Text

Insp Rept 50-443/89-82 on 890628-30.Major Areas Inspected: Event Sequence,Causes & Safety Significance Contrary to Test Procedure Reactor Tripping Criteria on 890622
	ML19325D661
Person / Time
Site:	Seabrook
Issue date:	07/28/1989
From:	Dudley N, Eselgroth P, Guenther F, Lois L, James Trapp, Wiggins J; NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:	;
Shared Package
ML19325C164	List: ML19325D661;
References
	50-443-89-82, NUDOCS 8910250253
	Download: ML19325D661 (42)
	v • d • e

p . .,.

>

'

"

,

.

,

-

. . - . <

,-

+,

,1

.

,

4

.

,

H

(:.V, , . '

p.

,

L

U.S. NUCLEAR REGULATORY COMMISSION

,

'

REGION I

<

Report No.: 50-443/89-82'

Docket No. 50-443

.

License..No. NPF-67'

Priority

Category C

Licensee: ~Public Service of New Hampshire

New Hampshire Yankee Division

Post Office Box 300

Seabrook, New Hampshire 03874

Facility Name: Seabrook Station Unit No. 1

' Inspection At: Seabrook, New Hampshire

Inspection Conducted: June 28-30, 1989

Inspectors:

_

M2fNf

P. W. Eseigp6th, Team Leader, R1

date

.

(See attached sheet)

.N. F. Oudley, Sr. Resident Inspector, R1

date

(See attached sheet)

L. Lois, Team Member, NRR

date

(See attached sheet)

J. M. Trapp, Team Member, R1

date

(See attached sheet)

F. Guenther, Team Member, NRR

date

j

Approved:

IMuM

7/d8

.WiggigTeamManager,RI

date

!

Inspection Summary:

See Executive Summary

i

!

hl'

J

g g,3G,u

-

$M 'IE5! SEh3

0

.

... .-.

-

..

- .-.

-

.

..

- -

.

,

-

4

-

3

.I -

5

'

U.S. NUCLEAR REGULATORY COMMISSION

REGION I

-

Report No.

50-443/89-82

Docket No..

50-443

License No.

NPF-67

Priority

Category C

-

Licensee:

Public Service of New Hampshire

,

New Hampshire Yankee Division

,

Post Office Box 300

Seabrook, New Hampshire 03874

Facility Name:

Seabrook Station Unit No. 1

Inspection At: Seabrook, New Hampshire

Inspection Conducted: June 28-30, 1989

Inspectors:

P. W. Eselgroth, Team Leader, R1

date

'?c u

'$ ,$'.

Es ,

7 -N - 87

~

'

N. F. Dudley, Senior' Resident, RI

date

.

L. Lois, Team Member, NRR

date

bb

~7 - -l - 8 9

J,p.Trapp,idamMember,RI

date

F. Guenther, Team Member, NRR

date

' Inspection Summary:

See Executive Summary

.

L.

!

l

1

l.

l

.

.~.

..

_

.

. . .

_

. . .

. . ,

..

r:

~

1

-

.

I

i

g

,

>

.

U.S. NUCLEAR REGULATORY COMMISSION

REGION I

I

Report No.

50-443/89-82

Docket No.

50-443

,

License No. .NPF-67

Priority

Category C

'

,

Licensee: Public Service of New Hampshire

.

New Hampshire Yankee Division

Post Office Box 300

Seabrook, New Hampshire 03874

Facility Name:

Seabrook Station Unit No. I

4

Inspection At:

Seabrook, New Hampshire

'

Inspection Conducted: June 28-30,1989

'

.

Inspectors:

P. W. Eselgroth, Team Leader, RI

date

-

.

N. F. Dudley, Senior Resident, RI

date

ka ~.6y b[

7!/7!D

L. Lois, Team Member, NRR

date

l'

l

J. M. Trapp, Team Member, RI

date

!

,'-

'8 4

H.7 ; y;

-

F. Guenther, Team Member, NRR

date

i

Inspection Summary:

See Executive Summary

l-

l

en-

.

,

.

j

". ' ,

,'

W:

'

-

L

'

A

s

h

I

li

'

i

TABLE OF CONTENTS

l

F

l

E592

'

b

'1. 0 Introduction

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

S'

)

,

.

i

1.1 Scope of Inspection

.

t

1.2 Team Composition

,

p

~ 2.0. Executive Summary . . . . . . . . . . . . . . . . . . . ...

5

4

o

. [l

2.1 Event Summary

[

2.2- Assessment Summary

1

3.0 Event Detcription . . . . . . . . . . . . . . . . . . . . .

9

'4.0

Plant and Equipment Performance . . . . . . . . . . . . . . .

10

>

4.1

Introduction. . . . . . . . . . . . . . . . . . . . . .

10

'

4.2 Plant Response . . . . . . . . . . . . . . . . . . . .

10'

4.2.1

RCP Trip to Steam Dump Valve (15-PV-3011 Failure Open

4.2.2 Steam Dump Valve MS-PV-3011 Failure Open to Closure of

All Steam Dump Valves

4.2.3 Steam Dump Valve Closure to Reactor Trip

4.2.4 Summary of Plant Equipment Response

4.3 Steam Dump Valves _. . . . . . . . . . . . .

13

.....

,

4.3.1

Introduction

4.3.2 MS-PV-3011 Failure to Modulate

4.3.3

Steam Dump Valve History

l

4.3.4

Valve Failure Cause

4.3.5 Licensee Short-Term Response

'

4.3.6 Licensee Long-Term Response

1

5.0 Personnel Activities and Performance

15

l

...........

l

15

5.1 Operating Crew .

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

.

L

-

_

..

_ ._

e n

..

EM

-

.-.

'

,

i

.j

E

-,

1

J

a.

.j .

,

.,

i

h-

!

.?

I.

TABLE OF CONTENTS

,

.P,1gg

,.

'

5.1.1

Organization.and Responsibilities

h

!

5.1.2. Training

.

!

5.1.3 Pre-Test Briefing

o

5.1.4 Crew Response

5.1.5. Performance Assessment

'

5.2 Sta rtup Te st Group . . . . . . . . . . . . . . . . . .

22

t

5.2.1 Organization and Responsibilities

i

5.2.2 Test Procedures

'

5.2.3 Training

'

5.2.4 ' Pre-Test Briefing

5.2.5 Test Group Response

.

5.2.6 Performance Assessment

,

.

L

5.3 Management and Support Staff . . . . . . . . . . . . .

26

-

Management and Other Support Personnel

1

L

5.3.1

5.3.2. Management Responsibilities

.

5.3.3 Management Response

1

l-

5.3.4

Performance Assessment

29

6.0 Safety Assessment .

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

6.1 Reactor Safety Significance of Event

'

6.2 Safety Significance of Personnel Performance

7.0 Exit Interview ............................................

30

_-

. . . .

.

..

.

-

. .-.

,

-

i

.

,

e,

.,.

,

'

,

.,

8.

... m

't

TABLE OF CONTENTS

>

,

, APPENDICES

Appendix A:

Chronology of. Events

Appendix B:

Chronology of Communications

'

-

Appendix C:

Individuals Interviewed

Appendix D:

Entrance Interview ~ Attendees

Appendix E:

Exit Interview Attendees

Appendix F:

NRC Observations Regarding Seabrook

Natural Circulation Test

' Appendix G:

Augmented Inspection Team (AIT) Charter

Appendix H:

Plant and Equipment Performance Figures

Appendix I:

Acronyms and Initialisms

'

,

5

e

!

I

,_

-,.

. . ~ .

-

o

.

-

'

.

a

.

(

.

,

s

'

~.

1.0 Introduction

1.1 Scope of Inspectio_n

In response to the performance of a natural circulation test at the

Saabrook Station Unit No.1 in a manner. contrary to the test

procedure reactor tripping criteria on June 22, 1989, the NRC formed

-

.

an Augmented Inspection Team (AIT) to determineithe event sequence,

causes and safety significance. This was accomplished by

establishing a chronology of the event (Appendix A) and accompanying

communications (Appendix B), and reviewing equipment perfonnan:e,

plant staff actions relative to this occurrence and applicable

'-

station procedures.

The NRC Team held an entrance interview with plant management and

support personnel on June 28, 1989 and performed the' inspection

t

'

during the period of June 28-30, 1989. An exit interview was

conducted with plant management on June 30, 1989.

Individuals

interviewed during the course of the inspection are listed in

Appendix C.

Attendees at the entrance and exit interviews are listed

in Appendixr.5 D and E.

