ML20150C007

Trends & Patterns Program - Special Rept,Operational Experience Feedback Evaluation,Rancho Seco Nuclear Generating Station,Restart

ML20150C007
Person / Time
Site:	Rancho Seco
Issue date:	03/31/1988
From:	Dennig R, Plumlee G NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
Shared Package
ML20150B986	List: ML20150B983 ML20150C007
References
TASK-AE, TASK-S802 AEOD-S802, NUDOCS 8803170245
Download: ML20150C007 (72)

Text

- - - _ - - - - - - -

~

1 i

Enclosure 1 AE00/S802 TRENDS AND PATTE0NS PROGRAM - SPECIAL REPORT OPERATIONAL EXPERIENCE FEEDBACK EVALUATION RANCHO SECO NUCLEAR GENERATING STATION, RESTART

'FARCH 1988 Trends and Patterns Analysis Branch Office for Analysis and Evaluation of Operational Data ,

Prepared by:

G.L. Plumlee III  !

Robert Dennig c

1 1

1 1

8803170245 8803KM3 )

DR ADOCK 05CK gg 2 ,

r. 90 .

t

l; CONTENTS Page

1. Purpose ........................................................ 1

2. Background ..................................................... 1

3. Methodclogy .................................................... 2

4. Results......................................................... 4 4.1 Summary of Conclusions and Suggestions' for Improvement..... 4 4.2 Detailed Di scussion of NUREG-1275 Lessons. . . . . . . . . . . . . . . . . . 6 FIGURES

1. SMUD/ Rancho Seco Restart Milestone Summary...................... 9

2. Rancho Seco Critical Path Summary............................... 10

3. Hot Shutdown Testing............................................ 12

4. Rancho Seco Power Ascension Schedule 0verview................... 13

5. Scope List Status............................................... 15 j

6. Power Ascension Testing.......................................... 17

-7. Safety and Performance Improvement Program (SPIP) Summary........ 28

8. Plant Performance and Management Improvement Program............. C-2 i

APPENDICES I i

l A. Reference Documents............................................. A-1 B. AE00 Team Intervi ews of SMUD Ma nagement. . . . . . . . . . . . . . . . . . . . . . . . . B-1 C. Plant Performance and Management Improvement Program (PP&MIP)... C-1 j D. Independent Review Informa ti on. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0-1 l

1 i

i

1. PURPOSE The Rancho Seco Nuclear Generating Station is preparing for the resumption of operation after an extended outage lasting more than two years. The purpose of this report is to provide answers to the following four que restart pions in support of NRC decision making for the Rancho Seco

1. Has Rancho Seco made a responsible attempt to learn from the startup experiences of other plants and from their own earlier experiences?

2. In what areas have they been effective in learning from experience and in preparing for restart (i.e., program strengths)?

3. In what areas have they been less effective in learning from experience and in preparing for restart (i.e., program weaknesses)?

4 In what areas should NRR and SMUD focus their attention while monitoring the 25-week startup and power ascension program?

2. BACKGROUND On August 4, 1987, the NRC staff briefed the Commission on the performance of commercial nuclear power plants during their first two years of licensed operation as documented in NUREG-1275. (Note that within this report, NUREG-1275, Volume 1 will subsequently be referred to as NUREG-1275.) This study reviewed the causes of unplanned reportable events (e.g., unplanned reactor scrants) during early operation for plants licensed between March 1983 and April 1986. It also derived a set of improvement lessons that appeared to have the greatest potential for smoothing the transition from construction to operation and improving performance during the startup program and early commercial operation. As a result of the briefing, the Commission directed that NUREG-1275 be promptly disseminated to plants that a currently in early licensed operation g approaching licensingthat It was determined or some are of these improvement lessons might be applicable for plent startup following a prolonged age and a copy of NUREG-1275 was transmitted by letter to Rancho Seco I

l l

1 M emorandum M G. Holahan (NRR) to T. Novak (AE00), Review of Rancho Seco Response to NUREG-1275 Lessons Learned, February 12, 1988.

U Staff Requirment Memorandum M8708048, August 18, 1987.

E etter, L E. Jordan (AE0D) to G.C. Andognini (SMUD), September 1,1987.

NRR is using NUREG-127hfor plants restarting after extensive outages, including Rancho Seco . On January 19, 1928 the Sacramento Municipial Utility District (SMUD) providg a formal review of Rancho Seco against the NUREG-1275 lessons to NRR , who in turn requested that AE00 review the rest.onse. AE0D staff visited the Rancho Seco site on February 16 and 17, 1988.

3. METHODOLOGY The basic method for the review of Rancho Seco readiness was to use operating experience documented in NUREG-1275 to benchmark Ranco Seco and to determine those lessons, if any, which had not been implemented in their program for restart.

NUREG-1275 contains lessons learned from plants going through their initial post OL startup and power ascension program. A number of the lessons were applicable to plants later in life and the degree to which these were applied was the result of the changes in the plant staff and equipment over the extended outage.

Following a two year outage, the status of the people, plant, and prcgrams at Rancho Seco place the facility somewhere between an NT0L plant and an exrerienced operating facility. Overall, one might view Rancho Seco tod y as a second unit NT0L at a two unit site: the facility gains some adyt.r.tages from the first unit (i.e., Rancho Seco prior to the shutdown), but enough is new that the experience may not translate into a smooth startup without some special effort to be more proactive and look forward toward startup operation. Major factors that contribute to Rancho Seco's resemblance to an NT0L are:

Installation of almost 600 plant modifications.

Rewrite of all emergency and casualty procedures, ana 80% of surveillance procedures.

A newly structured and staffed management team.

AE00 also used the NUREG-1275 approach to review reportable events occurring three otherduring the fiptg2 months following an extended outage for B&W plants l

Plant Outage Period i Three Mile Island 1 March 1979 - October 1985 Crystal River 3 March 1985 - August 1985*

Davis Besse June 1985 - December 1986

• Refueling outage N emorandum, M T. Rehm to T. Murley et al, Quarterly Program Review notes, l August 25, 1987.

E etter L AGM/NPP 88-37, J. Firlit (SMUD) to F. Miraglia (NRR), Rancho Seco Response to NUREG-1275, January 19, 1988.

E uick Q Review of Operating Experience for Three Mile Island 1 Davis Besse, and Crystal River 3 Following Restarts, ORNL, February 12, 1988.

E B&W Plants Event Trend Graphs, INEL, February 16, 1988.

. -i Reportable events (scrams, ESF actuations, technical specification (TS) violations and safety system failures) were trended by month following restart, and all LERs during for one year period following restart were reviewed to determine circumstances and causes. Major findings and conclusions for this reportable events review were: I TMI and Davis Besse power escalations were fairly smooth, i.e.,

few unplanned events or TS violations.

Crystal River operated for about 6 months following the refueling outage before being forced down due to reactor coolant pump shaft failure. During that period, the scram and ESF actuation rates were somewhat elevated, j Few problems were common to all three plants.

The experience at Davis Besse and Crystal River did emphasize the need for operator preparation to deal with overcooling events. Also, the Crystal River experience with a high number of spurious actuations of the newly installed Emergency Feedwater Initiation and Control (EFIC) system was directly relevant. Rancho Seco has taken advantage of this experience as will be discussed in detail in Section 4.

The analysis and lessons in NUREG-1275, experience with plant start-ups after publication of NUREG-1275, our review of restarts of TMI-1, Crystal River, and Davis Besse, the SMUD response to NUREG-1275 lessons, and the formulation of questions as followups to SMUD's response comprised bases for discussion with the licensee. These dicussions were accomplished through:

(1) A formal briefing by SMUD management on the background and current status of the overall SMUD program (Reference 1 of Appendix A).

(2) Attending the NRC-SMUD/ operations meeting on the Operations l Department Management Action Plan (Reference 2 of Appendix A). ,

(3) Meeting with mid-level managers from operations, training. I maintenance, plant performance, testing, scheduling and licensing as listed in Appendix B.

We also attended the NRR Operational Readiness Review Inspection (ORRI) team meeting on February 16, 1988 and the ORRI exit meeting on February 17, 1988. We also reviewed licensee documents listed as References 5 i through 15 in Appendix A.

, The results of our review are provided in Section 4. Section 4.1 provides 1 a summary response to the four questions stated in the Purpose. Section ,

4.2 contains a detailed discussion of Rancho Seco's situation as it )

relates to each NUREG-1275 Lesson.

1 l

l

4. RESULTS 4.1 Summary of Conclusions and Suggestions for Improvement (1) Has Rancho Seco made a responsible attempt to learn form the startup .

experiences of other plants and from their own earlier experiences? '

Rancho Seco has made extensive use of operational experience, both their own and from other plants, in formulating and implementing hardware and management improvements during the extended outage. The SMUD Plant Performance and Management Improvement Program (PP&MIP) (Refer to Appendix C) and related System Review and Test Program (SRTP) were based on a fairly exhaustive review of Rancho Seco experience and industry experience, including, for exarrple, scram reduction measures identified in the Babcock and Wilcox Owners Group (B&WOG) Safety and Performance ImprovementProgram(SPIP). SMUD has also tapped industr through the large number of independent progran reviews Refer(y to experience Appendix D). Rancho Seco also took advantage of the industry initial experience l with newly installed EFIC systems. Experience at Crystal River and other sites was translated into hardware, training, and maintenance improvements that should diminish the incidence of unplanned EFIC initiations following reactor startup.

The experience base used by Rancho Seco was not limited to startup I experience or viewed solely with the perspective of preparing for a startup. The SMUD review and NUREG-1275 shared the goals of reducing unplanned events. The lessons that emerge, especially for improving equipment, are largely the same (for example, NUREG-1275 found that the causes for scrams for new plants during startup were similar to those j found in the mature propulation). Thus, while traveling on different ,

paths, SMUD's PP&MIP and NUREG-1275 arrived at many of the same conclusions. .In addition, management stated that they had utilized NUREG-1275. As a result, with few exceptions, Rancho Seco has programs, plans, or procedures that address the lessons of NUREG-1275. The issue for restart becomes largely one of their current state of implementation 1 and the effectiveness of that implementation.

(2) In what areas have they been more effective in learnin and in preparing for restart (i.e., program strengths)g?from experience The preparation of the plant itself appears to be a strong point. The extensive review of operational experience, the resulting engineering changes, and the related pre-startup testing programs (component and system level tests) should eliminate many of the unplanned events (scrams, ESF actuations, system failures) that stem from hardware failures and design / installation problems. Also, ongoing construction (or major modifications) and late system turnover do not appear to be problems.

In the area of staff preparation for restart, licensed. operator training for nonnal operations appears to be a program strength, both in terms of the simulator training and the plans for extensive training at various plateaus during power ascension. Looking beyond the restart and power ascension, the involvement of plant operating and maintenance staff in the test program as well as the plans to move system test engineers into a permanent support staff should have benefits in power operation.

. - - ~ .

As a last point, it does not appear that the restart will be complicated by having to deal with last minute technical specification changes. Only three changes remain to be approved, and plant procedures and training have anticipated the changes.

(3) In what areas have they been less effective in learning from experience and in preparing for restart (i.e., program weaknesses)?

As evidenced by the Operations Department Management Action Plan dated February 12, 1988, Rancho Seco has beer less effective in stressing the importance of details, the need for discipline in following procedures, tight coordination throughout the plant staff, and the need for expedited resolution of problems (Section 411). In the Action Plan the plant operations staff has listed the actions they feel are needed to correct the situation, including improvement in surveillance, an area where our own review had raised a concern.

Rancho Seco has rewritten approximately 80% of its surveillance procedures during the outage, and experience has shown that adequate debugging of new procedures requires that they be performed as much as possible on actual systems by permanent plant staff. This was not a management goal. At this time, however, it appears that the licensee is conducting a concentrated review of new surveillance procedures that includes executing each procedure where plant conditions allow at least once before restarting the reactor (refueling interval surveillance excluded). The outcome of this effort will affect how many problems (personnel error, spurious actuations) will result from routine surveillance during restart and power ascension. Some additional inspection attention may be warranted.

Finally, startup experience has not been adequately reflected in the procedure for post-transient cause investigation and restart decision-making (see Section 4.2.2). That procedure (AP.28) allows a verbal restart decision by the Plant Superintendant if the preliminary post-trip investigation by the Shift Supervisor determines that the transient was i "Type I", e.g., a simple scram with expected plant response and no l complications. The more complex "Type II" event receives a broader l investigation with additional input from Engineering and Quality Control.  ;

Our concern arises frcm experience where inadequate investigation of I simple events resulted in frequent recurrence and additional unnecessary challenges to the plant and staff. Also, in NUREG-1275, it was mentioned that the root causes for these simple events were the same as those of significant events. Thus, we suggest that the licensee modify the procedure for post trip review (and by extension the review for all reportable events) to more fully ensure that the root cause corrective measures are fully defined prior to restart, implemented in a timely manner, and that the appropriate range of plant staff and management is involved in the analyses and decision-making.

Some minor areas for possible inprovement discussed in Section 4.2 are:

Strengthen the procedure for preparing for ma%tenance or I modification work by requiring walkdown of the area prior to i

the actural start of work (Section 4.2.4).

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

l

l Strengthen the closure of Restart Punchlist Forms by requiring signature by the Director, Nuclear Operations and Maintenance (Section 4.2.2).

Strengthen operational feedback to maintenance personnel (Section 4.2.5). ,

(4) In what areas should NRR and SMUD focus their attention while monitoring the 25 week startup and power ascension program?

NRR and SMUD should pay special attention to the following areas during restart and power ascension:

Thoroughness of root cause determination following unplanned reportable events.

l Performance of routine technical specification surveillance.

4.2 Detailed Discussion of NUREG-1275 Lessons 4.2.1 Lesson Establish an operating plant mentality well prior to initial criticality, i

Ensure that plant operations personnel have the responsibility for operating all equipment as early as possible in the completion process. Take early, complete control of the transition from construction to operation.

Have personnel who will be responsible for maintenance and

[surveillence] testing of plant systems after licensing begin these activities using post licensing procedures before fuel load. This lets procedures get debugged, and the plant staff gains experience in working under licensed conditions.

Stress the importance of details, the need for discipline in following procedures, the need for awareness of plant conditions and i the regulatory recuirements associated with these conditions, tight  ;

coordination throughout the plant staff, and the need for expedited resolution of problems.

Minimize continued construction activities after fuel load [ restart]

that may have an adverse impact on plant operations. Reduce plant staff to operational size, remove construction equipment, and establish housecleaning programs. Bring A/E, NSSS vendor key personnel onsite so that problems can be resolved promptly when discovered.

Discussion Operator involvement during testing has provided operating experience.

Operators are 1) involved in the testing action plans development; 2' in command and control of testing; 3) tasked with detailed test procedure reviews; l l l

4 and 4) involved in crew briefings before the test. Further, discussions with the Rancho Seco Operations Department Manager indicated that Operations has reviewed, walked down, and gone through each operating procedure step-by-step.

Caution statements are integrated during procedure development or revisison in efforts to prevent im dvertent transient initiation. For new system procedures, operators are trained on the procedures prior to completion of the modification utilizing draft procedures developed at the same time the design is developed. For reactor protection surveillances, i.e., those most likely to cause an inadvertant scram, trial runs have been implemented to verify adequacy.

