IR 05000271/1997005

Insp Rept 50-271/97-05 on 970601-07-19.Violations Noted. Major Areas Inspected:Licensee Operations,Engineering,Maint & Plant Support

ML20210S029
Person / Time
Site:	Vermont Yankee File:NorthStar Vermont Yankee icon.png
Issue date:	08/19/1997
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20210S011	List: IR 05000271/1997005 ML20210S008 ML20210S017
References
50-271-97-05, 50-271-97-5, NUDOCS 9709040351
Download: ML20210S029 (32)

Text

t i

U.S. NUCLEAR REGULATORY COMMISSION

REGION I

Docket No.

60 271 I

Licensee No.

DPR 28

_ Report No.

97 05

,

,

Licensee:

Vermont Yankee Nuclear Power Corporation l

Facility:

Vermont Yankee Nuclear Power Station Location:

Vernon, Vermont

!

I~

Dates:

June 1 July 19,1997 Inspectors:

William A. Cook, Senior Resident Inspector Edward C. Knutson, Resident inspector Keith A. Young, Reactor Engineer Jeff Laughlin, Emergency Preparedness Specialist Approved by:

Curtis J. Cowgill, Ill, Chief Projects Branch 5 Division of Reactor Projects

.

. f[9Nk ob271

-

-

~

PDR _

-

_

_

-

-

.

.

EXECUTIVE SUMMARY Vermont YarAce Nuclear Power Station NRC Inspection Report 60 271/97 05 This inspection included aspects of licensee operations, engineering, maintenance, and plant support. The report covers a seven week period of resident inspection. In addition, it includes the results of in office procedure reviews conducted by an emergency preparedness specialist.

QREA11201 The torus Inerting that was performed on May 0 was secured before the oxygen concentration had been reduced to less than four percent due to procedure inadequacies.

Although these inadequacies were not the result of the change to the procedure which removed the option for simultaneous inerting of the drywell and torus air spaces, the remaining method for inerting was not carefully scrutinized, from the perspective of an infrequently performed evolution, to ensure its technical adequacy. In addition, measures were not established to verify ihe adequacy of the torus purge prior to exceeding the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> allowed by TS. Since r verall containment oxygen concentration remained below the threshold defined in the F%R, the inspectors concluded that the safety consequences were not significant. (Section 01.2)

The oxygen concentration in the torus air space remained at approximately eight percent from May 8 until the morning of May 12, when it was reduced to approximately one percent by the second inerting operation. TS 3.7.A.7.b requires that, within the 24-hour period subsequent to placing the reactor in the RUN mode following a shutdown, the containment atmosphere oxygen concentration shall be reduced to less than four percent and maintained in this condition. In that the modo switch was placed in RUN on May 7 at 8:20 p.m. and the licensee operated the unit at power from 8:20 p.m. on May 8,1997 to the morning of May 12,1997 with containment atmosphere oxygen concentration greater that four percent, this was contrary to TS 3.7.A.7.b and was a violation. (Section 01.2)

The apparent cause of draining hydraulic control unit (HCU) 18 31, rather than HCU 14 31, was inadequate self checking by the responsible auxiliary operator prior to opening the HCU 18 31 accumulator leak off line isolation valve. This event did not constitute a violation of TS 3.3 F and, given the brevity of the event, the licensee's operational response was reasonable. However, the actions required by TS 3.3.F indicate that having two HCU accumulators inoperable within a nine rod square array is a significantly degraded condition. The personnel error of draining the wrong HCU accumulator was contrary to TS 6.5, " Plant Operating Procedures," and was a violation. However, since this event occurred prior to completing corrective actions for a recent similar violation, the NRC considered this to be another example of the prior violation. (Section 04.1)

Maintenanaq inspector review of the HCU 1011 scram solenoid pilot valve (SSPV) assembly replacement and associated post maintenance testing identified generally good conduct of il

,s

-

!

l

L the work and appropriate confirmatory post work testing.- Review of a number of

'

surveillance tests identified good operator and maintenance staff conduct of these evolutions. (Section M1.1)

A number of inspector observations involving material condition deficiencies and non-conforming conditiens cumulatively represented a failure of the plant staff to assure that conditions adverse to quality are promptly identified and corrected via an established

'

quality assurance process. The inspectors noted that for some of the conditions involving i

potential seismic qualification concerns, the licensee stated in Event Report (ER) No. 97-0785 that, "...the seismic qualification program does not have a clear identification of

what components are qualified, nor does it specify requirements on how to properly rnalntain that qualification." The failure to have identified and promptly corrected the material deficiencies and non conforming conditions identified by the inspectors during this inspection period, was contrary to 10 CFR 50, Appendix B, " Quality Assurance Criteria for Nuclear Power Plants and Fuel Reprocessing Plants," Criterion XVI, " Corrective Action,"

and was a violation. (Section M2.1)

.

Licensee evaluation of a deficient electrical conduit in the residual heat removal (RHR)

system was delayed due to problems in entering the deficiency into the corrective action program. Although the affected circuit was subsequently determined to be classified non-nuclear safety and have no impact on system operability, the potential existed that an avoidable degradation of this low pressure emergency core cooling system (ECCS)

capability could have occurred. Licensee response to this latter issue was appropriate.

,

(Section M2.2)

-

As a result of the April 24,1997 reactor scram root cause evaluation, procedures were being revised to perform instrument calibrations singly, in cases where an error could result in a half scram or full scram. A specific case where two average power range monitor (APRM) channels continue to be bypassed simultaneously was adequately justified.

(Section M3.1)

Licensee actions were appropriate in re establishing the pre calibration gain settings of the

local power range monitors (LPRMs) following the partial surveillance performed on April

!

24,1997. Following the return to power operations, LPRM calibrations were properly v

completed within the minimum frequency requirement of TS 4.1.A. (Section M8.1)

Enaineerino

An NRC Architec./ Engineering inspection completed its onsite review activities during this

'

inspection peric J. The results of this inspection will be presented in inspection report 50-271/97 201.

,

VY was slow to initiate the Basis for Maintaining Operation (BMO) process after concerns were identified with RHR and core spray (CS) system minimum flow requirements.

i Development of the BMO took longer than was targeted by the guidance. Although some delay was the result of concerns that were appropriately raised by the plant operations

,

review committee (PORC), the time required for development did not appear to be

commensurate with the safety significance of the issue. This issue will be discussed

<

iil

,

-

---,#

,

,,,,

c,m.,,, _, -,

.r -..

y..y,.~

.-_.,m.-,

,,y

. -

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

.v

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

y

!

.

further in the Architect / Engineering Team inspection report 50 271/97 201. (Section E1.1)

Inspector review of the 24 VDC system identified that Section 8.7 of the updated final safety analysis report (UFSAR) does not include the safety system designation of the 24 VDC system as non nuclear safety (NNS). This item remains unresolved pending further review of the design basis documentation. (Section E1.2)

The inspector was not able to determine the safety classification of the high pressure coolant injection / standby gas treatment (HPCl/SGT)line and the torus /drywell pumpback system. If the HPCl/SGT line is safety class 3, then it appears that the connection of the retired in place torus /drywell pumpback system is potentially inadequate and, that the system is not being maintained as safety class 3 and is therefore, potentially not isolable from the HPCl/SGT line. Pending determination of the safety classifications of this j

equipment, this issue remains unresolved. (Section E1.3)

The implementation of the VY BMO Guideline for dispositioning degraded and non-conforming conditions was appropriate for the recently discovered cable separation issues

- and the operability determinations were adequately supported. However, the inspector viewed the current VY processes (BMO Guidelines and AP 6002) for addressing degraded and non conforming conditions as warranting further refinement because these processes potentially employ a 10 CFR 50.59 safety evaluation when the condition does not warrant it. The NRC staff's review of the licensee's corrective actions and interim compensatory actions concluded that they were satisfactory (reference NRC memorandum, dated June 10,1997,in response to Task interface Agreement, " Technical Assistance in Support of Vermont Yankee Cable Separation Problem TAC No. M98629", dated May 30,1997).

However, unresolved item (URI 97 03 02) remains open pending NRC review of VY's root cause evaluation and long term corrective actions for the cable separation non.

conformances. (Section E1.4)

Two technicalissues are briefly discussed in this report which have been assigned an inspection follow item (IFI) including the HPCI suction valve Appendix J testing concerns and the primary containment interior coating concerns. (Section E7.1)

Plant Suncort

{

in-office review of emergency plan procedures OP 3542, revision 14, and OP 3536, revision 1 identified no concerns.

Iv

-

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

_-

t i

'

!

!

i (

TABLE OF CONTENTS

'

!

i

l

!

l E X EC UTIVE SU MM ARY............................................. ll l

i

!

