Text

Rev 0 to 890225 Steam Generator Leak Event Rept
	ML20245K731
Person / Time
Site:	North Anna, Big Rock Point File:Consumers Energy icon.png
Issue date:	04/26/1989
From:	; VIRGINIA POWER (VIRGINIA ELECTRIC & POWER CO.)
To:	;
Shared Package
ML19297H630	List: ML20245K724; ML20245K728; ML20245K731;
References
	NUDOCS 8905050155
	Download: ML20245K731 (127)
	v • d • e

Sk ih Q%N bk%id(n'\/y%yg%jh h'th \T 6 9s%%m$

sg 3 gA g'%e%x Q $p a s

gh N4 g qq Ng b N'a Vg w 3 y\o vNh)g Tj( yh k Q 1

(

N3 s w% w% %, m.aN%g k) %

North Anna Unit 1 February 25,1989 Steam Generator Leak Event Report April 26,1989 Revision 0 i

i l

l'irginia Electric and Potter Company

(

l

.L

I North Anna Unit 1 I February 25,1989 g Steam Generator Leak Event Report i

I l April 26,1989 Revision 0 I

I I

I Virginia Electric and Potter Company I

I 1 L

s- _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ _ - - _ - - _ _ _ _ _ _ _ _ - - - _ - _ - _ - - - - - - - - - - -

i Afg%g$

g b #n $

4y u%f "Q ff 1; %':)

k[\ g(

~

q{ \ ,

g + g \

North Anna Unit 1 Februaty 25,1989 Steam Generator Leak Event Report April 26,1989 Revision 0 l

Virginia Electric and Power Company I . _ . _ _ _ _ _ . _ _ _ _ _ _ . d

'i '

l h--

f 1

VIRGINIA ELECTRIC AND POWER COMPANY i

l I NORTH AIDIA UNIT 1 FEBRUARY 25, 1989 STEAM GENERATOR LEAK EVENT REPORT 1

APRIL 26, 1989 REVISION 0

)

i NORTH ANNA UNIT 1 FEBRUARY 25, 1989 STEAM GENERATOR LEAK EVENT REPORT TABLE OF CONTENTS

. CHAPTER PAGE I. EXECUTIVE

SUMMARY

A. Purpose 1 B. Brief Description of North Anna Power Station 1 C. Overview of the Event 2 D. Virginia Electric and Power Company Response 3 II. DESCRIPTION OF THE EVENT A. Introduction 5 B. Conditions Prior to the Event 5 C. Event Description and Analysis 7 D. Operatior Al Analysis 14 E. Procedure Utilization and Analysis 18 F. Safety Equipment Analysis 19 III. RADIOLOGICAL EFFECTS OF THE EVENT

'. A. Summary- 32 I B. Discussion of Release Pathways 33 C. Environment 31 Samples Obtained and Results 36 I

IV. EMERGENCY RESPONSE A. Emergency Plan Implementation 46 i B. Summary 50 I'

\

1 m-____________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ . _ _ _

! NORTH ANNA UNIT 1 FEBRUARY 25, 1989

! STEAM GENERATOR LEAK EVENT REPORT l TABLE OF CONTENTS CHAPTER PAGE l

V. SAFETY EVALUATION A. Safety Assessment Comparison to UFSAR Analysis 51 B. Core Integrity Evaluation 53 VI. STEAM GENERATOR EVALUATION A. North Anna Steam Generator Operating Experience 57 B. Failed Tube / Plug Identification and Examination 66 C. Plug Evaluation 75 D. Tube Failure 82 E. Steam Generator Inspection 86 F. Scope of Repair to Susceptible Mechanical Plugs 92 l

VII. BASIS FOR RETURN TO SERVICE 97 l

VIII. LESSONS LEARNED 103 5

IX. CONCLUSIONS 104 i

l l

11 ;

i

l

)

LIST OF FIGURES 7

l I FIGURE NUMBER DESCRIPTION I-1 Westinghouse Series 51 Steam Generator 11-1 Unit 1 Pressurizer Level ,

II-2 Unit 1 Pressurizer Pressure i II-3 Unit 1 Pressurizer 'C' S/G Pressure (first 600 seconds) 11-4 Unit 1 Pressurizer 'C' S/G Pressure (second 600 seconds)

II-5 Unit 1 'C' S/G Narrow Range Level II-6 Unit 1 'C' S/G Flows j II-7 Unit 1 Charging and Letdown Flow !

II-8 Picture of 'C' MFRV Controller w/ Fitting Break j II-9 Containment Sump Pumping frequency i II-10 Unit 1 Main Steamline 'C' Radiation Monitor I III-1 North Anna Power Station VG-179 Strip Chart Recording VI-1 Mechanical Plug Sketch l

1 1

i iii

___ J

f LIST OF TABLES TABLE NUMBER TITLE II-1 Unit 1 Reactor Trip and Tube Leak j 3equence of Events Summary 1

11-2 Operational Assessment Timeline III-l Calculated Activity Released During the Event III-2 Calculated Activity Released Through Vent Vent 'A' III-3 Calculated Activity Released from SDAFWP Exhaust III-4 Steam' Generator Activity Prior to and After Event III-5 Condenser Air Ejector Sample Activity III-6 Reactor Coolant Sample Data V-1 North Anna Unit 1 Steam Generator Tube Leak Comparison to Licensing Analysis i VI-1 North Anna Unit 1 Tube Plugging Summary VI-2 North Anna Unit 2 Tube Plugging Summary i (Mechanical Plugs) i, VI-3 Unit 1 Steam Generator 'C' Chemistry Sample Af ter Second Drain Down VI-4 Piug Head Area Calculation Remaining Wall at Failure l

i VI-5 Relative Time-to-Initiate PWSCC VI-6 Comparison of Required Versus Available Interface Loads l l

VI-7 North Anna Unit 2 Mechanical Plug Repair Plan VI-8 North Anna Unit 2 Mechanical Plug Repair Plan (Detail)

VII-1 Relative Time-to-Initiate PWSCC l l

j l

l iv

I. . EXECUTIVE

SUMMARY

/ A. Purpose

, This report'provides a detailed description of the February 25, 1989 steam l '

generat n leak event at North Anna Unit 1, and the ensuing evaluations and

. actions performed and planned by the Virginia Electric and Power Company (Virginia Power).

1 i The detailed sequence of events including the response-of the operators and j l

key plant equipment is discussed in Chapter II. The radiological effects of ,

l l.

the event are evaluated in Chapter III. An evaluation of the implementation'of the Emergency Plan is discussed in Chapter IV. The safety consequences and significance of the event are discussed and evaluated in Chapter V. The cause of the steam generator leak and the actions taken to address leak are discussed j

,- in Chapter VI. Finally, the lessons learned by Virginia Power as a result of !

l L

the event are discusr,ed in Chapter VIII.

The consequences of this event as it relates to Unit 2 are also discussed.

Corrective actions for Unit 2 are provided.

j- B. Brief Description of North Anna Power Station The North Anna Power Station is a two (2) unit station owned jointly by Virginia Electric and Power Company and Old Dominion Electric Cooperative and operated by Virginia Electric and Power Company Nirginia Power). It is located on the southern shore of Lake Anna, in Louina County, approximately 40 miles notth of Richmond, Virginia. Each unit includes a three-loop pressurized water reactor nuclear steam supply syrtem and turbine generator furnished by Westinghouse Electric Corporation. The balance of plant was designed and constructed by Virginia Power, with the assistance of Stone and Webster I

l Engineering Corporation. Each nuclear unit is licensed to operate at a rated thermal power of 2893 MW thermal.

7 Each unit has three (3) Westinghouse Series 51 steam generators illustrated l

on Fi;;ure I-1. These Series 51. steam generators have 3388 tubes, which have an L

outside diameter (OD) of 0.875 inches and'a wall thickness of 0.050 inches.

The tubing material is Alloy 600 in the mill annealed condition. The tubes are hardrolled above the bottom of the tubesheet, and were explosively expanded prior to oTeration to eliminate the open annular crevice that remained within the tubesheet. Each steam generator has seven (7) support plates which are carbon steel with drilled tube holes having a small clearance between the tube and support place.

Unit 1 achieved commercial operation in June 1978 and has completed operation of its seventh fuel cycle. Unit 2 achieved commercial operetion in December 1980 and has completed operation of its sixth fuel cycle.

C. Overview of the Event At 1406 hours0.0163 days 0.391 hours 0.00232 weeks 5.34983e-4 months on February 25, 1989, 'C' main feedwater regulating valve closed when a fitting between the. air regulator and the valve controller i fractured. The resulting loss of control air allowed the valve operator springs to close the valve. The closure of the 'C' main feedwater regulating i valve created a mismatch between the steam flow (leaving) and the (now reduced) t feedwater flow (incoming). An automatic reactor trip occurred at 1407 hours0.0163 days 0.391 hours 0.00233 weeks 5.353635e-4 months due to a steam flow /feedwater flow mismatch coincident with low steam generator level. At 1411 hours0.0163 days 0.392 hours 0.00233 weeks 5.368855e-4 months , during recovery, the operating shift observed that the l

pressurizer pressure did not return to its nominal pressure of 2235 psig. At 1426 hours0.0165 days 0.396 hours 0.00236 weeks 5.42593e-4 months , higher than normal radiation levels were detected in the air ejector monitor. Other radiation monitors were checked and indicated that 'C' l

l l

d main steamline had higher than normal radiation. The control room personnel then performed a preliminary reactor coolant system inventory. Other 1 observations, including the procedural guidance contained in an abnormal i l

procedure caused the Shif t Supervisor to enter the Emergency Plan. l 1An " Alert" was declared at 1525 hours0.0177 days 0.424 hours 0.00252 weeks 5.802625e-4 months and the process of bringing the unit to cold shutdown (Mode 5) began. At 1833 hours0.0212 days 0.509 hours 0.00303 weeks 6.974565e-4 months Unit 1 entered hot shutdown (Mode 4). The cooldown and depressurization continued until cold shutdown (Mode 5) was reached, at 2212 hours0.0256 days 0.614 hours 0.00366 weeks 8.41666e-4 months . The emergency condition was terminated at 2220 hours0.0257 days 0.617 hours 0.00367 weeks 8.4471e-4 months .

During the course of the event, effluents were released from Steam Generator 'C', prior to its isolation, through the condenser air ejector and auxiliary feedwater pump turbine exhaust. Doses calculated, based on estimated maximum release rates and total activity released, were well within Technical Specification limits for doses at and beyond the site boundary.

D. Virginia Electric and Power Company Response Following the te rmination of the emergency, Virginia Power's management immediately initiated recovery activities. An organization was established and l

l resources identified for evaluating the incident and recommending recovery ,

actions. Virginia Power's recovery organization var in place and functional I

immediately following termination of the emergency.

? The Virginia Power Recovery Organization was divided into four groups reporting to the Recovery Manager. Overall coordination of the event evaluation and recovery actions was the responsibility of the Recovery Manager and the Local Emergency Operations Facility (LEOF) Command Center. The Technical Evaluation group was responsible for the steam generator inspection program, failure analysis, steam generator integrity evaluation and tube

plugging criteria and recommendations. The Nuclear Safety and Licensing group was responsible for the event evaluation including evaluation of the response of the plant and emergency response organizations, as well as interfacing with the NRC. The Industry Interface group was responsible for interfacing with the media as well as INPO, State and Local authorities, and Old Dominion Electric Cooperative. The Health Physics group was responsible for evaluating the radiological effects of the event.

l I

FIGURE I-1 DEMISTERS SECONDARY TOR Of M MOISTURE SEPARATOR s a

) . =

PA '

Y,bb ' SECONDARY MANWAY OR!flCE RINGS j j% '

's

.; u j z. .

h %

l SWlRL VANE PRIMARY M0!STURE SEPARATOR - -

g%;7 ! UPPER SHELL

)

IN ,

FIEDWATER RlHG I FEDWATER INLET C

./ :,

g-

[ i

]

i 1N0ZILES ANTM8 Rail 0N BASS , . .

TU8E BUNDLE h:{ir,,

' . .f5> .

< :w% ,

'4 7.' :'$p QEl. ' s 3 6 LJt f [ LOWER SHEll WRAPPER .

\

V :M! &t TU8E SUPPORT PLATES 3 N[ tis'y

?

BLOWDOWN LINE E TU8E SHEET

\ N

.3 SECONDARY HAN0 HOLE s t/

A% =

PRIMARY MANWAY \ '

/ -TU8E LANE BLOCK PRIMARY COOLANT INLET #

l PRIMARY COOLANT OUTLET not to sau

@ SERIES 51 STEAM GENERATOR

II. DESCRIPTION OF THE EVENT A. Introduction l

The following narrative describes the operating events for the North' Anna Unit I steam generator tube leak which occurred on February 25, 1989. Table !

II-l provides a more co ::ise and detailed chronology of events. The narrative and sequence of events were compiled from a variety of sources including 1) the Sequence of Events Recorder (SER), 2) the Control Room P-250 process computer and alarm printouts, 3) the Emergency Response Facility Computer System (ERFCS) historical file, 4) interviews with licensed Control Room Operators (CRO) and Senior Reactor Operators (SRO), 5) CR0 and SRO logs, 6) Emergency Response Facilities logs, 7). strip charts from Control Room recorders and 8) General Electric Transient Analysis Recording System (GETARS). The response of important primary and secondary system parameters, including pressurizer level, Reactor Coolant System (RCS) pressure, RCS temperature, leaking steam generator l level and' flows and charging / letdown flow are provided on Figures 11-1 through j i

II-7 to assist the reader in following the event, and associated recovery actions.

/ B. Conditiins Prior to the Event Prior to the event, North Anna Unit I was operating at 76% power on a power coastdown. North Anna Unit 2 was shutdown for a refueling outage. Reactor and secondary system conditions were normal. Reactor Coolant System (RCS) leakrate measurements were taken on February 22, 1989 and indicated less than 0.5 j c gallons per minute unidentified leakage. At 1404 hours0.0163 days 0.39 hours 0.00232 weeks 5.34222e-4 months on February 25, 1989 l l I

l

the Shift Technical Advisor (STA) completed periodic test .(1-PT-46.3A),

i primary-to-secondary leak rate determination, and recorded radiation monitor I (N-16) leakrates of less than one (1) gallon per day (GFD) - Steam Generator- I i

'A', 2.39 GPD - Steam GeneratorB', less than one (1) GPD - Steam' Generator '

'C' and less than -one (1) GPD on 'the Main Steam header. The primary-to-secondary leakage, based on the air ejector radiation monitor, was 0.59 GPD. Two (2) main feedwater pumps, three (3) high pressure heater drain pumps, and two (2) condensate pumps were in service on the secondary system.

Operation of charging and letdown was normal with one (1) charging pump in operation. Service water and component cooling systems were normal with one (1) component cooling pump and two (2) service water pumps in service.

A grab sample from air ejector discharge was being taken every 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months , in accordance with Technical Specifications. A sample taken at 0050 hours5.787037e-4 days 0.0139 hours 8.267196e-5 weeks 1.9025e-5 months on February 25, 1989 showed no significant increase in air ejector activity from the previous sample, at 0035 hours4.050926e-4 days 0.00972 hours 5.787037e-5 weeks 1.33175e-5 months , on the previous day. Results of isotopic analysis of grab samples taken prior to the event were 5.36 GPD (Argon-41 Krypton-87, and Xenon-135 isotopes were analyzed). Unit 1 secondary. system l chemistry periodic test (1-PT-46.3C) was performed to determine primary to secondary leakage at 1610 hours0.0186 days 0.447 hours 0.00266 weeks 6.12605e-4 months on February 23, 1989. The following data was i

indicated for the steam generators at that time: j l 'A' steam generator - N.D. (none detected) l

> l

'B' steam generator - 1.8 GPD

'C' steam generator - equal to or less than 1.0 GPD Total steam generator leakage - 2.6 GFD j At the time of the event, the operating shift was manned by three (3)

Jenior Reactor Operators, including the Shift Supervisor, four (4) licensed J

Reactor Operators and seven (7) unlicensed Reactor Operators. A Shift

-Technical Advisor (STA) with an SRO license was also assigned to the operating shift. A Health Physics shift and Chemistry technicians were also onsite.

C. Event' Description And Analysis At.1406 hours0.0163 days 0.391 hours 0.00232 weeks 5.34983e-4 months , on February 25, 1989, 'C' main feedwater regulating valve (1-FW-FCV-1498)- closed when a fitting between the air regulator and the valve controller fractured (A picture of 'C' controller with the fitting . break is shown in Figure 11-8). The loss of instrument air to the valve controller allowed the valve operator springs to close the valve. With the reactor at 76%

power and the main generator supplying approximately 725 megawatts to the electrical grid, the closing of the main feedwater regulating (MFR) valve created a mismatch between the amount of steam coming from and feedwater going to the Steam Generator 'C'. The Control. Room Operator (CRO) at the controls (OATC) of Unit 1 immediately attempted to restore feedwater to the required amount, by opening the MFR valve and the bypass valve (1-FW-FCV-1499). 'The closed'MFR valve could not be reopened, and the bypass valve could not supply a sufficient amount of water to reduce the mismatch between steam and feedwater f

When the Steam Generator 'C' water level decreased to 25%, a reactor

~

l flows.

trip occurred at 1407 hours0.0163 days 0.391 hours 0.00233 weeks 5.353635e-4 months on steam flow / feed flow mismatch coincident with low steam generator level. The post event analysis determined that the GETARS f reporting system indicated a RM-MS-172 Main Steamline 'C' radiation spike at 1407:22, (Reference Figure II-10) that lasted for five (5) seconds. The s

l indication then abruptly returned to pretrip levels.

Two (2) licensed Reactor Operators, one (1) from the backboards watch and an extra CR0 from Unit 2, supplemented the Unit 1 operator to aid in stabilizing Unit 1 following the trip. The backboards CR0 assumed the duties

_7_

)

'i l

of controlling and_ monitoring the " primary side". The Unit 1 CR0 controlled feedwater (both main and auxiliary) and the extra CR0 from Unit 2 assumed the j l

'" balance of plant" duties. The three (3) CRos and the Senior Reactor Operator I (SRO) from Unit 2-immediately carried out the immediate actions of Emergency J Procedure EP-0, " Reactor Trip or Safety Injection". The Unit 1 SR0 arrived in the Control Room at the approximate time that EP-0, step 4 was completed. The step states, " Check if SI is actuated". (SI means Safety Injection.) _It was not, so the procedure reader, the Unit 2 SRO, transitioned to the " Response not Obtained" column, which states, " Verify that none of the following conditions requires SI have occurred:". After reviewing the stated parameters and l determining that no oSI was required, the Unit 1 SRO directed the procedure reader to transition to 1-ES-0.1, " Reactor Trip Response, Step 1".

I At 1411 hours0.0163 days 0.392 hours 0.00233 weeks 5.368855e-4 months , the Auxiliary Feed pumps were placed on recirculation and main feedwater supply to the steam generators was reestablished. The pressurizer level, shown in Figure II-1, decreased to approximately 15% and a letdown isolation occurred. Pressurizer pressure, shown in Figure II-2, stabilized at approximately 1950 psig, but did not return to its nominal pressure of 2235 psig. The CR0 increased charging flow rate to maximum, i

Pressurizer level stabilized and then started to increase toward its nominal 20% condition. At this time, letdown was returned to service. At l approximately 1417 hours0.0164 days 0.394 hours 0.00234 weeks 5.391685e-4 months , the CR0 monitoring the primary called the Unit 1 i

SRO's attention to the slightly below normal pressurizer level and pressure.