During the extended outage, approximately 80% of the plants surveillance procedures were upgraded. According to plant staff, as part of the approval process the procedure is walked down and any necessary revisions are made. i ISC surveillances do require an actual dry run as a debugging assurance. The emphasis has been on those procedures needed during the extended outage, and those where credit from performance of a Special Test Procedure (STP) could not be used to safisfy a surveillance requirement. However, per the Manager, Plant Performance, a recent review indicated that taking credit for STPs resulted in a large number of cases where the first performance of a revised surveillance procedure would occur during the restart and power ascension.

This finding and other recent difficulties with procedures resulted in a decision to restrict use of a STP for surveillance credit to those surveillances performed with a refueling interval. All other surveillance procedures are being assigned to lead individuals (for example, a Shift Technical Advisor or maintenance supervisor) who will, prior to restart, ensure that a walkdown is performed, revisions are made, and responsible personnel are briefed. Procedure rewrite status will be discussed at the Tuesday and Further, every Thursday (heat-upmeetings.

startup i.e., plant conditions will allow) procedure that canusing will be performed be run theprior to normal procedure at least once prior to restart. This will not allow every person (e.g. all shifts) to actually perform the procedure prior to restart. However, Rancho Seco adheres to an on-the-job-training (0JT) policy that requires any person who has not performed a procedure to be qualified by an individual who has. This process has resulted in fewer surveillance problems during actual performance.

Adherence to procedures and attention to detail is explicitly stressed in Administrative Procedure AP.23 "Conduct of the Operations Division " in the AP.4 series of clearence, test and caution authori:'ation procedures, and in Special Order #87-37, "Surveillance Procedure Scheduling / Performance."

Even with the operator involvement and procedures addressing attention to detail, recent areas for operations improvement have been identified. Recent events such as water spills and procedural noncompliance have led to I concerns. The Ranch Seco Operations Shift Management together with the Operations Staff Management recently developed an "Operations Departemnt Management Action Plan" to resolve errors which were attributable to  ;

personnel. This action plan addresses. i l

l i

I l

l

~ . -

l 4 k operations shift management buy-in to conduct of operations, t

• shift organization and responsibility, professionalism, attention to detail.

holding people accountable.

supervision in field and coordination in the control room, procedures and procedure use, operations surveillance procedures and work planning, equipment status control, and operations departement morale and attitude.

From a review of the Operations Department Management Action Plan and attendance at an NRC February 16, 1988, briefing on this plan, it became obvious that many self-improvement actions have been generated recently for completion either prior to resumption of testing, heatup, reactor startup, or power operation, and others are continuing on-going actions. Certain actions 4

that will be ongoing during reactor startup relate to operations procedure '

revisions which indicate that procedures may not be adequately debugged prior to startup, e.g.,1) a rewrite of Operations Surveillance Procedures SP 1, 2, and 3 (instrumentation surveillance procedures) will include a data sheet and efforts to make the steps clear and concise, 2) Operations will determine the format and content of system operating procedures by power operations, and 3) surveillance procedures that are performed by Operations will be rewritten into a new format that is "user friendly", and will facilitate preplanning,  ;

less errors, and greater attention to detail by power operations. ,

I Rancho Seco has completed 567 of 599 modifications including the addition of new systems such as the TDI diesels, control room / technical support center a HVAC, post accident sampling system, and EFIC. As indicated in Figure 1, most of their restart milestones are complete in regards to system testing except for the TDI diesels which have a high vibration problem remaining to be resolved. Cold shutdown testing completion has been delayed due to these TOI problems. However, only the loss of offsite power test remains (163 of 165 cold shutdown tests have been ccmpleted). Therefore, as indicated in Figure 2, only a couple of systems, e.g., fire protection (surveillances only remain to be performed) and the TDI acceptance test, would have been turned over to the operations personnel less than a month from planned restart. -

Numerous activities have been completed to minimize continued modification and maintainence activities that may have an adverse impact on plant operations. .

For example, as indicated above a majority of the modification work has been '

completed. As a result, reliance on contractor personnel has been reduced from approximately 1700 to 800. Additional effort is underway to reduce the numbers of temporary crafts personnel on the site.

! Conclusions Ongoing construction (or major modifications) and late systems turnover do not appear to be problems at Rancho Seco.

The involvement of the plant operating staff in the testing program should help avoid problems during restart and power ascension, i

,  ! l 1l '

i' ,  ! ,' l

, ~

Y R g' uI a

7

) A R A 7 6 2 g A

m A

7 M M gS gK 2 a g

6 2 M 0 yF '

3of s

KrS f

(4 U 0 B 3 7lllTr KaK n @gI{[ggygI l[1 E 1II[lIlBa S 0 F E F

2 7

5 N

A 4 4 E m E

f E J l2 T f N f NN 4

aAJ NA A QJN A O

T S

N A

J 7D o

J C

E 4s 1

a AJ J

NAM N

NJ AJ J MMM m E C D M mm C

em E

E D f L

I D C E

D M M t AX IM I f f

S DL 7 F V

N EMM NNN NNm 1

0 O M FW W _

T 0

/

N R 8 7

E E

A T 8 CO T T T S

E R

C O

C MMM O

E mMNN N N -

M L A O N T R 0 T l T AAA^ AAAAAFrF D' N l N S M O A R 0 K M C T E 9 A 1 A A r I

D GO N N T

AN 1

C C 777' 77777888 URR ER LL J

E 888* B8888888 R M I T U W0 R T OO WI L l

O CTT T. U SP l e O R T RN OU R O AAA UUCOEEAA M A AAAO F ONI T 01 E M T A N8 wP F NRY2GCTVCBRR JMM AAONDFMM U RE S Rt ED PT N NR RI T

S1 MC S Um C E P OT AM

/ U N t

a TI R0S 01 U L 3 7' 908359 10 C E E R EETNNE A t T E

I E R T I T A 222'1 23 1 2 2 A G U CCISOL CWET UO PO C A C DCEY R A

TO DE D SMLRT T AD/

V U . . :........ Y AL ECOF PAE O T U D T T LV C T T . R C FSSO C EEY T F A LC R A ONfO E C AHwA0V S

.... t H D E CA UO

...~ ...

C ..

E NY EE A MMmCnH R A . ~

~ . .

...... S N WETRY DD I

FO RO R

Ct c.UASN RnU TY CA G H Y H TRP CT S T AiS O

. . . RT E EDO N T STS MOLRA E T /oE T E ET F WW MtD A AC C

. ~

.T D

W MAAEES EBSNNSA C 1 MEAWNEC RD STCAL NO PMCECRH 5 A

. .. cS O OE E C LR O P R uE

. T . . .O .G. .

m e e a

.R .N m SL .

D .

tA . . Y .T I

N

. . IA* N'O"i .

.N.H .SE .

E .

R" I

T

.OY .T I

G .

T' C .

.TR . .

A N . .AAPN .

T .

USA OCE' T

%FU .MUDUW TNTO

.U COADP.

RH" D .U ACETU G L .C EHUT STY TNA'eC *S O .VSA FS T HR SAT E OECL C T.R ECN SATS N E D 1REDTLO O

T S

TD VEAF EN*A?

LA*

EILS NOPHRO TT NA T LTT E

1 E P RRD PR A I RC e L MLL r I

M OLL%TTOOTAANSATAG I 1 CFF'CSSCLSI S I T N u g

eem;eeeaemea i F

, ' O'

4

- - - - - ~ _ l . = m i,

j (5365 -

j RANCHO SECO "-

CRITICAL P ATH

SUMMARY

(SCHEDULE FORECAST - FEB 12, 1988) l 1988 FEBRUARY MARCH it'131*ll*l7

• l. I'*Ini'%I" 'isl'*l A'*l .I'%%I'*l24'*i>4'" 2. 'l:l'I. l *l. 'I. l *liol"I,21" ..I'*l .I'bi'%

I CR/TSC HVAC MMWWImmeFEB 5 I Toi TESDNC M FEB 4 1 TDI ACCEPTANCE . . ... . -

FEp 27

{

LOSS OF OFF3ITE POWER TEST FEB 25

, SPs REQUIRED FOR HEAT UP N@ FEB 17

- RCS HEAT UP . .. . ... .l . . . ........ 7FEB 29 O

HEAT UP SPs TO NOP/NOT . . . .. ....... I...... . . . . . > MAR 4 RCS AT HOT SHUTDOWN COND .I.....

. . . .. .. .... ........ ..... VMAR 5 RCS HYDRO /hESSV 6RCAVON . . .

...... .1. ... ...... . . . . . . = = MAR 9 LOSS OF INST AIR TEST . . . ... . ... .!...... .... .. .... . O MAR 6 PRESSURIZER /EMOV TEST . . .. . .....

.[..... ....... ...... - MAR 10 ERC HOT FUNCTIONAL TEST .l

. . . .. ...... .. ..... . ........ . . . - MAR 13 LOSS OF NNI/1CS POWER TEST . . ... ..... . 3. . . . . .... . . . ... . . . . . . - -

nRE PUWP CONTROLLER --mudmuu.mummuula FEB 23 MAR 16 APPENDIX R WOOS FEB 121 REACTOR START UP . . .

.l.....

... ...... ..... . ... .... ....... .... .V I MAR 20 l

I I

M ACTUAL I I SCHEDULE Figure 2

i r i Rancho Seco has rewritten approximately 80% of its surveillance procedures during the outage, and experience has shown that adequate debugging of new procedures requires that they be performed as much as possible on actual systems by permanent plant staff. This was not a management goal. At this time, however, it appears that the licensee is conducting a concentrated review of new surveillance procedures that includes executing each procedure where .

plant conditions allow at least once before restarting the reactor (refueling '

interval surveillance excluded). The outcome of this effort will affect how many problems (personnel error, spurious actuations) will result from routine surveillance during restart and power acension.

The Operations Department Management Action Plan appears to address the 1 NUREG-1275 issues such as the need to stress the importance of details, the i need for discipline in following procedures, the need for awareness of plant '

conditions and regulatory requirements associated with these conditions, tight coordination throughout the plant staff, and the need for expedited resolution  ;

of problems. However, it is also recognized that these efforts to reestablish an operating mentality have been implemented late in the Rancho Seco restart t program. i 4.2.2 Lesson ,

Conduct a deliberate, evenly paced, thorough and well-planned preoperational '

j and startup test program.

]

Conduct thorough reviews and dry runs for planned testing and allow

, time for additional testing during either the preoperational or startup testing program. Emphasize planning to reduce the frequency .

! of unplanned scrans and unnecessary ESF actuations. A detailed '

l review of operational experience of similar plants should be a principle guide to the areas needing additional attention.

4 Minimize the number of deficiencies and outstanding items carried forward. Establish a policy of ccmplete resolution before proceeding.

Discussion i 1 l The degree of pre-test planning varies in proportion to the magnitude or complexity of the test. At a minimum, plant staff are required to complete i AP.82, Enclosure 8.1 "Test Briefino Checklist," before the perfomance of any i Special Test Procedure (STP). The checklist includes a pre-briefing l requirement for a walkdown of systems and equipment affected by the STP, and numerous items such as possible malfunctions and recovery for discussion during ,

l the test briefing. i l

l Plant staff indicated that dry runs can be used for major integrated test i

! evolutions. Dry runs were conducted for the loss of offsite power (LOOP) test, and will be conducted for the remote shutdown test after the 25% reactor power )

trip. Other major evolutions planned during the testing program include: loss of instrument air test, EFIC hot functionel tests, and loss of NNI/ICS power i test before reactor start up; reactor trip /EFIC post heatup tests after i reactor startup (see Figures 3 & 4). Dry runs for these tests are not built 4

into the schedule, but may be performed if all involved personnel have time j available, i

! 1

o R

) e TT OY E RN I CU WO I 2 AA EC OS PN 6 1 RITDC E 1 G I 4 ORNS I

[ TCAA i

A I .

~

S S T I GEE gR P

I 1 N O

G I T

A G

I S

N I

l R

U S P

0

~

T T D L E GIS G 5 4

S 2 I

R S

K V E ISP P -

E E WL I E

E L

E 0 P?5 5 00iC2 l

OA W R 09 7 *.

• N 2 PN U O ) S I ! 3 S N NT NC I

( E R

W /N SU CFG I E

E R

/

P E

O SI T

F ON TI N H T

R U F D S E O HU T S T 1

Y T

A 5 T LSC I S L

E R

E P F4FFF 3

U AOF TLEI C T

A -

1 M

E 5100022 H O-T -- M I

T 34455 533 _

X _

S a l O R

P ABCOE _

T I P

A O

H )

I I

j/

I I

I _

3 _

e _

E 0 C B A r u _

g _

i PL3N CONDITION F .

' C. _

(4 81 .

RANCHO SECO POWER ASCENSION SCHEDULE OVERVIEW 1988 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC I I I f f I I if f I I I I I I I I I I I I I I il I I I I if f f I !,

E REACTOR STARTUP . . .. .V 3/20 E LOW POWER TESTING . ..

i i 4/5 E 25% POWER LEVEL -

.... . O 4 /11 i

E REACTOR LOW POWER TRIP .V 4/11 REMOTE S;I._ 70WN TEST AFTER 25% POWER TRIP

, E 2nd Rx STARTUP & l l l l 40% POWER PLATEAU . . . .s 1 6/10 OPERATOR TRAINING

' E MGMT APPROVAL TO INCREASE POWER LEVEL . .. .. . . . .. .V 6/10 1

5 65% POWER PLATEAU .. .

. . .i i 7/16 OPERATOR TRAINING MAIN B MGMT APPROVAL TO "" P F RM ANCE l

• INCREASE POWER LEVEL - . . .. . .... . . .V 7/16 I

E 80% POWER PLATEAU . . . . i i s/21 cPER. TRNG ICS TUNING E MGMT APPROVAL TO PERFORM REACTOR TRIP . . . .

E REACTOR TRIP AND .Vla/21 EFIC POST HEATUP TESTS . . . . .

.V s/22 3 3rd Rx STARTUP TO 92% POWER PLATEAU .. .

. . . V s/29 5 92% POWER PLATEAU . . . . . . .

1 9/2s E FULL POWER . ........

. V 10/9 ACTUAL i iSCHEDULE Figure 4

The Rancho Seco staff has devoted considerable effort to avoiding problems with EFIC by using operational experience from other B&W plants, e.g., an I&C engineering supervisor witnessed the EFIC startup testing at Arkansas Nuclear One; and an I&C engineer and a technical training instructor were sent to Crysta! River 3 in conjunction with initial EFIC operation at that facility.

In particular, Crystal River 3 experienced a large number of spurious EFIC initiations and overcooling events. Use of this experience has resulted in training Rancho Seco operators (via the EFIC simulator) to expect different response times from EFIC level control; in emphasizing the need to place an EFIC channel in maintsuance bypass before working on it; in a review of the need for a delay on low level initiation (decided not to incorporate this specific delay, but variable delays exist on other setpoints); and in the need to change the SG level operating setpoint (decided this was not needed due to hardware differences in SG level instrumentation).