T A BLE O F CO NT ENTS............................................... v i

l Summ ary of Plant St atus............................................ 1

i 1. Operations

.................................................... 1

Conduct of Operations.... _................................ 1

01.1 Control Rod Pattern Exchange

.........................

l-01.2 (Closed) Unresolved item (97 04 02h High Oxygen

!

l Concentration in the Torus Air Space During Plant Operations... 2 i

!

Operator Knowledge and Performance......................... 4

04.1 Hydraulic Control Unit Momentarily Made Inoperable due to

Personnel Error.................................... 4

ll. Maintenance

,

..................................................

l M1 Conduct of Maintenance

..................................6 l

M1.1 Maintenance Observations............................ 6

i M1.2 Surveillance Observations

............................

.

M2

'Melntenance and Material Condition of Facilities and Equipment

!

,

......

!

M2.1-Maintenance Related Inspection Observations.............. 8

!

M2.2 Broken Conduit for RHR System Annunciator Circuit.......... 9

!

M3 Maintenance Procedures and Documentation.................. 11

M3.1 Nuclear instrument Procedures That Affect Both Channels of

= the Reactor Protection System........................ 11 l

M8 Miscellaneous Maintenance issues.......................... 12

-

l

M8.1 Partial LPRM Calibration of April 24,1997................

l 111. En gi n e e ring.................................................. 13

.

,

E1 Conduct of Engineering

..................................

l

.E1.1 Residual Heat Removal and Core Spray Systems Minimum

-

Flow lasue

.....................................

E1.2 Seismic Design Considerations for the *24 Volt DC Power System.........................................

E1.3 Interface of a Retired in Place System with the High Pressure

-

Conlant injection System................. -........... 17 E1.4 (Update) URI 97 03 02: Electrical Cable Separation Review

.. 18 E7 Quality Assurance in Engineering Activities..,...............,. 21

E7.1 Review of Licensee Identified Engineering issues........... 21 E8 Miscellaneous Engineering issues........................... 22

!

l E8.1 (Closed) Unresolved item (96 08 01): East and West L

- switchgear room carbon dioxide suppression system design d e f icie n cy....................................... 22 E8.2 (Closed) Inspection Follow item (96 09-03): Residual heat

v

,

l

?

l

-

,

,.,

, _

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

-,,.., _ _ _ _ _

,

,

,,.,.,

_

,,.. _ ~ _ -.,

, -, _.,,.

.

removal pump start interlock problem................... 23 IV. Plant Support

................................................ 23 P3 Emergency Procedures and Documentation.................... 23 P3.1 In-Office Review of Licensee Procedure Changes........... 23 V.

M a na g em e nt M e eting s........................................ 24 X1 Exit Meeting Summary................................... 24 X3 Review of Updated Final Safety Analysis Report (UFSAR).......... 24

,

INSPECTION PROCEDURES USED..................................... 25 ITEMS OPENED AND CLOSED

....................................... 25 PARTI AL LIST OF PERSONS CONTACTED............................... 26 LIST O F AC RO NY M S U S E D......................................... 26 I

vi

_

Report Details Summary of Plant Statug At the beginning of the inspection period, Vermont Yankee (VY) was operating at 100 percent reactor power. On two occasions, power was reduced to 80 percent to support maintenance on scram solenoid pilot valves (hydraulic control unit (HCU) 1011 on June 11, and HCUs 10 35 and 18 35 on July 17). The plant operated at 100 percent reactor power for the remainder of the inspection period with the exception of power reductions to conduct planned rod pattem changes (June 11,12, and July 1) and to conduct planned

'

surveillance testing.

During the inspection period an NRC Architect / Engineering review completed its onsite activities. The results of this inspection will be presented in inspection report 50 271/97-201.

During the inspection period, the licenseo made six non emergency notifications to the NRC operations conter to report events as required by 10 CFR 50.72, as follows: On June 10, a one hour notification (EN 32456, retracted on July 18) that non nuclear safety class

}

pressore regulators were in use on service water system temperature control valves used for alternate emergency diesel generator cooling (a condition outside design basis); on June 13, a one hour notification (EN 32482) that the original primary containment point was not qualified or tested in accordance with the FSAR and could experience premature blistering (a condition outside design basis); on June 26 and July 11, one hour notifications (ens 32544 and 32608) that ths Ames Hill transmitter had been found inoperable; on July 1, a one hour notification (EN 32664) that the control room ventilation system had been declared inoperable for adjustment of an air inlet damper (control room ventilation is not addressed in technical specifications, but the licensee considered the condition to be outside design basis); and, on July 1, a one hour notification (EN 32566) that three of the four safety relief valves, replaced during the 1996 refueling outage, had subsequently demonstrated actuation times exceeding the maximum allowable (a condition outside design basis).

LOperations

Conduct of Operations'

01.1 Control Rod Pattern Exchanae (71707)

The inspector observed the power reduction to 80 percent and the rod pattern exchange on June 11. Power reduction was achieved by reducing recirculation flow, and was done in a controlled manner. The rod pattern exchange was directed by reactor engineering departmert personnel under the supervision of the senior control room operator and the shift supervisor. Licensee management was also present in the control room during the evolution. The exchange was performed in a Topical headings such as 01, M8, etc., are used in accordance with the NRC standardized reactor inspection report outline. Individual reports are not expected to address all outline topic :

.

deliberate manner, as demonstrated by the performance of average power range monitor (APRM) gain adjustment factor (GAF) adjustments twice during the I

exchange. Upon completion of the exchange, a single rod scram time test was

,

performed on rod 1011, and power was returned to 100 percent.

01.2 (Closed) Unresolved item (97 04-02h High Oxygen Concentration in the Torus Air Space During Plant Operations t

a, lunngstion Scone (92901)

.

On May 12,1997, a weekly surveillance to measure oxygen concentration in the i

torus air space revealed that the concentration was greater than the four percent limit established by technical specification (TS) 3.7.A.7. This event was discussed in inspection report 97 04 and was left unresolved, pending further review of the cause of the event, b.

Observations and Findinas Following discovery of the elevated torus oxygen concentration, the inspectors confirmed that the combined drywell and torus air space oxygen concentration was t

appropriately calculated to be 4.6 percent. This concentration was less than the

!

minimum 5 percent oxygen concontration necessary for rapid recombination (an

,

exothermic reaction) of hydrogen and oxygen (reference UFSAR Section 5.2.6.3),

t Therefore, the safety function of the inerted primary containment was maintained.

.

throughout this event, During this inspection period, the inspector reviewed licensee event report (LER) 97+

' 011, revision 0, "The primary containment torus was not inerted to technical

'

.

,

specification requirements due to an inadequate procedure which resulted in an'

insufficient nitrogen inerting purge flowrate." The LER indicated that the nitrogen purge flow rate used during the initial torus inerting operation on May 8 was approximately 20,000 standard cubic feet per hour (SCFH). The licensee had subsequently determined that this value was approximately 22 percent of what it had been using the previous procedural revision (that is, purging the drywell and the torus at the same time). The licensee concluded that the cause of the event was,

"an inadequate procedure which led to the use of Insufficient nitrogen purge flow rate resulting in the torus not being inerted to less than four percent."

The inspector reviewed operating procedure OP 2115, "Primarv Containment."

Revision 37 to OP 2115, dated April 24,1997, had eliminated the option to purge the drywell and the torus at the same time; revision 38, dated May 6,1997 (the revision in use at the time of the event), incorporated a requirement to maintain the torus vent system rupture disc isolated. Neither of these procedure revisions changed the procedural steps used to initiate nitrogen flow and establish the purge flow rate. This indicated that the procedural problems which led to the inadequate torus purge had existed during previous containment inerting operations, but were not evident because the drywell and torus had been inerted at the same time.

Specifically, the procedure did not contain sufficient information on purge flow rate

.

.

t

3 requirements. The proceduto contained a caution that, "the purge flow must be limited to 00,000 SCFH," but otherwise discussed flow in terms of containment pressure. The procedure did not include required or recommended flow rates, nor estimated times required to complete drywell and torus inerting at given flow rates.

Additionally, the guidance for increasing the purge rate by starting a standby gas treatment (SGTI fan was based on an indication that was valid only for purge operations that included the drywell. Specifically, the procedure directed that a SGT fan be started when drywell pressure reached 0.2 psig; however, the intent of the step was to start the SGT fan when pressure in the volume being purged (drywell or torus) reached 0.2 psig.

