ML20140G829

From kanterella
Jump to navigation Jump to search
Insp Repts 50-295/97-02 & 50-304/97-02 on 970206-0402. Violations Noted.Major Areas Inspected:Operations,Maint, Engineering & Plant Support
ML20140G829
Person / Time
Site: Zion  File:ZionSolutions icon.png
Issue date: 06/04/1997
From:
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:
Shared Package
ML20140G777 List:
References
50-295-97-02, 50-295-97-2, 50-304-97-02, 50-304-97-2, NUDOCS 9706170096
Download: ML20140G829 (27)


See also: IR 05000295/1997002

Text

.

-

-.

- .-

. . .

. . . _ .

-

-

. . . . -

..

. ~ .

l-

.

i

l

U.S. NUCLEAR REGULATORY COMMISSION

l

l

l

REGION lli

!

Docket Nos:

50-295, 50-304

l

License Nos:

DPR 39, DPR-48

Report No:

50-295/97-02, 50-304/97-02

Licensee:

Commonwealth Edison Company

Facility:

Zion Nuclear Plant, Units 1 and 2

!

!

Location:

101 Shiloh Boulevard

Zion, Illinois 60099

Dates:

February 6 through April 2,1997

Inspectors:

A. Vegel, Senior Resident inspector

D. R. Calhoun, Resident inspector

i

E. W. Cobey, Resident inspector

I

J. Yesinowski, Illinois Department of

Nuclear Safety (IDNS) Inspector

Approved by:

Michael E. Parker, Acting Chief

Reactor Projects Branch 2

)

i

I

l

i

~.

[

DR

AD

G

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

__._-

_.

_ . _ . . _ _ _ . _ _ _ _ _ _ . _

,

.

.

.

EXECUTIVE SUMMARY

Zion Nuclear Plant, Units 1 and 2

NRC Inspection Reports 50-295/9742; 50-304/97-02

This inspection included aspects of licensen operations, maintenance, engineering, and

plant support. The report covers an eight week period of inspection activities by the

resident staff.

Licensee performance during this inspection period was characterized by fNguent

operational events. Of particular concern was the licensee's continuing inability to

implement timely and effective corrective actions for identified deficiencies and preclude

)

their recurrence. Also, operations' performance was characterized by poor control room

l

operating practices and inadequate supervision, as demonstrated in both the reactivity

management event and the reactor coolant system loop flow instrumentation being taken

out-of-service without recognizing the Technical Specification (TS) implications. Recurring

1C containment spray pump starting problems were caused by equipment deficiencies, and

in one case,i.1 adequate maintenance activities. Engineering staff follow-up of the slow

start times was not thorough, as reflected in an operability issue assessment not being

performed within the required time frame, in addition, failure to follow radiation protection

procedures continued to be a problem, as evidenced by an unauthorized entry into a locked

high radiation area.

Ooerations

The inspectors concluded that the February 21,1997, reactivity management

-

event, occurred due to significant weaknesses in command and control, ineffective

communications, and poor reactivity management during shutdown activities. (NRC

Augmented Inspection Team Report 50 295/97-06)

The inspectors concluded that the failure of the control room operators to recognize

-

the TS implications of removing safety-related instrumentation from service was

indicative of weak operations staff knowledge of TS requirements. In addition, this

event demonstrated similar weak operating practices, which included poor

command and control, and training deficiencies, that contributed to the occurrence

of the February 21,1997, reactivity management event. (NRC Augmented

Inspection Team Follow-up Report 50 295/97-07, 50-304/97-07)

An event concerning an unexpected high cooldown rate of the Unit 2 pressurizer

.

surge line involved poor communications between operators and licensee

management. (Section 01.1)

An apparent violation was identified involving the failure to implement timely and

-

effective corrective actions for a previous instance of undetected gas accumulation

h the reactor coolant system in September 1996. Several generic communications

existed which described similar industry events. (Section 01.5)

I

2

,

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

_.

. . . _ _ _

_ . - . _ _

i

-

(

An apparent violation was identified involving the failure to have procedures for

-

l

'

extended operation in cold shutdown conditions and for operating procedures to

include measures to dia.gnose or prevent the undetected accumulation of gas in the

reactor coolant system. (Section 01.5)

Two apparent violations were identified involving the failure to make a four-hour

-

non-emergency report and subm.t a written Licensee Event Report within 30 days,

for a condition that alone could have prevented the fulfillment of the safety function

to remove residual heat. (Section 01.5)

Maintenance

Poor electrical maintenance work practices resulted in a maintenance technician

-

I

receiving an electrical shock. (Section M1.1)

The inspectors identified an example of a violation involving the failure to implement

-

timely and effective corrective actions for NRC Violation 50-295/93023-01, 50-

304/93023-01, which pertained to the lack of requirements or acceptance criteria

~

to evaluate the acceptability of 1C containment spray pump starting delays.

(Section M1.2)

The inspectors identified a violation concerning 1C containment spray pump testing,

-

in that, both starting circuits were not tested to ensure that design requirements

were met. (Section M1.2)

l

l

The inspectors identified an example of a violation involving the failure to implement

-

!

timely and effective correctivo actions for a series of equipment failures caused by

over-torquing. (Section M1.3)

Enaineerina

The inspectors identified a violation involving the failure to address the operability

-

of the 1C containmut spray pump for a slow starting condition in a timely manner.

(Section M1.2)

Plant Suooort

i

The inspectors determined that a poor operating work practice contributed to the

-

Unit 1 power reduction, since station personnel allowed the refueling water storage

tank boron concentration to trend toward the TS limit over a six-month period

without taking effective action to correct the trend. (Section R1.1)

1

l

A violation was identified involving the failure to have direct oversight of and

{

-

positive control over each personnel entry into a locked high radiation area.

i

(Section R1.2)

t

!

3

1

i

i

l

i

.

,

_

,

. - ,

.

.

.

.-

-

-

1

.

'

.

l

l

Report Details

Summarv of Plant Status

'

l

Unit 1 was at or near 100 percent power at the beginning of this inspection period. On

l

February 10,1997, power was reduced to approximately 45 percent in response to a

l

Technical Specification (TS) action statement requirement for a high boron concentration

i

in the refueling water storage tank. Power was maintained at approximately 40 percent

until February 21,1997, when the reactor was shutdown in accordance with a TS action

statement for an inoperable containment spray pump. During this shutdown evolution, an

,

improper control rod manipulation occurred due to significant weaknesses in command and

'

control, ineffective communications, and poor reactivity management during the shutdown

activities. As a result of this event, the plant was placed in a cold shutdown condition and

depressurized. On February 24,1997, an Unusual Event was declared due to the plant

exceeding a TS action statement requirement for reactor coolant system loop

instrumentation being removed from service. On March 11,1997, offsite power and

shutdown cooling was lost due to a trip of the Unit 1 system auxiliary transformer.

'

Control room operators promptly restored shutdown cooling, but offsite power was not

restored until March 14,1997.

Unit 2 remained shutdown during the inspection period in support of the 14th refueling

outage. On February 28,1997, an Unusual Event was declared due to an unknown device

being entangled in the offsite feed for the Unit 2 system auxiliary transformer.

Licensee performance during this inspection period was characterized by frequent

operational events. Of particular concern was the licensee's continuing inability to

implement timely and effective corrective actions for identified deficiencies and preclude

their recurrence. Also, operations' performance was characterized by poor control room

operating practices and inadequate supervision, as demonstrated in both the reactivity

marmgement event and the reactor coolant system loop flow instrumentation being taken

out-of-service without recognizing the TS implications. Recurring 1C containment spray

pump starting problems were caused by equipment deficiencies, and in one case,

inadequate maintenance activities. Engineering staff follow-up of the slow start times was

not thorough, as reflected in an operability issue assessment not being performed within

the required ti ne frame. In addition, failure to follow radiation protection procedures

continued to be a problem, as evidenced by an unauthorized entry into a locked high

radiation area.

