ML20135D007

Insp Repts 50-245/96-08,50-336/96-08 & 50-423/96-08 on 960827-1025.Violations Noted.Major Areas Inspected: Operations,Maint,Engineering & Plant Support

ML20135D007
Person / Time
Site:	Millstone
Issue date:	12/03/1996
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20135C977	List: ML20135C974 ML20135C995 ML20135D007
References
50-245-96-08, 50-245-96-8, 50-336-96-08, 50-336-96-8, 50-423-96-08, 50-423-96-8, NUDOCS 9612090203
Download: ML20135D007 (91)

See also: IR 05000245/1996008

Text

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

.

.

U.S. NUCLEAR REGULATORY COMMISSION

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

Report Nos.: 96-08 96-08 96-08

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

Licensee: Northeast Nuclear Energy Company

P. O. Box 128

Waterford, CT 06385

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

Inspection at: Waterford, CT

l

Dates: August 27,1996 - October 25,1996

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

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

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

A. L. Burritt, Resident Inspector, Unit 1

D. P. Beaulieu, Resident inspector, Unit 2

R. J. Arrighi, Resident inspector, Unit 3

R. J. Urban, Project Engineer, SPO

J. T. Shediosky, Senior Reactor Analyst

J. T. Furia, Senior Radiation Specialisi

L. M. Harrison, Reactor Engineer

V. L. Rooney, Project Manager, Unit 3 (NRR) l

J. H. Williams, Senior Operations Engineer l

N. Maguire-Moffitt, Contractor (PNL)

Approved by: Jacque P. Durr, Branch Chief

Special Projects Office

l

)

l

9612090203 961203

PDR ADOCK 05000245

G PDR s

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

O

I

TABLE OF CONTENTS

EXEC U TIV E S U M M ARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv

U1.1 Operations .................................................. 1

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

U108 Miscellaneous Operations issues (92700) ................. 3

U 1. Il M ai nte na nc e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

U1 M1 Conduct of Mdntenance ............................. 3

U1 M3 Maintenance Procedures and Documentation . . . . . . . . . . . . . . . 6

U1 M4 Maintenance Staff Knowledge and Performance . . . . . . . . . . . . . 9

U1 M8 Miscellaneous Maintenance issues . . . . . . . . . . . . . . . . . . . . . 11

U 1. lli Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12

U1 El Conduct of Engineering . ........................... 12 ,

U1 E8 Miscellaneous Engineering issues . . . . . . . . . . . . . . . . . . . . . . 13

U2.1 Operations ................................................. 18

U2 01 Conduct of Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

U2 O2 Operational Status of Facilities and Equipment . . . . . . . . . . . . . 18

1

1

U 2. ll M ainte na nc e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

U2 M2 Maintenance and Material Condition of Facilities and

Equipment ...................................... 19 i

U2 M8 Miscellaneous Maintenance issues . . . . . . . . . . . . . . . . . . . . . 20  !

1

l

U 2. lli Engi ne e ri ng . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

U2 E1 Conduct of Engineering ............................. 27

U2 E8 Miscellaneous Engineering issues . . . . . . . . . . . . . . . . . . . . . . 29

U3.1 Operations ................................................. 39

U3 01 Conduct of Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

US O2 Operational Status of Facilities and Equipment . . . . . . . . . . . . . 40  ;

U3 05 Operator Training Qualification . . . . . . . . . . . . . . . . . . . . . . . . 41 I

U3 07 Quality Assurance in Operations . . . . . . . . . . . . . . . . . . . . . . . 45

U3 08 Miscellaneous Operations issues (92700) ................ 46

U 3. Il M ainte na nc e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

U3M8 Miscellaneous Maintenance issues ..................... 48

U 3. Ill E ngi ne e ring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

U3 E2 Engineering Support of Facilities and Equipment . . . . . . . . . . . . 53

U3 E8 Miscellaneous Engineering Issues . . . . . . . . . . . . . . . . . . . . . . 56

IV Plant Support ................................................. 61

R8 Miscellaneous Radiological Protection and Chemistry Issues . . . 61

ii

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

.

!

i

. l

S2 Status of Security Facilities and Equipment ............... 65

S8 Miscellaneous Security and Safeguards issues . . . . . . . . . . . . . 65  ;

P6 Miscellaneous EP Issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 j

F1 Control of Fire Protection Activities . . . . . . . . . . . . . . . . . . . . . 67 '

F4 Fire Protection Staff Knowledge and Performance .......... 70

F7 Quality Assurance in Fire Protection Activities . . . . . . . . . . . . . 72

F8 Miscellaneous Fire Protection issues . . . . . . . . . . . . . . . . . . . . 73 1

!

V. M a nagement M eetings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

l

X1 Exit Meeting Su m m ary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 '

X3 Management Meeting Summary . . . . . . . . . . . . . . . . . . . . . . . 74

i

r

(

,

t

I

,

l.

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

. -

' *

EXECUTIVE SUMMt.nY

j Millstone Nuclear Power Station

Combined Inspection 245/96-08; 336/96-08;423/96-08

l

l Operations

f

e Shift turnovers at Unit 1 were conducted in a professional manner. The format for

the turnover briefings was consistent between operations crews with an exchange  ;

. of information necessary to convey plant status to the oncoming shift. (Section l

i

U1.01.2) )

.

e The failure to promptly address two potentially safety significant issues indicates a

lack of a questioning attitude by the Unit 1 staff. The issues concerned a fuel I

bundle that was found not fully seated in the spent fuel rack and the handling of the '

cracked fuse ferrules problem. Furthermore, the failure of the management review l

3 team to ensure the appropriate sensitivity and response to these issues indicates j

poor management oversight of emergent issues. The delays in processing adverse '

,

condition reports (ACRs) creates additional vulnerabilities that could prevent the  ;

i

prompt assessment of safety significant issues. The continued identification of I

discrepant conditions in the spent fuel pool indicates the need to accelerate the

evaluation portion of the spent fuel pool cleanup / recovery, plan. All discrepant I

! conditions warrant identification and evaluation in the short term to ensure that the l

collective impact of these issues is addressed. (Section U1.01.3)

1

i e At Unit 2, overall operator performance in the monitoring and operation of systems

necessary for safe shutdown was found to be good. (Section U2.01.1)

.

It

e The NRC found that the licensee failed to lock open the refueling pool drain valves,

as required by the final safety analysis report (FSAR), to prevent water

4 accumulation in the refueling pool rather than the containment sump following a

loss of coolant accident. The resulting operating procedure change was inadequate

in that it f ailed to ensure that the drain valves were locked open and is considered

i an apparent violation. (Section U2.02.1)

i

! e in addition, the licensee failed to perform the technical specification (TS) required

'

monthly containment boundary valve lineup at Unit 2 because: (1) The valve lineup

procedure did not include all required valves and (2) operators documented as not

i applicable the valves inside containment while at power even though the TS did not

,

allow for this exemption. This was characterized as a violation. In the resulting

licensee event report, the licensee committed to review other TS required valve

'

lineups to ensure required valves were included. Although the licensee mistakenly

stated and the corrective action tracking system indicated that the TS review was

complete, the NRC found that the review had not been performed. The failure to

! implement this corrective action was also characterized as an apparent violation.

4

(Section U2.M8.4)

e A Region i evaluation of the Millstone Unit 3 licensed operator requalification

training (LORT) program concluded that the facility's performance was good. The

,

iv

i

t

i

-- - -

__ _ _ __ _ _ _ __ __

.

i

~

,

'

six written examinations reviewed met all guidelines and contained no repeat

questions among the examinations; an indication of high quality examinations.

i Operator performance and facility evaluations on the dynamic simulator were i

eatisfactory. Remediation plans for weak operator performance were effective, but

1

some minor discrepancies were identified. (Section U3.05.1)

i e Initiallicensee actions to address the concerns raised by the discovery of cracked

fuse ferrules were inadequate until the Nuclear Oversight Organization became

involved in corrective action followup. Subsequently, resolution of the issue was

<

not prompt until NRC intervention prompted the proper focus and direction by the

licensee. (Section U3.07.1)

Maintenance

i

j * The Unit 1 maintenance activities associated with the restoration of control rod

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

were well planned and carefully executed. Team work and procedure adherence

were emphasized by a strong management presence throughout the work. Good

-

management oversight was noted throughout the two day evolution. (Section

U1.M 1.1 )

!

l * A system engineer identified a fluctuation in the field voltage for the gas turbine

generator during a Unit 1 surveillance test. Although the identification of the

, voltage fluctuation was good, the engineer's assessment of the significe.ce of that

!

problem was less than adequate, and the issue was not raised to the attention of

the operations shift manager for input. This indicates a reluctance on the part of

'

the licensee's staff to raise potential problems to management. While the system

engineer did ultimately document the concern in an ACR, the three-week delay

j prevented a proper assessment of the issue. Once identified, maintenance and

l engineering worked to resolve the problem and correct the deficiency. (Section

, U1.M1.2)

,

• The quality of the Unit 1 work order packages continues to 'be a problem with many

automated work orders (AWOs) returned to planning or not completed due to the

inadequacy of the work package. Work is scheduled using two software tools, P-2

-

and PMMS, which present coordination problems, since system engineers and other

plant work groups input to the P-2 schedule and operations works from the PMMS

i frozen schedule. Management oversight and engineering input and support to the

4

work control process are weak. Specifically, management did not question work

that was not completed on time or work that was scheduled and not started.

(Section U1.M3.1)

$

.

e The failure to provide a troubleshooting plan and troubleshooting guidelines with a

4

Unit 1 AWO package is an apparent violation of Technical Specification 6.8.1. The

control functions of a differential pressure (DP) transmitter were not properly

assessed by the I&C planner or the supervisor in charge of the work. Additionally,

the supervisor missed a second opportunity to identify the error when questions

about the clearance adequacy were raised by the technician. A similar event

i

,

v

,

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

t .

'

occurred on May 10,1995, while working on DPT-4-121 for the 'B' travelling

screen DP transmitter. The issue of inadequate corrective actions for the May 1995

event, should be addressed as part of the licensee's response to the procedural

violation stated above. (Section U1.M4.1)

• At Unit 2, due to poor maintenance and the absence of routine testing, the

circulating water pump trip, which is designed to protect the auxiliary feedwater

pumps from flooding, was found inoperable due to a miswired power supply cable.

j This item remains open to allow NRC review of the licensee's planned design basis

j change to not credit this trip. (Section U2.M8.3)

• The failure to include SPEC 200 electronics in reactor protection system and

'

engineered safeguards actuation system response time testing at Unit 2 is

considered unresolved to allow further review of how this error was missed during

multiple modifications and to evaluate whether cables must also be included.

(Section U2.M8.5)

! * The initial corrective actions regarding the failure to test an enclosure building i

, filtration system (EBFS) interlock at Unit 2 were good; however, overall corrective

action was considered unacceptable. Due to inadequate follow-through, the

Licensee Event Report (LER) commitment to complete a generic letter review in

September 1996 was not completed. (Section U2.M8.6)

) * Although the licensee reported that Unit 1 heavy loads, as well as Unit 2 heavy ,

loads, have been lifted over a Unit 2,480 Vac vital switchgear room, neither the

'

, LER nor the corrective action tracking system reflected the need for Unit 1

corrective actions. This was characterized as an apparent violation. (Section

j U2.M8.7)

* Numerous Unit 3 inservice testing program deficiencies were identified and reported

by the licensee. The licensee initially failed to perform a root cause evaluation of

l the programmatic concerns because of the improper classification of the relevant

, adverse condition report. Further, the total scope of the discrepancies was not

initially communicated to the NRC and the development of a comprehensive .I

corrective action plan was not in evidence. This issue remains unresolved pending

,

s completion of all required testing. (Section U3.M8.1)

l

!

j Engineering

> * The Unit 1 implementation of SPROC 96-1-35, " Control Room Temperature

,

Monitoring with Normal Ventilation Secured," was well coordinated' with adequate

I controls to ensure personnel safety in the main control room during the test. The

!

licensee's safety evaluation, performed for the special procedure, concluded that

.

the test did not create an unreviewed safety question. The inspector reviewed the i

safety evaluation and determined it to be adequate. (Section U1.E1.1)

4 * The licensee concluded in a Unit 1 operability evaluation that the SGTS remained

'

operable with outside ambient temperatures of 2 20 F based on the results of a

vi

1

l

.

.

surveillance test procedure. The licensee reassessed their original SGTS operability

determination and determined that it was not correct. The incorrect operability

determination was the result of inadequate SGTS testing. Substituting the lower l

draw-down capability into the operability determination calculation resulted in the '

SGTS being operable down to only 45 F for the "B" train and 30oF for the "A"

train. A review of ambient temperature since July 5,1995, revealed that on several

occasions during power operation the ambient temperature dropped below 45 F, l

with a low of 26 F. The failure to maintain the SGTS operable under all conditions '

is an apparent violation of Technical Specification 3.7.B.1, Containment Systems.

(Section U1.E8.3)

• At Unit 1 the licensee failed to implement the design modifications necessary to

bring the CRD system into design compliance within the NRC specified time period.

This is an apparent violation of Appendix B, Criterion 16, " Corrective Actions." 1

The licensee also failed to expand the scope of their review based on the significant I

number of CRD design and maintenance deficiencies identified extemal to

containment, in the early 1980's. Further, it appears the licensee failed to

implement a commitment when the CRD evaluation was limited to only the portions  !

of the system external to the containment based on their response to the NRC's '

request for additional information. The licensee failed to demonstrate the operability

of the CRD system, consistent with GL 91-18 methods, following the identification

of seismic deficiencies found during the cycle 14 refueling outage. This issue is

also unresolved pending the revision of the calculations related to CRD operability

prior to the cycle 14 refueling outage. The licensee could not demonstrate that all

of the seismic issues identified during the SEP re-evaluation were resolved. This

issue is unresolved pending NRC verification of the closure for all items identified.

(Section U1.E8.4)

• Following a review of the licensee's plans for entering Mode 6 at Unit 2, the NRC j

had concerns regarding the licensee's intent to perform a core offload using i

systems which, although operable, had known discrepancies that were contrary to i

the current operating license. Although no violations of NRC requirements were l

identified, this is considered to be a significant weakness in light of recent attention  !

-

given to compliance with the current design and licensing basis. (Section U2.E1.1) l

• Contrary to the Unit 2 technical specifications, both trains of containment air l

hydrogen monitors were inoperable because: (1) The NRC identified that there '

would be insufficient containment air flow past the thermal conductivity cell at low

containment pressures, and (2) The hydrogen monitors were taking a non-

representative suction from ductwork of the non-vital containment auxiliary

recirculating fans. This was characterized as an apparent violation. The NRC also

found that due to an inadequate review of the steam generator replacement

modification, the licensee failed to identify that the design basis and licensing basis

time period for placing the hydrogen monitors in service and taking an containment

atmosphere sample could not be met. The failure to establish adequate design

control measures for this modification was considered an apparent violation. In

addition, the NRC found that the Unit 2 FSAR had not been updated to reflect the

licensing basis regarding the amount of time following an accident that the

vii

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

-

i

"

-

hydrogen monitors would be placed in service. This was also characterized as an

apparent violation. (Section U2.E8.1)

• The licensee discovered that both trains of the Unit 2 reactor building closed cooling

4

water system could become inoperable following certain design basis events if

inventory were to be lost through the common primary makeup water system fill

line. This licensee-identified and corrected concern was characterized as a non-

cited violation. (Section U2.E8.3)

• A single failure vulnerability created by the hydrogen purge valve /EBFS heater

!

"

interlock was discovered at Unit 2 during the performance of an EBFS surveillance

reflected a good questioning attitude by plant personnel. This design deficiency

4

was characterized as a non-cited violation. (Section U2.E8.5)

, * The failure of the service water backwash valve solenoids to meet the FSAR flood

4

control requirements was a licensee-identified design discrepancy and is considered

a non-cited violation. (Section U2.E8.6)

,

• The appropriateness of removing the startup rate trip from the reactor protection

'

system in 1978 is considered unresolved pending NRC review of the licensee's

disposition of this concern. (Section U2.E8.7)

,

* The licensee's review of NRC IN 88 24 and GL 91-15 was inadequate. As a result, I

4

the potential exists for specific solenoid-operated valves (SOVs) to fail to perform )

'

their intended safety functions because of excessive operating pressure '

differentials. This can result from failures of non-qualified air regulators installed in j

the instrument air system located upstream of the SOVs. This potential design error

is considered to be an unresolved item. (Section U3.E2.1)

,'

• NRC Review of UFSAR and 1994 supporting documents revealed that more recent

data on the time for operator actions in the event of a steam generator tube rupture

.

(SGTR) had been obtained by the licensee. It was determined to not be

conservative and appeared not to have been reported to NRC. The UFSAR analysis

still used the shorter operator response time. Subsequent to the inspection, facility

preliminary analyses of the SGTR using'the more recent data determined that the

accident consequences were acceptable. The facility planned to finaiize these

analyses and submit the results to the NRC. This issue represents an unresolved

j ltem, pending NRC review of the updated analyses. (Section U3.05.2)

Plant Support

• A problem in the oversight and control of one aspect of the licensee's improving

Station Performance (ISP) program was identified. Licensee management allowed

! the site Material Condition Program (MCP) to be discontinued without recognition

<

that it was part of the ISP program. There was no comprehensive MCP in effect at

any of the units for a period of approximately nine months. The licensee's

. characterization that none of the Unit 3 issues identified in the site MCP were

,

required to be corrected prior to restart was determined to be appropriate. The

viii

1

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

.

l' review of the Unit 1 and Unit 2 MCP is considered an item for further inspection

4 followup. (Section U3.R8.1)

1

e in general, engineering work had not been prioritized or resolved commensurate

.

1

with its significance, and there was a lack of direction for resolving fire protection

issues. (Section F1.1)

j_ * Recently implemented initiatives by the Site Fire Protection Department were

noteworthy including the permit processes for comprehensively assessing plant

i conditions prior to authorizing fire risk work. (Section F1.2)

• Fire fighting training effectiveness showed some weakness, as demonstrated by a

drill failure on a scenario the operators had previously seen during training. The

remedial drill demonstrated adequate staff fire-fighting skills. (Section F4.2)

e Quality assurance audits of fire protection were limited in scope, sometimes

incorrectly characterized findings as low significance, and failed to follow up on

previously identified issues. (Section F7.1)

• The licensee continues to maintain a very effective program for the transportation of

radioactive materials. The Radwaste Remediation Project at Unit 1 continues to

make progress in addressing the material condition deficiencies in the Unit 1 liquid

waste processing systems and facilities. However, a lack of appropriate

management focus on liquid radwaste processing systems continues to exist at

Unit 3. (Section R8.2)

ix

=-

.

C

,

.

Report Details

Summarv of Unit 1 Status

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

licensee continues to review the plant's level of compliance with regulatory requirements,

and compliance with their established design and licensing basis, associated with an NRC

request pursuant to 10 CFR 50.54(f) and Confirmatory Orders. The primary focus of

operations this period was the restoration of the reactor water cleanup and control rod

drive systems, as well as rolling the main turbine. An increased emphasis was also placed

on activities associated with placing out of service systems in a layup condition.

U1.1 Operations

U101 Conduct of Operations

01.1 General Comments (71707)

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

plant operations to ensure that licensee's controls were effective in achieving continued

safe operation of the facility. The inspectors observed that proper control room staffing

was maintained, access to the control room was properly controlled, and operator behavior

'

was commensurate with the plant configuration and plant activities in progress in general,

the conduct of operations was professional and safety-conscious; specific events and

noteworthy observations are detailed in the sections below.

01.2 Shift Turnover Briefina

a. Insoection Scoce (71707)

The inspector observed shift turnover briefings in order to verify that all necessary

information concerning plant systems status was discussed during shift turnover and

i understood by the oncoming shift.

b. Observations and Findinas

.

The inspectors observed shift turnover briefings periodically throughout the inspection

period. Different shift managers were observed in order to assess the consistency

between shift crews. Overall, shift turnovers were conducted in a professional manner.

The briefing was attended by the control room operators, plant operators, and

representatives from the health physics, chemistry and radioactive waste departments.

Each of the participants were given the opportunity to report on plant status and provide

turnover information from the previous shifts. The control room operators reviewed the

shutdown risk assessment chart and reported on systems credited for safe shutdown. The

shift manager discussed plant conditions and reviewed the operability and availability

determinations for compensatory actions required by the shift. The ACR initiated on the

previous shift are discussed, as well as any issues documented in the " Shift Manager's

Notes," a running list of operational information. The shift managers discussed planned

activities for the day and any expected evolution or surveillance tests that needed to be

performed. The inspectors noted that the same format was used by each shift manager

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

.

.

2

and that the information prsented was consistent between shifts. Additionally, it was

apparent from the questions by the shift crews that the turnover briefing was a dialogue

with a detailed exchange of information between shift management and the plant workers.

c. Conclusions

.

Shift turnovers were conducted in a professional manner. The forrnat for the turnover

briefings were consistent between operations crews with an exchange of information l

necessary to convey plant status to the oncoming shift.  !

i

4

01.3 Manaaement Oversiaht of Emeraent issues

! I

i a. Insoection Scooe (71707)

l A review of licensee self identified issues was performed to verify that each item received

j prompt management assessment and followup as necessary. l

l

j

b. Observations and Findinas

l

,

On September 11,1996, a site wide ACR M3-96-0759 was initiated to address fuses

i

issued from the warehouse that were found to have cracked ferrules. This issue was

assigned to the warehouse for resolution. Two weeks later QAS initiated ACR M1-96-

} 0608 for the failure to address the possible plant effects from this issue at each of the

, units. The lack of prompt follow-up to the initial ACR and the oversight by the quality

J

assurance department is further discussed in section U.lli.O.7.1 of this report.

, The second ACR on fuses was not assessed by the Unit 1 management review team (MRT)

until five days after being issued. The ACR was again assigned to the warehouse for

'

' resolution; however, this time each unit was assigned to assess the possible impact on -

, plant operations and document this assessment in an operability determination (OD). The  !

i Unit 1 MRT assigned a 30 day deadline for completion of the assessment and development l

of the OD. Two days later the inspectors discussed the lack of urgency assigned to this

issue, which could impact systems credited for operability or shutdown risk availability,

with the plant management. Subsequently, the licensee fully assessed and documented

{ the issue within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of the discussion with the inspector,

r

Within the same week and subsequent to the inspectors intervention on the fuse issue, a

a second instance was identified in which the licensee did not promptly assess and followup

, on a potentially safety significant issue. In this case a fuel bundle was found not fully

seated in the spent fuel rack during the recovery of entangled control rod blades (section

1

U 1.M 1.1 ). ACR M1-96-0646 was generated and reviewed by the management team;

however, no immediate follow-up inspections or evaluations were performed. The licensee

did not determine how wide spread the problem was nor was the potential for fuel damage

or effect on the criticality margin assessed. The licensee subsequently periormed

1 additional inspections and identified a total of 56 fuel assemblies that were not fully seated

1 in the spent fuel racks. An evaluation was performed to address all relevant issues

! including: the effect of a bundle drop on the fuel bundle, the fuel rack and spent fuel pool

i liner, seismic response of the fuel racks, the criticality margin, fuel assembly cooling, and

.

.

3

water shielding. As corrective action for this issue, ACR M1-9S-0697 was initiated to

address the failure to perform a prompt evaluation of the unseated fuel assembly. The

inspector reviewed the licensees subsequent assessment of the unseated fuel bundles and

found the evaluation to be adequate.

c. Conclusions

The failure to promptly address these two potentially safety significant issues indicates a

lack of a questioning attitude by the unit staff. Furthermore, the failure of the management

review team to ensure the appropriate sensitivity and response to these issues indicates

poor management oversight of emergent issues. The delays in processing ACRs creates

additional vulnerabilities that could prevent the prompt assessment of safety significant

issues. The continued identification of discrepant conditions in the spent fuel pool

indicates the need to accelerate the evaluation portion of the spent fuel pool

cleanup / recovery plan. All discrepant conditions warrant identification and evaluation in

the short term to ensure the collective impact of these issues are addressed.

U108 Miscellaneous Operations issues (92700)

08.1 (Closed) LER 50-245/96-02: operation with reactor pressure in excess of the limits

described by the design basis. This event was discussed in inspection Report 50-245/95-

42. No new issues were revealed by the LER.

08.2 (Closed) LER 50-245/96-41: the deviation from technical specifications when

standby gas treatment was secured and the reactor building ventilation system was

unisolated to ventilate the drywell following a small electrical fire in the drywell. This

event was discussed in Inspection Report 50-245/96-06, section U1.01.3." No new issues

were revealed by the LER.

08.3 (Closed) LER 50-245/96-07: the gas turbine generator was determined to be

inoperable for more than the allowed outage time, as a result of a fuel forwarding pump

being degraded and also out of service for maintenance. This event was discussed in

Inspection Report 50-245/96-01, section 1.2.2. No new issues were revealed by the LER.

U1.Il Maintenance

U1 M1 Conduct of Maintenance

M 1.1 Soent Fuel Pool Tri-Nuclear Filter Removal

a. Insoection Scoce (62707)

On July 16,1996, plant personnel were removing a temporary filter assembly from the

spent fuel pool when a wire rope, attached to the filter assembly, was entangled with

control rods that were suspended from the spent fuel pool equipment rail. During this

inspection period, the inspector observed activities associated with removal of the

- . = - - - -

.

.