The lesson on minimizing the number cf deficiences carried forward and striving for complete resolution applies both to known equipment deficiencies carried into the startup program and to equipment problem resolutions which occur during the startup program itself. The former includes such things as corrective maintenance backlog and systens and equipment deficiencies uncovered by cold shut down testing when resolution might be postponed. The 7atter encompasses determination of causes of unplanned events, in parti:ular root cause determination of unplanned reactor scrams SMUD has developed a Restart Scope List (RSL) and a long Range Scope List (LRSL) tiat together provide accounting for all open items, both managerial /

atministrative and system-related. Those items wnich must be accomplished prior to restart are on the RSL. The items on the RSL and their status are reveiwed periodically with the Region, and any addition or deletion from the RSL is also reviewed by the Region. Figure 5 shows the status of the RSL and LRSL items as of February 11, 1988, when 54 items remained cn the RSL.

According to the licensee, the corrective maintenance backlog has baen reduced from about 5000 work requests in December 1987 to approximately 1200 as of February 15, 1988. The stated management goal is to reduce the backlog to as ,

low as reasonably possible by restart of the reactor. The plant staff are )

coordinatino with NRC regional staff and resident inspectors to justify any '

work requests that are not completed by restart. About 95% of these work requests were in the non-restart category as of mid-February.

A preventative maintenance (PM) program has been established with a goal of ensuring that all of the PMs on Category 1 equipment are current at restart.

Category 1 encompasses 3,925 I&C components, 1,685 electrical components, and 600 mechanical components.

During a power ascension progrdm there can be schedule pressures that work to circunscribe the time and effort spent in unplanned event followup and root cause determination. Thus, the true root cause may go undatected resulting in recurrence. This leads to questions about the time allotted for the restart and power ascension, the management attitude toward thorough resolution of problems, and programs for problem resolution and event cause determination.

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

( (53533 .

~

SCOPE LIST STATUS (AS OF 02/11/88)

RESTART SCOPE LIST TOTAL CLOSED OPEN MANAGEMENT ITEMS 720 706 14

, SYSTEM RELATED ITEMS 369 329 40

?

RSL TOTALS 1089 1035 54 i

LONG RANGE SCOPE LIST (POST-RESTART) TOTAL CLOSED OPEN MANAGEMENT ITEMS 1535 317 1218 SYSTEM RELATED ITEMS 545 95 450 LRSL TOTALS 2080 41 2 1668 i

Figure 5

The power ascension schedule from reactor startup to full power is scheduled to last roughly six months (see Figure 6). This program includes one week testing at both the 0 and 25 percent power plateau, eight weeks at the 40 percent power plateau, and 5 weeks each at the 65, 80 and 92 percent power plateaus. As indicated by plant staff, the durations at various plateaus are driven in large part by the plan to provide extensive operation training on such things as main feed pump performance.

Plant Operating Procedure B.1A, "Control of Heatup and Operating Procedure for Power Escalation Testing," controls the progression of startup and integrates the operating, special test, and surveillance procedures required for startup.

The procedure contains hold points where management from all departments and the CE0-nuclear are required to certify readiness to proceed (raise power to the next plateau or to perform a planned scram). The procedure also requires the generation of a Restart Punchlist Form (RPF), whereby, operational concerns considered significant by Operations Management that are discovered during testing or completion of normal plant operations will be documented.

The RPF addresses issues which require management &cisions and provides high visibility to the decision process.

The Director, Nuclear Operations and Maintenance or his designee is responsible for reviewing each RPF and determining the actions necessary to address the issue raised. When action is completed the Director reviews the disposition, but is not required to note his approval and closure through signature (this is at his discretion).

Ranche Seco staff offered their recent decision to suspend all testing as an indication of managements intention to follow a policy of full resolution before proceeding in the face of schedule pressures.

At some new plants, the determination of root causes for unplanned events was hampered by not taking advantage of the expertise of the entire staff. For example, post-scram root cause determination was carried out by operations personnel without consultation with engineering staff. On this point SMUD noted two features of their program that would counteract this. First, the Plant Support Group in the Nuclear Engineering Department has a function of providing on-call around the clock assistance in problem resolution. Second, SMUD is employing a systems engineering approach in which engineers currently responsible for modification and testing of systems will remain so throughout power ascension, but will eventually transition into the permanent engineering staff in the Plant Performance Department. Both of these features were cited as strengthening the communication between Operations and Engineering.

A major area where root cause determination is essential is for unplanned scrams. Procedure AP.28, "Post Trip Transient Report," covers the root cause determination process and specifies who has authority to restart the reactor.

The responsibility for ceuse determination resides with the Shift Supervisor or an SR0 designee, and an STA.

If as a result of the investigation the scram is classified as a "Type I" event, i.e., a simple trip with "expected" plar.t response and no complications, then the Shift Supervisor can recommend to the Plant Superintendent that the plant be restarted. The decision rests with the Plant

i POWER ASCENSION TESTING i 1

% / FULL POWER

• I ~92%

y so (5 wks) p 70

/ ~s7%

(5 suks) - i

, O go I ~g5x _

~ W l (5 wu) _

E so R

[ 0

~40%

m l (8 wh) - ,

~

E L to gx P!anned Trip and Remote Hot Shutdown Test Planned Trip Test o } (t wk)

Criticality 0%

(1 wk)

=

=

APPROXIMATELY SIX (6) MONTHS DURA 110N I

E 28 ADDITIONAL TESTS i RESULTS REQUIRED

- ICS TUNING

- PERSONNEL / PROCEDURES

CHECKOUT i

Figure 6 2

+ _ . . . - _ -. . .

Superintendent. For other more complex trips (Type II), the Operations Superintendent and Engineering and Quality Control Superintendent are contacted and direct a further investigation. The procedure for Type II '

events contains the note:

NOTE Sources of expertise that should be considered include nuclear steam supply vendors, vendor engineers, onsite engineering staff, corporate engineering staff, and other experienced operations and maintenance personnel.

For Type II events, the restart decision is still left to the Plant Superintendent, who may at his discretion convene the Plant Review Conmittee (PRC) to review the post-trip investigation prior to restart.

Conclusions Rancho Seco appears to have in place procedures and programs that appropriately address the issue of a controlled, evenly paced test program. They have also taken advantage of other cperational experience to determine the need for additional testing, e.g., EFIC.

Decisions on what items can be carried into the restart program have been reviewed on a continuing basis by NRC inspectors. The licensee has also indicated a goal of minimizing the number of corrective maintenance actions that will be open by the time of restart. This satisfactor'ly addresses the concerns about proceeding with restart with a large number of known open items which could distract staff form the restart and power ascension itself.

The committment to a thoroug'i *oot cause analysis of unplanned events during restart and power ascension st ald be strengthened. Two places where this can be done are in the close out of RPF items and in post transient investigation (Procedure AP.28). The Director, Nuclear Operations and Maintenance should formally note approval and closure of RPFs through signature. More importantly, we suggest thbt the licensee modify the procedure for post trip review (and by extension the review for all reportable events) to more fully ensure that the root cause corrective measures are fully defined prior to restart, implemented in a timely manner, and that the appropriate range of plant staff and management is involved in the analyses and decision-making.

During the remainder of the current fuel cycle all transients could be considered as Type II events. This will help ensure that full advantage is taken of the total staff expertise, and that narrow diagnosis of "simple" unplanned events will not occur.

4.2.3 Lesson Use the finalized Technical Specifications (TS) to generate and validate (e.g., against the as-built plant) surveillance testing procedures as early as possible. In this regard, great discipline should be exercised to restrict the number of last-minute changes in the proposed TSs. Once final draft TSs are issued, the licensee should begin to incorporate TS requirements into plant procedures instead of waitina until the last few changes have been implemented. In conjunction with this activity, have plant staff (as opposed to NSSS vendor or special startup group) perform all surveillances.

e Discussion Plant staff indicated that only three TS revisions rerrain to be approved:

proposed amendment (PA) 138 (organization), PA 155 (RETS), and PA 164 (TS upgrade). SMUD f ndicated that both the radiological effluent TSs and the TS upgrade, i.e., to be equivalent with the B&W Standard Technical Specification, approvals will have the greatest impact on restart since all related proccdures have been revised to reflect the proposed revisions and are currently being utilized in preparation for their required implementation.

< The staff stated that the organizations responsible for procedures have been provided with all proposed TSs, and all procedures written do conform to the revised TSs. Procedures which do not comply with the present TSs are being issued and implemented based on proposed TS amendments. These procedures are re-reviewed for conformance upon issuance of the TS amendment.

All surveillances are performed by plant staff, according to the staff.

The status of surveillance pr(cedures was previously discussed under Section 4.2.1.

Conclusions It does not appear that the restart will be complicated by having to deal with last minute TS changes.

4.2.4 Lesson Improve administrative control of surveillance. For example:

o Since problems have been experienced when work has been performed in the vicinity of instrument racks during plant operation, licensees should evaluate the location and nature of work activities during operation in terms of adverse effects on plant operation and take j appropriate administrative actions, l 1

o Implement schemes to separate channel testing, such as a specific l day of the week assigned to work on each channel, and to identify i the channel in test, such as posting on control room panels. 1 o Blend engineering staff into the I&C organization, o Flao, categorize, and schedule surveillance according to risk of l scrams or other ESF actuations. '

o Organize the I&C staff to establish accountability for specific equipment.

Discussion l

Rancho Seco's response to NUREG-1275 indicated, in part, that, "Administrative .

Procedure AP.44, "Plant Modifications - ECN Implementation," describes the l interface between various organization on site whc are impacted by a design change. It requires that all affected groups participate in a walkdown of the

job before implementation. At this time, any interference with, or adverse impact on other systems is identified." However, our review of AP.44, Revision 13, dated January 6,1988, indicates that walkdowns are only required; 1) during plant modification implementation planning (Section 6.1),

2) during interface planning on a case-by-case basis if exceptions to Section 6.2.2 exist, 3) prior to plant modification turnover (Section 6.3), and 4) prior to ECN closecut (Section 6.6). Definition 3.7, "ECN Walkdown," also indicates that walkdowns are only conducted for ECN planning, ECN turnovers, and ECN releases as necessary. There appears to be no requirement to walkdewn the area prior to construction / modification start to ensure operating equipment protection, e.g., addequate : " ' Jing is available that would prevent climbing on sensitive equipment, s intended by the NUREG-1275 recommendation.

Rancho Seco maintenance activities are addressed in Maintenance Administrative Procedure MAP-006 "Work Request Planning," Section 5.3.3, which requires that the discipline planner walkdown the work site during the planning of corrective maintenance work requests. This walkdown should identify any problems relevant to the NUREG-1275 findings.

SMUD indicated that Rancho Seco Administrative Procedure RSAP-0803, "Work Requests," requires that work on power block and safety-related systems be reviewed by the Shift Supervisor and approved prior to starting work. His s

operating experience and plant knowledge should be helpful in identifying problems of this type during his review. However, our review of RSAP-0803, Revision 3, dated November 11, 1987, indicates that a physical walkdown of the area by the shift supervisor is not required. Therefore, unless he physically observes the work area, he could not readily predict problems such as those noted by previous operating experiences, e.g, climbing on control cables resulted in a plant scram.

SMUD indicated that separate surveillance and preventive maintenance procedures are written for individual instrumentation channels which provides the separation of channels desired, thus reducing the risk of scrams from multiple channel testing. Rancho Seco Administration Procedure AP.48, "Safe Test Authorization Procedure," Revision 4, dated August 21, 1986, also requires that prior to any work, maintenance, or testing being performed on plant systems, the Shift Supervisor (or his designated representative) and the responsible Job Supervisor shall review the necessity for a Test Authorization. This review shall determine whether a Test or Clearance Authorization is required to ensure protection to individuals and equipment.

This review shall also determine any special limiting conditions applicable to the work to be performed. These controls should ensure that no multi-channel or cross-channel testing is performed by controlling which surveillance procedure is authorized for performance and when.

SMUD also indicated that after restart the bulk of the surveillance testing and preventive maintenance work will be performed on back-shifts when interference from other activities is minimized.

Rancho Seco currently has a Maintenance Department I&C engineering organization, however, plans are underway to consolidate all engineering functions in the nuclear departments to prevent overlap and effort duplication. A systems engineering group within the Plant Performance Department will include I&C engineers who will provide nc~ssary IAC maintenance group support.

Surveillance procedures involving channel / subsystem testing of the RPS, SFAS, and EFIC that have the potential of causing a reactor trip have been identified according to plant staff. These procedures have been reviewed and walked down (per SPIP recomrrendation TR-111-RPS - a B&WOG recommendation) to ensure that statements are included in the Limits and Precautions or Prerequisites Sections to minimize the risk of a reactor trip. SMUD's initial response to NUREG-1275 indicated that inadequate precautionary statements would be revised with no further clarification as to their schedule for completion.

A subsequent discussion with the Rancho Seco Plant Performance Department Manager indicated that after they made this initial response, they reconsidered the possibility of performing a new procedure during operations and causing an inadvertent scram. Therefore, lead people, e.g., STAS, have been assigned to ensure that new procedures are debugged (refer to Section 4.2.1 for additional details on debugging new surveillance).

Accountability for cutain complex I&C equipment is accomplished at Rancho Seco via one 180 crew (six people) specifically assigned to conduct surveillance testing, corrective maintenance, and preventive maintenance on the following systems: Radiation Monitoring, Meteorological, and Security.

This crew does not rotate on shift, thus allowing them the time they need to test and repair these very complicated systems. The remaining crews have non system specific experience and knowledge, and provide 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> coverage on the remaining I&C equipment.

Conclusions Rancho Seco has policies and procedures that are generally responsive to the NUREG-1275 lesson on administrative control of surveillance. However, the AP.44 and RSAP-0803 procedures could be strengthened by requiring actual review of a planned work area prior to beginning work.

4.2.5 Lesson Give high visability to the source (i.e. , organizational element) of unplanned scrams (and other unplanned events) caused by human error and establish performance goals.

Ensure that operating experience feedback programs: (a) combine internal events and relevant events from similar plants, (b) communicate them directly to the appropriate first level supervisors and working level staff at the l plant on a periodic basis, including prior to startup, and (c) address l preventive measures. For example, segregate the trip and ESF actuations data l involving human errors from recent plant startups into the specific positions, organizational or functional element, working activity, systems and components, time of day, etc. Feed this information back at the icwest levels so that the experience of others, the complexity of what is being done, and the ramifications of errors can be seen and appreciated.

Discussion Rancho Seco's Administrative Procedure AP.50, "Operational Assessment (Operating Experience) Program," Pevision 3, dated May 15, 1987, partially implement.e item I.C.5 of the TMI Action Plan, NUREG 0737, regarding procedures

for evaluation, action, and feedback of operating experience originating both within and outside of SMUD. The complete SMUD program consists of this procedure and: 1) Nuclear Licensing's feedback from vendors and the NRC; 2) feedback utilizing plant trip evaluations per AP.28; 3) feedback utilizing the Rancho Seco's Plant Review Committee review of Rancho Seco LERs; 4) feedback utilizing Operations Special Orders ;,er AP.24; and 5) feedback to the industry by input to Nuclear Network from any Rancho Seco plant supervisor. AP.50 provides for:

o evaluation of events classified by INP0 as "significant",

o evaluation of events classified by selected plant staff as "significant",

o determination of the applicability to Rancho Seco, o development and distribution of proposed actions, o tracking of the rejection of proposed actions or their transformation into planned actions, and o dissemination to Nuclear Managers of information uncovered by the process above for discretionary group training.