1. Ooerations

01

Conduct of Operations

01.1 Pressurizer Surae Line Unexpected Cooldown Rate

a.

Insoection Scoce (71707)

The licensee observed an unexpected high cooldown rato of the Unit 2 pressurizer

surge line following a reactor coolant pump (RCP) swap. The inspectors

4

'

\\

.

l

interviewed engineering and operations personnel and reviewed applicable

j

documentation including problem identification form (PlF) 97-0659.

b.

Observations and Findinos

l

On February 6,1997, with the plant in Mode 5 (cold shutdown), the licensee

conducted a swap of RCPs to support maintenance activities on the "B" loop

i

l

isolation valve,2MOV RC8002D. The control room (CR) operators started the 2D

'

RCP and secured the 2B RCP. Approximately,25 minutes following this evolution,

the operators noted that the surge line temperature had decreased from 418 to

l

216*F. The surge line temperature eventually stabilized at approximately 178'F,

which was the nominal reactor coolant temperature. The CR operators informed

operations management of the surge line cooldown and initiated an investigation

j

into the cause of the event. Subsequently, the licensee determined that the cause

of the event was the initiation of backflow through the surge line due to the

switching of RCPs.

Following the occurrence of the event, the operators questioned whether the

unexpected cooldown rate was reportable, due to being a condition outside TS

requirements. Technical Specification 3.3.2, "RCS [ Reactor Coolant System]

Pressurization System Integrity," states that the RCS coo ldown rates shall be

maintained within the limits specified in the Pressure and Temperature Limits

Report, and the pressurizer cooldown rate shall not exceed 200*F per hour. The

licensee subsequently concluded that the TS cooldown limits were not exceeded

based on the following information: (1) the RCS temperature remained stable

throughout the transient; (2) the pressurizer bulk temperature initially decreased

only 3*F; and (3) the surge line temperature decrease of approximately 240*F was

within the design basis. Specifically, Westinghouse Standard 1.3, " Systems

Standard Design Criteria NSSS [ Nuclear Steam Supply System] Design Transients,"

dated April 5,1971, states that the pressurizer surge line nozzle was designed to

withstand a step change of 320*F six times per plant heatup; a total of 1200 plant

occurrences.

Following event occurrence, licensee management informed the inspectors that the

potential for a temperature transient was discussed by the CR operators during the

pre-evolutionary brief. The operators discussed their concem with the onsite

Westinghouse representative and the engineering staff. However, the operators did

not discuss the concern with licensee management until after the event,

c.

Conclusions

The inspectors concluded that based on the review of licensee's design basis

documentation, TS were not violated and the consequences of the observed high

cooldown rate was minimal. The failure of the CR operators to communicate to

licensee management their concerns with the potential for a temperature transient

to occur while switching RCPs, was reflective of poor communications and

inadequate command and control.

I

5

i

-

l

i

.

01.2 Reactivity Manaaement Event

On Febrcry 21,1997, the operations staff performed an unplanned shutdown of

Unit 1 in response to the expiration of a TS limiting condition for operation action

statement concerning the 1C containment spray pump. During the shutdown, the

primary nuclear station operator manipulated control rods in a nonconservative and

'

uncontrolled manner. The inspectors concluded that the improper control rod

manipulation occurred due to significant weaknesses in command and control,

ineffective communications, and poor reactivity management during shutdown

activities. An NRC Augmented inspection Team was dispatched to the site to

investigate the event. The results of the inspection were documented in NRC

Augmented inspection Team Report 50-295/97-06 and in NRC Augmented

Inspection Team Followup inspection Report 50-295/97-07, 50-304/97-07.

01.3 Declaration of Unusual Event Due to imoronerly Removina the RCS Loon

Instrumentation from Service

On February 24,1997, the licensee identified that the TS action requirements for

removal of all three RCS loop "A" flow instruments had been exceeded.

Specifically, CR operators removed the flow instruments from service on

February 22,1997, without recognizing the TS implications. As a result, the

licensee exceeded the 48-hour requirement for the plant to be placed in a cold

shutdown condition. The licensee subsequently declared an Unusual Event in

accordance with the General Station Emergency Plan. On February 25,1997, the

plant was placed in a cold shutdown condition and the Unusual Event was

terminated. inspector assessment of licensee performance with respect to this

event was documented in NRC Inspection Report 50-295/97-07; 50-304/97-07.

01.4 Unusual Event Declared for Unknown Device Entanaled in Offsite Feed for the Unit

2 System Auxiliarv Transformer (SAT)

On February 28,1997, the licensee identified an unknown device entangled in the

offsite feed to the Unit 2 SAT. As a result, the licensee declared an Unusual Event

due to the potential for the degradation of plant safety (unknown nature of the

!

l

device) and the fact that outside agencies were called for assistance. The

inspectors concluded that plant management demonstrated the proper safety focus

in their declaration of an Unusual Event and the voluntary activation of the

Technical Support Center. The details of this event were discussed in NRC

inspection Report 50-295/97-08; 50-304/97-08.

01.5 Undetected Accumulation of Gas in the Reactor Vessel

l

a.

Insoection Scone (71707 and 37551)

l

On March 8,1997, the licensee identified the presence of a gas bubble in the

Unit 2 reactor vessel head. The inspectors interviewed licensee personnel,

reviewed applicable plant procedures and documentation, and evaluated the

6

_.

_

.

-

-

-

-

- .-

.-

--

-

.

--. .

. _ - .

.

.

(

-

.

!'

.

licensee's response to generic information on similar industry events and to a

previous Zion Station event in September 1996.

b.

Observations and Findinas

Seouence of Events:

On September 1,1996, the licensee identified a gradual decrease in reactor

-

vessel water level on Unit 1 as indicated on the reactor vessel level

l

indicating system (RVLIS) narrow range channels. The licensee

)

subsequently determined that a bubble existed in the reactor vessel head.

On September 2,1996, the licensee vented the Unit 1 reactor vessel head

l

-

and determined that approximately 1,028 gallons of water were required to

'

l

refill the vessel. Subsequently, the licensee initiated a root cause

investigation of the event.

On October 30,1996, engineering personnel approved the root cause

-

investigation, including recommended corrective actions, and forwarded it to

j

operations for approval.

l

On November 4,1996, Site Quality Verification (SOV) personnel initiated an

+

unresolved issue corrective action record (CAR) on this event. The purpose

l

of the CAR was to track the corrective actions.

l

1

On December 14,1996, the operations manager postponed addressing the

-

i

root cause investigation until following completion of the Unit 2 refueling

outage, which had been in progress since mid-September 1996. This was

due to the fact that Unit 1 was at power and Unit 2 was pressurized with

l

reactor coo l ant pumps energized, and in these plant conditions the

'

accumulation of gas in the reactor vessel was not an immediate concern.

l

On January 8,1997, SOV personnel updated the CAR to recognize that the

'

corrective actions were overdue (as of November 30,1996); however, no

action was initiated to address this issue.

'

On February 16,1997, the licensee completed Unit 2 depressurization.

-

On February 18,1997, the licensee took the Unit 2 reactor vessel level

+

indicating system (RVLIS) out-of-service (OOS) per Maintenance Instruction

(MI) 1, " Draining the Reactor Coolant System for Refueling or Maintenance,"

for calibration. Since the Unit 2 refueling outage had been significantly

l

extended, instrument maintenance (IM) department personnel wanted to

l

zero the surve;!!ance clock on the RVLIS instrumentation prior to the plant

l

conditions being thanged.

i

On February 19,1997, IM department personnel released the OOS to be

j

-

cleared on Unit 2 RVLIS.

s

7

--

- . .- - - - .