I Since the Auxiliary Feedwater system had been supplying all three (3) steam generators and because of low-low levels existing at this time, the operators

} l l assumed that the slow response for level and pressure was due to the cooldown from cold Auxiliary Feedwater being injected into the steam generators.

l i

l l

l At 1425 hours0.0165 days 0.396 hours 0.00236 weeks 5.422125e-4 months , both the Source Range Nuclear Instruments (N-31 and N-32),

were manually reinstated because of apparent under-compensation problems with the' Intermediate Range Nuclear Instrument (N-35 and N-36). When high voltage l was applied to N-32, the indication of the channel was not normal and it was declared inoperable. At 1426 hours0.0165 days 0.396 hours 0.00236 weeks 5.42593e-4 months , the licensed operator maintaining the l backboards noted an increasing level on the air ejector radiation monitor (1-RM-SV-121). The monitor periodically alarmed high and high-high. When the high-high alarm was received, operators verified that the air ejector discharge swapped from the vent stack to the containment. The Unit 1 SRO directed that 4 Abnormal Procedure AP-5.1, Unit 1 Radiation Monitoring system, be entered.

This procedure directs Health Physics personnel to sample the discharge of the air ejector and also for Chemistry to sample the steam generator blowdown for l increased activity. However, since low-low levels were still present in the steam generators, the blowdown trip valves were closed and sampling could not !

be performed. The STA, when the air ejector radiation monitor alarmed, checked the other radiation monitors for abnormal indications. The steam generator blowdown radiation monitors did not show any signs of increasing activity, but blowdown was still secured at this time. The N-16 monitors and recorder both indicated zero leakage. The Main Steam Line Radiation Monitors, 1-RM-MS-170, 171, 172, for 'A'i 'B' and 'C' steamlines, respectively, were checked. The 'C' steamline indicated approximately 0.15 mR/hr with 'A' and 'B' showing approximately 0.02 mR/hr each. The STA informed the SRO of these indications.

l With the plant maintaining a relatively constant RCS temperature and pressure, the CR0 and STA did a preliminary balance between charging and letdown. They determined that there was an apparent loss of inventory from the RCS of approximately 60 to 70 gpm. With this and the other radiation monitor

l l

1 and primary system indications, the SRO consulted Abnormal Procedure AP-24.1 and 24.2. These procedures are provided to direct the operators on a course of

)

action to deal with a leaking steam generator tube (AP-24.1 is for a large tube leak and AP-24.2 is for a small tube leak). It was also noted that the containment sump pumping frequency had increased (Figure 11-9). It was

) suspected that this was caused by the air ejector discharge being diverted to containment. So, at this time, both possibilities (i.e., a steam generator tube leak and a RCS leak to containment) were still being considered as

! possible causes of the loss of inventory. The leak from the RCS was still in the range of 60 to 70 gpm. The pressurizer level was being maintained at approximately 20% with one charging pump and the pressurizer pressure was l l

slowly increasing.

By 1440 hours0.0167 days 0.4 hours 0.00238 weeks 5.4792e-4 months , with the low-low levels in all steam generators now cleared, i

the turbine-driven Auxiliary Feedwater pump steam supply valves, 1-MS-TV-lllA l l

and 'B', closed. This effectively stopped the majority of the release-of radioactive steam to the environment. The MFR valves to all generators were manually isolated and feedwater was being supplied to all steam generators via the bypass valves. The blowdown isolation valves, between the steam generator blowdown tank and the containment trip valves, were manually closed, then the blowdown trip valves were reopened. This allowed Chemistry to obtain samples

! from the steam generators and returned the steam generator blowdown radiation i

monitors to service. In addition, the SRO directed the CR0 to reduce letdown flow by placing a lower flow orifice into service.

Over the next few minutes, the motor driven Auxiliary Feedwater pumps were secured. Subsequently, the Containment Gaseous Radiation Monitor (1-RM-RMS-160) alarmed and the SRO dirented the appropriate section of Abnormal 1

- - - - - - _ l

Procedure AP-5.1 be entered. It was-suspected that this condition of. increased !

radiation in the containment was a result of the air ejector discharging to i I

containment, but the possibility still existed that the RCS leak could be from-the RCS to containment, instead of through the steam generator. j At 1450. hours, the Control Room was notified that personnel exiting the Auxiliary Building were being delayed because of gaseous contamination.

Operation's personnel immediately began walkdowns of the Auxiliary Building to determine the source of the contamination. Control Room indication of the

' Auxiliary Building sump did not show an increased level. No charging and letdown system leaks were identified.

At 1506 hours0.0174 days 0.418 hours 0.00249 weeks 5.73033e-4 months , Steam Generator 'C' blowdown radiation monitor alarmed a high-high condition, and again the SRO directed the appropriate section of AP-5.1 be entered. At about this time, it was noted that Steam ' Generator 'C' level was decreasing, with all feed to it secured. This condition was noted by the CR0 and relayed to the SRO. This was confusing, but the operators concluded that this was possible if the leak to the steam generator was small enough to be masked by the amount of steam being drawn from the generator.

Letdown was then isolated in order to search for other potential leak paths.

f Chemistry technicians reported that the Steam Generator .' C ' blowdown sample showed "high" activity as compared to 'A' or 'B'. A report was received in the

! Control Room that Health Physics survey of main steamlines showed that the 'C' i-steamline was reading approximately 250 times the reading of the 'A' and 'B' t

i steamlines. The main steamline radiation monitors showed 'C' slightly above

'A' and 'B'.

1 I

At 1516 hours0.0175 days 0.421 hours 0.00251 weeks 5.76838e-4 months , the SRO directed the CR0 to commence a RCS boration and another operator was dispatched to close 1-MS-95 (This is the isolation valve I

11 - .

'from 'C' main steamline going to 1-MS-TV-111A and B). This was performed in conjunction with AP-24.1. With positive identification of the steam generator and the magnitude of the leak, the Shift Supervisor decided to enter the' Emergency Plan. This was conservative, since the RCS leakage was .now calculated at approximately 75 gpm. Although only one charging pump was needed to maintain Pressurizer-level at 20%. The Emergency Action Level of Emergency Plan Implementation Procedure EPIP-1.01 requires that an " Alert" condition be declared if: RCS leakage is 50 gpm or greater and more than one charging pump is required to maintain pressurizer level.

At 1525 hours0.0177 days 0.424 hours 0.00252 weeks 5.802625e-4 months , an " Alert" condition was declared and the process of bringing the unit to cold shutdown (Mode 5) began in accordance with AP-24.1.

The amount of boron to achieve " Cold Shutdown Boron Concentration" was added to the RCS and the Makeup system was switched from the normal system to the Refueling Water Storage Tank. Then the cooldown of the plant was started. With Abnormal Procedure AP-24.1, the cooldown would continue until RCS temperature reached 500'F, but the SRO noted that by cooling down the leaking steam generator would result in an increased RCS leakage, as the steam generator secondary pressure decreased. The need to deviate the procedure to isolate Steam' Generator 'C' at a higher temperature was discussed among the CR0s, SR0s, and STA. The SRO directed that the Steam ' Generator 'C' trip valve, i

1-MS-TV-101C, be closed at approximately 520*F (whereas AP-24.1 does not ' call for the trip valve to be closed until 500*F on the affected generator is reached). With the Steam' Generator 'C' trip valve closed, the unit cooldown I

continued.

l With the plant cooling down, it became necessary to start an additional charging pump in order to maintain the pressurizer level at 20%. This pump was used for a peried of 19 minutes and then was secured.

-At 1550 hours0.0179 days 0.431 hours 0.00256 weeks 5.89775e-4 months , the Station Manager assumed the duties of Station Emergency i Manager. With'in 10 minutes, the Technical Support Center (TSC) was fully manned. The Station Emergency Manager relocated to the TSC at 1615 hours0.0187 days 0.449 hours 0.00267 weeks 6.145075e-4 months and the facility was-then declared fully operational.

During'the process of the Station Manager assuming control of the emergency organization, the operating shift transitioned from AP-24.1 to ES-3.1, " Post Steam' Generator Tube Rupture Cooldown Using Backfill." The Steam' Generator 'C' water level also increased to greater than 70% during this time, and the SRO directed the auxiliary feedwater pumps be started in anticipation of losing the main feedwater pumps on a high-high steam generator water _ level (75%). Steam Generator 'C' level reached 72% and then started to decrease. As the level in Steam Generator 'C' was returning to the normal operating band, the auxiliary feedwater pumps were secured after approximately 30 minutes of operation on recirculation.

At 1833 hours0.0212 days 0.509 hours 0.00303 weeks 6.974565e-4 months , Unit 1 entered Mode 4 (hot shutdown). During the next three (3) hours, the unit continued to be cooled down and the Residual Heat Removal (RHR) system was prepared to be placed in service. During the process of l

l getting RHR aligned to remove decay heat, one of the inlet valves, 1

1-RH-M0V-1701, failed to stay open. After Maintenance personnel investigated, l the operating procedure was modified and the valve was opened via a jumper.

l RHR was placed in service at 2143 hours0.0248 days 0.595 hours 0.00354 weeks 8.154115e-4 months .

i i The cooldown and depressurization continued until Mode 5 (cold shutdown) '

was reached, at 2212 hours0.0256 days 0.614 hours 0.00366 weeks 8.41666e-4 months . The emergency condition was te rminated at 2220 hours0.0257 days 0.617 hours 0.00367 weeks 8.4471e-4 months .

At approximately 2300 hours0.0266 days 0.639 hours 0.0038 weeks 8.7515e-4 months , the CR0 noted an alarm for high vibrations on the Reactor Coolant Pump 'C'. Personnel monitoring the vibration !

l

instrumentation observed reading of greater than 30 mils, and the pump was j i

secured.

At 2038 hours0.0236 days 0.566 hours 0.00337 weeks 7.75459e-4 months on February 27, 1988, the RCS loop to Steam' Generator 'C' was isolated, and inspection and recovery operations were initiated.

D. Operational Analysis The purpose of this section is to analyze the performance of the Operations and Health Physics shifts on duty during the event and the transitions of key emergency management responsibilities.

The Unit 1 Control Room Operators' (CR0s), Senior Reactor Operstors' (SR0s), and the Shift Technical Advisor's (STA) overall response to the . steam generator tube leak was effective. After the reactor tripped, the immediate actions (as required by EP-0, Revision 1, Reactor Trip or Safety Injection) were performed in a timely manner. Pressurizer level was restored and-prensuriser pressure was stabilized within 20 minutes of the reactor trip. The cooldown to Mode 5 (cold shutdown) was conducted in a controlled manner and without incident. The operational performance exhibited is attributed to the classroom and simulator training received on steam generator tube ruptures which is structured to integrate the Emergency Operating Procedures. Emergency Plan Implementing Procedures, and the Critical Safety Functions.

The Emergency Plan was effectively implemented by the interim Station

[

Emergency Manager (Unit I SRO). A conservative emergency classification was i

made to mobilize the emergency organization. The predesignated communicators for the NRC and State / Local government performed their initial notifications I

and updates, as required by the Emergency Plan.

l 7

During the early stages of the event, the Shift Supervisor consulted by .

l telephone with the Station Manager and the Superintendent of Operations. Upon

]

. arrival, of the Station Manager, the interim Station Emergency Manager began an orderly and detailed turnover. After assuming the role of Station Emergency Manager, the Station Manager located to the already mobilized Technical Support Center.

The operating Shift Supervlsor interface with Health Physics was effective.

The interim Radiological Assessment Director was given a detailed briefing on the radioactive release pathways. Samples were requested for the air ejector discharge. However, these samples took approximately 40 minutes to obtain,

.because of moisture in the sample line piping. Normally, the time required to-obtain and analyze these samples, is approximately 10 minutes.

The operations shift, was quickly supplemented by the " Relief Office" crew, which was.already on site. The " Relief Office" crew (operations personnel assigned to . task other than direct day to day operations of the plant) consisted of two (2) licensed CR0s and three (3) experienced unlicensed operators which were utilized by the Shift Supervisor and interim Station Emergency Manager, as required. When the Operational Support Center was activated, extra operational personnel were also available.

The STA responded quickly to the Control Room and provided technical support to the Shift Supervisor. The STA monitored the Safety Parameter Display System (SPDS) and the Critical Safety Functions (CSF). No Orange or l Red level CSF conditions occurred. The STA also monitored the Inadequate Core

)

Cooling Monitoring System, which provided RCS inventory and core subcooling data, and assisted the CR0 in plotting the cooldown rate information. The STA's initial calculations of the primary leakrate were within 20 gallons per I minute of final calculations performed later.

s i

Major milestones for the reactor trip and subsequent steam generator tube leak are listed in the operational assessment timeline, Table II-1. Because the sma]l magnitude of the primary-to-secondary leakage, identification that a problem existed was delayed. During the July 15, 1987 steam generraor tube rupture event, the leckage was immediately identified because of the large

}, magnitude of the leak and since the tube rupture was the initiating event which ,

i lead' to the reactor trip and safety injection. The February 25, 1989 tube leak event occurred after the reactor trip. The subsequent normal cooldown to 547*F l l

cf the reactor coolant system (RCS) masked.any apparent leakage until the RCS l

pressure and temperature were stabilized. With indications which conid have been interpreted as a RCS leak to either a steam generator or to the containment, the operators took the required amount of time to make the proper determination.

The amount of time required to identify Steam Generator 'C' as the leakage path was increased significantly by the inability to obtain the steam generator blowdown and condenser air ejector samples. The quantity of water in the exhaust piping of the air ejector prevented Health Physics technicians from obtaining and analyzing the air sample, which took approximately 40 minutes. !

The steam generator blowdown sample could not be obtained until after the steam generators low-low level had cleared and then the blowdown system was realigned j 1

I for sampling. The steam generator low-low levels did not clear until 34 minutes after the reactor trip occurred. At that point, operators were l i l standing by to realign the system for sampling. Chemistry technicians aligned '

the sampling system to obtain the sample. The Chemistry technician utilized a hand-held frisker, RM-14, as a quick check of the sample to determine relative I contamination of the three (3) samples. Steam Generator 'C' sample was found l I l-i

to' be relatively more contaminated than the S/G 'A' or 'B' blowdown samples.

When the results of the samples reached the Control Room, the Shift Supervisor made the determination that the RCS leak was going to Steam Generator 'C'.

i With a positive identification of Steam' Generator 'C', the Shift Supervisor l commenced the isolation, cooldown and depressurization of the RCS in accordance.

with procedure 1-AP-24.1. The correct identification of the steam generator with the leak is critical because, once a steam generator is isolated, it can not be easily unisolated during the event. Procedure 1-AP-24.1 does not require immediate closure of the main steam trip valve of the affected steam generator, to completely isolate it. Instead, the procedure instructs the operator to cooldown to 500*F in the RCS before closing the affected trip valve. This would delay the time to complete the isolation of the steam generator.

When the cooldown of the RCS was commenced, the pressure in each of the steam generators followed the saturation temperatures in each of the steam generators. The Shift Supervisor recognized that the decreasing pressure in Steam Generator 'C' was creating a greater differential pressure between the l

l RCS and the steam generator. This resulted in a larger primary-to-secondary l

leakrate. At thet point, the decision was made to deviate procedure 1-AP-24.1 and close the 'C' main steam trip valve to complete isolation of the Steam Generator 'C'. Once the cooldown was completed, RCS pressure was reduced to the Steam' Generator 'C' pressure, thus, the primary-to-secondary leakage was effectively stopped.

l

)

i E. Emergency Operating Procedures (E0Ps) and Abnormal Procedures (APs)

Utilization and Analysis The revision one E0Ps being utilized at North Anna Power Station were l l

implemented April 30, 1987. The procedures used for the February 25, 1989 steam generator tube leak are discussed below-l l

Immediate operator actions of EP-0 (Reactor Trip or Safety Injection) '

were performed and the transition made to ES-0.1 (Reactor Trip Response) within a few minutes after the trip of the reactor. Within the next half-hour, procedure ES-0.1 was completed. No problems occurred with these procedures.

As conditions developed, procedure AP-5.1 (Unit 1 Radiation Monitoring System) was entered three (3) different times. This abnormal procedure was followed when high-high alare.s were received on the air ejector, containment, and blowdown radiation monitors. Appropriate sections of the procedure were used and aided in determining the location of the leak (i.e., that the leak from the reactor coolant system (RCS) was through the steam generator).

Once the Shift Supervisor determined the path of the RCS leakage was through Steam Generator 'C', AP-24.1, (Large Steam Generator Tube Leak Requiring Immediate and Rapid Unit Shutdown) was entered to deal with the transient. As the procedure directs, the cooldown is commenced with the affected generator unisolated and the main steam trip valve not closed, until 500*F is obtained in the RCS. This temperature is selected to ensure that steam generator's relief valves, safeties or PORV, do not lift. While reviewing the remaining portion of AP-24.1 and ES-3.1 (Post-LGTR Cooldown Using Backfill), the Shift Supervisor decided the best course of action to minimize I release of radioactivity was to isolate Steam Generator 'C' before reaching the l required temperature to stop radioactive steam from entering the condenser and I

1 to minimize- the release. In addition, this action resulted in a lower primary-to-secondary leakrate.

l Four major areas of concerns with AP-24.1 include:

1. Clarification of initial conditions of the plant,

2. Transition from AP to appropriate EP,

3. . Process for identification of affected steam generator, and i

4. Methodology for cooldown and depressurization from EP-3, " Steam Generator Tube Rupture" was missing from the AP.

Changes and enhancements to upgrade AP-24.1 are being made.

Once AP-24.1 was completed, selection of th( post-SGTR cooldown'was made and ES-3.1 was entered. The plant was placed in Mode 5 (cold shutdown) within approximately 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months of the trip.

F. Safety Equipment Analysis The purpose of this section is to analyze the performance of safety related or equipment important-to-safety required during the event and subsequent cooldown of Unit l'. The majority of safety equipment required to function during the event did perform its function as required. Components which did not perform'as designed include the following:

The 'C' Main feedwater regulating valve (1-FW-FCV-1498) closed when the instrument air, used to maintain the valve in the required position, was lost when the small cooper tubing broke between the air regulator and the valve positioner controller. The failure of the tubing has been determined to be a

( fatigue' fracture, probably induced by cyclic vibration. The loss of air causes

.the valve to actuate to its required safety position, which is closed. Based on l'

I the results of a root cause evaluation, air lines associated with the Unit 1 and 2 Main Feedwater Regulating Valves will be modified, as necessary, to ;

reduce the possibility of fatigue failures.

l-I

j The discharge piping of the main condenser air ejector 1 nitored for radiation by Radiation Monitor 1-RM-SV-121. When an alarm is received.

Health Physics is directed to obtain a local sample from the air ejector discharge. The presence of water at the sample point prevented obtaining the required sample. The moisture problem may be due to the absence of a check valve on the air ejector drain line and is currently being evaluated by

}

engineering.

The outside blowdown isolation valve for Steam' Generator 'C',

1-BD-TV-100F, failed to close when the auxiliary feedwater pumps started. This l

valve is one of two valves in series. The other valve, 1-BD-TV-100E, the inside containment valve, did close. The apparent cause of valve 1-BD-TV-100F failing to reposition appears to be a sticking solenoid problem. Repairs to I the valve will be completed before Unit 1 is returned to service.

When the Source Ranges were reinstated, N-31 responded normally, but N-32 failed to provide normal indications and was declared inoperable. The detector was found to be defective and was replaced. The investigation to determine the root cause is ongoing.

)

l While the reactor coolant system was being cooled down, the Residual Heat Removal System (RHR) was being prepared to be placed in service to bring RCS temperature down to below 200*F. In attempting to place the RHR system in a proper lineup, an inlet isolation valve 1-RH-MOV-1701 failed to stay open.

The cause of the malfunction was determined to be a failed comparator card which has been replaced.

.~_ _ _ __-- ____ _

l l

TABLE 11-1 1:

SEQUENCE OF EVENTS

SUMMARY

l 0800 - UNIT-1 AT 76% POWER, 725 MW IN COASTDOWN.

UNIT 2 IN MODE 5.