Appendix F contains the statement of NRC

observers present during the June 22nd test. Appendix G is the

memorandum of assignment of the AIT to this Seabrook Unit 1 event.

.

Appendix H contains plant and equipment performance figures,

1.2 Team Co:nposition

The team was composed of a team leader and four headquarters and

I

regional specialists with expertise in plant operations, reactor core

and plant systems, operator training, test programs and management

control s .

2.0 Executive Summary

2.1 Event Summary

On June 22, 1989, the plant conducted the natural circulation test of

the primary system which is part of the reactor testing program.

This test gathers prirnary system data under controlled conditions to

derconstrate the ability of the reactor coolant system to remove decay

'

heat using natural circulation.

!

At the initiation of the test, the reactor was operating at about 2%

power and heat was being removed from the plant by the steam dump

valves.

Shortly after the point in the test where the reactor

l'

coolant pumps were turned off to establish natural circulation of

l

[

the primary system coolant, one of the three controlling group steam

dump valves (MS-PV-3011) malfunctioned and went to the fully open

l-

l

position, resulting in a rate of heat removal from the primary system

beyond what was planned for the test.

At this time none of the

credefined criteria in the natural circulation test procedure

I

. . . , .

.

_ _ _ , .

_

.

. .

.

_- . , .

. _ , -

- .

.

v

>.

t

..

.

.c

' 6-

-

3

.

U

'

i>

(1-ST-22) for termination of the test were exceeded nor was the

MS-PV-3011 position problem recognized. The presence of this

equipment problem and the accompanying rate of heat removal resulted

"

in a primary coolant average temperature transient that resulted in

the coolant level in the pressurizer decreasing towards one of the

. test's reactor trip criteria at the pressurizer 17% level.

.

Prior to pressurizer level reaching the 17% point, the Unit Shift

Supervisor (USS), a Senior Reactor Operator, informed the Test

'

Director (TD)- that one of "your limits" is being approached. When

.

pressurizer level reached 17% (at which letdown is automatically

L

!

isolated and pressurizer heaters are deenergized) the Senior Control

Room Operator (SCRO), whu is the primary side reactor operator,

informed the USS of this, but did not mention the associated reactor

i

'

f

trip requirement. At this point the USS conferred with the TD; steam-

'

1

dump valve MS-PV-3011 had been shut; the pressurizer level decrease

had been halted; and, pressurizer level had begun to increase.

It

-

was on the basis of the pressurizer level recovery taking place that

the US$ decided to allow the reactor to continue to operate in support

of the test.

However, the USS did not correlate the isolation of

letdown indication with.the loss of pressure control and the need

-

to trip the reactor in accordance with the pressurizer 17% level

.

criterion,

i

An increasiag reactor coolant pressure transient was now developing

due to the closure of the malfunctioning steam dump valve

(MS-PV-3011) with the subsequent recovery of pressurizer level, and

the USS directed that the reactor be tripped due to primary plant

pressure approaching the test procedure trip criterion.

The shift

crew then carried out the emergency operating procedures for a

I

reactor trip and the natural circulation test (1-ST-22) was

,

terminated.

2.2 Assessment Summary

i

f

The conclusion of the AIT staff regarding the licensee's response

to the plant transient resulting from the malfunction of one of the

steam dump valves is that reactor plant safety was never in question.

I

and with the exception of the significant error of not tripping the

I

reactor at the point first called for by the test procedure and loss

of pressure control due to letdown isolation and pressurizer heater

deenergization, the operating staff performed well.

The following summary of assessments is provided with references to

the sections of the report where further details are documented:

The actual plant dynamic resconse was reviewed and compared to

o

the post trip review predicted response. The plant responded as

predicted in the June 22nd natural circulation testing including

the very mild overcooling event which resulted from stoam dump

valve MS-PV-3011's failure to properly modulate.

(Section 4.1)

l

,_

.

,

,, -

- ._ --

.

_ __

,

,-

c

..

.

,

-7-

'

,.

Plant equipment was not ready to support the June 22nd test.

'

o

Prior to commencing the test, a test prerequisite to confirm the

'

availability of the steam dump system was signed off.

However, there was an open work order for post maintenance

stroke testing of steam dump valve KS-PV-3011. (Section 4.3.3)

The interviews of the Unit Shift Supervisor (USS), Senior

o

'

Control Room Operator (SCRO) and Control Room Operators (CRO)

found them to be highly competent individuals, clearly aware of

their assignments for safe operation of the plant.

In

particular, the USS communicated that he had no doubts about

'

being the one responsible for conduct of the test in a safe and

controlled manner.

(Section 5.1.1)

Training relative to the conduct of the natural circulation test

o

which covered details of the expected plant response had been

accomplished about a year prior to the test. The AIT found no

evidence that such training had been repeated or refresher

training given since that time. (Section 5.1.2)

'

A review of the pre-test briefing that was conducted for the

o

operators by the Test Director determined that it was inadequate

with respect to covering the details of the testing to be

performed and thoroughly reviewing the reactor trip criteria.

(Section 5.1.3)

,

The operating crew was observed to be conducting plant

o

operations in a controlled, unfrenzied manner prior to, during

the test and following the reactor trip when the applicable

emergency operating procedures were entered and carried out

appropriately.

(Section 5.1.4)

During the Low Power Testing program prior to the June 22nd

o

,

event, as well as during this event, there was no evidence

of pressure applied by management or anyone else to complete

testing at the expense of controlled, safe operation of the

plant.

In fact, the NRC has been aware of personnel assigned to

shift operating and test responsibilities having received

direction from management to proceed with testing in a

>

l

controlled manner and specifically to not permit themselves to

,

feel rushed into completing evolutions.

(Section 5.1.4)

ihe USS did not trip the reactor at the 17% pressurizer level, as

o

l

called for in the test procedure (1-ST-22). He stated his reason

,

was that the decrecsing pressurizer level was under control and

)

I.

turning around. The AIT concluded that a cause of this event was

I

the lack of importance and/or sense of ownership placed on test

I

procedure requirements by the USS as compared to his other operating

l

. . .

.-

..

---

.

. . . .

-

.

.. . -

- - . .

.

1

p<

b

.

, , .

(v

.,

p

.g.

,

p

,

<

i

F

requirements such as those contained in Technical Specifications

l

and plant operating procedures.- Two other operators interviewed

also indicated the perception of a hierarchy'of importance for

procedural requirements'between test procedures and plant

operating procedures.

These misunderstandings on the part of

r

the operators demonstrated an absence of recognition of test

procedure criteria as controlling requirements for

.

,

operation under testing conditions.

(Section 5.1.4)

E

The Shift Superintendent (SS) did not provide effective

o

supervisory involvement in the conduct of this test. (Section

5.1.5)

P

From the interviews of operating crew personnel it has been-

o

,

'

concluded that these personnel now. recognize and understand that

the proper action was to have tripped the reactor before the

1-ST-22 trip criterion on pressurizer level was exceeded.

(Section 5.1.5)

The startup test group had responsibility to interrupt or

v

o

terminate the test in the event that required plant conditions

i

were not maintained.

However, no such recommend 4 tion was made

to the shift operating crew by the test group even though the

Startup Manager was made aware of the NRC's concern about the

plant being below a manual trip criterion. The overall

direction given by the test organization during the performance

of this test was inadequate.

(Section5.2.5)

From the interviews of startup group personnel it has been

j

o

concluded that these personnel now recognize and . understand that

the proper action was to have terminated the test and recommend

i

to the operating crew that the reactor be tripped before the

!

1-ST-22 trip criterion on pressurizer level was 2xceeded.

I

(Section 5.2.5)

During the conduct of 1-ST-22 and at the time when plant

o

c.onditions had reached the reactor trip criterion associated

with pressurizer level, there were several plant management

representatives in the control room with the responsibility and

l

authority to terminate test and plant operations when approved

procedures are not being followed. When members of management

,

I

having specific responsibility and authority relative to safe

'

i

operation of the plant are present in the control room, their

L

presence in no way dilutes the responsibilities of the operating

crew and test group personnel assigned to shift. However, by

virtue of the particular responsibilities and authorities that

they do possess relative to safe plant operations, there is a

responsibility

particularly during unique testing situations -

,

l

to keep themselves informed of key limits for operation and

'

plant status relative to those limits and to take appropriate

l

action relative to plant operation whenever others they have

-

l

1

l-

1'

.

_

,

- - . .

.

-

--,

. .

,

s

, w

,

-

.

t.

assigned to do this have not done so. This was not done by the

i

management members present.