The procedure was written to initially establish a low flow rate purge, exhausting through one idle train of the standby gas treatment (SGT) system. Once sufficient pressure had built up, the SGT fan was started. The purge flow rate was then increased to maintain containment pressure within a specified band When the

.

torus purge was commenced on May 8, operators noted that the oxygen

!

concentration, as indicated by the hydrogen / oxygen (H2/02) monitor in the control room, was lowering significantly at the initiallow flow purge rate. This was apparently due to the close proximity of the H2/02 monitor sample point to the purge intet. Such non representative samples are not observed when purging the drywell (or the torus and the drywell at the same time) because, during those operations, the H2/02 monitor is aligned to the drywell and draws from four widely separated sample points. In addition, cooling fans in the drywell promote mixing of the nitrogen with the air, whereas mixing in the torus is only due to the nitrogen flow. The high purge flow rate (historically approaching the 90,000 SCFH limit)

required to produce the desired drywellinerting rate was also adequate to inert the torus; therefore, problems with high torus oxygen concentration were never encountered during previous use of the procedure to inert the drywell and torus

.

-

simultaneously.

The operators were satisfied that the inerting rate, using the low flow purge rate, would be adequate to reach the required oxygen concentration within the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> allowed by TS 3.7.A.7 b. Drywell pressure did not increase to 0.2 psig, so the SGT fan was never started. Consequently, inadequate mixing of the nitrogen with the torus air (due to the low flow purge rate), and the non representative atmosphere that was being sampled by the H2/02 monitor, led operators to secure from inerting the torus when the actual oxygen concentration was approximately eight percent, c.

Conclusions

>

The torus inerting that was performed on May 8 was secured before the oxygen concentration had been reduced to less than four percent due to procedure inadequacies. Although these inadequacles were not the result of the change to the procedure which removed the option for simultaneous inerting of the drywell and torus air spaces, the remaining method for inerting was not carefully scrutinized, from the perspective of an infrequently pr,rformed evolution, to ensure its technical adequacy. In addition, measures were not established to verify the adequacy of the torus purge prior to exceeding the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> allowed by T :

'

.

The oxygen concentration in the torus air space remained at approximately eight percent from May 8 until the morning of May 12, when it was reduced to approximately one percent by the second inerting operation. The inspectors further concluded that since the overall containment oxygen concentration remained below the threshold defined in the UFSAR that the safety consequences were not significant. TS 3.7.A.7.b requires that, within the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period subsequent to placing the reactor in the RUN mode following a shutdown, the containment atmosphere oxygen concentration shall be reduced to less than four percent and maintained in this condition, in that the mode switch was placed in RUN on May 7 at 8:20 p.m. and the licensee operated the unit at power from 8:20 p.m. on May 8, 1997 to the morning of May 12,1997 with containment atmosphere oxygen concentration greater that four percent, this was contrary to TS 3.7.A.7 b and was considered a violation. (VIO 97 05 01)

Operator Knowledge and Performance 04.1 Hydraulic Control Unit Momentarily Made inocerable due to Personnel Error a.

Insnection Scone (71707)

On June 17, control room operators received an accumulator trouble alarm for

.hydraulle control unit (HCU) 14 31. An auxiliary operator (AO) was dispatched to determine whether the cause was high water levelin the accumulator leak off line or low nitrogen pressure. These activities led to the operator draining the wrong HCU, The inspectors reviewed this event and the licensee's actions, b.

Observations and Findinns The accumulator trouble alarm is caused by either high water level or low nitrogen pressure in the accumulator drain piping. These Indications are available at the HCU bank, but not in the control room; the 44 individual local alarms for the HCU bank are in a 4x11 matrix on a common local alarm panel, in accordance with operating procedure OP 2111, " Control Rod Drive System," section P, "CRD Accumulator Alarm (CRP 9 5), the cause was found to be high water level; however, the AO identified the affected HCU from the label above the alarm indication (HCU 18 31),

rather than the correct label below the indication. The AO proceeded to drain the leak off line for HCU 18 31. Simultaneously, the accumulator trouble alarm for HCU 18 31 was received in the control room (in this case, due to low nitrogen pressure as seen by the detector connected to the leak off line). When the AO observed that there was no water present, he realized that he was draining the wrong accumulator and secured the lineup. With the lineup returned to normal, the HCU 18 31 l

accumulator trouble alarm cleared; the alarm condition had existed for approximately 30 seconds. The AO then proceeded to drain the correct HCU accumulator (14 31).

- A presence of water in the leak off line could indicate f ailure of the accumulator seal, and low nitrogen pressure could indicate that a nitrogen leak has developed.

Either of these conditions could cause a loss of accumulator function. However, I

.

.

'.

this alarm is typically due to the slow accumulation of water leakage past the seal.

Draining the occumulator leak off line is accomplished in a manner such that the HCU remains functional. However, an HCU is considered to be administratively inoperable by the licensee while an alarm condition exists.

Technical Specification (TS) 3.3.D allows for one rod accumulator to be inoperable provided that no other control rod in the nine rod square array around this rod has an inoperable accumulator, During the brief period that the HCU 18 31 pressure switch was isolated while the HCU 14 31 accumulator high water level condition existed, two HCU accumulators within a nine rod square array were inoperable (although functional). TS 3.3.F requires that, if specification 3.3.D cannot be met, an orderly shutdown shall be initiated and the reactor shall be in the cold shutdown condition within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The control room operators recognized the TS implications of the error, but concluded that no action was required because the condition no longer existed. HCU 18 31 was recorded both as being made _

,

!

inoperable and returned to service at 3:10 p.m. The AO then proceeded to drain the HCU 14 31 accumulator, which was completed at 3:20 p.m.

As corrective action, the AO was counseled by the shift supervisor on the need to use self checking techniques, and the operations department manager reviewed the event with shift operating personnel. An event report (ER 97 0766) was generated to enter this event into the licensee's corrective action program. At the close of the inspection period, this ER was stillin process.

Technical Specification 6.5.A requires that, " Detailed written procedures, involving both nuclear and non nuclear safety...shall be prepared and approved. All procedures shall be adhered to." The operation to drain HCU 18 31 was contrary to operating procedure OP 2111, " Control Rod Drive System,' section P, "CRD Accumulator Alarm (CRP 9 5)," which states, "Since this is a common alarm, it is necessary to check local panels 25 4 or 25 22 to determine which accumulator has malfunctioned..." The operation of draining HCU 18 31, rather than HCU 14 31,

was contrary to TS 6.5.A. and was a violation, c.

Conclusions The apparent cause of draining HCU 18 31, rather than HCU 14 31, was inadequate self checking by the responsible auxiliary operator prior to opening the HCU 18 31 accumulator leak off line isolation valve. This event did not constitute a violation of TS 3.3.F and, given the brevity of the event, the licensee's operational response was appropriate. However, the actions required by TS 3.3.F indicate that having two HCU accumulators inoperable within a nine-rod square array is a significantly degraded condition. The personnel error of draining the wrong HCU accumulator was a violation of TS 6.5, " Plant Operating Procedures," and was considered another example of a similar violation cited in inspection report 97 04.

.

.

O ll. Maintentnat M1 Conduct of Maintenance M 1.1 Maintenance Observations a.

lunection Senna (02707)

The inspectors observed portions of plant maintenance activities to verify that the correct parts and tools were utilized, the applicable industry code and technical specification requirements were satisfied, adequate measures were in place to ensure personnel safety and prevent damage to plant structures, systems, and components, and to ensure that equipment operability was verifled upon completion of post maintenance testing.

b.

Observations. Findinas, and Conclusions The inspector observed portions of the scram solenoid pilot valve replacement for

'

hydraulic control unit 1011, conducted June 11.

On June 9, an Instrumant and Controls (l&C) technician noted that the scram solenoid pilot valve (SSPV) assembly for hydraulic control unit (HCU) 1011 was -

making a buzzing noise. As discussed in inspection report 97 04, failure of a rod to scram in April * 997, was determined to be due to binding of one of the SSPV solenoid plungers. Buzzing was demonstrated to be an indicator of incipient failure.

The HCU 1011 SEPV had a history of intermittent buzzing, and had one solenoid plunger replaced during the April 1997 maintenance outage. Given that the vendor's root causu evaluation was still under development, the licensee decided to replace the entire SSPV assembly.

On June 11, reactor power was reduced to 80 percent for a scheduled rod pattern adjustment, as well as to support single rod scram time testing and replacement of the rod 1011 SSPV. Prior to the replacement, rod 1011 was individually scrammed to verify that it had been operable. The resultant scram time was comparable to the original acceptance testing value. The SSPV was replaced and the rod was scram tested satisfactorily.

The inspector noted that the replacement SSPV was on a work cart in the immediate vicinity of HCU 1011, and that there were no foreign material exclusion (FME) covers on the air line connection ports. The work procedure included a requirement for FME covers to be installed upon the completion of verification testing and prior to installation. While the vabio could be considered to be in the process of being installed (and therefore no longer need FME covers), the inspector considered that these covers had been removed primaturely, given that the original valve had not yet been removed. The inspector discussed this observation with the Instrument and Controls (l&C) supervisor, and verified that use of FME plugs was required by the work control document. Mis-use of the FME covers was found to be an example of poor material condition controls. Other material condition

._

_

._

,

.

concerns are described in section M2 of this report. The inspector had no additional Concerns.