4

,

Tri-Nuclear filter from the spent fuel pool. The evolution included the movement of the

displaced control rods in order to lower the Tri-Nuclear filter to facilitate cutting of the

entangled wire rope. The inspector attended the pre-job briefings, and discussions were

held with maintenance workers, operators, and supervisors. The inspector observed the

performance of special procedure (SPROC), SPROC 96-1-36, Control Rod Restoration and

Tri-Nuclear Filter Removal,

b. Observations and Findinas

On October 1,1996, the inspector observed the pre-job briefing for the Tri-Nuclear filter

removal. All work groups were present for the briefing that was conducted by the

maintenance first line supervisor (FLS). Two work orders were prepared for the work to

cover the movement of the control rod blades and the Tri-Nuclear removal. Two special

procedures were used for the evolution SPROC 95-1-25, " Temporary Fuel Pool Filter

Operation," and SPROC 96-1-36, " Control Rod Restoration and Tri-Nuclear Filter Removal."

Both of the SPROCs were reviewed, step-by-step, during the briefing. Health Physics

personnel provided an overview of the potential significant radiation hazard associated with

the handling of control rod blades and reviewed the requirements of the radiation work

permit. During the briefing, roles and responsibilities were clearly delineated for all

activities. The first SPROC, 96-1-36, was used to install a control rod containment net

fabricated from stainless steel wire cables. The net was placed under the suspended group l

1

of six control rods, which were suspended twelve inches to three foot off the floor of the i

spent fuel pool. The control rods were lowered simultaneously, using come-alongs from

the refuel bridge, as the Tri-Nuclear filter was lowered using the auxiliary refuel floor crane.

Once the control rod blades were resting on the floor of the spent fuel pool, the wire rope j

attached to the filter assembly and tangled on the control rods was cut, freeing the  ;

Tri-Nuclear. During the performance of the recovery procedures, an underwater video I

camera was placed in-the spent fuel pool to monitor the evolution. The high resolution I

color camera provided excellent coverage of the work. The filter assembly was removed

from the spent fuel pool and placed in a transportation box on the refuel floor using SPROC

95-1-25. ,

1

Management oversight was provided by a FLS and general supervisor from maintenance,

and a health physics supervisor present at all times. Additionally, a quality assurance

inspector observed the activities. The performance of several practice evolutions ensured

that the actual work went as planned with no deficiencies noted. During the work, a

4

particular emphasis was placed on team work and procedure adherence. Se final phase

of the recovery plan will be completed at a later date.

I

c. Conclusions

The maintenance activities associated with the restoration of control rod blades and the

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

carefully executed. Team work and procedure adherence was emphasized by a strong

management presence throughout the work. Good management oversight was noted

throughout the two day evolution.

_ _ . _ _ _ _ _ _ _ -

_ _ . _ _ _ . _ . _ _ _ _ _ . . __

<

!

-

! 5

'

M1.2 Gas Turbine Generator Surveillance Testina

i i

! a. Insoection Scone (37551)

-

The inspector reviewed an ACR M1-96-0599, concerning the performance of a gas turbine

, generator surveillance test. During the test, the system engineer noted that the generator

i field voltage periodically dropped from 150 Vdc to 110 Vdc and then recovered. The

inspector conducted interviews, reviewed documentation, and observed follow-up testing.

i

l b. Observations and Findinas

.

j On August 29,1996, surveillance procedure 668.2, " Gas Turbine Emergency Fast Start,"

2

was performed to verify the fast start function of the gas turbine generator. The turbine

l was started and the generator loaded without incident, however, during the test it was

l noted that generator field voltage dropped and recovered almost immediately. The system ,

i engineer also noted that frequency and severity of these transients decreased as the 1

! surveillance progressed. The test was completed and the gas turbine was declared

l available in the shutdown risk assessment. Generator field voltage was not part of the

! acceptance criteria for the surveillance test and therefore, the test was considered  !

! satisfactory. It should be noted that field voltage is only indicated locally, with no remote

j

'

indication in the main control room. Following the test, the system engineer obtained a

computer printout of indicated field voltage during the test. This printout was reviewed by

j the system engineer and a supervisor from the generation test group, and they concluded

that the field voltage fluctuations were not a problem. This issue was not discussed with

the shift manager following the test and as a result, operations was not consulted about

, the significance of the field voltage problem.

!

On September 23,1996, prior to the next scheduled surveillance test on the gas turbine,

i the system engineer decided that an ACR should be generated to document the field

f voltage problem. The system engineer informed the inspector that after reconsidering his

j original assessment of the significance of the field voltage problem, he determined that an

'

ACR was required. In response to the ACR,' the shift manager requested that the system

engineer provide an Availability Determination for the gas turbine. Following this i

determination, the gas turbine was declared unavailable due to the possibility that the I

j voltage would not be stable.

i

! On September 27,1996, the inspector observed the performance of surveillance procedure

j SP 668.2 on the gas turbine. During that test, generator load was increased in stages to I

} allow the representatives from the generation test and technical support departments to

I- evaluate the performance of the voltage regulator and the effects as seen by the field

j . voltage indicator. The load was increased to 8 Mw electric when some oscillations in the

i field voltage were noted. The surveillance continued until the turbine was fully loaded at j

} 11 Mw. Throughout the remainder of the test, some small transients were seen where

j. field voltage dropped from 155 Vdc to 140 Vdc. The test was completed satisfactorily

j and an Availability Determination was written that stated if there was a loss of power, the

i gas turbine would start and it would achieve rated voltage and frequency, and the terminal

! voltage would remain stable at 4160 Vac, therefore, the gas turbine was available to

j support shutdown risk. The determination also stated that the loads necessary to maintain

. .

1

- , . _ , _ -

.

.

6

shutdown cooling of Units 1 and 2 are estimated to be approximately 3-5 Mw and at these

loads, the gas turbine was demonstrated stable with field voltage at 155 Vdc.

Subsequent discussions with General Electric, the gas turbine vendor, indicated that the

field voltage problem could have been caused by dirt on contacts in the potentiometers in

the voltage regulator circuit. The potentiometers were cleaned and a retest surveillance

was completed with no indicated fluctuations in field voltage. An ACR, M1-96-0607 was

generated to document the fact that the field voltage problem was not properly addressed

for approximately three weeks.

c. Conclusions

A system engineer appropriately identified a fluctuation in the field voltage for the gas

turbine generator during a surveillance test. At the time, the engineer's assessment of the

significance of that problem was less than adequate, and he did not raise the issue to the

attention of the operations shift manager for his input. While the system engineer did

ultimately document his concern in an ACR, the three-week delay prevented a proper

assessment of the issue. Once identified, maintenance and engineering worked to resolve

the problem and correct the deficiency.

U1 M3 Maintenance Procedures and Documentation

M 3.1 Work Control Process l

a. Insoection Scooe (62700)

The inspector reviewed the work management and control process at Unit 1 with a focus

on planning, automated work order (AWO) quality, scheduling, and engineering support.

The inspection consisted of a review of work control procedures, interviews, and direct

observation of the process.

b. Observations and Findinas

in June of 1996, plant management instituted an interim work management process, as

part of an initiative to establish schedule adherence. The goal of the process was to

produce a credible schedule which allowed work groups to accomplish their schedule

activitier in accordance with the published schedule. The interim process called for

freezing the work scope one week before the start of the work week. On August 20,

1996, the work scope was frozen an additional week. Under the new process, all AWOs

that are ready to be worked during the T-2 work week are delivered to operations work

control by 2:00 p.m. on Wednesday, two weeks before the actual work is scheduled. By

this process, " ready to work" means that the AWO is developed with all prerequisite

actions such as parts procurement, or procedure revisions completed. The Tuesday and

Thursday planning / schedule meetings take place each week and are attended by managers

from all the major plant disciplines. During these meetings, a multi-disciplined review is

performed on all associated work activities, with a final review of the work list for the T-2

work week, and the T-1 frozen work list. Operations work control is then responsible for

performing a review of the work packages, retest requirements, and preparation of

_ __

_ . _

.

.

7

clearances in advance of the work. Additionally, they hang and remove clearance TAGS in

support of work and interface with the control room.

Plannina and AWO Quality

1

The quality of work order packages being received by the work groups continues to be a j

problem (see NRC Inspection Report 50-245/96-01). Interviews with the plant staff

indicated that approximately 50% of the AWOs that leave planning are returned or not

completed due to the adequacy of the work package. Additionally, as stated above,

design problems are not being resolved before the work packages are sent to the work  !

groups. This places an additional burden on the first line supervisor (FLS) to review and

correct packages before work can begin. In an effort to resolve the AWO quality problem, 1

l&C supervisors meet twice a week with the I&C planner and review the AWO when they

, are stillin the development stage. The l&C department assigned a technician to assist the

planner with system walkdowns and the scoping of work.

The inspector reviewed a special procedure associated with a control room ventilation test.

There were two AWOs written for the test, one for l&C and one for maintenance. The l

AWOs had the special test procedure included in the packages, which were in a draft form,

requiring PORC approval before implementation. According to the planner, he did not have  ;

sufficient time to include an approved version of the special procedure in the work package l

for the FLS review and approval. The procedure was sent to the PORC on the Wednesday

the AWO was to be included in the frozen work list. As a result, the FLSs reviewed and

signed the work packages with a draft special procedure. In this case, the engineering

delay in preparing the special procedure resulted in the planner rushing to meet the frozen

work schedule. Having the FLS review a work package with draft documentation is a poor

practice, since the PORC review could have changed the final approved procedure. The

problem of scheduling AWOs on the frozen work list when in fact, they are not " ready to

work," appears to be a common practice. Some planners indicated that AWOs are

'

routinely placed on the two-week frozen work list when they may be waiting for parts or

have engineering issues that need to be resolved before work can begin. As a result, if

work does not start as scheduled, it is re-scheduled with no discussion about why the

work package was not complete and ready to be worked.

Schedulina Work

.

t

Currently, work is scheduled using two software tools, project-2 (P-2) and the plant

maintenance management system (PMMS). The P-2 schedule is maintained by the

scheduling department and is a logic driven tool. The PMMS schedule is maintained by the

"

work planning department, and is based on work start dates from a listing of AWOs,it is

not logic-tied. This presents challenges to the staft because the two systems only

communicate in one direction (PMMS downloads to P-2). Changes in the P-2 schedule are

not automatically reflected in the PMMS schedule, they must be manually entered. PMMS

is used to input items on the Tuesday and Thursday Planning Schedule, as well as the

AWO frozen work list. The use of two scheduling tools presents coordination problems

since system engineers and other plant work groups input to the P-2 schedule and

operations works from the PMMS frozen schedule. This places an additional burden on the

planners to ensure that all P-2 scheduled work is captured on the PMMS schedule. This

l

.

8

coordination problem is then reflected down to the worker level since a first line supervisor

needs to review three schedules, P-2, Tuesday / Thursday, and the frozen work list

schedule, to determine what work he needs to prepare for in the future.

The inspector attended a Thursday planning / scheduling meeting. Each work group

representative performed a line-by-line review of the work list and provided a status on

each item. The inspector noted a lack of dialogue concerning why work was not

completed on time or work that was scheduled and not started. Schedule adherence did

not appear to be a concern for the group. Missed work was just reschedu!sd with no

effort to determine why an activity was not performed in accordance with the c<:hedule. A

problem with work coordination was also identified by operations. The reactor water

cleanup system was scheduled to be taken out of service on two consecutive weeks for

two different activities. A last minute schedule change combined the activities into one

week. In another instance, an AWO was signed, reedy to work, and sent to ope /ations

work control; shortly after the AWO was lost. There was no discussion about what

happened to the AWO, the direction was just to reprint the AWO and re-schedule the

'

work. Recently, the planning supervisor initiated a performance monitoring system, which

uses key performance indicators, to track the status of AWOs and trouble reports (TR).

,

Enaineerino Sucoort *

The inspector reviewed the engineering departments role in the work control process. Both

system engineers and design engineering are required to input to the process. Interviews

with planning and maintenance personnel indicated the system engineers do not take an

active role in the planning and scheduling of work on their respective sys'tems. Some

system engineers are not aware of the outstanding TRs for their systems. In some cases,

the newer engineers cannot access PMMS due to their unfamiliarity with the system. I

Planners normally contact the system engineers only on an as-needed basis to help resolve

engineering related issues or assist in system walkdowns.

Design engineering support is also lacking, as evidenced by a large number AWOs requiring

design engineering disposition. Discussions with one of the mechanical planners indicated

that as many as one-half of the corrective maintenance mechanical AWOs require design

engineering support to resolve problems in the field. Work packages ~o being delayed in  ;

the planning stages, and work in the field is delayed due to the need for design engineering l

support. The areas of design basis configuration and in-field testing requirements need 1

additional engineering support and involvement on a daily basis. The inspector reviewed ,

AWO M1-96-09409, concerning the shutdown cooling heat exchanger. The heat l

exchanger relief valve was being replaced with a relief valve that received a commercial

grade dedication, and the maintenance FLS noted that the AWO did not specify the valve's  !

reset pressure setting for it functional test. The FLS and the mechanical planner were l

working on resolving this issue with procurement and design engineering, and after three I

weeks the issue had not been resolved.

l

1

___ _ _ _ _ _ . ._. _

.

.

9

c. Conclusion

The inspector concluded that the quality of work order packages continues to be a problem

with many AWOs returned to planning or not completed due to the adequacy of the work

I package. This places an additional burden on the FLS to review and correct packages

before work can begin, or approve work orders based on draft documents. Work is

scheduled using two software tools, P-2 and PMMS, which presents coordination problems

<

since system engineers and other plant work groups input to the P-2 schedule and

operations works from the PMMS frozen schedule. Engineering input and support to the

,

work control process is weak as evidenced by: system engineers not taking an active role

in the planning and scheduling of work on their respective systems, and work packages

!

being delayed in the planning stages and work in the field due to the need for design

.

engineering support. A lack of management oversight in the process was noted by the

inspector. Specifically, management did not question work that was not completed on

time or work that was scheduled and not started.

'

U1 M4 Maintenance Staff Knowledge and Performance

M4.1 Troubleshootina Activities

4

a. Insoection Scooe (62700)

! The inspector reviewed automated work order (AWO) M1-96-08061, concerning the

travelling screen 'E' differential pressure (DP) transmitter DPT-4-124. Both the local and

remote indications for the transmitter were indicating downscale, and it was suspected

'

that the transmitter had failed or that the sensor tubing was plugged. The work order was

written to troubleshoot the instrument and determine the cause of the indication being

downscale.

,

b. Observations and Findinos

The work order required that out-of-service stickers be placed on the local and remote

indicators. A note in the work order stated that the DP transmitter was out of service so

that a treubleshooting plan was not applicable. The clearance sheet, with the AWO,

, required that yellow caution tags be placed on the four start-stop control switches for the

circulating water pumps, which stated " travelling screen differential pressure

i

instrumentation work is in progress, screen DP alarms and/or circulating water pump trips

i may occur." At the time of the actual work, no circulating water pumps were in service.

,

During the troubleshooting activities, and subsequent repair of the sensing tube, indicated

DP went high causing the travelling screens to go into fast speed and the screen wash

4

pumps to start. While the technicians had informed operations personnel of the potential

for these components to operate, it was not clearly identified in the AWO and therefore,

was considered an unexpected response.

.

Two adverse conditions reports (ACRs) were initiated as a result of this activity. The first

ACR, M1-96-0606 was written by the technician prior to the start of the work, and stated

l that the yellow tags did not provide for personnel safety and had nothing to do with the

$

.

,

10

AWO stated problem. Since personnel safety was not a concern for this work, the

technician did not think it was appropriate to sign the block on the clearance verification

that stated " clearance adequate for personnel safety." He also stated in the ACR that he

<

had discussed this issue with both his first and second line supervision and nothing was i

changed on the AWO. The technician then discussed his objections with operations

personnel, and they removed the yellow tags from the verification sheet. In the end, the

technician signed the verification sheet "under protest" and completed the work. The

AWO planner later informed the inspector that the yellow tags were added as a reminder to

the control room operators that any DP alarms received, were from the DP troubleshooting

work. The second ACR, M1-96-0617 was written by the tecimician's FLE after the work

was completed, and documented the fact that the AWO did not ioatify the potential to

start the screen wash pumps and cause the screens to go into fast speed.

The inspector reviewed the AWO, ACRs, work control procedures, and conducted

interviews in response to this event. Work Control Process, WC-1, Revision 2, states that

'

if troubleshooting could impact plant operation or in service equipment, then a l

troubleshooting plan is required. A troubleshooting plan is not required for equipment

which is out of service and could not affect plant operation. Additionally, troubleshooting

guidelines shall be included in all troubleshooting AWO packages. The I&C planner who

initiated the AWO thought that placing out-of-service stickers on the indications and yellow

caution tags on the circulating water pump control switches had taken the system out of

service and therefore, a troubleshoot plan was not required. He apparently overlooked the

control function for the screens and spray wash pumps. This error was not detected

during the FLS's review of the AWO prior to assigning the work.

c. Conclusions

The failure to provide a troubleshooting plan and troubleshooting guidelines with the AWO

package is an apparent violation (eel 245/96-08-01) of Technical Specification 6.8.1,

which states that written procedures shall be implemented covering the activities

recommended in Appendix A to Regulatory Guide (RG) 1.33, " Quality Assurance Program

Requirements (Operations)," dated February 1978. Section 9.c. of Appendix A to RG 1.33

states procedures for the repair or replacement of equipment should be prepared prior to

beginning work. The control functions of the DP transmitter were not properly assessed

by the I&C planner or the FLS in charge of the work. Additionally, the FLS missed a

second opportunity to identify the error when questions about the clearance adequacy

were raised by the technician. l&C supervision's response to the concerns of the

technician was less than adequate and resulted in the technician being directed to

complete the work, in spite of his obvious apprehension.

A similar event occurred on May 10,1995, while working on DPT-4-121 for the 'B'

travelling screen DP transmitter. The travelling screens transferred to fast speed

unexpectedly when working on tubing to DPT-4-121. The licensee initiated an ACR to

. document the event. At that time, the causes of the event were an inadequate job review

and that the work order package did not identify that travelling screens would transfer to

high speed. The corrective actions for the ACR were to route the ACR to appropriate

personnel as lessons learned, and have work planning identify the expected plant response

based on input from the performance department. The I&C planner for the 'E' DP

.. _ . _ ._ _ . _ _ . __. _ ___ . . _ _ _ , .

.

.

11

transmitter work was not aware of the event that occurred in May of 1995, nor had he

seen the ACR. The issue of inadequate corrective actions for the May 1995 event, should

be addressed as part of the licensee's response to the procedural violation stated above.

U1 M8 Miscellaneous Maintenance issues

M 8.1 (Ocen) Unresolved item 50-245/96-06-03: ISI Proaram Review

a. Insoection Scooe (92902)

The inspector reviewed the licensee's commitments to Generic Letter 88-01 for the

augmented ultrasonic (UT) inspection program.

b. Observations and Findinas

During refueling outage 15, the licensee identified six reactor coolant components (RCAJ-

2, RCBJ-1 A, RRJJ-4, RREJ-4, RRCJ-4 and CUBJ-18) with flaws that were placed inservice

between 1984 and 1995, without flaw analysis as required by ASME Section XI,1986

Edition, Paragraph IWB-3640. The ASME Section XI analysis was not performed on the

components because the NDE Level 111 inappropriately evaluated the indications to be

geometry. The weld operability calculations were not reviewed on site and the issue was

left unresolved pending further NRC review.

By memorandum dated August 2,1996, the Director of Reactor Projects, Region I, initiated

a Task Interface Agreement with the engineering division at headquarters requesting a

review of the weld operability calculations performed by Northeast Utilities engineering ,

staff. )

c. Conclusion

By NRC memorandum dated October 9,1996, the Materials and Chemical Engineering

Branch's (NRR) evaluation of the subject weld operability calculations was submitted to the

regional office closing the TIA between NRR and Region 1. The NRR staff reviewed the

licensee's calculations and found them acceptable. The inspectors concur with the results

of the NRR review and no further action by the staff is necessary.

The observations from inspection Report 50-245/96-03 regarding: 1) The IGSCC program

does not specify a methodology to evaluate unresolved UT indications; 2) The examination

procedure and calibration blocks for the UT examination are not specified; and 3) the

IGSCC program does not provide guidelines for tracking and trending UT indications remain

open pending licenses corrective actions and NRC review.

l

.

.

12

U1.Ill Enaineerina

U1 E1 Conduct of Engineering

E1.1 Control Room Ventilation Testina

a. Insoection Scoce (37551)

The inspector reviewed pmceduras and documents related to collecting temperature and

damper closure data required for input into the engineering evaluation of control room

habitability. The document review included a special procedure SPROC 96-1-35, " Control

Room Temperature Monitoring with Normal Ventilation Secured," and a safety evaluation

prepared for the SPROC. The inspector also observed portions of the test being performed

in the Unit 1 control room.

b. Observations and Findinas

l

During a surveillance test to verify a control room isolation on a high radiation signal, the l

licensee noted that the control room ventilation system transfer fans (HVT-10A and HVT- ,

11) did not trip. An adverse condition report (ACR) 01215 and licensee event report (LER) l

96-033 were initiated to address the problem. The licensee's radiological assessment '

calculation, RAB-MP1CR-1996, determined that control room personnel radiological

exposure limits would not be exceeded following a design basis accident if self contained I

breathing apparatus are donned and control room in leakage is less than 1000 cfm. The

calculation also assumes that 1-HVD-8C, control room outside air supply damper closed in

10 seconds. Control room in leakage has never been verified, and in light of the fact that

the transfer fan woulo not trip, the 1000 cfm in leakage assumption was tmestioned. A i

possible system configuration to reduce the control room leakage to less t a 1000 cfm, I

was to secure all control room ventilation fans. With this configuration, coohng for the

control room is lost, resulting in a control room temperature increase. SPROC 96-1-35 was

developed to monitor and record control room temperature data with normal control room

ventilt. tion secured.

The inspeuor observed the pre-test briefing for the evaluation on September 12,1996.

The briefing was conducted in the main control room by the system manager. Industrial

hygiene personnel were present and provided input to the SPROC concerning the proper

use of safety equipment and air sampling devices, in order to monitor the control room air

quality during the test. They also provided termination criteria for minimum oxygen levels

and maximum levels for carbon dioxide and carbon monoxide. Emergency medical

personnel were also available to respond in the event of any medical problems, as a result

of the elevated room temperature.

The test was coordinated by system engineers who were present throughout the test. The

control room ventilation system was isolated at 8:15 am on September 14,1996. The

outside air supply damper,1-HVD-8C, closed within the required 10 seconds. The test

was terminated when the control room bulk average air temperature reached 95oF

(maximum temperature allowed by the SPROC), on September 15, at 12:15 am. The

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

• !

t

13  ;

inspector noted that no abnormal plant indications were experienced in the control room

and all plant systems functioned normally for the duration of the test. ,

c. Conclusions

l

The inspector concluded that the implementation of SPROC 96-1-35 was well coordinated

with adequate controls to ensure personnel safety in the main control room during the test. ,

The licensee's safety evaluation, performed for the special procedure, concluded that the

test did not create an unreviewed safety question. The inspector reviewed the safety '

evaluation and determined it was adequate. An additional SPROC will be developed and .

'

performed to verify actual control room in leakage at a later date. In the interim, an

Operability Determination was initiated that established compensatory actions to prevent

potential challenges to Unit 1 control room habitability. i

i

U1 E8 Miscellaneous Engineering issues

E.8.1 (Closed) Unresolved item 50-245/95-25-01: a single failure vulnerability with the j

loss of normal power (LNP) logic that would prevent both emergency power sources from l

performing their design safety function. This item was characterized as unresolved t

pending further NRC review and the assessment of the long term corrective actions. The i

issue was subsequently addressed in NRC report 245/95 31, section 1.4 and in a Decembei

7,1995 correspondence which forwarded a violation. The unresolved item is closed and ,

further NRC review of the corrective actions wi!! be documented with violation 50-245/95- l

31-03.  !

E.8.2 (Closed) Unresolved item 50-245/95-25-02: deficiencies with the standby gas l

treatment (SGT) system including a vulnerability while venting containment and to a single '

failure of a fan which could have resulted in a loss of the system function, and thereby

induce a loss of secondary containment integrity following an accident. This item was i

characterized as unresolved pending further NRC review and the assessment of the long l

term corrective actions. The issue was subsequently addressed in NRC report 245/95-31, {

section 1.4 and in a December 7,1995 correspondence which forwarded a violation. The  ;

unresolved item is closed and furthsr NRC review of the corrective actions will be

documented with violation 50-245/95-31-04.

E8.3 (Closed) Unresolved item 50-245/95-44-01: SGTS Ooerability Determination

On July 5,1995, procedure SP 624.1," Secondary Containment Tightness Test," was

performed with the reactor building exhaust fans running, in an alignment which simulated

recirculation flow through an idle standby gas treatment system (SGTS) train. Using the

test results, the licensee conclu'ied in an operability evaluation that the SGTS remained

operable with outside ambient temperatures of 2 20 F. On November 19,1995, as

required by Technical Specification 4.7.C, the licensee performed procedure SP 624.1 with

and without the main exhaust fans operating. Although both test results were acceptable,

the negative pressure attained in the reactor building was enhanced by the operation of the

main exhaust fans. This test was performed with all reactor building supply and exhaust

dampers closed. Investigation revealed that both reactor building exhaust dampers (HV-3

.

,

14

and 4) were not sealing properly. Since the July 1995, a containment draw down test

was performed ivith the main exhaust fans running, it masked the fact that the exhaust

dampers were !eaking.

The licenses reassessed their original SGTS operability determination. Substitution of the

lower draw down capability into the operability determination calculation resulted in the

SGTS being operable down to only 45 F for the "B" train and 30 F for the "A" train. A

review of ambient temperature since July 5,1995, revealed that on several occasions

during power operation the ambient temperature dropped below 45 F, with a low of 26 F.

