IR 05000369/1993020

Insp Repts 50-369/93-20 & 50-370/93-20 on 930901-05.Review of Insp Findings Continuing.Major Areas Inspected:Events Surrounding Leak Repair Procedure That Led to 930831 Steam Leak Inside Containment of Unit 2 Reactor

ML20057E958
Person / Time
Site:	McGuire, Mcguire
Issue date:	09/23/1993
From:	Casto C, Gibson A NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML20057E944	List: IR 05000369/1993020 ML20057E943
References
50-369-93-20, 50-370-93-20, NUDOCS 9310130418
Download: ML20057E958 (41)

Text

.-

.-.

.

-

.

-#

.

UNITED STATES

/pn Mog%

NUCLEAR REGULATORY COMMISSION

'

3 -

REGION 11

101 MARIETTA STREET, N.W., SUllE 2900 in -

E ATLANTA, GEORGIA 30323 0199

8

%

4***.*/

%

Report Nos.: 50-369/93-20 and 50-370/93-20 Licensee: Duke Power Company 422 South Church Street Charlotte, NC 28242 Docket Nos.: 50-369 and 50-370 License Nos.: NPF-9 and NPF-17 Facility Name: McGuire 1 and 2 Inspection Conducted:

September 1-5, 1993

/

Ad 9/>P/ G Team Leader:

-, o m-Date Sig'ned C.'Tisto, 'Chitf'

'

'

Test Program Section Division of Reactor Safety Team Members:

J. Arildsen, Human Factor Specialist, NRR R. Baldwin, Operator Licensing Examiner, RII

-

G. Hornseth, Materials Engineer, NRR W. Orders, Senior Resident Inspector, Robinson RII

\\

,

j

'

J

,

h U /D

'

Approved by:

/

N A.~Gibson, Director Ddte Signed l

!

Division of Reactor Safety

!

,

I

.

9310130418 930928 I'

PDR ADDCK 05000369<<

.

G PDR L i

,,-..

.. -.

.

5

t TABLE OF CONTENTS

'

l.

August 31, 1993, Steam Leak Inside Containment Page i

I.

INTRODUCTION - AIT FORMATION AND INITIATION........

A.

Background...................................

B.

AIT Formation................................

C.

AIT Charter..................................

I II.

EVENT DESCRIPTION..................................

A.

Event Overview...............................

B.

Detailed Sequence of Events..................

C.

Initial Conditions...........................

D.

Event Initiation.............................

E.

Operator Response............................

F.

Licensee Response............................

III.

REVIEW 0F CONTRIBUTING FACTORS.....................

A.

Management Controls..........................

B.

Maintenance Activities.......................

C.

Procedures...................................

D.

Teamwork / Command and Control.................

!

E.

Communications...............................

F.

Staffing.....................................

G.

Human-System Interfaces......................

H.

Training.....................................

i IV.

SAFETY CONSEQUENCES / SIGNIFICANCE...................

A.

ASME Code....................................

B.

Safety System Performance....................

C.

Manually Blocking Ice Condenser Doors........

'

D.

Technical Speci fications.....................

E.

Conclusions..................................

V.

ROOT CAUSES........................................

A.

Lack of Adequate Measures to Ensure Proper Val ve Re a s sembly...............................

B.

Low Sensitivity to High-Risk Evolutions.......

C.

Poor Control Room Command and Control.........

2.

Significant Event Investigation Team Evaluation...........

3.

Confi rmation of Action Letter.............................

'

.

4.

Long Term Corrective Actions..............................

5.

Exit Meeting.............................................. 30

,

h

.

"

.-

.

.-

.

.

.

.

.

,

'

,

.

,

!

Appendixes Appendix A Persons Contacted

'

'

Appendix B Procedures Reviewed and other References Appendix C AIT Charter Appendix D Acronyms

.

Figures Figure 1 Kerotest Valve Figure 2 Kerotest Parts List Figure 3 McGuire Ice Condenser Containment System

.

I

!

'!

i i

t i

J

.

.

.

.

1.

August 31, 1993, Steam Leak Inside Containment I.

INTRODUCTION -

AUGMENTED INSPECTION TEAM (AIT) FORMATION AND INITIATION A.

Background McGuire units 1 and 2 are Westinghouse 4-loop pressurized water reactors with ice condenser containment systems. The units are located on Lake Norman, 18 miles North of Charlotte, North Carolina in Huntersville, North Carolina.

Unit I and Unit 2 went commercially operational on December 1, 1981, and March 1, 1984, respectively.

On Tuesday August 31, 1993, Unit 2 was returning from a refueling outage, the core was replenished with 1/3 new fuel and the licensee intended to return the unit to service. At approximately 0144 hours0.00167 days <br />0.04 hours <br />2.380952e-4 weeks <br />5.4792e-5 months <br />, the licensee notified the NRC Headquarters duty officer of a 1" steam leak in the containment of the Unit 2 reactor. This event was recorded as event #25994.

The Resident Inspector was also notified of the event and responded to the site.

At 0900 hours0.0104 days <br />0.25 hours <br />0.00149 weeks <br />3.4245e-4 months <br /> on Tuesday, the Region II office, along with various representatives of NRC Headquarters, the Office of Nuclear Reactor Regulation (NRR), and the Office for Analysis and Evaluation of Operational Data, held a conference call with the licensee. During that conference call, the licensee indicated that operators were in the process of bringing the reactor to Mode 4 conditions to repair the leak.

The licensee also indicated that they had inadvertently exited Mode 4 and reentered Mode 3 during the process of cooling down the reactor.

Further, the licensee discussed the decision to hold, or block closed, if necessary, two doors in the ice condenser system.

These doors were opening because of a reduction in the cold air driving head that normally keeps the doors shut. This driving head was lost when the doors opened in response to the steam leak.

B.

AIT Formation On the morning of Wednesday, September 1,1993, the Regional Administrator, after further briefing by the regional and resident staff ano consultation with senior NRC management, directed f.he formation of an AIT from Region II and NRR per:nnnel.

The AIT was to be headed by a Region II Reactor Safety Division Section Chief. The bases for the formation of the AIT was to gain a clearer understanding of the management decisions and control of activities that led to the steam leak, personnel injury, containment pressurization and ice condenser activation.

.

.

.

.

.

-

'

.

.

.

.

C.

AIT Charter - Inspection Initiation Refer to Appendix C

i II.

Event Description l

A.

Event Overview

'

At approximately 0038 hours4.398148e-4 days <br />0.0106 hours <br />6.283069e-5 weeks <br />1.4459e-5 months <br /> on August 31, 1993, a radiation protection technician reported to the control room from inside the Unit 2 containment that there was a steam leak in the area around the "A" steam generator.

Operators had

,

received an " Ice Condenser Door Open" alarm just prior to

!

this telephone call. Two vendor personnel were performing a leak repair procedure on the "A" steam generator secondary i

side drain valve, 2CF-130 in the containment. Control room

operators immediately recognized an increase in containment temperature and pressure. This occurred during the leak repair when an end cap downstream of the drain valve blew off while technicians attempted to remove it and install a

-

Modified Threaded Injection Cap (MTIC) on the drain line.

'

During the steam leak, all personnel inside containment were accounted for and departed the containment. One vendor employee suffered second degree burns to his legs from the steam emission. He was not contaminated and was released from the hospital that day.

In the control room, the operators noted that containment pressure increased to +0.45 psig and that an annunciator alarmed indicating that some of the 24 ice condenser bay doors had opened. The steam leak continued throughout the event until the reactor was placed in Mode 5 at 2241 hours0.0259 days <br />0.623 hours <br />0.00371 weeks <br />8.527005e-4 months <br />. At 0100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> on September 1, i

'

1993, an MTIC was installed on the drain line and the leak stopped.

During the cooldown to Mode 4, the operators had inadvertently allowed the Reactor Coolant System (RCS)

temperature to increase above the Mode 3 limit of 350 degrees Fahrenheit.

B.

Detailed Sequence of Events Date/ Time Description 08/05/93 Mode 5 Outage Repaired valve 2CF-130 Steam Generator Drain i

valve per Work Order #92041833.

!

08/30/93 a.m.

Operations personnel note problem around pipe cap downstream of 2CF-130 and wrote a work request to repair. Some information about the damage pipe threads was mistakenly entered incorrectly on the work request.

i

.

.

.

t

a.m.

Work order written and planned by planner to repair pipe cap. The work order did not mention damaged threads on the pipe cap.

a.m.

Mechanical Maintenance technician performed a functional verification of 2CF-130 and identified leakage around the pipe cap.

1330 At this time the reactor is in Mode 3 approximately 528 degrees Fahrenheit, RCS pressure 1770 psig.

Operators commenced RCS Pressurization for

'

PT/2/A/4450/01A, (NC Leak Rate Test).

r 1400 Per procedure, operators secured RCS

'

pressurization at 2365 psig, RCS temperature 557 degrees Fahrenheit.

1645 Depressurized RCS to normal pressure. RCS pressure 2235 psig, RCS Temperature 557 degrees Fahrenheit.

08/31/93-0000 Vendor technician attempted to repair leaking pipe cap using on-line leak seal method.

0038 Control room annunciator " Ice Condenser Lower

!

Inlet Doors Open" alarms. Radiation Protection i

reported that pipe cap had blown off drain line at valve 2CF-130. One vendor was burned and

,

transported to University Hospital. No contamination was present.

i Called Operations Unit Manager to site.

0115 Commenced RCS depressurization due to Containment pressure greater than 0.3 psig.

Containment Average Temperature reached 133 degrees Fahrenheit; Pressure reached 0.45 psig.