M1.2 Euty.cillance Observations

a.

Insoection Scot;g (01720)

The inspectors observed portions of surveillance tests to verify proper calibration of test instrumentation, use of approved procedures, performance of work by qualified personnel, conformance to LCOs, and correct post test system restoration,

,

b.

Observations. Findinas, and Conclusions l

The inspector observed all or portioas of the following surveillance tests:

Heactor Core Isolation Cooling System Quarterly Surveillance, observed June

,

l 4,1997 The test whs completed satisfactorily with exception of vibration data, which did not get recorded due to a defective probe. This was not determined until after completion of testing. The surveillance was re performed and vibration readings were satisfactory. Other than the vibration readings, data from the first run was used as the surveillance of record, since the second run was a warm start.

Vibration readings are taken after the unit has warmed up, so the initial run did not affect the validity of those readings. The inspector observed no discrepancies in the conduct of this surveillance; failure of the vibration monitoring equipment could not have been foreseen, and subsequent actions to obtain the data were appropriate.

High Pressure Coolant injection System Quarterly Surveillance, observed

June 4,1907 The surveillance was completed satisfactorily with exception of verification of proper operation of two check valves in the turbine exhaust drain, which directs drain flow to the torus. This IST requirement is satisfied by not receiving an exhaust drain pot high level alarm; in this case, the alarm was received. The alarm coincides with operation of a high level dump valve which directs the turbine exhaust drains to the gland exhaust condenser, which in turn, goes to the HPCI

'

pump suction. Although the pump was capable of performing its safety function, it was declared inoperable pending correction of the condition. The problem was found to be a mechanical failure in the drain pot. This was corrected, and the HPCI pump was returned to service the following day. The inspector considered that declaring the HPCl pump inoperaole, while maintaining it available for emergency use until the start of corrective maintenance, was appropriate.

APRM GAF adjustment, observod June 11,1997

GAF adjustments were performed in conjunction with the rod pattern exchange, as discussed in section 01.1 of this report. No problems were noted in the conduct of

_

_

.

these adjustments; procedural observations are discussed in section M2.1 below.

Single rod scram time test of rod 10-11, observed June 11,1997

The scram time test was performed prbr to and after replacement of the associated SSPV assembly. The inspector attenced the pre job brief for the initial scram time test, and observed that it was thorough. No problems were observed during the conduct of either the pre-or postsmaintenance tests.

l

"B" emergency diesel generator monthly surveillance, observed June 24,

1997 During the test, the fuel oil trartsfer pump failed to operate when required for normal makeup to the day tank. The B" emergency diesel generator (EDG) fuel oil day tank low level annunciator subsequently alarmed in the control room, and an

,

operator was dispatched to manually operate the transfer pump. When this attempt was also unsuccessful, the "B" EDG was unloaded and secured. The "B" EDG was de:lared inoperable and the licensee entered a seven day limiting condition for operation per TS 3.5.H.1. Investigation revealed that a neutral wire in the control

,

(

panel for the day tank level control circuitry had broken off from its termination lug.

L The w!!e was re-terminated and the pump tested satisfactorily. The "B" EDG was declared operable later the same day, and the monthly surveillance was completed on the morning of Juaa 25. No other problems were noted during the conduct of this surveillance.

The inspectors noted that several of the surveillance activities observed evidenced material condition problems with either the test instrumentation or the equipment

" being tested. Other examples of material condition problems are described in

+

section M2 of this report.

M2 Maintenance and Material Condition of Facilities and Equipment M2.1 Maintenance-Related insoection Observations (62707)

During routine plant tours, the inspector noted the following degraded material conditions:

Five hangers in the containment air dilution-(CAD) system were found.to be

not properly made up. All were loose U-bolt clamps on piping in the area of the two CAD air receivers. An event report was generated (97-0791) and engineering evaluated the CAD system to be degraded, but operable. The deficiencies were promptly corrected, A one-quarter inch gap was found between the end battery cell and the

seismic support rack for one of 16 groups of cells in the "A" un-interruptible power supply (UPS) battery. An event report was generated (97-0781), and an immediate operability assessment was performed which concluded that the condition did not adverselv impact battery operability. The rack was

.

.

. _ _ -

_

~.... _ _ _ _

_

. _ _ _._

_ _ _.

. _.

__

.

'

.

,

promptly adjusted to be snug against the end battery cell, and a walkdown was performed to verify proper adjustment of other seismic battery racks.

• '

Cracked caps were found on the positive terminab of four cells in the UPS batteries (cells 41 and 136 in the A battery, and cells 28 and 191 in the B-

,

battery). This condition indicates deterioration of the terminal's electrolyte seal. However, this condition was not an immediate operability concern.

Work order requests (29619 and 29620) were initiated to replace these cells during the 1998 refueling outage.

Corrosion was noted at the bottom of the high pressure coolant injection

(HPCI) gland seal exhaust fan casing. The absence of significant rust stains or water tracks leading to the area made the inspector suspect that the

,

' source of the corrosion may have been inside the fan casing, and that it had

,

progressed through the casing wall. The licensee initiated a work order request to investigate the condition during the planned HPCI system outage in December 1997.

,

- A conduit to a class 1E junction box (B 504-SI) was missing the first hanger

(unistrut clamp) from the junction box; there were no other hangers in the immediately vich'ity, and the conduit was easily movable by hand. A priority

-

one work order (29538) was generated and the condition was corrected later the same day. However, the inspector observed that no event report was-initiated and system operability considerations were not addressed.

A class 1E conduit (11186E Sil) was missing two hangers (unistrut clamps)

' and was easily movable by hand. The licensee promptly determined that the

'

'

" conduit run contained only an inactive circuit (power to the RCIC condensate

= pump motor heater)/ however, a work order was initiated to re install the hangers.

The above inspector observations cumulatively represent a failure of the plant staff to assure that conditions adverse to quality are promptly identified and corrected via

an established quality assurarice process. The inspectors noted that for some of the conditions involving potential seismic qualification concerns, the licensee stated in Event Report (ER) No. 97-0785 that, "...the seismic qualification program does not nave a clear identification of what components are qualified, nor does it specify requirements on how to properly maintain that qualification." The failure to have +

identified and promptly corrected the material deficiencies and non-conforming conditions stated above, was contrary to 10 CFR 50, Appendix B, "_ Quality-Assurance Criteria for Nuclear Power Piants and Fuel Reprocessing Plants," Criterion XVI, " Corrective Action," and was a violation. (VIO 97-05-02)

M2.2 Broken Conduit for RHR System Annunciator Circuit a.

Insnection Scope, Observations, and Findinas (62707)

During a routine inspection in the torus room on July 7, the inspector noted a

-..

-

.

.

'

.

conduit that was pulled out of the wall, exposing about six inches of the contained conductor. The conductor was not under tension, however, the first conduit hanger from the disconnect was not made up, and the potential for disruption by a seismic event existed. The inspector traced the conduit to the motor operators for "B" and

"D" residual heat removal (RHR) pump torus suction isolation valves. Tt.e inspector reported the condition to the shift supervisor; in addition, to avoid confusion over its location, the inspector pointed out the deficient condition to an auxiliary operator.

On July 8, the inspector noted that a work order request had been generated concerning the conduit in question. During review of new work order requests at the morning planning meeting, the licensee recognized this as a potential RHR system operability issue and the systems engineering staff was tasked to examine the condition. The inspector was concerned that the core spray (CS) system surveillance testing was to be performed later that day and were the conduit determined to affect RHR system operability, this would potentially result in two loops of low pressure emergency core cooling systems (ECCS) being unavailable.

Accordingly, following the morning planning meeting, the inspector discussed this concern with licensee management.

i Following completion of the CS system surveillance testing on July 8, the inspector noted that the conduit deficiency still existed. After discussion with the licensee, it was determined that the licensee had not yet examinea the specific problem that had been pointed out by the inspector. The correct conduit had been examined prior to the CS system surveillance test, but in a different location. A minor deficiency at that location had been evaluated as not affecting circuit operability.

VY initiated events reports (ERs) on both the inspector identified material

.

discrepancy and the delay in initiating an ER that led to the condition not being

-

piomptly evaluated, in addition,-an operations department night order book entry was made, addressing the mis-identification and failure to initiate an ER, and the issue was discussed by the operations department manager with all shift operating personnel. The conduit was subsequently determined to be non-nuclear safety (NNS) class, supplying inputs to an alarm circuit that does not affect RHR system operability; therefore, seismic qualification was not applicable. VY determined that the condition did not affect the ervironmental qualification of the associated RHR valves. The wall penetration was determined to be missing a fire seal, however, it was not an Appendix R or TS required fire barrier. This deficiency was compensated for by an existing fire watch, c.