The program was originally carried out oy the STAS who report organizationally to the AGM - Nuclear Power Plant Production. This arrangement proved unsatisfactory siro : 1) the STAS had higher priority duties that interfered with the feedback program, and 2) a potential conflict of interest existed since the AGM - Nuclear Power Plant Production is also responsible for implementation of any actions generated by the feedback program. As a result the responsibility was transfered to Licensing under the AGM Technical and Administrative Services to ensure that the program is properly administered, each recomendation is tracked to closure, and an independent effectiveness evaluation of action followup is performed.

In addition to the feedback program described above, both the Maintenance Department and the Nuclear Quality Department have programs for trend analysis and root cause determination.

The Rancho Seco Training Department has their own program for obtaining documents for training references and as resource documents to be used in their evaluation of training commitments for implementation. Training Department administrative procedures TDAP-5100, Obtaining Documents for Action, "TDAP-5110, "Change Action Monitoring, Evaluating, & Implementing," and TDAP-5130, "ECN Implementation," have been established to route to training the )

range of information on plant experience and activities as well as l industry-related items, such as INP0 SOERs, SERs, O&M Reminders, Nuclear  ;

Network, NRC Inspection Reports, NRC Bulletins and Circulars, ANI Evaluation j Reports, Plant Incident Critiques and Reports, and B&W Bulletins. l l

One outcome of the the experience review by Training is the periodic l preparation of required reading packages for licensed operators. Each operator has a time limit for reviewing this material and must indicate compliance by signature. This mechanism provides direct feedback of experience to operators. However, no similar mechanism exists for maintenance (mechanical and electrical) and I&C staff to review experiences relevant to their specialties.

Change action, i.e., the activity, including documentation, evaluation and implementation of item (s) that affect nuclear training program development or revision, is a process whereby items are reviewed and evaluated by Training Deqartment supervisors and instructors to determine the value and training impact of each item. Emphasis is placed on identifying root causes of problems and applying the lessons learned to Rancho Seco's operating practices. When a' change action results in equipment or procedure changes, the training programs are revised to ensure the operators, technicians, and maintenance personmnel understand the basis for the change and how it affects their specialty. All change actions related to industry events are tracked. <

on the Training Departments Work Track system per Administrative Procedure TDAP-8120, "WORKTRAK System Employment."

Conclusions Rancho Seco's operational experiece feedback progran is generally responsive '

to the NUREG-1275 lesson. However, the program could be strengthened by providing direct experience feedback to maintenance and I&C staff in a way similar to that in place for operators.

4.2.6 Lesson A number of improvement lessons are directed at training as follows:

o Establish as a major goal an increased conmitment to training in performing surveillance testing, calibration, and troubleshooting activities well prior to operations. I&C training initiatives, such as repeated practice for those surveillance testing activities that could cause a transient and which should be conducted on actual in-plant equipment on live systems prict to operations, should be emphasized. An additional action to improve surveillance testing suggested by licensee staffs was training for I&C personnel in valving instrumentation in and out of service.

o Emphasize training for routine operations involving power level ,

changes and the associated conmunications among shift personnel i (i.e., feed flow and turbine evolutions) that have historically caused trips. Accelerated programs / efforts appear appropriate for newly licensed plants regarding steam generator level control.

Emphasize the need for site specific simulators to include, prior to ,

st6ctup, the best achievable fidelity of the simulator to the plant l regarding feedwater effects (lead / lag characteristics of level l indication and control methods), and include provisions to continue to improve fidelity as the startup progresses.

o Establish extensive, detailed training for all segments of the onsite plant staff, including ISC technicians, maintenance ,

mechanics, security staff, operations, and management. This  !

training would emphasize: (a) the applicability of the various TSs to the changing plant modes of operation and associated schedules, j (b) the relationship of the TSs to the plant procedures, (c) the NRC  ;

requirements for reportability of violations, and (d) the basis for l the TSs and discussion of LC0 requirements.

. . - . -=

. s Discussion Training for I&C technicians has been through Rancho Seco's continuing training program. Reportability, special training has been conducted on several modified systems, e.g., Emergency Feedwater Initiation and Control (EFIC), ICS, and Controlatron. Certain I&C technicians who either installed the new equipment or performed the initial surveillance are assigned to checkout nthers via a Restart Personnel Qualification Guide. This guide provides special training for problem areas and activities that have not been performed since the plant has been shutdown. On-the-job qualification cards are required to be signed off, and the maintenance department is required to ensure that qualified personnel are available to perform the work.

Accortiing to the Rancho Seco training manager, considerable effort has been expended -in training operators on new modifications. Over 400 hours0.00463 days <br />0.111 hours <br />6.613757e-4 weeks <br />1.522e-4 months <br /> of modification training and 240 hours0.00278 days <br />0.0667 hours <br />3.968254e-4 weeks <br />9.132e-5 months <br /> of simulator training per licensed operator has been completed. Their training is conducted at a B&W simulator shared by other B&W plants and apparently closely models the Rancho Seco design. The training maniger indicated that the most recent simulator training consisted of one solid week at the end of last year. This consisted of training on normal startups and shutdowns as well as abnonnal ICS operation. One crew's unsatisfactory performance resulted in their returning for additional training.

The simulator had been modified to incorporate the new AFW modifications (e.g., a new EFIC panel was added). EFIC's response characteristics were modeled, thus providing the operators heads up experience as to what to expect. The training manager indicated that preventing operators from taking j manual control of feedwater during certain abnormal operations would be the '

most difficult challenge due to EFIC's slow response time. The operations manager indicated that the Rancho Seco turbine bypass valve (T8V) response was different, i .e. , the TBVs were not modeled exactly. However, the differences are recognized and discussed during simulator training. This simulator training was evaluated by both SMUD management and INP0. SMUD's long-range i plan is to obtain a site-specific, plant-referenced simulator by mid-1989.

Specific operator hands-on training for field items, such as emergency feedwater initiation controls, auxiliary feedwater valves, turbine bypass valves, and atmospheric dump valves is being conducted according to plant staff. A power ascension program with plateaus reserved for operator hands-on training has been proposed.

The present operator training programs at Rancho Seco reportedly provide extensive training on watchstanding principles to non-licensed operators, reactor operators, senior reactor operators, shift supervisors, and shift technical advisors. These programs cover crew communications, use of procedures, reporting responsibilities, and other watchstanding requirements, and have been accredited by the National Nuclear Accrediting Board in April 1986.

As dise.ussed previously in Section 4.2.1, even with all the above emphasis on operator training, recent areas for operations improvement have been identified. Recent events such as water spills and procedural noncompliance have led to conduct of operation concerns. For exemple, a February 3, 1988,

. i event that occurred during the Loss of Offsite Power Test (LOOP) performance was caused from a mispositioned switch. In addition, the licensee discovered three additional problems during the LOOP test involved another mispositioned switch. These problems resulted in suspension of further startup testing.

The Rancho Seco operations management has recently developed an "Operations Departirent Management Action Plan" to resolve errors which are attributable to personnel.

The licensee's response to NUREG-1275 did indicate extensive training in the technical specifications and NRC requirements areas, however, recent operations personnei problems have been identified in these areas. The recently implemented Operations Department Management Action Plan does recognize attention to detail, operations surveillance procedures and work planning, and equipment status control as causes for some of their operational people problems. Some of the corrective actions in these areas to be completed by heatup, for example, have been: 1) to revise the Work and Test Authorization Procedure to track the status of Technical Specification Limiting Conditions for Operation (LCO) work in progress, 2) to provide the control room on a daily basis, a current dynamic schedule for scheduled and overdue surveillances that are reouired to be performed, and 3) better define and verify systems status files.

Conclusions The operator training for normal operations appears to be a strong point, both in terms of simulator training and plans for extensive training at various plateaus during power ascension. However, the Operations Department Action Plan indicates that improvments in shift connunication and procedure changes to improve awareness of plant status are needed.

4.2.7 Lesson Focus on the 80P prior to operation and early in the life appears to provide a high return regarding the reduction of unplanned scrams and ESF actuations.

Within this area, attention could be given to:

o Conducting additional reviews of feedwater and turbine control and bypass systems to identify sensitivities and plant-specific characteristics that could contribute to transients or the ability of the system to cope with expected transients.

o Conductina setpoints on acomponents systematic such review asofpumps equipsuction (ment-protective logics and trip, time delay, vibration trip) or power supplies to identify areas where a time delay or additional channels for coincidence could reduce the potential for unnecessary transients for spurious actuations. Give special attention to first-of-a-kind features not incorporated in earlier designs. Additional examples obtained from the plants visited include the main steam reheater drain high level trip and other turbine protective trips.

Discussion The licensee's response to NUREG-1275 indicates that many improvements are being implemented prior to restart. These improvements apparently resulted either directly or indirectly from Rancho Seco's PP&MIP (refer to Appendi < C fordetails). Certain areas or sources of the PP&MIP's Systematic Assessment Process appeared to-have been geared to 80P improvements, e.g. , precursor reviews, deterministic failure consequence reviews, and plant staff interviews.

The precursor review's intent was to determine 'f any unidentified precursor conditions exist at Rancho Seco that are capable of causing needless reactor trips, unsatisfactory post-trip responses or will induce challenges to safety systems, i.e., conditions at Rancho Seco or elsewhere that caused or could have caused an adverse safety situation to develop, and for which appropriate preventive measures should be taken to protect against the occurrence of that or a similar event or circumstance at Rancho Seco in the future. A total of approximately 1,396 primary documents, e.g., all Rancho Seco LERs, NRC Bulletins and Generic Letters, and Significant Operating Experience Reports (SOER) issued by INP0 back to 1978, and Rancho Seco Trip reports back to 1974 '

were reviewed. An estimated 3,000 secondary or supporting documents, e.g.,

NUREGs, Rancho Seco Occurrence Description Reports (0DR), and NRC Information Notices, underwent prescreening and selected documents were reviewed. A total of 457 recommendations were generated and categorized in the following areas:

o Major secondary plant systems, o Reactor systems, o Control sptems, o Emergency core cooling systems, o Support systems, o Misc secondary plant systems, o General administrative and building items, o Generic valves, and o Misc. recomendations.

The deterministic failure consequence review's intent was similar to that of the precursor review except a systematic method of looking at Rancho Seco's response due to component failures was utili7ed based on failure analysis versus precursor reviews. A deliberate look at Class 1 safety-related systems l was initiated, however, some non safety-related systens were also analyzed. l The loss of electrical power, instrument air, and ICS/NNI control failure modes were evaluated to determine impact upon operations. As an example, during the loss of electrical power analysis each breaker off the motor control center (MCC) or panel was "failed" individually, and the consequences of each failure determined. This process was repeated for all MCCs and panels ,

off a ccmmonn 480V bus. Once individual load losses were evauluated, the loss l of the source was analyzed. The guidance available to the operator to address the failure was then evaluated for adequacy, and recommendations to correct any plant physical or procedural inadequacies were incorporated and tracked within the PP&MIP. Approximately 1,293 recommendations were generated from i which some 15% addressed 80P problems and were categorized under the following systems / areas:

4 i

3 s

o ICS/NNI (9) o InstrumentAir(234) o Vital Bus o Electrical Distribution o Seal Injection Makeup o IE Bulletin 79-27 (3)

The plant staff interviews were conducted with, e.g., plant operators, operations technical support engineers, plant engineers and technicians, training instructors, design engineers, and quality department personnel.

These personnel were encouraged to identify system, component, or operational related problems, or concerns which they were aware of. Some 130-140 different personnel were interviewed resulting in approximately 1,600 recommendations divided among all areas from safety-related, non safety-related, or 80P systems / components, down to management issues. The program was mora 80P oriented than the others discussed above.

Conclusions Rancho Seco has made extensive use of operational experiences in seeking ways to 'mprove B0P 9erformance. Their review fully satisfies the intent of the NUR;.4-1275 lesen.

4.2.8 Lesson Install test jacks and bypass switches at appropriate points in actuation circuitry.

Discussion Test jacks and bypass switches during surveillance activities has been j identified by SMUD as a recommendation for further action based on their -

review of NUREG-1275.

Conclusions The NUREG-1275 lesson was one specific example of a method for reducing surveillance related unplanned actuations. Such modifications may not be necessary in all cases. Thus, Rancho Seco's response meets the intent of the NUREG-1275 lesson.

4.2.9 Lesson Implement on a priority basis vendor or licensee trip reduction measures.

Licensee trip reduction programs should focus on safety-related equipment as well as on 80P equipment.

Discussion Rancho Seco has adopted the B&WOG Safety and Perfonnance Improvement Program (SPIP). Figure 7 is a summary of the SPIP at Rancho Seco. The SPIP items are divided into "key" and "non-key" items. Key items were those that the B&WOG classified as the most important and beneficial in enhancing plant l

n n kiin B ain & & E E E E E E & E G E & &

( ' "l$ D SAFE.TY AND PERFORMANCE

~

IMPROVEMENT PROGR AM (SPIP) SUMM ARY a 207 ITEMS ISSUED BY B&WOG FOR CONSIDER ATION

-74 KEY ITEMS; 133 NON-KEY ITEMS a 150 ITEMS APPLICABLE TO RANCHO SECO

-66 KEY ITEMS; 84 NON-KEY ITEMS E 57 ITEMS ALREADY IMPLEMENTED BY RANCHO SECO

[ -33 KEY ITEMS; 24 NON-KEY ITEMS w

s 22 ITEMS PARTIALLY COMPLETE

-16 KEY ITEMS; 6 NON-KEY ITEMS a 71 ITEMS ARE BEING FURTHER EVALUATED FOR IMPLEMENTATION AFTER RESTART

-17 KEY ITEMS; 54 NON-KEY ITEMS a 8 NEW ITEMS ADDED JANU ARY 1988 BY B&WOG Figure 7

- . _- _ . . . = - - - . . . _ -

1 a '

a safety, and should be implemented first. Out of the 66 key items determined applicable to Rancho Seco, 58 will have been implemented prior to restart.

Each item is tracked to closure via the Rancho Seco Action tracking system.

Conclusions Rancho Seco appears to have aggressively committed themselves at the onset to implementing the B&WOG recommendations. The intent of this NUREG-1275 lesson appears to have been met.

4.2.10 Lesson Pay attention to the design and installation of equipment located ir the vicinity of radiation monitors and associated cabling to ensure that adequate grounding of equipment, cable shielding, etc., are provided to prevent the occurrence of EMI, which can trigger this extremely sensitive instrumentation.

Discussion In response to NUREG-1275, the licensee indicated that the following precautions have been taken to limit the effect of electromagnetic interference (EMI) on radiation monitors:

1. The analog signal cables have been routed separately from AC power and control circuits.

2. The installations have been designed based upon design guide Nuclear Engineering Procedure NEP 5204.43, "Instrumentation Systems, Shielding and Grounding, and Surge Protection."

3. The manufacturer's recommendations for limiting the effect of EMI have been followed.

Conclusions Rancho Seco appears to have taken steps that would prevent EMI type inter-ference on components that have historically caused unplanned actuations. The intent of this NL' REG-1275 lesson appears to have been met.