Entered required action for Technical Specification 3.6.1.4.

0140 Ice Condenser temperature greater than 27 degrees Fahrenheit; declared Ice Bed inoperable in Technical Specification Action Item log (TS 3.6.5.1).

0145 Called Superintendent of Operations to site.

.

_. -..

l

.'

'

.

0300 Shift Manager dispatched an ice condenser team to evaluate ice condenser system.

I 0348 Shift Manager reported that bay 21 had indication of ice melting. One or two doors were cracked open and would not close.

0445 Shift Manager reported that up to 6 feet of ice i

had melted in bay 21 and that 4 doors are not

'

fully closed.

0500 Decision from the Control Room to hold closed the two ice condenser doors in bay 22 by personnel.

Doors in bay 22 were declared inoperable and listed in the Technical Specification Action Item Log.

0650 Unit entered Mode 4, less than 350 degrees Fahrenheit.

Attempted to install pipe cap on leaking drain.

Attempt was unsuccessful due to the volume of steam leaking.

It did not appear that the valve 2CF-130 was slowing the flow at all.

0735 During operator turnover entered Mode 3 due to inadvertent heat up. Approximately 357 degrees Fahrenheit.

0835 Unit re-entered Mode 4.

-0900 Bay 22 Ice Condenser doors no longer require -

,

blocking.

0926 Exceeded administrative limit for Pressurizer cooldown. Cooldown remained within Technical Specification limit.

1245 Decision made to continue cooldown to Mode 5.

2240 Entered Mode 5.

09/01/93 1400 Successfully completed installation of pipe cap (MTIC) on 2CF-130. All leak flow ceased.

,

C.

Initial Conditions On August 30, 1993, the licensee was in the process of returning Unit 2 to service after a 60 day refueling outage.

_ _

,

.

.

~

5 The outage was shortened by 10 days and licensee personnel f

were in the process of conducting a containment walkdown.

At 1430 hours0.0166 days <br />0.397 hours <br />0.00236 weeks <br />5.44115e-4 months <br /> on August 30, letdown containment isolation valve 2NV-1 unexpectedly shut. This required operators to place the Excess Letdown System in service. Operators were keeping RCS pressure high enough to ensure the proper Excess Letdown flow rate.

During the heatup process, two steam leaks were noted inside containment. One leak was on 2NC-14, Letdown System Manual

'

Isolation valve and another on 2CF-130,

"A" Steam Generator Drain valve. Operations personnel observed steam issuing from the pipe cap downstream of 2CF-130. At that time, Operations developed a work request to repair the pipe cap.

A Mechanical Maintenance Technician inspected the pipe cap in accordance with the work order and noted the leak around i

the pipe cap.

The operations person had recognized that the pipe cap nipple had damaged threads. This observation was

inadvertently erased by the work request computer and the

'

work request incorrectly suggested that the valve was damaged and omitted the information on the pipe threads.

Consequently, this critical piece of information was not conveyed to the ver. dors who would be scheduled to perform the leak repair procedure.

A plan to repair the leaking pipe cap was developed by a component engineer and sent to the work control center for approval.

Four methods of leak repair options were considered. The plan chosen called for vendor personnel to use a MTIC to seal the end cap and stop the leak.

Operations personnel were informed that leak repair would occur inside containment.

D.

Event Initiation i

During the review of this event, the AIT interviewed the vendors who attempted to repair the leaking pipe cap downstream of 2CF-130 and the operators who responded to the event (Appendix A). The purpose of these discussions was to determine what happened during and immediately after the event with regard to vendor actions and operator actions.

The following outline of events was generated by those

-

interviews and is not meant to be a totally exhaustive description of all actions surrounding the response to the event.

At approximately 0000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> on August 31, 1993, two vendor personnel entered containment to begin work on the leaking pipe cap downstream of 2CF-130.

Simultaneously, co-workers were repairing a steam leak from 2NC-14, Letdown System Manual Isolation valve in the same area. Also, a i

.

.

-

'

-

.

t

4 containment pre-walkdown was in progress. The two vendors observed steam flow around the pipe end cap. One of the workers deflected the steam plume by placing a wrench in the

-

path of the steam flow to enable observation of the

'

condition of the pipe end cap. This was only marginally effective and the worker could not determine the condition of the pipe cap.

In accordance with the leak seal procedure, the worker checked 2CF-130 in the closed position. His co-worker verified this check. Also in i

accordance with the procedure, they attempted to tighten the

pipe cap to ensure proper installation.

!

After the preliminary checks were completed, the vendors began removal of the end cap. There was significant structural interference with this process. The worker could only turn the cap about an eighth of a turn before he had to reposition the pipe wrench. While he was turning the pipe cap the worker observed the steam flow.

The flow had not diminished and did not increase. After approximately 1 to 1-1/8th turns the pipe cap separated from the pipe and a large steam plume erupted from the pipe.

Both workers immediately turned to leave the area. A co-worker who was working on 2NC-14 moved towards the area of 2CF-130 to

,

determine the extent of the personnel hazard. All of the ver. dor workers assembled and immediately left the area.

'

While leaving the area one worker was burned by the escaping steam / water mixture.

Operators in the control room first became aware of the steam leak by an alarm indicating the ice condenser doors were open. This was followed immediately by a rapid increase in containment pressure and temperature. Personnel

!

performing a Rod Drop Time Test in containment also called

,

the control room to inform them of a large noise and

'

possible steam leak inside containment. Operations ordered a containment evacuation.

During the containment evacuation, all personnel inside containment assembled at the personnel airlock where an

'

accountability survey was conducted. Operations personnel also responded at the airlock to help with the evacuation.

i All personnel were accounted for and allowed to leave the containment building. The injured worker went to the dressing area where he stayed for several hours with ice on his injuries before being transported to University Hospital.

E.

Operator Response Following the call, at 0038 hours4.398148e-4 days <br />0.0106 hours <br />6.283069e-5 weeks <br />1.4459e-5 months <br />, from Radiation Protection

,

in the reactor building notifying the control room of the

'

steam leak on the drain line (2CF-130), the control room l

.

.

4

operators immediately entered AP/2/A/5500/01, Steam Leak procedure. The operators noted that reactor building pressure had increased and peaked at approximately 0.45 psig. Based on the containment pressure, a Technical Specification cooldown and depressurization was initiated at 0115 hours0.00133 days <br />0.0319 hours <br />1.901455e-4 weeks <br />4.37575e-5 months <br /> in order to comply with Technical Specification 3.6.1.4.

.

Performing the Steam Leak procedure eventually led the operators to shut down the unit in accordance with OP/2/A/6100/02, Controlling Procedure for Unit Shutdown, Enclosure 4.2.

The operators were unable to perform a normal shutdown due to the closing of Letdown Isolation valve 2NV-1 power supply the day before. This complicated the shutdown, in that, normal Letdown was not available and the plant was on Excess Letdown. Operating on Excess Letdown reduced the Letdown flow rate and hindered unit maneuverability by limiting RCS temperature control. Using Excess Letdown required a higher RCS pressure. Also, RCS temperature must be maintained above 320 degrees Fahrenheit for Low-Temperature-Over-Pressure concerns.

However, RCS temperature cannot exceed 350 degrees Fahrenheit to remain in Mode 4 conditions.

It was the operators' impression that operating with the plant in this condition caused a tighter control band at higher RCS pressure and lower RCS temperature than is normally encountered.

The control room operators initiated the reactor coolant system cooldown in a controlled fashion and were diligent in maintaining cooldown rates to within administrative and Technical Specification limits.

The Unit entered Mode 4 at 0650 hours0.00752 days <br />0.181 hours <br />0.00107 weeks <br />2.47325e-4 months <br /> on Tuesday, August 31, 1993.

Plant parameters were maintained in the allowable region of the Over-Pressure Protection (Low Pressure Setpoint) curve.

Operators noted plant conditions; RCS pressure approximately 1800 psig and RCS coolant temperature less than 350 degrees Fahrenheit. Upon Shift turnover at approximately 0730 hours0.00845 days <br />0.203 hours <br />0.00121 weeks <br />2.77765e-4 months <br />, the off going crew secured the cooldown by closing the steam dumps using the steam dump potentiometer. Once the closed position was reached, the valves were closed a little more. This was to ensure the Over-Pressurization Protection Curve was not violated during the shift turnover.

The oncoming operator was worried about violating this cooldown curve since the plant was at a higher pressure than he was accustomed. The operator at the controls did not closely monitor plant conditions as was required while in this mode of operation. When realizing the plant started to heat up the operator adjusted the setpoint potentiometer and did not observe a positive indication of RCS cooldown. This was done on two occasions without his observation of plant

-

.

.

.

.

8 response. At approximately 0745 hours0.00862 days <br />0.207 hours <br />0.00123 weeks <br />2.834725e-4 months <br />, the plant entered Mode 3 due to the inadvertent heat up.

The control room operators were not aware of the Mode change from Mode 4 to Mode 3 until a report from the reactor building questioned why the ambient temperature and leak rate from 2CF-130 were increasing in the lower level of reactor containment. At this point the Operations Staff Manager printed out a graph of the heatup of the unit. When this print out was provided to the operators it was realized that a mode change had occurred. The operators once again commenced a reactor system cooldown in response to the heatup. At approximately 0835 hours0.00966 days <br />0.232 hours <br />0.00138 weeks <br />3.177175e-4 months <br />, the unit entered Mode 4.

During the subsequent cooldown, the administrative limit of 64 degrees Fahrenheit per hour for pressurizer cooldown was exceeded and the Operations Shift Manager was notified.