The incorrect operability 4 termination was the result of inadequate SGTS testing. The

failure to test each SGl ' train independently, with the main exhaust fans operating,

masked the fact that both reactor building exhaust dampers were not sealing properly and

resulted in the licensee incorrectly determining that the system remained operable with

outside ambient temperatures 2 20oF.

Therefore, if actuation of the SGTS had occurred during the time period that the outside

ambient temperature was less than 45 F, coincident with a loss of normal power (reactor

building exhaust fans tripped) and a single fai!ure in one train, the required negative

pressure may not have been maintained throughout the secondary containment thus

resulting in a reduction of the systems's ability to perform its intended safety function. l

The failure to maintain the SGTS operable under all conditions is an apparent violation (eel l

245-96-08-02) of Technical Specification 3.7.B.1, Containment Systems. Unresolved item

254/95-44-01 is administratively closed, in lieu of the apparent violation.

E8.4 (Closed) LER 50-245/96-40.

(Ocen) Acoarent VIO 245/96-08-03 Failure to Perform Control Rod Drive System i

Seismic Modifications

l

a. Insoection Scone

The licensee identified, in LER 245/96-40, that the control rod drive (CRD) system piping

was not seismically qualified as a result of an original design error. The CRD insert and

withdraw lines, which form part of the reactor coolant boundary, extend between the

hydraulic control units located in the reactor building and the control rod drive mechanisms

attached to the reactor vessel. The inspector reviewed the circumstances leading to LER

245/96-40 and the subsequent corrective actions.

b. Observations and Findinas

As part of the Systematic Evaluation Program (SEP) implemented in the early 1980's,

plants licensed prior to the implementation of the general design criterion (GDC) were

required to confirm the seismic adequacy of selected structures, systems, and components

(SSCs). At Millstone Unit 1, the seismic reevaluation focused on the integrity of the  ;

reactor coolant boundary and the capability of essential SSCs necessary to shutdown the  !

reactor and maintain it in a safe condition following a seismic event.

l

The NRC staff and consultant review, documented in a May 28,1981 letter, identified a

number of open items including the CRD hydraulic system, that required additional

.

1

.

.

15

inforrnation to evaluate the seismic design adequacy of the system. In a subsequent setter,

the licensee responded by stating that, "the CRD system tubing and supports will be

evaluated for the Millstone Unit 1 site specific seismic response spectra. Field surveys will

be performed to confirm the as-built configuration of the tubing runs and supports." In an

April 28,1982 letter, the licensee stated in part, that " detailed as-built drawings based on

plant walkdowns were generated for the insert and withdraw lines and supports. These

data were used to demonstrate operability of the system during a design basis

earthquake." The review demonstrated that the system and certain support frames were

operable, but that they exceeded certain ASA B31.1 Code allowables. However,

numerous maintenance items such as loose or missing pipe clamps, identified during the

walkdowns, were corrected prior to demonstrating operability. Subsequently, the NRC

concluded, in a June 30,1982 letter, that in part, " based on the licensee's responses to

the SEP seismic related safety issues, all safety related piping systems are capable of

withstanding the postulated safe shutdown earthquake loads pending the completion of all

necessary modifications required by lE Bulletins 79-02 and 79-14." Continued plant

operation was found to be ecceptable until early 1983 when the modifications were

scheduled for implementation. However, the licensee did not implement the design ,

modifications necessary to bring the CRD system into design compliance within the

specified time period. At the end of this inspection period, the necessary CRD

modifications had not been implemented.

A review of the licensee's evaluation of the CRD system, performed by a third party,

revealed that the scope of the review only encompassed the portions of the system I

external to the drywell. Although,.LER 96-40 discussed the failure to implement the

modifications necessary to bring the CRD system into design compliance relative to seismic i

capacity, it did not address the failure to assess the CRD piping inside the containment. '

Further, in the LER, the licensee failed to discuss why the modifications were not

implemented. The inspector also noted that the immediate notification per 10 CFR 50.72

was not performed until 25 days after the issue was identified. For historical issues

requiring a detailed evaluation there is no specific time limit for determining the reportability

per 10 CFR 50.72; however, this issue provides further evidence that the licensee is not

appropriately focused to ensure that prompt reportability determinations are made.

The inspector performed a review of other seismic issues identified during the SEP re-

evaluation and found that the licensee could not demonstrate that all of these issues were

resolved. ACR M1-96-0696 was written to address these potentially outstanding seismic

issues related to the structuralintegrity and functionality of numerous components. For

example, the suspect components include: the low pressure heat exchangers, electrical

distribution transformers, control room electrical panels, and electrical cable raceways,

in May 28,1993 the licensee issued LER 93-004, which identified seismic deficiencies

with the main steam line flow venturis. As part of the long term corrective actions for

these seismic deficiencies, additional walkdowns were performed in the drywell during the

following refueling outage. The walkdowns identified numerous maintenance and design

deficiencies associated with several systems. These deficiencies included 14 CRD

maintenance related issues such, as loose or missing pipe clamps and an inadequate pipe

support. However, the seismic evaluations performed as part of the long term corrective

actions did not identify the previous corrective actions which were not implemented

I

l

l

I

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

~

i .

,

16

I following the SEP re-evaluation. In addition, the licensee failed to supplement LER 93-004,  ;

previously committed, with the additional deficiencies identified in the subsequent '

i walkdowns performed during the cycle 14 refueling outage. The licensee wrote an ACR to

address the failure to report subsequent seismic deficiencies and plans to supplement LER

93-004 within the next few weeks. They also plan to perform an immediate review of  ;

J

other LERs for similar circumstances.

!

A review of the operability assessment performed following the identification of the CRD

, maintenance and design related seismic discrepancies discussed above, found that a j

!

rigorous evaluation had not been performed consistent with the guidance provided in  ;

Generic Letter 91-18. Although the pipe analysis was consistent with the GL 91-18 i

methods and criteria, the licensee did not properly evaluate the pipe support loading

associated with the as found conditions. The licensee did not consider the integrated l

,

effect of the loose clamps in conjunction with the undersized support. The licensee also 1

failed to consider the additional seismic loading on active supporM as a result of other  !

supports being inactive due to loose or missing hardware. The licensee plans to re-

!

evaluate the support loads and revise the applicable calculations consistent with GL 91-18 i

l methods. The licensee also initiated an ACR to address the inadequate operability {

determination and to assess the need for additional procedural guidance related seismic

operability evaluations.

!

c. Conclusions

The licensee failed to implement the design modifications necessary to bring the CRD

system into design compliance within the NRC specified time period. This is an apparant,

violation (eel 245/96-08-03) of Appendix B Criterion 16, " Corrective Actions." The

licensee also failed to expand the scope of their review based on the significant number of

CRD design and maintenance deficiencies identified external to containment, in the earl) l

'

1980's. Further it appears the licensee failed to implement a commitment when the CRD l

evaluation was limited to only the portions of the system external to tha containment  !

based on their response to the NRCs request for additional information, j

j The licensees LER did not discuss why the modifications were not implemented and  ;

'

consequently did not provide corrective actions for the cause. Further, the immediate l

notification per 10 CFR 50.72 was not performed until 25 days after the issue was i

identified and the reason for this delay was also not addressed. Trie delay in the initial I

reporting provides further evidence that the licensee is not appropriately focused to ensure

prompt reportability determinations are made. I

!

The licensee failed to demonstrate the operability of the CRD system, cons stent with GL

91-18 methods, fol!owing the identification of seismic deficiencies found during the cycle i

14 refueling outage. This issue is unresolved (URI 245/96-08-04) pending the revision of

the calculations related to CRD operability prior to the cycle 14 refueling outage. in

addition, this issue is unresolved pending the licensee's assessment of the need for

additional procedural guidance related seismic operability evaluations.

.

.

17

The licensee could not demonstrate that all of the seismic issues identified during the SEP

re-evaluation were resolved. This issue is unresolved (URI 245/96-08-05) pending NRC

verification of the closure for allitems identified.

,

!

,

18

Report Details

Summarv of Unit 2 Status

Unit 2 remained in cold shutdown throughout the inspection period. The unit has been

shut down since February 20,1936, due to uncertainty with the licensee's compliance

with the plant design and licensing bases. A comprehensive recertification process is

being conducted to support plant restart.

U2.1 Operations

U2 01 Conduct of Operations

01.1 General Comments (71707)

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

plant operations. Overall operator performance in the monitoring and operation of systems

necessary for safe shutdown was found to be good. However, as discussed below, the

licensee did not comply with the design basis requirement to lock open the refueling pool

drain valves during operation. In addition, several concerns associated with the adequacy

and conduct of the monthly containment integrity surveillance were noted and are

discussed in Section M8.4 of this report.

U2 O2 Operational Status of Facilities and Equipment ,

l

02.1 Lockina Onen the Refuelina Pool Drain Line

a. Insoection Scooe

The NRC evaluated whether the licensee locked open the refueling pool drain line as stated

in the final safety analysis report (FSAR).

b. Observations and Findinas

FSAR, Section 6.4.3.1, which describes the operation of the containment spray system

during emergency conditions, states that the refueling pool drain line isolation valves (2-

RW-24A&B) are locked open during the operating cycle to prevent the refueling pool from

capturing water. Following a loss of coolant accident, the ability to cool the core would

be lost if a sufficient amount of inventory accumulated in the refueling pool rather than

draining to the containment sump where the emergency core cooling system pumps take

suction. Operating procedure OP 2305, " Spent Fuel Pool Cooling and Purification

System," provides instructions for draining the refueling pool following refueling activities

and therefore, is the procedure that positions the drain valves prior to operation. The NRC

found that the refueling pool draining instructions, as well as the refueling water

purification system valve lineup (OPS Form 2305-2) leaves the valves in the open but not

locked open position as required by the FSAR.

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

.

.

19

This concern was discussed with the licensee who changed Section 5.27 of procedure OP

2305, which provides instructions to drain the refueling pool to the refueling water storage

tank (RWST), as well as the valve lineup, to lock open the valves. However, the NRC

found that licensee failed to change Section 5.28 of the procedure, which drains the

refueling pool to the liquid radiological waste system. Since Section 5.28 makes no

reference to valves 2-RW-24A&B the valves could have been left in the closed position.

The failure to change Section 5.28 is significant because this section, rather than Section

5.27, is normally the last section to be performed (thereby dictating the final position of

the drain valves) because water remaining in the refueling pool after the RWST is full must

be drained to the liquid radiological waste.

c. Conclusion

The NRC found that procedure OP 2305 was inadequate in that it failed to lock open the

refueling pool drain valves as required by FSAR Section 6.4.3.1.10 CFR 50, Appendix B,

Criterion XVI, requires that measures be established to assure that conditions adverse to ,

quality, such as deficiencies and deviations, are promptly identified and corrected. After  ;

the deviation was identified, the resulting corrective action was inadequate in that the  !

change to procedure OP 2305 failed to ensum the refueling pool drain valves are locked

open during the operating cycle. This is an apparent violation. (eel 336/96-08-06)

U2.ll Maintenance

U2 M2 Maintenance and Material Condition of Fetilities and Equipment

M 2.1 Soent Fuel Pool and Refuelino Pool Skimmer Systems

'

a. Insoection Scooe

The inspector reviewed the current status of the spent fuel pool (SFP) and refueling pool

skimmer systems, including a review of the maintenance history for each of their

associated pumps, P21 and P96.

b. Qbservations and Findinos

The SFP and refueling pool skimmer systems, whose function is to remove accumulated

dust from the surfaces of these pools, is composed of skimmer assemblies, a pump, a tank

and two filters. The SFP skimmer system is designed to always be in service while the

refueling pool skimmer system is only used during outages when the refueling pool is

flooded. The high and low level limit switches on the skimmer tanks serve to cycle the

skimmer pumps on and off. During the past several years, the licensee has encountered

problems with the operation of these systems, apparently caused by sticking tank level

switches. If the high level switch sticks, the pump will not cycle on and the system won't

operate. If low level switch sticks, the pump will not shut off and the tank will empty,

thus causing the pump to cavitate. Because the systems have not been operating

properly, they have been tagged out of service. In the mean time, the I;censee relies on

the SFP and refueling pool purification systems to maintain the pool purity and clarity.

1

.

.

20

Due to the unavailability of replacement parts, the licensee has attempted to repair the i

systems by cleaning the limit switches which has proven unsuccessful. The licensee has  !

also increased the setpoint of the low level switch on the refueling pool skimmer system,

but the system won't be tested until the next time the refueling pool is filled. If adjusting

the low level setpoint does not alleviate the problem, the licensee is considering a

modification that would replace the switches with an improved design. i

The inspector reviewed the maintenance history for pumps P21 and P96 for the last 10

years. The maintenance history did not reflect a high failure rate for either pump that could

have been caused by cavitation. However in late 1994, the SFP skimmer pump was found

to have a leaking mechanical seal, a loose and damaged suction plate, and a scored shaft

and impeller, which could have been caused by cavitation. No other major problems

occurred since this repair, but the systems were most likely tagged out of service a

majority of the time since then,

c. Conclusion .

l

The SFP and RP skimmer systems are non-safety-related systems that are not required to j

be in operation. The purification systems currently maintains the purity and clarity of the  !

refueling pool when flooded and the SFP. The licensee is considering a modification to the  ;

level switches that will enable these systems to operate as they were designed. i

U2 M8 Miscellaneous Maintenance issues )

M 8.1 (Closed) LER 50-336/96-14: Missed Technical Soecification Surveillances

This event involved the failure to conduct weekly checks of shutdown safety parameters

required by four separate technical specifications (TSs). The cause of the event was the

erroneous deletion of the required procedure from the work planning schedule and a

programmatic weakness that allowed such mistakes to go unnoticed. Missed surveillances

have been a chronic problem at Unit 2. NRC Inspection Report 50-336/96-04 documents

continuing examples (including this event) as evidence of inadequate and untimely

corrective action for previous violations of TS surveillance requirements. This event 1

recurred prior to the completion of alllicensee corrections at Unit 2, and therefore is

considered an additional example of the prior violations. On August 2,1993, the licensee

docketed a response to NRC's concern with the acequacy of surveillance tracking

corrective actions. These additional actions will be reviewed as part of the closure of open

item VIO 336/95-38-01. Further, licensee demonstration of improved effectiveness in

corrective action programs has been elevated to an NRC restart verification issue for all the

Millstone units.

M8.2 (Closed) LER 50-336/96-18: Dearaded Seals in Environmental Enclosures

This event, reported on April 19,1996, involved the degradation of door seats in

environmental enclosures for two auxiliary building motor control centers. NRC Inspection

Report 50-336/96-201 reviewed this problem in detail. The inspector further noted that

the LER was incomplete in that the licensee had not bounded the scope of the causes, nor

finalized the required corrective action. As of October 9,1996, the licensee had not i

. - . . . -

~

l

.

21

submitted the supplemental LER committed to in the original LER. This weakness in quality

of original input and follow-through on LER supplements was previously detailed in NRC

Inspection Report 50-336/96-06. NRC followup on the integrity of environmental

enclosures will be tracked by open item eel 336/96-201-20. In addition, other problems

with the Unit 2 program for environmental qualification of electrical equipment will be

reviewed under open item eel 336/96-06-12.

M8.3 (Onen) LER 50-336/96-16: Inocerable Circulatina Water Pumo Trio Function

a. Insoection Scoce

'

The scope of this inspection was a review of Licensee Event Report 50-336/96-16.

b. Observations and Findinas

This LER described that the common power supply cable to the circulating water (CW)

pump trip level switches was found to be miswired. The licensee could not identify a

maintenance activity that affected this function, and no tests are routinely conducted to

verify the operability of the trip logic. As a result, the CW pump trip logic was inoperable

for an undetermined duration. The Unit 2 Updated Final Safety Analysis Report (FSAR),

Section 9.7.1.2.1, states that the CW pump trip logic protects safety-related systems in

the turbine building from flooding by preventing overflow from the condenser pit. With the

trip logic not functional, the motor-driven auxiliary feedwater pumps could have been

threatened if a CW pipe or expansion joint ruptured.

The licensee repaired the wiring errors, and verified that current work control practices

require appropriate post-maintenance testing prior to restoration of this circuit to service.

The testing, which is yet to be completed, is being tracked by the Mode 3 hold punchlist.

In addition, the licensee has committed to implement periodic functional testing of this

logic on a 36-month frequency. As of October 23,1996, this activity had not been

completed.

The NRC was concerned that this logic, designed to protect safety-related equipment, was

inoperable for a prolonged period without licensee awareness. The licensee stated that

turbine building flooding analyses have shown that overflowing the condenser pit does not

threaten safety-related equipment and that they plan to change the FSAR to remove the

reference to the CW pump trip protection in the plant's design basis,

c. Conclusion

Due to poor maintenance and the absence of routine testing, the circulating water pump

trip, which is designed to protect the auxiliary feedwater pumps from flooding, was found

inoperable. This item remains open to allow NRC review of the licensee's planned design

basis change to not credit this trip.

. . _ __ , . .- . --

.

.

22

'

,

M8.4 (Closed) LER 50-336/96-23 & LER 50-336/96-26: Failure to Perform Surveillance on

Certain Containment isolation Valves

!

a. Insoection Scooe

'

The scope of this inspection included a review of Licensee Event Report (LER) 50-336/96-

23 and LER 50-336/96-26.

b. Observations and Findinas

,

'

'

LER 50-336/96-23 discussed that the licensee discovered that several valves that were

classified as containment isolation boundaries did not receive a monthly verification to

.

ensure they were in the closed position as required by Technical Specification (TS)

'

4.6.1.1.a. The licensee determined that the cause of this event was an inadequate

procedure in that it did not include several valves that must be verified closed to maintain

i

containment integrity. The licensee reported this event in accordance with 10 CFR

50.73(a)(2)(1)(B), as a condition prohibited by the plant's technical specif; cations.

i

The procedural discrepancies identified in this event were discovered by a licensee task

'

team assigned to review TS compliance. The Final Safety Analysis Report (FSAR), Table

5.2-11, lists the containment structure isolation valve information. It was noted that some

valves listed in the FSAR table were not included in the monthly surveillance Procedure

OPS Form 2605A-1, " Verifying Containment Integrity." Valves that were not included in

the surveillance were 3/4 inch and smaller vent and drain valves, a 2-inch isolation valve

(2-CH-517) located on the pressurizer auxiliary spray line, and two 8 inch main steam line

atmospheric dump valves (2-MS-190 A&B).

1

The safety significance of this event was low because all but two 3/8 inch drain valves

were verified closed by other surveillance procedures, although not on a monthly cycle. As

an initial corrective action the licensee checked and found all the valves in question to be

closed. The LER stated that they plan to change procedure OPS Form 2605A-1 to add the

excluded valves. The NRC found that the proposed procedure change had not yet been

, completed but was completed shortly after the NRC questioned the status of the procedure

'

change. In addition, the LER stated other TS surveillances containing requirements to

j verify valve position are being reviewed to identify any valves that are potentially not

included within appropriate surveillance procedures. The corrective action tracking system

item that was generated to track this LER commitment indicated that the review of other

TS surveillances had been completed. However, the NRC found that the review had not

3

been performed.

4

i LER 50-336/96-26 also discussed that the surveillance requirements of TS 4.6.1.1.a had

not been met. TS 4.6.1.1.a requires a monthly verification to ensure that all penetrations,

not capable of being closed during accident conditions, "are closed by valves, blind flanges

or deactivated automatic valves secured in their positions . . . " Similarly, the LER

discussed that the requirements of TS 4.5.2.a.7, " Emergency Core Cooling Systems,"

i (ECCS) and TS 4.7.3.1.a.5, " Reactor Building Closed Cooling Water," (RBCCW) which

require a monthly verification of the correct position of valves that are not " locked, sealed

i or otherwise secured in position," had also not been met. For each of these TS

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

'

.

! .

I 23

surveillance, the licensee found that while the plant was at power, operators had been

marking "N/A" the verification of manual valves located inside containment. The licensee

i reported this event in accordance with 10 CFR 50.73(a)(2)(1)(B) as a condition prohibited

I by TS.

l

l The licensee determined that the cause of this event was a historical interpretation of the

!

TS that resulted in operating practices that were not consistent with TS requirements.

This operating practice for the containment integrity surveillance was consistent with the

Final Safety Analysis Report, Section 5.2.8.4.2, which indicated that manual containment

isolation valves located inside containment need only be inspected during plant shutdowns.

Since the radiation level near several of the valves inside containment is approximately 1

rem / hour gamma and 200 mrem / hour neutron, monthly entry into these areas would be

undesirable. However, neither TS 4.6.1.1.a nor the associated TS basis allows manual

valves located inside containment to be exempted from the monthly containment integrity

verification surveillance. Regarding TS 4.5.2.a.7 and TS 4.7.1.3.a.5, the LER stated that

operators marked "NA" the ECCS and RBCCW valves inside containment based on the fact

that the containment personnel hatch was " locked" and " sealed".

As corrective actions, the manual valves that were not included in these surveillances were

verified to be in their correct position and operators were briefed that the practice of

marking "N/A" valves was contrary to TS requirements. Prior to this event, on January

22,1996, a license amendment request for TS 4.6.1.1.a was submitted which specifies

that the position of the manual containment isolation valves inside containment be verified

prior to changing from mode 5 to 4, but does not require visual verification while at power.

c. Conclusion

TS 4.6.1.1.a. states that valves that are classified as containment isolation boundaries

receive a monthly verification to ensure they are in the closed position. The failure of  !

procedure OPS Form 2605A-1 to include all of the required containment isolation valves

and the practice of marking "NA" containment isolation valves inside containment while at

power even though TS did not reflect this exemption are considered to be violations. (VIO

336/96-08-07)

10 CFR 50, Appendix B, Criterion XVI, requires that measures be established to assure

that conditions adverse to quality, such as deficiencies, deviations, and nonconformances

are promptly identified and corrected. The failure of the licensee to implement the

corrective action committed to LER 50-336/96-23 of reviewing other TS surveillances

containing requirements to verify valve position to identify any valves that are potentially

not included within appropriate surveillance procedures is considered an apparent violation.

(eel 336/96-08-08)

,

I

l

.

l

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

. l

.

24

4

'

M8.5 [ Closed) LER 50-336/96-24: Comoonents Omitted from Resoonse Time Testina

1

a. Insoection Scooe

The scope of this inspection was a review of Licensee Event Report 50-336/96-24.

b. Observations and Findinas

2

The licensee discovered that certain sensor channels of the reactor protection system

(RPS) and the engineered safeguards actuation system (ESAS) were inoperable because the i

response time testing required by TSs 4.3.1.1.3 and 4.3.2.1.3 was inadequate. Response

I

time testing is performed to determine the amount of time it takes for a signal to travel

from the sensor (such as a pressure or temperature instrument) to the end device and

3 initiate action (for the RPS this is considered tripping of the reactor trip breakers and for

i ESAS this is considered the initiation of safety equipment for accident mitigation.) I

l Information from various plant sensors is processed through the Split Package Electronic  !

Control (SPEC) 200 electronics circuitry and is sent to the RPS and ESAS. The total

response time is determined by measuring the times of individual sections of the circuit

between the sensor and final device and adding up the individual times. However the

licensee had incorrectly assumed that the signal response time of the SPEC 200 electronics

was zero seconds and therefore, this time was not measured or factored into the total

response time. By adding the estimated SPEC 200 response time of 225 milliseconds to

the previous surveillance results of 20 to 50 milliseconds (which did not include SPEC

200), the licensee determined that they would have still satisfied the most limiting overall

time response criteria of 400 milliseconds. This condition was reported in accordance with

10 CFR 50.73(a)(2)(i)(B), as an operation or condition prohibited by TS.

The licensee determined that the cause of this event was that the RPS and ESAS i

surveillance procedures were inadequate for failing to take into account the SPEC 200 l

response time. No immediate operator corrective actions were necessary because the

affected SPEC 200 sensor channels were not required to be operable in mode 5. The

licensee committed in the LER to revise the appropriate surveillance procedures prior to

plant startup to incorporate response time testing and acceptance criteria for the SPEC 200

electronics.

The NRO reviewed LER 50-336/96-24 and had two concerns. The first concern relates to

the licensee's determination that the root cause of the event was an inadequate procedure.

Although the NRC agrees that the procedure was inadequate, there are also concerns

regarding the modification process that were not addressed in the LER. SPEC 200 is a

digital system that was installed on severalinstruments at a time over a several year period

beginning in the early 1980's to replace the original analog system. Due to an inadequate

review of the modifications, the licensee failed to take into account the effect of the SPEC

200 electronics on RPS and ESAS time response testing. This is of particular concern  ;

because each group of instruments that was modified had its own plant design change

record and therefore, the licensee had multiple opportunities over a several year period to

identify the response time testing concern.

l

.

-

\

25 l

Secondly, the NRC noted that in addition to the SPEC 200 electronics, the cabling (and

possibly other components) between the sensor and final device is also not included in the

total response time. TS definition 1.26 states that the reactor trip system response time l

shall be the time interval from when the monitored parameter exceeds its trip setpoint at l

the channel sensor until electrical power is interrupted to the control rod drive mechanism.