- . _ . _ . . . - . .

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

-

'

.

.*

!

.

.

'

.

.

On February 20,1997, the licensee vented the Unit 2 reactor head.

.

,

On February 21,1997, IM department pe sonnel completed the calibration

,

+

of Unit 2 RVLIS instrumentation. However, the operators did not retum

RVLIS to service since Mi-2, " Reactor Coolant System Fill and Vent," which

was being utilized by the operators to provide guidance in the existing plant

conditions, did not require RVLIS to be in-service until after the reactor

coolant system was solid and pressurized.

On March 3,1997, the licensee completed Unit 1 depressurization.

-

,

On March 7,1997, a Unit 2 nuclear station operator questioned the fact that

+

the computer trend plots for pressurizer level and the volume control tank

-

(VCT) were not trending in opposite directions as expected. Operations and

j

engineering personnel identified that a void in the vessel head could have

been the cause. Subsequently, operators reduced VCT pressure to inhibit

l

the transfer of nitrogen from the VCT to the reactor vessel on both units.

l

!

On March 8,1997, the licensee vented the Unit 2 reactor vessel head and

-

placed RVLIS in-service. The initial estimate of water to refill the void was

t

6500 gallons. The licensee subsequently determined that approximately

6913 gallons were necessary to refill the reactor head void. The estimate

i

for reactor vessel level was 588' 4.5" which was 2' 7.5" below the reactor

vessel flange.

i

i

The licensee also vented the Unit 1 reactor head. Approximately 1090

gallons of water were required to fill the void. Review of RVLIS trend data

for Unit 1 indicated a decreasing reactor vessel level for several days.

However, since the operators were not required to monitor RVLIS in these

plant conditions, the declining trend had not been identified. Consequently,

the licensee initiated periodic venting (daily) of the reactor head on both-

units.

The licensee determined that a void would have continued to grow in the

reactor vessel until the surge line tap in the loop (approximately 584' 6"

which is 5' 8" above the top of active fuel) was exposed before pressurizer

levelinstrumentation would have revealed the presence of a void in the

vessel. Once the surge line tap was exposed, gas would have been able to

migrate to the pressurizer and would have resulted in an indicated dcctoase

on pressurizer level instrumentation. However, residual heat remova! (RHR)

i

pump suction could have been affected at 584' 8", which would have been

-

prior to reaching the point where pressurizer level would have provided

i

!

direct indication of a void in the reactor vessel. In addition, the licensee

l.

subsequently determined that gas accumulation in the steam generators

i

would have prevented the operation of the preferred alternate method of

l

reactor coolant system (RCS) cooling due to the obstruction of natural

j

circulation flow.

!

ll

'

8

.

,

, ..

-

..

.-_ _.

.

_ - . -

- - -

. .

~ -.-

.

-

i

.

lt

I

.

l

l

On March 18,1997, independent Safety Engineering Group personnel

-

i

submitted a revision to the unresolved issue CAR for the September 1996

'

event to upgrade the CAR to a Severity Level I finding (a significant

condition which does affect safety).

On March 25,1997, system engineering personnel recognized the ongoing

-

accumulation of gases in the Unit 2 reactor vessel and initiated a problem

identification form.

,

)

On March 28,1997, the licensee approved the Level I CAR and initiated a

-

!

troubleshooting plan to address the ongoing gas accumulation on Unit 2.

<

l

On March 29,1997, the licensee sampled the gases being vented on Unit 2.

-

The results indicated that air was a source: however, the licensee

questioned the validity of the sample due to the sampling methodology.

On April 3,1997, the licensee revised PT-0, Appendix E 3, " Operating

-

Surveillance Checksheet," and issued an Operating Special Procedure (OSP)

!97-014, " Maintaining RCS Conditions in Mode 5."

.

On April 4,1997, the licensee established a team to review the

-

circumstances surrounding the voiding of both Unit 1 and 2 reactor vessels

I

and evaluate the ongoing gas accumulation in Unit 2.

l

1

On April 6,1997, the licensee sampled the gases being vented on Unit 2.

-

l

The results indicated that the composition of the gases was consistent with

the composition of the gases in the VCT.

Untimely Corrective Action for a Previous Event

On September 1,1996, the licensee identified a gradual decrease in reactor vessel

water level on Unit 1, as indicated on the reactor vessel level indicating system

'

(RVLIS) narrow range channels. The licensee subsequently determined that a

bubble existed in the reactor vessel head. On the following day, the licensee

vented the Unit 1 reactor vessel head and determined that approximately 1.028

gallons of water were required to refill the vessel.

As a result, the licensee initiated a root cause investigation. The investigation with

recommended corrective actions was approved by engineering and forwarded to

,

'

operations for approval on October 30,1996. On December 14,1996, the

Operations Manager postponed addressing the investigation until following

completion of the Unit 2 refueling outage. This was due to the fact that Unit 1 was

at power and Unit 2 was pressurized with reactor coolant pumps running, and in

these plant conditions the accumulation of gas in the reactor vessel was not an

I

immediate concern. However, when the units were subsequently depressurized,

l

the investigation and corrective actions were not addressed. Had this investigation

9

,

l

l

l

t

I

-

.

'

?

!

,

1

'

been aggressively pursued, the March 1997 event could have been prevented. As

l

of April 9,1997, the root cause investigation and recommended corrective actions

had not been approved.

The failure to irnplement timely and effective corrective actions for a previous

instance of undetected gas accumulation in the reactor coolant system in

September 1996 is an apparent violation of 10 CFR Part 50, Appendix B, Criterion

XVI (Escalated Enforcement item (EEI) No. 50-295/97002-01, 50-304/97002-01).

Poor Evaluation and imolementation of Generic Industry information

Generic communications regarding similar industry problems and the licensee's

resulting actions are described below. The licensee failed to adequately evaluate

and address this information, and hence the licensee's program to review generic

industry information did not prevent the March 1997 event.

NRC Information Notice (IN) 94-36, " Undetected Accumulation of Gas in Reactor

Coolant Systern," was issued on May 24,1994. This IN described an event that

was similar to the events that occurred at Zion Station. The licensee's evaluation

of this IN in August 1994 was inadequate,in that, it did not address the following

aspects of the IN: (1) the susceptibility of monitoring reactor coolant system

inventory solely based on pressurizer levelinstrumentation; (2) operator knowledge

of RVLIS; and (3) a lack of thorough evaluation of previously published information

on similar events, in addition, the corrective actions that were designated were not

completely implemented as of April 9,1997.

Nuclear Safety Advisory Letter (NSAL)94-013, " Nitrogen Accumulation in RCS

During Mode 5 Operation," was issued on June 16,1994. The licensee's

evaluation of NSAL 94-013 was closed in reference to the evaluation being

conducted for NRC IN 94-36.

NRC IN 96-37, " inaccurate Reactor Water Level Indication and Inadvertent

Draindown During Shutdown," was issued on June 18,1996. This IN described an

event that was somewhat different from the subsequent gas accumulation events

at Zion. However, botr. were similar, in that, the loss of reactor coolant system

inventory went undetected due to the sole reliance on pressurizer level

irotrumentation, which in some cases provides inaccurate reactor vessel water level

indication. The licensee's evaluation of this IN on August 7,1996, stated that

RVLIS was not utilized for levelindication, and it did not address the susceptibility

of relying only on pressurizer levelinstrumentation to monitor the reactor coolant

I

system inventory, in addition, the evaluation specified system engineering review

for concurrence; however, system engineering personnel never reviewed this IN.

l

As a result, no corrective actions were designated.

l

NRC IN 96-65, " Undetected Accumulation of Gas in Reactor Coolant System and

Inaccurate Reactor Water Level Indication During Shutdown," was issued on

December 11,1996. The licensee's evaluation indicated a lack of complete

10

l

i

i

~

_

-

.