The operators did not exceed the Technical Specification limit for pressurizer cooldown.

Plant depressurization did not occur until temperature and

,

pressure were within the Reactor Coolant System cooldown limit curve, 1.6, of OP/2/A/6100/22 Enclosure 4.3, this commenced at approximately 308 degrees Fahrenheit. At approximately 1301 hours0.0151 days <br />0.361 hours <br />0.00215 weeks <br />4.950305e-4 months <br />, the operators used an air jumper on 2NV-1, Letdown Isolation valve, to open the valve and establish normal Letdown. This was performed to pressurize Letdown when placing the Residual Heat Removal system in service.

F.

Licensee Response 1.

Notifications After the event, the licensee made two 10 CFR 50.72 notifications to the NRC. One notification was made under 10 CFR 50.72(b)(1)(1)(A), shutdown in accordance

-

with Technical Specifications. The other notification was made under 10 CFR 50.72(b)(2)(ii) Engineered Safety function (ESF) operation. The AIT reviewed the Emergency Action Levels (EALs) provided in RP/0/A/5700/00 "McGuire Nuclear Station Emergency Action Levels," Enclosure 4.1, and concluded that a Notification of Unusual Event would be appropriate for the steam leak event reviewed in this Report.

There were two situations in Enclosure 4.1 that may require an emergency classification of a Notification of Unusual Event.

Both of these are classified under Event #4.1.10, "Other Abnormal Plant Conditions." The first states that an Unusual Event is declared if:

l l

,

.

.

.

.

"Less than minimum channels of ESF function operable. AND Load reduction or plant cooldown initiated in accordance with Tech Specs."

In this event, with the single ice condenser system inoperable, it could be interpreted that all channels of an ESF system were inoperable and a plant cooldown was initiated in accordance with Technical Specifications. Operators did review the EALs after this event. When they evaluated this criteria they interpreted this as applying only to instrumentation

,

channels and not " systems."

,

Another criteria existed under Event #4.1.10, "Other Abnormal Plant Conditions," that might be applicable

!

for the steam leak event reviewed in this Report.

This criteria states that a notification of Unusual

,

i Event will be made if:

"Both trains of any ESF function found inoperable (if caused by fire see event #4.1.7 -

Fires and Security Actions, Site Area Emergency

'

Classification). AND Load reduction or plant cooldown initiated in accordance with Tech Specs."

After reviewing both of these criteria, the AIT concluded that this event appeared to meet the intent of the EALs as a Notification of Unusual Event.

2.

Short-Term Corrective Actions

'

In response to the causes of the steam leak evaluated in this report and identified in their own evaluation, the licensee issued a report to the NRC on September

-

5,1993, outlining the following short-term corrective actions:

a.

Chartered a Mechanical Maintenance investigation of the event.

!

'

b.

Completed a valve disassembly investigation to determine the cause of mechanical failure of 2CF-130.

!

c.

Modified the Kerotest corrective maintenance procedure to add

,

clarifying steps.

i

.

.

.

-

]

j d.

Placed an administrative hold on the use of the vendor's on line leak seal repair method for pressurized

-

pipe caps on high energy systems.

e.

Placed the maintenance procedure for i

on-line leak sealing on administrative hold.

l f.

Clarified management oversight responsibilities for on-line leak i

sealing activities.

g.

Briefed Mechanical Maintenance managers to heighten their awareness

'

regarding training and qualification of vendors, contractors and interface personnel.

,

h.

Issued a Special Order to suspend

,

removal of pipe caps from high energy systems.

i.

Modified the Post Maintenance i

Testing on 2CF-130.

j.

Issued a memo to all Operations

{

personnel requiring that Operations accompany vendor personnel when interfacing with Operations for leak

repair activities.

k.

Modified shift briefings to reduce the number of personnel in the control room.

1.

Documented the decision to hold two ice condenser doors closed during

)

the recovery from this event.

m.

Reviewed all Kerotest valve maintenance worked during the outage e

to ensure proper assembly.

n.

Replaced solenoid wires on 2NV-1 and perform functional checks.

The AIT concluded that these short-term measures were adequate.

I

,

.

.

a

III.

REVIEW 0F CONTRIBUTING FACTORS A.

Management Controls The AIT evaluated the management controls which allowed the unsuccessful attempt to perform a temporary leak repair on the pipe cap downst am of valve 2CF-130. To fully evaluate i

the applicable management controls the AIT first analyzed the maintenance history of the valve for event precursor information.

On March 14, 1992, the operations staff originated Work Request #146808 to repair a leak in the pipe cap downstream of valve 2CF-130.

Later that day a mechanical maintenance technician loosened the cap, applied thread sealant and re-tightened cap. He noted that the pipe threads were damaged.

The cap still leaked after his efforts.

On March 15, 1992, a vendor technician removed the pipe cap, cut off about 3/8" of damaged threads, thread-chased the pipe nipple, and installed a new pipe cap. The cap still leaked.

On March 16, 1992, the same vendor technician removed the

pipe cap, installed a leak repair pipe cap (MTIC) and

,

injected it with sealant which stopped the leak. Work

'

Request #146808 was converted to Work Order #92041833, to repair the valve during the next outage.

On the afternoon of August 4,1993, a mechanical maintenance technician began disassembly of valve 2CF-130 pursuant to Work Order #92041833. On the following morning, a different mechanical maintenance technician completed the re-assembly of the valve and signed in the procedure that it had been cycled.

Later that same morning, a different mechanical maintenance technician verified that the pipe cap had been reinstalled.

On August 8, 1993, operators began filling the 2A steam generator.

On August 30, 1993, a non-licensed operator (NLO) discovered that the pipe cap was leaking. The reactor was at full operating temperature and pressure at this time, and the steam generator pressures were at approximately 1100 psig.

Later that morning, an operations shift supervisor i

instructed an NLO to verify that the valve was closed and tighten the pipe cap.

The NLO attempted to tighten the valve in the closed

'

direction but got no movement.

He then tried to tighten the

I

.

.

$

.

,

'

pipe cap. The leak did not change and he stopped attempting to tighten the cap because he determined that the threads were damaged.

'

Later that same morning, an operations staff engineer originated Work Request #93039766 to leak repair the pipe

,

cap. He included the information about the pipe threads being damaged, but placed some of the information about the damaged pipe threads in the incorrect location on the computer generated Work Request form. Thus when a planner originated Work Order #93063230 to perform the leak repair, the information about the damaged threads was not identified on the Work Order.

Shortly after midnight on August 31, 199, a vendor technician attempted to leak repair the pipe cap using procedure HP/0/A/7650/077, On-Line Leak Sealing Initial Injection. During repair efforts, the pipe cap ejected, due to what was subsequently determined to be damaged /

inadequate cap thread engagement.

In summary, the licensee knew: the pipe threads were damaged

.

in 1992; the pipe nipple was cut off and threads chased in

'

1992 but still leaked; the pipe cap was leak repaired in 1992; an operator had reported that the cap / nipple had damaged threads on August 30, 1993; the valve was leaking by; and the system was at full temperature and pressure.

Furthermore, the work order to leak repair the pipe cap stated that the pipe cap would be removed.

The AIT evaluated the review and approval process for the Work Requests / Work Orders to determine the level of management involved in this maintenance evolution.

Work Request #93039766 to leak repair the pipe cap was generated by an operations staff engineer.

Work Order #93063230 was generated from Work Request #93063230 by a maintenance planner.

The operations approval to begin work was given by an

,

operations support engineer.

Before beginning work, the

vendor tech contacted the shift operations manager.

It should be noted that the shift operations manager was the acting plant manager at the time because the plant manager and all superintendents were off site. Thus the highest

,

ranking member of management involved in this decision was the acting plant manager.

After reviewing the pertinent evidence, and discussing the sequence of events with the personnel involved, the AIT

-

_,.

.

.-

...

.

..=

.

.

-

-.

.

.

.

.

.

concluded that management decisions to undertake this high risk maintenance activity were made without the benefit of important facts regarding the safety consequences of leak i

repair plan. Supervisory review in this process was pro

,

forma rather than active involvement. The AIT concluded

that the rationale behind this tacit approval approach was j

based in large part.upon the licensee's belief that " leak i

.coair" is a benign evolution and their trust in the vendor

'

who, for the most part, operates as would a licensee erployee.

B.

Maintenance Activities Valve 2CF-130, Figures 1 and 2, Steam Generator Shell Side

Drain, is a 1 inch nominal pipe size, stainless steel body,

!

angle globe valve manufactured by Kerotest Manufacturing j

,

Company.

It is a manual valve used to drain (to the

t containment sump) the "A" steam generator shell. Valves of

!

this type are commonly employed as manual drain and vent l

valves at McGuire.

In the case of valve 2CF-130 it is an

American Society of Mechanical Engineering (ASME) Class 2

,

l component.

j

.

McGuire Work Order #92041833 covered the outage maintenance activities for the repair and testing of the subject valve.

j Contained in the work package were detailed maintenance i

instructions and a list of references for in-place

'

corrective maintenance of Kerotest globe valves. These i

detailed instructions included step-by-step check-offs and

Quality Control hold points.

l Upon disassembly, the mechanic noted some scratches on the i

valve seat. These scratches were removed by lapping, a

{

commonly used repair technique for slightly damaged valve.

seats. The lapping procedure was detailed in the work order

<

package. The disc seating surfaces were found to be in an unacceptable condition. Consequently, it was replaced with i

a new disc assembly.

.