1404 - 1-PT-46.3A, PRIMARY TO SECONDARY LEAK RATE DETERMINATION,

( RECORDED N-16 RADIATION MONITOR LEAK RATES OF LESS THAN 1 GFD 'A' STEAM GENERATOR , 2.39 GPD 'B' STEAM GENERATOR,'LESS THAN 1 GFD l

'C' STEAM GENERATOR, AND LESS THAN 1 GPD MAIN STEAM HEADER.

PRIMARY TO SECONDARY LEAKAGE, BASED ON THE AIR EJECTOR RADIATION MONITOR WAS 0.59 GFD.

l l

1406:49 - 1-FW-FCV-1498, 'C' MAIN ~2EDWATER REGULATING VALVE FAILED' CLOSE DUE TO THE FAILURE OF A FITTING IN THE AIR SUPPLY LINE.

1407:16 -

REACTOR TRIP ON 'C' STEAM GENERATOR LOW LEVEL WITH STEAM FLOW j

~ GREATER THAN FEEDWATER FLOW MISMATCH. ENTERED 1-EP-0,- REACTOR TRIP OR SAFETY INJECTION.

1407:18 -

CONTROL ROOM OPERATOR BACKED UP AUTOMATIC REACTOR TRIP WITH l

MANUAL REACTOR TRIP. AUXILIARY FEEDWATER PUMPS RECEIVED ,

AUTOMATIC START SIGNAL AND STEAM GENERATOR BLOWDOWN ISOLATE ON

'C' STEAM GENERATOR LOW-LOW LEVEL. I l

TABLE II-1 (CON'T) 1407:22 - RM-MS-172 FOR STEAM GENERATOR 'C' MAIN STEAM LINE REGISTERS RADIATION SPIKE FOR 5 SECONDS AND THEN RETURNS TO PRETRIP LEVEL.

1407:23 - ' B ' STEAM GENERATOR LOW-LOW LEVEL.

1407:24 - 'A' STEAM GENERATOR LOW-LOW LEVEL 1408 - ENTERED 1-ES-0.1, REACTOR TRIP RESPONSE.

MAIN STEAM LINE RADIATION ALARM RECEIVED BY THE SPDS SYSTEM AND CLEARED. SUBSEQUENT TO THE SPDS ALARM (UNABLE TO ESTABLISH THE EXACT TIME) A CONTROL BOARD MAIN STEAM LINE RADIATION ALERT ALARM WAS RECEIVED, ACKNOWLEDGED AND CLEARED. THE U1 SRO CHECKED ALL MAIN STEAM LINE RADIATION MONITORS AND NORMAL INDICATIONS WERE NOTED ON ALL STEAM LINES.

l 1411 -

SECURED AUXILIARY FEEDWATER FLOW TO ALL STEAM GENERATORS.

(AUXILIARY FEEDWATER PUMPS RUNNING ON RECIRC.) STEAM GENERATORS BEING FED USING THE MAIN FEEDWATER PUMPS.

l LETDOWN ISOLATED AUTOMATICALLY.

l l 1414 - 'B' STEAM GENERATOR LOW-LOW LEVEL CLEARED.

I I

l

TABLE II-1 (CON'T)

LETDOWN RESTORED 1417 -

CONTROL ROOM OFERATOR NOTICED THAT RCS PRESSURE WAS N0T i

INCREASING AS FAST AS EXPECTED AND MORE CHARGING FLOW THAN EXPECTED WAS NEEDED TO RESTORE PRESSURIZER LEVEL. ONE CHARGING PUMP SUFFICIENT TO RESTORE PRESSURIZER LEVEL TO 20%.

l 1425 -

SOURCE RANGE NEUTRON DETECTORS MANUALLY REINSTATED DUE TO UNDER COMPENSATION OF INTERMEDIATE RANGE DETECTORS. N-32 FAILED- TO INDICATE WHEN REINSTATED.

1426 -

AIR EJECTOR RADIATION MONITOR ALARMS HI AND HI-HI. INDICATION WAS SWINGING ERRATICALLY. 1-AP-5.1, UNIT 1 RADIATION MONITORING SYSTEM, ENTERED. HEALTH PHYSICS IS DIRECTED TO SAMPLE AIR EJECTOR EXHAUST. CHEMISTRY IS DIRECTED TO SAMPLE STEAM GENERATOR I

l BLOWDOWN.

THE SHIFT TECHNICAL ADVISOR CHECKED OTHER RADIATION MONITORS FOR INDICATIONS OF PRIMARY-TO-SECONDARY LEAKAGE. BLOWDOWN RADIATION MONITORS DID NOT INDICATE AN INCREASE IN ACTIVITY SINCE BLOWDOWN WAS ISOLATED. N-16 RADIATION MONITORS DID NOT INDICATE AN INCREASE IN ACTIVITY SINCE THE REACTOR HAD TRIPPED. THE 'C' MAIN STEAM LINE RADIATION MONITOR SHOWED A SLIGHT INCREASE IN ACTIVITY. (INDICATED APPROXIMATELY 0.15 MR/HR, NORMAL INDICATION 1

TABLE II-l (ON'T)

APPROXIMATELY 0.04 MR/HR. 'A' AND 'B' MAIN STEAMLINES INDICATED

]

0.02 MR/HR.)

)

.THE dIR EJECTOR DISCHARGE WAS VERIFIED TO SWAP TO CONTAINMENT ON THE HI-HI RADIATION ALARM. WHEN THE HI RADIATION ALARM CLEARED.

THE. RADIATION MONITOR HI-HI ALARM WAS RESET AND THE AIR EJECTOR DISCHARGE SWAPPED BACK TO THE ATMOSPHERE. (THIS SEQUENCJ OF ACTIONS WAS REPEATED NUMEROUS TIMES UNTIL THE AIR EJECTOR DISCHARGE WAS MANUALLY DIVERTED TO CONTAINMENT AT 1511. THE TIME THE AIR EJECTOR WAS DISCHARGING TO THE ATMOSPHERE IS ESTIMATED TO.

HAVE BEEN LESS TRAN TEN MINUTES.)

- 1-AP-24.1, LARGE STEAM GENERATOR TUBE LEAK REQUIRING IMMEDIATE AND RAPID UNIT SHUTDOWN, WAS CONSULTED.

1427 -

PRIMARY SYSTEM RESPONSE INDICATED A LOSS )F INVENTORY OF APPROXIMATELY 50 GPM. SUSPECTED 'C' STEAM GENERATOR AS THE CAUSE OF LEAKAGE BASED ON THE AIR EJECTOR RADIATION MONITOR AND 'C' MAIN STEAM LINE RADIATION MONITOR INDICATIONS. THE 'C' STEAM GENERATOR LEVEL TREND DID NOT SUPPORT A POSITIVE DIAGNOSIS AT THIS TIME. AN INCREASE IN CONTAINMENT PUMPING FREQUENCY WAS ALSO NOTED.

\

TABLE 11-1 (CON'T) .

'C.' STEAM GENERATOR LOW-LOW LEVEL CLEARED.- .

AIR EJECTOR RADIATION MONITOR EXHAUST SAMPLING HINDERED BY WATER IN LINES. WATER. IN LINES CAUSED THE ERRATIC AIR EJECTOR RADIATION MONITOR INDICATIONS.

1432 - NOTICED TV-BD-100F, 'C' STEAM GENERATOR OUTSIDE CONTAINMENT l

ISOLATION VALVE WAS OPEN. IT SHOULD HAVE BEEN CLOSED. VALVE L COULD NOT BE CLOSED MANUALLY. TV-BD-100E, 'C' STEAM GENERATOR 1

INSIDE CONTAINMENT ISOLATION VALVE, WAS VERIFIED CLOSED.

l l

, 1440 -

'A' STEAM GENERATOR LOW-LOW LEVEL CLEARED. STEAM DRIVEN l

AUXILIARY FEEDWATER PUMP SECURED. (1-MS-TV-111A & B CLOSED.)

MAIN FEEDWATER REGULATING VALVE MANUAL ISOLATION VALVES CLOSED.

- BLOWDOWN TRIP VALVES OPENED, SAMPLING COMMENCED. (BLOWDOWN FLOW TO THE BLOWDOWN TANK HAD BEEN MANUALLY ISOLATED.)

BLOWDOWN RADIATION MONITORS RETURNED TO SERVICE.

LETDOWN FLOW REDUCED BY PLACING A LOWER CAPACITY FLOW ORIFICE IN SERVICE TO DECREASE CHARGING FLOW.

l' l l

f 1

TABLE 11-1 (CON'T) i 1442 -

'A' AND 'B' AUXILIARY FEEDWATER PUMPS SECURED AND RETURNED TO AUTOMATIC.

1443 -

HIGH RADIATION ALARM RECEIVED AND LOCKED IN ON RM-RMS-160, CONTAINMENT GASEOUS RADIATION MONITOR. AP-5.1 ENTERED.

1445 -

CLOSED 'C' FEEDWATER REGULATING VALVE BYPASS VALVE AND ITS MANUAL ISOLATION VALVE.

1 1448 -

HIGH RADIATION ALARM RECEIVED AND CLEARED ON RM-VG-103, A STACK RADIATION MONITOR.

1450 - 'C' STEAM GENERATOR LEVEL TREND WAS CONSTANT.

CONTROL ROOM RECEIVED INFORMATION THAT PERSONNEL IN AUXILIARY BUILDING WERE BEING DELAYED IN EXITING DUE TO SLIGHT GASEOUS CONTAMINATION. '

EXTENSIVE WALKDOWN OF THE AUXILIARY BUILDING WAS COMMENCED.

AUXILIARY BUILDING PARAMETERS HAD BEEN MONITORED AND DID NOT SHOW SIGNS OF A LEAK.

1506 -

'C' STEAM GENERATOR BLOWDOWN RADIATION MONITOR HIGH/HIGH ALARM l

RECEIVED. AP-5.1 ENTERED. j 1

i 1

I TABLE II-I (CON'T) 1511 -

LETDOWN WAS SECURED IN AN ATTEMPT TO ISOLATE POSSIBLE LEAK PATHS.

- 'C' STEAM GENERATOR LEVEL OBSERVED TO BE DECREASING.

CONTROL ROOM RECEIVED INFORMATION THAT PERSONNEL EXITING THE SECURITY BUILDING WERE BEING DELAYED IN EXITING DUE TO SLIGHT GASEOUS CONTAMINATION.

AIR EJECTOR DISCRARGE MANUALLY DIVERTED TO CONTAINMENT DUE TO CONTINUED ERRATIC INDIC/ TION.

1516 -

COMMENCED EMERGENCY BORATION IN ACCORDANCE WITH 1-AP-24.1 1517 MS-95, 'C' STEAM GENERATOR STEAM SUPPLY TO THE STEAM DRIVEN AUXILIARY FEEDWATER PUMP, WAS MANUALLY ISOLATED.

1520 -

CHEMISTRY REPORTED 'C' STEAM GENERATOR BLOWDOWN SAMPLE SHOWS HIGH ACTIVITY. 'C' STEAM GENERATOR POSITIVELY IDENTIFIED AS THE SOURCE OF LEAKAGE.

- EPIP-1.01, EMERGENCY MANAGER CONTROLLING PROCEDURE, ENTERED.

l 1521 -

HEALTH PHYSICS REPORTS CONTACT READINO FROM THE 'C' MAIN STEAM TRIP VALVE OF 5 MR/HR. READING THE DAY BEFORE WAS 0.02 MR/HR.

4 i

I 1

l

i TABLE 11-1 (CON'T) 1525 -

LEAKAGE ESTIMATED TO EE APPR0XIMATELY 65 GPM. STATION EMERGENCY MANAGER DECLARED AN ALERT BASED ON A REACTOR COOLANT SYSTEM LtAK !

0F GREATER TRAN 50 GPM. (THIs WAS A CONSERVATIVE ACTION BECAUSE THE EPIPS REQUIRE AN ALERT TO BE DECLARED WHEN REACTOR COOLANT SYSTEM LEAKAGE IS GREATER THAN 50 GPM AND MCRE THAN ONE /1)

CHARGING PL'MP IS NEEDED TO MAINTAIN PRESSURIZER LEVEL.)

f 1526 -

REACTOR COOLANT SYSTEM C00LDOWN COMMENCED IN ACCORDANCE WITH 1-AP-24,1.

1531 -

CHARGING PUMP SUCTION MAKUALLY SWAPPED TO RWST FROM VCT.

1544 -

WITH REACTOR COOLANT TEMPERATURE AT APPR0XIMATELY 520 DEGREES F.

THE SHIFT SUPERVISOR EVALUATED THE EFFECTS OF COOLING DOWN WITH THE 'C' STEAM GENERATOR MAIN STEAM TRIP VALVE (1-MS-TV-101C)

OPEN. DECISION WAS MADE TO CLOSE THE VALVE AND CONTINUE TO COOL DOWN.

l l

1550 -

STARTED A SECOND CHARGING PUMP TO KEEP UP WITH INVENTORY REDUCTION DUE TO C00LDOWN.

l STATION MANAGER ASSUMES POSITION OF STATION EMERGENCY MANAGER.

l 1

'1 1

TABLE I1-1 (CON'T) i 1600 -

STARTED AUX FEEDWATER PUMPS ON RECIRCULATION IN THE ANTICIPATION OF LOSS OF MAIN FEEDWATER ('C' STEAM GENERATOR LEVEL AT 72%).

'C' STEAM GENERATOR LEVEL STARTED TO SLOWLY DECREASE AS PRESSURE l 1

IN THE REACTOR COOLANT SYSTEM WAS REDUCED TO BELOW STEAM l GENERATOR PRESSURE.

TECHNICAL SUPPORT CENTER (TSC) FULLY MANNED. i

- ENTERED 1-ES-3.1, POST-SGTR C00LDOWN USING BACKFILL.

1009 -

SECURED SECOND CHARGING PUMP. THE PUMP WAS NO LONGER NEEDED TO MAINTAIN PRESSURIZER LEVEL.

1613 -

LETDOWN RESTORED.

1615 -

STATION EMERGENCY MANAGER RELOCATED TO TSC AND TSC IS DECLARED l OPERATIONAL.

1619 -

SECURED EMERGENCY B0 RATION.

1 1631 -

SECURED AUXILIARY FEEDWATER PUMPS.

1833 -

UNIT ENTERED MODE 4.

l TABLE I1-1 (CON'T) l l

l 2143 -

PLACED RHR SYSTEM IN SERVICE. 1 2212 -

UNIT ENTERED MODE 5.

2220 -

ALERT TERMINATED 1

f i

}

TABLE II-2 OPERATIONAL ASSESSMENT TIMELINE CLOCK TIME TIME FROM Rx TRIP HOURS MINUTES I REACTOR TRIP 1407 0 RELEASE THRU SDAFW 1440 33 PUMP EXHAUST ISOLATED RELEASE THRU AIR EJECTOR 1511 64 EIHAUST ISOLATED S/G IDENTIFICATION 1516 69 S/G COOLDOWN COMMENCED 1526 79 S/C ISOLATION 1544 97 S/G PRESSURE EQUALIZED 1600 113 WITH RCS PRESSURE i l

i l

l l

[

FIGURE II-1 m

M h

@ W C

(It

.. a O C<

N@ CY x=

- sn N ..

W c

C CD N

Wa I ;

' } i i N

G

. t- co r : ) i C ! l i I , i M C l l i i j l

W : -

3 l

1*

s CD H 1 ,

( j ca 1 ;? 1 :

M.. C

i Cbc0: ,

cT j j ... . .. .j -

p. . . . ..- e (0 a !! ) )- i 8.

ar . ,

- N d,l' j i I. !

y .. , .

} l <

dN f,'. I I I

W ** i f i I j I CT ] l g

-H -

I ( l .

>- ., ......>o..~.._ >

cp

& 4 s : i a y a WC l ,'

I.

I. ) i.

N 19 u

3 ; "

CD : : i ) j T l i e I e <

e

.r.4 < l } j -

! N u , : : : , =

N : I. . .

s i . i O I 1' ,

e

< {' i . 0*

U l t s C r.

; i ( CD T 'r4 wv CL ,

- t ) .

9 l

I s

i Im l

m" -

t 5e l I i. ... ax j

j

. 3a e ,

g 4

} ( j '

4Jr l

, r 3 g .

r } f .f .

l i j 1 i 3

< ; L.... - ^

.....-.~.--r-.- O i } l' l i T N

f, ,

e ,

> (f) y i : , ; O M f h g'., : : .{

i n' o ck , , , , , , . . . .! ...l. : -

M HN '4 H O. G. G. C. C.

g t N CD CD @

h4 a a gg m N m 1.N3333d 1

_ _ _ _ _ _ _ - - - _ _ _ - _ - _ _ l

9 8

5 29

8 2: 0 5 .

2 0 E: 0 T2 6 A

D E

8T 5A 9D 0

93 .

54 4 4 t 0

8 3: 4 0 -

2 n 4:

17

?

E:

1 i4 "

" 0 I1 .

T 4i0

+ , E T R 6 U 1A p 3 S S

8 ;

E

R 4 ; P

1 R R m

Z P

0 H ,

. 5 C i iN i -

0 5 4 4 P -

- {4 2 1

- I hl < -

R t T

~-

MJ T --

P r

r r

g _

g g -

0

+

_ 7

' ( 0 2

g 1 E 7 N ;

_ u g

- E$

SS u AE -

BE

- DF

.n

^ - " * - . . - - . -

0 O 0 0 1EN .

0 0 0 0 0 0 T 4 6 8 0 2 IR 2 1 0 0 9 NI 2 2 2 2 1 UF e yu_

lli !l

TS SD RN 9 I O 8 FC

E FS 5 O 29 0 D0

8 N6 2: E 0 5 .

2 . > 0 E: - -

- 0 T2 . -

6 A .

D- .

E .

8T ".

5A ._- ,

8D r

. . 0 93 .

54 I' - , 0 8

3: '

4 0 " ,

2 -,

4: a" 17 u

a m

E E: a -

m R M4 a. .

U

. 0 S I1 -

. ES T - .

0 RE

- 2 ? UR T ._ ' *_

6 SP 1A '

3 S

'.. EG 8 " R/

PS 4 " .

1 "

R' ZC R -

, P' 0

H ' .

. 24 C a' " 09

~

' 0 44 4

P .

2 11 I '

R -

. TT T n, - m PP

. m. _

, m E ,. .,

0 _

_^'

u .

1 4 4.' 3 0 S ,

. ., , 2 S -

=

.,, 1 E =

=.

= ..

S F

D E u .

6 t N S-A S

i i

. O C

- B D

C ~,

i i,

.. E S

E 0 0 0 0 1 . . . . .

0 0 0 0 O T 2 0 8 o

~

I# ,4c 9 6 2 o.

NZ 2 1 1 1 UF o m n.

~

1

i l

FIGURE II-4 m

CO

'l Nm END OF GETARS l

- G DATA (1200 SEC) cu ..

O

e i w

H eu c

j ,

O j , I w < ,- :

Ww i .

oc .1 : n

@O

! i

r n.

U ': G e co dv .....

..j {..

...... [

rm m

y ..

He ::: ': *

.. cu ::l :

T .. It .

v4 N O i.

)

w ..

i'.

i;. .

o zy 1e : x g 1:. :- G :s g ac ,, e m

j ...... . . . .: ;.. cp g fj g

ae h n e o p:

m-

. ,, , cn e y i CO .i r ll .

WC

%N qr cm

ev4 g

[

. e-

. l Nu

:p c-r :- i' o . t cu O '

eo

.1l 1 : s vv

c. 1:: 1 e HH

- 1 :: : cu ,

i . I i E

]) : FH i H a : .

g \

] - : '

J t -

J g I ' '

f .

.. <.1,

r n,.

, G.

....- ; r.. . ....... ,.

t-- cp

( . : (u i

[ :: v4 i

.. t. '.

i l n m wD U :- O M it : Z

\ . o g%

i .