ConclyJ ons i

The deficient electrical conduit in the RHR system was not properly entered into the corrective action program. Although the affected circuit was subsequently determined to be classified non-nuclear safety and have no impact on system operability, the potential existed that a degradation of the low pressure emergency core cooling system capability could have occurred. Licensee response to this latter issue was found to be agpropriat :

.

M3 Maintenance Procedures and Documentation M3.1 Nuclear instrument Procedures That Affect Both Channels of the Reactor Protection System a.

Insoection Scoce (61726)

While observing the rod pattern exchange on June 11, the inspector noted that the procedure (OP 4400) for changing the average power range monitor (APRM) gain adjustment factor (GAF) resulted in two APRM channels being simultaneously placed in bypass. The two bypassed APRM channels input to the two channels of the reactor protection system (RPS). The inspector was concerned that this practice was not consistent with proposed corrective action for an earlier plant event, b.

Observations and Findinas As described in inspection report 97-04, a reactor scram occurred on April 24, 1997, in part due to testing that involved two RPS channels simultaneously. Based on the inspectors concern, the licensee performed a root cause analysis to l

understand why prior corrective actions may not have been fully effective. The

!

licensee's evaluation of the inspector concern indicated that the lessons learned from the past calibration error, had not been factored into Reactor Engineering's calibration procedures. The previous error involved procedures used by a different organizational department. The corrective actions at that time had not included a review of similar tasks performed by the Reactor Engineering group though. The-(

evaluation recommended additional corrective actions included a review of all

- Reactor Engineering proceduras that have the potential to cause a half or full scram, if performed incorrectly. However, it also recommended making such procedures

" Continuous Use" procedures, rather than ensuring that only one channel of instrumentation was tested at a time. At the time of this observation, OP-4400 had been revised to require continuous use, in subsequent discussions of this matter, the licensea indicated that it was nonetheless their intent to revise procedures such that only one channel of instrumentation was being worked on at a given time. The licensee pointed out that the basis for bypassing APRM channels during GAF adjustments was different than during calibrations. During calibration, the APRM must be bypassed to prevent test signals from generating trips of the associated RPS channel, however, since GAF adjustments are performed with the instrument in the " Operate" mode, the APRM is bypassed to prevent personnel error, while making this adjustment, from generating i

an RPS trip signal. The reason that two APRMs are bypassed is that there are two l

channels in each APRM cabinet. The second APRM (the other channel in the cabinet) is bypassed against the possibility that the technician may adjust the wrong channel, in either case, the consequences of an error would be immediate if the

'

APRM was not bypassed. Having the APRMs bypassed prevents an acknowledged error (such as the screwdriver slipped while making the adjustment) from causing a half scram, and affords the opportunity for the error to be caught by operations

t

.

personnel before it can cause a more significant problem. The inspector considered this basis and proposed action to be reasonable.

The licensee indicated that the LPRM calibration procedure that was being performed at the time of the April 24 reactor scram had b6an revised. Now, such calibrations affect only one APRM channel at a time. The inspector verified that this procedural revision was in place. The licensee projects that the review and revision of additional procedures that have the potential to cause a half scram or full scram I

will be completed by the end of the year.

c.

Conclusions As a result of the April 24,1997 reactor scram root cause evaluation, procedures were being revised to perform instrument calibrations singly, in cases where an error could result in a half scram or full scram. A specific case where two APRM channels continue to be bypassed simultaneously was adequately justified.

M8 Miscellaneous Maintenance issues M8.1 Partial LPRM Calibration of Aoril 24,1997 a.

Elackoround and Insoection Scoce (92902)

The local power range monitor (LPRM) system consists of 80 miniature fission chamber-type neutron detectors that are positioned at various fixed locations in the reactor core. The LPRMs provide indication of local neutron flux to the average power range monitor (APRM) system, the rod block monitor (RBM) system, and the

'

"e < process computer.' LPRMs are grouped by axial and radial location to provide a' ~

representative indication of neutron flux to the six APRM channels. The APRMs-provide indication of core average thermal power, and also input to the reactor protection system. The RBM system develops indication of average local power from LPRMs around a selected control rod and prevents withdrawal of that rod when local power is above a preset limit. LPRM inputs to the process computer are used to develop core thermal performance indicators to verify that core thermal performance is within established limits.

LPRMs must be calibrated periodically due to depletion of the fissile detection media. Calibration data is obtained from the traversing in-core probe (TIP) system, which uses moveable neutron detectors to measure the in-core flux distribution.

LPRM gain adjustments are derived from the TlP data; these adjustments are made during the LPRM calibration procedure, and are subsequently verified using the TIP system. LPRM calibrations can only be performed while the reactor is operating at power, sint a the minimum sensitivity of the LPRM detector is on the order of one percent reactor power.

On April 24,1997, a reactor scram occurred due to cperator error while performing LPRM calibrations. This surveillance is required by Technical Specification (TS) 4.1.A, and is performed during power operations at a frequency not to exceed 1000 an w

.

....

.

'

.

effective full power hours (EFPHs). Since the surveillance of April 24 was only partially complete at the time of the scram, the inspector reviewed the licensee's actions to resolve the partial completion of this surveillance and to verify compliance with TS 4.1.A.

b.

Observations and Findinos The inspector reviewed the calibration data sheets from operating procedure OP-4406, "LPRM Calibration and Functional Check," performed on April 24,1997.

This record indicated that gain adjustments weie performed on LPRMs that input to average power range monitor (APRM) channels A and D; LPRMs in the remaining four channels of APRMs had not been adjusted. Since the LPRM calibration could not be completed with the reactor shut down, the licensee backed out of the calibration procedure by resetting the gains of the affected LPRMs to their original values (that is, the values that they were set at prior to adjustment on April 24).

This was done under work order (WO) control. The inspector reviewed WO Nos.

97 03204 and 97 03205, "Miscalibration of APRM A (and -D)." The inspector verified that: (1) all of the LPRMs that were adjusted during the aborted April 24 calibration were readjusted under these WOs, and (2) the "as lef t current" values recorded in these WOs matched the "as found test currents" as recorded in the April 24 calibration data sheets. The LPRM readjustments were completed on April 24.

Following plant startup on May 7 and return to full power operation, LPRM calibrations were tuccessfully completed on May 19. The reactor core exposure at that time, as recorded in the TIP system operating log, was approximately 4000 EFPH. Core exposure at the time of the previous LPRM calibration, performed on -

March 25, was approximately 3100 EFPH. The inspector concluded that the LPRM

'

calibration of March 25 was still valid on April 24, and therefore, that the decision to restore from the partial LPRM calibration by reestablishing the March 25 gain settings was appropriate.

c.

Conclusions Licensee actions wers appropriate in re-establishing the pre-calibration gain settings of the LPRMs following the partial surveillance performed on April 24,1997.

Following the return to power operations, LPRM calibrations were properly completed within the minimum frequency requirement of TS 4.1.A.

111. Enaineerina E1 Conduct of Engineering E1.1 Residual Heat Removal and Core Sorav Systems Minimum Flow issue a.

Backaround and Inspection Scoce (92903)

An NRC Architect / Engineering (A/E) Team review of the residual heat removal (RHR)

___ _ _ _

.

system identified that the current RHR pump minimum flow rate of 350 gpm was significantly less than the pump manufacturer's recommended 2700 gpm flow rate.

The pump manufacturer had communicated this criteria to the licensee via a letter dated November 13,1986. The 1986 recommendation specifically limited short-term minimum flow (2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> or less in a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period) to 2075 gpm and continuous minimurn flow (in excess of 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />) to 2700 gpm. Similarly, the core spray (CS) pump minimum flow rate of 300 gpm was determined to be significantly less than the currently recommended values of 900 gpm and 1275 gpm for both short term and continuous minimum flow.

Renewed discussions with the pump vendor during this inspection period, resulted in a May 21,1997 letter to the licensee that stated a minimum flow of 350 gpm for 30 minutes, for a one-time plant event, was supported by the vendor (Sulzer Bingham Pumps Inc.). Based upon the above, the VY staff verbally concluded that the RHR pumps remain operable (capable of fulfilling their safety function during design basis accident conditions). Contributing to this determination was that, to date, the pumps have experienced only short duration (less than one minute)

minimum flow conditions at 350 gpm during routine surveillance testing and operational use (torus cooling and shutdown cooling initiation). In addition, review of station operating procedures identiHed that the RHR pumps are secured, if not l

required to maintain reactor vessel level, or the pump discharge flows are increased

'

beyond minimum flow requirements if realigned for torus cooling. Lastly, the licensee initiated the Basis for Maintaining Operation (BMO) process to make a formal operability determination and to establish a corrective action plan for long-term resolution of this issue.

A/E Team characterization of this inspection finding and the apparent long-standing-design issue will be documented via inspection report 50-271/97 201.. During this inspection period, the inspectors observed the VY staff's dispositioning of this A/E Team finding, b.