4.2.11 Lesson Thoroughly test new or unique plant features, such as new RPS systeins, electrical systems, etc., prior to fuel load to reduce unanticipated failures or unexpected erratic behavior. Emphasize planning to reduce the frequency of unplanned scrams and unnecessary ESF actuations.

Discussion The EFIC, TOI Diesel, Control Room / Technical Support Center HVAC, and Post Accident Sampling Systems are new systems installed at Rancho Seco.

The EFIC System, for example, will be fully preoperationally tested. The Cabinet Functional Test (STP.665), the Cold Functional Test (STP.666), and the Cold Preoperational Test (STP.667) have been performed. A Hot Functional Test (STP.1113) and a Post Critical High Decay Heat Test (STP.668) will be performed. EFIC will also be tested during the Loss-of-Instrument Air Test (STP.1056)andtheloss-of-Offsite-PowerTest(STP-961). Operators have been trained on the simulator on the new EFIC panel.

The TDI Emergency Backup Diesel Generators, for example, were fully preoperationally tested in STP.1009A/B, preceded by 27 Special Test Procedures for individual component and sub-system functions. These diesels and associated equipment will undergo post modification and integrated system phase Il retests (STP.1134A/B). They will also be tested during the Loss-of-Offsite L Power Test (STP.961).

As indicated previuusly in Section 4.2.1, testing on the HVAC and Post Accident Sampling System were completed satisfactorially per Rancho Seco's restart milestone sumary.

Conclusions The extensive review of operational experience, the resulting engineering changes, and the related pre-startup testing (component and system level tests) and controlled power ascension programs should eliminate many of the unplanned events that stem from new or unique plant features. The intent of this NUREG-1275 lesson has been met. ,

4.2.12 Lesson Incorporate scram prevention measures such as:

o Develop a color coding scheme for single point scram components whose misoperation could cause a scram (for example, pressure sensing lines).

o Install cages or covers over switches or racks that could provide trip signals.

Discussion ,

Scram preventative measures (color coding scheme for components) has been identified for further action. SMUD also indicated that 207 toggle switch guards have been installed on panels /switchgear to protect from inadvertant ,

trip signal generation. '

Conclusions The NUREG-1275 lesson provided specific exarrples of methods to prevent scrams based on human factor considerations. SMUD, based on their PP&MIP, has I indicated a cemitment to reduce inadvertent scrams. SMUD's response appears '

to be responsive to the intent of this NUREG-1275 lesson, l

l l

i

~ - - - - - - - - - - - - - - - - , - - , - - - - , - -

y . .at_= _ , - ,a. . 2 ,

0 APPENDIX A REFERENCE DOCUMENTS

REFERENCE DOCUMENTS

1. SMUD/NRC (NRR/ Region V) Restart Meeting, February 17,1988 (Briefing Slides).

2. Rancho Seco Operations Department Management Action Plan, Revision 0, February 12, 1988.

3. NUREG-1286, "Safety Evaluation Report Related to the Restart of Rancho Seco Nuclear Generating Station, Unit 1, Following the Event of December 26, 1985," October 1987.

4. SMUD/NRR Meeting, September 23, 1987 (Briefing Slides).

5. Rancho Seco New Surveillance Procedure Completion Master Tracking List, Mode 5 Procedures (Hea tup), February 9,1988.

6. Rancho Seco Administrative Procedures:

AP.4A, "Safe Clearance Procedure Danger Tags," Rev. 5, November 14, 1986.

AP.48, "Safe Test Authorization Procedure," Rev. 4, August 21, 1986.

AP.4C, "Caution Tag Authorization Procedure," Rev. 2, June 11, 1986.

AP.18, "Plant Housekeeping and Inspection," Rev. 2, April 22,1987.

AP.23.00, "Conduct of the Operations Division," Rev. 0, March 6,1987.

AP.23.01, "Organization and Scope," Rev. 2, January 13, 1988.

AP.23.02, "Shift Turnover," Rev.1, November 19, 1987.

AP.23.03, " Logs and Rounds," Rev. 1, January 25, 1988.

AP.23.04, "Shift Routine," Rev. 1, November 3, 1987.

AP.23.05, "Control Room Activities," Rev. 0, March 6, 1987.

AP.23.06, "Operation Procedures," Rev. 1, July 3, 1987.

AP.23.07, "Control of Special Order," Rev. O, March 6, 1987.

AP.23.08, "Reporting / Notification," Rev.1, October 2,1987.

AP.23.09, "Coninunications," Rev. 0, March 6,1987.

AP.23.10, "Fquipment Maintenance and Operating Standards," Rev. O, March 6, 1987.

AP.23.11, "Clearances, Te:;ts, and Cautions," Rev. O, March 6,1987.

AP.23.12, "Control of Operator Aids," Rev. O, March 7,1987.

AP.23.14. "Control of Wquipment and System Status," Rev.1, November 3, 1987.  !

AP.23.17, "Independent Verification," Rev. 0, March 7, 1987.

AP.23.18, "Working Hours, Vacation and Sick Leave," Rev. O, March 7, 1987. 1 AP.23.22, "Simulator Training," Rev. 0, March 7, 1987.

AP.23.23, "Operations Advisor Program," Rev. O, March 6,1987.

AP.23.24, "Shift Operations Responsibilities During Special Testing,"

Rev. 0, May 15, 1987.

AP.23.25, "Shift Tecf.nical Advisor Transient Evaluation Guidelines,"

Rev. 0, July 17, 1987.

AP.28, "Post Trip Transient Report," Rev. 9, April 11,1986.

AP.44, "Plant Modifications - ECN Implementation," Rev.13, January 6, 1988.

AP.50, "Operational Assessment (Operating Experience) Program," Rev. 3, May 15, 1987.

A-1

s Appendix A con't AP.69, "Nuclear Network Usage at Rancho Seco," Rev. O, May 15, 1987.

AP.82, Enclosure 8.1, "Test Briefing Checklist," Rev. 4.

AP.303.04, "Cross-Indes of Technical Specifications and Surveillance Procedures," Rev. 0, January 16, 1987.

AP.303.05, "Technica Specification'to Surveillance Procedure Cross Reference Data Base Update Guideline," Rev. O, December 4, 1986.

7. Training Department Administrative Procedures:

TDAP-5100, "0btaining Documents for Action," Rev. 2, June 12,1987.

TDIP-5110, "Change Action Monitoring, Evaluating, & Implementing," Rev.1, July 3, 1987.

T0lP-5130, "ECN Implementation," Rev. O, June 12, 1987.

8. Maintenance Department Administrative Procedures:

MAP-0002, "Control of Maintenance Activities," Rev. O, July 2,1987.

MAP-0006, "Work Request Planning," Rev.1, July' 27,1987.

MAP-0009, "Preventive Maintenance Program," Rev.1, October 13, 1987.

FAP-0017, "Root Cause Determination," Rev. 0, June 12, 1987.

9. Licensing Department Administrative Procedure LDAP-0004, "Incident Analysis," Rev. O, October 14, 1987.

10. Rancho Seco Administrative Procedure RSAP-0803, "Work Request," Rev. 3, November 21, 1987.

11. Plant Operating Procedure 8.1A, "Control of Heatup and 0perating Procedure for Power Escalation Testing," Rev. 1, February 3, 1988.

12. Rancho Seco Operations' Special Order 87-37, "Surveillance Procedure Scheduling / Performance," August 10, 1987.

13. Rancho Seco Precursor Review Task Closure Report, August 29, 1986.

14. Rancho Seco Procedure RPQG, "Restart Personnel Qualification Guide," Rev.1.

15. Rancho Seco Daily Outage Schedule, February 17, 1988.

A-2 l

1 APPENDIX B AEOD TEAM INTERVIEWS OF SMUD MAMGEMENT l

l l

l l

AE00 TEAM INTERVIEWS OF SMUD MANAGEMENT Manager Department Bill Ken:per Operations Steve Redeker Operations Lee Fossum Scheduling Joe Flynn Plant Peris.rmance Paul Turner Training Bill Stahl Testing Dave Brock Maintenance Steve Crunck Licensing i

I l

l B-1 j

APPENDIX C Plant Performance and Management Improvement Program (PP&MIP)

0 t

In response to the December 26, 1985, overcooling event, SMUD initiated a comprehensive Plant Performance and Management Improvement Program (PPAMIP).

The broad-based PP&MIP wa., developed in recognition of Rancho Seco's generally poor overall performance in the time leading up to the transient. The PP&MIP included the systematic assessment of approximately 4000 recommendations for corrective action and the extensive review and testing (i.e., the System Review and Test Program (SRTP)) of 33 selected plant systems to demonstrate the functional capability of systems important to the safe operation of Rancho Seco.

As shown in Figure 8, the PP&MIP was structured to achieve: 1) a reduction in the number of reactor trips, 2) a reduction in challenges to safety systems,

3) assurance that the plant remains within allowed ranges of reactor coolant system pressures and temperatures immediately following a reactor trip, 4) assurance of license requirements compliance, 5) a minimum need for operator actions outside the control room, and 6) improvement in plant reliability and availability. Issues and deficiencies related to plant performance and recommendations for corrective action.came primarily from three sources: 1) the Department Managers Hardware and Programmatic Recommendations, 2) the Management Process Review Recommendations, and 3) the Systematic Assessment Process Recommendations. Further, recommendations for additional plant modifications or corrective actions developed as "feedback" from the SRTP, a key element in the implementation phase. The Department Managers Hardware and Programmatic Recommendations were developed for an assessment of the plant design, management, operations, and administrative system deficiencies based on existing reports and evaluations (e.g., NUREG-1195, INP0 Audit Reports, and the November 1984 Rancho Seco "Maragement Appraisal Report"). The Management Process Review Recommendations were developed by a group that was established to review previous (i.e, the last five years) management audits and assessments and licensee responses and commitments related to these assessments, assessments of current management processes and fuctions, and abstracted assessments of management from other elements of the PP&MIP. The Systematic Assessment Process was developed to perform a detailed review of the Rancho Seco design and experience as well as appropriate industry experience. The Systematic Assessment Process developed recomendations from six areas or sources: 1) precursor reviews, 2) plant staff interviews, 3) deterministic failure consequence reviews, 4) B&W0G Safety and Performance Improvement Program (SPIP), 5) December 26, 1985 overcooling event and NUREG-1195, and 6) selected projects (e.g., Motor Operated Valve Refurbishemnt Program, Implementation of IE Bulletin 85-03).

The objective of the SRTP was to ensure that systems important to safety are ,

ready to perform their intended function by meeting the following:

o Evaluate all system problems identified by the PP&MIP.

o Develop an integrated program of corrective actions for implementation which will address system problems.

o Identify those systems that require special consideration under the SRTP.

o Identify system functions important to the safe operation of the plant.

o Develop and implement a testing program that will demonstrate how well those functions important to safe plant operation work.  !

1 C-1

m l.

k.

I' i

Phase 2 Phase 3 Phase 4

. Phase 1 i n.

ISSUE IMPLEMENTATION CLOSURE INPUT b EVALUATION fs DISPOSITION PROCESS MANAGEMENT Restert Testing t.

p Dept. Mors. t' Hardware & J' Programmat ic Modifications & I' 7

Maintenance Recommendations ,

- Improvements Management '

Operations b Administrative o 1 y 3 g Process 8 Improvements y

l f Management Performance ,

Process  ; Analysis OA Review I Geoup Recommendatione Verification i J L i Vehdation Fehng m Troching j L Systematic SYSTEMS REVIEW Assessment Recommendetsons. Systems Engr.

Process 2 Review and -

Functional Regmts.

Recommendations Resolut.on Board Test Regmts.

(O Cl-12)

• I i I i

! Figure 8 Plant Performance and Management Improvement Program Figure 8

All systems at Rancho Seco were reviewed under the SRTP and divided into two categories. The first category, "selected systems," includes 33 systems that are important to safe. operation. The second category, "additional systems,"

includes the 44 remaining major systems at the plant. The review concentrated on the selected systems.

An augmented system review and test program (ASRTP) inspection was conducted by the NRC staff and its consultants to complement the programmatic evaluatin of the SRTP decribed above. The objectives of the inspection were to evaluate the effectiveness of 1) the SRTP process and results, and 2) the licensee's established programs for ensuring safety during plant operations after restart.

As a result of the ASRTP inspection findings, SMUD committed to resolve all identified issues (refer to NRC Inspection Report 50-312/86-41),and implemented the following major actions: 1) quality assurance vertical audit, 2)AFWsystemupgrade,3)anEngineeringActionPlant,4)expandedASRTP (i.e., complete an ASRTP type inspection of all 33 select systems before restart - evaluates the overall effectiveness of programs established to ensure safety after plant restart), 5) B&W SRTP review, 6) surveillance procedure technical review, 7) preventive maintenance program verification, 8) operator readiness program, and 9) technical specification compliance / verification program. The expanded ASRTP (EASRTP) was accomplished on all 33 systems by teams of Rancho Seco employees and independent consultants utilizing their new vertical audit techniques. The EASRTP identified 294 new issues which were integrated into the Rancho Seco Action Plan for resolution.

C-3

l 4

f APPENDIX D INDEPENDENT REVIEW INFORMATION

i I

List of Outsic'e/Independant Reviews '

Crgenization Da_te __

_ ___ _ Purpose Operational Readiness 11/87 - Current Review of plant readiness Review Cunf ttee (ORRC)

Independant Review Cemittee 3/86 - 6/87 Review of Nuclear Activities Nuclear Advise;y Corsittee 5/87 - Current Ad Hoc Review of Nuclear (NAC) Activities ,

Technical Oversight Co2tittee 9/87 - Current Review of Engineering /

(T0C) Dasign Function bel 9/87 Review Test Program of 13 systcus LRS 4/86 Ascoss K,terial t Control Program Ralative to LER 85-14 LRS 11/84 Managetrent Appraisal Ganeral D)Tatics 84 - 87 Assessment of Chemistry & RP Th<xas Laronge Inc. 9/87 P. Chemistry Staff Facility L Progrc:2 S. Levy Inc. S/87 Audit Design Configurant Process for I:sdification on 4 Safoty System 3 B?chtel Western Pcwer Company G/87 Evaluate calcuntion for l Dasign chcngas UES 1/83

• EASTRP 33 Systr.s reviewed I

SMUD personnel participated in evaluation teams but progran directed by UES.

FOR g oRi'<T M D-1

List of Outside Agency Reviewers Agency __

Deto Subject ANI 12/849/87 Property / Fire /all risk insurance ANI 1/26-28/88 Liability Insurance ANI 11/18/A7 Property Insurance Inspection INP0 1/22/88 Plant & Corporate Evaluation 10/96, 12/86, 4/87, 11/87 Operator evaluation at Simulator 7/06 Evaluation of Training 7/87, 11/87 RP & Chemistry (IAC, Electrical &

Mechanical) Maintenance & Tech Staff ANI 2/86,9/86,9/87 Training Evaluation Jerry Holmen & Assoc. 2/87 Required Training Evalut a of Operators 9/87 Hot operator Evaluation General Physics 2/87 Training Evaluation of Operators Stone & Webster 7/87 Fire Protection Training A Systcas Evaluation United Engineering Srvs. 6/87 - 8/87 Schedule content & process understanding UES - Cost ichedule 7/87 -9/87 Cost Schedule (twf N a week) Integration On-Shift Management 4/8f L&W type SRos Industry Advisor prcvide addition depth &

Management Experience B&W 9/87 Review E0Ps for evaluation against B&WOG E0P Technical Basis Doctaments General Physics 2/81 Training Evaluation of Operators I FOR WFORMAT!DN ONL 0-2

FOR INF0iUUfb List of Outside People For Aid In Experience (Deputy Department Managers or Equivalent and above and special consultants.)