' O m -

oQ . e : w

, ac . , , i i . , y a g e s s y s. . . e. .

g o e c3 o e START OF I

}. T O M 'D N SECOND 600

) Z cu 'o e e H SECONDS cu H H H H D

91Sd s

FIGURE II-5 m

o..

to OJ @

.. m Na e.

to N o W .. o sca m C

l g

Q w

Co r ,

cc ' !

o.. o

$* (f) '

c.v : ....,.. e a ce> ..

o e

@ G o

~w ,

<r .. G as i z w .. ,

h re '

e 5

-a -

a c::

H g + . o y v4 C m e N

o E e

% +v \

)

u 2 -

1 <

, o .

l .

a +.

-e ,

,. g TA o *H a , y *v 1 Ho e .

HH r Jx j 1 l i =Jr

< r

. . . . e i

, 1 v 01 l Vs 4

o w- z oW t !

o aa -

o m w

a. . . . . . , g d

g\ o o o o o o.

i p% . . . .

o o

l

~ o o o o 34 ov4 CD o v N 1N3033d

.l l

i l

1 1

FIGURE II-6 1 l

l l

m CD -

D N@

- 0)

N .. g D .

.s . . G 1 W .,,

i y .. ,--h e.

i r.

v. Q i sta

v. q) '

)

C ; I'I y;.

c 1l j :,

w '

15 t CO H <

O CI '

i n H l 0)Q '

,y

.cr

r O OO

..r .

ll '

i

'[...'fl c Ot I ll i

"i, .

. [ co cn .. i, i.

, e .

. .*.E, T

.. ruG) .,,

- i.n v

db 1 -

a >-

P.

um 3 I oD i

',ii --.

'.. iI .'.

p y .

a :.:

w ..

a o ,, 1:- 1 C4 $& Y.

O =>

ma l -

< - e ,': ,

ud H !.- -y" l', , ,, ............

.$(v) $@

[C i l1 dh t ,

5 U2!: ee n..! >r ss 1y,; to tn th ,

[

T l !' ', i r ou

,I ,' ',,

(g y 1, ,, .; [ --

i i

i l l i, .

.s i

% .ll }

, , . ,4 i ,

c3 1

,- l g g I i< u u o.1 6 U

,'l l . . . . . - - * -

ww " -~ g @@^

-~=

4~ ;

, @ m ,.A

<ll } ! it -** WW Y1 '.

Q. b i- g

- ;ll e

H i CD HHH W

l- LLE JI w 11, l

< i i

! I ! i i l o

!lll c.

) 'i i i

,t -

l l 1 t 3

; l !

I _ q '

,- .' . . . . ,i . . <i

,l l o i

\ '

i oJ 1 lli ,A

l. s j : il I l ll,

} ;

o t (h o

(

4 3

W

) l Z j

%g ; j lll- Q l q ;

' ,, 'L O C

CD )

t i

.I I; W Q . . . ,l, ,' . . - ' . . . <'

',' c) w 24' , , e e* o o 2 . .

g

'D e T M " H,'N

l1i!Jl1ilI i !l i

@g E" 9

8 5

29 -

8 0 -

2: .

5 0 2 0 E: -

2 T2 ~ 1 A

D E

8T 3A ~

8D 0

93 54 4~ 0 6

3: 9 0

2 4:

W 17

OW

- LO F L E: ' F M4 0 G NN I1 I W T 4' 0 GO T

.I 2 RD AT 1A 7 l i E CL 8 1 1 4

1 N W

.- TT I I R NN O .

DW UU TO -

._ - 0 EL E H LF C .;i , j ,, 0 2e7 8 151 P -9 -

4 11(

I -

1B R -

TTT T G FFR N _

~G~I

~RW .

Av v

AO

~ilC :t L 0 1_

. 0

~ 4 N/

W -

g,. -

2 C -

y- ~ -

6 _

~ - S D

~ - D E T -

N SE

_ ~ O AL w ~

2 C B E D& ~ w. . , . .. . . - S 1 S N 0 O 0 TI 6 0 . . O. O.

0

- 0 I

0 0 0 NN U C 00 6 2 8 4 2 1 1 .

rde l lllllllll l

FICURE.II-3 (1-FW-FCV-1498) n q .

pg l .

..p * '

,s,t 3

4, 3 ,

m 3e4

-}k 84jx .';k{% yty' . .?"f 4 ~

.wp.

y y.

,p ,

pe4w: ..

m . ,

. r ' f,f , ,

., om . h,., %

', fr 9.., .

} , . . y 4 r, -

7

. s e _

n; s ~

1

.2 j+g

. .?, ~

l a >s $ v i

. O' ,s! .e e

%,. a.

?.,

. tg$%l 4 mG

' W ..b $ hjjp s r "$ -f , nj . ?

g'?%. '-lf}ll e

'4% . . q;y.

~

,& w .gP

' W 'MMW., '

ge.,...

N N- -).. .

4 i. :

I II5kkMg . ? )

.. y wq; ggg m ~. z p p:

b~

e ,

1 m

~ ~

4 t- 4 ,

l * -

$g ;'^ ' {,(

, 3..~--- p a :g p , TI ,

4 mm m . *

~

~k'j:Qgp'sfi f Q j aAy% .,

, M* % % g f # g g ea pi j a ;3gg y a:e.;ns ,.

m-

- W W. - . .c ca w y e g; M d

(,:: ~.-g, gg *

/

r w.;.g-g .

> v . j*p

., . a - . A.

k

. %,. s t.i$6hj:; k -

p (3,,. w b N.

y -

y.gh: ' ~ pg,_7 s,3,(; < .y+.

..; . p .c - ;n:y--~%w. w

m. w. ++,

. .x e A. ~' .

,i

p;%yha:%m;>%,, p4 "-.,eg +

., , f T 2 E,, g.y - -

m, .

sg.: .

.-+

J

FIGURE II-9 CONTAINMENT SUMP PUMPING FREQUENCY i

._.h.2 8

t l

.._&__. _,. _ _ a. _.

e l l

.. f _. __ f .. I *

N..J ; ..

4

. .y ._.

i i

..p I . _'- -

___! .. i I y

i . _. 7. . . x L. _,,_

i .i I _ J.. ca

y

~

0 01 09 ! 09 , ,1 OF .:J 02 1 0

_ ,. p .

r -_ . -

oa CONTAINMENT r- .

.l_.. . ..L

. l_ . _ . . . . _.. d .

j

[8 _

h Wbc-

^

SUMP LEVEL m; ,.{

7' p,

'-~1-~ -

l , .

l , ,l T- -

'- K

_.i

.L. _,e a_l . #,

i l

. i e

- .q. .. __7_. .._,.__

r- -- .

,'i I

-j l i. ,

a _.j eu , ,

.. , ._.7,._ _ . .i, . .7_- ___ . . .

s; ~

i

-_ sl I

i tr .a 1 0 01

__i._09 l 8

1 l 09

. _-_ 10F-@. -

O I !

--QKI --

.p.. .__

. .- r \

r----- j --

\ ,

. .y -

, .qq- - 7. .

l ; ,I l

. ,,3 w .' m a .gi

. ! .W '

I

@ __e; i

l .,_. l' 1 , .

g l

'j i .; ,,, !

_a i i l:

.l ,' '

3 ' j i! h le 4."l 1

lpqg

I

..cl;i:, .

g 0 01 09 : ,

09 '

.!- .OF '

O' i

,;,_r g.}

f "I -

' ' 5

'l' e m !_.

,)

. i; i, c 6,. - - -

-l

.' I l-i me, FpH 5I 7 p i f* -

'I i'

g --

f: q' I ;I 8 ii' , ' . ' .i i.' ii : j "t:3'""8~~

1, .n -

, i l . il n - _=

p'

,Il : g- - *;ll 'j- Y

-i;. i' , ,,. . :,: .gi o .., , , 4as,. pdese., 4

,1

.l..

,.'. .i . I ..i. h ,,'..i .

.i ,

a.-

, ii.

. , gm I 0 01 , -

0,9, . ,:n '

i 09 l, . 'l, 07

., 9 p. . O

' .i ; l ; :': ..

,r-->

l \.. + cy . . l. @% .

Og

.) ! ,. 4, ! j =

I

l

-t

.i '

=; p - - - + - ---

l i ;, a i w+ i

'l -{ NU ' '

I.

.._ . . . . _ _ _ . _ . ... W, b.='iis-=l -

, l

_ .1 .

_., ..,.__ __ _ .. . i.... _..

_. a . . _ . . .

I .. _._: _ _ _ . _ . . . _ . : .. . ,--

i ~ ~.1. *v=

i l

l l I

i FIGURE 11-10 m i CO D

OJ m

" 0)

N ..

D O N .

Wa I::

1. O em j} E S !

O- j, y , :. .

x CO H j ,i P DC Si

, C

n. 2

03. C J, ,i h.

1 Ch (ej D ,y *;

3 3

(,, . .

.:s t'. ... .

. .' e m c

s s ,.

,M y

a "N ;I <.

t .. it g

1'. ,

$M rifs i! I2 1 ::

. F

'J ,j w .. ,

-, u~

Ct 1 ,

= ::o wv4 1 .'.' 9 O H ~O l' .: 4.. ..; t.

-e dH

,,4 g 9i O i j . 'D U (Q

sr L

8. ) $awc z8J

y H 1r E' 2

- *C M 'l [

'.k 5.

.. - 5

% 4

t ! --

3' Z )! j .:

U

' i - o .

OJ is

{\ ...-. $.- "..

i;-

1 c v4 rs 1

I: g 0) r4 m Cv A "'I: .

I CD H - .

EH l

1 a 1 ..

xx

'I

. a,-

e a.: ,

- t

. i

- e . .y

if G.

1 l

$-- .r

.' l c l J 'l GJ L~ : I e( v4 A h" -

>. 3 0 : t' t

J it u)

\

l Y

> kt .

O

~

C : <

i

s '.

O m i

' j ,' ): O c; : :

i.

i < t . . .I .

i .

] ;* W i , .

i-- see c) d h,

3 N m v O t h D (' . . . . . )

v w e i a

/ .'d l I i

III. RADIOLOGICAL EFFECTS

'A.- Summary

' Releases. of noble gases, radioiodines, .particu' late and tritium were

evaluated for : applicable airborne pathways. The.following activities were calculated to have been released:

Noble Gases: 5.28E+0 C1.

Radiciodines: 9.44E-8 Ci Particulate: 4.54E-8 Ci Tritium: 1.00E-2 Ci Dose rates, calculated based on estimated maximum release rates, were well within Technical Specification allowable dose rates at and beyond the site boundary:

Tech Spec Dose Rate Limit Event Dose Rate % Tech Spec Total Body, 500 mrem /yr 1.2E+1 mrem /yr 2.3E+0 Skin of Whole Body, 3000 mrem /yr 2.4E+1 mrem /yr 8.2E-1 Critical organ, 1500 mrem /yr 3.0E-2 mrem /yr 2.0E-3 Doses, calculated based on estimated total activity released, were well within Technical Specification limits for doses at and beyond the site boundary:

Tech Spec Quarterly Dose Limit Event Dose % Tech Spec. :)

Noble Gas Gamma Air Dose, 10 mrad, 2.0E-3 mrad 2.0E-2 i

Noble' Gas Beta Air Dose, 20 mrad 2.7E-3 mrad 1.4E-2 Critical Organ Dose, 15 mrem 1.1E-6 mrem 7.3E-6 Environmental air samples and TLDs collected after the event did not L indicate any measurable increase in radioactivity or radiation levels at the monitoring locations.

l I

B. Discussion of Release Pathways Release pathways evaluated for airborne activity include Vent Vent 'A' .snd the Steam Driven Auxiliary Feedwater Pump.(SDAFWP) exhaust. Vent Vent 'A' discharges include activity originating from the Unit 1 Condenser Air Ejector (CAE) which is considered a potentially significant source of noble gas releases, if primary to secondary leakage is present. Vent Vent 'A' is monitored by the wide range Kaman Monitors, RM-VG-179-1 and RM-VG-i*9-2.

SDAFWP exhaust steam is considered .a potential source of radioiodines and particulate since the steam used to drive the turbine could be contaminated (turbine steam discharges directly to atmosphere.) SDAFWP exhaust piping is monitored by Nuclear Research Monitor RM-MS-176. Each pithway Js discussed in the following evaluations.

Since the Main Steam PORVs or Main Steam Safety Valves were not used or activated during the event, it is assumec that neither of these potential pathways are applicable.

1. Vent Vent 'A' Releases The noble gas activity released through Vent Vent 'A' was determined based on the Kaman monitoring system RM-VG-179 strip chart data. The applicable portion of the chart is shown in Figure III-1. Chart data indicates an increase in Vent Vent 'A' activity at about 14:26 which coincides with the alarm on CAE Radiation Monitor SV-121. Therefore, 14:26 is considered the start time for assessing the Vent Vent 'A' noble gas release for the event.

The release rate remains essentially constant until about 15:03 when an increase in the release rate by a factor of 10 is indicated. The decrease at ,

1 !

l 15:13 is assumed to result frou the manual divert of CAE discharge to ',

containment. (The sequence of events summary indicates the divert at 15:11, l

which allowing for a time delay, coincides with chart indication.) The release rate.-then gradually trends downward with a minor reduction indicated at.15:50 and a slight increase at 16:13 until between 17:17 to 17:24 when release rates are similar to those prior to the event. The time period for which noble-gas was being released through Vent Vent 'A' and considered applicable to the event is then from 14:26 to 17:24, for a total time of 178 minutes.

The total noble gas release through Vent Vent 'A' was determined by manually integrating the relsase rate chart curve, which shows noble . gas release' rate in uC1/sec Xe-133 equivalent. A correction for detector response was evaluated based on the isotopic composition from the 14:55 CAE grab sample, which indicated that the detector would be over-responding by a factor of 1.47 due to the energy dependence of the plastic scintillator beta detector. The total integrated activity released was adjusted by a reduction of 1.47 to account for detector response. Table III-5 summaries the results of the CAE grab sample, including nuclides, uCi/cc, and the percent of total activity each nuclide represents.

Dose factors provided in the ODCM were used to calculate site boundary total- body and skin dose rates and the gamma and beta air doses resulting from Vent Vent ' A' no::zle gas releases. For consistency in calculating dose rates, the release rate was averaged over the period of one (1) hour which included the maximum release rates. A total body dose rate of 1.7E+1 mrem /yr was obtained, which is about 2.33% of the Technical Specification limit of 500 l mrem /yr. A skin dose rate of 2.44E+1 mrem /yr was obtained which is about 0.82%

of the Technical Specification limit of 3000 mrem /yr. Gamma and beta air doses

)

[ were calculated to be 2E-3 and 2.7E-3 mrad, which are about 0.02% and 0.014% of the Tect ical Specification litr.its of 10 and 20 mrad per calendar quarter for

) k gamma and beta air dose respectively.

I l

2. Steam Driven Auxiliary Feedwater Pump Exhaust Releases l Releases through the SDAFWP exhaust (also known as the Terry turbine) were considered to have occurred during the time period 14:07 (when the AFW pumps were automatically started) to 14:40 (when the SDAFWP was secured), for a total release time of 33 minutes. A steam flow rate of 182,400 grams per minute total (about 400 pounds per minute or 48 gpm, maximum design flow rate) from all three steam generators is assumed, with each contributing one-third of the total flow rate.

Particulate and radiciodines were calculated based on steam generator sample activity, and assuming a partition factor of 0.01 is applicable to the concentration of activity in the steam versus activity in steam generator liquid. The nuclide concentrations used are provided in Table III-4. Two different concentrations are used, based on the assumption that for the period from 14:07 to 14:26 (19 minutes) the steam generator activity is reasonably represented by samples obtained on February 23, 1989. For the period from 14:26 to 14:40 (14 minutes) the steam generator activity is represented by samples obtained at 14:50 on February 25, 1989.

Tritium release was calculated assuming a tritium concentration of 1.66E-3 uCi/ml in each steam generator for the 33 minutes of pump steam flow. This concentration is based on the tritium analysis available on Steam Generator 'C' q

and is considered to be conservative.

Noble gas releases were calculated assuming the period from 14:26 to 14:40 j s

l (14 minutes) is the time of potentially significant noble gas releases from this pathway. Since the SDAFWP exhaust monitor did not indicate any activfty significantly above background (background about 0.02 mR/hr), the release rate was conservatively estimated based on the fraction of steam flow going to the l 1

) .

]

SDAWP versus that going to the condenser. A maximum steam flow to the SDAWP (48 gpm) and a minimum feedwater flow rate of 340 gpm is assumed, for a ratio of about 1:7. This is the same ratio which should be expected for the noble gas release rates. Based on the maximum instantaneous release rate indicated for Vent Vent 'A' (5880 uC1/sec), an estimated SDAWP exhaust release rate of 832 uCi/sec is obtained.

The maximum release rates of particulate, iodines and tritium occurred from tht SDAWP exhaust, and are provided in Table III-3 (noble gas values are provided as information).

Dose factors provided in the ODCM were used to calculate a site boundary )

critical organ dose rate and the do . to the maximum exposed member of the public resulting from the SDAWP exhaust releases. For consistency in calculating dose rates, the release rate was averaged over a period of one (1) hour. A dose rate of 2.96E-2 crem/yr was obtained, which is approximately 2.0E-3% of the Technical Specification limit of 1500 mrem /yr critical organ dose rate. The critical organ dose was calculated to be 1.09E-6 mrem. This dose is about 7.3E-6% of the Technical Specification limit of 15 mrem / calendar quarter critical organ dose.

C. Environm?ntal Samples Obtained and Results During the event, wind direction was generally westerly (blowing from the west). Airborne particulate and iodine samples were collected from environmental air samplers in the following sectors: E, SE, and SSE.

Environmental TLDs were also collected from the following sectors: NNW, N, NNE, NE, ENE, E, ESE, and SE. The air samples and TLDs were forwarded to the vendor and expedited analyses were obtained.

I E

l I

i l

i Air sample results indicated airborne radiciodine less than detectable (all less than 2E-2 pCi/m ) and gross beta between 3.6E-2 to 5.9E-2 pCi/m which are typical environmental levels. No detectable gamma emitters were present.

TLD results for the exposure period of 54 days (January 4, 1989 thru February 27, 1989) varied from 7.1 to 15.8 mR. If these exposures are converted to exposure per average month (30.4 days), the values vary from 4.0 to 8.9 mR per month. These values are within typical environmental levels around the station, based on the environmental monitoring program which has been in place prior to station operation.

I i

i i

I

( l l

1 1

1

)

TABLE III-1 CALCULATED ACTIVITY RELEASED DURING THE EVENT Ralease Rate Nuclide Curies Released Curies Per Second Ar-41 4.75E-3 9.44E-7 Kr-85m 1.11E -2 2.20E-6 Kr-87 1.06E-2 2.10E Kr-88 1.80E-2 3.57E-6 Xe-133m 7.66E-2 1.52E-5 Xe-133 2.61E+0 5.19E-4 Xe-135m 1.67E-1 3.31E-5 l

Xe-135 2.33E+0 4.63E-4 Xe-138 4.70E-2 9.33E-6 I Sub-Total 5.28E+0 1.05E-3 1-131 1.18E-8 3.28E-12 I-132 1.89E-8 5.25E-12 l 1

1-133 1.90E-8 5.28E-12 i I-134 1.55E-3 4.31E-12 l l

I-135 2.92E-8 8.11E-12 ;

Sub-Total 9.44E-8 2.62E-11 l l

l l

l

) ,

s

____-_.m___..___-_-.._____..--. _ . _ _ _ _

I TABLE III-l (CONT'D.)