! .*

l

'

l

l

understanding of the reactor coolant system levelinstrumentation. Specifically, the

evaluation indicated that RVLIS was capable of providing actuallevelindication.

However, the licensee concluded that the inaccurate reactor vessel level

instrumentation concern discussed in the IN was not on issue because refueling

level instrumentation (which indicates a level based on a reference to the

l

l

pressurizer) was available. Although this evaluation was conducted on

l

February 24,1997, when the plant conditions existed for the undetected

!

accumulation of gas to occur on Unit 2, the licensee did not recognize or question

the need for immediate corrective actions. The resulting recommendations were

not implemented prior to the March 1997 event.

inadeouate Procedures

The inspectors determined that no operating procedure existed which provided

guidance on maintaining the plant in Mode 5 (cold shutdown) for an extended

period of time. Consequently, operators were using Maintenance Instruction (MI) 2,

" Reactor Coolant System Fill and Vent," to provide operating guidance for the

!

existing plant conditions. The purpose of Mi-2 was to describe the operations

necessary to control the filling and venting of the primary system. As a result,

l

operators did not place RVLIS in-service following completion of instrument

calibration on February 21,1997, since Mi-2 did not specify RVLIS to be placed in-

l

service until after the plant was solid and pressurized. Without RVLIS in-service, no

direct indication of reactor vessel water level was available to the operators, which

prevented the timely identification of gas accumulation in the reactor vessel.

In addition, the inspectors noted that none of the operating procedures utilized

during cold shutdown conditions included measures to diagnose or prevent the

undetected accumulation of gas in the reactor coolant system. Specifically:

1

PT-0, Appendix E-3, " Operating Surveillance Checksheet" (Nuclear Station

j

-

Operator's shiftly surveillance while in Mode 5) did not include measures to

identify the accumulation of gas in the reactor vessel head, such as

monitoring RVLIS. Consequently, when RVLIS was trending downward on

Unit 1 in early March 1997, operators failed to identify that a void was being

j

created in the reactor head.

Abnormal Operating Procedure (AOP) 6.3, " Loss of RHR [ Residual Heat

-

Removal] Shutdown Cooling," relied on pressurizer levelinstrumentation for

determining reactor vessel water level. Pressurizer level was not an accurate

indication of reactor vessellevel under the voiding conditions.

Consequently, had the accumulation of gas continued to the point where

!

reactor vessel level was below 584' 8" (the point below which shutdown

cooling could have been affected), Abnormal Operating Procedure (AOP) 6.3

would not have identified the cause of the loss of shutdown cooling to have

i

l

been low reactor vessellevel. If shutdown cooling was not able to have

been restored, then AOP 6.3 would have directed alternate cooling to havo

been established. Alternate cooling methods were listed in order of

,

preference beginning with RCS cooling using steam generators. However,

l

11

i

i

_

.- _

__

_

_

_ .

.

_ _

._

i

e

<

.

l

the obstruction of natural circulation flow by the accumulation of gas in the

steam generator U-tubes could have prevented RCS cooling by this altcrnate

l

method. As reactor vessel water level continued below 584' 6" (the point

below which pressurizer levelinstrumentation provided direct indication of a

gas bubble in the reactor vessel), AOP 6.3 would have directed the addition

l

of inventory to raise reactor vessellevel. Therefore, due to the failure of

AOP 6.3 to utilize RVLIS for determining reactor vessel water level, the

recovery from a loss of shutdown cooling would have been significantly

l

complicated.

)

The failure to have procedures for extended operation while in cold shutdown

l

conditions and for operating procedures to include measures to diagnese or prevent

!

the undetected accumulation of gas in the reactor coolant system is an apparent

l

violation of 10 CFR Part 50, Appendix B, Criterion V (eel No. 50-295/97002-02,

50-304/97002-02).

Lack of Ooerator/Encineerina Knowledae on the Ooeration of RVLIS

Based on review of the licensee's evaluation of several generic communications

that pertained to this issue and interviews of senior reactor operators, reactor

operators, operating procedure writers, and engineering personnel, the inspectors

concluded that the licenses staff lacked an understanding of the operation of

RVLIS. In addition, the licensee did not demonstrate an appreciation for the

importance of monitoring RVLIS while in cold shutdown conditions. The inspectors

determined that operators only received training on the operation of RVLIS during

the systems phase of initial license training. In addition, operator training had not

covered the use of RVLIS in cold shutdown conditions. During one training cycle,

NRC IN 94-36 and the associated revision of the Mi-1 series procedures were

covered. This revision included the addition of a precaution that indicated under

certain plant conditions nitrogen could have accumulated in the reactor vessel head

and steam generator tubes to the point where natural circulation was inhibited

and/or gas binding of ths RHR pumps may have occurred. However, this training

was not sufficient to raise the operators sensitivity to the importance of monitoring

RVLIS while in cold shutdown conditions.

Failure to Reoort Event

The inspectors identified that the licensee failed to make the appropriate prompt

report to the NRC following the identification of the undetected gas accumulation in

the reactor vessel head on March 8,1997. Specifically,10 CFR Part 50.72(b)(2)(iii)(B) requires a four-hour event report be made for any event or

condition that alone could have prevented the fulfillment of the safety function to

remove residual heat. Also,10 CFR Part 50.73(a)(2)(v)(B) requires a written

l

Licensee Event Report be submitted within 30 days from the discovery of the event

l

for any event or condition that alone could have prevented the fulfillment of the

l

safety function to remove residual heat. This condition (undetected gas

[

accumulation in the RCS) could have potentially caused the loss of both trains of

shutdown cooling prior to the gas bubble reaching the size where pressurizer level

r

12

i

i

1

. _ . _ - _

__._____ _ _ _ _ _ .

. _ _ _ . _ _ . . _ _ _

-

.

.,

.

,

l

would have provided direct indication of reactor vessel water level. Additionally,

j

the accumulation of gas in the steam generators would have prevented the

preferred alternate method of RCS cooling due to the obstruction of natural

circulation flow.

,

The failure to make a four-hour non-emergency report and submit a written

t.

!

Licensee Event Report within 30 days, for a condition that alone could have

prevented the fulfillment of the safety function to remove residual heat, are

apparent violations of 10 CFR Part 50.72(b)(2)(iii)(B) and 10 CFR Part

'

50.73(a)(2)(v)(B) (eel No. 50-295/97002-03, 50-304/97002-03 and eel No. 50-

j

295/97002-04, 50-304/97002-04 respectively).

Site Quality Verification (SOV) and Manacement Oversiah!

SQV and management involvement in addressing the up iotected gas accumulation

l

in the reactor vessel events was fragmented and protre aced, which resulted in

l

untimely corative actions for significant conditions adverse to quality. The

untimely resp ose was reflected in: (1) the delay 'a reviewing and approving the

l

root cause inn 'tigation and recommended corsctive actions for the September

l

1996 event, wluch was not approved as o' April 9,1997;(2) SQV's acceptance

I

that the root cause investigation and recommended corrective actions for the

l

September 1996 event were overdue and were not being actively pursued in

l

January 1997; (3) the delay in providing the operators an approved procedure

j

governing plant operation in cold shutdown for an extended period of time; (4) the

ten day delay in reviewing and approving the Severity Leveli Corrective Action

Record submitted by the independent Safety Engineering Group; (5) the delay in

,

approving and implementing a troubleshooting plan to address the ongoing gas

l

'

accumulation in the Unit 2 reactor vessel; and (6) the delay in establishing a team

of personnel to evaluate the circumstances surrounding the March 1997 event and

the ongoing gas accumulation on Unit 2.