TS definition 1.27 states that the engineered safety feature (ESF) response time shall be

that time interval from when the monitored parameter exceeds its ESF actuation setpoint

at the channel sensor until the ESF equipment is capable of performing its safety function

(i.e., the valves travel to their required positions, pump discharge pressures reach their

required values, etc.). The implication of the definitions is that response time of the entire l

. circuit, including the cabling (and other possible components), should be measured.  ;

c. Conclusion

The concerns regarding response time testing are considered unresolved to allow further

licensee and NRC review of: (1) how the effect of SPEC 200 electronics on RPS and ESAS

response time was overlooked during the licensee's review of several modifications that

installed SPEC 200 drawers and cabinets, and; (2) whether the response time of the entire

circuit, including cabling (and possibly other components) between the sensor and final

device, is required to be measured and included in the total response time. (URI 336/96-

08-09)

M8.6 (Open) LER 50-336/96-25: Enclosure Buildino Filtration System /Auxiliarv Exhaust

Actuation System Interlock Not Tested

a. Insoection Scone

The scope of this inspection included a review of Licensee Event Report 50-336/96-25.

b. Observations and Findinas

While performing a review in response to Generic Letter 96-01, " Testing of Safety-Related

Logic Circuits," the licensee identified that an interlock between the enclosure building

filtration actuation system (EBFS) and the auxiliary exhaust actuation system (AEAS) was

not periodically tested. The EBFS/AEAS interlock is designed to ensure that filtration and

exhaust of the enclosure building takes precedence over filtration and exhaust of the spent

fuel pool area by closing four ventilation dampers that take suction from the spent fuel pool

area. If the EBFS/AEAS interlock does not function as designed, EBFS would be rendered

inoperable because the system may not be capable of fulfilling its design safety function

during accident conditions, which includes maintaining the enclosure building under a

measurable negative pressure of approximately 0.25 inches water gage. Technical

Specifications 4.3.2.1.1 and 4.9.13 require that surveillance testing be performed on EBFS

and AEAS actuation instrumentation. The licensee reported the failure to include the

testing of the EBFS/AEAS interlock in accordance with 10 CFR 50.73(a)(2)(i)(B) as a

condition prohibited by TS.

The licensee determined that the cause of this event was a programmatic deficiency that

failed to identify the surveillance testing necessary to fully demonstrate EBFS operability.

, .

.

,

26

As an initial corrective action, surveillance procedure SP 2614D, "AEAS Operability

Verification," was changed to periodically test the EBFS/AEAS interlock. Utilizing the

'

revised procedure, the interlocks on both trains were tested and were found to be

operable. As a longer term corrective action, the LER stated that there was an ongoing

1

assessment of TS surveillance requirements to address the concerns identified in Generic

Letter 96-01 and that this review was scheduled to be completed by September 30,1996.

i However, the NRC found that the September 30,1996, commitment was not met. The

licensee stated that the specified due date was based on their desire to complete the

Generic Letter 96 01 review prior to plant startup, which they believed at the time could

occur in October or November 1996. Although the licensee was aware that September

4

30,1996, would not be met, the NRC was not informed of this and at the end of the

inspection period, the commitment date had not been changed.

'

c. Conclusion

The licensee's efforts in identifying that the EBFS/AEAS interlock did not receive a periodic

surveillance were good, in addition, the licensee's follow-up of revising and performing the

surveillance to test interlock was timely in that the test was completed 12 days after the

condition was identified. Despite the fact that the initial corrective actions for this

licensee-identified violation were good, overall corrective action was considered

unacceptable; due to inadequate follow-through, the licensee's commitment to complete

their review of Generic Letter 96-01 was not completed.

M8.7 (Closed) LER 50-336/96-27: Turbine Buildino Crane not Evaluated for imoact of

Heavv Loads on Safetv-Related Switchaear

a. Insoection Scone

The scope of this inspection included a review of Licensse Event Report 50-336/96-27.

b. Observations and Findinas

This LER discussed that previous reviews of the controls for handling heavy loads, such as

the response to Generic Letter 81-07, " Control of Heavy Loads," had not considered the

loads lifted using the turbine building cranes. A historical review of past lifts identified that

in 1983 the main turbine low pressure exhaust hoods were lifted above the facility 1,480

Volt vital switchgear room while the facility 2,480 Volt vital switchgear was out of

service. The LER also stated that Unit 1 utilizes the same laydown area for their main

turbine exhaust hood and must also traverse the Unit 2, facility 1,480 volt vital

switchgear room. The licensee reported this condition in accordance with 10 CFR

50.72(B)(2)(iii)(B) as an event that alone could have prevented the fulfillment of a safety

function needed to remove residual heat.

As corrective actions, the LER states that the turbine building crane procedure will be

revised so that a heavy load lift will not compromise the operation of safe shutdown

equipment. Although both Units 1 & 2 utilize the same two turbine building cranes, each

unit has their own operating procedure for the cranes. Discussions with the licensee

indicated that they do not plan to prohibit the lift of heavy loads over the vital switchgear

. _ . _

l

l

l

1

l

.

27

to room. Instead, they plan to determine an acceptable height of the lifts over switchgear i

room based on the weight and footprint of the heavy loads. I

The NRC noted that the corrective actions specified in the LER did not mention Unit 1

corrective actions. In addition, the LER states that they plan to revise the turbine building i

crane "procc dare," rather than " procedures," even though the Unit 1 and Unit 2 operating

procedure needed to be changed. Therefore, the NRC evaluated whether the licensee's

corrective action tracking system reflected the need to change Unit 1 operating

procedures. The inspector found that although the adverse condition report that initially  ;

documented the concern stated that Unit 1 heavy loads over the Unit 2 switchgear room l

were also a concern, there was no item in the corrective action tracking system for Unit 1 )

to address this concern. Unit 1 personnel indicated they were not aware of the need to '

change their operating procedures. Discussions with Unit 2 personnel indicated that they

planned to notify Unit 1 of the need to change their operating procedures after Unit 2

personnel had evaluated and completed the necessary changes for the Unit 2 operating

procedure. Following discussions with the NRC, the licensee entered an item into their j

corrective action tracking system for Unit 1 to change their operating procedures. l

c. Conclusion

10 CFR 50, Appendix B, Criterion XVI, requires that measures be established to assure

that conditions adverse to quality, such as deficiencies, deviations, and nonconformances

are promptly identified and corrected. Although the licensee's efforts in identifying this

heavy load vulnerability were good, the failure to implement appropriate tracking measures

to ensure that necessary Unit 1 operating procedures were changed is considered an

apparent violation. This is a concern because lifting a Unit 1 heavy load over the Unit 2

switchgear room is equally as safety significant as lifting a Unit 2 heavy load. The NRC did

not consider the licensee's stated plans to notify Unit 1 personnel after the Unit 2 '

procedure was changed were sufficient considering the fact that the LER did not indicate

that Unit 1 corrective actions were necessary. (eel 336/96-08 10)

M8.8 (Closed) LER 50-336/96-30: Failure to Perform Reauired Testina of the Hiah

Pressure Safetv iniection Discharae Check Valves

The issues discussed in Licensee Event Report 50-336/96-30 were discussed in detail in

NRC Inspection Report 50-336/96-06. This LER is considered administratively closed.

U2.lli Enaineerina

U2 El Conduct of Engineering

E1.1 Disposition of License Deviations to Sucoort Entry into Mode 6 and Core Offload

a. Insoection Scoce

The NRC evaluated the licensee's plans for dispositioning identified licensing basis and

design basis deviations on those systems necessary to support entry into Mode 6 and core

offload.

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

.

.

28

b. Observations and Findinas

With Unit 2 currently in Mode 5, the licensee is planning to enter Mode 6 and perform a

full core offload to support a repair of a low pressure safety injection system injection  !

valve, 2-SI-645. As a result of the ongoing design basis review efforts, the licensee

generated a list of the design basis and licensing basis discrepancies on those systems j

necessary to support entry into Mode 6 and core offload. For example, there were no

records that substantiated the seismic qualification of the makeup line to the SFP or l

sections of the SFP cooling system. '

The NRC noted that prior to Mode 6 and core offload, the licensee planned to perform

operability determinations to address the identified discrepancies. The licensee did not

plan to disposition the discrepancies by either modifying the plant to reflect the

design / licensing basis or by changing the design / licensing basis in accordance with 10 CFR

50.59 until later in the outage. The NRC did not consider that placing the plant in a

condition in which a known discrepancy exists that is contrary to their license to be

acceptable. Based on the NRC's concern, at the end of the inspection period, the licensee

was in the process of completing their review of the systems needed for Mode 6, and is

planning to prepare safety evaluations to disposition the identified license discrepancies.  !

Following entry into Mode 6, the licensee plans to complete their review and disposition of

discrepancies for those systems necessary to support core offload, such as the spent fuel

pool cooling system.

I

c. Conclusion

Following a review of the licensee's plans for entering Mode 6 at Unit 2, the NRC had

concerns regarding the licensee's intent to perform a core offload using systems which,

although operable, had known discrepancies that were contrary to the current operating

license. Although no violations of NRC requirements were identified, this is considered to

be a significant weakness in light of recent attention given to compliance with the current

design and licensing basis.

E1.2 Unit 2 Valve List l

a. Insoection Scoce

The inspector reviewed the status of the Unit 2 valve list, which is intended to be a single

document that lists Unit 2 valves along with specific identifying information and

specifications. The valve list would be used as a configuration management tool to

number new valves added to systems during plant modifications, which in turn would be

used to update piping and instrument diagrams (P&lDs) and operations valve lineup

procedures.

b. Observations and Findinas

The Unit 2 valve list is not yet complete but is currently being assembled. The back-end

portion of the valve list is complete, which means that new valves added to plant systems

would be properly numbered, which would result in updates to the P&lDs and operations

-

l

'

.

29

valve lineup procedures, if necessary. The front-end portion of the valve list is currently

working but is not yet complete. Completing this portion of the valve list requires

consolidating several lists into one. The lists that are currently in existence are the original

valve list provided bv Bechtel at the time of construction, a hand-written list of valves that

were added since plant construction, a computer aided drawing list of valves, and a valve

list generated as a result of P&lD walkdowns done over the last several years.

c. Conclusion

The valve list in its current format is acceptable and should ensure that the configuration

management of valves added to Unit 2 in the future could be accomplished properly, as

well as updates to P&lDs and valve lineup procedures. At this time, locating specific

information on valves can be accomplished.

U2 E8 Miscellaneous Engineering Issues

E 8.1 (Uodate - Unresolved item 50-336/96-01-05): Containment Hvdroaen Monitors and

i Post Accident Samole System inocerable

a. Insoection Scoce

This unresolved item involved eight concerns regarding the licensee's understanding of the

functioning of the hydrogen monitors, post-accident sampling system (PASS), and

containment radiation monitors. The eight concerns were characterized as unresolved

pending further NRC and licensee review. These concerns were safety significant in that

both trains of hydrogen monitors and PASS have been inoperable. The scope of this

inspection was the first two issues which involved: (1) The post-LOCA containment

pressure / temperature profile that was performed to support the steam generator

replacement modification and its affect on safety-related systems and; (2) The ability of the

hydrogen monitors and PASS to satisfy their licensing basis and design basis. When

reviewing these two items, the NRC found that the licensee was stillin the process of

completing the corrective actions which involves a design modification to the hydrogen

monitor and PASS. Therefore, the purpose of this update is to disposition the concerns for

the first two items covered by the unresolved item from an enforcement perspective.

Accordingly, this report contains no additional technical concerns that were not previously

described in NRC Inspection Report 50-336/96-01.

E8.1.1 Hvdroaen Monitors Rendered Inocerable due to insufficient Air Flow

The unresolved item addressed that hydrogen monitors would be unable to perform their

design function due to insufficient air flow past the thermal conductivity cell when

containment pressure is low. Each of two, redundant hydrogen monitors takes suction

from and discharge back to containment. The containment radiation monitors (RE-8123A

& B and RE-8262A & B) branch off the hydrogen monitor suction piping and the PASS

system branches off the discharge piping. Each hydrogen monitor is capable of taking a

sample from the containment radiation monitor sample line (Stream 2) or directly from the

containment building dome area (Stream 1). The Stream 2 sample lines contains two

automatic isolation valves (2-AC-47 & 2-EB-89 and 2-AC-12 & 2-EB-88) actuated by a

l

i

l

4

'

30

containment isolation actuation signal (CIAS) at 4.75 psig, as are the sample return

isolation valves (2-AC-20 & 2 AC-15). The containment isolation is necessary because

leakage in excess of reviewed limits would occur through the radiation monitor pump shaft

seals if exposed to containment building pressures greater than the monitor design

operating pressure (10 psig). The post-incident hydrogen control procedure allows

operators to open the associated containment isolation valves following a LOCA when

conteinment pressure decreases to 10 psig.

The containment air enters the hydrogen monitors through pressure regulators that the

licensee stated in LER 336/96-10 are set for 10 psig which is contrary to FSAR Section

6.6.2.2, which states that the pressure regulator limits sample pressure to less than 5 psig.

The sample is then throttled to 100 cc/ minute (assuming a 10 psig inlet pressure) using a

needle valve and passes through the analyzer thermal conductivity cell which senses

hydrogen concentration. The majority of the air sample bypasses the thermal conductivity

cell through a throttled bypass line set to 6500 cc/ minute to provide sufficient flow to

PASS. This bypass flow plus the flow past the thermal conductivity cellis returned to the

containment building by the sample pump via the containment radiation monitoring system

flow path. Downstream of the thermal conductivity cells is a vacuum regulating valve

(PCV-7852 for facility 1 and PCV-7856 for facility 2) that is designed to maintain a

vacuum at the cells of 75 to 80 inches of water (approximately 12.2 p_gia).

The NRC noted that the needle valve upstream of the thermal conductivity cells, as well as

the downstream vacuum regulating volves, are set using a calibration gas pressure of 10 l

psig. Therefore, the hydrogen monitors were checked for proper operation using only

single calibration gas pressure of 10 psig even though the hydrogen monitors must operate

with a pressure range of 0 to 10 psig. The NRC found that as containment pressure

decreased to O psig, the vacuum regulating valve throttles closed to maintain backpressure

and stops air flow through the thermal conductivity cell. As a result, nearly all the

containment air flow is diverted through the bypass line rendering the hydrogen monitors

inoperable. This is safety significant because the containment hydrogen levels could be

excessive, which would necessitate operator action, while the hydrogen analyzers

indicated no hydrogen was present.

E8.1.2 Hydrooen Monitor Post-Accident Samolino, and Containment Radiation

Monitor Inocerable due to Non-Recresentative Samole Points

i

The unresolved item also involved the f act that the sample points for the containment

building radiation monitor, the hydrogen analyzers, and the PASS system were found to be

located in the suction ductwork of the containment auxiliary circulating fans. These fans

are not designed for post-LOCA operation and are not powered from a vital power supply.

If the fans are not operating post-LOCA, the sample taken from the duct would not be

representative of the containment atmosphere, rendering these systems inoperable.

Correspondence to the NRC concerning NUREG 0737, was consistently in error, in

identifying the samp!e points as being the suction piping of the post-incident recirculation

fans, not the containment auxiliary circulating fans. This was an original design deficiency

that existed since initial plant operation in 1975.

1

l

.

1

l

, i

31

E 8.1.3 Excessive Time Delav in Placino the Hydrocen Monitors in Service

The unresolved item involves an NRC review of the design and licensing basis regarding

the amount of time before operators could place the hydrogen monitors in service following

a loss of coolant accident (LOCA). The design basis, as specified in the Final Safety l

Analysis Report (FSAR), Section 6.6.3.1, states that the hydrogen monitoring system is I

manually initiated within 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> following the accident. Although there have been

numerous pieces of docketed correspondence between the NRC and the licensee since

l

1980 regarding the hydrogen monitors, the current licensing basis is not well defined. '

On October 31,1980, the NRC issued NUREG 0737, "TMl Action Plan Requirements."

Licensees were required to perform the necessary actions to demonstrate their ability to

meet the requirements or be granted an exemption. NUREG 0737, item II.F.1.6,

" Containment Hydrogen Monitor," required that a continuous indication of hydrogen

concentration ln the containment atmosphere be provided in the control room. A j

continuous indication of hydrogen concentration is not required during normal operation. If j

an indication is not available at all times, continuous indication and recording must be '

functioning within 30 minutes of the initiation of safety injection. Based on several

licensee submittals in 1981 and 1982 which described their ability to satisfy this TMi

Action Plan item, in 1983, the NRC issued a safety evaluation report and an Order that

reflected that the licensee completed this item.

The licensee provided several submittals to the NRC, dated December 31,1981; April 16,

June 25, December 29,1982; and March 8, and March 28,1983, that demonstrated their j

compliance with NUREG 0737, item II.F.1.6 The licensee was unaware at this time that  ;

they were not able to meet the 30 minute requirement and therefore, this exception was j

not reflected in their submittals. Based on a review of the licensee's submittals, on June

27,1983, the NRC issued a safety evaluation report (SER) that concluded that the

requirements of NUREG 0737, Item II.F.1.6, have been met and the action item was

considered resolved. On March 14,1983, NRC issued an Order confirming the licensee's

commitments on post-TMl related issues that indicated that Item II.F.1.6 was completed.

In a March 27,1984, letter to the NRC, the licenseo stated that a deviation from item

II.F.1.6 had been identified that was not discussed in their previous submittals and

therefore was not reflected in the associated safety evaluation report or confirmatory

Order. The letter indicated that they were unable to meet the 30 minute requirement for

hydrogen concentration indication because the hydrogen monitors share a common line

with the containment radiation monitors whose design makes it undesirable and impractical

to open this penetration until containment pressure is between 0 and 10 psig. The letter

described that containment pressure for the design basis LOCA will be less than 10 psig in

approximately three hours after the initiation of the event.

In 1982, the NRC issued Supplement 1 to NUREG 0737, " Requirements for Emergency

Response Capability.," NUREG 0737, Supplement 1, paragraph 6.1, requires licensees to

provide measurements and indication of Type A,B,C,D, and E variables listed in Regulatory

Guide (RG) 1.97 (Rev. 2), " Instrumentation for Light-Water Cooled Nuclear Power Plants

and Environs Conditions During and Following an Accident." RG 1.97 recommends that

hydrogen monitoring instrumentation remain functional for containment pressures from -5

. - - - - _ - ._ .

.

O

32

psig to the maximum design pressure. In a 1984 letter to the NRC, the licensee described

their inability to satisfy the requirement that the hydrogen monitors remain functional for

containment pressures up to the maximum design pressure because the analyzers were

designed for operation under a positive pressure up to 10 psig. Referring to the licensee's

March 27,1984, letter, the NRC granted an exemption to RG 1.97 stating that the

operational limit of 10 psig has been shown by the licensee to be acceptable because the

hydrogen concentration instrumentation is not necessary until after the containment

,

pressure has decayed to less than 10 psig. Although the licensee's March 27,1984, letter

stated that containment pressure will be less than 10 psig in approximately three hours

following a LOCA, the NRC did not specifically mention the 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> because the RG 1.97

item involved only the hydrogen monitor design pressure.

The inspector determined that the NRC did not explicitly approve a deviation from the

requirement in NUREG 0737, item II.F.1.6, that the hydrogen monitor be functional within

30 minutes of the initiation of safety injection even though the NRC was appropriately

notified of the deviation. However, the NRC granted an exemption to RG 1.97, which as

enacted by Supplement 1 to NUREG 0737, allowing the hydrogen monitor to be placed in

service when containment pressure dropped below 10 psig. Since the NRC referenced the

licensee's March 27,1984, letter, placing the hydrogen monitors in service approximately

three hours after a LOCA was implicitly found acceptable. It is therefore reasonable to

conclude that the current licensing basis requires the hydrogen monitors to be in operation

within approximately three hours of an accident.

The actual time delay of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> to place the hydrogen monitors in service differed from

both the design basis (12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />) and the licensing basis (approximately three hours). The

NRC found that due to an inadequate review of the steam generator replacement

modification, the licensee f ailed to identify the that design basis and licensing basis i

regarding the time needed to place the hydrogen monitors in service could no longer be

met. The engineering evaluation for the modification included ABB-CE Calculation 006- .

AS92-C-010, " Millstone Unit 2 LOCA Containment Pressure / Temperature Analysis for l

Steam Generator Replacement," dated October 15,1992. The results of the calculation

{

showed that it would take approximately 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> following a LOCA for containment

pressure to drop below 10 psig.  ;

E8.1.4 Time Delav in Drawina a Containment Post-Accident Air Samole

Similar to the hydrogen monitors described above, the unresolved item also addressed the

fact the licensee did not meet their design basis regarding the amount of time necessary to

sample and analyze the containment atmosphere. The containment atmosphere samples I

are collected by the PASS via connections on the discharge of the hydrogen monitor

sample pump. FSAR, Section 9.6.3.1, states the containment air PASS has the capability

of collecting a sample of containment air as required by NUREG 0578, Appendix A, Section l

'

2.1.8.A. NUREG 0578 states that the licensee should be able to obtain a containment

atmosphere sample in less than 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and analyze the sample for radioisotopes within the

following two hours under accident conditions. The three hour time period starts from the

time it is determined that a containment air sample is needed. Similarly, NUREG 0737,

item II.B.3, " Post Accident Sampling Capability," required the licensee to have the i

capability to promptly obtain reactor coolant and containment atmosphere samples. The I

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

.

i

,

33

4

combined time allotted for sampling and analysis should be three hours or less from the

time a decision is made to take a sample.

'

in a licensee letter dated November 1,1982, the licensee stated that the entire sampling

operation including preparation, sample recirculation, sample isolation, purge of the system

'

piping, sample retrieval, transport to the chemistry laboratory, and analysis, for both

reactor coolant and containment air samples can be completed within 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br />. In letters

, dated January 12 and April 19,1984, the licensee provided additional information -

describing their ability to satisfy the requirements of the NUREG 0737, item ll.B.3. In an

SER dated June 14,1984, the NRC stated that based on a review of the licensee's letters

dated January 12 and April 19,1984, the provisions of NUREG 0737, item II.B.3 were

satisfied. The SER stated that the licensee has provided the capability to promptly sample

i

and analyze containment atmosphere samples within three hours from the time the

decision is made to take the sample.

i

1

The NRC determined that the current design and licensing basis is that the containment air l

i be sampled and analyzed in three hours. Similar to the hydrogen monitors, the NRC found

l that due an inadequate review of the steam generator replacement modification, the

-

licensee failed to identify that the design basis and licensing basis regarding the time ,

'

needed to take a containment atmosphere sample could not be met during the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />

following a LOCA when containment pressure was above 10 psig.

E8.1.5 Conclusion

Technical Specification 3.6.4.1 requires that two independent hydrogen monitors be

operable. FSAR, Section 6.6.2.1, requires that two, full capacity hydrogen concentration

! monitoring systems are provided outside the containment for periodic or continuous

2

analysis of hydrogen concentration of the containment atmosphere and that uniform mixing

of the containment post-accident atmosphere is provided by the post-accident recirculation

system. The fact that both hydrogen monitors were inoperable due to: a. insuf ficient

a

containment air flow past the thermal conductivity cell at low containment pressures and,

, b. taking a non-representative suction from ductwork of the non-vital containment auxiliary

recirculating fans, is considered an apparent violation. (eel 336/96-08-11)

Appendix B to 10 CFR 50, Criterion 111, " Design Control", requires measures to be

established to assure that applicable regulatory requirements and design bases be correctly

j translated into specifications, procedures, and instructions. Due to an inadequate review

of the steam generator replacement modification, the licensee failed to identify that design

basis and licensing basis regarding the time necessary to place the hydrogen monitors in-

service and, the time needed to take a containment atmosphere sample, could not be met

! during the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> following a LOCA when containment pressure was above 10 psig.

, The failure to establish adequate design control measures for the steam generator

replacement modification is considered an apparent violation. (eel 336/96-08-12)

10 CFR 50.71(e) requires the licensee to periodically update the FSAR originally submitted

as part of the application for the open. ting license. The failure of the licensee to update

<

the Unit 2 FSAR, Section 6.6.3.1, which states that the hydrogen monitoring system is

,

manually initiated within 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> following the accident, to reflect the licensing basis time

.

.

34

of approximately 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> described in the licensee's letter to the NRC dated March 27,

1984, is considered an apparent violation. (eel 336/96-08-13)

E8.2 (Closed) LER 50-336/96-17: Hydroaen Monitorina System Dearadation Not Restored

This LER discussed an NRC finding associated with using operator compensatory measures

as a permanent corrective action for addressing a single failure vulnerability in the

hydrogen monitoring system. This issue is discussed in detail in NRC Inspection Reports

50-336/96-01 and 96-201. Final resolution of the system deficiencies will be tracked by

open item eel 336/96-201-41. This LER is closed.

E8.3 (Ciosed) LER 50-336/96-20: Potential Loss of Reactor Buildina Closed Coolina

Water to the Primarv Makeuo Water System

a. Insoection Scoce

The scope of this inspection included a review of Licensee Event Report 50-336/96-20.

b. Observations and Findinas

The licensee discovered that both trains of the reactor building closed cooling water

(RBCCW) system could become inoperable following certain design basis events if

inventory were to be lost through the common primary makeup water (PMW) system fill

line. The RBCCW system has a single surge tank which serves both trains. An installed

weir inside the surge tank ensures a minimum of approximately 1000 gallons are available

for each train. Each of the three RBCCW pumps has a minimum flow recirculation line that

ties into a common line that directs flow back to the top of the RBCCW surge tank.

Makeup to the RBCCW surge tank is provided by the PMW system which taps into the

common minimum flow recirculation line.

The licensee discovered that if the non-vital PMW pump were lost and the PMW makeup

line isolation valve was to fail open on loss of air, the PMW system would depressurize

allowing recirculation line flow to be diverted to the PMW surge tank. This scenario is

unlikely since there is an installed non-QA check valve within a few feet of the RBCCW

surge tank level control valve air operator that would have prevented significant backflow

through the PMW line. However, since this is a non-tested /non-QA check valve, its

presence cannot be credited. A failure of the non-seismic PMW system would also result

in the loss of RBCCW inventory. The licensee reported this condition in accordance with

10 CFR 50.72(b)(2)(i), as a condition that would have resulted in the plant being seriously

degraded or being in an unanalyzed condition.