I Na-24 3.04E-9 8.44E-13 Rh-106 1.39E-8 3.86E-12 Cs-134 6.04E-10 1.68E-13 Cs-137 1.03E-9 2.86E-13 Cs-138 1.97E-8 5.47E-12 Sub-Total 3.83E-8 1.06E-11 H-3 1.00E-2 2.78E-6 Sub-Total 1.00E-2 2.78E-6 I

I I

I

1 1

TABLE III-2 CALCUIATED ACTIVITY RELEASEP THROUCd VENT VENT 'A' Release Rate Nuclide Curies' Released Curies Per Second Ar-41 4.32E-3 8.25E-7 Kr-8am 1.01E-2 1.93E-6 Kr-87 9.60E-3 1.83E-6 Kr-88 1.63E-2 3.12E-6 Ke-133m 6.96E-2 1.33E-5 Ie-133 2.38E+0 4.54E-4 Ie-135m 1.52E-1 2.90E-5 Ie-135 2.12E+0 4.05E-4 Ie-138 4.27E-2 8.16E-6 Sub-Total 4.80E+0 9.17E-4

) .e.

)

TABLE.III-3 j CALCULATED ACTIVITY RELEASED FROM SDAFWP EIHAUST Pelease Rate Nuclidt Curies Released Curies Per Second 1 i

Ar-41 4.28E-4 1.19E-7 i

Kr-85m 9.98E-4 2.77E-7 I Kr-87 9.50E-4 2.64E-7 Kr-88 1.62E-3 4.50E 4' Xe-133m 6.89E-3 1.91E-6 Xe-133 2.35E-1 6.53E-5 Xe-135m 1.50E-2 4.17E-6 )

Ze-135 2.10E-1 5.83E-5 Xe-136 4.23E-3 1.13E-6 Sub-Total 4,75E-1 1.32E-4 l

I-131 1.18E-8 3.28E-12 l I-132 1.89E-8 5.25E-12 L

I-133 1.90E-8 5.28E-12 1-134 1.55E-8 4.31E-12 I-135 2.92E-8 8.11E-12 Sub-Total 9.44E-8 2.62E-11

I

'1 1

i l l l

TABLE III-3 (CONT'D.) j Na-24 1.02E-8 2.83E-12 Rh-106 1.39E-8 3.86E-12 Cs-134 6.04E-10 1.68E-13 j i

Cs-137 1.03E-9 2.86E-13 !

Cs-138 1.97E-8 5.47E-12 Sub-Total 4.54E-8 1.26E-11 l l

l H-3 1.00E-2 2.78h-6 Sub-Total 1.00E-2 2.78E-6 f

t I

t 1

i

}

I

TABLE III-4 STEAN GENERATOR ACTIVITY PRIOR TO AND AFTER EVENT Activity: February 23, 1989 (Time: 1610)

'A' 'B' 'C' Nuclide uCi/ml uC1/ml uCi/ml H-3 Note 1 Note 1 Note 1 Na-24 ND 6.89E-8 ND.

Rh-106 ND ND ND I-131 ND 1.85E-7 6.91E-8 I-132 ND 4.06E-7 ND I-133 ND l.*iE-7 9.67E-8 I-134 ND ND ND l I-135 ND 4.58E-7 ND Cs-134 ND ND ND Cs-137 ND ND ND Cs-138 ND ND ND l Activity: February 25, 1989 (Time: 1450)

'A' 'B' 'C' Nuclide uCi/ml uCi/a1 uCi/ml H-3 Note 1 Note 1 1.66E-3 !

Na-24 ND ND 2.64E-7 !

Rh-106 ND ND 1.63E-6 I-131 ND 8.25E-7 2.22E-7 I-132 ND 2.55E-7 1.41E-6 '

I-133 ND 5.44E-7 9.87E-7 I-134 ND ND 1.82E-6 I-135 ND 4.75E-7 2.33E-6 Cs-134 ND 7.10E-8 ND Cs-13't ND 1.21E-7 ND l

Cs-138 ND ND 2.32E-i l NOTE 1: H-3 concentrations are conservatively assumed to be equal to the measured H-3 concentration of 'C' Steam

' Generator sample of February 25.

)

ND: Indicates these nuclides were not detected in the sample analyzed.

1 i

TAB 1E-III-5' CONDENSER AIR EJECTOR SAMPLE ACTIVITY- q Sample Date and Time: February 25, 1989 at 14:55 Nuclide uCi/ml % of Total Ar-41 1.28E-5 0.09 Kr-85m 2.98E-65 0.21 Kr-87 2.29E-5 0.20 I Kr-88 4.83E-5 0.34 Xe-133m 2.05E-4 '1.45 ;

Xe-133 6.98E-3 49.50 ]

Xe-135m 4.46E-4 3.16 Xe-135 6.23E-3 44.17 Xe-138 1.25E-4 0.89 TOTALS 1.41E-3 100.00 l

I E

i l

1

I 1

1 i

TABLE III-6 l REACTOR COOLANT SAMPLE DATA

. Sample Date and Time for Noble Cases: February 20, 1980 at 01:05.

Sample Date and Time for other: February 24, 1989 at 00:26.

Nuclide uCi/a1 Ar-41 4.24E-2 Kr-85m 1.59E-2 Kr-87 1.85E-2 Kr-88 3.41E-1 Xe-133m 2.74E-5 Xe-133 8.79E-2 Xe-135m 1.33E-2 Xe-135 9.57E-2 Na-24 5.06E-3 Rh-106 3.81E-2 I-131 3.91E-3 !

I-132 3.69E-2 I-133 1.86E-2 I I-134 6.80E-2 l

I-135 3.60E-2 Cs-134 LLD l Cs-137 LLD 1~

Cs-138 1.41E-1 l

l-1

) ,

... ... ... (,.~ ~~. i 8 '

- J .

I i

i c FIGUltE III-l 2 i j i I I f -.

l g

i i

i m . ii.. ii.

(

. ; ._._p. ... . . g . . a.

( ,p i j % '

( ..,

i 1 i . :

J, l , I I q q . b l

di.. .... ii.. . .. . _ _ , .

.. . j t.. p .

,t .,

...a .

3 e4 l l l E' _.j. -i q j - 1'. ;

o l l S ci. .l-

' 1

., ,, s

. .. ii.. -

... .l.. a

, I, l ..

e l J '

ir.

- , f

k i

-y .

1 O .... i.. i .. .. _... ..- . .

y .

. .,4 8 , , ,

l 1 .., ,

@ j J_ ., .

i s

I ! j L l f e

t.:..

.. ei.. ii.. .i , ... .. s q

.i

, s, i s

3 .h,..

l-' : ,

l T

tts e .. . ..

m ,-

V 4 . .

l a .l s ,

m n.

)' j- i

.t ..

4

,4 i,

d i ,

i 5, 4 4 .. s i e ,

.i. . . .s. s 3, ,,. s ... i 7 ...

i -

~ j g '

l__ j . t. .

H i- \

H j t i 1 '

s <

r 3 j

_.4 . } ., m , l . g t.. .i s 3 l , 1 l T I c 1 m i- , -.I..- .. 1 , i 3 i .. < j l 1, h ! . ,. r s t .( i I

g i .i l

.t. . . i. ._.... M. ,i. '

c. . l i, J H I 5 i l

( i u ,

j i - .. J-l l. L i E - i- j '- ' l { l ! (=~t s ... ... ... .f. / ... .. r I } .. r- , 3 l l , ,1 l . . ~ , f ; I l

. i

. ' .I ' . . . . ~. .a 4 l r ! ,1 $g I g....I.. . i , e. .j - .. ii...i. .. ... \ r ., e'- 4 -~ 5 -l 7 .=n j - { . ___.._l = , ,tl'. It*. i t ., .i .. s.a g ;. -- .-- _.- 1 I, .i i l 4__ ' { .r , a FIGURE III-l (CONT'D.) , r_ _ _y ( I ' . i } i t . S. ' 1, d '... ~. l ,.. .i . - . ,i . ! p i I }* ! , 0 4 f L.__J_.- l l ' { ! ity %. i ! ) d .p_u>... .. . . g , i - q 1 I ' ' ' t l ..; i } l h l . < ._)[ ! , ! - i l < l l I f i i . { , ,3. ,3 .. ... .. i I , i + l I i d y ! I i a: - t i O g, .. ,,(.. , ,h.. n 3 . ,;,._ w\ ...

t. . . , u .,

O , M , } i < ' ..[ l l ) [~ , l g i @ 11 i k i

; 1 ci.. ii..

u 1- ! i i i . . . ip>r. ... .. . l} , ,- , p F ! I l l ' ' I I g,

, ( 8 H

si.. ...._ i,.. i eij,. <> . . . 1, t.. 6 s $j . l_ j b ,j! % ' **a t;; ' f

l *'

b I {, l I l to ,--)-- / ; i , ; I P, . m i i , i I i } d t - l I,_ i. ti.. e t'. . ^ P4 , p I *1 11.. ._ i. 4_ y. + -l l 6 ? i . . 0 ! * ) 1 ! 'i p _ J, . . i l 1 ( i { i e

l g ! I I 4

l o s .... .... ,,.. ., s. , -. . .p--- . ... I . . . !. , , 4 j , 8 $., "d -,l- < l' l : l , j l , 1 .j- . Ee R' - - - CO ! l i i g i - 'l ' ' 3, - ! ' 1 M ! ! m g~j} t .#.. s .' . [ i y .'.. ca. i l

{

,j j ~' f j j g i i' i.. l ' I 9. ' I ' ._ L . m , 1 I i: 9.. l 6.. ) i ! '. i ! .*. t . k.. . .. t . ' ' i e, , .- q .- I ~, - ' f g . . . _ . . . . ~ . i I o 2 ' ' o . .i . J D . - . **' '** ' O> ' 3 I ' 1' .b.,** m

j i , l

__. .. \' . . . _ _ ... . ' . . ... . . q. ... . . . , l . ( ' ~. . - - .. 't 4 i l I j. g.. 64 ..e, -Cd a l' - { I 9 ! t h' 1 [ i u 1 {;. f r.. ,.. s.. ... }. ., 48' (; L -l I a .q M e I l .' . l # f i i. i I g 1 , f< ,,.. ,,.. ,,.. .,l. 7 1 f 1 IV. EMERGENCY RESPONSE A. ' Emergency Plan Implementation j

1. . Staff and Responsibility .

] Specific assignments to the emergency organization were made in accordance ] with the Station Emergency Plan and were adequate to respond to the event. - Timely activation, staffing, and operation of. Emergency Response Facilities ! i was accomplished.

2. . Emergency Response Support and Resources Support resources (i.e. Westinghouse) were requested and effectively used !

I during the event. State resources that participated at the LEOF were-- adequately accommodated. Response agencies demonstrated their ability to augment the station response.

3. Emergency Classification The emergency classification and action level scheme, as provided in the 4

i Emergency Plan and implementing procedures, were effective for the "

l classification of the event. Emergency Plan and implementing procedures effectively provide for initial and continuing assessment and response actions during the emergency.

4. Notification Methods and Procedures ,

1 Procedures established for notification of federal, state, and local , l \ response organizations and emergency personnel were promptly and effectively implemented. Content of initial and follow-up messages was adequate and consistent with the requirements of the Emergency Plan. Furthermore, provisions in place to alert Virginia Power's augmented I 4 emergency response organization were executed and found to be effective for contacting primary and alternate responders. The response times for some of the positions were not in accordance with the planned time frames. This has been identified in the critique process as an improvement item, i Notifications to the NRC were initiated at 1556 hours0.018 days 0.432 hours 0.00257 weeks 5.92058e-4 months and were continuously ; I maintained until 2133 hours0.0247 days 0.593 hours 0.00353 weeks 8.116065e-4 months , when the NRC terminated the requirement for continuous communication. The termination notification was made to the NRC at 2222 hours0.0257 days 0.617 hours 0.00367 weeks 8.45471e-4 months . State and local notifications were performed according to the following schedule: MESSAGE # TIME 1 1534 2 1604 3 1643 4 1713 5 1745 6 1812 0.425 hours 0.00253 weeks 5.82165e-4 months . ) i ) l } 47 - l i l Corporate call-out was initiated at 1542 hours0.0178 days 0.428 hours 0.00255 weeks 5.86731e-4 months . Station accountability was initiated at 1530 hours0.0177 days 0.425 hours 0.00253 weeks 5.82165e-4 months and completed at 1617 hours0.0187 days 0.449 hours 0.00267 weeks 6.152685e-4 months . All personnel were accounted for by 1650 hours0.0191 days 0.458 hours 0.00273 weeks 6.27825e-4 months . This delay in accountability has been identified for review and improvement to expedite the accountability process.

5. Emergency Communicate;ns Communications between emergency facilities, principal response organizations and emergency personnel were promptly established and continuously maintained throughout the event. Necessary equipment and procedures were in place in order to provide adequate communications. No significant equipment malfunctions occurred.

6. Media and Public Information Information concerning the emergency was made available to the media for dissemination to the public. Four press releases were issued during the event (at 1740, 1845, 2015, and 2240 hours0.0259 days 0.622 hours 0.0037 weeks 8.5232e-4 months , respectively). In addition, press briefings were conducted at the Joint Public Information Center and

) Local Media Center. The above facilities were activated at 1705 and 1800 hours0.0208 days 0.5 hours 0.00298 weeks 6.849e-4 months , respectively. The Rumor Control Network was established and successfully used throughout the event. l l r

7. Emergency Facilities and Equipment i

Emergency facilities were activated and adequately staffed'during the event and the facilities were sufficiently equipped to support. tne response effort. The activation times for the Emergency Response Facilities are-provided below: ) Technical Support Center - 1615 hours0.0187 days 0.449 hours 0.00267 weeks 6.145075e-4 months Operational Support Center - 1603 hours0.0186 days 0.445 hours 0.00265 weeks 6.099415e-4 months Local Emergency Operations Facility - 1730 hours0.02 days 0.481 hours 0.00286 weeks 6.58265e-4 months Corporate Emergency Response Center - 1700 hours0.0197 days 0.472 hours 0.00281 weeks 6.4685e-4 months Joint Public Information Center - 1705 hours0.0197 days 0.474 hours 0.00282 weeks 6.487525e-4 months Local Media Center - 1800 hours0.0208 days 0.5 hours 0.00298 weeks 6.849e-4 months It should be noted that the. Technical Support Center (TSC) was fully manned prior to 1615 hours0.0187 days 0.449 hours 0.00267 weeks 6.145075e-4 months . However, the Station Emergency manager did not relocate from the Control Room to the TSC until 1615 hours0.0187 days 0.449 hours 0.00267 weeks 6.145075e-4 months and, therefore, i' the TSC was not " officially" activated until that time. L The emergency response computer system was used extensively by all the emergency response centers. The addition of dedicated computer operators in each of the facilities to :-t as advisors'provided the necessary expertise to use the computer's full capabilities.

8. Accident Assessment Methods, systems, and equipment for assessing and monitoring actual onsite i consequences of the event were adequate. Assessment and response h activities included both an engineering assessment of plant status and an J assessment of radiological hazards. In-plant and onsite monitoring teams were promptly dispatched and effectively used throughout the event.

i-k _ _-_ _ _-_ _ ___ __ _ ___ __ a

9. Recovery Planning

) Prior to termination of the event, the fomation of a Recovery Organization j i was initiated. Upon event termination, this team met and was tasked with { l developing specific plans and procedures for recovery. The Recovery Organization was established under the direction of the Vice President - Power Engineering Services and Manager - Nuclear Operations Support, and was directed to initiate the following activities:

a. Monitor and ensure stable plant conditions were maintained through the following shift.

b. Review and verify Health Physics dose assessment calculations.

c. Initiate post-trip review and root cause analysis,

d. Formulate both short and long-term response actions,

e. Prepare briefings for State and Local governmental authorities.

f. Establish industry interface for dissemination of event information.

g. Conduct an event critique to encompass all phases of the emergency response.

B. Summary The February 25, 1989 steam generator tube leak event demonstrated that emergency response personnel within the Virginia Power organization are capable of satisfactorily responding to an emergency of this nature. Furthermore, the NRC and emergency organizations within the Commonwealth of Virginia demonstrated the capability to respond to requests for assistance [ during such an emergency. In conclusion, the North Anna Power Station and ) those emergency response agencies that supported it effectively mitigated the transient event and adequately protected the health and safety of the ) public. } l l _ - - _ - _ _ _ _ _ _ _ _ _ - - _ _ J V. SAFETY EVALUATION A, Safety Assessment Comparison to UFSAR Analysis The initiating occurrence for the February 25, 1989 transient was a partial I loss .of feedwater (i.e. to one (1) steam generator) at reduced power. Loss of feedwater is sn ANS Condition II (moderate frequency) event analyzed in Section 15.2.8 of. the UFSAR. The analysis demonstrates that the transient does not violate any limits which could potentially compromise either fuel integrity (i.e. DNB limits not violated) or Reactor Coolant System (RCS) integrity (i.e. RCS faulted stress pressure limits not exceeded and liquid discharge through the pressurizer relief and/or safety valves does not occur). The second phase of the February 25, 1989 event involved a steam generator tube leak, which is bounded by the analysis of steam generator tube rupture, which is discussed in Section 15.4.3 of the UFSAR. The most recent analysis of l. tube rupture was submitted on September 15, 1987 as part of the steam generator downcomer modifications approval request. The discussion below presents a l comparison of each phase of the event (feedwater flow reduction and tube leak) ~ l with the corresponding safety analyses. The actual feedwater flow reduction event was much less severe than the , l l event examined in the UFSAR. Significant areas where the UFSAR analysis bounds the event include the following: 0 ) 1)- Feedwater flow reduction: the February 25, 1989 event involved loss of j l main feedwater flow to a single steam generator, compared to the UFSAR assumption of loss of all normal feedwater. I i _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ __ _ __ J l

2) Initial power level: the UFSAR analysis assumes an initial power level of 102% of rated. Due to the coastdown operation in progress, the actual initial power was about 76% of rated.

3) Source of protection: the UFSAR analysis assumes the anticipatory trip on low steam generator level coincident with steam / feed flow mismatch does not function, and reactor trip occurs on a conservative low-low steam generator level trip setpoint of 0% of narrow range span. In the actual event, trip occurred on low steam generator level (about 25% span) coincident with a high steam / feed mismatch.

4) Inherent reactivity feedback: The analysis assumes a conservative beginning of cycle (B0C) moderator temperature coefficient of +6,0x10(-5) i delta-k/k-F. Due to the end of life, low boron conditions during coastdown, the actual MTC was on the order of -3.5x10(-4) delta-k/k-F. The i

large negative MTC makes heatup events such as a feedwater flow reduction inherently self-limiting, which is , reflected in the limited RCS heatup observed prior to the trip (on the order of one (1) degree F).