Safety Consecuences

Although there were no actual consequences of these events on the health and

safety of the public or plant staff, the inspectors concluded that these events

'

l

involved potential safety consequences. The identification of the gas accumulation

)

in the reactor vessel was an example of an operator's questioning attitude towards

unexpected, conflicting it'dications. However, it was fortuitous that the gas

l

accumulation was detected prior to it reaching a size which could affect RHR

system operation. Specifically, given the appropriate plant conditions and the

operating practices at that time (not requiring RVLIS to be in-service while in cold

shutdown and, if in-service, not monitoring RVLIS), the potential existed for a void

i

to have been created in the reactor vessel that could have impacted decay heat

removal prior to being detected by pressurizer levelindication. Additionally, the

accumulation of gas in the steam generators would have prevented the preferred

2

altemate method of RCS cooling due to the obstruction of natural circulation flow.

i

i

13

l

l

l

'

.

-

.

<

-

t

c.

Conclusions

The inspectors concluded that the undetected gas accumulation in the reactor

coolant system was due to: (1) untirnely corrective action for a previous event:

(2) poor evaluation and implementation of generic industry information; (3) lack of

procedural guidance for extended operation in cold shutdown conditions; (4)

operating procedures utilized during cold shutdown conditions did not include

measures to diagnose or prevent the undetected accumulation of gas in the reactor

coolant system; and (5) operations and engineering personnel did not have an

understanding of the importance of monitoring RVLIS while in cold shutdown

conditions.

II. Maintenance

M1

Conduct of Maintenance

M1.1 Electrician Received Electrica' Shock

a.

Insoection Scoce (62707)

A contractor electrician received an electrical shock while troubleshooting a service

building heating, ventilation,.and air conditionir,g unit (HVAC), OSV078. The

inspectors reviewed the licensee's documentation of the event including problem

i

identification form (PlF) 97-0829.

b.

Observation and Findinas

On February 16,1997, a contractor electrician was performing troubleshooting of

the service building HVAC unit, OSV078. The electrician performed the work under

a personal out-of-service. While troubleshooting, the electrician placed his hand

into a control panel with exposed conductors and received an electrical shock. The

electrician was transported to a local hospital for examination and was

subsequently released.

c.

Conclusion

As previously documented in NRC Inspection Report 50-295/96-20; 50-304/96-20,

poor electrical maintenance work practices resulted in the inadvertent actuation of

two valves in the containment spray system. The occurrence of the electrician

receiving an electrical shock on February 16,1997, was another example of poor

electrical maintenance work practices which could have significant personnel safety

consequences. The inspectors were concerned that previous corrective actions to

address poor work practices in the vicinity of energized components have not been

effective in improving performance in this area. At the end of the inspection period,

,

the licensee was in the process of developing corrective actions to address the

l

February 16,1997 event.

i

t

l

l

14

_ _ _ _

  • -

t

.

'.

,

! *

M1.2 1C Containment Sorav (CS) Pumn Testino

,

s.

Insoection Scope (627071

l

The inspectors reviewed the circumstances related to problems encountered during

l

1C CS pump testing. The inspectors reviewed relevant 1C CS pump surveillance

documentation, PlF 97-0798, and operability assessment #ER9701035. In

addition, the inspectors interviewed licensed operators, the system engineer, design

!

engineers, and licensee management.

'

b.

Qhservations and Findinas

During this inspection period numerous problems were experienced with the 1C CS

pump. The 1C CS pump is one of three CS pumps per unit, but, unlike the other

l

motor driven pumps, the 1C CS pump is driven by a diesel engine. The diesel

engine is started by one of two independent starting circuits, with each circuit

receiving power from a dedicated battery. Each battery also has a dedicated

battery charger which maintains the associated battery' replenished.

On February 11,1997, the 1C CS pump was taken OOS for maintenance. Work

performed included motor control center (MCC) inspections and fuel oil day tank

level indication calibration. The maintenance on the level indication required fuel oil

day tank draining and refilling.

On February 12,1997, during performance of the post maintenance test in

accordance with PT-6C-ST, " Containment Spray C Pump System Test and

Checks," Revision 5, the operators noted a delay in the stort time of the pump.

The operators observed that the pump took more than five seconds to start.

However, since the surveillance procedure did not require the pump start to be

timed, the exact start time delay was not recorded. The delay in the 1C CS pump

start was caused by the engine not being able to start on the selected battery, in

this case, the #2 battery. The enginn subsequently did start on the alternate

battery (#1) due to an automatic ratcheting device in the start circuitry. The

automatic ratcheting device is designed such that when one battery has an

undervoltage condition, the circuit automatically transfers to the remaining battery

bank and locks out the degraded battery. The lockout can be manually reset using

a reset push button.

Following the observed slow start the operators secured the pump and started it a

second time as required by the surveillance procedure. Again, the operators

observed a similar start time delay. The diesel tried to start on the degraded

i

battery again because an operator had apparently reset the locked out battery.

Operators eventually secured the pump and requested assistance from the system

j

engineer in evaluating pump operability. The system engineer subsequently

,

determined that the start delay was caused by the #2 battery being partially

'

discharged and not capable of starting the diesel. The battery had been

inadvertently discharged during maintenance activities on the MCCs since the OOS

in support of the work de-energized the battery charger.

l.

15

l

- - - - .

.

-

-

...

-

--.

-

-

-

-

.

- . . - - ,_ - . .

.

-

-. ...~.

..

- . - - . _ -

. - - - -

. - -

a

g

.

,

t

.

.

'

After discussions with the system engineer, the shift engineer declared the 1C CS

,

pump operable based on: (1) Zion Operability Determination Manual (ZODM)

CS-3.3, Revision 7, which required only 1 of 2 batteries to be operable, and (2)

PT-6C-ST, which did not have a starting time acceptance criteria. Therefore, since

l

PT-6C-ST was satisfied, the 1C CS pump was considered operable.

t

[

On February 14,1997, the system engineer determined that a formal operability

i

!

evaluation was needed to address the time delay caused by the ratcheting of the

i

starting circuit. Operability assessment #ER9701035 was initiated to document

engineering's conclusion that the 1C CS pump was operable based upon the results

!

of previously completed calculations. Specifically, engineering calculation 22S-B-

006M-080, which was performed to justify a slow start of the 2C CS pump,

allowed for a CS pump start delay of 18.3 seconds before the analyzed time

requirement to respond to a postulated loss of coolant accident (LOCA) would be

exceeded. On February 16,1997, based on re-evaluation of the operability

assessment, the licensee initiated an Operability Isex Form, Appendix B, to further

evaluate and conect the #2 battery degraded condition.

On February 19,1997, the licensee declared the 1C CS pump inoperable during

performance of PT-6C-ST when the pump required approximately 60 seconds to

start. Upon declaring the 1C CS pump inoperable, the licensee entered TS 3.6.1.C

and D limiting condition for operation (LCO) action requirements which required the

l

'

pump to be restored within 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> or the plant be placed in hot shutdown within

an additional four hours. The operators documented in the control room logs that

the pump was inoperable at 10:40 a.m.; however, the pump was actually

-

inoperable at 10:20 a.m. The discrepancy was subsequently noticed on

February 21,1997, by the shift engineer and the discrepancy was corrected prior

to the LCO action requirements being exceeded.

The licenses subsequently determined that the cause for the slow start of the 1C

CS pump was clogging of the fuel oil filter. The licensee flushed the fuel oil day

tank and removed all debris from the tank. At the end of this inspection period, the

cause of the fuel oil filter fouling was still under investigation.

.

On February 21,1997, maintenance personnel overfilled the day tank causing a fuel

oil spill. The cleanup of the spill delayed restoration of the CS pump to service.

Following completion of maintenance activities, the pump was retested in

accordance with PT-6C-ST. Durin0 the test, the CS pump experienced a slow start

time of approximately 20 seconds. The licensee subsequently determined that the

I

start delay was caused by bumt contacts on a starter circuit relay. The licensee

replaced the affected relay and the starting solenoids.

