of Nuclear Plant Operations IPE Individual Plant Evaluation ISEG Independent Safety Engineering Group JUMA Joint Utilities Management Assessment LER(s) licensee event report (s)

LOCA loss of coolant accident M&TE measuring and test equipment MCB- main control board MMOD minor modification MRT- Management Review Team MSEE maintenance support engineering evaluation MSR moisture separator reheaters l

i l

l.

55 NCAT Nuclear Committee Assessment Team NCR(s) nonconformance report (s)

NGP(s) nuclear guidance procedure (s)

NNECO Northeast Nuclear Energy Company NORVP Nuclear Oversight Verification Plan NOV(s) Notice of Violation NRR Nuclear Reactor Regulation NSAB nuclear safety assessment board NSIC Nuclear Safety Information Center NSSS nuclear steam system supply NU Northeast Utilities NUMARC Nuclear Management and Resources Council NUQAP Northeast Utilities Quality Assurance Program NUREG Nuclear Regulation NUSCO Northeast Utilities Service Company ,

OCA Office of Congressional Affairs I OEDO Office of Executive Director for Operacions OSTI Operational Safety Team Inspection P&lD(s) Piping and Instrument Drawing PAO Public Affairs Office PDCR plant design change request PDDS Plant Design Data System PDR Public Document Room PIR plant incident report PLC post-LOCA cooling PLS Westinghouse Precaution, Limitations and Setpoints PMMS production maintenance management system PORC plant operation review committee l Pzr pressurizer  !

QA quality assurance QAS Quality and Assessment Services RG Regulatory Guide SBO station blackout SCWE Safety Conscious Work Environment Program SFR Safety Functional Requirements Manual SIL significant item list SORC site operations review committee SPDS Safety Parameter Display System SPO Special Projects Office SW service water TS(s) technical specification (s)

UFSAR updated final safety analysis report l URl(s) unresolved item (s)

USO(s) unresolved safety question (s)

Vac volts, alternating-current Vdc volts, direct-current VIO violation

l 56 l VTM vendor technical manual WC work control WP&OM work planning & outage management I

I l

l l

l l

T6m: Hers;sre some' comments in ths form 'of a red line/ strike ~ outiersiori.BTheseiare

~

~

mostly all1 editorial,;so;take whstever you see fit.' .On my c'ouple of. findings,11 added some clarification to ensure that they_were accurate.- LJim.Higgins Mr. Martin L. Bowling Recovery Officer - Millstone Unit 2 c/o Mr. H. Miller Northeast Nuclear Energy Company P.O. Box 128 Waterford, CT 06385-0128

Dear Mr. Bowling:

This letter provides the preliminary results of the NRC Region I team inspection of Northeast Utilities (NU) controls in identifying, resolving and preventing issues that degrade the quality of plant operations or safety at Millstone Unit 3. This team inspection was performed -onsite from February 9 through February 20,1998. The detailed findings of the team inspection will be documented in inspection report 50-423/97-82. The inspection team leader provided you with the results of the inspection at a public meeting on February 26,1998.

Inspection Scope Our inspection examined the management processes that are.used to provide direction to the plant staff In; order to facilitate effective and safe plant operations. This was accomplished by reviewing your -goals and expectations, communications and teamwork, receptiveness to problems brought forward, performance monitoring, and theiryour commitment to resolve safety committee recommendations and audit and

~:entandit/ assessment findings.

Our inspection also assessed the adequacy of your corrective actions program including processes for identification, analysis and resolution of plant deficiencies. We also evaluated your orgenbetion organization's responsiveness in dealing with issues brought forward by employees through various channels including your employee concerns program. We examined the backlog of open problem reports to verify that safety significant issues are being tracked to completion. We reviewed the process to prioritize corrective actions based on risk, and evaluated your process for assessing the effectiveness of corrective actions.

The inspectors evaluated your process for site and departmental self assessments. We reviewed your corrective actione in place for several significant self assessments and third party audits including actions applicable to Unit 3 from the'ACR-7007 Event Response Team Report, and the actions to improve the Nuclear Oversight Department taken in response to the 1996 Joint Utilities Management Assessment. In addition, we examined the effectiveness of your Performance and Evaluations Group in their audit, surveillance and quality control function.

We also observed -the Nuclear Safety Advisory Board (NSAB) and ! c; the on-site safety committees, the Site Operations Review Comtmttee (SORC) and the Unit 3 Plant Operations Review Committee (PORC). We else-reviewed -your operating experience l

. i l.. . u ._ _ . . _ . . . _ _ . _ _ __ .___. _ _.._ _ ...____

2 M. L. Bowling program, including the programs for evaluation of industry data and site experience.

Overall Assessment of the Corrective Actions Program The team found a structured framework in place that provided a strong definition for the corrective actions program. Also, there is evidence of e-good dea! cfsignificant management attention applied to infuse quality into the program implementation process.

The team observed a general improving trend in quality over the last year for most program aspects, thatur,-includind issue identification, classification, analysis and actions to prevent recurrence. However, deficient conditions were found to exist m some specific; root cause ana!ycicanalpses and in ecmc corrective actions.

Our evere!! :::ce mert of-the-eenestive action pregicm it, !!heugh0ysrall we have seen evidence efthat the corrective action program beginning to functien,ls functionir@,1but it is

~

clear that the program will c6ntinue to require careful monitoring by NU. For example; the team found a dieproportionalnotable number of ' additional. issues in a relatively small sample size of Condition Reports; after NU had completed an extensive self assessment.

Preliminary Findings identified by the NRC Team The NRC inspection team identified regulatory issues within the scope of the areas examined. OwBased on~our preliminary evaluation-iedicates that/ the following findings are being considered as potential violations.

l In the area of Design Control, we found problems in the Master Setpoint List (MSL), in that

~

l it was inconsistent andldid not contain all of the setpoint and the calculation i referencereferences as required by NGP 5.23. Also, a potential deficiency with the sub-cooling margin setpoints was found, in addition, the team questioned the adequacy of setpoint control based on our observations.

There were two procedural problems identified with maintaining accurate design basis documents. The Safety Functional Requirements Manual and the Design Basis Summaries were not maintained as required by NGP 5.28 and PI-29, respectively.

The team also found problems with meeting the organizational independence required by Section 6 of the Millstone Unit 3 Technica? Specifications because of the reporting l

relationship between the Radiation Protection Manager and the Maintenance Manager.

1 These requirements are established in TS 6.2.1.d, the Updated Final Safety Analysis Report and Regulatory Guide 8.8 Also, the independence required for persons performing the Independent Safety Engineering Group (ISEG) reviews for human performance issues did not appear to meet the intent of:TS 6.2.3.3.

The team found severalindividual problems within the Millstone Corrective Action process.

These problems include -incomplete corrective actions and root cause analyses, as illustrated by the repetitive air binding of the Boric Acid Transfer pumps, issue closure without completion of all corrective actions, as was the case with Design Change Notice

~

3 *I i M. L. Bowling l (DCN) review required by CR M&97 050M3-97-0506,and also the closure of an l Automated Work Order (AWO) to correct service water flow instrument anomalies that was associated with Operator Work Around 96-03. We note that completion of the DCN review was an Adverse Condition Report (ACR) 7007 and Configuration Management Program restart commitment.

The team also identified some procedural problems with assigning inappropriate significance levels for Condition Reports. These included the following: incomplete action on Generic Letter (GL) 89-13, Service Water Fouling, and GL 90-03, Vendor Technical Information Program. These are commitments to the NRC and should have been classified as at least Level 2. We also observed that some. actions [onlthe Condition Reports for the incomplete Generic Letter issues had been inappropriately coded as ' Deferred' until after plant restart.

Additional Findings in the management area, the team found that management communications methods with the plant staff were a strength. There was a common understanding of management's expectations by plant personnel. However, it was noted that a strategic plan and vision statement on where the plant is headed afewere stillin draft. This is considered a weakness [in view of the fact that the current management has been in place since late

~

1996f Hcrd to saylgiven^the multiple management changes since/96]. Overall, the Nuclear Group Policies and Standards, were considered good. .^'though, teemworkTeamwork initiatives at the first line supervisor and above were developed, there is-aand now need to extend thicbe extended to the worker level.

Observations and interviews show that managers and supervisors encourage employees to identify problems. The plant staff feels that management is receptive to problems brought forward, and individuals generally characterized the environment as improved and currently receptive to problem identification. There is no reluctance or rccervctienciessvation expressed by individuals to identify problems:: through the Corrective Actions Condition Report process, to the Employee Concerns Program (ECP)l or to the NRC.

The handling of individual HIRD cases by the Employee Concerns Program and the Safety Conscious Work Environment (SCWE) program is adequately responsive to specific case needs. Both technical and humer cidehuman problems are generally well addressed. The ECP case intakes, and the Employee Concerns Oversight Panel (ECOP) oversight activities and surveys, are used to identify potential or actual HIRD problems or organizational units which exhibit barriers to free identification and reporting of problems. These are positive contributions to the overall process. These mechanisms are effective, especially for the more egregious issues identified as problem areas. However, NU has not adequately dealt with trends, common causes and the overall occurrence rate of HIRD allegations to ECP organization-wide. A significant backlog of HIRD allegations were pending investigation and the backlog and emergent HIRD allegations had not been analyzed by NU for broad trends or patterns, and common causes. As a result, the actions taken to date had not ]

been effective in curtailing an adverse trend in the incidence of HIRD allegations. Further, the SCWE processes have not yet been formalized, thetasthus the program lacks structure

i l

i 4

M. L. Bowling in the form of procedures, formal processes and documentation requirements.

(NU has not been effective in preventing the ongoing emergence rate of HIRD allegations to ECP. The incident rate of HIRD allegations and management-related or induced chilling effect events has not diminished significantly. Management actions to reduce the overall rate of HIRD allegations have not been effective, except for specific HIRD cases or explicitly identified SCWE Problem Areas.y--;Seems repetitive [ delete?] l l

ConcerningEcorrectiviaction'sl,(we"notedLa generally _ low threshold;for' recording issues;but also;noted la tendency toLassign a lower] Significance; Level classification thanl appropriate and to l wave root causelanalysisifor similar issues;

~ 1 Concer$gWefareTconcerned that thispecticelm'ay; overlook trendsTandLmay result in  !

ineffective corrective actions; in the issue addressed earlier concerning air binding of the Boric Acid Transfer pumps, NU has re-evaluated their earlier root cause analysis and conclusions and began re-analysis of the issue at the conclusion of the inspection period. However, the team noted that NU failed to recognize the potential unreviewed safety question (USQ) which resulted from their initial conclusion of a more restrictive Boric Acid Tank level requirement relative to the Technical Specification. The team also observed that the deportability evaluation of the event was incomplete and that NU failed to consider industry opercting experience in their evaluation of the problem.

[The team also found that your Corrective Actions program had inadequate controls over combining Condition Reports such that their issues were preserved and that they maintained their appropriate significance level. This weakness was evident in the handling of multiple Condition Reports concerning deficiencies in the Nonconformance Report process. We consider the poor control of this activity as a program weakness. -- Move ~t'o CA program section.)

We found that long term compensatory measures are in effect for fire protection systems because surveillance testing which verifies operability of these systems has been suspended. Compensatory measures are taken to allow restoration of a fire protection

~

system. But in this case, are being appropriately used to substitute for long term system inoperability. -The team also found that the failure to conduct the surveillance was inappropricic beccuse it sets a low standard of performance for plant personnel.

The team found that the self assessments were generally of high quality. NU has taken a strong initiative in areas like the Nuclear Oversight Restart Verification Assessment.

The NRC team found that there has been an improvement within the traditional quality assurance had quality control functionfunctions of the Nuclear Oversight orgeeirat+e+4s Audit crd Evaluation Grouporganization. The number of Auditors has significantly increased as well as their qualifications and knowledge level has increased. Audit program procedures are acceptable. There are four new audit managers. Oversight has the opportunity to concur with the corrective actions taken for audit findings and l

l

5 M. L. Bowling nonconformance reports. There is good interface with the line organization concerning the Nuclear Oversight Restart Verification Plan.

The NRC team found that the Independent Safety Engineering Group (ISEG) has performed high quality plant reviews and Operational Experience (OE) reviews. Hcuever, the number of ISEG reviews done in 1997 was only 12, down from 24 the previous year. This appears to be a result of the larheiOE workload being performed by the group. The ISEG group has reduced the backlog of OE issues from several hundred to approximately 40 for Unit 3.