(Section5.3.3)

The initial management thrust following this event appeared to

o

be to resolve any equipment problems necessary to resume

.testinD. An in-depth review of the cause or causes leading to

the improper conduct of the 1-ST-22 natural circulation test

-

apparently did not take place prior to an initial management

decision to resume testing. A thorough review of this event

was not completed by the licensee until after the NRC raised

this issue with licensee management. (Section 5.3.3)

.

3.0 Event Description

.

The following description of the event was determined through

observations, interviews with the operators and review of the plant

computer traces and printouts. A chronology of the event is presented in

Attachment A.

'

.

On June 22, 1989, the plant was at about 2% rated power in preparation for

the performance of tne natural circulation test, which was intended to

y

demonstrate the ability of the reactor coolant system to remove decay heat

using natural circulation. At approximately 12:19 p.m., the reactor

coolant pumps were tripped. The loop average temperatures began to

increase, as' expacted, and the pressurizer level and pressure began to

increase. At 12:25 p.m.,

the steam dump valves began to modulate open and

one valve failed full open resulting in a rapid cooldown of the primary

system. During the cooldown, pressurizer level dropped below 17% at 12:29

p.m.

This caused an automatic isolation of letdown and deenergization of

- the pressurizer heaters.

The steam dump valve was manually shut at 12:31

p.m. and the cooldown was terminated.

Level in the pressurizer did not go

below 14%.

l

Pressurizer level increased above 17% at 12:34 p.m. and a corresponding

l

pressure overshoot occurred. At 12:35 p.m. the reactor was manually

i

tripped due to primary plant pressure approaching the test procedure trip

l'

criteria.

The pressure rise was terminated prior to reaching the auto-

matic trip set point due to the manual reactor trip. A reactor coolant

I~

pump was started and primary plant temperatures were stabilized at 12:50

p.m.

At no time during the transient was a reactor protection or

engineered safeguards features actuation setpoint reached.

The natural circulation test contains a manual trip criterion which states

that the reactor must be tripped if pressurizer level decreases below 17%.

l

NRC inspectors recognized that a manual trip was not initiated when

i

pressurizer level dropped below 17% and informed the Startup Manager, the

L

Test Director, and the Assistant Operations Manager of the requirement to

trip the reactor.

However, no apparent steps were taken to direct the

tripping of the reactor prior to the manual reactor trip for increasing

i

I

primary pressure.

1

l

..

.

..

- -

..

.-.

-

- . .

E ,

.-

,

'

,

-

, e

,i

>

.r.

l[

f

g

4

10-

h

<

,

j

4.0 Plant and Equipment performan g

)

4.1

Introduction

This section covers plant dynamic response including the' steam dump

valves.. In accordance with the AIT charter the objective is to

" determine the expected plant response during a transition to natural

,

circulation cooling and compare it to the actual plant dyncmic

i

response observed during the event." In addition " assess the scope

and quality of ... licensee identified concerns and corrective

actions."

Information was collected through interviews with PSNH employees and

from GETARS (General Electric Transient Analog Recorder System).

This segment of the report is divided into two major parts:

1.

Plant response to an overcooling transient, and

.

2.

Mechanical and electrical instrumentation aspects of the first

steam-dump valve-bank, before, during and after the test.

'

4.2 Plant Response

For this test the reactor was heavily borated at 1150 ppm boron with

all control rods fully withdrawn except for bank D rods which were

'

positioned at 130 steps out of the core (full out equals 228 steps).

There were no safety systems or safety functions bypassed for this

test however, the plant was being operated under special test

conditions which allowed the reactor to be critical at power without

reactor coolant pumps operating. All low reactor coolant flow trips

are automatically blocked below the P-7 permissive setpoint

(approximately 10% power).

There is adequate data collection by the plant's GETARS system

computer to reconstruct the vital plant parameter behavior with the

exception of the valve MS-PV-3011 response. The reason for the lack

of valve MS-PV-3011 data is that a connecting link of the (Bailey)

-

positioner of the valve feedback mechanism had become disconnected

which affected both valve operation as well as the computer indica-

'

tions.

For the primary coolant system transient there are three distinct

time segments, that is:

1.

From the RCP trip to the steam dump valve opening, 12:18:50 to

12:25:56 (5465 to 5891 sec in GETARS indication)

2.

From the opening of the steam dump bank to their closing

12:25:56 to 12:31:06 (5891 to 6202 see in GETARS indication)

3.

From the steam dump valve closing to reactor trip 12:31:06 to

12:35:54 (6202 to 6489 sec in GETARS indication)

.

__

___

_

.

. .

^

i

'

t

s

. ,

..

.

-11-

Total transient time 12:35:54 - 12:18:50 = 17 min-4 sec.

The following'three report subsections discuss each of the distinct

time segments in detail.

-

'

4.2.1

RCp Trip to Steam Dump Valve MS-pV-3011 Failure Open-

After the RCPs were tripped, the total heat input into the primary

coolant system decreased by about 12 MWt, the total heat input from

the primary pumps. The reactor was already at about 2.2% power and

the steam dump valves were in manual because they were used to dis-

.

pose of the reactor total heat input of about 86 MWt, (i.e., 74 MWt

of nuclear heat and 12 MWt of primary pump input). With reactor

power at about 2.2% of rated power, Th and Tc loop temperatures

showed the initiation of natural circulation with Th rising to about

570*F and Tc dropping to 545'F and Tavg rising by a few degrees in

all loops.

(See Figures 1.1 to 1.4, Appendix H).

Pressurizer level

-

and pressurizer pressure increased as the average reactor coolant

r

temperature increased due to loss of forced circulation, see Figure

2 (Appendix H). Steam generator level stayed constant as well as

1.he charging and letdown flows, see Figure 3.

Steam generator

pressure began decreasing due to cooldown, see Figure 5.

Decreased

primary circulation rate caused coolant and fuel temperature in-

.

creases in the core which in turn decreased core power due to

doppler feedback, see Figure 4.

The core configuration for the test

(control rod position and boron concentration) were such that the

moderator temperature coefficient was about zero thus, doppler was

'

the only feedback. About 5 minutes into the test, core flow was

removing the generated heat, thus Th stopped rising. However, Tc

continued falling due to steam dumping. Therefore, Tavg began to

decrease.

Steam generator pressure was also decreasing due to the

mismatch between steam dumping level and power and heat production

before the RCP trip (at 86 MWt) and after the PCP trip (at 74 MWt).

With decreasing reactor power and Tavg, pressurizer pressure and

level began to decrease and at 12:24:56 (6 min 6 see into the test)

the (condenser) steam dump valve control was lost due to the Lo-Lo

Tavg interlock at 550*F (P-12). This occurred because this inter-

lock operates on the narrow range Tavg signal which is located on

a bypass loop and without forced circulation cools faster than the

reactor coolant.

P-12 was bypassed through control room switches

and the operator regained steam dump manual control through the

first valve bank on the steam pressure mode. As soon as P-12 was

bypassed the valves attempted to return to their existing demand

position at about 5%.

However, MS-PV-3015 was blocked due to a

pre-existing excessive air leak, valve MS-PV-3011 went open (i.e.,

probably f ailed to modulate) and only valve MS-PV-3019 operated

properly. Within 40 seconds valve MS-PV-3019 closed, but MS-PV-3011

most likely stayed open.

In this brief time interval steam demand

increased, charging flow continued to increase and pressurizer level

.

- - .

. - - . . .

- - -

.

. . - , -

.

- - . .

.

..

. .

- .

- . . . - - . -

-

lt . 3

[

o

.

.;

. ' .

t

-12-

'

"

0

3

continued to decrease. The operator. responding to decreasing pres-

surizer level, further decreased the letdown flow, see Figures 2,

3, and 5.

As soon as valve MS-PV-3011 closeo, steam generator level

showed a small rapid increass for a few seconds, see Figure 2.

i

In this first time segment the reactor responded as expected. All

major parameters varied in the expe:ted direction and within ex-

,

pected ranges.

Post-event inspection showed that MS-PV-3011

probably failed to modulate.

(

4.2.2

Steam Dump' Valve MS-PV-3011 Failure Open to Closure of All Steam

Dump Valves

Six seconds after MS-PV-3011 went closed the operator manually be-

'

gan to open the first steam-dump valve-bank to initiate energy dis-

posal. The valve-bank valves are supposed to modulate in unison

4

,

with instruments in manual pressure control. The control board sig-

nals for valves MS-PV-3015 and MS-PV-3019 were correct and as ex-

pected. However, valve MS-PV-3011 went fully open and stayed in

,

that position as was verified a few minutes later by actual obser.-

,

vition.

l

As steam dumping continued, pressurizer pressure and level decreased.

and the operator responded by increasing charging flow, decreasing

(to almost zero) the letdown flow and closing the main steam drains.