5) Pressurizer pressure control: the operation of the pressurizer pressure l l

) spray acted in concert with the mild nature of the RCS temperature transient to limit the pressure increase due to the partial loss of feedwater to less than 10 psi. This compares to an increase of about 250 ! psi in the analysis case. Following the reactor trip the operators identified a primary-to-secondary i leak rate which was about an order of magnitude below the UFSAR tube rupture and the 1987 Unit i event. Table V-1 presents a summary evaluation of the tube leak portion of the event as compared to the current licensing basis analysis. The current ~j 1 i l l analysis was performed as a result of the 10CFR50.59 review of the downcomer flow resistance plate (DFRP) modification following the 1987 tube rupture .. event. As'can be seen, the February'25, 1989 event.was bounded in severity by ) I the analysis for all significant parameters. While operator response times to d identify and isolate the leak were longer than assumed in the analysis, this is I readily attributable to much smaller leak rates occurring in the actual event. ' For perspective, comparisons of total primary-to-secondary mass transfer prior .to the various isolation actions are presented. Nots also that the tube uncovery time (conservatively estimated f rom narrow range level data) was slightly longer than assumed in the dose analysis; again this is attributable to the low actual leak rate. Note that the integrated break flow during the estimated uncovery time was about an order of magnitude below that assumed in the dose calculation, i B. Core Integrity Evaluation , o : During the entire event, the plant was not in any condition which could i- have compromised the integrity of the fuel. The core thermal limits 'were not ! ' violated at any time during the event. In fact, the closest approach to core thermal limits was during initial operation prior to the steam generator low level reactor trip (only a slight increase in RCS average temperature resulted i i from the loss of feed flow to Steam' Generator 'C'). Due to the reduced core l \ l power resulting from coastdown operation, substantial margins to the thermal ! limits were being maintained. I l p L i TABLE V-1 NORTH ANNA UNIT 1 STEAM GENERATOR TUBE LEAK COMPARISON TO LICENSING ANALYSIS Consideration February 25, 1989 Licensing Analysis Event Assumption or Result

1. Tube rupture area Approximately 10-15% of a Double-ended rupture full double ended rupture, of a single tube eg,1

2. Initial break flow Approx. 60-70 gpm (based on Approx. 710 gpm charging-letdown balance) (UFSAR Fig. 15.4-32)

I

3. Integrated primary-to- Prior to: Calculated:

j secondary mass transfer - Turbine Driven AFW Pump 115,000 lbm Isolation 14,000 lbm Used for dose - Air Ejector Exhaust calculation: divert to containment 132,000 lbm 27,000 lbm - Steam Gen. C Isolation j 40,000 lbm - Break flow termination 47,000 lbm

4. Initial coolant 0.0204E-6 Ci/gm 1% failed fuel = l activity (Dose (Measured 2/24/89) 4.0E-6 Ci/gm Eq. 1-131)

) , } k i.f, . I N I-TABLE V-1 (CONT'D.) NORTH ANNA UNIT 1 STEAM GENERATOR TUBE LEAK r COMPARISON TO LICENSING ANALYSIS I )L r k - Consideration February 25, 1989 Licensing Analysis [; Event Assumption or Result h t-b

5. Release path for Condenser air ejectors / Main steam safety

[ radionuclides AFW turbine exhaust valves to environment . 6. Operator-response 114 minutes /47,000 lbm 30 minutes /132,000 lbm V time to identify (based on avg. flow of [ accident type and 70 gpm) terminate break . flow to faulted generator / total primary / secondary . mass transfer !

7. Steam generator 11.4 minutes /4700 lbm 10 minutes /44,000 lbm tube uncovery (conservative estimate duration / integrated based on loss and i

break flow with recovery of narrow 1 tubes uncovered range level indication) j i l l TABLE V-1 (CONT'D.) NORTH ANNA UNIT 1 STEAM GENERATOR TUBE LEAK COMPARISON TO LICENSING ANALYSIS l } \ I Consideration February 25, 1989 Licensing Analysis Event Assumption or Result

8. Offsite Power Available Unavailable 1

9. Site boundary dose

- whole body

0.297 REM **

i - thyroid

26.7 REM ***

L

No dose rate above background actually measured

- Assumes 1* failed fuel /no iodine spike

- - Assumes pre-accident iodine spike i

l I _ 3s _ VI. STEAM GENERATOR 3 VALUATION A. Operating Experience

1. Unit 1 Steam Generator Operating Experience Since startup in 1978, Unit 1 Ms been on All Volatile Treatment (AVT) chemistry. During February 1979, powdex resin was inadvertently introduced into the steam generators. During the resin intrusion event, the res'-

decomposed to form sulfuric acid, lowering the steam generator pH to 6.02 and increasing the cation conductivity to 25.0 micro-mhos. Two (2) other less severe intrusion events occurred in July and September of 1979. During the September 1979 refue'.ing outage, tube denting was identified in the hot leg, and a boric acid treatment was initiated. Unit I has operated in accordance with Westinghouse recommendations for use of a boric acid treatment since 1980. Most of the secondary plant feedwater heaters were originally supplied with copper alloy tubing. A program for elimination of copper in the secondary ] plant is currently in progress. 1 Subsequent inspections have confirmed denting at the hot leg tube support plate locations. However, the denting has not re sult e c' in extensive tube i deformation. Recent comparisons of data taken in 1984 and 1987 indicate that the growth in both the number of dented intersections and extent to which > l intersections have been deformed has been essentially arrested. } Tube degradation involving primary side cracking of Row I and 2 U-bends has also been observed (during r.he 1979 inspection on Row I and during the 1987 ) Preventive plugging of Row 1 tubes and stress relief of inspection on Row 2). } - ) i l l I unplugged Row 2 tubes in the three steam generators has been accomplished to address these issues. Unit I was removed from service for refueling outages beginning in December 1980 and again in May 1982. During both outages, all three steam generators j were examined using eddy current in accordance with the facility Technical Specifications. No significant indications were observed. During the May, 1982 refueling outage, partial tube end repair was performed in Steam Generators 'A' and 'C'. Broken split pins from the upper internals package of the reactor vessel had migrated to these steam generators and damaged the tube ends. This damage only affected the ability to insert an eddy current probe. 1 The integrity of the steam generator and its ability to perform its intended function were tot compromised. The first significant indication of steam generator tube corrosion on North Anna Unit I was in December 1983. Primary to secondary leakage, in Steam Generators 'B' and 'C', increased to a maximum value of 396 gallons per day (gpd). The unit was shut down January 10, 1984, and a visual inspection of both steam generators during a leak test showed a total of five (5) leaking tubes and four (4) leaking explosive plugs between the two steam generators. A total of 579 tubes were inspected in Steam Generator 'B' during the outage. Four (4) tubes were identified with greater than 40% thru-wall indications at > the tube support plates while 13 tubes were identified as having significant distortion. All leaking tubes exhibited eddy current indications, but none of the significantly distorted indications were identified as leakers. In Steam , Generator 'C', a total of 552 tubes were inspected. Four (4) tubes contained I deep eddy current indications and were plugged. Also, one (1) tube was plugged preventively. Additionally, two (2) tubes were identified has having thru-wall ) i indications. The unit returned to service in February and operated until the scheduled May 1984 refueling outage. T average was increased from 580.3 degrees F to 582.8 degrees F in March 1983 and to 587.8 degr-aes F in April 1984. Steam generator activities during the May 1984 refueling outage included complete eddy current inspection in all three (3) steam generators and an attempted tube removal effort. A total of eight (8) tubes in Steam Generator ' A' , one (1) tube in Steam Generator 'B', and four (4) tubes in Steam Generator 'C' exhibited eddy current indications greater than 40% t h ru-wall . Additionally, two (2) tubes in Steam Generator 'A' and cne (1) tube in Steam Generator 'C' were plugged as a preventive measure. Complex signal distortions arising from denting, copper, and magnetite were also observed during this outape, but not recorded. Two (2) attempts were made during the outage to remove a tube from Steam Generator 'A', but both were unsuccessful. The unit returned to service in September 1984 and operated until August 2, 1985 when the unit was shutdown again due to primary to secondary leakage. Primary to secondary leakage was first detected in February 1985 in trace amounts. Noticeable step changes occurred in April 1985. The leakage gradually increased to a maximum value of 213 gpd in Steam Generator 'A' in j late July. The unit came off-line on August 2, 1985, and an inspection of Steam Generator 'A' was performed. A video inspection of the tubesheet was performed while the steam generators were filled and pressurized. A total of l three (3) tubes were identified as leaking. Subsequently, a total of 830 tubes were eddy current tested, twelve (12) of which had pluggable indications l (eleven (11) of these were greater than 90% thru-wall) . Tubes with distorted indications were also identified in this inspection, but none were plugged. Thirteen (13) tubes were removed from service, including one (1) tube which was preventively plugged. ] The unit returned to service in mid-August, 1985 with trace leakage in 1 Steam Generators 'B' and 'C', The leakage suddenly increased after approximately five days on line to approximately 90 gpd where it remained until i the Nove:dber 1985 refueling outage. Steam generator activities performed i during the November 1985 outage included complete eddy current of all three (3) steam generators, and the removal of two (2) tubes containing a total of four (4) support plate intersections from Steam Generator 'C'. The eddy current testing program resulted in the plugging of 43 tubes for indications greater than Technical Specifications limits. Thirty (30) additional tubes containing strong distorted indications were also removed f rom service for preventive purposes. The unit returned to service in January 1986 with no primary to secondary leakage. Trace, intermittent leakage was detected in Steam Generator 'A' beginning in February 1986. During the November 1985 refueling outage, of North Anna Unit 1, eddy current inspection results at the hot leg support plate intersections of all three (3) steam generators revealed tubes with eddy current signals in which the interpretation of the extent of tube wall penetration was precluded due to the level of signal distortion caused by tube denting. The eddy current analysis at that time concluded that the signal characteristics were representative of tube wall degradation but because of signal distortion, depth of penetration could not be quantified. Subsequently, in an effort to characterize the type and extent of tube wall degradation in the steam generators, Virginia Power had two (2) tubes extracted from Steam Generator 'C' and metallurgically examined. The tube examination program conducted included detailed physical measurements of the tubes at the support plate intersection j l locatio: 3, characterization of the tube metallurgical structure, chemical 1 )i analysis of the tube surface deposits, and corrosion testing of the tubing. As a result of the tube examination program, the tube wall degradation has been characterized as outside diameter (0D) and inner diameter (ID) initiated stress corrosion cracking (SCC) located within and extending beyond the tube support plate thickness in conjunction with tube denting. ) Unit 1 was returned to service following the November 1985 refueling outage with only a small amount of primary-to-secondary leakage detected during operation near the end of cycle. A core uprate was performed in October 1986, on Unit I which increased the rated thermal power from 2775 MWt to 2893 MWt i which reduced T average to 586.8*F. The current bounding design values for T H I and T therefore are 621.2*F and 552.3*F respectively, at a total vessel flow C of 278,400 gpm. The unit was shutdown for refueling in April 1967. During the outage, an extensive 100% eddy current program was performed employing the most advanced eddy current techniques available. In addition, a tube strcas relief l demonstration program was conducted at the support plate and all available Row 2 U-bends were stress relieved. Indications were found at the tube support plate, the antivibration bars in the area of the steam generator U-bends, and i L the tubesheet. A total of 263 tubes were plugged during the outage. In addition, two (2) tubes were removed from Steam Generator 'A' for laboratory 1 examination of the tubesheet indications. The removed tubes, as well as, the demonstration program for tube stress relief at the tube support plates were evaluated. (LATER - Results) i Because greater than one (1) percent of the tubes in the initial tube inspection sample group, in the April 1987 outage, were found to be defective, NRC approval of the actions taken to return the steam generators to an operable status was required before returning Unit 1 to service. A meeting was held I t l j with the NRC on June 3, 1987 to discuss-the results of the steam generator tube inspections and the Virginia Power plans to evaluate the two (2) tubes that I were removed from Steam Generator 'A'. Virginia Power also committed at this l meeting to impose a 100 gallon per day limit on primary to seconde.ry leakage l for the duration of the upcoming fuel cycle. NRC approval for restart of Unit j 1 was given at this meeting. l On July 15, 1987 after being in service for 16 days following the sixth j . l ) refueling outage, the Unit 1 Steam Generator 'C' experienced a ruptured tube. Evaluation of the event was documented in the North Anna Unit 1 July 15, 1987 ) Steam Generator Tube Rupture Event Report Revision 2, dated February 12, 1988. ) The cause of the tube ruyture was established to be fatigue. Corrective action ! included installation of a downcomer flow resistance ring, preventive plugging of potentially susceptible Rows 8, 9, 10, and 11 tubes, and implementation of an augmented leakage surveillance program which included expansion of the steam I generator inspection program and installation of an N-16 detector on each of i the main steam lines. Following the repairs, the unit was returned to service i in October 1987. During the last cycle, the unit was continuously on line for a period of 198 days prior to the unit trip on February 25, 1989. In February 1989, Steam Generator 'B' primary to secondary leakage increased slightly to i approximately two (2) gpd and Steam Generators 'A' and 'C' were less than one (1) gpd. L A historical summary of the numbers of tubes plugged in each steam generator by outage date is given in Table VI-1.

2. Unit 2 Steam Generator Operating Experience The Unit 2 steam generators were placed in service in August 1980 and have

! operated for six fuel cycles. An All Volatile Treatment (AVT) of the secondary l side water was used from initial operation. Most of the secondary plant feedwater heaters were originally supplied with copper alloy tubing. A program for elimination of copper in the secondary plant is currently in progress. ] The unit experienced a powdex resin intrusion from the condensate polishers in August 1981, after which the unit was shut down and drained. In February s 1984, the unit ev.perienced a sulfuric acid intr"sion from regeneration of the i makeup u *r system after which the unit was again shut down and drained. The first operating cycle was completed March 1982. During steam generator eddy current inspections minor denting was discovered with most of it occurring in Steam Generator 'C'. As a result, a program of high concentration boric acid soaks (50 ppm) has been used since the beginning of the second cycle with continuous on line addition of 5-10 ppm boric acid ever since. Unit 2 was placed in service for the second operating cycle in June 1982 and was operated until April 1983. The third cycle began in May 1983 and was completed in August 1984. During this cycle, T average was increased from 580.3 *F to 582.8 'F. As a preventive measure, one (1) tube was plugged due to cold leg thinning and two (2) due to the presence of foreign objects in the secondary side. Cycle 4 ran from November 1984 to February 1986. During this cycle, T average was increased to 587.8 *F. Eddy current inspection was performed in all three (3) steam generators. Based on the eddy current inspections, a total i of nine (9) tubes were plugged (6 tubes in Steam Generator 'C'). Of these, eight (8) tubes were plugged due to PWSCC and one (1) tube due to cold leg i thinnin;. This is the first time tube plugging was required due to PWSCC and h the second occurrence of cold leg peripheral tube thinning at the tube support plate. No new dents were identified but some slight increase in dent size was noted. i 1 The fifth fuel cycle ran from April 1986 through August 1987. A core uprate was performed during September 1986 which increased rated thermal power from 2775 MWt to 2693 MWt. This core uprate reduced T average to 586.8'F. The current bounding design values for T and T are 621.2*F and 552.3*F, H C respectively, at a total vessel flow of 278,400 gpm. As a result of the Unit 1 l I steam generator tube rupture event, in July 1987, modifications required for Unit I were also made to Unit 2 as a preventive measure. Included in the modifications was the installation of 118 sentinel plugs in tubes considered susceptible to fatigue failure. The initial leak inspection discovered a number of leaking explosive plugs and one leaking mechanical plug. The mechanical plug was removed and replaced with a new mechanical plug (Figure VI-1). Westinghouse stated in a report, dated March 1989, that upon laboratory analysis the mechnical plug was determined to have circumferential cracks above the expander attributed to PWSCC. The cracks were above the expander between the eighth and ninth lands which is the same location of the failure in the Unit 1 plug. The failed mechanical plug was in Row 3, Column 6 on the hot leg side of Steam Generator 'A'. The plug was fabricated from material heat NX 3962. This is the same heat which was associated with the Unit 1, February 1989 leak event. As a result, of the eddy current inspections, 37 tubes were plugged. Of these, plugging was required due to PWSCC (5 tubes), tight radius U-bend PWSCC (5 tubes), cold leg thinning (16 tubes), AVB wear (7 tubes), and outside diameter PWSCC (1 tube). One (1) additional tube was conservatively L I plugged due to a free span indication observed in previous inspections. Two (2) tubes were plugged because of the presence of a foreign object. The > majority of the 37 tubes that were plugged had not been inspected during the . I previous outage and the actual degradation may have been gradual, particularly j i for cold-leg thinning and AVB wear indications. No progression of denting was , detected. Unit 2 returned to service in November 1987 for the sixth fuel cycle. The unit completed the cycle with 372 days of continuous operation. The current I refueling outage started on February 20, 1989. (LATER-results of current EC inspections). Unit 2 has a history of minimal primary to secondary leakage. A historical summary of the number of tubes plugged with mechanical plugs in each steam generator by outage date is provided in Table VI-2. L l > 1 l l i ! l i l E. Failed Tube / Plug Identificatic a and Examination

1. Steam Generator Cooldown, Flush, and Purge l

A recovery organization was established following the Unit I steam generator tube leak event on February 25, 1989, to direct the s >am generator recovery process. The recovery organization also evaluated actions required to be taken in the Unit 1 Steam Generators 'A' and 'B' and Unit 2 steam generators. After the event, the Unit 1 Steam Generator 'C' contained a mixture of j 1 primary and secondary coolant. The concentration of boric acid was j approximately 34 ppm. This sample was based upon surface samples taken on March 1, 1989 above the feed ring. When the secondary side was vented, hydrogen was not detected. The Reactor Coolant Loop 'C' was isolated on the primary side by closing the loop stop valves. Over the following days, gases were vented through the waste gas processing system to the process vent system. The recovery process was carefully contro11cd in accordance with approved procedures. In parallel, the entire reactor coolant system was established in a cold shutdown condition and degassed with Steam Generators 'A' and 'B' placed in a wet layup condition. In order to make Steam Generator 'C' accessible for evaluation while maintaining tight chemistry controls, the steam generator was systematically vented, drained, and refilled under a nitrogen purge until the level of boron in the secondary side was less than one (1) ppm. Acceptable boron levels were i i achieved during the second drain down. The analysis of the chemistry sample of the second drain down is shown on Table VI-3. The approximate location of the tube leak was then de rmined while draining of the secondary side to the primary through the tube leak. Drainage continued by gravity until ide level j _ _ i reached 58 percent wide range (WR). This level placed the leak location near the seventh and uppermost tube support plate. Upon achieving the required boron concentrate n limit, the steam generator was filled to the 58 percent wide range level with a nitrogen blanket. This fill was performed to provide additional shielding (ALARA practice) for workers in the containment. l l TABLE VI NORTH ANNA UNIT 1 TUBE PLUGGING

SUMMARY

TOTAL' OUTAGE DATE STEAM GENERATOR TUBES-

'A' ' _B' 'C' September 1979 94 94. 96 284 January- 1984 0 4 5 3 May 1984 10 1 5 16

' August- 1985 13 0- 0 13 November 1985 9 17 47 73 April 1987 83 62 118 263 l

July 1987 105 39 -109 253 i

February 1989 l l

(Later) l-TOTAL ----- (Lat e r)

Z (Later).

i f

}

t I

l TABLE VI-2 NORTH ANNA -

UNIT 2 L-I- TUBE PLUGGING

SUMMARY

-(MECHANICAL PLUCS)

' PLUGS STEAM GENERATOR l-OUTAGE INSTALLED HEAT 'A' 'B' 'C TOTAL 8/87 Hot Leg NX 4523 40 7 4 52 NX 3513 14 - -

14 L

l NX 3962 -

28 61 89 155 l

Cold Leg NX 4523 9 2 5 16 l

[

i NX 3962 -- 4 17 -

21 37 Cold Leg NX 4523 11 27 27 65 (Sentinel NX 3962 7 2 17 26 l

Plugs) NX 3513 ,

27 - -

27 118 3/86 Hot Leg NX 3962 2 1 6 9 l Cold Leg NX 3962 2 1 6 9_

18 3/84 Hot Leg NX 2387 1 - 2 3 i j

Cold Leg NX 2387 1 --

2 3_

6 l I l

' i' j

TABLE VI-3 UNIT 1 STEAM GENERATOR 'C' CHEMISTRY' SAMPLE

~AFTER SECOND DRAIN DOWN 9

I I

I l

l pH 7.56 l Conductivity 4.10 Sodium, Na 145 ppb Chlorides, C1 12 ppb Fluorides, F </= 1 ppb Sulfate, SO 176 ppb 4

Boron, 'B'- </=.1 ppm l

l

}-

[ !

,l i

1 s.