Verification of proper disc to seat contact required the l

disc to be disassembled. The disc assembly consisted of i

four pieces: disc, disc cap, spring and spring guide. The i

disc assembly for this valve has the general appearance of a i

spring check type disc. Disassembly of the disc allows

-

verification of seat contact without presenting the

difficulty of having to compress the disc spring to make

!

.

seat contact. The' mechanic verified proper seating of the

!

[

disc by obtaining a 360 degree seat contact.

-

>

k An error occurred during reassembly of the disc assembly.

l i

This involved inversion of the spring and spring guide when i

reinstalled on to the disc cap and disc. The assembly was i

,

~f

-

,

..

.

.

-

,

-..

.

..

.\\

.

-

.

-

-

.

.

.

.

i

then inserted into the valve body with proper orientation of the disc toward the seat. However, the inverted spring and spring guide pieces escaped notice and presented no immediate indication of mis-assembly.

Reassembly of the valve proceeded according to the detailed instructions.

The instruction step for assembling the disc assembly was weak in that it was not detailed (" reassemble disc"). There was no instruction to verify proper reassembly of the disc assembly. The incorrectly assembled disc did not present a radically different appearance when the AIT later compared it to the valve sectional diagram during this investigation.

The valve assembly diagram is part of the repair package, but it was some pages behind the text. Also, its use was not referenced in the work package.

A Quality Control technician was present during the reassembly. The procedure called for the technician to verify the valve disc dimensions after assembly. He also was to verify certain other general steps.

During the reassembly the Quality Control technician became involved in the rebuilding process. He helped the mechanic by hand.ng him tools and helping with other parts of the task.

i After valve assembly was completed, there was a specific instruction, with check-off, to verify mechanical operation by opening and closing the valve (stroke the valve). This item was initialed by the mechanic. The Quality Control technician also verified that the valve was stroked after reassembly.

This step is key in determining the proper assembly of the valve.

Ensuring the valve can be fully stroked confirms proper installation by the mechanic. There was substantial evidence that the valve could not be stroked by the mechanic after he incorrectly reassembled the valve. This conclusion was drawn from observations discussed below which were made during the failure analysis of the valve conducted in the plants' " hot" shop.

Prior to disassembly in the hot shop, it was noted that the valve stem was frozen in position. The valve could be neither opened nor closed, even though vitual inspection of

'

the seat and disc area through the valve discharge nipple revealed the valve to be roughly 1/2 to 3/4 open.

Disassembly revealed the fact that the inverted spring guide effectively blocked the valve disc in the open position.

,

The disc was held away from the seat by the spring retainer

"

acting as a rigid spacer between the disc cap and a land on the inside of the valve body.

Furthermore, the valve could not be opened. That was due to the yoke not being properly

.

.

.

.

.

i

,

seated against the intended valve body land.

Instead, the stem assembly became jammed on the now rigid, out of position disc assembly. When the yoke was torqued to the specified 800 foot-pounds, the torque jammed the stem against the disc. This froze the valve stem in one position and prevented any stroking of the valve.

Even if the valve

-

stem were somehow not jammed, the valve still would not have operated. Once the yoke was removed from the valve body, the valve stem moved freely. Jamming of the stem against the blocked disc was further verified by later reassembly of the as-found components.

Incorrect assembly of the disc assembly made it impossible to stroke the valve.

The final weakness in the licensees' overall procedure ensured that gross leakage through the valve would escape detection until a unit start-up was attempted. Since the valve is 1 inch nominal pipe size, it is exempt from a post-repair ASME Section XI hydrotest, per paragraph IWA-4400(a)(5) of the 1980 edition, the ASME Code of record for Unit 2.

Thus, the licensee found no requirement to perform

'

a post-repair leak test. The only surveillance for leakage in this case is the post outage leakage inspection, conducted during Mode 3, under Work Order #93012110.

Leakage during preoperational actions, such as filling the steam generators, escaped notice. This was due to the presence of the installed pipe cap on the discharge nipple of the valve. A pipe cap is routinely attached to a threaded nipple on the discharge of all vent and drain valves to capture small nuisance weepage past the valve seat. This is common power p' ant housekeeping practice.

However, in this case, the pipe cap masked the fact the valve was effectively open, and its presence led the licensee to assume the valve was closed and only weeping.

C.

Procedures The AIT reviewed the procedures associated with this event and concluded that, other than the limited instruction in the maintenance procedure for kerotest valve reassembly, procedures were adequate and used properly. However, the AIT made the following observations regarding the procedures: McGuire's procedures were not substantially different from other licensee's procedures in these areas;

!

there was no procedural control of pipe cap installation or removal; and there was no preestablished criteria, guidance, or procedure regarding the holding shut of ice condenser doors.

,

D.

Teamwork / Command and Control The AIT evaluated the human factors aspects associated with

\\

l

.

.

i

\\

.

l

l the event including teamwork / command and control. McGuire i

operators, operations staff, maintenance personnel, and

,

vendor personnel involved in the event were interviewed, and equipment human-system interfaces, procedures, training / qualification records, lesson plans, and logs associated with the event were reviewed.

A strong sense of cooperation, mutual respect, and support was exhibited by each shift crew involved in the event.

!

However, poor command, control, and communications significantly inhibited the shift crew's ability to effectively act as a team in monitoring and controlling shutdown plant operations.

Significant weaknesses were noted in plant shutdown

,

operations command and control.

Supervision and clear control of plant maintenance and operations was weak.

'

The licensee exhibited poor control of the temporary leak repair of the drain pipe cap for valve 2CF-130. None of the

,

on-shift control room operators were cognizant that the

'

repair procedure involved pipe cap removal and installation of a MTIC vice a lower risk, routine repair technique where the pipe cap is drilled and leak sealant injected into the pipe of the existing installed cap. Although knowledgeable of the repair being scheduled for the shift, none of the on-shift control room operators was aware of the commencement of the work which was coordinated by the Shift Manager i

outside of the control room. No briefing was conducted detailing this procedure. No requirement was in place for

!

plant maintenance personnel to monitor the temporary leak repair. Once having been cleared to enter containment, vendor personnel worked in the containment independent of plant supervision. The Unit Operations Coordinator and the

Planner were not fully sensitive to the risk involved and did not ensure adequate controls were in place to perform a i

l safe leak repair.

l The licensee failed to exhibit adequate command and control

concerning the holding shut of the ice condenser doors.

After the containment pressure had returned to the normal i

range following 1% steam leak, the operations management present in the co.A L room (Superintendent of Operations, Operations Staff Manager, Shift Supervisor, and Shift Manager) discussed the concern that personnel were reported to be holding shut ice condenser doors and such action needed to be evaluated with respect to Technical Specifications. The decision was made to hold ice condenser doors shut and mechanically block them shut as necessary.

This was intended to conserve ice and enhance the ability of the ice condenser to respond to any further challenges.

The Superintendent of Operations ensured concurrence of the l

.

.

-

,

L

,

'

Shift Supervisor, the licensed operator on shift, and the i

discussion concluded with the Superintendent of Operations, Operations Staff Manager, and Shift Supervisor understanding that the ice condenser doors would be held shut or blocked shut. Contrary to their understanding, the Shift Manager believed that the ice condenser doors could and should be

,

allowed to open and then be shut with each occurrence. As the director of this evolution, the Shift Manager left the control room and proceeded to give positive direction that the ice condenser doors were to be shut whenever they

'

opened.

He did not require them to be held shut

'

continuously.

Consequently, the ice condenser work party in containment shut the doors whenever they opened until approximately 1100 hours0.0127 days <br />0.306 hours <br />0.00182 weeks <br />4.1855e-4 months <br /> when the doors remained shut of their own accord. The on-shift control room operators remained unaware of the specific action being taken by the

,

ice condenser work party. The following day, licensee management determined that, except for one person leaning on a set of doors for a brief period of time, the doors were r

never held shut nor blocked shut until the plant was in Mode 5.

.

During the inadvertent plant heat up, the Shift Supervisor and Control Room Senior Reactor Operator (CRSRO) failed to provide the control room operators with clear direction of activities affecting plant operation and failed to keep themselves informed of plant status as required by Operations Management Procedure 1-13, " Conduct of Operations."

In particular, these supervisors did not provide adequate direction to control room operators regarding the plant

,

l temperature / pressure parameters to be maintained.

Additionally, both the Operator at the Controls (0AC) and the Balance of Plant (B0P) operator used poor judgement and i

failed to maintain the plant less than 350 degrees

Fahrenheit to prevent transition to Mode 3.

The increase in 2CF-130 leak rate prompted a containment work crew to recognize changing plant conditions and notify the operations staff. The operations staff, in turn, provided the Shift Supervisor and CRSRO with their initial indication of a plant heat up.

The BOP had recorded temperature i

greater than 350 degrees Fahrenheit, and the OAC, cognizant

'

of plant temperature increasing, was taking action to decrease temperature which peaked at 366 degrees Fahrenheit.

However, neither recognized the transition to Mode 3.

,

Subsequently, the Shift Supervisor, CRSRO, OAC, and B0P

'

discussed plant status and recognized that the plant was in Mode 3.

l Several factors contributed to poor operations command and

'

control during the inadvertent heat up.

The crew had i

.

.

'

!

.

returned to assume the shift expecting to start up vice

cooldown, and the inadvertent heat up occurred during the shift turnover and initial crew briefing. During AIT

conducted interviews, the oncoming shift operators stated

these expectations and that there was a high work tempo at the beginning of the shift.