In addition Th began to dacrease and Tc showed a sharp downturn,

resulting in decreasing Tavg. At this time nuclear power generation

shows a slight upturn from a minimum of about 1.4% due to excess

heat removal and fuel cooldown. Steam generator pressure decreased

due to excessive cooldown see Figures 2, 3 and 5.

l

This plant behavior, that is, excess cooling of the RCS, was caused

by valve MS-PV-3011's failure to modulate and being fully open.

The valve failure was established by visual inspection during the

transient. As this trend continued, at 12:28:54 the pressurizer

level fell below 17% (which is a procedural reactor trip level),

and, as a result, letdown isolated and pressurizer heaters de-

energized, causing the loss of normal pressurizer pressure con-

a

trol. At this point in time the pressurizer pressure was 2192

psia and reached a minimum value of 2179 psia. Pressurizer level

,

continued to fall and reached a minimum of 14.5% at which point

the operator closed the dump valves. Valves MS-PV-3011 and MS-

l

PV-3019 went fully closed.

In this time period the plant responded as expected in view of the

l

excess cooling of the RCS.

Each valve when fully open discharges

i

about 3.3% of total steam load, thus, with reactor power at about

!.

1.5%, valve MS-PV-3011 fully open and valve MS-PV-3019 partially

l

open the heat loss was at times over 4.0% and the primary system

j

heat loss exceeded the heat input from the reactor.

However, valve

MS-PV-3011 failed to modulate and the operator failed to trip the

!

-

--

. ... - . _,

.

--

.

-

.

-

--

-

- - - . - . - . . .

-13-

'

l

reactor as required.

(Note:

the valve failure to modulate is

discussed in paragraph 4.3.)

4.2.3

Steam Dump Valve Closure to Reactor Trip

Within a few seconds of steam dump valve closing, pressurizer pres-

sure and level began rising,

Charging rate was at 122 gpm (about

1% level / minute) and letdown was isolated.

Likewise, steam gene-

)

'

rator pressure and level began to rise after a small dip in the

i

level and an upturn in the pressure, see Figures 2, 3 and 5.

j

Reactor power leveled off at about 2.5%. As pressurizer level and

,

pressure increased rapidly the operator realized that pressure was

i

getting close to 2340 psia (another procedural trip requirement),

i

At 12:35:54 the reactor was tripped at a reactor pressure of about

2310 psia and the operators entered emergency operating procedure

l

E-0 in response to a reactor trip. The rise of pressurizer level

i

to 17% was recorded at 12:33:55.

Nerefore, the reactor stayed

!

below 17% for about 5 min.

l

!

In this time segment, the reactor coolant system responded as would

,

be predicted due to reduced cooling at high charr.ing rate and zero

!

letdown.

4.2.4

Summary of Plant Ecutoment Response

f

,

The plant response during operations t-iated to the natural circu-

!

lation test was as would be predicted, and all plant parameters be-

!

haved normally. A steam dump valve MS-PV-3011 failure to modulate

c

caused an unanticipated cooling of the reactor coolant system. All

phenomena were explainable and no unexplained parameter values were

i

observed.

i

4.3 Steam Dump Valves

4.3.1

Introduction

t

The origin of the primary cooling transient was the malfunction of

steam dump valve MS-PV-3011, which stuck open and failed to modu-

late.

This section reviews valve performance, operating record,

failure root cause and licensee short term and long term response.

Most of the information regarding the valves was obtained from

post-event examination.

i

t

4.3.2

MS-PV-3011 Valve Failure to Modulate

!

Post-event examination revealed that a connecting link nut to a

>

positioner arm fell off. This mechanism was providing the feedback

,

and the disconnected link explains the failure to modulate.

Never-

e

theless, valve MS-PV-3011 was able to respond to the final fully-

closed signal from the control room.

,

f

-- ,

,

, ,,- , . , , . , , - - , -, - - , , . . , , . . , , - - . ,

.

-.

- - - - , . , . -

.n-

-

. - . . . , - - , - - -

'

-14-

4.3.3

Steam Dump Valve Hittery

Interviews with the system support personnel, revealed the

l

following with respect to the steam dump valves:

1

MS-PV-3011 was stroked after the natural circulation test

.

and failed to operate properly due to mechanical binding.

Examination after removal of the valve mechanism revealed

that the binding was caused by stem misalignment and

,

interference with a guide bushing.

At the beginning of the preparation for the natural

<

circulation test, valve MS-PV-3011 was not ready to support

'

the test since work order WR87 WOO 5592 was still open for a

stroke test at NOP/NOT.

In spite of this, a test pre-

requisite to confirm the availability of the steam dump

system was signed off.

i

)

There is no indication as to when the linkage in valve

i

MS-PV-3011 failed,

it had been tested earlier from the

i

control room for close#open positioning, however, this

test would not reveal the linkage problem.

After the June 22nd event, binding was also found in valve

MS-PV-3019 but not enough to prevent open/close motion or

modulation.

Post-event testing of cil steam dump valves revealed that

seven of the twelve valves showed binding, scored stems

!

loose linkage or tight linkage.

In general, the history of

steam dump valvo system work orders indicates that there is

a valve maintenance or design ~ problem.

.

!

4.3.4

Valve Failure Cause

i

It is concluded that the MS-PV-3011 steam dump valve failure

l

cause is apparently inadequate valve maintenance or design.

!

Licensee personnel failed to follow through on a pending work

order and failed to recognize and resolve a maintenance problem

,

with the steam dump valves.

In addition the licensee failed to

l

adhere to test procedures by failing to assure that the required

j

test prerequisites and initial conditions were met before

p

commencing the test.

!

v

,

'

l

.

-

.

. _ .

-

- - - -

.

. .

- .

.

-

- .

a

t

,

-

,

-15-

4.3.5

Licensee Short-Term Response

The following actions were taken or initiated by the licensee

while the AIT was on the site:

i

dismantled the valve HS-PV-3011 mechanism for shop testing

i

'

called a vendor representative to the site

.

initiated extensive diagnostit. testing for all steam dump

'

valves,

replaced the valve MS-PV-3011 actuator with a new unit from

j

'

storage, and

i

performed a comprehensive logic circuit test.

These actions envelope an appropriate review of the behavior of

'

the steam dump valves and constitute a technically sound, prompt

!

and adequate response to the specific valve problem.

l

4.3.6

Licensee Lono-Term Response

,

Licensee personnel expressed their intent for a complete

detailed and in-depth investigation of the valve problem so as

'

to be able to take the appropriate corrective action.

,

It is the team's understanding that the licensee will

investigate:

generic failure rate data base for this type of valve

t

'

seek to verify whether valve usage (including surveillance

testing) is related to failure frequency, and

review (and if necessary revise) the current valve

i

maintenance and Surveillance program.

[

These actions appear to be appropriate.

t

r

5.0 personnel Activities and performance

,

5.1 Operatino Crew

-

5.1.1 Organization and Responsibilities

'

Seabrook Station's normal control room shift crew composition

and the crew composition that existed during the day shift on

June 22, 1989 were reviewed. Normally, while in mode 2

-

(startup) operations, the Unit I control room optrations staff

'

,

would consist of a Shift Superintendent (SS), a Unit Shift

Supervisor (USS), a Supervisory Control Room Operator (SCRO) and

a Control Room Operator (CRO). Both the SS and the USS possess

l

a senior reactor operator license and the SCR0 and CR0 must be

licensed as reactor operators or senior reactor operators.

This is consistent with the minimum requirements for licensed

operators per shift for on-site staffing of nuclear power units

-

-.

.

.. . _ .

. .

. - . .

.

-

..

_.

.

!

.

-16-

l

specified in 10 CFR S0.54 and in the facility's Technical

!

'

Specifications.

In anticipation of performing the natural

l

circulation test on the morning of June 22, 1989, the normal

!

shift complement was augmented with additional CR0s to assist in

performing various control room functions: one operator was

[

held over from midnight shift to assist in acknowledging

secondary alarms; one was assigned to control steam generator

!

,

level and reactor coolant system temperature; a third operator

was assigned responsibility for turbine shell and chest warming

(he was never used, however); and a fourth additional operator

i

monitored the radiation monitor panels. This crew augmentation

,

allowed the operators normally assigned to the shift to

i

concentrate on the reactor and the primary plant.