NRC team members attended meetings of the Site Operations Review Committee (SORC),

the Unit 3 Plant Operations Review Committee (PORC) and the Nuclear Safety Advisory Board (NSAB). All three safety committees operated effectively, the members were prepared for the meeting and added quality to the issue being addressed. [The team observed two minor examples of nonconservative interpretation and implementation of the technical specifications regarding staffing qualifications and delegation of responsibility. --

Drop " minor" or if really minor, drop sentence from letter]

The team reviewed NU actions on several Significant items List (SIL) issues, and has recommended that all four be closed. Theselfour issues are SIL ltem 41, ACR-7007 issues

~

relevant to Unit 3?and issue concerning trending of'NCRs/Silfitem ,73/concsrning the Technical .Specificat_ ion audit program; end SIL:ltem 73/ concerning the; adequacy of the )

Nuclear Oversight Program. I The SPO staff willinclude these findings within NRC Inspection Report 50-423/97-82, which will provide the final observations, findings, and any enforcement actions to which you will be required to respond based on the results of the subject inspection. No response the issues discussed in this letter are required at this time; however any potential enforcement items which warrant prompt corrective actions should be addressed in a timely manner rather than waiting for the final report.

Should you have any questions or comments regarding the issues discussed in this letter, please contact me at (610) 337 5126or Tom Shediosky at (610) 337-xxxx.-

Sincerely, Wayne D. Lanning I Deputy Director, inspections Special Projects Office Office of Nuclear Reactor Regulation Docket Mc. 50-423 cc: See next page

9 6

M. L. Bowiing l

i l g:\tempfile\405quic.wpd i

l l

t l

1 l

l

~

Management Debrief for Millstone MC 40500lnspection Corrective Actions inspection 50-423/97-82 Objective: To evaluate the effectiveness of the NU controls in identifying, resolving and preventing issues that degrade the quality of plant operations or safety.

Management:

Hiah Level Goals and Expectations:

A Upper level meetings with plant personnel were effective.

4 Management communications methods were a strength.

4 There was a common understanding of management's expectations by plant personnel.

• Strategic plar. and vision statement on where the plant is headed are in draft. This is a weakness.

1. Overall, the " Nuclear Group Policies and Standards," were considered good; the lack of a " Nuclear Group Mission and Vision" statement, under development, was a weakness.

Organizational Communications and Teamwork:

A Vertical Communications good.

v Lateral Communications need improvement.

A Communications were adequate relative to identifying safety issues.

1. Teamwork initiatives at the first line supervisor and above were developed, strongest at the upper levels. Need to be reinforced.

v Teamwork at the worker level needs improvement.

Manaaement Receptiveness of Problems Broyaht Forward:

• Observations and interviews show that managers and supervisors encourage employees to identify problems.

A- The plant staff feels that management is receptive to problems brought forward (West).

individuals generally characterized the environment as improved and currently receptive to problem identification (Beckman).

v interview results indicate that some plant staff may not fully trust plant management.

They have expressed a " wait-and-see" view of management. They therefore pursue other alternatives to rasing problem rather than with their management.

II

' x. [- .h

l.

l 1

l Evaluation of Performance Monitorino:

A Performance indicators to evaluate the subject programs were considered excellent. O A The effectiveness ant the quality of information on performance was considered satisfactory. (Not discussed at management debrief).

A Fire Protection Surveillance Testing has been suspended intentionally , pending personnel training in the Fire Brigade Department. NU has implemented long term compensatory measures because system operability is not being verified through testing.

This is an inappropriate use of the compensat< :y maasures allowance of the TRM.

(

Reference:

M3-97-3035, M3-97 3981, M3-97-4246, M3-97-4394, M3-97-4618).

(This was discussed with the Organizational Independence issue).

v The usefulness of personnel error data, provided to line management needs improvement. That is data broken down by type of error, or normalized data to provide information. (Not discussed at management debrief).

v The high number of LER-related human errors was considered a weakness and needs to be further examined.

l Manaaement Commitment to Resolve Safety Committee Recommendations, and Audit and Self Assessment Findinas:

A Self-assessment recommendations on reactor engineering problems had been addressed.

(Not discussed at management debrief).

A PORC recommendations on a contingency plan to repair a valve were resolved. (Not discussed at management debrief).

A SPDS-related issues raised by the team were resolved. (Not discussed at management debrief).

v Radiation Protection Manager reports to the Maintenance Manager, this lacks the organizational independence specified by TS 6.2.1.d, the FSAR and Regulatory Guide 8.8. (Discussed as an example of TS Section 6 organizational independence issue).

(Other references CR M3-97-1875, and audit recommendations MP-97-AO6-02).

e !SEG independent verification of human error reduced as much as practical, does not meet TS 6.2.3.3. (Discussed as a second example of TS Section 6 organizational independence issue).

v NU Self-assessment findings "...the failure to fulfill ANSI /ANS 3.5 standards requirements. ." was considered a weakness regarding simulator and plant fidelity

(

Reference:

Self Assessment 97-004). (Not discussed at management debrief, Dr. West will discuss issue with NRR staff reviewing simulator certification).

l \

1 v lt was considered a weakness that design engineering determines whether modifications require human factors review (

Reference:

NGP 5.25).

Organizational Independence:

v Two issues of failing to meet OrganizationalIndependence requirements of the TS:

Radiation Protection Manager reports to the Maintenance Manager, this lacks the organizational independence specified by TS 6.2.1.d, the FSAR and Regulatory i Guide 8.8. And, l l

lSEG independent verification of human error reduced as much as practical, does f not meet TS 6.2.3.3. l l

1 This appears to be a Violation of the requirements of TS Section 6.

Inaoorooriate Use of Compensatory Measures:

v Fire Protection Surveillance Testing has been suspended intentionally , pending personnel training in the Fire Brigade Department. NU has implemented long term l

compensatory rneasures because system operability is not being verified through testing. ]

We understand that these measures may be in place as long as until June of this year. l The NRC assumes that Fire Protection compensatory measures will be implemented j when fire protection systems or components are not available due to equipment issues l

which are being addressed. However, compensatory measures are not an appropriate j long-term substitute for system operability.  !

5 Safety Conscious Work Environment Proarams:

l l

We examined the organizational structure supporting the Safety Conscious Work Environment (SCWE), the Employee Concerns (ECP) and the Human Resources (HR) groups, for the effects of recent re-alignments. We understand the reasons for these actions. We recommend that efforts be made to stabilize the your organization in this area to promote the effectiveness of these groups. 9 v The SCWE processes have not yet been formalized. This is the same finding as MC 40001 inspection. This has resulted in some of the management actions being handled personally by the Recovery Managers and other senior managers.

A The handling of issues by the Employee Concerns Program and the Safety Conscious Work Envirunnent program seems to be responsive to specific case needs.

A Performance indicators and Corrective Action program statistics reflect adequate levels of participation.

A SCWE problem area corrective actions for Harassment, intimidation, Retaliation and Discrimination (HIRD) issues are adequate and effective. i a NU is monitoring a number of organizational units and activities which pose the potential of becoming Problem Areas.

i I

4 Monitoring is being done by line management via normalline activities, the Employee Concerns Oversight Panel (ECOP) and during the daily SCWE meetings.

(Check with Don on including ECOP).

n There are no formal action plans, but actions are in place or are planned by management.

v However, programs do not appear to be effective in addressing high incident rate of HIRD issues in ECP system. (Not included on February 20, pending additional information and analysis. Check with Don).

I Corrective Actions:

Our observations are organized along with the process for which corrective actions develop.

Identification and Classification:

4 We observed a low threshold for identification of issues, that is Significance Level 3, Enhancement issues.

4 The Management Review Team (MRT) activities processing of Condition Reports (CRs) was judged to be effective by NRC Team members attending meetings.

v Find out if the RHR Minimum it ; Valve closing was a CR previously, identified and not acted upont v However, there were several examples where the Significance Level had been classified lower than assessed by the NRC team, or the MRT had approved changing Level to a l level that was inappropriately low. Examples of this are:

v Condition Reports (CR's) issued for incomplete action on GL 90-03, Vendor Information Technical Program (VITP), and GL 89-13, Service Water Fouling, l were classified as Significance Level 3, an enhancement item, although they both

! concerned NRC commitments for previous inspection findings. The team also noted that these issues were also included on the Deferred items List (inform Larry Scholl).

v A CR (M3-97-0530) addressing electrical separation issues in of electrical separation in Control Room, Main Control Board, was changed from Significance Level 1, a significant condition edverse to quality, to Level 2, a condition adverse to quality. There was no basis for the change.

v Recently ISEG intervened in Main Transformer switchyard work because equipment was found running unattended, safety questions on equipment lifting practices and because of degraded backup power for Unit 3. However the CR was classified as Significance Level 3, an improvement item, which is inappropriately low. This was not identified or challenged by the ISEG group.

v CRs were written on issues involving the NCR program, the CRs addressing these I

issues were classified as either Significance Level 2 or 3. In handling these CRs after initial processing, the MRT had assigned the Root Cause Analysis to be completed with another CR, and also had combined the Level 1 action items with a Level 3 CR, which was contained only improvement items. The MRT subsequently recognized the error and changed the CR the Significance Level 2.

The Corrective Actions program does not limit the combination of CR issues, nor limit CRs of greater significance being combined into CRs of low significance.

The NRC team considers the absence of controls in this matter to be a program weakness.

v NU has recently begun using an additional risk significance classification for CRs.

However, this classification is not tied to the PRA, Maintenance Rule risx significance or information from the IPE.

Root Cause Analvsis:

A Review of Level 1 CR Root Cause analysis has observed an improving trend from mid-1996 to the present. Significant improvement was noted in analysis performed about a year ago. The additional review and grading program for RCA has added quality. 9 Inacorooriate Corrective Actions:

v Actions to complete the verification of 408 volt molded case circuit breaker overload current magnetic trip settings have been deferred until the end of the next refueling outage, RFO 6, for inaccessible motors. Additionalinformation on the risk significance of those electrical motors is needed for final evaluation of this issue.

Inaooropriate Closure of Condition Report:

v A CR concerning trips of the Boric Acid Transfer Pump was improperly closed as the corrective action, increasing the minimum tank level administratively, is also an Unreviewed Safety Question, because the Technical Specification for minimum tank

' level was less conservative than the new administratively controlled minium level. The

! NRC team also questioned the completeness of the Root Cause analysis because it did not consider possible air or gas intrusion paths, other than tank level. (The Technical Specification is for minimum volume for reactivity purposes, if the system required a higher level for reliable operation, that is volume above the minimum for reactivity purposes).

I v A CR fur the issue of Operating Experience (OE) procedures not being SORC approved, was combined with another CR for the issue of ISEG procedures not being approved.

The issue was closed following approval of OE procedures; however, the ISEG procedures remain unapproved.

• A CR related to procedures, tools and equipment needed to support EOPs, which were not available in the plant. The corrective actions for this CR were not tied to a key event and were outside of the restart at the time that it was approved. This issue was also identified during an NRC review of the deferred items and was then designated as a Mode 2 issue.

l

l 4

v There wa1 a lost opportunity for earlier detection of material deficiencies in High Energy Line Break (HELB) doors because of narrow corrective actions for CRs which a series of CRs HELB door closure issues.

v Although not categorized as a CR, an operator " work-around" that involved service water flow indication, was closed before all actions were complete (Reference OWA 96-003). A CR was issued M3-98-0942.

Significant items List issues SILltem 41-1 concerning the findings of the root cause team for FSAR and licensing basis inaccuracies ACR 7007, and two items related to M3-97-1839 and ACR 13302 is Closed.

In addition the process for UFSAR changes was reviewed and determined to be functioning. However open findings are:

v The Safety Functional Requirements Manual was discovered lacking configuration control. This documents system characteristics for input to the FSAR Chapter 15 analysis. Several revisions were not processed correctly, DCNs were not issued and entered into the Generation Records Tracking System (GRITS). This is a Violation Failure to maintain design control.

v Errors were found in the Annunciator Response Procedures and the Master Set point List (MSL); also the MSL did not contain all necessary information, that is set points.

Problems were identified with two of ten items checked, the Set point for saturation subcooling margin, which was found at 15 F, this apparently is a nonconservatively low Set point because the Safety Parameter Display System Set point is established at 32 F.

The second item concerned the empty data base value for pressurizer level. This is a Violation, Failure to maintain design basis control. Also, a general question of Set point control is open, v Design Basis Summary (DBS) control is deficient. DBS have not yet been updated although they have collected DCNs for the last two years. A Gap Report was constructed. DBS do not exist for two Maintenance Rule (MR) systems (Group 1 & 2).

Charging Volume Control (CVCS), there is a DBS for the Safety injection portion. A DBS for Emergency Lighting was not written when included as a MR system. Also, there is no mechanism to update the DBS when the MR systems change.

v There is a large backlog of drawing changes, approximately 3,000, and plans to work off this backlog extend out two years. This issue was passes to the OSTI to evaluate.

v Or>e Action Request (AR) under CR M3-97-0506 was closed improperly without completing all reviews. The action was to review administrative DCNs for changes that should have received safety evaluations, but did not. The review was to be made for l the past 5 (or 7) years, the action was closed, however only one year was reviewed.

This is a Violation, failure to follow procedures. This is a Restart item, also.

l

1. A few areas of the ACR/CR closeout package were found not to fully address the concerns. The documentation was upgraded satisf actorily during the inspection.