Wase

_ Cour;any/Backgroun_d Purpose __ Data M

Red Carlton B4W 20 yeaes NSS Adytser Test 10/87 Design & Test Program Program Bill Stahl Nutech Calaway Test conduct 6/87  :

1 Toa Childress Volt Yartous Startup Test Support 9/87 I Operations Engineering Marshall Weaver Duke Onconee Mechanical Engineer 9/87 Modification Manager Supervisor Charlie Aycock Duke Construction Mgr. Restart Engineering 9/87 i Managor l T. Telford QMSC Mgr. Nuclear Engr. Engineering Support Mgr.10/87 Dir. CA, US WRC-NRR Richard Harlow MACK Configuration Control 2/87 D. Humenansky UES Tech Assistant EASRTP 7/87 US WRC - INPO l

Materials Control '

Bob Ferguson LRS Materials Continued Assessment 4/86 of Material Program i Plant Support Joteph Flynn INP0 Maintenance / Mgr. Plant Perform nce 10/87 Training -

> nsumers Power

_ Operations William O' Conner Toledo Eiigineering Evaluate Ops 11/87 staff consultant Dept, readiness for restart D-3

.1..

Wa'ze_ _ Company / Background Purpose Date Matatenance Dave Brock Nutech Maintenance Mgr. Manager 6/87 Crystal River Maintenance Consultant & SN0G Startup Schedul_ing_& Outage Management Gary Shenker UES Scheduling Deputy Implementatt - 9/87 Manager Nar.ance Managemnt Supervisor i

John Strange IMPELL Pacific Director */67 Manageant Sys. Management Systems Grant Yelliot N4T Procedure Manecer Procedure 8/87 Development Development gnagtent_ Safety Review Ctraittee P. Herbert Indep. Codes & Standards MSRC Member 1/87 T. Petersen General Dyna. des /RP MSRC Member 9/87 W. Hartley MAC Operations t 0A MSRC Member 9/87

_Q:!ality Assura".ce i

T. Hardy KAC Operations Superviser Quality 6/86  !

Startup Outage & Support Quality  !

Services R. Roth DLRE Mgr. Quality Supervisor 1/88 Quali1;y Engineering i

Mechanical l M. Parenteau Bechtel Startup, 0A Surveillance t

6/86  !

Quality Supervisor i

. - - . ~ , .

n e '

@ Q-j {-Tj h $h . ' !I I... . .1 I5.-.-

l 0-4 i l

l

Enclosure 2 B&W Plant Restart Data Review l

l 1

o w.

Quick Review of Operating Experience For Three Mile Island 1, Davis-Besse, and Crystal River 3 Following Restarts Introduction The Nuclear Operations Analysis Center (NOAC) at Oak Ridge National Laboratory was requested to review the operating experience as reported in licensee events reports (LERs) of three B&W plants for one year following their respective restarts after extended outages. The three plants are Three Mile Island 1, Davis-Besse, and Crystal River 3. The objective of this review was to identify any trends or unu.aual occurrences from these three plants, individually or collectively, for the one-year period following restart tilat AEOD might highlight for th.e benefit of Rancho Seco, another B&W. plant, that will soon resume operations following a lengthy outage.

Review of Three Mile Island 1 Number of events reviewed - 14.) For

~

(Restart date - 10/3/85.

date, THI-l the one-year period following the 10/3/85 restart apparently did not experience any significant or recurring problems, equipment failures, or personnel errors, as reported in LERs. The types of events reported for the most part did not appear to be directly related to the fact the plant had just restarted from an extended outage. However, four events that may have sorae potential generic implication for the restart of Rancho Seco are summarised below. I LER 289/86-00E. (Event date: 1/4/86) - The reactor was at 22%

power and esca:ating following a planned reactor trip on 1/2/86 when the turbine tripped followed by a reactor trip. The turbine trip was caused bg high moisture separator level due to an undetected failurn in a valve controller in the the moisture heater drain flow path. The heater brain backed up into separator y

causing the high level and subsequent turbine trip. The air ports on the valve cor. troller were clogged with dirt / rust.

LER 289/86-003 (Event date: 1/14/86) - The undervoltage relay associated with the shunt trip feature for a control rod drive breaker unit failed its surveillance testing. The failure was due to the incorrect positioning of a jumper internal to the relay. The 48/125 VDC jumper was believed to have been in the 48 VDC position instead of the 125 VDC position when the relay was shipped from the factnry T: .e test procedure for initial installation of the relay did not specify checking the position

of the jumper. The relay passed accept,nce tests and functioned i

4 1

NOAC Operating Experience Review February 10, 1988 for TMI-1, DB, and CR-3 several times after installation with the jumper in the incorrect position.

LER 289/86-008. (Date of event -

4/22/86) - The unit had just achieved hot shutdown following an outage for eddy current testing. Transformer tap changes had been completed on "A" auxiliary transformer and preparation for "B" transformer changes were in progress to support plant startup. When the "D" bus was transferred by closing the feeder breaker, an electrical' fault occurred tripping open both incoming breakers to the "D" bus.

When the breaker tripped open the fault did not clear. At TMI, automatic transfer schemes exist to transfer loads between transformers. The effect on Unit 2 was minor since all TMI-2 loads were transferred 'to "A" auxiliary transformer from "B".

This auto transfer scheme for Unit 1 was ' defeated at this time because the control switches for all the breakers that would have closed to supply power from the "B" auxiliary transformer were in pull-to-lock as part of the sequence for taking the "B" transformer out of service for the tap changes. When the "A" auxiliary transformer was de-energized, all four reactor coolant pumps tripped, all secondary AC powered equipment tripped, and the reactor tripped. Having lost main feedwater , the control room operator started the electric driven emergency feedwater pump to supply water *o the steam generators. The emergency feed water pump did not au.) start because the start circuit had been defeated per the design and operating procedure.- (This ceque'nce.

cf events is performed as a prerequisite to criticality and.is not performed with the reactor critical.) A pr.ocedure was to be prepared at TMI to incicde the proce.ss of changing euxiliary transformer taps making all required transfers prior to placing any breakers in the pull-to-lock position. This would allow an automatic bus transfer to occur if required.

L.ER 289/86-010. (Date of event - 4/23/86) - While at 9% power during power escalation in the same startup as the previous event, the reactor tripped o r. high reactor coolant system e.;ssure resulting from inadcquste main feedwater flow while transferring the main feedwater pump turbine steam supply from the auxiliary boiler to main steam supply. The manual action taken by the operator during the transition was too slow to maintain adequate feedwater to the stea a generators.

2 m

NOAC Operating Experience Review February 10, 1988 for TMI-1, DB, and CR-3 Review of Davis-Besse (Restart date - 12/22/86. Number of events reviewed -

16.)

There are three problems that Davis-Besse experienced' of potential concern to Rancho Seco -

one involving improper, previous maintenans3, one involving problems with the moisture separator reheaters, and one concerning over-cooling following

• tribs.

LER 346/87-001. (Date of event - 1/1/87) - While at 40% power, the reactor tripped due to the loss of one main feedwater pump turbine (MFPT) which tripped as a result of high. vibration due to bearing failure. Low oil flow . to the bearings occurred when a pipe plug was lodged between the lower portion of the bearing and -

the oil supply tubing. A pipe plug was missing from an open port -

directly across from the oil supply inlet. It was the same size as the missing plug. The pipe plug could not have physically entered this region without disassembly / reassembly of the MFPT block-bearing assembly. Review of the maintenance history did not provide a conclusive answer on when the pipe plug was left~in the supply system.

Moisture separator reheater problems. On two occasior.s following reactor trips (LER 346/87-010 on 8/27/87 and LER. 346/87-015 on -

12/7/87), moisture separator reheater reheat steam supply valves did not close. In bsth instances the valves failed to close to due to failure of the same pressure switch. The switch was  ;

repaired following the first failure and replaced following the l second. This pressure switch was being evaluated .to . determine l appropriate preventative maintenance requirements.

Over-cooling events following tri_na. On at least two occasions following reactor trips (346/86-043 on 12/22/86 and 346/87-010 on 8/27/87), Davis-Besse experienced cooldown transients. The first occurred when use of main feedwater in manual caused a steam generator overshoot of the post trip target level. .The steam demand exceeded the heat input from the reactor coolant pumps.

The second occurred due primarily to the failure to close of the moisture separator reheat steam supply valve noted above. This created an additional steam load on the steam generators. Davis-Besse noted that trips occurring since restart have involved over-cooling and was to have completed a review of these events by December 30, 1987.

3

NOAC Operating Experience Review February 10, 1988 for TMI-1, DB, and CR-3 Review of Crystal River 3 (Restart date - 8/20/85. Number of events reviewed - 34.) A review of the LERs submitted by Crystal River 3 in the year following its restart revealed three recurring event's: cracks in equipment, spurious emergency feedwater actuations, and voluntary violations of technical specifications. The first of these recurring events, cracks in equipment, may have been related to the extended outage. The remaining two trends are probably not related to the outage.

Cracks in ecuipment. On three occasions during the period reviewed, Crystal River 3 experienced pipe or piping support cracks (LERs 302/85-015, 320/85-024, and 302/86-014) and twice experienced pump shaft failures (LERs 302/86-001 and 302/88-003).

The first pipe leak occurred on the restart date. It involved a leak on a drain line for the main turbine high pressuce crossover line. The leak was caused by fatigue failure induced by vibrations from modified turbine governor valves that had been recently installed. The remaining two pipe problems involved the nuclear services sea water system. On 10/28/85, cracked concrete support pedestals revealed design errors in the support system for the sea water discharge piping and leaks developed in the sea water discharge piping on 8/12/86 as the result of erosion caused by Juspended solids in the unfiltered sea water.

The two pump shaft failures occurred on 1/1/86 and 2/2/86. The first occurred on a reactor coolant pump due to fatigue. The crack initiations were caused by mechanical and thermal stresses but came to a stop due to a lack of driving force. After a long pause, the crack growth continued due to broken impeller-to-shaft bolts and normal bending loads. The second shaft failure occurred on startup following repairs to the reactor coolant pumps. A decay heat removal pump shaft broke due to torsional fatigue induced by excessive loading. The excessive loading was most likely caused by pump air entrainment due to vortexing which occurred during operation with a low reactor coolant level.

Spurious emergency feedwater actuations. The second recurring problem at Crystal River 3 in the year of operation following restart after an extended outage involved spurious actuation of the emergency feedwater (EFW) system. On uoven occasions (LERs 302/85 020, 302/85-023, 302/85-026, 302/85-027, 302/86-001, and  ;

302/86-008), a spurious signal indicatinn low steam generator j level caused EFW actuation. The spurious signal had a variety of 1 causes but were the result of a upset condition elsewhere in the l secondary system. Each tim the RCS was over-cooled. A recent I update of LER 302/86-00' (1 r date 12/9/87) indicated that the l l

NOAC Operating Experience Review February 10, 1988 for TMI-1, DB, and CR-3 utility has attempted to solve this problem by installing a 2-second time delay on initiating EFW.

In addition to these EFW actuations occurring after the restart (criticality date of 8/20/85), three LERs reported similar actuations immediately prior to 8/20/85. LER 302/85-012 (date of event - 8/9/85) reported 12 EFW actuations - 6 due to a small margin between the normal steam generator level control setpoint ~

and the EFW low level actuation setpoint and 6 due to pressure surges resulting from valve mr.nipulations for the level transmitters. LER 302/85-013 (dt.te of event 8/15/85) described an EFW actuation occurring when maintenance to repair a leak on a line feeding two level sensors resulted in a depressurization of the line causing the two level transmitters to indicate higher than actual level. The EFW overfill protection circuit actuated.

Two EFW actuations were reported in LER 302/85-014 (date of event 8/16/85) due to steam to feedwater flow imbalances due to personnel errors while control of steam generator level was in the manual control mode.

Voluntary technical specification violations. The third recurring event involved four soluntary entries into technical specification violation. During three events (LERs 302/85-022, 302/85-029, and 302/86 33), Crystal River 3 disabled two redundant channels of oae emergency feedwater initiation and control system were disabled to allow maintenance personnel to replace failed instrumentation. The fourth event (LER 302/86-011) involved deactivation of the control rod indication circuitry in order to replace three relays. The unit posted additional personnel to monitor process conditions each time. No adverse effects were reported.

LER_302/85-016. (Event date: 8/20/85) - In addition to the t'.ree recurring events, one isolated event occarred that was the result of maintenance activities that occurred during the extended outage. During the restart on 8/20/85, the unit developed a heat '

removal cyclic transient while raising reactor power because of sluggish feedwater control valve response. Feedwater control was l in manual at the time. Operators switched control to automatic and back to manual in an attempt to control the perturbations. '

I However, all attempts were unsuccessful and the reactor scrammed on high RCS pressure. An investi stion revealed that the feedwater control valve air line had been crimped and an air control valve was mechanically misaligned within the c neumatic controller. No additional problems with the feedwater control valve were reported during the year following the 8/20/85 restart.

l 1

e i

NOAC Operating Experience Review February 10, 1988 .

for TMI-1, DB, and CR-3 Conclusions This review of the operating experience for a one-year period following the restarts of Three Mile Isl'and 1, Davis-Besse, and Crystal River 3 included 64 LERs. Few recurring problems were noted that were directly related to the extended outages. No problems were common to all three units. However, isolated occurrences that were unique to each unit may provide the operating personnel at Rancho Seco insights to the types of problems that they may encounter during their own restart.

1 o

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

1 UNPLANNED REACTOR SCRAMS ,

THREE MILE ISL. 1 15 1000 E3 Eo#went m o. er l M Rxe&res  ;

a Pescru ere' C g a t . ne . , - v 00>

g u 10 _

N W* M M '

_ 600 o

% I O

L

_o C)

U O

% _ 4C:0 <.

'c

~

3 5 -

z

.. 200 0 h -- -

0 10 / S 5 2 4 6 8 10 12 14 16 13 20 22 24 26 Months CUMULATIVE UNRLANNED REACTOR SCR/ 'iS .

THREE M:LE ISL. 1 15 m

E 8 10 _

u -

O C)

> ^0 0 0 z /

o -

a ,

1 E  !

j 5 _

: == .

1 1

l 1

0 - , , i l 0 5000 10000 15000 20000 Cumulative critical hours I

-<-..,<...,em.-

UNPLANNED REACTOR SCRAMS CRYSTAL RIVER 3 15 1000 ESS Eq.opwt ra other M Proct &Jt$

M Persenr+i err c' M L*b FC e n $00 E

h 10

_ 600 5

% I u E a) t D =

E

_ 400 c a 5 .

o Z

_ 200 ,

O h 4

. . . . ... . i . . .. . . . . . . . . . . - 0 OS/85 2 4 6 8 10 12 14 16 13 20 22 24 26 23 Months  !

l l

CUMULAT!VE UNPLANNED REACTOR SCRAMS )

15 CRYSTAL RIVER 3 t

. i m

E 8 10 _

u _

O I 9  :  :  : ==  : :  :  : .