The inspectors reviewed a number of facility licensee documents

!

in an effort to determine the operating shift crew's responsi-

,

bilities during normal operations and upset conditions and

during the startup test program.

j

Section 2.3 of the Seabrook Operations Management Manual (OPMM)

discusses the control room command function and states that the

$$ is the senior on-shift manager and is responsible for the

,

control room command function.

It goes on to state that the SS

may, and normally will, delegate this responsibility for each

unit to its respective USS. The SS, under Section 3.3.2 of the

!

OPMM, retains the authority to assume command of the control

room, or to order the shutdown of the reactor when, in his

judgement, such action is required to protect the safety of the

unit or the health and safety of the public.

Furthermore .the

i

,

SS is responsible for the safety and operation of the unit

!

equipment, in accordance with approved Station procedures.

-

Section 1.1 of the OPMM provides an overview of shift operations

and states that the SS maintains a broad perspective of

i

conditions affecting the status and safety of the unit, while

i

the USS maintains a comprehensive perspective of operational

conditions affecting the safety of the unit and is in charge of

'

.

the control room during emergencies.

Section 4.2.4 of the Startup Test Program Description (STPD)

states that the station staff will perform its normal job func-

tions as required to support plant operations and the startup

test program. Although the Test Director has the primary

responsibility for the execution of the test, the station

operating crew has the responsibility for the proper operation

of equipment, systems, and the plant and reserves the right to

,

take appropriate corrective actions whenever unsafe or unsatis-

factory conditions exist. A determination by the SS or USS that

i

a test would place the plant in an unacceptable condition is

identified in Section 4.3.5 of the STPD as an event which

constitutes grounds for a test interruption.

_.

.----

.

- .

.

..

.

-

.

i

~

.

-37-

,

Section 1.5 of the Seabrook Station Management Manual (SSMM)

{

addresses the issue of procedural adherence and states that

where a procedure exists, it shall be considered guidance

l

regarding the method of performing a function.

Procedures shall

l

be followed, but not without question.

If a procedure directs

,

an action contrary to what is considered proper, the operator

should question the procedure and seek resolution with appro-

[

.

priate supervisory personnel.

It states, however, that 6

,

procedure being questioned should not be deviated from on the

i

basis that it is being questioned.

5.1.2 Training

The inspection team reviewed the operators' startup test program

l

training completed in preparation for low power testing and

other aspects of the licensed operator training program which

'

may have had a bearing on this event.

l

During the period from April 14 to May 23, 1986, the facility's

plant reference simulator was used to train all the operators on

}

the tests that would be run during the startup program. The $$,

'

USS and the two CR0s having primary plant responsibility during

,

the natural circulation test were verified as having completed

!

.

that training. Additional classroom training on the low power

test program was conducted as part of the licensed operator

i

requalification training program during the period from

September 12 to October 21, 1988. This course was observed by

i

an NRC inspector and was addressed in Inspection Report 88-13.

l

The four-hour course, which was conducted by the Assistant

,

Startup Program Manager, provided a detailed description of the

.

startup testing program. The course topics included program

l

administration, organization, test equipment, and applicable

procedures, including 1-ST-22, the Natural Circulation Test.

'

The training also provided the operators with an awareness of

the startup test program structure.

The licensed operator initial and requalification training

programs were reviewed to determine whether deficiencies in

diagnostic and team training or in command and control and

procedural compliance training may have contributed to the event

,

on June 22.

It was determined that these subjects are addressed

i

in classroom and simulator training during the initial and

i

requalification training programs. Operations Training Standard

i

Number 3, dated January 1989, states that procedural adherence

is required with deviation allowed only after procedure changes

have been made or in the case of the emergency response

l

procedures by invoking 10 CFR 50.54(x).

This training standard

is endorsed by both the Operations and Training Managers.

Interviews with a Training Department representative indicated

that the operators are trained to comply with all approved

station procedures, regardless of whether they are operations

.

_

.

-- - - -

_

.

- -

- . .

-

.

..

-

-.

!

,

t

,

-18-

!

t

procedures, administrative procedures or test procedures. The

'

CR0s are instructed to advise the USS when and if reactor trip

criteria are approached and/or exceeded and to trip the reactor

l

unless directed otherwise by the U$S.

!

>

5.1.3 Pre-Test Briefing

The licensee's requirements for operating crew briefings were

'

reviewed to determine whether those conducted in preparation for

the natural circulation test were adequate.

Section 1.8 of the

,

OPMM addresses shift evolution briefings and states that they

!

shall be conducted for individuals involved in the performance

!

of the evolution.

The detail of the briefing depends on the

!

complexity, logistics or number of people involved in the

!

evolution.

Evolutions involving many individuals, especially

from two or more departments or disciplines, may require large

formal briefings or planning sessions.

It goes on to state that

complex evolutions requiring close coordination of individuals

should include the following five elements:

review of the

appropriate section of the procedure by key individuals;

!

examination of each inc'.ividual's specific involvement and

responsibility; discussion of expected results or performance;

i

'

review of precautions, limitations, emergency actions to be

taken if contingencies arise; and assurance that everyone

understands the required interface and communications required.

l

!

The inspection team interviewed the operators involved with the

conduct of the Natural Circulation Test,1-ST-22, on June 22,

1989, and it was determined that the operators were not briefed

!

as a crew prior to commencing the test procedure. The operating

7

crew members were individually briefed by the Test Director (TD)

during the early hours of their shift.

Copies of the procedure

,

j

had been distributed to the operators the preceding day but not

l

all the operators had taken the time to review it in detail; the

,

USS reviewed the procedure on the morning of June 22nd. The

l

primary plant CR0s were given copies of the manual reactor trip

!

criteria, Attachment 9.3 of the test, just prior to commencing

i

the test. One of the CR0s and the SS never received an

individual briefing. The $$ did, however, read the procedure

three days before the test was attempted, but he did not have

a copy available to him at the time it was being performed,

r

Immediately prior to commencing the test, the TD provided a

general overview briefing of the test objectives and procedure

geared for the management observers and operators from other

crews present in the back of the control room. This briefing

i

provided a brief overview of the test and was not directed to

the operator: performing the test.

I

!

-

.

..

-

- -

-

-.

..

--- -

71

-

..

\\

-19-

i

'

Discussions with NRC inspectors who were present during earlier

phases of the low power testing program indicated that the

.

pre-test briefings for the natural circulation test were less

1

thorough than others had been; previous tests h

also generally

included some sort of pre-shift group briefing i tiher than

relying solely on individual briefings. The SS did not find out

j

until after the test was aborted that the operators in his crew

i

'

had net been prcperly briefed, and the USS indicated during his

I

interview that while other pre-test briefings have been short,

)

they have generally been more thorough than what was done for

the natural circulation test.

,

i

5.1.4 Crew Response

l

'

A detailed description of the event is provided in Section

4.0 of this inspection report and a chronology of significant

.

events is provided in Appendix A.

A chronology of communica-

tions during the event is provided in Appendix B.

,

It became evident during the operator interviews that the

primary CR0s and the USS were aware that pressurizer level was

decreasing and approaching the 17% manual reactor trip criterion

!

specified in Attachment 9.3 to 1-ST-22. The $$, not being as

familiar with the test trip criteria as the rest of the

,

operating crew and not having a copy of the procedure to which

'

he could refer, suspected that level had decreased to less than

or equal to 17% when he heard letdown isolate, but he did not

associate the letdown isolation with a manual trip requirement.

The primary CR0s and the USS were aware that letdown had

i

isolated at 17% pressurizer level and that the manual reactor

t

trip criterion had been satisfied,

i

The question of why the operators did not promptly trip the

reactor when they realized that pressurizer level had decreased

below the 17% trip criterion was pursued by the inspection team

-

in the interviews.

The primary CR0s knew that the US$ was aware

of the level control problems and that he was also aware, as

they were, of the requirement to manually trip the reactor.

However, the CR0s never actually recommended to the USS that

the reactor be tripped.

Interviews with the CR0s indicated

that they were generally aware of discussions taking place

.

1

between the USS and the Test Director (TD) regarding the loss

of pressurizer level. The US$ informed the TD that pressurizer

level had decreased below "your limit."

In the interim, the

i

US$ directed the primary CR0s to monitor. level and to report

when it reached 15%. At about this time the control room

received a report from an operator in the plant that one of

the condenser steam dump valves had failed full open. The valve

was promptly closed, terminating the cooldown transient and

reversing the pressurizer level decrease at approximately 14.5%.

Both level and pressure began to recover quickly after closing

'

,

-

. - -

- _ - - - . .