Observations and Findinas The inspectors noted that an Event Report (ER) had not been generated until June 6 for this issue. Although, Operations Department instructions (the non-intent procedure change process) were implemented on May 22,1997 for both OP-2124,

"RHR System," Revision 42, dated April 24,1997, and OP-2123, " Core Spray System," Revision 26, dated October 18,1996. The prccedures precautions were revised to minimize operation of the RHR and CS pumps in the minimum flow mode and to establish system flow rates of 2700 gpm and 1275 gpm, respectively, l

promptly after planned pump starts. OP 2123 and OP-2124 previously had precaution statements which limited minimum flow at these values to a maximum of two hours. The inspector determined that these precautions were added to the operating procedures in 1992.

On June 6,1997, Event Report No. 97-0694 was initiated and, based upon another A/E Team observation involving the accuracies of the RHR and US pump flow instrumentation, recommended caution tags be affixed to the RHR and CS pump

_

7. ~

'

.

switches in the control room. The caution tags addressed the e=tablishment of 4,000 gpm and 1,750 gpm, RHR and CS systems, respectively, vi minimum flow due to flow instrument inaccuracies. These caution tags were subsequently revised on June 9,1997 to establish these flow rates within 30 minutes based upon the May 21,1997 letter from the pump vendor.

Based upon inquiries made by the inspectors concerning the basis for RHR and CS system operabluty, the BMO process was initiated for this issue on June 11. The BMO guidelines establish a seven-day period for the completion of the BMO, unless the issue warrants a more timely completion (high safety significance)

commensurate with a more restrictive limiting condition for operation (LCO) action statement. The inspector learned that the engineering staff ran into difficulties in completing their reviews by June 18 and requested (and received) an extension from the plant manager to June 20.

l The inspectors observed the Plant Operations Review Committee meeting held on

,

June 20,1997 to review, in part, the proposed "RHR and CS pump minimum flow"

!

BMO No. 97 29. The inspectors witnessed good discussions by the PORC members in probing the completeness and accuracy of the draft document. The inspectors noted that the PORC rejected the evaluation because of inadequate information to support the conclusions.

On June 25, the NRC staff held a conference call with the VY staff and management to discuss the progress being made in resolving the RHR minimum flow issue, and in particular, to understand the delay in approving the written BMO.

The licenaee reviewed the chronological sequence of events in addressing this issue and summarized tneir current basis for concluding the RHR and CS systems remain operable. The basis included credit for: (1) emergency operating procedures which provide adequate guidance for RHR and CS pump operation during design basis accidents scenarios; (2) the caution tags affixed to the pump switches on June 6 (revised on June 9) provide appropriate interim operations guidance and, (3)

Department instructions implemented on May 22,1997, provide adequate precautions to minimize RHR and CS pump minimum flow pump operation. The BMO was subsequently approved on June 27.

c.

Conclusions VY was slow to initiate the BMO process after concerns were identified with RHR-and CS system minimum flow requirements. Development of the BMO took longer than was targeted by licensee guidance. Although some delay was the result of concerns that were appropriately raised by PORC, the time required for development did not appear to be commensurate with the safety significance of the issue. This issue will be discussed further in inspection report 50-271/97-20.

'

.

E1.2 Egismic Deslan Considerations for the 24 Volt DC Power System a.

Insoection Scoce (71707,37551)

During a routine plant tour, the inspector noted that the 124 volt DC (VDC) power system batteries were not seismically mounted. Since batteries are generally only included in safety related (and therefore seismically qualified) power systems, the inspector questioned whether this condition was acceptable, b.

Observations and Findinas The i24 volt DC (VDC) power system provides power for the neutron monitoring system source and intermediate range instruments, and for portions of the process

,

!

radiation monitoring system. It consists of two subsystems, A and B, each of which is made up of two 12-celllead acid storage batteries and two battery chargers. The subsystems are normally powered from opposite trains (SI and Sil) of the safety related AC power distribution system. FSAR, Section 8.7, "i24 V DC j

Power System," does not directly indicate whether the system is safety related or

~

non-nuclear safety (NNS); however, NNS is implied, because it discusses only the power generation (as opposed to safety) objective and design bases. Although the FSAR indicates that, "the battery racks are designed to withstand design basis earthquake effects," it does not discuss the seismic requirements of the batteries themselves. However, the inspector noted that the FSAR description of battery mounting in the ECCS 24 VDC power system (a safety class electrical (SCE)

system, for which % batteries are seismically mounted) makes the exact same statement.

~The' bases for requiring a power system to bo safety class electrical are 1) that its supplies power to a safety system, and 2) that power is required to achieve the safety function of that system. The intermediate range neutron monitoring subsystem is a safety system, because it generates a reactor trip signal to prevent fuel damage resulting from abnormal operational transients that occur while operating in the intermediate power range. However, it also generates a reactor trip signal if it is de-energized; because it fails safe on loss of electrical power, the intermediate range neutron monitoring subsystem does not require an SCE power supply. The source range neutron monitoring subsystem is used for indication only, and is not classified as an SCE system.

The process liquid radiation monitoring system receives power from the i24 VDC power system. FSAR, Section 7.12.4 states that the safety design basis of the system is to "... provide a clear indication to operations personnel whenever the radioactivity level in the stream reaches or exceeds pre-established operational limits for the discharge of radioactive material to the environs." As indicated in the FSAR definitions, a safety design basis indicates that the system is a safety system.

Since power is required for the system to perform its safety function (indication),

the power system should be safety class electrical; therefore, the i24 VDC power i

system should be seismically qualified. However, the safety evaluation portion of FSAR, Section 7.12.4 states, "The Process Liquid Radiation Monitoring system

'

--

-

. -

... - - - - -. - - -..

-. - - - - - -

.

-

,

'

i i

.

instrumentation has been classified as NNS per the Vermont Yankee Safety Classification Manual. This instrumentation is not relied upon to accomplish any safety functions."

c.

Conclusions inspector review of the 24 VDC system identified that Section 8.7 of the UFSAR

,

does not include the safety system designation of the 24 VDC system as non-

'

nuclear safety (NNS). Based on the initial review, the inspectors could not

. determine if the system is appropriately classified. This item remains unresolved

'

!

pending further review of the design basis documentation. (URI 97 05-03)

E1.3 Interface of a Retired in-Place System with the Hioh Pressure Coolant inlection System a.

Insoection Scooe (71707,37551)

The torus /drywell pumpback system was an original plant system that was used to establish differential pressure between the torus and the drywell. When the

-

'

containment nitrogen inerting system was subsequently installed, tho torus /drywell

'

pumpback system was no longer required, and was retired in place. The inspector L

examined the system to determine the adequacy of this process.

b.

Observations and Findinos

The inspector noted that the torus /drywell pumpback system three-inch diameter suction pipe connected to the high pressure coolant injection (HPCI) three inch

c.

c,~ %

. diameter gland exhauster discharge to the standby gas treatment system. Given*

that this pipe connects two safety systems, the inspector considered it likely that it

,

was a safety class pipe. The substantial piping supports on this line suggested that I-it was a safety class line, and therefore would c*so be seismically qualified.

,

However, the inspector noted that the pumpback system was not seismically qualified. Specifically, the system pump (RRU-18) was not fastened to the floor,

'

and a capped section of smaller piping near the pump was not supported and could be easily moved by hand. The inspector was concerned because there is no

' isolation valve between RRU-18 or the unsupported smaller pipe, and the pumpback

< system connection to the HPCl/SGT line (3" HPCI-13). Isolation is provided by a blank flange on the RRU-18 discharge, and a plug on the end of the smaller pipe.

The inspector checked the piping and instrumentation diagrams (P&lDs) to determine the safety class of torus /drywell pumpback system and the HPCl/SGT line. The HPCI system P&lD (G191169) shows the HPCl/SGT line as safety class 3, and has no class break for the connecting line to the torus /drywell pumpback system. However, the primary containment and atmosphere control system P&lD (G191175) shows the HPCl/SGT line as NNS, and shows no safety classification for the torus /drywell pumpback system. VY initiated an ER (97-0821) to investigate this inconsistency. During initial investigation, VY noted additional inconsistent information concerning the safety classifications of this equipment. At the close of

.

.

the inspection period, the event report investigation was stillin process, c.

Conclusions The inspector was not able to determine the safety classification of the HPCl/SGT line and the torus /drywell pumpback system. If the HPCl/SGT line is safety class 3, then it appears that the connection of the retired-in-place torus /drywell pumpback system is potentially inadequate and, that the system is not being maintained as safety class 3 and is therefore, potentially not isolable from the HPCl/SGT line.

Pending determination of the safety classifications of this equipment, this issue remains unresolved. (URI 97 05-04)

E1.4 (Uodate) URI 97-03-02: Electrical Cable Seoaration Review

>

a.