The licensee subsequently retested the 1C CS pump satisfactorily at 1:58 p.m. on

February 21,1997. However, the pump could not be declared operable since the

system was stillin the test configuration. As a result, operators subsequently

tripped the reactor at 2:15 p.m., five minutes prior to the expiration of the LCO

action requirements. As discussed in the Augmented Inspection Team

(AIT)lnspection Report 50-295/97-06 and AIT Followup Report 50-295/97-07;

f

16

-

.-.

.

..

..

_

__

__

_ _

_ ___

,

.

.

.

50-304/97-07, the CS maintenance and testing activities directly contributed to the

plant shutdown and reactor trip in which the improper control rod manipulation

event occurred.

The inspectors reviewed the Updated Final Safety Analysis Report (UFSAR) for CS

start times with respect to main steam line break (MSLB) and LOCA analyses. The

UFSAR LOCA analysis requires the initiation of CS flow to the spray nozzles within

110 seconds after the containment high high pressure setpoint is reached. The CS

pump start time delay caused by the ratcheting of the starting circuit did not result

in the required 110 second actuation time being exceeded. Concerning the CS

system response time required in the MSLB analysis, the inspectors review of

supporting calculations was in progress at tne end of the inspection period.

l

Pending inspector review of the licensee's MSLB analysis calculations to determine

the impact of the 1C CS ratcheting time delay on the ability of the CS system to

perform its design function, this item is considered an inspector Follow-up item (IFl

50-295/97002-05).

The inspectors reviewed the applicable CS pump test procedures including PT-6C-

ST-RT, " Containment Spray C Pump System Tests and Checks," Revision 8. The

inspectors determined that this test, which was utilized to obtain CS pump start

time data once every 18 months, was inadequate. Specifically, the test only

required response time data to be obtained for one of the two engine starting

circuits. As a result, the potential existed that the most conservative starting time

was not recorded. The starting time obtained by this test was used as an input for

Technical Staff Surveillance Procedure 15.6.136/146, " Matrix Procedure for Train B

Reactor Protection and Safeguards Response. Time Testing Un'it 1." This matrix

procedure was utilized to verify that the system response time was within the

I

bounds of the analyzed design limits.

Also during review of 1C CS pump test procedures PT-6C-ST and PT-6C-ST-RT, the

inspectors identified that acceptance criteria for the pump start time was not

delineated in the procedures. The inspectors were concerned that PT-6C-ST did not

require the starting time to be measured and neither procedure enntained

acceptance criteria for starting times. Consequently, starting time delays would not

have been identified or analyzed.

inspectors had previously identified the absence of starting time criteria in

procedure PT-6C-ST in Novumber 1993, as documented in Notice of Violation

50-295/93023-01, 50-304/93023-01. The licensee's corrective actions included

increased training of engineering personnel on operability daterminations and

revision of the ZODM. However, as reflected in the inspectors identifying the same

FT-6C-ST inadequecies on February 21,1997, the licensee's corrective actions

wero inadequate.

1

l

l

During the course of inspection follow-up of the 1C CS pump failures, the

inspectors requested a copy of the Appendix B to operability assessment

  1. ER9701035 which was initiated on February 16,1997. During the month of

,

March, the inspectors made numerous additional requests for the completed

17

l

.-

_

__

_

._

_ _ . . .

_ _ . __

_ . _ .

_. __

e

\\

.

l

Appendix .B without success. On April 3,1997, the inspectors were informed by

the licensee that the Appendix B had not been approved by the Operating Engineer

until March 25,1997. ZODM-0, " Operability Determination Manual," Revision 9,

requires that Appendix B, " Operability issue Form," shall be completed within five

days from initial discovery of the deficiency. The licensee exceeded the prescribed

completion time for the Appendix B operability assessment by approximately 36

days.

c.

Conclusiong

The inspectors concluded that the recurring 1C CS pump starting problems were

caused by equipment material condition deficiencies and poor control of

maintenance activities. ~ Spocifically, an OOS in support of maintenance activities

resulted in a battery being unexpectedly discharged, which caused the 1C CS slow

Wrt on February 12,1997. Once the starting delays were identified, investigation

ano P;1 solution of the problem.s were delayed by poor maintenance practices and

'

untimely engineering evaluation of conditions. These ~ weaknesses were reflected in

the occurrence of the fuel oil spill and the significant delay in completing Appendix

B of the operabiHty assessment. In addition, the inspectors identified weaknesses

in the 1C CS test procedures, some of which were previously identified but never

,

'

corrected.

As a result of the inspectors review of CS pump testing activities, several violations

I

of NRC requirements were identified. Periodic test PT-6C-ST-RT " Containment

Spray C Pump System Tests and Checks" was inadequate to demonstrate that the

CS pump would have performed satisfactorily in service; in that both battery

starting circuits were not tested, contrary to the requirements of 10 CFR Part 50,

Appendix B, Criterion XI, " Test Control" (50-295/97002-06, 50-304/97002-06).

In addition, the inspectors identified that licensee corrective actions were

inadequate for a previous NRC violation concerning PT-6C-ST not having any

requirements or acceptance criteria to evaluate starting time delays during CS

testing. The failure of the licensee to correct the previously identified testing

inadequacies was considered a violation of 10 CFR Part 50, Appendix, B, Criterion

XVI, " Corrective Actions" (50-295/97002-07a, 50-304/97002-07a).

The failure of the licensee staff to complete the Appendix B for the operability

assessment of the 1C CS pump starting time delay, that occurred on February 12,

within five days as required by ZODM-0, " Operability Determination Program," was

considered a violation of 1'O CFR Part 50, Appendix B, Criterion V, " Instructions,

l

Procedures, and Drawings" (50-295/97002-08).

!

!

M1.3 Ineffective Corrective Actions for Several Over-torauino Events

a.

Insoection Scone (62707)

t

On February 11,1997, when the 2A emergency diesel generator (EDG) was

i

!

returned to service following maintenance activities, the licensee discovered the

!

i

l

18

l

-

.

.

. - -

-

-

. -

-

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

- . - -

j

-

.

l*

i

!

(

o.

J,

)

(

lube oil cooler channel head was cracked. On February 28,1997, during

reassembly of the OB fire pump discharge piping, the piping connection between

the diesel engine fire pump and the discharge check valve cracked. The inspectors

reviewed the applicable work instructions and evaluated the licensee's corrective

j

l

actions.

l

b.-

Observations and Firidinas

On February 11,1997, when the 2A EDG was returned to service following

i

maintenance activities, the lube oil cooler channel head was discovered to have

been cracked. The licensee determined that during reassembly of the lube oil

cooler, the service water to lube oil channel head flange had been over-torqued.

Maintenance procedure P/DG002/3 3, " Diesel Generator Lube Oil Cooler

l

Maintenance," Revision 4, specified the torque to have been 131 ft-lbs: however,

'

since the flange was a cast iron raised face flange, the correct torque could have

been 47 ft-lbs. As a result of this event, the licensee replaced the cracked channel

l

head, revised maintenance procedure P/D002/3-3, and reduced the actual torque on

the fasteners for both the jacket watet and lube oil cooler channel heads on each of

the other EDGs.

,

!

A similar event occurred on the 2B EDG lube oil cooler channel head in

February 1995 following replacement of the lube oil cooler service water isolation

valve,2SWOO33. The valve was replaced in accordance with Work Request No.

!

950006910-01. The licensee attributed the cause of the failure to have been

flange misalignment, .but did not recognize that the flange was over-torqued.

l

Consequently, the licensee's corrective actions were not sufficient to preclude the

j-

subsequent failure of the 2A EDG lube oil cooler channel head on February 11,

l

1997.