2. Identification of the Leaking Tube and Plug With the primary side now accessible, the location of the leak was determined by the draining of water from the failed tube. The tube was made to leak by increasing the secondary water level to slightly above the 58% level (WR). The leakage was located on the hot leg side in tube Row 3, Column 60 (R3C60). The leakage was observed as a continuous steady stream of water.

Estimates of a leak rate of eight (8) gpm under static head had been established. This is based upon the rate at which the steam generator had drained during the previous fill and drain activ 2y.

After identifying R3C60 as the tube causing the event, the tube was temporarily plugged and the standard pre-eddy current leak test was performed.

This is a procedure performed by Virginia Power prior to the start of eddy current examination to identify any leaking tubes.

Procedurally, the secondary side of the steam generator is pressurized with water to approximately 200 psi and a video scan is made of the tubesheet from the p rima.cy side. Tube R24C8 was served to have dripped one drop of water.

The origin of the drop could not be determined with certainty (i.e., may have j been residual water off the tubesheet or from the tube); therefore, it was decided to include tube R24C8 in the study.

With these locations established, previous current examinations were reviewed. Tube R3C60 had been plugged in the Fall of 1985 for a reported 65%

thru-wall indication at the first support plate on the hot leg side.

Determination of the thru-wall percentage was based upon the convention of eddy current signal analysis existing in 1985. In 1987, the rules for analysis were changed to better describe the types of indications being found at North Anna.

In accordance with the instructions developed in 1987, the bobbin coil data

from this current outage was interpreted as a distorted indication (DI).

Similarly, the R3C60 data from 1985, interpreted with the 1987 rules, would also be called a DI.

Tube R24C8 had been plugged in the Spring of 1987 for as a result of an indication above the tubesheet on the hot leg side. This indication had been identified as a possible indication (PI) by the rotating pancake coil eddy current probe in 1987. Data from this current outage revealed no change in the evaluation of this indication.

3. Description of the Westinghouse Mechanical Plug As stated, both tubes R3C60 and R24C8 suspected of leaking had been plugged during past. outages. These plugs are designed and installed to maintain the pressure boundary. The fact that at least one (1) plug was leaking heavily, caused the recovery effort to include this type of plug in its review.

[

j a.c.e All Alloy 600 material orders for plugs, since June 1988, have specified a tighter carbon content control and a mill anneal temperature sufficiently high to ultimately allow adequate carbide distribution at the grain boundaries. A I

I continuous chromium-carbide distribution at the grain boundaries along with the microstructure, resulting from the Westinghouse thermal treatment, is essential for maximum resistance of Alloy 600 to PWSCC.

Captured inside the plug is a component known as the expander. It is manufactured from Carpenter Custom 455 (ASTM-A-564). !

I l

)

l l

1 l

9

)

a,C,e 1

1

4. Removal of the Plugs 4 l

An additional investigation of the plugs and the tubes R3C60 and R24C8

! i I could not continue until the plugs were removed. Both plugs were removed from t the cold leg side, in one piece. The threads of the plug R3C60 were stripped, l l

thus causing the use of the spear. Both cold leg plugs in the tube were sent j i

to the Westinghouse R&D laboratory, in Pittsburgh, for analysis.

Removal of the plugs from the hot leg side was not as straight forward. l l

l The plug, in tube R24C8, failed above the threads in the skirt area, due to tensile overload. At a later date, the remainder of the plug was removed by drilling. Both pieces were sent to the Westinghouse R&D laboratory for l

i examination.

1 The plug, in tube R3C60, was not immediately pulled from the tube. First, I it was determined by probing with a stainless steel rod that: (1) the top 'of the plug was no longer in place and had moved beyond the top of the tubesheet, and (2) the expander had travelled approximately [ ]*' during the original installation of the glug. Also, a video scan was made of the plug remnant as it set in the tube in the post event state to view the fracture surface located above the expander. A quarter (1/4) inch 90' angle Welsh-Allyn video probe was used for this purpose.

After completion of this activity, the plug remnant was removed by the standard pull method. Prior to shipment, to the Westinghouse R&D laboratory, an initial inspection of the fracture surface was made by a Virginia Power l

metallurgist.

4 i

l l

i ii C. Plug l Evaluatiow_ I

1. Field analysis of the R3C60 plug remt. ant ;

The video scan of the plug fracture surface made while the plug was still in place indicated the plug had failed 360* on essentially the same horizontal plane, at an elevation just above the top of the expander, as installed. Some minor areas of 45' tearing were noted running from the surface of the fracture down into the plug's body. The remainder of the plug's surface did not show I any cracking. This includes the expander, the area between the bottom of the expander and the removal threads, and the removal thread area. )

A Virginia Power metallurgist visually examined the plug in the field (Figure VI-l depicts a sketch of a typical mechanical plug). The failed plug exhibited a flat circumferential fracture at the root of the second land. The fracture appeared to be perpendicular to the axis of the plug. The fracture surface was covered with an even undisturbed oxide layer, with no indication of any rubbing or fretting of the surface. There did not appear to be any necking I down of the plug in the vicinity of the fracture nor was there any obvious indication of a shear lip. There was no evidence of a ductile type fracture i

apparent to the naked eye. In addition, there was no apparent cracking ;

transverse to the main fa11ure or no evidence of any significant branching of the main fracture. None of the fracture morphology or orientation was inconsistent with the postulated stress corrosion cracking failure mechanism. I

2. Laboratory Analysis of R3C60 Plug Remnant The Westinghouse R&D laboratory performed an analysis of the plug section 1

removed from tube R3C60 on the hot leg side. Westinghouse reported that the )

actual failure of the plug was due to tensile overload of a very small remaining wall section. This remaining wall section was 360* around the plug 1

1 1

i 1

'thereby causing the plug to sti11' maintain a primary to secondary coolant seal even though the thickness of the wall.section was four (4) mils or less '(1.7 1

mils ' average). Table VI-4 documents the geometry of this sealing area. The design wall thickness of the plug in the area of failure, is approximately [ ,

j a.c.e TABLE VI-4 PLUG HEAD AREA CALCULATION REMAINING WALL AT PAILURE REMAINING ANGLE PLUG ID WALL PLUG OD AREA (DEG) (IN) (MILS) (IN) (IN )

0-20 a,c.e 3 a c.e 0.00034 20-65 4 0.00102 65-140 2 0.00085 140-160 1 0.00011 ;

160-230 0 0.00000 230-290 1 0.00034 290-330 2 0.00045 330-360 3 0.00051 -

i TOTAL AREA 0.00362 l

} - 7e - ,

l l

. _ - - _ _ _ _ _ _____--______-_-___Q

The fracture surface is located between the 8th and 9th land, at the point of reentry of the 9th land into the plug body (The 9th land is the second land f

! from the closed end of the plug). Evident on the fracture surface is extensive j wall reduction, due to primary water stress corrosion cracking (PWSCC).

l Analysis indicates that the corrosion cracking proceeded through the wall on at least six (6) fronts, all located on the same relative horizontal plane. These intergranular cracks linked-up due to tensile overload to form the 360*

macro-crack. The macro-crack has penetrated between 91 to 98% thru-wall, for approximately 320* of the circumference. For the remaining 40*, the PWSCC was nearly 100% thru-wall with only local areas separating the intergranualar crack front from a region of highly cold worked material inclined 45* from the fracture. The highly cold worked material may have been created during or prior to plug expansion. Below the fracture face of the plug (R3C60) remnant, there were ID initiated axial cracks which extended from the top of the 1st land to the bottom of the 4th land. This is where the bottom edge of the plug expander was located, The cracks tended to curl circumferentially near their lower crack tips. One of the larger axial cracks was examined in detail and founded to be 0.145 inches long and 0.025 inches deep; it was intergranular. j

3. Laboratory Results of R24C8 Hot Leg and the Cold Leg Plugs a) R24C8 Hot Leg The plug (R24C8) failed due to tensile overload in the area above the removed threads during the removal process. The fracture surface was I characterized by a 45* shear lip, 360* around. However, the Westinghouse R&D laboratory also identified several small axial cracks at the shear lip interface which are attributed to PWSCC. At least one of these cracks was approaching 50% thru-wall; however, most were approximately 20% thru-wall.

i

The above anflysis was performed on the lower section of the plug. It was necessary to remove the upper section by drilling. A large amount of the plug was consumed by the drilling process; thereby, making conclusive analysis of the plug state prior to drilling, impossible. Approximately, 180 degrees of the wall section was removed from the plug after drilling and the remaining section had undergone substantial ID wall reduction from the drilling.

However, a circumferential stress corrosion of ID origin, 0.6 inches long and estimated to have been 42 mils deep was found below the top most land.

l b) Cold Leg Plugs of Tubes R3C60 and R24C8 l

Laboratory analysis of these cold leg plugs did not identify any evidence of cracking being induced by PWSCC or other service induced mechanism.

4. PWSCC in Alloy 600 Mechanical Plugs Primary water stress corrosion cracking (PWSCC) attacks Alloy 600 by a mechanism which is not yet fully understood. However, it is known that the process is affected by the temperature, time at temperature, stress and the microstructure of the material. Temperature and stress are interrelated factors with time, which vary inversely (e.g. the higher the temperature or stress, the less the time to failure). It also appears from work performed by Westinghouse, each particular microstructure will have its own curve of temperature and time relationships. These curves can vary greatly, in relative time to failure, because of microstructure.

The microstructure is therefore the key to obtaining the maximum resistance to PWSCC in Alloy 600. Two (2) factors affect the attainment of optimum microstructure. They are carbon content and thermal treatment history. Prior to June of 1988. Westinghouse did not place any controls on either the carbon content or the mill anneal process, other than what was specified by the American Society of Mechanical Engineers in their Material Specification SB-166. This was not adequate to ensure that the post manufacturing thermal treatment, of holding the material between [ ]

a,c.e would have its desired effect of improving PWSCC resistance to an acceptable level.

The Westinghouse thermal treatment is designed to cause the growth of a chromium-carbide at the grain boundaries and to allow free chromium to migrate to the resulting chromium depleted region near the grain boundaries. This process requires that sufficient carbon has been returned to solutien by the mill anneal. If the mill anneal temperature is too low, then the carbon will remain tied to chromium-carbides dispersed throughout the grain and not be available to form the carbides needed at the grain boundaries. It was reported that Westinghouse believes many of these heats of material processed prior'to-a,c.e June 1988, received a mill anneal in the range of [ ] A mill anneal in the range of ( ) "' would have been required for optimum resistance to PWSCC.

Since June of 1988, Westinghouse altered its acceptance criteria for the new heats of material by placing tighter controls on carbon content versus mill anneal temperature. They also only accept the final product based upon the attainment of the desired microstructure. This microstructure is defined by Westinghouse as having the grain boundaries essentially completely decorated with the desired chromium carbide.

Westinghouse has also undertaken efforts to document the grain structure which exists in their plugs in service and to determine the relative failure time of these heats at various temperatures. This is based on testing representative heats in both pure and aggressive steam, in mock-ups. By using

) )

the results of these tests, they developed a scaling factor based on the relative time to failure of the samples and kinetic calculations. For example, if " failure time" is defined as the time to obtain (near) thru-wall cracking, a variation of approximately [ ]

is observed between heats of material with the preferred microstructure and heats with the least preferred microstructure. Table VI-5 documents the results of their work and its relationship to North Anna.

I

)

TABLE VI-5.

Relative Time-To-Inititate PWSCC Plug Grain Boundary Scaled Time vs. Temperature *F Heat # Carbide Distribution 618'F (App. T ) 556*F (App. T )

H C NX3962 Less than :1emicontinuous a,b,c a,b c NX3513 Less than semicontinuous NX4523 Semicontinuous NX5222 Semicontinuous NX2387 Continuous NX1989 Continuous i

NX2205 Continuous 1

l NX6323 Continuous NX9789 Continuous 1 .

l 4

l l

Both plugs in R3C60 were manufactured from heat NX3962 and both plugs in 1

I R24C8 vere manufactured from heat NX4523.

1

D. Tube Failure

1. Introduction The failure of the tube R3C60 on February 25, 1989 was the reruit of three conditions:

Plug R3C60 hot leg was no longer able to restrain the primary pressure due to FWSCC.

Tube R3C60 was dry (i.e. free of water).

Force necessary to hold the plug top in place, being supplied by the tubesheet flexure during operation, was released by the event.

2. Plug Interaction with the Tube Sheet The remaining ligament on plug R3C60 Hot Leg was not sufficient to retain the primary pressure for 2250 psia. Based on the measured ligament of the plug remnant (1.7 mil avg.) the material would need to have a tensile st ess I capacity of 227 ksi to contain the primary pressure. The actual materia.

ultimate stress limit is between 80-120 ksi. Therefore, supplemental forces were required to prevent the failure of the piug.

)

These supplemental forces were applied to the plug land (s) above the plane of the PWSCC cracking and came from two sources. Primarily, they result from the pressure differential across the tubesheet during operation. This difference in pressure causes the tubesheet to bow placing the material between the neutral axis of the tubesheet and the primary loop pressure into compression, thus placing additional compressive loads on the land (s). Also, the primary loop pressure within the center of the plug is compressing the land (s) against the tube, thus creating additional compressive loads.

Using a finite element model, Westinghouse determined the value of these

. additional compression loads being applied to a plug during operation. These

loads vary across the tubesheet since displacement due to bowing (flexure) is greater at the center of the tubesheet than at outer points. The loads were applied primarily to the 9th land of the plug which was above tiie fracture surface. Table VI-6 shows the results of this model for a plug at the center of the tubesheet and at the outer part of the tubesheet with an average remaining ligament of 1.7 mils (e.g. R3C60). The plug (R3C60) remnant was located more toward the center of the tubesheet. Four (4) different operational conditions were modelled.

)

l

)

l l

)

TABLE VI-6 COMPARISON OF REQUIRED VERSUS AVAILABLE INTERFACE LOADS (Available Load Less Required Load, lbs.)

Condition Central Outer Steady State a,c Feedline Break Hot Shutdown Feedwater Isolation

"- " Negative Sign Identifies Insufficient Force to Resist Failure Available Load = Fg+Fg+F7 Fg= restraining' force due to remaining uncracked ligament around plug.

Fg = fractional resisting force on the land (s) due to primary

, loop pressure within tb*. center of the plug.

F = Interfacial force on land due to tubesheet bow.

7 Esquired Load = F = pressure force pushing up; i.e. primary j 2 pressure x Fr I

f If Fg+Fg+Fy - F, < 0, plug top will release.

}

3. Failure of Tube R3C60 The February'25, 1989 event wa's a feedwater isolation event which caused an approximate 380 psi pressure reduction across the tubesheet. This pressure drop allowed the plug to be released into a tube having only atmospheric pressure contained within it. The plug top propelled through the tube wall at the tangent point and then struck the next adjacent tube, R4C60, causing i

denting in the wall of that tube. Because of the closeness of tube R4C60, the plug section was unable to leave R3C60 but rather moved along the extrados of R3C60 opening a hole estimated to be three (3) inches in length and 3/4 inches in width. The plug section came to rest in the tube wall. This information was obtained from a video scan using a Welsh-Allyn 3/8" diameter probe, which was inserted into both the hot leg and cold leg sides of. tube R3C60 and into imoved for analysis).

~

the het leg of tube,R4C60 (Note: The tube will not be The resultant tube leak rate was computed to be 74 gallons per minute. The actual leak rate was limited by the cold leg plug and the expander (and plug remnant) remaining - in place within the failed hot leg plug. The expander has an opening of [ "'C

] This size opening is in agreement with a l

7 calculation performed at the time of the event (i.e., 0.250 inch simple opening would leak 77 gpm).

,l

) The expander can be expected to remain in place during these events.

Westinghouse has determined the force of flow through the expander openinj; during an event is approximately 600 lbs. Based on actual field experience, the l load required to dislodge the expander is between [ ]

). a,c on the average.

l

4. Eddy Current Examination of R3C60,'R24C8 and Neighbors An eddy current program was performed on tubes R3C60, R24C8 and their neighbors. The program consisted of a bobbin examination of both the hot legs and cold legs, an 8 x 1 program of the hot legs and a profilometry program of the hot legs. Other than damage caused by the plug section to tubes R3C60 and R4C60 and flaws identified by previous eddy current indications within R3C60 and R24C8, no new indications were identified except for anti-vibration bar wear on tube R24C9. These AVB indications were either 23% or 26% thru-wall.

The dent on tube R4C60 was sufficient to prevent the passage of any probe over the U-bend larger than a 0.580 inch bobbin probe. No evidence of any thru-wall damage to tube R4C60 was found.

E. Steam Generator Inspection

1. Eddy Current Inspection Methods and Scope Prior to the event, a plan was formed to nondestructively examine each steam generator using various eddy current testing (ECT) techniques. Other than the Eddy Current testing stated above, no additional testing was added because of the event. Eddy current testing is the principal method used for performing tube inspections. This inspection method involves the insertion of a test coil inside the tube that traverses the tube length. The test coil is excited by alternating current, which creates a magnetic field that induces eddy currents in the tube wall. Disturbances of the eddy currents caused possibly by flaws in the tube wall will produce corresponding changes in the j electrical impedance as seen at the test coil terminals. Instruments are used to translate these changes in test coil impedance into output voltages which can be monitored by the test operator. The depth of the flaw can be estimated by the observed phase angle response. The test equipment is calibrated using tube specimens containing artificially induced flaws of known depth.

)

It was planned to use (standard) bobbin coil, the 8x1 probe, eddy current profilometry and the rotating pancake coil (RPC) probe. The bobbin coil probe detects ID and OD axial flaws with good resolution. It is used to implement the facility Technical Specification required sampling program. The 8x1 probe locates both ID and OD circumferential indications and ID and OD axial indications. The profilometry probe helps to describe the extent of denting.

Finally, the RFC probe is similar to the 8x1 probe detection capability, but moves at a slower speed. RPC was used to verify signals of interest from the bobbin and 8x1 probes.

The following initial inspection plan was implemented for Unit 2: (" nit i later) i All 3 steam generators

-- standard bobbin coil probe of all tube portions open for inspection

-- 8x1 probe of the hot legs and tha cold legs through the seventh tube support plate for Row 8 thru 12 fto check for fatigue damage)

-- 8xl probe of an additional 20% of hot legs through the first tube support plate hot leg.

-- RPC of selected indications

-- RFC of Row 2 U-bends

) -- profilometry of selected intersections

2. ECT Analyst Qualification 9

A specific steam generator eddy current data analysis program was

! implemented for the inspection plan discussed above. -

program ensured that

)

the evaluation of the eddy current data would be of the highest quality l possible. The program contains five (5) elements which are designed to meet this goal. They are:

1

a). demonstration of the ability to correctly interpret data by each' 4

analyst; l

b)' data evaluation by two analysts working independently (of each other);

c) reporting of any indication on which agreement of interpretation was difficult to obtain; d) written " analysis rules" which provide instructions about interpretation of signals specific to North Anna; and, e) all complex indications are considered relevant until dispositi,oned by RPC or until more definitive rules for bobbin or 8x1 are established.

Analysts evaluations were categorized as follows:

a) for bobbin, clear indication (in percent, % of penetration depth),

' distorted indication (DI), and tubesheet indication (TI); and, b) for 8x1, possible indication (PI)

RPC was used to verify those indications determined to be DI, TI, or PI.

3. ECT Data Evaluation Methods The date acquisition system consists of a Remote Positioner (SM-10 or l

ROSA), MIZ-18 Data Acquisition System and a DDA-4 Display and Evaluation System j to monitor data acquisition.

SM-10 and ROSA are computerized positioners designed to automatically locate and verify the position of steam generator tubing. Their functioning is I verified by two (2) remotaly operated cameras. Benchmarks such as plugged or marked tubes are used to ensure the accuracy of the remote positioner.

The MI2-18 acquisition system is attached to the positioner via a conduit system. Eddy current calibration standards are connected in series with the

! .conduft to facilitate system c.alibration.