-.

..-

.

. .

-

.-.

-

. - -

. - . - .

-

!

,

-

~20-

i

l

!

the failed open steam dump valve, and the operators quickly

!

tried to restore pressurizer pressure control capability.

l

Without pressurizer spray or letdown capability, pressure rapidly

increased past 2300 psig and was approaching the high pressure

reactor trip setpoint of 2385 psig. Realizing that pressure

>

was continuing to rise, the USS directed that the reactor be

.

manually tripped. The total elapsed time from the point when

!

pressurizer level decreased below 17% until the operators

!

manually tripped the reactor was approximately five minutes.

l

lt was apparent from the operator interviews that there was no

,

doubt in their minds that the command and control function in

l

the control room rested with the USS and not with the TD. The

l

USS informed the TD that level was at 17% and decreasing but

i

t

he failed to recognize that the test procedure 17% pressurizer

level trip criterion required him to direct chutdown of the

i

reactor at this point without further discussion or deliberation.

During his interview, the USS indicated that he did not trip the

'

reactor because other operating procedures do not require a trip

,

until a lower pressurizer level. Since pressurizer level

,

appeared to be stabilizing as it passed through 17%, he made the

decision not to insert a manual reactor trip at that time.

It

,

was only after the steam dump valve was closed and pressurizer

!

pressure began to rapidly increase toward the automatic trip

i

setpoint that the US$ decided that recovery from the transient

l

was not feasible and a manual trip was necessary.

}

!

l

The NRC inspectors who were present du"ing the natural circula-

i

!

tion test witnessed the crew's response to the reactor trip and

their performance of the emergency operating procedures. No

performance deficiencies were noted during this post trip

response.

l

Through observations and interviews the inspectors determined

i'

the Emergency Operating Procedures (EOPs) were adequately

implemented following the manual reactor trip.

The Emergency

,

Operating Procedures are normally implemented with two operators

!

l

at the control panel, however during the natural circulation

I

test there were four operators at the control panel.

No prior

discussion had been held by the USS as to how the operators were

l

to implement the E0Ps. At the inception of EOP implementation

.

I

the operator's recognized the need to adjust to the situation

!

and reached an unspoken agreement that only two of the four

l

operators would conduct the E0P procedure. As a result one

of the additional operators who was designated as the Shift

'

Technical Advisor (STA) performed the EOP control board

manipulations.

If the E0P recovery had been extended, the

inspectors were uncertain whether this operator would have been

free to perform his STA responsibilities.

Forty-five minutes

'

af ter the manual trip the NRC was notified in accordance with

-

- , -

.

- - - -

- - - - .

. -

- . - .

.-. -

[

i

-

,

!

'

-21-

!

i

,

10 CFR 50.72 by the Shift Superintendent. This was well within

,

the four hour reporting requirement,

j

,

5.1.5 performance Assessment

'

The operating crew did not comply with an explicit procedural

!

requirement to manually trip the reactor even though they were

.

fully aware that the established trip criterion had been

!

exceeded.

The CR0s should have recommended to the USS that the

!

reactor be tripped before level exceeded the 17% pressurizer

l

level criterion. The operator interviews revealed an apparent

I

tendency by some of them to place higher priority on satisfying

some procedural requirements than others. Some of the operators

'

had attached a greater safety significance and importance to

!

complying with a Technical Specification or emergency operating

procedure requirement than, for instance, a test procedure

!

requirement.

Subsequent to arriving at this conclusion from the

!

operator interviews, the team viewtd a video tape of the

June 22nd natural circulation test in which the US$, when

discussing the pressurizer level problem with the TD, referred

,

to the 17f4 pressurizer level reactur trip criterion as "your

?

limit".

The USS apparently felt comfortable that the situation

was under control since he had not yet approached a lower level

i

trip criterion established in the emergency operating pro-

'

cedures.

This hierarchical approach to procedural compliance

is not endorsed by the facility's administrative policies nor

!

by the operators' licensing and continuing training programs.

'

i

The pretest briefing conducted for 1-ST-22 appears to have been

>

-

inadequate in that the operators were never formally briefed as

'

a group to address the five elements identified in Seabrook

[

Station's OPMM.

All complex evolutions, particularly those

!

involving new or infrequently performed tasks, should be

thoroughly briefed. The fact that the natural circulation test

simulator training had been performed over three years earlier

i

and the classroom training was almost a year old should have

t

provided added incentive to ensure that the operators receive

some refresher training and were thoroughly briefed prior to

l

commencing the test.

The fact that the operators on shift that

morning had not routinely worked together as a crew should have

,

emphasized the need to examine, during the pretest briefing,

each operator's specific involvement and responsibility and

understanding of the required interfaces and communications.

These observations and the assignment of the STA function to a

'

panel operator, discussed in the previous section, indicate the

,

need for more thorough planning and preparation to have been

'

done prior to this test.

Although the SS normally serves an oversight function in the

control room, his level of awereness, knowledge and involvement

of shift evolutions was not commensurate with the significance

-

. -

- . - .

. . .

-

- . - -

. .

l

\\

v

'

.gg.

L

and complexity of this test.

The fact that the natural circu-

'

,

lation test is one of the first evolutions performed with a

critical reactor and the fact that the test involves abnormal

operating conditions should have been sufficient to raise the

$$'s level of awareness and involvement.

i

NDTE:

The video taping referenced above was done by the licensee for

)

use by the training department in future training sessions. The

-

inspection team found the video tape to be supportive of the

information obtained from the interviews and the resultant

conclusions.

,

5.2 Startup Test Group

5.2.1 Organization and Responsibilities

The organization and responsibilities of the startup staff are

!

delineated in the Startup Test Program Description (STPD), Rev.

2.

The Startuo organization is led by the Startup Manager. The

Startup Manager has tFe overall responsibility for the initial

startup program and reports to the Station Manager.

The

l

Startup Supervisor reports to the Startup Manager.

The Startup

i

Supervisor is responsible for detailed coordination of the

startup test program.

Reporting to the Startup Supervisor

are the Shift Test Directors. The Shift Test Director's

responsibilities include in part to insure required test

conditions are established in a safe and prudent manner, and

i

meintained as necessary for test performance. The startup

staff normally present in the control room during startup test

performance are the Shift Test Director and the Test Director.

The Test Director reports to the Shift Test Director and is

t

responsible to perform individual startup tests.

At the time of the Natural Circulation Test Performance,

the Startup Supervisor was the Acting Shift Test Director.

During the performance of the Natural Circulation Startup

l

Test, 1-ST-22, the Startup Manager, Shif t Test Director,

.

and Test Director were all present in the Control Room.

The responsibilities of the Startup Staff and the Station

Operating crew for specific activities are provided in Table 1-1

.

of the Startup Test Program Description.

Test coordination and

i

direction activity is designated as being the responsibility of

the Startup Test Department.

The responsibility for systems and

equipment operations is delegated to the Station Operating crew.

4

Section 4.3.5 of the Startup Test Program Description states

l

that the Startup Supervisor or the Shift Test Director will

t

I

determine if a startup test should be interrupted. An example

l

of events which may warrant a test interruption provided in the

"

Test Program Description is the inability to maintain plant

1

-

.

.~.

n

,

.

)

.

'

-23-

,

!

conditions as specified in the startup test.

Section 4.3.6 of

l

the Startup Test Program Description states that the Startup

i

Manager, Startup Supervisor, or Shift Test Director will

i

determine if a test will be terminated. An example of a test

1

.

termination event is if the performance of a test procedure

i

'

reveals design or equipment deficiencies which prevent the

i

objectives and/or acceptance criteria from being met. During

'

.

the natural circulation test, a plant condition (pressurizer

)

level greater than 17%) was not maintained due to the steam dump

!

valve problem which prevented test objectives from being met.

'

However, no interruption or termination action was taken by the

i

Test Organization,

i

l

5.2.2 Test Procedures

i

>

Methods to change Startup Test Procedures are described in the

Startup Test Program Description. Test procedure changes may be

made utilizing two methods.

For major changes, a procedure

revision is required.

Procedure revisions undergo extensive

i

review and comment cycles including review by Westinghouse.

The

procedure is recon. mended for approval by the Startup Manager and

reviewed by the Station Operations Review Com:nittee (SORC) prior

!

to being approved by the Station. Manager.

Field procedure

'

.

changes fall into two categories: intent changes and non-intent

changes.

Procedure changes which involve a change of intent

,

!

must be reviewed and approved by the Startup Supervisor, SORC,

and the Station Manager prior to being implemented.