For example, ACR 7007 identified 104 issues,13 were designated Unit 1 only.

However, there was no justification to eliminate some of these 13 which appeared

to be generic to Unit 3. However, all of these issues were found to be overlapped by actions in Unit 3 issues.

SIL ltem 41-2 also tracked as URI 95-81-01, concerned trending of NCRs. This issue is Closed.

SIL ltem 731 also tracked as NOV 96-05-12, concerned failing to have an effective five year Technical Specification audit program. This SIL ltem is Closed; however, the Violation remains Open pending corrective actions regarding the audit planning and tracking matrix.

SIL ltem 73-2 also tracked as IFl 96-06-17, concerning an inadequate Oversight Program based on the 1996 Joint Utilities Management Audit. This issue is Closed.

A The team found that the number of Auditors has significantly increased; their qualifications and knowledge level has increased.

A Audit program procedures are acceptable; audits and audit checklists are acceptable.

There are four new audit managers.

A Oversight has ihr opportunity to concur with the corrective actions taken for audit findings and NCds .

A There is good follow-up on Audit Findings, QA re-audits for an effectiveness review during the next audit.

A There is good interface with the line organization concerning the Nuclear Oversight Restart Verification Plan.

A The NRC team observed initiatives being taken by QC inspector during a hold point inspection.

l # An NRC team member identified a foreign rnaterial exclusion (FME) concern after observing tape in the Unit 2 spent fuel pool FME area which were not noted during previous QC closeout inspections. Although the tape was noted by the NRC team member, the QC inspector was focused on severalissues of higher safety ]

! significance and performed satisfactorily in those areas.

I Safety Committees NRC team members attended meetings of the Site Operations Review Committee (SORC),

I the Unit 3 Plant Operations Review Committee (PORC) and the Nuclear Safety Advisory Board (NSAB) .

A All three safety committees operated effectively, the members were prepared for the meeting and added quality to the issue being addressed.

v The NSAB qualifications matrix did not credit any member as having metallurgy experience, but credited the expertise of an alternate member. The board reassessed their qualifications af ter the NRC team concluded thst the Technical Specification was not met because no member possessed metallurgy experience. Based on this concern,

w the Board re-evaluated its qualifications and concluded that the previous screening qualifications were too conservative. The previous screening required an academic degree, or direct experience to be counted as a metallurgist. However this screening was in excess of the Technical Specification. Based on the new screening criteria, two members were identified with metallurgy experience. Although the NRC team did not contest the actual board members qualifications, the evaluation and initial resolution of the discrepancy did not meet the Technical Specification. This is an example of r>on-conservative interpretation and implementation of the technical specifications.

Independent Safety Engineering Group 4 The ISEG has performed high quality plant reviews and Operational Experience (OE) reviews.

v The number of ISEG reviews done in 1997 was only 12, down from 24 the previous year. This appears to be a result of the OE workload being performed by the grcup.

1. The ISEG group has reduced the backlog of OE issues from several hundred to approximately 40 for Unit 3. However, the amount of tvork represented by the remaining issues is significant. Additionally, several of the remaining issues have a high probability of identifying safety significant issues. One example is NRC IN 97-87, which involves identifying unreviewed safety questions related to the use of manual actions in place of autornatic actions in emergency operating procedures.

The review for this issue was completed for Unit 2 which identified several areas of I

concern. Additionally a previous NRC inspection identified an issue involving manual action for a control room ventilation system that cannot be accomplished I with the safety analysis. Based on these factors the priority of evaluating this issue appears inappropriate. (Mike Brothers added that the IN is being evaluated by Unit 3 personnel, additionally).

1. Site implementation and use of OE was mixed. OE is not consistently being used l

by the working groups at this time. A key reason is that the site wide procedure to establish expectations for the use of OE was not yet issued. Once issued the  ;

departments will still have to develop implementing procedures. An exception is Unit 3 Health Physics which has established access to the nuclear network safety database.

l Self Assessments A The team found that the self assessments were generally of high quality. Continued l attention should be applied to the line departmental self assessments in order to improve  ;

their quality.

1 I

I l

1

l h

l l

l

. CORRECTIVE ACTION l Corrective Actions --CA1:

l l Scope:

I I Assess the adequacy of the corrective action program. This assessment included I

the evaluation of programs for the identification, analysis, and resolution of pla',

f deficiencies.

Findings:

The team conducted discussions with the licensee and performed document reviews of the corrective actions program contained in the recently issued (September 30, I

1997) revision 5. Prior to 1995, the program, employing what was known as the Plant incident Report, was essentially incapable of performing at the standard for the industry. This was a program that captured an average of only 300 items per year for Unit 3 at a high threshold level for events or reportable conditions. This l program was superseded by the site-wide Adverse Condition Report which improved the " capture" threshold to approximately 4,000 items per year. In 1996 and 1997, the licerisee's program was moved closer to industry practices with revisions 2,4, and 5 (revision 3 was never issued). These revisions, among other l changes, resulted in a multi-disciplined management review, as well as the requirement that the shift manager review discovered conditions for operability and l deportability. The revised condition reporting program also strengthened accountability, provided for enhancement items as well as adverse conditions, and defined management expectations.

The team's general conclusions regarding the adequacy of the licensee's corrective I action program were that the program elements concerning identification, CR initiation, and CR processing were performing at a good performance level. The CR program elements concerning root cause, corrective actions, and failure recurrence were considered to be operating at an acceptable level, but with room for improvement. The remaining element, effectiveness, involves trending and self-assessment, as well as effectiveness review. The first two attributes (trending and ,

self assessmer;t) show indication of being performed at an accepta'a le level. While the effectiveness reviews, by their nature of being performed at a later date, cannot yet be fully evaluated due to the newness of the current program requirements.

The team evaluated the performance of the licensee's program for identification of adverse conditions. As indicated by the range of CRs reviewed, and discussions l with a tiroad spectrum range of plant personnel, the team agreed that the licensee

! has attained a low threshold for initiating Condition Reports. The fourth quarter continued the high volume trend of CRs initiated for Unit 3, with a total of 1,621 CRs written during the fourth quarter. An increased awareness by plant staff and a decreased threshold for initiating a CR, appear to be the causal factors for the high vobme of Cas initiated. An additional reason for the high volume of CRs is the l lCAVP (Independent Corrective Action Verification Program). I

'/  ;

} , .

i

r I  !

I l'

It was also the team view that the licensee's analysis of reported conditior.s (CRs), j while generally adequate, tended to be narrowly focused. Several examples to l support this are discussed below. Among those examples is the resolution of l l

closure problems with High Energy Line Break (HELB) doors.

l During discussions with the licensee it was learned that the Independent Safety l Engineering Group (ISEG) had not been involved in any reviews of the CR program, l its status, or the impact of the backlog of hi]h priority CRs. This determination

! further reinforces the team view that ISEG may need to rebalance its workload and reemphasize a focus on plant and management contrcl issues, i

Conclusions:

The team evaluation of the overall program is that it is performing at an acceptable level; however a large number of issues were raised from a relatively small sample of CRs reviewed. This may indicate that further licensee rnanagement attention is ,

warranted.

l l

The licensee has attained a low threshold for initiating Condition Reports. An {

increased awareness by licensee personnel and a decreased threshold for initiating a j CR have resulted in a high volume of CRs initiated. 1 ISEG may need to rebalance its workload and reemphasize a fccus on plant and management control issues in order to appropriately implement their oversight functions.

[ Corrective Actions - CA2:

This plan item involved a review of the effectiveness of the NU programs for corrective action taken in response to a sample of employee concerns issues.

During a team status meeting we were informed by Tom that another group was looking at this area and we not were responsible for inspecting it.)

Corrective Actions --CA3:

Scope:

The team reviewed a sample of root cause analysis and equipment f ailure evaluations to determine the adequacy of the process. For less significant issues the team reviewed a sampling of the apparent cause determinations. The team also independently verified the significance of condition reports for significance and that apparent root cause determinations had been performed where required.

Findings:

For the Level 2 and Level 3 CRs reviewed, the team did not disagree with the significance nor the apparent root causes. Although generally adequate, the inspectors review of Level 1 CRs disclosed a tendency for root cause analysis to be

narrowly focused and several cases where the root cause was waived. Although  !

exercising waivers are within procedural constraints, their use can also lead to missed opportunities for case-specific, in-depth evaluation. In the case of CR M3-97-0652, which described how design interface distribution and transmittal control of design information did not meet the requirements of Criterion lli and V of Appendix B, the root cause analysis was waived by the MRT with no accompanying explanation. The corrective actions involved training U-3 design engineers on DCR/MMOD requirements and emphasized the need for attention to detail. As determined by the team the corrective actions appeared to be narrowly confined to l the items associated with the DCM deficiencies.

(

l For ACR M3-97-0558 (issued on February 20,1997), the licensee's DBDP (Tom fill in: Design Basis Document ...??) document for the Chemical and Volume Control System had non-conservative assumptions related to maximum temperatures for the l'

letdown heat exchanger and charging flow. The root cause was waived and referred to other CRs with related similar situations. The inspectors regard this l waiver as a missed opportunity to thoroughly evaluate the issue. However, the corrective actions appear to adequately address the design deficiency.

Another instance invc!ved ACR M3-97-0409, a Level B ACR (issued on February 4, 1997), that documented concerns for sump water level calculated head losses.

Although the cause of the event was addressed in an LER (LER 97-015),no root cause was performed for the CR. Nevertheless, the proposed corrective actions appear adequate to resolve identified design deficiencies and a modification review l is scheduled following reanalysis with actions to be completed prior to mode

change.

Conclusions:

The team's review of CRs disclosed a tendency for the root cause analysis to be narrowly focused and identified several cates where the root cause was waived.

l Notwithstanding the procedural legality of waiving the analysis, in light of the narrow focus of corrective actions, the team determined that this represented missed opportunities to thoroughly evaluate the issue.

Corrective Actions --CA4 & CA5:

Scope:

The team evaluated the technical resolution of a sample of safety significant issues for timeliness and effectiveness of corrective actions as well as the backlog of open condition reports to verify that safety significant items were being tracked to completion. The teams evaluations also included Interviews with supervisors regarding the closure rate and reviewing the process to prioritize corrective actions based on risk.

Findings:

4 During the inspection preparation trip in December 1997, the team questioned the apparently low closure rate associated with the i;ceasee-defined, high-priority Level 1 CRs. At that time, a review of the approximately 260 items classified as Level 1, j and by the licensee's definition, a "Significant Condition Adverse to Quality," l showed only 25% of the items to be closed. Subsequently, various supervisors, including the root cause supervisor and several QA supervisors, were interviewed regarding the CR process in general and the apparent low closure rate in particular.

From these discussions and a review of the current status of the Unit 3 CR backlog (as of February 16,1998), the following information was obtained:

Section 1.8.5 of the licensee's procedure, RP-4, Corrective Action Program, states that evaluation due dates of 30 days from determination of assignment duties are established. The number of overdue evaluations in January 1997 was 825 items.

This number had been reduced to ris low as eight items by July of 1997. Since November of 1997 to the present, the average backlog number has been about 20-30 items. This reflects a partial recovery of the management of this portion of the process. A rery high volume of CRs have been and currently are belng initiated; as the volume af CRs being generated moderates,it is anticipated that this average backlog number will be further reduced.

Good efforts have also been made in the reduction of restart-related CRs. Starting in the same period, January 1997, restart CRs were at 575 items. Due to the i licensee's emphasis and resource focus applied to the reduction of the "over due evaluation" backlog, these CRs increased to more than sixteen hundred items ,

during the period when the evaluations were being driven down. Starting from July i 1997, with the " freeing up" of resources from the efforts to reduce the "over days evaluations," the licensee was able to drive down the restart CRs to less than 400 items currently.

The licensee's procedure RP-4 requires CRs to be assigned to one of three significance levels; Significant Adverse Condition (Level 1), Adverse Condition (Level 2), and improvement item (Level 3). The licensee has in use a risk I

significance classification system that further subdivides risk into four categories. I Snncifically, Attachment 4, Risk Significance, to RP-4 states that in addition to a du rmination of whether a condition adverse to quality is significant, further l attnbution is warranted to assist the CR Owner and MDMRT in prioritizing the corrective actions. The four risk significant categories are defined as: Priority 1, Risk Critical-This category would result in consequences that are severe and unaccepta'ble in either human, societal, political, or monetary terms; Priority 2, important to Safety - This category can result in risk to the reactor systems, industrial safety, public health and safety, or the environment; Priority 3, Compliance - This category corrects common and repetitive non-compliance with j laws, procedures, or normally accepted standards and expectations as defined by l station standards and regulatory agencies; Priority 4, Good Management - this j j category identifies an isolated condition adverse to quality or an opportunity for I program enhancement.