'

i

l

In addition, the inspectors noted that during the past year, the licensee experienced

i

at least two other equipment failures due to the equipment having been over-

torqued. On June 20,1996, the 1 A condensate pump discharge flange cracked

during reassembly; and, on July 31,1996, the OC cribhouse sump pump discharge

check valve flange cracked during reassembly. In both of these instances, the

joints were raised face cast iron, The licensee repaired the equipment and

incorporated each uf the instances into mechanical maintenance lessons learned in

the case of the 1 A condensate pump, the licensee subsequently revised the

,

'

procedure to include the correct torque value,

On February 28,1997, during reassembly of the OB fire pump discharge piping in

accordance with Work Request No. 940037310-01, the piping connection between

the diesel engine fire pump and the discharge check valve cracked due to being

over-torqued. The licensee's investigation concluded that the flange had been

torqued to 370 ft-lbs; however, since the flange was raised face cast iron, it should

have been torqued to 76 ft-lbs.

l

.

.

l

In response to a series of equipment failures on February 29,1997, caused by

over-torquing, the licensee issued a stop work for mechanical maintenance on

i

19

\\

-.

-

. ._.

_

.

.

--

1=

.

_

.

-

-

-

.-.

.

-

_

_ -

-

.

r

i e

e

l

February 28,1997, pending the development and implementation of corrective

actions. Prior to allowing the mechanical maintenance department to return to

work, the licensee completed short-term corrective actions which included:

(1) training on maintenance procedure P/M014-1N, " Safety Related and Section XI

Code Mechanical Closure Report"; (2) the verification of correct torque values in

mechanical maintenance procedures; and (3) the implementation of a torque

verification form to be completed by the first line supervisnr for each work activity

prior to reassembly. Additional long term corrective actions included the

verification of torque values for all maintenance procedures and work packages

conducted during the current Unit 2 outage (scheduled completion date of April 30,

1997) and the completion of a training course for mechanical maintenance

l

personnel on the identification of flange and fitting materials, flange configurations,

closure procedures, and the use of torque wrench adapters and extensions

l

(scheduled completion date of December 31,1997).

c.

Conclusions

The inspectors concluded that in the case of the 2B EDG failure in February 1995, a

significant condition adverse to quality, the causes were not effectively determined

i

and corrective actions taken to preclude recurrence were inadequate, as evidenced

l

by the subsequent failure associated with the 2A EDG on February 11,1997. This

'

is considered an example of a violation of 10 CFR Part 50, Appendix B, Criterion

XVI, " Corrective Actions" (50-295/97002-07b, 50-304/97002-07b). In addition,

the corrective actions taken for several failures of nonsafety-related equipment with

!

the same cause (over-torquing of cast iron raised face flanges) were not effective at

preventing a subsequent failure of a safety-related piece of equipment.

l

III. Enoineerina

l

E1

Conduct of Engineering

E1.1

Notification of inocerable Enaineered Safety Feature Busses Due to Breakers in an

Unaualified Seismic Position.

On February 13,1997, the station made a one hour non-emergency report in

accordance with 10 CFR Part 50.72(b)(1)(ii)(A) after determining that several 4KV

l

and 480V buses were inoperable because their associated breakers were not in a

l

seismically qualified position. The licensee's breaker operating practices allowed

l

the breakers to be in a removed / racked out position, which is an unqualified

i

position. The licensee's immediate corrective actions included placing all breakers

!

in a seismically qualified position. This issue is considered an Inspection Follow-up

item (50-295/97002-09, 50-304/97002-09) pending NRC review of the licensee's

long-term corrective actions.

<

20

l

l

l

l

.

.

~

.

9

e

.

E8

Miscellaneous Engineering issues

1

E8.1

(Closed) Unresolved item 50-295/96020-07. 50-304/96020-07: Review the

i

consequences of the TS required boron concentration samples with the residual

heat removal (RHR) sample valves wired incorrectly.

,

Technical Specification 4.2.1.A.2 requires that when reactor coolant system

pressure is less than 200 psig, boron concentration in the operating RHR loop be

used to verify shutdown margin at least once a shift. The inspectors reviewed the

licensee's root cause investigation report, No. 304-200-97-CAOS-0097, the RHR

system configuration, and the previous year's boron concentration sample results.

The inspectors concluded that even though the RHR sample valves had been wired

incorrectly since original construction, the configuration of the RHR system was

such that both trains were normally cross-connected and sufficient mixing was

available for a boron concentration sample from either train to be a representative

sample for the operating loop. In addition, the inspectors reviewed previous sample

results and did not identify any abnormalities. This Unresolved item is closed.

IV. Plant Suonort

R1

Radiological Protection and Chemistry (RP&C) Controls

R 1.1 Hiah Boron Concentration in the Refuelina Water Storaae Tank (RWST)

a.

Insnection Scoce (71750)

On February 10,1997, the licensee reduced power on Unit 1 after entering a

shutdown LCO as a result of RWST in. con concentration being greater than the TS

limit. The inspectors reviewed RWST boron sample results over the last nine

months and interviewed personnel from the operations and chemistry departments.

b.

Observations and Findinas

On February 8,1997, RWST boron concentration sample results indicated 2600

ppm. Technical Specification 3.8.1.F specifies RWST boron concentration to be

between 2400 and 2600 ppm. Operators added approximately 9000 gallons of

primary water to the RWST and placed the RWST on recirculation for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> prior

to obtaining a subsequent sample. On February 10,1997, the post recirculation

sample indicated a boron concentration of 2605 ppm, which was greater than the

TS limit. As a result, the licensee commenced actions to place Unit 1 in hot

shutdown as required by TS. After recirculating the RWST a second time using one

of the containment spray pumps, the licensee obtained acceptable boron sample

results and terminated the plant shutdown.

The licensee subsequently determined that the cause for the high boron

concentration samples was stratification of the water in the RWST. As a result of

insufficient circulation, the boron tended to concentrate on the lower portions of

the tank, which resulted in the high sample results, even after significant dilutions.

21

-

1

.

r

l

s

'

,

I

The inspectors were informed by the licensee that the RWST boron concentration

l

sample results had been trending toward the upper TS limit since October 1996.

l

However, actions taken were ineffective to stop the trend.

C.

Conclusion

i

The inspectors concluded that the poor operating practice of allowing the boron

concentration to trend towards the upper TS limit without taking appropriate action

resulted in the licensee unnecessarily placing Unit 1 in an LCO and contributed to

'

the occurrence of the Unit 1 power reduction. Station personnel did not effectively

address the cause for the unexplained rise in boron concentration sample results

until the TS limit was exceeded.

R1.2 Unauthorized Entry into Locked Hiah Radiation Area (LHRA.]

a.

Insoection Scoce (7125.Q1

The licensee identified that s maintenance mechanic made an unauthorized entry

into a locked high radiation area. The inspectors interviewed licensee personnel and

reviewed applicable procedures and documentation.

b.

Observations and Findinas

On April 1,1997, radiation protection personne! issued maintenance contractors

the key for the Unit 1 vertical pipe chase (VPC) room which was posted as a LHRA,

an area with a dose rate greater than 1,000 mrem /hr. Radiation protection

personnel briefed the contractors on the LHRA access requirements before issuing

the key to one of the contractors who was designated as the key custodian.

Subsequently, custody of the key was transferred to another contractor, who was

also trained on LHRA access controls. In addition, the contractors had also been

informed of the LHRA access requirements by their foreman during a pre-job

briefing.

Upon aniving at the job site, the contractors propped open the VPC room door to

accomplish welding activities. In accordance with Zion Administrative Procedure

(ZAP) 610-02, "High Radiation Area Access Control," Revision 3, the key custodian

was required to provide direct oversight of and positive control over each personnel

entry into the area. However, he became involved in other work activities which

distracted him from providing continuous control over entrance into the VPC. As a

result, a maintenance mechanic entered and exited the VPC without being observed

by the key custodian. The mechanic was in the VPC for approximately five minutes

and received a dose of 2 mrem.