Turnover checklists were completed and reviewed, and turnovers were conducted individually by position in

-

accordance with Operation Management Procedure 2-2, " Shift Turnover." However, the times for turnovers were set by mutual agreement between the two operators (within a reasonable time frame for shift relief) and without report of relief to the Shift Supervisor or CRSRO. After turnover, the shift crew briefing was conducted in accordance with procedures, however, no mention was made of any upper temperature limitation. The briefing was conducted in the control room with all shift licensed and non-licensed operators.

There was an indication of inappropriate task assignment in that the OAC appeared to be overburdened during the heatup event. The OAC was responsible for obtaining and understanding pertinent information discussed in the briefing while simultaneously monitoring and controlling plant parameters. Additionally, he was involved in coordinating with Radiological Protection personnel and directing the frequent pumping of the in-core instrument sump and the containment floor and equipment sump which was necessitated by the leak. The OAC stated in an AIT interview that he "was very busy in the crowded control room." Furthermore, due to the non-routine operation of the plant on Excess Letdown with elevated pressure during the plant cooldown, he was focused on maintaining temperature above the over-pressure protection limit of 320 degrees Fahrenheit (referring to Reactor Coolant System Cooldown Li.'tations Curve 1.6 of OP/2/A/6100/22, Unit 2 Data Book)

,

and was not concerned with the 350 degree Fahrenheit transition point to Mode 3.

E.

Communications

,

'

Prior to and during the event, significant weaknesses were noted in plant oral and written communications.

The threads on the drain pipe cap for valve 2CF-130 were damaged and part of the threaded pipe was cut off in the reworking and initial installation of the pipe cap. This

,

information was not effectively communicated to the vendor personnel performing the pipe cap removal in either the lengthy work package or by briefing. Vendor personnel

,

assumed proper pipe cap installation.

In fact, the vendor

'

.

.

.

personnel stated that they would not have performed that particular work item if they had foreknowledge of the damaged threads.

Furthermore, the work request for repairing the leak (in 1993) on the pipe cap was incorrectly entered with the work request title continuing onto the problem statement line. This led to a error with the work package description of the problem.

Consequently, information was lost in the computer and there was inadequate corrective action in that the pipe threads were not repaired. Additionally, it led to the mischaracterization of thread condition which could have informed the vendor personnel of the potential problem with the pipe cap removal.

Poor communications were exhibited regarding the ice condenser operation and the holding shut of the ice condenser doors.

Initial reports of ice condenser doors opening incorrectly misidentified the doors to bay 22 as the

<

doors to bay 21.

Furthermore, Operations Management Procedure 1-12, " Operations Communications Standards,"

provides clear guidance on general communications practices.

However, the Shift Supervisor and Superintendent of Operations failed to provide clear direction to the Shift Manager regarding the holding shut of ice condenser doors.

This resulted in the Shift Manager giving direction other than that expected by the Shift Supervisor and Superintendent of Operations.

Poor control room communications practices were demonstrated during the inadvertent plant heat up. During turnover the

.

'

0AC operators failed to address all temperature limits for the plant, and during the shift briefing the Shift Supervisor and CRSR0 failed to communicated the plant's temperature band to the operators. As a result the oncoming OAC was unaware of the need to control temperature less than 350 degrees Fahrenheit. Additionally, the OAC failed to inform the Shift Supervisor and CRSRO that the plant was heating up.

F.

Staffing Shift staffing during the event met both 10 CFR 50.54(m) and plant administrative requirements.

Excessive working hour or fatigue did not appear to be factors in this event, i

However, the duties and responsibilities assigned to the OAC may have been excessive to the point that he was distracted from his responsibility to monitor instrumentation displays and know and comply with limits designated in the Operating Procedure.

.

.

.

.

.

G.

Human-System Interfaces Human-system interfaces associated with plant indications and controls associated with this event were adequate.

Specifically, there were no human-system interfaces that significantly contributed to the inadvertent heat up d ring

shift change. Although, there was no direct control room indication of plant mode; there was digital, analog, and

plant computer indication of plant temperature.

The AIT noted that on the ice condenser door monitoring system's local indication panel most of the zone indications of open/ shut doors and the personnel access door are not

.

clearly labeled.

Furthermore, the switches on the i

individual ice condenser doors do not provide a clear status

'

of individual door position in that the doors may indicate shut while actually unseated and allowing audible air flow which in this event was reported as an open door.

H.

Training The AIT identified three specific training weaknesses, they were: (1) Both crews' OACs demonstrated a weakness in their understanding of steam dump operation during cooldown. They received steam dump system training in April 1991 and are scheduled for review in September / October 1993. (2) The OAC failed to effectively observe and respond to plant feedback from his manipulations which resulted the heat up peaking at i

366 degrees Fahrenheit. (3) All the control room operators

,

failed to recognize indications of the plant changing from

Mode 4 to Mode 3.

A plant cooldown per the normal shutdown

operating procedure was included in Lesson Plan OP-MCC-SRT-N01 which the crew exercised in licensed operator requalification training in January 1993. Several operators stated in interviews that more realistic simulator training on plant cooldown which incorporates the numerous routine i

evolutions and communications would be beneficial.

IV. SAFETY CONSEQUENCES / SIGNIFICANCE

-

A.

ASME Code Repair plans similar to the one attempted for the pipe cap downstream of 2CF-130 are beyond the applicability and

'

intent of the ASME Code. This is because the usual

,

application for such methods is the temporary alleviation of gasket / packing leaks.

Per the ASME Code, these are non-structural elements. Consequently, they are exempted from many ASME Code requirements, one of them being leak-tightnes )

.

e

~

,

An area not addressed by the ASME Code is the performance of maintenance / modifications on an operating system. The ASME i

Code, by implication, assumes a system is either under construction or not operating. No Code rules exist for repairs performed during operation.

The regulatory status of these methods has been dependent upon the ASME Code status.

Licensees often use an ASME Code exclusion for gaskets to employ on-line leak repair methods, therefore, there is little regulatory applicability for these methods. After review of the repair procedure attempted for 2CF-130, the AIT found no ASME Code concerns.

B.

Safety System Performance Safety System performance was reviewed by the AIT.

Specifically, the AIT reviewed McGuire Technical Specifications and Final Safety Analysis Report (FSAR) to determine the ice condenser functioned appropriately in response to this event. Additionally, the AIT reviewed Technical Specifications applicable to the ice condenser and the conditions of the plant to determine whether the specifications or lack of specifications contributed to the handling of the event.

Plant conditions prior to time of the 2CF-130 main steam

leak were: RCS pressure 1770 psig and RCS temperature 528

'

degrees Fahrenheit. Thus, the Unit was operating in Mode 3 in accordance with OP/2/A/6100/01, Controlling Procedure of Unit Startup. The AIT determined there were no limitations on the ice condenser or ice condenser doors prior to the event.

The AIT determined that the ice condenser (Figure 3)

operated as designed.

It was determined that 20 of 24 doors came off their open seats and two of those 20 doors would not remain closed once building pressure had decreased to normal parameters. One group of doors (four bays) did not indicate open when verified on the Inlet Door Monitoring System annunciator panel located inside containment.

It was

determined that these four doors did not move sufficiently to cause the limit switches to indicate an open door in that grouping of doors.

This observation was not noted until visual inspection by the ice condenser crew after entry into the lower ice condenser to inspect door position and condition of the ice condenser. At the time of the event it was believed that the four bays of doors did not open.

However, these doors did move slightly due to the increased building pressure. Upon lower ice condenser inspection it was noted by the ice condenser crew, that a " whistling" noise was heard in the vicinity of the doors. The doors did not move enough to allow the instrumentation to make up to

,

.

.

.

.

.

.

,

give an open indication. This " whistling" would be representative of the doors being repositioned by the increased building pressure. The other 20 bay doors opened as expected.

Most of the doors were only cracked open, but there were four doors (two bays) that opened from one to two feet. The doors to bays 21 and 22 would not close on their

'

own accord. The AIT determined this based upon a subsequent interview with the ice condenser crew supervisor.

The AIT reviewed the FSAR Chapter 6.2.2.8, for design operation of the ice condenser inlet doors.

It was determined that the door panels are provided with tension spring mechanisms that produce a small closing torque on the door panels as they open. The magnitude of the closing torque is equivalent to providing approximately a one pound per square foot pressure drop through the inlet ports with the door panels open to a position equivalent to the full port flow area. The zero load position of the spring mechanisms is set such that, with zero differential pressure across the door panels, the gasket holds the door slightly open. This setting provides assurance that all doors will

'

!

open slightly, upon removal of cold air head, therefore eliminating significant inlet maldistribution for very small incidents. All doors will open to allow venting of energy to the ice condenser for any leak rate which results in a divider deck differential pressure in excess of the ice

'

condenser cold head. The FSAR states that under zero differential pressure conditions all doors remain 3/8 inch open. This is consistent with the performance of the doors during the event.

Further, the AIT determined from data taken during PT/0/A/4200/32, " Periodic Inspection of Ice i

Condenser Lower Inlet Doors," performed on August 25, 1993, that the doors associated with bay 22 had the lowest pressure test for opening.

C.

Manually Blocking Ice Condenser Bay Doors Manually snutting the doors in bays 21 and 22 actually helped preserve the integrity of the ice. After pressure

<

and temperature had returned to near normal values the shift crew decided to hold the doors shut. The conditions of the

<

plant indicated that the leak had decreased and the

,

continuing energy addition could and was being removed by

the Containment Ventilation system.

In accordance with FSAR Chapter 6.2, " Containment Systems", the Containment structure was designed to be able to withstand a design

basis accident with the following assumptions and

'

conditions:

(1) The design blowdown is 324.2 E6 British

!