Observations by the team of the MDMRT members review and evaluation of CRs showed that both ti" vel categorization and the risk-significance priority classification were gned to CRs generated each day. However, discussions,

~

l 1

with the Corrective Actions Department and others, revealed that this last classification has not been acted upon beyond the assignment of the classification by the MDMRT. The licensee envisions further implementation of this concept in the future. In that this concept is not a "hard" requirement by the procedure, nor by regulatory requirement, this is not a violation. However, at this stage of  !

implementation, it is ineffective in accomplishing its intent and may create a wrong I mind set regarding the spirit, as well as the rule of piccedural compliance. It was also noted by the team that the nomenclature has the potential to create program l

confusion in that the title " risk significance" is already a defined term with respect to the maintenance rule.

Also, the inspectors did not discover any evidence, through review of RP-4 or )

interviews, that the licensee was availing themselves of information from their IPE, )

nor from maintenance rule risk rankings, to assign CR classifications. Additionally, there was no evidence that risk-information was utilized for approving extensions to action due dates.

Conclusions:

i Efforts have been made to reduce the number of CR evaluations open over 30-days.

This reflects a partial recovery of the management of this portion of the process.

Good efforts have also been made in the reduction of restart-related CRs. Progress on closing Level 1 CRs has accelerated since late last year.

The CR risk significance classification process is incomplete. At the current state of implementation, it is ineffective in accomplishing its intent. It was also noted by the team that there is the potential to create program confusion in that the term " risk significance" is already in widespread use in the industry for an entirely different program: the maintenance rule. Furthermore, there was no evidence that the licensee was using risk information in the prioritization of issues in the conective action process.

Condition Reports:

Overall, the team considered corrective actions to be narrowly focused. Examples were identified where the timeliness of corrective actions has been hampered by narrowly focused corrective actions. One example of a narrowly focused resolution of plant deficiencies involved the corrective actions for unclosed HELB (High Energy Line Break) doors, and a missed opportunity for timely and necessary corrective actions. During January 1997, six Level 1 CRs were written on problems involving HELB doors and the failure to close them properly. Corrective actions focused on correcting personnel errors as the resolution to the recurring problem of unclosed doors. Subsequently, in August of 1997, CR M3-97-2567 was written. This CR j addressed and described the physical deterioration of the doors and stated that problems ranged from missing gasket sections, damaged thresholds, and damaged gasket bars, to large gaps around the gasket seals.

The origination of this CR (M3-97-2567)was not driven by the HELB door closure problems but rather by a different review (CMP PI 21, " Engineering Topical Areas

Reviews"). This later CR pointed out the lack of a PM program, the lack of regular inspections, and the lack of inspection criteria.

An effective corrective actions program would expand the scope of corrective actions to include investigations for, and evaluations of, related problems (and also to related systems, equipment, procedures, personnel actions, and applicability at other site units); this would be especially true for repetitive problems.

Thus, the expectation was that the large number of CRs written in the January time frame would have instigated a broader look at the problem by management, and the discovery of these degraded physical conditions in a more timely manner. It is problematic whether the physical conditions and lack of a PM program would have been identified by the corrective action program alone without the " Engineering Topical Reviews" process.

This was indicative of a narrowly focused corrective action addressing personnel issues and a missed opportunity for the corrective action process to detect existent problems.

1 l

i L_______________________________...___

i SELFASSESSMENT Inspection Scope The stated purpose of the licensee's self-assessment program is to identify areas of I concern and to improvo performance utilizing the established corrective action program. Specifically, the programmatic controls described in unit procedure U3 l CA 11, "Self-Astesment", Revision 1, direct the performance of pre-planned  ;

department self assessments, by qualified individuals, with the objective of j achieving higher standards of quality and performance at Millstone Unit 3. This l procedure further states that the primary responsibility for the performance of self-assessment activities resides within the line organizations, including identification '

and resolution of deficiencies.

Observations and Findings in order to evaluate the effectiveness of the self-assessment program the team reviewed a selected sample of (20 out of approximately 110) recently completed U3 self-assessments. These self-assessments included a cross section of U3 department evaluations as well as follow-up reviews which were performed during 1997 and early 1998. As a result of these reviews the team determined that there was a significant variance in the quality and technical depth of these self-assessments depending on (1) the time frame in which the assessment was performed and (2) the organization performing the activity. In particular the self-assessments which were performed earlier in 1997 tended to be narrowly focused with limited corrective actions specified. Examples of this category of self-assessments included the following reports:

• 3MNTC-SA-97-17, QC Hold point Requirements, February 5,1997

• 3MNTC-SA-97-01,in-Field Procedure Compliance, May 7,1997

• 3 CAD-SA-97-03, Root Cause Evaluation Quality, April 30,1997 However, as noted by the team, the technical adequacy of self-assessments tended to improve in the latter half of 1997 and into the first quarter of 1998. Self-assessment reports which reflected this improving trend included the following examples:

• 3MNTC-SA-97-07, Health Physics Department, June 24,1997

• 3TS-SA-97-13, Functional Requirements in Safety Systems Preoperational Testing, September 30,1997 l

• 3 CAD-SA-97-14, Assessment of MP3 and '

iel 1 Condition i

Reports, January 22,1998 1

i l

l l 1

As determined by the team the overall improvement in the quality of U3 self-assessments was atuibutable to the establishment of definitive management expectations regarding the need for performance improvement, an emphasis on self-assessment training and enhanced procedural controls. Additionally, at the time of the inspection the licensee was in the process of implementing a self-assessment program that incorporated both station and unit specific elements. Relative to this issue the team ascertained that the station unit and support organizations have developed departmental self-assessment plans with each major support group performing formal assessments using a common approach. Accordingly, department self-assessments consist of an ir.-depth evaluation of significant line and staff activities, which are performed by teams of knowledgeable individuals in l accordance with defined assessment plans. The assessment results are used to I identify program strengths, findings and areas for improvement. The licensee's program has also been expanded to evaluate the effectiveness of the self-assessments in order to improve the quality and consistency of these activities.

The team also evaluated the effectiveness of NU's program to identify and correct  !

operator " work-arounds." The purpose of the work-around program is to identify )

and assess equipment deficiencies that adversely affect plant operations. These deficiencies are characterized as items that may degrade the operator's ability to react to plara transients. The inspection effort involved the review of the procedural 7trols contained in OP 3260E," Program for Resolution of Operator Work-Arourd s", Revision 0, examination of Self-Assessment 30PS-SA-97-04,

" Effectiveness. of The Operator Workaround Program" and the evaluation of recently completed modification packages related to operator work-arounds.

The inspection team determined that discrepancies associated with operatcr work-arounds are not documented on CR's nor are they required to be tracked on the licensee's AITTS system, but are administered under a separate program. In particular, procedure OP 3260E, directs that operator work arounds, initiaily identified in the shift turnover log, be screened for their cumulative impact, including operability and deportability considerations. This screening process is performed by the Unit Supervisor, the Shift Manager and finally by the Operations Manager, ,

Based on this screening process, the items that are classified as operator burdens or '

work-arounds are separated from the category of non-conforming conditions which l would require corrective actions in accordance with the CR program. Additionally, these items which are tracked using the Operations Performance Database, are prioritized and penodically reviewed by the operations department, and they are reviewed on a biweekly basis by unit management personnel.

Subsequent to reviewing the completed modification work packages for six operator work-arounds, the team performed a system walk-down in order to confirm the implementation of equipment modifications and the adequacy of the completed l

work activities. Based on the team's reviews, and the results of system walk-i downs, it was determined that the equipment modifications associated with five of the operator work-arounds had been appropriately completed or were awaiting testing following component re-work. However, during the review of the documentation related to operator work-around Number 96-03,(correction of flow indication anomalies on service weter instrumentation 3SWP-Fi-059 A, B, and C),

the team determined that the Trouble Report (TP) tags had been removed from the l

o l

. l l 1 l

flow indicators and that the automated work order (AWO) associated with this modification had been inappropriately closed. prior to the completion of all specified work, Specifically, the final setpoint calibrati >ns for flow indicators 3SWP-FI-059 i i A, B and C had not been accomplished prior te removing the TR tags and closing the AWO, which is contrary to the work control requirements of Procedure U3 WC1, " Unit 3 Work Management," Revision 1, Section 1.8.7.

1 Following the identification of this issue, the licensee initiated CR M3-98-0942,in accordance with procedure RP-4, to document this deficiency and to effect l resolution.

Two contributing factors related to the above noted violation involved the (1) the lack of formal training for operations personnd on procedure OP 3260E and (2) the fact that Action Requests (A/Rs) are not initiated to track the status of each operator work-around. These specific issues had been previously identified in Self-Assessment Report 3 OPS-SA-97-04 and although a CR (M3-97-3632) had been initiated on October 21,1997 to address the assignment of A/Rs to track discrete l operator work-around items, the corrective actions for this item were not scheduled l for completion until September 30,1998. Given that both of these contributing factors had been previously identified in a self-assessment report, this missed opportunity to avert an item of noncompliance is identified as a weakness within the self-assessment program.

l Conclusions in order to determine if the self-assessment program was being adequately implemented, the team reviewed a selected sample of recently completed self-assessment reports. Based on this review, it was determined that the self-assessment program had established appropriate administrative controls which provided for the tracking of information to detect declining performance and adverse trends. The team concluded that, in general, the self-assessment program was being adequately implemented and that the associated recommendations were beneficial in identifying areas for enhancement and improved performance.

During the inspection the team identified one item of non-compliance related to the implementation of NU's program to document and 60rrect operator work-arounds.

Specifically, this issue involved the premature closure of the AWO associated with the reconciliation of operator work-around 96-03, prior to the completion of the required re-calibration of Service Water instruments 3SWP-F1-059 A, B and C. i Subsequent to tho identification of this example of failure to follow procedures, the licensee initiated CR M3-98-0942 to document the condition.

l l

Inspection Report inout for Jim Hiaains Portion of MP3 40500 Inspection i

1. Inspection Area: Self-Assessment - SA-5 l

a. Insnection Scope l

l Evaluate NU's actions taken in response to ACR-7007- Event Response Team. This issue l primarily concerned the inaccuracies in the Updated Final Safety Analysis Report (FSAR) and design basis documents. It is designated by NRC as SIL ltem 41.

l

b. Observations and Findinos )

l 4 Backoround:

ACR 7007 was written by the licensee in January,1996 to address the high level concern

! that "The UFSAR, syrtem descriptions and design basis documents contain inaccuracies."

! An Event Response Team was chartered to determine the causes of these inaccuracies. l This team used root cause ant.tysis methods to identify causes and contributing factors.

This team developed the ACR 7007-Event Response Team Report, dated fwnary 22, 1996, that identified conclusions, corrective actions, and comments in Sectio 5,6, & 7 of the report. It also included a significant amount of backup material that ler >these conclusions. The report presented a thorough analysis of the problems and a o isonable set of conclusions and corrective actions. The identified causes and contributing factors were very broad and far reaching, necessitating extensive corrective actions. These were further elaborated upon by the licensee as described below.

In July,1996 the Fundamental Cause Assessment Tearn (FCAT) and the Nuclear Committee Assessment Team (NCAT) determined and reported to the NU Board of Trustees that the root cause of decline in Millstone performance was that senior executives at NU from the CEO to senior nuclear site executives were ineffective over a number of years in providing vision, direction, and leadership necessary for the management of the NU nuclear power program.

Over the last two years the licensee has embarked on a number of broad programs and changes to address these concerns and improve performance. These have included the configuration management program (CMP) and major changes in organization and management at tiie Millstone site, in order to address the specifics of ACR 7007, the licensee extracted all identified issues from the Event Response Team Report. These were compiled into a main item list of 104 issues, dated Nov. 5,1996. In June,1997, Unit 3 issued CR M3-97-1839to address the Unit 3 aspects of ACR 7007 and the Event Response Team Report. This document sorts the 104 issues into five major areas in order to track and address them effectively. The

' l I

3 n* i - .

o five areas are: leadership, self-assessment, corrective actions, configuration control, and oversight. These categories are the same that have been identified by the corrective actions department. Additionally, they are the same as five of the eleven key site-wide issues for restart that are being tracked by NU and that are addressed in the periodic letter to the NRC titled, Progress Towc;J Restart Readiness at Millstone Station (NRC Briefing Book).

In order the provide more detailed tracking and specification of corrective actions, subcategories were established below the five categories. All of the 104 issues were then assigned to the categories and subcategories. [However, three of the 104 issues were noted to have no action required, and 15 were noted to be applicable to MP1 only.] A succinct high levelissue was developed for each of the subcategories, and corrective actions were defined and approved to address each high levelissue. Action request (AR) numbers were assigned for all corrective actions. An important point to note is that this ACR and the related corrective actions addressed primarily the configuration control aspects of the five major areas and did not try to fully address other aspects of leadership, self-assessment, corrective actions and oversight. There will still be other NRC assessments, such as the ICAVP assessment and NRC Senior Management Assessments, of these five issues and the other remaining Key issues.

As part of the CMP program, under Pl 2, " Unit Specific Assessments," Unit 3 performed a self~ assessment to determine any other areas similar to those in ACR 7007 that needed to be addressed. ACR 13302 was written to address the findings from that assessment.