Since the RBCCW system was currently needed to support shutdown cooling, the

operability of the system was reestablished by tagging closed the PMW isolation valve.

The licensee later decided to reestablish automatic makeup capability to the RBCCW surge

tank by isolating the RBCCW recirculation line for each pump and reopening the PMW line

isolation valve. The licensee determined that isolating the recirculation lines was

acceptable because procedural controls ensure a flow path for the RBCCW pump discharge

without crediting the recirculation lines. However, as a longer term corrective action, the

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

.

1 ,

35

'

licensee has a priority "A" design modification that they are scheduled to complete during

the current mid-cycle shutdown.

The licensee discovered this deficiency during a RBCCW system design review which was

a licensee initiative that was planned in October 1995, and implemented in April 1996.

Therefore, this initiative was planned prior to the March 7,1996, letter in which the NRC

required the licensee to perform design reviews to ensure facility operation is in accordance

with the plant's operating license. The RBCCW design review was prompted when a

service water system design review revealed potential concerns with the RBCCW system. ,

'

The cause of the RBCCW/PMW design deficiency was an oversight in the original design of )

l RBCCW recirculation line. This condition was not likely to have been identified by routine '

licensee efforts such as normal surveillance or quality assurance activities.

i c. Conclusion

!

4

The licensee's interim corrective action of isolating the RBCCW recirculation line to i

reestablish RBCCW operability until the modification is completed was found to be I

"

acceptable. The NRC determined that the RBCCW recirculation line design vulnerability

was a design control discrepancy associated with the plant's original design and is l

] considered a violation of 10 CFR 50 Appendix B, Criterion lil, " Design Control." In '

accordance with Section Vll.B.3 of the Enforcement Policy, the NRC may refrain from

proposing a violation involving a past problem, such as an engineering or design deficiency,

provided that the violation meets the following criteria: a. It was licensee-identified as a i

result of its voluntary initiative; b. It was or will be corrected, including immediate i

corrective actions and long term corrective actions to prevent recurrence within a I

reasonable time following identification; and c. It was not likely to be identified by routine

licensee efforts such as normal surveillance or quality assurance activities. As described

above, each of these criteria are satisfied and therefore, this violation is being treated as a

Non-Cited Violation, consistent with Section Vll.B.3 of the NRC Enforcement Policy.

E8.4 (Closed) LER 50-336/96-21: Snubber Failure in Pressurizer Sorav Line

This snubber failure was discussed in detailin NRC Inspection Report 50-336/96-01.

Licensee Event Report 50-336/96-21 is administratively closed.

E8.5 (Closed) LER 50-336/96-22: Hvdroaen Purae Valve Sinale Failure Vulnerability

a. Insoection Scooe

The scope of this inspection was a review of Licensee Event Report 50-336/96-22.

b. Observations and Findinas

On April 23,1996, the licensee discovered a single failure vulnerability of the hydrogen

purge valve interlock to the enclosure building filtration system (EBFS) heaters. During a

containment purge to reduce hydrogen concentration following a loss of coolant accident,

the hydrogen purge valve interlock is designed to deenergize the both trains of EBFS

heaters when any one of four purge valves (two per train) is opened to prevent a hydrogen

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

'

.

d

36

explosion as air passes through the EBFS. This creates a single failure vulnerability

"

because the interlock on a single purge valve deenergizes the EBFS heaters in both trains.

The EBFS heaters are designed to maintain the relative humidity of the air at less than 60

'

percent to reduce the buildup of moisture on the EBFS charcoal absorbers and HEPA filters.

The technical specification surveillance verifies the charcoal efficiency at a relative

<

humidity of 95 percent which assumes no heaters are available. The safety significance of

this design deficiency is low because EBFS is capable of performing its design function

j without crediting operation of the heaters. This is based on the fact that although the

containment air could be at 100 percent humidity, the purge flow of 50 standard cubic feet

i

per minute (SCFM) would be diluted with the exhaust flow from the enclosure building of

5950 SCFM. The HEPA filters are designed to operate with 100 percent humidity.

The cause of this condition was an ovqsight in the original design of the interlock. As

corrective actions, the licensee installeo a temporary modification to defeat the hydrogen

purge valve /EBFS heater interlock which was necessary to support their plans involving the

movement of spent fuel. In addition, the licensee stated in the LER that a permanent  ;

design change was planned that eliminates this single failure vulnerability prior to startup

from the current mid-cycle outage.

c. Conclusion

The fact that this design vulnerability was discovered during the performance of an EBFS

surveillance reflects a good questioning attitude by plant personnel. The NRC determined

that the single failure vulnerability created by the hydrogen purge valve /EBFS heater

interlock was a design control deficiency associated with the plant's original design and is

considered a violation of 10 CFR 50 Appendix B, Criterion Ill, " Design Control." This

licensee-identified and corrected violation is being treated as a Non-Cited Violation,

consistent with Section Vll.B.1 of the NRC Enforcement Policy.

E8.6 (Closed) LER 50-336/96-28: Potential Service Water Strainer Backwash Valve

Failure Durino Flood Conditions

a. Insoection Scord

The scope of this inspection included a review of Licensee Event Report 50-336/96-28.

b. Observations and Findinos

The licensee discovered that potential intake structure flooding could render both trains of

service water (SW) inoperable because electrical shorts of the SW strainer backwash valve

solenuids would result in the loss of the automatic backwash control circuit as well as the

SW strainer motor starters. The inability to backwash the SW strainers to clear debris

buildup would eventually result in an excessive restriction of SW flow. The Final Safety

Analysis Report (FSAR), Section 2.5.4.2.4, " Flood Protection of Electrical Equipment,"

requires that the power and control cables necessary for trouble-free SW pump operation

that are below the 22 ft, elevation are of tight construction. Since the elevation of the

backwash valve solenoids is approximately 15 ft., the licensee reported this condition in

.

d

37

accordance with 10 CFR 50.73(a)(2)(ii)(B) as a condition that resulted in the plant being i

outside of its design basis, i

The licensee determined that the cause of the event was an inadequate original design in

that flood protection requirernents were not incorporated into the design of the backwash

valve solenoid. As corrective actions, abnormal operating procedure AOP 2560, " Storms,

High Winds, and High Tides," was changed to specify that the solenoid cables be i

disconnected when flooding conditions are expected which causes the backwash valves to

{

fail open and allows the strainer motor to function normally. In addition, the licensee

committed in LER 50-336/96-28 to perform a modification prior to entering mode 4 that

relocates the solenoids above the 22 ft. elevation. j

i

c. Conclusion ,

I

The NRC determined that the failure the SW backwash valve solenoids to meet the FSAR

flood control requirements was a design control discrepancy associated with the plant's

original design and is considered a violation of 10 CFR 50 Appendix B, Criterion Ill, " Design )

Control." This licensee-identified and corrected violation is being treated as a Non-Cited l

Violation, consistent with Section Vll.B.1 of the NRC Enforcement Policy.

E8.7 (Closed) LER 50-336/96-29: Removal of the Startuo Rate Trio Feature was  !

Potentially Non-Conservative

l

l

a. Insoection Scoce  !

l

The scope of this inspection included a review of Licensee Event Report 50-336/96-29.

b. Observations and Findinas

l

1

The licensee discovered that the analysis associated with the removal of the high startup

rate reactor trip in 1978 did not thoroughly evaluate the consequences of a potential

control rod withdrawal accident starting with the reactor subcritical. The original plant

design basis reflected that the reactor protection system (RPS) high startup rate trip was

used for equipment protection only and no credit was assumed for this trip in accident

analyses. The variable over-power trip was the design basis method credited for protection  ;

in the event of a control rod withdrawal accident. In 1977, the licensee decided that the

high startup rate trip and its associated setpoint should be removed from technical

specifications because spurious actuations of the startup rate trip modules were causing

reactor trips. The licensee's technical specification change request indicated that no

credit was taken for the high startup rate trip in the safety analysis and that no safety limit

was directly related to the trip. Technical Specification Amendment No. 38, which

removed this trip, was issued on April 19,1978. The licensee physically removed the high

startup rate module from the reactor protection system shortly thereafter.

j During their current ongoing review of the plant's design basis, the licensee discovered

l that there were certain plant conditions in which the variable over-power trip would not

l initiate soon enough to protect the core. If the control rod withdrawal event began when

j the core was subcritical, in mode 3, and boron concentration was approximately 300 to

.

.

38

500 ppm lower than required, local core power would exceed analyzed values before the

variable over-power trip could shut down the reactor. The licensee reported this in

accordance with 10 CFR 50.72(b)(2)(iii)(A) as a condition that alone could have prevented

the fulfillment of the safety function of systems needed to shut down the reactor.

On July 19,1996, ASEA Brown-Boveri - Combustion Engineering (ABB-CE) informed the

licensee that the plant's original design basis incorporated the protective action of the high l

startup rate trip to provide protection for subcritical rod withdrawal events. ABB-CE

believes the original design basis evaluation was incorrect because they assumed that rod

withdrawal events starting from zero percent power were most limiting and therefore, did j

not evaluate the event starting from a subcritical condition.

j

i

'

The licensee is in the process of evaluating whether the postulated subcritical rod

withdrawal event is considered credible. If they conclude that the startup rate trip is ,

required, the licensee committed to implement necessary corrective actions during the I

current outage. The licensee committed in LER 50-336/96-29 to submit an LER l

supplement describing the results of their analysis and any necessary corrective actions.

At the end of the inspection period, the licensee was still in the process of evaluating the  ;

startup rate trip concern. I

c. Conclusion i

The concern described in LER 50-336/96-29 regarding the appropriateness of removing the

startup rate trip from the reactor protection system in 1978 is considered unresolved to

allow NRC review of the licensee's disposition of this concern. (URI 336/96-0814)

.

et

.

"

,

39

Report Details

Summarv of Unit 3 Status

Unit 3 remained in cold shutdown throughout the inspection period. On October 5,1997,

the plant experienced an inadvertent heatup of 1.8cF due to a malfunction with a

component cooling water temperature control valve. Operators immediately identified and

corrected this condition. The licensee continues to implement configuration management

program activities, engineering reviews, and docketed. correspondence assessments to

verify compliance with their established design and licensing basis as required by the NRC

prior to plant restart.

U3.1 Operations

U301 Conduct of Operations

01.1 General Comments (71707)

The inspectors conducted frequent reviews of control room activities, plant shutdown risk

management controls, and ongoing unit operations in mode 5. In general, the conduct of

operations was professional and safety conscious, with evidence of the appropriate

consideration of operability determinations (ODs) associated with adverse condition

reports. The inspector observed a plant operations review committee (PORC) meeting (3-

96-198), evaluating the processing and approval of ODs, bypass-jumpers (i.e., temporary

modifications), and procedural revisions. Where applicable, e.g., for OD MP3-210-96,

acceptable compensatory measures were implemented to ensure the continued operability

of systems that were determined to not be fully qualified under specific analyzed

conditions, e.g., seismic events.

The inspector also noted proper consideration of the relevant safety criteria and technical

specification (TS) controls in the isolation of the "C" reactor coolant loop for the conduct j

of steam generator (S/G) eddy current testing (ECT) and tube plugging activities.  ;

Conservative planning was noted, to include taking boron concentrations samples and l

performing evaluations to assure that the restoration of the loop would not effectively

result in a positive reactivity addition to the reactor core. Ultimately, a decision was made

to defer loop restoration until an appropriate technical specification revision could be

processed. Upon completion of the steam generator "C" ECT, the licensee submitted a

special report to the NRC, in accordance with TS 4.4.5.Sa, documenting the plugging of

two additional S/G tubes due to anti-vibration bar wear. As reported, the appropriate

material (Inconel 690) was used in the mechanical plug installation. The inspector noted

that as of the current cycle 6 of Unit 3 operation, only 41 S/G tubes have required

s

plugging to date, out of the total number of tubes (over 22,500)in all four S/Gs.

< l

l

. - _ . .-

.

4

.

40

U3 02 Operational Status of Facilities and Equipment

O 2.1 Technical Soecification (TS) Consistency (71707)

a. Insoection Scooe

The inspector reviewed a plant operations review committee (PORC) agenda that was

scheduled to discuss Final Safety Analysis Report (FSAR) change request 95-MP3-62,

which documents the evaluation of an event affecting the design basis of the spent fuel

pool. The licensee believes that the analyzed event represents a situation beyond the

design basis of the plant and the recommended changes do not reflect an unreviewed

st.fm question. However, since it appears that a revision to Bases for TS 3/4.7.4 would

be r4uired, in addition to the FSAR changes, the inspector reviewed the licensee's

documentation to assess the consistency of the plant TS with its design basis, as well as

the technical adequacy of the supporting licensee analysis.

b. Observations and Findinas

The service water (SWP) system at Unit 3 is designed with two pumps per train. TS

3/4.7.4 does not specify the number of pumps required for a SWP train to be considered

operable. A licensee reportability evaluation (i.e.,90-049) evaluated this question and

dett.ained that only one service water pump is necessary to maintain a SWP train

operable. However, under post-LOCA conditions and a single failure of the redundant SWP

train, one SWP pump does not have sufficient flow capacity to cool both the emergency

core cooling system (ECCS) loads and the reactor plant component cooling (CCP) system.

Since the CCP system provides cooling to the spent fuel pool, spent fuel pool cooling is

considered unavailable for a period of time (i,e., four hours is discussed in the Safety

Evaluation Report, SER) under these postulated event conditions.

If the event scenario assumes that one of the available service water pumps had been

taken out of service for maintenance activities, the length of time necessary to restore

spent fuel pool cooling directly relates to how quickly a second SWP can be returned to

service in the operable train. The inspector reviewed the Bases for TS 3/4.7.4, FSAR

Sections 3.1.2.44 and 9.1.3, SER Section 9.2.1, and 10 CFR 50, Appendix A, Criterion 44

and noted some inconsistencies in the assumption that a second SWP in each train is not

necessary to support the operability of the train. These documents imply that the

continued operation of all safety-related equipment is supplied by sufficient SWP cooling

under all accident conditions. As discussed above, the spent fuel pool cooling, a safety "

related function, is unavailable during certain post-LOCA times if only one SWP pump can

provide the cooling flow.

Therefore, a key issue appears to be whether removal of a SWP pump for maintenance

constitutes an activity which needs analysis for some allowable outage time (AOT) defined

by the unit TS. Currently, TS 3/4.7.4 only specifies an AOT fer one SWP loop out of

service, but does not specify the number of pumps required per loop. The licensee's

proposed FSAR chcnge would come under the purview of the NRC Office of NRR for

further review. Nevertheless, the inspector indicated that the questions involving the need

.

.

41

for additicnal SWP flow capacity than can be provided by one pump need to be addressed

by the licensee prior to plant startup. Further review of this issue is merited.

c. Conclusion

Continued NRC review of the licensee's technical bases for considering a SWP train / loop

operable with only one operable SWP pump is considered an inspector follow-up item (IFl

50-423/96-08-15). Licensee consideration of the second pump as an " installed spare"

merits acceptance if the post-LOCA scenario, planned for documentation in the FSAR

change, leading to the loss of spent fuel pool cooling can, in fact, be addressed as an

event beyond the design basis of the Unit 3 plant.

U3 05 Operator Training Qualification

05.i Ucensed Ooerator Reaualification Trainina (LORT) Proaram Evaluation (71001)

a. Insoection Scope

Two NRC inspectors evaluated Millstone's LORT program, including UFSAR Chapter

13.2 commitments, using Inspection Procedure (IP) 71001 during the wepk of

September 16,1996. The inspectors evaluated the following areas: adequacy of

written and operating tests; administration of operating tests; effectiveness of the

training feedback program; adequacy of the remedial training program; conformance

with license conditions; and program administration.

A review of examination material was performed per the requirements of Millstone

Unit 3 training procedure "MP3 Licensed Operator Requalification Training," Rev.1,

dated July 12,1996. The procedure states that the program complies with the

requirements of 10 CFR 55.59, "Requalification.*' The material was evaluated using

the guidelines in NUREG 1021, Rev. 7, the Examiner Standards. Annual written

exams were not administered during the LORT cycle when the inspectors were on

site.

b. Observations and Findinas

Test Adeauacv

Six annual written examinations administered in 1995 were reviewed. The exam

questions were operationally oriented and safety significant. Topics identified in 10

CFR 55.41 and 10 CFR 55.43 were adequately covered on the exams. The exams

contained no overlap of questions, and they exhibited appropriate variation between

the reactor operator (RO) and senior reactor operator (SRO) level exams. Exam

questions were generally written at higher cognitive levels (comprehension /

application). However, there were some memory level knowledge questions or

direct look up questions on the exams.

The simulator evaluation bank had 37 scenarios. Twenty-three of these scenarios

were available for use and fourteen were in various stages of revision. The LORT

__ _ _ _ _ .___. ____ _ _ .__ _ . ._ _ __ _

=  :

l

.

42

coordinator indicated the bank was continually being upgraded by revisir.g at least

two scenarios and developing one new scenario during each six-week training cycle.

Selection of scenarios for each exam is based on picking scenarios that meet the

sample plan theme and have not been seen by the crew in the last year. t

l

Four simulator scenarios were reviewed by the inspectors which comprised the i

operating exams administered to one operating crew and one staff crew during

cycle 96-6. The scenarios met the qualitative and quantitative attributes of the

Examiner Standard. When placed in sets, the scenarios were capable of evaluating

the full range of competencies.  :

Annual job performance measure (JPM) operating exams were not administered

during this cycle so the individual JPMs for this cycle were not reviewed. A brief

review of the JPM bank indicated that 151 JPMs were in the bank. None of the

JPMs were time critical. The facility indicated they would review their task lists to {

see if any JPM should be time critical.  ;

Examination Administration i

Four simulator scenarios and the evaluation process were observed by the  !

inspectors. Both crews performed satisfactorily on the scenarios. One operator

failed his operating exam because he did not take timely action to isolate a steam

leak. The evaluators included both operations and training management. The

evaluations were thorough and provided constructive feedback to the crews and

individual crew members. The inspectors determined that the facility practice of ~

having discussions and critiques with the crew shortly after the exam was good.

The inspectors noted that operations management took an active, effective part in

the training and evaluation process.

'

Command and control during the scenarios was generally good, however, there

were instances where the shift manager approached the control boards and became

involved in event diagnosis and did not maintain a position of oversight. This was

significant because one scenario included a failure of the reactor to automatically i

trip. The crew did not recognize that two first out annunciators were lit for

approximately 20-25 seconds. If the shift manager had remained in a position of

oversight, he would have had a better opportunity to recognize the first out

annunciators sooner than the 20-25 seconds it took for the shift technical advisor

(STA) to identify them. Briefing sessions were held at appropriate times when

conditions allowed. Sometimes it was not evident when the briefing session w~as

finished. l

Trainina Feedback

Sevaral mechanisms were used by the facility to obtain operator feedback. The {

feedback system appeared to be effective. A memorandum from the assigned

mentor to the shift manager provided the response to individual operators'

comments. The effectiveness of the process appeared to depend strongly on the

individual shift manager. Nonetheless, the inspectors interviewed several operators

. . _ _ _ _ . _ _ __ . _ _ _ _ _ . . __ . _ __ . _.

.

4

.

43

and all of the operators believed they were informed of how their comments were

handled. The inspectors noted that the staff crews have not had a mentor assigned

i to them and, therefore, have not had this feedback process.

Remediation Trainina

'

The Millstone 3 LORT program appeared to be effective in identifying operators

requiring remediation and in developing appropriate remediation plans to address

! individual or crew weaknesses. Records associated with five remediation plans

! were reviewed. Two discrepancies were identified:

.

Several plans expected the operators to review questions missed on their

written exams. For each question missed, the individual was directed to

identify the basis for the correct answer and why the distractors were

incorrect. They were also directed to identify the references used to support

their response. Operators were not completely satisfying this requirement in i

that they did not provide justification as to why distractors were incorrect.

In one case, the operator circled the correct answer and did not appear to

provide supporting information at all. The inspectors determined that e

closer review of the remediation packages needed to be performed to ensure

all aspects of the plan are satisfactorily completed.

Two remediation plans had documentation in the files as being conipMted,  !

however, there were no signatures or dates on the plans that the plan had j

ever been completed. The LORT coordinator indicated to the inspectot that '

he had verified by telephone conversations that the remediation was  ;

completed, but there was no documentation to support this conclusion. I

Current practice places an unsigned copy of the plan in the file to document

development, but there was no documentation to indicate completion. The ,

inspectors judged this to be a weak approach. '

c. Conclusions

Two inspectors evaluated Millstone Unit 3 LORT program and concluded that the

program met NRC requirements. Operator performance on the dynamic simulator

exams was satisfactory, and facility evaluations of operator performance were

acceptable and included effective operations department management.

Written exam questions met the guidelines and contained no overlap of questions,

an indication of a high quality examination. The simulator scenarios also met the

guidelines.

Some minor discrepancies were identified regarding remediation plans which were

not being cc,mpleted and documented in a thorough manner.

.

.

44

05.2 UFSAR Commitments

The inspectors determined that Millstone Unit 3 met their LORT program

commitments as found in the UFSAR. Nonetheless, as part of the UFSAR review,

the inspector identified that operator actions discussed in the analyses of steam

generator tube ruptures (SGTRs) appeared to have changed in a nonconservative

direction, and the UFSAR had not been updated to reflect the current information

nor had the NRC been notified of the changes.

Specifically, UFSAR section 15.6.3.1, Identification of Causes and Accident

Description, for SGTR accident analyses states that " Consideration of the

indications provided at the control board, together with the magnitude of the break

flow, leads to the conclusion that the accident diagnostics and isolation procedure

can be completed within 30 minutes of initiation for the design basis event." This

conclusion appeared to be supported by analyses submitted in a November 7,1994

letter in response to an April 13,1994 NRC request for the faci!ity to show that the

operator action times assumed in the SGTR analysis were realistic and echievable.

However, in the November 7,1994, letter Northeast Utilities noted that "In order to

assure ourselves that Millstone Unit No. 3 continues to meet the SGTR analysis

basis, NNECO has determined to revalidate the operator response times assumed in

the plant specific SGTR analysis by conducting additional operating crew design

basis SGTR simulations. This action is expected to be completed by April 1995."

The NRC issued a safety evaluation report (SER) dated December 22,1994,

accepting the SGTR analysis based, in part, on the times provided in the November

1994 letter.

The inspector found that when these simulations were performed in early 1995, the

results showed that the average times for operators to isolate the leak (i.e., equalize

primary and secondary pressures for the ruptured SG) were longer, on the order of

41 minutes versus 30 minutes. However, there appeared to have been no

subsequent reanalyses of the event, updating of the UFSAR, or communication to

the NRC regarding the revised results.

In a November 12,1996 telephone conversation with the inspector and a Region I

manager, the licensee, including a licensing representative, stated that preliminary

analyses performed subsequent to the inspection had determined that the

consequences of the extended operator action times were still acceptable, based in

part on additional considerations such as throttling of auxiliary feedwater flow to .

the ruptured SG. In addition, the facility representatives stated that efforts were in

progress to finalize the analyses and to submit the results to NRC within months.

Based on this information, the acceptability of extended operator action times for

SGTR analysis represents an unresolved item (URI 423/96-08-16).

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

.

'

.

1 45

U307 Quality Assurance in Operations

07.1 Followuo of Questionable Fuse Quality (92901)

l

l

a. Insoection Scone j

l

On September 11,1996, a site-generic adverse condition report (ACR) M3-N-0759 was

-

issued to document the discovery that fuses with cracked ferrules had been issued for i

installation. Corrective actions for this ACR were directed to the future inspection of fuses I

prior to issuance and an inspection of the warehouse stock for similar defects.

Subsequently, a Quality and Assessment Services (OAS) review of the ACR follow-up

activities determined that the corrective measures failed to identify all defective fuses and

4

did not evaluate possible plant effects. This concern was documented in ACR M1-96-

0608 on September 25,1996. The inspector discussed this issue with QAS personnel on

September 27,1996, and conducted further review of the continuing licensee investigation

and evaluation activities.

b. Observations and Findinas

! At the time of QAS identification of corrective action concerns regarding this problem,

2

additional fuses with cracked ferrules were found in the warehouse stock supply, no

engineering evaluation had been performed of the impact that the cracked ferrules might

have on the fuse function, no investigation had been conducted to determine whether any

defective fuses were installed in the units, and the vendor had not been contacted to

initiate a review of 10 CFR 21 applicability to this problem. The fuses of initial concern

were supplied by Gould Shawmut (Amptrap catalog numbers). The inspector reviewed the

pertinent ACRs and a letter from Gould Shawmut, dated October 28,1994, indicating that l

. split ferrules were the result of stress corrosion cracking caused by internal stresses in the

i

brass ferrule material remaining from the manufacturing process. In August,1994, Gould

Shawmut addressed this problem by changing the ferrule material to pure copper.

j information was subsequently provided to the inspector indicating that the cracked ferrule

issue had originally been identified by QA auditors for the Browns Ferry Nuclear Plant in

1993 during a supplier evaluation at the Gould Shawmut manufacturing facility. Testing

!

associated with the Browns Ferry audit led Gould Shawmut to conclude that the split

, ferrules would not compromise the integrity of the fuses to the point of adverse impact

upon interrupting capability.