Backoround and Insoection Scoce (37551)

As documented in inspection report 97 03, the VY staff identified a number of electrical cable routings which were in conflict with VY Updated Final S:faty Analysis Report (UFSAR), Section 8.4.6, " Cable Installation and Separation Criteria." To support continued full power operations of the facility, the VY staff initiated the Basis for Maintaining Operation (BMO) process to establish the technical basis for their safety system operability determination and to document their corrective action plan per 10 CFR Part 50, Appendix B.

BMO No. 97-13, " Separation of Lighting Panel Feeder Cables," was completed and reviewed by the Plant Operations Review Committee (PORC) on April 9,1997 and approved for use by the Plant Manager on April 10. It was subsequently revised

(Revision'1) on April-28,1997 to encompass 67 additional non-nuclear safety (NNS)

instrumentation-and control cables that were similarly routed in Si and Sil divisional cable trays and the title of the BMO was revised to " Separation of NNS Power and Control Cables." On May 2, the licensee identified (Event Report No. 97 0470) two additional NNS cables routed in Si and Sil manholes in the switchgear room. These additional electrical cable soparation discrepancies were incorporated into Revision 2 of BMO No. 97-13, dated May 12,1997.

The inspector examined BMO No. 97-13 to assess the adequacy of the technical basis for continued safe plant operations and the supporting VY safety evaluation.

(No. 97-14, "50.59(a)(2) Safety Evaluation for BMO 97-13").

b.

Observations and Findinas As previously documented in inspection report 97-03, section E1.1, the VY staff promptly evaluated these non-conforming cable separation conditions-for operability and reportability. The inspector considered the VY staff's reportability assessments per 10 CFR 50.72 to be timely and proper. However, the operability assessments for these electrical separation issues have evolved since initial discovery on March-27,1997.

I

.

'

.

Ooerability Assessment Review The 480V feeder cables from motor control center (MCC) 8A to LP-1SR and LP-N/E-1 A (cables C40232A and C40245A, respectively) were assessed for operability in Revision 0 to BMO No. 9713. Cable C40232A was de-energized and protective tagged because its lighting panelloads were non-essential for power operations.

This action alleviated any immediate operability impact. Both cables were original installation and did not have two isolation devices in series "right after nne another,' as recommended in the original construction EBASCO Separation Criteria (Revision 3, dated June 1971) for non-safeguards power cables run in Si and Sil raceways. The two protective devices in series are designed to prevent a single active failure from affecting the isolatinn function. Following discovery, cable C40245A was maintained energized because of the essential balance of plant lighting circuit loads. The VY staff credited the safety class electrical feeder breaker at MCC 8A and the supply breaker to MCC 8A for interim e!ectrical fault and fire propagation protection until cable C40245A could be re routed. The VY staff also credited the East / West switchgear room wall as a one-hour fire barrier, as both cables penetrate through this wall before crossing into the other divisional safety class electrical cable tray. The inspector found this interim operability sssessment adequate and noted that the licensee initiated actions (commitment tracking item BMO 9713-01) to re-route these cables as soon at practical. This operability assessment and corrective action plan was consistent with Generic Letter (GL) 91-18 and 10 CFR Part 50, Appendix B.

The initial operability assessment for the 67 NNS cables, which was documented in the associated ERs, was technically based upon the fact that these NNS cables were 120 VAC, or less, instrumentation and control cables installed during original construction. ' These indication,-annunciation,- and control circuits were identified to

'be protected by at least two devices such as a breaker and fuse, two breakers in series, or a fuse and control power transformer. The low voltage and amperage signal carried by these circuits was considered below that necessary to cause damage via short circuits to adjacent cables.

NRC staff review of the initial operability assessment for these 67 NNS cables identified concerns with the completeness of the VY staff's assessment of the existing isolation devices. Specifically, the NRC staff was concerned with the crediting of non-safety related protective tripping devices. During a telephone

_

conference call on April 25,1997 between the NRC and VY staffs, the licensee agreed to conduct a more detailed analysis to ensure that at least two safety class, or equivalent electricalisolation devices were available for each cable. Where upstream MCC feeder breakers were relied upon, the analysis should include an evaluation of the MCC's electricalloads to ensure safety related systems or components were not adversely impacted by the tripping of that safety class breaker. The summary of this detailed analysis was documented in BMO 97-13, Revisions 1 and 2, and resulted in the re-routing of eight cables prior to restart from the forced outage in April 1997. The balance of the 67 NNS cables were recommended to be re-routed at a later time to comply with the VY licensing basis.

The inspector confirmed the re routing of the eight cables (C1102C, C112312,

_

'

.

.

C1600D, C1603F, C1981U, C1232D, C1220E, and C1226N) in accordance with the station Work Order and plant design change request procedures.

The operability assessments and the completed and planned corrective actions documented in BMO 97-13, Revisions 1 and 2 wore appropriate and in accordance with GL 91 18 and 10 CFR Part 50, Appendix B.

Safetv Evaluation Review '

in accordance with VY Operability Determination / Basis for Maintaining Operation Guideline, Revision 5, dated January 1996, the VY staff is obligated to determine if a 10 CFR 50.59 Safety Evaluation is required as an initial step in the BMO process.

To appropriately make that determination, the VY staff conducts a "10 CFR 50.59(a)(1) screening evaluation" per Administrative Procedure (AP) 6002, Revision 5, dated April 11,1997. A detailed safety evaluation, if necessary, is referred to by the VY staff as a "10 CFR 50.59(a)(2) Safety Evaluation." The 50.59(a)(2)

evaluation is used to determine if the licensee can make the change to their UFSAR without prior approval of the NRC and if the change involves a Technical

.

Specifications (TS) change or an Unreviewed Safety Question (USO). By answering

"Yes" to any of the questions in the 10 CFR 50.59(a)(1) screening evaluation, the licensee is obligated to conduct the 10 CFR 50.59(a)(2) safety evaluation.

The inspector determined that Revision O to BMO 9713 did not result in a 10 CFR 50.59(a)(2) Safety Evaluation. However, the 10 CFR 50.59(a)(1) screening evaluation credited compensatory actions (protective tagging the LP 1SR feeder

}

breaker) to answer "No" to a number of the screening quesdons. The inspector viewed the protective tagging compensatory measure as a possible temporary

"

modification to the facility, pending a permanent change by re-routing-the cable" As such, this compensatory measure may have warranted a 50.59 evaluation.

Discussions with station representatives and review of the affected circuit clearly indicated that the de-energization had no adverse safety impact. The inspector subsequently verified that Revisions 1 and 2 to BMO 97-13 both included a 10 CFR 50.59(a)(2) safety evaluation (No. 97-14) which concluded no TS change or USO l

was introduced with the identified non-conforming cable separation condition and l

the compensatory measures taken.

NRC staff review of Safety Evaluation No. 97-14 determined that the VY staff may have incorrectly concluded that there was no USQ for the non-conforming cables-not yet re-routed. The non-conforming cable separation conditions, alone, conflict with the operating license and UFSAR, and therefore, may render the plant outside established safety analysis'and margins. However, the valid operability assessrnent is sufficient basis (per Generic Letter 91 18) to justify continued planc operation, provided the non-conforming condition is promptly entered into the licensee's 10 CFR 50, Appendix B corrective action program and resolved in a timely manner.

The NRC staff determined that the compensatory measure taken to ensure the feeder cable to LP-1SR remained de-energized should have more appropriately been supported by a 10 CFR 50.59(a)(2) evaluation in BMO 97-13, Revision 0, because this was a deliberate change made to the facility to address (on an interim basis) the

___

-_-_-

_

_

.

'.

non-conforming cable separation condition, c.

Conclusions VY's implementation of the BMO process for dispositioning the non conforming cable separation conditions was appropriate and the operability determinations were adequately supported. However, the inspector viewed the current process as weak, noting the potential existed to inappropriately perform 10 CFR 50.59 reviews. The NRC staff's review of the licensee's corrective actions and interim compensatory actions concluded that they were satisfactory (reference NRC memorandum, dated June 10,1997,in response to Task Interface Agreement, " Technical Assistance in Support of Vermont Yankee Cable Separation Problem TAC No. M98629", dated May 30,1997). However, this unresolved item remains open pending NRC review of VY's root cause evaluation and long term corrective acticns for the cable separation non-conformances.

E7

= Quality Assurance in Engineering Activities E7.1

. Review of Licensee-ldentified Engineerina lssues

,

l The licensee's Design Basis Documentation (DBD) and improved Technical Specifications (ITS) projects have the potential for identifying inconsistencies

!

between the design, licensing, and operating bases of plant structures, systems,

!

and components. Such inconsistencies will be documented in this section of the report and tracked to resolution as inspection follow items.