ACR 13302 was included as part of the ACR 7007 package and was also reviewed for this inspection.

General Discussion:

The inspector reviewed the lists, discussed above, and verified that the issues had been appropriately extractett from the Event Response Team Report and that each of the 104 issues had been assigned to at least one of the categories / subcategories. The decision of

'no action required

• on three issues was judged appropriate. For the 15 issues, designated as applicable only to MP1, the documentation was not clear as to why they were not also applicable to MP3. The licensee acknowledged this, performed further review of these items, and documented the review in Memo PES-98-055, Clarification for Corrective Action Plan for CR M3 97-1839, Feb.1: ,1998. This memo provided justification for the 13 of the 15 Unit 1 items. The other two were determined to be generic to all units, however, it was further determined that the Unit 3 corrective actions, al ready completed for ACR 7007, encompassed these two items as well.

The inspector also reviewed each subcategory and verified that the issues (assigned to the l subcategory's defined high levelissue) and the specified corrective actions were l appropriate. A few areas were noted, requiring further justification, and questions were passed on to the licensee. With the exception of those noted below, all were satisfactorily resolved.

The inspector also reviewed, on a sampling basis, the documentation provided with the CR for the completion of the defined corrective actions. Some of the reviewed areas were 2

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ _ - - _ _ _ _ I

then selected for further in depth followup. AdditionaHy, the inspector performed i independent review in the general area of the high level concerns to verify the thoroughness of the corrective actions. Some areas addressed in more detailincluded:

the FSAR update process, setpoint control, an independent contractor audit of the Design Control Manual, and the Design Basis Summaries.

The inspector also verified that all actions for each of the three CRs were appropriately closed or that remaining open items are not significant and scheduled for closure on an acceptdsle time frame. Some areas designated for completion post startup were  ;

questioned. The licensee provided documentation that the work had been completed already or justification that its deferral was appropriate, except as noted below.

Additionally, the inspector reviewed an internal audit of the corrective actions associated with ACR 7007, "The Independent Review Team Report on the Effectiveness of I Correctiveness Actions Associated with ACR 7007, Rev.1,6/17/97," and noted that the findings were addressed by extensive actions taken by the licensee on ACR 7007 since the l audit.

Specific Areas and issues:

l

1. FSAR Changes The inspector discussed the area of FSAR change identification and control with licensee representatives and reviewed procedure RAC 03, Rev. O, Changes and Revisions to Final Safety Analysis Reports. The inspector also reviewed: selected FSAR Change Requests, I the comput! r tracking system for FSAR changes, current related to the FSAR changes in process, the process for initiation of a FSAR change, the review and approval process, the facilitatory / owner of the FSAR and specific FSAR sections, the tracking of in-process changes, and the overall scheme for FSAR updates during the current outage. The licensee has processed a large number of changes (e.g., 596 in 1997) during the current outage due to CMP issue identification (called an updating FSAR change) and due to the large number of modifications being completed (called an upfront FSAR change).

Approvals for the upfront changes are processed together with the modification, ensuring that appropriate reviews are given and that the FSAR is kept current. This large number of changes has resulted in several update submittels to the NRC during the past year and has measurably improved the content of the FSAR.

RAC 03, Attachment 7 contains a listing of the manager of the primary responsible discipline for every section of the FSAR, thus facilitating proper revew of change requests.

The procedure also has mechanisms for ensuring that in-process changes are consistent with each other. Time limits are set and, if processing time frames of the procedure are i not met, then a CR is issued. The inspector reviewed the licensee's tracking and trending l

data. The area of FSAR change control was judged acceptable.

l l

2. DCM Review One of the issues associated with ACR 7007 related to design control and the Design Control Manual (DCM). As a result of this, one of several actions was to contract MDM l Services Corp, to perform an independent review of the DCM. MDM issued the report, 3

" Strategic Overview of Millstone Configuration Control Processes by MDM Services Corporation, Final Report," July 15,1997. In this report were a number of recommendations, which the licensee addressed in Memo PES-97-412, dated Dec. 31, 1997. However, a number of the recommendations were rejected with no justification.

the licensee revisited the area and provided a new evaluation that adequately addressed all recommendations or provided justification for not addressing them.

3. NRC Commitments The area of " licensee commitments to NRC" was addressed in ACR 7007. The inspector also noted that Level 1 CR M3-97-1759," Trend identified in the area of NRC commitments," was issued in early 1997 to address then current problems in this area.

The licensee placed considerable effort into the identification of past regulatory commitments made to the NRC, in docketed correspondence, through the implementation of procedure PI 6, Licensing Reviews. Ongoing con +rols for commitments were established in several procedures, including DC 18, "NRC Communication," U3RP10 Outgoing Regulatory Correspondence Processing and Validation," and RAC 06, " Regulatory Commitment Management Program." The inspector reviewed the list of CRs generated as a result of the Pl 6 reviews and sampled the actions taken to resolve them. No problems were identified.

4. Updating of Drawings The inspector discussed the area of drawing control and updating with the MP3 design engineering personnel. The updating process is contro%d by the DCM, Chapter 7, and by EDI 30250. These documents define categories of importance for incorporation of changes into the more important drawings more quickly. Category 1 or Operations Critical drawings requira incorporation within 30 days of the DCN being released. Category 2A drawings have a 90 day guideline for incorporating outstanding changes, however, there is currently a backlog of about 3000 drawings. The licensee had no cancrete plan for eliminating this backlog. In response to inspector questions, the licensee issued Memo M3-DE-98-0090that forecast a plan to work off the backlog over a two year period. This i area was passed on to the OSTI team for further evaluation.

5. Design Change ControlIssues issue 4, and the similar issues 45 and 59, were not specifically addressed in the ACR 7007 close out package. These issues related to a potential weakness in programs that may l have allowed drawing changes without generating a PDCR or a DCR. The licensee reviewed this area and prepared a documentation package to demonstrate that other controls existed to prevent such an occurrenu. Further, the CMP utilized PI-29, Unit 3 P&lD Walkdowns, to review P&lDs for discrepancies, and then evaluate and correct them.

One closely related concern, documented in ACR 7007, was that the process for controlling drawings had been weak , allowing drawings to be changed without a DCN or DCR and not ensuring that other design documents were updated. As a followup to this concern, an Engineering Self-Assessment Report (3-ESAR-97-001)was performed to assess Category 8 Administrative DCNs for this general problem. The ESAR found that )

l 4

l l -----._--__-.--------_---------------U

certain DCNs exceeded allowable criteria in that an MMOD or DCR may have been required to properly approve and document the change in question on the administrative DCN. ACR M3-97-0506 was written to take corrective actions on this specific finding. The inspector reviewed the CAP anr d Leumented close out actions to address the CAP. The inspector noted that not all of the required actions were clearly documented as being complete. The inspector requested additional documentation from the licensee to show that the actions had been completed. The licensee's review then found that all of the actions for one assignment, AR 97003960-05,had in fact not been completed, even though the AR had been closed. This is contrary to RP 4, Step 1.12.4, which states that actions are to be closed out "WHEN assignment is complete." The two actions that did not appear to be completed nue: performing an MSEE reconciliation of 199 DCNs that document as-built condit!.ms, and reviewing the 142 DCNs that initiated work and then generating an MSEE, Mi.iOD, or DCR as appropriate.

As a result, the licensee issued CR M3-98-0921, which documents the incorrect closure.

AR 97003960-05 specifies four actions.

6. Setpoint control The inspector evaluated the design control area of setpoint control. The licensee does not maintain all of the setpoints in one common database or area. However, Specification SP-ST-EE-329, Use and Control of Master Setpoint index, Rev. 2,10/15/97, provides a roadmap of the various documents, control mechanisms, and databases involved. Change control for the setpoints is typically via the DCM. NGP 5.23, Plant Design Data System (PDDS) Data Packages states that the Master Setpoint List provides the setpoint calculation number and process and instrumentation setpoints for components designated by an instrument mark number. In order to evaluate this area, the inspector selected ten annunciator windows and their respective setpoints, as noted in the corresponding alarm response procedure (ARP). The inspector requested the licensee to provide the design information for each of the ten instruments involved, that would include the setpoint value and the basis for the setpoint (e.g., the calculation).

Based on the licensee's response, the inspector noted that the information for the ten selected setpoints was not contained in one consistent location but rather was in the ,

Master Setpoint List, the Technical Specifications, the Westinghouse Precaution, i Limitations and Setpoints (PLS) document (WCAP-10072), a Surveillance Procedure, and a calculation. Further, one of the ARPs was found to contain an incorrect instrument as the l initiating device for the alarm. The licensee wrote CR M3-98-0805,"Pzr Level alarm j initiating device incorrect in ARP," to address this finding. Also, for one of the alarms, the 1 Saturation Trouble alarm for the Inadequate Core Cooling System, the Master Setpoint List referenced an incorrect calculation. The licensee was unable to find a calculation that provided the basis for the setpoint of 15 degrees. Upon further exploration, the licensee determined that the setpoint should have been changed to 32 degrees (which would agree with a similar alarm in the Safety Parameter Display System - SPDS), as part of PDCR M3-93-121 (and as recommended by Memo NE-93-SAB-263 dated 6/14/93). This 32 degree setpoint is derived in the Millstone 3 Emergency Operating Procedure (EOP) Setpoint Documentation, Calculation # W3-517-981-RE, Rev. 6, dated 9/17/97. The licensee  !

issued, CR M3-98-0935 to address this finding. {[ Tom: Could cite this against - 10 CFR l 5

t

50, App. B, Criterion til that states b part that, Measure shall be established to assure that the design basis is correctly translated into specifications, drawings, and procedures.]}

The Saturation Trouble Alarm in question is described in the FSAR, Section 4.4.6.5, instrumentation for Detection of inadequate Core Cooling, as backups to the primary subcooled/superheat display on SPDS.

The ins,pector noted that the methods for control and documentation of setpoint information appeared inconsistent, difficult to retrieve at times, and had the potential for allowing incorrect information to persist. In addition, the licensee performed an Engineering Self-Assessment of the Setpoint Control Topical Area,3-ESAR-97-015, dated 5/2/97 and this seif-assessment also noted weaknesses in the Setpoint Control Program that have not been addressed to date.

7. Design Basis Summaries (DBSs)

As part of the changes to the design control process made in addressing ACR 7007, the licensee eliminated their Design Basis Documentation Packages (DBDPs) and System Descriptions and established Design Basis Summaries (DBSs). These were created per procedure U3 Pl 29, Development of Millstone Unit 3 Design Bases Summary Documents.

U3 PI 29 states that the objectives of the DBSs, include providing a documented reference:

for use in the design process for future modifications, to support technical reviews and safety evaluations, to support operability evaluations and determinations for continued operations, and to support review of Technical Specification changes and FSAR changes.

It further states that the DBSs are integral to the Unit 3 restart program and are prepared for the Unit 3 Maintenance Rule (MR) Group 1 and 2 systems. The inspector reviewed the set of DBSs against the list of MR systems and reviewed portions of selected DBSs. The inspector also reviewed methods for updating the DBSs, which are contained in the Design Control Manual, Chapter 11, Attachment 5, instructions for Controlling and Revising Design Basis Summaries, Rev. 6, Change 2.

The inspector noted that there was no DBS for the Emergency Lighting System that was moved from MR Group 3 to Group 2 during the summer of 1997 after the original DBS list was developed. The inspector also questioned how the licensee would ensure that any changes to the list of MR systems was reflected into the DBSs. The licensee initiated CR M3-98-0892 to address this issue. The inspector also questioned the lack of coverage for l the full Chemical & Volume Control System (CVCS) in a DBS since this was a MR Group 1

/ Group 2 system. Portions of the CVCS were included in the Emergency Core Cooling l Cystem (ECCS) DBS, but much of the CVCS system was not included in any DBS.

The inspector noted that the original DBSs were prepared based on design information that had a freeze date of June,1996. The DCM update controls began sometime after that resulting in a gap of coverage of about one year. This was identified in Memo MP3-DE 1616 and is being tracked by Design Engineering, CRs, and by Oversight. While not yet updated, the gaps are clearly identified and are being satisfactorily tracked.

The inspector also questioned the overall control of DBSs, since the Pl procedures are being phased out and the DCM only addresses a revision process. The licensee presented the Millstone NPS, Programs and Engineering Standards, Configuration Management Plan, 6

l

Revision 1,9/23/97 which addressed the transition process from the CMP and the PI procedures to a permanent organization. This document is to ensure no gaps in processes and that the going forward products are clearly identified.

8.- SFR manual The MP-3 Safety Functional Requirements (SFR) manual was developed to identify the key system level requirements that are refl.ected in the safety analysis. This provides design input and assumption information, primarily for the NSSS equipment. The actual plant NSSS calculations are proprietary and are maintained by Westinghouse. The SFR was developed by NU and was reviewed and commented on by Westinghouse. Chapter 2 of the manual addresses systems, Chapter 3 addresses the safety analyses of the FSAR -

Chapter 15 analyses, and Chapter 4 addresses three programs (fire protection, SBO, and refety grade cold shutdown. The manua appears to provide a valuable tool for the plant j design personnel.