Non-intent

procedure changes (e.g. editorial changes) must be reviewed and

l

approved by the Startup Supervisor and the Unit Shift Supervisor

,

(or another SRO) prior to implementation.

In the event that the

l

,

Startup Supervisor is unavailable the Shift Test Director may

'

provide this review and approval.

Non-intent changes are

reviewed by 50RC within 14 days of implementation.

l

A review of the Seabrook startup test procedures is documented

i

j

in NRC inspection reports 50-443/86-31, 86-48 and 88-13.

Each

!

of the inspection reports describes a small number of minor

i

l

procedure changes which, if incorporated, would more clearly

!

or correctly state procedural steps and test objectives.

The applicable changes described in the inspections were

incorporated into the startup test procedures prior to

implementing the Zero Power Test Program. The inspections

concluded that the startup test procedures were well prepared

and technically sound. Also, the number of test procedure

p

changes made during the test program thus far appears to be

less than other comparable facility test programs.

l

.-

._ .

.

-

..

- .

.

.- - -

_

-.

_

L

-

.

,

,'

'

-24-

.

j..

-

.

,

5.2,3 Trainino

'

Training requirements for the Shif t Test Director and the Test

!

'

Director are provided in the Startup Test Program Description,

Rev. 2, Section 5.2, " Personnel Training." The Shift Test

Director, and Test Director were provided training in those

l

aspects of the programppplicable to procedure compliance, test

,

performance, and test documentation. This training was provided

to students as a formal classroom lecture.

In addition, the

j

Shif t Test Directors and Test Directors were also provided

!

training on selected transients which might be expected as

abnormal occurrences during various startup tests. This

training covered general transient conditions which could occur

5

and did not explicitly cover cooldown transients during natural

circulation.

The transient training was provided on a self

l

study basis, without formal training handouts or lesson plans,

,

with an examination given at the end of the self study period.

Both the Shift Test Director and the Test Director met the

i

training requirements' described in the Startup Test Program

i

Description.

In addition, both the Shift Test Director and Test

Director attended an additional course on transient analysis

-

which was conducted in the simulator. No members of the Startup

i

Organization are operator license holders at the Seabrook

Station, nor are they required to be.

,

5.2.4

Pre-Test Briefino

Procedure 1-ST-22, Rev 2., " Natural Circulation Test " Step

i

,

3.2 states that " Personnel involved with the performance of

'

this procedure have been briefed on the procedure content

and informed of their respective duties." The Test Director

provided information copies of test procedure 1-ST-22 Rev. 2,

.

,

to the primary desk, the Unit Shif t Supervisor's desk, and the

!

f.

Shif t Superintendent's desk a few days prior to initiating this

procedure.

The actual execution of the pre-test briefing and

i

sign-off of the procedural step occurred a few hours prior to

,

l.

initiating the test by the Test Director speaking with the

l

,

l

licensed operators individually on shift. The briefings were

very short according to the operators, but did cover the manual

i

,

l

trip criteria. The Test Director supplied copies of the manual trip criteria and expected plant response, Attachment 9.3 of

l

1-ST-22, to the primary operator, reactor controls operator, and

l

Unit Shif t Supervisor just prior to the test. The control board

'

operators and the Unit Shift Supervisor responsible for shift

operation, stated in interviews following this event that they

-

were made aware of the 17% pressurizer level trip criterion

during the pre-test briefing with the exception of one of the

-

two control board operators assigned to assist the shift crew

who stated he was not briefed and was not aware of the manual trip criteria.

l

1

_._ _ __. _ _ . . .

,

. _ . ,

.

_ _ . , _ . . . _ . . .

l

-

,

G

(

)

,

~25-

'

)

i

!

5.2.5 Test Group Response

l

The startup test group crew response was derived from

1

observations made by the inspectors during the event and

interviews held with the startup group staff following the

event.

The Startup personnel present in the control room during

i

the event were the Startup Manager, Shift Test Director, Test

,

Director, and other supporting Startup Engineers. Of the

l

Startup Staff only the Test Director was positioned inside the

'

operating area, other members of the Startup Group witnessed the

test from inside the control room but outside the operating

area.

Prior to the event the Test Director was communicating

i

test instructions with the operating staff and monitoring test

,

data. Af ter the reactor coolant pumps were tripped, per the

'

test procedure, and the test initiated, the Test Director

primarily monitored computer and panel indications. The Test

Director stated that during the event he was aware that the

pressurizer level had decreased below the manual reactor trip

criteria of 17% when it was announced by the control board

operator that letdown had isolated. At this point the Test

i

Director did not recommend to the Unit Shift Supervisor to trip

the reactor.

He indicated to the Unit Shift Supervisor that he

,

would monitor computer trends for Tavg to assure that the 15

l

minute Technical Specification on Lo-Tavg was not violated.

Performing this task essentially removed the Test Director

!

from the overview t,f plant status.

During the period when

l

pressurizer level was below 17%, an NRC inspector monitoring

l

the test activities expressed a concern to the Test Director

,

l

that the pressurizer level was below the manual trip criteria.

Following this communication the Test Director stated that he

'

told the Unit Shift Supervisor that the NRC has a problem with

being below the manual trip criterion. The Test Director stated

that the USS said he was handling it. At no time during this

test did the Test Director recommend that the operators manually

,

trip the plant.

,

The Shift Test Director was monitoring test activities from

,

outside the operating area. The Shift Test Director stated that

he was aware of the manual trip criterion for pressurizer level

I

and that the value was exceeded during the test.

He stated that

he focused on attempting to analyze the on going transient and,

l

therefore, did not provide an advisory role to the operating

staff. At no time during this transient were recommendations

provided by the Shift Test Director to the operating staff.

The Startup Mar,ager was monitoring test activities from outside

the operating area.

The Startup Manager stated that he first

became aware of the pressurizer level being below the 17% trip

l

criterion when informed of such by the NRC inspector monitoring

-

startup testing.

The Startup Manager did not communicate the

-

.

- - . -.

-

.

- . - .

.

.- -

. -

-.

.

4

~26-

,

inspector's concerns to the operating crew or other members of

the Startup Staff.

5.2.6 Performance Assessment

l

The inspection team concluded that the pre-test briefing

performed by the Test Director was conducted in a fragmented and

.

abbreviated manner. Due to the interactions which occur between

,

plant systems and operator actions, it is important to perform

the operator pre-test briefings as a group rather than in a

piece meal fashion.

Three levels of the Startup organization were aware that the

pressurizer level was below the manual trip criterion during the

transient.

Only after the NRC inspector voiced a concern did a

startup organization member (Test Director) indicate a concern

to the operating staff. At no time during the performance of

.

this test did any member of the startup group communicate to

the station operating staff a recommendation to interrupt or

terminate the test procedure. The technical guidance provided

by the startup organization to the operating staff during this

event was inadequate.

In general, the startup organization became more occupied with

individual tasks at the expense of maintaining at least one

individual with overall responsibility for overview and

technical input to the Unit Shift Supervisor for conduct of

the test procedure.

5.3 Management and Support Staff

5.3,1 Management and Other Support Personnel

During the performance of the test there were approximately

seven managers in the main control room.

The Vice President

of Nuclear Production was the most senior manager present.

The Operations Manager and the Assistant Operations Manager

were the only managers in the control board area.

Approximately twenty licensed operators, in addition to the

operating crew, were in the main control room to observe the

natural circulation test to fulfill the commitment in Final

Safety Analysis Report (FSAR) request for additional information

response.

These operators remained outside the control board

area, did not become involved with plant operations and

maintained a quiet presence throughout the test.

5.3.2 Management Responsibilities

The responsibilities of the Station Manager and the Assistant

Station Manager, both of whom were present in the control room

L

+

,

L

-27-

F

during the test, are delineated in the Nuclear Production

Management Manual (NPMM). The Station Manager is responsible

for ensuring the station is operated and maintained in

accordance with applicable requirements and he serves as

chairman of the Station Operation Review Committee ($0RC). The

Station Manager has the authority to direct reactor shutdown

when conditions may endanger equipment status or the health and

.

safety of the public. The Assistant Station Manager is

responsible for n.aintenance of the programs and procedures

needed to operate the station in accordance with applicable

requirements and he also has the authority to direct the reactor

to be shutdown.

The responsibilities for the Operations Manager and the

Assistant Operations Manager are delineated in the Operations

Management Manual (OPMM).

The Operations Manager is responsible

to direct operating activities in a safe and reliable manner,

supervise the Assistant Operation Manager and he is a member of

t

50RC.