HPCI Suction Aooendix J Proaram Weaknesses

• -

e a

~The primary containment barrier for the HPCI suction line is the first isolation valve from the torus, V23 58. The licensee identified that, in some situation, the condensate storage tank suction check valve, V23-32, may also be required to act as a primary containment barrier. However, V23-32 is not included in VY's Appendix J (Primary Reactor Containment Leakage Testing for Water-Cooled Power Reactors) program. VY developed BMO 97 28 to justify interim operation while the issue was being resolved. No compensatory actions were required as a result of the BMO. (IFl 97 05-05)

Primarv Containment Coatino Qualification issue

The licensee determined that some of the original coatings used to line the interior of the primary containment may not be fully qualified. These coatings could peel off during a design basis LOCA and could potentially clog the ECCS strainers in the torus. VY developed BMO 97-31 to justify interim operation while the issue was being resolved. No compensatory actions were required as a result of the BMO. (IFl 97 05-06)

_.

.. -.

i

-

.

..

E8 Miscelleneous Engineering Issues E 8.1 - (Closed) Unresolved item (96-08-01): East and West switchgear room carbon dioxide suppression system design deficiency.

This unresolved item tracked the licensee's resolution of the failure of the East and West switchgear rooms' high pressure carbon dioxide (CO2) fire suppression systems to demonstrate (via testing) adequate deep seated fire suppression capability.' As previously discussed in inspection reports 96 03, 96-08, and 9611, a design change was being proposed to resolve this design issue. Inspector follow-up determined that the high pressure CO2 banks located in the East and West switchgear rooms were replaced with a single low pressure CO2 tank and all new associated actuation valves and distribution manifold. The old high pressure detection and actuation circuits were rnodified and used for the new low pressure system. The inspector verified that the old high pressure systems remain

- functional, but via manual actuation only. The inspector noted that the West bank of high pressure CO2 still provides second discharge (back-up) for the cable spreading room.

The inspector verified that the applicable operating procedure (OP-2186, Fire Suppression Systems) was appropriately revised and operators were trained on the revisions pertaining to the new low pressure switchgear CO2 suppression system.

l-The inspector's review included a walkdown of the new system with a licensed operator. Currently, a Technical Specification (TS) hourly compensatory firewatch j_

remains in affect until the low pressure system is recognized in the plant's Technical l

Specifications via an amendment. The inspector confirmed that the submission of a L'

'TS amendment request to revise the high pressure to low pressure system

'

~

L operability and surveillance requirements is pending.- The licensee had previously=

withheld this amendment request, pending submission of their request to remove the entire Fire Protection Prngram requirements from TS, in accordance with Gener!o Letter 86-10.

As documented in VY's voluntary Licensee Event Report (LER) No. 96 20, dated October 17,1996, the f ailure of the high pressure CO2 fire suppression system to achieve 50 percent CO2 concentration and maintain greater than or equal to 44 percent concentration for a minimum of 30 minutes (for suppression of a postulated

_

deep-seated fire in the switchgear rooms), is contrary to National Fire Protection Academy (NFPA):12-1977 requirements, as stipulated in the VY Fire Protection Program and required by 10 CFR 50, Appendix R, and BTP APCSB 9.51 and Appendix A. The safety significance of this design deficiency was low. This conclusion was based upon the fact that in the event of a postulated deep-seated -

fire in the switchgear, the high pressure CO2 system would still have initiated and the plant fire brigade would have responded to ensure proper extinguishing of the fire. Licensee corrective actions have been appropriate (including the compensatory fire watches) and timely, given the effort necessary to completely replace the high pressure system with a properly designed, installed, and satisf actorily tested low pressure system which satisfies NFPA Code requirements. Accordingly, this

.

.

,

licensee identified and corrected violation is being treated as a Non Cited Violation, consistent with Section Vll.B.1 of the NRC Enforcement Poliev This unresolved item is closed.

E8.2 (Closed) Insoection Follow Item (96-09-03): Residual heat removal pump start interlock problem.

,

This inspector follow item involved the licensee's discovery on St 7,1996 that the limit switch settings for the "D" residual heat removal (RHh, pmp shutdown cooling suction motor operated isolation valve (RHR V1015D) were improperly set. As documented in LER No. 96 21, dated October 7,1996, the licensee's root cause evaluation led them to conclude that the lack of sufficient procedural controls to ensure proper limit switch settings may have potentially impacted (common cause failure) several other safety related motor operated valves

.

(MOVs) which had recently undergone similar maintenance. Subsequent VY staff review of the other susceptible MOVs' identified that no other limit switches were

- improperly set.

inspector follow up concluded that this event was of low safety significance because the failed start of the "D" RHR pump due to the limit switch setting error, did not impact the RHR low pressure coolant injection safety function. The other potentially affected MOVs were found with alllimit switch settings proper, and therefore their safety functions were likewise unaffected. The licensee did take prompt and thorough corrective actions to identify the cause and address the procedural deficiency, including the completion of maintenance personnel training.

Lastly, the absence of any similar MOV limit switch setting problems indicated that this not a recurring problem. This licensee identified and corrected violation is being

" '

' treated as a Non-Cited Violation,- consistent with Section Vll.B.1 of the NEQ m

Enforcement Poliev. -This inspection follow item is closed.

IV. Plant Suonort P3 Emergency Procedures and Documentation P3.1 In Office Review of Licensee Procedure Chanaes An in-office review of revisions to the emergency plan and its implementing procedures submitted by the licensee was completed..The specific revisions reviewed were:

OP 3542, Revision 14 Emergency Actions to Ensure initial Accountability and Security Response OP 3536, Revision 1 In-Plant Air Sample Analysis With Abnormal Conditions Based on the licensee's determination that the changes do not decrease the overall effectiveness of the emergency plan, and that it continues to meet the standards of 10 CFR 50.47(b) and the requirements of Appendix E to Part 50, NRC approval is r

.

( /

'

,

'

.

not required for those changes. Implementation of those changes will be subject to inspection in the future.

V.-

Management Meetings X1 Exit Meeting Summary The inspectors met with licensee representatives periodically throughout the inspection and following the conclusion of the inspection on July 30,1997. At that time, the puipose and scope of the inspection were reviewed, and the preliminary findings were presented. The licensee acknowledged the preliminary inspection findings.

X3 Review of Updated Final Safety Analysis Report (UFSAR)

l A recent discovery of a licensee operating its f acility in a manner contrary to the UFSAR description highlighted the need for a special focused review that compares r

plant practices, procedures, and parameters to the UFSAR description. While perfonning the inspections discussed in this report, the inspectors reviewed the applicable portions of the UFSAR that related to the areas inspected. Discrepancies that were noted were documented in the applicable section of the above report.

l

..

.

- _ _ _ _ _

',

INSPECTION PROCEDURES USED 62707 Maintenance Observations 61726 Surveillance Observations 71707 Plant Operations 37551 On Site Engineering 92901 Follow Up - Plant Operations ITEMS OPENED AND CLOSED OPEN VIO 97 05-01 Violation of TS 3.7.A.7.b VIO 97-05 02 Failure to implement appropriate mecsures to identify and correct degraded and nonconforming conditions.

URI 97-05 03 Seismic design considerations for the,24 volt DC power system URI 97-05-04 -

Interface of a retired-in-place system with the high pressure coolant injection system IFl 97 05-05 HPCI suction Appendix J Program weaknesses IFl 97-05 06 Primary containment coating qualification issue CLOSED URI 97-04-02 High oxygen concentration in the Torus air space during plant operations URl-96 08-01 -

East and west switchgear room CO2 supression system design

-

deficiency

- IFI 96-09-03 Residual heat removal system pump start interlock UPDATED URI 97-03-02 Electrical cable separation review

,

.

t

. ',

PARTIAL LIST OF PERSONS CONTACTED

' G. Maret, Plant Manager F. Helin, Tech. Services Superintendant E. Lindamood, Director of Engineering M. Balduzzi, Operatlons Superintendent K. Bronson, Operations Manager

_

M. Watson, Piialntenance Manager M. Desilets, Radiation Protection Manager R. Gardi's, Chemistry Manager G. Morgan, Security Manager C. Nichols, Electrical and Controls Maintenance Manager

,

LIST CF ACRONYMS USED BMO Basis for Maintaining Operation CFR Code of Federal Regulation CR control room CS core spray EDG emergency diesel generator ER Event Repon GL Generic Letter-HPCI high pressure coolant injection IFl Inspector follow item IN Information Notice LCO Limiting Condition for Operation LER--

Licensee Event Report LPCI low pressure coolant injection MCC motor control center NRC Nuclear Regulatory Commission NNS Non-nuclear safety PORC Plant Operations Review Committee QA Quality Assurance RHR residual hcat removal-RP-radiation protection -

TS Technical Specifications UFSAR Updated Final Safety Analysis Report URI unresolved item VY Vermont Yankee i

. _ _ _ _ l