During a discussion between the licensee and the inspector on September 30,1996, the

licensee stated that it had started marking (" red tagging") and segregating the susceptible

fuses (i.e., pre-August,1994) in the warehouse, had started the Part 21 evaluation

3

process, and had initiated operability determinations (Ods) for each of the three Millstone

, units. The following day, the inspector met with the licensee and was informed that

a

additional actions (e.g., independent testing of some suspect fuses) and development of an

integrated action plan on this issue were being planned. Subsequently, a licensee

inspection of warehouse fuse supplies identified a non-QA fuse, supplied by Bussman, with

cracked ferrules. In response to this finding, the licensee expanded the scope of its

investigation and initiated revisions to the existing ODs. Some filed walkdowns of installed

'

j fuse applications were also performed.

,

,

_ _

e

.

46

The licensee continued discussions with the various suppliers of fuses to Millstone Station

and committed to recall all fuses designated ior QA use back from field storage for {

inspection. Additional information, received from the South Texas Project, was reviewed

and determined to be consistent with the determinations that the operability of fuses with i

split ferrules was not a problem. Functional and structural testing to confirm this position

are still planned. The inspector noted that by October 1,1996, the licensee had developed i

and was implementing a detailed plan to address all concerns related to the identified

cracked fuse ferrules. Emphasis on a comprehensive resolution of this issue was provided

by the Nuclear Oversight Organization, with indications that NRC inspector questioning i

focused licensee management attention on the need to address more timely corrective

measures. While licensee corrective actions were continuing, as of the end of this

inspection period, the inspector had no additional questions at this time regarding the

installed fuses or on the future licensee plans to complete the assessment of this issue. l

c. Conclusions

initial licensee actions to address the concerns raised by the discovery of the cracked fuse

ferrules (i.e., ACR M3-96-0759) were inadequate until the Nuclear Oversight Organization

became involved in the ACR corrective action followup. Even then, evidence that the

licensee was appropriately focused on the resolution of this issue in a timely manner with a

detailed plan was not forthcoming. The NRC inspectors began questioning the problem

status and actions, and in the end, the licensee determined that the suspect fuses were

acceptable for use. This position is being verified by additional testing, the results of

which will be reviewed during a future inspection as inspectian followup item (IFl 50-

423/96-08-17). However, during the period of time when the uncertainties regarding fuse

quality were first raised, licensee efforts to address these concerns could have been more

focused and directed to a more prompt resolution.

U3 08 Miscellaneous Operations issues (92700)

08.1 (Closed) LER 50-423/96-02: This LER documented a historical condition in which

Technical Specification (TS) 4.4.1.6.2 was violated. Specifically, the licensee failed to

verify the reactor shutdown margin within 30 minutes of unisolating a reactor coolant loop.

The time requirement of 30 minutes was exceeded by between 3 and 13 minutes

depending upon interpretation of when the reactivity determination was made and when

the cold leg stop valve was considered open. This issue was previously discussed in NRC

Inspection Report 50-423/96-05 and remained open pending completion of the licensee's

corrective actions. The inspector verified that the necessary procedural changes had been

made to the specified procedure and that the personnel involved in the event had been

briefed. To prevent recurrence, the licensee plans on reviewing other department

surveillance forms and changing the procedure handbook review checklist to ensure

surveillances contain adequate data such that the listed acceptance criteria can be

validated.

This licensee identified and corrected violation is being treated as a Non-Cited Violation,

consistent with Section Vll.B.1 of the NRC Enforcement Policv. This LER is closed.

.

.

47

08.2 (Closed) LER 50-423/96-14: This LER documented that several TS surveillances for

the emergency diesel generators had been performed during plant operation, versus

shutdown, contrary to the requirements of TS 4.8.1.1.2.g. The surveillances were

subsequently performed with the unit shutdown and the diesels declared operable. The

inspector audited the licensee's surveillance tracking systems and identified that although

the primary tracking system had been updated to indicate the performance of the

surveillance every refueling outage, the backup tracking system " access" was not. The

licensee generated an adverse condition report M3-96-0669 to document and resolve this

concern. The licensee had submitted a proposed TS change in June 1995 to change the

surveillance frequency to delete the requirement to perform the diesel surveillances during

shutdown. This licensee identified and corrected violation is being treated as a Non-Cited

Violation, consistent with Section Vll.B.1 of the NRC Enforcement Policv. This LER is

closed.

08.3 (Closed) LER 50-423/96-17: During a review of NRC Information Notice 96-06, the

licensee identified that the emergency diesel generator (EDG) enclosure ventilation system

would not automatically open after the passage of a tornado. This condition would result

in the temperature within the EDG enclosure being higher than design and affect the

operability of the EDGs. This condition was corrected by revising procedures for restoring

the EDG ventilation. This LER is closed.

08.4 (Closed) VIO 50-423/95-42-05 Inadeauste Corrective Actions. Reactivity Transient

from Ooeration of the Chemical and Volume Control System

,

Adverse Condition Reoort (ACR) 05715. Review Reactor Power increase

a. Insoection Scooe (71707. 92901)

The inspector reviewed additional information on the licensee's corrective actions following

i a reactivity transient and resultant reactor power increase from operation of the chemical

-

and volume control system (CVCS). This event was addressed during two prior NRC

inspections (Reference inspection Report No. 50-423/95-42 and 50-423/96-05).

b. Observations and Findinas

Additional actions have been taken by the licensee, and the inspector has reviewed

additional information concerning this event. The licensee has included appropriate

information on boron dilution ents that address industry experience within their licensed

operator training program. Or. rice flow plates associated with flow indicator 3CHS-Fl136

have been installed correctly. The licensee has included improvements to OP 3304A that

provide additional barriers intended to prevent an inadvertent boron dilution.

Aaalysis of the operation of the high temperature divert valve (3CHS"TCV129) and its

actuator, failed to identify a method for bypassing the demineralizers. The inspector

confirmed that the valve position is provided by valve stem operated position switches

which would readily provide control board indication of the valve repositioning, which

complies with the FSAR failure modes and effect analysis.

.

.

48

c. Conclusions

The licensee has appropriately addressed the issues identified from review of the

November 1995 boron dilution event. Their corrective actions appear to include the issues

associated with the 1993 and 1994 reactivity transient. This item is closed.

08.5 LQlqqqd) VIO 50-423/95-42-05 Inadeauste Corrective Actions. Reactor Coolant

Pumo Seal Water iniection Filter Housina Gasket Failure

Adverse Condition Report (ACR) 04199. Seal Iniection Filter Gasket Failure

a. Insoection Scoce (71707. 92901)

The inspector reviewed the licensee's corrective actions following a failure of a reactor

coolant pump (RCP) seal water injection filter housing 0-ring gasket. This event was

addressed during two prior NRC inspections (Reference inspection Report No. 50-423/95-

42 and 50-423/96-05).

b. Observations and Findinas

The licensee has reviewed the past performance problems concerning failure of the RCP

seal water injection filter housing 0-ring gasket with the vendor. Based on information

obtained from the vendor, a standard notation will be added to work orders that specify

lubrication of the gasket and the correct torque value for the housing cover. Information

on appropriate O-ring gasket lubrication material that is suitable for RCP sealinjection

environment is pending from Westinghouse.

c. Conclusions

The licensee has appropriately addressed the issues identified from their review of the

previous failures of the RCP seal water injection filter housing O-ring gasket. This item is

closed.

U3.Il Maintenance

U3 M8 Miscellaneous Maintenance issues

M 8.1 (Closed) LER 50-423/96-21.

(Onen) URI 423/9.6-08-20 Inadeauate IST Proaram Controls

a. Insoection Scoce (61726)

The licensee's review of the Inservice Test (IST) Program identified numerous

programmatic deficiencies which were documented in LER 423/96-21. The review was

performed as a part of the 10 CFR 50.54 effort to verify the plant would be operated

within the design and licensing basis. The inspector reviewed the circumstances leading to

to LER and the subsequent corrective actions.

._. . .

.

.

49

b. Observations and Findinas

The licensee's review of the IST program identified multiple deficiencies including:

incomplete implementation of licensing commitments, omission of valves from the

program, testing inadequacies for valves in the program, incomplete or missing

dccumentation in the IST manual and IST surveillance procedures and inadequate process

controls procedures. The adverse condition report (ACR) which documented these

numerous IST programmatic discrepancies was classified as a level "D" ACR, the lowest

possible level. However, following discussions with the inspector, the licensee recognized

that the ACR was improperly classified and upgraded the issue to a level "C" ACR which

requires that a causal factor analysis be performed. The consequence of this misclassified

ACR could have been an inadequate level of management attention and the issue closeout

would not have been subject to a corrective actions review by the management review

team.

The licensee did not perform a root cause evaluation following the determination that this

programmatic issue was reportable nor when the ACR level was upgraded. The cause of

the event was attributed to a lack of resources required to implement and maintain an

effective program; however, specific causes were not identified. Although corrective

actions were developed to address the symptoms of the problem, no corrective actions

were targeted for the cause of the breakdown nor why the issues were not identified by

, the licensee oversight processes. Based on interviews, the inspector determined that a

lack of understanding of all component safety functions was a contributing factor and may

be applicable to other programs. Following discussions with the inspector, the licensee

determined that a root cause evaluation was necessary and committed to perform the

evaluation by January 31,1997. Further, at the conclusion of the inspection period, the

licensee also planned to revise the ACR procedure such that issues determined to require

i

an LER would require a "B" level ACR, and would typically result in a root cause

evaluation.

None of the components identified in the LER that were not properly tested had been

'

tested by the end of the inspection period. In some cases, plant modifications are required

to allow the appropriate testing. Further, based on interviews with the cognizant engineer,

,

the inspector determined that the LER did not provide a complete list of components that

were not properly tested. The licensee did not plan on supplementing the LER with the

complete list of discrepant tests, the results of the valve testing or specific corrective

actions. The licensee subsequently committed to supplement the LER with a complete list

of discrepancies and specific corrective actions by January 31,1997. In addition, the

results of the valve testing will be provided in a supplemental LER prior to plant startup.

The inspector determined that an operability determination (OD) had not been performed

for the equipment currently credited as operable. For example, several check valves and

excess flow check valves in the emergency diesel air start system had not been properly

tested and were not specifically assessed for operability. The inadequate testing of these

valves was previously identified in ACR M3-96-0063, written June 7,1996 and ACR M3-

96-0285, written June 28,1996. In both of these cases the shift manager determined

that these valves, along with numerous other valves, were operable during the initial ACR

screening. Consequently, ACR M3-96-1054 was initiated to address this missed

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

,

.

.

) 50

operability determination, and the licensee assessed the emergency diesel air start system

i valves and found all valves to be operable. The inspector reviewed the operability

determination and found that much of the discussion in the " bases for maintaining

operability" did not provide a bases for operability and some of the assertions were

i

unsupported. For example, as a bases for the excess flow check valve operability the OD

stated " valve leakage from a restricted 0.5 inch line will not prevent the diesel generator

from starting in the required time frame." However, when asked, the licensee was uncble

to support the accident sequence assumptions used to draw the OD conclusion. The

,

licensee subsequently revised the OD and initiated another ACR M3-96-1105, to address

j the inappropriate operability determination. In addition, the licensee tested the excess flow

check valves and found that they were operable.

d

-

The inspector determined that the valves identified as being tested in only one direction

were not listed in the initial program indicating there was a period of time in which neither

'

the open or close safety function was tested. The licensee plans to evaluate why the

valves were added subsequent to the initial program approval and determine whether

additional reporting is necessary.

c. Conclusions

The mis-classified ACR, the failure to perform a root cause evaluation of the programmatic

issues, and the failure to establish comprehensive corrective actions indicates continued

problems with the licensee's corrective actions process and oversight of that process. This

issue is unresolved (423/96-0818) pending the completion of all required testing,

performance of root cause evaluation, and the implementation of comprehensive corrective

actions. In addition, a regulatory disposition of LER 423/96-24 will be performed when

this unresolved item is addressed due to the similarity of the issues. The failure to include

all discrepancies in the initial LER indicates a weakness in assessing reportable issues and

development of appropriate LERs.

The licensee failed to appropriately assess the operability of the diesel air start system

valves on two separate occasions. Subsequently, when operability was assessed the OD

did not provide an adequate bases to support the conclusion. Although the valves were

ultimately found to be operable, this issue demonstrated weaknesses in the assessment of

equipment operability.

M8.2 (Closed) LER 50-423/96-24. Inadeauate IST Testino

a. Insoection Scooe (61726)

LER 423/96-24 documented that the main steam supply valve to the turbine driven

auxiliary feedwater pump (3 MSS *MOV17D) had not been disassembled and visually

inspected during the cycle 5 refueling outage as required by the inservice test (IST)

program. The inspector reviewed the circumstances leading to the LER ano the

subsequent corrective actions.

_. _ _ .. . . ._ _. _ _ _ __ _ _ . _ _

e

j

.

51

b. Observations and Findinos ,

Valve 3 MSS *MOV17D is one of three main steam supply valves to the turbine driven

!

auxiliary feedwater pump. One of the functions of this motor operated stop check valve is

,

to close in the event of a main steam line break upstream of the valve, to prevent the

diversion of steam. The purpose of the visual inspection program is to verify that the valve

can perform the isolation function. The licensee attributed the missed IST inspection to

inadequate program controls and an inadequate program review as a result of personnel

error. The licensee used an informal computer generated spreadsheet to schedule all check

,

valve inspections. Valve 3 MSS *MOV17D was scheduled for inspection during the cycle 5

refueling outage; however, the inspection was coupled with a work order to repair the

valve actuator which was subsequently canceled. The work order did not identify that the

valve inspection was part of the IST program. The management review of this canceled

maintenance item, prior to startup, also failed to identify the required valve inspection as a

result of the coding error.

As corrective action, the licensee performed the required IST inspection and identified

minor damage to the valve seating surface. However, the licensee concluded the function

of the check valve to close and prevent the diversion of main steam flow pressure was not

affected. A review of other IST program required check valve inspections revealed no

others were missed. The licensee plans to make programmatic enhancements to the check

valve program based on the results of the configuration management process review.

However, at the end of the inspection period no specific enhancements had been

identified.

c. Conclusions

Technical Specification 4.0.5 requires that in service testing of ASME Class 1,2,3 pumps

and valves shall be performed in accordance with Section XI of the ASME Boiler and

Pressure Vessel Code. The licensee's IST program which implements the requirements of

ASME section XI and required that valves 3 MSS *MOV17A, B, & D be partially

disassembled, inspected, and manually exercised on a staggered sampling basis each

refueling outage. However, the licensee failed to inspect 3 MSS *MOV17D, the valve

scheduled for the cycle 5 refueling outage. This is a violation of the IST requirements

since none of the valves within the specified group were inspected during the refueling

outage. Although the valve subsequently passed the inspection and no other check valve

inspections were missed, a regulatory disposition of this issue will be made with

unresolved item (423/96-08-18) due to the similarity of the issues and the appearance of a

common root cause related to inadequate program controls. The failure to identify

corrective actions in more than three months following the identification of this issue

indicates continued problems with the licensee's corrective actions process and oversight

of that process.

<

.

- _ _ _

e

.

52

M8.3 (Closed) Violation 423/95-38-01:

a. Insoection Scooe (92902)

Violations 336/95-38-01 and 423/95-38-01 involved instances where Unit 2 and Unit 3

failed to implement timely corrective actions to address weaknesses in the licensee's

surveillance tracking program. There had been severalinstances at Units 2 and 3 where

the licensee failed to perform the required technical specification (TS) surveillance within

the required time interval.

Follow-up inspection of these issues in NRC Inspection Report 50-423/96-04 revealed that

corrective actions were delayed resulting in recurrent problems and incomplete

implementation of licensee commitments. In addition, some actions planned to be taken by

the licensee to correct these discrepancies were contrary to those stated in the response to

the violations. A request for additional information was requested by the NRC on actions

the licensee planed to take to resolve these discrepancies. The inspector reviewed the

licensee's response and the implementation of the corrective actions,

b. Observations and Findinas

As action to prevent recurrence, the licensee indicted that formal procedural guidance had ,

been developed that defined roles and responsibilities of personnel scheduling and tracking l

surveillances. The licensee assigned a work planning surveillance planner the responsibility  !

to track and schedule all none conditional surveillances with a frequency of greater than 24 )

hours. As a backup, each department was responsible to track those surveillances under '

their control using an independent tracking mechanism.

The inspector reviewed licensee procedure WC-9, " Station Surveillance Program," and

verified that it captured these requirements. The inspector also verified that a work

planning surveillance planner was assigned for Unit 3, and that TS surveillances were being

tracked, using an independent tracking mechanism, by the responsible Unit 3 departments.

Additional corrective actions required that control room operators initiate a timer / alarm

feature and log conditional surveillances which are not captured by procedural controls in

the shift mangers log. In addition, the operator training re-qualification program was to be

updated to include scenarios that evaluate the operator's ability to apply TS in various ,

operational situations.  !

The inspector reviewed licensee procedure 3-OPS-10.2 and verified that it required the use

of a timer for conditional surveillances, and that shift managers have, in practice, been

logging conditional surveillances in the shift managers log. The inspector also reviewed

selected Unit 3 simulator exercises and verified that they included scenarios that test the

operator's ability to comply with conditional surveillances dunng simulated plant transients.

The inspector reviewed Unit 3 adverse condition reports (ACRs) generated since June

1996 to determine the effectiveness of these corrective actions. The review identified six

issues related to surveillance discrepancies. A review of these issues revealed that one

issue was attributed to inadequate scheduling of surveillances. ACR M3-96-0396 was

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

'

o

.

-

53

i

generated on July 12,1996, and documented a historicalissue regarding a missed in-

j service test surveillance (reference licensee event report 423/96-24). The other

'

surveillance discrepancies dealt with either: a program deficiency (a localleak rate test of

! the fuel transfer bellows was not performed, reference LER 50-423/96-23), inadequate

procedures which implement conditional surveillances, and a personnel performance issue.

'

The corrective actions taken to address the surveillance scheduling concern would not

i have prevented these specific issues. The licensee generated a level 'B' ACR to determine

l and address the root cause of these discrepancies.

j C. Conclusions

The NRC determined that the corrective actions were adequate to prevent the occurrence

of missed surveillances due to scheduling concerns. Therefore this issue is closed. j

However, the licensee still needs to address the programmatic concerns with the IST j

program and the procedural discrepancies that resulted in missed conditional surveillances. '

Closure of this item does not address the overall effectiveness and timeliness of the

licensee's corrective action program. This latter issue must be demonstrated to be

effective by the license prior to start-up of the unit and is included in the licensee's Nuclear

Excellence Plan.

U3.Ill Enaineerina

U3 E2 Engineering Support of Facilities and Equipment

E2.1 (Closed) LER 50-423/96-31 - Potential Failure of Solenoid-Ooerated Valves (SOVs)

due to Overoressurization l

a. Insoection Scoce (37551) .

On September 6,1996, the licensee reported that the potential exists for 37 ASCO l

solenoid valves to fail to operate properly and complete their intended safety function if

subjected to full air system pressure. Specifically, the potential exists for non-qualified air

pressure r6gulators, located upstream of each of the affected SOVs, to fail and result in the

SOVs experiencing full air system pressure (110 psig). This pressure is higher than the

rated maximum operational pressure differential (MOPD) of 60 psig for each of the

identified SOVs. This issue was identified during a review of NRC Generic Letter (GL) 91-

15, " Operating Experience Feedback Report - Solenoid-Operated Valve Problems at U.S.

Reactors," as part of the licensee's corrective actions concerning the potential common-

mode failure of the Target Rock SOVs used in the auxiliary feedwater system (reference-

NRC Inspection Report 50-423/96-201).

The solenoid MOPD issue was identified in NRC information Notice (IN) 88-24, " Failure of

Air-Operated Valves Affecting Safety-Related Systems," and GL 91-15. The inspector

requested the background documentation to evaluate the licensee's previous corrective

actions in response to these NRC correspondences, and to track the final resolution of this

issue.

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

4

. l

54

i

b. Observations and Findinas i

Review of NRC Correspondence

in response to IN 88-24, the licensee identified that there were approximately 20 SOVs

installed in safety-related configurations that had the potential to fail because of

overpressurization. However, the licensee did not consider this to be an immediate safety

issue since the SOVs and regulators were believed to have been purchased as a qualified

unit, and there were no other common-mode failure scenarios identified in the IN which

t ould initiate regulator failure, in addition, since the SOVs are checked for operability on a

regular basis and no problems had been identified, replacement of the SOVs on an

emergency basis was not planned. Since these solenoids were reaching the end of their

environmental equipment qualification (EEO) life, the licensee opened commitment number

3-88-0030 to replace the solenoids with valves of a higher MOPD rating by the end of the

third refueling outage (RFO3). As part of this commitment, the licensee also planned to

explore other possible regulator failure modes to determine if any could result in a failure of

the SOVs to perform their intended safety function. The commitment was closed on June

17,1996.

Generic letter 91-15 also identified the potential SOV failure mechanism because of MOPD.

The licensee had received NRC GL 91-15, but apparently had not reviewed it for

applicability. A review of the controlled routing revealed that it had been distributed for

information only. The generic letter did not require any specific action or response from

the licensee; however, it was expected that the licensee would review the information for

applicability and consider actions, as appropriate, to' avoid similar problems.

The inspector noted that similar problems with the licensee's review of industry

experience, specifically vendor information and NRC information notices (IN), had been

identified by the NRC in 1993, but that this has not recurred as a recent concern.

Corrective actions for these issues included establishing procedure controls such that

vendor information and NRC IN be tracked and evaluated by the appropriate plant

department for concurrence and resolution. However, since NRC generic letters and

bulletins usually require a response from the licensee, no procedural changes were taken.

A review of the licensee's Regulatory Compliance Manual (RCM), Chapter 2, " incoming

Correspondence," revealed a potential problem with the disposition of GLs and bulletins.

All correspondence from regulatory agencies are reviewed by the licensing department to

determine if NU action is required; however, only IN are required to be forwarded to the

Nuclear Safety Engineering Group for review and evaluation. Therefore, the inspector

concluded that if a GL or bulletin is received that does not require a response, a review for

applicability and possible action may not be performed.

In response to this concern, the licensee issued an adverse condition report (ACR) to revise

the RCM and to review Gls that did not require a response to the NRC retrospectively for a

period of 15 years. As interim corrective action, the licensing department generated an

action tracking item to ensure that all regulatory correspondences are reviewed for

applicability.

.

55

Qualification of Air Pressure Reaulators

The inspector requested the licensee to provide the list of 20 SOVs identified to be

replaced as part of commitment 3-88-0030 to determine whether the 37 valves currently

reported included those identified in 1988, and verify that the regulators and SOVs were

purchased as qualified units. The 'icensee was unable tc, identify any documentation listing

the 20 SOVs or identify, from a review of maintenance records, any SOVs that had been

replaced because of the potential overpressurization concern. The licensee had replaced,

during RFO3, 30 ASCO SOVs due to the solenoids reaching the end of their EEQ life, but

only two had MOPD ratings of 60 psig. These two were replaced with solenoids that had

the same MOPD rating. An evaluation had been performed by the licensee that extended

the EEQ life for the other ASCO SOVs. The inspector reviewed commitment 3-88-0030

and noted that it only addressed replacement of the solenoids due to EEO concerns. The

licensee's review of the planned maintenance management system (PMMS) bill of material

and material equipment parts list (MEPL) database revealed that a number of the regulators

and SOVs were not classified as qualified components, and that several safety-relate

regulators had been downgraded via MEPL evaluations. The licensee plans to review the

purchase requisitions and work history for these components to determine if the

solenoids / regulators were purchased or replaced with non-qualified components.

i

Because of the discrepancy in the number of SOVs that could potentially be affected if

subjected to full air system pressure, the licensee generated a list of all safety-related i

SOVs from the PMMS database and performed a field walkdown to obtain the

manufacturers nameplate data. The field walkdown revealed that there are 48 safety- j

related SOVs with MOPD ratings of 60 psig. The list of SOVs originally reported on l

September 6,1996, was generated from the plant document design system computer l

database. This database is not accurate and is in conflict with the PMMS database. The

licensee is reviewing the original design specification for the SOVs and regulators to

determine how they were procured, tested, and certified. The inspector requested that the

licensee review the other Northeast Utility unit's database to confirm that the numbers of

ASCO SOVs originally identified in 1988 was correct. An ACR was written to address this

concern.

Of the 48 SOVs identified, no immediate corrective actions were required to be taken since

the components affected were either not required for operation in the current plant

condition (mode 5), or compensatory measures had previously been taken. As corrective

action for this issue, the licensee committed to maintain the plant in Mode 5 or less until

modifications are complete to correct the potential overpressure condition on all of the

susceptible SOVs.

c. Conclusion

The NRC determined that the licensee's review and followup of NRC correspondence was

inadequate. A detailed review for potential components affected and of purchase records

was not performed to determine the full scope of the problem. In addition, a licensee

commitment was inadvertently closed out without replacing the subject SOVs or

performing a review to determine potential regulator failure mechanisms. As a result, the

licensee incorrectly determined that the ASCO SOVs installed in the plant were not

.

.

56

susceptible to the MOPD phenomena since the regulators were believed to be quality

components. if non-safety parts have been installed in the plant or components

downgraded, then the safety function of the component may be impacted. Compensatory

measures have been taken for the present plant condition. This issue is considered to be

an unresolved item (URI 423/96-08-19).

U3 E8 Miscellaneous Engineering Issues

E8.1 [CJhlsed) URI 423/96-05-14: Soent Fuel Pool (SFP) Decav Heat Removal Analvsis

and Final Safety Analvsis Reoort (FSAR) Update

This issue involved an inspector question whether the FSAR update, submitted subsequent

to the analysis associated with Ameridment 60 to the Unit 3 license, contained all the

information relevant to the affected SFP cooling safety evaluation. By letter dated July 8,

1996, the NRC requested an explanation of the licensee position relative to the need for an

update to Table 9.1-2 of the FSAR, which contains pertinent data on the SFP cooling

analysis. The licensee responded by letter on August 9,1996. During this inspection, the

inspector confirmed that the licensee intends to submit a discussion of a normal full core

offload of fuel, addressing all of the variables documented in Table 9.1-2,in the FSAR

update on this subject. Further, the inspector verified that the licensee's analysis

supporting the new data is to be based upon the assumption of a single failure. Since the

licensee has committed to the FSAR revisions that address the original questions and

concerns raised by the inspector, this unresolved item is considered closed.