Thermal Units (BTUs) and (2) a mass of 498,200 pounds is delivered into the containment. This analysis assumes a

,

.

- - -

-

.

.

.

W

preexisting reactor core power of 102% of rated thermal power for decay heat generation. Hot metal energy was also a consideration in the analysis.

Actual plant conditions after the event were as follows:

The decay heat load was very low due to the refueling outage, and the stored energy in the RCS was much reduced due to the cooldown to below 400 degrees Fahrenheit.

Blocking two of the 24 doors would only minimally decrease the functional capability of the ice condenser, since the remaining 22 doors were still functional / operational. With the small loss of ice coupled with continued localized melting of bay 22, the decision to hold the doors shut was determined to be an acceptable method to prevent further ice condenser bed degradation.

If a loss of Coolant Accident (LOCA) had occurred while these two doors were held shut, the ice condenser would still have performed its intended function. A large break LOCA with the reactor coolant system at 400 degrees Fahrenheit or less would release RCS inventory and energy at a lower rate then if the reactor had been at full power.

'

Since there is a lower pressure differential between the RCS and the containment there is much less RCS inventory that would be lost through the break. The initial containment pressure peak would therefore be much lower for a large-break LOCA with the RCS at a lower pressure then for a large-break LOCA from full power conditions.

It is also noted that the initial containment pressure peak is not the limiting pressure peak for the full power case.

The long-term pressure peak for a LOCA, which occurs following ice melt, would also be much less reduced due to the reduced heat level.

Since the conditions of the RCS were significantly below those of an operating reactor and the analysis for the ice condenser is designed for higher power

operations, this lower energy case is bounded by the at-power analysis.

Even with the ice bed slightly decreased due to the event, the ice melt following a large-break LOCA would take much longer due to the low level of decay heat.

The AIT has concluded that holding the doors was an acceptable action to take in order to preserve and maintain the ice condenser integrity.

D.

Technical Specifications At approximately 0140 hours0.00162 days <br />0.0389 hours <br />2.314815e-4 weeks <br />5.327e-5 months <br /> on August 31, 1993, the ice condenser was determined inoperable due to Technical Specification 3.6.5.1.c.

An entry was made in the Technical Specification log documenting the ice condenser as inoperable. Technical Specifications require:

i

.

.

.

.

.

'

"A maximum ice bed temperature of less than or equal to 27 degrees Fahrenheit."

-

The ice bed temperature exceeded the 27 degrees Fahrenheit and reached a temperature of 32 degrees Fahrenheit on point

1. 4 of the ice condenser system recorder in the control room.

During this period the amount of ice melted was indeterminate. The licensee did not enter Technical Specification 3.6.5.1.d., which requires:

"A total ice weight of at least 2,099,790 pounds at a 95% confidence" at any time during the event. There was no engineering evaluation to determine how much ice melted prior to returning the ice condenser to an operable status. Once the operating crew noted that the temperature of the ice bed had returned to less than 27 degrees Fahrenheit, the ice bed was considered operable. The ice condenser was logged as being operable in the Technical Specification Actinn Item Log (Tsail) at 0710 hours0.00822 days <br />0.197 hours <br />0.00117 weeks <br />2.70155e-4 months <br />, August 31, 1993. According to the log, it took approximately six hours to return the ice condenser to an operable status.

This decision was based upon non-technical assessment from input of the ice condenser crew supervisor and operations staff. They based this decision on the perceived amoult of ice melt during the event, the duration of the event and the plant conditions during and after the event. There were no technical specification surveillances performed to determine the operability of the ice condenser and that the ice weight i

would meet Technical Specification 3.6.5.1.d.

In order for the licensee to assess this parameter, at least Technical

,

Specification surveillance, 4.6.5.1.b.2, should have been performed to determine the actual ice mass present in the ice condenser and more specifically in those baskets in bays 21 and 22.

Since severe ice melting occurred in bay 22 and slight ice melting occurred in bay 21, this would have been necessary to evaluate if the ice condenser contained the Technical Specification minimum weight at a 95% minimum confidence level. The licensee did not accomplish this surveillance until September 1, 1993. Technical Specification 3.6.5.1 was logged in the Technical Specification Action Item Log (Tsail) on August 31, 1993,

,

for tracking only, awaiting engineering evaluation of the i

ice bed bay weight.

An analysis from Duke Power Company Nuclear Engineering to the McGuire Nuclear Station on September 2, 1993, provided justification that the ice condenser was able to " provide a very sufficient post-accident heat sink." Upon subsequent i

.

.

.

.

'

measurements it was determined that the total amount of ice that melted was approximately 2400 pounds. This was considered to be a "small" loss of ice during the steam leak event and was also recognized as only a " minimal dec ease from the initial inventory." The AIT agreed with this

'

analysis. Subsequent analysis performed by Component Engineering determined that the most predominant melting occurred in four baskets in bay 22. However, indications of melting were observed in bays 12, 19, 20, 21, 22, 23 and 24.

The four baskets in bay 22 were found to be below the 1081 pound limit required by Technical Specification surveillance

.

4.6.5.1.b.2.

These four ice baskets were replenished with

'

'

new borated ice.

It was assumed the ice condenser had sufficient amount of ice following the event which was based upon the severity of event and minimal amount of ice that melted.

Comparing this to the design basis LOCA, the amount of time the event took place, and the likelihood of other accidents during Mode 3 operations, there was enough ice available to mitigate all other accidents that could be postulated.

It was subsequently determined from Nuclear Engincaring that based upon plant conditions being below 442 degrees Fahrenheit a blowdown would release energy at a far lower rate than if the reactor had been at full thermal power.

The AIT determined that Technical Specification 4.5.6.3.1, requires the inlet door monitoring system be determined operable by performing a CHANNEL CHECK "...within 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> after receiving an " Ice Condenser Inlet Door Open" alarm on the control room annunciator portion of the s.g tem."

This CHANNEL CHECK was not performed during the required time period.

E.

Conclusions After reviewing the facts and root causes of the event on

,

August 31, 1993, the AIT reached the following conclusions:

'

The work plan used to attempt repair of 2CF-130 was flawed.

.

It was flawed because: (a) approvals were given without due consideration to the risk involved; (b) there was pour

coordination between the approving officials; and (c)

j approving officials lacked sufficient knowledge of the plan j

to make proper decisions.

-

Plant safety systems functioned as expected and as designed.

The ice condenser system reduced containment pressure and temperature and the bay doors opened and closed as designed.

During the repair of valve 2CF-130, inadequate control measures existed to ensure proper reassembly of the valv.

.

.

.

Maintenance procedures lacked specificity to delineate how the disk assembly should be reassembled.

Existing maintenance controls were ineffectual.

Quality Control was not independent enough to ensure proper assembly.

No management guidance, procedure or other management controls, exists for the control of pipe caps. Operations, Radiation Protection or Maintenance personnel can install / remove pipe caps. This contributed to the unexpected discharge of the pipe cap during the leak repair process.

Holding closed, or systematically closing, the ice condenser bay doors posed minimal safety significance. Had a subsequent event occurred the doors were not prevented from operating.

'

During the inadvertent heatup, the operating team failed to maintain cognizance and control of the shut down reactor.

Poor management direction, poor judgement by the operator and shift turnover led to a lapse in attention by the crew.

Team work, command and control and human performance were

[

satisfactory during the response to the steam leak. There were no human factors concerns for egress from the building or for the operating crews' response to the steam leak.

Poor communications occurred during the evolution to hold the ice condenser bay doors closed. Control room operators

'

were not fully aware that the ice condenser bay doors were only intermittently shut.

V.

ROOT CAUSES A.

Lack of Adequate Measures to Ensure Proper Valve Reassembly A key step in reassembling the Kerotest valve procedure was missing. This, in combination with a lack of awareness by the maintenance and quality control technicians, lead to the

.nisassembly of Kerotest valve 2CF-130.

.

B.

Low Sensitivity to High-Risk Evolutions Personnel involving in developing and approving the repair plan for on-line leak repair of 2CF-130 failed to recognize i

the potentially adverse consequence in using an MTIC procedure on an in-service system. This evolution was treated as routine and was not given proper significance by those individuals.

Several of the reviewers made decisions using incomplete informatio.

.

.

,

C.

Poor Control Room Command and Control Although complicated by shift turnover, Shift Supervisors failed to give explicit instructions to the Reactor Operators on the cooldown process.

In addition, all control room personnel failed to recognize the impending mode change. The reactor operator failed to seek assistance when plant conditions were not responding to his manipulations.

2.

Significant Event Investigation Team Evaluation At 0845 hours0.00978 days <br />0.235 hours <br />0.0014 weeks <br />3.215225e-4 months <br /> on August 31, 1993, just hours after the steam leak event, the site requested a SEIT investigation from the corporate office.

Separately, a McGuire site management team was formed to investigate and determine actions necessary to recover from the event. Throughout the recovery period, these two teams functioned together to direct the site response.

-

On the afternoon of August 31, 1993, the SEIT team arrived on site and began their investigation. The SEIT was led by the Manager of the Safety Assurance department from the corporate Nuclear Services Department (NSD). Also on the SEIT was an NSD operations representative, one NSD maintenance support representative and a Operational Events analyst from NSD. The site provided a member of the Safety Review group to assist the SEIT.

This team remained on site until their exit on September 3, 1993. At their exit they provided a draft report of their findings that included four recommendations for corrective action prior to the plant entering Mode 4. Their report also provided 12 recommendations for long-term I

corrective actions.

Each of the recommendations were incorporated in the site response to this event.

Further, the site took many other corrective actions beyond what the SEIT suggested.