The SFR manual, (MP Unit 3, Design Basis Documentation Package, Safety Functional Requirements, DBDP-MP3-SFR)is controlled by procedure NGP 5.'28, Design Basis Documentation Packages, Rev. 3,10/15/97. Step 1.1.2 of NGP 5.28 states that if changes are required they must be documented as DCNs and the DCN numbers entered into GRITS (Generation Records information and Tracking System). GRITS is the on-line interactive database system that provides the enrrent design and revision status for the Millstone facility. The original Revision O of the SFR was issued 12/30/94. Rev.1 addressed Westinghouse review comments was issued 12/11/96. Rev. 2 was issued on 11/20/97 to incorporate CMP changes. However, Revisions 1 and 2 were issued with Engineering Record Correspondence per NGP 5.31 rather titan NGP 6.28, and as a result no DCNs were issued and GRITS was not updated. As of February,1998, GRITS still showed Rev. O as the latest version. This constitutes a failure to follow procedures. The licensee issued CR M3-98-0861 to address this problem. The inspector questioned which version had been supplied to the ICAVP contractor, and the licensee stated that the current ]

Rev. 2 had been supplied.

11. Inspection Area: Corrective Actions - CA-4

a. Inspection Scope Select a sample of CRs, that have been generated by the licensee based on NRC inspection findings over the last several months at Millstor'e. Review the resolution of these CRs to evaluate both timeliness and effectiveness.

l-

b. Observations and Findinas

( The inspector selected a sample of 19 CRs that had been initiated over the general time frame from August to December,1997 (with a couple from 1996). These were reviewed for timeliness, appropriateness, and effectiveness of corrective actions. The inspector noted that the CR process generally appeared effective and serving its purpose. Several 7

t l

1

areas were noted for further followup and are discussed below.

1. Many of the CRs had all of the Action Requests (ARs) completed but the overall CR was still open. The licensee stated that due to the high volume of CRs, as the CMP program is coming to a conclusion, there is presently a backlog of 600 to 700 CRs of this type in the system awaiting closure by the Corrective Actions Department.

2. In the computerized on-line CR system, all of the pertinent .ARs do not appear as associated with their particular CR. The licensee stated that this is an historical problem that is in the process of being corrected but will still take more time to backfit the appropriate links.

3. There were noted to be some gaps in the tracking of Non-Conformance Reports (NCRs) and the ARs associated with NCRs. The licensee was aware of this problem and is in the process of addressing it (reference Memo MP3-CAD 98-004). (Tom:

since so many other people on the team were following this and writing about it, I did not pursue it any further.)

4. There appeared to be some tendency to schedule some required corrective actions too far.into the future and to delay or postpone corrective actions.

5. The inspector noted four CRs that involved f ailure to conform to NRC commitments made in Generic Letters (GLs) and that were apparently misclassified as a Level 3 versus a Level 2 CR and/or contained ARs that were deferred post-startup. RP 4, Rev. 5, Attachment 3, CR Initiation and Classification Guidelines, includes in the i Level 2 guidelines as: an external station c.ommitment not adhered to; or a deficiency in material that, if left uncorrected, could affect safe reliable plant {

operation. The specifics were as follows.

CR M1-97-1914is a level 1 CR (related to GL 90-03 and appUcable to all three Units). Corrective action #1 (AR 97020123-02)of this CR was inappropriately classified as deferrable post-startup; however, the portions of this AR that related to compliance with the commitment were completed prior to startup in December,

, 1997. CR M3-97-4672is related to GL 89-13 and was inappropriately classified as l

Level 3. Further, none of the actions were coded as needing to be completed prior to startup. At the time of the inspection allitems but one were noted as complete.

The one remaining item appears that it should be completed prior to startup. CR M3-97-4346 contains a deficiency in material (inadequate corrosion control) that, if left uncorrected, could affect safe reliable plant operation. It is related to GL 89-13 and was inappropriately classified as Level 3. Some of the corrective acticns appear that they should be completed before startup. CR M3-97 3501 is also related to GL 8913 and documents an adverse condition. It is uassified as a Level 2, however the corrective actions were not tied to startup. Scme of the corrective  !

actions appear that they should be completed before startup. j l

The licensee issued CR M3-98-0933 to address these findings and performed a preliminary review to determine the cause. They noted several breakdowns in their program related to corrective actions for NRC commitments and beoan actions to j 8

i

i

., l both correct current instances and prevent further such occurrences. On March 9, )

1998 the licensee reclassified CR M3-97-4346 and CR M3-97-4672 as level 2 CRs. j l

1 I

l i

9

o

'4 1

• Independent Oversight 1 Assess the quality of the NU Quality Assurance (QA) involvement in operations, maintenance and surveillance and engineering. Verify that the QA organization adequately implements the QA program.

XX. Nuclear Oversiaht

a. Inspection Scoce The team reviewed the activities of the Audits and Evaluation Group of the Nuclear Oversight Organization. This review involved discussions with auditors, inanagers, the director of the audit group and the Vice President, Nuclear Oversight. Documents reviewed included but were not limited to the Nuclear Oversight Group procedures, a sampling of recently performed audits, tracking and review of audit findings, adequacy of audit findings, scheduling of audits, training and qualifications of auditors, adequacy of staffing , adequacy of audit corrective actions, and Nuclear Oversight Group self assessment in addition to the Audit and Evaluations group, the following activities were also reviewed:

• Nuclear Oversigh. Group actions improve on deficiencies identified by independent outside assessments of quality assurance (OA) performed during 1996. Specifically addressea were a sampling of actions taken by the licensee to respond to criticisms identified in the Joint Utilities Management Assessment (JUMA) performed in July, 1996;

• Surveillance and quality controlinspections performed by the Performance Evaluation Group concerning repairs made to a Unit 3 emergency diesel generator; and,

• Various activities performed by Oversight implementing the nuclear Oversight Verification Plan (NORVP).

l

b. Observations and Findinas  ;

i The review of Nuclear Oversight involved personnel, programs, procedures, audits, audit j follow up, audit scheduling, certain aspects Performance Evaluation and NORVP. The observations and findings for each area are discussed separately below:

• Personnel Since 1996, the auditor staff has been increased from five to twenty auditors. The audit i staff has a significant background various technical areas including operations. In addition, there is an auditor qualification and training program in place. There are now four audit managers over the areas of operations, maintenance, engineering and technical support, and plant support; and an overall director of Audits and Evaluations. This increase in audit management attention and audit personnel has increased assurance of the quality of audit t

/. .  ! 'FI 'l

t I

2 findings. There is now a stronger interface with the line organization. Licensee audit teams typically have four or five members and take up to two weeks to perform. For this l

reason, support for the audit team is periodically obtained from contractors, the line organization, or other groups within the Nuclear Oversight organization.

• Procedures The team reviewed the following procedures:

NOOP 1.05, Self-Assessment Process, Revision 0, June 30,1997 NOOP 1.06, Nuclear Oversight Resolutions issues, Revision 0, November 12,1997 NOOP 2.01, Nuclear Oversight Audits, Revision 2, November 20,1997 l

- NOOP 2.02, Qualification / Certification of Audit team Leaders and Orientation of Team Leaders, Revision 2, November 20,1997 NOQP 3.02, Analysis of Quality Program Performance, Revision 0, August 29, 1997 NOOP 3.03, Nuclear Oversight Assessments, Revision 0, May 12,1997 NGP 3.19, Procedure to Stop work, Revision 1, July 30,1997 I

RP-4, Corrective Action Program, Revision 5, September 5,1997 I The team observed that all the procedures listed above were established or significantly revised during 1997. However, nat all were new procedures as they may have existed another form in the previously established QA program. The procedures appeared to be comprehensive, clearly written and user friendly. There were procedures for all aspects of the program. It was determined from the review of the procedures that a new audit program had been established with more strength than the previous OA audit system.

• Audits The team reviewed the following audits and audit checklists-l l MP-97-A11-03, Software QA, performed November 3-14,1997 l

- MP-97-A09-01, Fire Protection, Performed September 8- 26,1997 I 1

i MP-97-A05-02, Chemistry, Performed May 27 - June 6,1997 )

1 MP-97-A10-07, " Operating Experience" Program, November 10-17,1997 l

- Audit checklist for Audit MP-97-A09-02, Security l

l

3

- Audit checklist for Audit MP-97-A06-03, Systematic Approach to Training Procedure NOOP 2.01 has strengthened the audit process by more clearly defining audit expectations, audit checklists and the makeup of audit teams. A review of the above audits and audit checklist indicates significant improvement in the audit process. Audits are generally two weeks in length and performed by a team. This has led to a more in depth look into each area.

Audit findings are issued as level 1 condition reports (CRs) causing them to receive a high level of attention as required procedure RP-4. An initial 7 day audit report is issued with the CR findings. The audited organization must respond within 30 days and the audit group must agree with proposed corrective action as stated on then initial CR response.

After agreement on the CR response to the findings, a more complete 45 day audit report is issued. This process ensures significant QA involvement in the corrective action process. Less significant audit observations are issued as level 2 CRs.

In the past, audit exits were apparently given a low level of attention and were meagerly attended and rarely attended by appropriate management. The team reviewed a sampling of audit exit attendance sheets and observed that the exits were well attended by both line staff and managers.

• Control of Status of Audit Findinas l

As stated above, audit findings are controlled by the CR process. It is the responsibility of l

the line organization to provide corrective actions. The corrective action group for each l unit maintains a status of open CRs. However, they do not necessarily track CRs as audit findings. All due or overdue level 1 CRs are treated equally. The team had a concern that audit findings may be lost in the large number of open level 1 CRs.

The audit group has established its own computer tracking system of open audit findings.

This information is provided to audit managers and certain line personnel. It is used as a ,

tool to track overdue or inadequate audit corrective actions. The audit open item is tracked  !

by both the CR number and associated assignment request numbers (ARs). Recently the report has been improved to clearly show a summary of the open or incomplete issue.

This audit finding tracking system appears to provide an effective mechanism to ensure l that audit findings are not lost in the ' shuffle" and to ensure that audit managers have a tool for managing the follow up of audit findings.

All audit findings receive a follow up for adequacy of corrective actions; although this may not be a 100% verification of each action. Audit observations are followed up on a sampling basis. Audit follow ups may be done independently or during the next audit of I the subject area. Inadequate corrective actions are identified by the issuance of a new CR.

The team noted that this was done concerning corrective actions identified in a fire protection audit. The details of these audits are further discussed in NRC Inspection 97-84.

• Audit Schedulina and Plannina (Open VIO FO-423/96-05-11)

4 The team observed that an audit schedule had been established for 1998 and a projected schedule for 1999. These schedules just covered general areas to be audited and not detailed audit objectives. Actual audit planning is done by the team leader in conjunction with the manager for that area. Audit planning is established in NOQP 2.01. If not already done, the adequacy of corrective actions for previous audits in the area to be audited. The team verified that such audit planning is accomplished for each audit.

On May 29,1996, the licensee wrote an ACR that TS 6.8.4.e, " Accident Motoring instrumentation," may have been missed. Based on this ACR, a review of audits of the technical specifications, and a concern that TS audits were neither comprehensive nor well documented previously detailed in Inspection Report 50-423/94-28, violation 50-423/96-05-12, was issued stating that "The failure to audit technical specification section 6.8.4.e within a five year is a violation of TS 6.5.3.7..." In its response dated September 16, 1996, the licensee stated in part that "The process for the independent review of "The TS Audit Matrix" will be revised to require a review of "what has been done" versus "what has been scheduled." In the future, personnel will enter information into a tracking database after the has been completed and after the independent reviewer has checked "what has been done." Procedure...will be revised by October 31,1996."

The team venfied NOQP 2.01 has been revised to require an audit commitment database.

A database has been established for general audit commitments and a separate one for all technical specifications. Further discussions indicated that while a data base had been established to determine what technical specifications had been audited and what needed to be audited, it was difficult to use as a " tracking and scheduling" tool. The licensee had already discovered the same thing during a recent self assessment of the audit program.

Memorandum AE-98-4037, dated January 19,1998, issued report no. 97-AE-08, "Self-Assessment of the Audits and Evaluation Group's Audit Commitment Tracking Process."

The self-assessment was performed December 1-5,1997. The executive summary of this .

assessment stated, in part, the following:

I "The audit commitment database, as it is presently configured, does not meet the needs of the Audits & Evaluation (A&E) Group. This database is not a tracking database but is strictly a database for recording information. As a result, the A&E Group can not use this database to track when a commitment was addressed, by what audit and the deadline for addressing the commitment again. There is not a high level of confidence that all audit l

commitments listed in the 45 day audit report have been completely addressed...ATLs l [ audit team leaders} and Managers are not consistently using the database and,in some

! cases, do nct completely understand the purpose of the database. Finally, the database is l I

missing some data and contains some data that is no longer relevant."