E8.2 (Closed) LER 50-423/96-12-01: Containment Local Leak Rate Test Results in Excess

of Technical Soecification Limit

a. Insoection Scope (92700, 61720)

The inspector reviewed the results of the containment penetration local leak rate tests, the

results of which were presented in Licensee Event Reports (LER) 96-012-00 and -01. The

combined leakage rate for penetrations subject to Type C testing exceeded Technical

Specification 3.6.1.2.b limit of 0.6 La and the secondary containment bypass leakage

paths exceeded Technical Specification 3.6.1.2.c limit of 0.042 La. The inspector

reviewed the licensee's corrective actions and reviewed the test results for trends,

b. Observations and Findinos

The licensee found that through tests conducted during the current outage, the combined

leakage rate measured at containment penetrations exceeded Technical Specification

4.6.1.1.c limit of La. Although, the licensee reported a combined leakage rate of

498,959.1 sccm, valves at both the purge supply (penetration number 86) and low

pressure injection (penetration 93) were found leaking beyond their capability to measure

at Pa of 53.27 psia (38.57 psig). A six inch low pressure safety injection inboard isolation

check valve,3SIL*V6 was also found with excessive leakage that exceeded their ability to

measure.

_ - - - . _ -

.

.

57

The forty-two inch containment purge supply penetration consists of three valves,

3HVU*CTV33A, 3HVU*CTV32A, and a thirty inch branch line valve, 3HVU*V5. The

licensee reported that the leakage from this penetration alone exceeded both Technical

Specification 3.6.1.2.b Type C limit of 0.6 La and Technical Specification 3.6.1.2.c

secondary containment bypass leakage limit. Type C refers to tests required by 10 CFR

Part 50, Appendix J, paragraph II.H.

The licensee's root cause investigation (Reference: ACR 10430) focuses on possible

inadequacies in the purge valve "T" ring adjustment and the associated hardware;

however, there were no actions identified that provide assurance that this problem will be

corrected for the future. Prior to failing on May 29,1996, penetration 86 failed localleak

rate tests on April 17,1995, and November 23,1987. The licensee points out, in the root

cause analysis, that because the valve is cycled open and closed before being tested, the

test results may not reflect the its condition during the operating cycle. However, since

the purge valves were opened following a successful test on May 13,1995, the valves

may not have been leak tight during the last operating cycle. The licensee will test the

containment purge penetration at the beginning of an outage, prior to opening the valves to

obtain as-found data; likewise, as-left test data will be obtained after the valves are closed,

prior to entry into mode four.

Of the other penetrations reported with high leakage rates, six involved check valves. In

most cases the leak rates were reduced following a flush with demineralized water to

remove boron deposits or other foreign material from the valves. Although, the licensee

has proposed routine flushing of some penetrations, it is not clear that they have taken an

integrated approach to improve performance of the containment system. For example, in

addition to the successive failure of the purge supply penetration valve, the present as-left

test results for quench spray valve 3OSS*V4 (penetration number 100), safety injection

valves 3SIL*V13 and 3SlH'V24 (penetration number 94), and low pressure injection valve

3SIL*V6 (penetration number 93) are greater than the as-left results from the last refueling

outage.

All valves were repaired and retested successfully, the "T" ring seat of the purge valve

was readjusted, and the check valves were flushed with demineralized water. Several

check valves were disassembled and cleaned. The overall as-left leakage for all Type C

penetrations is 73,356.5 sccm; and 4,424.3 sccm for those that bypass the secondary

,

containment. In both cases these totals are higher than the as-left leakage rates from the

last refueling outage.

In reporting containment penetration leakage rates, the licensee includes the purge supply

(penetration number 86) with those which bypass the secondary containment. Technical

Specification 3.6.1.2.c limits this bypass leakage to 0.042 La. However, the purge line

leakage was not included with the combined Type C leakage. Technical Specification

3.6.1.2.b limits the combined leakage rate for all penetrations and valves subject to Type B

and C tests to less than 0.6 La when pressurized to Pa. Although the licensee included

purge valve leakage with the more restrictive bypass leakage limit, its leak rate also

contributes to the combined result of the Type B and C tests.

a

.

58

c. Conclusions

Two penetrations were found with leakage rates in excess of that which could be

measured at Pa,53.27 psia or 38.57 psig. One of these two is the forty-two inch

containment purge supply penetration, leakage through which also bypasses the secondary

containment. This is the second consecutive failure of that purge supply penetration. The

other found with excessive leakage was the low pressure injection penetration where a

check valve was found to have foreign material that prevented the disk from closing fully.

This was one of six check valves found leaking excessively because of boron deposits or

other foreign material fouling the seating surface.

The licensee has proposed changes to the purge valve maintenance procedure and the test

schedule. Also, all the subject leaking valves were repaired and retested, with the

resulting total Type C valve leakage found to be well below the technical specification

limit. The licensee has also taken additional corrective measures that appear to address

the concern for the "as-found" test results from outage to outage. The licensee's

hardware repairs and changes in testing methodology are directed toward preventing a

recurrence of such problems in the future. However, the inspector noted that such

corrective action enhancements were necessary because of both the higher, although

acceptable, current "as-left" leakage rates and the fact that past corrective measures had

not been effective in precluding the problems that were reported in this LER. Licensee

commitments for future valve inspections, as necessary, and continued analyses of the

valve leakage criteria and test result trends appear prudent.

E8.3 (Closed) LER 50-423/96-23: Failure to Perform Containment Local Leak Rate Tests

of the Fuel Transfer Tube Exoansion Bellows

a. Insoection Scooe (92700,61720)

The licensee discovered their failure to include a leakage test of the fuel transfer tube

expansion bellows within the containment test program. The inspector reviewed their

analysis of the event along with the corrective actions.

b. Observations and Findinas

The licensee performed an independent assessment of the containment system and

discovered that the fuel transfer tube expansion bellows (containment penetration number

88) had not been tested as a Type B local leak rate test since initial startup of the unit.

Technical Specification 4.6.1.2.d requires that Type B and C tests be conducted at

intervals no greater than 24 months. The Final Safety Analysis Report (FSAR), Section

6.2.6.2, specifies that all containment piping penetrations fitted with expansion bellows

require Type B testing in accordance with 10 CFR Part 50, Appendix J, Paragraph II.G.

The licensee's assessment concluded that all other Type B and C penetrations are included

in their surveillance procedures ano are tested.

As corrective action, the licensee indicated that penetration 88 would be added to their

Appendix J program, the associated surveillance procedure revised, and the penetration j

tested prior to unit restart. In addition, the licensee plans to revise the FSAR to clarify the '

!

i

I

.

.

59

requirements associated with the Appendix B testing of the fuel transfer bellows. The

inspector verified that the surveillance procedure and FSAR were in the process of being

changed, and that testing of the penetration had been scheduled to be tested prior to unit

restart.

Although the licensee concluded, in their analysis of this event, that since the results of

containment integrated leak rate tests were successful, the integrity of the expansion

bellows was verified. The inspector noted that the fuel transfer tube penetration consists

of a tube, the actual transfer tube, which is within an outer tube that is attached to the

containment liner. The double tube assembly is sealed at either end with expansion

beilows. There is a third expansion bellows assembly, approximately midway between the

two ends, to allow movement relative to the containment structure and the fuel storage

pool structure. The containment integrated leak rate test does not test the overall integrity

of this expansion bellows system.

c. Conclusions

Failure to localleak rate test the fuel transfer containment penetration expansion bellows

assembly is a violation of Technical Specification 4.6.1.2.d. However, this licensee

identified and corrected violation is being treated as a Non-Cited Violation, consistent with

Section Vll.B.1 of the NRC Enforcement Policy. This LER is closed.

E8.4 (Closed) URI 50-423/94-32-02: Comoonent Coolina Water Svstem Temoerature

Limitations: (Closed) LER 50-423/96-13-01: (Ooen) 50-423/96-08-22

a. Insoection Scooe

The reactor plant component cooling water (CCP) system is currently designed with a

maximum operating temperature of 115 F and maximum stress analyzed operating limits

for portions of the system piping set at 125 F. Previous NRC inspection reports (50-

423/94-32 and 95-20) documented an unresolved item regarding licensee evaluations that

determined that the system would be subjected to higher temperatures, particularly when

subjected to safety grade cold shutdown (SGCS) scenarios. Further, in June,1996, the

licensee determined that a design deficiency in the residual heat removal system (RHS)

could result in CCP temperatures above the 125 F used in the system stress analysis.

This was reported by the licensee in accordance with 10 CFR 50.73 and supplemented on

August 19,1996 in LER 96-13-01. Additionally, a NRC Special Team inspection (IR 50-

423/96-201) identified cases where the CCP piping temperatures had exceeded the 115

degree operating limit in apparent violation of operating procedures (eel 50-423/96-201-

22). During this current inspection, the inspector reviewed licensee engineering activities

to resolve the technical concerns associated with the above issues and findings and

assessed ongoing plant design modification activities.

b. Observations and Findinas

The inspector reviewed past licensee engineering activities (1995) to analyze the subject

CCP piping to a temperature of 125 degrees F, with an associated 75 degree F service

water temperature limit. Assuming a maximum operating temperature of 115 degrees F,

.

o

.

60

the licensee's analysis determined that short duration transients above this limit would be

acceptable. With the recent (1996) identification of the RHS failure modes that could

significantly raise the CCP temperatures, additional engineering work was required and

design modifications were determined to be necessary. Activities in support of two design

change record (DCR) modifications are currently in progress, as follows:

e DCR M3-96075 - structural modifications to the CCP and service water systems to

support hi0her analyzed temperatures.

e DCR M3-96065 - hardware and instrumentation changes to address the RHS flow

control issue, given the single failure considerations.

The intent of the current licensee design activities is to qualify the CCP system for the

higher limiting temperatures, while implementing system modifications to ensure that the

design is consistent with the expected limits. This would also restore the affected system

to compliance with design basis criteria, with regard to SGCS considerations, for unit

restart. In reviewing the licensee's plan and design intent, the inspector raised specific

questions to the cognizant design engineer regarding the service water flow assumptions

used in the analysis, how spent fuel pool cooling is addressed as a CCP heat load, and the

failure modes evaluation of the RHS heat exchanger flow and bypass flow control valves.

Based upon the licensee response to these questions, the inspector determined that the

current engineering activities are proceeding in a direction which appears likely to resolve

the technical concerns.

However, the inspector also noted that the SGCS concerns, relative to CCP temperature

design restrictions, have documented origins in correspondence dating back to 1985. The

fact that this issue only now appears to be headed to resolution is a matter that merits

licensee management attention, particularly as design basis review activities continue with

Unit 3 remaining in cold shutdown status. The inspector determined that additional

inspection of the DCR implementation progress, as well as further assessment of overall

licensee corrective measures in this area, are necessary to satisfactorily resolve the

technical concerns associated with this longstanding issue.

c. Conclusions

Based upon the licensee's actions to address the historical aspects of the open questions

on CCP temperature limitations, open items, URI 50-423/94-32-02 and LER 50-423/96-13-

01, are considered closed. Continued licensee progress in the implementation of the

planned DCRs and additional corrective actions, as required, are considered to be

unresolved (URI 50-423/96-08-20) pending further NRC review.

-_. _ -.

.

e

61

IV Plant Support

(Common to Unit 1, Unit 2, and Unit 3)

R8 Miscellaneous Radiological Protection and Chemistry issues

R 8.1 Material Condition Proaram (MCP)

a. Insoection Scone (71750)

)

Adverse condition report (ACR) 10172 was initiated on May 2,1996, by a member of the  !

Unit 2 quality assessment service department, to document that the status of the MCP j

was indeterminate. Specifically, recent plant walkdowns had not met the MCP manual

requirements and the site MCP was discontinued without replacement.

The site MCP was developed by the licensee to identify both major and minor deficiencies

like housekeeping and equipment deficiencies for tracking end trending to facilitate

improvements in the material condition of the plants. The identification of an item in this I

program did not relieve the individual of their responsibility to document it in one of the l

other types of tracking mechanisms. Therefore, the more significant MCP findings should i

have been documented and tracked for resolution by those tracking mechanisms. l

However, there were significant plant issues that were only being tracked by the site MCP '

(i.e. Unit 1 liquid radwaste issue; reference NRC Inspection Report 245/96-03). The MCP

is also part of the licensee's improving Station Performance (ISP) program that was

developed in 1995 to address weakness in various licensee programs. The inspector l

reviewed the overall status of the MCP, and the specific Unit 3 MCP issues to determine if i

any of the items needed to be addressed prior to Unit 3 startup.

b. Observations and Findinas

The site MCP was terminated in January 1996 after the person responsible for managing

this program left the employment of Northeast Utilities, and no replacement was identified

to continue these duties. Licensee management had decided that the program would be

discontinued as a site wide program, with responsibility moving to the specific unit line '

function so that the individual Unit Directors could have more control. However, no unit

specific programs were implemented.

Unit 3 health physics (HP) department had a database which documented HP type

discrepancies, but not other department issues it wasn't until the June / July time frame

that the HP database was used to document other types of plant issues. However, there

was na procedure / instruction which documented that this process was used to satisfy the

Unit 3 MCP. As part of the planned corrective actions for ACR 10172, each unit was to

develop, by October 1996, an individual unit program and transfer the identified site MCP

issues into the new program. Unit instruction 3-Ul-2.01, " Material Condition," was

developed and implemented on September 4,1996, to satisfy this commitment for Unit 3.

The inspector obtained the site MCP database and noted that there were a total of 2613

items identified. Of this total,43% were indicated as being closed out. Those items

remaining in an open status assigned to Unit 3 totaled approximately 400. The inspector

.

.

62

verified that Unit 3 developed a unit instruction and that the items listed in the site MCP

were incorporated into this new database. A review of the Unit 3 instruction revealed that

it specified that if an adverse condition could not be immediately corrected, then a trouble

report or work request would t,e generated and the condition entered into the material

condition database.

Discussions with the licensee revealed that the site or Unit 3 database had not been

reviewed against the Unit 3 startup criteria. A review was subsequently performed by the

licensee that determined no restart items were identified. The inspector reviewed the

specific Unit 3 discrepancies listed on the site MCP for items that could potentially affe;

plant / equipment operation. Most items were general cleanliness and painting issues that

would not be required to be corrected prior to the unit restart. The inspector discussed a

number of items with the licensee that appeared to be the most safety significant and

walked down severalitems to ensure they were properly dispositioned. These included:

o Piping on the north side of the condensate storage tank (CST) was not insulated.

This piping was part of the nitrogen supply to the CST and therefore insulation was

not required for freeze protection.

e Piping and supports in the 24'6" level of the auxiliary room were listed as needing

painting. Inspection of this area revealed that some of the piping and supports had

been painted. The piping and supports that had not been painted revealed no

indication of pipe boundary or support degradation.

e Service water valves 3SWP'V674 and 'V673 were indicated to be badly corroded.

This was surface rust / oxidation on bronze valves and therefore does not affect the

pressure boundary. However, the nuts attaching the valves to the piping appeared

to be badly corroded. ACR 3-96-295 had been written to document this condition.

The licensee stated that the fasteners were not corroded to the point where the

structural integrity of the valve was affected. A work order was written to replace

the nuts.

c. Conclusion

A problem in the oversight and control of one aspect of the licensee's ISP program was

identified as evidenced by the fact that the site MCP was discontinued without

replacement. There was no comprehensive MCP in effect at any of the units for a period

of approximately nine months. The licensee's characterization of the Unit 3 issues

identified in the site MCP was appropriate as no Unit 3 startup items were identified. The

review of the Unit 1 and Unit 2 MCP is considered an item for further inspection followup.

(IFl 245/96-08-21 and 336/96-08 21).

R8.2 Radwaste and Transoortation

a. Insoection Scone (86750)

An inspection of the licensee's programs for the collection, treatment, processing and

transportation of solid radioactive waste and other radioactive materials in accordance with

'

.

.

63

NRC Inspection Procedure 86750, was conducted by a region-based specialist inspector

during this inspection period. Areas inspected included the licensee's program in quality

assurance as it supports the radwaste and transportation program.

b. Observations and Findinas

The licensee's program for the transportation of radioactive materials and free-release of

materials from the site is under the direction of the Waste Services Department. The

Department Manager reports through the Director, Nuclear Maintenance and Outage

Services to the Vice President - Work Services. On September 18,1996, the licensee

announced a major organizational restructuring. Although certain position / job titles were

changed, the basic organizational structure for the Waste Services Department remained

unchanged, and the senior management overseeing this department remained the same,

in support of the Waste Services Department is a site health physics group of six

technicians who perform all surveys related to shipping and the free-release program.

These technicians were part of the site Radiation Protection Department under the general

direction of the Unit 1 Radiation Protection Supervisor. Following the management

reorganization, at the time of this inspection, no decision as to the management oversight

of this group had been determined.

Liquid radwaste processing within each of the units was the responsibility of the respective

Operations Departments. At Unit 1, a recent change in the Operations Assistant

responsible for liquid radwaste operations occurred. The new Operations Assistant

formerly served as a senior station health physicist in charge of the Radwaste Remediation

Project. At Unit 2, the plant equipment operator (PEO) responsible for liquid radwaste

processing was scheduled to retire shortly, and was in the process of training a second

PEO to be the lead for Unit 2 liquid radwaste. At Unit 3, the PEO responsible for liquid

radwaste remained the same, as did the lead engineer for radwaste systems. Tours of the

radwaste facilities by the inspector, together with interviews of plant personnel and a

review of a recently concluded Electric Power Research Institute (EPRI) Unit 3 radwaste

study, indicated that the Unit continues to provide minimal management attention to the

liquid radwaste processing systems. Issues involving total suspended solid in the liquid

waste streams have repeatedly occurred over the past several months. While options to ,

deal with this issue have been developed by the PEO and lead engineer, minimal attention l

has been focused by management and the issue remains unresolved. This has led to a

J

noticeable increase in the amount of radwaste generated at the unit. l

i

The inspector reviewed nine radioactive waste shipments made in 1996, against the j

applicable criteria contained in Titles 10 and 49, Code of Federal Regulations (CFR), and l

the State of South Carolina issued site license to Chem Nuclear Systems, Inc. for the

operation of the Barnwell Low Level Radioactive Waste Management Facility. These

shipments were found to be well documented, and to meet all the applicable regulatory

requirements. All of the shipments were accepted at the Barnwell facility, without

comment, by the State of South Carolina.

The inspector also verified that the licensee ensures that each recipient of radioactive

,

materials is authorized to receive the type, form and quantity of material being sent.

_

.

.

64

Copies of receiving company's NRC or agreement state license is maintained by the Waste

Services Department and reviewed before each shipment of radioactive materials.

The licensee composited waste samples and plant smears for scaling factor analysis, and

sends these to an independent laboratory (ThermoNuTech) on an annual basis for Class B

and Class C waste streams and biennially for Class A waste streams. Additionally, in

accordance witti Procedure RW-46041, Rev 2, " Compliance with 10 CFR 61 - Waste

Classification," monthly plant primary coolant samples are analyzed and the results

provided to the Waste Services Department to determine if changes in plant conditions

necessitate the need to submit new samples for scaling factor analysis. Results of these

analyses are maintained by the Waste Services Department, and the scaling factors are

entered into various spreadsheets utilized in the preparation of shipping papers and

radwaste manifests. This portion of the licensee's program is conducted in accordance

with licensee procedure RW-46042, Rev 2, "Use and Application of Work Services

Computer Spreadsheets."

The inspector discussed, with cognizant licensee personnel, the programs established to

minimize the volume of radwaste generated by the licensee. The licensee has made

significant progress in the replacement of plastic and tape materials with reusable items to

significantly reduce the volume of waste generated. Additionally the licensee maintains a ,

" Green is Clean" free release program for non-contaminated materials being released from

the radiologically controlled areas (RCA). All dry active waste (DAW) is placed in either

radioactive or non-radioactive bags which are then sorted by the licensee. Radioactive

materials are then bulk loaded into SeaVans for further waste processing by a vendor.

Non-radioactive materials are surveyed twice prior to free release. Significant reductions at

all three units has occurred over the last 5 years in both total radwaste generated and in

the volume of radwaste disposed. Increases in radwaste generated at Units 1 and 3 in

1996 are the result of the extended shutdowns at these plants, and the wastes being

removed from the Unit 1 Liquid Radwaste facility as part of the Radwaste Remediation

Project.

The licensee's Quality Assurance (OA) Department provides support to the Waste

Services' radwaste and transportation program through the use of QA hold points

throughout the Waste Services' procedures and by conducting biennial audits of the

Process Control Program (PCP), as required by each unit's Technical Specifications. The

inspector reviewed the most recent PCP audit, dated April 16,1996 (Audit No. A24062)

and a Nuclear Utilities Procurement issues Council (NUPIC) audit of Chem Nuclear

Systems, Inc. (CNSI) (Audit No. 9401013) conducted in May 1994. The PCP audit team

included a technical specialist from another utility. No issues of safety significance were

identified in this audit. The NUPIC audit is utilized to verify the acceptability of CNSI as a

vendor of NRC certified shipping casks, in accordance with 10 CFR 71, Subpart H.

c. Conclusions

Solid radioactive waste controls are intended to minimize the volume of radioactive waste

and ensure safe transportation of radioactive materials. Based on aspects observed during

this inspection and discussed above, the inspector concluded that the licensee's program

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

.

.

65

was very effective in meeting these safety objectives. Contir ued problems in the area of

management oversight of liquid radwaste, especially at Unit 3, contique to exist, however.

!

R8.3 Miscellaneous Radioloaical & Chemistrv lssues '

A recent discovery of a licenste operating their facility in a manner contrary to the Updated

Final Safety Analysis Report (UFSAR) description highlighted the need for a special focused

review that compares plant pra:tices, procedures and/or parameters to the UFSAR  ;

descriptions. I

While performing the inspections discussed in this report, the inspector reviewed the

applicable portions of the UFSAR that related to the areas inspected. The inspector  ;

verified that the UFSAR wording was consistent with the observed plant practices, l

procedures and/or parameters, except in the area of Unit 1 liquid radwaste processing,

which has been previously discussed in NRC Inspection Reports 50-245/95-35, 95-38, 96-

03, and was the subject of a pre-decisional enforcement conference held on March 11,

1996.

S2 Status of Security Facilities and Equipment l

S 2.1 Devitalization of Protected Areas

l

The inspector reviewed the licensee's compensatory plan used to devitalize the "B"

emergency diesel generator area and verified it was in accordance with the licensee's

Physical Security Plan. Security procedure 5018-A, " Protected Area and Vital Area Access

and Egress Control," allows areas to be devitalized when requested by the shift manager

(SM) if the equipment in the room is out of service for maintenance or the equipment is not

required by technical specifications for accident mitigation. The inspector verified that the

SM had requested that the "B" diesel room be devitalized for maintenance activities, and

that the equipment was not required for the specific plant condition (mode 5). Once the

area was devitalized, signs were posted at the entrance to the room and this condition was

logged by the SM. Discussions with the operations department representatives responsible

for restoring the room to a vital area revealed that they were cognizant of the steps

necessary for revitalizing this area.

S8 Miscellaneous Security and Safeguards issues

S8.1 (Closed) LER 50-245/96-47: an unauthorized contractor was allowed protected area

access by a co-worker. This event was discussed in Inspection Report 50-245/96-06,

section U1.S1.1. No new issues were revealed by the LER.

r

.

e

l

66

P8 Miscellaneous EP issues

l

P8.1 Millstone Call-in Drill

a. Insoection Scope (82701)

Review the licensee's drill objectives and evaluation report of the September 29,1996,

Millstone Station call-in drill. Additionally, evaluate the requirements for staffing of the

station emergency response organization (SERO).

b. Observations and Findinas

The inspector conducted an in-office review of the licensee's drill objectives and evaluatic,n

report. The call-in drill for September 29,1996, had the following six objectives:

(1) demonstrate the capability to promptly notify station on-call response personnel of

emergency classifications; (2) demonstrate the capability to initiate and maintain

communications between appropriate emergency response personnel; (3) demonstrate the

capability to adequately brief additional personnel when utilized; (4) demonstrate the

capability to staff the station emergency response organization in accordance with staffing

requirements identified in the emergency plan and procedures (30 minutes responders

including dose assessment personnel, 60 minutes for staffing emergency facilities' by

primary responders and 120 minutes for support staff); (5) demonstrate the conduct of a

drill between the hours of 6:00 p.m. and 4:00 a.m.; and (6) demonstrate the use of the

common operating procedure for severe weather operations.

The drill was conducted from 6:00 p.m. to 8:30 p.m. on September 29,1996. According

to the licensee's drill evaluation report, the simulated alert was declared at 6:20 p.m. and j

the emergency response notification system (ERNS) was activated at 6:29 p.m. The SERO  ;

personnel were required to report to their assigned facilities, i.e., the Technical Support l

Center (TSC), Operational Support Center (OSC) and the Emergency Operation Facility

(EOF). The TSC/OSC was declared activated at 7:20 p.m. and the EOF was declared

activated at 7:27 p.m. These activations were within the NRC's 60-minute goal for l

staffing emergency response facilities after notification to emergency responders. l

The drill evaluation identified strengths in command and control exhibited by the Unit 3

shift manager, operations crew and on-shift director of station emergency operations, and

timely notifications by the shift technician. A fitness-for-duty issue was identified with

regard to one of the responders. It was appropriately handled by station security.  !