Upon conclusion of the AIT inspection, the Team evaluated the effectiveness of the SEIT. A comparison of the findings between the AIT and SEIT evaluations shows several differences in the two evaluations.

A description of the major differences is listed below.

-

The SEIT report did not identify operator turnover in its evaluation or recommendations as a problem. Additionally, during the SEIT exit they concluded that operator turnover was not a problem in this event. The AIT concluded that the turnover contributed to the confusion in the control room and distracted

'

the operators.

i

-

Discussion with the SEIT revealed that they did not consider that operating on Excess Letdown complicated the operating conditions for the operators. The AIT concluded that the operators were fixated on the lower boundary of RCS temperature and this did contribute to the inadvertent heatup.

Interviews with the operating crew supported this conclusio.

l

.

'

-

The SEIT report states the operator response and management decision making were appropriate during the event (excluding mode change). The AIT found poor decision making by operations personnel in approving the repair plan.

Further, the AIT made several other observations that the SEIT did not mention.

Some of these include:

-

The ice condenser was inappropriately declared operable without the required surveillance evaluation.

-

The four-hour surveillance, in accordance with Technical Specifications 4.5.6.3.1, for ice condenser doors was not performed.

-

The AIT found that there is no program for field supervisory audits of visiting maintenance workers in the field.

-

The AIT found that the Quality Control inspector lost his independence for verifying the reassembly of Vale 2CF-130 when he became part of the re-work effort.

After observation and evaluation of the SEIT team the AIT concluded that the licensee's SEIT evaluation was most effective early in the event.

It provided an aid to the site in focusing on the proper areas of investigation just after the event and before a complete site investigation team could begin work. Discussions with senior site managers revealed that their expectations of the SEIT were fully met.

Although the SEIT scope and evaluation had limitations, the licensee developed other responses that compensated for these limitations.

3.

Confirmation of Action Letter A Confirmation of Action Letter (CAL) documenting the licensee's commitments made as a result of the circumstances surrounding the steam leak event was issued by NRC Region II on September 1, 1993. The letter

'

confirmed the licensee's commitment to take the following actions:

a.

Obtain concurrence of the Regional Administrator prior

[

to entering Mode 2.

b.

Conduct a comprehensive investigation to determine all aspects of the August 31, steam leak event.

c.

Fully evaluate the recommendations of the Significant Event Investigation Team and implement appropriate corrective action in a timely manner.

In addition to the general items in the CAL, Region II had several specific corrective actions requested of the licensee. After

!

'

incorrectly diagnosing one damaged steam generator tube on Unit 1, the Region II and NRP. requested a complete reevaluation of the steam

i

.

.

.

.

generator tube data for Unit 1.

The licensee was asked to assess the impact of the incorrect diagnosis on Unit 2 prior to restart. The NRC also expected the licensee to complete all steam leak repairs before entering Mode 2.

Finally, the NRC desired resolution of Information Notice 93-02, " Debris Plugging of Emergency Core Cooling System Suctions," before concurring on a startup to Mode 2.

On September 8,1993, the licensee, Region II and NRR personnel held a conference call to discuss the items in the CAL. During that telephone call the licensee stated that all short-term corrective actions were completed and that all steam leaks on Class I systems were repaired. At that point the licensee had completed 84% of the steam generator data review. The licensee then asked for concurrence for entry into Mode 2 and that concurrence was given by the Regional Administrator.

4.

Licensee Long Term Corrective Actions After evaluating the causes of the steam leak event, the licensee issued a report to the NRC nn September 5,1993, outlining plans to initiate the following long-term corrective actions:

a.

Complete a full Mechanical Maintenance Human Performance Evaluation Study of this event.

b.

Develop a means to emphasize to valve technicians the consequences of improper spring guide installation.

c.

Perform a Human Factors review of the Kerotest corrective maintenance procedure.

d.

Given specific event training to all Mechanical Maintenance technicians who may be required to work with Kerotest activities.

e.

Evaluate the repair methods used which involve removal of pipe caps on pressurized systems, f.

Revise the on-line leak seal injection procedure to include pre-job briefings.

g.

Review all maintenance procedures for infrequently j

performed evolutions to include pre-job briefings.

'

h.

Review on-line leak seal procedures for Human Factors concerns and technical accuracy.

i.

Evaluate the effectiveness of current Post Maintenance Testing methods on unisolable components.

j.

Develop site specific administrative guidelines that

address control of non-assigned individuals and organizations performing work on the site.

l

.

.

.

.

.

F

-l k.

Change the work. planning procedure to include guidance to planners to ensure critical as-found information is

,

,

included in work packages.

1.

The specifics of the 2CF-130 work package will be reviewed with the work planners.

m.

Develop a method to identify maintenance activities on inservice systems that have the potential to result in

'

unit trips or transients, unisolable leaks, or rendering safety systems or components inoperable.

l Ensure these jobs include a pre-job briefing.

n.

Develop a reading package describing the event and addressing communications for all licensed personnel.

,

o.

Develop a method of bringing impending mode changes to the control room team's attention.

p.

Develop a case study of this event for all licensed personnel.

q.

Conduct coaching, counseling and remediation on communications for the individual operator involved.

r.

Train all licensed operators on the effects of cooling down while on Excess Letdown.

s.

Clarify the Annunciator Response Procedure for the ice condenser bay door alarm.

t.

Develop a process to better the integrity of pipe cap threaded connections.

I u.

Investigate cost effective methods of improving the performance of vent and drain connections.

Several of the above actions have been completed as interim measures in response to the event evaluated in this report. The long-term corrective actions are scheduled to be completed by March 1994. The AIT found the long-term corrective actions adequate measures in response to this event.

5.

Exit Interview With Licensee Management Inspection scope and evaluations were summarized during an exit interview on September 7, 1993, with those persons listed in Appendix A.

j The NRC described the areas inspected and discussed in detail the

'

inspection results covering the steam leak event on August 31, 1993.

No proprietary material is contained in this report.

No dissenting comments were received from the licensee. Afterwards, a media briefing was held at the licensee's facility.

j

.

. -

.

,

'

\\

Appendix A Persons Attending Exit Interview i

J. Allgood, MNS/ Safety Review Group T. Ariow, MNS/ Safety Review Group D. Baxter, MNS/NGD/ Operations A. Beaver, MNS/NGD/ Operations J. Boyle, MNS/ Work Control Superintendent D. Bumgardner, MNS/NGD/ Operations W. Byrum, MNS/NGD/ Radiation Protection B. Caldwell, MNS/ Trainer K. Crane, MNS/ Regulatory Compliance R. Cross, MNS/ Regulatory Compliance T. Curtis, MNS/NGD/ System Engineering P. Davis, MNS/ Business Management

'

E. Estep, MNS/NGD/ Safety Assurance E. Geddie, MNS/NGD/ Station Manager G. Gilbert, MNS/NGD/ Safety Assurance J. Grogan, General Service Department P. Guill, MNS/ Regulatory Compiiance R. Hall, MNS/ Engineering Manager B. Hamilton, MNS/NGD/ Operations

,

F. Hayes, MNS/ Human Resource R. Hofmann, MNS/ Safety Review B. Isenhour, MNS/ Safety Review P. Keros, NGD/ Human Resource L. Kunka, MNS/ Regulatory Compliance G. Massey, MNS/IAE General Supervisor B. Matthews, MNS/ Electrical Engineer T. McHeeken, MNS/ Site Vice President M. Mullen, MNS/Com.aunity Relations M. Nazar, MNS/ Station Manager Staff M. Pacetti, NGD/ Mechanical / Nuclear Engineer B. Parrott, MNS/ Safety Review T. Pedersen, MNS/ Safety Review Group N. Pope, MNS/ Superintendent IAE R. Rhodes, MNS/ Community Relations R. Roberts, General Service Department M. Sample, Manager Safety Assurance G. Savage, General Office / Corporate Communications R. Sharpe, MNS/ Regulatory Compliance Manager R. Sipe, MNS/ Community Relations B. Taylor, MNS/ General Service Department / Customer Support

.

R. Travis, MNS/NGD/ Component Engineer M. Tuckman, NGD/ Senior Vice-President, Nuclear Services R. Vigor, MNS/ General Service Department / Maintenance J. Washam, MNS/ Safety Review

"

R. White, Jr., MNS/NGD/ Mechanical Maintenance

.

)

i

.

.

.

.

,

.

Persons Interviewed B. Hamilton Superintendent of Operations G. Blake Unit 2 Coordinator A. Lindsey Operations Coordinator J. Lukowski Shift Manager ("A" Shift)

E. Faggart Shift Manager ("C" Shift)

J. Rumfelt Shift Supervisor ("A" Shift)

J. Pressley Shift Supervisor ("C" Shift)

D. McCorkle Assistant Shift Supervisor ("A" Shift)

G. Mills Assistant Shift Supervisor ("A" Shift)

G. Vellers Assistant Shift Supervisor ("C" Shift)

J. Howard Nuclear Control Operator ("A" Shift OAC)

K. Ribelin Nuclear Control Operator ("C" Shift 0AC)

D. Helton Nuclear Control Operator ("A" Shift B0P)

>

B. Moore Nuclear Control Operator ("C" Shift BOP)

E. Wilkinson Nuclear Control Operator ("A" Shift RO)

e D. McGinnis Director of Operator Training M. Wunderlich Utility Support Specialist Inc.

D. Kolb Utility Support Specialist Inc.

R. Rider Utility Support Specialist Inc.

i i

.1

.

<

.

.

,

'

.

.

Appendix B i

Procedures Reviewed

!