He has the authority to order the shutdown of the reactor

.

when action is required to protect the safety of the station or

!

the health and safety of the public. The Assistant Operations

Manager has the responsibility for safe operation of the unit's

equipment and directs the activities of the members of the

operating crews.

He also has the authority to order shutdown of

the reactor.

5.3.3 Manacement Response

Of the managers interviewed, two were aware, during the test,

that pressurizer level had dropped below 17%. The Station

.

Manager was the only manager interviewed that knew of the

existence of a trip criterion on pressurizer level but was

unfamiliar with the exact criterion. Most of the staff

members interviewed are members of the 50RC which had reviewed

and approved the natural circulation test procedure.

Through interviews with the management staff and review of

management responsibilities in the NPMM and the OPMM it was

determined that four of the managers interviewed had the

authority to direct a reactor shutdown. However, none of these

managers communicated to the USS a need to trip the reactor when

i

pressurizer level decreased below 17%.

i

The Station Manager stated he was not sure why the USS did not

trip the reactor but believed it was due to the training the USS

had received in the simulator. The Operations Manager stated

that the US$ did not trip the reactor because the USS knew the

cooldown was causing the pressurizer level drop and that the US$

knew the cooldown was under control.

During an interview conducted on June 24, 1989, the Station

4

is -

- - . --

--. - . . . . .

.-m.

.

.__

_

-

. _ _ - .

,

'

i

-

.28-

i

?

r

\\

Manager stated that he recognized soon after the reactor trip

l

on June 22, 1989, that the failure to follow procedures was a

j

significant problem but had been unable to conduct a full

,t

discussion of the problem with his management team prior to

i

meeting with the VP Nuclear Production at approximately 5:00

(

p.m.

'

i

During a conference call on June 22, 1989 with the Region I

'

Branch Chief at 6:00 p.m. the VP Nuclear Production indicated

,

that the procedural compliance issue would be looked at and put

i

in proper perspective and that if the event occurred again he

I

would expect the operators to trip the reactor. The VP of

!

Nuclear Production initially indicated a desire to restart the

.

reactor early the next morning but agreed to postpone reactor

}

startup until after a follow-up conference call.

5

During the follow up conference call at 7:30 a.m. on

June 23, 1989, the licensee outlined the planned modification to

f

their management manucis that would provide additional guidance

on the implementation of procedures and outlined the briefings

!

that were planned with all shift crews to present the new

6

guidance. As a result of a subsequent phone call between the

!

Deputy Regional Administrator and the President of New Hampshire

!

Yankee Division a Confirmatory Action Letter was issued

requiring that a complete review and analysis of the event be

formally prepared and presented to the NRC prior to reactor

!

restart.

Immediately after the phone conversation the license's

Event Evaluation Team, the Human Performance Evaluation System

i

team, and the Independent Review Team were assigned to perform

'

separate evaluations of the event.

l

5.3.4

Performance Assessment

'

The initial management thrust following this event appeared

to be to resolve any equipment probitms necessary to resume

testing. An in-depth review of the cause or causes leading to

f

the improper conduct of the 1-ST-22 natural circulation test

!

.

apparently did not take place prior to an initial management

decision to resume testing. An extensive review of this event

was not completed by the licensee until after the NRC raised

this issue with licensee management.

During the conduct of 1-ST-22 and at the time when plant

conditions had reached the reactor trip criterion associated

i

with pressurizer level, there were several plant management

!

representatives in the control room with the responsibility

and authority to terminate the test and plant operations when

approved procedures are not being followed.

This was not done.

When a member of management having specific responsibility and

authority relative to safe operation of the plant is present

in the control room, their presence in no way dilutes the

..

._

_

.-.

- - - . - - . .

-

- _.

.

-

.

t

'

i

.

t

-29-

l

,

-

r

[

l

r

responsibilities of the operating crew and test group personnel

l

assigned to shift. However, by virtue of the particular

!

responsibilities and authorities that they do possess relative

l

to safe plant operations, there is a responsibility -

particularly during unique testing situations - to keep them-

i

selves informed of key limits for operation and plant status

i

relative to those limits and to take appropriate action relative

i

'

to plant operation whenever others they have assigned to do this

have not done so. Plant management present did not do this in

'

the case of the 17% pressurizer level trip criterion that was

exceeded. The reactor was subsequently shut down by the USS

!

when the transient response of another parameter, primary plant

.

pressure, caused the USS to take this action.

l

6.0 Safety Assessment

6.1 Reactor Safety Significance of the Event

,

!

The aspect:, of this event which cause the plant t.ansient to be

.

different from the intended natural circulation cest transient are

i

'

the failure of valve MS-PV-3011 to modulate and the fact that the

operators did not manually trip the reactor based on pressurizer

level. The excess cooling of the reactor coolant system is of

[

little or no reactor plant safety significance in that it is very

!

'

minor by comparison to other analyzed events (steam line break,

inadvertent initiation of a coolant loop, etc.) and these have

'

been analyzed and shown to be acceptable. The June 22nd event

,

is, therefore, totally bounded by these other analyzed events.

.

6.2 Safety Significance of Personnel performance

,

The failure of the operating crew to trip the reactor when required

by the test procedure during the June 22nd test; the failure of test

'

group personnel to recommend tripping of the reactor at that point

and the failure of management present in the control room to

,

exercise their responsibilities in this situation, despite the fact

-

the plant was being operated under a Technical Specification Special

Test Exception, is safety significant. Also, the apparent willingness

of management to proceed with testing following the June 22nd

occurrence without first completing a thorough review and causal

factor assessment is safety significant.

-

Test procedures often involve placing the plant in unusual

conditions for operation, conditions which are not routinely

'

experienced nor necessarily adequately covered by normal operating

procedures. Use of test procedures results in operation under an

'

approved margin of safety only when strictly followed. These test

procedures are carefully developed, utilizing industry experience

and expertise, are carefully reviewed and only approved after

confidence is established in their ability to assure plant s&fety.

The conduct of tests such as the natural circulation test in which

,

--

-

7-

~

-,-e

,., . ,

, , ,

,

, - , , - - . , , - - , -

- , - , . -

w e

- - , - - ,

--

a

.

l

-30-

,

the reactor is critical without reactor coolant pumps operating is

an example of testing under unusual conditions, conditions which call

for a heightened sensitivity to plant status and attention to strict

procedural adherence. Neither shift operators, the key test group

personnel nor the managers present in the control room during the

June 22nd test demonstrated an adequate understanding of this.

'

As stated in the previous section, the particular plant transient

which resulted from the combination of the steam dump valve

equipment problem and failure to follow the test procedure reactor

trir criterion did not significantly challenge the plant margin to

safety.

However, the operational practice exhibited by the

personnel in the control rcom was unacceptable.

The AIT concluded that all operations, test group and management

personnel interviewed now recognize that testing can proceed only if

done so in accordance with the test procedure requirements and that

if testing should for any reason proceed otherwise the test

procedure must first be formally revised. The AIT found no

indications of uncertainty or equivocation about this during the

site visit.

7.0 Exit Interview

On June 30, 1989 a preliminary exit interview was held with licensee

management to review the observationr,and assessments of the AIT. The

licensee was informed at the time that this interview might not be the

final exit for this inspection.

During this inspection, the NRC

inspectors received no comments from the licensee that any of their

inspection items or issues contained proprietary information. No written

material was provided to the licensee during this inspection.

On July 5,1989, the team briefed regional management on the results of

the inspection.

The licensee was informed by NRC Region I management that

f

the above exit interview would be considered the final exit.

,

r

i

5

t

!

,

-.

.

.-

.

..

_ _

_ - .

. _

_ - . .

.

'

.

l

i

'

APPENDIX A

CHRONOLOGY OF EVENTS

INSPECTOR

i

TIME:

EVENT:

OBSERVATIONS / ACTIONS:

'

,

12:18:00

Pre-RCP trip conditions

Pressurizer (PIR) pressure

i

2237.2 psig, PZR level 25.54%

,

Wide Range T cold $55.8 F.

t

All rods withdrawn except

bank 0 (step 133)

i

12:18:50

RCP breakers are opened

,

All 12 RCS low flow alarms

are received

12:25:15

Steam dumps control lost due

to permissive P-12 actuation

at $50'F Low Tavg

12:25:19

Steam dumps 3019 and 30))

began to open when P-12 was

manually bypassed.

.,

All steam dumps in bank should

i

modulate together

!

12:25:21

Steam dump valve 3011 closes

12:25:21

Steam dump valve 3011 opens

,