It should be noted that the commitment database includes all audit commitment sources such as NRC regulations, the FSAR,10CFR50 Appendix B, ANSI Standards, regulatory guides, INPO, etc. and not just technical specifications. the licensee was ultimately able to generate a list of all TS in their database and those already audited,; however, this was not being used as a scheduling tool. The Director of A&E stated there is commitment to NSAB ,

to resolve this issue by June ,1998. Although much of the corrective action for the l violation has been accomplished, this violation remains open pending the development of i i

l 5

( an effective scheduling tool tom ensure that all technical specifications will be audited l within a five year period. In addition, the remaining commitment database is lengthy and

, cumbersome. The licensee stated they intended to " scrub" the database of unnecessary and obsolete commitments.

• Nuclear 'Oversiaht Review Plan (NORVP) {

The NORVP was established to provide independent oversight for readiness to restart and for heat up (mode 4). The NORVP meets at least weekly and frequently interacts with the line organization. Among the documents reviewed were the following:

• Work Control and Planning Brief (2/5/98)- Presents work control and planning for restart from a Nuclear Oversight perspective

• Nuclear Oversight Restart Verification Plan - Presents Nuclear Oversight numerical evaluation of progress in the areas leadership, corrective action, NSAB/ oversight, configuration management, engineering, maintenance /l&C, regulatory compliance, j radiation protection, conduct of operations, security, procedure quality & adherence, '

work control & planning, training, conduct of operations, and fire protection

• Millstone 3 Mode 4 Attribute Summary,2/9/98

• 50.54(f) Recovery Oversight status report for the week ending 2/7/98 Review of the above documents and discussions with Nuclear Oversight personnel indicate significant oversight involvement in the Millstone recovery process. The i reports were indicative of an extensive Oversight review. Nuclear Oversight meets three times a week to discuss station progress toward restart. Deficiencies observed in the startup process are discussed with the line organization. This process appears to be effective to ensure that senior management is provided with an independent evaluation of the restart process.

Self As.sessment Self-assessments have become a significant portion of the licensee's self improvement program. All Millstone organizations are required to perform self assessments including the Nuclear Oursight Group. As noted above the Oversight self assessment procedure is NOOP 1.05, Self-Assessment Process, Revision 0, June 30,1997. The team reviewed the following self assessments: j l l

• AE-97-S2 Final Self Assessment of : Audits and Evaluations Compliance With NOOP 2.01, Rev.0, " Nuclear Oversight audits, Dated May 16 1997 l-

• AE-97-S2 Self-Assessment Report of Training and Certification of Lead Auditors / Auditors in Training and Technical Specialists, Dated May 19,1997

• Self-Assessment Report of Performance Evaluation Strategic Plan Development and  !

Assessment of Organizational Effectiveness I

l l

- l 6  !

97-PE Self-Assessment Report of Applying PE QC Hold Points and AWOs by PE QC AWO Reviewers, Dated January 6,1998 97-PE Self-Assessment Report of Performance Evaluation's Work Process, Dated January 12,1998 97-AE-08 Self-Assessment of the Audits & Evaluation Groups's Audit Commitment Tracking Process )

t I

The team observed that the self assessments were in-depth and effective. The self assessment process allows for process deficiencies to be identified and brought forward in a formal manner. A response ro the NSAB is required and corrective action commitments must be made. The self-assessment process is proactive rather waiting for problems to {

arise or to be discovered by outside assessment groups. As noted above self assessment 97-AE-08 noted significant weaknesses in the audit scheduling process. " DRAFT" Nuclear Oversight Plans for Audits and Evaluation for 1996-2000, identifies future planned self assessments.

• Performance Evaluation The Nuclear Oversight Performance Evaluation Group performs surveillance of work activities and quality control hold point inspections for specific work activities. The surveillance technicians and quality controlinspectors comprise two separate groups. The term surveillance as described in this, paragraph applies to periodic oversight of various work activities and does not refer to surveillance testing as required by the Technical Specifications. Surveillance activities are controlled by the following procedures:

NOQP - 4.02, Performance, Reporting and Follow-up of Surveillance Activities and Field Observations at the Millstone Station, Revision 1, Dated May 20,1997

• NOOP - 4.09, Planning, Scheduling, and Administration of Quality Surveillance Activities, Revision 1, Dated December 19,1997 There are 27 surveillance technicians , of which, nine are assigned to Millstone Unit 3 activities. There are 23 QC inspectors. For both the technicians and inspectors there is a mix of permanent employees and contract specialists. Most technicians and QC inspectors were former maintenance workers or I&C technicians thus giving them credibility with the personnel performing the maintenance personnel.

The team reviewed documentation of the following completed surveillance activities:

l

• MP3-P-97-132, System Engineering Communication, September 23,1997

• MP3-P-97-105, Cortical Maintenance, Dated September 23,1997 ,

I

• MP3-P-97-123, Shielding Program, Dated October 7,1997 I

L l

l 3

f 7 l

• MP3-P-97-136, Control of Overtime, Dated November 3,1997 l

MP3-P-97-149, Procurement - Vendor Control, Dated December 4,1997

• MP3-P-97-153, AWO Quarterly Review, Dated December 6,1997 MP3-P-97-118, Material Condition - Field Walkdown - Housekeeping / Material storage, Dated December 30,1997 j l

MP3-97-154, Minor Modifications, Dated January 8,1998 r

• MP3-P-98-001, Conduct of Operations, Dated January 13,1998 l l

• MP3-P-98-005, AWO Quarterly Review, Dated January 19,1998.

Surveillance activities for major areas and known scheduled activities are scheduled l approximately six months in advance. The team reviewed the current six month schedule which called for 27 surveillance for Unit 3 alone. Surveillance for emergent maintenance l activities are scheduled as they occur. Not all jobs are reviewed; but, there is an attempt to provide some oversight of most major work activities.

During this inspection, the team watched one surveillance activity in the field. Testing of q the Unit 3 "B" Emergency Diesel Generator was stopped due to a leak on the air start line.

The leak had been caused by chafing between the air line and the fuelline. AWO M3 2900 was issued to replace and reroute both lines to eliminate the tubing contact. While there was no specific procedure for this job, generic procedure CMP 721 A, " Installation of Instrument Tubing, Fittings and Supports" was used as the basis for the tubing installation.

OC hold points were established by this procedure.

The team witnessed a nuclear surveillance of portions of the above work. The surveillance I comprehensive and reviewed all aspects of the job. There was good interface between the surveillance technician and the maintenance workers performing the job. Although not witnessed by the team, the surveillance technician briefed the workers as to the results of his oversight at job completion. Field observation checklist MP3-P-98-004-F18 was issued on February 12,1998, giving the results of the surveillance. The surveillance identified one deficiency concerning the fact that a systems engineer at the job sight did not have l controlled cnpy of the drawing in use. However, the maintenance workers did.

The team also observed a QC inspection of portions of this job. As part of the hold point <

the inspector verified tubing connections and tubing bend radiuses; and, then signed off a verification sheet. The OC inspector stated that he also looks at more than just what id called for in the hold point. Overall, the OC inspection appeared to be acceptable.

The teams evaluation of both the surveillance and QC inspection activity is that they were effective and comprehensive.

c. Conclusions l

r

~

8 The team determined that the Nuclear Oversight is effective in performing audits, general plant oversight, and work surveillance activities. Considerable improvement is noted since independent assessments identified considerable weaknesses two years ago in the performance of QA activities. Procedures and audits have improved. Audit findings are stronger and there is good control in the follow up of audit findings. Audit scheduling has improved but there are still weaknesses that have to be resolved. There is much better communication between the line organization and nuclear oversight.

YY. (Closed) Inspector Follow Item (IFI) 06-17. Joint Utilities Manaaement Assessment (JUMA) Concernina the Effectiveness of QA: (Open) Violation 423/96-05-12: Failure to Audit All Technical Specifications Within a Five Year Period: (Closed) MC 0350. Restart Checklist items C.1.4.a.b &c end C.2.1.c:

(Closed - SIL ltem 73)

The JUMA was performed during June,1996. A subsequent NRC inspection Report,96-06, referring to the JUMA results stated, in part, that:

" The JUMA team concluded that the audit, surveillance and inspection programs at Millstone were not effective in the implementation of their mission statement and the resolution of identified problems. The [JUMA) team attributed these problems to:

• Lack of support for the QA organization by the executive and line management

• Lack of an effective action program" The NRC made the licensee JUMA corrective actions an IFl in IR 96-06.

Paragraph XX of this report noted a increase in the effectiveness of the Nuclear Oversight Organization and increased support for the Nuclear Oversight organization by the NU l

President and CEO. There is increased interaction with the line organization and Oversight is involved with corrective actions to deficiencies identified by them. The licensee showed the team documentation that they had been responsive to all JUMA concerns. Paragraph

?? of this report demonstrates improvement in the licensee's corrective action process. It is not the intent of this inspection to make judgement as to the adequacy of the licensee's response to the JUMA. However, because of licensee actions identified during this inspection, IFl 96-0617 is considered closed.

The licensee response to violation 96-05-12 concerning the scheduling of audits of the technical specifications is discussed in paragraph XX. Although, the licensee has made considerable progress in this area, they are not yet effectively using their database to schedule TS audits. As stated in paragraph XX, this violation remains open pending completion of alllicensee corrective actions, but, based on corrective actions already taken, it is closed for purposes of SIL No. 73.

NRC Manual Chapter 0350, " Restart Approval Checklist" include the following items:

6 l

9

• C.1.4.a Effectiveness of quality assurance program

• C.1.4.b Effectiveness of industry experience review program

• C.1.4.c Effectiveness of licensee's independent review group

• C.2.1.c Management involvement in self-assessment and independent self-assessment capability Each of the above areas are addressed in various parts of this inspection report and have been found to be sufficiently acceptable to close these areas. Based on the above, SIL No.

73 is closed.

ZZ (Closed) Unresolved item (URI) 50-423/95-81 Lack of Trendino of Non-conformance Reports (NCRs) and Level "D" ACRs : Update SIL No. _41 Inspection Report 95-81 stated, in part, the following: "...QA did not trend any NCR or look for adverse trends which may be discernible from such data... the lack of level "D" ACRs may mask a recurring problem and its significance. Formalizing the practice of trending NCRs and level "D" ACRs would address a potential weakness in the program, and make the process less reliant on individual analysis and perception of the significance of the recorded problem.

The acceptability of the lack of trending of NCRs and verifying the effectiveness and adequacy of the ACR database by the Quality Assurance Department remains unresolved..."

Since IR 95-85 was performed CRs have taken the place of ACRs and level "D" ACRs have been eliminated. The team verified that NGP 3.05, "Non-Conformance reports" has been revised to have NCRs troeded in accordance with procedure RP-4, " Corrective Action Programs." Also a CR is wntte.n for each NCR issued. Procedure RP-4, Paragraph 1.17, states "at least quarten/, PSFORM trend analysis of AITTS database related to CRs and ISSUE trend report within 30 days of the end of the of the quarter. Trend analysis shall include human errors associated with NOVs and LERs...BRIEF appropriate levels of organizational management of results of trend analyzed trend data... ENSURE CRs are initiated for any adverse trends identified through periodic trending" The team reviewed recent trend reports and verified that CRs and NCRs (which are trended separately) are now trended at least quarterly. Based on the above review, URI 95-81-01 is closed. This

updates SIL No. 41.

.o Richard A. Rasmussen -Independent Oversight Millstone Nuclear Power Station Unit 3 Inspection Report No: 50-423/97-82 This report covers the following inspection plan sections:

Independent Oversight 3 Independent Oversight 4 Independent Oversight 5 06.1 Performance of the Nuclear Safety Assessment Board

a. Insoection Scooe (40500)

The performance of the Nuclear Safety Assessment Board (NSAB) was assessed by reviewing board meeting minutes, observing a board meeting, and interviewing board members,

b. Observations and Findinas The NSAB consisted of senior plant managers and two independent contractors. All of the board members met the Technical Specification (TS) requirements for education and background. Section 6.5.3 of technical specifications specifically stated the areas of expertise required for representation on the NSAB. To assure compliance with these requirements the board secretary maintained a matrix of qualifications of the board members. The list, as originally presented to the NRC, did not indicate any regular members as having metallurgy experience.

As a result, an alternate member with specific metallurgy experience was added to the list of alternate board members in February,1997. However, based on the review of the qualification matrix the NRC concluded that the technical specification was not met because no regular member possessed metallurgy experience.

The licensee re-evaluated the qualification matrix and concluded that their previous screening of qualifications was too conservative. The previous screening criteria required an academic degree or direct experience to be counted as a metallurgist.

However, this screening was in excess of the technical specification requirements.

Based on the new screening criteria two current members were identified with metallurgy experience. Although the NRC did not contest the actual board member's l qualifications, the evaluation and resolution of this discrepancy in 1996/1997did not l meet the literal interpretation of the technical specifications. This was an example of non-conservative interpretation and implementation of technical specifications.