Additionally, several areas for improvement were identified for equipment, procedures,

communications, and drill control. The licensee's drill evaluation report indicated that all of

the drill comments and areas for improvement were placed in the emergericy preparedness

tracking system.

The inspector reviewed the applicable sections of emergency preparedness administrative

procedure (EPAP) 1.15, Revision 1, for Millstone Station and Administrative Control

Procedure (ACP) 1.0-6, Revision 15, for the Haddam Neck plant to determine if there were

adequate controls placed on the assignment of personnel to the SEROs. Both EPAP-1.15

and ACP 1.0-6 have detailed checklists to determine the qualifications of responders for

.

.

67

assignment. These checklists ensure that the assignee has completed the required

emergency plan and other necessary training, understands the requirements and

responsibilities of an SERO member, understands the fitness-for-duty requirements, and

lives within an area that enables the assignee to respond within the required time limit.

Additionally, the procedures require emergency response assignees to stay within an area

that allows response within the time limit when on call. The inspector also reviewed a

memorandum, dated July 21,1995, signed by responsible managers and sent to the SERO

staffs for the Millstone Station and the Haddam Neck Plant. The memorandum is explicit

in establishing the responsibilities and obligations for the on-call SERO staff.

c. Conclusion

Based on review of the licensee's evaluation report, the call-in drill met NRC goals and

requirements and the NRC-approved emergency plan commitments. The administrative

procedures for the Millstone Station and Haddam Neck Plant appear to be adequate for

ensuring that the established times for staffing the emergency response facilities are met.

F1 Control of Fire Protection Activities

F1.1 Proaram Review

a. Insoection Scope (64704)

The inspector reviewed program documentation established to provide the fire protection f

policy and strategy for protecting structures, systems, and components important to safety 1

and for personnel, procedures, and equipment required to implement the program. In I

addition, the inspector interviewed selected staff to assess their knowledge of certain fire ;

protection requirements, l

l

b. Observations and Findinos

Nuclear group procedure 2.14, Revision 9, " Nuclear Plant Fire Protection Program"

established Northeast Utilities fire protection program. The inspector found that this

program procedure failed to provide appropriate guidance and supporting details for

implementing and controlling the requirements of the fire protection program, as described

in NRC Branch Technical Position 9.5.1, " Fire Protection Program." Specifically, the lines

of communication pertaining to fire protection were not defined between various positions,

departments, or unit organizations; the responsibilities presented failed to reflect the

current organization; no single point of control and contact for prograrn implementation had

been established; and no identification was made of implementing procedures or processes

such as permit systems. In addition, the inspector noted that no design basis document

existed for the fire protection program that provided design and licensing requirements for

any of the three Millstone plants.

In contrast to the above program document, the inspector reviewed a recently completed

Millstone station procedure, Work Control (WC) 7, Revision 1, " Fire Protection Program."

WC7, although not referenced in procedure 2.14, provided valuable information to plant

staff regarding proper fire protection implementation. WC7 provided instructions for

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

.

I

'

i

. 1

4

,

68

{ controls of combustibles, ignition sources (hotwork), impairments of fire detection or

l

2 suppression systems, and firewatches. The inspector found that numerous maintenance

!

<

workers were cognizant of their responsibilities as presented in WC7. Staff interviewed

]

credited the newly established Site Fire Protection Department's efforts for making them  !

aware of the instructions and recently implemented program requirements including the

j permit processes described in report section F1.2. The inspector noted that a self-

{ assessment performed by the Site Fire Protection department, dated June 23,1996,

j identified weaknesses in procedure 2.14 as discussed above. At the time of this  !

inspection, a fire protection program manual was being developed by the Site Fire j

j Protection Department to better define the fire protection program. )

3 Engineering support for the fire protection program was found by the inspector to be

j inconsistently implemented and assigned among the three units. Based on a review of  !

design work planned, in-progress, and completed to support fire protection and interviews

with various engineering staff personnel and supervisors, the inspector found that:

,

! * initiatives being taken to improve the fire protection program were not adopted by

$ all units;

i

2

• fire protection system engineers were also assigned to support numerous other i

j plant systems;

>

.

1

e a narrow interface existed among ech unit's fire p.otection engineering personnel,  ;

and

'

]

1

i e at Units 2 and 3, design work was performed by a number of staif engineers in lieu

of a dedicated fire protection design engineer.

'

The inspector noted that no fire protection coordinator position existed at Millstone and

system engineers had not established performance indicators to gauge the effectiveness of

i fire protection.

j

The inspector found that, in general, an appropriately low threshold had been established

l by plant staff for documenting potential discrepancies. However, engineering work, as

] documented in adverse condition reports (ACRs) had not been prioritized or resolved

l commensurate with their significance. These ACR issues included: the use of combustible

1

materials for fire area separation; recurring emergency light battery problems; and the

implementation of a corrosion control and monitoring program for the fire water system.

The inspector discussed this lack of program oversight with senior licensee management.

The Engineering Vice-President agreed with the inspector's concerns and stated that

actions were currently being planned by the licensee to resolve su6 issues prior to any

unit start-up.

c. Conclusions

<

The inspector concluded that the fire protection program procedure was marginally

acceptable in that it failed to provide good direction and guidance on specific functional

_

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

'

l

?

.

69

responsibilities and details to staff for implementing and controlling the requirements of the

fire protection program. Engineering support for the fire protection program was

inconsistently implemented and assigned, and no program oversight was evident. i

'

F1.2 Permit Processes

i

a. Insoection Scooe (64704)

The inspector reviewed the adequacy of the established controls for preventing fires during

fire risk activities, including the introduction and use of combustible materials.

b. Obsery.ations and Find!nas

The inspector found that the administrative processes for control of ignition sources and

combustible materials appropriately required special authorization prior to the introduction  ;

of the fire risk. Good work instructions have been implemented that meet the requirements

of NRC Branch Technical Position 9.5-1, " Fire Protection Program," and accomplish the '

following:

• prohibit bulk storage of combustible materials in safety-related areas;

i

e govern the handling and limited the use of combustible materials:

• maintain periodic housekeeping inspections; and 3

e control impairments and establ% appropriate compensatory measures, j

l

The inspector reviewed the permit log records and held discussions with Site Fire -

Protection department persor,,el to assess the recently implemented process for

authorizing and tracking permits. Permits were required for the storage and use of 1

combustible materials, hotwork activities including welding, grinding, and cutting, and

impairments of any fire protection equipment including, but not limited to, fire doors, fire

barrier penetrations, any sprinkler or detection ':ystem, or portion thereof. The inspector

found that the licensee had implemented an efficient review process for assessing plant

conditions prior to authorizing fire risk work. This review process utilized a software

program that enabled an assessment of all fire risk related activities by fire area, as

identified by each unit Fire Hazard Analysis report. The software program identified all

open combustible, impairment, and hotwork permits to the reviewer for consideration, prior

to the authorization of any additional work activities.

I

c. Conclusions

The inspector concluded that the licensee had established very good measures for

minimizing fire risk due to the introduction of ignition sources and combustibles. The

review process implemented was noteworthy, and allowed for comprehensive fire

protection assessments of fire area conditions.

!

l

.

.

70

F4 Fire Protection Staff Knowledge and Performance

F4.1 Fire Briaade Personnel

a. Insoection Scooe (64704)

The inspector performed a review of the training provided for fire brigade members, lesson

and pre-fire plans, and completed training and fire drill rosters of selected personnel to

verify their qualification for duty,

b. Observations Findinas

The inspector verified that the licensee had developed a training program that required the

following criteria:

e announced and unannounced drills;

e a minimum of two drills per year for each fire brigade member;

e at least or.e backshift drill per year per brigade member;

• maintenance of training records; and

• training and retraining at prescribed frequencies.

The inspector found that lesson plans selected for review, as identified in Attachment F,

presented training materialin an organized and clear manner. In general, pre-fire plans

were in place that identified firefighting equipment, fire area layouts, and potential hazards;

however, plans were difficult to read due to reproduction quality and the complexity of the

drawings.

The inspector verified that five fire brigade members, selected for review from each of the

three plants, had successfully completed the required training courses and drills.

c. Conclusion

The inspector concluded that proper training procedures had been developed and

implemented to qualify fire brigade personnel for duty.

F4.2 Fire Briaade Drills

a. insoection Scope (64704)

The inspector observed an unannounced fire brigade drill to evaluate the brigade's

effectiveness and understanding of fire attack strategies. The drill was conducted to

demonstrate the following:

• an understanding of the fire attack strategy;

.

,

71

• the ability to assess the fire properly;

• an awareness of vital equipment in the area;

• effective communication with other fire brigade members;

• an awareness of additional hazards in the fire area; and

• search and rescue techniques.

b. Observations and Findinas

The inspector observed a fire drill on August 29,1996, that involved a hydrogen seal oi!

fire in the Millstone Unit 2 turbine building. This drill scenario duplicated the simulated

conditions experienced by brigade personnel during annual hands-on training at the

Northeast Utilities training facility mock-up.

.

The performance of the fire brigade personnel during this drill was unsatisfactory and

subsequently categorized as a drill failure by the training department instructor in charge of

the drill. The bases for the failure included:

• use of an inappropriate suppressant type on the fire;

• untimely response to fire scene;

e poor command and control by fire brigade leader;

e poor teamwork by brigade members (i.e., failure to have a beck-up hose line in

place, failure to have brigado back-up present during search and rescue efforts and

fire attack, failure to verify each other's proper attire / readiness); and  !

e poor communications among brigade members.

The inspector found the quality of the critique following the drill to be marginally sufficient i

for providing constructive feedback to the brigade regarding individual performance.

Statements made by the drill critiquer sometimes contradicted previous statements l

regarding brigade performance. in addition, the inspector noted that poor support was  !

provided by security and chemistry during the fire drill. Security personnel were not posted

to limit access near the fire area or direct drill participants along safe pathways to the fire.  !

Chemistry personnel did not attend the drill.

The licensee appropriately initiated ACR No. M2-96-0333 and developed a remediation

plan to include a repeat drill within thirty days, as required by 10 CFR Part 50. Accendix R.

In addition, the licensee held a second unannounced fire drill on September 4,1996, that

was also observed by the inspector.

Results of this second fire drill were satisfactory. The inspector noted that support I

provided to the fire brigade by the training and operations department exceeded all

j

,

.

72

previous drills completed over the last three years as documented in drill evaluation

reports. In addition, the training department implemented improvements to the fire brigade

drill critique reports prior to the end of this inspection. These improvements provided

clarity for making drill determinations as either satisfactory or unsatisfactory. Brigade

members demonstrated an adequate understanding of fire attack strategies during the

second drill.

c. Conclusions

The inspector concluded that the training program provided for fire brigade members was

adequate for preparing them to combat fires. However, the effectiveness of measures

taken by training, including drill activities, to support brigade member readiness was

demonstrated by mixed performance observed during different drills.

F7 Quality Assurance in Fire Protection Activities

F7.1 Audits and Surveillances

a. Insoection Scoce (64704)

l

The inspector reviewed the three most recent audits completed to satisfy the technical ,

specification requirements and surveillances to evaluate the effectiveness of fire protection l

measures, equipment, program implementation, and probiern identification and resolution.- j

1

b. Observations and Findinas l

l

t

The inspector found that audits:

j

e were limited in scope

• often categorized findings as a low significance, thereby not requiring root cause

analyses, and

• failed to follow-up on previously identified issues.

In addition, the inspector found that weak corrective actions were accepted by the quality

assurance 1QA) department to resolve identified concerns.

More specifically, audits sampled had very limited reviews of only three or four different

program areas, usually including housekeeping, fire drill, and combustible material storage.

On two occasions, work orders for opening sprinkler piping failed to include a WC-7

required inspection by the chemistry department or fire protection engineer to determine if

mejor biological fouling of the pipe was present. Corrective actions accepted by QA

consisted of the issuance of a memo stating that supervisors were required to discuss this

issue with their personnel. However, tne inspector noted that this issue was categorized

as low significance, and subsequently did not require a root cause analysis for determining

why personnel were not familiar with fire protection requirements as presented in WC-7.

As a low significance finding, no followup was required to verify the adequacy of

1

f

i J

~

.

73

corrective actions taken nor why personnel were unfamiliar with WC-7 requirements or

other fire protection program requirements.

Only one fire protection surveillance had been conducted by the OA department fire

protection during the past three years.

Although the inspector identified deficiencies with QA's oversight of fire protection, the

licensee was already aware of overall QA ineffectiveness. For example, the adequacy of

Northeast Utilities QA program had recently been reviewed by a Joint Utility Management ,

Assessment (JUMA) team. The results of this review by JUMA concluded that Northeast i

Utilities' audit, surveillance, and inspection programs were ineffective based on the lack of

management support and an adequate corrective action program. Millstone Nuclear Power

Station, Unit No. 3 Operational Readiness Plan Punchlist. dated August 13,1996, further

identifies deficiencies within the QA department. Resolution of OA programmatic issues

was beyond the scope of this inspection and will be addressed as a restart item by the

NRC Management Oversight Team.

c. Conclusions 1

!

The inspector concluded that GA audits failed to provide valuable assessments of the fire

protection program. Although technical specification requirements were satisfied for

conducting the annual program reviews, the root causes for discrepancies identified were

not pursued.

F8 Miscellaneous Fire Protection issues

F8.1 pocuments Reviewed

A list of fire protection documents reviewad is included as Attachment F to this inspection

report.

F8.3 Overan Conclusion

The inspector concluded that the fire protection and prevention program was,in general,

lacking appropriate direction for prioritizing and resolving identified issues. Positive

inspection findings identified regarding permit systems and procedure WC 7 reflected well

on the newly established Site Fire Protection Department; however, engineering support for

the fire protection program was found to be inconsistently implemented and assigned.

V. Manaaement Meetinaq

X1 Exit Meeting Summary

The inspectors presented the inspection results to members of licensee management at the

conclusion of the inspection during individual unit and visiting inspector exit meetings. The

licensee acknowledged the findings presented.

S

.

74

X 1.1 Final Safety Analysis Reoort Rev6a'

A recent discovery of a licensee operating their facility in a manner contrary to the updated

final safety analysis report (UFSAR) description highlighted the need for additional

verification that licensees were complying with UFSAR commitments. All reactor

inspections will provide additional attention to UFSAR commitments and their incorporation

into plant practices, procedures, and parameters.

While performing the inspections' which are discussed in this report, the inspectors

reviewed the applicable portions of the UFSAR that related to the areas inspected.

Inconsistencies were noted between the wording of the UFSAR and the plant practices,

and procedures and/or parameters observed by the inspectors, as documented in Sections,

U2.01.2, U2.M8.3, U2.M8.4, U2.E8.1, U2.E8.6, U3.01.1, U3.05.2, U3.E8.1, U3.E8.3,

and R8.3.

While performing the inspections discussed in Section F of this report, the inspector

4

reviewed the applicable portions of the UFSAR that related to the areas inspected. This

included portions of the following:

e Section 9.5.1 of Millstone Unit 1 UFSAR;

e Section 9.10 of Millstone Unit 2 UFSAR; and

e Section 9.5.1 of Millstone Unit 3 UFSAR.

The inspector noted that at the time of this inspection the fire protection system engineers

for each Millstone unit were currently verifying the wording of each UFSAR for consistency

with known plant practices, procedures and/or parameters regarding fire protection

including alllicensing documents.

X3 Management Meeting Summary

, During this inspection period, Northeast Utilities (NU) Nuclear Group underwent a major

l reorganization. A number of management changes occurred, resulting in a Recovery

>

Organization with the following senior personnel assuming the designated management

positions:

B.D. Kenyon President and Chief Executive Officer

T.C. Feigenbaum Executive Vice President and Chief Nuclear Officer

D.M. Goebel Vice President, Nuclear Oversight

F.C. Rothen Vice President, Work Services

J.K. Thayer Nuclear Engineering and Support Recovery Officer

J. McElwain Millstone Unit 1 Recovery Officer

M. Bowling Millstone Unit 2 Recovery Officer

J. P. Cowan Millstone Unit 3 Recovery Officer

On October 1,1996, new recovery organizations were established at all three units.

Representatives from PECO Energy, Virginia Power, and Carolina Power and Light were

9

.

75

contracted / loaned by these nuclear facilities to lead and provide support personnel for this

recovery effort at Units 1,2, and 3 respectively. The supporting personnel will augment

the NU unitized organizations as determined by the loaned Recovery Officer. The loan of

the Nuclear Engineering and Support Recovery Officer from the Vermont Yankee Nuclear

Power Corporation became effective on October 21,1996. i

.

i

i

j

.

3

.

76

INSPECTION PROCEDURES USED

IP 37551: Onsite Engineering

IP 61720: Containment of Local Leak Rate Testing

IP 61726: Surveillance Observations

IP 62700: Maintenance Program implementation

IP 62707: Maintenance Observations

IP 64704: Fire Protection Program

IP 71001: Licensed Operator Requalification Program Evaluation

IP 71707: Plant Operations

IP 71750: Plant Support Activities

IP 86750: Solid Radioactive Waste Management and Transportation of

Radioactive Materials

IP 92700: Onsite follow-up of Written reports of Nonroutine Events at Power

Reactor Facilities

IP 92901: Follow-up Operations

IP 92902: Follow-up Maintenance

_ _ _ _

• t

.

77

ITEMS OPENED, CLOSED, AND UPDATED

Opened

eel 245/96-08-01 U1.M4.1 Troubleshooting Activities

eel 245/96-08-02 U 1.E 8.3 SGTS Operability Determination

eel 245/96-08-03 U 1.E 8.4 CRD System Modifications

URI 245/96-08-04 U 1.E 8.4 CRD Operability Calculations  !

URI 245/96-08-05 U1.E8.4 SEP Seismic issues

eel 336/96-08-06 U 2.02.1 Refueling Pool Drain Line

l

VIO 336/96-08-07 U2.M8.4 Containment isolation Valve Surveillance

eel 336/96-08-08 U2.M8.4 LER Corrective Action Failure

URI 336/96-08-09 U 2.M 8.5 Response Time Testing of Components

eel 336/96-08-10 U2.M8.7 Heavy Load Concern and Corrective Action

eel 336/96-08-11 U2.E8.1 Inoperable Hydrogen Monitor l

eel 336/96-08-12 U2.E8.1 Inadequate Design Control of Modification

eel 336/96-08-13 U 2.E 8.1 FSAR Update for Hydrogen Monitor System ,

URI 336/96-08-14 U2.E8.7 Removal of Startup Rate Trip l

I

IFl 423/96-0815 U 3.02.1 TS Consistency - SWP Train

URI 423/96-08-16 U 3.0 5.2 Operator Action - SGTR Analysis

IFl 423/96-08-17 U3.07.1 Questionable Fuse Quality - Cracked Ferrules

URI 423/96-08-18 U 3.M 8.1 &2 Inadequate IST Program Controls

URI 423/96-08-19 U 3.E2.1 Potential Overpressurization Failure of SOVs

URI 423/96-08-20 U 3.E8.4 CCP system Temperature Limits

IFl 245/96-08-21 &

336/96-08-21 RS.1 Material Condition Program

Closed Sp_c_ tion

URI 245/95-25-01 U 1.E 8.1

URI 245/95-25-02 U1.E8.2

URI 245/95-44-01 U 1.E 8.3

VIO 423/95-42-05 U3.08.4 & 5

VIO 423/95-38-01 U 3.M 8.3

URI 423/96-05-14 U 3.E8.1

URI 423/94-32-02 U3.E8.4

Uodated

URI 245/96-06-03 U 1.M 8.1

URI 336/96-01-05 U 2.E8.1 4

LER 336/96-16 U2.M 8.3

.

_i .J. ---4+= 4m J.---d-.M 4-4,A.e+-e- 4 we - &=4+- J wm-*hsA4-- 4 A AJ4a_- -e.- J a .4 .4 As- -M=dAA-~.A

$

1

el

'

78

l The followina LERs were also closed durina this insoection: {

i

!- DN 50-245

'

96-02

96-07

96-40

96-41

96-47

DN 50-336

l

96-14 96-24

96-17 96-25

96-18 96-26

96-20 96-27

96-21 96-28

96-22 96-29

96-23 96-30

DN 50-423

96-02 96-21

96-12-01 96-23

96-13-01 96-24

96-14 96-31

96-17

9

J

79

,

LIST OF ACRONYMS USED

ACP Administrative Control Procedure

ACR adverse condition report

AEAS auxiliary exhaust actuation system

AOP abnormal operating procedure

AOT allowable outage time

ASA American Standards Association

ASME American Society of Mechanical Engineers

AWO automated work order

BAST boric acid storage tank

CCF reactor plant component cooling

CFR Code of Federal Regulations

CIAS containment isolation actuation signal

CNSI Chem Nuclear Systems, Inc.

CRD control rod drive

CVCS chemical and volume control system

CW circulating water

DAW dry active waste

DCR design change record

DP differential pressure

DRP Division of Reactor Projects

EA escalated enforcement

EBFS enclosure building filtration actuation system

ECCS emergency core cooling system

ECT eddy current testing

EDG emergency diesel generator

eel escalated enforcement item

EEQ electrical equipment qualification

EOF Emergency Operation Facility

EPAP emergency preparedness administrative procedure

EPRI Electric Power Research institute

ERNS emergency response notification system

ESAS Engineered Safeguards Actuation System

ESF engineered safety feature

FLS first line supervisor

FSAR Final Safety Analysis Report

GDC general design criterion / criteria

GL Generic Letter

ICAVP Independent Corrective Action Verification Program

IE lE (Office of Inspection and Enforcement) bulletin

IFl inspector follow item

Ins Information Notices

IP inspection procedure

Irs inspection Reports

ISP improving station performance

IST in-service testing

JPM job performance measure

.

A

..

9

4

80

JUMA Joint Utility Management Assessment

LER licensee event report

LNP loss of normal power

LOCA loss of coolant accident

MCP materials condition program

MEPL material, equipment, and parts list

MOPD maximum operational pressure differential

MRT management review team

NDE non-destructive examination

, NNECO Northeast Nuclear Energy Company

NRC Nuclear Regulatory Commission i

NRR Office of Nuclear Reactor Regulation

NSIC Nuclear Safety Information Center

NUPIC Nuclear Utilities Procurement Issues Council  !

NUREG Nuclear Regulation i

OCA Office of Congressional Affairs ,

OD operability determination {

, OP operating procedure

OSC Operational Support Center

P&lD piping & instrumentation diagrams

PAO Public Affairs Office

PEO plant equipment operator

l PDR Public Document Room

PMMS planned maintenance management system

PMW primary makeup water i

PNL Pacific Northwest Laboratory

PORC plant operation review committee

QA quality assurance i

QAS Quality and Assessment Services l

RBCCW reactor building closed cooling water

RCM Regulatory Compliance Manual

RCP reactor coolant pump

RF03 third refueling outage

RG Regulatory Guide

RHS residual heat removal system

RI Region i

RO reactor operator

RPS reaction protection system

RWST refueling water storage tank

SER safety evaluation report

SCFM standard cubic feet per minute

SFP spent fuel pool

SGCS safety grade cold shutdown

SGTR steam generator tube ruptures

SGTS standby gas treatment system

SM shift manager

SOVs solenoid-operated valves

SP surveillance procedure

9

4

81

l

SPO Special Projects Office

j

SRO senior reactor operator t

SSCs selected structures, systems, and components

STA shift technical advisor

,

.SWP plant service water

TIA Task Interface Agreement

TMI Three Mile Island -

TR trouble reports

TS technical specifications

TSC Technical Support Center

URis unresolved items '

VDC volts, direct-current

VIO

,

= violation  :

WC work control

P

.

I

'l

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

h

, ATTACHMENT F

Fire Protection Documents Reviewed

l Procedures (No./ Revision / Title):

NTM-7.205, Rev. 0 Fire Brigade Initial /Requalification Program

NGP-2.14, Rev. 9 Nuclear Plant Fire Protection Program

WC-7, Rev.1 Fire Protection Program

Lesson Plans:

TTB-FBI-LOO 6 Communications

TTB-FBI-LO17 Fire Fighting

TTB-fbi-LO11 Selected Scenarios

TTB-fbi-LOO 1 Introduction to Fire Brigade Leader

TTB-FBLI-LOO 6 Table Top Exercise

Surveillances:

SP 680Y Rev.1 Sprinkler System Inspection

SP 680N Rev.10 18-Month, Fire Barrier Penetration inspection

SP 2734D Rev.4 Fire Penetration Seal Inspection

SP 680M Rev.8 Fire Pumps Operability Demonstration

MP 3780AE Rev.5 Emergency Lighting PM Procedure

MP 790.2 Rev.16 Emergency Light inspection

MP 3780AF Rev.2 Emergency DC Lighting Discharge Test

SFP 10 Rev. O Fire Prevention inspections

SP 2618F Rev.10 Fire Pump Performance Test

SP 2618K Rev.5 Fire Protection System Valve Lineup

Verification

SP 3641 A.4 Rev.8 Fire Protection Water System Functional Test

and Deluge Spray Nozzle Operability

IC 2439A hev.3 Fire Suppression System Test

QA Audits Reviewed:

No.A24037 1993 MPS Fire Protection / Loss Prevention

No.A24042 1994 Triennial Fire Protection

No.A24050 1995 Fire Protection Millstone Station

UFSARs

Millstone Units 1,2, and 3

Technical Reauirements Manuals

Millstone Units 1,2, and 3

Fire Protection Safetv Evaluation Reoorts

Fire Hazard Analvses

Millstone Units 1,2, and 3