.:7 E

[

D 5 .

O ~

O i i 0 5000 10000 15000 Curnulative critical hours

- - co., to.o i= ~ m

UNPLANNED REACTOR SCRAMS DAVIS-BESSE 15 1000 M Coipmeet e other a proceases 3 Persorrel e ce m a no.- - 800 v' 10 *

• 3

_ 600 o '

- C o

L _O c> U

-9 400 'c c .

O g5 _

t 200 a _

0 , ,, , , 0 12 / 8 6 2 4 6 8 10 12 14 16 'S 20 22 24 Months CUMULATIVE UNRLANNED REACTO: SCRA MS 15 DAVIS-BESSE a . , ,

D c

10 _

o

.2

-o 2

5 o _ - -

1 l

0 e

_. i i ,

l 0 2000 4000 6000 8000 10000 Cumulative critical hours

- - r .,-.ot- - ,

SCRAM EVENVS RWdt P( llAE DOCKET LER.NUMER EVDT.HTE CAUSESCRAM 1 ilftEE MILE IR. I 299 28985003 12/91/85 EIDIP Wil 2 i)StEE MILE IR. I 289 2Et94002 81/H/N E9UIP Wil 3 T)ftEE MILE 1R. I 2t9 28 W 004 03/15/84 PROCEDUR Wit 4 T R EE MILE ISL. 1 289 28984010 N/23/84 PRSNLERR AUTO 5 T)ftEE MILE ISL.1 289 28 m 0!! M/02/N EQUIP AUTO

& i)EtEE MILE ISL. I 281 28987004 05/02/97 PRSEERR WTO 7 TWtEE MILE ISL.1 289 28987006 04/12/87 PRSNLERR WTO 8 T M EE MILE ISL. 1 289 28987000 09/16/87 EQUIP AUTO 9 CRYSTAL AlVER 3 302 30285016 08/20/05 E00lP AUTO 10 CRYSTAL RIVER 3 302 30205015 08/20/85 EQUIP AUTO 11 CRYSTAL RIVER 3 302 30285020 10/09/85 PROCE80R MANUAL 12 CRYSTAL tlVER 3 302 30285023 10/26/85 E901P AUTO 13 CRY 874L RIVER 3 302 30285025 !!/08/85 Ult'N0tAl WTO 14 CRYSTAL RIVER 3 302 30285026 11/22/85 PRSE ERR AUTO 15 CRYSfAL RIVER 3 302 30285028 12/03/85 EQUIP WTO 16 CRYSTAL RIVER 3 302 30284001 01/01/84 EQUIP AUTO 17 CRYSTAL RIVER 3 302 30287009 07/02/87 EQUIP AUTO 18 CRYSTAL RIVER 3 302 30287011 07/10/87 EDU!P AUTO 19 DAVIS KSSE 346 34686043 12/22/86 PRSNLERR AUTO 20 DAVIS-8 ESSE 346 34687001 01/01/87 EQUIP AUTO 31 DAVIS 8 ESSE 346 34657006 03/13/87 PR$le, ERR AUTO 32 H VIS 8 ESSE 346 34687010 08/21/87 NATURALP AUTO 23 DAVIS 8 ESSE 346 34687011 09/06/87 EQUIP AUTO 24 DAVIS-8 ESSE 34 6 34687015 12/07/87 EQUIP AUTO I

l l

l

u ,- y .

t n n - - - -

- 1. o I

e k

5 f

I d 9

I i la o a 3

E

$3

• g lI 5

l e a

.= n -

a sla 3 E 2

5 t

l!:Es!

0

m g

:a I

l3 .g  ! :i g;e 5

!y

• IE "

E$G mE 92 2g"E.

E - I E 6

=

j 5 3o 3  ! i. 38 3 g lI9I $e E

• s-I $a5

. p kW a Be

! hd !E e a 5 w 6

• I.9<E 5*ge E -

=

E I n A E !$ 5 1 a e i a i* e ; ! 1 lF

~

g$Eg$- g

k. -g3

! h N. E. !. 5, 5e Et -

X E

5

r. aEe 51

-,e s8 9 8i!!

a lE( w a, *g5 5 ,a $. g,1 n

W l5a 9 -

i ; *

- I 4 E m 8 N he j E j 3 s, 3b"[!lE F 2 i j 0S a E 2 -

8G A E D =

-g5 [I g g l* i; 85 q

i !il!l$jN

{E .

II E

!!!Ii!!!i

- E 9 r lailIiIlaaal!!Iaaaii1ll8a 5

I

• I g $  ; 5g g g g g 3 5 E 3W - e e afI5$$

c n g

• g e5i5555m a = g a a s e a n c c a s a a e g g a l

1 w

- i .! . ee:

18 .a l l l .: . . I I=i = e i: . . .

inis Iea:ievi g . _ .

aieii::i! __ ._-

.liIiIIiiiai11iiIIi1Iiiii O

f3n.l. NEE - - n n& n n.52E2UEiiiiii55ii a a

, , a, u, a, e a a a

, w , , a, Ean 3E!

1 affIlIIIIIttttttttt1iii1IE o- - - - - - - - - , = = = = a a =,= a n n n r, ,

l

l

. 1 I

ESF ACTUATIONS '

THREE MILE ISL. 1 40 1000 i ES Eggment go otw 3 Procedures e ag PesoW wrw C E (M noen - 800

-c 30 _

0 3u m b

O . 600 o r

o 20 _

6

.9 D

O _ 400 Y E O o

2 10 _

. 200 0 . , i i . .

mm e m 0

. . . > .. . . . .. . 6 i .

10 / 8 5 2 4 6 8 10 12 14 M 13 2^ 22 24 26 Months CUMULATIVE ESF A:TUATIO: 5 '

'5 THREE M:LE ISL.

m c

.9

~

E10 _

~

u O

O

.2

.O_

a E5 a

O

o. e. 7 ....

0 , . . . . , , i i i,i,iiiiiii iiiiii 10 / 8 5 2 4 6 8 10 12 14 16 18 20 22 24 26 Months

...,. c -

ESF ACTUATIONS 40 CRYSTAL RIVER 3 1000 E Equpment rzg othe 3 Procedures c- 3 Persomel error C M Ur*nown - 800

.E 30 _

o D

~ $

a 8 i _ 600 o

.c Y) 20 _

o o

w N _

400 k c

o a

Z 10 _

_ 200 0 . . ' . ' h3 A

. . . .m . . .

a, , , ,m. i i i i . . i i i i i . . i Em 0 03/85 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Months CUMULATIVE ESF ACTUATIONS 20 CRYSTAL RIVER 3

. i i . . . i i . .i,ii.i,ii,.i,,,iii, c.

c

.c 15 _

c a

S  : : : : : : : : : -

, - . . . /.

.2 -

~o a

E 2

C 5 -

0 i . > i i i i iiiiiiiiiiiii,,,7 i i i i i i 08/85 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Months i ., ..,. i m .

ESF ACTUATIONS t

DAVIS-BESSE 40 1000 M Emern ent as otte e proceas u (S m PersoW error C

m m no.n - 800

.2 30 _

o '

3 M a "

8 _ 600 o

.c o 20 _ i o

1 O

N _ 400Ui T

3 I z 10 _

_ 200 0 mm. i . m

. i i.E ..,, . . . i i i . .

O 12 / 8 6 2 4 6 8 10 12 14 16 15 20 22 24 Months CUMULATIVE ESF ACTUATIONS DAVIS-BESSE 15 . . . . i>> > i i , , , , i i . . .

0 5

.=

o 10 -

Uo O

.2

~b

'3 3 5 -

o O i i i i i i i i i i iii i i i i i i i i 12 / 8 6 2 4 6 8 10 12 14 16 18 20 22 24 Months isse umhee teseesq taeeen taas m h6

691 M]'tGNW Retards PL lWE DOCKli LD RURBER Evivi,Mit CAust i TIItti RILE 1R. I 209 209N000 N/21/N EQUIP 2 Tlftt! R!LE 1R. I 299 299Nell N/02/N EIUIP 3 iletti Ritt ik. I 299 29987001 01/09/17 PROCEDUR 4 TMtti RlLE ISL. I 299 2B987002 03/02/17 E90!P 5 CRYSTAL RIVER 3 302 30285019 10/04/85 PRSEERR 6 CRYSTE A!VER 3 302 30295020 10/09/05 EWIP 7 CRYSTAL RIVER 3 302 30295023 10/26/85 EDUlP I CRYSTAL RIVER 3 302 30295026 11/22/05 (Gu!P t CRYSTAL RIVER 3 302 30295021 11/23/85 PRSEERR  !

10 CRYSTAL RIVER 3 302 30295027 11/23/85 PRSEERR I

!! CRYSTE RIVER 3 302- 30295020 12/03/05 ElulP  !

12 CRYSTAL RIYDt 3 302 30295031 12/07/85 EDUlP 13 CRYSTAL IlvtR 3 302 30286001 01/01/N EDUlf 14 CRYSTAL RIVER 3 302 30286005 04/22/N PtSEERR 15 CRYSTAL RIVER 3 302 30286000 N/20/M UKWCW 16 CRYSTAL RIVER 3 302 30286021  !!/12/N UEEN 1 17 CRYSTAL AlVER 3 302 '30297021 10/14/87 EQUIP 10 CRYSTAL RIVER 3 302 30297025 10/16/87 PRSEER 19 CRYSTAL RIVER 3 302 30297027 10/26/87 PRSNLERR 20 CRYSTAL RIVER 3 302 30297022 !!/06/87 PR$NLERR  !

21 CRYSTALAlVER3 302 30287030 11/20/87 UwWCW 22 CRYSTAL RIVER 3 302 30297028 12/05/87 PRSEERR 23 CRYSTAL RIVER 3 302 01/07/98 (9UIP 24 CRYSTAL RIVER 3 302 01/07/98 EQUIP 25 CRYSTAL RIVER 3 302 01/09/88 EDU!P 26 DAVIS 4tSSE 346 34696043, 12/22/M PRSEERR 27 DAVIS-BESSE 346 34687001 01/01/87 (BulP 21 HV15-8tSSE 346 34687006 03/13/87 PRSWLEPR 29 DAYlS DES $t 146 34687011 09/06/I7 E9JIP 30 DAV!S DESSE 346 34687011 09/06/87 EDUIP

ansse et,mut P 61A,iWata Qsuut (61 anM PMRN1 F5DI 1754112 EMftATI SMllaf2 EMTWi3 564TWI4 Sv52RATI 5%!51LAl2 Sv51511Af3 SY51tME CAE Catt':7 i n.E M In a .x. .ua/m .K a o Mme (M' 2 DEU RM IL 8 3 ences cap 7 tuC/t see a gi (708E (MP CSF O 3 DIEI eM C. I 3 487001 47423 C/st.O OK Q mtue sguttas 4 1:3E NM It i Mus7002 97E2 83/s2/87 9K O MIe4 EMP S Of574 tic 3 3mpit eDC 14/st/E5

& IMITAL tie 3 icwune 0223 T3 K p

't mise ft541ap O 1a/99/95 %K an so ss Man't Eur $rts Pe 1 CR$it Ele 3 IrmM CD47 10/3WE5 f3 K In CD WUmE EMP e" -

8 D5ft #16 3 302Ec3 11/22/55 22 K m trUs4 EMP f DYETE 816 3 30 3 3027 10980 8K m 31/23/B TE571as ru9ttp te Of5tt RIO 3 303Xfl7 31/23/E3 9K In R5ftes restLC 33 IM5ft #16 3 3tC203 02CF 12/93/t3 T3 K m Mbit (MP 9410ts a pr5t= sic 3 m2rma em a/oue eK m a u se n5115 (MP 13 Dv5TA RIO 3 30 3 6001 0 213 tim /u MK en M4rt (MP

. Oms t c 3 m3ee 0.n am/is eK D n51n. ,,,p O I3 Dv5it AI6 3 303n000 E183 46/25/a6 3K an mmE tumme le DY514 816 3 3c23 02: Geses II/12/a6 I4 K m as SJ MmE tasmasi 17 DV5TE R16 3 30 3 7021 10 3 30/14/E7 0K D 18 DY574 #16 3 303703 10'43 30/IVE7 0K 2 B A WOnfE fa:!P O N W M mlNO FW9L06 It DYSTE t!& 3 3IC57027 10C6 10/26/87 8K II to P = mients P99 tis -

D D9574 R16 3 30 3 7022 IQh03 t t/06/87 0K m a D,5is nie 3 m3mm ion. wwi7 eK = m = =

Mm7E PrAInn O en ci Mmm inns.

22 Dv514 #16 3 3037tus Basi 12/03/57 9 JE A BD EP D W W trUd4 PMim 23 Dv57= ere 3 ms? eueres eK in 2e IMSTE 816 3 11130 01/07/ss 0K in Mis.w tur O (PueE IMP 3 DY514 816 3 1I172 Ot/Ot/W lK an MMT Fit?

D thW!5455[ 346anos3 07276 13 K m E 27 Devil-stS5t 12/22/06 JT 5J M)n4 Pr#1m O 3aeroct 47E2 01/01/w? 40 K = 94 M5 Mm!E EMP 3 Dan 5435r 3ae sh OW 36 03/13/87 72 K p as 2T 5J MD4 PW94 t*

29 atf4E55E 34e870tl PP9 M/06/87 0K m N BU.5455E 34e470t t 09M7 pron /87 300 K q

D en MIdit EMP O MM4 IMP i b w8'e=

edh.tW Q 5 A . A % rs.J sJ = F- .f.. 4ft.v O.

6 t - D e b 0-e i'" y cc - Cmp. 62.:.3 IJ.C. .

.TE. csrAs 6 4 !GR L 5-f%NY f>P " LPI (Pue)

C4= Cmt. % CL' QMCl9 & O 64 = HP I Uu80 VF=A %

SB: r%;- / fa U Uh WAc JT S FRC5 (BM Ve= q i.,E H VA C. M 5

• MSW g; j

O O

~

6 9-O O

O

SAFETY SYSTEM FAILURES THREE MILE ISL.1 40 1000 E Eppment rs other a Procedtres a Personne enor m m no.n - 800 S 30 -

C e v.

> h

_ 600 5C 0

20 -

~

o b .2 D

E _ 400 'E o C Z

10 _

_ 200 0 m 0

. . . i . . . i,ii , . -. . . . . . . , , . .

10 / 8 5 2 4 6 8 10 12 14 '6 13 20 22 24 26 Months C.UMULATIVE SAFETY S 5TEt.1 FA! LURES THREE MLE '5_. 1 15

. . , . 4 ....... . . . . . . . i .

v.