Portions of a NSAB meeting were observed on January 29,1998. The meeting consisted largely of presentations to the board by various managers regarding departmental readiness for restart. Several board members asked probing questions and displayed significant knowledge of the issues and obvious preparation for the meeting. However, participation by the members varied notably with several members remaining mostly silent throughout the meeting.

The board identified a potential safety issue with fire protection systems that have

[/ l f/]

2 outstanding surveillance tests. The board questioned the acceptability of prolonDed use of compensatory measures. The fire protection program manager was requested to respond to the board at the next meeting. Action items assigned during meetings were tracked by the NSAB secretary and the status of open items were included as part of the meeting minute packages. Closure of open items required consensus of the board membership.

Another example of the NSAB providing appropriate oversight of plant activities was weaknesses identified in the area of training. As a result of NSAB intervention, the training on site was stopped pending program improvements. Additionally, the board thoroughly probed the area of operational experience (OE). To validate a presentation by the OE program manager, the board questioned various department managers on the use of OE by their departments during subsequent presentations.

The NSAB audit program was appropriately implemented by the Audits and Evaluation Department. The scope of the audit plan was reviewed by the board, and audit results were presented to the board by the auditors,

c. Conclusions The NSAB was effective in reviewing activities on site and identifying potential nuclear safety issues. The implementation of the NSAB met the technical specification requirements, however the 1997 resolution of an issue of membership qualifications did not meet technical specifications as presented. This was identified by the NRC as a weakness in the implementation of section six of the technical specifications.

06.2 Performance of the Plant Operations Review Committee

a. Insoection Scope (4050,0)

The performance of the Plant Operations Review Committee (PORC) was assessed by reviewing committee meeting minutes, observing board meetings, and interviewing board members.

b. Observations and Findinas During observed PORC meetings, PORC membership met the technical specification requirements. PORC members were prepared for the issues on the agenda and asked techni::al questions of the presenters. The questions focused on safety and compliance with regulatory issues. The PORC meetings were conducted in a professional manner with an emphasis on clear communications between the committee and the presenters.

Station expectations for the conduct of PORC were high as demonstrated by several assessments of PORC performance. The PORC process was being evaluated at the station level as including the performance of the committee as well as the personnel presenting items to the committee. Instances such as presenters being unable to I

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

3 answer PORC questions and PORC-rejected documents were considered weaknesses.

An effort was underway to develop a feedback process to track and reduce poor presentations to PORC.

PORC issues were tracked in the plant action request tracking system by the PORC secretary. A review of the backlog indicated timely resolution of PORC issues.

c. Conclusions PORC was effective in accomplishing the reviews required by technical specifications.

Station evaluation of PORC performance and the standards being demonstrated set high standards for the documentation and review of technicalissues.

06.3 Performance of the Site Operations Review Committee

a. Lnsoection Scope (405QQ)

The performance of the Site Operations Review Committee (SORC) was assessed by reviewing committee meeting minutes, observing board meetings, and interviewing board members.

b. Observations and Findinas During observed SORC meetings, SORC membership met the technical specification l requirements. However, technical specifications designate that the Senior Vice President and Chief Nuclear Officer (CNO) shall be the SORC chairperson. A review of SORC meeting minutes demonstrated that the chairperson responsibility was normally delegated to the Director of Unit Operations. The basis given for this delegation was the fact that the Chief Executive Officer (CEO) was acting as the CNO and a potential conflict existed because technical specifications state that disagreements between the CNO and the SORC shall be brought to the attention of the CEO. The team did not take issue with delegation of SORC chairperson responsibilities, i

l SORC members were prepared for the issues on the agenda and asked technical questions of the presenters. Members adequately represented the site-wide perspective of SORC. This site presents a particular challenge with the difference in license requirements between the units. SORC members displayed a combination of j knowledge to integrate site-wide license and technical requirements. This was evident i

in a discussion of fire protection issues which required detailed knowledge related to all three units. During one meeting the inspector noted four occasions that safety issues which required further evaluation were identified by the committee.

j SORC issues were tracked in the plant action request tracking system by the SORC secretary. A review of the backlog indicated timely resolution of SORC issues.

c. Conclusions i

4 SORC met the technical specification requirements and was effective in identifying potential safety issues.

06.4 Effectiveness of Quality Control Oversiaht l

a. Incoection Scone (405001 j i

The effectiveness of the Quality Control (OC) department was assessed by reviewing procedures, interviewing personnel, and accompanying inspectors on inspections,

b. Observations and Findinas The QC Inspectors interviewed were experienced and qualified in their areas of expertise. The licensee utilized a mix of staff and contract personnel to perform the QC inspection function. QC Inspectors were knowledgeable of the site work control process and documentation. All of the inspectors stated they would stop work if required. During observations of inspections in the field, two inspectors stopped jobs due to questions with proposed signoffs. All of the interviewed QC inspectors stated that they now felt they had management support to stop jebs if required.

Although the observed inspections were generally rigorous, the NRC observed one instance that indicated insp< 7tions performed as part of the spent fuel pool foreign material exclusion (FME) program were not adequately performed. The purpose of the inspection was a monthly requirement to sign for a list of objects allowed within the FME barrier and to determine that r o other objects existed. The NRC inspector noted a row of tape pieces hanging from t' o underside of the spent fuel pool bridge. The tape was not listed as authorized and was not noticed by the QC inspector. Although the tape was not noted by the QC inspector, the QC inspector did raise other concerns with staging that was in the area and not tracked with the list. A condition report was written to document the tape, however, the response went to the reactor engineer and was being treated as only a technicalissue. Through interviews of the QC manager the inspector identified that the QC inspector issue was not being addressed.

The QC support group, which was developed to standardize the inspections performed by the QC inspectors, was a strength. The QC support group reviewed all work packages to identify hold points prior to the packages going to the field. The group l deve!oped standardized inspection points for many routine work activities and group members coordinate closely with each other to assure standardization. Additionally, the group had a rotational assignment that was filled by an inspector from the QC inspection staff. The purpose of this position was twofold. It brought the QC inspector's perspective into the process and trained the inspectors on the group's policies and procedures. This was essential because emergent work packages developed after hours were processed by the field inspectors in the absence of the QC support group, i.________________.____

5

c. Conclusions Quality Control was generally effective in performing the required in-plant inspections.

However, the spent fuel pool FME inspection was noted as an inspection that was not adequately performed. The QC support group was effective in establishing and standardizing the use of QC hold points in work packages.

++++++++++++CUTHERE++++++++++++++++

Update SIL ltem 73 (Cut and Paste into Norm's write-up on this SIL. Some of the terminology may need to be modified to be consistent with the Normster.)

An issue identified by the JUMA audit team involved QC inconsistently assigning hold points in work packages. This issue was addressed by the licensee by establishing the QC support group. The procedures for reviewing work packages and assigning hold points were formalized. The QC support group reviewed all work packages to identify hold points prior to the packages going to the field. The group developed standardized inspection points for many routine work activities and work members coordinate closely with each other to assure standardization. Additionally, the group had a rotational assignment that was filled by an inspector from the QC inspection staff. The purpose of this position was twofold. It brought the QC irispectors perspective into the process and trained the inspectors on the group's policies and procedures. This was essential because emergent work packages developed after hours were processed by the field inspectors in the absence of the QC support group.

The inspector reviewed several work packages and concluded that the inspection points were appropriately selected. Based on the program improvements and implementation of the process by the QC support group, the inspector concluded that the issues identified in the JUMA report were adequately addressed.

+++++++++++++CUTHERE+++++++++++++++++

06.5 Performance of the Independent Safety Enoineerino Group

a. Inspection Scoce (40500)

The performance of the Independent Safety Engineering Group (ISEG) was evaluated by I reviewing ISEG reports, reviewing the operating experience (OE) program implementation, reviewing ISEG backlogs, and interviews of personnel,

b. Observations and Findinas Millstone unit 3 technical specifications required an ISEG consisting of four full-time personnel to perform independent reviews of plant operations. The unit 3 ISEG consisted of three full-time engineers, one part-time contractor, and a supervisor. The ISEG charter required the group to use operating experience when reviewing plant i

I 6

l operations and to make detailed recommendations to improve safety and reduce human )

errors. in implementing this charter, the licensee elected to make the ISEG the site group responsible for reviewing and implementing OE. OE typically comes in the form i of industry group reports, vendor notifications, and NRC information. l1 Over the past tvso years the'ISEG focus appears to have shifted from the ISEG activity of independent reviews of plant operations to the performance of OE reviews. A large backlog of unreviewed OE items were processed which resulted in a reduction of the number of ISEG activities performed. The number of ISEG reviews done in 1997 was only 12, down from 24 the previous year. j Reviews of plant operations performed by the ISEG group were documented in reports.

Reports reviewed by the inspector indicated that the ISEG performed critical reviews ,

and made appropriate recommendations. Examples included an ISEG review of high voltage switchyard work, and an evaluation of work activities in the spent fuel pool.

The ISEG group used the action request (AR) process to track the implementation of recommendations and performed closeout inspections of each item. j 1

The inspector followed up on the condition report from the ISEG review of the high voltage switchyard work. This ISEG evaluation identified several significant issues '

which involved personnel safety and the potentialloss of off-site power. The initial response to the issue was good, however the condition report was processed as only a level three. This was not identified or challenged by the ISEG group. NRC review of this issue indicated that the condition report should have been processed at a higher level.

By focusing on operating experience, the ISEG group reduced the backlog of OE issues from several hundred to approximately 40 for unit 3. A sample of OE packages was reviewed and found generally thorough and complete. However, one weakness was identified in the review conducted for NRC IN 97-14 related to spent fuel pool cooling.

The information notice suggested reviewing siphon breakers used to prevent spent fuel pool draining. The OE review confirmed that the siphon breakers were installed, however the reviewer stated no future inspection or maintenance was required. The inspector considered the issue of potential fouling of the siphon break orifices or the l need for periodic inspections was not adequately evaluated. The licensee was l reviewing this issue at the close of the inspection.

Although the backlog of OE issues had been significantly reduced over the past year, l the amount of work represented by the remaining issues was significant. Several of the issues remaining in the backlog had a high probability of identifying safety significant issues. One example is NRC IN 97-78, which involved identifying unreviewed safety questions related to the use of manual actions in place of automatic actions in emergency operating procedures (EOP's). The review of this issue was completed for unit 2 which identified several areas of concern. Additionally, a previous NRC inspection identified an issue involving manual action for a control room ventilation system that cannot be accomplished within the time assumed in the safety analysis.

Based on these factors the priority of evaluating this issue appears inappropriate.

4 7

Site implementation and use of OE was mixed. OE is not consistently being used by the working groups at this time. A key reason is that the site wide procedure to establish the expectations for the use of OE was in development and not yet issued.

Once a site-wide procedure is issued, the departments still have to develop implementing procedures. The OE Minute, a daily publication listing relevant OE items, appeared effective in disseminating OE to the site on a daily basis. However, the info mation w0s not easily retrieved after the f act. Industry and NRC information was -

sent to appropriate personnel (system engineers, operations) ft r information, however the site responsibility for evaluating and identifying actions was with the ISEG group.

Access to the industry nuclear network data base was improving, but still !acked in some areas such as system engineering.

l

c. Conclusions l l

The ISEG was staffed and met the technical specification requirements. Hcwever, the j implementation of the OE reviews by the ISEG was limiting the number of independent reviews of plant operations performed by this group. ISEG independent reviews and ,

ISEG reviews of OE were generally thorough. Corrective actions for ISEG items were tracked and verified by the ISEG prior to closure.

Examples of ooor CR reviews l

CR M3-97-2898, dated September 2,1997, documented that a nuclear oversight audit identified procedures, tools, and equipment needed to support emergency operating procedures were not available in the plant, The corrective actions for this CR were not tied

! to a key event and were outside of the scheduled restart date at the time they were

! approved. This issue was also picked up in the NRC review of differed items and was moved to a mode 2 issue.

CR M2-97-0510, dated April 1,1997, documented an ISEG activity that identified j significant work controlissues in the high voltage switchyard. These issues involved personnel safety and loss of off-site power. The initial response to the issue was good, however the associated CR was processed as only a level three. This was not identified or  ;

challenged by the ISEG group. Although listed as a unit 2 CR, the loss of off-site power J had the potential to effect all of the units.

CR M3-97-3974, dated November 11,1997, documented an audit finding that the  ;

ISEG/OE procedure, NOQP 3.04 was not reviewed by SORC as required by technical l specifications. The ISEG response was to get SORC review of the new OE procedure which was being developed. However, this response was inadequate because the ISEG l procedure also needed SORC review by the same technical specification.

CR M2-98-0419 documented seven pieces of tape that were identihed by the NRC during a OC foreign materialinspection of the unit two spent fuel pool. The CR was assigned to reactor engineering as a technical issue and was not evaluated by OC as an inspector performance issue.