AP/2/A/5500/01 Steam Leak AP/2/A/5500/12 Loss of Letdown, Charging or Seal Injection AP/2/A/5500/24 Loss of Containment Integrity AP/2/A/5500/35 ECCS Actuation During Plant Shutdown MP/0/A/7650/077 On-Line Leak Sealing Initial Injection

'

OMP 1-2 Use of Procedures OMP 1-6 Independent Verification

,

OMP 2-4 Reactor Operator and Unit Supervisors Logbook

'

OMP 2-5 Technical Specifications Action Items Logbook i

OMP l-12 Operations Communications Standards

,

OMP l-13 Conduct of Operations OMP 2-2 Shift Turnover Procedure OP/2/A/6100/01 Controlling Procedure for Unit Startup OP/2/A/6100/02 Controlling Procedure for Unit Shutdown OP/2/A/6100/22 Unit 2 Data Book OP/2/A/6200/01 Chemical and Volume Control System

,

PT/0/A/4200/32 Periodic Inspection of Ice Condenser Inlet Doors i

PT/2/A/4700/10 Shift Turnover Verification RP/0/A/5700/00 MNS Emergency Action Levels SD 2.8.2 Operability Determination

,

SD 3.1.4 Conduct of Operations

,

SD 4.4.2 Control of Temporary Modifications

,

Nuclear Station Modification Manual McGuire Nuclear Station Directives McGuire Nuclear Station Operating Manual MNS Emergency Plan Implementing Procedures

'

,

HNS Technical Specifications MNS Final Safety Analysis Report

-

Employee Training History Files Other References Reviewed

,

,

1.

McGuire Nuclear Station, " Problem Investigation Process,"

Serial No.1-M93-0756.

2.

Nuclear Maintenance Applications Center, "On-Line Leak Sealing - A Guide for Nuclear Power Plant Maintenance Personnel," July 1989.

j 3.

Shao, Lawerence C., "Use of Sealing Fluids on Primary Pressure Boundary Components," Letter from NRC Director of Engineering &

Systems Technology to James L. Milhoan, NRC Director, Division of t

Reactor Safety, Region IV, January 25,1988.

4.

USNRC, Generic Letter 90-05, " Guidance for Performing Temporary Non-Code Repair of ASME Code Class 1, 2 and 3, Piping."

5.

Duke Power Company, "McGuire Nuclear Station, Unit 2, Docket No.

50-370,"

Report from Duke Power to the USNRC, September 5, 1993.

>

b

-

.

.

~

.

.

.

.

~

i l

i

<

6.

Ice Condenser Engineering Design Guide No. 1.

Ice Condenser Door Position Monitoring System.

.

Training Documents Reviewed 1.

" Containment Systems," McGuire operator lesson plan, Rev. 8, January 29, 1991.

2.

" Upper and Lower Containment Ventilation," McGuire operator lesson plan, f

Rev. 6, April 6, 1989.

3.

Requalification Training Lesson Plan, Normal Plant Shutdown, OP-MCC-SRT-N01.

!

4.

" Administrative Procedures and Controls: Operations Procedures," McGuire operator lesson plan, Rev. 4, August 3,1990.

,

5.

" Administrative Procedures and Controls: Station Directives," McGuire operator lesson plan, Rev.10, March 5,1991.

,

6.

" Administrative Procedures and Control: Operations Management Procedures," McGuire operator lesson plan, Rev. 7, November 14, 1990.

7.

McGuire Simulator Certification Submittal, January 1991.

,

8.

MC-0P-TG-13, " Simulator Configuration Management Guideline."

9.

Standard 2401.1, Rev.4, " Mechanical Maintenance Training and Qualifications Overview." Standard 2404.0, Rev. 2, "McGuire Review of

,

Support Personnel Training and Qualifications."

,

10.

Task qualification Report - Repair of Kerotest "Y" Type Globe Valve (task #MM-0T-0714).

MM-MC-SFT-193 Mechanical Maintenance Training HM-MC-ADM-001 MM Work Request and Administrative Controls Training MM-MC-SFT-292 Attention to Detail Training HM-TC-ASS-VMA-E6-IG Valve Maintenance I Training

MM-TC-ASS-VMB-E-1 Valve Maintenance II Training HM-MC-VLV-REF Valve Refresher Training

!

,

t i

.

-

- -

.

.

.

Appendix C AIT Charter The Charter for the AIT was prepared on September 1, 1993. The special inspection commenced with an Entrance Meeting on September 1, 1993. The Charter for the AIT specified that the following tasks be completed:

1.

Develop and validate the sequence of events associated with the

'

August 31, 1993, Unit 2 steam leak and containment pressurization

'

event. This sequence should begin with plant conditions immediately prior to the event and extend until the plant had fully recovered from the transient.

2.

Evaluate the management controls which led to the attempt to perform a temporary leak repair on the CF-130 drain pipe cap and evaluate the appropriateness of the repair plan.

3.

Evaluate the safety significance of the event with regard to system performance including the ice condenser and plant proximity

,

to safety limits as defined in the Technical Specifications.

'

4.

Evaluate operator response during the event including the circumstances which led to the inadvertent heatup to Mode 3.

5.

Determine the root cause of each equipment malfunction and personnel error.

Include failure of 2CF-130.

6.

Evaluate the management controls which led to personnel

'

intentionally holding closed ice condenser bay doors when an apparent open demand was present.

7.

Evaluate the effectiveness of the licensee's Significant Events Investigation Team.

8.

Prepare a special inspection report documenting the results of the above activities within 30 days of completion.

?

l

.

i

.

-

-

.

l

.,

.

i

i

Appendix D

Acronyms

AIT Augmented Inspection Team

[

MNS McGuire Nuclear Station

!

NRC Nuclear Regulatory Commission PSIG Pounds per Square Inch Gauge SEIT Significant Event Investigation Team

!

!

!

,

!

,

l

!

,

n I

i i

)

,

i

-

--

-

-

,

-

-

.,.,

- " *

.

-

Figure 1

'

,

\\

,

.

Kerotest Valve Crossection

'

(See Figure 2)

.

-

/

)

xN

>

+

\\

r

%'

N

'@f.

-

_

l N,

s g

h.

eN

' y

,

y j

.o/N x

%

-

v,7

,

\\,

'N

,

1. 0 s

/,

'-

,,

N N i s

D g*<

Sr

'

-

i L'

l s

(

\\(

'

'

'

'

's

,

'x x

-\\

x x

,

f

\\.,

he

/

[y;':'

,

'

-

,

' k*

'

'

i: x

'

-,

.

, "'

y

'

,

j

-

~

e

. f i m

!

.

,

.

.

,

,#

/'

.

T/,/

• '

s

e

x

. _ _

___.__.

\\

!

-

\\i

-

r

'd'

I I

N

.'

t*'4

-

-

,

p_c.gvec g

i

-

N

I cri

"c"

,

's PJ

'

s w

m

_

,

O

_..

o g<

--

VI.

~

-

-f_

t

(

i

%r M-m-q~

%

rJ

$

't!g

_

g

____

_.

sg.

K1

\\n

-

,

mov-i)

.

-

d,(M 1.

,-

,,

,

,

.

,

,

--

!

!

'

.

w.

5k tem

.

1<u..a.s

.m w. m, n

,m

.

,

... u

,

,

OP 6 4 tM 888==--

ISL M

' ' ' ' - star r/4

A

-. t T.

'

ERN: MCOO147Q

• ~"~

-

'

e.:

misine om mir ce arr2..

,s im

-

L

.

- - - -

- - - - -

-

-

-

- -

~

- _

_--

q

,...

l e

Figure 2 j

Kerotest Parts List

>

I l

-

Parts List

_.

Item Name of Part

,

No.

Body

Disc

Disc Cap

Spring Guide

Spring

'

Pin

Diaphragm

.

Bonnet

Stem Head

Bearing Assembly

Stem

Packing Ring

Gland Follower l

>

Gland Plate

.

Yoke r

Gland Screw Cap

,

Handle

,

Handle Lock Nut

Nameplate

,

I

!

!

!

't i

_

.

,

.,.

...

.

%

l l

,'

Figure 3 t/ Guire ice Condenser Systen C

s

'

U IT (

- - - - '

Ib/

.

h

-

-CRANE

'

'

h(

h INTERMED! ATE

!

n,.

!

'

DECK p

,

,

'#

i

,

AIR DISTRIBUTION I

'

gO DUCTS

'

!i l

i

.*

ICE BASKET lP l

I. *

CRANE i

n j

'

'

y i

dl'

, '.

WALL

,..

LATTICE FRAME EN RATOR l,!

!!

l j

f

!

i,

-

ENCLOSURE

'

,

.

,"

SHELL-

l

.

%

-

,

p i

i t

i o

I l

'

t,t y Lui@

J.4

-

WALL PANELS

},'

l

-

,

(

-

q r

.

If f

STE AM GENERATOR p

'

TURNING

'

N

,q

','q

,

,

VA NES N's

,

l i

!

'

-

e

.

D l

5\\

M' bad %

,

.

,

.

t g

,

'f I

I I

.

q 4 /'

<d r

sp l

a Jj

>

M DIVICER DECK s

,

,ih LOWER

.,.

'*

.

INLET DOORS

,,

~i*

'

j

-

LOWER

'

SUPPORT FLOOR

' ' ' '

ICE CONDENSER (NF)

t. ICE REGION EQUIPMENT I" MC-CNT-NF-6

! " ' 10-l6 - 86

]

'

"' FSAR VOL.6

!

'

CMR l'" DWA

!

au= e '

T R AINING USE ONLY

!

.

.