C O

> 10 ~

O o

.2

_o a

E 3

g 5 _

O - l ; ; ; ; ; ; ; ;-; ; ; ; ; ;-; ; ; ; ; ; ;

6 4 ia i 4 l l l l l I i & 4 3 I 4 I 4 i i i i i i i 10 / 8 5 2 4 6 8 10 12 14 16 18 20 22 24 26 Months

~ r n ., t . u - -

SAFETY SYSTEM FAILURES CRYSTAL RIVER 3 40 - 1000 m Eospment s ot*

3 Proced.ses a pesonnel wrc' e in ne.n - 80n-i 30 _

E cn o '3 i . 600 o o .c i

20 _ f

-o e3 .Y 9 -

_ 400 T E O 10 _

_ 200 0 , , , , ,

m  %, , ,

e um, m, 0

, ,,,ii,i , i i i , , , , i 08/85 2 4 6 8 10 12 14 16 13 20 22 2' 26 28 Months CUMULATIVE SAFETY SYSTEM FA URES CRYSTAL RIVER 3 15 i . , , , , , , i , ,. i i . . , , , , , , , , , , ,

c-E I g 10 _ <

t l

e  ;

.2 I

-o a l E l j 5 _

i l

.  : ....d ~~~~

0 --

I e e i i i i i i i iiI iI I I i i i i i 4 i I I i i i 1 08/85 2 4 6 8 10 12 14 16 18 20 22 24 26 28 ,

Months l

~ ~ w t .n e - ~

SAFETY SYSTEM FAILURES -

t DAVIS-BESSE 40 000  :

s tet a other W Procedies 3 Pesomet wrc' m w no.n - '00 a -

C o 4 M

E $

_ 600 o

.C O

20 .

'5 i b .9 j {

_ 400 T a U O _

_ 200 i

0 mm .. 0

. . . ... . imm . . . . . . . . . . . . . .

12 / 8 6 2 4 6 8 O 14 2 20 22 12 16 24 Months i

i CUMULATIVE SAFETY SYSTEM FA URES

,3 DAVIS-BESSE 2

C '

• D -

e v

.2 o

3  :'

E 3 5 0 -

1 l

1 o a O

. ......f i i ii. iiiiii.. . .ii . .

12 / 8 6 2 4 6 8 0 12 14 16 1S 20 22 24 I Months l

-- r - ., . ., n ~

i

' Page No.* 1' $ f, \ I $ ,J es W.lv & S

02/MID 7

' CRYSTAL R!iER 3 20: 30:!!c27 11/23/g5 [pV!e Ani} jar,l[ e,;ent, rettsater AUTILIA:',(' ERIE 4:?FE!:WAi!8

!vstess Srn :  !)SIEP :st' CRTSTR R!iER 3  :; :::1600;.- 02/02/S6 Eiv!F Resi c al Fest Res:va: Sistees AE!!!'.I FIAT REP;,t SY!i!=

Srcus IFuR) [$PE;!E O E!!S CON:

CRYSTAL R!vtR 3 302 30:37002 02/27/87 EQUlP Engineered Safety Features - EN6!IEERED SAFETY FEATUP.!S Instrueentation ACTUAi!ON SYSTEM CRYSTAL RIVER 3 302 30217015 07/27/87 FRS4LERR Engineered Safety Features En6thEERED SAFETY FEAft9ES Instrueentation ACTUAi!CN SYSTER

!CRYSTALRIVER3 302 30:57016 08/03/97 PRSWLERR Peactor trip Instr n entation PLANT PROTECT!DN SYSTEM -

' CRYSTAL RIVER 3 302 10652 !!/11/87 Ett!' fee *;ency AC/DC P:eer Systees EMER6ENCY CNS!TE POWER SL* FLY Srcuo !YSTEM DAVIS-!ES!! 346 34687001 12/22/56 PRSNl. ERR Integrated Contre! Systee INTESRAI D CONTROL SYSTEM Instrueentation DAVIS BESSE 346 34697013 10/16/B7 FRSNLERR Control Rece Energeccy C0hTROL BUILDlhS/C04 TROL Ventilation Srstes Broup C09FLEI ENVIRONMENTAL CChiROL ,

SYSTER THREE MILE ISL. 1 2$5 28986007 03/26/86 ECU!F !ssential tes:ressed Air ESSENTIAL AIR SYSTEM Systes 5rcup I

4 l

.t ,

i l

l l

4 d

.,; --- ~ - + , -,---w, ., - ., ---,m , , - >. ---

TECHNICAL SPECIFICATION VIOLATIONS THREE MILE ISL. 1 40 1000 m to., m t VA Otter M Procedures M Pe Somel errcr g

[ M Ur,noen - "00

.c 30 .

2 e

.9 =

> _ 600 o

.c O 20 .

-o g

.o _ 400 'c s

E C o

Z l 10 _

_ 200 0 m i , i .

mme . i , , ,

m ,

M M, m C 10 / 8 5 2 4 6 8 10 12 14 16 13 20 22 24 26 Months CUWLATiVE TECH'.M SPE'FICATIOF. VIOLA 7!Ok5 7.uc,. ~c v,r

.- . i .t . . 3

a i a e a . i 4

l m

C o

1 I

_o 10 _

i

.9

> l l

v . . .. .

.2 g

5 e: .f " " '

E 5 l 2 _ .-

I o ... / l O I 4 I i i ii4 I iI I I I I i i

& I I i I I I i i i i 10 / 8 5 2 4 6 8 10 12 14 16 18 20 22 24 26 Months

- - .o . , < = m

TECHNICAL SPECIFICATION VIOLATIONS CRYSTAL R'lVt:R 3 40 _ 1000 IE Eppenent 555 other a Procea.ses M Persomel error p a u ne.n - 800 o 30 -

=

-O C 3

's i _ 600 o

.c o 20 _

e u .9

$ _ 400 'E E o a

Z 10 _

_ 200 0 ,

,m, , $7 m , $mymE. ,f3 MN d?f,E 7 ,

0 08/85 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Months CUMULATIVE TECHN! CAL SPECIFICi lC', VIOLAT!ONS CRYSTAL RiiER 3 30 i i . . , , i. , i i , . . . .

25 _

m c

.9

_o 20 -

.9

,$ 15 _

~s a

h10 -

U 5 _

.....}

O . e.

i i iiiii ,iii,,,iiiiiiii,, .

08/85 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Months

~.,<..,c.-

TECHNICAL SPECIFICATION VIOLATIONS DAVIS-BESSE 40 1000 RB Egtsprrent Gm on 3 Procedtres g 3 Persow errce a m ,,,, . 800 c 30 _

ho $

a 3 . 600 o r

o 20 _

6 u .Y

$ - 400 T E C a

Z 10 -

- 200 0 i E . .

En. . . . . . 4 .

m . i I . . . 1 . i . . i 0

12 / 8 6 2 4 6 8 10 12 14 16 13 20 22 24 Months CUMULATIVE TECHNICAL SPECFICATiON VIOLATION 5 DAVIS-EESSE 15 l

e c

.E  !

? 10 -

.9 l

> 1 l

o

.2

-)

E p .-. .

,a s - -

0  : -

j l

v 0 . i I I I i i i i i I i I 4 I I I i i l i i I 4 12 / 8 6 2 4 6 8 10 12 14 16 18 20 22 24 Months

- s n, u .. - ~ ~

1 l

l nerd 6 (hME ' LERJi"Il4 E'4WT.DAll iS.VIA CAJ:t 39 THREE MILE ISL. 1 28995004 11/26/85 v!0L PkSMAR 24 T;:.EE MILE ISL. 1 29996001 01/03/86 VIOL PASEERR 26 THREE n:LE ISL 1 28986004 02/27/86 Y!DL PR$V. ERR 25 THFi! M:.E ISL. 1 26966005 03/11/i6 V I .. EDU!5 27 THREI =:iE ISL. 1 26986012 09/04/f6 i:0L FA!EEF8 15 ThRif MH E ISL. 1 26987003 04/02/37 VIOL FR$r. IRA 6 1 6 Ei ':LE ISL. I 26987005 05/06/87 i::L FRS4!;5 7 Ta !! *:.! 15L. 1 I!ii7009 09/27/i7 .::L FRSLIR; 33 CRIS*;. R D ER 3 302iS0;! 10/0eli5 .:A FA3Eis; 23 CRYSTA. A h ER 3 302i6006 04/26/36 iiDL FRO:iL; 37 CRYSIL RIVER 3 30286010 06/15/Bo VIOL FASEftR 35 CRYSTAL RIVER 3 30266012 07/31/86 VIOL FRSEERR 34 CRYSTAL RIVER 3 30266013 08/15/86 VIOL FRSEERR 36 CRYSTAt RIVER 3 30286015 06/14/96 VIOL PROCENR 33 CRYSTA. R hER 3 30286017 10/09/06 i!?L FRSEERR 32 CRYSTA. RIVER 3 30286019 10/2;/i6 i!:L F ::ED':.

31 Chi!a. RhE; 3 30256020 09/29/66 V:1 FFILEER 30 CRYSI:. RIVER 3 30286022 11/19/06 ii:L Fi!LIRR 29 CRYSTA n!VER 3 30286026 !!/18/66 V! L PROCENR 26 CRYSik RIVER 3 30296027  !!/13/96 V:X EGUF 16 CSYifA. R WER 3 30267003 03/27/07 V0 PRi%!;;

21 ChYSTA; RIVER 3 30287004 01/29/37 VI:t FRSEERR 10 CRYSTA; ;lVER 3 30297005 01/29/E7 V::L PR::EDLF 16 CRYSTA. Rul; 3 302S706 04/15/E7 i:A PRC !IU; 17 CRrS7A. ;biR : 302i7006 04/20/i7 .::L PA::iha 4 C0!*;. :DI: 3 30257010 06/12/!? i:X FX::::A 2 CE'!?A. : bi; ; 30227012 07/02i57 i: A FFiLi;;

I lb i! . : h ii 3 30257014 07/22/i7 i::. Fi!E*:"

3 CRfST:. : ht; 3 30:87016 06/03/97 V:k FA::i :

5 CD$ia. Rn!R 3 30237020 09/03/67 i::L FFC:!%;

6 C h i":. : bE: :  :':57023 Oi/;0/i? i::. F;::!:.:

4 C' ! M. :.i; : 302i' 4 6 11 :i/!' .::.  : it::'

13 CPS';. ' 3 !0"i702E 11/:iii7 .::. 75ih: -

22 D& :3 i !h 34:37002 0;/07/57 i :. F;iLI:;

11 DAki$-}!!!E 34667005 01/27/57 *::. FEiEiti 12 Ceili liiiE !46i7M5 03is7/97 v!i :s;LER3 19 th:5 !!!!E 24607007 03/25/!7 .::. FFili::

14 jii!!!E 78:5700i 04/05/67 i::' . F;::!:."

20 . ,IS-fE!!E 24657014 11/:5/!7 v:L FASti;F l

l l

l l

I

-,y -

e ,.

, y e *

• a h e E a o-v e

= e.

s a =

8

• d S Q w

V l E  :

e W

g m

5

c. a.:

3 I, i e 5 5 e

W$a i5 -

EE gs l 0 Y s

• n

'4 N0 dY ?s d lla yaI l .

i d = 1i i  :*E 3 i l $$

5N Gd j ii !5r  !  !

• u

, a da d a a a g a a g* -

g s

a e N l hn -

h!  ! IW e 0 , 8a i8 3di I  !. ,! 8EgI j u

8 - = sew rg g =

, er W

nl-!

a w.ge ea x .

1 w

I s ;. r a

a s

I e;g 5 f "+ z:'::

~

D :1 @ M ;R E ~! .: 5 = d r - 8a ra = "s aE- =  :

!]- *s * '

i, ,5 g w~

• 3 3

a l* "" _ g g ai e d g#a5d[n od ga

=a dv$

gdg g*I -

g

. t xd=8g+d as 8

g 3 ' l ]s r n

• w .E g g d

Ber ss e =

2 3 e a 8

e g ggs a s 3 = e* liyI

- d a" lci4 a

g e[t air

- =

laE m

s s=i e

g 8

8 m 3 i s rgn lg 5  ! 5 =p gca a- -a a a

d g g v g .: g =I e *

• G g "

g a

W

'a !i d 2 *

&9 E4%!!e33!Ig!5 i ? ;

m i is@jf35 e: ag=8  !!

g9!

lll$sa!i s

5 E = a 1 -

- E e = 0 =

!

• gl *r.

• 8.ial j

-g E

l

." sE eg]d: eg 5 55E,3 l

~

r a - '

e,- e i 8 e !IrEi E E "A V =

a agIEE g hI 8a5 = I, E 3fea e - e e a

g3IIiIle g -y & , 3 iy - en 1 e e 3 i e85 i - $.

r e s *

- M g h! B 3 e g 3 , e f g -

n p g ? gg d e !

A J * ~ 5 E $ 5 8 E a g5 x g $ % s - # g -

e3 - ? $m5 i !-< ! E E Ee ir i i g i g

• l r n E

• g i *r 5 E@E B i 5

i  ;!jZ*r r

! gi*5 I r$ i i r E Id! = 6 . g*

~4 e i r e3 8 g-y =y g* s# ,9 8 w i =

g a ' xs= 5 E I lI r e aelr - -

sa  : I y e ;= vgaI

= - I l E ] es lE ay E

=

n a at >

855 En 5ii!l5sei***fggfgg s5fg* 3 5

- = -

a g a a

C#

E n E' l e r I = = = = *=

s e-5le 8

• Ee i

u =

W 53E*o i

5- oy e

a E-W2 g s, .

a a

e5ra n ar rar =vi- 5si a.

= t r  !. 8 ,

• r=

j=rrr i 3g j g l g -

E e .

t1 w s s s a w r a e w w w s . c w w w w a a e e c w w w e h5 9 59 59 59 5555l9 9 59 i 9 59 9 559 9 9 9 5 a

assenssenessnesensassessenesses ge e o s c - --e a o u - a a c - s s e - - - - - a a a g b -

.m . . . . , .e. e. 0. . . . , . . . . . .

9i 69  ! s G#

9e.,...B s6e a a6 .Q#6 s s aea3=a a.08 . s s s s e 2 e = = = = =

=s a e $.

. ,, rR. . n - ~ . n ~ . ~

9111111111111111110111111111111 SEEEEEEE.EESEEEEEEEjEEEEEEEEEiii

.a n =a m

= n n w n - ~-n -

= - n -,- n n -

a n = a a = = =

• 5 A Y O - - -

r. n n n n a n m m n n ,a n n n n n <-

l m

u i 3 8 m

L's D

= e y a p i si.

ss i s:

da a =n iX [ 5I nn , as dI e 1

• a.

2 1i, ,

d E 4 y e i e .s ,.

.. =

s

,, n*

5-

• la '

lf l,a g

,f s

9 8 g

$ g8fI E3 a; s.

s j e st e-am

-E 1rs2 gng!:~ gw <s aI8a- E9s

~

A l E

r 5 5Ew I *5 i -

xasigei3:g=2 n x = -

y E z 1

Bil,l5! ag - a5  : !

i i

s -

I g x a: :8 I E E* a n E EE g 1 5g !ge i. . ;

a W

r -

i 8 E 1 I

g W

B g -[=s 8slEll_

gg

. .d vs:

i ,

1 1 l

E yeE*ignIEiI 5Ig,-a - l e

rm g l x s s ! ;l as,l w E l

i

. . . . . . . . .. 1 eae9 9 sn99 9 9W5Wec!###

ns -

aa=aaae B g4G 4g 9W *3.8 G: se 6 EG j ,

u o a sgs s =

11111111111 EEE E aExi ni R

= n c5 Eln , - a S

e u O O O O O O O O V U .

, 9 O O O O

- -