The M12-18 functions as a probe pusher and an analog-to-digital converter.

l

-hh-

Analog signals are converted to a digital format within the MIZ-18 for display on a DDA-4 Evaluation System.

Prior to probing tubes, the system is set up and configured. The MIZ-18 then acquires 100 percent of the eddy current data within the configuration parameters.

The DDA-4 Evaluation System is used to monitor data acquisition and to evaluate the results of eddy current testing. During data acquisition the DDA-4 is used to verify the extend of examination and to ensure the eddy current probes are functioning properly.

Because the data are acquired digitally, the data analyst can independently calibrate and mix any of the 16 channels of eddy current information acquired.

Several software packages have been written for DDA-4 which allows the evaluation of standard bobbin, 8x1, and RPC eddy current data.

When data are to be evaluated by the eddy current data analyst, he calibrates the DDA-4 in accordance with a written procedure utilizing the calibration standards recorded on the data. He then reviews each tube. The results of the review are stored on a floppy disk. The two reviews are compared and the differences are arbitrated via a resolution process.

Following resolution, the final results are stored on the "Supertubin" Data Base system.

4. ECT Inspection Results (Unit 2)

The inspection plan discussed earlier was implemented to collect data for 1

) analysis of the current steam generator condition. This data was then analyzed and, based on the plugging criteria discussed later in this report, appropriate

) tubes will be removed from service.

l l

Because of the identification of two indications at the first support l I

plate, in Steam Generator 'C', Unit 2, the 8x1 first support plate program in that generator was extended to 100%. After an additional similar indicativa was found in Unit 2 Steam Generator 'A', the 8x1 to first support plate program was extended to 100% in both the Unit 2 Steam Generators 'A' and 'B'.

Tubes to be removed from service based on the ECT inspection program are listed below.

Unit 2 Steam Generator 'A' Row Column 15 37 33 39 24 45 12 46 Unit 2 Steam Generator 'B'

"~

I 17 12 27 16 39 45 13 48 39 49 I 2 51

~

15 51 2 53 2 57 2 59 I - e0 I_ _ _ __ _

I l

Unit 2 Steam Generator 'C' Row Column

]

l 24 13 21 24 26 24 8 25 32 29 29 32 33 35 5 38 40 42 35 43 17 44 2 45 29 48 3

37 58 2 61

)

5. Steam Generator Tube Plugging Criteria Based on Eddy Current Inspection Plugging criteria have been established to remove tubes from service that 1

f may become unserviceable prior to the next inspection. This action assures that tubes accepted for continued service will retain adequate structural

i i

margins against a gross tube failure' under normal operating and accident conditions. The following plugging criteria were establi.=hed:

- plug all clear indications greater than 40% (This complies with the Facility Technical Specifications.);

- plug all confirmed axial indications on which depth of degradation cannot l be determined and,

- plug all confirmed circumferential identified at the tubesheet.

F. Scope of Repair for Susceptible Mechanical Plugs The previously described evaluation of mechanical plug material heats provided the data necessary to categorize the various heats in terms of operating life expectancy. This classification method resulted in the ;

requirement to replace mechanical plugs fabricated from heats NX3962, NX4523, and NX3513 located in the hot leg tubesheet except those installed in tubes centaining sentinel plugs. This criteria was used in developing the scope of work for both Units 1 and 2.

The Unit 1 scope of work results in the repair of 435 hot leg plugs.

The Unit 2 Scope of Work results in replacement of 51 hot leg plugs.

1. Unit 1 Steam Generator Repairs Because of the extent of repair work required in Unit 1 (435 total hot leg plugs), ALARA considerations warranted the consideration of alternate means of repair for the Unit I work. Virginia Power conceived the approach of installing a plug in the existing mechanical plugs (plug in a plug or " PIP" utilizing the threads already existing in the mechanical. The purpose of the design is to limit the primary-to-secondary leakage which would prevent the I failure of the tube due to plug failure. Use of a PIP would eliminate the work required in the channel head for removing mechanical plugs and therefore would

significantly reduce the exposure levels required to repair the steam generators.

Westinghouse was tasked with evaluating and testing potential designs and recommending the optimum design which could be qualified as permanent repair.

The design required analysis to ensure provisions were included to prevent the P1P. from becoming loose and resulting in a loose part in the primary system.

The analysis also required evaluation and protection against potentially inherent tube failure modes. For instance, primary water buildup in the PIP / mechanical plug interface could potentially act as a propellant that could drive a failed plug cap against the tube creating a tube leak similar to that which resulted from the February 25, 1989 leak event.

(LATER -

description of the PIP design and qualification process:

installation method (s): and pertinent information related to installation mettods, industry experience.)

In addition to eliminating potential failure of the mechanical plugs, it was necessary to stabilize the Unit 1 Steam Generator 'C' tube R3C60 which failed as a result of the February 25, 1989 tube leak event. Tube R3C60 was plugged with nechanical plugs selected from a currently approved material heat.

The top of rae failed plug in tube R3C60 hot leg was left lodged in place in the tube U-bend. Analysis showed that during operation that the plug top would remain in place and, if it did become dislodged, it would take 8 years of operation for an adjacent tube to develop a thru-wall eddy current indication.

As a precaution, the tubes adjacent to Row 3 Column 60 (except R2C60 which is already plugged) will be mechanically plugged on the hot leg and sentinelly f plugged on the cold leg to protect against failure of these tubes. Sentinel plugging will provide the means to monitor the adjacent tubes to determine if any degradation occurs during subsequent operation.

2. Unit 2 Steam Generator Repairs Due to the limited number of Unit 2 mechanical plugs which were included in the scope of plug repair (51 hot leg plugs) and the lack of a qualified PIP design at the time the work was performed, the Unit 2 repair consisted of removal of the suspect plugs and replacement with plugs fabricated from a currently approved plug material heat less susceptible to PWSCC.

The distribution of the mechanical plugs between the three (3) steam.

generators as shown on Tables VI-7 and VI-8 resulted in the decision to use the Westinghouse robotic (ROSA) equipment where possible in Steam Generator 'C' and pull and replace the plugs in Steam Generators 'A' and 'B' by manual means using hydraulic plug pulling equipment. Initial attempts to remove plugs using hydraulic plug pullers were unsuccessful due to failure of the plugs near the second land from the tubesheet when the expander was being relieved or the pulling was in progress. As a result, the remaining plugs were drilled out.

Tube ends from which plugs were removed required additional inspections to ensure remaining conditions were suitable for the replacement mechanical plug.

Plugs from material heat NX6323, which have a hot leg margin scaling factor of 8' 'C

[ ] (See Table VI-5) were used for all replacement plugs on the hot leg side.

For tubes plugged due to the results of the eddy current inspection program, material heat NX6323 was primarily used on both the hot leg and cold-leg sides. However, NX5222 was used to plug 15 tubes on the cold leg of Steam Generator 'C'. Material heat NX5222 has a scaling factor of [ ]

8'

'" on the cold leg side. This was the only available material when the decision to begin the plugging operation in Steam Generator 'C' cold leg was made ('C' cold leg was the first plugging operation).

1

TABLE VI-7 NORTH ANNA UNIT 2 MECHANICAL PLUG REPAIR PLAN STEAM GENERATOR OUTAGE 'A' 'B' 'C' TOTAL 8/87 Hot Leg w/o Cold Leg 9 6 22 37 Sentinel (3962, 3513, 4523) 8/87 Hot Leg w/ Cold Leg 2 1 2 5 Sentinel (Fatigue Susceptible) 3/86 Hot Leg 2 1 6 9 (3962)

TOTAL 13 8 30 51 I

)-

) 1

\-

I j TABLE VI-8 NORTH ANNA UNIT 2 1

MECHANICAL PLUG REPAIR PLAN (DETAIL) 8/87 3/86 8/84 TOTAL Total Plugs 310 18 6 334

< Good Heats) < 0> < 0> < 6> < 6>

Total 310 18 0 328

< Sentinels > <118) < 0) < 0> (118)

Total 192 18 0 210

< Hot Leg w/ Sentinels) <118) < 0> < 0> <118)

~

Total 74 18 0 92

< Cold Leg good for at least one cycle > < 37 < 9) < 0> < 46)*

Total 37 9 0 46 Fatigue Susceptible 5 0 0 5 Total 42 9 0 51

Includes Heats 3962, 3513, and 4523

}

) - ee -

1

.j

,i,J j

j=5

)

n io t

)

n r

o o it P r d o e t P

d e d n e e a

p h

s d e n x a h e n

b u

t e p t

c U

(

T f

e x h E(

s o

e b e

u p

o e e b b u u k T T T T S g 8aL g

l u :

, f, ' [

P g t

l a t i

h c

n a

c I

Vj l

$e{,

l e _

M

i<

VII, BASIS FOR- RETURN TO SERVICE 1 l

)1 1

i In order for a plug top to be released with sufficient energy to penetrate the wall of the tube, the following must occur:

- 360 degree circumferential crack must proceed until s 0.005 inch ligament

.is left without leakage,

- Tube above the plug must be dry

- If plug is held together by interfacial load, then load must change relative to the pressure load.

- Primary pressure has sufficient access to the plug to accelerate the plug section so that it has a force in excess of [ ] a,e uponstri5tingthe tube.

. The basis'for return to service is to deny at least one of these required conditions.

1) PWSCC sufficient to reduce plug wall to less than 0.005 inches.

Primary water stress corrosion cracking (PWSCC) attacks Alloy 600 by a mechanism which is not yet fully understood. Ilowever, it is known that the 1

l process is affected by the temperature of service, the time at temperature, stress and the microstructure of the material. Temperature, time and stress ,

I are interrelated factors which vary inversely (e.g. the higher the temperature or stress, the less the time to failure). It also appears from work performed by Westinghouse each particular microstructure has its own curve of temperature

f. and time relationships. These curves can vary greatly in relative time to failure.

f I I !

I

Westinghouse has completed efforts to document the microstructure of the various plugs existing in North Anna and to determine the relative failure times of these heats at the nominal TH and TCtemperatures. This is based upon testing of representative heats in both pure steam and aggressive steam, in mock-ups.

From the results of these tests, Westinghouse developed a scaling factor based on the relative. test time to failure and kinetic calculations. For example, if

" failure time" is defined as the time to obtain (near) thru wall cracking, a

,e factor of approximately [ ] a, is observed between the failure time of the heatt with the preferred microstructure and the failure time of the heats with the least preferred microstructure. Table VII-1 documents the results of their work as it relates to North Anna.

TABLE VII-1 Relative Time-to-Initiate PWSCC Plug In North Grain Boundary Scaled Time vs. Temperature 'F Heat Anna Unit Carbide 618'F (Thot) 556*F (Tcold)

- ~~ ~ ~

NI3962 1,2 Less than semicontinuous s,b,c NI3513 1,2 Less than semicontinuous j l

NI4523 1,2 Semicontinuous NI5222 2 Semicontinuous l

K11989 1 Continuous N12205 1 Continuous l N12387 1,2 Continuous l NI6323 1,2 Continuous NI9789 i Continuous

) - - _

i i

Based on the history of NX3962 at North Anna, the base "1" in the above table is approximately equal to a fuel cycle.

Virginia Power has accepted the concept of scaled time factor developed by Westinghouse. However, based on the North Anna experience with heats NX3962 and NX4523 Virginia Power has chosen for the next fuel cycle of Unit 2, not to use these heats or heats with similar microstructure on the hot legs of the Unit 2 steam generators (Unit 1 later). The heats in this group are NX3962, NX3513, NX4523, and NX5222. However, industry experience and the scaled time factors do provide adequate data to allow the continued use of these four heats on the cold leg. To date, the industrf has not identified PWSCC in any plug installed on the cold leg side of a steam generator. Additionally, at North Anna, none of the heats have been in service for longer than two fuel cycles.

Therefore, this continued use on the cold leg for an additional cycle is justified by the scaled time factor with a margin of "'

[ ] cycles or more.

The heats which have a continuous chromium-carbide microstructure are acceptable for service, based on the scaled time factor, on either the hot leg or cold leg. At North Anna Unit 2, none of the heats with continuous microstructure have been in service for longer than three fuel cycles.

Therefore, their continued use on the hot leg for an additional fuel cycle is

'C justified by the scaled time factor with a margin of [ ] "' cycles. (The margin on the cold leg would be [ ]"' 'c )

l Westinghouse presently has additional laboratory testing underway which I

will provide more data for input into the scaled time factors. Additionally, l

[ field data will continue to' increase as more of the plugs are reworked in the

. field. Virginia Power will continue to evaluate this data and its effect on the scaled time factors.

l 4

2) Tube B. hind the Plug is Dry

),

The tube may be filled with water from three sources - leaking plug, a l

sentinel plug, or from a leaking tube defect. In the first case, a leaking ]

plug, the pressure behind the plug would be equal to the pressure in the tube i or greater. Therefore, no potential would exist to transport the plug top {

toward the tube. However, North Anna will not leave a known, leaking mechanical plug in service.

In the case of the sentinel plug, the driving potential for the plug would only be equal to the differential pressure between the inlet and outlet side of the tube sheet. At this pressure, the plug would have a force of less than [

l a.c; far less than the [ ] "'" required to rupture a tube.

Some tubes in North Anna 2 have been plugged due to leaking defects.

However, the actual leak rate cannot be predicted. Therefore, no susceptible

~ heat plug wi11' remain in service even though leakage from a known tube defect would allow water on the tube side of the plug.

3) Limiting access of primary pressure to the plug remnant Sufficient amounts of pr' mary pressure fluid must have continuing access to the plug section to accelerate it to the minimum force of ( ) **"

required to fail the tube. If a device, a plug-in-a-plug, can be installed into the existing mechanical plug, which would deny this level of access, then the plug section could not fail the tube by impact. Such a device is presently l under development by Westinghouse, but not available for use on North Anna Unit 2.

- 100 - .

i

) I

_- o

2. Plug removal required related to Unit 2 North Anna Unit 2 has had three (3) refueling outages prior to the 1989 l outage which required tube plugging. During the Fall of 1984, six plugs total were installed in all three (3) steam generators. These plugs were of heat NX2387 (remaining scale factor of 13). All will remain in service. All hot leg plugs installed during the 1986 and 1987 outages now have negative scaling factors. These hot leg plugs will be removed and replaced unless the hot leg I plug is in a tube with a sentinel plug on the cold leg ade. A total of 51 hot leg plugs will be removed from all three steam generators - 13 in S/G 'A'; 8 in S/G 'B'; and 30 in S/G 'C'.

Virginia Power will continue to evaluate the state of the mechanical plugs !

each refueling outage based upon the scaled time factor presented in this l section. If the scaled time factor falls below one (1), the plug will eititer be removed from service or otherwise made serviceable by other acceptable means (e.g. installation of a plug in a plug design).

3. UFSAR Bounding Condition Based upon the above discussion, Virginia Power will remove from service all plugs which it has determined can fail in such a manner to cause a failure of a tube similar to the failure of R3C60 in Steam Generator 'C', Unit 1.

Therefore, Virginia Power is confident a repeat of this event will not occur in the next fuel cycle. Similar corrective action will be taken each outage to l

ensure continued safe operation. However, if such an event was to occur, the leak rate would be controlled by the plug expander to approximately 74 gallons per minute. This was demonstrated by this event and supported by calculations 1

f and data obtained from Westinghouse. As stated, the expander must remain in place to control the leak rate. Flow through the expander during an event will

- 101 -

) i

_ i

exert a force of approximately 600 lbs. toward dislodging the expander. From actual field data, it is known that a force between [ ] a,e i, required to dislodge the expander. Therefore, the expander can be expected to stay in place.

In order to exceed the UFSAR boundary conditions from a tube rupture, nine (9) of the plugs would'have to fail in a manner similar to R3C60, Unit 1. This is considered an unlikely event.

Operational Aspects The procedures used during the February 25,-1989 event have been evaluated and enhancements identified. These enhancements will .be incorporated' into.

applicable procedures prior to. restart. Also, assessment of the licensed {

operator training program was made in order' to determine if the training-provided for -steam generator tube leaks was adequate. No additional training requirements were identified except to complete briefing on the February 25, !

1989 event. These briefings will be completed prior to restart and will include operator response, the procedure changes discussed above, the tube failure mechanism, and results of steam generator inspections and repairs.

1

(

~ 1 l

I L - 102 - ;

i l

VIII. LESSONS LEARNED I l

1) The augmentation of licensed operator training after the July 15, 1987 SGTR has been effective in maintaining operator skills to diagnose and mitigate 1

steam generator tube leaks. l l

4

2) The diagnosis of the affected s te am generator was complicated by the i relative low primary to secondary leak rates, low radioactivity in the secondary system, and time required to obtain and analyze S/G blowdown and condenser air ejector exhaust.

3) The principle mechanism for the detection of primary to secondary leakage at power is N-16 Radiation Monitors. The N-16 Radiation Monitors could not be used during this event because the primary to secondary leak did not occur until af ter the reactor trip.

4) The use of an Abnormal Procedure rather than Emergency Operating Procedures identified differences in the mitigation techniques.

5) The time to complete accountability is extended when a large number o,f contractors are working in the station as was the case since Unit 2 was in a refueling outage.

l l 6) Tube plugging records should identify the specific plug material heat installed in each tube.

- 103 - -

I l

l IX. CONCLUSIONS j k

A. The overall response of personnel, procedures and equipment to the steam l generator leak event was excellent. All safety related equipment, with the exception of those noted in the report performed as designed. The leaking I

steam generator was isolated without overfilling the steam generator or lifting any of the steam generator relief or safety valves. Also, pressurizer level was restored and reactor coolant pressure was stabilized without going solid in the pressurizer or lifting a pressurizer relief or safety valve. The operational performance exhibited is attributed to the classroom and simulator training received on steam generator tube ruptures which is structured to integrate the Abnormal Procedures, the Emergency Operating Procedures, Emergency Plan Implementing Procedure, and the Critical Safety Functions.

The licensed operators on shift at the time of the event found the Emergency Operating Procedures to be very effective.

Several areas within Abnormal Procedures were identified as possibly needing further review and this review has been completed.

The emergency response by the various emergency organizations was equally effective. The interim Station Emergency Manager correctly classified the l emergency in a timely manner and notifications to the NRC, State and Local authorities were performed as required by the Emergency Plan. As each of th,e 1

emergency facilities was activated, an orderly turnover of responsibilities occurred. Once activated, each facility fulfilled its duties and responsibilities as required by the Emergency Plan.

l B. Calculations of radioactivity released during the event show a minimal release. This negligible release is attributed to the timely manner in which

- 104 -

the operating shift controlled the plant and isolated the various possible radioactive release paths, as well as the very low activity in the RCS at the time of the event.

C. The consequences of this leak event were bounded by the Chapter 15 accident analysis in the UFSAR in terms of maximum break flow rate, integrated primary-to-secondary mass transfer and offsite dose consequences. The core was effectively and adequately cooled throughout the event and the integrity of the fuel was not compromised in any way.

D. The plug in the hot leg of R3C60 in the 'C' Steam Generator, Unit 1, failed due to primary water stress corrosion cracking (PWSCC). The failure of the plug in the hot leg of R3C60 in Steam Generator 'C' of Unit I was the result of the following conditions existing simultaneously:

I 360' degree cracking to a near thru-wall state in an Alloy 600 material failing due to PWSCC.

Dry tube behind the plug.

Sufficient potential behind the failed plug remnant to accelerate it to a force level greater than [ ] a,c ,

l l

Continued operation of the steam generator is justified, because, at least one of the above conditions has been eliminated for each plugged tube remaining l in each steam generator.

- 105 -

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

ML20245K731

Contents

Text

SUMMARY

SUMMARY

SUMMARY

SUMMARY

SUMMARY

Navigation menu

ML20245K731

Text

SUMMARY

SUMMARY

SUMMARY

SUMMARY

SUMMARY

Navigation menu

Search