The maintenance mechanic had previously been in the Unit 2 VPC, which was not

posted as a LHRA, and was entering the Unit 1 VPC to verify valve packing

information. The maintenance mechanic entered the VPC area since the door was

open and he did not see the radiological postings which would have prevented him

from entering.

22

-

--

-

.-

.- - . -.-

..

- - - . - -

-- ---- -

.

.

l ,~,

.

?

e

ie

i

'.

i

I

l.

'

On April 2,1997, the maintenance mechanic contacted the radiation protection

'

l

(RP) department to question the Unit 2 VPC access controls. As a result, the

l

mechanic's unauthorized entry into the Unit 1 VPC was identified. Radiation

'

Protection personnel informed RP management of the problem; however, RP

management failed to inform senior plant management of the unauthorized entry

until April 3,1997. The licensee's immediate corrective actions included: (1)

initiating a PlF; (2) performing a preliminary investigation; (3) restricting the key

custodian from the radiological controlled area; and (4) issuing a standing order on

April 3,1997, which required RP personnel to be the key custodian and control

access to locked high radiation areas.

I

c.

Conclusion

i

The inspectors concluded that the licensee's preliminary investigation was

appropriate; however, RP management did not notiny senior plant management of

.

the unauthorized entry into a LHRA in a timely manner.

ZAP 610-02, "High Radiation Area Access Control," Revision 3, requires that

entrances to accessible high radiation areas with radiation levels greater than

1,000 mrem /hr be locked or be controlled by a key custodian who has direct

l

oversight of and positive control over each personnel entry into the area. The

l

failure of the key custodian to have direct oversight and positive control of

personnel entry into the Unit 1 VPC, which was a locked high radiation area, is

considered a violation of Technical Specification 6.2.2.B (50-295/97002-10).

l

l

l

V. Manaaement Meetinas

X1

Exit Meeting Summary

The inspectors presented the inspection results to members of licensee

management at the conclusion of the inspection on April 9,1997. The licensee

acknowledged the findings presented.

The inspectors asked the licensee whether any materials examined during the

inspection should be considered proprietary. No proprietary information was

identified.

j

X3

Management Meeting Summary

l

l

l

The Deputy Executive Director for Regulatory Programs and the Director for Nuclear

!

Reactor Regulation toured the Zion facility and met with licenseo management on

'

March 26,1997.

!

i

!

v

i

l

23

l

l

._.

_

._

_

_ . . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . _ . . . _ . _ .

. _

_ _ _

. _ _ _ _ . _ . _ . _ . . _ _ _ _ . . ._

_

[

'

! c.

I

..

Partial List of Persons Contacted

i

J.icensee

!

)

J. Mueller, Site Vice President

i

l

K. Dickerson, Executive Assistant

j

l

R. Godley, Regulatory Assurance Mainager

!

M. Weis, Support Services Director -

R. Zyduck, Site Quality Verification Director

l

T. Patterson, Unit 1 Plant Manager

)

R. Starkey, Unit 2 Plant Manager

'

K. Hansing, Unit 1 Operations Manager

!

G. Vanderheyden, Unit 2 Operations Manager

M. Schimmel, Unit 2 Maintenance Manager

B. Giffin, System / Component Engineering Manager

R. Budowle, Site Quality Verification

i

M. Madigan, Site Ouality Verification

W. Stone, Regulatory Assurance Supervisor

C. Allen, Regulatory Assurance

i

D. Beutel, Regulatory Assurance

NaC

M. Dapas, Chief, Reactor Projects Branch

l

l

1

l

I

4

l

l

!

!

'

'

)

I

i

i

24

.

I

i

i

'

_

_

,

, _ -

$

-

2

i

o

,

I

List of Insoection Procedures Used

IP 37551

Engineering

lP 62707

Maintenance Observation

IP 71707

Plant Operations

IP 71750

Plant Support Activities

List of items Ooened, Closed, and Discussed

Onened

50-295/304-97002-01

EEi

Failure to imp!ement timely and effective corrective

action for a previous instance of undetected gas

accumulation in the reactor coolant system in

September 1996

50-295/304-97002-02

eel

Failure to have procedures for extended operation while

in cold shutdown conditions and for operating

procedures to include measures to diagnose or prevent

the undetected accumulation of gas in the reactor

coolant system

50-295/304-97002-03

eel

Failure to make a four-hour non-emergency report, for a

condition that alone could have prevented the

fulfillment of the safety function to remove residual

heat in accordance with 10 CFR Part 50.72(b)(2)(iii)(B)

50-295/304-97002-04

eel

Failure to submit a written Licensee Event Report within

30 days, for a condition that alone could have

prevented the fulfillment of the safety function to

remove residual heat in accordance with 10 CFR Part 50.73(a)(2)(v)(B)

50-295/97002-05

IFl

Review the MSLB analysis to determine the impact of

the 1C CS pump ratcheting time delay on the ability of

the CS system to perform its safety function

50-295/304-97002-06

VIO

Failure to test both battery starting circuits for the "C"

containment spray pumps to ensure that the design

starting time requirements were met

50-295/304-97002-07a

VIO

Failure to implement timely and effective corrective

actions for NRC Violation 50-295/93023-01, 50-

304/93023-01

50-295/304-97002-07b

VIO

Failure to determine the cause and implement effective

corrective actions to preclude recurrence for the

cracked lube oil cooler channel head on the 2B

emergency diesel generator

25

_ _ _ _ - _ _ _ _ - _

__

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ ____

9

.

!s

a

g

50-295/97002-08

VIO

Failure to complete appropriate operability assessment

within five days of discovery of the slow starting time

on the 1C CS pump

50-295/304-97002-09

IFl

Review of long-term corrective actions for breaker

position seismic qualification

50-295/97002-10

VIO

Failure to have direct oversight und positive control

over each personnel entry into a locked high radiation

area

Closed

50-295/304-96020-07

URI

Review the adequacy of boron concentration samples

with the RHR sample valves wired incorrectly

.

4

0

26

. _ . .

. _ .

.

.

. ._ .

..

.. _ . .

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

[*

l2

,

.

1

I

list of Am. .. ...s

!

l

AOP ' Abnormal Operating Procedure

CAR Corrective Action Record

CR

Control Room

l

CS

Containment Spray

l

EDG

Emergency Diesel Generators

l

eel

Escalated Enforcement item

l

HVAC Heating, Ventilation and Air Conditioning

IDNS lilinois Department of Nuclear Safety

IFl

inspector Follow-up Item

IN

information Notice

IM

instrument Maintenance

LCO

Limiting Condition for Operation

LHRA Locked High Radiation Area

LOCA Loss of Coolant Accident

'

MI

Maintenance Instruction

MCC Motor Control Center

'

MSLB Main Steam Line Break

NRC- Nuclear Regulatory Commission

NSAL Nuclear Safety Advisory Letter

NSSS Nuclear Steam Supply System

OOS Out-of-Service

OSP Operational Special Procedure

PDR

Public Document Room

PlF

Problem identification Form

RCP

Reactor Coolant Pump

RCS

Reactor Coolant System

RHR

Residual Heat Removal

RP

Radiation Protection

RP&C Radiological Protection & Chemistry

RVLIS Reactor Vessel LevelIndicating System

RWST Refueiing Water Storage Tank

SAT

System Auxiliary Transformer

SQV Site Quality Verification

TS

Technical Specifications

UFSAR Updated Final Safety Analysis Report

URI

Unresolved item

VCT Volume Control Tank

VIO

Violation

VPC

Vertical Pipe Chase

ZAP

Zion Administrative Procedure

ZODM Zion Operability Determination Manual

i

l

27

_

_